We characterized the skin and gill microbiota of different age groups of farmed European seabass and gilthead seabream using 16S rRNA amplicon high-throughput sequencing. By taking into account potential environmental and seasonal effects, our study shows that fish age, in particular sexual maturation and growth, impact skin and gill microbial diversity (Table 1; Figure 1), composition (Additional file 4; Figure 3, 5) and predicted microbial functions (Additional file 6; Figure 4).
Microbial diversity across age groups
Fish growth and sexual maturation are usually accompanied by extreme morphological and physiological changes [e.g., 44, 45]. Importantly, some of these changes reported for the skin and gills have been suggested to also affect their microbiota. For example, changes in epidermal structure derived from sexual maturation (e.g., increases in the number, size and activity of the mucous cells) have been detected in several fish species [e.g., 44, 46], and suggested to increase infection rate with Saprolegnia fungus in sea trout and brown trout [47]. Likewise, changes in the hormones expressed in the skin alter the biochemistry of the skin mucous and also potentially affect its microbiota [48]. Fish growth and sexual maturation also impact gill morphology and function in some fish species. For example, the ability to osmoregulate at different salinities was seen to increase throughout the developmental stages (larva to juvenile) of seabass [45]. Additionally, body size was also identified as the main factor affecting morphological variation in gill rakes and gill pore size in the Silver Carp and Gizzard Shad, suggesting that the overall filtering ability of these species is related to size and maturation [49]. Importantly, a recent study showed that body weight increase is accompanied by higher microbial community structuring in the skin and gill of rabbit fish [36]. We thus hypothesize that such physiological and morphological changes occurring during fish growth have led to the changes in microbial diversity, composition and predicted functionality observed in the present study.
The seabass skin and gill microbiotas of older age groups showed significantly higher alpha-diversity than those of early juveniles. Additionally, a higher percentage of significant differences in the relative abundance of the most abundant phyla and genera occurred between early juveniles and older age groups (67±27% and 55±38% in the skin and gill, respectively; Additional file 5). This suggests that the skin and gill microbiotas of the seabass were highly dynamic, diversifying with age. Conversely, the skin and gill microbiotas of the seabream juveniles and adults showed similar alpha-diversity means, although a high percentage of the most abundant phyla and genera varied between age groups (70±42% and 63±18% in the skin and gill, respectively; Additional file 5). Variation in microbiota alpha-diversity between different age groups has been previously reported for many fish species. For example, studies on the zebrafish and salmon gut microbiota, have reported differences between mature and immature life stages; however, those differences also coincide with other major ecological changes in the fish, such as diet [17] or environment transitions [20]. Moreover, the relative abundance of predominant bacterial groups also changed with aging in other fish species [e.g., 14, 17, 20].
We detected significant differences in microbial structure across all age groups in both species. Similar results have been also reported in other fish (e.g., several reef fish species [14, 35] and Salmo salar [19]), being particularly evident in longitudinal studies encompassing several months [13, 17, 20]. High inter-individual variability within age groups was also previously reported for other fish species [e.g., 14, 28, 30]. However, our results showed that fish age only explained a low percentage of the variation in the bacterial community structure (R2 < 0.1). This suggests that microbial differences between age groups are small at the community level, but clearly noticeable at the species level, with a high proportion of the predominant bacterial taxa (61±39% and 46±32% in the skin and gill of seabass, respectively; 70±42% and 63±18% in the skin and gill of the seabream, respectively) changing their abundance with sexual maturation. Although our statistical models accounted for sampling dates as a random factor, other biotic and abiotic factors (e.g., variation in the environment and individual weights) could be responsible for most of the variation observed in community structure [e.g., 12, 19, 28].
Microbial predicted functional diversity across age groups
The predicted functional analysis suggests that distinct significantly enriched metabolic pathways are expressed in skin and gill microbiotas of both fish species across age groups. Although metabolic information is particularly limited for fish microbiotas, studies on other vertebrates, mainly in humans and their gut microbiotas, are starting to shed light on the beneficial outcomes specific microbial metabolic functions have on the host health and physiology [e.g. 50].
Notwithstanding that present results should be interpreted with caution since PICRUSt2 analysis is limited by the currently available genomes and biased towards human health microorganisms [51], one could suggest that some of the enriched metabolic pathways found in our analysis could also improve the seabass and seabream health and physiology. The protective role of the microbiota is often related to the production of secondary metabolites that provide chemical defense and mediate bacterial diversity [4]. Secondary metabolites with antimicrobial activity have been previously isolated from microbial species inhabiting the gut microbiome of fish [52]. Here, the biosynthesis of secondary metabolites that have been associated with antimicrobial activity, including hemiterpene [e.g., 53], were enriched in mature adults of both species. Additionally, the biosynthesis of chlorophyll and several amino acids, herein enriched in older age groups of both fish species, have also been found to be expressed in the skin and gut of healthy humans [e.g., 54; 55]. Importantly, amino acid biosynthesis was reported in the gut microbiota of grass carp when fed a protein-deficient diet, suggesting a metabolic role of the gut microbiota towards fish nutrition [56]. The biosynthesis of vitamins, here enriched in older age groups of seabass, has been found beneficial for human skin (e.g., Saxena et al., 2018) and gut mucosa, including folate and thiamine [57]. In addition, polyamines are bacterial metabolites known to have several benefits towards gut mucosa recovery [58]. These pathways were also enriched in the juveniles of seabream.
It is also worth noticing that some of the enriched metabolic pathways detected in the present study could be driven by the high environmental variability of the Alvor estuary where these fish were reared. In estuaries, salinity variations occur on a daily basis due to tides and pollutants can be prevalent [e.g., 59]. Biosynthesis of fatty acids and unsaturated fatty acids were two of the predicted metabolic pathways enriched in the microbiota of mature seabream. These same pathways have also been enriched in previous analyses of the skin and gut microbiota in the atlantic salmon [60, 61] and in the skin microbiota of the common snook [62] when transitioning between freshwater and seawater. Additionally, two of the predicted metabolic pathways identified in both fish species were related to degradation of toxic compounds. Specifically, biodegradation of the highly prevalent toxic pollutants toluene and chloroaromatic compounds by bacteria is essential to remove them from the environment and to prevent absorption through the skin and gills in aquatic animals [63-65].
Following alpha-diversity patterns, fish from older age groups, particularly in the seabass, had greater enrichment of predicted functions related mainly to the biosynthesis and degradation of compounds; as well as, to a lesser extent, metabolism and energy cycles. We then hypothesize that the increase in microbial diversity observed as fish ages leads to wider functional diversity. This could prove beneficial to those fishes, given the key physiological modifications older fish groups are experiencing during sexual maturation and growth.
Fish and water microbiota comparisons
The water microbiotas of fishponds were significantly distinct and more diverse than the skin and gill microbiota of both fish species, regardless of their age. It is known that free-living microbial communities retain higher richness than host-associated communities [31], with many studies showing a higher bacterial diversity in water relative to fish skin [28, 30, 36, 66-68], gills [14, 36], gut [7, 15, 18, 21, 69], stomach [36], hindgut [36] and whole larvae [22]. Although some studies in fish have shown that the microbial communities found in the water tend to be recovered in the larval gut microbiota [17, 21], others have also shown that water microbiota does not influence directly the microbiota of the fish mucosa [7, 8, 13-15, 18, 19, 22, 28, 30, 34, 36, 66-68, 70, 71]. Importantly, a previous study of the skin microbiotas of seabass and seabream also showed significant differences with planktonic communities [67]. However, in that study only a low number of Operational Taxonomic Units (3%) was shared between skin and water microbiota; whereas in the present study higher percentages of ASVs were shared between the skin (14%±1) and the gill (15%±1) of both fish species and the surrounding water.
Microbial dissimilarities depicted by PCoAs showed that, although significantly different, the skin microbiota of both species clustered more closely to the water microbiota than the gill microbiota. However, only a small percentage of the variation (PC 1 – average 18%±2; PC 2 – average 10%±1) was explained by this analysis. On the other hand, the results from the Mantel tests showed a significant (p<0.03) correlation between the water and gill microbiotas, but not between the water and skin microbiotas. This suggests that although both skin and gill are permanently in contact with water, the gill environment may be more susceptible to variations in the water microbiota.