Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Choice of DNA extraction method affects stool microbiome recovery and subsequent phenotypic association analyses.

Author(s): Fernández-Pato A, Sinha T, Gacesa R, Andreu-Sánchez S, Gois MFB, Gelderloos-Arends J, Jansen DBH, Kruk M, Jaeger M, Joosten LAB, Netea MG, Weersma RK, Wijmenga C, Harmsen HJM, Fu J, Zhernakova A, Kurilshikov A

Publication: Sci Rep, 2024, Vol. 14, Page 3911

PubMed ID: 38366085 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   This study compared the yield and purity of total DNA and DNA from the microbiome between specimens that underwent extraction with the AllPrep DNA/RNA Mini Kit (AllPrep) compared to the QIAamp Fast DNA Stool Mini Kit (QIAamp). The analysis was conducted using fecal specimens from two different Dutch cohorts. Potential associations between microbial composition, extraction method effects, and volunteer characteristics were also investigated. Fecal specimens were self-collected at home by 292 participants as part of the Lifelines-DEEP (LLD) follow-up (diagnosis not specified) and from 453 healthy participants from the 500 Functional Genomics cohort. Within 15 min of collection, specimens were frozen, transported on dry ice, and stored at -80°C until DNA extraction. DNA was extracted using the AllPrep DNA/RNA Mini Kit or the QIAamp Fast DNA Stool Mini Kit using a modification for the QIAcube instrument that increased the lysis temperature from 70°C to 95°C. DNA yield was quantified by spectrophotometer in all specimens from the LLD cohort and in specimens from the 500FG that underwent extraction with the QIAamp Fast DNA Stool Mini Kit.; DNA yield was quantified by the Quant-iT PicoGreen dsDNA assay in specimens from the 500FP cohort that underwent extraction with the  AllPrep DNA/RNA Kit. DNA purity was evaluated using a spectrophotometer (A260/A280).  Sequencing libraries were prepared from DNA extracted with the QIAamp Fast DNA Kit using the NEBNext DNA Library Prep Kit or from DNA extracted with the AllPrep DNA/RNA Kit with the Nextera XT DNA Library Preparation Kit off-site.

   Summary of Findings:

   The AllPrep kit yielded more concentrated DNA than the QIAamp kit in the LLD cohort (205.20 versus 64.91ng/μl, P<0.001). In the 500FG cohort, a similar trend was observed (205.20 versus 64.91 ng/μl), but a comparison wasn’t possible due to differences in quantification methods. The ratio of absorbance at 260 nm to 280 nm of the isolated DNA was higher when DNA was extracted from LLD specimens using the QIAamp than the AllPrep kit (1.99 versus 1.89, P<0.001), but there was no difference among specimens from the 500FG cohort. Extraction of DNA using the QIAamp kit rather than the AllPrep kit resulted in a significantly higher read depth from fecal specimens in the 500FG cohort (25.19 versus 11.59, P<0.001) but no difference was observed for specimens in the LLD cohort (12.91 versus 13.33).

   The average number of microbial species that were identified per sample was higher when extraction was with the AllPrep kit than the QIAamp kit in specimens from the LLD cohort (197.78 versus 181.64, P<0.001), but the opposite was observed in specimens from the 500FG cohort (172.77 versus 177.80, P< 0.001). The Alpha diversity indices (Shannon and Inverse Simpson) were higher in both cohorts when DNA extraction was with the AllPrep than the QIAamp kit (P<0.001, all). While species richness was correlated with read depth in the 500FG cohort, neither of the alpha diversity indices were correlated with read depth. Species richness, Shannon indices, and Inverse Simpson diversity indices were each significantly correlated between specimens extracted using the AllPrep and QIAamp kits in the 500 FG cohort (R=0.85, P<2.2e^-16, R=0.68, P<2.2e^-16, and R=0.58, P<2.2e^-16, respectively). In the LLD cohort, only species richness and Shannon indices were significantly correlated between specimens extracted using the AllPrep and QIAamp kits (R=0.61, P<2.2e^-16 and R=0.24, P<2.5e^-5, respectively). In both cohorts, PCoA based on the Aitchison distances clustered specimens based on DNA extraction method. The effects of patient sex, age, and body mass index on Atchinson distances were much smaller than that of extraction method. PERMANOVA analysis identified a significant effect of DNA isolation method on the microbial community composition at the species-level. At the phylum-level, there were significant differences between the two extraction methods in the relative abundances of most of the phyla identified. More genera and species were identified in >5% of LLD specimens when extraction was with the AllPrep than the QIAamp kit (331 versus 301 genera and 596 versus 529 species) but the opposite was observed in specimens from the 500FG cohort (286 versus 297 genera and 508 versus 521 species). While 464 and 494 species were differentially abundant between extraction methods in LLD and 500FG, the relative abundance of species were generally correlated between extraction methods (P<0.001). The most prevalent taxa also differed between extraction methods, with more being found in 90% of specimens when extraction was with the AllPrep than the QIAamp kit (32 versus 25 in LLD, and 28 versus 25 in 500FG). DNA extracted from a mock community using the AllPrep kit better reflected the known microbial composition than when extraction was with the QIAamp kit. Correlations between volunteer characteristics (sex, age, BMI, smoking, allergies, pets, sport, and consumption of fruit, vegetables, sugary drinks, milk, caffeinated drinks, or alcohol) and the relative abundance of bacterial species, alpha (Shannon Index) or beta diversity were found but were dependent on the extraction method used and the cohort. Importantly, only 53.5-75.3% of associations with patient characteristics were significant across methods/cohorts.

    

   Biospecimens

   • Bodily Fluid - Feces

   Preservative Types

   • Frozen

   Diagnoses:

   • Normal
   • Not specified

   Platform:

   Analyte	Technology Platform
   DNA	Spectrophotometry
   DNA	Next generation sequencing
   DNA	Fluorometry

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Preaquisition	Patient gender	Female Male
   Analyte Extraction and Purification	Analyte isolation method	AllPrep DNA/RNA Mini Kit QIAamp Fast DNA Stool Mini Kit
   Preaquisition	Patient diet	Dietary factors including consumption of fruit, vegetables, sugary drinks, milk, caffeinated drinks, or alcohol investigated
   Preaquisition	Patient age	Mean 51.7 years (range not specified) Mean 28.5 years (range not specified)
   Preaquisition	Patient body mass index	Range Not specified

   View more