A meta-analytic investigation of linkage and association of common leptin receptor (LEPR) polymorphisms with body mass index and waist circumference

Abstract

Methods: We analyzed data pooled from nine studies on the human leptin receptor (LEPR) gene for the association of three alleles (K109R, Q223R and K656N) of LEPR with body mass index (BMI; kg/m2) and waist circumference (WC). A total of 3263 related and unrelated subjects from diverse ethnic backgrounds including African-American, Caucasian, Danish, Finnish, French Canadian and Nigerian were studied. We tested effects of individual alleles, joint effects of alleles at multiple loci, epistatic effects among alleles at different loci, effect modification by age, sex, diabetes and ethnicity, and pleiotropic genotype effects on BMI and WC.

Results: We found that none of the effects were significant at the 0.05 level. Heterogeneity tests showed that the variations of the non-significant effects are within the range of sampling variation.

Conclusions: We conclude that, although certain genotypic effects could be population-specific, there was no statistically compelling evidence that any of the three LEPR alleles is associated with BMI or WC in the overall population.

Introduction

The role of homozygosity for inactivating mutations of the leptin receptor (LEPR) in producing extreme obesity syndromes in laboratory animals is established.1 Heterozygosity for Lepr mutations in mice and rats also results in increased in fat stores.2,3 Additionally, a small number of extremely obese humans from consanguineous pedigrees have been identified who are obese due to homozygosity for inactivating mutations of LEPR.4 However, instances of obesity due to inactivating mutations of LEPR in humans are quite rare. The question of whether more common polymorphisms of the LEPR gene confer increased susceptibility to obesity remains an open and important issue in the molecular physiology of human body weight. A relatively large number of studies5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24 have now addressed this question (see Table 1). As shown, there are marked differences among these studies in the conclusions reached. A few studies have detected significant effects on adiposity-related phenotypes in the primary sample or sub-samples, but others have not. In such situations, pooling data from several studies, which can include raw data pooling or meta-analysis of summary statistics from individual studies, can be useful for assessing the likelihood and magnitude of the association between the allelic variants and phenotypes of interest. To the extent that the non-significant results are due to type 2 errors, pooling data from multiple studies greatly increases power and reduces the probability of such an error. Alternatively, to the extent that some of the apparently significant results are simply due to type 1 errors resulting from a large number of independent studies being done on a related question, data pooling potentially reduces the type 1 error rate by conducting overall tests for all studies combined. Data pooling can be an especially useful for objectively and quantitatively addressing whether study-to-study variations in results are simply due to random sampling error or to deterministic factors across studies. In the latter case, data pooling may identify which factors (eg age, sex, ethnic background of subjects, etc) influence the outcome of the constituent studies.25

Table 1 Leptin receptor (LEPR) in humans: linkage and association studies

We report a pooled analysis/meta-analysis of the association of common LEPR polymorphism's with body mass index (BMI; kg/m2) and waist circumference (WC). In conducting the analysis, we pooled raw data from studies in which we have collaborated. This approach has two advantages. First, by relying only on raw data we avoid many of the statistical problems related to pooling summary statistics.26 Second, because studies were selected for inclusion on the basis of having been conducted by collaborating groups, rather than on the basis of having been published, publication bias is avoided.27 In the process of conducting this meta-analysis, many complex statistical challenges arose, which are described in a separate paper on methodology.28 The present paper focuses on the substantive findings of biological relevance that may be important to obesity researchers and clinicians, keeping methodological details to a minimum.

Methods

Samples

A total of 3263 individuals were included in this meta-analysis (Table 2). Sixty-two percent are related to one or more subjects in the data set. The largest number of generations among the family pedigrees in the pooled data was two. The subjects were ethnically diverse, ie African-American, Caucasian, Danish, Finnish, French Canadian and Nigerian; approximately 50% were female (Table 3).

Table 2 Data sets for the pooled analysis
Table 3 Descriptive statistics from the 3263 observations

Genotyping

Genomic DNA samples were prepared from whole blood in the laboratories providing each dataset. While several LEPR allelic variants have previously been described, three exonic polymorphisms that result in amino acid substitutions in the LEPR gene comprise the combined data set used in the present analysis: K109R (coding exon 2); Q223R (coding exon 4); and K656N (coding exon 12). Detection of polymorphisms was by PCR-restriction fragment length polymorphism analysis or SSCP screening (see the individual referenced data sets in Table 2 for the methods used).

Statistical methods

The statistical methods are described in detail in a separate paper.28 In brief, we tested for association of the LEPR polymorphisms with the phenotypes using both univariate and multivariate tests. We conducted association tests and joint tests for both linkage and association using transmission disequilibrium tests for quantitative traits.29 Finally, we tested for linkage using the new and old versions of the genetic model-free Haseman–Elston test30 using identity in state (IIS) rather than identity by descent (IBD) because we were interested in the effects of the specific polymorphisms rather than in general linkage to the genetic region of LEPR. In each situation, the individual and joint effects of each polymorphism were assessed. Both main effects of the polymorphisms, and their interactions (ie epistasis), were examined. Incomplete data were handled in two ways: (1) data were analyzed excluding cases with incomplete data for the analysis in question; and (2) analyses were repeated handling the incomplete information by multiple imputation.31 For the association studies involving related individuals, non-independence of observations was accommodated by the use of the ASSOC program in the S.A.G.E. software.32

All of these analyses were adjusted for covariates such as age, its polynomials and sex essentially through residual analyses. To test significance of heterogeneity among the estimates, we employed the chi-square Q-test statistic of Hedges and Olkin.33 Funnel plot of standard errors (s.e.) vs estimates from all studies and all loci for each of BMI and WC was used for visual exploration of potential publication bias. In addition, Kendall's rank correlation between the estimates and the standard errors was used as a formal test of publication bias as suggested by Begg and Mazumdar.34

Results

Descriptive statistics are presented in Table 3. The range of age is large, as are the ranges of BMI and WC. The estimated allele frequencies of the alleles at exon 2, exon 4 and exon 12 were 0.23, 0.48 and 0.20, respectively. The maximum likelihood test35 for departure from Hardy–Weinberg equilibrium showed no evidence for a departure for alleles at any exon: exon 2 (P=0.285); exon 4 (P=0.597); and exon 12 (P=0.537). The multiple imputation resulted in little change in statistical significance: exon 2 (P=0.960); exon 4 (P=0.770); and exon 12 (P=0.374). However, as expected, the three exons are all in significant in pair-wise linkage disequilibrium regardless of the imputation: exon 2 vs exon 4 (P<0.001); exon 2 vs exon 12 (P<0.001); and exon 4 vs exon 12 (P<0.001).

Results of the ASSOC analyses are presented in terms of differences of the estimated effects of the two genotypes, the ‘wild-type’ homozygote and the heterozygote, on BMI and WC, separately, from those of the ‘mutant’ homo-zygous genotypes after adjusting for age and sex (see Table 4). For example, subjects heterozygous (K109R) for the exon 2 allele had a meta estimate effect size of 0.03 on BMI when compared to subjects with K109K genotype. No single effect was significant from either the individual studies or the meta-analysis. We also assessed the significance of the phenotypic variation due to genotypic variation, by exon (data not shown). The effects were also not significant for either phenotype for any exon from individual studies or from the meta-analysis; P-values ranged from 0.10 to 0.95.

Table 4 Estimated effects and standard errors (with reference to mutant homozygotes) of each genotype on body mass index (BMI) and waist circumference (WC) at each locus adjusted for age and sex

Potential effect modifications were evaluated by testing significance of interaction effects among genotypes and main covariates using backward elimination ordinary least squares (OLS) regression. Only the interaction effect between R109R at exon 2 and sex was significant (P=0.027) for BMI, implying that male subjects with R109R genotype at exon 2 have significantly higher BMI than the other subjects. However, the contribution of this interaction effect to the variations of BMI is minimal (increase in r2<0.01%). The non-significant allele-by-environment interaction effects suggest that the genotypic effects, if any, are not significantly modified by the main covariates such as diabetes, sex, age and ethnicity. Finally, multivariate analyses for pleiotropic effects of the alleles on BMI and WC adjusted for sex, ethnicity, diabetes and age polynomials indicated that the polymorphisms had no statistically significant simultaneous effects on BMI and WC (see Heo et al28 for details).

Results of the Hedges–Olkin chi-square test statistic Q are listed in Table 4. The non-significant heterogeneity indicates that study-to-study variation in outcome is consistent with simple random sampling variation.

The funnel plots (Figure 1) for both BMI and WC show that the estimates are fairly symmetric about 0 and fanning out, as the s.e.s get larger. Furthermore, Kendall's rank correlation did not support publication bias for either BMI (0.058, P=0.553) or WC (−0.060, P=0.627).

Figure 1
figure1

Funnel plot of standard errors vs estimates of genotypic effects on (a) BMI and (b) WC obtained from all studies and loci (Table 4).

Discussion

The lack of association between the LEPR amino acid substitutions and obesity indices, despite a large sample, suggests that the substitutions do not affect these phenotypes. While amino acid substitutions may result in either non-functioning or poorly functioning proteins or even functional proteins (if the effect is ‘silent’), it is important to note that among complex traits with multiple pathways, such as obesity, the absence of association does not necessarily indicate a lack of effect. It may simply be that persons with the amino acid substitution compensated by other means, or that additional genotypic factors may be involved and need to be taken into account before the phenotype becomes manifest. Another possibility is ‘hyper’ functioning, ie a protein that has greater functional activity than ‘wild-type’. This is a less common result of mutation, but is a formal possibility with precedent. Thus, we might have detected a hyper-functioning ‘leanness’ allele, although there was no strong evidence for this.

We acknowledge that the Hedges–Olkin chi-square test statistic is often less powerful than a direct test of heterogeneity as a function of a specific measured moderator variable. However, we believe that the results of the tests (Table 4) are relatively strong in suggesting that the variation due to sources (such as different populations and genotyping in different laboratories) is not substantial.

With respect to possible ‘collaboration’ bias, although it is theoretically possible, it is not obvious to us exactly how such bias would operate or that it would be likely, given that our collaborators were not selected on the basis of the results of their LEPR studies. However, the patterns on the funnel plots (Figure 1) and the non-significant Kendall's rank correlations imply that there are no substantial biases due to ‘collaboration’ and/or unpublished reports.

The lack of association, of course, does not rule out the possibility that the three alleles may influence intermediate traits, or phenotypes, not examined as part of these analyses. Moreover, these results might have been very different if we had examined LEPR along with other genes known to influence energy homeostasis (eg CART, LEP, MC4R). This sort of contingent/epistatic analysis (see Horikawa et al36 for an example) is required if we are to fully understand the genetics of complex traits such as human weight homeostasis.

References

  1. 1

    Chua S, Leibel RL . Obesity genes: molecular and metabolic mechanisms Diabetes Rev 1997 5: 2–7.

  2. 2

    Chung WK, Power-Kehoe L, Chua M, Chu F, Aronne L, Huma Z, Sothern M, Udall JN, Kahle B, Leibel RL . Exonic and intronic variation in the leptin receptor (OBR) of obese humans Diabetes 1997 46: 1509–1511.

  3. 3

    Zhang YY, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM . Positional cloning of the mouse obese gene and its human homolog Nature 1994 372: 425–432.

  4. 4

    Clement K, Vaisse C, Lahlou N, Cabrol S, Pelloux V, Cassuto D, Gourmelen M, Dina C, Chambaz J, Lacorte JM, Basdevant A, Bougneres P, Lebouc Y, Froguel P, Guy-Grand B . A mutation in the human leptin receptor gene causes obesity and pituitary dysfunction Nature 1998 392: 398–401.

  5. 5

    Considine RV, Considine EL, Williams CJ, Hyde TM, Caro JF . The hypothalamic leptin receptor in humans, identification of incidental sequence polymorphisms and absence of the db/db mouse and fa/fa rat mutations Diabetes 1996 19: 992–994.

  6. 6

    Norman RA, Leibel RL, Chung WK, Power-Kehoe L, Chua SC, Knowler WC, Thompson DB, Bogardus C, Ravussin E . Absence of linkage of obesity and energy metabolism to markers flanking homologues of rodent obesity genes in Pima Indians Diabetes 1996 45: 1229–1232.

  7. 7

    Echwald SM, Sorensen TD, Sorensen TI, Tybaerg-Hansen A, Andersen T, Chung WK, Leibel RL, Pedersen O . Amino acid Variants in the human leptin receptor: lack of association to juvenile onset obesity Biochem Biophys Res Commun 1997 233: 248–252.

  8. 8

    Francke S, Clement K, Dina C, Inoue H, Behn P, Vatin V, Basdevant A, Guy-Grand B, Permutt MA, Froguel P, Hager J . Genetic studies of the leptin receptor gene in morbidly obese French Caucasian families Hum Genet 1997 100: 491–496.

  9. 9

    Gotoda T, Manning BS, Goldstone AP, Imrie H, Evans AL, Strosberg AD, McKeigue PM, Scott J, Aitman TJ . Leptin Receptor Gene variation and obesity: lack of association in a white British male population Hum Mol Genet 1997 6: 869–876.

  10. 10

    Matsuoka N, Ogawa Y, Hosoda K, Matsuda J, Masuzaki H, Miyawaki T, Azuma N, Natsui K, Nishimura H, Yoshimasa Y, Nishi N, Thompson DB, Nakao K . Human leptin receptor gene in obese Japanese subjects: evidence against either obesity causing mutations or association of seauence variants with obesity Diabetologia 1997 40: 1204–1210.

  11. 11

    Silver K, Walston J, Chung WK, Yao F, Parikh VV, Andersen RE, Cheskin LJ, Elahi D, Muller D, Leibel RL, Shuldiner AR . The Gln223Arg and Lys656Asn Polymorohisms in the human leptin receptor do not associate with traits related to obesity Diabetes 1997 46: 1898–1900.

  12. 12

    Hasstedt SJ, Hoffman M, Leppert MF, Elbein SC . Recessive inheritance of obesity in familial non-insulin-dependent diabetes mellitus, and lack of linkage to nine candidate genes Am J Hum Genet 1997 61: 668–677.

  13. 13

    Thompson DB, Ravussin E, Bennet PH, Bogardus C . Structure and sequence variation at the human leptin receptor gene in lean and obese Pima Indians Hum Mol Genet 1997 6: 675–679.

  14. 14

    Oksanen L, Kaprio J, Mustajoki P, Kontula K . A common pentanucleotide polymorphism of the 3-untranslated part of the leptin receptor gene generates a putative stem-loop motif in the mRNA and is associated with serum insulin levels in obese individuals Int J Obes Relat Metab Disord 1998 22: 634–640.

  15. 15

    Rolland V, Clement K, Dugail I, Guy-Grand B, Basdevant A, Froguel P, Lavau M . Leptin receptor gene in a large cohort of massively obese subjects: no indication of the fa/fa rat mutation. Detection of an intronic variant with no association with obesity Obes Res 1998 6: 122–127.

  16. 16

    Norman RA, Tataranni PA, Pratley R, Thompson DB, Hanson RL, Prochazka M, Baier L, Ehm MG, Sakul H, Foroud T, Garvey WT, Burns D, Knowler WC, Bennet PH, Bogardus C, Ravussin E . Autosomal genomic scan for loci linked to obesity and energy metabolism in Pima Indians Am J Hum Genet 1998 62: 659–668.

  17. 17

    Roth H, Korn T, Rosenkranz K, Hinney A, Ziegler A, Kunz J, Siegfried W, Mayer H, Hebebrand J, Grzeschik K . Transmission disequilibrium and sequence variants at the leptin receptor gene in extremely obese German children and adolescents Hum Genet 1998 103: 540–546.

  18. 18

    Chagnon YC, Chung WK, Perusse L, Chagnon M, Leibel RL, Bouchard C . Linkages and associations between the leptin receptor (LEPR) gene and human body composition in the Quebec Family Study Int J Obes Relat Metab Disord 1999 23: 278–286.

  19. 19

    De Silva AM, Walder KR, Aitman TJ, Gotoda T, Goldstone AP, Hodge AM, De Courten MP, Zimmet PZ, Collier GR . Combination of Polymorphisms in OB-R and the OB gene associated with insulin resistance in Nauruan males Int J Obes Relat Metab Disord 1999 23: 816–822.

  20. 20

    Bray MS, Boerwinkle E, Hanis CL . Linkage analysis of candidate obesity genes among the Mexican-American population of Starr County, Texas Genet Epidemiol 1999 16: 397–411.

  21. 21

    Chagnon YC, Wilmore JH, Borecki IB, Gagnon J, Perusse L, Chagnon M, Collier GR, Leon AS, Skinner JS, Rao DC, Bouchard C . Associations between the leptin receptor gene and adiposity in middle-aged Caucasian males from the HERITAGE Family study J Clin Endocrinol Metab 2000 85: 29–34.

  22. 22

    Endo K, Yanagi H, Hirano C, Hamaguchi H, Tsuchiya S, Tomura S . Association of Trp64Arg polymorphism of the beta-3-adrenergic receptor gene and no association of Gln223Arg polymorphism of the leptin receptor gene in Japanese schoolchildren with obesity Int J Obes Relat Metab Disord 2000 24: 443–449.

  23. 23

    Del Giudice EM, Perrone L, Forabosco P, Devoto M, Carbone MT, Calabro C, Ditoro R . Linkage study of early-onset obesity to leptin receptor gene in Italian children Nutr Res 2000 20: 1059–1063.

  24. 24

    van der Kallen CJ, Cantor RM, van Greevenbroek MM, geurts JM, Bouwman FG, Aouizerat BE, Allayee H, Buurman WA, Lusis AJ, Rotter JI, de Bruin TW . Genome scan for adiposity in Dutch dyslipidemic families reveals novel quantitative trait loci for leptin, body mass index and soluble tumor necrosis factor receptor superfamily 1A Int J Obes Relat Metab Disord 2000 24: 1381–1391.

  25. 25

    Draper D, Graver DP, Goel PK, Greenhouse JB, Hedges LV, Morris CN, Tucker JR, Watenaux CM . Combining information: statistical issues and opportunities for research National Academy Press: Washington, DC 1992.

  26. 26

    Rosenthal R . Meta-analytic procedures for social research Sage: Newbury Park, CA 1991.

  27. 27

    Allison DB, Faith M . Publication bias in obesity treatment trials? Int J Obes Relat Metab Disord 1996 20: 931–937.

  28. 28

    Heo M, Leibel RL, Boyer BB, Chung WK, Koulu M, Karvonen MK, Pesonen U, Rissanen A, Laakso M, Uusitupa MIJ, Chagnon Y, Bouchard C, Donohoue PA, Burns TL, Shuldiner AR, Silver K, Andersen RE, Pedersen O, Echwald S, Sorensen TIA, Behn P, Permutt MA, Jacobs KB, Elston RC, Hoffman DJ, Allison DB . Pooling analysis of genetic data: the association of leptin receptor (LEPR) polymorphisms with variables related to human adiposity Genetics 2001 159: 1163–1178.

  29. 29

    Allison DB, Heo M, Kaplan N, Martin ER . Development of Sibling-based Tests of Linkage in the Presence of association for quantitative traits that do not require parental information Am J Hum Genet 1999 64: 1754–1764.

  30. 30

    Elston RC, Buxbaum S, Jacobs KB, Olson JM . Haseman and Elston revisited Genet Epidemiol 2000 19: 1–17.

  31. 31

    Rubin DB . Multiple imputations in sample surveys—a phenomenological Bayesian approach to nonresponses In Proceedings of the survey research methods section American Statistical Association 1978 pp 20–34.

  32. 32

    S.A.G.E. Statistical analysis for genetic epidemiology, release 3.1 Computer program package available from the Department of Epidemiology and Biostatistics, Rammelkamp Center for Education and Research, MetroHealth Campus, Case Western University, Cleveland 1997.

  33. 33

    Hedges LV, Olkin I . Statistical methods for meta-analysis Academic Press: New York 1985.

  34. 34

    Begg CB, Mazumdar M . Operating characteristics of a rank correlation test for publication bias Biometrics 1994 50: 1088–1101.

  35. 35

    Lynch M, Walsh B . Genetics and analysis of quantitative traits Sinauer Associates: Sunderland, MA 1985.

  36. 36

    Horikawa Y, Oda N, Cox NJ, Li X, Orho-Melander M, Hara M, Hinokio Y, Lindner TH, Mashima H, Schwarz PE, del Bosque-Plata L, Horikawa Y, Oda Y, Yoshiuchi I, Colilla S, Polonsky KS, Wei S, Concannon P, Iwasaki N, Schulze J, Baier LJ, Bogardus C, Groop L, Boerwinkle E, Hanis CL, Bell GI . Genetic variation in the gene encoding calpain-10 is associated with type 2 diabetes mellitus Nature Genet 2000 26: 163–175.

  37. 37

    Tanizawa YA, Riggs C, Dagog-Jack S . Isolation of the human LIM/homeodomain gene islet-1 and identification of a sample sequence repeat polymorphism Diabetes 1994 43: 935–941 [Published erratum appears in Diabetes 1994; 43: 1171.]

  38. 38

    Donohoue PA, Burns TL, Mendoza MCB . Lys656Asn variant of the leptin receptor gene (LEPR) and the β-3 adrenergic receptor (β3AR) gene linked to body mass index in humans: the Muscatine study Pediatr Res 2000 47: 127A.

Download references

Acknowledgements

This study was supported in part by the NIH grants R01DK51716, R01DK52431, R01ES09912, F33DK09919, P30DK26687, DK25295 and P41RR03655. We thank Drs Jose Fernandez and Gary Gadbury for their valuable input.

Author information

Correspondence to DB Allison.

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • body mass index
  • waist circumference
  • meta-analysis
  • leptin receptor polymorphism
  • linkage
  • association

Further reading