Abstract
Objective
To investigate polymorphisms of the vascular endothelial growth factor (VEGF) gene in patients with rheumatoid arthritis (RA) and their relationship to clinical features.
Method
A total of 198 unrelated Chinese individuals were enrolled in this study, including 98 patients with RA and 100 healthy controls. Eight different polymorphisms of the VEGF gene were analyzed using Sequenom MassArray platform.
Result
All 8 polymorphisms were in Hardy-Weinberg equilibrium in controls. The frequencies of rs833070 A allele and rs325010 C allele were elevated in the patients with RA compared with the controls. There were increased genotype frequencies in GA of rs833070, GC of rs3025030, CT of rs3025039 and decreased genotype frequencies in GG of rs833070, GG of rs3025030, CC of rs3025039 in the patients with RA compared with the controls. The frequencies of haplotype GA in rs2010963 and rs833070 were higher in the patients with RA than in the controls. There was no significant difference in the genotype or allele frequencies in the RA group sorted by complications, serum markers, or age of onset.
Conclusion
Our data suggested a trend of association between VEGF gene polymorphisms and RA, and patients who carried the haplotype GA of rs2010963 and rs833070 were more susceptible to RA. Our study was performed in a small population, and further studies in other populations are needed to confirm these results.
Background
Rheumatoid arthritis (RA) is a chronic and complex autoimmune disease and a common inflammatory disorder with complicated etiology. Although the causes of RA are still unclear, it is believed that both environmental and genetic factors may play important roles in RA. Many genes, such as tumor necrosis factor receptor-associated factor 1, complement component 5, cytotoxic T-lymphocyte-associated protein 4, and peptidyl arginine deiminase type IV have been found to be correlated with the susceptibility of RA.1–3
Angiogenesis is an essential process in the proliferative synovitis, and abundant new vessels are found within hyperplastic synovial tissue.4,5 Vascular endothelial growth factor (VEGF), a potent endothelial cell-specific angiogenic factor, possesses a function of angiogenesis and increasing vascular permeability. Vascular endothelial growth factor may play complicated roles in the development of RA.
It has been reported that serum VEGF concentration was significantly elevated in RA groups compared to controls, and evidence also illustrated that serum VEGF concentration was correlated with disease activity and inflammation markers such as erythrocyte sedimentation rate and C-reactive protein and even associated with destructive changes.6–8
Abundant data revealed that VEGF gene single-nucleotide polymorphisms (SNPs) were associated with abnormal production of VEGF protein in many diseases such as various cancers and renal diseases.9–11 However, the association of VEGF gene SNPs with RA was not well identified and controversial. Two promoter polymorphisms of VEGF have been investigated in a Spanish population and may not play a relevant role in RA pathogenesis in Spanish population. 12 Intriguingly, Han et al. 13 investigated the same 2 polymorphisms and 2 additional ones in Korean population, and their data suggested that VEGF gene might may play a role in the development of RA. It was indicated that the association of VEGF SNPs with RA might depend on racial background, and genetic susceptibility to RA might be different in different populations. To detect the association of VEGF SNPs with RA in Chinese population, we used Sequenom chip-based matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry platform to analyze 8 VEGF gene SNPs, which localized in noncoding regions (rs2010963 in 5′untranslated regions; rs833070, rs3024994, rs3025000, rs3025010, rs3025030, and rs3025035 in intron region; rs3025039 in 3′untranslated regions).
Materials and Methods
Study Population
In the present study, 198 unrelated Chinese individuals, 98 patients and 100 controls, were enrolled. The 98 patients with RA, whose condition was diagnosed by the 1987 revised classification criteria of the American College of Rheumatology, 14 were recruited from China Medical University from December 2008 to August 2009, whereas patients who had tumors, diabetes, high blood pressure, and fatty liver were rejected. The control group of 100 unrelated healthy individuals who had no known medical problems on a health screening questionnaire was enrolled. All individuals gave informed consent for study participation. The patients with RA included 18 men and 80 women aged 32 to 68 years (mean, 40.3 years), and the mean duration of the disease was 3 years. The control group included 22 men and 78 women aged 25 to 55 years (mean, 39.0 years). The condition of the patients of RA complicated with interstitial lung disease was diagnosed by typical clinical and radiological signs, 15 and the condition of those with SS was diagnosed by revised international classification criteria for SS. 16
Genotyping of the VEGF Gene Polymorphisms
Genomic DNA was isolated from peripheral blood mononuclear cells using TIANamp Blood DNA Kit according to the manufacturer's instructions. Eight polymorphism loci were selected, and primers were designed by MassARRAY Assay Design 3.1. The polymerase chain reaction (PCR) primers for rs2010963, rs833070, rs3024994, rs3025000, rs3025010, rs3025030, rs3025035, and rs3025039 were 5′-ACGTTGGATGAAGTCGAGGAAGAGAGAGA-3′ (forward) and 5′-ACGTTGGATGAGCAGGTCACTCACTTTGCC-3′ (reverse); 5′-ACGTTGGATGAAGTTCACAGCACCCGAACA-3′ (forward) and 5′-ACGTTGGATGCCCTGGTTTGCATTCCTTTG-3′ (reverse); 5′-ACGTTGGATGTCAGACTTCTAGTCTCGTTC-3′ (forward) and 5′-ACGTTGGATGATGGGCACAGAATCCTTCTC-3′ (reverse); 5′-ACGTTGGATGTTGTCCCATCTGGGTATGGC-3′ (forward) and 5′-ACGTTGGATGGGTTTGATCCGCATAATCTG-3′ (reverse); 5′-ACGTTGGATGCTTTCTTCCCTGTGACAGAC-3′ (forward) and 5′-ACGTTGGATGCCCTTCAAGAGAACCAGAGC-3′ (reverse); 5′-ACGTTGGATGAAAATGTGTGGGCTGCTTGG-3′ (forward) and 5′-ACGTTGGATGACACACTGAAGGAGCTGTAG-3′ (reverse); 5′-ACGTTGGATGGGTTTGTGTGAAGTGACCTG-3′ (forward) and 5′-ACGTTGGATGTATTCCCAGATACAGCCAGC-3′ (reverse); 5′-ACGTTGGATGACTCTGCGCAGAGCACTTTG-3′ (forward); and 5′-ACGTTGGATGATGGCGAATCCAATTCCAAG-3′ (reverse).
Single-nucleotide polymorphism genotyping was performed by Shanghai Benegene Biotechnology Co, Ltd using MassARRAY system (Sequenom, San Diego, CA) via the MALDI-TOF mass spectrometry method according to the manufacturer's instructions. Briefly, the DNA samples to be examined were diluted to 5 ng/µL, and 1 µL of DNA was combined with 0.95 µL of water, 0.625 µL of PCR buffer containing 15-mmol/L MgCl2, 1 µL of 2.5 mmol/L deoxy-ribonucleoside triphosphate, 0.325 µL of 25 mmol/L MgCl2, 1 µL of PCR primers, and 0.1 µL of 5 units/µL HotStar Taq (Qiagen, Shanghai, China). The reaction was incubated at 94°C for 15 minutes followed by 45 cycles at 94°C for 20 seconds, 56°C for 30 seconds, 72°C for 1 minute, and a final incubation at 72°C for 3 minutes. After PCR amplification, remaining deoxy-ribonucleoside triphosphates were dephosphorylated by adding 1.53 µL of water, 0.17 µL of SAP buffer, and 0.3 units of shrimp alkaline phosphatase (Sequenom). The reaction was placed at 37°C for 40 minutes, and the enzyme was deactivated by incubating at 85°C for 5 minutes. After shrimp alkaline phosphatase treatment, the single primer extension over the SNP was combined with 0.755 µL of water, 0.2 µL of 10× iPLEX buffer, 0.2 µL of termination mix, 0.041 µL of iPLEX enzyme (Sequenom), and 0.804 µL of 10-umol/L extension primer. The single-base extension reaction was carried out at 94°C for 30 seconds and then 94°C for 5 seconds, followed by 5 cycles of 52°C for 5 seconds and 80°C for 5 seconds, total 40 cycles, then 72°C for 3 minutes. The reaction mix was desalted by adding 6 µg of cation exchange resin (Sequenom), mixed, and resuspended in 25 µL of water. The completed genotyping reactions were spotted onto a 384-well spectroCHIP (Sequenom) using MassARRAY Nanodispenser (Sequenom) and determined by the MALDI-TOF mass spectrometer. Genotype calling was performed in real time with MassARRAY RT software version 3.0.0.4 and analyzed using the MassARRAY Typer software version 3.4 (Sequenom).
Statistical Analysis
Hardy-Weinberg equilibrium for each polymorphism was analyzed using the Fisher exact test. The differences in genotype distribution and allele frequency among the groups were examined for statistical significance by the χ2 test and the Fisher exact test where necessary. Haplotypes were determined using the Haploview 4.0 software (Broad Institute of Massachusetts Institute of Technology and Harvard, Cambridge, MA). When typed by complications, serum markers, or age of onset, the genotype distribution and the allele frequency among groups were also examined using the χ2 test and the Fisher exact test where necessary. A P < 0.05 was considered statistically significant. Bonferroni correction was applied when appropriate. Analysis was conducted using the Haploview 4.0 and SPSS 11 softwares.
Results
Coincidence rates of Hardy-Weinberg equilibrium were measured within 8 SNPs of control group using the Fisher exact test. It revealed that observed values were consistent with expected value within 8 SNPs, which did not show a significant deviation from the Hardy-Weinberg equilibrium (P > 0.05; Table 1).
Hardy-Weinberg Equilibrium Coincidence Rate Test of All SNPs of Controls
The distribution of genotypes and allele frequencies is shown in Table 2. We compared the differences in genotype distribution and allele frequencies among groups and observed that both genotypes and allele frequencies of the rs2010963, rs3024994, rs3025000, and rs3025035 were not significantly different between the RA group and the controls (P > 0.05).
Vascular Endothelial Growth Factor Genotype Distributions (%) and Allele Frequencies in Patients With RA and Controls
Allele frequencies within rs833070 and rs3025010 were obviously different between the RA group and the controls; carriers of the A allele in rs833070 and C allele in rs3025010 were more frequent in the RA group compared with the controls (33.2% vs 21.5%, odds ratio [OR] = 1.756, P < 0.05; 37.2% vs 28%, OR = 1.526, P = 0.05). In addition, rs833070 GA genotype frequency was higher in the patients with RA than in the controls (GA vs GG+AA, 45.9% vs 31%; OR = 2.178; P < 0.05), whereas genotype GG frequency was lower (GG vs GA+AA, 43.9% vs 63%; OR = 2.178; P < 0.05). We also observed that the genotype GC frequency in rs3025030 was higher in the patients with RA compared with the controls (GC vs GG+CC, 38.8% vs 20%; OR = 2.047; P < 0.05), whereas genotype GG was lower (GG vs GC+CC, 58.2% vs 74%; OR = 2.047; P < 0.05). Meanwhile, the genotype CT in rs3025039 was more frequent in the RA group compared with the controls (CT vs CC+TT, 39.8% vs 21%; OR = 2.047; P < 0.05), whereas genotype CC was less (CC vs CT+TT, 58.2% vs 74%; OR = 2.047; P < 0.05).
Haplotype frequencies of 8 SNPs of VEGF biallelic polymorphisms were determined by Haploview 4.0. Phase has been estimated to correctly reconstruct the haplotypes. 17 Haplotyping was performed, and 2 blocks of rs2010963 and rs833070, rs3025000 and rs3025010 were proposed (Fig. 1), and the strength of linkage disequilibrium between SNPs was presented as values of D′. Among the 6 haplotypes, haplotype GA of rs2010963 and rs833070 was exclusively more frequent in the RA group compared with the controls (32.5% vs 21.4%, P < 0.05; Table 3).
Block of linkage disequilibrium of SNPs constructed by Haploview 4.0.
Distribution of Haplotypes Within rs2010963 and rs833070, rs3025000, and rs3025010 as Proposed by Haploview 4.0
In the patients with RA, 4 SNPs, rs833070, rs3025010, rs3025030 and rs3025039, which might be related to susceptibility of RA, were used to perform genotype distribution and allele frequency analysis, classifying by complication with interstitial lung disease, erosivity, or SS and serum marker of anti-cyclic citrullinated peptide antibodies(+) or rheumatoid factor(+). We found that only allele rs833070 was associated with erosion detected by x-ray. There was no association between smoking and polymorphisms. The results showed that there was no statistical significance when performing intergroup comparisons (Table 4). When RA was sorted by age of onset and divided into 2 groups, 50 years or younger and older than 50 years, no statistical significance of genotype distribution and allele frequency was observed either (Table 5).
Vascular Endothelial Growth Factor Genotype Distributions and Allele Frequencies in Patients With RA Sorted by Complications and Serum Markers
Vascular Endothelial Growth Factor Genotype Distributions and Allele Frequencies in Patients With RA Sorted by Age of Onset
Discussion
Vascular endothelial growth factor is the only mitogen that specifically acts on endothelial cells and has been shown to stimulate the elongation, network formation, and branching of nonproliferative endothelial cells in culture that are deprived of oxygen and nutrients. 18 Recently, numerous data suggested that VEGF might play a crucial role in RA and might be a marker in the progress of RA. High levels of the cytokine VEGF have been found in the synovial joints of both early and established RA. 19 Maeno et al. 20 investigated peripheral blood VEGF levels in the polyarticular juvenile RA (JRA) and found that VEGF levels correlated with the number and severity of injured joints.
Vascular endothelial growth factor messenger RNA (mRNA) and protein expression have been localized to the lining layer and endothelial cells in rheumatoid synovial tissue.21,22 Vascular endothelial growth factor might play a complex reciprocal causative role in the development of RA. Once the autoimmune process has been initiated in the early stage of RA, inflammation within the joint structure may induce relative hypoxia of synovial tissue, which promotes the synthesis of VEGF from rheumatoid synovial tissue. 23 Both hypoxia and VEGF may aggravate tissue injury by increase in vascular permeability and acceleration of the inflammatory process, which may be crucial in the initiation of RA. 12
Han et al. 13 observed an association of the T allele with the development of RA. Previous studies have shown that VEGF plasma level was lower in carriers of the rs3025039 T allele than in noncarriers, whereas rs3025039 C allele was one of the core sequences for the potential binding of papillomavirus regulator E2, and the C to T change at position rs3025039 resulted in the loss of the core-binding sequence for this transcription factor, which provided a probably analogical interaction with the 5′-untranslated regions-binding proteins of VEGF gene and played a role in the transcriptional regulation.24–27
Our results were similar to the study by Han to some extent. Higher frequencies of the haplotype CT at rs3025039 might contribute to lower VEGF protein expression. It is contrary to the high levels of VEGF protein in RA, and genotype difference in the other SNPs might work. We also found that rs833070 GA and rs3025030 GC genotype frequencies were higher, whereas rs833070 GG and rs3025030 GG genotype frequencies were lower in the patients with RA compared to the controls. Our data also revealed that the frequencies of rs833070 A allele and RS325010 C allele were increased in the patients with RA, which might account for high levels of VEGF protein in RA. Interestingly, in our another study about VEGF polymorphisms and systemic lupus erythematosus (SLE), we observed that rs833070 A allele was more frequent in patients with SLE than in healthy controls, and rs833070 A allele frequency in SLE complicated with RA was the same in the patients with SLE without RA. These findings suggested that the role of rs833070 polymorphism should not be ignored in many autoimmune system diseases including RA and SLE.
Despite our interesting and novel findings in the present study, some limitations of our study should be stated. Our study was performed in a small population (98 patients and 100 controls), which represented a major limitation of our study. In this case, it allowed the detection of the expected effect sizes with a power of only approximately 35% and might be underpowered to perform any stratification analysis. After multiple testing (Bonferroni correction), most SNPs might not be statistically associated with RA. Further studies in a big population are needed to confirm our findings.
However, our study at least showed a trend of association between 8 new VEGF SNPs and RA in Chinese population and revealed novel linkage disequilibrium relationships between these SNPs. Our approach might provide a new understanding on the pathogenesis of RA and add evidence that the association of VEGF with RA might be different in different racial populations, which might be useful for investigators in this field to perform in-depth analysis and might provide potential basis for future cure of RA in different racial backgrounds.