Publications

Context and Objective Obesity and SLC2A9 genotype are strong determinants of uric acid levels. However, data on SLC2A9 variants and weight loss induced changes in uric acid levels are missing. We examined whether the changes in uric acid levels two- and ten-years after weight loss induced by bariatric surgery were associated with SLC2A9 single nucleotide polymorphisms (SNPs) in the Swedish Obese Subjects study. Methods SNPs (N = 14) identified by genome-wide association studies and exonic SNPs in the SLC2A9 gene locus were genotyped. Cross-sectional associations were tested before (N = 1806), two (N = 1664) and ten years (N = 1201) after bariatric surgery. Changes in uric acid were compared between baseline and Year 2 (N = 1660) and years 2 and 10 (N = 1172). A multiple testing corrected threshold of P = 0.007 was used for statistical significance. Results Overall, 11 of the 14 tested SLC2A9 SNPs were significantly associated with cross-sectional uric acid levels at all three time points, with rs13113918 showing the strongest association at each time point (R2 = 3.7−5.2%, 3.9×10−22≤p≤7.7×10−11). One SNP (rs737267) showed a significant association (R2 = 0.60%, P = 0.002) with change in uric acid levels from baseline to Year 2, as common allele homozygotes (C/C, N = 957) showed a larger decrease in uric acid (−61.4 µmol/L) compared to minor allele carriers (A/X: −51.7 µmol/L, N = 702). No SNPs were associated with changes in uric acid from years 2 to 10. Conclusions SNPs in the SLC2A9 locus contribute significantly to uric acid levels in obese individuals, and the associations persist even after considerable weight loss due to bariatric surgery. However, we found little evidence for an interaction between genotype and weight change on the response of uric acid to bariatric surgery over ten years. Thus, the fluctuations in uric acid levels among the surgery group appear to be driven by the weight losses and gains, independent of SLC2A9 genotypes.

Obesity, especially intraabdominally deposited fatness, is associated with reduced insulin sensitivity. However, it is not well established whether this association is confounded by genetic factors. We studied 23 monozygous twin pairs (14 female, 9 male), 33–59 yr old, who had, on the average, 18 kg intrapair difference in body weight. A 75-g oral glucose tolerance test with glucose and insulin measurements at 30-min intervals was performed, and fat distribution was determined with magnetic resonance imaging. The pairs were divided into two groups by the gender-specific median of the abdominal visceral fat area (AVF) in the obese co-twins. In the high-AVF pairs, the mean area under curve (AUC) for glucose (mmol × min/L) was 758 vs. 968 (P = 0.001), AUC for insulin (mU × min/L) was 4320 vs. 8741 (P = 0.001), and insulin sensitivity index (mg × L × L/mmol × mU× min) was 71.5 vs. 45.9 (P < 0.001) in the lean and obese co-twins, respectively. In the low AVF pairs, the mean AUC for glucose was 669 vs. 706 (not significant), AUC for insulin was 3323 vs. 4241 (not significant), and the sensitivity index was 85.2 vs. 73.7 (P = 0.04) in the lean and obese co-twins, respectively. In subjects who are genetically identical but who are discordant for body mass, only those who differ most in visceral fat level are characterized by major alterations in insulin sensitivity and glucose tolerance.

No abstract is provided for this article.

&lt;i&gt;Background/Aims:&lt;/i&gt; Genome-wide association studies have led to the identification of several susceptibility genes for type 2 diabetes mellitus (T2DM). The objective of this study was to test the hypothesis that the associations between single nucleotide polymorphisms (SNPs) in these genes and adiposity and glucose homeostasis-related phenotypes are influenced by dietary fat intake. &lt;i&gt;Methods:&lt;/i&gt; Thirty-three SNPs in 9 T2DM genes &lt;i&gt;(CDKAL1, CDKN2A/B, HHEX, HNF1B, IGF2BP2, KCNJ11, SLC30A8, TCF7L2&lt;/i&gt; and &lt;i&gt;WFS1)&lt;/i&gt; were tested in a maximum of 669 subjects from the Quebec Family Study. Subjects were measured for several adiposity indices and underwent a 75-gram oral glucose tolerance test. Total fat intake was estimated from a 3-day dietary record. &lt;i&gt;Results:&lt;/i&gt; We observed 13 significant (p ≤ 0.01) SNP-dietary fat interactions. Among them, &lt;i&gt;IGF2BP2&lt;/i&gt; rs4402960, alone or in interaction with dietary fat intake, influenced abdominal total fat (ATF: SNP effect, p = 0.006, interaction effect, p = 0.009) and abdominal visceral fat (AVF: SNP effect, p = 0.007, interaction effect, p = 0.01). Similarly, &lt;i&gt;TCF7L2&lt;/i&gt; rs12573128 alone or in interaction with dietary fat intake, influenced insulin sensitivity (SNP effect and interaction effect, p ≤ 0.008) and glucose tolerance (SNP effect p ≤ 0.009 and interaction effect, p ≤ 0.01). &lt;i&gt;Conclusion:&lt;/i&gt; These results suggest that gene-dietary fat interactions may influence glucose homeostasis-related phenotypes and play an important role in determining the increased risk of diabetes associated with the T2DM susceptibility genes.

Familial risk ratios for high blood pressure were estimated in a representative sample of the Canadian population. The sample consisted of 14,069 participants 7–69 years of age from 5,753 families participating in the 1981 Canada Fitness Survey. Resting systolic (SBP) and diastolic (DBP) blood pressures were adjusted for the effects of body mass index using regression procedures. Varying degrees of high blood pressure were defined as the 75th, 85th, and 95th percentiles of age- and sex-specific values. Age- and sex-standardized risk ratios (SRRs) were calculated comparing the prevalences in the general population to those in spouses and first-degree relatives of probands with high blood pressure. SRRs for the 95th percentile were, for SBP and DBP, respectively, 1.37 and 1.45 in spouses and 1.33 and 2.36 in first-degree relatives of probands. SRRs decrease with decreasing percentile cut-offs used to define high blood pressure (95th > 85th > 75th), and SRRs are generally higher in first-degree relatives than in spouses, particularly for DBP. The results indicate significant familial risk for high blood pressure in the Canadian population, and the pattern of SRRs suggests that genetic factors may be responsible for a portion of the risk. Am. J. Hum. Biol. 13:620–625, 2001. © 2001 Wiley-Liss, Inc.

The potential long-term impact of low glycaemia on body fat accumulation has not been verified. Therefore, we examined the effects of low glucose concentrations on long-term energy balance and weight gain in humans.Two sets of analyses were realized in order to verify this objective. First, Study 1 consisted of 259 participants between 20 and 65 years of age selected from Phase 2 of the Quebec Family Study (QFS). The association between glucose concentrations at the end of an oral glucose tolerance test (OGTT) and changes in body mass was analysed prospectively (mean follow-up of 6 years). In addition, Study 2 consisted of 44 obese participants (20 men and 24 women) randomly assigned to a 15-week weight loss program in either a drug therapy group (fenfluramine) or a placebo group coupled with energy intake restriction. The focus of this study was the relationship between glycaemic control at the end of the treatment and post-treatment weight regain.In Study 1, the glucose concentrations at 120 min of the OGTT were negatively correlated with weight gain over 6 years (r=-0.42, p<0.01). In Study 2, the weight loss program induced a mean reduction in body weight of 10 kg in the fenfluramine and placebo groups. In participants who returned for a follow-up visit (mean=81 weeks after the intervention), the glucose area below fasting values (GABF) at the end of the OGTT increased with weight loss (p<0.01) and was correlated with weight regain (r=0.74, p<0.01).Lower glucose concentrations at the end of an OGTT are correlated with weight gain over time. Large amounts of weight loss are associated with low glycaemia at the end of an OGTT. These low glucose concentrations are strong predictors of the amount of weight regained after weight loss.

For analysis of the main effects of SNPs, meta‐analysis of summary results from individual studies has been shown to provide comparable results as “mega‐analysis” that jointly analyzes the pooled participant data from the available studies. This fact revolutionized the genetic analysis of complex traits through large GWAS consortia. Investigations of gene‐environment (G×E) interactions are on the rise since they can potentially explain a part of the missing heritability and identify individuals at high risk for disease. However, for analysis of gene‐environment interactions, it is not known whether these methods yield comparable results. In this empirical study, we report that the results from both methods were largely consistent for all four tests; the standard 1 degree of freedom (df) test of main effect only, the 1 df test of the main effect (in the presence of interaction effect), the 1 df test of the interaction effect, and the joint 2 df test of main and interaction effects. They provided similar effect size and standard error estimates, leading to comparable P ‐values. The genomic inflation factors and the number of SNPs with various thresholds were also comparable between the two approaches. Mega‐analysis is not always feasible especially in very large and diverse consortia since pooling of raw data may be limited by the terms of the informed consent. Our study illustrates that meta‐analysis can be an effective approach also for identifying interactions. To our knowledge, this is the first report investigating meta‐versus mega‐analyses for interactions.

Fat-free mass (FFM) consists mostly of skeletal muscle and bone tissues, and identification of the genes and molecular mechanisms involved in the control of FFM would have implications for the understanding of sarcopenia and potentially osteoporesis associated with aging, as well as the response to starvation, refeeding, anorexia, and any other conditions in which lean body mass is important. A genome-wide search for genes related to body leanness has been completed in the Québec Family Study (QFS). Microsatellite markers (N = 292) from the autosomal chromosomes were typed. The mean spacing of the markers was 11.9 centimorgans (cM) (range, <0.1 to 41). FFM was calculated from percent body fat, derived from underwater weighing, and body weight and was adjusted by regression for age and sex effects before analysis. A maximum of 336 sib paris or 609 pairs of extended relatives were analyzed using single-point Haseman-Elston regression (SIBPAL and RELPAL) and multipoint variance component (SEGPATH) linkage analyses. Significant linkages were observed on chromosomes 15q25-q25 for5 a CA repeat within the insulin-like growth factor 1 receptor (IGF1R) gene (Lod score = 3.56) and at 18q12 with D18S877 (Lod score = 3.53) and D18S535 (Lod score = 3.58), 2 makers located 10 cM apart. A moderately significant linkage was also observed on chromosome 7p15.3 with the marker D7S1808 (Load score = 2.72). The most obvious candidate genes within the regions identified by these linkages include the IGF1R on 15q and neuropeptide Y (NPY) and growth hormone—releasing hormone (GHRH) receptor on 7p. On 18q, the melanocortin receptor 4 (MC4R) is not likely the candidate gene for the observed linkage. This study represents the first genome-wide search for genes that may be involved in the regulation of the lean component of body mass in humans.

The purpose of the study was to assess the role of adiposity in the enhanced insulin sensitivity observed in endurance athletes (EA). An oral glucose tolerance test (75 g glucose) was administered to nine EA and to 23 sedentary subjects (SS). Two different strategies were used to investigate the problem. First, body composition indicators and Vo2max were correlated with the delta insulin and delta glucose areas measured for 180 minutes following glucose ingestion. These correlation analyses were performed for the two groups combined (n = 32). No significant correlations were observed between either fat weight or percent body fat versus delta insulin, delta glucose or delta glucose/delta insulin areas. Moreover, no significant correlation was observed between the several subcutaneous fat indicators and delta insulin, delta glucose and delta glucose/delta insulin areas. The second strategy consisted of comparing EA to SS when percent body fat difference was eliminated. This was achieved by two different methods, first by covariance analysis and second by comparing subsamples of trained and non-trained subjects paired with respect to percent body fat. These two comparisons revealed that even when adiposity was equal between the groups, a significantly greater insulin sensitivity was observed in the EA group (p less than 0.01). The present results suggest that adiposity is not the determining factor for the increased insulin sensitivity of trained subjects.

Objective: Previous studies show a wide range in the percentage of variance in leptin levels attributable to genetic factors. These studies differ markedly with respect to ethnicity, study design, and statistical methodology. Therefore, the purpose of this study was to investigate heterogeneity hypotheses across ethnic groups and by adiposity level, using the same statistical methods. Research Methods and Procedures: Samples included black vs. white (HERITAGE Family Study) and random vs. obese (Québec Family Study) individuals from 432 families (1432 individuals). Heritability for leptin, alternatively adjusted for age and sex and then for age, sex, and adiposity was estimated with the use of familial correlations. Heterogeneity in the magnitude of the familial resemblance between samples and the effect of adjusting for adiposity was explored. Results: Heritability did not vary across samples stratified by adiposity level or ethnic group or across adjustment schemes. Maximal heritability, the percentage of additive phenotypic variability due to all familial sources, was 32%. Discussion: Whereas leptin and adiposity were highly correlated within individuals, removing the effects of adiposity did not significantly alter the magnitude of the familial component for leptin. Moreover, this effect did not vary as a function of ethnicity (black vs. white) or adiposity level. Thus, no evidence for heterogeneity was detected. However, a comparison among previous studies raises questions concerning possible genetic heterogeneity in other ethnic groups in which complex interactions among leptin, adiposity, and diabetes status may be important.

The relations of tissue-type plasminogen activator (tPA) and plasminogen activator inhibitor (PAI) to cardiorespiratory fitness and high-density lipoprotein (HDL) subclasses were examined in 150 men aged 52 to 62 years. PAI activity was analyzed by a chromogenic assay, and tPA activity bioimmunologically. Total HDL was fractionated into three subclasses by density gradient ultracentrifugation, and concentrations of cholesterol, phospholipids, cholesterol esters and triglycerides were measured in each fraction. Maximal oxygen uptake (˙VO2max) was assessed by analyzing respiratory gases during a maximal exercise test. The mean (95%CI) tPA activity was 0.46 (0.42;0.50) IU · ml-1, PAI activity 19.4(17.5;21.3) AU · ml-1, total HDL cholesterol 43 (41;45) mg· dl-1 and ˙VO2max 30.6 (29.4;31.7) ml · kg-1 · min-1. ˙VO2max associated with PAI (r=-0.352, p<0.001) and tPA activities (r=0.31, p<0.001). PAI (r=-0.31 to -0.42) and tPA activities (r=0.37 to 0.47) were associated with total HDL, HDL2b and HDL2a cholesterol, cholesterol esters and phospholipids (p<0.001 for all), but not with these components in HDL3 subfraction. PAI (r=-0.20, p=0.014) and tPA (r=0.25, p=0.002) correlated with triglycerides in HDL2b, while PAI was directly and tPA inversely related (p<0.001 for both) to triglycerides in HDL3 subfraction. When age, ˙VO2max, sum of skinfolds, HDL2b cholesterol and HDL2b triglycerides were entered in multiple stepwise regression analyses, HDL2b cholesterol and ˙VO2max explained 20% of the variance in PAI activity. In the same model, HDL2b cholesterol, ˙VO2max and age explained 27% of the variance in tPA activity. In summary, tPA and PAI activities were mainly associated with larger than HDL2a particles, which are hypothesised to be more antiatherogenic than HDL3 subclass. Moreover, relation of cardiorespiratory fitness to fibrinolytic factors seem to be independent of plasma HDL.