IREB2, CHRNA5, CHRNA3, FAM13A & hedgehog interacting protein genes polymorphisms & risk of chronic obstructive pulmonary disease in Tatar population from Russia

Chronic obstructive pulmonary disease (COPD) is a complex chronic inflammatory disease of the respiratory system affecting primarily distal respiratory pathways and lung parenchyma. It manifests with partially reversible bronchial obstruction and shows a progressive course with lung emphysema and increasing respiratory failure1. Smoking is generally considered as a principal risk factor for COPD: the disease develops in 20-30 per cent of smokers1. An important internal risk factor is hereditary predisposition. Genetic mechanisms underlying COPD have been extensively investigated all over the world2.

Genome-wide association studies (GWAS) have identified several loci associated with COPD, in particular, in chromosomal regions 15q25.1 near cholinergic receptor, nicotinic, alpha 3/5 (CHRNA3/5) and iron-responsive element binding protein 2 (IREB2), the chromosome 4q24 region near family with sequence similarity 13, member A (FAM13A) and chromosome 4q31 region near hedgehog interacting protein (HHIP)34567. Single nucleotide polymorphisms (SNPs) of these genes have been found to be associated with COPD, lung function parameters and smoking behaviour in different Caucasian and Mongoloid populations891011.

The gene cluster CHRNA3/A5/B4 encodes nicotinic acetylcholine receptor subunits alpha 3, alpha 5 and beta 4. These genes are expressed in the central nervous system and in bronchial epithelium these play a key role in the formation of nicotine addiction1012131415. IREB2 is located in the same chromosomal region 15q25.1 and encodes iron-responsive element-binding protein 2, which is involved in maintaining iron balance in lung tissues1116. Intracellular iron concentration plays an important role in oxidative stress16. IREB2 expression is modulated by hypoxia, which commonly accompanies COPD16. FAM13A, a gene located on 4q22, comprises 25 exons and encodes FAM13A, a protein with a key role in signal transduction9. It is known that hypoxia enhances FAM13A expression17. HHIP is located on 4q31.21-4q31.3 and encodes a transmembrane glycoprotein that binds three members of the Νedgehog signalling pathway, which all play an important role in lung development18. COPD-associated polymorphisms were identified in the HHIP enhancer; substitutions at these sites result in a decrease in the promoter activity1920.

The molecular genetic basis of COPD remains largely unclear, data obtained from different populations and ethnic groups often disagree2891011. These problems are largely related to the complex nature of this disease and to the strong genetic diversity of human populations21. The frequency distribution of the IREB2, CHRNA5, CHRNA3, FAM13A and HHIP polymorphisms and their association with COPD has not yet been investigated in populations of Russia. This study was aimed at investigating the association of COPD with polymorphisms of IREB2, CHRNA5, CHRNA3, FAM13A and HHIP in a Tatar population from Russia.

Material & Methods

The study protocol was approved by the Local Ethical Committee of Institute of Biochemistry and Genetics of Ufa Scientific Center of Russian Academy of Sciences (IBG USC RAS), Ufa, Russia (Ufa, Protocol No 17, December 7, 2010). Written informed consent was obtained from all participants. The patients and controls were selected from December 2010 to January 2013 from the pulmonary departments of the Ufa City Hospitals №13, №18, №21 (Ufa, The Republic of Bashkortostan, Russia). The laboratory work was conducted in the Genomics Department, IBG USC RAS Ufa, Russia. The blood samples (4 ml) were collected from unrelated patients with COPD (affected group) and unrelated control group (unaffected group) age, sex and ethnically matched. Ethnic origin (up to the third generation) of all the participants was derived by direct interviews with examined persons.

Inclusion and exclusion criteria: The diagnosis of COPD was made according to the International Classification of Diseases tenth revision (ICD 10)22 and following the recommendations of the GOLD (2011)1. For all patients with COPD, the diagnosis was based on the medical histories and the results of general, clinical and special tests (chest X-ray, spirometry measures and fibrobronchoscopy), physical examination and laboratory tests. Patients were excluded from the study if they had diagnosis of asthma and lung cancer. The predicted values for forced vital capacity (FVC), forced expiratory volume in 1 sec (FEV₁) and FEV₁/FVC ratio were generated using previously defined prediction equations as detailed to the European Coal and Steel Community2324. All COPD patients had post-bronchodilator FEV₁/FVC values of <70 per cent. The study group consisted of 511 unrelated COPD patients.

The control group comprised of 508 unrelated age, sex and ethnicity (Tatar population) matched healthy residents of Ufa with no history of chronic diseases such as respiratory system pathology and allergic diseases in the anamnesis. These individuals came to the Ufa City Hospitals №13, №18, №21 (Ufa, Russia) for regular medical examination. All control group individuals had normal lung function (FEV₁/FVC >70%, FEV₁ >80%) (Table I).

Table I Characteristics of the study and control groups

Sample size: The sample size was calculated by Quanto software25. The sample size (n=511 for study group and n=508 for control group) was sufficient to detect the association of the examined five candidate genes and COPD with more than 80 per cent power (Power: 95.53%, disease prevalence, 7%, error: 5%). Based on minor allele frequency (MAF) of six candidate SNPs: IREB2 (rs13180), CHRNA5 (rs16969968), CHRNA3 (rs1051730), CHRNA3 (rs6495309), FAM13A (rs7671167), HHIP (rs13118928) in Caucasians (HapMapCEU)21, a power calculation was performed for the study. To detect an odds ratio (OR) of 2.0, assuming a power of 80 per cent, significance level 5 per cent, disease prevalence, 7 per cent, 225 persons were required in each group for IREB2, 218 persons for CHRNA5, 218 persons for CHRNA3 (rs1051730), 138 persons for CHRNA3 (rs6495309), 163 persons for FAM13A, 215 persons for HHIP. To account for possible variations in the genotype distribution in small datasets, 511 patients were included in COPD group and 508 individuals in control group.

Genotyping: Genomic DNA was isolated from peripheral blood leucocytes using the standard phenol-chloroform extraction procedure26. Six SNPs: rs13180 (IREB2), rs16969968 (CHRNA5), rs1051730, rs6495309 (CHRNA3), rs7671167 (FAM13A), rs13118928 (HHIP) were examined by real-time PCR, with the use of commercial kits (TaqMan SNP discrimination assays) custom designed by (http://testgen.ru, “TestGene” LLC, Ulyanovsk, Russia). Real-time PCR amplified in 20 μl of the reaction mixture containing 1 μl of genomic DNA (concentration 30 ng/μl) and PCR Master Mix containing 200 nM of each dNTP, 67 mM Tris-HCl (pH 8.8 at 25°C), 16.6 mM (NH₄)₂ SO₄, 0.01 per cent Tween - 20, 2 mM MgCl₂, 500 nM primers, 250 nM fluorogenic probes and 1.5 units of Taq polymerase (Thermo Fisher Scientific Inc., USA). PCR was performed by initial denaturation for 2 min at 95°C, 40 cycles of denaturation for 10 sec at 94°C, annealing for 1 min at different annealing temperatures (Table II). Accumulation of specific PCR-product by hybridization and cleavage of double-labelled fluorogenic probe during amplification was detected with a BioRad CFX96 instrument (Bio-Rad Laboratories Inc., USA). End-point fluorescence and genotype discrimination were determined according to the BioRad CFX96 protocol, using CFX Manager software. Individuals with each of the three possible genotypes for each SNP were confirmed by sequencing by kit (“TestGene” LLC, Ulyanovsk, Russia) and included on each genotyping tray as control. For quality control, 5 per cent duplicates and blank controls were also taken up along with the samples in each experiment. The genotyping was blind to case or control status of the samples. The data were excluded after examining missingness, reproducibility and inbreeding. All subjects with a genotype call rate of <95 per cent were removed. Subsequently, SNPs were filtered according to their proportion of missing, MAF or deviation from Hardy-Weinberg-Equilibrium (HWE)27.

Table II Sequences of the amplification primers and fluorogenic probes and polymerase chain reaction conditions

Statistical analysis: For the quantitative traits, the mean values and standard deviations were calculated; the group comparison was performed with a non-parametric Mann–Whitney U-test. The frequencies of qualitative traits were compared using the Pearson's chi-square analysis. Statistical analysis was carried out with the Statistica v. 6.0 programme (StatSoft Inc., Tulsa, OK, USA). MAF and the agreement of the genotype distribution to the HWE (χ²), the association analysis using the basic allele test and the calculation of the OR for the rare allele of each locus and the significance of inter-group differences in allele and genotype frequencies (Chi-square test for sample heterogeneity and the P value) and Cochran-Armitage trend test were performed with PLINK v. 1.0728. To control Type I error rate, Bonferroni correction for multiple comparison was performed. Logistic regression was used to detect the association of SNPs and haplotypes of linked loci in different models, accounting for quantitative and binary traits [gender, age, pack-years, smoking status, body mass index (BMI)]. The significance of the obtained model accounting for all variables was verified by the significance of the likelihood ratio test (P_adj). The best model was chosen using the Akaike's information criterion (AIC). For each significant locus (P<0.05), the model with lowest AIC was chosen. Linear regression analyses were performed to estimate the relationship between SNPs and quantitative phenotypes, such as lung function parameters and pack-years. The regression analysis was performed with PLINK v. 1.07 and SNPStats packages2729. The linkage disequilibrium (LD) structure in the CHRNA3/5 and IREB2 region and haplotype frequencies, the standard LD coefficient (D’), the inter-group differences in the haplotype frequencies were calculated with Haploview 4.229.

Results

Systematic quality control procedures were performed to obtain a high quality of data. Subsequently, SNPs were filtered according to their proportion of missings, MAF or deviation from HWE within the controls. For the control group, the following results were obtained: IREB2 (rs13180) (P=0.123, MAF=0.410), CHRNA5 (rs16969968) (P=0.457, MAF=0.271), CHRNA3 (rs1051730) (P=0.494, MAF=0.238), CHRNA3 (rs6495309) (P=0.984, MAF=0.204), FAM13A (rs7671167) (P=0.141, MAF=0.465), HHIP (rs13118928) (P=0.954, MAF=0.334). The data on the allele and genotype frequency distribution of six SNPs: IREB2 (rs13180), CHRNA5 (rs16969968), CHRNA3 (rs1051730, rs6495309), FAM13A (rs7671167), HHIP (rs13118928) in COPD and control groups were obtained (Table III).

Table III Allele and genotype frequencies of the IREB2, CHRNA5, CHRNA3, FAM13A and HHIP polymorphisms in chronic obstructive pulmonary disease (COPD) patients and control subjects

The COPD and control groups differed significantly in the allele and genotype frequency distributions of IREB2 (rs13180) (P=0.00001, OR=0.67 for allele test and P=0.00001, OR=0.64 for Cochran-Armitage test), CHRNA5 (rs16969968) (P=0.0001, OR=1.42 and P=0.00034, OR=1.41) and CHRNA3 (rs1051730) (P=0.0002, OR=1.44 and P=0.00015, OR=1.47) (Table III). At the next stage, the association was analyzed using various models. By this approach, the IREB2 (rs13180) showed association with COPD in recessive model (P_adj=0.0001, P_cor=0.0006, OR=0.61) and additive model (P_adj=0.00001, P_cor=0.00006, OR=0.64) (Table IV). The COPD risk was higher in homozygous and heterozygous carriers of the rare A allele of CHRNA5 (rs16969968) (P_adj=0.001, P_cor=0.006, OR=1.47 in dominant model). The highest level of significance (P_adj=0.0001, P_cor=0.0006) and the lowest AIC were obtained for the additive model, the risk of COPD was associated with each copy of the rare A allele (OR=1.41) (Table IV). In the COPD group, the frequency of AA genotype of the CHRNA3 (rs1051730) two-fold increase (9.8 vs. 4% in control, P_adj=0.0002, P_cor=0.0012, OR=2.63, in recessive model). The lowest Akaike's information criteria testing the association of rs1051730 was obtained for the additive model with corresponding P_adj=0.0001 (P_cor=0.0006, OR=1.47) (Table IV).

Table IV Association between IREB2, CHRNA5, CHRNA polymorphisms and chronic obstructive pulmonary disease (COPD)

Haplotype association analysis: Since IREB2 (rs13180), CHRNA5 (rs16969968), CHRNA3 (rs1051730, rs6495309) belong to the same LD block on chromosome 15q25, we analyzed the haplotype frequencies of these polymorphisms in COPD and controls groups. The pairwise LD values (D’ - Lewontin's coefficient, r² - correlation coefficient between the two loci) were calculated for rs13180, rs16969968, rs1051730, rs6495309 in chromosome 15q25. Strong LD between SNPs were found (Figure and Table V). The strong levels of LD between rs16969968 and rs1051730 (D’=0.783, r² =0.516) were observed. Both SNPs were significantly associated with COPD in the Tatar population with similar ORs.

Close

Figure

Visualization of linkage disequilibrium among single-nucleotide polymorphisms on chromosome 15q25 in Tatar population from Russia (linkage disequilibrium values are presented as D’ and linkage disequilibrium block, calculated with Haploview 4.2 software).

Export to PPT

Table V Linkage disequilibrium between the CHRNA3/5 and IREB2 genes polymorphic markers

Based on the results of the haplotype frequency analysis of CHRNA5 (rs16969968), CHRNA3 (rs1051730, rs6495309), it was observed that the COPD group differed significantly from control individuals in their haplotype frequency distribution (Table VI). The percentage of the A-A-G haplotype by rs16969968, rs1051730, rs6495309 was higher among COPD patients (25.8% in COPD vs. 20.74% in control, P_adj=0.0078, OR=1.44). Haplotype analysis demonstrated significant association of IREB2 (rs13180) and CHRNA3 (rs1051730) C-G haplotype with COPD [P_adj=0.0005, OR=0.61 95% confidence interval (CI) 0.47-0.81] (Table VI).

Table VI Association of IREB2, CHRNA5 and CHRNA3 haplotypes with chronic obstructive pulmonary disease (COPD)

Association of IREB2, CHRNA3/A5, FAM13A and HHIP polymorphisms with COPD stratified by smoking status: No significant gene was observed by environment interactions in the logistic regression analysis of six studied SNPs with smoking status and pack-years of smoking. Genotype and environment interactions were also analyzed by comparing the OR values for studied SNPs obtained for the groups of smokers and non-smokers.

The COPD risk in smokers was associated with CHRNA5 (rs16969968) (P_adj=0.002, P_cor=0.012, OR=1.44) and CHRNA3 (rs1051730) and (P_adj=0.00004, P_cor=0.00024, OR=2.11) in additive model and IREB2 (rs13180) in recessive model (P_adj=0.00001, P_cor=0.00006, OR=0.49) (Table VII). The A-A-G haplotype by rs16969968, rs1051730, rs6495309 (P_adj=0.0071, OR=1.56 95% CI 1.13-2.16) and of Ρ-G haplotype by IREB2 (rs13180) and CHRNA3 (rs1051730) (P_adj=0.0001, OR=0.58 95% CI 0.42-0.80) were significantly associated with COPD in smokers.

Table VII Association analysis of IREB2, CHRNA5, CHRNA3 polymorphisms with chronic obstructive pulmonary disease (COPD) in smokers

Single markers and haplotype association analysis in non-smokers did not demonstrate any significant association with COPD. The groups of COPD patients, stratified by smoking status - smokers (n=392) and non-smokers (n=119) were also compared. The groups of COPD patients did not differ by any of the studied polymorphisms.

Genetic association results between the IREB2, CHRNA3/A5, FAM13A and HHIP polymorphisms and lung function parameters: The relationship between the rs13180 (IREB2), rs16969968 (CHRNA5), rs1051730, rs6495309 (CHRNA3), rs7671167 (FAM13A), rs13118928 (HHIP) and lung function parameters was investigated. The rs16969968 (CHRNA5) was associated with FEV_1% predicted in dominant model (b=−3.81, s.e.=1.48, P_adj=0.005). The presence of GA genotype for the rs1051730 (CHRNA3) was associated with a 4.36 per cent decrease in FEV_1% predicted (b=−4.36, s.e.=1.72, P_adj=0.0019) (Table VIII).

Table VIII Association results between, CHRNA5 and CHRNA3 polymorphisms and lung function in chronic obstructive pulmonary disease (COPD) patients

There was no significant association between the rs13180 (IREB2), rs16969968 (CHRNA5), rs1051730, rs6495309 (CHRNA3), rs7671167 (FAM13A), rs13118928 (HHIP) and FEV₁/FVC predicted in the COPD patients.

Discussion

The purpose of our study was to analyze the contribution of IREB2, CHRNA5, CHRNA3, FAM13A and HHIP polymorphisms to COPD in the ethnically homogenous Tatar population from Russia. It was found that rs13180 (IREB2), rs16969968 (CHRNA5) and rs1051730 (CHRNA3) were associated with COPD in the Tatar population. The minor alleles of both rs16969968 (CHRNA5), rs1051730 (CHRNA3) and the A-A-G haplotype by rs16969968, rs1051730, rs6495309 of CHRNA3/A5 were significantly associated with COPD.

Polymorphisms of rs16969968 (CHRNA5) and rs1051730 (CHRNA3) were significantly associated with lung function (FEV_1% predicted) after adjusting for age, sex, BMI, pack-years and smoking status. Similar results were observed in a Chinese population303132. Our data also corroborated with the results obtained in several Caucasian populations456. A meta-analysis by Zhang et al14 showed that A allele of rs1051730 (CHRNA3) was a COPD susceptibility factor, both for respiratory airway obstruction and for emphysematous destruction of lung parenchyma. Shpagina et al33 did not find any association between rs1051730 (CHRNA3) and occupational chronic obstructive lung disease in Russian population from Novosibirsk. Siedlinski et al34 confirmed the existence of direct effects of the CHRNA3, IREB2, FAM13A and HHIP loci on COPD development. This study indicated that the association of the CHRNA3 locus with COPD was significantly mediated by smoking status, and IREB2 associated with COPD independent of smoking. Lococo et al35 demonstrated that the variants on the gene cluster CHRNA3/A5/B4 were associated with nicotine addiction and antismoking therapy side effects. We did not observe any significant association of these loci with the smoking index. The rs13180 (IREB2), rs16969968 (CHRNA5) and rs1051730 (CHRNA3) were significantly associated with COPD only in smokers, which might be due to the insufficient size of the non-smoker sample in our study.

Analysis of the allele frequency distribution patterns at the CHRNA5 (rs16969968) and CHRNA3 (rs1051730) among the Tatar ethnic groups in comparison with the worldwide populations indicated that significant interethnic differences were found among the Tatar population and Mongoloid and Africans populations, having a minimum frequency of rare alleles of both markers [3.6% in Chinese (HapMap-HCB), 1.2% in Japanese (HapMap-JPT), 0 per cent in Africans (HapMap-YRI) for rs16969968; and 3.5, 1.2 and 9.7% for rs1051730]21 (Table IX). Analysis of previously published data concerning CHRNA5 (rs16969968) and CHRNA3 (rs1051730) markers allele frequencies distribution showed that the Tatars ethnic group from Russia differed significantly in terms of these markers from the general population of Caucasians (HapMap-CEU21), in which the rs16969968 and rs1051730 rare allele frequency was 38.5 per cent. Tatars and Indians are in an intermediate position, for them the frequency of minor allele does not exceed 28 per cent. Allele frequencies in Tatars were similar in the prevalence of the polymorphic variants of CHRNA5 (rs16969968) and CHRNA3 (rs1051730) markers among Indians (HapMap-GIH)21 (27.12 vs. 21.0% and 23.79 vs. 21.0%).

Table IX Minor allele frequencies (%) of the IREB2, CHRNA5, CHRNA3, FAM13A and HHIP polymorphisms in control Tatar population and in worldwide populations

We also analyzed the potential COPD association with the rs6495309 polymorphism of the CHRNA3 promoter in Tatar population from Russia. This locus is particularly interesting, since it alters the promoter affinity to the octamer-binding transcription factor (Oct-1), thus affecting CHRNA3 expression31. However, in the population studied, rs6495309 (CHRNA3) was not significantly associated with COPD or its progression. It has been shown to be associated with COPD in populations from China and Korea3637. It was also linked with lung cancer risk and prognosis in a Chinese population37. Significant associations of rs6495309 with COPD observed in Mongoloid populations may be related to the high frequency of the A allele, which ranges from 48.8 per cent in the Japanese to 52.4 per cent in the Chinese21. On the other hand, rs6495309 allele and genotype frequencies in Tatars from Russia are similar to those found in other Caucasian populations21, where A is the minor allele with frequencies of below 20 per cent (Table IX).

IREB2 polymorphism rs13180 was also associated with COPD in the population studied. The C-G haplotype by rs13180 and rs1051730 SNPs of IREB2 and CHRNA3 was a protective factor for COPD in the present study. IREB2 polymorphisms have been reported to be associated with lung function parameters and with COPD in a nonsmoking Chinese population30. In a Polish population, rs13180 was associated with severe COPD10. Chappell et al11 confirmed the involvement of IREB2 polymorphisms in predisposition to COPD in Caucasian populations. The SNPs in the IREB2 showed associations with COPD in case-control and family-based studies510303132. Two key processes involved in COPD pathogenesis and lung tissue damage, proteolysis and oxidative stress, are related to each other via intracellular iron homeostasis511. It has been shown that IREB2 is actively expressed in the lungs, whereas aberrations of iron balance can cause oxidative stress and lung tissue inflammation5.

Our study did not confirm an association between the rs13118928 (HHIP) and COPD in Tatars population from Russia. However, the frequencies of the minor G allele and the GG genotype in the Tatar population were similar to those observed in other Caucasian populations821. GWASs and subsequent replication studies showed that HHIP polymorphisms were associated with COPD and lung function parameters1920. According to a study by Zhou et al8, rs13118928 contributed to the risk of severe COPD in smokers in a Polish population. In another study, rs13118928 was not associated with COPD in the Chinese, but rs12504628 was associated with FEV₁/FVC, while rs10519717 of HHIP affected the severity of the disease38.

In our study, rs7671167 polymorphism of FAM13A was not associated with COPD development, pulmonary function parameters and smoking index. rs7671167 was shown to be associated with COPD in former smokers in a Chinese population9 and with COPD and lung function parameters in a population of Southern India39. The G allele of rs7671167 was shown to decrease the risk of COPD and lung cancer3. Choo et al40 demonstrated an association between the emphysematous COPD phenotype.

In conclusion, our results showed association of rs13180 (IREB2), rs16969968 (CHRNA5) and rs1051730 (CHRNA3) with COPD and pulmonary function in Tatar population from Russia. Further studies involving other ethnic groups and populations of Russia are required to verify the genetic associations detected in GWASs of COPD.