Comparison of two DNA extraction methods widely used in aquatic microbial ecology

In recent years, the rapid advances of culture-independent methods and new molecular tools have revolutionized our understanding of microbial biodiversity and ecological functions. DNA extraction from microbial communities is a critical step in this process and several methods have been proposed and used, but the influence of the extraction method on the outcome and ultimately on ecological inferences from the results is not yet precisely determined. Here, we compared two of the most commonly used extraction methods in aquatic microbial ecology, and investigated whether the two methods yielded comparable results for community ecology analyses. We extracted DNA from 15 different shallow lakes with phenol:chloroform, a classical and widely used extraction method, and with the PowerSoil DNA isolation Kit, often suggested as the standard DNA extraction method, with some adaptations for aquatic environments. We found that although only 5% of all OTUs showed significant differences in pairwise comparisons (using the 15 lakes as replicates), these OTUs accounted for >35% (on average) of the relative abundance. Diversity and richness did not differ significantly between the two extraction methods, but the beta-dispersion of the communities indicated that the organic extraction yielded more homogeneous communities, while the kit extraction generated variability. Consequently, we conclude that despite the small number of OTUs with significant differences, their impact on the community composition obtained was not negligible, and therefore the results from these two extraction methods were not comparable.