Introduction
In aquatic animals, gut microbial communities shift with host development and living environments. Understanding the mechanism by which the environment impacts the gut microbial communities of aquatic animals is crucial for assessing and managing aquatic ecosystem health. Initially, fish embryos develop in a relatively constant bacteria-free environment (i.e., within the egg), although some environmental microbes quickly colonize the egg surface from the surrounding water after spawning. After hatching, environmental microbes colonize the gut and after the first feeding, new microbial communities are introduced into the gut with the diet, increasing microbial diversity. The gut microbiota further shifts with host development and with changes in diet. At an early stage, the gut microbiota is influenced mainly by the introduction of environmental microbes with water and diet; however, as the immune system and nutrition metabolism develop, gut microbes are selected and enriched gradually and can be altered with nutritional elements. For this reason, the research project aimed to explore the impact of novel ingredients on these gut communities using metabarcoding analysis of feed- and gut-associated microbial communities.
Materials and Methods
Feeding trial and sampling: the feeding trial was set at a recirculation aquaculture system (RAS) and the growth phase of the trial lasted for 84 days. Briefly, fish were fed with two isoenergetic and isonitrogenous experimental diets in triplicate (3 tanks/diet). A control diet containing only fish and plant proteins (CTRL), and an insect diet containing 10% of Hermetia illuncens meal (Hi10) were tested in rainbow trout (Oncorhynchus mykiss). At the end of feeding trial two fish per tank were sampled (6 fish/diet) and the whole intestine was aseptically removed. The fecal matter containing transient (allochthonous) intestinal bacteria were collected and the resident (autochthonous) microbiota was obtained by scraping the intestinal mucosa of whole gut length and excluding pyloric caeca.
Bacterial DNA extraction and sequencing: DNA was extracted from three samples per feed and six biologic samples of feces-mucosa per each dietary fish group. For the characterization of microbial communities, the V4 region of the 16S ribosomal RNA gene was amplified, using the oligonucleotides:
515F: 5′-GTGYCAGCMGCCGCGGTAA-3’ | 806R: 5′-GGACTACNVGGTWTCTAAT-3 ′
The amplicon library preparation and sequencing were carried out using an Illumina MiSeq platform and protocol (#15044223 rev. B). The raw sequencing data were analyzed with QIIME 2™ (v. 2018.4) pipeline at the default setting (Bolyen et al., 2018). The Silva database was used for the recognition and taxonomic classification of the sequences obtained. The taxonomical classification was performed down to the genus level. OTUs (Operational Taxonomical Unit) assigned to chloroplasts and mitochondria were removed from the analysis as of eukaryotic origin. The relative abundance (%) of each bacterial taxon was determined for every single sample (fish or feed) and for every experimental group (CTRL and Hi 10%). Starting from sequencing data processed by the QIIME pipeline the relative abundance of each OTU was calculated. Alpha (within a single sample) and beta diversity (between samples) have been performed using QIIME alpha-phylogenetic and beta-phylogenetic commands. The “observed OTUs”, “Shannon”, “Faith-PD”, and “Evenness” alpha diversity indices were calculated at the same level of rarefaction, i.e. considering the sample size with a lower number of sequences. Beta diversity was calculated using both the weighted (presence /absence/abundance) and unweighted (presence/absence) UniFrac distance matrix. The UniFrac distances between the microbial communities of the individual samples were visualized by means of two-dimensional scatter plots (PCoA).
Results
Metabarcoding analysis of feed-associated bacteria showed that the microbial communities of CTRL feed differed from those of experimental feed Hi10. The microbial community structure for each feed was outlined at the phylum, class, order, family, and genus. The microbiome profile of all samples comprised 6 different phyla, 11 classes, 17 orders, 25 families, and 30 genera. The number of reads taxonomically classified according to SILVA database was 168,562. Six hundred and thirteen OTUs at 99% identity were identified in feed samples and classified into 134 different taxa. Good’s coverage in all samples reached a value of 0.99, indicating that sequencing coverage was achieved and that the OTUs found were representative of intestinal microbial communities. To elaborate on alpha rarefaction analysis, samples were normalized at a sequencing depth of 3,900 reads. At the order level fish fed diet containing 10% of insect meal showed a significantly higher relative abundance of Bacillales (34%) in their intestinal microflora compared to the control group (1.7%). This important enrichment in Bacillales was due to an increased number of bacteria belonging mainly to Bacillaceae (22%) families in the intestine of fish fed Hi10 diet. Accordingly, at genus taxonomic level, Bacillus, Oceanobacillus, Virgibacillus, and Paenibacillus Corynebacterium were more abundant in Hi10 samples, while fish fed control diet showed higher percentage of Shigella and Shewanella genera, the last one practically absent in fish fed Hi10 diet.
Conclusions
In agreement with the majority of studies on the partial substitution of fishmeal with insect meal (Terova et al., 2019; Rimoldi et al., 2019; Bruni et al., 2018; Antonopoulou et al., 2019; Huyben et al., 2019), the experimental diet with 10% Hermetia illucens (Hi10) had an important effect in modulating the intestinal microbial communities of trout. In particular, Hi10 diet increased gut microbial richness (Observed OTUs index) and numerically the diversity (Shannon diversity index p=0.05) as compared to control fish. In general, high gut microbial richness and diversity are considered desired features because they are usually associated with the health status of the host. Based on the ratio of Firmicutes:Proteobacteria, fish fed Hi10 diet showed a dominance of Firmicutes that has been associated with dietary plant ingredients that are rich in fibers, whereas an animal protein-based diet usually favors the dominance of Proteobacteria (Terova et al., 2019; Ingerslev et al., 2014 a; 2014b). Dietary inclusion of insect meal influenced the microbial intestinal profiles of trout both qualitatively and quantitatively. Indeed, multivariate analysis of bacterial communities revealed a significant relationship between diet and microbiota associated with fish intestines.