Analysis of aberrant methylation on promoter sequences of tumor suppressor genes and total DNA in sputum samples: a promising tool for early detection of COPD and lung cancer in smokers

Chronic obstructive pulmonary disease (COPD) is a cluster of heterogenic disorders, characterized by expiratory flow limitation [1] chronic bronchitis and emphysema [2‐5]. The impact of COPD in the world population is significant; according to a study of the World Bank and the World Health Organization, COPD affects 210 million people worldwide, and if the current trend remains unchanged, by 2020 COPD will be the third cause leading to death in the world [6]. Cigarette smoke is the most commonly encountered risk factor for this disease [6], as well as in the development of a COPD co-morbidity, the lung cancer (LC) – the deadliest cancer in men and women around the world [7], [8]. Several studies have shown that smokers who develop COPD have a higher risk to develop LC than no smokers [9], [10].

Since early detection of COPD is an almost impossible task in asymptomatic subjects, and the radiologic and cytological methods employed for detection of LC lack sensitivity and accuracy in early stages of the disease [11], [12], the diagnosis is late in most cases, resulting in a low overall survival rate for LC (<15%) [11].

Therefore, the development of methods based in the analysis of new diagnostic markers, with high sensitivity and prognostic value for early detection of COPD and LC, in high-risk subjects (chronic smokers with GOLD 0), will be beneficial to improve quality of life – there is evidence that smokers diagnosed with COPD are more successful in quitting [13] – and to prevent the further development of LC, reducing in this way the mortality of both lung diseases [13], [14].

There is strong evidence towards about key role of DNA methylation and mutation events in airway epithelial cells in the early development of COPD. Some molecular alterations in DNA are produced by reactive oxidant species (ROS) found in tobacco smoke [15], [16]: e.g. the tobacco specific nitrosamine 4-(methylnitrosamine)-1-(3-pyridyl)-1-butanone (NNK) is a precursor of the alkylating agents leading to the methylation of guanosine residues in DNA [17] (Figure 1).

The inflammation in the lung airway resulting from those molecular alterations may contribute to COPD pathogenesis and may predispose to the development of LC [18], [19]. Previous work in our group, carried out in Chilean subjects by methylation-specific PCR (MSP), demonstrated a high prevalence of CDKN2A promoter methylation in squamous cell lung carcinomas (SQCs) and adenocarcinomas (AdCs) [20], [21]. Although the whole pattern of DNA methylation in bronchial epithelial cells associated with COPD is poorly known, the methylation of the afore mentioned promoters has been described in sputum, bronchial epithelium and brush biopsy of active and former smokers, subjects at high-risk for developing LC [19], [22].

On the other hand, previous studies have shown that: i) cancer patients have a higher amount of cell-free DNA in plasma, serum, sputum and other body fluids compared to healthy subjects [23], [24], and; ii) many active and former smokers have increased bronchial secretions (sputum) containing exfoliated cells (from the lower respiratory tract) [24].

Inspired in the above results we propose non-invasive methodology for early detection of COPD on smokers at risk (stage GOLD 0) to develop this disease, as well as LC, based on: i) the quantification of the amount of DNA in sputum samples, and, ii) the detection the methylation pattern of gene-promoters of TSGs critical to tumor suppression – cyclin-dependent kinase inhibitor 2A (CDKN2A, encoding p16^INK4a protein regulator of progression through the G1 phase of the cell cycle [25]), MGMT, encoding the O(6)-methylguanine-DNA methyltransferase (a key DNA-repairing enzyme for removal of methyl group from mutagenic O(6)-methylguanine [26]), and CDH1 (encoding the H-cadherin, key for intracellular adhesion [27]).

In this paper we show the results of the analysis of sputum samples from patients with COPD, LC, as well as in healthy subjects (total DNA and frequency of CDKN2A, CDH1 and MGMT promoter methylation), and how them correlate with the development of COPD and LC. Regarding the significant impact of COPD and smoking in Chile, – 18,000,000 inhabitants, 1.6 millions of COPD patients [28] and 4,800,000 smokers [29] – this paper is the first step in the development of an useful tool to detect, control and prevent the lung destruction and the development of LC in our country.

Table 1, shows a summary of demographic and clinicopathological features of patients and CTR. Nineteen of the 26 LC patients had COPD and 7 presented normal spirometry. The GOLD stages are indicated in the Table1. Twenty-three smoker and ex-smoker were diagnosed with COPD; 4 were classified as GOLD 0 and 19 were classified as GOLD (stages 1 to 4). Thirty-three CTR, presented normal spirometry. All patients were diagnosis between 2006–2007 years.

The cytological examination was carried out in samples from all patients (COPD and LC) and CTR, as indicated in Table 1. Malignant cells were present in sputum samples of 12 LC patients (46%), while 6 cases presented mild to severe atypia. Normal cells were present in sputum samples of 8 LC patients. The COPD patients (smokers and ex-smokers) had an increased inflammatory cell numbers (neutrophil and macrophages) compared with CTR group without COPD (78% vs 12%).

The goal of study was verify our hypothesis and assess the usefulness of TSGs promoter methylation and total DNA measurement in sputum samples, as biomarkers of early stages of COPD as GOLD 0. Our results showed an increase of amount of DNA in sputum samples from smokers or ex-smoker with lung diseases (COPD or LC), compared with control subjects never-smoker (CTR). However, if the correlation between total DNA in the sputum sample and occurrence of LC is not clear. Although there is evidence that patients with cancer have larger amounts of DNA in the septum compared with non-cancer patients [36]. The DNA may have several sources: airway epithelial cells, macrophages, inflammatory cells and in some cases, only a minor fraction from tumor cells [37], [38]. The patients with cancer have larger amount of DNA circulating in plasma or serum compared to healthy subjects; probably the DNA is released into circulation after cell lyses triggered by necrosis or apoptosis during tumor development [37], [38]. Our results showed a higher DNA concentration in sputum from LC and COPD patients compared with CTR group (15.95 to 20.35 vs 2.75 ng/mL). All patients with COPD and without cancer, at the moment of diagnosis were classified as GOLD 0 to 4 by spirometry. Six COPD patients developed cancer after five years from the initial COPD diagnosis at 2006–2007.We also evaluated the frequency of CDKN2A, CDH1 and MGMT promoter methylation in sputum samples of the groups under study. Few studies show the feasibility to use DNA from sputum samples, for the detection of molecular alterations [23]. In this study, we confirmed the aberrant methylation of CDKN2A and MGMT promoter regions in LC (73% and 65%, respectively) and COPD (70% and 65%, respectively) compared with CTR (9% and 6%, respectively). There was no significant difference of aberrant methylation at CDH1promoter between LC (35%), COPD (40%) patients and CTR (32%). We found a high percentage of CDKN2A and MGMT promoter methylation in COPD and LC patients, observing a non-statistically significant difference in both pathologies. The high frequency of promoter methylation had strong association with high smoke frequency present in all patients studied. Previously, several studies have shown a variable frequency of gene promoter methylation in LC: CDKN2A (range: 23 -70%), CDH1 (range: 40-50%), and MGMT (range: 16-47%) [39‐41]. Discrepancies in the rates of methylation are mainly due to the use of different specimens (fresh tumors, paraffin-embedded tissue, cell lines, bronchioalveolar lavage and sputum) containing a diverse DNA concentration or different analytical approaches [41‐43]. Recently, Belinsky et al. (2006) reported the feasibility to use the sputum to determine the promoter methylation in 14 genes [42]. However, the cohort used in this study all LC patients and control had subjects a smoking history of ≤50 packs/year (current and former). In this report, all control subjects showed aberrant promoter methylation in all the genes studied. In addition, was demonstrated that CDKN2A and/or MGMT promoter methylation can be detected in DNA obtained from sputum in 100% of patients with SQC up to 3 years before clinical diagnosis [43]. A similar frequency of CDKN2A promoter methylation was observed in our study, as indicated above. Our results, confirmed previous findings using a different study group: the methylation of CDKN2A and MGMT were more frequent in smoker patients with LC compared with never smokers [44], [45]. Nevertheless, our results about the frequencies of aberrant promoter methylation of CDKN2A and MGMT genes between smokers and never-smokers do not agree with another previously reported [45]. The explanation for the observed discrepancies among different studies is unclear. However, we can speculate that these discrepancies are due to the small number of never-smokers involved in these studies [45‐47]. In our report, only eight patients with LC were never-smokers during their lifetime. A key element of our analysis is the meaning of “never-smoker”. A never-smoker is a patient who smoked no cigarettes during his lifetime, also discussed previously [46]. However, in several studies the never-smoker has been defined as a patient who has smoked less than 100 cigarettes [47‐49]. The effect of the different analytical approaches for DNA purification from sputum, employed in previous studies cannot be ruled out. Also determined, that he frequency of promoter methylation of CDKN2A and MGMT was associated with a 22.9 and 29.1-fold increase in the risk to develop cancer in smoker COPD patients. This result is interesting, since Belinsky et al.,[42] reported only 6.5-fold increase in the risk of LC development, through detection of three o mores genes methylated in the sputum. In our cases, both smoking frequency and aberrant promoter methylation of CDKN2A and MGMT gene were associated with progression of COPD diseases in the group of patient without LC. Patients with GOLD > 2 diagnosed as negative to LC in the 2007 developed LC at the beginning of 2012. Other patient with GOLD 0 at the time of diagnosis has LC actually have been GOLD 1 by spirometry. All this patients have smoked between 20 and 35 packs of cigarettes/year for over 40 years during his life. Therefore, these results show a positive correlation between H-S-Freq and H-GOLD with a high prevalence to develop LC.

Although in this study not investigated the effect of cigarette smoking on cellular structures of the bronchial airway. We found have increased sputum inflammatory patterns (neutrophil and macrophage) numbers compared with subjects without COPD nonsmoker. This study showed that there are no significant differences in the cellular profile in sputum samples between patients with COPD who are active smokers and those who have ceased smoking. Similar observation has been demonstrated in other studies [50].

It is known that gene promoter methylation is a very early event in tumoral progression, so our findings probably suggest and confirm that molecular alterations as methylation occur in a temporal-dependent manner during a long time before COPD arising or LC development. A recent review [19], described that oxidants generated endogenously, such as air pollutants or cigarette smoke, generate ROS, producing an inflammatory response in lung cells by activation of stress kinases as c-Jun or NF-Kβ and phosphoinositide-3 kinase (PI-3 K) and other signal transduction pathways, leading to enhanced gene expression of pro-inflammatory mediators [51]. Recently, Damiani et al.[52] have proposed that epithelial molecular alterations in bronchial cells - such as an increase of DNA methyltransferase expression, histone modification, DNA methylation and p53 and K-ras genes mutations - are induced by tobacco carcinogens. These epigenetic and genetic alterations deregulate the expression of genes involved in the anti-inflammatory response and block the DNA repair, contributing to malignant transformation [53], [54]. The involvement of environmental factors – e. g: pollution exposition - in the TSGs methylation pattern observed in our study cannot be ruled out. In fact, it has been reported that pollutant exposure in modern-cities, is a major factor leading to aberrant promoter methylation of TSGs [55]. Therefore, tobacco smoking is a necessary but not sufficient condition for LC development; the synergistic effect between smoking and environmental pollutants must be studied in the future.

Finally, the most significant factor for survival in lung cancer is the stage of disease at diagnosis. Therefore, our results suggest that DNA present in the sputum is an useful non-invasive sample to identify epigenetic alterations such as gene-promoter methylation. These results highlight a potential value of this MSP-sputum molecular marker for early detection of LC in high-risk individuals as smoker at risk (GOLD 0), and may be a complement to current diagnostic techniques used to diagnosis of lung diseases.