In the last decades, research on the use of insects as FM replacers in aquafeed is rapidly evolving. Several reviews have been published on insects nutritional value, environmental low impact, and food safety, all attributes that could contribute to make aquaculture system more productive and sustainable [4,6,7,35].
In terms of fish growth, the research of our group, as also reported by Chemello et al. [31], confirms what has been found in previous studies, i.e. the complete or partial substitution of dietary fishmeal with TM does not affect rainbow trout growth performance and fillet quality [10–12]. Similarly, TM was successfully utilised and well accepted by several marine fish species, such as European sea bass (Dicentrarchus labrax) [36], gilthead sea bream (Sparus aurata) [37], blackspot sea bream (Pagellus bogaraveo) [38], and red sea bream (Pagrus major) [13].
While the effects of dietary FM/TM replacement on fish growth performances have been widely investigated, less evidence is available on the effects on host commensal bacterial communities. Therefore, this study aimed at evaluating the effects of total replacement of FM with T. molitor larvae meal on rainbow trout gut and skin microbiome. In particular, skin microbiome is underexplored in fish as well as in most farmed animals.
The data showed no major effects of FM substitution with TM meal on species richness and diversity of both gut mucosa- and skin mucus-associated bacteria. In line with our results, the inclusion of hydrolysed TM meal did not affect the total number of digesta-associated bacteria in sea trout (Salmo trutta m. trutta) [39]. In contrast, in the study of Józefiak et al. [40], the total number of intestinal bacteria increased in rainbow trout fed a diet in which FM was partially replaced by TM in comparison to control fish that were fed a FM-based diet.
Interestingly, Antonopoulou and colleagues [17] reported that the dietary inclusion of T. molitor larvae meal led to a five-fold increase of Simpson dominance D index, and to a two-fold decrease of the Shannon H index in rainbow trout gut microbiota, but not in sea bream and sea bass microbiota in which the same diversity indices remained practically unchanged. This evidence suggests a species-specific impact of insect meal on gut bacterial communities. Equally, in our previous studies, we found an increase of bacteria species richness and diversity in intestinal microbiome of trout fed diets with partial replacement of FM with Hermetia illucens meal [15,16].
Regardless of the diet type, marked differences in terms of alpha diversity were found between gut and skin microbiota, being the latter characterized by higher microbial diversity and richness. Previous studies on trout and other fresh water species displayed a similar trend with a lower alpha diversity in the gut than in the skin mucosal surface [22,41,42]. Unfortunately, in contrast to high number of studies focused on fish gut microbiome, the skin mucus microbiome remains largely underexplored.
Initially, fish skin is colonized by bacteria present in the water, but over time, the superficial mucus harbors an increasingly divergent microbial community [41,43]. Like in intestine, the balance between members of skin microbial community, i.e., commensals, symbionts or pathogenic bacterial strains, collectively forming skin microbiome, is important to preserve fish health. It is well known that factors such as diet, water quality, seasonality, host physiology, infections, and stress can shape the composition of fish microbiomes and influence the balance of the microbic ecosystems [28–30].
Our metabarcoding analysis showed that rainbow trout skin microbiome was largely dominated by Proteobacteria, and especially Gammaproteobacteria, which constituted approximately half of the bacterial taxa found. This result is in agreement with previous studies on other fish species regardless of the technique used for bacterial identification [21–23,25,26,44–46]. Gammaproteobacteria class includes several potentially pathogenic bacterial species for fish, such as Vibrioanguillarum, and Photobacterium damselae. Actually, there are several evidences supporting the role of fish skin microbiota as an important niche for mucosal pathogen evolution in nature [44]. For instance, potentially pathogenic Vibrio, such as Vibrio anguillarum and Vibrio cholerae, monopolize skin microbiome of wild eel (Anguilla anguilla) from estuary and wetland [44]. Other accidental pathogens identified in wild eel have been Pseudomonas aeruginosa, Stenotrophomonas maltophilia, Achromobacter xylosoxidans, and Aeromonas veronii. Similarly, skin microbiome of coral reef fish showed a significant enrichment in Gammaproteobacteria, especially Vibrionaceae [26].
Although in the present study trout skin microbiome was dominated by the Gammaproteobacteria’s family of Aeromonadaceae instead of Vibrionaceae, at genus level, Pseudomonas, Stenotrophomonas and Citrobacter were present in our samples likewise in wild and farmed eel skin microbiome [44]. This result is quite interesting, since previous studies have indicated that fish skin microbiome is species-specific, both in terms of bacterial diversity and bacterial community structure, showing significantly lower variability between individuals from the same species than between those of different species [21,26].
The low frequency of Vibrio genera in trout skin microbial community could be explained by the fact that trout is a freshwater fish while Vibrio are mainly marine bacterial genera. It is widely accepted, indeed, that the skin of fish harbors a complex and diverse microbiota that closely interacts with the microbial communities of the surrounding water.
In line with our data, Lowrey and co-workers [22] reported that Proteobacteria and Bacteroidetes were the most abundant phyla of rainbow trout skin microbiota, however at genus level they found a skin bacterial community consistently composed by Flectobacillus. These apparently controversial evidences are inevitable since, up to date, few studies have investigated skin microbiome in freshwater fish, and it is not yet known if it fundamentally differs from that of marine fish [45].
With regard to skin microbial community composition, the two dietary groups did not display distinctive features, except for a decrease in the relative abundance of Deefgea genus (family Neisseriaceae) in skin microbiome of trout fed with insect meal. Changes in the skin microbiota of fish in response to stressors, such as hypoxia have been previously observed, in brook charr (Salvelinus fontinalis), in which probiotic-like bacteria decreased after stress exposure [47]. Studies in salmonids have also shown that parasitic infections or other microbial aetiological agents (e.g. viruses) may perturb skin microbiota [25].
In agreement with our recent study in rainbow trout [16], metagenomic analysis indicated that Tenericutes was the most abundant phylum in trout intestine, regardless of the diet. Specifically, within this phylum, the Mollicutes, mainly represented by Mycoplasmataceae family, were the dominant class. The Tenericutes are among the protagonists of gut symbionts of rainbow trout, indicating that they are possibly related to the metabolism of the host [22,48,49]. Although diet is the most important external factor affecting the gut microbiota composition, in this case we observed only a weak dietary modulation of intestinal bacterial communities. The only changes due to dietary FM substitution with TM meal were a decreased number of Proteobacteria and, at family level, a reduced number of taxa assigned to Ruminococcaceae and Neisseriaceae.
In line with our results, Antonopoulou et al. [17] reported that T. molitor meal replacement affected the dominant intestinal phyla less in rainbow trout than in sea bream and sea bass. In contrast, there are several evidences that FM replacement with insect meal from black soldier fly (Hermetia illucens) larvae positively modulates gut microbiota of rainbow trout by increasing the proportion of lactic acid bacteria (LAB), which are generally considered as beneficial microorganisms and frequently used as probiotics in fish and other vertebrates diet [15,16,50].
Actually, there is a study stating that the inclusion of 20% TM meal in the diet increased the intestinal population of Lactobacillus and Enterococcus genera in rainbow trout juveniles [51]. The increase of LAB by dietary insect meal could be related to the prebiotic properties of chitin. Chitin is an insoluble linear polysaccharide (a biopolymer of N-acetyl-β-D-glucosamine) that confers structural rigidity to insects’ exoskeleton. Partial or full enzymatic deacetylation of chitin produces chitosan. Both chitin and chitosan are hardly digested by the majority of fish [18]; therefore, once consumed, the fermentation of both polysaccharides is largely performed by gut microbiota. The lack of enrichment in intestinal LAB during the present study was an unexpected result, especially when compared to what has been previously observed in the intestine of trout fed with diets containing H.illucens larvae meal [15,16]. The main effect of the dietary inclusion of this type of insect meal was a significant increase of Firmicutes at the expense of Proteobacteria phylum. The dietary administration of TM meal caused instead only a decrease in relative amount of Proteobacteria without any increase in Firmicutes.