Distribution of ascaridoid nematodes (Nematoda: Chromadorea: Ascaridoidea) in fish from the Barents Sea

Nematodes were isolated from the host’s tissue, washed in deionized water, and preserved in 70% ethanol. Each parasite was divided into 3 parts. The anterior and posterior parts were preserved in 70% ethanol with 5% glycerol. Nematodes were cleared by evaporation of the mixture of ethanol and glycerin, mounted and examined under a light microscope. The identification of worms was based on key morphological characters, i.e. the structure of labiae and ventriculus, the tail shape, the absence or the presence of the mucron, and the position of the excretory pore (Fagerholm 1991). The middle sections of specimens examined were fixed in 70% ethanol and used for molecular identification. Parasites that were excessively damaged for microscopic observations, or difficult to identify using morphological characters, were identified using PCR-RFLP. The same procedure was applied to 25% of nematodes to confirm the morphological identification (Zhu et al. 2000; Kijewska et al. 2002; Dzido 2011).

After the morphological identification, the nematode DNA was isolated as described by Hoarau et al. (2002). Due to the small abundance of Pseudoterranova decipiens s.l., all specimens of this species were identified using molecular characters.

Because of their large numbers, 27% of Contracaecum osculatum s.l., 5% of Hysterothylacium aduncum, and 2% of Anisakis simplex were identified using PCR-RFLP. The low proportion of A. simplex examined based on DNA was supported by the results of our previous study in the Barents Sea (Karpiej et al. 2013), during which only a single species representing the A. simplex complex was found in the study area, namely A. simplex s.s. In total, parasites from 74 fish individuals (48 Greenland halibut and 26 specimens of by-catch species) were identified using molecular characters.

Amplification of the rDNA marker was carried out according to Zhu et al. (1998). PCR products were separated electrophoretically on 1% agarose gels and visualized by staining with ethidium bromide. The PCR products (ITS1-5.8S-ITS2 fragment of rDNA) were digested with Alul, Rsal, TaqI, Hhal, Hinfl, and BsuRI restriction enzymes (Fermentas, Vilnius, Lithuania). The sibling species of the C. osculatum complex were determined by digesting the rDNA region with RsaI and XbaI (Fermentas, Vilnius, Lithuania) separately (Table 1). The products were separated electrophoretically on 4% agarose gels and visualized by staining with ethidium bromide (Sambrook et al. 1989). The pUC Mix Marker (0.5 μg μl^-1; Marker 8, Fermentas) was used as a molecular weight marker. The obtained patterns were compared using a molecular key (Zhu et al. 2000; Kijewska et al. 2002; Dzido 2011).

Table 1

Digestion patterns given as approximate lengths of the main DNA fragments (bp), which characterize three sibling species of C. osculatum s.l. The missing fragments are represented by short and not diagnostic fragments (total length).

All fish were divided into the following groups: Greenland halibut (overall), Greenland halibut (“deep” water samples), Greenland halibut (“shallow” water samples), and by-catch species. The prevalence and mean abundance (Bush et al. 1997) were calculated for each group separately. All statistical procedures were performed using STATISTICA 12.0 (StatSoft, USA). The non-parametric test (Spearman’s rank correlation) was used because the distribution of parasites deviated from the normal one. The infection abundance–fish length covariance and correlation were examined as well. In addition, the data were tested for differences in the infection intensity and abundance of parasites between the two areas sampled (the northern and the southern one) and two depth categories surveyed (“deep” and “shallow”).

The most abundant ascaridoid in the Barents Sea was A. simplex, present in all fish species examined. The prevalence of this parasite in Greenland halibut was lower than that observed by Wierzbicka (1992) (81.1% vs. 95.3%), which could be caused by different seasons of sampling performed in the Barents Sea (autumn vs. spring). Michalsen & Nedreaas (1998) observed differences in the Greenland halibut stomach content depending on the sampling season: blue whiting, Micromesistius poutassou (Risso, 1827), the squid Gonatus fabricii (Lichtenstein, 1818), unidentified fish, herring, Clupea harengus Linnaeus, 1758, and Greenland halibut dominated in autumn, while G. fabricii, heads of Greenland halibut as offal from fisheries, shrimps, and cod remains (possibly offal) dominated in spring. The comparison of data from the same region with a five-year interval (Karpiej et al. 2013 vs. recently reported results) enables us to assume that the Greenland halibut infection with A. simplex s.s. in this part of the Barents Sea has been slightly declining (prevalence 86.2% and abundance 39 according to Karpiej et al. 2013 vs. 81.1% and 36 determined in this work). Dwyer et al. (2010) suggested that many factors such as year, length, and sex may cause a different composition of the halibut diet. Moreover, miscellaneous items in the halibut diet during different seasons may determine the parasite fauna of this species.

It is very likely that the diet of halibut depends on many factors, including the depth and the region, which in turn can affect the anisakid infection level. In the Greenland halibut caught in the “shallow” waters, the prevalence of A. simplex s.s. was only slightly different (10%) compared to the halibut caught in the “deep” waters of the Barents Sea. On the other hand, the abundance of A. simplex s.s. was over three times higher in the group of halibut from the “shallow” waters (Table 2) compared to that from the “deep” waters. This observation is consistent with the fact that intermediate hosts of A. simplex s.s. are primarily pelagic organisms. According to Dalpadado & Skjoldal (1996), the distribution of dominant species of Arctic krill (Thysanoessa longicaudata and T. inermis) – the first intermediate hosts for A. simplex (see Smith 1971), shows their concentration south of the Svalbard Bank (below 76°N). This pattern of distribution affects the accumulation of parasites in secondary intermediate and paratenic hosts, which are the prey for bigger piscivorous fish. Moreover, a significant increase in the abundance of A. simplex in relation to the increased length of Greenland halibut suggests that this parasite is accumulated also in Reinhardtius hippoglossoides, known to change its diet with the age and size of the fish (Karpiej et al. 2013). The smaller fish acquire A. simplex s.l. by consuming crustaceans and cephalopods, which are the initial paratenic hosts of this nematode. Subsequently, halibut individuals gradually consume more fish as they grow, i.e. paratenic hosts providing anisakid larvae (Karpiej et al. 2013). Moreover, large individuals of Greenland halibut, generally distributed at greater depths than the small ones (Bowering & Chumakov 1989), had different prey items in the diet, with a higher contribution of offal and larger fish.

This anisakid species was dominant in not-benthic species of fish (prevalence of 100% in G. morhua and over 70% in other species). This observation is valuable from a practical aspect – A. simplex s.s. and other anisakids pose a consumer health hazard if accidentally eaten alive. People should be aware that the consumption of some fish species is associated with the risk of anisakidosis. Moreover, this anisakid may adversely affect the product quality if abundant in the raw material, which may result in economic losses.

In the Northern Hemisphere, adult P. bulbosa occurs in the bearded seal in both the North Atlantic and Pacific Oceans, while its larval stages infect the flatfish such as the American plaice, Greenland halibut, Atlantic halibut Hippoglossus hippoglossus (Linnaeus, 1758), fourhorn sculpin Myoxocephalus quadricornis (Linnaeus, 1758) and rarely the Pacific cod Gadus macrocephalus Tilesius, 1810 (see Bristow & Berland 1992; Brattey & Davidson 1996; Mattiucci et al. 1998). The hosts identified so far for P. bulbosa are mostly benthic and demersal, with the exception of G. morhua, which is benthopelagic (Paggi et al. 1991). This supports the hypothesis that P. decipiens s.l. in the Northern Hemisphere has a benthic life cycle with larvae not being able to swim (Bristow & Berland 1992). Moreover, the results of the presented study confirm that, contrary to P. decipiens A and B, the intermediate stage of P. bulbosa occurs primarily in the liver of the American plaice (H. platessoides) and the Greenland halibut (Paggi et al. 1991). The distribution ranges and food preferences of these two pleuronectids overlap (Bristow & Berland 1992). Of all the fish studied from the Barents Sea, the American plaice, which is a strongly demersal species, was characterized by the highest prevalence and abundance of P. bulbosa (Table 2). Nevertheless, the observed values were lower than the prevalence and abundance values calculated by Martell & McClelland (1995) for H. platessoides sampled on Sable Island Bank. The Atlantic cod was the second species of fish where P. bulbosa is very frequently found. In this fish, the prevalence of the above worm reached 50.0%, which is much higher than observed for P. decipiens s.l. in G. morhua by Sobecka et al. (2011) in the area of Bear Island. Moreover, the Atlantic cod was the most heavily parasitized species among all fish examined during this study, with the infection prevalence equal to 100% for A. simplex s.s., the C. osculatum complex, and H. aduncum. The above-mentioned observation could be an effect of continuous accumulation of parasites in the Atlantic cod and ontogenetic shift in their diet (Link et al. 2009). The mean length of the cod analyzed by Sobecka et al. (2011) was shorter than the length determined for specimens examined in this study (weight: 3478 vs. 4478 g and length: 738 vs. 420 mm), and the study carried out by Sobecka et al. (2011) was limited to 300 m, while our study – to 660 m. Intensification of parasitic infections may depend on many factors, but most likely it represents a continuous accumulation of parasites in larger cod (Mouritsen et al. 2010).

A similar situation applies to H. aduncum whose larvae were found in many fish species, including capelin Mallotus villosus (Müller, 1776) (see Levsen et al. 2016) and a wide spectrum of other fish species (Balbuena et al. 1998; 2000). These species probably transmit H. aduncum directly to Reinhardtius hipopoglossoides or indirectly to piscivorous fish eaten by Greenland halibut. The list of its prey items contains more than 20 species of fish belonging to Gadiformes, Clupeiformes, Scorpaeniformes, etc. (Froese & Pauly 2017). Infections with H. aduncum in the Barents Sea are common and the parasite usually completes its life cycle in fish that feed on crustaceans. This assumption is supported by the observation that H. aduncum was recorded only in smaller Greenland halibut that feed on crustaceans. Only 4 specimens (abundance = 0.04) were found in 110 individuals of Greenland halibut, whereas the abundance was higher (2.5) in other fish species. On the other hand, H. aduncum was numerous in Atlantic cod, which in the Barents Sea feeds mainly on hyperiids and small fish (Dolgov et al. 1992). Also many S. mentella (42%) were infected with H. aduncum, which is caused by the diet of this species, containing mainly euphausiids, hyperiids, cephalopods, chaetognaths, and small fish (Froese & Pauly 2017).

McClelland et al. (1985) suggested that C. osculatum s.l. competed with the proliferation of P. decipiens s.l., because the seasonal and geographical reduction in the number of P. decipiens s.l. in gray seals often coincided with increases in the abundance of C. osculatum s.l. This negative correlation between the abundance of the two anisakid nematodes was not supported by the data reported by Brattey & Stenson (1993). According to the aforementioned authors, the differences may well result from dissimilarities in the diet among gray seals, as they tend to acquire infection with larval P. decipiens s.l. by preying more often on benthic fish (McClelland et al. 1990). Others are more likely to feed on pelagic fish species, such as capelin or cod that harbor larvae of C. osculatum s.l. (see Palsson & Beverly-Burton 1984). In our study, the proportions between these two anisakid species in the examined fish are clearly visible and associated with their diet and biology. We observed that C. osculatum s.l. is very numerous in both species, i.e. the Atlantic cod and the Greenland halibut, and also strongly parasitized the American plaice (Table 2), while P. bulbosa occurs in the same fish species in smaller numbers. Compared to the results by Karpiej et al. (2013), we also observed the increase in the Greenland halibut infection with these two parasites (abundance for C. osculatum s.l. 7.0 vs. 11.2 and for P. bulbosa 0.2 vs. 0.5). The said observation appears to contradict the previous assumption by McClelland et al. (1985).

It is interesting to note that fish species infected predominantly with the C. osculatum complex are mainly demersal or benthopelagic feeders. Moreover, the highest mean intensity of infection with the C. osculatum complex was observed in the longest specimens of the American plaice (39–44 cm). Also in the case of Greenland halibut, the mean intensity of infection was mostly observed in fish with a length of 38–48 cm. According to Rodríguez-Marín et al. (1995), crustaceans, mollusks and small fish accounted for a significant part of the diet of Greenland halibut specimens with a length (TL) ranging from 40 to 49 cm. We cannot exclude the possibility that some of invertebrates, too big for shorter fish, are the main vector for Contracaecum sp. Based on the list of food items in the fish diet and the size of invertebrates as a limit for shorter fish, we suspect that representatives of Isopoda and Polychaeta are intermediate or paratenic hosts for the C. osculatum complex. Usually they live at the bottom, even below 1000 m, and they are widespread in most oceanic waters around the world, including waters with reduced salinity like the Baltic Sea. An alternative hypothesis assumes that C. osculatum is acquired by halibut indirectly, by consuming fish feeding on small invertebrates, like gammarids being a dietary component of many fish, including Greenland halibut from the length class of 39–49 cm (Hovde et al. 2002). These invertebrates are widely distributed and could transmit larvae of C. osculatum A to a wide spectrum of secondary hosts. This could explain why the distribution and abundance of C. osculatum A does not depend on the location or the depth. The presented hypothesis also explains the fluctuation in the abundance of C. osculatum A according to the length classes. The first peak of abundance for the shortest length class may result from feeding on invertebrates (Michalsen & Nedreaas 1998). The second peak observed in the higher length class could be the effect of the accumulation of parasites and feeding on smaller fish species. The prey species included M. villosus, which is one of the most preferred species in the diet of the Greenland halibut (Bowering & Lilly 1992). This species, abundant in the Barents Sea, is carrying the larval stages of C. osculatum B (Levsen et al. 2016). The size of capelin suggests that only halibut representing the middle and upper length classes can feed on capelin. Contrary to the results of Karpiej et al. (2013), the northern halibut population seems to be less exposed to C. osculatum B larvae. The inconsistency of the results obtained in the five-year period may result from the dynamics of environmental conditions, especially those governed by the Gulf Stream (Sato et al. 2014). It also supports the hypothesis that the acquisition of C. osculatum B depends on the distribution of intermediate hosts as well. This explains why the abundance of C. osculatum B remains on the low level but consequently increases with the length of Greenland halibut.

The distribution of parasites in different hosts depends on the diet composition determined by the depth zone and the spectrum of preferred paratenic hosts occurring at each geographical location.