Review of experimental and natural invertebrate hosts of sealworm ( Pseudoterranova decipiens ) and its distribution and abundance in macroinvertebrates in eastern Canada

Experimental and natural invertebrate intermediate hosts of sealworm ( Pseudoterranova decipiens ) as well as transmission experiments of sealworm from invertebrates to fish are reviewed and summarized. Experimental hosts include copepods, mysids, cumaceans, isopods, amphipods, decapods, annelids, and molluscs. Invertebrates collected from eastern Canada between 1989 and 1995 were checked for nematode infections by microscopic examination of dissected animals or enzymatic digestion of bulk samples. Third-stage larval sealworm were found in mysids ( Neomysis americana , Mysis stenolepis ) from Passamaquoddy Bay, the Bras d’Or Lakes, inshore Cape Breton, Sable Island and Sable Island Bank. Infected amphipods ( Amphiporeia virginiana, Americorchestia megalophthalma, Gammaruspp.) were found only on Sable Island. Typical infection rates in macroinvertebrates were 1-4/1000. No sealworm infections were found in approximately 18,000 amphipods examined from Sable Island Bank, the site of the most heavily infected fishes in eastern Canada. In Wallace Lake, a brackish pond on Sable Island, infection rates were much higher in mysids than in amphipods. Estimates of rates of transmission of sealworm from invertebrates to fish were derived from infection levels in Wallace Lake and feeding experiments involving sticklebacks and invertebrate prey. It is concluded that mysids may be much more important than amphipods in transmitting sealworm to fish hosts. Marcogliese, D. J. 2001. Review of experimental and natural invertebrate hosts of sealworm (Pseudoterranova decipiens ) and its distribution and abundance in macroinvertebrates in eastern Canada. NAMMCO Sci. Publ. 3:27-37.

R Re ev vi ie ew w o of f e ex xp pe er ri im me en nt ta al l a an nd d n na at tu ur ra al l i in nv ve er rt te eb br ra at te e h ho os st ts s o of f s se ea al lw wo or rm m ( (P Ps se eu ud do ot te er rr ra an no ov va a d de ec ci ip pi ie en ns s) ) a an nd d i it ts s d di is st tr ri ib bu ut ti io on n a an nd d a ab bu un nd da an nc ce e i in n m ma ac cr ro oi in n-v ve er rt te eb br ra at te es s i in n e ea as st te er rn n C Ca an na ad da a David J. Marcogliese Environment Canada,St. Lawrence Centre,105 McGill Street,7th Floor,Montreal,Quebec,Canada H2Y 2E7 A AB BS ST TR RA AC CT T Experimental and natural invertebrate intermediate hosts of sealworm (Pseudoterranova decipiens) as well as transmission experiments of sealworm from invertebrates to fish are reviewed and summarized.Experimental hosts include copepods, mysids, cumaceans, isopods, amphipods, decapods, annelids, and molluscs.Invertebrates collected from eastern Canada between 1989 and 1995 were checked for nematode infections by microscopic examination of dissected animals or enzymatic digestion of bulk samples.Third-stage larval sealworm were found in mysids (Neomysis americana, Mysis stenolepis) from Passamaquoddy Bay, the Bras d'Or Lakes, inshore Cape Breton, Sable Island and Sable Island Bank.Infected amphipods (Amphiporeia virginiana, Americorchestia megalophthalma, Gammarus spp.) were found only on Sable Island.Typical infection rates in macroinvertebrates were 1-4/1000.No sealworm infections were found in approximately 18,000 amphipods examined from Sable Island Bank, the site of the most heavily infected fishes in eastern Canada.In Wallace Lake, a brackish pond on Sable Island, infection rates were much higher in mysids than in amphipods.Estimates of rates of transmission of sealworm from invertebrates to fish were derived from infection levels in Wallace Lake and feeding experiments involving sticklebacks and invertebrate prey.It is concluded that mysids may be much more important than amphipods in transmitting sealworm to fish hosts.Marcogliese, D. J. 2001.Review of experimental and natural invertebrate hosts of sealworm (Pseudoterranova decipiens) and its distribution and abundance in macroinvertebrates in eastern Canada.NAMMCO Sci.Publ.3:27-37.

I IN NT TR RO OD DU UC CT TI IO ON N
S ealworms (Pseudoterranova decipiens) are parasitic nematodes (Anisakidae) whose adults infect stomachs of seals.Eggs are passed with the feces and sink to the bottom at a rate of 1.01 x 10 -4 m/s (McConnell et al. 1997), where free-living larvae hatch.There is debate over how many moults occur in the egg, and whether an L 2 or an L 3 stage larva hatches (Koie et al. 1995, Measures andHong 1995).Development and hatching are temperature, but not salinity, dependent (Brattey 1990, Measures 1996).These larvae are infective to invertebrates, which transmit the larvae to fish when ingested (McClelland et al. 1983).Seals typically acquire the parasite by ingesting infected fish.
Three sibling species designated Pseudoterranova decipiens A, B and C have been found in the North Atlantic, with types B and C occurring in Canadian waters (Paggi et al. 1991).Type B (P. decipiens sensu stricto after Paggi et al. 2000) is found primarily in grey seals (Halichoerus grypus), but also in harbour seals (Phoca vitulina) in the western North Atlantic, and type C is found only in bearded seals (Erignathus barbatus) (Paggi et al. 1991).Bristow and Berland (1992) suggested that the data on sealworm distribution and abundance from the Gulf of St. Lawrence must be interpreted carefully because all three seal species are found there.However, this is incorrect as the bearded seal occurs rarely in the Gulf, being confined to colder waters off Labrador and to the north.All sealworms recovered from seals and fish off Nova Scotia, Newfoundland, and the Gulf of St. Lawrence are P. decipiens B (Paggi et al. 1991, Brattey and Davidson 1996, Brattey and Stenson 1993).Thus, surveys of invertebrates, fish and seals in the regions of eastern Canada discussed herein are dealing with only one species of sealworm.In European waters, the situation is more complex, with considerable geographic overlap of sibling species A and B (Paggi et al. 1991).
A variety of microcrustaceans and macroinvertebrates have been infected experimentally with the sealworm (McClelland 1982, McClelland 1990).Nonetheless, until recently, records of natural infections of sealworm in invertebrates were sporadic and important questions pertaining to sealworm transmission remained unanswered.Herein, experimental and natural infections of sealworm in invertebrates are reviewed and integrated together with new developments in understanding the sealworm life cycle.

E EX XP PE ER RI IM ME EN NT TA AL L R RE EC CO OR RD DS S
McClelland (1982) successfully infected 12 species of harpacticoid copepods and one cyclopoid with larvae hatched from eggs obtained from gravid female sealworms recovered from harbour and grey seals (Table 1).The success rate of experimental infections varied from 58% to 100% in prevalence and 1.25 to 18.10 in mean intensity among the species and stages tested.Mature females developed the heaviest infections.Calanoid copepods (Eurytemora sp. and Pseudocalanus sp.) did not become infected, though one unpublished record of an infected calanoid exists (see Burt in McClelland 1995).Larvae in infected copepods demonstra-ted considerable growth, but little morphological change and no moult was detected (McClelland 1982).
A variety of macroinvertebrates have been infected experimentally with sealworm.These include five species of decapods, three species of amphipods, and single species of polychaete, nudibranch, mysid, cumacean, and isopod (Table 1) (Jarecka et al. 1988, McClelland 1990, Burt 1994).Among these hosts, the cumaceans, isopods and decapods displayed a haemocytic response which destroyed the parasite (McClelland 1990).The amphipods Gammarus oceanicus, Gammarus lawrencianus and Unciola irrorata and crab zoea could be infected upon direct exposure to newlyhatched larvae (Jarecka et al. 1988, McClelland 1990).However, serial transmissions involving benthic copepods (Halectinosoma, Tisbe, Ameira and Paracyclopina spp.) as initial hosts proved much more efficacious (McClelland 1990).All other macroinvertebrates only became infected upon exposure to infected copepods, but not to newly-hatched larvae (McClelland 1990).McClelland (1990) used adult gravid female sealworms from grey seals as a source of infective larvae in his experiments, whereas Jarecka et al. (1988) used worms from both grey and harbour seals with varying success.Burt (1994) infected G. oceanicus with hatched larvae produced by worms from grey seals, but was unable to obtain infections with those from harbour seals.
Sealworms did not survive beyond 21 days after free-living larvae were intraperitoneally injected or administered orally into rainbow trout (Oncorhynchus mykiss) (Smith et al. 1990), implying that invertebrate intermediate hosts are a requirement for completion of the sealworm life cycle.In a further set of experiments, McClelland (1995) demonstrated that only worms larger than 1.41 mm in length, and usually larger than 2.00 mm, were infective to fish.As sealworms cannot attain this size in microinvertebrate hosts such as copepods, these results imply that macroinvertebrates are an essential component of transmission.However, efficiency of transmission improves when a microcrustacean host is incorporated into the life cycle (McClelland 1995).
Table 1.List of invertebrates experimentally infected with the sealworm (Pseudoterranova decipiens).Direct transmission refers to exposing the invertebrate to the free-living infective larvae.Serial transmission refers to exposing the invertebrate to an infected precursor invertebrate host, such as a copepod.Naturally infected invertebrates have been found in both Canadian and European waters (Fig. 1).These include mysids and gammarid amphipods in eastern Canada (Scott and Black 1960, McClelland 1990, Marcogliese 1992a, 1993a, 1993b, 1996a, Marcogliese and Burt 1993, Jackson et al. 1997), and gammarid and caprellid amphipods, prawns, mysids, and polychaetes in Europe (Uspenskaya 1960, Val'ter and Popova 1974, Val'ter 1978, 1987, Lick 1991) (Table 2).In addition, infected isopods and mysids have been found in the stomach contents of Atlantic cod (Gadus morhua) in Norway and American plaice (Hippoglossoides platessoides) from Sable Island Bank respectively (Bjorge 1979, Martell andMcClelland 1995).However, until recently, little quantitative data on infections were available.No records of invertebrates infected with sealworm are available from the Pacific Ocean or the southern hemisphere.
Normally, determination of infection in invertebrates requires dissection and microscopic examination of individuals (Fig. 2), a very time-consuming and laborious process, perhaps explaining the paucity of data.Using a modified Baermann apparatus (Marcogliese 1992a, 1993a, 1993b, 1996a, Marcogliese and Burt 1993, Jackson et al. 1997), large numbers of invertebrates can be processed fresh to obtain live nematodes, a technique permitting proper fixation of parasites for morphometric study.The drawback of this system is that only total number of worms in the sample may be determined, but multiple infections per host can not.Thus, mean abundance (mean number of parasites per animal) can be determined correctly, but not necessarily prevalence (percentage of animals infected).However, abundances are usually so low in invertebrate intermediate hosts that multiple infections are extremely rare (though a single mysid containing two sealworms was found by microscopic examination from Wallace Lake, Sable Island, in 1994).Rates of infection in the natural environment tend to be in the order of 0.2-7.5 sealworms per 1000 invertebrates (Table 3).Generally, when quantita-Table 2. List of invertebrates found naturally infected with sealworm (Pseudoterranova decipiens).

Species Location Reference
Crustacea tive data are available, they suggest that mysids are more important than amphipods as intermediate hosts of sealworm in eastern Canada.Naturally infected amphipods were found only on Sable Island (Marcogliese 1993a(Marcogliese , 1993b(Marcogliese , 1996a)), the home of the largest breeding colony of grey seals in the Northwest Atlantic, and in the Northwest Arm, near Halifax, Nova Scotia (McClelland 1990).The finding of infected in-vertebrates in the Northwest Arm is puzzling because there is no seal population in that vicinity.In contrast, infected mysids were found in or on Sable Island, Sable Island Bank, Passamaquoddy Bay, St. Ann's Bay and the Bras d'Or Lakes (Fig. 3).When amphipods and mysids were collected from the same site, mean abundance in mysids was higher than in amphipods (Marcogliese 1992a(Marcogliese , 1993a(Marcogliese , 1996a)).Wallace Lake (1992) 0.5 Marcogliese (1996a) Gammarus spp.

N NE EG GA AT TI IV VE E R RE EC CO OR RD DS S
Though polychaetes have been implicated in the sealworm life cycle, no sealworm were found from 4,819 polychaetes collected from Sable Island Bank in 1989-90 (Marcogliese 1996b).Nor were sealworms found in 17,800 amphipods belonging to 36 species from Sable Island Bank and vicinity, including 3,874 specimens of Unciola irrorata; 1,254 mud shrimp (Crangon septemspinosa); 780 hermit crabs (Pagurus spp.); 1,147 cumaceans; 735 isopods; or 170 megalops of Cancer irrorata (Marcogliese 1993a, 1996b, Jackson et al. 1997).Most of these specimens were collected from areas close to Sable Island, the waters around which contain the highest levels of infection in fish in eastern Canada (McClelland et al. 1985(McClelland et al. , 1987)).Infected mysids (Neomysis americana), however, were collected from the same locality (Marcogliese 1996a).Nor were sealworms found in 3,500 amphipods from Passamaquoddy Bay, New Brunswick, includ-ing 1,047 specimens of Marinogammarus obtusatus and 1,664 of Caprella linearis, though one infected mysid (Mysis stenolepis) was found (Marcogliese andBurt 1993, Marcogliese 1996b).The two latter amphipods were collected either directly on or near harbour seal haul-out sites.In addition, no sealworm were found in 1,360 Gammarus lawrencianus and Gammarus oceanicus collected in Metis Bay, in the St. Lawrence estuary (Jackson et al. 1997).No sealworm were found from over 800 crab zoea and 294 isopods (Idotea balthica) collected in the Bras d'Or Lakes, where infected mysids also were found (Jackson et al. 1997).In addition, 405 specimens of I. balthica from Prince Edward Island, an area of high sealworm abundance in fish, were not infected (Jackson et al. 1997).
Many of the organisms mentioned in the preceding paragraph or closely related species have previously been recorded as natural or ex- perimental hosts for sealworm (Val'ter 1978, 1987, Bjorge 1979, McClelland 1990) (See Tables 1 and 2).Among amphipods, M. obtusatus and a caprellid, Caprella septentrionalis, were infected in the White Sea (Val'ter, 1978(Val'ter, , 1987)), and G. lawrencianus and U. irrorata, in the Northwest Arm, Nova Scotia (McClelland 1990).The isopod Idotea neglecta was found among fish stomach contents (Bjorge 1979).Successful laboratory infections include G. lawrencianus, G. oceanicus, U. irrorata, C. septemspinosa and crab zoea (Jarecka et al. 1988, McClelland 1990, Burt 1994).Yet, negative results may be biased due to limited sample sizes and other confounding factors.However, when comparative quantitative data are available, as from various sites in eastern Canada, mysids typically are infected with sealworm while other potential hosts often are not, the only exception being samples directly on Sable Island.These results further imply that mysids are more important than other organisms in sealworm transmission.

T TR RA AN NS SM MI IS SS SI IO ON N T TO O F FI IS SH H
On Sable Island Bank, a comparative quantitative survey of three sympatric pleuronectids with varying degrees of sealworm abundance demonstrated a strong association of sealworm infection with the amount of suprabenthic prey, and to a lesser extent epibenthic prey, in the diet (Martell and McClelland 1995).American plaice had the heaviest infections, followed by yellowtail flounder (Pleuronectes ferrugineus) and lastly, winter flounder (Pleuronectes americanus).American plaice principally consume cumaceans, amphipods, mysids, and pagurids; yellowtail flounder primarily polychaetes, cumaceans, amphipods, and tunicates; and winter flounder mainly polychaetes, crustaceans, and tunicates (Martell and McClelland 1994).
Length of sealworms from naturally infected macroinvertebrates varies from 1.1 to 9.0 mm (McClelland 1990, Marcogliese 1992a, 1993a, 1993b, Marcogliese and Burt 1993, Jackson 1995, Martell and McClelland 1995).McClelland (1995) demonstrated that a minimal size of sealworm larvae must be attained in the invertebrate host to allow successful transmission to fish.However, in his experiments larger fish tended to be refractory to infection via infected invertebrates, leading him to propose that the small larvae in invertebrates may have difficulty in establishing in larger fish because of the extensive migration required.Moreover, small nematodes are rarely found in naturally infected commercial fishes, but do occur in experimentally-infected and wild-caught small benthophagous fishes (McClelland 1995).By implication, these small fishes may be important as bridges between macroinvertebrates and large piscivorous fish (McClelland and Martell 2001).
The discovery of infected populations of threespine (Gasterosteus aculeatus) and fourspine (Apeltes quadracus) sticklebacks in brackish ponds frequented by harbour seals on Sable Island (Marcogliese 1992b) permitted examination of the sealworm life cycle and transmission dynamics in an enclosed relatively simple ecosystem consisting of few invertebrate and fish species (Marcogliese 1996a).Mysids (Neomysis americana) were more heavily infected than amphipods (Gammarus spp.), their infec tion levels being positively correlated with those in fourspine sticklebacks.Both threespine and fourspine sticklebacks fed preferentially on mysids over amphipods.Given the infection rates of N. americana and Gammarus spp.and their respective densities in Wallace Lake, densities of infected invertebrates could be calculated.The density of infected mysids (1.5 per m 2 ) was greater than that of infected amphipods (0.36 per m 2 ) in the pond.These results imply that the probability of a fish encountering an infected mysid is greater than that of an infected amphipod, despite the fact that amphipods were almost three times as numerous as mysids.By integrating the data on infection levels in invertebrates with information derived from feeding experiments, the rate of transmission of sealworm from invertebrates to fish could be determined.Transmission rate was about 100 times higher via mysids than via amphipods, and fish accumulated parasites at a rate of about 1 per month from mysids.These rates may be similar to those experienced by small benthophagous fish at sea, given that infection densities in Wallace Lake sticklebacks are similar to those of certain small benthophagous fishes on the Scotian Shelf (McClelland 1995), that my-sids are preferred prey, and that infection levels are comparable between mysids from Wallace Lake and Sable Island Bank (Table 3).
Distribution records of sealworm in invertebrate hosts from eastern Canada, both from areas of high sealworm abundance in fish and from Wallace Lake, Sable Island, suggest mysids are extremely important intermediate hosts of P. decipiens.Indeed, mysids are probably more important than amphipods in transmission of sealworm to fish, but the role of amphipods in the transmission process should neither be ignored nor underestimated.

A AC CK KN NO OW WL LE ED DG GE EM ME EN NT TS S
I thank Dr. Jane Cook for reading and commenting on the manuscript.

Fig
Fig. 1.A variety of methods are used to sample invertebrates for parasite analysis, including dip nets, traps and benthic sleds.Photo: D. Marcogliese