Vol. 515: 275–279, 2014 MARINE ECOLOGY PROGRESS SERIES Published November 18 doi: 10.3354/meps11008 Mar Ecol Prog Ser

NOTE

Feeding ecology of the largest mastigoteuthid squid species, Idioteuthis cordiformis (Cephalopoda, Mastigoteuthidae)

Heather E. Braid*, Kathrin S. R. Bolstad

Institute for Applied Ecology New Zealand, School of Applied Sciences, Auckland University of Technology, Private Bag 92006, Auckland 1142, New Zealand

ABSTRACT: The squid Idioteuthis cordiformis is consumed by several apex predators, but nothing is known of its own feeding ecology. Its anomalous size and morphology within the family Mastig- oteuthidae suggest that its diet may also be unusual, but material suitable for dietary studies has not been available previously. Herein, using several opportunistically collected specimens from New Zealand waters, gut contents were examined using DNA barcoding, revealing that this spe- cies feeds upon birdbeak dogfish Deania calcea and, apparently, snapper Lutjanus sp. Stable- isotope analysis of 15N and 13C confirmed that D. calcea occupied a trophic level immediately below that of the squid. These data suggest that this squid may prey actively upon relatively large pelagic species; alternatively, it could be net feeding and/or scavenging. I. cordiformis has 15N sig- natures comparable to those of the colossal squid and higher than those of the giant squid, which could indicate a high trophic position; however, further research is needed to conclusively distin- guish predation from scavenging.

KEY WORDS: Idioteuthis cordiformis · Mastigoteuthidae · New Zealand · DNA barcoding · Stable isotopes

Resale or republication not permitted without written consent of the publisher

INTRODUCTION New Zealand in the greatest danger of local ex - tinction. This benthic, bathypelagic species, found The largest species in the family Mastigoteuthidae around northern New Zealand, is associated with is Idioteuthis cordiformis; many times larger than any seamounts, including those in the Bay of Plenty and other species in this family, it attains more than 1 m in on the Chatham Rise (Freeman et al. 2010), where mantle length and 75 kg in mass (S. O’Shea pers. this species has been caught between 750 and comm.). It is known from Sumatra (Chun 1908), 1500 m (Braid 2013). Its occurrence on seamounts Japan (Sasaki 1929), the Philippines (Voss 1963) and makes it vulnerable to deep-sea fishing activities that New Zealand, where it was recently listed as nation- target these structures. Although fishing is believed ally endangered (Hitchmough et al. 2005). However, to have strong effects on the New Zealand I. cordi- after the re-evaluation of the threat classification formis population (Freeman et al. 2010), the actual system, the status for I. cordiformis was elevated to impacts have not been verified due to limited data. nationally critical (Freeman et al. 2010), meaning A variety of different marine species are known to that it is considered to be within the top 10 species in consume I. cordiformis. It has been found in the diets

*Corresponding author: [email protected] © Inter-Research 2014 · www.int-res.com 276 Mar Ecol Prog Ser 515: 275–279, 2014

of blue marlin Makaira nigricans (Shimose et al. zing (1) gut contents from fresh specimens using 2012), sperm whales Physeter macrocephalus (Evans DNA barcoding, and (2) food sources and trophic & Hindell 2004), swordfish Xiphias gladius (Watan- position using stable isotopes 13C and 15N. abe et al. 2009) and neon flying squid Ommastrephes bartramii (Watanabe et al. 2008). Conversely, very little is known of the diet of any mastigoteuthid, MATERIALS AND METHODS although one species (Mastigoteuthis psychrophila) has been reported to consume euphausiids (Kear This study was based on opportunistically col- 1992). The tentacular morphology of most mastigo- lected samples because fresh or frozen material was teuthids (unexpanded clubs with microscopic suck- required. Four frozen individuals of Idioteuthis ers) suggests that they are passive predators, dan- cordiformis were available for gut content and sta- gling their tentacles in search of prey (Dilly et al. ble isotope analysis; one lacked collection data, and 1977). However, the diet of I. cordiformis (a mastigo- the others were collected from West Norfolk Ridge teuthid with unusually large tentacular suckers for by the R/V ‘Tangaroa’, during the 2003 New this family, making the tentacle club more compara- Zealand and Australia Norfolk Ridge and Lord ble to those of large ommastrephids) remains com- Howe Rise Biodiversity Voyage (NORFANZ). Speci- pletely unknown. mens were kept frozen at −20°C until analysis Previously, morphological analysis of the gut con- (Table 1). Individual prey items (pieces of scale, tents has been successful for some large squids (Bol- bone or tissue) were removed and stored individu- stad & O’Shea 2004, Ibáñez et al. 2008, Stewart et al. ally in 100% EtOH. Eight items were chosen from 2013), but this approach relies on identifiable hard each gut for analysis based on morphological differ- parts, which are often lacking in squid gut contents, ences (different tissue types or colours). The DNA since any items that successfully traverse the esoph- barcode region (Hebert et al. 2003) was amplified agus have been finely masticated by the beak (Jack- for 8 items from each of 2 gut contents of 2 individu- son et al. 2007). However, DNA analysis can identify als (2 were empty) using primers and methods fol- squid prey from even fragmentary remains and soft lowing Braid et al. (2012), except that DNA was tissue (e.g. Braley et al. 2010, Braid et al. 2012). Gut extracted using the Xytogen Animal Extraction Kit contents identified by either method can give insight and sequenced with Folmer et al. (1994) primers. into specific recently consumed prey, while stable Edited sequences were uploaded to the Barcode of isotopes can give a more general, long-term view of Life Data Systems (BOLD) under the project name the trophic role of a species and have been used to ‘Idioteuthis cordiformis Gut Contents’ (project code study the trophic interactions of other squids (e.g. DBICG) and compared against the BOLD COI Spe- Cherel & Hobson 2005, Cherel et al. 2008). Stable cies Database (Ratnasingham & Hebert 2007) in isotopes for nitrogen can be used to infer trophic January 2014. Species-level identifications were position because the tissue of predators is enriched made using a tree-based identification method by with 15N compared to their prey, while carbon, which examining neighbor-joining trees. remains relatively constant, can be used to geo- Muscle samples from the fins of the 4 I. cordiformis graphically pinpoint sources of primary productivity. specimens were used for stable-isotope analysis for This research aimed to gain the first known insight carbon and nitrogen (Table 1). Approximately into the feeding ecology of I. cordiformis by analy - 700 mg of fin tissue was dried at 60°C for 48 h. The

Table 1. Stable isotope analysis of New Zealand Idioteuthis cordiformis gut contents. No collection data were available for NIWA 84390; gear type was orange roughy trawl (ORT) or ratcatcher (RTC); trophic level was estimated based on the equation from Cherel et al. (2008). ML: mantle length; indet.: sex indeterminate

Specimen ID Sex ML Latitude Longitude Depth (m) Date δ15N δ13C Estimated (mm) (°S) (°E) and gear type (d/mo/yr) (‰) (‰) trophic level

NMNZ M.306356 Indet. 405 34.24 168.35 1195−1200 (ORT) 3/06/2003 16.121 −18.160 5.9 NMNZ M.306355 m 513 34.57 168.94 1013−1340 (RTC) 3/06/2003 16.258 −18.043 6 NMNZ M.306358 m 549 33.78 167.49 1431−1460 (ORT) 29/05/2003 15.538 −18.284 5.7 NIWA 84390 f 820 −− − −16.602 −17.831 6.1 Braid & Bolstad: Idioteuthis cordiformis ecology 277

tissue was ground using a mortar and pestle under of this fluid suggested that it was from myctophid liquid nitrogen. Samples were sent to the Waikato fishes, which are known to contain a high level of oil Stable Isotope Unit, University of Waikato, Hamilton, (Phleger et al. 1999). New Zealand. Analysis was performed using a fully The shark sequence was unexpected because automated Europa Scientific 20/20 isotope analyzer. mastigoteuthids are generally believed to be passive δ13C was calculated using precalibrated C4 sucrose predators (Dilly et al. 1977) that likely — based on that is cross-referenced to PeeDee belemnite. δ15N the diet of M. psychrophila — consume euphausiids was calculated using a urea standard traceable to (Kear 1992). The sequence of the birdbeak dogfish atmospheric nitrogen. Trophic levels were calculated Deania calcea matched most closely with those of δ15 according to the equation TL = [( Nx − 3.4)/3.2] + specimens from the Tasman Sea, which is near the 2.0 from Cherel et al. (2008), where TL is trophic capture location for these I. cordiformis specimens level and x is the individual squid. (Norfolk Ridge). Deania calcea, which is relatively abundant in the same depths where I. cordiformis is often collected (Zintzen et al. 2011), was also caught RESULTS during the NORFANZ voyage (Clark et al. 2003). While initially surprising, it is not unreasonable to Of the 4 Idioteuthis cordiformis individuals, only 2 find sharks as prey of a cephalopod. Dogfish skin contained gut contents suitable for DNA analysis, has been found in the gut contents of another large while the others contained only red-orange oil. Of squid in New Zealand waters, Onykia ingens the 16 samples extracted, 10 PCR amplicons were (J. McKinnon pers. comm.). The considerable mac- obtained, and from these, 2 barcodes were success- eration of the gut contents prevented any estimation fully sequenced and identified as prey. One se - of prey size; however, a large squid could consume quence was a 100% match to the shark Deania cal- a small shark, and reports also exist of squid cea. The other sequence was a 100% match to actively preying on animals of sizes comparable to, Lutjanus sp. (this sequence was a 100% match to or even larger than, themselves (e.g. MBARI 2013). BOLD records for 2 species: L. lemniscatus and L. ful- The shark may also have been scavenged. Scaveng- vus). I. cordiformis showed high trophic levels, rang- ing would be a rare primary feeding strategy, since ing from 5.7 to 6.1 (Table 1). to date only one cephalopod has been revealed as a detritivore (the vampire squid Vampyroteuthis infer- nalis; see Hoving & Robison, 2012). Another possi- DISCUSSION bility is that the squids analysed here had fed on prey caught in the same trawl net, as has been This study is of particular importance because Idio- reported for other large squids, such as Dosidicus teuthis cordiformis specimens rarely have gut con- gigas (Ibáñez et al. 2008). tents (S. O’Shea pers. comm.), and such contents have The snapper sequence (Lutjanus sp.) is from a shal- not been previously reported; the only other known low-water reef species (Newman & Williams 1996). gut contents were otoliths from orange roughy However, the 2 potential species matches for the Hoplostethus atlanticus anecdotally observed in a sequence from I. cordiformis gut contents — L. lem- single specimen (D. Stevens pers. comm.). This spec- niscatus and L. fulvus — are both found in the Great imen was a fe male of 510 mm mantle length col- Barrier Reef (Newman & Williams 1996), which is lected from the Challenger Plateau, and the otoliths west of the Norfolk Ridge. Some deep-sea fishes, were from 2 fish that were both approximately 30 cm such as rattails (Coryphaenoides armatus and C. ya - in standard length; this may represent net feeding quinae), have been shown to scavenge shallow- because the I. cordiformis was collected during an water carrion (Drazen et al. 2008). However, the orange roughy survey. Lu & Williams (1994) experi- presence of snapper could also be due to a different enced similar difficulties in examining Mastigo- kind of sampling bias: secondary predation can be a teuthis psychrophila for gut contents: of 19 speci- significant source of error with molecular prey detec- mens, 12 had empty guts and the rest were missing tion (Harwood et al. 2001, Sheppard et al. 2005) and their viscera. In our specimens and in preserved could also potentially explain the Lutjanus sequence. material, a large amount of red-orange oil was found More research is needed to conclusively determine in the stomachs (see Braid 2013, Fig. 59C); Phillips et the feeding behavior of I. cordiformis. al. (2001) found a similarly high oil content in the gut Nitrogen, which becomes enriched as it passes fluid of Moroteuthis ingens. Their fatty acid analysis through the food chain, can be analysed to provide 278 Mar Ecol Prog Ser 515: 275–279, 2014

insight into the trophic level at which animals are feeding (Fry 2006). I. cordiformis, rather unex- pectedly, had an estimated trophic level comparable to that of the colossal squid Mesonychoteuthis hamiltoni and higher than that of the giant squid Architeuthis dux (Cherel et al. 2008), indicating that it occupies a high trophic position in the system. Another mastigo- teuthid, M. psychrophila, which is relatively small, was also previ- ously found to have higher δ15N values than those of the giant squid (Cherel & Hobson 2005). Relatively elevated δ15N values can be caused by scavenging, which was found in the gastropod Fig. 1. Trophic level for Idioteuthis cordiformis, other squids, and fish prey items Cyclope neritea (Camusso et al. of I. cordiformis. Trophic levels from Cherel et al. (2008) are included for Archi- 1998). teuthis dux, Me sonychoteuthis hamiltoni, Onykia in gens and Taningia danae. Trophic levels were calculated from nitrogen stable isotope levels (15N) using the Unlike nitrogen, the isotope equation from Cherel et al. (2008) for Deania calcea (Pethybridge 2010), Dosidi- value for carbon remains rela- cus gigas (Ruiz-Cooley et al. 2006), I. cordiformis and Vampyroteuthis infernalis tively constant as it moves through (Cherel et al. 2009). Trophic levels for Hoplostethus at lanticus and the Lutjanus the food chain, making carbon a species are from Froese & Pauly (2011) helpful tool for tracing the origin of primary production for a system (Fry 2006). Car- Acknowledgements. We thank the Museum of New Zealand Te Papa Tongarewa, especially B. Marshall, and National bon and nitrogen stable-isotope values have been Institute of Water and Atmospheric Research (NIWA), par- previously identified for D. calcea spe cimens from ticularly S. Mills, K. Schnabel, D. Stevens and D. Stotter. southeastern Australia (Pethybridge 2010). These Many thanks to R. Hanner at the University of Guelph for isotope values place D. calcea one trophic position laboratory space and reagents. Thanks to J. McKinnon from lower than I. cordiformis, with a similar carbon value the University of Otago; A. B. Evans, J. Kelly, M. Jones, B. Lorigan and S. Pointing from the Auckland University of (Pethybridge 2010), which supports it as potential Technology; J. Gross, A. Naaum and N. Serrao from the Uni- prey of I. cordiformis (Fig. 1). versity of Guelph; A. Rajendram from the University of This is the first investigation into the ecology of Waikato. Thanks also to S. O’Shea for advice and to Auck- any mastigoteuthid using stable isotopes and DNA land University of Technology for financial support. barcoding for gut-content analysis. DNA barcoding did successfully identify prey items, and stable-iso- LITERATURE CITED tope values indicated that this species could be a top predator; however, the gut contents revealed Bolstad KS, O’Shea S (2004) Gut contents of a giant squid Architeuthis dux (Cephalopoda: Oegopsida) from New herein could reflect scavenging or net feeding Zealand waters. NZ J Zool 31: 15−21 (Ibáñez et al. 2008, Hoving & Robison 2012). Braid HE (2013) Systematics and ecology of the New Cephalopod trophic interactions in the Southern Zealand Mastigoteuthidae (Cephalopoda, Oegopsida). Ocean are poorly known (Cherel & Hobson 2005); MAppSc dissertation, Auckland University of Technol- however, it is important to study the ecological role ogy, Auckland Braid HE, Deeds J, DeGrasse SL, Wilson JJ, Osborne J, Han- that this species plays to assess the impacts that its ner RH (2012) Preying on commercial fisheries and accu- population reduction could have on other deep-sea mulating paralytic shellfish toxins: a dietary analysis of species around New Zea land. Future research invasive Dosidicus gigas (Cephalopoda: Ommastrephi- should investigate the stable-isotope values and gut dae) stranded in Pacific Canada. Mar Biol 159:25−31 Braley M, Goldsworthy SD, Page B, Steer M, Austin JJ contents of other organisms within the local ecosys- (2010) Assessing morphological and DNA-based diet tem to better understand the trophic interactions analysis techniques in a generalist predator, the arrow between species and help predict the impacts that squid Nototodarus gouldi. Mol Ecol Resour 10:466−474 could occur from disturbances. Camusso M, Martinotti W, Balestrini R, Guzzi L (1998) C and Braid & Bolstad: Idioteuthis cordiformis ecology 279

N stable isotopes and trace metals in selected organisms Kear AJ (1992) The diet of Antarctic squid: comparison of from the river Po delta. Chemosphere 37: 2911−2920 conventional and serological gut contents analyses. Cherel Y, Hobson KA (2005) Stable isotopes, beaks and J Exp Mar Biol Ecol 156: 161−178 predators: a new tool to study the trophic ecology of Lu CC, Williams R (1994) Contribution to the biology of cephalopods, including giant and colossal squids. Proc R squid in the Prydz Bay region, Antarctica. Antarct Sci 6: Soc B 272: 1601−1607 223−229 Cherel Y, Ducatez S, Fontaine C, Richard P, Guinet C (2008) MBARI (2013) Midwater Ecology Expedition, November, Stable isotopes reveal the trophic position and mesopela- 2013: Day 5 — What’s for dinner? www.mbari.org/ gic fish diet of female southern elephant seals breeding expeditions/ Midwater2013-Fall/logbook/day5.htm (ac- on the Kerguelen Islands. Mar Ecol Prog Ser 370:239−247 cessed 1 Aug 2014) Cherel Y, Ridoux V, Spitz J, Richard P (2009) Stable isotopes Newman SJ, Williams DM (1996) Variation in reef associ- document the trophic structure of a deep-sea cephalopod ated assemblages of the Lutjanidae and Lethrinidae at assemblage including giant octopod and giant squid. different distances offshore in the central Great Barrier Biol Lett 5: 364−367 Reef. Environ Biol Fishes 46: 123−138 Chun C (1908) Über Cephalopoden der Deutschen Tiefsee- Pethybridge HR (2010) Ecology and physiology of deepwa- Expedition. Zool Anz 33: 86−89 ter chondrichthyans off southeast Australia: mercury, Clark M, Roberts C, Williams A, Last P (2003) Voyage report stable isotope and lipid analysis. PhD dissertation, Uni- of a biodiversity survey of seamounts and slopes of the versity of Tasmania, Hobart Norfolk Ridge and Lord Howe Rise (NORFANZ), May- Phillips KL, Jackson GD, Nichols PD (2001) Predation on June 2003. CSIRO, Hobart myctophids by the squid Moroteuthis ingens around Dilly PN, Nixon M, Young JZ (1977) Mastigoteuthis — the Macquarie and Heard Islands: stomach contents and whip-lash squid. J Zool 181: 527−559 fatty acid analyses. Mar Ecol Prog Ser 215: 179−189 Drazen JC, Popp BN, Choy CA, Clemente T, De Forest L, Phleger CF, Nelson MM, Mooney BD, Nichols PD (1999) Smith KL (2008) Bypassing the abyssal benthic food web: Wax esters versus triacylglycerols in myctophid fishes macrourid diet in the eastern North Pacific inferred from from the Southern Ocean. Antarct Sci 11: 436−444 stomach content and stable isotopes analyses. Limnol Ratnasingham S, Hebert PD (2007) BOLD: the barcode of Oceanogr 53:2644−2654 life data system (www. barcodinglife. org). Mol Ecol Evans K, Hindell MA (2004) The diet of sperm whales (Phy- Notes 7: 355−364 seter macrocephalus) in southern Australian waters. Ruiz-Cooley RI, Markaida U, Gendron D, Aguíñiga S (2006) ICES J Mar Sci 61:1313−1329 Stable isotopes in jumbo squid (Dosidicus gigas) beaks to Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R (1994) estimate its trophic position: comparison between stom- DNA primers for amplification of mitochondrial cyto - ach contents and stable isotopes. J Mar Biol Assoc UK 86: chrome c oxidase subunit I from diverse metazoan inver- 437−445 tebrates. Mol Mar Biol Biotechnol 3: 294−299 Sasaki M (1929) A monograph of the dibranchiate cepha- Freeman DJ, Marshall BA, Anyong ST, Wing SR, Hitch- lopods of the Japanese and adjacent waters. College of mough RA (2010) Conservation status of New Zealand Agriculture, Hokkaido Imperial University, Sapporo marine invertebrates, 2009. NZ J Mar Freshw Res 44: Sheppard SK, Bell J, Sunderland KD, Fenlon J, Skervin D, 129−148 Symondson WOC (2005) Detection of secondary preda- Froese R, Pauly D (2011) FishBase. www.fishbase.org tion by PCR analyses of the gut contents of invertebrate (accessed 30 Mar 2014) generalist predators. Mol Ecol 14:4461−4468 Fry B (2006) Stable isotope ecology. Springer, New York, NY Shimose T, Yokawa K, Saito H, Tachihara K (2012) Sexual Harwood JD, Phillips SW, Sunderland KD, Symondson difference in the migration pattern of blue marlin, WOC (2001) Secondary predation: quantification of food Makaira nigricans, related to spawning and feeding chain errors in an aphid-spider-carabid system using activities in the western and central North Pacific Ocean. monoclonal antibodies. Mol Ecol 10:2049−2057 Bull Mar Sci 88: 231−250 Hebert PDN, Cywinska A, Ball SL, deWaard JR (2003) Bio- Stewart JS, Field JC, Markaida U, Gilly WF (2013) Behav- logical identifications through DNA barcodes. Proc R Soc ioral ecology of jumbo squid (Dosidicus gigas) in relation Lond B 270:313−321 to oxygen minimum zones. Deep-Sea Res II 95:197−208 Hitchmough R, Bull L, Cromarty P (2005) New Zealand Voss GL (1963) Cephalopods of the Philippine Islands. threat classification systems list. Department of Conser- Smithsonian Institution, Washington, DC vation, Wellington Watanabe H, Kubodera T, Ichii T, Sakai M, Moku M, Seitou Hoving HJT, Robison BH (2012) Vampire squid: detritivores M (2008) Diet and sexual maturation of the neon flying in the oxygen minimum zone. Proc R Soc Lond B 279: squid Ommastrephes bartramii during autumn and 4559−4567 spring in the Kuroshio−Oyashio transition region. J Mar Ibáñez CM, Arancibia H, Cubillos LA (2008) Biases in de - Biol Assoc UK 88:381−389 termining the diet of jumbo squid Dosidicus gigas Watanabe H, Kubodera T, Yokawa K (2009) Feeding ecol- (D’Orbigny 1835) (Cephalopoda: Ommastrephidae) off ogy of the swordfish Xiphias gladius in the subtropical southern-central Chile (34°S−40°S). Mar Res 62: region and transition zone of the western North Pacific. 331−338 Mar Ecol Prog Ser 396: 111−122 Jackson GD, Bustamante P, Cherel Y, Fulton EA and others Zintzen V, Roberts CD, Clark MR, Williams A, Althaus F, (2007) Applying new tools to cephalopod trophic dynam- Last PR (2011) Composition, distribution and regional ics and ecology: perspectives from the Southern Ocean affinities of the deepwater ichthyofauna of the Lord Cephalopod Workshop, February 2−3, 2006. Rev Fish Howe Rise and Norfolk Ridge, south-west Pacific Ocean. Biol Fish 17: 79−99 Deep-Sea Res II 58:933−947

Editorial responsibility: Christine Paetzold, Submitted: April 22, 2014; Accepted: August 19, 2014 Oldendorf/Luhe, Germany Proofs received from author(s): October 23, 2014