UNRAVELING THE CORE GILL MICROBIOME OF Sparus aurata BETWEEN WILD AND FARMED POPULATIONS

Introduction

 In teleosts , skin, gills and gut are the major mucosal surfaces (MS) which also act as host-microbiota interface, where they can form mutualistic relationships.  To date, numerous studies have revealed the importance of host microbiomes and the various factors (host associated or environmental) affecting their dynamics (Egerton et al. 2018). Despite that there is progress in research on fish microbiome there is limited knowledge on the microbial structure of natural fish populations and any differences that may exist in the composition between wild and farmed fish species. Wild fish microbiota can be used as a fish health indicator to improve production through innovative applications of microbiomes-tailored products , but also for other interventional strategies, such as conservation practices. In this study, we analysed the gill microbiomes of 20 wild-caught and 32 farmed Mediterranean gilthead seabream (S. aurata) specimens by 16S rRNA gene sequencing to detect structural and inferred functional differences between the two different sets of individuals.

Materials and Methods

 In total 52 tissue samples of gills  were  collected from S . aurata individuals. The specimens obtained by experimental/scientific fishing  in the Ionian Sea (Wild: n = 20 ; weight = 147.2 g ± 44.7)  and by a distantly located (Aegean Sea coast in central Greece)  commercial aquaculture open-sea unit [Farmed n = 32; weight = 127.4 g ± 185.7 (average ± SD)]. The farmed fish species were collected in 4 different time points across the aquaculture production cycle (between ~100 to 500 days after their  transfer in the sea cages). After dissection with sterilized equipment ,  and prior to their storage, samples were rinsed with sterile particle free seawater, to reduce surrounding environment associated bacteria. W ater samples (n=8) were taken from the sea cages and filtered by using 0.2-μm isopore  membrane filters to enable comparisons of the microbial communities in the water column with those present on the fish gill. Total DNA was isolated from the collected samples using the DNeasy PowerSoil Pro Kit (Qiagen) and 16S rRNA gene sequencing was applied for the identification of their bacterial targeting the V3–V4 regions of the 16S rRNA gene. Sequencing was performed on a MiSeq Illumina instrument (2 × 300 bp). Raw data were analyzed on the MOTHUR v.1.48.0 (Schloss at al. 2009) . PICRUSt analysis  (Douglas et al. 2020)  was  applied to predict the genetic functions of the detected bacterial taxa. Statistical analyses performed in Past4 software (Hamme et al. 2001).

Results & Discussion

A total of 52 gill samples, collected from both wild-caught and farmed populations, along with 8 water samples,  were analyzed for this study. The bioinformatic analysis yielded 4678 operational taxonomic units (OTUs) , after subsampling to the size  (n = 10313) of the smallest group . The wild group microbiota presented the highest number of unique OTUs, while  254 (5.9% of the total OTUs) were  shared among the microbiomes of the two S . aurata groups. Overall, OTUs represented 33 bacterial phyla. Proteobacteria and Bacteroidota accounting for 75.78% of the OTUs abundance. The most abundant OTUs were identified as unclassified Burkholderiales  sp. (Gills) and Synechococcus  sp. (Water). PERMANOVA analysis detected significant differences (PERMANOVA: p

<

0.001 using Bray–Curtis dissimilarity indices , 9999 permutations) in community composition between the gill microbial communities of the two S . aurata groups (Wild and Farmed) but also  between host (Gill) and environmental samples (Water). The clear separation between S . aurata samples and Sea water, supporting previous findings suggesting that S . aurata and other fish species harbours distinct microbial communities compared to those found in their surrounding environment (Legrand et al. 2018; Quero et al. 2022). According to SIMPER analysis the average dissimilarity between Wild and Farmed group was 83.86% with only 6 OTUs presenting a cumulative contribution of 49.9%. It is noteworthy that in ordination of samples, based on the same distance matrix, all samples from the farmed group (except of those obtained during the initial sampling, which had spent the  minimum  time in the sea) clustered closely with those of the Wild group. The relative abundance of the predicted MetaCyc pathways from PICRUST analysis revealed that the two groups do not differ significantly (ANOSIM, p = 0.0611 ) and their most abundant (relative abundance ³ 1%) pathways are involved in Aerobic Respiration, Fatty Acid ,  Lipid  and Phospholipid Biosynthesis, suggesting that despite the structural differences of the gill’s microbiota, these bacterial communities have no specific metabolic pathways. The same analysis revealed that the predicted metabolic pathways of the water samples shared similarities with  the gill  microbiota of the farmed group (ANOSIM, p = 0.12), but there were significant different when compared to the microbiota of the wild (ANOSIM, p = 0.0001).

Conclusion

Knowing wild fish microbiota, could contribute to unravel the natural microbiome variation and enhance our knowledge on microbial diversity and host-microbe interactions. Our study revealed that gill microbial community composition differs significantly between  Wild and Farmed populations. However, these differences tend to decrease as farmed populations spend more time in sea. These results suggest that host genetic background and the habitat status, due to human intervention, affect the bacterial dynamics in S . aurata gills. In addition, the existence of a core microbial community and the similarities in metagenomic functions strongly imply a coevolutionary relationship between S . aurata and its gill microbiome. These findings highlight the importance of thoroughly exploring these host-microbe interactions and taking them into account when developing conservation strategies and designing microbiome-based approaches.

References

 Douglas GM, Maffei VJ, Zaneveld JR, Yurgel SN, Brown JR, Taylor CM, Huttenhower C, Langille MGI. PICRUSt2 for prediction of metagenome functions. Nat Biotechnol . 2020 Jun;38(6):685-688. doi: 10.1038/s41587-020-0548-6. PMID: 32483366; PMCID: PMC7365738.

 Egerton S, Culloty S, Whooley J, Stanton C, Ross RP. The Gut Microbiota of Marine Fish. Front Microbiol . 2018 May 4;9:873 . doi: 10.3389/fmicb.2018.00873. PMID: 29780377; PMCID: PMC5946678.

Hammer Ø, Harper D, Ryan P (2001) Past: Paleontological statistics software package for education and data analysis. Palaeontol . Electr. 4, 9.

 Legrand TP, Catalano SR, Wos -Oxley ML, Stephens F, Landos M, Bansemer MS, Stone DA, Qin JG, Oxley A (2018) The inner workings of the outer surface: skin and gill microbiota as indicators of changing gut health in yellowtail kingfish. Front Microbiol 15(8):2664

Quero GM, Piredda R, Basili M, Maricchiolo G, Mirto S, Manini E, Seyfarth AM, Candela M, Luna GM (2022) Host-associated and Environmental Microbiomes in an Open-Sea Mediterranean Gilthead Sea Bream Fish Farm. Microb Ecol https://doi.org/10.1007/s00248-022-02120-7

Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB, Lesniewski RA, Oakley BB, Parks DH, Robinson CJ , Sahl JW, Stres B, Thallinger GG, Van Horn DJ, Weber  CF (2009)  Introducing mothur:  Open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl. Environ. Microbiol . 2009, 75, 7537–7541.