More Than Meets the Eye: Cigarette Smoke Induces Genomic Changes in the Small Airway Epithelium Independent of Histologic Changes

Cigarette smoke–induced lung diseases, including lung cancer and chronic obstructive pulmonary disease (COPD), are leading causes of morbidity and mortality. The airway “field of injury” hypothesis suggests that exposure to a disease or environmental insult, such as cigarette smoke, leads to molecular alterations throughout the whole respiratory system, and that these alterations occur even in the absence of histologic changes. This concept, well developed in the cancer literature, suggests exposure-associated molecular alterations can be measured in histologically normal airway epithelium by gene expression profiling (1). These genomic signatures can then be used both to gain insights into disease mechanisms and to generate biomarkers for disease onset, progression, prognosis, and treatment.

In COPD, the earliest pathological changes appear to occur in the small airways (2–4). Cigarette smoke induces squamous cell metaplasia and mucous cell hyperplasia in the small airway epithelium (SAE) (5, 6). Further, there is evidence of decreased SAE repair (7), suggesting a detrimental effect of cigarette smoke on basal cells (BCs), the airway stem or progenitor cells (8). Although cigarette smoke–induced, SAE-specific molecular alterations have been identified (9–11), whether these molecular alterations precede these early pathologic changes is less well studied. The progression of this early injury to the heterogeneous pathologic changes in COPD, including emphysema and bronchitis, is also poorly understood, especially in former smokers.

In this issue of the Journal, Yang and colleagues (pp. 340–352) advance our understanding of the cigarette smoke–induced airway field of injury (12). They focus on molecular alterations induced in the SAE compared with the larger bronchi, leveraging the group’s small airway brushing collection technique. By comparing global gene expression profiles of the large and small airway epithelium from healthy control patients, they developed proximal and distal airway transcriptome signatures (P- or D-signatures). Using immunohistochemistry, the authors established that the genomic differences between regions was not simply a result of distinct compositions of known cell types by demonstrating that certain proximal gene expression markers are expressed by ciliated cells, a cell type also abundant in the distal airways in which these genes have lower expression. They next compared the SAE gene expression of smokers with and without COPD with that of nonsmokers. Smokers exhibited a down-regulation of ∼50% of D-signature genes compared with nonsmokers, whereas P-signature genes were up-regulated. These smoking-induced SAE molecular alterations were termed “distal-to-proximal repatterning.” The study further shows that the degree of proximalization was associated with lung function (FEV₁/FVC ratio) and age in healthy smokers, suggesting these genomic lesions have functionally measurable consequences.

As pathway analysis revealed EGFR as a major upstream regulator of the P-signature genes, the authors demonstrated evidence for its relevance in vitro by culturing primary human BCs at an air–liquid interface. Treatment of proximal airway BC cultures with an EGFR inhibitor decreased the expression of P-signature genes and increased D-signature genes. SAE BC cultures exhibited opposite changes when treated with EGF. EGF was further found to be up-regulated in the SAE of smokers, a finding reproduced by exposing cultures to cigarette smoke extract.

The changes induced in vitro by cigarette smoke extract support the concept that SAE proximalization represents early injury from smoking. The observation on immunohistochemistry that one P-signature gene, UPK1B, is induced in SAE ciliated cells from smokers, in areas free from squamous metaplasia and mucous cell hyperplasia, further suggests SAE proximalization precedes the development of histologic lesions.

Taken together, Yang and colleagues report an extensive SAE transcriptome analysis, which identifies proximalization of distal airways as at least partially a result of cigarette smoke–induced EGF signaling. However, as a result of the study design, it is difficult to determine which of the observed effects are specifically associated with COPD pathogenesis. Although they did use SAE samples from current smokers with and without COPD, samples obtained from participants with COPD had higher smoke exposure (significantly higher pack-years and slightly higher nicotine and cotinine levels, although of unclear statistical significance). The COPD participants were also older, which was important, as age was associated with increased proximalization of the SAE in healthy smokers. Any inferences made specifically about COPD are thus confounded by smoke exposure and age. Furthermore, no former smokers or never smokers with COPD were included in the analysis. Additional studies are necessary to determine which, if any, of these proximalization-associated alterations are reversed by smoking cessation. Nonetheless, this study provides a deeper understanding of how cigarette smoke exposure regionally affects the airway field of injury, which is important for all smoking-related lung diseases.

The underlying mechanism of this SAE proximalization remains unknown. Nicotine, the most well-characterized cigarette ingredient, may alter airway BC proliferation and differentiation (13). However, the detrimental effects of the other toxic components of cigarette smoke should not be neglected. Noncoding RNA (e.g., microRNA) regulation, as well as epigenetic and microbiome alterations, may all influence exposure-related gene expression alterations. Authors from the current study previously identified SAE alterations in microRNA expression between healthy smokers and nonsmokers (14). DNA methylation has also been reported to be altered in SAE of COPD, which was linked to altered gene expression (15). A recent study reported a decreased diversity of microbiota derived from lower airways in COPD (16), and although unique smoking-induced changes in the distal airway microbiome have not been identified, one might speculate that such changes occur secondary to smoke exposure or an altered SAE in the development of COPD.

Individual variability in genetics, epigenetics, and the microbiome, in addition to the intensity of smoking and exposure to airborne pollutants, may also explain the striking heterogeneity observed in the degree of smoking-dependent proximalization of the SAE. An understanding of this heterogeneity has the potential to lead to the development of biomarkers for COPD risk and precision therapies.

Between proximalization of the SAE and translation to the clinic, there remain important gaps to fill. Although there is indirect evidence that repatterning is an early smoking-dependent lesion occurring before the traditionally observed histologic changes, proximalization might also or instead be a separate type of injury in a subgroup of smokers. Longitudinally collected airway brushings and samples from healthy and diseased areas of lung are required to establish repatterning as a precursor lesion. Employing matched samples from the distal and proximal compartments may also be crucial (as opposed to each sample from a distinct participant as in this study) to control for the heterogeneity identified across participants. If SAE proximalization is a precursor to COPD, the implications for treatment will be predicated in part on knowing whether the early changes reverse on smoking cessation. Finally, if proximalization is indeed an early form of smoking-induced injury that can progress to COPD, therapeutic strategies will likely need to target molecules downstream of EGFR or in the other pathways enriched in the proximal airway signature (e.g., oxidative stress and extracellular matrix-associated pathways), as EGFR inhibition in COPD was not successful in one randomized controlled trial (17).