- Email: [email protected]
Deletion Polymorphism Within the 3′ Untranslated Region of COL1A2 and Chronic Venous Insufficiency
Basic Science Analysis of the Association Between an Insertion/Deletion Polymorphism Within the 30 Untranslated Region of COL1A2 and Chronic Venous Insufficiency Yiqi Jin, Guoxiong Xu, Jian Huang, Dayong Zhou, Xianchen Huang, and Liming Shen, Jiangsu, China
Background: Chronic venous insufficiency (CVI) is a common cause of leg pain and swelling and is commonly associated with varicose veins. It has significant socioeconomic consequences and is among the most common problems encountered in surgical practice. Although our current understanding of the pathogenesis of CVI is far from clear, there is a growing body of evidence suggesting a genetic contribution to the etiology of CVI. Methods: By analyzing 254 CVI cases and 508 healthy controls in a Chinese population, we used a candidate gene approach to evaluate the association between a 7ebase pair insertion/deletion (indel) polymorphism (rs3917) in the 30 untranslated region (30 UTR) of the alpha2 type I collagen gene (COL1A2) and CVI susceptibility. Logistic regression was used to analyze the association between rs3917 and CVI risk, adjusted for sex and age. Computational modeling was used to predict potential molecular mechanisms underlying the association. Results: Logistic regression analysis revealed that subjects carrying indel or deletion/deletion genotypes had a significantly increased risk for CVI than individuals carrying insertion/insertion genotypes (adjusted odds ratio, 1.64; 95% confidence interval [CI], 1.10e2.45; P ¼ 0.010). Carrying the 7ebase pair deletion allele was associated with a 1.60-fold risk for CVI (95% CI, 1.11e2.31; P ¼ 0.008). Computational modeling suggests that the rs3917 insertion allele lies within a predicted binding site (seed region) for microRNA-382 and that the deletion allele alters the affinity of microRNAemRNA binding by disrupting the local structure of COL1A2 mRNA, presumably allowing for upregulated COL1A2 expression. Conclusions: Taken together, our data suggest that common genetic variations in COL1A2 may influence CVI risk, possibly through microRNA-382emediated regulation. Replication of our studies in other populations will strengthen our understanding of this association.
INTRODUCTION Supported by grants from the Suzhou Science and Technology Fund Planning Project (no. SYSD2012102). Department of Vascular Surgery, Suzhou Municipal Hospital, Affiliated Suzhou Hospital of Nanjing Medical University, Jiangsu, China. Correspondence to: Liming Shen, MD, Department of Vascular Surgery, Suzhou Municipal Hospital, Affiliated Suzhou Hospital of Nanjing Medical University, Suzhou, Jiangsu 215002, China; E-mail: [email protected] Ann Vasc Surg 2013; 27: 959–963 http://dx.doi.org/10.1016/j.avsg.2013.04.001 Ó 2013 Elsevier Inc. All rights reserved. Manuscript received: February 4, 2013; manuscript accepted: April 3, 2013; published online: July 10, 2013.
Chronic venous insufficiency (CVI) is characterized by a functional abnormality in the venous system and is usually associated with more severe diseases including edema, skin changes, or venous ulcers.1,2 The most common manifestation of CVI is superficial venous dysfunction in the lower extremities, leading to the development of varicose veins. The prevalence of varicose veins is estimated to be 10e 15% for men and 20e25% for women in Western populations. CVI has significant socioeconomic consequences and is among the most common 959
960 Jin et al.
problems encountered in surgical practice.3,4 Epidemiologic studies have shown that there are several risk factors for CVI, including genetic predisposition, older age, standing occupations, obesity, physical activity, multiple pregnancies, and connective tissue abnormalities.5e8 The most widely accepted causes for primary varicose vein pathophysiology include primary valvular incompetence and vein wall weakness.3 Despite extensive research in recent years, our current understanding of the pathogenesis of CVI is far from clear. There is a growing body of evidence suggesting a genetic role in the etiology of CVI.9 It has been estimated that the additive genetic component is approximately 17%, which implies a strong genetic component in primary venous failure. However, the exact mechanisms by which varicosis occurs are still unclear.10 Tumor necrosis factore a block haplotypes have been shown to be associated with chronic venous leg ulcers in white patients.11 Common genetic variations in the promoter region of matrix metalloproteinase 9 (MMP-9) and tissue inhibitor of metalloproteinase 2 (TIMP-2) are associated with varicose veins in the Chinese population.12 In addition, suppressive subtractive hybridization experiments produced a profile of differentially expressed genes in varicose veins, such as matrix gla protein, which may be involved in extracellular matrix remodeling.13 The alpha-2 type I collagen (COL1A2) gene is 1 of 2 genes encoding type I collagen polypeptides that represent the major constituent of human organ systems, including blood vessels and many interstitial tissues. Type I collagen consists of 2 alpha-1 and 1 alpha-2 chains. The COL1A2 gene is located in the 7q22.1 locus and encodes the proealpha-2 chain protein. Several experimental and clinical studies suggest that varicose veins may be linked to collagen dysfunction. For instance, type I collagen mRNA in fibroblasts and smooth muscle cells was shown to be increased in human varicose veins.14e16 Recently, a 7ebase pair (bp) insertion/deletion (indel) polymorphism (rs3917) was identified in the 30 untranslated region (30 UTR) of COL1A2 and was shown to be associated with hepatocellular carcinoma susceptibility.17 As far as we know, no previous study has examined the association between the 7-bp indel polymorphism and CVI susceptibility. The present study investigates whether rs3917 modifies the occurrence of CVI in a Chinese population and explores the potential molecular mechanisms underlying this association.
Annals of Vascular Surgery
METHODS Study Populations All subjects enrolled in this study were ethnically homogenous and of Han ethnicity. CVI patients (n ¼ 254) confirmed by clinical and Doppler ultrasound examination were selected from the Department of Vascular Surgery of Suzhou Municipal Hospital between 2011 and 2012. Those patients suffering from diabetes, peripheral arterial disease, leukemia, iron deficiency anemia, hemolytic anemia, or autoimmune diseases or from severe cardiac, hepatic, renal, or pulmonary insufficiency were excluded to control for potential confounding factors. The clinical classification of patients was determined using a revised clinical, etiology, anatomy, and pathophysiology classification.18 The control group (n ¼ 508) consisted of healthy subjects lacking any evidence or history of vascular disease who were randomly selected from a communitybased screening program for noninfectious diseases that was conducted in the Jiangsu province. The control subjects were each matched to a CVI case based on age (±5 yrs) and sex. Each subject provided written informed consent and underwent an in-person interview. The interview included a standard questionnaire with demographics factors, including smoking and drinking status. After the interview, a 5-mL venous blood sample was collected from each subject. ‘‘Smokers’’ were classified as individuals who smoked almost every day for >1 yr until the time of the interview; the ‘‘nonsmokers’’ had either never smoked or seldom did. Subjects were considered ‘‘drinkers’’ if they consumed 1e2 alcohol drinks per week for >1 year. ‘‘Nondrinkers’’ either never drank or seldom did. The study design was approved by the Ethical Committee of Suzhou Municipal Hospital. DNA Extraction and Genotyping Genomic DNA was extracted from blood samples using a genomic DNA purification kit (Qiagen, Venlo, the Netherlands). The genotyping method was previously described.17 Briefly, DNA fragments containing rs3917 were amplified by polymerase chain reaction (PCR) using a pair of genotyping primers (F: 50 -CTGTGGAACCATGGAAGAAG-30 ; R: 50 -GTATTGAGTTGTATCGTGTGG-30 ). PCR products were subsequently analyzed by electrophoresis on a 7% nondenaturing polyacrylamide gel and visualized by silver staining. The rs3917 genotypes were determined by the number and positions of
Vol. 27, No. 7, October 2013
Genetic variations within COL1A2 confer CVI risk 961
Table I. Clinical characteristics of chronic venous insufficiency cases and controls Case Characteristics
N ¼ 254
Age, yr (mean ± SD) Sex Male Female Family history of VV Yes No Smoking status Smokers Nonsmokers Drinking status Drinkers Nondrinkers Clinical classification C2e3, S C4e6, S
59.1 ± 11.3
Control Frequency (%)
N ¼ 508
Frequency (%)
P value
0.95a
58.9 ± 11.5 242 266
47.6 52.3
1.00b
27.2 72.8
119 389
23.4 76.6
0.26b
84 170
33.1 66.9
178 330
35.0 65.0
0.59b
65 189
25.6 74.4
121 133
47.6 52.3
35 219
13.8 86.2
69 185
VV, varicose veins. a Two-sided 2-sample t-test between cases and controls. b 2 c test for differences between cases and controls.
the bands on the gels. The 7-bp deletion allele yielded a 168-bp band and the insertion allele yielded a 175-bp band. To validate the genotyping method, we also analyzed 20 randomly selected DNA samples by direct sequencing; the results of these 2 methods were 100% concordant. Approximately 10% of the case and control samples were randomly selected and tested in duplicate by independent technicians, resulting in 100% concordance. In Silico Prediction of Micro-RNA Binding on rs3917 Mature human miRNA sequences were obtained from the miRNA database (miRBase; available at: http://microrna.sanger.ac.uk). A 30-bp region comprising rs3917 was analyzed for hybridization of putative miRNAs using miRanda software with default parameters.19 Statistical Analysis The genotype distribution was analyzed for Hardye Weinberg equilibrium using the c2 test. Logistic regression was used to analyze the association between rs3917 genotypes and CVI risk, adjusted for both sex and age. Because of the limited number of deletion/deletion genotypes, they were integrated with the indel genotypes using a dominant model. These statistical analyses were implemented in SAS software (version 8.0; SAS Institute, Cary,
NC). P < 0.05 was considered statistically significant. All statistical tests were 2-sided.
RESULTS The characteristics of the CVI patients and controls are summarized in Table I. There were no statistically significant differences between the cases and controls in the distribution of sex, age, or smoking or drinking status. An example output from genotyping and sequencing assays of rs3917 is shown in Figure 1A and B. The genotypic frequencies of rs3917 for both cases and controls agreed with HardyeWeinberg equilibrium (P ¼ 0.068 and 0.227, respectively). Allelic and genotypic frequencies for cases and controls are shown in Table II. Under a dominant model, the heterozygous and homozygous deletion/deletion genotypes were associated with an increased risk of CVI compared with the homozygous insertion/insertion genotype (adjusted odds ratio [OR], 1.64; 95% confidence interval [CI], 1.10e2.45; P ¼ 0.010). At the allelic level, the 7-bp deletion allele was also associated with an increased risk of CVI compared with the insertion allele (OR, 1.60; 95% CI, 1.11e2.31; P ¼ 0.008). Based on bioinformatics analysis, the rs3917 insertion allele lies within a predicted binding site (seed region) for human miR-382 (Fig. 1C), suggesting that miR-382 binds tightly to COL1A2 mRNA transcripts containing the 7-bp insertion allele and negatively regulates COL1A2 expression. However,
962 Jin et al.
Annals of Vascular Surgery
Fig. 1. Genotyping and sequencing results of rs3917 and bioinformatics predictions. (A) Different genotypes analyzed using polymerase chain reaction studies. Lanes 3e7, 9, 11, and 13dhomozygous (insertion/insertion); lanes 1, 8, 10, and 12dheterozygous (insertion/deletion);
lane 2dhomozygous (deletion/deletion). (B) Sequencing results of rs3917. The underlined base pair indicates the insertion sequence. (C) The microRNAeCOL1A2 mRNA binding model and position of the deletion allele (arrow).
Table II. Associations between rs3917 genotype and chronic venous insufficiency susceptibility Genetic model
Genotype
Cases, n (%)
Controls, n (%)
OR (95% CI)a
Dominant
Ins/ins Ins/del + del/del Ins allele Del allele
198 50 + 6 446 62
432 71 + 5 935 81
1.00 1.64 1.00 1.60
Additiveb
(78.0) (22.0) (87.8) (12.2)
(85.0) (15.0) (92.0) (8.0)
(reference) (1.10e2.45) (reference) (1.11e2.31)
P value
0.010 0.008
CI, confidence interval; del, deletion; ins, insertion; OR, odds ratio. a Adjusted for sex and age. b Assuming an additive effect of the deletion allele.
the deletion allele of rs3917 could alter the affinity of the microRNAemRNA binding by disrupting the local COL1A2 mRNA structure, possibly allowing COL1A2 expressional upregulation.
DISCUSSION Using a candidate gene approach, we tested for the first time the possible association between a 7-bp indel polymorphism within the 30 UTR of COL1A2 and CVI susceptibility in a cohort of 254 CVI patients and 508 control patients from China. Our studies suggest that rs3917 is associated with CVI susceptibility, possibly through miR-382emediated regulation. Therefore, rs3917 may potentially be used as a molecular marker for the early detection of CVI. Vein wall weakness is one of the most widely accepted causes of primary varicose vein pathophysiology. Because collagen is the matrix component of veins that provides strength, veins with collagen
dysregulation may be prone to stretch and thereby turn into varicose veins. It has been shown that collagen content is significantly increased in varicose veins compared with normal veins.20,21 We hypothesized that genetic variations in COL1A2 may alter the transcriptional activity, lead to differential COL1A2 expression, and potentially contribute to a variety of pathological processes, including CVI. miRNAs are a class of short noncoding RNAs with sequence-specific posttranscriptional regulatory functions.22 Computer alignment by miRanda software predicted that the transcript bearing the insertion allele binds tightly to miR-382, thereby negatively regulating COL1A2 expression (Fig. 1C). However, binding of mRNA transcripts containing the 7-bp deletion allele would be disrupted, allowing COL1A2 expressional upregulation. This prediction is consistent with the results of our case control studies, which indicate that the deletion
Vol. 27, No. 7, October 2013
allele confers a higher risk of developing CVI. Nevertheless, how genetic variability at this locus influences COL1A2 expression still needs to be fully elucidated at both genetic and functional levels. Finally, some limitations of the study should be addressed. First, the present study lacks detailed information on the prognosis of the CVI patients, which prevents further analysis of the effect of rs3917 on CVI prognosis. Second, the association study was conducted in a Chinese population. Previous studies have shown that there are significant genetic differences in different populations for specific polymorphisms. For example, a 14-bp indel polymorphism in exon 8 of the human leucocyte antigeneG (HLA-G) gene showed marked variation among the 3 ancient tribes in China.23 Moreover, a pentanucleotide repeat polymorphism within inducible nitric oxide synthase (NOS2) promoter region also displayed highly significant differences in allelic frequencies among ethnically diverse populations.24 Therefore, we cannot exclude the possibility that rs3917 may only be applicable to this particular population group. Replication studies in other ethnic groups are necessary to fully establish the role of rs3917 in CVI and the role of other genes implicated in CVI susceptibility.
CONCLUSION In summary, we have identified a new candidate molecular marker for CVI in the Chinese population. The 7-bp indel polymorphism might be involved in the pathogenesis of CVI through hsamiR-382emediated regulation and may be a promising marker for personalized diagnosis and CVI therapy. REFERENCES 1. Rabe E, Pannier F. Clinical, aetiological, anatomical and pathological classification (CEAP): gold standard and limits. Phlebology 2012;27:114e8. 2. Beebe-Dimmer JL, Pfeifer JR, Engle JS, et al. The epidemiology of chronic venous insufficiency and varicose veins. Ann Epidemiol 2005;15:175e84. 3. Fan CM. Venous pathophysiology. Semin Intervent Radiol 2005;22:157e61. 4. Tran NT, Meissner MH. The epidemiology, pathophysiology, and natural history of chronic venous disease. Semin Vasc Surg 2002;15:5e12. 5. Oklu R, Habito R, Mayr M, et al. Pathogenesis of varicose veins. J Vasc Interv Radiol 2012;23:33e9.
Genetic variations within COL1A2 confer CVI risk 963
6. Lim CS, Davies AH. Pathogenesis of primary varicose veins. Br J Surg 2009;96:1231e42. 7. Naoum JJ, Hunter GC, Woodside KJ, et al. Current advances in the pathogenesis of varicose veins. J Surg Res 2007;141: 311e6. 8. Raffetto JD, Khalil RA. Mechanisms of varicose vein formation: valve dysfunction and wall dilation. Phlebology 2008;23:85e98. 9. Krysa J, Jones GT, van Rij AM. Evidence for a genetic role in varicose veins and chronic venous insufficiency. Phlebology 2012;27:329e35. 10. Fiebig A, Krusche P, Wolf A, et al. Heritability of chronic venous disease. Hum Genet 2010;127:669e74. 11. Tan JH, Price P, Gut I, et al. Characterization of tumor necrosis factor-a block haplotypes associated with susceptibility to chronic venous leg ulcers in Caucasian patients. Hum Immunol 2010;71:1214e9. 12. Xu HM, Zhao Y, Zhang XM, et al. Polymorphisms in MMP-9 and TIMP-2 in Chinese patients with varicose veins. J Surg Res 2011;168:e143e8. 13. Cario-Toumaniantz C, Boularan C, Schurgers LJ, et al. Identification of differentially expressed genes in human varicose veins: involvement of matrix gla protein in extracellular matrix remodeling. J Vasc Res 2007;44:444e59. 14. Sansilvestri-Morel P, Rupin A, Badier-Commander C, et al. Chronic venous insufficiency: dysregulation of collagen synthesis. Angiology 2003;54:S13e8. 15. Sansilvestri-Morel P, Rupin A, Jaisson S, et al. Synthesis of collagen is dysregulated in cultured fibroblasts derived from skin of subjects with varicose veins as it is in venous smooth muscle cells. Circulation 2002;106:479e83. 16. Sansilvestri-Morel P, Rupin A, Badier-Commander C, et al. Imbalance in the synthesis of collagen type I and collagen type III in smooth muscle cells derived from human varicose veins. J Vasc Res 2001;38:560e8. 17. Zhu Z, Jiang Y, Chen S, et al. An insertion/deletion polymorphism in the 3’ untranslated region of type I collagen a2 (COL1A2) is associated with susceptibility for hepatocellular carcinoma in a Chinese population. Cancer Genet 2011;204:265e9. 18. Ekl€ of B, Rutherford RB, Bergan JJ, et al. Revision of the CEAP classification for chronic venous disorders: consensus statement. J Vasc Surg 2004;40:1248e52. 19. John B, Enright AJ, Aravin A, et al. Human microRNA targets. PLoS Biol 2004;2:e363. 20. Gandhi RH, Irizarry E, Nackman GB, et al. Analysis of the connective tissue matrix and proteolytic activity of primary varicose veins. J Vasc Surg 1993;18:814e20. 21. Elsharawy MA, Naim MM, Abdelmaguid EM, et al. Role of saphenous vein wall in the pathogenesis of primary varicose veins. Interact Cardiovasc Thorac Surg 2007;6:219e24. 22. Pillai RS. MicroRNA function: multiple mechanisms for a tiny RNA? RNA 2005;11:1753e61. 23. Tao Y, Chen J, Yao Y, et al. Distribution of HLA-G 14-bp insertion/deletion polymorphism in six Chinese ethnic groups. Int J Immunogenet 2013;40:93e8. 24. Xu W, Humphries S, Tomita M, et al. Survey of the allelic frequency of a NOS2A promoter microsatellite in human populations: assessment of the NOS2A gene and predisposition to infectious disease. Nitric Oxide 2000;4:379e83.
Background: Chronic venous insufficiency (CVI) is a common cause of leg pain and swelling and is commonly associated with varicose veins. It has significant socioeconomic consequences and is among the most common problems encountered in surgical practice. Although our current understanding of the pathogenesis of CVI is far from clear, there is a growing body of evidence suggesting a genetic contribution to the etiology of CVI. Methods: By analyzing 254 CVI cases and 508 healthy controls in a Chinese population, we used a candidate gene approach to evaluate the association between a 7ebase pair insertion/deletion (indel) polymorphism (rs3917) in the 30 untranslated region (30 UTR) of the alpha2 type I collagen gene (COL1A2) and CVI susceptibility. Logistic regression was used to analyze the association between rs3917 and CVI risk, adjusted for sex and age. Computational modeling was used to predict potential molecular mechanisms underlying the association. Results: Logistic regression analysis revealed that subjects carrying indel or deletion/deletion genotypes had a significantly increased risk for CVI than individuals carrying insertion/insertion genotypes (adjusted odds ratio, 1.64; 95% confidence interval [CI], 1.10e2.45; P ¼ 0.010). Carrying the 7ebase pair deletion allele was associated with a 1.60-fold risk for CVI (95% CI, 1.11e2.31; P ¼ 0.008). Computational modeling suggests that the rs3917 insertion allele lies within a predicted binding site (seed region) for microRNA-382 and that the deletion allele alters the affinity of microRNAemRNA binding by disrupting the local structure of COL1A2 mRNA, presumably allowing for upregulated COL1A2 expression. Conclusions: Taken together, our data suggest that common genetic variations in COL1A2 may influence CVI risk, possibly through microRNA-382emediated regulation. Replication of our studies in other populations will strengthen our understanding of this association.
INTRODUCTION Supported by grants from the Suzhou Science and Technology Fund Planning Project (no. SYSD2012102). Department of Vascular Surgery, Suzhou Municipal Hospital, Affiliated Suzhou Hospital of Nanjing Medical University, Jiangsu, China. Correspondence to: Liming Shen, MD, Department of Vascular Surgery, Suzhou Municipal Hospital, Affiliated Suzhou Hospital of Nanjing Medical University, Suzhou, Jiangsu 215002, China; E-mail: [email protected] Ann Vasc Surg 2013; 27: 959–963 http://dx.doi.org/10.1016/j.avsg.2013.04.001 Ó 2013 Elsevier Inc. All rights reserved. Manuscript received: February 4, 2013; manuscript accepted: April 3, 2013; published online: July 10, 2013.
Chronic venous insufficiency (CVI) is characterized by a functional abnormality in the venous system and is usually associated with more severe diseases including edema, skin changes, or venous ulcers.1,2 The most common manifestation of CVI is superficial venous dysfunction in the lower extremities, leading to the development of varicose veins. The prevalence of varicose veins is estimated to be 10e 15% for men and 20e25% for women in Western populations. CVI has significant socioeconomic consequences and is among the most common 959
960 Jin et al.
problems encountered in surgical practice.3,4 Epidemiologic studies have shown that there are several risk factors for CVI, including genetic predisposition, older age, standing occupations, obesity, physical activity, multiple pregnancies, and connective tissue abnormalities.5e8 The most widely accepted causes for primary varicose vein pathophysiology include primary valvular incompetence and vein wall weakness.3 Despite extensive research in recent years, our current understanding of the pathogenesis of CVI is far from clear. There is a growing body of evidence suggesting a genetic role in the etiology of CVI.9 It has been estimated that the additive genetic component is approximately 17%, which implies a strong genetic component in primary venous failure. However, the exact mechanisms by which varicosis occurs are still unclear.10 Tumor necrosis factore a block haplotypes have been shown to be associated with chronic venous leg ulcers in white patients.11 Common genetic variations in the promoter region of matrix metalloproteinase 9 (MMP-9) and tissue inhibitor of metalloproteinase 2 (TIMP-2) are associated with varicose veins in the Chinese population.12 In addition, suppressive subtractive hybridization experiments produced a profile of differentially expressed genes in varicose veins, such as matrix gla protein, which may be involved in extracellular matrix remodeling.13 The alpha-2 type I collagen (COL1A2) gene is 1 of 2 genes encoding type I collagen polypeptides that represent the major constituent of human organ systems, including blood vessels and many interstitial tissues. Type I collagen consists of 2 alpha-1 and 1 alpha-2 chains. The COL1A2 gene is located in the 7q22.1 locus and encodes the proealpha-2 chain protein. Several experimental and clinical studies suggest that varicose veins may be linked to collagen dysfunction. For instance, type I collagen mRNA in fibroblasts and smooth muscle cells was shown to be increased in human varicose veins.14e16 Recently, a 7ebase pair (bp) insertion/deletion (indel) polymorphism (rs3917) was identified in the 30 untranslated region (30 UTR) of COL1A2 and was shown to be associated with hepatocellular carcinoma susceptibility.17 As far as we know, no previous study has examined the association between the 7-bp indel polymorphism and CVI susceptibility. The present study investigates whether rs3917 modifies the occurrence of CVI in a Chinese population and explores the potential molecular mechanisms underlying this association.
Annals of Vascular Surgery
METHODS Study Populations All subjects enrolled in this study were ethnically homogenous and of Han ethnicity. CVI patients (n ¼ 254) confirmed by clinical and Doppler ultrasound examination were selected from the Department of Vascular Surgery of Suzhou Municipal Hospital between 2011 and 2012. Those patients suffering from diabetes, peripheral arterial disease, leukemia, iron deficiency anemia, hemolytic anemia, or autoimmune diseases or from severe cardiac, hepatic, renal, or pulmonary insufficiency were excluded to control for potential confounding factors. The clinical classification of patients was determined using a revised clinical, etiology, anatomy, and pathophysiology classification.18 The control group (n ¼ 508) consisted of healthy subjects lacking any evidence or history of vascular disease who were randomly selected from a communitybased screening program for noninfectious diseases that was conducted in the Jiangsu province. The control subjects were each matched to a CVI case based on age (±5 yrs) and sex. Each subject provided written informed consent and underwent an in-person interview. The interview included a standard questionnaire with demographics factors, including smoking and drinking status. After the interview, a 5-mL venous blood sample was collected from each subject. ‘‘Smokers’’ were classified as individuals who smoked almost every day for >1 yr until the time of the interview; the ‘‘nonsmokers’’ had either never smoked or seldom did. Subjects were considered ‘‘drinkers’’ if they consumed 1e2 alcohol drinks per week for >1 year. ‘‘Nondrinkers’’ either never drank or seldom did. The study design was approved by the Ethical Committee of Suzhou Municipal Hospital. DNA Extraction and Genotyping Genomic DNA was extracted from blood samples using a genomic DNA purification kit (Qiagen, Venlo, the Netherlands). The genotyping method was previously described.17 Briefly, DNA fragments containing rs3917 were amplified by polymerase chain reaction (PCR) using a pair of genotyping primers (F: 50 -CTGTGGAACCATGGAAGAAG-30 ; R: 50 -GTATTGAGTTGTATCGTGTGG-30 ). PCR products were subsequently analyzed by electrophoresis on a 7% nondenaturing polyacrylamide gel and visualized by silver staining. The rs3917 genotypes were determined by the number and positions of
Vol. 27, No. 7, October 2013
Genetic variations within COL1A2 confer CVI risk 961
Table I. Clinical characteristics of chronic venous insufficiency cases and controls Case Characteristics
N ¼ 254
Age, yr (mean ± SD) Sex Male Female Family history of VV Yes No Smoking status Smokers Nonsmokers Drinking status Drinkers Nondrinkers Clinical classification C2e3, S C4e6, S
59.1 ± 11.3
Control Frequency (%)
N ¼ 508
Frequency (%)
P value
0.95a
58.9 ± 11.5 242 266
47.6 52.3
1.00b
27.2 72.8
119 389
23.4 76.6
0.26b
84 170
33.1 66.9
178 330
35.0 65.0
0.59b
65 189
25.6 74.4
121 133
47.6 52.3
35 219
13.8 86.2
69 185
VV, varicose veins. a Two-sided 2-sample t-test between cases and controls. b 2 c test for differences between cases and controls.
the bands on the gels. The 7-bp deletion allele yielded a 168-bp band and the insertion allele yielded a 175-bp band. To validate the genotyping method, we also analyzed 20 randomly selected DNA samples by direct sequencing; the results of these 2 methods were 100% concordant. Approximately 10% of the case and control samples were randomly selected and tested in duplicate by independent technicians, resulting in 100% concordance. In Silico Prediction of Micro-RNA Binding on rs3917 Mature human miRNA sequences were obtained from the miRNA database (miRBase; available at: http://microrna.sanger.ac.uk). A 30-bp region comprising rs3917 was analyzed for hybridization of putative miRNAs using miRanda software with default parameters.19 Statistical Analysis The genotype distribution was analyzed for Hardye Weinberg equilibrium using the c2 test. Logistic regression was used to analyze the association between rs3917 genotypes and CVI risk, adjusted for both sex and age. Because of the limited number of deletion/deletion genotypes, they were integrated with the indel genotypes using a dominant model. These statistical analyses were implemented in SAS software (version 8.0; SAS Institute, Cary,
NC). P < 0.05 was considered statistically significant. All statistical tests were 2-sided.
RESULTS The characteristics of the CVI patients and controls are summarized in Table I. There were no statistically significant differences between the cases and controls in the distribution of sex, age, or smoking or drinking status. An example output from genotyping and sequencing assays of rs3917 is shown in Figure 1A and B. The genotypic frequencies of rs3917 for both cases and controls agreed with HardyeWeinberg equilibrium (P ¼ 0.068 and 0.227, respectively). Allelic and genotypic frequencies for cases and controls are shown in Table II. Under a dominant model, the heterozygous and homozygous deletion/deletion genotypes were associated with an increased risk of CVI compared with the homozygous insertion/insertion genotype (adjusted odds ratio [OR], 1.64; 95% confidence interval [CI], 1.10e2.45; P ¼ 0.010). At the allelic level, the 7-bp deletion allele was also associated with an increased risk of CVI compared with the insertion allele (OR, 1.60; 95% CI, 1.11e2.31; P ¼ 0.008). Based on bioinformatics analysis, the rs3917 insertion allele lies within a predicted binding site (seed region) for human miR-382 (Fig. 1C), suggesting that miR-382 binds tightly to COL1A2 mRNA transcripts containing the 7-bp insertion allele and negatively regulates COL1A2 expression. However,
962 Jin et al.
Annals of Vascular Surgery
Fig. 1. Genotyping and sequencing results of rs3917 and bioinformatics predictions. (A) Different genotypes analyzed using polymerase chain reaction studies. Lanes 3e7, 9, 11, and 13dhomozygous (insertion/insertion); lanes 1, 8, 10, and 12dheterozygous (insertion/deletion);
lane 2dhomozygous (deletion/deletion). (B) Sequencing results of rs3917. The underlined base pair indicates the insertion sequence. (C) The microRNAeCOL1A2 mRNA binding model and position of the deletion allele (arrow).
Table II. Associations between rs3917 genotype and chronic venous insufficiency susceptibility Genetic model
Genotype
Cases, n (%)
Controls, n (%)
OR (95% CI)a
Dominant
Ins/ins Ins/del + del/del Ins allele Del allele
198 50 + 6 446 62
432 71 + 5 935 81
1.00 1.64 1.00 1.60
Additiveb
(78.0) (22.0) (87.8) (12.2)
(85.0) (15.0) (92.0) (8.0)
(reference) (1.10e2.45) (reference) (1.11e2.31)
P value
0.010 0.008
CI, confidence interval; del, deletion; ins, insertion; OR, odds ratio. a Adjusted for sex and age. b Assuming an additive effect of the deletion allele.
the deletion allele of rs3917 could alter the affinity of the microRNAemRNA binding by disrupting the local COL1A2 mRNA structure, possibly allowing COL1A2 expressional upregulation.
DISCUSSION Using a candidate gene approach, we tested for the first time the possible association between a 7-bp indel polymorphism within the 30 UTR of COL1A2 and CVI susceptibility in a cohort of 254 CVI patients and 508 control patients from China. Our studies suggest that rs3917 is associated with CVI susceptibility, possibly through miR-382emediated regulation. Therefore, rs3917 may potentially be used as a molecular marker for the early detection of CVI. Vein wall weakness is one of the most widely accepted causes of primary varicose vein pathophysiology. Because collagen is the matrix component of veins that provides strength, veins with collagen
dysregulation may be prone to stretch and thereby turn into varicose veins. It has been shown that collagen content is significantly increased in varicose veins compared with normal veins.20,21 We hypothesized that genetic variations in COL1A2 may alter the transcriptional activity, lead to differential COL1A2 expression, and potentially contribute to a variety of pathological processes, including CVI. miRNAs are a class of short noncoding RNAs with sequence-specific posttranscriptional regulatory functions.22 Computer alignment by miRanda software predicted that the transcript bearing the insertion allele binds tightly to miR-382, thereby negatively regulating COL1A2 expression (Fig. 1C). However, binding of mRNA transcripts containing the 7-bp deletion allele would be disrupted, allowing COL1A2 expressional upregulation. This prediction is consistent with the results of our case control studies, which indicate that the deletion
Vol. 27, No. 7, October 2013
allele confers a higher risk of developing CVI. Nevertheless, how genetic variability at this locus influences COL1A2 expression still needs to be fully elucidated at both genetic and functional levels. Finally, some limitations of the study should be addressed. First, the present study lacks detailed information on the prognosis of the CVI patients, which prevents further analysis of the effect of rs3917 on CVI prognosis. Second, the association study was conducted in a Chinese population. Previous studies have shown that there are significant genetic differences in different populations for specific polymorphisms. For example, a 14-bp indel polymorphism in exon 8 of the human leucocyte antigeneG (HLA-G) gene showed marked variation among the 3 ancient tribes in China.23 Moreover, a pentanucleotide repeat polymorphism within inducible nitric oxide synthase (NOS2) promoter region also displayed highly significant differences in allelic frequencies among ethnically diverse populations.24 Therefore, we cannot exclude the possibility that rs3917 may only be applicable to this particular population group. Replication studies in other ethnic groups are necessary to fully establish the role of rs3917 in CVI and the role of other genes implicated in CVI susceptibility.
CONCLUSION In summary, we have identified a new candidate molecular marker for CVI in the Chinese population. The 7-bp indel polymorphism might be involved in the pathogenesis of CVI through hsamiR-382emediated regulation and may be a promising marker for personalized diagnosis and CVI therapy. REFERENCES 1. Rabe E, Pannier F. Clinical, aetiological, anatomical and pathological classification (CEAP): gold standard and limits. Phlebology 2012;27:114e8. 2. Beebe-Dimmer JL, Pfeifer JR, Engle JS, et al. The epidemiology of chronic venous insufficiency and varicose veins. Ann Epidemiol 2005;15:175e84. 3. Fan CM. Venous pathophysiology. Semin Intervent Radiol 2005;22:157e61. 4. Tran NT, Meissner MH. The epidemiology, pathophysiology, and natural history of chronic venous disease. Semin Vasc Surg 2002;15:5e12. 5. Oklu R, Habito R, Mayr M, et al. Pathogenesis of varicose veins. J Vasc Interv Radiol 2012;23:33e9.
Genetic variations within COL1A2 confer CVI risk 963
6. Lim CS, Davies AH. Pathogenesis of primary varicose veins. Br J Surg 2009;96:1231e42. 7. Naoum JJ, Hunter GC, Woodside KJ, et al. Current advances in the pathogenesis of varicose veins. J Surg Res 2007;141: 311e6. 8. Raffetto JD, Khalil RA. Mechanisms of varicose vein formation: valve dysfunction and wall dilation. Phlebology 2008;23:85e98. 9. Krysa J, Jones GT, van Rij AM. Evidence for a genetic role in varicose veins and chronic venous insufficiency. Phlebology 2012;27:329e35. 10. Fiebig A, Krusche P, Wolf A, et al. Heritability of chronic venous disease. Hum Genet 2010;127:669e74. 11. Tan JH, Price P, Gut I, et al. Characterization of tumor necrosis factor-a block haplotypes associated with susceptibility to chronic venous leg ulcers in Caucasian patients. Hum Immunol 2010;71:1214e9. 12. Xu HM, Zhao Y, Zhang XM, et al. Polymorphisms in MMP-9 and TIMP-2 in Chinese patients with varicose veins. J Surg Res 2011;168:e143e8. 13. Cario-Toumaniantz C, Boularan C, Schurgers LJ, et al. Identification of differentially expressed genes in human varicose veins: involvement of matrix gla protein in extracellular matrix remodeling. J Vasc Res 2007;44:444e59. 14. Sansilvestri-Morel P, Rupin A, Badier-Commander C, et al. Chronic venous insufficiency: dysregulation of collagen synthesis. Angiology 2003;54:S13e8. 15. Sansilvestri-Morel P, Rupin A, Jaisson S, et al. Synthesis of collagen is dysregulated in cultured fibroblasts derived from skin of subjects with varicose veins as it is in venous smooth muscle cells. Circulation 2002;106:479e83. 16. Sansilvestri-Morel P, Rupin A, Badier-Commander C, et al. Imbalance in the synthesis of collagen type I and collagen type III in smooth muscle cells derived from human varicose veins. J Vasc Res 2001;38:560e8. 17. Zhu Z, Jiang Y, Chen S, et al. An insertion/deletion polymorphism in the 3’ untranslated region of type I collagen a2 (COL1A2) is associated with susceptibility for hepatocellular carcinoma in a Chinese population. Cancer Genet 2011;204:265e9. 18. Ekl€ of B, Rutherford RB, Bergan JJ, et al. Revision of the CEAP classification for chronic venous disorders: consensus statement. J Vasc Surg 2004;40:1248e52. 19. John B, Enright AJ, Aravin A, et al. Human microRNA targets. PLoS Biol 2004;2:e363. 20. Gandhi RH, Irizarry E, Nackman GB, et al. Analysis of the connective tissue matrix and proteolytic activity of primary varicose veins. J Vasc Surg 1993;18:814e20. 21. Elsharawy MA, Naim MM, Abdelmaguid EM, et al. Role of saphenous vein wall in the pathogenesis of primary varicose veins. Interact Cardiovasc Thorac Surg 2007;6:219e24. 22. Pillai RS. MicroRNA function: multiple mechanisms for a tiny RNA? RNA 2005;11:1753e61. 23. Tao Y, Chen J, Yao Y, et al. Distribution of HLA-G 14-bp insertion/deletion polymorphism in six Chinese ethnic groups. Int J Immunogenet 2013;40:93e8. 24. Xu W, Humphries S, Tomita M, et al. Survey of the allelic frequency of a NOS2A promoter microsatellite in human populations: assessment of the NOS2A gene and predisposition to infectious disease. Nitric Oxide 2000;4:379e83.