This Article Abstract Full Text (PDF) All Versions of this Article: 91/9/3446    most recent Author Manuscript (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hansen, L. Articles by Reinhardt, R. R. Search for Related Content PubMed PubMed Citation Articles by Hansen, L. Articles by Reinhardt, R. R. Related Collections Diabetes and Insulin

The Pro12Ala Variant of the PPARG Gene Is a Risk Factor for Peroxisome Proliferator-Activated Receptor-/ Agonist-Induced Edema in Type 2 Diabetic Patients

Novo Nordisk A/S (L.H., R.T.P.), DK-2880 Bagsværd, Denmark; Novo Nordisk Pharmaceuticals Inc. (K.W., R.R.R.), Princeton, New Jersey 08540; Genaissance Pharmaceuticals, a wholly owned subsidiary of Clinical Data, Inc. (M.A.), New Haven, Connecticut 06524; and Royal Veterinary and Agricultural University (C.T.E.), DK-1871 Copenhagen, Denmark

Address all correspondence and requests for reprints to: Lars Hansen, M.D., D.M.S.C., Krogshoejvej 53A, 9E2.48, DK-2880 Bagsvaerd, Denmark. E-mail: larh{at}novonordisk.com.

	Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Context: Activation of peroxisome proliferator-activated receptors (PPARs)- by thiazolidinediones (pioglitazone, rosiglitazone) and dual-acting PPAR/ agonists (pargluva, ragaglitazar) is a widely used pharmacological principle to treat insulin resistance and type 2 diabetes. Clinically, however, fluid retention and edema are worrying side effects with these drugs.

Objective: The objective of the present study was to investigate any variation in the PPARG and PPARA genes associated with the risk of fluid retention and development of peripheral edema in patients with type 2 diabetes treated with the dual-acting PPAR/ agonist ragaglitazar.

Design: Single-nucleotide polymorphism and haplotype analyses of the PPARA and PPARG genes were performed on DNA obtained from 345 type 2 diabetic patients randomized to 26-wk monotherapy with the dual-acting PPAR/ agonist ragaglitazar.

Results: At 26 wk, edema was recorded in 48 of the patients (14%) treated with ragaglitazar, and Cox regression analyses identified the common Pro12Ala variant of the PPARG gene as biologically the most important risk factor (hazard ratio 4.42, P = 0.0081) for edema. Other risk factors included female gender (hazard ratio 3.34, P = 0.0005) and weight change during treatment (hazard ratio 1.20, P = 0.0017).

Conclusions: A population-attributable risk of approximately 50% for the Pro12Pro genotype indicates that testing for the Pro12Ala of the PPARG gene in addition to the already identified clinical risk factors may become a useful tool to further reduce the risk of PPAR agonist-induced fluid retention and edema in patients with type 2 diabetes.

	Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
THE INSULIN-SENSITIZING effect of peroxisome proliferator-activated receptor (PPAR)- agonists pioglitazone (Actos, Eli Lilly and Co., Indianapolis, IN) and rosiglitazone (Avandia, GlaxoSmithKline, Harlow, Essex and Greenford, Middlesex, UK) belonging to the class of thiazolidinediones (TZDs) has been established as a useful target in the treatment of insulin-resistant type 2 diabetes (1). Clinically, however, there is growing recognition that fluid retention and edema are common side effects with these drugs (www.fda.gov/medwatch/safety/2002/summary-actos-avandia.pdf), and concerns have been raised as to the importance of this effect on the development of congestive heart failure. At present, no predictive tests for this TZD-induced fluid retention and risk of edema are available, but the American Heart Association and American Diabetes Association have identified clinical risk factors in a consensus statement (2).

Physiological studies in healthy subjects have shown that pioglitazone-induced fluid retention is secondary to increased renal salt reabsorption (3). In addition, clinical studies in patients with type 2 diabetes treated with rosiglitazone indicate that the increase in renal salt reabsorption occurs in the collecting ducts because rosiglitazone-induced edema is amenable to spironolactone rather than loop diuretic treatment (4). Further substantive evidence of a direct renal effect of PPAR activation comes from mouse models in which deletion of the PPARG in the collecting ducts protects these animals from TZD-induced fluid retention (5, 6).

In humans, a splice variant of the PPARG gene, PPARG2, contains a common amino acid polymorphism Pro12Ala (carrier frequencies 8–20%, depending on the population) that reduces the ligand-induced activity of the PPAR protein by 30–50%, depending on the cell lines (7, 8). Other, very rare mutations in the PPARG gene are associated with extreme insulin resistance, familial partial lipodystrophy, and severe diabetes in humans (9), and these mutations also have attenuated transcriptional responses to activation with pioglitazone and rosiglitazone (10).

Because peripheral edema is an important clinical sign of fluid retention, we therefore hypothesized that inherited differences in PPAR activity caused by the common Pro12Ala variant could be involved in the development of the likewise relatively common side effect of peripheral edema that we previously observed during treatment of type 2 diabetes patients with the dual PPAR/ agonist ragaglitazar in a phase 2 dose-finding clinical trial (11). In the present study (a phase 3 clinical trial), we used a data set of 345 type 2 diabetes patients randomized to monotherapy with either 1 or 2 mg/d of ragaglitazar for up to 26 wk. At 26 wk, edema (predominantly peripheral/pedal) was recorded in 48 of the patients (14%) treated with ragaglitazar, compared with nine patients (2.6%) of 344 in the control groups randomized to either 2000 mg/d metformin or 20 mg/d glyburide. Because ragaglitazar is a dual PPAR/ agonist, we genotyped both the PPARG and the PPARA gene for single nonsynonymous (Pro12Ala of PPARG and Leu162Val of PPARA) as well as synonymous and noncoding single-nucleotide polymorphisms (SNPs) across the genes to generate haplotypes of each gene. The effect of the PPARG and PPARA on risk of edema was evaluated by logistic and Cox regression analyses.

	Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Subjects

Five hundred fifty-four patients with type 2 diabetes, who consented to give DNA, were randomized in a double-blind, parallel, active-controlled trial to monotherapy for 52 wk with 1 mg/d of the dual-acting PPAR/ agonist ragaglitazar (n = 167), 2 mg/d of the dual-acting PPAR/ agonist ragaglitazar (n = 178), 20 mg/d diabeta (glyburide) (n = 99), or 2000 mg/d glucophage (metformin) (n = 110). Only the 345 patients randomized to ragaglitazar monotherapy were included in the present time-to-edema analyses (Table 1). After 26 wk of treatment, the trial was stopped prematurely and the whole ragaglitazar development program was discontinued due to preclinical findings.

View this table: [in this window] [in a new window]   TABLE 1. Baseline characteristics according to Pro12Ala genotype of type 2 diabetes patients randomized to 1 or 2 mg/d of dual PPAR/ agonist treatment

Edema was reported as AEs at the discretion of the investigator, and there was no formal definition or grading system applied (ankle circumference). The investigators and patients were told before enrollment that edema was an expected finding for this class of drugs (14), and changes in this area were solicited at each office visit. If a patient complained of edema or weight gain, then the investigator would evaluate the complaint and decide whether it fit the definition of an AE. For each edema event, the investigator recorded the following parameters on the adverse event form: description of event, onset and outcome date, severity, relation to trial product, changes to trial product, outcome, and whether the event was serious. All the AEs were coded using the preferred term according to Novo Nordisk Adverse Reaction Dictionary (based on the World Health Organization Adverse Reaction Dictionary).

Genotyping

Genotyping of Pro12Ala was performed by allele-specific PCR on DNA extracted from frozen whole-blood samples (Genaissance Pharmaceuticals Inc., New Haven, CT). Haplotypes were generated by MALDI-TOF genotyping of seven SNPs in PPARG (including Pro12Ala) and six single nucleotide polymorphisms in PPARA. Haplotypes were estimated by SNPHap (http://www-gene.cimr.cam.ac.uk/clayton/software/snphap.txt) and were numbered in a descending order according to frequencies of the most common: h1, h2, h3, h4, and a reference hx consisting of all other haplotypes.

Statistics

Logistic regression including age, body mass index (BMI), Pro12Ala (PPARG), Leu162Val (PPARA), treatment duration, ethnicity, gender, and treatment dose as covariates was used as a first model to evaluate risk factors for edema. A Cox regression was then used to model time to edema and further evaluation of the time dependence of risk factors. Our initial model of the hazard of edema included both time-independent continuous covariates from baseline (age, BMI, baseline measurements of weight, diastolic blood pressure, triglycerides, free fatty acids, high-density lipoprotein cholesterol, white blood cell count, and absolute neutrophil count) and time-independent categorical covariates [sex (male/female), treatment (1 or 2 mg), Pro12Ala polymorphism (mutant or wild type), race (Caucasian, Hispanic, or other), and an interaction between Pro12Ala and sex] together with time-dependent continuous covariates as changes in weight, diastolic blood pressure, triglycerides, free fatty acids, high-density lipoprotein cholesterol, white blood cell count, and absolute neutrophil count. Population-attributable risk was calculated using an odds ratio (OR) estimate of the Pro12Ala (OR 2.68) from the logistic regression (with correction for gender and BMI): population-attributable risk = p(OR – 1)/1 + p(OR – 1), where p is the prevalence of the Pro12Pro genotype in the nonedema group.

	Results

Top Abstract Introduction Subjects and Methods Results Discussion References

 
PPARG (Pro12Ala) and risk of edema

In the 345 patients randomized to ragaglitazar, 19.5% of the patients carried the Pro12Ala variant. Logistic regression analysis (Fig. 1) indicated a reduced risk of edema in carriers of the Pro12Ala variant (P = 0.05), whereas baseline BMI (P = 0.01), gender (P < 0.01, and high dose (2 mg) (P = 0.01) associated with an increased risk of edema. We then hypothesized that the most susceptible patients would develop edema earlier, so when taking time to event into account (Fig. 2), Cox regression models identified the Pro12Pro genotype as the biologically most important of the risk factors for ragaglitazar-induced edema with a hazard ratio of 4.42 (P = 0.0081) and female gender with a hazard ratio of 3.34 (P = 0.0005). Also, weight gain from baseline contributed to the risk of edema with a hazard ratio of 1.20 (P = 0.0017), whereas the least fall in absolute neutrophil leukocyte count reduced the risk with a hazard ratio of 0.738 (P = 0.0063). In the first round, ragaglitazar dose (1 or 2 mg/d) and baseline BMI also increased the risk of edema, but their contributions became nonsignificant when the time-dependent variable weight change during treatment was included in the model (Table 2). Ethnicity did not contribute to the risk in any of the models.

View larger version (13K): [in this window] [in a new window]   FIG. 1. Logistic regression on the impact of the Pro12Ala variant of the PPARG gene and baseline BMI on the probability of developing edema in type 2 diabetic patients during treatment with the dual-acting PPAR/ agonist ragaglitazar. OR estimate corrected for gender and BMI is 2.68 for the Pro12Pro genotype. Wt, Pro12Pro genotype (n = 152 females + 126 males); Mt, Pro12Ala and Ala12Ala genotypes (n = 36 females + 31 males).

View larger version (11K): [in this window] [in a new window]   FIG. 2. Kaplan-Meier plot showing the impact of the Pro12Ala variant in the PPARG gene on the time (in days) to development of edema in type 2 diabetic patients during treatment with the dual-acting PPAR/ agonist ragaglitazar. Wt, Pro12Pro genotype (n = 152 females + 126 males); Mt, Pro12Ala and Ala12Ala genotypes (n = 36 females + 31 males).

View this table: [in this window] [in a new window]   TABLE 2. Maximum likelihood estimates of risk factors for PPAR / agonist-induced edema in patients with type 2 diabetes

To exclude the possibility that our findings could be attributed to a random SNP effect, we genotyped seven different SNPs covering the promoter and coding region of the PPARG (including the Pro12Ala). This procedure, based on all available genotypes (Tables 3 and 4), generated the haplotypes h1 (n = 489), h2 (n = 125), h3 (n = 54), h4 (n = 23), and hx (n = 55) of the PPAR2 gene. On a closer look (Table 4), it became evident that the 77 alleles of the 12Ala variant that we previously had analyzed as one group in the Cox-regression model could now be split into two groups based on the combination of the adjoining SNPs in the PPARG gene and constituting 70 and 30%, respectively, of the Ala alleles: one group with 54 alleles included in the h3 haplotype and another group with 23 alleles included in the h4 haplotype. When these two groups, the h3 and h4 haplotypes, subsequently were included in the Cox regression model instead of the single Pro12Ala variant, the hazard ratio for developing edema dropped from the initial 4.42 by the Pro12Pro genotype down to 2.77 for the h3 haplotype (representing 70% of the 12Ala alleles). In the absence of an independent data set for replication, this drop in hazard ratio when the Pro12Ala single group is divided into two independent groups (haplotype groups h3 and h4) lends confidence to our data that the Pro12Ala is most likely the risk marker within the PPAR gene.

One nonsynonymous polymorphism (Leu162Val) is known for the PPARA gene, and 7% of the patients carried this variant. Logistic regression did not show association between this variant and the risk of developing edema (P = 0.869), nor did Cox regression (P = 0.913, Table 2). Similarly, when the PPARA gene was analyzed with haplotypes, none of the PPARA haplotypes (Tables 3 and 4) showed any association with the risk of developing edema. Therefore, the present data do not indicate that variation in the PPARA gene plays any role for the pharmacodynamic risk of edema in type 2 diabetes patients treated with a dual PPAR/ agonist.

	Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Ragaglitazar, the drug investigated in the present study, is a dual-acting PPAR/ agonist and strictly does not belong to the same TZD class of drugs as the PPAR agonists pioglitazone (Actos) and rosiglitazone (Avandia) because it also contains significant PPAR (fibrate) activity. However, both PPAR agonists (2, 3, 4) and dual-acting PPAR/ agonists like ragaglitazar and muraglitazar, also known as Pargluva (www.fda.gov/ohrms/dockets/ac/05/briefing/2005-4169B2_01_01-BMS-Pargluva.pdf), share the same side effect of fluid retention and edema.

In accordance with the mouse models (5, 6), our data confirm at the genetic level that the fluid retention induced by PPAR/ agonist drugs in humans is mediated through the PPAR-activating component (and not through its PPAR-activating component) and that its severity (edema) is influenced by variation in the PPARG gene and a 4-fold increase in hazard ratio of edema for patients carrying the Pro12Pro genotype. The 12Ala variant of the PPAR protein has a 30–50% lower biological activity (dependent on the assay), compared with the 12Pro protein (7), and therefore, individuals carrying the Pro12Ala variant may be expected to be less sensitive to PPAR agonist-induced salt retention and also relatively protected against the development of edema.

The hazard ratio of 3.3 for female gender in the present study is similar to the OR of 4.2 recently reported for female gender and risk of rosiglitazone-induced fluid retention (12) and, together with gain in weight, these are well-known clinical risk factors for TZD-induced edema (2). The size of the drop in absolute neutrophil blood cell count, which is a sensitive marker of the antiinflammatory potency of PPAR agonists (13) was also independently associated with increased risk of edema.

Although we cannot exclude that ragaglitazar as a dual-acting PPAR/ agonist acts differently from the classical PPAR agonists, our finding of the Pro12Ala as a pharmacogenetic risk marker of edema in the present study may become clinically relevant as fluid retention continues to be a problem for patients with type 2 diabetes prescribed a PPAR agonist. A conservative estimate of the population-attributable risk based on the prevalence of the risk genotype, Pro12Pro, of approximately 80% in the general population, indicates that more than half (57%) of edemas induced by PPAR agonist (TZDs) treatment could be eliminated if Pro12Pro patients either were not prescribed these drugs or were treated as recommended by the American Diabetes Association and American Heart Association (2).

Biologically, our study emphasizes the importance of PPARG gene variability in a wide range of clinical conditions from rare insulin resistance syndromes (9) to type 2 diabetes (8) and the metabolic syndrome (14). Our study suggests that in addition to the prediction of type 2 diabetes (15), the Pro12Ala may be used as a clinically relevant pharmacogenetic risk marker for edema in patients prescribed a PPAR agonist.

Analysis of the Pro12Ala was not included in two recent major (all together more than 5000 patients) outcome studies of TZD treatment of type 2 diabetes and cardiovascular disease (16, 17), but if our findings in the present study can be replicated, genotyping for the Pro12Ala may offer an attractive tool to optimize treatment with PPAR agonists, increase patient safety, reduce hospitalization and other health contacts, reduce prescription of diuretics (adding their own side effects) due to edema, and thus reduce the overall community cost of treatment with PPAR agonists (18).

	Acknowledgments

 
Special thanks go to all the patients in the NN622-1341 clinical trial who consented to donate blood samples for the present genetic study and the investigators who made this study possible.

	Footnotes

 
First Published Online July 5, 2006

Abbreviations: AE, Adverse event; BMI, body mass index; OR, odds ratio; PPAR, peroxisome proliferator-activated receptor; SNP, single-nucleotide polymorphism; TZD, thiazolidinedione.

Received March 16, 2006.

Accepted June 27, 2006.

	References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. Yki-Järvinen H 2004 Thiazolidinediones. N Engl J Med 351:1106–1118[Free Full Text]
2. 2003 A consensus statement from the American Heart Association and American Diabetes Association: thiazolidinediones use, fluid retention, and congestive heart failure. Circulation 108:2941–2948
3. Zanchi A, Chiolero A, Maillard M, Nussberger J, Brunner H-R, Burnier M 2003 Effects of the peroxisome proliferator-activated receptor- agonist pioglitazone on renal and hormonal responses to salt in healthy men. J Clin Endocrinol Metab 89:1140–1145
4. Karalieddde J, Starkei GM, Lorand DF, Stewart MG, Viberti G 2005 Management of rosiglitazone related fluid retention. Diabetes 54(Suppl 1):A20
5. Zhang H, Zhang A, Kohan DE, Nelson RD, Gonzalez FJ, Yang T 2005 Collecting duct-specific deletion of peroxisome proliferator-activated receptor blocks thiazolidinedione-induced fluid retention. Proc Natl Acad Sci USA 102:9406–9411[Abstract/Free Full Text]
6. Guan YF, Hao C, Cha DR, Rao R, Lu W, Kohan DE, Magnuson MA, Redha R, Zhang Y, Breyer MD 2005 Thiazolidinediones expand body fluid volume through PPAR stimulation of ENAC-mediated renal salt absorption. Nat Med 11:861–866[CrossRef][Medline]
7. Deeb SS, Fajas L, Nemoto M, Pihlajamäki J, Mykkänen L, Kuusisto J, Laakso M, Fujimoto W, Auwerx J 1998 A Pro12Ala substitution in PPAR2 is associated with decreased receptor activity, lower body mass index and improved insulin sensitivity. Nat Genet 20:284–287[CrossRef][Medline]
8. Altshuler D, Hirschhorn JN, Klannemark M, Lindgren CM, Vohl M-C, Nemesh J, Lane CR, Schaffner SE, Bolk S, Brewer C, Tuomi T, Gaudet D, Hudson TJ, Daly M, Groop L, Lander ES 2000 The common PPAR Pro12Ala polymorphism is associated with decreased risk of type 2 diabetes. Nature Genetics 26:76–80[CrossRef][Medline]
9. Barroso I, Gurnell M, Crowley VEF, Agostini M, Schwabel JW, Soos MA, Maslen GLI, Williams TDM, Lewis H, Schafer AJ, Chatterjee VKK, O’Rahilly S 1999 Dominant negative mutations in human PPAR associated with severe insulin resistance, diabetes mellitus and hypertension. Nature 402:880–883[Medline]
10. Agostini M, Gurnell M, Savage DB, Wood EM, Smith AG, Rajanayagam O, Garnes KT, Levinson SH, Xu HE, Schwabe JW, Willson TM, O’Rahilly S, Chatterjee VK 2004 Tyrosine agonists reverse the molecular defects associated with dominant-negative mutations in human peroxisome proliferator-activated receptor . Endocrinology 145:1527–1538[Abstract/Free Full Text]
11. Saad MF, Greco S, Osei K, Lewin AJ, Edwards C, Nunez M, Reinhardt RR 2004 Ragaglitazar improves glycemic control and lipid profile in type 2 diabetic subjects. Diabetes Care 27:1324–1329[Abstract/Free Full Text]
12. Karraliedde J, Hider A, Donaldson J, Stewart M, Viberti G2005 Predictors of rosiglitazone induced fluid retention. Diabetologia 48(Suppl 1):A281; 773
13. van Wijk JP, Cabezas MC, Joven J, Rabelink TJ, Koning EJ 2005 Effects of rosiglitazone on postprandial leucocytes and cytokines in type 2 diabetes. Atherosclerosis 186:152–159
14. Meirhaeghe A, Cottel D, Amouyel P, Dallongeville J 2005 Association between peroxisome proliferator-activated receptor haplotypes and the metabolic syndrome in French men and women. Diabetes 54:3043–3048[Abstract/Free Full Text]
15. Lyssenko V, Almgren P, Anevski D, Orho-Melander M, Sjögren M, Saloranta C, Tuomi I, Groop L 2005 The Botnia Study Group: genetic prediction of future type 2 diabetes. PLOS Med 2:1299–1308
16. Inzucchi SE, Masoudi FA, Wang Y, Kosiborod M, Foody JM, Setaro JF, Havranek EP, Krumholz HM 2005 Insulin-sensitizing antihyperglycemic drugs and mortality after acute myocardial infarction. Diabetes Care 28:1680–1689[Abstract/Free Full Text]
17. Dormandy JA, Charbonnel B, Eckland DJA, on behalf of the PROactive investigators 2005 Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive Study (Prospective Pioglitazone Clinical Trial in Macrovascular Events): a randomised controlled trial. Lancet 366:1279–1289[CrossRef][Medline]
18. Semenkovich CF 2005 TZDs and diabetes: testing the waters. Nat Med 11:822–824[CrossRef][Medline]

This article has been cited by other articles:

				N. Goenka, C. Kotonya, M. D. Penney, H. S. Randeva, and J. P. O'Hare Thiazolidinediones and the renal and hormonal response to water immersion-induced volume expansion in type 2 diabetes mellitus Am J Physiol Endocrinol Metab, April 1, 2008; 294(4): E733 - E739. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 91/9/3446    most recent Author Manuscript (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hansen, L. Articles by Reinhardt, R. R. Search for Related Content PubMed PubMed Citation Articles by Hansen, L. Articles by Reinhardt, R. R. Related Collections Diabetes and Insulin