Abstract
Exopolysaccharides produced by bacterial species and present in feces are extremely inhibitory to DNA restriction and can cause discrepancies in metagenomic studies. We determined the effects of different DNA extraction methods on the apparent composition of the gut microbiota using Illumina MiSeq deep sequencing technology. DNA was extracted from the stool from an obese female using 10 different methods and the choice of DNA extraction method affected the proportional abundance at the phylum level, species richness (Chao index, 227 to 2,714) and diversity (non parametric Shannon, 1.37 to 4.4). Moreover DNA was extracted from stools obtained from 83 different individuals by the fastest extraction assay and by an extraction assay that degradated exopolysaccharides. The fastest extraction method was able to detect 68% to 100% genera and 42% to 95% species whereas the glycan degradation extraction method was able to detect 56% to 93% genera and 25% to 87% species. To allow a good liberation of DNA from exopolysaccharides commonly presented in stools, we recommend the mechanical lysis of stools plus glycan degradation, used here for the first time. Caution must be taken in the interpretation of current metagenomic studies, as the efficiency of DNA extraction varies widely among stool samples.
Highlights
Extracted if a sample has been frozen[16]
PCR inhibitors are common in stool samples, little attention has been paid to the potential biases introduced by exopolysaccharides produced by the microorganisms of the gut microbiota[13]
None of the different DNA extraction methods that were used resulted in 100% comparable bacterial community compositions
Summary
16S rRNA gene sequencing of the faecal microbial community of an obese subject using 10 different DNA extraction protocols. With DNA extraction method 1, we detected more phyla than with extraction method 5 for 30 stools (36%) and less phyla for 39 stools (47%); for 16 stools (19%), both assays resulted in an equal number of plyla (Supplementary Fig. 6). With DNA extraction method 1, we detected more genera than with extraction method 5 for 28 stools (35%) and less genera for 33 stools (41%); for 20 stools (25%), both assays resulted in an equal number of genera (Supplementary Table 4 and Supplementary Fig. 7). The microbial richness estimated by the Chao[1] index and biodiversity, as assessed using a nonparametric Shannon index, revealed large differences between the two DNA extraction methods for the stools samples tested (Supplementary Table 4). Extraction method 1 was able to detect 68% to 100% genera and 42% to 95% species whereas extraction method 5 was able to detect 56% to 93% genera and 25% to 87% species
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
