Methylation of SERPINA1 gene promoter may predict chronic obstructive pulmonary disease in patients affected by acute coronary syndrome

Diagnostic biomarkers for detecting chronic obstructive pulmonary disease (COPD) in acute coronary syndrome (ACS) patients are not available. SERPINA1, coding for the most potent circulating anti-inflammatory protein in the lung, has been found to be differentially methylated in blood cells from COPD patients. This study aimed to investigate the methylation profile of SERPINA1 in blood cells from ACS patients, with (COPD+) or without COPD (COPD−). Blood samples were from 115 ACS patients, including 30 COPD+ and 85 COPD− according to lung function phenotype, obtained with spirometry. DNA treated with sodium bisulfite was PCR-amplified at SERPINA1 promoter region. Methylation analysis was carried out by sequencing the PCR products. Lymphocytes count in ACS patients was recorded at hospital admission and discharge. SERPINA1 was hypermethylated in 24/30 (80%) COPD+ and 48/85 (56.5%) COPD− (p < 0.05). Interestingly, at hospital discharge, lymphocytes count was higher in COPD− patients carrying SERPINA1 hypermethylated (1.98 × 103 ± 0.6 cell/µl) than in COPD− carrying SERPINA1 hypomethylated (1.7 × 103 ± 0.48 cell/µl) (p < 0.05). SERPINA1 is hypermethylated in blood cells from COPD+ patients. COPD− carrying SERPINA1 hypermethylated and high lymphocytes count may be at risk of COPD development. Therefore, SERPINA1 hypermethylation may represent a potential biomarker for predicting COPD development in ACS patients.


Introduction
Chronic obstructive pulmonary disease (COPD) is a chronic inflammatory lung disease that causes obstructed airflow from the lungs [1]. It is caused by long-term exposure to particulate matter, most often from cigarette smoke [1]. Acute coronary syndrome (ACS) results from acute obstruction of a coronary artery [2]. Several studies suggest that ACS patients with concomitant COPD have a poor prognosis [3][4][5]. However, COPD is frequently undiagnosed in ACS patients, with a significant delay in the treatment and a negative impact on short-and long-term prognosis [5]. The gold standard for COPD diagnosis is the spirometry.
SERPINA1 gene encodes for Alpha-1 antitrypsin (AAT), a blood protease of 52 kDa constitutively released from the hepatocytes [6]. AAT is involved in inflammatory processes [7,8]. Indeed, this protein plays a protective role on the healthy cells adjacent to the inflamed tissue where it inhibits different proteases, including the elastase produced by neutrophils [7]. The activity of AAT is very high in the lower respiratory tract where it provides more than 90% of the defenses against the elastolytic load of neutrophils [9,10]. The relevance of AAT in COPD is evident in individuals carrying mutations in SERPINA1 gene where absence or alteration of the protein in association with cigarette consumption predisposes to the risk of developing COPD [11].
SERPINA1 gene is 12.2 kb and maps on chromosome 14q32.1 [12]. It consists of 7 exons called IA, IB, IC and II-III-IV-V, and six introns [13]. Transcriptional regulation occurs in exons IA, IB and IC, in a tissue-specific manner [13]. IA and IB regulate transcription in the monocytes and macrophages and IC in hepatocytes [14,15]. SERPINA1 is also an inducible gene upon activating its inflammation-responsive promoter in hepatocytes, monocytes and macrophages [7,16,17]. High levels of circulating AAT, fourfold to sixfold higher than baseline levels, are present during the course of inflammation, infections and late pregnancy [18][19][20][21].
In a recent study we have performed a comprehensive assessment of SERPINA1 gene promoter methylation profile in peripheral blood mononuclear cells (PBMCs) from healthy subjects [21]. We showed that SERPINA1 gene promoter is hypermethylated in healthy individuals, such as blood donors, but hypomethylated in pregnant women at the third trimester of pregnancy. These findings suggest that SERPINA1 gene is normally silenced in blood cells of healthy individuals, but it is induced under emergency conditions, such as late pregnancy, when women are highly exposed to the risk of inflammations and infections [21].
Based on our recent findings, in this study we aimed to verify whether methylation of SERPINA1 gene promoter may differ between ACS patients with COPD (COPD+) and without COPD (COPD−). To this aim, the methylation profile of SERPINA1 gene promoter was investigated in blood cells from COPD+ and COPD− patients.

Sample population
Blood samples were collected from 115 ACS patients (mean age ± standard deviation [SD], 65 ± 9 yrs) at the University Hospital of Ferrara, Ferrara, Italy. The inclusion criteria comprised smokers or former smokers (≥ 10 pack/years) patients hospitalized with a clinical diagnosis of ACS, as defined by current European guidelines [22]. Exclusion criteria included previous diagnosis of COPD and/or asthma, known pulmonary diseases other than COPD, ongoing pneumonia, ongoing heart failure, documented or suspicion of malignant disease, life expectancy < 1 year, recent thoracic trauma.

Definition of undiagnosed COPD
Clinical and laboratory data as well as blood samples were collected for all patients during hospitalization. Two months after discharge, all patients underwent spirometry to verify the presence of COPD. A spirometry was performed by two expert pulmonologists and revised by an independent reviewer blinded to patients' clinical conditions and outcomes. Spirometry was performed according to standardized procedures [23]. Briefly, COPD was diagnosed in the presence of: (1) exposure to risk factors for the disease (all patients enrolled in the study were current or former smokers); (2) presence of chronic respiratory symptoms (mainly shortness of breath, cough and sputum) and (3) post-bronchodilator fixed ratio forced expiratory volume at first second (FEV1)/forced vital capacity (FVC) < 0.7.

Blood collection and leukocyte count
A venous blood sample was collected from all ACS patients at the time of both admission and hospital discharge. Leukocyte count was performed by flow cytometry with the automated cell counter Sysmex XN. When needed, a blood smear was prepared for microscopic evaluation and accurate quantification of total lymphocytes count.

DNA extraction and DNA PCR suitability
DNA was isolated from total blood using QIAamp DNA Blood Mini Kit (Qiagen, Milan, Italy), as described [24]. After purification, DNA was quantified by spectrophotometric reading (NanoDrop 2000, Thermo Scientific) [21] and then evaluated for its PCR suitability by amplification for β-globin gene sequence [25].

Treatment of DNAs with sodium bisulfite and SERPINA1 PCR
Methylation analysis was carried out by DNA treatment with sodium bisulfite, using the Epitect Bisulfite kit (Qiagen, Milan, Italy), as reported [21]. Treated DNA was purified with DNA purification columns (Epitect Bisulfite kit, Qiagen, Milan, Italy) and then subjected to PCR amplification for SERPINA1 gene promoter (GenBank accession number: NG_008290.1). Briefly, 150 ng of DNA was amplified using the primers forward 5′-TTT TGG TTT AGT TTA GGA TTT TGA GG-3′ and reverse 5′-ACC TAC CAA TTA TTA ATA CCA AAT CTA TAC -3′ [21,26]. These primers amplify a promoter region of 375 bp (Gen-Bank accession number: NG_008290.1, position 4711-5085), which contains 8 CpG dinucleotides. Among these CpGs, CpG number 1 (CpG-1) and CpG-8 have been previously found hypomethylated in association with lower average lung function phenotypes and COPD [26], while CpG-6 belongs to a CCG CCC -box regulatory consensus sequence of the promoter region (Fig. 1a) [21]. PCR program was: 10 min of denaturation at 95 °C followed by 40 cycles of 1 min at 95 °C, 1 min at 65 °C and 2 min at 72 °C and a final extension for 5 min at 72 °C [21]. A PCR-negative control containing distilled water without DNA, was included per reaction. PCR products were run onto 2% agarose gel electrophoresis and stained with ethidium bromide.

DNA sequencing of SERPINA1 PCR products
To evaluate the methylation levels of SERPINA1 promoter, PCR products were sequenced by direct sequencing [27]. Briefly, PCR products were purified using the QIAquick PCR Purification Kit (Qiagen, Milan, Italy) and then subjected to direct sequencing using the automated ABI-Prism-3130X DNA sequencer (Applied Biosystems, Monza, Italy) [25], as performed before [28]. SERPINA1 sequences showing more than or equal to 5 out of 8 methylated CpGs (≥ 50%) were considered hypermethylated [27,29].

Statistical analysis
Continuous data were tested for normal distribution with the Kolmogorov-Smirnov test. Normally distributed values were presented as mean ± SD and compared by t test [30]. Otherwise, median [interquartile range] and Mann-Whitney U were used. The Chi-square trend test with Yate's correction [31] was used to compare the observed SERPINA1 epigenotypes, i.e., DNA hypermethylation and hypomethylation, among COPD+ and COPD− groups. Lymphocyte values were analyzed with D' Agostino-Pearson test for normality, and groups were compared employing the nonparametric Mann-Whitney U test [32,33]. Spearman correlation analysis was applied to evaluate age-related variations in SERPINA1 promoter methylation by matching the level of methylation to the age of each individual [30]. Multiple linear regression model has been used to estimate the associations between SERPINA1 epigenotypes and lymphocytes count. Statistical analyses were carried out employing the R software package (version 3.5.0) and using the Graph-Pad Prism for Windows (version 5.0, GraphPad) [34]. p values less than 0.05 (p < 0.05) were considered statistically significant [35].

SERPINA1 gene promoter methylation analysis
In order to profile the methylation status, SERPINA1 was PCR amplified in a promoter region containing 8 CpG dinucleotides (Fig. 1a). PCR amplifications were efficiently obtained in all blood samples from all patients (Fig. 1b).
As CpG-6 dinucleotide has been recently associated with the epigenetic regulation of SERPINA1 gene by promoter methylation in blood cells from healthy individuals [21], we specifically investigated the methylation status of CpG-6 in SERPINA1 hypermethylated and methylated and CpG-6 un-methylated was counted in the two groups (Fig. 3a). CpG-6 was found to be methylated in 68/72 (94.4%) hypermethylated ACS samples and in 9/43 (20.9%) hypomethylated ACS samples (p < 0.0001; Fig. 3b).
Moreover, as the mean age was significantly higher in COPD+ (n = 30, mean age ± SD: 70 ± 7 years old) than in COPD− (n = 85, mean age ± SD: 63 ± 9 years old) (p < 0.001), age-related changes in SERPINA1 promoter methylation of ACS patients were assessed. Results indicate a lack of correlation between age and SERPINA1 promoter methylation in ACS patients (r = − 0.071 and p > 0.05) as well as between age and SERPINA1 methylation in both COPD+ (r = − 0.073, p > 0.05) and COPD− groups (r = − 0.166, p > 0.05). These results indicate that SERPINA1 promoter methylation is age independent obstructive pulmonary disease, FEV1: forced expiratory volume at first second, FVC: forced vital capacity, RHSQ: respiratory health screening questionnaire, HDL: high-density lipoprotein, CAD: coronary artery disease, CV: cardiovascular, BMI: body mass index, BSA: body surface area. u/μl: cell/μl, NLR: neutrophil/ lymphocyte ratio Table 1 (continued) Fig. 2 Bisulfite-polymerase chain reaction (PCR) sequencing of the SERPINA1 promoter of DNAs from blood samples belonging to acute coronary syndrome (ACS) patients a with chronic obstructive pulmonary disease (COPD+) and b without COPD (COPD−). In both panels: stratification of bisulfite-PCR sequences according to ACS subgroup and SERPINA1 hyper-and hypomethylation status. Filled-in and clear circles represent methylated and unmethylated CpGs, respectively. The CpGs within the SERPINA1 promoter are numbered across the top of the grid. Each row represents one PCR product/analyzed sequence. Samples showing more than or equal to 5 out of 8 (≥ 50%) methylated CpGs were considered hypermethylated; c frequencies of SERPINA1 promoter hypermethylation in COPD+ and COPD− samples. *p < 0.05 in ACS patients with and without COPD. Thus, age did not represent a possible cofounder factor for SERPINA1 methylation analysis, as reported in healthy subjects [21].
In addition, in order to exclude diabetes as a confounding factor, a correlation between diabetes and SER-PINA1 promoter methylation was assessed. The whole pool of ACS patients (n = 115) was stratified in patients with (n = 28) and without (n = 87) diabetes, and SER-PINA1 hypermethylation rates were compared. A total of 67.9% (19/28) and 60.9% (53/87) of ACS patients with and without diabetes, respectively, presented SER-PINA1 hypermethylation (p > 0.05). ACS patients were afterward stratified according to diabetes status in COPD+ with (n = 9) and without diabetes (n = 21) as well as in COPD− with (n = 19) and without diabetes (n = 66), and SERPINA1 hypermethylation rates were compared. SERPINA1 was hypermethylated in 100% (9/9) of COPD+ patients with diabetes and in 71.4% (15/21) of COPD+ patients without diabetes (p > 0.05). Moreover, 52.6% (10/19) and 57.6% (38/66) of COPD− patients with and without diabetes, respectively, presented SERPINA1 hypermethylation (p > 0.05). Thus, no differences in SER-PINA1 methylation status were observed when analyzing groups according to diabetes. These results indicate that the presence of diabetes did not suppose an excluding factor in our study.
Considering that the level of neutrophil count and neutrophil/lymphocyte ratio (NLR) is important prognostic predictors to assess the degree of inflammation [39,40], these parameters were evaluated in ACS patients with and without COPD at both hospital admission and discharge. Neutrophil count was 5.95 × 10 3 ± 2.06 cell/μl in COPD+ and 6.11 × 10 3 ± 2.66 cell/μl in COPD− at hospital admission (p > 0.05). At hospital discharge, neutrophil count was 4.91 × 10 3 ± 1.36 cell/μl in COPD+ and 5.56 × 10 3 ± 1.84 cell/μl in COPD− (p > 0.05). NLR resulted to be 3.78 × 10 3 ± 2.51 cell/ μl in COPD+ and 3.47 × 10 3 ± 2.17 cell/μl in COPD− at hospital admission (p > 0.05), and 3.39 × 10 3 ± 1.88 cell/ μl in COPD+ and 3.24 × 10 3 ± 1.44 cell/μl in COPD− at hospital discharge (p > 0.05). Furthermore, no differences in both neutrophil count and NLR were found when comparing hospital admission and discharge values, for both COPD+ and COPD− groups (p > 0.05). These results indicate that measuring for neutrophil count and/or NLR in ACS patients does not enable to differentiate between COPD+ and COPD− ACS patients and cannot help in predicting which ACS patients are at risk of developing COPD.

Discussion
In the present study, we investigated the methylation status of SERPINA1 gene promoter in blood samples from ACS patients with (COPD+) and without (COPD−) COPD. Our results show that SERPINA1 promoter (See figure on next page.) Fig. 3 a Acute coronary syndrome (ACS) patient's stratification according to the number of CpGs methylated in SERPINA1, i.e., hypermethylation and hypomethylation. Methylated and unmethylated CpG-6 in SERPINA1 hypermethylated and hypomethylated samples, respectively, are reported. Samples showing more than or equal to 5 out of 8 (≥ 50%) methylated CpG dinucleotides were considered hypermethylated. Filled-in and clear circles represent methylated and unmethylated CpGs, respectively. The CpGs within the SERPINA1 promoter are numbered across the top of the grid. Each row represents one PCR product/analyzed sequence. Vertical red box highlights CpG-6 dinucleotides across hypermethylated and hypomethylated samples. b Frequencies of methylated CpG-6 in SERPINA1 promoter hypermethylated and hypomethylated ACS samples. ****p < 0.0001 Rotondo et al. Clin Epigenet (2021) 13:79 methylation is higher in COPD+ patients compared to COPD−. Therefore, SERPINA1 gene promoter hypermethylation may be a potential biomarker for detecting COPD in ACS patients. SERPINA1 is constitutively expressed in hepatic cells [6]. SERPINA1 is also expressed in inducible manner in hepatocytes and blood cells during the course of the inflammation, infections and late pregnancy [16,17,[19][20][21]. In PBMCs of healthy individuals the SERPINA1 promoter is hypermethylated except in women in advanced stage of pregnancy, suggesting the gene silencing under normal physiological conditions and the gene expression induction when circulating AAT is needed to be increased [19][20][21]. In this study, SERPINA1 gene promoter was found to be hypermethylated in 62.6% of ACS patients. The prevalence of SERPINA1 hypermethylation was higher in COPD+ (80%) than in COPD− (56.5%) patients (p < 0.05). This result suggests an association between SERPINA1 hypermethylation and COPD+ leading to the hypothesis that lack of AAT expression in blood cells may play a role in the development of COPD in ACS patients. The relevance of AAT in COPD is well known in smokers carrying mutations in SERPINA1 gene where protein deficiency synergizes with smoking in exerting strong adverse effects on lung function, causing COPD [11]. As ACS patients enrolled in this study are all smokers, the adverse effects on lung functions may be mediated by altering promoter methylation in SERPINA1. In addition, we specifically analyzed methylation status at CpG-6 site. The dinucleotide CpG-6 falls within a CCG CCC box regulatory consensus region of SERPINA1 promoter. The CCG CCC box is located 46 nucleotides upstream of the transcription start site in SERPINA1 promoter, and it is recognized by transcription factors (TFs) to recruit RNA polymerase, thereby inducing transcription [41]. Methylation at CpG-6 could prevent CG-box-TFs interaction, thereby inducing gene transcriptional repression [42]. In our previous study, we have found that methylation status of CpG-6 reflects the methylation status of SERPINA1, being methylated in SERPINA1 hypermethylated samples and unmethylated in SERPINA1 hypomethylated samples [21]. In this study, nearly 95% of SERPINA1 hypermethylated samples had CpG-6 methylated, and 79% SERPINA1 hypomethylated samples had CpG-6 unmethylated. These results extend our previous observations and confirm that CpG-6 may play a pivotal role in epigenetic regulation of SERPINA1 in blood cells [21]. Therefore, the methylation status of CpG-6 may represent a useful marker of expression of SERPINA1 in blood cells.
As high levels of circulating lymphocytes play a role in COPD pathogenesis [36][37][38], lymphocytes count was analyzed in COPD+ and COPD− patients at hospital admission and discharge. Lymphocytes levels were similar in COPD+ and COPD− patients, both at hospital admission and discharge. Nevertheless, COPD− patients carrying SERPINA1 hypermethylated had circulating lymphocytes higher than COPD− patients carrying SER-PINA1 hypomethylated, at hospital discharge (p < 0.05). This result suggests that COPD− patients carrying SER-PINA1 hypermethylated may counteract inflammation less efficiently than COPD− patients carrying SERPINA1 hypomethylated. Therefore, we hypothesize that SER-PINA1 hypermethylation may represent a risk factor for developing COPD in ACS patients at risk (smokers) with high levels of circulating lymphocytes in the presence of pharmacological treatments following percutaneous coronary intervention. In this view, ACS patients at risk of developing COPD might be identified by combining epigenetic and hematologic parameters. Follow-up studies should be carried out to verify whether hypermethylated SERPINA1 and lymphocyte count may be useful markers to predict the development of COPD in COPD− patients.
Few studies have explored SERPINA1 methylation in COPD pathogenesis. Our findings are partially in line with a recent meta-analysis study of tobacco-smoke exposed children and adult smokers, which reported association between methylated CpGs located about 32 kb downstream of SERPINA1 gene and decline in lung functions [26,43]. Our data are discordant compared to a previous family-based study of smoking subjects with and without a history of COPD, which found two hypomethylated CpG sites, the CpG-1 and CpG-8 of this study, associated with lower average lung function phenotypes and COPD [26]. It should be pointed out that comparisons between our data and those reported above are difficult to be made, as both patient enrolment criteria and methodological approaches used herein are different. Indeed, we enrolled ACS patients, who were stratified according to the presence/absence of COPD, while previous data were obtained from predominantly smoking adults with/without COPD [26]. About methods, unlike previous investigations that studied epigenome-wide associations, our study was specifically designed to assess the methylation profile of SERPINA1 gene promoter in blood cells from ACS patients.
Our study presents some limitations. First, we studied the epigenetic regulation of SERPINA1 through promoter methylation without validation analyses, such as AAT mRNA and protein expression. However, in our previous study conducted on healthy subjects we determined an association between variations in SERPINA1 promoter methylation in PBMCs and changes in AAT circulating levels [21], while promoter methylation of SERPINA1 in association with gene expression inhibition has been demonstrated in animal models [44][45][46]. Second, we did not analyze mutations in SERPINA1 gene. Since about 1% of individuals carry mutations in SERPINA1, it is possible that some COPD+ may be associated with mutations in this gene. Considering that most of the gene mutations occur in the coding region [47], we may infer that SER-PINA1 gene regulation by methylation in blood cells is not affected.
In conclusion, the present study shows, for the first time, that SERPINA1 gene promoter is hypermethylated in blood cells from COPD+ patients, compared to COPD−. Further, we show that CpG-6 methylation status reflects the methylation status of SERPINA1 promoter. We also found that COPD− patients with SER-PINA1 hypermethylated present higher lymphocytes levels, in the presence of pharmacological treatments following percutaneous coronary intervention, than COPD− patients with SERPINA1 hypomethylated. Collectively, our data indicate that SERPINA1 hypermethylation may play a role in the development of COPD in ACS patients. Therefore, SERPINA1 promoter methylation may be a potential biomarker for detecting/predicting COPD in ACS patients.