Background
Chronic obstructive pulmonary disease (COPD) is a complex genetic disorder that is characterized by a reduction in lung function with airflow obstruction [
1]. Although cigarette smoking is a common risk factor for COPD, cigarette smoking can differentially affect lung function, and not all cigarette smokers develop COPD [
2‐
4]. The response to cigarette smoking as well as other environmental factors is significantly influenced by a complex array of genetic factors [
1,
5]. Four large-scale genome-wide association studies have been performed in multiple populations and successfully identified numerous genetic variants consistently associated with COPD (Additional file
1: Table S1) [
6‐
9]. In addition to the GWAS, candidate gene approaches also significantly contributed to the identification of risk factors of COPD in the individual cohort of patients [
10‐
12].
The
FGF7 gene encodes keratinocyte growth factor (KGF), a member of the FGF family that are involved in various biological processes, including embryonic development, morphogenesis, cell growth, tumor growth, and tissue repair [
13,
14]. Recent studies have demonstrated a significant association of genetic variants at the
FGF7 gene in COPD patients of Spanish, Native American, Norwegian (2940 cases and 1380 controls in total, rs12591300 and rs4480740) [
10], and Chinese Han (279 cases and 367 controls in total, rs10519225) ancestry [
12]. Due to the relatively small sample size in the Chinese Han study, further evaluation of the genetic association of the
FGF7 gene in an independent cohort of Chinese COPD is needed.
KGF, encoded by the
FGF7 gene, is mainly related to the repair of the lung, and that is mostly due to their capacity to stimulate alveolar and bronchial epithelial cell proliferation [
15,
16]. Although the potential role of
FGF7 in influencing the risk of COPD is poorly understood, functional studies have been performed to investigate gene expression abnormalities of the
FGF7 in patients with COPD [
17]. A study showed that the KGF levels were not notably different between patients with COPD and healthy controls in bronchoalveolar lavage (BAL) fluid or in serum, which may be due to the limitation of the KGF detection method used in the samples [
17]. Also, studies on the role of human recombinant KGF in modulating lung function have also been conducted in cell-based assays and mouse models. The expression of KGF increases after lung injury in humans and minimizes lung injury in experimental animals [
18,
19]. These data further suggested an essential role of fibroblast growth factor signaling as well as the KGF protein in the development and the treatment of COPD [
14,
15,
18,
20,
21].
Human genetic variations and epigenetic mechanisms play a critical role in regulating the expression of the FGF7 gene. Further assessment of genetic association and mechanistic characterization of the COPD-associated functional variants of the FGF7 gene are critical steps to understand the disease mechanisms. In the current study, therefore, we employed a combination of approaches, including bioinformatic analysis of candidate functional variants, functional evaluation of transcription factor binding of variant by electrophoretic mobility shift assay (EMSA), gene expression assays of FGF7 using real-time quantitative polymerase chain reaction (RT-qPCR), chromatin conformation capture followed by RT-qPCR (3C-qPCR), and luciferase enhancer activity assays to characterize the COPD-associated candidate causal variant. The current study provides significant insight into the functional variants of the FGF7 gene in influencing risk for COPD.
Methods
Subjects
In this study, a total of 258 patients with COPD and 311 matched non-COPD population controls were enrolled. The control subjects were healthy donors with Chinese Han descent. The subjects in both groups were self-reported Chinese Han individuals and were recruited from Wenzhou in the Zhejiang Province of China. Each subject was interviewed by interviewers who collected the patients’ demographic data and information related to risk factors, such as smoking cigarettes. The clinical analyses were performed at Wenzhou Medical University; diagnosis of patients was performed according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria. All patients were subjected for clinical testing and evaluation of genetic variations. Patients were excluded from this study if they had other respiratory diseases (bronchial asthma, bronchiectasis, cancer, or pulmonary tuberculosis). Control subjects were excluded if they had a history of lung disease, atopy, an acute pulmonary infection in the 4 weeks before the enrolment for this study. This study was approved by Institutional Review Board of Jilin Agricultural University and Ethics Committee of the Second Affiliated Hospital, Wenzhou Medical University. The investigator explained the purpose and risks of the study and provided the subject with a copy of the information sheet.
SNP selection
Based on the LD and haplotype block analysis using HaploReg 4, we included 68 SNPs in LD with the three tag SNPs with an r2 0.8 for an initial screening. Further bioinformatic analyses of the list of SNPs were performed to exclude predicted SNP motif changes less than 5. By applying the filter, 17 SNPs remained for further analysis. Because a SNP rs12905203 is expected to change 19 binding motifs for multiple transcription factors, we included this candidate functional variant along with the reported Asian COPD-associated variant for this genetic association study.
SNP genotyping
Isolation of genomic DNA from PBMCs was performed using a DNA extraction and purification kit (TAKARA, Dalian, China), according to the manufacturer’s instructions. Sequences of the PCR primers are listed in Additional file
3: Table S3. The genotypes at selected SNPs for each genomic DNA sample with PCR amplified target regions were determined using a TaqMan SNP genotyping assay kit (Thermo Fisher Scientific Inc., Waltham, MA, USA). The assay followed the protocol in the manual, and PCR reactions were conducted on an OpenArray™ real-time PCR instrument (Applied Biosystems™, Foster City, CA, USA).
Statistical analysis
The demographic and clinical data between the COPD patients and the controls were analyzed using the chi-square test and Student’s t-test. Hardy-Weinberg equilibrium was calculated with a goodness of fit chi-squared test to assess the observed and expected genotype frequencies. The differences between the COPD patients and the control subjects in the context of the genotypes were analyzed using one-way analysis of variance and logistic regression for multivariate analysis. Age, sex, and smoking were used as covariates in the multivariate analyses. Statistical analyses were conducted using SPSS version 21.0 and Microsoft Excel.
Antibodies, plasmid DNAs, and cell lines
Expression constructs expressing human c-Fos (pLX304-FOS-V5) and c-Jun (pCLXSN-c-JUN) were purchased from Addgene (Addgene Headquarters, Cambridge, MA, USA). Luciferase activity assay backbone DNA pGLuc-mini-TK was purchased from New England Biolabs (Ipswich, MA, USA). Anti-KGF, anti-c-Fos, and anti-c-Jun antibodies were purchased from Abcam (Abcam, Inc. Shanghai, China). Sixty-nine skin fibroblast cell lines were established from healthy Chinese donors.
RNA isolation and quantitative RT-PCR
Total RNA from fibroblast cell lines was isolated using a Total RNA isolation and purification kit (Invitrogen Inc., Carlsbad, CA, USA) according to the manufacturer’s instructions. cDNA from each sample was synthesized using iScript cDNA Synthesis Kits (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Quantitative RT-PCR assays were performed using Power SYBR Green Master Mixture (Applied Biosystems™, Foster City, CA, USA) to determine the mRNA expression of the FGF7 gene in each sample, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as the internal control. Fold changes were calculated according to the ΔΔCT method.
Electrophoretic mobility shift assays
Fifty base pair non-risk or risk DNA probes were synthesized and end-labeled with a Biotin-DNA labeling kit (Thermo-Fisher). Nuclear protein extracts were prepared from fibroblast cells transfected with plasmid DNA expressing c-Fos, c-Jun, or control vector. A portion of the cells was stimulated with PMA/Ionomycin (50 ng/ml, 500 ng/ml) for 2 h. Incubation of nuclear proteins with labeled probes was performed for 25 min at 37 °C in the binding buffer (1 μg poly dI-dC, 20 mM HEPES, 10% Glycerol, 100 mM KCl, and 0.2 mM EDTA, pH 7.9). DNA-protein complexes were resolved on non-denaturing acrylamide gels.
ChIP and qPCR
ChIP assays were performed using the Millipore Magna ChIP A kit (Millipore, Billerica, CA, USA) according to the manufacturer’s instructions. The chromatin-protein complexes were immune-precipitated by antibodies specific for c-Fos, c-Jun, or rabbit IgG (negative control) overnight at 4 °C. DNA was eluted from the immune-precipitated chromatin complexes, reverse-crosslinked, purified by a DNA purification kit (TAKARA, Dalian, China), and subjected to real-time PCR analysis. Sequences of primers are listed in the Additional file
4: Table S4
Luciferase assay
We cloned 200 bp of DNA sequence surrounding the variant rs12905203 into a minimal TK promoter luciferase plasmid. Each plasmid was transiently co-transfected with a pGL3-promoter transfection efficiency control plasmid for normalization. Luciferase assays were performed in fibroblast cells. Forty eight hours post-transfection, cells were treated with 50 ng/ml PMA/500 ng/ml Ionomycin for additional 48 h. The cells were lysed using a lysis butter provided by the Kit, and the luciferase activity was measured using the Luciferase Assay System (Promega, Madison, MA, USA).
Discussion
In this report, we used a combination of approaches, including genetic association testing and bioinformatic analysis to fine-map the COPD-associated variants in
FGF7 locus. Three SNPs have shown significant associations of the
FGF7 in COPD (rs4480740, rs12591300, and rs10519225) [
10,
12]. We evaluated SNPs at the
FGF7 locus that are in LD with the reported SNPs and selected candidate functional SNPs for genetic association testing. We demonstrated that multiple variants were significantly associated with Chinese COPD while the most significant associated SNP is rs12905203, located upstream of the promoter of the
FGF7 gene. To our knowledge, this is the second report on the genetic association between the
FGF7 gene and COPD in Chinese Han. In this study, we showed that previously reported COPD-associated risk variants are in strong LD with the functional variant rs12905203. These data suggested a single effect at the FGF7 locus is associated with COPD in Chinese Han. A drawback of our study is of the failure to detect genetic associations of SNPs those were not in LD with the reported tag-SNPs, while there might be additional independent association signals. Additionally, significant associations of the
FGF7 gene with COPD in this study might prove to be falsely positive due to the relatively small sample size, but even with a larger sample, functional characterization of the variant would be required.
The
FGF7 gene, which encodes KGF, was initially identified in cultured human embryonic lung fibroblasts [
10,
16]. The gene plays a vital role in protecting airway epithelium from oxidant injury that is potentially related to the pathogenesis of COPD [
10,
16]. Genetic alterations of the
FGF7 gene could affect the expression of the gene or function of the encoded protein, KGF. Previous studies demonstrated that FGF7 gene plays an important role in protecting against oxidative stress response. Increases in
FGF7 expression associated with disease severity may indicate a higher burden of injury [
18,
19,
25]. We hypothesize that the risk G allele of the functional variant resulted in a decreased expression of
FGF7 and that might influence antioxidant mechanisms protecting against deleterious effects of smoking on the lung. Furthermore, it was unclear whether the
FGF7 gene plays a role in disease susceptibility through its role in epithelial development by influencing epithelial responses to cigarette smoke [
2,
3,
13,
25]. Because it is unclear whether increased FGF7 expression is a marker of exposure to oxidant injury or a cause of epithelial damage, we further characterized the role of candidate functional SNP rs12905203 on FGF7 expression.
The most significant association variant lies immediately upstream of
FGF7 in a putative enhancer element. The variant results in a substitution of the consensus binding sequences for multiple transcription factors, including AP-1 [
22]. We functionally assessed binding of the variant to AP-1 transcription factors and demonstrated that the COPD-associated risk G allele significantly reduces the binding that is accompanied by a reduction of mRNA expression of the FGF7 gene in fibroblast cells. Stimulation of the fibroblast cells with PMA/Ionomycin or overexpression of AP-1 transcription factors significantly increases the expression of
FGF7 gene, suggesting a role of the AP-1 transcription factors in regulating the transcription of FGF7.
In conclusion, we investigated the association between the FGF7 variants and the risk of COPD in the Chinese Han population. The variant rs12905203 resides in an enhancer element that binds c-Fos, and c-Jun transcription factors promote the expression of FGF7. Impaired binding of c-Fos and c-Jun proteins to the risk allele of the rs12905203 inhibits the activity of the enhancer, resulting in reduced KGF expression. These results provide genetic and functional evidence supporting a causal role for the variant rs12905203 in the genetic predisposition to COPD.
Acknowledgments
We thank the COPD patients and the matched controls that participated in this study. We are grateful to the research coordinators and physicians that helped in the recruitment of all participants.