Longitudinal Effects of 1-Year Smoking Cessation on Human Bronchial Epithelial Transcriptome

Abstract

Smoking is known for its adverse health effects, which include increased risk for various pulmonary diseases, such as lung cancer and COPD.1Beane J. Sebastiani P. Liu G. Brody J.S. Lenburg M.E. Spira A. Reversible and permanent effects of tobacco smoke exposure on airway epithelial gene expression.Genome Biol. 2007; 8: R201Crossref PubMed Scopus (174) Google Scholar Approximately 80% of patients who experience COPD are smokers.2Willemse B.W. ten Hacken N.H. Rutgers B. Lesman-Leegte I.G. Postma D.S. Timens W. Effect of 1-year smoking cessation on airway inflammation in COPD and asymptomatic smokers.Eur Respir J. 2005; 26: 835-845Crossref PubMed Scopus (251) Google Scholar Smoking cessation is the most successful intervention to improve respiratory symptoms in smokers with COPD and smokers with no COPD.2Willemse B.W. ten Hacken N.H. Rutgers B. Lesman-Leegte I.G. Postma D.S. Timens W. Effect of 1-year smoking cessation on airway inflammation in COPD and asymptomatic smokers.Eur Respir J. 2005; 26: 835-845Crossref PubMed Scopus (251) Google Scholar Cross-sectional studies have been done to describe the transcriptome changes that are associated with smoking.3Spira A. Beane J. Shah V. et al.Effects of cigarette smoke on the human airway epithelial cell transcriptome.Proc Nat Acad Sci. 2004; 101: 10143-10148Crossref PubMed Scopus (487) Google Scholar A longitudinal study by Willemse et al2Willemse B.W. ten Hacken N.H. Rutgers B. Lesman-Leegte I.G. Postma D.S. Timens W. Effect of 1-year smoking cessation on airway inflammation in COPD and asymptomatic smokers.Eur Respir J. 2005; 26: 835-845Crossref PubMed Scopus (251) Google Scholar found a reduction in mast cell populations in asymptomatic smokers after 1-year of smoking cessation. In contrast, specific inflammatory cell populations remained the same or increased in smokers with COPD. The current letter is based on transcriptome-wide data in bronchial biopsies from the same study cohort. We first assessed the effects of smoking cessation on the bronchial transcriptome in asymptomatic smokers vs smokers with COPD, and no significant association was found. Thus, the present study investigated the transcriptomic changes in bronchial biopsies of before and after 1-year smoking cessation in smokers, regardless of disease status, to investigate the overall effect of smoking cessation. The initial study consisted of 63 subjects with and without COPD, of which 33 successfully quit smoking for 1 year. The 11 patients with COPD patients and five asymptomatic smokers who successfully have completed the 1-year smoking cessation program and had bronchial biopsies available with good quality RNA (before and after 1-year smoking cessation) data were used for the current analysis.2Willemse B.W. ten Hacken N.H. Rutgers B. Lesman-Leegte I.G. Postma D.S. Timens W. Effect of 1-year smoking cessation on airway inflammation in COPD and asymptomatic smokers.Eur Respir J. 2005; 26: 835-845Crossref PubMed Scopus (251) Google Scholar The Groningen University Medical Centre’s local medical ethics committee approved the study; all participants gave their written informed consent. Bronchial biopsy specimens were collected as described previously2Willemse B.W. ten Hacken N.H. Rutgers B. Lesman-Leegte I.G. Postma D.S. Timens W. Effect of 1-year smoking cessation on airway inflammation in COPD and asymptomatic smokers.Eur Respir J. 2005; 26: 835-845Crossref PubMed Scopus (251) Google Scholar; RNA was extracted with the use of the all prep kit from Qiagen according to the manufacturer’s instructions, and bulk-RNA-seq was performed with the use of Illumina NovaSeq6000 (Illumina Inc) paired-end sequencing. A differential gene expression analysis was performed with the use of EdgeR to examine the gene expression changes with smoking cessation. A Benjamini–Hochberg corrected P < .05 and fold change > 1.5 was considered statistically significant. The RNA-Seq counts were normalized with the use of the “Voom” R statistical software package (LogCPM) to visualize the heatmap. There were a total of 21,532 genes in the dataset after filtering for low reads in EdgeR was completed. We explored biologic pathways associated with genes using the g:Profiler web-based tool. Gene set variation analysis (GSVA)4Hänzelmann.S CR, Guinney J. Gene set variation analysis for microarray and RNA-Seq data. BMC Bioinformatics. Forthcoming.Google Scholar was performed in a second and independent study of current vs never smokers (NORM cohort5ClinicalTrails.gov NCT00848406.Google Scholar) to examine whether changes on smoking cessation drives the bronchial transcriptome towards never smokers with the use of RNA-seq data from bronchial biopsy specimens.6Imkamp K. Berg M. Vermeulen C.J. et al.Nasal epithelium as a proxy for bronchial epithelium for smoking-induced gene expression and expression Quantitative Trait Loci.J Allergy Clin Immunol. 2018; 142: 314-317.e315Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar This study cohort consists of 40 never smokers and 37 current smokers, with 18.75 mean packyears (SD, 14.84). This never smokers group comprised 20 male patients and 20 female patients with 104.82 mean postbronchodilator FEV1 percentage predicated (SD, 10.67 SD); the current smokers consist of 22 male patients and 15 female patients with 102.8 mean postbronchodilator FEV1 percentage predicated (SD, 9.89). GSVA formed the enrichment scores by creating an eigenvalue for a set of genes in a given sample in the context of sample population distribution. The significantly up-regulated genes (false discovery rate [FDR], < 0.05; FC, 1.5) that were derived from the differential expression analysis between before and after 1 year of smoking cessation were used to create a positive enrichment score, whereas the significantly down-regulated genes (FDR < 0.05; FC, ≤ 1.5) from the same analysis were used to generate the negative enrichment score in a current and never smokers gene expression matrix. Finally, the cellular deconvolution method was applied with the use of the support vector regression approach to investigate how smoking cessation affects cellular compositions in biopsy specimens by integrating our study transcriptional profiles with previously published bronchial biopsy single-cell RNA sequencing data7Vieira Braga F.A. Kar G. Berg M. et al.A cellular census of human lungs identifies novel cell states in health and asthma.Nat Med. 2019; 25: 1153-1163Crossref PubMed Scopus (363) Google Scholar with AutoGeneS,8Aliee H. Theis F.J. AutoGeneS: automatic gene selection using multi-objective optimization for RNA-seq deconvolution.Cell Syst. 2021; 12: 706-715.e704Abstract Full Text Full Text PDF Scopus (14) Google Scholar which will give the cell-specific signatures. This cellular deconvolution method will integrate the cell-specific gene expression signatures from the single-cell RNA sequencing with bulk RNA-sequencing data based on the cell-specific marker genes and provide a relative proportion of the cell-specific expression in the bulk RNA-sequencing data. The statistical analyses were done in R statistical software (version 3.5.3) with the use of the EdgeR package (version 3.24.3), GSVA (version 1.38.2), and CIBERSORT, respectively.4Hänzelmann.S CR, Guinney J. Gene set variation analysis for microarray and RNA-Seq data. BMC Bioinformatics. Forthcoming.Google Scholar,9Robinson M.D. McCarthy D.J. Smyth G.K. EdgeR: a bioconductor package for differential expression analysis of digital gene expression data.Bioinformatics. 2010; 26: 139-140Crossref PubMed Scopus (21051) Google Scholar,10Chen B. Khodadoust M.S. Liu C.L. Newman A.M. Alizadeh A.A. Profiling tumour-infiltrating immune cells with CIBERSORT.Methods Mol Biol. 2018; 1711: 243-259Crossref Scopus (1085) Google Scholar This CIBERSORT method is an R-based package that falls under reference-based cellular deconvolution and provides a relative estimation of cell type abundance in a heterogenous bulk RNA sequencing sample with references to cell-specific gene expression information that is obtained from single-cell RNA sequencing.” The 11 patients with COPD and five asymptomatic smokers that were included in the present analysis had a median age of 57 years (range, 46 to 63) and 49 years (range, 45 to 57), with median packyears of 36 (range, 15 to 66) and 23 (range, 17 to 32), respectively. We found that the expression of 213 genes was significantly altered after 1 year of smoking cessation. Table 1 represents the top 25 significant genes in the differential gene expression analysis; the volcano plot in Figure 1A shows the distribution of the genes with their significance. Among these differentially expressed genes, 139 genes were lower expressed, which were enriched for xenobiotic metabolism, mainly detoxification and oxidative-stress responses (ALDH3A1, ADH7, SLC7A11), Aldo-keto reductase activity (AKR1C2, AKR1C3, AKR1B10), mucin production (MUC5AC, MUC2), and Nrf2 pathway activity (NQO1, GCLC) like biologic pathways. This is in accordance with previous findings that showed that cigarette/tobacco smoke increases xenobiotic metabolism by detoxifying gas and tar phase xenobiotics in cigarette smoke.1Beane J. Sebastiani P. Liu G. Brody J.S. Lenburg M.E. Spira A. Reversible and permanent effects of tobacco smoke exposure on airway epithelial gene expression.Genome Biol. 2007; 8: R201Crossref PubMed Scopus (174) Google Scholar After smoking cessation, the most significantly lower expressed gene was ALDH3A1, which protects airway epithelial from cigarette smoke-induced DNA damage and cytotoxicity.1Beane J. Sebastiani P. Liu G. Brody J.S. Lenburg M.E. Spira A. Reversible and permanent effects of tobacco smoke exposure on airway epithelial gene expression.Genome Biol. 2007; 8: R201Crossref PubMed Scopus (174) Google Scholar This agrees with the highly expressed nature of ALDH3A1 in previous studies that were conducted in the presence of cigarette smoke as a phase I xenobiotic-metabolizing enzyme.1Beane J. Sebastiani P. Liu G. Brody J.S. Lenburg M.E. Spira A. Reversible and permanent effects of tobacco smoke exposure on airway epithelial gene expression.Genome Biol. 2007; 8: R201Crossref PubMed Scopus (174) Google Scholar Nagaraj et al11Nagaraj N.S. Beckers S. Mensah J.K. Waigel S. Vigneswaran N. Zacharias W. Cigarette smoke condensate induces cytochromes P450 and aldo-keto reductases in oral cancer cells.Toxicol Lett. 2006; 165: 182-194Crossref PubMed Scopus (128) Google Scholar previously showed that genes that belong to the Aldo-keto reductase family (AKR1C1, AKR1C3, and AKR1B10) are expressed highly in oral cancer cells after cigarette smoke exposure, encoding proteins involved in the detoxification of various toxic aldehydes and ketone compounds in cigarette smoke. Our analysis shows down-regulation in these Aldo-keto reductase family genes after smoking cessation, which indicates that there is no longer a need to express these genes in the airways once the toxic compounds are removed. The 74 genes higher on smoking cessation include genes that regulate cell-to-cell adhesion (POSTN, VIM, PCDH17), genes regulating cell differentiation (CNN3, ETS1), and apoptosis-related genes BIRC3. Previous studies found that serum POSTN protein levels increased in former smokers immediately after quitting smoking, which aligns with our findings.12Carpaij O.A. Muntinghe F.O.W. Wagenaar M.B. et al.Serum periostin does not reflect type 2-driven inflammation in COPD.Respir Res. 2018; 19: 112Crossref Scopus (7) Google Scholar The heatmap in Figure1B represents the pattern of significant differentially expressed genes among each subject. These differentially expressed genes on 1-year smoking cessation were then investigated in bulk-RNA-seq of current vs never smokers with the use of GSVA to examine whether patterns observed on smoking are reversed towards normal by smoking cessation. Those genes lower on smoking cessation were lower in never smokers than current smokers (Figure 1C). At the same time, genes higher on smoking cessation were also higher in never smokers compared with current smokers. These findings suggest that gene expression patterns 1 year after smoking cessation reverse towards the direction of never smokers (Figure 1D). The pathway analysis discovered that the top functional pathways associated with lower expressed genes after smoking cessation include metabolism of xenobiotics by cytochrome P450 (KEGG:00980), chemical carcinogenesis (KEGG:05204), ferroptosis (KEGG:04216), NRF2 pathway (WP: WP2884), Meta pathway biotransformation phase I and II (WP: WP702) and oxidative-stress (WP: WP408) (FDR, <0.05). The Nrf2 pathway plays a significant role in the presence of cigarette smoke to regulate cellular protective responses to oxidative and electrophilic stress-induced damage by increasing the expression of antioxidant genes such as NQO1.13Sidhaye V.K. Holbrook J.T. Burke A. et al.Compartmentalization of anti-oxidant and anti-inflammatory gene expression in current and former smokers with COPD.Respir Res. 2019; 20: 190Crossref Scopus (14) Google Scholar The single-cell RNA sequencing-based cellular deconvolution findings in Figure 1E show the cellular composition shifts on smoking cessation. Among the different cell types, the percentage of goblet cells was significantly lower after quitting smoking (FDR, 0.0004); although basal (FDR, 0.001), fibroblasts (FDR, 0.0013), arterial endothelial cells (FDR, 0.0002), and macrophages (FDR, 0.004) were significantly higher after 1-year smoking cessation. We performed a sub analysis solely in patients with COPD and found similar results. Cigarette smoke exposure alters airway epithelial cell composition and induces hyperplasia of the goblet cells.14Utiyama D.M. Yoshida C.T. Goto D.M. et al.The effects of smoking and smoking cessation on nasal mucociliary clearance, mucus properties and inflammation.Clinics. 2016; 71: 344-350Crossref Scopus (22) Google Scholar The mucus hypersecretion reduction may be due to the removal of stress stimulus with smoking cessation. This removal may significantly reduce the goblet cell numbers and the expression of their signature genes, such as MUC5AC.14Utiyama D.M. Yoshida C.T. Goto D.M. et al.The effects of smoking and smoking cessation on nasal mucociliary clearance, mucus properties and inflammation.Clinics. 2016; 71: 344-350Crossref Scopus (22) Google Scholar Conversely, the higher relative percentage of basal cells after smoking cessation may be due to active repair in the airways after smoke withdrawal15Crystal R.G. Airway basal cells. The “smoking gun” of chronic obstructive pulmonary disease.Am J Respir Crit Care Med. 2014; 190: 1355-1362Crossref PubMed Scopus (73) Google Scholar; however, this may also be due to the loss of goblet cell percentage that leads to the increment of other cell populations. The present study reveals the effects of 1-year smoking cessation at the transcriptomic level in airways by indicating reversible trends of oxidative stress and detoxification mechanisms. At the same time, mucus hypersecretion-associated gene expression and percentage goblet cell change that were observed in deconvolution findings further confirm that goblet cell hyperplasia and mucus production may be reversible, even to some extent in smokers with COPD with smoking cessation. Future experiments can be done to consider the reversible effects of smoking cessation to help build up COPD-related health guidelines to improve patient awareness.