Microbiomes help compose the “holobiont” or the total individual, which includes microbiota co-adapted with the macro-organismal host. Thus, microbiomes can contribute to organismal health through roles in the production of protective secondary metabolites (Nakatsuji et al., 2017; Rothschild et al., 2018), modulating host immunity (Apprill, 2017), or indicating a shift to a disease state (Caporaso et al., 2011; Llewellyn et al., 2014; Nelson et al., 2015; Sanders et al., 2015; Colston and Jackson, 2016; Lax et al., 2017). The term holobiont can refer to any sort of relationship between the microbial community and the host which is not simply limited to symbiotic relationships (Simon et al., 2019). While there is no direct support for holobiont coevolution in sharks, the general concept as shown in other organisms is worth considering (Zilber-Rosenberg and Rosenberg, 2008; Pita et al., 2018; Pratte et al., 2018; Freed et al., 2019; van der Loos et al., 2019; Carthey et al., 2020). Additionally, various anatomical parts within an individual organism can harbor significantly different microbiome communities. In humans, for example, individuals have significantly different microbiome compositions by anatomical location, likely stemming from a complex combination of host behavior, habitat, pH, diet, and varying life stage exposure to microbes (Human Microbiome Project Consortium, 2012a,b; Lloyd-Price et al., 2017; Carlson et al., 2018).

Sharks (class Chondrichthyes; subclass Elasmobranchii) spark biological interest for multiple reasons: sharks represent one of the oldest jawed vertebrate lineages dating back to over 450 million years, many species and populations are in decline due to overexploitation, they play important ecological roles as upper trophic level predators, and display intriguing biological features such as extreme longevity and the presence of molecular adaptations related to genome stability in some species (Dulvy et al., 2014; Nielsen et al., 2016; Marra et al., 2019). Despite the evolutionary antiquity, unique adaptations, and biological and ecological diversity of elasmobranchs, their microbiomes have received little attention with only a handful of studies published thus far (Apprill, 2017; Doane et al., 2017, 2020; Johny et al., 2018; Pogoreutz et al., 2019). A recent report found that the skin microbiome of the thresher shark (Alopias vulpinus) was significantly different and distinguished from the water column, while still hosting some taxa which were also represented in the surrounding water (Doane et al., 2017). In another interesting study, the microbiomes on the skin of four elasmobranch species, which are physiologically marked by dermal denticles and lower mucus secretions, had a higher phylosymbiotic signal than the skin microbiomes of teleost fishes (Doane et al., 2020). Pogoreutz et al. (2019) examined the microbiome of healthy and compromised skin on blacktip reef sharks (Carcharhinus melanopterus), finding significant community differences in individuals from different geographic areas, but not between healthy and injured skin on the same individuals. Johny et al. (2018) found nearly 25% of the gut microbiome of the deep-sea shark Centroscyllium fabricii was unable to be taxonomically classified at the phylum level, suggesting a large proportion of still uncharacterized microbial diversity in that shark microbiome.

Although rare, shark bites on humans do occasionally occur, with 64 unprovoked bites documented worldwide in 2019. The frequency of shark bites is higher in some regions of the world, including Florida, United States, due to large local human populations engaging in a high volume of recreational activities in the ocean (Florida Museum, 2020). Although not often fatal, shark bites can lead to an increased risk of bacterial infection, which are often treated with broad-spectrum antibiotics due to lack of pathogen knowledge (Unger et al., 2014). Characterizing the microbiome of shark teeth may help future bite treatment become more effective (e.g., vancomycin, fidaxomicin, and sarecycline) and targeted to narrower microbial groups.

Here, we provide a comparative assessment of the microbiome communities across four different anatomical sites (gills, teeth, skin, and cloaca) within each of five shark species common to South Florida waters (Florida Sea Grant, 2013): nurse (Ginglymostoma cirratum), lemon (Negaprion brevirostris), sandbar (Carcharhinus plumbeus), Caribbean reef (Carcharhinus perezii), and tiger (Galeocerdo cuvier) sharks. These five species occupy variable, but sometimes overlapping habitats (nearshore benthic to neritic waters, coral reef ecosystems, pelagic waters), and all have been implicated, at various levels, in human bite incidents. We tested three hypotheses about the microbiomes of these five shark species: (1) shark microbiomes are distinct from the surrounding seawater environment, (2) microbiomes vary in community composition among shark species, and (3) microbiomes of different anatomical locations within host species vary in microbial composition.

In this study, we found that shark-associated microbial communities were distinct from the surrounding seawater and were significantly different among the species sampled. Previous studies indicate that holobiont microbiomes significantly differ from surrounding non-symbiont communities—e.g., seawater (Thompson et al., 2017; Freed et al., 2019). Based on our findings, some microbial taxa clearly overlap between the microbial community of the surrounding seawater and host shark microbial communities. We showed that overall microbial community composition was significantly different between shark species. Additionally, we found that anatomical location microbiomes varied significantly in composition when comparing between shark species.

Previous work has shown the presence of a community which co-evolves with the host in other organisms and systems (Zilber-Rosenberg and Rosenberg, 2008; Pita et al., 2018; Pratte et al., 2018; Freed et al., 2019; van der Loos et al., 2019; Carthey et al., 2020). Our findings support specialized microbial communities partially explained by anatomical location and shark species. Different shark species could inhabit niches or exhibit behaviors which could be related to the differential microbial communities characterized here. Environmental influences on natural microbial communities could include pollutants, migration patterns, pH, and salinity. For example, sharks are known to inhabit depths of 90 m or deeper during migration but move to shallower waters for parturition (Beck, 2016). The frequency of this behavior could contribute to the overall microbiome community composition sharks, as environmental factors could change with water depth. For example, tiger sharks have a proposed gestation period of 12 months (Castro, 2009), while nurse sharks have a much shorter 5–6 month long gestation period (Castro, 2000). Earlier environmental and host associated microbiome profiles of S. Florida, which our laboratory has been systematically characterizing, may add to the more routinely monitored environmental parameters (Thomas et al., 2016; Freed et al., 2019).

When comparing samples from all anatomical location for differences by shark species, there was no significant difference in overall microbial community diversity or richness (Table 1 and Figure 3). There was a significant difference in the composition by shark species, suggesting that there are factors which this study did not account for in addition to species which are contributing to unique communities. Our data support this, with variance partitioning showing that richness (95.7%), diversity (89.4%), and composition (89.7%) have substantial residuals which could in part be explained by environmental differences such as pollution, temperature, pH, etc.

When microbial communities of anatomical locations while also considering shark species were compared across all samples, there was a significant difference in the diversity of these communities, but not in richness (Table 1 and Figure 3). Community composition was also significantly different between anatomical location when considering host species. Our analyses found that only 5.2% of the variability in richness is explained by host species and anatomical location, suggesting that other environmental factors have important effects on richness of shark-associated microbial communities. Additionally, 10.7% of variability in diversity and 10.3% in composition are explained by host species and anatomical location. When considering richness (9 vs. 1%), diversity (9 vs. 1%), and composition (5 vs. 4%) anatomical location is the main driver, suggesting associated specialized microbial niches. This further emphasizes that these shark-associated communities are heavily dependent upon other environmental and perhaps behavioral factors which would allow for specialized niches.

Distinct microbial communities on shark teeth which are less diverse than other anatomical locations studied were drivers for compositional differences among anatomical locations based on pairwise comparisons (Table 2), leading to a more in-depth examination of the teeth microbial communities between host species. Previous studies show species, individual, or ethnicity specific microbial communities (Mason et al., 2013; Hernandez-Agreda et al., 2018; Ribeiro and Arnold, 2019). We showed that shark teeth host species-specific microbiomes which are unique in diversity, richness, and composition (Figure 7). It has been shown that differences in diet can result in differential oral microbiomes in other organisms (Adler et al., 2016; Fan et al., 2018), but this issue has been minimally explored in sharks. It is possible that differential diets effect the teeth microbiomes of sharks. For example, nurse sharks (Ginglymostoma cirratum) which are primarily benthic dwellers, frequently sedentary, and show high site fidelity (Castro, 2000), while tiger sharks (Galeocerdo cuvier), which occupy the mid to surface of the water column and often migrate vast distances (Lea et al., 2015). Tiger sharks also have highly cosmopolitan diets which include bivalves, teleosts, reptiles, and mammals and show ontogenetic variability (Dicken et al., 2017). However, dietary habits are not well-characterized for all shark species and can vary by individual and geographically (Simpfendorfer et al., 2001), so it is difficult to make conclusions about teeth microbial communities based only on host species.

With a larger sample size and more data, teeth microbiomes could reveal microbial taxa which could serve as biomarkers to identify shark species involved in a bite incident, or to improve bite treatment, which would be useful for forensics and ecological contexts. Potential infections after any deep bite wound are of concern. However, extensive profiling of the oral microbiome has occurred mostly for humans, with fewer current studies of other animals (Roggenbuck et al., 2014; Rojas et al., 2020). Bacteria isolated from infected bite wounds can often reflect the oral flora of the organism responsible for the bite, as exhibited in fish, mammals, reptiles, and birds (von Graevenitz et al., 2000; Abrahamian and Goldstein, 2011). Several microbial taxa found to be associated with shark teeth in this study belong to taxonomic groups that have been causes for concern in other animal bite wounds, such as Streptococcus, Staphylococcus, Corynebacterium, Enterococcus, and Haemophilus. The taxa Vibrio, Salmonella enterica, Psychrobacter, and Halomonas were also found, and are mostly associated with bites from aquatic organisms and reptiles. Vibrio has previously been found to be a concern predominantly in shark bites (Abrahamian and Goldstein, 2011).

Vibrio has been cultured previously from shark teeth, with Vibrio parahaemolyticus, Vibrio alginolyticus, and Vibrio fluvialis being present in a beached white shark’s teeth (Carcharodon carcharias) (Buck et al., 1984). Additionally, Vibrio carchariae has previously been isolated from an infected shark bite wound in a human victim (Pavia et al., 1989). Wounds from another two infected shark bites were sampled and cultured to show Vibrio parahaemolyticus in one patient and Vibrio alginolyticus in the other (Royle et al., 1997). Bite wounds are known to include bacteria which are found on the skin of the victim or in the surrounding environment (Abrahamian and Goldstein, 2011), but our data shows that Vibrio was not found in high relative abundance in the waters sampled in the vicinity of the shark captures.

South Florida sharks host bacterial microbiomes distinct from the surrounding seawater environment and differ among taxa in microbial community richness and diversity. Microbiomes by host species appear driven by differences in composition, while microbial communities compared by anatomical sample locations are driven by the diversity and composition of the community. Host-associated microbial communities vary in composition when considering both anatomical location and host species. Because we show that teeth microbial communities are dependent upon host species, data generated in this study could be applicable to improvement in shark bite treatment and could serve as a biomarker to identify shark species involved in a bite incident which would be useful for ecological context. Future research should focus on bacteria found in shark teeth to determine if those strains present are truly pathogenic (following Koch’s postulates or carry viable virulence markers, etc.) to provide tangible insights to bite treatment or for future use as biomarkers for identifying shark species involved in bite events.

The datasets generated for this study can be found in the online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

The animal study was reviewed and approved by the Nova Southeastern University Animal Care and Use Committee, IACUC Protocol 2017.11.MS1.

RS performed the experimental methods and the statistical analysis with assistance from CE. RS wrote the first draft of the manuscript. All authors contributed to the conception and design of the study, contributed to the manuscript revisions, read, and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.