The diversity of the gastrointestinal bacterial community and its relationship to Atlantic salmon health and productivity

The Atlantic salmon (Salmo salar L.) gastrointestinal tract has a dynamic microbial community and its structure has been suggested to be a factor influencing fish health and productivity. Less productive salmon growth performance during times of temperature stress has been suggested to be linked to various factors including diet quality and the incumbent gastrointestinal tract microbial community. As salmon farming is one of Tasmania's largest agrisector industries, there is a strong motivation to better understand, protect and sustain salmon health and productivity. The aim of this thesis was to providing strategic information in relation to how gastrointestinal microbiology relates to farm environmental conditions and diet composition. Overall, this study provides improved understanding of farmed salmon gastrointestinal bacterial communities, what drives their dynamism and describes the relative impact of diet, environment and farm management factors. The relationship of Atlantic salmon gastrointestinal tract bacteria to environmental factors, in particular water temperature within a commercial mariculture system, was investigated. Salmon gastrointestinal tract bacterial communities of commercially farmed in south-eastern Tasmania was analysed, over a 13 month period across a standard commercial production farm cycle, using 454 16S rRNA-based pyrosequencing. Faecal bacterial communities were highly dynamic but largely similar between randomly selected fish. In post-smolt, the faecal bacteria population was dominated by Gram-positive fermentative bacteria, however by mid-summer members of the family Vibrionaceae predominated. As fish progressed towards harvest, a range of different bacterial genera became more prominent corresponding to a decline in Vibrionaceae. The sampled fish were fed two different commercial diet series with slightly different protein, lipid and digestible energy levels; however the effect of these differences was minimal. The overall data demonstrated dynamic hind gut communities in salmon that were related to season and fish growth phases but were less influence by differences in commercial diets used routinely within the farm system studied. This study provides understanding of farmed salmon gastrointestinal bacterial communities and describes the relative impact of diet, environmental and farm factors. Less productive Atlantic salmon growth performances have been linked to temperature stress, diet and indirectly linked to the incumbent GI tract microbial community. To obtain knowledge that may aid management of salmon production during abnormally warm summer periods as well as to better understand salmon GI tract microbial community dynamics a feeding trial was performed over a summer period. The diets were tested over a 5 month period in relation to a commercial standard diet that has intermediate protein levels (45:25 protein:lipid, 35% fishmeal, IntPro). Modified diets tested included a low (10% w/v) fish meal content diet (LFM diet); a diet with a high protein and energy content (50:20 ratio, 21.4 MJ/kg digestible energy, HiPro) and a low protein, low energy diet (40:30, 19.7 MJ/kg, LoPro). A six point categorical scoring system was developed to describe expressed digesta consistency, where a low score describes 'normal' faeces and a high score denotes casts (pseudofaeces), or empty hind gut. The faecal score‚ÄövÑvp was used as a proxy for gut function. Faster growing fish generally had lower faecal scores and this was due to accelerated growth of sexually immature subpopulations while the diet cohorts showed comparatively little difference in terms of faecal score though the overall lowest scores were observed after 5 months with the LoPro diet. The GI tract bacterial communities assessed with 16S rRNA amplicon pyrosequencing were dynamic over time with the LoPro diets most strongly shifting the community structure in relation to the commercial standard diet used. During the summer period the LoPro diet cohort and to a lesser extent all other cohorts with standard fish meal content had transient increases in GI tract community diversity mainly represented by an increased abundance of anaerobic (Bacteroidia and Clostridia) and facultatively anaerobic (lactic acid bacteria) taxa. The digesta had enriched populations of these groups in relation to faecal casts. The majority of samples (60-86%) across all diet cohort faecal communities were eventually dominated by the marine-derived, bile-tolerant marine facultatively anaerobic genus Aliivibrio. The results suggest that time (incorporating seasonal changes in temperature) and diet is potentially related to faecal microbial community structure. Categorization of the digesta via the faecal scoring system revealed strictly anaerobic taxa were comparatively more abundant in firmer, normal faecal samples that are also rich in plant chloroplast material suggesting significant diet digestion had occurred therein. Anaerobes were comparatively much less populous in pseudofaeces, which is generally associated with poor feeding. These community shifts possibly through formation of different levels of metabolites and/or immune system stimulation could influence salmon physiology and farm-level performance outcomes. In order to better understand microbial changes within the salmon GI tract at the dietary level, the microbial community dynamics were assessed within a simple in vitro growth model system. In this system the growth and composition of bacteria were monitored within diet slurries held under anaerobic conditions inoculated with salmon faecal samples. This system was assessed using total viable bacteria counts (TVC), automated ribosomal intergenic spacer analysis (ARISA), and 16S rRNA pair-end Illumina-based sequence analysis. A total of 5 complete diets were tested including low fish meal (LM), low protein (LP), high protein (HP), a commercial standard diet with intermediate protein and lipid content (CS) and a CS diet version where fish oil was completely replaced with poultry oil (PO). In addition plant meals (lupin kernel meal and pea extract, referred to as the LK and PE diets) were tested in isolation to determine if plant-derived material promotes the growth of specific bacteria. The in vitro model cultures were incubated at 20¬¨‚à´C to simulate warm summer temperatures. The microbial growth in the diet slurries after a lag phase of ~3 h grew over a 24 h period with a progressive decline in pH. TVC counts indicated growth on MA and TCBS plates were equivalent indicating most bacteria that grew were bile salts tolerant. ARISA and Illumina sequencing data revealed there was very clear separation between the complete diets and the LK and PE plant meal diets suggesting bacteria that grew were very distinct. The sequencing analysis showed in the case of the complete diets those members of the genera Aliivibrio, Vibrio and Photobacterium became greatly predominant. However based on replicated experiments there was evident stochasticity of what exact species became dominant. Vibrionaceae may have become predominant due to their rapid growth capacity, relatively high abundance within the starting faecal material and salt tolerance though several other bacterial taxa were also present in great abundance initially. The LK and PE diets only allowed the growth of the aerobic genus Sphingomonas no other faecal-associated bacterial grew including Vibrionaceae suggesting a combination of protein and lipid diet components structure the salmon GI tract community. Taken together the data suggests acyclic dynamism in farmed Atlantic salmon GI tract populations is the norm with largely Vibrionaceae predominant beyond the post-smolt phase. Vibrionaceae are successful owing to their bile tolerance and rapid growth on diet nutrients. In future studies, to achieve demonstrable farm-level performance outcomes via diet manipulation focus should be placed on dietary additives and experimental strategies instigated to more stably manipulate and influence GI tract communities, especially if probiotic supplementation is intended. Such experiments will need to be performed in tandem with deeper investigations of salmon physiological responses (cell biology, gene expression, protein profiles) during both optimal feeding periods and during periods in which feeding is suppressed or halted due to thermal stress and/or behavioural changes and correlate these responses to GI tract microbial communities. Furthermore, the functional role of GI tract bacteria needs to be more deeply examined to determine if metabolites and/or cellular interactions influence salmon immune or hormonal system responses. Integrated, these approaches can potentially lead to deeper understanding of how salmon GI tract bacteria interact within salmon in relation to environmental drivers and also give clues of how management strategies can be altered to maximise production during change for the long term.