G protein, operation as a molecular transducer, was necessary for different biological signals outside of a cell transmit into the inside of the cell. G protein was composed by α, β, and γ subunits, and β and γ subunits forming a functional monomer [1]. The beta-3 subunit is one of the most important components of intracellular signal transduction in cells and was encoded by G protein β-3 gene (GNB3) [2]. Once activated, α and β subunits dissociated from the receptor and the state of serious intracellular effecter systems changed [3] [4]. Therefore, G protein plays an important role in intracellular signal transduction, and once hurt may induce cells out of control, which could be one of the mechanisms of tumorigenesis.

The GNB3 gene, consists of 12 exons, located on chromosome 12p13 [5]. The polymorphism C825T of GNB3 located in exon 10. In previous studies, polymorphism of GNB3 C825T has been reported to have association of variety disease such as obesity, heart disease [6] [7] [8], hypertension and lately It is reported that the polymorphism was correlated with the cerebrovascular risk independent of blood pressure [9]. However, the GNB3 C825T allele, was reported to have association with G protein activation, which could resulting in increased cell proliferation [10] [11]. Some researchers indicated that the polymorphism C825T of GNB3 could be a potential candidate biological marker of cancer risk [12], for that the wrong synthesis of G protein was associated with signaling processes inside of cells, as well as cell growth and replication control [2] [13] [14] [15]. In recent years, the association between C825T polymorphism of GNB3 and cancers, including breast cancer [16] [17], prostate cancer [18], thyroid carcinomas [19], bladder cancer [20], gastric cancer [21], cholangiocarcinoma [22], glioma [23], and even lymphocytic leukemia [24] has been studied, however, the results were inconsistent.

Therefore, to determine whether the GNB3 C825T polymorphism was associated with different cancers and whether the polymorphism could proved to be one potential cancer marker, we preformed this meta-analysis, which may be important for the previous diagnosis of cancers and may helpful for researchers who interested in the association between GNB3 gene and cancer.

Publication search
To acquire all the studies that have association of the GNB3 C825T polymorphism with cancer risk, we searched the PubMed, Wanfang, CNKI (China National Knowledge Infrastructure) database, using the terms "GNB3 polymorphism" and "G Protein β3 polymorphism" and "cancer" up to March 31, 2014 without language restrictions. The searching work were performed by two reviewers independently to ensure the correctness of our work.The retrieved literatures was scrutinized to ensure whether data on the topic of interest were included.

Inclusion criteria
Studies included in our meta-analysis had to fit the following criteria: (1) studies that evaluated the relationship between the GNB3 C825T polymorphism and cancer; (2) a case-control study; (3) the available allele frequency of the GNB3 C825T allele; (4) providing sufficient data of GNB3 C825T polymorphism to calculate the odds ratio (OR) with 95% confidence interval (95% CI). Accordingly, case-only studies, reviews, or studies without usable data were all excluded.

Data extraction
The following information from each eligible study was extracted carefully for our analysis: first author's name, year of publication, cancer type, country of origin, ethnicity, total number of cases and controls, number of cases and controls with the GNB3 C825T polymorphism (CC genotype with CT and TT genotypes). We examined the extracted information by two authors independently to ensure that the job compiled without man-made faults.

Statistical analysis
We used χ2 test to test whether genotype frequencies of control groups were in Hardy-Weinberg equilibrium (HWE). Then, we employed the odds ratios (OR) and 95% confidence intervals (95% CIs) of the recessive genetic model (TT vs CT/TT), dominant genetic model (CT/TT vs CC), and homozygote comparison (TT vs CC), heterozygote comparison (CT vs CC) as well as allele T vs. allele C in cases and controls to assess the association between the GNB3 C825T polymorphism and cancer risk. Stratified analysis were performed in the subgroups of the same cancer type or the same race which consisting of more than two studies. We employed the Q-statistic and I-squared statistic to determine the degree of heterogeneity, p<0.05 in Q-statistic or I-squared statistic >50% was regarded as significant heterogeneity. When there was no statistical heterogeneity, we used the fixed-effect model and when there was a statistical heterogeneity, we used the random-effect model. All statistical analyses were performed with the STATA package version 11.0 (Stata Corporation, USA).

To conduct sensitivity analysis, we deleting a single study each time involved in the meta-analysis to identify the potential influence of the individual dataset on the pooled ORs. We employed the Begg's funnel plots and Egger's test to assess the potential publication bias and the asymmetry of the funnel plot, respectively [25] [26].

Study characteristics
A total of 20 studies were retrieved in Pubmed by the keywords mentioned earlier. Among these studies, we extracted 6097 subjects involving 2246 cases and 3851 controls in nine available literatures [12] [16] [17] [18] [19] [21] [22] [27][28] , and we excluded the other literatures for that two studies have no data we wanted [20] [29], four were not case-control studies [23] [24] [30] [31], and five paid no attention to the relationship between GNB3 polymorphism and cancers [32] [33] [34] [35] [36]. To amplify the sample size, we included the two studies which were not in agreement with HWE [12] [22] in our meta-analysis, and the results were not changed compared to the results that excluded the two studies. In the nine studies, there were six case-control studies of Caucasian, two of Asians, and one of Latino. The characteristics of each case-control study are summarized in Table 1.

Quantitative synthesis
In the meta-analysis of all involved studies, the Q-test showed there was no heterogeneity in all genetic models of the nine studies except the genetic model TT versus CC/CT, therefore we used random effects model in the TT versus CC/TT model and the fixed effects model in the rest of genetic models to calculate the combined effects. Finally, the OR (95% CI) values of the genetic models in whole samples were: T versus C: 1.03(0.95–1.12) P=0.29 (for heterogeneity); TT versus CC: 1.10 (0.91–1.33) P=0.14; TC versus CC: 1.03(0.91–1.12) P=0.13; CT/TT versus CC: 1.04 (0.95–1.12) P=0.87; TT versus CC/CT: 1.01(0.78–1.31) P = 0.03. These results indicated in a certain extent that the individuals with TT homozygote have an increased risk of cancer compared with those the CC homozygote and TC heterozygote carriers, but there was no statistical significance.

Then we performed a stratified analysis, as the results given in Table 2, no statistical association between GNB3 C825T polymorphism and cancer risk was observed either by ethnicity or by cancer type. When we preformed the stratified analysis by breast cancer in Caucasian, however, we come to a conclusion that the homozygote genotype TT was associated with significantly increased breast cancer risk compared with the homozygote genotype CC (OR=1.44, 95% CI: 1.02–2.04), and CT/CC (recessive model OR=1.49, 95% CI: 1.07–2.09), but no statistical significance was observed when we compared CT versus CC (OR=0.93, 95% CI: 0.75–1.15) and TT/TC versus CC (dominant model, OR=1.01, 95% CI: 0.83–1.23).

Sensitivity analysis and Publication bias
We performed a sensitivity analysis to ensure the confidence for the results, although there was no significant heterogeneity in most genetic models except the genetic model CT/TT versus TT. After exclusion of either individual study, there was little modification of the estimates with pooled ORs ranging from 0.95 to 1.12 (Figure 1), this showed clearly that the results of our meta-analysis was believable.

We employed Begg's test and a funnel plot to estimate the publication bias of the studies included in our meta-analysis. And the result showed that there was no significant publication bias for GNB3 C825T polymorphism, and the funnel plot showed a symmetrical distribution of the studies (Figure 2).

G protein is one of the most important members of cell receptors, was closely related to mitosis and cellular growth [37]. GNB3 gene is essential for the synthesis of G protein β3 subunit. A splice variant could be induced by the C825T polymorphism of GNB3 gene, which can lead to a deletion of 41 amino acids of the β3 subunit [2]. In the previous studies, researchers had concerned about the association between GNB3 C825T polymorphism and other diseases such as obesity [38], hypertension [2] [39], cardiovascular disease [40]. However, in the past decade, increasing researchers were attracted by the potential relationship between the GNB3 C825T polymorphism and cancer, so that the association between GNB3 with its genetic polymorphism and risk of cancer has been widely studied.

From previous studies, we cannot obtain a clear conclusion that whether the GNB3 C825T polymorphism was associated with cancer risks, even the same type of cancer. For instance, Safarinejad et al. indicated that the frequency of the GNB3 825T allele in patients with prostate cancer was significantly higher than in controls, for that patients with prostate cancer who had the TT genotype were at 2.52 times higher risk for prostate cancer than the CC genotype referent group (OR 2.22, 95% CI: 1.18–4.22, p=0.008)[18], however, Eisenhardt et al. suggested that there was no association between prostate cancer and the polymorphism of the GNB3 C825T [21]. So a meta-analysis was needed to certify the association between this polymorphism and cancer risk.

In our meta-analysis, we involved a total of nine case-control studies in Caucasians, Asians, and Latinos including five different types of cancer. To investigate the role of GNB3 C825T polymorphism in cancer, we calculated the effect of different genetic models involving T vs C, TT vs CC, TC vs CC, CT/TT vs CC (dominant genetic model) and TT vs CT/TT (recessive genetic model). Finally, our study suggested that there was no association existed between the GNB3 C825T polymorphism and cancer risk in the overall population. Since the studies involved in our meta-analysis including just one from Latinos, when performed a stratified analysis by ethnicity, we only calculated the samples from Caucasians and Asians. However, the results were also indicated that there were no significant association between the polymorphism and cancer risk in Caucasians and Asians. Then we performed a stratified analysis by the types of cancer, including breast cancer, prostate cancer, and thyroid tumor, and no significant results were obtained, too. But when we performed an analysis of breast cancer in Caucasians, a significant association was observed under the genetic models homozygote comparison (TT vs CC) and recessive genetic model (TT vs TT/CC), the results indicated that the homozygote TT may increase the risk of breast cancer among Caucasians.

Limitations of the meta-analysis existed and should be discussed. First, some relevant studies did not including in our analysis because the raw data were incomplete. Second, Siffert et al. analyzed the distribution frequencies of GNB3 C825T and indicated an existence of different genotypic frequencies among different ethnic group [2], since seven of the nine studies in our analysis were performed in Caucasian, when to assess the whole effects between the GNB3 polymorphism and cancer risk, more studies are needed in other ethnic population to exclude the effect of different genotypic frequencies among different ethnic groups. Third, to expanding the sample size, two studies which were not in HWE were not excluded, although the results were not changed. Even though the above limitations, however, this meta-analysis we performed had some advantages. First, to the best of our knowledge, this is the first meta-analysis which comprehensively assessed the association between the GNB3 gene C825T polymorphism and cancer risk. Second, the substantial data we used in this analysis were select strictly from different studies which could increase the statistical power of the analysis significantly. Third, we indicated that there existed no publication bias suggesting that the whole pooled result should be unbiased.

In conclusion, this meta-analysis indicating that the GNB3 C825T polymorphism was not associated with cancer risk in whole population, but could increase the risk of breast cancer in Caucasian. Bounded by the sample size and source of the ethnic group, more information is needed in the future to ensure our results.

This work was supported by grants from the National High Technology Research and Development Program of China (863) (2012AA02A508), International Cooperation Program (2012DFA30470), National Natural Science Foundation of China (81272792, 81172389, 81372709, 81302185, 81101901,81302184), Jiangsu Province's Natural Science Foundation (BK2011847 and 20131019), Jiangsu Province's Key Provincial Talents Program (RC2011051), Jiangsu Province's Key Discipline of Medicine (XK201117), Jiangsu Provincial Special Program of Medical Science (BL2012028), and Program for Development of Innovative Research Team in the First Affiliated Hospital of NJMU, and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

1. Olate J, Allende JE. Structure and function of G proteins. Pharmacol Ther 1991;51(3):403–19.   [CrossRef]   [Pubmed]    
2. Siffert W, Rosskopf D, Siffert G, et al. Association of a human G-protein beta3 subunit variant with hypertension. Nat Genet 1998 Jan;18(1):45–8.   [CrossRef]   [Pubmed]    
3. Johnson GL, Dhanasekaran N, Gupta SK, Lowndes JM, Vaillancourt RR, Ruoho AE. Genetic and structural analysis of G protein alpha subunit regulatory domains. J Cell Biochem 1991 Oct;47(2):136–46.   [CrossRef]   [Pubmed]    
4. Wieland T, Mittmann C. Regulators of G-protein signalling: Multifunctional proteins with impact on signalling in the cardiovascular system. Pharmacol Ther 2003 Feb;97(2):95–115.   [CrossRef]   [Pubmed]    
5. Levine MA, Modi WS, O'Brien SJ. Chromosomal localization of the genes encoding two forms of the G protein beta polypeptide, beta 1 and beta 3, in man. Genomics 1990 Oct;8(2):380–6.   [Pubmed]    
6. Hegele RA, Anderson C, Young TK, Connelly PW. G-protein beta3 subunit gene splice variant and body fat distribution in Nunavut Inuit. Genome Res 1999 Oct;9(10):972–7.   [Pubmed]    
7. Kato N, Sugiyama T, Morita H, Kurihara H, Yamori Y, Yazaki Y. G protein beta3 subunit variant and essential hypertension in Japanese. Hypertension 1998 Nov;32(5):935–8.   [Pubmed]    
8. Klintschar M, Stiller D, Schwaiger P, Kleiber M. DNA polymorphisms in the tyrosine hydroxylase and GNB3 genes: Association with unexpected death from acute myocardial infarction and increased heart weight. Forensic Sci Int 2005 Oct 29;153(2–3):142–6.   [CrossRef]   [Pubmed]    
9. Casiglia E, Tikhonoff V, Boschetti G, et al. The C825T GNB3 polymorphism, independent of blood pressure, predicts cerebrovascular risk at a population level. Am J Hypertens 2012 Apr;25(4):451–7.   [CrossRef]   [Pubmed]    
10. Pietruck F, Moritz A, Montemurro M, et al. Selectively enhanced cellular signaling by Gi proteins in essential hypertension. G alpha i2, G alpha i3, G beta 1, and G beta 2 are not mutated. Circ Res 1996 Nov;79(5):974–83.   [Pubmed]    
11. Siffert W, Rosskopf D, Moritz A, et al. Enhanced G protein activation in immortalized lymphoblasts from patients with essential hypertension. J Clin Invest 1995 Aug;96(2):759–66.   [CrossRef]   [Pubmed]    
12. Paleari RG, Peres RM, Florentino JO, et al. Reduced prevalence of the C825T polymorphism of the G-protein beta subunit gene in women with breast cancer. Int J Biol Markers 2011 Oct-Dec;26(4):234–40.   [CrossRef]   [Pubmed]    
13. Rosskopf D, Busch S, Manthey I, Siffert W. G protein beta 3 gene: Structure, promoter, and additional polymorphisms. Hypertension 2000 Jul;36(1):33–41.   [Pubmed]    
14. Vallar L. Oncogenic role of heterotrimeric G proteins. Cancer Surv 1996;27:325–38.   [Pubmed]    
15. Hayakawa T, Takamura T, Abe T, Kaneko S. Association of the C825T polymorphism of the G-protein beta3 subunit gene with hypertension, obesity, hyperlipidemia, insulin resistance, diabetes, diabetic complications, and diabetic therapies among Japanese. Metabolism 2007 Jan;56(1):44–8.   [CrossRef]   [Pubmed]    
16. Krippl P, Langsenlehner U, Renner W, et al. The 825C>T polymorphism of the G-protein beta-3 subunit gene (GNB3) and breast cancer. Cancer Lett 2004 Mar 31;206(1):59–62.   [CrossRef]   [Pubmed]    
17. Menzel HJ, Sarmanova J, Soucek P, et al. Association of NQO1 polymorphism with spontaneous breast cancer in two independent populations. Br J Cancer 2004 May 17;90(10):1989–4.   [CrossRef]   [Pubmed]    
18. Safarinejad MR, Safarinejad S, Shafiei N, Safarinejad S. G Protein beta3 subunit gene C825T polymorphism and its association with the presence and clinicopathological characteristics of prostate cancer. J Urol 2012 Jul;188(1):287–93.   [CrossRef]   [Pubmed]    
19. Sheu SY, Handke S, Bröcker-Preuss M, et al. The C allele of the GNB3 C825T polymorphism of the G protein beta3-subunit is associated with an increased risk for the development of oncocytic thyroid tumours. J Pathol 2007 Jan;211(1):60–6.   [Pubmed]    
20. Eisenhardt A, Siffert W, Rosskopf D, et al. Association study of the G-protein beta3 subunit C825T polymorphism with disease progression in patients with bladder cancer. World J Urol 2005 Sep;23(4):279–86.   [CrossRef]   [Pubmed]    
21. Eisenhardt A, Scherag A, Kempin M, Jockel KH, Rubben H. Genotype of the GNB3 C825T polymorphism, A risk factor for the development and course of prostate cancer?. Urologe A 2011 Sep;50(9):1137–42. [Article in German].   [CrossRef]   [Pubmed]    
22. Fingas CD, Katsounas A, Kahraman A, et al. Prognostic assessment of three single-nucleotide polymorphisms (GNB3 825C>T, BCL2-938C>A, MCL1-386C>G) in extrahepatic cholangiocarcinoma. Cancer Invest 2010 Jun;28(5):472–8.   [CrossRef]   [Pubmed]    
23. El Hindy N, Adamzik M, Lambertz N, et al. Association of the GNB3 825T-allele with better survival in patients with glioblastoma multiforme. J Cancer Res Clin Oncol 2010 Sep;136(9):1423–9.   [CrossRef]   [Pubmed]    
24. Nuckel H, Frey U, Aralh N, Durig J, Duhrsen U, Siffert W. The CC genotype of the C825T polymorphism of the G protein beta3 gene (GNB3) is associated with a high relapse rate in patients with chronic lymphocytic leukaemia. Leuk Lymphoma 2003 Oct;44(10):1739–43.   [CrossRef]   [Pubmed]    
25. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997 Sep 13;315(7109):629–34.   [CrossRef]   [Pubmed]    
26. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics 1994 Dec;50(4):1088–101.   [CrossRef]   [Pubmed]    
27. Shibata T, Tahara T, Yonemura J, et al. The G-protein beta3 polymorphism is associated with diffuse type gastric cancer in Japanese. Asian Pac J Cancer Prev 2010;11(5):1195–9.   [Pubmed]    
28. Sheu SY, Görges R, Ensinger C, et al. Different genotype distribution of the GNB3 C825T polymorphism of the G protein beta3 subunit in adenomas and differentiated thyroid carcinomas of follicular cell origin. J Pathol 2005 Dec;207(4):430–5.   [Pubmed]    
29. Lazúrová I, Spiŝáková D, Wagnerová H, et al. Clinically silent adrenal adenomas - their relation to the metabolic syndrome and to GNB3 C825T gene polymorphism. Wien Klin Wochenschr 2011 Oct;123(19-20):618–22.   [CrossRef]   [Pubmed]    
30. Lehnerdt GF, Franz P, Bankfalvi A, et al. Association study of the G-protein beta3 subunit C825T polymorphism with disease progression an overall survival in patients with head and neck squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev 2008 Nov;17(11):3203–7.   [CrossRef]   [Pubmed]    
31. Clar H, Langsenlehner U, Krippl P, et al. A polymorphism in the G protein beta3-subunit gene is associated with bone metastasis risk in breast cancer patients. Breast Cancer Res Treat 2008 Oct;111(3):449–52.   [CrossRef]   [Pubmed]    
32. Pemberton TJ, Mehta NU, Witonsky D, et al. Prevalence of common disease-associated variants in Asian Indians. BMC Genet 2008 Feb 4;9:13.   [CrossRef]   [Pubmed]    
33. Dravecka I, Lazurova I, Habalova V. The prevalence of Gly972Arg and C825T polymorphisms in Slovak women with polycystic ovary syndrome and their relation to themetabolicsyndrome. Gynecol Endocrinol 2010 May;26(5):356–60.   [CrossRef]   [Pubmed]    
34. Vargens DD, Almendra L, Struchiner CJ, Suarez-Kurtz G. Distribution of the GNB3 825C>T polymorphism among Brazilians: impact of population structure. Eur J Clin Pharmacol 2008 Mar;64(3):253–6.   [CrossRef]   [Pubmed]    
35. Sun A, Ge J, Siffert W, Frey UH. Quantification of allele-specific G-protein beta3 subunit mRNA transcripts in different human cells and tissues by Pyrosequencing. Eur J Hum Genet 2005 Mar;13(3):361–9.   [CrossRef]   [Pubmed]    
36. Chang HW, Yen CY, Liu SY, Singer G, Shih Ie M. Genotype analysis using human hair shaft. Cancer Epidemiol Biomarkers Prev 2002 Sep;11(9):925–9.   [Pubmed]    
37. Cohen LH, Pieterman E, van Leeuwen RE, et al. Inhibitors of prenylation of Ras and other G-proteins and their application as therapeutics. Biochem Pharmacol 2000 Oct 15;60(8):1061–8.   [CrossRef]   [Pubmed]    
38. Siffert W, Forster P, Jöckel KH, et al. Worldwide ethnic distribution of the G protein beta3 subunit 825T allele and its association with obesity in Caucasian, Chinese, and Black African individuals. J Am Soc Nephrol 1999 Sep;10(9):1921–30.   [Pubmed]    
39. Siffert W, Rosskopf D, Erbel R. Genetic polymorphism of the G-protein beta3 subunit, obesity and essential hypertension. Herz 2000 Feb;25(1):26–33. [Article in German].   [Pubmed]    
40. Yamamoto M, Abe M, Jin JJ, et al. Association of GNB3 gene with pulse pressure and clustering of risk factors for cardiovascular disease in Japanese. Biochem Biophys Res Commun 2004 Apr 9;316(3):744–8.   [CrossRef]   [Pubmed]