The role of the folate cycle genes in development of uterine fibroids

Uterine fibroids are neoplasms of uterine smooth muscle. The incidence rate of uterine fibroids in women of reproductive age is 25–30% [1, 2]. Management strategies for uterine fibroids mainly involve surgical interventions [3, 4], and this is associated with significant economic costs of health care for women of reproductive age.

Many studies are devoted to the mechanisms of development of proliferative diseases of the uterus [5, 6]. There are some published data on the genetic factors for development of this disease, but they have not yet been fully studied [7–9]. The folate cycle genes can be associated with the incidence of uterine fibroids. This may be due to reduced enzymatic activity of the folate cycle leading to impairment of folic acid metabolism and high level of homocysteine, which has an independent mutagenic and prooxidative effect [10, 11]. Moreover, in hyperhomocysteinemia, prostacyclin synthesis decreases and proliferation in vascular smooth muscle cells increases [12–14], that may increase the risk of uterine fibroids.

The aim of research was to study association between polymorphic variants of the folate cycle genes and uterine fibroids.

Materials and methods

The study included women with uterine fibroids alone (n=188) and women in the control group (n=789), who are Russian residents of the Central Black Earth Region of Russia and have no blood kinship.

The studied groups were formed in the Department of Gynecology of St. Joasaph Clinical Hospital in Belgorod Region of Russia. The inclusion criteria were the patients with uterine fibroids alone verified by ultrasound scan and hysteroscopy followed by histological examination of the obtained material. The control group included women, who had no proliferative diseases of the uterus and underwent prophylactic medical examination in the Diagnostic Department of the Perinatal Center. All participants of the study have signed informed consent for publication of their data. The study was carried out under control of the Ethics Committee of the Medical Institute of Belgorod State National Research University.

Genomic DNA isolation from the peripheral blood leukocytes was performed by standard procedure based on the phenol-chloroform extraction. Genotyping of 5 polymorphic loci of genes of the folate cycle was performed: МТRR c.66А>G rs1801394, SHМТ1 c.1420С>Т rs1979277, ТYМS c.*19С>Т rs699517, ТYМS c.*89А>G rs2790, МТR c.2756А>G rs180508. According to published data [15] these polymorphisms have a significant regulatory potential.

Analysis of genomic polymorphic loci of the folate cycle genes was performed by real-time polymerase chain reaction using synthesized oligonucleotide primers and probes (LLC Syntol, Russia) [16].

Statistical analysis

Distribution of alleles and genotypes in the 2 groups was evaluated using chi-square test (χ2 test) with 2×2 contingency tables and the Yates continuity correction. Odds ratio (OR) and 95% confidence interval (95% CI) were used for detection of associations between allelic variants and formation of uterine fibroids.

Software for Windows STATISTICA 10.0 was used for statistical data processing. Analysis of associations between the combination of alleles and genotypes of the studied polymorphic variants of genes and the incidence of uterine fibroids was carried out by АРSampler (https://sourceforge.net/projects/apsampler/) using Markov chain Monte Carlo and Bayesian nonparametric statistics [17, 18]. Permutation test (p-value=pperm) was used to validate the detected associations. Statistically significant p-value was pperm<0.05.

Evaluation of the effect of polymorphic variants on gene expression was performed in silico [19] using the Genotype-Tissue Expression (GTEx) portal data base (https://www.gtexportal.org/). The results with p<8×10-5, FDR ≤0.05 were included in the study. Canalization and relationship between the variants of polymorphic alleles and gene transcription was analyzed using coefficient β of the linear regression, which is indicative of normalized gene expression changes per one polymorphic (alternative) genetic variant. [20].

Regulatory effects of polymorphic loci were detected using web-based SNP selection tools SNPinfo (https:// npinfo.niehs.nih.gov/).

Results and discussion

Comparative analysis of allelic and genotypic frequencies in the studied groups is shown in Tables 1 and 2. Distribution of the frequencies of alleles and genotypes for all studied polymorphic loci was in conformity with Hardy-Weinberg equilibrium (Tables 1, 2).

Analysis of distribution of alleles and genotypes of SHМТ1 c.1420С>Т rs1979277, МТRR c.66А>G rs1801394, МТR c.2756А>G rs180508, ТYМS c.*19С>Т rs699517, ТYМS c.*89А>G rs2790 showed differences in allelic frequency in locus SHМТ1 rs1979277 C among the patients with uterine fibroids (261/368, 70,92%) and women in the control group (970/1490, 65.11%, χ2=4.21, р=0.04, OR=1.30, 95% CI 1.01;1.69).

Statistically significant differences in other studied genes of the folate cycle were not detected in the two groups (p>0.05).

Analysis of allelic and genotypic combinations at the studied loci revealed low occurrence of the combination of the SHMT1 rs1979277 TT genotype and МТR rs180508 G allele among the patients with uterine fibroids (3/176, 1.71%) versus the control group (40/721, 5.54%, р=0.01 OR=0.29, 95% CI 0.09; 0.96).

It was found that among the patients with uterine fibroids, the frequency of combination of SHМТ1 rs1979277 СС genotype and ТYМS rs699517 С allele (90/184, 48.91%) was significantly high versus the control group (294/730, 40.27%, р=0.02, OR=1.41 95% CI 1.02; 1.96).

At the same time, the combination of SHMT1 rs1979277 TT genotype and ТYМS rs699517 С allele occurs 3.7 times less often in women with uterine fibroids (2/183, 1.09%) versus the control group (30/742, 4.04%) and is a protective factor for development of uterine fibroids alone (р=0.03, OR=0.26, 95% CI 0.06;1.10).

Three polymorphisms (p<8×10-5, FDR≤0.05) related to mRNA expression level (cis-eQTL) of four genes in different organs were detected using in silico the GTEx portal data base (http://www.gtexportal.org/).

It was found that MTRR gene polymorphic variant rs1801394 A was associated with low levels of MTRR expression in ovaries (β=-0.42, FDR≤0.05), thyroid gland (β =-0.13, FDR≤0.05), fibroblasts (β=0.26, FDR≤0.05), basal ganglia β=-0.47, FDR≤0.05) and subcutaneous adipose tissue (β=-0.16, FDR≤0.05).

TYMS gene polymorphic variant rs699517 Т was associated with a high level of genes expression: RP11-806L2.6 in fibroblasts (β=0.23, FDR≤0.05), thyroid gland (β=0.38, FDR≤0.05), adrenal glands (β=0.38, FDR≤0.05), ENOSF1 in fibroblasts (β=0.24, FDR≤0.05) and low level of gene transcription: ENOSF1 in skeletal muscles (β=-0.40, FDR≤0.05), and TYMS in ovaries (β=-0.28, FDR≤0.05).

ТYМS gene polymorphic variant rs2790 G was associated with a high level of genes expression: ENOSF1 in fibroblasts (β=0.29, FDR≤0.05) and adrenal glands (β=0.43, FDR≤0.05), RP11-806L2.6 in thyroid gland (β=0.40, FDR≤0.05) and low level of genes expression: ENOSF1 in skeletal muscles (β=-0.49, FDR≤0.05), thyroid gland (β=-0.90, FDR≤0.05), uterus (β=-0,51, FDR≤0,05), and ТYМS in ovaries (β=-0.29, FDR≤0.05).

Thus, GTEx portal data base indicate a significant effect of the studied polymorphisms on expression levels of MTRR, RP11-806L2.6, ENOSF1, TYMS genes in female reproductive organs (ovaries, etc.) and organs that regulate their activity (thyroid gland, adrenal glands, subcutaneous adipose tissue, etc.) and respectively are involved in the pathogenesis of uterine fibroids [21–24].

Polymorphic locus МТRR c.66А>G rs1801394 is located in the region of exon splicing enhancer (ESE) of the soluble cytokine receptor TNFα (spr55) (score=2.82). Regulatory potential (RegPotential) of locus MTRR rs1801394 equals to 0.25 in GTEx portal data base (SNPinfo https://manticore.niehs.nih.gov/). It should be noted that the value of the regulatory potential is indicative of the level at which the locus is involved in regulation of gene expression. We detected in silico that the value of the regulatory potential for locus rs1801394 was 0.25. This indicates its significant role in regulation of transcription activity of МТRR gene.

Polymorphic locus ТYМS c.*19С>Т rs699517 participates in synthesis of microRNA (miRNA) hsa-miR-151-3p (Score=144 points, Energy=-18.84), hsa-miR-28-5p (Score=165 points, Energy=-18.04), hsa-miR-338-5p (Score=156 points, Energy=-17.16), hsa-miR-498 (Score=141 points, Energy=-18.51), hsa-miR-548a-3p (Score=141 points, Energy=-15.15), hsa-miR-708 (Score=156 points, Energy=-17.16), hsa-miR-548e (Score=140 points, Energy=-12.36). The regulatory potential of the studied polymorphism ТYМS c.*19С>Т rs699517 was 0.12.

Polymorphic locus ТYМS c.*89А>G rs2790 participates in synthesis of miRNA hsa-miR-1248 (Score=151 points, Energy=-13.54), hsа-miR-192 (Scоre=140 points, Energy=-9.42), hsа-miR-215 (Scоre=140 points, Energy=-9.70), hsа-miR-515-3p (Scоre=143 points, Energy=-21.97). The regulatory potential of the studied polymorphism ТYМS c.*89А>G rs2790 was 0.20.

In the course of this study we found that the combination of polymorphic variants of the folate cycle genes are associated with formation of uterine fibroids. There are published data about association of МТR rs180508 G allele with a high homocysteine level [25]. In addition, polymorphic variants МТR c.2756А>G rs180508 and МТRR c.66А>G rs1801394 were found in women with rapidly growing uterine fibroids.

The data in literature confirm association of SHМТ1 c.1420С>Т (rs1979277) with development of uterine fibroids alone [26]. This polymorphic locus can affect the catalytic activity of the enzyme serine hydroxymethyltransferase, which is a cofactor for thymilate synthase and glycine [27]. This can lead to impaired biosynthesis of thymidylate and, as a consequence, cause pathological cell proliferation [28]. A number of studies have shown association between the SHMT1 c.1420C> T (rs1979277) polymorphism and development of intrauterine growth retardation [29], non-Hodgkin's malignant lymphomas [30], and rectal cancer [31].

Conclusion

Our study showed that the folate cycle genes: ТYМS c.*19С>Т rs699517, ТYМS c.*89А>G rs2790, SHМТ1 c.1420С>Т rs1979277, МТR c.2756А>G rs180508, МТRR c.66А>G rs1801394 play important role in formation of uterine fibroids alone. SHМТ1 rs1979277 C allele is a risk factor for development of uterine fibroids (OR=1.30). Also, a risk factor for development of uterine fibroids is the combination of SHМТ1 rs1979277 СС genotype and ТYМS rs699517 C allele (OR=1.41). However, the combination of the SHMT1 rs1979277 TT genotype and TYMS rs2790 G allele is a protective factor for development of uterine fobroids alone (OR=0.26).