Use of stable isotope ratio analysis to investigate the biology and clinical significance of seal parasites

Abstract

Stranded seals are often infected with a broad range of parasites, although whether they are the cause of significant morbidity or an incidental finding is usually unclear. In this study we used stable isotope ratio analysis, a method frequently used to investigate food webs, to explore the extent to which common seal parasites feed directly on host tissue and fluids or absorb host-derived metabolites, which in turn may give an indication of their potential impact on the host's health. The trophic discrimination factor Δ¹⁵N for the nasal mite, Halarachne halichoeri, was mostly positive, ranging between −0.015 and 3.2‰ (n = 6), while for the Acanthocephalan worm, Corynosoma strumosum and the anisakid nematode, Pseudoterranova decipiens, Δ¹⁵N ranged between −4.2 and −2.0‰ (n = 7), and between −6.7 and −0.8‰ (n = 5) respectively. In the case of the lungworm, Otostrongylus circumlitus, Δ¹⁵N measured between −5.6 and 0‰ for worms collected in the stomach (n = 5), between −1.1 and 0.2‰ for worms collected from the heart (n = 3), between −0.7 and 1.9‰ for worms situated in the lungs (n = 4). Based on Δ¹⁵N, parasites could be clearly divided into those that were on a higher trophic level than their host suggesting a predator–prey-like relationship, and those that were not. It is hypothesized that Δ¹⁵N may be indicative of the clinical significance of parasite–host associations.

Introduction

Stable isotope ratio analysis, a methodology conventionally used to investigate trophic relationships within food webs, is based on the fact that the tissues of animals are enriched, at a given rate, with heavy nitrogen (¹⁵N) and carbon (¹³C) isotopes through food consumption, assimilation and excretion. As a result, the ratios of heavy-to-light isotopes in tissues, expressed as δ ¹⁵N and δ ¹³C (‰), are specific to the individual and determined by their diet. Consumers and their food sources can be inferred by determining the difference of their respective δ ¹⁵N and δ ¹³C values, the so-called trophic discrimination factor, Δ (Post, 2002; Caut et al., 2009). A review of over 260 animal-diet discrimination factors reported overall mean values of 2.75‰ for Δ¹⁵N and 0.75‰ for Δ¹³C; however, the authors stressed that trophic discrimination factors, such as the ratios of heavy and light isotopes themselves, varied significantly between animal taxa, the environment and animal inhabits (i.e. terrestrial, freshwater or marine habitat) and the tissues that were being examined (Caut et al., 2009). In recent years, stable isotope ratio analysis has been employed to investigate the trophic relationships between hosts and their parasites, initially with the expectation to record trophic discrimination factors indicative of classical predator–prey relationships, i.e. a Δ¹⁵N of around 3.4‰ and a Δ¹³C of 0.5–1‰. However, such Δ¹⁵N and Δ¹³C values turned out to be the exception rather than the rule, as many parasite–host relationships are characterized by low, zero or even negative Δ¹⁵N and Δ¹³C values. Moreover they vary with parasite taxa, with arthropod parasites generally showing enriched Δ¹⁵N and Δ¹³C values in relation to their hosts, and cestodes’ depletion (Thieltges et al., 2019). These trophic associations are generally thought to be a reflection of the particular feeding habits and metabolic processes of parasites (Table 1) (Nachev et al., 2017; Thieltges et al., 2019; Kamiya et al., 2020; Riekenberg et al., 2021). It is likely that a better understanding of the trophic relationships between parasites and hosts will provide information regarding the parasites’ life style and the likelihood that they cause disease in their host.

Table 1. Proposed explanations for typical and atypical trophic discrimination factors (Δ¹⁵N and Δ¹³C) in host–parasite relationships

Adapted from Riekenberg et al. (2021).

To our knowledge this is the first study that employs stable isotope ratio analysis to investigate the trophic relationships of grey seals (Halichoerus grypus) and harbour seals (Phoca vitulina) and their parasites. We focused on nasal mites (Halarachne halichoeri), Acanthocephala (Corynosoma strumosum) and the anisakid and metastrongyloid nematodes, Pseudoterranova decipiens and Otostrongylus circumlitus as they are some of the most common parasites of pinnipeds in the North Atlantic. Halarachne halichoeri inhabits the (upper) respiratory tract of its host. It has a direct life cycle and is transmitted by direct contact or sneezing. In contrast, the 3 worm species have indirect life cycles. Seals become infected by ingesting crustaceans or benthic fish that serve as intermediate or paratenic hosts. Corynosoma strumosum inhabits the small intestine, P. decipiens the stomach and O. circumlitus the lung of its final host (Measures, 2001; Raga et al., 2009; Alonso-Farré et al., 2012). In spite of their ubiquity, several important knowledge gaps remain regarding the biology of these parasites, such as the extent of their tissue migration, when and how they derive nutrition during their life cycles and the degree by which they cause disease in their host. This is particularly the case for O. circumlitus, the lungworm which has been associated with significant respiratory disease if present in large numbers and/or concurrently with other helminthic or bacterial infections (Measures, 2001; Lehnert et al., 2010). Moreover its migratory route from the gut to the lungs following infection of the final host is still poorly understood, although it is suspected to be via the circulation (Barnett et al., 2019). In this study, stable isotope ratio analysis of seal parasites, host tissue and luminal content collected at the site of infection was employed to determine whether the technique can be used to explore further details of the biology and life style of these parasites.

Materials and methods

Sample collection and parasite identification

Parasites and tissue samples were collected during post-mortem examination of 4 stranded grey seals and 2 stranded harbour seals which had been brought to Seal Rescue Ireland and had subsequently died. Nasal mites were identified on the basis of morphology and anatomical location and all the collected specimens were used for stable isotope analysis. For Acanthocephala a subsample of whole worms was used for polymerase chain reaction (PCR) identification and the remainder for stable isotope ratio analysis. Nematodes were cut into 2 segments with one-third used for molecular identification and two-thirds for stable isotope analysis. For molecular identification, DNA was extracted using the QIAamp DNA Mini Kit following the manufacturer's instructions. Acanthocephala were identified based on a fragment of the small subunit ribosomal RNA gene sequence (ssrDNA) (Waeschenbach et al., 2007; Hermosilla et al., 2018), nematodes by amplification of the second internal transcribed spacer (ITS2) (Gasser et al., 1993). All primers and PCR conditions are provided in Table 2. Following Sanger sequencing in both the directions using the amplification primers, sequences were subjected to BLAST and aligned with each other and published sequences using Clustal Omega.

Table 2. Primers and PCR conditions used in the PCR analysis of nematodes and Acanthocephala

Stable isotope ratio analysis

Samples subjected for bulk stable isotope ratio analysis included whole nasal mites, whole Acanthocephala specimens, the remaining two-thirds of nematode worms and duplicate samples of all tissues and (where available) luminal content at the anatomical sites where parasites were collected in each of the seals. Samples were individually freeze-dried, powdered using pestle and mortar and aliquoted into 1 mg portions in miniature tin capsules (Elemental Microanalysis D1028). They were then shipped to Iso-Analytical Limited, Crewe, UK for dual carbon and nitrogen isotope analysis by elemental analysis-isotope ratio mass spectrometry, using a Europa Scientific 20-20. The reference material used was IA-R068 (soy protein, δ ¹³C_V−PDB = −25.22‰, δ ¹⁵N_AIR = 0.99‰). IA-R069 (tuna protein, δ ¹³C_V−PDB = −18.88‰, δ ¹⁵N_AIR = 11.60‰) was run during the analysis of these samples as comparable quality control check samples, with a measurement precision (s.d., n = 12) of 0.03 and 0.04‰ for δ ¹³C and δ ¹⁵N, respectively.

Average δ ¹⁵N and δ ¹³C ± standard deviation (s.d.) were calculated for all seal tissues, luminal content and parasite species. The trophic discrimination factors Δ¹⁵N and Δ¹³C for each parasite–host sample combination was calculated by subtracting the δ ¹⁵N or δ ¹³C of the individual host tissue/luminal content collected from the same site where the parasite was found from the δ ¹⁵N or δ13C value of individual parasites using the following formulas:

$$\eqalign{& \hbox{Trophic discrimination factor}\, \Delta^{15}{\rm N} \cr & \quad \quad = \delta^{15}{\rm N}\, ( {\rm parasite} ) \,-\,\delta^{15}{\rm N} \, ( \hbox{host tissue or luminal content} ) \cr & \hbox{Trophic discrimination factor}\, \Delta^{13}{\rm C} \cr & \quad \quad = \delta^{13}{\rm C}\, ( {\rm parasite} ) \,-\,\delta^{13}{\rm C}\, ( \hbox{host tissue or luminal content} ) } $$

Results

Parasite identification

Based on their morphology and anatomical location, nasal mites were identified as H. halichoeri. PCR amplicons of all 7 Acanthocephala matched the published small subunit ribosomal RNA gene sequence of C. strumosum (e.g. OK096765). Out of a total of 16 analysed nematodes, 5 worms that had been collected from the stomach were identified as P. decipiens, based on 389 bp fragments (identical to HF680317). The remaining 12 worms produced 561 bp PCR amplicons which matched published sequences of O. circumlitus (e.g. KM458088). Of these, 4 were found in the lung, 5 in the stomach and 3 in the right ventricle of the heart and pulmonary artery.

Stable isotope analysis

Nitrogen isotopic compositions and trophic discrimination factors Δ¹⁵N

A total of 6 mites and nasopharyngeal tissue samples from 4 seals (3 grey seals and 1 harbour seal) were subjected to stable isotope ratio analysis (in this case no luminal fluid was available). δ ¹⁵N values of the mites measured 18.9‰ ± 1.4 s.d. on average, those of the seal nasal tissues were 17.3‰ ± 1.2 s.d. (Table 3). Δ¹⁵N, i.e. the difference in δ ¹⁵N of the parasites and the individual host tissues was 1.6 ‰ ± 1.4 s.d. on average (Fig. 1), ranging between −0.015 and 3.2‰ (Fig. 2). For 5 out of 6 mites Δ¹⁵N was positive.

Table 3. δ ¹⁵N and δ ¹³C values (average ± s.d.) of (a) seal tissues, luminal content and (b) parasites

Figure 1. Δ¹⁵N (‰) of Halarachne halichoeri, Corynosoma strumosum, Pseudoterranova decipiens and Otostrongylus circumlitus in relation to (A) host tissue and (B) luminal content. Panels (C) and (D) show Δ¹⁵N (‰) of O. circumlitus worms located in the stomach, heart and lungs. ‘X’ indicates the mean, the middle line the median and the box the data between the 25th and 75th percentiles.

Figure 2. Δ¹⁵N (‰) of (A) H. halichoeri, (B) C. strumosum, (C) P. decipiens and (D) O. circumlitus in relation to host tissue (blue) and luminal content (orange). Numbers on the Y-axis refer to individual parasites.

Overall 7 Acanthocephalan worms, C. strumosum (all collected from grey seals) and 5 Anisakid nematodes, P. decipiens (collected from 3 grey seals and 1 harbour seal) were included in the study. Average δ ¹⁵N values for C. strumosum were 13.7‰ ± 1.4 s.d., for P. decipiens 12.8‰ ± 0.8 s.d. Regarding the respective host tissues, δ ¹⁵N for seal intestinal tissue was 16.7‰ ± 1.5 s.d. (luminal content: 16.3‰ ± 1.5 s.d.) and for the stomach tissue 16.2‰ ± 1.2 s.d. (luminal content: 16.5‰ ± 1.9 s.d.) respectively (Table 3). For both worm species, Δ¹⁵N was always negative with regards to host tissue and to luminal content. In the case of C. strumosum Δ¹⁵N ranged between −4.2 and −2.0‰ (average: −3.0‰ ± 0.6 s.d. with regards to host tissue; −2.6‰ ± 0.6 s.d. with regards to luminal content), in the case of P. decipiens between −6.7 and −0.8‰ (average −3.0‰ ± 1.4 s.d. with regards to host tissue; −3.5‰ ± 2.9 s.d. with regards to luminal content) (Figs 1 and 2).

Twelve O. circumlitus lungworms were collected from 3 grey and 2 harbour seals. δ ¹⁵N for O. circumlitus measured 15.8‰ ± 1.7 s.d. on average. In this case, Δ¹⁵N varied with the location where the worms were found. For lungworms collected from the stomach Δ¹⁵N ranged between −5.6 and 0‰, for worms collected from the heart between −1.1 and 0.2‰, and for worms collected from the lungs, between −0.7 and 1.9‰ (Fig. 2). For all of the lungworms and tissues collected from the stomach Δ¹⁵N was negative (except for a single value which was 0), and all but 1 Δ¹⁵N value for lungworms collected from the heart were negative. In contrast, all worms in the lungs had a positive Δ¹⁵N with regards to the host tissue samples, and in all but 1 case also with regards to the luminal content.

Carbon isotopic compositions and trophic discrimination factors Δ¹³C

Average δ ¹³C values for all parasite species and seal tissues are also provided in Table 3. For Halarachne halochoeri, Δ¹³C, i.e. the difference in δ ¹³C between nasal mites and individual host tissues ranged between −0.2 and −2.7‰ (average −2.0‰ ± 1.0 s.d.) (Fig. 3) and was negative for all 6 mites (Fig. 4). For C. strumosum, most Δ¹³C values were positive (average 0.2‰ ± 0.7 s.d.), while for P. decipiens most were negative (average 0.4‰ ± 2.0 s.d.) (Figs 3 and 4). For O. circumlitus, Δ¹³C again varied depending on where the worms were collected. While in this case the pattern was not as clear-cut as for Δ¹⁵N, there was a tendency for positive Δ¹³C values for worms found in the stomach and heart and negative Δ¹³C values for worms in the lungs (Fig. 4). The difference in δ ¹³C between parasite and host tissue was greatest for worms in the stomach (overall average 0.3‰ ± 0.8 s.d.).

Figure 3. Δ¹³C (‰) of H. halichoeri, C. strumosum, P. decipiens and O. circumlitus in relation to (A) host tissue and (B) luminal content. Panels (C) and (D) show Δ¹³C (‰) of O. circumlitus worms located in the stomach, heart and lungs. ‘X’ indicates the mean, the middle line the median and the box the data between the 25th and 75th percentiles.

Figure 4. Δ¹³C (‰) of (A) H. halichoeri, (B) C. strumosum, (C) P. decipiens and (D) O. circumlitus in relation to host tissue (blue) and luminal content (orange). Numbers on the Y-axis refer to individual parasites.

Biplots of δ ¹³C vs δ ¹⁵N values for host tissues, luminal content and the various parasite species are provided as Supplementary materials (Figs S1 and S2).

Discussion

Halarachne halochoeri mites inhabit the upper respiratory tract, particularly the nasal passages and nasopharyngeal mucosa of seals. They are transmitted directly via sneezing or direct nasal contact when seals congregate on land (Fay and Furman, 1982; Raga et al., 2009; Alonso-Farré et al., 2012). Of all parasites investigated in this study, nasal mites were the only group that consistently recorded positive Δ¹⁵N values with regards to their host's nasopharyngeal tissues, with 2 out of 6 mites exceeding Δ¹⁵N values of 3.1‰. This trophic relationship indicates a classical predator–prey relationship, confirming previous observations that H. halochoeri, similar to other parasitic mites, feeds directly on host tissue (Alonso-Farré et al., 2012). Similar Δ¹⁵N values have been reported for other ectoparasites including ixodid ticks (Schmidt et al., 2011; Heylen et al., 2019) and parasitic copepods (Kamiya et al., 2020). In contrast Δ¹³C values were negative (with 4 out of 6 mites measuring −2.5‰ or less). Reduced or close to zero Δ¹³C values have also been reported in ticks and parasitic copepods relative to their host tissues (Schmidt et al., 2011; Heylen et al., 2019; Kamiya et al., 2020), although none were quite as low as those recorded here. It has been argued that parasites often have low δ ¹³C values because many utilize host-derived lipids and fatty acids directly instead of synthesizing them de novo (Riekenberg et al., 2021), and these host-derived lipids are typically depleted in heavier isotopes because, for kinetic reasons, lighter isotopes are favoured during lipid synthesis (Fagan et al., 2011; Nachev et al., 2017). It is also important to point out that we compared δ ¹³C values from whole mites (containing a range of carbon-rich metabolites, storage and structural elements) with samples of the hosts’ nasopharyngeal mucosa, which would have been more homogenous and presumably contained fewer lipid stores. This was indicated by the C:N ratios of the 2 sample types (which measured 5.9 ± 0.4 s.d. for H. halochoeri and 3.6 ± 0.3 s.d. for seal nasopharyngeal tissue on average, data not shown). No lipid removal step was included in our analysis. Another factor that may play a part in the metabolism of ectoparasites is their endosymbionts. In the case of ticks, symbionts are known to serve as sources of vitamins and energy, affecting many aspects of the tick's biology and life cycle (Bonnet et al., 2017). It is possible that parasitic mites have similar symbiotic relationships which may affect their lipid biosynthesis pathways and isotope composition.

Seals become infected with the Acanthocephalan or thorny-headed worm, C. strumosum, by ingesting infected nearshore amphipod intermediate hosts or demersal fish such as cod or herring which can act as paratenic hosts (Nickol et al., 2002; Raga et al., 2009). Inside the small intestine of seals which serve as final hosts, the adult worms attach to the mucosa using an array of hooks on their retractable proboscis. In contrast to parasitic mites, which due to their exoskeleton and anatomical location are unable to absorb nutrients directly via their body surface, Acanthocephala resemble cestodes in that they lack a mouth and digestive system and absorb all nutrients via their body surface. This nutrient absorption is facilitated by a specialized tegument consisting of a complex syncytium with a lacunar transport system of interconnecting canals and an internal lipid storage system (Deplazes et al., 2016). Interestingly, Δ¹⁵N and Δ¹³C values measured for C. strumosum were exactly the opposite to those recorded for H. halochoeri, i.e. they were strongly negative with regards to Δ¹⁵N and mostly positive with regards to Δ¹³C. This was true in relation to the hosts’ intestinal tissue as well as the luminal content. Negative Δ¹⁵N values have also been reported for Acanthocephala species that infect fish as final hosts (Nachev et al., 2017; Kamiya et al., 2020). Strong depletion in δ ¹⁵N indicates that rather than absorbing half-digested food in the hosts’ intestine indiscriminately, which would place them on the same trophic level as the host, Acanthocephala selectively absorb isotopically lighter host- or host microbiome-derived molecules (Nachev et al., 2017), possibly facilitated by the specialized tegumental syncytium. Alternatively δ ¹⁵N depletion in adult C. strumosum may be due to peptides and amino acids absorbed by previous life stage in the intermediate or paratenic host, amphipods or fish, which themselves would have been on a lower trophic level than the final seal host. The comparatively high Δ¹³C values observed in the present study are puzzling. These compare to more varied but much lower Δ¹³C values reported in Acanthocephalan parasites in fish ranging from negative to close to zero (Nachev et al., 2017; Kamiya et al., 2020). Generally, δ ¹³C enrichment resulting in Δ¹³C values of 0.5–1‰ are thought to indicate direct ingestion of host tissues. Four of the 7 worms analysed in this study were in this bracket with regards to the host's intestinal tissues, 2 were negative (−0.8 to −0.6‰), while one was close to zero. Even higher positive Δ¹³C values, ranging up to 3.6‰, were recorded in relation to the intestinal content, suggesting either specialized use of a fraction of intestinal content or selective use of protein- or carbohydrate-derived materials from the host tissues. It is also possible that the δ ¹³C values of at least some of the worms reflected lipids absorbed from the previous host. The C:N ratio of the parasites was very similar to that of the hosts’ intestinal tissues (3.7 ± 0.1 s.d. for C. strumosum and 3.5 ± 0.2 s.d. for the seal tissues, respectively).

Both, P. decipiens and O. circumlitus have indirect life cycles. The larvae of the anisakid nematode, P. decipiens, undergo development in the gut and haemocoel of benthic crustaceans and macroinvertebrates, before infecting benthophagous fish, where they migrate to somatic muscle tissue (McClelland, 2002). Seals become infected by predation on these smaller fish or on larger demersal fish that have ingested them. In the final host, the adult worms attach to the stomach wall (mostly in the fundus) with their heads embedded in the tissues. Regarding the lungworm, O. circumlitus, the life cycle is not yet fully known. While it is thought to largely resemble that of P. decipiens, an invertebrate host may not be required (Lehnert et al., 2010). Larval stages have been detected in the mucosa, mesentery and pancreas of turbot which are thought to serve as intermediate hosts. It has been suggested that once the worm has been ingested by the final seal host, it migrates to the lungs via the hepatic portal system, heart and pulmonary circulation (Measures, 2001; Barnett et al., 2019). However, the fact that we detected quite a number of O. circumlitus worms in the stomach (with identities confirmed by PCR) indicates that the parasite might spend a considerable amount of time in the stomach before embarking on its tissue migration.

The Δ¹⁵N values of both nematode species were strongly negative when they were collected from the seals’ stomach, resembling those of C. strumosum. Whether this means that they derive their nutrition in a similar way remains to be seen. While the nematode cuticle would preclude the worms from absorbing nutrients across the body surface, nematodes with very small buccal capsules such as P. decipiens and O. circumlitus are thought to feed on mucosal fluid, products of host digestion and cell debris (Taylor et al., 2016a). Consequently low δ ¹⁵N values in these worms may be due to selective absorption of δ ¹⁵N-depleted host metabolites. Alternatively they may be a reflection of nutrients absorbed from previous hosts by earlier life stages. Again, Δ¹³C values were more variable and inconsistent, with some highly positive and some highly negative values recorded for both stomach tissue and contents. The C:N ratio for both nematode species was also quite high (5.5 ± 1.4 s.d. for P. decipiens and 5.0 ± 1.0 s.d. O. circumlitus) possibly as a result of carbon-rich energy stores in the form of glycogen (accounting for about 10% of the dry weight of the worm) and chitin in the cuticle.

What was interesting was the shift in δ ¹⁵N values in O. circumlitus worms collected from other anatomical sites. As the worms migrated from the stomach to the heart and lungs their δ ¹⁵N ratios increased leading to less negative Δ¹⁵N values in relation to heart tissue and mostly positive Δ¹⁵N values in relation to lung tissue (ranging from 0.05 to 1.9‰). This implies a gradual shift to a higher trophic level and an adaptation to a more predatory life style associated with the more pathogenic nature of O. circumlitus compared to that of the other worm species. The fact that Δ¹⁵N did not quite reach the threshold of 3.4‰, considered the hallmark of a predator–prey relationship, may be due to the fact that increased protein quality in the diet of a top predator, such as a seal, can result in a smaller ‰ difference between the diet and the consumer (Riekenberg et al., 2021). It is also interesting to note that the gradual shift of δ ¹⁵N values of O. circumlitus worms in the various tissues indicates that the parasites in the stomach were not flushed there from the lungs via the trachea just before or after death, but had apparently been there for a significant period of time. Whether this delayed migration is part of the normal life cycle or only occurs during very heavy infestations (as has been reported for other histiotrophic parasites) remains to be determined. Δ¹³C values were mostly negatively correlated with Δ¹⁵N values, with positive ratios in the heart and negative ones in the lungs.

In conclusion, this is the first study, to our knowledge, that investigates the stable isotope composition of common seal parasites. Based on Δ¹⁵N values the parasite–host associations could be clearly divided into those that resembled a trophic predator–prey relationship (H. halochoeri in the nasopharyngeal passages and O. circumlitus in the lungs) and those that did not (C. strumosum in the intestine, and P. decipiens and O. circumlitus in the stomach). It is tempting to suggest that Δ¹⁵N values are indicative of the level of pathogenicity associated with a parasite. Parasites with positive Δ¹⁵N values that feed directly on host tissue would be expected to be more invasive and more likely to cause tissue damage than parasites that absorb host molecules. As a result they would be expected to affect their hosts’ health more severely, both directly and indirectly by eliciting host responses which may contribute to their pathogenicity. It may be worthwhile to re-examine the potential pathogenicity of nasal mites in seals, especially considering the fact that they can cause severe clinical signs in cats and dogs if present in large numbers (Taylor et al., 2016b).

In contrast to Δ¹⁵N, Δ¹³C values were much more difficult to interpret. This was probably at least partly due to the fact that we compared parasite bulk samples containing a whole complement of organs and tissues to relatively homogenous seal tissue samples. Finally there were no obvious benefits to including luminal content in addition to host tissue samples in the analysis.