Methylated Genes in Sputum Among Older Smokers With Asthma

Abstract

Objective:

The epigenetic basis for human asthma is not well studied, particularly among older adults. This study investigated the methylation profiles in sputum DNA among older adults with asthma, using a population of smokers.

Methods:

This was a cross-sectional study using the Lovelace Smokers Cohort, a population of former and current smokers aged ≥ 40 years in New Mexico. One hundred eighty-four smokers with asthma were compared with 511 smoker control subjects with a similar smoking history, after carefully excluding those with COPD. Environmental exposures were assessed by a standard questionnaire. Postbronchodilator spirometry was performed. Induced sputum was analyzed for the methylation prevalence of 12 selected asthma-related genes using nested methylation-specific polymerase chain reaction assay.

Results:

Asthma was associated with a greater number of methylated genes and, specifically, with methylated protocadherin-20 gene in sputum DNA compared with control subjects with a similar smoking history. These associations remained significant after adjustment for covariates as well as Bonferroni correction. A synergistic interaction was noted between two methylated genes (protocadherin-20 and paired box protein transcription factor-5α) in sputum DNA on the odds for asthma (P = .009). Interestingly, the epigenetic-asthma associations were not explained by the environmental factors studied. Further, methylated genes in sputum DNA, including the protocadherin-20 gene, identified a symptomatically more severe asthma phenotype in a subgroup analysis.

Conclusions:

Asthma is associated with methylation of selected genes, such as protocadherin-20 gene, in sputum DNA. If future studies establish causality, novel demethylating interventions to prevent and treat asthma among older smokers may be possible.

Asthma, an obstructive inflammatory airway disease affecting 39.9 million people in the United States,¹ may have an epigenetic basis.² Epigenetics is the study of changes in gene function through heritable changes in chromatin structure without a change in DNA sequence. Epigenetic changes (involving promoter methylation or demethylation of cytosines at cytosine-phosphate-guanine [CpG] sites in genes) regulate diverse cellular functions, including inflammation.³ Epigenetic changes are also potentially modifiable (unlike genetic mutations) and provide an exciting possibility of novel interventions to prevent and treat asthma. Epigenetic patterns in sputum cells that include leukocytes and epithelial cells from the aerodigestive tract, as opposed to those in whole blood cells, may be more relevant to studies of airway diseases. We have developed a sensitive and specific approach, the nested methylation-specific polymerase chain reaction (PCR) assay, that allows interrogation for the presence of methylated alleles of genes selectively silenced in sputum cells.^4,5

Most asthma studies involve children and young adults and exclude older adults. Adult-onset asthma currently forms the majority of asthma cases seen at and after middle age, particularly among women.⁶ Unlike pediatric-onset asthma, adult-onset asthma is more strongly associated with female sex, obesity, nonatopic state, and smoking.⁶ Although cigarette smoking is a known risk factor for adult-onset asthma, most asthma studies carefully exclude smokers for fear of overlap with COPD, another obstructive lung disease.^6‐11 Thus, asthma in older adults, particularly smokers, remains a common but woefully understudied disease. Our objective was to identify profiles of methylated genes in sputum DNA of subjects with asthma, using a population of older adults with past or current history of smoking but without COPD. If future studies establish causality, novel epigenetic interventions to prevent and treat asthma in older adults may be possible.

Materials and Methods

Study Population

In this cross-sectional study, subjects were drawn from 1,861 eligible participants enrolled between 2001 and 2007 in the Lovelace Smokers Cohort, a well-characterized cohort in New Mexico.^12,13 This large cohort disproportionately enrolled women who had ever smoked to study the susceptibility to the development of obstructive lung disease, since women are underrepresented in most studies of airflow obstruction in the United States.¹⁴ The catchment area was Albuquerque, New Mexico, and its surrounding communities, comprising a diverse population of approximately 700,000 persons living at altitudes of approximately 1,500 m above sea level. Participants were recruited from the community through newspaper or television advertisements and were paid a small stipend for their participation. This study was approved by the Western Institutional Review Board (Olympia, Washington; WIRB Protocol #20031684).

Inclusion Criteria

Subjects were included if they were 40 to 75 years of age, smokers (former or current) with ≥ 10 pack-years of smoking history, and able to understand English. In addition, subjects were required to undergo spirometric testing and sputum induction in the same time frame as obtaining the medical history. Of the 1,861 subjects enrolled in the original cohort, we included 184 smokers with asthma and 511 smoker control subjects in this study.

Asthma was defined by a self-reported provider diagnosis of ever having asthma and/or spirometric reversibility. Subjects with asthma who also met the definition of COPD (see later in this section) were excluded. Reversibility was defined by an increase in absolute value by ≥ 12% and 200 mL compared with baseline in FEV₁ and/or FVC during a single testing session with bronchodilator administration.¹⁵ COPD was defined by a self-reported provider diagnosis of having COPD and/or presence of irreversible spirometric airflow obstruction (ie, lack of reversibility in a setting of postbronchodilator FEV₁/FVC value of ≤ 70%). The control subjects were smokers, both current and former, as defined by the absence of all of the following criteria: (1) any self-reported lung disease, (2) spirometric airflow obstruction (defined by a prebronchodilator FEV₁/FVC ratio above both the Hankinson-defined lower limit of normal¹⁵ and GOLD [Global Initiative for Chronic Obstructive Lung Disease]-defined value of 70%¹⁶), and (3) bronchodilator reversibility (as defined here).

Exclusion Criteria

We carefully excluded those with COPD (168), asthma-COPD overlap state (161), chronic bronchitis and other lung diseases (119), missing data (35), minimal smoking history (108), and test samples of inadequate quality (572). For this purpose, asthma-COPD overlap state was defined by those subjects who met the definitions described here of both asthma and COPD. Chronic bronchitis was defined by a self-report of a persistent cough and phlegm production for at least 3 months per year in 2 consecutive years. “Other lung diseases” were defined by self-report of provider-diagnosed lung disease that did not meet the discrete definitions described here of asthma, COPD, chronic bronchitis, and asthma-COPD overlap state.

Study Measurements

Outcome measures and assessment of exposure variables were obtained at the same visit for all participants. All tests were conducted at Lovelace Scientific Resources, Albuquerque, New Mexico. Information related to demographics, respiratory diseases, quality of life, and smoking was obtained by self-report from all study participants via a questionnaire, based on the validated American Thoracic Society (ATS) 1978 Adult Questionnaire and the St. George Respiratory Questionnaire (SGRQ).^17,18 BMI was measured using standardized methods.¹⁹

Prebronchodilator and postbronchodilator spirometry were obtained on all subjects by registered respiratory therapists, strictly adhering to the 1994 ATS guidelines.²⁰ After completion of prebronchodilator spirometry, all subjects were given two puffs of albuterol (90 μg/spray metered dose inhaler) with a LiteAire dual valve spacer (Thayer Medical Corporation), and spirometry was repeated after 15 min.²⁰ Participants were requested not to take any inhalers for 4 h prior to their appointment. Vmax Encore 22 (Viasys Respiratory Care) and KoKo (Ferraris Respiratory) spirometers were used. Both machines met the 1994 ATS recommendations and were calibrated daily and checked at three different injection speeds, as per the ATS guidelines.²⁰ Additionally, respiratory therapists were monitored and periodically recredentialed as part of a standardized laboratory proficiency testing plan. Only spirometric tests that met the ATS criteria were included in the analyses.

Induced sputum was collected and stored in Saccomanno fixative. Three slides were made for each sputum sample to check for adequacy, as defined by the presence of lung macrophages or Curschmann spirals.²¹ Adequate sputum samples from each study subject were taken for DNA isolation by protease digestion followed by phenol chloroform extraction and ethanol precipitation. Nested methylation-specific PCR assays were used to detect methylated alleles in DNA recovered from the sputum samples, as previously described.⁵ We studied the promoter methylation of a panel of 12 genes selected due to their known involvement with oxidative damage, DNA repair pathways, and cell fate determination of epithelial or hematopoietic cells that are closely associated with chronic inflammatory conditions, as found in patients with asthma.^22‐24 These genes included PCDH20 (protocadherin 20 gene); SULF2 (extracellular sulfatase 2 gene); GATA4 and GATA5 (GATA binding protein-4 and -5 transcription factor genes); PAX5α and PAX5β (paired box protein 5 transcription factor genes); p16; MGMT (O⁶-methylguanine-DNA methyltransferase gene); RASSF1α (Ras association domain family member 1α gene), DAPK (death-associated protein kinase gene); DAL1 (differentially expressed in adenocarcinoma of the lung 1 gene); and JPH3 (junctophilin 3 gene). The nested methylation-specific PCR assay, as published by Belinsky et al,^4,25 designs stage II methylation-specific primer pairs for each gene to include three to five CpGs in each primer, thus, six to 10 CpGs for each gene. Stringent annealing temperatures are used in the PCR reaction such that amplification of the allele only occurs if the majority of the CpGs under the primer are methylated. The specificity of this approach has been confirmed by the bisulfite sequencing of the PCR products, and those results show that the CpG sites between the primers that comprise the amplified sequence are indeed methylated.^4,25 The methylation-specific PCR assays were thus optimized to be highly specific for the region amplified. As an additional quality control measure, a subset of sputum samples that gave positive methylation products were reanalyzed with 100% validation by a second method using restriction enzyme digestion that can discriminate methylation status of CpGs within the resulting PCR product.^4,25

Statistical Analysis

Summary statistics, including means, SDs, medians, and interquartile ranges for continuous variables and proportions for categorical variables, were obtained. The χ² test was used for analysis of categorical variables and the t test for continuous variables. Multivariable binary logistic regression models were used for binary outcomes. Covariates considered in the adjusted models included sex, age, heavy smoking, current smoking, obesity, educational attainment (high school or better), wood smoke exposure, and Hispanic ethnicity. Heavy smoking was defined as ≥ 40 pack-years (based on the observed mean cut point of 39.4 pack-years in the cohort). All analyses were conducted in SAS 9.2. A two-sided P value of < .05 was generally considered statistically significant. However, to adjust for multiple comparisons involving individual genes, a Bonferroni correction for the 12 genes was applied using a P value of < .0042 to determine statistical significance.

Results

Smokers with asthma were more likely to have higher SGRQ scores (ie, greater respiratory symptoms, activity difficulties, and impact on daily life); greater prevalence of postbronchodilator spirometric obstruction²⁶; and lower absolute postbronchodilator FEV₁/FVC ratio, as compared with smoker control subjects (Table 1). In addition, smokers with asthma were more likely to be obese than smoker control subjects, in both unadjusted and adjusted analyses (Tables 1, 2, respectively). Of particular note, the asthma and control groups were not different with respect to their smoking history.

Table 1.

—Unadjusted Characteristics of Smokers With Asthma Compared With Smoker Control Subjects

Characteristic	Asthma (n = 184)	Smoker Control Subjects (n = 511)	P Value
Male sex	26.6	23.7	.43
Age, y	53.3 ± 9.2	54.5 ± 9.1	.15
Hispanic ethnicity	23.4	22.3	.77
≥ High school education	69.6	69.1	.90
Obese	41.3	29.4	.003
BMI, kg/m2	29.6 ± 6.4	28.3 ± 5.9	.01
Pack-y of cigarettes	33.6 ± 16.6	35.0 ± 17.2	.30
Current cigarette smoke exposure	60.3	59.9	.92
Wood smoke exposure	27.2	22.3	.18
SGRQ total score	21.2 ± 16.5	16.8 ± 14.3	.001
SGRQ symptom score	31.1 ± 21.5	25.7 ± 20.6	.002
SGRQ impact score	11.6 ± 13.6	7.9 ± 10.6	< .001
SGRQ activity score	30.0 ± 23.8	26.1 ± 22.3	.046
FEV1/FVC < LLN per NHANES III	1.1	0.0	.02
Absolute FEV1/FVC ratio	75.3 ± 5.5	77.0 ± 5.1	< .001
% predicted FEV1, NHANES III	93.9 ± 13.6	94.3 ± 13.2	.74
Rapid % FEV1 decliners	32.4	28.1	.33

Data are presented as % or mean ± SD. Spirometric tests were obtained postbronchodilator; obese status was defined by BMI ≥ 30 kg/m²; LLN uses the NHANES III reference standards.²⁶ Rapid decliners were defined by the highest tertile of percent annual decline in FEV₁ obtained from the difference between the baseline and final follow-up visits among 142 and 413 subjects with asthma and control subjects, respectively. LLN = lower limit of normal; NHANES = Third National Health and Nutrition Examination Survey; SGRQ = St. George Respiratory Questionnaire.

Table 2.

—Multivariable Analysis of the Association Between Selected Environmental Characteristics and Asthma Compared With Smoker Control Subjects

	Asthma (n = 184)
Exposure Variable	OR (95% CI)	P Value
Male sex	1.18 (0.80-1.76)	.41
Age, y	0.99 (0.97-1.01)	.27
Hispanic ethnicity	0.98 (0.64-1.52)	.94
≥ High school education	1.04 (0.70-1.53)	.86
Obese (BMI ≥ 30 kg/m2)	1.71 (1.20-2.44)	.003a
Current cigarette smoke exposure	0.97 (0.67-1.42)	.89
Heavy cigarette smoke exposure	0.86 (0.55-1.33)	.49
Wood smoke exposure	1.22 (0.82-1.82)	.32

Heavy cigarette smoke exposure is defined by pack-y of cigarettes ≥ 40; for age, the OR represents a 7% increase in the odds of COPD per 1 year of age. Of note, the obesity association remained significant even after additional simultaneous adjustment for the methylation status of the four genes PCDH20 (protocadherin 20 gene), SULF2 (extracellular sulfatase 2 gene), GATA4 (GATA binding protein-4 transcription factor gene), and PAX5α (paired box 5 protein transcription factor gene).

^aRepresents significant associations.

As compared with smoker control subjects, sputum DNA from smokers with asthma showed a greater mean number of methylated genes as well as a greater prevalence for at least four (of 12) genes methylated (Table 3). Examination of individual genes showed that the prevalence for methylation of PCDH20 was associated with asthma (Table 3, unadjusted analyses). The former association remained significant even after Bonferroni correction for multiple comparisons for 12 genes. In multivariable analyses, the association of asthma with either methylation of ≥ 4 genes or of individual PCDH20 remained significant and unchanged even after adjustment for environmental factors as covariates (Table 4).

Table 3.

—Unadjusted Prevalence for Gene Methylation in Sputum DNA Among Smokers With Asthma Compared With Smoker Control Subjects

Gene Methylation	Asthma (n = 184)	Smoker Control Subjects (n = 511)	P Value
Total number of methylated genes	2.6 ± 2.2	2.2 ± 2.0	.02
Methylation index ≥ 4 genes	31.5	23.1	.02
PCDH20 methylation	46.2	34.1	.0035
SULF2 methylation	38.0	29.9	.04
GATA4 methylation	41.8	34.6	.08
PAX5α methylation	13.6	12.1	.61
p16 methylation	14.7	16.4	.58
MGMT methylation	27.2	24.1	.40
RASSF1α methylation	1.6	0.2	.03
DAPK methylation	19.0	15.1	.21
GATA5 methylation	19.0	17.2	.58
PAX5β methylation	8.7	7.4	.58
DAL1 methylation	9.2	7.8	.56
JPH3 methylation	25.0	24.5	.88

Data are presented as prevalence (%) or mean ± SD. Bonferroni correction for multiple comparisons involving 12 genes requires a P value < .0042 for analysis of individual genes. DAL1 = differentially expressed in adenocarcinoma of the lung 1 gene; DAPK = death-associated protein kinase gene; GATA5 = GATA binding protein-5 transcription factor gene; JPH3 = junctophilin 3 gene; MGMT = O⁶-methylguanine-DNA methyltransferase gene; PAX5β = paired box 5 protein transcription factor gene; RASSF1α = Ras association domain family member 1α gene. See Table 2 legend for expansion of other abbreviations.

Table 4.

—Multivariable Analysis of the Association Between Gene Methylation in Sputum DNA and Asthma Compared With Smoker Control Subjects

	Asthma (n = 184)
	Unadjusted		Adjusted
Exposure Variable	OR (95% CI)	P Value	OR (95% CI)	P Value
Methylation index ≥ 4 genes	1.53 (1.06-2.23)	.03	1.56 (1.06-2.28)	.02
PCDH20 methylation	1.66 (1.18-2.34)	.0035	1.71 (1.20-2.44)	.003
SULF2 methylation	1.44 (1.01-2.04)	.04	1.46 (1.01-2.09)	.04
GATA4 methylation	1.36 (0.96-1.92)	.08	1.34 (0.94-1.91)	.10
PAX5α methylation	1.14 (0.69-1.87)	.61	1.14 (0.68-1.90)	.62

The variables adjusted included sex, age, heavy smoking, current smoking, obesity, educational attainment (high school or better), wood smoke exposure, and Hispanic ethnicity. Four of a panel of 12 genes are depicted in this table. Bonferroni correction for multiple comparisons involving 12 genes requires a P value < .0042 for analysis of individual genes. See Table 2 legend for expansion of abbreviations.

Smokers with asthma with methylated PCDH20 were indistinguishable from those without methylation in terms of their lung function but showed significantly greater concomitant methylation of most other genes in their sputum, particularly SULF2 (e-Table 1 ^{(379.3KB, pdf)} ). Further, subjects with “symptomatically most severe” asthma when compared with subjects with “least severe” asthma (ie, those in the highest vs lowest tertiles of SGRQ symptom subscale scores) were more likely to show gene methylation in their sputum DNA, particularly of PCDH20 and SULF2 (Table 5); the latter association remained significant even after the Bonferroni correction for multiple comparisons for 12 genes.

Table 5.

—Unadjusted Comparison of Characteristics of Patients With Asthma in the Highest Tertile of SGRQ Symptom Subscale Scores, Compared With Those in the Lowest Tertile Group

Characteristic	High Tertile Symptom Subscale Score (n = 61)	Low Tertile Symptom Subscale Score (n = 60)	P Value
Male sex	27.9	18.3	.21
Age, y	49.5 ± 7.5	56.0 ± 8.7	< .001
Hispanic ethnicity	29.5	16.7	.09
≥ High school education	54.1	73.3	.03
Obese (BMI ≥ 30 kg/m2)	42.6	36.7	.50
BMI, kg/m2	30.1 ± 7.1	28.1 ± 5.6	.09
Current cigarette smoke exposure	85.3	30.0	< .001
Pack-years of cigarettes	32.6 ± 15.8	32.7 ± 13.3	.98
Wood smoke exposure	39.3	16.7	.006
SGRQ symptom score	56.2 ± 14.3	9.3 ± 6.8	< .001
Absolute FEV1/FVC ratio	75.9 ± 5.7	74.9 ± 5.6	.36
FEV1 % predicted (NHANES III)	95.5 ± 14.1	95.3 ± 13.0	.94
Total number of methylated genes	3.0 ± 2.1	2.0 ± 2.1	.003
Methylation index ≥ 4 genes	49.2	21.7	.002
PCDH20 methylation	57.4	38.3	.04
SULF2 methylation	55.7	28.3	.002
GATA4 methylation	45.9	38.3	.40
PAX5α methylation	14.8	15.0	.970

Data are presented as prevalence (%) or mean ± SD. Low tertile was defined by SGRQ symptom subscale scores < 18.6 and high tertile by score ≥ 36. Expanded table is e-Table 2 ^{(379.3KB, pdf)} . See Table 1 and 2 legends for expansion of abbreviations.

Additionally, no significant gene-environment interactions for PCDH20 were noted among smokers with asthma. On the contrary, a synergistic (multiplicative) interaction between the methylation of PCDH20 and PAX5α on the adjusted odds for asthma was observed (P = .009) (Fig 1).

Figure 1.

A significant synergistic (multiplicative) interaction between methylated PCDH-20 gene and PAX-5α gene in sputum was seen on the adjusted odds for asthma among smokers (P = .009). PAX-5α = paired box 5 protein transcription factor; PCDH-20 = protocadherin 20.

Additional analyses were performed to address whether environment-asthma associations were explained by epigenetics and, conversely, whether epigenetic-asthma associations were explained by the environment. The association between obesity and asthma did not change substantially after adjusting for the epigenetic findings in sputum (Table 2 footnote). Similarly, the epigenetic-asthma associations also did not change after adjusting for obesity and the remaining covariates (Table 4).

Alternative Analytic Strategy

Subgroup analyses among women smokers alone revealed similar epigenetic-asthma associations as for the entire group (no significant sex-epigenetic interactions were noted on asthma). Use of more stringent clinical definitions of asthma generally showed similar results, but the associated decrease in sample size among cases resulted in loss of power (data not shown). On the other hand, our use of a less stringent definition of asthma (expanded to include all obstructive lung diseases—asthma, COPD, asthma-COPD overlap state, and chronic bronchitis) resulted in similar methylation associations as described here with asthma alone (e-Table 3 ^{(379.3KB, pdf)} ). Similar methylation associations were also seen when subjects with asthma or asthma-COPD overlap state were together compared with smoker control subjects (data not shown).

Discussion

Older adult smokers with asthma demonstrate epigenetic changes in sputum DNA, specifically PCDH20 methylation, when compared with control subjects with a similar smoking history. We further demonstrate a synergistic interaction between two methylated genes (PCDH20 and PAX5α) in sputum DNA on the odds for asthma in this population. Interestingly, the epigenetic-asthma associations are not explained by the environmental factors studied, nor are the obesity-asthma associations explained by the methylated genes in sputum. Further, the methylation of genes in sputum DNA identifies a symptomatically more severe disease phenotype among subjects with asthma.

Cadherins, including protocadherins, have functions in cell adhesion and signal transduction pathways,²⁷ and protocadherin-1 gene (a related member of the cadherin family) has been previously associated with asthma in both mice and humans.^28,29 Although PCDH20 is expressed by human lung tissue,³⁰ its potential role in regulating asthma pathophysiology needs further evaluation. Interestingly, our finding that PCDH20 methylation may be associated with a symptomatically more severe asthma phenotype supports the clinical relevance of studying gene methylation status in sputum (Table 5). Nevertheless, the cross-sectional nature of our study does not allow us to establish the direction of the association (ie, whether gene methylation precedes airway disease or is a consequence of airway disease).

Our study showed obesity to be associated with asthma in older adults, thus confirming findings previously described among children and young adults. Although obesity is associated with epigenetic effects,^31,32 we did not find that it explained away the epigenetic associations with asthma. Since epigenetic modifications may constitute a memory of prior environmental exposures including those occurring in utero,³³ we may not have had sufficiently detailed information about prior exposures to explain our epigenetic associations in this study. Alternatively, epigenetic events may be triggered by the diseased phenotype itself, thus bypassing environmental triggers.³²

Interestingly, subjects with asthma with methylation of PCDH20 also showed concomitant methylation of most other genes in their sputum (e-Table 1 ^{(379.3KB, pdf)} ). Further, we demonstrate that multiple methylated genes jointly affect the odds for asthma (Fig 1). This phenomenon of multiple methylated genes in respiratory samples has not been previously reported in asthma but is similar to the field effect described with cancer.³⁴

The strengths of our study include the use of postbronchodilator spirometry to assess reversibility, strict adherence to the 1994 ATS guidelines in the performance of spirometry, state-of-the-art validated PCR assays to measure methylation, use of sputum cells to study methylation instead of blood or buccal cells (whose epigenetic patterns may be irrelevant to the airway), exclusion of those with COPD or asthma-COPD overlap state, and conservative Bonferroni correction for multiple comparisons. To our knowledge, this is the first study of sputum methylation profiles in subjects with asthma.

We also recognize several limitations to our study. Our study cohort may not be representative of all smokers in New Mexico. However, the smoking behavior in this study is consistent with that observed in representative surveys of New Mexico.³⁵ Our definitions of asthma are not confirmed by methacholine bronchoprovocation and are thus subject to misclassification bias. However, this misclassification is likely to be nondifferential between the methylated and unmethylated groups and unlikely to produce a spurious effect. Although we have differentiated smokers into smoker control subjects, asthma, COPD, chronic bronchitis and asthma-COPD overlap state, we recognize that such discrete differentiation may not always be possible in clinical practice. Our results may be generalizable primarily to the asthma phenotype that occurs in the older, overweight/obese smoker, particularly women. However, this phenotype is clinically relevant since it often represents difficult-to-manage disease.³⁶ Although we arbitrarily selected 12 biologically relevant genes based on their associations with inflammation and DNA repair pathways, it is likely that there are many other genes associated with inflammation and oxidative stress in asthma. However, ours is one of the few sentinel studies that argue for studying the potential role for methylated genes in sputum of subjects with asthma. Thus, longitudinal epidemiologic and functional genetic studies are needed in the future to establish a possible causal association between environmental exposures, airway epigenetic changes, and incident asthma.

Supplementary Material

Online Supplement

Abbreviations

ATS

American Thoracic Society

CpG

cytosine-phosphate-guanine

DAL1

differentially expressed in adenocarcinoma of the lung 1 gene

DAPK

death-associated protein kinase gene

GATA4

GATA binding protein-4 transcription factor gene

GATA5

GATA binding protein-5 transcription factor gene

JPH3

junctophilin 3 gene

MGMT

O⁶-methylguanine-DNA methyltransferase gene

PAX5α and PAX5β

paired box 5 protein transcription factor genes

PCDH20

protocadherin 20 gene

PCR

polymerase chain reaction

RASSF1α

Ras association domain family member 1α gene

SGRQ

St. George Respiratory Questionnaire

SULF2

extracellular sulfatase 2 gene