Robust host source tracking building on the divergent and non-stochastic assembly of gut microbiome in wild and farmed large yellow croaker
Abstract BackgroundRevealing the potential divergence of gut microbiome between farmed and wild fishes, and its underlying mechanism are informative to improve its mariculture, as well as establish the molecular marker of host source tracking, which is an alternative to the yet-to-be-established host genetic marker. A candidate for testing the feasibility is the large yellow croaker, Larimichthys crocea , which is carnivorous and ranking the top maricultural fish in China with depleted wild resource and frequently farmed individuals escaping and fry releasing for wild stock enhancement. ResultsThe rectums of wild (n=212) and farmed (n=79) individuals from multiple batches were collected for the profiling of gut bacterial communities. The farmed individuals had a higher alpha diversity and lower bacterial loading than the wild individuals. The gut microbiota of the two sources exhibited divergence and high inter-batch variation, featured by the dominance of Psychrobacter spp. in the wild group. Predicted function of gut microbiome and representative isolates suggested that diet could be a key factor for the divergence, which was linked to the high ratio and diverse source of carbohydrate in formulated feed and low pH of rectum contents in farmed fishes. The non-stochastic distribution patterns of the core gut microbiota of the wild and farmed individuals indicated the feasibility of microbiota-based host source tracking through machine learning algorithm. Random forest classifier building on the divergence and non-stochastic assembly of gut microbiome was robust in host source tracking for individuals from all batches including a newly introduced batch. ConclusionsOur study revealed the divergence of the gut microbiota between wild and farmed croakers and suggested that diet change is an underlying key factor for the divergence. As the first time, we verified that with less biased datasets and non-stochastic pattern, gut microbiota can be robustly applied to the tracking of host source even in carnivorous fish.