Introduction
Fish closely interact with their environment and the microorganisms that inhabit it. Accordingly, increasing attention is being given to the interplay between microbiota, host, and environment. While most studies focus on intestinal bacteria, other microbial niches—such as the skin and the surrounding water—are gaining importance as less or non-invasive biomarkers for monitoring fish health and welfare. There is now evidence that certain behavioural patterns can be modulated by targeting the gut microbiota (Olorocisimo et al., 2023). The relationship between the skin microbiota and specific behaviours has been less studied, although correlation analyses have also supported an association between skin microbiota and reactive or proactive behaviour in gilthead sea bream (Sparus aurata) reared at high stocking densities (Toxqui-Rodriguez et al., 2024). To further explore this line of research, the present study aims to characterize how seawater microbial dynamics influence and interact with the skin and rearing water microbiomes throughout the production cycle of gilthead sea bream. Moreover, by understanding the interactions among these three interconnected microbiomes, we explore the potential of using rearing water microbiota as a non-invasive proxy for monitoring host-associated microbial shifts in fish fed different diets formulations.
Methods
Gilthead seabream juveniles were reared under natural photoperiod and temperature conditions (40°5’N; 0°10’E) at IATS facilities in an open-flow system. During the trial, fish were fed with a CTRL fish meal-based diet, or alternative feeds with processed animal protein (PAP), or insect meal + bacterial protein (ALT) as main novel fish meal replacers. Fish skin and rearing tank water microbiome were sampled at four time points: t0-May (13.2g); t1-July (70.3g); t2-November (295.9g); and t3-February (357.5g). At each sampling point, 9-12 fish per diet were sampled for skin mucus, and water samples were filtered (0.22 µm) to collect bacterial biomass. In parallel, to study seasonal fluctuations in the inlet seawater of the system, water samples were collected monthly during a two-year period from the inlet pipe (Jan-2023 to Dec-2024). After DNA extraction, full-length 16S rRNA (V1–V9) was amplified and sequenced using Oxford Nanopore (PromethION P2). Reads were basecalled (Dorado v0.9, SUP), and taxonomic assignment was performed with minimap2 using SILVA v138.2 as reference database.
Results and discussion
Microbiome composition of inlet and rearing water showed a dominance of three main phyla: Pseudomonadota , Bacteroidota , and Cyanobacteria. Similarly, the skin microbiota maintains the dominance of the first two phyla, while the presence of Cyanobacteria was significantly lower, being replaced by the Bacillota phylum. In terms of OTU prevalence, 130 taxa at genus level were shared consistently between water and the skin mucus. Interestingly, the abundances of 94 taxa (approximately 70% of the total shared microbiota) were positively or negatively correlated between skin and rearing water microbiomes. To refine this set of bacteria select ed across time , we applied a time series analysis (RAIN analysis; Coenen et al., 2020) to seawater inlet samples collected across two consecutive years. As a result, among a list of 102 genera (p < 0.01) whose abundances were positively (Figure 1A) or negatively (Figure 1B) correlated with seawater temperature, 24 taxa exhibited the same seasonal variation in the skin mucus and rearing water microbiomes. These bacteria (representing 21.1 and 16.8% of the total water and skin abundance) are relevant because apart of being consistent in our facilities environment, can be measured in the rearing water as a non-invasive proxy of a host-associate microbiome.
In addition to time, beta-diversity analysis of skin microbiome revealed that diet represented the secondary source of variation. In fact, even with lower influence compared to temporal variable, LEfSe analysis identified 11 bacterial biomarkers associated with the diet including Alteromonas , Cobetia and various members of Paracoccaceae family . Notably, Alteromonas—the most abundant diet-discriminant genus—has previously been associated with a reactive behaviour in fish under high-stocking densities (Toxqui-Rodriguez et al., 2024) . Interestingly, the ALT diet , enriched in this genus , has also been associated with this behavioural trait, reinforcing a possible link between diet, behaviour, and the skin microbiome.
Concluding remarks
This work reveals the cross-talk between the microbiomes of rearing water and fish skin mucus, supporting the potential use of rearing water microbiota as a practical, non-invasive tool for monitoring host-associated microbial shifts. These results reinforce the idea that water microbiome analysis can complement fish health monitoring strategies, offering a non-invasive approach to track broad microbial fluctuations within aquaculture environments.
Funding
This work was supported by MCIN with funding from European Union NextGeneration EU (PRTR-C17.I1) and by Generalitat Valenciana (THINKINAZUL/2021/024).
References
Coenen et al. Front. Genet. 11:310 (2020); Olorocisimo et al. Neuropharmacology 109401 (2023) ; Toxqui-Rodríguez et al. Microorganisms 12(7), 1360 (2024).