The Ascidian Styela Plicata As a Potential Bioremediator of Bacterial and Algal Contamination of Marine Estuarine Waters

THE ASCIDIAN STYELA PLICATA AS A POTENTIAL BIOREMEDIATOR OF

BACTERIAL AND ALGAL CONTAMINATION OF MARINE ESTUARINE

WATERS

Lisa Denham Draughon

A Dissertation Submitted to the Faculty of

The Charles E. Schmidt College of Science

In Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

Florida Atlantic University

Boca Raton, Florida

May, 2010

ACKNOWLEDGEMENTS

I wish to express my sincerest gratitude to those who made this research possible;

Ocean Restoration Corporation Associates (ORCA), the Link Foundation, and Florida

Atlantic University (FAU) for their financial fellowships, and for the many hands that

played a part in getting it all done. A special thank you goes to my committee members;

Dr. James X. Hartmann (chairperson) and Dr. David Binninger of FAU, Dr. Ed Proffitt

and Dr. John Scarpa of Harbor Branch Oceanographic Institute at FAU, and Dr. Ray

Waldner of Palm Beach Atlantic University. You were all such an integral part of helping me achieve this goal. Special thanks also go to my friend Karen Halloway-

Adkins who was always willing to help locate specimens and provide equipment when she had it. Mike Calinski of ORCA made trips from the Gulf Coast to deliver specimens to me. Sherry Reed of the Smithsonian Marine Station in Ft. Pierce was extremely helpful by keeping an eye out for the tunicates and at times gathering them for me. Dr.

Patricia Keating was invaluable for her help in flow-cytometry analysis. Dr. Peter

McCarthy of Harbor Branch Oceanographic Institute at FAU provided the bacteria for the filtration rate experiment. Dr. Esther Guzman and Tara Pitts, also from HBOI-FAU, were extremely helpful during the ATP assay. Ashley Callahan, your help in Panama is so much appreciated. And finally but definitely not least, I want to thank all the undergraduate students that assisted during the experiments, especially Catherine (Katie)

Lord and Danielle (Dani) Nangle. I can’t thank you enough.

PREFACE

This dissertation, which focuses on the potential use of the tunicate Styela plicata for bioremediation, is organized into one general background chapter followed by individual chapters that have already been or will be submitted for publication. For this reason, there will be redundancy in most of the chapter introductions and discussions as different facets of the tunicate’s use are explored.

Portions of Chapter Two have been compiled and reprinted with permission from two separate publications:

Draughon, L., Scarpa, J., Keating, P., & Hartmann, J. (2008). Potential estuarine water

quality improvement via marine invertebrate bioremediation. In M. Theophanides &

T. Theophanides (Eds.), Environmental Awareness and Management (pp. 97-112).

Athens: ATINER.

and

Draughon, L. D., Scarpa, J., & Hartmann, J. X. (2010). Are filtration rates for the rough

tunicate Styela plicata independent of weight or size? Journal of Environmental

Science and Health, Part A: Toxic/Hazardous Substances and Environmental

Engineering, 45(2), 168 - 176.

During the summer of 2008, south Florida was hit with Tropical Storm Fay that resulted in massive rain totals along the Indian River Lagoon counties. Subsequently the salinities in the Lagoon decreased for a prolonged period of time resulting in the loss of

local Styela plicata populations. The same salinity decreases were found along the Gulf

Coast of Florida, as well. For this reason, the tunicate Phallusia nigra was used in the

last experiment (Chapter Six). These specimens were collected during the month of June,

2009 in Bocas del Toro, Panama.

ABSTRACT

Author: Lisa Draughon

Title: The Ascidian Styela plicata as a Potential Bioremediator of

Bacterial and Algal Contamination of Marine Estuarine Waters

Institution: Florida Atlantic University

Dissertation Advisor: Dr. James X. Hartmann

Degree: Doctor of Philosophy

Year: 2010

The tunicate Styela plicata (Ascidiacea) was investigated for its potential use in bioremediation of bacteria and microalgae blooms from estuarine waters. Salinity tolerances, filtration rate, substrate selection, and effects on digested bacteria and

ichthyofaunal communities were examined. If acclimated slowly, Styela can be placed in salinities as low as 24 parts per thousand (ppt) before increased fatalities result. An immediate decrease in salinity from 32 ppt to 20 ppt was not detrimental at short term exposure (four days). One average-sized (~40g) Styela, exposed to 105 and 106 bacteria

or microalgae mL-1, can filter as much as 4.7 L hr-1 and 3.3 L hr-1, respectively.

Individual tunicate filtration rates varied from hour to hour, independent of organism

size, indicating that filtration rates for this species would be better reported on a

population basis rather than on an individual weight or size basis. Bacterial viability in

tunicate feces was assessed by ATP analysis. Bacteria were found to be non-viable after

vii

passage through the tunicate digestive tract. Filtration of eggs or larvae of recreationally

or commercially important fish is of concern. The smallest eggs (~0.6mm) reported in

the Indian River Lagoon belong to Cynoscion nebulosus (spotted seatrout) and Bairdiella

chrysoura (silver perch). Over the course of two hours, 72.5% of 0.5 mm glass beads

(simulated fish eggs) were initially retained, but many were later expelled by the tunicates. For 1.0 mm beads, 49.4% were initially retained and for 2.0 mm beads, 43.5%

were initially retained. Neither the size of the oral siphon nor the size of the overall animal was correlated to bead retention. Substrate selection was investigated using the

tunicate Phallusia nigra. Newly hatched larvae preferred settling on wood (53.7%), as

compared to polyethylene (30.9%), high-density polyethylene (13.0%), and polyvinyl

chloride (2.4%). Overall, results of this study indicate S. plicata is very effective at

removing bacteria and microalgae from estuarine waters. However, further testing is

warranted regarding the unwanted removal of fish eggs and larvae before utilization of S.

plicata for bioremedial purposes.

viii

DEDICATION

I dedicate this work and this book to my wonderful and supportive family who

encouraged me all along the way, especially to my wonderful husband, Randy. I couldn’t

have done this without him. Most times, he worked multiple jobs to help make ends

meet, never once complaining. He ate late dinners, leftovers, and sometimes put up with

“fend-for-yourself” night. He is truly as perfect of a man as I have ever met.

He’s not the only one who had to eat late meals at times. I also dedicate this book

to Sammie, my daughter, whose homework often times stumps me. I hope I was there

for her enough when she needed me. She is such a blessing in our lives.

My Dad passed away on Feb. 2, 2003 before he even saw me complete my undergraduate degree. He would have been very proud to know I achieved this. Still, I think he does know. Daddy, this is also for you.

Many, like my husband, Mom, and daughters Kelly, Kasey, and Sammie prayed a lot for me. Thank you to all of you.

Most important, I thank God, who in a very confusing way brought me to this place. I would have never believed it but He works in strange ways. For those who know me, you know the story. For those of you who don’t, I’d love to share it with you

some day.

TABLE OF CONTENTS

List of Tables ...... xiv

List of Figures...... xv

CHAPTER ONE: General Introduction...... 1

Problems Facing the Indian River Lagoon ...... 1

Styela plicata History...... 5

Species Populations ...... 6

Styela plicata Filtration Advantages...... 7

CHAPTER TWO: Potential Estuarine Water Quality Improvement Via Filtration

Benefits of Tunicate Styela plicata (LeSueur 1823)...... 9

Abstract...... 9

Introduction...... 10

Materials and Methods...... 13

Collection and Acclimation of Organisms...... 13

Experimental Design and Statistical Analysis ...... 14

Algal Filtration Rate ...... 14

Establishment of Bacterial Filtration Rate...... 16

Establishment of bacterial region for flow cytometry analysis...... 16

Bacterial filtration rate...... 18

Preparation of bacterial water samples for flow cytometry...... 19

Calculation of Filtration Rate...... 19

Results and Discussion ...... 20

Filtration Rate - Algae ...... 20

Filtration Rate- Bacteria...... 25

Analysis of Food Types and Treatments ...... 26

Filtration Rates versus Animal Weight...... 28

Reporting Filtration Rates...... 32

Conclusion ...... 34

CHAPTER THREE: Long and Short Term Salinity Tolerance ...... 36

Abstract...... 36

Introduction...... 37

Materials and Methods...... 39

Long-term Salinity Tolerance Adaptation ...... 39

Short-term Salinity Tolerance Shock...... 40

Statistical Analysis...... 41

Results and Discussion ...... 42

Long-term Salinity Tolerance Adaptation ...... 42

Short-term Salinity Tolerance Shock...... 44

Conclusion ...... 46

CHAPTER FOUR: Post-Digestion Bacterial Viability ...... 48

Abstract...... 48

Introduction...... 49

Materials and Methods...... 52 xi

Collection and Acclimation of Specimens...... 52

Experimental Design...... 53

ATP assay and calculations...... 54

Results and Discussion ...... 55

Conclusion ...... 60

CHAPTER FIVE: Potential Effects of Styela plicata on Ichthyofaunal Communities... 61

Abstract...... 61

Introduction...... 62

Materials and Methods...... 65

Collection and Acclimation of Specimens...... 65

Experimental Apparatus and Procedure...... 66

Statistical Analysis...... 67

Results and Discussion ...... 67

Statistical Results...... 67

Bead Retention...... 68

0.5 mm bead retention...... 69

1.0 mm bead retention...... 70

2.0 mm bead retention...... 72

Conclusion ...... 74

CHAPTER SIX: Artificial Substrate Selection by Phallusia nigra ...... 75

Abstract...... 75

Introduction...... 76

Materials and Methods...... 79 xii

Biofilm Preparation of Artificial Substrates ...... 79

Collection and Acclimation of Adult Specimens...... 79

Experimental Design...... 79

Statistical design...... 79

Substrate setup...... 80

In Vitro Fertilization ...... 80

Counting Larval Recruits...... 82

Results and Discussion ...... 82

Conclusion ...... 85

CHAPTER SEVEN: Project Conclusions and Future Directives...... 86

xiii

LIST OF TABLES

CHAPTER TWO

Table 2.1 Calculated filtration rates for individual tunicates exposed to microalgae;

hourly and specimen average…………………….…………………………… 24

Table 2.2 Calculated filtration rates for individual tunicates exposed to bacteria

Escherichia coli; hourly and specimen average ……………………………… 26

CHAPTER FIVE

Table 5.1 Commercially and recreationally important fish species of the Indian

River Lagoon. Egg and larvae sizes with protective measures in place ………. 65

Table 5.2 Size of tunicate specimen and results of 0.5mm bead retention ……………..70

Table 5.3 Size of tunicate specimen and results of 1.0 mm bead retention …...………. 72

Table 5.4 Size of tunicate specimen and results of 2.0 mm bead retention …...………. 73

xiv

LIST OF FIGURES

CHAPTER 2

Figure 2.1 Determination of the gating region to identify a population of Escherichia

coli within the flow cytometer ……………………...…………..……...………..17

Figure 2.2 Fluorescence of E. coli in the FL1 region of the flow cytometer using an

overlay histogram ………….……………………….…………….………………18

Figure 2.3 Demonstration of the histogram region that identifies E. coli via removal

by membrane filtration …………………………………………………………..18

Figure 2.4 Remaining concentrations of microalgae following filtration by

S. plicata ……………………………………………………………………….. 21

Figure 2.5 Mean cell concentrations of microalgae remaining suspended in seawater

without filtrating S. plicata present …………………………………...…………22

Figure 2.6 Mean filtration rates of S. plicata exposed to two different concentrations

of algae and bacteria .……………………...………………………...……..... 27

Figure 2.7 Scatter plot of S. plicata size versus filtration rates …………..………….... 30

Figure 2.8 Regression analysis of whole animal wet weight in grams and dry organ

weight ……………………………………………………………………..……. 31

CHAPTER 3

Figure 3.1 Mortality of Styela plicata held at stable salinity versus decreasing

salinities ……………………………………………………………...…….…... 43

Figure 3.2 Survival function versus days alive for specimens maintained at 32 ppt

and at decreasing salinities ………………….……………………………..…... 44

Figure 3.3 Number of living animals per day in each of four salinities during the

short term shock exposure study ……………………………………………….. 45

CHAPTER FOUR

Figure 4.1 Chemical reaction of an ATP assay..………...…………………………… 52

Figure 4.2 Collection site for specimens used in the ATP assay ...……..……….……..53

Figure 4.3 Hourly bacterial counts in tunicate control beakers …………..…………… 57

Figure 4.4 Averages of bacterial cell density over time in controls and experimental

beakers ……..……………………………………………………………………59

Figure 4.5 Log plot of hourly bacterial counts per mL over the course of the four hour

ATP experiment ………………………………………………...……………… 59

CHAPTER FIVE

Figure 5.1 Experimental apparatus for deliverance of 0.5, 1.0, and 2.0 mm glass

beads ………………………………………………………………………….... 66

Figure 5.2 Close up view of the oral tentacles lining the incurrent siphon of tunicate

Styela plicata………………………………………………………….………... 68

CHAPTER SIX

Figure 6.1 Fertilized eggs of Phallusia nigra after in vitro fertilization ……………… 80

Figure 6.2 Ten hour old Phallusia nigra embryo, just prior to hatching …………….... 81

Figure 6.3 Four day old Phallusia nigra larval recruits on marine plywood under

microscopy ……………………….……………………………………...………82

xvi

Figure 6.4 Percentages of Phallusia nigra larval settlement on different substrate

types ……………………………………..………………………………………84

xvii

CHAPTER ONE: GENERAL INTRODUCTION

Development and anthropogenic disturbance have negatively impacted estuarine benthic and ichthyofaunal communities with the destruction of mangrove, salt marsh, and seagrass habitats that are considered nursery grounds for many organisms (Beck et al.,

2001; Halpern et al., 2007; Heck et al., 2003; Lipcius et al., 2005; Lotze et al., 2006).

The loss of these habitats results in the subsequent loss of many filter-feeding benthic

invertebrates (Bingham, 1992; Sime, 2005). These filter-feeders assisted in providing water clarity by removing particulates from the water column (Jørgensen, 1966b;

Petersen & Svane, 2001). The reduction of the filtration services has directly contributed

to declining water quality (Dame et al., 2002) and associated increased harmful algal

blooms, fish kills, beach closures, and oxygen depletion (Peterson et al., 2006; Worm et

al., 2006).

Problems Facing the Indian River Lagoon

Of immediate concern to Florida is the Indian River Lagoon (IRL), which spans

five eastern counties from Martin northward to Volusia. The 251 kilometer-long lagoon

is a bar-built estuary comprised of three bodies of water: the Mosquito Lagoon to the

north and the Banana River and Indian River to the south, with a watershed comprising

3,575 square kilometers (USEPA, 2006). It is one of the most biologically diverse

estuaries in the world, supporting over 3,000 species of plants and animals (Smithsonian

Marine Station at Ft. Pierce, 2008) including 21 federally endangered or threatened

species (USEPA, 2006). The IRL embodies mangrove dominated marshes and swamps,

wetland and non-wetland forests, scrub land, salt marshes, shellfish growing areas, oyster

bars, sand bars, sand/mud and salt flats, submerged aquatic vegetation, drift algae, deep-

water areas, and spoil islands (USEPA, 2006).

The IRL is artificially connected to Lake Okeechobee by the St. Lucie River for

the purpose of water supply and flood control. High nutrient-containing fresh water exits

Lake Okeechobee through a series of canals and the St. Lucie River and empties into the

southern end of the IRL, resulting in extreme salinity fluctuations, eutrophication, and

sedimentation from excess total suspended solids (TSS). The St. John’s River Water

Management District (2004) reported that turbidity in the IRL has increased with

amplified suspended particulates in the water. Agricultural run-off of fertilizers has

caused elevated nitrogen and phosphorus concentrations enabling algal blooms and an increase in chlorophyll a. The combination of increased turbidity and phytoplankton has prevented sunlight penetration to sufficient depth. This has resulted in the loss of seagrass beds (Morris & Virnstein, 2004).

It is estimated that six billion people will inhabit the coastal regions of the world by the year 2025, almost equal to the total world population today (Hameedi, 1997;

Weber, 1994). The increased discharge of pollutants and industrial wastes as the coastal

population continues to expand will exacerbate the problems of reduced water clarity,

sediment loading, and eutrophication with the eventual loss of nursery grounds.

In addition to increasing turbidity is the problem of decreasing water circulation.

A lock at Port Canaveral, along with the construction of bridges, has reduced water flow

in the IRL system, which already had minimal tidal exchange (Geiger et al., 2004; Indian

River Lagoon National Estuary Program, 1996). Poor water circulation and increasing

water temperature results in elevated algal and bacterial counts; a threat to human health

(Florida Department of Health, 2000; Steidinger et al., 1999). Recent cyanobacterial

blooms, which are often toxic (Steidinger et al., 1999), cause damage to the nervous

system and liver of animals that ingest the cyanobacteria, potentially resulting in death

and massive fish kills (Hauserman, 2007). Bacterial outbreaks result in Florida beach closures, which affect both industry and tourism (Florida Department of Health, 2000).

Bacteria commonly found during these outbreaks are fecal coliforms, indicative of human and animal waste. Low levels of fecal coliforms are commonly found in natural waterways while high levels often indicate sewage contamination (Madigan et al., 2000).

Although the IRL is currently in a state of crisis, it is only one of many Florida aquatic sites of concern (Florida Department of Health, 2000). In 1998 Florida began monitoring water quality along its coastal beaches. Currently 34 counties are monitoring fecal coliforms such as Enterococcus sp. on a weekly basis (Florida Department of

Health, 2000). The State Coordinator of the Healthy Beaches Program in the Florida

Department of Health explained that elevated bacterial levels resulted in the conducting county to either: 1) run a confirmatory re-test followed by a public advisory or warning or 2) immediately issue a public advisory if the county could not re-test (Polk, pers. comm., 2006). An effective and ecologically sound method to remove these harmful bacteria has not yet been developed. Therefore, the only current solution is to allow the bacterial level to subside on its own. Beaches have been closed several days while waiting for this natural process to occur (Florida Department of Health, 2000).

Means of controlling contamination problems vary from mechanical, such as

cordoning and partitioning areas by barriers, to remediation by invertebrates. As cordoning does not result in over-all water quality improvements, the search for the ideal organism(s) for bioremediation was the target of the present study.

Numerous studies have been conducted in recent years utilizing bivalves as indicators of metal contamination and other pollutants (Arnold et al., 2006; Damiens et al., 2006; Hull et al., 2006; Inoue et al., 2006) as well as to improve water quality

(Kohata et al., 2003; Newell et al., 2005). These studies indicate the effectiveness of bivalves in reducing large quantities of organic and inorganic particulate matter (Officer et al., 1982). However, there is concern that large volumes of organic and inorganic nutrients may be excreted back into the water column in the form of pseudofeces, which stimulates primary production by phytoplankton (Kohata et al., 2003; Orlova et al.,

2004), and may further cause decreased water clarity. Asmus and Asmus (1991) discovered that mussels reduced phytoplankton by 37 ± 20% between upstream and downstream areas of a mussel bed; however following a potential N:C ratio of 16:106 estimated by the study, and assuming the ammonium released by the mussels would entirely be taken up by the phytoplankton, the excreted nutrients released by the bed had a potential of stimulating even higher primary production than the uptake. Additionally, bivalves are capable of particle selection based on size (Defossez & Hawkins, 1997) or composition (Kennedy et al., 1996; Ward et al., 1997; 1998; Ward & Shumway, 2004).

With particles >4 µm, bivalves have a particle retention efficiency of approximately

100%. However, the retention efficiency drops to 20% with 0.6 µm particles (Stuart &

Klumpp, 1984), making them ineffective against bacterial blooms.

Another group of organisms that warrant investigation to remove particles from

the marine environment are the tunicates (Chordata: Ascidiacea), also known as ascidians or sea squirts. Tunicates are filter-feeding animals capable of non-discriminate removal of inert, inorganic and organic particles (Jørgensen, 1966b) including bacteria (Flood &

Fiala-Médioni, 1981) from marine waters. Some benthic solitary and colonial tunicates are capable of retaining particles as small as 0.6 µm with approximately 100% efficiency

(Stuart & Klumpp, 1984). Additionally, they do not create pseudofeces (Fiala-Médioni,

1974), which make tunicates a more desirable organism for the bioremediation of estuarine waters. The locally prevalent tunicate species, Styela plicata (Lesueur) is the target of this research.

Styela plicata History

A solitary tunicate, S. plicata was first described when found on the hull of a ship entering the Delaware River in Philadelphia, Pennsylvania, USA in 1823 (Van Name,

1912). Specimens were found during a dredge off Cedar Key, in Levy County, Florida as early as 1877 (US National Museum of Natural History, 2002). In 1884, two separate excursions reported finding S. plicata in the Caribbean (U.S. Virgin Islands) and near

Marco Island, Florida (US National Museum of Natural History, 2002). In 1887 and

1902 the species was noted at St. Martins Reef and Key West, respectively, in the Florida

Keys (US National Museum of Natural History, 2002). A voyage by the R/V Pelican reported finding S. plicata in 1940 off the coast of Jacksonville, Florida, and it was again noted in 1950 in Key Biscayne (US National Museum of Natural History, 2002).

Additional reports of species coverage have been noted throughout the sub-tropical world

(Carballo & Naranjo, 2002; Cima et al., 1996; da Rocha & Kremer, 2005; Lambert &

Lambert, 1998; Thiyagarajan & Qian, 2003; Tujula et al., 2001).

Although reported as an introduced species to Florida through ship fouling or

ballast water (Mook, 1983), populations are now well established in the IRL complex.

David Mook (1983) was the first to report finding S. plicata in the IRL in a review encompassing surveys from 1976 through 1981 . At that time, he reported their presence in the St. Lucie Inlet, the Ft. Pierce Inlet, Link Port, Vero Beach, and the Sebastian Inlet.

During this investigation, specimens were also collected from the Mosquito Lagoon,

Little Jim Island, Bear Point, the waters behind the Smithsonian Marine Station in Ft.

Pierce, and in the Lake Worth Lagoon at MacArthur Park. It is now reported by the

United States Geological Survey as having a native range from North Carolina to Florida

and the West Indies (Fuller, 2006).

Species Populations

As with all benthic invertebrates, populations are limited by available space for

larval settlement (Dayton, 1971; Mook, 1983; Stachowicz et al., 2002; Svane & Petersen,

2001) and by the presence of proper environmental cues (Svane & Petersen, 2001). The

equilibrium dynamics of heterogenic assemblages are controlled by interspecific

competition that is constantly changing both spatially and temporally (Connell & Glasby,

1999; Roughgarden et al., 1988; Svane & Petersen, 2001). Annual patterns and densities

are questionable and variable due to fluctuations in environmental conditions (Svane &

Petersen, 2001), especially when large salinity and temperature fluctuations occur in the

IRL (Mook, 1983), as well as hydrodynamics (Young & Braithwaite, 1980b). This has

been witnessed by the author following the long term drop in salinity from Tropical

Storm Fay in 2008. Waters along the IRL complex remained at a salinity of 10 parts per thousand (ppt) for extended periods resulting in drastic decreases of S. plicata populations as well as other benthic assemblages (personal observation).

Styela plicata Filtration Advantages

Styela plicata has a remarkable ability to filter greater volumes of water than some of the other ascidian species due to its complex branchial structure (Fiala-Médioni,

1978b). Its feeding mechanism explains the high efficiency of particle capture.

MacGintie (1939) was the first to describe this mechanism in detail. A current of water is drawn in through the incurrent siphon into the branchial basket, facilitated by beating cilia that line the stigmata. While feeding, mucous glands situated along the endostyle secrete a sheet of mucus that is propelled around the surface of the branchial basket by a second type of cilia. As water enters the cavity, particulates are trapped in this mucous layer. The mucus continually moves toward the dorsal groove on the opposite side of the branchial basket from the endostyle where it is formed into an unbroken thread and fed into the esophagus at the base of the basket. This thread remains intact upon entering the stomach where it is folded back and forth. As the thread nears the pyloric valve it becomes semi-liquid and enters the intestine. The particle depleted water in the branchial basket exits through gill slits to the excurrent siphon where it is expelled (Meglitsch,

1967). When the animal is not feeding, the mucous lining is absent and particles in the water can be seen entering the incurrent siphon and immediately exiting the excurrent siphon (MacGinite, 1939). Extensive studies of the mucous film of S. plicata conducted by Flood and Fiala-Médioni (1981) determined that it could filter and retain particles

larger than 0.5 µm. The ability of S. plicata to remove these bacterial sized particles, its abundant population world-wide, and its physical competence are the basis for selection of this species in the present study use in a bioremediation.

CHAPTER TWO: POTENTIAL ESTUARINE WATER QUALITY

IMPROVEMENT VIA FILTRATION BENEFITS OF TUNICATE STYELA

PLICATA (LESUEUR 1823)

Abstract

The subtropical rough tunicate Styela plicata was examined for its prospective reduction

of microalgal and bacterial particle concentration in estuarine water. Filtration rates were

determined from individual tunicates exposed to the microalgae Nannochloropsis sp.

(n=14) and the bacteria Escherichia coli (n=12) in concentrations of 105 and 106 cells

mL-1 over four hours. Results revealed that one average-sized (~ 40 g) S. plicata,

exposed to 105 algal cells mL-1 had an average filtration rate of 3.1 L hr-1and those

exposed to 106 algal cells mL-1 had an average filtration rate of 3.3 L hr-1, with 100% particle retention. Average filtration rates for tunicates exposed to bacteria at 105 cells mL-1 were 4.7 L hr-1 and were 2.3 L hr-1 in the 106 treatment, with 100% particle

retention. However, individual tunicate filtration rates differed by 3.5 L hr-1 for

microalgae and 2.3 L hr-1 for bacteria. No relationship was found between hourly

filtration rate and whole animal weight (r2= 0.0001) or dry organ weight (r2= 0.0067),

indicating that filtration rates for this species should not be reported on a live weight

basis. Results suggest that filtration should be quantified by averaging the filtration rate

of a population of animals over time, especially when establishing models for

bioremedial use. It is estimated that 200 rough tunicates could fully remove 105 bacteria

mL-1 from 22,600 L each day; size and concentration of suspended particulates, water

flow, and temperature would affect the rate of bacterial removal. Controlled densities of

filter-feeders, such as Styela plicata, strategically placed in contaminated areas could

substantially reduce unwanted bacteria and microalgae, thus improving water quality.

INTRODUCTION

Healthy marine ecosystems are dependant on maintenance of the delicate balance

between the intricate interactions of organisms from microscopic marine diatoms to

macroscopic marine mammals (Costanza & Mageau, 1999). Catastrophic changes to the

entire system can occur when this balance is interrupted (Marques et al., 1997). The past

century of development and anthropogenic disturbances have negatively impacted

estuarine benthic and ichthyofaunal communities with the destruction of mangrove, salt

marsh, and seagrass habitats, that are considered nursery grounds for many organisms

(Beck et al., 2001; Heck et al., 2003; Lipcius et al., 2005; Lotze et al., 2006). The major

causal factor of a worldwide decline in seagrass populations is reduced water clarity and

eutrophication (Short & Wyllie-Echeverria, 1996), which can result in elevated microbial and microalgal growth (Aslan-Yilmaz et al., 2004). Mangrove and salt marsh habitats are additionally stressed through anthropogenic disturbances and shoreline modifications as coastal populations increase (Kennish, 2002). It is estimated that six billion people will inhabit the coastal regions of the world by the year 2025, almost equal to the total world population today (Hameedi, 1997; Weber, 1994). The increased discharge of pollutants and industrial wastes as the coastal population expands will exacerbate the problems of reduced water clarity, sediment loading, and eutrophication with the eventual loss of aquatic animal nursery grounds.

Loss of habitat correlates with a loss of filter-feeding benthic invertebrates that inhabit those grounds (Bingham, 1992). Filter-feeders assist in providing water clarity by removing particulates from the water column (Bone et al., 2003; Jørgensen, 1966b;

Petersen & Svane, 2001). The loss or reduction of filtration benefits provided by suspension feeders has potentially contributed to declining water quality (Dame et al.,

2002) with a direct correlation in the increase of harmful algal blooms, bacterial outbreaks, fish kills, beach closures, and oxygen depletion (Worm et al., 2006). A recent study of Florida Bay revealed a direct correlation between phytoplankton and cyanobacterial blooms with the large scale decline in dominant sponge populations; the microbial blooms occurred without any increase in nutrient loads (Peterson et al., 2006).

Alterations in water current due to the construction of seawalls, bridges, locks, and marinas result in pockets with poor water circulation, often stagnant, producing elevations of potentially pathogenic bacteria, such as fecal coliforms. Low levels of fecal coliforms are commonly found in natural waterways while high levels often indicate sewage contamination (Madigan et al., 2000) that results in human health risks. Methods to control these problems vary from mechanical, such as cordoning and partitioning areas by barriers, to remediation by invertebrates. As cordoning does not result in over-all water quality improvements, the search for the ideal organism(s) for bioremediation was the target of this investigation.

Historically, bioremediation by filter-feeding benthic invertebrates has been successful (Dame et al., 1984; Officer et al., 1982; Ulanowicz & Tuttle, 1992). Therefore considerable effort has been devoted to studying their use for water quality improvements

(Akhtar et al., 2003; Arnold et al., 2006; Damiens et al., 2006; Giangrande et al., 2005;

Hull et al., 2006; Inoue et al., 2006; Kohata et al., 2003; Milanese et al., 2003; Takeda &

Kurihara, 1994). Studies by Officer et al. (1982) indicate effectiveness of bivalves in

reducing large quantities of organic and inorganic particulate matter. However, there is

concern that even larger volumes of organic and inorganic nutrients may be excreted

back into the water column in the form of pseudofeces. Bivalve excrement stimulates

primary production of phytoplankton (Kohata et al., 2003; Orlova et al., 2004), a further

cause of decreased water clarity. Asmus and Asmus (1991) discovered that mussels

reduced phytoplankton biomass by 37 ± 20% between upstream and downstream areas of

a mussel bed. However the excreted nutrients released by the bed could potentially stimulate even higher primary production than the uptake. Additionally, bivalves are capable of particle selection based on size (Defossez & Hawkins, 1997) or composition

(Kennedy et al., 1996; Ward et al., 1997; Ward et al., 1998; Ward & Shumway, 2004).

With particles >4 µm, bivalves have a particle retention efficiency approaching 100%; however, the retention efficiency drops to 20% with particles less <0.6 µm (Stuart &

Klumpp, 1984) making them ineffective against bacterial blooms.

Milanese et al. (2003) examined the sponge Chondrilla nucula as a potential bioremediator. Their extrapolated results estimated that a one meter square patch of this sponge could retain up to 7.0 x 1010 Escherichia coli cells hr-1.

Another group of organisms that warrants investigation for its members’ abilities to remove particles from the marine environment is the tunicates (Chordata: Ascidiacea),

also known as ascidians or sea squirts. Tunicates are filter-feeding animals capable of

non-discriminate removal of inert, inorganic and organic particles (Jørgensen, 1966b)

including bacteria (Flood & Fiala-Médioni, 1981) from marine waters. Some benthic

solitary and colonial tunicates are capable of retaining particles as small as 0.6 µm with

approximately 100% particle efficiency (Stuart & Klumpp, 1984). Additionally, they do

not create pseudofeces (Fiala-Médioni, 1974), which make tunicates a more desirable

organism for the bioremediation of estuarine waters.

The rough tunicate Styela plicata is suggested for bioremedial use due to its

ability to remove micro-particles, such as bacteria and algae, from the marine

environment. S. plicata has a global distribution in most tropical and sub-tropical marine

environments (Carballo & Naranjo, 2002; Cima et al., 1996; Lambert & Lambert, 1998;

Thiyagarajan & Qian, 2003; Tujula et al., 2001) and was reported by Fiala-Médioni

(1978a) to have a greater capacity for water filtration than other ascidians due to a more

complex branchial basket. Global availability and high filtration rates make S. plicata a

target species for bioremedial investigation of micro-particle contamination of estuarine

waters.

MATERIALS AND METHODS

Collection and Acclimation of Organisms

Individual S. plicata specimens ranging in weight from 16.8 - 55.7g were

collected from the Indian River Lagoon near Ft. Pierce, Florida, USA, in June and July,

2005. Each organism was carefully removed and placed in an aerated 19 L bucket of

seawater (28 ppt) and transported to the lab. Tunicates were manually cleaned of

commensal organisms and placed unsecured in fiberglass tanks (3.1 m x 0.66 m x 0.3 m) with flow-through treated seawater at 28 ppt for 48 hours to remove any remaining debris prior to experimentation. Treated seawater was ozonated, filtered (5 µm) and UV-treated

to reduce and remove any micro-organisms present. During this time, tunicates were fed

the microalgae Isochrysis aff. galbana (clone T-Iso) at a concentration of 1,000 cells

mL-1 to keep gut contents to a minimum. Only healthy individuals exhibiting open

siphons and consistent brown color were used for experiments.

Experimental Design and Statistical Analysis

All statistical analyses were conducted using SAS 9.1 software. Regression

analyses were conducted on 1) dry organ weight (DOW) versus whole animal wet weight

(WAWW) , 2) mean hourly filtration rate of each individual tunicate vs. whole organism

wet weight in both the algae and bacteria experiments, and 3) dry organ weight vs. mean

hourly filtration rate in the bacteria experiment. The a priori H0 was that “weight does

not have an effect on the volume of water cleared hr-1 by a tunicate”. The response

variable “Filtration Rate” was analyzed by repeated-measures ANOVA (critical

alpha=0.05) for comparison of the filtration rates of two food types (microalgae and

bacteria) and concentrations (105 and 106 cells mL-1) over time. Triplicate counts of

hourly densities in the water samples were averaged for comparison. Differences were

considered significant for probabilities (p< 0.05). The a priori H0 was that “no

significant difference exists between individual hourly or individual average filtration

rates for the two food types and concentrations over time”.

Algal Filtration Rate

Nannochloropsis sp. 1 cultures (clone CCMP531, Bigelow Laboratory for Ocean

Sciences, Maine, USA) were filtered twice through a 20 µm screen to remove any large particle clumps and then placed in a water bath at 45° C for 30 minutes to heat-kill them

(i.e., stop proliferation). Immediately after heat killing, average cell counts from

triplicate 10 mL samples were determined using a hemacytometer and phase contrast

microscopy. Nannochloropsis sp. glows a consistent turquoise green color,

differentiating it from other particles present in the sample.

Fourteen tunicates were placed individually in 8 L beakers containing 6 L of

treated seawater. Beakers were randomly placed in an ambient flow-through water bath

to provide water temperature stabilization. Specimens were secured by a monofilament

loop around the body and tied to 12 cm X 12 cm plastic grids and allowed to acclimate

until each resumed open siphon conditions. Heat-killed Nannochloropsis sp. was added

to the beakers to obtain either 105 or 106 cells mL-1 (n=7 for each concentration). Six

additional control beakers were established in the same fashion with the absence of a tunicate (n=3 for each algae concentration) to monitor any microalgae loss due to adherence to the plastic beaker or sedimentation.

Water temperature, salinity, and pH remained constant for the experimental period at 26° C, 28 ppt, and 8.5, respectively. Aeration was supplied through a pipette secured to the side of each replicate beaker to prevent direct exposure of the tunicate to air bubbles. As microalgae were added, beakers were tapped lightly on the sides causing the tunicates to close their siphons so that a 10 mL water sample could be drawn at ‘Hour

0’ (Hr0) prior to any filtration. Samples were drawn from the center of the beaker after stirring and subsequent samples were taken at 1 hour intervals in this same fashion for a total of seven samples (six hours) per beaker. Samples were fixed with 1 mL of 10% buffered formalin after completion of each hourly sampling and refrigerated at 4°C until processing. Nannochloropsis density in all samples was determined within 96 hours in

all replicates by taking the mean of three individual hemacytometer counts for each

sample.

Establishment of Bacterial Filtration Rate

Establishment of bacterial region for flow cytometry analysis.

Flow cytometry enables detection of a ‘population of interest’ by the demarcation

of regions (i.e. gating) on a double parameter histogram. E. coli was used in the bacterial

filtration experiment as it is one of the problematic fecal coliforms present in the IRL.

Additionally, it is similar in size (0.72 µm x 3.1 µm) (Pierucci, 1978) to the microalgae

Nannochloropsis (2-4 µm) (Rocha et al., 2003). In three trials a pure culture of E. coli

grown in Luria broth was used to determine the gating region. Aliquots of this culture

were collected, centrifuged at 20,000xg for 15 minutes, and the pellets resuspended in

0.85% NaCl. Dilutions of these samples were prepared as 1X, 1/100X, and 1/1000X.

Samples (1 mL) of each dilution were analyzed with a Becton Dickinson FACSCalibur

flow cytometer. A standard 10,000 events was collected for each sample. A control

sample of buffer alone was also analyzed. A double parameter histogram was established

using Forward Scatter (FS) log vs. Side Scatter (SS) log channels. Fig. 2.1 A–D reveals a

decrease in total events with increasing dilutions, indicating a properly established

region. Further validation of the established region was achieved by staining E coli

samples with SybrII green dye (Molecular Probes, Eugene OR). E. coli samples of 106 cells mL-1 were centrifuged at 20,000xg for 15 minutes, resuspended and washed twice in

a solution of 0.85% NaCl followed by re-suspension in 30 mM potassium citrate and

SybrII green dye at a dilution of 10-4 (Lebaron et al., 1998). Aliquots (1 mL) of both

stained and unstained E. coli samples were analyzed using the same FS log vs. SS log region previously established and a FL1 histogram. Acquisition was stopped at 20 seconds in each case. A region was created on FL1 excluding the autofluorescence of the control unstained sample (Fig. 2.2). The stained samples showed 91.35 % positive cells.

These results were further confirmed by filtering the stained E. coli samples through a 0.2

µm filter, thus eliminating the bacteria (Fig. 2.3). As shown in Fig. 2.3, the gated events decreased from 91.35% to 2.5% after filtration. These three experiments indicate that the flow cytometry method developed for determining the E. coli population was valid.

4800 events 2133 events

A B

1378 events

C D

Figure 2.1 A-D Individual histograms of known pure samples of E. coli prepared in dilutions to determine flow cytometric gating region. A – 1X sample shows 4800 events within gated region. B – 1/100X dilution shows 2133 events in gated region. C – 1/1000X dilution shows 1378 events in gated region. D – Overlay histogram of the three dilutions. The small spike to the right is calibration microbeads (6µm) reflecting the small calibration gate visible in the upper right corner of histograms A – C.

Figure 2.2 Overlay histogram of FL1 region, gated on E. coli. Light peak is unstained control of sample buffer alone. Dark peak in region M1 is E. coli stained with SybrII Green bacterial stain. Histogram statistics indicate 91.35% positive cells in targeted M1 region.

Figure 2.3 Overlay histogram of stained unfiltered E. coli (Region M1, light gray peak) vs. E. coli filtered through a 0.2 µm filter (dark gray peak) that eliminates bacteria. M1 gated peak indicates 91.35% positive cells prior to filtration and decreases to 2.5% after filtration.

Bacterial filtration rate.

Twelve tunicates were placed individually in 8 L beakers with 6 L of treated

seawater as described in the above section, “Algae Filtration Rate”. Three control

beakers with no tunicates were also established. The experimental method of the

bacterial filtration experiment was the same as that described for the algae filtration

experiment with changes only to numbers of replicates. Ethanol-killed E. coli (Oie et al.,

1999) were added to each beaker to provide one of two treatment concentrations, 105 and

106 cells mL-1 (n=6 for each concentration). Control beakers were established containing

105 cells mL-1 of ethanol-killed E. coli (n=3). Hourly water samples (10 mL) were taken in the same fashion as in the algae experiment for a total of seven samples (six hours) per beaker. Water samples were fixed with 1 mL 10% formalin and refrigerated at 4º C until preparation for flow cytometry.

Preparation of bacterial water samples for flow cytometry.

Water samples were vortexed to assure even mixing. One mL was removed from each sample and centrifuged at 20,000xg for 15 minutes. The supernatant was decanted and the pellet resuspended in 1 mL 0.85% NaCl. This procedure was repeated, but the pellet was resuspended in 1 mL 30mM potassium citrate containing 0.1 µl SyberII Green dye (10-4 dilution). Samples were analyzed with a Becton Dickinson FACSCalibur flow cytometer gated on the previously established region. Acquisition parameters were set to run a standard 20 seconds in each case. Triplicate acquisitions were obtained per water sample. Only the third acquisition was used in each case to assure uniformity and untainted results from cross contamination of sample tubes.

Calculation of Filtration Rate

Filtration rate is defined as the specific volume of water that has been completely cleared of suspended particles over a defined amount of time. With the algae filtration experiment, these particles were counted as cells mL-1 while the bacteria filtration

experiment measured flow cytometer events mL-1. Filtration rates were calculated

following Quayles’s formula, modified by Willemsen for the adjustment of particle

sedimentation as suggested by Coughlan (1969):

E E C2 C2 F = (V0 + Vt) (ln (S 0/S t) - ln (S 0/S t)) / (2nt)

-1 where F = filtration rate in mL hr ; V0 = volume of water remaining in experimental

beaker after withdrawal of sample at time 0; Vt = volume of water in experimental beaker

before withdrawal of sample at time t; n = number of animals; t = duration of experiment;

E E -1 S 0 , S t = suspension loads in cells or events mL in experimental beaker at times 0 and t;

C2 C2 -1 S 0 , S t = suspension loads in cells or events mL in control beaker at times 0 and t.

RESULTS AND DISCUSSION

Filtration Rate - Algae

Initial cell counts taken at Hour 0 (Hr0) were lower than target concentrations of

105 and 106 cells mL-1. Actual Hr0 densities for microalgae were 8.3 x 104 cells mL-1 ±

2.3 x 104 cells mL-1 s.d. for the 105 cells mL-1 treatment and 6.8 x 105 cells mL-1 ± 6.8 x

104 cells mL-1 s.d. for the 106 cells mL-1 treatment Microalgal cell concentrations fell

below the threshold of reliable hemacytometer counting of 100 cells mL-1 after only three

hours in the 105 cells mL-1 treatment and four hours in the 106 cells mL-1 treatment. For

this reason, statistical analyses of filtration rates are only from (Hr0) through Hr3 for both

concentrations. Change in mean cell count mL-1 hr-1 for each treatment was analyzed by

regression analysis over time: cell count mL-1 = 79881 - 23571(hour), (r2=0.7109,

p<0.0001, n=27) for the 105 cell mL-1 treatment and

cell count mL-1 = 600265 -143162(hour), (r2=0.8289, p<0.0001, n=35) for the 106 cell

mL-1 treatment (Fig. 2.4).

120,000 ) -1

100,000 A

80,000 sp. (Cellsml 60,000

40,000

20,000

Nanochloropsis Nanochloropsis 0 Time 0 1 Hour 2 Hours 3 Hours Sample Periods

)

-1 800,000 700,000 B 600,000

500,000 sp. (Cells ml

400,000

300,000

200,000

100,000 Nanochloropsis Nanochloropsis 0 Time 0 1 Hour 2 Hours 3 Hours 4 Hours Sample Periods Figure 2.4 Mean (± s.d., n=7) cell concentrations over time for tunicates Styela plicata , exposed to two different concentrations of the microalgae Nannochloropis sp. A) 105 algal cells mL-1 treatment. Linear decrease (y = 79881 - 23571(hour), (r2=0.7109)) in the average hourly suspension load of algal 6 -1 2 cells, (n=27). B) 10 cells mL treatment. Linear decrease (y = 600265 -143162(hour), (r =0.8289)) in the average hourly concentration of cells remaining, (n=35).

Calculated hourly filtration rates for individual tunicates exposed to microalgae

ranged from -2,644 mL hr-1 to 4,295 mL hr-1 (mean=1,663 mL hr-1, n=7) in the 105 cell mL-1 treatment and -570 mL hr-1 to 5,750 mL hr-1 (mean=2,836 mL hr-1, n=7) in the 106 cell mL-1 treatment. Negative filtration rates would indicate the addition of particulates

instead of depletion. Such addition would have to come from either feces or gametes. As

each tunicate began the experiment with a cleared digestive tract, gametes would be the

only other source. This situation was not observed. However, close examination of the

algal control beakers revealed erratic hourly densities instead of the expected steady

decline due to settlement or affinity to the plastic beaker (Fig. 2.5).

900,000 800,000 700,000 600,000 500,000 400,000

Cell Counts 300,000 200,000 100,000 0 Time 0 1 Hour 2 Hour 3 Hour 4 Hour A Time Control 1 Control 2 Control 3 Average

140,000

120,000

100,000

80,000

60,000

Cell Counts Cell 40,000

20,000

0 Time 0 1 Hour 2 Hour 3 Hour 4 Hour B Time Control 1 Control 2 Control 3 Average Figure 2.5 Hourly cell counts of microalgae control beakers containing no tunicates (n=3 for each control) showing hourly fluctuation. Darker lines indicate moving average. A. 105 cell mL-1 treatment controls. B. 106 cell mL-1 treatment controls.

A possible explanation for the non-homogeneity of Nannochloropsis sp. in the control beakers is the absence of the water circulation caused by the tunicates themselves.

A strong current of water enters the incurrent siphon at the top of the animal and exits the sides adding to the mixing effect. The purpose of the controls was to account for loss of particulates from sources other than filtration. As there was no obvious loss from sedimentation or affinity, the data were re-analyzed with the control beakers omitted from the calculations. Filtration rates calculated without the controls ranged from 0.0

mL-1 (when the tunicate had a closed siphon or did not have the mucous net extended

across the pharyngeal basket) to 4,812 mL hr-1 (mean=3,065 mL hr-1 ± 1,264 s.d., n=7) in

the 105 cell mL-1 treatment and 771 mL hr-1 to 5,950 mL hr-1 (mean=3,252 mL hr-1 ±

1,039 s.d., n=7) in the 106 cell mL-1 treatment (Table 2.1).

The average filtration rate per organism exposed to 105 algal cells mL-1 with

controls factored in was 1,663 mL hr-1 whereas the106 treatment specimens filtered 2,744 mL hr-1. Filtration rates without controls were more similar and higher; 3,065 and 3,252

mL hr-1 for the 105 and 106 treatments, respectively. It is believed that filtration rates

calculated without controls are more reliable in these experiments.

TABLE 2.1 – Calculated filtration rates for individual tunicates exposed to the microalgae Nannochloropsis sp.; per hour and specimen average. # = specimen number; WAWW = Whole Animal Wet Weight in grams; HFR = Hourly Filtration Rates in mL hr-1 AFR = Average Filtration Rate in mL hr-1 for individual specimens over the course of three or four hours.

105 Cell mL-1 Treatment 106 Cell mL-1 Treatment WAWW WAWW # # (g) HFR AFR (g) HFR AFR

1 20 0 1729 8 30 2224 3324 3114 3705 2074 4048 3321

2 21 1336 2329 9 32 3582 2471 1066 2541 4586 1765 1994

3 48 2578 3770 10 17 4994 3862 4145 3427 4586 3523 3504

4 23 4148 4246 11 26 771 1744 4812 1997 3778 1903 2303

5 24 4719 4460 12 34 4405 4802 3885 3927 4776 5950 4925

6 19 4561 3684 13 24 3421 3914 3285 3737 3205 3756 4743

7 39 1091 1236 14 21 3075 2645 2618 2317 0 2558 2630

Filtration Rate- Bacteria

Individual hourly filtration rates for tunicates exposed to the bacteria E. coli

ranged from 1,990 mL hr-1 to 7,505 mL hr-1 for the 105 cell mL-1 treatment and from

-1,211 mL hr-1 to 4,970 mL hr-1 in the 106 treatment (Table 2.2). The negative filtration

rate of -1211 mL hr-1 involved one specimen out of 12 and was recorded during the first

hour. No gametes were found in the water sample although it is possible that there was

some type of contamination on the animal’s test when first placed in the beaker. As this

was an isolated incident, the data were kept and utilized in calculations. The calculated

average filtration rates for individual tunicates exposed to bacteria ranged from 3,845 to

6,068 mL hr-1 per animal for the 105 treatment and from 750 to 4,458 mL hr-1 per animal

for the 106 cells mL-1 treatment (Table 2.2). Regression analysis for the 105 cell mL-1

treatment revealed a significant negative relationship (r2=0.8854) between time and flow

cytometry events with a model of Events = 4,018.3 -1,325(hour). Linear regression analysis for the 106 cell mL-1 treatment similarly resulted in a significant negative

relationship (r2=0.6758) with a model of Events = 26,418 -5,998.6(hour). Less variability

was found in the bacteria control beakers from one hour to the next than was found in the

microalgae controls. Therefore, very little difference was found in the filtration rate per

organism calculated with or without the controls. Tunicates exposed to the 105 cells mL-1

treatment filtered 4,654 mL hr-1 when controls were factored in and 4,353 mL hr-1 without controls. The 106 treatment resulted in filtration rates of 2,225 mL hr-1 with

controls and 2,185 mL hr-1 without.

TABLE 2.2 – Calculated filtration rates for individual tunicates exposed to bacteria E. coli; per hour and specimen average. # = specimen number; WAWW = Whole Animal Wet Weight in grams; HFR = Hourly filtration rates in mL hr-1 AFR = Average filtration rate in mL hr-1 for individual specimens over the course of three or four hours.

105 Cell mL-1 Treatment 106 Cell mL-1 Treatment WAWW WAWW # (g) HFR AFR # (g) HFR AFR

15 43 4515 5005 21 39 1175 1570 6010 1721 4489 2076 1308

16 42 1990 3845 22 28 2591 2369 4339 2100 5208 2519 2265

17 38 4628 4112 23 36 832 1078 4198 1309 3509 1240 931

18 58 4565 4692 24 39 2462 3550 5549 3803 3962 4374 3560

19 50 4170 4200 25 56 -1211 750 4396 654 4034 1354 2201

20 35 7505 6068 26 48 3281 4458 5609 4802 5091 4970 4778

Analysis of Food Types and Treatments

Filtration rates calculated without the controls for the two different algal densities

(105 and 106 cells mL-1) were very similar: 3,065 and 3,252 mL hr-1 per animal,

respectively (Fig. 2.6). However, a 50% reduction in filtration rate was observed in the bacterial treatment at the higher concentration. The importance of bacteria as a supplemental food source for invertebrates has long been known (Jørgensen, 1966a). The difference found in filtration rates in this study may be attributed to the nutritional needs

of the organisms being met more rapidly with bacteria than with microalgae. It may also

indicate an overload of material on the mucous net.

5000

4500 4653.8

-1 4353.3 4000

animal 3500 -1

3000 3251.7 3065.0

2500 2744.1

2295.7 2000 2185.4

1500 1662.8

1000

500 Avg. Filtration Rate in mL hr mL in Rate Filtration Avg.

0 5 6 5 6 5 6 5 6 10 10 10 10 1 10 10 10 10 With Controls No Controls With Controls No Controls Microalgae Bacteria

Figure 2.6 Mean filtration rates per animal for both microalgal treatments (n=7) and bacterial treatments (n=6) with and without control densities factored in. Average filtration rates increased with increasing microalgal densities but decreased with increasing bacterial densities.

For extrapolation purposes of the potential environmental impact that S. plicata would have in bioremediation, the combined mean filtration rate for both microalgal

treatments calculated with the controls was 2,250 mL hr-1 per animal. Without including

the controls, the combined mean was 3,158 mL hr-1 per animal. The combined mean for

the bacterial treatment with controls was 3,475 mL hr-1 per animal (Fig. 2.6).

Repeated measures analysis of combined filtration rate data met the assumptions

of normality of residuals (p=0.4136) and sphericity (p=0.2836). A Wilkes-Lambda test indicated hourly filtration rates changed significantly with time (F 2,21 =10.69, p=0.0006),

with the largest difference occurring between Hr0 and Hr1 (p=0.0004). Slightly less

difference was found between Hr1 and Hr2 (p=0.0157). No interactions between

time*treatment (F 2,21 = 0.37, p< 0.6921), time*food type (F 2,21 =2.94, p=0.0746) and time*food type*treatment (F 2,21 = 2.71, p=0.0899) were found. The mean values of

filtration rate averaged over all three sampling times were not significant between

individual test subjects for the two concentrations (p=0.1389). However, a significant

difference in filtration rate averaged over all three sampling times was found between

subjects for the two food types (p=0.0122) and food type*treatment (p=0.0015)

indicating conflicting filtration responses to different food types and concentrations.

Tunicates had responded in a similar way to both microalgal density treatments; however,

a 50% reduction in filtration rate was found between the two bacterial density treatments.

Filtration Rates versus Animal Weight

Armsworthy et al. (2001) found that the tunicate Halocynthia pyriformis was able

to regulate the diameter of the incurrent siphon in response to food quantity and quality in

an apparent effort to maintain a constant filtration rate. This is not the case with all

ascidians. The filtration rate of Ciona intestinalis decreases with increasing food supply

when fed a microalgal diet of Rhodomonas sp. (Petersen et al., 1995). These same results

were found by Sisgaard et al. (2003) when they repeated the same experiment. However

they found non-uniform filtration rates when C. intestinalis was fed a low quality diet of

decaying Spartina grass Zostera marina particles (<100 µm).

Non-uniform filtration rates were found in the present study with S. plicata

independent of organism size (Fig. 2.7). As seen in Table 2.2, the highest average

filtration rate (6,068 mL hr-1) came from a specimen weighing 35.3 grams (specimen

#20), and the lowest average filtration rate of 750 mL hr-1 came from a 55.7 gram specimen (#25). Specimen #2 in the microalgae experiment (Table 2.1) had the largest difference in filtration rates from one hour to the next with 1,066 to 4,586 mL hr-1

between Hr2 and Hr3. Specimen #16 in the bacteria experiment exhibited a change from

1,990 to 4,339 mL hr-1 between Hr0 and Hr1 (Table 2.2). No pattern could be detected

in the filtration rates of an individual tunicate, either increasing or decreasing, as

particulates were depleted from hour to hour.

7000.0

6000.0 -1

5000.0

4000.0

3000.0

2000.0

1000.0 Avg. Filtration Rate in mL hr

0.0 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 Whole Animal Wet Weight in Grams

Figure 2.7 Scatter plot of Styela plicata size (whole animal wet weight in grams) vs. filtration rates in mL hr-1. No relationship was found (r2=0.0001, n=26).

Specimens selected for use in this study differed in size up to 3.5-fold. It would

be difficult to find a mature adult specimen smaller than that used for this study

(16grams); however it is possible to find specimens larger than 58 grams. It is doubtful if

a wider variety of sizes would have produced different results than those found here.

Regression analysis revealed a strong positive relationship between WAWW and

DOW (r2 = 0.7361) with a resulting linear regression model of

DOW = 0.0156(WAWW) + 0.02426 (Fig. 2.8). Dry organs represent 1.62% of the

WAWW (± 0.02% s.d., n=15). However, no significant relationship was found between

wet weight and average filtration rate of tunicates exposed to algae or bacteria (r2=

0.0035 and 0.0036 respectively), nor with dry organ weight and average bacterial filtration rate (r2=0.2637).

1.200

1.000

0.800

0.600

DryWeight Organ 0.400

0.200

0.000 0 10203040506070 Whole Animal Wet Weight

Figure 2.8 Regression analysis of whole animal wet weight (WAWW) in grams and dry organ weight (DOW) in grams with a resulting linear regression model of DOW = 0.0156(WAWW) + 0.02426; (r2=0.7361).

Many studies have shown the effects of body size on the physiological aspects of

energetics (Fisher, 1976; Jiang et al., 2008a, 2008b; Petersen et al., 1995) and it was not

the intent of the present study to dispute such fundamental principles. However, it is

important to elucidate the inconsistencies found in average filtration rates of individual S. plicata versus size over the course of this experiment, as shown in Tables 2.1, 2.2 and

Figs. 2.7 and 2.8. Other inconsistencies in metabolic processes have been found in

similar species. Jiang et al. (2008a) found an inconsistency in the oxygen uptake of S.

clava. However, they used two to four animals per replicate, which would mask individual fluctuations in oxygen consumption.

Reporting Filtration Rates

Filtration rates, clearance rates, pumping rates, and flow rates have historically

been based on organism wet weight or dry organ weight (Fiala-Médioni, 1978b;

Jørgensen, 1949; Olivas-Valdez & Cáceres-Martínez, 2002; Petersen & Riisgård, 1992;

Petersen et al., 1995; Randløv & Riisgård, 1979; Thompson et al., 1978). Hecht (1916)

reported the rate at which carmine particles were drawn into the tunicate Ascidia atra in

terms of mL hr-1 g-1 wet weight. This method was identified as problematic by Jørgensen

(1949) when studying the feeding rates (his term for filtration rate) of several marine

organisms, including ascidians. He stated, “Feeding-rates are expressed as per mgm. of

amino-nitrogen of the experimental animals instead of the normally used weight and

length or similar measurement, which provide an unsatisfactory basis for comparisons

between animals”. He did not elaborate further as to why the normally used basis was

unsatisfactory. Despite these objections, weight basis continued to be used by Goldberg et al. (1951) when they studied the uptake of vanadium in tunicates.

Carlisle (1966) reported filtration rates both in terms of organic nitrogen and wet

weight. Walne (1972) studied five species of bivalves, reporting the filtration rate in

terms of both length and mean dry meat weight. Weight-based reporting of filtration

rates, clearance rates, pumping rates, and flow rates became widely accepted by 1978 and

continues to be used by researchers (Armsworthy et al., 2001; Fiala-Médioni, 1978b;

Kowalke, 1999; Petersen & Riisgård, 1992; Petersen et al., 1995; Randløv & Riisgård,

1979; Robbins, 1983) with findings often in disagreement (Jiang et al., 2008b; Randløv &

Riisgård, 1979; Robbins, 1983). Inconsistent filtration rates from hour to hour,

irrespective of size, such as those found in the present study, may explain these

differences.

Other inconsistencies could also explain these differences, including variations in

experimental methods, (i.e., maintaining constant particles concentrations versus

measuring depleting concentrations over time). However, other causes are likely a factor,

(i.e., numbers of replicates, variations in size of organisms, and environmental factors).

In Fiala-Médioni (1978b), only three replicates of the same appoximatly sized S.

plicata (7.0 to 7.5 cm in length by 4.0 to 4.5 cm width) were used to determine filtration

rate. The present study utilized tunicates from 16.8 to 57.8 grams with six to seven

replicates per concentration. Fiala-Médioni’s filtration rate findings for S. plicata were

8.76 mL hr-1 dry weight of organs. This equates to 4.1 L hr-1 per animal, which is 33%

higher than the rate of 3.1 L hr-1 per animal reported here. In Jørgensen’s (1949) study,

two to three C. intestinalis of approximately six to seven cm in length were used to

establish a feeding rate of 552-750 mL hr-1 when fed colloidal graphite particles. The

filtration rate of this species was also studied by Randløv and Riisgård (1979) utilizing

the microalgae Dunaliella marina at a starting concentration of approximately 1.0 x 104 cells mL-1 with 21 animals. They based their results on total dry weight

0.84 (F = 46.4 Wtotal ). Dry organs were found to make up 30.4% of the total dry weight.

However, the total length of the animals was not reported making it difficult to compare

to Jørgensen’s (1949) findings.

There are other factors that affect filtration rates. Petersen and Riisgård (1992)

repeated the Randløv and Riisgård experiment on C. intestinalis, a cold water species,

with manipulation of prior starvation, temperature, and season (over 10 months).

Filtration rates were dependant on temperature with a positive correlation from 4 to 21°C

and a negative correlation above 21°C. Their findings varied from as low as 390 mL hr-1

to 1,880 mL hr-1 (n = 16). Kim and Moon (1998) found similar results when studying the

effects of water temperature on S. clava, S. plicata,and the blue mussel Mytilus edulis.

The critical temperature in all three cases was 29° C. Tunicates used in the present study

were maintained at 26° C, well within their acceptable range. Other environmental

factors may also be significant to filtration rates, such as hydrology (water chemistry and

current speeds), and particulate density and would require further study.

CONCLUSION

Previous ascidian filtration rate studies have been based on volume filtered hr-1 g-1 dry organ weight (Armsworthy et al., 2001; Fiala-Médioni, 1978b; Randløv & Riisgård,

1979). However, no relation between organism weight and filtration rate was evident in the present study, suggesting filtration rates of S. plicata should not be reported in weight-based units. Instead, the filtration rate average of a population of specimens may be a more effective method of reporting.

Most of the specimens tested in the present study exhibited inconsistent filtration from hour to hour. Filtration rates ranged from 1.6–3.3 L hr-1 with algae and 2.2 – 4.7 L

hr-1 with bacteria. No selection based on particle size was observed; both particles are

approximately the same size (2-4 µm for the microalgae and 0.7 x 3 µm for the bacteria).

Filtration rates increased when the concentration of algal food increased, both with and

without the controls. However, filtration rates decreased with an increasing

concentration of bacteria as a food source. These results are not expected if tunicates are

non-discriminate filter-feeders as claimed by Jørgensen(1966b). One possible explanation

may be that the nutritional value of bacteria satisfies the metabolic needs of the tunicate

more sufficiently than does algae, resulting in less water filtered per unit time once this need is met. Further testing is needed to determine if this is the case.

By extrapolating the results of this study, 200 S. plicata would be able to filter as

much as 940 L hr-1 or 22,560 L day-1 of bacterially contaminated waters at a

concentration of 105 cells mL-1. These same 200 organisms would filter 100% of the

particulates in as much as 660 L hr-1 or 15,840 L day-1 of algae at a concentration of 106

cells mL-1. One hundred percent clearance of particulates is not necessary for

bioremediation. A two to three-fold reduction would restore the water to healthy

standards in most cases, making the potential impact even higher. This filtration ability of S. plicata warrants further investigation for its potential positive impact on the environment.

CHAPTER THREE: LONG AND SHORT TERM SALINITY TOLERANCE

Abstract

The rough tunicate Styela plicata was exposed to various salinities through long-term

adaptation and short-term shocks. Determination of the response of the organism to

salinity changes was evaluated because the tunicate has been considered for

bioremediation of microalgal and bacterial blooms. Tunicates were collected from the

wild and acclimated for four days preceding experimentation. Six 100 L vessels

contained ten tunicates each and were separated into two treatment levels (n=3 per

treatment) to determine long-term salinity adaptation; 1) decreasing salinities of 1 ppt

every four days and 2) control specimens maintained at 32 ppt. S. plicata maintained at

32 ppt had a survival expectancy from 40 days to the conclusion of the experiment (58

days), within 95% confidence limits, (mean 46.4 days, ± 2.5 s.e.). This compares to only

36 to 46 days (mean 40.7 days ± 2.1 s.e.) of survival expectancy, within 95% confidence

limits, for the decreasing salinities. The critical safe salinity was 24 ppt for long-term

adaptation. Short-term salinity shocks (i.e., step changes and exposed for maximally 4

days) from 32 ppt to 20 ppt, resulted in 40% mortality after four days. Exposure to 16

ppt resulted in 87% mortality within 24 hours and exposure to 12 ppt resulted in 100%

mortality within 24 hours. Biostructures with tunicates for permanent remediation should

not be placed where salinities fall below the critical safe salinity limit of 24 ppt for extended periods of time. Tunicates may be implemented in emergency situations of

microalgal or bacterial blooms in salinities as low as 20 ppt however, fatalities of the animals will occur. Their overall bioremedial effects on the blooms may outweigh the

loss of tunicates.

INTRODUCTION

As coastal regions become increasingly populated, estuaries are progressively

affected by coastal development (Beck et al., 2001; Halpern et al., 2007; Heck et al.,

2003; Lipcius et al., 2005; Lotze et al., 2006). Organic loads from fertilizer run-off, increasing turbidity due to dredging, construction, industry, recreation, and the

destruction of natural wetlands has had a negative impact on the ecological balance of the

Indian River Lagoon (IRL) along Florida’s east coast. The IRL’s natural habitat of

wetlands and seagrass beds was once teeming with benthic filter-feeders (Bingham,

1992). Historically, remediation by these filter-feeders was effective (Dame et al., 1984;

Officer et al., 1982; Ulanowicz & Tuttle, 1992). As these habitats were destroyed and

replaced by bridges, locks, and seawalls, the populations of these filterers decreased. A

decline in water quality has been attributed to the loss of filtration benefits of filter-

feeders according to Dame et al. (2002) and has been directly correlated to increases in

bacteria, toxic algae, anoxia, fish kills, and consequential beach closures (Worm et al.,

2006). Problems such as these force scientists to come up with new methods for

controlling water quality.

The utilization of introduced biostructures to replace the loss of filter-feeder

habitat is a possible solution. These biostructures would be populated by organisms that

could remediate the water by removing seston, both organic and inorganic. Bivalves,

such as oysters, clams, and mussels have the ability to remove large amounts of

suspended particulates (Grizzle et al., 2006; Kohata et al., 2003; Newell, 1988; Newell et

al., 2005; Officer et al., 1982), but the retention efficiency with particles <4 µm is low

(Stuart & Klumpp, 1984) making them ineffective against bacteria and some microalgae.

Draughon et al. (2008) examined the use of the rough tunicate Styela plicata for

the purpose of estuarine bioremediation with fecal coliform bacteria and microalgae. S.

plicata is found in the IRL as well as most sub-tropical and tropical marine waters globally (Carballo & Naranjo, 2002; Cima et al., 1996; Lambert & Lambert, 1998;

Thiyagarajan & Qian, 2003; Tujula et al., 2001). Laboratory tests of filtration rates

showed a population of 14 animals averaged 3.2 L hr-1 per animal with bacteria and 3.6 L

hr-1 per animal with microalgae with 100% retention efficiency (Draughon et al., 2008).

If biostructures housing approximately 200 organisms were strategically placed in areas of bacterial or microalgal bloom, roughly 15,000 to 17,000 liters per day would be filtered with 100% of all bacterial and microalgal particles removed.

Areas of potential placement depend on various environmental factors. In the present study, the possible implementation of S. plicata as a bioremediation agent includes examining the tunicate’s salinity tolerance levels, both with long-term adaptation exposure and short-term shock exposure. Conditions in the IRL, as with most estuarine bodies of water, are in a constant state of flux due to weather, temperature, and discharges of fresh water from rivers, lakes, and canals. In one week’s time during

September of 2008, the salinity at the mouth of the St. Lucie River Estuary, centrally located in the IRL, dropped from 31.0 ppt to 15.3 ppt due to Tropical Storm Fay. The salinity in this location remained at or below 20 ppt for four weeks. Over the next four months the salinity slowly rose to over 35 ppt at this same site (Florida Oceanographic

Society, 2004). Additional regional variation occurs with proximity to ocean-access inlets and freshwater inputs. Annual averages can vary as well. To anticipate the locations where S. plicata could be used for bioremediation, its salinity tolerances must be considered.

MATERIALS AND METHODS

Long-term Salinity Tolerance Adaptation

Specimens of S. plicata were collected during June, 2006, from the IRL near Ft.

Pierce, Florida (27° 27’ 52” N and 80° 17’ 41” W). Salinity was 30.8 ppt and water temperature was 27.5ºC at the time of collection. Ten specimens were each placed in six

100L tanks containing 40L ozonated, UV-sterilized, and 5 µm-filtered seawater. Starting salinity, pH and temperature were 32 ppt, 8.3, and 28.5º C, respectively. Water temperature was allowed to fluctuate with ambient air similar to that found in the estuary.

Dissolved oxygen (DO) was maintained above 80% by aeration stones placed in the bottom of the tanks. Treatments were randomly assigned to tanks along two rows with three control tanks remaining at 32 ppt and the other three reduced in unison by 1 ppt every four days, for a total of 58 days. A 50% water exchange was performed every four

+ days to maintain pH and lower NH3 and NH4 levels. Tunicates were allowed an

acclimation period of four days prior to any experimentation. A diet of DT’s Live

Marine Phytoplankton (DT’s Phytoplankton Farm, Sycamore, IL, USA) was provided at

a concentration of 1 x 106 cells mL-1 per tunicate every other day during acclimation and

the experiment.

Specimens were tested for responsiveness to the salinity conditions every two

days by lightly touching them with a small plastic rod between the incurrent and atrial

siphons. Rapid retraction of the siphons was indicative of good health. Slow siphon

retraction by any individual was noted and the animal considered stressed. Subsequent responsiveness tests with that animal were given special attention. Failure of an animal to retract the siphons after two consecutive testing periods (four days) was considered

evidence that the animal had expired. The date and salinity when the organism first

failed to respond were noted, and taken as the time of death. Salinity levels were reduced

by 1 ppt in the test tanks every four days until 100% mortality was experienced in all

experimental tanks.

Short-term Salinity Tolerance Shock

Specimens were gathered along the northern seawall of the ship channel of

Harbor Branch Oceanographic Institute at Florida Atlantic University in Ft. Pierce, FL

(27° 32’ 05” N and 80° 21’ 07” W) during the month of July, 2008. Environmental

salinity, water temperature, and pH were 31.3 ppt, 30.0ºC, and 8.0 respectively, at the time of collection. Specimens were cleaned of commensal organisms and placed in 100L

tanks containing ozonated, UV-sterilized, and 5 µm-filtered seawater for 18 hours for acclimation. Twenty 8L beakers (four treatments with five replicates) containing 6L seawater were randomly placed in 3.1 m x 0.66 m x 0.3 m tanks with flow-through water to create a water bath and maintain temperature near 26°C. Each beaker contained three tunicates. Treatment levels were 32 ppt (controls), 20 ppt, 16 ppt, and 12 ppt. Salinities were prepared by mixing filtered seawater with freshwater purified by reverse osmosis.

Aeration was provided through a pipette tip placed at the bottom of the beaker secured

away from the tunicates to prevent interference with filtration. Tunicates were fed a diet

of the microalgae Chaetoceros gracilis at a concentration of 1 x 105 cells mL-1 beaker -1

per day. Each beaker received a total water exchange daily to compensate for any

remaining uneaten food. Mortality was monitored on a daily basis using the same

method as done in the long-term adaptation experiment. Additionally, any tunicate found

floating at the surface of the water was relocated back to the bottom of the beaker. Those

that would not stay on the bottom were allowed to float but this was usually indicative of

a stressed animal. Death usually followed soon thereafter.

Statistical Analysis

Statistics for both the long-term adaptation study and short-term shock study were

analyzed using SAS version 9.1. Survival data for the long-term study of controls versus decreasing salinities utilized the Kaplan-Meier method with PROC LIFETEST to determine mean survival times within 95% confidence levels. Death times were right- hand censored since death may have occurred any time within a two day window.

Survival analysis, sometimes referred to as failure analysis, is a collection of statistical methods used to analyze the timing and occurrence of events, (i.e., economical trends, equipment failures, arrests, earthquakes, and deaths) (Allison, 2008). As some factors of the individual animals are unknown and will affect the time of death, (i.e., age or health of the animal when taken into captivity), results are reported as a mean time of death plus or minus standard error, within a time frame. The method works by assigning a probability (survivor function) of surviving beyond a specific time. PROC LIFETEST then compares the survivor functions between the different subgroups in a sample. The

group with the higher survivor function lives longer (Allison, 2008).

In the short-term study, PROC MIXED was utilized with salinities as a fixed

effect nested within beakers to determine the presence of a significant effect of days alive

between salinities.

RESULTS AND DISCUSSION

Long-term Salinity Tolerance Adaptation

When S. plicata is in its natural environment, a wide variety of food composition

is available, enabling all nutritional requirements to be met. Duplication of this situation

in a laboratory is difficult making long-term captive maintenance difficult. Additionally,

S. plicata only lives approximately one year from hatching to death (Goodbody, 1962).

The tunicates’ age at time of collection was unknown; however, all were adults. Of the

sixty animals examined (thirty as controls and thirty experimental with decreasing

salinity) rates of failure (death) were very similar at the onset of the experiment when salinity rates were comparable. As seen in Figure 3.1, the first fatality occurred in the experimental group on day 10 at 30 ppt, followed by the first fatality among the controls

(32 ppt) on day 16. From 28 ppt down to 25 ppt (i.e., days 20 through 30), equal numbers of fatalities occurred in both the control and experimental groups). A noticeable difference in mortality began after exposure to 24 ppt for four days (i.e., 38 days after the experiment started). At this time, the experimental group had exhibited a cumulative mortality of 46.7%, compared to only 33.3% in the controls. By day 46, fatalities in control groups were extensive, with only 11 specimens remaining, compared to 8 in the experimental groups, likely due to time in captivity.

Day 38 30

25 Day 46 20 Day 10 Day 16

Number of Living Organisms Living of Number 5

9 8 7 6 2 1 0 32 32 31 30 2 2 2 2 25 24 23 2 2 2 19 Salinity in ppt

living control specimens living experimental specimens

Figure 3.1 Mortality of Styela plicata held at stable salinity versus decreasing salinities. Difference in fatality rates occurred after exposure to 24 parts per thousand (ppt). NOTE: Data points represent 2 day intervals. Salinity decreased 1 ppt every four days.

Survival analysis revealed an expected survival time for tunicates maintained at

32 ppt between 40 days and the conclusion of the experiment at 58 days (within 95% confidence limits). Mean survival time was 46.4 days ± 2.5 s.e. This compares to only

36 to 46 days with a mean survival time of 40.7 days ± 2.1 s.e. (within 95% confidence limits) for those subjected to decreasing salinities. Salinities on day 36 were 24 ppt and decreased by 1 ppt down to 22 ppt by day 46 (Fig 3.2) indicating the critical safe salinity for survival was 24 ppt.

1.2

1.0

0.8

0.6

Survival Function 0.4

0.2

0.0 0 2 4 6 8 10121416182022242628303234363840424446485052545658 Days

Controls Decreasing Salinities Figure 3.2 Survival function (probability of surviving beyond a specific time) versus days alive for controls remaining at 32 ppt and decreasing salinities. Light gray box represents survival expectancy within 95% confidence limits for 32 ppt with mean of 46.4 days (upward facing arrow). Darker gray box represents survival expectancy within 95% confidence limits for decreasing salinities. Downward facing arrow is mean of 40.7 days, corresponding to 23 ppt. Error bars represent standard error. NOTE: Survival expectancy should not be confused with life expectancy. Tunicate starting age upon onset of the experiment was unknown.

Short-term Salinity Tolerance Shock

Tunicates involved in the short-term study were examined daily for a response when touched, as was described above for the long-term study. An animal with a closed siphon during examination was considered deceased if it was floating in the vessel and failed to respond to any subsequent stimulation. Tunicates were considered “potentially alive” until other evidence of death appeared, such as turning white or becoming coated with slime. The lowest salinity tested, 12 ppt, resulted in the death of all exposed animals within the first 24 hours. In the16 ppt salinity treatment, only two tunicates survived the

first 24 hours and one of them survived until 48 hours (Fig. 3.3). Nine out of the 15

tunicates exposed to a sudden salinity drop to 20 ppt were still alive after four days; however, some were slow to respond, exhibiting a stressed state. They were returned to the 32 ppt salinity upon completion of the four-day experiment where they seemed to

recover after 24 hours and resumed normal filtration, responding rapidly to stimulation.

All tunicates in the control beakers at 32 ppt were alive at the end of the four-day experiment (Fig. 3.3).

4 Living Specimens Living 2

0 24 48 72 96 Hours

ppt 32 ppt 20 ppt 16 ppt 12

Figure 3.3 Number of living animals per day in each of four salinities during the short term salinity shock exposure study. Bars denote potential, yet doubtful living specimens.

Salinity was found to have a significant effect on survival (F 3,16 = 57.47,

p<0.001) when all four treatments were analyzed. As little to no survivorship existed in

the 12 ppt and 16 ppt treatments, a contrast statement between the 20 ppt and 32 treatments was added. Although 60% of the S. plicata survived the four-day exposure to

20 ppt, the survival difference between tunicates maintained at 20 ppt and 32 ppt was

significant (F 1,16 = 56.03, p<0.0001).

CONCLUSION

S. plicata survived exposure to slowly decreasing salinities from 32 to 24 ppt signifying tolerance and resilience within this range. This type of exposure in the natural

environment would be similar to slow, steady rains over a period of days. As long as

salinities remain at or above 24 ppt, no high mortality rates would be expected. Below

this salinity fatalities increased, indicating the lower salinity limit had been reached.

These findings correspond to those published by Thiyagarajan and Qian (2003), who

found that fertilized eggs of S. plicata in Hong Kong did not develop at or below 26 ppt.

Prolonged exposure to salinities below 24 ppt may result in a massive die-off, causing

increased bioload from dead and decaying animals. For this reason, the use S. plicata for the purpose of long term bioremediation in salinities below 24 ppt is not recommended.

Lower salinities can be tolerated without catastrophic results if the exposure is rapid but temporary. Of the 15 test animals examined, 60% survived a sudden plunge

from 32 ppt to 20 ppt for a maximum period of four days. Sudden changes such as this

may be necessary when moving animals bred in captivity to areas of elevated algal and

bacterial levels for the purpose of bioremediation, especially if these areas are near

freshwater canals or rivers. Prolonged exposure in these locations is not recommended.

Filtration benefits provided by S. plicata is substantial and bioremediation efforts

utilizing their ability warrants investigation. However, due to salinity fluctuations in

estuarine environments, close monitoring is suggested to avoid sudden massive die-offs

that may increase overall bioload. In areas of more stable salinities, communities on pre-

seeded substrates could be introduced for long term remediation. In such cases, these

communities should self-perpetuate with breeding seasons in Spring and Fall (West &

Lambert, 1976) maintaining populations.

CHAPTER FOUR: POST-DIGESTION BACTERIAL VIABILITY

Abstract

The tunicate Styela plicata has been suggested for use as a bioremediator of bacterial contamination having a filtration capacity of 4.7 L hr-1 with 100% efficiency of removing

Escherichia coli at a concentration of 105 cells mL-1. The filtration and removal

capacities are of limited remedial use if the bacteria removed are still viable and simply

repackaged in a mucous-bound fecal pellet. This study investigated the viability of E.

coli before and after digestion by Styela plicata. Viability was assessed by measuring

microbial ATP present in a sample of E. coli before filtration, during filtration, and after

re-suspension of bacteria within expelled fecal pellets. The ATP assay has the ability to

quantify living from non-living microbial cells because only live bacteria produce

ATP. Four treatments were established; 1) treated seawater only control (n=3), 2)

tunicate only in treated seawater control (n=3), 3) 5 x 105 cells mL-1 of Escherichia coli in treated seawater control (n=3), and 4) tunicates with addition of 5 x 105 cells mL-1 of

Escherichia coli in treated seawater (n=9). Hourly water samples were taken from each replicate for five hours and analyzed for ATP. Initial bacterial density in the water only beakers averaged 4.53 x 105 cells ml-1 ± 2.61 x 105 cells mL-1 s.d. and decreased after the first hour to 1.47 x 105 cells ml-1 ± 4.45 x 104 cells mL-1 s.d. where it remained relatively

stable through the fifth hour with an ending density of 1.54 x 105 cells mL-1 ± 4.05 x 104

cells mL-1 s.d. Initial bacterial density in beakers containing only tunicates averaged

2.44 x 105 cells mL-1 ± 5.33 x 104 cells mL-1 s.d., ending with an average of 2.85 x 104 cells mL-1 ± 1.13 x 104 cells mL-1 s.d. Initial bacterial density in beakers containing only bacteria averaged 2.85 x 106 cells mL-1 ± 2.76 x 105 cells mL-1 s.d. with an increase to

2.99 x 106 cells mL-1 ± 3.58 x 105 cells mL-1 s.d. at the end of the five hour experiment.

Average starting density of bacteria in beakers containing tunicates with the addition of

E. coli was 2.35 x 106 cells mL-1 ± 6.53 x 105 cells mL-1 s.d. which decreased by approximately one order of magnitude within two hours in seven out of the nine replicates. Of the two remaining replicates, 62% and 42% of the bacteria was depleted within the first two hours. No significant change in mean ATP amounts as a result of time was found, however MANOVA revealed a significant overall treatment effect

F (15, 28.007) = 10.38 p<0.0001 and time*treatment effect F (12, 29. 395) =2.23, p=0.0381. Water samples with resuspended fecal pellets produced by the tunicates during the course of the experiment showed an absence of detectable microbial ATP.

This indicated that bacteria removed from the marine water by S. plicata were effectively killed by the tunicate during the filtration/digestion process or by processes that continue after expulsion of the fecal pellet. This is evidence that S. plicata not only removes bacteria from a water column but also renders it incapable of replicating and further indicates the potential bioremedial benefits provided by the organism.

INTRODUCTION

The viability of bacteria after digestion was investigated in the present study. The bacterial filtration capability of the tunicate Styela plicata is substantial; 200 animals have the potential of removing 105 bacteria mL-1 from 22,600 L each day (Draughon et al., 2008). The absence of viable bacterial cells after the completion of the digestive

process is of key importance to the success of the bioremedial use of S. plicata.

As reviewed by Bone et al. (2003), the tunicate digestive system acts as a particulate filter and has been well studied by many researchers. Mucus is secreted by an endostyle along the ventral side of the branchial sac (pharynx) and is moved by cilia along a rectangular meshwork of filaments toward the dorsal lamina. Particulate is trapped in the mucus web as water enters the incurrent siphon, circulates through the pharynx, and exits the atrial siphon. This particulate-embedded mucus is formed into a continuous string (Goodbody, 1974; Jørgensen, 1966b) at the dorsal lamina, where it is fed into the esophagus and continues through the digestive tract. The size of particulates present in the mucous strand varies with species, depending on the complexity of the pharynx and the size of the mesh pore. Styela plicata, of the order Stolidobranchiata, has considerably larger mesh pores than most other ascidians. A study conducted by Flood and Fiala-Médioni (1981) found the transverse strands of the mesh net to be only ~10 nm in diameter, with the longitudinal strands ~20 nm. The longitudinal and latitudinal sides of the rectangular pore were often two to three times that of the other tested stolidobranchs. The open fraction of the net, or pore size, was 97-98%, yet the width of the net pores in S. plicata was only ~0.5 µm, sufficiently small enough to filter bacteria

(Flood & Fiala-Médioni, 1981). According to Wallace and Malas (1976), the elongated, rectangular shape of the pores represents the most economical way of creating a filtering net with the least amount of matter. Tranter and Smith (1968) added that this shape also represents the least amount of energy expenditure for pumping. The physiology of this net enables S. plicata to effectively filter the bacteria in the present study.

Difficulty of counting living aquatic bacteria was first documented in the 1930’s

(Butkevich & Butkevich, 1936; Butkevich, 1932, 1938; Radsimovsky, 1930) and has been the subject of numerous studies and reviews (Amann et al., 1995; Brock, 1987; Bull

& Hardman, 1991; Hugenholtz et al., 1998; Jannasch & Jones, 1959; Roszak & Colwell,

1987; Staley & Konopka, 1985; Yasumoto-Hirose et al., 2006). Staley and Konopka

(1985) described “The Great Plate Anomaly”, noting that direct microscopic counting of bacteria often exceeds cultured colonies on plates by orders of magnitude, which may be attributable to various factors (Jannasch & Jones, 1959). For this reason, the detection of viable bacteria after digestion by S. plicata was determined by adenosine triphosphate

(ATP) assay, which has been validated as an accurate way to isolate and count only viable bacteria (Bautista et al., 1997; Deininger & Lee, 2001; Deininger & Lee, 2005;

Holm-Hansen & Booth, 1966; Kimmich et al., 1975; Trudil et al., 2000).

The quantification by ATP assumes that only living organisms produce ATP and there is little to no ATP production in non-living organisms. Background ATP from non- living material was examined in a study by Holm-Hansen and Booth (1966) when they killed various algae and bacteria by heat, freezing, and cyanide. Any residual ATP was found negligible.

The ATP assay works by coupling ATP with D-luciferin and O2 in the presence of firefly luciferase resulting in the oxidation of luciferin and the admission of light

(Fig. 4.1). The intensity of the light is directly proportional to the amount of ATP present and can be calculated down to the pmol (10-12 mole) (BioThema, 2007) with a luminometer.

luciferase ATP + D-luciferin + O2

AMP + PPi + Oxyluciferin + CO2 + light

Figure 4.1 Chemical reaction during ATP assay. ATP present in living organisms reacts with D-luciferin in the presence of oxygen and firefly luciferase to result in emission of light detectable by a luminometer. MATERIALS AND METHODS

Collection and Acclimation of Specimens

Specimens were collected along the ship channel wall of Harbor Branch

Oceanographic Institute at Florida Atlantic University during April, 2008 (27°

32’05.28”N, 80°21’07.42”W, Fig. 4.2) and were immediately cleaned of any commensal organisms. Two to three animals (approximate 60 g total weight) were secured by fishing line to individual 12 cm x 12 cm plastic grids (1.2 cm x 1.2 cm openings) covered with 1 mm mesh and were placed in 8 L beakers containing 6 L of ozonated, UV- irradiated, and filtered (5 µm) seawater. Aeration was provided through a pipette tip secured away from the animals. Specimens were allowed to acclimate for 48 hours prior to any experimentation. During this time, food was withheld to allow clearance of gut contents.

Figure 4.2 Specimen collection site along northern ship channel wall of Harbor Branch Oceanographic Institute at FAU, Ft. Pierce, FL (27° 32’05.28”N, 80°21’07.42”W).

Experimental Design

Four treatments were established; WC) a control containing only treated seawater

as described above (n=3), TC) a control containing a tunicate in treated seawater to

monitor additional bacteria originating on the animals, themselves (n=3), BC) a control

containing 5 x 105 cells mL-1 of Escherichia coli in treated seawater (n=3), and EXP) the

tunicates with the addition of 5 x 105 cells mL-1 of E. coli in treated seawater (n=9).

Beakers were each placed on individual stir plates along two rows. Each contained a 12 x 12 cm plastic grid (1.2 cm x 1.2 cm openings) covered in 1 mm mesh. A magnetic stir bar was placed in the bottom of each beaker to enhance circulation and to resuspend any produced fecal pellets. Aeration was provided through a pipette. Each beaker was covered with plastic film to exclude atmospheric bacteria from settling in the beakers

during the experiment.

Initially, 5 x 105 cells mL-1 of E. coli were administered to each of the nine EXP

beakers and the three BC beakers. The TC and WC beakers did not receive bacteria.

Hourly water samples from all beakers were analyzed in triplicate in a BMG Labtech

NOVOstar Multifunctional Microplate Reader (luminometer) using 96 well plates.

Water samples were taken for five hours in all treatments except EXP beakers containing

tunicates that did not produce a fecal pellet. In these beakers, analysis was completed

through four hours, due to a shortage of reagents. The averages of the triplicate samples

were analyzed using SAS 9.1 software. A repeated-measures design over five hours

(critical alpha=0.05) was used to analyze the amounts of ATP over time. Differences

were considered significant for probabilities (p< 0.05). The ‘within subject’s main effect

test’ a priori H0 was that “there is no significant difference in the mean ATP over time”.

The ‘between subject’s a priori H0 was that “ATP did not change over time differently in

the different treatments”. The residuals were normal for Repeated Measures ANOVA

(p=0.7815), however the data did not meet the assumption of sphericity (p=0.0012).

Therefore, MANOVA was performed. Regression analysis was also performed between the ln of cells mL-1 and time. The method used to convert ATP amounts to numbers of

bacterial cells is explained in the following section.

ATP assay and calculations.

The assay used to determine ATP was purchased from BioThema Luminescent

Assays (Handen, Sweden) and is specific for intracellular ATP analysis of microbial

cells. The assay is multi-step beginning with the lysing of somatic cells, which are

enzymatically destroyed. Intracellular ATP released from these cells is extracted and assayed, leaving microbial cells and other contents.

The amount of ATP (pmol) in each sample was calculated using the following formula recommended by BioThema (2007) :

ATPsmp = I smp1/ (Ismp1 + std – Ismp2)

where ATPsmp = the amount of ATP (pmol) in the sample, I smp1= sample + lyophilized

reagent containing luciferase and luciferin, Ismp1 + std = I smp1 + ATP standard, and

Ismp2 = light emission immediately prior to addition of ATP standard.

The amount of ATP in one bacterial cell is 2 x 10 -18 mol (BioThema, 2007; Satoh et al., 2004). From ATPsmp, the number of bacterial cells per well could be determined

using the following formula:

Cellswell = (ATPsmp / M) / ATPbac

= where Cellswell = number of bacterial cells in the 20 µl well, ATPsmp the amount of ATP

(pmol) in the sample, M = number of pmol per mole, and ATPbac = amount of ATP in one

bacterial cell. To determine the amount of ATP mL-1, this amount was multiplied by 50.

RESULTS AND DISCUSSION

Hourly water samples for all controls were analyzed for the duration of the five

hour experiment, however, reagents became sparse during Hr5, necessitating the analysis

of water samples only in EXP beakers in which a fecal pellet had been produced.

Therefore, some graphs will appear through Hr4 and some through Hr5. Additionally,

the preparation time of the triplicate water samples for the 96 well plates was extensive.

Therefore, there is no Hr0 reading. The first ATP assay was approximately one hour

after the placement of bacteria in the beakers. The seawater only (WC) beakers had an

average background bacterial cell density of 4.53 x 105 cells mL-1 ± 2.61 x 105 s.d., (n=3)

at the Hr1 reading that dropped to 1.47 x 105 cells mL-1 ± 4.45 x 104 s.d. by Hr2. The WC

starting bacterial density exceeded the TC control of 2.44 x 105 cells mL-1 ± 5.33 x 104

s.d., (n=3). It was expected to be lower as there should have been a higher amount of

bacteria in the beakers containing water plus tunicates than beakers with just water alone.

Upon examination of the individual WC beakers, it was noted that WC2 started with a

bacterial concentration that was three times higher than WC1 or WC3. This was

potentially caused by contamination in the beaker or some other unknown contaminant.

The data were retained but are suspect, as the exact cause of the higher reading is

unknown. Subsequent average WC readings were similar to the Hr2 reading at 1.71 x

105, 1.59 x 105, and 1.54 cells mL-1 for Hr3 through Hr5 respectively.

Figure 4.3 reveals erratic bacterial counts each hour in the TC beakers, potentially indicating sporadic release of bacteria from areas of the animal’s test.

Tunicate #1 experienced a sudden increase in bacterial concentration; almost triple that of

the previous hour, without production of a fecal pellet. The increase went from 2.3 x 105 cells mL-1 at Hr3 to 6.2 x 105 cells mL-1 at Hr4. The excess bacteria were quickly filtered by the animal with a decrease to 3.5 x 104 cells mL-1 by Hr5. The mean load of all TC

replicates were 2.44 x 105 cells mL-1 ± 5.33 x 104 cells mL-1 s.d. at Hr1 and remained

comparatively steady at 2.25 x 105 cells mL-1 ± 9.30 x 104 cells mL-1 s.d., 2.12 x 105 cells mL-1 ± 9.39 x 104 cells mL-1 s.d., and 3.14 x 105 cells mL-1 ± 2.69 x 105 cells mL-1 s.d., for

Hrs 2-4. Between Hr4 and Hr5, the mean density decreased by one order of magnitude

(2.85 x 104 cells ± 1.13 x 104 cells mL-1 s.d.). Overall, the mean background bacterial

density in the TC beakers decreased over the five hours following a predictive equation of

“density” = 409938 - 62556(Hr); (r2 = 0.7628).

More uniform filtration from hour to hour by the tunicates in the TC controls was

expected. However, as bacteria were released from the tunic, concentrations in the water

column changed. As increases occurred, filtration of the excess bacteria by the tunicates

followed, bringing the concentrations back down (Fig. 4.4). Even though these

fluctuations in concentrations occurred, the overall trend for the averages in the TC

controls was relatively steady; indicating that background bacteria from the animal tests

would not interfere with the experiment.

7.00E+05

6.00E+05 y = -62556x + 409938 R2 = 0.7628

5.00E+05

4.00E+05 TC1 TC2 TC3 3.00E+05 Linear (TC2) Bacteria in Cells per mL in Cells Bacteria 2.00E+05

1.00E+05

0.00E+00 Hr 1 Hr 2 Hr 3 Hr 4 Hr 5 Hourly Measurement

Figure 4.3 Hourly bacterial counts in tunicate control beakers over the course of the five hour experiment. Dashed line is trend representing predictive equation of “density” = 409938 -62556(Hr) (r2 = 0.7628). Sudden increase in bacterial counts at Hr 4 in specimen #1 was from some unknown contaminant likely originating from the tunicate test and was not from the production of a fecal pellet.

The BC controls would establish changes in the bacterial populations over the course of the five hour experiment due to replication or die off. Alkan et al. (1995) noted that, under normal circumstances, bacteria do not usually replicate in marine

environments. However it is known that bacteria can replicate in this environment under

certain conditions (Rozen & Belkin, 2001; Troussellier et al., 1998). The initial

background mean (n=3) bacterial density at Hr1 was 2.82 x 106 cells mL-1 ± 2.76 x 105

cells mL-1 s.d. Fig. 4.4 reveals an increase in cell counts, (insufficient to indicate cell division that would interfere with the experiment). Average density for Hr2 was 2.86 x

106 cells mL-1 ± 2.74 x 105 cells mL-1 s.d., Hr3 at 2.97 x 106 cells mL-1 ± 4.11 x 105 cells mL-1 s.d., Hr4 at 3.1 x 106 cells mL-1 ± 2.33 x 105 cells mL-1 s.d., and 2.99 x 106 cells

mL-1 ± 3.58 x 105 cells mL-1 s.d. at Hr5.

The addition of 5 x 105 cells mL-1 of E. coli added to the background bacteria

from the water and on the tunicate resulted in Hr1 readings in the EXP treatment ranging

from 1.55 x 106 to 3.34 x 106 cells mL-1 with a mean (n=9) of 2.35 x 106 cells

mL-1 ± 6.53 x 105 cells mL-1 s.d. Regression analysis of hour versus the natural log of

bacterial density revealed the average hourly depletion of bacteria, followed a predictive

equation of “ln(cells mL-1 ) = 15.041 – 0.06146(hour)” (r2=0.6387, n=9), with the

majority of tunicates removing the maximum bacterial load by Hr3 (Fig. 4.5). The first

fecal pellet was produced by specimen #2 at Hr4. The pellet fell to the bottom of the

beaker and was resuspended by the magnetic stirbar preceding subsequent ATP

measurements. No detectible spike in ATP was found. Bacterial counts in this replicate

were 3.19 x 105 bacterial cells mL-1 at Hr3 that decreased to 2.08 x 105 at Hr4. After the

pellet was resuspended the sample was retested at Hr5 and had 1.72 x 105 cells mL-1.

Production of small fecal pellets was common after Hr4 in several beakers, but resulted

in no subsequent ATP spikes. MANOVA revealed no significant change in the mean

ATP over time (F 4,11 = 2.24, p=0.1314) however, there was a significant overall

treatment effect (F 15, 28.007 = 10.38 p<0.0001), and interaction of time with treatment,

(F 12, 29. 395 = 2.23, p=0.0381).

3.50E+06

3.00E+06

2.50E+06 -1

2.00E+06

1.50E+06 Bacterial Cells mL Cells Bacterial 1.00E+06

5.00E+05

0.00E+00 Hr 1 Hr 2 Hr 3 Hr 4 Time WC TC BC EXP Figure 4.4 Averages of bacterial cell density over time in water only controls (WC) (n=3), tunicate only controls (TC) (n=3), bacterial only controls (BC) (n=3), and experimental beakers containing tunicates plus addition of 5 x 105 cells mL-1 of E. coli (EXP) (n=9).

1.00E+07

1.00E+06

1.00E+05

-1 Series1 Series2 1.00E+04 Series3 Series4 Series5 Series6 1.00E+03 Series7 Series8 Series9 Bacterial Cells mL Cells Bacterial 1.00E+02

1.00E+01

1.00E+00 Hr 1 Hr 2 Hr 3 Hr 4 Time Figure 4.5 Log plot of hourly bacterial counts mL-1 over the course of the four hour experiment in each of nine test specimen beakers. Regression analysis of hour versus ln of cells mL-1 revealed average depletion of bacteria per hour followed a predictive equation of “ln(cells mL-1 )=15.041– 0.06146(hour)”, (r2=0.6387, n=9).

CONCLUSION

ATP assay is an extremely sensitive way of determining viable bacteria in a sample. The mean of 2.35 x 106 cells mL-1 of bacteria at Hr1 in the beakers containing tunicates plus bacteria was reduced by one order of magnitude in all but two beakers within the first two hours by S. plicata. No increase in ATP production from the resuspended fecal pellets was found over time, though several fecal pellets were produced over the course of the experiment. This indicates ingested bacteria were rendered non-viable during the digestion process within the tunicate, further indicating the potential usefulness of S. plicata as a bacterial bioremediator.

CHAPTER FIVE: POTENTIAL EFFECTS OF STYELA PLICATA ON

ICHTHYOFAUNAL COMMUNITIES

Abstract

The rough tunicate Styela plicata (Ascidiacea) has been suggested as a bioremediator of

bacterial and algal contamination in estuarine waters, however it may affect the

ichthyofaunal community by removing sperm, eggs or embryos, such as pelagic Sciaenid

eggs that are as small as 0.6 mm and sciaenid larvae as small as 1.3 mm. The present

study investigated particles, simulating fish eggs, which could be retained by the tunicate.

Three sizes of glass beads (0.5 mm, 1.0 mm, and 2.0 mm diam.) were utilized. Fifty 0.5

mm and 1.0 mm beads and 20 2.0 mm beads were delivered to specimens (n=18) one minute apart, with each size delivered on separate days. Tunicates were monitored for bead retention for two hours. After ingestion, 72.5% of the 0.5 mm beads were retained,

49.4% of the 1.0 mm beads were retained, and 43.5% of the 2.0 mm beads were retained.

Many of these beads were later rejected when the animals were laid on their sides. When examining the percentage of particles expelled by the tunicates, neither siphon diameter

(F 1, 54 = 0.95, p= 0.3342) nor height (F 1,54 = 1.2, p=0.2781) were found to be significant

factors, nor was their an interaction between siphon diameter and height

(F 2,54 = 0.57 p= 0.4525). Percentage of particles expelled according to particle size was

also marginally not significant (F (1,54) = 3.17, p= 0.0512).

INTRODUCTION

The Indian River Lagoon (IRL), an estuary system along the northern and central

east coast of Florida, is home to one of the most diverse arrays of ichthyofauna found in

any estuary in the United States (Gilmore, 1995; Reyier & Shenker, 2007; Virnstein,

1995) and is home to 685 fish species (SFWMD et al., 2007). It is estimated that $30

million of revenue is generated annually from the IRL’s commercial fisheries with

additional millions spent on recreational boating and fishing (Montgomery & Smith,

1983; SFWMD et al., 2007), making it a very closely monitored industry. Over the past

century, fish population densities in the IRL declined due to anthropogenic disturbances

(Gilmore, 1995), salinity changes (Lirman et al., 2008; SFWMD et al., 2007), and loss of

seagrass beds (Cardoso et al., 2004; Fletcher & Fletcher, 1995). These disturbances, along with overfishing, have caused some species to be classified as threatened or endangered (Smithsonian Marine Station at Ft. Pierce, 2005b). Stock enhancements and other protective measures, such as bag and size limits or season limits are in place for several species (MyFlorida.com, 2010) in an attempt to control these declines (Table

5.1).

The tunicate Styela plicata (Ascidiacea) has been suggested as a bioremediator of bacterial and algal contamination in the IRL resulting from agricultural runoff, poor circulation, and elevated temperatures (Draughon et al., 2008). Prior to bioremedial implementation in vivo, it is necessary to investigate whether the effort would result in unintentional filtration and removal of eggs and larvae from commercially or recreationally important ichthyofaunal species by the tunicate.

A recent survey of the northern IRL complex revealed several species of sportfish

larvae from the family Sciaenidae, including black and red drums (Pogonias cromis and

Sciaenops ocellatus, respectively), southern kingfish (Menticirrhus americanus), spotted

seatrout (Cynoscion nebulosus), and Atlantic croaker (Micropogonias undulates) (Reyier

& Shenker, 2007). In addition to these commercially and recreationally important fish

are nine species of Lutjanidae (snappers) (Smithsonian Marine Station at Ft. Pierce,

2005a) and four different species of Centropomidae (snooks) (Smithsonian Marine

Station at Ft. Pierce, 2005b). Also potentially present is the extremely rare and

endangered shortnose sturgeon Acipenser brevirostrum (Acipenseridae) (National Marine

Fisheries Service, 1998). Hence, protection of this species is of extreme importance.

Not all of the fishes of commercial, recreational, and endangered importance have eggs or larvae present in the estuary. Members of Lutjanidae spawn offshore in

aggregates with only juveniles greater than 10 mm entering the estuary to seek shelter

amid shallow water (Smithsonian Marine Station at Ft. Pierce, 2005a; Thompson &

Munro, 1974; Wicklund, 1969). Centropomus undecimalis (common snook) spawn at

the mouths of rivers and canals with the larvae moving immediately into fresh water until

reaching >40mm (Gilmore et al., 1983; Lewis III, 1991; MyFlorida.com, 2010;

Smithsonian Marine Station at Ft. Pierce, 2005b). Other sport fish such as tarpon

Megalops atlanticus (Megalopidae), Florida Pompano Trachinotus carolinus

(Carangidae), and sheepshead Archosargus probatocephalus (Sparidae) are found as

juveniles within the estuary, but spawn offshore (Crabtree, 1995; Crabtree et al., 1992;

Gilbert, 1986; Gilhen et al., 1976; Jennings, 1985; Muller et al., 2002).

Of primary concern are eggs and larvae from members of the Sciaenidae. Species

such as silver perch Bairdiella chrysoura, spotted seatrout Cynoscion nebulosus,

weakfish Cynoscion regalis, black drum Pogonias cromis, and red drum Sciaenops

ocellatus use the estuary as spawning grounds (Adams & Tremain, 2000; Connaughton &

Taylor, 1996; Murphy et al., 1998; Paperno et al., 2006; Roumillat & Brouwer, 2004).

Other Sciaenidae species such as Atlantic croaker Micropogonias undulates, and spot

Leiostomus xanthurus spawn offshore with larvae migrating into the estuary (Center for

Quantitative Fisheries Ecology, 2010). Most members of the Sciaenidae lay pelagic eggs

that could potentially be filtered by bioremedial clusters of S. plicata if conditions were

right. Many of these eggs are similar in size and morphology making field identification

difficult (Daniel & Graves, 1994; Joseph et al., 1964). However, spotted seatrout

Cynoscion nebulosus has the smallest eggs (0.6 mm) and the smallest hatchling larval size (1.3 – 1.6 mm) of the species that utilize the estuary for spawning grounds (Table

5.1).

The minimum size limitation for particulate removal by S. plicata was studied by

Flood and Fiala-Medioni (1981) and determined to be small enough to trap bacteria;

however, maximum size limits were not established. While studying sensory cells,

Mackie et al. (2006) determined that polystyrene beads of 105 - 125 µm diameter were

almost always accepted by a related species, S. montereyensis. Particles of 355 – 425 µm

were usually accepted if delivered one at a time, but were always rejected if delivered in

clumps or with larger beads of 500 – 600 µm. It was hypothesized that the size of

particles removed by S. plicata would be similar to S. montereyensis.

Table 5.1 Egg and larvae sizes of commercially or recreationally important fish species found in the Indian River Lagoon, along with any protective measures applicable. Excluded from this table are Lutjanidae, Megalopidae, Carangidae, and Sparidae as spawning normally occurs offshore with only juveniles present in the estuary.

Florida Hatchling FAMILY/ Common C or Protective Estuary Eggs (mm) Larvae References (*) Genus species Names R Measures Spawning (mm)

SCIAENIDAE C 0.63 – 0.78 Bairdiella chrysoura silver perch None Yes 1.5 – 1.8 [1-3] (bait) P 0.6 – 0.8 D Cynoscion spotted P C R SL, S, BL Yes 1.3 – 1.6 [1,3] nebulosus seatrout (depending on salinity) 0.68 1.18 Cynoscion regalis weakfish C R SL, BL Yes 1.49 – 1.99 [1-3] P 1.5 – 1.7 but arrives Leiostomus 0.72 – 0.87 spot C R None Rare in estuary [2,3] xanthurus P B 10 – 24 mm Menticirrhus southern 0.63 – 0.87 C R None No 1.7 avg. [2-3,6-7] americanus kingfish P Spawns offshore. Micropogonias Atlantic 0.35 C R None Larvae 2.0 [2-3, 8-11] undulates croaker P move to estuary 0.8 – 1.0 Pogonias cromis black drum R SL, BL Yes 1.9 – 2.4 [3] P 0.8 – 0.98 Sciaenops D P [3] red drum R SL, BL Yes 4.0 – 6.0 ocellatus (depending on salinity) ACIPENSERIDAE Acipenser shortnose 3.0 E E No 7.0 – 11.0 [3] brevirostrum sturgeon D CENTROPOMIDAE Larvae move to 0.6 – 0.8 Centropomus [3-4,12] snook R SL, S, BL fresh water P 1.4 – 1.5 undecimalis until >40mm Key to Table terms: C = Commercial value, R = Recreational value, E = Endangered, SL = Size limits, S = Seasonal limits, BS = Bag limits, E = Endangered species (no catch allowed), P = Pelagic, D = Demersal, B = Buoyant.

* [1] (Daniel & Graves, 1994) [2] ("Fishbase," 2009) [3] (Robins et al., 1980) [4] (Smithsonian Marine Station at Ft. Pierce, 2005b) [5] (Turner et al., 1985) [6] (Georgia Department of Natural Resources Coastal Resources Division, 2003) [7] (Hildebrand & Cable, 1934) [8] (Texas Parks and Wildlife, 2009) [9] (Center for Quantitative Fisheries Ecology, 2010) [10] (Poling & Fuiman, 1997) [11] (Whitaker, 2005) [12] (MyFlorida.com, 2010)

MATERIALS AND METHODS

Collection and Acclimation of Specimens

Styela plicata were gathered off Captiva Island, Florida (26°29’06.36” N,

82°10’56.90”W) during the month of October, 2008. Specimens were placed in a five

gallon bucket containing aerated seawater and transported to Jupiter, Florida where they

were cleaned of any commensal organisms and placed in clean Instant Ocean artificial

seawater (28 ppt). Animals were fed a diet of DT’s Live Marine Phytoplankton (DT’s

Phytoplankton Farm, Sycamore, IL, USA) at a concentration of 1 x 106 cells mL-1 per day and were allowed to acclimate for 18 hours prior to any experimentation.

Experimental Apparatus and Procedure

Individual tunicates were placed in 4 L beakers containing 3 L artificial seawater (28 ppt). A ring stand positioned a 300 mm glass stem funnel above the incurrent siphon opening (Fig. 5.1). The height of the animal and the diameter of the incurrent siphon of each tunicate was measured and recorded.

Three treatments of glass beads (0.5, 1.0, and 2.0 mm diameter) were individually delivered directly into the Figure 5.1 Experimental apparatus for deliverance of 0.5, 1.0, and 2.0 incurrent siphons. Over a period of three days (one mm glass beads directly into the incurrent siphons of test specimens. day per size), each tunicate (n=18) received fifty of the 0.5 and 1.0 mm beads and twenty of the 2.0 mm beads, one at a time, allowing up to one minute for expulsion of each. After experimentation with a specimen, animals were isolated for two hours to monitor any late bead expulsions before being placed back in a

10 gallon aquarium. Upon the conclusion of all the tests, the animals were monitored for an additional two weeks in the 10 gallon aquarium for continued late expulsions. At this

point, it was impossible to know which bead came from which animal, therefore total

rejection counts only after the initial expulsions are reported.

Statistical Analysis

All statistical analyses were performed using SAS 9.1 software. Two-way

ANOVA (critical alpha=0.05) was used on ranked data (percentage expulsion).

Differences were considered significant for probabilities (p< 0.05). Regression analyses were conducted on 1) siphon diameter versus animal height, 2) height versus percent beads expelled and, 3) siphon diameter versus percent beads expelled.

RESULTS AND DISCUSSION

Statistical Results

There was no relationship between animal height and oral siphon diameter (r2 =

0.2222, n=18), oral siphon diameter and percent particles expelled (r2= 0.0310, n=18),

nor animal height and percent particles expelled (r2 = 0.0333, n=18).

After ranking the data on the percent of particles expelled, assumptions were met

for two-way ANOVA; residuals were normal (p<0.0759) and variances were

homogenous (p<0.0687). No significant difference was found between the percent beads

retained and main effects (siphon diameter: F (1, 54) = 0.95, p= 0.3342;

height: F (1,54) = 1.2, p=0.2781; nor siphon size x height: F (2,54) = 0.57 p=0.4525). The

difference between percent of beads retained and particle size was also, not significant:

F (1,54) = 3.17, p=0.0512.

Bead Retention

Acceptance of the glass beads of all sizes by individual specimens was extremely erratic (Table 5.2). Some specimens appeared more sensitive than others to the triggering of the oral tentacles (Fig. 5.2), resulting in rejection of almost all particles, while others appeared unbothered by the intrusion. Still others appeared to become desensitized after a number of bead introductions, ceasing to reject the particles. The repeated use of the same specimens for the different bead sizes, although a day apart, may have increased this desensitization as a result of fatigue. Sizes of the test specimens were not the determining factor of particle rejection.

S. plicata initially accepted 72.5% of the 0.5 mm glass beads; however, a considerable difference in bead materials exists. Glass beads sank easily into the animal

having a density of 2.4 – 2.8 g cm3 (Giancoli,

1998). Polystyrene beads, as used by Mackie et

al. (2006) have a density of approximately 1.03

g cm3 (Algers et al., 2004), and may not have

been as difficult to expel. This is noteworthy,

considering that continued rejection of the glass

beads occurred when the animals were Figure 5.2 Close up view of the oral tentacles lining the incurrent siphon of tunicate Styela plicata. Triggering of positioned on their sides. However, as the these tentacles causes expulsion of ingested particles. purpose of the present study was to determine the effects on the ichthyofaunal community by filtration of S. plicata, it is doubtful that fish eggs would survive being captured in the mucous net of the tunicate for any extended time.

Of the 1.0 mm beads, 49.4% were initially accepted but many of these were later

rejected when the animals were placed on their sides. When the 2.0 mm beads were

introduced, 43.5% were initially retained. However, within the first 24 hour period, an

additional 122 were rejected. These beads were the heaviest and several animals made

visible efforts to expel them without success until they were placed on their side. There

was a clear overall increase in attempts to expel particles as the bead size increased.

0.5 mm bead retention.

Specimen #1, a relatively large subject (80 mm in height with an 8 mm incurrent siphon), rejected the first twenty 0.5 mm beads in a row, but subsequently started to

accept them sporadically. Specimen #7, average sized (65 mm in height with a 5 mm

incurrent siphon), vigorously rejected the introduction of any beads, retaining only two,

having an even more forceful reaction to the 1.0 mm beads (see below). Specimen #2,

one of the largest specimens (90 mm in height with a 10 mm incurrent siphon), retained

49 out of 50 particles for the first two hours while Specimen #18, one of the smallest, (45

mm in height with a 3 mm incurrent siphon) also retained 49 out of 50.

In total, 900 0.5 mm beads were introduced to 18 test specimens with 653 beads

(72.6%) being retained for the first two hours. After this time, the animals were placed

back in a common ten gallon aquarium. Continual expulsion of beads occurred for the

next several days when the animals were allowed to lie on their sides. Many ejections

included numerous beads clustered in mucus. This may indicate that the weight of the

beads became too much to reject once they built up inside the animals, but were ejected

after placing the animals on their sides. Due to the small size of the 0.5 mm beads, a final

count of expelled beads could not be taken. Some were clearly visible on the glass

bottom but those entangled in mucus and fecal pellets or caught along the folds of a test

animal could not be accurately accounted for. It is estimated that a minimum of half of

the 0.5 mm beads originally believed to be retained were later rejected. As some fecal

pellets produced within the next days had the presence of a few 0.5 mm glass beads,

digestion was indicated but appeared to be minimal.

Table 5.2 Size of test specimens and results of 0.5 mm bead expulsion within the first two hours after bead introduction. Height, Percent rounded 0.5 mm Oral to beads Siphon closest initially Specimen in mm 5 mm expelled 1 8 80 62 2 10 90 2 3 4 90 6 4 10 60 6 5 4 60 0 6 7 65 58 7 5 65 96 8 6 70 0 9 4 60 60 10 4 30 30 11 8 80 0 12 4 45 20 13 6 90 4 14 6 90 6 15 4 70 6 16 5 80 54 17 4 35 82 18 3 45 2

1.0 mm bead retention.

Acceptance of 1.0 mm beads was equally erratic as the 0.5 mm beads, ranging from 0 – 100% in the first two hours. As was the case with the 0.5mm beads, organism

size was not a factor. The smallest specimen (#10) only expelled 10 out of 50 of the total

beads introduced, yet specimen #17 of similar size initially expelled 40 out of 50 (Table

5.3). Specimen #7 had been extremely sensitive to the 0.5 mm beads and continued this sensitivity with the 1.0 mm. After receiving only eight beads, it erupted in a violent expulsion of three beads along with a small portion of the branchial basket lining. The incurrent siphon remained tightly shut, trapping this section of the branchial basket for the remainder of the experiment. The atrial siphon continued to open and close normally during this time. As a result, this specimen did not receive its balance of 1.0 mm beads.

A total of 858 1.0 mm beads were delivered and resulted in 424 (49.4%) being retained beyond two hours. As was the case with the 0.5 mm beads, many continued to be rejected over the next few days when the animals were laying on their sides. In addition to those beads that were orally rejected, several were found as part of fecal pellets. It is not possible to determine how many were “digested” versus rejected with the experimental setup.

Table 5.3 Size of test specimens and results of 1.0 mm bead expulsion within the first two hours after bead introduction.

Height, Percent rounded 1.0 mm Oral to beads Siphon closest initially Specimen in mm 5 mm expelled 1 8 80 16 2 10 90 58 3 4 90 6 4 10 60 0 5 4 60 78 6 7 65 76 7 5 65 62.5* 8 6 70 74 9 4 60 100 10 4 30 20 11 8 80 98 12 4 45 6 13 6 90 18 14 6 90 76 15 4 70 0 16 5 80 86 17 4 35 80 18 3 45 68 * Specimen #7only received eight beads before closing the incurrent siphon for the rest of the day making additional bead introductions impossible.

2.0 mm bead retention.

Specimens only received 20 beads of 2.0 mm size, due to the weight of the larger

beads. With four specimens, beads were rejected and expelled out the top of the incurrent siphons but did not clear the animal and instead, landed along the rim and were later ingested again. In such a case, the second introduction of the bead was added to the total count of beads introduced to the individual, (i.e. animal received 21 or more beads instead of 20). As was the case with the 0.5 mm and 1.0 mm beads, specimen size was not a factor in the retention of the beads. Specimens 3, 10, 11, and 17 expelled all 2.0 mm beads. Specimens 8 and 15 accepted only one bead, while specimens 4, 13, 14, and

16 accepted all 2.0 mm beads introduced to them. Specimens 5 and 18 were not tested

due to the unhealthy appearance of the animals (Table 5.4).

In total, 336 2.0 mm beads were introduced with 146 (43.5%) being retained for the first two hours. Within 24 hours after the experiment, an additional 122 beads were orally rejected resulting in only 24 out of 336 beads being retained. Within two weeks, all 24 missing beads were orally expelled by the tunicates.

Table 5.4 Size of test specimen and results of 2.0 mm bead expulsion within the first two hours after bead introduction.

Height, Percent rounded 2.0 mm Oral to beads Siphon closest initially Specimen in mm 5 mm expelled 1 8 8 75 2 10 9 20 3 4 9 100 4 10 6 0 5 4 6 * 6 7 6.5 60 7 5 6.5 80 8 6 7 96 9 4 6 70 10 4 3 100 11 8 8 100 12 4 4.5 20 13 6 9 0 14 6 9 0 15 4 7 95 16 5 8 0 17 4 3.5 100 18 3 4.5 * * denotes specimens not tested.

CONCLUSION

Based on the results of this study, many of the pelagic Sciaenidae eggs and larvae

could be vulnerable to filtration by the tunicate S. plicata. However, results of this

experiment merely determined the rejection of inorganic glass beads as experimentation

with actual fish eggs was not possible. If these beads had been organic, they may not have been rejected. As a result, no definitive conclusion can be drawn on the true outcome of fish eggs and larvae by the filtration by S. plicata. Until such experimentation can be conducted in vivo, it is suggested that bioremedial efforts with high numbers of S. plicata be approached with caution, especially during spawning times of fishes and invertebrates and in those areas used as spawning grounds.

CHAPTER SIX: ARTIFICIAL SUBSTRATE SELECTION BY PHALLUSIA

NIGRA

Abstract

This study investigated larval settlement preferences of the tunicate Phallusia nigra.

Adult specimens were gathered from the Vista del Marina, Bocas del Toro, Panama, for

in vitro fertilization. Fertilized eggs were allowed to incubate 10 hours before being added to five 10 gallon aquaria containing four artificial substrates: white polyvinyl chloride (PVC), black polyethylene (PE), light gray high-density polyethylene with a structurally foamed surface (HDPE), and dark brown/black marine plywood. Substrates naturally floated at the surface except PVC, which was suspended by wire from the top to simulate surface flotation. Each substrate was placed in the aquaria for five days prior to experimentation to allow formation of a biofilm. Aquaria were supplied with flow- through filtered marine seawater, were aerated by airstone, and were under constant lighting conditions. Seawater flow was turned off upon addition of fertilized eggs and remained off for 48 hours, until after larval hatching, settlement, and metamorphosis began. Substrates were demarcated, each containing a randomly chosen 3,025 mm2 region in which larvae were to be counted under dissection microscopy. Four days after the addition of embryos to the aquaria, a total of 3,097 larvae had settled within demarcated regions of the test substrates. Marine plywood was the preferred substrate with an overall average settlement of 53.7%, followed by 30.9% on PE, 13.0% on HDPE,

and 2.4% on PVC, indicating a strong preference to the two darker substrates.

INTRODUCTION

All previous investigations during the course of this dissertation study have utilized the solitary tunicate Styela plicata. During the summer of 2008, a large tropical storm (TS Fay) moved slowly across Florida depositing large amounts of rain into Lake

Okeechobee. The lake is surrounded by the 70 year old Herbert Hoover Dike, a 140 mile long, 30 foot tall mound that protects surrounding communities from flooding. During the week of the storm, areas in the northern watershed sent storm water into the lake at a rate of 13 million gallons per minute. Water levels rose two feet according to the South

Florida Water Management District (SFWMD, 2009) causing necessary discharges of freshwater into the Caloosahatchee River Estuary, to the west, and the St. Lucie River

Estuary, to the east (SFWMD, 2008a). Data recorded by the South Florida Water

Management District (2008b) revealed plummeting salinities. The westernmost salinity monitoring station is Shell Point, where the Caloosahatchee River meets the Gulf of

Mexico. Prior to TS Fay, (August 19, 2008), the surface salinity at this station averaged

22.2 ppt. The average salinity in this area the following week was 5.3 ppt. On the

Atlantic side, the easternmost monitoring station of the St. Lucie River is the A1A Bridge where it meets the Indian River Lagoon (IRL). Testing at this site during the same two weeks revealed the average salinity dropped from 19.7 ppt to 3.8 ppt. Open water in the

IRL was recorded at 31.9 ppt decreasing to 15.3 ppt (SFWMD, 2008b). Salinity conditions on both coasts remained poor for several weeks consequently destroying local populations of S. plicata. For this reason, the final study in this series of investigations was completed with Phallusia nigra Savigny (Ascidiacea, Ascidiidae), another solitary

ascidian found in South Florida waters. Its local Florida population was also affected but

specimens were available in Bocas del Toro, Panama at the Smithsonian Tropical

Research Institute, where S. plicata is not found.

Phallusia nigra has a wide, warm water distribution including the Atlantic Ocean,

Indian Ocean, Red Sea, Mediterranean Sea, and the Indo-West Pacific (da Rocha et al.,

1999; Millar, 1958; Van Name, 1945). It is characterized by a distinctive smooth black

test, void of epibionts, growing upwards to 10 - 12 cm (Goodbody, 1962). The absence

of overgrowth enables easy identification. A member of the family Ascidiidae, P. nigra

was originally classified in the genus Ascidia by Van Name (1945). It was later placed in

the genus Phallusia due to the discovery of accessory openings in the neural gland duct

(Monniot, 1972), which are only visible upon dissection of very large specimens. P.

nigra was chosen as a substitute organism despite a few morphological differences

between it and S. plicata. The major differences are found in the branchial sac and

gonads. S. plicata is a member of the family Styelidae characterized by numerous folds

in the branchial sac. Members of Ascidiidae do not have these folds. The gonads of

Styelidae are located beside the branchial sac whereas in Ascidiidae, they are located

inside the gut loop. Both are of the same approximate size in the estuarine waters of the

IRL under normal salinity circumstances and are attached to similar substrates (personal

observation).

Ascidians begin the first few hours to days of their lives as free swimming larvae,

seeking suitable substrate before metamorphosing into sessile adults. Selection of this

substrate is crucial to post-settlement survival (Hunt & Scheibling, 1997; Young & Chia,

1984) and is usually driven by light and gravity cues (Vazquez & Young, 1998). Once

hatched, ascidian larva are usually photopositive and geonegative, sometimes switching phototaxis just before settlement (Thorson, 1964) on darker surfaces (Goodbody, 1963).

However, this behavior is variable (Svane & Young, 1989; Thorson, 1964).

Larval recruitment and settlement studies have told us much about larval behavior. Young and Braithwaite (1980a) found that Chelyosoma productum settled either very closely to established juveniles or directly on the tunic of conspecific adults, apparently void of the photonegative settlement tendency found in some other tunicates.

Manríquez and Castilla (2007) found that the solitary tunicate Pyura cliensis aggregated according to the presence of conspecifics. Flores and Faulkes (2008) established a preference of texture by Ascidia interrupta showing the existence of mechanoreceptive sensory neurons. Regardless of a species’ settlement cues, larvae availability is usually not the limiting factor of new recruits to benthic communities (Olafsson et al., 1994) but rather food and substrate availability (Fréchette & Lefaivre, 1990). For bioremediation purposes, enhancement of substrate would be necessary to support an increase in ascidian numbers for a positive effect on water clarity.

The current study examines substrate selection by P. nigra on four artificial substrates; white polyvinyl chloride (PVC), black polyethylene (PE), light gray high- density polyethylene with a structurally foamed surface (HDPE), and dark brown to black marine plywood. Construction of artificial substrate habitats for use in bioremediation could theoretically consist of any of these substances, with wood being the least desirable due to its tendency to decompose.

MATERIALS AND METHODS

Biofilm Preparation of Artificial Substrates

Biofilm preparation and subsequent experimentation was completed at the wet lab of the Smithsonian Tropical Research Institution (STRI) in Bocas del Toro, Panama.

Five small pieces each of PVC pipe, PE sheets, HDPE sheets, and marine plywood were placed in ten gallons of bubble bead filtered (International Filter Solutions model BBF

XF20000) flow-through sea water taken from a depth of 20 feet, to allow biofilm formation for five days. Flow-through water was set to a slow trickle. All materials except PVC floated, necessitating daily rotation to allow even formation of the film.

Collection and Acclimation of Adult Specimens

Twenty-four adult Phallusia nigra were collected from the Vista del Marina,

Bocas del Toro, Panama (9° 20’ 09.57” N, 82° 14’ 48.84” W) along dock pilings and the concrete dock slab. Animals were transported in plastic collection bags to the wet lab at

STRI, placed in a prepared tide pool tank containing filtered flow-through sea water, and allowed to acclimate for 48 hours. Specimens were kept under a constant light condition to prevent gamete release (Lambert & Brandt, 1967; West & Lambert, 1976).

Experimental Design

Statistical design.

The response variable “percent settlement on each substrate type” was analyzed by one-way ANOVA (critical alpha=0.05) for comparison of the larval settlement preferences on the four differing substrates. Triplicate counts of larval settlement within

previously established regions of demarcation were averaged for comparison.

Differences were considered significant for probabilities (p< 0.05). The a priori H0 was that “no significant difference exists between the preferences of substrates by the settling larva of Phallusia nigra”.

Substrate setup.

Five replicate 38 L tanks containing ten liters of filtered seawater each contained all four substrate types on which biofilms had formed. A 3025 mm2 surface area was

marked on each substrate. PVC was marked with graphite pencil. The other three

substrates were scored with a sharp knife. PE, HDPE, and marine plywood all floated at

the surface interface. PVC pipe was taped at both ends to make a closed tube and was

suspended by wire from the side of the tank to assure the demarcated area would rest in

the proper position at the surface.

In Vitro Fertilization

Nine of the largest and healthiest adult specimens collected were taken to the

STRI dry lab for in vitro fertilization.

Tunics were removed from each living

specimen exposing the oviduct and sperm

duct along the outside surface of the body

wall, opening just posterior to the anus. The

oviduct lies more toward the surface of the

Fig. 6.1 Fertilized eggs of P. nigra after in animal and is slightly creamier in color than vitro fertilization. Eggs were allowed to develop for ten hours prior to distribution to the sperm duct, which is a brighter white experimental aquaria. 80

color. Animal bodies were dried on paper towels. Eggs were removed from five of the animals by nicking the oviduct with a sharp dissection needle, carefully avoiding the

sperm duct. A dry glass pipette tip was slid

along the oviduct and eggs were extracted into a

250 mL beaker of bead-filtered sea water. The

same procedure was performed for removal of

sperm from four different animals. Sperm were

placed into a separate 50 mL beaker of bead- 100 µm filtered seawater. Several drops of the Fig. 6.2 Ten hour old Phallusia nigra embryo, just prior to hatching. homogenous sperm solution were added to the

beaker of eggs. The sperm/egg mixture was stirred and allowed to sit undisturbed for one hour. After this time, excess sperm were removed from the eggs by washing four times with seawater through a 100 µm opening nylon mesh. After ten hours (Fig. 6.1), just prior to hatching (Fig 6.2), fertilized eggs were distributed to the five aquaria containing experimental substrates. Seawater to the aquaria was turned off at this time, as newly hatched larvae of P. nigra do not feed until approximately 45 hours after settlement which takes three to six hours under normal laboratory conditions (Goodbody, 1963). After 48 hours in the aquaria, seawater was turned on to a minimal trickle. The water temperature was 29° C before and after water flow began. For the next 48 hours the aquaria were undisturbed, allowing growth time for the newly attached larvae.

Counting Larval Recruits

Attached larvae on each substrate

type were manually counted in triplicate

under dissection microscopy (Fig. 6.3).

Larvae appeared as gelatinous globules

with the otoliths and ocelli observable. On

light backgrounds, larvae were stained

with drops of hematoxylin (undiluted) to 250 µm enable easier viewing. In each tank, the Fig. 6.3 Four day old larval recruits on marine total counts of larvae that settled within plywood under microscopy.

the combined demarcated surfaces were considered 100% of total settlement per

replicate.

RESULTS AND DISCUSSION

Overall, 3,097 larvae settled within the demarcated regions of the substrates in all

of the replicates with an average of 53.7% ± 21.3% settling on wood. PE had the second

highest average settlement percentage at 30.9% ± 11.7%, followed by 13.0% ± 18.6% on

HDPE, and 2.4% ± 1.4% on PVC (Fig. 6.4). This pattern was unmistakable in replicate

numbers 1, 2, 4, and 5, which resulted in an average of 61.9% preferring marine

plywood, 31.3% with PE, 4.5% with HDPE, and 2.3% preferring PVC. The third

replicate had surprising results. Only 21.1% preferred marine plywood and 29.6%

preferred PE. The highest preference was shown to HDPE in this tank at 46.7%. PVC

received 2.6% preference (Fig. 6.4 C). Upon careful reexamination of substrates taken

from this replicate, an abnormality was noted in the surface of the HDPE plate. The edge

of the demarcated region was along a cut edge of the plate, which was not completely smooth. It had a slight lip curling under, clearly visible under the dissection microscope.

Most of the larval recruits on this plate were tucked up under this lip and were missed

upon first counting. Subsequent counts revealed the densely packed conglomerate. As

explained previously, newly hatched larvae are photopositive, allowing the surface of the

water to be easily found. This is not for food availability, as might be expected, as

tunicate larvae are lecithotrophic and do not feed until after completing metamorphosis

(Glaser & Anslow, 1949) but may be a dispersal tactic (Thorson, 1964). Conversely, P.

nigra has been shown to reverse phototaxis from photopositive to photonegative just

before settlement (Goodbody, 1963; Thorson, 1964). This may be a possible explanation

for the apparent hiding of the recruits under the lip of the HDPE plate in the well-lit

laboratory. However, this does not explain the absence of normal numbers of recruits on

marine plywood, which one might expect to be second highest in settlement. It is

possible that plywood inconsistencies resulting from the process of gluing and lamination

could offer an explanation. This deviation caused the data to be abnormal and non-

homogenous even after transformations and ranking, making statistical analysis

impossible. When the replicate was omitted, and data transformed by taking the log of

the proportion of settlement on each substrate type within a tank, residuals were normal

(p< 0.4158) and variances were homogenous (p<=0.3509). ANOVA indicated that

substrates were significantly different (F 3,16 = 33.64, p<0.0001).

Marine plywood had the largest sample variance of 454.5, with a range of

larval recruitment percentage of 21.1 to 75.2%, and a mean of 53.7%. Sample size power

PVC PVC HDPE 1.2% HDPE 2.1% 5.5% 6.9% PE PE 18.1% 21.2%

Wood Wood 75.2% 69.9%

A B

PVC PVC HDPE 1.2% 2.6% 1.2%

PE 29.6% PE HDPE 43.6% 46.7%

Wood 54.0%

Wood 21.1% C D

HDPE PVC 4.7% 4.7%

PE 42.3% Wood 48.3%

Fig 6.4 A – E Percentages of larval settlement on different substrate types in individual replicate tanks. The overall highest settlement was found on marine plywood followed by polyethylene (PE) in all but tank 3. The lowest settlement was on polyvinyl chloride (PVC) in all five replicates. A) Tank 1 results. B) Tank 2 results. C) Tank 3 results - High Density Polyethylene (HDPE) plates had the highest settlement, unlike all other treatments, followed by PE. Marine plywood had the third highest settlement. D) Tank 4 results. E) Tank 5 results. analysis revealed this experiment would require 72 replicates to get within 10% of the population mean:

n=72

454.5(1.994)2/(5)2 = 72.3

This number of replicates was not possible in the current study but it is recommended that a follow-up study be performed with as many replicates as possible and with careful attention given to the substrates. Each surface should be identical in size and it is suggested that the edges be smoothed on each. The type of wood selected should be uniform in color and texture, as much as can be controlled. Lighting should be from several angles to prevent shadows. Also, more time for larval growth before counting would be desirable.

CONCLUSION

Of the four types of artificial substrate examined, P. nigra preferred marine plywood, followed by PE, HDPE, and PVC. This is not an unexpected result as tunicates in their natural habitat can be seen inhabiting mangrove proproots, wooden dock pilings, boat hulls, and other wooden substrates. Marine plywood and PE were also, the darkest substrates offered. PVC was the least settled upon in all replicates, although P. nigra and other species of tunicates can be found attached to PVC pipes in marinas (personal observation). PVC may not be a favored substrate, but the pressure to settle within a short time is evidently stronger than the cues for substrate selection itself.

Construction of artificial substrate habitats for use in bioremediation should consist of lightweight sturdy materials. Although marine plywood was the preferred substrate by the tunicates, it is highly degradable and would be the least reliable as a bioremediation platform. Polyethylene, a more steadfast substrate, is lightweight, dark in color, and was acceptable to the tunicate larvae. Results of this study indicate that it may be the best material of those tested, for construction of tunicate habitats for use in bioremediation.

CHAPTER SEVEN: PROJECT CONCLUSIONS AND FUTURE DIRECTIVES

Environmental conditions of the Indian River Lagoon complex have dramatically

changed over the past century due to a variety of anthropogenic disturbances. Seawalls,

bridges, marinas, and other man-made structures have partially replaced the natural environment of mangroves and salt marshes. Suspended particulate matter has increased

as the substrates for filter feeding organisms decreased (Dame et al., 2002). Pockets of

poorly circulated water with elevated water temperatures frequently experience harmful

algal blooms, which lead to fish kills, beach closures, oxygen depletion (Peterson et al.,

2006; Worm et al., 2006) and elevated bacterial counts often consisting of fecal coliforms

(Florida Department of Health, 2000).

Successful methods of restoring the health of the estuarine environment are necessary as coastal communities become more populated. Studying the methods that have historically worked to maintain the environment reveals the vitally important role of filter-feeding organisms. An accidental observation of the filtering effects of turbidity by the tunicate Styela plicata led to a hypothesis that they may be a good target species for the bioremediation of bacterial and algal cells from the estuarine environment.

Utilization of tunicates for bioremediation is virtually an unexplored area and should be thoroughly investigated.

As this species is considered highly fouling, all studies were performed in a

laboratory setting. Filtration rates were established first, to determine whether further

study was warranted. Previous studies on the filtration rate of S. plicata were not

intentionally focused on their use in bioremediation, but instead, were for the purpose of

broadening our understanding of the species’ physiology and morphology. A filtration

rate of S. plicata was first reported by Fiala-Médioni (1978b) as 8.76 mL hr-1 per gram of

dry organ weight or 4.1 L hr-1 per animal. Her results were based on three replicates

using tunicates of the same approximate size with a constant food concentration over a

12-hour period. Filtration rates for this species found in the present study (3.2 L hr-1 per

animal) were 33% lower, but were still substantial enough to warrant a closer look at the

tunicate for potential bioremedial use. As filtration rates were found to vary extensively

from specimen to specimen, it is suggested that extrapolation for a population of S.

plicata over time is a more appropriate method of reporting, especially for the purpose of

bioremedial modeling. Based on the findings of this study, 200 average-size S. plicata

could filter as much as 660 L hr-1 or 15,840 L day-1 of microalgae at a density of106 cells mL-1 with 100% efficiency.

The tunicates examined in the current study responded to the two concentrations

of microalgae (105 and 106 cells mL-1) in a similar manner, yet they responded very

differently to the same two concentrations of bacteria. Filtration rates based on E. coli

removal were 4.7 L hr-1 in the 105 cells mL-1 treatment, but decreased by 50% to 2.3 L hr-1 in the 106 cells mL-1 treatment. This difference in response has a few possible

explanations. It could mean that the tunicates were able to differentiate between the

particles and are not non-discriminate filter-feeders as originally suggested by Jørgensen

(1966b). There could potentially be receptors within the branchial basket that remain

undiscovered, which cause the tunicate to cease or slow filtration to something that is potentially hazardous to the animal. Although possible, this is not likely the reason. It may mean that the nutritional needs of the animal were met more quickly with bacteria than with microalgae. Additionally, it also may mean that the lipopolysaccharides

(endotoxins) (Alexander & Rietschel, 2007) in the outer membrane of the gram negative

E. coli at the higher concentration interfered with the tunicate’s filtration efforts.

Repeating the current study with a non-toxic gram positive bacterium could test this possibility.

After the establishment of filtration rates for bacteria, the question was posed as to whether the bacteria had been assimilated by the tunicate and permanently removed from the water column or whether it was merely repackaged in a mucous-bound fecal pellet available for resuspension. If the animal was to be used for bioremedial purposes, the bacteria had to be permanently removed. Difficulties in culturing marine bacteria was first documented by (Butkevich & Butkevich, 1936; Butkevich, 1932, 1938;

Radsimovsky, 1930) and continues to be the topic of many studies (Amann et al., 1995;

Brock, 1987; Bull & Hardman, 1991; Hugenholtz et al., 1998; Jannasch & Jones, 1959;

Roszak & Colwell, 1987; Staley & Konopka, 1985; Yasumoto-Hirose et al., 2006). This difficulty necessitated the analysis of post-digested bacteria by ATP assay. ATP production was measured before and after the resuspension of fecal pellets. As only living organisms produce ATP (Holm-Hansen & Booth, 1966), this method of analysis has been shown to be an effective way to measure indirectly, viable bacteria (Bautista et al., 1997; Deininger & Lee, 2001; 2005; Holm-Hansen & Booth, 1966; Kimmich et al.,

1975; Trudil et al., 2000). No increases in ATP levels after fecal pellet production were

found in any treatment (n=9) over the course of a five-hour experiment. This indicated

that bacteria were assimilated during digestion and were effectively removed from the

water column, further supporting its potential use as a bioremediator.

Additional studies on the filtration rates of S. plicata should be conducted using a

benign marine dinoflagellate compared to the red tide dinoflagellate Karenia brevis to

determine the effects of the brevetoxin on the tunicate. If effects were minimal, studies

using S. plicata for treatment of red tide outbreaks are warranted.

With high filtration rates that have been suggested as non-discriminate

(Jørgensen, 1966b), comes the potential accidental filtration of particles essential to the

estuary. The commercially and recreationally important members of the ichthyofaunal

family Sciaenidae (the drums) lay pelagic eggs in the IRL, making them vulnerable to

filtration by S. plicata. The smallest of these eggs have diameters of 0.6 mm and are

produced by Cynoscion nebulosus (spotted seatrout). To simulate the potential negative

effects on fish eggs by the filtration of S. plicata, particle ingestion and retention were

examined using glass beads of three sizes, 0.5 mm, 1.0 mm, and 2.0 mm. Of the 0.5 mm

beads, 72.5% were retained for at least two hours. Of the 1.0 mm beads which more

closely resemble the size of some larvae found in the estuary, 49.4% were retained for the

first two hours. Many beads were orally expelled in subsequent hours; however, if these

had been fish eggs or larvae instead of glass beads, their survival during this time is

questionable. Conversely, areas needing bioremediation may already be contaminated

with either bacteria, algae, or both and existing fish eggs may be non-viable or absent due

to the pollution (Austin, 1999). The benefits of water quality improvements by the

placement of tunicates in these areas may out-weigh the risk of fish egg removal.

Incurrent siphon diameter and animal height exhibited no significant correlation with particle ingestion which appeared erratic as some specimens were more sensitive than others, indicating other stochastic factors are involved in particle size selection. As glass beads are inorganic and are heavier than fish eggs, it is suggested that this experiment be repeated with actual fish eggs. If results are similar to those in this study, placement of populations of S. plicata should be carefully analyzed prior to introduction in areas of estuarine fish spawning.

Although the potential filtration benefits of the tunicate indicate a substantial positive impact on the environment, the area where S. plicata can be utilized for bioremedial purposes is limited. Long-term placement is not recommended in areas where salinity frequently falls below 24 ppt for extended periods of time. However, if average salinities are higher, areas prone to frequent bacterial or algal blooms, such as

Dubois Park near the Jupiter Inlet, may benefit from the permanent placement of S. plicata populations to decrease the frequency of the outbreaks.

Sudden shock exposure to 20 ppt from 32 ppt resulted in 60% survival of tunicates up to four days. Although this is not an acceptable salinity for long-term exposure, short-term placement of tunicates in extreme algal or bacterial blooms may still have enough positive impact on water quality to improve conditions. Additionally, tunicates in the current study were acclimated at 32 ppt, prior to experimentation. For future aquaculture purposes, the maintenance salinity may be kept lower to prevent such extreme shocks from higher to lower salinities. Suggested maintenance salinities would be at or near 28 ppt as thriving populations are found in situ at this salinity (personal observation).

Aquaculture of S. plicata would require additional substrate on which newly

hatched larva could settle. This substrate would need to be light-weight for easy

movement throughout the estuary, especially in response to algal or bacterial blooms.

Four experimental substrates were investigated for future aquaculture purposes: marine

plywood, polyethylene, high-density polyethylene, and polyvinyl chloride. Larvae

settled on marine plywood in preference to other substrates, although all substrates had

some settlement.

The low overall selection preference of PVC was not expected. Specimens for

this study were collected from the Vistas del Marino, Bocas del Toros, Panama. In this area, adult specimens of P. nigra were observed attached to PVC on a frequent basis.

Pressure to find a suitable substrate within a very limited amount of time, is likely more

important than the animal’s substrate preference cues. This may indicate that the type of

substrate offered to new larva is less important than originally thought. As polyethylene

is not prone to decay as in the case of wood, it may be a suitable substrate to offer for

aquaculture purposes and for use in bioremediation.

As the results of the current studies indicate potential positive impacts on the environment by the bioremediation of S. plicata, field studies are now recommended.

These studies should be conducted within contained areas, utilizing Secchi disc and water

quality measurements taken before and after bioremediation. Within these controlled

areas, the combined effects of food type availability, water turbidity, water current speeds

(hydrology), and proximity to other organisms is recommended for further research. Any

in situ studies outside of containment should be conducted using sterile tunicates (i.e. triploid animals). S. plicata is considered a highly fouling organism; although, as with

other benthic invertebrates, it is substrate limited (Dayton, 1971; Mook, 1983;

Stachowicz et al., 2002; Svane & Petersen, 2001). Lack of substrate controls the populations in the natural environment. Introduction of increased populations, even on a temporary basis, could cause an imbalance in the other benthic invertebrates present in the estuary. Caution should be taken to prevent increases in S. plicata abundance, especially during in situ investigations. Until the environmental effects of increasing populations are known, the use of triploid (sterile animals) for experimentation should be explored.

For future investigations, a field study should involve the use of two similar marinas of approximately equal sizes, water circulation patterns, water depths, and levels of contaminants. One would be equipped with bioremedial platforms containing approximately 200 tunicates each, while the other would have no tunicates other than those naturally present. The number of introduced platforms would be calculated from the size of the water body and results of the hydrology and mixed particle concentrations experiments.

If using S. plicata for the purpose of bioremediation continues to appear promising, its use may help solve the problems of bacterial and microalgal blooms in estuarine waters. Historically, nature has always made the best technology. Investigating the bioremedial use of S. plicata and other marine invertebrates should be extensively considered.

LITERATURE CITED

Adams, D. H., & Tremain, D. M. (2000). Association of large juvenile red drum,

Sciaenops ocellatus, with an estuarine creek on the Atlantic coast of Florida.

Environ. Biol. Fishes, 58(2), 183-194.

Akhtar, N., Saeed, A., & Iqbal, M. (2003). Chlorella sorokiniana immobilized on the

biomatrix of vegetable sponge of Luffa cylindrica: a new system to remove

cadmium from contaminated aqueous medium. Bioresour. Technol., 88(2), 163-

165.

Alexander, C., & Rietschel, E. T. (2007). Bacterial lipopolysaccharides and innate

immunity. J. Endotoxin Res., 7(3), 167-202.

Algers, J., Suzuki, R., Ohdaira, T., & Maurer, F. H. J. (2004). Characterization of free

volume and density gradients of polystyrene surfaces by low-energy positron

lifetime measurements. Polymer 45(0), 4533-4539.

Alkan, U., Elliott, D. J., & Evison, L. M. (1995). Survival of enteric bacteria in relation to

simulated solar radiation and other environmental factors in marine waters. Water

Res., 29(9), 2071-2080.

Allison, P. D. (2008). Survival analysis using SAS®: A practical guide. Cary, NC: SAS

Institute Inc.

Amann, R., Ludwig, W., & Schleifer, K. (1995). Phylogenetic identification and in situ

detection of individual microbial cells without cultivation. Microbiol. Rev., 59(1),

143-169.

Armsworthy, S. L., MacDonald, B. A., & Ward, J. E. (2001). Feeding activity, absorption

efficiency and suspension feeding processes in the ascidian, Halocynthia

pyriformis (Stolidobranchia: Ascidiacea): responses to variations in diet quantity

and quality. J. Exp. Mar. Biol. Ecol., 260(1), 41-69.

Arnold, W. R., Cotsifas, J. S., & Corneillie, K. M. (2006). Validation and update of a

model used to predict copper toxicity to the marine bivalve Mytilus sp. Environ.

Toxicol., 21(1), 65-70

Aslan-Yilmaz, A., Okus, E., & Ovez, S. (2004). Bacteriological indicators of

anthropogenic impact prior to and during the recovery of water quality in an

extremely polluted estuary, Golden Horn, Turkey. Mar. Pollut. Bull., 49(11-12),

951-958.

Asmus, R. M., & Asmus, H. (1991). Mussel beds: limiting or promoting phytoplankton?

J. Exp. Mar. Biol. Ecol., 148(2), 215-232.

Austin, B. (1999). The effects of pollution on fish health. J. Appl. Microbiol. Symposium

Supplement, 85(0), 234S-242S.

Bautista, D. A., Kozub, G., Jericho, K. W. F., & Griffiths, M. W. (1997). Evaluation of

Adenosine Triphosphate (ATP) bioluminescence for estimating bacteria on

surfaces of beef carcasses. J. Rapid Methods Autom. Microbiol., 5(1), 37-45.

Beck, M. W., Heck, K. L., Jr., Able, K. W., Childers, D. L., Eggleston, D. B., Gillanders,

B. M., Halpern, B., Hays, C. G., Hoshino, K., Minello, T. J., Orth, R. J., Sheridan,

P. F., & Weinstein, M. P. (2001). The identification, conservation, and

management of estuarine and marine nurseries for fish and invertebrates.

BioScience, 51(8), 633-641.

Bingham, B. L. (1992). Life histories in an epifaunal community: Coupling of adult and

larval processes. Ecology, 73(6), 2245-2259.

BioThema (2007). Microbial ATP kit (pp. 2). Handen, Sweden: BioThema.

Bone, Q., Carre, C., & Chang, P. (2003). Tunicate feeding filters. J. Mar. Biol. Assoc.

UK, 83(5), 907-919.

Brock, T. D. (1987). The study of microorganisms in situ: Progress and problems. In M.

Fletcher, T. R. G. Gray & J. G. Jones (Eds.), Ecology of microbial communities

(pp. 1-17). Cambridge: Society for General Microbiology.

Bull, A. T., & Hardman, D. J. (1991). Microbial diversity. Curr. Opin. Biotechnol. , 2(3),

421-428.

Butkevich, N. V., & Butkevich, V. S. (1936). Multiplication of sea bacteria depending on

the composition of the medium and on temperature. Microbiology (Moscow, Russ.

Fed.), 5(0), 322-343.

Butkevich, V. S. (1932). Zur Methodik der bakteriologischen Meeresuntersuchungen und

einige Angaben tiber die Verteilung der Bakterien im Wasser und in den Boden

des Barents Meeres. Trans. Oceanogr. Inst. Moscow, 2(0), 5-39

Butkevich, V. S. (1938). On the bacterial populations of the Caspian and Azov Seas.

Microbiology (Moscow, Russ. Fed.), 7(0), 1005-1021.

Carballo, J. L., & Naranjo, S. (2002). Environmental assessment of a large industrial

marine complex based on a community of benthic filter-feeders. Mar. Pollut.

Bull., 44(7), 605-610.

Cardoso, P. G., Pardal, M. A., Lillebø, A. I., Ferreira, S. M., Raffaelli, D., & Marques, J.

C. (2004). Dynamic changes in seagrass assemblages under eutrophication and

implications for recovery. J. Exp. Mar. Biol. Ecol., 302(2), 233-248.

Carlisle, D. B. (1966). The ciliary current of Phallusia [Ascidiacea] and the squirting of

sea squirts. J. Mar. Biol. Assoc. UK, 46(1), 125-127.

Center for Quantitative Fisheries Ecology (2010). Atlantic croaker. Retrieved Jan. 26,

2010, from

http://www.odu.edu/sci/cqfe/Research/Chesapeake%20Bay/Atlantic%20croaker/

Atlantic_croaker.htm

Cima, F., Ballarin, L., Bressa, G., Martinucci, G., & Burighel, P. (1996). Toxicity of

organotin compounds on embryos of a marine invertebrate (Styela plicata;

Tunicata). Ecotoxicol. Environ. Saf., 35(2), 174-182.

Connaughton, M. A., & Taylor, M. H. (1996). Drumming, courtship, and spawning

behavior in captive weakfish, Cynoscion regalis. Copeia, 1996(1), 195-199.

Connell, S. D., & Glasby, T. M. (1999). Do urban structures influence local abundance

and diversity of subtidal epibiota? A case study from Sydney Harbour, Australia.

Mar. Environ. Res., 47(4), 373-387.

Costanza, R., & Mageau, M. (1999). What is a healthy ecosystem? Aquat. Ecol., 33(1),

105-115.

Coughlan, J. (1969). The estimation of filtering rate from the clearance of suspensions.

Mar. Biol., 2(4), 356-358.

Crabtree, R. E. (1995). Relationship between lunar phase and spawning activity of

tarpon, Megalops atlanticus, with notes on the distribution of larvae. Bull. Mar.

Sci., 56, 895-899.

Crabtree, R. E., Cyr, E. C., Bishop, R. E., Falkenstein, L. M., & Dean, J. M. (1992). Age

and growth of tarpon, Megalops atlanticus larvae in the eastern Gulf of Mexico,

with notes on relative abundance and probable spawning areas. Environ. BioI.

Fishes, 35(4), 361-370.

da Rocha, R. M., & Kremer, L. P. (2005). Introduced ascidians in Paranaguá Bay, Paraná,

southern Brazil. Rev. Bras. Zool., 22(4), 1170-1184. doi:10.1590/S0101-

81752005000400052

da Rocha, R. M., Lotufo, T. M. d. C., & Rodrigues, S. d. A. (1999). The biology of

Phallusia nigra Savigny, 1816 (Tunicata: Ascidiacea) in southern Brazil: spatial

distribution and reproductive cycle. Bull. Mar. Sci., 64(1), 77-87.

Dame, R., Bushek, D., Allen, D., Lewitus, A., Edwards, D., Koepfler, E., & Gregory, L.

(2002). Ecosystem response to bivalve density reduction: management

implications. Aquat. Ecol., 36(1), 51-65.

Dame, R. F., Zingmark, R. G., & Haskin, E. (1984). Oyster reefs as processors of

estuarine materials. J. Exp. Mar. Biol. Ecol., 83(3), 239-247.

Damiens, G., Mouneyrac, C., Quiniou, F., His, E., Gnassia-Barelli, M., & Romeo, M.

(2006). Metal bioaccumulation and metallothionein concentrations in larvae of

Crassostrea gigas. Environ. Pollut., 140(3), 492-499.

Daniel, L. B., & Graves, J. E. (1994). Morphometric and genetic identification of eggs of

spring-spawning sciaenids in lower Chesapeake Bay. Fish. Bull., 92(2), 254-261.

Dayton, P. K. (1971). Competition, disturbance, and community organization: The

provision and subsequent utilization of space in a rocky intertidal community.

Ecol. Monogr., 41(4), 351-389.

Defossez, J. M., & Hawkins, A. J. S. (1997). Selective feeding in shellfish: size-

dependent rejection of large particles within pseudofaeces from Mytilus edulis,

Ruditapes philippinarum and Tapes decussatus. Mar. Biol., 129(1), 139-147.

Deininger, R. A., & Lee, J. (2001). Rapid determination of bacteria in drinking water

using an ATP assay. Field Anal. Chem. Technol., 5(4), 185-189.

Deininger, R. A., & Lee, J. (2005). Rapid detection of bacteria in drinking water. In A.

Omelchenko, A. A. Pivovarov & W. J. Swindall (Eds.), Modern tools and

methods of water treatment for improving living standards (Vol. 48, pp. 71-78).

Dnepropetrovsk, Ukraine: Springer Netherlands.

Draughon, L., Scarpa, J., Keating, P., & Hartmann, J. (2008). Potential estuarine water

quality improvement via marine invertebrate bioremediation. In M. Theophanides

& T. Theophanides (Eds.), Environmental awareness and management (pp. 97-

112). Athens: ATINER.

Fiala-Médioni, A. (1974). Ethologie alimentaire d'invertébres benthiques filtreurs

(ascidies). II. Variations des taux de filtration et de digestion en fonction de

l'espèce. Mar. Biol., 28(3), 199-206.

Fiala-Médioni, A. (1978a). Filter-feeding ethology of benthic invertebrates (ascidians). 3.

Recording of water current in situ-rate and rhythm of pumping. Mar. Biol., 45(2),

185-190.

Fiala-Médioni, A. (1978b). Filter-feeding ethology of benthic invertebrates (ascidians). 4.

Pumping rate, filtration rate, filtration efficiency. Mar. Biol., 48(3), 243-249.

Fishbase (2009). Retrieved Jan. 22, 2010, Froese, R. and D. Pauly: Retreived from

http://www.fishbase.org, version (11/2009)

Fisher, T. R. (1976). Oxygen uptake of the solitary tunicate Styela plicata. Biol. Bull.,

151(2), 297-305.

Fletcher, S. W., & Fletcher, W. W. (1995). Factors affecting changes in seagrass

distribution and diversity patterns in the Indian River Lagoon complex between

1940 and 1992. Bull. Mar. Sci., 57(1), 49-58.

Flood, P. R., & Fiala-Médioni, A. (1981). Ultrastructure and histochemistry of the food

trapping mucous film in benthic filter feeders (Ascidians). Acta Zool.-Stockholm,

62(1), 53-65.

Flores, A. R., & Faulkes, Z. (2008). Texture preferences of ascidian tadpole larvae during

settlement. Mar. Freshwater Behav. Physiol., 41(3), 155-159.

Florida Department of Health (2000). Florida Healthy Beaches Program. Retrieved

Mar. 2, 2006, from http://esetappsdoh.doh.state.fl.us/irm00beachwater

Florida Oceanographic Society (2004). St. Lucie River Estuary Water Quality Data,

Retrieved from http://www.floridaoceanographic.org/water.htm

Fréchette, M., & Lefaivre, D. (1990). Discriminating between food and space limitation

in benthic suspension feeders using self-thinning relationships. Mar. Ecol. Prog.

Ser., 65(0), 15-23.

Fuller, P. (2006, April 12, 2005). Styela plicata. USGS Nonindigenous Aquatic Species

Database Retrieved October 4, 2006, from

http://nas.er.usgs.gov/queries/FactSheet.asp?speciesID=1293

Geiger, S. P., Arnold, W. S., & Frisher, M. (2004). Abundance of hard clam larvae in the

Indian River Lagoon. J. Shellfish Res., 23(1), 292-293.

Georgia Department of Natural Resources Coastal Resources Division (2003).

Management Plan: Southern kingfish (Whiting). Georgia Department of Natural

Resources: Retreived from

http://crd.dnr.state.ga.us/assets/documents/marfish/whitingfmp07.pdf

Giancoli, D. C. (1998). Physics, principles with applications (5th ed.). Upper Saddle

River, NJ: Prentice Hall.

Giangrande, A., Cavallo, A., Licciano, J., Mola, E., Pierre, C., & Trianni, L. (2005).

Utilization of the filter feeder polychaete Sabella Spallanzanii Gmelin

(Sabellidae) as bioremediator in aquaculture. Aquacult. Int., 13(1-2), 129-136.

Gilbert, C. (1986). Species profiles: life histories and environmental requirements of

coastal fishes and invertebrates (south Florida): Florida pompano. U.S. Fish

and Wildlife Service Biological Report 82(11.42): U.S. Army Corps of Engineers

TR-EL-82-4.

Gilhen, J., Gruchy, C. G., & McAllister, D. E. (1976). The sheepshead, Archosargus

probatocephalus, and the feather blenny, Hypsoblennius hentzi, two additions to

the Canadianantic ichthyofauna. Can. Field Nat., 90, 42-46.

Gilmore, G. R. (1995). Environmental and biogeographic factors influencing

ichthyofaunal diversity: Indian River Lagoon. Bull. Mar. Sci., 57(1), 153-170.

Gilmore, R. G., Donohoe, C. J., & Cooke, D. W. (1983). Observations on the distribution

and biology of east-central Florida populations of the common snook

Centropomus undecimalis (Bloch). Fla. Sci., 46, 313-336.

Glaser, O., & Anslow, G. A. (1949). Copper and ascidian metamorphosis. J. Exp. Zool.,

111(1), 117-139.

100

Goldberg, E. D., McBlair, W., & Taylor, K. M. (1951). The uptake of vanadium by

tunicates. Biol. Bull., 101(1), 84-94.

Goodbody, I. (1962). The biology of Ascidia nigra (Savigny). I. Survival and mortality in

an adult population Biol. Bull., 122(1), 40-51.

Goodbody, I. (1963). The biology of Ascidia nigra (Savigny). II. The development and

survival of young ascidians. Biol. Bull., 124(1), 31-44.

Goodbody, I. (1974). The physiology of ascidians. In F. S. Russell & S. M. Yonge (Eds.),

Advances in marine biology (Vol. 12, pp. 1-149): Academic Press, Inc., Harcourt

Brace Jovanovich.

Grizzle, R. E., Greene, J. K., Luckenbach, M. W., & Coen, L. D. (2006). A new in situ

method for measuring seston uptake by suspension-feeding bivalve molluscs. J.

Shellfish Res., 25(2), 643-650.

Halpern, B. S., Selkoe, K. A., Micheli, F., & Carrie V. Kappel (2007). Evaluating and

ranking the vulnerability of global marine ecosystems to anthropogenic threats.

Conserv. Biol., 21(5), 1301-1315.

Hameedi, J. J. (1997). Strategy for monitoring the environment in the coastal zone. In B.

U. Haq (Ed.), Coastal zone management imperative for maritime developing

nations (pp. 111-142). Rotterdam, Netherlands: Kluwer Academic Publishers.

Hauserman, J. (2007). Florida's Coastal and Ocean Future - a blueprint for economic

and environmental leadership: Caribbean Conservation Corporation and Sea

Turtle Survival League, Clean Water Network of Florida, Environmental Defense,

Natural Resources Defense Council, National Wildlife Federation, Reef Relief,

The Surfrider Foundation, The Ocean Conservancy. Retrieved from

101

http://www.nrdc.org/water/oceans/florida/flfuture.pdf

Hecht, S. (1916). The water current produced by Ascidia atra Lesueur. J. Exp. Zool.,

20(3), 429-434.

Heck, K. L., Hays, G., & Orth, R. J. (2003). Critical evaluation of the nursery role

hypothesis for seagrass meadows. Mar. Ecol. Prog. Ser., 253(0), 123-136.

Hildebrand, S. F., & Cable, L. E. (1934). Reproduction and development of whitings or

kingfishes, drums, spot, croaker and weakfishes or sea trouts, Family Sciaenidae,

of the Atlantic coast of the United States. Bull. U.S. Bur. Fish., 48(0), 41-117.

Holm-Hansen, O., & Booth, C. R. (1966). The measurement of adenosine triphosphate in

the ocean and its ecological significance. Limnol. Oceanogr., 11(4), 510-519.

Hugenholtz, P., Goebel, B. M., & Pace, N. R. (1998). Impact of culture-independent

studies on the emerging phylogenetic view of bacterial diversity. J. Bacteriol.,

180(18), 4765-4774.

Hull, M. S., Cherry, D. S., & Neves, R. J. (2006). Use of bivalve metrics to quantify

influences of coal-related activities in the Clinch River watershed, Virginia.

Hydrobiologia, 556(1), 341-355.

Hunt, H. L., & Scheibling, R. E. (1997). Role of early post-settlement mortality in

recruitment of benthic marine invertebrates. Mar. Ecol. Prog. Ser., 155(1), 269-

301.

Indian River Lagoon National Estuary Program (1996). The Indian River Lagoon: Our

heritage at risk—The Indian River Lagoon comprehensive conservation and

management plan. Sponsored by the St. Johns and South Florida Water

Management Districts in cooperation with the US Environmental Protection

102

Agency Melbourne, FL.

Inoue, S., Abe, S.-I., Oshima, Y., Kai, N., & Honjo, T. (2006). Tributyltin contamination

of bivalves in coastal areas around northern Kyushu, Japan. Environ. Toxicol.,

21(3), 244-249.

Jannasch, H. W., & Jones, G. E. (1959). Bacterial populations in sea water as determined

by different methods of enumeration. Limnol. Oceanogr., 4(2), 128-139.

Jennings, C. A. (1985). Species profiles: life histories and environmental requirements of

coastal fishes and invertebrates (Gulf of Mexico) - Sheepshead. U.S. Fish and

Wildlife Service Biological Reports 82(11.29): U.S. Army Corps of Engineers, TR

EL-82-4.

Jiang, A., Guo, J., Cai, W., & Wang, C. (2008a). Oxygen consumption of the ascidian

Styela clava in relation to body mass, temperature and salinity. Aquacult. Res.,

39(14), 1562-1568.

Jiang, A., Lin, J., & Wang, C. (2008b). Physiological energetics of the ascidian Styela

clava in relation to body size and temperature. Comp. Biochem. Physiol. A,

149(2), 129-136.

Jørgensen, C. B. (1949). Feeding-rates of sponges, lamellibranchs and ascidians. Nature,

163(4154), 912.

Jørgensen, C. B. (1966a). Biology of suspension feeding. Oxford, Great Britain:

Pergamon Press.

Jørgensen, C. B. (1966b). Tunicata. In G. A. Kerkut (Ed.), Biology of suspension feeding

(pp. 112-116). Oxford, Great Britain: Pergamon Press.

Joseph, E. B., Massmann, W. H., & Norcross, J. J. (1964). The pelagic eggs and early

103

larval stages of the black drum from Chesapeake Bay. Copeia, 1964(2), 425-434.

Kennedy, V. S., Newell, R. I. E., & Eble, A. F. (Eds.). (1996). The eastern oyster

Crassostrea virginica. College Park, MD.

Kennish, M. J. (2002). Environmental threats and environmental future of estuaries.

Environ. Conserv., 29(1), 78-107.

Kim, Y. S., & Moon, T. S. (1998). Filtering rate with effect of water temperature and size

of two farming ascidians Styela clava and S. plicata, and a farming mussel

Mytilus edulis. J. Korean Fish. Soc., 31(2), 272-277.

Kimmich, G. A., Randles, J., & Brand, J. S. (1975). Assay of picomole amounts of ATP,

ADP, and AMP using the luciferase enzyme system. Anal. Biochem., 69(1), 187-

206.

Kohata, K., Hiwatari, T., & Hagiwara, T. (2003). Natural water-purification system

observed in a shallow coastal lagoon: Matsukawa-ura, Japan. Mar. Pollut. Bull.,

47(1-6), 148-154.

Kowalke, J. (1999). Filtration in antarctic ascidians - striking a balance. J. Exp. Mar.

Biol. Ecol., 242(2), 233-244.

Lambert, C. C., & Brandt, C. L. (1967). The effect of light on the spawning of Ciona

intestinalis. Biol. Bull., 132(2), 222-228.

Lambert, C. C., & Lambert, G. (1998). Non-indigenous ascidians in southern California

harbors and marinas. Mar. Biol., 130(4), 675-688.

Lebaron, P., Parthuisot, N., & Catala, P. (1998). Comparison of blue nucleic acid dyes for

flow cytometric enumeration of bacteria in aquatic systems. Appl. Environ.

Microbiol., 64(5), 1725-1730.

104

Lewis III, R. R. "Coastal habitat restoration as a fishery management tool". Paper

presented at the A Symposium on Conservation of Coastal Fish Habitat,1991,

March 7-9 Baltimore, Maryland. Ed. R. H. Stroud. National Coalition for Marine

Conservation, Inc., Print.

Lipcius, R. N., Seitz, R. D., Seebo, M. S., & Colon-Carrion, D. (2005). Density,

abundance and survival of the blue crab in seagrass and unstructured salt marsh

nurseries of Chesapeake Bay. J. Exp. Mar. Biol. Ecol., 319(1-2), 69-80.

Lirman, D., Deangelo, G., Serafy, J., Hazra, A., Hazra, D. S., Herlan, J., Luo, J.,

Bellmund, S., Wang, J., & Clausing, R. (2008). Seasonal changes in the

abundance and distribution of submerged aquatic vegetation in a highly managed

coastal lagoon. Hydrobiologia, 596(1), 105-120.

Lotze, H. K., Lenihan, H. S., Bourque, B. J., Bradbury, R. H., Cooke, R. G., Kay, M. C.,

Kidwell, S. M., Kirby, M. X., Peterson, C. H., & Jackson, J. B. C. (2006).

Depletion, degradation, and recovery potential of estuaries and coastal seas.

Science, 312(5781), 1806-1809.

MacGinite, G. E. (1939). The method of feeding of tunicates. Biol. Bull., 77(3), 443-447.

Mackie, G. O., Burighel, P., Caicci, F., & Manni, L. (2006). Innervation of ascidian

siphons and their responses to stimulation. Can. J. Zool., 84(8), 1146 - 1162.

Madigan, M. T., Martinko, J. M., & Parker, J. (2000). Public health and water quality. In

P. Corey (Ed.), Brock biology of microorganisms (9th ed., pp. 974-976). Upper

Saddle River, NJ: Prentice-Hall.

Manríquez, P. H., & Casilla, J. C. (2007). Roles of larval behaviour and microhabitat

traits in determining spatial aggregations in the ascidian Pyura chilensis. Mar.

105

Ecol. Prog. Ser., 332(0), 155-165.

Marques, J. C., Pardal, M. A., Nielsen, S. N., & Jorgensen, S. E. (1997). Analysis of the

properties of exergy and biodiversity along an estuarine gradient of

eutrophication. Ecol. Model., 102(1), 155-167.

Meglitsch, P. A. (1967). Invertebrate zoology. New York, NY: Oxford University Press.

Milanese, M., Chelossi, E., Manconi, R., Sara, A., Sidri, M., & Pronzato, R. (2003). The

marine sponge Chondrilla nucula Schmidt, 1862 as an elective candidate for

bioremediation in integrated aquaculture. Biomol. Eng. , 20(4-6), 363-368.

Millar, R. H. (1958). Some ascidians from Brazil. Annals Mag. Nat. Hist. Ser. , 13(1),

497-514.

Monniot, C. (1972). Ascidies phlébobranches des Bermudes. Bull. Mus. Nat. His. Nat.,

3rd ser. 82(61), 939-948.

Montgomery, J. R., & Smith, N. P. (1983). Future of the Indian River System. Fla. Sci.,

46(3-4), 129-131.

Mook, D. (1983). Indian River fouling organisms: a review. Florida Scientist, 46(0),

162-167.

Morris, L., & Virnstein, R. (2004). The demise and recovery of seagrass in the northern

Indian River Lagoon, Florida. Estuaries Coasts, 27(6), 915-922.

Muller, R. G., K.Tisdel, & Murphy, M. D. (2002). The 2002 update of the stock

assessment of Florida Pompano (Trachinotus carolinus). Florida Fish and

Wildlife Conservation Commission, Florida Marine Research Institute, St.

Petersburg, FL.

Murphy, M. D., Adams, D. H., & Tremain, D. M. (1998). Direct validation of ages

106

determined for adult black drum, Pogonias cromis in east-central Florida, with

notes on black drum migration. Fish. Bull., 96(0), 382-387.

MyFlorida.com (2010). Basic recreational saltwater fishing regulations Retrieved Jan. 25,

2010, MyFWC.com Florida Fish and Wildlife Conservation Commission:

Retreived from http://myfwc.com/RULESANDREGS/SaltwaterRules_index.htm

National Marine Fisheries Service (1998). Recovery plan for the shortnose sturgeon

(Acipenser brevirostrum) Silver Spring, MD: National Marine Fisheries Service.

Newell, R. I. E. (1988). Ecological changes in Chesapeake Bay: Are they the result of

overharvesting the American oyster, Crassostrea virginica? In E. Krome & M.

Haire (Eds.), Understanding the estuary: Advances in Chesapeake Bay research.

(Vol. 129, pp. 536-546). Baltimore, Maryland: Chesapeake Research Consortium

Publication 129 (CBP/TRS 24/88).

Newell, R. I. E., Fisher, T. R., Holyoke, R. R., & Cornwell, J. C. (2005). Influence of

eastern oysters on nitrogen and phosphorus regeneration in Chesapeake Bay,

USA. In R. F. Dame & S. Olenin (Eds.), The comparative roles of suspension-

feeders in ecosystems (Vol. 47, pp. 93-120): Springer Netherlands.

Officer, C. B., Smayda, T. J., & Mann, R. (1982). Benthic filter feeding: A natural

eutrophication control. Mar. Ecol. Prog. Ser., 9(0), 203-210.

Oie, S., Kamiya, A., Tomita, M., Katayama, A., Iwasaki, A., & Miyamura, S. (1999).

Efficacy of disinfectants and heat against Escherichia coli O157:H7. Microbios,

98(389), 7-14.

Olafsson, E. B., Peterson, C. H., & Ambrose Jr., W. G. (1994). Does recruitment

limitation structure populations and communities of macro-invertebrates in

107

marine soft sediments: The relative signficance of pre- and post-settlement

processes. In H. Barnes, A. D. Ansell, R. N. Gibson & M. Barnes (Eds.),

Oceanography and marine biology: An annual review (Vol. 32, pp. 65-109).

London: University College London Press.

Olivas-Valdez, J. A., & Cáceres-Martínez, J. (2002). Infestation of the blue mussel

Mytilus galloprovincialis by the copepod Pseudomyicola spinosus and its relation

to size, density, and condition index of the host. J. Invertebr. Pathol., 79(2), 65-

71.

Orlova, M., Golubkov, S., Kalinina, L., & Ignatieva, N. (2004). Dreissena polymorpha

(Bivalvia: Dreissenidae) in the Neva Estuary (eastern Gulf of Finland, Baltic Sea):

Is it a biofilter or source for pollution? Mar. Pollut. Bull., 49(3), 196-205.

Paperno, R., Tremain, D., Adams, D., Sebastian, A., Sauer, J., & Dutka-Gianelli, J.

(2006). The disruption and recovery of fish communities in the Indian River

Lagoon, Florida, following two hurricanes in 2004. Estuaries Coasts, 29(6),

1004-1010.

Petersen, J. K., & Riisgård, H. U. (1992). Filtration capacity of the ascidian Ciona

intestinalis and its grazing impact in a shallow fjord. Mar. Ecol. Prog. Ser., 88(0),

9-17.

Petersen, J. K., Schou, O., & Thor, P. (1995). Growth and energetics in the ascidian

Ciona intestinalis. Mar. Ecol. Prog. Ser., 120(0), 175-184.

Petersen, J. K., & Svane, L. (2001). Filtration rate in seven Scandinavian ascidians:

implications of the morphology of the gill sac. Mar. Biol., 140(2), 397-402.

Peterson, B. J., Chester, C. M., Jochem, F. J., & Fourqurean, J. W. (2006). Potential role

108

of sponge communities in controlling phytoplankton blooms in Florida Bay. Mar.

Ecol. Prog. Ser., 328(0), 93-103.

Pierucci, O. (1978). Dimensions of Escherichia coli at various growth rates: model for

envelope growth. J. Bacteriol., 135(2), 559–574.

Poling, K. R., & Fuiman, L. A. (1997). Sensory development and concurrent behavioural

changes in Atlantic croaker larvae. J. Fish. Biol., 51(0), 402-421.

Radsimovsky, R. (1930). Vorliufige Angaben uiber die Dichtigkeit der bakteriellen

Besiedlung einiger Gewässer. Trav. Sta. Biol. Dniepre, 5(0), 385-402.

Randløv, A., & Riisgård, H. U. (1979). Efficiency of particle retention and filtration rate

in four species of ascidians. Mar. Ecol. Prog. Ser., 1(0), 55-59.

Reyier, E. A., & Shenker, J. M. (2007). Ichthyoplankton community structure in a

shallow subtropical estuary of the Florida Atlantic coast. Bull. Mar. Sci., 80(2),

267-293.

Robbins, I. J. (1983). The effects of body size, temperature, and suspension density on

the filtration and ingestion of inorganic particulate suspensions by ascidians. J.

Exp. Mar. Biol. Ecol., 70(1), 65-78.

Robins, C. R., Bailey, R. M., Bond, C. E., Brooker, J. R., Lachner, E. A., Lea, R. N., &

Scott, W. B. (Eds.). (1980). A list of common and scientific names of fishes from

the United States and Canada (4th ed.). Bethesda, MD: American Fisheries

Society Special Publication No. 12.

Rocha, J. M. S., Garcia, J. E. C., & Henriques, M. H. F. (2003). Growth aspects of the

marine microalga Nannochloropsis gaditana. Biomol. Eng., 20(4-6), 237-242.

Roszak, D. B., & Colwell, R. R. (1987). Survival strategies of bacteria in the natural

109

environment. Microbiol. Mol. Biol. Rev., 51(3), 365-379.

Roughgarden, J., Gaines, S., & Possingham, H. (1988). Recruitment dynamics in

complex life cycles. Science, 241(4872), 1460-1466.

Roumillat, W. A., & Brouwer, M. C. (2004). Reproductive dynamics of female spotted

seatrout (Cynoscion nebulosus) in South Carolina. Fish. Bull., 102(3), 473-487.

Rozen, Y., & Belkin, S. (2001). Survival of enteric bacteria in seawater. FEMS

Microbiol. Rev., 25(5), 513-529.

Satoh, T., Kato, J., Takiguchi, N., Ohtake, H., & Kuroda, A. (2004). ATP amplification

for ultrasensitive bioluminescence assay: Detection of a single bacterial cell.

Biosci. Biotechnol. Biochem., 68(6), 1216-1220.

SFWMD (2008a). Special Edition, Executive Director's Message. The Ripple Effect

September, 2008 Retrieved July 14, 2009, from

https://my.sfwmd.gov/pls/portal/docs/PAGE/COMMON/NEWSR/ENEWS/RIPP

LE/ARCHIVE/RIPPLE_INDEX_200808_SPEC_TSFAY.PDF

SFWMD (2008b). Weekly Environmental Recommendation for Systems Operations

Retrieved July 14, 2009, from

https://my.sfwmd.gov/portal/page/portal/pg_grp_sfwmd_hesm/portlet_techsumfil

es/lors08252008/wres_inp_aug2508.html

SFWMD (2009). 2009 South Florida Environmental Report (SFER) Retrieved Feb. 22,

2010, from

https://my.sfwmd.gov/portal/page/portal/PG_GRP_SFWMD_SFER/PG_SFWMD

_SFER_HOME

SFWMD, SJRWMD, & Indian River Lagoon National Estuary Program (2007). Indian

110

River Lagoon Program Retrieved Dec. 15, 2009, from

http://www.sjrwmd.com/itsyourlagoon/pdfs/IRL_Natural_Treasure_book.pdf

Short, F. T., & Wyllie-Echeverria, S. (1996). Natural and human induced disturbance of

seagrasses. Environ. Conserv., 23(1), 17-27.

Sime, P. (2005). St. Lucie Estuary and Indian River Lagoon conceptual ecological model.

Wetlands, 25(4), 898-907.

Sisgaard, S. J., Petersen, J. K., & Iversen, J. J. L. (2003). Relationship between specific

dynamic action and food quality in the solitary ascidian Ciona intestinalis. Mar.

Biol., 143(6), 1143-1149.

Smithsonian Marine Station at Ft. Pierce (2005a). The Indian River Lagoon Species

Inventory Retrieved Jan. 7, 2010, from

http://www.sms.si.edu/irLspec/Lutjan_analis.htm

Smithsonian Marine Station at Ft. Pierce (2005b). The Indian River Lagoon Species

Inventory. Retrieved Jan. 7, 2010, from

http://www.sms.si.edu/irLspec/Centro_undeci.htm

Smithsonian Marine Station at Ft. Pierce (2008). Field Guide to the Indian River Lagoon

Retrieved Nov. 20, 2008, from

http://www.sms.si.edu/IRLFieldGuide/IRLBiodiv.htm

Stachowicz, J. J., Fried, H., Osman, R. W., & Whitlatch, R. B. (2002). Biodiversity,

invasion resistance, and marine ecosystem function: Reconciling pattern and

process. Ecology, 83(9), 2575-2590.

Staley, J. T., & Konopka, A. (1985). Measurement of in situ activities of

nonphotosynthetic microorganisms in aquatic and terrestrial habitats. Annu. Rev.

111

Microbiol., 39(0), 321-346.

Steidinger, K. A., Landsberg, J. H., Tomas, C. R., & Burns, J. W. (1999). Harmful algal

blooms in Florida. Retreived from

http://research.myfwc.com/engine/download_redirection_process.asp?file=HAB_

whitepaper2006_UPDATE.pdf&objid=26925&dltype=article

Stuart, V., & Klumpp, D. W. (1984). Evidence for food-resource partitioning by kelp-bed

filter feeders. Mar. Ecol. Prog. Ser., 16(0), 27-37.

Svane, I., & Petersen, J. K. (2001). On the problems of epibioses, fouling and artificial

reefs, a review. Mar. Ecol. , 22(3), 169-188.

Svane, I., & Young, C. M. (1989). The ecology and behavior of ascidian larvae.

Oceanogr. Mar. Biol. Annu. Rev., 27(1), 45-90.

Takeda, S., & Kurihara, Y. (1994). Preliminary study of management of red tide water by

the filter feeder Mytilus edulis galloprovincialis. Mar. Pollut. Bull., 28(11), 662-

667.

Texas Parks and Wildlife (2009). Atlantic Croaker (Micropogonias undulatus). Retrieved

Jan. 27, 2010, Texas Parks and Wildlife Department: Retreived from

http://www.tpwd.state.tx.us/huntwild/wild/species/croaker/

Thiyagarajan, V., & Qian, P.-Y. (2003). Effect of temperature, salinity and delayed

attachment on development of the solitary ascidian Styela plicata (Lesueur). J.

Exp. Mar. Biol. Ecol., 290(1), 133-146.

Thompson, M., & Munro, J. L. (1974). The biology, ecology, exploitation and

management of Caribbean reef fishes; scientific report of the O.D.S./U.W.I.

fisheries. Ecology Research Project 1969-1973. Part V. The biology, ecology and

112

bionomics of Caribbean reef fishes: V.D. Lutjanidae (snappers): Zool. Dept.

Univ. West Indies. Kingston, Jamaica. Res. Rep. 3:1-69.

Thompson, R. J., Bayne, C. J., Moore, M. N., & Carefoot, T. H. (1978). Haemolymph

volume, changes in the biochemical composition of the blood, and cytological

responses of the digestive cells in Mytilus californianus Conrad, induced by

nutritional, thermal and exposure stress. J. Comp. Physiol. [B], 127(4), 287-298.

Thorson, G. (1964). Light as an ecological factor in the dispersal and settlement of larvae

of marine bottom invertebrates. Ophelia, 1(1), 167-208.

Tranter, D. J., & Smith, P. E. (1968). Filtration performance. In D. J. Tranter (Ed.),

Zooplankton sampling, Part I: Reviews on zooplankton sampling methods (pp.

27-56). Paris: Unesco.

Troussellier, M., Bonnefont, J.-L., Courties, C., Derrien, A., Dupray, E., Gauthier, M.,

Gourmelon, M., Joux, F., Lebaron, P., Martin, Y., & Pommepuy, M. (1998).

Responses of enteric bacteria to environmental stresses in seawater. Oceanolog.

Acta 21(6), 965-981.

Trudil, D., Loomis, L., Pabon, R., Hasan, J. A. K., & Trudil, C. L. (2000). Rapid ATP

method for the screening and identification of bacteria in food and water samples.

Moscow Univ. Chem. Bull. , 41(0), 27-30.

Tujula, N., Radford, J., Nair, S. V., & Raftos, D. A. (2001). Effects of tributyltin and

other metals on the phenoloxidase activating system of the tunicate, Styela

plicata. Aquat. Toxicol., 55(3-4), 191-201.

Turner, J. T., Tester, P. A., & Hettler, W. F. (1985). Zooplankton feeding ecology. Mar.

Biol., 90(1), 1-8.

113

Ulanowicz, R. E., & Tuttle, J. H. (1992). The trophic consequences of oyster stock

rehabilitation in Chesapeake Bay. Estuaries, 15(3), 298-306.

US National Museum of Natural History (2002). Department of Systematic Biology:

Invertebrate Zoology collections database. Retrieved Dec. 10, 2006, US National

Museum of Natural History: Retreived from http://goode.si.edu/mcs/iz/ query.php

USEPA (2006, June 9, 2009). National Estuary Program - Indian River Lagoon Retrieved

Sept. 21, 2006, from http://www.epa.gov/owow/estuaries/programs/irl.html

Van Name, W. G. (1912). Simple ascidians of the coasts of New England and

neighboring British provinces. Proceedings of the Boston Society of Natural

History (Vol. 34, pp. 439-619 ). Boston, MA: Boston Society of Natural History.

Van Name, W. G. (1945). The North and South American ascidians. BulI. Am. Mus. Nat.

Hist, 84(0), 1-476.

Vazquez, E., & Young, C. M. (1998). Ontogenetic changes in phototaxis during larval

life of the Ascidian Polyandrocarpa zorritensis J. Exp. Mar. Biol. Ecol., 231(2),

267-277.

Virnstein, R. W. (1995). Anomalous diversity of some seagrass-associated fauna in the

Indian River Lagoon, Florida. B. Mar. Sci., 57(1), 75-78.

Wallace, B. J., & Malas, D. (1976). The fine structure of capture nets of larval

Philoptamidae (Trichoptera) with special emphasis on Dolophilodes distinctus.

Canad. J. Zool., 54(10), 1788-1802.

Walne, P. R. (1972). The influence of current speed, body size and water temperature on

the filtration rate of five species of bivalves. J. Mar. Biol. Assoc. U. K, 52, 345-

374.

114

Ward, J. E., Levinton, J. S., Shumway, S. E., & Cucci, T. (1997). Site of particle

selection in bivalve mollusc. Nature, 390(0), 131-132.

Ward, J. E., Levinton, J. S., Shumway, S. E., & Cucci, T. (1998). Particle sorting in

bivalves: in vivo determination of the pallial organs of selection. Mar. Biol.,

131(2), 283-292.

Ward, J. E., & Shumway, S. E. (2004). Separating the grain from the chaff: particle

selection in suspension- and deposit-feeding bivalves. J. Exp. Mar. Biol. Ecol.,

300(1-2), 83-130.

Weber, P. (1994, March/April). It comes down to the coasts. World Watch, 7, 20-29.

West, A. B., & Lambert, C. C. (1976). Control of spawning in the tunicate Styela plicata

by variations in a natural light regime. J. Exp. Zool., 195(2), 263-270.

Whitaker, J. D. (2005). Comprehensive Wildlife Conservation Strategy. South Carolina

Department of Natural Resources: Retreived from

http://www.dnr.sc.gov/cwcs/pdf/Croaker.pdf

Wicklund, R. (1969). Observations on spawning of lane snapper. Underwater Nat., 6(2),

40.

Worm, B., Barbier, E. B., Beaumont, N., Duffy, J. E., Folke, C., Halpern, B. S., Jackson,

J. B., Lotze, H. K., Micheli, F., Palumbi, S. R., Sala, E., Selkoe, K. A.,

Stachowicz, J. J., & Watson, R. (2006). Impacts of biodiversity loss on ocean

ecosystem services. Science, 314(5880), 787-790.

Yasumoto-Hirose, M., Nishijima, M., Ngirchechol, M., Kanoh, K., Shizuri, Y., & Miki,

W. (2006). Isolation of marine bacteria by in situ culture on media-supplemented

polyurethane foam. Mar. Biotechnol., 8(3), 227-237.

115

Young, C. M., & Braithwaite, L. F. (1980a). Larval behavior and post-settling

morphology in the ascidian, Chelyosoma productum Stimpson. J. Exp. Mar. Biol.

Ecol., 42(2), 157-169.

Young, C. M., & Braithwaite, L. F. (1980b). Orientation and current-induced flow in the

stalked ascidian Styela montereyensis. Mar. Biol., 159(2), 428-440.

Young, C. M., & Chia, F.-S. (1984). Microhabitat-associated variability in survival and

growth of subtidal solitary ascidians during the first 21 days after settlement. Mar.

Biol. , 81(1), 61-68.

116