Insight into the Short-Finned Squid Illex coindetii (Cephalopoda: Ommastrephidae) Feeding Ecology: Is There a Link Between Helminth Parasites and Food Composition?

Predatory cephalopods, with their important role in marine food webs (Clarke, 1996) are significant intermediate/paratenic hosts in the transmission and life cycles of many parasites. Previous studies have showed their susceptibility to infection by a number of parasitic groups in the waters of all major oceans and seas (Pascual, Gestal et al., 1996; Gestal et al., 1999; González et al., 2003; Pascual et al., 2007). Many parasites, particularly those belonging to Digenea, Cestoda, and Nematoda, exploit trophic transmission of their infective stages ingested by an intermediate host, followed many times by a paratenic host in the life cycle.

Marine food webs have extremely long food chains, including also large invertebrate predators, such as chaetognaths, coelenterates, and decapods. Typically, they are more long-lived than copepods, isopods, and amphipods that may be used as first intermediate host, resulting in a necessity to develop transport or paratenic hosts as part of the parasites' life cycle. Paratenic hosts serve as a means for bridging trophic and temporal ecological gaps between intermediate and definite hosts, without engaging in parasite development into another life cycle stage (Abollo et al., 1998; Marcogliese, 2005). This permits the infective stage to persist in the environment for longer periods, increasing the chance of ingestion by the next host. Many marine parasites thus are transferred from 1 intermediate, or paratenic, host to another via predation.

Squids are especially common as paratenic hosts of helminth parasites, particularly to those that have elasmobranchs and mammals as final hosts (Hochberg, 1983). Anisakid nematode larvae and cestode plerocercoids are most frequently transferred through the trophic chain by oegopsid squids (Pascual, Gonzalez et al., 1996; Abollo et al., 1998). Although several studies on host–parasite systems with cephalopods as hosts have been completed recently (Abollo et al., 2001; Pascual et al., 2005; Mladineo and Bočina, 2007; Nigmatullin et al., 2009), our understanding of the interactions between diet, feeding behavior, parasitic disease, and transmission pathways in cephalopod–parasite systems is still incomplete.

All known exploited species of cephalopods harbor parasites—fast-growing, predatory ommastrephid squids, like Illex coindetii (Verany, 1839), are no exception (Pascual et al., 1994, 1995, 2005). This broadtail shortfin squid is an economically important species, widely distributed throughout the Mediterranean Sea, in the eastern Atlantic from the south of Britain to Namibia, and in the western Atlantic from the Caribbean, Gulf of Mexico, and to the Straits of Florida (Sánchez et al., 1998). It is a monocyclic species with a 1-yr life span (Laptikhovsky and Nigmatullin, 1993) that primarily feeds on fish, namely, small mesopelagic sternoptychids and myctophids (Lordan et al., 1998). In the Adriatic Sea, squids constitute a large portion of the total cephalopod catch, with annual landings of approximately 400 tons and short-finned squid, I. coindetii, represents a valuable demersal resource as 1 of the most important cephalopod species caught by trawlers (Ceriola et al., 2007).

However, despite its commercial significance and importance in the trophic ecology, nothing is known about the host–parasite relationships of this species in the Adriatic waters. Moreover, the only report of a cephalopod parasite in the Adriatic basin refers to Aggregata octopiana (Apixomplexa, Coccidia) infection in the reared Octopus vulgaris (Mladineo and Jozić, 2005; Mladineo and Bočina, 2007). Furthermore, there are a limited number of studies carried out on cephalopod–parasite systems in the Mediterranean waters as well (Gestal et al., 1999). Although it was reported that I. coindetii from the northwest Spain harbors several parasite species, namely, tetraphyllidean cestodes, copepod crustaceans, and an anisakid nematode (González et al., 2003), the only Anisakis spp. infestation was reported in the Tyrrhenian Sea (western Mediterranean) by Gestal et al. (1999).

Therefore, the present study was designed (1) to develop an inventory of parasitic helminths in the commercially important broadtail shortfin squid, I. coindetii; (2) to follow the population dynamics of these helminths; (3) to investigate the relationship between parasite infestation and host sex, maturity, and seasonality, with particular focus on the host feeding behavior; and (4) to identify isolated anisakid larvae at the species level with the use of molecular tools, in order to generate further insight into the parasites' zoogeography.

MATERIALS AND METHODS

RESULTS

The overall prevalences of nematode A. pegreffii and cestode Phyllobothrium sp. parasitizing I. coindetii (N  =  439) were 30.5% and 2.3%, respectively (Table I). Overall abundance of parasites was 0.7. Host size ranged from 42 to 216 mm DML (mean ± SE mean: 118.68 ± 1.25), with the smallest and largest infected squid measuring 79 and 216 mm DML, respectively. Larval stages of both parasites were mainly located in the hosts' digestive tract. Anisakid larvae were identified as third-stage larvae based on the presence of a boring tooth, an excretory pore placed below the tooth, and a short tail with a rounded tip (Grabda, 1976). Anisakis pegreffii larvae were found coiled and encysted on the external wall of stomach (43.61%) or on gonads and other reproductive organs (17.11%), and the rest of parasites were found inside the stomach (5.06%), on the digestive gland (0.72%), or free in the mantle cavity (1.69%) (Fig. 1). Plerocercoids of Phyllobothrium sp. were found in the host stomach. They were identified based on the morphology of bothridia according to Euzet (1994). Phyllobothrium sp. larvae have a scolex with 4 large bothridia with curled margins and a round anterior accessory sucker.

View Figure

• Download as Powerpoint
• Download Figure

Table I Total mean prevalence (%) with Stern's exact 95% confidence limits (CI), mean intensity and mean abundances with bootstrap 95% CI, and variance-to-mean ratio of parasites Anisakis pegreffii and Phyllobothrium sp. isolated from Illex coindetii through the seasons. For each season, data on host length are given as mean dorsal mantle length (DML) ± standard error (SE). Exponent of the negative binomial (k) showed no statistical difference between observed and expected frequencies at P  =  0.05*.

View Fullscreen

Individuals of A. pegreffii were found to be highly aggregated across host populations, with most individuals harboring low numbers of parasites and only few with many of them (Fig. 2). The greatest number of parasites (N  =  12) were harbored in 2 mature females of 195 and 154 mm DML, and 1 mature male of 154 mm DML. Prevalence, mean intensity, and mean abundance for both sex and maturity stages are presented in Table II. Fisher's exact test showed no significant differences of A. pegreffii prevalences between males and females (P  =  0.755). Furthermore, the bootstrap t-test showed no significant differences between the sexes in mean intensity (t  =  1.167, P  =  0.2665) and in mean abundance (t  =  0.711, P  =  0.459). Prevalences of Phyllobothrium sp. infecting opposite sexes were significantly different (Fisher's exact test, P  =  0.048), with a higher proportion in male hosts (Table II). The bootstrap t-test showed significant difference in mean abundance between the sexes (t  =  −2.448, P  =  0.0355).

View Figure

• Download as Powerpoint
• Download Figure

Table II Prevalence (P), mean intensity (I), and mean abundance (A) ± standard deviation (SD) (min-max) for Anisakis pegreffii and Phyllobothrium sp. infesting Illex coindetii females and males with maturity stages data pooled for sexes (1—immature; 2—maturing; 3—mature). Data on host length are given as mean dorsal mantle length (DML) ± standard error (SE).

View Fullscreen

A highly significant correlation was found between host length (DML) and abundance of A. pegreffii (Pearson's r  =  0.913, P  =  0.011), but not for Phyllobothrium spp. Although there was a positive relationship between host maturity and parasite abundances, it was not significant (for A. pegreffii, Pearson's r  =  0.975, P  =  0.141; for Phyllobothrium sp., Pearson's r  =  0.866, P  =  0.333). For A. pegreffii infection, a comparison of prevalences and mean abundances showed highly significant differences between all maturity stages (P < 0.05).

Seasonal oscillations of A. pegreffii prevalence and mean abundance are shown in Figure 3a; total values of prevalence, intensity, and abundance for each season are presented in Table I. Significant seasonal variation were detected in prevalences (Fisher's exact test: P < 0.05), mean abundances, and intensities of A. pegreffii (bootstrap t-test: P < 0.05). Anisakid infection occurred through all seasons, with lowest prevalence in summer (13.1%) and highest in autumn 2008 (55.0%) (Table I). A strong positive correlation between the mean intensity of infection and the degree of parasite aggregation (var/mean ratio) was observed (Pearson's r  =  0.987, P  =  0.002). Phyllobothrium sp. was absent in the winter of 2008, with lowest prevalence in spring 2008 (0.8%) and highest in autumn 2008 (8.3%) (Table I).

View Figure

• Download as Powerpoint
• Download Figure

Analysis of stomach contents (N  =  429) showed that I. coindetii mainly preys on fish (61.4%), crustaceans (30.0%), and cephalopods (28.4%). A total of 126 stomachs had no prey remains (29.37%). The most frequent prey species were pearlsides (Maurolicus muelleri) (16.83%), and other less frequent teleosts included capelin (Trisopterus minutus), silvery pout (Gadiculus argenteus), greater forkbeard (Phycis blennoides), and horse mackarel (Trachurus trachurus) (Table III). Fish comprised the greatest proportion of diet in all seasons, except in summer, when cephalopods represented the primary prey group (Fig. 3b). Euphausiids were eaten more frequently in summer and spring, and were absent from other seasons. Repletion ratio was smallest in autumn 2007 (51.0%) and highest in summer (90.5%) (Fig. 3b). The relative proportion of the main prey categories changed with squid size, with cephalopod prey decreasing and fish prey increasing with squid increase in size (Fig. 4). Interestingly, in summer, the lowest prevalence and abundance of A. pegreffii was observed. Comparing seasonal occurrence of nematode and prey, prevalence of A. pegreffii correlated positively only for fish prey (Pearson's r  =  0.634, P  =  0.251). The analysis of similarities (ANOSIM) of diet composition by squid sex, maturity stages, and size (DML) showed no differences (R  =  −0.008, P  =  0.903; R  =  0.003, P  =  0.287 and R  =  0.008, P  =  0.22, respectively). However, significant seasonal differences in the diet were found (ANOSIM: R  =  0.067, P  =  0.001), i.e., pairwise comparisons between the summer of 2008 and all other seasons, and the autumn of 2008 and all other seasons.

View Figure

• Download as Powerpoint
• Download Figure

Table III Percentage occurrence (% FO) of prey items recovered from stomach contents of Illex coindetii from the Adriatic Sea.

View Fullscreen

The partial CO₂ mitochondrial sequence of the isolated anisakid revealed a 498-bp sequence. Multiple sequence alignment showed that this isolate belongs to A. pegreffii (bootstrapping 2,000 generations). No ambiguities were present in the 8 isolates. Genetic divergence based on p-distance values showed that overall average between anisakid was 0.121, whereas squid isolates diverged from 0.006 to 0.027. The nucleotide frequencies of A. pegreffii were 45.7% T, 11.2% C, 21.0% A, and 22.1% G. Isolates share 491 identical pairs, and 6 transitional and 1 transversional pairs, with a ratio of 6.8. Neighbor-joining analysis generated a consensus tree resulting from 2,000 bootstrap replicates, which clustered squid Anisakis isolates within the A. pegreffii clade (Fig. 5). Tree topology rooted by P. ceticola, confirmed nematode branching in 2 main sister clades, with the first including A. paggie, A. brevispiculata, and A. physeteris, and the second polyphyletic clade that includes A. ziphidarum, A. simplex clade, as well as with A. pegreffii. All squid isolates of the nematode formed a well-supported clade with A. pegreffii, most closely related to A. simplex branch.

View Figure

• Download as Powerpoint
• Download Figure

DISCUSSION

Although Anisakis spp. and Phyllobothrium sp. have been already reported for I. coindetii inhabiting the northeastern Atlantic (Pascual et al., 1995; Pascual, Gonzalez et al., 1996; Abollo et al., 1998, 2001; González et al., 2003) and Mediterranean waters (Gestal et al., 1999), the present study is their first report in broadtail shortfin squid, I. coindetii, from the Adriatic Sea, thus characterizing the nematode larvae at the species level. Furthermore, the present study provides the first data regarding the feeding ecology of this species in the Adriatic basin. Both isolated species of the parasites have shown development at the larval stage, suggesting a paratenic nature of I. coindetii as a host species. No significant differences were found between host sex and prevalence and intensity of A. pegreffii; however, prevalences of Phyllobothrium sp. infection between sexes were significantly different. It is interesting to note the substantially higher proportion of infected males, with only a single infected female. Such observation is difficult to explain without further study in a larger host population, although it might be related to some ecological idiosyncrasy in the male population.

Observed prevalences of A. pegreffii in this study (30.5%) are somewhat higher when compared to those of Pascual et al. (1995) (10.6%), Pascual, Gonzalez et al. (1996), and Abollo et al. (1998, 2001) (11.0%) in the northeastern Atlantic. Moreover, it seems that the A. pegreffii infestation is considerably higher then that the 4.8% reported for western Mediterranean by Gestal et al. (1999). However, such a low prevalence could be due to the smaller sample size (N  =  42) and the majority of immature squid (88%) in the sample. With respect to Phyllobothrium sp., the observed prevalence (2.3%) is much lower than in the northeastern Atlantic waters where prevalences of 61.8% (Pascual et al., 1995) and 45.9% were reported (Pascual, Gonzalez et al., 1996). The lower number of the cestode's definitive hosts in Adriatic waters, or smaller availability of the first intermediate or paratenic hosts that the squid feeds upon, could attribute to the observation.

A marked difference between I. coindetii individuals in the number of parasites they harbor was observed. It is evident that A. pegreffii has a strongly aggregated distribution, which is rather common among parasites (Rohde, 1993). The anisakid infection correlated well with host size, showing increasing trend in prevalence, intensity, and abundance with host length and consequently with maturity. The smallest infected squid measured 79 mm DML, indicating that I. coindetii acquires the infection at an early age and accumulates the parasite with further growth, increasing its opportunities to acquire even greater number of anisakids. Pascual, Gonzalez et al. (1996) reported Anisakis sp. infestation in I. coindetii longer than 120 mm DML, with large mature animals harboring the greatest number of parasites (N  =  150). However, the highest number of A. pegreffii observed in the present study was only 12. This discrepancy could be due to our smaller mean squid size (118 mm DML) compared to Pascual, Gonzalez et al. (1996) (173 mm DML), because factors that directly affect parasitic infestation are host size, its life span, and dietary habits.

Broadly speaking, a helminth fauna depends on both host habitat and diet, and, therefore, clearly demonstrates the role of the host in the local food webs. Cestode plerocercoids and larval nematodes are passively transmitted through trophic interactions, and most trophically transmitted parasites are associated with a particular niche and host diet (see Marcogliese, 2005). The final hosts of tetraphyllidean cestodes are elasmobranchs, where plerocercoids are commonly found encysted in the subcutaneous blubber, usually in the abdominal area (Agustí et al., 2005), and adults occur in the spiral valves (Euzet, 1994). Anisakis spp. have a complex life cycle, passing through a number of hosts to completion in marine mammals, where they mature in the digestive tract (Grabda, 1976). Anisakid nematodes are known as extreme generalists, occurring in both pelagic and benthic hosts, which contributes to their widespread distribution (Marcogliese, 1996). Although previous studies suggested that I. coindetii acquires anisakids by ingesting infected prey, mainly euphausiids as anisakid intermediate hosts or small bony fishes as parasite paratenic hosts (Pascual, Gonzalez, et al., 1996; Abollo et al., 2001), no investigation linking this squid species feeding habits and parasite loads has been conducted. Recently, Nigmatullin et al. (2009) reported on diet composition and helminth infection of neon flying squid Ommastrephes bartramii in the southeastern Pacific, however, without assessing possible diet–parasite relationships. The diet of I. coindetii has been previously investigated in several regions, revealing its omnivorous, but primarily piscivorous, feeding behavior (Castro and Hernández-García, 1995; Rasero et al., 1996; Lordan et al., 1998). We observed the same feeding pattern; i.e., the most frequently consumed prey were fishes, namely, pearlside, followed by crustaceans and cephalopods. Overall, the diets of males and females were similar and there was no statistically significant variation in the dietary habits during squid ontogeny. However, fish and cephalopods did comprise larger part of smaller juvenile I. coindetii diet, whereas the larger specimens mostly preyed upon fish, suggesting a gradual shift from cephalopod to fish feeding with squid size increase. In most species, ontogenetic diet shifts are mainly associated with change in habitat use (Werner and Gilliam, 1984) or in size-prey selection, because of morphological changes of beaks (Castro and Hernández-García, 1995).

In the present study, we observed a highly significant seasonal variation in diet consumption, suggesting that the diet varied in response to seasonal changes in availability of prey communities. Congruent to this pattern, seasonal variation in anisakid intensity also occurred. Temporal variations in parasite intensities are rather common in highly seasonal aquatic systems and can be generated by variation in both host physiology, e.g., immune function, or host exposure to parasite infective stages that may be driven by host-related factors, such as diet (see Wilson et al., 2002). Accordingly, we found that the seasonal variations in anisakid intensity are most likely affected by dietary habits of I. coindetii. Because in each season squid of all sizes and maturity stages were sampled, the notion that observed differences in prevalence are related to specific host factors can be rejected. The highest prevalence was seen in autumn, when the fish prey, mostly M. muelleri, comprised the greatest proportion of diet, and the lowest prevalence was in summer, corresponding with the majority of cephalopods in the diet. This helped to assign the Adriatic broadtail shortfin squid not as a first, but as a second, paratenic host for the anisakid, unlike what was suggested earlier (Mattiucci et al., 2001). This is further reinforced by the observation that squids that had euphausiids in stomach contents that are not infected by anisakid or cestodian larvae.

Anisakis pegreffii also represents the first molecular characterization of the nematode at the species level in the Adriatic Sea squid. Its occurrence in the Adriatic is not surprising, since this is the most frequent anisakid species in the Mediterranean and its adjacent seas (100% prevalence in a sample of 116 Adriatic fish) (Mattiucci et al., 2008). Interestingly, Mattiucci et al. (2001) suggested that rather than Mediterranean fish, squids are the main intermediate hosts of A. physeteris—an anisakid rarely found in the Mediterranean, but abundant in the north Atlantic horse mackerel T. trachurus. However, those authors did not confirm their observation using molecular tools, and our study evidenced that short-finned squid has been the paratenic host of A. pegreffii and not A. physeteris. To our knowledge, this is the first molecular identification of an anisakid from the Adriatic squids, adding to the nematode's ubiquitous and trophic-associated character.

In conclusion, trophic relationships vary over time and space, and animals with size and age-related diet change often alter the parasites they harbor, indicating ontogenic shifts in the feeding behavior (Marcogliese, 2005). Evidence from the helminth community of the Adriatic I. coindetii population favor ontogenic shifts in its feeding behavior due to morphological change of beaks and not because of habitat change. Furthermore, the consistent presence of only 2 helminth taxa indicates decisive feeding patterns, coevolved predator–prey relationships, and stability in an ecosystem (Marcogliese, 2005). Because parasites are recognized as an important part of food-web structure, contributing in food-web networking and nestedness (Lafferty et al., 2006), we believe that further investigations of their cephalopod community as parasite hosts will not only contribute to the knowledge of food-web structure, but will elucidate other biotic relationships of squids in the Adriatic Sea as well.