Smoking modulates different secretory subpopulations expressing SARS-CoV-2 entry genes in the nasal and bronchial airways

Coronavirus Disease 2019 (COVID-19) is caused by severe acute respiratory syndrome coronavirus 2 SARS-CoV-2), which infects host cells with help from the Viral Entry (VE) proteins ACE2, TMPRSS2, and CTSL1-4. Proposed risk factors for viral infection, as well as the rate of disease progression, include age5,6, sex7, chronic obstructive pulmonary disease7,8, cancer9, and cigarette smoking10-13. To investigate whether the proposed risk factors increase viral infection by modulation of the VE genes, we examined gene expression profiles of 796 nasal and 1,673 bronchial samples across four lung cancer screening cohorts containing individuals without COVID-19. Smoking was the only clinical factor reproducibly associated with the expression of any VE gene across cohorts. ACE2 expression was significantly up-regulated with smoking in the bronchus but significantly down-regulated with smoking in the nose. Furthermore, expression of individual VE genes were not correlated between paired nasal and bronchial samples from the same patients. Single-cell RNA-seq of nasal brushings revealed that an ACE2 gene module was detected in a variety of nasal secretory cells with the highest expression in the C15orf48+ secretory cells, while a TMPRSS2 gene module was most highly expressed in nasal keratinizing epithelial cells. In contrast, single-cell RNA-seq of bronchial brushings revealed that ACE2 and TMPRSS2 gene modules were most enriched in MUC5AC+ bronchial goblet cells. The CTSL gene module was highly expressed in immune populations of both nasal and bronchial brushings. Deconvolution of bulk RNA-seq showed that the proportion of MUC5AC+ goblet cells was increased in current smokers in both the nose and bronchus but proportions of nasal keratinizing epithelial cells, C15orf48+ secretory cells, and immune cells were not associated with smoking status. The complex association between VE gene expression and smoking in the nasal and bronchial epithelium revealed by our results may partially explain conflicting reports on the association between smoking and SARS-CoV-2 infection.