Introduction
Digeneans of the family Gymnophallidae are common parasites of seabirds. Many of them use bivalves as first and second intermediate hosts (e.g., Benito et al. Reference Benito, Izagirre, Lekube, Zaldibar, Villalba, De Montaudouin, Daffe, Soto and de Cerio2023; Marchiori et al. Reference Marchiori, Quaglio, Franzo, Brocca, Aleksi, Cerchier, Cassini and Marcer2023; Montenegro et al. Reference Montenegro, Romero and González2021). The higher in the Arctic, the less we know about the transmission of gymnophallids. However, even limited existing data suggest that gymnophallid life cycles may occur there in diverse and flexible ways, presumably as an adaptation to the cold climate.
For example, a group of related Parvatrema species have developed several strategies to deal with environmental conditions in the sub-Arctic (Galaktionov et al. Reference Galaktionov, Gonchar, Postanogova, Miroliubov and Bodrov2024). Within the second intermediate host, they reproduce parthenogenetically and, oddly enough, form cercariae – but only in some species, while other abandon this free-living stage. One more shift is from inhabiting the extrapallial space to tissue parasitism. Another case is Gymnophallus choledochus Odhner, 1900, which interrupts cercariae emergence in winter and switches to metacercarial development within sporocysts in the first intermediate host (Loos-Frank Reference Loos-Frank1969). This obviously reduces the risks for the free-living stages in the most unfavourable seasonal conditions, and the question arises if this kind of life cycle adaptation also occurs in other gymnophallids.
Gymnophallus minor Ryzhikov, Reference Ryzhikov1963, was first described from the common eider in the Gulf of Anadyr (Chukotka) (Ryzhikov Reference Ryzhikov1963) and later recorded by Bishop and Threlfall (Reference Bishop and Threlfall1974) in common eiders from Newfoundland and Labrador. The maritae (sexual adults) of this species are special in having an approximately equal size of the pharynx and ventral sucker, while in all other species of Gymnophallus, the ventral sucker is about twice as large as the pharynx. Such a clear morphological differential character facilitates the accurate identification in the field, and we have repeatedly found G. minor in eiders in the Arctic and sub-Arctic. These findings have now been analysed together with molecular data and recent finds of the gymnophallid intramolluscan stages in the bivalve Liocyma fluctuosa (A. Gould, 1841) on Svalbard. This allowed us to assess the transmission pathways of this parasite in Arctic coastal ecosystems and potential life cycle adaptations.
Material and methods
We studied the morphology in maritae of Gymnophallus minor from three sources: our own sampling, material shared by colleagues, and museum collections. Sampling involved obtaining wild birds in accordance with local regulations, dissection, and recovery of worms.
We identified the maritae of G. minor from common eiders Somateria mollissima (Linnaeus, 1758) in the southwest of West Spitsbergen (Bellsund Fjord; 77°30’N 14°35’E) in 1990; from common eiders in the Pechora Sea (off the Vaygach Island; 69°50’N 59°26’E) in 2010 and 2017; and also from the king eider Somateria spectabilis (Linnaeus, 1758) in the Pechora Sea (Khaipudyr Bay; 68°43’N 60°3’E) in 2020. We preserved these worms in 70% ethanol, stained them with boric carmine, and whole-mounted. The mounts were deposited in the Collection of Helminths, section Trematoda, of the Zoological Institute, Russian Academy of Sciences (RAS) (ZISP, no. 3745, 3746, 3748, 3749).
Two specimens of ethanol-preserved gymnophallid maritae from the common eider collected in Ivanovskaya Bay, East Murman (Barents Sea) were kindly provided by Vadim V. Kuklin. These were as well stained with boric carmine, whole-mounted, and deposited in the ZISP collection (no. 3747).
We also studied the collection of the whole-mounted gymnophallid maritae from common eiders in Iceland, kindly provided by Karl Skírnisson, and the syntypes of G. minor from the trematode collection of the A. N. Severtsov Institute of Ecology and Evolution (IEE) RAS (no. 141, 863–870).
One of the two G. minor specimens collected from the king eider in Khaipudyr Bay (Pechora Sea) in 2020 was used for DNA extraction using Chelex 100 (BioRad, USA) (for protocol, see, for example, Krupenko et al. Reference Krupenko, Kremnev, Gonchar, Uryadova, Miroliubov, Krapivin, Skobkina, Gubler and Knyazeva2022).
Gymnophallid intramolluscan stages were found in the bivalve Liocyma fluctuosa (A. Gould, 1841) collected in September 2016 and March 2017 at tidal and subtidal sites near Ny-Ålesund (Svalbard): sites A, B, subtidal, 78°55’20’’N, 11°58’30’’E; site E, tidal, 78°55’35’’N, 11°54’35’’E. They were photographed and used for DNA extraction with a DNeasy Blood & Tissue Kit (Qiagen, Hilden, Germany). The sample codes A49, A54, B6, E1, E23, and E54 correspond to different infected L. fluctuosa individuals collected at sites A, B, and E.
We amplified the partial 5.8S rDNA and the complete ITS2 using the primer pair 5.8S-ITS2 and 28S-ITS2, as described in Cremonte et al. Reference Cremonte, Gilardoni, Pina, Rodrigues and Ituarte2015. For the partial 28S rDNA, we used two primer pairs: digl2 and 1500R (Olson et al. Reference Olson, Cribb, Tkach and Littlewood2003; Tkach et al. Reference Tkach, Grabda-Kazubska, Pawlowski and Swiderski1999), and C2’B and D2 (Bayssade-Dufour et al. Reference Bayssade-Dufour, Jouet, Rudolfova, Horák and Ferté2006; Vân Le et al. Reference Vân Le, Lecointre and Perasso1993), and PCR conditions as described in Krupenko et al. Reference Krupenko, Uryadova, Gonchar, Kremnev and Krapivin2020. For the partial 18S rDNA, we used two pairs of primers covering the overlapping fragments: 652 and 28, and 18S1A and 32 (Hernández-Mena et al. Reference Hernández-Mena, García-Varela and de León2017). PCR products (for the parasite samples originating from the bivalves only) were cleaned using a Qiaquick PCR purification kit (Qiagen, Hilden, Germany). Sequencing was performed with the PCR primers in both directions by an automated ABI 3500xl genetic analyser (Applied Biosystems, USA) at the Research Resource Centre ‘Molecular and cell technologies’ (St. Petersburg University) and Eurofins Genomics (Ebersberg, Germany).
Chromatograms were inspected, analysed, edited, assembled, and aligned in Geneious Prime (https://www.geneious.com). For the comparisons, we used relevant sequences from GenBank that matched the lengths of the newly obtained ones. To estimate the phylogenetic position of our samples, we used the maximum likelihood (ML) analysis implemented as RAxML-NG v. 1.1.0 (Kozlov et al. Reference Kozlov, Darriba, Flouri, Morel and Stamatakis2019) in raxmlGUI 2.0 (Edler et al. Reference Edler, Klein, Antonelli and Silvestro2021), with ‘ML + transfer bootstrap + consensus’ approach, random seed 317646, and 10,000 bootstrap replicates. The best substitution model was estimated prior to the analysis in the same program with ModelTest-NG (Darriba et al. Reference Darriba, Posada, Kozlov, Stamatakis, Morel and Flouri2020). We also performed the Bayesian inference (BI) analysis in MrBayes3.2.7a (Ronquist et al. Reference Ronquist, Teslenko, Van Der Mark, Ayres, Darling, Höhna, Larget, Liu, Suchard and Huelsenbeck2012) with 1,000,000 iterations and the ‘allcompat’ option. The dataset was aimed to balance between the alignment length and the number of species: we concatenated two fragments, the partial 18S rDNA (the best substitution model TIM2+I+G) and 5.8S–ITS2–28S rDNA (GTR+G). In MrBayes, TIM2 was replaced by GTR.
Results and discussion
We recovered G. minor maritae from the eiders in the Barents and Pechora Sea on several occasions, but data on infection prevalence and intensity are limited. In Bellsund Fjord in 1990, one adult female S. mollissima was dissected and found infected, intensity ca. 70 individuals (two dissected ducklings were uninfected). Off the Vaygach Island in 2010 and 2017, five adult female S. mollissima were dissected, two of them infected, intensity 2–9 individuals (five dissected ducklings were uninfected). In the Khaipudyr Bay, eight adult female S. spectabilis were dissected, four of them infected, intensity 1–4.
The studied maritae varied in size (Table 1). We believe these variations could be introduced while handling material (i.e., the extent of the body contraction during the fixation, and the extent of coverslip pressure during mounting) or have natural origins (marita age, host- and geography-induced variation). Their morphological characters were uniform and complied with the diagnosis from the original G. minor species description by Ryzhikov (Reference Ryzhikov1963). The brief description below is based on our specimens.
Morphometric parameters of Gymnophallus minor maritae from different hosts and geographic regions
N, number of measured individuals. *, data after Ryzhikov (Reference Ryzhikov1963).
Worms oval, oral sucker subterminal, about 1.5–2 times larger than ventral sucker (Table 1, Figure 1a, b). Pharynx large, roundish, roughly equal in size to ventral sucker. Seminal vesicle bipartite. Ovary pre-testicular, oval. Testes symmetrical, left testis slightly larger than right one. Vitellaria paired, follicular, located at sides and slightly posterior to ventral sucker. In young worms, uterus loops with eggs only in forebody, anteriorly to intestinal caeca. When filled with eggs, uterus loops moving posteriorly at sides of body, reaching approximately level of testes. Eggs oval, about 1.5 times longer than wide.
A–C, Representative microphotographs of life cycle stages of Gymnophallus minor: phase contrast view of the marita from the slide ZISP 3745 (host Somateria mollissima, Spitsbergen) (A); phase contrast view of the marita from the slide ZISP 3748 (host Somateria spectabilis, Pechora Sea) (B); live view of a sporocyst containing metacercariae along with cercarial embryos and developing cercariae (C); abbreviations: ph, pharynx; m, metacercariae; ov, ovary; t, testis; sv, seminal vesicle; vs, ventral sucker; scale bars: A, B—100 μm; C—300 μm. D, ML phylogenetic tree based on concatenated fragments of 18S and 5.8S–ITS2–28S rDNA (1,722 bp); bootstrap support values printed at nodes, followed after slash by the posterior probabilities inferred in the Bayesian analysis of the same dataset; scale bar shows substitutions per site; sequences obtained from GenBank are given with their accession numbers and originate from Shchenkov et al. Reference Shchenkov, Sokolov, Tsushko and Denisova2022 (G. deliciosus) and Cremonte et al. Reference Cremonte, Gilardoni, Pina, Rodrigues and Ituarte2015 (all other species).
We detected no marked differences between the G. minor maritae from our samples and the syntypes from the IEE collection. Gymnophallid maritae from the common eiders from Iceland (collection of K. Skírnisson) also matched G. minor in their morphological characters.
Since DNA sequences were obtained from a single marita individual, the possibility of dealing with a mixture of species cannot be excluded – even though morphological data on all the examined maritae were consistent. Among trematodes, cryptic diversity is a common feature, probably found more frequently than in other helminths (Pérez-Ponce de León and Poulin, Reference Pérez-Ponce de León and Poulin2018). This should be kept in mind during future studies, but currently there are no contradictions to calling the species in question G. minor and to interpreting its life cycle the way we do below.
Gymnophallid sporocysts containing cercariae (Figure 1c) were found in the hepatopancreas of 11 out of 48 (23%) specimens of L. fluctuosa. These sporocysts were vermiform and moved only slowly. Some sporocysts also contained metacercariae (Figure 1c) which were similar in size to those observed in the extrapallial space of these molluscs. For molecular studies, we used material from six infected L. fluctuosa: sporocysts, cercariae, and metacercariae samples from B6 and E1; sporocyst samples from A54 and E54; and cercariae samples from A49 and E23. Intensity of metacercariae infection for B6 exceeded 500 individuals.
To genetically match the life cycle stages, we obtained two 5.8S–ITS2 (536 bp) and 11 28S rDNA (1256–1280 bp) sequences (submitted to GenBank under the accession numbers PQ526380–PQ526381 and PQ526368–PQ526378, respectively). The sequences were identical among the marita, sporocyst, cercariae, and metacercaria samples, with two nuances. First, sequences from samples A49, A54, and E1 had a single T peak instead of two distinct peaks (TT in other samples); this probably was an artefact. Second, there were three cases of clear double peaks consistently seen in both forward and reverse reads: two in 28S rDNA (marita and E54) and one in 5.8S–ITS2 (E54). We report these double peaks using IUPAC codes, as it was suggested in Blasco-Costa et al. (Reference Blasco-Costa, Cutmore, Miller and Nolan2016). In all double peak cases, one peak matched the nucleotide found in all the other sequences, and one peak differed. There are several reports of such polymorphism in trematodes – for example, in 28S rDNA (Achatz et al. Reference Achatz, Curran, Patitucci, Fecchio and Tkach2019), ITS2 (Diaz et al. Reference Diaz, Bray, Cutmore, Ward and Cribb2015; Womble et al. Reference Womble, Orélis-Ribeiro and Bullard2015), both ITS1 and ITS2 (Ahasan et al. Reference Ahasan, De Elías-Escribano, Artigas, Alam, Mondal, Blair, Chowdhury, Bargues and Mas-Coma2023). Though it may indicate hybridization (Ahasan et al. Reference Ahasan, De Elías-Escribano, Artigas, Alam, Mondal, Blair, Chowdhury, Bargues and Mas-Coma2023), we could not test this. Overall, we consider that the observed subtle variations do not violate the otherwise complete similarity among the sequences.
ITS2 sequences are generally conservative within species in Gymnophallidae, with a couple of modest exceptions (Galaktionov et al. Reference Galaktionov, Gonchar, Postanogova, Miroliubov and Bodrov2024) and one case with suspected cryptic diversity (Feis et al. Reference Feis, Thieltges, Olsen, de Montaudouin, Jensen, Bazaïri, Culloty and Luttikhuizen2015). The interspecific divergence in the ITS2 marker among three species of the genus Gymnophallus (our data for G. minor; GenBank data for G. australis, JN381028, and G. choledochus, JN381028) is 70 to 77 substitutions per a 550-bp alignment. Even with a single polymorphic position, our data fall within the expected intraspecific level. The same is true for the 28S rDNA data.
The DNA sequencing data therefore suggest the conspecificity of the marita, sporocyst, and metacercaria samples. Thus, the life cycle of G. minor was elucidated, with the bivalve L. fluctuosa as the first and second intermediate host, and the eider ducks S. mollissima and S. spectabilis as the definitive host species (perhaps among some other marine anatids). This conclusion is consistent with data on the spatial distribution patterns of the host species.
Most likely, G. minor is a circumpolar boreal–Arctic species, with the distribution range generally limited by that of its first intermediate host, L. fluctuosa. This bivalve is found in all Siberian seas, in the North Pacific and the North Atlantic. Eiders, the definitive hosts of G. minor, are also common across this range. However, G. minor seems not to reach the high Arctic because we have not found it at Franz Joseph Land (Galaktionov et al. Reference Galaktionov, Węsławski and Stempniewicz2021), though L. fluctuosa has been registered there (Golikov and Scarlato Reference Golikov, Scarlato and Scarlato1977). One explanation could be the low density of L. fluctuosa in this area (1–10 individuals/m2; Golikov and Scarlato Reference Golikov, Scarlato and Scarlato1977) in comparison to southwestern Spitsbergen (an average of 185.2 or 190 individuals/m2, depending on the sediment type; Różycki and Gruszczyński Reference Różycki and Gruszczyński1991). Also, the latitude does not favour transmission via the free-swimming cercariae which are sensitive to harsh environmental conditions (Galaktionov Reference Galaktionov2017). Only the life cycle of Microphallus pseudopygmaeus Galaktionov, 1980, without any free-living stages, is completed there (Galaktionov et al. Reference Galaktionov, Węsławski and Stempniewicz2021).
Although some conditions appear to be too cold for G. minor to persist, in less severe conditions, it shows considerable flexibility in the life cycle. Its sporocysts produce swimming cercariae that can facilitate transmission by spending several hours in the external environment and infecting the second intermediate host. However, our observations indicate that cercariae may also stay inside the first intermediate host and develop into metacercariae there. Finally, it was not uncommon to observe metacercariae inside the sporocysts (Figure 1c). So, we hypothesize that the life cycle may follow either the two-host or three-host scenario. In the first case, transmission is guaranteed if the definitive host eats the first intermediate host. In the second case, free-swimming cercariae enhance dispersal by encysting in other bivalves of the same or, potentially, other species. In addition to L. fluctuosa, gymnophallid metacercariae were found near Ny-Ålesund in Macoma calcarea, Astarte sp., Eunucula tenuis, Yoldia hyperborea, Nuculana pernula, and most commonly in Mya truncata (Mathias Wegner, unpublished). Some of them could serve as the alternative second intermediate hosts for G. minor, but we do not yet have evidence to support this.
This resonates with the observation of Loos-Frank (Reference Loos-Frank1969) that in winter at the North Sea, cercariae of G. choledochus do not emerge from their host Cerastoderma edule (Linnaeus, 1758). Instead, they develop into the metacercariae inside the sporocysts, perhaps because chances for transmission by free-swimming cercariae are scarce in the cold season (Loos-Frank Reference Loos-Frank1969). This explanation makes sense for G. minor, too, but applies to conditions in Ny-Ålesund not just in winter but throughout the year. Such an adaptive mechanism may be common in gymnophallids. Notably, this phenomenon does not occur in temperate regions: for example, gymnophallids in the Mediterranean seem to never harbour metacercariae within the sporocysts (Bartoli and Gibson Reference Bartoli and Gibson2007).
To complete the story, we confirmed the phylogenetic position of G. minor within the genus Gymnophallus (Figure 1d). The partial 18S rDNA sequence obtained for the marita sample was 1763-bp long (GenBank accession number PQ526379). The concatenated alignment after trimming was 1722 bp (778-bp long 18S rDNA partition and 944 bp-long 5.8S–ITS2–28S rDNA partition). The topology in the ML and BI trees was consistent, with the comparable support values (Figure 1d). Five species of Gymnophallus formed a well-supported monophyletic clade, with G. minor being the sister-species of G. minutus.
To sum up, we have discovered that the bivalve species L. fluctuosa very likely serves as first and second intermediate host of G. minor. Availability of alternative life-cycle scenarios, involving two or three hosts, may enhance the transmission of this parasite in extreme environmental conditions. Our data contribute to a better overall understanding of the bird parasites’ circulation in Arctic ecosystems.
Acknowledgements
We thank our colleagues Vadim Kuklin (Murmansk Marine Biological Institute, Russia) and Karl Skírnisson (University of Iceland, Keldur) for sharing their material with us. Our fieldwork was possible thanks to the facilities of the White Sea Biological Station and the R/V ‘Professor Vladimir Kuznetsov’ (Zoological Institute of the Russian Academy of Sciences) and the Educational and Research Station ‘Belomorskaia’ (St Petersburg University). We thank Eike Petersen for help with the molecular work. We acknowledge the research resource centre ‘Molecular and cell technologies’ (St Petersburg University) for the sequencing facilities.
Financial support
This study was funded by the Russian Science Foundation (K.G. and A.G., grant number 23-14-00329). Primary sample processing, including DNA extraction, was supported by the research programme of the Zoological Institute of the Russian Academy of Sciences (project number 125012800903-5).
Competing interest
None.
Ethical standard
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional guides on the care and use of laboratory animals.
