Deepwater Horizon Spill

Home , Anotopterus pharao, Bald notothen, Cotylorhiza tuberculata, Eel life history, Guachanche barracuda, King snake eel

Appendix by Ballengée et al. (B. Ballengée with T.Gardner, J. Rudloe, B. Schiering and P. Warny) 2012 26,160 preserved specimens representing 370+ species or >2.4% of the known 15,419 species of the Gulf of Mexico

DIAGRAM OF INSTALLATION…………………………………………………………2

SPECIES LIST BY TROPHIC LEVEL………………..……………………………..….3-9

SELECT PUBLICATIONS

Diagram of Collapse

Tier 7 ?

Tier 6 6.3 E 6.4 E E E 6.2 6.1 6.5

Tier 5 5.11 E E 5.6 5.2 E E 5.10 E 5.9 5.4 E 5.13 5.7 E E 5.1 E E 5.3 5.8 5.5 E 5.12 E

Tier 4 4.3 E 4.9 4.17 E 4.4 E 4.2 4.16 E 4.27 4.21 E 4.5 E 4.11 4.10 E E 4.30 4.26 4.29 4.14 E 4.18 4.25 E 4.8 E 4.12 E 4.24 E 4.28 E 4.6 E 4.31 E 4.15 4.1 4.23 4.22 4.19 E 4.7 4.20 E 4.13

Tier 3 E 3.36 E 3.21 E 3.45 3.6 E 3.11 3.1 3.39 3.46 E 3.15 3.40 E 3.42 3.22 E 3.38 E 3.32 3.27 E 3.30 3.35 E 3.8 E 3.49 E E 3.10 E 3.7 3.16 3.54 3.2 E 3.55 3.12 3.9 3.43 3.28 3.50 3.37 E 3.18 E 3.33 3.31 E 3.20 E 3.25 3.56 3.17 3.4 3.19 E 3.53 E 3.24 3.5 3.41 E 3.3 E 3.13 3.48 3.34 3.47 E 3.26 3.14 3.29 3.44 3.23 E 3.51 3.52

Tier 2 2.21 2.92 E 2.101 2.59 2.47 2.29 2.99 2.5 2.70 2.104 2.45 2.24 2.75 E 2.77 2.23 2.38 2.50 2.43 E 2.9 E 2.36 2.28 2.106 E 2.42 2.64 2.35 2.49 2.37 2.11 2.34 E 2.52 2.62 2.39 E 2.82 2.87 2.46 2.32 2.65 2.48 2.58 2.102 2.60 2.61 2.98 2.18 E 2.22 2.78 E 2.57 2.14 2.105 2.69 E 2.1 2.76 2.68 E 2.91 2.66 E 2.51 2.73 2.86 2.80 2.74 2.13 2.56 2.88 2.30 2.20 2.3 2.41 E 2.79 E 2.83 2.63 2.4 2.54 2.96 2.100 2.103 2.27 2.8 2.16 2.89 2.95 2.94 2.6 2.2 E 2.40 2.97 2.72 E 2.44 2.26 2.15 2.25 E 2.84 2.31 2.33 2.81 2.55 2.19 2.90 2.17 2.10 2.85 2.12 2.67 2.93 2.53

Tier 1 1.155 1.30 1.36 1.87 1.150 E 1.136 1.78 1.140 1.98 1.23 1.57 1.48 1.47 E 1.40 1.84 E 1.7 1.71 E 1.61 1.76 1.152 1.12 1.142 1.20 1.32 1.110 1.37 1.8 1.107 1.52 1.85 1.4 1.153 1.28 1.100 1.83 1.95 1.104 1.15 1.24 E 1.35 E 1.126 1.129 1.41 1.103 1.19 E 1.53 1.145 1.89 1.69 1.26 1.135 1.6 1.134 1.16 1.154 1.13 1.132 1.60 1.130 1.3 1.149 1.70 1.39 1.119 1.74 1.127 1.151 1.29 1.64 1.106 1.38 1.17 1.124 1.10 1.58 1.133 1.101 1.21 1.143 1.122 1.111 1.34 1.118 1.99 1.144 1.56 1.55 E 1.116 1.65 1.22 1.115 1.81 1.73 1.27 1.51 1.43 1.62 E 1.102 1.125 1.92 1.88 1.25 1.66 1.18 1.108 1.14 E 1.77 1.131 1.117 1.148 1.49 1.44 E 1.59 1.128 1.42 1.68 1.45 1.105 1.96 1.80 1.11 1.54 1.113 E 1.9 1.120 1.33 1.2 1.123 1.72 1.137 1.46 1.112 1.139 1.90 1.5 1.75 1.97 1.91 1.146 E 1.109 1.141 1.1 1.86 1.156 1.63 1.121 1.94 1.138 1.67 1.147 1.82 1.79 1.31 1.93 1.50 1.114

2 Specimen list by trophic level

Level Reference Species Scientific Name Type Quantity Origin Year 1 Col.1.1 Flower coral species Mussa spp. Partial specimens 5 Pet-trade 2012 1 Col.1.2 Organ pipe coral Tubipora musica* Partial specimens 15 Pet-trade 2012 1 Col.1.3 Thin finger coral Porites furcate Partial specimens 8 Pet-trade 2012 1 Col.1.4 Elkhorn coral Acropora palmate Partial specimens 1 Pet-trade 2012 1 Col.1.5 Frilly lettuce coral Lobophytum crassum Partial specimens 3 Pet-trade 2012 1 Col.1.6 Scarlet coral species Dichocoenia spp. Partial specimens 1 Pet-trade 2012 1 Col.1.7 Boulder coral species Monstastrea spp. Partial specimens 10 Pet-trade 2012 1 Col.1.8 Brain coral species Diplora spp. Partial specimens 1 Pet-trade 2012 1 Col.1.9 Cauliflower coral species Pocillopora spp. Partial specimens 3 Pet-trade 2012 1 Col.1.10 Table coral Acropora hyacinthus Partial specimens 23 Pet-trade 2012 1 Col.1.11 Branch coral Acropora florida Partial specimens 1 Pet-trade 2012 1 Col.1.12 Brownstem coral Pocillopora verrucosa Partial specimens 22 Pet-trade 2012 1 Col.1.13 Lace coral species Family stylasteridae Partial specimens 3 Pet-trade 2012 1 Col.1.14 Branching coral species Madracris spp. Partial specimens 9 Pet-trade 2012 1 Col.1.15 Blue ridge coral Heliopora coerula Partial specimens 4 Pet-trade 2012 1 Col.1.16 Large-cupped fungal coral Scolymia lacera Partial specimens 3 Pet-trade 2012 1 Col.1.17 Rooster tail conch Lobatus gallus Shells only 1 Wholesale supplier 2012 1 Col.1.18 West Indian fighting conch Strombus pugilis Shells only 4 Wholesale supplier 2012 1 Col.1.19 Florida fighting conch Strombus alatus Shells only 11 Wholesale supplier 2012 1 Col.1.20 Incongruous ark Anadara brasiliana Whole specimens 24 Markets 2012 1 Col.1.21 Southern quahog Mercenaria campechiensis notata Whole specimens 63 Markets 2012 1 Col.1.22 Stout tagelus Tagelus plebeius Whole specimens 10 Markets 1999 1 Col.1.23 Common razor clams Ensis directus Whole specimens 9 Markets 1999 1 Col.1.24 Propeller clam Cyrtodaria siliqua Whole specimens 20 Markets 1999 1 Col.1.25 Soft shell clams Mya arenaria Whole specimens 24 Markets 1999 1 Col.1.26 Neapolitan triton Cymatium parthenopeum Whole specimens 4 Markets 1999 1 Col.1.27 Northern moon snails Lunatia heros Whole specimens 6 Markets 1999 1 Col.1.28 Dwarf razor clams Ensis megistus Whole specimens 93 Markets 2000 1 Col.1.29 Hawkwing conch Strombus raninus Shells only 6 Wholesale supplier 2012 1 Col.1.30 Shark eye sea snail (didyma) Polinices duplicatus Shells only 51 Wholesale supplier 2012 1 Col.1.31 Graceful fig sea snail Ficus gracilis Shells only 2 Wholesale supplier 2012 1 Col.1.32 Florida crown conch Melongena corona Shells only 6 Wholesale supplier 2012 1 Col.1.33 Lettered olive sea snail Oliva sayana Shells only 40 Wholesale supplier 2012 1 Col.1.34 Florida calico scallop Argopecten gibbus Shells only 5 Wholesale supplier 2012 1 Col.1.35 White scallop Argopecten irradians Shells only 11 Wholesale supplier 2012 1 Col.1.36 Nucleus scallop Argopecten nucleus Shells only 28 Wholesale supplier 2012 1 Col.1.37 Fairyland snail sea snail Achatina achatina Shells only 10 Wholesale supplier 2012 1 Col.1.38 Pillow stinking sponge Ircinia strobilina Partial specimens 1 Wholesale supplier 2012 1 Col.1.39 Florida hermit-crab sponge Pseudospongosorites suberitoides Whole specimens 34 Wholesale supplier 2012 1 Col.1.40 Stinking vase sponge Ircinia campana Whole specimens 1 Wholesale supplier 2012 1 Col.1.41 Half-naked pen clam Atrina seminude Shells only 3 Ecological surveys 2005 1 Col.1.42 Atlantic plate limpet Lottia testudinalis Shells only 19 Wholesale supplier 2012 1 Col.1.43 Yellow sea fan species Gongonia spp. Whole specimens 1 Wholesale supplier 2012 1 Col.1.44 Green sea fan species Gongonia spp. Whole specimens 1 Wholesale supplier 2012 1 Col.1.45 Purple sea fan species Gongonia spp. Whole specimens 3 Wholesale supplier 2012 1 Col.1.46 Spiny oyster Spondylus americanus Partial shells 12 Pet-trade 2012 1 Col.1.47 Red sea lettuce Halymenia floresia Partial specimens 25 Markets 2012 1 Col.1.48 Tar balls N/A N/A 10 Ecological surveys 2011 1 Col.1.49 Mixed cone sea snail species Conus spp. Shells only 21 Wholesale supplier 2012 1 Col.1.50 Turrid snail species Crassispira spp. Shells only 839 Wholesale supplier 2012 1 Col.1.51 Yellow sea snails Littoraria spp. Shells only 753 Wholesale supplier 2012 Pourtale's abalone 1 Col.1.52 Haliotis pourtalesii** Shells only 51 Markets 2012 (pearlized) 1 Col.1.53 Knobbed triton snail Cymatium muricinum Shells only 146 Markets 2012 1 Col.1.54 Nerite snail species Nerita spp. Whole specimens 613 Markets 2012 1 Col.1.55 Eastern auger Terebra dislocata Shells only 17 Wholesale supplier 2012 1 Col.1.56 White donax or Coquina clam Donax variabilis Shells only 556 Wholesale supplier 2012

3 Specimen list by trophic level

Level Reference Species Scientific Name Type Quantity Origin Year 1 Col.1.57 Rockweed Fucus vesiculosus Partial specimens 12 Ecological surveys 2012 1 Col.1.58 Rough file clam Lima scabra Shells only 1 Wholesale supplier 2012 1 Col.1.59 Southern quahog Mecenaria campechiensis Partial Shells 6 Pet-trade 2012 Crown conch (smooth and 1 Col.1.60 Melongena corona Shells only 90 Wholesale supplier 2012 horned forms) 1 Col.1.61 Skate species egg cases Raja spp. Whole specimens 81 Ecological surveys 2001 1 Col.1.62 Pleated sea squirt Styela plicata Whole specimens 11 Markets 1999 Deep sea hexactinellid 1 Col.1.63 Scolymastra spp. Partial specimens 1 Wholesale supplier 2012 “glass” sponge species 1 Col.1.64 Five-holed keyhole urchin Mellita quinquiesperforata Whole specimens 9 Wholesale supplier 2012 1 Col.1.65 Black sun coral Tubastraea micrantha Partial specimens 3 Pet-trade 2012 1 Col.1.66 Cat's paw coral Pocillopora palifera* Partial specimens 14 Pet-trade 2012 1 Col.1.67 Finger coral Acropora humilis* Partial specimens 7 Pet-trade 2012 1 Col.1.68 Plate coral species Acropora spp. Partial specimens 2 Pet-trade 2012 1 Col.1.69 Purple barnacle Conchylepes conchylepes* Partial specimens 5 Pet-trade 2012 1 Col.1.70 Gulf fire coral Millepora alcicornis Partial specimens 3 Pet-trade 2012 1 Col.1.71 Spiral Babylon sea snail Babylonia spirata* Whole specimens 84 Markets 2012 1 Col.1.72 Endive murex sea snail Murex endiva* Shells only 1 Pet-trade 2012 1 Col.1.73 Millipede conch Lambis millepeda** Shells only 6 Wholesale supplier 2012 1 Col.1.74 Horned starfish Protoreaster nodosus** Whole specimens 27 Wholesale supplier 2012 1 Col.1.75 Brown spiny sea star Echinaster spinulosus Whole specimens 49 Wholesale supplier 2012 1 Col.1.76 Inflated sea biscuit Clypeaster rosaceus Whole specimens 2 Wholesale supplier 2012 1 Col.1.77 Flat sea biscuit Clypeaster subdepressus Whole specimens 24 Wholesale supplier 2012 Knobbed whelk egg case 1 Col.1.78 Busycon carica Eggs 159 Ecological surveys 2001 (string) 1 Col.1.79 Knobbed whelk Busycon carica Whole specimens 6 Markets 2012 1 Col.1.80 Channeled whelk Busycotypus canaliculatus Whole specimens 10 Markets 2012 1 Col.1.81 Common or Waved whelk Buccinum undatum* Whole specimens 28 Markets 2012 Goose neck barnacles 1 Col.1.82 species and Shipworm Bankia spp. & Lepas spp. Whole specimens >40 Ecological surveys 2012 species with wood 1 Col.1.83 Giant sea roach Bathynomus giganteous Whole specimens 1 Ecological surveys 2012 1 Col.1.84 Cannonball jellyfish Stomolophus meleagris Whole specimens 2 Ecological surveys 2012 Atlantic horseshoe crab 1 Col.1.85 Limulus polyphemus Eggs 241 Ecological surveys 1999 (eggs) 1 Col.1.86 Colorful sea rod Diodogorgia nodulifera Whole specimens 1 Wholesale supplier 2012 1 Col.1.87 Brown rock sea cucumber Holothuria glaberrima Whole specimens 2 Markets 2001 1 Col.1.88 Furry sea cucumber Astichopus multifudus Whole specimens 2 Markets 1999 1 Col.1.89 Five-toothed sea cucumber Actinopyga agassizii Whole specimens 2 Ecological surveys 2012 1 Col.1.90 Moon jellyfish Aurelia aurito Partial specimens 3 Markets 1999 1 Col.1.91 Black encrusting tunicate Botrylloides nigrum Colonies 4 Markets 2001 1 Col.1.92 Mangrove tunicate Ecteinascidia turbinata Partial colonies 12 Markets 2000 1 Col.1.93 Sea pork species Amaroucium spp Whole specimens 1 Ecological surveys 2012 Primitive deep sea stalked 1 Col.1.94 Litoscalpellum spp. Colonies 1 Ecological surveys 2012 barnacle species 1 Col.1.95 Portuguese man of war Physalia physalis Whole specimens 1 Ecological surveys 2012 1 Col.1.96 Mottled sea hare Aplysia brasiliana Whole specimens 1 Ecological surveys 2012 1 Col.1.97 Atlantic black sea hare Aplysia morio Whole specimens 1 Ecological surveys 2012 1 Col.1.98 Hauff’s alcyonidium Alcyonidium hauffi Partial specimens 1 Ecological surveys 2000 1 Col.1.99 Lettuce bryozoan Thalamoporella gothica Partial specimens 1 Ecological surveys 2000 1 Col.1.100 Gulfweed Sargassum spp. Partial specimens 12 Ecological surveys 2000 Sea lettuce macro-algae 1 Col.1.101 Ulvaria spp. Partial specimens 67 Markets 2012 species 1 Col.1.102 Pincushion urchin species Lytechinus spp. Whole specimens 12 Markets 2012 1 Col.1.103 Spotted linckia starfish Linckia multiflora** Whole specimens 3 Wholesale supplier 2012 1 Col.1.104 Brown sargassum weed Sargassum natans Partial specimens 5 Markets 2000 1 Col.1.105 Paper scallop Euvola papyracea Whole specimens 7 Markets 2012 1 Col.1.106 Gulf mole crabs Emerita brasiliensis Whole specimens 116 Ecological surveys 2001 1 Col.1.107 Thin sea lettuce Ulva lactuca Partial specimens 47 Ecological surveys 2001 1 Col.1.108 Soft spaghettiweed Liagora farinose Partial specimens 4 Ecological surveys 2012

4 Specimen list by trophic level

Level Reference Species Scientific Name Type Quantity Origin Year Florida halemenia red macro- 1 Col.1.109 Halymenia floridana Partial specimens 3 Ecological surveys 2012 algae 1 Col.1.110 Intricate brown macro-algae Rosenvingeo intricate Partial specimens 4 Ecological surveys 2012 1 Col.1.111 Green encrusting tunicate Symplegma viride Whole specimens 54 Ecological surveys 2012 1 Col.1.112 Feather duster worm species Sabellastarte spp. Partial specimens 5 Pet-trade 2004 1 Col.1.113 Line nemerteans worm Lineus spp. Whole specimens 1 Markets 2001 White shrimp with oil staining 1 Col.1.114 with one specimen with Litopenaeus setiferus Whole specimens 2 Anonymous shrimper 2012 assemtrical eyes 1 Col.1.115 Common Shore Shrimp Palaemonetes vulgaris Whole specimens 64 Markets 2001 1 Col.1.116 Ivory barnacles Chelonibia patula Whole specimens 67 Ecological surveys 2001 1 Col.1.117 Northern brown shrimp Farfantepenaeus aztecus Whole specimens 1 Markets 2001 1 Col.1.118 Northern white shrimp Penaeus setiferus Whole specimens 212 Markets 2012 1 Col.1.119 Southern pink shrimp Farfantepenaeus duorarum Whole specimens 5 Ecological surveys 2012 1 Col.1.120 Malaysian prawn Macrobrachium rosenbergii Whole specimens 1 Markets 2001 1 Col.1.121 Pacific giant clam species Tridacna spp.* Shells only 21 Pet-trade 2012 1 Col.1.122 Silver lip conch Strombus lentiginosus* Whole specimens 6 Wholesale supplier 2012 1 Col.1.123 Tulip sea snail Turbonilla curta Whole specimens 5 Wholesale supplier 2012 1 Col.1.124 Asian shore crab Hemigrapsus sanguineus Whole specimens 36 Ecological surveys 2006 Southern and American 1 Col.1.125 Modiolus spp. Whole specimens 56 Ecological surveys 2012 horse mussel species 1 Col.1.126 Variable coquina clams Donax variabilus Whole specimens 58 Ecological surveys 2012 1 Col.1.127 Bruised nassa Nassarius vibex Whole specimens 31 Ecological surveys 2012 1 Col.1.128 Caribbean carpet anemone Stichodactyla helianthus Whole specimens 1 Pet-trade 2012 1 Col.1.129 Curlycue anemone Bartholomea annulata Whole specimens 1 Pet-trade 2012 1 Col.1.130 Cassiopea jellyfish Cotylorhiza tuberculata Whole specimens 1 Pet-trade 2012 Atlantic horseshoe crab 1 Col.1.131 Limulus polyphemus Whole specimens 1 Pet-trade 2012 (early instar) 1 Col.1.132 Basket sea-star Gorgonocephalus eucnemis Whole specimens 1 Pet-trade 2012 Flamescallop or Red file shell 1 Col.1.133 Lima scabra Whole specimens 1 Pet-trade 2012 clam 1 Col.1.134 Brittle Star species Ophioderma spp. Whole specimens 1 Pet-trade 2012 1 Col.1.135 Serpent star species Hemipholis spp. Whole specimens 2 Pet-trade 2012 Sponge spider crab species 1 Col.1.136 Macrocoeloma spp. & Tedania ignis Whole specimens 1 Pet-trade 2012 & Fire sponges 1 Col.1.137 Green reef crab Mithrax sculptus Whole specimens 1 Pet-trade 2012 1 Col.1.138 Ridged slipper lobster Scyllarides nodifer Whole specimens 1 Ecological surveys 2012 Deep-water slipper lobster 1 Col.1.139 Scyllarides spp. Whole specimens 1 Ecological surveys 2012 species 1 Col.1.140 Yellowline arrow crab Stenorhynchus seticornis Whole specimens 1 Ecological surveys 2012 1 Col.1.141 West Indian sea urchins Tripneustes ventricosus Whole specimens 3 Pet-trade 2012 Flat-clawed hermit crab 1 Col.1.142 inside Northern moon snail Pagurus pollicaris inside Lunatia herosWhole specimens 1 Markets 2012 shell Acadian hermit crab inside 1 Col.1.143 Pagurus acadianus inside CymatiumWhole parthenopeum specimens 1 Markets 2012 Neapolitan triton snail shell 1 Col.1.144 Brown spiny sea star Echinaster spinulosus Whole specimens 1 Pet-trade 2012 1 Col.1.145 Mud fiddler crab Uca pugnax Whole specimens 1 Pet-trade 2012 1 Col.1.146 Purple marsh crab Afrithelphusa monodosa Whole specimens 1 Pet-trade 2005 1 Col.1.147 Mangrove land crab Ucides cordatus Whole specimens 1 Pet-trade 2004 Mixed southern gulf 1 Col.1.148 Littorina spp. Shells only 18,584 Wholesale supplier 2012 periwinkle species 1 Col.1.149 Caribbean mud fiddler crab Uca rapax Whole specimens 65 Markets 2012 1 Col.1.150 Sea nettle jellyfish species Chrysaora spp. Whole specimens 3 Ecological surveys 2012 1 Col.1.151 Pink vase sponge Niphates digitalis Whole specimens 1 Retail store 2012 1 Col.1.152 Boring sponge species Cliona spp. Partial specimens 8 Retail store 2012 1 Col.1.153 Red beard sponge Microciona prolifera Partial specimens 1 Retail store 2012 1 Col.1.154 Rope sponge species Aplysina spp. Partial specimens 2 Retail store 2012 White shrimp with lesions 1 Col.1.155 and one specimen with Whole specimens 2 Anonymous shrimper 2012 missing eye "sOil" (marsh sediment mixed 1 Col.1.156 with Macondo oil and Partial specimens 1 Ecological surveys 2012 chemical dispersants)

5 Specimen list by trophic level

Level Reference Species Scientific Name Type Quantity Origin Year 2 Col.2.1 Lady crab Ovalipes ocellatus Whole specimens 12 Markets 2005 2 Col.2.2 Rock crab Cancer irroratus Whole specimens 3 Markets 2012 2 Col.2.3 Green crab Cancer maenas Whole specimens 10 Markets 2005 Rugose or Purple swimming 2 Col.2.4 Callinectes exasperates Whole specimens 1 Markets 2005 crab 2 Col.2.5 Common mantis shrimp Squilla empusa Whole specimens 1 Markets 2005 2 Col.2.6 Hogchoker Trinectes maculatus Whole specimens 2 Markets 2005 2 Col.2.7 Mediterranean sand lance Gymnammodytes cicerelus* Whole specimens 82 Markets 2005/2012 2 Col.2.8 Rimspine searobin Peristedion thompsoni Whole specimens 2 Ecological surveys 2012 2 Col.2.9 Atlantic sturgeon (juvenile) Acipenser oxyrinchus Whole specimens 1 Research institute 2005 Banded coral shrimp or 2 Col.2.10 Stenopus hispidus Whole specimens 1 Pet-trade 2012 Cleaner shrimp-like decapod 2 Col.2.11 Atlantic guitarfish Rhinobatos lentiginosus Whole specimens 1 Ecological surveys 2012 2 Col.2.12 Smooth skate Malacoraja senta Whole specimens 1 Ecological surveys 2012 2 Col.2.13 Clear-nose skate Raja eglanteria Whole specimens 1 Markets 2005 2 Col.2.14 Lesser electric ray Narcine bancroftii Whole specimens 1 Ecological surveys 2012 2 Col.2.15 Red goatfish Mullas auratus Whole specimens 7 Markets 2012 2 Col.2.16 Ratfish Hydrolagus species Whole specimens 1 Wholesale supplier 2014 2 Col.2.17 Blueback herring Alosa aestivalis Whole specimens 13 Markets 2012 2 Col.2.18 Balao halfbeak Hemiramphus balao Whole specimens 1 Markets 2012 2 Col.2.19 Southern eagle ray (juvenile) Myliobatis goodei Whole specimens 1 Ecological surveys 2012 2 Col.2.20 Underworld windowskate Fenestraja plutonia Whole specimens 1 Ecological surveys 2012 2 Col.2.21 Masked pufferfish Arothron diadematus* Whole specimens 1 Pet-trade 2005 2 Col.2.22 Regal angelfish Pygoplites diacanthus** Whole specimens 1 Pet-trade 2005 2 Col.2.23 Six barred angel (juvenile) Pomacanthus sexstriatus** Whole specimens 1 Pet-trade 2005 2 Col.2.24 Sargassum triggerfish Xanthichthys ringens Whole specimens 1 Pet-trade 2005 2 Col.2.25 Gray triggerfish Balistes capriscus Whole specimens 1 Ecological surveys 2012 2 Col.2.26 Queen parrotfish Scarus vetula Whole specimens 1 Pet-trade 2005 2 Col.2.27 Sharpnose puffer Bamthigaster rostrata Whole specimens 1 Pet-trade 2005 2 Col.2.28 Sea lamprey Petromyzon marinus Whole specimens 2 Markets 1999 2 Col.2.29 Thornbacked boxfish Tetrosomus gibbosus** Whole specimens 1 Pet-trade 2005 2 Col.2.30 Violet goby Gobioides broussoneti Whole specimens 15 Markets 2012 2 Col.2.31 White trevally Pseudocaranx dentex Whole specimens 8 Markets 2012 2 Col.2.32 Pygmy filefish Monacanthus setifer Whole specimens 1 Pet-trade 2005 2 Col.2.33 Long-horned cowfish Lactoria cornuta Whole specimens 1 Wholesale supplier 1980 2 Col.2.34 Longsnout seahorse Hippocampus reidi Whole specimens 1 Pet-trade 2005 2 Col.2.35 Northern pipefish Synganathus fuscus* Whole specimens 30 Markets 2012 2 Col.2.36 Chain pipefish Synganathus louisanae Whole specimens 15 Markets 2012 2 Col.2.37 Lined seahorse Hippocampus erectus Whole specimens 2 Markets 1999 2 Col.2.38 Checkered puffer fish Sphoeroides testudineus Whole specimens 40 Ecological surveys 2012 2 Col.2.39 Striped burrfish Chilomycterus schoepfii Whole specimens 2 Pet-trade 1990 2 Col.2.40 Spotlight parrotfish Sparisoma viride Whole specimens 1 Markets 2012 2 Col.2.41 Redband parrotfish Sparisoma aurofrenatum Whole specimens 3 Markets 2012 2 Col.2.42 Porcupinefish Diodon hystrix Whole specimens 1 Pet-trade 1990 2 Col.2.43 Princess parrotfish Scarus taeniopterus (juvenile) Whole specimens 1 Ecological surveys 2011 2 Col.2.44 Opossum pipefish Microphis brachyurus lineatus Whole specimens 8 Markets 2000 2 Col.2.45 Gulf pipefish Syngnathus scovelli Whole specimens 9 Ecological surveys 2011 2 Col.2.46 Sargassum pipefish Syngnathus pelagicus Whole specimens 31 Ecological surveys 2002 2 Col.2.47 Atlantic Moonfish Selene setapinnis Whole specimens 3 Markets 2005 2 Col.2.48 Lookdown Selene vomer Whole specimens 3 Markets 2012 2 Col.2.49 Blue-striped Angelfish Chaetodontoplus septentrionalis** Whole specimens 1 Pet-trade 2003 2 Col.2.50 Pearl wrasse Anampses cuvier** Whole specimens 1 Pet-trade 2004 2 Col.2.51 Yellow tang Zebrasoma flavescens* Whole specimens 2 Pet-trade 2003 2 Col.2.52 Spotfin jawfish species Opistognathus spp. Whole specimens 1 Pet-trade 2004 2 Col.2.53 Striped killifish Fundulus majalis Whole specimens 177 Markets 2000

6 Specimen list by trophic level

Level Reference Species Scientific Name Type Quantity Origin Year 2 Col.2.54 Flathead mullet Mugil cephalus Whole specimens 12 Ecological surveys 2011 2 Col.2.55 Naked goby (female & male) Gobiosoma bosci Whole specimens 2 Ecological surveys 2011 2 Col.2.56 Tidewater silverside Menidia peninsulae Whole specimens 10 Ecological surveys 2011 2 Col.2.57 Golden topminnow Fundulus chrysotus Whole specimens 1 Ecological surveys 2011 2 Col.2.58 Texas cichlid Herichthys cyanoguttatus Whole specimens 1 Ecological surveys 2011 2 Col.2.59 Sailfin molly (female & male) Poecilia latipinna Whole specimens 2 Ecological surveys 2011 2 Col.2.60 Rainwater killifish Lucania parva Whole specimens 5 Ecological surveys 2011 Dwarf livebearer or Least 2 Col.2.61 Heterandria formosa Whole specimens 6 Ecological surveys 2011 killifish 2 Col.2.62 Rough silverside Membras martinica Whole specimens 4 Ecological surveys 2011 2 Col.2.63 Green sunfish Lepomis cyanellus Whole specimens 3 Ecological surveys 2011 2 Col.2.64 High hat (juvenile) Pareques acuminatus Whole specimens 1 Pet-trade 2012 2 Col.2.65 Foureye butterfly fish Chaetodon capistratus Whole specimens 2 Pet-trade 2012 2 Col.2.66 Dwarf seahorse Hippocampus zosterae Whole specimens 2 Wholesale supplier 2012 2 Col.2.67 Pancake batfish Halieutichthys aculeatus Whole specimens 2 Ecological surveys 2012 2 Col.2.68 Wahoo (larvae) Acanthocybium solandri Whole specimens 61 Markets 2000 2 Col.2.69 Gulf killifish Fundulus grandis Whole specimens 8 Ecological surveys 2002 2 Col.2.70 Striped Mullet Mugil cephalus Whole specimens 4 Markets 2012 2 Col.2.71 Inland silverside Menidia beryllina Whole specimens 4 Ecological surveys 2011 2 Col.2.72 Atlantic silverside Menidia menidia Whole specimens 816 Markets 2002 2 Col.2.73 Spanish sardine Sardinella aurita Whole specimens 7 Markets 2012 2 Col.2.74 Dusky pipefish Synganathus floridae Whole specimens 3 Ecological surveys 2001 2 Col.2.75 Gulf menhaden Brevoortia patronus Whole specimens 15 Markets 2012 2 Col.2.76 Fantail mullet Mugil trichodon Whole specimens 5 Markets 2012 2 Col.2.77 Mummichog Fundalulus heteroclitus* Whole specimens 147 Markets 2004 2 Col.2.78 Blackwing flyingfish Hirundichthys rondeletii Whole specimens 3 Markets 2004 2 Col.2.79 American gizzard shad Dorosoma cepedianum Whole specimens 3 Markets 2012 2 Col.2.80 Butterfish Peprilus triacanthus Whole specimens 14 Markets 2012 2 Col.2.81 Harvestfish Peprilus paru Whole specimens 7 Markets 2012 2 Col.2.82 Blackcheek tonguefish Symphurus plagiusa Whole specimens 5 Markets 2012 2 Col.2.83 Blue reef chromis damselfish Chromis cyanea Whole specimens 2 Pet-trade 2012 2 Col.2.84 Creole wrasse Clepticus parrae Whole specimens 1 Pet-trade 2012 2 Col.2.85 Sharksucker Echeneis naucrates Whole specimens 1 Ecological surveys 2012 2 Col.2.86 Atlantic thread herring Opisthonema oglinum Whole specimens 1 Markets 2012 2 Col.2.87 Whitecheek surgeonfish Acanthurus nigricans* Whole specimens 1 Pet-trade 2003 2 Col.2.88 Cunner Tautoglabrus adspersus Whole specimens 5 Markets 2005/2012 2 Col.2.89 Mountain mullet Agonostomus moniticola Whole specimens 29 Markets 2004 2 Col.2.90 Giant or Red hermit crab Petrochirus diogenes Whole specimens 1 Ecological surveys 2012 2 Col.2.91 Rosy lobsterette Nephropsis rosea Whole specimens 1 Ecological surveys 2012 Red swamp crayfish or 2 Col.2.92 Procambarus clarkia Whole specimens 1 Markets 2012 Louisiana crawfish Common Australian yabby 2 Col.2.93 Cherax destructor** Whole specimens 1 Markets 2002 crayfish Longear sunfish or Creek 2 Col.2.94 Lepomis megalotis Whole specimens 4 Markets 1999 perch 2 Col.2.95 Yellowtail coris wrasse Coris gaimard** Whole specimens 1 Pet-trade 2003 2 Col.2.96 Cownose ray Rhinoptera bonasus Partial specimens 1 Ecological surveys 1996 Caribbean spiny lobster 2 Col.2.97 Panulirus argus Whole specimens 1 Ecological surveys 1999 (larvae) 2 Col.2.98 Coral mithrax crab Mithrax coryphe Whole specimens 5 Ecological surveys 1999 2 Col.2.99 Whitetail damselfish Dascyllus aruanus* Whole specimens 1 Pet-trade 2003 2 Col.2.100 Lined porcelain crab Petrolisthes galathinus Whole specimens 1 Pet-trade 2012 Bigclaw pistol or snapping 2 Col.2.101 Alpheus heterochaelis Whole specimens 1 Ecological surveys 2012 shrimp 2 Col.2.102 Mosquitofish Gambusia affinis Whole specimens 70 Ecological surveys 2012 Armored catfish or 2 Col.2.103 Hypostomus ssp. Whole specimens 1 Pet-trade 2012 Plecostomus species 2 Col.2.104 Bay anchovy Anchoa mitchilli Whole specimens 20 Markets 2000 2 Col.2.105 Birdmouth Wrasse Gomphosus caeruleus** Whole specimens 1 Pet-trade 2004

7 Specimen list by trophic level

Level Reference Species Scientific Name Type Quantity Origin Year 3 Col.3.1 Freshwater drum Aplodinotus grunniens Whole specimens 1 Ecological surveys 2011 3 Col.3.2 Permit Trachinotus falcatus Whole specimens 1 Markets 2012 3 Col.3.3 Florida pomano Trachinotus carolinus Whole specimens 2 Markets 2012 3 Col.3.4 Atlantic needlefish Strongylura marina Whole specimens 1 Markets 2005 3 Col.3.5 Red lionfish Pterois volitans Whole specimens 1 Pet-trade 2012 3 Col.3.6 Atlantic blue-clawed crab Callinectes sapidus Whole specimens 9 Markets 2005 3 Col.3.7 Three-toed amphiuma Amphiuma tridactylum Whole specimens 1 Pet-trade 2000 3 Col.3.8 White perch Morone Americana* Whole specimens 2 Markets 2000 Royal dorade or 3 Col.3.9 Mediterranean gilt-head Sparus aurata* Whole specimens 4 Markets 2012 seabream 3 Col.3.10 Rainbow smelt Osmerus mordax* Whole specimens 5 Markets 2012 3 Col.3.11 Ocellated frogfish Fowlerichthys ocellatus Whole specimens 1 Ecological surveys 2012 3 Col.3.12 Winter flounder Pseudopleuronectes americanus* Whole specimens 2 Markets 2005 3 Col.3.13 Summer flounder Paralichthys dentatus Whole specimens 1 Markets 2012 3 Col.3.14 Gulf flounder Paralichthys albigutta Whole specimens 1 Markets 2002 3 Col.3.15 Southern flounder Paralichthys lethostigma Whole specimens 1 Markets 2003 3 Col.3.16 Windowpane flounder Scophthalmus aquosus Whole specimens 1 Markets 2012 3 Col.3.17 Hardhead catfish Ariopsis felis Whole specimens 1 Ecological surveys 2013 3 Col.3.18 Spotted pig grunt Orthopristis chrysoptera Whole specimens 1 Ecological surveys 2012 3 Col.3.19 White grunt Haemulon plumierrii Whole specimens 1 Ecological surveys 2012 3 Col.3.20 Sailor’s choice grunt Haemulon parra Whole specimens 1 Ecological surveys 2012 3 Col.3.21 Puddingwife Halichoeres poeyi Whole specimens 1 Markets 2000 3 Col.3.22 Bigeye Priacanthus arentus Whole specimens 1 Markets 1999 3 Col.3.23 Short bigeye Pristigenys alta Whole specimens 1 Markets 2006 3 Col.3.24 Striped searobin Prionotus evolans Whole specimens 1 Markets 2012 3 Col.3.25 Southern kingfish Mentricirrhus americanus Whole specimens 4 Markets 2012 3 Col.3.26 Northern kingfish Mentricirrhus saxitilis Whole specimens 1 Ecological surveys 2004 3 Col.3.27 Atlantic croaker Micropogonias undulates Whole specimens 1 Markets 2012 3 Col.3.28 Deep-sea lantern fish species Diaphus or Lampanyctus spp. Whole specimens 3 Markets 2012 3 Col.3.29 Pinfish Lagodon rhomboids Whole specimens 2 Ecological surveys 1999 3 Col.3.30 Sea bream Archosargus rhomboidalis Whole specimens 1 Ecological surveys 1999 3 Col.3.31 Yellowfin morjarra Gerres cinereus Whole specimens 1 Ecological surveys 1999 3 Col.3.32 Atlantic saury Scomberesox saurus Whole specimens 4 Markets 2012 3 Col.3.33 Atlantic chub mackerel Scomber colias Whole specimens 2 Markets 2012 3 Col.3.34 Mackerel scad Decapterus macarellus Whole specimens 3 Markets 2012 3 Col.3.35 Bigeye scad Selar crumenophthalmus Whole specimens 3 Markets 2012 3 Col.3.36 Fat sleeper Dormitator maculatus Whole specimens 1 Markets 2012 3 Col.3.37 Tautog or Blackfish Tautoga onitis Whole specimens 1 Markets 2012 3 Col.3.38 Chinese softshell turtle Pelodiscus sinensis** Whole specimens 1 Markets 2012 3 Col.3.39 American eel Anguilla rostrata Whole specimens 3 Markets 2012 3 Col.3.40 Shrimp eel Ophichthus gomesii Whole specimens 1 Ecological surveys 2012 Asian swamp or 3 Col.3.41 Monopterus spp. Whole specimens 3 Markets 2002/2012 Synbranchidae eel species 3 Col.3.42 Common Atlantic octopus Octopus vulgaris Whole specimens 1 Markets 2012 3 Col.3.43 Atlantic Oval Squid Sepioteuthis sepioidea Whole specimens 1 Markets 2000 Gulf deep-water octopus 3 Col.3.44 Benthoctopus spp. Whole specimens 1 Ecological surveys 2012 species 3 Col.3.45 Caribbean reef octopus Octopus briareus Whole specimens 1 Ecological surveys 2012 3 Col.3.46 Brownstripe octopus Octopus burryi Whole specimens 2 Markets 2000 3 Col.3.47 Caribbean reef scorpion fish Scorpaenodes caribbaeus Whole specimens 1 Pet-trade 2012 3 Col.3.48 Chestnut Moray eel Enchelycore carychroa Whole specimens 1 Pet-trade 2012 3 Col.3.49 Spanish grunt Haemulon macrostomum Whole specimens 1 Ecological surveys 2012 3 Col.3.50 Northern sennet Sphyraena boreali Whole specimens 1 Ecological surveys 1999 3 Col.3.51 Deepreef scorpionfish Scorpaenodes tredecimspinosus Whole specimens 12 Pet-trade 2012 Bonnethead shark or 3 Col.3.52 Sphyrna tiburo Whole specimens 1 Ecological surveys 2012 shovelhead 3 Col.3.53 Blackbelly rosefish Helicolenus dactylopterus Whole specimens 3 Ecological surveys 2012 3 Col.3.54 Butter hamlet Hypoplectrus unicolor Whole specimens 1 Pet-trade 2004 3 Col.3.55 Clown leaflip soapfish Pogonoperca punctate** Whole specimens 1 Pet-trade 2004 3 Col.3.56 Longfin squid Loligo pealei Whole specimens 1 Markets 2012

8 Specimen list by trophic level

Level Reference Species Scientific Name Type Quantity Origin Year 4 Col.4.1 St Pierre’s fish or John Dory Zeus faber Whole specimens 1 Markets 2012 4 Col.4.2 Guachanche barracuda Sphyraena guachancho Whole specimens 4 Markets 2000 4 Col.4.3 Yellow-tail snapper Ocyurus chrysurus Whole specimens 1 Markets 2012 4 Col.4.4 Cottonwick Haemulor melanurum Whole specimens 3 Markets 2000 4 Col.4.5 Conger eel Conger oceanicus Whole specimens 2 Markets 2000 4 Col.4.6 Gulf toadfish Opsanus beta Whole specimens 1 Ecological surveys 2012 4 Col.4.7 Gray snapper Lutjanus griseus Whole specimens 2 Markets 2000 4 Col.4.8 Red snapper Lutjanus campechanus Whole specimens 2 Markets 2000 4 Col.4.9 Freckled stargazer Gnathagnus egreglus Whole specimens 1 Ecological surveys 2012 4 Col.4.10 Vermilion snapper Rhomboplites aurorubens Whole specimens 3 Markets 2012 4 Col.4.11 Barmundi or Asian seabass Lates calcarifer** Whole specimens 1 Markets 2012 4 Col.4.12 Oyster toadfish Opsanus tau Whole specimens 1 Markets 2012 4 Col.4.13 Bearded brotula Brotula barbata Whole specimens 1 Ecological surveys 2012 Coney Grouper or Leopard 4 Col.4.14 Cephalopholis fulva Whole specimens 1 Markets 2012 Hind 4 Col.4.15 Spinycheek scorpionfish Neomerinthe hemingway Whole specimens 3 Ecological surveys 2012 4 Col.4.16 Inshore lizardfish Synodus foetens Whole specimens 1 Markets 2000 Deep-sea dagger tooth or 4 Col.4.17 Anotopterus pharao Whole specimens 3 Ecological surveys 2012 Pharaoh fish 4 Col.4.18 Pilot fish Naucrates doctor Whole specimens 3 Markets 2012 4 Col.4.19 Black sea bass Centropristis striata Whole specimens 2 Markets 2012 Chain catshark or Chain 4 Col.4.20 Scyliorhinus rotifer Whole specimens 1 Ecological surveys 2012 dogfish 4 Col.4.21 Caribbean ocellated moray Gymnothorax ocellatus Whole specimens 1 Pet-trade 2012 4 Col.4.22 Hourglass moray eel Muraena clepsydra* Whole specimens 1 Pet-trade 2004 4 Col.4.23 Bluefish Pomatomus saltatrix Whole specimens 6 Markets 1999 4 Col.4.24 Great northern tilefish Lopholatilus chamaeleonticeps Whole specimens 1 Markets 2012 4 Col.4.25 Wrenchman Pristipomoides aquilonaris Whole specimens 1 Markets 2000 4 Col.4.26 Deep-sea squid species llex spp. Whole specimens 1 Ecological surveys 2012 4 Col.4.27 Mutton hamlet Alphestes afer Whole specimens 1 Markets 2000 4 Col.4.28 Graysby Cephalopholis cruentata Whole specimens 1 Markets 2000 4 Col.4.29 Red hind grouper Epinephelus guttatus Whole specimens 1 Markets 2012 4 Col.4.30 Redfish species Sebastes spp. Whole specimens 4 Markets 2012 Deep-water needlefish or 5 Col.5.1 Tylosurus crocodilus Whole specimens 1 Ecological surveys 2012 Houndfish 5 Col.5.2 Blue runner Caranx crysos Whole specimens 3 Markets 2012 5 Col.5.3 Crevalle jack Caranx hippos Whole specimens 2 Markets 2012 5 Col.5.4 Blackfin goosefish Lophius gastrophysus Whole specimens 1 Markets 2012 Atlantic cutlassfish or 5 Col.5.5 Trichiurus lepturus Whole specimens 3 Markets 2012 Ribbonfish 5 Col.5.6 Burmese python Python molurus bivittatus Whole specimens 1 Pet-trade 2012 5 Col.5.7 Greater amber jack Seriola dumerili Whole specimens 1 Markets 2012 Dusky smooth-hound or 5 Col.5.8 Mustelus canis Whole specimens 1 Markets 2000 Smooth dogfish shark 5 Col.5.9 Atlantic Spanish mackerel Scomberomorus maculatus Whole specimens 1 Markets 2012 5 Col.5.10 Yellowfin grouper Mycteroperca venenosa Whole specimens 1 Markets 2012 5 Col.5.11 Yellowmouth grouper Mycteroperca interstitialis Whole specimens 1 Markets 2000 5 Col.5.12 Red grouper Epinephelus morio Whole specimens 1 Markets 2012 5 Col.5.13 Atlantic striped bass Morone saxatilis Whole specimens 2 Markets 2000 6 Col.6.1 Blacktip shark (juvenile) Carcharhinus limbatus Whole specimens 1 Ecological surveys 2012 6 Col.6.2 Mako shark species Isurus spp. Partial specimens 1 Ecological surveys Historic 6 Col.6.3 Atlantic wreckfish Polyprion americanus Partial specimens 1 Markets 2012 6 Col.6.4 Wahoo Acanthocybium solandri Partial specimens 1 Markets 2012 Mahi-mahi or Common 6 Col.6.5 Coryphaena hippurus Partial specimens 1 Markets 2012 dolphinfish TOTAL 26,160

* non-native species ** non-native species anecdotally reported but not scientifically confirmed or species with a high-probability of introduction from the pet-trade or seafood industry

9 Vol. 63: 101–109, 2011 AQUATIC MICROBIAL ECOLOGY Published online March 31 doi: 10.3354/ame01482 Aquat Microb Ecol

OPENPEN FEATURE ARTICLE ACCESSCCESS Effects of COREXIT® EC9500A on bacteria from a beach oiled by the Deepwater Horizon spill

Leila J. Hamdan1,*, Preston A. Fulmer2

1Marine Biogeochemistry Section, Code 6114, U.S. Naval Research Laboratory, Overlook Ave. SW, Washington, DC 20375, USA 2Bioenergy and Biofabrication Section, Code 6115, U.S. Naval Research Laboratory, Overlook Ave. SW, Washington, DC 20375, USA

ABSTRACT: Hydrocarbon-degrading bacteria are important for controlling the fate of natural and anthropogenic hydrocarbons in the marine environment. In the wake of the Deepwater Horizon spill in the Gulf of Mexico, microbial communities will be important for the natural attenuation of the effects of the spill. The chemical dispersant COREXIT® EC9500A was widely deployed during the response to the Deepwater Horizon incident. Although toxicity tests confirm that COREXIT® EC9500A does not pose a significant threat to invertebrate and adult fish populations, there is limited information on its effect on microbial communities. We determined the composition of the microbial community in oil that had been Oil from the Deepwater Horizon spill washing ashore on freshly deposited on a beach in Louisiana, USA, as a Elmer's Island, Louisiana, USA, in May 2010. Inset: viability result of the Deepwater Horizon spill. The metabolic assay for the hydrocarbon-degrading bacterium Acinetobac- activity and viability in cultures obtained from oil ter venetianus before (left) and after (right) exposure to COREXIT, revealing inhibition by the chemical dispersant. samples were determined in the absence and presence of COREXIT® EC9500A at concentrations rang- Photo: Warren Wood; inset: Leila J. Hamdan ing from 0.001 to 100 mg ml–1. In length heterogeneity PCR (LH-PCR) fingerprints of oil samples, the most abundant isolates were those of Vibrio, followed by KEY WORDS: Deepwater Horizon · Gulf of Mexico · hydrocarbon-degrading isolates affiliated with Acine- Dispersant · Hydrocarbon degraders · Vibrio · to bac ter and Marinobacter. We observed significant COREXIT · Toxicity reductions in production and viability of Acineto - Resale or republication not permitted without bacter and Marinobacter in the presence of the dis- written consent of the publisher persant compared to controls. Of the organisms examined, Marinobacter appears to be the most sensi- INTRODUCTION tive to the dispersant, with nearly 100% reduction in viability and production as a result of exposure to The mobile offshore drilling unit ‘Deepwater Hori- concentrations of the dispersant likely to be encoun- zon’ experienced an explosion on April 20, 2010, and tered during the response to the spill (1 to 10 mg sank 2 d later (Coastal Response Research Center ml–1). Significantly, at the same concen tration of dis- 2010). As a result of the explosion, and failure of a persant, the non-hydrocarbon-degrading Vibrio isolates proliferated. These data suggest that hydro- blowout preventer on the sea floor, crude oil began carbon-degrading bacteria are inhibited by chemical leaking from a broken riser pipe into the Gulf of Mexico dispersants, and that the use of dispersants has the at an estimated rate of 19 000 to 70 000 barrels (~3000 to potential to diminish the capacity of the environment 11 000 m3) of oil d–1 (Labson et al. 2010). This resulted in to bioremediate spills. the largest oil spill in the coastal waters of the USA.

*Email: [email protected] © Inter-Research 2011 · www.int-res.com Form Approved Report Documentation Page OMB No. 0704-0188

Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number.

1. REPORT DATE 3. DATES COVERED MAR 2011 2. REPORT TYPE 00-00-2011 to 00-00-2011

4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER Effects of COREXIT EC9500A on bacteria from a beach oiled by the 5b. GRANT NUMBER Deepwater Horizon spill 5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S) 5d. PROJECT NUMBER

5e. TASK NUMBER

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION U.S. Naval Research Laboratory,Marine Biogeochemistry Section, Code REPORT NUMBER 6114,Washington,DC,20375

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)

11. SPONSOR/MONITOR’S REPORT NUMBER(S)

12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution unlimited

13. SUPPLEMENTARY NOTES

14. ABSTRACT Hydrocarbon-degrading bacteria are important for controlling the fate of natural and anthropogenic hydrocarbons in the marine environment. In the wake of the Deepwater Horizon spill in the Gulf of Mexico, microbial communities will be important for the natural attenuation of the effects of the spill. The chemical dispersant COREXIT? EC9500A was widely deployed during the response to the Deepwater Horizon incident. Although toxicity tests confirm that COREXIT? EC9500A does not pose a significant threat to invertebrate and adult fish populations there is limited information on its effect on microbial communities. We determined the composition of the microbial community in oil that had been freshly deposited on a beach in Louisiana, USA, as a result of the Deepwater Horizon spill. The metabolic activity and viability in cultures obtained from oil samples were determined in the absence and presence of COREXIT? EC9500A at concentrations ranging from 0.001 to 100 mg ml?1. In length heterogeneity PCR (LH-PCR) fingerprints of oil samples, the most abundant isolates were those of Vibrio, followed by hydrocarbon-degrading isolates affiliated with Acine - to bac ter and Marinobacter. We observed significant reductions in production and viability of Acineto - bacter and Marinobacter in the presence of the dis - persant compared to controls. Of the organisms examined, Marinobacter appears to be the most sensitive to the dispersant, with nearly 100% reduction in viability and production as a result of exposure to concentrations of the dispersant likely to be encountered during the response to the spill (1 to 10 mg ml?1). Significantly, at the same concen tration of dispersant the non-hydrocarbon-degrading Vibrio isolates proliferated. These data suggest that hydro - carbon-degrading bacteria are inhibited by chemical dispersants, and that the use of dispersants has the potential to diminish the capacity of the environment to bioremediate spills.

15. SUBJECT TERMS 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF ABSTRACT OF PAGES RESPONSIBLE PERSON a. REPORT b. ABSTRACT c. THIS PAGE Same as 9 unclassified unclassified unclassified Report (SAR)

Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18 102 Aquat Microb Ecol 63: 101–109, 2011

Natural seepage of hydrocarbons is an important 2010). Furthermore, the dispersants themselves may source of carbon for benthic environments in the Gulf impact hydrocarbon-degrading microorganisms. of Mexico and is a structuring factor for benthic and Thus, the goal of this study was to understand the near-bottom microbial communities (National Re - impact of COREXIT® EC9500A (henceforth referred search Council Committee on Oil in the Sea 2003). In to as ‘COREXIT’) on bacterial viability and metabolic marine environments, bacteria are the predominant function in oiled beach samples from the area affected degraders of hydrocarbons (Leahy & Colwell 1990) and by the Deepwater Horizon spill. These samples may be thus are of great importance in controlling the fate of ideal candidates for such analysis because microbial natural and anthropogenic seepage. In such environ- communities have been acclimatized to conditions of ments, microbial communities contain members pre- high contamination by hydrocarbons, environmental disposed to hydrocarbon metabolism (Aharon & Fu weathering and possibly in situ exposure to disper- 2000, Lanoil et al. 2001). Microbial assemblages associ- sants. Future studies of the impact of dispersants on ated with hydrocarbon seeps are well characterized in microbial communities should also include analysis of Gulf of Mexico sediments (Hollaway et al. 1980, ‘pristine’ environmental communities. Studies of both Aharon & Fu 2000, Lanoil et al. 2001, Mills et al. 2003, exposed and pristine communities will be of impor- Joye et al. 2004, Reed et al. 2006, Lloyd et al. 2010, tance in understanding the long-term affects of the Orcutt et al. 2010). The impact of natural seepage on spill and the use of chemical dispersants in aquatic water-column communities has been relatively less environments. Previous studies have documented studied, although microorganisms with the ability to clear evidence of the deleterious effect of the disper- degrade hydrocarbons are ubiquitously distributed in sant on wildlife and microbial communities (Lindstrom waters along continental shelves (Venkateswaran et al. & Braddock 2002, Couillard et al. 2005, Jung et al. 1991). Because large-scale impacts of hydrocarbons on 2009). Due to the unknown effects of the large-scale Gulf of Mexico beaches are rare, little is known about use of dispersant on microbial communities in affected the microbial communities which live in proximity to areas, more study is warranted on the effects of the beached oil. widespread application of dispersant. In the USA, the Environmental Protection Agency (EPA) maintains a list of chemicals and spill-mitigating devices which may be deployed during an oil spill in MATERIALS AND METHODS coastal waters of the USA; this is a part of the National Contingency Plan (NCP) (US EPA 2010b). NCP- Sample collection and handling. Beached oil sam- approved dispersants are applied to break up masses ples were collected on May 22, 2010, from the south of oil and reduce the formation of surface slicks. This is end of the Elmer’s Island Wildlife Refuge (EIWR), done to reduce the incidence of oil coating on popula- Louisiana, USA, on the seaward facing shore. Personal tions of birds, mammals and invertebrates. In the wake accounts from EIWR staff indicate that accumulations of of the Deepwater Horizon spill, dispersants were oil reached the beach mid-day on the previous day. deployed widely in surface and sub-surface waters During sampling, small brown pea-sized floating drop- (Coastal Response Research Center 2010). As of early lets of oil were observed in the water within 0.5 m of the June 2010, more than 990 000 gallons (~3700 m3) of high-water mark, and pools of thick oil were deposited dispersant were used in the response. The most at the high-water mark along the beach. Oil on the commonly used dispersant was COREXIT® EC9500A beach appeared in several different forms, related to (Nalco). There is limited knowledge regarding the how long it had been deposited (EIWR staff personal communities that metabolize oil in the water column account). Oil that had been on the beach longer than and on coastal beaches. Even less is known about the 24 h had soaked into the sand. Fresher oil rested on the impact of dispersants in general—and specifically surface of the sand. Freshly deposited oil was collected COREXIT® EC9500A—on hydrocarbon-degrading in cellulose acetate butyrate core liners, capped at both microbial communities in the Gulf of Mexico. ends, and later sub-sampled into sterile 50 ml poly- Dispersants have been suggested as a means to ethylene tubes. Samples were maintained at ambient improve microbial biodegradation of oil contamination temperature during transport from the field (24 h) and in the water column by forming small oil droplets with subsequently held in the dark on ice during shipping to high surface-to-volume ratios which increase their the laboratory (48 h). The total time between sample lability to microplankton (Brakstad 2008). However, collection and analysis was ~72 h. Because of these this same process is likely to increase the concentration hold times, alterations in the structure of the microbial and lability of spill-related toxic compounds in the community may have occurred; this may have resulted water column, which may, in turn, affect the microbial in the enhancement of some phylotypes, masking of the hydrocarbon-degrading community (Zahed et al. appearance of phylotypes accounting for less than 1% Hamdan & Fulmer: Effects of COREXIT® EC9500A on bacteria 103

of the population in LH-PCR and culture analysis, or Length heterogeneity polymerase chain reaction loss of cultivability of others. (LH-PCR). Genomic DNA was extracted from ~500 mg of Bacterial abundance. A modification of the method oil and oil-water mixtures using the Bio 101 FastDNA® of Hobbie et al. (1977), described in Hamdan & Jonas SPIN kit for soil. DNA was quantitated on a 1% agarose (2006), was used to determine bacterial abundance. Oil gel with ethidium bromide and diluted with DEPC- and oil-water (liquid surrounding the oil) was diluted treated water such that ~10 ng of DNA was used as tem- 1:100 with buffered sterile seawater, stained with plate for LH-PCR. Environmental samples (oil and oil- 0.1%, acridine orange, collected on black polycarbon- water) and isolates were analyzed by LH-PCR. ate filters of pore size 0.2 µm (Osmonics) and observed Hamdan et al. (2008) provides a detailed description at 1000× magnification. of LH-PCR. Briefly, amplification of variable regions Cultivation conditions. Marine agar (MA) (Difco) V1 and V2 of the small subunit rRNA gene was per- was used as a complete medium for bacterial growth. formed using the primers 6-FAM-27F (5’-6-FAM-AGA Bushnell-Haas (BH) agar (Difco) supplemented with GTT TGA TCM TGG CTC AG-3’) and 355R (5’-GCT- 1% w/v hexadecane (Fisher Scientific) was used as a GCC TCC CGT AGG AGT-3’). Controls accompanied selective medium to isolate hydrocarbon degraders. reactions to determine PCR efficiency and calibrate All media were prepared according to the manufac- peak intensity. PCR mixtures consisted of 1× Gold turer’s specifications. Culturable bacteria were as ses - buffer, 2.5 mM MgCl2, 0.2 mM dNTPs, 0.5 U AmpliTaq sed using standard microbial culture techniques. Gold, 0.5 µM primers, 0.01% BSA, and diethylpyrocar- Approximately 10 µl of oil or oil-water was plated on bonate (DEPC)-treated water. PCR was performed on a MA and BH using a quadrant streak method. Plates GeneAmp System 9700 (Applied Biosystems) with the were incubated at 30°C. Colonies were isolated based following program: 20 to 30 cycles of 95°C (30 s), 48°C on morphology and re-plated to produce pure cultures. (30 s), 72°C (2 min plus 5 s per cycle) and final exten- Isolation of genomic DNA and 16S rDNA sequenc- sion at 72°C (30 min). PCR product was diluted, mixed ing. To identify members of the microbial community, with ILS-600 (Promega) and HiDi formamide and individual colonies from MA and BH plates were loaded on an ABI 3130xl Analyzer (Applied Biosys- added to 5 ml of marine broth (MB) and incubated for tems). LH-PCR was performed on the isolates, and the 18 h at 30°C. Genomic DNA was isolated by alkaline resulting amplicons were virtually aligned with LH- lysis. For this, 2 ml of MB culture was centrifuged at PCR amplicons from the environmental samples so 4000 × g. The medium was removed and cells were that the latter could be attributed to one or more iso- resuspended in 1 ml sterile phosphate-buffered saline lates. In this manner, the contribution of isolates to the (PBS). Cells were pelleted again at 4000 × g and resus- environmental community was determined. LH-PCR pended in 50 mM Tris-Cl (pH 8.0), 10 mM EDTA, and was also performed using universal archaeal primers. 1 mg l–1 (1 µg ml–1) RNase A. Cells were then lysed in No product was obtained with archaeal primers 1HKF a buffer containing 200 mM NaOH and 1% sodium and 589R (see Litchfield et al. 2005 for details); thus, dodecyl sulfate (SDS), and the solution was neutralized we conclude that most of the environmental commu- with 3.0 M potassium acetate. Lysates were cen- nity belongs to the bacterial domain. trifuged at 15 000 × g to remove cell debris, and super- Bacterial viability. The effect of COREXIT (Nalco, natants were collected. DNA was precipitated with iso- batch # SLOE 1184) on bacterial growth was determined propyl alcohol and centrifuged at 15 000 × g, washed using a BacLight™ (Invitrogen) Live/Dead Bacterial with ethanol, and dried at room temperature prior to Viability assay. COREXIT is a mixture of light petroleum resuspension in 10 mM Tris-Cl (pH 8.5). DNA (1 µg) distillates (10 to 30%), propylene glycol (1 to 5%) bu- was used as a template for a PCR as follows: 1 U Fail- tanedioic acid, 2-sulfo-1,4-bis(2-ethylhexyl) ester, safe Enzyme mix (Epicentre Biotechnologies), 2× Fail- sodium salt (10 to 30%), in addition to unspecified safe Premix E, 2 mM 27F 16s rDNA universal bacterial amounts of propanol and sorbitan (www.nalco. com, primer (5’-AGA GTT TGA TCC TGG CTC AG-3’), and www.epa.gov). Assays were conducted according to the 2 mM 1492R 16s rDNA universal bacterial primer (5’- manufacturer’s instructions, including standardization ACG GCT AGC TTG TTA CGA CTT-3’). Reactions curves for each strain. To determine the toxicity of were run as follows: 94°C for 5 min; 20 cycles of 94°C COREXIT for isolates, 10-fold serial dilutions of for 30 s, 50°C for 30 s, and 72°C for 90 s; 72°C for 7 min. COREXIT, ranging from 1:10 to 1:100 000, in MB were PCR product was visualized on a 1% agarose gel con- added to wells of a 96-well microtitre plate. Approxi- taining ethidium bromide. Bands of ~1500 bp size were mately 106 colony-forming units (cfu) from a mid-log excised and purified using a gel purification kit (Qia- phase culture of each isolate (in triplicate) were added to gen). Excised PCR products were sequenced by Gene- the wells and incubated at 30°C for 18 h. Cells were pel- wiz and checked for homology to known sequences in leted at 3000 × g and resuspended in sterile 0.85% NaCl, GenBank using the BLASTn algorithm. followed by staining using the BacLight™ kit. Bacterial 104 Aquat Microb Ecol 63: 101–109, 2011

viability was determined using a FLx800 (BioTek) micro - abundance on the oil sample were not possible even plate reader. when dispersed, as the sample matrix clogged filters, To determine the effect of COREXIT on hydrocarbon making accurate measurements impossible. Het- utilization by isolates with known hydrocarbon- erotrophic secondary production in the oil-water was degrading capabilities, we used Bushnell-Haas (BH) 2.7 × 109 cells l–1 d–1, indicating a high standing stock of agar supplemented with 1% w/v hexadecane. cells in the stationary phase of growth. COREXIT was added to the medium–hexadecane mix Eight isolates were identified from the environmen- at ratios of 1:10, 1:25 and 1:50 COREXIT:hexadecane. tal samples (oil and oil-water) (Table 1). Most isolates Cultures were assayed as above. were members of the class Gammaproteobacteria. Heterotrophic bacterial production. Production was With the exception of the isolates related to Exiguo- measured as leucine incorporation (Smith & Azam bacterium arabatum and Acinetobacter venetianus, all 1992). Briefly, 0.50 ml aliquots of oil, oil-water and MB matched most closely (>96%) with environmental cultures were added to microcentrifuge tubes (3 exper- clones obtained from marine or coastal waters. Among imental and 1 abiotic control) that contained [3H-4,5]- the isolates were 3 known hydrocarbon degraders L-leucine (154 mCi mmol–1) and incubated for 1 h at (Table 1). A total of 15 LH-PCR amplicons were ob - 25°C. Incubations were terminated by the addition of served in the samples (Table 1). The Simpson’s Index 100% trichloroacetic acid (TCA). Samples were cen- (D) was used to estimate bacterial diversity. Diversity trifuged to pellet cells and washed with 5% TCA and was relatively low in the oil and oil-water samples ethanol to remove unincorporated radiolabel. Radioac- (0.15 and 0.20, respectively), and 4 amplicons ac - tivity was determined on a Beckman-Coulter LS6500 counted for the majority of the LH-PCR peak area. The liquid scintillation counter. MB cultures of isolates con- 4 peaks that accounted for the majority of peak abun- taining no COREXIT and containing 1 of 2 dilutions of dance corresponded with 3 isolates and 1 unidentified COREXIT (final conc. ~1 and 10 mg ml–1 in MB) were amplicon (Table 1). Vibrio sp. was highly abundant in used. These were held for 48 h and sampled at t = 0, 6, both samples and, alone, accounted for ≥31% of the 12, 24 and 48 h. peak area in both samples. Experiments to assess the acute toxicity of COREXIT for environmental isolates are summarized in Fig. 1. RESULTS The addition of COREXIT resulted in varying levels of reduction in cell viability, and an increase in cell num- Cell abundance in the oil-water fraction of the bers in some cases. A reduction in live cells was seen sample was 2.9 × 1011 cells l–1. Measurements of cell for all isolates at all concentrations tested with the

Table 1. Summary of isolates obtained from oil and oil-water samples and LH-PCR (length heterogeneity PCR) analysis of samples and isolates

Isolate Phylogenetic group Nearest Alignment Amplicon Hydrocarbon Cumulative peak relative (%) length degrader abundance (%) (GenBank (bp) Oil-water Oil Accession no.)

Acinetobacter venetianus Gammaproteobacteria AM909651 99 338 Yes 4 4 Exiguobacterium arabatum Bacillales FM203124 99 359 No <1 0 Marinobacter Gammaproteobacteria GQ901055 100 341 Yes 12 12 hydrocarbonoclasticus Pseudidiomarina sp. Gammaproteobacteria EF212001 96 357 No <1 <1 Pseudoalteromonas sp. Gammaproteobacteria GQ245921 99 344 No 17 13 Pseudomonas Gammaproteobacteria EU440977 99 336 Yes <1 <1 pseudoalcaligenes Shewanella algae Gammaproteobacteria DQ386137 98 346 No 5 10 Vibrio sp. Gammaproteobacteria EU834012 99 353/354 Yes 31 32 Unmatched amplicon 302 Unknown 0 2 Unmatched amplicon 310 Unknown 5 3 Unmatched amplicon 312 Unknown 9 16 Unmatched amplicon 320 Unknown 1 0 Unmatched amplicon 322 Unknown 10 3 Unmatched amplicon 327 Unknown 2 2 Unmatched amplicon 329 Unknown 1 3 Hamdan & Fulmer: Effects of COREXIT® EC9500A on bacteria 105

Fig. 1. Toxicity of COREXIT® EC9500A to isolated strains. Dispersant dilutions are 1:10 to 1:100 000 w/v. Values in parentheses are the concentration of dispersant in each treatment. Percentages of live cells in dispersant-amended cultures are relative to the controls. Error bars are ±1 SD

exception of Vibrio sp. Dilutions of COREXIT of 1:10 statistically significant differences between treatments and 1:100 w/v in Marine broth resulted in near total and controls. In the Acinetobacter venetianus and cell death for all isolates (≥99%). However, at the Marinobacter hydrocarbonoclasticus cultures, addition 1:1000 dilution, corresponding to the addition of of the dispersant moderately stimulated production 0.964 mg ml–1 COREXIT, there was an increase in live compared to the controls. In the Vibrio sp. culture, a cells of Vibrio sp. compared to the control, and a decrease in live cells (40 to 90%) in all other cultures (relative to controls). To address the effects of COREXIT on hydrocarbon degraders, dilutions of COREXIT were added to minimal medium supplemented with hexadecane as a carbon source. COREXIT was added at the concentrations suggested by the Environmental Protection Agency (EPA), i.e. ratios of 1:10, 1:25 and 1:50 COREXIT:hexadecane (US EPA 1995); the results are summarized in Fig. 2. Such ratios of COREXIT to hydrocarbons may have been encountered during surface and deepwater application in the response to the spill, but dispersant concentrations may have been significantly lower within the water column and on affected beaches. Thus, we suggest that these concentrations represent the maximum encountered in the environment. As before, the addition of COREXIT was toxic to all hydrocarbon degraders in a dose-dependent manner. Fig. 2. Toxicity of COREXIT® EC9500A to hydrocarbon A second experiment was conducted to determine degraders when present at concentrations suggested by the Environmental Protection Agency (EPA). Dispersant dilutions the impact of COREXIT on heterotrophic secondary are 1:10, 1:25 and 1:50 ratios of COREXIT to hexadecane. production. The results of this experiment are summa- Values in parentheses are the concentration of dispersant in rized in Fig. 3. At t = 0 h in all 3 cultures, there were each treatment. Error bars are ±1 SD 106 Aquat Microb Ecol 63: 101–109, 2011

8x109 hydrocarbonoclasticus remained similar in all treat- A Control + COREXIT® EC9500A (1:1000) ments until t = 12 h. After t = 12 h, a biofilm appeared + COREXIT® EC9500A (1:100) 6x109 slightly above the medium in treatments containing COREXIT, and production declined significantly, while in the control it remained consistent for the dura- 4x109 tion of the experiment. In Vibrio sp., production in all treatments declined during the first 12 h of the experi- 2x109 ment. After t = 12 h, a concentration-dependent decline in production was observed in the dispersant treatments while the control rebounded. A biofilm was 0 also observed in COREXIT-amended Vibrio sp. tubes )

–1 after t = 12 h.

d 0 1020304050 –1 8x1010 B DISCUSSION

6x1010 LH-PCR revealed a relatively low-diversity population. This is not surprising given the nature of the samples and that others document low diversity in oil-rich 4x1010 environments (Orcutt et al. 2010). Experiments to determine microbial responses to oil contamination using pristine marine sediments demonstrate rapid 10 2x10 increases in the abundance of hydrocarbon degraders (Brakstad 2008). Thus, we hypothesized that the environmental samples would be enriched in microorgan- 0 0 1020304050isms capable of hydrocarbon degradation. Over 80 bacterial genera have been confirmed to degrade hydrocarbons (Head et al. 2006, Brakstad Heterotrophic secondary production (cells l C 2008). The most commonly observed hydrocarbon 10 3x10 degraders in aquatic environments belong to the Gam- ma proteobacteria and include the genera Pseudo - monas, Acinetobacter, Marinobacter and Alcanivorax 10 2x10 (Atlas 1981, Venkateswaran et al. 1991, Head et al. 2006). Fewer reports indicate that Vibrio spp. are involved in hydrocarbon degradation (Atlas 1981, 1x1010 Venkateswaran et al. 1991). Virtual alignment of LH-PCR amplicons with isolate amplicons reveals that a total of 17% of peak abun- 0 dance was associated with 3 known hydrocarbon de- 0 1020304050graders (Table 1). The most abundant of these was Time (h) Marinobacter hydrocarbonoclasticus. The M. hydrocar- Fig. 3. Time course experiment tracking heterotrophic bacter- bonoclasticus isolate exhibited 100% homology with an ial production in 3 cultures: (A) Acinetobacter venetianus, isolate obtained from coral tissue, infected with Black (B) Marinobacter hydrocarbonoclasticus, and (C) Vibrio sp. Band Disease, obtained proximate to the Gulf of Mex- No sample for t = 48 h for A. venetianus + dispersant at 1:100 was available due to evaporative loss. Error bars are ±1 SD ico (Richardson et al. 2009). M. hydrocarbonoclasticus can degrade a variety of aliphatic and aromatic hydrocarbons and produce a nondialyzable bioemulsifier moderate decline in production was observed at t = when grown on hydrocarbons (Gauthier et al. 1992). 0 h. At t = 12 h COREXIT clearly inhibited A. vene- One of the main features that distinguish the genus tianus secondary production compared to the controls; Marinobacter from closely related genera is the ability such inhibition remained significant in the highest of these bacteria to tolerate high levels of salt and to dilution of COREXIT until t = 24 h. Interestingly, in the grow at temperatures up to 45°C. These factors suggest 1:1000 dispersant dilution, production of A. venetianus that the location in which the sample was found—and re bounded to the level of the control at t = 24 h and high ambient temperatures on the beach at the time of exceeded the control at t = 48 h. Production by M. collection—may have selected for this isolate. Hamdan & Fulmer: Effects of COREXIT® EC9500A on bacteria 107

The second most abundant hydrocarbon degrader in to some adult test organisms, dispersants may be the environmental fingerprint was related 99% to highly toxic to communities directly involved in natural Acinetobacter venetianus. Other studies have demon- hydrocarbon bioremediation. strated that A. venetianus proliferates in oil-degrading In the second live:dead experiment, in which hexa- consortia and is capable of metabolizing complex decane was added to growth media (Fig. 2), we con- marine hydrocarbon mixtures (Vaneechoutte et al. firm that cell death results from the application of the 2009). Some strains of A. venetianus produce bioemul- dispersant, not from limitation of hydrocarbons. Inter- sifiers to aid in hydrocarbon metabolism. Pseudomonas estingly, in the presence of hexadecane, Pseudomonas pseudoalcaligenes, a confirmed naphthalene degrader pseudoalcaligenes exhibited the greatest resistance to (Garcia-Valdes et al. 1988), was detected in the LH- COREXIT toxicity, suggesting that the presence of PCR profile, although its amplicon accounted for less hydrocarbons may reduce the impact of COREXIT on than 1% of peak abundance. some populations. Vibrio sp. accounted for the majority of LH-PCR peak In microcosm studies conducted on Alaskan North abundance. The Vibrio isolate was 99% related to 4 Slope crude oil treated with COREXIT® EC9500A, the Vibrio phylotypes: Vibrio natriegens, alginolyticus, fluvi- dispersant did increase the surface area of crude oil alis and the pathogenic vulnificus. All GenBank entries droplets and enhanced microbial droplet colonization which matched the Vibrio sp. isolate are capable of rapid (Lindstrom & Braddock 2002). However, the disper- growth, with generation times of less than 10 min (Aiyar sants resulted in a negligible increase in biodegrada- et al. 2002). Each is capable of forming biofilms, which tion of oil when compared to non-dispersed oil. may explain the biofilm observed in the production COREXIT can be a highly labile substrate for microbial tubes. V. fluvialis and V. natriegens have been identified growth and metabolism (Zahed et al. 2010), and in hydrocarbon-degrading communities in experimental increases in carbon mineralization in oil samples may seawater meso cosms (Venkateswaran et al. 1991); be attributed to mineralization of the chemical disper- however, their role as hydrocarbon degraders is not well sant alone (Lindstrom & Braddock 2002, Brakstad recognized. In experimental studies, V. natriegens 2008). This may explain the results observed for Acine- was capable of metabolizing insoluble surfactants at tobacter venetianus where, after t = 24 h, production in air–water interfaces (Salter et al. 2009). This ability to the highest dilution of COREXIT rebounded to levels metabolize surfactants may explain why the Vibrio sp. that exceeded the control (Fig. 3). However, in the case isolate showed greater viability in the presence of of Marinobacter hydrocarbonoclasticus, no such recov- COREXIT compared to other phylotypes in this study. ery occurred. In this respect, COREXIT may impart At some concentrations, all isolates were susceptible positive effects on hydrocarbon-degrading cultures to impairment as a result of exposure to COREXIT. The capable of withstanding its initial toxicity. Because addition of dispersant, at all concentrations, resulted in some Vibrio spp. are capable of metabolizing disper- a significant reduction in the live:dead cell ratio in all sant (Salter et al. 2009), it is reasonable to hypothesize isolates tested—with the exception of Vibrio sp., that the increase in viability in cultures of Vibrio sp. in which, of the organisms examined, appear to be the the live/dead staining analysis may be due to metabo- most tolerant to COREXIT (Figs. 1 & 2). lism of COREXIT. However, even within this group, a Because of the importance of microorganisms to the reduction in heterotrophic secondary production was natural attenuation of hydrocarbons, toxicity tests observed in the presence of COREXIT (Fig. 3), were focused on hydrocarbon-degrading isolates. although the impact on Vibrio sp. was less significant COREXIT at concentrations <1 mg ml–1 was capable of than that on M. hydrocarbonoclasticus. The high killing at least 60% of cells in cultures of the 2 most turnover rates observed for numerous Vibrio spp. may abundant hydrocarbon-degrading isolates observed in likewise explain the reduced toxicity of COREXIT over the LH-PCR profile (Fig. 2). This is a significant finding other isolates. given that the 1:50 dilution, ~0.2 mg ml–1, is within the concentration range for COREXIT® EC9500A classified as ‘practically non-toxic’ to the standard toxicity CONCLUSIONS test organism Menidia beryllina (US EPA 2010a). The result for M. beryllina was included in a recent Envi- Hydrocarbon degradation in the marine environ- ronmental Protection Agency (EPA) report identifying ment is dependent on the ability of microorganisms to the environmental hazards associated with the use of utilize hydrocarbons for growth and metabolism. The National Contingency Plan (NCP)-approved disper- results of the current study demonstrate that microbial sants during the Deepwater Horizon response. The populations are susceptible to toxicity from the use of results of the current study demonstrate that, at con- COREXIT® EC9500A when applied at prescribed concentrations which do not present a significant hazard centrations. While the short-term goals of dispersants 108 Aquat Microb Ecol 63: 101–109, 2011

may be achieved, this study provides evidence that for counting bacteria by fluorescence microscopy. Appl COREXIT® EC9500A differentially impacts hydrocar- Environ Microbiol 33:1225–1228 bon-degrading microorganisms. Although toxicity Hollaway SL, Faw GM, Sizemore RK (1980) The bacterial community composition of an active oil field in the north- testing with bacteria is only a minor component of the western Gulf of Mexico. Mar Pollut Bull 11:153–156 development of new dispersants, these experiments Joye SB, Boetius A, Orcutt BN, Montoya JP, Schulz HN, Erick- demonstrate the importance of understanding the son MJ, Lugo SK (2004) The anaerobic oxidation of impact of dispersants on microbial communities, as methane and sulfate reduction in sediments from Gulf of Mexico cold seeps. Chem Geol 205:219–238 there is potential to diminish the capacity of the Jung JH, Yim UH, Han GM, Shim WJ (2009) Biochemical environment to mitigate spills. changes in rockfish, Sebastes schlegeli, exposed to dispersed crude oil. Comp Biochem Physiol Part C: Toxicol Pharmacol 150:218–223 Acknowledgements. We thank W. Wood (NRL) for acquiring Labson VF, Clark RN, Swayze GA, Hoefen TM and others the samples used in this study, P. Gillevet and M. Sikaroodi (2010) Estimated lower bound for leak rates from the (George Mason University) for assistance with LH-PCR analy- Deepwater Horizon spill—Interim report to the Flow Rate sis, B. Ringeisen (NRL) for helpful discussions and support of Technical Group from the Mass Balance Team. U.S. Geo- this work, R. Jonas (George Mason University) for comments logical Survey Open-File Report 2010–1132 on the draft manuscript and L. Tender (NRL) for providing the Lanoil BD, Sassen R, La Duc MT, Sweet ST, Nealson KH dispersant used in these experiments. We acknowledge the (2001) Bacteria and archaea physically associated with Naval Research Laboratory base program for support of this Gulf of Mexico gas hydrates. Appl Environ Microbiol 67: project. 5143–5153 Leahy JG, Colwell RR (1990) Microbial degradation of hydrocarbons in the environment. Microbiol Mol Biol Rev 54: LITERATURE CITED 305–315 Lindstrom JE, Braddock JF (2002) Biodegradation of petro- Aharon P, Fu B (2000) Microbial sulfate reduction rates and leum hydrocarbons at low temperature in the presence sulfur and oxygen isotope fractionations at oil and gas of the dispersant Corexit 9500. Mar Pollut Bull 44: seeps in deepwater Gulf of Mexico. Geochim Cosmochim 739–747 Acta 64:233–246 Litchfield CD, Sikaroodi M, Gillevet PM (2005) The microbial Aiyar SE, Gaal T, Gourse RL (2002) rRNA promoter activity in diversity of a solar saltern on San Francisco Bay. In: the fast-growing bacterium Vibrio natriegens. J Bacteriol Gunde-Cimerman N, Oren A, Plemenita$ A (eds) Adapta- 184:1349–1358 tion to life at high salt concentrations in Archaea, Bacteria, Atlas RM (1981) Microbial degradation of petroleum hydro- and Eukarya. Springer, Dordrecht, p 59–69 carbons—an environmental perspective. Microbiol Rev Lloyd KG, Albert DB, Biddle JF, Chanton JP, Pizarro O, Teske 45:180–209 A (2010) Spatial structure and activity of sedimentary Brakstad OG (2008) Natural and stimulated biodegradation of microbial communities underlying a Beggiatoa spp. mat in petroleum in cold marine environments. In: Margesin R, a Gulf of Mexico hydrocarbon seep. PLoS ONE 5(1):e8738 Schinner F, Marx JC, Gerday C (eds) Psychrophiles: Mills HJ, Hodges C, Wilson K, MacDonald IR, Sobecky PA from biodiversity to biotechnology. Springer Verlag, (2003) Microbial diversity in sediments associated with Berlin, p 389–407 surface-breaching gas hydrate mounds in the Gulf of Coastal Response Research Center (2010) Deepwater Horizon Mexico. FEMS Microbiol Ecol 46:39–52 dispersant use. Meeting Report. University of New Hamp- National Research Council Committee on Oil in the Sea shire, Durham, NH (2003) Oil in the sea III: inputs, fates and effects. National Couillard CM, Lee K, Legare B, King TL (2005) Effect of dis- Academies Press, Washington, DC persant on the composition of the water-accommodated Orcutt BN, Joye SB, Kleindienst S, Knittel K and others (2010) fraction of crude oil and its toxicity to larval marine fish. Impact of natural oil and higher hydrocarbons on micro- Environ Toxicol Chem 24:1496–1504 bial diversity, distribution, and activity in Gulf of Mexico Garcia-Valdes E, Cozar E, Rotger R, Lalucat J, Ursing J (1988) cold-seep sediments. Deep-Sea Res II 57:2008–2021 New naphthalene-degrading marine Pseudomonas Reed AJ, Lutz R, Vetriani C (2006) Vertical distribution and strains. Appl Environ Microbiol 54:2478–2485 diversity of bacteria and archaea in sulfide and methane- Gauthier MJ, Lafay B, Christen R, Fernandez L, Acquaviva M, rich cold seep sediments located at the base of the Florida Bonin P, Bertrand JC (1992) Marinobacter hydrocarbono- Escarpment. Extremophiles 10:199–211 clasticus gen. nov., sp. nov., a new, extremely halotolerant, Richardson LL, Miller AW, Broderick E, Kaczmarsky L, Gan- hydrocarbon-degrading marine bacterium. Int J Syst tar M, Stanic´ D, Sekar R (2009) Sulfide, microcystin, and Bacteriol 42:568–576 the etiology of black band disease. Dis Aquat Org 87: Hamdan LJ, Jonas RB (2006) Seasonal and interannual 79–90 dynamics of free-living bacterioplankton and microbially Salter I, Zubkov MV, Warwick PE, Burkill PH (2009) Marine labile organic carbon along the salinity gradient of the bacterioplankton can increase evaporation and gas trans- Potomac River. Estuar Coast 29:40–53 fer by metabolizing insoluble surfactants from the air–sea- Hamdan LJ, Gillevet PM, Sikaroodi M, Pohlman JW, Plum- water interface. FEMS Microbiol Lett 294:225–231 mer RE, Coffin RB (2008) Geomicrobial characterization of Smith DC, Azam F (1992) A simple, economical method for gas hydrate-bearing sediments along the mid-Chilean measuring bacterial protein synthesis rates in seawater margin. FEMS Microbiol Ecol 65:15–30 using 3H-leucine. Mar Microb Food Webs 6:107–114 Head IM, Jones DM, Roling WFM (2006) Marine microorgan- US EPA (U.S. Environmental Protection Agency) (1995) isms make a meal of oil. Nat Rev Microbiol 4:173–182 COREXIT® EC9500A NCP Product schedule. http://www. Hobbie JE, Daley RJ, Jasper S (1977) Use of Nuclepore filters epa.gov/osweroe1/content/ncp/products/corex950.htm Hamdan & Fulmer: Effects of COREXIT® EC9500A on bacteria 109

US EPA (2010a) Comparative toxicity of eight oil dispersant Evol Microbiol 59:1376–1381 products on two Gulf of Mexico aquatic test species. http:// Venkateswaran K, Iwabuchi T, Matsui Y, Toki H, Hamada E, www.epa.gov/bpspill/reports/phase2dispersant-toxtest.pdf Tanaka H (1991) Distribution and biodegradation poten- US EPA (2010b) National Contingency Plan product sched- tial of oil-degrading bacteria in North Eastern Japanese ule. http://www.epa.gov/emergencies/content/ncp/prod- coastal waters. FEMS Microbiol Lett 86:113–121 uct_schedule.htm Zahed MA, Aziz H, Isa M, Mohajeri L (2010) Effect of initial Vaneechoutte M, Nemec A, Musilek M, van der Reijden oil concentration and dispersant on crude oil biodegrada- TJK and others (2009) Description of Acinetobacter tion in contaminated seawater. Bull Environ Contam venetianus ex Di Cello et al. 1997 sp. nov. Int J Syst Toxicol 84:438–442

Editorial responsibility: Antje Boetius, Submitted: August 25, 2010; Accepted: December 9, 2010 Bremen, Germany Proofs received from author(s): February 18, 2011 Home Search Collections Journals About Contact us My IOPscience

Oil carbon entered the coastal planktonic food web during the Deepwater Horizon oil spill

This article has been downloaded from IOPscience. Please scroll down to see the full text article. 2010 Environ. Res. Lett. 5 045301 (http://iopscience.iop.org/1748-9326/5/4/045301)

View the table of contents for this issue, or go to the journal homepage for more

Download details: IP Address: 193.191.134.1 The article was downloaded on 19/04/2011 at 10:24

Please note that terms and conditions apply. IOP PUBLISHING ENVIRONMENTAL RESEARCH LETTERS Environ. Res. Lett. 5 (2010) 045301 (6pp) doi:10.1088/1748-9326/5/4/045301 Oil carbon entered the coastal planktonic food web during the Deepwater Horizon oil spill

William M Graham1,2,3, Robert H Condon1, Ruth H Carmichael1,2, Isabella D’Ambra1,2, Heather K Patterson1,2,LauraJLinn1 and Frank J Hernandez Jr1,2

1 Dauphin Island Sea Lab, Dauphin Island, AL 36528, USA 2 Department of Marine Sciences, University of South Alabama, Mobile, AL 36688, USA

E-mail: [email protected]

Received 27 September 2010 Accepted for publication 1 November 2010 Published 8 November 2010 Online at stacks.iop.org/ERL/5/045301

Abstract The Deepwater Horizon oil spill was unprecedented in total loading of petroleum hydrocarbons accidentally released to a marine ecosystem. Controversial application of chemical dispersants presumably accelerated microbial consumption of oil components, especially in warm Gulf of Mexico surface waters. We employed δ13C as a tracer of oil-derived carbon to resolve two periods of isotopic carbon depletion in two plankton size classes. Carbon depletion was coincident with the arrival of surface oil slicks in the far northern Gulf, and demonstrated that subsurface oil carbon was incorporated into the plankton food web.

Keywords: zooplankton, petroleum hydrocarbon, stable isotope, Gulf of Mexico S Online supplementary data available from stacks.iop.org/ERL/5/045301/mmedia

1. Introduction present quantitative data collected as a rapid-response effort to track carbon isotopic signal in two size classes representing a Following the sinking of Deepwater Horizon (DWH) on pathway into the bulk zooplankton community of the northern 22 April 2010, an estimated 780 000 m3 of Sweet Louisiana Gulf of Mexico. The present study reflects only the initial Crude (SLC) and 205 000 mT of methane [1] were released steps of a larger and continuing laboratory experimental and into the northern Gulf of Mexico over an 85 d period. General field effort aimed at understanding how oil affects pelagic agreement exists that ∼25% was directly recovered or burned communities of the northern Gulf and vice versa effects of the at sea, leaving ∼75% to be degraded naturally or with the aid biological community on the fate of the oil. of chemical dispersants [2]. Recent publications document the scope of deep subsea oil and methane along the northern Gulf slope [1, 3, 4], but scant evidence exists for the presence of 2. Methodology subsea oil in warm (>25 ◦C), shallow shelf waters. We employed δ13C as a tracer of oil-derived carbon A large pool of isotopically depleted carbon from incorporation into the lower marine food web across the middle dispersed oil and methane is presumably available for and inner continental shelf. During JuneÐAugust 2010, we biological consumption via prokaryotic consumers [5]. followed two plankton size classes: the nominally 1 μmÐ Isotopic depletion extending into marine zooplankton grazers, 0.2 mm ‘small suspended particulate’ and the >0.2Ð2 mm a pathway mediated by the microbial food web [6], is a ‘mesozooplankton’ fractions, with the former considered likely good proxy for food web modification by the spill. Here we food for the latter. The study region, >100 km north of the 3 Author to whom any correspondence should be addressed. DWH well head, had three defined northward pulses of surface

Table 1. The location and depth of surface and deep sampling stations in the Gulf of Mexico and reference sites within Mobile Bay, AL, including mean (±SD) salinity at each depth. n.d. = no data, dash = not deep enough to collect separate Deep sample. Symbols next to station names are for reference to figure 1. Surface Deep Station, symbol Latitude Longitude Depth (m) Mean salinity (psu) Depth (m) Mean salinity Gulf sites T35, 29.7989 −88.2083 1 27 ± 33336± 0 T20, 30.0902 −88.2116 1 25 ± 31833± 4 T10, • 30.1609 −88.1229 1 24 ± 3832± 3 Buoy M (BM), 30.1306 −88.1097 1 23 ± 415n.d. Mobile Bay reference sites Cedar Point Reef, + 30.3256 −88.1327 1 12 ± 6—— Sand Reef, − 30.2772 −88.1052 1 18 ± 6—— oil: two prior to the well’s 15 July shut-in, and one in August 2.2. Stable isotope analysis (figures 1(A), (B), movie S1 available at stacks.iop.org/ERL/ 13 5/045301/mmedia). Samples were collected in surface and Bulk carbon (C) stable isotope ratios (δ C, ) were measured bottom waters of the Gulf of Mexico at four shelf and two inner by continuous flow isotope ratio mass spectrometry at the Mobile Bay reference sites (figure 1(A), table 1). University of Utah Stable Isotope Facility (USA) and on a Picarro cavity ringdown spectrometer coupled to a Costech 2.1. Plankton and suspended particle collection elemental analyzer at Dauphin Island Sea Lab (DISL). Bulk carbon isotopic composition in organisms reflects both short- ( μ . ) Suspended particulates 1 mÐ0 2mm were collected using term energy stores (i.e., lipids) and relatively longer turnover 1.7 l vertical Niskin bottles deployed at target depths of one in tissues [7]. Since lipids are isotopically depleted and meter above the bottom and one meter below the surface at do not necessarily reflect time-integrated diet of organisms, stations T10, T20, and T35 at two-week intervals from 2 June variation in lipid content may introduce bias into stable isotope to 15 August, 2010. At Buoy M, and the two bay reference analyses. Established mathematical normalization techniques sites, Cedar Point Reef and Sand Reef, only water from one allow correction of δ13C values in lipid-rich samples, but meter below the surface was collected owing to shallowness of preserve sample integrity for other analyses [7]. Here, the water column; these were collected at 1Ð2 week intervals. bulk δ13C values in mesozooplankton were lipid-corrected At the Sand Reef reference station, samples were collected using a 1 l horizontal sampler. Air-filled balloons, released according to [7], after comparison to C:N in mesozooplankton δ13 at depth from bottle ends, were used to avoid surface oil samples (figure 2(A). Comparison of C values to the C:N ( −1) contamination during sampling. Water was vacuum filtered and relative C content mg l in suspended particulates (5 psi) onto pre-combusted GF/F filters, and dried to a indicated no correction was needed for the smaller fraction constant weight at 60 ◦C. (figures 2(A) and (B)). C and N content were obtained during Mesozooplankton were collected at the same stations and stable isotope analysis. with the same timing as above with openingÐclosing plankton nets (3.5 m long, 0.25 m diameter), using 333 μm for surface 2.3. Source crude oil samples and bottom and 202 μm for oblique samples at Gulf sites. A μ 202 m mesh (1 m long, 0.5 m diameter) ring plankton net We analyzed δ13C of oil in both weathered and fresh condition. was used at BM and reference sites (mesozooplankton were not Weathered surface oil was collected in the nearshore waters collected at Sand Reef). Samples were rinsed with ultrapure ◦ off Dauphin Island, Alabama, on 11 June 2010, and stored water, dried at 60 C, and homogenized by mortar and pestle. in the dark at 5 ◦C. Prior to carbon stable isotope analysis, Additional historical pre-spill mesozooplankton samples ◦ the weathered oil was further dried at 60 C for 48 h to from Buoy M and suspended particulate samples from T35, remove residual water. This sample was analyzed for δ13C T20 and Buoy M were collected in MayÐAugust of both 2008 at the University of Utah Stable Isotope Facility. Fresh and 2009 (all data within each station were pooled as ‘pre- Source Oil B (SOB) supplied by BP was collected on spill’. Collections were similar to those described for 2010. These samples were historically analyzed only for carbon (C) 22Ð23 May 2010, from the riser insertion tube on board and nitrogen (N) content; no pre-spill stable isotope data exist the drill ship Enterprise. Accompanying documentation for for continental shelf stations. All mesozooplankton samples two samples (SOB-20100716-067 and SOB-20100716-130) are being processed for community assemblage changes with reported Nalco EC9323A defoamer was injected topsides, and respect to the spill; however, assemblage analysis is beyond the subsea injections included methanol with 10 000 ppm VX9831 scope of this study. That said, a cursory review of the samples oxygen scavenger/catalysts solution. Fresh oil was stored ◦ for presence of contaminating oil droplets revealed the samples at 5 C until analysis by saturating a small piece of pre- were clear of both oil and resuspended sediments. In addition, combusted GF/F filter with oil and analyzed using a Picarro the zooplankton samples were dominated by organisms typical cavity ringdown spectrometer coupled to a Costech elemental of springÐsummer assemblages such as calanoid copepods. analyzer at DISL.

2 Environ. Res. Lett. 5 (2010) 045301 W M Graham et al

Figure 1. (A) Sample sites, symbols relating to (D) and (E); box deﬁnes area used to calculate oil % coverage in (B) with example of peak oil coverage 28 June 2010. A full animation of oil movement around these sites can be found in the supplementary data (movie S1 available at stacks.iop.org/ERL/5/045301/mmedia). (B) Timing of three shoreward pulses of oil, IÐIII (cf movie S1 available at stacks.iop.org/ERL/5/ 045301/mmedia). (C) Daily averaged river discharge into Mobile Bay. (D) δ13C values for mesozooplankton fraction (0.2Ð2 mm). (E) δ13C values for suspended particulate fraction (1 μmÐ0.2 mm). Both (D) and (E) referenced against δ13C values from Mobile Bay and weathered and fresh SLC oil. Error bars show standard deviation.

3 Environ. Res. Lett. 5 (2010) 045301 W M Graham et al

Figure 2. Analysis of C stable isotope ratios. (A) Bulk δ13C in mesozooplankton (0.2Ð2 mm) and suspended particulates (1 μmÐ0.2 mm) compared to C:N in surface, deep, and oblique (mesozooplankton only) samples. (B) δ13C compared to carbon content in suspended particulates. (C) Corrected δ13C in mesozooplankton and δ13C in smaller suspended particulates compared to salinity at surface and deep sampling locations. (D) Bulk δ13C in suspended particulates compared to chlorophyll a (chl a) in water samples from stations T10, T20, and T35.

2.4. Chlorophyll a Claiborne Dam on the Alabama River (# 02429500). There is an estimated 5Ð10 d flow-dependent lag between readings at Whole water samples collected at Stations T10, T20 and T35 gauging stations and reference sites in Mobile Bay [9]. Daily (1 m below surface and 1 m above bottom) were stored on ice in mean discharge is reported in figure 1(C) (not including lags the dark and filtered in the laboratory onto GF/F filters within as it did not change the pattern). Salinity was measured at 2 h of collection. Extracted chlorophyll a was fluorometrically Gulf stations T10ÐT35 using a Sea Bird SBE 25 conductivityÐ determined with a Turner Designs fluorometer [8]. temperatureÐdepth (CTD) probe and at Buoy M and the two Mobile Bay reference sites using a handheld YSI 85 salinity 2.5. Oil proximity data probe. Surface layer slick distribution was defined from Geographical Information System (GIS) data (ftp://satepsanone.nesdis.noaa. 2.7. Statistical analyses gov/OMS/disasters/DeepwaterHorizon/). Per cent coverage of Analysis of variance (ANOVA) was used for comparison oil and distance from the nearest slick to station T20 were of δ13C values, C:N, and chl a among sample types. If measured by mapping oil layers in ESRI ArcMap v9.3 GIS ANOVAs were significant, post hoc pairwise comparison of software. means using Tukey’s test of variability were performed. All linear correlations were tested using the Z-test. These analyses 2.6. Freshwater discharge were performed in StatView 5.0.1. An additional three-way ANOVA with location, sample depth and condition (pre- or To determine potential effects of freshwater discharge on post-spill years) was performed on C:N values for suspended 13 δ C, riverine discharge into Mobile Bay was estimated by particulates and mesozooplankton using Minitab 15. All data adding daily discharge rates at two US Geological Survey were normally distributed and did not require transformation, gauging stations (http://waterdata.usgs.gov/usa/nwis/sw), the and also conformed to the assumption of homogeneity of Coffeeville Dam on the Tombigbee River (#02469765) and the variance.

4 Environ. Res. Lett. 5 (2010) 045301 W M Graham et al 3. Results and discussion suspended particles, and some primary consumers in Mobile Bay and elsewhere [13, 14]. δ13C values in mesozooplankton δ13C depletion occurred in each size fraction at middle and were not related to salinity (figure 2(C)), and δ13C in suspended inner shelf stations coincident with two sequential northward particulates showed a weak positive correlation only when pulses of surface oil slicks from DWH (figures 1(D), (E)). surface and bottom samples were considered together (r = Relative to early June, an isotopic shift of −1to−4 0.37, P = 0.03). This finding is consistent with the relatively (toward weathered and fresh oil, −27.23 ± 0.03 and low discharge to Mobile Bay during most of the sampling −27.34 ± 0.34, respectively) occurred during the peak of period (figure 1(C)). The similar patterns of δ13C depletion in areal coverage of oil over the sites (figures 1(B), (D), (E)). both mesozooplankton and small suspended fractions, despite Recovery from this depletion to the pre-spill baseline was 2Ð decreasing discharge to Mobile Bay and little or no relationship 4 wks. A third pulse of residual oil occurred in late July, to salinity during the period of greatest oil proximity and 13 and depleted δ C was observed in mid-August at the furthest coverage in the region, supports an oil-derived C source offshore stations. Depletion and recovery cycles on the order mediating this shift as opposed to a freshwater-derived source of a few weeks are consistent with published warm water from Mobile Bay. petroleum hydrocarbon decay timescales [10]. Since phytoplankton were a component of the mixed δ13 The apparent oil-related C depletion occurred in both small suspended particulate fraction (1 μmÐ0.2 mm), isotopic fractions and throughout the water column. The pattern depletion of C in this fraction and subsequently in the δ13 was consistent despite the differences in both CandC:N mesozooplankton could result from dominance by isotopically δ13 between the two size fractions. C and C:N values differed depleted phytoplankton. Given the range of δ13C values typical between mesozooplankton and suspended particulates, with in marine phytoplankton (−20 to −24;[15, 16]) and the δ13 − . ± . mesozooplankton having heavier C( 20 44 1 37 lack of evidence for a significant freshwater influence during − . ± . . ± . compared to 23 19 1 26 )andlowerC:N(48 0 6 the period of depletion, phytoplankton alone are unlikely to . ± . compared to 6 9 0 9) than suspended particles (ANOVA: account for the observed depletion (figures 1(D), (E)). The δ13 = . < . = . < C:F4,84 23 29, P 0 001; C:N:F4,84 43 04, P concentration of chlorophyll pigments extracted from water . 0 001; figure 2(A). Comparisons among surface, bottom, and samples collected at stations T10 through T35 varied relatively oblique samples did not differ for either size fraction (Tukey’s little during the study period and did not indicate a bloom of > . post hoc test: P 0 05 for all comparisons). shelf phytoplankton (1.50 ± 1.02 μgl−1; figure 2(D)). Chl a δ13 Bulk C values were correlated with C:N in surface concentration also was not related to δ13C of the suspended (r =−0.64), bottom (r =−0.61), and oblique (r =−0.63) particulate fraction (F , = 2.33, P = 0.14). Surface mesozooplankton samples (P < 0.01) for all comparisons; reg1 24 and bottom samples were not different and were combined figure 2(A), consistent with a mesozooplankton fraction for further analysis (two-way ANOVA with station and depth (largely composed of animals) and requiring correction for as variables: F , = 1.30, P = 0.29) (figure 2), indicting lipid-related depletion of δ13C[7]. In contrast, δ13C 3 26 that a change in phytoplankton abundance made no measurable in suspended particulates was not correlated with C:N contribution to the isotopic depletion of C shown in figure 1. (figure 2(A)), and was weakly correlated with the relative To elucidate whether the observed δ13C depletion was due C content in the sample only when surface and bottom to the contaminating presence of oil in the water column or to fractions were considered together (r =−0.44, P < 0.01; the assimilation and incorporation of oil-derived C by resident figure 2(B)). These findings suggest a small particle fraction biota, we compared C:N for mesozooplankton and suspended of mixed composition, including algal and detrital matter that particulate fractions during pre- (2008Ð2009) and post-oil spill did not demand lipid correction despite a higher C:N ratio than (2010) years between May and August. The expectation was mesozooplankton [7, 11]. The mean correction applied to bulk that presence of SLC oil on or inside the animals would yield δ13C in mesozooplankton samples was 1.49 ± 1.73.The anomalously high C:N values. For the suspended particulate relative shift in sample values can be seen by comparing panels fraction, there was no difference in C:N (by weight) between A and C in figure 2. any of the sampling stations within and across pre- and post- spill years (three-way ANOVA with station, pre- and post- 3.1. Discounting masking effects or sample contamination spill years, and sample depth as variables: F9,89 = 1.46, In comparison to reference sites inside Mobile Bay, offshore P = 0.18) (figure 3). Similarly, there was no significant depletion of δ13C was not related to timing of freshwater difference in mesozooplankton C:N at station BM in pre- and discharge from the Bay, phytoplankton blooms, or direct post-oil spill years (ANOVA: F1,17 = 0.69, P = 0.42), and contamination of samples with external oil. Corrected δ13C while we do not have pre-spill C:N data for stations T35, T20 in mesozooplankton and bulk δ13C values in suspended and T10, post-spill C:N data from these sites were the same particulates were compared to salinity at each station to detect as those at station BM (ANOVA: F1,68 = 2.30, P = 0.09) potential freshwater influence (figure 2(C)). The hydrology of (figure 3). Combined, these results suggest that the depleted Mobile Bay is dominated by freshwater inputs, which lead to C isotope values were not driven by direct oil contamination salinity stratification [12] and may convey isotopically light in the samples (e.g., oil micro-droplets collected on the filter). suspended particles and biota from the upper reaches of the That similar results were found for both mesozooplankton Bay to the Gulf [13]. δ13C in reference oil samples was similar and suspended particulate fractions suggests oil-derived C was to δ13C typically found in freshwater-derived vegetation, transferred through the food web.

5 Environ. Res. Lett. 5 (2010) 045301 W M Graham et al the Richard C Shelby Center for Ecosystem-Based Fisheries Management and the Northern Gulf Institute’s rapid-response funding from BP’s Gulf Research Initiative. We thank J Burchﬁeld, M Miller, K Weis, J Hermann, B Dzwonkowski, K Park, K Robinson, L McCallister and the crew of the R/V E O Wilson for laboratory, ﬁeld and logistical support.

References

[1] Adcroft A, Hallberg R, Dunne J P, Samuels B L, Galt J A, Barker C H and Payton D 2010 Geophys. Res. Lett. 37 L18605 [2] Kerr R A 2010 A lot of oil on the loose, not so much to be found Science 329 734Ð5 [3] Camilli R, Reddy C M, Yoerger D R, Van Mooy B A S, Jakuba M V, Kinsey J C, McIntyre C P, Sylva S P and Maloney J V 2010 Tracking hydrocarbon plume transport and biodegradation at Deepwater Horizon Science 330 201Ð4 [4] Hazen T C et al 2010 Deep-sea oil plume enriches indigenous oil-degrading bacteria Science 330 204Ð8 [5] Spies R B and DesMarais D J 1983 Natural isotope study of Figure 3. Particulate organic carbon to particulate nitrogen ratios (C:N) of (A) mesozooplankton (0.2Ð2 mm) and (B) smaller trophic enrichment of marine benthic communities by suspended particulates (1 μmÐ0.2 mm) collected at stations T35, petroleum seepage Mar. Biol. 73 67Ð71 T20, T10 and BM (cf figure 1 and table 1). There was no significant [6] Sherr E B and Sherr B F 1991 Planktonic microbes: tiny cells at difference between surface and bottom C:N within each station, thus the base of the ocean’s food webs Trends Ecol. Evol. 6 50Ð3 data were pooled for all pre- and post-spill analyses (ANOVA: [7] Post D M, Layman C A, Arrington D A, Takimoto G, F1,97 = 0.04, P = 0.85). ND indicates data were not available. Quattrochi J and Montana C G 2007 Getting to the fat of the matter: models, methods and assumptions for dealing with lipids in stable isotope analyses Oecologia 152 179Ð89 4. Conclusions [8] Welschmeyer N A 1994 Fluorometric analysis of chlorophyll a in the presence of chlorophyll b and phaeopigments Limnol. Oceanogr. 39 1985Ð92 Carbon isotopic depletion in mesozooplankton and suspended [9] Schroeder W W 1979 Dispersion and Impact of Mobile River particulate samples throughout the water column (figures 1(D) System Waters in Mobile Bay, Alabama (Water Resources and (E)) indicates trophic transfer of oil carbon into the Research Institute vol 37) (Auburn, AL: Auburn University) planktonic food web. A similar response found in benthic p48 communities around natural seeps [5] suggests that carbon [10] Atlas R M 1981 Microbial degradation of petroleum isotopic shifts in the plankton fractions are likely due to hydrocarbons: an environmental perspective Microbiol. Rev. the duration and magnitude of depleted carbon released 45 180Ð209 into the system. These data provide strong evidence that [11] S¿reide J E, Tamelander T, Hop H, Hobson K A and labile fractions of the oil extended throughout the shallow Johansen I 2007 Sample preparation effects on stable C and N isotope values: a comparison of methods in Arctic marine water column during northward slick transport and that food web studies Mar. Ecol.-Prog. Ser. 328 17Ð28 this carbon was processed relatively quickly at least two [12] Schroeder W W, Dinnel S P and Wiseman W J Jr 1990 Salinity trophic levels beyond prokaryotic hydrocarbon consumers stratification in a river-dominated estuary Estuar. Coasts given our understanding of microbial-zooplankton trophic 13 145Ð54 linkages [6, 17]. Further, this study provides a launching [13] Goecker M E, Valentine J F, Sklenar S A and Chaplin G I 2009 point for follow up experimental laboratory and field exercises Influence from hydrological modification on energy and aimed at understanding the fate and transport of petroleum nutrient transference in a deltaic food web Estuar. Coasts hydrocarbons in marine planktonic ecosystems under the 32 173Ð87 influence of natural or human-mediated chemical dispersion. [14] Michener R H and Schell D M 1994 Stable isotope ratios as tracers in marine aquatic food webs Stable Isotopes in Ecology and Environmental Science ed K Lajtha and R Acknowledgments H Michener (Oxford: Blackwell Scientific) pp 138Ð58 [15] Moncreiff C A and Sullivan M J 2001 Trophic importance of This work was supported by a grant from the National epiphytic algae in subtropical seagrass beds: evidence from multiple stable isotope analyses Mar. Ecol. Prog. Ser. Science Foundation through the RAPID program funding for 215 93Ð106 oil spill research (OCE-1043413). Additional funding and [16] Fry B 2006 Stable Isotope Ecology (New York: Springer) p 308 resources were from the Alabama Department of Conservation [17] Azam F, Fenchel T, Field J G, Gray J S, Meyer-Reil L A and and Natural Resources, Marine Resources Division, and the Thingstad F 1983 The ecological role of water-column National Oceanic and Atmospheric Administration through microbes in the sea Mar. Ecol. Prog. Ser. 10 257Ð63

6 GEOPHYSICAL RESEARCH LETTERS, VOL. 39, L01605, doi:10.1029/2011GL049505, 2012

Macondo-1 well oil-derived polycyclic aromatic hydrocarbons in mesozooplankton from the northern Gulf of Mexico Siddhartha Mitra,1 David G. Kimmel,2,3 Jessica Snyder,2 Kimberly Scalise,1 Benjamin D. McGlaughon,2 Michael R. Roman,4 Ginger L. Jahn,4 James J. Pierson,4 Stephen B. Brandt,5 Joseph P. Montoya,6 Robert J. Rosenbauer,7 Thomas D. Lorenson,7 Florence L. Wong,7 and Pamela L. Campbell8 Received 7 October 2011; revised 7 December 2011; accepted 11 December 2011; published 14 January 2012.

[1] Mesozooplankton (>200 mm) collected in August and disaster than public knew, 2010, http://mcbi.org/news/PR- September of 2010 from the northern Gulf of Mexico show Norse-Amos-2010.pdf). Chemical evidence of subsurface evidence of exposure to polycyclic aromatic hydrocarbons oil from the leak was found as far away as 30 km south of the (PAHs). Multivariate statistical analysis revealed that M-1 well [Diercks et al., 2010]. Oil which is a complex distributions of PAHs extracted from mesozooplankton mixture of hydrocarbons and other chemicals, contains were related to the oil released from the ruptured British numerous PAHs [Connell, 1997; National Research Council, Petroleum Macondo-1 (M-1) well associated with the R/V 2002]. These PAHs may be used as chemical fingerprints of Deepwater Horizon blowout. Mesozooplankton contained specific types of oil released into natural environments 0.03–97.9 ng gÀ1 of total PAHs and ratios of fluoranthene [Blumer, 1976]. For example, PAH distributions, or relative to fluoranthene + pyrene less than 0.44, indicating a liquid abundances of low and high molecular weight PAHs, have fossil fuel source. The distribution of PAHs isolated from been used to fingerprint oil and determine the provenance of mesozooplankton extracted in this study shows that the various oil spills in the environment [Bennett et al., 2000; 2010 Deepwater Horizon spill may have contributed to Stout et al., 2001; Christensen et al., 2004]. Despite exten- contamination in the northern Gulf of Mexico ecosystem. sive efforts at completely delineating the extent of the DWH Citation: Mitra, S., et al. (2012), Macondo-1 well oil-derived oil spill, there have been no studies published to date polycyclic aromatic hydrocarbons in mesozooplankton from the addressing whether or not the spill introduced oil-derived northern Gulf of Mexico, Geophys. Res. Lett., 39, L01605, toxic compounds into the pelagic food web of the nGOM. doi:10.1029/2011GL049505. [3] Mesozooplankton are useful sentinel organisms for oil-derived pollution [Carls et al., 2006] that serve as food 1. Introduction for early life stages of fish and shrimp. Furthermore, they act as conduits for the movement of oil-derived contamination [2] An estimated 4.93 million barrels (1 barrel = 42 and other persistent organic pollutants through the marine US gallons) of crude oil were released into the Northern food web [Clayton et al., 1977; Borgå et al., 2004; Sobek Gulf of Mexico (nGOM) from the British Petroleum et al., 2006; Hallanger et al., 2011a, 2011c]. The objective (BP) Macondo-1 (M-1) site (Federal Interagency Solutions of this study was to extract and analyze PAHs in mesozoo- Group, Oil budget calculator—Deepwater Horizon, 2010, plankton collected in the nGOM after the 2010 DWH spill. http://www.restorethegulf.gov/sites/default/files/documents/ The null hypothesis of this study was that the relative PAH pdf/OilBudgetCalc_Full_HQ-Print_111110.pdf), the loca- distributions in mesozooplankton collected throughout the tion of the R/V Deepwater Horizon (DWH) blowout. The Gulf of Mexico would not resemble the relative PAH dis- total extent of the surface oil slick, derived from wind, ocean tributions in oil collected from the M-1 well. currents, aerial photography, and satellite imagery, was estimated to be 180,000 km (J. Amos, BP spill was greater 2. Methods 2.1. Mesozooplankton Sampling 1 Department of Geological Sciences, East Carolina University, [4] The mesozooplankton samples in this study were col- Greenville, North Carolina, USA. 2 lected from within 20 km from the M-1 site and at distal Department of Biology, East Carolina University, Greenville, North Carolina, USA. stations grouped around 300 km south of the M-1 site 3Institute for Coastal Sciences and Policy, East Carolina University, (Figure 1). Mesozooplankton were sampled with a Multiple Greenville, North Carolina, USA. Open and Closing Net Environmental Sampling System 4 Horn Point Laboratory, Center for Environmental Science, University (MOCNESS) [Wiebe et al., 1976]. The MOCNESS had a of Maryland, Cambridge, Maryland, USA. 2 m 5Oregon Sea Grant, Oregon State University, Corvallis, Oregon, USA. 1m opening and a mesh size of 200 m. Five to ten minute 6School of Biology, Georgia Institute of Technology, Atlanta, Georgia, tows were conducted at the surface, and forty to sixty minute USA. tows were conducted between 500 m and the surface in 7Pacific Coastal Marine Science Center, U.S. Geological Survey, Menlo August and September of 2010 (auxiliary material, Table S1).1 Park, California, USA. 8 All mesozooplankton (>200 mm) were collected from the Water Resources, National Research Program, U.S. Geological Survey, Menlo Park, California, USA. cod end of the nets and immediately frozen in glass vials.

L01605 1of7 L01605 MITRA ET AL.: DWH PAHS IN GOM MESOZOOPLANKTON L01605

Figure 1. Map of sampling area. Oil extent as of May 2010. Numbered symbols represent stations from which samples were analyzed for PAHs in this study.

This procedure generally resulted in two sets of samples tins and transferred to vials to be extracted. Following from each station: one group of mesozooplankton samples determination of the sample weight, two mL of deuterated from the surface water and one group integrated across the PAH surrogate standard (in acetone) were added to each top 500 m water depth. oil, zooplankton, and seep sample. Then, zooplankton samples were macerated in a Sentry tissue macerator after 2.2. Oil, Surface Slick, and Seep Samples which PAHs were extracted from them using an Acceler- [5] A sample of M-1 subsurface oil was provided by B & ated Solvent Extractor (ASE) or sonication using hexane: B Laboratory, College Station, Texas. The well oil was acetone (1:1, v:v). All extracts were concentrated by rotary- obtained by BP from the riser insertion tube aboard the evaporation followed by a N2 stream. Extracts were puri- drillship Discoverer Enterprise on May 21, 2010, and was fied via silica gel and sodium sulfate chromatography on an absent of any defoamer or dispersant. All samples were open column. The aromatic fraction was eluted with 75 mL collected, processed, and shipped under standard chain-ofof 80:20 hexane:methylene chloride. Zooplankton PAH custody protocols according to methods listed in the USGS concentrations are reported on a wet weight basis. Addi- National Field Manual for the Collection of Water-Quality tional details of sample collection and extractions, includ- Data (NFM) (http://pubs.water.usgs.gov/twri9A/) as well as ing extraction recoveries, may be found in the auxiliary other USGS standard operation procedures [Wilde et al., material. 2010]. Surface slick samples were obtained manually in pre-cleaned I-Chem jars on May 8. The only available nat- 2.4. Data Analysis ural seep samples from the Gulf of Mexico were collected in [7] Data analyses were conducted using R statistical soft- 2002. These had been stored frozen at À20 °C at the USGS ware version 2.10.1 (©R Foundation for Statistical Com- Menlo Park. Santa Barbara Channel samples, which con- puting, 2009). The proportion of each of the 24 PAHs was sisted of sludge, seep oil, and produced oil, were collected in calculated in each mesozooplankton extract by dividing 2001, 2005, and 2008 and frozen until they were extracted individual PAH concentrations by the total concentration. for PAHs in this study. We created a data matrix of columns containing proportional concentrations of each PAH compound and rows repre- 2.3. Polycyclic Aromatic Hydrocarbon (PAH) senting individual samples. Relationships between the PAH Extraction distributions in mesozooplankton and oil samples were [6] Less than a gram of each oil or seep sample was characterized using a two-stage multivariate analysis. First, weighed out in pre-cleaned (450 °C for 4h) aluminum foil we performed an agglomerative, hierarchical cluster analysis

2of7 L01605 MITRA ET AL.: DWH PAHS IN GOM MESOZOOPLANKTON L01605 of a matrix of distances calculated using the dist.prop func- analysis [Cox and Cox, 1994] to help visualize relationships tion in the R library ade4 [Dray and Dufour, 2007], we in PAH distributions between mesozooplankton, DWH oil, calculated the distances between each row based on the suite and seep samples (Figure 2, bottom). For a majority of the of PAH compounds. The cluster analysis was conducted mesozooplankton samples, PAHs detected in the nGOM using the hclust function using Ward’s linkage. To identify mesozooplankton scaled closely to either Cluster C (PAHs particular groupings of data, we used the resultant cluster in the DWH surface slicks) or Cluster D (PAHs in oil from analysis fit and cut the tree into 4 distinct groups using the the M-1 riser pipe). The PAHs in the nGOM seep and SBC cutree function. Using the resulting 4 groupings from this samples clustered together, but did not scale closely to DWH analysis, we calculated the average (Æ SD) proportion of oil-derived PAH distributions (Cluster B). Our results indi- each PAH within each group. A non-metric multidimen- cate that oil derived from the DWH incident was associated sional scaling (NMDS) analysis was performed on the with mesozooplankton collected as far as 180 km from the matrix of calculated distances using the isoMDS function M-1 well. Zooplankton in Cluster A do not have a PAH in the R library MASS [Venables and Ripley, 1999]. signature associated with DWH riser oil or DWH-derived surface slicks. This observation suggests that the oil release from the DWH incident, although spatially-extensive, may 3. Results and Discussion have been patchy or that these zooplankton were exposed [8] There were no systematic trends in PAH distributions to oil, but no longer have a relative PAH distribution that in nGOM mesozooplankton as a function of radial distance matches that of the DWH oil. from the M-1 well or with depth in the water column from [11] The PAH distributions detected in some nGOM which the organisms were collected. The lack of any dif- mesozooplankton in this study match the PAH distributions ference in PAH distributions in mesozooplankton collected extracted in oil released from the broken riser pipe collected from surface or deeper (0–500 m) waters is not surprising from the M-1 well and from surface oil slicks originating from given that many species of mesozooplankton exhibit diel the DWH incident (Figure 3). A two-sample Kolmogorov- vertical migration patterns [Haney, 1988]. In contrast to oil Smirnov goodness-of-fit (KS-GOF) test comparing the mean spills occurring at the sea surface, petroleum hydrocarbons PAH distributions revealed no differences in PAH signatures originating from the M-1 well were subjected to several between Cluster B and C ( p =0.68),ClusterBandD(p = unique environmental processes that may account for the 0.45), and Cluster C and D ( p = 0.26). Mesozooplankton unique PAH distributions within each cluster group. First, PAH signatures from Cluster A (Figure 3a), although rela- petroleum was released from the Macondo well at 1.5-km tively abundant in naphthalene, did not possess the ratios of depth. This resulted in partitioning of hydrocarbons into the fluorene:phenanthrene resembling those in the DWH surface aqueous phase in the absence of atmospheric evaporation slicks or oil from the broken riser pipe (Figures 3b and 3c). [Reddy et al., 2011]. Furthermore, the dispersants applied to Comparisons of these PAH distributions showed significant this spill may have enhanced aqueous dissolution of oil differences when compared with the KS-GOF, Cluster A and droplets, affecting overall water column concentrations of B(p = 0.03), Cluster A and C ( p = 0.004), and Cluster A and PAHs in a manner in which there was no systematic geo- D were not statistically different at a = 0.05, but did have a graphical pattern with distance from the M-1 well. low p-value of 0.07. The PAH distributions from the SBC [9] Cluster analysis suggests that PAH distributions in all and the nGOM seep sample had much higher proportions of samples analyzed could be divided into four distinct groups 1-methyl and 2-methylnapthalene (Figure 3b) than the M-1 (Figure 2, top). Group A corresponded to mesozooplankton samples or the nGOM surface slicks (Figures 3c and 3d). The that were dissimilar to any oil samples analyzed in this dissimilarity in these particular PAHs between leaking oil study. The PAHs that were detected in natural seeps in the from the M-1 well and the natural GOM seep samples can be Gulf of Mexico (Mississippi Canyon, Figure 1, Station 64) explained by chemical heterogeneity resulting from varying clustered in Group B along with PAH distributions in pro- oil source rock, age, and migration history that likely occurs duced oil, crude oil and tar balls from several locations in between the seep location and the M-1 site [Aharon et al., the Santa Barbara Channel. The distributions of PAHs 1997; Hood et al., 2002]. in mesozooplankton from the other stations led to their [12] We are not aware of any existing studies examining emplacement in Groups C and D. Distributions of PAHs in PAH distributions in zooplankton collected from the water Groups C and D were similar to PAH distributions in from column of the nGOM, therefore we have no means of the DWH surface slick and from the broken M-1 riser pipe, comparing PAH body burdens in nGOM zooplankton col- respectively. lected prior to the spill. Distributions of PAHs in fluvial [10] We used a non-parametric multidimensional scaling suspended sediments such as those exported from the analysis on the distance matrix calculated for the cluster Mississippi River generally contain elevated relative

Figure 2. (top) Agglomerative, hierarchical cluster analysis of PAH distributions in oil from the broken riser pipe at the M-1 well (DWH.D), Deepwater Horizon surface slick oil (DWH.S), mesozooplankton collected from surface tows (ZP.S) and tows from 0–500 m water depth in nGOM (ZP.D). A natural seep sample from the nGOM (GMS) and oil samples derived from the Santa Barbara channel (SBC). The four letters in both plots represent the 4 groups identified in the cluster analysis (see text for explanations). (bottom) Non-metric multi-dimensional scaling (NMDS) biplot of PAH distributions in oil from the broken riser pipe at the M-1 well (3, oval), Deepwater Horizon surface slick oil (1,2, circles), nGOM seep samples (64, squares) and oil samples derived from a variety of locations in the Santa Barbara channel (56–63, hexagons). Other numbers are representative of mesozooplankton samples (see Figure 1 for collection location). The stress of the NMDS fit was 0.14. The principal variables structuring the data are the PAH signatures within each group.

3of7 L01605 MITRA ET AL.: DWH PAHS IN GOM MESOZOOPLANKTON L01605

Figure 2

4of7 L01605 MITRA ET AL.: DWH PAHS IN GOM MESOZOOPLANKTON L01605

Figure 3. Mean ratio of individual PAHs to total PAHs (R) within groups identified using the cluster analysis. (a) Meso- zooplankton samples only, (b) Gulf of Mexico seeps and Santa Barbara Channel, (c) mesozooplankton samples and surface slick oil, and (d) mesozooplankton samples and oil from the leaking riser pipe. abundances of higher molecular weight PAHs [Mitra et al., mesozooplankton samples and that the PAH distributions in 2002] than found in mesozooplankton analyzed in this study. these samples are petrogenic. Thermodynamic modeling of PAH distributions on par- [13] There are numerous factors that may affect PAH body ticles has demonstrated that PAHs are strongly-sorbed onto burdens in zooplankton isolated for this study. Unlike the Mississippi River and Gulf of Mexico sediments, which elevated levels of PAHs accumulated in mesozooplankton contain a non-trivial fraction of combustion-derived black collected in Port Valdez, Alaska [Carls et al., 2006], the carbon [Mitra and Bianchi, 2003]. Furthermore, ratios of mesozooplankton PAH concentrations in our study (Table 1) fluoranthene:pyrene in these mesozooplankton samples are are similar to that found in zooplankton collected near other less than 0.5 (Table 1), as is expected in PAH distributions oil spills in temperate and tropical environments globally originating from a petrogenic source [Yunker et al., 2002]. [Davenport, 1982; Guzman del Proo et al., 1986]. Several Taken together, this suggests that PAHs from suspended sedi- processes (e.g. exchange with water through passive parti- ments did not contribute to the PAH signal detected in these tioning, ingestion of contaminated food, and production of

Table 1. Table of PAH Concentrations in Each Hierarchal Cluster Groupa Cluster Sample Type N Mean Standard Deviation Minimum Maximum flu/flu + pyr A mesozooplankton 19 17.4 28.9 1.64 120.0 0.19 Æ 0.23 B Santa Barbara Channel seep samples 8 1.23E + 06 6.91E + 05 4.96E + 05 2.61E + 06 0.25 Æ 0.14 nGOM seeps 4 1.19E + 04 1.15E + 04 1.60E + 03 2.70E + 04 0.29 Æ 0.11 C mesozooplankton 7 29.0 26.1 11.1 85.7 0.17 Æ 0.22 surface slicks 2 5.45E + 05 5.40E + 05 5.439 + 05 0.41 D mesozooplankton 26 13.8 19.7 2.04 97.9 0.44 Æ 0.29 M-1 well 1 3.61E + 06 3.61E + 06 3.61E + 06 0.44 aConcentrations are given in ng gÀ1 wet weight. N signifies number of samples extracted from within that group.

5of7 L01605 MITRA ET AL.: DWH PAHS IN GOM MESOZOOPLANKTON L01605 fecal pellets and eggs), occurring simultaneously, may affect 109(5), 568–579, doi:10.1130/0016-7606(1997)109<0568:RDOSHS>2.3. the final body burden of PAHs derived from oil spills [Sobek CO;2. Bennett, A., T. S. Bianchi, and J. C. Means (2000), The effects of PAH con- et al., 2006; Berrojalbiz et al., 2011]. Moreover, all of these tamination and grazing on the abundance and composition of microphy- processes may vary as a function of ambient temperature and tobenthos in salt marsh sediments (Pass Fourchon, LA, USA): II: The use seasonality [Hallanger et al., 2011b]. of plant pigments as biomarkers, Estuarine Coastal Shelf Sci., 50(3), 425–439, doi:10.1006/ecss.1999.0572. [14] Mesozooplankton species in microcosm experiments Berrojalbiz, N., S. Lacorte, A. Calbet, E. Saiz, C. Barata, and J. Dachs have been shown to excrete PAHs on time scales of days (2009), Accumulation and cycling of polycyclic aromatic hydrocarbons [Berrojalbiz et al., 2009]. However, we detected the pres- in zooplankton, Environ. Sci. Technol., 43(7), 2295–2301, doi:10.1021/ ence of PAHs in mesozooplankton collected in August and es8018226. Berrojalbiz, N., J. Dachs, M. Ojeda, M. C. Valle, J. Castro-Jimenez, early September 2010, well after the M-1 well was capped J. Wollgast, M. Ghiani, G. Hanke, and J. M. Zaldivar (2011), Biogeo- on 15 July, 2010. This was surprising given that mesozoo- chemical and physical controls on concentrations of polycyclic aromatic plankton population turnover times may be quite rapid in the hydrocarbons in water and plankton of the Mediterranean and Black Seas, Global Biogeochem. Cycles, 25, GB4003, doi:10.1029/ warmer waters of the Gulf of Mexico (e.g. Acartia tonsa has 2010GB003775. a generation time of 7 days at 25 °C [Heinle, 1966]). We Blumer, M. (1976), Polycyclic aromatic compounds in nature, Sci. Am., offer several possible explanations for the persistence of a 234,34–45, doi:10.1038/scientificamerican0376-34. low but DWH-derived PAH signal so long after capping of Borgå, K., A. T. Fisk, P. F. Hoekstra, and D. C. G. Muir (2004), Biological and chemical factors of importance in the bioaccumulation and trophic the well. First, the mesozooplankton samples in this study transfer of persistent organochlorine contaminants in Arctic marine food consisted of several species homogenized together; thus, webs, Environ. Toxicol. Chem., 23(10), 2367–2385, doi:10.1897/03-518. cross-species metabolism of PAHs may vary resulting in Camilli, R., C. M. Reddy, D. R. Yoerger, B. A. S. Van Mooy, M. V. Jakuba, J. C. Kinsey, C. P. McIntyre, S. P. Sylva, and J. V. Maloney (2010), relatively lower body burdens than seen in many single Tracking hydrocarbon plume transport and biodegradation at Deepwater species laboratory studies. Second, PAHs from DWH oil Horizon, Science, 330, 201–204, doi:10.1126/science.1195223. may have remained in the system at significant levels long Carls, M. G., J. W. Short, and J. Payne (2006), Accumulation of polycyclic after the well was capped. Lastly, mesozooplankton may aromatic hydrocarbons by Neocalanus copepods in Port Valdez, Alaska, Mar. Pollut. Bull., 52, 1480–1489, doi:10.1016/j.marpolbul.2006.05.008. have been accumulating PAHs in their bodies and passing Christensen, J. H., A. B. Hansen, G. Tomasi, J. Mortensen, and O. Andersen them across generations via eggs, which are relatively lipid (2004), Integrated methodology for forensic oil spill identification, rich compared to individual mesozooplankton. Determining Environ. Sci. Technol., 38(10), 2912–2918, doi:10.1021/es035261y. À1 Clayton, J. R., Jr., S. P. Pavlou, and N. F. Breitner (1977), Polychlorinated which of these explanations is responsible for the ng g biphenyls in coastal marine zooplankton: Bioaccumulation by equilib- observed levels of PAHs in nGOM mesozooplankton is rium partitioning, Environ. Sci. Technol., 11(7), 676–682, doi:10.1021/ beyond the scope of this study. However, our study es60130a008. Connell, D. (1997), Basic Concepts of Environmental Chemistry, Lewis, demonstrates that there was a signature distribution of Boca Raton, FL. DWH-derived PAHs in zooplankton. Cox, T., and M. Cox (1994), Multidimensional Scaling, Chapman and Hall, [15] As of August 2010, the U.S. National Incident London. Davenport, J. (1982), Oil and planktonic ecosystems, Philos. Trans. R. Soc. Command Center estimated that 26% of the residual oil B, 297, 369–384, doi:10.1098/rstb.1982.0048. could be found either on or below the surface as light sheen Diercks, A.-R., et al. (2010), Characterization of subsurface polycyclic and weathered tar balls, washed ashore or collected from the aromatic hydrocarbons at the Deepwater Horizon site, Geophys. Res. shore, or buried in sand and sediments (Federal Interagency Lett., 37, L20602, doi:10.1029/2010GL045046. Dray, S., and A. Dufour (2007), The ade4 package: Implementing the Solutions Group, Oil budget calculator, 2010). Although a duality diagram for ecologists, J. Stat. Software, 22,1–20. subsurface oil plume was identified [Camilli et al., 2010] Graham, W. M., R. H. Condon, R. H. Carmichael, I. D’Ambra, H. K. Patterson, and its composition recently elucidated [Reddy et al., 2012], L. J. Linn, and F. J. Hernandez (2010), Oil carbon entered the coastal planktonic food web during the Deepwater Horizon oil spill, Environ. Res. Lett., the ultimate fate of the oil and its presence in the ecosystem 5, 045301, doi:10.1088/1748-9326/5/4/045301. has yet to be comprehensively determined. The presence of Guzman del Proo, S. S., E. A. Chavez, F. M. L. Alatriste, S. de la Campa, this PAH signature in nGOM mesozooplankton samples in L. G. De la Cruz, R. Duadarrama, A. Guerra, S. Mille, and D. Torruco the patterns noted in our study suggests that the spatial and (1986), The impact of the Ixtox-1 oil spill on zooplankton, J. Plankton Res., 8, 557–581, doi:10.1093/plankt/8.3.557. temporal extent of the 2010 spill in the nGOM was extensive Hallanger, I. G., A. Ruus, N. A. Warner, D. Herzke, A. Evenset, M. Schoyen, and patchy. A recent study reported a depleted d13C isotopic G. W. Gabrielsen, and K. Borga (2011a), Differences between Arctic and signature in coastal mesozooplankton collected north of the Atlantic fjord systems on bioaccumulation of persistent organic pollutants in zooplankton from Svalbard, Sci. Total Environ., 409(14), 2783–2795, M-1 well showing that carbon in oil collected at depth was doi:10.1016/j.scitotenv.2011.03.015. incorporated up to two trophic levels above prokaryotic Hallanger, I. G., N. A. Warner, A. Ruus, A. Evenset, G. Christensen, hydrocarbon consumers and into the planktonic food web D. Herzke, G. W. Gabrielsen, and K. Borga (2011b), Seasonality in contaminant accumulation in Arctic marine pelagic food webs using trophic [Graham et al., 2010]. That study, combined with ours, magnification factor as a measure of bioaccumulation, Environ. Toxicol. suggests that the potential for movement of DWH-derived Chem., 30(5), 1026–1035, doi:10.1002/etc.488. carbon and PAHs to higher trophic levels, existed after the Hallanger, I. G., A. Ruus, D. Herzke, N. A. Warner, A. Evenset, E. S. well at M-1 had been capped. Heimstad, G. W. Gabrielsen, and K. Borga (2011c), Influence of season, location, and feeding strategy on bioaccumulation of halogenated organic contaminants in Arctic marine zooplankton, Environ. Toxicol. Chem., [16] Acknowledgments. The authors thank the National Science 30(1), 77–87, doi:10.1002/etc.362. Foundation RAPID grants OCE-1043249, OCE-1047736, OCE-1057461, Haney, J. F. (1988), Diel patterns of zooplankton behavior, Bull. Mar. Sci., and the captain and crew of the R/V Oceanus. 43(3), 583–603. [17] The Editor thanks two anonymous reviewers for their assistance in Heinle, D. R. (1966), Production of a calanoid copepod, Acartia tonsa,in evaluating this paper. the Patuxent River estuary, Chesapeake Sci., 7,59–74, doi:10.2307/ 1351126. Hood, K. C., L. M. Wenger, O. P. Gross, and S. C. Harrison (2002), Hydro- carbon systems analysis of the northern Gulf of Mexico: Delineation References of hydrocarbon migration pathways using seeps and seismic imaging, Aharon, P., H. P. Schwarcz, and H. H. Roberts (1997), Radiometric dating of in Surface Exploration Case Histories: Applications of Geochemistry, submarine hydrocarbon seeps in the Gulf of Mexico, Geol. Soc. Am. Bull.,

6of7 L01605 MITRA ET AL.: DWH PAHS IN GOM MESOZOOPLANKTON L01605

Magnetics, and Remote Sensing, edited by D. Schumacher and L. A. Wilde, F. D., S. C. Skrobialowski, and J. S. Hart (2010), Sampling protocol LeSchack, AAPG Stud. Geol., 48,25–40. for post-landfall Deepwater Horizon oil release, Gulf of Mexico 2010: Mitra, S., and T. S. Bianchi (2003), A preliminary assessment of polycyclic Addendum to standard USGS methods for the collection of water, aromatic hydrocarbon distribution in the lower Mississippi River benthic invertebrates, and microorganisms, U.S. Geol. Soc. Open File and Gulf of Mexico, Mar. Chem., 82, 273–288, doi:10.1016/S0304- Rep., O2010-1191, U.S. Geological Survey, Reston, VA. 4203(03)00074-4. Yunker, M. B., S. M. Backus, E. Graf Pannatier, D. S. Jeffries, and Mitra, S., T. S. Bianchi, B. A. McKee, and M. Sutula (2002), Black carbon R. W. Macdonald (2002), Sources and significance of alkane and PAH from the Mississippi River: Quantities, sources, and potential implications hydrocarbons in Canadian arctic rivers, Estuarine Coastal Shelf Sci., for the global carbon cycle, Environ. Sci. Technol., 36(11), 2296–2302, 55(1), 1–31, doi:10.1006/ecss.2001.0880. doi:10.1021/es015834b. National Research Council (2002), Oil, in The Sea: Inputs, Fates, and – S. Mitra and K. Scalise, Department of Geological Sciences, East Effects, pp. 89 117, Natl. Acad. of Sci., Washington, D. C. Carolina University, Greenville, NC 27858, USA. ([email protected]) Reddy, C. M., et al. (2012), Composition and fate of gas and oil released D. G. Kimmel, B. D. McGlaughon, and J. Snyder, Department of to the water column during the Deepwater Horizon oil spill, Proc. Natl. Biology, East Carolina University, Greenville, NC 27858, USA. Acad. Sci. U. S. A., doi:10.1073/pnas.1101242108, in press. G. L. Jahn, J. J. Pierson, and M. R. Roman, Horn Point Laboratory, Sobek, A., G. Cornelissen, P. Tiselius, and Ö. Gustafsson (2006), Passive par- Center for Environmental Science, University of Maryland, Cambridge, titioning of polychlorinated biphenyls between seawater and zooplankton, MD 21613, USA. a study comparing observed field distributions to equilibrium sorption – S. B. Brandt, Oregon Sea Grant, Oregon State University, Corvallis, OR experiments, Environ. Sci. Technol., 40(21), 6703 6708, doi:10.1021/ 97331–2131, USA. es061248c. J. P. Montoya, School of Biology, Georgia Institute of Technology, Stout, S. A., A. D. Uhler, and K. J. McCarthy (2001), A strategy and Atlanta, GA 30332, USA. methodology for defensibly correlating spilled oil to source candidates, – T. D. Lorenson, R. J. Rosenbauer, and F. L. Wong, Pacific Coastal Environ. Forensics, 2(1), 87 98, doi:10.1006/enfo.2001.0027. Marine Science Center, U.S. Geological Survey, Menlo Park, CA 94025, Venables, W. N., and B. D. Ripley (1999), Modern Applied Statistics With USA. S-PLUS, 3rd ed., Springer, New York. P. L. Campbell, Water Resources, National Research Program, U.S. Wiebe, P. H., K. H. Burt, S. H. Boyd, and A. W. Morton (1976), Multiple Geological Survey, Menlo Park, CA 94025, USA. opening-closing net and environmental sensing system for sampling zooplankton, J. Mar. Res., 34(3), 313–326.

7of7 Impact of the Deepwater Horizon oil spill on a SPECIAL FEATURE deep-water coral community in the Gulf of Mexico

Helen K. Whitea,1, Pen-Yuan Hsingb, Walter Choc, Timothy M. Shankc, Erik E. Cordesd, Andrea M. Quattrinid, Robert K. Nelsone, Richard Camillif, Amanda W. J. Demopoulosg, Christopher R. Germanh, James M. Brooksi, Harry H. Robertsj, William Sheddk, Christopher M. Reddye, and Charles R. Fisherb aDepartment of Chemistry, Haverford College, Haverford, PA 19041; bDepartment of Biology, Pennsylvania State University, University Park, PA 16802; cBiology Department, eDepartment of Marine Chemistry and Geochemistry, fApplied Ocean Physics and Engineering, and hDepartment of Geology and d g THE DEEPWATER HORIZON SCIENCE APPLICATIONS IN Geophysics, Woods Hole Oceanographic Institution, Woods Hole, MA 02543; Biology Department, Temple University, Philadelphia, PA 19122; Southeast OIL SPILL SPECIAL FEATURE Ecological Science Center, US Geological Survey, Gainesville, FL 32653; iTDI-Brooks International Inc., College Station, TX 77845; jDepartment of Oceanography and Coastal Sciences, Coastal Studies Institute, Louisiana State University, Baton Rouge, LA 70803; and kBureau of Ocean Energy Management, US Department of the Interior, New Orleans, LA 70115

Edited by Paul G. Falkowski, Rutgers, State University of New Jersey, New Brunswick, NJ, and approved February 28, 2012 (received for review November 1, 2011) To assess the potential impact of the Deepwater Horizon oil spill on (1, 2). Numerous coral colonies were discovered at this location offshore ecosystems, 11 sites hosting deep-water coral communities and many were partially or completely covered in a brown, floc- were examined 3 to 4 mo after the well was capped. Healthy coral culent material (hereafter referred to as floc). They showed signs communities were observed at all sites >20 km from the Macondo of recent and ongoing tissue damage (Fig. 2) not observed else- well, including seven sites previously visited in September 2009, where at this time (Fig. 1) or in the previous 10 y of baseline where the corals and communities appeared unchanged. However, studies in the Gulf of Mexico (GoM) (3–5). Between December 8 at one site 11 km southwest of the Macondo well, coral colonies and 14, 2010 additional surveys were performed with the deep presented widespread signs of stress, including varying degrees of submergence vehicle (DSV) Alvin at MC 294 and a newly dis- tissue loss, sclerite enlargement, excess mucous production, bleached covered site 22 km to the ESE of the Macondo well in MC 388 commensal ophiuroids, and covering by brown flocculent material (1,850 m depth). Visible signs of recent impact or stress were not SCIENCES (floc). On the basis of these criteria the level of impact to individual evident in the corals imaged at MC 388. ENVIRONMENTAL colonies was ranked from 0 (least impact) to 4 (greatest impact). Of To determine whether the cause of the overall decrease in the 43 corals imaged at that site, 46% exhibited evidence of impact coral health at MC 294 was related to the Deepwater Horizon oil on more than half of the colony, whereas nearly a quarter of all of the spill, the floc covering the corals and nearby sediment was ex- > corals showed impact to 90% of the colony. Additionally, 53% of amined for the presence of petroleum hydrocarbons originating ’ these corals ophiuroid associates displayed abnormal color and/or from the Macondo well. Determining the source of petroleum attachment posture. Analysis of hopanoid petroleum biomarkers iso- hydrocarbons in these samples posed a significant challenge. The fl lated from the oc provides strong evidence that this material con- complexity of the petrogeochemical signatures in the GoM en- tained oil from the Macondo well. The presence of recently damaged vironment is considerable (6). Specific crude oils can be differ- and deceased corals beneath the path of a previously documented entiated from their source rock groups using biomarkers (mo- plume emanating from the Macondo well provides compelling evi- fi lecular fossils), which are highly resistant to abiotic and biotic dence that the oil impacted deep-water ecosystems. Our ndings processes and have been invaluable tools for characterizing and underscore the unprecedented nature of the spill in terms of its mag- fingerprinting crude oils (7). For example, sterane biomarkers are nitude, release at depth, and impact to deep-water ecosystems. derived primarily from marine phytoplankton and vary depending on geologic age. Hopanes, which are another class of biomarkers, hopane | sterane | Paramuricea | sediment can be used individually or in concert with sterane distributions to provide even greater certainty in characterizing oils (7). The use etween October 15 and November 1, 2010, approximately 6 of biomarkers by the petroleum industry and subsequently in Bmonths after the Deepwater Horizon blowout and 3 months environmental forensics has, however, been performed in much after the Macondo well was capped, nine sites hosting deep-water different environments than the Deepwater Horizon spill, where coral communities were examined with the remotely operated oil and gas at 105 °C were released at pressure into 5 °C seawater vehicle (ROV) Jason II. This effort was part of an ongoing study at ∼1,400 m depth (2). We used traditional 1D gas chromatog- funded by the Bureau of Ocean Energy Management (BOEM) raphy (GC) and comprehensive two-dimensional gas chroma- ’ and the National Oceanic and Atmospheric Administration s tography (GC×GC, as in refs. 8, 9 and 10) to analyze floc and Ocean Exploration and Research program. These sites, located sediment samples from MC 294. These samples were compared between 93.60 °W and 87.31 °W and between −27.42 °N and −29.16 °N (Fig. S1), were >20 km from the Macondo well, ranged in depth from 290 to 2600 m, and hosted coral communities in- Author contributions: H.K.W., T.M.S., E.E.C., A.W.J.D., C.R.G., and C.R.F. designed research; cluding scleractinian, gorgonian, and antipatharian corals. At H.K.W., P.-Y.H., W.C., T.M.S., E.E.C., A.M.Q., R.K.N., R.C., A.W.J.D., C.R.G., C.M.R., and C.R.F. these sites, no visible evidence of impact to the corals and asso- performed research; J.M.B., H.H.R., and W.S. contributed new reagents/analytic tools; ciated communities was observed (Fig. 1). However, on Novem- H.K.W., P.-Y.H., W.C., T.M.S., E.E.C., A.M.Q., R.K.N., C.M.R., and C.R.F. analyzed data; and H.K.W. and C.R.F. wrote the paper. ber 2, 2010, the ROV Jason II investigated an area in lease blocks fl Mississippi Canyon (MC) 294 and 338, 11 km to the SW of the The authors declare no con ict of interest. site of the Deepwater Horizon spill. This area was explored be- This article is a PNAS Direct Submission. cause 3D seismic reflectivity data (Fig. S1) suggested there was Data deposition: The octocoral and ophiuroid sequences reported in this paper have been deposited in the GenBank database (accession nos. JQ241244–52, JQ411462–9 and a strong likelihood of hard grounds, and hence likely coral sub- JQ771615–JQ771617) and all images have been submitted to the US National Oceano- strate present. Its location (28.40N, 88.29W, 1,370 m) also placed graphic Data Center (accession no. 0084636). it in the path of a 100-m-thick deep-water plume of neutrally 1To whom correspondene should be addressed. E-mail: [email protected]. buoyant water enriched with petroleum hydrocarbons from the This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. Macondo well that was documented at 1,100 m in June 2010 1073/pnas.1118029109/-/DCSupplemental. www.pnas.org/cgi/doi/10.1073/pnas.1118029109 PNAS Early Edition | 1of6 A

0 50 100 Kilometers

Fig. 1. Healthy deep-water coral communities observed in November 2010 from various sites >20 km from the Macondo well (shown as a star on map). (A) Paramuricea sp. type E and Asteroschema sp. at 360 m depth in Garden Banks (GB) 299; (B) Paramuricea sp. type E and Asteroschema sp. at 440 m depth in Mississippi Canyon (MC) 751; (C) Paramuricea sp. type A and Eumunida picta at 530 m depth in Green Canyon (GC) 354; (D) Paramuricea sp. type E and Asteroschema sp., along with a brisingid basket star at 360 m depth in Viosca Knoll (VK) 906; (E) P. biscaya and A. clavigerum at 2,300 m depth in Desoto Canyon (DC) 673. Phylogenetic species identifications of corals and associated ophiuroids are given in Fig. S2. with oil collected from directly above the broken riser pipe at the indications of stress: bare skeleton above the basal region, loose Macondo well (2) and field samples from surface water and salt tissue or heavy mucous hanging from the skeleton, and/or cov- marshes in areas oiled by the Deepwater Horizon spill. erage with brown flocculent material (Fig. 3). Eighty-six percent Here we report on the analyses of the visible impact to the of the coral colonies imaged in the central area exhibited signs of gorgonian corals and coral associates at MC 294 based on in situ impact. Forty-six percent exhibited impact to at least 50% of the video imagery, shipboard microscopic analyses, and petroleum colony (impact level 3 or 4), and 23% of the colonies sustained biomarker analysis of the floc adherent to the coral. In addition, impact to more than 90% of the colony (impact level 4). we compare the petroleum hydrocarbon content and biomarkers Between the November and December 2010 research cruises, with the surrounding surface and subsurface sediments and changes in condition were assessed for all corals or portions of compare the condition of the corals and associates between colonies for which high-resolution imagery was available from November and December 2010 visits. similar perspectives. Although differences in camera placement Results and Discussion Gorgonian and other corals present at MC 294 are pre- November December dominantly found in a central area 10 × 12 m in extent, composed 2010 2010 of two adjacent carbonate slabs. Scattered boulders surround this region over an area of 50 × 50 m, and some of the isolated boulders host one or two additional coral colonies. The majority of the colonial corals were Paramuricea biscaya, with one or two colonies of Swiftia pallida, Paragorgia regalis, Acanthogorgia aspera, and Clavularia rudis (Fig. S2). The majority of these colonies exhibited signs of stress response, including excessive mucous production and retracted polyps, which have been observed in corals experimentally exposed to crude oil (11). Impact to the corals was quantified from close-up images (<1 m away) for 43 of the 58 coral colonies identified in the central area (Fig. S3) (not fl all of the corals could be approached for close-up imaging with Fig. 2. Impacted corals at MC 294. Brown occulent material and tissue loss is observed on the larger coral, A10, in November and December 2010. Al- ROV Jason II or DSV Alvin without disturbing other colonies). though there is no evidence of recovery on A10, note that the tips of some The level of impact to individual colonies was ranked from branches that were living in November were still living in December 0 (least impact) to 4 (greatest impact) according to the percent- (arrows). Coral A14 in the red box was the only colony showing apparent age of the colony exhibiting one or more of the following visual signs of recovery from coverage by the floc between visits.

2of6 | www.pnas.org/cgi/doi/10.1073/pnas.1118029109 White et al. 14 30% ionization detector (GC-FID; Table 1). An unresolved complex SPECIAL FEATURE 12 mixture (UCM) with n-alkane carbon range of C15–C42 indicates 23% 23% 10 the presence of weathered petroleum (e.g., ref. 8; Table 1). Slight 8 variations in UCM carbon range and distributions of n-alkanes 14% 6 among samples showed no consistent relationship to the pure 9% 4 Macondo Well oil (described in ref. 2; Table 1). Rather, it is evident that the n-alkanes in the samples represent input from Number of corals 2 fi 0 a mixture of sources such as plants, bio lms, and differentially 0 1 2 3 4 weathered subsurface hydrocarbons, including some that may Impact ranking have come from natural seeps. Acoustic mapping cruises per- Fig. 3. Impact assessment for coral colonies (n = 43) where high-quality formed from late May to August 2010 mapped several natural THE DEEPWATER HORIZON SCIENCE APPLICATIONS IN images could be obtained from at least the November or December 2010 gas seeps in near proximity to both the Macondo well and the OIL SPILL SPECIAL FEATURE cruise. The levels of impact are ranked according to the proportion of a coral sample sites presented here, which could provide additional exhibiting obvious tissue damage, bare skeleton above basal region, or sources of subsurface hydrocarbons (14). covered by brown flocculation: rank 0 (0–1%), rank 1 (<10%), rank 2 (10– Polycyclic aromatic hydrocarbon (PAH) distributions from 50%), rank 3 (50–90%), rank 4 (>90%). Numbers above bars are percentages coral E3 and sediment sample 4664 0–2 cm show good corre- of corals in a rating relative to all assessed colonies. spondence to Macondo well oil, with similar relative abundances of naphthalene, phenanthrene, and their alkylated derivatives as well as dibenzothiophenes, benzo[a]anthracene, and chrysene. on the two underwater vehicles, lighting, and quality of images The remaining coral samples are inconclusive owing to the small limited the size of this data set to 18 colonies, neither progression quantity of sample available for analysis, as well as the fact that of the visible damage nor clear evidence of recovery or growth these samples have been extensively weathered, as evidenced by was apparent in the majority of corals. Possible recovery was the dominance of biodegradation-resistant chrysene in all extracts. noted for one colony (A14, highlighted by box in Fig. 2). The fi fl Petroleum systems in the GoM do not display signi cant dif- relatively light covering of oc over more than 50% (impact ferences in the presence or absence of specific biomarkers; instead, level 3) of this colony in November was ranked as less than 10% differences in the relative amounts of biomarkers present have impacted (impact level 1) by the time it was revisited in De-

previously allowed sources to be determined (15, 16). Analysis of SCIENCES cember, when extended polyps were visible in areas that had been

biomarkers such as hopanes is critical because these compounds ENVIRONMENTAL fl partially covered with oc in November. are more resistant to biodegradation and water washing than Sampling of a P. biscaya coral (E3) in December enabled mi- n-alkanes and PAHs and provide insight into petroleum source fl croscopic analysis to be made after removal of the oc. Varying determination (as in ref. 17). At the Macondo well, oil sampled degrees of tissue loss and sclerite enlargement were observed from above the broken riser pipe (2) contains abundant hopa- (Fig. S4). The skeleton was bare and entirely devoid of tissue at noids, diasteranes, and steranes (Fig. S7). Hopanoid biomarker the base and along the main axis of the colony. At increasing ratios have been calculated for comparison with coral and sedi- distances from the basal point of attachment, less extensive tissue ment samples, as well a reference surface water (S1) and two loss resulted in the exposure of the calcite skeletal elements that reference coastal water (M1 and M2) samples (shown in Fig. 1 and are normally embedded in the tissue layers and coenenchyme. described in ref. 18). These reference samples represent Macondo These sclerites were still in their normal form of a polyp but well oil that has undergone vertical transport from the seabed to appeared enlarged. The localized alteration of growth form, in- the ocean surface (∼1,400 m) and subsequent lateral dispersion cluding excessive secretion of gorgonin and sclerite production to over ranges of 1–175 km, respectively (Table 1). form granuloma-like structures, has previously been observed in Comparison of the hopanoid portion of the GC×GC chro- gorgonians as an acute stress response (12, 13). Near the tips of matographic plane for the Macondo well oil to the S1 and M1 some branches, which were not covered by the floc in situ, a few samples indicates a high degree of similarity (Fig. S8 A–C). This polyps on this coral appeared normal. similarity is also seen in the floc from coral B8 (Fig. S8D) and in Coral associates at MC 294 included 13 actinarian anemones the surface sediment sample taken in the immediate vicinity of and 78 Asteroschema clavigerum (a symbiotic ophiuroid). Of the 52 the corals (core 4664 0–2 cm; Fig. S8E). Slight but significant individual corals examined for coral associates, 25% hosted none, differences in hopanoid biomarker ratios are observed, by con- 2% hosted actinarian anemones, and 73% hosted A. clavigerum, trast, both in comparable core-top sediments collected away with 70% of the ophiuroids present on P. biscaya, 18% on the from the impacted corals at the MC 294 site (core 4662 0–2 cm; single individual of P. regalis, and 12% on A. aspera. A. clavigerum Table 1 and Fig. S8F) and at greater depths (2–5 cm and 5–10 is typically tan to red in color (Fig. 1); however, at this site only cm; Table 1) in the core 4664 sediments. Further, the concen- 47% were tan to red, whereas 44% had distinctly white arms (Fig. trations of oil present in the uppermost sediments of core 4664 2), and 9% (all hosted by P. biscaya), were bleached almost entirely (0–2 cm) are much higher (9.25 mg/g; Table 1) than the deeper white. In November, 27% of the ophiuroids displayed behaviors sediments (2–5 cm and 5–10 cm) in the same core, which range other than their normal attached posture of arms tightly coiled in concentration from 0.02 to 0.03 mg/g (Table 1). They are also around their coral host (Fig. 1). Between visits, 13% of the higher than the oil concentrations observed in surface sediments ophiuroids transitioned from tightly to loosely coiled (i.e., Fig. 2). (0–2 cm) collected away from the impacted corals at the MC 294 Two ophiuroids (Fig. S5) transitioned from tightly coiled to a site (3.46 mg/g; Table 1), where a bimodal n-alkane distribution posture with splayed out arms, a previously undocumented be- indicative of inputs from mixed sources is observed. Significant havior in this species. variations in sediment oil concentrations have been previously The floc samples collected (>72 μm in size) were removed documented in the GoM, particularly in areas of known natural from the surface of the corals in situ and when filtered were oil seepage such as Green Canyon, where oil concentrations may found to contain dead coral polyp fragments, detached sclerites, be as high as 39.0 mg/g (19). The oil concentration and bio- and small brown droplets (Fig. S6). Solvent extracts of all of the marker data from sediments collected away from the impacted floc examined were dominated by C16 and C18 saturated and corals and sediments at depth at MC294, are, however most unsaturated fatty acids and cholesterol, which are dominant consistent with long-term background inputs of oil derived from lipids in biological tissue. Petroleum residues were also present petroleum sources that are quite distinct to that present in the and quantified via 1D gas chromatography coupled to a flame most superficial (hence, recent) core-top sediments and floc

White et al. PNAS Early Edition | 3of6 Table 1. Oil content, alkane distribution, and hopanoid biomarker ratios of brown ﬂocculent material and sediment samples compared with Macondo Well oil Oil content*

Oil content* (mg/g dry weight UCM n-alkane C17-n-alkane/ C18-n-alkane/ Carbon Preference † Sample (mg/g extract) sediment) carbon range pristane phytane Index (CPI) Ts/(Ts+Tm) C29-Ts/NH

Source oil ‡ Macondo well NA§ NA NA 1.71 2.33 0.86 0.59 0.49 { Surface water samples

1 km from well (S1) NA NA C12–C40 2.21 2.66 0.81 0.59 0.50

175 km from well (M1) NA NA C16–C40 1.89 2.72 0.83 0.59 0.50

175 km from well (M2) NA NA C16–C40 1.89 2.72 0.83 0.58 0.51 Flocculent material samples jj MC 294 coral (B8) 310 ND C15–C40 2.42 2.69 1.15 0.60 0.43

MC 294 coral (F6) 8.0 ND C17–C34 2.41 2.21 1.09 0.58 0.48

MC 294 coral (A5) 74 ND C16–C41 2.43 1.87 0.38 0.59 0.45

MC 294 coral (E3) 73 ND C15–C42 1.34 2.28 1.22 0.58 0.48 Sediment samples

MC 294 4662 0–2 cm 630 3.46 C15–C37 0.42 0.47 1.12 0.57 0.46

MC 294 4664 0–2 cm 570 9.25 C17–C42 0.44 0.29 0.99 0.59 0.50

MC 294 4664 2–5cm 68 0.03 C16–C42 0.56 0.81 1.31 0.52 0.38

MC 294 4664 5–10 cm 120 0.02 C11–C42 0.80 1.21 1.37 0.54 0.42

Abbreviations for biomarkers: C29-Ts,18α(H),21β(H)-30-norneohopane; NH, 17α(H),21β(H)-30-norhopane; Tm, 17α(H)-22,29,30-trinorhopane; Ts, 18α(H)- 22,29,30-trinorneohopane. *Oil content was calculated by integration of the UCM observed via GC-FID. † CPI = Σ(odd numbered alkane abundances from n-C23 to n-C35)/Σ(even numbered alkane abundances from n-C22 to n-C34). ‡Described in ref. 2. §Not applicable to sample as pure oil was collected. { Described in ref. 18. jj Not determined due to collection protocol of flocculent onto filters, which did not allow for dry weights of the flocculent material to be taken post collection and before extraction. samples collected from site MC 294. Similarly, a comparison of cally inhabit areas exposed to a moderate current regime (28). the sterane portion of the GC×GC chromatographic plane for The presence of a deep-water coral community dominated by the Macondo well oil and floc from the coral samples also shows recently impacted, visibly unhealthy, and recently dead individ- significant differences, particularly in the relative distributions of uals (as evidenced by skeletons free of encrusting organisms), DiaC29βα-20S, C27αββ-20R, and C27αββ-20S in steranes (e.g., B8, together with ophiuroid symbionts with unhealthy color and Fig. S9B). Although preferential loss of steranes and diasteranes atypical posture, provides evidence of a recent waterborne im- relative to hopanoids would not be expected from traditional pact. Although the spatial and temporal proximity of this impact biodegradation sequences, this trend has been observed pre- to the Deepwater Horizon oil spill might be coincidental, the viously for oil that undergoes severe weathering in energetic and normal longevity of deep-water corals and the lack of visual ev- aerobic conditions (20). This could result from either biodeg- idence of impact to deep-water corals elsewhere in the GoM radation or chemical and physical processes arising from the pre- suggest that this may not be the case. Importantly, even though cipitation of the wax component of oil at the low temperatures there are multiple inputs of oil to the GoM, the use of hopanoid present in the deep GoM (21, 22). Wax formation may have biomarker compositions and ratios in the floc collected from the resulted from turbulent mixing of the well’s hot source-jet fluids surface of corals allows us to establish a connection to the oil spill with the surrounding cold seawater, fractionating constituent even though other biomarkers for characterizing oil in these hydrocarbons according to their molecular characteristics (1, 2). environments (e.g., PAHs and sterane biomarker ratios) are af- Nevertheless, the data from the hopanoids, which have a greater fected by severe weathering (20) and, hence, are not robust under fidelity, confirm the presence of Macondo well oil in the floc and the conditions of this spill. surrounding surface sediment samples. The constant, albeit rel- The data suggest the Deepwater Horizon oil spill impacted atively low level, input of hydrocarbons from natural seepage in a community of deep-water corals near the Macondo well. The the GoM may also complicate these biomarker ratios (14). numerous apparently healthy deep-water coral communities in other parts of the GoM may indicate that the localized impact in Conclusions MC 294 found to date, is not part of a much larger, acute, GoM- Observations of recently damaged corals and the presence of wide event. However, life in deep-water coral ecosystems is Macondo well oil on corals indicates impact at a depth of 1,370 known to operate at a slow pace. Consequently it is too early to m, 11 km from the site of the blowout. This finding provides fully evaluate the footprint and long-term effects of acute and insight into the extent of the impact of the spill, which is signif- subacute exposure to potential waterborne contaminants result- icantly complicated by physical mixing processes (23) and frac- ing from the Deepwater Horizon oil spill. tionation of the oil constituents (24). Because deep-water corals Materials and Methods are sessile and release mucous that may trap material from the water column, these corals may provide a more sensitive indicator Discovery. Areas for exploration were chosen according to examination of 3D seismic data in the BOEM database. Areas of high reflectivity and bathymetric of the impact from petroleum hydrocarbons than marine sedi- relief were targeted for visual examination, and during the ROV dive, ment cores and may record impacts from water masses passing onboard sonar was used to find exposed carbonates that might host corals. through a community, even if no deposition to the sediment occurs. Deep-water colonial corals exhibit extreme longevity as Image Analyses. A down-looking mosaic (as in ref. 29) was constructed from sessile adults (hundreds to thousands of years; 25–27) and typi- 379 partially overlapping images, taken 3 m above the seafloor using

4of6 | www.pnas.org/cgi/doi/10.1073/pnas.1118029109 White et al. a Nikon E995 camera in pressure housing mounted on the ROV Jason II. genetic comparison. DNA was extracted from frozen or preserved (95%

Individual coral colonies were labeled (Fig. S3). Close-up images of individual ethanol) ophiuroids living on MC 294 corals using the Qiagen DNeasy kit. SPECIAL FEATURE corals from a side-looking perspective were obtained from frame grabs us- A fragment of the mitochondrial 16S rRNA (16S) gene was amplified with ing a dedicated NDSF/AIVL Adimec 2000 HDTV digital video camera moun- the universal primers 16SarL and 16SbrH, sequenced, and aligned using ted on the ROV Jason II vehicle frame and the starboard manipulator of DSV SEQUENCHER 4.8 (Gene Codes Corporation) and previously described Alvin. Close-up imagery was used for assessment of impact to all corals that methods (34). Phylogenetic inference (and subsequent species identification) could be approached by ROV Jason II or DSV Alvin without damaging other was conducted using parsimony (PAUP* 4.0b10), neighboring-joining dis- corals. Bare skeleton above the coral’s basal region, obviously damaged tance-based and Bayesian approaches associated with Geneious (version 5). tissue (strands of mucous or loose tissue hanging from the skeleton), and fl areas covered by oc were scored as impacted. Levels of impact were Sediment and Floc Collection. Floc was collected at depth through 1.5-m-long fi broadly binned into ve categories according to the percentage of the im- precleaned tubing into a 4-L carboy, where it was collected onto two 15-cm – aged portion of the colony showing impact: rank 0, 0 1% of the colony diameter precombusted glass fiber filters (GF/C, >1 μm) mounted between – – THE DEEPWATER HORIZON impacted; rank 1, 1 10% of the colony impacted; rank 2, 10 50% of the SCIENCE APPLICATIONS IN two layers of Nytex (63 μm). The majority of the floc did not sorb to the filters OIL SPILL SPECIAL FEATURE – > colony impacted; rank 3, 50 90% of the colony impacted; or rank 4, 90% of and remained suspended in the seawater collected in the carboy. Once the colony impacted. In three cases in which the ranking category changed onboard ship, this material was immediately filtered onto 47-mm-diameter between the November and December visits, reexamination of images and precombusted GF/F filters (0.7 μm), which were then placed in combusted foil fi the original video did not substantiate signi cant changes in the corals, and and frozen at –20 °C before further analysis. Push cores (6.35 cm diameter) the higher-quality images obtained from DSV Alvin in December were used were used to collect sediment samples. Immediately after reaching the ship’s for the overall rankings (Fig. 3). All images can be accessed through the US deck, the cores were sectioned as 0–2cm,2–5 cm, and 5–10 cm intervals into National Oceanographic Data Center (accession no. 0084636). combusted glass jars and then frozen at −20 °C until analysis. Macondo well reference oil was collected directly above the well on June 21, 2010 (as in ref. Coral and Invertebrate Associate Collection. Corals and their associate 2). This sample is the reference Macondo well oil with which samples are ophiuroids were collected by the ROV Jason II and DSV Alvin in October and compared throughout the study. Other samples (described in ref. 18) were fi December 2010, respectively. The manipulator claws were modi ed with also obtained on May 31, 2010 from a saltmarsh ≈175 km from the spill near a cutting blade to aid in the collection of host corals and attached ophiuroid Cocodrie Louisiana (29.29 °N; −90.49 °W) as a 2-cm-diameter droplet of sur- associates. Individuals were collected into temperature-insulated bioboxes face oil (M1) and from a scraping of Spartina alterniflora saltmarsh grass on the sea floor and processed immediately after recovery of the vehicles. taken within meters of the droplet (M2). Another sample (S1) was collected Approximately 2 to 3 cm of a coral branch or arms of individual ophiuroids on June 20, 2010 with the R/V Endeavor from a 1-cm-thick layer of oil floating were subsampled and frozen at −80 °C or in 70% ethanol for shore-based on the surface water at 28.74 °N, 88.38 °W (18). All samples were placed in morphological and genetic analyses. Voucher specimens were either pre- SCIENCES combusted glass jars and frozen until further analysis. served in 95% ethanol or dried. ENVIRONMENTAL Oil Analysis. Samples were solvent extracted and purified with fully activated Microscopic Examination. Tissue necrosis and the presence of bare skeleton silica gel. Extracts were analyzed for hydrocarbons via GC-FID, gas chroma- were documented on a Leica S6D microscope with an attached Nikon tography–mass spectrometry (GC-MS), and comprehensive GC×GC. For D300 camera. quantification and identification, GC×GC was coupled to an FID (GC×GC- FID), and identities of biomarkers were confirmed by coupling with MS Octocoral Identification. Octocorals were identified to the lowest possible (GC×GC-MS). SI Materials and Methods provides a complete discussion of taxon using molecular barcodes and morphological characters (following these analyses. refs. 30–32). DNA was extracted from frozen or preserved (95% ethanol) specimens using the Qiagen DNeasy kit. The 5′ end of the mitochondrial msh ACKNOWLEDGMENTS. We thank the crew and captains of the R/V Atlantis gene and the COI+igr region were PCR amplified (33). Sequences were and R/V Ron Brown; the pilots and crew of DSV Alvin and ROV Jason II;and edited, combined with related sequences from GenBank, and aligned by J. Abbassi, C. Carmichael, O. Chegwidden, D. Cowart, C. Doughty, T. Ender- ClustalW, resulting in a 1,430-bp region. Bayesian phylogenetic inference lein, P. Etnoyer, J. Frometa, K. Halanych, K. Reuter, M. Rittinghouse, A. Sen, was conducted using the GTR+G+I model (MrBayes v3; number of gen- C. Sheline, K. Stamler and J. Thoma for their contributions to this work. erations = 2,000,000; sample frequency = 100; burnin = 5,000). Because the This work was supported by Bureau of Ocean Energy Management Contract COI+igr region has not been previously amplified for many octocorals, these 1435-01-05-CT-39187 (to TDI-Brooks), the National Oceanic and Atmospheric fi regions were coded as missing data for the appropriate GenBank specimens. Administration Of ce of Ocean Exploration, the Census of Marine Life (ChEss Project), USGS-Terrestrial, Water, and Marine Environments Program through fi fi the BOEM/Outer Continental Shelf, Northeast Gulf of Mexico Deep Offshore Ophiuroid Identi cation. Ophiuroids were identi ed to the lowest possible Reef Ecology, Lophelia II Study and National Science Foundation RAPID taxon using morphological and genetic data. Morphological examination of Grants OCE-1045131 (to H.K.W.), OCE-1045083 and OCE-1064041 (to C.R.F.), type and voucher specimens was conducted at the Smithsonian Institution OCE-1043976 (to C.M.R.), OCE-1045025 (to R.C.), OCE-1045329 (to T.M.S.), (Washington, DC), and DNA from voucher specimens was obtained for OCE-1044289 (to C.R.G.), and OCE-1045079 (to E.E.C.).

1. Camilli R, et al. (2010) Tracking hydrocarbon plume transport and biodegradation at 11. Ducklow HW, Mitchell R (1979) Bacterial populations and adaptations in the mucus Deepwater Horizon. Science 330:201–204. layers on living corals. Limnol Oceanogr 24:715–725. 2. Reddy CM, et al. (2011) Composition and fate of gas and oil released to the water 12. Goldberg WM, Makemson JC, Colley SB (1984) Entocladia endozoica sp nov, a path- column during the Deepwater Horizon oil spill. Proc Natl Acad Sci USA, 10.1073/ ogenic Chlorophyte: Structure, life history, physiology, and effects on its coral host. – pnas.1101242108. Biol Bull 166:368 383. 3. Schroeder WW, et al. (2005) Cold-Water Corals and Ecosystems, eds Freiwald A, 13. Petes LE, Harvell CD, Peters EC, Webb MAH, Mullen KM (2003) Pathogens compromise Roberts JM (Springer, Heidelberg), pp 297–307. reproduction and induce melanization in Caribbean sea fans. Mar Ecol Prog Ser 264: – 4. Cordes EE, et al. (2008) Coral communities of the deep Gulf of Mexico. Deep Sea Res 167 171. 14. Weber TC, De Rovertis A, Greenaway SF, Smith S, Mayer L, Rice G (2011) Estimating oil Part I Oceanogr Res Pap 55:777–787. concentration and flow rate with calibrated vessel-mounted acoustic echo sounders. 5. Lessard-Pilon SA, Podowski EL, Cordes EE, Fisher CR (2010) Megafauna community Proc Natl Acad Sci USA, 10.1073/pnas.1108771108. composition associated with Lophelia pertusa colonies in the Gulf of Mexico. Deep 15. Sassen R, Sweet ST, DeFreitas DA, Morelos JA, Milkov AV (2001) Gas hydrate and Sea Res Part II Top Stud Oceanogr 57:1882–1890. crude oil from the Mississippi Fan Foldbelt, downdip Gulf of Mexico Salt Basin: Sig- 6. Cole GA, et al. (2001) Constraining source and charge risk in deepwater areas. World nificance to petroleum system. Org Geochem 32:999–1008. Oil 222:69–77. 16. Warburton GA, Zumberge JE (1983) Determination of petroleum sterane distribution 7. Peters KE, Walters CC, Moldowan JM (2005) The Biomarker Guide: Biomarkers and by mass spectrometry with selective metastable ion monitoring. Anal Chem 55: Isotopes in Petroleum Exploration and Earth History (Cambridge Univ Press, Cam- 123–126. bridge, UK), Vol 2. 17. Cometa PA, Rafalskaa JK, Brooks JM (1993) Sterane and triterpane patterns as di- 8. Frysinger GS, Gaines RB, Xu L, Reddy CM (2003) Resolving the unresolved complex agnostic tools in the mapping of oils, condensates and source rocks of the Gulf of mixture in petroleum-contaminated sediments. Environ Sci Technol 37:1653–1662. Mexico region. Org Geochem 20:1265–1296. 9. Reddy CM, et al. (2002) The West Falmouth oil spill after thirty years: The persistence 18. Carmichael CA, et al. (2012) Floating oil-covered debris from Deepwater Horizon: of petroleum hydrocarbons in marsh sediments. Environ Sci Technol 36:4754–4760. Identification and application. Environ Res Lett 7:015301. 10. Nelson RK, et al. (2006) Tracking the weathering of an oil spill with comprehensive 19. Sassen R, et al. (1994) Organic geochemistry of sediments from chemosythetic com- two-dimensional gas chromatography. Environ Forensics 7:33–44. munities, Gulf of Mexico slope. Geo-Mar Lett 14:110–119.

White et al. PNAS Early Edition | 5of6 20. Wang Z, Fingas M, Owens EH, Sigouin L, Brown CE (2001) Long-term fate and per- 27. Prouty NG, Roark EB, Buster NA, Ross SW (2011) Growth-rate and age distribution of sistence of the spilled metula oil in a marine salt marsh environment degradation of deep-sea black corals in the Gulf of Mexico. Mar Ecol Prog Ser 423:101–115. petroleum biomarkers. J Chromatogr A 926:275–290. 28. Roberts JM, Wheeler AJ, Freiwald A (2006) Reefs of the deep: The biology and ge- 21. Strom-Kristiansen T, Lewis A, Daling PS, Hokstad JN, Singsaas I (1997) Weathering and ology of cold-water coral ecosystems. Science 312:543–547. dispersion of naphthenic, asphaltenic, and waxy crude oils. Int Oil Spill Conf631–636. 29. Pizarro O, Singh H (2003) Toward large-area mosaicing for underwater scientific 22. Daling PS, Faksness LG, Hansen AB, Stout SA (2002) Improved and standardized applications. IEEE J Oceanic Eng 28:651–672. methodology for oil spill fingerprinting. Environ Forensics 3:263–278. 30. Madsen FJ (1970) Remarks on Swiftia rosea (Grieg) and related species (Coelenterata, 23. Valentine DL, et al. (2012) Dynamic autoinoculation and the microbial ecology of a deep Gorgonaria). Steenstrupia (Cph) 1:1–10. water hydrocarbon irruption. Proc Natl Acad Sci USA, 10.1073/pnas.1108820109. 31. Grasshoff M (1977) Die Gorgonarien des ostlichen Nordatlantik und des Mitteleeres III 24. Ryerson TB, et al. (2012) Chemical data quantify Deepwater Horizon hydrocarbon Die Familie Paramuriceidae (Cnidaria, Anthozoa). Meteor Forsch-Ergebnisse 27:5–75. flow rate and environmental distribution. Proc Natl Acad Sci USA, 10.1073/ 32. Sanchez J (2005) Systematics of the bubblegum corals (Cnidaria: Octocorallia: Para- pnas.1110564109. gorgiidae) with description of new species from New Zealand and the Eastern Pacific. 25. Andrews AH, et al. (2002) Age, growth and radiometric age validation of a deep-sea, Zootaxa 1014:1–72. habitat-forming gorgonian (Primnoa resedaeformis) from the Gulf of Alaska. Hy- 33. McFadden CS, et al. (2011) Limitations of mitochondrial gene barcoding in Octocor- drobiologia 471:101–110. allia. Mol Ecol Resour 11:19–31. 26. Roark EB, Guilderson TP, Dunbar RB, Fallon SJ, Mucciarone DA (2009) Extreme lon- 34. Cho W, Shank TM (2010) Incongruent patterns of genetic connectivity among four gevity in proteinaceous deep-sea corals. Proc Natl Acad Sci USA 106:5204–5208. ophiuroid species on North Atlantic Seamounts. Mar Ecol (Berl) 31:121–143.

6of6 | www.pnas.org/cgi/doi/10.1073/pnas.1118029109 White et al. Environmental Toxicology and Chemistry, Vol. 24, No. 5, pp. 1219±1227, 2005 ᭧ 2005 SETAC Printed in the USA 0730-7268/05 $12.00 ϩ .00

COMPARATIVE TOXICITY OF TWO OIL DISPERSANTS, SUPERDISPERSANT-25 AND COREXIT 9527, TO A RANGE OF COASTAL SPECIES

ALAN SCARLETT,² TAMARA S. GALLOWAY,*³ MARTIN CANTY,³ EMMA L. SMITH,² JOHANNA NILSSON,³ and STEVEN J. ROWLAND² ²School of Earth, Ocean, and Environmental Sciences, and ³School of Biological Sciences, University of Plymouth, Drake Circus, Plymouth, PL4 8AA United Kingdom

(Received 2 July 2004; Accepted 5 November 2004)

AbstractÐThe acute toxicity of the oil dispersant Corexit 9527 reported in the literature is highly variable. No peer-reviewed data exist for Superdispersant-25 (SD-25). This study compares the toxicity of the two dispersants to a range of marine species representing different phyla occupying a wide range of niches: The marine sediment-dwelling amphipod Corophium volutator (Pallas), the common mussel Mytilus edulis (L.), the symbiotic snakelocks anemone Anemonia viridis (ForskaÊl), and the seagrass Zostera marina (L.). Organisms were exposed to static dispersant concentrations for 48-h and median lethal concentration (LC50), median effect concentration (EC50), and lowest-observable-effect concentration (LOEC) values obtained. The sublethal effects of 48-h exposures and the ability of species to recover for up to 72 h after exposure were quanti®ed relative to the 48-h endpoints. Results indicated that the anemone lethality test was the most sensitive with LOECs of 20 ppm followed by mussel feeding rate, seagrass photosynthetic index and amphipod lethality, with mussel lethality being the least sensitive with LOECs of 250 ppm for both dispersants. The results were consistent with current theory that dispersants act physically and irreversibly on the respiratory organs and reversibly, depending on exposure time, on the nervous system. Superdispersant-25 was found overall to be less toxic than Corexit 9527 and its sublethal effects more likely to be reversible following short-term exposure.

KeywordsÐDispersants Anemonia viridis Corophium volutator Mytilus edulis Zostera marina

INTRODUCTION es of Warren Spring Laboratory Speci®cation LR 448(OP) and has been approved as a type-2, as well as a type-3, dispersant Faced with the prospect of an oil spill coming ashore or under test quali®cation CSR 4600/8902798 (Oil Slick Dis- passing over reefs, decisions have to be made swiftly as to persants Ltd. product pro®le, cited May 5, 2004; http:// how best to deal with the situation. One option is to use chemical dispersants to break up the slick into a large number of www.croftpark.co.uk/osd-products.html). Superdispersant-25 small droplets. Once broken up, the slick poses less of a phys- has been tested by the Center for Environment, Fisheries, and ical risk to seabirds or marine mammals but may transfer oil Aquaculture Science and has been found to be of low toxicity into the water column and possibly to the benthos. Within to Crangon crangon (brown shrimp) for use at sea and on estuaries, inlets, enclosed bays, or shallow water reefs, the beaches, and Patella vulgata (common limpet) for use on concentration of the dispersants alone may be suf®cient to rocky shores; it is licensed under the Ministry of Agriculture, cause toxic effects. In the United Kingdom, dispersants cannot Fisheries, and Food, Food and Environment Protection Act 53/ be used in water less than 20 m deep or within one nautical 98 (Oil Slick Dispersants Ltd. product pro®le, cited May 5, mile of such [1] without the permission of the Department for 2004; www.croftpark.co.uk/osd-products.html). Corexit 9527 Environment Food and Rural Affairs; similar rules relating to has been tested extensively in the laboratory and used on oil sensitive habitats such as coral reefs and mangroves exist in spills since 1978 [4]. A considerable number of toxicity reports tropical regions [2]. Hence, the option to use dispersants within exist concerning a wide variety of species, reviewed by estuaries, inlets, and shallow water does exist and it is in such George-Ares and Clark [3]. Thus Corexit 9527 provides a circumstances that dif®cult decisions on how best to protect useful comparative toxicant for the study of SD-25. the environment and commercial operations have to be made. The use of dispersants within enclosed bodies of water may The handling of large volumes of dispersant under dif®cult pose a threat to a diverse range of species. This study compares conditions may result in accidental release of potentially toxic the toxicity of SD-25 with that of Corexit 9527 to the marine chemicals into the sea. Research into the toxicity of dispersants sediment-dwelling amphipod Corophium volutator (Pallas), has been reported widely [3,4] and companies continue to the blue mussel Mytilus edulis (L), the symbiotic snakelocks improve the ef®ciency of the chemicals and reduce their tox- anemone Anemonia viridis (ForskaÊl), and the seagrass Zostera icity. In the United Kingdom, the oil dispersant Superdisper- marina (L.). The mudshrimp C. volutator is distributed widely sant-25 (SD-25) is now the Maritime and Coastguard Agency's around the coasts of western Europe and northeast America, main stockpiled chemical for spraying onto oil slicks at sea. and is signi®cant in structuring and sustaining the ecology of No data exist within peer-reviewed literature for SD-25. How- near-shore sediment communities [5,6]. Corophium volutator ever, SD-25 in association with oil meets all the relevant claus- is now used commonly as a European acute toxicity test organism [7±12]. Amphipods occupying a similar niche exist in * To whom correspondence may be addressed other regions, e.g., Ampelisca abdita (Mills) also are used for ([email protected]). toxicity testing. Blue mussels mainly occur on exposed rocky

1219 1220 Environ. Toxicol. Chem. 24, 2005 A. Scarlett et al. shores and are distributed widely from the Arctic to the Med- 476) and transported to the laboratory within 1 h. The outer iterranean with related species distributed worldwide. Mussels leaves were removed so that the three youngest leaves re- have been used for long-term monitoring projects such as the mained. These leaves were wiped with paper to remove epi- Global Mussel Watch [13], ®eld surveys [14], and Scope for phytes and shortened to a maximum length of 300 mm above Growth studies [15], allowing a large body of knowledge to the sheath. Roots were standardized to three rhizome segments. accumulate regarding their acute and sublethal responses to The plants were placed in 20-L tanks ®lled with ®ltered 34 Ϯ stressors. Snakelocks anemones occur on intertidal rocky 1 psu seawater without any sediment or additional nutrient. shores and are associated closely with seagrass beds [16]. As Other conditions were as above. symbionts possessing zooxanthellae, anemones have been used as surrogate organisms for the study of coral organisms (e.g., Preparation of test solutions [17,18]). Eelgrass, Z. marina, is a marine angiosperm with a Corexit 9527 was supplied gratis from the U.S. Minerals worldwide distribution and is protected strictly under the Berne Management Service and SD-25 was obtained from the U.K. Convention [19]. North Atlantic populations have suffered Maritime and Coastguard Agency. Nominal exposure concen- great losses during the last century [20] and, as a consequence, trations were prepared by direct syringe injection or pipetting Z. marina is now deemed to be scarce. Eelgrass plants grow of dispersant into seawater adjusted to the required salinity for well under laboratory conditions and have proved to be a the toxicity tests as appropriate (25 Ϯ 1 psu for Corophium, robust test species if chlorophyll fast ¯uorescence transient is 34 Ϯ 1 psu for other spp.) and vortex mixing for 3 min at used to measure plant health [21]. Taken together, these species high speed. represent a broad range of organisms that might be affected by dispersant use. Some degree of extrapolation to other sim- Rationale and general test conditions ilar species can be justi®ed from the toxicity data obtained. Toxicity tests of dispersants have been performed using The aim of this study was to compare the acute and sub- short-term, static, continuous ¯ow-through and spiked declin- lethal toxicities of the two dispersants, Superdispersant-25 and ing ¯ow-through exposures. Exposure conditions have a con- Corexit 9527, following 48-h static exposures. The extent of siderable in¯uence on the reported toxicity values with spiked recovery of each species was documented for up to 72 h with declining exposures giving much higher LC50 values, ap- regard to changes in the no-observable-effect concentration proximately 3 to 23 times greater, than continuous exposures (NOEC) and LOEC values. Behavioral responses of test ani- [3]. Here we used static exposures of 48-h duration noting that mals also were recorded. the length of exposure does not necessarily produce an incip- ient LC50 for any of the species tested during the 48-h ex- MATERIALS AND METHODS posure and longer exposure, therefore, may result in lower Collection and maintenance of organisms LC50 values. As well as acute toxicity, the ability of organisms Corophium volutator and sediment were collected from an to recover in clean seawater for up to 72 h was assessed relative intertidal area of the Avon estuary near Aveton Gifford, South to the 48-h endpoints. The species-speci®c sublethal endpoints Devon, United Kingdom (ordinance survey grid reference: SX for each test are described in the following paragraphs. 683 467). Amphipods were sieved from the upper 5 cm of Test vessels (2-L Pyrex beakers) were maintained at 15 Ϯ sediment and transported back to the laboratory within 1 h, 1ЊC with a 12:12 h light:dark cycle. Beakers were sealed loose- where they were placed in 5-L culture tanks lined with ®eld- ly with Para®lm௡ M (Pechiney Plastic Packaging, Menasha, collected, sieved (Ͻ300-␮m) sediment. The tanks were ®lled WI, USA) and aerated via a glass Pasteur pipette. Dissolved with ®ltered seawater 25 Ϯ 1 practical salinity units (psu), oxygen, pH, temperature, and salinity were measured after 0, which was aerated and maintained at 15 Ϯ 1ЊC with a 12:12- 24, and 48 h in one replicate from each treatment. Dissolved h light:dark cycle. The animals were fed weekly with ®ve drops oxygen was measured in all replicates at the beginning and standard aquarium invertebrate food (Waterlife Invert Food, end of the experiment. Three replicate vessels per treatment Waterlife Research Industries, Longford, UK; Liquifry Marine, were used in all tests except the mussel feeding-rate bioassay Interpret Ltd., Dorking, UK; Roti-Rich, Florida Aqua Farms, in which nine replicate vessels per treatment were used. Test Dade City, FL, USA; and dried algae) and the water was re- results were analyzed statistically using one-way analysis of placed 24 h after feeding. Amphipods were maintained under variance after checking for normality and homogeneity of var- the above conditions for one to two weeks after removal from iances; 48-h LC50 values were derived using the trimmed the ®eld to acclimate them to experimental conditions. Spearman-KaÈrber method. The EC50 values were obtained Mussels were collected from Port Quin, on the North Corn- from regression of log-transformed data. wall coast, United Kingdom (ordinance survey grid reference: SW 972 905). Organisms of 30-mm Ϯ 5 mm length were Corophium volutator removed carefully from the rock by cutting the byssal threads In order to obtain a concentration-response relationship, and transported back to the laboratory within 2 h. Any epifauna nominal exposure concentrations of 0, 50, 125, 175, 213, 250, and epiphytes were removed from the mussels, which then 375, and 500 ppm Corexit 9527 and SD-25 were prepared as were placed in 20-L tanks ®lled with ®ltered 34 Ϯ 1 psu above. Sieved sediment (Ͻ300 ␮m), approximately 160 ml, seawater. Other conditions were as above. was placed in the test vessels to a depth of 15 mm and 1 L Anemones were collected from Jennycliff Bay, Plymouth of test solution added and allowed to settle for 2 h before 20 Sound on the South Devon coast, United Kingdom (ordinance amphipods (size range 3±7 mm) were introduced via a plastic survey grid reference: SX 491 523) and transported back to Pasteur pipette to each test vessel, a total of 60 amphipods per the laboratory within 1 h. Maintenance conditions were same treatment. Beakers were inspected for amphipod activity and as for the mussels. behavior after 18, 24, and 42 h. At the end of the exposure Eelgrass was collected from the Yealm estuary, South Dev- period of 48 h, the sediment was sieved (300 ␮m) and the on, United Kingdom (ordinance survey grid reference: SX 530 number of live, dead, and missing individuals recorded. Miss- Comparative toxicity of Superdispersant-25 and Corexit 9527 Environ. Toxicol. Chem. 24, 2005 1221

ing individuals were assumed to have died. Surviving am- pended in buffer. Centrifugation was repeated until samples phipods were placed in clean sediment and seawater for a 24- were free of host tissue. The resulting pellet was resuspended h recovery period to assess their ability to burrow. in buffer to give a cell density of about 1 ϫ 106 cells mlϪ1. Zooxanthallae numbers were determined by replicate counts Mytilus edulis (n ϭ 5) and converted to zooxanthallae per wet weight of Nominal exposure concentrations of 0, 80, 130, 200, 250, tentacle. and 320 ppm Corexit 9527 and SD-25 were prepared as above. Zostera marina Three mussels, length 30 mm Ϯ 5 mm, were placed in the test vessels with 1 L of test solution, a total of nine mussels per Because mortality is not a practical parameter for a plant treatment. At the end of the exposure period, the mussels were toxicity test, the photosynthetic ef®ciency of eelgrass was ex- categorized as: Closed, open but able to close when stimulated, amined using the chlorophyll fast ¯uorescence JIP transient or open but failed to close when stimulated. Mussels in the (from 50 ␮s to 1 s; J is a characteristic step at 2 ms, I is an former two categories were placed in clean seawater for a intermediate step at 30 ms, and P is maximum ¯uorescence) further 72 h. Open mussels that failed to respond to stimuli measurements [23,24]. Three plants per exposure vessel were were considered dead and were excluded from the recovery exposed to a concentration range of 0, 80, 130, 200, 320, and trial. Following the recovery period of 72 h, mussels were 500 ppm dispersant, a total of nine plants per treatment, fol- reassessed. An additional test was performed with a single lowed by a 24-h recovery period in clean seawater. Plants were toxicant treatment of 50 ppm for both dispersants plus controls dark-adapted for 5 min using a clip system before illumination using only one mussel per test vessel but with nine replicates with 660 nm light from a light-emitting diode source built into per treatment. At the end of the 48-h exposure period, the the ¯uorometer sensor. Fluorescence measurements at F0 (50 ␮ feeding rate of the mussels was assessed based on methodology s), F300␮s Fj (2 ms), and Fm (tFmax) were recorded using a described by Donkin et al. [22]. Mussels were placed individ- Handy PEA plant ef®ciency analyzer (Hansatech Instruments, ually in 400-ml glass beakers containing 350 ml of clean sea- Kings Lynn, Norfolk, UK) after 24 and 48 h. The Biolyzer᭧ water. After a 10-min acclimation period with slow vortex software (Ronald Maldonado-Rodriguez, Bioenergetics Lab- mixing, 500 ␮lofIsochrysis sp. algal solution was added to oratory, University of Geneva, Geneva, Switzerland, 2002) give approximately 2 ϫ 104 cells mlϪ1. A 20-ml water sample was used to load the full ¯uorescence transients and to cal- was removed immediately from all the beakers upon the ad- culate the JIP parameters according to the equations of the dition of the algae and retained in vials for algae enumeration. JIP-test [23,24]. The effect of the dispersants on the plant's Further samples were taken after 15 and 30 min. Algal particles photosynthetic apparatus was assessed by statistical compar- were analyzed using a Beckman Z2 Coulter particle count and ison of the main parameters: Yield of primary photochemistry size analyzer (Beckman Coulter, Wycombe, UK), which was expressed in ratio of variable to maximal ¯uorescence (Fv/ set to count particles between 3 to 10.43 ␮m. From the loss Fm), performance index (PI), and the area between the ¯uo- of algal particles during the 30-min period, the feeding rates rescence curve and the maximal ¯uorescence intensity. Ten of the mussels were determined. measurements per exposure vessel were taken at random points on the leaves avoiding any necrotic lesions (minimum of 10 Anemonia viridis mm distance). The plants then were placed in clean seawater Scoping tests had indicated that anemones were highly sen- and further measurements recorded after 24 h. sitive to both dispersants; therefore, a lower concentration RESULTS range of 0, 10, 20, 40, 80, and 160 ppm was used followed by a 24-h recovery period. One anemone per test vessel was Physical conditions exposed to 1 L of test solution, a total of three anemones per Dissolved oxygen was above 70% saturation, pH in the treatment. At the end of the 48-h exposure period, the anem- range 7.8 to 8.2, salinity 25 Ϯ 1 psu (Corophium test) or 34 ones were classi®ed as having extended or retracted tentacles. Ϯ 1 psu (other spp. tests), and temperature 15 Ϯ 1ЊC during The tentacles were stimulated gently with a glass rod and their the exposure period in all test vessels. response recorded. Anemones that were insensitive to stimuli were classed as moribund because they would be unable to Corophium volutator feed or defend themselves from predation and, therefore, were No activity was observed after 18 h at nominal dispersant unlikely to survive in their natural environment. The anemones concentrations at or below 125 ppm, or within the controls. then were placed in clean seawater for a 24-h recovery period At dispersant concentrations of 175 ppm and above, individ- and reassessed. A con®rmatory test was performed using three uals were showing signs of stress: Either crawling on the sed- anemones per test vessel at 20 and 30 ppm, a total of nine iment surface or swimming erratically. The activity caused anemones per treatment. The recovery period was extended to greater turbidity within treatments compared to the controls. 72 h to test if further recovery was possible. The effect of the After 24 h, several moribund individuals were visible in ex- dispersants on the density of zooxanthallae within the tentacles posure treatments of 375 and 500 ppm for both Corexit 9527 of A. viridis also was assessed using anemones exposed to the and Superdispersant-25. After 42 h of exposure, moribund 0 to 160 ppm concentration range following 48-h exposure individuals also were visible at and above 175 ppm for both and 24-h recovery. Zooxanthallae were separated from the host dispersants, although this clearly was greater within the Cor- material after homogenization of weighed tentacles in a Whea- exit 9527 treatments. By the end of the experiment, little ac- ton glass homogenizer in 1 ml of buffer solution (50 nM HE- tivity was observed, resulting in a reduction in turbidity, es- PES, 1 nM ethylenediaminetetraacetate, pH 7.4). Whole ho- pecially at the higher concentrations. mogenate was centrifuged for 3 min at 735 G (Eppendorf Mortality was less than 1% within the controls and 100% 5415C, Eppendorf UK, Cambridge, UK). The supernatant was at the highest nominal concentration of 500 ppm for both discarded and the insoluble plant material retained and resus- Corexit 9527 and SD-25 (Fig. 1). Although there was a sharp 1222 Environ. Toxicol. Chem. 24, 2005 A. Scarlett et al.

Table 2. Responses of the mussel Mytilus edulis after 48-h dispersant exposures (48 h) and a 72-h recovery period (Rec.) Mussels from all concentrations were pooled (Superdispersant-25 [SD-25] n ϭ 45 and Corexit 9527 n ϭ 45; Control n ϭ 9 and classi®ed as closed (closed), open and responsive to stimuli (active), or open and unresponsive to stimuli (dead)

Closed (%) Active (%) Dead (%)

48 h Rec. 48 h Rec. 48 h Rec.a

Control 0 0 100 100 0 0 SD-25 44 58 36 0 20 42 Corexit 9527 67 51 11 0 22 49

a Dead mussels from the 48-h exposure were excluded from the recovery vessels but are recorded in the recovery column.

Fig. 1. Mortality (mean Ϯ standard error) of the mudshrimp Coro- close promptly when touched) were able to close very slowly phium volutator following a 48-h static exposure to the oil dispersants and then remain closed. Therefore, only those that absolutely Corexit 9527 and Superdispersant-25 (each treatment contained three failed to close when touched were deemed to have died (Fig. replicates of 20 amphipods). 2a). At the lowest concentration of 80 ppm, the bivalves exposed to Corexit 9527 experienced about 20% mortality but fall in survival of C. volutator exposed to 175 ppm Corexit variation was high and signi®cant mortality ( p Ͻ 0.05) only 9527, a more gradual decline was observed within the SD-25 occurred at 250 ppm for Corexit 9527±and SD-25±exposed treatments. The LC50 (48-h) values of 159 (95% con®dence organisms (Table 1). High variation within treatments occurred limits 145±173 ppm) and 260 ppm (95% con®dence limits at all treatment levels and reliable LC50 values could not be 240±282 ppm) were calculated for Corexit 9527 and SD-25, calculated. Following 72 h in clean seawater, the number of respectively (Table 1). The NOEC was 125 ppm for both dis- dead mussels had increased, but NOEC and LOEC values were persants. Mortality of Corexit 9527 exposed amphipods was unchanged (Fig. 2b). Sublethal effects were assessed at dis- signi®cantly ( p Ͻ 0.05) greater than those exposed to SD-25. persant concentrations of 50 ppm for 48 h using a feeding rate All of the surviving individuals exposed to SD-25 concentra- bioassay (n ϭ 9). Two mussels within the Corexit 9527 ex- tions up to 175 ppm were able to swim normally and were posure died and were omitted from the bioassay. Feeding rates able to rebury in clean sediment, although swimming activity were reduced signi®cantly ( p Ͻ 0.05) for both dispersant ex- still was observed 3 h following introduction into the clean posures with SD-25 rates reduced to 9.8% of controls and seawater; all survived the recovery period. Survivors from Corexit 9527±exposed mussels only 2.6% (Fig. 3). Corexit- Corexit 9527 also were able to rebury and survive the recovery exposed mussels had feeding rates signi®cantly ( p Ͻ 0.05) period with exposures up to 125 ppm. However, above 175 less than that of SD-25±exposed organisms. ppm, 100% of amphipods failed to recover from Corexit 9527. Anemonia viridis Mytilus edulis The anemones were highly sensitive to both dispersants Control mussels appeared active and healthy (i.e., open and with 55% of SD-25±exposed and 100% of Corexit 9527±ex- responsive to stimuli) throughout the test period but all disposed organisms insensitive to stimuli at 20 ppm (Fig. 4a). At persant-exposed organisms were classi®ed as either closed or 40 ppm for both dispersants, all organisms had retracted ten- dead at the end of the 72-h recovery period (Table 2). As- tacles that failed to respond to stimuli and were assessed to sessment of mortality after 48 h was problematic because some be moribund. Above 80 ppm, anemone tissue was starting to mussels that initially appeared dead (i.e., open and failing to decompose. The concentration range between no effect and

Table 1. Comparison of toxicity estimates for Superdispersant-25 and Corexit 9527 with four marine species

Corexit 9527 concn. (ppm) Superdispersant-25 concn. (ppm)

LC50 or Con®dence LC50 or Con®dence Species Test NOECa LOECb EC50c limits NOEC LOEC EC50 limits

Anemonia viridis No response to 10 20 15d NAe 10 20 20a NA stimuli Corophium volutator Mortality 125 175 159 145±173 125 175 260 240±282 Mytilus edulis Mortality 200 250 Ð Ð 200 250 Ð Ð M. edulis Feeding rate Ͻ50 50 NA Ͻ50 50 NA Zostera marina JIP-testf PIg Ͻ80 80 55 28±150 Ͻ80 80 386 339±439 a NOEC ϭ no-observed-effect-concentration. b LOEC ϭ lowest-observed-effect-concentration. c LC50 ϭ median lethal concentration; EC50 ϭ median effect concentration. d Interpolated from data only, not derived from model. e NA ϭ not applicable. f JIP-test ϭ measurements acquired from the fast ¯uorescence transient. g PI ϭ performance index. Comparative toxicity of Superdispersant-25 and Corexit 9527 Environ. Toxicol. Chem. 24, 2005 1223

Fig. 2. Mortality (mean Ϯ standard error) of the mussel Mytilus edulis following a 48-h static exposure (a) and a further 72-h recovery period (b) to the oil dispersants Corexit 9527 and Superdispersant-25 (each treatment contained three replicates of three mussels).

Fig. 4. Percentage of moribund Anemonia viridis (mean Ϯ standard 100% mortality was too small to calculate LC50 values using error), following a 48-h static exposure (a) and a further 24-h recovery the Spearman-KaÈrber method. However, interpolation of the period (b) to the oil dispersants Corexit 9527 and Superdispersant- 25 (the 20- and 30-ppm treatments contained three replicates of three data indicated EC50 values of about 20 ppm for SD-25 and anemones; all other treatments contained three replicates of one anem- about 15 ppm for Corexit 9527. The NOEC and LOEC (48- one). h) values were 10 ppm and 20 ppm, respectively for both dispersants (Table 1). Following recovery in clean seawater, all of the anemones previously exposed to 30 ppm SD-25 and below were able to respond to stimuli, as did all organisms previously exposed to Corexit 9527 at 20 ppm (Fig. 4b); however, 89% of the anemones previously exposed to 30 ppm Corexit 9527 failed to respond to stimuli (Fig. 4b). No signi®cant differences were observed in zooxanthallae densities.

Zostera marina All of the main parameters, Fv/Fm, PI, and area, were reduced at the lowest exposure of 80 ppm after 24 and 48 h for both dispersants (Fig. 5); the NOEC for both dispersants was Ͻ80 ppm. The PI was the most-sensitive parameter giving 48- h EC50 values of 386 ppm and 55 ppm for SD-25 and Corexit 9527, respectively. At the lowest exposure concentration there was no signi®cant difference between the dispersants; however, at 130 ppm and above, Corexit was signi®cantly ( p Ͻ 0.05) more toxic for all main parameters. Leaves of the Cor- exit-exposed plants turned brown at 200 ppm and the outer leaves started to become detached, leaving only the more protected inner leaf from which to take measurements. As well as the main parameter values obtained from the PEA, the JIP- Fig. 3. Feeding rates (mean Ϯ standard error) of the mussel Mytilus edulis following a 48-h static exposure of 50 ppm to the oil dispersants test allows calculation of various bio-physical expressions, Corexit 9527 and Superdispersant-25 (each treatment contained nine such as speci®c ¯uctuations and yields, and phenomenological replicates of one mussel). ¯uctuations. To visualize the effect of the dispersants on the 1224 Environ. Toxicol. Chem. 24, 2005 A. Scarlett et al.

Fig. 5. Photosynthetic parameters (Fv/Fm), performance index (PI), and area (mean Ϯ standard error), measured in the leaves of Zostera marina after 24- and 48-h exposure, and 24-h recovery in clean seawater, to the oil dispersants Corexit 9527 and Superdispersant-25 (each treatment contained three replicates of three plants from which 10 measurements were taken). eelgrass leaves, pipeline models have been calculated and to recover, e.g., mean PI values were 0.84 (SE 0.01) after the drawn on the basis of experimental signals from the 200-ppm 48-h exposure to 80 ppm and fell to 0.76 (SE 0.07) at the end concentrations and compared to control leaves (Fig. 6). The of the 24-h recovery period (Fig. 4). SD-25±exposed plants were able to recover slightly after 24 h in clean seawater, e.g., mean PI values rose from 0.88 (stan- DISCUSSION AND CONCLUSION dard error [SE] 0.01) after 48 h at 80 ppm exposure to 1.07 This is, to our knowledge, the ®rst report of toxicity data (SE 0.04) following recovery. Corexit-exposed plants failed for SD-25 that reveals its acute toxicity to a range of marine Comparative toxicity of Superdispersant-25 and Corexit 9527 Environ. Toxicol. Chem. 24, 2005 1225

[27]. Surfactants can bind to and disrupt cellular phospholipid bilayers altering the transmembrane sodium gradient [28]. A review of Corexit 9527 toxicity data by George-Ares and Clark [3] covering 28 reports and 37 aquatic species, found no apparent trend in the reported sensitivity between taxa. Toxicity (24±96±h LC50 or EC50) ranged from 1.6 to Ͼ1,000 ppm. Amphipod LC50 values varied between 3 ppm for Al- lorchestes compressa [29] and Ͼ175 ppm for Boekosimus sp. [30], but no data for C. volutator were reported. Mollusc (adults) LC50 values were in the range 33.8 ppm for the sand snail Polinices conicus [29] to 2,500 ppm for the scallop Ar- gopecten irradians [31]; however, no data for Mytilus edulis were reported. The class Anthozoa (anemones and true corals) were not represented in the reviewed papers and the only seagrass was Thalassia testudinum with a 96-h LC50 of 200 ppm [32]. In general, the reviewed data show that Corexit 9527 is of low acute toxicity to most species, although embryo and larval stages are more sensitive. The results from this study are consistent with the literature data but suggest an order of sensitivity, based on NOEC and LOEC values, of: Anthozoa Ͼ macrophyte Ͼ crustacean Ͼ mollusc for both Corexit 9527 and SD-25, although the order of sensitivity partly is due to the differences in test methodology and endpoints. Both dispersants caused a sharp increase in mortality over a narrow concentration range; this was most pronounced for Corexit 9527. Therefore, it is of more use to refer to NOEC or LOEC than LC50 values. The NOEC values for A. viridis were much lower than that observed for the other species tested (Table 1) and the sensitivity of the anemones was similar to that of the embryo or larval stages of molluscs and ®sh [33,34]. Of the anemones with extended tentacles that were unresponsive to stimuli following 48-h dispersant exposure of 20 ppm Fig. 6. Comparison of membrane models of the Photosystem II (PSII) and, therefore, classed as moribund (Fig. 4a), all were able to apparatus in control plants (top), and 48-h exposure to 200 ppm Su- perdispersant-25 (middle) and Corexit 9527 (bottom). The models recover when placed in clean seawater (Fig. 4b). All of the represent the speci®c activities expressed as ¯uctuation per reaction SD-25±exposed anemones also were able to recover from the center (RC). The relative magnitude of each activity or ¯uctuation is 30-ppm exposure, but this was true of only 11% of Corexit- shown by the width of the corresponding arrow. Absorption ¯ux exposed anemones. The ability of the organisms to recover (ABS) is proportional to the concentration of antenna chlorophyll and the average antenna size is given as ABS/RC. This expresses the total implies that the dispersants act reversibly on the neural re- absorption ¯ux of PSII antennae chlorophyll divided by the number ceptors at low concentrations but cause irreversible membrane of active, in the sense of primary quinone (QA)-reducing, reaction damage at higher concentrations, consistent with reported sur- centers. The absorption and trapping by PSII units with a heat sink factant toxicity modes of action [27,28]. No effect on zoox- center (non-Q ±reducing) is indicated as the hatched parts of the A anthallae density was found and the cells appeared undamaged arrows ABS/RC and energy ¯ux for trapping (TRo/RC). The antenna belonging to the PSII units with heat sink centers is drawn in black even when their host tissue was damaged severely. It is likely and the antenna that belongs to the active centers is drawn in white. that the host tissue protected the plant cells and that, during The degree of stress, thus, is indicated by the reduction in ABS/RC short-term exposure, the algae do not migrate from their hosts and TRo/RC leading to a greater dissipation of energy in the form of heat (Dl /RC) and a decrease in the useful energy available to the as happens with coral bleaching episodes. Anemones have o been used as a surrogate organism for symbiotic coral assem- plant (ETo/RC). blages [17,18], although it is not known if the skeletal cup of reef-building corals would afford the same protection for the species to be signi®cantly lower ( p Ͻ 0.05), in the majority zooxanthallae as the anemone. of tests, or equivalent (mussel mortality and anemone tests) The M. edulis mortality data were very variable, but Corexit to the widely used dispersant Corexit 9527. Dispersants con- 9527 clearly was more toxic at low concentrations (Fig. 2a). tain surfactants, which may be nonionic or anionic, dissolved, The total percentage of SD-25±exposed mussels that were or suspended in solvents. Corexit 9527 is a mixture of both closed at the end of the 48-h exposure period increased from nonionic (48%) and anionic (35%) surfactants in an aqueous 44 to 58%, but Corexit-exposed closed mussels decreased from solvent containing ethylene glycol monobutyl ether (17%); the 67 to 51% with a corresponding rise in the percentage dead surfactants include ethoxylated sorbitan mono- and trioleates (Table 2). A very pronounced effect was observed in the feed- (nonionic) and sodium dioctyl sulfoccinate (anionic) [25], but ing rates of mussels exposed to 50 ppm of both dispersants, little is known about the constituents of SD-25 except that it with Corexit-exposed organisms' ®ltration rates reduced so is a blend of glycol and glycol ether solvents, combined with much that they were not signi®cantly greater than system nonionic and anionic surfactants [26]. Dispersants are thought blanks (Fig. 3). Although the highly reduced ®ltration rates to act physically and irreversibly on the respiratory organs and could be a behavioral response to the dispersants, the mussels' reversibly, depending on exposure time, on the nervous system failure to open within the 72-h recovery period following the 1226 Environ. Toxicol. Chem. 24, 2005 A. Scarlett et al.

48-h concentration-response test (Table 2) suggests that the erally is of lower toxicity than Corexit 9527 to a range of mussels may have been suffering physiological damage as they coastal species, although equivalent in terms of NOEC and would be expected to recover from nonspeci®c narcosis. Al- LOEC values; also, at maximum concentrations likely to be though there was no sublethal component to the C. volutator found at sea, any sublethal effects upon organisms are more test, it was observed that many individuals failed to burrow likely to be reversible. quickly when exposed to Ն175 ppm dispersants. It was impossible to tell if it was the swimming amphipods that died later, although these likely had a greater exposure than those AcknowledgementÐThis study was ®nanced by the Maritime and that burrowed rapidly into the sediment. Briggs et al. [8] used Coastguard Agency, the Department for Environment Food and Rural Affairs, the Department of Trade and Industry, and Minerals Man- turbidity caused by stressed Corophium as a measure of tox- agement Service Grant RP 480. The Photosynthesis Membrane model icity and found a good correlation between turbidity during was derived using Biolyzer௡ software provided by R. Maldonado- the ®rst 24 h of exposure and mortality after 10 d. The failure Rodriguez, Bioenergetics Laboratory, University of Geneva, Geneva, to burrow and increased swimming activity observed at mod- Switzerland. erate contamination levels, are most likely an escape behavioral response rather than a physiological response because REFERENCES they occur with a range of toxicants [8,35]. The dispersant 1. Maritime and Coastguard Agency. 2002. Contingency Planning concentrations in this study that caused the greatest turbidity for Marine Pollution Preparedness and Response: Guidelines for Ports, 2002. Southampton, UK. were within the con®dence limits of the calculated LC50 values 2. Etkin DS. 1998. Factors in the dispersant use decision-making (Table 1). Under test conditions where escape was impossible, process: Historical overview and look to the future. Proceedings the behavioral response not to burrow probably proved to be of the Twenty-First Arctic and Marine Oil Spill Program Tech- detrimental, but in the wild it may enable the organisms to nical Seminar. Environment Canada, Ottawa, ON, pp. 281±304. 3. George-Ares A, Clark JR. 2000. Acute toxicity of two Corexit escape from localized areas of high concentration. dispersants. Chemosphere 40:897±906. Both dispersants disrupted the Photosystem II (PSII) ap- 4. U.S. National Research Council. 1989. Using Oil Dispersants on paratus, known to be a primary casualty during stress condi- the Sea. National Academy, Washington, DC. tions [36], within the leaves of Z. marina and this was found 5. Limia JM, Raffaelli D. 1997. The effects of burrowing by the to be largely irreversible, especially for Corexit 9527 (Fig. 5). amphipod Corophium volutator on the ecology of intertidal sediments. J Mar Biol Assoc UK 128:147±156. The PI is a standard composite of several JIP-test parameters 6. Raffaelli D, Milne H. 1997. An experimental investigation of the and has been shown to be a highly sensitive measure that is effects of shorebird and ¯at®sh predation on estuarine inverte- correlated strongly with other measurements of plant health brates. Estuar Coast Shelf Sci 24:1±13. [24]. Indeed, at concentrations Ն200 ppm Corexit 9527, it was 7. Bat L, Raffaelli D. 1998. Sediment toxicity testing: A bioassay approach using the amphipod Corophium volutator and the poly- evident by visual inspection that the leaves were damaged chaete Arenicola marina. J Exp Mar Biol Ecol 226:217±239. severely and the outer leaves were becoming detached. Leaves 8. Briggs AD, Greenwood N, Grant A. 2002. Can turbidity caused of Z. marina possess a thin cuticle [37] that may afford a by Corophium volutator (Pallas) activity be used to assess sed- degree of protection from the dispersants, although it is clear iment toxicity rapidly? Mar Environ Res 55:181±192. that 24 h was suf®cient for photosynthesis to be affected. Algae 9. Roddie BD, Thain JE. 2001. Biological effects of contaminants: Corophium sp. sediment bioassay and toxicity test. Techniques and spores with less protection likely are more sensitive, as in Marine Environmental Sciences 28. International Council for found by Singer et al. [34], exposing zoospores of giant kelp, the Exploration of the Sea, Copenhagen, Denmark. Macrocystis pyrifera, to Corexit 9527. 10. Center for Environment, Fisheries, and Aquaculture Science. Extrapolating the results from this and other studies sug- 2001. Monitoring and surveillance of nonradioactive contaminants in the aquatic environment and activities regulating the gests that the most sensitive organisms to dispersants are those disposal of wastes at sea, 1998. Aquatic Environment Monitoring with the least protective tissue or shell. As well as the spore, Report 53. Center for Environment, Fisheries, and Aquaculture embryo, and larval stages, it is likely that unprotected organ- Science, Lowestoft, UK. isms such as nudibranchs and seaslugs would be vulnerable 11. Conradi M, Depledge MH. 1998. Population responses of the marine amphipod Corophium volutator (Pallas, 1766) to copper. and, therefore, extra care may need to be taken where rare or Aquat Toxicol 44:34±45. endangered species are present. Under current dispersant-use 12. Ciarelli S, Vonck WAPMA, van Stralen NM. 1997. Reproduc- guidelines that exist in the United Kingdom and elsewhere ibility of spiked-sediment bioassays using the marine benthic am- [1,2], dispersant concentrations are unlikely to be suf®ciently phipod, Corophium volutator. Mar Environ Res 43:329±343. high to cause harm to aquatic organisms within open waters. 13. Goldberg ED. 1975. The Mussel WatchÐa ®rst step in global marine monitoring. Mar Pollut Bull 6:111±113. Measured concentrations at sea are rare but Bocard et al. [38] 14. Galloway TS, Sanger RC, Smith KL, Fillmann G, Readman JW, reported a maximum concentration of 13 ppm. However, if Ford TE, Depledge MH. 2002. Rapid assessment of marine pol- dispersants are used or spilled accidentally within estuaries, lution using multiple biomarkers and chemical immunoassays. enclosed bays, or shallow coral reefs, it may be possible for Environ Sci Technol 36:2219±2226. 15. Widdows J, Donkin P, Brinsley MD, Evans SV, Salkeld PN, concentrations to become elevated long enough to cause mor- Franklin A, Law RJ, Waldock MJ. 1995. Scope for growth and tality of sensitive species and stress to more tolerant animals contaminant levels in North Sea mussels Mytilus edulis. Mar Ecol and plants. Any harm to aquatic species as a result of dispersant Prog Ser 127:131±148. use must be measured against the potential harm that may arise 16. Connor DW, Brazier DP, Hill TO, Northen KO. 1997. Marine Nature Conservation Review: Marine Biotope Classi®cation for from the toxic or physical effects of oil. The dif®cult decision Britain and Ireland, Vol 1ÐLittoral Biotopes. Version 97.06, of when, where, and under what conditions to use dispersants Report 229. Joint Nature Conservation Committee, Peterborough, will remain as long as oil is transported via the sea. Spill UK. response agencies are better able to make such decisions when 17. Morrall CE, Galloway TS, Trapido-Rosenthal HG, Depledge MH. 2000. Characterization of nitric oxide synthase activity in the provided with relevant ecotoxicological data. To our knowl- tropical sea anemone Aiptasia pallida. Comp Biochem Physiol edge, these are the ®rst reported results concerning the toxicity B 125:483±491. of Superdispersant-25; it has shown that this dispersant gen- 18. Whitehead LF, Douglas AE. 2003. Metabolite comparisons and Comparative toxicity of Superdispersant-25 and Corexit 9527 Environ. Toxicol. Chem. 24, 2005 1227

the identity of nutrients translocated from symbiotic algae to an 29. Gulec I, Leonard B, Holdway DA. 1994. Oil and dispersed oil animal host. J Exp Biol 206:3149±3157. toxicity to amphipods and snails. Spill Science & Technology 19. Davison DM, Hughes DJ. 1998. An Overview of Dynamics and Bulletin 4:1±6. Sensitivity Characteristics for Conservation Management of Ma- 30. Foy MG. 1982. Acute lethal toxicity of Prudhoe Bay crude oil rine SACs, Vol IÐZostera Biotopes. UK Marine SACs Project. and Corexit 9527 to Arctic marine ®sh and invertebrates. Tech- Scottish Association for Marine Science, Argyll, UK. nology Development Report EPS 4-EC-82-3. Environment Can- 20. Vergeer LHT, Aarts TL, deGroot JD. 1995. The wasting disease ada, Ottawa, ON. and the effect of abiotic factors (light-intensity, temperature, sa- 31. Ordsie CJ, Garofalo GC. 1981. Lethal and sublethal effects of linity) and infection with Labyrinthula zosterae on the phenolic short-term acute doses of Kuwait crude oil and a dispersant Cor- content of Zostera marina shoots. Aquat Bot 52:35±44. exit 9527 on bay scallops, Argopecten irradians, and two pred- 21. Scarlett A, Donkin P, Fileman TW, Evans SV, Donkin ME. 1999. ators at different temperatures. Mar Environ Res 5:195±210. Risk posed by the antifouling agent Irgarol 1051 to the seagrass 32. Baca BJ, Getter CD. 1984. The toxicity of oil and chemically Zostera marina. Aquat Toxicol 45:159±170. dispersed oil to the seagrass Thalassia testudinum. In Allen TE, 22. Donkin P, Widdows J, Evans SV, Worrall CM, Carr M. 1989. ed, Oil Spill Chemical Dispersants: Research, Experience, and Quantitative structure activity relationships for the effect of hy- Recommendations. STP 840. American Society for Testing and drophobic organic chemicals on rate of feeding by mussels (My- Materials, Philadelphia, PA, pp 314±323. tilus edulis). Aquat Toxicol 14:277±294. 33. Singer MM, Smalheer DL, Tjeerdema RS, Martin M. 1990. Tox- 23. Strasser RJ, Srivastava A, Govindjee B. 1995. Polyphasic chlo- icity of an oil dispersant to the early life stages of four Californian rophyll a ¯uorescence transient in plants and cyanobacteri. Pho- marine species. Environ Toxicol Chem 9:1389±1397. tochem Photobiol 61:32±34. 34. Singer MM, Smalheer DL, Tjeerdema RS, Martin M. 1991. Ef- 24. Hermans C, Smeyers M, Rodriguez RM, Eyletters M, Strasser fects of spiked exposure to an oil dispersant on the early life stages of four marine species. Environ Toxicol Chem 10:1367± RJ, Delhaye JP. 2003. Quality assessment of urban trees: A com- 1374. parative study of physiological characterization, airborne imag- 35. Kravitz MJ, Lamberson JO, Ferraro SP, Swartz RC, Boese BL, ing, and on-site ¯uorescence monitoring by the OJIP-test. J Plant Specht DT. 1999. Avoidance response of the estuarine amphipod Physiol 160:81±90. Eohaustorius estuarius to polycyclic aromatic hydrocarbon±con- 25. Canevari GP. 1971. Oil slick dispersants and methods. Patent taminated, ®eld-collected sediments. Environ Toxicol Chem 18: 3,793,218. U.S. Patent Of®ce, Washington, DC. As cited in Scelfo 1232±1235. GM, Tjeedema RS. 1991. A simple method for determination of ௡ 36. Morales F, Belkhodja R, Abadia A, Abadia J. 2000. Photosystem Corexit 9527 in natural waters. Mar Environ Res, 31:69±78. II ef®ciency and mechanisms of energy dissipation in iron-de®- 26. Ayles Fernie International. 2003. Superdispersant-25 Safety Data cient, ®eld-grown pear trees (Pyrus communis L.). Photosynth Sheet. Kent, UK. Res 63:9±21. 27. Wells PG. 1984. The toxicity of oil spill dispersants to marine 37. Larkum AWD, Roberts G, Kuo J, Strother S. 1989. Gaseous organisms: A current perspective. In Allen TE, ed, Oil Spill movement in seagrasses. In Larkum AWD, McComb AJ, Shep- Chemical Dispersants: Research, Experience, and Recommen- herd SA, eds, The Biology of Seagrasses with Special References dations. STP 840. American Society for Testing and Materials, to the Australian Region. Elsevier, Amsterdam, The Netherlands, Philadelphia, PA, pp 177±202. pp 686±722. 28. Borseth JF, Aunaas T, Denstad JP, Nordtug T, Olsen AJ, Schmid 38. Bocard C, Castaing G, Gatellier C. 1984. Chemical oil dispersion R, Skjaervo G, Zachariassen KE. 1995. Transmembrane sodium in trials at sea and in laboratory tests: The key role of dilution energy gradient and calcium content in the adductor muscle of processes. In Allen TE, ed, Oil Spill Chemical Dispersants: Re- Mytilus edulis L. in relation to the toxicity of oil and organic search, Experience, and Recommendations. STP 840. Philadel- chemicals. Aquat Toxicol 31:263±276. phia, PA, pp 125±142. Genomic and physiological footprint of the Deepwater SPECIAL FEATURE Horizon oil spill on resident marsh ﬁshes

Andrew Whiteheada,1, Benjamin Dubanskya, Charlotte Bodiniera, Tzintzuni I. Garciab, Scott Milesc, Chet Pilleyd, Vandana Raghunathane, Jennifer L. Roacha, Nan Walkere, Ronald B. Walterb, Charles D. Ricef, and Fernando Galveza

Departments of aBiological Sciences, cEnvironmental Sciences, and eOceanography and Coastal Sciences, and dCoastal Studies Institute, Louisiana State University, Baton Rouge, LA 70803; bDepartment of Chemistry and Biochemistry, Texas State University, San Marcos, TX 78666; and fDepartment of Biological Sciences, Clemson University, Clemson, SC 29634 THE DEEPWATER HORIZON SCIENCE APPLICATIONS IN Edited by Paul G. Falkowski, Rutgers, The State University of New Jersey, New Brunswick, NJ, and approved September 1, 2011 (received for review OIL SPILL SPECIAL FEATURE June 13, 2011) The biological consequences of the Deepwater Horizon oil spill are spatially and temporally comprehensive but of low resolution unknown, especially for resident organisms. Here, we report and chemistry data are of high resolution but patchy in space and results from a field study tracking the effects of contaminating time. Ocean surface oil was remotely detected through the oil across space and time in resident killifish during the first 4 analysis of images from synthetic aperture radar (SAR) (14). mo of the spill event. Remote sensing and analytical chemistry Proximity of the nearest oil slick to each field site (e.g., Fig. S1) identified exposures, which were linked to effects in fish charac- was measured for each day that SAR data were available, from terized by genome expression and associated gill immunohisto- May 11 through August 13, 2010, to approximate the location, chemistry, despite very low concentrations of hydrocarbons timing, and duration of coastal oiling (Fig. 1C). Although surface remaining in water and tissues. Divergence in genome expression oil came close to many of our field sites in mid-June, only the coincides with contaminating oil and is consistent with genome Grande Terre (GT) site was directly oiled (Fig. 1 B and C). responses that are predictive of exposure to hydrocarbon-like Although the GT site had been clearly contaminated with crude chemicals and indicative of physiological and reproductive impair- oil for several weeks before our sampling (Fig. 1C and Fig. S2) ment. Oil-contaminated waters are also associated with aberrant and retained much oil in sediments (Dataset S2), only trace SCIENCES protein expression in gill tissues of larval and adult fish. These data

concentrations of oil components were detected in subsurface ENVIRONMENTAL suggest that heavily weathered crude oil from the spill imparts water samples collected from the GT site on June 28, 2010, and significant biological impacts in sensitive Louisiana marshes, some tissues did not carry abnormally high burdens of oil constituents of which remain for over 2 mo following initial exposures. at any site or time point (Dataset S2). Despite a low chemical signal for oil in the water column and tissues at the time of ecological genomics | ecotoxicology | microarray | RNA-seq | sampling, we detected significant biological effects associated toxicogenomics with the GT site postoil. We sampled multiple tissues from adult Gulf killifish (average ollowing the Deepwater Horizon (DWH) drilling disaster on weight of 3.5 g) from each of six field sites for each of three time FApril 20, 2011, in the Gulf of Mexico, acute oiling and the points [only the first two time points for the Mobile Bay (MB) resulting mortality of marine wildlife were evident. In contrast, site] spanning the first 4 mo of the spill event (Fig. 1C). We the sublethal effects, critically important for predicting long- compared biological responses across time (before, at the peak, term population-level impacts of oil pollution (1), have not and after oiling) and across space (oiled sites and sites not oiled) been well described following the DWH disaster. Here, we re- and integrated responses at the molecular level using genome port the results of a 4-mo field study monitoring the biological expression profiling with complimentary protein expression and effects of oil exposure on fish resident in Gulf of Mexico coastal tissue morphology. Genome expression profiles, using micro- marsh habitats. arrays and RNAseq, were characterized for livers because the Gulf killifish (Fundulus grandis) were used as our model spe- organ is internal and integrates xenobiotic effects from multiple cies because they are among the most abundant vertebrate ani- routes of entry (gill, intestine, and skin), and because liver is the mals in Gulf of Mexico-exposed marshes (2–4). Furthermore, the primary tissue for metabolism of toxic oil constituents. Tissue Atlantic-distributed sister species to F. grandis (Fundulus heter- morphology and expression of CYP1A protein, a common bio- oclitus) has a narrow home range and high site fidelity, especially marker for exposure to select polycyclic aromatic hydrocarbons during the summer (5, 6), and, among fishes, it is relatively (PAHs), was characterized for gills, the organ that provides the sensitive to the toxic effects of organic pollutants (7). Although greatest surface area in direct contact with the surrounding home range and toxicology studies are lacking for F. grandis,we aquatic environment. In addition, we exposed developing infer that F. grandis is also relatively sensitive to pollutants and embryos to field-collected water samples to document bio- exhibits high site fidelity, such that the biology of this species is availability and bioactivity of oil contaminants for this sensitive likely affected primarily by the local environment, given the re- early life stage. cent shared ancestry of F. grandis with F. heteroclitus (8) and similar physiology, life history, and habitat (9–13). We sampled from populations resident in Gulf of Mexico-exposed marshes Author contributions: A.W. and F.G. designed research; A.W., B.D., C.B., T.I.G., S.M., C.P., – V.R., J.L.R., N.W., R.B.W. and F.G. performed research; C.D.R. contributed new reagents/ before oil landfall (May 1 9, 2010), during the peak of oil analytic tools; A.W., B.D., C.B., T.I.G., S.M., C.P., V.R., N.W., R.B.W. and F.G. analyzed data; landfall (June 28–30, 2010), and after much of the surface oil was and A.W., B.D., C.B., and F.G. wrote the paper. no longer apparent 2 mo later (August 30–September 1, 2010) at The authors declare no conflict of interest. six field sites from Barataria Bay, Louisiana, east to Mobile Bay, This article is a PNAS Direct Submission. Alabama (Fig. 1 and Dataset S1). Data deposition: Microarray data have been deposited to ArrayExpress (accession no. E-MTAB-663). Results and Discussion 1To whom correspondence should be addressed. E-mail: [email protected]. Remote sensing and analytical chemistry were used to charac- This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. terize exposure to DWH oil, where remote sensing data are 1073/pnas.1109545108/-/DCSupplemental. www.pnas.org/cgi/doi/10.1073/pnas.1109545108 PNAS Early Edition | 1of5 Fig. 1. Location of field study sites and incidence of oil contamination. (A) Location of field sampling sites, which include Grand Terre (GT), Bay St. Louis (BSL), Belle Fontaine Point (BFP), Bayou La Batre (BLB), Mobile Bay (MB), and Fort Morgan (FMA). Color coding is consistent with other figures. The red star indicates the DWH spill site. (B) Photograph (by A.W.) of the GT field site on June 28, 2010, showing contaminating oil and minnow traps in the marsh. (C) Proximity of nearest surface oil to each field site was determined by SAR, where rows are field sites and columns are days. Light gray represents no data, and black represents the nearest surface oil at a distance of >4 km; the increasing intensity of red indicates closer proximity of oil. Three field sampling trips are highlighted (blue boxes). BSL; BFP; FMA.

The oiling of the GT site at the end of June 2010 is associated such as cytochrome P450s, cytochrome B5, and UDP-glucur- with a clear functional genomic footprint. Of the 3,296 genes onosyltransferase (Fig. 2F, set 1), for which increased transcrip- included in our analysis, expression of 1,600 and 1,257 genes tion is particularly diagnostic of exposure to select hydrocarbons varied among field sites and throughout the time course, re- (20). Indeed, many genes that are transcriptionally induced or spectively (P < 0.01) (Dataset S3). For the 646 genes that varied repressed by AHR activators (dioxins, PCBs, and PAHs) show in expression only among sites (no significant time effect or site- induction or repression at the GT site coincident with crude oil by-time interaction), site variation followed a pattern of pop- contamination (Fig. 2F, set 1). An independent measure of ge- ulation isolation by distance, which is consistent with neutral nome expression, RNAseq, also indicates AHR activation in GT evolutionary divergence (Fig. 2A) and population genetic expec- fish from June 28, 2010, compared with reference RNA (e.g., up- tations (15). Most importantly, 1,500 genes indicated a pattern of regulation of cytochrome P450s, UDP-glucuronosyltransferase site-dependent time course expression (significant interaction, (UGT), and AHR itself; Fig. 2E). In parallel, up-regulation of false discovery rate <0.01), where the trajectory of genome ex- CYP1A protein was detected in gills from GT fish sampled pression through time was divergent at the GT site compared postoil and in early life-stage fish following controlled exposures with all other sites (Fig. 2 B and C), particularly at the second to GT waters (Figs. 3 and 4). These data appear to be diagnostic time point, which coincides with oil contamination (Fig. 1C). of exposure to the toxic constituents in contaminating oil (PAHs) Previous studies have identified genes that are transcription- at a sufficient concentration and duration to induce biological ally responsive to planar polychlorinated biphenyl (PCB) expo- responses in resident fish. Sustained activation of the CYP1A sures in killifish (16). Planar PCBs, dioxins, and PAHs (the gene (Figs. 2F and 3) was predictive of persistent exposure to primary toxic constituents in crude oil) are all mechanistically sublethal concentrations of crude oil components and negative related insofar as they exert biological effects, in whole or in part, population-level impacts in fish, sea otters, and harlequin ducks through aryl-hydrocarbon receptor (AHR) signaling pathways; following the Exxon Valdez oil spill (reviewed in 1), although indeed, morpholino knockdown of the AHR is protective of the PAH toxicity may be mediated through AHR-independent toxic effects of PAHs and PCBs in killifish (17), and exposures to pathways as well (21). PCBs and PAHs induce common genome expression responses Transcriptional responses in other sets of coexpressed genes in flounder (18). Of the genes that were transcriptionally re- offer insights into the potential biological consequences of con- sponsive to PCB exposures (16), 380 were included in the current taminating oil exposure at the GT site. Several gene ontology analysis. Expression of this subset of genes is predictive of (GO) categories were enriched in the subset of genes that transcriptional divergence in fish from the GT site coincident showed GT-specific expression divergence coincident with site- with oil contamination compared with other field sites (Fig. S3), and time-specific oil contamination (Dataset S4). GO enrich- especially for the top 10% of PCB-responsive genes (Fig. 2D). ment indicates activation of the ubiquitin-proteasome system Transcriptional activation of these planar PCB-responsive genes (Fig. 2F, set 2), which, among diverse functions, is important for in developing killifish embryos is predictive of induction of de- cellular responses to stress, cell cycle regulation, regulation of velopmental abnormalities, decreased hatching success, and de- DNA repair, apoptosis, and immune responses (22). The AHR creased embryonic and larval survival (16, 19). This set of genes protein itself plays a role as a unique ligand-dependent E3 includes members of the canonical battery of genes that are ubiquitin ligase that targets sex steroid (estrogen and androgen) transcriptionally induced by ligand-activated AHR signaling, receptor proteins for proteasomal destruction, thereby impairing

2of5 | www.pnas.org/cgi/doi/10.1073/pnas.1109545108 Whitehead et al. SPECIAL FEATURE THE DEEPWATER HORIZON SCIENCE APPLICATIONS IN OIL SPILL SPECIAL FEATURE

Fig. 2. Genome expression between field sites and across time. Field sites include Grand Terre (GT), Bay St. Louis (BSL), Belle Fontaine Point (BFP), Bayou La Batre (BLB), Mobile Bay (MB), and Fort Morgan (FMA). GT was the only site to be directly oiled, which occurred between the first and second sampling times (Fig. 1 and Dataset S2). (A) For genes that vary only among sites (no expression change with time or interaction), pairwise site-specific transcriptome divergence along principal component (PC) 1, as a function of pairwise geographical distance, shows a pattern consistent with isolation by distance. (B) Tra- jectory of genome expression responses through time for each of six field sites from the preoil sample time (dot at base of arrow) through the peak-oil sample time (middle dot), to the latest postevent sample time (dot at head of arrow) following PC analysis of genes showing statistically significant main effects (site and time) and interaction terms. (C) Divergence along PC1 is isolated, where bars for each site from left to right represent sampling times from the earliest to the latest. (D) Expression divergence along PC1 for the subset of genes that is dose-responsive to PCB exposure (top 10% of PCB-responsive genes). (E) RNAseq fi data showing genes up- and down-regulated (x axis positive and negative, respectively) in sh from GT sample time 2 (coincident with oil) compared with SCIENCES reference RNA, where select genes are identified. (Inset) Genes are dramatically down-regulated at GT (detailed RNAseq data are presented in Dataset S5). (F) ENVIRONMENTAL Expression levels for specific genes (rows) and treatments (columns), where cell color indicates up-regulation (yellow) or down-regulation (blue) scaled according to site-specific expression level at the preoil sample time, for genes with divergent expression at the GT site. Genes are grouped into functional categories, and scale bars indicate N-fold up- or down-regulation. normal cellular responses to sex hormones in reproductive tis- in embryos, can impair cardiac performance in adulthood (30). sues, and this response can be activated by planar PAHs (23). The adult fish sampled in situ from the oil-contaminated GT site Significant down-regulation of transcripts for egg envelope pro- showed divergent regulation of several genes involved in blood teins zona pellucida (ZP3 and ZP4) and choriogenin (ChgHm vessel morphogenesis and heme metabolism coincident with oil and ChgH) that we detect at the GT site coincident with oil contamination (Fig. 2F, set 3). Multigeneration field studies are exposure (Fig. 2F, set 1) may be linked to this AHR-dependent necessary to confirm cardiovascular effects from DWH oil con- proteolytic pathway because their transcription is estrogen-de- tamination of marshes that coincided with spawning. pendent (24, 25) and is down-regulated by exposure to PAHs in fish (25–27). In corroboration, RNAseq detects dramatically down-regulated ZP, ChgH, and vitellogenin transcripts in GT fish (Fig. 2E). Although the transcriptional response that we detect is in male fish, these proteins are synthesized in male livers (reviewed in 25, 27) and down-regulation is consistent with antiestrogenic effects from exposure to PAHs (28). Possible impacts on reproduction merit attention because water only from the GT site induced CYP1A protein in the gills of developing killifish (Fig. 3) at low concentrations of total aromatics and alkanes (Dataset S2) and more than 2 mo after initial oiling, indicating persistent bioavailability of PAHs. Marsh contamination with DWH oil coincided with the spawning season for many marsh animals, including killifish (29), and reproductive effects are predictive of long-term population-level impacts from oil spills (1). Controlled exposures of developing killifish to water collected from GT on June 28 and August 30, 2010, induced CYP1A protein expression in larval gills relative to fish exposed to GT water preoil and exposed to Bayou La Batre (BLB) site water that was not oiled (Fig. 3). This response is consistent with the location and Fig. 3. CYP1A protein expression (dark red staining) in larval killifish gills timing of oil contamination, and it indicates that the remaining oil (24 d postfertilization) exposed to waters collected from GT (oiled) and BLB constituents dissolved at very low concentrations at GT after (not oiled) during development. (Magnification 40×, scale bars = 10 μm.) landfall (Dataset S2) were bioavailable and bioactive to de- CYP1A expression is elevated in the lamellae of larvae exposed during developing fish. Although exposures to PAHs stereotypically induce velopment to waters collected from GT postoil (trips 2 and 3) compared with fi background levels of CYP1A expression in larvae exposed to GT water preoil cardiovascular system abnormalities in developing sh at rela- (trip 1), compared with CYP1A in fish exposed to waters collected from BLB tively high concentrations (e.g., 21), none were observed in these (which was not directly oiled), and compared with CYP1A in fish reared in animals. However, even very low-concentration exposures during laboratory control water. Nuclei were stained using hematoxylin (blue). development, insufficient to induce cardiovascular abnormalities Analytical chemistry of exposure waters is reported in Dataset S2.

Whitehead et al. PNAS Early Edition | 3of5 Coastal salt marsh habitats are dynamic and stressful, where changes in environmental parameters, such as temperature, hypoxia, and salinity, can continuously challenge resident wildlife. Regulation of ion transport in fish is particularly important for facilitating homeostasis in response to the salinity fluctuations that are common in estuaries. We found altered regulation of multiple ion transport genes in fish from the GT site coincident with oil contamination (Fig. 2F, set 4). For example, V- type proton ATPases are up-regulated and Na+,K+-ATPase subunits and tight-junction proteins are down-regulated, coincident with oiling at the GT site, in the absence of substantial changes in environmental salinity (Dataset S2). Other genes important for osmotic regulation in killifish (31) are also divergently down-regulated at the GT site, including type II iodothyronine deiodinase (DIO2), transcription factor jun-B (JUNB), and arginase 2 (ARG2). In corroboration, RNAseq data show down-regulation of DIO2, JUNB, and ARG2 in GT fish compared with reference fish (Fig. 2E). Although the physiological consequences of oil exposures are typically studied in isolation, it is reasonable to predict that exposure to oil may compromise the ability of resident organisms to adjust physiologically to natural stressors. Induction of CYP1A protein expression is a hallmark of AHR signaling pathway activation, making it a sensitive biomarker of exposure to select planar PAHs and other hydrocarbons (20). Although the liver is the key organ for CYP1A-mediated metabolism of these substrates, gill tissues represent the most proximate site of exposure to PAHs. As a result of direct contact with the environment and the nature of the gill as a transport epithelium, the gill may be a more sensitive indicator of exposure to contaminants than the liver (32). CYP1A protein was markedly elevated in GT fish postoil compared with GT fish preoil and compared with fish from other field sites that were not directly oiled (Fig. 4). CYP1A induction was localized pre- dominantly to pillar cells of the gill lamellae and within undifferentiated cells underlying the interlamellar region, which may have contributed to the filamental and lamellar hyperplasia observed during trips 2 and 3, as well as the gross proliferation of the interlamellar region observed during trip 2 in GT fish (Fig. 4). These effects imply a decrease in the effective surface area of the gill, a tissue that supports critical physiological functions, such as ion homeostasis, respiratory gas exchange, systemic acid- base regulation, and nitrogenous waste excretion (33). Currently, the degree to which oil-induced effects may interact with com- fl monly encountered challenges, such as uctuations in hypoxia Fig. 4. CYP1A protein expression in adult killifish gills (dark red staining) and salinity, to compromise physiological resilience is unclear. sampled in situ from all sampling times (columns) and locations (rows). By integrating remote sensing and in situ chemical measures of Locations include Grand Terre (GT), Bay St. Louis (BSL), Belle Fontaine Point exposure, and linking these with integrated measures of bi- (BFP), Bayou La Batre (BLB), Mobile Bay (MB), and Fort Morgan (FMA). ological effect (genome expression and tissue morphology), we (Magnification 40×, scale bars = 10 μm.) The MB site was only sampled on provide evidence that links biological impacts with exposure to trips 1 and 2, and gills from trip 1 at the BLB site were not available for contaminating oil from the DWH spill within coastal marsh processing. Fish gills from the GT site during trips 2 and 3 showed high CYP1A habitats. Although body burdens of toxins are not high, consis- expression and an elevated incidence of hyperplasia of the lamellae and interlamellar space on the gill filaments coincident with oil contamination. tent with reports indicating that seafood from the Gulf of Mexico CYP1A protein was elevated at the GT site postoil (trips 2 and 3) compared is safe for consumption (34), this does not mean that negative with GT preoil (trip 1) as well as with other field sites, none of which were biological impacts are absent. Our data reveal biologically relevant directly oiled. Nuclei were stained using hematoxylin (blue). Exact site loca- sublethal exposures causing alterations in genome expression and tions and sampling dates are reported in Dataset S1. tissue morphology suggestive of physiological impairment per- sisting for over 2 mo after initial exposures. Sublethal effects were predictive of deleterious population-level impacts that persisted were caught by minnow trap, and tissues were excised immediately. Liver was preserved in RNAlater (Ambion, Inc.) for genome expression (microarray over long periods of time in aquatic species following the Exxon fi Valdez spill (1) and must be a focus of long-term research in and RNAseq) analysis. Gill tissues were xed in situ in buffered zinc-based formalin Z-Fix (Anatech LTD). Succinct methods follow, and more detailed the Gulf of Mexico, especially because high concentrations of methods are available online. hydrocarbons in sediments (Dataset S2) may provide a persistent Satellite imagery (SAR) was analyzed to provide estimation of the timing, source of exposures to organisms resident in Louisiana marshes. location, and duration of coastal oil contamination. The calculated distance from each field sampling site to the nearest oil slick was calculated from the Methods “straight-line” distance from the global positioning system position of the The locations (latitude and longitude) of our field sampling sites and dates for station (Dataset S1) to that of the observed oil across any and all intervening sampling at each site are summarized in Dataset S1. Gulf killifish (F. grandis) geographical barriers (e.g., Fig. S1).

4of5 | www.pnas.org/cgi/doi/10.1073/pnas.1109545108 Whitehead et al. Analytical chemistry of water, tissue, and sediment samples was performed each differentially expressed target using a negative binomial method with to offer detailed characterization of exposure to contaminating oil (data P values adjusted by the Benjamini–Hochberg procedure. SPECIAL FEATURE reported in Dataset S2). Sample dates and locations are summarized in Gill tissues were sampled from all field sites for morphological analysis and Dataset S1. All sample extracts were analyzed using GC interfaced to an MS immunohistochemical analysis of CYP1A protein expression. Gill tissues from detector system. Spectral data were processed by Chemstation Software the first and second gill arches were sectioned along the longitudinal axis at (Agilent Technologies), and analyte concentrations were calculated based a thickness of 4 μm and probed with mAb C10-7 against fish CYP1A (40). on the internal standard method. Sections were counterprobed using the Vectastain ABC immunoperoxidase Genome expression across sites and time was characterized using custom oligonucleotide microarrays. Genome expression was measured in liver tis- system (Vector Laboratories), utilizing the ImmPACT Nova RED peroxidase sues from five replicate individual male fish per site-time treatment (5 bi- substrate kit (Vector Laboratories) to visualize the CYP1A protein in red. ological replicates) hybridized in a loop design, including a dye swap. Data Tissue sections were counterstained with Vector Hematoxylin QS (Vector were lowess-normalized and then mixed model-normalized using linear Laboratories). fi THE DEEPWATER HORIZON mixed models to account for xed (dye) effects and random (array) effects. F. grandis embryos obtained from parents not exposed to oil (collected SCIENCE APPLICATIONS IN OIL SPILL SPECIAL FEATURE Normalized data were then analyzed using mixed model ANOVA, with from Cocodrie, LA) were exposed to water samples from the GT and BLB “site” [Grand Terre (GT), Bay St. Louis (BSL), Belle Fontaine Point (BFP), sites collected subsurface on the dates indicated in Dataset S1. Following Bayou La Batre (BLB), Mobile Bay (MB), and Fort Morgan (FMA)] and fertilization, 20 embryos were randomly transferred in triplicate to one of “ ” sampling time (sampling trips 1, 2, and 3) (Dataset S1) as main effects, the six field-collected waters (2 field sites × 3 time points) at 3 h post- including an interaction (site-by-time) term. The false discovery rate was fertilization. Embryos were also exposed to a laboratory control consisting estimated using Q-value (35). Principal components analysis was performed of artificial 17 parts per thousand (ppt) water. Larvae were sampled at 24 using MeV (36). GO enrichment was tested using DAVID (37). d postfertilization and fixed in Z-Fix solution. Sectioning and staining were For RNAseq, transcript abundance was compared between liver mRNA from three replicate fish (RNA was not pooled) from the GT site from June 28, as described in the previous section. 2010, and mRNA from two control samples. All RNA samples were sequenced fi on the Illumina Gene Analyzer platform (Expression Analysis, Inc.). Following ACKNOWLEDGMENTS. K. Carman helped facilitate early eld studies. The quality control filtering, quantitative transcript abundance analysis was authors thank R. Brennan, D. Roberts, E. McCulloch, Y. Meng, A. Rivera, C. Elkins, H. Graber, R. Turner, D. Crawford, and M. Oleksiak, for technical performed by mapping sequence reads to target sequences (6,810 unique F. assistance. Funding was from the National Science Foundation (Grants DEB- heteroclitus target EST sequences, Dataset S5) using the Bowtie short read 1048206 and DEB-1120512 to A.W., Grant EF-0723771 to A.W. and F.G., and alignment software (38). A custom Perl script determined the number of Grant DEB-1048241 to R.B.W.), the National Institutes of Health (R15- fragments mapped to each target sequence. The Bioconductor package ES016905-01 to C.D.R.), and the Gulf of Mexico Research Initiative (A.W., fi

DESeq (version 2.8) (39) was used to determine the statistical signi cance of F.G., and N.W.). SCIENCES ENVIRONMENTAL

1. Peterson CH, et al. (2003) Long-term ecosystem response to the Exxon Valdez oil spill. 21. Incardona JP, et al. (2005) Aryl hydrocarbon receptor-independent toxicity of weath- Science 302:2082–2086. ered crude oil during fish development. Environ Health Perspect 113:1755–1762. 2. Rozas LP, Reed DJ (1993) Nekton use of marsh-durface habitats in Louisiana (USA) 22. Glickman MH, Ciechanover A (2002) The ubiquitin-proteasome proteolytic pathway: deltaic salt marshes undergoing submergence. Mar Ecol Prog Ser 96:147–157. Destruction for the sake of construction. Physiol Rev 82:373–428. 3. Rozas LP, Zimmerman RJ (2000) Small-scale patterns of nekton use among marsh and 23. Ohtake F, et al. (2007) Dioxin receptor is a ligand-dependent E3 ubiquitin ligase. adjacent shallow nonvegetated areas of the Galveston Bay Estuary, Texas (USA). Mar Nature 446:562–566. Ecol Prog Ser 193:217–239. 24. Modig C, et al. (2006) Molecular characterization and expression pattern of zona 4. Subrahmanyam CB, Coultas CL (1980) Studies on the animal communities in 2 North pellucida proteins in gilthead seabream (Sparus aurata). Biol Reprod 75:717–725. Florida salt marshes. 3. Seasonal fluctuations of fish and macroinvertebrates. Bull Mar 25. Yu RMK, Wong MML, Kong RYC, Wu RSS, Cheng SH (2006) Induction of hepatic Sci 30:790–818. choriogenin mRNA expression in male marine medaka: A highly sensitive biomarker 5. Lotrich VA (1975) Summer home range and movements of Fundulus heteroclitus for environmental estrogens. Aquat Toxicol 77:348–358. – (Pisces: Cyprinodontidae) in a tidal creek. Ecology 56:191 198. 26. Holth TF, et al. (2008) Differential gene expression and biomarkers in zebrafish (Danio 6. Teo SLH, Able KW (2003) Habitat use and movement of the mummichog (Fundulus rerio) following exposure to produced water components. Aquat Toxicol 90:277–291. – heteroclitus) in a restored salt marsh. Estuaries 26:720 730. 27. Sanchez BC, Carter B, Hammers HR, Sepúlveda MS (2011) Transcriptional response of 7. Van Veld PA, Nacci DE (2008) Toxicity resistance. The Toxicology of Fishes, eds Di hepatic largemouth bass (Micropterus salmoides) mRNA upon exposure to environ- – Giulio RT, Hinton DE (Taylor and Francis, Boca Raton, FL), pp 597 641. mental contaminants. J Appl Toxicol 31:108–116. 8. Whitehead A (2010) The evolutionary radiation of diverse osmotolerant physiologies 28. Thomas P (1990) Teleost model for studying the effects of chemicals on female re- in killifish (Fundulus sp.). Evolution 64:2070–2085. productive endocrine function. J Exp Zool Suppl 4(Suppl 4):126–128. 9. Able KW, Hata D (1984) Reproductive behavior in the Fundulus heteroclitus-F. grandis 29. Greeley MS, Macgregor R (1983) Annual and semilunar reproductive-cycles of the complex. Copeia (4):820–825. Gulf killifish, Fundulus grandis, on the Alabama Gulf Coast. Copeia (3):711–718. 10. Kneib RT (1997) The role of tidal marshes in the ecology of estuarine nekton. 30. Hicken CE, et al. (2011) Sublethal exposure to crude oil during embryonic de- Oceanography and Marine Biology: An Annual Review 35:163–220. velopment alters cardiac morphology and reduces aerobic capacity in adult fish. Proc 11. Nordlie FG (2006) Physicochemical environments and tolerances of cyprinodontoid Natl Acad Sci USA 108:7086–7090. fishes found in estuaries and salt marshes of eastern North America. Reviews in Fish 31. Whitehead A, Roach JL, Zhang S, Galvez F (2011) Genomic mechanisms of evolved Biology and Fisheries 16:51–106. physiological plasticity in killifish distributed along an environmental salinity gradi- 12. Rozas LP, Lasalle MW (1990) A comparison of the diets of Gulf killifish, Fundulus grandis ent. Proc Natl Acad Sci USA 108:6193–6198. Baird and Girard, entering and leaving a Mississippi brackish marsh. Estuaries 13:332–336. 32. Levine SL, Oris JT (1999) CYP1A expression in liver and gill of rainbow trout following wa- 13. Weisberg SB, Lotrich VA (1982) The importance of an infrequently flooded intertidal terborne exposure: Implications for biomarker determination. Aquat Toxicol 46:279–287. marsh surface as an energy source for the mummichog Fundulus heteroclitus:An 33. Evans DH, Piermarini PM, Choe KP (2005) The multifunctional fish gill: Dominant site experimental approach. Mar Biol 66:307–310. 14. Brekke C, Solberg AHS (2005) Oil spill detection by satellite remote sensing. Remote of gas exchange, osmoregulation, acid-base regulation, and excretion of nitrogenous – Sensing of Environment 95:1–13. waste. Physiol Rev 85:97 177. 15. Williams DA, Brown SD, Crawford DL (2008) Contemporary and historical influences 34. State of Louisiana Department of Health and Hospitals (2011) Louisiana Seafood on the genetic structure of the estuarine-dependent Gulf killifish Fundulus grandis. Safety Surveillance Report (Louisiana Department of Health and Hospitals, Baton Mar Ecol Prog Ser 373:111–121. Rouge, LA). fi 16. Whitehead A, Pilcher W, Champlin D, Nacci D (2011) Common mechanism underlies 35. Storey JD, Tibshirani R (2003) Statistical signi cance for genomewide studies. Proc repeated evolution of extreme pollution tolerance. Proc R Soc B, 10.1098/rspb.2011.0847. Natl Acad Sci USA 100:9440–9445. 17. Clark BW, Matson CW, Jung D, Di Giulio RT (2010) AHR2 mediates cardiac terato- 36. Saeed AI, et al. (2006) TM4 microarray software suite. Methods Enzymol 411:134–193. genesis of polycyclic aromatic hydrocarbons and PCB-126 in Atlantic killifish (Fundulus 37. Huang W, Sherman BT, Lempicki RA (2009) Systematic and integrative analysis of heteroclitus). Aquat Toxicol 99:232–240. large gene lists using DAVID bioinformatics resources. Nat Protoc 4:44–57. 18. Williams TD, et al. (2008) Transcriptomic responses of European flounder (Platichthys 38. Langmead B, Trapnell C, Pop M, Salzberg SL (2009) Ultrafast and memory-efficient flesus) to model toxicants. Aquat Toxicol 90:83–91. alignment of short DNA sequences to the human genome. Genome Biol 10:R25. 19. Whitehead A, Triant DA, Champlin D, Nacci D (2010) Comparative transcriptomics 39. Anders S, Huber W (2010) Differential expression analysis for sequence count data. implicates mechanisms of evolved pollution tolerance in a killifish population. Mol Genome Biol 11:R106. Ecol 19:5186–5203. 40. Rice CD, Schlenk D, Ainsworth J, Goksoyr A (1998) Cross-reactivity of monoclonal 20. Varanasi U (1989) Metabolism of Polycyclic Aromatic Hydrocarbons in the Aquatic antibodies against peptide 277-294 of rainbow trout CYP1A1 with hepatic CYP1A Environment (CRC, Boca Raton, FL), p 341. among fish. Mar Environ Res 46:87–91.

Whitehead et al. PNAS Early Edition | 5of5 BP Sees a Return to Grandeur as Gulf Fishermen Reel From Disaster | Bridge The Gulf Project 4/30/12 10:38 PM

BP Sees a Return to Grandeur as Gulf Fishermen Reel From Disaster

Fri, 2012-04-27 13:22 The second memorial of the nation’s worst oil catastrophe has come and gone, forever linked to Earth Day and seared into the psyches of millions of Gulf residents and fishermen. In recent weeks, the media has unleashed a torrent of stories about the devastating impacts (http://www.southernstudies.org/2012/04/troubled-waters-gulf-communities-still-reeling-two-years-into-bp-disaster.html) of the nation’s worst oil spill disaster; deaths, disease and deformities in the fisheries (http://www.tampabay.com/news/environment/water/oil-from-deepwater-horizon-spill-still- causing-damage-in-gulf-2-years/1225134) ; a two-year record-setting die off (http://www.pbs.org/newshour/rundown/2012/04/baby-dolphin-die-offs- continue-in-the-gulf.html ) in dolphin populations; medical emergencies and family health crises in coastal communities (http://www.huffingtonpost.com/2012/04/20/gulf-oil-spill-anniversary-children_n_1438959.html) ; and ongoing Congressional wrangling over tens of millions of dollars in fines needed to save and rebuild (http://www.nola.com/opinions/index.ssf/2012/03/resolve_to_pass_the_restore_ac.html) the rapidly disappearing Gulf coast.

But it won’t be long before these stories fade from the consciousness of a nation once riveted to the volcanic well spewing out Louisiana crude a mile below the sea. Instead we will see more stories like this one BP published in the Alabama Press-Register last week: After Two Years, The Grandeur of the Gulf is Returning. (http://blog.al.com/press-register- commentary/2012/04/bp_exec_after_two_years_grande.html)

"These days, we don’t see oily sheens and miles of orange containment boom; we see sparkling water and clean sand, dotted with deck chairs and beach towels. On the horizon, we don’t see an armada of ships skimming oil; we see fishing vessels at work gathering the day’s catch. And, in the skies and on the ground, we don’t see planes and large cleanup crews; we see birds and other wildlife at play.

"But one thing is clear: Many of the dire predictions for the Gulf, made in the days and weeks after the accident, have not turned out to be true. Indeed, after two years of hard work alongside local, state and federal officials, the scientific community and the people of the region, substantial progress has been made. And the grandeur of the Gulf is steadily returning."

You can expect the media and the airwaves to be clogged with happy talk about the Gulf in the months ahead. We all wish it were true, but the facts—and perceptions of those toiling in the fisheries—just don't support it. After reading BP’s latest polemic, veteran Alaska marine toxicologist and author Riki Ott (http://www.rikiott.com/) remembered Exxon's tactics after the Valdez disaster in an email this week;

"Reminds me of when the state of Alaska officials under gag order NOT to speak about impacts complained about Exxon's conclusion that just b/c wildlife are "cavorting" on beaches, everything is "hunky-dory" in Prince William Sound. Exxon flew in 3 or 4 British "scientists" (biostitutes) to fly-over Prince William Sound and visit beaches in spring 1990. Exxon sponsored them to go to our state capital and tell all 60 legislators about the "remarkable recovery of Prince William Sound.'"

BP is sponsoring a $500 million scientific initiative (http://www.gulfresearchinitiative.org/) to investigate the impacts of the Gulf oil catastrophe, but it's way too early to be waving the victory flag. Fishermen in the Gulf of Mexico still are reeling after the Deepwater Horizon blew up, and "grandeur" isn't exactly the term they're using down there. Here's how Louisiana's Plaquemines Parish charter fisherman Ryan Lambert described it in the Times Picayunne (http://www.nola.com/news/gulf-oil- spill/index.ssf/2012/04/fishing_guides_say_their_busin.html) this week:

"'The oil may have stopped flowing, but the spill still goes on down here every day,' said Lambert, owner and operator of Cajun Fishing Adventures, a sprawling lodge and charter business in Buras. 'My fishing business is still down 50 to 60 percent, we're still finding oil and tar balls on the beach and in the marsh, people still think the fish are polluted (http://www.nola.com/news/gulf-oil- spill/index.ssf/2012/04/2_years_after_gulf_oil_spill_l_1.html) , and now we can't find speckled trout in nearly the numbers we had before the spill. So don't tell me the disaster is over. Maybe for BP it is. Maybe for the oil business people it is. But for me and other charter businesses, it's never stopped.'"

Shrimp with no eyes and tumors caught in Barataria Bay, April, 2012. Photo by Xuan Chen

http://bridgethegulfproject.org/node/633 Page 1 of 4 BP Sees a Return to Grandeur as Gulf Fishermen Reel From Disaster | Bridge The Gulf Project 4/30/12 10:38 PM

So which is it, gasping for air or returning to grandeur? My bet is most people outside the Gulf will believe the latter as BP continues its PR assault on the national airwaves. Although thousands of fishermen are struggling, their stories are easily overpowered by the massive economic and political forces aligned against them. This Sen. David Vitter (R-LA) missive to his constituents pretty much sums up where the Powers That Be stand in the Gulf:

"The good news is that I don’t think anyone would have predicted that the Gulf would have rebounded to where it is today. That goes for our tourism industry, which is thriving, and of course our Gulf seafood, which is as safe and delicious as ever…”

Yes, that's the same seafood that NRDC's Miriam Rotkin-Ellman and Gina Solomon reported on in Environmental Health Perspectives (http://ehp03.niehs.nih.gov/article/info%3Adoi%2F10.1289%2Fehp.1104539R) last year; the very same seafood that can expose vulnerable populations like children and pregnant women to up to 10,000 times the allowable levels of cancer-causing polycyclic aromatic hydrocarbons (PAHs). You can read Miriam's excellent Gulf health update here (http://switchboard.nrdc.org/blogs/mrotkinellman/bp_oil_disaster_two_years_late.html) .

Of course, it's no surprise that Sen. Vitter ignores this. The oil and gas industry ranks numero uno on his list of top campaign contributors (http://www.opensecrets.org/politicians/summary.php?cid=n00009659#ind) . Most of his colleagues down there float in the same boat. But that's not the boat Gulf fishermen are working in, the ones pulling up tumor and oil-encrusted shrimp (http://www.aljazeera.com/indepth/features/2012/04/201241682318260912.html) . It's not the reality residents who still witness record numbers of dolphins (http://www.pbs.org/newshour/rundown/2012/04/baby-dolphin-die-offs-continue-in-the-gulf.html) washing ashore on beaches littered with tar balls after every storm.

As the carnage continues, BP and other oil industry behemoths are busier than ever drilling in the Gulf, dragging their mammoth platforms into deeper water where they can jam cement-reinforced pipelines into even more remote high-pressure subsea oil and gas deposits. Oil slicks are routinely reported (http://onwingsofcare.org/protection-a-preservation/gulf-of-mexico-oil-spill-2010/gulf-2012/244-shell-macondo- mars-ursa-green-canyon-gulf-oil-wings-care.html) in drilling areas offshore. Can the disaster happen again?

You bet. Lured by the high prices and increased global demand, more and more rigs are at work, while Congress still has passed no new drilling safety laws. NRDC's David Pettit reports (http://switchboard.nrdc.org/blogs/dpettit/what_if_another_bp_deepwater_h.html) that while some new federal drilling regulations are in place, we may be setting ourselves up for a "repeat performance." For instance, in an Orlando Sentinel article, (http://articles.orlandosentinel.com/2012-04-17/news/os-bp-gulf-spill-anniversary-20120414_1_deepwater-horizon-rig-gulf-of- mexico-oil-bob-dudley) Pettit questioned why the industry doesn't redesign the critical blowout preventers that failed in the first place.

"You can have 10 of them, and if they are all subject to being jammed by a pipe doing something that we know actually happened [once before], then you're not much safer."

Requiring twice as many safety mechanisms that failed once before doesn’t exactly inspire confidence. But that’s life in our oil-addicted, rush-to-drill world. Our country’s energy policy is stuck on dig, extricate and burn. The petrochemical industry will become more dangerous as we rush to extract harder to find deposits of crude in places like the harsh environs of the Arctic, where Shell Oil plans (http://switchboard.nrdc.org/blogs/rkistner/arctic_oil_drilling_threatens.html) to start drilling this summer. Have we learned nothing from the historic BP blowout?

But there is some good news; disasters can change attitudes, even in the oil-dominated Gulf of Mexico. The BP debacle blew a hole through Gulf residents' confidence in the oil industry and changed perspectives on the importance of environmental protections, according to a recent University of New Hampshire study reported in Science Daily (http://www.sciencedaily.com/releases/2012/04/120412105227.htm) :

"If disasters teach any lessons, then experience with the Gulf oil spill might be expected to alter opinions about the need for environmental protection. About one-fourth of our respondents said that as a result of the spill, their views on other environmental issues such as global warming or protecting wildlife had changed," said Lawrence Hamilton, professor of sociology at the University of New Hampshire. This proportion rose to 35 percent among those most affected economically by the spill. People reporting changed views also expressed greater concern about sea level rise due to climate change, more support for a moratorium on deepwater drilling, and were more likely to favor alternative energy rather than increased oil exploration."

Shrimp with tumors bought in New Orleans grocery store. Photo by Mac MacKenzie.

Changes in attitudes can lead to positive change in the oil patch. But there's also another way to get Big Oil to change its behavior: jail http://bridgethegulfproject.org/node/633 Page 2 of 4 BP Sees a Return to Grandeur as Gulf Fishermen Reel From Disaster | Bridge The Gulf Project 4/30/12 10:38 PM

time. Oil industry experts say fines and penalties do little to reduce reckless decisions and risky policies that threaten the lives and livelihoods of fishermen and workers. Pro Publica reporter Abrahm Lustgarten (http://www.propublica.org/site/author/Abrahm_Lustgarten/) laid it all out succinctly in this New York Times op-ed (http://www.nytimes.com/2012/04/20/opinion/a-stain-that-wont-wash-away.html? _r=1&partner=rss&emc=rss) :

"What the gulf spill has taught us is that no matter how bad the disaster (and the environmental impact), the potential consequences have never been large enough to dissuade BP from placing profits ahead of prudence. That might change if a real person was forced to take responsibility — or if the government brought down one of the biggest hammers (http://www.propublica.org/article/epa-officials-weighing-sanctions-against-bps-us-operations) [4] in its arsenal and banned the company from future federal oil leases and permits altogether. Fines just don't matter."

Although low-level prosecutions (http://www.usatoday.com/money/industries/energy/story/2012-04-24/bp-oil-spill-arrest-justice-department/54504158/1) have just been announced, nobody expects many BP execs to get rolled up by the Department of Justice anytime soon. But this catastrophe also wasn’t an ordinary industrial accident. It coated the coasts of four states with oil and has damaged the lives of countless residents. Some fishermen in the Gulf say they still relish the thought of former BP CEO Tony Hayward (http://www.usatoday.com/money/industries/energy/story/2012-04-24/bp-oil-spill-arrest-justice-department/54504158/1) caught in the wheels of American justice, a fanciful dream as they struggle to rebuild their lives.

While BP fights a long war in the courts, the battle over public opinion remains hot and intense. Despite the million dollar ad campaigns and the political rhetoric about the Gulf returning to normal, many—especially in the fishing community—are facing a new reality. Their lives on the water have changed in ways no one comprehends, in ways many fear will never be the same.

The grandeur of the Gulf will return. But it’s up to people along its shores—and to all of us—to decide what kind of future lies in store; a marine environment stressed by an oil-soaked sea or a healthy ocean preserved for distant generations. The critical decisions are still to come.

Rocky Kistner

(javascript:void(0);)

Comments

NBA Reports (/node/633#comment-10132) Submitted by Anonymous on Sun, 2012-04-29 02:18. CHICAGO (AP) -- Bulls legend Derrick Went up by will Michael Vick Nike Jersey (http://www.nfleaglesproshop.com/nike-michael-vick-jersey/) certainly pass up the rest of the time caused by a divided anterior cruciate plantar fascia throughout the quit leg. Went up by ended up being made it easier for off of the the courtroom delayed throughout Chicago's 103-91 playoff-opening win in the Philadelphia 76ers in Wednesday, an astounding hit for the crew eyeing a new champion manage. They won 3 things along with ended up being participating in more like your league's reigning MVP immediately after lost 35 online games as a consequence of incidents in the standard time, nevertheless the time located a stop because Bulls ended up wrapping up an amazing get. Went up by crumbled on the terrain immediately after they driven your isle using with regards to 1: 20 quit plus the Bulls primary by simply 12. They ended up being getting a layup while they located a new jump-stop along with seemed to adjust the head because 76ers' Spencer Hawes along with Lavoy Allen revolving around, moving Nike Michael Vick Jersey (http://www.nfleaglesproshop.com/nike-michael-vick-jersey/) past off of to your teammate ahead of the cumbersome clinching. Crew health care workers quickly in a rush out and about along with offered help for you to Went up by for several minutes while they ended up being writhing throughout ache at the baseline ahead of aiding your ex on the locker place. Went up by ended up being arrive at the hospital, wherever MRI benefits established your Bulls' worst type of anxieties. NEW MEXICO (AP) -- Knicks novice defend Iman Shumpert will certainly pass up the rest of the postseason immediately after bringing a new quit leg plantar fascia inside 3 rd fraction involving Brand-new York's Sport 1 decline on the New mexico High temperature in Wednesday. Your Knicks released Shumpert in addition took the side to side meniscus. Medical procedures are going to be timetabled throughout The big apple along with Shumpert can be supposed to pass up about 6-8 a few months. Shumpert, whom commenced a final 19 Rolando McClain Nike Jersey (http://www.nflraidersproshop.com/rolando-mcclain-nike-jersey/) online games in the standard time immediately after Jeremy Lin got leg medical procedures, ended up being taking your soccer ball up the appropriate sideline along with tried out for you to dribble guiding the when they ripped way up along with quickly selected in the quit leg. They ended up being produced in the Knicks' locker place pertaining to examination along with after to your infirmary on an MRI. WALTHAM, Muscle size. (AP) -- Boston ma Boston celtics defend Beam Allen remains to be unclear in case he can be capable of participate in inside playoff opener resistant to the The atlanta area Hawks in On the. Allen haven't played out throughout a fortnight caused by a painful appropriate foot. They ended up being taking pictures all-around ahead of train while using crew in Wednesday along with explained although experience many of the "smaller parts" in the training. Nevertheless instructor Doc Waters explained Allen may not train along with they don't feel Nike Rolando McClain Jersey (http://www.nflraidersproshop.com/rolando-mcclain-nike-jersey/) although participate in throughout Sport 1, sometimes. Your 36-year-old taking pictures defend explained they gotten a new cortisone picture in Wed. In case the idea just weren't to the playoffs, they explained, he had by now always be obtaining medical procedures. All the others for the Boston celtics described throughout balanced to the closing train prior to crew foliage pertaining to The atlanta area. THE BIG APPLE -- Your In Pacers' Honest Vogel plus the Memphis Grizzlies' Lionel Hollins right now ended up referred to as your NBA Western along with Developed Seminar Motor coach buses in the Thirty day period, respectively, pertaining to online games played out throughout The spring. Vogel brought about In with an Western Conference- ideal 12-3 (#(. 900) file which include a new league-leading 5-0 level while travelling, your Pacers top earning talent out of the house considering that 2003-04. Your Pacers' 12 victories ended up your team's nearly all is the winner in a very thirty day Darren McFadden Nike Jersey (http://www.nflraidersproshop.com/darren-mcfadden-nike-jersey/) period considering that Late 2003 (13-2). Through The spring, your Pacers averaged 102. 7 things -- virtually six to eight things earlier mentioned their time regular -- along with graded 6th all round throughout Plus-Minus Credit score Differential (+5. 7). Hollins carefully guided your Grizzlies to your 13-3 file throughout The spring, tying your franchise's level to the nearly all is the winner in a very thirty day period (Drive, 2004). One of several Grizzlies' 13 victories ended up is the winner in Oklahoma Area, which in turn clicked your Thunder's six-game earning talent, along with in New mexico. Memphis' season-ending household win versus Holiday ended up being http://bridgethegulfproject.org/node/633 Page 3 of 4 BP Sees a Return to Grandeur as Gulf Fishermen Reel From Disaster | Bridge The Gulf Project 4/30/12 10:38 PM

your team's franchise-tying 11th direct get in FedExForum, along with clinched homecourt advantages inside 1st rounded in the playoffs. Various other nominees pertaining to Instructor in the Thirty day period ended up Atlanta's Lewis Drew, Boston's Doc Waters, Denver's George Karl, Brand-new York's Henry Woodson, San Antonio's Gregg Popovich, along with Utah's Tyrone Corbin. WALTHAM, Nike Darren McFadden Jersey (http://www.nflraidersproshop.com/darren-mcfadden-nike-jersey/) Muscle size. (AP) -- Paul Pierce ended up being rear in train to the Boston ma Boston celtics in Feb 5th, every day immediately after lightly making your regular-season finish which has a sprained quit major feet. Pierce explained it turned out any small trouble along with they don't count on the idea for you to hassle your ex inside first-round string resistant to the The atlanta area Hawks that will commences On the. Pierce quit Thurs night evening of sport first inside 1st fraction plus the Boston celtics explained they has not been planning to give back, nevertheless they ended up being rear for the the courtroom delayed inside subsequent fraction. Beam Allen would not train in Feb 5th. Instructor Doc Waters explained they don't recognize no matter whether Allen would be able to participate in in On the. Allen features have missed a fortnight which has a painful appropriate foot.

http://bridgethegulfproject.org/node/633 Page 4 of 4 PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

RESEARCH ARTICLE Response of Coastal Fishes to the Gulf of Mexico Oil Disaster

F. Joel Fodrie1*, Kenneth L. Heck Jr.2 1 Institute of Marine Sciences and Department of Marine Sciences, University of North Carolina at Chapel Hill, Morehead City, North Carolina, United States of America, 2 Dauphin Island Sea Lab and Department of Marine Sciences, University of South Alabama, Dauphin Island, Alabama, United States of America

Abstract

The ecosystem-level impacts of the Deepwater Horizon disaster have been largely unpredictable due to the unique setting and magnitude of this spill. We used a five-year (2006–2010) data set within the oil-affected region to explore acute consequences for early-stage survival of fish species inhabiting seagrass nursery habitat. Although many of these species spawned during spring-summer, and produced larvae vulnerable to oil-polluted water, overall and species- by-species catch rates were high in 2010 after the spill (1,989±220 fishes km-towed−1 [µ ± 1SE]) relative to the previous four years (1,080±43 fishes km-towed−1). Also, several exploited species were characterized by notably higher juvenile catch rates during 2010 following large-scale fisheries closures in the northern Gulf, although overall statistical results for the effects of fishery closures on assemblage-wide CPUE data were ambiguous. We conclude that immediate, catastrophic losses of 2010 cohorts were largely avoided, and that no shifts in species composition occurred following the spill. The potential long-term impacts facing fishes as a result of chronic exposure and delayed, indirect effects now require attention.

Citation: Fodrie FJ, Heck KL Jr (2011) Response of Coastal Fishes to the Gulf of Mexico Oil Disaster. PLoS ONE 6(7):

e21609. doi:10.1371/journal.pone.0021609

Editor: Steven J. Bograd, National Oceanic and Atmospheric Administration/National Marine Fisheries Service/Southwest

Fisheries Science Center, United States of America

Received: March 7, 2011; Accepted: June 2, 2011; Published: July 6, 2011

Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original

author and source are credited.

Funding: The authors acknowledge support from the National Marine Fisheries Service, National Oceanic and

Atmospheric Administration Marine Fisheries Initiative and Northern Gulf Institute. The funders had no role in study

design, data collection and analyses, decision to publish, or preparation of the manuscript.

1 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

Competing interests: The authors have declared that no competing interests exist.

* E-mail: [email protected]

INTRODUCTION

Prevailing models of ecological impacts resulting from oil pollution are being revised after the April 2010 release of ~4.4 million barrels [1] of oil into the northern Gulf of Mexico (GOM). In part, this is a legacy of the Exxon Valdez accident as a watershed environmental catastrophe, and the extensive research on acute and chronic impacts of the resulting inshore oil pollution [2]. Unlike the 0.25–0.5 million barrels released by the Valdez [2], however, the Deepwater Horizon (DH) disaster hemorrhaged oil into the open ocean at 1500 m depth over a protracted 84-day period [1]. As a critical step toward new model development applicable for detecting impacts of the DH spill, rigorous observational data at organismal through community levels are needed to guide ecosystem-based toxicology.

We have already learned that a significant fraction of the oil released into the GOM from the Macondo well did not rise to the surface, and this has implications for the ecosystem-level responses we should anticipate. Rather, oil was emulsified at the well head due to turbulent mixing, reduced buoyancy at depth, and addition of Corexit 9500 dispersant. Subsequently, mid-water hydrocarbon plumes [3] have been observed with stimulation of petroleum-degrading bacteria [4]. With this now understood, we revisit some early concerns regarding impacts for nearshore fisheries [5].

During the DH spill, near-surface waters lacked any reliable refuge from oil pollution, as slicks/sheens occurred at the immediate surface and oil was emulsified throughout the water column. For many fishes, including commercially valuable snappers (Lutjanidae) and groupers (Serranidae), spawning occurs during the spring or summer (table S1), and eggs, larvae and post-larvae would have relied upon near-surface waters overlaying the continental shelf during the DH spill [6]–[7]. Furthermore, eggs/larvae and oil can be transported by the same hydrodynamic and atmospheric processes, enhancing the probability of oil encounters for many species. Because the population ecology of marine species with bipartite life histories is disproportionately affected by the health and survival of early life stages [8], understanding how eggs, larvae and newly-settled juveniles coped with the DH spill is essential for quantifying ecosystem responses.

We hypothesized that the strength of juvenile cohorts spawned on the northern GOM continental shelf during May–September 2010 in the northern GOM would be negatively affected by egg/larval-oil interactions. Oiled seawater contains toxic compounds such as polycyclic aromatic hydrocarbons (PAHs) which, even after weathering, can result in genetic damage, physical deformities and altered developmental timing for fish eggs/larvae [9]–[10]. These effects may be

2 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

induced at very low (~1 ppb PAHs) levels of exposure when persistent over days to weeks [11]–[12] - timescales relevant for larval development and descriptive of the DH spill. Additionally, emulsified oil droplets could mechanically damage the feeding and breathing apparatus of relatively fragile larvae and further decrease individual fitness. Unfortunately, observing egg/larval mortality, growth or migration in situ is an enduring challenge for biological oceanographers, as eggs/larvae are simply too dilute and experience relatively high instantaneous mortality, even in undisturbed systems [13].

In the absence of direct observations on eggs and larvae, juvenile abundance data provide valuable indices of the acute, population-level responses of young fishes to the spill. Although indirect evidence [14], early juvenile abundances are the integrated products of early life-history processes such as fertilization, larval growth/mortality, and settlement [6]–[8]. Therefore, effects of oil pollution on early life stages should be detectable in time series data as shifts in the abundance of recently settled juvenile fishes. We tested these predictions using 2006–2010 survey data collected from the Chandeleur Islands, LA, to Saint Joseph Bay, FL (Fig. 1), representing most of the nearshore region directly impacted by oil. In contrast to the difficulties of surveying marine larvae, quantitative measures of juvenile abundances are tractable due to the tendency of settled fish to aggregate in specialized nursery habitats [15]. In the northern GOM, many fish species, such as those in the drum (Sciaenidae), snapper and grouper families have juveniles that are routinely collected from shallow-water seagrass meadows they use as primary nurseries [16].

Figure 1. Sampling region and study sites. Map of juvenile fish sampling stations, divided among four survey areas: Chandeleur Islands (blue circles), Gulf Islands (green circles), Grand Bay (orange circles) and Florida Bays (red circles). 1. Chandeleur Is., LA; 2. Ship Is., MS; 3. Horn Is., MS; 4. Petit Bois Is., MS; 5. Dauphin Is., AL; 6. Grand Bature Shoal, AL; 7. Point Aux Pines, AL, 8. Big Lagoon, FL; 9. Pensacola Bay, FL; 10. Choctawhatchee Bay, FL; 11. St. Andrew Sound, FL; 12. St. Joseph Bay, FL. The spread of surface oil during the 84-day spill is also shown (brown shading). Image at lower right shows juvenile gray snapper (L. griseus), spotted seatrout (C. nebulosus) and pipefish (Syngnathus spp.).

doi:10.1371/journal.pone.0021609.g001

Our dataset consisted of 853 individual trawl samples taken between July 15 and October 31 of 2006–2010 within seagrass meadows of the northern GOM (tables S2, S3). We collected 167,740 individual fishes representing 86 taxa, and examined catch-per-unit-effort (CPUE) data for all species pooled together, as well as separately for each of the 20 most abundant species. We also tested for post-spill community-level shifts in seagrass-associated fish assemblages using multivariate analyses [17]. We recognized that not all species were at equal risk for oil exposure due to variation in spawning timing and larval distributions (tables S1, S4). Furthermore, some species may have experienced release from fishing pressure due to large-scale fishery closures [18]

3 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

during the spill (table S5), perhaps enhancing their larval production during the summer spawning season. Therefore, we also considered how these factors affected species-specific CPUEs during 2010. In all analyses, comparisons among years were considered as a proxy for the effects of oil disturbance (2006–2009 as undisturbed, 2010 as disturbed).

RESULTS

Within the oil-affected GOM, a five-year survey of seagrass-associated fish communities did not indicate reductions in juvenile abundances following the spill. Rather, of the twenty most commonly collected fish species, twelve were characterized by statistically higher catch rates in 2010 relative to 2006–20009 (α = 0.05; Table 1). Among the remaining eight taxa, pre- and post-spill catch rates were statistically indistinguishable. Across our entire study region, CPUE increased from 1,080±43 fishes km-towed−1 (µ ± 1SE) during 2006–2009 to 1,989±220 fishes km-towed−1 in 2010. When resolved among four geographical areas (Chandeleur Islands, Gulf Islands, Grand Bay, Florida Bays; Fig. 1), overall catch rates of juvenile fishes, as well as CPUE of the most abundant species, pinfish (Lagodon rhomboides), were consistently higher during 2010 than in 2008 or 2009, and in some areas were higher in 2010 than all previous years (Fig. 2A–B; fig. S1, S2, S3; table S6).

Figure 2. Catch rates of juvenile fishes, 2006–2010. Catch rates among years and sampling areas (Chandeleur Islands, Gulf Islands, Grand Bay and Florida Bays) for: (A) all fishes pooled; (B) pinfish (L. rhomboides), (C) gray snapper (L. griseus), and (D) spotted seatrout (C. nebulosus). CPUE data in panels B–D are presented on a log scale, and all data are shown as means of trawl samples (µ + 1SE).

doi:10.1371/journal.pone.0021609.g002

Table 1. Relative frequencies and CPUE data for abundant fishes collected during sampling in seagrass meadows of the northern GOM.

doi:10.1371/journal.pone.0021609.t001

The species composition of juvenile fish assemblages was unaltered in each sampling area during the months following the DH disaster (Fig. 3). Prior to the spill, similarities among individual trawl samples (SIMPER) ranged from 50.3% at the Chandeleur Islands to 52.9% within Florida Bays (table S7). By comparison, similarity percentages between pre- (2006–2009) and post-spill (2010) samples were correspondingly high, ranging from 43.4% within Grand Bay to 50.8% at the Chandeleur Islands. Furthermore, pinfish, silver perch (Bairdiella chyrsoura), mojarras (Eucinostomus spp.), pigfish (Orthopristis chrysoptera) and spotted seatrout (Cynoscion nebulosus) drove similarity patterns both before and after the spill (table S7). Species richness (S, up 15%,

p<0.001), diversity (ES(20), up 11%, p = 0.006; H′, up 18%, p<0.001) and evenness (J′, up 11%, p = 0.003) among trawl samples were all slightly elevated during 2010 relative to 2006–2009

4 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

averages (table S8; fig. S4), indicating that high CPUEs in 2010 were broad based.

Figure 3. Community composition of seagrass-associated fish communities, 2006–2010. Multi-dimensional scaling plots for seagrass-associated fish communities prior to (2006–2009; colored symbols) and following (2010; black circles) the DH spill. Data for (A) Chandeleur Islands, (B) Gulf Islands, (C) Grand Bay and (D) Florida Bays are presented separately. Each datum represents a single trawl sample.

doi:10.1371/journal.pone.0021609.g003

When averaged across species, there was little statistical evidence that either exposure risk or release from fishing pressure significantly affected CPUEs during 2010. When comparing 2010 CPUE data against pre-spill catch rates, we did observe that fishes characterized by moderate (spring spawning, nearshore larvae) or high risk (spring-summer spawning, larvae distributed across the continental shelf) exhibited decreases in CPUE following the spill at the Chandeleur Islands and Grand Bay (Fig. 4A). However, no statistically significant differences were found as a function of

egg/larval risk (F4,848 = 1.410, p = 0.242) or sampling areas (F3,849 = 0.999, p = 0.440; table S9). Similarly, release from fishing pressure on spawning fishes could be implicated, although not proven, as a determinant of post-spill CPUEs. Along the Chandeleur and Gulf Islands, increases in catch rates during 2010 relative to 2006–2009 were 800% and 950% higher, respectively (Fig. 4B), for species identified in state and federal management plans than for species not harvested by fishermen (table S5). No similar patterns were recorded within Grand Bay or Florida Bays, however,

and effects of fishing pressure (F1,851 = 1.510, p = 0.223) and area (F3,849 = 1.397, p = 0.225) on CPUE responses were not significant.

Figure 4. Larval risk and fishery closure impacts. Effects of (A) egg/larval vulnerability and (B) harvest pressure on the responses of fishes to the DH spill. Response of individual species calculated as the ratio of 2010 versus 2006–2009 CPUE data. Data are presented on a log scale as group means (µ + 1SE), with ratios >1 indicating that 2010 catch rates were elevated relative to 2006–2010 data.

doi:10.1371/journal.pone.0021609.g004

DISCUSSION

Collectively, no significant, acute impacts on the strength of juvenile cohorts within seagrass habitats were detected following the DH disaster. This was true for all species examined, bolstering our confidence in the conclusion that ecosystem-level injuries were not severe for this community of fishes. Unfortunately, our assessment cannot be compared to the most analogous spill, the IXTOC 1

5 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

blowout [5], due to a paucity of formal scientific investigation following that accident (The 1979 IXTOC I blowout at 3600 m depth, 80-km north of the Yucatán Peninsula, was a ~3.5-million-barrel spill.). The most parsimonious explanation for our data is that these fishes were resilient to the spill, possibly due to the retention of a large proportion of spilled oil at depth. As such, these data add to a developing literature [3]–[4] in which the acute impacts of the spill may be concentrated in the deep ocean rather than shallow-water, coastal ecosystems that were the focus of early concern [5]. For instance, gray snapper (Lutjanus griseus) larvae were abundant in surface waters (0–25-m deep) over the continental shelf from July through September [19], and were among the most likely individuals to have contacted oil-polluted water. Still, catch rates of gray snapper juveniles following the spill were high relative to the four previous years (up 82%, Fig. 2C; area * pre/post spill context interaction p<0.001, table S6).

When averaged across species - and in some cases across species with vastly different life histories - there were no statistically significant differences in the response of fished or unfished species to the spill (or their responses to subsequent management actions; i.e., fishery closures). Still, there were notable patterns suggesting that certain species may have been released from harvest pressure during 2010, subsequently enhancing spawning activity and post-spill cohort sizes despite any potentially negative oil impacts. For example, spotted seatrout spawn during summer [20], but many mature individuals are typically removed by recreational fishers before reproducing. Following the fishery closures in 2010, we recorded order-of-magnitude higher juvenile abundances of spotted seatrout at the Chandeleur and Gulf Islands, as well as elevated catch rates throughout our survey region (Fig. 2D; area, pre/post spill context and 2-way interaction p<0.001, table S6).

Consistent with the patterns observed in the species-by-species catch data and analyses of ‘risk” or ‘fishing” effects, there were no significant post-spill shifts in community composition and structure, nor were there changes in any of several biodiversity measures. While natural recruitment variability can make it difficult to detect population-level impacts for any one species following large-scale disturbance [14], our whole-community analyses and results are likely robust against these concerns.

Several other factors could have contributed to the high catch rates of seagrass-associated fishes in 2010 despite large-scale oil pollution. For instance, fishes may be uniquely buffered against oil pollution due to their mobility or foraging ecology [21]–[22]. Also, the major predators of fish eggs/larvae (e.g., gelatinous zooplankton) may have been impacted by the spill, thereby reducing natural mortality rates for coastal fishes [23]. Regardless of the mechanism(s) involved, thus far the potential for 2010 cohorts to support regional fisheries over the next several years has persisted despite the spill. This information is critical for projecting the mode and tempo of ecological and economic recovery in the oil-affected GOM, as well as guiding future conservation/restoration activities to mitigate oil-spill injuries.

While these data provide reason for early optimism, attention should now turn to possible chronic effects of oil exposure on fishes as well as delayed indirect effects cascading through the post-spill

6 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

GOM. Fish may suffer growth, survival or reproductive penalties years after exposure to oil [24], or experience altered migratory behaviors [25]. Oil sequestered in sediments may also affect species laying benthic eggs for several years [26]. More broadly, ecosystems experiencing large-scale disturbance can carry or build instabilities over protracted periods that can eventually result in delayed collapses of fisheries stocks [27].

Improved threat assessment for energy exploration [28] and process-oriented studies of ecosystem responses will be long-term initiatives resulting from the DH spill. In the short term, however, observational data collected over relevant spatial and temporal scales are invaluable for guiding and evaluating targeted studies of oil toxicology [29]. For fish species experiencing multiple stressors such as habitat degradation [30] harvest pressure [31], climate change [16], and now oil pollution, rigorous baseline survey data and new syntheses are needed to enact effective ecosystem-based management.

MATERIALS AND METHODS

Sampling

We analyzed changes in northern Gulf of Mexico (GOM) seagrass-associated fish communities during the last 5 years by comparing survey data obtained either prior to (2006–2009) or immediately following the Deepwater Horizon disaster (2010). The survey region extended approximately 340 km along the coastline, including a significant portion of the inshore area most affected by oil (Fig. 1.). Each year, we made repeated sampling trips to 12 sites, extending from the Chandeleur Islands, LA, to St. Joseph Bay, FL (29.68–30.72°N, 85.30–89.10°W). Sampling occurred within mixed seagrass meadows that serve as primary nursery habitat for many juvenile fishes that have recently settled from the water column following a 5–45 day larval period [6], [16]. Our samples were collected from seagrass mosaics that included turtle grass (Thalassia testudinum), shoal grass (Halodule wrightii), widgeon grass (Ruppia maritima), and manatee grass (Syringodium filiforme), along with scattered unvegetated patches (table S3).

During each year, trawls were conducted from July 15 through October 31 in order to record the abundances and composition of fishes during the period when seagrass meadows are utilized as primary nurseries by recently settled juveniles (refer to table S1 for reproductive seasons of common fishes). Fishes were collected using a 5-m otter trawl (2.0-cm body mesh; 0.6-cm bag mesh; 0.3×0.7-m doors) with conventional 4-seam balloon design including float and lead lines but without tickler chains. Trawls consisted of 3.9±0.1 (µ ± 1SE) minute tows behind small (<7 m) research vessels traveling at 3.3+0.1 kilometers hour−1. Overall, 853 samples were taken (table S2), and the trawl covered a linear distance of 184.7 kilometers during our sampling. These trawls occurred in depths of 0.5–2.5-m, and were conducted during daylight hours. During our surveys, species were enumerated in the field unless species-level identifications were not easily made. Unidentified specimens were transported to the lab where meristics were used by at least two different technicians to identify each individual. In cases in which species could not be identified,

7 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

specimens were classified to the lowest taxonomic level possible. Typically, fishes were 20–100 mm long (standard length), indicative of recently-spawned juveniles. However, we did not record individual sizes for all species, and, for pipefishes (Syngnathus spp.) and hard-headed catfish (Arius felis), we did observe that a small proportion of our catch included reproductive adults. For four species: gray snapper (50.5±0.8 mm [µ ± SE]), lane snapper (Lutjanus synagris; 55.7±0.7 mm), spotted seatrout (60.8±1.1 mm) and gag grouper (Mycteroperca microlepis; 157.5±3.2 mm); we did record the sizes of all individuals throughout our surveys. Based on our own otolith analyses (Fodrie unpublished) and published reports of first-year growth among these four species (age-1 sizes: gray snapper ~109 mm; lane snapper ~140 mm; spotted seatrout ~127 mm; gag grouper >198 mm), we calculated that >96% of individuals were captured in the same year as they were spawned (including 2010).

Once enumerated, fishes were entered in to an Excel database, and abundance data were converted into catch-per-unit-effort (CPUE) data based on the linear distance over with each trawl occurred. All statistical analyses were applied to these CPUE data. Our complete CPUE dataset is included as a separate appendix in our supporting information (dataset S1). This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. Our sampling protocol was approved by the Committee on the Ethics of Animal Experiments of the University of North Carolina at Chapel Hill (Permit Number: 10-114.0).

Statistical analyses

We investigated differences in the catch rates of seagrass-associated fishes (all species pooled as well as the 20 most abundant species individually) by unpaired t-tests comparing pre- (2006–2010) and post-spill (2010) data (Table 1), as well as 2-way ANOVAs in which sampling area (Chandeleur Islands, Gulf Islands, Grand Bay, Florida Bays) and context (pre- versus post-spill) were fixed factors (table S6). Regions were defined by basic geomorphology and location, local water clarity, local salinity, and local seagrass composition [32]. Because variances were stable among groups, no data transformations were required prior to analyses.

We analyzed similarities and differences in fish communities among years (2006–2009 versus 2010) within each sampling area (each area considered separately) using non-metric multidimensional scaling (MDS), based on Bray-Curtis similarity indices among all individual trawl samples (4th root-transformed data). Pairwise comparisons between trawl samples across years were conducted with analysis of similarity (ANOSIM) and similarity (or dissimilarity) percentages (SIMPER) using PRIMER 5.2.2 software (PRIMER-E Ltd; [33]).

We also examined patterns of species diversity among regions and years by computing the following

measures for each trawl sample: S, number of species collected; ES(20), species richness rarefied to

20 individuals; H′, Shannon-Weiner diversity index (loge); and J′, Pielou's evenness measure (PRIMER 5.2.2 software). We investigated differences in community diversity via 2-way ANOVAs in

8 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

which sampling area (Chandeleur Islands, Gulf Islands, Grand Bay, Florida Bays) and context (pre- versus post-spill) were fixed factors. Because variances were stable among groups, no data transformations were required prior to analyses.

These approaches are proscribed in earlier syntheses for detecting environmental impacts [17]. Critiques of employing parametric testing to detect ecosystem injury exist due to interannual variability and reduced statistical power [14], although those concerns have focused on analyses involving single species.

We determined the relative probability for oil-larvae encounters (‘risk’) for the twenty most commonly captured fishes, and used these data to explore how individual species responded differently to large-scale oil pollution in the northern GOM. Information on the seasonal timing of spawning and distribution of larvae from shore to the outer margin of the continental shelf was collected from the literature (tables S1 and S4), and used to define 4 levels of risk (in addition to an ‘unknown’ [n = 4] category containing species for which no data were available). ‘Low’ risk species (n = 6) included those in which larvae remained inside estuaries, either in the plankton or as benthic egg masses, regardless of spawning season. ‘Moderate-Low’ risk species (n = 4) were defined by having either 1) larvae distributed in estuaries as well as across the continental shelf, or 2) larvae distributed across the continental shelf, but not likely during the spill period (i.e., April– September). Only two ‘Moderate’ risk species were identified: pigfish (Orthopristis chrysoptera) spawn throughout summer, and have larvae distributed within nearshore waters; while flounder (Paralichthys spp.) have larvae distributed across the continental shelf, with a protracted spawning that extends into April or May (i.e., some overlap with the oil spill). ‘High’ risk species (n = 4) spawn offshore and have larvae distributed across the continental shelf. Furthermore, spawning data indicates that these species would have produced larvae sometime during the DH spill (April–Sept in our classification scheme). Based on these risk guilds, we examined the response of fishes to the spill by calculating the ratio of 2010 CPUE data (averaged) to 2006–2009 CPUE data (averaged) for each species. Following these calculations, ratios >1 indicate that average 2010 catch rates were higher than during the previous 4 years, while ratios <1 indicate that average 2010 catch rates were lower than during the previous 4 years. Using each species as a replicate measure, we used ‘risk’ and region (Chandeleur Islands, Gulf Islands, Grand Bay, Florida Bays) as fixed factors in a 2-way ANOVA that compared 2010 CPUE: 2006–2009 CPUE trends. Because variances were stable among groups, no data transformations were required prior to analyses.

Similarly, we determined whether species were likely to have experienced significant release from harvest pressure following large-scale closures in the northern GOM, and examined how this may have affected CPUE data in 2010. For each of the twenty most commonly caught fish, we designated species as ‘fished’ if they were included in any state or federal management plan as of Dec 31, 2010 (table S5), or identified as <1% (by biomass) of bycatch in shrimp trawl fisheries within the northern GOM (table S5). As before, we examined the response of fishes to the spill by calculating the ratio of 2010 CPUE data (averaged) to 2006–2009 CPUE data (averaged) for each species. Using each species as a replicate measure, we used ‘fishing pressure’ (with fished species including

9 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

species that are targeted or captured as incidental bycatch at significant levels) and region (Chandeleur Islands, Gulf Islands, Grand Bay, Florida Bays) as fixed factors in a 2-way ANOVA that compared 2010 CPUE: 2006–2009 CPUE trends. Because variances were stable among groups, no data transformations were required prior to analyses.

All univariate tests were conducted using StatView 5.0.1 software (SAS Institute Inc.). Because each statistical analysis applied to separate and easily distinguishable hypotheses, we made no corrections to experiment-wise alpha for any of the univariate (total fishes CPUE, individual fishes CPUE, risk guilds, harvest guilds, diversity) or multivariate (ANOSIM) tests we conducted [34].

SUPPORTING INFORMATION

Figure S1.

Catch rates of all fishes, pooled together, among sampling areas prior to (2006–2009) and following (2010) the Deepwater Horizon disaster.

(DOCX)

Figure S2.

Catch rates of individual species, among sampling areas prior to (2006–2009) and following (2010) the Deepwater Horizon disaster. Data are presented for the 20 most abundant species.

(DOCX)

Figure S3.

Catch rates among sampling areas and years for the 20 most abundant species collected during trawl surveys.

(DOCX)

Figure S4.

Diversity measures for seagrass-associated fish communities within sampling areas affected by the Deepwater Horizon disaster.

(DOCX)

Table S1.

Summary table for CPUE data (fish kilometer-towed−1) of fishes prior to (2006–2009) and following (2010) the DH disaster.

(DOCX)

10 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

Table S2.

Distribution of trawl samples among sampling areas (Chandeleur Islands, Gulf Islands, Grand Bay, Florida Bays) and years (2006–2010).

(DOCX)

Table S3.

Quantitative description of seagrass habitats sampled throughout the northern Gulf of Mexico during 2006–2010.

(DOCX)

Table S4.

Information used to determine the likelihood of larvae contacting oiled water during the summer of 2010.

(DOCX)

Table S5.

Summary table for the management status of the 20 most abundant fishes collected during our survey program.

(DOCX)

Table S6.

Summary table of the effects of sampling area and year (context: pre- versus post-spill) on the catch rates of the 20 most abundant fishes collected during surveys in northern Gulf of Mexico seagrass meadows.

(DOCX)

Table S7.

Comparisons of community structure between catch data prior to (2006–2009) or immediately following (2010) the Deepwater Horizon disaster (ANOSIM and SIMPER).

(DOCX)

Table S8.

Summary table of the effects of sampling area and year (context: pre- versus post-spill) on the

diversity (S, ES(20), H′, and J′) of trawl samples collected within northern Gulf of Mexico seagrass meadows.

11 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

(DOCX)

Table S9.

Summary table of the effects of sampling area, larval risk and harvest pressure on the change in catch rates of individual species for pre- (2006–2009) and post-spill (2010) data.

(DOCX)

Dataset S1.

Complete CPUE data obtained for 2006–2009 trawl surveys within seagrass meadows of the northern Gulf of Mexico.

(XLSX)

ACKNOWLEDGMENTS

We are extremely grateful to the students and technicians who aided in the field, especially C. Baillie, M. Brodeur, J. Myers, O. Rhoades and S. Williams. B. Raines supplied the image of juvenile fishes in Fig. 1. Constructive comments and support were provided by S. Powers, C. Peterson, J. Kenworthy, and 2 anonymous reviewers.

AUTHOR CONTRIBUTIONS

Conceived and designed the experiments: FJF KLH. Performed the experiments: FJF KLH. Analyzed the data: FJF. Wrote the paper: FJF KLH.

REFERENCES

1. Crone TJ, Tolstoy M (2010) Magnitude of the 2010 Gulf of Mexico Oil Leak. Science 330: 634–643.

2. Peterson CH, Rice SD, Short JW, Esler D, Bodkin JL, et al. (2003) Long-term ecosystem response to the Exxon Valdez Oil Spill. Science 302: 2082–2086.

3. Camilli R, Reddy CM, Yoerger DR, Van Mooy BAS, Jukuba MV, et al. (2010) Tracking hydrocarbon plume transport and biodegradation at Deepwater Horizon. Science 330: 201–204.

4. Hanzen TC, Dubinsky EA, DeSantic TZ, Andersen GL, Piceno YM, et al. (2010) Deep-sea oil plume enriches indigenous oil-degrading bacteria. Science 330: 204–208.

5. Kerr R, Kintisch E, Stokstad E (2010) Will Deepwater Horizon set a new standard for catastrophe? Science 328: 674–675.

12 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

6. Able KW, Fahay MP (2010) 566 p.

7. Miller BS, Kendall AW (2009) 376 p.

8. Hjort J (1914) Fluctuation in the great fisheries of northern Europe reviewed in the light of biological research. Rapports et Procès-Verbaux des Réunions du Conseil International pour l'Exploration de la Mer 20: 1–228.

9. Kocan RM, Hose JE, Brown ED, Baker TT (1996) Injury to the early life history stages of Pacific herring in Prince William Sound after the Exxon Valdez oil spill. Can J Fish Aquat Sci 53: 2366–2375.

10. Tuvikene A (1995) Responses of fish to polycyclic aromatic hydrocarbons (PAHs). Ann Zool Fennici 32: 295–309.

11. Carls MG, Rice SD, Hose JE (1999) Sensitivity of fish embryos to weathered crude oil: Part I. Low-level exposure during incubation causes malformations, genetic damage, and mortality in larval pacific herring (Clupea pallasi). Environ Toxicol Chem 18: 481–493.

12. Brown ED, Baker TT, Hose JW, Kocan RM, Marty GD, et al. (1996) Injury to the early life history stages of Pacific herring in Prince William Sound after the Exxon Valdez oil spill. Am Fish Soc Sym 18: 448–462.

13. Levin LA (2006) Recent progress in understanding larval dispersal: new directions and digressions. Integr Comp Biol 46: 282–297.

14. Hilborn R (1996) Detecting population impacts from oil spills: a comparison of methodologies. Am Fish Soc Sym 18: 639–644.

15. Beck MW, Heck KL, Able KW, Childers DL, Eggleston DB, et al. (2001) The identification, conservation, and management of estuarine and marine nurseries for fish and invertebrates. Bioscience 51: 633–641.

16. Fodrie FJ, Heck KL, Powers SP, Graham WM, Robinson KL (2010) Climate-related, decadal-scale assemblage changes of seagrass-associated fishes in the northern Gulf of Mexico. Glob Change Biol 16: 48–59.

17. Underwood AJ (1996) pp. 151–175. in Detecting Ecological Impacts: Concepts and Applications in Coastal Habitats, Schmitt RJ, Osenberg CW (Academic Press).

18. NOAA Fisheries Service Deepwater Horizon Oil Spill Information. Available: http://sero.nmfs.noaa.gov

13 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

/deepwater_horizon_oil_spill.htm. Accessed: 2011 Jun 13.

19. D'Alessandro EK, Sponaugle S, Serafy JE (2010) Larval ecology of a suite of snappers (family: Lutjanidae) in the Straits of Florida, western Atlantic Ocean. Mar Ecol Prog Ser 410: 159–175.

20. Holt GJ, Holt SA (2000) Vertical distribution and the role of physical processes in the feeding dynamics of two larval sciaenids Sciaenops ocellatus and Cynoscion nebulosus. Mar Ecol Prog Ser 193: 181–190.

21. Peterson CH, Kennicutt MC, Green RH, Montagna P, Harper DE, et al. (1996) Ecological consequences of environmental perturbations associated with offshore hydrocarbon production: a perspective on long-term exposures in the Gulf of Mexico. Can J Fish Aquat Sci 53: 2637–2654.

22. Edgar GJ, Marshall PA, Mooney P (2003) The effect of the Jessica grounding on Galápagos reef fish communities adjacent to the wreck site. Mar Pollut Bul 47: 296–302.

23. Graham WM, Condon RH, Carmichael RH, D'Ambra I, Patterson HK, et al. (2010) Oil carbon entered the coastal planktonic food web during the Deepwater Horizon oil spill. Environ Res Lett 5: 045301.

24. Heintz RA, Rice SD, Wertheimer AC, Bradshaw RF, Thrower FP, et al. (2000) Delayed effects on growth and marine survival of pink salmon Oncorhynchus gorbuscha after exposure to crude oil during embryonic development. Mar Ecol Prog Ser 208: 205–216.

25. Wertheimer AC, Heintz RA, Thedinga JF, Maselko JM, Rice SD (2000) Straying behavior of adult pink salmon (Oncorhynchus gorbuscha) exposed as embryos to weathered Exxon Valdez crude oil. Trans Am Fish Soc 129: 989–1004.

26. Bue BG, Sharr S, Seeb JE (1998) Evidence of damage to pink salmon populations inhabiting Prince William Sound, Alaska, two generations after the Exxon Valdez Oil Spill. Trans Am Fish Soc 127: 35–43.

27. Pearson WH, Elston RA, Bienert RW, Drum AS, Antrim LD (1999) Why did the Prince William Sound, Alaska, Pacific herring (Clupea pallasi) fisheries collapse in 1993 and 1994. Can J Fish Aquat Sci 56: 711–737.

28. U. S. Department of the Interior, Minerals Management Service Environmental Division, “Oil-Spill Risk Analysis: Gulf of Mexico Outer Continental Shelf (OCS) Lease Sales, Eastern Planning Area, 2003–2007 and Gulfwide OCS Program, 2003–2042” OCS Report MMS 2002–069.

29. Sagarin R, Pauchard A (2010) Observational approaches in ecology open new ground in a changing world. Front Ecol Environ 8: 379–386.

14 of 15 3/19/12 4:07 AM PLoS ONE: Response of Coastal Fishes to the Gulf of Mexico O... http://www.plosone.org/article/info:doi/10.1371/journal.pone.0...

30. Diaz RJ, Rosenberg R (2008) Spreading dead zones and consequences for marine ecosystems. Science 321: 926–929.

31. Baum JK, Myers RA (2004) Shifting baselines and the decline of pelagic sharks in the Gulf of Mexico. Ecol Lett 7: 135–145.

32. Byron D, Heck KL (2006) Hurricane effects on seagrasses along Alabama's Gulf Coast. Estuar Coast 29: 939–942.

33. Clark KR, Gorley RN (2001) PRIMER v5: User Manual/Tutorial. Plymouth: Plymouth Marine Laboratory.

34. Moran MD (2003) Arguments for rejecting the sequential Bonferroni in ecological studies. Oikos 100: 403–405.

All site content, except where otherwise noted, is licensed under a Creative Commons Attribution License.

15 of 15 3/19/12 4:07 AM Roundtable Gulf of Mexico Oil Blowout Increases Risks to Globally Threatened Species

Claudio Campagna, Frederick T. Short, Beth A. Polidoro, Roger McManus, Bruce B. Collette, Nicolas J. Pilcher, Yvonne Sadovy de MitchesoN, Simon N. Stuart, and Kent E. Carpenter

Fourteen marine species in the Gulf of Mexico are protected by the US Endangered Species Act, the Marine Mammal Protection Act, and the Migratory Bird Treaty Act. As the British Petroleum oil spill recovery and remediation proceed, species internationally recognized as having an elevated risk of extinction should also receive priority for protection and restoration efforts, whether or not they have specific legal protection. Forty additional marine species—unprotected by any federal laws—occur in the Gulf and are listed as threatened on the International Union for Conservation of Nature’s (IUCN) Red List. The Red List assessment process scientifically evaluates species’ global status and is therefore a key mechanism for transboundary impact assessments and for coordinating international conservation action. Environmental impact assessments conducted for future offshore oil and gas development should incorporate available data on globally threatened species, including species on the IUCN Red List. This consideration is particularly important because US Natural Resource Damage Assessments may not account for injury to highly migratory, globally threatened species.

Keywords: IUCN Red List, Gulf of Mexico, oil spill, threatened species

primary concern following the British Petroleum categories have an elevated risk of extinction, and we argue A Deepwater Horizon oil well blowout and the resulting they should receive priority for protection and restoration oil pollution in the Gulf of Mexico is the damage to marine efforts in the Gulf, whether or not they have specific legal plants and animals—especially to those already considered protection from any government entity in the region. The vulnerable. Several US federal and state statutes protect Gulf oil blowout is likely to worsen the threat status of some coastal and marine species of special concern found in the of these species as more of the spill’s impacts manifest. Gulf of Mexico, including 14 marine species protected by United States law requires restoration to prevent condi- the US Endangered Species Act (ESA), the Marine Mammal tions of natural resources damaged by the oil pollution, and Protection Act, and the Migratory Bird Treaty Act. Addition- restoration is overseen by NOAA’s (the National Oceanic ally, species protected by Mexican and Cuban laws must be and Atmospheric Administration) Natural Resource Dam- considered. age Assessment (NRDA; NOAA 2010a). The primary legal The International Union for Conservation of Nature authority for assessing damages and providing for recov- (IUCN) Red List of Threatened Species (IUCN 2010) results ery of coastal and marine species is the Oil Pollution Act, from a rigorous scientific process to assess the relative which is implemented by the NRDA. Under the Damage extinction risks of species globally, using widely accepted Assessment Remediation and Restoration Program, NRDA standards (Mace et al. 2008, Hoffmann et al. 2010). As such, trustees determine whether injury to public trust resources the IUCN Red List categories and criteria are the most has occurred, as well as the type, amount, and methods of respected international system for classifying global extinc- restoration needed. tion risk at the species level (De Grammont and Cuarón According to the most recent revision of the Mexican list 2006, Rodrigues et al. 2006, Carpenter et al. 2008, Mace et al. of threatened and protected species (NOM 2002, 2006), all 2008, Schipper et al. 2008). In addition to the 14 species marine mammals and marine turtles are accorded some protected by US law, the IUCN Red List identifies a further degree of protection status in Mexico (e.g., classified as in 39 species in the Gulf (table 1) as belonging to one of three danger of extinction, as threatened, or under special protec- threatened categories: critically endangered, endangered, or tion). Other than mammals and turtles, only three species are vulnerable (IUCN 2001). All species in Red List threatened protected in Mexico: subspecies of two seabirds present in

BioScience 61: 393–397. ISSN 0006-3568, electronic ISSN 1525-3244. © 2011 by American Institute of Biological Sciences. All rights reserved. Request permission to photocopy or reproduce article content at the University of California Press’s Rights and Permissions Web site at www.ucpressjournals.com/ reprintinfo.asp. doi:10.1525/bio.2011.61.5.8 www.biosciencemag.org May 2011 / Vol. 61 No. 5 • BioScience 393 Roundtable Roundtable

Table 1. Marine species in International Union for Conservation of Nature threatened Red List categories (critically endangered, endangered, or vulnerable) that have a distribution directly overlapping the area of the oil spill, or that are found in the greater Gulf region extending from Texas to Miami, Florida. Red List category Common name Protection Red List category Common name Protection species name status species name status

Critically endangered Vulnerable (continued) Lepidochelys kempii Kemp’s ridley turtle ESA-E Epinephelus flavolimbatus Yellowedge grouper Eretmochelys imbricata Hawksbill turtle ESA-E Epinephelus niveatus Snowy grouper Dermochelys coriacea Leatherback turtle ESA-E Mycteroperca interstitialis Yellowmouth grouper Thunnus thynnus Atlantic bluefin tuna, Lachnolaimus maximus Hogfish western stock Alopias superciliosus Bigeye thresher shark Epinephelus drummondhayi Speckled hind Alopias vulpinus Common thresher shark Epinephelus itajara Atlantic goliath grouper Carcharhinus longimanus Oceanic whitetip shark Epinephelus nigritus Warsaw grouper Carcharhinus obscurus Dusky shark Pristis pectinata Smalltooth sawfish ESA-E Carcharhinus plumbeus Sandbar shark Pristis perotteti Largetooth sawfish Carcharhinus signatus Night shark Narcine bancroftii Lesser electric ray Centrophorus granulosus Gulper shark Acropora cervicornis Staghorn coral ESA-T Cetorhinus maximus Basking shark Acropora palmate Elkhorn coral ESA-T Carcharodon carcharias Great white shark Endangered Isurus oxyrinchus Shortfin mako Balaenoptera borealis Serving whale ESA-E, MMPA Isurus paucus Longfin mako Balaenoptera musculus Blue whale ESA-E, MMPA Carcharias taurus Sand tiger shark Balaenoptera physalus Finback whale ESA-E, MMPA Odontaspis ferox Small-tooth sand tiger shark Pterodroma caribbaea Jamaica petrel Rhincodon typus Whale shark Pterodroma hasitata Black-capped petrel MBTA Sphyrna zygaena Smooth hammerhead Caretta caretta Loggerhead turtle ESA-T Squalus acanthias Spiny dogfish Chelonia mydas Green turtle ESA-E, ESA-T (by range) Gymnura altavela Butterfly ray Sphyrna lewini Scalloped hammerhead shark Agaricia lamarcki Lamarck’s sheet coral Sphyrna mokarran Great hammerhead shark Montastraea franksi Montastraea coral Montastraea annularis Boulder star coral Dendrogyra cylindrus Pillar coral Montastraea faveolata Mountainous star coral Dichocoenia stokesii Elliptical star coral Mycetophyllia ferox Rough cactus coral Vulnerable Oculina varicose Large ivory coral Trichechus manatus Manatee ESA-E, MMPA Halophila baillonii Clover seagrass Physeter macrocephalus Sperm whale ESA-E, MMPA

ESA-E, endangered under the Endangered Species Act (ESA); ESA-T, threatened under the ESA; MBTA, listed on the Migratory Bird Treaty Act; MMPA, listed on the Marine Mammal Protection Act. Source: IUCN 2010. See the supplementary table online at dx.doi.org/10.1525/bio.2011.61.5.8. the Gulf of Mexico (Pelecanus occidentalis and Oceanodroma have significant populations, spawning aggregations, or nest- leucorhoa) and the smalltooth sawfish Pristis( pectinata). No ing sites in the Gulf region. Therefore, greater threats in this species-level protection occurs in Cuba comparable to the region may have implications for the species’ global survival. US ESA, but there are laws protecting biodiversity (e.g., Ley Other species (e.g., Kemp’s ridley turtle, Lepidochelys kempii; No. 81 Del Medio Ambiente; Ministerio De Ciencia, Tecno- the western Atlantic population of bluefin tuna, Thunnus logia Y Medio Ambiente Resolucion No. 111/96). thynnus) breed only in the Gulf, and oil spill damage exacer- The Gulf of Mexico has exceptionally high marine biodi- bates previously existing threats to these species. versity, with 15,419 recorded species, of which 10% (1511) IUCN Red List assessments are being expanded to evaluate are endemic (Felder and Camp 2009). This diversity is partly more marine species (http://sci.odu.edu/gmsa/ ), including attributable to the Gulf’s geographic position within the some in the Gulf of Mexico. The IUCN has assessed 322 transition zone between temperate and tropical waters. Some species in the Gulf of Mexico to date, 53 of which are in threatened species in the Gulf (e.g., whale shark, Rhincodon threatened categories (table 1); an additional 29 are listed typus; loggerhead turtle, Caretta caretta) occur globally but as near threatened (see the supplementary table online

394 BioScience • May 2011 / Vol. 61 No. 5 www.biosciencemag.org Roundtable Roundtable

at dx.doi.org/10.1525/bio.2011.61.5.8). The IUCN assess- The whale shark is listed as vulnerable on the IUCN Red ments include all Gulf marine mammals (5 of 28 species List but is not protected by the ESA. Found worldwide in threatened), sea turtles (all 5 species threatened), seagrasses tropical and warm temperate waters, many individuals (2 of 9 threatened or near threatened), mangroves (0 of 6 aggregate in the Gulf of Mexico in the summer months. The threatened), reef-building corals (11 of 60 threatened or whale shark is the largest fish in the world; it feeds almost near threatened), wrasses (1 of 20 threatened), sharks and entirely on plankton, crustaceans, and small fishes. It is one rays (43 of 131 threatened or near threatened), seabirds (3 of only three filter-feeding species of shark, sieving zoo- of 40 threatened or near threatened), and groupers (11 of 22 plankton as small as 1 millimeter in diameter through the threatened or near threatened). Groupers are of particular fine mesh of its gill rakers. The shark’s feeding behavior puts concern; three species are classified as critically endangered it at high risk from the oil itself and the oil dispersants used on the Red List and the Atlantic goliath grouper (Epineph- in the Gulf. Although relatively little is known about the biol- elus itajara) is listed as near extinction. ogy of the whale shark, populations will probably be slow to An oil spill of this magnitude threatens many species recover from disturbances given the species’ estimated long already listed under IUCN threatened categories—more life span, slow reproductive rate (Pauly 2002), and naturally species than are currently protected by the ESA. In 1996, the low abundance outside of mating aggregations. IUCN assessed the western stock of the Atlantic bluefin tuna The Kemp’s ridley sea turtle is listed as critically endan- as critically endangered, and the Convention on Biological gered on the Red List and is also protected by the ESA. Diversity recently petitioned the US Department of Com- This turtle nests exclusively in the Gulf and is the rarest sea merce to protect the species under the ESA (CBD 2010). turtle in the world. Of the threatened marine species that There are two spawning populations of bluefin tuna, one frequent the Gulf, only the Kemp’s ridley depends on Gulf in the Gulf of Mexico and the other in the Mediterranean shores for nesting, and most of its young develop in Gulf Sea. Although there is extensive mixing of the populations waters. Although it appears that the 2010 hatchlings did not on both sides of the Atlantic, particularly on the feeding encounter the spill directly, the timing of the oil spill could grounds off the eastern coast of North America, individu- not have been worse for this species, clashing as it did with als hatched in the Gulf of Mexico return there to spawn the turtles’ key reproductive period. The vast majority of (spawning site fidelity). Peak spawning in the Gulf occurs sea turtles found dead since the spill were Kemp’s ridleys from mid-April to June, unfortunately during the period (NOAA 2010b). The Kemp’s ridley was just on the road of the British Petroleum oil spill in 2010. Like tuna, many to recovery after a population collapse a few decades ago other species in threatened Red List categories—that are not that drove it near extinction; the species now faces a new protected by the ESA—require protection and remediation environmental hurdle. from the oil spill. The West Indian manatee (Trichechus manatus) is listed Seagrasses are a unique group of 72 undersea flowering as vulnerable by the IUCN and is considered endangered plants found in coastal seas globally. In the Gulf of Mexico, under US law. Manatees are found in the Gulf and around there are nine seagrass species, and these plants provide the coastline of Florida, in the range of the oil spill. Manatees crucial structural habitat and nursery grounds for many may be affected by air quality and oil at the surface, which recreationally and commercially important fish and inver- they encounter as they emerge to breathe; oil irritating their tebrates (including Gulf pink shrimp and brown shrimp), skin and eyes; the consumption of seagrass species—their as well as waterfowl. Some seagrasses, as indicated by their primary food—that are covered in oil; and chemical oil common names (e.g., turtle grass and manatee grass) are the dispersants that may also have a toxic effect. The Florida primary food for already threatened species of sea turtles and manatee (Trichechus manatus latirostris), a subspecies of the manatees. The seagrass habitat, and the proliferation of the West Indian manatee, is additionally threatened by loss of species it supports, may be at risk as a result of the oil spill; habitat, entanglement with fishing gear, and increased boat- three diminutive seagrass species of the genus Halophila are ing activity, as well as extreme cold temperatures that killed most threatened. Halophila baillonii is listed as vulnerable 10% of the population during the winter of 2009–2010. The and Halophila engelmanni is listed as near threatened on the Florida manatee subspecies was listed as endangered in 2008 Red List (Short et al. 2011), and Halophila johnsonii is listed by the IUCN. on the ESA. The limited global distributions of these spe- The trends in species declines are clearly worrying, par- cies, restricted primarily to Gulf and Florida waters in the ticularly because the Gulf was already a system affected by cases of H. engelmanni and H. johnsonii, mean their risk of various risk factors before the oil blowout occurred. How global extinction could be elevated by the oil spill. Halophila can we adequately address the threats of oil and gas develop- baillonii, already rapidly declining in the Caribbean, is the ment and steward the Gulf’s biological diversity? Priorities fourth most threatened seagrass species in the world. Potential at this stage must focus on species with high commercial damages to these seagrasses from the oil pollution in the Gulf value, species critical to the integrity of coastal and marine should be assessed, and recovery actions for these species ecosystems in the Gulf, species with populations in decline should be aided by funding available from the Oil Pollution before the blowout, and species now recognized as in greater Act and other sources. danger of extinction. Because marine species in particular

www.biosciencemag.org May 2011 / Vol. 61 No. 5 • BioScience 395 Roundtable Roundtable may be underrepresented by the ESA (Wilcove and Master Acknowledgments 2005), the ongoing NRDA in the Gulf of Mexico—as well as The majority of marine species assessments conducted environmental impact assessments conducted for offshore through the International Union for Conservation of Nature oil and gas development—should include available data on (IUCN) Species Survival Commission are made through globally threatened species, including the expanding species the Global Marine Species Assessment, with core funding data sets on the IUCN Red List. Species information on the provided by Tom Haas and the New Hampshire Chari- Red List can serve as a standardized mechanism to identify table Foundation. We thank numerous partners who helped and coordinate conservation and mitigation priorities, espe- compile information, including BirdLife International; cially for highly migratory and transboundary species. The SeagrassNet; the Groupers and Wrasses, Tunas and Billfishes, US Department of the Interior must reevaluate the “low Sharks, and Marine Turtles IUCN Species Specialist Groups; risk” status currently attributed to pollution from routine Jonnell Sanciangco and Suzanne Livingstone (Global Marine operations of oil and gas development, as well the poten- Species Assessment); and Cynthia Taylor (Sirenia Red List tial impacts of catastrophic events such as oil spills, in its Authority Focal Point). Thanks to Cathy Short for editing compliance with the National Environmental Policy Act, the the manuscript. This article is Jackson Estuarine Laboratory ESA, and other applicable domestic and international laws. contribution no. 498. Species identified as threatened with extinction on the IUCN Red List may become even more threatened as a result of the oil spill. Such elevations in threatened status are part References cited Carpenter KE, et al. 2008. One-third of reef-building corals face elevated of the spill’s impacts and as such are damages that must be extinction risk from climate change and local impacts. Science 321: recognized and compensated. The six threatened grouper 560–563. species on the Red List that occur in the Gulf, for example, [CBD] Center for Biological Diversity. 2010. Petition to List the Atlantic currently receive no protection under the ESA or Mexican Bluefin Tuna (Thunnus thynnus) as Endangered under the United States law, despite their status as a major food resource in the Endangered Species Act. CBD. (2 February 2011; www.nmfs.noaa.gov/ pr/pdfs/species/cbd_bluefintunapetition_5242010.pdf) region and a high-value restaurant menu item. De Grammont PC, Cuarón AD. 2006. An evaluation of threatened species Gulf-occurring animals and plants protected by the ESA categorization systems used on the American continent. Conservation (and other US laws) should be priorities for federal dam- Biology 20: 14–27. age assessments; as such, these laws should help restore the Felder DL, Camp DK, eds. 2009. Biodiversity, vol. 1. Gulf of Mexico Origin, natural resources injured by the release of oil or hazardous Waters, and Biota. Texas A&M University Press. Hoffmann M, et al. 2010. The impact of conservation on the status of the substances. Although the methodology of assessment and world’s vertebrates. Science 330: 1503–1509. the names of threatened categories may differ among laws, [IUCN] International Union for Conservation of Nature. 2001. IUCN Red List assessments, and criteria, the IUCN Red List is a highly cred- Categories and Criteria, version 3.1. (2 February 2011; www.iucnredlist. ible source of species requiring particular attention—both org/technical-documents/categories-and-criteria/2001-categories-criteria). for damage assessment and for special consideration for ———. 2010. IUCN Red List. (2 February 2011; www.iucnredlist.org). Mace GM, Collar NJ, Gaston KJ, Hilton-Taylor C, Akçakaya HR, Leader- future regulations of oil and gas development. As a result Williams N, Milner-Gulland EJ, Stuart SN. 2008. Quantification of of the rapid increase in IUCN assessments during the last extinction risk: The background to IUCN’s system for classifying threat- few years, we now know there are many threatened marine ened species. Conservation Biology 22: 1424–1442. species in the Gulf that are not currently protected by US [NOAA] National Oceanic and Atmospheric Administration. 2010a. law (table 1). Threatened species not yet listed in national US Natural Resource Damage Assessment, Damage Assessment Remediation and Restoration Program. (2 February 2011; www.darrp. legislation should nevertheless be the subject of damage noaa.gov) assessments, targeted research, and monitoring, as well as ———. 2010b. Sea Turtles, Dolphins, and Whales and the Gulf of Mexico recovery efforts when needed. Oil Spill. NOAA Office of Protected Resources. (2 February 2011; www. Although understanding has improved of the medium- and nmfs.noaa.gov/pr/health/oilspill.htm) long-term impacts from oil pollution on animal and plant [NOM] Norma Oficial Mexicana. 2002. NOM-059-Ecol, Diario Oficial de la Federación Tomo DLXXXII 4: 1–80. physiologies, there is still much we do not know. Globally, Pauly D. 2002. Growth and mortality of the basking shark Cetorhinus maxi- countries must improve risk assessments of offshore oil and mus and their implications for management of whale sharks Rhincodon gas development, and seek to expand and regularize damage typus. Pages 199–208 in Fowler SL, Reed TM, Dipper FA, eds. Elasmo- and impact assessments, domestically and internationally. branch Biodiversity, Conservation and Management. Proceedings of These impacts must be systematically considered to establish the International Seminar and Workshop, July 1997, Sabah, Malaysia. IUCN. priorities for research and monitoring that will best ensure Rodrigues ASL, Pilgrim JD, Lamoreux JF, Hoffmann M, Brooks TM. 2006. effective species and system recovery. Although the research The value of the IUCN Red List for conservation. Trends in Ecology and agenda should focus on the United States’ immediate needs, Evolution 21: 71–76. its development should also support similar efforts in other Schipper JS, et al. 2008. The status of the world’s land and marine mammals: regions of the world in identifying species of priority concern. Diversity, threat, and knowledge. Science 322: 225–230. Short FT, et al. 2011. Extinction risk assessment of the world's seagrass The IUCN Red List is continually improved and revised under species. Biological Conservation. Forthcoming. strict standards and criteria, and its value in assessing the global Wilcove DS, Master LL. 2005. How many endangered species are there in the conservation status of biological diversity steadily expands. United States? Frontiers in Ecology and the Environment 3: 414–420.

396 BioScience • May 2011 / Vol. 61 No. 5 www.biosciencemag.org Roundtable Roundtable

Claudio Campagna ([email protected]) is with the Wildlife Conservation Museum of Natural History, in Washington, DC. Nicolas J. Pilcher is with Society in New York, New York. Frederick T. Short is with the Department the Marine Research Foundation in Sabah, Malaysia. Yvonne Sadovy de of Natural Resources and the Environment, University of New Hampshire, Mitcheson is with the School of Biological Sciences, University of Hong Kong, Jackson Estuarine Laboratory, in Durham. Beth A. Polidoro, Roger McManus, in China. Simon N. Stuart is with the IUCN Species Survival Commission, at and Kent E. Carpenter are with the Global Marine Species Assessment, the United Nations Environment Programme World Conservation Monitoring Marine Biodiversity Unit, International Union for Conservation of Nature Centre, in Cambridge, United Kingdom; the Department of Biology and Bio- (IUCN) Species Programme, Department of Biological Sciences, at Old chemistry, University of Bath, in the United Kingdom; the Al Ain Wildlife Park Dominion University, in Norfolk, Virginia. Roger McManus is also with the and Resort, in Abu Dhabi, United Arab Emirates; and Conservation Interna- Global Marine Species Assessment, IUCN Species Survival Commission, Perry tional, in Arlington, Virginia. Campagna, Short, Polidoro, Collette, Pilcher, Institute for Marine Science, in Jupiter, Florida. Bruce B. Collette is with the Sadovy, and Carpenter are also with the IUCN Species Survival Commission National Marine Fisheries Service Systematics Laboratory, at the National Marine Conservation Subcommittee in Gland, Switzerland.

www.biosciencemag.org May 2011 / Vol. 61 No. 5 • BioScience 397 LETTER

Underestimating the damage: interpreting cetacean carcass recoveries in the context of the Deepwater Horizon/BP incident Rob Williams1, Shane Gero2,LarsBejder3, John Calambokidis4, Scott D. Kraus5, David Lusseau6, Andrew J. Read7, & Jooke Robbins8

1Marine Mammal Research Unit, University of British Columbia, Vancouver, Canada 2Department of Biology, Dalhousie University, Halifax, Canada 3Centre for Fish and Fisheries Research, Cetacean Research Unit, Murdoch University, Western Australia 4Cascadia Research Collective, Olympia, WA, USA 5New England Aquarium, Boston, MA, USA 6School of Biology, Aberdeen University, Aberdeen, Scotland, UK 7Nicholas School of the Environment, Duke University, Beaufort, NC, USA 8Humpback Whale Studies Program, Provincetown Center for Coastal Studies, Provincetown, MA, USA

Keywords Abstract Anthropogenic impacts; dolphin; Deepwater Horizon; Gulf of Mexico; mortality; oil; Evaluating impacts of human activities on marine ecosystems is difficult when strandings. effects occur out of plain sight. Oil spill severity is often measured by the number of marine birds and mammals killed, but only a small fraction of carcasses Correspondence are recovered. The Deepwater Horizon/BP oil spill in the Gulf of Mexico was Rob Williams, Current address: Sea Mammal the largest in the U.S. history, but some reports implied modest environmental Research Unit, Scottish Oceans Institute, St Andrews Fife KY16 8LB. Tel: +44 (0)1334 impacts, in part because of a relatively low number (101) of observed ma- 462630; Fax: +44 (0)1334 463443. rine mammal mortalities. We estimate historical carcass-detection rates for 14 E-mail: [email protected] cetacean species in the northern Gulf of Mexico that have estimates of abundance, survival rates, and stranding records. This preliminary analysis suggests Received that carcasses are recovered, on an average, from only 2% (range: 0–6.2%) 23 September 2010 Accepted of cetacean deaths. Thus, the true death toll could be 50 times the number 15 February 2011 of carcasses recovered, given no additional information. We discuss caveats to this estimate, but present it as a counterpoint to illustrate the magnitude of Editor misrepresentation implicit in presenting observed carcass counts without simi- Leah Gerber lar qualification. We urge methodological development to develop appropriate multipliers. Analytical methods are required to account explicitly for low prob- doi: 10.1111/j.1755-263X.2011.00168.x ability of carcass recovery from cryptic mortality events (e.g., oil spills, ship strikes, bycatch in unmonitored fisheries and acoustic trauma).

National Oceanic and Atmospheric Association 2010). Introduction Not surprisingly, perhaps, this spill has been compared The Deepwater Horizon/BP oil spill in the Gulf of Mexico to other acute environmental disasters, such as the 1989 was not only the largest in the US history (Machlis & Exxon Valdez oil spill (EVOS). In the case of EVOS, the McNutt 2010) but was also the first to release oil at mortality of sea otters became emblematic of environ- the sea floor (over 1.5 km below sea level) and to in- mental impact, as well as a contentious effort to agree volve the widespread use of dispersants below the surface on compensation (Ehrenfeld 1990; Estes 1991). In con- (Mascarelli 2010). However, many media reports have trast, the Deepwater Horizon/BP event has not left such suggested that the spill caused only modest environ- an iconic symbol of devastation. As of November 7, mental impacts (Grunwald 2010; Walsh 2010), in part 2010, “only” 101 cetacean (whale, dolphin, and porpoise) because of a low number of observed wildlife mortal- carcasses1 had been detected across the Northern Gulf ities, especially marine mammals (Unified Area Com- mand from the U.S. Fish and Wildlife Service and the 1 http://www.nmfs.noaa.gov/pr/health/oilspill.htm

228 Conservation Letters 4 (2011) 228–233 Copyright and Photocopying: c 2011 Wiley Periodicals, Inc. R. Williams et al. Low probability of cetacean carcass recovery of Mexico. The critical issue is, therefore, how to in- viously leave no carcass at all. Shore recoveries may be terpret this relatively low number of carcass recoveries very site-specific, such that the likelihood of a carcass in terms of impact to populations. The Gulf of Mexico drifting to shore varies with the geography of the coast- is a semienclosed subtropical sea that forms essentially line itself (Faerber & Baird 2010). As such, “oiled” car- one ecosystem with many demographically independent casses detected subsequent to the Deepwater Horizon/BP cetacean populations (Mullin & Fulling 2004). Some of event are expected to represent a small fraction of total these cetacean populations, such as killer whales (Orci- mortality in the Northern Gulf of Mexico. nus orca), false killer whales (Pseudorca crassidens), melon- Given the magnitude of the spill and complexity of the headed whales (Peponocephala electra), and several beaked response, quantifying the ecological impacts will take a whale species, appear to be quite small, are poorly stud- long time. To contribute to this effort, we examined his- ied, or are found in the pelagic realm where they could torical data from the Northern Gulf of Mexico to evalu- have been exposed to oil and yet never strand. Small, ge- ate whether cetacean carcass counts in this region have netically isolated populations of bottlenose dolphins (Tur- previously been reliable indicators of mortality, and may siops truncatus)could have experienced substantial losses therefore accurately represent deaths caused by the Deep- either inshore or offshore. water Horizon/BP event. In an ideal world, one would simply compare post- spill to prespill abundance estimates. But, it is rare to Methods have good knowledge of long-term trends in wildlife abundance (Bonebrake et al. 2010). Abundance of many We estimated historical carcass-detection rates for 14 marine mammal populations has been monitored for species of cetaceans in the northern Gulf of Mexico for decades, but the low precision of most cetacean abun- which species-specific estimates of abundance (Waring dance estimates would prevent us from detecting all but et al. 2009a, b; Mullin & Fulling 2004) species-level adult- the most catastrophic declines using conventional null- survival rates (Taylor et al. 2007a), and stranding records hypothesis testing (Taylor et al. 2007b). As a result, it exist (Waring et al. 2009a, b). Estimates of mortality were would not be very informative to compare pre- and post- generated for each species by multiplying recent abun- spill abundance estimates for populations of cetaceans in dance estimates by the species-specific mortality rate. An the northern Gulf of Mexico. An alternative approach is annual carcass-recovery rate was then estimated by divid- to count the number of carcasses recovered, acknowl- ing the mean number of observed strandings each year by edging that these recoveries were subject to a number our estimate of annual mortality (Table 1). First, an over- of processes (e.g., sinking, decaying, scavenging, drifting) all pooled carcass-recovery rate was calculated across all that reduce detection probability, and then adjusting the cetacean species (n = 14) in the Gulf of Mexico for which counts upward to estimate total mortality. This is the ap- data was available by using the expected number of proach that is commonly taken to estimate the effects of deaths across all species and the total number of observed power lines on bird mortality, for example, in which it carcasses across all species. Next, species-specific carcass- has been shown in one instance that carcass counts un- recovery rates were calculated using only species-specific derestimate total mortality by 32% (Ponce et al. 2010). values and a mean (n = 14) of those was taken across This also appears to be the approach being taken to as- species. Species-specific carcass-recovery estimates were sess impacts of the Deepwater Horizon Incident on whales not generated for Bryde’s whale (Balaenoptera brydei), and dolphins, with the important caveat that the car- bottlenose dolphins (Tursiops sp.), and Fraser’s dolphins cass counts appear to be presented at face value, with (Lagenodelphis hosei) due to uncertainties in their abun- no attempt to extrapolate to total mortality (Unified Area dance and/or population structure (Waring et al. 2009b). Command from the U.S. Fish and Wildlife Service and No attempt was made to estimate carcass-recovery rates the National Oceanic and Atmospheric Association 2010; in the two taxonomic groups that are not identified to Grunwald 2010; Walsh 2010). species in the field during raw data collection: Kogia Cetacean carcasses do not necessarily strand along (a pooled estimate for two species, dwarf and pygmy coastlines or remain afloat long enough to be detected at sperm whales), or mesoplodonts (a pooled estimate for sea. The probability of detecting the death of a marine a genus of similar-looking beaked whales) (Waring et al. mammal depends on a wide range of physical and bi- 2009b). ological factors, including: behavioral responses prior to death, proximity of the carcass to shore (or at-sea ob- Results servers), decomposition rates and processes, water temperature, wind regime, and local currents (Epperly et al. Our analysis suggests that an average of 4,474 individual 1996). Cetaceans subject to natural predation would ob- cetaceans died annually between 2003 and 2007 from all

Conservation Letters 4 (2011) 228–233 Copyright and Photocopying: c 2011 Wiley Periodicals, Inc. 229 Low probability of cetacean carcass recovery R. Williams et al.

Table 1 Population parameters and illustrative species-speciﬁc carcass-recovery rates for 14 species from the Gulf of Mexico.

Northern Gulf of Population Adult- Estimated annual Mean observed Carcass-detection Mexico population estimatea CVa survival rateb mortalityc annual strandingsa rate (%)

Sperm whale 1665 0.20 0.986 23.3 0.8 3.4 Cuvier’s beaked whale 65 0.67 0.95 3.3 0.2 6.2 Atlantic spotted dolphind 37611 0.28 0.95 1880.6 2.4 0.13 Pantropical spotted dolphin 34067 0.18 0.95 1703.4 0.8 0.05 Striped dolphin 3325 0.48 0.95 166.3 0.8 0.48 Spinner dolphin 1989 0.48 0.95 99.5 1 1.0 Rough-toothed dolphind 2653 0.42 0.95 132.7 5.8 4.4 Clymene dolphin 6575 0.36 0.95 328.8 0.6 0.18 Killer whale 49 0.77 0.99 0.5 0 0 False killer whale 777 0.56 0.99 7.8 0 0 Pygmy killer whale 323 0.60 0.95 16.2 0.2 1.2 Melon-headed whale 2283 0.76 0.99 22.8 1.4 6.1 Risso’s dolphin 1589 0.27 0.95 79.5 2.8 3.5 Short-finned pilot whale 716 0.34 0.986 10.0 0.2 2.0 Average of all species 2.0 Pooled across all species (n = 14) 93,687 – – 4,474 17 0.4 aPopulation abundance and stranding data (2003–2007) were taken from (Waring et al. 2009b), unless otherwise noted. bPopulation-level estimates are preferable but generally unavailable, so data taken from (Taylor et al. 2007a). cCalculated as the abundance multiplied by mortality rate ( = 1–survival rate). dPopulation abundance and stranding data (2002–2006) were taken from (Waring et al. 2009a). natural and anthropogenic causes. However, during that long-term studies of killer whales off the coasts of British period, an average of only 17 cetacean carcasses were de- Columbia and Washington State, in which populations tected annually along the northern Gulf of Mexico. This are censused completely every year, carcasses from con- would suggest that the overall pooled rate of carcass re- firmed deaths of known individuals are recovered only covery for cetaceans in the Gulf of Mexico is approx- 6% of the time (Fisheries and Oceans Canada 2008). Sim- imately 0.4% of the total estimated mortality. Table 1 ilarly, low-detection rates have been estimated for car- breaks down the recovery rates by species. Carcasses were casses of eastern gray whales (Eschrichtius robustus, <5%, recovered only from a mean of 2.0% (range: 0–6.2%) Heyning & Dahlheim 1990), North Atlantic right whales of cetacean species deaths along the northern Gulf of (Eubalaena glacialis, 17%, Kraus et al. 2005), and har- Mexico. The disparity between this value and the over- bor porpoises (Phocoena phocoena, <1%, Moore & Read all pooled value likely results from undue influence of 2008), all of which occur in near-shore waters. Beached poorly studied and relatively rare species (e.g., Cuvier’s carcasses of other pelagic marine vertebrates have been beaked whale and melon-headed whale; Table 1) with shown to be equally poor indicators of mortality (for ex- high estimated carcass-recovery rates that are weighted ample, 7–13% recovery rates for four species of sea tur- equivalently and treated as reliably in this average as es- tle, Epperly et al. 1996). As such, raw carcass counts alone timates from species that are common and well studied. are not reliable indicators of the magnitude of mortality We have reason to believe that the Cuvier’s beaked whale for these species. recovery rate is positively biased. The original abundance We do not claim to have calculated definitive multi- estimate is thought to be an underestimate by a factor pliers for this spill. Instead, our aim is to show plausi- of 2 to 4, based on the assumption of certain track line ble ranges for those multipliers, in order to illustrate how detection (Mullin & Fulling 2004). Our carcass-recovery much they would affect our perception of the ecological rate for deep-diving whales would then be biased high by damages caused by Deepwater Horizon incident and why a factor of 2–4. this topic is worthy of additional resources for methodological development. Consider, for example, one sperm whale being detected as a carcass, and a necropsy identi- Discussion fied oiling as a contributing factor in the whale’s death. Our results indicate that carcass-recovery rates are his- If the carcass-detection rate for sperm whales is 3.4% torically low for cetaceans in the Gulf of Mexico. Stud- (Table 1), then it is plausible that 29 sperm whale deaths ies of other populations show similar recovery rates. In represents the best estimate of total mortality, given no

230 Conservation Letters 4 (2011) 228–233 Copyright and Photocopying: c 2011 Wiley Periodicals, Inc. R. Williams et al. Low probability of cetacean carcass recovery additional information. If, for example, 101 cetacean car- carcasses. Given that many cetaceans are highly social, casses were recovered overall, and all deaths were at- entire clusters, schools, pods, matrilines, or groups of an- tributed to oiling, the average-recovery rate (2%) would imals could have been affected (Williams et al. 2009). Al- translate to 5,050 carcasses, given the 101 carcasses de- though we used recent population estimates, it has yet tected (Table 1). As the necropsy results emerge, we can to be determined how many animals in each population evaluate whether this prediction is high or low, but the were actually exposed to the spill. Finally, the location of sheer scope for underestimation builds a compelling case, the spill and the subsequent response effort likely affected in our view, for additional work. The vast majority of the probability of detecting associated deaths. These are carcasses recovered appear to have been bottlenose dol- the factors that must be carefully considered as efforts to phins.1 As necropsy results emerge and the need for re- assess population impacts continue. We present our his- covery plans debated, we encourage such discussions to toric recovery rates as starting points for discussion, but explicitly take into account the probability that the num- caution that incorrect multipliers may result in estimated ber of dolphins stranded represented something on the mortalities exceeding the number of animals that were order of only 2% of the number of animals killed. The ever in the vicinity of the spill (Parrish & Boersma 1995). potential is high for the spill to have caused catastrophic Estimating the correct multipliers will require an interdis- impacts on small, localized populations of bottlenose dol- ciplinary research effort to combine oceanographic and phins in the Gulf. We note that coastal and offshore forms cetacean habitat modeling to assess exposure risk and of bottlenose dolphins are found off California, with the likely deaths caused by exposure. This research is needed, coastal carcass having a 50-fold greater probability of but currently lacking from research priorities emerging stranding than an offshore one (Perrin et al. 2010). from the oil spill mitigation and recovery efforts. Even in the case of EVOS, the large number of ob- The issue of carcass-detection rates is not merely of served deaths was acknowledged to represent only a frac- academic interest. Our results are directly relevant to as- tion of the total mortality (Estes 1991). Two approaches sessment of ecological damages caused by the Deepwa- were taken to estimate total mortality in Prince William ter Horizon/BP oil spill, but also have global relevance Sound: (1) a comparison of pre- and postspill popula- for litigation and marine conservation policy. Given that tion size; and (2) extrapolations from recovered carcasses environmental restitution in the United States can be to total mortality from a multiplier based on the prob- based on a violation system (Alexander 2010), carcass- ability of recovering a carcass (Garshelis 1997). Our es- recovery rates must be explicitly considered when evalu- timates of carcass-recovery rates were calculated from ating the impacts of such disasters. In the case of EVOS, the best available data, but we caution against using his- legal damages placed the value of each sea otter killed toric (i.e., pre-spill) carcass-recovery rates to generate a at US$80,000, or the cost of rehabilitating each oiled ot- simple multiplier to assess total mortality in the Deep- ter (Estes 1991; Garshelis 1997). In terms of broader rec- water Horizon/BP Incident. On the one hand, consider- ommendations for marine policy, we note that carcass able efforts were expended by government agencies and counts are used in many countries, including the United others to search for marine mammal carcasses after the States, to monitor human impacts on cetacean popula- spill, which could raise recovery rates above those esti- tions. The tools that managers use in the United States mated here. Fortunately, a comparison of pre- and post- to estimate and limit the impacts of human activities on spill search effort ought to be among the most tractable stocks relies upon “potential biological removal” (PBR), factors to account for when calculating carcass-recovery a calculation that determines how many animals can be rates. On the other hand, there are several arguments to removed from a stock before causing harm. The PBR esti- suggest that our carcass-recovery rates are biased high. mate, under the Marine Mammal Protection Act (MMPA) First, we estimated the number of carcasses using adult- depends on reasonably unbiased and precise estimates of survival rate; had we included juvenile and calf mor- human-caused mortality (Wade 1998). In contrast, the tality, the total number of carcasses would have been effects of many human impacts are only witnessed op- substantially higher and our estimated carcass-recovery portunistically, such as a carcass being discovered on a rate substantially lower. The point estimate is strongly in- beach. The issue arises when policymakers, legislators, or ﬂuenced by some optimistic values for Cuvier’s beaked biologists treat these carcass-recovery counts as though whale and melon-headed whale (Table 1). Abundance of they were complete counts or parameters estimated these elusive species is biased low, due to well-known from some representative sample, when in fact, they difﬁculties in estimating track line detection probability are opportunistic observations. Our study suggests that (g(0)) for deep-diving species. Some of these cetaceans these opportunistic observations should be taken to esti- represent prey species: our denominators include animals mate only the bare minimum number of human-caused that would have been preyed upon and not ended up as mortalities. This work suggests that carcass counts alone

Conservation Letters 4 (2011) 228–233 Copyright and Photocopying: c 2011 Wiley Periodicals, Inc. 231 Low probability of cetacean carcass recovery R. Williams et al. are unreliable indicators of either natural or anthro- Acknowledgments pogenic sources of mortality. It is vital to develop a framework that explicitly accounts for the low probabil- Lynne Barre, Dee Boersma, and Dave Thompson gave ity of recovering carcasses, if we are to accurately assess valuable comments at an early stage of the development the sustainability of all cryptic forms of human-caused of this manuscript. We thank Leah Gerber, Tim Ger- mortality. rodette and Barb Taylor for their careful reviews. Human impacts on marine ecosystems and marine mammals are growing both in type and scale (Kraus & References Rolland 2007; Clausen & York 2008; Duce et al. 2008; Doney 2010; Hoegh-Guldberg & Bruno 2010; Tittensor Alexander, K. (2010) The 2010 oil spill: criminal liability under et al. 2010). Establishing the proper spatial and temporal wildlife laws. Congressional Research Service Report to R41308 scales, at which to assess the impacts of acute events, is Congress, Washington D.C., ,p.10. Bejder, L., Samuels, A., Whitehead, H., Gales, N. (2006) further complicated by potential long-term effects and a Interpreting short-term behavioural responses to lack of basic population-specific information (Bejder et al. disturbance within a longitudinal perspective. Anim Behav 2006). This highlights the need for long-term population 72, 1149–1158. monitoring, such as that mandated by the U.S. MMPA Bonebrake, T.C., Christensen, J., Boggs, C.L., Ehrlich, P.R. (Bonebrake et al. 2010). In the first year after the 1989 (2010) Population decline assessment, historical baselines Exxon Valdez spill, the AT1 group of “transient” killer and conservation. Conserv Lett 3, 371–378. whales experienced a 41% loss; there has been no re- Clausen, R., York, R. (2008) Economic growth and marine production since the spill (Matkin et al. 2008). Although biodiversity: influence of human social structure on decline the cause of the apparent sterility is unknown, the lesson of marine tropic levels. Conserv Biol 22, 458–466. serves as an important reminder that immediate death Doney, S.C. (2010) The growing human footprint on is not the only factor that can lead to long-term loss coastal and open-ocean biochemistry. Science 328, of population viability. The recent disaster in the Gulf 1512–1516. of Mexico provides an important opportunity to assess Duce, R.A., LaRoche, J., Altieri, K., et al. (2008) Impacts of whether or not the intensity of monitoring conducted atmospheric anthropogenic nitrogen on the open ocean. in the Gulf of Mexico is sufficient to detect even catas- Science 320, 893–897. trophic effects (Taylor et al. 2007b). If line-transect sur- Ehrenfeld, D. (1990) The lessons of Valdez. Conserv Biol 4,1–2. vey data are found wanting, we see value in explor- Epperly, S.P., Braun, J., Chester, A.J., et al. (1996) Beach ing new passive acoustic monitoring methods to detect strandings as indicators of at-sea mortality of sea turtles. trends in relative abundance (Marques et al. 2009, Rojas- Bull Mar Sci 59, 289–297. Bracho et al. 2010). These could be especially useful for Estes, J.A. (1991) Catastrophes and conservation: lessons 254 rare or pelagic species, or those for which g(0) estima- from sea otters and the Exxon Valdez. Science , 1596. tion is particularly problematic. Accurate assessment of Faerber, M.M., Baird, R.W. (2010) Does the lack of observed impacts also must consider how species are likely im- beaked whale strandings in military exercise areas mean no impacts have occured? A comparison of stranding and pacted, whether acutely on contact with oil or over detection probabilities in the Canary and main Hawaiian the longer term through toxicity or habitat degradation islands. Mar Mamm Sci 26, 602–613. (Lovett 2010; Schrope 2010b). If support for longer term Fisheries and Oceans Canada. (2008) Recovery strategy for the assessments dwindles as the time passes and public at- northern and southern resident killer whales (Orcinus orca) in tention moves elsewhere (Schrope 2010a), then chronic Canada. Fisheries and Oceans Canada, Ottawa, Canada. effects will remain unknown. In such cases, only immedi- Garshelis, D.L. (1997) Sea otter mortality estimated from ately observable effects, such as the number of carcasses, carcasses collected after the Exxon Valdez oil spill. Conserv have and will be used to determine the impact of an Biol 11, 905–916. event, and synergistic and lagged effects will not be con- Grunwald, M. (2010) The BP spill: has the damage been sidered. Our findings suggest that assessments of the im- exaggerated? Time Magazine. Available from http://www. pact of anthropogenic events based solely on the numbers time.com/time/nation/article/0,8599,2007202,00.html. of carcasses recovered are deceptively biased. A better Accessed 4 October 2010. understanding of carcass-recovery rates and the degree Heyning, J.E., Dahlheim, M.E. (1990) Strandings and to which they underestimate actual mortality, is critical incidental takes of gray whales. Report to the International. to assessing the true consequences of oil spills and other Whaling Commission, SC/A90/G2, 16 pp. human activities known to cause cryptic mortality, such Hoegh-Guldberg, O., Bruno, J.F. (2010) The impact of climate as ship strikes, certain fisheries interactions, and acoustic change on the world’s marine ecosystems. Science 328, trauma. 1523–1528.

232 Conservation Letters 4 (2011) 228–233 Copyright and Photocopying: c 2011 Wiley Periodicals, Inc. R. Williams et al. Low probability of cetacean carcass recovery

Kraus, S.D., Brown, M.W., Caswell, H., et al. (2005) North U.S. Department of Commerce, NOAA Technical Atlantic right whale in crisis. Science 309, 561–562. Memorandum NMFS, NOAA-TM-NMFS-SWFSC-459. 39 Kraus, S.D., Rolland, R.M. (2007) The urban whale pages. syndrome. Pages 488–513 in S.D. Kraus, R.M. Rolland, Schrope, M. (2010a) The lost legacy of the last great oil spill. editors. The urban whale: North Atlantic right whales at the Nature 466, 304–305. crossroads. Harvard University Press, Cambridge, Schrope, M. (2010b) Oil spill cruise finds a field of dead coral. Massachusetts. Nature News Online. Available from: http://www.nature. Lovett, R.A. (2010) Oil spills toxic trade-off. Nature News com/news/2010/101105/full/news.2010.589.html Online. Available from: http://www.nature.com/ Accessed November 11 2010. news/2010/101110/full/news.2010.597.html. Accessed 11 Taylor, B.L., Chivers, S.J., Larese, J., Perrin, W.F. (2007) November 2010. Generation length and percent mature estimates for IUCN Machlis, G.E., McNutt, M.K. (2010) Scenario-building for the assessments of cetaceans. Administrative Report LJ-07-01, deepwater horizon oil spill. Science 329, 1018–1019. Southwest Fisheries Science Center, 8604 La Jolla Shores Marques, T.A., Thomas, L., Ward, J., DiMarzio, N., Tyack, P.L. Blvd., La Jolla, CA 92038, USA. 24 pp. (2009) Estimating cetacean population density using fixed Taylor, B.L., Martinez, M., Gerrodette, T., Barlow, J., Hrovat, passive acoustic sensors: an example with Blainville’s Y.N. (2007b) Lessons from monitoring trends in abundance beaked whales. JAcoustSocAm125, 1982–1994. of marine mammals. Mar Mamm Sci 23, 157–175. Mascarelli, A. (2010) Deepwater horizon: after the oil. Nature Tittensor, D.P., Mora, C., Jetz, W., et al. (2010) Global 467, 22–24. patterns and predictors of marine biodiversity across taxa. Matkin, C.O., Saulitis, E.L., Ellis, G.M., Olesiuk, P., Rice, S.D. Nature 466, 1098–1101. (2008) Ongoing population-level impacts on killer Unified Area Command from the U.S. Fish and Wildlife whales Orcinus orca following the ‘Exxon Valdez’ oil spill Service and the National Oceanic and Atmospheric in Prince William Sound, Alaska. Mar Ecol Prog Ser 356, Association. (2010) Deepwater/Horizon response 269–281. consolidated fish and wildlife collection report. Available Moore, J.E., Read, A.J. (2008) A bayesian uncertainty from: http://www.restorethegulf.gov/sites/default/files/ analysis of cetacean demography and bycatch mortality Consolidated%20Wildlife%20Table%20100410.pdf. using age-at-death data. Ecol Appl 18, 1914–1931. Accessed 5 October 2010. Mullin, K.D., Fulling, G.L. (2004) Abundance of cetaceans in Wade, P.R. (1998) Calculating limits to the allowable the oceanic northern Gulf of Mexico, 1996–2001. Mar human-caused mortality of cetaceans and pinnipeds. Mar Mamm Sci 20, 787–807. Mamm Sci 14, 1–37. Parrish, J.K., Boersma, P.D. (1995) Muddy waters. Amer Sci Walsh, B. (2010) Oil spill: the well is dead. Time Magazine. 83, 112–115. Available from: http://ecocentric.blogs.time.com/2010/ Perrin, W.F., Thieleking, J.L., Walker, W.A., Archer, F.I., 09/19/oil-spill-the-well-is-dead/. Accessed 4 October 2010. Robertson, K.M. (2010) Common bottlenose dolphins Waring, G.T., Josephson, E., Fairfield-Walsh, C.P., (Tursiops truncatus) in California waters: Cranial Maze-Foley, K. (2009a) U.S. Atlantic and Gulf of Mexico differentiation of coastal and offshore ecotypes. Mar Mamm marine mammal stock assessments – 2008. NOAA Tech Sci, doi: 10.1111/j.1748-7692.2010.00442.x. Memo NMFS NE 210, p. 440. Ponce, C., Alonso, J.C., Argandona, G., Garcia Fernandez, A., Waring, G.T., Josephson, E., Maze-Foley, K., Rosel, P.E. Carrasco, M. (2010) Carcass removal by scavengers and (2009b) U.S. Atlantic and Gulf of Mexico marine mammal search accuracy affect bird mortality estimates at power stock assessments – 2009. NOAA Tech Memo NMFS NE lines. Anim Conserv 13, 603–612. 213, p. 528. Rojas-Bracho, L., Jaramillo-Legoretta, A., Cardenas, G., et al. Williams, R., Lusseau, D., Hammond, P. (2009) The role of (2010) Assessing trends in abundance for vaquita using social aggregations and protected areas in killer whale acoustic monitoring: within refuge plan and outside refuge conservation: the mixed blessing of critical habitat. Biol research needs. Workshop Report—October 19–23, 2009. Conserv 142, 709–719.

Article

pubs.acs.org/est

1 Multitissue Molecular, Genomic, and Developmental Eﬀects of the

2 Deepwater Horizon Oil Spill on Resident Gulf Killiﬁsh (Fundulus

3 grandis) ,† ‡ ‡ § † 4 Benjamin Dubansky,* Andrew Whitehead, Jeﬀrey Miller, Charles D. Rice, and Fernando Galvez

† 5 Louisiana State University, Department of Biological Sciences, 208 Life Sciences Building, Baton Rouge, Louisiana 70803, United 6 States ‡ 7 University of California, Davis, Department of Environmental Toxicology, 4121 Meyer Hall, Davis, California 95616, United States § 8 Clemson University, Department of Biological Sciences, 132 Long Hall, Clemson, South Carolina 26634, United States

9 *S Supporting Information

10 ABSTRACT: The Deepwater Horizon oil rig disaster resulted 11 in crude oil contamination along the Gulf coast in sensitive 12 estuaries. Toxicity from exposure to crude oil can affect 13 populations of fish that live or breed in oiled habitats as seen 14 following the Exxon Valdez oil spill. In an ongoing study of the 15 effects of Deepwater Horizon crude oil on fish, Gulf killifish 16 (Fundulus grandis) were collected from an oiled site (Grande 17 Terre, LA) and two reference locations (coastal MS and AL) 18 and monitored for measures of exposure to crude oil. Killifish 19 collected from Grande Terre had divergent gene expression in 20 the liver and gill tissue coincident with the arrival of 21 contaminating oil and up-regulation of cytochrome P4501A 22 (CYP1A) protein in gill, liver, intestine, and head kidney for 23 over one year following peak landfall of oil (August 2011) 24 compared to fish collected from reference sites. Furthermore, laboratory exposures of Gulf killifish embryos to field-collected 25 sediments from Grande Terre and Barataria Bay, LA, also resulted in increased CYP1A and developmental abnormalities when 26 exposed to sediments collected from oiled sites compared to exposure to sediments collected from a reference site. These data 27 are predictive of population-level impacts in fish exposed to sediments from oiled locations along the Gulf of Mexico coast.

5 28 ■ INTRODUCTION fish populations. However, sublethal effects of PAH exposure 50 such as impairment of corticosteroid secretion, immune 51 29 As a result of the explosion of the Deepwater Horizon oil dysfunction, gill damage, impaired growth and reproduction, 52 30 platform, as much as 700 million liters of crude oil were 1 and reduced cardio-respiratory capacity are capable of reducing 53 31 released from the Macondo well. Despite cleanup efforts, this an animal’s ability to effectively respond to environmental 54 32 oil was widely distributed along shorelines of Louisiana and, to 4,6−14 stressors. Furthermore, PAHs cause developmental 55 33 a lesser extent, Mississippi, Alabama, and Florida. During the abnormalities in larval fish such as craniofacial defects, edema, 56 34 initial response and cleanup effort, visible oil was reported on reduced size, and cardiovascular defects result in a decrease in 57 35 the water surface, along beaches, and in marshes as early as May fitness that can persist into adulthood, affecting the productivity 58 36 2010, which coincided with the spawning season for many 4,5,15−17 37 marine and estuarine fish species. By August 2010, much of the and survivorship of a population. Multiple studies 59 38 surface oil slick had dissipated or was removed, and visible oil following the Exxon Valdez oil spill (EVOS) linking crude oil to 60 ff fi 39 on the beaches and marsh became far less noticeable. However biological e ects in sh were reported, providing benchmarks 61 ff 40 to this date, a considerable amount of weathered oil likely for the evaluation of remediation e orts and recovery status of 62 ff 5 fi ff 41 remains deposited in the sediment, serving as a reservoir for a ected areas. Examination of adult sh for biological e ects 63 2−4 42 persistent exposure to resident species. Since crude oil and molecular markers of exposure to oil, along with 64 fi 43 contains chemicals that are toxic to fish, such as polycyclic developmental end points in embryos and larval sh both in 65 fi 44 aromatic hydrocarbons (PAHs), it is probable that residual oil oiled eld sites and in laboratory-based exposures, was 66 45 from the Deepwater Horizon oil spill (DHOS) will affect 46 populations of fish living and breeding in contaminated Received: January 29, 2013 47 locations. Revised: April 18, 2013 48 Fish deaths from toxic acute exposure to PAHs certainly Accepted: April 22, 2013 49 occur and are often used as a measure of the effects of oil on

4,5 19 67 predictive of population-level impacts in fish populations. Bioinformatics Resources, and gene interaction network 129 68 Few reports have emerged to date using a resident species analysis was performed using Ingenuity Pathway Analysis 130 69 either in field or laboratory-based bioassays of exposure and software (Ingenuity Systems, Inc.). 131 70 effect in areas known to be affected by the DHOS. Immunohistochemistry. All tissues were processed as 132 3 71 This study follows the first report to describe the biological described in Whitehead et al., 2012, for immunohistochemical 133 72 effects in populations of resident killifish exposed in situ to visualization of protein abundance and distribution of 134 3 73 crude oil from the DHOS. Here, we present evidence of cytochrome P4501A (CYP1A), a marker of exposure to AhR- 135 12,20,21 74 exposure to xenobiotics in multiple tissues of adult fish, effects active PAHs, using monoclonal antibody (mAB) C10−7. 136 75 that persist for over one year following the landfall of DHOS Tissues adhered to poly L-lysine coated slides were probed with 137 76 oil. We also show developmental defects in embryos exposed to mAb C10−7 using Vectastain ABC immunoperoxidase system 138 77 sediments collected from oil-impacted sites in coastal Louisiana (Vector Laboratories). CYP1A was visualized with NovaRed 139 78 in 2010 and 2011. Collectively, the long-term exposure of these (Vector Laboratories) and counter-stained with Hematoxylin 140 79 fish to persistent PAHs bound in the sediment and the QS (Vector Laboratories). Gill tissues from T1, T2, and T3 141 3 80 reduction in fitness of developing embryos due to early life were previously used in Whitehead et al., 2012. Slides from 142 81 exposure to these sediments could be predictive of persistent these previously used tissues were produced and imaged 143 4 82 population impacts as seen after the EVOS. alongside newly processed liver, intestine, and head kidney 144 tissues from the same fish. 145 83 ■ MATERIALS AND METHODS Sediment Collection and Embryo Exposures. Sediment 146 84 Sampling of Adult Fish. Adult Gulf killifish, an abundant, was collected from GT on June 16, 2010, coincident with peak 147 85 nonmigratory baitfish, were collected using wire minnow traps oiling, and approximately one year later on August 28, 2011, 148 86 during four trips between May 2010 and August 2011, a time when much of the visible oil had dissipated (Table S2, 149 87 frame that would have coincided with the peak spawning period Supporting Information). A second location was sampled on 150 3 88 of this species in the field (Table S1, Supporting Information). August 16, 2011, on a small island in South Wilkinson Bay, LA 151 89 The first three trips in 2010, as described in Whitehead et al., (WB), where visible oiling was reported from July 2010 to 152 3 90 2012, consisted of five sites identified as unoiled reference sites December 2012 by the National Oceanographic and Atmos- 153 91 and a sixth site (Grand Terre Island, LA) directly impacted by pheric Administration (NOAA) during the Shoreline Cleanup 154 92 contaminating oil. Here, we utilize remaining tissues from that and Assessment Technique (SCAT) surveys (www.gomex. 155 93 study collected at two of the five reference sites, Bay St. Louis, erma.noaa.gov). A third location was chosen in the north of Bay 156 94 MS (BSL) and Bayou La Batre, AL (BLB), and the oiled Sansbois (NBS) on August 3, 2011, where no oil was reported 157 95 location, Grande Terre Island, LA (GT) on trip one (T1) prior by SCAT surveys and this sample served as our nonoiled 158 96 to oiling, on trip two (T2) at the peak of oiling, and on trip control sediment (see below). 159 97 three (T3) after peak oiling. Additional tissues were collected Sediments were collected using a stainless steel sediment 160 98 from GT during a fourth trip (T4) in August 2011, one year sampler (WILDCO). Approximately 10 cm sediment cores 161 99 after T3. were removed along the shoreline every 0.5 m, and combined 162 100 Transcriptomics. Gill tissues were excised immediately and mixed in an aluminum container, then aliquoted into 950 163 fl 101 from five field-captured male fish of reproductive age (>6 cm), mL amber glass bottles with polytetra uoroethylene (PTFE) 164 102 preserved in RNA-later (Ambion, Inc.), and stored at −20 °C lined lids. Samples were stored on ice for transport to Louisiana 165 103 until nucleic acid extraction. Microarray data collection was as State University. Sediments collected in 2010 were stored at 166 3 104 reported in Whitehead et al., 2012. Briefly, total RNA was −20 °C until collection of 2011 sediments, when they were 167 105 extracted using Trizol reagent (Invitrogen; Life Technologies stored with the later at 4 °C until use. 168 106 Corp.); antisense RNA (aRNA) was prepared (Ambion Amino Sediment Analytical Chemistry. Analytical chemistry of 169 107 Allyl Messageamp II aRNA amplification kit), then coupled sediments used in embryo exposures was conducted as reported 170 3 108 with Alexa Fluor dyes (Alexa Fluor 555 and 647; Molecular in Whitehead et al., 2012. Briefly, sediments were solvent- 171 109 Probes), and hybridized to custom oligonucleotide microarray extracted and extracts analyzed by gas chromatography 172 110 slides (Agilent eArray design ID 027999). The microarray was interfaced with a mass spectrometer. Spectral data were 173 111 designed from F. heteroclitus expressed sequences and included analyzed by Chemstation Software (Agilent Technologies, 174 112 probes for 6800 target sequences. This same microarray Inc.). 175 113 platform was previously used in studies of PCB exposures in Embryo Collection. Male and female Gulf killifish were 176 18 114 F. heteroclitus, in a field study of the liver response to the collected from Cocodrie, Louisiana (29.254175°, 177 3 115 DHOS in F. grandis and a laboratory study of gill and liver −90.663559°) and held in a 1500 L recirculating tank 178 116 responses to weathered Louisiana crude oil in F. grandis (A.W., containing 10 ppt artificially formulated seawater (AFS) 179 117 unpublished data). (Instant Ocean). Embryos were collected on Spawntex 180 118 Raw data were sequentially normalized by lowess, mixed spawning mats (Aquatic Ecosystems), which were placed in 181 119 model analysis, and quantile methods in JMP Genomics (SAS, the tanks at sundown, and removed one hour after sunrise. 182 120 Inc.), then log2 transformed. Statistical analysis was performed Eggs were removed and held in a 950 mL Pyrex dish until use. 183 121 using mixed models in JMP Genomics, where main effects were Sediment Exposures. Laboratory mesocosms were pre- 184 122 specified as “time” (including our four sampling time-points) pared by first saturating sediments with 10 ppt AFS prior to 185 123 and “site” (including our three field sites), including an adding 150 ± 5 mL of this mixture to graduated 950 mL Pyrex 186 124 interaction term. Five biological replicates were included within dishes. This mixture was then overlaid with 75 mL AFS. 187 125 each treatment. A transcriptional response that was different Suspended sediment was allowed to settle for 1 h prior to 188 126 between sites throughout the oiling event was identified by insertion of sampling baskets containing embryos. 189 127 statistically significant (p < 0.01) time-by-site interaction. Gene Sampling baskets were constructed using virgin polytetra- 190 128 ontology enrichment analysis was performed using David fluoroethylene (PTFE) pipe stock (4B Plastics, Baton Rouge, 191

B dx.doi.org/10.1021/es400458p | Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article

192 LA) and PTFE mesh with 250 μm openings (Macmaster-Carr). for 94% of the genes that showed site-dependent expression, 254 193 Two interlocking rings were machined (4B Plastics, Baton whereas, for only 6% of these genes, the outlier was one of the 255 194 Rouge, LA) and fitted together to clamp PTFE mesh tightly reference sites. The departure in expression at GT from other 256 195 across the rings to create a filter basket that rested on the field sites occurred primarily at the second sampling time-point 257 196 sediment for the duration of the exposure and elevated the (Figure 1A and C). This indicates that divergence in genome 258 f1 197 embryos approximately 5 mm above the sediment−water expression is tightly associated with the timing and location of 259 198 interface. major oiling events in the field. This is consistent with the 260 199 At the start of experiments, 20−54 embryos were randomly divergence in genome expression in the liver, which was also 261 200 distributed to each PTFE basket, which was positioned on top tightly coupled with the timing and location of oil 262 3 201 of sediment within a Pyrex dish. Four mesocosms were created contamination (Figure 1 and Whitehead et al., 2012 ). 263 202 for GT sediments from 2010 (N = 146), five from GT Gene ontology categories that were significantly enriched 264 203 sediments from 2011 (N=132), three from WB sediments within the set of oiling-associated genes in gill include “response 265 204 collected in 2011 (N = 60), and five from NBS sediments to wounding”, “inflammatory response”, “acute phase response”, 266 205 collected in 2011 (N = 132). Mesocosms were placed on an and “cytochrome P450”. PCBs are mechanistically related to the 267 206 orbital shaker at 29 rpm to simulate tidal and wind movement toxic components of oil (PAHs), insofar as toxicity is largely 268 207 of water, without causing an observable increase in turbidity. mediated through the aryl hydrocarbon receptor (AhR) 269 22 208 Animals were kept at room temperature (20−22 °C) on a signaling pathway. Genes that are PCB dose-responsive in 270 18 209 natural light cycle. Hatching of Gulf killifish embryos typically killifish are significantly enriched within this set of oil- 271 210 occurs between days 10−14 postfertilization at this temper- associated genes in the gill (p < 0.01, Fisher’s exact test). Up- 272 211 ature, so embryos were observed daily for mortality and regulation of a canonical set of genes is diagnostic of activation 273 212 hatching for 21 days prior to termination of an experiment. of the AhR pathway. Gene targets of the activated AhR pathway 274 213 Every other day, 25 mL of water was removed from the are among the genes that have oil-associated expression, 275 214 mesocosms and replaced with fresh 10 ppt AFS. Preliminary including CYP1A1, CYP1B1, GCHFR, CYB5, and NUPR1. 276 215 experiments using reference sediments and this high water to Among the top five biological functions implicated by network 277 216 biomass ratio showed no increase in dissolved oxygen, and pathway analyses for gill and gill-only genes (Figure 1D, 278 217 ammonia, or pH when water was replaced every second day. pink + green genes) were “dermatological diseases”, “immuno- 279 fi fl 218 On day eight, the lter baskets were brie y removed from the logical disease”, and “cancer”. Acute phase response signaling 280 219 mesocosms and placed under a stereomicroscope, where the was also implicated (p < 0.0001) among the top canonical 281 220 heartbeats of eight embryos per treatment were measured. Each pathways for gill tissue-specific genes (Figure 1D, pink genes). 282 221 day, newly hatched larvae were preserved in either Z-Fix Inflammatory signaling is becoming increasingly recognized as 283 ff 222 bu ered zinc formalin (Ameresco) for histological processing as an important mechanism mediating the toxic effects of AhR 284 23 223 above or in RNAlater (Ambion, Inc.) for future genomics work. agonists. The molecular mechanism of AhR ligand-activated 285 23 224 Additional exposures using NBS 2011, WB 2011, and GT 2010 inflammation is cell-type specific, where the inflammatory 286 225 sediments were conducted under identical conditions to response is facilitated by cytokines (e.g., TNF) and chemo- 287 226 measure larval length at hatch. Our supply of GT sediment kines, both of which are implicated in the gill-specific gene 288 227 from 2011 was depleted at the time of these exposures; thus, cluster (Figure 1D). Such AhR activation can mediate immune 289 24 228 GT 2011 sediment was not included in this test. Larval length modulation, which may increase health risks for animals 290 229 at hatch was measured using Zeiss AxioVision 4.8 software on a encountering persistent pathogen challenge in the wild. These 291 230 Zeiss Lumar stereomicroscope. Embryonic heartbeat, mortality, patterns of expression provide clear evidence that killifish were 292 231 hatching success, and larval length were evaluated using exposed to the toxic components of oil, and that gill is a 293 232 XLSTAT version 2012.6.08 and were compared using the sensitive target of such exposure. 294 − ’ 233 Kruskal Wallis test, followed by Dunn s pairwise comparison. Patterns of liver gene expression (from Whitehead et al., 295 3 234 Cumulative daily hatching success was analyzed using 2012 ) were compared with patterns in gill to uncoverexpres- 296 235 generalized linear mixed models (GLMM) with the GLIMMIX sion responses that were unique between tissues and common 297 236 procedure (SAS, Inc.) to compare daily hatching between between tissues. More genes were diagnostic of the oiling event 298 237 treatment and day. in the liver than in the gill (434 versus 248, respectively (Figure 299 238 Imaging. Microscope slides were observed and imaged on a 1A and B)). However, the degree of up- or down-regulation of 300 239 Nikon Eclipse 80i compound microscope using a Nikon DS-Fi1 genes in response to the oiling event was more dramatic in gill 301 240 camera and NIS-Elements BR 3.10 software. Images were than in liver tissue (Figure 1C). Most of the genes that showed 302 241 balanced globally in Photoshop CS3 (Adobe) for levels using statistically significant divergence in expression only in the liver 303 242 the curves tool for white balance or the levels function. showed the same trend in gill (Figure 1A, bottom panel), 304 though the reciprocal pattern (correlation between liver and gill 305 243 ■ RESULTS AND DISCUSSION fi patterns for genes with signi cant response only in the gill) was 306 244 Genome Expression Analysis. In gills, 374 genes were not apparent (Figure 1A, top panel). This more dramatic 307 245 divergently expressed between field sites throughout the time- transcriptional response in the gill may be reflective of this 308 246 course of the oiling event (significant time-by-site interaction; p organ’s direct contact with the contaminated external environ- 309 247 < 0.01) (gene expression data are archived under the EBI ment. In contrast, the liver is not in direct contact with the 310 248 Array-Express accession number E-MTAB-1622; see the environment, and its attenuated response relative to the gill 311 249 Supporting Information Microarray Excel file for results of may be reflective of mechanisms of chemical uptake at epithelia 312 250 statistical analyses for each gene). Of these, the vast majority and the complex internal dynamics metabolism. 313 251 (94%) were different in their expression between the oil- Immunohistochemistry of Adult Fish Sampled in Situ. 314 252 exposed site (GT) relative to the two reference sites (BSL and Immunolocalization of CYP1A in gill, liver, intestine, and head 315 253 BLB). That is, the response at the oiled GT site was the outlier kidneys from fish collected in situ from three locations (GT, 316

C dx.doi.org/10.1021/es400458p | Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article

exposure to crude oil at GT. GT fish collected after the landfall 320 of oil (T2−T4) had elevated CYP1A protein within the 321 interlamellar region of the filament and in the pillar cells of the 322 lamellae compared to fish gills collected prior to oiling (T1), or 323 fish collected from the reference sites (Figure 2A and 324 f2

Figure 2. Distribution of CYP1A protein (burgundy staining) in gills, liver, head kidneys, and intestines from adult killifish collected from Figure 1. Transcriptomics. Patterns of expression for the genes that Bay St. Louis (BSL), Bayou La Batre (BLB), and Grande Terre (GT) showed site-dependent expression throughout the oiling event (genes at four time points (T1−T4) beginning prior to landfall of oil in May with significant time-by-site interaction). (A) Clusters of genes with fi fi 2010 (T1), at the peak of oiling in June 2010 (T2), after peak oiling in gill-speci c (top cluster), liver-speci c (bottom cluster), and common August 2010 (T3), and one year after landfall of oil in August 2011 at (middle cluster) expression response between tissues, where columns GT only (T4). (A) Increase in CYP1A in the gill lamellae (chevrons) are mean expression for a treatment, rows are genes, and color of cells and in the epithelia of the interlamellar regions of the gill filament indicate fold up-regulation (yellow) or down-regulation (blue) relative (star) as well as an increase in hyperplasia in the gill lamellae and in to preoil controls per site. Columns are organized by consecutive time- the interlamellar regions of the gill filaments. (B) Livers from GT points within sites, where sites are Grand Terre (GT), Bay St. Louis showed the highest expression of CYP1A at T3 and T4 time points, (BSL), and Bayou La Batre (BLB). Three consecutive time-points are and an increase in CYP1A was observed in BSL fish at T3. (C) preoil, peak oil, and postoil (left to right) in 2010. Gills were sampled Increased staining of epithelial cells of the kidney tubules (arrow at a fourth time-point at the GT site one year later (August 2011). (B) heads) and an increase in CYP1A-positive vascular endothelial cells Venn diagram indicating number of oil-associated genes expressed per were found in GT fish. (D) GT intestinal tissues show an increase in tissue. (C) Plots of principal component 1 (PC1) from principal CYP1A in the epithelial cells and CYP1A-positive vascular endothelial components analysis of the trajectory of transcriptome change through cells (arrow heads) in the lamina propria and submucosa. All tissues time and between sites for liver (open symbols) and gill (closed sectioned at 4 μm thickness and imaged with a 20× objective. Arrows symbols). Red, green, and blue represent the trajectories of time- fi fi = vascular endothelial cells, chevrons = gill laments, asterisks = course response (base through head of arrows represent rst through buccopharyngeal cavity, arrow heads = kidney tubules. Scale bar = 50 last sampling times) at sites GT, BSL, and BLB, respectively. Top, μm. All slides were counterstained with hematoxylin (blue). Gill middle, and bottom panels represent genes that were gill-specific, fi tissues from T1, T2, and T3 were previously used to create images common to both tissues, and liver-speci c, respectively, mirroring the published in Whitehead et al., 2012,3 although the gill images clusters represented in the heatmap. In brackets is the proportion of presented here (T1, T2, and T3) are new images from archived tissues variation accounted for by PC1. (D) Interaction networks for genes fi fi that were processed alongside gills from T4, and head kidney, that were gill-speci c (pink), liver-speci c (blue), or common between intestine, and liver tissues depicted in this figure. tissues (green) in their transcriptional response. Bold symbols represent genes included in the analysis, whereas other genes form connections with one degree of separation. The common (green) set 3 Whitehead et al., 2012 ). Increased hyperplasia along the 325 of genes are separated into three clusters, where the middle cluster is fi equally connected to the gill and liver networks, but the left and right lamental and lamellar epithelia of the gill was also present in 326 clusters primarily associate with the liver or gill clusters, respectively. GT fish postoil compared to fish from GT preoil and reference 327 sites. Liver tissues from GT fish collected during T3 and T4 328 also had elevated CYP1A protein, in contrast to fish collected 329 317 BSL, BLB) during peak oiling, one month after peak oiling, and during T2, when CYP1A was less abundant (Figure 2B) 330 318 one year after peak oiling (Table S1, Supporting Information) compared to fish collected from reference sites. Distribution 331 319 reveals a pattern of CYP1A protein expression indicative of and abundance of CYP1A protein in the head kidney (Figure 332

D dx.doi.org/10.1021/es400458p | Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article

Figure 3. Hatching success and mortality of embryos exposed to sediments collected from an unoiled reference location in North Bay Sansbois in 2011 (NBS 2011) and from oiled locations in Wilkinson Bay in 2011 (WB 2011) and Grande Terre Island (GT 2011 and 2010). Embryos were fertilized and observed daily for 21 days for mortality and hatching. (A) Percentage of hatched and unhatched larvae and percentage of mortalities. Unhatched embryos are defined as those that did not hatch by day 21 postfertilization. Mortalities are animals that died within the 21 day period. Asterisks indicate significant difference compared to the NBS reference sediment exposure (P ≤ 0.001). (B) After day 13, embryos exposed to GT 2010 and GT 2011 sediments hatched significantly less compared to embryos exposed to WB or NBS sediments. Error bars indicate standard error and are one sided to prevent overlapping and crowding of symbols. Asterisks indicate significant difference compared to the NBS reference sediment exposure (P ≤ 0.05).

333 2C) and intestine (Figure 2D) also suggest exposure to PAHs CYP1A in the gill can indicate water-borne exposure to 374 334 in crude oil in GT fish collected after the arrival of oil (T2− PAHs, whereas CYP1A increase in the intestine can indicate 375 335 T4), when compared to fish from reference sites. CYP1A dietary uptake and metabolism of PAHs as they cross the 376 2,25,28 336 expression, as found pervasively in the vascular endothelial cells intestinal epithelium. Additionally, fish in hyperosmotic 377 337 of intestine and head kidney of GT fish, is a hallmark of AhR environments drink to absorb water across the posterior 378 2,20,25 338 pathway activation. Head kidney tissues also had increased intestine leading to the accumulation of PAHs via the 379 2 339 CYP1A protein localized in the tubular epithelial cells, while gastrointestinal tract. CYP1A expression in the head kidney 380 340 intestinal epithelial cells were heavily stained in all GT fish is indicative of exposure of a central part of the teleost immune 381 341 collected after the arrival of oil, in contrast to the near absence system to immunotoxic PAHs and also suggests systemic 382 31,32 342 of staining in head kidney and intestine from fish collected from circulation of PAHs. The up-regulation and distribution of 383 343 BSL and BLB. CYP1A expression was detected in some CYP1A protein throughout the gill, intestine, head kidney, and 384 344 reference fish tissues, consistent with the endogenous CYP1A liver seen in GT fish, in contrast to that seen in fish collected 385 345 protein expression in most tissues associated with diverse from reference sites, is consistent with exposure to PAHs in 386 20,26 346 biological functions. The AhR can be activated by multiple contaminating oil, and with persistence of PAHs in sediments 387 2,5 347 anthropogenic and natural environmental stressors and enabling long-term exposure to resident biota. 388 348 endogenous AhR ligands to induce the expression of CYP1A. Developmental Impacts of Field-Collected Sediments. 389 349 However, CYP1A expression is far greater when the AhR is Total PAH (tPAH) and alkane content in GT sediments from 390 350 activated by xenobiotic ligands and the relative increase in 2010 and 2011, and WB sediments from 2011 were elevated 391 351 CYP1A protein distributed throughout multiple tissues in GT compared to NBS sediments (Tables S3−S4 and Figures S1− 392 352 fish is consistent with exposure to PAHs coincident with the S3, Supporting Information). SCAT data confirms visible oil at 393 23 353 arrival of contaminating oil (Figure 2). GT and WB throughout December 2012, while no oil was 394 354 The persistence of increased CYP1A protein expression in reported at NBS. Based on analytical chemistry data, GT 395 355 GT fish more than one year after initial oiling is not surprising sediments collected during 2010 had almost a 10-fold higher 396 356 since crude oil can remain bound in sediments, facilitating the tPAH concentration compared to sediments collected at that 397 357 slow and long-term release of PAHs from sediments and from site a year later. WB sediments collected in 2011 contained 398 2,5 358 dietary accumulation frombenthicfoodsources. By approximately 21% less tPAHs than do GT 2011 sediments 399 359 examining sensitive biological responses in resident species collected at the same time, whereas NBS had negligible tPAH 400 360 with high home-range fidelity (such as the Gulf killifish), it is content that was less than 1% of the tPAHs found in WB 401 361 possible to determine the distribution and persistence of sediments. 402 362 exposures to the toxic components of oil across space and time. Throughout the 21-day exposure of embryos to these 403 363 Typically, responses of liver are used for estimating exposure sediments, many of the GT sediment-exposed embryos failed 404 364 to PAHs due to its capacity for xenobiotic transformation of to hatch (Figure 3). Beginning at day 13, embryos exposed to 405 f3 365 blood-borne toxicants, although the gill, intestine, and head GT 2010 and GT 2011 sediments had significantly fewer 406 366 kidney are also sensitive indicators of exposure to AhR-inducing hatching events compared to embryos exposed to WB or NBS 407 367 chemicals, and have become increasingly popular for their sediments (P < 0.05). Accordingly, percent hatch within 21 408 20,26−28 368 utility in determining exposure to PAHs. However, days postfertilization for embryos exposed to GT sediments 409 369 assessing multiple tissues for CYP1A distribution can provide collected in 2010 and 2011 was significantly reduced compared 410 370 insights into the route of exposure based on the differential to those exposed to sediments collected in 2011 from NBS or 411 371 expression between organs and their functional roles. The gill WB (P < 0.05). In comparison, there was no significant change 412 372 and intestine are transport epithelia capable of metabolizing in embryonic mortality between treatments. Those embryos 413 25,29,30 373 AhR-inducing xenobiotics, including PAHs. Elevated that did hatch in the GT sediments were significantly smaller, 414

E dx.doi.org/10.1021/es400458p | Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article

Figure 4. Early life-stage phenotypic effects of embryos exposed to sediments collected from an unoiled reference location in North Bay Sansbois in 2011 (NBS 2011) and from oiled locations in Wilkinson Bay in 2011 (WB 2011) and Grande Terre Island (GT 2011 and 2010). (A) Larval length at ≤24 h posthatch was significantly lower for GT 2010 sediment exposed fish. GT 2011 exposed fish were unavailable for length measurements. (B) Heart rate of embryos at 8 DPF. Error bars indicate standard error. Asterisks in both graphs indicate a significant difference compared to the NBS reference sediment exposure (P ≤ 0.001).

415 had pronounced bradycardia during embryonic development (P is likely that the dosage of toxicants needed to elicit observable 459 f4 416 < 0.05) (Figure 4), and had poor vigor at hatch. Yolk-sac and phenotypic effects is higher than that found in WB sediments 460 417 pericardial edema were observed in fish exposed to GT but lower than that found in GT sediments collected at the 461 418 sediments, but they were not observed in WB or NBS larvae same time in 2011(Figures S3−S4 and Figures S1−S2, 462 419 (data not shown). These data are consistent with the Supporting Information), or that other physical or chemical 463 420 characteristic developmental impairments associated with characteristics of these sediments contribute to the effects seen 464 421 early life-stage exposures to crude oil and PAHs and are in these animals. 465 3 fi 422 correlated with an increase in PAHs (and alkanes) in the In Whitehead et al., 2012, adult Gulf killi sh collected at GT 466 423 sediments, coincident with oiling attributed to the DHOS in 2010 exhibited molecular and protein-level responses that 467 424 (Tables S1−S4 and Figures S1−S3, Supporting Informa- are diagnostic of exposure to crude oil. In that study, although 468 3,5,33,34 425 tion). Variation in physiochemical properties of the PAH concentrations were similarly low in tissues and water 469 426 sediments may influence the toxicity of crude oil to developing samples between all sites, PAHs were elevated in GT sediments 470 3 427 embryos, and sex, breeding potential, and other variables are relative to reference sites. The methods used in the 471 12,35 428 known to influence teleost response to crude oil exposure. experiments reported here were designed to assess the potential 472 fi 429 However, these factors are likely minor compared to the for eld-collected sediments from oiled sites to elicit lethal or 473 ff fi 430 influence of oil concentration between sites (Tables S1−S4 and sublethal e ects. By utilizing eld-collected sediments, we 474 431 Figures S1−S3, Suppporting Information). The effects seen in attempted to characterize the developmental potential of Gulf 475 fi 432 these developing fish indicate that resident developing Gulf killi sh at locations that received oil from the DHOS. The 476 433 killifish embryos were exposed to crude oil from these results suggest that sediments were a persistent source of 477 434 sediments (see below) for at least two breeding seasons and biologically available AhR-activating toxicants at oiled sites for 478 435 that this exposure may affect future population demographics at over one year following the landfall of oil. Furthermore, since 479 fi 436 locations where crude oil is present. PAHs in eld-collected water and tissues of resident animals 480 481 437 Immunohistochemistry of Sediment-Exposed Larvae. have tended to be below the detection limits of analytical 482 438 Larvae collected at <24 h posthatch had elevated CYP1A chemistry in areas that received contaminating oil, biological responses appear to be more sensitive indicators of 483 439 protein when exposed to oiled sediments (WB 2011, GT 2010, 3,36 contamination. 484 440 and GT 2011) compared to larvae exposed to unoiled (NBS) f5 441 sediment (Figure 5 and Figure S4, Supporting Information). 442 Elevated CYP1A protein in oiled sediment-exposed larvae was ■ IMPLICATIONS 485 443 found in vascular endothelial cells throughout the body, and in Integration of diverse end points spanning multiple levels of 486 444 the gill, buccopharyngeal epithelium, liver, head kidneys, and biological organization in both laboratory and field studies with 487 fi 445 heart (Figure 5B). NBS sediment-exposed sh showed slightly indigenous organisms as site-specific indications of ecosystem 488 446 elevated CYP1A in kidney tubules, heart, and liver tissue, health enriches the understanding of the effects of environ- 489 447 although elevated expression of CYP1A was not observed in mental change. In seeking to characterize the effects of the 490 448 other tissues. Larvae exposed to GT sediments that had DHOS on at-risk fish, a field study examining the health of 491 449 substantially more CYP1A protein than the other sediment- populations of Gulf killifish was initiated prior to the landfall of 492 fi 450 exposed sh, including stronger staining of the external oil and is ongoing. Here, we presented evidence of exposure to 493 451 epithelial surfaces. Staining in the external epithelia was present PAHs in adult fish coincident with the oil contamination from 494 452 in WB larvae, albeit to a much lesser extent than that of GT the DHOS. Genome expression profiling indicates significant 495 453 sediment-exposed larvae, probably because PAH and alkane divergence in genomic responses of fish from an oiled location 496 454 concentrations were elevated in WB sediments at levels just compared to reference sites. These responses are diagnostic of 497 455 below that of GT 2011 sediments (Tables S1−S2 and Figures exposure to the toxic components of oil and are highly 498 456 S1−S3, Supporting Information). Despite the increased CYP1A correlated with CYP1A protein expression responses in the 499 457 expression in WB sediment-exposed larvae, no bradycardial or gills, liver, head kidney, and intestine of adult and larval fish. 500 458 developmental effects were noted in these animals (Figure 4). It Exposure to sediments from oiled locations caused cardiovas- 501

F dx.doi.org/10.1021/es400458p | Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article

which may translate into longer-term eﬀects at the population 507 level for Gulf killiﬁsh and other biota that live or spawn in 508 similar habitats. 509

■ ASSOCIATED CONTENT 510 *S Supporting Information 511 Tables of ﬁeld sampling sites, analytical chemistry data from 512 sediments, and comparison of representative sediment-exposed 513 larva. This material is available free of charge via the Internet at 514 http://pubs.acs.org. 515

■ AUTHOR INFORMATION 516 Corresponding Author 517 *E-mail: [email protected]. 518 Author Contributions 519 The manuscript was written through contributions of all 520 authors. All authors have given approval to the ﬁnal version of 521 the manuscript. 522 Notes 523 The authors declare no competing ﬁnancial interest. 524

■ ACKNOWLEDGMENTS 525 We thank the College of Science and the Office of Research 526 and Economic Development at Louisiana State University for 527 generously helping to fund initial field work and critical bridge 528 funding support. Thanks to David Roberts and Eve McCulloch 529 for field assistance. Special thanks to Captain Les Barios for 530 Figure 5. Distribution of CYP1A protein (burgundy staining) in larval transportation and shelter in Barataria Bay, Dr. E. William 531 tissues at ≤24 h posthatch exposed throughout development to Wischusen and the students enrolled in Biology 1208 532 sediments collected from an unoiled reference location in North Bay laboratories sections 13 and 14 in the spring of 2011 for help 533 Sansbois in 2011 (NBS 2011), from a mildly oiled location in Wilkinson Bay in 2011 (WB 2011), and a heavily oiled location at in monitoring embryo mortality, hatch, and phenotypic 534 Grande Terre Island in 2010 and 2011 (GT 2011 and 2010). (A) characteristics, and M. Scott Miles for analytical chemistry of 535 Representative image of larvae at <24 h posthatch that were exposed sediment samples. This research was funded in part by the Gulf 536 to WB sediment throughout embryonic development. CYP1A was of Mexico Research Initiative (to F.G. and A.W.), the National 537 mostly found localized in gill (G), intestine (I), head kidney (HK), Science Foundation (DEB-1048206 and DEB-1120512 to 538 liver (L), heart (H) endothelium, vascular endothelial cells (arrows), A.W.), and the National Institutes of Health (R15- 539 and in epithelial cells lining the buccopharyngeal cavity (asterisks) in ES016905−01 to C.D.R.). 540 fish exposed to oiled sediments (i.e., WB or GT sediments). Fish exposed to unoiled sediment from NBS showed staining for CYP1A in ■ ABBREVIATIONS 541 endothelia within the heart and kidney tubules (see Supporting Information, S2). Larvae were sectioned at 4 μm. Figure depicts a DHOS, Deepwater Horizon oil spill; EVOS, Exxon Valdez oil 542 montage of 12 images captured using a 20× objective. (B) Tissue spill; AhR, Aryl hydrocarbon receptor; PAH, polycyclic 543 sections from larvae at <24 h posthatch. Gill tissues from fish exposed aromatic hydrocarbon; GT, Grande Terre Island; BSL, Bay 544 to WB 2011, GT 2011, and GT 2010 sediments had CYP1A-positive St. Louis, MS; BLB, Bayou La Batre, LA; T1, Trip 1; T2, Trip 545 vascular endothelial cells and increased CYP1A in epithelial and pillar 2; T3, Trip 3, T3; CYP1A, cytochrome P4501A; mAb, 546 fi cells in the gill laments and lamellae. Epithelial cells lining the monoclonal antibody; WB, Wilkinson Bay; NOAA, National 547 fi buccopharyngeal cavity also showed increased CYP1A in these sh. Oceanographic and Atmospheric Administration; SCAT, 548 Intestine showed light staining for CYP1A protein in the epithelial Shoreline Cleanup and Assessment Technique; NBS, North 549 cells and endothelial cells in the submucosa in these fish. Head kidneys Bay Sansbois; PTFE, polytetrafluoroethylene; AFS, artificially 550 tubules and vascular endothelial cells in fish exposed to oiled sediment had higher CYP1A staining compared to fish exposed to unoiled NBS formulated seawater; DPF, days postfertilization; BPM, beats 551 2011 sediment. Hepatocytes of liver tissue of fish exposed to oiled per minute; tPAHs, total PAHs 552 sediments showed increased CYP1A expression compared to those of fish exposed to unoiled NBS sediment. Images captured through a 40× ■ REFERENCES 553 objective. Arrows = vascular endothelial cells, chevrons = gill lamellae, (1) Crone, T. J.; Tolstoy, M. Magnitude of the 2010 Gulf of Mexico 554 asterisks = bucopharyngeal cavity, arrow heads = kidney tubules. All oil leak. Science 2010, 330 (6004), 634−634. 555 tissues from A and B were sectioned at 4 μm. Scale bar =50 μm. All (2) Woodin, B. R.; Smolowitz, R. M.; Stegeman, J. J. Induction of 556 slides were counterstained with hematoxylin (blue). cytochrome P4501A in the intertidal fish Anoplarchus purpurescens by 557 prudhoe bay crude oil and environmental induction in fish from Prince 558 William Sound. Environ. Sci. Technol. 1997, 31 (4), 1198−1205. 559 502 fi cular defects in embryonic sh, delayed hatching, and reduced (3) Whitehead, A.; Dubansky, B.; Bodinier, C.; Garcia, T. I.; Miles, 560 503 overall hatching success. Those larvae that do hatch are smaller S.; Pilley, C.; Raghunathan, V.; Roach, J. L.; Walker, N.; Walter, R. B.; 561 504 and have yolk-sac and pericardial edema. These data include Rice, C. D.; Galvez, F. Genomic and physiological footprint of the 562 505 results that encompass two breeding seasons and indicate that Deepwater Horizon oil spill on resident marsh fishes. Proc. Natl. Acad. 563 506 contaminating oil from the DHOS impacts organismal fitness, Sci. 2012, 109 (50), 20298−20302. 564

G dx.doi.org/10.1021/es400458p | Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article

565 (4) Peterson, C. H.; Rice, S. D.; Short, J. W.; Esler, D.; Bodkin, J. L.; (21) Rice, C. D.; Schlenk, D.; Ainsworth, J.; Goksøyr, A. Cross- 633 566 Ballachey, B. E.; Irons, D. B. Long-term ecosystem response to the reactivity of monoclonal antibodies against peptide 277−294 of 634 567 Exxon Valdez oil spill. Science 2003, 302 (5653), 2082−2086. rainbow trout CYP1A1 with hepatic CYP1A among fish. Mar. Environ. 635 568 (5) Short, J. W.; Rice, S. D.; Heintz, R. A.; Carls, M. G.; Moles, A. Res. 1998, 46 (1−5), 87−91. 636 569 Long-term effects of crude oil on developing fish: Lessons from the (22) Clark, B. W.; Matson, C. W.; Jung, D.; Di Giulio, R. T. AHR2 637 570 Exxon Valdez oil spill. Energy Sources 2003, 25 (6), 509−517. mediates cardiac teratogenesis of polycyclic aromatic hydrocarbons 638 571 (6) Claireaux, G.; Desaunay, Y.; Akcha, F.; Auperin, B.; Bocquene, and PCB-126 in Atlantic killifish (Fundulus heteroclitus). Aquat. Toxicol. 639 − 572 G.; Budzinski, F. N.; Cravedi, J. P.; Davoodi, F.; Galois, R.; Gilliers, C.; 2010, 99 (2), 232 240. 640 573 Goanvec, C.; Guerault, D.; Imbert, N.; Mazeas, O.; Nonnotte, G.; (23) Matsumura, F. The significance of the nongenomic pathway in 641 574 Nonnotte, L.; Prunet, P.; Sebert, P.; Vettier, A. Influence of oil mediating inflammatory signaling of the dioxin-activated Ah receptor 642 − 575 exposure on the physiology and ecology of the common sole (Solea to cause toxic effects. Biochem. Pharmacol. 2009, 77 (4), 608 626. 643 576 solea): Experimental and field approaches. Aquat. Living Resour. 2004, (24) Stevens, E. A.; Mezrich, J. D.; Bradfield, C. A. The aryl 644 − hydrocarbon receptor: A perspective on potential roles in the immune 645 577 17 (3), 335 351. − 578 (7) Davoodi, F.; Claireaux, G. Effects of exposure to petroleum system. Immunology 2009, 127 (3), 299 311. 646 647 579 hydrocarbons upon the metabolism of the common sole (Solea solea). (25) VanVeld, P. A.; Vogelbein, W. K.; Cochran, M. K.; Goksoyr, A.; Stegeman, J. J. Route-specific cellular expression of cytochrome 648 580 Mar. Pollut. Bull. 2007, 54 (7), 928−934. P4501A (CYP1A) in fish (Fundulus heteroclitus) following exposure to 649 581 (8) Claireaux, G.; Davoodi, F. Effect of exposure to petroleum aqueous and dietary benzo[a]pyrene. Toxicol. Appl. Pharmacol. 1997, 650 582 hydrocarbons upon cardio-respiratory function in the common sole 142 (2), 348−359. 651 583 (Solea solea). Aquat. Toxicol. 2010, 98 (2), 113−119. (26) Ortiz-Delgado, J. B.; Segner, H.; Sarasquete, C. Cellular 652 584 (9) Hontela, A. Interrenal dysfunction in fish from contaminated distribution and induction of CYP1A following exposure of gilthead 653 585 sites: In vivo and in vitro assessment. Environ. Toxicol. Chem. 1998, 17 seabream, Sparus aurata, to waterborne and dietary benzo[a]pyrene 654 586 (1), 44−48. and 2,3,7,8-tetrachlorodibenzo-p-dioxin: An immunohistochemical 655 587 (10) Frederick, L. A.; Van Veld, P. A.; Rice, C. D. Bioindicators of approach. Aquat. Toxicol. 2005, 75 (2), 144−161. 656 588 immune function in creosote-adapted estuarine killifish, Fundulus (27) Jonsson, E. M.; Brandt, I.; Brunstrom, B. Gill filament-based 657 589 heteroclitus. J. Toxicol. Environ. Health, Part A 2007, 70 (17), 1433− EROD assay for monitoring waterborne dioxin-like pollutants in fish. 658 590 1442. Environ. Sci. Technol. 2002, 36 (15), 3340−3344. 659 591 (11) Rice, C. D.; Kergosien, D. H.; Adams, S. M. Innate Immune (28) Jonsson, E. M.; Abrahamson, A.; Brunstrom, B.; Brandt, I. 660 592 Function as a Bioindicator of Pollution Stress in Fish. Ecotoxicol. − Cytochrome P4501A induction in rainbow trout gills and liver 661 593 Environ. Safety 1996, 33 (2), 186 192. following exposure to waterborne indigo, benzo[a]pyrene and 662 594 (12) Spies, R. B.; Stegeman, J. J.; Hinton, D. E.; Woodin, B.; 3,3′,4,4′,5-pentachlorobiphenyl. Aquat. Toxicol. 2006, 79 (3), 226− 663 595 Smolowitz, R.; Okihiro, M.; Shea, D. Biomarkers of hydrocarbon 232. 664 596 exposure and sublethal effects in embiotocid fishes from a natural (29) Andersson, T.; Part,̈ P. Benzo[a]pyrene metabolism in isolated 665 597 petroleum seep in the Santa Barbara Channel. Aquat. Toxicol. 1996, 34 perfused rainbow trout gills. Mar. Environ. Res. 1989, 28 (1−4), 3−7. 666 − 598 (3), 195 219. (30) Kleinow, K. M.; James, M. O.; Tong, Z.; Venugopalan, C. S. 667 599 (13) Marty, G. D.; Hinton, D. E.; Short, J. W.; Heintz, R. A.; Rice, S. Bioavailability and biotransformation of benzo[a]pyrene in an isolated 668 600 D.; Dambach, D. M.; Willits, N. H.; Stegeman, J. J. Ascites, premature perfused in situ catfish intestinal preparation. Environ. Health Perspect. 669 601 emergence, increased gonadal cell apoptosis, and cytochrome P4501A 1998, 106 (3), 155−166. 670 602 induction in pink salmon larvae continuously exposed to oil- (31) Nakayama, A.; Riesen, I.; Kollner, B.; Eppler, E.; Segner, H. 671 603 contaminated gravel during development. Can. J. Fish. Aquat. Sci. Surface marker-defined head kidney granulocytes and B lymphocytes 672 604 1997, 75 (6), 989−1007. of rainbow trout express benzo[a]pyrene-inducible cytochrome 673 605 (14) Heintz, R. A.; Rice, S. D.; Wertheimer, A. C.; Bradshaw, R. F.; P4501A protein. Toxicol. Sci. 2008, 103 (1), 86−96. 674 606 Thrower, F. P.; Joyce, J. E.; Short, J. W. Delayed effects on growth and (32) Brand, D. G.; Fink, R.; Bengeyfield, W.; Birtwell, I. K.; 675 607 marine survival of pink salmon (Oncorhynchus gorbuscha)after McAllister, C. D. Salt water-acclimated pink salmon fry (Oncorhynchus 676 608 exposure to crude oil during embryonic development. Mar. Ecol.: gorbuscha) develop stress-related visceral lesions after 10-day exposure 677 609 Prog. Ser. 2000, 208, 205−216. to sublethal concentrations of the water-soluble fraction of north slope 678 610 (15) Incardona, J. P.; Collier, T. K.; Scholz, N. L. Oil spills and fish crude oil. Toxicol. Pathol. 2001, 29 (5), 574−584. 679 611 health: exposing the heart of the matter. J. Expo. Sci. Environ. Epidemiol. (33) Incardona, J. P.; Collier, T. K.; Scholz, N. L. Defects in cardiac 680 612 2011, 21 (1), 3−4. function precede morphological abnormalities in fish embryos exposed 681 613 (16) Hicken, C. E.; Linbo, T. L.; Baldwin, D. H.; Willis, M. L.; Myers, to polycyclic aromatic hydrocarbons. Toxicol. Appl. Pharmacol. 2004, 682 − 614 M. S.; Holland, L.; Larsen, M.; Stekoll, M. S.; Rice, S. D.; Collier, T. 196 (2), 191 205. 683 615 K.; Scholz, N. L.; Incardona, J. P. Sublethal exposure to crude oil (34) Incardona, J. P.; Carls, M. G.; Day, H. L.; Sloan, C. A.; Bolton, J. 684 616 during embryonic development alters cardiac morphology and reduces L.; Collier, T. K.; Scholz, N. L. Cardiac arrhythmia is the primary 685 617 aerobic capacity in adult fish. Proc. Natl. Acad. Sci. 2011, 108 (17), response of embryonic Pacific Herring (Clupea pallasi) exposed to 686 − 618 7086−7090. crude oil during weathering. Environ. Sci. Technol. 2009, 43 (1), 201 687 619 (17) de Soysa, T. Y.; Ulrich, A.; Friedrich, T.; Pite, D.; Compton, S.; 207. 688 689 620 Ok, D.; Bernardos, R.; Downes, G.; Hsieh, S.; Stein, R.; Lagdameo, M. (35) Martinez-Gomez, C.; Vethaak, A. D.; Hylland, K.; Burgeot, T.; 690 621 C.; Halvorsen, K.; Kesich, L.-R.; Barresi, M. Macondo crude oil from Kohler, A.; Lyons, B. P.; Thain, J.; Gubbins, M. J.; Davies, I. M. A guide to toxicity assessment and monitoring effects at lower levels of 691 622 the Deepwater Horizon oil spill disrupts specific developmental biological organization following marine oil spills in European waters. 692 623 processes during zebrafish embryogenesis. BMC Biol. 2012, 10 (1), 40. ICES J. Mar. Sci. 2010, 67 (6), 1105−1118. 693 624 (18) Whitehead, A.; Pilcher, W.; Champlin, D.; Nacci, D. Common (36) Ylitalo, G. M.; Krahn, M. M.; Dickhoff, W. W.; Stein, J. E.; 694 625 mechanism underlies repeated evolution of extreme pollution Walker, C. C.; Lassitter, C. L.; Garrett, E. S.; Desfosse, L. L.; Mitchell, 695 626 tolerance. Proc. R. Soc. B: Biol. Sci. 2012, 279 (1728), 427−433. K. M.; Noble, B. T.; Wilson, S.; Beck, N. B.; Benner, R. A.; 696 627 (19) Huang, D. W.; Sherman, B. T.; Lempicki, R. A. Systematic and Koufopoulos, P. N.; Dickey, R. W. Federal seafood safety response to 697 628 integrative analysis of large gene lists using DAVID bioinformatics the Deepwater Horizon oil spill. Proc. Natl. Acad. Sci. 2012, 698 629 resources. Nature Protocols 2009, 4 (1), 44−57. DOI: 10.1073/pnas.1108886109 . 699 630 (20) Sarasquete, C.; Segner, H. Cytochrome P4501A (CYP1A) in 631 teleostean fishes. A review of immunohistochemical studies. Sci. Total 632 Environ. 2000, 247 (2−3), 313−332.

H dx.doi.org/10.1021/es400458p | Environ. Sci. Technol. XXXX, XXX, XXX−XXX REPORTS

References and Notes 13. See supplementary materials on Science Online. in the Drosophila media kitchen at UC Davis. Funded by NIH 1. D. J. Begun, H. A. Lindfors, M. E. Thompson, 14. N. Sheth et al., Nucleic Acids Res. 34, 3955–3967 grant GM084056 (D.J.B.) and NSF grant 0920090 (C.D.J and A. K. Holloway, Genetics 172, 1675–1681 (2006). (2006). D.J.B.). Author contributions: L.Z. and D.J.B. conceived and 2. M. T. Levine, C. D. Jones, A. D. Kern, H. A. Lindfors, 15. V. N. Uversky, Protein Sci. 11, 739–756 (2002). designed the experiments; L.Z., P.S., and C.D.J. generated D. J. Begun, Proc. Natl. Acad. Sci. U.S.A. 103, 16. D. Hebenstreit et al., Mol. Syst. Biol. 7, 497 (2011). data; L.Z. performed data analysis; P.S. and L.Z. did molecular 9935–9939 (2006). 17. D. Schwartz, Genetics 67, 411–425 (1971). experiments; L.Z. and D.J.B. wrote the paper; and L.Z., D.J.B., 3. D. J. Begun, H. A. Lindfors, A. D. Kern, C. D. Jones, 18. R. M. Graze et al., Mol. Biol. Evol. 29, 1521–1532 (2012). and C.D.J. revised the paper. Illumina reads produced in Genetics 176, 1131–1137 (2007). 19. J. M. Smith, J. Haigh, Genet. Res. 23,23–35 (1974). this study are deposited at NCBI BioProject under accession 4. Q. Zhou et al., Genome Res. 18, 1446–1455 (2008). 20. N. L. Kaplan, R. R. Hudson, C. H. Langley, Genetics 123, number PRJNA210329. 5. D. G. Knowles, A. McLysaght, Genome Res. 19, 887–899 (1989). 1752–1759 (2009). 21. J. Hermisson, P. S. Pennings, Genetics 169, 2335–2352 Supplementary Materials 6. W. Xiao et al., PLOS ONE 4, e4603 (2009). (2005). www.sciencemag.org/content/343/6172/769/suppl/DC1 7. J. Cai, R. Zhao, H. Jiang, W. Wang, Genetics 179, 22. D. J. Begun et al., PLOS Biol. 5, e310 (2007). Materials and Methods 487–496 (2008). 23. C. H. Langley et al., Genetics 192, 533–598 (2012). Supplementary Text 8. A. R. Carvunis et al., Nature 487, 370–374 (2012). Figs. S1 to S10 9. T. J. A. J. Heinen, F. Staubach, D. Häming, D. Tautz, Acknowledgments: We thank the Begun lab for valuable Tables S1 to S17 Curr. Biol. 19, 1527–1531 (2009). comments, especially N. Svetec, J. Cridland, A. Sedghifar, References (24–58) 10. S. Chen, Y. E. Zhang, M. Long, Science 330, 1682–1685 T. Seher, and Y. Brandvain. We also thank three anonymous (2010). reviewers and M. W. Hahn, C. H. Langley, J. Anderson, and 11 November 2013; accepted 7 January 2014 11. J. A. Reinhardt et al., PLOS Genet. 9, e1003860 (2013). A. Kopp for comments and suggestions. We acknowledge Published online 23 January 2014; 12. T. F. C. Mackay et al., Nature 482, 173–178 (2012). Phyllis Wescott (1950–2011) for more than 30 years of service 10.1126/science.1248286

may have spawned in oiled habitats, including Crude Oil Impairs Cardiac yellowfin tuna, dolphin fish, blue marlin, and swordfish (15). Excitation-Contraction Coupling in Fish To more precisely define the mechanisms of crude oil cardiotoxicity and to evaluate the poten- Fabien Brette,1 Ben Machado,1 Caroline Cros,1 John P. Incardona,2 tial vulnerability of eggs, larvae, and juveniles in Nathaniel L. Scholz,2 Barbara A. Block1* the vicinity of the DWH spill, we assessed the impact of field-collected DWH oil samples on Crude oil is known to disrupt cardiac function in fish embryos. Large oil spills, such as the in vitro cardiomyocyte preparations dissociated Deepwater Horizon (DWH) disaster that occurred in 2010 in the Gulf of Mexico, could severely from the hearts of bluefin tuna (T. orientalis) and affect fish at impacted spawning sites. The physiological mechanisms underlying such potential yellowfin tuna (T. albacares). Juvenile tunas were cardiotoxic effects remain unclear. Here, we show that crude oil samples collected from the DWH caught at sea and held in captivity at the Tuna spill prolonged the action potential of isolated cardiomyocytes from juvenile bluefin and yellowfin Research Conservation Center and the Monterey tunas, through the blocking of the delayed rectifier potassium current (IKr). Crude oil exposure also Bay Aquarium (16). decreased calcium current (ICa) and calcium cycling, which disrupted excitation-contraction The cardiotoxic effects of four distinct envi- coupling in cardiomyocytes. Our findings demonstrate a cardiotoxic mechanism by which crude oil ronmental samples of MC252 crude oil were as- affects the regulation of cellular excitability, with implications for life-threatening arrhythmias sessed as water-accommodated fractions (WAFs) in vertebrates. prepared in Ringer solution for marine fish (16). Oil samples were collected under chain of custody rude oil is a complex chemical mixture Numerous studies on crude oils and PAHs, par- during the DWH spill response effort. The samples containing hydrocarbons (aliphatic and ticularly in the aftermath of the Exxon Valdez spill, included riser “source” oil (sample 072610-03), riser Caromatic) and other dissolved-phase or- have described embryonic heart failure, brady- oil that was “artificially weathered” by heating at ganic compounds. Toxicity research on crude oil cardia, arrhythmias, reduction of contractility, 90° to 105°C (sample 072610-W-A), and two skimmed constituents has focused mainly on polycyclic and a syndrome of cardiogenic fluid accumu- oil samples: “slick A” (sample CTC02404-02), aromatic hydrocarbons (PAHs) (1, 2), pervasive lation (edema) in exposed fish embryos (6, 7). collected 29 July 2010, and “slick B” (sample environmental contaminants that are also found These severe effects are lethal to embryos and GU2888-A0719-OE701), collected 19 July 2010 in coal tar, creosote, air pollution, and land-based larval fishes (8–10) and could be due to atrio- by the U.S. Coast Guard cutter Juniper.High- runoff. In the aftermath of oil spills, PAHs can ventricular conduction block (11). energy WAFs were prepared in a commercial persist for many years in marine habitats and Despite recent progress using zebrafish and blender that dispersed oil droplets to mimic re- thereby create pathways for lingering biological other experimental models to study PAH car- lease conditions at the MC252 well head (16). exposure and associated adverse effects. diotoxicity (12), the mechanisms that underpin As expected from previous studies (11, 12), the PAH toxicity is structure-dependent, and the the physiological effects on cardiac function and total sum (∑) of PAHs declined in WAFs from carcinogenic, mutagenic, and teratogenic proper- changes in cardiac morphology during develop- source oil to the surface-weathered samples, owing ties of many individual PAHs are known (3, 4). ment are not known. The Deepwater Horizon to loss of naphthalenes, whereas the total concen- Developing fish are particularly vulnerable to (DWH) oil spill released >4 million barrels of trations of three-ringed PAHs (e.g., phenanthrenes) dissolved PAHs in the range of ~100 parts per crude oil during the peak spawning months for increased proportionately (fig. S1 and table S1). billion (ppb or mg/liter) down to ≤10 mg/liter. Atlantic bluefin tuna (Thunnus thynnus) in 2010. PAH concentrations were in a range expected to Consequently, PAH toxicity to fish early life stages This large and long-lived species reaches a mass cause cardiotoxicity in intact embryos and consist- is an important contributor to both acute and long- of 650 kg over a life span of 35 years or more ent with the ∑PAHs measured in some surface term impacts of environmental disasters (2, 5). (13), and the Gulf of Mexico population of blue- water samples during the DWH oil spill (up to fin tuna is severely depleted (14). Electronic- 84 mg/liter) (16, 17). WAFs in Ringer solution tagging data confirm that bluefin tuna spawn were perfused over freshly dissociated, isolated 1Department of Biology, Stanford University, Hopkins Marine Station, Pacific Grove, CA 93950, USA. 2Northwest Fisheries in the vicinity of the DWH spill, which indicates tuna cardiomyocytes, and we assessed the effects Science Center, National Oceanic and Atmospheric Adminis- that bluefin tuna embryos, larvae, juveniles, and of these oil-containing solutions on excitation- tration, Seattle, WA 98112, USA. adults were likely exposed to crude oil–derived contraction (EC) coupling using electrophysio- *Corresponding author. E-mail: [email protected] PAHs (14). Many other Gulf of Mexico pelagics logical and Ca2+-imaging techniques.

772 14 FEBRUARY 2014 VOL 343 SCIENCE www.sciencemag.org REPORTS

Patch-clamp recordings revealed a strong ef- potential amplitude (figs. S2 and S3). This sug- (IKr)(19)—are known to cause or aggravate ven- fect of DWH source oil and weathered oil on gests that IK1, the background current responsible tricular arrhythmias, which can result in torsade bluefin and yellowfin tuna cardiomyocytes’ ac- for resting membrane potential, and INa,thecur- de pointes and/or sudden death (20). tion potential duration (APD) (Fig. 1 and fig. S3). rent responsible for the upstroke of the action Overall, the effects of MC252 oil WAFs on A concentration-dependent lengthening of the potential, are not modified by crude oil. All four cardiomyocyte action potentials in bluefin and APD waveform was observed in both tuna spe- oil samples significantly increased the time for yellowfin tunas were similar. However, the car- cies. APD at 90% repolarization (i.e., equivalent repolarization from APD30 to APD90. This in- diotoxic potency of each oil sample correlated to the QT interval on an electrocardiogram) was crease in triangulation (Fig. 1, I to L, and fig. S3, closely with the concentrations of three-ringed significantly increased across all four oil sam- E and F) is a strong predictor of fatal cardiac ar- PAHs rather than total ∑PAHs (fig. S4), as evi- ples at ∑PAH concentrations ranging from 4 to rhythmia (18). Pharmacological agents that cause denced in particular by the extensively weathered 61 mg/liter (table S1). The source and weathered a cardiac repolarization disorder by lengthening slick B sample (fig. S1 and table S1), which in- oils significantly decreased the APD at 10% re- cardiomyocyte APD, as well as congenital muta- creased both APD and triangulation without af- polarization (APD10) (Fig. 1). WAF exposures did tions of hERG (human ether-à-go-go–related gene fecting resting membrane potential or amplitude not influence other action potential parameters, or KCNH2)channels—the mammalian homolog (Fig. 1 and figs. S2 and S3). In some cardiomyo- such as resting membrane potential and action to the fish delayed rectifier potassium current cytes, WAFs caused unstable action potentials after depolarizations (fig. S5B). Such arrhythmias were not observed among ventricular cells in Ringer solution over an equivalent recording duration (fig. S5A). The functional effects of PAHs on fish cardiac rhythmicity suggest that components of crude oil interfere with EC coupling, which links electrical excitation to contraction in cardiomyocytes (21, 22). Depolarization of the cardiac sarcolemmal membrane opens voltage-gated ion channels, including L-type Ca2+ channels, which results in Ca2+ entry into the cytosol. This Ca2+ transient triggers the release of additional Ca2+ from internal stores [sarcoplasmic reticulum (SR)] by means of a Ca2+-induced Ca2+ release mechanism (CICR) (23–25). The rise in intracellular Ca2+ activates the contractile machinery within the cardiomyocyte. Critical for action potential repolarization are the opening and closing of voltage-gated Na+, Ca2+,andK+ channels, which renew the EC coupling process at every heartbeat. The repolarization of the tuna cardiomyocyte action potential in- volves a delicate balance of inward and outward ionic currents. Thus, cardiac action potential prolongation could be due to a decrease in outward current, an increase in inward current, or both. To distinguish between these possibilities, we used electrophysiological analyses (voltage clamp) to investigate the influence of slick B (as a representative oil sample of all four WAFs) on the major outward currents (IK) and inward calcium current (ICa) in isolated cardiomyocytes. We characterized the rapid component of the delayed potassium current (IKr) in the bluefin tuna using electrophysiological and pharmacological techniques as previously described (i.e., E-4031–sensitive current (26)]. In the bluefin tuna ventricular cardiomyocyte, IKr amplitude and tail current were reduced in a concentration-dependent manner in response to exposures to slick B WAF (Fig. 2, A to C) with a half maximum inhibitory concentration (IC50)of51T 6 mg ∑PAHs per liter T Fig. 1. Effect of oil WAFs on action potential characteristics from bluefin tuna ventricular car- and a Hill coefficient of 1.19 0.11. Perfusion diomyocytes. (A to D) Action potentials in controls (black) and with ascending concentrations of source with surface oil (slick A) also decreased IKr in oil (blue traces), artificially weathered (orange traces), slick A (green traces), and slick B (red traces) bluefin tuna ventricular cardiomyocytes (fig. S6) T m WAFs. (E to H) APD (expressed as a percentage of control) at 10, 50, and 90% repolarization in control with a similar IC50 (53 31 g/liter) and Hill (black bars) and with ascending concentrations of source (n = 9), artificially weathered (n = 8), slick A (n =7), coefficient (1.16 T 0.43). In yellowfin tuna, ex- and slick B (n =7).(I to L) Action potential triangulation (expressed as a percentage of control; posure of ventricular cardiomyocytes to slick B calculated as APD90 – APD30) in control (black bars) and with ascending concentrations of source, WAF also significantly decreased IKr tail currents artificially weathered, slick A, and slick B. (E) to (L): Means T SEM. *P <0.05. IC50 =61T 12 mg/liter, Hill coefficient = 0.84 T

www.sciencemag.org SCIENCE VOL 343 14 FEBRUARY 2014 773 REPORTS

0.11) (fig. S7). Taken together, these data show sponse to oil is Ca2+-dependent and not voltage- Intracellular Ca2+ transients in bluefin tuna car- that DWH crude oils significantly decrease IKr dependent. The decrease in ICa amplitude and diomyocytes were recorded using confocal currents in both species. The effect of slick B slowing of inactivation might have countervail- microscopy and Ca2+-sensitive dye (Fluo-4). Ex- 2+ (22 mg ∑PAHs per liter) on the IKr current- ing effects on Ca entry during the plateau phase posures to each oil sample (source, artificially voltage (I-V) relation is shown in Fig. 2D. WAF of the action potential, as measured from action weathered source, slick A, and slick B at 30, perfusion reduced IKr amplitudes across all volt- potential waveforms in response to physiological 18, 7, and 11 mg ∑PAHs per liter, respectively) ages without affecting the shape of the I-V curve pulses (29). significantly decreased the Ca2+ transient am- 2+ 2+ (Fig. 2E). In addition, IKr tail currents were de- The entry of Ca via ICa during action po- plitudes and slowed the decay of the Ca tran- creased at all voltages without shifting the curve tentials was similar among controls and ventric- sients in bluefin tuna ventricular cardiomyocytes (Fig. 2F). Bluefin tuna ventricular cardiomyocytes ular cardiomyocytes perfused with slick B WAF (Fig. 4). This reduction in Ca2+ transient am- exposed to source oil (61 mg/liter) and yellowfin (22 mg ∑PAHs per liter) for bluefin (fig. S13) and plitudes would decrease contractility and would tuna ventricular cardiomyocytes exposed to slick yellowfin tunas (fig. S14). Overall, the absence reduce cardiac output at the scale of the whole B (22 mg/liter) showed comparable blockade of of an effect of crude oil on Ca2+ entry during a heart. A diminished cytosolic Ca2+ transient could 2+ IKr (figs. S8 and S9). Thus, dissolved constituents physiological pulse is attributable to (i) an in- be a consequence of reduced extracellular Ca of MC252 crude oil do not affect the voltage- crease in APD, allowing more time for Ca2+ influx, a smaller Ca2+ release from SR internal dependent (gating) properties of the K+ chan- entry; (ii) a leftward shift in the activation prop- stores, or both (24, 30). nel but rather inhibit outward conductance in erties of Ca2+ channels; and (iii) a slowing of The direct measurements of Ca2+ transients in + the open state, most likely by blocking the K ICa inactivation. Although our findings are not cardiomyocytes indicate there may be inhibitory channel pore. This mechanism would be consist- sufficient to explain action potential prolonga- effects of oil on SR Ca2+ release and/or reuptake. ent with the observed prolongation of the cardio- tion, they show that DWH crude oil significantly To investigate the possible SR sites of interaction, myocyte action potential. To confirm this, we decreases ICa amplitude in cardiomyocytes of cardiomyocytes were exposed to pharmaco- perfused tuna ventricular cardiomyocytes with tunas. L-type Ca2+ channels play a key role in ini- logical inhibitors of SR Ca2+ release channels 2+ the specific IKr blocker, E-4031 (2 mM in Ringer tiating the critical CICR from SR internal stores; (5 mM ryanodine) and Ca adenosine triphos- solution). As anticipated, E4031 significantly pro- thus, the next series of experiments were designed phatase (ATPase) pumps (2 mM thapsigargin) 2+ longed APD90, consistent with IKr shaping the to measure whole-cell Ca cycling in isolated for at least 30 min before exposures to WAFs. repolarization of bluefin and yellowfin tuna cardio- cardiomyocytes exposed to DWH oils. Under pharmacological blockade, the four dis- myocytes (fig. S10). ICa also plays a critical role in cardiomyocyte APD (27, 28). Exposure to the weathered slick B surface sample significantly decreased the amplitude of ICa (Fig. 3, A to C) in a concentration- dependent manner, with an IC50 of 36 T 7 mg ∑PAHs per liter and a Hill coefficient of 0.76 T 0.13 for bluefin tuna cardiomyocytes. Note that slick B WAF also slowed the inactivation decay of ICa (Fig. 3, D and E) and thereby allowed more Ca2+ entry during depolarization (27). As indicated by quantification of Ca2+ entry, there was a small, but not significant, decrease in charge passing through the channel during the square pulse (Fig. 3, F and G). I-V relations (Fig. 3H) revealed an inhibitory effect of slick B WAF on Fig. 2. Effect of oil WAF + ICa across all voltages, with a slight influence on (slick B) on K current (IK) the shape of the I-V curve (Fig. 3I, top), which from bluefin tuna ventri- suggested a change in the voltage-dependent prop- cular cardiomyocytes. (A) 2+ erties of Ca channels. Perfusion with the slick B IK recorded in control condi- WAF shifted the activation curve toward more tion (black trace), with ascend- hyperpolarized potentials (by ~7 mV) (Fig. 3I, ing concentrations of slick B 2+ I bottom), which allowed more Ca entry at neg- WAF (red traces) or the K block- ative potentials. As with bluefin tuna, slick B er E4031 (2 mM, gray trace). WAF (22 mg ∑PAHs per liter) also inhibited I (Inset) Voltage step to record Ca I I in ventricular cardiomyocytes of yellowfin tuna K.(B) Kr tail in control (black (IC =46T 5 mg/liter, Hill coefficient = 1.01 T bar) and with ascending con- 50 centrations of slick B (red bars, 0.09) (fig. S11). n =9).(C) Change in I tail To further explore the influence of DWH oil Kr (expressed as a percentage of on the voltage-dependent properties of cardiac 2+ control) with ascending con- Ca channels, ICa was measured in bluefin car- I V 2+ centrations of slick B. (D) - diomyocytes with Ba as a charge carrier. In I 2+ relation of K in control (black the absence of Ca -dependent inactivation, trace), slick B (22 mg/liter, the channel inactivates primarily via voltage- red trace), and with E4031 (2 dependent processes (27). Similar to the effects mM, gray trace). (E and F) I-V on ICa, slick B WAF significantly decreased the relation of IKr [circle in (E)] amplitude of IBa but did not slow the inactiva- and tail IKr [triangle in (F)] in tion of the current (fig. S12). This suggests that control (black trace) and with the observed change in ICa inactivation rate in re- slick B (22 mg/liter, red trace, n = 9). (B), (C), (E), and (F): Means T SEM. *P <0.05.

774 14 FEBRUARY 2014 VOL 343 SCIENCE www.sciencemag.org REPORTS

2+ tinct crude oil samples had no significant effect an additional influx of ICa via L-type Ca chan- with crude oil are similar to the physiological on the amplitude of the cytosolic Ca2+ transient nels during action potential prolongation, consist- effects of the antimalarial drug halofantrine, a (Fig. 4D). This indicates that the oil-induced ent with the ICa results in Fig. 3. chemical with structural similarities to three- decrease in Ca2+ transient amplitude observed Our experimental findings provide a mecha- ringed PAHs that causes K+ channel inhibition in the absence of blockers is due to a disruption nistic underpinning for cardiac-specific physio- and cardiac arrhythmias (31). Our results indi- of SR Ca2+ release and/or reuptake from inter- logical defects previously reported and reinforce cate compounds in DWH oil produce a cardio- nal stores. However, these toxic effects of oil on the findings that crude oil has deleterious phys- toxic mechanism that have direct effects on ion intracellular Ca2+ cycling were partially offset by iological impacts on fish hearts (10). Our results channels involved in the EC coupling and cardiac

2+ Fig. 3. Effect of oil WAF (slick B) on Ca current (ICa) from bluefin tuna ventricular cardiomyocytes. (A) ICa recorded in control condition (black trace) and in ascending concentrations of slick B WAF (red traces). (Inset) Voltage step to record ICa.(B) ICa amplitude in control (black bar) and with ascending concentrations of slick B (red bars, n =7).(C) Change in ICa amplitude (expressed as a percentage of control) with ascending concentrations of slick B (red bars). (D) Time to decline to 37% of ICa peak (T0.37)incontrolandwith ascending concentrations of slick B. (E) Change in T0.37 (expressed as a percentage of control) with ascending concentrations of slick B (red bars). (F) ICa charge in control and with ascending concentrations of slick B. (G) Change in ICa charge (expressed as a percentage of control) with ascending concentrations of slick B WAF. (H) I-V relations of ICa in control condition (black trace) and with slick B (22 mg/liter, red trace). (Inset) Voltage step to record ICa. (I) I-V relation of ICa in control (black trace) and with slick B (22 mg/liter, red trace, n =7).(J) Availability-voltage relation (normalized conductance G/Gmax)ofICa in control and with slick B (22 mg/liter, n = 7). (B) to (G), (I), and (J): Means T SEM. *P <0.05.

Fig. 4. Effect of oil WAFs on Ca2+ transients from bluefin tuna ventricular cardiomyocytes. (A)Ca2+ transients recorded in control (black trace) and with source (30 mg/liter, blue trace); artificially weathered (18 mg/liter, orange trace); slick A (7 mg/liter, green trace); and slick B (11 mg/liter, red trace) WAFs, respectively. (B )Ca2+ transients amplitude as fluorescence divided by baseline fluorescence (F/F0) and (C) tau, the decay time constant of Ca2+ transients, in control (n =37)andsource(n = 29), artificially weathered (n = 27), slick A (n =25),and and slick B (11 mg/liter, red trace), respectively. (E)Ca2+ transients amplitude 2+ 2+ slick B (n = 21), respectively. (D)Ca transients recorded in ryanodine (Rya) (F/F0)and(F) tau of decay of Ca transients in ryanodine and thapsigargin and thapsigargin (Tg) (black trace) and with source (30 mg/liter, blue trace); (n =22)andsource(n = 17), artificially weathered (n = 18), slick A (n = 17), and artificially weathered (18 mg/liter, orange trace); slick A (7 mg/liter, green trace); slick B (n = 16), respectively. (B), (C), (E), and (F): Means T SEM. *P <0.05.

www.sciencemag.org SCIENCE VOL 343 14 FEBRUARY 2014 775 REPORTS

contractility of cardiomyocytes. These pathways References and Notes 26. G. L. Galli, M. S. Lipnick, B. A. Block, Am. J. Physiol. in cardiac muscle cells are highly conserved across 1. M. G. Carls, J. P. Meador, Hum. Ecol. Risk Assess. Int. J. Regul. Integr. Comp. Physiol. 297, R502–R509 (2009). 15, 1084–1098 (2009). 27. F. Brette, J. Leroy, J. Y. Le Guennec, L. Sallé, all vertebrates, which explains the common, ca- 2. C. H. Peterson et al., Science 302, 2082–2086 (2003). Prog. Biophys. Mol. Biol. 91,1–82 (2006). nonical crude oil toxicity syndrome observed in a 3. C. E. Boström et al., Environ. Health Perspect. 110 28. F. Brette et al., Biochem. Biophys. Res. Commun. 374, diversity of fish species from habitats that range (suppl. 3), 451–488 (2002). 143–146 (2008). from tropical freshwater (zebrafish) to boreal ma- 4. J. J. Stegeman, J. J. Lech, Environ. Health Perspect. 90, 29. F. Brette, L. Sallé, C. H. Orchard, Biophys. J. 90, 101–109 (1991). 381–389 (2006). rine (herring). 5. J. P. Incardona et al., Proc. Natl. Acad. Sci. U.S.A. 109, 30. H. A. Shiels, E. White, Am. J. Physiol. Regul. Integr. In conclusion, the oil-induced disruption of E51–E58 (2012). Comp. Physiol. 288, R1756–R1766 (2005). cardiomyocyte repolarization via K+ channel 6. M. G. Carls, S. D. Rice, J. E. Hose, Environ. Toxicol. Chem. 31. H. Tie et al., Br. J. Pharmacol. 130, 1967–1975 (2000). blockade and sarcolemmal and SR Ca2+ cycling 18, 481 (1999). 32. C. E. Hicken et al., Proc. Natl. Acad. Sci. U.S.A. 108, 7.R.A.Heintz,J.W.Short,S.D.Rice,Environ. Toxicol. Chem. – should call attention to a previously underap- 7086 7090 (2011). 18, 494 (1999). 33. J. M. Blank et al., J. Exp. Biol. 207, 881–890 (2004). preciated risk to wildlife and humans, particularly 8. J. P. Incardona, T. K. Collier, N. L. Scholz, Toxicol. Appl. – from exposure to cardioactive PAHs that are also Pharmacol. 196, 191 205 (2004). Acknowledgments: We thank D. Bers and K. Ginsburg of 9. J. P. Incardona et al., Environ. Health Perspect. 113, University of California, Davis; R. Kochevar of Stanford relatively enriched in air pollution. IKr inhibition – 1755 1762 (2005). University; and R. Ricker of NOAA for constructive advice by DWH crude oil from the MC252 well was 10. J. P. Incardona et al., Environ. Sci. Technol. 43, 201–207 on an earlier draft of the manuscript. We are also grateful to (2009). robust, and its properties are consistent with di- C. Farwell, A. Norton, and E. Estess of the Monterey Bay + 11. J. H. Jung et al., Chemosphere 91, 1146–1155 (2013). rect channel pore block. These K channel targets Aquarium, along with husbandry teams for maintenance of 12. J. P. Incardona, T. K. Collier, N. L. Scholz, J. Expo. Sci. bluefin and yellowfin tunas and handling of oil experiments and their unique gating properties play a critical Environ. Epidemiol. 21,3–4 (2011). role in cardiac action potential repolarization and 13. J. D. Neilson, S. E. Campana, Can. J. Fish. Aquat. Sci. 65, within Tuna Research and Conservation Center; the Government of Mexico for permitting the collection of tunas are highly conserved across the animal king- 1523–1527 (2008). 14. B. A. Block et al., Nature 434, 1121–1127 (2005). in their waters; T. Dunn and the crew of the F/V Shogun; dom (20). These results lead us to believe that 15. J. R. Rooker et al., PLOS ONE 7, e34180 (2012). and H. Shiels, G. Galli, and S. Thompson for experimental help PAH cardiotoxicity was potentially a common 16. Materials and methods are available as supplementary during preliminary studies of PAH compounds. This work was form of injury among a broad range of species materials on Science Online. funded as a contributing study to the DWH–MC252 incident during and after the DWH oil spill. The early 17. A. R. Diercks et al., Geophys. Res. Lett. 37, L20602 (2010). Natural Resource Damage Assessment. Additional funds were 18. L. M. Hondeghem, L. Carlsson, G. Duker, Circulation 103, provided by the Monterey Bay Aquarium Foundation and life stages of fish and other vertebrates may have 2004–2013 (2001). Stanford University. All experiments with crude oil were been particularly vulnerable, given that even a 19. J. M. Nerbonne, R. S. Kass, Physiol. Rev. 85, 1205–1253 approved by Stanford University Institutional Animal Care transient and sublethal effect of PAHs on the em- (2005). and Use Committee procedures. All data necessary to bryonic heartbeat can cause permanent second- 20. P. Kannankeril, D. M. Roden, D. Darbar, Pharmacol. Rev. understand this manuscript are presented in the main text 62, 760–781 (2010). or supplementary materials. ary changes in heart shape and cardiac output (32). 21. D. M. Bers, Nature 415, 198–205 (2002). Moreover, the underlying ion channel currents that 22. M. B. Cannell, H. Cheng, W. J. Lederer, Science 268, Supplementary Materials drive the electrical properties of cardiomyocytes in 1045–1049 (1995). www.sciencemag.org/content/343/6172/772/suppl/DC1 tunas and mammals (such as heart rates), are sim- 23. H. A. Shiels, E. V. Freund, A. P. Farrell, B. A. Block, Materials and Methods Figs. S1 to S14 ilar (26, 33). Thus, we suggest the extension of our J. Exp. Biol. 202, 881–890 (1999). 24. H. A. Shiels, A. Di Maio, S. Thompson, B. A. Block, Table S1 – current oil toxicity results to mammalian, cardio- Proc. Biol. Sci. 278,18–27 (2011). References (34 41) myocytes may be warranted to better understand 25. M. Vornanen, H. A. Shiels, A. P. Farrell, Comp. Biochem. 3 July 2013; accepted 11 December 2013 PAH threats to human health. Physiol. A Mol. Integr. Physiol. 132, 827–846 (2002). 10.1126/science.1242747

on enumeration of cell types and their precise Massively Parallel Single-Cell definition, which can be controversial (4–7)and is based in many cases on indirect association of RNA-Seq for Marker-Free Decomposition function with cell-surface markers (5–8). Perhaps the best understood model for cellular differentiation and diversification is the hematopoietic sys- of Tissues into Cell Types tem. The developmental tree branching from hematopoietic stem cells toward distinct immu- 1 2,3 1 1 Diego Adhemar Jaitin, * Ephraim Kenigsberg, * Hadas Keren-Shaul, * Naama Elefant, nological functions was carefully worked out 1 1 1 2,3 1 Franziska Paul, Irina Zaretsky, Alexander Mildner, Nadav Cohen, Steffen Jung, through many years of study, and effective cell- 2,3 1 Amos Tanay, †‡ Ido Amit †‡ surface markers are available to quantify and sort the major hematopoietic cell types. Even in this In multicellular organisms, biological function emerges when heterogeneous cell types form well-explored system, however, it is becoming complex organs. Nevertheless, dissection of tissues into mixtures of cellular subpopulations is increasingly difficult to explain modern genome- currently challenging. We introduce an automated massively parallel single-cell RNA sequencing wide and in vivo data with refined cell types’ (RNA-seq) approach for analyzing in vivo transcriptional states in thousands of single cells. hierarchy and functions that extend beyond the Combined with unsupervised classification algorithms, this facilitates ab initio cell-type classical myeloid and lymphoid cell types. For characterization of splenic tissues. Modeling single-cell transcriptional states in dendritic cells and example, dendritic cells (DCs) are antigen-presenting additional hematopoietic cell types uncovers rich cell-type heterogeneity and gene-modules activity in steady state and after pathogen activation. Cellular diversity is thereby approached through 1Department of Immunology, Weizmann Institute, Rehovot inference of variable and dynamic pathway activity rather than a fixed preprogrammed cell-type 76100, Israel. 2Department of Computer Science and Applied hierarchy. These data demonstrate single-cell RNA-seq as an effective tool for comprehensive Mathematics, Weizmann Institute, Rehovot 76100, Israel. 3 cellular decomposition of complex tissues. Department of Biological Regulation, Weizmann Institute, Rehovot 76100, Israel. *These authors contributed equally to this work. nderstanding the heterogeneous and defined cell types that are used to dissect cell †These authors contributed equally to this work. stochastic nature of multicellular tissues populations along developmental and functional ‡Corresponding author. E-mail: [email protected] Uis currently approached through a priori hierarchies (1–3). This methodology heavily relies (AT); [email protected] (IA)

776 14 FEBRUARY 2014 VOL 343 SCIENCE www.sciencemag.org Deepwater Horizon crude oil impacts the developing hearts of large predatory pelagic fish

John P. Incardonaa,1, Luke D. Gardnerb, Tiffany L. Linboa, Tanya L. Browna, Andrew J. Esbaughc, Edward M. Magerc, John D. Stieglitzc, Barbara L. Frencha, Jana S. Labeniaa, Cathy A. Laetza, Mark Tagala, Catherine A. Sloana, Abigail Elizurd, Daniel D. Benettic, Martin Grosellc, Barbara A. Blockb, and Nathaniel L. Scholza aEcotoxicology Program, Environmental Conservation Division, Northwest Fisheries Science Center, National Oceanic and Atmospheric Administration, Seattle, WA 98112; bHopkins Marine Station, Department of Biology, Stanford University, Pacific Grove, CA 93950; cDivision of Marine Biology and Fisheries, Rosenstiel School of Marine and Atmospheric Sciences, University of Miami, Miami, FL 33149-1098; and dGenecology Research Centre, Faculty of Science, Health, Education and Engineering, University of the Sunshine Coast, Maroochydore DC, QLD 4558, Australia

Edited by Karen A. Kidd, University of New Brunwsick, Saint John, BC, Canada, and accepted by the Editorial Board February 24, 2014 (received for review November 6, 2013) The Deepwater Horizon disaster released more than 636 million L respectively) (14, 15). The Atlantic bluefin tuna (Thunnus thynnus) of crude oil into the northern Gulf of Mexico. The spill oiled upper population from the Gulf of Mexico is currently at a historically surface water spawning habitats for many commercially and eco- low level (16), and was recently petitioned for listing under the logically important pelagic fish species. Consequently, the devel- US Endangered Species Act. For these and other pelagics, the oping spawn (embryos and larvae) of tunas, swordfish, and other extent of early-life stage loss from oiled spawning habitats is an large predators were potentially exposed to crude oil-derived important outstanding question for fisheries management and polycyclic aromatic hydrocarbons (PAHs). Fish embryos are gener- conservation. ally very sensitive to PAH-induced cardiotoxicity, and adverse The developing fish heart is known as a sensitive target organ changes in heart physiology and morphology can cause both acute for the toxic effects of crude oil-derived polycyclic aromatic and delayed mortality. Cardiac function is particularly important hydrocarbons (PAHs) (4). Of the multiple two- to six-ringed for fast-swimming pelagic predators with high aerobic demand. PAH families contained in crude oil, the most abundant three- Offspring for these species develop rapidly at relatively high tem- ringed compounds are sufficient to drive the cardiotoxicity of peratures, and their vulnerability to crude oil toxicity is unknown. petroleum-derived PAH mixtures. These compounds (fluorenes, We assessed the impacts of field-collected Deepwater Horizon (MC252) oil samples on embryos of three pelagic fish: bluefin tuna, dibenzothiophenes, and phenanthrenes) directly disrupt fish yellowfin tuna, and an amberjack. We show that environmentally cardiac function (17, 18), thereby interfering with the inter- realistic exposures (1–15 μg/L total PAH) cause specific dose- dependent processes of circulation and heart chamber forma- dependent defects in cardiac function in all three species, with tion. Exposure of fish embryos to PAH mixtures derived from circulatory disruption culminating in pericardial edema and other crude oil slows the heartbeat (bradycardia) and reduces con- secondary malformations. Each species displayed an irregular tractility (17, 19–21). The underlying mechanism was recently atrial arrhythmia following oil exposure, indicating a highly con- shown to be blockade of key potassium and calcium ion channels served response to oil toxicity. A considerable portion of Gulf wa- involved in cardiac excitation-contraction coupling (22). These ter samples collected during the spill had PAH concentrations exceeding toxicity thresholds observed here, indicating the poten- Significance tial for losses of pelagic fish larvae. Vulnerability assessments in other ocean habitats, including the Arctic, should focus on the The 2010 Deepwater Horizon (MC252) disaster in the northern developing heart of resident fish species as an exceptionally sen- Gulf of Mexico released more than 4 million barrels of crude oil. sitive and consistent indicator of crude oil impacts. Oil rose from the ocean floor to the surface where many large pelagic fish spawn. Here we describe the impacts of field- oil spill | damage assessment | heart development | embryology collected oil samples on the rapidly developing embryos of warm-water predators, including bluefin and yellowfin tunas he Deepwater Horizon disaster resulted in the release of more and an amberjack. For each species, environmentally relevant Tthan 4 million barrels (636 million L) of oil into the offshore MC252 oil exposures caused serious defects in heart develop- waters of the northern Gulf of Mexico between April 10 and July ment. Moreover, abnormalities in cardiac function were highly 14, 2010 (1). Although subsurface application of dispersant near consistent, indicating a broadly conserved developmental crude the wellhead resulted in retention of a considerable portion of oil oil cardiotoxicity. Losses of early life stages were therefore in the bathypelagic zone (2), oil also traveled to the upper sur- likely for Gulf populations of tunas, amberjack, swordfish, face waters where it formed a large and dynamic patchwork of billfish, and other large predators that spawned in oiled surface slicks (e.g., covering an estimated 17,725 km2 during May 2010) habitats. (3). In the decades following the last major US oil spill (the 1989 Exxon Valdez spill in Alaska), developing fish embryos have been Author contributions: J.P.I., L.D.G., A.J.E., M.G., and B.A.B. designed research; J.P.I., L.D.G., shown to be especially vulnerable to the toxicity of crude oil (4). T.L.L., T.L.B., A.J.E., E.M.M., J.D.S., B.L.F., J.S.L., C.A.L., M.T., and C.A.S. performed research; A.E. and D.D.B. contributed new reagents/analytic tools; J.P.I. and C.A.S. analyzed data; The northern Gulf provides critical spawning and rearing hab- J.P.I. and N.L.S. wrote the paper; J.P.I., D.D.B., M.G., B.A.B., and N.L.S. supervised the studies; itats for a range of commercially and ecologically important T.L.L., E.M.M., and J.D.S. handled logistics; and B.L.F. and J.S.L. provided logistical and pelagic fish species, and the timing of oil release into the eco- administrative support. system from the damaged Deepwater Horizon/MC252 well co- The authors declare no conflict of interest. incided with the temporal spawning window for bluefin and This article is a PNAS Direct Submission. K.A.K. is a guest editor invited by the Editorial yellowfin tunas, mahi mahi, king and Spanish mackerels, greater Board. and lesser amberjack, sailfish, blue marlin, and cobia (5–13). Freely available online through the PNAS open access option. Yellowfin tuna (Thunnus albacares) and greater amberjack 1To whom correspondence should be addressed. E-mail: [email protected]. (Seriola dumerili) contribute to important commercial fisheries This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. (48,960,000 pounds in 2010 and 4,348,000 pounds in 2004, 1073/pnas.1320950111/-/DCSupplemental.

E1510–E1518 | PNAS | Published online March 24, 2014 www.pnas.org/cgi/doi/10.1073/pnas.1320950111 collective effects of PAHs during embryonic and larval stages can used the only land-based captive broodstock available in the PNAS PLUS influence the structure and function of the adult fish heart in world for experiments. Similarly, we relied on a commercial ways that permanently reduce cardiac performance (23), po- broodstock of yellowtail amberjack and a research broodstock of tentially leading to delayed mortality. Consistent with this, mark- yellowfin tuna, the latter the only worldwide source of fertilized recapture studies on pink salmon following the Exxon Valdez spill embryos for this species. Embryos were exposed to high-energy found that transient and sublethal exposures to crude oil at very water-accommodated fractions (HEWAFs) (20) that generated low levels during embryogenesis reduced subsequent marine PAH concentrations and compositional profiles closely match- survival to adulthood by 40% (24, 25). Exposures to relatively ing water samples collected during active MC252 crude oil higher PAH concentrations cause embryonic heart failure and release phase. death soon after fish hatch into free-swimming larvae (19, 20, 23). These effects occur at a total PAH concentration range as Results low as 1–10 μg/L for more sensitive species (26, 27), levels as PAH Concentrations and Composition in HEWAF Preparations. Em- much as an order-of-magnitude lower than those measured in bryos were exposed to two different oil samples, one collected some samples collected both at depth and at the surface during from surface-skimming operations in the open Gulf of Mexico the Deepwater Horizon active spill phase (28, 29). (slick A) and one taken from the source pipe attached to the The above crude oil cardiotoxicity syndrome has been exten- damaged well head that was subsequently weathered artificially sively characterized in zebrafish embryos exposed to several geo- (Methods) to remove most of the volatile monoaromatic com- logically distinct oils (17, 21, 23, 30, 31), including the Mississippi pounds (artificially weathered source oil; AW source). The two Canyon 252 (MC252) crude oil released from the blown out samples were weathered to different degrees, as reflected pri- Deepwater Horizon wellhead (20, 32). Similar effects have been marily in the relative ratios of naphthalenes to tricyclic PAHs reported for temperate marine and anadromous species, such as and four-ringed chrysenes (Fig. S1). Relative to unweathered Pacific herring (19, 26, 27, 33) and pink salmon (34, 35), fol- fresh-source oil (Fig. S1A), the AW-source oil showed an in- lowing exposure to Alaska North Slope crude oil. Although crease in C2–C4 naphthalenes and C1–C4 phenanthrenes (Fig. zebrafish are a tropical freshwater model species, the embryos S1B). The slick A sample collected from the Gulf surface waters of herring and salmon assessed in the aftermath of the Exxon was considerably more weathered, with ∼90% depletion of all Valdez spill develop at cold temperatures (4–12 °C) over rela- naphthalenes per unit mass and ∼50% depletion of tricyclic tively long intervals (weeks to months). In contrast, pelagic spe- PAHs (Fig. S1C). However, the tricyclic PAHs represented 65% cies spawning in the warm surface waters of the northern Gulf of of the total ΣPAHs of slick A but only 34% of ΣPAHs in fresh Mexico (e.g., 24–29 °C) develop rapidly (24–48 h to hatch) (36, source oil. 37). The influence of development duration on PAH uptake and Each of the three species tested was exposed to independently toxicity, if any, is not well understood. The higher temperatures generated HEWAFs. PAHs in HEWAFs were measured at the characteristic of waters in the Gulf of Mexico may also influence onset of exposure only for tuna embryos, and at both the be- how the chemical composition of crude oil in surface habitat(s) ginning and end of exposure for amberjack embryos (Fig. 1 and changes over time (i.e., weathers). Processes that determine Fig. S2). HEWAFs were unfiltered, and measurements therefore weathering are generally accelerated at higher temperatures, po- included PAHs that were both dissolved and droplet-associated. tentially influencing the fraction of cardiotoxic PAHs that is bio- HEWAFs used in laboratory exposures for each species closely available for uptake by floating fish embryos in the mixed layer matched the PAH composition and concentrations measured in and thermocline regions. To address these information gaps, con- representative water samples collected from the upper pelagic trolled laboratory exposures are necessary to determine the sensi- zone during the active spill phase, with ΣPAH concentrations in tivity of Gulf species to Deepwater Horizon crude oil. the range of 3–14 μg/L at the highest tested oil load (Fig. 1; To assess potential early life-stage losses from large pelagic measured PAHs are provided in Table S1). HEWAFs used for predator populations that were actively spawning in habitats bluefin (Fig. 1A) and yellowfin tuna (Fig. 1B) exposures were affected by the Deepwater Horizon spill, we determined the similar to water samples collected in the Gulf of Mexico in late effects of field-collected MC252 oil samples on the development and early June 2010, respectively, and the amberjack exposure of embryos from representative warm water open-ocean fish closely matched a representative sample from late May 2010 species. Our approach extended earlier work in zebrafish, a lab- (Fig. 1C). Notably, C1-phenanthrenes are one of the most abun- oratory model species and Pacific herring, a marine nearshore dant three-ringed PAH classes in MC252 crude oil samples (Fig. spawner (19, 27, 38). Zebrafish and herring both produce large S1). Across all tests, the slick A HEWAF exposure to yellowfin demersal embryos that are relatively easy to manipulate (i.e., tuna had the lowest concentration of C1-phenanthrenes (0.36 μg/L collect, dechorionate, and image at consistent ontogenetic in- at the highest tested HEWAF dilution) (Fig. 1B). Among all of tervals). In contrast, Gulf pelagic species produce small, fragile, the water samples collected within and external to the spill zone buoyant embryos that develop relatively rapidly (on a timescale (between April 28 and July 30, 2010), 2.3% (61 of 2,647) con- of hours relative to days or weeks for zebrafish and herring, retained C1-phenanthrene concentrations above 0.36 μg/L. Levels spectively) and are not amenable to dechorionation. Moreover, in many samples were considerably higher, with mean (± SEM) the embryos hatch into buoyant larvae. Normally present in infinite- and median C1-phenanthrene concentrations of 1.76 ± 0.64 volume pelagic habitats, they are very sensitive to any form of (n =61)and0.55μg/L, respectively. For samples collected in physical contact, thereby complicating conventional embryology the pelagic zone of the northern Gulf of Mexico (above 200-m in small-volume laboratory cultures. Finally, access to embryos depth, an area bounded by latitude 29.6–27.25°, longitude –90.6 SCIENCES is difficult, with only a few land-based facilities capable of main- to –87.09; a rectangular area of roughly 96,000 km2 centered taining spawning broodstocks throughout the world. over the wellhead, extending offshore from 73 to 333 km), a ENVIRONMENTAL In the present study we overcome the aforementioned chal- larger fraction (4.6%, 25 of 548) had concentrations of C1- lenges for focal pelagic species that included yellowfin tuna, phenanthrenes ≥ 0.36 μg/L. Southern bluefin tuna (Thunnus maccoyii), and yellowtail am- PAH concentrations declined during each exposure. For ex- berjack (or kingfish, Seriola lalandi). The yellowfin tuna are the ample, in the amberjack tests the PAH profile at the start of the same species that spawn in the Gulf of Mexico, and the other exposure generally matched the profile for the whole AW-source two species are closely related congenerics to T. thynnus and oil, albeit with a slightly higher level of the more water-soluble S. dumerili, respectively. Controlled bluefin tuna spawning is ex- naphthalenes (Fig. S2A). By the end of incubation (Fig. S2B), the ceptionally difficult to achieve in a husbandry facility, and we naphthalenes, fluorenes, and dibenzothiophenes were mostly lost

Incardona et al. PNAS | Published online March 24, 2014 | E1511 3.0 1.0 development in unexposed control tuna and amberjack embryos. A Thomas Jefferson Cruise 3, 3.5 m depth 2.5 0.8 PAHs 8.0 ppb Heart development was very similar among bluefin and yellowfin 2.0

field lab tunas and amberjack, both in terms of morphogenetic sequence 0.6 1.5 and timing in relation to developmental stage. There were par- 0.4 1.0 allels and contrasts in comparison with the well-described de- 0.5 0.2 velopment of the zebrafish heart (39). In all three species, the

0.0 0.0 heart was first visible microscopically in late segmentation-stage 3.0 1.0 bluefin tuna exposure, AW-source HEWAF embryos (e.g., 24 somites) (Fig. S3B) as a cone that was sym- 2.5 0.8 PAHs 8.5 ppb metrical along the anterior-posterior axis with a ventrally located 2.0 0.6 base (Fig. S3D). At the time the first cardiomyocyte contractions 1.5 were observed, some left-right asymmetry was already apparent 0.4 1.0 as the cardiac cone began to rotate to position the atrial opening 0.5 0.2 on the left side (Fig. S3E). By the free-tail stage (several hours 0.0 0.0 later), the heart showed regular, rapid contractions of both the

3.0 1.0 atrium and ventricle, and was rotated nearly 90° so that the lu- B Brooks-McCall Cruise 6, 2 m depth men of the atrium was visible in left lateral views (amberjack, 2.5 0.8 PAHs 3.7 ppb 2.0 Fig. S3 F and G; yellowfin tuna, Fig. S3 J and K). After the 0.6 field lab 1.5 hatching stage, the heart elongated along the anterior-posterior

g/L) 0.4 µ 1.0 axis as the head extended and yolk absorbed, bringing the atrium posterior to the ventricle, but still slightly on the left side (am- 0.5 0.2 berjack, Fig. S3 H and I; yellowfin tuna, Fig. S3 L–O). Cardiac 0.0 3.0 1.0 development in bluefin tuna was virtually indistinguishable from yellowfin tuna exposure, Slick A HEWAF 2.5 0.8 PAHs 3.4 ppb yellowfin tuna, and was not photographically documented be- 2.0 cause of a limited availability of embryos and time constraints 0.6 1.5 for sample processing. PAH concentration (ppb, PAH 0.4 1.0

0.5 0.2 Gross Morphological Defects in Response to MC252 Oil Exposure.

0.0 0.0 Exposures were carried out at temperatures appropriate for broodstock maintenance and routine hatchery rearing for each 3.0 1.0 1.1 C Weatherbird II Cruise 1, surface grab species (bluefin tuna and amberjack, 25 °C; yellowfin tuna, 27 °C), 2.5 0.8 PAHs 13.2 ppb which generally resulted in high survival rates for unexposed 2.0 0.6 field lab (control) embryos. Bluefin tuna embryos had the lowest control 1.5 survival at 60 ± 5% (n = 4). The control survival rates for yel- 0.4 1.0 lowfin tuna and amberjack were 72 ± 9% (n = 6) and 93 ± 3% 0.5 0.2 (n = 4), respectively. 0.0 0.0 Exposure to each MC252 sample type (source- or surface- 3.0 1.0 amberjack exposure, AW-source HEWAF collected) produced a virtually identical suite of defects in the 2.5 0.8 PAHs 13.8 ppb pelagic fish embryos (Figs. 2–4) that were consistent with the 2.0 0.6 previously described effects of other crude oils in other fishes, 1.5 0.4 including neritic spawning marine species such as herring. Each 1.0 0.2 pelagic species showed comparable morphological responses, 0.5 marked by accumulation of pericardial edema, expanding to 0.0 0.0 yolk-sac edema in more severely affected larvae (representative F0 F1 F2 F3 P3 P4 P0 P1 P2 AY N3 N0 N1 N2 N4 D0 D1 D2 D4 D3 C1 C2 C3 C4 C0 BP AC AN IDP FL0 FP1 PY0 BZP BBF BKF BAA BEP BAP DBA PER

Fig. 1. HEWAF preparations produced environmentally relevant PAH exposures. PAH profiles in HEWAFs used in pelagic embryo exposures are compared with water samples collected during the active spill phase of the Deepwater Horizon/MC252 incident. Water samples shown are representative of 78 samples collected during May–July 2010 that had comparable PAH compositions to HEWAFs, 44 of which had ΣPAH levels that exceeded the highest concentrations tested in embryo exposures. These data were generated as part of the NRDA sampling program carried out by the Deepwater Horizon oil spill trustee team, and are publicly available (Methods). (A) Field sample from late June 2010 (Upper) and the artificially weathered source oil HEWAF used in a bluefin tuna test (Lower). (B) Field sample from early June 2010 (Upper) and the slick A HEWAF used in a yellowfin tuna test (Lower). (C) Field sample from late May 2010 (Upper) and the artificially weathered source oil HEWAF used in the amberjack test (Lower). Abbreviations for PAHs are listed in Table S1 and all values for individual PAHs are provided in Dataset S1. Focal compounds are naphthalenes (N0–N4, purple), fluorenes (F0–F3, blue), dibenzothiophenes (D0–D4, green), phenanthrenes (P0–P4, Fig. 2. Gross morphology of hatching stage larvae exposed to MC252 olive), and chrysenes (C0–C4, orange). HEWAFs oil during embryonic development. Embryos were exposed from shortly after fertilization to 12–16 h after hatching. Unexposed controls incubated in clean water are shown in A–C, and oil-exposed (highest-dose from the HEWAF, most likely because of volatilization, whereas ′– ′ ′ ∼ tested) in A C . Representative examples are shown for (A, A ) bluefin tuna 25% of the phenanthrenes remained. exposed to artificially weathered source oil (1 mg/L oil, 8.5 μg/L ΣPAH), (B, B′) yellowfin tuna exposed to slick A (2 mg/L oil, 3.4 μg/L ΣPAH), (C, C′)amberjack Normal Cardiac Development in Tunas and Amberjack. We charac- exposed to artificially weathered source oil (1 mg/L oil, 13.8 μg/L ΣPAH). (Scale terized the timing and general morphological sequence of heart bars, 1 mm.)

E1512 | www.pnas.org/cgi/doi/10.1073/pnas.1320950111 Incardona et al. examples shown in Fig. 3). Notably, all three species also showed PNAS PLUS similar types of extracardiac defects at the highest exposure concentration (8.5, 3.4, and 13.8 μg/L for bluefin, yellowfin, and amberjack, respectively) (Fig. 2). These morphological abnormalities included a lack of actinotrichia (fin ray precursors), reduction in the outgrowth of the finfolds or finfold blisters (Fig. 4), a dorsal or upward curvature of the body axis, and marked reduction in the growth of the eye (Fig. 3). Bluefin tuna showed the highest percentage of larvae with the entire suite of defects at a ΣPAH level of 8.5 μg/L (73%) (Table 1). In the pair of assays conducted with yellowfin tuna embryos, one assay included the addition of the broad-spectrum antibiotic oxytetracycline (10 mg/L) to inhibit potential bacterial growth. The presence or absence of oxytetracycline in the exposure water (see, for example, Fig. S5) did not influence the morphological effects of MC252 crude oil (see below). Similar effects on finfold outgrowth and a dorsal flexion of the body axis were also observed in yellowfin tuna, although the lower exposure level (ΣPAH 3.4 μg/L) resulted in a lower frequency of this suite of defects (15.6%) (Table 1).

MC252 Oil Caused Pericardial Edema at ΣPAH Concentrations in the Range of 1–15 μg/L, Irrespective of Weathering. Pericardial fluid accumulation (edema) caused by a failing circulatory system is a consistent and sensitive indicator of the petrogenic PAH cardiotoxicity syndrome. Both tuna species and amberjack showed edema and cardiac-specific abnormalities in response to both naturally and artificially weathered MC252 oil. In fish early life stages, accumulating fluid distorts the rounded yolk mass into Fig. 4. Caudal finfold defects in oil-exposed larvae. Higher magnification views are shown for bluefin tuna (A, A′), yellowfin tuna (B, B′), and amberjack (C, C′) from the same exposure shown in Fig. 1. A–C show the tails of controls and A′–C′ show the corresponding caudal fins of oil-exposed larvae. Arrowheads indicate the margins of the caudal finfolds in controls, and arrows indicate actinotrichia. The specific oil exposures (source and concentration) correspond to those in Figs. 2 and 3. (Scale bars, 0.5 mm.)

a shape that is either angular or flattened anteriorly. Distortion of the anterior end of the yolk mass was therefore used to quantify edema in tunas and amberjack. For all three species, the percentage of hatched larvae showing pericardial and yolk-sac edema in response to oil exposure was dose-dependent (Fig. 5A). The EC50 values (Table 2) were lower for the smaller (≤1 mm) bluefin tuna (0.8 μg/L ΣPAHs) and yellowfin tuna (2.3 μg/L) embryos relative to the larger (1.2 mm) amberjack embryos (12.4 μg/L). Based on the intersection between the upper 95% confidence limit of controls and the 95% confidence band for individual concentration-response nonlinear regressions (Fig. 5A), the thresholds for edema were in the range of 0.3–0.6 μg/L, 0.5– 1.3 μg/L, and 1.0–6.0 μg/L for bluefin tuna, yellowfin tuna, and amberjack, respectively (Table 2). Based on C1-phenanthrenes, ∼20% (112 of 548) of water samples collected in the northern Gulf pelagic zone had concentrations above these lower thresholds. The larger volume amberjack embryo exposures yielded similar results, with ΣPAH 13.5 μg/L causing significant edema (Fig. S4A). However, the single-replicate design for larger volumes (Methods) precluded corresponding EC50 determinations.

Exposure to MC252 Oil Caused Defects in Cardiac Function. The effects of MC252 crude oil on heart rate and rhythm were SCIENCES Fig. 3. Oil-induced circulatory failure and corresponding fluid accumulation assessed using digital video analyses. Heart rate was measured in ENVIRONMENTAL (edema) in pelagic fish embryos. Higher magnification views are shown for a single assay each for bluefin tuna and amberjack, and two in- (A, A′) bluefin tuna exposed to artificially weathered source oil (1 mg/L oil, dependent assays for yellowfin tuna. Each species showed a 8.5 μg/L ΣPAH); (B, B′) yellowfin tuna exposed to slick A (2 mg/L oil, 3.4 μg/L Σ ′ dose-dependent slowing of heart rate (Fig. 5B), or bradycardia, PAH), (C, C ) amberjack exposed to artificially weathered source oil (1 mg/L following exposure to either artificially weathered source oil oil, 13.8 μg/L ΣPAH). Controls are shown in A–C and oil-exposed larvae in A′–C′. Arrows indicate the anterior edges of the yolk masses, which are dis- (bluefin tuna and amberjack) or slick A (yellowfin tuna). The torted and forced posteriorly in the oil-exposed larvae; white arrowheads IC50 values for oil-induced bradycardia were somewhat higher indicate the location of the heart. Reduction of the eyes is indicated in the than the EC50 values for edema (Table 2), and showed a similar oil-exposed embryos with asterisks. (Scale bar, 0.2 mm.) relationship to egg size. The IC50s for reduction in heart rate

Incardona et al. PNAS | Published online March 24, 2014 | E1513 Table 1. Occurrence of extracardiac morphological defects in relation to egg size and ΣPAH Species Egg diameter (mm) Percent with tail and axial defects (%) ΣPAH (μg/L)*

Bluefin tuna 1 73.4 ± 8.6 8.5 (9.4) Yellowfin tuna 1 15.6 ± 6.0 3.4 (3.9) Amberjack 1.2 29 ± 4.6 13.8 (ND)

*Values in parentheses are sum of 50 PAHs listed in Table S1. were ΣPAH 7.7 and 6.1 μg/L for the smaller bluefin and yel- 3.1% of control animals, but oil exposure from fertilization lowfin tuna embryos, respectively, and 18.2 μg/L for the larger through the hatching stage lengthened the systolic phase and in- amberjack embryos. Based on the intersection between the lower creased the frequency of larvae with systole longer than diastole in 95% confidence limit of controls and the 95% confidence band a concentration-dependent manner (Fig. 5C). for individual concentration-response nonlinear regressions (Fig. Similarly, in hatching-stage amberjack larvae, 96.7 ± 3.3% of 5B), the exposure thresholds for bradycardia were in the range of unexposed controls had heartbeats with a systolic phase that was 4.0–8.5 μg/L and 1.0–2.6 μg/L for bluefin and yellowfin tunas, shorter on average (115 ± 8 ms) than the corresponding diastolic respectively, and 2.2–6.5 μg/L for amberjack (Table 2). The lower phase (200 ± 0 ms). Amberjack larvae exposed to the highest thresholds in yellowfin tuna and amberjack are most likely be- MC252 treatment (ΣPAH 13.8 μg/L) had systolic and diastolic cause of higher power associated with a larger sample size and phases lasting 350 ± 69 and 158 ± 16 ms, respectively, and the lower variability in control heart rates. Bradycardia was also frequency of larvae with prolonged systole also increased in observed in amberjack embryos exposed in larger volumes, at a concentration-dependent manner (Fig. 5C). The calculated a ΣPAH concentration as low as 1.2 μg/L (Fig. S4B). Addition of EC50s for prolongation of systole in yellowfin tuna and amber- oxytetracycline to the exposure solution did not influence func- jack were ΣPAH 2.6 and 8.6 μg/L, respectively. Based on the tional cardiotoxicity in yellowfin tuna embryos (i.e., the in- upper 95% confidence limits of controls, exposure thresholds for hibitory effects of increasing ΣPAH concentrations on yellowfin arrhythmia were 0.026–0.13 μg/L and 0.27–1.0 μg/L for yellowfin heart rates were comparable in the presence or absence of the tuna and amberjack, respectively (Fig. 5C). Based on C1-phen- antibiotic) (Fig. S5C). anthrenes, 30% (165 of 548) of water samples collected in the In addition to a slowing heartbeat, oil exposure resulted in an northern Gulf pelagic zone had concentrations above these irregular arrhythmia observable in all three species (Fig. 5 C and lower thresholds. D and Movie S1). These dose-dependent effects were quantified The slower basal heart rate in amberjack larvae also allowed for yellowfin tuna and amberjack embryos. For both species, the an assessment of beat-to-beat variability, or heart rate regularity durations of atrial systole (contraction) and atrial diastole (re- (19). The average heart rate of unexposed amberjack controls was laxation) were assessed by digital video analysis. In hatched ∼180 beats per minute, with a beat-to-beat variability (Methods)of yellowfin tuna larvae, systole was shorter than diastole in 91.4 ± only 16 ± 2ms(Fig.5D). Heart rhythm irregularities increased

AB100 bluefin tuna 300 80 yellowfin tuna 60 amberjack 200 40 bluefin tuna 20 100 yellowfin tuna

percent with edema 0

heart rate (beats/min) amberjack -20 0 0.01 0.1 1 10 0.01 0.1 1 10 PAH (µg/L) PAH (µg/L) C D 100 yellowfin tuna 160 amberjack 80 amberjack 120 60 80 40 40 20 percent with arrhythmia

0 interbeat variability (msec) 0 0.01 0.1 1 10 control 4.5 µg/L 13.8 µg/L 13.8 µg/L no edema no edema edema PAH (µg/L)

Fig. 5. Dose-dependent heart failure and defects in cardiac function following MC252 crude oil exposure. Dose–response curves are shown for (A) occurrence of heart failure leading to edema and (B) bradycardia in southern bluefin tuna, yellowfin tuna, and amberjack. (C) Occurrence of arrhythmia defined as prolongation of atrial systole in yellowfin tuna and amberjack. (D) Quantification of the regularity of the heart beat in amberjack by beat-to-beat variability. Data are mean ± SEM. Red horizontal lines (A–C) indicate estimations for PAH exposure thresholds based on the overlap of the 95% confidence limits of controls and the 95% confidence band of the nonlinear regression. Asterisks in D indicate groups significantly different from control (α < 0.05). Because only control and highest concentrations were measured for southern bluefin tuna, intermediate PAH concentrations shown in A and B are predictions based on dilution.

E1514 | www.pnas.org/cgi/doi/10.1073/pnas.1320950111 Incardona et al. PNAS PLUS Table 2. Calculated EC50 and IC50 values for cardiotoxic endpoints Edema Heart rate Arrhythmia

2 2 2 Species EC50 95% CI R Threshold IC50 95% CI R Threshold EC50 95% CI R Threshold

Bluefin tuna 0.8 (0.9) 0.6–1.1 0.9 0.3–0.6 7.7 (8.5) 5.0–11.9 0.7 4.0–8.5 NC NC Yellowfin tuna 2.3 (2.5) 2.0–2.8 0.8 0.5–1.3 6.1 (7.5) 3.3–11.2 0.5 1.0–2.6 2.6 (2.9) 2.1–3.4 0.9 0.026–0.13 Amberjack 12.4 (ND) 10.9–14.1 0.9 1.0–6.0 18.2 12.7–26.0 0.8 2.2–6.5 8.6 7.0–10.6 0.9 0.27–1.0

Values are micrograms per liter ΣPAH, calculated using nonlinear regression models as described in Methods; first ΣPAH value is sum of 40 analytes, values in parentheses are sum of 50 analytes (Table S1). CI, confidence interval; NC, not able to calculate; ND, not determined. with both crude oil-exposure concentration (beginning at 4.5 μg/L arrhythmia, with thresholds near or below 1-μg/L ΣPAH for ΣPAH), as well as the severity of the cardiotoxicity phenotype. both yellowfin tuna and the larger amberjack embryos. Larvae Heartbeat irregularity was significant in larvae exposed to 13.8 μg/L with arrhythmia did not have gross pericardial edema. However, ΣPAH, and larvae with visible edema were more severely affected at lower exposure concentrations, the pathophysiological pro- (Fig. 5D). cesses leading to arrhythmia are likely to also disrupt cardiac morphogenesis [e.g., calcium ion homeostasis (40)], potentially Discussion leading to permanent impacts on cardiac form and function (23) The Deepwater Horizon disaster is the largest oil spill yet to occur and delayed mortality (25). Our analysis shows that a consider- in the pelagic zone of an oceanic ecosystem. Crude oil released able portion of upper pelagic water samples (nearly one-third) at depth eventually rose to warm mixed layer and surface waters collected during the active spill phase had PAH concentrations of the northern Gulf of Mexico during the spawning windows above thresholds for arrhythmias in the developing hearts of for bluefin tuna and many other large predator species (e.g., tunas and amberjack. However, this approximation may be low mackerel, amberjack, sailfish, marlin, mahi mahi, and other tunas). given the limited sampling in areas with high surface oiling be- These pelagic fish all produce fertilized eggs that float in the cause of worker health and safety concerns and other vessel upper layers of the water column. It is therefore highly likely access restrictions. that the early life stages of many northern Gulf pelagics were In conjunction with previous studies, the findings here dem- exposed to crude oil. Despite differences in temperature and onstrate that the response of teleost embryos to petroleum- species-specific developmental rate, we show here that warm- derivedPAHsinboththelaboratoryandthefieldishighly water pelagic embryos are similarly sensitive to crude oil car- conserved among species tested thus far. Tunas and amberjacks diotoxicity, with an injury phenotype that aligns closely with develop at higher water temperatures, and yet they display heart previous reports for temperate and boreal species. Additionally, failure and other abnormalities that are remarkably similar to the higher exposure concentrations produced extracardiac de- those previously reported for species, such as herring and fects very similar to those observed in other species (zebrafish, salmon, which develop at very cold temperatures. Pacific herring herring) with several crude oils, including MC252 (17, 20, 21). In were a focal species for natural resource injury assessments fol- particular, finfold defects and reduced fin growth appear to be lowing the Exxon Valdez (41) (Alaskan waters) and Cosco Busan specific effects of crude oil exposure, and not necessarily because oil spills (38) (California Current). Herring early life stages ex- of developmental delay in embryos with severe edema. For exposed to both oil types, in both the field and the laboratory, ample, reduced finfold outgrowth was observed in MC252- developed cardiotoxic defects in the form of bradycardia (38) exposed zebrafish embryos that only had mild to moderate and pericardial edema (38, 41), with strikingly similar results to edema (20). those reported here for subtropical spawning pelagic fish. The For all three species assessed here, exposures to ΣPAHs remarkably consistent morphological and physiological respon- below 15 μg/L caused cardiotoxicity in the form of heart failure ses to oil across diverse fish species indicate that the core and corresponding edema. Our exposure protocols produced mechanisms of PAH-induced cardiotoxicity are conserved. Namely, PAH concentrations and mixture compositions that overlapped the cardiotoxic injury stemming from embryonic exposure to with measured PAHs in northern Gulf surface waters during the crude oil observed in the scombrid and carangid species in this active spill phase, with field detections in many cases at levels study is essentially identical to the response of a boreal clupeid that were considerably higher than those used here in experi- (herring) (19), representing families that are separated by roughly ments [e.g., up to 84 μg/L ΣPAH (28, 29)]. Despite relative dif- 100 million y of evolution (42). Our findings thus have implica- ferences in weathering, each of the MC252 crude oil samples tions beyond the upper pelagic zone of the Gulf of Mexico, and produced concentration-dependent, stereotypical crude oil car- are likely to be indicative of sensitivity to oil over a wider range of diotoxicity to the embryos of tunas and amberjack. The spatial fish species spawning in other habitats contaminated by MC252 extent of injury to fish early life stages may therefore have been crude oil. large, in response to both fresher oil proximal to the wellhead Previous studies have commonly used flow-through columns and more weathered oil further afield. packed with oiled gravel to generate PAH exposures (19, 26, 27). As anticipated from previous studies, pericardial edema was In zebrafish embryos, the effects of two geologically distinct oils SCIENCES a sensitive indicator of reduced cardiac function or heart failure (Alaskan and Iranian) were indistinguishable irrespective of in the tunas and amberjack. Oil exposures yielding ΣPAH whether PAHs originated from oiled gravel columns or HEWAFs ENVIRONMENTAL concentrations in the range of 2–3 μg/L caused gross edema in (17, 21). The similarity of the injury phenotype observed here in ≥50% of exposed tuna embryos. Yolk-sac larvae with severe pelagic embryos and larvae, in the presence or absence of a broad- edema would not be expected to survive into the feeding stage spectrum antibiotic, provides further evidence that crude oil car- (20, 23), indicating that the early life stages of pelagic fish that diotoxicity to fish embryos is both independent of exposure spawned in or near Gulf of Mexico surface waters containing methodology and consistent with direct actions of individual PAH ΣPAHs at levels above a few micrograms per liter may have compounds on the developing fish heart (17, 18, 43). been lost from wild populations. Importantly, the most sensitive Subtle variations in heart failure and edema among pelagic physiological measurement indicative of cardiotoxicity was atrial species are likely the result of a combination of factors. For

Incardona et al. PNAS | Published online March 24, 2014 | E1515 example, PAHs would be expected to accumulate more rapidly allowed to separate for 1 h, but were not filtered, and thus contained in smaller eggs with higher surface-to-volume ratios, yielding whole-particulate oil in addition to dissolved PAHs. higher effective tissue concentrations and hence more severe toxicity at the same exposure concentration. Although tissue Embryo Exposures. Experiments with each species were approved by animal PAH concentrations could not be measured in these pelagic care and use committees at Stanford University (protocol no. 14706) and the University of Miami (protocol no. 12–011 RENEWAL 09–001). Fertilized species, this has been shown for other species with different egg eggs were obtained from yellowfin tuna (Achotines Laboratory, Peninsula sizes (e.g., herring and salmon) (26, 34). Species may also differ Azuero, Panamá), Southern bluefin tuna (Cleanseas Tuna, Arno Bay, South in the degree of metabolic capacity at a given time point within Australia), and yellowtail amberjack (Cleanseas Tuna, Arno Bay, South Aus- prehatching development. Relative to tunas, zebrafish have a tralia). For yellowfin tuna, eggs were removed from an egg collector at- larger embryo (>1 mm) and hatch at a more advanced stage of tached to the broodstock tank and placed in a 20-L bucket. Water was organogenesis. Zebrafish exposed to Deepwater Horizon oil pri- saturated with dissolved oxygen (100% D.O.) using an airstone and non- marily show reduced ventricular contractility rather than bra- viable eggs were allowed to settle for 15 min, after which floating eggs were dycardia and atrial arrhythmia (20). Also important is the scooped off the surface using a soft mesh net and placed into another 20-L μ ∼ – interaction of individual components of complex PAH mixtures bucket of 1- m filtered/UV-sterilized seawater at a density of 300 500 eggs – per liter. After gentle flushing with clean 1-μm filtered/UV sterilized sea- with multiple cellular targets in the fish heart (17 19, 21, 43), water for 10–15 min, formalin [37% (vol/vol) formaldehyde solution] was ap- each of which may vary across species with different cardiovas- plied to the eggs at a final concentration of 100 ppm for 1 h as a prophylactic cular physiology. Crude oil HEWAFs similar to those used here treatment. Aeration was supplied at a low rate to maintain dissolved oxygen were shown to block key voltage-gated potassium and calcium levels at or above saturation levels (6.1–6.5 mg/L at 27–29 °C) throughout ion channels controlling excitation-contraction coupling in fish formalin treatment. Eggs were then rinsed using 1-μm filtered/UV-sterilized cardiomyocytes (22), and the precise impacts on cardiac function seawater for at least five full volume changes before use in exposure assays. may depend on the relative levels of these targets in different Temperature was maintained within 1 °C of the spawning tank temperature species. In addition, results from studies on alkylated phenan- throughout the collection and formalin treatment periods. For bluefin tuna threnes and more weathered crude oil mixtures enriched with and amberjack, fertilized eggs were collected from the spawning captive broodstock by hatchery operators and treated to reduce microbes using higher levels of alkylated tricyclic PAHs suggest mechanisms of nonformaldehyde proprietary methods per standard practice at the facility cardiotoxicity that are downstream of aryl hydrocarbon receptor for routine larval production. For yellowfin tuna exposures, oil exposures (AHR) activation in cardiac muscle cells (21, 43). The AHR is and embryological analyses were carried out at the Achotines Laboratory the ligand-activated transcription factor that induces the up- adjacent to the captive broodstock operation. For bluefin tuna and amberjack regulation of protective PAH-metabolizing enzymes (e.g., cyto- exposures, bagged oxygenated eggs were transported in seawater in chrome P4501A, CYP1A). Inappropriate AHR activation in a cooler ∼1 h to a laboratory at the Lincoln Marine Science Center (Port heart muscle cells also underlies the cardiotoxicity of some Lincoln, South Australia). All embryos were staged and sorted under higher molecular weight PAHs, as well as dioxins and poly- Olympus and Nikon stereomicroscopes. For yellowfin tuna, exposures began chlorinated biphenyls (43–47). Species-specific developmental during the cleavage stage. For bluefin tuna and amberjack, embryos un- variation in ion channel expression and other targets for PAHs in dergoing normal gastrulation were selected and exposure was started at 50% epiboly. Two exposures were carried out sequentially with yellowfin cardiac tissues may underlie the observed subtle differences in tuna, one with an addition of 10 ppm oxytetracycline to the water used for MC252 oil-induced functional defects in the developing hearts of processing eggs and HEWAF dilutions for exposure. tunas, amberjack, and zebrafish. Exposures were carried out in 1-L beakers with densities of 20 embryos per In closing, determining early life-stage cardiotoxic injury for liter (bluefin and yellowfin tuna) or 80 embryos per liter (amberjack). Amberjack open-ocean spawners during a large spill is practically impossible embryos were also exposed in 10-L plastic buckets with an exposure vol- given the logistical constraints of sampling live but fragile yolk- ume of 7 L and the same embryo density. Because of a limited availability of sac larvae and identifying individual species with uncertain de- embryos, bluefin tuna was restricted to a single assay. Assays with beakers velopmental staging in mixed populations of zooplankton. used either four replicates (bluefin tuna, amberjack) or six replicates (yel- Moreover, land-based facilities capable of maintaining captive lowfin tuna) per treatment (unexposed control and four concentrations of broodstocks of large pelagic predators are rare, making experi- HEWAF). Amberjack exposure in buckets used a single replicate per treatment (control and four exposure concentrations) with an embryo density of 79 ± 4 ments challenging. For example, the capacity to spawn bluefin per liter (mean ± SEM; n = 5). At source hatcheries, each pelagic species was tunas in captivity in controlled land-based facilities has only been typically incubated at densities of ∼100 embryos per liter in large tanks with accomplished recently, at a single location. At this time, there- constant upwelling agitation, allowing very high hatch rates of morpho- fore, the studies described here constitute the only feasible logically normal larvae. Initial studies were carried out with bluefin tuna means for providing an environmentally realistic assessment of embryos during a narrow window of availability, and used a static in- the potential impacts to the offspring of bluefin and yellowfin cubation protocol. Bluefin tuna embryos could only be incubated at low tunas that spawned in or near the Deepwater Horizon spill zone. density (20 per liter), with hatch rates lower than that achieved in large Finally, these studies demonstrate the feasibility of collecting volume culture. In addition, it was determined that newly hatched larvae quantitative information for species that are logistically very could not be viably mounted and imaged without anesthesia, and a protocol was developed that relied on imaging of anesthetized larvae 6–10 h post- difficult to handle and study. A similar approach can be used for hatch. Subsequent experiments with amberjack and yellowfin tuna embryos ecologically and commercially important fishes in other ocean determined that hatch rates were markedly improved with gentle agitation habitats where oil exploration and extraction is ongoing or in- of beakers that kept hatching-stage embryos away from the air–water increasing, such as the Arctic. terface and beaker bottoms. In amberjack assays, beakers were agitated on a modified horizontal shaking water bath and buckets were agitated with Methods magnetic stir bars. Hatching rates could not be precisely quantified for buckets Oil Samples and WAF Preparation. All oil samples were collected under chain because of the large numbers of embryos (> 500 per bucket), but generally of custody during Deepwater Horizon response efforts. Samples used in appeared similar to beakers. Despite agitation on an orbital shaker, yellowfin toxicity assays included MC252 source oil that was artificially weathered by tuna hatch rates still declined at densities above 20 per liter. heating with gentle mixing to 90–105 °C until the mass was reduced by Dilutions of a 1:1,000 stock WAF were distributed to the replicate beakers 33–38% (sample 072610-W-A), and slick A (sample CTC02404-02), a spill- for each concentration tested, and embryos added using glass transfer pipets. zone sample collected July 29, 2010 from a barge holding mixed oil off- Depending on the timing of incubation to hatch for each species, exposures loaded from a number of different skimmers. HEWAFs were prepared using were carried out without WAF renewal up to several hours after hatch. All a commercial stainless steel blender as described in detail elsewhere (20). exposures for the data shown were carried out in temperature-controlled HEWAF stocks (1:1,000 dilutions of oil) and further dilutions for exposure rooms (bluefin tuna and amberjack, 25 °C; yellowfin tuna, 27 °C). Water incubations were prepared in high-quality seawater used for culture pur- quality was monitored daily, and included measurements of pH, dissolved poses at each facility supplying fertilized eggs. All HEWAF preparations were oxygen, salinity, ammonia, and temperature.

E1516 | www.pnas.org/cgi/doi/10.1073/pnas.1320950111 Incardona et al. Image Collection. For all data collection, room temperatures were held at the treatment group to assess “tank” effect and to avoid pseudoreplication. If PNAS PLUS same temperature as exposures. Larvae were anesthetized with clove oil ANOVA was significant for effect of treatment (P < 0.05), means were before capture and imaging. The dilution of clove oil required to immobilize compared between controls and treatment groups (exposure concen- larvae was titrated for each species. Clove oil was diluted 1:10,000 into ex- trations) using Dunnett’stest(α = 0.05). posure beakers, with 10 min allowed for anesthetic effect before initiation of imaging. Only one beaker at a time was imaged, so that all replicates had Analytical Chemistry. PAH concentrations were determined by gas chroma- equal exposure time in clove oil. For all imaging (with or without anesthetic), tography/mass spectrometry (GC/MS) by either Columbia Analytical Services two to three larvae were captured at a time and transferred to a Petri dish, (CAS; bluefin and yellowfin tuna) or the Northwest Fisheries Science Center where they were individually mounted on top of 2% methylcellulose in (NWFSC; amberjack). The 250-mL water samples were stabilized with either seawater. This process ensured rapid imaging of specimens while avoiding hydrochloric acid (CAS samples) or dichloromethane (NWFSC samples), stored potential temperature elevation on the microscope stage. Two stereoscope at 4 °C until shipment, and shipped to each analytical laboratory by expe- stations allowed sequential processing of two replicates at a time, and larvae dited courier. Samples were then extracted with dichloromethane and were imaged continuously until all replicates were completed. Replicates processed for GC/MS. Details of sample extraction, cleanup and GC/MS were processed in random order, except control replicates that were evenly analysis, including limits of detection, are provided elsewhere (38). PAH spaced throughout the entire time of processing, which ranged from 5 to 8 h. concentrations were determined for each dilution of the dose–response For amberjack embryos exposed in buckets, 10 hatched larvae were randomly series. For the bluefin tuna exposure, PAHs were measured in only the × captured from each treatment and imaged within 1 h. Images (640 480 pixels) highest concentration and control samples because of sample loss during were collected using FireI-400 industrial digital video cameras (Unibrain) shipping between Australia and the United States. Measured PAHs are mounted on Nikon SMZ800 stereomicroscopes, using MacBook laptops and provided in Table S1. For consistency with previously published work, BTV Carbon Pro software. Images were calibrated using a stage micrometer. a conventional list of 40 analytes was used for PAH composition figures and calculations of ΣPAH. A larger list of 50 analytes was used to provide sup- Image Analysis and Statistics. For scoring the presence or absence of edema, plemental EC50/IC50 values comparable to other ongoing Deepwater Horizon still frames and videos were assessed for the shape of the yolk mass. Larvae Natural Resource Damage Assessment (NRDA) studies. Individual PAH mea- were scored as normal if the anterior portion of the yolk sac was smooth and surements for all tests are provided in Dataset S1. HEWAF PAH data were rounded with a bullet-shaped tip and if there were no obvious indentations compared with field sample PAH data generated within the larger NRDA on the yolk sac because of pressure from fluid buildup in the pericardial area. field sampling program. The NRDA sampling plans, protocols, and raw data Edema was scored positive if the anterior portion of the yolk sac was concave are publicly available through links at www.gulfspillrestoration.noaa.gov/ or pushed to a sharp point, or if indentations indicated by dark, angular lines oil-spill/gulf-spill-data. For comparison with PAH levels in field-collected were seen pushing on the yolk sac because of pressure from fluid buildup in water samples, the source file for all validated Gulf water sample PAH values the pericardial area. For all species, there was a range of normal yolk-sac was obtained at http://54.243.205.138/gulfspillrestoration/qmmatrix/Water_ shapes in control fish. Yolk sacs that did not have a perfect rounded, bullet chem_export.zip. The file Water_chem_export.csv was filtered in Microsoft shape (e.g., slightly blunted or semipointed) were still considered within the Excel using columns “matrix” (water), “chemname” (C1-phenanthrenes/ range of normal. Because amberjack embryos exposed in buckets were all anthracenes), “sampdate” (20100428–20100730), “upperdepth” (greater imaged within a narrow time window (1 h) and were thus at the same de- than or equal to 0 m), “lower depth” (less than or equal to 200 m), “latitude” velopmental stage, pericardial areas were measured in lateral images as (≥27.25, ≤29.6), and “longitude” (≥−90.6, ≤−87.09). These filtering criteria a quantitative assessment of edema as described elsewhere (20). resulted in a list of 548 pelagic water samples from a total of 2,647. The Cardiac function was assessed in ≥10-s digital video clips of all viable map locations of water samples can be found at http://gomex.erma.noaa. embryos collected from each replicate exposure. Heart rate was determined gov/erma.html#x=-88.25810&y=27.03211&z=6&layers=19129. For compari- by simply counting the total number of heart beats in a given video clip son of overall PAH composition between HEWAFs and field samples (Fig. 1), played back at half speed. Arrhythmia was assessed by relating the onset of a subset of 77 water samples was selected that showed weathering as a contraction and relaxation for cardiac chambers to individual video frames, relative loss of naphthalenes. The ΣPAH concentrations for these samples and counting the total number of frames for each phase. The actual mean ranged from 2.6–354.4 μg/L, with a mean (±SD) of 22.3 ± 42.7 μg/L and a duration of systole and diastole was determined for amberjack from four median value of 13.2 μg/L. Water samples from the field that showed randomly selected videos for control and high-dose treatment groups. Du- similar PAH concentration and composition (i.e., weathering state) to HEWAF ration of systole and diastole was not determined for yellowfin tuna preps (represented in Fig. 1) were thus in the low end of the range. The data because of higher heart rates. For amberjack, heart rate (beat-to-beat) subset used for this comparison is publically available as a package at variability was determined by the number of video frames between the http://54.243.205.138/gulfspillrestoration/PNAS/NOAA_WAF_PNAS_2014.zip. initiation of contractions for an entire video clip, and calculated as described elsewhere (19), measured in 10 randomly selected videos from each replicate (e.g., Movie S1). ACKNOWLEDGMENTS. The authors thank Adam Miller, Morten Deichmann, and Craig Foster of Cleanseas Tuna for providing bluefin tuna and amberjack embryos, and advice on culture conditions; Erin Bubner and Bob Delaine for Statistical Analysis. Parametric statistical analyses were performed using JMP facilities access and logistical support at the Lincoln Marine Science Center; (versions 6.0.2 and 8.0.1 for Macintosh; SAS Institute), and dose–response Tor Linbo and Robbie Schallert for assistance with bluefin tuna and amberjack relationships analyzed by nonlinear models using Prism 6.0b for Macintosh embryo exposures; the Inter-American Tropical Tuna Commission members (GraphPad Software). For nonlinear regressions, ΣPAH data were log- Guillermo Campeán, Richard Deriso, Daniel Margulies, Vernon Scholey, and transformed and a normalized-response model was used for occurrence data the staff at the Achotines Laboratory for providing access to the yellowfin (percent with edema or arrhythmia); heart rate data were not normalized. tuna broodstock and laboratory facilities; Bernadita Anulacion, Daryle Boyd, and Ron Pearce for assistance with polycyclic aromatic hydrocarbon analyses; For parametric analyses, occurrence data (arrhythmia) (Fig. 5D) were ana- and National Oceanic and Atmospheric Administration National Ocean Ser- lyzed by one-way ANOVA. If ANOVA was significant for effect of treatment vice staff and contractors for reviewing the experimental design and a draft < (P 0.05), means were compared between controls and treatment groups of the manuscript. M.G. is a Maytag professor of ichthyology. This work was (exposure concentrations) using Dunnett’s post hoc test (α = 0.05). Heart-rate funded as a contributing study to the Deepwater Horizon/MC252 Incident data (Fig. S5) were analyzed by one-way ANOVA with replicate nested in Natural Resource Damage Assessment.

1. Crone TJ, Tolstoy M (2010) Magnitude of the 2010 Gulf of Mexico oil leak. Science 6. Arocha F, Lee DW, Marcano LA, Marcano JS (2001) Update information on the SCIENCES 330(6004):634.

spawning of yellowfin tuna, Thunnus albacares, in the western central Atlantic. Col ENVIRONMENTAL 2. Kujawinski EB, et al. (2011) Fate of dispersants associated with the Deepwater Horizon Vol Sci Pap ICCAT 52(1):167–176. – oil spill. Environ Sci Technol 45(4):1298 1306. 7. Grimes CB, Finucane JH, Collins AL, DeVries DA (1990) Young king mackerel, Scom- 3. Labson VF, et al. (2010) Estimated Lower Bound for Leak Rates from the Deepwater beromorus cavalla, in the Gulf of Mexico, a summary of the distribution and occur- Horizon Spill—Interim Report to the Flow Rate Technical Group from the Mass rence of larvae and juveniles, and spawning dates for Mexican juveniles. Bull Mar Sci Balance Team, U.S. Geological Survey Open–File Report (US Geological Survey, 46(3):640–654. Reston, VA), p 4. 8. McEachran JD, Finucane JH, Hall LS (1980) Distribution, seasonality and abundance of 4. Incardona JP, Collier TK, Scholz NL (2011) Oil spills and fish health: Exposing the heart of the matter. J Expo Sci Environ Epidemiol 21(1):3–4. king and Spanish mackerel larvae in the northwestern Gulf of Mexico (Pisces: Scom- 5. Lang KL, Grimes CB, Shaw RF (1994) Variations in the age and growth of yellowfin bridae). Northeast Gulf Science 4:1–16. tuna larvae, Thunnus albacares, collected about the Mississippi River plume. Environ 9. Block BA, et al. (2005) Electronic tagging and population structure of Atlantic bluefin Biol Fishes 39(3):259–270. tuna. Nature 434(7037):1121–1127.

Incardona et al. PNAS | Published online March 24, 2014 | E1517 10. Teo S, et al. (2007) Annual migrations, diving behavior, and thermal biology of At- 29. Wade TL, et al. (2011) Analyses of water samples from the Deepwater Horizon oil lantic bluefin tuna, Thunnus thynnus, on their Gulf of Mexico breeding grounds. Mar spill: Documentation of the subsurface plume. Monitoring and Modeling the Deep- Biol 151:1–18. water Horizon Oil Spill: A Record-Breaking Enterprise, eds Liu Y, MacFadyen A, Ji Z-G, 11. Rooker JR, et al. (2012) Distribution and habitat associations of billfish and swordfish Weisberg RH, Geophys Monogr Ser (AGU, Washington, DC), 195:77–82. larvae across mesoscale features in the Gulf of Mexico. PLoS ONE 7(4):e34180. 30. Carls MG, et al. (2008) Fish embryos are damaged by dissolved PAHs, not oil particles. 12. Murie DJ, Parkyn DC (2008) Age, growth and sex maturity of greater amberjack Aquat Toxicol 88(2):121–127. (Seriola dumerili) in the Gulf of Mexico. Marine Fisheries Research Initiative (MARFIN) 31. Hatlen K, et al. (2010) Natural sunlight and residual fuel oils are an acutely lethal Program, Final Report NA05NMF4331071 (National Marine Fisheries Service, combination for fish embryos. Aquat Toxicol 99(1):56–64. St. Petersburg, FL). 32. de Soysa TY, et al. (2012) Macondo crude oil from the Deepwater Horizon oil spill 13. Thompson BA, Wilson CA, Render JH, Beasley M, Cauthron C (1991) Age, growth, and disrupts specific developmental processes during zebrafish embryogenesis. BMC Biol reproductive biology of greater amberjack and cobia from Louisiana waters. Marine 10:40. Fisheries Research Initiative (MARFIN) Program, Final Report NA90AAH-MF722 33. Barron MG, Carls MG, Short JW, Rice SD (2003) Photoenhanced toxicity of aqueous (National Marine Fisheries Service, St. Petersburg, FL). phase and chemically dispersed weathered Alaska North Slope crude oil to Pacific – 14. National Marine Fisheries Service (2010) Available at www.st.nmfs.noaa.gov/sisPortal/ herring eggs and larvae. Environ Toxicol Chem 22(3):650 660. sisTimeSeriesPdf.jsp?tsPdfIndex=1&entityName=Greater%20amberjack%20%20Gulf% 34. Heintz RA, Short JW, Rice SD (1999) Sensitivity of fish embryos to weathered crude oil: 20of%20Mexico&asmtId=2758&asmtYear=2010&asmtMonth=December. Accessed Part II. Increased mortality of pink salmon (Oncorhynchus gorbuscha) embryos in- March 10, 2014. cubating downstream from weathered Exxon Valdez crude oil. Environ Toxicol Chem – 15. National Marine Fisheries Service (2011) Available at www.st.nmfs.noaa.gov/sisPortal/ 18(3):494 503. sisTimeSeriesPdf.jsp?tsPdfIndex=1&entityName=Yellowfin%20tuna%20%20Western 35. Marty GD, et al. (1997) Ascites, premature emergence, increased gonadal cell apoptosis, and cytochrome P4501A induction in pink salmon larvae continuously exposed %20Atlantic&asmtId=3143&asmtYear=2011&asmtMonth=October. Accessed March to oil-contaminated gravel during development. Can J Zool 75(6):989–1007. 10, 2014. 36. Margulies D, et al. (2007) Spawning and early development of captive yellowfin tuna 16. Taylor NG, McAllister MK, Lawson GL, Carruthers T, Block BA (2011) Atlantic bluefin (Thunnus albacares). Fish Bull 105:249–265. tuna: A novel multistock spatial model for assessing population biomass. PLoS ONE 37. Ditty JG, Shaw RF, Grimes CB, Cope JS (1994) Larval development, distribution, and 6(12):e27693. abundance of common dolphin, Coryphaena hippurus, and pompano dolphin, C 17. Incardona JP, et al. (2005) Aryl hydrocarbon receptor-independent toxicity of equiselis (family, coryphaenidae), in the northern Gulf of Mexico. Fish Bull 92(2): weathered crude oil during fish development. Environ Health Perspect 113(12): 275–291. 1755–1762. 38. Incardona JP, et al. (2012) Unexpectedly high mortality in Pacific herring embryos 18. Incardona JP, Collier TK, Scholz NL (2004) Defects in cardiac function precede mor- exposed to the 2007 Cosco Busan oil spill in San Francisco Bay. Proc Natl Acad Sci USA phological abnormalities in fish embryos exposed to polycyclic aromatic hydro- 109(2):E51–E58. carbons. Toxicol Appl Pharmacol 196(2):191–205. 39. Staudt D, Stainier D (2012) Uncovering the molecular and cellular mechanisms of 19. Incardona JP, et al. (2009) Cardiac arrhythmia is the primary response of embryonic heart development using the zebrafish. Annu Rev Genet 46:397–418. Pacific herring (Clupea pallasi) exposed to crude oil during weathering. Environ Sci 40. Rottbauer W, et al. (2001) Growth and function of the embryonic heart depend upon Technol 43(1):201–207. the cardiac-specific L-type calcium channel alpha1 subunit. Dev Cell 1(2):265–275. 20. Incardona JP, et al. (2013) Exxon Valdez to Deepwater Horizon: Comparable toxicity 41. Marty GD, Hose JE, McGurk MD, Brown ED, Hinton DE (1997) Histopathology and – – of both crude oils to fish early life stages. Aquat Toxicol 142 143:303 316. cytogenetic evaluation of Pacific herring larvae exposed to petroleum hydrocarbons 21. Jung J-H, et al. (2013) Geologically distinct crude oils cause a common cardiotoxicity in the laboratory or in Prince William Sound, Alaska, after the Exxon Valdez oil spill. – syndrome in developing zebrafish. Chemosphere 91(8):1146 1155. Can J Fish Aquat Sci 54(8):1846–1857. 22. Brette F, et al. (2014) Crude oil impairs cardiac excitation-contraction coupling in fish. 42. Betancur-R R, et al. (2013) The tree of life and a new classification of bony fishes. PLOS – Science 343(6172):772 776. Curr, 10.1371/currents.tol.53ba26640df0ccaee75bb165c8c26288. 23. Hicken CE, et al. (2011) Sublethal exposure to crude oil during embryonic de- 43. Scott JA, Incardona JP, Pelkki K, Shepardson S, Hodson PV (2011) AhR2-mediated; velopment alters cardiac morphology and reduces aerobic capacity in adult fish. Proc CYP1A-independent cardiovascular toxicity in zebrafish (Danio rerio) embryos ex- – Natl Acad Sci USA 108(17):7086 7090. posed to retene. Aquat Toxicol 101(1):165–174. 24. Heintz RA (2007) Chronic exposure to polynuclear aromatic hydrocarbons in natal 44. Carney SA, Peterson RE, Heideman W (2004) 2,3,7,8-Tetrachlorodibenzo-p-dioxin habitats leads to decreased equilibrium size, growth, and stability of pink salmon activation of the aryl hydrocarbon receptor/aryl hydrocarbon receptor nuclear populations. Integr Environ Assess Manag 3(3):351–363. translocator pathway causes developmental toxicity through a CYP1A-independent 25. Heintz RA, et al. (2000) Delayed effects on growth and marine survival of pink salmon mechanism in zebrafish. Mol Pharmacol 66(3):512–521. Oncorhynchus gorbuscha after exposure to crude oil during embryonic development. 45. Clark BW, Matson CW, Jung D, Di Giulio RT (2010) AHR2 mediates cardiac terato- Mar Ecol Prog Ser 208:205–216. genesis of polycyclic aromatic hydrocarbons and PCB-126 in Atlantic killifish (Fundulus 26. Carls MG, Rice SD, Hose JE (1999) Sensitivity of fish embryos to weathered crude oil: heteroclitus). Aquat Toxicol 99(2):232–240. Part I. Low-level exposure during incubation causes malformations, genetic damage, 46. Incardona JP, Day HL, Collier TK, Scholz NL (2006) Developmental toxicity of 4-ring and mortality in larval Pacific herring (Clupea pallasi). Environ Toxicol Chem 18: polycyclic aromatic hydrocarbons in zebrafish is differentially dependent on AH re- 481–493. ceptor isoforms and hepatic cytochrome P4501A metabolism. Toxicol Appl Pharmacol 27. Incardona JP, et al. (2012) Potent phototoxicity of marine bunker oil to translucent 217(3):308–321. herring embryos after prolonged weathering. PLoS ONE 7(2):e30116. 47. Incardona JP, Linbo TL, Scholz NL (2011) Cardiac toxicity of 5-ring polycyclic aromatic 28. Diercks A-R, et al. (2010) Characterization of subsurface polycyclic aromatic hydro- hydrocarbons is differentially dependent on the aryl hydrocarbon receptor 2 isoform carbons at the Deepwater Horizon site. Geophys Res Lett 37:L20602. during zebrafish development. Toxicol Appl Pharmacol 257(2):242–249.

E1518 | www.pnas.org/cgi/doi/10.1073/pnas.1320950111 Incardona et al. Footprint of Deepwater Horizon blowout impact to deep-water coral communities

Charles R. Fishera,1, Pen-Yuan Hsinga, Carl L. Kaiserb, Dana R. Yoergerb, Harry H. Robertsc, William W. Sheddd, Erik E. Cordese, Timothy M. Shankf, Samantha P. Berleta, Miles G. Saundersa, Elizabeth A. Larcoma, and James M. Brooksg aDepartment of Biology, The Pennsylvania State University, University Park, PA 16802-5301; bDepartment of Applied Ocean Physics and Engineering, Woods Hole Oceanographic Institution, Woods Hole, MA 02543-1050; cCoastal Studies Institute, Department of Oceanography and Coastal Sciences, Louisiana State University, Baton Rouge, LA 70803; dBureau of Ocean Energy Management, United States Department of the Interior, New Orleans, LA 70115; eBiology Department, Temple University, Philadelphia, PA 19122; fRedfield Laboratory, Woods Hole Oceanographic Institution, MA 02543; and gTDI-Brooks International Inc., College Station, TX 77845

Edited by Charles H. Peterson, University of North Carolina at Chapel Hill, Morehead City, NC, and accepted by the Editorial Board June 27, 2014 (received for review February 25, 2014) On April 20, 2010, the Deepwater Horizon (DWH) blowout oc- in the vicinity of the wellhead and then determine the status of curred, releasing more oil than any accidental spill in history. Oil the corals in these communities. release continued for 87 d and much of the oil and gas remained Locating deep-water coral communities in the GoM is a la- in, or returned to, the deep sea. A coral community significantly borious process as these communities are rare, relatively small, impacted by the spill was discovered in late 2010 at 1,370 m depth. and there is no known remote-sensing method to unambiguously Here we describe the discovery of five previously unknown coral locate them. Most corals require a stable, hard substrate upon communities near the Macondo wellhead and show that at least which to settle and grow (11). However, most of the sea floor in two additional coral communities were impacted by the spill. Al- the deep GoM is soft sediment. The primary exception in the though the oil-containing flocullent material that was present on deep northern Gulf are authigenic carbonates which are formed corals when the first impacted community was discovered was as an indirect byproduct of anaerobic hydrocarbon degradation by bacteria in areas with hydocarbon seepage (12, 13). Authigenic SCIENCES largely gone, a characteristic patchy covering of hydrozoans on

carbonates form hardgrounds that are often suitable for a variety ENVIRONMENTAL dead portions of the skeleton allowed recognition of impacted of attached megafauna and associated biological communities, colonies at the more recently discovered sites. One of these com- including in some cases, corals (14). munities was 6 km south of the Macondo wellhead and over 90% of the corals present showed the characteristic signs of recent impact. Discovering Coral Communities in the Deep GoM The other community, 22 km southeast of the wellhead between Because of the massive hydrocarbon reserves in the northern 1,850 and 1,950 m depth, was more lightly impacted. However, the GoM, much of the sea floor has been surveyed by energy com- discovery of this site considerably extends the distance from panies using seismic reflectivity, and copies of these data are Macondo and depth range of significant impact to benthic macro- faunal communities. We also show that most known deep-water Significance coral communities in the Gulf of Mexico do not appear to have been acutely impacted by the spill, although two of the newly discovered Deepwater Horizon communities near the wellhead apparently not impacted by the spill The blowout released more oil and gas have been impacted by deep-sea fishing operations. into the deep sea than any previous spill. Soon after the well was capped, a deep-sea community 13 km southwest of the wellhead was discovered with corals that had been damaged oil spill | octocoral | Paramuricea | autonomous underwater vehicle | anthropogenic impact by the spill. Here we show this was not an isolated incident; at least two other coral communities were also impacted by the spill. One was almost twice as far from the wellhead and in he explosion of the Deepwater Horizon (DWH) drilling rig at 50% deeper water, considerably expanding the known area of Tthe Macondo wellhead site created an oil spill with charac- impact. In addition, two of four other newly discovered coral teristics unlike those of previous major oil spills where the re- communities in the region were fouled with commercial fishing lease occurred either on the ocean surface or at shallow depths line, indicating a large cumulative effect of anthropogenic ac- (1, 2). Because of the physics of the release, as well as the ex- tivities on the corals of the deep Gulf of Mexico. tensive use of dispersants, much of the oil and gas remained at depth (3–6). In addition, weathering, burning, and application of Author contributions: C.R.F., C.L.K., D.R.Y., H.H.R., W.W.S., E.E.C., and T.M.S. designed dispersants to surface slicks resulted in a return of additional research; C.R.F., P.-Y.H., C.L.K., D.R.Y., W.W.S., E.E.C., T.M.S., S.P.B., M.G.S., E.A.L., and J.M.B. performed research; C.R.F., P.-Y.H., C.L.K., D.R.Y., S.P.B., M.G.S., and E.A.L. analyzed hydrocarbons to the deep sea (5, 7, 8). The potentially toxic data; and C.R.F. wrote the paper. hydrocarbons and dispersants had the potential to impact nu- Conflict of interest statement: The cruises and some of the analyses were funded by merous deep-sea communities that are inherently difficult to National Oceanic and Atmospheric Administration and BP as part of the Deepwater assess. In October 2010, beginning 90 d after the wellhead was Horizon (DWH) oil spill Natural Resource Damage Assessment (NRDA). Neither the DWH NRDA Trustees nor BP had a role in sample processing, data analysis, decision to capped, we visited 13 deep-water coral sites spread over a depth publish, or preparation of the manuscript. Preapproval to submit the manuscript for range of 350–2,600 m and from 87.31° to 93.60° W in the Gulf of publication was provided by representatives of the NRDA Trustees and independently Mexico (GoM), and did not detect visual indications of acute by the Bureau of Ocean Energy Management (BOEM). W.W.S. is an employee of BOEM. effects to coral communities at any of these sites. However, on This article is a PNAS Direct Submission. C.H.P. is a guest editor invited by the Editorial November 2, 2010, we discovered a previously unknown coral Board. community 13 km away from the Macondo wellhead that had Data deposition: The coral image dataset has been deposited at Penn State ScholarSphere, https://scholarsphere.psu.edu/collections/pv63g1177 (Gulf of Mexico Research Initiative clearly suffered a recent severe adverse impact, and oil forensics GRIDC#R1.x132.136:0020). indicated that hydrocarbons found on corals at the site originated 1To whom correspondence should be addressed. Email: [email protected]. from the Macondo wellhead (9, 10). Following that discovery, This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. we made a systematic effort to discover additional communities 1073/pnas.1403492111/-/DCSupplemental. www.pnas.org/cgi/doi/10.1073/pnas.1403492111 PNAS Early Edition | 1of6 possessed by the Bureau of Ocean Energy Management (BOEM) Assessment of Anthropogenic Impact to Corals through the permit process. The 3D seismic data can be used to In November 2011, we used an ROV to assess potential impact locate areas where hard substrate is present within the top 8 m of from the spill to nine sites in addition to the site already de- the sea floor and visualize conduits that deliver hydrocarbons to termined to have been impacted in Mississippi Canyon (MC) 294 the sea floor and fuel the production of authigenic carbonates (all sites are referred to by the BOEM 3 × 3 nm lease block (15). Inspection of the 3D seismic dataset at BOEM in New designation). These included five sites discovered using towed Orleans, LA revealed 488 potential hardground sites, ranging cameras and the AUV Sentry as described above and other sites in size from 0.0003 to 26 km2, within 40 km of the Macondo discovered previously (Fig. 1 and Table 1). With the exception of the Vioska Knoll (VK) sites and Atwater (AT) 357 that each wellhead (Fig. S1 A and B). From these we selected sites for harbor thousands of coral colonies, all octocoral colonies en- further consideration based on (i) the association of potential countered were photographed from a horizontal perspective from exposed hardgrounds with local topographic highs and/or sides within 1–3 m of the colonies using a digital still camera. Images of of slopes or canyons where the probability of exposure to en- octocoral colonies of sufficient resolution were then digitized as hanced currents is elevated (enhanced currents are both favor- described in Hsing et al. (18) with all branches coded as colonized able for removal of sediment from potential hardgrounds and for by hydroids, otherwise obviously impacted (covered with floccu- delivery of food to corals) (11); (ii) proximity to the Macondo lent material, with bare skeleton, excessive mucous production, or wellhead; and (iii) depth, favoring sites deeper than 900 m be- sloughing tissue), or not obviously impacted (which included dis- cause models and data on the deep-water hydrocarbon plumes colored branches and branches without expanded polyps). These from the DWH suggested impact to be most likely below this characterizations were performed independently by two observers depth (3, 16). A subset of 29 sites representing multiple areas in and the averages of their results are shown for all sites in Table S1. different directions from the Macondo wellhead, and four sites A time series of coral images from MC294 starting in November 2010 allowed documentation of the changes in the further away to the west-southwest, the direction where models appearance of corals confirmed to have been impacted from the and data had suggested the furthest excursion of deep-water oil spill in 2010 (9, 18). This temporal study allowed us to recognize plumes (3, 16, 17), were chosen for further investigation (Fig. 1). corals at other sites that were impacted in the same time frame as Twenty-five sites were imaged using either a towed or drift those confirmed to have been impacted by the spill, even though camera system tethered to a ship. This approach identified corals the adherent flocculent material originally present on the impacted at one site, and this site as well as three others where carbonates corals was normally no longer present. At this later point in time, were imaged were further investigated using the autonomous octocorals originally impacted to over 20% of their colony were underwater vehicle (AUV) Sentry. often patchily colonized by hydroids (18), a feature not seen on The AUV Sentry obtained high-resolution bathymetry of the deep-water octocorals at sites distant from the Macondo wellhead. sites and after automated processing at sea, small areas with Because the impacted corals at MC294 had not lost appreciable small-scale bathymetric relief, such as would be caused by exposed numbers of branches by November 2011, we did not include coral “ ” boulders or slabs, were identified and then imaged by Sentry stumps or dead octocoral colonies without small branches in our (Fig. 2). In addition, side-scan sonar data from an AUV survey analyses of impact from the DWH spill to the other sites. Most of the sites surveyed did not show visible evidence of acute supplied by BP (formerly known as British Petroleum) of a 2 recent impact to the colonial coral communities. Two coral sites to 375-km area around the Macondo wellhead was used to locate five the north of the Macondo wellhead in lease blocks VK906 and additional areas for image collection by Sentry.Atotalof20AUV VK826, in water depths ranging from 380 to 550 m, are 37 and 58 images from four new sites and two new areas near previously km from the Macondo wellhead, respectively. These shallower sites discovered coral sites included colonial corals. These sites and each harbor thousands of coral colonies, had been visited nu- several others were targeted for further investigation using a merous times by our research group before the spill occurred, remote-operated vehicle (ROV; Schilling ultra heavy-duty model). and continued to show no visible signs of recent impact to the

Fig. 1. Potential and confirmed deep coral sites investigated for this study. Black stars indicate sites with coral communities discovered before the 2010 and 2011 exploration efforts. Sites in green were imaged with a towed or drift camera system and corals were confirmed with the drift camera at the site marked with a green star. Sites in red were imaged with targeted AUV Sentry surveys as described in the text and the red stars indicate sites where coral communities were discovered. Inset shows the relation of AT357 to the rest of the sites.

2of6 | www.pnas.org/cgi/doi/10.1073/pnas.1403492111 Fisher et al. Fig. 2. Multibeam map and imaging tracklines from survey of MC036 site. High-resolution map from an AUV Sentry multibeam survey of a portion of the MC036 hardground area identified from a 3D seismic survey. The red lines indicate the track lines flown by Sentry at an altitude of 5 m to collect digital images

for identification of areas hosting corals. (Lower SCIENCES

Right) Image of corals taken at this site by Sentry. ENVIRONMENTAL scleractinian (Lophelia pertusa), octocoral (primarily Callogorgia Two of the newly discovered sites had limited rocky outcrops americana, but also including Paramuricea type A and C), or and few coral colonies. The site in lease block MC203 at 951 m antipatharian (primarily Leiopathes glaberrima)coralspresentat water depth hosted 19 coral colonies and the other in lease block the sites. A previously discovered site 183 km to the southwest in MC507 about 55 km southwest of Macondo at 1,040 m water lease block AT357 at 1,050 m water depth was imaged in detail for depth harbored 10 coral colonies. There was fishing line among the the first time, to our knowledge, in 2011. This site was found to corals at both of these sites, and one coral at the MC507 site which harbor thousands of coral colonies and is the largest community of was tangled in long line had a patchy injury pattern similar to what corals at a depth greater than 1,000 m currently known in the GoM. was present at MC294 (but not as extensive; Fig. 3). Although This site is dominated by the octocoral Paramuricea sp. B3 (19) and there were other corals at this site with large areas of dead skel- the scleractinian coral Madrepora cf. prolifera. There was no visual eton, the absence of small branches on the dead portions was not evidence indicating recent impact to this community observed consistent with the very recent impact as seen at MC294 and more during the ROV dives at this site (18) (Table S1). Another newly likely reflects historical impact from fishing line at this isolated site. discovered site in MC036 has extensive areas of hardground and There was extensive evidence of recent impact to the corals only a portion of it was explored by Sentry. Seventeen corals were at the newly discovered site in MC297. This site is 6 km to the discovered in one corner of the surveyed area (Fig. 2) but there was south-southeast of the Macondo wellhead at 1,560 m water no consistent visual evidence of recent impact to these corals. depth, 13 km from the impacted site in MC294. A total of 68 Similarly, we found no evidence of widespread impact to the octocoral colonies were photographed at this site in two areas octocoral community at another site only 18 km to the north of separated by about 370 m. Sixty-three of the coral colonies im- Macondo at 880 m water depth (MC118) (although a single small aged at this site shared the characteristic appearance of the colony with large areas of dead skeleton was observed). adverse impact from the DWH spill present in corals at MC294

Table 1. Coral sites surveyed in November 2011 for impact from the DWH spill Distance to No. of octocorals No. with the Macondo imaged with digital damage to Site Latitude Longitude Depth, m wellhead, km still camera >5% of colony

AT357 27.5867 −89.7048 1,050 183 52 1 MC036 28.9354 −88.2014 1,090 27 17 1 MC118 28.8527 −88.4920 880 18 16 2 MC159 28.7872 −88.6347 920 27 19 1 MC294 28.6722 −88.4765 1,370 13 54 39 MC297 28.6825 −88.3450 1,560 6 68 49 MC344 28.6337 −88.1698 1,850 22 30 7 MC507 28.4857 −88.8509 1,040 55 10 2 VK906 29.0694 −88.3774 390 37 na nd VK826 29.1560 −88.0165 500 58 na nd

na, not applicable—corals at these sites were not individually imaged; nd, none detected in the video surveys.

Fisher et al. PNAS Early Edition | 3of6 middle and lower left portion of Fig. 4A (from 2010). By November 2011 most of the flocculent material was gone and hydroids had colonized portions of the skeleton with no living material (Fig. 4B). The extensive patchy growth of hydroids on these paramuricid octocorals is very distinctive and was not present on any of the corals at other sites described above or at other sites visited previously. In Fig. 4 C and D are two of the corals from the site in MC297 photographed in November 2011 that exhibit this characteristic patchy covering of epizoic hydroids. Hsing et al. (18) reported that hydroid colonization only occurred on corals originally impacted on over 20% of their surface and only on portions that showed obvious signs of impact when first visited 3 mo after the well was capped. Thus, colonization by hydroids is an indication of impact to a significant portion of the colony. The level of impact to the coral population at MC297 is comparable to what was experienced by the coral community at MC294. Like MC294, most of the coral colonies still exhibited signs of impact in November 2011. The rate of very heavy (>50%) impact to coral colonies in November 2011 was approximately twice as high at MC297 (16%) compared with MC294 (8%). Two completely dead or hydroid-covered colonies Fig. 3. A colony of a paramuricid coral at MC506 tangled in fishing line. still retained many small branches, an attribute consistent with recent impact. Similar to the community at MC294, the visible effects were patchy not only across the site, but also on individual at the same point in time. Forty-nine coral colonies showed ev- coral colonies. This pattern suggests that the impacting agent was idence of recent impact to over 5% of the colony and 38 of the not evenly dispersed in the bottom water, but rather present as corals displayed evidence of recent impact to over 10% of the microdroplets or particles; whether this represents small droplets colony (Table 1 and Table S1). Fig. 4 A and B shows one of of oil/dispersant (20, 21) or oil-containing marine snow (8) is the impacted corals from MC294 as it first appeared when dis- not known. covered in November 2010, and its appearance in November In a response to White et al. (9), Boehm and Carragher (22) 2011. In November 2010, a brown flocculent material covered suggested that the impact to the community of MC294 could much of the coral with live yellow coral tissue evident in the have been due to causes such as slope failure that may have

Fig. 4. Recently impacted corals from MC294 and MC297. (A) Colony of Paramuricea biscaya at MC294 as it appeared in November 2010, 3 mo after the well was capped. (B) The same coral in November 2011. (C and D) Colonies of P. biscaya at MC297 in November 2011. Note the extended polyps and apparently healthy yellow tissue on most branches, and the patchy brown hydroid growth on other portions of the colonies.

4of6 | www.pnas.org/cgi/doi/10.1073/pnas.1403492111 Fisher et al. occurred coincidently in the same time frame as the DWH abundant taxa at all of the sites below 1,000 m, normally main- blowout, or other local causes. The discovery of this second tain living polyps over their entire surface. These attached col- community, at the same stage of postimpact injury progression onies obtain food from, and exchange respiratory gases with, the 13 km away from MC294, indicates that the impact cannot be bottom water bathing their exposed surfaces. In essence, they are explained by a more localized event. That this is the only coral constantly sampling the water surrounding them. If impacted by community discovered to date that is closer to the Macondo waterborne agents, they cannot move nor cover their exposed wellhead than the MC294 community and the fact that most tissues except by exuding a thin layer of mucous. If there is a other communities further away do not show similar visible signs significant impact to a portion of the colony, it may be recorded of impact provide additional evidence linking the current state of as damaged tissue, bare skeleton, or epizoic encrustation on that both communities to the DWH blowout. portion of the colony (9, 18, 27). If a colony dies, its skeleton Although most other deep-water coral communities we have remains attached to the sea floor for years, slowly losing smaller visited in the GoM did not show widespread visual evidence of branches and providing a record of its existence and death. Be- recent acute impact, evidence of injury was found at one other cause these colonial animals normally live for many hundreds of site near the Macondo wellhead. This site in lease block MC344 years, natural death is a rare event (19). As a result, these types at 1,850–1,950 m depth is 22 km east of the wellhead. At this site of corals are reliable visual biomonitors of anthropogenic impact 30 corals were photographed. The visual evidence of impact to to the deep-sea benthos. this community in November 2011 was much less severe than Thetimecourseandsequenceofeventsthatoccuroverthefirst that observed at MC294 or MC297. Fourteen coral colonies 18 mo after deep-sea octocoral colonies are acutely impacted by exhibited evidence of recent impact noted by both observers. a toxic waterborne agent has been well described (18). This docu- Most of these had only small areas impacted, either because the mented timeline allowed for the recognition of acute impact to other corals were quite small or because only small portions of the sites, even after the initial appearance of the impact had changed, corals were impacted, although six corals showed visible evidence and the original causative agent(s) may have been removed with of impact to over 10% of the colony. Although largely minor, sloughing mucous and tissue, by currents, or by microbial activity. visible effects were widespread at this site. At this site there was The distinctive appearance of the corals at some sites near the also very little colonization of the impacted corals by hydroids, Macondo wellhead, and the absence of corals with these anomalous perhaps reflecting a lighter initial impact and a slower pro- features at all other sites, leads to the conclusion that the acute gression to this stage, or a difference in hydroid colonization impact to deep benthic megafauna communities was not limited to the one site discovered shortly after the event, but rather extended to abilities associated with the difference in location and depth. SCIENCES at least two other deep coral habitats, and perhaps more in the re-

However, the patchy nature of the impact and the appearance of ENVIRONMENTAL gion of the Macondo wellhead that have yet to be discovered. the impacted branches are consistent with what was observed at We have now also carefully monitored corals at numerous MC294 and MC297 (Fig. 5). We conclude that this site was also other deep-water sites all over the northern GoM and found no impacted at the same time and in the same way as the two sites compelling evidence of acute impact from the spill at any coral closer to the Macondo wellhead, albeit not as heavily. sites between 400 and 850 m depth or more than 30 km from Although early National Oceanic and Atmospheric Adminis- Macondo. Although it is still possible that other sites will be tration models and empirical studies suggested that the deep- discovered, the extensive survey and sampling reported here water plume of hydrocarbons from the blowout moved pre- suggest that we have constrained the footprint of acute impact dominantly to the southwest, later models suggest a more dynamic to deep-water coral communities in the GoM from the DWH pattern of swirling flow from the wellhead that could readily blowout. However, it may still be many years before the effects transport hydrocarbon rich fluids to the east in the direction of of subacute impact are manifested in the deep-water coral MC344 (23, 24). The data presented here and in White et al. (9) communities of the wider GoM. indicate that impact from oil and/or dispersant from the DWH – With our ever-increasing population and technological ad- spill occurred at depths greater than 1,000 1,300 m as predicted by vancements, anthropogenic impact to deep-sea habitats and bi- most models. ota will likely continue to increase. Although far removed from surface and coastal waters, and from the consciousness of most Corals as Biomonitors of Anthropogenic Impact to the people, deep-sea environments play numerous roles in the health Deep Sea of the world’s oceans. Many species of fishes and sharks use deep Deep-sea corals are among the longest-lived animals on the corals as spawning grounds or sites for deposition of eggs (28) planet, with some species living thousands of years (25, 26). (Fig. S2). Healthy deep-sea sediments are remarkably high in Many octocorals, like the paramuricids that were the most biodiversity, and important in global carbon and nitrogen cycling, decomposition processes, and energy flow to higher trophic-level consumers (29–31). Perhaps most importantly, we know relatively little about deep-sea fauna and communities, and therefore the full spectrum of ecosystem services derived from deep-sea biota and habitats is largely unknown. Accumulating baseline data on conditions in different deep-sea habitats as well as monitoring for changes in these habitats will prove to be critical when scientists are asked to evaluate the inevitable impacts these ecosystems will experience and provide input on mitigation. Methods Site Selection. The acoustic amplitude maps used in this study were generated from 3D seismic data acquired by the oil industry and provided to BOEM as required by the permitting process. Although these data were shot and recorded primarily for exploration targets thousands of meters below the seafloor, they are also useful in characterizing changes in seafloor lithology. High positive anomalies are acoustically faster than both seawater and soft bottom mud, resulting in strong responses on the amplitude maps. High positives can be indicators of hydrocarbon migration pathways that have distorted the seismic Fig. 5. (A and B)TwoParamuricea sp. from MC344 in November 2011 with response on vertical cross-sections. Typically, at historic or current hydrocarbon apparently healthy and visibly unhealthy and dead portions of the colonies seep sites, high positive response is also associated with the presence of apparent. authigenic carbonates formed as a byproduct of bacterial activity in shallow

Fisher et al. PNAS Early Edition | 5of6 subsurface sediments. These rocky, calcium carbonate substrates are suitable Quantification of Impact to Coral Colonies. The sites surveyed by Sentry and habitats for corals, but only where ocean bottom currents are adequate to towed cameras, plus previously known sites in the vicinity of the Macondo keep unconsolidated hemipelagic mud off the top of the rock. wellhead, were revisited with an ROV. The corals imaged by Sentry were Sites for exploration were selected at the BOEM office in New Orleans located again and the surrounding areas searched using sonar to locate using their 3D seismic database, which covers over 90% of the northern GoM exposed carbonates that potentially host corals. At each site except AT357, continental slope. The data were analyzed for surface reflectivity or ampli- VK826, and VK906, we attempted to image every coral encountered. At ’ tude using Schlumberger s Geoframe software on a Dell workstation. The AT357, VK826, and VK906 we actively searched among the thousands of Geoframe’s program ASAPwas used to further define the seafloor horizon, corals present at these sites for colonies that were covered with flocculent followed by the manual review for gaps in the seafloor identifications. Po- material, colonized by hydroids, or contained portions of dead skeleton. tential coral sites were identified by areas of high positive amplitude Corals were also opportunistically imaged during survey of these sites. The (reflectivity), appropriate bathymetry (near crests of canyons or steep slopes or on local highs), and subsurface profiles that identified faults and other visible impact to each coral colony was quantified on high-resolution digital – fluid and gas migration pathways that could supply hydrocarbons for shal- images acquired from an ROV within 1 3 m of the coral. Following the low microbial activity and production of authigenic carbonates. methods of Hsing et al. (18), all portions of coral branches not obscured by associated organisms were categorized into one of the following categories Initial Site Survey. Twenty-five sites selected from examination off the 3D and digitized using Fiji 1.4 (32) or Inkscape 0.48.2 (33) software: (i) covered seismic data were surveyed using a camera system deployed on a tether from by hydroids: branches obviously covered by hydroids; (ii) other impacted a surface ship. One system, the TDI Brooks Drift Camera system, used a 14.7 branches, i.e., branches not covered with hydroids but either covered by floc, mega-pixel Pentax digital camera with strobe illumination and the second, showing excess mucous, tissue damage, or bare skeleton; and (iii) no visible the WHOI TowCam SN 6004 used a 3.3 megapixel color camera with strobe impact, i.e., branches not clearly in either of the other two categories. All illumination. Both were flown at a height of 2–5 m above the bottom of the corals were digitized by three independent observers instructed to use cat- sea floor. Over 44,500 images were collected with these systems and three egory iii by default unless visible impact as described above, or hydroid sites with abundant carbonates and one with colonial corals were identified growth, was clear in the image. Because an ANOVA showed the differences for further investigation. These four sites and five others were further sur- between observers to be small, simple arithmetic means of their digitized veyed by the AUV Sentry using a Reson7125 400 kHz multibeam echo values were used for downstream analyses (Table S1). sounder from an altitude of 25 m to obtain high-resolution bathymetry, with a postprocessed pixel size of 0.5 m. After automated processing at sea, ACKNOWLEDGMENTS. We thank A. Fundis and G. Kurras for oversight small areas with small-scale bathymetric relief, such as would be caused by of the towed camera system; and A. Chopra, W. Cho, A. Govindarajan, exposed boulders or slabs, were identified and surveyed again with the S. Herrera, A. Lukasiewicz, T. Muric, C. Munro, and M. Zelenevich for assistance Sentry vehicle collecting overlapping digital images from 5 m above the sea with data collection. We also thank I. MacDonald for providing the digital floor in a series of small grid surveys centered on the features of interest still camera used for the in situ imaging, the crews of the ships and deep (Fig. 2). During the 16 imaging deployments of the AUV Sentry, over 68,000 submergence vehicles used for this study, and BP for access to AUV bathymetry photographs were collected and colonial corals identified in 20 images. and side-scan data. This work was supported by the Assessment and Restoration Sentry’s geodetic position was determined by ultrashort baseline acoustic Division of the National Oceanic and Atmospheric Administration (NOAA) and Gulf of Mexico Research Initiative funding to support the Ecosystem navigation from the support vessel combined with the vehicle’s internal Impacts of Oil and Gas Inputs to the Gulf (ECOGIG) consortium administered estimates of its relative motion provided by a combination of a Doppler by the University of Mississippi. The research cruises and some analyses were velocity log and an inertial navigation system. The resulting vehicle geodetic funded by the NOAA and BP as part of the Deepwater Horizon oil spill position was determined to within 5 m, allowing targets to be directly ac- Natural Resource Damage Assessment. This is Contribution 195 from the quired on subsequent ROV dives. ECOGIG consortium.

1. Socolofsky SA, Adams EE, Sherwood CR (2011) Formation dynamics of subsurface www.restorethegulf.gov/sites/default/files/documents/pdf/OSAT_Report_FINAL_17DEC. hydrocarbon intrusions following the Deepwater Horizon blowout. Geophys Res Lett pdf. Accessed July 17, 2014. 38(9):L09602. 18. Hsing P-Y, et al. (2013) Evidence of lasting impact of the Deepwater Horizon oil spill 2. Peterson CH, et al. (2012) A tale of two spills: Novel science and policy implications of on a deep Gulf of Mexico coral community. Elem Sci Anthr 1:0000012. an emerging new oil spill model. Bioscience 62(5):461–469. 19. Doughty CL, Quattrini AM, Cordes EE (2014) Insights into the population dynamics of 3. Camilli R, et al. (2010) Tracking hydrocarbon plume transport and biodegradation at the deep-sea coral genus Paramuricea in the Gulf of Mexico. Deep Sea Res Part II Top Deepwater Horizon. Science 330(6001):201–204. Stud Ocean 99:71–82. 4. Joye SB, MacDonald IR, Leifer I, Asper V (2011) Magnitude and oxidation potential of 20. Adcroft A, et al. (2010) Simulations of underwater plumes of dissolved oil in the Gulf hydrocarbon gases released from the BP oil well blowout. Nat Geosci 4(7):160–164. of Mexico. Geophys Res Lett 37(18):L18605. 5. Reddy CM, et al. (2012) Composition and fate of gas and oil released to the water 21. Paris CB, et al. (2012) Evolution of the Macondo well blowout: Simulating the effects column during the Deepwater Horizon oil spill. Proc Natl Acad Sci USA 109(50): of the circulation and synthetic dispersants on the subsea oil transport. Environ Sci 20229–20234. Technol 46(24):13293–13302. 6. Ryerson TB, et al. (2012) Chemical data quantify Deepwater Horizon hydrocarbon flow 22. Boehm PD, Carragher PD (2012) Location of natural oil seep and chemical finger- rate and environmental distribution. Proc Natl Acad Sci USA 109(50):20246–20253. printing suggest alternative explanation for deep sea coral observations. Proc Natl 7. Hazen TC, et al. (2010) Deep-sea oil plume enriches indigenous oil-degrading bacteria. Acad Sci USA 109(40):E2647, author reply E2648. Science 330(6001):204–208. 23. Valentine DL, et al. (2012) Dynamic autoinoculation and the microbial ecology of 8. Passow U, Ziervogel K, Asper V, Diercks A (2012) Marine snow formation in the aftermath a deep water hydrocarbon irruption. Proc Natl Acad Sci USA 109(50):20286–20291. of the Deepwater Horizon oil spill in the Gulf of Mexico. Environ Res Lett 7(3):035301. 24. Spier C, Stringfellow WT, Hazen TC, Conrad M (2013) Distribution of hydrocarbons 9. White HK, et al. (2012) Impact of the Deepwater Horizon oil spill on a deep-water released during the 2010 MC252 oil spill in deep offshore waters. Environ Pollut 173: coral community in the Gulf of Mexico. Proc Natl Acad Sci USA 109(50):20303–20308. 224–230. 10. White HK, et al. (2012) Reply to Boehm and Carragher: Multiple lines of evidence link 25. Roark EB, Guilderson TP, Dunbar RB, Fallon SJ, Mucciarone DA (2009) Extreme lon- deep-water coral damage to Deepwater Horizon oil spill. Proc Natl Acad Sci USA gevity in proteinaceous deep-sea corals. Proc Natl Acad Sci USA 106(13):5204–5208. 109(40):E2648. 26. Prouty NG, Roark EB, Buster NA, Ross SW (2011) Growth rate and age distribution of 11. Roberts JM, Wheeler A, Freiwald A, Cairns S (2009) Cold-Water Corals: The Biology and deep-sea black corals in the Gulf of Mexico. Mar Ecol Prog Ser 423:101–115. Geology of Deep-Sea Coral Habitats (Cambridge Univ Press, Cambridge, UK). 27. Cerrano C, et al. (2000) A catastrophic mass-mortality episode of gorgonians and 12. Joye SB, et al. (2004) The anaerobic oxidation of methane and sulfate reduction in other organisms in the Ligurian Sea (North-western Mediterranean), summer 1999. sediments from Gulf of Mexico cold seeps. Chem Geol 205(3):219–238. Ecol Lett 3(4):284–293. 13. Pancost RD, et al. (2005) Lipid biomarkers preserved in hydrate-associated authi- 28. Baillon S, Hamel J-F, Wareham VE, Mercier A (2012) Deep cold-water corals as nurs- genic carbonate rocks of the Gulf of Mexico. Palaeogeogr Palaeoclimatol Palaeoecol eries for fish larvae. Front Ecol Environ 10(7):351–356. 227(1):48–66. 29. Danovaro R, et al. (2008) Exponential decline of deep-sea ecosystem functioning 14. Fisher C, Roberts H, Cordes E, Bernard B (2007) Cold seeps and associated communities linked to benthic biodiversity loss. Curr Biol 18(1):1–8. of the Gulf of Mexico. Oceanography (Wash DC) 20:118–129. 30. Tyler PA (2003) Disposal in the deep sea: Analogue of nature or faux ami? Environ 15. Roberts HH, Shedd W, Hunt J, Jr (2010) Dive site geology: DSV ALVIN (2006) and ROV Conserv 30(1):26–39. JASON II (2007) dives to the middle-lower continental slope, northern Gulf of Mexico. 31. Armstrong CW, Foley NS, Tinch R, van den Hove S (2012) Services from the deep: Steps Deep Sea Res Part II Top Stud Oceanogr 57(4):1837–1858. towards valuation of deep sea goods and services. Ecosyst Serv 2:2–13. 16. Diercks A-R, et al. (2010) Characterization of subsurface polycyclic aromatic hydro- 32. Schindelin J, et al. (2012) Fiji: An open-source platform for biological-image analysis. carbons at the Deepwater Horizon site. Geophys Res Lett 37(20):L20602. Nat Methods 9(7):676–682. 17. Operational Science Advisory Team (2010) Sub-sea and sub-surface oil and 33. The Inkscape Team (2011) Inkscape 0.48.2. Avalailable at www.inkscape.org. Accessed dispersant detection: Sampling and monitoring. Unified Area Command. Available at July 16, 2014.

6of6 | www.pnas.org/cgi/doi/10.1073/pnas.1403492111 Fisher et al. Supporting Information

Fisher et al. 10.1073/pnas.1403492111

Fig. S1. Three-dimensional seismic dataset showing potential hardground sites (shown in red in A) in the vicinity of the Macondo wellhead, which is represented by a yellow star. The red outline in A and B is the 40-km radius around the well.

Fig. S2. Paramuricea sp. with a shark egg case at Mississippi Canyon 159.

Other Supporting Information Files

Table S1 (DOCX)

Fisher et al. www.pnas.org/cgi/content/short/1403492111 1of1 This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Article

pubs.acs.org/est

PAH Exposure in Gulf of Mexico Demersal Fishes, Post-Deepwater Horizon Susan M. Snyder,*,† Erin L. Pulster,‡ Dana L. Wetzel,‡ and Steven A. Murawski†

† University of South Florida, College of Marine Science, St. Petersburg, Florida 33701, United States ‡ Mote Marine Laboratory, Sarasota, Florida 34236, United States

*S Supporting Information

ABSTRACT: Following the 2010 Deepwater Horizon (DWH) blowout, we surveyed offshore demersal fishes in the northern Gulf of Mexico (GoM) in 2011−2013, to assess polycyclic aromatic hydrocarbon (PAH) exposure. Biliary PAH metabolites were estimated in 271 samples of golden tilefish (Lopholatilus chamaeleonticeps), king snake eel (Ophichthus rex), and red snapper (Lutjanus campechanus), using high performance liquid chromatography with fluorescence detection. Mean concentration of naphthalene metabolites in golden tilefish (240 μgg−1) was significantly higher (p = 0.001) than in red snapper (61 μgg−1) or king snake eel (38 μgg−1). Biliary naphthalene metabolite concentration decreased over the study period in red snapper (58%) and king snake eel (37%), indicating likely episodic exposure, while concentrations were persistently high in golden tilefish. Naphthalene metabolite levels measured in golden tilefish are among the highest concentrations measured in fishes globally, while concentrations for red snapper and king snake eel are similar to pre-DWH levels measured in GoM species. In contrast, concentrations of benzo[a]pyrene metabolites were similar for all three species (p = 0.265, mean 220 ng g−1) and relatively low when compared to GoM, global data and previous oil spills. These data support previous findings that fish life history and physiology play significant roles in exposure and uptake of PAH pollution.

■ INTRODUCTION each route of exposure will change with physicochemical properties of each PAH, such as the octanol−water partition The Deepwater Horizon (DWH) blowout in the Gulf of Mexico ffi 20 (GoM) released 4.9 million barrels of crude oil between April coe cient (Kow). Direct exposure to PAH-contaminated 1 − sediment and a diet of benthic prey is suspected to be an 20th and July 15th, 2010. Through multiple mechanisms, 4 21−24 fl important source of exposure for demersal organisms. 31% of the oil residue settled on the northern GoM sea oor fi and its effects are apparent in sediment cores taken at sites Following uptake, the hepato-biliary system in shes works to − 25 polluted by the spill with negative impacts on the benthos.2 5 metabolize and eliminate PAHs. The metabolites are accumulated in the bile which is stored in the gall bladder. In Given the persistence of DWH oil in the environment, there is fi a need to understand the interactions with, and impacts on fish sh, metabolism rapidly converts up to 99% of PAH molecules and other biota, particularly demersal fishes living in contact into a more hydrophilic PAH metabolite leading to rapid − with sediments that may contain residual DWH oil.2,6 8 excretion. Therefore, the concentration of parent PAHs in Polycyclic aromatic hydrocarbons (PAHs) are considered the routinely monitored tissues, such as muscle and liver, often reveal only trace levels of contamination and are not necessarily most toxic component of crude oil to marine life and are 20,26−30 ubiquitous pollutants in the marine environment.9,10 Hydro- good quantitative indicators of PAH exposure. The carbon analysis of DWH oil collected from the wellhead presence of biliary PAH metabolites represents an early marker − 11,12 of relatively short-term exposure to PAHs from all routes of documented 3.8 4.0% PAHs by weight. With 4.9 million ff barrels of oil released during the DWH blowout, there was a exposure and has been associated with known sublethal e ects fi 22,28,31,32 large episodic pulse of PAHs associated with DWH into the in sh. fi GoM in 2010, with the potential to negatively impact fishes. We studied three demersal sh species, the burrow-forming fi Exposure to PAHs has been linked with a variety of sublethal golden tile sh (Lopholatilus chamaeleonticeps), the mud-dwell- fi effects in fish, including DNA damage, hepatic lesions and ing king snake eel (Ophichthus rex) and the reef sh, red neoplasia, epidermal lesions, immunosuppression, cardiotox- snapper (Lutjanus campechanus). These species represent a fi icity, reduced adult fitness, altered and reduced growth, gradient of likely sediment associations, with golden tile sh “toxicant-induced starvation”, disrupted cell membranes, gill being heavily associated with sediments, king snake eel likely abnormalities, osmoregulatory imbalance, endocrine disruption, − decreased fecundity and reduced survival to maturity.10,13 17 Received: April 13, 2015 Routes of exposure to PAHs in demersal fishes include Revised: June 8, 2015 ingestion, ventilation over the gills and dermal uptake and may Accepted: June 11, 2015 act simultaneously.10,18,19 However, the relative importance of

© XXXX American Chemical Society A DOI: 10.1021/acs.est.5b01870 Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article being moderately/heavily associated with the sediments, and their metabolites, alkylated derivatives of PAHs, their red snapper being more distantly associated with sedi- metabolites and N-, S-, and O-containing compounds with − ments.33 35 Studying differential PAH exposure between fish the same aromatic structure.29 The HPLC-F bile screening with varying levels of association with the sediment, including method has been validated by comparison to quantitative GC/ fishes that ingest sediment, has shown that the level of direct MS methodology following oil spills and in environmental contact with contaminated sediment leads to differential body monitoring, demonstrating strong correlation between the two burdens of PAHs.36 methods for NPH equivalents (r = 0.94, p < 0.0001), PHN Golden tilefish are demersal, nonmigratory and relatively equivalents (r = 0.93, p < 0.0001) and BaP equivalents.29,30,50,51 long-lived fish that are commercially important in the The method is routinely used in a number of biomonitoring − GoM.37 39 Golden tilefish have been observed living in a studies to estimate PAH exposure in fish and is frequently the variety of shelter habitats, however, their primary habitat is a preferred method over the more rigorous analytical approaches large funnel-shaped burrow.38,40 Living in depositional environ- following an oil spill where there is a need to test large numbers fi of samples for exposure in a timely and cost-effective ments, the lling-in of a burrow is rapid, therefore, substantial − maintenance of a burrow is required, consequently, golden manner.29,30,47,49,51 53 tilefish frequently bioturbate sediments with their mouths and bodies.34,37,40 This incidental sediment ingestion could ■ MATERIALS AND METHODS potentially be a key route of exposure to PAH pollution for Collection of Samples. During 2011−2013, extensive the species. In addition, golden tilefish are a species of interest demersal longline surveys evaluating fish disease were due to the highest frequency of external skin lesions (7%), conducted in the GoM, from the West Florida Shelf (WFS) observed during a 2011 disease survey.13 While cause and effect to west of the Mississippi River.13 In 2011, fish were caught via have not been established, it has neither been rejected.13 demersal longlining on chartered commercial fishing vessels Little information is available about king snake eel life between June and August at depths ranging from 38−180 m history. They are described as “mud eels” and “obligate mud (Figure 1). From 2012, all fish were caught via demersal dwellers” that are associated with soft-bottom habitat and are often concentrated near oil rigs.35,41 The limited literature does not suggest king snake eel form permanent burrows comparable to those of golden tilefish but they may burrow into sediments to hide or to search for food as is similar to the behavior of other Ophichthid eels.42 Additionally, 82% of king snake eel habitat overlaps with the area that received some oil following DWH blowout, making it a species of high interest relative to oil pollution effects.43 Red snapper are a commercially and recreationally important demersal reef fish associated more with vertical structure (e.g., natural and artificial reefs, offshore oil infrastructure) than the bottom sediments.33,44 All three study species are known to eat benthic prey during parts of their life cycle.41,44,45 The objective of this study was to estimate equivalent concentrations of metabolites of three common PAHs found in DWH crude oil, naphthalene (NPH), phenanthrene (PHN), Figure 1. Location of sampling stations conducted in the northern and benzo[a]pyrene (BaP) as a biomarker of recent PAH Gulf of Mexico in 2011 (white), 2012−2013 (red, gray, yellow) and exposure in three demersal fish species in the GoM following the Deepwater Horizon (DWH) blowout. Red markers denote red the DWH blowout and primarily designed to monitor relative snapper stations grouped as northern Gulf of Mexico (nGoM) concentrations over time. This research is part of a much larger stations. Yellow markers denote red snapper stations grouped as West study of fish disease and contamination in the GoM and Florida Shelf (WFS) stations. Adapted by permission. Copyright © 2015 Esri, DeLorme, GEBCO, NOAA, NGDC. All rights reserved. ongoing analyses include quantifying PAHs and alkylated homologues in muscle and liver tissue, exposure studies to assess sublethal effects, sediment analysis, and risk assessment. longlining in the month of August, onboard the R/V Knowing the relative level of PAH exposure is important in Weatherbird II, at depths ranging from 34−389 m. Additionally, understanding the link between PAH exposure, accumulation in red snapper samples were obtained from the Madison-Swanson edible tissues and resulting sublethal effects measured, fishery closed area, by rod and reel in June of 2013 and 2014, at particularly since PAHs exert their toxic effects after a depth of 85 m (Figure 1). metabolism.22,46,47 In addition, this study provides ample At each longline sampling station, an average of 495 baited baseline data, which can be used to measure future environ- #13 circle hooks were attached to 91-kg-test leaders and to 3.2 mental impacts in the GoM. mm galvanized steel (2011, 2012), or 544-kg-test monofilament In three years following the DWH blowout (2011−2013), main line (2013). Bait used was cut fish (Atlantic mackerel we collected bile from the three species to screen bile samples (Scomber scombrus)) or various squid. Temperature−time- for relative concentrations of biliary PAH metabolites and depth recorders (Star: Oddi CDST Centi) were deployed at fluorescent aromatic compounds (FACs). This study used a the beginning and end of each longline set. Latitude, longitude, widely accepted semiquantitative method, high performance depth, and weather conditions were also recorded at the liquid chromatography with fluorescence detection (HPLC-F), beginning and end of each set. Average soak time was 2 h. to estimate concentrations of biliary PAH metabolites.29,48,49 Target fish for the longline surveys included the three species This methodology estimates the concentration of parent PAHs, analyzed in this study. The target fish species were processed at

B DOI: 10.1021/acs.est.5b01870 Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article

Table 1. Summary Statistics for Biliary Naphthalene (μgg−1) and Benz[a]pyrene (ng g−1) Metabolite Equivalent Concentrations for Red Snapper, Golden Tileﬁsh and King Snake Eel Caught in the Northern Gulf of Mexico, Separated by a Year of Catch

− year naphthalene equivalents (μgg 1) species mean median range SD n mean length (cm) % female : % male 2011 red snapper 120 110 41−470 78 30 65 60:40 2012 red snapper 61 54 20−130 27 15 58 60:40 golden tileﬁsh 240 230 110−340 61 24 67 n/a king snake eel 38 24 11−88 27 23 151 n/a 2013 red snapper 51 48 13−140 27 63 65 47:53 golden tileﬁsh 220 230 22−480 110 72 65 63:37 king snake eel 24 16 3.6−210 34 44 140 n/a

− year benzo[a]pyrene equivalents (ng g 1) 2011 species mean median range SD n red snapper 280 260 94−590 140 30 2012 red snapper 220 170 68−540 150 15 golden tilefish 170 140 51−470 110 24 king snake eel 260 160 46−850 150 23 2013 red snapper 380 300 310−1500 330 63 golden tilefish 370 220 71−3030 450 72 king snake eel 160 130 34−880 150 44 aSD, standard deviation, n, number of samples analyzed. Mean length (cm) and sex ratio (% female: % male) is also provided, and are the same for individuals measured for naphthalene and benzo[a]pyrene equivalents. Sex ratio is not provided for king snake eel (n/a = not applicable), as sex is difficult to determine, and sex is not provided for 2012 golden tilefish catch. time of capture for standard and total lengths, weight, sex and a Chromatograms were recorded at representative wavelength subsample of the catch were selected to be sampled for bile, pairs of 292/335 nm for the NPH equivalents (2−3 ring liver, muscle, otoliths and other tissues. Bile was collected by FACs), 260/380 nm for the PHN equivalents (3−4 ring FACs) dissecting the gall bladder away from the liver, cutting the bile and 380/430 nm for the BaP equivalents (4−5 ring FACs). All duct, and draining the fluid via the duct into 8 mL combusted peaks within the portion of the chromatogram where amber vials. The samples were immediately frozen. In the metabolites elute (6−19 min), were integrated for each laboratory, bile samples were stored at −40 °C until analysis. wavelength pair, summed and FACs were calculated using Laboratory Analysis. All 2011 bile samples (n = 30) were external standards of the respective parent compounds, NPH, PHN, and BaP, to convert sample area (fluorescence response) analyzed at the Northwest Fisheries Science Center (NWFSC), − Seattle, WA. The 2012 (n = 62) and 2013 (n = 179) bile to PAH equivalents (ng g 1) bile wet weight using the 48 samples were analyzed at Mote Marine Laboratory (MML), following calculation: Sarasota, FL. Prior to analysis of the 2012 and 2013 samples, an standard concentration integrated sample area(6− 19min) uL of standard injected × × interlaboratory comparison was completed to validate methods, standard mean area density of bile uL of sample injected precision and accuracy between the two laboratories. The where the density of bile is 1.03 g mL−1.21 All equivalent interlab comparison used three bile samples from 2011, from fi fi ff concentrations were reported to two signi cant gures. di erent species (cobia (Rachycentron canadum), greater Quality Assurance/Control. Quality assurance was moni- amberjack (Seriola dumerili), red snapper) and over a wide tored in four ways. (1) An interlaboratory comparison of three − μ −1 range of concentrations (for NPH: 41 240 gg ; for PHN: 2011 bile samples was performed between MML and the − μ −1 − −1 6.3 46 gg , for BaP: 97 310 ng g ). Prior to the interlab NWFSC discussed above. (2) A methanol solvent blank was comparison, accuracy was monitored as part of the quality run prior to every field sample. The area of the methanol blank fi assurance plan at the NWFSC using a sh bile control sample was subtracted from the area of the field sample that was μ −1 (bile of Atlantic salmon (Salmo salar) exposed to 25 gmL subsequently analyzed. (3) Each field sample was run in of Monterey Crude oil for 48 h). There was successful duplicate, with a CV of less than 15%. If the CV between interlaboratory agreement for the three bile samples, with a CV duplicates was greater than 15%, the sample was run again in of less than 15%, for PAH equivalents for NPH, PHN, and BaP. triplicate until the CV reached less than 15%. (4) A continuing Bile samples were then analyzed using the semiquantitative bile calibration was used to monitor instrument stability throughout screening HPLC-F method following NWFSC Environmental the entire analysis by running the quantifying standards of 29,48,49 Chemistry program protocols described below. parent PAHs of NPH (2.5 μgmL−1), PHN (1 μgmL−1) and Untreated bile samples (3 μL) were injected directly onto BaP (250 ng mL−1) every 12 field samples, making sure the CV the HPLC-F system (MML: Agilent Technologies, Series 1100, remained less than 15%. Santa Clara, CA; NWFSC: Waters, Milford, MA) equipped Data Analysis. Nonparametric, or permutation-based, with a C-18 reverse-phase column (Synergi 4 μm Hydro-RP hypothesis tests were run as it is unrealistic that biological 80A, Phenomenex, Torrence, CA), with the column temper- data meet the assumptions (e.g., normal distribution, random ature held at 50 °C. Fluorescent aromatic compounds were sampling) of traditional parametric analyses, such as analysis of eluted at 1 mL/min using a linear gradient from 100% solvent variance.54,55 If a potential explanatory variable was categorical A (0.5% acetic acid in water) to 100% solvent B (methanol). (e.g., year), a nonparametric multivariate analysis of variance

C DOI: 10.1021/acs.est.5b01870 Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article

(npMANOVA), also known as a permutation-based analysis of variance when used on univariate data, was run using α = 0.05 to reject the null hypothesis. Between-group dispersion was checked for homogeneity prior to each npMANOVA and if dispersions were nonhomogenous, the data were transformed to minimize heterogeneity. If the npMANOVA was significant, a pairwise npMANOVA was run to assess subset’s significant differences. An adjusted p-value, using the Holms−Bonferroni transformation, was used to test the null hypothesis for a pairwise npMANOVA, using α = 0.05. ■ RESULTS AND DISCUSSION Biliary PAH Metabolite Concentrations. This study examined 271 bile samples, from 26 longline stations, collected over the three years following the DWH blowout, including 96 golden tilefish, 67 king snake eel, and 108 red snapper (Table Figure 3. Biliary benzo[a]pyrene metabolite concentrations (ng g−1) 1). The concentrations of biliary NPH and PHN equivalents for golden tilefish (2012: n = 24; 2013: n = 72), red snapper (2011: n varied linearly within individuals (r = 0.92, p = 0.001), for all = 30; 2012: n = 15; 2013: n = 63) and king snake eel (2012: n = 23; three species, and within sites, therefore, only results for NPH 2013: n = 44), sampled in 2011, 2012, and 2013 in the northern Gulf equivalents are analyzed and discussed to avoid redundant of Mexico. information. The strong correlation between NPH and PHN metabolite concentration has been found in other studies snapper for both 2012 (p = 0.003) and 2013 (p = 0.003), six suggesting a common petrogenic PAH source.51,56,57 The times higher than king snake eel in 2012 (p = 0.003) and nine correlation between NPH and BaP metabolites was also times higher than king snake eel in 2013 (p = 0.003). This significant (r = 0.30, p = 0.001), although much weaker than difference persists overall and at the four longline stations the correlation between NPH and PHN, therefore, results for where golden tilefish and king snake eel co-occurred. Golden BaP metabolite concentration are analyzed and presented tilefish and red snapper have never been caught at the same separately. The low molecular weight (LMW) PAHs, station on our longlining cruises due to differences in habitat specifically NPH and PHN, are present in relatively high and depth range. Red snapper mean NPH metabolite concentrations in DWH crude oil while BaP is found in much concentration is consistently higher than king snake eel smaller concentrations, or not quantified above the detection concentrations, with mean concentrations five times higher in limits of the instrument.11,12 2012 (p = 0.012) and two times higher in 2013 (p = 0.003). Species-SpecificDifferences in Biliary PAH Metabolite This pattern also occurred at two of three longline stations Concentration. For both 2012 and 2013 data, there is a where red snapper and king snake eel co-occurred. significant difference in NPH metabolite concentrations In contrast to results for NPH, no significant differences in between all three study organisms, with golden tilefish having BaP metabolite concentration were found among the three significantly higher concentrations of biliary NPH metabolites species for 2012 (Figure 3; p = 0.265). However, for 2013, than king snake eel or red snapper (Figure 2; p = 0.001, p = average BaP metabolite concentration is two times higher in 0.001 respectively). Mean NPH metabolite concentration in golden tilefish and red snapper compared to king snake eel golden tilefish was, on average, four times higher than red (Figure 3; p = 0.003, p = 0.003 respectively), while golden tilefish and red snapper concentrations are statistically similar (p = 0.751). The uptake of PAHs by organisms is generally governed by a number of complex mechanisms including chemical bioavailability, species-specific metabolism, and biotransformation, diet, trophic level, and habitat use. Bioavailability is a function of both species-specific metabolic processes and uptake, as well as physicochemical properties. Molecular weight and Kow have been found to be negatively correlated with bioavailability of 58 individual PAHs. The less hydrophobic LMW PAHs (log Kow < 4) tend to dissolve more rapidly than high molecular weight (HMW) compounds (low Kow > 4), thereby increasing their bioavailability to marine organisms. Preferential uptake of LMW PAHs, by two species of benthic fish, has been documented and attributed to both the bioavailability of the compounds, as well as biotransformation mechanisms, such as higher biotransformation rates of HMW compounds compared − to LMW PAHs.59 61 However, in contrast, higher rate of μ −1 metabolism of LMW PAHs has also been found and attributed Figure 2. Biliary naphthalene metabolite concentrations ( gg ) for ff 62,63 golden tilefish (2012: n = 24; 2013: n = 72), red snapper (2011: n = to di erences in Kow. In addition, research suggests that the 30; 2012: n = 15; 2013: n = 63) and king snake eel (2012: n = 23; use of chemical dispersants, such as Corexit 9500, increases the uptake, bioavailability and bioconcentration of PAHs by 2013: n = 44), sampled in 2011, 2012, and 2013 in the northern Gulf − of Mexico. exposed fish.64 66 The use of chemical dispersants following

D DOI: 10.1021/acs.est.5b01870 Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article the DWH blowout, therefore, could have influenced both the metabolite concentration. Male red snapper were found to have bioavailability and potentially the toxicity of PAH to exposed higher biliary NPH metabolite concentrations than females (p fishes. = 0.021) with the mean concentrations of 61 μgg−1 and 44 μg The significant difference in NPH metabolite concentrations g−1, respectively. Differences in biliary PAH metabolite among the three fish species, which persists even when species concentration between sex has been documented and are sampled at the same station, indicates these differences are hypothesized to be due to sex hormone differences in PAH not a sampling artifact or due to spatial heterogeneity of metabolism (e.g., enzyme induction and activity), and exposure. The much higher concentrations of NPH metabolites regulation of monooxygenase activity by estrogens, leading to in golden tilefish as compared to the other two species is likely male fish having higher concentrations of biliary FACs.10,53,73 due to their burrowing lifestyle, incidental sediment ingestion, Larger king snake eel were found to have lower concentrations and their diet of benthic prey (e.g., benthic invertebrates and of both biliary NPH and BaP metabolites (NPH: r = −0.40, p = demersal fishes), which may result in additional routes of 0.001; BaP: r = −0.60, p = 0.002). A similar trend with length exposure to sedimented PAH contamination. As the solubility was seen for golden tilefish for BaP metabolites (r = −0.33, p = of PAHs decreases with increasing molecular weight, the 0.004). The weak but significant negative relationship between bioaccumulation potential for LMW compounds from sediment biliary BaP and fish length for golden tilefish suggests diet may tends to be greater than the HMW compounds.59 Previous play a larger role in exposure. Golden tilefish exhibit an studies on organic contaminants also demonstrated negative ontogenetic shift in diet, from benthic invertebrates, to a higher correlations between biota-sediment accumulation factors proportion of fish, which may be a mechanism of changing fi (BSAF) and log Kow, suggesting decreasing bioavailability PAH exposure with sh length, as invertebrates have lower 67 ffi with increasing log Kow. Furthermore, HMW compounds can metabolism and elimination e ciencies for PAHs compared to be physically bulky limiting their ability to pass through lipid vertebrates. Very little is known about king snake eel, therefore, membrane barriers, resulting in lower body burdens. The more their life history, physiology or possibly the same ontogenetic hydrophobic HMW PAHs bind more readily with particulates, shift in diet may explain the similar trend, however, no diet dissolved organic material (DOM) and oil droplets, also studies have been published for the species. hindering their bioavailability to marine organisms. Recent Temporal Variation in Biliary PAH Metabolite studies of the DWH blowout document a hydrocarbon- Concentration. Between 2012 and 2013, there was no contaminated marine snow event and found evidence of rapid significant change in golden tilefish NPH metabolite concen- deposition of associated oil-particle aggregates and degraded oil tration (Figure 2; p = 0.367), with the mean concentration to the seabed.3,5,68 When dissecting golden tilefish, it is evident declining 8% from 240 μgg−1 to 220 μgg−1. However, there that they ingest large volumes of sediment, seen in the digestive was a significant increase in golden tilefish BaP metabolite tract, buccal cavity and gills, while obvious sediment ingestion is concentration over the two years (Figure 3; p = 0.025), with not observed in the king snake eel or red snapper digestive the mean concentration increasing from 170 ng g−1 to 370 ng tracts (personal observation). Together the observed hydro- g−1. Between 2012 and 2013, the mean concentration of NPH carbon-contaminated marine snow and degraded oil deposition metabolites in king snake eel declined by 37%, from 38 μgg−1 on the seafloor, differences in bioavailability, and species- to 24 μgg−1, although this difference was not significant specific metabolism and uptake help to explain the higher levels (Figure 2; p = 0.063). There was, however, a significant of NPH observed in golden tilefish. decrease in king snake eel BaP metabolite concentration over We hypothesize king snake eel may have lower biliary PAH the two years (Figure 3; p = 0.025), decreasing 38% (260 ng metabolite concentrations than the other species analyzed due g−1 to 160 ng g−1). to their inordinate epidermal mucus production, inefficiency of Over the three-year period following the DWH event, there metabolism or peculiarities of their life history. King snake eel was a significant decrease in mean red snapper NPH metabolite may eliminate LMW PAHs through mucus, as they produce concentration (Figure 2; p = 0.001). Between 2011 and 2012, prodigious amounts relative to the other species (personal there was a significant 49% decrease from a mean concentration observation). Exposure studies have documented high levels of of 120 μgg−1 to 61 μgg−1 (p = 0.003). Between 2012 and NPH and metabolites in the epidermal mucus of both rainbow 2013, there was a continued decrease in the mean trout (Oncorhynchus mykiss) and starry flounder (Platichthys concentration, from 61 μgg−1 to 51 μgg−1, although the stellatus), exposed to [3H]naphthalene, suggesting epidermal difference between 2012 and 2013 was nonsignificant (p = mucus is an important pathway of LMW PAH excretion from 0.194). Red snapper BaP metabolites remained at similar the body.69,70 Alternatively, mucus may provide a physical concentrations over the three years (Figure 3; p = 0.282). barrier against dermal uptake. King snake eel may also be less The significant declines in NPH metabolite concentration in efficient at metabolizing PAHs, although, hepatic ethoxyresor- red snapper is indicative of an episodic exposure event to NPH ufin-O-deethylase (EROD) activity, a biomarker of exposure to prior to 2011.13 A similar, although nonsignificant, decrease in organic contaminants, in eel of another family, the European eel king snake eel NPH metabolite concentration between 2012 (Anguilla anguilla), was found to be comparable to a teleost, and 2013 also suggests episodic exposure to elevated PAHs. For European flounder (Platichthys flesus), suggesting that eels can both species, these data support a scenario of increased NPH metabolize PAHs.71 King snake eel life history, although contamination in the environment coincident with the DWH unknown, may play a role in contaminant exposure and blowout and a decrease in exposure reflected in decreasing metabolism, as other Ophichthid eels (Myrophis punctatus) have biliary NPH metabolite concentration in following years. In complex life history and sexual maturation, which may influence comparison, the persistent, and significantly higher, concen- enzymatic activity.71,72 trations of biliary NPH metabolites in golden tilefish suggests Fish length, weight, and sex have been previously correlated that the source of NPH for golden tilefish did not decrease over with biliary PAH concentrations.52,53 In this study, fish sex and time, since biliary PAH metabolites indicate short-term length were examined as factors influencing biliary PAH exposure to PAHs.28,31 If PAHs exist in sediments associated

E DOI: 10.1021/acs.est.5b01870 Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article with golden tilefish burrows and are being sequestered by among the most contaminated fish, only falling below pink continued sedimentation, the digging behavior of golden tilefish salmon (Oncorhynchus gorbuscha, 480 μgg−1) sampled may re-expose these animals to PAHs, while other co-occurring immediately after the Exxon Valdez oil spill and an assortment species may not be exposed to the same routes and thus the of species from a polluted channel (Saõ Sebastiaõ channel, 290 same effective exposure levels. Crescent gunnel (Pholis laeta), a μgg−1)inSaõ Paulo, Brazil, that serves as the largest petroleum demersal fish, collected from sites polluted by the Exxon Valdez terminal in that country, frequently experiencing oil spills and oil spill, showed elevated concentrations of LMW biliary PAH discharge.56,76 Golden tilefish NPH metabolite concentrations metabolites 10 years after the event, indicating that oil can are about 1.5 times higher than those reported from inshore persist in the environment resulting in persistently elevated fish (primarily Atlantic croaker (Micropogonias undulatus)) levels of biliary PAH metabolites.74 The combination of caught in the northern GoM both before and after Hurricane sediment enrichment from the hydrocarbon-contaminated Katrina (190 μgg−1, 150 μgg−1, respectively) and 6.5 times marine snow event following the DWH and the slow higher than white sturgeon (Ancipenser transmontanus,32μg dissolution rates of HMW PAHs from oil droplets, particulates g−1) caught downstream of the Columbia River oil spill.76,77 and DOM, help to explain the increasing concentration of Red snapper and king snake eel biliary NPH metabolite biliary BaP metabolites for golden tilefish, as the bioavailability concentrations (120 μgg−1 (from 2011), 38 μgg−1 (from of PAHs in oil droplets or DOM increases over time.75 2012), respectively) comparatively rank much lower, closer to Spatial Variation in Biliary PAH Metabolite previous GoM data sampled from an assortment of species Concentration. Spatial variability in biliary PAH metabolite offshore of Texas in 1993 (110 μgg−1), with king snake eel concentration was evaluated by comparing results for the concentrations being similar to the relatively unpolluted site in northern GoM to the WFS for red snapper sampled in 2013 the Columbia River.77,80 For biliary BaP metabolites, all three (Figure 1). Red snapper caught on the WFS had significantly species have comparatively low concentrations (170−280 ng lower biliary PAH metabolites, for both NPH (p = 0.001) and g−1), which are very similar to the 1993 GoM data (200 ng BaP(p = 0.001), compared to red snapper caught in the g−1), and much lower than the highly polluted estuaries (580− northern GoM, closer to the Mississippi River, extant oil 2900 ng g−1) and inshore fish samples taken before and after infrastructure and the DWH event (Figure 4). Additionally, the Hurricane Katrina (1400−1600 ng g−1).76,77,79,80 Comparisons of PAH data from these three species post- DWH reveal that the level of LMW PAH exposure was extensive, particularly for golden tilefish, even two years and more following the event. While the levels of LMW PAHs in golden tilefish prior to the DWH are unknown, the 1993 study off of Texas, which included anchor tilefish (Caulolatilus intermedius), had a much lower average of LMW PAHs (116 μg g−1 for NPH), and perhaps indicated the residual pollution levels in areas where oil and gas rigs are located.80 In contrast, the BaP metabolite concentrations in our study were lower than most other studies. Benzo[a]pyrene is primarily derived of from combusted hydrocarbons (e.g., from urban and industrial Figure 4. Comparison of red snapper biliary polycyclic aromatic − pollution), thus, the relatively high levels of BaP in studies of hydrocarbon metabolite concentration for naphthalene (μgg1) and polluted estuaries, and inshore GoM, serve more a function of −1 benzo[a]pyrene (ng g ) from two regional groups sampled in 2013, depth and distance from shore, compared to the offshore West Florida Shelf (WFS, n = 14) and northern Gulf of Mexico locations sampled in this study.76,78 While large quantities of (nGoM, n = 49). For both NPH and BaP metabolite concentrations, p DWH oil were burned at sea, it is unknown what portion of the = 0.001. oil resulted in BaP contamination, although the quantity of BaP in DWH source oil was very low, if present at all.11,12 northern GoM stations showed a decrease in biliary PAH The data synthesized in this study constitute one of the concentration over time, for both NPH and BaP metabolites, largest biliary PAH data sets for fishes and the largest for the suggesting episodic exposure to an elevated source of PAHs, GoM. Nearly 300 bile samples were analyzed from three GoM whereas biliary PAH concentrations from the Madison− demersal fish species over three years following the DWH Swanson fishery closed area on the WFS have not decreased blowout. Significant interspecies differences exist between the between years. Therefore, these stations on the WFS may three species in concentrations of LMW biliary PAH represent baseline biliary PAH metabolite concentration for red metabolites. Golden tilefish exhibit the highest known snapper in the GoM. No significant spatial trends existed for concentrations of biliary NPH metabolites of data available golden tilefish and king snake eel biliary PAH data, as these on GoM fishes, and were among the highest concentrations in species were not caught in large numbers on the WFS. comparable studies globally. Differences in habitat, physiology, Comparison to Historical Biliary PAH Data. We and diet most likely account for the observed differences in conducted a metadata analysis of previous studies that used LMW PAH exposure, however, differences in bioavailability, similar HPLC-F methods to quantify biliary PAH metabolite and species-specific rates of uptake, metabolism, and elimi- concentrations in fish, including studies of three other oil spills, nation, all play a role. While king snake eel may directly three previous GoM studies, six polluted estuaries, and one encounter polluted sediments as well, their physiology may “pristine” site upstream of the 1984 Columbia River oil result in avoidance (e.g., mucus barrier) and elimination of − spill51,56,57,76 80 (Figure 5). In comparison to all of these, ingested PAHs, although, this is conjecture. Red snapper, in golden tilefish biliary NPH metabolite concentrations (240 μg contrast, are not as intimately coupled with sediments, leading g−1), collected two years hence of the DWH blowout, were to respiration and diet as the more likely exposure routes, with

F DOI: 10.1021/acs.est.5b01870 Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article

Figure 5. Comparison of biliary naphthalene (left, thin white bars), phenanthrene (left, thin black bars) and benzo[a]pyrene (right) metabolite concentration between post-Deepwater Horizon golden tilefish (n = 24) sampled in the northern Gulf of Mexico (GoM) in 2012, red snapper (n = 30) sampled in 2011 and king snake eel (n = 23) sampled in 2012, to other oil spills, polluted estuaries and one pristine site on the Columbia River. Data from this study are denoted with stars. species-specific metabolism and excretion most likely lending to should document return to baseline conditions in areas not the lower biliary PAH concentrations. To definitively under- subjected to persistent levels of PAH contamination. stand these species-specificdifferences, further analysis of muscle and liver tissue from the field-caught individuals in this ■ ASSOCIATED CONTENT study, will be analyzed for PAHs and alkylated homologues to *S Supporting Information better understand the distribution of PAHs between bile, liver, A summary table of the information conveyed in Figure 5. The and muscle, and to better understand PAH source. Controlled Supporting Information is available free of charge on the ACS exposure studies could also assist in understanding these Publications website at DOI: 10.1021/acs.est.5b01870. mechanisms. Temporal trends were observed for interspecies variation of ■ AUTHOR INFORMATION biliary NPH metabolite concentrations. The high concen- Corresponding Author trations of biliary NPH persisted over time in golden tilefish, *E-mail: [email protected]. whereas they declined in red snapper and king snake eel. The Author Contributions fi statistically signi cant, exponential decrease over time of biliary The manuscript was written through contributions of all NPH metabolites in red snapper suggests exposure to LMW authors. All authors have given approval to the final version of PAH pollution from an episodic event likely occurred in the the manuscript. 13 GoM prior to 2011. The same, albeit lower, decrease over Notes time was seen in NPH and BaP biliary PAH metabolite The authors declare no competing financial interest. concentration for king snake eel samples. It is possible (likely) that the episodic event was the DWH blowout, since it is ■ ACKNOWLEDGMENTS unlikely that the other sources of PAHs to the GoM would We thank the owners, captains and crew of the F/V Pisces and decrease substantially (e.g., 50%) in the same region at the 13 the R/V Weatherbird II; the field team, including E. Herdter, A. same time. Concentrations of biliary PAH metabolites were fi fi Wallace, K. Deak, S. Gilbert, S. Grasty, and the shermen; the signi cant higher in the northern GoM, closer to the DWH Murawski Lab, G. Ylitalo and B. Anulacion for their guidance event and Mississippi River, compared to the WFS, for 2013 and the 2011 bile analysis at the NOAA Northwest Fisheries red snapper samples. The declining concentrations of PAHs Science Center; and I. Romero, P. Schwing, D. Hollander and with distance from the DWH well site is consistent with this K. Able for their collaboration. This research was made possible event being the source of elevated PAHs in red snapper in part by grants from BP/The Gulf of Mexico Research observed. Initiative, through its Center for Integrated Modeling and While the collection and analysis of biliary PAHs in fish to Analysis of Gulf Ecosystems (C-IMAGE), the State of evaluate temporal trends in contamination reveals much about Louisiana, and NOAA Grant NA11NMF4720151-Systematic the source and magnitude of pollution in the GoM, these Survey of Finfish Disease Prevalence in the Gulf of Mexico. studies would have been aided by the availability of pre-DWH baseline data. The lack of such baseline complicates but does ■ ABBREVIATIONS not obviate the assessment of PAH pollution source and DWH Deepwater Horizon magnitude. Spatially relevant, precise and replicated baseline GoM Gulf of Mexico data would have been useful in detecting the levels of exposure PAH polycyclic aromatic hydrocarbons of GoM fishes. Nevertheless, temporal changes in contaminant NPH naphthalene levels reveal much about the ephemeral and more persistent PHN phenanthrene levels of contamination, as suggested by long-term studies of BaP benzo[a]pyrene the Exxon Valdez oil spill. Continued monitoring of GoM fishes FAC fluorescent aromatic hydrocarbons

G DOI: 10.1021/acs.est.5b01870 Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article

HPLC-F high performance liquid chromatography with during embryonic development alters cardiac morphology and reduces fluorescence detection aerobic capacity in adult fish. Proc. Natl. Acad. Sci. U.S.A. 2011, 108 WFS West Florida Shelf (17), 7086−7090. NWFSC Northwest Fisheries Science Center (15) Heintz, R. A. Chronic exposure to polynuclear aromatic MML Mote Marine Laboratory hydrocarbons in natal habitats leads to decreased equilibrium size, growth, and stability of pink salmon populations. Integr. Environ. Assess. LMW low molecular weight Manage. 2007, 3 (3), 351−363. HMW high molecular weight (16) White, P. A.; Robitaille, S.; Rasmussen, J. B. Heritable BSAF biota-sediment accumulation factor reproductive effects of benzo a pyrene on the fathead minnow − ffi Kow octanol water partition coe cient (Pimephales promelas). Environ. Toxicol. Chem. 1999, 18 (8), 1843− DOM dissolved organic matter 1847. (17) Brette, F.; Machado, B.; Cros, C.; Incardona, J. P.; Scholz, N. L.; ■ REFERENCES Block, B. A. Crude oil impairs cardiac excitation-contraction coupling in fish. Science 2014, 343 (6172), 772−776. (1) McNutt, M. K.; Chu, S.; Lubchenco, J.; Hunter, T.; Dreyfus, G.; (18) Collier, T. K.; Stein, J. E.; Goksoyr, A.; Myers, M. S.; Gooch, J. Murawski, S. A.; Kennedy, D. M. Applications of science and W.; Huggett, R. J.; Varanasi, U. Biomarkers of PAH exposure in oyster engineering to quantify and control the Deepwater Horizon oil spill. toadfish (Opsanus tau) from the Elizabeth River, Virginia. Environ. Sci. Proc. Natl. Acad. Sci. U.S.A. 2012, 109 (50), 20222−20228. 1993 − (2) Schwing, P. T.; Romero, I. C.; Brooks, G. R.; Hastings, D. W.; , 2 (3), 161 177. Larson, R. A.; Hollander, D. J. A Decline in Benthic Foraminifera (19) Nichols, J. W.; McKim, J. M.; Lien, G. J.; Hoffman, A. D.; following the Deepwater Horizon Event in the Northeastern Gulf of Bertelsen, S. L.; Elonen, C. M. A physiologically based toxicokinetic model for dermal absorption of organic chemicals by fish. Fund. Appl. Mexico. PLoS One 2015, 10 (3), e0120565. − (3) Hastings, D. W.; Schwing, P. T.; Brooks, G. R.; Larson, R. A.; Toxicol. 1996, 31 (2), 229 242. (20) Law, R. J.; Hellou, J. Contamination of fish and shellfish Morford, J. L.; Roeder, T.; Quinn, K. A.; Bartlett, T.; Romero, I. C.; − Hollander, D. J., Changes in sediment redox conditions following the following oil spill incidents. Environ. Geosci. 1999, 6 (2), 90 98. BP DWH blowout event. Deep Sea Res., Part II 2015, 10.1016/ (21) Krahn, M. M.; Rhodes, L. D.; Myers, M. S.; Moore, L. K.; j.dsr2.2014.12.009. Macleod, W. D.; Malins, D. C. Associations between metabolites of (4) Valentine, D. L.; Fisher, G. B.; Bagby, S. C.; Nelson, R. K.; Reddy, aromatic compounds in bile and the occurence of hepatic lesions in English sole (Parophrys vetulus) from Puget Sound, Washington. Arch. C. M.; Sylva, S. P.; Woo, M. A. Fallout plume of submerged oil from − Deepwater Horizon. Proc. Natl. Acad. Sci. U.S.A. 2014, 111 (45), Environ. Contam. Toxicol. 1986, 15 (1), 61 67. 15906−15911. (22) Myers, M. S.; Johnson, L. L.; Collier, T. K. Establishing the (5) Romero, I. C.; Schwing, P. T.; Brooks, G. R.; Larson, R. A.; causal relationship between polycyclic aromatic hydrocarbon (PAH) Hastings, D. W.; Ellis, G.; Goddard, E. A.; Hollander, D. J. exposure and hepatic neoplasms and neoplasia-related liver lesions in English sole (Pleuronectes vetulus). Hum. Ecol. Risk Assess. 2003, 9 (1), Hydrocarbons in deep-sea sediments following the 2010 Deepwater − Horizon blowout in the northeast Gulf of Mexico. PLoS One 2015, 10 67 94. (5), e0128371. (23) Malins, D. C.; McCain, B. B.; Brown, D. W.; Chan, S. L.; Myers, (6) Lin, Q.; Mendelssohn, I. A. Impacts and recovery of the M. S.; Landahl, J. T.; Prohaska, P. G.; Friedman, A. J.; Rhodes, L. D.; Deepwater Horizon oil spill on vegetation structure and function of Burrows, D. G.; Gronlund, W. D.; Hodgins, H. O. Chemical pollutants coastal salt marshes in the northern Gulf of Mexico. Environ. Sci. in sediments and diseases of bottom-dwelling fish in Puget Sound, − Technol. 2012, 46 (7), 3737−3743. Washington. Environ. Sci. Technol. 1984, 18 (9), 705 713. (7) Michel, J.; Owens, E. H.; Zengel, S.; Graham, A.; Nixon, Z.; (24) Malins, D. C.; McCain, B. B.; Brown, D. W.; Varanasi, U.; Allard, T.; Holton, W.; Reimer, P. D.; Lamarche, A.; White, M.; Krahn, M. M.; Myers, M. S.; Chan, S. L. Sediment associated Rutherford, N.; Childs, C.; Mauseth, G.; Challenger, G.; Taylor, E. contaminants and liver diseases in bottom-dwelling fish. Hydrobiologia − Extent and degree of shoreline oiling: Deepwater Horizon Oil Spill, Gulf 1987, 149,67 74. of Mexico, USA. PLoS One 2013, 8 (6), e65087. (25) Lee, R. F.; Sauerheb, R.; Dobbs, G. H. Uptake, metabolism and (8) Silliman, B. R.; van de Koppel, J.; McCoy, M. W.; Diller, J.; discharge of polycyclic aromatic hydrocarbons by marine fish. Mar. − Kasozi, G. N.; Earl, K.; Adams, P. N.; Zimmerman, A. R. Degradation Biol. 1972, 17 (3), 201 &. and resilience in Louisiana salt marshes after the BP−Deepwater (26) Varanasi, U.; Stein, J. E.; Nishimoto, M., Biotransformation and Horizon oil spill. Proc. Natl. Acad. Sci. U. S. A. 2012, 109 (28), 11234− disposition of polycyclic aromatic hydrocarbons (PAH) in fish. In 11239. Metabolism of Polycylic Aromatic Hydrocarbons in the Aquatic (9) Hylland, K. Polycyclic aromatic hydrocarbon (PAH) ecotoxicol- Environment; Varanasi, U., Ed.; CRC Press, Inc.: Boca Raton, FL, ogy in marine ecosystems. J. Toxicol. Environ. Health, Part A 2006, 69 1989; pp 94−143. (1−2), 109−123. (27) Ylitalo, G. M.; Krahn, M. M.; Dickhoff, W. W.; Stein, J. E.; (10)Tuvikene,A.Responsesoffishtopolycyclicaromatic Walker, C. C.; Lassitter, C. L.; Garrett, E. S.; Desfosse, L. L.; Mitchell, hydrocarbons (PAHs). Ann. Zool. Fenn. 1995, 32 (3), 295−309. K. M.; Noble, B. T.; Wilson, S.; Beck, N. B.; Benner, R. A.; (11) BP Gulf Coast Restoration Organization: Gulf Science Data, Koufopoulos, P. N.; Dickey, R. W. Federal seafood safety response to 2014. http://gulfsciencedata.bp.com/. the Deepwater Horizon oil spill. Proc. Natl. Acad. Sci. U.S.A. 2012, 109 (12) Reddy, C. M.; Arey, J. S.; Seewald, J. S.; Sylva, S. P.; Lemkau, K. (50), 20274−20279. L.; Nelson, R. K.; Carmichael, C. A.; McIntyre, C. P.; Fenwick, J.; (28) Beyer, J.; Jonsson, G.; Porte, C.; Krahn, M. M.; Ariese, F. Ventura, G. T.; Van Mooy, B. A. S.; Camilli, R. Composition and fate Analytical methods for determining metabolites of polycyclic aromatic of gas and oil released to the water column during the Deepwater hydrocarbon (PAH) pollutants in fish bile: A review. Environ. Toxicol. Horizon oil spill. Proc. Natl. Acad. Sci. U.S.A. 2012, 109 (50), 20229− Pharmacol. 2010, 30 (3), 224−244. 20234. (29) Krahn, M. M.; Myers, M. S.; Burrows, D. G.; Malins, D. C. (13) Murawski, S. A.; Hogarth, W. T.; Peebles, E. B.; Barbeiri, L. Determination of metabolites of xenobiotics in the bile of fish from Prevalence of external skin lesions and polycyclic aromatic hydro- polluted waterways. Xenobiotica 1984, 14 (8), 633−646. carbon concentrations in Gulf of Mexico Fishes, post-Deepwater (30) Krahn, M. M.; Burrows, D. G.; Ylitalo, G. M.; Brown, D. W.; Horizon. Trans. Am. Fish. Soc. 2014, 143 (4), 1084−1097. Wigren, C. A.; Collier, T. K.; Chan, S. L.; Varanasi, U. Mass (14) Hicken, C. E.; Linbo, T. L.; Baldwin, D. H.; Willis, M. L.; Myers, spectrometric analysis for aromatic compounds in bile of fish sampled M. S.; Holland, L.; Larsen, M.; Stekoll, M. S.; Rice, S. D.; Collier, T. after the Exxon Valdez oil spill. Environ. Sci. Technol. 1992, 26 (1), K.; Scholz, N. L.; Incardona, J. P. Sublethal exposure to crude oil 116−126.

H DOI: 10.1021/acs.est.5b01870 Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article

(31) Le Du-Lacoste, M.; Akcha, F.; Devier, M.-H.; Morin, B.; Northeast Region, Northwest Fisheries Science Center, 1986; pp 1− Burgeot, T.; Budzinski, H. Comparative study of different exposure 25. routes on the biotransformation and genotoxicity of PAHs in the (49) Johnson, L. L.; Ylitalo, G. M.; Myers, M. S.; Anulacion, B. F.; flatfish species, Scophthalmus maximus. Environ. Sci. Pollut. Res. 2013, Buzitis, J.; Collier, T. K. Aluminum smelter-derived polycyclic aromatic 20 (2), 690−707. hydrocarbons and flatfish health in the Kitimat marine ecosystem, (32) Meador, J. P.; Buzitis, J.; Bravo, C. F. Using fluorescent aromatic British Columbia, Canada. Sci. Total Environ. 2015, 512, 227−239. compounds in bile from juvenile salmonids to predict exposure to (50) Krahn, M. M.; Burrows, D. G.; Macleod, W. D.; Malins, D. C. polycyclic aromatic hydrocarbons. Environ. Toxicol. Chem. 2008, 27 Determination of individual metabolites of aromatic compounds in (4), 845−853. hydrolyzed bile of English sole (Parophrys vetulus) from polluted sites (33) Gallaway, B. J.; Szedlmayer, S. T.; Gazey, W. J. A life history in Puget Sound, Washington. Arch. Environ. Contam. Toxicol. 1987, 16 review for red snapper in the Gulf of Mexico with an evaluation of the (5), 511−522. importance of offshore petroleum platforms and other artificial reefs. (51) Krahn, M. M.; Ylitalo, G. M.; Buzitis, J.; Bolton, J. L.; Wigren, C. Rev. Fish. Sci. 2009, 17 (1), 48−67. A.; Chan, S. L.; Varanasi, U. Analyses for petroleum related (34) Able, K. W.; Grimes, C. B.; Cooper, R. A.; Uzmann, J. R. Burrow contaminants in marine fish and sediments following the Gulf oil construction and behavior of tilefish, Lopholatilus chamaeleonticeps, spill. Mar. Pollut. Bull. 1993, 27, 285−292. in Hudson Submarine Canyon. Environ. Biol. Fishes 1982, 7 (3), 199− (52) Tomy, G. T.; Halldorson, T.; Chemomas, G.; Bestvater, L.; 205. Danegerfield, K.; Ward, T.; Pleskach, K.; Stern, G.; Atchison, S.; (35) Clark, S. T. Age, growth, and distributions of the giant snake eel, Majewski, A.; Reist, J. D.; Palace, V. P. Polycyclic aromatic Ophichthus rex, in the Gulf of Mexico. Bull. Mar. Sci. 2000, 67 (3), hydrocarbon metabolites in Arctic Cod (Boreogadus saida) from the 911−922. Beaufort Sea and associative fish health effects. Environ. Sci. Technol. (36) Levengood, J. M.; Schaeffer, D. J. Polycyclic aromatic 2014, 48 (19), 11629−11636. hydrocarbons in fish and crayfish from the Calumet region of (53) Vuorinen, P. J.; Keinanen, M.; Vuontisjarvi, H.; Barsiene, J.; southwestern Lake Michigan. Ecotoxicology 2011, 20 (6), 1411−1421. Broeg, K.; Forlin, L.; Gercken, J.; Kopecka, J.; Koehler, A.; Parkkonen, (37) Twichell, D. C.; Grimes, C. B.; Jones, R. S.; Able, K. W. The role J.; Pempkowiak, J.; Schiedek, D. Use of biliary PAH metabolites as a of erosion by fish in shaping topography around Hudson Submarine biomarker of pollution in fish from the Baltic Sea. Mar. Pollut. Bull. Canyon. J. Sediment. Petrol. 1985, 55 (5), 712−719. 2006, 53 (8−9), 479−487. (38) Grimes, C. B.; Turner, S. C. The complex life history of tilefish (54) Anderson, M. J. A new method for non-parametric multivariate Lopholatilus chamaeleonticeps and vulnerability to exploitation. Am. analysis of variance. Austral Ecol. 2001, 26 (1), 32−46. Fish. Soc. Symp. 1999, 23,17−26. (55) Manly, B. F. J.; Manly, B. F. J. Randomization and Monte Carlo (39) Katz, S. J.; Grimes, C. B.; Able, K. W. Delineation of tilefish, Methods in Biology, 1991; Vol. i-xiii,pp1−281. Lopholatilus chamaeleonticeps, stocks along the United States east coast (56) da Silva, D. A. M.; Buzitis, J.; Krahn, M. M.; Bicego, M. C.; and in the Gulf of Mexico. U.S. Fish Wildl. Serv. Fish. Bull. 1983, 81 Pires-Vanin, A. M. S. Metabolites in bile of fish from Sao Sebastiao (1), 41−50. Channel, Sao Paulo, Brazil as biomarkers of exposure to petrogenic (40) Grimes, C. B.; Able, K. W.; Jones, R. S. Tilefish, Lopholatilus polycyclic aromatic compounds. Mar. Pollut. Bull. 2006, 52 (2), 175− chamaeleonticeps, habitat, behaviro and community structure in Mid- 183. Atlantic and Southern New England waters. Environ. Bio. Fishes. 1986, (57) Hom, T.; Varanasi, U.; Stein, J. E.; Sloan, C. A.; Tilbury, K. L.; 15 (4), 273−292. Chan, S.-L. Assessment of the exposure of subsistence fish to aromatic (41) Burgess, G. H. Biological, Fishery, and Product Assessments of the compounds after the Exxon Valdez oil spill. Am. Fish. Soc. Symp. 1996, Keoghfish, an Underutilized and Unmanaged Gulf of Mexico Resource; 18, 856−866. University of Florida: Floirda Museum of Natural History. (58) Wang, H.-S.; Man, Y.-B.; Wu, F.-Y.; Zhao, Y.-G.; Wong, C. K. (42) Bozzano, A. Vision in the rufus snake eel, Ophichthus rufus: C.; Wong, M.-H. Oral bioaccessibility of polycyclic aromatic Adaptive mechanisms for a burrowing life-style. Marine Biology 2003, hydrocarbons (PAHs) through fish consumption, based on an in 143 (1), 167−174. vitro digestion model. J. Agric. Food Chem. 2010, 58 (21), 11517− (43) Chakrabarty, P.; Lam, C.; Hardman, J.; Aaronson, J.; House, P. 11524. H.; Janies, D. A. SM: A web-based application for visualizing the (59) Baumard, P.; Budzinski, H.; Garrigues, P.; Sorbe, J. C.; Burgeot, overlap of distributions and pollution events, with a list of fishes put at T.; Bellocq, J. Concentrations of PAHs (polycyclic aromatic risk by the 2010 Gulf of Mexico oil spill. Biodiversity Conserv. 2012, 21 hydrocarbons) in various marine organisms in relation to those in (7), 1865−1876. sediments and to trophic level. Mar. Pollut. Bull. 1998, 36 (12), 951− (44) McCawley, J. R.; Cowan, J. H., Jr., Seasonal and size specific diet 960. and prey demand of red snapper on Alabama artificial reefs. In (60) Varanasi, U.; Gmur, D. J. Hydrocarbons and metabolites in American Fisheries Society Symposium; Patterson, W. F., Cowan, J. H., English sole (Parophrys vetulus) exposed simultaneously to [3H]- Fitzhugh, G. R., Nieland, D. L., Eds. ; 2007; Vol. 60,pp77−104. benzo[a]pyrene and [14C]naphthalene in oil-contaminated sediment. (45) Steimle, F. W. Essential fish habitat source document. Tilefish, Aquat. Toxicol. 1981, 1 (1), 49−67. Lopholatilus chamaeleonticeps, life history and habitat characteristics. U.S. (61) Nichols, J. W.; Hoffman, A. D.; ter Laak, T. L.; Fitzsimmons, P. Dept. of Commerce, National Oceanic and Atmospheric Admin- N. Hepatic clearance of 6 polycyclic aromatic hydrocarbons by isolated istration, National Marine Fisheries Service, Northeast Region, perfused trout livers: Prediction from in vitro clearance by liver s9 Northeast Fisheries Science Center ; Available from the National fractions. Toxicol. Sci. 2013, 136 (2), 359−372. Technical Information Service: Woods Hole, MA, 1999. (62) Burkhard, L. P.; Lukasewycz, M. T. Some bioaccumulation (46) Hellou, J.; Leonard, J.; Collier, T. K.; Ariese, F. Assessing PAH factors and biota-sediment accumulation factors for polycyclic exposure in feral finfish from the Northwest Atlantic. Mar. Pollut. Bull. aromatic hydrocarbons in lake trout. Environ. Toxicol. Chem. 2000, 2006, 52 (4), 433−441. 19 (5), 1427−1429. (47) Krahn, M. M.; Ylitalo, G. M.; Buzitis, J.; Chan, S. L.; Varanasi, U. (63) Jonsson, G.; Bechmann, R. K.; Bamber, S. D.; Baussant, T. Rapid high-performance liquid chromatographic methods that screen Bioconcentration, biotransformation, and elimination of polycyclic for aromatic compounds in environmental samples. J. Chromatogr. aromatic hydrocarbons in sheepshead minnows (Cyprinodon varie- 1993, 642 (1−2), 15−32. gatus) exposed to contaminated seawater. Environ. Toxicol. Chem. (48) Krahn, M. M.; Moore, L. K.; Macleod, W. D. Standard Analytical 2004, 23 (6), 1538−1548. Procedures of the NOAA National Analytical Facility, 1986: Metabolites (64) Ramachandran, S. D.; Hodson, P. V.; Khan, C. W.; Lee, K. Oil of Aromatic Compounds in Fish Bile, NMFS-NWC-102; U.S. Depart- dispersant increases PAH uptake by fish exposed to crude oil. ment of Commerce, N. O. a. A. A., National Marine Fisheries Service, Ecotoxicol. Environ. Saf. 2004, 59 (3), 300−308.

I DOI: 10.1021/acs.est.5b01870 Environ. Sci. Technol. XXXX, XXX, XXX−XXX Environmental Science & Technology Article

(65) Couillard, C. M.; Lee, K.; Legare, B.; King, T. L. Effect of dispersant on the composition of the water-accommodated fraction of crude oil and its toxicity to larval marine fish. Environ. Toxicol. Chem. 2005, 24 (6), 1496−1504. (66) Milinkovitch, T.; Kanan, R.; Thomas-Guyon, H.; Le Floch, S. Effects of dispersed oil exposure on the bioaccumulation of polycyclic aromatic hydrocarbons and the mortality of juvenile Liza ramada. Sci. Total Environ. 2011, 409 (9), 1643−1650. (67) Maruya, K. A.; Lee, R. E. Biota-sediment accumulation and trophic transfer factors for extremely hydrophobic polychlorinated biphenyls. Environ. Toxicol. Chem. 1998, 17 (12), 2463−2469. (68) Passow, U.; Ziervogel, K.; Asper, V.; Diercks, A., Marine snow formation in the aftermath of the Deepwater Horizon oil spill in the Gulf of Mexico. Environ. Res. Lett. 2012, 7, (3). (69) Varanasi, U.; Gmur, D. J. Metabolisms and disposition of naphthalene in Starry flounder (Platichthys stellatus). Fed. Proc. 1978, 37 (3), 813−813. (70) Varanasi, U.; Uhler, M.; Stranahan, S. I. Uptake and release of napthalene and its metabolites in skin and epidermal mucus of salmonids. Toxicol. Appl. Pharmacol. 1978, 44 (2), 277−289. (71) Rotchell, J. M.; Bird, D. J.; Newton, L. C. Seasonal variation in ethoxyresorufin O-deethylase (EROD) activity in European eels Anguilla anguilla and flounders Pleuronectes flesus from the Severn Estuary and Bristol Channel. Mar. Ecol.: Prog. Ser. 1999, 190, 263− 270. (72) Able, K. W.; Allen, D. M.; Bath-Martin, G.; Hare, J. A.; Hoss, D. E.; Marancik, K. E.; Powles, P. M.; Richardson, D. E.; Taylor, J. C.; Walsh, H. J.; Warlen, S. M.; Wenner, C. Life history and habitat use of the speckled worm eel, Myrophis punctatus, along the east coast of the United States. Environ. Biol. Fishes 2011, 92 (2), 237−259. (73) Navas, J. M.; Segner, H. Modulation of trout 7-ethoxyresorufin- O-deethylase (EROD) activity by estradiol and octylphenol. Mar. Environ. Res. 2000, 50 (1−5), 157−162. (74) Jewett, S. C.; Dean, T. A.; Woodin, B. R.; Hoberg, M. K.; Stegeman, J. J. Exposure to hydrocarbons 10 years after the Exxon Valdez oil spill: Evidence from cytochrome P4501A expression and biliary FACs in nearshore demersal fishes. Mar. Environ. Res. 2002, 54 (1), 21−48. (75) Baussant, T.; Sanni, S.; Jonsson, G.; Skadsheim, A.; Borseth, J. F. Bioaccumulation of polycyclic aromatic compounds: 1. Bioconcentra- tion in two marine species and in semipermeable membrane devices during chronic exposure to dispersed crude oil. Environ. Toxicol. Chem. 2001, 20 (6), 1175−1184. (76) Hom, T.; Collier, T. K.; Krahn, M. M.; Strom, M. S.; Ylitalo, G. M.; Nilsson, W. B.; Papunjpye, R. N.; Varanasi, U., Assessing seafood safety, in the aftermath of Hurricane Katrina. In American Fisheries Society Symposium; McLaughlin, K. D., Ed., 2008; Vol. 64,pp73−93. (77) Krahn, M. M.; Kittle, L. J.; Macleod, W. D. Evidence for exposure of fish to oil spilled into the Columbia River. Mar. Environ. Res. 1986, 20 (4), 291−298. (78) Myers, M. S.; Stehr, C. M.; Olson, O. P.; Johnson, L. L.; McCain, B. B.; Chan, S. L.; Varanasi, U. Relationships between toxicopathic hepatic lesions and exposure to chemical contaminants in English sole (Pleuronectes vetulus), Starry flounder (Platichthys stellatus), and White croaker (Genyonemus lineatus) from selected marine sites on the Pacific Coast, USA. Environ. Health Perspect. 1994, 102 (2), 200−215. (79) Johnson, L. L.; Ylitalo, G. M.; Arkoosh, M. R.; Kagley, A. N.; Stafford, C.; Bolton, J. L.; Buzitis, J.; Anulacion, B. F.; Collier, T. K. Contaminant exposure in outmigrant juvenile salmon from Pacific Northwest estuaries of the United States. Environ. Monit. Assess. 2007, 124 (1−3), 167−194. (80) McDonald, S. J.; Willett, K. L.; Thomsen, J.; Beatty, K. B.; Connor, K.; Narasimhan, T. R.; Erickson, C. M.; Safe, S. H. Sublethal detoxification responses to contaminant exposure associated with offshore production platforms. Can. J. Fish. Aquat. Sci. 1996, 53 (11), 2606−2617.

J DOI: 10.1021/acs.est.5b01870 Environ. Sci. Technol. XXXX, XXX, XXX−XXX Marine Pollution Bulletin 98 (2015) 259–266

Contents lists available at ScienceDirect

Marine Pollution Bulletin

journal homepage: www.elsevier.com/locate/marpolbul

Exposure to Deepwater Horizon weathered crude oil increases routine metabolic demand in chub mackerel, Scomber japonicus ⇑ Dane H. Klinger a, , Jonathan J. Dale b, Benjamin E. Machado b, John P. Incardona c, Charles J. Farwell d, Barbara A. Block b a Emmett Interdisciplinary Program in Environment and Resources, Stanford University, Stanford, CA 94305, USA b Biology Department, Hopkins Marine Station, Stanford University, Paciﬁc Grove, CA 93950, USA c Northwest Fisheries Science Center, National Oceanic and Atmospheric Administration, Seattle, WA 98112, USA d Tuna Research and Conservation Center, Monterey Bay Aquarium, Monterey, CA 93940, USA article info abstract

Article history: During the 2010 Deepwater Horizon incident, the continuous release of crude oil from the damaged Received 7 November 2014 Macondo 252 wellhead on the ocean floor contaminated surface water habitats for pelagic fish for more Revised 18 June 2015 than 12 weeks. The spill occurred across pelagic, neritic and benthic waters, impacting a variety of Accepted 23 June 2015 ecosystems. Chemical components of crude oil are known to disrupt cardiac function in juvenile fish, Available online 22 July 2015 and here we investigate the effects of oil on the routine metabolic rate of chub mackerel, Scomber japonicus. Mackerel were exposed to artificially weathered Macondo 252 crude oil, prepared as a Water Keywords: Accommodated Fraction (WAF), for 72 or 96 h. Routine metabolic rates were determined pre- and Oil spill post-exposure using an intermittent-flow, swim tunnel respirometer. Routine energetic demand PAHs Aerobic metabolism increased in all mackerels in response to crude oil and reached statistical significance relative to unex- Scomber japonicus posed controls at 96 h. Chemical analyses of bile from exposed fish revealed elevated levels of fluorescent WAF metabolites, confirming the bioavailability of polycyclic aromatic hydrocarbons (PAHs) in the exposure Biliary metabolites WAF. The observed increase in metabolic demand is likely attributable to the bioenergetic costs of contaminant detoxification. These results indicate that short-term exposure (i.e. days) to oil has sub-lethal toxicity to mackerel and results in physiological stress during the active spill phase of the incident. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction estimated to be in the billions of dollars over the next decade (Sumaila et al., 2012; Upton, 2011). Understanding how oil impacts The 2010 Deepwater Horizon (DWH) disaster released an esti- the physiology of juvenile and adult pelagic fish is critical for dis- mated 780 million liters of crude oil into the northern Gulf of cerning population impacts. Mexico over 87 days (McNutt et al., 2012), making it the largest In addition to the more easily observed histopathological accidental oil spill in history (Eckle et al., 2012). During the DWH impacts on aquatic organisms (e.g. integumental lesions; Hargis event, crude oil released from the seafloor rose up through the et al., 1984), crude oil contains numerous chemical components slope waters and spread throughout mesoscale eddies in the east- that are toxic when dissolved in the water column. Of these, poly- ern Gulf of Mexico, creating a relatively prolonged exposure win- cyclic aromatic hydrocarbons (PAHs) are the most prevalent and dow in the pelagic zone, where many fish species utilize habitat extensively studied, primarily because they are relatively water in the mixed layer. The Gulf of Mexico has a large and diverse soluble and thus available for uptake by oil-exposed organisms assemblage of pelagic fishes, with representatives from nearly all (NRC, 2003). Metabolism of PAHs varies by compound, concentra- of the major taxonomic families including mackerels, tunas, mar- tion, exposure duration, organism, life-stage, and tissue type, but lins, swordfish, sunfish, and carangids (Hoese and Moore, 1998). many metabolites of PAHs are rapidly excreted into the urine Thus, economic losses due to closures of recreational and commer- and secreted into the bile of juvenile and adult fish (Collier et al., cial fisheries and negative impacts on fish populations are 2014; Tierney et al., 2013). Decades of studies have shown that PAHs cause a range of sub-lethal effects in fish, including neoplasia, immunotoxicity, ⇑ Corresponding author at: Department of Ecology and Evolutionary Biology, Princeton University, Princeton, NJ 08544, USA. reduced growth and condition, and reduced reproductive success E-mail address: [email protected] (D.H. Klinger). (as reviewed in Collier et al., 2014). Crude oil-derived PAHs are also http://dx.doi.org/10.1016/j.marpolbul.2015.06.039 0025-326X/Ó 2015 Elsevier Ltd. All rights reserved. 260 D.H. Klinger et al. / Marine Pollution Bulletin 98 (2015) 259–266 directly toxic to the fish heart, as documented for the developing embryos of zebrafish (Carls et al., 2008; de Soysa et al., 2012; Hicken et al., 2011; Incardona et al., 2013), herring (Hose et al., 1996; Incardona et al., 2009), flounder (Collier et al., 2014), and tunas (Incardona et al., 2014), as well as in cardiomyocytes isolated from the hearts of juvenile bluefina and yellowfin tunas (Brette et al., 2014). In the case of growth, sublethal toxicity at the scale Davison et al. (1992) dos Santos et al. (2006) Davoodi and Claireaux (2007) Claireaux and Davoodi (2010) Christiansen et al. (2010) Prasad (1987) Milinkovitch et al. (2012a) Mager et al. (2014) This Study of individual fish can reduce the intrinsic growth and abundance Davison et al. (1993) Cohen et al. (2001) of wild populations (Spromberg and Meador, 2005, 2006). Whole-organism bioenergetics, typically measured using respirometry, provides an integrative physiological measure of sub-lethal toxicity (McKenzie et al., 2007; Whitehead, 2013). Previous respirometry studies on PAH-exposed teleosts have yielded varying results, with reported increases, decreases, and no change in metabolic rates (Table 1). However, direct comparisons across respiration studies are confounded by differences in focal species, PAH exposure protocols, water temperatures, and Metabolic rate change References Increased MR (4 days) respirometry protocols. Previous studies have generally focused decreased MR at low DO on fish from freshwater, estuarine, and benthic marine habitats, ) 1 where oil spills are common. Much less is known about the À sub-lethal impacts of PAHs on pelagic fish. gL l In this study we examine the impacts of oil on chub mackerel PAH low (

(Scomber japonicus) as a representative of the scombrid fishes ) 1 (tunas, mackerels, and bonitos). Scombridae are pelagic predators À gL l 47.5 3.5 No change in MR (3 days), with unique physiological specializations and life history traits PAH high ( that potentially increase their oil exposure risk, including relatively high rates of activity, mobility, and endurance. They swim continuously and with their mouths open, allowing the forward motion of the body to create a continuous flow of water across the gills (obligate ram ventilation; Magnuson, 1979). To meet a high oxygen Stirring – WSF n/r n/r Increased MR Stirring – WSF n/r n/rWAF No change in MR at normal DO, 30 4.2 No change in MR Mixing method demand, scombrids require a robust respiratory system and a well-developed circulatory system with specializations for rapid oxygen delivery. These include gills with extraordinarily large surface areas and thin membranes, a high mitochondrial content in the muscle, visceral, and gill tissues, and increased cardiac function relative to many other teleost fish (Block and Stevens, 2001; Galli et al., 2011). Large gill surface area could increase the risk of oil 0.125 oil: 2500 water WAF 1 fuel: 200 water1 fuel: 200 watern/r None None 0.039 0.039 0.039 WSF 0.039 No change SMR, Decreased AMR Decreased MR 40 4.3 Decreased MR water water dilutions 20 g oil: 300 L water Multiple 3.3 0.5 No change in MR impairing ventilation and associated metabolic processes. 1 oil: 9 water Stirring – WAF 384 68 No change in MR Among scombrids, chub mackerel is a favorable model species for assessing metabolic rate because they are relatively easy to acquire and maintain in captivity (Mendiola et al., 2008). Accordingly, they have been used successfully in prior flume and respiration studies (e.g. Beamish, 1984; Dickson et al., 2002; weathered oil No. 2 No. 2 North Sea petroleum oil Additive Concentration Donley and Dickson, 2000; Sepulveda and Dickson, 2000). They crude oil are also commercially important and made up 15% of global Scrombidae landings in 2013 (FAO, 2015). Chub mackerel live in subtropical waters (10–27 °C) of the Pacific and Indian Oceans (Fishbase, 2014b). Evolutionarily, they are closely related to 4 weeks Exposure duration Atlantic chub mackerel (Scomber colias) from the pelagic–neritic waters of the Gulf of Mexico and the Mediterranean Sea (Catanese et al., 2010; Cheng et al., 2011; Collette et al., 2001; Fishbase, 2014a). Here, we exposed chub mackerel to weathered 1.15, 1.77 24 h, 12 days Naphthalene n/r n/r274 300 3 and 4 days 150 DWH Increased MR mass (g) MC252 crude oil at a concentration of 0.125:2500 oil:seawater to 34.4 48 h Arabian crude evaluate the impact of PAH exposure on pelagic predator bioenergetics.

2. Materials and methods Bald notothen 50Florida pompano 2–72 h Diesel fuelBald oil notothen 100% 75.9 WSF of 1 oil: 9 Mahi-mahi 7 days 0.404–0.812mackerel Diesel 24 fuel h oil 33% of a WSF of 1 oil: 2 DWH slick oil 0.4%, 1.2%, and 2% WAF Common name Mean ﬁsh 2.1. Experimental organisms Australian bass 6.5mullet 3 days Bass Strait

Chub mackerel (mean ± standard deviation: fork length = 28.4 ± 1.3 cm, weight = 273.8 ± 45.7 g) were collected in 18–21 °C surface waters off the southern coastline of California, in the California Current Large Marine Ecosystem. At this length and borchgrevinki carolinus borchgrevinki hippurus novemaculeata Pagothenia Trachinotus Solea soleaSolea soleaBoreogadus saida Common sole PolarPentius cod sophore Common 146 solePagothenia 1020 Minor carp 30 5 days 2.68 5 daysCoryphaena 60 Heavy min, Scomber fuel japonicus 10–168 h Heavy Pacific fuel chub Crude oil 0.2–2.5 oil: 1000 water Shaking n/r n/r Decreased MR Species weight, fish were sub-adult or mature (Hernandez and Ortega, Macquaria Liza aurata Golden gray Table 1 2000). Mackerel were transported by truck to the Tuna Research Respiration studies on themetabolic effects rate, of WAF exposure = to water oil accommodated fraction, on WSF the = metabolic water rates soluble of fraction. teleost fishes. AMR = active metabolic rate, DO = dissolved oxygen, DWH = Deepwater Horizon, n/r = not reported, MR = metabolic rate, SMR = standard D.H. Klinger et al. / Marine Pollution Bulletin 98 (2015) 259–266 261 and Conservation Center (Hopkins Marine Station, Stanford University, Pacific Grove, CA). The fish were acclimated for at least 1 month, at 20 °C, in a 22,300 L tank with a turn-over rate of one volume per hour. Fish which were successfully trained to swim in a respirometer (see below for respirometer protocol) were moved to a 3780 L holding tank with a turnover rate of one volume per hour. Fish were held under natural photoperiod, with a dim moon-light over the tanks which gave enough light for mackerel to see the tank walls. Mackerel were fed a mixture of euphausiids and chopped market squid (Loligo opalescens) and sardines (Sardinops sagax) three times per week. Temperature was maintained at 20 °Cin all tanks throughout the experiment.

2.2. Preparation of Water Accommodated Fraction (WAF)

WAFs were prepared in a 57 L stainless steel mixing cask with an electric motor (GE 3583, 1/4 hp, 115 VAC, Fairfield, CT) run at Fig. 2. Mean oxygen consumption rate, ± standard deviation, of individual Pacific low speed for four hours (design specifics were modified from mackerel exposed to oil for 72 and 96 h. Barron et al., 2003). An 18 inch stainless steel drywall paddle was attached in series to the motor through a ½ in. drill chuck (Jacobs Multi CraftÒ,½Â 20 in. thread, Sparks, MD) and ½ in. water samples were taken every 24 h until completion of the expo- motor arbor (1/2 Â 20 in.). The motorized mixing apparatus was sure protocol. Subsurface water samples (550 ml) were collected affixed to the stainless steel lid by a wooden stand, with the shaft by submerging a closed separatory funnel into the exposure tank of the drywall paddle running through a hole drilled into the lid. and opening the stopcock. The water sample was divided into Prior to each preparation, the stainless steel cask and drywall pad- two equal aliquots of 250 ml. One aliquot was filtered through dle were cleaned with three rinses each of acetone and then 2.7 and 0.7 lm glass microfiber filters (GF/D and GF/F, respectively, dichloromethane and thoroughly wiped down with paper towels. Whatman/GE Healthcare, Piscataway, NJ) held in a glass filter Exposures were run at a nominal concentration of 25 ppb total holder apparatus (VWR International, Radnor, PA). The filtered PAHs in water. For each WAF preparation, 40 L of filtered seawater and unfiltered water samples were then stored at 4 °C for water at 20 °C was siphoned from the holding tank directly into the mix- chemistry analysis. Water quality in the exposure tank was also ing cask. 125 ml of DWH-MC252 artificially weathered oil was monitored during oil exposures. Parameters included temperature measured in a dedicated 150 ml glass graduated cylinder and then and dissolved oxygen (YSI ProODO™ probe, Yellow Springs, OH), combined with the seawater in the mixing cask. Artificial weather- pH (Litmus paper strips, Cole Parmer, Vernon Hills, IL), salinity ing included gently mixing the sample while heating it to 90– (refractometer, Grainger Industrial, Salinas, CA) and ammonia Ò 105 °C to reduce the mass by 33–38% (Incardona et al., 2014). (Hach Ammonia test kit, model NI-SA, Loveland, CO). Upon completion of mixing in the cask, the WAF preparation was Temperature, dissolved oxygen and pH were all measured on a poured into a recirculating exposure tank, which was identical in daily basis, while salinity and ammonia were measured at the design and volume to the holding tank. beginning and end of each exposure.

2.3. Water sampling and water quality analysis 2.4. Exposure protocol

Water samples were taken from the exposure tank starting Prior to each oil exposure, an initial seawater rinse of 1000 L 20 min after introduction of the WAF preparation. Subsequent was run through the exposure tank’s recirculating system to flush out any standing water. Following removal of the rinse water, 2500 L of seawater was pumped into the exposure tank, which was maintained at a temperature of 20 ± 0.2 °C with a dual heater (Titanium I10 heater, GLO-QUARTZ, Mentor, OH) and chiller system (Cyclone chiller, Aqua-Logic, San Diego, CA). The WAF was added to the exposure tank and allowed to mix for 20 min. Fish were then collected from the holding tank and placed in the exposure tank. For each experiment, two flume-trained fish (see respirometer protocol) were exposed simultaneously in the experimental regime, one fish for 72 h and the other for 96 h. The tank was covered with a clear, plastic sheet for the duration of the exposures. After each exposure duration, one fish was removed from the exposure tank and transferred to the respirometer.

2.5. Respirometer protocol

For conducting the respiration trials, a 10 L intermittent flow, Fig. 1. Example of baseline routine and post-exposure respiration trials for an swim tunnel respirometer (Loligo, Denmark) was used to measure individual mackerel. This individual was exposed to oil for 96 h. The dashed black metabolic rates both before and after exposure to oil. The line marks the end 4.5 h period upon entry to the respirometer when the fish was allowed to recover from the stress of handling. Data during this period was not respirometer was supplied with seawater from a 1000 L reservoir. included in the analysis. Dissolved oxygen concentration and temperature was measured 262 D.H. Klinger et al. / Marine Pollution Bulletin 98 (2015) 259–266 with a fiberoptic oxygen dipping probe and temperature probe, 3. Results respectively (Presens, Germany), and recorded with Oxyview data acquisition software (Presens, Germany). Fish were introduced into the oil exposed seawater tank envi- Measured metabolic rates are often impacted by a fish’s lack of ronment that was visibly brown in color and turbid. The volatile familiarity with swimming in a flume, requiring acclimation and fumes around the tank were noxious to humans without respira- training prior to experimental trials (Blank et al., 2007a,b). An tory masks, and once fish were inside the tank they were difficult extensive training regime was conducted with all mackerel prior to locate. When visually observed, paired fish swam together to exposure trials to establish the routine metabolic rates within within the tank. the flume environment for each individual. Training included an The fish were exposed in the tank environment to a total initial phase where each individual mackerel was introduced into dissolved (filtered) PAH concentration that declined over the the respirometer and allowed to swim at 1 body length per second course of the exposure trials from a high concentration of for at least 4 h. During this initial training period, fish were manu- 47.5 ± 13.7 lglÀ1 (mean ± standard deviation) at the start of the ally assisted as necessary to ensure that they did not repeatedly hit experimental procedure to a low of 3.5 ± 1.3 lglÀ1 at 72 h, and or rest on the back grate of the respirometer, possibly resulting in 3.7 ± 1.1 lglÀ1 at 96 h (Fig. 4). The PAH concentration decline in injury to their caudal fins. This is a common problem when mack- the tank was greatest within the first 24 h and concentrations erel are first introduced to the respirometer, and many individual remained relatively stable for the remainder of the exposure. The fish had to be trained several times before they were able to swim total unfiltered PAH concentration went from 114.3 ± 41.4 lglÀ1 consistently for over 24 h. at the start of exposure trials to 91.6 ± 31.2 lglÀ1 after 72 h and Once fish successfully passed the training exercises, they were 71.0 ± 26.9 lglÀ1 after 96 h (Fig. 5). The unfiltered PAH concentra- returned to the holding tank for at least 72 h. For pre-exposure tions remained relatively stable for 48 h and then began to decline. routine metabolic rate (baseline) trials, fish were fasted for at least The variation in unfiltered concentrations between experiments 48 h and then introduced into the respirometer for 24 h at a speed was greater than for the filtered concentrations. Salinity, pH, and of 1 body length per second. Before oxygen respiration measure- ammonia remained constant and within biologically optimal ments began, fish were allowed to acclimate to the respirometer ranges throughout all experiments. for 4.5 h to allow the effects of stress of handling and excitation from introduction to the flume to subside. After the pre-exposure trial, fish were again returned to the holding tank for at least 72 h. Fish were then randomly assigned exposure durations and the respiration protocol was repeated post-exposure (see Fig. 1). Differences in pre- and post-exposure routine metabolic rates at each experimental duration were tested using paired t-tests.

2.6. Bile protocol

To assess the internal dose of oil compounds in exposed fish, PAH metabolites were measured in the bile to provide a relative measure of PAH uptake in addition to the nominal exposure concentration in the surrounding water (Beyer et al., 2010). Bile was collected from oil-exposed fish post-experiment and after eutha- nization. As a control, bile was also collected from 3 trained mackerel after swimming in the exposure tank without oil and then being subjected to the respirometer protocol. To collect bile, fish were euthanized by pithing and then the body cavity was opened with a clean scalpel. Using a new scalpel Fig. 3. Mean oxygen consumption rate, ± standard deviation, for pre- and post-oil exposures. Exposure durations were 72 and 96 h. An asterisk (⁄) indicates a blade, the internal organs were severed at the esophagus and the significant difference between pre- and post-exposure metabolic rates. entire internal organ mass was removed and placed on aluminum foil. The gall bladder was identified as the elongated green sac-like organ, and if there was blood on the outside of the gall bladder it was rinsed with distilled water from a squirt bottle. The gall bladder was gripped at the bile duct and then gently separated from the liver with a scalpel. To collect bile from the gall bladder the blind end was positioned over the mouth of an amber vial and punctured with a fine scalpel. Bile was allowed to drip into the vial, and if more was needed than dripped out, the gall bladder was dipped into the vial and the blunt side of the scalpel blade was placed against it and the vial mouth and the gall bladder was pulled up so as to force the remaining liquid into the vial. The vial was placed immediately into a À20 °C freezer. All tools were rinsed with deionized water and then with isopropanol, and scalpel blades were discarded. The bile samples were sent to the NOAA laboratory in Seattle, WA on ice, and metabolites of biologically pertinent oil compounds (phenanthrene, benzo(a)pyrene, naphthalene and pyrene) were analyzed along with total protein concentration, following protocols described in (Yanagida et al., 2012). Fig. 4. Mean ± standard deviation of total dissolved (filtered) PAH concentrations. D.H. Klinger et al. / Marine Pollution Bulletin 98 (2015) 259–266 263

compared to the control mackerel (Fig. 6). On average metabolites of phenanthrene increased from 1.73 mg/ml in the control ﬁsh to 1133 mg/ml in ﬁsh exposed to oil for 72 h, benzo(a)pyrene increased from 0.05 mg/ml to 6.9 mg/ml, naphthalene increased from 7.03 mg/ml to 3200 mg/ml, and pyrene increased from 0.12 mg/ml to 52 mg/ml. The mackerel exposed for 96 h also had increased concentrations of metabolites of phenanthrene, benzo(a)pyrene, naphthalene and pyrene at 293 mg/ml, 1.9 mg/ml, 970 mg/ml, and 14 mg/ml, respectively.

4. Discussion

This study differs from other respiration exposure studies in that it: (1) exposes sub-adults and adults of a pelagic species for multiple days, (2) employs a respirometry protocol that reduces confounding factors such as SDA, stress from handling, Fig. 5. Mean ± standard deviation of unfiltered PAH concentrations. and stress from unfamiliarity with the respirometer, and (3) uses a modified method of fractionating oil into seawater at large volumes. Mackerel exposures were conducted to discern the Pairs of trained mackerel were exposed to oil in four trials. whole-animal metabolic response and potential for injury upon Mean pre-exposure routine metabolic rate for mackerels was swimming through a patch of pelagic ocean habitat contaminated À1 À1 167 ± 22 mg O2 kg h (N = 8 fish) when swimming at 1 body with oil droplets. Notably, mackerel were capable of swimming in length per second at 20 °C. In all cases, there was an increase in oil over a period of 72–96 h. During this interval, fish schooled routine metabolic rates following oil exposure (Fig. 2). Exposure with conspecifics and were successful in negotiating the 2.5 m to oil for 96 h resulted in a statistically significant increase in meta- diameter tank despite turbid conditions. Thus, the basic swim- À1 À1 bolic rates (mean post-exposure RMR = 249 ± 69 mg O2 kg h , ming behavior of these obligate ram ventilators was not evidently Paired T-test, T-value = 3.21, P = 0.049). Oxygen consumption also impaired at the concentrations of oil tested. When examined over increased in 72 h exposures, but the increase was not statistically 72 h, there was a trend toward higher metabolic rates, but this À1 À1 significant (mean post-exposure RMR = 224 ± 56 mg O2 kg h , was not significantly different from controls. For the longer 96 h Paired t-test, T-value = 2.77, P = 0.070) (Fig. 3). exposure interval, however, the animals exhibited an increase in The bile of mackerel that were exposed for 72 h showed the oxygen consumption that was significantly elevated from routine greatest increase in each of the four PAH metabolites analyzed, metabolic rate.

Fig. 6. Mean ± standard deviation of biliary concentrations of metabolites of four PAHs, (A) phenanthrene, (B) benzo(a)pyrene, (C) naphthalene, and (D) pyrene, in control mackerel and mackerel exposed to oil for 72 and 96 h. 264 D.H. Klinger et al. / Marine Pollution Bulletin 98 (2015) 259–266

The increase in metabolism at 96 h was most likely due to a ( Liza aurata) exposed to lower PAH concentrations (0.5–3.3 lgLÀ1 physiological stress response as well as an induction of xenobiotic relative to 3.5–47.5 lgLÀ1 here) for a shorter period of time (48 h metabolizing enzymes. The stress response is associated with relative to 72 and 96 h) showed no significant change in metabolic increases in circulating catecholamine and cortisol, which in turn rates (Milinkovitch et al., 2012a). The higher concentration of PAHs increase energy production via increases in plasma glucose and longer exposure time in our study could explain the differing and lactate (George et al., 2013). Catecholamines bind to results. Australian bass (Macquaria novemaculeata) exposed to b-adrenoreceptors, amplifying oxygen uptake pathways and rate higher concentrations of PAHs (68–384 lgLÀ1 relative to 3.5– of delivery (Randall and Perry, 1992). The aryl hydrocarbon recep- 47.5 lgLÀ1 here) showed no change in metabolic rate, but fish tor (AHR) signaling pathway is responsible for induction of xenobi- were only allowed to acclimate in the respirometer for 15 min, otic metabolizing enzymes such as CYP1a (Kim et al., 2013). increasing the likelihood that stress from handling affected the Upregulation of these metabolizing enzymes and associated ATP results (Cohen et al., 2001). Mackerel in our study were trained demands for detoxification would also result in the increased in multiple trials and then acclimated to the respirometer for metabolic rates observed in our study. Given that stress may have 4.5 h before final routine measurement, thereby minimizing any increased with the longer duration exposures, future experiments influence of handling stress. Young of the year mahi-mahi should consider monitoring cortisol in tandem with metabolism. (Coryphaena hippurus) exposed to oil and similar concentrations The 24 h period in which some fish were alone in the exposure of PAHs showed no change in standard or active metabolic rates tank, following removal of the 72 h exposure fish, could have also (Mager et al., 2014), but mahi were only exposed to oil for 24 h contributed to elevated stress. (relative to 72 or 96 h here), meaning the shorter exposure time could explain the difference in results. 4.1. Comparison with previous studies 4.2. Bile The routine metabolic rates measured for unexposed mackerels À1 À1 in this study (mean = 167 ± 22 mg O2 kg h ) are similar to those Each of the four PAH metabolites measured in the bile of mack- reported for mackerel in previous studies conducted at similar erel increased in the oil exposure fish, but concentrations were 70% temperatures (Dickson et al., 2002; Shadwick and Steffensen, lower in fish exposed for 96 h relative to fish exposed for 72 h. This 2000). Fish behavior and swimming in both the exposure tank is consistent with the decline in dissolved waterborne PAH concen- and the respirometer did not appear abnormal or labored. trations during the exposure, and indicates that PAHs taken up Prior studies show increases, decreases, and no change in meta- during the initial exposure period induced a metabolic response. bolic rates as a result of exposure to oil and PAHs (Table 1). As Biliary PAH concentrations also increased in golden gray mullet observed here for mackerel, PAH exposures have previously been exposed to oil (Milinkovitch et al., 2012b). shown to increase metabolic rates in the polar Antarctic bald notothen (Pagothenia borchgrevinki) as well as the tropical Florida 4.3. Dissociation of oil in seawater using the Water Accommodated pompano (Trachinotus carolinus). Bald notothen were exposed to Fraction (WAF) method much higher concentrations of oil than used in the current study (277:2500 relative to 0.125:2500 oil:seawater here) although Oil is a complex chemical mixture containing hundreds or thou- exposure durations were similar (Davison et al., 1992). Florida sands of compounds. In order to expose aquatic organisms to the pompano were exposed to only one PAH, naphthalene, and at toxicological compounds contained within oil in an environmen- much higher concentrations (300–150 ppb naphthalene relative tally realistic exposure profile – one that mimicked, to the extent to 3–47 ppb total PAHs here) (dos Santos et al., 2006). possible, conditions during the DWH oil spill – it was necessary By contrast, several prior studies reported a decrease in meta- to thoroughly mix oil into water. The most established and consis- bolic rates after exposure to oil. For example, polar cod tent method for mixing oil into water is the creation of a WAF of oil (Boreogadus saida) exposed to similar concentrations of PAHs but (Incardona et al., 2013; Singer et al., 2000). This method blends oil for shorter and longer durations (1 h and 4 weeks, respectively) and water together at high speeds for a defined amount of time, showed decreased metabolic rates. However, control and exposed producing a thoroughly mixed sample of oil dissolved in a small fish were both fed prior to respirometry, and thus the measured amount of water (typically less than 4 L). The WAF can then be metabolic rates included the energetic cost of specific dynamic used to either expose an organism directly or be filtered or diluted action (SDA) (Christiansen et al., 2010). Exposure to PAHs may to a desired concentration. have impeded digestion or assimilation, resulting in decreased We modified an existing method for mixing larger volumes of SDA and associated metabolic rates relative to the control, which water (40 L) and oil together (from Barron et al., 2003), so that included unimpeded digestion and SDA. In our study, fish were the WAF contained oil droplets in the solution, similar to condi- fasted for both their baseline and exposure runs, eliminating the tions that fish might have encountered in pelagic habitats of the confounding effects of SDA. In one study where oil was simply Gulf of Mexico during the DWH oil spill (Adcroft et al., 2010; added to the surface of the water at much higher concentrations Diercks et al., 2010). This method also increases the scale of the (12.5:2500 oil:seawater relative to 0.125:2500 in our study), adult WAF, making it possible to expose larger juvenile and adult pelagic common sole (Solea solea) exposed in this way for 5 days showed a fish in greater volumes of seawater. decline or no change in metabolic rates (Claireaux and Davoodi, When measuring PAH levels in unfiltered and filtered seawater 2010; Davoodi and Claireaux, 2007). The lack of mixing and higher samples, the concentrations are consistently higher in unfiltered concentration of oil could have reduced oxygen diffusion across the samples because they contain the dissolved as well as the particu- gills, thereby limiting metabolic rates (Milinkovitch et al., 2012a). late fraction of PAHs. On a total mass basis, however, the filtered Similarly, minor carp (Pentius sophore) exposed to crude oil fraction is proportionally more toxic to fish because dissolved without thorough mixing exhibited decreased metabolic rates PAHs absorb through the gills more easily (McKim and Erickson, (Prasad, 1987). 1991). The particulate (droplet) PAHs are more likely to have a Other studies report no change in metabolic rates following physical effect – i.e., oiling the gills, thereby interfering with oxy- exposure to oil. Bald notothen exposed to oil for 7 days showed gen uptake (Claireaux and Davoodi, 2010). Both toxicological and no change in metabolic rate, but rates did decrease under low dis- physical effects of oil may have an impact on metabolism and thus solved oxygen conditions (Davison et al., 1993). Golden gray mullet both must be taken into consideration. D.H. Klinger et al. / Marine Pollution Bulletin 98 (2015) 259–266 265

The unfiltered PAH concentrations in the exposure tank varied Blank, J.M., Morrissette, J.M., Farwell, C.J., Price, M., Schallert, R.J., Block, B.A., 2007b. slightly between each experiment (Fig. 5). This variability is likely Temperature effects on metabolic rate of juvenile Pacific bluefin tuna Thunnus orientalis. J. Exp. Biol. 210 (23), 4254–4261. due to differences in the amount of oil that dissolves sufficiently in Block, B.A., Stevens, E.D. (Eds.), 2001. Tunas: Physiology, Ecology and Evolution. the seawater. After the WAF mixed for the allotted time (4 h) there Academic Press, San Diego, CA. were often droplets of oil on the sides of the stainless steel cask Brette, F., Machado, B., Cros, C., Incardona, J.P., Scholz, N.L., Block, B.A., 2014. Crude oil impairs cardiac excitation–contraction coupling in fish. Science 343 (6172), and top of the mixing vessel that did not go into solution. 772–776. Droplets also appeared on the walls of the exposure tank during Carls, M.G., Holland, L., Larsen, M., Collier, T.K., Scholz, N.L., Incardona, J.P., 2008. exposure trials. These globular droplets varied in size and shape Fish embryos are damaged by dissolved PAHs, not oil particles. Aquat. Toxicol. 88 (2), 121–127. and indicate that certain volumes of oil did not go into solution. Catanese, G., Manchado, M., Infante, C., 2010. Evolutionary relatedness of mackerels The filtered (dissolved) PAH concentrations were highly consis- of the genus Scomber based on complete mitochondrial genomes: strong tent across each of the experiments. This can be attributed to the support to the recognition of Atlantic Scomber colias and Pacific Scomber japonicus as distinct species. Gene 452 (1), 35–43. fact that the mechanical action of the stirring paddle is consistent Cheng, J., Gao, T., Miao, Z., Yanagimoto, T., 2011. Molecular phylogeny and evolution in dissociating similar amounts of dissolved PAHs in each experi- of Scomber (Teleostei: Scombridae) based on mitochondrial and nuclear DNA ment. While a certain amount of oil globules will stick to the sides sequences. Chin. J. Oceanol. Limnol. 29 (2), 297–310. of the mixing vessel, the globules that remain in the water are Christiansen, J., Karamushko, L., Nahrgang, J., 2010. Sub-lethal levels of waterborne petroleum may depress routine metabolism in polar cod Boreogadus saida more consistently dissociated during each WAF. The concentra- (Lepechin, 1774). Polar Biol. 33 (8), 1049–1055. tions of unfiltered and filtered toxic PAHs in the WAF mixture were Claireaux, G., Davoodi, F., 2010. Effect of exposure to petroleum hydrocarbons upon fairly consistent between trials, indicating that the modified WAF cardio-respiratory function in the common sole Solea solea. Aquat. Toxicol. 98 (2), 113–119. method is suitable for conducting replicate experiments. Cohen, A., Nugegoda, D., Gagnon, M.M., 2001. Metabolic responses of fish following exposure to two different oil spill remediation techniques. Ecotoxicol. Environ. Saf. 48 (3), 306–310. 5. Conclusions Collette, B.B., Reeb, C., Block, B.A., 2001. Fish Physiology: Tuna: Physiology, Ecology, and Evolution. In: Block, B.A., Stevens, E. (Eds.). Academic Press, pp. 1–33. Collier, T.K., Anulacion, B.F., Arkoosh, M.R., Dietrich, J.P., Incardona, J.P., Johnson, L.L., Experiments examining the effects of crude oil on mackerel Ylitalo, G.M., Myers, M.S., 2014. Fish Physiology. In: Tierney, K.B., Farrell, A.P., indicate oil exposure has a significant impact on metabolic rates Brauner, C.J. (Eds.). Academic Press, pp. 195–255. after four days. As an active, pelagic species, mackerel have Davison, W., Franklin, C.E., McKenzie, J.C., Dougan, M.C.R., 1992. The effects of acute exposure to the water soluble fraction of diesel fuel oil on survival and increased aerobic potential and tissues enriched in mitochondrial metabolic rate of an antarctic fish (Pagothenia borchgrevinki). Comp. Biochem. capacity. Their high movement and endurance ability provides a Physiol. Part C: Comp. Pharmacol. 102 (1), 185–188. mechanism through which they may be able to actively move Davison, W., Franklin, C.E., McKenzie, J.C., Carey, P.W., 1993. The effects of chronic away from contaminated regions. This study indicates short-term exposure to the water soluble fraction of fuel oil on an antarctic fish Pagothenia borchgkevinki. Comp. Biochem. Physiol. Part C: Comp. Pharmacol. 104 (1), 67– (i.e. days) exposure to oil has sub-lethal toxicity to chub mackerel, 70. increasing energy expenditures and consuming energy that could Davoodi, F., Claireaux, G., 2007. Effects of exposure to petroleum hydrocarbons otherwise be used for growth or reproduction. However, we cannot upon the metabolism of the common sole Solea solea. Mar. Pollut. Bull. 54 (7), 928–934. yet evaluate the long-term consequences of oil exposure on life- de Soysa, T.Y., Ulrich, A., Friedrich, T., Pite, D., Compton, S., Ok, D., Bernardos, R., time fitness. Longer post-exposure observation might yield more Downes, G., Hsieh, S., Stein, R., Lagdameo, M.C., Halvorsen, K., Kesich, L.-R., information about the potential effects of oil on survival and fit- Barresi, M., 2012. Macondo crude oil from the Deepwater Horizon oil spill disrupts specific developmental processes during zebrafish embryogenesis. ness. Oil exposure may have substantial effects on growth, fecun- BMC Biol. 10 (1), 40. dity or gamete production which would not be evident in this Dickson, K.A., Donley, J.M., Sepulveda, C., Bhoopat, L., 2002. Effects of temperature study. Additionally, aerobic scope (the metabolic range between on sustained swimming performance and swimming kinematics of the chub mackerel Scomber japonicus. J. Exp. Biol. 205 (7), 969–980. resting and maximal metabolic rates) may have been impacted, Diercks, A.-R., Highsmith, R.C., Asper, V.L., Joung, D., Zhou, Z., Guo, L., Shiller, A.M., limiting aerobic capacity for challenging behaviors such as preda- Joye, S.B., Teske, A.P., Guinasso, N., Wade, T.L., Lohrenz, S.E., 2010. tor avoidance, prey capture, and reproduction. Characterization of subsurface polycyclic aromatic hydrocarbons at the Deepwater Horizon site. Geophys. Res. Lett. 37 (20), L20602. Donley, J.M., Dickson, K.A., 2000. Swimming kinematics of juvenile kawakawa tuna Acknowledgements Euthynnus affinis and chub mackerel Scomber japonicus. J. Exp. Biol. 203 (20), 3103–3116. dos Santos, T.d.C.A., Ngan, P.V., de Arruda Campos Rocha Passos, M.J., Gomes, V., We thank N. Scholz of NOAA for advice on earlier draft of this 2006. Effects of naphthalene on metabolic rate and ammonia excretion of manuscript and D. Coffey and A. Norton of the Monterey Bay juvenile Florida pompano, Trachinotus carolinus. J. Exp. Mar. Biol. Ecol. 335 (1), Aquarium for fish husbandry during the experiment. This study 82–90. Eckle, P., Burgherr, P., Michaux, E., 2012. Risk of large oil spills: a statistical analysis was funded as part of the DWH–MC252 incident Natural in the aftermath of Deepwater Horizon. Environ. Sci. Technol. 46 (23), 13002– Resource Damage Assessment. All experiments were approved by 13008. Stanford University Institutional Animal Care and Use Committee FAO, 2015. Fishery and aquaculture statistics (global capture production 1950– 2012). FishStatJ, . procedures. Fishbase, 2014. Scomber colias. Froese, R., Pauly, D. (Eds.). Fishbase, 2014. Scomber japonicus. Froese, R., Pauly, D. (Eds.). Galli, G., Lipnick, M., Shiels, H., Block, B., 2011. Temperature effects on Ca2+ cycling References in scombrid cardiomyocytes: a phylogenetic comparison. J. Exp. Biol. 214, 1068–1076. Adcroft, A., Hallberg, R., Dunne, J.P., Samuels, B.L., Galt, J.A., Barker, C.H., Payton, D., George, N., Peter, V.S., Peter, M.C.S., 2013. Physiologic implications of inter- 2010. Simulations of underwater plumes of dissolved oil in the Gulf of Mexico. hormonal interference in fish: Lessons from the interaction of adrenaline with Geophys. Res. Lett. 37 (18), L18605. cortisol and thyroid hormones in climbing perch Anabas testudineus Bloch. Gen. Barron, M.G., Carls, M.G., Short, J.W., Rice, S.D., 2003. Photoenhanced toxicity of Comp. Endocrinol. 181, 122–129. aqueous phase and chemically dispersed weathered Alaska North Slope crude Hargis Jr, W.J., Roberts Jr, M.H., Zwerner, D.E., 1984. Effects of contaminated oil to Pacific herring eggs and larvae. Environ. Toxicol. Chem. 22 (3), 650–660. sediments and sediment-exposed effluent water on an estuarine fish: acute Beamish, F.W.H., 1984. Swimming performance of three southwest Pacific fishes. toxicity. Mar. Environ. Res. 14 (1–4), 337–354. Mar. Biol. 79, 311–313. Hernandez, J.J.C., Ortega, A.T.S., 2000. Synopsis of Biological Data on the Chub Beyer, J., Jonsson, G., Porte, C., Krahn, M.M., Ariese, F., 2010. Analytical methods for Mackerel (Scomber japonicus Houttuyn, 1782). FAO Fisheries, p. 77. determining metabolites of polycyclic aromatic hydrocarbon (PAH) pollutants Hicken, C.E., Linbo, T.L., Baldwin, D.H., Willis, M.L., Myers, M.S., Holland, L., Larsen, in fish bile: a review. Environ. Toxicol. Pharmacol. 30 (3), 224–244. M., Stekoll, M.S., Rice, S.D., Collier, T.K., Scholz, N.L., Incardona, J.P., 2011. Blank, J.M., Farwell, Charles J., Morrissette, Jeffery M., Schallert, Robert J., Block, Sublethal exposure to crude oil during embryonic development alters cardiac Barbara A., 2007a. Influence of swimming speed on metabolic rates of juvenile morphology and reduces aerobic capacity in adult fish. Proc. Natl. Acad. Sci. 108 Pacific bluefin tuna and yellowfin tuna. Physiol. Biochem. Zool. 80 (2), 167–177. (17), 7086–7090. 266 D.H. Klinger et al. / Marine Pollution Bulletin 98 (2015) 259–266

Hoese, H.D., Moore, R.H., 1998. Fishes of the Gulf of Mexico: Texas, Louisiana, and Milinkovitch, T., Lucas, J., Le Floch, S., Thomas-Guyon, H., Lefrançois, C., 2012a. Effect Adjacent Waters. Texas A&M University Press, College Station, TX. of dispersed crude oil exposure upon the aerobic metabolic scope in juvenile Hose, J.E., McGurk, M.D., Marty, G.D., Hinton, D.E., Brown, E.D., Baker, T.T., 1996. golden grey mullet Liza aurata. Mar. Pollut. Bull. 64 (4), 865–871. Sublethal effects of the (Exxon Valdez) oil spill on herring embryos and larvae: Milinkovitch, T., Thomas-Guyon, H., Lefrançois, C., Imbert, N., 2012b. Dispersant use morphological, cytogenetic, and histopathological assessments, 1989–1991. as a response to oil spills: toxicological effects on fish cardiac performance. Fish Can. J. Fish. Aquat. Sci. 53 (10), 2355–2365. Physiol. Biochem., 1–6 Incardona, J.P., Carls, M.G., Day, H.L., Sloan, C.A., Bolton, J.L., Collier, T.K., Scholz, N.L., NRC, 2003. Oil in the Sea III: Inputs, Fates, and Effects. National Academies Press. 2009. Cardiac arrhythmia is the primary response of embryonic pacific herring Prasad, M.S., 1987. Toxicity of crude oil to the metabolism of freshwater minor carp (Clupea pallasi) exposed to crude oil during weathering. Environ. Sci. Technol. 43 Puntius sophore. Bull. Environ. Contam. Toxicol. 39 (2), 188–193. (1), 201–207. Randall, D., Perry, S., 1992. Fish Physiology. In: Hoar, W., Randall, D., Farrell, A. Incardona, J.P., Swarts, T.L., Edmunds, R.C., Linbo, T.L., Aquilina-Beck, A., Sloan, C.A., (Eds.). Academic Press, New York. Gardner, L.D., Block, B.A., Scholz, N.L., 2013. Exxon Valdez to Deepwater Sepulveda, C., Dickson, K.A., 2000. Maximum sustainable speeds and cost of Horizon: comparable toxicity of both crude oils to fish early life stages. Aquat. swimming in juvenile kawakawa tuna Euthynnus affinis and chub mackerel Toxicol. 142–143, 303–316. Scomber japonicus. J. Exp. Biol. 203 (20), 3089–3101. Incardona, J.P., Gardner, L.D., Linbo, T.L., Brown, T.L., Esbaugh, A.J., Mager, E.M., Shadwick, R.E., Steffensen, J.F., 2000. The cost of efficiency of aerobic locomotion in Stieglitz, J.D., French, B.L., Labenia, J.S., Laetz, C.A., Tagal, M., Sloan, C.A., Elizur, chub mackerel. Am. Zool. 40 (6), 1208. A., Benetti, D.D., Grosell, M., Block, B.A., Scholz, N.L., 2014. Deepwater Horizon Singer, M.M., Aurand, D., Bragin, G.E., Clark, J.R., Coelho, G.M., Sowby, M.L., crude oil impacts the developing hearts of large predatory pelagic fish. Proc. Tjeerdema, R.S., 2000. Standardization of the preparation and quantitation of Natl. Acad. Sci. 111 (15), E1510–E1518. water-accommodated fractions of petroleum for toxicity testing. Mar. Pollut. Kim, R.-O., Kim, B.-M., Hwang, D.-S., Au, D.W.T., Jung, J.-H., Shim, W.J., Leung, K.M.Y., Bull. 40 (11), 1007–1016. Wu, R.S.S., Rhee, J.-S., Lee, J.-S., 2013. Evaluation of biomarker potential of Spromberg, J.A., Meador, J.P., 2005. Relating results of chronic toxicity responses to cytochrome P450 1A (CYP1A) gene in the marine medaka, Oryzias melastigma population-level effects: modeling effects on wild chinook salmon populations. exposed to water-accommodated fractions (WAFs) of Iranian crude oil. Comp. Integr. Environ. Assess. Manage. 1 (1), 9–21. Biochem. Physiol. Part C: Toxicol. Pharmacol. 157 (2), 172–182. Spromberg, J.A., Meador, J.P., 2006. Relating chronic toxicity responses to Mager, E.M., Esbaugh, A.J., Stieglitz, J.D., Hoenig, R., Bodinier, C., Incardona, J.P., population-level effects: a comparison of population-level parameters for Scholz, N.L., Benetti, D.D., Grosell, M., 2014. Acute embryonic or juvenile three salmon species as a function of low-level toxicity. Ecol. Modell. 199 (3), exposure to Deepwater Horizon crude oil impairs the swimming performance 240–252. of mahi-mahi (Coryphaena hippurus). Environ. Sci. Technol. 48 (12), 7053–7061. Sumaila, U.R., Cisneros-Montemayor, Andrés M., Dyck, A., Huang, L., Cheung, W., Magnuson, J.J., 1979. Fish Physiology. In: Hoar, W.S., Randall, D.J. (Eds.). Academic Jacquet, J., Kleisner, K., Lam, V., McCrea-Strub, A., Swartz, W., Watson, R., Zeller, Press, pp. 239–313. D., Pauly, D., 2012. Impact of the Deepwater Horizon well blowout McKenzie, D.J., Garofalo, E., Winter, M.J., Ceradini, S., Verweij, F., Day, N., Hayes, R., on the economics of US Gulf fisheries. Can. J. Fish. Aquat. Sci. 69 (3), 499– van der Oost, R., Butler, P.J., Chipman, J.K., Taylor, E.W., 2007. Complex 510. physiological traits as biomarkers of the sub-lethal toxicological effects of Tierney, K.B., Kennedy, C.J., Gobas, F., Gledhill, M., Sekela, M., 2013. Fish Physiology. pollutant exposure in fishes. Philos. Trans. Roy. Soc. B: Biol. Sci. 362 (1487), In: Keith, B., Tierney, Farrell, A.P., Brauner, C.J. (Eds.). Academic Press, pp. 1–52. 2043–2059. Upton, H.F., 2011. The Deepwater Horizon oil spill and the Gulf of Mexico fishing McKim, J.M., Erickson, R.J., 1991. Environmental impacts on the physiological industry. Congr. Res. Serv., 7–5700 mechanisms controlling xenobiotic transfer across fish gills. Physiol. Zool. 64 Whitehead, A., 2013. Interactions between oil-spill pollutants and natural (1), 39–67. stressors can compound ecotoxicological effects. Integr. Comp. Biol. 53 (4), McNutt, M.K., Camilli, R., Crone, T.J., Guthrie, G.D., Hsieh, P.A., Ryerson, T.B., Savas, 635–647. O., Shaffer, F., 2012. Review of flow rate estimates of the Deepwater Horizon oil Yanagida, G., Anulacion, B., Bolton, J., Boyd, D., Lomax, D., Paul Olson, O., Sol, S., spill. Proc. Natl. Acad. Sci. 109 (50), 20260–20267. Willis, M., Ylitalo, G., Johnson, L., 2012. Polycyclic aromatic hydrocarbons and Mendiola, D., Yamashita, Y., Matsuyama, M., Alvarez, P., Tanaka, M., 2008. Scomber risk to threatened and endangered chinook salmon in the Lower Columbia River japonicus, H. is a better candidate species for juvenile production activities than Estuary. Arch. Environ. Contam. Toxicol. 62 (2), 282–295. Scomber scombrus, L. Aquacult. Res. 39 (10), 1122–1127. Journal of Marine Science and Engineering

Article Effects of Exposure of Pink Shrimp, Farfantepenaeus duorarum, Larvae to Macondo Canyon 252 Crude Oil and the Corexit Dispersant

Susan Laramore 1,*, William Krebs 1,2,† and Amber Garr 1,3,† 1 Harbor Branch Oceanographic Institute at Florida Atlantic University, 5600 US 1 North Fort Pierce, FL 34946, USA 2 Colorado Catch, LLC., PO Box 210, Sanford, CO 81151, USA; [email protected] 3 Fountain Valley School of Colorado, 6155 Fountain Valley School Road, Colorado Springs, CO 80911, USA; [email protected] * Correspondence: [email protected]; Tel.:+1-772-242-2525 † These authors contributed equally to this work.

Academic Editor: Merv Fingas Received: 1 January 2016; Accepted: 2 February 2016; Published: 8 March 2016 Abstract: The release of oil into the Gulf of Mexico (GOM) during the Deepwater Horizon event coincided with the white and pink shrimp spawning season. To determine the potential impact on shrimp larvae a series of static acute (24–96 h) toxicity studies with water accommodated fractions (WAFs) of Macondo Canyon (MC) 252 crude oil, the Corexit 9500A dispersant, and chemically enhanced WAFS (CEWAFs) were conducted with nauplii, zoea, mysid, and postlarval Farfantepenaeus duorarum. Median lethal concentrations (LC50) were calculated and behavior responses (swimming, molting, light sensitivity) evaluated. Impacts were life stage dependent with zoea being the most sensitive. Behavioral responses for all stages, except postlarvae, occurred at below LC50 values. Dispersants had the greatest negative impact while WAFs had the least. No short-term effects (survival, growth) were noted for nauplii exposed to sub-lethal CEWAFs 39 days post-exposure. This study points to the importance of evaluating multiple life stages to assess population effects following contaminant exposure and further, that the use of dispersants as a method of oil removal increases oil toxicity.

Keywords: Farfantepenaeus duorarum; shrimp; DWH; MC252 crude oil; Corexit 9500A dispersant

1. Introduction The Gulf of Mexico (GOM) has some of the most productive coastal bodies of water in the world, making it a major source for the U.S. seafood industry and the most economically important of all domestic commercial seafood harvesting sectors [1]. One of the most important GOM fisheries is the shrimp industry, extending from Brownsville, Texas to Key West, Florida. In 2010 the GOM provided 68% of U.S.-harvested shrimp with a total dockside value of $281 million [2]. The fishery consists of three major species: brown shrimp (Farfantepenaeus aztecus), pink shrimp (Farfantepenaeus duorarum), and white shrimp (Litopenaeus setiferus) [3]. Both the pink and white shrimp began migrating offshore to spawn in the spring, with continued spawning migration throughout the summer (pink shrimp) and fall (white shrimp), while the spawning season for brown shrimp is less defined in terms of season [4–6]. Fertilized eggs pass through nauplii, zoea, and mysis stages in offshore waters before migrating back to coastal estuaries as postlarvae within three to four weeks, throughout the spring and summer, dependent on species. On April 20, 2010 the Deepwater Horizon (DWH) oil platform exploded resulting in 200 million gallons of oil being released into the GOM until the well was capped on July 15 [7,8]. It is estimated that 100,000 km2 of the GOM was affected by the spill, which coincided

J. Mar. Sci. Eng. 2016, 4, 24; doi:10.3390/jmse4010024 www.mdpi.com/journal/jmse J. Mar. Sci. Eng. 2016, 4, 24 2 of 18 with the spring spawning season of a number of key GOM species, including shrimp [3–6,8]. In an effort to contain the spill and prevent the oil from reaching the shoreline, booms, skimmers, burning, direct recovery, and dispersants were used [7]. It has been calculated that 1.9 million gallons of dispersant (Corexit 9527 and Corexit 9500A) were used [9]. Dispersants do not remove oil from water but act to break the oil into smaller droplets that are more readily dispersed into the water column [10]. While dispersant use decreases the amount of surface oil lessening the amount of oil that reaches shorelines, the small dispersed droplets that remain in the water column are now made available to pelagic organisms that inhabit the water column [7]. Several studies have shown negative impacts of oil or dispersed oil exposure on various invertebrates, including mollusks [11–15], echinoderms [16–18], and crustaceans [11,13,19–22]. Other studies have focused on determining the effect of dispersants on marine organisms [10,23–26]. Most studies concentrate on one stage of development. Early life stages are typically more sensitive to pollutants than juveniles or adults and may be impacted at concentrations that, at least on the surface, do not cause acute mortality in juveniles or adults. Yet, in the long term survival may be impacted by behavioral modiﬁcations such as reduced activity that may affect predator avoidance and food intake [27–31]. The aim of our study was to determine what concentration of MC252 oil, Corexit dispersant and chemically dispersed MC252 would adversely affect the survival, development, and behavioral responses of the four major larval stages of shrimp (nauplii, zoea, mysis, and postlarvae). Behavioral responses included swimming activity, light response, feeding, and molting.

2. Experimental Section

2.1. Animals

Various life stages: nauplii (stage N1,N5), proto-zoea (stage Z1,Z3), mysis (stage M1,M2), and six-day-old post-larvae (Pl6) of shrimp (Farfantepenaeus duorarum) were obtained from two commercial shrimp facilities in Florida (Scientiﬁc Associates, Indiantown and Pine Island Aquafarms, St. James City, FL, USA).

2.2. Solution Preparation Oil and dispersant solutions for all experiments were prepared with MC252 oil (British Petroleum Company, BP PLC, London, UK) or Corexit 9500A dispersant (Nalco/Exxon Energy Chemicals, Sugarland, TX, USA). Solutions were prepared following CROSERF procedures [32,33]. Prior to solution preparation, crude oil was physically weathered in the lab for 24 h by placement of oil in a beaker on a stir plate and mixing with a magnetic stir bar in the dark in a chemical fume hood. Stock solutions of water accommodated fractions (WAFs) of crude oil (2 g L´1), dispersant (2 g L´1) and chemically enhanced WAFs (CEWAFs) (1:10 ratio) were prepared in 2 L ﬂasks of ﬁltered, UV treated seawater (28 ppt), covered, and mixed at moderate intensity (25% vortex) for 24 h. Stock solutions were allowed to settle for 3 h prior to preparation of working solutions.

2.3. PAH Analysis Samples of oil and dispersed oil stocks (2 g L´1) used in the acute toxicity experiments were preserved in glass jars with dichloromethane (1:10 v/v) and extracted using modified EPA method 3510C (Mote Marine Laboratory, Sarasota, FL). Polycyclic aromatic hydrocarbons (PAHs, parent compounds, and homologues) were analyzed using GC/MS (Agilent 7890A/5975C), modified EPA method 8260. Total petroleum hydrocarbons (TPH) n-C9 to n-C42 were analyzed using a GC with a flame ionization detector (FID, Agilent 7890A, Agilent Technologies Inc., Santa Clara, CA, USA). J. Mar. Sci. Eng. 2016, 4, 24 3 of 18

2.4. Acute Toxicity Bioassays

2.4.1. Survival (Determination of LC50 Values)

Acute static toxicity tests (May 2011) were conducted with N2,Z1, and M1 using nominal WAF ´1 concentrations of 0, 6.25, 12.5, 25, 50, and 100 mg L , and for Pl6 shrimp using nominal WAF concentrations of 0, 50, 100, 200, and 400 mg L´1 . CEWAF concentrations of 0, 6.25, 12.5, 25, 50, and 100 mg L´1, and dispersant concentrations of 0, 1.25, 2.5, 5, 10, and 50 mg L´1 were used for all four life stages, with ﬁve replicates per treatment for each of the three solutions. Nauplii (N = 15) were placed in ﬁnger bowls containing 50 mL of the appropriate solution. All other life stages (N = 15 Z1; N = 12 M1, Pl6) were placed in 1000 mL beakers containing 600 mL of the appropriate solution. All containers were placed in incubators (28 ˝C, 12:12 h light:dark cycle). Shrimp were fed once per day. Nauplii and zoea were fed a mixture of Chaetocerous gracilis and Isochrysis galbana, mysis were fed rotifers, and postlarvae were fed a pelleted diet (Shrimp PL 40-9, Zeigler Bro. Inc., Gardners, PA, USA). Survival was assessed at 24, 48, 72, and 96 h. Lethal concentrations (LC50) were determined using the trimmed Spearman-Karber method (ToxCalcv5.0).

2.4.2. Behavioral Responses Several experiments were conducted to evaluate behavioral responses. Activity level (swimming behavior) and molting frequency were evaluated for M1 and Pl6 stages for both WAF (0, 100, 200, 400, 800, and 1200 mg L´1) and CEWAF (0, 6.25, 12.5, 25, 50, and 100 mg L´1) exposures, with five replicates per treatment group, and 12 shrimp per replicate. Activity level was scored on a scale of 1–4: 1 = actively moving, 2 = moderately active, 3 = lethargic/moving appendages only, 4 = dead. Molting frequency was calculated as the percent of shrimp that molted compared to the total number of shrimp. Subsequent behavioral response experiments were conducted for CEWAF exposures only for N5, Z1,Z3, and M2 stages. Concentrations used varied based on life stage evaluated, with five replicates per treatment group, 15 shrimp per replicate. CEWAF concentrations of 0, 6.25, 12.5, 25, 50, and 100 mg L´1 were used for nauplii and mysis stages, while concentrations were adjusted to 0, 3.125, 6.25, 12.5, and ´1 25 mg L for the more sensitive proto-zoeal stages as determined by the LC50 experiments. Behavioral parameters assessed included activity and molting as defined above, and feeding and photo-taxic response. Feeding was scored on a 1–4 scale: 1 = actively feeding (food in gut, fecal strands), 2 = 50% or less feeding, 3 = 25% or less feeding, 4 = 0% feeding. Photo-taxic response was evaluated by placing a light source to one side of the container and noting the proportion of shrimp that were attracted to the light (N5,Z1) or avoided the light (Z3,M2). Photo-taxic response was scored on a 1–3 scale: for N5, Z1—1 = actively moving towards light, 2 = sluggish response, 3 = no response; for Z3, M2—1 = actively moving away from light, 2 = slow avoidance response, 3 = no response. The proportion of shrimp that underwent metamorphosis to the next stage was also noted in these experiments.

2.5. Sub-Lethal Toxicity Bioassays Approximately 10,000 L. duorarum nauplii were evenly divided between one of six 13 L buckets containing either filtered, UV treated HBOI salt well water (N = 3) or 23 mg L´1 CEWAF (N = 3). During the 24 h exposure, shrimp were fed Isochrysis galbana during experimental exposure at a rate of 15,000 (or ˆ103) cells/mL. Surviving shrimp from both control buckets and treatment buckets were sieved, combined and then redistributed into one of four 400-L larval rearing tanks (two control, two treatment) containing filtered, UV treated HBOI salt well water. On day 1 (24 h exposure) and on alternate days, seven to eight shrimp were randomly removed from each of the four tanks (15 control, 15 exposed) for 39 days, collected and placed in vials containing 10% NBT formalin. After a 24 h fixation period, shrimp were placed in 70% ethanol, examined microscopically, and photographed (Infinity 2 digital camera, Luminera Co., Sachse, TX, USA). Developmental stage was recorded and length measurements averaged for each data point using Infinity Analyze (Luminera Co.). J. Mar. Sci. Eng. 2016, 4, 24 4 of 18

3. Results

3.1. PAH Analysis The total PAH level in the CEWAF stock solution (1429 µg L´1) was three times greater than that of the WAF stock solution (452 µg L´1) while the TPH level (62,613 µg L´1) in the CEWAF solution was 25 times greater than that of the WAF stock solution (2467 µg L´1) (Table 1). The predominant compound was napthalene, which made up 83.5% of the compounds in the WAF and 65% of the compounds in the CEWAF stock solution. Compounds containing three and four carbon rings (e.g., anthracene, ﬂuorene, pyrene, chrysene, and phenanthrene) were approximately two times greater in the CEWAF compared to the WAF solution.

Table 1. Individual PAH and total TPH and PAH concentrations (µg L´1) of 2 g L´1 stock solutions of water accommodated fractions (WAF) and chemically enhanced water accommodated fractions (CEWAF) used to prepare working solutions used in the acute toxicity experiments.

2 ppt CEWAF 2ppt WAF Target Compounds C rings µg L´1 µg L´1 Napthalene (C0-C4) 2 925.96 377.66 Acenaphthylene 2 6.40 0.05 Acenaphthene 2 0.61 0.67 Fluorene (C0-C4) 3 102.92 14.65 Anthracene (C0-C4) 3 235.38 30.14 Phenanthrene 3 34.79 8.36 Fluoranthene 3 1.53 0.18 Chrysene (C0-C4) 4 32.6 4.02 Pyrene (C0-C4) 4 61.5 6.71 Benzo[A]anthracence 4 0.21 0.14 Napthobenzothiophene (C0-C4) 4 1.32 0.16 Dibenzothiophene (C0-C4) 5 5.47 5.6 Benzo[B]fluorene 5 0.72 0.09 Benzo[B]fluoranthene 5 0.00 0.07 Benzo[K]fluoranthene 5 0.43 0.00 Benzo[E]pyrene 5 0.61 0.11 Benzo[A]pyrene 5 0.00 0.02 Perylene 5 0.77 0.13 Dibenzo[A,H]anthracene 5 0.00 0.01 Indeno[1,2,3-Cd]pyrene 6 0.01 0.00 Benzo[G,H,I]perylene 6 0.00 0.02 Total PAH in µg L´1 1428.64 451.92 Total Petroleum Hydrocarbon 62,613.50 2466.57 C9-C42 in µg L´1

3.2. Acute Toxicity Bioassays

3.2.1. Survival (Nominal LC50 Values) Dispersants had the greatest impact on survival of all larval stages while WAFs had the least, with the proto-zoeal (Z1) stage exhibiting the greatest sensitivity and the postlarval (Pl6) stage the least sensitivity to all three contaminants (Table 2). The dispersant had the greatest impact on Z1 shrimp ´1 ´1 (3.1 mg L , LC50, 24 h; 2.5 mg L LC50, 48 h; 100% mortality, 72 h), with all other stages having ´1 similar LC50 values at 24 h (21–33 mg L ) (Table 2). LC50 values continued to decrease for nauplii (N2) ´1 and mysis (M1) over time, but not for Pl6 (22–28 mg L ). CEWAFs, likewise, had the greatest impact ´1 on Z1 shrimp (15.4 mg L , LC50, 24 h; 100% mortality, 48 h), with all other stages having similar ´1 LC50 values at 24 h (81.5–100 mg L ) (Table 2). LC50 values continued to decrease for all stages over ´1 ´1 time, but less for Pl6 (44 mg L , LC50, 96 h) than for M1 (8.5 mg L , LC50, 96 h). WAFs had the least J. Mar. Sci. Eng. 2016, 4, 24 5 of 18

´1 ´1 impact on all life stages, with Z1 being the most sensitive (67.4 mg L , LC50, 24 h; 25.5 mg L LC50, 48 h; 100% mortality, 72 h) and Pl6 the least, with no LC50 value determined for concentrations tested (>400 mg L´1, 96 h) (Table2).

Table 2. Lethal concentration (LC50) values for shrimp exposed to oil (WAF), dispersant (Corexit 9500A) and oil/dispersant mixture (CEWAF) as determined by a trimmed Spearman-Karber method (ToxCalc). ´1 Reported values include nominal LC50 (95% CL) (mg L ) and corresponding PAH and TPH levels (µg L´1). Non-determined values are indicated by ND; Non-calculated values are indicated by NC.

Time WAF LC50 PAH TPH CEWAF LC50 PAH TPH Corexit LC50 24 h Nauplii >100 NC NC 81.5 (75.3, 88.1) 58 2,551 33.3 (31.8, 34.9) Zoea 1 67.4 (39, 100) 15 83 15.4 (11.6, 20.4) 11 470 3.1 (0.7, 13.7) Mysis 1 >100 NC NC 84.6 (74.9, 95.7) 60 2,649 20.9 (19.2, 22.7) PL 6 >400 NC NC 99.7 (77.3, 100) 71 3,121 28.4 (24.2, 33.3) 48 h Nauplii >100 NC NC 41.5 (36.3, 47.5) 30 1,299 18.6 (16.8, 20.5) Zoea 1 25.5 (22.5, 28.9) 6 31 ND NC NC <2.5 Mysis 1 >100 NC NC 47.4 (41.3, 54.3) 34 1,484 18.3 (16.1, 20.8) PL 6 >400 NC NC 70.4 (56.4, 87.9) 50 2,204 26.5 (22.4, 31.3) 72 h Nauplii ND NC NC ND NC NC ND Zoea 1 21.2 (17.7, 25.5) 5 26 ND NC NC ND Mysis 1 >100 NC NC 31.9 (28.6, 35.7) 23 999 8.3 (6.8, 10.1) PL 6 >400 NC NC 49.2 (40, 60.5) 35 1,002 22.4 (20.8, 23.9) 96 h Nauplii ND NC NC ND NC NC ND Zoea 1 23.3 (20.9, 26) 5 29 ND NC NC ND Mysis 1 29.7 7 37 8.5 (7.1, 10.1) 6 266 2.6 (2.2, 3.0) PL 6 >400 NC NC 44 (36.5, 53.2) 31 1,377 22.5 (21.4, 23.8)

3.2.2. Survival (Determined PAH and TPH LC50 Values) PAH and TPH values for nominal CEWAFs and WAFs could only be compared at 24 h for Z1 and 96 h for M1 (Table 2). Toxicity of CEWAFs and WAFs were similarly toxic when PAH concentrations were compared, however WAFs were more toxic than CEWAFs when TPH concentrations were compared.

3.2.3. Behavioral Response—WAF & CEWAF (M1,Pl6)

Activity (swimming ability) was signiﬁcantly decreased for M1 exposed to CEWAF and WAF ´1 ´1 (Figures 1 and 2). CEWAFs decreased M1 activity at 50 mg L (36 µg L PAH) at 24 h (F5,24 = 103.6, ´1 ´1 p < 0.0001), 12.5 mg L (18 µg L PAH) at 48 and 72 h (F5,24 = 25.93, p < 0.0001; F5,24 = 26.53, ´1 ´1 p < 0.0001) and 6.25 mg L (4 µg L PAH) at 96 h ((F5,24 = 119.73, p < 0.0001 (Figure 1). WAFs ´1 ´1 ´1 decreased M1 activity at 800 mg L (181 µg L PAH) at 24 h (F5,24 = 28.11, p < 0.0001), 400 mg L ´1 ´1 ´1 (90 µg L PAH) at 48 h (F5,24 = 9.62, p < 0.0001) and 100 mg L (23 µg L PAH) at 72 and 96 h (F5,24 = 12.38, p < 0.0001; F5,24 = 25.05, p < 0.0001 ) (Figure2). J.J. Mar. Mar. Sci. Sci. Eng. Eng. 2016 2016, ,4 4, ,24 24 66 of of 19 19

−1 −1 == 9.62,9.62, pp << 0.0001)0.0001) andand 100100 mgmg LL−1 (23(23 gg LL−1PAH) PAH) atat 7272 andand 9696 hh ((FF5,245,24 == 12.38,12.38, pp << 0.0001;0.0001; FF5,245,24 == 25.05,25.05, pp

55 2424 h h 48 h Z 48 h zz ZZ zzzzcc Z 44 7272 h h YZYZ yy CC 9696 h h YY bcbc

bcbc 3 a b 3 XX a XX xx b Activity Activity 1 1 B M 2 w B M 2 w ABAB A A A aa AA A 11

00 00 6.25 6.25 12.5 12.5 25 25 50 50 100 100 CEWAFCEWAF (mg (mg L L−−11))

FigureFigure 1.1.1. Average AverageAverage activity activityactivity level levellevel (˘ S.D.) (±S.D.)(±S.D.) of F. ofof duorarum F.F. duorarumduorarummysis 1mysismysis (M1) shrimp11 (M(M11)) larvae shrimpshrimp exposed larvaelarvae to exposedexposed chemically toto chemicallyenhancedchemically water enhanced enhanced accommodated water water accommodated accommodated fractions of fractions fractions MC252 of crude of MC252 MC252 oil (CEWAF). crude crude oil oil (CEWAF). (CEWAF). Treatment Treatment Treatment groups consisted groups groups consistedofconsisted ﬁve replicates ofof fivefive replicates withreplicates 12 shrimp withwith 1212 each: shrimp shrimp 1 = each: active,each: 11 = 2= active, =active, moderately 22 == moderatelymoderately active, 3 active, active, = lethargic, 33 == lethargic, lethargic, and 4 = and dead.and 44 =Numerical= dead.dead. NumericalNumerical representations representationsrepresentations indicate statistical indicateindicate statistical comparisonsstatistical comparisonscomparisons of exposure ofof periods. exposureexposure Statistical periods.periods. differences StatisticalStatistical differencesweredifferences seen at were were all exposure seen seenl atl at al al times exposure exposure (p < 0.0001). times times ( (pp < < 0.0001). 0.0001).

55 2424 h h yy y YYYyyybc bc 4848 h h YY y bcbc YYYyyybc BB bc 44 YY 7272 h h 9696 h h BB abab 3 3 XX ab Activity ab Activity 1 1 M M 22 xx A AA AA A AA aa 11

00 00 100 100 200 200 400 400 800 800 1200 1200 WAFWAF (mg (mg L L−−11))

Figure 2. Average activity level (˘S.D.) of F. duorarum mysis 1 (M1) shrimp larvae exposed to water FigureFigure 2.2. AverageAverage activityactivity levellevel (±S.D.)(±S.D.) ofof F.F. duorarumduorarum mysismysis 11 (M(M11)) shrimpshrimp larvaelarvae exposedexposed toto waterwater accommodated fractions of MC252 crude oil (WAF). Treatment groups consisted of ﬁve replicates with accommodatedaccommodated fractionsfractions ofof MC252MC252 crudecrude oiloil (WAF).(WAF). TreatmentTreatment groupsgroups consistedconsisted ofof fivefive replicatesreplicates 12 shrimp each: 1 = active, 2 = moderately active, 3 = lethargic, and 4 = dead. Numerical representations withwith 1212 shrimpshrimp each:each: 11 == active,active, 22 == moderatelymoderately active,active, 33 == lethargic,lethargic, andand 44 == dead.dead. NumericalNumerical indicate statistical comparisons of exposure periods. Statistical differences were seen at all exposure representationsrepresentations indicate indicate statistical statistical comparisons comparisons of of exposure exposure periods. periods. Statistical Statistical differences differences were were seen seen times (p < 0.0001). atat all all exposure exposure times times ( (pp < < 0 0.0001)..0001).

Activity of Pl6 was not affected by exposure to CEWAFs or WAFs at concentrations tested ActivityActivity ofof PlPl66 waswas notnot affectedaffected byby exposureexposure toto CEWAFsCEWAFs oror WAFsWAFs atat concentrationsconcentrations testedtested (Figures 3 and 4). There were signiﬁcant differences in activity in Pl6 exposed to CEWAFs at 48 h (Figures(Figures 3 3 and and 4). 4). There There were were significant significant differences differences in in activity activity in in Pl Pl66 exposed exposed to to CEWAFs CEWAFs at at 48 48 h h ( (FF5,245,24 (F5,24 = 4.56, p = 0.0046) and 96 h (F5,24 = 15.73, p = 0.0001), however this was not dose dependent == 4.56,4.56, pp == 0.0046)0.0046) andand 9696 hh ((FF5,245,24 == 15.73,15.73, pp == 0.0001),0.0001), howeverhowever thisthis waswas notnot dosedose dependentdependent (Figure(Figure 3).3). (Figure 3). There were signiﬁcant differences in activity in Pl6 exposed to WAFs at 48 h (F5,24 = 3.81, ThereThere werewere significantsignificant differencesdifferences inin activityactivity inin PlPl66 exposedexposed toto WAFsWAFs atat 4848 hh ((FF5,245,24 == 3.81,3.81, pp == 0.0111),0.0111), p = 0.0111), however activity was not dose dependent (Figure 4). howeverhowever activityactivity waswas notnot dosedose dependentdependent (Figure(Figure 4).4). J. Mar. Sci. Eng. 2016, 4, 24 7 of 19

J.J. Mar. Mar. Sci. Sci. Eng. Eng.2016 2016,,4 4,, 24 24 77 of of 18 19 5 48 h

5 4872 h h Y Y 96 h 4 72 h Y A XZ Y 4 96 h XZ X A XZ 3 AB Z XZ AB X Activity B 6 3 AB

Pl AB B Z B Activity 2 B 6 Pl 2 B B 1

1 0 0 6.25 12.5 25 50 100 0 CEWAF (mg L−1) 0 6.25 12.5 25 50 100 CEWAF (mg L−1) Figure 3. Average activity level (±S.D.) of F. duorarum postlarval (Pl6) shrimp exposed to chemically

Figureenhanced 3. Average water accommodated activity level (˘ fractionsS.D.) of F. of duorarum MC252 crudepostlarval oil (CEWAF). (Pl6) shrimp Treatment exposed groups to chemically consisted Figure 3. Average activity level (±S.D.) of F. duorarum postlarval (Pl6) shrimp exposed to chemically enhancedof five replicates water accommodated with 12 shrimp fractions each: 1 of = MC252active, 2 crude = moderately oil (CEWAF). active, Treatment 3 = lethargic, groups and consisted 4 = dead. enhanced water accommodated fractions of MC252 crude oil (CEWAF). Treatment groups consisted ofNumerical ﬁve replicates representations with 12 shrimp indicate each: statistical 1 = active, comparisons 2 = moderately of exposure active, 3periods = lethargic, (p ≤ 0.0046, and 4 =48 dead. h; p = of five replicates with 12 shrimp each: 1 = active, 2 = moderately active, 3 = lethargic, and 4 = dead. Numerical0.3642, 72 representationsh; p ≤ 0001, 96 h). indicate statistical comparisons of exposure periods (p ď 0.0046, 48 h; Numerical representations indicate statistical comparisons of exposure periods (p ≤ 0.0046, 48 h; p = p = 0.3642, 72 h; p ď 0001, 96 h). 0.3642, 72 h; p ≤ 0001, 96 h). 48 h 5 72 h 48 h 5 96 h 4 72 h B AB96 h 4 B AB 3 AB A AB A AB A Activity 3 A 6 AB 2 Pl Activity 6 2 Pl 1

1 0 0 100 200 400 800 1200 0 −1 0 100 200WAF (mg L 400) 800 1200

WAF (mg L−1) Figure 4. Average activity level (˘S.D.) of F. duorarum postlarval (Pl6) shrimp exposed to MC252 water accommodatedFigure 4. Average fractions activity of level crude (±S.D.) oil (WAF). of F. duorarum Treatment postlarval groups consisted(Pl6) shrimp of exposed five replicates to MC252 with water 12 shrimpaccommodated each: 1 = active, fractions 2 = of moderately crude oil active,(WAF). 3 =Treatment lethargic, groups and 4 = consisted dead. Significant of five replicates differences with were 12 Figure 4. Average activity level (±S.D.) of F. duorarum postlarval (Pl6) shrimp exposed to MC252 water seenshrimp at 48 each: h (p = 1 0.0111).= active, 2 = moderately active, 3 = lethargic, and 4 = dead. Significant differences were accommodated fractions of crude oil (WAF). Treatment groups consisted of five replicates with 12 seen at 48 h (p = 0.0111). shrimp each: 1 = active, 2´ =1 moderately active, 3 = lethargic, and 4 = dead. Significant differences were CEWAFs ě 12.5 mg L caused a significant increase in molting of M1 (Figure 5). Significant seen at 48 h (p = 0.0111). −1 differencesCEWAFs were ≥ seen12.5 at mg both L 24 caused (F5,17 = a 7.71, significantp = 0.0006) increase and 48 in h molting (F5,17 = 63.79,of M1 p(Figure< 0.0001). 5). Significant differences were seen at both 24 (F5,17 = 7.71, p = 0.0006) and 48 h (F5,17 = 63.79, p < 0.0001). CEWAFs ≥ 12.5 mg L−1 caused a significant increase in molting of M1 (Figure 5). Significant differences were seen at both 24 (F5,17 = 7.71, p = 0.0006) and 48 h (F5,17 = 63.79, p < 0.0001). J. Mar.J. Sci. Mar. Eng. Sci.2016 Eng., 20164, 24, 4, 24 8 of 198 of 18

2.00 24 h 1.75 48 h 72 h AB 1.50 96 h 1.25 Molts b b

of b 1.00 B b B 0.75 AB

Proportion 0.50 A A a 0.25 a 0.00 0 6.25 12.5 25 50 100 CEWAF Crude Oil (mg L−1)

Figure 5. Proportion of F. duorarum M1 shrimp larvae (˘S.D.) that molted following exposure to Figure 5. Proportion of F. duorarum M1 shrimp larvae (±S.D.) that molted following exposure to CEWAFs.CEWAFs. Treatment Treatment groups groups consisted consisted of ﬁve replicatesof five replicates with 12 shrimp with 12 each. shrimp Numerical each. representationsNumerical indicaterepresentations statistical comparisons indicate statistical of exposure comparisons periods. of exposure Signiﬁcant periods. differences Significant were differences seen at were 24 andseen 48 h (p ď 0.0006).at 24 and 48 h (p ≤ 0.0006).

3.2.5. Behavioral Response—CEWAF (N5, Z1, Z3, M2) 3.2.4. Behavioral Response—CEWAF (N5,Z1,Z3,M2)

Exposure of nauplii (N5) to CEWAFs impacted swimming ability, feeding activity and phototaxic Exposure of nauplii (N5) to CEWAFs impacted swimming ability, feeding activity and phototaxic response (Table 3). Activity (swimming ability) of N5 shrimp significantly decreased (F4,25 = 98.8, p < response (Table 3). Activity (swimming ability) of N shrimp significantly decreased (F = 98.8, 0.0001) after 24 h exposure to all concentrations of 5CEWAF tested (12.5–100 mg L−1) in a4,25 dose p < 0.0001) after 24 h exposure to all concentrations of CEWAF tested (12.5–100 mg L´1) in a dose dependent matter. At 48 h, activity of N5 shrimp was significantly different (F4,25 = 98.8, p=.0262) only F p dependentat 100 matter.mg L−1. Feeding At 48 h, activity activity was of N likewise5 shrimp reduced was significantly (F4,10 = 35.75, differentp < 0.0001) ( at4,25 24= h 98.8, at all CEWAF=.0262) only ´1 at 100concentrations. mg L . Feeding Phototaxtic activity response was likewise was also reduced reduced (F at4,10 both= 35.75, 24 (F4,25p <= 81.45, 0.0001) p < at 0.0001) 24 h at and all 48 CEWAF h concentrations.(F4,15 = 7.67, p Phototaxtic = 0014), in a response dose dependent was also manner, reduced with atexposed both 24 shrimp, (F4,25 being= 81.45, slowerp < to 0.0001) respond and at 48 h (F4,15 both= 7.67, 24 pand = 0014), 48 h. There in a dose was no dependent difference manner, in metamorphosis with exposed from shrimp,nauplii to being zoea slowerstages by to 48 respond h, at bothexcept 24 and for at 48 100 h. Theremg L−1. was no difference in metamorphosis from nauplii to zoea stages by 48 h, except for at 100 mg L´1. Table 3. Behavioral response of various larval stages of F. duorarum exposed to nominal CEWAF concentrations. Activity level is ranked on a 1–4 scale (1 = active, 4 = no response); feeding is ranked Table 3. Behavioral response of various larval stages of F. duorarum exposed to nominal CEWAF on a 1–4 scale (1 = 100% feeding, 4 = 0% feeding); phototaxic response is ranked on a 1–3 scale, which concentrations. Activity level is ranked on a 1–4 scale (1 = active, 4 = no response); feeding is ranked indicates attraction to light (1 = 100% attracted, 3 = 0% response) for nauplii (N5) and zoea (Z1), but on a 1–4 scale (1 = 100% feeding, 4 = 0% feeding); phototaxic response is ranked on a 1–3 scale, which light avoidance (1 = 100% avoidance, 3 = 0% response) for zoea (Z3) and mysis (M2). For each indicatesconcentration attraction a total to light of five (1 replicates = 100% attracted, consisting 3of =15 0% shrimp response) each were for averaged. nauplii (N Letters5) and indicate zoea (Z1), but lightsignificant avoidance differences (1 = 100% between avoidance, behavioral 3 responses = 0% response) for concentrations for zoea (Z at 3each) and time mysis point. (M 2). For each concentration a total of five replicates consisting of 15 shrimp each were averaged. Letters indicate Activity Phototaxic significantLarval differencesTime Concentration between behavioral responsesFeeding for concentrations % at each time point. Level (1–3) ± Metamorphosis Stage (h) (mg L−1) (1–4) ± S.D. molts Larval Concentration Activity(1–4) Level ± S.D. Feeding PhototaxicS.D. Time (h) % molts Metamorphosis Stage (mg L´1) (1–4) ˘ S.D. (1–4) ˘ S.D. (1–3) ˘ S.D. N5 24 0 1.03±0.05a 1.0 ± 0.0a ‐ 1.02 ± 0.05a N5‐Z1 N5 24 0 1.03˘0.05 a 1.0 ˘ 0.0 a - 1.02 ˘ 0.05 a N5-Z1 12.5 2.13 ± 0.05b 3.33 ± 0.5b ‐ 2.13 ± 0.05b N5‐Z1 12.5 2.13 ˘ 0.05 b 3.33 ˘ 0.5 b - 2.13 ˘ 0.05 b N5-Z1 25 25 2.352.35˘ 0.35 ± 0.35bc bc 3.673.67˘ 0.5± 0.5b b ‐ - 2.272.27 ± 0.41˘ 0.41b b N5‐N5-Z1Z1 50 2.28 ˘ 0.14 bc 4.0 ˘ 0.0 b - 2.28 ˘ 0.14 b N5-Z1 50 2.28 ± 0.14bc 4.0 ± 0.0b ‐ 2.28 ± 0.14b N5‐Z1 100 2.98 ˘ 0.04 c 4.0 ˘ 0.0 b - 3.0 ˘ 0.0 c N5-Z1 48 0100 1.332.98˘ 0.52 ± 0.04a c 4.0- ± 0.0b ‐ -3.0 1.5 ± ˘0.00.58c a N5‐Z1Z1-Z2 12.5 2.17 ˘ 0.98 a - - 2.25 ˘ 0.29 b Z1-Z2 48 0 1.33 ± 0.52a ‐ ‐ 1.5 ± 0.58a Z1‐Z2 25 2.23 ˘ 0.74 a - - 2.34 ˘ 0.45 bc Z1-Z2 50 3.19 ˘ 0.95 ab - - 2.53 ˘ 0.67 bc Z1-Z2 100 3.83 ˘ 0.41 b - - 3.0 ˘ 0.0 c Z1 J. Mar. Sci. Eng. 2016, 4, 24 9 of 18

Table 3. Cont.

Larval Concentration Activity Level Feeding Phototaxic Time (h) % molts Metamorphosis Stage (mg L´1) (1–4) ˘ S.D. (1–4) ˘ S.D. (1–3) ˘ S.D. Z1 24 0 1.67 ˘ 0.82 a - 4% 1.67 ˘ 0.82 a Z1-Z2 (4:1) 3.125 2.29 ˘ 0.56 ab - 12% 2.33 ˘ 0.58 ab Z1 6.25 2.54 ˘ 0.56 b - 21% 2.63 ˘ 0.38 b Z1 12.5 2.92 ˘ 0.13 bc - 21% 2.92 ˘ 0.13 bc Z1 25 3.0 ˘ 0.0 c - 21% 3.0 ˘ 0.0 c Z1 48 0 2.0 ˘ 1.55 - - 1.67 ˘ 1.03 a Z1-Z2 3.125 2.17 ˘ 1.47 - - 1.83 ˘ 0.98 a Z1-Z2 6.25 3.0 ˘ 1.55 - - 2.5 ˘ 0.77 ab Z1-Z2 12.5 3.17 ˘ 1.33 - - 2.5 ˘ 0.84 ab Z1-Z2 25 4.0 ˘ 0.0 - - 3.0 ˘ 0.0 b - 72 0 2.5 ˘ 1.64 a - - 2.0 ˘ 1.1 Z2 3.125 2.75 ˘ 1.47 a - - 2.33 ˘ 1.03 Z1-Z2 6.25 3.75 ˘ 0.61 ab - - 2.75 ˘ 0.61 Z1-Z2 12.5 4.0 ˘ 0.0 b - - 3.0 ˘ 0.0 - 25 4.0 ˘ 0.0 b - - 3.0 ˘ 0.0 - Z3 24 0 1.0 ˘ 0.0 a 1.0 ˘ 0.0 0.80% 1.0 ˘ 0.0 a - 3.125 1.0 ˘ 0.0 a 1.0 ˘ 0.0 1.60% 1.0 ˘ 0.0 a - 6.25 1.0 ˘ 0.0 a 1.0 ˘ 0.0 0% 1.0 ˘ 0.0 a - 12.5 2.0 ˘ 0.0 b 1.0 ˘ 0.0 2.80% 3.0 ˘ 0.0 b - 25 2.0 ˘ 0.0 b 1.0 ˘ 0.0 1.60% 3.0 ˘ 0.0 b - 48 0 1.02 ˘ 0.04 1.0 ˘ 0.0 9.1% - Z3-M1 (3:2) 3.125 1.0 ˘ 0.0 1.0 ˘ 0.0 15% - Z3-M1 (1:4) 6.25 1.0 ˘ 0.0 1.0 ˘ 0.0 19.0% - Z3-M1 (1:4) 12.5 1.0 ˘ 0.0 1.0 ˘ 0.0 13.3% - M1 25 1.02 ˘ 0.04 1.0 ˘ 0.0 14.1% - Z3-M1 (2:3) 72 0 1.13 ˘ 0.31 - 0% 1.29 ˘ 0.71 ab - 3.125 1.7 ˘ 1.1 - 0% 1.7 ˘ 1.1 ab - 6.25 1.03 ˘ 0.05 - 0% 1.03 ˘ 0.05 a - 12.5 1.12 ˘ 0.12 - 0% 2.39 ˘ 0.47 b - 25 1.07 ˘ 0.05 - 1.60% 2.03 ˘ 0.03 b - M2 24 0 1.0 ˘ 0.0 a 1.0 ˘ 0.0 a 4% 1.0 ˘ 0.0 a M2 12.5 1.2 ˘ 0.45 a 1.0 ˘ 0.0 a 1% 1.0 ˘ 0.0 a M2 25 1.84 ˘ 0.19 b 2.2 ˘ 0.45 b 28% 3.0 ˘ 0.0 b M2 50 1.9 ˘ 0.12 b 3.6 ˘ 0.55 c 37% 3.0 ˘ 0.0 b M2 100 2.0 ˘ 0.0 b 4.0 ˘ 0.0 c 51% 3.0 ˘ 0.0 b M2 48 0 1.0 ˘ 0.0 a 1.0 ˘ 0.0 a 2% 1.0 ˘ 0.0 a - 12.5 1.2 ˘ 0.45 a 1.0 ˘ 0.0 a 4% 2.0 ˘ 0.0 b - 25 1.35 ˘ 0.41 a 2.0 ˘ 0.0 b 28% 2.0 ˘ 0.0 b - 50 1.98 ˘ 0.08 b 2.6 ˘ 0.55 bc 2% 3.0 ˘ 0.0 c - 100 2.28 ˘ 0.04 b 3.4 ˘ 0.55 c 0% 3.0 ˘ 0.0 c - 72 0 1.8 ˘ 1.1 a 1.0 ˘ 0.0 a 0% 1.0 ˘ 0.0 a M3 12.5 2.29 ˘ 0.40 a 1.0 ˘ 0.0 a 0% 2.2 ˘ 0.45 b - 25 2.6 ˘ 0.55 ab 3.2 ˘ 1.1 b 3% 2.8 ˘ 0.45 bc - 50 3.0 ˘ 0.0 b 4.0 ˘ 0.0 b 0% 3.0 ˘ 0.0 c - 100 3.0 ˘ 0.0 b 4.0 ˘ 0.0 b 0% 3.0 ˘ 0.0 c -

The behavior responses of proto-zoeal larvae (Z1,Z3) were impacted at lower concentrations of CEWAFs than were nauplii and mysis larvae (see below), and Z1 larvae tended to exhibit these responses at levels lower than did Z3 larvae (Table 3). Activity of Z1 (F4,25 = 6.68, p = 0.0008) and Z3 ´1 (F4,25 = 707, p < 0.0001) shrimp was signiﬁcantly decreased at 6.125 and 12.5 mg L , respectively, at 24 h. No signiﬁcant differences were seen for either zoeal stage at 48 h, or for Z3 shrimp at 72 h, ´1 however, Z1 shrimp exposed to 12.5 and 25 mg L for 72 h (F4,25 = 2.96, p = 0.039) were dead or moribund. Feeding activity was not monitored for Z1 shrimp, however no difference was seen with ´1 Z3 shrimp (Table 3). Molting frequency increased following exposure to 3.125 mg L CEWAFs at 24 h for Z1 and 48 h for Z3 shrimp (Table 3). Phototaxic response was reduced for Z1 shrimp at both ´1 24 (F4,25 = 7.78, p = 0.003) and 48 h (F4,25 = 2.77, p = 0.05) at 6.125 and 12.5 mg L respectively, and ´1 for Z3 shrimp at 24 (F4,25 = 3751, p < 0.0001 ) and 72 h (F4,25 = 5.12, p = 0.0036) at 12.5 mg L . All Z1 control shrimp developed to Z2 stage by 72 h, while some of the shrimp in all exposed groups were still in stage Z1. An interesting pattern was seen in Z3 exposed shrimp. A greater percentage of shrimp J. Mar. Sci. Eng. 2016, 4, 24 10 of 18

´1 exposed to the highest CEWAF concentrations (12.5 and 25 mg L ) developed to M1 stage than did control Z3 shrimp, or shrimp exposed to lower CEWAF concentrations (Table3). Exposure of M2 larvae to CEWAFs affected swimming ability, feeding response, phototaxic ´1 response and molting (Table 3). Swimming ability was affected at 25 mg L at 24 h (F4,20 = 21, ´1 p = 0.0007) and at 50 mg L at 48 (F4,20 = 13.8, p < 0.0001) and 72 h (F4,20 = 3.0, p = 0.043). Feeding ´1 behavior was affected at 25 mg L at all exposure times (F4,20 = 99, p < 0.0001, 24 h; F4,20 = 45, p < 0.0001, 48 h; F4,20 = 48.9, p < 0.0001, 72 h) and shrimp exposed to higher concentrations had ceased feeding at 72 h. Molting frequency increased following exposure at 24 h for concentrations ´1 ´1 25–100 mg L (Table 3). Light avoidance was affected at 25 mg L at 24 h (F4,20 = 956, p < 0.0001) ´1 and at 12.5 mg L at 48 (F4,20 = 1483, p < 0.0001) and 72 h (F4,20 = 44.2, p < 0.0001). No apparent lag in development to M3 was noted between control and exposed groups.

3.3. Long Term Sublethal Effects

3.3.1. Survival No difference in survival was seen between control and treatment groups. Survival was approximately 25% for both groups at day 39 post exposure.

3.3.2. Growth No significant difference was seen in growth. Growth, as defined by total body length, was not significantly different between control and exposed groups from N5 to Pl28 (F1,12 = 0.42, p = 0.5302) (FigureJ. Mar. Sci.6). Eng. 2016, 4, 24 11 of 19

FigureFigure 6.6. Average growth growth (±S.D.) (˘S.D.) of of F.F. duorarum duorarum shrimpshrimp nauplii nauplii (N (=N 15)= 15)exposed exposed to sub to‐ sub-lethallethal −´1 1 concentrationsconcentrations (23 mg L ) of) of CEWAFs CEWAFs for for 24 24 h ( hF1,12 (F 1,12= 0.42,= 0.42, p = 0.5302).p = 0.5302).

3.3.3.3.2.3. Developmental Developmental Stages AA slight slight developmentaldevelopmental delay delay was was seen seen between between the the control control and and exposed exposed treatments treatments on day on dayfive. ﬁve. Development from Z3 to M1 proceeded at a slower pace in the exposed groups resulting in a delayed Development from Z3 to M1 proceeded at a slower pace in the exposed groups resulting in a delayed development from M2 to M3 from day seven to nine. By day 11, development was similar and both groups reached Pl1 (Figure 7).

Figure 7. Development of F. duorarum nauplii (N = 15) exposed to sublethal concentrations of dispersed oil (23 mg L−1) for 24 h. Life stage, (y axis), were assigned a numerical function for graphical J. Mar. Sci. Eng. 2016, 4, 24 11 of 19

Figure 6. Average growth (±S.D.) of F. duorarum shrimp nauplii (N = 15) exposed to sub‐lethal

concentrations (23 mg L−1) of CEWAFs for 24 h (F1,12 = 0.42, p = 0.5302).

3.2.3. Developmental Stages J. Mar. Sci. Eng. 2016, 4, 24 11 of 18 A slight developmental delay was seen between the control and exposed treatments on day five. Development from Z3 to M1 proceeded at a slower pace in the exposed groups resulting in a delayed development from M2 to M3 fromfrom day seven to nine. By day 11, development was similar and both groups reached PlPl11 (Figure7 7).).

Figure 7.7.Development Development of ofF. duorarumF. duorarumnauplii nauplii (N = 15)(N exposed= 15) exposed to sublethal to sublethal concentrations concentrations of dispersed of ´1 oildispersed (23 mg oil L (23) mg for L 24−1) h.for Life24 h. stage,Life stage, (y axis), (y axis), were were assigned assigned a numericala numerical function function for for graphicalgraphical representation. 1 = nauplii V, 2 = zoea 1, 3 = zoea 2, 4 = zoea 3, 5 = mysis 1, 6 = mysis 2, 7 = mysis 3, 8 = postlarvae.

4. Discussion Exposure of various stages (nauplii, zoea, mysis, and postlarvae) of F. duorarum shrimp larvae to MC252 surrogate oil from the Deepwater Horizon well, and the primary dispersant, Corexit 9500A, used during the spill, adversely affected survival and behavior. Zoea (Z1) were more sensitive to contaminant effects than other life stages. Dispersant exposure had a more pronounced affect, than did water accommodated fractions of crude oil (WAFs) or chemically enhanced WAFs (CEWAFs), and affected all larval stages equally and negatively. CEWAFs generally had a more negative impact than WAFs. Effects were dose and exposure dependent, with short-term sublethal effects resulting in slight developmental delays, with no longer term consequences to growth or survival seen in the laboratory. The water column is the most likely route of contaminant uptake following an oil spill and therefore the toxicity of oil within the water column are commonly measured by analyzing the amount of oil contaminants within the WAFs and CEWAFs. Oil used was artiﬁcially weathered, to decrease the amount of volatile organic compounds (benzene, toluene, ethylbenzene, xylene) that tend to evaporate readily, in order to more closely mimic the type of oil that most organisms in the water column would likely encounter. Concentrations of WAFs and CEWAFs used in this study represent moderate to maximum environmentally relevant levels based on reported literature. Oil levels ranging from 20 to 600 mg L´1 and dispersed oil levels ranging from 25 to 75 mg L´1 have been reported in the water column 24 h after a spill and may be a magnitude greater immediately following a spill [34]. Dispersants are used in relatively few spill events due in part to unfavorable conditions that are necessary for dispersants to work effectively [35]. Due to the magnitude of the Deepwater Horizon event large amounts of dispersant were used. Although the majority of water column organisms were likely exposed to either oil or dispersant and oil mixtures during the event it is possible that some organisms were J. Mar. Sci. Eng. 2016, 4, 24 12 of 18 inadvertently exposed to dispersants alone. Levels of dispersant used in these experiments were considered relevant based on reported literature. Dispersant concentrations of 0.1–15 mg L´1 have been reported in the ﬁeld [10,31,36], although initial dispersant concentrations are generally below 10 mg L´1 (maximum range 5–15 mg L´1), dropping to less than 1 mg L´1 in a few hours [24,36].

4.1. Acute Toxicity Effects

4.1.1. Survival Survival is often used as the endpoint to determine oil toxicity effects. Previous studies have reported lethal concentration (LC50) values for crustaceans, including shrimp, following acute exposure to oil, however, researchers typically focus on only one life stage and one contaminant (oil, dispersed oil, or dispersant). Early life stages tend to be more susceptible to toxic compounds than adults, which may negatively affect populations in affected areas, especially in the short-term. This study is unique in studying the effects of various shrimp larval stages simultaneously, to oil, dispersed oil, and a dispersant, and reporting behavioral responses along with LC50 values. We found that larval shrimp mortality varied dependent on developmental stage, and was not age dependent as zoea were more sensitive than nauplii. This is likely the result of differing feeding modalities at these two larval stages. Nauplii have undeveloped mouth parts and rely on their yolk sac for nutrition, while zoea are indiscriminate feeders and consume anything large enough to enter their mouth, and mysis seek out and capture their food [37]. Feed sources provided during this study varied: nauplii and zoea were provided with algae, mysis with rotifers, and postlarvae with commercial pellets. Exposure of WAFs or CEWAFs through either the water column or exposed feed resulted in similar alterations of metabolic enzymes in ﬁsh [38]. The addition of small quantities of feed may have resulted in larvae being exposed to oil contaminants through both the water column via the gills and the digestive tract via ingestion. We believed that administration of some feed was necessary to eliminate the likelihood of starvation as the cause of death as larvae, unlike postlarvae and juvenile shrimp, need to eat continuously. Preliminary experiments conducted with untreated and unfed larvae resulted in notable lethargy at 24 h and 50%–90% mortality of nauplii and zoea, respectively, at 48 h. We noted that dispersant exposure negatively impacted all four larval stages at similar concentrations, although zoea were the most adversely affected, with all Z1 shrimp dead by 48 h. Our reported values for 96 h exposures for F. duorarum M1 and Pl6 for Corexit 9500A, were similar to ´1 those previously reported for Corexit 9527 for L. setiferus postlarvae (96 h LC50, 12–31 mg L )[11,13]. ´1 Similar LC50 values (3.5–83 mg L ) have been reported following 48 to 96 h of exposure of other postlaval and juvenile crustaceans to Corexit 9500 [10,26,32,39,40]. Exposure to nominal concentrations of CEWAFs resulted in increased mortality compared to WAFs in our study. The majority of researchers have concluded that chemically dispersed oil is more toxic than physically dispersed oil [35]. However, reporting methods (nominal, PAH, TPH), may impact researchers conclusions [41]. In our study, when PAH, rather than nominal values were compared, the toxicity of WAFs and CEWAFs were equivalent, although WAFs appeared to be more toxic when TPHs were compared, similar to that seen for L. setiferus juveniles [13]. The increase in toxicity of CEWAFs is attributed to increased availability of PAHs in the water column through the creation of a large number of small oil droplets [35]. The PAH levels in the prepared CEWAF stocks in our study were three times greater and the TPH levels 25 times greater than in the WAF stocks. Oil droplets were observed in fecal strands and the digestive tract and fecal strands of some zoea and mysis larvae exposed to CEWAF concentrations ě 25 mg L´1 indicating ingestion of oiled particulates in larvae that were still feeding. At 50–100 mg L´1 oil was noted on appendages and molts of some shrimp larvae, implicating narcosis (PAH toxicity) and perhaps restricted motility as likely causes of mortality. Although previous researchers have compared survival of crustaceans exposed to WAFs and CEWAFs by reporting LC50 values, there is little consistency in the reporting method (nominal, TPH, J. Mar. Sci. Eng. 2016, 4, 24 13 of 18

PAH) which makes comparison difficult. In this study, we attempted to make cross-comparison easier by listing LC50 values for all three parameters. The majority of early crustacean studies were conducted with larval mysid shrimp, Americanus (Mysiopsis) bahia, where TPH values were reported, ´1 ´1 and 96 h LC50 values ranged from 0.15–83 mg L WAFs and 0.5–120 mg L CEWAFs [26,42,43]. That these results differ somewhat may be explained by variation in exposure methods used (constant, spiked, static renewal), however, each study reported a similar toxicity for WAFs and CEWAFs based on comparison of TPHs. Similar results have been reported for Americamysis (Holmesimysis) costata (1–35 mg L´1 WAF, 8–33 mg L´1 CEWAF, 96 h) and L. setiferus juveniles (6.5 mg L´1 WAFs, 5–7.5 mg L´1 CEWAFs, 96 h) [13,25,44]. In contrast, we report a 96 h TPH toxicity with larval F. duorarum (0.029–0.037 mg L´1 WAFs, 0.27–1.38 mg L´1 CEWAFs), indicating that WAFs were more toxic. This is in contrast to 96 h LC50 PAH values, in which little difference in toxicity was seen for WAFs and CEWAFs. In the wake of the Deepwater Horizon event, reported concentrations of TPAHs in May 2010 varied greatly dependent on site and sampling depth and ranged from 0 to 146 mg L´1 at the wellhead, 0 to 0.9 mg L´1 in field collected WAFs, 0 to 18 mg L´1 in field collected CEWAFs, and 0 to 0.17 µg L´1 in shoreline samples [45–47]. Following capping of the well in July concentrations in collected sample were significantly lower at all sites and depths.

4.1.2. Behavior Factors in addition to mortality need to be considered when assessing contaminant effects, as behavioral responses, such as swimming ability, and response to stimuli affect the ability to locate prey or escape predation. Some researchers have reported behavioral inhibitory (IC50) effects following exposure to lower levels of oil contaminants than at which mortality (LC50) occurs. Changes in behavior due to sublethal exposures are considered to be the most sensitive indicators of environmental disturbance, and yet are among the least studied effects with regards to toxicity [48]. A variety of behavioral responses of marine organisms to pollutants, including oil, such as motivation (e.g., feeding response), sensory responses (e.g., phototaxis), and motor activity (e.g., swimming performance) are given in a summary of early work [49]. Examples of depressed feeding responses associated with PAHs have been shown in a variety of invertebrates including rotifers, crabs, and shrimp [50–52]. Examples of differential phototaxic responses associated with invertebrates have been reported with crabs and barnacles [27,53]. Invertebrates, such as crabs, shrimp, and barnacles, have also been shown to exhibit erratic swimming behavior in response to oil contaminants [27,53,54] and it has been postulated that differential sensory and motor responses that resulted in differential depth distribution might affect larval distribution and recruitment via directional current activity [53]. Some researchers have reported behavioral effects, such as reduction of swimming ability, at lower than LC50 concentrations [15,28]. Others have reported similar LC50 and EC50 values following exposure to oil contaminants, including swimming ability, settlement behavior and burying behavior [27,31,39]. Decreased swimming behavior is likely a result of narcosis typically seen in acute toxicity of high short-term exposures to naphthalene [55]. Narcotic chemicals affect the lipid bilayer in membranes reducing activity and the ability to react to stimuli, which may ultimately lead to mortality [56]. However narcosis does not account for other reported toxic effects such as deformities, edema, and cardiovascular effects [57] Regardless of whether this is the result of narcosis, or some other phenomenon the end result is that decreased swimming ability results in decreased ability to find food or escape predation, either of which will likely reduce survival. In this study, swimming ability was significantly decreased for M1 larvae exposed to both CEWAF and WAF at all exposure times. Exposure to CEWAFs caused an initial decrease in activity at concentrations that were two times less than LC50 values at 24 h and three to four less than LC50 values at 48 and 72 h. Similar results were seen with N5,Z1,Z3, and M2 larvae exposed to CEWAFs. However, results were stage dependent, in that Z1 larvae were the most sensitive, whereas, older shrimp (Pl6) did not exhibit reduced swimming ability at sublethal concentrations of either CEWAFs or WAFs. Although significant differences in activity in Pl6 shrimp exposed to CEWAFs at 48 and J. Mar. Sci. Eng. 2016, 4, 24 14 of 18

96 h and to WAFs at 48 h occurred, they were not dose dependent and thought to be due to water quality issues. Phototaxic response, whether attraction to light (N5,Z1) or avoidance (Z3,M2) followed a similar pattern to that seen with swimming ability, and this manifested itself in reduced feeding response for N5,Z1 stages.

4.1.3. Molting Molting frequency increased in response to CEWAFs at 24 or 48 h post-exposure. Response was stage dependent with Z1 and Z3 larvae responding at lower concentrations than M1 or M2 larvae. It is postulated that this behavior is a stress response to compounds present in CEWAFs or an attempt by the shrimp to rid itself of oil adhering to the carapace or appendanges. Molting was not associated with metamorphosis to the next stage, as, except for the Z3 stage, metamorphosis occurred at the same rate as the controls or was somewhat delayed, and control shrimp molted less frequently. An unwanted side effect of this response might be increased susceptibility to oil contaminants. Crustaceans are more susceptible to environmental stressors, including oil pollutants during molting, and may experience increased mortality [58,59]. In the present study, molting frequency was not followed after 72–96 h post-exposure. PAHs has been shown to increase the length of the intermolt period, resulting in decreased molting in a variety of invertebrates [58,60,61]. The use of additional measurements, such as biochemical endpoints, provide researchers with another set of tools for evaluating oil toxicity, allowing additional means of assessing the potential consequences of oil exposure for marine organisms, such as shrimp.

4.2. Sublethal Effects Survival following sublethal exposures of invertebrates varies based on species and life stage [21,54,62,63]. In the present study, F. duorarum nauplii exposed to sub-lethal amounts (23 mg L´1, LC10) of CEWAFs for 24 h showed no difference in survival compared to controls over 39 days and was approximately 25% for both groups. Shrimp were cultured in larval tanks specific to the penaeid shrimp industry and industry operational procedures followed. Due to the sensitivity of handling the zoeal stage, tanks are not drained, water is instead added to partially filled tanks to alleviate water quality issues. Unfortunately, Artemia proliferated in the tanks, competing with shrimp for resources, and causing low survival in tanks regardless of exposure. Delayed morality has been reported by other researchers in shrimp and crab larvae and embryos exposed to low levels of WSF for short periods with zoea being more sensitive than later developmental stages [21,54,62]. Other research has shown no survival effects in crab zoea following either short term exposure or continuous exposure to low concentrations of oil [63]. Developmental delays have been reported for invertebrate larvae exposed to PAHs [21,63–65]. Developmental delays in invertebrates are typically accompanied by an increased period of intermolt following exposure to oil [58,64]. We saw a slight developmental lag in the exposed group between Z3 and M1 resulting in delayed development to subsequent mysis stages, but both groups developed in postlarvae at roughly the same time. Other researchers have reported similar findings. Zoeal stages of the mud crab Rhithropanopeus harrisii increased following prolonged exposure to chronic, low levels of WSF, however short term exposure had no impact [63,64]. Minor differences (one day lags) in development times were also reported in Pandalus borealis larvae in early stages of development but effects decreased at later stages [21]. The more time spent in pelagic larval stages, as would occur as a result of delayed development, may result in increased likelihood of predation, impact dispersion, increase time to maturity, and therefore negatively affect population growth rate [66]. In our study, despite slight developmental lags, no lasting growth effects were seen in the exposed group at day 39 (Pl28). This is consistent with results reported in the literature for Cancer irroratus larvae exposed to WAFs, R. harrisii larvae exposed to low concentrations of naphthalene, and grass shrimp Palaemonetes pugio exposed to sublethal WAFs [59,67,68]. Other research has indicated decreased growth following exposure to oil. Decreased growth was reported in DWH exposed juvenile brown J. Mar. Sci. Eng. 2016, 4, 24 15 of 18 shrimp Farfantepenaeus aztecus but not in juvenile white shrimp L. setiferus exposed to the same waters [22].

5. Conclusions This study shows that the concentrations of oil released and dispersant used during the DWH event could have negatively affected penaeid shrimp in the GOM, whether through altered behavioral responses, delayed development, or mortality. Even though the spill occurred during the spring spawning season and likely affected shrimp larvae at select locations, GOM shrimp populations as a whole do not appear to have been affected long term, perhaps in part due to ﬁshery closures that were put in place following the spill [69].

Acknowledgments: This research was made possible by a grant from BP/The Gulf of Mexico Research Initiative through the Florida Institute of Oceanography #4710-1101-00B. We thank Sloane Wendell, BP GCRO Reference Material Account Manager for providing us with surrogate MC252 oil #SO-20110212-HMPA9-008; David Giddings, Strategic Technology Manager of Nalco Co., Sugarland Texas, for providing us the Corexit 9500A dispersant; and to FAU student Kaitlin Gallagher (currently UConn). This is HBOI-FAU publication # 1989. Author Contributions: Susan Laramore and Amber Garr conceived and designed the experiments; Susan Laramore and William Krebs performed the experiments; Amber Garr analyzed the data; Susan Laramore and William Krebs wrote the paper; Amber Garr edited the paper. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. National Marine Fisheries Service (NMFS). Fisheries of the United States 2003. In National Marine Fisheries Service, Office of Science and Technology; NMFS: Silver Spring, MD, USA, 2004. 2. National Marine Fisheries Service (NMFS). Fisheries of the United States 2010. In National Marine Fisheries Service, Office of Science and Technology; NMFS: Silver Spring, MD, USA, 2011. 3. Zimmerman, R.J.; Minello, T.J.; Rozas, L.P. Saltmarsh linkages to productivity of penaeid shrimps and blue crabs in the Northern Gulf of Mexico. In Concepts and Controversies in Tidal Marsh Ecology; Weinstein, M.P., Kreeger, D.A., Eds.; Kluwer Academic Publishers: Dordrecht, The Netherlands, 2000; pp. 293–314. 4. Linder, M.J.; Anderson, W.W. Growth, Migration, Spawning and Size Distribution of Shrimp: Penaeus setiferus; US Government Printing Office: Washington, DC, USA, 1956; Volume 56, pp. 555–645. 5. Renfro, W.C. Life History Stages of Gulf of Mexico Brown Shrimp; United States Bureau of Commercial Fisheries: Galveston, TX, USA, 1964; pp. 94–98. 6. Cummings, W.C. Maturation and spawning of the pink shrimp, Penaeus duorarum Burkenroad. Trans. Am. Fish. Soc. 1961, 90, 462–468. [CrossRef] 7. National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling. Deep Water: The Gulf oil disaster and the future of offshore drilling. 2011; Report to the President. Available online: http://docs.lib.noaa.gov/noaa_documents/NOAA_related_docs/oil_spills/DWH_report-to-president.pdf (accessed on 16 November 2011). 8. Kelly, T.R. Environmental health insights into the 2010 Deepwater Horizon (BP) oil blowout. Environ. Health Insights 2010, 4, 61–63. [CrossRef] 9. Federal Interagency Solutions Group. Oil Budget Calculator: Deepwater Horizon, November 2010. Available online: http://www.restorethegulf.gov/sites/default/files/documents/pdf/OilBudgetCalc_Full_HQ- Print_111110.pdf (accessed on 16 November 2011). 10. George Aryes, A.; Clark, J.R. Aquatic toxicity of two Corexit dispersants. Chemosphere 2000, 40, 897–906. [CrossRef] 11. Fucik, K.W.; Carr, K.A.; Balcom, B.J. Toxicity of oil and dispersed oil to the eggs and larvae of seven marine fish and invertebrates from the Gulf of Mexico. In The Use of Chemicals in Oil Spill Response; ASTM STP 1252; Lane, P., Ed.; American Society for Testing and Materials: Philadelphia, PA, USA, 1995; pp. 135–171. 12. Le Hir, M.; Hily, C. First observations in a high rocky-shore community after the Erika oil spill (December 1999, Brittany, France). Mar. Pollut. Bull. 2002, 44, 1243–1252. [CrossRef] J. Mar. Sci. Eng. 2016, 4, 24 16 of 18

13. Liu, B.; Romaire, R.P.; Delaune, R.D.; Lindau, C.W. Field investigation on the toxicity of Alaska North Slope crude oil (ANSC) and dispersed ANSC crude to Gulf killifish, Eastern oyster and white shrimp. Chemosphere 2006, 62, 520–526. [CrossRef][PubMed] 14. Saco-Alvarez, L.; Bellas, J.; Nieto, O.; Bayona, J.M.; Albaiges, J.; Beiras, R. Toxicity and phototoxicity of water-accommodated fraction obtained from Prestige fuel oil and Marine fuel oil evaluated by marine bioassays. Sci. Total Environ. 2008, 394, 275–282. [CrossRef][PubMed] 15. Laramore, S.; Krebs, W.; Garr, A. Effects of Macondo Canyon 252 oil (naturally and chemically dispersed) on larval Crassostrea virginica (Gmelin, 1791). J. Shell. Res. 2014, 33, 709–718. [CrossRef] 16. Temara, A.; Gulec, I.; Holdway, D.A. Oil induced disruption of foraging behavior of the asteroid keystone predator, Coscinasterias muricata (Echinodermata). Mar. Biol. 1999, 133, 501–507. [CrossRef] 17. Fernandez, N.; Cesar, A.; Salamanca, M.J.; DelValls, T.A. Toxicological characterization of the aqueous soluble phase of Prestige fuel oil using the sea urchin embryo bioassay. Ecotoxicology 2006, 15, 593–599. [CrossRef] [PubMed] 18. Beiras, R.; Saco Alvarez, L. Toxicity of seawater and sand affected by the Prestige fuel oil spill using bivalve and sea urchin embryogenesis bioassays. Water Air Soil Poll. 2006, 177, 457–466. [CrossRef] 19. Martínez-Jerónimo, F.; Villaseñor, R.; Ríos, G.; Espinosa-Chavez, F. Toxicity of the crude oil water soluble fraction and kaolin-adsorbed crude oil on Daphnia magna (Crustacea: Anomopoda). Arch. Environ. Contam. Toxicol. 2005, 48, 444–449. [CrossRef][PubMed] 20. Roth, A.M.F.; Baltz, D.M. Short-term effects of an oil spill on marsh edge fishes and decapod crustacean. Estuar. Coasts 2009, 32, 565–572. [CrossRef] 21. Bechmann, R.K.; Larsen, B.K.; Taban, I.C.; Hellgren, L.I.; Moller, P.; Sanni, S. Chronic exposure of adults and embryos of Pandalus borealis to oil causes PAH accumulation, initiation of biomarker responses and an increase in larval mortality. Mar. Poll. Bull. 2010, 60, 2087–2098. [CrossRef][PubMed] 22. Rozas, L.P.; Minello, T.J.; Miles, M.S. Effect of Deepwater Horizon oil on growth rates of juvenile penaeid shrimps. Estuar. Coasts 2014, 37, 1403–1414. [CrossRef] 23. Swedmark, M.; Granmo, A.; Kollberg, S. Effects of oil dispersants and oil emulsions on marine animals. Water Res. 1973, 7, 1649–1692. [CrossRef] 24. Wells, P.G. The toxicity of oil dispersants to marine organisms: A current perspective. In Oil Spill Chemical Dispersants: Research, Experience and Recommendations; ASTM International: Philadelphia, PA, USA, 1984; pp. 177–202. 25. Singer, M.M.; Aurand, D.; Coelho, G.; Bragin, G.E.; Clark, J.R.; Sowby, M.L.; Tjeerdema, R.S. Making, measuring and using water-accommodated fractions of petroleum for toxicity testing. In Proceedings of the 2001 International Oil Spill Conference, American Petroleum Institute, Tampa, FL, USA, 26–29 March 2001; pp. 1269–1274. 26. Fuller, C.; Bonner, J. Comparative toxicity of oil, dispersant and dispersed oil to Texas marine species. In Proceedings of the 2001 International Oil Spill Conference, American Petroleum Institute, Tampa, FL, USA, 26–29 March 2001; pp. 1243–1248. 27. Wu, R.S.S.; Lam, P.K.S.; Zou, B.S. Effects of two oil dispersants on phototaxis and swimming behavior of barnacle larvae. Hydrobiologia 1997, 352, 9–16. [CrossRef] 28. Barron, M.G.; Carl, M.G.; Short, J.W.; Rice, S.D. Photoenhanced toxicity of aqueous phase and chemically dispersed weathered Alaska North Slope crude oil to Pacific herring eggs and larvae. Environ. Toxicol. Chem. 2003, 22, 650–660. [CrossRef][PubMed] 29. Georgiades, E.T.; Holdway, D.E.; Brennan, S.E.; Butty, J.S.; Temera, A. The impact of oil derived products on the behavior and biochemistry of the eleven armed asteroid Coscinasterias muracata, Echinodermata. Mar. Environ. Res. 2003, 55, 257–276. [CrossRef] 30. Lopez, B.A.; Lopez, D.A. Moulting frequency and behavioural responses to salinity and diesel oil in Austromegabalanus psittacus (Molina) (Cirripedia: Balanidae). Mar. Freshw. Behav. Physiol. 2005, 38, 249–258. [CrossRef] 31. Greco, G.; Corra, C.; Garaventa, F.; Chelossi, E.; Faimali, M. Standardization of laboratory bioassays with Balanus amphitrite larvae for preliminary oil dispersants toxicological characterization. Chem. Ecol. 2006, 22, 163–172. [CrossRef] J. Mar. Sci. Eng. 2016, 4, 24 17 of 18

32. Clark, J.R.; Bragin, G.E.; Febbo, R.J.; Letinski, D.J. Toxicity of physically and chemically dispersed oils under continuous and environmentally realistic exposure conditions: Applicability to dispersant use decisions in spill response planning. In Proceedings of the 2001 International Oil Spill Conference, American Petroleum Institute, Tampa, FL, USA, 26–29 March 2001; pp. 1249–1255. 33. Barron, M.G.; Ka'aihue, L. Critical evaluation of CROSERF test methods for oil dispersant toxicity testing under subarctic conditions. Mar. Pollut. Bull. 2003, 46, 1191–1199. [CrossRef] 34. National Research Council (NRC). Committee on understanding oil spill dispersants. In Oil Spill Dispersants Efﬁcacy and Effects; National Academies Press: Washington, DC, USA, 2005; p. 400. 35. Fincas, M. A Review of Literature Related to Oil Spill Dispersants 1997–2008; Prince William Sound Regional Citizens Advisory Council: Anchorage, AK, USA, 2008; p. 155. 36. Trudel, K. Environmental risks and trade-offs in Prince William Sound. In Proceedings of the Dispersant Application in Alaska: A Technical Update, Anchorage, AK, USA, 1998; pp. 159–188. 37. Treece, G.D.; Fox, J.M. Design, Operation and Training Manual for an Intensive Culture Shrimp Hatchery; Publication TAMU-SG-93–505; Texas A&M University, Sea Grant College Program: Galveeston, TX, USA, 1993; p. 187. 38. Cohen, A.M.; Gagnon, M.M.; Nugegoda, D. Alterations of metabolic enzymes in Australian bass, Macquaria novemaculeata, after exposure to petroleum hydrocarbons. Arch. Environ. Contam. Toxicol. 2005, 49, 200–205. [CrossRef][PubMed] 39. Gulec, I.; Leonard, B.; Holdway, D.A. Oil and dispersed oil toxicity to amphipods and snails. Spill Sci. Technol. Bull. 1997, 4, 1–6. [CrossRef] 40. Fuller, C.; Bonner, J.; Page, C.; Ernest, A.; McDonald, T.; McDonald, S. Comparative toxicity of oil, dispersant and oil plus dispersant to several marine species. Environ. Toxicol. Chem. 2004, 23, 2941–2949. [CrossRef] [PubMed] 41. Perkins, R.A.; Rhoton, S.; Behr-Andres, C. Toxicity of dispersed and undispersed, fresh and weathered oil to larvae of a cold water species, Tanner crab (C. bairdi) and standard warm water test species. Cold Reg. Sci. Technol. 2003, 36, 129–140. [CrossRef] 42. Rhoton, S.; Perkins, R.A.; Braddock, J.F.; Behr-Andres, C. A cold weather species response to chemically dispersed fresh and weathered Alaska North Slope crude oil. In Proceedings of the 2001 International Oil Spill Conference, American Petroleum Institute, Tampa, FL, USA, 26–29 March 2001; pp. 1231–1236. 43. Wetzel, D.L.; Van Fleet, E.S. Cooperative studies on the toxicity of dispersants and dispersed oil to marine organisms: A 3 year Florida study. In Proceedings of the 2001 International Oil Spill Conference, American Petroleum Institute, Tampa, FL, USA, 26–29 March 2001; pp. 1237–1241. 44. Singer, M.M.; George, S.; Lee, I.; Jacobson, S.; Weetman, L.L.; Blondina, G.; Tjeerdema, R.S.; Aurand, D.; Sowby, M.L. Effects of dispersant treatment on the acute aquatic toxicity of petroleum hydrocarbons. Environ. Contam. Toxicol. 1998, 34, 177–187. [CrossRef] 45. Boehm, P.D.; Cook, L.L.; Murray, K.J. Aromatic hydrocarbon concentrations in seawater: Deepwater Horizon oil spill. In Proceedings of the 2011 International Oil Spill Conference, American Petroleum Institute, Portland, OR, USA, 23–26 May 2011; Volume 1, Abstract 371. 46. Echols, B.S.; Smith, A.J.; Gardinali, P.R.; Rand, G.M. Acute aquatic toxicity studies of Gulf of Mexico water samples collected following the Deepwater Horizon incident (2 May 2010 to 11 December 2010). Chemosphere 2015, 120, 131–137. [CrossRef][PubMed] 47. Allan, S.E.; Smith, B.W.; Anderson, K.A. Impact of the Deepwater Horizon oil spill on bioavailable polycylic aromatic hydrocarbons in Gulf of Mexico coastal waters. Environ. Sci. Techno. 2012, 46, 2033–2039. [CrossRef] [PubMed] 48. Gerhardt, A. Aquatic behavioral ecotoxicology—Prospects and limitations. Hum. Ecol. Risk Assess. 2007, 13, 481–491. [CrossRef] 49. Eisler, R.; Carney, G.C.; Lockwood, A.P.M.; Perkins, E.J. Behavioural responses of marine poikilotherms to pollutants (and discussion). Philos. Trans. R. Soc. Lond. B Biol. Sci. 1979, 286, 507–521. [CrossRef][PubMed] 50. Takashashi, F.T.; Kittredge, J.S. Sublethal effects of the water soluble component of oil: chemical communication in the marine environment. In The Microbial Degradation of Oil Pollutants; Ahearn, D.G., and Mers, S.P., Eds.; Louisiana State University: Baton Rouge, LA, USA, 1973; pp. 259–264. 51. McCain, B.B.; Malins, D.C. Effects of petroleum hydrocarbons on selected demersal ﬁshes and crustaceans. In Sediment Toxicity Assessment; Burton, G.A., Ed.; Lewis Publishers: Boca Raton, FL, USA, 1992; pp. 315–325. J. Mar. Sci. Eng. 2016, 4, 24 18 of 18

52. Jensen, L.K.; Carroll, J. Experimental studies of reproduction and feeding for two Artic dwelling Calanus species exposed to crude oil. Aquatic. Biol. 2010, 10, 261–271. [CrossRef] 53. Bigford, TE. Effects of oil on behavioral response to light, pressure and gravity in larvae of the rock crab Cancer irroratus. Mar. Biol. 1977, 43, 137–148. [CrossRef] 54. Brodersen, C. Rapid narcosis and delayed mortality in larvae of king crabs and kelp shrimp exposed to the water soluble fraction of crude oil. Mar. Environ. Res. 1987, 22, 233–239. [CrossRef] 55. Rice, S.D.; Short, J.W.; Brodersen, C.C.; Mecklenburg, T.A.; Moles, D.A.; Misch, C.J.; Cheatham, D.L.; Karinen, J.F. Acute toxicity and uptake-depuration studies with Cook Inlet crude oil, Prudhoe Bay crude oil, No. 2 fuel oil, and several subarctic marine organisms. In Northwest Fisheries Center Auke Bay Fisheries Laboratory, Processed Report; National Marine Fisheries Service, NOAA: Juneau, AK, USA, 1976; p. 90. 56. Van Wezel, A.P.; Opperhuizen, A. Narcosis due to environmental pollutants in aquatic organisms—Residue-based toxicity, mechanisms and membrane burdens. Crit. Rev. Toxicol. 1995, 25, 255–279. [CrossRef][PubMed] 57. Barron, M.G.; Carls, M.G.; Heintz, R.; Rice, S.D. Evaluation of ﬁsh early life stage toxicity models of chronic embryonic exposures to complex polycyclic aromatic hydrocarbon mixtures. Toxicol. Sci. 2004, 78, 60–67. [CrossRef][PubMed] 58. Weis, J.S.; Cristini, A.; Rao, K.R. Effects of pollutants on molting and regeneration in Crustacea. Amer. Zool. 1992, 32, 495–500. [CrossRef] 59. Johns, D.M.; Pechenik, J.A. Inﬂuence of water accommodated fraction of No. 2 fuel oil on energetics of Cancer irroratus larvae. Mar. Biol. 1980, 55, 247–254. [CrossRef] 60. Karinen, J.F.; Rice, S.D. Effects of Prudhoe bay crude oil on molting tanner crabs, Chionoecetes bairdi. Mar. Fish. Rev. 1974, 36, 31–37. 61. Cantelmo, A.; Lazell, R.; Mantel, L. The effects of benzene on molting and limb regeneration in juvenile Callinectes sapidus. Mar. Bio. Lett. 1981, 2, 333–343. 62. Cucci, T.L.; Epifanio, C.E. Long term effects of water soluble fractions of Kuwait crude oil on the larval and juvenile development of the mud crab Eurypanopeus depressus. Mar. Biol. 1979, 55, 215–220. [CrossRef] 63. Laughlin, R.B., Jr.; Ng, J.; Guard, H.E. Hormesis: A response to low environmental concentrations of petroleum hydrocarbons. Science 1981, 211, 705–707. [CrossRef][PubMed] 64. Laughlin, R.B. Jr.; Young, L.G.L.; Neff, J.M. A long term study of the effects of water soluble fractions of No. 2 fuel oil on the survival, development rate and growth of the mud crab Rhithropanopeus harrisii. Mar. Biol. 1978, 47, 87–95. [CrossRef] 65. Bang, H.W.; Lee, W.; Kwak, L.S. Detecting points as developmental delay based on the life-history development and urosome deformity of the harpacticoid copepod, Tigriopus japonicus sensu lato, following exposure to benzo(a)pyrene. Chemosphere 2009, 76, 1435–1439. [CrossRef][PubMed] 66. Worboys, M.A.; Leung, K.M.Y.; Grist, E.P.M.; Crane, M. Time should be considered in developmental ecotoxicity test. Mar. Poll. Bull. 2002, 45, 92–99. [CrossRef] 67. Laughlin, R.B., Jr.; Neff, J.M. Ontogeny of respiratory and growth responses of larval mud crabs Rhithropanopeus harrisii exposed to different temperatures, salinities and naphthalene concentrations. Mar. Ecol. Prog. Ser. 1981, 5, 319–332. [CrossRef] 68. Gunderson, D.T.; Kristanto, S.W.; Curtis, L.R.; AL-Yakoob, S.N.; Metwally, M.; Al-Aljimi, D. Subacute toxicity of the water soluble fractions of Kuwait crude oil and partially combusted crude oil on Menidia beryllina and Palaemontes pugio. Arch. Environ. Contam. Toxicol. 1996, 31, 1–8. [CrossRef] 69. Van der Ham, J.L.; de Musert, K. Abundance and size of Gulf shrimp in Louisiana’s coastal areas following the Deepwater Horizon oil spill. PLOS ONE 2014, 9, e108884. [CrossRef]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/). See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/301481290

Chronic Toxicity of Unweathered and Weathered Macondo Oils to Mysid Shrimp (Americamysis bahia) and Inland Silversides (Menidia beryllina)

Article in Archives of Environmental Contamination and Toxicology · April 2016 DOI: 10.1007/s00244-016-0280-x

4 authors, including:

Brandi Echols Abraham J Smith Chevron Florida International University

11 PUBLICATIONS 39 CITATIONS 8 PUBLICATIONS 12 CITATIONS

SEE PROFILE SEE PROFILE

Piero R Gardinali Florida International University

88 PUBLICATIONS 1,313 CITATIONS

SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Abraham J Smith letting you access and read them immediately. Retrieved on: 07 August 2016 Arch Environ Contam Toxicol DOI 10.1007/s00244-016-0280-x

Chronic Toxicity of Unweathered and Weathered Macondo Oils to Mysid Shrimp (Americamysis bahia) and Inland Silversides (Menidia beryllina)

1 1 2 1 B. Echols • A. Smith • P. R. Gardinali • G. M. Rand

Received: 12 August 2015 / Accepted: 4 April 2016 Ó Springer Science+Business Media New York 2016

Abstract Chronic, 21–28-day toxicity tests of Macondo respectively. The effect (LOEC, IC25) and no-effect source (Massachusetts, or MASS) and weathered Slick A exposure concentrations (in TPAHs) from the standardized (CTC) and Slick B (Juniper) oils field collected during the laboratory toxicity studies with un-weathered and weath- 2010 Deepwater Horizon (DWH) Incident in the Gulf of ered oils are discussed relative to the actual exposure Mexico (GOM) were conducted using standardized pro- concentrations in the GOM in 2010. The exposures eval- cedures. Standard species, Americamysis bahia and Meni- uated in the laboratory toxicity tests represent the highest dia beryllina, were evaluated for changes in survival and concentrations of total PAHs that were rarely observed in growth during daily static-renewal tests. Both species water column samples collected in the GOM during the demonstrated an increased sensitivity to low-energy water release and post release periods of the DWH incident. accommodated fractions (WAFs) of un-weathered MASS oil, with growth and survival decreasing as oil loading rate increased from 0.01 to 1.0 g/L. Survival and growth of On April 20, 2010, the blowout on the Deepwater Horizon mysid shrimp exposed to weathered oil (Slick A and Slick (DWH) drilling rig in the Gulf of Mexico (GOM) produced B) did not differ from that of test controls. In contrast, a prolonged (87 day) subsurface release of oil from the survival and growth of inland silversides declined relative well at the Macondo lease block 252 at approximately to that of test controls at loading rates of 1 g/L for both 1500 m depth (Lehr et al. 2010). The oil subsequently rose weathered oils. Based on the concentration of total poly- to the surface, where natural attenuation occurred as a cyclic aromatic hydrocarbons (TPAH42), no observed direct result of volatilization, oxidation, biodegradation, effect concentrations were lower for inland silverside sur- hydrolysis, evaporation, and several other biological and vival (5.00–7.61 lg/L) and growth (\2.02 to \7.61 lg/L) physical mechanisms (Fingas 2011). Additionally, a dis- in chronic exposures to Slick B and Slick A weathered oils persant (COREXIT 9500) was used at the wellhead and compared with mysids (4.75–17.9 lg/L). Average TPAH also was applied aerially and by boat on drifting surface oil concentrations in full strength WAFs followed the weath- to augment natural physical dispersion (NAS 2013). As a ering trend, with 165 ± 17.2, 17.9 ± 0.480, and result, oil was present in the immediate reaches of the 4.75 ± 0.521 lg/L for MASS, Slick A, and Slick B oils, water column and on the sea surface during and after the discharge period (Boehm et al. 2016). The different types and stages of oil (e.g., weathered oil, dispersed weathered & B. Echols [email protected] oil) present in the water column may have posed a hazard to local aquatic organisms. 1 Earth & Environment, Ecotoxicology and Risk Assessment Macondo oils (source and field collected weathered), Laboratory, Southeast Environmental Research Center, along with oil-dispersant mixtures have demonstrated a Florida International University, North Miami, FL 33181, USA range of adverse effects from acute lethality (Almeda et al. 2013; Echols et al. 2016a, b) to sublethal chronic effects 2 Department of Chemistry, EARL, Southeast Environmental Research Center, Florida International University, such as cardiac toxicity (Brette et al. 2014; Incardona et al. North Miami, FL 33181, USA 2014), mutagenicity (Paul et al. 2013), developmental 123 Arch Environ Contam Toxicol deformities (Barron 2012; Incardona et al. 2013; Dubansky collected from the barge on July 26, 2010. Slick A (CTC) et al. 2013), etc. However, sublethal chronic effect studies, weathered oil sample was collected from the CTC barge on in general, have not been based on standardized procedures July 29, 2010. The CTC barge received skimmed oil from and often fail to demonstrate reproducibility. The use of other, unknown, skimming vessels (Schuett 2010; Faksness widely accepted standardized aquatic toxicity test proce- et al. 2015). Slick B (Juniper) weathered oil was collected dures typically used by regulatory authorities for making from the skimming vessel ‘‘Juniper’’ on July 19, 2010. decisions on safety assessment provides greater assurance Based on publicly available data (Gulf Science Data of data quality and the repeatability of results for complex website 2014), the Slick A and Slick B oils show total PAH mixtures, such as oil. At the time this manuscript was (TPAH) depletion of approximately 65 and 83 %, respec- prepared, no information on chronic exposures and toxi- tively, compared with un-weathered source oil. PAH cological effects based on standard aquatic toxicity test depletion was calculated relative to the compound hopane procedures existed in the literature for Macondo oils. In due to the relative recalcitrant nature of this compound general, such chronic toxicity information on crude oils and (Prince et al. 1994). associated polycyclic aromatic hydrocarbons (PAHs) is limited (Lee et al. 2015). Recently, in a review by the Preparation of the Water Accommodated Fractions Royal Society of Canada on the ‘‘Behaviour and Envi- (WAFs) ronmental Impacts of Crude Oil Released into Aqueous Environments’’ (Lee et al. 2015), it is suggested that Synthetic seawater (SSW) was used for cultures, WAF additional research is needed to determine ‘‘critical expo- preparations and dilutions of WAFs (i.e., marine mix plus sure periods for impacts on the reproductive biology of bioelements plus crystal sea marine mix bioassay labora- sexually maturing fish’’ (Lee et al. 2015) and that ‘‘models tory formula sea salts (80:20)) with deionized water, and of chronic toxicity must be developed from results of adjusted to 25 ppt, and aged before use. WAFs prepared for chronic toxicity tests and not from acute toxicity tests via the following studies followed methodology based on application factors’’ (Lee et al. 2015). Therefore, the previous methods described in the literature, e.g., Chemical objective of this study was to assess the chronic toxicity of Response to Oil Spills: Ecological Effects Research Forum source and field-collected Macondo oil(s) using standard- (CROSERF) from 1994 to 2000 (as discussed in Singer ized aquatic toxicity tests with two standard test species et al. 2000, 2001b; Singer and Jacobson 2001a), Neff (mysids and inland silversides). The mysid shrimp and (1999), Rhoton et al. (2001), and the report Critical Eval- inland silverside were chosen, because they are the most uation of CROSERF Test Methods for Oil Dispersant widely used marine standard toxicity test species, and they Toxicity Testing under Subarctic Conditions (Barron and are the recommended test species under USEPA guidance Ka’aihue 2003). for Whole-Effluent Toxicity testing (USEPA 2002). Oil was weighed before and after being dispensed to Additionally, in a recent review of available toxicity the saltwater to estimate an accurate loading rate. This information for the calculation of species sensitivity dis- step is particularly important for the weathered oils, tributions (SSDs) for petroleum products and oil dispersant which were highly viscous and were dispensed into WAF exposures, it was reported that the largest toxicity database using spatulas, instead of gas-tight syringes, which were available for aquatic toxicity of PAHs exists for mysids and used for fresh crude oil. WAFs were prepared in 4-liter inland silversides (Barron et al. 2013). (L) glass aspirator bottles with a bottom dispensing port. Aspirator bottles were filled with 3.8 L SSW to account for a relative headspace-to-water volume ratio of *20 %, Materials and Methods which has been shown to increase reproducibility between WAFs (Singer et al. 2000). Aspirator bottles were sealed Field Collected Source Oils immediately after the addition of oil using DuraSealÒ film (Diversified Biotech). WAFs were mixed on digital stir Water accommodated fractions (WAFs) for toxicity tests plates for 20 h followed by a 4-h settling period. No were prepared using un-weathered Macondo oil (Barge visible vortex-by-depth was observed during mixing. Massachusetts) and two field-collected weathered oil Using this low-energy WAF (LE-WAF) preparation pro- samples, referred to as Slick A (or ‘‘CTC’’) and Slick B (or cedure provides a more stable exposure media, ensures ‘‘Juniper’’). The Barge Massachusetts received un-weath- low potential for creating oil droplets in the water column, ered oil from the Enterprise Producer, which collected and thus minimizes the contribution of any additional crude oil directly from the subsea containment system toxicity from oil droplets. WAFs were prepared under positioned directly over the well. Massachusetts oil, dark conditions in a temperature-controlled incubator at referred to by the reference code ‘‘MASS’’ hereafter, was 25 ± 2 °C. 123 Arch Environ Contam Toxicol

Individual WAFs were prepared using a multiple ratio ratio and a LPC were used for each oil type. On day 0 of method (Singer et al. 2000), which is consistent with each test, 15 \12-h old silversides and 10 juvenile mysids CROSERF in which different oil/water volume ratios or oil were distributed randomly to each replicate test chamber loadings (0.01, 0.1, 1.0 g oil/L) are used for the WAF (1.45-L glass chambers and 260-mL glass jars, respec- exposure series for tests. The oil-to-water ratio 1:10,000 tively) of each oil treatment study and subsequently placed (0.1) is considered environmentally realistic in the upper in an environmental chamber at 25 ± 1 °C with a pho- surface water layer after use with chemical dispersants toperiod of 16 h light: 8 h dark. WAF water renewal (Neff 1999). replacement occurred every 24 h. Every 24 h, the DO, Initial water-quality parameters were measured and temperature, salinity, and pH in each treatment along with recorded [temperature, dissolved oxygen (DO), and salin- survival was recorded. ity]. WAFs were prepared fresh daily before each use, Starting on day 15 for mysid shrimp, the number of decanted, and used immediately. Toxicity studies were reproductively active females was noted, and any young initiated on the same day as the first WAF solutions were produced (F1 generation) were counted and transferred to decanted. A saltwater-only laboratory performance control clean saltwater. Second-generation mysids were held for (LPC) WAF was used to assess test organism health and 96 h and fed. On day 21 (for MASS and Juniper oils) and test acceptability requirements. After measuring water day 28 (for CTC), the final survival data were recorded and quality parameters, water samples were collected for the study was terminated. Surviving mysid were examined chemistry analyses from each loading of 0.01, 0.1, and individually under a dissecting scope and the number of 1.0 g/L WAF and the LPC WAF for incoming water (days males, females with young developing in their brood sac, 0, 7, 14, 21, and 27) and outgoing water (days 1, 8, 15, 21, and number of females with eggs developing in the ovi- and 28) during renewal. ducts were recorded. The number of fecund females per replicate, the mean fecundity per replicate, and the mean Test Organisms fecundity per treatment were calculated. Fecundity of mysids was too variable precluding the possibility of Inland silversides (Menidia beryllina) and mysid shrimp quantifying and analyzing data. (Americamysis bahia) were cultured at Environmental Individual surviving females were rinsed in DI water, Enterprises, USA (Slidell, LA). Both species were main- placed on a tared weighing dish, and dried at 60 ± 4 °C for tained in the laboratory and exposed to the WAF treatments 24 h. Dried mysids were weighed and the individual dry including the untreated LPC in environmental chambers at weight for each mysid, the mean dry weight per replicate, 25 ± 1 °C with a photoperiod of 16 h light: 8 h dark. and the mean dry weight per treatment were calculated. Mysid were \24-h old, and silversides were \12-h old Statistical endpoints reported were survival and growth. post-hatch when toxicity studies were initiated. The wet weight (mg) of each surviving fish was recorded at the end of the test (day 28). The individual wet Chronic Toxicity Study Methods weights were used to calculate the mean wet weight per replicate and the mean wet weight per treatment. Toxicity studies were conducted with the oils in accor- The sensitivity of both mysids and silversides, based on dance with ASTM (2008) for 28-days early life-stage 7-day standard reference toxicant tests with potassium studies with silversides and with USEPA (2002) for 21–28 chloride (Sigma Chemical) indicated that mysids met all days life-cycle studies with mysids. For mysids, a 28-days the quality control parameters based on the no observable study was conducted with Slick A oil; however, because effect concentrations (NOECs), lowest observable effect mysid shrimp propensity to jump increases as they mature, concentrations (LOECs), and IC25 (the concentration tests with MASS and Slick B oils were conducted for expected to result in 25 % reduction in an endpoint, during 21 days. Silverside studies were conducted in covered exposure period) for survival and growth. rectangular PyrexÒ glass chambers (21 9 11 9 7 cm, total Toxicity study statistical endpoints for the chronic volume 1.45 L) with 500 mL of test solution. Mysid studies studies are expressed as oil loading rate (g oil/L). Survival were conducted in glass jars of 260 mL of total volume and growth endpoints collected during these studies were with 220 mL of test solution for days 0–14. On day 15, analyzed using CETIS (Tidepool Scientific, v.1.8.4.30). 355-lm NitexÒ mesh screened caps were added to the tops Treatments with statistically significant survival effects are of the jars (Slick B and MASS tests only) to prevent loss of excluded routinely from sublethal effects (e.g., growth) mysids due to jumping from test chambers. Test jars were statistical analysis to maintain the power of those tests. The then submerged into 600-mL beakers containing 460 mL progression of statistical analysis followed the U.S. EPA of test solution to maintain submersion of the mysids. flowcharts as described in Short-term Methods for Esti- Three replicates of each WAF treatment for each loading mating the Chronic Toxicity of Effluents and Receiving 123 Arch Environ Contam Toxicol

Waters to Marine and Estuarine Organisms, EPA-812-R- duplicates (when available), and standard reference 02-014 (U.S. Environmental Protection Agency 2002). materials. Survival endpoint NOECs and LOECs were calculated using contingency tables and hypothesis testing with Fisher Quality Assurance/Quality Control exact tests using Bonferroni–Holm adjustment. Growth endpoint NOECs and LOECs were calculated using mul- All of the biological and chemistry data presented in this tiple comparison analysis with Bonferroni adjusted t tests paper were subjected to formal data verification and vali- (a = 0.05) or Wilcoxon tests with Bonferroni adjustment. dation prior to use. All laboratory data were reviewed by an Point estimates (ICp) for growth data were first attempted external (independent) auditor to ensure that each study with nonlinear regression analysis, and if the data were was performed in accordance with the protocol and labo- unable to fit an appropriate model, Linear Interpolation was ratory standard operating procedures (SOPs). Sample pro- used. Additional analysis and visual comparisons were cessing activities, analytical procedures, and storage and conducted and displayed in the figures using Minitab 16 holding times for the chemical analyses were consistent (MinitabÒ, 16.2.4) and SigmaPlot 11 (Build: 11.0.0.75). with the Analytical Quality Assurance Plan, Mississippi Canyon 252 (DWH) Natural Resource Damage Assess- Analytical Methods ment, Version 2.2 (NOAA 2010). Batch QA/QC was used to assess method precision and accuracy by analyzing One liter (2 9 500 mL) samples were collected immedi- laboratory blank(s), fortified blank(s), sample duplicates ately following decanting of each WAF into amber glass (when available), and standard reference materials. jars with minimal headspace for parent and alkylated PAHs and saturated hydrocarbons (SHC) analyses. Additional samples (3 9 40 mL) were collected with no headspace for Results and Discussion volatile organic compound (VOC-BTEX) analysis. Samples of each WAF were sent to Battelle (Battelle, Water Quality Analysis Duxbury, MA) for trace volatile and semi-volatile hydrocarbon analyses. All samples were collected in certified, Water quality parameters (i.e., temperature, DO, pH, precleaned glass containers and refrigerated immediately at salinity) were monitored daily for each of the six chronic or below 4 °C during storage and transit to the analytical studies. Test temperatures ranged from 23.9 to 26.4 °C, pH laboratory. Samples were shipped overnight with full ranged from 7.6 to 8.1, DO ranged from 4.5 to 7.7 mg/L, chain-of-custody documentation. and salinity ranged from 24.8 to 26.2 ppt for all tests. All WAF samples were processed by liquid–liquid extraction with methylene chloride and analyzed by Bat- Analytical Chemistry telle for PAHs, SHC, and for diesel range organics by gas chromatography-flame ionization detector (GC-FID; SW- Table 1 shows the summary of TPAH42 (summed as the 846 Method 8015—modified). Parent and alkylated PAHs total of 42 parent and alkylated measured PAHs) mean and hopane were analyzed by gas chromatography/mass analytical results for 100 % WAFs (no vortex) prepared spectrometry in selected ion monitoring (GC/MS-SIM) with artificial saltwater. Preparations containing MASS oil using modifications of SW-846 Method 8270. Due to their were dominated by volatile monoaromatic components strong molecular ion response, GC/MS-SIM yields sensi- (BTEX) and low molecular weight PAHs (2–3 rings) and tive detection of PAHs at low concentrations. Analyte contained small amounts of SHC limited by their solubility concentrations were calculated based on representative (Table 2). The trend in the PAHs was consistent with the deuterated surrogate standards (Naphthalene-d8, Ace- degree of weathering and the water partitioning of the naphthene-d10, Phenanthrene-d10, and benzo(a)pyrene- remaining LMWPAHs (low molecular weight PAHs) in the d12) and surrogate standard recoveries were calculated field-collected slicks. The weathered oil samples contained based on internal standards chrysene-d12 and fluorene-d10. relatively high concentrations of the 2- to 3-ring low- Alkylated homologues were quantitated based on the molecular-weight polycyclic aromatic hydrocarbons response factors of the parent compounds. Samples for (LPAHs) in the WAF fraction relative to the other mea- volatile organic compounds were analyzed by Battelle by sured TPAH constituents (Table 1). The naphthalene’s are purge-and-trap GC/MS (Battelle SOP 5-245, a modifica- the dominant components of the LPAHs in the weathered tion of SW-846 Method 8260). Although individual anal- oil WAFs due to their high water solubility. Average cal- yses may have additional quality assurance requirements, culated TPAH concentrations in the WAFs prepared at test batch quality assurance/quality control included at least the initiation and subsequent renewals ranged from analysis of laboratory blank(s), fortified blank(s), sample 4.75 ± 0.521 lg/L in the 1.0 g/L Slick B WAF to 123 Arch Environ Contam Toxicol

Table 1 Summary of polycyclic aromatic hydrocarbon analysis for 100 % WAF from chronic toxicity studies with mysids (A. bahia) and inland silversides (M. beryllina) conducted using unweathered (MASS) and weathered (Slick A, Slick B) Macondo oils Oil ID-loading rate (g/L) Americamysis bahia Menidia beryllina Mean (lg/L) Mean (lg/L) LPAH HPAH TPAH LPAH HPAH TPAH

MASS-0.01 43.9 ± 4.27 0.221 ± 0.013 44.1 ± 4.26 43.9 ± 4.27 0.221 ± 0.013 44.1 ± 4.26 MASS-0.10 131 ± 20.0 0.333 ± 0.106 132 ± 20.1 131 ± 20.0 0.333 ± 0.106 132 ± 20.1 MASS-1.00 164 ± 17.3 0.707 ± 0.351 165 ± 17.2 164 ± 17.3 0.707 ± 0.351 165 ± 17.2 Slick A-0.01 7.42 ± 0.966 0.195 ± 0.014 7.61 ± 0.974 7.42 ± 0.966 0.195 ± 0.014 7.61 ± 0.974 Slick A-0.10 12.0 ± 0.736 0.320 ± 0.028 12.3 ± 0.746 12.0 ± 0.736 0.320 ± 0.028 12.3 ± 0.746 Slick A-1.00 17.3 ± 0.506 0.586 ± 0.051 17.9 ± 0.480 17.3 ± 0.506 0.586 ± 0.051 17.9 ± 0.480 Slick B-0.01 2.11 ± 0.261 0.150 ± 0.018 2.26 ± 0.268 1.90 ± 0.826 0.117 ± 0.035 2.02 ± 0.859 Slick B-0.10 3.34 ± 0.363 0.224 ± 0.039 3.56 ± 0.398 3.01 ± 0.681 0.193 ± 0.048 3.20 ± 0.720 Slick B-1.00 4.43 ± 0.452 0.326 ± 0.069 4.75 ± 0.521 4.68 ± 1.46 0.320 ± 0.121 5.00 ± 1.57 Calculated total PAH = sum of 42 polycyclic aromatic hydrocarbons Calculated LPAH = sum of low molecular weight (2–3 rings) polycyclic aromatic hydrocarbons Calculated HPAH = sum of high molecular weight (4–6 rings) polycyclic aromatic hydrocarbons

Table 2 Summary of SHC and BTEX analytical results for 100 % WAF from chronic toxicity studies with mysids and inland silversides conducted using unweathered (MASS) and weathered (CTC, Juniper) Macondo oils Oil ID-loading rate Americamysis bahia Menidia beryllina Mean SHC (lg/L) Mean BTEX (lg/L) Mean SHC (lg/L) Mean BTEX (lg/L)

MASS-0.01 4.45 ± 4.06 11.8 ± 7.26 4.45 ± 4.06 11.8 ± 7.26 MASS-0.10 91.0 ± 59.5 338 ± 73.4 91.0 ± 59.5 338 ± 73.4 MASS-1.00 215 ± 244 2812 ± 249 215 ± 244 2812 ± 249 Slick A-0.01 0.733 ± 0.694 0.354 ± 0.611 0.733 ± 0.694 0.354 ± 0.611 Slick A-0.10 0.722 ± 1.06 0.649 ± 1.32 0.722 ± 1.06 0.649 ± 1.32 Slick A-1.00 0.993 ± 1.37 1.01 ± 1.25 0.993 ± 1.37 1.01 ± 1.25 Slick B-0.01 1.21 ± 0.856 0.106 ± 0.212 0.927 ± 1.40 0.028 ± 0.056 Slick B-0.10 3.22 ± 3.11 0.085 ± 0.169 0.757 ± 0.713 ND Slick B-1.00 3.58 ± 1.11 0.041 ± 0.051 1.92 ± 2.49 ND Calculated total SHC = sum of saturated hydrocarbons Calculated BTEX = sum of benzene, toluene, ethyl benzene, and xylenes concentrations Mass and Slick A tests conducted for both species at the same time; Slick B tests conducted at different times for species ND non-detect

165 ± 17.2 lg/L in the 1.0 g/L MASS WAF. The analyt- concentration of TPAH42 (lg/L) increased from 44.1 lg/L ical results for the 100 % WAF for MASS oil loading rate in the 0.01 g/L oil loading concentration to 165 lg/L at 1.0 g/L in the present study is consistent with other TPAH42 with a loading rate of 1.0 g/L of MASS oil analysis for studies of full-strength WAFs at this loading (Figs. 1, 2, 3, 4). Lowest observed effect concentrations rate (Echols et al. 2016a). (LOECs) were calculated based on the TPAH42 (lg/L) at test initiation for changes in survival and growth. MASS oil

Chronic Toxicity Studies exposures resulted in LOECs of 132 lg/L TPAH42, for both endpoints for mysids (Table 3). Mysid mortality was Under chronic exposure test conditions for both species, not signiﬁcant in chronic tests of Slick A (p = 0.6468) and mortality increased and growth decreased as the Slick B (p = 0.9567) oils, with calculated LOECs of[17.9

123 Arch Environ Contam Toxicol

Fig. 1 Concentration responses (% mortality) for Americamysis bahia exposed to water accommodated fractions of Macondo weathered (Slick A ‘‘CTC’’ and Slick B ‘‘Juniper’’) and source (MASS) oils. Mortality results for mysids exposed to MASS oil are Fig. 3 Concentration responses (% mortality) for Menidia beryllina plotted on the secondary x axis. Concentration is expressed as the exposed to water accommodated fractions of Macondo weathered (Slick A ‘‘CTC’’ and Slick B ‘‘Juniper’’) and source (MASS) oils. total polycyclic aromatic hydrocarbons (TPAH42, lg/L) measured at test initiation. Error bars represent the standard error of the mean Mortality results for inland silversides exposed to MASS oil are plotted on the secondary x axis. Concentration is expressed as the total polycyclic aromatic hydrocarbons (TPAH42, lg/L) measured at test initiation. Error bars represent the standard error of the mean

Fig. 2 Concentration responses (growth as mean dry weight (mg)) for Americamysis bahia exposed to water accommodated fractions of Macondo weathered (Slick A ‘‘CTC’’ and Slick B ‘‘Juniper’’) and source (MASS) oils. Growth results for mysids exposed to MASS oil Fig. 4 Concentration responses (growth as mean wet weight (mg)) are plotted on the secondary x axis. Concentration is expressed as the for Menidia beryllina exposed to water accommodated fractions of total polycyclic aromatic hydrocarbons (TPAH42, lg/L) measured at Macondo weathered (Slick A ‘‘CTC’’ and Slick B ‘‘Juniper’’) and test initiation. Error bars represent the standard error of the mean source (MASS) oils. Concentration is expressed as the total polycyclic aromatic hydrocarbons (TPAH42, lg/L) measured at test initiation. Error bars represent the standard error of the mean and [4.75 lg/L TPAH42, respectively for mysids (Table 3). Differences in mysid growth also were not sig- [4.75 lg/L TPAH42, representing the highest concentra- niﬁcantly different between controls and Slick A and Slick tion (1.0 g/L oil loading) for both oils. Average control B WAF exposures, because LOECs also were [17.9 and survival of mysids was within acceptable limits ([80 %), 123 Arch Environ Contam Toxicol

Table 3 Summary of statistical Endpoint Americamysis bahia Menidia beryllina endpoints for chronic studies with mysid shrimp MASS Slick A (CTC) Slick B (Juniper) MASS Slick A (CTC) Slick B (Juniper) (Americamysis bahia) and the inland silverside (Menidia Survival beryllina) NOEC 44.1 17.9 4.75 44.1 7.61 5.00 LOEC 132 [17.9 [4.75 132 12.3 [5.00 IC25 84.3 [17.9 [4.75 61.2 9.81 4.64 Growth NOEC 44.1 17.9 4.75 \44.1 \7.61 \2.02 LOEC 132 [17.9 [4.75 44.1 7.61 2.02 IC25 129 [17.9 [4.75 40.4 7.76 3.85

Endpoints are expressed at TPAH42 (lg/L)

ranging from 83.3 % in the MASS un-weathered oil studies NOECs, LOECs, and IC25 point estimates for these to 90.0 % for the Slick B chronic test. chronic studies were generated because of their use in U.S. Inland silversides showed comparable results to mysids federal regulations and in the calculation of acute-to- for both survival and growth for MASS oil, but this species chronic ratios (USEPA 1991; Rand 1995), although we was more sensitive than mysids to the Slick A and Slick B recognize their limitations (Crane and Newman 2000). For weathered oils, using growth as the endpoint (Figs. 2, 4; survival as the endpoint, the NOECs and LOECs for MASS Table 3). Growth, based on average wet weight (mg) at test and Slick B exposures were similar for both species. termination, was significantly different (p values ranging However, for survival and growth, silverside NOECs and from\0.0001 to 0.0288) than controls for each loading rate LOECs compared to mysids were lower for Slick A than for all three oils evaluated in this study. Average control for Slick B oil exposures. Oil with high fractions rich in survival of inland silversides for the MASS, Slick A, and three to five-ringed PAHs (higher molecular weight com- Slick B tests were 77.8, 93.3, and 71.1 %, whereas survival pounds: see Table 1) including their alkyl homologues are in the highest test treatment of the three oils were 2.2, 0.0, associated with chronic toxicity (Anderson et al. 1974; and 51.1 %, respectively. Neff 1999; Carls et al. 1999; Heintz et al. 1999; Wu et al. The mean wet weight and %CV of surviving silversides 2012). Table 1 shows that the Slick A-weathered oil con- in controls for the MASS oil, Slick A, and Slick B tests tains higher concentrations of HPAH than Slick B-weath- were 19.74 mg (24.6 %), 21.74 mg (25.2 %), and ered oil. Silversides may be more sensitive to these 19.04 mg (18.8 %), respectively. The CVs for wet weight components in chronic exposures than mysids; a result were consistent with acceptable CVs for chronic tests which is partially supported by the lower IC25s for sil- (BSAB 1994; Pastorak et al. 1994). versides than mysids for all oils and both endpoints As noted earlier, the oil-to-water ratio represented by (Table 3). Acute toxicity values from standardized tests the 0.1 g/L loading is considered more environmentally were summarized for mysids and silversides for petroleum realistic in the upper surface water layer after use with products and dispersants to develop SSDs (Barron et al. chemical dispersants (Neff 1999). The 1.0 g/L thus rep- 2013). Silversides were ‘‘1.5- to 3.3-fold less sensitive than resents a worst-case exposure scenario. At the lowest mysids based on ranking within the SSD and 1.6- to 5.8- loading rate (0.01 g/L), mysid survival and growth were fold based on toxicity values.’’ Because there are limited not affected and inland silverside survival was not chronic toxicity standardized tests, we can neither confirm affected, but as illustrated in Fig. 4 growth decreased in the sensitivity of the silversides compared with mysids nor each of the three oil WAF exposures for inland silversides can we develop an SSD based on chronic toxicity values compared to their controls. For growth as the endpoint for petroleum products. with mysids, both the weathered oils were not toxic at the The study design of the laboratory toxicity studies used highest loading rate (1.0 g/L) and they also were less static-renewal exposure systems (i.e., decreasing exposure toxic than the MASS oil. Measured TPAH concentrations concentrations) to approximate field exposure conditions, in WAFs of Slicks A and B producing chronic effects on in which weathering and chemical dispersion reduce growth and survival of fish in this study were within the volatile fractions to detect the extent of toxicity reduction range of 0.3–60 lg/L TPAH reported from chronic produced by decreasing concentrations of monoaromatic exposure studies with fish in the literature (Lee et al. and diaromatic compounds, because continuous exposure 2015). tests may overestimate the potential toxicity that may occur 123 Arch Environ Contam Toxicol in the environment (Clark et al. 2001). To further under- The total PAH concentrations in undiluted 100 % WAFs stand and extrapolate the laboratory exposure-chronic for all laboratory testing exposure media (i.e., up to 1.0 g/L effects results relative to the field exposures, we used the loading rate) represent the highest 2.0 % or less of the TPAH analyses from the GOM water column samples *10,800 total PAH concentrations measured in water (n = 10,828) collected in 2010 from the DWH incident column samples collected in the GOM during the DWH oil (OSAT 2010). Table 4 summarizes the 50–90th percentile spill in 2010. The latter is based on the empirical data TPAH concentrations using all of the 10,828 water samples discussed in the Summary Report for Sub-Sea and Sub- (i.e., nearshore, offshore, deepwater) combined from the Surface Oil and Dispersant Detection: Sampling and GOM in 2010. The estimated 90th centile concentration Monitoring (OSAT 2010) as well as the field data collected was used as an ‘‘exposure descriptor’’ (Solomon et al. as part of NRDA activities. Therefore, the exposures 1996), because it is used in the conduct of aquatic eco- evaluated in the laboratory toxicity tests represent the logical risk assessments. The 90th percentile concentration highest concentrations of total PAHs that were rarely estimate assumes that 90 % of the exposure concentration observed in water column samples collected in the Gulf of samples will be below this descriptor if it comes from an Mexico during the Deepwater Horizon oil spill incident in exposure distribution that is unbiased and that accurately 2010. represents the concentrations found for a location during a given time period (Giddings et al. 2005). Table 1 sum- Acknowledgments This work was supported by BP Exploration & marizes the TPAH concentrations in WAF (100 %) expo- Production Inc. and the BP Gulf Coast Restoration Organization. These studies were conducted at Environmental Enterprises USA sures for the laboratory chronic toxicity tests. Comparison (Slidell, LA). This is contribution number 786 from the Southeast of the lowest TPAH concentrations from WAF exposures Environmental Research Center at Florida International University. for NOECs and survival for both species (4.75–5.00 lg/L) and the 90th percentile concentration indicates that there is a239 difference in TPAH concentrations between the 100 % WAF exposures (4.75–5.00 lg/L) used in the lab- References oratory toxicity tests (with Slick B) and the 90th percentile TPAH concentration (0.210 lg/L) of all field samples. Almeda R, Wambaugh Z, Wang Z, Hyatt C, Liu Z, Buskey EJ (2013) Interactions between zooplankton and crude oil: toxic effects and There also is a 109 difference between the lowest TPAH bioaccumulation of polycyclic aromatic hydrocarbons. PLoS exposures for NOECs and growth (\2.02 lg/L, in silver- ONE 8:e67212 sides) in the laboratory and the 90th percentile TPAH Anderson JW, Neff JM, Cox BA, Tatem HE, Hightower GM (1974) concentration of all field samples. Characteristics of dispersions and water-soluble extracts of crude and refined oils and their toxicity to estuarine crustaceans and In the most updated analyses of TPAHs (sum of 50 fish. Mar Biol 27:75–88 parent and alkylated PAHs) to characterize the distribution ASTM E1191-03a (2008) Standard guide for conducting life-cycle of water exposures in the GOM from the DWH incident toxicity tests with saltwater mysids. ASTM International, West (from May 5-December 2010), the following was shown Conshohocken Barron MG (2012) Ecological impacts of the Deepwater Horizon oil (Boehm et al. 2016): 39 % of the samples had no spill: implications for immunotoxicity. Toxicol Pathol detectable TPAHs; geometric mean of exposure concen- 40(2):315–320 trations for samples (n = 5408) from during the release Barron MG, Ka’aihue L (2003) Critical evaluation of CROSERF test period (May–August 3) was 0.023 lg/L and for samples methods for oil dispersant toxicity testing under subarctic conditions. Mar Pollut Bull 46:1191–1199 (n = 5999) from after the release period (August 4-De- Barron MG, Hemmer MJ, Jackson CR (2013) Developments of cember 2010) was 0.004 lg/L; 99 % of the samples during aquatic toxicity benchmarks for oil products using species post release period contained less than 1.0 lg/L TPAH; sensitivity distributions. Integr Environ Assess Manag 99:1–6 and TPAH concentrations [1 lg/L in deep waters 20 km Biomonitoring Science Advisory Board (BSAB) (1994) Criteria for acceptable variability of marine chronic toxicity test methods. from the well head were ‘‘infrequent.’’ Washington Department of Ecology BSAB Report #1 Boehm PD, Murray KJ, Cook LL (2016) Distribution and attenuation Table 4 TPAH 50–90th percentile concentrations from Gulf of of polycyclic aromatic hydrocarbons in Gulf of Mexico seawater Mexico water samples (n = 10,828) from the Deepwater Horizon Oil Accident. Env Sci Tech 50:584–592 a Field collected percentiles TPAH (tot *-50) lg/L Brette F, Machado B, Cros C, Incardona JP, Scholz NL, Block BA (2014) Crude oil impairs cardiac: excitation-contraction cou- 90th 0.210 pling in fish. Science 343:772–776 75th 0.044 Carls MG, Rice SD, Hose JE (1999) Sensitivity of fish embryos to 50th BDL (0.00597) weathered crude oil: Part I. Low level exposure during incubation causes malformations, genetic damage, and mortality BDL below detection limit in larval Pacific herring (Clupea pallasi). Environ Toxicol Chem a Data from field sample database (OSAT 2010) 18:481–493 123 Arch Environ Contam Toxicol

Clark JR, Bragin GE, Febbo EJ, Letinski DJ (2001) Toxicity of NOAA (2010) Analytical quality assurance plan. Mississippi Canyon physically and chemically dispersed oils under continuous and 252 (Deepwater Horizon) Natural Resource Damage Assess- environmentally realistic exposure conditions: applicability to ment. Version 2.2 dispersant use decisions in spill response planning. In: Proceed- OSAT (2010) Summary report for sub-sea and sub-surface oil and ings of the 2001 international oil spill conference, pp 1249–1255 dispersant detection: sampling and monitoring. Operational Crane M, Newman MC (2000) What level of effect is a no observed Science Advisory Team, Unified Area Command, Deepwater effect? Environ Toxicol Chem 19:516–519 Horizon MC252, 17 December 2010, pp 35 Dubansky B, Whitehead A, Miller J, Rice CD, Galvez F (2013) Multi- Pastorak RA, Anderson JW, Butcher MK, Sexton JE (1994) West tissue molecular, genomic, and developmental effects of the coast marine species chronic protocol variability study. Colum- Deepwater Horizon oil spill on resident Gulf killifish (Fundulus bia Aquatic Sciences (PTI Contract No. C333-03-01) grandis). Environ Sci Technol 47(10):5074–5082 Paul JH, Hollander D, Coble P, Daly KL, Murasko S, English D, Echols BS, Smith A, Gardinali P, Rand G (2016a) An evaluation of Basso J, Delaney J, McDaniel L, Kovach CW (2013) Toxicity select test variables potentially affecting acute oil toxicity. Arch and mutagenicity of Gulf of Mexico waters during and after the Env Contam Tox 70:392–405 Deepwater Horizon oil spill. Environ Sci Technol 47:9651–9659 Echols BS, Smith AJ, Gardinali PR, Rand GM (2016b) The use of Prince RC, Elmendorg DL, Lute LR, Hsu CS, Haith CE, Senius JD ephyrae of a scyphozoan jellyfish, Aurelia aurita, in the aquatic et al (1994) 17.Alpha. (H)-21.beta. (H)-hopane as a conserved toxicological assessment of Macondo oils from the Deepwater internal marker for estimating the biodegradation of crude oil. Horizon incident. Chemosphere 144:1893–1900 Environ Sci Technol 28:142–145 Faksness LG, Altin D, Nordtug T, Daling PS, Hansen BH (2015) Rand GM (1995) Fundamentals of aquatic toxicology-effects. Envi- Chemical comparison and acute toxicity of water accommodated ronmental fate and risk assessment. Taylor & Francis, Boca fraction (WAF) of source and field collected Macondo oils from Raton, FL the Deepwater Horizon spill. Mar Poll Bull 91:222–229 Rhoton SL, Perkins RA, Braddock JF, Behr-Andres C (2001) A cold- Fingas M (2011) Oil spill science and technology: prevention, weather species’ response to chemically dispersed fresh and response, and cleanup (Gulf Professional Publishing, Burlington weathered Alaska North Slope crude oil. In: Proceedings, 2001 MA, 2011), 1192 pp international oil spill conference, American Petroleum Institute, Giddings JM, Hall LW, Anderson TA, Hosmer AJ, Kendall RJ (2005) Washington, DC, pp 1231-1236 Atrazine in North American surface waters: a probabilistic Schuett K (2010) Source oil sample provenance (Updated Oct7 2010) aquatic ecological risk assessment. Society of Environmental Singer MM, Jacobson S (2001a) Acute effects of fresh versus Toxicology and Chemistry, Pensacola Gulf Science Data http:// weathered oil to marine organisms: California findings. In: gulfsciencedata.bp.com Proceedings of the 2001 international oil spill conference, API, Heintz R, Short J, Rice S (1999) Sensitivity of fish embryos to Washington, DC, pp 1263–1268 weathered crude oil: Part II. Increased mortality of pink salmon Singer MM, Aurand D, Bragin GE, Clark JR, Coelho GM, Sowby (Oncorhynchus gorbuscha) embryos incubating downstream ML, Tjeerdema RS (2000) Standardization of the preparation from weathered Exxon Valdez crude oil. Environ Toxicol Chem and quantitation of water-accommodated fractions of petroleum 18:494–503 for toxicity testing. Mar Pollut Bull 40:1007–1016 Incardona JP, Swarts TL, Edmunds RC, Linbo TL, Aquilina-Beck A, Singer MM, Aurand D, Coelho GM, Bragin GE, Clark JR, Sowby Sloan CA, Scholz NL (2013) Exxon Valdez to Deepwater ML, Tjeerdema RS (2001b) Making, measuring, and using water Horizon: comparable toxicity of both crude oils to fish early life accommodated fractions of petroleum for toxicity testing. stages. Aqua Toxicol 142:303–316 Proceedings, 2001 international oil spill conference. API, Incardona JP, Gardner LD, Linbo TL, Brown TL, Esbaugh AJ, Mager Washington, DC EM, Scholz NL (2014) Deepwater Horizon crude oil impacts the Solomon KR, Baker DB, Richards P, Dixon KR, Klaine SJ, La Point developing hearts of large predatory pelagic fish. Proc Nat Acad TW, Kendall RJ, Giddings JM, Giesy JP, Hall LW Jr, Weisskopf Sci 111:E1510–E1518 C, Williams M (1996) Ecological risk assessment of atrazine in Lee K, Boufadel M, Chen B, Foght J, Hodson P, Swanson S, Venosa North American surface waters. Environ Toxicol Chem A (2015) Expert panel report on the behaviour and environmen- 15:31–76 tal impacts of crude oil released into aqueous environments. U.S. Environmental Protection Agency (USEPA) (1991) Technical R Soc Can, Ottawa, ON. ISBN 978-1-928140-02-3 support document for water quality-based toxics control. U.S. Lehr BP et al. (2010) Oil Budget Calculator: Deepwater Horizon. Environmental Protection Agency, Washington, D.C. EPA-505- Technical Document: A Report to the National Incident Com- 2-90-001. 335 pp mand, November 2010. 217 pp U.S. Environmental Protection Agency (USEPA) (2002) Short-term NAS (2013) An ecosystem services approach to assess the impacts of methods for estimating the chronic toxicity of effluents and the Deepwater Horizon oil spill in the Gulf of Mexico. Chapter 4, receiving waters to marine and estuarine organisms, 3rd edn. Oil Spill Response Technologies. Committee on the effects of U.S. Environmental Protection Agency, Washington, D.C. EPA- the Deepwater Horizon Mississippi Canyon 252 oil spill on 821-R-02-014. 464 pp ecosystem services in the Gulf of Mexico, Ocean Science Board, Wu D et al (2012) Comparative toxicity of four chemically dispersed NRC. National Academy Press, Washington, DC and undispersed crude oils to rainbow trout embryos. Environ Neff JM (1999) Toxicity effects. Element 8. In: Robertson SB (ed) Toxicol Chem 31:754–765 Guidelines for the scientific study of oil spill effects. Robertson Environmental Services Publishers, Langley, pp 1–31

123 Biodiversity Data Journal 4: e8728 doi: 10.3897/BDJ.4.e8728

General Article

Five Years Later: An Update on the Status of Collections of Endemic Gulf of Mexico Fishes Put at Risk by the 2010 Oil Spill

Prosanta Chakrabarty‡,§, Glynn A. O’Neill|, Brannon Hardy |, Brandon Ballengee¶

‡ Louisiana State Unviersity Museum of Natural Science, Baton Rouge, Louisiana, United States of America § National Science Foundation, Arlington, Virginia, United States of America | Louisiana State University, Baton Rouge, United States of America ¶ Louisiana State Unviersity Museum of Natural Science, Baton Rouge, United States of America

Corresponding author: Prosanta Chakrabarty ([email protected])

Academic editor: Pavel Stoev

Received: 04 Apr 2016 | Accepted: 11 Aug 2016 | Published: 18 Aug 2016

Citation: Chakrabarty P, O’Neill G, Hardy B, Ballengee B (2016) Five Years Later: An Update on the Status of Collections of Endemic Gulf of Mexico Fishes Put at Risk by the 2010 Oil Spill. Biodiversity Data Journal 4: e8728. doi: 10.3897/BDJ.4.e8728

Abstract

Background

The 2010 Gulf of Mexico Oil Spill took place over 180,000 square kilometers during a 12- week period over five years ago; however, this event continues to influence the development and distribution of organisms in and around the region of the disaster. Here we examine fish species that may have been most affected by noting their past distribution in the region of the spill and examining data of known collecting events over the last 10 years (five years prior to the spill, five years post spill).

New information

We found that more than half of the endemic ﬁsh species of the Gulf (45 of 77)

© Chakrabarty P et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 2 Chakrabarty P et al.

Keywords

BP, Deepwater Horizon, Macondo, ichthyology, ﬁsh

Introduction

The 2010 Gulf of Mexico Oil Spill (also called the Deepwater Horizon/BP disaster/oil spill, or Macondo blowout among others) was the largest accidental spill of oil in history (Crone and Tolstoy 2010, Rabalais 2014). Coupled with the fact that it occurred in the deep sea (>1000 m depth) and with the coordinated release of more than a million gallons of dispersant, it is one of the greatest pollution events in history (Goodbody-Gringley et al. 2013). The long lasting effects of the spill are still not fully understood even five years after the event. There is considerable evidence that some species continue to be physically and developmentally challenged by the impact of the spill, particularly fishes (Whitehead et al. 2011; Incardona et al. 2014; Dubansky et al. 2013; Brette et al. 2014; Mager et al. 2014; Alloy et al. 2016). However, population studies of fishes remain poorly examined (Fodrie et al. 2014). Although, fisheries for commercial species are better studied, the ichthyofauna as a whole has received little attention. Chakrabarty et al. (2012) listed fish species in need of conservation concern based on their known distribution in relation to the historical surface position of the oil spill. Here we reexamine the distribution of all 77 known endemic Gulf fish species five years after the spill based on collection records (as a reminder endemic means in this context, species only found in the Gulf of Mexico). We compare these post-spill records with those from five years prior to the spill.

These collection records are obtained from natural history museum records of specimen collections. Museum collections are a vital source for biological records (Drew 2011; Rocha et al. 2014). They maintain a record of the world’s biodiversity by keeping specimens recorded from a certain area and time allowing comparisons to be made across time and space. With these collections one can compare a changing fauna before and after a catastrophic event, such as an oil spill. The correct identification of specimens is also vital (Chakrabarty et al. 2013), as museum collections are maintained by taxonomists and the specimens and comparative material are at hand, the identifications from these collections are more trustworthy than those from ship records or other sources where specimens are discarded. Here we use these collection records to examine the affects of the 2010 Gulf of Mexico Oil Spill on the endemic fishes of the region. Five Years Later: An Update on the Status of Collections of Endemic Gulf ... 3

Methods and Results

The occurrence records of the 77 endemic species of the Gulf of Mexico were tallied using The Global Biodiversity Information Facility and FishNet2 from October-December of 2015. Duplicate events from the two databases were deleted (duplicates were discovered if they had the same museum catalog numbers). A scatter plot graph was then created in Microsoft Excel showing collections five years prior to the 2010 Oil Spill and five years post spill. Only collections records from the Gulf of Mexico were counted (assuming for these endemics that records from outside the region are likely misidentifications).

Scatter plots of endemic fishes from the Gulf of Mexico are shown below with the “Number of Occurrence(s)” on the y-axis vs. the “Number of Years” on the x-axis. Species are listed in alphabetical order. Endemic species that have few or no collections records do not have a scatterplot but details about their last collecting events are presented. The scientific name is also presented followed by common name (when there is one) and family. Spill zone overlap information is from Chakrabarty et al. (2012). If the scientific name has changed in the past five years we show both the old and new names. Conservation information about “Resilience” is taken from FishBase (Froese and Pauly 2016). Resilience is based upon the time it takes to double the species population and are as follows: Very Low (minimum of 14 years to double population); Low (4.5-14 years to double population); Medium (1.4-4.4 years to double population); High (less than 15 months to double population).

1) Alosa alabamae - Alabama Shad – Clupeidae (1% range overlap with spill zone). Resilience: Medium (Fig. 1)

Figure 1. Alosa alabamae

2) Alosa chrysochloris - Skipjack Shad – Clupeidae (2% range overlap with spill zone). Resilience: Medium (Fig. 2) 4 Chakrabarty P et al.

Figure 2. Alosa chrysochloris

3) Anacanthobatis folirostris - Leaf-nose Leg Skate – Anacanthobatidae (79% range overlap with spill zone). Resilience: Low. – last time collected: 2004

4) Atherinella schultzi - Chimalapa Silverside – Atherinopsidae (No range overlap with spill zone). Resilience: High – collected once (2013) since 2005

5) Atractosteus spatula – Alligator Gar – Lepisosteidae (No range overlap with spill zone). Resilience: Very low (Fig. 3)

Figure 3. Atractosteus spatula Five Years Later: An Update on the Status of Collections of Endemic Gulf ... 5

6) Bollmannia communis – Ragged Goby – Gobiidae (41% range overlap with spill zone). Resilience: High (Fig. 4)

Figure 4. Bollmannia communis

7) Bollmannia eigenmanni – Shelf Goby – Gobiidae (64% range overlap with spill zone). Resilience: Medium – last time collected: 1988

8) Brevoortia gunteri – Finescale Menhaden – Clupeidae (2% range overlap with spill zone). Resilience: Medium (Fig. 5)

Figure 5. Brevoortia gunteri 6 Chakrabarty P et al.

9) Brevoortia patronus – Gulf Menhaden – Clupeidae (11% range overlap with spill zone). Resilience: Medium (Fig. 6)

Figure 6. Brevoortia patronus

10) Calamus arctifrons – Grass Porgy – Sparidae (No range overlap with spill zone). Resilience: Medium (Fig. 7)

Figure 7. Calamus arctifrons Five Years Later: An Update on the Status of Collections of Endemic Gulf ... 7

11) Calamus campechanus – Campeche Porgy – Sparidae (No range overlap with spill zone). Resilience: Medium – last time collected: 1987

12) Chasmodes longimaxilla – Stretchjaw Blenny – Blenniidae (No range overlap with spill zone). Resilience: High – last time collected: 1983

13) Chriolepis benthonis – Deepwater Goby – Gobiidae (No range overlap with spill zone). Resilience: High – last time collected: 1953

14) Chriolepis vespa – Wasp Goby – Gobiidae (No range overlap with spill zone). Resilience: High – last time collected: 1970

15) Citharichthys abbotti – Veracruz Whiﬀ – Paralichthyidae (No range overlap with spill zone). Resilience: High – last time collected: 2001

16) Coryphaenoides mexicanus – Mexican Grenadier – Macrouridae (54% range overlap with spill zone). Resilience: Medium (Fig. 8)

Figure 8. Coryphaenoides mexicanus

17) Coryphopterus punctipectophorus – Spotted Goby – Gobiidae (No range overlap with spill zone). Resilience: High – last time collected: 1982

18) Ctenogobius claytonii – Mexican Goby – Gobiidae (No range overlap with spill zone). Resilience: High – collected once (2005) since 2005

19) Cynoscion arenarius – Sand Weakﬁsh – Sciaenidae (12% range overlap with spill zone). Resilience: Medium (Fig. 9) 8 Chakrabarty P et al.

Figure 9. Cynoscion arenarius

20) Dipturus olseni – Spreadﬁn Skate – Rajidae (29% range overlap with spill zone). Resilience: Low – collected twice (2005) since 2005

21) Dipturus oregoni – Hooktail Skate – Rajidae (80% range overlap with spill zone). Resilience: Low – last time collected: 1987

22) Eptatretus minor – Hagﬁsh – Myxinidae (23% range overlap with spill zone). Resilience: Low – collected twice (2005) since 2005

23) Eptatretus springeri – Gulf hagﬁsh – Myxinidae (54% range overlap with spill zone). Resilience: Low – collected once (2010) since 2005

24) Etmopterus schultzi – Fringeﬁn Lanternshark – Etmopteridae (90% range overlap with spill zone). Resilience: Low – collected ﬁve times (2006) since 2005

25) Eustomias leptobolus – Stomiidae (40% range overlap with spill zone). Resilience: High – last time collected: 1960

26) Exechodontes daidaleus – Zoarcidae (No range overlap with spill zone). Resilience: High – last time collected: 1989

27) Floridichthys carpio – Goldspotted killiﬁsh – Cyprinodontidae (No range overlap with spill zone). Resilience: High (Fig. 10) Five Years Later: An Update on the Status of Collections of Endemic Gulf ... 9

Figure 10. Floridichthys carpio

28) Fundulus grandis – Gulf Killiﬁsh – Fundulidae (13% range overlap with spill zone). Resilience: High (Fig. 11)

Figure 11. Fundulus grandis

29) Fundulus jenkinsi – Saltmarsh Topminnow – Fundulidae (4% range overlap with spill zone). Resilience: High (Fig. 12) 10 Chakrabarty P et al.

Figure 12. Fundulus jenkinsi

30) Fundulus persimilis – Yucatán Killiﬁsh – Fundulidae (No range overlap with spill zone). Resilience: High – collected twice in 2005

31) Fundulus pulvereus – Bayou Killiﬁsh – Fundulidae (18% range overlap with spill zone). Resilience: High (Fig. 13)

Figure 13. Fundulus pulvereus

32) Fundulus xenicus (formerly Adinia xenica) – Diamond Killiﬁsh – Fundulidae (13% range overlap with spill zone). Resilience: Low (Fig. 14) Five Years Later: An Update on the Status of Collections of Endemic Gulf ... 11

Figure 14. Fundulus xenicus

33) Gambusia yucatana – Yucatan Mosquitoﬁsh – Poeciliidae (No range overlap with spill zone). Resilience: High (Fig. 15)

Figure 15. Gambusia yucatana

34) Gobiosoma longipala – Twoscale Goby – Gobiidae (No range overlap with spill zone). Resilience: High – collected 2 times (2012) since 2005

35) Gordiichthys ergodes – Irksone Eel – Ophichthidae (No range overlap with spill zone). Resilience: Medium (Fig. 16) 12 Chakrabarty P et al.

Figure 16. Gordiichthys ergodes

36) Gordiichthys leibyi – String Eel – Ophichthidae (No range overlap with spill zone). Resilience: Medium – last time collected: 2004

37) Gunterichthys longipenis – Gold Brotula – Bythitidae (88% range overlap with spill zone). Resilience: Low – last time collected: 2002

38) Gymnachirus texae – Gulf of Mexico Fringed Sole – Achiridae (16% range overlap with spill zone). Resilience: High – collected once (2012) since 2005

39) Halichoeres burekae – Mardi Gras Wrasse – Labridae (No range overlap with spill zone). Resilience: High – collected twice (2006) since 2005

40) Halieutichthys intermedius – Louisiana Pancake Batﬁsh – Ogcocephalidae (68% range overlap with spill zone). Resilience: High – collected ﬁve times (2010) since 2005

41) Heteroconger luteolus – Yellow Garden Eel – Congridae (No range overlap with spill zone). Resilience: Medium – last time collected: 2004

42) Hyperoglyphe bythites – Black Driftﬁsh – Centrolophidae (82% range overlap with spill zone). Resilience: Medium – collected once (2008) since 2005

43) Hypleurochilus caudovittatus – Zebratail Blenny – Blenniidae (Insuﬃcient data) Resilience: High – last time collected: 2004

44) Hypleurochilus multiﬁlis – Featherduster Blenny – Blenniidae (25% range overlap with spill zone). Resilience: High – last time collected: 2001

45) Ijimaia antillarum – Ateleopodidae (8% range overlap with spill zone). Resilience: Unknown – last time collected: 2004 Five Years Later: An Update on the Status of Collections of Endemic Gulf ... 13

46) Jordanella ﬂoridae – Flagﬁsh – Cyprinodontidae (No range overlap with spill zone). Resilience: Low (Fig. 17)

Figure 17. Jordanella ﬂoridae

47) Jordanella pulchra (previously Garmanella pulchra) – Yucatán ﬂagﬁsh – Cyprinodontidae (No range overlap with spill zone). Resilience: High – collected 10 times (2005) since 2005

48) Lepisosteus oculatus – Spotted Gar – Lepisosteidae (0.2% range overlap with spill zone). Resilience: Medium (Fig. 18)

Figure 18. Lepisosteus oculatus 14 Chakrabarty P et al.

49) Leucoraja lentiginosa – Freckled Skate – Rajidae (53% range overlap with spill zone). Resilience: Low – collected once (2012) since 2005

50) Lupinoblennius nicholsi – Highﬁn Blenny – Blenniidae (No range overlap with spill zone). Resilience: High – last time collected: 2000

51) Lycenchelys bullisi – Zoarcidae (50% range overlap with spill zone). Resilience: Medium – last time collected: 1999

52) Menidia clarkhubbsi – Texas Silverside – Atherinopsidae (No range overlap with spill zone). Resilience: High – last time collected: 2000

53) Menidia colei –Golden Silverside – Atherinopsidae (No range overlap with spill zone). Resilience: High – collected 29 times (2005) since 2005

54) Menidia conchorum – Key Silverside – Atherinopsidae (No range overlap with spill zone). Resilience: High – last time collected: 1978

55) Microdesmus lanceolatus – Lancetail Wormﬁsh – Microdesmidae (43% range overlap with spill zone). Resilience: High – last time collected: 1980

56) Monopenchelys acuta – Redface Moray – Muraenidae (No range overlap with spill zone). Resilience: High (Fig. 19)

Figure 19. Monopenchelys acuta

57) Mustelus sinusmexicanus – Gulf Smooth-hound – Triakidae (43% range overlap with spill zone). Resilience: Low (Fig. 20) Five Years Later: An Update on the Status of Collections of Endemic Gulf ... 15

Figure 20. Mustelus sinusmexicanus

58) Neoopisthopterus cubanus – Cuban Longﬁn Herring – Pristigasteridae (Insuﬃcient data). Resilience: High – last time collected: N/A

59) Ogcocephalus pantostictus – Spotted Batﬁsh –Ogcocephalidae (3% range overlap with spill zone). Resilience: Low (Fig. 21)

Figure 21. Ogcocephalus pantostictus

60) Ogilbia cayorum – Key Brotula – Bythitidae (No range overlap with spill zone). Resilience: Low (Fig. 22) 16 Chakrabarty P et al.

Figure 22. Ogilbia cayorum

61) Oneirodes bradburyae – Oneirodidae (100% range overlap with spill zone). Resilience: High – last time collected: 1954

62) Ophichthus omorgmus – Dotted Snake Eel – Ophichthidae (Insuﬃcient data). Resilience: Medium – last time collected: 1999

63) Ophichthus rex – King Snake Eel – Ophichthidae (82% range overlap with spill zone). Resilience: Very low – collected once (2009) since 2005

64) Opsanus pardus – Leopard Toadﬁsh – Batrachoididae (38% range overlap with spill zone). Resilience: Low (Fig. 23)

Figure 23. Opsanus pardus Five Years Later: An Update on the Status of Collections of Endemic Gulf ... 17

65) Parasaccogaster rhamphidognatha (previously Saccogaster rhamphidognatha) – (100% range overlap with spill zone). Resilience: High – last time collected: N/A

66) Parmaturus campechiensis – Campeche Catshark – Pentanchidae (Insuﬃcient data). Resilience: Low – last time collected: 1970

67) Prionotus longispinosus – Bigeye Sea Robin – Triglidae (50% range overlap with spill zone). Resilience: Medium (Fig. 24)

Figure 24. Prionotus longispinosus

68) Prionotus martis – Gulf of Mexico Barred Sea Robin – Triglidae (5% range overlap with spill zone). Resilience: High (Fig. 25)

Figure 25. Prionotus martis 18 Chakrabarty P et al.

69) Prionotus paralatus – Mexican Sea Robin – Triglidae (Insuﬃcient data). Resilience: High (Fig. 26)

Figure 26. Prionotus paralatus

70) Raja texana – Roundel Skate – Rajidae (11% range overlap with spill zone). Resilience: Low (Fig. 27)

Figure 27. Raja texana

71) Sanopus reticulates – Reticulate toadﬁsh – Batrachoididae (Insuﬃcient data). Resilience: Medium – last time collected: 1977 Five Years Later: An Update on the Status of Collections of Endemic Gulf ... 19

72) Sphoeroides parvus – Least Puﬀer – Tetraodontidae (Insuﬃcient data). Resilience: High (Fig. 28)

Figure 28. Sphoeroides parvus

73) Sphoeroides spengleri – Bandtail Puﬀer – Tetraodontidae (.4% range overlap with spill zone). Resilience: High (Fig. 29)

Figure 29. Sphoeroides spengleri

74) Stemonosudis bullisi – Paralepididae (Insuﬃcient data). Resilience: High – last time collected: 1960 20 Chakrabarty P et al.

75) Syngnathus aﬃnis – Texas Pipeﬁsh – Syngnathidae (No range overlap with spill zone). Resilience: High – last time collected: 1983

76) Trichopsetta ventralis – Sash Flounder – Bothidae (31% range overlap with spill zone). Resilience: Medium (Fig. 30)

Figure 30. Trichopsetta ventralis

77) Varicus marilynae – Orangebelly Goby – Gobiidae (No range overlap with spill zone). Resilience: High – last time observed: 1974

Discussion

The continued influence of an oil spill that occurred more than five years ago on the Gulf of Mexico is evident (Incardona et al. 2014; Alloy et al. 2016; Schaefer et al. 2015); however, data about population status, or even tangible proof of the continued existence of many of the Gulf’s endemic fish species, is lacking. More than half (45) of the 77 endemic species from the Gulf of Mexico have not been officially collected since the 2010 spill. Of these, nine species have not been collected since before 1980, eight species have not been collected since the 1980s, and two not since the 1990s. Although there is a focus on fisheries data for commercially important species post-spill, the endemic species examined here are among the Gulf species we know the least about. Even with the data presented here our study of collections records must be viewed as a small glimpse into the true effects of the spill. Collections records are not a true estimate of population dynamics; however, in the case of rare and poorly studied species (as is the case with these endemics) – it is our best estimate. Five Years Later: An Update on the Status of Collections of Endemic Gulf ... 21

The species we should perhaps be most concerned for are the 14 that have collection records in the ﬁve years before the spill, but lack records post-spill (2010-2015). Among these are Fundulus jenkinsi collected 306x, Menidia colei (29x), Jordanella pulchra (10x), Ogilbia cayorum (6x), and Etmopterus schultzi and Monopenchelys acuta both collected 5x. Gambusia yucatana was collected 14x in the last 10 years, and all but one of those was pre-spill.

Other species appear to be more common post-spill, with most of the collections occuring in the last five years (rather than the 2005-2010 period): Trichopsetta ventralis (6 of 8 collections post-2010), Sphoeroides parvus (83 of 109), Prionotus longispinous (203 of 206), Prionotus paralatus (74 of 76), Opsanus pardus (6 of 7), Ogcocephalus pantostictus (6 of 6), Gobiosoma longipala (2 of 2). It should be noted that all the collections of Halieutichthys intermedius are post-spill because this species was described in 2012 (Ho et al. 2012) and most museums have not updated their records for this species. Some of the species that had higher collections numbers post spill may have been influenced by the closing of fisheries during and after the immediate period of the oil spill (Schaefer et al. 2015). Although not directly targeted for fisheries these species may have increased in number because they were not collected as by-catch when fishing was closed. Also the increased interest in collecting and studying Gulf species post spill may have increased efforts to identify and catalogue these species. We also note here that the collections efforts pre- and post-spill were likely not equal. We therefore cannot do a statistical sampling comparison based on collecting effort.

There are some notable trends among and within groups as well. Of the six eels in the study (Elopomorpha Families: Ophichthidae, Muraenidae, Congridae) only one species, Ophichthus rex had a high percentage of its range in the region of the spill (82%) and it has been collected once since the spill. However, eel species in general are very rare in collections, and little or no data about any of the endemic eels from the Gulf of Mexico is known (9 total collection records, all post spill).

Of the seven cartilaginous fishes (Elasmobranchii Families: Anacanthobatidae, Rajidae, Etmopteridae, Triakidae) most had a high proportion of their range in the area of the spill zone but most have post-spill collections. The exception being the rare Anacanthobatis folirostris, which has no collection records since 2004. These elasmobranchs all have low resiliency, with populations doubling time between 4.5-14 years (Froese and Pauly 2016). Most members of the small but diverse members of gobies (Gobioidei) and blennies (Blennioidei) lack sufficient information (in being collected mostly before 2005), as is the case for most of the ten coral associated endemic Gulf species (Table 1). Inshore brackish fishes such as those in the families Lepisosteidae, Clupeidae, Atherinopsidae, Fundulidae, Poeciliidae, and Cyprinodontidae, were mainly out of the area of the immediate spill (i.e., little overlap with the region of the spill as initially measured) and are among the most collected species among Gulf endemics (Table 1 ). However, although the collections may be high, the documented developmental impairment of near shore species points to the fact that even these species are not out of harms way (Dubansky et al. 2013). Additionally, 22 Chakrabarty P et al. the influence of the oil slick at the surface on pelagic larvae and in the deep-sea on individuals that are rarely seen will never be completely known (Fodrie and Heck 2011).

Table 1. Summary of species occurrence records (based on GBIF and FishNET2), and habitat types (from McEachran 2009; Chakrabarty et al. 2012). Taxa that were deemed ‘‘Species of Greatest Concern’’ by Chakrabarty et al. (2012) are in bold. These species had 35% of their historical occurrence records in the region of the oil spill.

Species: Family Occurrences: Occurrences: Habitat Scientiﬁc name 2010-present 2005-present

Alosa alabamae Clupeidae 12 24 Bay and Near Shore, Anadromous, Neritic

Alosa chrysochloris Clupeidae 47 177 Bay and Near Shore, Anadromous, Neritic

Anacanthobatis Anacanthobatidae 0 0 Slope folirostris

Atherinella schultzi Atherinopsidae 1 1 Bay and Near Shore, Estuarine

Atractosteus spatula Lepisosteidae 15 29 Bay and Near Shore, Neritic, Estuarine

Bollmannia communis Gobiidae 4 5 Demersal, Soft Substrates

Bollmannia Gobiidae 0 0 Demersal eigenmanni

Brevoortia gunteri Clupeidae 9 17 Bay and Near Shore, Neritic, Estuarine

Brevoortia patronus Clupeidae 85 180 Bay and Near Shore, Neritic, Estuarine

Calamus arctifrons Sparidae 9 32 Demersal, Seagrass

Calamus campechanus Sparidae 0 0 Demersal

Chasmodes longimaxilla Blenniidae 0 0 Demersal, Coral Reef

Gobiidae 0 0 Demersal

Citharichthys abbotti Paralichthyidae 0 0 Demersal, Soft Substrates

Coryphaenoides Macrouridae 2 2 Benthopelagic, Slope, mexicanus Abyssal

Coryphopterus Gobiidae 0 0 Demersal, Coral Reef punctipectophorus

Ctenogobius claytonii Gobiidae 0 1 Demersal, Bay and Near Shore, Estuarine Five Years Later: An Update on the Status of Collections of Endemic Gulf ... 23

Cynoscion arenarius Sciaenidae 33 90 Demersal, Beach and Shoreline, Soft Substrates

Dipturus olseni Rajidae 0 2 Demersal, Slope

Rajidae 0 0 Slope

Eptatretus minor Myxinidae 0 2 Slope, Soft Substrates, Burrower

Eptatretus springeri Myxinidae 1 1 Slope, Soft Substrates, Burrower

Etmopteridae 0 5 Slope

Eustomias leptobolus Stomiidae 0 0 Mesopelagic

Exechodontes daidaleus Zoarcidae 0 0 Benthic, Slope

Floridichthys carpio Cyprinodontidae 3 17 Bay and Near Shore, Estuarine, Seagrass

Fundulus grandis Fundulidae 97 292 Bay and Near Shore, Estuarine, Seagrass

Fundulus jenkinsi Fundulidae 0 306 Bay and Near Shore, Estuarine

Fundulus persimilis Fundulidae 0 2 Bay and Near Shore, Estuarine

Fundulus pulvereus Fundulidae 35 69 Bay and Near Shore, Estuarine

Fundulus xenicus Fundulidae 20 92 Bay and Near Shore, Estuarine

Gambusia yucatana Poeciliidae 1 14 Bay and Near Shore, Estuarine

Gobiosoma longipala Gobiidae 2 2 Demersal, Soft Substrates

Gordiichthys ergodes Ophichthidae 0 3 Demersal, Burrower, Soft Substrates

Gordiichthys leibyi Ophichthidae 0 0 Demersal, Soft Substrates, Burrower

Gunterichthys Bythitidae 0 0 Demersal, Bay and longipenis Near Shore, Burrower

Gymnachirus texae Achiridae 1 1 Demersal, Soft Substrates

Halichoeres burekae Labridae 0 2 Coral Reef

Halieutichthys Ogcocephalidae 5 5 Benthic, Soft intermedius Substrates

Heteroconger luteolus Congridae 0 0 Demersal

Hyperoglyphe bythites Centrolophidae 0 1 Benthopelagic

Hypleurochilus Blenniidae 0 0 Demersal, Soft caudovittatus Substrates 24 Chakrabarty P et al.

Hypleurochilus multiﬁlis Blenniidae 0 0 Demersal, Coral Reef

Ijimaia antillarum Ateleopodidae 0 0 Benthic, Slope

Jordanella ﬂoridae Cyprinodontidae 19 40 Bay and Near Shore, Estuarine, Seagrass,

Jordanella pulchra Cyprinodontidae 0 10 Bay and Near Shore, Estuarine

Lepisosteus oculatus Lepisosteidae 84 146 Neritic, Bay and Near Shore, Estuarine

Leucoraja lentiginosa Rajidae 1 1 Demersal, Slope

Lupinoblennius nicholsi Blenniidae 0 0 Demersal

Lycenchelys bullisi Zoarcidae 0 0 Benthic, Slope

Menidia clarkhubbsi Atherinopsidae 0 0 Bay and Near Shore, Estuarine

Menidia colei Atherinopsidae 0 29 Bay and Near Shore, Estuarine

Menidia conchorum Atherinopsidae 0 0 Bay and Near Shore, Coral Reef

Microdesmus Microdesmidae 0 0 Demersal, Bay and lanceolatus Near Shore, Burrower

Monopenchelys acuta Muraenidae 0 5 Demersal, Coral Reef

Mustelus Triakidae 2 0 Soft Substrates sinusmexicanus

Neoopisthopterus Clupeidae 0 0 Neritic, Bay and Near cubanus Shore, Beach and Shoreline, Estuarine

Ogcocephalus Ogcocephalidae 6 6 Demersal pantostictus

Ogilbia cayorum Bythitidae 0 6 Demersal, Hard Substrate

Oneirodes bradburyae Oneirodidae 0 0 Bathypelagic

Ophichthus omorgmus Ophichthidae 0 0 Benthic, Slope, Soft Substrates

Ophichthus rex Ophichthidae 0 1 Demersal, Soft Substrates, Burrower

Opsanus pardus Batrachoididae 6 7 Demersal, Hard Substrates

Parasaccogaster Bythitidae 0 0 Benthic, Slope, Soft rhamphidognatha Substrates

Parmaturus Scyliorhinidae 0 0 Slope, Soft Substrates campechiensis

Prionotus Triglidae 203 207 Demersal, Soft longispinosus Substrates

Prionotus martis Triglidae 24 26 Demersal Five Years Later: An Update on the Status of Collections of Endemic Gulf ... 25

Prionotus paralatus Triglidae 74 76 Demersal, Benthic, Slope

Raja texana Rajidae 2 6 Demersal

Sanopus reticulatus Batrachoididae 0 0 Coastal Surface and Epipelagic, Demersal

Sphoeroides parvus Tetraodontidae 83 109 Demersal, Bay and Near Shore

Sphoeroides spengleri Tetraodontidae 50 93 Demersal, Coral Reef, Seagrass

Stemonosudis bullisi Paralepididae 0 0 Mesopelagic

Syngnathus aﬃnis Syngnathidae 0 0 Benthopelagic, Bay and Near Shore, Seagrass

Trichopsetta ventralis Bothidae 6 8 Demersal, Benthic, Soft Substrates

Varicus marilynae Gobiidae 0 0 Demersal

More than quarter of the Gulf of Mexico endemic fish species (20) had greater than 35% of their historical records in the area of the spill zone (Chakrabarty et al. 2012; those in bold text in Table 1). These species were identified by Chakrabarty et al. (2012) as being in the highest potential impact category. Of these species half (10 species) still lack any collection records post spill. We note that both GBIF and FishNET are not perfect records of all collecting events or even all museum collections. Also we note that these databases are dynamic and change on a near daily basis as museum records are uploaded and updated. For that reason the data in this paper should be taken as a snapshot of the information available at this time. It is clear more work needs to be done tofind and potentially protect these endemic taxa. Future work will include citizen science projects by the authors (see Acknowledgements) and others, that will target Gulf endemics and add data, museum records, and increase community awareness. We hope this study helps focus conservation efforts on those species that lack the most information, or that have not been collected post-spill.

Acknowledgements

We thank the National Academies Keck Futures Initiative for funding to BB and PC - Crude Life: A Citizen Art and Science Investigation of Gulf of Mexico Biodiversity after the Deepwater Horizon Oil Spill.

References

• Alloy M, Baxter D, Stieglitz J, Mager E, Hoenig R, Benetti D, Grosell M, Oris J, Roberts A (2016) Ultraviolet Radiation Enhances the Toxicity of Deepwater Horizon Oil to Mahi- mahi (Coryphaena hippurus) Embryos. Environmental Science & Technology 50 (4): 2011‑2017. DOI: 10.1021/acs.est.5b05356 26 Chakrabarty P et al.

• Brette F, Machado B, Cros C, Incardona JP, Scholz NL, Block BA (2014) Crude Oil Impairs Cardiac Excitation-Contraction Coupling in Fish. Science 343 (6172): 772‑776. DOI: 10.1126/science.1242747 • Chakrabarty P, Warren M, Page L, Baldwin C (2013) GenSeq: An updated nomenclature and ranking for genetic sequences from type and non-type sources. ZooK e ys 346: 29‑41. DOI: 10.3897/zookeys.346.5753 • Chakrabarty P, Lam C, Hardman J, Aaronson J, House P, Janies D (2012) SpeciesMap: a web-based application for visualizing the overlap of distributions and pollution events, with a list of fishes put at risk by the 2010 Gulf of Mexico oil spill. Biodiversity and Conservation 21 (7): 1865‑1876. DOI: 10.1007/s10531-012-0284-4 • Crone TJ, Tolstoy M (2010) Magnitude of the 2010 Gulf of Mexico Oil Leak. Science 330 (6004): 634‑634. DOI: 10.1126/science.1195840 • Drew J (2011) The Role of Natural History Institutions and Bioinformatics in Conservation Biology. Conservation Biology 25 (6): 1250‑1252. DOI: 10.1111/ j.1523-1739.2011.01725.x • Dubansky B, Whitehead A, Miller J, Rice C, Galvez F (2013) Multitissue Molecular, Genomic, and Developmental Effects of the Deepwater Horizon Oil Spill on Resident Gulf Killifish (Fundulus grandis). Environmental Science & Technology 47 (10): 5074‑5082. DOI: 10.1021/es400458p • Fodrie FJ, Heck K (2011) Response of Coastal Fishes to the Gulf of Mexico Oil Disaster. PLoS ONE 6 (7): e21609. DOI: 10.1371/journal.pone.0021609 • Fodrie FJ, Able KW, Galvez F, Heck KL, Jensen OP, Lopez-Duarte PC, Martin CW, Turner RE, Whitehead A (2014) Integrating Organismal and Population Responses of Estuarine Fishes in Macondo Spill Research. BioScience 64 (9): 778‑788. DOI: 10.109 3/biosci/biu123 • Froese R, Pauly D (2016) FishBase. FishBase 1: 1. URL: www.fishbase.org • Goodbody-Gringley G, Wetzel D, Gillon D, Pulster E, Miller A, Ritchie K (2013) Toxicity of Deepwater Horizon Source Oil and the Chemical Dispersant, Corexit® 9500, to Coral Larvae. PLoS ONE 8 (1): e45574. DOI: 10.1371/journal.pone.0045574 • Incardona JP, Gardner LD, Linbo TL, Brown TL, Esbaugh AJ, Mager EM, Stieglitz JD, French BL, Labenia JS, Laetz CA, Tagal M, Sloan CA, Elizur A, Benetti DD, Grosell M, Block BA, Scholz NL (2014) PNAS Plus: From the Cover: Deepwater Horizon crude oil impacts the developing hearts of large predatory pelagic fish. Proceed ings of the National Academy of Sciences 111 (15): E1510‑E1518. DOI: 10.1073/ pnas.1320950111 • Mager E, Esbaugh A, Stieglitz J, Hoenig R, Bodinier C, Incardona J, Scholz N, Benetti D, Grosell M (2014) Acute Embryonic or Juvenile Exposure to Deepwater Horizon Crude Oil Impairs the Swimming Performance of Mahi-Mahi (Coryphaena hippurus). Environmental Science & Technology 48 (12): 7053‑7061. DOI: 10.1021/es501628k • Rabalais N (2014) Assessing Early Looks at Biological Responses to the Macondo Event. BioScience 64 (9): 757‑759. DOI: 10.1093/biosci/biu132 • Rocha LA, Aleixo A, Allen G, Almeda F, Baldwin CC, Barclay MVL, Bates JM, Bauer AM, Benzoni F, Berns CM, Berumen ML, Blackburn DC, Blum S, Bolanos F, Bowie RCK, Britz R, Brown RM, Cadena CD, Carpenter K, Ceriaco LM, Chakrabarty P, Chaves G, Choat JH, Clements KD, Collette BB, Collins A, Coyne J, Cracraft J, Daniel T, de Carvalho M, Queiroz Kd, Dario FD, Drewes R, Dumbacher JP, Engilis A, Erdmann MV, Eschmeyer W, Feldman CR, Fisher BL, Fjeldsa J, Fritsch PW, Fuchs J, Getahun A, Five Years Later: An Update on the Status of Collections of Endemic Gulf ... 27

Gill A, Gomon M, Gosliner T, Graves GR, Griswold CE, Guralnick R, Hartel K, Helgen KM, Ho H, Iskandar DT, Iwamoto T, Jaafar Z, James HF, Johnson D, Kavanaugh D, Knowlton N, Lacey E, Larson HK, Last P, Leis JM, Lessios H, Liebherr J, Lowman M, Mahler DL, Mamonekene V, Matsuura K, Mayer GC, Mays H, McCosker J, McDiarmid RW, McGuire J, Miller MJ, Mooi R, Mooi RD, Moritz C, Myers P, Nachman MW, Nussbaum RA, Foighil DO, Parenti LR, Parham JF, Paul E, Paulay G, Perez-Eman J, Perez-Matus A, Poe S, Pogonoski J, Rabosky DL, Randall JE, Reimer JD, Robertson DR, Rodel M-O, Rodrigues MT, Roopnarine P, Ruber L, Ryan MJ, Sheldon F, Shinohara G, Short A, Simison WB, Smith-Vaniz WF, Springer VG, Stiassny M, Tello JG, Thompson CW, Trnski T, Tucker P, Valqui T, Vecchione M, Verheyen E, Wainwright PC, Wheeler TA, White WT, Will K, Williams JT, Williams G, Wilson EO, Winker K, Winterbottom R, Witt CC (2014) Specimen collection: An essential tool. Science 344 (6186): 814‑815. DOI: 10.1126/science.344.6186.814 • Schaefer J, Frazier N, Barr J (2015) Dynamics of Near-Coastal Fish Assemblages following the Deepwater Horizon Oil Spill in the Northern Gulf of Mexico. Transactions of the American Fisheries Society 145 (1): 108‑119. DOI: 10.1080/00028487.2015.1111253 • Whitehead A, Dubansky B, Bodinier C, Garcia TI, Miles S, Pilley C, Raghunathan V, Roach JL, Walker N, Walter RB, Rice CD, Galvez F (2011) Genomic and physiological footprint of the Deepwater Horizon oil spill on resident marsh ﬁshes. Proceedings of the National Academy of Sciences 109 (50): 20298‑20302. DOI: 10.1073/pnas.1109545108