Association of MTHFR C677T polymorphism and type 2 diabetes mellitus (T2DM) susceptibility

Abstract Introduction Methylenetetrahydrofolate reductase (MTHFR) is essential in mediating folate metabolism, and thus plays an important role in diabetes and diabetic complications. MTHFR C677T (rs1801133 C>T) polymorphism has been proposed to be linked with type 2 diabetes mellitus (T2DM) susceptibility. However, the conclusions are inconsistent. Therefore, we rechecked their linkage aiming to obtain a more reliable estimation by performing an updated meta‐analysis. Methods We searched electronic databases PubMed, EMBASE, CNKI, and Wanfang to obtain studies updated to October 2019. Results After carefully screening, we finally incorporated 68 studies with 10,812 cases and 8,745 controls. The genotype frequency of C677T polymorphism was analyzed pooled to generate odds ratios (ORs) and 95% confidence intervals (CIs). Pooled results presented that MTHFR C677T polymorphism was significantly associated with T2DM under homozygous (OR = 1.64, 95% CI = 1.39–1.94), heterozygous (OR = 1.38, 95% CI = 1.20–1.59), recessive (OR = 1.41, 95% CI = 1.23–1.61), dominant (OR = 1.47, 95% CI = 1.27–1.70), and allele (OR = 1.37, 95% CI = 1.23–1.52) genetic models. Stratified analysis demonstrated that C677T genotype was associated with T2DM in Asian populations, but not Caucasian and African populations. Conclusion Our results indicated that MTHFR C677T polymorphism confers to T2DM, especially in Asian populations. Much more large‐scale case–control studies are needed to strengthen such conclusion in the future.

Previous reports showed that individuals with insufficient intake of folic acid were more likely to have T2DM. Folate is a methyl group donor in the synthesis of intracellular methylation reactions and de novo deoxynucleoside (Blount et al., 1997). When folate deficiency, the DNA stability will be impaired (Duthie, 1999). Methylenetetrahydrofolate reductase (MTHFR) is a folate-metabolizing enzyme that participates in folic acid circulation and DNA synthesis (Friso et al., 2002). MTHFR catalyzes the irreversible reduction of 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate (Niclot et al., 2006). Dysfunction or low activity of MTHFR may decrease the level of methyl pool; consequently, it inhibits the successful deoxynucleoside synthesis and intracellular methylation reactions (Rozen, 1997).
The human gene MTHFR (OMIM number: 607093) is located on chromosome 1p36.3. Of all the identified SNPs, C677T (Ala222Val, rs1801133 C>T) is one of the most investigated genetic variations (Adinolfi et al., 2005;Liew & Gupta, 2015). The C677T polymorphism is a C to T transition at base pair 677, which results in the amino acid transition from Ala to Val. Such amino acid transition significantly decreases the activity of MTHFR (Weisberg, Tran, Christensen, Sibani, & Rozen, 1998). Recent data suggested that there exist an association between C677T and the susceptibility of T2DM. However, the role of C667T in risk of T2DM was discrepant. Several meta-analyses that were conducted to solve this conflicting role somehow failed. To get a precise estimation, we re-analyzed the role of MTHFR C677T on T2DM via including larger eligible investigations.

| Literature search
We carried out a comprehensive literature search in the following databases: PubMed, EMBASE, CNKI, and Wanfang. The searching was updated to October 2019 without any language limitations. The combination of the following search terms was adopted: 'MTHFR or methylenetetrahydrofolate reductase', and 'polymorphism or polymorphisms or SNP or single nucleotide polymorphism or variant' and 'diabetes or mellitus or diabetes mellitus or T2DM'. To expand the included studies, we also retrieved eligible references from the selected studies. The GenBank reference sequence and version number for the gene is: MTHFR (NM_005957.5).

| Inclusion/exclusion criteria
We set the following criteria when performing the selection work: (a) evaluating the association of MTHFR C677T polymorphism with T2DM risk; (b) case-control design; (c) odds ratios (ORs) and their 95% confidence intervals (CIs) were able to obtain; and (d) reports Hardy-Weinberg equilibrium (HWE). Exclusion criteria were as follows: (a) reviews or meta-analyses; (b) case-only studies or case reports; and (c) duplicate publications.

| Data extraction
We arranged three authors to handle data extraction: two authors to extract data independently and one author to resolve the disagreement. The following data were selectively extracted from each study: first author's surname, year of publication, country, ethnicity, genotyping methods, and genotypic distribution. The stratification analysis was conducted by ethnicity (Asians, Caucasians, and Africans) and HWE (HWE <0.05 and HWE >0.05).

| Study characteristics
The study characteristics of the final selected studies were presented in Table 1. A total of 68 studies with 10,812 cases and 8,745 controls were included in our final meta-analysis. Among these eligible studies, 52 were done on Asians, 11 studies were done on Caucasians, and five studies were done on Africans. As to the HWE, genotype distribution in the controls of 52 studies was agreed with the HWE, and 16 studies were not. We also classified the studies into low-quality studies (48 studies) and high-quality studies (20 studies) by quality score.  Figure 2 illustrated the main results of the current meta-analysis. We adopted five genetic models to assess the association between MTHFR C677T and T2DM: homozygote model TT versus CC, heterozygous model CT versus CC, recessive model TT versus CT/CC, dominant model CT/TT versus CC, and allele model T versus C. There was a significant association between MTHFR C677T polymorphism and T2DM under homozygous (OR = 1.64, 95% CI = 1.39-1.94), heterozygous (OR = 1.38, 95% CI = 1.20-1.59), recessive (OR = 1.41, 95% CI = 1.23-1.61), dominant (OR = 1.47, 95% CI = 1.27-1.70), and allele (OR = 1.37, 95% CI = 1.23-1.52) genetic models in a random-effects model.

T A B L E 1 (Continued) T A B L E 2 Meta-analysis of the association between MTHFR
C677T polymorphism and T2DM susceptibility

| Heterogeneity and sensitivity analysis
As shown in Table 1, substantial heterogeneities could be found among all the genetic models (p < .001) for the MTHFR C677T. Therefore, the random-effect model was used to calculate the pooled ORs and 95% CIs for all the models. Sensitivity analysis using sequential leave-one-out strategy was carried out to explore the influence of a single study on the pooled ORs. The omission of each study did not impact the recalculated ORs, indicating the credibility and reliability of our results (Figure 3).

| Publication bias
Begg's funnel plot and quantitative Egger's test were adopted to test the publication bias of the current meta-analysis. As indicated by the symmetrical shape of the Begg's funnel plots, no significant publication bias was observed ( Figure 4). Moreover, Egger's test also suggested the nonexistence of publication bias among the studies (data not shown).

| DISCUSSION
To our knowledge, the current meta-analysis represents the largest and most comprehensive one regarding the relationship between MTHFR C677T and T2DM so far. Our F I G U R E 3 Sensitivity analysis of the association between MTHFR C677T polymorphism and T2DM. Each point represents the recalculated OR after deleting a separate study. OR, odds ratio; T2DM, type 2 diabetes mellitus | 9 of 13 MENG Et al.
analysis provided strong evidence that MTHFR C677T was significantly associated with T2DM, especially in Asians. Sensitivity analysis indicated that there was no significant change in the overall results by removing one study in each turn. Publication bias analysis also showed that the results are convincible.
MTHFR C677T is a functional genetic variation that leads to amino acid substitution from alanine to valine (Ueland, Hustad, Schneede, Refsum, & Vollset, 2001). Such amino acid shift was illustrated to compromise the enzyme activity to nearly 50%, compared to the wild-type MTHFR enzyme (Weisberg et al., 1998). Although the relationship between T2DM susceptibility and MTHFR C677T genotype has been largely investigated, contradictory conclusions still remain. In 2006, no evidence of association was found by F.I.V. Errera et al. between the 677TT genotype of MTHFR and T2DM, in Brazilian populations (Errera et al., 2006). In a study conducted in China in 2014, Wang et al. found that C677T in the MTHFR may influence the risk of T2DM (Wang et al., 2014). Recently, in a case-control study conducted in the population of Brazilian with 47 T2DM cases and 78 controls by Flavio Fontes Pirozzi et al. (Pirozzi et al., 2018), no correlation was found between the MTHFR C677T in the development of T2DM.
Due to the divergent results among single-country studies, several systematic meta-analyses have been undertaken to determine conclusively whether MTHFR C677T is associated with the risk of T2DM. In 2013, Chinese academics Zhong, Rodriguez, Yang, and Li (2013) conducted a meta-analysis regarding MTHFR C677T and T2DM. Their meta-analysis included 4,855 T2DM patients and 5,242 controls. However, they failed to obtain clear evidence of a significant association of MTHFR C677T and T2DM across all 39 studies conducted in 15 countries. They also failed to provide compelling evidence of an association specifically for African, Asian, or Caucasian populations. Interestingly, Khalid et al. (Al-Rubeaan et al., 2013) observed that there was a significant relationship between MTHFR C677T polymorphism and T2DM in Arab population, in 2013. In 2014, Zhu et al. (2014) conducted an updated meta-analysis in Chinese population aiming to better identify the role of C677T polymorphism in T2DM. They included 29 studies with 4,656 T2DM patients and 2,127 controls. They detected a significant relationship between MTHFR C677T polymorphism and T2DM in the Chinese Han population.
Genotype frequencies at the C677T locus of MTHFR vary widely by ethnicity (Errera et al., 2006;Yilmaz et al., 2004), raising the possibility that any association between this SNP and the risk of T2DM may likewise depend on ethnicity. Thus, we further put our focus on ethnic stratification analysis based on the groups that emerged from our literature searches: African, Asian, and Caucasian. Our analysis provided strong evidence that MTHFR C677T was significantly associated with T2DM in Asians, but not in Caucasians or Africans. Thus, it is necessary to identify the role of C677T in different ethnicities.
Several weaknesses should be pointed out before interpreting our conclusion. First, selection bias could not be avoided as only the articles in English and Chinese were analyzed. Other studies written in other languages were unable to include. Second, analyzing one SNP in MTHFR was far more enough, as the development of T2DM was associated with multiple SNPs in multiple genes. Third, we also failed to determine the role of other potential influential factors in the initiation of T2DM. These potential influential factors such as life style, environment exposures, and gene-environment interactions were reported to be associated with T2DM. Fourth, it is inevitable to avoid several shortages such as F I G U R E 4 Funnel plot analysis for assessing publication bias for MTHFR C677T polymorphism under homozygous model. Each point represents a separate study for the indicated association misclassified genotypes, unwell-matched sources of controls, and inconsistent qualities of the included studies, due to the retrospective nature of meta-analysis. Finally, between-study heterogeneity was found in all comparisons, which may compromise the reliability of conclusion.

| CONCLUSION
In all, this meta-analysis provides a precise conclusion that MTHFR C677T polymorphism was significantly associated with T2DM, especially in Asian populations. Further welldesigned, large-scale, and in-depth studies are warranted to check such relationship.

ACKNOWLEDGMENTS
This study was supported by grants from School fund of Guilin Medical University (No. 2017KY0487).