Epidermal Microbiomes of Triakis Semifasciata Are Consistent Across Captive and Wild Environments

doi:10.21203/rs.3.rs-948364/v1

Background: Characterizations of sharks-microbe systems in wild environments have outlined patterns of species-specific microbiomes; however, whether captivity affects these trends has yet to be determined. We used high-throughput shotgun sequencing to assess the epidermal microbiome belonging to leopard sharks (Triakis semifasciata) in captive (Birch Aquarium, La Jolla California), semi-captive (<1 year in captivity; Scripps Institute of Oceanography, California) and wild environments (Moss Landing and La Jolla, California).

Results: Here we report captive environments do not drive microbiome composition of T. semifasciata to significantly diverge from wild counterparts as life-long captive sharks maintain a species-specific epidermal microbiome resembling those associated with semi-captive and wild populations. Major taxonomic composition shifts observed were inverse changes of top taxonomic contributors across captive duration, specifically an increase of Pseudoalteromonadaceae and consequent decrease of Pseudomonadaceae relative abundance as T. semifasciata increased duration in captive conditions. Moreover, we show captivity did not lead to significant losses in microbial α-diversity of shark epidermal communities. Finally, we present a novel association between T. semifasciata and the Muricauda genus as MAGs revealed a consistent relationship across captive, semi-captive, and wild populations.

Conclusions: Our report illustrates the importance of conservation programs for coastal fishes as epidermally-associated microbes of near-shore shark species do not suffer detrimental impacts from long or short-term captivity. Our findings also expand on current understanding of shark epidermal microbiomes, explore the effects of ecologically different scenarios on benthic shark microbe associations, and highlight novel microbial associations that are consistent across captive gradients.

General Microbiology

Chondrichthyes

shark skin microbiome

metagenome-assembled genomes

captivity

leopard shark

Triakis semifasciata

The eukaryotic organism harbors an astounding number of microbes both internally and externally to supplement nutrient breakdown, thwart pathogenic colonization, and provide defense via molecular byproducts, and is known as the microbiome. The eukaryotic host and the respective microbiome has been reclassified as a meta-organism, representing a functional system that adapts to changing environments and external threats by cultivating a synergistic interdependence with microscopic symbionts (Bosch and McFall-Ngai 2011). The epidermal microbiome of the marine metaorganism reflects both environmental conditions and host-microbiome interdependent effects, in addition to directly interfacing with seawater, while remaining distinct from the reservoir of microbes in the water column. The epidermal microbiomes belonging to triple fin fish (Forsterygion capito) experiencing polluted conditions, for example, have lower levels of biodiversity compared to fish residing in pristine environments (Montenegro et al. 2020). Likewise, microbial profiling of stranded leopard sharks (Triakis semifasciata) revealed infections caused by the pathogenic bacteria Miamiensis avidus, thus causing the mass die-offs of T. semifasciata populations in the San Francisco Bay the summer of 2017 (Retallack et al. 2019). While the presence of unicellular species is telling, metabolic potentials of epidermal microbiomes is equally descriptive of environmental conditions. The epidermal microbiomes belonging to thresher sharks (Alopias vulpinus) encoded higher levels of genes associated with heavy metal resistance (Doane et al. 2017, 2020), indicating an increased supply of cobalt, zinc, cadmium, and iron from the host as Chondrichthyan (cartilaginous) fishes are known to attain heavy metals in the integumentary tissues (Jeffree et al. 2006). Therefore, the taxonomy and functional profiles of epidermal microbiomes to serve as indicators of environmental conditions, disease, and the provision of host-supplied nutrients. To this point, data for microbiomes belonging to captive, vertebrate hosts is still lacking, and is exceptionally dearth for the Chondrichthyan clade.

Extant sharks (Chondrichthyes subclass: Elasmobranchii, superorder: Selachimporhpa) are vertebrates showcasing an ancient host-associated microbiome evolution in addition to harboring diverse biophysical and physiological traits such as the ability to naturally produce bioluminescence via photophores (Gruber et al. 2016) and the ability to detect the electromagnetic fields produced by all living organisms using ampullae of Lorenzini (Kalmijn 2019). The shark epidermis, in particular, is exceptional due to the presence of overlapping dermal placoid scales (denticles) connected to numerous, horizontally-arranged layers of muscles that are known to contract, thereby optimizing the build-up of hydrostatic pressure during fast swimming (Bechert, Bruse, and Hage 2000; Meyer and Seegers 2012). The physiology of the shark epidermis supports protective functions similar to those of Actinopterygii (Ray-fin fish), whereby continuous shedding of denticles, in concert with low drag and the flexion of the scales, reinforces antifouling efforts. The shark epidermis is not completely free of biofilms; although elasmobranchs produce a reduced mucosal layer relative to teleost fish, several species including T. semifasciata, A. vulpinus, the whale shark (Rhincodon typus), and the round ray (Urobatis helleri) have been observed to have a microbiome that mirrors the respective evolutionary trajectory of the host (Doane et al. 2020). This concept by which the microbiomes reflect host phylogeny is called phylosymbiosis. The combination of the physical and chemical barrier that is the mucosal layer, acting jointly with a maintained microbiome across evolutionary time, may account for the few reported cases of infections witnessed in wounded, wild black-tip reef shark (Carcharhinus melanopterus) populations (Pogoreutz et al. 2019), where pathogens were absent in the epidermal microbiome. Investigations into shark-associated epidermal microbiomes, therefore, are necessary to supplement the scarcity of information concerning the shark metaorganism.

While host-related factors including biological sex, health status (Egan et al. 2018), and diet. (Antwis et al. 2014) regulate microbial recruitment and retainment, environmental such as geography, pH, and salinity also contribute to the composition of microbial communities and functions (Haggerty and Dinsdale 2017; Ross, Rodrigues Hoffmann, and Neufeld 2019). Investigating microbiomes of sharks in captivity environments such as public aquariums, provide consistent conditions with reduced introduction of natural microbial communities due to water filtration, stable macroscopic species interactions, regular diets, and lower variations of abiotic factors that influence the composition of both host-associated, and water column, microbiomes. Of the 1370 species of Elasmobranches, only a small percentage of associated microbial communities have been characterized including those previously mentioned in addition to the nurse (Ginglymostoma cirratum), lemon (Negaprion brevirostris), sandbar (Carcharhinus pleumbeus), caribbean reef (Carcharhinus perezii), and tiger (Galeocerdo cuvier) sharks and the thornback (Raja clavate) and round (Urobatis halleri) stingrays (Doane et al. 2017, 2020; Storo et al. 2021). A single study of a captive population was performed with cownose rays (Rhinoptera bonasus) exclusively residing in an aquarium touch tank (Kearns, Bowen, and Tlusty 2017). Herein, we investigate the impact that a captive environment exerts on the epidermal microbiota associated with captive T. semifasciata individuals and compare the composition and functional potentials of microbiomes belonging to wild sharks that live in nearshore habitats.

The T. semifasciata species was chosen as our model to investigate the change in microbiome composition across environments as they commonly inhabit both aquaria year-round and aggregate along the coast of southern California in the spring, summer, and fall months (Nosal et al. 2013). The T. semifasciata elasmobranch species has also been shown to have a phylosymbiotic relationship with their epidermal microbiomes (Doane et al. 2020). However, describing the epidermal microbiomes of T. semifasciata across captive and wild environmental gradients has yet to be accomplished. Therefore, the proximity of T. semifasciata populations along coastal waters and those concomitantly housed in nearby research facilities provided us a unique opportunity to sample epidermal microbiomes of captive (>1 year in captivity), temporarily captive (<1 year in captivity), and wild individuals using shotgun metagenomics. Our analyses show the taxonomic composition of the epidermal microbiomes belonging to T. semifasciata remain consistent between captive sharks and wild peers and irrespective of captive conditions. Furthermore, we show broad functional genes remain largely unchanged across environments and host lifestyles, as associated bacteria perform comparable functions, differing in few specific functions only which may reflect environmental conditions and resources and host-derived nutrients. Last, by constructing metagenomes assembled genomes, we confirm the presence and novel association of specialist microbes that are consistent across wild and captive sharks.

Sampling of Metagenomes

Wild T. semifasciata were sampled in the summers of 2013 (n = 6) and 2017 (n = 8) at La Jolla Shores in San Diego, California (32.868342, -117.255304), and the summer of 2015 (n = 5) at Moss Landing in Monterey County, California (36.801588, -121.791970). Wild sharks were sampled during the summer aggregations that are characteristic of T. semifasciata populations (Nosal et al. 2013, 2014). Wild sharks were initially caught using a handline with baited barbless circle hook and, using a scoop net, brought onboard for sampling and subsequent release. Four T. semifasciata (n = 4) were sampled in short term captivity (< 1 year) during the summer of 2016 while cohoused in a tank at the Scripps Research Institute of Oceanography in San Diego, California (32.865402, -117.252945). Captive T. semifasciata (n = 4) were sampled in the summer of 2018 at the Birch Aquarium at Scripps Institution of Oceanography in La Jolla, California (32.865658, -117.250535) to represent microbiomes belonging to T. semifasciata long held (>1 year) in captivity. Captive sharks were corralled by divers into a pen and immobilized in a sling during sampling.

To consistently describe the shark epidermis-associated microbiome, microbes were collected between the pectoral fin and dorsal fin along the lateral line on the left side of the organism (Fig. 1.A) using a blunt, closed-circuit syringe. The syringe is prefilled with the 100 kDa filtered seawater that was then flushed over the epidermis to dislodge microbes without introducing microbes from the environment (Figure 1.B; Doane et al. 2017). The approximate 200 mL of captured microbe was collected on a sterivex, where one sterivex per individual was obtained each collection date.

To describe the water-associated microbial communities, bulk water samples were taken: ~60 L of tank or ocean water were simultaneously collected and filtered once through a nylon mesh sieve (200 µm pore size) to remove large debris and eukaryotic organisms and again using tangential flow filtration (TFF, 100 kDa) to further concentrate the microbial consortia (Dinsdale et al. 2008). The resulting sample of approximately 500 mL composed the concentrated sample and was processed and collected on a Sterivex filter (0.22 µm).

DNA Extraction and Metagenome Sequencing and Annotation

Upon collection, cells were lysed within the sterivex filters and purified using a Macherey Nagel Tissue Kit. Microbiome DNA was then prepared for shotgun metagenomic sequencing using the Swift 2S Plus Kit (Swift Biosciences). Purified DNA was sequenced using an Illumina MiSeq sequencer with Mi-Seq v3 Reagent Kit on site at San Diego State University as previously performed (Dinsdale et al. 2013; Doane et al. 2017; Haggerty and Dinsdale 2017). Following sequencing, reads were processed for quality control via PrinSeq to remove artificial duplicates, and reads having >10 N’s and <60 bp as per pervious metganemomic workflows (Schmieder and Edwards 2011). High quality, paired reads were uploaded to Metagenomic Rapid Annotations using Subsystems Technology (MG-RAST) to call taxonomic and functional gene assignments. MG-RAST accomplishes gene calling and protein prediction with BLAST comparisons to the NCBI and SEED protein databases (Overbeek et al. 2014). Sequencing annotations were conducted using the following parameters: e-value >10⁻⁵, 70 percent identity, and > 60 bp alignment length.

MAG Assembly and Annotation

Metagenome assembly genomes were constructed to identify whether metagenomes from the sharks across captivity and wild environments contributed sequences to the genomes, thus potentially identifying shark specific microbes. The MAGs were generated from metagenomes belonging to 38 individual T. semifasciata across captivity status using MEGAHIT (Li et al. 2015). Contigs >1000 bp were run through METABAT2 binning program to reconstruct MAGs. CheckM was applied to each MAG and for the remainder of this article, only MAGs with at least 80 % completeness and less than 10 % contamination (confirmed via CheckM; Papudeshi et al. 2017; Parks et al. 2015, 2017) will be presented and discussed following quality control metrics outlined in previous investigations (Nayfach et al. 2020; Robbins et al. 2019). Bacterial delineations of MAGs were compared across several taxonomic classification algorithms including FOCUS (Silva et al. 2014), CheckM, and the Pathosystems Resource Integration Center (PATRIC; Wattam et al. 2014), and confirmed by comparing average nucleotide identity (ANI; Yoon et al. 2017) or DNA Kmer-based evidence of similarity surpassing 90 %, reflective of standard criteria (Chun and Rainey 2014). Of the 54 bins, eight met baseline requirements for further analysis, with three MAGs taxonomically classified with high confidence (> 90 %) and three identified with low (< 90 %) confidence. We utilized the advanced analysis and visualization tool Anvi’o to visually explore and interpret our assembly-based metagenomes (Eren et al. 2015). Functional gene annotation of MAGs was performed through PATRIC using the BLAT alignment tool.

Statistical Analyses

All taxonomic data was normalized to produce relative abundance data and was then fourth root transformed to reduce variation among samples. The α-diversity indices of microbial community richness, evenness, and diversity were measured using Margalef’s (d), Pielou’s (J’), and Inverse Simpson’s (1/λ) indices, respectively (Chao and Shen 2003; Death 2008; Pielou 1966) to discern the effect captivity exerts over associated microbial community biodiversity. To measure the intraspecific similarities, and between group dissimilarities, similarity percentages breakdowns (SIMPER) were calculated (Clarke 1993). To measure variation community compositions between environmental groups a permutation multivariate analysis of variance (PERMANOVA) using 999 random permutation tests was utilized (Kelly et al. 2015). Association mapping between metagenomes across environments were visualized via non-metric multidimensional scaling (nMDS) generation with correlation overlays (Kruskal 1964). Tests for differences in group dispersion or homogeneity of the multivariate variations were generated using a PERMDISP analysis (Anderson 2006; Clarke and Gorley 2015).

From the epidermal microbiomes of 27 T. semifasciata individuals, 31 114 584 sequenced reads were identified as bacteria and archaea spanning 27 phyla, 42 classes, 92 orders, 208 families, and 564 genera (Table 1). The total number of families present for T. semifasciata metagenomes ranged from 164 to 208 phyla. Water-associated microbiomes significantly differed from host-associated microbiomes in taxonomic composition (PERMANOVA: Family, Pseudo-F _{df = 1, 31} = 2.056, P(perm) < 0.05). The water column microbial communities had little within-group variation (SIMPER analysis; 72.96 similarity) and high dissimilarity and are not compared further.

Table 1

Base pair and sequence count for T. semifasciata across environments.
Sample	Base Pair Count	Sequence Count
Captive 1	13,254,680	36,040
Captive 2	165,572,548	563,299
Captive 3	129,410,990	454,887
Captive 4	140,813,416	512,408
Semi-Captive 1	194,456,844	683,590
Semi-Captive 2	114,129,299	364,329
Semi-Captive 3	106,800,378	351,397
Semi-Captive 4	147,769,235	491,188
Wild 1	263,295,773	1,298,868
Wild 2	262,597,687	1,176,351
Wild 3	258,939,703	1,226,960
Wild 4	653,532,367	2,531,410
Wild 5	640,031,260	2,434,754
Wild 6	673,699,889	2,549,181
Wild 7	397,391,784	1,368,629
Wild 8	273,016,668	1,039,793
Wild 9	291,042,619	1,045,530
Wild 10	291,228,603	1,001,807
Wild 11	369,123,866	1,254,908
Wild 12	211,033,215	710,743
Wild 13	279,527,632	914,034
Wild 14	195,358,159	648,568
Wild 15	109,131,187	328,208
Wild 16	94,795,257	316,833
Wild 17	186,632,228	623,111
Wild 18	342,901,940	1,246,300
Wild 19	217,511,438	709,664

Microbial community richness (d), evenness (J’), and overall diversity (1/λ) were similar across captivity status (Table 2). To understand the taxonomic composition and variation across environments we then characterized each group of metagenomes. The major contributors to taxonomic classes of epidermal microbiomes belonging to T. semifasciata long held in captivity were Alteromonadales (57.1% ± 4.29 S.E.), Burkholderiales (5.91% ± 0.79) and Sphingomonodales (5.51% ± 0.85), belonging to Gamma-, Beta-, and Alphaproteobacteria clades respectively. Of the marine generalists, the Pseudoalteromonadaceae family were the most abundant family present (30.1% ± 2.42), followed by Alteromonodales (20.01% ± 1.60), Sphingomonadaceae (4.48% ± 0.73), and Shewanellaceae (2.62% ± 0.14; Figure 2). The epidermal microbiomes of T. semifasciata inhabiting the southern coast of California harbored different abundances of major clades than their captive counterparts; fine-scale taxonomic resolution revealed differences between captive and wild benthic shark microbiomes were accounted for by major taxonomic contributors (>1% relative abundance), with smaller populations having no significant impact on microbiome compositions. While the Pseudoalteromonadaceae family dominates the skin microbiomes of T. semifasciata residing in captivity (21.4 ± 1.45 S.E.), the Flavobacteriaceae family contributed the most to metagenome compositions belonging to wild T. semifasciata (10.13 % ± 1.98), followed by Pseudomonadaceae (8.86 % ± 1.87) and Alteronomonadaceae (7.26 % ± 0.94). The Pseudomonadaceae (18.75 % ± 2.87) and Flavobacteriaceae (8.38 % ± 0.337) families were among the top three major taxonomic contributors to T. semifasciata microbiomes for semi-captive shark microbiomes, with the Moraxellaceae (11.26 % ± 2.12) family comprising significantly more of the microbial composition than in other groups.

Table 2

Table 1 Average biodiversity indices for epidermal microbiomes belonging to T. semifasciata across environments.
Host Environment	Margalef’s (d) Index ± S.E.M.	Pielou’s (J) Index ± S.E.M.	Inverse Simpson (1/λ) Index ± S.E.M.
Captive	41.21 ± 4.15	0.59 ± 4.3E-02	70.50 ± 3.54
Semi-captive	41.53 ± 3.54	0.624 ± 3.37E-02	73.30 ± 2.27
Wild	40.07 ± 1.44	0.581 ± 2.88E-02	68.29 ± 2.29

While no statistical differences were observed between wild and neither captive nor semi-captive epidermal microbiomes (PERMANOVA Family level, Pseudo-F _{df =2, 18} =, P(perm) > 0.1); Genus level, Pseudo-F _{df =2, 18} =, P(perm) > 0.1), SIMPER analyses revealed trends along captive durations: microbiomes belonging to captive sharks are more similar to those associated with semi-captive sharks (87.99 Average Similarity (A.S.) = 100 - Average Dissimilarity) than to wild sharks (84.27 A.S.), and semi-captive sharks are more similar to captive sharks than to wild sharks (85.6 A.S.). Dissimilarity between captive and wild T. semifasciata was attributed to higher abundances of Pseudoalteromonoadaceae in captive microbiomes, consistently driving differences between the two groups (3.04 Diss/S.D.). Taxonomic clades accounting for consistent differences between semi-captive and wild sharks were Alcanivoraceae (2.79 Diss/S.D.), Planctomycetaceae (2.47 Diss/S.D.), and Moraxallanceae (2.38 Diss/S.D.), and were higher in proportional abundance in semi-captive metagenomes. The constant contributor of dissimilarity between captive and semi-captive individuals was attributed to Halomonodaceae (3.74 Diss/S.D.), which was approximately twice the relative abundance in semi-captive epidermal microbiomes. There was no difference in the microbiome variation of T. semifasciata epidermal microbiomes residing in each environment (PERMDISP: Pseudo-Family, F _{df =2, 18} = 1.508, P(perm) > 0.1, Genus, Pseudo-F _{df =2, 18} = 1.32, P(perm) > 0.1). The microbial families of T. semifasciata epidermal microbiomes did not form distinct clusters for each environment when visualized using an nMDS; the structure of T. semifasciata epidermal microbiome data across captivity showed no difference in coefficients of similarity and several microbial families were found to be highly correlated (> 0.85; Figure 3). These families include Pseudomonadaceae, Shewanellaceae, and Sphingomonodaceae, and represent major taxonomic contributors (>1%) for all three groups of T. semifasciata. Less represented clades (<1%) related to the spread of metagenomes include Beijerinckiaceae, Brucellaceae, Rickettsiaceae, and Bacteroidaceae as illustrated in nMDS overlays (Figure 3).

Comparisons of functional gene potentials of T. semifasciata epidermal microbiomes across environments

There were between 661 and 778 metabolic categories identified in all T. semifasciata metagenomes. Water column metagenomes functional potentials were significantly distinct from T. semifasciata epidermal microbiomes (PERMANOVA: subsystem level 1, Pseudo-F _{df = 1, 31} = 1.93, P(perm) < 0.05) and were not compared further.

Of the 27 major gene pathways identified in T. semifasciata metagenomes, six contributed >5% of reads, with the two top contributors (>10%) involving carbohydrate (12.2% ± 0.196 S.E.) and amino acid (11.0% ± 0.151) metabolism. Analyses of functional levels revealed no statistical differences between all groups (PERMANOVA: subsystems level 3, Pseudo-F _{df 2, 18} = 0.859, P(perm) > 0.5). In addition, differences in similarity between groups were insignificant, as wild metagenome metabolic profiles were comparable (SIMPER: wild vs semi-captive, 90.02 A.S.; captive vs semi-captive, 91.42 A.S.) with the greatest dissimilarity measured between wild and semi-captive shark populations (90.4 A.S.) when analyzing metabolic pathways (SEED subsystem level 2). The functions contributing consistently to differences between wild and semi-captive epidermal microbiomes, as identified by SIMPER analysis, were genes involved in gene transfer agents (1.95 % contribution, 1.95 Diss/S.D.), motility and non-flagellar swimming (1.69 %, 1.38 Diss/S.D.), and protein secretion system type VII (1.47 %, 1.35 Diss/S.D.).

Although β-diversity analyses revealed no statistical differences between functional profiles of metagenomes belonging to T. semifasciata epidermal microbiomes across environments, several functional pathways were found to differ. Captive shark metagenomes featured increases in relative abundances of genes; in carbohydrate synthesis (14.06% ± 0.24 S.E.) and utilization, such as the more specific fermentation (0.959% ± 0.06; Figure 4) pathway (SEED subsystem: level 2). Specific functional pathways (SEED subsystem: level 3) featuring utilization of simple sugars and sugar alcohols were also significantly increased in captive metagenomes including fructose and mannose inducible phosphotransferase system (PTS; 0.458% ± 0.01), methylglyoxal metabolism (0.37 ± 0.01), and mannitol metabolism (0.192% ± 0.26, Figure 5). Genes coding for enzymes involved in the breakdown of these saccharides i.e., beta-glucosidase (0.48% ± 0.01), were correspondingly increased in captive shark epidermal metagenomes compared to semi-captive and wild counterparts, in addition to the subsequent alcohol synthesis including acetone, ethanol, and butanol (0.20% ± 0.09). Pathways involved in heavy metal acquisition were likewise increased in captive shark microbiomes, as indicated by increased genes involved in heme/hemin uptake and utilization (0.19% ± 0.03), iron acquisition (2.65% ± 0.17). Furthermore, pathways involved in virulence, disease, and defense were significantly more abundant (5.85% ± 0.03), including specific genes related to periplasmic stress (0.02% ± 0.01), capsular polysaccharide biosynthesis and assembly (0.13% ± 0.013), murein hydrolytic activities (0.341% ± 0.01), and the antibiotic resistance gene BlaR1 family regulatory sensory-transducer disambiguation (0.34% ± 0.32). Conversely, higher functional potentials involved in vitamin synthesis (7.22% ± 0.145 S.D., p < 0.01) and nitrogen metabolism (1.82% ± 0.001) were observed in semi-captive samples. For T. semifasciata individuals, no significant gene pathways differed between semi-captive or wild metagenomes.

Metagenome Assembled Genomes constructed from Microbial Communities Associated with T. semifasciata

Cross assembly of the 27 T. semifasciata metagenomes yielded 54 MAGs containing 241 814 contigs greater than 1 kilobase pairs, with N50 of 735 bp and N75 of 583 bp. Of these, nine high quality MAGs were constructed spanning seven known bacterial phyla (Supplementary Table 1). While all T. semifasciata metagenomes were involved in MAG assembly, three groupings of contributions to MAG generation can be observed in Figure 6, where more even mean coverage of MAG contribution by T. semifasciata populations across environment is visualized. Among the groups, group 1 in Figure 6 highlights heavy contributions from both captive and wild metagenomes, while not harboring any of the nine high quality MAGs, while group 2 contains two MAGs further investigated. Finally, group 3 has an even spread of mean contribution from each metagenome environments and contains three high quality MAGs. The number of contigs wild shark hosts contributed to MAG assembly was greater (87.1% ± 4.88 S.E.) than both semi-captive (7.45% ± 6.86) and long-held captive sharks (5.49% ± 4.50) and is due to number of reads in the metagenomes.

Of the nine qualifying MAGs, two (Bin 27 and Bin 9) were identified as belonging to the Muricauda genus, an increasingly classified child taxon of Flavobacteriaceae family. Bin 27 had an average nucleotide identity match of 99.75% with the Muricauda antarctica species (Yoon et al. 2017), while identification of the species of Bin 9 remains incompletely verified (<90% similarity) at species level, with the highest resemblance matched to Muricauda reustringensis (84% similarity). The DNA G+C content of Bin 27 and Bin 9 was 45.17% and 41.72% respectively, both falling within the acceptable range reported for taxa belonging to the Muricauda genus, i.e. 41-45.4 mol% (Arun et al. 2009). Bin 27 featured a 95.77% complete genome, with 4 106 genes, 4.23 % genome redundancy, and a total length of 4 285 655 bp. The Bin 9 MAG was calculated to have a 91.55 % complete genome composed of 4 673 genes, with more genomic redundancy (7.04%) and a longer total genome length of 4 625 638 bp (Figure 7). Following genome annotation, no resistance or susceptibility to antibiotics were found in Bin 27 encoding for Muricauda antarctica. However, two virulence features were discovered: GTP-binding and nucleic acid-binding protein YchF (fig|1055723.17.peg.1964), and ferric uptake regulation protein FUR (fig|1055723.17.peg.1945).

Of the remaining seven MAGs, two were of unknown taxonomic origin with no similar genomes (Bin 36 and 47). In decreasing order of confidence, the five remaining MAGs were identified as Zunongwangia atlantica (99.75%, Bin 30), Roseivirga pacifica (96.5%, Bin 30), Leeuwenhoekiella blandensis (63.9%, Bin 22), Micavibrio spp. (39.2%, Bin 13), and Pseudomonas spp. (34.4%, Bin 12; Supplementary Table 1). Finally, Bin 36 most closely resembled a member of the Fluviicola genus (23.34% similarity), while Bin 47 matched most (20% similarity) to the Thalassospira genus.

Captive environments aim to conserve populations while balancing a complex ecosystem of micro- and macro-organisms. These efforts necessitate characterization of elasmobranch microbiomes across environments to understand the effect captive exerts on the meta-organism. We provide the first comparisons of taxonomic compositions and functional potentials between metagenomes associated with captive, semi-captive and wild T. semifasciata populations and show that captive environment exerts no detrimental pressure to T. semifasciata epidermal microbiome composition. In addition, there was no effect on biodiversity across environments. However, we did observe changes in microbial community structure as sharks increased captive status, and only more specific (SEED subsystem: level 2 and 3) functional pathways varied within metagenomes.

Epidermal Microbiome Taxonomic Structure as a Product of Captivity Duration and Host Factors

Here we show the composition of T. semifasciata populations differ in proportion of major contributors (>10% relative abundance) between captive and wild environments; we observed the Pseudoalteromonadaceae relative abundance to be significantly higher in captive samples, while their presence was only maintained in wild microbiomes. Consequently, we observed a shift in taxonomic compositions as T. semifasciata populations increased durations in captivity, including decreasing proportions of Pseudomonadaceae and Moraxellaceae, as Flavobacteriaceae and Pseudoalteromonadaceae became the dominant phyla in captive environments. The Pseudomonadaceae family, which belongs to the higher Gammaproteobacteria phylum and encompasses a diverse clade of microbes known to perform essential tasks such as nitrogen fixation and contaminant degradation. The observed change from microbiomes dominated by generalists (Pseudomonadaceae) in wild environments to opportunists (Pseudoalteromonadaceae, Flavobacteriaceae) in captive metagenomes may signify wild environments drive the selection of fast-growing associates by providing an abundance of resources. Likewise, the increased relative abundance of the copiotrophic Pseudomonadaceae in wild groups may reflects unclean, near-shore conditions resulting from anthropogenic forces. The mostly aerobic Flavobacteriaceae family of microbes are physiologically similar to the Pseudomonadaceae and Pseudoalteromonadaceae families (Agronomique, Nakagawa, and Holmes 2002). However, the increased relative abundance of Flavobacteriaceae in captive metagenomes may reflect a reduced available nutrient as microbes utilizing slow growth to navigate starvation naturally outcompete coptiotrophic rivals.

The epidermal microbiome alpha-diversity showed no significant differences across host environment, with less abundant taxa (<1%) accounting for little dissimilarity. Thus, we theorize the environment influences the relative abundance of the most abundant taxa while the hosts recruit taxon and maintain community diversity of epidermal microbiomes. Conversely, host factors such as biological sex, size, age, and pregnancy are known to play a role in shaping microbial associations the effect captive environments exert over major contributors to epidermal microbiomes suggests these and other host factors help maintain recurrent compositions that are worth exploring in future analyses. For example, the increased presence of Moraxellaceae in semi-captive sharks may reflect the transitional period from wild to captive lifestyles or a compounding effect of pregnancy as the semi-captive population was gravid during sampling. Therefore, while we have provided an important baseline representing the microbiomes of wild, semi-captive, and captive T. semifasciata populations, the gravid status of hosts may play a larger role in microbial community composition than the environment and individual host factors including age, sex, and weight should be investigated further.

Metabolic Potentials of Captive Shark Microbiomes Reflect Environmental Conditions

The epidermal microbiomes of captive T. semifasciata exhibited different metabolic potential although only at SEED subsystem level 2 and 3 and these higher-resolution variations are attributed to environmental differences. For example, captive T. semifasciata epidermal metagenomes had higher levels of several pathways involved in carbohydrate metabolism including mannitol utilization and fructose and mannose inducible phosphotransferase. While increased abundances of carbohydrate metabolism pathways in captive shark microbiomes are indicative of high levels of simple sugars in the environment, higher observed levels methylglyoxal pathway (MG) potential in captive metagenomes suggests excess nutrients available for captive microbes. The MG pathway produces aldehyde methylglyoxal instead of adenosine triphosphate, a toxic electrophile to cells. The MG pathway is therefore a low-energy-yielding glycolysis bypass that must be tightly regulated and is theorized to occur as a high-risk mechanism to harbor excess nutrients when catabolic rates increase, specifically sugar phosphate levels (Weber, Kayser, and Rinas 2005). Correspondingly, alcohol synthesis pathways i.e., acetone butanol ethanol synthesis gene pathway, metabolically follow carbohydrate metabolism and were also elevated in captive samples. Adjacent to nutrient metabolism, genes (SEED subsystems: Level 3) annotated for mycolic acid synthesis, peptide methionine sulfoxide reductase, and chorismite synthesis were all measured higher in captive metagenomes, indicating higher rates of cell wall synthesis, protein processing, and amino acid synthesis respectively. From these findings we surmise microbes associated with captive sharks benefit from stable abiotic conditions in aquaria, utilizing an overabundance of nutrients and experiencing increased catabolism compared to wild counterparts as a result of scheduled aquaria feeding and supplement offerings.

While captive metabolic pathways indicated the amount of environmentally available carbohydrates exceed wild environmental levels, limited heavy metal availability was equally inferred. Several specific, key functions found to be significantly higher in captive samples revealed captive environments to have iron limiting conditions i.e. metabolic pathways such as iron acquisition, heme uptake and hemin utilization indicate a need for bacteria to sequester iron when the heavy metal is not readily found in the environment (Gunn et al. 2019). Competitive microbial interactions in captive environments could also be inferred from increased bacteriophage and vibrio-related genes such as murein hydrolases i.e. hydrolytic enzymes known to cleave bacterial cell walls during final stages of the lytic cycle (Moak and Molineux 2004), and aspartate amino transferase i.e. enzymes key for bacteriophage infection (Chen et al. 2013). Last, shark microbiomes sampled in the captive environment displayed overall higher rates of protein misfolding within bacterial cell walls (Merdanovic et al. 2011), as seen in periplasmic stress gene. Overall, captive environmental conditions prompt higher rates of nutrient-hoarding, phage interactions, and heavy-metal sequestering among microbes associated with elasmobranch hosts. Further metagenomics studies should determine which factors besides captivity e.g., temperature, light, etc, also promote these effects. In addition, while this data was collected from metagenomics, these findings could be further explored by metatranscriptomics.

MAGs reveal Novel, Constant Microbial Associations with T. semifasciata

Construction of genomes confirmed several microbial associations with T. semifasciata in addition to metabolic contributions these microbes encode. For one such association, we confirmed a novel and continuous relationship with two bacteria belonging to the Muricauda genus associated with T. semifasciata across captative environments. Historically, this genus of marine microbes is genomically underexplored and free living, isolated from both seawater and marine sediments in Antarctica (Lee et al. 2012; S. Q. Liu et al. 2018; Yang et al. 2013; Zhang et al. 2018) and more recently from the gills of shrimp in Okinawa (L. Liu et al. 2020). Therefore, this is the first instance of Muricauda genus being associated with a vertebrate host organism and may be a shark-specific microbe. The associative driving force between the Muricauda genus and Chondrichthyes is not immediately apparent: while other deep-sea, marine bacteria are capable of utilizing trimethylamine N-oxide (TMAO; a well-known organic compounds used by sharks and other fish for osmotic regulation), as a carbon and nitrogen source, the Muricauda genus was shown to lack the necessary enzymatic activity to reduce the macromolecule (Yin et al. 2019). Although some members of the Bacteroidetes phylum are characterized as opportunistic pathogens, the Muricauda antarctica species is deemed a low pathogen risk.

Additional MAG annotation belonging to the Bacteroidetes phylum was Bin 21, identified as Roseivirga pacifica. The Roseivirga genus has been isolated from seawater and sea urchins (Strongylocentrotus intermedius), an organism featured in benthic shark diets and a likely source of recruitment (Nedashkovskaya et al. 2005). The annotation of Bin 30 as Zunongwangia atlantica confirmed another microbe discovered in a deep-sea environment (Shao et al. 2014), and like the Muricauda genus, is also non-motile and strictly aerobic. Bin 22, identified as Leeuwenhoekiella blandensis, similarly aerobic, prefers warmer growth conditions (28-30˚ C; Pinhassi et al. 2006) and may be explained by aggregations of the sharks that our in summer months in shallower waters.

While we were unable to annotate Bins 36 and 47 with confidence, the MAGs were found to contain the least contamination of the nine investigated MAGs. Uncharacterized MAGs are thought to be novel species as their sequences were highly complete. Future annotation of these genomes will no doubt reveal unique host-microbe associations and implications for host-microbe interactions. It should be noted the assembly of MAGs revealed a significantly larger contribution of contigs from wild samples, suggesting higher microbial loads in wild environments.

We investigated the associated microbial communities belonging to T. semifasciata housed since birth at the Birch Aquarium at Scripps Institution of Oceanography in La Jolla, a population held temporarily (<1 year) at the Scripps institute of Oceanography, and wild populations in La Jolla and Moss Landing, California during summer months. We found these leopard shark epidermal microbiomes to be taxonomically distinct from respective environmental water columns with some inter-location variability. Our investigations into the impact of captivity on microbiome profiles revealed no significant differences in alpha-diversity indices, proportional abundance of taxonomy or function, and therefore captivity exert no detriment on biodiversity of shark-associated epidermal microbiomes.

We also measured the similarity of epidermal microbiomes community compositions belonging to T. semifasciata across environments to be the same, albeit with varied structures, as beta-diversity measures showed no significant difference between the groups. We also discovered as sharks are held in captivity for longer durations, their microbiomes proportionally deviate from wild hosts; our results show captive environments influence relative abundances of key generalist bacteria while hosts regulate the presence of microbes. Last, metagenome assembled genomes from T. semifasciata epidermal microbiomes identified and confirmed novel and consistent associations between the Muricauda, Zunongwangia, Roseivirga and Leeuwenhoekiella genera and the shark hosts. All shark groups contributed to the generation of the genomes (Figure 6), confirming the persistent presence of these microbes associated with T. semifasciata.

Abbreviations:

Not applicable

Ethical Approval and Consent to participate:

Leopard shark samples was conducted under ethics approval from the SDSU ethics board APF 17-11-010D, APF 18-05-007D, APF # 014-05-011D, and APF # 11-02-004D. Birch aquarium sharks were sampled under the ethic approval by the Birch Aquarium ethic board.

Department of Fish and Game Scientific collecting permit: 9893, Dr. Andrew Nosal, Scripps Institute of Oceanography, secured June 4th, 2016. Department of Fish and Game Scientific collecting permit: 12847, Dr. Michael Doane, Secured November 30th, 2014.

Consent for publication:

Not applicable

Availability of supporting data:

All data will be publicly released upon publication

Competing interests:

The authors declare no conflict of interest.

Funding:

We acknowledge support from S. Lo and B. Billings.

Authors' contributions:

AG collected field samples, analyzed the data, and wrote the manuscript, NP conducted bioinformatics, MD collected field samples, and constructed the ideas, CJ, EK, MM, captured sharks and prepared samples and provide feedback on the manuscript, LL and EK provided sequencing and analysis, AN facilitated the catching of leopard sharks and access to leopard sharks are Scripps Research Facility, MT and JN provided support for sampling Birch Aquarium Sharks and the design of the experiment, ED conceptualized the idea, analyzed, wrote and lead the project.

Acknowledgements: The authors would like to thank the Birch Aquarium at Scripps for allowing us to conduct research on site alongside staff. We acknowledge the S. Lo and B. Billings Fund for Global Shark Conservation for their support for the project. Students in the SDSU undergraduate ecological metagenomic class provide the sequencing of the leopard shark metagenomes.

Authors' information:

1. Department of Biology, San Diego State University, San Diego, California, USA, 2. College of Science and Engineering, Flinders University, Bedford Park, South Australia, Australia, 3. Birch Aquarium at Scripps Institution of Oceanography, University of California San Diego, USA, 4. Scripps Institute of Oceanography, La Jolla, California, USA

Agronomique R, Nakagawa Y, and Barry Holmes. 2002. “Proposed Minimal Standards for Describing New Taxa of the Family Flavobacteriaceae and Jean-Franc by the Members of the Subcommittee on the Taxonomy of Flavobacterium and Cytophaga -like Bacteria of the International Committee on Systematics Of.”: 1049–70.
Anderson MJ. Distance-Based Tests for Homogeneity of Multivariate Dispersions. Biometrics. 2006;62(1):245–53.
Antwis RE, et al. Ex Situ Diet Influences the Bacterial Community Associated with the Skin of Red-Eyed Tree Frogs (Agalychnis Callidryas). PLoS ONE. 2014;9(1):1–8.
Arun AB, et al. Muricauda Lutaonensis Sp. Nov., a Moderate Thermophile Isolated from a Coastal Hot Spring. Int J Syst Evol Microbiol. 2009;59(11):2738–42.
Bechert DW, Bruse M, Hage W. Experiments with Three-Dimensional Riblets as an Idealized Model of Shark Skin. Exp Fluids. 2000;28(5):403–12.
Bosch TCG, McFall-Ngai MJ. Metaorganisms as the New Frontier. Zoology. 2011;114(4):185–90. http://dx.doi.org/10.1016/j.zool.2011.04.001.
Chao A, Shen TJ. S Index of Diversity When There Are Unseen Species in Sample. Environ Ecol Stat. 2003;10:429–43. https://link.springer.com/content/pdf/10.1023%2FA%3A1026096204727.pdf.
Chen Y, Wei D, Yiqian Wang, and Zhang X. The Role of Interactions between Bacterial Chaperone, Aspartate Aminotransferase, and Viral Protein during Virus Infection in High Temperature Environment: The Interactions between Bacterium and Virus Proteins. BMC Microbiol. 2013;13(1):1. BMC Microbiology.
Chun J, Rainey FA. Integrating Genomics into the Taxonomy and Systematics of the Bacteria and Archaea. Int J Syst Evol Microbiol. 2014;64(PART 2):316–24.
Clarke KR. 1993. “Non-Parametric Multivariate Analyses of Changes in Community Structure.” (1988): 117–43.
Clarke KR, Gorley RN. 2015. “PERMANOVA+ Primer V7: User Manual.” Primer-E Ltd., Plymouth, UK: 93.
Death R. “Margalef’s Index.”. In: Encyclopedia of Ecology, Five-Volume Set. Elsevier Inc.; 2008. pp. 2209–10.
Dinsdale EA, et al. Microbial Ecology of Four Coral Atolls in the Northern Line Islands” ed. Niyaz Ahmed. PLoS ONE. 2008;3(2):e1584. https://dx.plos.org/10.1371/journal.pone.0001584.
Dinsdale EA, et al. Multivariate Analysis of Functional Metagenomes. Front Genet. 2013;4(APR):1–25.
Doane MP, et al. The Skin Microbiome of the Common Thresher Shark (Alopias Vulpinus) Has Low Taxonomic and Gene Function β-Diversity. Environmental Microbiology Reports. 2017;9(4):357–73.
Doane MP, et al. 2020. “The Skin Microbiome of Elasmobranchs Follows Phylosymbiosis, but in Teleost Fishes, the Microbiomes Converge.” Microbiome 8(1): 93. https://microbiomejournal.biomedcentral.com/articles/10.1186/s40168-020-00840-x (November 9, 2020).
Egan S, et al. The Inner Workings of the Outer Surface: Skin and Gill Microbiota as Indicators of Changing Gut Health in Yellowtail Kingfish. Front Microbiol. 2018;8:2664. .
Eren A, Murat, et al. Anvi’o: An Advanced Analysis and Visualization Platformfor ’omics Data. PeerJ. 2015;2015(10):1–29.
Gruber DF, et al. Biofluorescence in Catsharks (Scyliorhinidae): Fundamental Description and Relevance for Elasmobranch Visual Ecology. Sci Rep. 2016;6(January):1–16. http://dx.doi.org/10.1038/srep24751.
Gunn JS, et al. 2019. “Heme Uptake and Utilization by Gram-Negative Bacterial Pathogens.” Frontiers in Cellular and Infection Microbiology | www.frontiersin.org 1: 81. .
Haggerty JM, Elizabeth AD. Distinct Biogeographical Patterns of Marine Bacterial Taxonomy and Functional Genes. Glob Ecol Biogeogr. 2017;26(2):177–90.
Jeffree RA, Warnau M, Teyssié JL, Markich SJ. Comparison of the Bioaccumulation from Seawater and Depuration of Heavy Metals and Radionuclides in the Spotted Dogfish Scyliorhinus Canicula (Chondrichthys) and the Turbot Psetta Maxima (Actinopterygii: Teleostei). Sci Total Environ. 2006;368(2–3):839–52.
Kalmijn AdJ. 2019. “Electric and Magnetic Field Detection in Elasmobranch Fishes Author (s): Ad. J. Kalmijn Published by : American Association for the Advancement of Science Stable URL : Https://Www.Jstor.Org/Stable/1689048 Detection in Elasmobranch in Elasmobranch Fish.” 218(4575): 916–18.
Kearns PJ, Bowen JL, Tlusty MF. The Skin Microbiome of Cow-Nose Rays (Rhinoptera Bonasus) in an Aquarium Touch-Tank Exhibit. Zoo Biology. 2017;36(3):226–30.
Kelly BJ, et al. 2015. “Power and Sample-Size Estimation for Microbiome Studies Using Pairwise Distances and PERMANOVA.” Bioinformatics. http://github.com/brendankelly/micropower.
Kruskal JB. 1964. “MULTIDIMENSIONAL SCALING BY OPTIMIZING GOODNESS.” (1).
Lee S, Young S, Park TK, Oh, and Jung Hoon Yoon. Muricauda Beolgyonensis Sp. Nov., Isolated from a Tidal Flat. Int J Syst Evol Microbiol. 2012;62(5):1134–39.
Li D, et al. MEGAHIT: An Ultra-Fast Single-Node Solution for Large and Complex Metagenomics Assembly via Succinct de Bruijn Graph. Bioinformatics. 2015;31(10):1674–76.
Liu L, et al. Muricauda Alvinocaridis Sp. Nov., Isolated from Shrimp Gill from the Okinawa Trough. Int J Syst Evol Microbiol. 2020;70(3):1666–71.
Liu S, Qi, et al. Muricauda Iocasae Sp. Nov., Isolated from Deep Sea Sediment of the South China Sea. Int J Syst Evol Microbiol. 2018;68(8):2538–44.
Merdanovic M, et al. Protein Quality Control in the Bacterial Periplasm. Annu Rev Microbiol. 2011;65:149–68.
Meyer W, Seegers U. Basics of Skin Structure and Function in Elasmobranchs: A Review. J Fish Biol. 2012;80(5):1940–67.
Moak M, and Ian J. Molineux. Peptidoglycan Hydrolytic Activities Associated with Bacteriophage Virions. Mol Microbiol. 2004;51(4):1169–83.
Montenegro D, Astudillo-García C, Hickey T, Lear G. A Non-Invasive Method to Monitor Marine Pollution from Bacterial DNA Present in Fish Skin Mucus. Environ Pollut. 2020;263:114438. https://doi.org/10.1016/j.envpol.2020.114438.
Nayfach S, et al. 2020. “A Genomic Catalog of Earth’s Microbiomes.” Nature Biotechnology.
Nedashkovskaya OI, et al. Roseivirga Echinicomitans Sp. Nov., a Novel Marine Bacterium Isolated from the Sea Urchin Strongylocentrotus Intermedius, and Emended Description of the Genus Roseivirga. Int J Syst Evol Microbiol. 2005;55(5):1797–800. http://ijs.sgmjournals.org.
Nosal AP, et al. Demography and Movement Patterns of Leopard Sharks (Triakis Semifasciata) Aggregating near the Head of a Submarine Canyon along the Open Coast of Southern California, USA. Environ Biol Fishes. 2013;96(7):865–78.
———. 2014. “Aggregation Behavior and Seasonal Philopatry in Male and Female Leopard Sharks Triakis Semifasciata along the Open Coast of Southern California, USA.” Marine Ecology Progress Series 499: 157–75.
Overbeek R, et al. The SEED and the Rapid Annotation of Microbial Genomes Using Subsystems Technology (RAST). Nucleic Acids Res. 2014;42(D1):206–14.
Papudeshi B, et al. Optimizing and Evaluating the Reconstruction of Metagenome-Assembled Microbial Genomes. BMC Genom. 2017;18(1):1–13.
Parks DH, et al. CheckM: Assessing the Quality of Microbial Genomes Recovered from Isolates, Single Cells, and Metagenomes. Genome Res. 2015;25(7):1043–55.
Parks D, et al. Recovery of Nearly 8,000 Metagenome-Assembled Genomes Substantially Expands the Tree of Life. Nature Microbiology. 2017;2(11):1533–42. http://dx.doi.org/10.1038/s41564-017-0012-7.
Pielou EC. The Measurement of Diversity in Different Types of Biological Collections. J Theor Biol. 1966;13(C):131–44.
Pinhassi J, et al. Leeuwenhoekiella Blandensis Sp. Nov., a Genome-Sequenced Marine Member of the Family Flavobacteriaceae. Int J Syst Evol Microbiol. 2006;56(7):1489–93. https://research.venterinstitute.org/moore/.
Pogoreutz C, et al. Similar Bacterial Communities on Healthy and Injured Skin of Black Tip Reef Sharks. Animal Microbiome. 2019;1(1):1–16.
Retallack H, et al. Metagenomic Next-Generation Sequencing Reveals Miamiensis Avidus (Ciliophora: Scuticociliatida) in the 2017 Epizootic of Leopard Sharks (Triakis Semifasciata) in San Francisco Bay, California, USA. J Wildl Dis. 2019;55(2):375–86.
Robbins SJ, et al. A Genomic View of the Reef-Building Coral Porites Lutea and Its Microbial Symbionts. Nature Microbiology. 2019;4(12):2090–100.
Ross AA, Rodrigues Hoffmann A, and Josh D. Neufeld. 2019. “The Skin Microbiome of Vertebrates.” Microbiome 7(1): 79. https://microbiomejournal.biomedcentral.com/articles/10.1186/s40168-019-0694-6.
Schmieder R, and Robert Edwards. Quality Control and Preprocessing of Metagenomic Datasets. Bioinformatics. 2011;27(6):863–64. https://academic.oup.com/bioinformatics/article-lookup/doi/10.1093/bioinformatics/btr026.
Shao R, et al. 2014. “Zunongwangia Atlantica Sp. Nov ., Isolated from Deep-Sea Water.” : 16–20.
Silva G, Gueiros Z, Daniel A, Cuevas BE, Dutilh, Robert AE. FOCUS: An Alignment-Free Model to Identify Organisms in Metagenomes Using Non-Negative Least Squares. PeerJ. 2014;2:e425. https://peerj.com/articles/425.
Storo R, Easson C, Shivji M, and Jose V. Lopez. Microbiome Analyses Demonstrate Specific Communities Within Five Shark Species. Front Microbiol. 2021;12(February):1–10.
Wattam AR, et al. 2014. “PATRIC, the Bacterial Bioinformatics Database and Analysis Resource.” Nucleic Acids Research 42(D1). http://www.niaid.nih.gov/topics/biodefenserelated/biodefen.
Weber J, Kayser A, and Ursula Rinas. Metabolic Flux Analysis of Escherichia Coli in Glucose-Limited Continuous Culture. II. Dynamic Response to Famine and Feast, Activation of the Methylglyoxal Pathway and Oscillatory Behaviour. Microbiology. 2005;151(3):707–16. http://mic.sgmjournals.org.
Yang C, et al. Muricauda Zhangzhouensis Sp. Nov., Isolated from Mangrove Sediment. Int J Syst Evol Microbiol. 2013;63(PART6):2320–25.
Yin Q, et al. Contribution of Trimethylamine N-Oxide on the Growth and Pressure Tolerance of Deep-Sea Bacteria. Journal of Oceanology Limnology. 2019;37(1):210–22.
Yoon SH, et al. A Large-Scale Evaluation of Algorithms to Calculate Average Nucleotide Identity. Antonie van Leeuwenhoek International Journal of General Molecular Microbiology. 2017;110(10):1281–86.
Zhang X, et al. Muricauda Indica Sp. Nov., Isolated from Deep Sea Water. Int J Syst Evol Microbiol. 2018;68(3):881–85.

SupplementaryTable1.docx

Epidermal Microbiomes of Triakis Semifasciata Are Consistent Across Captive and Wild Environments

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Comparisons of functional gene potentials of T. semifasciata epidermal microbiomes across environments

Metagenome Assembled Genomes constructed from Microbial Communities Associated with T. semifasciata

Discussion

Epidermal Microbiome Taxonomic Structure as a Product of Captivity Duration and Host Factors

Metabolic Potentials of Captive Shark Microbiomes Reflect Environmental Conditions

MAGs reveal Novel, Constant Microbial Associations with T. semifasciata

Conclusion

Declarations

References

Supplementary Files

Status:

Version 1