Differential Effect of Active Smoking on Gene Expression in Male and Female Smokers

Abstract

Smoking is the second leading cause of preventable death in the United States. Cohort epidemiological studies have demonstrated that women are more vulnerable to cigarette-smoking induced diseases than their male counterparts, however, the molecular basis of these differences has remained unknown. In this study, we explored if there were differences in the gene expression patterns between male and female smokers, and how these patterns might reflect different sex-specific responses to the stress of smoking. Using whole genome microarray gene expression profiling, we found that a substantial number of oxidant related genes were expressed in both male and female smokers, however, smoking-responsive genes did indeed differ greatly between male and female smokers. Gene set enrichment analysis (GSEA) against reference oncogenic signature gene sets identified a large number of oncogenic pathway gene-sets that were significantly altered in female smokers compared to male smokers. In addition, functional annotation with Ingenuity Pathway Analysis (IPA) identified smoking-correlated genes associated with biological functions in male and female smokers that are directly relevant to well-known smoking related pathologies. However, these relevant biological functions were strikingly overrepresented in female smokers compared to male smokers. IPA network analysis with the functional categories of immune and inflammatory response gene products suggested potential interactions between smoking response and female hormones. Our results demonstrate a striking dichotomy between male and female gene expression responses to smoking. This is the first genome-wide expression study to compare the sex-specific impacts of smoking at a molecular level and suggests a novel potential connection between sex hormone signaling and smoking-induced diseases in female smokers.

Keywords: Gene expression analyses; Microarray; Smoking and immune response; Smoking and sex; Smoking carcinogenesis.

Figure 1

Multidimensional scaling plot summarizing gene expression differences between smokers and non-smokers. Each point represents an individual sample, and the distance between two points reflects the overall similarity in expression of the selected set of genes in those two samples. A) Separation of smokers and non-smokers according to 300 genes differentially expressed between all smokers and non-smokers, irrespective of their sex. The points are colored as red (male non-smoker), pink (female non-smoker), green (male smoker) and cyan (female smoker). B) Separation of male smokers (red) and non-smokers (green) using a set of 175 genes differentially expressed in male smokers and non-smokers. C) Separation of female smokers (red) and non-smokers (green) using a set of 247 genes differentially expressed in female smokers and non-smokers.

Figure 2

Hierarchical clustering of smoking correlated genes corresponding to IPA functional categories in the context of pathways and networks, biological function and/or diseases (presented in Table 2). A supervised average linkage clustering of 47 genes in male (A) and 111 genes in female (B) smokers is presented. The annotation of all genes in clustered order is presented in SI Table 6.

Figure 3

Gene product interaction network of immune and inflammatory responsive genes associated with smoking in females generated from the information in the Ingenuity knowledge base, version 7.6. Genes or gene products are represented as nodes, and the biological relationship between two nodes is represented as an edge (line). Solid lines represent direct relationship and dashed line represents indirect relationship between nodes. The intensity of node color indicates the degree of up- (red) or down- (green) regulation in smokers. Grey nodes represent molecules not altered in smokers added by Ingenuity to show network connections. The yellow node is a cigarette smoke toxicant manually added to the network. The shape of each node indicates the gene product’s functional class as shown in the key. The genes involved in immune/inflammatory responses are highlighted in yellow, and those involved both in immune/inflammatory responses and the NK cell signaling pathway are highlighted in purple. Immune/inflammatory genes known to be influenced by female hormones (progesterone and estrogen) are linked with yellow edges.

Figure 4

Enrichment of the over-expressing oncogenic form of KRAS, most commonly mutated oncogenes in lung cancer, in male non-smokers vs smokers (A) and female non-smokers vs smokers (B). In response to smoking, 21 genes associated with KRAS oncogene were overexpressing in male smokers and that of 41 genes in female smokers. C) heat map showing the top 50 genes/features representing oncogenic signature gene-sets over-represented in male and female smokers (red=up-regulated, white=average expression, blue=down-regulated). Natural killer cell-mediated cytotoxicity genes down-regulated in female smokers are marked with asterisks along the right edge.