Introduction

Prostate cancer (PCa) remains the most commonly occurring non-cutaneous carcinoma. It can be considered as the leading cause of cancer-related deaths among men in Western countries1. Although the incidence and mortality rate of PCa in Asian populations is lower than that in Western countries2, they have been increasing remarkably in recent years3. In China, the overall incidence is also increasing. PCa has become the most common solid tumor in urban male individuals. Therefore this highlights the need of strategies to prevent advanced PCa4,5. Up to now, various factors, such as age, hormone exposure and gene mutations, have been proved to be correlated with the development of PCa6,7. Particularly, growing evidence has indicated that genetic factors may play a crucial role in the etiology of PCa. For instance, the susceptibility of PCa would be increased around two to fivefold in patients suffering from Lynch syndrome. This disease is caused by gene mutations including PMS2 (Postmeiotic segregation increased—2) and MLH1 (Mutl homolog—1)8,9,10.

Previous studies have shown that folate metabolizing genes play a central role in carcinogenesis. They are involved in the process of DNA repair and methylation11,12. Methionine synthase is located on chromosome 1 (1q43) and with 34 exons. It encodes a core enzyme in folate pathway with 1,265 amino acids (molecular weight 140.5 kDa)13. Methionine synthase rs1805087 A/G variant is the most common mutation. It can lead to an aspartic acid to glycine transition at position 919 of the polypeptide chain14,15,16. Previous researches have indicated that this variant is involved in DNA methylation and elevation of homocysteine levels, therefore regulating the enzymatic activity of Methionine synthase17. The association of Methionine synthase rs1805087 A/G variant with susceptibility to PCa was investigated by several studies; however, their conclusions were divergent. In 2009, a systemic analysis evaluated the association between Methionine synthase rs1805087 A/G variant and susceptibility of PCa; nevertheless, they indicated no significant effects of this mutation on PCa risk18. From then on, another research assessed the association between this polymorphism and PCa susceptibility in Han Chinese population. They revealed that Methionine synthase rs1805087 A/G polymorphism was independently associated with PCa by down-regulating the potential of methylation19. The aim of the present study was to explore the association between Methionine synthase rs1805087 A/G variant and PCa susceptibility in larger sample size using multiple analyses to acquire convincing conclusion18,19,20,21,22,23,24,25,26,27,28,29,30,31,32.

Methods

Identification of relevant literature

We conducted a comprehensive literature search according to EMbase, PubMed, Google Scholar, Web of Science, and Chinese SinoMed databases. The keywords are as follows: (MTR OR METH OR methionine synthase) AND (variant OR single nucleotide polymorphism) AND (prostate cancer OR tumor) (last search updated on May 01, 2020). In addition, the reference lists of reviews or supplementary material of source documents were also retrieved for further research.

Inclusion criteria and exclusion criteria

Studies were enrolled in our analysis according to the following criteria: (a) evaluating the relationship between PCa and Methionine synthase rs1805087 A/G polymorphism; (b) containing adequate information for all genotypes; and (c) case–control studies. Furthermore, studies should be removed if: (a) no control population was included; (b) research focuses on other diseases rather than cancer; (c) repeated previous publications.

Data extraction and quality assessment

Relevant data were independently screened by two of the authors according to the selection criteria. The quality assessment of the included studies was investigated by Newcastle–Ottawa Scale (NOS). The NOS score ranges from 0 to 9 stars. A research can be considered as high-quality if it obtained seven or more stars. The following items were extracted: the first author’s name, publication year, race, source of control, score of NOS, genotype frequency, age range, Hardy–Weinberg Equilibrium (HWE) of cases and controls, sample size, and method. G-allele is a minor allele (mutated gene) for Methionine synthase rs1805087 A/G variant. On the other hand, the D-allele is a wild type and considered to be a low-risk allele. Five genetic models were selected in the current study: allelic comparison (G-allele vs. A-allele), homozygote model (GG vs. AA), heterozygote comparison (GA vs. AA), dominant contrast (GG + GA vs. AA), and recessive model (GG vs. GA + AA).

Statistical analysis

The strength of relationship between Methionine synthase rs1805087 A/G variant and risk of PCa was assessed by odds ratios (ORs) and 95% confidence intervals (CIs). We adopt Z-test to measure statistical significance of ORs. Assumption of heterogeneity was calculated by Chi-square-based Q-test. Fixed-effect model (Mantel–Haenszel method) was used when P value of Q-test was more than 0.05; otherwise, the random-effect model was used33,34. Subgroup analyses were conducted by race, sample size of case, quality assessment, and source of controls. P value for HWE was evaluated by web-based program (https://ihg2.helmholtz-muenchen.de/cgibin/hw/hwa1.pl) 35. P value > 0.05 revealed an HWE balance. Sensitivity analysis was adopted to explore the effect of single study on the OR by sequential exclusion of individual study36. I2 value < 50% indicates no statistical heterogeneity among studies. We used the Begg's funnel plot and Egger's test to evaluate the publication bias37,38. P > 0.05 indicates no statistical significance. All the statistical process was conducted using software STATA v11.0 (Stata Corporation, College Station, TX).

Study population and immunohistochemical staining (IHS)

Totally, 200 pathologically confirmed PCa patients were enrolled from our centers. Distribution of characteristics of these patients has been listed in our previous study42. The written informed consent was required from each participant. The above study protocol has been approved by ethics committee of the Affiliated Changzhou No.2 People's Hospital of Nanjing Medical University and Affiliated Hospital of Jiangnan University. IHS was used to test the Methionine synthase expression in PCa cases enrolled in our centers. We incubated the paraffin section of PRAD in hydrogen peroxide (1%), and then washed it in phosphate buffer saline (PBS). Goat serum was utilized to block binding of non-specific proteins. These sections were incubated with anti Methionine synthase antibody at 1:200. Immunoreactive site was brown using diaminobenzidine. All methods in the present study were conducted in accordance with the relevant guidelines and regulations.

In silico analysis of Methionine synthase expression

We employed the online database to explore the expression of Methionine synthase in PCa and control counterparts (https://gemini.cancer-pku.cn/) 39. We also used The Cancer Genome Atlas (TCGA) samples to investigate the expression in PCa based on patients’ age, Gleason score, and nodal metastasis status. This database contains expression profiles of 52 PCa subjects and 496 controls (https://ualcan.path.uab.edu/analysis.html). We evaluated the effects of Methionine synthase rs1805087 A/G variant by SNAP tool (https://rostlab.org/services/snap/). String online server was also employed to assess the network of Methionine synthase interaction (https://string-db.org/) 47.

Ethics approval and consent to participate

The present study was approved by ethics committee of the Affiliated Changzhou No.2 People's Hospital of Nanjing Medical University and ethics committee of Affiliated Hospital of Jiangnan University.

Results

Characteristics of relevant studies

A total of 15 case–control studies were included in our study. The PRIZMA statement has been presented in the Supplementary material Table 1. Totally, 10,666 PCa patients and 40,750 controls were included in the current analysis (Table 1). Subgroup analyses were based on the following criteria: (a) race: 10 studies were conducted in Europeans, 3 studies were based on Asian populations, only one study was on Africans and South Americans; (b) source of controls: there were 6 HB studies and 9 PB studies in the present analysis; (c) quality assessment: 12 studies were of high quality and 3 studies were of low quality. (d) Sample size of case: 9 were large sample size studies and the rest were small sample size studies. Furthermore, we investigated the minor allele frequencies (MAF) in worldwide populations according to genome database (https://www.ncbi.nlm.nih.gov/snp). In Global, G = 0.2118; Europeans, G = 0.1961; Americans, G = 0.1890; Africans, G = 0.2679; East Asians, G = 0.1098; Ashkenazi Jewish, G = 0.1660; and Others, G = 0.2123 (Fig. 1).

Table 1 Study characteristics of Methionine synthase rs1805087 A/G variant included in this analysis.
Full size table
Figure 1
figure 1

Minor allele frequency (MAF) of Methionine synthase rs1805087 A/G polymorphism in worldwide populations according to genome database.

Full size image

Systemic analysis

In the overall analysis, we observed that Methionine synthase rs1805087 A/G variant was associated with increased risk of PCa (Table 2) under three genetic models. For GG + GA vs. AA comparison: OR = 1.08, 95% CI = 1.02–1.14, P value for heterogeneity = 0.183, P = 0.007, I2 = 24.5 (Fig. 2A); for G-allele vs. A-allele contrast: OR = 1.06, 95% CI = 1.01–1.11, Pheterogeneity = 0.122, P = 0.013, I2 = 30.9; and for GA vs. AA model: OR = 1.08, 95% CI = 1.02–1.14, P value for heterogeneity = 0.353, P = 0.009, I2 = 8.9. In stratification analysis by race, similar positive results were revealed in Asian descendants. During GG vs. AA model: OR = 1.93, 95% CI = 1.14–3.26, Pheterogeneity = 0.390, P = 0.014, I2 = 0; for GG vs. GA + AA model: OR = 1.72, 95% CI = 1.02–2.89, Pheterogeneity = 0.767, P = 0.041, I2 = 0. In stratified analysis based on quality assessment, a positive relationship was revealed between Methionine synthase rs1805087 A/G variant and PCa risk in high quality studies (allelic contrast: OR = 1.05, 95% CI = 1.01–1.11, Pheterogeneity = 0.417, P = 0.029, I2 = 2.8; heterozygous model: OR = 1.07, 95% CI = 1.01–1.13, Pheterogeneity = 0.696, P = 0.020, I2 = 0; dominant comparison: OR = 1.07, 95% CI = 1.01–1.13, Pheterogeneity = 0.576, P = 0.018, I2 = 0), but not in low quality studies (allelic contrast: OR = 1.27, 95% CI = 0.81–1.98, Pheterogeneity = 0.019, P = 0.295, I2 = 74.8; heterozygous model: OR = 1.31, 95% CI = 0.80–2.16, Pheterogeneity = 0.046, P = 0.281, I2 = 67.5; dominant comparison: OR = 1.35, 95% CI = 0.79–2.32, Pheterogeneity = 0.019, P = 0.270, I2 = 74.8) (Fig. 2B). Furthermore, in subgroup analysis based on source of control and sample size, similar findings were obtained between Methionine synthase rs1805087 A/G variant and PCa susceptibility in hospital-based studies (allelic contrast: OR = 1.15, 95% CI = 1.02–1.29, Pheterogeneity = 0.137, P = 0.018, I2 = 40.3; heterozygous model: OR = 1.15, 95% CI = 1.01–1.32, Pheterogeneity = 0.245, P = 0.041, I2 = 25.3; dominant comparison: OR = 1.16, 95% CI = 1.02–1.32, Pheterogeneity = 0.133, P = 0.023, I2 = 40.9, Fig. 3A) and studies with larger sample size (allelic contrast: OR = 1.05, 95% CI = 1.00–1.11, Pheterogeneity = 0.226, P = 0.039, I2 = 27.8; GA vs. AA model: OR = 1.07, 95% CI = 1.00–1.13, P value for heterogeneity = 0.309, P = 0.036, I2 = 16.2; GG + GA vs. AA model: OR = 1.07, 95% CI = 1.01–1.13, P value for heterogeneity = 0.277, P = 0.029, I2 = 20.8, Fig. 3B).

Table 2 Stratified Stratification analysis of Methionine synthase rs1805087 A/G polymorphism on the susceptibility to PCa.
Full size table
Figure 2
figure 2

Forest plot shows odds ratio for the association between Methionine synthase rs1805087 A/G variant and PCa risk in subgroup analysis by race (A) and quality assessment (B) (GG + GA vs. AA, fixed-effects).

Full size image
Figure 3
figure 3

Forest plot of Methionine synthase rs1805087 A/G polymorphism in stratified analyses by source of control (A) and sample size (B).

Full size image

In silico analysis and IHS for Methionine synthase expression

We employed in silico tool to investigate the expression of Methionine synthase among 496 PCa patients and 52 controls. The overall result showed that the methionine synthase expression is down-regulated in both young and old PCa subjects (P < 0.05, Fig. 4A). In addition, we assessed whether the level of Methionine synthase expression affected the OS time of patients with PCa. As shown in Fig. 4B, PCa patients with high expression of Methionine synthase may have a shorter OS time in the first 5 years. Nevertheless, with the passage of time, this positive relationship would not be statistically significant (P > 0.05). TGCA database40 showed evidence that the expression of Methionine synthase in PCa cases with Gleason score 6 to 9 was lower than that in normal subjects (P < 0.05, Fig. 5A). In addition, the Methionine synthase expression was both down-regulated in PCa patients with and without lymph node metastasis (P < 0.05, Fig. 5B). In order to assess whether the Methionine synthase rs1805087 A/G variant could impact the protein expression, we used SNAP tool to predict the mutation of Methionine synthase (Fig. 6A). The mutation score is 3, which indicate that Methionine synthase rs1805087 A/G variant is deleterious and effect (Fig. 6B). As described in Fig. 7A, at least 30 proteins are predicted to participate in the protein–protein interaction with Methionine synthase in Homo sapiens. The top 10 proteins are: MTRR: Methionine synthase reductase; MTHFR: Methylenetetrahydrofolate reductase; MMADHC: Methylmalonic aciduria and homocystinuria type D protein, mitochondrial; CBSL: Cystathionine beta-synthase-like protein; MMACHC: Methylmalonic aciduria and homocystinuria type C protein; CTH: Cystathionine gamma-lyase; AHCY: Adenosylhomocysteinase; SHMT1: Serine hydroxymethyltransferase; MAT1A: S-adenosylmethionine synthase isoform type-1; MTHFD1: Methylenetetrahydrofolate dehydrogenase (Fig. 7B). In order to further investigate the Methionine synthase expression in PCa tissues, we adopted IHS to assess its expression among PCa participants in our hospitals. Compared with the earlier stage, the down-regulated expression of Methionine synthase was found in more advanced PCa participants (stage >  = T3b versus =  < T2c, P < 0.05, Fig. 8).

Figure 4
figure 4

In silico analysis of the expression of Methionine synthase. The expression of Methionine synthase is down-regulated in prostate adenocarcinoma tissue (both =  < 60 years and > 60 years group, A). The relationship between the expression of Methionine synthase and OS time among PCa patients (B).

Full size image
Figure 5
figure 5

The expression of Methionine synthase in PCa based on patients' Gleason score (A) and nodal metastasis status (B).

Full size image
Figure 6
figure 6

Evaluation of Methionine synthase rs1805087 A/G variant by SNAP tool (A). The mutation score is 3, which indicate that Methionine synthase rs1805087 A/G variant is deleterious and effect (B).

Full size image
Figure 7
figure 7

Methionine synthase crosstalk with other protein evaluated by String server. At least 30 proteins are predicted to participate in the protein–protein interaction with Methionine synthase in Homo sapiens (A). The top 10 proteins are: MTRR Methionine synthase reductase, MTHFR Methylenetetrahydrofolate reductase, MMADHC Methylmalonic aciduria and homocystinuria type D protein, mitochondrial, CBSL Cystathionine beta-synthase-like protein, MMACHC Methylmalonic aciduria and homocystinuria type C protein, CTH Cystathionine gamma-lyase, AHCY Adenosylhomocysteinase, SHMT1 Serine hydroxymethyltransferase, MAT1A S-adenosylmethionine synthase isoform type-1, MTHFD1 Methylenetetrahydrofolate dehydrogenase (B).

Full size image
Figure 8
figure 8

Tissue expression of Methionine synthase in PCa subjects. Compared with the earlier stage, the down-regulated expression of Methionine synthase is found in more advanced PCa participants (stage >  = T3b versus =  < T2c, P < 0.05).

Full size image

Sensitivity analysis and publication bias

The sensitivity analysis was conducted to investigate the effect of individual study on the OR. As described in Fig. 9A, no single study would significantly impact the overall OR. Both the Begg’s funnel plot (Fig. 9B) and Egger’s tests (Fig. 9C) were performed to assess the publication bias. No evidence of publication bias was revealed among any of the genetic models. For G-allele vs. A-allele: t = 2.01, P = 0.066; GA vs. AA: t = 2.03, P = 0.064; GG vs. AA: t = 1.95, P = 0.073; GG + GA vs. AA: t = 2.06, P = 0.060; and GG vs. GA + AA model: t = 1.92, P = 0.077.

Figure 9
figure 9

Publication bias analysis of Methionine synthase rs1805087 A/G variant and PCa risk. Sensitivity results were assessed by removing every single study in turn (A). No evidence of publication bias can be revealed through both Begg's funnel plot (B) and Egger's test(C).

Full size image

Discussion

Identification of genetic mutations that related to susceptibility of carcinoma is useful to predict cancer risk and reveal the pathogenesis of various tumors41,42. PCa remains one of the most commonly occurring carcinomas among men in Western countries. It has been clarified that several risk factors, such as family history, hormone exposure, lifestyle, may be associated with susceptibility of PCa43,44,45,46. Previous articles showed that Methionine synthase may be involved in the development of PCa29. However, the assoiation between Methionine synthase rs1805087 A/G variant and PCa risk remains ambiguous. For instance, a case–control study based on Iran descendants found that Methionine synthase rs1805087 A/G variant could influence stability and activity of its expression22. This polymorphism might be related to PCa risk in male individuals. However; another researchers did not indicate similar conclusions26. A meta-analysis conducted 10 years ago revealed no significant associations between Methionine synthase rs1805087 A/G variant and PCa risk18. From then on, more and more researchers ascertained their results in different populations19,20,22,23,25,26. Hence, in our systemic analysis, all eligible studies based on the inclusion criteria were summerized to ascertain a precise conclusion on the association between Methionine synthase rs1805087 A/G variant and PCa risk. Totally, 10,666 PCa patients and 40,750 controls were included to investigate the Methionine synthase polymorphism.

In our analysis, 15 case–control studies were included. The overall results revealed a positive association between Methionine synthase rs1805087 A/G variant and susceptibility of PCa under three genetic models. The conclusions acquired from our study were: person carrying the Methionine synthase G-allele may have an increased PCa risk. In stratification analysis by race, we observed similar positive results in Asian populations. Methionine synthase rs1805087 A/G variant might elevate PCa susceptibility as seen in high quality studies, hospital-based studies, and that with large sample size. Furthermore, we used online gene expression mini-database to investigate Methionine synthase expression in PCa and control counterparts. We found that Methionine synthase expression is down-regulated in both young and old PCa subjects. Compared with the normal subjects, the down-regulated expression of Methionine synthase was found in PCa cases with Gleason score 6 to 9, which was consistent with results of our IHS analysis. We also evaluated whether the Methionine synthase expression level influence the PCa patients' OS time. PCa patients with high Methionine synthase expression may have a shorter OS time in the first 5 years. However, with the passage of time, this positive relationship would not be statistically significant. In addition, we used SNAP tool to evaluate the variation of Methionine synthase. The mutation score is 3, which indicate that Methionine synthase rs1805087 A/G variant is deleterious and effect.

Although considerable resources have been generated to evaluate association between Methionine synthase rs1805087 A/G variant and PCa susceptibility, there are some limitations that should be considered. To start with, only three Asian studies involving Methionine synthase rs1805087 A/G variant and PCa risk were retrieved, which indicates that the total participants count for Asian population remains relatively low for more comprehensive analysis. Second, all included case–control studies were retrospective, which may cause selection bias during the process of analysis. Third, lacking of some original data such as family history, lifestyle, and smoking exposure, may limit the efficacy to further calculate adjusted OR. Fourth, studies with only English or Chinese language were included; therefore, some articles written in other languages may be missing, which may cause some bias in risk estimation. In spite of the limitations, there are several advantages in our analysis. First, HWE is very significant while evaluating genetic variations. P value for HWE is more than 0.05 in most of the included studies, which shows that the conclusion from the included studies is very stable. Second, studies with larger sample size particularly improved the statistical efficiency. Third, we indicated no publication bias while evaluating Methionine synthase rs1805087 A/G variant, so the conclusions of the present analysis are more convincing. As described in Fig. 7A, at least 30 proteins are predicted to participate in the protein–protein interaction with Methionine synthase in Homo sapiens. Nevertheless, there is not enough research on their further mechanism in PCa. Further researches are warranted to explore the interactions in more details.

Conclusions

Taken together, our study indicated that Methionine synthase rs1805087 A/G variant is associated with PCa susceptibility, especially in Asian descendants, hospital-based studies, high quality studies and that with large sample size. Moreover, Methionine synthase rs1805087 A/G variant may be related to the prognosis of PCa. Further studies containing more information such as lifestyle and smoking exposure, are warranted to assess this association in details.

Availability of data and materials

All the data analyzed in the present study is included in the manuscript.