FTO T/A and Peroxisome Proliferator-Activated Receptor-γ Pro12Ala Polymorphisms but Not ApoA1 −75 Are Associated with Better Response to Lifestyle Intervention in Brazilians at High Cardiometabolic Risk

Abstract

Background:

The role of obesity-related polymorphisms on weight loss and inflammatory responses to interventions is unclear. We investigated associations of certain polymorphisms with response to a lifestyle intervention.

Methods:

This 9-month intervention on diet and physical activity included 180 Brazilians at high cardiometabolic risk, genotyped for the fat mass and obesity-associated (FTO) T/A, peroxisome proliferator-activated receptor-γ (PPARγ) Pro12Ala, and ApoA1 −75G/A polymorphisms. Changes in metabolic and inflammatory variables were analyzed according to these polymorphisms.

Results:

The intervention resulted in lower energy intake and higher physical activity. Anthropometric measurements, 2-hr plasma glucose, insulin, high-density lipoprotein cholesterol (HDL-C), and apolipoprotein B (ApoB) improved significantly for the total sample, and these benefits were similar among genotypes. Only variant allele carriers of FTO T/A decreased fasting plasma glucose after intervention (99.9±1.3 to 95.6±1.4 mg/dL, P=0.021). Mean blood pressure reduced after intervention in variant allele carriers of the PPARγ Pro12Ala (109.4±2.1 to 101.3±2.1 mmHg, P<0.001). Improvement in lipid variables was not significant after adjustment for medication. Only the reference genotype of PPARγ Pro12Ala increased apolipoprotein A1 (ApoA1) after intervention (134.3±2.4 to 140.6±2.3 mg/dL, P<0.001). Only variant allele carriers of FTO reduced C-reactive protein (CRP) concentration (0.366±0.031 to 0.286±0.029 mg/dL, P=0.023).

Conclusion:

In Brazilian individuals, the FTO T/A polymorphism induces a favorable impact on inflammatory status and glucose metabolism. The reference genotype of PPARγ Pro12Ala seems to favor a better lipid profile, while the variant allele decreases blood pressure. Our data did not support benefits of the variant allele of ApoA1 −75G/A polymorphism. Further studies are needed to direct lifestyle intervention to subsets of individuals at cardiometabolic risk.

Introduction

E pidemics of noncommunicable chronic diseases have encouraged programs of lifestyle modifications.^1,2 Due to the well-known impact of certain dietary patterns on health, the majority of the initiatives are based on dietary changes aiming at controlling weight gain. Some nutrients such as saturated and trans-fatty acids, as well as increased body adiposity are able to trigger inflammatory mechanisms that contribute to deteriorate insulin sensitivity.^3
–5 In parallel to environmental factors, genetic characteristics may be determinants of obesity and/or may influence the response to interventions on lifestyle to prevent co-morbidities. In fact, some heterogeneous responses have been observed in lifestyle interventions worldwide.⁶

Several single-nucleotide polymorphisms (SNPs) have been described in association with the risk for obesity and metabolic diseases.^7
–9 There is consistent evidence that the cluster of diseases such as type 2 diabetes mellitus (T2DM), hypertension, and dyslipidemia may occur via inflammation and insulin resistance.^10
–12 However, little is known about the influence of certain SNPs on the weight loss and inflammatory responses to intervention programs.

Among the genetic determinants of obesity, the SNP in intron 1 of the fat mass and obesity-associated (FTO) gene is one of the most studied. The FTO rs9939609 (T/A) polymorphism was shown to be associated with both obesity and T2DM in many populations, including the Brazilian.^13,14 The FTO–diabetes association is abolished by adjustment for body mass index (BMI), suggesting that this association is mediated through its effect on adiposity.^4,8 The extent to which this inherited predisposition to obesity may influence the response to dietary interventions is scarcely investigated.

Also, the interest on SNPs regarding the nuclear hormone receptor peroxisome proliferator-activated receptor-gamma (PPAR-γ) is justified considering its role in regulating adipocyte differentiation, lipid and glucose homeostasis, and insulin sensitivity.^15
–17 Among these SNPs, the PPARγ Pro12Ala polymorphism, the isoform expressed only in adipose tissue, has been associated with lower BMI, enhanced insulin sensitivity, and lower risk for T2DM.^18,19 To our knowledge, it is not clear how carriers of the PPARγ Pro12Ala polymorphism respond to intervention on dietary habits and physical activity.

Because apolipoproteins are indicative of cardiovascular risk, the impact of polymorphisms in these genes has been investigated. Apolipoprotein A1 (ApoA1) is the major high-density lipoprotein (HDL)-associated apolipoprotein. Whether polymorphisms in the ApoA1 gene are associated with increased ApoA1 transcription and lipid metabolism disturbances is controversial. The −75G/A SNP was associated with increased ApoA1 transcription in some studies^20,21 but not in others.^22,23 Scarce data are available on its association with glucose metabolism and its relationship with response to dietary interventions.²⁴

Assuming that part of the heterogeneous response to lifestyle interventions among populations may be explained by genotypic factors, we investigated whether certain metabolic diseases–related polymorphisms influence the response profile of Brazilian individuals. This study assessed the association of the FTO T/A, PPARγ Pro12Ala, and ApoA1 −75G/A SNPs with changes in metabolic variables and C-reactive protein (CRP) following lifestyle intervention in Brazilians at high cardiometabolic risk.

Methods

This was a 9-month intervention study on dietary habits and physical activity for prevention of T2DM carried out during 2009–2010. The study protocol was approved by the local ethical committee, and individuals provided informed consent before entering the study. Individuals aged from 18 to 80 years were screened at the Healthcare Unit of the School of Public Health, University of São Paulo, Brazil. Those with at least one risk factor were invited to a clinical examination and laboratory tests including a 75-gram oral glucose tolerance test (OGTT). Inclusion criteria were prediabetes (impaired fasting glucose, defined as fasting glucose between 100 and 125 mg/dL, and/or impaired glucose tolerance as 2-hr glucose between 140 and 200 mg/dL) or metabolic syndrome criteria according to International Diabetes Federation definition for Latin American populations.²⁵ For the purpose of this study, only two categories of skin colors were defined —white and non-white—the latter being composed of mulattos and blacks. Asian descendants were excluded, as well as those with a medical history of neurological or psychiatric disturbances, thyroid, liver, renal, and infectious diseases. From a total of 230 eligible individuals, 180 agreed to participate. Thirty-four participants were lost during follow-up and 8 DNA samples with low quality that resulted in poor genotyping calls were excluded.

The intervention on lifestyle consisted of medical consultations every 3 months, where counseling for changing living habits was given. Our intervention goals were similar to those previously proposed.^1,2 Individuals also attended group sessions, where topics on healthy diet, physical activity, and psychosocial stress management were discussed with a multiprofessional team. At the end of the ninth month, examinations were repeated. Compliance rate to the intervention program did not differ between subgroups of individuals stratified according to the genotypes.

Dietary data were collected using 24-hr food recalls by trained nutritionists. Physical activity level was assessed by the long version of the international Physical Activity Questionnaire.²⁶ Individuals that achieved 150 min per week of total physical activity were considered active.

Height and weight were measured using standard protocols. BMI was subsequently calculated. Waist circumference (WC) was measured at the midpoint between the bottom of the rib cage and above the top of the iliac crest during minimal respiration. Blood pressure was measured three times, in a sitting position, by automatic device (Omron model HEM-712C, Omron Health Care, Inc, USA). Mean values of the two last measurements were considered for calculation of mean blood pressure (MBP): Systolic blood pressure+(diastolic blood pressure×2)/3.

Venous blood samples were collected after overnight fasting and immediately centrifuged. Plasma glucose was measured by the glucose-oxidase method, and lipoproteins were determined enzymatically by automatic analyzer. Aliquots were frozen at −80°C for further determinations of apolipoproteins, inflammatory markers, and for genotyping. ApoA1 and ApoB concentrations were determined using turbidimetry. CRP and interleukin 6 (IL-6) concentrations were determined by immunoenzyme chemiluminescent assay (Immulite, Los Angeles, CA). Serum insulin was determined by immunometric assay using a quantitative chemiluminescent kit (AutoDelfia, Norton OH). Whole blood was used for DNA extraction for determination of the SNPs: FTO T/A (rs9939609), PPARγ Pro12Ala (rs1801282), and ApoA1 −75G/A (rs670).

Genotyping

Genomic DNA was extracted from whole blood using the salting out method.²⁷ After extraction, the viability of the extracted material was visualized in a 1% agarose gel, and a concentration was obtained by spectrophotometer Nanodrop 8000 (Thermo Scientific, Waltham, MA). Genotyping of the selected polymorphisms was performed using an allele-specific polymerase chain reaction or amplification-refractory mutation.²⁸ Detection of the products was made using fluorescence resonance energy.²⁹

Statistical analysis

Clinical and laboratory data were expressed as means and mean standard errors (SE). A chi-squared test was used to evaluate the Hardy–Weinberg equilibrium (P>0.05). We applied generalized estimating equations (GEE) for longitudinal data³⁰ for 13 biomarkers separately. All models were controlled for age and sex, as well as for medication when necessary. For all models, the effects of the intervention (before vs. after 9 months) of the polymorphisms (with vs. without), as well as of the interaction between the intervention and the presence of the polymorphism, were evaluated. When interaction was detected, comparisons before and after intervention by polymorphism categories were made; when it was not detected, we performed the comparison for whole sample only. A Bonferroni correction for pairwise comparison was used. We considered the normal distribution (with link identity) for weight, waist circumference, mean blood pressure, low-density lipoprotein cholesterol (LDL-C), ApoA1, ApoB, fasting, and 2-hr plasma glucose; and gamma distribution (with link logarithm) for triglycerides, total cholesterol, HDL-C, insulin, and CRP. The effects of specific medication for mean blood pressure, as well as lipid-lowering medication for triglycerides, total cholesterol, HDL, LDL, ApoB, and ApoA1, were also considered. Statistical analyses were performed using the PASW Statistics 18 for Windows.

Results

In the sample of 138 individuals, 66.7% were women and 30.4% nonwhite; the mean age was 56.6±0.99 years and BMI was 30.4±0.48 kg/m². A total of 62.3% had prediabetes and 90.6% had metabolic syndrome. They decreased energy intake (1814±58 to 1530±48 kcal, P<0.001) and increased total physical activity (186±14 to 220±575 min/week, P<0.001) after intervention. Genotype groups were in Hardy–Weinberg equilibrium.

The frequency of the variant allele of FTO T/A in the sample was 42.5%; the distribution of TT, TA, and AA genotypes was 36.3%, 42.2%, and 21.5%, respectively. The variant allele of PPARγ Pro12Ala was present in only 8% of the total sample and distribution of ProPro, ProAla, and AlaAla genotypes was 83.0%, 16.2%, and 0.8%, respectively. The frequency of the variant allele ApoA1 −75G/A was 16% and the distribution of GG, GA, and AA genotypes was 68.7%, 30.5%, and 0.8%, respectively. In 34 individuals who were lost to follow up, the frequencies of the variant alleles and genotypes distributions were similar. The individuals with at least one variant allele were grouped and compared with those with the reference genotype. Weight, waist circumference, 2-hr plasma glucose, HDL-C, ApoB, and insulin decreased significantly after the 9-month intervention in the total sample, but no significant interaction between the intervention and the genotypes of FTO T/A, PPARγ ProAla, and Apo1 −75G/A after adjustments was found (Tables 1, 2, and 3).

Table 1.

Mean Values (SE) of Variables in Baseline and After 9 Months of Lifestyle Intervention According to FTO T/A Genotypes

		n	Baseline	9 months	P
Weight (kg)	TT	49	81.1 (2.3)	79.3 (2.3)	—
	TA+AA	85	78.74 (1.5)	76.9 (1.5)	—
	Total sample	134	79.9 (1.3)	78.1 (1.3)	<0.001
Waist circumference (cm)	TT	49	100.6 (1.7)	99.6 (1.7)	—
	TA+AA	85	99.5 (1.3)	98.5 (1.3)	—
	Total sample	134	100.0 (1.1)	99.1 (1.1)	0.008
Mean blood pressure (mmHg)^a	TT	47	109.8 (1.9)	107.3 (1.7)	—
	TA+AA	77	107.7 (1.4)	105.2 (1.3)	—
	Total sample	124	108.7 (1.3)	106.3 (1.1)	0.013
Fasting plasma glucose (mg/dL)^*	TT	49	98.9 (1.5)	99.6 (1.8)	NS
	TA+AA	85	99.9 (1.3)	95.6 (1.4)	0.021
	Total sample	134	99.5 (1.0)	97.6 (1.2)	0.121
2-hr plasma glucose	TT	49	121.0 (3.6)	114.6 (3.8)	—
	TA+AA	85	116.7 (2.9)	110.4 (2.8)	—
	Total sample	134	118.9 (2.5)	112.5 (2.6)	0.007
Triglycerides (mg/dL)	TT	49	158.6 (10.5)	155.4 (10.2)	—
	TA+AA	85	163.4 (11.5)	160.2 (11.1)	—
	Total sample	134	161.0 (9.3)	157.8 (9.0)	NS
Total cholesterol (mg/dL)	TT	49	199.8 (7.2)	196.2 (7.1)	—
	TA+AA	85	201.1 (6.3)	197.5 (6.6)	—
	Total sample	134	200.5 (5.8)	196.8 (6.0)	NS
HDL-C (mg/dL)	TT	49	40.1 (1.5)	44.1 (1.6)	—
	TA+AA	85	41.5 (1.6)	45.6 (1.8)	—
	Total sample	134	40.8 (1.3)	44.9 (1.5)	<0.001
LDL-C (mg/dL)	TT	49	128.0 (6.5)	121.5 (6.6)	—
	TA+AA	84	126.5 (6.1)	120.1 (6.5)	—
	Total sample	133^b	127.3 (5.6)	120.8 (5.8)	NS
Apolipoprotein A1 (mg/dL)	TT	49	133.5 (2.8)	138.3 (2.6)	—
	TA+AA	85^c	135.3 (2.7)	140.2 (2.8)	—
	Total sample	134	134.4 (2.4)	139.2 (2.3)	<0.001
Apolipoprotein B (mg/dL)	TT	49	94.0 (4.0)	82.7 (3.8)	—
	TA+AA	85^c	90.7 (3.5)	79.5 (3.8)	—
	Total sample	134	92.4 (3.2)	81.1 (3.3)	<0.001
Insulin (μUl/mL)	TT	49	11.0 (0.96)	8.7 (.71)	—
	TA+AA	85	10.0 (0.89)	7.9 (.80)	—
	Total sample	134	10.5 (0.67)	8.3 (.58)	<0.001
C-reactive protein (mg/dL)^**	TT	44	.277 (.033)	.332 (.048)	NS
	TA+AA	77	.366 (.032)	.286 (.030)	0.009
	Total sample	121^d	.318 (.023)	.308 (.028)	NS

Model considering a significant interaction between medication and intervention (medication since base line with significant reduction).

At 9 months n=130.

At baseline n=132.

At baseline n=113.

P value of interaction (SNP*intervention)=0.038.

P value of interaction=0.023.

SE, standard error; FTO, fat mass and obesity-associated; NS, not significant; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol.

Table 2.

Mean Values (SE) of Variables at Baseline and After 9 Months of Lifestyle Intervention According to PPARγ Pro12Ala Genotypes

		n	Baseline	9 months	P
Weight (kg)	ProPro	111	79.5 (1.4)	77.6 (1.4)	—
	ProAla+AlaAla	23	79.1 (2.7)	77.3 (2.7)	—
	Total sample	134	79.3 (1.5)	77.5 (1.5)	<0.001
Waist circumference (cm)	ProPro	110	99.7 (1.2)	98.8 (1.2)	—
	ProAla+AlaAla	22	99.3 (2.4)	98.3 (2.3)	—
	Total sample	132	99.5 (1.3)	98.6 (1.3)	0.010
Mean blood pressure (mmHg)^a,*	ProPro	103	108.2 (1.4)	107.1 (1.2)	NS
	ProAla+AlaAla	21	109.4 (2.1)	101.3 (2.1)	<0.001
	Total sample	124	108.9 (1.3)	104.3 (1.2)	<0.001
Fasting plasma glucose (mg/dL)	ProPro	111	100.4 (1.1)	98.0 (1.2)	—
	ProAla+AlaAla	23	96.1 (2.3)	93.7 (2.3)	—
	Total sample	134	98.3 (1.4)	95.8 (1.4)	0.041
2-hr plasma glucose	ProPro	111	119.3 (2.6)	113.2 (2.8)	—
	ProAla+AlaAla	23	115.9 (5.1)	109.8 (4.8)	—
	Total sample	134	117.6 (3.1)	111.5 (2.9)	0.010
Triglycerides (mg/dL)	ProPro	111	161.1 (9.4)	158.4 (9.3)	—
	ProAla+AlaAla	23	158.2 (15.6)	155.5 (14.4)	—
	Total sample	134	159.7 (10.9)	157.0 (10.0)	NS
Total cholesterol (mg/dL)	ProPro	111	201.9 (5.8)	198.4 (5.9)	—
	ProAla+AlaAla	23	187.1 (9.2)	183.8 (9.3)	—
	Total sample	134	194.3 (6.4)	190.9 (6.6)	NS
HDL-C (mg/dL)	ProPro	111^b	41.1 (1.4)	45.2 (1.6)	—
	ProAla+AlaAla	23	39.5 (2.0)	43.5 (2.3)	—
	Total sample	134	40.3 (1.5)	44.3 (1.7)	<0.001
LDL-C (mg/dL)	ProPro	111	128.2 (5.6)	122.4 (5.8)	—
	ProAla+AlaAla	22	115.4 (8.5)	109.7 (8.7)	—
	Total sample	133^c	121.8 (6.2)	116.1 (6.4)	NS
Apolipoprotein A1 (mg/dL)^**	ProPro	111	134.3 (2.4)	140.7 (2.3)	<0.001
	ProAla+AlaAla	23	133.6 (4.2)	131.4 (4.6)	NS
	Total sample	134^d	133.8 (2.7)	136.0 (2.8)	NS
Apolipoprotein B (mg/dL)	ProPro	111	92.4 (3.3)	81.1 (3.4)	—
	ProAla+AlaAla	23	91.0 (5.0)	79.7 (5.0)	—
	Total sample	134^d	91.7 (3.6)	80.4 (3.7)	<0.001
Insulin (μUl/mL)	ProPro	111	10.6 (0.7)	8.4 (0.6)	—
	ProAla+AlaAla	23	9.3 (1.1)	7.4 (0.9)	—
	Total sample	134	10.0 (0.7)	7.9 (0.6)	<0.001
C-reactive protein (mg/dL)	ProPro	101	0.345 (0.025)	0.314 (0.028)	—
	ProAla+AlaAla	21	0.277 (0.049)	0.252 (0.046)	—
	Total sample	122^e	0.309 (0.031)	0.281 (0.031)	NS

Model considering a significant interaction between medication and intervention (medication since base line with significant reduction).

At 9 months n=133.

At 9 months n=130.

At baseline n=132.

At baseline n=114.

P value of interaction (SNP*intervention)=0.002.

P value of interaction=0.009.

SE, standard error; PPARγ, peroxisome proliferator-activated receptor-γ; NS, not significant; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol.

Table 3.

Mean Values (SE) of Variables at Baseline and After 9 Months of Lifestyle Intervention According to ApoA1 −75G/A Genotypes

		n	Baseline	9 months	P
Weight (kg)	GG	89	79.7 (1.5)	77.9 (1.5)	—
	GA+AA	41	78.8 (2.7)	76.9 (2.7)	—
	Total sample	130	79.3 (1.5)	77.4 (1.5)	<0.001
Waist circumference (cm)	GG	89	100.1 (1.2)	99.1 (1.2)	—
	GA+AA	41	98.5 (2.0)	97.5 (2.0)	—
	Total sample^a	130	99.3 (1.2)	98.3 (1.2)	0.005
Mean blood pressure (mmHg)^b	GG	83	109.0 (1.5)	106.3 (1.4)	—
	GA+AA	37	107.6 (1.8)	104.9 (1.7)	—
	Total sample	120	108.3 (1.3)	105.6 (1.1)	0.009
Fasting plasma glucose (mg/dL)	GG	89	99.7 (1.2)	97.5 (1.3)	—
	GA+AA	41	99.5 (1.8)	97.3 (1.9)	—
	Total sample	130	99.6 (1.2)	97.4 (1.2)	NS
2-h plasma glucose	GG	89	117.2 (2.8)	111.4 (3.0)	—
	GA+AA	41	122.1 (4.5)	116.3 (4.4)	—
	Total sample	130	119.7 (2.8)	113.8 (2.8)	0.016
Triglycerides (mg/dL)	GG	89	162.9 (9.9)	160.1 (10.0)	—
	GA+AA	41	159.4 (12.9)	156.7 (12.1)	—
	Total sample	130	161.1 (9.8)	158.4 (9.3)	NS
Total cholesterol (mg/dL)	GG	89	199.9 (6.0)	195.7 (6.0)	—
	GA+AA	41	208.2 (7.9)	203.9 (7.8)	—
	Total sample	130	204.0 (6.1)	199.8 (6.1)	NS
HDL-C (mg/dL)	GG	89	40.1 (1.5)	44.3 (1.6)	—
	GA+AA	41	43.0 (2.2)	47.5 (2.4)	—
	Total sample	130	41.5 (1.5)	45.9 (1.6)	<0.001
LDL-C (mg/dL)	GG	89	127.1 (5.7)	120.8 (5.8)	—
	GA+AA	41	131.6 (7.2)	125.2 (7.4)	—
	Total sample	130	129.3 (5.8)	123.0 (6.0)	NS
Apolipoprotein A1 (mg/dL)	GG	89	133.2 (2.6)	138.1 (2.5)	—
	GA+AA	41	139.2 (3.5)	144.0 (3.5)	—
	Total sample	130	136.2 (2.4)	141.0 (2.3)	<0.001
Apolipoprotein B (mg/dL)	GG	89	93.0 (3.5)	81.3 (3.5)	—
	GA+AA	41	94.1 (4.3)	82.4 (4.3)	—
	Total sample	130	93.5 (3.4)	81.9 (3.4)	<0.001
Insulin (μUl/mL)	GG	89	10.3 (0.7)	7.9 (0.6)	—
	GA+AA	41	11.0 (1.2)	8.5 (0.9)	—
	Total sample	130	10.6 (0.7)	8.2 (0.6)	<0.001
C-reactive protein (mg/dL)	GG	81	0.329 (0.026)	0.307 (0.030)	—
	GA+AA	37	0.331 (0.042)	0.310 (0.041)	—
	Total sample	118^c	0.330 (0.027)	0.308 (0.028)	NS

At 9 months n=128.

Model considering a significant interaction between medication and intervention (medication since base line with significant reduction).

At baseline n=110.

SE, standard error; ApoA1, apolipoprotein A1; NS, not significant; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol.

As shown in Table 1, only variant allele carriers of FTO T/A showed a reduction of fasting plasma glucose after intervention (99.9±1.3 to 95.6±1.4 mg/dL, P=0.021). There was no significant interaction between the intervention and PPARγ Pro12Ala (Table 2), and ApoA1 −75G/A (Table 3) polymorphisms.

Mean blood pressure reduced after the 9-month intervention in variant allele carriers of the PPARγ Pro12Ala (109.4±2.1 to 101.3±2.1 mmHg; P<0.001) but not in the reference genotype group (Table 2). Such a result indicates an interaction between the PPARγ Pro12Ala polymorphism and the intervention. Blood pressure decreased (113.4±1.7 to 105.0±1.6 mmHg; P<0.001) among individuals who were taking medications since the beginning of intervention. No significant interaction between the intervention with the FTO T/A (Table 1) or with the ApoA1 −75 G /A polymorphism (Table 3) was detected.

Decreases in mean values of triglycerides, total cholesterol, and LDL-C after intervention were not significant after adjustment for lipid-lowering medication. Interactions between the intervention with the polymorphisms were not found.

The reference genotype group PPARγ Pro12Ala had a significant rise in ApoA1 concentration after intervention (134.3±2.4 to 140.7±2.3 mg/dL, P<0.001) but not the variant allele carriers group (Table 2), showing an interaction effect between the intervention and this polymorphism. Regarding the FTO T/A and ApoA1 −75G/A polymorphisms, the difference in ApoA1 mean values between baseline and after 9 months were observed only for total sample (Tables 1 and 2). For the ApoA1 −75 G/A polymorphism, despite a higher increase in mean values of ApoA1 concentration in variant allele after intervention, this increase was not different between two groups of polymorphism.

Variant allele carriers of FTO T/A had lower CRP concentrations after intervention (0.366±0.032 to 0.286±0.030 mg/dL, P=0.009), but not the reference genotype group (Table 1), indicating an interaction effect between the intervention with this polymorphism. For the PPARγ Pro12Ala and ApoA1 −75 G/A polymorphisms, no interaction with intervention was detected (Tables 2 and 3).

Discussion

Our results showed diverse responses to intervention when analyzing the sample as a whole and as subgroups stratified according to the SNPs categories, suggesting a role for genotype in the response to lifestyle interventions. The high admixture of the Brazilian population limits adequate classification by ethnic group and the comparisons of our findings with other populations. Contrasting with the great number of reports on the frequencies of SNPs associated with body adiposity and metabolic disturbances, their association with response to intervention on lifestyle is rare. Despite the heterogeneity of our population, the frequencies of the variant alleles studied did not markedly differ from data reported in literature.^31
–33 The FTO T/A, PPARγ Pro12Ala, and ApoA1 −75G/A polymorphisms were selected considering that they have been closely related to adiposity and obesity-associated diseases.

FTO T/A polymorphism

In contrast with case–control and cross-sectional studies conducted with childhood and obese adults in US and European populations,^34,35 Brazilians carriers of the FTO T/A SNP and noncarriers showed similar body adiposity. Our data reinforced previous findings in T2DM Brazilian individuals and in Asian Indian and Caucasian Danish populations, in whom no association of this polymorphism with body adiposity and other components of the metabolic syndrome was detected.^36
–38 The association of FTO T/A SNP with response to lifestyle intervention was investigated in the Finnish Diabetes Prevention Study, similar to our results, in which the variant allele did not modify the effect of lifestyle changes on body weight.³⁹ More recently, data from the Diabetes Prevention Program showed that the A allele of this SNP was associated with long-term weight loss after an intervention on lifestyle.⁴⁰

In contrast with a previous case–control study conducted in a Japanese population, in which the FTO T/A SNP was associated with glucose intolerance,⁴¹ in our sample, its presence seems to induce beneficial metabolic response to intervention. Although values of fasting plasma glucose were within the recommended limits, only variant allele carriers improved glucose metabolism, as their plasma glucose levels reduced after intervention. Moreover, despite similar mean values of weight and waist circumference at baseline and after intervention in both genotype groups, the reduction in CRP concentrations only in variant allele carriers suggests that the presence of SNPs could favor an antiinflammatory response. Contrasting with our results, no effect of the A allele on CRP was shown and an increase in other inflammatory markers was observed in two Scandinavian studies.^38,42

PPARγ Pro12Ala polymorphism

Some studies have shown that the Pro12Ala substitution was associated with obesity in several populations,^43
–45 but there are many others showing opposite results.^46
–48 In our sample of Brazilian individuals, there was no effect of the intervention and no effect of interaction between the intervention and PPARγ Pro12Ala for body adiposity.

Our results suggest an association of the Ala allele with greater reductions in blood pressure after a 9-month intervention compared with the reference genotype. This is in agreement with recent report showing that PPARγ may have pleiotropic vascular effects in addition to its known blood glucose-lowering effect, which are protective against the progression of hypertension and atherosclerosis.⁴⁹ Also, Frederiksen et al.⁴⁶ reported that homozygosity of the Ala allele reduced diastolic blood pressure in a Danish population. In another Caucasian sample, lower blood pressure levels were found in women with the Ala allele⁵⁰; in our sample, gender adjustment did not alter its significant association with mean blood pressure. These findings are opposite to a study in a Chinese population in which Ala allele carriers had a significant higher risk of hypertension than noncarriers.⁵¹

Concerning lipid metabolism, our results showed benefits from the intervention on reference genotype of PPARγ Pro12Ala polymorphism. At baseline and after intervention, both genotypic groups had similar total cholesterol and LDL-C concentrations, but only the reference genotype group increased ApoA1 values. Our findings are discordant with a Chinese study reporting lower cholesterol levels in Ala allele carriers, although no result on ApoA1 was provided.⁵²

Apo A1 −75G/A polymorphism

Several studies conducted in different populations have shown that this polymorphism was associated with benefits in lipid metabolism, specifically with greater ApoA1 concentrations among variant allele carriers.^20,21,53 In contrast with previous cross-sectional studies, at the baseline of our study the mean values of ApoA1 were not significantly higher in variant carriers. An apparent favorable response of lipid profile to intervention of variant allele carriers was not significant after adjusting for lipid-lowering medication, raising the importance of multiple adjustments for adequate interpretation of the impact of SNP on outcomes.

The main limitation was our sample size, particularly considering the multiple adjustments made, but power analyses (data not shown) for each biomarker showed satisfactory power. If low power is the case in our study, lack of significant results could be found. Further studies are needed to verify if this was a problem of power or if the differences are really not present. Strengths of our study include the longitudinal design and the availability of traditional risk factors and inflammatory markers measurements, allowing a comprehensive look at the impact of polymorphisms on the response to intervention on lifestyle.

In summary, we conclude that, in a sample of Brazilian individuals, the FTO T/A polymorphism induces a favorable impact on inflammatory status and glucose metabolism. The reference genotype of PPARγ Pro12Ala seems to favor a better lipid profile, whereas the variant allele decreases blood pressure. Our data did not support benefits on the lipid profile of the variant allele of ApoA1 −75G/A polymorphism. Studies like ours of polymorphisms of genes involved in the control of body adiposity and related metabolic disturbances provide the means to achieving the goal of optimizing the health via nutritional intervention of individuals at cardiometabolic risk.

Footnotes

Acknowledgment

This study was supported by the Foundation for Research Support of the State of São Paulo.

Author Disclosure Statement

No competing financial interests exist.

References

Tuomilehto

, Lindström

, Eriksson

et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. New Engl J Med, 2001; 344:1343–1350.

Knowler

, Barrett-Connor

, Fowler

et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. New Engl J Med, 2002; 346:393–403.

Esposito

, Ceriello

, Giugliano

. Diet and the metabolic syndrome. Metab Syndr Relat Disord, 2007; 5:291–295.

De Lorenzo

, Del Gobbo

, Premrov

et al.

Normal-weight obese syndrome: Early inflammation?

Am J Clin Nutr, 2007; 85:40–45.

Saravanan

, Haseeb

, Ehtesham

et al. Differential effects of dietary saturated and trans-fatty acids on expression of genes associated with insulin sensitivity in rat adipose tissue. Eur J Endocrinol, 2005; 153:159–165.

Ramachandran

, Snehalatha

, Mary

et al.

The Indian Diabetes Prevention Programme shows that lifestyle modification and metformin prevent type 2 diabetes in Asian Indian subjects with impaired glucose tolerance (IDPP-1)

Diabetologia, 2006; 49:289–297.

Curti

MLR

, Jacob

, Borges

et al. Obesity, inflammation and dyslipidemia: implications for nutrigenetics. J Obes, 2011; v2011ID 497401.

Loos

RJF

, Bouchard

. The first gene contributing to common forms of human obesity. Obes Rev, 2008; 9:246–250.

Corella

, Ordovas

. Nutrigenomics in cardiovascular medicine. Circ Cardiovasc Genet, 2009; 2:637–651.

10.

Steinberg

, Paradisi

, Hook

et al. Free fatty acid elevation impairs insulin-mediated vasodilation and nitric oxide production. Diabetes, 2000; 49:1231–1238.

11.

Boden

. Fatty acid-induced inflammation and insulin resistance in skeletal muscle and liver. Curr Diab Rep, 2006; 6:177–181.

12.

Fantuzzi

, Mazzone

. Adipose tissue and atherosclerosis. Arterioscler Thromb Vasc Biol, 2007; 27:996–1003.

13.

Ramos

, Casanova

, Maturana

et al. Variations in the fat mass and obesity-associated (FTO) gene are related to glucose levels and higher lipid accumulation product in postmenopausal women from southern Brazil. Fertil Steril, 2011; 96:974–979.

14.

Hinney

, Vogel

, Hebebrand

. From monogenic to polygenic obesity: Recent advances. Eur Child Adolesc Psychiatry, 2010; 19:297–310.

15.

Jones

. Peroxisome proliferator-activated receptor (PPAR) modulators: Diabetes and beyond. Med Res Rev, 2001; 21:540–552.

16.

Torra

, Chinetti

, Duval

et al. Peroxisome proliferator-activated receptors: From transcriptional control to clinical practice. Curr Opin Lipidol, 2001; 12:245–254.

17.

Danawati

, Nagata

, Moriyama

et al. A possible association of Pro12Ala polymorphism in peroxisome proliferator-activated receptor γ 2 gene with obesity in native Javanese in Indonesia. Diabetes Metab Res, 2005; 21:465–469.

18.

Tonjes

, Stumvoll

. The role of the Pro12Ala polymorphism in peroxisome proliferator-activated receptor gamma in diabetes risk. Curr Opin Clin Nutr Metab Care, 2007; 10:410–414.

19.

Gouda

, Sagoo

, Harding

et al. The association between the peroxisome proliferator-activated receptor-g2 (PPARG2) Pro12Ala gene variant and type 2 diabetes mellitus: A HuGE review and meta-analysis. Am J Epidemiol, 2010; 171:645–655.

20.

Paul-Hayase

, Rosseneu

, Robinson

et al. Polymorphisms in the apolipoprotein (apo) AI-CIII-AIV gene cluster: Detection of genetic variation determining plasma Apo AI, Apo CIII and Apo AIV concentrations. Hum Genet, 1992; 88:439–446.

21.

Talmud

, Ye

, Humphries

. Polymorphism in the promoter region of the apolipoprotein AI gene associated with differences in apolipoprotein AI levels: The European Atherosclerosis Research Study. Genet Epidemiol, 1994; 11:265–280.

22.

Juo

, Wyszynski

, Beaty

et al. Mild association between the A/G polymorphism in the promoter of the apolipoprotein A-I gene and apolipoprotein A-I levels: A meta-analysis. Am J Med Genet, 1999; 82:235–241.

23.

Lopez-Miranda

, Ordovas

, Espino

et al. Influence of mutation in human apolipoprotein A-1 gene promoter on plasma LDL cholesterol response to dietary fat. Lancet, 1994; 343:1246–1249.

24.

Chen

, Mazzotti

, Furuya

et al. Apolipoprotein A1 gene polymorphisms as risk factors for hypertension and obesity. Clin Exp Med, 2009; 9:319–325.

25.

Alberti

, Zimmet

, Shaw

. International Diabetes Federation: A consensus on Type 2 diabetes prevention. Diabetic Med, 2007; 24:451–463.

26.

Craig

, Marshall

, Sjöström

et al. International Physical Activity Questionnaire: 12-country reliability and validity. Med Sci Sports Exerc, 2003; 35:1381–1395.

27.

Miller

, Dykes

, Polesky

. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res, 1988; 16:1215.

28.

Newton

, Heptinstall

, Summers

et al. Amplification refractory mutation system for prenatal diagnosis and carrier assessment in cystic fibrosis. Lancet, 1989; 2:1481–1483.

29.

Livak

, Flood

, Marmaro

et al. Oligonucleotides with fluorescent dyes at opposite ends provide a quenched probe system useful for detecting PCR product and nucleic acid hybridization. PCR Methods Appl, 1995; 4:357–362.

30.

Liang

K-Y

, Zeger

. Longitudinal data analysis using Generalized Linear Models. Biometrika, 1986; 73:13–22.

31.

. PPARg2 Polymorphism and human health. PPAR Res, 2009; 2009:849538.

32.

HapMap (haplotype map of the human genome). http://www.ncbi.nlm.nih.gov/SNP/.

33.

Chen

, Mazzotti

, Furuya

et al. Apolipoprotein A1 gene polymorphisms as risk factors for hypertension and obesity. Clin Exp Med, 2009; 9:319–325.

34.

Scuteri

, Sanna

, Chen

et al. Genome-wide association scan shows genetic variants in the FTO gene are associated with obesity-related traits. PLoS Genet, 2007; 3:e115.

35.

Frayling

, Timpson

, Weedon

et al. A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity. Science, 2007; 316:889–894.

36.

Steemburgo

, de Azevedo

, Gross

et al. The rs7204609 polymorphism in the fat mass and obesity-associated gene is positively associated with central obesity and microalbuminuria in patients with type 2 diabetes from Southern, Brazil. J Ren Nutr, 2012; 22:228–236.

37.

Ranjith

, Pegoraro

, Shanmugam

. Obesity-associated genetic variants in young Asian Indians with the metabolic syndrome and myocardial infarction. Cardiovasc J Afr, 2011; 22:25–30.

38.

Zimmermann

, Skogstrand

, Hougaard

et al. Influences of the common FTO rs9939609 variant on inflammatory markers throughout a broad range of body mass index. PLoS One, 2011; 6:e15958.

39.

Lappalainen

, Tolppanen

, Kolehmainen

et al. Finnish Diabetes Prevention Study Group. The common variant in the FTO gene did not modify the effect of lifestyle changes on body weight: the Finnish Diabetes Prevention Study. Obesity, 2009; 17:832–836.

40.

Delahanty

, Pan

, Jablonski

et al. Genetic predictors of weight loss and weight regain after intensive lifestyle modification, metformin treatment, or standard care in the Diabetes Prevention Program. Diabetes Care, 2012; 35:363–366.

41.

Hotta

, Kitamoto

et al. Association of variations in the FTO, SCG3 and MTMR9 genes with metabolic syndrome in a Japanese population. J Hum Genet, 2011; 56:647–651.

42.

Lappalainen

, Kolehmainen

, Schwab

et al. Association of the FTO gene variant (rs9939609) with cardiovascular disease in men with abnormal glucose metabolism—the Finnish Diabetes Prevention Study. Nutr Metab Cardiovasc Dis, 2011; 21:691–698.

43.

Mattevi

, Zembrzuski

, Hutz

. Effects of a PPARG gene variant on obesity characteristics in Brazil. Braz J Med Biol Res, 2007; 40:927–932.

44.

Morini

, Tassi

, Capponi

et al. Interaction between PPARgamma2 variants and gender on the modulation of body weight. Obesity, 2008; 16:1467–1470.

45.

Mirzaei

, Akrami

, Golmohammadi

et al. Polymorphism of Pro12Ala in the peroxisome proliferator-activated receptor gamma2 gene in Iranian diabetic and obese subjects. Metab Syndr Relat Disord, 2009; 7:453–458.

46.

Frederiksen

, Brodbaek

, Fenger

et al. Studies of the Pro12Ala polymorphism of the PPAR gamma gene in the Danish MONICA Cohort: Homozygosity of the Ala allele confers a decreased risk of the insulin resistance syndrome. J Clin Endocrinol Metab, 2002; 87:3989–3992.

47.

Franks

, Jablonski

, Delahanty

et al. Pro12Ala variant at the peroxisome proliferator-activated receptor gamma gene and change in obesity-related traits in the Diabetes Prevention Program. Diabetologia, 2007; 50:2451–2460.

48.

Milewicz

, Tworowska-Bardziñska

, Dunajska

. Relationship of PPAR gamma2 polymorphism with obesity and metabolic syndrome in postmenopausal Polish women. Exp Clin Endocrinol, 2009; 117:628–632.

49.

Marchesi

, Schiffrin

. Peroxisome proliferator-activated receptors and the vascular system: Beyond their metabolic effects. J Am Soc Hypertens, 2008; 2:227–238.

50.

Passaro

, Dalla Nora

, Marcello

et al. PPARγ Pro12Ala and ACE ID polymorphisms are associated with BMI and fat distribution, but not metabolic syndrome. Cardiovasc Diabetol, 2011; 10:112.

51.

Kim

, Lee

, Valentine

. Association of Pro12Ala polymorphism in the peroxisome proliferative-activated receptor gamma2 gene with obesity and hypertension in Korean women. J Nutr Sci Vitaminol, 2007; 53:239–246.

52.

Yang

, Hua

, Liu

et al. Association between two common polymorphisms of PPARgamma gene and metabolic syndrome families in a Chinese population. Arch Med Res, 2009; 40:89–96.

53.

Pagani

, Giudici

, Baralle

et al. Association of a polymorphism in the Apo AI gene promoter with hyperalphalipoproteinemia. Eur J Epidemiol, 1992; S8:54–58.