Funding sources National Natural Science Foundation of China (no. 81130032, no. 81172749 and no. 81101189) and the Xijing Hospital Support Fund (no. XJZT10Z05).
Conflicts of interest None declared.
J.C., Q.S. and X.W. contributed equally to this work.
Recent evidence has revealed an elevation of total homocysteine (tHcy) in patients with vitiligo. Methylenetetrahydrofolate reductase (MTHFR) is one of the main enzymes regulating homocysteine (Hcy) metabolism. Thus, polymorphisms of MTHFR could potentially contribute to the development of vitiligo by affecting MTHFR activity and tHcy levels.
To evaluate the potential association between MTHFR polymorphisms and vitiligo susceptibility.
In total, 1000 patients with vitiligo and 1000 age- and sex-matched controls were enrolled in this hospital-based case–control study. Two single-nucleotide polymorphisms of the MTHFR gene (rs1801133 C>T and rs1801131 A>C) were selected and genotyped using polymerase chain reaction (PCR)–restriction fragment length polymorphism and allele-specific PCR, respectively. The MTHFR activity concentration and tHcy level in serum were measured by enzyme-linked immunosorbent assay.
We found that allele T of rs1801133 in the MTHFR gene was associated with a significantly reduced risk of vitiligo (adjusted odds ratio 0·58, 95% confidence interval 0·43–0·76, P <0·001). In addition, the patients with vitiligo had a lower activity concentration of MTHFR and higher level of tHcy than the controls. Correlation between these markers and the risk of vitiligo was also observed. Furthermore, the individuals with a no-risk genotype (CT + TT) of rs1801133 and higher activity concentration of MTHFR or lower level of tHcy had a significantly decreased risk of vitiligo.
Our data suggest that MTHFR gene polymorphisms may play a vital role in genetic susceptibility to vitiligo.
Vitiligo is an acquired pigmentary disorder of the skin characterized by loss of functional epidermal melanocytes and the presence of white macules. The prevalence rate of this disease varies from 0·1% to 2·0% among different ethnic groups worldwide, without predilection for sex. Although various hypotheses have been proposed, the aetiology of vitiligo is not clear.[2-6] Recently, accumulating studies have suggested that oxidative stress caused by reactive oxygen species (ROS) may be the triggering event in the melanocyte impairment seen in vitiligo.[7-13] An increase in ROS has been found in the epidermis of patients with vitiligo, and it plays a key role in the onset and progression of melanocyte damage. It has been suggested that excess ROS production by metabolic deregulation leads to toxic damage of melanocytes, and this process is involved in vitiligo.
Methylenetetrahydrofolate reductase (MTHFR) is the main regulatory enzyme for homocysteine (Hcy) metabolism. MTHFR catalyses reduction of 5,10- methylenetetrahydrofolate to 5-methyltetrahydrofolate. 5-Methyltetrahydrofolate acts as a methyl donor during synthesis of methionine from Hcy, which is an intermediary amino acid formed during the conversion of methionine to cysteine. Consequently, the level and function of MTHFR is vital and indispensible to maintain Hcy homeostasis. It has been reported that Hcy with abnormal metabolism and quantity or function is associated with many diseases, including vitiligo. In clinical studies, recent evidence demonstrated that the level of total Hcy (tHcy) was significantly elevated in patients with vitiligo, and was more obvious in those patients in the active stage of vitiligo.[2, 17-19] The abnormal Hcy level was found to be significantly associated with distribution and bilaterality of lesions in vitiligo.[17-19] Additionally, depigmentation of the skin and hair was observed in patients with homocystinuria, which was described as ‘pigmentary dilution’. The studies mentioned above strongly suggest an association between the abnormality of Hcy and depigmentation diseases, especially vitiligo. Further molecular studies demonstrated that oxidation of tHcy produced ROS, which caused oxidative stress in melanocytes. In vitro and in vivo studies showed that elevated Hcy induced apoptosis by downregulating antioxidant proteins such as glutathione peroxidase 1 and catalase, and increasing ROS generation via activation of p38 mitogen-activated protein kinase.[22-24] Thus, it is easy to speculate that abnormal Hcy modulated by MTHFR may contribute to melanocyte apoptosis in vitiligo through induction of oxidative stress.
The human MTHFR gene is mapped to chromosome 1p36·3 and consists of 11 exons and 10 introns (Figure S1a; see Supporting Information). It has been reported that two common polymorphisms (rs1801133 and rs1801131) lead to missense mutations, which may result in a change of MTHFR enzyme activity and influence the level of Hcy.[26-28] The polymorphism rs1801133 is a C>T missense mutation in exon 4 of MTHFR and results in an alanine-to-valine change at codon position 222. Another missense mutation, rs1801131 (A>C), located in exon 7, induces an amino acid change from glutamic acid to alanine. These two polymorphisms in MTHFR have been implicated in several diseases;[29-31] thus, it is possible that functional polymorphisms in MTHFR may be associated with vitiligo risk.
Taking all of these considerations together, we hypothesized that the missense mutations of MTHFR may have an impact on MTHFR expression or enzyme activity, and further influence Hcy homeostasis, which may trigger oxidative stress in melanocytes and contribute to vitiligo risk. In order to test this hypothesis, we performed genotyping of rs1801133 and rs1801131 in MTHFR, to determine the clinical significance of MTHFR activity and Hcy level in patients with vitiligo in our hospital-based case–control study in a Han Chinese population.
Materials and methods
The study subjects were recruited from Xijing Hospital between August 2007 and January 2011. Only Han Chinese subjects (who comprise > 90% of the population of China) were included in this study to avoid variations in genotype frequencies among different ethnic groups. The controls were recruited from the health centre for health examinations and had no blood relationship to the patients with vitiligo. Control subjects were frequency matched to the cases by age (± 5 years) and sex (Table 1). We used a standard questionnaire to obtain demographic and characteristic information (sex, onset age, stage, family history and accompanying autoimmune diseases). After clinical examination, patients with vitiligo were classified as having the segmental and nonsegmental types according to the new classification. The patients with segmental vitiligo were excluded from the present study. The response rate was approximately 90% among patients and controls for participation in this study. After signing informed consent forms, each subject donated 5 mL of blood, which was used for genomic DNA extraction and serum collection. The research protocol was designed and performed according to the principles of the Declaration of Helsinki and was approved by the ethics review board of the Fourth Military Medical University.
Table 1. Distribution of selected variables in cases of vitiligo and controls
Cases (n =1000)
Controls (n =1000)
Average age (years), mean ± SD
24·9 ± 12·8
25·9 ± 10·8
Female/male, n (%)
465 (46·5)/535 (53·5)
463 (46·3)/537 (53·7)
Early onset/late onset, n (%)
598 (59·8)/402 (40·2)
Active/stable, n (%)
803 (80·3)/197 (19·7)
With/without family history, n (%)
152 (15·2)/848 (84·8)
With/without autoimmune diseases, n (%)
27 (2·7)/973 (97·3)
Polymorphisms and genotyping
Genomic DNA was extracted from the peripheral blood using a DNA isolation kit (Tiangen Biotech Co. Ltd, Beijing, China). The genotypes of rs1801133 were determined using polymerase chain reaction–restriction fragment length polymorphism (PCR-RFLP) analysis. PCR amplifications were generated using the following primers: forward 5′-AGGACGGTGCGGTGAGAGTG-3′ and reverse 5′-TGAAGGAGAAGGTGTCTGCGGGA-3′ (product of 198 bp). PCR was performed using 50 ng of DNA as a template under the following conditions: 95 °C for 3 min, 35 cycles of 95 °C for 30 s, annealing temperature of 61 °C for 30 s and 72 °C for 30 s, and the final extension step at 72 °C for 10 min. After affinity membrane purification, the PCR products were subjected to cycle sequencing with the respective forward and reverse primers. The HinfI restriction enzyme (Fermentas GmbH, St Leon-Rot, Germany) was used to delineate the rs1801133 genotypes. In the PCR-RFLP method, HinfI was used to digest the 198-bp PCR amplification product of rs1801133, which resulted in two fragments of 175 bp and 23 bp for the T allele (Figure S1b; see Supporting Information). The rs1801131 polymorphism was genotyped by adapted allele-specific PCR using the following primers: forward 5′-TCTTTGTTCTTGGGAGCGG-3′ and reverse 5′-CGAAGACTTCAAAGACACTTG-3′ for the A allele of rs1801131; forward 5′-TCTTTGTTCTTGGGAGCGG-3′ and reverse 5′-CGAAGACTTCAAAGACACTTT-3′ for the C allele of rs1801131 (product of 298 bp). PCR was performed under the following conditions: one cycle of 94 °C for 2 min, 35 cycles of 95 °C for 30 s, 58 °C for 30 s and 72 °C for 30 s, and one cycle of 72 °C for 10 min (Figure S1b). The PCR system without DNA templates was employed and performed as a negative reaction control in the rs1801131 genotyping. The amplified products were analysed using agarose gel 3% with ethidium bromide (0·4 mg mL−1) by electrophoresis. More than 10% of the samples were randomly selected and genotyped again by the same method and sequenced using an ABI PRISM 3700 automatic sequencer (Invitrogen, Carlsbad, CA, U.S.A.) to test the discrepancy rate, and the results were 100% concordant.
Measurement of methylenetetrahydrofolate reductase activity concentration and total homocysteine level
The whole blood of each subject was allowed to coagulate before centrifugation (2000 g for 15 min at 4 °C). Serum was collected and stored at −80 °C until analysis. The MTHFR activity concentration and tHcy level in serum were measured by enzyme-linked immunosorbent assay using an MTHFR assay kit and a tHcy assay kit (Yaji Biosystems, Shanghai, China). Quantitation was performed by external calibration using standards.
Bioinformatics and statistical analyses
Statistical power analysis was performed for the power and sample size calculations of the case–control study using PSSETUP3 software (PS: Power and Sample Size Calculation, Department of Biostatistics, Vanderbilt University, Nashville, TN, U.S.A.). We used the χ2-test to evaluate the differences in frequency distributions of selected demographic variables, including each allele and genotype of rs1801133 and rs1801131. Statistical analysis of the MTHFR activity concentration and tHcy level was performed with the unpaired t-test using GRAPHPAD PRISM 5 software (GraphPad Software, La Jolla, CA, U.S.A.). The means ± SD of the calculated values were used for documentation. Unconditional univariate and multivariate logistic regression analyses were performed to obtain the crude and adjusted odds ratios (ORs) for the risk of vitiligo and their 95% confidence intervals (CIs). Multivariate adjustments were made for age and sex. Trend analysis was performed using the χ2-test and logistic regression was adjusted by age and sex. P-values < 0·05 were considered statistically significant. All tests were two sided for statistical significance and were carried out using SAS software (version 9.1; SAS Institute, Cary, NC, U.S.A.).
Characteristics of study subjects
This study included 1000 Han Chinese patients with vitiligo and 1000 age- and sex-matched controls (P = 0·065 for age; P =0·928 for sex). The frequency distributions of selected characteristics of the cases and controls are shown in Table 1. Among the cases, there were 803 (80·3%) patients with active vitiligo and 197 (19·7%) with stable vitiligo. The patients were considered to have a family history of vitiligo if they had one or more first- to third-degree relatives with the condition (n =152, 15·2%). Patients whose age of onset was earlier than 20 years were considered to have early-onset vitiligo (n =598, 59·8%). In addition, there were 27 (2·7%) patients with vitiligo accompanied by autoimmune disease, including four cases of psoriasis, eight cases of hyperthyroidism, five cases of alopecia areata, five cases of rheumatoid arthritis, one case of scleroderma, three cases of atopic dermatitis and one case of diabetes mellitus.
Genotype distributions of the rs1801133 and rs1801131 polymorphisms between cases and controls
The rs1801133 and rs1801131 genotype distributions between patients with vitiligo and controls, and their associations with the risk of vitiligo, are shown in Table 2. The observed MTHFR genotype frequencies among the controls were in agreement with the Hardy–Weinberg equilibrium (P =0·874 for rs1801133; P =0·385 for rs1801131). The difference in frequencies of the rs1801133 polymorphism was statistically significant between the cases and controls (P <0·001). The frequency of the T variant allele was significantly lower among patients with vitiligo than among controls (34·3% vs. 39·9%, P <0·001). When we used the CC genotype as a reference, a statistically significant decreased risk of vitiligo was associated with the TT genotype (adjusted OR 0·58, 95% CI 0·43–0·76). The frequency of the combined T variant genotypes (CT + TT) was significantly lower among cases than among controls (57·8% vs. 63·7%, P <0·01) and was associated with a significant decrease in risk of vitiligo (adjusted OR 0·78, 95% CI 0·65–0·93). However, the differences in rs1801131 genotype distributions between cases and controls were not statistically significant (P =0·405).
Table 2. Genotype and allele frequencies of the rs1801133 and rs1801131 polymorphisms in cases and controls and their association with the risk of vitiligo
OR, odds ratio; CI, confidence interval. aThe observed genotype frequencies among the control subjects were in agreement with the Hardy–Weinberg equilibrium (χ2 = 0·025, P =0·874 for rs1801133; χ2 = 0·754, P =0·385 for rs1801131). bFor rs1801133, the statistical power is 0·957; for rs1801131, statistical power is 0·205. cTwo-sided χ2-test for distributions of either genotype or allele frequencies between the cases and controls. dORs were obtained from a logistic regression model with adjustment for age and sex.
CT + TT
AC + CC
Activity concentration of methylenetetrahydrofolate reductase and level of total homocysteine in patients with vitiligo and controls
According to the statistical formula n = Z2σ2d−2 (n = sample size, Z = confidence interval, d = sampling error range, σ = SD = 0·5), we calculated the required sample size and employed computer-generated random numbers to select 129 patients with vitiligo from the experimental group and 129 age- and sex-matched normal samples from the control group for serum MTHFR and tHcy measurement. The selected 129 patients and 129 controls were representative of the subjects overall (1000 cases and 1000 controls) with regard to epidemiological variables, clinical features and genotype frequency (data not shown). As shown in Figure 1a, the activity concentration of MTHFR in patients with vitiligo was significantly lower than that of controls (335·72 ± 91·45 U L−1 vs. 400·15 ± 97·48 U L−1; t =5·48, P <0·001). Furthermore, our results showed that patients with vitiligo had higher serum tHcy levels than controls (10·59 ± 1·75 μmol L−1 vs. 8·94 ± 1·79 μmol L−1; t =7·50, P <0·001) (Fig. 1b).
Logistic regression analysis of methylenetetrahydrofolate reductase activity concentration and total homocysteine level in patients with vitiligo and controls
As shown in Table 3, we performed a logistic regression analysis of MTHFR activity concentration and tHcy level in patients with vitiligo and controls. When the activity concentration of MTHFR was dichotomized by the median of the controls, we found that a decreased risk of vitiligo was associated with a higher activity concentration of MTHFR (> 386·88 U L−1; adjusted OR 0·47, 95% CI 0·29–0·79). Furthermore, when the activity concentration of MTHFR was trichotomized by the controls, a correlation between decreased risk of vitiligo and elevated MTHFR activity concentration was evident: lower tertile (< 348·47 U L−1), middle tertile (348·47–449·64 U L−1) and upper tertile (> 449·64 U L−1) activities were associated with adjusted ORs of 1·00, 0·51 (95% CI 0·28–0·90) and 0·26 (95% CI 0·13–0·50; Ptrend < 0·001), respectively.
Table 3. Logistic regression analysis of methylenetetrahydrofolate reductase (MTHFR) activity concentration and total homocysteine (tHcy) level in patients with vitiligo and controls
When the serum tHcy level was dichotomized by the median of the controls, we found that an increased risk of vitiligo was associated with a higher level of tHcy (> 9·07 μmol L−1; adjusted OR 2·84, 95% CI 1·68–4·78). In addition, when the tHcy level was trichotomized by the level of the controls, a correlation between increased risk of vitiligo and elevated tHcy level was evident: lower tertile (< 8·01 μmol L−1), middle tertile (8·01–9·96 μmol L−1) and upper tertile (> 9·96 μmol L−1) levels were associated with adjusted ORs of 1·00, 3·60 (95% CI 1·59–8·16) and 8·30 (95% CI 3·80–18·11; Ptrend < 0·001), respectively.
Risk of vitiligo associated with the rs1801133 genotypes by methylenetetrahydrofolate reductase activity concentration and total homocysteine level
We further estimated the risk of vitiligo associated with the rs1801133 genotypes by MTHFR activity concentration and tHcy level (Table 4). The rs1801133 genotypes were divided into two categories: risk genotype (CC) and no-risk genotype (CT + TT). When those who had the risk genotype of rs1801133 and a lower activity concentration of MTHFR (≤ 386·88 U L−1) were used as the reference, the individuals with the no-risk genotype and a higher activity concentration of MTHFR (> 386·88 U L−1) had a significantly decreased risk of vitiligo (adjusted OR 0·49, 95% CI 0·24–0·99). Consistent with the preceding results, when the activity concentration of MTHFR was divided into tertiles according to the activity concentration of the controls, a correlation between decreased vitiligo risk and increased activity concentration of MTHFR was obvious (Ptrend < 0·01). The rs1801133 (CT + TT) genotype with higher activity concentration of MTHFR was associated with increased protection against vitiligo (348·47–449·64 U L−1, adjusted OR 0·52, 95% CI 0·22–1·20; > 449·64 U L−1, adjusted OR 0·28, 95% CI 0·11–0·70).
Table 4. Risk of vitiligo associated with the rs1801133 genotype by activity concentration of methylenetetrahydrofolate reductase (MTHFR) and level of total homocysteine (tHcy)
In addition, individuals with the no-risk genotype of rs1801133 and a lower tHcy level (≤ 9·07 μmol L−1) had a significantly decreased risk of vitiligo (adjusted OR 0·36, 95% CI 0·17–0·76) than those subjects who had the risk genotype and a higher tHcy level (> 9·07 μmol L−1). Consistent with the preceding results, when the tHcy level was divided into tertiles according to the level of the controls, a correlation between decreased vitiligo risk and lower tHcy level was found (Ptrend < 0·001). The individuals with the no-risk genotype and lower-tertile tHcy level (< 8·01 μmol L−1) had more clearly decreased risk of vitiligo (adjusted OR 0·12; 95% CI 0·04–0·34) compared with those who had the risk genotype of rs1801133 and upper-tertile tHcy level (> 9·96 μmol L−1).
In this hospital-based case–control study, we investigated the associations of the single-nucleotide polymorphisms rs1801133 and rs18011131 in the MTHFR gene with the risk of vitiligo in the Han Chinese population. The genotypic frequencies of the MTHFR polymorphisms found in our results were consistent with the previous data reported in Chinese populations. Our data demonstrate that rs1801133 is associated with a significantly reduced risk of vitiligo, and the T allele of rs1801133 acts as a protective allele. In addition, we found that the patients with vitiligo had a lower activity concentration of MTHFR and higher level of tHcy than the controls. Logistic regression analysis of the MTHFR activity concentration and tHcy level showed dose relationships between decreased risk of vitiligo and higher activity concentration of MTHFR or lower level of tHcy in the rs1801133 CT/TT genotype carriers, especially in patients with vitiligo.
The polymorphism rs1801133 C>T leads to an alanine-to-valine substitution at position 222 of the MTHFR protein. This amino acid substitution is at the bottom of the (βα)8 barrel in the N-terminal 40-kDa catalytic domain of the MTHFR protein. This domain plays a major role in the biochemical properties of the MTHFR protein. It was demonstrated that carriers of both homozygous and heterozygous rs1801133 mutations exhibited high levels of enzyme activity that rescued the biochemical phenotype of the MTHFR knockout cells. However, the underlying mechanism remains unclear and needs to be explored further in molecular research. The association of MTHFR polymorphisms with vitiligo risk has been studied in the Turkish population. However, no evidence was found that MTHFR mutations (rs1801133 and rs1801131) were related to either vitiligo susceptibility or Hcy level in the Turkish population study, negative results from which may mostly be due to a relative smaller sample size of 40 cases and 40 controls. In contrast, our study showed that the T allele of rs1801133 was associated with a significantly reduced risk of vitiligo. This discrepancy may be partially due to the different sample sizes or ethnic groups between the two studies.
The polymorphism rs1801131 A>C results in an amino acid change from glutamic acid (Glu) to alanine (Ala) at position 429 of the MTHFR protein. This amino acid substitution is in the C-terminal regulatory domain of the MTHFR protein, which is involved in stabilizing the enzyme, but does not greatly regulate the enzyme activity. The Glu429Ala protein has been shown to have similar properties to the wild-type enzyme. The effects of rs1801131 on MTHFR activity or Hcy level in human diseases remain controversial. In our study, we did not find any statistical relationship between rs1801131 and vitiligo risk, which is consistent with the study of vitiligo in the Turkish population.
On this basis, our study suggests that the T allele in the rs1801133 polymorphism plays a role in protection against vitiligo. Individuals carrying the rs1801133 protective allele and with higher activity concentration of MTHFR or lower tHcy level may have a decreased risk of developing vitiligo in the Han Chinese population. Nevertheless, it remains unknown whether the MTHFR variants may be functional or in linkage disequilibrium with others. Thus, it is necessary to test this hypothesis in further studies.
In summary, we provide evidence that MTHFR polymorphisms may influence the risk of vitiligo in the Han Chinese population, and correlation may exist between the MTHFR activity concentration or tHcy level and risk of vitiligo. These results are beneficial to better understanding of vitiligo pathogenesis associated with MTHFR or tHcy, and to development of vitiligo therapy in the future. However, larger sample-based studies may be needed to confirm these findings. Furthermore, functional studies may also be required to explore further the mechanisms governing these effects.