Abstract
In brief
Reproductive microbiomes contribute to the successful embryogenesis of offspring but are poorly studied in non-mammalian species that exhibit pregnancy. This study characterises the male pregnant seahorse brood pouch microbiome and identifies potential microbial maternal contributions to the pouch, providing insights into the sources and adaptive value of the embryonic microbial environment.
Abstract
Seahorses demonstrate an unusual reproductive strategy, in which males incubate embryos inside a complex ‘brood pouch’ until parturition, analogous to mammalian viviparity. In many species, a ‘normal’ reproductive microbiome ensures successful embryogenesis and enables parents (usually mothers) to provide their offspring with their initial microbiome. In male-pregnant seahorses, embryos may receive microbiomes from both parents: from the paternal brood pouch and from the maternal eggs. Using the pot-bellied seahorse (Hippocampus abdominalis), we employed 16S rRNA sequencing to explore the reproductive microbiome. We aimed to compare the microbiome of the male pregnant pouch to the male non-pregnant pouch and external skin, and to identify bacterial taxa found exclusively in the pregnant pouch that could be derived maternally from the microbiome of eggs. Our findings demonstrate that the pregnant brood pouch microbiome is compositionally distinct from the non-pregnant pouch and external skin. The pouch microbiome also has characteristics of resistance to colonisation by pathogens, including a low species richness, high species evenness and diversity and very low abundance of Vibrio, a genus that includes fish skin pathogens. Thirteen bacterial taxa appear exclusively in the pregnant pouch, relative to the non-pregnant pouch, and seven of these overlapped with taxa present in or on the eggs. The possible supplementation of brood pouch microbiome with egg-associated micro-organisms hints at a maternal microbial contribution to male pregnancy. This characterisation of the pregnant seahorse pouch microbiome provides a platform for further research into its function and possible adaptive value during male pregnancy.
Introduction
Syngnathids (seahorses, pipefish and seadragons) demonstrate advanced parental care through brooding (parental incubation of eggs in or on the body, outside the female reproductive tract), a process analogous to mammalian gestation that is termed ‘male pregnancy’ (Whittington & Friesen 2020). Male seahorses have the most complex brooding anatomy in Syngnathidae (Whittington & Friesen 2020, Harada et al. 2022). Their highly specialised brood pouches incubate embryos until their release as free-swimming neonates (Carcupino et al. 2002). The seahorse pouch is a physiologically modified extension of the tail skin that opens at a single pore (Kawaguchi et al. 2017). Made up of a thick, vascularised dermis containing a placental layer, it facilitates physiological exchange between fathers and embryos during pregnancy (Whittington & Friesen 2020, Whittington et al. 2022). The pouch tissue changes morphologically and physiologically during pregnancy and facilitates respiratory gas exchange (Dudley et al. 2021), nutrient transfer (Skalkos et al. 2020, 2024) and putatively waste removal and immunological protection (Melamed et al. 2005, Roth et al. 2012, Whittington et al. 2015).
During mating, the seahorse brood pouch fills with non-sterile seawater, and the addition of nutrient-rich eggs, some of which break, likely produces an environment conducive to microbial proliferation (Whittington & Friesen 2020, Zhao et al. 2023). The male pouch microbiome may play a vital role in pregnancy and embryo development akin to gestational microbiomes in classical female viviparity. In viviparous animals, reproductive tract-associated microbiomes directly influence host reproductive success, as disruptions to the normal microbiome (dysbiosis) can harm offspring (Comizzoli et al. 2021, Rowe et al. 2021). Commensal micro-organisms can defend against pathogens by producing antimicrobial agents and via competitive exclusion (Tachedjian et al. 2017, Rowe et al. 2021). Reproductive microbiomes also facilitate vertical transmission, the inheritance of micro-organisms from mother to offspring (Funkhouser & Bordenstein 2013). Vertical transmission can immunologically benefit offspring by limiting pathogen invasion (Kerwin et al. 2019, Bunker et al. 2021), seeding commensal micro-organisms (Mika et al. 2021) and priming the embryonic immune system (Beemelmanns et al. 2019). In seahorses, the pouch microbiome may similarly protect embryos and enable vertical transmission.
In addition to acquisition of micro-organisms from direct contact with the male pouch microbiome, eggs may act as vessels for maternal vertical transmission to offspring. Thus, the seahorse gestational microbiome could be plausibly sourced both paternally and maternally. Research on egg-mediated microbial transfer primarily focuses on oviparous animals with limited parental care (Funkhouser & Bordenstein 2013, Nyholm 2020). In oviparous species, commensal micro-organisms in or on eggs can protect offspring against pathogenic fouling (Liu et al. 2014, Kerwin et al. 2019, Bunker et al. 2021). Beneficial micro-organisms are packaged inside eggs and/or inoculate their surface for maternal germline transmission (Cary & Giovannoni 1993, Hosokawa et al. 2013, Van Veelen et al. 2018, Bunker et al. 2021). Micro-organisms contained inside eggs are absorbed by developing embryos while egg surface micro-organisms are acquired after hatching (Nyholm 2020). A study in broad-nosed pipefish (Syngnathus typhle) demonstrated that micro-organisms inside eggs colonise the offspring gut (Tanger et al. 2024). Therefore, both the female egg and male pouch microbiome warrant investigation as potential sources of embryonic micro-organisms.
While the non-pregnant brood pouch microbiome has been characterised in lined seahorses (Hippocampus erectus) (Zhao et al. 2023), the pregnant seahorse pouch microbiome and its potential protective role for embryos remain unexplored. Previous studies in pipefish have examined the egg microbiome and its contributions to embryos (Beemelmanns et al. 2019, Tanger et al. 2024), but we lack such investigations in seahorses, which have more complex, closed pouches that permit fathers more control over the embryonic environment (Whittington & Friesen 2020).
We used the pot-bellied seahorse (Hippocampus abdominalis) to address two main questions regarding the sources of the seahorse gestational microbiome. First, does the male pregnant pouch harbour a unique microbiome compared to the non-pregnant pouch and external skin? Second, are there micro-organisms exclusive to the pregnant pouch that are not otherwise found in the non-pregnant pouch, and are these taxa on and/or in female eggs? For our first aim, we characterised the male pregnant pouch by assessing bacterial species richness, diversity, community structure and taxonomic composition, comparing it to the male non-pregnant pouch and external skin. For our second aim, we identified micro-organisms unique to the male pouch and assessed their overlap with the taxonomic profile of the female eggs to identify any potential egg-derived taxa.
Methods
Animal husbandry and breeding
Captive bred, reproductively mature H. abdominalis were obtained from Seahorse Australia (TAS, Australia) (University of Sydney Animal Ethics Committee; approval number: 2021/1995). Animals were housed in recirculating aquaria under standard conditions, including artificial seawater, at The University of Sydney (Australia), as previously described (Whittington et al. 2013). Tanks were cleaned twice weekly, during which water quality parameters were monitored. The water used in the aquaria was prepared in a controlled laboratory environment, which may influence the microbiome differently than natural environments. Animals were fed thawed frozen Mysis relicta shrimp (Ocean Nutrition, USA) six days each week and held under a summer light regime (15.5h light:8.5h darkness) to stimulate natural breeding behaviours. Males were tagged with coloured bead ‘necklaces’ for identification. To obtain pregnant males, we co-housed four non-pregnant males and five females in a deep 750 L breeding tank for three-day periods (Woods 2000). After mating, males were transferred to 170 L shallow-water tanks, separated from females. To determine which males were pregnant after three days in the breeding tank, we assessed their reproductive status using established behavioural testing methods (Whittington et al. 2013). We conducted behavioural trials between 08 30 and 10 30 h, when seahorses are most reproductively active, over three consecutive days (Masonjones & Lewis 1996, Whittington et al. 2013). We repeated this breeding and behavioural testing cycle until we obtained five putatively pregnant males. We housed putatively pregnant males separately from non-pregnant males until mid-pregnancy (14–17 days post-fertilisation), roughly corresponding to 70% of embryonic development (Sommer et al. 2012).
Sample collection
Animals were euthanised by gradual overdose of ethanol, followed by decapitation and pithing, in accordance with protocols approved by the University of Sydney Animal Ethics Committee (approval number: 2021/1995) and the American Veterinary Medical Association 2020 (Leary et al. 2020). Microbial samples were collected using sterile flocked swabs (Rongye Technology, China) and stored immediately at −20 °C until DNA extraction.
We collected two microbiome samples (repeated measures) from each male: one external skin surface swab (skin outside of the pouch) and one internal brood pouch swab (Fig. 1i, ii, iii, iv). We took the first sample by swabbing the skin surface of each male. We then opened brood pouches by making an incision with a sterile razor blade from the pouch opening to the base. We took the second sample by swabbing the brood pouch tissue surface (non-pregnant males) or the surface of the pouch and interspersed embryos (pregnant males 14–17 days post-fertilisation). In both cases, we passed the swab head 10 times over a ∼1 cm2 area. We collected 20 microbiome samples in total from four experimental groups: i) non-pregnant external skin samples and iii) non-pregnant internal pouch (n = 5 non-pregnant males), and ii) pregnant external skin samples and iv) pregnant internal pouch (n = 5 pregnant males) (Fig. 1i, ii, iii, iv).
Diagram of swab samples collected for microbiome analysis from male and female Hippocampus abdominalis adults. (i) Non-pregnant external skin, outside of pouch (male). (ii) Pregnant external skin, outside of pouch (male). (iii) Non-pregnant internal pouch (male). (iv) Pregnant internal pouch containing mid-stage embryos (14–17 days post-fertilisation) (male). (v) Ovary with intact eggs (female). (vi) Ovary with punctured eggs (female). Blue box indicates experimental groups used to address Aim 1 (comparison of male non-pregnant and pregnant external skin and internal pouch microbiomes). Yellow box indicates experimental groups used to address Aim 2 (to identify bacterial taxa exclusive to the male pregnant pouch and assess their presence on and/or in female eggs). Dashed line indicates swabbing area and dotted line indicates where incisions were made to dissect animals.
Citation: Reproduction 169, 4; 10.1530/REP-24-0159
We collected two microbiome samples (repeated measures) from each female: one egg surface swab and one egg surface plus egg contents swab (Fig. 1v and vi). We opened the abdominal cavity of each female and cut the membrane of one randomly selected ovary using sterile surgical scissors to expose the intact eggs. We took the first sample by swabbing the external surface of the eggs. We took the second sample by puncturing the egg membranes using sterile surgical scissors and swabbing the released internal contents. For all female samples, we passed the swab head 10 times over a ∼0.5 cm2 area (due to the size constraints of the ovary). We collected ten microbiome samples in total from two experimental groups: v) egg surface samples and vi) egg surface plus egg contents (n = 5 females) (Fig. 1v and vi).
DNA extraction, amplification and sequencing
We extracted DNA from swab samples using the DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany), with modifications to the lysis step to optimise DNA recovery from gram-positive bacteria. Specifically, the swab head was vortexed in 180 μL Buffer ATL for two minutes to dislodge bacterial cells. The swab was then removed with sterile forceps and 20 μL Proteinase K was added. The mixture was incubated at 56 °C for ∼18 h with light agitation (600 rpm). We used 30 μL UltraPure DNase/RNase-Free Water (Invitrogen, USA) for elution (DNA concentrations available in Supplementary Table 1: (see section on Supplementary materials given at the end of the article)). We included unused swabs as negative controls in each batch of extractions and quantified them to check for contamination. DNA from animal swabs were submitted to the Ramaciotti Centre for Genomics (University of New South Wales, Australia) for amplification using primers 341F (5′-CCTACGGGNGGCWGCAG-3′) and 805R (5′-GACTACHVGGGTATCTAATCC-3′) (Klindworth et al. 2012), targeting the V3-V4 regions of the bacterial 16S rRNA gene. All samples were uniquely barcoded, and amplicon libraries were sequenced on an Illumina MiSeq 2000 (Illumina, USA) using a 2 × 300 bp kit. A negative no-template control and positive ZymoBIOMICS Microbial Community DNA Standard (Zymo Research, USA) were included for quality control.
Data processing and statistical analysis
We processed sequences using the Mothur (https://mothur.org; v.1.43.0) pipeline, implemented in Galaxy, according to the MiSeq SOP recommendations for 16S rRNA paired-end reads (Schloss et al. 2009). We merged paired-end reads into contigs and aligned unique reads against the SILVA 16S database release 138.1 (Quast et al. 2013), removing all sequences that did not align to the V3-V4 16S region (SILVA alignment position 6,388–25,316). An average of 4.2% of sequences across all samples (min. = 0.8%, max. = 12.5%) were removed (Supplementary Table 2). We included a pre-clustering step to group sequences, allowing up to 4 bp differences between sequences in each group, for the 465 bp long V3-V4 region (Schloss et al. 2011). Chimeric sequences were identified and cleaned from the data using the UCHIME algorithm in Mothur (Edgar et al. 2011). We clustered representative sequences into operational taxonomic units (OTUs) at genetic distances of 0.03, based on the naïve Bayesian classifier (Wang et al. 2007). Taxonomy was assigned based on the Ribosomal Database Project reference taxonomy (Cole et al. 2014). To control for variation in sequencing coverage, we rarefied the data by subsampling reads to the lowest sequencing depth in our sample set (14,138 reads for male samples and 16,212 reads for female samples). To assess sequencing adequacy, rarefaction curves were generated (Supplementary Fig. 1) and Good’s coverage scores were calculated (estimates ranged from 98.1 to 99.2%).
To address our first aim, we calculated alpha diversity of male external skin and internal pouch samples in Mothur using the command summary.single with two community richness calculators, observed number of OTUs and the Chao1 index (Chao 1984), and one community richness and evenness metric, the Shannon index (Shannon 1948). In a community, richness refers to the number of species, evenness to their relative abundances and diversity to both richness and evenness. Observed number of OTUs is a simple count of the number of OTUs in each sample, while the Chao1 index estimates species richness by counting the number of species and emphasising rare taxa that only appear once or twice (Chao 1984). The Shannon index is a measure of species diversity that takes into account both evenness and species richness (Shannon 1948). To test the effect of pregnancy and sampling location on each alpha diversity metric, we fitted a linear mixed-effect model using the lmer function in the package lme4 (Bates et al. 2015) using a Gaussian error function. We tested this model using the lmerTest package (Kuznetsova et al. 2017). Fixed factors included ‘sampling location’ (with two levels: skin (=0) and pouch (=1)) and ‘reproductive status’ (with two levels: non-pregnant (=0) and pregnant (=1)), with a random factor of ‘animal’ to account for repeated measures, as one skin and one pouch sample were taken from the same individual. Although the sample size for each group (n = 5) did not provide sufficient statistical power to formally include an interaction term in our main effects analysis, we conducted pairwise comparisons to test the differences between our four experimental groups using the estimated marginal means with Tukey’s HSD adjustment in the emmeans package (https://cran.r-project.org/web/packages/emmeans/index.html). To analyse and visualise beta diversity, we generated a Bray–Curtis dissimilarity matrix using dist.shared (Bray & Curtis 1957), to estimate differences in diversity and community structure between different male microbiomes. We tested main factor effects using PERMANOVA (999 permutations), using the adonis function in the vegan package (https://cran.r-project.org/web/packages/vegan/index.html). We tested differences between experimental groups using the pairwise.adonis2 function in the pairwiseAdonis package (https://github.com/pmartinezarbizu/pairwiseAdonis). Taxonomic composition of male microbiomes was explored by joining the taxonomic profile (tax.summary output) and summary of per-sample OTU occurrence (mothur.shared output) data frames to calculate relative abundances of taxa in each experimental group.
To address our second aim, we compared taxonomic profiles between non-pregnant and pregnant pouch microbiomes. We identified taxa exclusive to the pregnant pouch by searching for taxa that were absent in all non-pregnant pouch samples but present in most (at least 60%, 3 out of 5) of the pregnant pouch samples. A 60% cutoff allowed us to include more taxa in the core microbiome, given our small sample size. We then determined whether these taxa were present in any female egg samples (egg surface and egg surface plus egg contents). For each taxon present, we recorded the number of egg surface samples and egg surface plus egg contents samples it occurred in. To explore taxonomic composition of egg microbiomes, we calculated the relative abundance of bacterial taxa within each experimental group.
Results
Aim 1: male non-pregnant and pregnant external skin and internal pouch microbiomes
Alpha and beta diversity
The mean observed OTUs of external skin microbiomes was significantly higher than that of internal pouch microbiomes, while the mean observed OTUs of non-pregnant microbiomes was no different from that in pregnant microbiomes (Table 1, Fig. 2A). The pregnant internal pouch microbiome had the lowest number of observed OTUs out of all four groups (Fig. 2A). The mean Chao1 index of non-pregnant microbiomes was significantly higher than that of pregnant microbiomes, while the mean Chao1 index of external skin microbiomes was no different to that of internal pouch microbiomes (Table 1, Fig. 2B). The pregnant internal pouch microbiome had the lowest Chao1 index out of all four groups (Fig. 2B). As compared to the non-pregnant internal pouch, the pregnant internal pouch microbiome had a significantly lower Chao1 index (Table 1). The mean Shannon’s index of external skin microbiomes was no different to that of internal pouch microbiomes, and the mean Shannon’s index of non-pregnant microbiomes was also no different to that of pregnant microbiomes (Table 1, Fig. 2C). The pregnant internal pouch microbiome had a similar Shannon’s index compared to the other three groups (Fig. 2C).
Results from linear mixed-effect models on effects of ‘reproductive status’ (non-pregnant vs pregnant) and ‘sampling location’ (skin vs pouch) on three measures of alpha diversity in Hippocampus abdominalis males. (A) Random term ‘animal’ (individual) was included to account for repeated measures (i.e., skin and pouch specimens taken from the same individual, see the section on Methods). Also shown are pairwise comparisons between microbiomes of the four male experimental groups: non-pregnant external skin (n = 5), non-pregnant internal pouch (n = 5), pregnant external skin (n = 5) and pregnant internal pouch (n = 5).
| Factor | Estimate | SE | df | t value | P value (>|t|) |
|---|---|---|---|---|---|
| Observed OTUs | |||||
| Main effects | |||||
| Intercept | 2,798.4 | 210.8 | 17.0 | 13.28 | 2.11 × 10−10 *** |
| Sampling location (pouch) † | −1,092.9 | 243.4 | 17.0 | −4.49 | 2.32 × 10−4 *** |
| Reproductive status (pregnant) ‡ | −109.5 | 243.4 | 17.0 | −0.45 | 0.658 |
| Pairwise experimental group comparisons | |||||
| Non-pregnant skin – non-pregnant pouch | 717 | 326 | 8.0 | 2.20 | 0.204 |
| Non-pregnant skin – pregnant skin | −267 | 329 | 16 | −0.81 | 0.849 |
| Non-pregnant skin – pregnant pouch | 1,202 | 329 | 16 | 3.66 | 0.010* |
| Non-pregnant pouch – pregnant skin | −983 | 329 | 16 | −2.99 | 0.039* |
| Non-pregnant pouch – pregnant pouch | 486 | 329 | 16 | 1.48 | 0.474 |
| Pregnant skin – pregnant pouch | 1,469 | 326 | 8.0 | 4.50 | 8.70 × 10−3 ** |
| Chao1 index | |||||
| Main effects | |||||
| Intercept | 15,846.0 | 1,658.3 | 14.2 | 9.56 | 1.41 × 10−7 *** |
| Sampling location (pouch) | −879.4 | 1845.8 | 9.0 | −0.48 | 0.645 |
| Reproductive status (pregnant) | −6,279.2 | 1948.4 | 8.0 | −3.22 | 1.22 × 10−2 * |
| Pairwise experimental group comparisons | |||||
| Non-pregnant skin – non-pregnant pouch | −2,513 | 2,188 | 8.0 | −1.15 | 0.672 |
| Non-pregnant skin – pregnant skin | 2,887 | 2,488 | 15.2 | 1.16 | 0.660 |
| Non-pregnant skin – pregnant pouch | 7,159 | 2,488 | 15.2 | 2.88 | 0.050 |
| Non-pregnant pouch – pregnant skin | 5,400 | 2,488 | 15.2 | 2.17 | 0.176 |
| Non-pregnant pouch – pregnant pouch | 9,672 | 2,488 | 15.2 | 3.89 | 6.90 × 10−3 ** |
| Pregnant skin – pregnant pouch | 4,272 | 2,188 | 8.0 | 1.95 | 0.281 |
| Shannon index | |||||
| Main effects | |||||
| Intercept | 2.805 | 0.282 | 12.4 | 9.94 | 2.90 × 10−7 *** |
| Sampling location (pouch) | −0.156 | 0.268 | 9.0 | −0.58 | 0.576 |
| Reproductive status (pregnant) | 0.752 | 0.351 | 8.0 | 2.14 | 6.47 × 10−2 |
| Pairwise experimental group comparisons | |||||
| Non-pregnant skin – non-pregnant pouch | 0.3336 | 0.392 | 8.0 | 0.852 | 0.829 |
| Non-pregnant skin – pregnant skin | −0.5738 | 0.447 | 15.2 | −1.283 | 0.587 |
| Non-pregnant skin – pregnant pouch | −0.5962 | 0.447 | 15.2 | −1.333 | 0.557 |
| Non-pregnant pouch – pregnant skin | −0.9074 | 0.447 | 15.2 | −2.029 | 0.221 |
| Non-pregnant pouch – pregnant pouch | −0.9298 | 0.447 | 15.2 | −2.079 | 0.204 |
| Pregnant skin – pregnant pouch | −0.0225 | 0.392 | 8.0 | −0.057 | 1.000 |
P (>|t|) < 0.001.
P (>|t|) < 0.01.
P (>|t|) < 0.05.
Reference category: external skin.
Reference category: non-pregnant.
DF, degrees of freedom.
Mean richness measured by (A) observed OTUs, (B) Chao1 index, and mean diversity measured by (C) Shannon’s index, of four experimental group microbiomes in Hippocampus abdominalis males: non-pregnant external skin (n = 5), non-pregnant internal pouch (n = 5), pregnant external skin (n = 5) and pregnant internal pouch (n = 5). Error bars represent ±standard deviation about the mean. Lettering denotes significant differences.
Citation: Reproduction 169, 4; 10.1530/REP-24-0159
The three alpha diversity statistics revealed differing patterns; while only sampling location significantly affected the mean observed OTUs, only reproductive status significantly affected the Chao1 index of microbiomes. The Shannon’s index was not significantly affected by either of the two tested factors. While the number of observed OTUs was relatively low in the non-pregnant internal pouch, it did have a high Chao1 index, indicating that it harboured a high number of OTUs that appeared only once or twice. In the pregnant pouch, despite its low richness with both the observed OTUs and Chao1 index metrics, it had a similar Shannon’s diversity index to that of the other groups. As Shannon’s diversity accounts for both richness of species and evenness between species, this result indicates that the pregnant pouch microbiome had a highly even community relative to the other groups in the study.
Beta diversity describes the differences in diversity and community structure of microbiomes between samples. Microbiome community structure was significantly different between external skin and internal pouch and also significantly different between non-pregnancy and pregnancy (Table 2, Fig. 3). The pregnant internal pouch microbiome is significantly different from the non-pregnant internal pouch and both the pregnant and non-pregnant external skin (Table 2, Fig. 3).
Results from PERMANOVA and pairwise comparisons on Bray–Curtis distance matrix, with 999 permutations, representing the beta diversity between male Hippocampus abdominalis microbiomes: non-pregnant external skin (n = 5), non-pregnant internal pouch (n = 5), pregnant external skin (n = 5) and pregnant internal pouch (n = 5). The two factors ‘sampling location’ and ‘reproduction status’ were included in the model, and the strata term ‘animal’ (individual) was used to account for repeated measures taken from the same individual.
| Term | df | Sum of squares | R 2 value | F value | P (>F) |
|---|---|---|---|---|---|
| PERMANOVA | |||||
| Sampling location | 1 | 0.591 | 0.138 | 3.133 | 2.00 × 10−3 ** |
| Reproductive status | 1 | 0.475 | 0.111 | 2.516 | 2.00 × 10−3 ** |
| Residual | 17 | 3.207 | 0.751 | ||
| Pairwise comparisons | |||||
| Non-pregnant skin – non-pregnant pouch | 1 | 0.237 | 0.146 | 1.366 | 0.184 |
| Residual | 8 | 1.388 | 0.854 | ||
| Non-pregnant skin – pregnant skin | 1 | 0.244 | 0.208 | 2.107 | 0.069 |
| Residual | 8 | 0.926 | 0.792 | ||
| Non-pregnant skin – pregnant pouch | 1 | 0.736 | 0.329 | 3.926 | 0.014* |
| Residual | 8 | 1.500 | 0.671 | ||
| Non-pregnant pouch – pregnant skin | 1 | 0.330 | 0.181 | 1.769 | 0.043* |
| Residual | 8 | 1.492 | 0.819 | ||
| Non-pregnant pouch – pregnant pouch | 1 | 0.446 | 0.178 | 1.729 | 0.037* |
| Residual | 8 | 2.065 | 0.822 | ||
| Pregnant skin – pregnant pouch | 1 | 0.570 | 0.262 | 2.842 | 5.00 × 10−3 ** |
| Residual | 8 | 1.603 | 0.738 |
P (>F) < 0.01.
P (>F) < 0.05.
DF, degrees of freedom.
Samples within both the non-pregnant and pregnant external skin microbiomes were clustered tightly together (Fig. 3), indicating that external skin microbiome community structure does not change greatly with pregnancy. While the non-pregnant pouch shared some similarity to the external skin microbiomes, the pregnant pouch microbiomes were distinct from the external skin microbiomes (Fig. 3).
Principal coordinates analysis (PCoA) plot based on the Bray–Curtis distance matrix, depicting beta diversity across the four male microbiomes in Hippocampus abdominalis: non-pregnant external skin (n = 5), non-pregnant internal pouch (n = 5), pregnant external skin (n = 5) and pregnant internal pouch (n = 5). Ellipses illustrate 70% spread around the centroids within each experimental group.
Citation: Reproduction 169, 4; 10.1530/REP-24-0159
Taxonomic composition of the male skin and brood pouch microbiomes
Male skin and brood pouch microbiomes were dominated by the phyla Bacteroidetes (66.0%) and Proteobacteria (50.7%). The most abundant classes observed were Flavobacteriia (53.1%), Gammaproteobacteria (28.8%), Alphaproteobacteria (24.5%) and Saprospiria (11.9%) (Fig. 4A).
Taxonomic class composition of (A) male Hippocampus abdominalis microbiomes in four experimental groups (first bar): non-pregnant external skin (n = 5), non-pregnant internal pouch (n = 5), pregnant external skin (n = 5) and pregnant internal pouch (n = 5), and at individual sample level (second bar). (B) Female H. abdominalis microbiomes in two experimental groups (first bar): egg surface (n = 5) and egg surface plus egg contents (n = 5), and at individual sample level (second bar). Relative abundance of bacterial classes with over 1% abundance is shown.
Citation: Reproduction 169, 4; 10.1530/REP-24-0159
While we did not conduct statistical tests on taxonomic group abundance, the class Gammaproteobacteria appeared substantially more abundant in internal pouch microbiomes than on the external skin (Fig. 4A). At a finer taxonomic resolution, Gammaproteobacteria in the two internal pouch groups consisted of different genera. Within class Gammaproteobacteria, the non-pregnant internal pouch had a much higher relative abundance of the genera Vibrio (16.9%) and Oceaniserpentilla (4.35%) than in the three other male microbiomes (where abundances of both were <1.00%). In the pregnant internal pouch, the genera Acinetobacter (9.46%) and Pseudomonas (8.06%) occurred in high relative abundance as compared to the other male microbiomes (where abundances of both were <1.00%). The class Saprospiria was less abundant in the internal pouch microbiomes than on the external skin (Fig. 4A), with this difference mostly attributed to a lower relative abundance of an unclassified taxon in the order Saprospirales in the non-pregnant internal pouch (3.16%) and pregnant internal pouch (3.68%) as compared to the non-pregnant external skin (8.63%) and pregnant external skin (7.85%). The relative abundance of family Flavobacteriaceae, in class Flavobacteriia, was lowest in the pregnant internal pouch (38.3%) as compared to the non-pregnant internal pouch (49.6%), non-pregnant external skin (52.9%) and pregnant external skin (51.6%).
Aim 2: bacterial taxa exclusive to the male pregnant pouch and overlap with female eggs
We identified 13 taxa that were not present in the non-pregnant pouch microbiome but were present in most (60%) of the pregnant pouch microbiome samples (Tables 3 and 4). Seven of the 13 taxa exclusive to the pregnant pouch, compared to the non-pregnant pouch, were also present in egg samples (Table 5). The pregnant pouch taxa Cellulophaga, Kangiella, Maritalea, Reyranella, Rhizobiaceae (unclassified genus) and Oligoflexales (unclassified family and genus) were not present in any egg samples (Table 5). There were no taxa exclusive to just egg surface plus egg contents samples.
Taxonomic classification of 13 bacterial taxa exclusive to the male Hippocampus abdominalis pregnant brood pouch microbiome in comparison to the non-pregnant pouch. The lowest level of taxonomic classification of each taxon is shown in bold.
| Phylum | Class | Order | Family | Genus |
|---|---|---|---|---|
| Bacteroidetes | Chitinophagia | Chitinophagales | Chitinophagaceae | Sediminibacterium * |
| Flavobacteriia | Flavobacteriales | Flavobacteriaceae | Cellulophaga | |
| Saprospiria | Saprospirales | Lewinellaceae | Flavilitoribacter * | |
| Planctomycetes | Planctomycetacia | Pirellulales | Lacipirellulaceae | Bythopirellula * |
| Proteobacteria | Alphaproteobacteria | Parvularculales | Parvularculaceae | Marinicaulis * |
| Rhizobiales | Devosiaceae | Maritalea | ||
| Rhizobiaceae | (Unclassified) | |||
| Rhodobacterales | Rhodobacteraceae | Roseivivax * | ||
| Rhodospirillales | Reyranellaceae | Reyranella | ||
| Betaproteobacteria | Burkholderiales * | (Unclassified) | (Unclassified) | |
| Deltaproteobacteria | Desulfobacterales | Desulfobacteraceae * | (Unclassified) | |
| Gammaproteobacteria | Oceanospirillales | Kangiellaceae | Kangiella | |
| Oligoflexia | Oligoflexales | (Unclassified) | (Unclassified) |
Taxa that were present in female Hippocampus abdominalis egg microbiomes.
Description of microbial characteristics, habitat and ecology of bacteria exclusive to the male Hippocampus abdominalis pregnant brood pouch microbiome (n = 5) and not present in non-pregnant brood pouch samples (n = 5). The lowest taxonomic classification is used for identification of each bacterial taxon.
| Taxon | Description of microbial characteristics | Habitat | Ecology |
|---|---|---|---|
| Sediminibacterium | Gram-negative rods. Strictly aerobic to facultatively anaerobic (Kim et al. 2013b, 2016). Bacteria are motile by gliding | Sediments and soil in or around aquatic environments, including from fishbowls (Sediminibacterium aquarii) (Kim et al. 2013b, 2016, Sethuraman et al. 2022). | Putative commensalism with algae (Sethuraman et al. 2022) |
| Cellulophaga | Gram-negative, gliding, aerobic rods forming iridescent colonies (Chapelais-Baron et al. 2017). Antifouling and algicidal (Skerratt et al. 2002) | Associated with marine algae and isolated from sediments, marine animals and seaweeds (Lafleur et al. 2015, Chapelais-Baron et al. 2017) | Inhibits Pseudomonas aeruginosa growth (Lafleur et al. 2015) |
| Flavilitoribacter | Rod-shaped (García-López et al. 2019) | F. nigricans from beach sand (García-López et al. 2019) | Unknown |
| Bythopirellula | Gram-negative, oval to pear-shaped bacteria that bud (Storesund & Øvreås 2013, Wiegand et al. 2020) | Bythopirellula goksoyri is from deep sea iron hydroxide deposits (Storesund & Øvreås 2013) and Bythopirellula polymerisocia from a riverbank (Wiegand et al. 2020) | Unknown |
| Marinicaulis | Gram-negative, aerobic or facultatively anaerobic motile rods (Yu et al. 2018, Wang et al. 2019) | M. aureus and M. flavus both isolated from seawater (Yu et al. 2018, Wang et al. 2019) | Unknown |
| Maritalea | Gram-negative, motile, strictly aerobic rods with flagella (Hwang et al. 2009, Fukui et al. 2012) | Found from coastal seawater and in marine plankton and algae cultures (Hwang et al. 2009, Fukui et al. 2012) | Unknown |
| Roseivivax | Gram-negative, aerobic, motile rods with subpolar flagella (Suzuki et al. 1999, Chen et al. 2012) | Found on charophyte algae, epiphytes (Suzuki et al. 1999) and coral (Chen et al. 2012) | Unknown |
| Reyranella | Gram-negative, non-motile rods with aerobic growth (Lee et al. 2017) | Both aquatic sources (rivers and cooling towers) (Pagnier et al. 2011) and terrestrial sources (forest and agricultural soil) (Kim et al. 2013a, Lee et al. 2017) | Reyranella massiliensis is an intra-cellular bacteria of amoeba (Pagnier et al. 2011) |
| Kangiella | Gram-negative, non-motile rods (Yoon et al. 2004, Lee et al. 2013), with enriched protein degradation ability (Wang et al. 2018) | From marine environments including marine organisms, coastal seawater, tidal flat sediments and deep-sea sediments (Lee et al. 2013, Wang et al. 2018) | Unknown |
| Desulfobacteraceae (family) | Gram-negative, anaerobic and morphologically varied (Galushko & Kuever 2020). Members metabolise sulphate to sulfide (Kuever 2014) | Found in various aquatic habitats including in freshwater, saline or hypersaline waters, in sediment or on aquatic organisms (Ahn et al. 2009, Kuever 2014) | Unknown |
| Rhizobiaceae (family) | A family of gram-negative rods, predominantly aerobic and highly heterogeneous (Carareto Alves et al. 2014) | Highly diverse, but mostly associated with soil and plants (Carareto Alves et al. 2014) | Facilitate nitrogen fixation in plants (Carareto Alves et al. 2014) |
| Burkholderiales (order) | Highly varied gram-negative bacteria. Strictly aerobic to facultatively anaerobic (Garrity et al. 2005) | Highly diverse habitats (Garrity et al. 2005) | Pathogenic and non-pathogenic (Garrity et al. 2005) |
| Oligoflexales (order) | Gram-negative, pleomorphic, obligately aerobic bacteria, with filamentous stages (Hahn et al. 2017) | Oligoflexus tunisiensis from desert sand (Nakai et al. 2014) and Pseudobacteriovorax antillogorgiicola from coastal coral (McCauley et al. 2015) | Unknown |
Presence of bacterial taxa exclusive to the male pregnant brood pouch compared to the non-pregnant pouch in female egg surface (ES; n = 5) and egg surface plus egg contents samples (ES+C; n = 5) in Hippocampus abdominalis.
| Bacterial taxa | Presence in egg samples | |
|---|---|---|
| ES | ES+C | |
| Sediminibacterium | 3/5 | 5/5 |
| Bythopirellula | 3/5 | 2/5 |
| Marinicaulis | 3/5 | 2/5 |
| Burkholderiales (unclassified family, genus) | 3/5 | 0/5 |
| Flavilitoribacter | 2/5 | 1/5 |
| Desulfobacteraceae (unclassified genus) | 1/5 | 1/5 |
| Roseivivax | 1/5 | 0/5 |
| Cellulophaga | 0/5 | 0/5 |
| Kangiella | 0/5 | 0/5 |
| Maritalea | 0/5 | 0/5 |
| Oligoflexales (unclassified family, genus) | 0/5 | 0/5 |
| Reyranella | 0/5 | 0/5 |
| Rhizobiaceae (unclassified genus) | 0/5 | 0/5 |
Taxonomic composition of female egg microbiomes
Egg microbiomes were largely dominated by the phylum Proteobacteria (91.2%), followed by Bacteroidetes (6.99%) and, at class level, were composed mostly of Gammaproteobacteria (73.4%), Alphaproteobacteria (17.2%) and Flavobacteriia (5.26%) (Fig. 4B).
Egg surface and egg surface plus egg contents microbiomes had very similar taxonomic compositions (Fig. 4B). Together, the egg microbiomes were dominated by the genus Vibrio (52.5%), followed by Ruegaria (9.96%), Acinetobacter (7.19%), an unclassified taxon in the class Gammaproteobacteria (5.96%) and Pseudomonas (4.12%). Like the male pregnant internal pouch, the female egg microbiomes harboured a relatively high abundance of the genera Acinetobacter and Pseudomonas compared to all other male microbiomes. Vibrio occurred in high relative abundance in both the non-pregnant internal pouch and egg microbiomes, although these taxonomic abundance differences were not explicitly tested.
Discussion
Our study compared the male pregnant seahorse pouch microbiome with the non-pregnant pouch and external skin and identified bacterial taxa exclusive to the pregnant pouch, potentially derived from eggs. The pregnant pouch microbiome was compositionally unique, with a lower species richness but higher evenness as compared to the other male microbiomes. The abundance of Vibrio was high in the non-pregnant pouch but very low in the pregnant pouch. We also identified 13 bacterial taxa unique to the pregnant pouch and inferred that seven of these taxa were potentially derived from eggs, suggesting a maternal microbial contribution to the embryonic environment.
We determined that the male pregnant brood pouch harbours an internal microbiome distinct from the external skin and non-pregnant pouch. The brood pouch encloses after mating (Berglund & Rosenqvist 1990), which we postulate creates a barrier against external bacterial immigration, fostering an isolated microbial habitat. The microbial community shift from non-pregnancy to pregnancy may result from physiological changes. For instance, endocrine changes can mediate reproductive microbiome structure (Comizzoli et al. 2021). Glucocorticoids modulate the reproductive microbiome in other animals, including rhinoceros and humans (Amabebe & Anumba 2018, Antwis et al. 2019). Glucocorticoids also increase during seahorse pregnancy to stimulate pouch growth (Scobell & Mackenzie 2011). Sex hormones, including oestradiol and progesterone, also modify reproductive tissues by enhancing their epithelial receptivity to beneficial bacteria (Stumpf et al. 2013). The same hormones stimulate epithelium proliferation of pregnant seahorse pouch tissue (Oconer et al. 2003), which may contribute to the pouch microbiome shift with pregnancy we report here. Nutritional changes in the pouch likely also influence the microbiome. Pregnant syngnathid pouch fluid contains lipid-rich yolk from fragmented eggs (Linton & Soloff 1964, Ripley & Foran 2009). This increased nutrient density may enhance the dominance of saprophytic bacteria that thrive by metabolising organic compounds including lipids and amino acids. Acinetobacter and Pseudomonas are such saprophytes (Cousin 1999, Kämpfer 2014), explaining why these taxa were highly abundant in nutrient-rich eggs and inside the pregnant pouch in our study. The salinity inside the pouch changes to match external marine conditions in late pregnancy (Oconer et al. 2006), which may also alter the resident microbiome. Our use of artificial seawater might have cultivated a slightly different pregnant pouch microbiome compared to wild seahorses experiencing natural environmental salinity fluctuations. There are multiple plausible physiological mechanisms that may contribute to the observed shift in brood pouch microbiome and further investigations into these relationships will help to clarify how the seahorse gestational microbiome is formed. We note that our swabbing approach samples the overall gestational environment but does not allow differentiation of the pregnant pouch versus embryonic microbiomes. Future studies aiming to isolate taxa tightly associated with the embryos or pouch could use surface sterilised whole embryos or dissected pouch tissue directly.
We observed that the male pregnant pouch microbiome exhibits low species richness, high species evenness and low Vibrio abundance, patterns which suggest that the pouch environment is pathogen limited. First, the pregnant pouch has a low richness, especially of rare taxa, compared to the non-pregnant pouch. A low microbial richness in the gestational environment is also evident in marsupial pouches, such as those of the tammar wallaby (Macropus eugenii) and the southern hairy-nosed wombat (Lasiorhinus latifrons) to protect developing young (Old & Deane 1998, Weiss et al. 2021). Like seahorse embryos, marsupial young spend a large proportion of their development inside a pouch. In marsupials, antimicrobial secretions help form this protective environment (Weiss et al. 2021). Similarly, antimicrobial peptides are also secreted into the seahorse pouch (Melamed et al. 2005), and pouch-expressed genes encoding other immune factors, such as cytokines and pathogen recognition receptors, are also upregulated during pregnancy (Wu et al. 2021, Jiang et al. 2022). These immune factors may reduce pathogenic challenges inside the seahorse pouch and reduce embryo mortality, a hypothesis congruent with our finding of a low occurrence of rare species in the pregnant pouch microbiome. Our study used captive laboratory-reared seahorses fed frozen food. While captive and wild fish may have a similar core microbiome (Roeselers et al. 2011), captivity may limit their exposure to rare micro-organisms compared to wild counterparts inhabiting diverse marine microbiomes and consuming varied diets (Clavere-Graciette et al. 2022, Ortega-Kindica et al. 2024). Future studies on wild seahorses are warranted to confirm whether these patterns persist in natural marine environments.
We also found that the pregnant pouch microbiome maintained high species evenness (and therefore diversity), which may confer pathogen colonisation resistance. Laboratory-controlled studies of bacterial communities have shown that increased evenness enhances resistance to invasion by alien species (De Roy et al. 2013). From a marine perspective, in oysters, surface microbiomes of higher evenness exhibit resistance against the pathogenic ostreid herpesvirus (Clerissi et al. 2020). In the broad-nosed pipefish (Syngnathus typhle), a highly diverse internal pouch microbiome is thought to protect embryos (Tanger et al. 2024). The high level of evenness observed in the pregnant seahorse pouch microbiome in our study may similarly suggest resilience against pathogenic invasion. Finally, while Vibrio abundance was high in the non-pregnant pouch (and eggs), it was very low in the pregnant pouch (and external skin). Vibrio comprises a large number of opportunistic marine pathogens, including V. harveyi, V. alginolyticus and V. splendidus, which are linked to diseases of the external tissues, such as the skin and gills, in seahorses (Alcaide et al. 2001, Balcázar et al. 2010, Binh et al. 2016). In our study, which used healthy (captive) seahorses, Vibrio abundance was indeed low on the external skin. It is notable that Vibrio is limited in the pregnant pouch, despite its high abundance in the non-pregnant pouch and further enrichment with Vibrio from the eggs, indicating that the pregnant pouch environment may actively reduce Vibrio abundance. Alternatively, microbe–microbe interactions could limit Vibrio proliferation, although this hypothesis requires further investigation into the pouch microbiome’s community dynamics. The low taxonomic resolution of Vibrio in our study is a constraint and future research should examine species and strain level patterns to enable more concrete biological inferences. Still, our findings highlight an interesting starting point to test whether the pregnant pouch microbiome reduces the microbial burden for offspring. Functional studies are needed to resolve the adaptive significance of the pregnant pouch microbiome. Testing wild seahorses could validate whether low Vibrio abundance occurs in the pouch under natural conditions. In addition, examining immune gene expression in pregnant pouch tissue in both captive and wild seahorses could clarify whether low Vibrio abundance correlates with heightened local immune activity.
While Vibrio abundance was low in the male pregnant pouch, this genus dominated female egg microbiomes. Although some Vibrio species are associated with seahorse skin disease, which may explain the genus’s low abundance in the pregnant pouch, several Vibrio species are commensal and abundant in the seahorse gut microbiome (Balcazar et al. 2010, Tanu et al. 2012, Wang et al. 2020). It is possible that gut commensal Vibrio species are contained inside the egg coat, enabling ingestion and early gut colonisation in offspring, while the pregnant pouch environment surrounding the eggs suppresses the abundance of Vibrio skin pathogens, to prevent embryo skin infections post-hatching. A study of the broad-nosed pipefish (Syngnathus typhle) tracked the source of micro-organisms that colonise offspring, and posited that the paternal brooding environment mostly influences the offspring external microbiome, while maternal transmission via eggs is responsible for internal colonisation of the offspring gut (Tanger et al. 2024). Our results are consistent with this concept, suggesting a similar mechanism in seahorses. To test this hypothesis, the embryo microbiome before and after hatching from the egg coat, which occurs within the pouch (Sommer et al. 2012), should be compared to reveal whether gut commensals are abundant before hatching, and whether skin commensals are enriched and/or skin pathogens are reduced after hatching. Comparison of surface sterilised embryos versus unsterilised embryos would also enable differentiation between the internal (gut) and external (skin) embryonic microbiomes (Tanger et al. 2024).
We identified the genera Cellulophaga and Sediminibacterium in the pregnant pouch (but not the non-pregnant pouch), the latter taxon being potentially maternally derived. These taxa are interesting due to their antimicrobial characteristics. Cellulophaga inhibits pathogenic Pseudomonas spp., including Pseudomonas aeruginosa, by disrupting virulent colony formation (Lafleur et al. 2015, Chapelais-Baron et al. 2017). While Pseudomonas aeruginosa pathogenicity is more widely acknowledged in freshwater fish, it is present in coastal seawater, where it can infect marine species (Kimata et al. 2004, El-Dakroury et al. 2020). The presence of Cellulophaga spp. in the pregnant pouch may protect embryos from such infections. While we recognised Pseudomonas to be relatively abundant in the pregnant pouch, potentially due to its ability to proliferate in nutrient-rich environments, the species in this genus are diverse, and the lack of taxonomic resolution here prevents detailed inferences about its interaction with Cellulophaga. Sediminibacterium, which occurred in a high proportion of egg samples, belongs to the family Chitinophagaceae, which is known for its ability to inhibit fungal growth by degrading chitin, a fungal cell wall component (Rosenberg 2014), and secreting antibiotic compounds against gram-positive pathogens (Beckmann et al. 2017). Chitinophagaceae presence is protective in some fish: increases in Chitinophagaceae abundance in sea bass (Lateolabrax maculatus) gut and three-spined stickleback (Gasterosteus aculeatus) water are associated with host disease resistance (Deng et al. 2021, Fuess et al. 2021). In our study, Sediminibacterium may share these microbicidal features, thereby protecting embryos. That Sediminibacterium may be transmitted to the pouch via the eggs demonstrates a potential maternal contribution to embryo protection.
To the best of our knowledge, this is the first study to explore the microbiome inside the pregnant male seahorse brood pouch. We found that the pregnant pouch harbours a distinct microbiome, compositionally different from the male non-pregnant pouch and external skin. The microbiome possibly attenuates the pathogenicity of the gestational environment, as the abundance of the genus Vibrio is greatly reduced. The low richness of rare bacteria and high diversity of the pregnant pouch microbiome suggest resistance to pathogen colonisation, possibly providing immunological protection for offspring. The eggs may also supplement the pouch with beneficial bacteria, such as Sediminibacterium, and prime offspring with gut commensals. Our characterisation of the seahorse brood pouch microbiome provides a valuable foundation for further investigation of the function of the gestational microbiome in male pregnancy.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/REP-24-0159.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the work reported.
Funding
This work was supported by a University of Sydney Research Accelerator (SOAR) Prize and Australia and Pacific Science Foundation Funding to CMW.
Author contribution statement
JW, CMW and CEG conceived and designed the study. JW performed the experiments and analysed data with guidance from CEG, CMW and ZS. JW, CMW and CEG wrote the manuscript, with editing from ZS. CMW obtained funding for this study. All authors provided feedback throughout the study to produce the data, analyses and final manuscript.
Data availability
All raw 16S amplicon reads are available in the NCBI SRA database under accession number PRJNA1227541.
Acknowledgements
We thank the members of the Applied and Evolutionary Zoology Group, including K Kambersky for assistance with animal husbandry, J Herbert for assistance with sample collection and C Songsomboon and M Hodgson for their helpful technical advice. We also thank C Friesen for the use of his reagents in DNA extraction.
References
Ahn YB , Kerkhof LJ & Häggblom MM 2009 Desulfoluna spongiiphila sp. nov., a dehalogenating bacterium in the Desulfobacteraceae from the marine sponge Aplysina aerophoba. Int J Syst Evol Microbiol 59 2133–2139. (https://doi.org/10.1099/ijs.0.005884-0)
Alcaide E , Gil-Sanz C , Sanjuán E , et al. 2001 Vibrio harveyi causes disease in seahorse, Hippocampus sp. J Fish Dis 24 311–313. (https://doi.org/10.1046/j.1365-2761.2001.00297.x)
Amabebe E & Anumba DOC 2018 Psychosocial stress, cortisol levels, and maintenance of vaginal health. Front Endocrinol 9 568. (https://doi.org/10.3389/fendo.2018.00568)
Antwis RE , Edwards KL , Unwin B , et al. 2019 Rare gut microbiota associated with breeding success, hormone metabolites and ovarian cycle phase in the critically endangered eastern black rhino. Microbiome 7 27. (https://doi.org/10.1186/s40168-019-0639-0)
Balcázar JL , Gallo-Bueno A , Planas M , et al. 2010 Isolation of Vibrio alginolyticus and Vibrio splendidus from captive-bred seahorses with disease symptoms. Antonie Leeuwenhoek 97 207–210. (https://doi.org/10.1007/s10482-009-9398-4)
Balcazar JL , Lee NM , Pintado J , et al. 2010 Phylogenetic characterization and in situ detection of bacterial communities associated with seahorses (Hippocampus guttulatus) in captivity. Syst Appl Microbiol 33 71–77. (https://doi.org/10.1016/j.syapm.2009.11.005)
Bates D , Mächler M , Bolker B , et al. 2015 Fitting linear mixed-effects models using lme4. J Stat Software 67 1–48. (https://doi.org/10.18637/jss.v067.i01)
Beckmann A , Hüttel S , Schmitt V , et al. 2017 Optimization of the biotechnological production of a novel class of anti-MRSA antibiotics from Chitinophaga sancti. Microb Cell Fact 16 143. (https://doi.org/10.1186/s12934-017-0756-z)
Beemelmanns A , Poirier M , Bayer T , et al. 2019 Microbial embryonal colonization during pipefish male pregnancy. Sci Rep 9 3. (https://doi.org/10.1038/s41598-018-37026-3)
Berglund A & Rosenqvist G 1990 Male limitation of female reproductive success in a pipefish – effects of body-size differences. Behav Ecol Sociobiol 27 129–133. (https://doi.org/10.1007/bf00168456)
Binh DT , Quyen V , Sang T , et al. 2016 Vibriosis in cultured seahorse(Hippocampus spp.) in Khanh hoa province, Vietnam. Int J Innov Stud Aquat Biol Fish 2 43–50. (https://doi.org/10.20431/2455-7670.0202005)
Bray JR & Curtis JT 1957 An ordination of the upland forest communities of Southern Wisconsin. Ecol Monogr 27 325–349. (https://doi.org/10.2307/1942268)
Bunker ME , Elliott G , Heyer-Gray H , et al. 2021 Vertically transmitted microbiome protects eggs from fungal infection and egg failure. Anim Microbiome 3 43. (https://doi.org/10.1186/s42523-021-00104-5)
Carareto Alves LM , de Souza JAM , Varani AdM , et al. 2014 The Family Rhizobiaceae. In The Prokaryotes, 4th ed. pp 419–437. Eds E Rosenberg , EG DeLong , S Lory , E Stackebrandt & F Thompson . Heidelberg, Berlin: Springer. (https://doi.org/10.1007/978-3-642-30197-1_297)
Carcupino M , Baldacci A , Mazzini M , et al. 2002 Functional significance of the male brood pouch in the reproductive strategies of pipefishes and seahorses: a morphological and ultrastructural comparative study on three anatomically different pouches. J Fish Biol 61 1465–1480. (https://doi.org/10.1006/jfbi.2002.2160)
Cary SC & Giovannoni SJ 1993 Transovarial inheritance of endosymbiotic bacteria in clams inhabiting deep-sea hydrothermal vents and cold seeps. Proc Natl Acad Sci U S A 90 5695–5699. (https://doi.org/10.1073/pnas.90.12.5695)
Chao A 1984 Non-parametric estimation of the classes in a population. Scand J Stat 11 265–270.
Chapelais-Baron M , Goubet I , Duchaud E , et al. 2017 Draft genome sequence of the iridescent marine bacterium Cellulophaga lytica CECT 8139. Genome Announc 5 e00811–e00817. (https://doi.org/10.1128/genomea.00811-17)
Chen MH , Sheu SY , Chen CA , et al. 2012 Roseivivax isoporae sp. nov., isolated from a reef-building coral, and emended description of the genus Roseivivax. Int J Syst Evol Microbiol 62 1259–1264. (https://doi.org/10.1099/ijs.0.032961-0)
Clavere-Graciette AG , Mcwhirt ME , Hoopes LA , et al. 2022 Microbiome differences between wild and aquarium whitespotted eagle rays (Aetobatus narinari). Anim Microbiome 4 34. (https://doi.org/10.1186/s42523-022-00187-8)
Clerissi C , De Lorgeril J , Petton B , et al. 2020 Microbiota composition and evenness predict survival rate of oysters confronted to pacific oyster mortality syndrome. Front Microbiol 11 311. (https://doi.org/10.3389/fmicb.2020.00311)
Cole JR , Wang Q , Fish JA , et al. 2014 Ribosomal Database Project: data and tools for high throughput rRNA analysis. Nucleic Acids Res 42 D633–D642. (https://doi.org/10.1093/nar/gkt1244)
Comizzoli P , Power ML , Bornbusch SL , et al. 2021 Interactions between reproductive biology and microbiomes in wild animal species. Anim Microbiome 3 87. (https://doi.org/10.1186/s42523-021-00156-7)
Cousin MA 1999 Pseudomonas|introduction. In Encyclopedia of Food Microbiology, vol 1, pp 1864–1867. Ed RK Robinson . Oxford: Elsevier. (https://doi.org/10.1006/rwfm.1999.1290)
De Roy K , Marzorati M , Negroni A , et al. 2013 Environmental conditions and community evenness determine the outcome of biological invasion. Nat Commun 4 1383. (https://doi.org/10.1038/ncomms2392)
Deng Y , Mao C , Chen H , et al. 2021 Shifts in pond water bacterial communities are associated with the health status of sea bass (Lateolabrax maculatus). Ecol Indic 127 107775. (https://doi.org/10.1016/j.ecolind.2021.107775)
Dudley JS , Hannaford P , Dowland SN , et al. 2021 Structural changes to the brood pouch of male pregnant seahorses (Hippocampus abdominalis) facilitate exchange between father and embryos. Placenta 114 115–123. (https://doi.org/10.1016/j.placenta.2021.09.002)
Edgar RC , Haas BJ , Clemente JC , et al. 2011 UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27 2194–2200. (https://doi.org/10.1093/bioinformatics/btr381)
El-Dakroury MF , El-Gohary MS & El-Gamal AM 2020 Bacterial causes for mortality syndrome in some marine fish farms with treatment trials. Pakistan J Biol Sci 23 1513–1522. (https://doi.org/10.3923/pjbs.2020.1513.1522)
Fuess LE , Den Haan S , Ling F , et al. 2021 Immune gene expression covaries with gut microbiome composition in stickleback. mBio 12 e00145. (https://doi.org/10.1128/mbio.00145-21)
Fukui Y , Abe M , Kobayashi M , et al. 2012 Maritalea porphyrae sp. nov., isolated from a red alga (Porphyra yezoensis), and transfer of Zhangella mobilis to Maritalea mobilis comb. nov. Int J Syst Evol Microbiol 62 43–48. (https://doi.org/10.1099/ijs.0.028704-0)
Funkhouser LJ & Bordenstein SR 2013 Mom knows best: the universality of maternal microbial transmission. PLoS Biol 11 e1001631. (https://doi.org/10.1371/journal.pbio.1001631)
Galushko A & Kuever J 2020 Desulfobacteraceae. In Bergey’s Manual of Systematics of Archaea and Bacteria, pp 1–6. Eds ME Trujullo , S Dedysh , B DeVos , B Hedlund , P Kampfer , FA Rainey & WB Whitman . Hoboken, New Jersey: John Wiley & Sons, Inc.. (https://doi.org/10.1002/9781118960608.fbm00193.pub2)
García-López M , Meier-Kolthoff JP , Tindall BJ , et al. 2019 Analysis of 1,000 type-strain genomes improves taxonomic classification of Bacteroidetes. Frontiers in Microbiology 10 2083. (https://doi.org/10.3389/fmicb.2019.02083)
Garrity GM , Bell JA & Lilburn T 2005 Class II. Betaproteobacteria class. nov. In Bergey’s Manual of Systematic Bacteriology, vol 2, 2nd ed. pp 575–922. Eds DJ Brenner , NR Krieg & JT Staley . Boston, Massachusetts: Springer US. (https://doi.org/10.1007/978-0-387-29298-4_2)
Hahn MW , Schmidt J , Koll U , et al. 2017 Silvanigrella aquatica gen. nov., sp. nov., isolated from a freshwater lake, description of Silvanigrellaceae fam. nov. and Silvanigrellales ord. nov., reclassification of the order Bdellovibrionales in the class Oligoflexia, reclassification of the families Bacteriovoracaceae and Halobacteriovoraceae in the new order Bacteriovoracales ord. nov., and reclassification of the family Pseudobacteriovoracaceae in the order Oligoflexales. Int J Syst Evol Microbiol 67 2555–2568. (https://doi.org/10.1099/ijsem.0.001965)
Harada A , Shiota R , Okubo R , et al. 2022 Brood pouch evolution in pipefish and seahorse based on histological observation. Placenta 120 88–96. (https://doi.org/10.1016/j.placenta.2022.02.014)
Hosokawa T , Hironaka M , Inadomi K , et al. 2013 Diverse strategies for vertical symbiont transmission among subsocial stinkbugs. PLoS One 8 e65081. (https://doi.org/10.1371/journal.pone.0065081)
Hwang CY , Cho KD , Yih W , et al. 2009 Maritalea myrionectae gen. nov., sp. nov., isolated from a culture of the marine ciliate Myrionecta rubra. Int J Syst Evol Microbiol 59 609–614. (https://doi.org/10.1099/ijs.0.002881-0)
Jiang H , Li CY , Zhang B , et al. 2022 Roles of interleukins in antibacterial immune defense of the brood pouch in the lined seahorse Hippocampus erectus. J Oceanology Limnology 40 235–244. (https://doi.org/10.1007/s00343-021-0310-z)
Kämpfer P 2014 Acinetobacter. In Encyclopedia of Food Microbiology, vol 1, 2nd ed. pp 11–17. Eds CA Batt & ML Tortorello . Oxford: Academic Press. (https://doi.org/10.1016/B978-0-12-384730-0.00002-1)
Kawaguchi M , Okubo R , Harada A , et al. 2017 Morphology of brood pouch formation in the pot-bellied seahorse Hippocampus abdominalis. Zoolog Lett 3 19. (https://doi.org/10.1186/s40851-017-0080-9)
Kerwin AH , Gromek SM , Suria AM , et al. 2019 Shielding the next generation: symbiotic bacteria from a reproductive organ protect bobtail squid eggs from fungal fouling. mBio 10 e02376. (https://doi.org/10.1128/mbio.02376-19)
Kim SJ , Ahn JH , Lee TH , et al. 2013a Reyranella soli sp. nov., isolated from forest soil, and emended description of the genus Reyranella Pagnier et al. 2011. Int J Syst Evol Microbiol 63 3164–3167. (https://doi.org/10.1099/ijs.0.045922-0)
Kim Y , Nguyen N , Weon H , et al. 2013b Sediminibacterium ginsengisoli sp. nov., isolated from soil of a ginseng field, and emended descriptions of the genus Sediminibacterium and of Sediminibacterium salmoneum. Int J Syst Evol Microbiol 63 905–912. (https://doi.org/10.1099/ijs.0.038554-0)
Kim Y , Kim B , Kang K , et al. 2016 Sediminibacterium aquarii sp. nov., isolated from sediment in a fishbowl. Int J Syst Evol Microbiol 66 4501–4505. (https://doi.org/10.1099/ijsem.0.001380)
Kimata N , Nishino T , Suzuki S , et al. 2004 Pseudomonas aeruginosa isolated from marine environments in Tokyo Bay. Microb Ecol 47 41–47. (https://doi.org/10.1007/s00248-003-1032-9)
Klindworth A , Pruesse E , Schweer T , et al. 2012 Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res 41 e1. (https://doi.org/10.1093/nar/gks808)
Kuever J 2014 The Family Desulfobacteraceae. In The Prokaryotes, 4th ed. pp 45–73. Eds E Rosenberg , EF DeLong , S Lory , E Stackebrandt & F Thompson . Heidelberg, Berlin: Springer. (https://doi.org/10.1007/978-3-642-39044-9_266)
Kuznetsova A , Brockhoff PB & Christensen RHB 2017 lmerTest package: tests in linear mixed effects models. J Stat Software 82 1–26. (https://doi.org/10.18637/jss.v082.i13)
Lafleur JE , Costa SK , Bitzer AS , et al. 2015 Draft genome sequence of Cellulophaga sp. E6, a marine algal epibiont that produces a quorum-sensing inhibitory compound active against Pseudomonas aeruginosa. Genome Announc 3 e01565. (https://doi.org/10.1128/genomea.01565-14)
Leary S , Underwood W , Anthony R , et al. 2020 Methods of Euthanasia by Species and Environment. In AVMA Guidelines for the Euthanasia of Animals, 2020 ed. pp 84–85. Schaumburg, Illinois: American Veterinary Medical Association. (https://www.avma.org/sites/default/files/2020-02/Guidelines-on-Euthanasia-2020.pdf)
Lee SY , Park S , Oh TK , et al. 2013 Kangiella sediminilitoris sp. nov., isolated from a tidal flat sediment. Int J Syst Evol Microbiol 63 1001–1006. (https://doi.org/10.1099/ijs.0.040691-0)
Lee H , Kim DU , Lee S , et al. 2017 Reyranella terrae sp. nov., isolated from an agricultural soil, and emended description of the genus Reyranella. Int J Syst Evol Microbiol 67 2031–2035. (https://doi.org/10.1099/ijsem.0.001913)
Linton JR & Soloff BL 1964 The physiology of the brood pouch of the male sea horse Hippocampus erectus. Bull Mar Sci 14 45–61.
Liu Y , De Bruijn I , Jack ALH , et al. 2014 Deciphering microbial landscapes of fish eggs to mitigate emerging diseases. ISME J 8 2002–2014. (https://doi.org/10.1038/ismej.2014.44)
Masonjones H & Lewis S 1996 Courtship behavior in the dwarf seahorse, Hippocampus zosterae. Copeia 1996 634. (https://doi.org/10.2307/1447527)
McCauley EP , Haltli B & Kerr RG 2015 Description of Pseudobacteriovorax antillogorgiicola gen. nov., sp. nov., a bacterium isolated from the gorgonian octocoral Antillogorgia elisabethae, belonging to the family Pseudobacteriovoracaceae fam. nov., within the order Bdellovibrionales. Int J Syst Evol Microbiol 65 522–530. (https://doi.org/10.1099/ijs.0.066266-0)
Melamed P , Xue Y , Poon JFD , et al. 2005 The male seahorse synthesizes and secretes a novel C-type lectin into the brood pouch during early pregnancy. FEBS J 272 1221–1235. (https://doi.org/10.1111/j.1742-4658.2005.04556.x)
Mika K , Okamoto AS , Shubin NH , et al. 2021 Bacterial community dynamics during embryonic development of the little skate (Leucoraja erinacea). Anim Microbiome 3 72. (https://doi.org/10.1186/s42523-021-00136-x)
Nakai R , Nishijima M , Tazato N , et al. 2014 Oligoflexus tunisiensis gen. nov., sp. nov., a Gram-negative, aerobic, filamentous bacterium of a novel proteobacterial lineage, and description of Oligoflexaceae fam. nov., Oligoflexales ord. nov. and Oligoflexia classis nov. Int J Syst Evol Microbiol 64 3353–3359. (https://doi.org/10.1099/ijs.0.060798-0)
Nyholm SV 2020 In the beginning: egg-microbe interactions and consequences for animal hosts. Phil Trans Biol Sci 375 20190593. (https://doi.org/10.1098/rstb.2019.0593)
Oconer EP , Herrera AA , Amparado EA , et al. 2003 Immunolocalization of hormones involved in male gestation in the seahorse, Hippocampus barbouri Jordan and Richardson 1908. Philippine Agric Sci 86 84–91.
Oconer EP , General SCP , Pascual CB , et al. 2006 Chemical components of the pouch fluid in the seahorse, Hippocampus barbouri Jordan and Richardson 1908 and their uptake in embryos. Asian Int J Life Sci 15 73–84.
Old JM & Deane EM 1998 The effect of oestrus and the presence of pouch young on aerobic bacteria isolated from the pouch of the tammar wallaby, Macropus eugenii. Comp Immunol Microbiol Infect Dis 21 237–245. (https://doi.org/10.1016/s0147-9571(98)00022-8)
Ortega-Kindica RCMH , Padasas-Adalla CS , Tabugo SRM , et al. 2024 Shotgun metagenomics reveals taxonomic and functional patterns of the microbiome associated with barbour’s seahorse (Hippocampus barbouri). Mar Biotechnol 26 835–841. (https://doi.org/10.1007/s10126-024-10330-y)
Pagnier I , Raoult D & La Scola B 2011 Isolation and characterization of Reyranella massiliensis gen. nov., sp. nov. from freshwater samples by using an amoeba co-culture procedure. Int J Syst Evol Microbiol 61 2151–2154. (https://doi.org/10.1099/ijs.0.025775-0)
Quast C , Pruesse E , Yilmaz P , et al. 2013 The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res 41 D590–D596. (https://doi.org/10.1093/nar/gks1219)
Ripley JL & Foran CM 2009 Direct evidence for embryonic uptake of paternally-derived nutrients in two pipefishes (Syngnathidae: Syngnathus spp.). J Comp Physiol B 179 325–333. (https://doi.org/10.1007/s00360-008-0316-2)
Roeselers G , Mittge EK , Stephens WZ , et al. 2011 Evidence for a core gut microbiota in the zebrafish. ISME J 5 1595–1608. (https://doi.org/10.1038/ismej.2011.38)
Rosenberg E 2014 The family Chitinophagaceae. In The Prokaryotes: Other Major Lineages of Bacteria and The Archaea, 4th ed. pp 493–495. Eds E Rosenberg , E F Delong , S Lory , E Stackebrandt & F Thompson . Berlin, Heidelberg: Springer. (https://doi.org/10.1007/978-3-642-38954-2)
Roth O , Klein V , Beemelmanns A , et al. 2012 Male pregnancy and biparental immune priming. Am Nat 180 802–814. (https://doi.org/10.1086/668081)
Rowe M , Veerus L , Trosvik P , et al. 2021 The reproductive microbiome: an emerging driver of sexual selection, sexual conflict, mating systems, and reproductive isolation. Trends Ecol Evol 36 98. (https://doi.org/10.1016/j.tree.2020.10.015)
Schloss PD , Westcott SL , Ryabin T , et al. 2009 Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol 75 7537–7541. (https://doi.org/10.1128/aem.01541-09)
Schloss PD , Gevers D & Westcott SL 2011 Reducing the effects of PCR amplification and sequencing artifacts on 16S rRNA-based studies. PLoS One 6 e27310. (https://doi.org/10.1371/journal.pone.0027310)
Scobell SK & Mackenzie DS 2011 Reproductive endocrinology of Syngnathidae. J Fish Biol 78 1662–1680. (https://doi.org/10.1111/j.1095-8649.2011.02994.x)
Sethuraman A , Stancheva R , Sanders C , et al. 2022 Genome of a novel Sediminibacterium discovered in association with two species of freshwater cyanobacteria from streams in Southern California. G3: Genes, Genomes, Genetics 12 2160–1836. (https://doi.org/10.1093/g3journal/jkac123)
Shannon CE 1948 A mathematical theory of communication. Bell Syst Tech J 27 379–423. (https://doi.org/10.1002/j.1538-7305.1948.tb01338.x)
Skalkos ZMG , Van Dyke JU & Whittington CM 2020 Paternal nutrient provisioning during male pregnancy in the seahorse Hippocampus abdominalis. J Comp Physiol B 190 547–556. (https://doi.org/10.1007/s00360-020-01289-y)
Skalkos ZMG , Van Dyke JU , Dowland SN , et al. 2024 Paternal protein provisioning to embryos during male seahorse pregnancy. Reproduction 167 e230420. (https://doi.org/10.1530/rep-23-0420)
Skerratt J , Bowman H , Hallegraeff G , et al. 2002 Algicidal bacteria associated with blooms of a toxic dinoflagellate in a temperate Australian estuary. Mar Ecol-Prog Ser 244 1–15. (https://doi.org/10.3354/meps244001)
Sommer S , Whittington CM & Wilson AB 2012 Standardised classification of pre-release development in male-brooding pipefish, seahorses, and seadragons (family Syngnathidae). BMC Dev Biol 12 39. (https://doi.org/10.1186/1471-213x-12-39)
Storesund JE & Øvreås L 2013 Diversity of planctomycetes in iron-hydroxide deposits from the Arctic Mid Ocean Ridge (AMOR) and description of Bythopirellula goksoyri gen. nov., sp. nov., a novel Planctomycete from deep sea iron-hydroxide deposits. Antonie van Leeuwenhoek 104 569–584. (https://doi.org/10.1007/s10482-013-0019-x)
Stumpf RM , Wilson BA , Rivera A , et al. 2013 The primate vaginal microbiome: comparative context and implications for human health and disease. Am J Phys Anthropol 152 119–134. (https://doi.org/10.1002/ajpa.22395)
Suzuki T , Muroga Y , Takahama M , et al. 1999 Roseivivax halodurans gen. nov., sp. nov. and Roseivivax halotolerans sp. nov., aerobic bacteriochlorophyll-containing bacteria isolated from a saline lake. Int J Syst Evol Microbiol 49 629–634. (https://doi.org/10.1099/00207713-49-2-629)
Tachedjian G , Aldunate M , Bradshaw CS , et al. 2017 The role of lactic acid production by probiotic Lactobacillus species in vaginal health. Res Microbiol 168 782–792. (https://doi.org/10.1016/j.resmic.2017.04.001)
Tanger IS , Stefanschitz J , Schwert Y , et al. 2024 The source of microbial transmission influences niche colonization and microbiome development. Proc Biol Sci 291 20232036. (https://doi.org/10.1098/rspb.2023.2036)
Tanu , Deobagkar DD , Khandeparker R , et al. 2012 A study on bacteria associated with the intestinal tract of farmed yellow seahorse, Hippocampus kuda (Bleeker, 1852): characterization and extracellular enzymes. Aquac Res 43 386–394. (https://doi.org/10.1111/j.1365-2109.2011.02841.x)
Van Veelen HPJ , Salles JF & Tieleman BI 2018 Microbiome assembly of avian eggshells and their potential as transgenerational carriers of maternal microbiota. ISME J 12 1375–1388. (https://doi.org/10.1038/s41396-018-0067-3)
Wang J , Lu Y , Nawaz MZ , et al. 2018 Comparative genomics reveals evidence of genome reduction and high extracellular protein degradation potential in Kangiella. Frontiers in Microbiology 9 1224. (https://doi.org/10.3389/fmicb.2018.01224)
Wang Q , Garrity GM , Tiedje JM , et al. 2007 Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl Environ Microbiol 73 5261–5267. (https://doi.org/10.1128/aem.00062-07)
Wang Z , Zhang Z , Li C , et al. 2019 Marinicaulis aureum sp. nov., isolated from a culture of the green alga Ulva prolifera, and emended description of the genus Marinicaulis. Antonie van Leeuwenhoek 112 159–165. (https://doi.org/10.1007/s10482-018-1139-0)
Wang R , Pan X & Xu Y 2020 Altered intestinal microbiota composition associated with enteritis in yellow seahorses Hippocampus kuda (Bleeker, 1852). Curr Microbiol 77 730–737. (https://doi.org/10.1007/s00284-019-01859-6)
Weiss S , Taggart D , Smith I , et al. 2021 Host reproductive cycle influences the pouch microbiota of wild southern hairy-nosed wombats (Lasiorhinus latifrons). Anim Microbiome 3 13. (https://doi.org/10.1186/s42523-021-00074-8)
Whittington CM & Friesen CR 2020 The evolution and physiology of male pregnancy in syngnathid fishes. Biol Rev 95 1252–1272. (https://doi.org/10.1111/brv.12607)
Whittington CM , Musolf K , Sommer S , et al. 2013 Behavioural cues of reproductive status in seahorses Hippocampus abdominalis. J Fish Biol 83 220–226. (https://doi.org/10.1111/jfb.12156)
Whittington CM , Griffith OW , Qi WH , et al. 2015 Seahorse brood pouch transcriptome reveals common genes associated with vertebrate pregnancy. Mol Biol Evol 32 3114–3131. (https://doi.org/10.1093/molbev/msv177)
Whittington CM , Buddle AL , Griffith OW , et al. 2022 Embryonic specializations for vertebrate placentation. Phil Trans Biol Sci 377 20210261. (https://doi.org/10.1098/rstb.2021.0261)
Wiegand S , Jogler M , Heuer A , et al. 2020 Updates to the recently introduced family Lacipirellulaceae in the phylum Planctomycetes: isolation of strains belonging to the novel genera Aeoliella, Botrimarina, Pirellulimonas and Pseudobythopirellula and the novel species Bythopirellula polymerisocia and Posidoniimonas corsicana. Antonie van Leeuwenhoek 113 1979–1997. (https://doi.org/10.1007/s10482-020-01486-3)
Woods CMC 2000 Preliminary observations on breeding and rearing the seahorse Hippocampus abdominalis (Teleostei: Syngnathidae) in captivity. N Z J Mar Freshw Res 34 475–485. (https://doi.org/10.1080/00288330.2000.9516950)
Wu YL , Zhang HX , Zhang B , et al. 2021 Molecular characterization of TLR22 and its role in immunological modification of the brood pouch of the lined seahorse, Hippocampus erectus. Aquaculture 539 736628. (https://doi.org/10.1016/j.aquaculture.2021.736628)
Yoon JH , Oh TK & Park YH 2004 Kangiella koreensis gen. nov., sp. nov. and Kangiella aquimarina sp. nov., isolated from a tidal flat of the Yellow Sea in Korea. Int J Syst Evol Microbiol 54 1829–1835. (https://doi.org/10.1099/ijs.0.63156-0)
Yu XY , Yu XD , Fu GY , et al. 2018 Marinicaulis flavus gen. nov., sp. nov., a novel stalked bacterium of the family Parvularculaceae. Int J Syst Evol Microbiol 68 2061–2067. (https://doi.org/10.1099/ijsem.0.002795)
Zhao Z , Liu Y , Jiang H , et al. 2023 Microbial profiles and immune responses in seahorse gut and brood pouch under chronic exposure to environmental antibiotics. Ecotoxicol Environ Saf 253 114711. (https://doi.org/10.1016/j.ecoenv.2023.114711)
