DNA EXTRACTION PROTOCOLS MAY INFLUENCE BIODIVERSITY DETECTED IN THE INTESTINAL MICROBIOME: A CASE STUDY FROM WILD PRUSSIAN CARP, Carassius gibelio

In this investigation, we examined the influence of different DNA extraction protocols on results obtained for intestinal microbiota of Prussian carp (Carassius gibelio). In our work, we use two common commercial kits to extract total DNA from seven fish intestines (mucosa and chyme separately): AxyPrep Multisource Genomic DNA Miniprep Kit (Union City, CA, USA) and KitDNA sorb B (Moscow, Russia).

It was shown that the microbiota of Prussian carp using the Axygen Kit was dominated by Proteobacteria and Firmicutes while using the DNA-sorb B kit the dominants were Proteobacteria, Firmicutes and Bacteroidetes. The NMDS using Bray-Curtis similarities between the samples showed a clear grouping of the samples from intestinal mucosa and content extracted by DNA-sorb B and of the same samples extracted by AxyPrep.

Moreover, ANOSIM is supportedby these results and indicated that the composition of OTUs was significantly different among all the communities, based on each factor: type of tissue and type of extraction (Table). For type of tissue, R value was 0.266 (P = 0.001), for type of extraction R value was 0.477 (P = 0.001).

For supporting our results we ran the meta-analysis (correspondence and non-specific correlation Gamma test at p≤0.05 analyses) of 25 published articles describing the biodiversity of freshwater fish. We found that intestinal microbiota of fish depend on many different factors, including the type of DNA extraction which is supported by our experimental results.

Microbial diversity, community structure, proportions of read numbers derived from each OTU and the total number of OTU's obtained by different DNA extraction protocols could lead to a bias in results obtained. Therefore, we recommend two significant points in bacterial community studies of fish gut: 1) whenever possible divide samples for microbiota analyses into their constituent parts, mucosa and chyme (intestinal content) to enable identification of autochthonous bacteria resident in the gut from those merely passing through; and 2) combine several extraction methods, if possible, to recover some of the observable biodiversity that is lost when using only one method. Consequently, knowing these possible sources of bias exist, researchers should be conservative in conclusions about structures of bacterial community of wild and cultured populations of fish species.

This work was supported by the Russian Foundation for Basic Research, project no.16-34-00309.