SEARCH

SEARCH BY CITATION

Dear Sir,

The genetic background of insulin resistance is believed to be polygenic with several genes simultaneously involved and each other interacting. We recently described the association between two APM1 SNPs (i.e. +45T>G and +276G>T, considered either singly or in combination as ‘TG’ haplotype) and insulin resistance-related features in nondiabetic Caucasians from centre-east coast Italy [1]. The +276G>T SNP also predicts serum adiponectin concentrations in the same population [2] and coronary artery disease in type 2 diabetes patients from the same region [3]. The functional missense Pro12Ala SNP in the PPARγ2 gene has been proposed to have a broad impact on the risk of insulin resistance and type 2 diabetes [4]. It has been recently described that the APM1 +45T>G SNP interacts with the PPARγ2 Pro12Ala SNP in modulating insulin resistance in subjects from a Japanese and Chinese hypertensive family cohort [5]. The possibility of such an interaction is indeed reinforced by the presence of a functional PPAR responsive element in the human APM1 promoter [6] and by the observation that troglitazone, an insulin sensitizer, which bind and activate PPARγ2, increases adiponectin serum concentrations and ameliorates insulin resistance in type 2 diabetes patients [7]. The aim of the present work was to verify whether the +45T>G and/or +276G>T SNPs, interact with the PPARγ2 Pro12Ala variant in modulating insulin resistance and related features also in Caucasians. The study group included 574 unrelated, nondiabetic subjects recruited among the employees of the Scientific Institute ‘Casa Sollievo della Sofferenza’ (San Giovanni Rotondo, Italy) who had fasting plasma glucose <7 mmol L−1 and were not taking medications known to interfere with insulin sensitivity and/or glucose metabolism. The study was approved by the local Ethical Committee. Anthropometric measures, mean blood pressure and fasting blood glucose, serum insulin and adiponectin were measured as previously described [1, 2]. APM1 +45T>G and +276G>T and PPARγ2 Pro12Ala polymorphisms were detected as indicated [1, 2]. The effect of genotypes on metabolic traits and other continuous variables were assessed by linear regression analysis using a model including age and gender. The effect of modification between APM1 ‘TG’ haplotype and PPARγ2 Pro12 Ala was tested by introducing an interaction term in the regression model. Fasting insulin, HOMAIR and serum adiponectin were analysed in the logarithms. A P value of <0.05 was considered as significant. SPSS for Windows v. 10.1. (SPSS Inc., Chicago, IL, USA) was used. SNPs +45T>G and +276G>T, considered in combination as the ‘TG’ haplotype were significantly associated with blood pressure, serum insulin and adiponectin levels and the insulin resistance index HOMAIR (Table 1) as previously described in a 30% smaller sample [1, 2]. At variance, Pro12Ala polymorphism had no effect per se on insulin resistance and related variables (Table 1). The same trend of differences observed in the whole population between subjects being or not homozygotes for the APM1 ‘TG’ haplotype on blood pressure, serum insulin and HOMAIR levels were maintained in PPARγ2 Pro12Pro carriers (n = 485) and PPARγ2 X12Ala (n = 65) individuals as well (Table 2), thus clearly indicating no interaction between APM1 and PPARγ2 genes in modulating these variables. Differences in serum adiponectin levels tended to show an opposite trend in the two PPARγ2 subgroups; however, also in this case, no evidence for a significant gene–gene interaction was obtained (Table 2). Worth nothing, with the present sample size, we have 80% power to detect a significant difference (P = 0.05) of 31 pmol L−1 in serum insulin and of 1.1 in HOMAIR index values. Therefore, although the real pathophysiological significance of smaller differences would be questionable, we cannot exclude the presence of a gene–gene interaction able to induce very subtle differences. The apparent discrepancy between our present finding and that of Yang et al. [5] may be explained by the different genetic background and health status of the samples studied (families of Chinese and Japanese hypertensive patients most of them taking antihypertensive medication, as compared with our healthy untreated Caucasians). In this context, it is worth nothing that in our study, in contrast to what observed in the Asian population [5], the PPARγ2 Pro12Ala polymorphism had no effect per se on insulin resistance and related variables. In addition, dietary differences (e.g. food intake and/or fatty acid composition of Asian versus Mediterranean diet) and their possible interaction with the PPARγ2 gene [8] may have played a role in the different results obtained.

Table 1.  Clinical characteristics of nondiabetic unrelated subjects according to carrying status of APM1 45-276 TG haplotype and PPRγ2 Pro12Ala
 Haplotype 45-276PPRγ2 Pro12Ala
TG/TGTG/X + X/XPPro/ProX/AlaP
  1. BMI, body mass index; MBP, mean blood pressure; FBG, fasting blood glucose; HOMAIR, homeostasis model assessment of insulin resistance. Data are mean ± SD. Significances are given after adjusting for age and gender. aSignificance was tested on log-transformed values.

N148 (26.7)407 (73.3) 568 (87.9)78 (12.1) 
M/F 61/87156/2510.55213/35531/470.70
Age (years) 37.8 ± 12.8 36.5 ± 11.50.26 36.9 ± 11.636.1 ± 12.30.58
BMI (kg m−2 ) 26.0 ± 4.5 25.2 ± 4.50.17 25.4 ± 4.426.2 ± 4.90.07
Waist (cm) 84.5 ± 11.7 82.2 ± 13.20.14 82.6 ± 12.583.9 ± 13.70.25
MBP (mmHg) 90.8 ± 9.7 88.4 ± 9.70.019 88.8 ± 9.888.2 ± 9.10.60
FBG (mmol L−1)  5.0 ± 0.5  4.9 ± 0.50.71  4.9 ± 0.54.9 ± 0.40.39
Serum insulin (pmol L−1) 61.7 ± 34.4 53.1 ± 30.70.004a 55.2 ± 31.658.1 ± 31.60.34a
HOMAIR  1.93 ± 1.2  1.65 ± 1.10.006a  1.7 ± 0.71.8 ± 0.50.45a
Adiponectin (μg mL−1) 42.7 ± 28.7 47.5 ± 29.30.045a 46.9 ± 29.640.5 ± 25.40.12a
Table 2.  Clinical characteristics of nondiabetic unrelated subjects according to the different combinations of APM1 45–276 TG haplotype and PPRγ2 Prol2Ala
 Pro12ProX12Ala
TG/TGTG/X + X/XTG/TGTG/X + X/XP*
  1. BMI, body mass index; MBP, mean blood pressure; FBG, fasting blood glucose; HOMAIR: homeostasis model assessment of insulin resistance. Genotypes were available at both positions for 555 study subjects. X denotes any haplotype other than TG. Data are mean ± SD. Significances are given after adjusting for age and gender. a Significance was tested on log-transformed values. *For interaction.

N134 (27.6)351 (72.4)13 (20.0)52 (80.0) 
M/F 54/80133/218 6/722/300.92
Age (years) 37.8 ± 13.2 36.5 ± 11.336.9 ± 9.936.3 ± 13.40.95
BMI (kg m−2 ) 25.9 ± 4.5 25.0 ± 4.426.3 ± 5.226.4 ± 4.90.50
Waist (cm) 84.3 ± 11.9 81.9 ± 12.985.4 ± 9.884.4 ± 15.00.66
MBP (mmHg) 90.8 ± 9.7 88.4 ± 9.789.1 ± 9.488.5 ± 9.60.53
FBG (mmol L−1)  5.0 ± 0.5  4.9 ± 0.54.9 ± 0.34.9 ± 0.40.65
Serum insulin (pmol L−1) 65.8 ± 36.9 56.2 ± 33.571.6 ± 42.460.1 ± 31.30.86a
HOMAIR  1.9 ± 1.2  1.6 ± 1.12.0 ± 1.21.7 ± 1.00.90a
Adiponectin (μg mL−1) 42.6 ± 28.6 48.7 ± 29.946.5 ± 30.437.5 ± 23.20.27a

In conclusion, our data suggest no interaction between APM1 and PPARγ2 genes in modulating insulin resistance and related features in Caucasians from Italy. As for all genotype–phenotype associations, data from several future studies on the same issue have to be accumulated before the results obtained can be claimed as definitive. More specifically, this study clearly underline the problem of trying to investigate interaction between genes whose genotype frequencies are rather rare and indicate the need of very large samples to be tested for this purpose. In this context, our present report might be useful for a future meta-analysis.

Acknowledgements

  1. Top of page
  2. Acknowledgements
  3. Conflict of interest
  4. References

This research was supported by Italian Ministry of Health grants Ricerca Finalizzata 2001, 2002, 2003 (to V. Trischitta) and Ricerca Corrente 2003, 2004 (to C. Menzaghi), NIH grant HL73168 and a Grant-in-Aid from the American Heart Association (to A. Doria), the Diabetes Genome Anatomy Project and by the Genetics Core of the DERC at the Joslin Diabetes Center.

Conflict of interest

  1. Top of page
  2. Acknowledgements
  3. Conflict of interest
  4. References

No conflict of interest was declared.

C. Menzaghi 1 , T. Ercolino 2 , L. Salvemini 1 , A. Coco 1 , G. Fini 1 , R. Di Paola 1 , A. Doria 2 , V. Trischitta 1,31 Unit of Endocrinology, Scientific Institute Casa Sollievo della Sofferenza, San Giovanni Rotondo (FG), Italy , 2 Research Division Joslin Diabetes Center, and Department of Medicine, Harvard Medical School, Boston, MA, USA and 3 Department of Clinical Sciences, University La Sapienza, Rome, Italy

References

  1. Top of page
  2. Acknowledgements
  3. Conflict of interest
  4. References
  • 1
    Menzaghi C, Ercolino T, Di Paola R et al. A haplotype at the adiponectin locus is associated with obesity and other features of the insulin resistance syndrome. Diabetes 2002; 51: 230612.
  • 2
    Menzaghi C, Ercolino T, Salvemini L et al. Multigenic control of serum adiponectin levels: evidence for a role of the APM1 gene and a locus on 14q13. Physiol Genomics 2004; 19: 1704.
  • 3
    Bacci S, Menzaghi C, Ercolino T et al. The +276 G/T single nucleotide polymorphism of the Adiponectin gene is associated with coronary artery disease in type 2 diabetic patients. Diabetes Care 2004; 27: 201520.
  • 4
    Stumvoll M, Haring H. The peroxisome proliferator-activated receptor-gamma2 Pro12Ala polymorphism. Diabetes 2002; 51: 23417.
  • 5
    Yang WS, Hsiung CA, Ho LT et al. Sapphire Study Group Genetic epistasis of adiponectin and PPARγ2 genotypes in modulation of insulin sensitivity: a family-based association study. Diabetologia 2003; 46: 97783.
  • 6
    Iwaki M, Matsuda M, Maeda N et al. Induction of adiponectin, a fat-derived antidiabetic and antiatherogenic factor, by nuclear receptors. Diabetes 2003; 52: 165563.
  • 7
    Phillips SA, Ciaraldi TP, Kong AP et al. Modulation of circulating and adipose tissue adiponectin levels by antidiabetic therapy. Diabetes 2003; 52: 66774.
  • 8
    Luan J, Browne PO, Harding AH et al. Evidence for gene-nutrient interaction at the PPARgamma locus. Diabetes 2001; 50: 6869.