Methods for the analysis of skin microbiomes: a comparison of sampling processes and 16S rRNA hypervariable regions

Abstract

AbstractBackgroundThe skin microbiome is increasingly recognised as a critical component of the innate skin immune response and is central to the pathogenesis of many inflammatory skin disorders including atopic dermatitis. Previous studies have not looked in detail at the impact of changing both sampling methodology and hypervariable region sequencing for skin microbiome analysis.ObjectivesWe set out to undertake a detailed analysis comparing microbe population diversity and resolution at the single species level by swab, tape, scrape and scrape then swab sampling assayed by hypervariable region sequencing.MethodsIn triplicate samples from the antecubital fossa were taken from healthy volunteers for 16s RNA analysis by primer amplification for hypervariable regions 1-3, and 3-4.ResultsAlpha (phylogenetic) diversity was the greatest with tape sampling and V1-3 analysis, whereas for V4, tape and swab were equivalent. Scrape sampling showed lower alpha diversity at both V1-3 and V4. All measures of beta diversity showed the scrape methodology yielded a lesser diversity than the others. Phyla composition was similar across all sampling methodologies. Minor differences in composition were noted between V1-3 and V4 sequencing, but V1-3 was optimal for identification of firmicutes (including staphylococci).ConclusionsIn the methodological planning of skin microbiome analysis skin scientists need to consider the microbes of interest to choose the optimal hypervariable region to sequence, and harmonisation of methodological approaches would be beneficial to the field. For detection of staphylococci on flexural skin, we would recommend tape sampling with analysis of V1-3.What’s already known about this topic?Previous microbiome studies have utilised many different methodologies, but a detailed comparison of their utility in skin research has not been undertakenMethodological choices in skin microbiome analysis affect both sensitivity to detect diversity and sensitivity to identify at a species levelOptimisation to detect diversity and species identification can require different approachesWhat does this study add?Our study compared a matrix the four most widely utilised skin sampling methodologies and the two principle means for hypervariable gene analysis.We show that alpha diversity is optimal with tape samplingAdditionally, in contrast to gut microbiome analyses, in the skin hypervariable region V1-3 analysis is superior to V4 when the species of interest are staphylococci.