SEARCH

SEARCH BY CITATION

Keywords:

  • G proteins;
  • genetics;
  • polymorphism;
  • population studies;
  • multiple logistic regression analysis

Abstract

  1. Top of page
  2. Abstract
  3. Research Methods and Procedures
  4. Acknowledgment
  5. References

We examined the association of the G-protein β3 subunit gene (GNB3) C825T polymorphism with overweight in Japanese workers. This cross-sectional study used multivariate analysis to investigate whether a polymorphism in the C825T polymorphism was associated independently with overweight when factors such as age and lifestyle were taken into account. The study in 1453 men and 1172 women involved identifying subjects with the C825T genotype using the polymerase chain reaction, followed by restriction fragment-length polymorphism analysis. Overweight was defined as a BMI ≥25 kg/m2. Genotype distributions for C825T in overweight men (CC = 80, CT = 162, TT = 80) and women (CC = 52, CT = 91, TT = 40) were not significantly different from normal-weight men (CC = 278, CT = 588, TT = 265) and women (CC = 242, CT = 549, TT = 198). The allele distributions were also not significantly different between either sex. The power of the study was estimated as 98% in men and 81% in women based on the allelic frequencies reported in a previous positive study in Chinese subjects. Multiple logistic regression analysis showed that the genotype was not significantly associated with overweight. In conclusion, this study indicated that the GNB3 C825T polymorphism is not a significant factor for overweight in Japanese people.

In recent years, many genetic variations have been identified and associated with congenital hereditary diseases, as well as with acquired chronic diseases such as hypertension and diabetes mellitus. Such chronic diseases are considered to be polygenetic and have a multifactorial trait, in that the onset of disease may be influenced by various genes interacting reciprocally in combination with host factors such as lifestyles and environmental factors.

The variants of the G-protein β3 subunit (GNB3)1 gene (e.g., the C825T polymorphism) have attracted renewed attention in recent years. The 825T allele is associated with a splice variant of GNB3 protein and enhanced G-protein activation (1), with ethnic distributions of 825T allele frequency ranging from 20% to 80% (2). Several studies have demonstrated that the 825T allele of the GNB3 gene C825T polymorphism is associated with essential hypertension or elevation of blood pressure (1, 3, 4, 5, 6, 7). Because obesity is an established major risk factor for hypertension, several studies have investigated whether the 825T allele increases the risk for obesity (8); these studies have demonstrated that this allele is associated with obesity or increased BMI in several ethnic groups (2, 9, 10, 11, 12). On the basis of these findings, Siffert speculated that the increased risk for obesity in 825T allele carriers may contribute to the greater prevalence of hypertension in these obese subjects (8). Other similar studies, however, have failed to demonstrate this association (13, 14, 15). We are aware of only one other study that has been carried out in the Japanese population that did not show a significant association between GNB3 825T and obesity (16). Because of these varying findings, we considered that the apparent association between the polymorphism of the GNB3 gene and obesity or raised BMI had not been demonstrated conclusively. From an epidemiological point of view, we believe that in order to determine the influence of genetic polymorphisms in the occurrence of a specific disease, it is necessary to undertake large-scale studies in the general population. This requirement led us to investigate in greater detail the relationship between the GNB3 gene C825T polymorphism and overweight in Japanese workers. This study included logistic regression analyses that examined whether the polymorphism in this gene was an independent determinant of overweight when factors such as age and lifestyle were taken into account.

The frequencies of genotype, allele, and other categorical variables grouped according to gender are shown in Table 1. There was no significant difference in the frequency of any of these genotypes or alleles between the overweight subjects and normal-weight subjects. In men, the percentage of subjects with a smoking habit was significantly higher in normal-weight compared with overweight subjects. In contrast, the percentage prevalence of alcohol consumption and habitual exercise was not significantly different between the normal-weight and overweight subjects in either sex. The means and SDs of age and BMI grouped according to gender are also shown in Table 1. In both sexes, overweight subjects had a significantly higher mean age compared with normal-weight subjects.

Table 1. . Frequencies of genotype, allele, and characteristics of other variables grouped according to gender
VariableMenWomen
 Normal weight (n = 1131)Overweight (n = 322) Normal weight (n = 989)Overweight (n = 183) 
 n%n%pn%n%p
Genotype          
 CC27824.68024.80.8424224.55228.40.34
 CT58852.016250.3 54955.59149.7 
 TT26523.48024.8 19820.04021.9 
Allele          
 C114450.632250.00.82103352.219553.30.73
 T111849.432250.0 94547.817146.7 
Alcohol consumption (six times a week or more)45540.211535.70.16191.921.10.76
Smoking habit (smoker)67759.917253.40.04464.763.30.56
Habitual exercise (absence)74165.519259.60.0678379.213372.70.06
 MeanSDMeanSDpMeanSDMeanSDp
Age (years)38.39.939.89.20.0138.59.342.48.6<0.001
BMI (kg/m2)21.81.927.02.2<0.00121.02.027.62.6<0.001

The results of the logistic regression analysis are summarized in Table 2. Although there was no relationship between genotype and overweight, several other factors were found to be independent determinants of overweight. These included age [odds ratio (OR) = 1.020 in men and OR = 1.054 in women] and absence of habitual exercise (OR = 0.615 in women). Neither smoking habit nor alcohol consumption in both sexes nor habitual exercise in men was associated significantly with overweight. Furthermore, the OR of the TT genotype to the CC or CT genotypes or the CC genotype to the TT or CT genotypes was not significant. When the data that were grouped according to gender and habitual exercise were evaluated by univariate and multivariate analysis, the influence of the C825T polymorphism was also not significant (Appendix 2, Tables 3–6; available at Obesity Research Online: http:www.obesityresearch.org).

Table 2. . The results of the logistic regression analyses grouped according to gender
 MenWomen
Independent variablesOdds ratio*95% Confidence intervalpOdds ratio*95% Confidence intervalp
  • *

    Odds ratio: the ratio of the former to the latter was estimated for categorical variables.

Genotype      
 CT/CC0.9380.691 to 1.2730.680.7610.521 to 1.1120.16
 TT/CC1.0330.724 to 1.4730.860.9040.570 to 1.4320.67
  Alcohol consumption (six times a week or more/less than 6 times a week)0.7750.591 to 1.0150.06   
  Smoking habit (smoker/nonsmoker)0.8140.631 to 1.0490.111.0480.430 to 2.5530.92
  Habitual exercise (absence/presence)0.7830.605 to 1.0140.060.6150.425 to 0.8890.01
  Age (years)1.0201.006 to 1.0330.0041.0541.034 to 1.074<0.001
Genotype      
 TT/CC + CT1.0780.807 to 1.4410.611.0840.735 to 1.6010.68
  Alcohol consumption (six times a week or more/less than 6 times a week)0.7750.591 to 1.0160.07   
  Smoking habit (smoker/nonsmoker)0.8150.632 to 1.0500.111.0190.419 to 2.4800.97
  Habitual exercise (absence/presence)0.7840.606 to 1.0150.070.6130.424 to 0.8860.009
  Age (years)1.0201.006 to 1.0330.0041.0531.033 to 1.074<0.001
Genotype      
 TT + CT/CC0.9680.725 to 1.2910.820.7990.558 to 1.1440.22
  Alcohol consumption (six times a week or more/less than 6 times a week)0.7740.591 to 1.0140.06   
  Smoking habit (smoker/nonsmoker)0.8140.632 to 1.0490.111.0440.429 to 2.5430.92
  Habitual exercise (absence/presence)0.7850.607 to 1.0160.070.6160.426 to 0.8900.01
  Age (years)1.0201.006 to 1.0330.0041.0541.034 to 1.074<0.001

Important features of this study include the fact that data were collected from >2000 subjects, and the influence of a wide range of confounding variables was examined by multiple logistic regression analyses. Other representative studies on the polymorphism in the C825T polymorphism have been multivariate studies that have been limited by using only age or sample collection (2) or age and gender as confounding variables (12). Another notable feature of our study was the composition of the study cohort, which included a large sample of workers from a Japanese factory. From an epidemiological point of view, we believe that it is important to establish the genotypic distribution and association of a polymorphism with overweight in the general population, and, accordingly, our study would be expected to provide a more accurate distribution of the GNB3 C825T polymorphism in the general Japanese population. Taken together, we consider that this study may be more informative than previous studies.

The univariate analyses showed that genotypic and allelic frequencies of GNB3 C825T were similar in overweight and normal-weight subjects in both sexes, with these frequencies being comparable with those reported in other studies of Japanese subjects (16). Multivariate analysis confirmed that the polymorphism in this gene was not associated with overweight, whereas age in both sexes and the absence of habitual exercise in women were the factors associated significantly with overweight. Furthermore, the influence of the C825T allele was also not significant when the data grouped according to habitual exercise were evaluated. This finding was contrary to the result of an earlier study (9). The power of our study was calculated and showed a power of 98% for the male group and 81% for the female group, indicating that the number of subjects in the present study was sufficient to enable detection.

Using univariate and multivariate analyses to examine the effect of numerous confounding factors, Brand et al. examined the association between the GNB3 C825T polymorphism and BMI and obesity in a Belgian population (10) and showed that the T825 allele of GNB3 was associated with the presence of obesity (BMI ≥30kg/m2) and BMI level in only men. On the other hand, their study also demonstrated that the genotypic and allelic frequencies were not significantly different between normal (BMI <25kg/m2) and overweight (BMI 25 to 29.9kg/m2) subjects. Because the sampling and analytical methods used in their study were appropriate from an epidemiological point of view, we considered their results to have been accurate and, therefore, important. In our study, only 1.7% of all men and 1.9% of all women had a BMI >30 kg/m2; therefore, there may have been a minor influence on the results if subjects with a BMI ≥30 kg/m2 were included within the overweight group. It is interesting to note that in another appropriately designed epidemiological study, we obtained results similar to those found in the present study; both studies used almost identical criteria, although ethnicity was different. Based on the power calculation, it would be necessary to study at least 10, 000 subjects to ascertain the association between the C825T polymorphism and obesity (BMI ≥30 kg/m2) in the Japanese population.

Rosskopf and colleagues characterized the entire GNB3 gene and defined new polymorphisms (17, 18), some of which almost showed linkage disequilibrium with C825T. Siffert (8) pointed out that the complex of these polymorphisms appeared to cause alternative splicing of the 825T allele. There is evidence that the haplotype of the C825T polymorphism with other polymorphisms of the GNB3 gene (i.e., C1429T) is different across ethnic groups (17). This difference may explain the discrepancies in the association of C825T and obesity or overweight across ethnic groups. Our study does not exclude the possibility that other variants around the C825T polymorphism may influence BMI levels and the prevalence of overweight in the Japanese population; therefore, future investigations should also investigate whether haplotypes of C825T and other GNB3 gene polymorphisms are associated with overweight. In conclusion, this study indicated that the GNB3 C825T polymorphism is not a significant factor for determining overweight in Japanese people and implies that therapy targeting this polymorphism would be ineffective for preventing overweight in the general Japanese population.

Research Methods and Procedures

  1. Top of page
  2. Abstract
  3. Research Methods and Procedures
  4. Acknowledgment
  5. References

Subjects

The study was a cross-sectional design. The target subjects were 3834 male and 2591 female subjects who worked in a zipper and sash factory in the Hokuriku district of Japan. All workers in this company underwent a legally required health checkup in 1997, which included measurement of height and weight and a self-administered questionnaire. Workers who had not provided sufficient data or who had not provided written, informed consent were excluded from the analyses. The final study cohort consisted of 1453 men and 1172 women. The study protocol was approved by the ethical review boards of Kanazawa Medical University and the Graduate School of Medicine, Chiba University.

Genotyping

The buffy coat was isolated from venous blood collected from each subject in 1997. The C825T genotype was then determined by direct polymerase chain reaction (PCR) amplification from the buffy coat using Ampdirect (Shimadzu, Kyoto, Japan) buffers, with the use of primers that have been described previously by Siffert et al. (1). Ampdirect buffers allow PCR amplification directly from blood samples (19). The PCR products were restricted with BseDI (Fermentas, Vilnius, Lithuania), separated on 2% agarose gels, and visualized under ultraviolet light by ethidium bromide staining.

Statistical Analyses

For analysis of categorical variables, alcohol consumption was classified as either drinking at least six times per week or drinking less than six times per week, smoking habit was classified as either smoking or nonsmoking, and habitual exercise was classified as either the presence or absence of regular exercise. Body weight was measured with participants wearing light clothing, and height was measured while participants stood barefoot with their heels together and backs against a wall. Overweight was defined as BMI ≥25.0 kg/m2. In the univariate analyses, the genotypic and allelic frequencies of C825T, alcohol consumption, smoking habits, and habitual exercise were compared between the overweight and normal-weight subjects using the χ2 test (genotype) and Fisher's exact test (other variables). As noted in Appendix 1 (available at Obesity Research Online: http:www.obesityresearch.org), the power of the study was estimated as 98% in men and 81% in women, based on the allelic frequencies reported in a previous positive study of Chinese subjects (2); we set the criterion for significance as α = 0.05 (two tailed). The means of age and BMI for both men and women were calculated and compared between overweight subjects and normal-weight subjects using nonpaired t tests. In the multivariate analyses, logistic regression was used to evaluate the effect of C825T genotype on overweight using the following confounding factors as the independent variables: age, alcohol consumption, smoking habit, and habitual exercise. In women, alcohol consumption was not included in the logistic model because there was only a small number of women who reported consuming alcohol regularly (Table 1). Because other studies have shown that the effect of the 825T allele on BMI is not seen in individuals with high physical activity, but occurs only in subjects with almost zero physical activity (9), we reanalyzed the genotypic and allelic frequencies of C825T in the overweight and normal-weight subjects using the χ2 test (genotype) and Fisher's exact test (allele) grouped according to gender and habitual exercise (Appendix 2; available at Obesity Research Online: http:www.obesityresearch.org). In addition, we analyzed the effect of the C825T genotype in overweight subjects by multivariate logistic regression grouped according to gender and habitual exercise using the same confounding factors as dependent variables (Appendix 2; available at Obesity Research Online: http:www.obesityresearch.org). The analyses were performed with SPSS 10.0J and SamplePower 1.0 software (SPSS Inc., Chicago, IL). Significance was considered as p < 0.05.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Research Methods and Procedures
  4. Acknowledgment
  5. References

This study was financially supported by grants from the Japan Society for the Promotion of Science [Grants-in-Aid for Scientific Research: Encouragement of Young Scientists (A) no. 12770175 and Encouragement of Young Scientists (B) no. 14770163].

Footnotes
  • 1

    Nonstandard abbreviations: GNB3, G-protein β3 subunit; OR, odds ratio; PCR, polymerase chain reaction.

References

  1. Top of page
  2. Abstract
  3. Research Methods and Procedures
  4. Acknowledgment
  5. References
  • 1
    Siffert, W., Rosskopf, D., Siffert, G., et al (1998) Association of a human G-protein beta3 subunit variant with hypertension. Nat Genet. 18: 4548.
  • 2
    Siffert, W., Forster, P., Jockel, K. H., et al (1999) 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. 10: 19211930.
  • 3
    Schunkert, H., Hense, H. W., Doring, A., Riegger, G. A., Siffert, W. (1998) Association between a polymorphism in the G protein beta3 subunit gene and lower renin and elevated diastolic blood pressure levels. Hypertension. 32: 510513.
  • 4
    Benjafield, A. V., Jeyasingam, C. L., Nyholt, D. R., Griffiths, L. R., Morris, BJ. (1998) G-protein beta3 subunit gene (GNB3) variant in causation of essential hypertension. Hypertension. 32: 10941097.
  • 5
    Beige, J., Hohenbleicher, H., Distler, A., Sharma, AM. (1999) G-Protein beta3 subunit C825T variant and ambulatory blood pressure in essential hypertension. Hypertension. 33: 10491051.
  • 6
    Hengstenberg, C., Schunkert, H., Mayer, B., et al (2001) Association between a polymorphism in the G protein beta3 subunit gene (GNB3) with arterial hypertension but not with myocardial infarction. Cardiovasc Res. 49: 820827.
  • 7
    Dong, Y., Zhu, H., Sagnella, G. A., Carter, N. D., Cook, D. G., Cappuccio, FP. (1999) Association between the C825T polymorphism of the G protein beta3-subunit gene and hypertension in blacks. Hypertension. 34: 11931196.
  • 8
    Siffert, W. (2003) G-protein beta3 subunit 825T allele and hypertension. Curr Hypertens Rep. 5: 4753.
  • 9
    Gutersohn, A., Naber, C., Muller, N., Erbel, R., Siffert, W. (2000) G protein beta3 subunit 825 TT genotype and post-pregnancy weight retention. Lancet. 355: 12401241.
  • 10
    Brand, E., Wang, J. G., Herrmann, S. M., Staessen, JA. (2003) An epidemiological study of blood pressure and metabolic phenotypes in relation to the Gbeta3 C825T polymorphism. J Hypertens. 21: 729737.
  • 11
    Dishy, V., Gupta, S., Landau, R., et al (2003) G-protein beta3 subunit 825 C/T polymorphism is associated with weight gain during pregnancy. Pharmacogenetics. 13: 241242.
  • 12
    Hegele, R. A., Anderson, C., Young, T. K., Connelly, PW. (1999) G-protein beta3 subunit gene splice variant and body fat distribution in Nunavut Inuit. Genome Res. 9: 972977.
  • 13
    Benjafield, A. V., Lin, R. C., Dalziel, B., Gosby, A. K., Caterson, I. D., Morris, BJ. (2001) G-protein beta3 subunit gene splice variant in obesity and overweight. Int J Obes Relat Metab Disord. 25: 777780.
  • 14
    Hinney, A., Geller, F., Neupert, T., et al (2001) No evidence for involvement of alleles of the 825-C/T polymorphism of the G-protein subunit beta 3 in body weight regulation. Exp Clin Endocrinol Diabetes. 109: 402405.
  • 15
    Poston, W. S., Haddock, C. K., Spertus, J., et al (2002) Physical activity does not mitigate G-protein-related genetic risk for obesity in individuals of African descent. Eat Weight Disord. 7: 6871.
  • 16
    Ohshiro, Y., Ueda, K., Wakasaki, H., Takasu, N., Nanjo, K. (2001) Analysis of 825C/T polymorphism of G proteinbeta3 subunit in obese/diabetic Japanese. Biochem Biophys Res Commun. 286: 678680.
  • 17
    Rosskopf, D., Busch, S., Manthey, I., Siffert, W. (2000) G protein beta 3 gene: structure, promoter, and additional polymorphisms. Hypertension. 36: 3341.
  • 18
    Rosskopf, D., Manthey, I., Siffert, W. (2002) Identification and ethnic distribution of major haplotypes in the gene GNB3 encoding the G-protein beta3 subunit. Pharmacogenetics. 12: 209220.
  • 19
    Nishimura, N., Nakayama, T., Tonoike, H., et al (2002) Various applications of direct PCR using blood samples. Clin Lab. 48: 377384.