Microbiome profile of the Antarctic clam Laternula elliptica

Abstract

The filter feeder clam Laternula elliptica is a key species in the Antarctic ecosystem. As a stenothermal benthic species, it has a poor capacity for adaptation to small temperature variations. Despite their ecological importance and sensitivity to climate change, studies on their microbiomes are lacking. The goal of this study was to characterize the bacterial communities of L. elliptica and the tissues variability of this microbiome to provide an initial insight of host-microbiota interactions. We investigated the diversity and taxonomic composition of bacterial communities of L. elliptica from five regions of the body using high-throughput 16S rRNA gene sequencing. The results showed that the microbiome of L. elliptica tended to differ from that of the surrounding seawater samples. However, there were no significant differences in the microbial composition between the body sites, and only two OTUs were present in all samples, being considered core microbiome (genus Moritella and Polaribacter). No significant differences were detected in diversity indexes among tissues (mean 626.85 for observed OTUs, 628.89 Chao1, 5.42 Shannon, and 0.87 Simpson). Rarefaction analysis revealed that most tissues reached a plateau of OTU number according to sample increase, with the exception of Siphon samples. Psychromonas and Psychrilyobacter were particularly abundant in L. elliptica whereas Fluviicola dominated seawater and siphons. Typical polar bacteria were Polaribacter, Shewanella, Colwellia, and Moritella. We detected the prevalence of pathogenic bacterial sequences, particularly in the family Arcobacteraceae, Pseudomonadaceae, and Mycoplasmataceae. The prokaryotic diversity was similar among tissues, as well as their taxonomic composition, suggesting a homogeneity of the microbiome along L. elliptica body. The Antarctic clam population can be used to monitor the impact of human activity in areas near Antarctic stations that discharge wastewater.

Supplementary Information

The online version contains supplementary material available at 10.1007/s42770-023-01200-1.

Introduction

The Antarctic clam Laternula elliptica (P. P. King, 1832) is widely distributed in Antarctica [1]. It is an abundant species that is deeply buried in the muddy sediments of shallow waters. As a filter-feeder, L. elliptica is a key species in Antarctica that enhances the organic carbon flux from the water column to the seabed [1]. This species can filter large amounts of seawater, inducing important contact with bacteria from the environment [2]. Its circumpolar distribution and abundance make it a good sentinel species for Antarctic climate change monitoring, considering the current and projected changes associated with climate change [3]. L. elliptica has been subject to many experiences about thermal tolerance as a typical stenothermal species [4, 5]. However, microbiological and genetic studies on symbiotic bacterial associations with Antarctic bivalves or L. elliptica are lacking, despite the importance of symbiosis for sensitive Antarctic benthic species. The possibly essential role of its associated microbiome in the adaptation to climate change and the diversity, abundance, and functions of symbiotic bacteria in Antarctica remain largely unknown [6, 7]. However, due to their sensitivity and fast response to changes, bacteria could be good indicators of the ecosystem of host health [8, 9].

Studies on the functional role of microbes in marine invertebrates have shown that they produce enzymes that help digestion [2], which are a source of secondary metabolites that constitute the chemical defense of the host [10], influence the stress response, and contribute to the protection of the host [11]. In groups such as Porifera, the holobiont influences the nitrogen, phosphorus, and sulfur cycles; uptake and conversion of dissolved organic matter; and CO₂ fixation [12].

Several studies on bivalve mollusks have shown that the gut microbiota is influenced by life stage, temporal variation, host taxonomy, and environmental factors [13–16]. In regard to the latter, the gut microbiota has been studied in the context of climate change (e.g., temperature and acidification) [17–19]. For instance, increased temperature significantly affected microbial diversity in the gut microbiota of mussels [16] and in oyster shifts in the microbiome depending on the host genotype [20, 21].

In Antarctica, the presence of chemoautotrophic microorganisms in sponges, such as amino-oxidizers, nitrifiers, and sulfur-oxidizers, helps them survive the winter in inhospitable environments and contributes to major nutrient cycles in the ecosystem [6, 22]. Recent characterization of the Antarctic krill microbiome (Euphausia suberba) highlighted these patterns, showing the presence of a symbiont of the order Mycoplasmoidales in the digestive tissues, which seems to improve the survival of low-quality water in other crustaceans [23]. Cui et al. [24] found that the microbiome of E. superba provides protection to its host by producing antibacterial metabolites against pathogenic microorganism’s and synthesizing cytotoxic metabolites in predatory mammals [24]. Moreover, Antarctic krill influences the composition and dispersal of Southern Ocean bacterial communities and the nutrient cycles simultaneously by migration horizontally and vertically by sinking pellets [23]. Despite the importance of pelagic and benthic Antarctic marine invertebrates, few studies have examined their microbiome. Only the gut microbiome of the Antarctic sea urchin (Abatus agassizii) and the pteropod gut contents of Limacina helicina antarctica have been described using sequencing techniques [25, 26].

In this study, we assessed the diversity, taxonomic composition, and structure of the bacterial communities of Laternula elliptica from five regions of the body: digestive gland, gill, mantle, and siphon. The organisms were collected from Fildes Bay, King George Island, Antarctica. We identified L. elliptica and seawater-associated bacteria using 16S rRNA Metabarcoding and bioinformatics tools.

Materials and methods

Sampling of Laternula elliptica tissues

Individuals of L. elliptica (n=5) were collected by SCUBA diving during the Antarctic summer in 2020 (LVI Antarctic Scientific Expedition), in shallow waters (approximately 15 m depth) at Fildes Bay (62° 12′ 19.34″ S, 58° 56′ 57.0052″ W), King George Island (Supplementary Material 1). Small pieces of four tissue types (mantle, digestive gland, gill, and siphon) were collected obtained from each organism with a sterile scalpel and then preserved in plastic tubes containing RNALater©, quick-frozen in liquid nitrogen, and stored at −80°C. Samples were transported at −20 °C to the Laboratorio de Biorecursos, Chilean Antarctic Institute (INACH) in Punta Arenas (Chile) for further processing. Fildes Bay seawater were sampled in January 2020, from the pump, taking water from the bay to aquariums of the Profesor Julio Escudero base without filtration system. Physical-chemical conditions of in situ Fildes Bay seawater were −0.51°C temperature, pH of 8.42, and salinity of 32.8‰.

DNA extraction, library preparation, and 16S rRNA gene sequencing

Genomic DNA was extracted from the collected specimens for 16S rRNA gene amplicon sequencing using the DNeasy PowerSoil Kit (MOBIO, Cat: 12888-100) following the manufacturer’s protocol. For each replicate, about 0.28 g (± 0.06 g) of clam tissue (to ensure obtaining the whole bacterial community) was homogenized using the Precellys© Evolution homogenizer (Bertin Technologies, Montigny-Le-Bretonneux, France). The same kit was used for DNA extraction from water samples. From each sample (n=4), a volume of 0.78±0.13 L of seawater was filtered through a 0.22-μm filter. These filters were previously homogenized with Precellys© Evolution homogenizer.

DNA concentration was assessed using Infinite 200 PRO® (Tecan, Switzerland), and DNA was quantified by measuring the absorbance at 260 nm (ultraviolet). DNA purity was checked by computing the (260/280) ratio.

Amplification of the full-length 16S rRNA gene (≃1400 bp) was performed by PCR using the forward primer27F 5′-AGAGTTTGATCCTGGCTCAG-3′ and the reverse primer1492R 5′-GGTTACCTTGTTACGACTT-3′. Each 15 μL of PCR reaction mixture contained 8.9 μL of nuclease free water, 0.6 μL of MgCl₂ solution (50 mM), 1.5 μL of 10× PCR buffer (–Mg), 0.75 μL of each primer (10 μM), 0.2 μL of the Taq polymerase DNA (5 U/μL), 0.3 μL of deoxynucleotide triphosphates (dNTP; 10 mM), and 2 μL of DNA template with a concentration <100 ng μL⁻¹. Samples with higher concentrations were diluted to 60 ng μL⁻¹. Touchdown PRC conditions were as follows: 1 cycle of denaturation at 95°C for 7 min, followed by 35 cycles of annealing and extension at 95°C for 45 s, 58°C for 30 s, 72°C for 2 min, and 10 s. Then, 1 cycle at 72°C for 7 min and finally PCR products were stored at 4°C until processing (Thermociclador SimpliAmp™, A24811). In order to validate the amplification of the 16S rRNA, amplicons were electrophoresed on 1.2% agarose gel with TAE buffer, containing 5 μL of GelRed™ 10,000× solution in DMSO and using a 1 kbDNA ladder.

Sequencing was performed by Macrogen Inc. (Seoul, South Korea) using the Miseq Illumina 300 bp technology, sequencing 100-kb reads per sample, to sequence the hypervariable regions V3-V4 of the bacterial gene 16S rRNA. The adaptors used to construct the libraries were Bakt_341F (CCTACGGGNGGCWGCAG) and Bakt_805R (GACTACHVGGGTATCTAATCC). Raw sequences were deposited at NCBI as BioProject with the accession number PRJNA970788.

Bioinformatic analysis

Raw sequence analyses were performed using the open-source software QIIME2 [27]. First, quality control (qc) of paired-end read libraries of the 16S rRNA V4–V5 region was performed to determine the degree of sequence integrity. Sequences were then denoised with the dada2-plugin [28], including merging, clustering, and trimming of the reads. For the latter, the cutoff at the 3′-end was chosen based on the qc graph, resulting in 242 nucleotides (nt) for the forward strand and 228 nt for the reverse strand. At the 5′-end, the primers were removed (forward strand: 20; reverse strand: 21).

The taxonomic classification was performed using the SILVA rRNA gene database version 138 [29]. After taxonomic affiliation, the OTUs assigned as chloroplasts (Cyanobacteria) and mitochondria (eukaryotes) were discarded from the two previously generated filtered tables and sequences. Analyses at the Phylum and Class levels were performed with the most abundant taxa to visualize their composition. The OTUs present in all individual samples were considered core OTUs. Alpha-diversity indices were then calculated using core metrics plugging, including Chao1, Shannon, and Simpson. OTU sharing was investigated through a Venn diagram using the Jvenn tool, a free online platform for data analysis [30].

Circos software online (http://www.circos.ca) was used to perform co-occurrence microbiome networks at Class and OTU levels using taxa present in at least one sample with at least 1% of relative abundance. Values were multiplied by 1000 and rounded off to perform the analysis.

Statistical analysis

Rarefaction and univariate measures of diversity indices (observed OTUs [Sobs], Chao1, Shannon, and Simpson) were calculated and plotted using R Studio scripts, and statistical differences were assessed by applying the Shapiro–Wilk and Barttlet tests to check normality assumption and variance homogeneity, respectively. Considering these results, we applied ANOVA or Kruskal–Wallis tests to assess significant statistical differences for parametric and non-parametric data, respectively. Non-metric multidimensional scaling (nMDS) was performed to visualize multivariate patterns in the microbiomes of different tissues of L. elliptica from Fildes Bay, King George Island, Antarctica. Statistical differences in microbiome composition were analyzed using permutational analysis of variance (PERMANOVA), based on Bray–Curtis matrices of double-square root transformed abundance data. Statistical differences were tested using 9999 permutations of data in PRIMER v7 [31].

Results

The present Illumina Miseq V3-V4 Prokaryotic 16S amplicon sequencing of 13 tissue samples and four seawater samples yielded 1,198,714 reads after merging, denoising, and chimera removal, representing 72.63% of total reads. These filtered reads were affiliated with 1622 bacterial and archaeal OTUs belonging to 49 different phyla (the most abundant in Fig. 1 and Supplementary Material 2), considering four different tissues for each individual (siphon, mantle, gill, and digestive gland) and seawater samples (Table 1). The most abundant phyla shared by all tissue and seawater samples were Proteobacteria and Bacteroidetes with 38.97 and 32.89%, respectively). Although that Spirochaetes have represented 8.52% of the total abundance in all samples, this phylum was present only in Laternula elliptica tissues, and not in seawater. Rarefaction analysis revealed that most curves reached a plateau suggesting that sequencing effort showed a high coverage for the majority of tissues analyzed, with the exception of Siphon organ which have continuous increase of observed OTUs with adding new samples (Supplementary Material 3).

Fig. 1.

Relative abundances of the most frequent Prokaryotic Phylum (a) and Class (b) levels associate to L. elliptica tissues and surrounding seawater. The number of samples were n = 4 for sea water, n = 3 for siphon, n = 5 for mantle, n = 3 for gill and n = 2 for digestive gland

Table 1.

Sample codes corresponding to seawater or tissues and the sequences quantities after all qiime2 filters

Code	Type of sample	Sequences
AqE1	SeaWater	60115
AqE2	SeaWater	63789
AqE3	SeaWater	64491
AqE4	SeaWater	56083
B2A	Gill	88168
B3	Gill	72950
B3A	Gill	73210
GD1A	DigestiveGland	61303
GD2A	DigestiveGland	68778
M1A	Mantle	66751
M2A	Mantle	74628
M3	Mantle	60538
M3A	Mantle	78784
M5	Mantle	74096
S2A	Siphon	73885
S3	Siphon	78814
S5	Siphon	82331

A highly similar prokaryotic assembly was observed at the Phylum and Class levels for all seawater samples (Fig. 1a, b). A relatively similar community structure was recorded for most external tissues (mantle and siphon) and seawater; however, two siphon samples showed higher number of phyla and classes. In contrast, seawater samples were clearly differentiated from internal tissues (gill and digestive gland). Seawater samples were dominated by Bacteroidia and Proteobacteria phyla at 65.87 and 26.52%, respectively. Meanwhile, an inversion of abundances occurred between these two phyla for two siphon samples with 25.32 and 42.76%, respectively, with the other siphon samples similar to seawater. For these two siphon samples dissimilar to seawater, an increase in Verrucomicrobiota, Planctomycetota, Desulfobacterota, and Nitrospinota phyla was detected. The same situation was observed in mantle tissues, with one sample being similar to seawater and others, with inversion of abundances of phylum Bacteroidia and Proteobacteria and an increase of other phyla, such as Spirochaetota, Fusobacteriota, and Campilobacterota. Regarding gill tissues, the differences in seawater, siphon, and mantle samples were clear, showing a decrease in the two main phyla and an important increase in Spirochaetota, Fusobacteriota, and Campilobacterota with 16.72, 27.43, and 16.43%, respectively. Concerning most internal tissues, the digestive gland, the prokaryotic community is entirely different, with an increase in Firmicutes phylum abundance for one sample and Fusobacteriota and Campylobacter for the other (Fig. 1a). Similar patterns were observed for the class level with one sample of siphon and another for mantle very similar to all seawater samples, and for the other siphon and mantle samples, an inversion of class occurred, with an increase in Gammaproteobacteria and a decrease in Bacteroidia. In addition, an increase in other classes such as Verrucomicrobiae, Nitrospinia, Desulfobulbia, and Zetaproteobacteria was observed in these two tissues. Gill tissue is very dissimilar in prokaryotic classes, with augmentation of Fusobacteriia and Spirochaetia classes. Finally, as expected, the digestive gland has a different prokaryotic composition with a high abundance of Clostridia, Fusobacteria, and Campylobacteria Classes (Fig. 1b). Four OTUs showed significant differences in relative abundance: OTU3 belonging to Flavobacteriaceae family, OTU8 belonging to Bacteroidia Class, OTU15 belonging to Rhodobacteraceae family, and OTU172 belonging to Lachnospiraceae family (Supplementary Material 4 for detailed information). These relative abundance differences indicate natural seawater and at least one of the tissues evaluated.

The core microbial community associated with the four tissues and water samples showed a small number of OTUs associated with L. elliptica in the seawater. Among the 1662 OTUs detected, only 21 were present in all samples (Fig. 2). These OUTs belonged to Proteobacteria (n = 10), Bacteroidota (n = 7), Campilobacterota (n = 3), and Fusobacteriota (n = 1).

Fig. 2.

Shared OTUs among the different tissues of L. elliptica and seawater samples, with at least one sequence in at least one sample of tissue or seawater. Barplots show the abundance of present OTUs for each sample type

The principal coordinates analysis plot corroborates our previous analyses in terms of comparing seawater samples to tissues with an external exposure; there is no clear distinction in OUT composition with respect to seawater. Seawater samples were localized on one side of the graph with two samples of external tissues (one for the siphon and another for the mantle), and on the other side on axis 1, the rest of the samples (Fig. 3). On axis 2, separation of samples was observed from external tissues (bottom of the plot) to internal tissues (top of the plot).

Fig. 3.

Bray-Curtis PCA of prokaryotic assembly at OTUs level associated with L. elliptica tissues and surrounding seawater

Regarding alpha diversity indices, no significant differences were revealed for the four analyses performed (Chao1, ObsOTUs, Shannon, and Simpson) between all prokaryotic communities considered in the present study (p-values > 0.05, Fig. 4). Alpha-diversity indices revealed that Siphon samples had the highest prokaryotic associated diversity, contrary to gill which had the lowest indices (Supplementary Material 5). Considering the OTU contributions to the differences between seawater and tissue samples, the eight OTUs with the highest contributions only represented 5.55% of the cumulative differences (Supplementary Table 1). In fact, the highest OTU difference contribution is 1.25 for OTU2 affiliated with Fluviicola uncultured Bacteroidetes/Chlorobii species of Bacteriodota phylum, which is highly abundant in seawater samples and poorly represented in tissue samples, with the exception of a mantle sample.

Fig. 4.

Alpha-Diversity indexes in different tissues of Laternula elliptica for observed OTUs (a), Chao1 (b), Shannon (c), and Simpson (d)

The co-occurrence network plot from Circos at the phylum level showed that two main phyla were present in all tissue samples (Fig. 5a) (Bacteroidia and Proteobacteria) and other phyla with high abundances, but not in all samples, such as Spirochaetota, Campilobacterota, and Firmicutes. At the Class level, a finer resolution is shown with the presence of three classes in all samples, but with different abundances, such as Bacteroidia, Gamma, and Alphaproteobacteria (Fig. 5b). Other classes with presence in most samples were Fusobacteria and Campylobacteria with moderate abundances, Spirochaetia was present in high abundance in three samples of mantle, and gill and Clostridia were only present and abundant in one sample of the digestive gland. At the finest scale, three main OTUs were detected in most samples (Fig. 5c), corresponding to OTU5 (genus Psychromonas), OTU3 (Family Flavobacteriaceae), and OTU4 (genus Psycrilyobacter) (Fig. 5c), with being OTU5 the most abundant. OTU1 was highly abundant in some mantle and gill samples, and five other OTUs were present in most samples but had low abundance, such as OTU13 (genus Moritella), OTU7 (genus Sulfitobacter), OTU14 (genus Polaribacter), OTU6 (family Arcobacteraceae), and OTU16 (genus Colwellia).

Fig. 5.

Relative abundances of microbial groups associated with the Antarctic soft clam Laternula elliptica. The circos plot displays the relative abundance of most abundant prokaryotic bacteria at Phylum level (a), Class level (b), and OTU level (c) from samples of different tissues and natural sea water. Each plot was generated from the OTU table grouped by phylum, class level, and OTU level with at least 1% of relative abundance in at least one tissue or natural sea water sample. Values were multiplied by 1000 and rounded without decimals in order to perform the circos plot analysis. The relative abundance of each categories is directly proportional to the width of each ribbon connecting prokaryotic taxa to its respective sample. Each phylum is assigned a specific color. The inner ring represents the total relative abundances (multiplied by 100) for a specific taxon and the proportion assigned to each locality. The outer ring represents the percentage of each taxon assigned to each tissue or natural sea water

Regarding the core community, only two OTUs (OTU13 and OTU14) were found in all 17 samples corresponding to the genera Moritella (Class Gamma-proteobacteria) and Polaribacter (Class Bacteroidia), and their compositions were 2.49 and 2.46%, respectively.

We identified 10 OTUs (OTU06, OTU109, OTU191, OTU349, OTU447, OTU672, OTU690, OTU929, OTU1002, and OTU1221) belonging to 8 families related to pathogenic bacteria present in seawater and tissue. The most abundant was Arcobacteraceae (2.36%), which was found in both samples. Pathogenic bacteria were present in low percentages. The genera identified in tissues were Pseudomonas (0.045%), Mycoplasma (0.027%), Escherichia-Shigella (0.016%), Serratia (0.001%), and Enterococcus (0.0007%). Pseudomonas was the only genus present in the seawater samples (Supplementary Material 6).

Discussion

In this study, we characterized the microbiota of a filter-feeder Antarctic clam using a next-generation sequencing approach for the first time, the recent development of next-generation sequencing (NGS) of the bacterial gene 16S rRNA and the diminution of costs have led to an increase in microbiome research [2]. In marine systems, most microbial research has been focused on tropical and temperate regions [32–34], where sponges are common examples of animal-microbial symbiosis and are therefore widely studied [8, 35]. Literature regarding polar species is scarce and mainly associated with Antarctic sponges [6, 22, 36, 37] and a few other invertebrate species, such as the Antarctic anemone and soft-shell clam Mya arenaria [38, 39]. We found that the microbial community associated with L. elliptica appears to be different from that in seawater, with relatively few shared OTUs in the tissues. This has been reported for other benthic marine invertebrates including sea anemones, sea cucumbers, sponges, Antarctic krill, sea urchins, and corals [12, 23, 40, 41]. In the pteropod (pelagic snail), Limacina helicina antarctica [26] was reported at the class level that Mollicutes, Alphaproteobacteria, and Bacteroidia comprised the largest proportion of the prokaryotic community and were present within all samples, while in the L. elliptica appear Gammaproteobacteria and Bacteroidia which were found to be predominant class.

The phylum Bacteroidota dominated seawater samples, whereas Proteobacteria (Gamma) was dominant in L. elliptica tissues, which is in accordance with the literature [23, 26, 42]. Moreover, Schwob et al. [25] found that these two phyla were the most abundant in sediments of the surface layer in Fildes Bay, King Georg Island. The genus Psychromonas was found to be dominant in the digestive gland (56%), mantle (39%), and gill (24%); however, it was rare in seawater samples. Its members are capable of hydrolyzing starch and other insoluble sugars and can produce ω-3 polyunsaturated fatty acids [43, 44]. Psychrilyobacter (Fusobacteriia) was the second most abundant bacterium in digestive gland (25%) and gills (18%). This genus is also anaerobic, halophilic, and psychrophilic, and is found in Pacific oysters at 4°C [45]. It has been described as a saccharolytic and fermentative bacteria [46]. Therefore, these genera may have important functions in the digestion and nutrition of L. elliptica. Proteobacteria, Fusobacteriia, and Campylobacterota were dominant in other invertebrate digestive systems, such as sea urchins [33, 42, 47]. Campylobacterota contains the families Arcobacter and Sulfurimonadaceae, and may be involved in sulfur oxidation, which could be an alternative energy metabolism in extreme environments, as suggested for the Antarctic anemone Edwardsiella andrillae and Antarctic sponges [37, 38]. Fluviicola was the dominant genus in seawater (32%) but was also abundant in the siphon (13%) and the mantel (7%), which is coherent because they are in constant contact with 29 seawater bacteria. The psychrophilic bacterium Colwellia was mostly present in seawater and was abundant in all tissues, except the digestive gland. Clarke et al. [23] found Colwellia was dominant (40%) in moults of Antarctic krill, being rare in the digestive gland but also into the seawater. Polaribacter (Flavobacteria), Shewanella, and Moritella (Gammoproteobacteria) are other psychrophilic bacteria that are mostly found in gill and seawater samples. These bacteria are generally found in polar seawater [48, 49] but are also present in holothurians [49], sea urchins [33, 50], sponges [51], and oysters [45]. Polaribacter species, originally described by Gosink et al. [48], are psychrophilic organisms isolated from polar marine environments. Shewanella species have been reported to cause animal diseases and to be pathogenic to humans [52], but also to produce nitrate reductase, which is important in the nitrogen cycle [53]. Important nitrite-oxidizing bacteria (NOB), such as the genera Nistropina, Nitrospira, and Nitrosomonas, and sulfur-reducing bacteria (SRB), such as the classes Desulfobacteria and Desulfobulbia, were only present in the siphon and mantle samples. NOB were previously found in several sponge species [9], and these nitrifiers can impact the ecology of the host as well as the ecosystem by converting ammonium into nitrate [54]. RSB were found in the gut digesta of sea urchins [25, 33]. Finally, Verrucomicrobiae was also predominant in the siphon, whereas it has been reported in the gut of the oyster Crassostrea virginica [55] and in the gill and gut of the zebra mussel Dreissena polymorpha [56]. These bacteria may have an important function in host nutrition but were not found in the digestive gland of L. elliptica. Once the OTUs shared between seawater and L. elliptica were deleted, Photobacterium appeared to be dominant in the digestive gland. This genus has also been found in mussels, oysters, and the tropical sea urchin gut [47, 57, 58]. Photobacterium is commonly reported in symbiotic associations with marine animals and is widely spread into oceans. Some of its members are linked to lipolytic activity [59], while others are pathogenic to animals and humans [60]. Bacteroides are ubiquitous in the digestive systems of animals [61]. It possesses both oxidative and fermentative types of metabolism, can utilize several substrates, and is often found in association 30 with mussels [53, 57]. The functions of the genera found in the digestive gland were consistent with those found in the gut of sea urchins by Hakim et al. [42]. Methylophaga and halophilic and methylotrophic bacteria were the dominant sea water-specific genera.

Metagenomic analyses have shown that fecal contamination in mollusk samples is mainly composed of indicator bacteria [62]. The source of fecal contamination in shellfish mollusks is humans or animals, indicating that the sampling area was contaminated by human and/or animal fecal sources [63]. This study reported the presence of human and multi-resistant bacteria in seawater samples from the Fildes Peninsula [64]. The Antarctic clam population sampled was close to the sewage outfalls of three Antarctic stations on Maxwell Bay (King George Island); however, it was also close to Ardley Island, where a penguin colony is located. The prevalence of Arcobacteraceae may be related to the high prevalence found in raw sewage and wastewater treatment plants [65]. Therefore, it is important to identify and manage the sources of contamination and take preventive actions to reduce pathogenic and multidrug-resistant bacteria in the seawater of Fildes Bay. Further studies of Antarctic clam populations from impacted sites are needed to validate the use of these organisms as indicator species. Finally, as it has been found in other invertebrates from other regions, the tissues associated microbiome can mediate mechanisms contributing to tolerance to climate change such as Ocean Acidification, warming, and potentially pollution. For example, in coral reefs, it has been identified several potential direct and indirect microbiome-mediated mechanisms that may contribute to environmental acclimatization in invertebrate species. These include increasing host-energy metabolism, reduction of oxidative stress, regulation of nutrients in host cells, and increased pathogen resistance [66]. In this regard, it will be important to evaluate host-associated microbiome metabolite production in Antarctic coastal waters near and far from human activities.

Author contribution

Conceptualization: MGA, RR, and CAC; methodology: GP, RR, AF, CAC, and MGA; formal analysis and investigation: GP, RR, CAC, and MGA; writing and editing: GP, RR, CAC, and MGA; and funding acquisition: RR and MGA.

Funding

This research was funded by the ANID-FONDECYT Grant Proyecto FONDECYT Iniciación No. 11190802 and the Marine Protected Areas Program (Number 24 03 052) of the Instituto Antártico Chileno. CAC was also funded by the ANID-Millennium Science Initiative Program (ICN2021_002). We would like to thank the Antarctic expedition department and Ignacio Garrido for diving support, as well as all logistic personnel at Profesor Julio Escudero Station.

Declarations

Conflict of interest

The authors declare no competing interests.