Christine Drengenes1,2, Tomas M L Eagan3,4, Ingvild Haaland3,4, Harald G Wiker4,5, Rune Nielsen3,4. 1. Department of Thoracic Medicine, Haukeland University Hospital, Bergen, Norway. Christine.Drengenes@gmail.com. 2. Department of Clinical Science, Faculty of Medicine, University of Bergen, Bergen, Norway. Christine.Drengenes@gmail.com. 3. Department of Thoracic Medicine, Haukeland University Hospital, Bergen, Norway. 4. Department of Clinical Science, Faculty of Medicine, University of Bergen, Bergen, Norway. 5. Department of Microbiology, Haukeland University Hospital, Bergen, Norway.
Abstract
BACKGROUND: Studies on the airway microbiome have been performed using a wide range of laboratory protocols for high-throughput sequencing of the bacterial 16S ribosomal RNA (16S rRNA) gene. We sought to determine the impact of number of polymerase chain reaction (PCR) steps (1- or 2- steps) and choice of target marker gene region (V3 V4 and V4) on the presentation of the upper and lower airway microbiome. Our analyses included lllumina MiSeq sequencing following three setups: Setup 1 (2-step PCR; V3 V4 region), Setup 2 (2-step PCR; V4 region), Setup 3 (1-step PCR; V4 region). Samples included oral wash, protected specimen brushes and protected bronchoalveolar lavage (healthy and obstructive lung disease), and negative controls. RESULTS: The number of sequences and amplicon sequence variants (ASV) decreased in order setup1 > setup2 > setup3. This trend appeared to be associated with an increased taxonomic resolution when sequencing the V3 V4 region (setup 1) and an increased number of small ASVs in setups 1 and 2. The latter was considered a result of contamination in the two-step PCR protocols as well as sequencing across multiple runs (setup 1). Although genera Streptococcus, Prevotella, Veillonella and Rothia dominated, differences in relative abundance were observed across all setups. Analyses of beta-diversity revealed that while oral wash samples (high biomass) clustered together regardless of number of PCR steps, samples from the lungs (low biomass) separated. The removal of contaminants identified using the Decontam package in R, did not resolve differences in results between sequencing setups. CONCLUSIONS: Differences in number of PCR steps will have an impact of final bacterial community descriptions, and more so for samples of low bacterial load. Our findings could not be explained by differences in contamination levels alone, and more research is needed to understand how variations in PCR-setups and reagents may be contributing to the observed protocol bias.
BACKGROUND: Studies on the airway microbiome have been performed using a wide range of laboratory protocols for high-throughput sequencing of the bacterial 16S ribosomal RNA (16S rRNA) gene. We sought to determine the impact of number of polymerase chain reaction (PCR) steps (1- or 2- steps) and choice of target marker gene region (V3 V4 and V4) on the presentation of the upper and lower airway microbiome. Our analyses included lllumina MiSeq sequencing following three setups: Setup 1 (2-step PCR; V3 V4 region), Setup 2 (2-step PCR; V4 region), Setup 3 (1-step PCR; V4 region). Samples included oral wash, protected specimen brushes and protected bronchoalveolar lavage (healthy and obstructive lung disease), and negative controls. RESULTS: The number of sequences and amplicon sequence variants (ASV) decreased in order setup1 > setup2 > setup3. This trend appeared to be associated with an increased taxonomic resolution when sequencing the V3 V4 region (setup 1) and an increased number of small ASVs in setups 1 and 2. The latter was considered a result of contamination in the two-step PCR protocols as well as sequencing across multiple runs (setup 1). Although genera Streptococcus, Prevotella, Veillonella and Rothia dominated, differences in relative abundance were observed across all setups. Analyses of beta-diversity revealed that while oral wash samples (high biomass) clustered together regardless of number of PCR steps, samples from the lungs (low biomass) separated. The removal of contaminants identified using the Decontam package in R, did not resolve differences in results between sequencing setups. CONCLUSIONS: Differences in number of PCR steps will have an impact of final bacterial community descriptions, and more so for samples of low bacterial load. Our findings could not be explained by differences in contamination levels alone, and more research is needed to understand how variations in PCR-setups and reagents may be contributing to the observed protocol bias.
Authors: Noha Youssef; Cody S Sheik; Lee R Krumholz; Fares Z Najar; Bruce A Roe; Mostafa S Elshahed Journal: Appl Environ Microbiol Date: 2009-06-26 Impact factor: 4.792
Authors: Robert P Dickson; John R Erb-Downward; Hallie C Prescott; Fernando J Martinez; Jeffrey L Curtis; Vibha N Lama; Gary B Huffnagle Journal: J Clin Microbiol Date: 2014-07-30 Impact factor: 5.948
Authors: James M Beck; Patrick D Schloss; Arvind Venkataraman; Homer Twigg; Kathleen A Jablonski; Frederic D Bushman; Thomas B Campbell; Emily S Charlson; Ronald G Collman; Kristina Crothers; Jeffrey L Curtis; Kimberly L Drews; Sonia C Flores; Andrew P Fontenot; Mary A Foulkes; Ian Frank; Elodie Ghedin; Laurence Huang; Susan V Lynch; Alison Morris; Brent E Palmer; Thomas M Schmidt; Erica Sodergren; George M Weinstock; Vincent B Young Journal: Am J Respir Crit Care Med Date: 2015-12-01 Impact factor: 21.405
Authors: Alexa A Pragman; Tianmeng Lyu; Joshua A Baller; Trevor J Gould; Rosemary F Kelly; Cavan S Reilly; Richard E Isaacson; Chris H Wendt Journal: Microbiome Date: 2018-01-09 Impact factor: 14.650
Authors: Catherine Lozupone; Adela Cota-Gomez; Brent E Palmer; Derek J Linderman; Emily S Charlson; Erica Sodergren; Makedonka Mitreva; Sahar Abubucker; John Martin; Guohui Yao; Thomas B Campbell; Sonia C Flores; Gail Ackerman; Jesse Stombaugh; Luke Ursell; James M Beck; Jeffrey L Curtis; Vincent B Young; Susan V Lynch; Laurence Huang; George M Weinstock; Kenneth S Knox; Homer Twigg; Alison Morris; Elodie Ghedin; Frederic D Bushman; Ronald G Collman; Rob Knight; Andrew P Fontenot Journal: Am J Respir Crit Care Med Date: 2013-05-15 Impact factor: 21.405
Authors: Sun Young Cho; Jeong-Hyun Choi; Seung Hyeun Lee; Yong-Sung Choi; Sung Wook Hwang; Young Jin Kim Journal: PLoS One Date: 2021-11-05 Impact factor: 3.240