K.A. Ziesemer, A.E. Mann, K. Sankaranarayanan, H. Schroeder, A.T. Ozga, B.W. Brandt, E. Zaura, A. Waters-Rist, M.L.P. Hoogland, D.C. Salazar-García, M. Aldenderfer, C. Speller, J. Hendy, D.A. Weston, S.J. MacDonald, G.H. Thomas, M.J. Collins, C.M. Lewis, C.L. Hofman, and C. Warinner. 2015. “Intrinsic challenges in ancient microbiome reconstruction using 16S rRNA gene amplification”. Nature (scientific reports) 5: 1-19.
 

Abstract:
Dental calculus, formed by periodic mineralization of dental plaque, is an ideal biomolecular reservoir as we seek to understand the ancient oral microbiome. To date, characterization of the ancient oral microbiome, as well as the ancient gut microbiome (i.e., coprolites), has primarily been accomplished through a phylotyping approach involving targeted amplification and sequencing of variable regions in the 16S rRNA gene. Specifically, the V3 region (E coli 341-534) of this gene, with a short amplification length and good taxonomic resolution, has been identified through in silico and in vitro analyses as an excellent candidate for ancient DNA amplification and community reconstruction.  Nevertheless, in practice this phylotyping approach often results in unusual taxonomic frequency data. In this study, we use targeted (amplicon) and non-targeted (shotgun metagenomics) sequencing methods on four archaeological dental calculus samples to better understand these discrepancies. The four samples were chosen from diverse geographic and temporal contexts: Middenbeemster, Netherlands (159 BP); Guadeloupe, Caribbean (700 BP); Samdzong, Nepal (1900 BP); and Camino del Molino, Spain (4000 BP). Through comparisons of microbial taxonomic counts from paired amplicon and shotgun sequencing datasets, we show preferential amplification of archaea and the candidate bacterial phylum TM7 and underamplification of Spirochaetes and many important bacterial genera (e.g., Streptococcus) in amplicon datasets. Through informatics analysis, we demonstrate that extensive length polymorphisms in the V3 region are a consistent and major cause of amplification dropout and taxonomic bias in ancient microbiome reconstructions based on amplicon sequencing. We conclude that systematic amplification bias confounds attempts to accurately reconstruct microbiome taxonomic profiles using 16S rRNA V3 amplicon data.

Find the article on Nature Scientific reports.