RAD51 135G>C genetic polymorphism and the risk of prostate cancer

Abstract

BACKGROUND

Prostate cancer is the second most common cancer in men worldwide. The incidence of this cancer is constantly and rapidly increasing in India. The genetic variations of RAD51 and its paralog may contribute to the development of cancers such as breast, ovarian, endometrial, colorectal, head and neck cancer, and acute leukemia but there is a scarcity of literature regarding the role of RAD51 135G>C polymorphism in prostate cancer.

AIM

To evaluate RAD51 135G>C polymorphism in patients with prostate cancer.

METHODS

For this research, 50 patients of prostate cancer and fifty healthy controls were enrolled. Genotyping for RAD51 135G>C polymorphism was done by polymerase chain reaction-restriction fragment length polymorphism while estimation of serum total prostate specific antigen (PSA) was done by ELISA. The data were analyzed using appropriate statistical tools, considering P < 0.05 as significant.

RESULTS

A significant association of GC genotype with prostate cancer was observed. No significant association of genotype with total PSA level, Gleason’s score and stage of disease (P > 0.05) could be established.

CONCLUSION

There appears to be a role of RAD51 gene polymorphism in prostate cancer, and GC genotype may have an association with prostate cancer.

Key Words: Single nucleotide polymorphism; RAD51 135G>C polymorphism; Prostate cancer; Risk of disease; GC genotype; Prostate specific antigen

Core Tip: Prostate cancer is the second most common cancer of men worldwide and it is an important public health concern in India. We compared 50 patients with prostate cancer with 50 healthy controls. The genotyping for RAD51 135 G>C polymorphism was done by polymerase chain reaction-restriction fragment length polymorphism while estimation of serum total prostate-specific antigen was done by ELISA.

INTRODUCTION

Prostate cancer is the second most commonly diagnosed noncutaneous malignancy and fifth leading cause of cancer-related mortality among men worldwide[1]. In India, the burden of this disease is substantial, with age-adjusted years lived with disability of 6.5 per 100000 males, making it the third highest among males, and an age-adjusted disability-adjusted life year of 46.9 per 100000 males, keeping it at the seventh highest among males. The incidence of prostate cancer in India is trending upwards and is further expected to rise due to population growth, especially of aging males, increase in life expectancy, improved screening facilities, and better awareness[2].

The etiology of prostate cancer is multifactorial. Based on epidemiological observations, at least four possible etiological factors have been reported in prostate cancer, including dietary and environmental factors, hormonal influence, oxidative stress and genetic predisposition. Genetic factors are one of the main causes and a single nucleotide polymorphism (SNP) is associated with increased risk of prostate cancer[3]. RAD51 is an evolutionary conserved highly polymorphic enzyme encoded by the RAD51 gene which is located on chromosome 15q15.1. RAD51 protein is a bacterial RecA recombinase homolog containing 339 amino acids. It plays a central role in homologous recombination (HR) of double strand break (DSB) repair. RAD51 protein colocalizes with BRCA2 protein in nuclear foci in mitotic cells. Interaction between BRCA2 and RAD51 is essential for error-free HR to repair DSBs. While BRCA2 is directly involved in RAD51-mediated repair, BRCA1 acts upstream from these pathways and is required for transport of RAD51 from cytoplasm to nucleus and to the DNA damage sites[4].

Five RAD51 paralogs, i.e. RAD51B, RAD51C, RAD51D, XRCC2, and XRCC3, have been identified in the human genome. The RAD51 paralogs can form two major complex RAD51B-RAD51C-RAD51D-XRCC2 (BCDX2) and RAD51- XRCC3 (CX3) and two subcomplexes RAD51B-RAD51C (BC) and RAD51D-XRCC2 (DX2)[5]. RAD51 polymorphism (rs1801320) implies that a G->C substitution mapping upstream at 135 bp from the transcription initiation site in the 5’ untranslated region (UTR). Although the functional consequences of this SNP remain to be clarified, a single nucleotide change in this CpG promoter island may upregulate RAD51 gene expression[6]. The genetic variations of RAD51 and its paralog may contribute to the development of different cancers such as breast, ovarian, endometrial, colorectal, acute leukemia, and head and neck cancer[7].

Association of RAD51 gene polymorphism and prostate carcinoma has not been studied extensively. A Polish study has found significant association of rs1801320 polymorphism of RAD51 gene with prostate cancer, suggesting its potential role as a biomarker[8]. In a Brazilian study on the relationship of XPD, RAD51, and APEX1 DNA repair genotypes with prostate carcinoma risk, it was found that only the XPD/Lys751Gln SNP significantly increased susceptibility to prostate cancer, and not the APEX1 and RAD51 SNPs; although combined XPD+RAD51 SNPs were highly associated with the disease[4].

Since the role of RAD51 135G>C polymorphism in prostate cancer is still inconclusive and no such study is available in Indian population to the best of our knowledge, this study was planned to evaluate the role of this polymorphism in prostate cancer.

MATERIALS AND METHODS

For the present study, 50 patients with newly diagnosed, histopathologically proven prostate carcinoma, irrespective of age, Gleason scoring, and staging were enrolled along with 50 age- and sex-matched healthy controls. Diagnosis of prostate carcinoma was established with the help of detailed history, clinical examination and histopathological examination. Informed consent was obtained from each individual for participating in this study and approval from the institutional ethical committee was obtained. Patients with any other cancer, prostatic illness or any chronic disease were excluded. Staging was done according to American Joint Committee on Cancer guidelines and histological aggressiveness was assigned using Gleason grading system[9,10]. Cases were grouped according to Gleason’s score ≤ 7 and > 7.

Extraction of genomic DNA from blood samples was done by proteinase K and phenol-chloroform method on the same day and DNA was stored at 20 °C after checking the purity[11]. The routine evaluation of the isolated DNA was done by agarose gel electrophoresis [using 0.8% agarose gel stained with ethidium bromide (10 μg/mL)]. After electrophoresis, the gel was visualized using a UV transilluminator. The purity of DNA was checked as per 260/280 (1.8-2.0) ratio of the extracted sample.

The rs1801320 polymorphism of RAD51 gene is characterized by a change of guanine (G) to cytosine (C) at 135 position in its untranslated region. The polymerase chain reaction (PCR) was run in 10 μL mixture containing 10 ng genomic DNA, 0.2 mmol/L each primer, 2.5 mmol/L MgCl₂, 1 mmol/L dNTP, 3U Taq DNA polymerase and 1 Solis BioDyne buffer B1. The thermal cycling was performed using the protocol: Initial activation at 95 °C for 12 min, followed by 30 amplification cycles consisting of denaturation at 95 °C for 30 s, annealing at 65 °C for 30 s and extension at 72 °C for 1 min, followed by a final extension at 72 °C for 10 min. For allele genotype, 10 μL each PCR product was digested with 2 UBstNI at 37 °C for 16 h. The genotypes were determined by running the digested product in 3% agarose gel with ethidium bromide (1 μL/mL) for UV visualization[8]. The details of primers, PCR conditions and the products of each genotype are shown in Tables 1 and 2. Estimation of total PSA was done by a solid phase enzyme immunoassay[11].

Table 1 Details of single nucleotide polymorphism analyzed.

Gene	Polymorphism	Other names	SNP position	Chromosome	Method
RAD51	Rs1801320	c.-98G>C, G135C	UTR-5, Exon	15:40695330	PCR-RFLP

PCR-RFLP: Polymerase chain reaction-restriction fragment length polymorphism; SNP: Single nucleotide polymorphism.

Table 2 Details of polymerase chain reaction-restriction fragment length polymorphism of the studied single nucleotide polymorphism.

SNP	Primer sequences	Annealing temperature (°C)	Product size (bp)	Enzyme	Genotype	Fragment sizes (bp)
Rs1801320	(F)5’TGGGAACTGCAACTCATCTGG3’, (R)5’GCGCTCCTCTCTCCAGCAG3’	65	157	BstNI	GG	7186
					GC	7186157
					CC	157

SNP: Single nucleotide polymorphism.

The data were compiled and analyzed by SPSS version 22.0, with 95% confidence level. The level of significance was kept at P < 0.05. Normality of the data was checked by Shapiro-Wilk test. The ages of the patients and controls were normally distributed, so they were compared by unpaired t test. PSA values were nonparametric (even among various genotype groups, such as GG, GC and CC but not GG); therefore, they were compared using the Kruskal-Wallis test. Comparison of genotype frequency among cases and controls, association of stages, Gleason’s scoring with genotypes and alleles were analyzed by χ² test.

RESULTS

The mean age of the patients was 67.86 ± 9.91 years (45-86 years) and that of controls was 65.48 ± 6.28 years (50-78 years). The difference in age between the groups was nonsignificant (P > 0.05, t = 1.43). Forty-nine (98%) patients and 48 (96%) controls were married (P = 0.5577). Nineteen patients presented with dysuria and six with urinary retention as the main symptom. Other presentations included lower back pain (n = 5), burning micturition (n = 5), hematuria (n = 4), increased frequency of urine (n = 4), urinary incontinence (n = 2), decreased output of urine (n = 2), decreased force of urine (n = 2) and diffuse bony pain (n = 1). Most of them presented with multiple complaints. Only one was asymptomatic. Twenty-nine (58%) patients had Gleason score ≤ 7 and 21 (42%) had a score > 7. Ten (20%), two (4%), 13 (26%) and 25 (50%) patients presented with stage A, B, C and D, respectively. The median level of serum PSA among the patients was 54.93 μg/L [interquartile range (IQR): 106.1] and the median serum level of PSA was 55 μg/L (IQR: 106.5), 51 μg/L (IQR: 86.83) and 296.56 μg/L (IQR: 403.69) for the genotype groups CC, GC and GG respectively (P > 0.05).

The distribution of genotype (GG, GC and CC) frequency in RAD51 (rs1801320) among prostate cancer patients was 4%, 70% and 26%, respectively and 76%, 14% and 10% among the controls (P < 0.05) (Table 3). The association of any genotype or allele with the PSA level, staging and Gleason’s score was not significant (Tables 4, 5, 6, 7, 8 and 9). Table 10 shows the odds ratio of different genotypes and alleles along with their significance levels.

Table 3 Distribution of genotype frequency in RAD51 (rs1801320) among prostate cancer patients and controls, n (%).

	Genotype	Cases	Controls	P, χ2d value
RAD51 Rs1801320	GG	2 (4)	38 (76)	1.377 × 10-12, 54.62
	GC	35 (70)	7 (14)
	CC	13 (26)	5 (10)

Table 4 Association of prostate-specific antigen with genotype in prostate cancer patients.

Genotype (n)	PSA level (μg/L) (median, IQR)	P, test statistic H
CC (13)	55, 106.5	P = 0.4, H = 1.84 at 95%CI (0-5.99)
GC (35)	51, 86.83
GG (2)	296.56, 403.69

CI: Confidence interval; PSA: Prostate-specific antigen.

Table 5 Association of prostate-specific antigen with allele in prostate cancer patients.

Allele	PSA level ( μg/L )			Total	P value, χ2d value
	< 4	4-10	> 10
G	3	3	33	39	0.818167, 0.40
C	7	5	49	61
Total	10	8	82	100

PSA: Prostate-specific antigen.

Table 6 Association of genotype with staging in prostate cancer patients.

Stage	Genotype			Total	P, χ2d value
	CC	GC	GG
A	2	8	0	10	0.649885, 4.2
B	1	1	0	2
C	4	18	1	23
D	6	8	1	15
Total	13	35	2	50

Table 7 Association of allele with staging in cases.

Stage	Allele		Total	P value, χ2d value
	G	C
A	8	12	20	0.769677, 1.13
B	1	3	4
C	20	26	46
D	10	20	30
Total	39	61	100

Table 8 Association of genotype with Gleason’s score.

Gleason’s score	Genotype			Total	P, χ2d value
	CC	GC	GG
≤ 7	10	19	0	29	0.087551, 4.87
> 7	3	16	2	21
Total	13	35	2	50

Table 9 Association of allele frequency with Gleason’s score.

Gleason’s score	Allele		Total	P, χ2d value
	G	C
≤ 7	19	39	58	0.132648, 2.26
> 7	20	22	42
Total	39	61	100

Table 10 Odds ratio of genotypes and alleles and corresponding statistical values.

Genotype/allele	OR	95%CI	Z-statistic	P
GG	0.0132	0.0028-0.0624	5.45	< 0.0001
GC	14.33	5.26-39.04	5.21	< 0.0001
CC	3.16	1.03-9.69	2.02	0.0438
C	7.64	3.95-14.75	6.05	< 0.0001
G	0.13	0.07-0.25	6.05	< 0.0001

CI: Confidence interval; OR: Odds ratio.

DISCUSSION

The mean age of patients was 67.61 ± 9.85 years (45-86 years). Cancer can be regarded as a disease of old age as the incidence of most cancers has been observed to increase with age. Age may be considered as a surrogate measure of the complex biological process of aging, which is believed to be one of the most important contributing factors to the pathogenesis of cancer. In the present study, age of the healthy individuals was matched with that of the patients to exclude any bias in the study[12].

The established marker for screening and diagnosing the disease is PSA. However, the actual significance of this screening is moderate because although serum level of PSA > 4 μg/L suggests a diagnosis of prostate cancer, almost 25% of cases with increased levels do not have the disease and ~20% of prostate cancer patients have normal PSA levels[13]. In the present study, the median level of serum PSA among the patients was 54.93 μg/L (IQR: 106.1) and the median serum level of PSA was 55 μg/L (IQR: 106.5), 51 μg/L (IQR: 86.83) and 296.56 μg/L (IQR: 403.69) for the genotype groups CC, GC and GG, respectively (P > 0.05). The serum level of PSA was not associated with the G or C allele. Nowacka-Zawisza et al[8] also found no association of total PSA value with any of the alleles. In another study, the same authors did not find any association of total PSA value and free PSA/total PSA ratio with A/G alleles of rs5030789 of RAD51 gene and A/G alleles of rs1799796 of XRCC3 gene[14]. Hardy-Weinberg equilibrium calculation in the controls showed the allele frequency of the wild type was 0.83 and that of the mutant was 0.17, with χ² value of 12.7 and P = 0.0004. This may be attributed to population admixing, and nonrandom mating in the local population. Moreover, the present study found no association of staging or Gleason’s score with the genotype or particular allele (P > 0.05 in all cases). Cypriano et al[4], in their study on the Brazilian male population, also found no association of this polymorphism with Gleason’s score or serum PSA value.

Various germline and somatic mutations have been found to be risk factors for prostate cancer. Various changes associated with the RAD51 have been investigated in different cancers. There are few studies investigating the association of RAD51 135G>C polymorphism in prostate cancer, and they were conflicting in their nature. Their results are summarized in Table 11. Cypriano et al[4] studied this polymorphism in the Brazilian male population. They found that wild-type RAD51 patients showed worse prognosis, possibly because of better repair of therapy-induced DNA damage by homologous recombination repair, while XPD/Lys751Gln patients showed high Gleason scores. Regarding general clinical findings, XPD and XPD+RAD51 showed good negative predictive value. They found no significant difference between cases and controls. Nowacka-Zawisz et al[8] reported that RAD51 rs1801320 (135G>C) polymorphism doubled the risk of prostate cancer as significant allele distribution was found between cases and controls. According to them, this polymorphism might be an independent biomarker in the Polish population. Besides, they did not find any association of polymorphic variants of RAD51B, XRCC2 and XRCC3 genes in their study. Dhillon et al[15] investigated this polymorphism in the Australian population and found no such correlation. Rather, they reported the association of prostate cancer with the hOGG C1245G polymorphism. Another study, in the Jordanian population, also highlighted the probable association of RAD51 135G>C polymorphism with prostate cancer, along with no significant association with another 5′-UTR polymorphism of RAD51, rs1801321 (G172T)[16]. Nowacka-Zawisza et al[14] investigated the other SNPs of RAD51, rs2619679, rs2928140 and rs5030789. They reported the association of rs5030789 with the occurrence of prostate cancer.

Table 11 Summary of literature reports regarding the association of RAD51 135G>C polymorphism with prostate cancer.

Ref.	Years	Population	Cases (%)			Controls (%)
			GG	GC	CC	GG	GC	CC
Cypriano et al[4]	2017	Brazillian	76.4	21.8	1.8	78.5	20	1.5
Nowacka-Zawisza et al[8]	2015	Polish	65.3	26.7	8	79.6	17.1	3.3
Dhillon et al[15]	2009	Australian	83.6	15.5	0.9	90.2	9.8	0
Al-Zoubi et al[16]	2018	Jordanian	92.5	1.3	6.2	88	10	2

However, the role of RAD51 gene and its product seems to be complicated. Mitra et al[17] reported overexpression of RAD51 in high-grade or aggressive prostate cancer and that mRNA and protein levels of RAD51 and other homologous recombination genes were increased in prostate cancer cell lines. Considering that RAD51 is involved in DNA repair, it should decrease the risk of harmful mutations from environmental toxic factors, indicating the positive impact of RAD51 overexpression in prostate cancer and other malignancies. Therefore, both the overexpression of RAD51 and its various polymorphisms are associated with cancer, which emphasizes the need for complete understanding of the role of the RAD51 regarding DNA repair.

The limitations of the study included small sample size and not measuring RAD51 protein in serum or tissues due to time and financial constraints. The small sample size reduces the statistical power, which invariably increases type II error and increases the probability of unrepresented allele frequency. Therefore, the study should be validated in a larger, multicenter study that will include the various strata of the population, decreasing the selection bias further.

CONCLUSION

The findings of this study suggest an association of GC genotype with prostate cancer, although no association of a particular genotype or allele with total PSA level, Gleason’s score and disease staging could be established. To understand the role of RAD51 135G>C SNP in prostate cancer better, further large-scale studies are required.