According to the estimates of the PLATINUM study, 16.9% of the population of Santiago de Chile aged 40 years or older have COPD,1 while preliminary results from our MaulEPOC study showed that 90% of COPD patients in Chile's Maule region were smokers or former smokers.2 One of the reasons for the high incidence of COPD in Chile could be the high prevalence of smoking in the population.3

Apart from alpha-1-antitrypsin deficiency, many other genetic variants associated with the disease have been described, including CHRNA3/5,4 a region also associated with nicotine addiction. This suggests that other smoking-related genes may also play an important role in the pathogenesis of COPD.5 In this respect, some single nucleotide polymorphisms (SNPs) in the FRMD4A gene have been described that are significantly associated with nicotine dependence in Asians,6 some of which were also replicated in Caucasian and African-American populations. To date, however, the FRMD4A gene has not been studied in Latin American subjects with and without COPD. The aim of this paper, therefore, was to analyze the contribution of SNPs located in the FRMD4A gene to COPD susceptibility in the Chilean population.

A total of 568 subjects (322 with COPD and 246 healthy controls) were recruited from the Hospital Regional de Talca between 2016 and 2018. They all signed an informed consent form approved by the ethics committee of the Maule Health Service. A diagnosis of COPD was confirmed according to the GOLD criteria,7 and clinically stable COPD patients and healthy controls underwent functional testing. A blood sample was also obtained for molecular testing and a standardized questionnaire2 was completed on social and demographic history, current treatment, comorbidities, CAT index, dyspnea grade according to the modified Medical Research Council scale, number of exacerbations in the last 2 years, and number of exacerbations requiring hospitalization. History of exposure to known COPD risk factors was also recorded, including smoking history (pack-year index) and exposure to biomass smoke in the home. Our study complied with the guidelines for studies in humans and was conducted in accordance with the ethical principles of the World Medical Association Declaration of Helsinki.

Genotyping was performed in 214 COPD patients and 193 healthy controls of the overall cohort. We selected 60 haplotype SNP markers (ht-SNP) based on linkage disequilibrium (LD) patterns located within the FRMD4A gene, using the HapMap dataset.8 Haplotype SNP markers were selected with the Haploview Tagger tool,9 according to the following criteria: low-frequency allele≥0.01 and r2>0.8, based on the HapMap populations (CEU=European reference population and MEX=Mexican native reference population). Genotyping was performed using the OpenArray®™ TaqMan platform (Applied Biosystems Inc., California, USA).

The sex variable was expressed by number of subjects, and differences between study groups were determined by the Chi-squared test. Continuous variables were expressed by mean and standard deviation. For the latter, an ANOVA variance analysis was performed, taking a p-value<0.05 as significant. These analyses were conducted with the statistical R package (R Foundation for Statistical Computing, Vienna, Austria). Allele frequencies between COPD patients and healthy controls were compared using the Chi-squared test, and odd ratios with 95% confidence intervals were calculated using the PLINK program.10

Significant differences were observed in the sex ratios between control subjects and COPD patients (p<0.05), although ages were similar in both groups (p>0.05; Table 1). COPD patients showed reduced FEV1, FEV1/FVC, DLCO, and oxygen saturation values, as well as poorer performance in the 6-minute walk test (p<0.05; Table 1). COPD patients also had a greater cumulative exposure to tobacco and biomass smoke (p<0.05; Table 1).

The selected SNPs showed a low-frequency allele>0.01, successful genotyping in at least 90% of the samples, and panmictic equilibrium, p>0.005, so all were included in the association study. To compare the frequencies of the different SNP markers, the Bonferroni correction was applied by multiplying the p-value by the number of SNPs analyzed in the study (n=60) to obtain the corrected p-value (Bonferroni p-value). p-Values below 0.05 were considered significant. Seven markers showed a tentative association with COPD susceptibility (p<0.05), of which rs10906545 and rs1218353 showed the highest association (Bonferroni p-value<0.05) (Table 2 [all subjects]). When only smokers were analyzed, both COPD patients and controls, rs10906545 and rs1534627 were significantly associated with COPD susceptibility (Table 2 [smokers only]). Since the intensity of smoking measured in pack-years was significantly higher in smokers with COPD than in smokers without COPD (36.77±16.22 versus 14.25±9.63, respectively, p<0.05), the association may be more closely linked to nicotine dependence (and, as such, to the cumulative exposure to tobacco smoke) than to intrinsic pathogenic mechanisms activated by precipitating factors. In contrast, no association was detected when only patients and controls with no smoking history were analyzed (data not shown).

Our results show for the first time that the SNPs rs10906545, rs1218353, and rs1343005 are associated with an increased risk of developing COPD among smokers. Although Yoon et al. previously reported that the rs4424567 polymorphism in FRMD4A was associated with nicotine dependence,6 it showed no association in our cohort or proximity to any of the SNPs that were significantly associated with COPD. Moreover, the rs1534627 SNP was only associated with COPD among smokers, and the association was lost when non-smokers were included, probably due to a statistical effect in which non-smokers offset the differences observed in smokers.

FRMD4A codes for a scaffolding protein that activates ADP-ribosylation factor 6 (Arf6)11 that is involved in cell signal transduction, membrane traffic, and actin cytoskeleton organization.12FRMD4A gene overexpression has been associated with squamous cell carcinoma of the head and neck and non-small cell lung carcinoma.13,14FRMD4A hypermethylation has also been reported in MCF-7 cells treated with benzopyrene,15 a polycyclic aromatic hydrocarbon present in many environmental pollutants, including biomass smoke and cigarette smoke.16 In the case of COPD, variations have been described in the DNA methylation pattern between patients with COPD or reduced lung function and control subjects.17,18 In this respect, a recent study by Morrow et al. revealed differences in FRMD4A methylation between subjects with and without COPD.19 In addition, FRMD4A has also been reported to be associated with nicotine dependence,14 and smoking during pregnancy has recently been shown to cause changes in the FRMD4A methylation pattern in newborns, and that these changes persist for many years after prenatal exposure.20

In conclusion, the present study is the first to describe an association between FRMD4A polymorphisms and the risk of developing COPD. Overall, our results and those of previous studies suggest that both genetic and epigenetic variations in the FRMD4A gene could be involved in the pathogenesis of COPD. Since FRMD4A is also associated with nicotine dependence, our results underscore the need to develop new strategies for smoking and COPD.