SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Disclosure
  7. REFERENCES

This study estimated the genetic and environmental determinants of plasma leptin and insulin levels and of obesity-related phenotypes. Included in this analysis were family members from 80 families living in kibbutz settlements, who participated in two examinations 8–10 years apart. We estimated that polygenes explained 30–50% of the adjusted leptin and insulin levels and 30–70% of the anthropometric phenotypes. This study demonstrated a significant genetic influence on longitudinal changes in leptin and BMI (h2 = 0.45) and small-to-moderate heritability estimates for changes in insulin and other obesity-related phenotypes. In bivariate genetic analyses, we observed positive genetic correlations between leptin and anthropometric phenotypes, suggesting that shared effects of the same sets of loci account for 20–30% of the additive genetic variance in these pairs of variables. Shared genetic factors also account for 20–25% of the additive genetic variance in insulin—anthropometric pairs of variables.

It has been shown that leptin, which is primarily secreted by adipocytes, is critical to feeding behaviors and weight maintenance. Several studies have reported that genetic factors influence leptin concentrations (1). The well-established statistical and physiological relationships between measures of leptin, insulin, and obesity-related phenotypes may suggest common underlying genetic and environmental influences.

Thus, the present investigation evaluates the evidence that genes and environmental factors that influence leptin and insulin levels exert pleiotropic effects on obesity-related phenotypes. In addition, we examine the genetic influences on changes in leptin, insulin, and anthropometric phenotypes over time. The study makes use of an unselected sample of family members who participated in two examinations, 8–10 years apart, in the Kibbutz Family Study.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Disclosure
  7. REFERENCES

Eighty kindreds were recruited and a total of 476 individuals were screened at visit 1 (ref. 2) and 379 subjects were re-examined at the visit 2 (79.6% response rate). At both visits, the family members completed a self-administrated medical questionnaire and a fasting blood samples were obtained and stored at −80 °C until assayed for plasma leptin and insulin concentrations by radioimmunoassays (Linco, St Charles, MO).

Variance components decomposition method, as implemented in SOLAR (3), was then used to estimate the unmeasured additive genetic and shared environmental components of the variability in leptin, insulin, and anthropometric traits. Maximum likelihood methods were used to simultaneously estimate the effects of covariates, heredity, and environment components. The significance of genetic effect was assessed by comparing the likelihoods of models in which this parameter was estimated with models in which it was constrained to zero.

A bivariate genetic analysis was implemented to examine the possible pleiotropic genetic effects and shared environmental influences on leptin and anthropometric traits and on insulin and anthropometric phenotypes. This approach uses estimated heritabilities of the two traits to partition the phenotypic correlations (ρP) between the pairs of phenotypes into two additive components attributable to pleiotropic genetic (ρG) and common environmental (ρE) effects.

An estimated correlation coefficient indicates genetic pleiotropy and/or shared environmental effects and its square is an estimate of the proportion of the additive genetic (environmental) variance in each trait of the pair phenotypes which is due to the effects of the same set of genes (environmental factors).

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Disclosure
  7. REFERENCES

Adjustment for sex and age accounted for 46.1, 3.6, 31.6, 15.3, and 58.7% of the total variation seen in plasma leptin, insulin, weight, BMI, and waist-to-hip ratio, respectively. Additional lifestyle variables added 0.5–6.2% to the explained variability.

The sources of variation in each of the phenotypes are summarized in Table 1. At both visits, polygenes explained 30–50% of the adjusted interindividual variability in leptin and insulin levels and 30–70% of the anthropometric phenotypes. When the analyses were repeated on levels adjusted for a set of demographic and lifestyle variables, BMI and waist-to-hip ratio, the heritability estimate was somewhat higher for leptin (h2 = 0.38) and lower for insulin (h2 = 0.28) (data not shown).

Table 1.  Genetic determinants of leptin, insulin, and anthropometric phenotypes
inline image

Table 1 presents high genetic influence on adjusted leptin change which did not changed after the adjustment for visit 1 leptin level. The heritability estimate for change in insulin levels was moderate and tended to be higher after further adjustment for visit 1 levels. The results also indicate that changes in BMI over 8–10-year period appear to be genetically influenced and the heritability estimates for changes in weight, waist, and waist-to-hip ratio the heritability estimates were relatively low.

Table 2 indicates that at visit 1, positive genetic correlations were observed for leptin with anthropometric phenotypes, suggesting that shared genetic effects accounted for 30% of the additive genetic variance in these pairs. A significant influence of the shared environmental factors on the relationship between these pairs of variables was also observed. At visit 2, shared genetic factors accounted only for 20% of additive genetic variance in these leptin—anthropometric pairs, and common environment factors accounted for 45–50% of the association between these pairs.

Table 2.  Additive genetic (ρG) and environmental (ρE) correlations from bivariate genetic analysisa of leptin and insulin with anthropometric phenotypes
inline image

Positive genetic correlations were also observed for the leptin-weight pair of changes, and leptin-BMI pair of change, suggesting that shared genetic effects accounted for 12–14% of the additive genetic variance in these pairs of phenotypes. Shared environmental effects accounted for ∼25% of the relationship between these pairs.

At visit 1, strong positive genetic correlations were observed for adjusted insulin–anthropometric pair of variables (20–25% of the additive genetic variance). Small influence of the shared environmental factors on the relationship between these pairs of variables was observed. At visit 2, shared genetic effects accounted for a smaller percentage and shared environmental factors accounted for a greater percentage of the relationship between these pairs of phenotypes. None of the genetic and environmental correlations for pairs of changes in insulin–anthropometric variables were statistically significant.

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Disclosure
  7. REFERENCES

The present study detected evidence for genetic effects on the levels of plasma leptin, insulin, and anthropometric phenotypes in an unselected sample of family members. The results at visit 1 and visit 2 for leptin and insulin were similar to heritability estimate obtained in other family studies (1,4).

The polygenic estimates (h2 = 0.5–0.7) which resulted from our analysis applied to adjusted obesity-related phenotypes are in agreement with the estimates in several twin and family studies (5).

In the present study, analysis of longitudinal changes in leptin and BMI revealed a significant polygenic influence. The genetic influence on change in insulin and other anthropometric phenotypes was modest. Several studies have shown that changes in body mass and adiposity have a genetic component (6). Association and linkage studies also provided evidence for a genetic influence on long-term weight change (7,8).

Because it has been suggested that insulin resistance represents the primary physiological defect underlying the metabolic syndrome, we can hypothesize that genes influencing both leptin and insulin levels would have pleiotropic effects on a number of obesity-related phenotypes. Indeed, our results indicated that shared genetic factors accounted for a considerable portion of the additive genetic variance in the leptin–anthropometric pairs. Utilizing various methodologies, several family studies (9,10), but not all (1), have also suggested pleiotropy in the leptin–anthropometric pairs.

Our results also suggest that insulin and anthropometric phenotype pairs were influenced by both shared genetic factors and environmental factors. Similarly, in the San Antonio Family Heart Study, high-genetic correlations were observed between insulin levels and obesity-related phenotypes (4).

In sum, the heritability of levels and changes in leptin-, insulin-, and obesity-related phenotypes, demonstrates the presence of genetic influences on the development of obesity and related traits. Mapping of genes underlying inter- and intraindividual variations in these phenotypes, and understanding their function(s), could lead to targeted intervention strategies to prevent obesity in the general population.

Disclosure

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Disclosure
  7. REFERENCES

The authors declared no conflict of interest.

REFERENCES

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Disclosure
  7. REFERENCES