Abstract
<title>Abstract</title> <p> Background Smoking is the single largest risk factor for ill-health and premature death globally. While direct toxic effects on the respiratory epithelium are well understood, smoking also changes the microbial environment, resulting in an increased risk of infections. Previous analyses of the nasal microbiome suggested an increased abundance of anaerobic, potentially pathogenic taxa, but were limited by small participant numbers and unclear comparison groups. In this study we use data from the German National Cohort (NAKO) to investigate the association between smoking and the anterior nares’ microbiome in a general population sample. Methods We analysed participants’ nasal swabs from 1,948 participants (16.8% current and 35.6% former smokers) at the NAKO baseline examination with 16S rRNA sequencing of the V3-V4 hypervariable regions. The USEARCH pipeline was applied for taxonomic profiling. Linear regression was performed for the association between smoking habits and species richness and evenness. PERMANOVA was used to assess the explained microbiome variance due to smoking behaviours. The pairwise Wilcoxon rank sum test, MaAsLin3 and a Random Forest model were employed to identify differentially abundant taxa. Latent Dirichlet allocation was utilised to identify groups of taxa potentially sharing biologically similar functions and their topic distribution was compared between smoking statuses. Results The most prevalent genera were <italic>Corynebacterium</italic> (100%), <italic>Staphylococcus</italic> (100%), and <italic>Cutibacterium</italic> (99.9%). Current smokers exhibited a higher diversity compared to never smokers (Shannon index: Cohen’s <italic>d</italic> = 0.20, 95% CI [0.07–0.32]; Simpson index: Cohen’s <italic>d</italic> = 0.23, 95% CI [0.11–0.36]). Differential abundance analysis identified a higher relative abundance of anaerobic pathobionts in smokers (current and former), most strongly for the taxa <italic>Anaerococcus, Finegoldia, and Peptoniphilus (a)</italic> . Latent Dirichlet allocation identified six topics whose distributions varied between smoking status and corresponded to the results from the differential abundance analysis. Former smokers showed patterns consistent with partial normalization. Conclusions We found in our population-based study on the association between smoking and the anterior nares’ microbiome higher microbial diversity and anaerobic abundance in smokers, compatible with a higher susceptibility for respiratory infections. Our results can inform the underlying mechanisms of why smokers are more susceptible to infection. </p>