Abstract
Background
Circulating branched-chain amino acids (BCAAs) and aromatic amino acids (AAAs) were recently related to insulin resistance and diabetes in prospective cohorts. We tested the effects of a genetic determinant of BCAA/AAA ratio on changes in body weight and insulin resistance in a 2-year diet intervention trial.
Methods and Results
We genotyped BCAA/AAA ratio associated variant rs1440581 near PPM1K gene in 734 overweight or obese adults who were randomly assigned to one of four diets varying in macronutrient content. At 6 months, we observed that dietary fat significantly modified genetic effects on changes in weight, fasting insulin and homeostasis model assessment of insulin resistance (HOMA-IR), after adjustment for confounders (all P for interaction ≤ 0.006). Further adjustment for weight change did not appreciably change the interactions for fasting insulin and HOMA-IR. In the high-fat diet group, the C allele was related to less weight loss and smaller decreases in serum insulin and HOMA-IR (all P ≤ 0.02 in an additive pattern); while an opposite genotype effect on changes in insulin and HOMA-IR was observed in low-fat diet group (P = 0.02 and 0.04, respectively). At 2 years, the gene–diet interactions remained significant for weight loss (P = 0.008); but became null for changes in serum insulin and HOMA-IR due to weight regain.
Conclusions
Individuals carrying C allele of BCAA/AAA ratio associated variant rs1440581 may benefit less in weight loss and improvement of insulin sensitivity than those without this allele when undertake an energy restricted high-fat diet.
Keywords: Branched-chain amino acids, gene-diet interaction, insulin resistance, weight loss
Introduction
Emerging evidence has shown that circulating amino acids may play an important role in the pathogenesis of metabolic disorders such as insulin resistance and type 2 diabetes (T2D)1-3. Recently, using metabolomic profiling methods, Wang et al. identified that high levels of circulating branched-chain amino acids (BCAAs) and aromatic amino acids (AAAs) predicted T2D in two prospective cohorts. These circulating amino acids were elevated up to 12 years before the onset of diabetes and associated higher insulin resistance4.
In line with these findings, several previous studies have shown associations of BCAAs or AAAs with obesity, insulin sensitivity5-7 and metabolic syndrome2. In addition, a recent study reported that circulating BCAAs levels significantly decreased after weight loss induced by gastric bypass surgery, but not by dietary intervention8,9; while in another 6 months behavioral/dietary intervention trial10, baseline plasma BCAAs were strong predictors of improvement in insulin sensitivity with weight loss11.
Blood levels of amino acids are partially determined by genetic factors. A recent genome-wide association study (GWAS) found a single-nucleotide polymorphism (SNP) rs1440581 near PPM1K gene (PP2C domain-containing protein phosphatase 1K) to be associated with higher serum valine levels; and the ratio of BCAA to AAA (Fischer’s ratio)12, which is characteristic of liver fibrosis and may contribute to hepatic encephalopathy13. Interestingly, PPM1K was recently identified as a susceptibility gene for T2D by systems genetics approch14. According to the Mendelian randomization theory15,16, a genetic variant could be a better marker than biomarkers in causal inference, because is less likely to be affected by confounding and reverse causation17. In the present study, we examined the effects of a circulating BCAA to AAA ratio (Fischer’s ratio) associated genetic variant on changes in weight and insulin resistance in the 2-year Preventing Overweight Using Novel Dietary Strategies (POUNDS LOST) trial, and particularly assessed gene-diet interactions.
Methods
POUNDS LOST trial
The POUNDS LOST trial was designed to compare the effects of energy-reduced diets with different compositions of macronutrients on body weight during 2 years of follow-up. The study design has been described in detail elsewhere18. Briefly, a total of 811 overweight and obese subjects (25 ≤ body mass index (BMI) ≤ 40 kg/m2) aged 30–70 years with no diabetes or unstable cardiovascular disease, no use of medications affecting body weight, and insufficient motivation, were randomly assigned to four diets; the target percentages of energy derived from fat, protein, and carbohydrate in the four diets were 20%, 15%, and 65%; 20%, 25%, and 55%; 40%, 15%, and 45%; and 40%, 25%, and 35%. Thus, two diets were low-fat (20%) and two diets were high-fat (40%), or two diets were average-protein (15%) and 2 diets were high-protein (25%), which constituted a 2-by-2 factorial design. After 2 years, 80% of the participant (n = 645) completed the trial. The study was approved by the human subjects committee at the Harvard School of Public Health and Brigham and Women’s Hospital and the Pennington Biomedical Research Center of the Louisiana State University System and by a data and safety monitoring board appointed by the National Heart, Lung, and Blood Institute. All participants gave written informed consent.
Study population
A total of 734 participants with genotype data available at baseline (91% of the participants in the POUNDS LOST trial; number of missing DNA sample = 66, unsuccessfully genotyped = 11) were included in the current study. The average age of the participants were 51 years, BMI were 33.6 kg/m2; 61% were women, 80% were white, 15% were black, 3% were Hispanic, and 2% were Asian or other ethnic groups by self report. There was no significant difference in the basic characteristics between the participants with and without genotype data. Among the 734 subjects at baseline, 658 and 598 subjects provided the information of body weight at 6 months and 2 years of intervention, respectively (the drop-out rate is 10% and 19%, respectively). For the biomarkers, 603 and 522 subjects had the serum insulin and glucose measurements at 6 months and 2 years of intervention, respectively. The drop-out rates are slightly greater than those for body weight, mainly due to missing blood samples or failure of measurement of the biomarkers.
Assessment of outcomes and covariates
Body height was measured at baseline. Body weight and waist circumference were measured in the morning before breakfast on 2 days at baseline, 6 months, and 2 years, as well as on a single day at 12 and 18 months. Blood pressure was measured on 2 days, at baseline, 6 months, and 2 years, by using an automated device (HEM-907XL; Omron). Dietary intake was assessed in a random sample of 50% of the participants by a review of 5-day diet record at baseline and by 24-hour recall during a telephone interview on 3 nonconsecutive days at 6 months and 2 years. Fasting blood samples, 24-h urine samples, and measurement of resting metabolic rate were obtained on 1 day. Concentrations of serum glucose, insulin, and urinary nitrogen were measured at the clinical laboratory at the Pennington Biomedical Research Center. BMI was calculated as weight in kilograms divided by height in squared meters. Insulin resistance was indicated as homeostasis model assessment of insulin resistance (HOMA-IR) calculated by the following equation: Fasting insulin (μU/ml) × fasting glucose (mg/dl) × 18.0−1 × 22.5−1.
Genotyping
DNA was extracted from the buffy coat fraction of centrifuged blood by using the QIAmp Blood Kit (Qiagen, Chatsworth, CA). We selected a novel SNP rs1440581 near PPM1K gene that was established to be associated with the serum valine levels and the ratio of circulating BCAA to AAA (Fischer’s ratio) as well in a recent GWAS12. The SNP was genotyped by using the OpenArray SNP Genotyping System (BioTrove, Woburn, MA) and the genotyping success rate was 99%. Replicated quality control samples (10%) were included in every genotyping plate with greater than 99% concordance19.
Statistical analysis
SAS version 9.1 (SAS Institute, Inc., Cary, NC, USA) was used to perform the data analysis. Variables with skewed distribution (serum insulin and HOMA-IR) were log-transformed before analysis. The Hardy-Weinberg equilibrium of rs1440581 genotypes and comparison of categorical variables at baseline were examined by χ2 test. Differences in continuous variables at baseline were tested using ANCOVA, with adjustment for age, sex and ethnicity. Multivariate general linear models were used to test the main effects and the potential interactions of genetic variation and diet intervention (high vs. low fat, or high vs. average protein diets) on changes in weight and insulin resistance at 6 months and 2 years. The covariates included age, sex, ethnicity, baseline BMI and values for the respective outcome were used in the analysis (Model 1). For the changes in serum insulin and HOMA-IR, weight change at 6 months or 2 years was further adjusted, respectively (Model 2). Multivariate linear models were used to examine the genetic effects according to different diet groups. Linear mixed models were used to test genetic associations with the trajectory of changes in weight or insulin resistance according to diet groups. Time was treated as a repeated measurement factor, with which genotype-time interaction terms were included in the mixed models. We used Quanto 1.2.4 (University of Southern California, Los Angeles, CA; http://hydra.usc.edu/gxe/) to estimate the detectable effect sizes of gene-diet interactions. The study had 80% power to detect gene–diet interaction effect sizes of 1.8 and 2.5 kg for weight loss, 0.14 and 0.16 log-transformed unit for changes in fasting insulin, and 0.16 and 0.18 log-transformed unit for changes in HOMA-IR at 6 months and 2 years, respectively. An additive genetic model was analyzed for rs1440581 variant. All the reported P values were two-sided and nominal, and a P value of 0.05 was considered statistically significant. We used Bonferroni adjustment to adjust for 4 multi tests [2 dependently measured outcome traits (body weight and insulin resistance) and 2 stratified analyses (high fat vs. low fat)] at each time point. Because HOMA-IR was calculated on the basis of insulin, and the endpoints at 6 month and 2 years were correlated, we did not treat them as independent tests20 . A P value of 0.0125 (0.05 / 4) was considered significant after adjustment of multi comparisons.
Results
Baseline characteristics of the study participants
The major allele frequency of PPM1K rs1440581 (C allele) was 55.4% in all the participants and 53.1% in the white individuals; and the genotype distribution fits Hardy-Weinberg equilibrium (both P ≥ 0.43). Table 1 shows the baseline characteristics of the participants according to PPM1K rs1440581 genotype. There was no significant difference in genotype frequency between males and females, and between diet groups varying in fat or protein (all P ≥ 0.18); while significant difference was observed across the ethnicity groups (P = 0.005). We did not observe significant difference in body weight, BMI, waist circumference, serum insulin levels or HOMA-IR across the genotypes of rs1440581 at baseline. Similarly, no genotypic difference was found for these variables in the white participants (Supplementary Table 1).
Table 1.
Baseline Characteristics of the Study Participants According to PPM1K rs1440581 Genotypes
| CC | CT | TT | P values † | |
|---|---|---|---|---|
|
|
||||
| (n = 231) | (n = 352) | (n = 151) | ||
| Age, years | 51 ± 9 | 51 ± 9 | 52 ± 9 | 0.74 |
| Female, n (%) | 144 (62.6) | 212 (60.6) | 91 (60.3) | 0.90 |
| Race or ethnic group, n (%) | ||||
| White | 170 (29.0) | 283 (48.2) | 134 (22.8) | 0.005 |
| Black | 48 (43.2) | 55 (49.6) | 8 (7.2) | |
| Hispanic | 9 (36.0) | 10 (40.0) | 6 (24.0) | |
| Asian or other | 4 (36.4) | 4 (36.4) | 3 (27.3) | |
| Height, cm | 168.6 (8.4) | 168.4 (9.1) | 169.3 (8.6) | 0.44 |
| Weight, kg | 93.7 (15.0) | 92.9 (16.0) | 93.2 (15.6) | 0.87 |
| Waist circumference, cm | 103.9 (12.5) | 103.1 (13.6) | 104.2 (12.7) | 0.95 |
| BMI, kg/m2 | 32.9 (3.9) | 32.6 (3.8) | 32.4 (3.9) | 0.46 |
| Blood pressure, mmHg | ||||
| Systolic | 119 (13) | 121 (14) | 119 (12) | 0.56 |
| Diastolic | 75 (9) | 76 (10) | 75 (9) | 0.35 |
| Glucose, mg/dl | 91 (11) | 92 (12) | 93 (12) | 0.29 |
| Insulin, μU/ml‡ | 10.80 (7.00 – 16.30) | 10.40 (6.90 – 14.80) | 10.55 (6.80 – 16.50) | 0.99 |
| HOMA-IR‡ | 2.40 (1.50 – 3.61) | 2.32 (1.51 – 3.56 ) | 2.40 (1.51 – 3.85) | 0.84 |
| Dietary intake per day | ||||
| Carbohydrate, % | 44.8 (7.3) | 44.7 (8.1) | 44.2 (7.2) | 0.94 |
| Fat, % | 37.2 (5.6) | 36.8 (6.1) | 37.2 (6.1) | 0.67 |
| Protein, % | 17.9 (3.4) | 18.3 (3.6) | 18.2 (3.6) | 0.46 |
| Energy, kcal | 2008 (569) | 1972 (579) | 1894 (504) | 0.05 |
| Urinary nitrogen, g | 12.3 (3.9) | 11.9 (4.2) | 12.7 (5.2) | 0.93 |
| Respiratory quotient | 0.84 (0.04) | 0.84 (0.04) | 0.85 (0.05) | 0.36 |
Data are means ± SD, median (IQR), or n (%).
P values were calculated by χ2 test for categorical variables, and multivariate analysis of covariance for continuous variables after adjusted for age, sex and ethnicity.
These variables were log-transformed before analysis. BMI, body mass index. HOMA-IR, homeostasis model assessment of insulin resistance.
Nutrients intake and biomarkers of adherence according to PPM1K rs1440581 genotype
Table 2 shows adherence markers of the participants. At 2 years, the percentage of dietary protein intake was slightly different among 3 genotypes (P = 0.04), which was consistent with a parallel difference for urinary nitrogen (P = 0.04). There were no significant differences in mean values of nutrient intakes and biomarkers of adherence at 6 months across the PPM1K rs1440581 genotype in each of the 2 fat-diet groups or protein-diet groups (P > 0.05).
Table 2.
Nutritient Intake and Biomarkers of Adherence According to PPM1K rs1440581 During the Intervention
| At 6 months |
At 2 years |
|||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CC | CT | TT | CC | CT | TT | |||||||
| Dietary intake per day* | n | n | n | n | n | n | ||||||
| Energy, kcal | 103 | 1639 ± 591 | 154 | 1611 ± 487 | 72 | 1618 ± 491 | 51 | 1585 ± 533 | 83 | 1466 ± 467 | 34 | 1576 ± 447 |
| Carbohydrate, % | 103 | 48.9 ± 9.7 | 154 | 51.7 ± 10.4 | 72 | 51.3 ± 11.0 | 51 | 48.5 ± 11.0 | 83 | 50.0 ± 10.6 | 34 | 48.0 ± 9.2 |
| Fat, % | 103 | 31.2 ± 7.8 | 154 | 29.0 ± 8.4 | 72 | 30.4 ± 8.5 | 51 | 33.1 ± 9.2 | 83 | 29.0 ± 7.9 | 34 | 31.1 ± 8.3 |
| Protein, % | 103 | 20.2 ± 4.4 | 154 | 20.1 ± 4.7 | 72 | 19.7 ± 4.3 | 51 | 18.7 ± 3.6 | 83 | 21.0 ± 5.1 | 34 | 20.7 ± 4.4 |
| Biomarkers of adherence | ||||||||||||
| Respiratory quotient† | 183 | 0.84 ± 0.04 | 279 | 0.84 ± 0.04 | 130 | 0.85 ± 0.04 | 139 | 0.83 ± 0.04 | 215 | 0.83 ± 0.04 | 107 | 0.83 ± 0.04 |
| Urinary nitrogen, g‡ | 167 | 11.7 ± 4.3 | 247 | 11.5 ± 4.9 | 121 | 11.7 ± 4.2 | 116 | 11.5 ± 3.8 | 162 | 12.0 ± 4.2 | 92 | 12.7 ± 5.0 |
Data are expressed as means ± SD.
Data were included for 329 participants at 6 months; and 168 at 2 years, respectively.
Data were included for 592 participants at 6 months; and 461 at 2 y, respectively.
Data were included for 535 participants at 6 months; and 370 at 2 y, respectively.
Interaction between diet intervention and PPM1K rs1440581 on weight loss and insulin resistance
After adjustment for age, sex, ethnicity, diet intervention, baseline BMI and measurements of each outcome, PPM1K variant rs1440581 was not associated with changes in body weight or serum insulin or HOMA-IR at 6 months and 2 years (all P ≥ 0.09). Then we tested the interactions between the PPM1K rs1440581 and diet intervention (two-factorial comparisons: high- or low-fat diet, and high-or average protein) on changes in weight, fasting insulin, and HOMA-IR at 6 months and 2 years.
We observed that dietary fat (high vs low) significantly modified the effects of PPM1K rs1440581 on changes in body weight (P for interaction = 0.002), fasting insulin and HOMA-IR (both P for interaction = 0.006) at 6 months, after adjustment for age, sex, ethnicity and the baseline values for the respective outcome traits (Table 3, Supplementary Table 2). Further adjustment for weight change did not appreciably change the interactions on changes in fasting insulin and HOMA-IR. In the high-fat diet group, the C allele was related to less weight loss (P = 0.001 in an additive pattern), whereas no significant genetic effect was observed in the low-fat diet group (P = 0.53). Participants carrying the C allele had smaller decreases in serum insulin and HOMA-IR values than those without this allele in high-fat diet group (both P = 0.02), while an opposite effect was observed in participants assigned to the low-fat diet group (P = 0.02 and 0.04, respectively). When further adjusted for weight loss at 6 months, the gene-diet interactions on changes in serum insulin and HOMA-IR remained significant (both P for interaction ≤ 0.03). However, the genetic effects on changes in serum insulin and HOMA-IR in both low- and high-fat diet groups were significantly attenuated (Table 3); only in the low-fat diet group, the C allele carriers showed significant greater decrease of serum insulin (P = 0.03).
Table 3.
The Interactions of Dietary Fat and Genetic Variant PPM1K rs1440581 on Changes in Body Weight and Insulin Resistance
| At 6 months |
P for interaction |
At 2 years |
P for interaction |
|||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Low-fat Diet | High-fat Diet | Low-fat Diet | High-fat Diet | |||||||
| Weight loss, kg | n | n | n | n | ||||||
| CC | 98 | −7.2 ± 5.6 | 112 | −5.1 ± 4.9 | 85 | −4.4 ± 6.9 | 101 | −2.8 ± 6.6 | ||
| CT | 161 | −6.7 ± 5.9 | 148 | −6.7 ± 5.8 | 153 | −4.6 ± 8.3 | 131 | −3.6 ± 7.3 | ||
| TT | 71 | −6.7 ± 5.4 | 68 | −8.2 ± 6.2 | 66 | −3.1 ± 6.7 | 62 | −6.6 ± 7.7 | ||
| P values | ||||||||||
| Model 1 | 0.53 | 0.001 | 0.002 | 0.24 | 0.005 | 0.008 | ||||
| Change in log-insulin, μU/ml | ||||||||||
| CC | 89 | −0.33 ± 0.43 | 101 | −0.16 ± 0.43 | 76 | −0.20 ± 0.42 | 90 | −0.11 ± 0.45 | ||
| CT | 146 | −0.27 ± 0.45 | 136 | −0.30 ± 0.48 | 130 | −0.16 ± 0.45 | 107 | −0.12 ± 0.46 | ||
| TT | 69 | −0.14 ± 0.48 | 62 | −0.34 ± 0.44 | 60 | −0.05 ± 0.46 | 59 | −0.23 ± 0.50 | ||
| P values | ||||||||||
| Model 1 | 0.02 | 0.02 | 0.006 | 0.02 | 0.22 | 0.16 | ||||
| Model 2 | 0.03 | 0.38 | 0.01 | 0.049 | 0.59 | 0.23 | ||||
| Change in log-HOMA-IR | ||||||||||
| CC | 89 | −0.36 ± 0.48 | 101 | −0.17 ± 0.47 | 76 | −0.19 ± 0.45 | 90 | −0.07 ± 0.51 | ||
| CT | 146 | −0.29 ± 0.49 | 136 | −0.32 ± 0.53 | 130 | −0.15 ± 0.49 | 107 | −0.08 ± 0.52 | ||
| TT | 69 | −0.18 ± 0.54 | 62 | −0.38 ± 0.48 | 60 | −0.02 ± 0.48 | 59 | −0.21 ± 0.55 | ||
| P values | ||||||||||
| Model 1 | 0.04 | 0.02 | 0.006 | 0.01 | 0.20 | 0.13 | ||||
| Model 2 | 0.07 | 0.39 | 0.03 | 0.04 | 0.66 | 0.19 | ||||
Data are means ± SD. P values are after adjusted for age, sex, ethnicity and the baseline BMI and values for the respective outcome trait (Model 1). Model 2, based on Model 1 further adjusted weight loss at 6 months and 2 years, respectively. HOMA_IR, homeostasis model assessment of insulin resistance; PPM1K, PP2C domain-containing protein phosphatase 1K gene.
At 2 years, for weight loss, the gene-diet interaction remained significant (P for interaction=0.008). The C allele carriers had less weight loss than the non-carriers (P = 0.005) in the high-fat diet group. No significant genetic effect was observed on weight loss in the low-fat diet group. No significant gene-diet interaction was found on changes in serum insulin and HOMA-IR (both P ≥ 0.13) (Table 3).
In the white participants, both at 6 months and 2 years, we observed significant interactions between PPM1K rs1440581 genetic variant and dietary fat on changes in body weight, insulin, and HOMA-IR under the additive genetic models (all P ≤ 0.02). However, the interactions were attenuated after adjustment for weight loss (Supplementary Table 3). We did not find significant interactions between PPM1K rs1440581 variant and dietary protein on weight loss or changes in insulin resistance in the total sample or in whites (all P for interaction > 0.05).
Trajectory of changes in body weight, fasting insulin and HOMA-IR
We then performed linear mixed models to evaluate the genetic effect by intervention time in the two fat-diet intervention groups (Figure 1). We found significant interactions between PPMIK rs1440581 genotypes and intervention time on changes in weight (P for interaction = 0.01 among participants in the high-fat diet group). Participants with the CC genotype had consistently less long-term weight loss than those who carry the other two genotypes across the 2-year intervention (Figure 1, Panel B). No significant gene–time interactions were observed on changes in serum insulin or HOMA-IR levels (P for interaction = 0.39 and 0.34, respectively). The gene–time interactions were not observed in low-fat diet group (all P for interaction ≥ 0.10).
Figure 1.
Genetic effects of rs1440581 on trajectory of changes in body weight in participants assigned to the two fat-diets over 2 years. A. the low-fat diet group. B. the high-fat diet group. Data are adjusted means ± standard errors after adjusted for age, sex, ethnicity and baseline body weight. The numbers of the participants were shown in Table 3.
Discussion
In a 2-year prospective dietary intervention trial, we found significant interactions between dietary fat and a genetic variant rs1440581 near PPM1K gene associated with amino acid metabolites on weight loss and changes in insulin resistance. In the participants assigned to the high-fat diet group, the C allele carriers showed less weight loss and a smaller decrease of insulin resistance than the non-carriers; whereas an opposite effect on changes in insulin resistance was observed in the low-fat diet group. The gene-diet interaction on weight loss was persistent through 2 years of intervention. We did not observe significant interaction between dietary protein and rs1440581 genotype.
To the best of our knowledge, the present study is the first to examine the effects of a genetic determinant for BCAA to AAA ratio, and its potential interactions with dietary fat on weight loss and improvement of insulin sensitivity in a randomized dietary intervention trial. Our data are in line with previous studies showing a relation between BCAAs and AAAs with body weight, insulin resistance, and the risk of diabetes. In a recent study, Wang et al. reported that circulating BCAAs and AAAs significantly predicted future development of T2D and insulin resistance, suggesting a potential role of amino acids in the pathogenesis of diabetes4. Plasma BCAAs have also been associated with weight loss and improvement of insulin sensitivity in response to either the bypass gastric surgery8 or behavioral/dietary (DASH) intervention11. Our data suggest that these amino acids might modulate change in body weight and insulin sensitivity in people with different levels of fat intake.
In the present study, we found that carriers of the C allele showed less weight loss and a smaller decrease in serum insulin levels and insulin resistance index in participants with high-fat diet. The opposite changes were observed in those eating the low-fat diet. These findings were expected. Firstly, it has been proposed that BCAAs synergize with hyperlipidemia to contribute to the development of insulin resistance3, 21. Animal studies have shown that high-fat diet strengthened the effects of BCAAs supplementation on insulin resistance3. These studies support interplay between fat intake and BCAAs in progression of insulin resistance. Second, in the study by Kettunen et al12, the C allele of PPM1K rs1440581 associated with increased Fisher’s ratio and higher serum valine levels. Our study demonstrated that the C allele, which is associated with higher circulating BCAAs or AAAs levels, was related to less weight loss and improvement of insulin sensitivity in the context of high-fat diet; these observations are consistent with those found in the animal studies.
The PPM1K gene is located at 4q22 and encoded mitochondrial protein phosphatase 1K, which is a kind of branched chain α-ketoacid dehydrogenase phosphatase and plays a vital role on the metabolism of BCAAs22. It catalyzes oxidative decarboxylation of branched-chain α-ketoacid from leusine, isoleucine and valine. In a very recent study, PPM1K gene was identified by systems genetics approach as a T2D gene14. Functional study showed that silencing of PPM1K gene resulted in reduced glucose-stimulated insulin secretion. Further studies are warranted to investigate how dietary fat influences PPM1K gene function and circulating BCAAs levels.
Our data indicated that the gene-diet interaction on changes in insulin resistance at 6 months was significant even after adjusting for weight loss. However, at 2 years of intervention, the modification of diet-fat on the genetic effect of the C allele on weight loss in those eating the high fat diet was sustained, whereas the gene-diet interaction was no longer found for changes in insulin or HOMA-IR. The attenuation of the genetic effects at 2 years was partly attributable to weight regain due to the less adherence to the weight-loss diet after 6 months. Our results suggest that the biologic mechanisms underlying the relation of BCAA/AAA to weight loss and insulin resistance might be different.
Some limitations of the study should be addressed. First, we did not measure the circulating BCAAs levels in the study participants. This prevented the potential analysis of the relationship of genetic variants and circulating BCAAs levels, and the roles of circulating BCAAs in gene-diet interaction. However, a genetic marker could be a surrogate for the biomarker in yielding causal relationship according to Mendelian Randomization theory15,16. Even though, we acknowledge that measuring the circulating levels of BCAAs and AAAs would provide additional evidence and strengthen the conclusion; and will pursue such analyses in future. Second, HOMA-IR was used as an indicator of insulin resistance instead of euglycemic hyperinsulinemic clamp. However, the markers included in our study have been tightly related to the clamp and widely used in clinical practice23, 24. Third, the results may not be generalized to other ethnic groups since 80% of the participants were whites in the current study.
In conclusion, we found a significant interaction between the amino acid metabolites related genotype and dietary fat on weight loss and changes in insulin resistance in a large, 2-year intervention trial. Individuals carrying C allele of BCAA/AAA ratio associated variant rs1440581 may benefit less in weight loss and improvement of insulin sensitivity than those without this allele when undertake an energy restricted high-fat diet.
Supplementary Material
Acknowledgments
We appreciate the participants in the clinical trial for their participation and contribution to the research.
Sources of Funding This study was supported by grants from the National Heart, Lung, and Blood Institute (HL071981), the National Institute of Diabetes and Digestive and Kidney Diseases (DK091718), the General Clinical Research Center (RR-02635), the Boston Obesity Nutrition Research Center (DK46200), and United States – Israel Binational Science Foundation Grant2011036. Dr. Lu Qi was a recipient of the American Heart Association Scientist Development Award (0730094N).
Footnotes
Disclosure None.
Clinical Trial Registration Information — http:www.clinicaltrials.gov. Unique identifier: NCT00072995.
Reference
- 1.Felig P, Marliss E, Cahill GF., Jr Plasma amino acid levels and insulin secretion in obesity. N Engl J Med. 1969;281:811–816. doi: 10.1056/NEJM196910092811503. [DOI] [PubMed] [Google Scholar]
- 2.Huffman KM, Shah SH, Stevens RD, Bain JR, Muehlbauer M, Slentz CA, Tanner CJ, Kuchibhatla M, Houmard JA, Newgard CB, Kraus WE. Relationships between circulating metabolic intermediates and insulin action in overweight to obese, inactive men and women. Diabetes Care. 2009;32:1678–1683. doi: 10.2337/dc08-2075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Newgard CB, An J, Bain JR, Muehlbauer MJ, Stevens RD, Lien LF, Haqq AM, Shah SH, Arlotto M, Slentz CA, Rochon J, Gallup D, Ilkayeva O, Wenner BR, Yancy WS, Jr., Eisenson H, Musante G, Surwit RS, Millington DS, Butler MD, Svetkey LP. A branched-chain amino acid-related metabolic signature that differentiates obese and lean humans and contributes to insulin resistance. Cell Metab. 2009;9:311–326. doi: 10.1016/j.cmet.2009.02.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Wang TJ, Larson MG, Vasan RS, Cheng S, Rhee EP, McCabe E, Lewis GD, Fox CS, Jacques PF, Fernandez C, O’Donnell CJ, Carr SA, Mootha VK, Florez JC, Souza A, Melander O, Clish CB, Gerszten RE. Metabolite profiles and the risk of developing diabetes. Nat Med. 2011;17:448–453. doi: 10.1038/nm.2307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Zhao X, Peter A, Fritsche J, Elcnerova M, Fritsche A, Häring HU, Schleicher ED, Xu G, Lehmann R. Changes of the plasma metabolome during an oral glucose tolerance test: is there more than glucose to look at? Am J Physiol Endocrinol Metab. 2009;296:E384–393. doi: 10.1152/ajpendo.90748.2008. [DOI] [PubMed] [Google Scholar]
- 6.Shaham O, Wei R, Wang TJ, Ricciardi C, Lewis GD, Vasan RS, Carr SA, Thadhani R, Gerszten RE, Mootha VK. Metabolic profiling of the human response to a glucose challenge reveals distinct axes of insulin sensitivity. Mol Syst Biol. 2008;4:214. doi: 10.1038/msb.2008.50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Tai ES, Tan ML, Stevens RD, Low YL, Muehlbauer MJ, Goh DL, Ilkayeva OR, Wenner BR, Bain JR, Lee JJ, Lim SC, Khoo CM, Shah SH, Newgard CB. Insulin resistance is associated with a metabolic profile of altered protein metabolism in Chinese and Asian-Indian men. Diabetologia. 2010;53:757–767. doi: 10.1007/s00125-009-1637-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Laferrere B, Reilly D, Arias S, Swerdlow N, Gorroochurn P, Bawa B, Bose M, Teixeira J, Stevens RD, Wenner BR, Bain JR, Muehlbauer MJ, Haqq A, Lien L, Shah SH, Svetkey LP, Newgard CB. Differential metabolic impact of gastric bypass surgery versus dietary intervention in obese diabetic subjects despite identical weight loss. Sci Transl Med. 2011;3:80re82. doi: 10.1126/scitranslmed.3002043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Robert E, Gerszten TJW. Two Roads Diverge: Weight Loss Interventions and Circulating Amino Acids. Sci Transl Med. 2011;3:80ps15. doi: 10.1126/scitranslmed.3002377. [DOI] [PubMed] [Google Scholar]
- 10.Svetkey LP, Stevens VJ, Brantley PJ, Appel LJ, Hollis JF, Loria CM, Vollmer WM, Gullion CM, Funk K, Smith P, Samuel-Hodge C, Myers V, Lien LF, Laferriere D, Kennedy B, Jerome GJ, Heinith F, Harsha DW, Evans P, Erlinger TP, Dalcin AT, Coughlin J, Charleston J, Champagne CM, Bauck A, Ard JD, Aicher K. Weight Loss Maintenance Collaborative Research Group. Comparison of strategies for sustaining weight loss: the weight loss maintenance randomized controlled trial. JAMA. 2008;299:1139–1148. doi: 10.1001/jama.299.10.1139. [DOI] [PubMed] [Google Scholar]
- 11.Shah SH, Crosslin DR, Haynes CS, Nelson S, Turer CB, Stevens RD, Muehlbauer MJ, Wenner BR, Bain JR, Laferrère B, Gorroochurn P, Teixeira J, Brantley PJ, Stevens VJ, Hollis JF, Appel LJ, Lien LF, Batch B, Newgard CB, Svetkey LP. Branched-chain amino acid levels are associated with improvement in insulin resistance with weight loss. Diabetologia. 2012;55:321–330. doi: 10.1007/s00125-011-2356-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kettunen J, Tukiainen T, Sarin AP, Ortega-Alonso A, Tikkanen E, Lyytikäinen LP, Kangas AJ, Soininen P, Würtz P, Silander K, Dick DM, Rose RJ, Savolainen MJ, Viikari J, Kähönen M, Lehtimäki T, Pietiläinen KH, Inouye M, McCarthy MI, Jula A, Eriksson J, Raitakari OT, Salomaa V, Kaprio J, Järvelin MR, Peltonen L, Perola M, Freimer NB, Ala-Korpela M, Palotie A, Ripatti S. Genome-wide association study identifies multiple loci influencing human serum metabolite levels. Nat Genet. 2012;44:269–276. doi: 10.1038/ng.1073. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Soeters P, Fischer JE. Insulin, glucagon, aminoacid imbalance, and hepatic encephalopathy. Lancet. 1976;2:880–882. doi: 10.1016/s0140-6736(76)90541-9. [DOI] [PubMed] [Google Scholar]
- 14.Taneera J, Lang S, Sharma A, Fadista J, Zhou Y, Ahlqvist E, Jonsson A, Lyssenko V, Vikman P, Hansson O, Parikh H, Korsgren O, Soni A, Krus U, Zhang E, Jing XJ, Esguerra JL, Wollheim CB, Salehi A, Rosengren A, Renström E, Groop L. A systems genetics approach identifies genes and pathways for type 2 diabetes in human islets. Cell Metab. 2012;16:122–134. doi: 10.1016/j.cmet.2012.06.006. [DOI] [PubMed] [Google Scholar]
- 15.Voight B, Peloso GM, Orho-Melander M, Frikke-Schmidt R, Barbalic M, Jensen MK, Hindy G, Hólm H, Ding EL, Johnson T, Schunkert H, Samani NJ, Clarke R, Hopewell JC, Thompson JF, Li M, Thorleifsson G, Newton-Cheh C, Musunuru K, Pirruccello JP, Saleheen D, Chen L, Stewart A, Schillert A, Thorsteinsdottir U, Thorgeirsson G, Anand S, Engert JC, Morgan T, Spertus J, Stoll M, Berger K, Martinelli N, Girelli D, McKeown PP, Patterson CC, Epstein SE, Devaney J, Burnett MS, Mooser V, Ripatti S, Surakka I, Nieminen MS, Sinisalo J, Lokki ML, Perola M, Havulinna A, de Faire U, Gigante B, Ingelsson E, Zeller T, Wild P, de Bakker PI, Klungel OH, Maitland-van der Zee AH, Peters BJ, de Boer A, Grobbee DE, Kamphuisen PW, Deneer VH, Elbers CC, Onland-Moret NC, Hofker MH, Wijmenga C, Verschuren WM, Boer JM, van der Schouw YT, Rasheed A, Frossard P, Demissie S, Willer C, Do R, Ordovas JM, Abecasis GR, Boehnke M, Mohlke KL, Daly MJ, Guiducci C, Burtt NP, Surti A, Gonzalez E, Purcell S, Gabriel S, Marrugat J, Peden J, Erdmann J, Diemert P, Willenborg C, König IR, Fischer M, Hengstenberg C, Ziegler A, Buysschaert I, Lambrechts D, Van de Werf F, Fox KA, El Mokhtari NE, Rubin D, Schrezenmeir J, Schreiber S, Schäfer A, Danesh J, Blankenberg S, Roberts R, McPherson R, Watkins H, Hall AS, Overvad K, Rimm E, Boerwinkle E, Tybjaerg-Hansen A, Cupples LA, Reilly MP, Melander O, Mannucci PM, Ardissino D, Siscovick D, Elosua R, Stefansson K, O’Donnell CJ, Salomaa V, Rader DJ, Peltonen L, Schwartz SM, Altshuler D, Kathiresan S. Plasma HDL cholesterol and risk of myocardial infarction: a mendelian randomisation study. Lancet. 2012;380:572–580. doi: 10.1016/S0140-6736(12)60312-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Interleukin-6 Receptor Mendelian Randomisation Analysis (IL6R MR) Consortium HA, Casas JP. The interleukin-6 receptor as a target for prevention of coronary heart disease: a mendelian randomisation analysis. Lancet. 2012;379:1214–1224. doi: 10.1016/S0140-6736(12)60110-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Qi L. Mendelian randomization in nutritional epidemiology. Nutr Rev. 2009;67:439–450. doi: 10.1111/j.1753-4887.2009.00218.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sacks FM, Bray GA, Carey VJ, Smith SR, Ryan DH, Anton SD, McManus K, Champagne CM, Bishop LM, Laranjo N, Leboff MS, Rood JC, de Jonge L, Greenway FL, Loria CM, Obarzanek E, Williamson DA. Comparison of weight-loss diets with different compositions of fat, protein, and carbohydrates. N Engl J Med. 2009;360:859–873. doi: 10.1056/NEJMoa0804748. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Qi Q, Bray GA, Smith SR, Hu FB, Sacks FM, Qi L. Insulin receptor substrate 1 gene variation modifies insulin resistance response to weight-loss diets in a 2-year randomized trial: the Preventing Overweight Using Novel Dietary Strategies (POUNDS LOST) trial. Circulation. 2011;124:563–571. doi: 10.1161/CIRCULATIONAHA.111.025767. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Qi Q, Bray GA, Hu FB, Sacks FM, Qi L. Weight-loss diets modify glucose-dependent insulinotropic polypeptide receptor rs2287019 genotype effects on changes in body weight, fasting glucose, and insulin resistance: the Preventing Overweight Using Novel Dietary Strategies trial. Am J Clin Nutr. 2012;95:506–13. doi: 10.3945/ajcn.111.025270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Newgard CB. Interplay between lipids and branched-chain amino acids in development of insulin resistance. Cell Metab. 2012;15:606–614. doi: 10.1016/j.cmet.2012.01.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Wynn RM, Li J, Brautigam CA, Chuang JL, Chuang DT. Structural and biochemical characterization of human mitochondrial branched-chain α-ketoacid dehydrogenase phosphatase. J Biol Chem. 2012;287:9178–9192. doi: 10.1074/jbc.M111.314963. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Anderson RL, Hamman RF, Savage PJ, Saad MF, Laws A, Kades WW, Sands RE, Cefalu W. Exploration of simple insulin sensitivity measures derived from frequently sampled intravenous glucose tolerance (FSIGT) tests. The Insulin Resistance Atherosclerosis Study. Am J Epidemiol. 1995;142:724–732. doi: 10.1093/aje/142.7.724. [DOI] [PubMed] [Google Scholar]
- 24.Emoto M, Nishizawa Y, Maekawa K, Hiura Y, Kanda H, Kawagishi T, Shoji T, Okuno Y, Morii H. Homeostasis model assessment as a clinical index of insulin resistance in type 2 diabetic patients treated with sulfonylureas. Diabetes Care. 1999;22:818–822. doi: 10.2337/diacare.22.5.818. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.