International Journal of Obesity
SEARCH     advanced search my account e-alerts subscribe register
Journal home
Advance online publication
Current issue
Press releases
For authors
For referees
Contact editorial office
About the journal
For librarians
Contact Springer Nature
Customer services
Site features
NPG Subject areas
Access material from all our publications in your subject area:
Biotechnology Biotechnology
Cancer Cancer
Chemistry Chemistry
Dentistry Dentistry
Development Development
Drug Discovery Drug Discovery
Earth Sciences Earth Sciences
Evolution & Ecology Evolution & Ecology
Genetics Genetics
Immunology Immunology
Materials Materials Science
Medical Research Medical Research
Microbiology Microbiology
Molecular Cell Biology Molecular Cell Biology
Neuroscience Neuroscience
Pharmacology Pharmacology
Physics Physics
Browse all publications
October 2001, Volume 25, Number 10, Pages 1471-1473
Table of contents    Previous  Article  Next   [PDF]
Leptin resistance is associated with extreme obesity and aggregates in families
J H Leea, D R Reedb and R A Price

Center for Neurobiology and Behavior, Department of Psychiatry, University of Pennsylvania, Philadelphia, Pennsylvania, USA

Correspondence to: R A Price, Center for Neurobiology and Behavior, University of Pennsylvania, 415 Curie Blvd, CRB 135b, Philadelphia, PA 19104, USA. E-mail:

aCurrent address: Department of Epidemiology, Columbia University, New York, New York, USA.

bCurrent address: Monell Chemical Senses Center, Philadelphia, Pennsylvania, USA.


OBJECTIVE: To examine the determinants of plasma leptin levels and leptin resistance in a sample of extremely obese subjects and their relatives.

DESIGN and METHOD: We obtained plasma leptin values on 968 individuals from 218 families having both extremely obese and average weight members for obesity related variables. Multivariate regression analyses were used to identify predictors of both plasma leptin concentration and an index of leptin resistance. Family correlations and heritabilities were computed for plasma leptin and indices of leptin resistance.

RESULTS: Body mass index and sex predicted 63% of the variance in plasma leptin. Extremely obese subjects were most likely and average weight subjects least likely to be leptin resistant. Both leptin and leptin resistance aggregated within families.

CONCLUSION: Leptin resistance is strongly associated with extreme obesity and appears to be a heritable trait. The determination of its genetic causes will aid in understanding the role of leptin in common forms of human obesity.

International Journal of Obesity (2001) 25, 1471-1473


body mass index; familial risk; leptin resistance


Leptin is a hormone secreted by adipose tissue that acts in the central nervous system as a negative feedback signal to regulate appetite, metabolism and sexual maturation.1,2 Extremely obese individuals have high leptin levels, which suggests they are resistant to its anorectic and metabolic effects.1 Meta-analyses support linkage between obesity and the region containing the leptin gene.3 Further, several studies examining polymorphisms in the 5' region of the leptin gene and obesity reported positive associations.4,5,6 Together, these studies indicate that leptin may play a role in the development of common forms of obesity.

Materials and methods


We studied 968 individuals from 218 families ascertained from ongoing genetic studies of obesity at the University of Pennsylvania. A detailed description of the characteristics of the study participants was presented in a previous paper.7 To be included in the study, probands must have a body mass index (BMI; weight (kg)/height (m2)) greater than or equal to 40 kg/m2, have one or more siblings with a current BMI greater than or equal to 30 kg/m2, one or more siblings must have a maximum lifetime BMI less than 27 kg/m2, and at least one parent must have a maximum lifetime BMI less than 27 kg/m2.

Obesity phenotypes

We directly measured weight, height, waist and hip circumferences, and percentage body fat (bioelectrical impedance; Valhalla Scientific). The waist-to-hip ratio (WHR) was computed by dividing the circumference of the waist (cm) by the circumference of the hips (cm). Plasma leptin was assayed using a commercially available radioimmunoassay kit (Linco Research, St Louis, MO).

Leptin resistance

For use as an index of leptin resistance, we computed the ratio of leptin to BMI (LEP/BMI). This index measures leptin level while controlling for the contribution of BMI. We chose BMI rather than percentage body fat or fat mass because in multivariate models, BMI predicted the most variance; also, more cases were available for BMI than for percentage fat or fat mass. For qualitative comparisons, a threshold of the 90th percentile of the leptin to BMI ratio (adusted for age within race, sex and generation) was chosen because it yielded an adequate number of cases with a high leptin-to-BMI ratio. In analyses of family correlations, we also examined leptin residualized for BMI.

Statistical methods

We conducted multiple regression analyses to evaluate the relationship between leptin and BMI, percentage fat, fat mass, waist circumference, hip circumference and WHR.

We computed the odds ratios using logistic regression to assess the risk of leptin resistance (LEP/BMI greater than the 90th percentile) using extreme obesity (BMI greater than 40 kg/m2) and normal weight (BMI<27 kg/m2).

We computed familial correlation coefficients for leptin and LEP/BMI, both adjusted for age within race, generation and sex. We also examined the leptin measure residualized for BMI. We applied equal weight to each nuclear family. The correlations were computed using the computer program FCOR2 from the SAGE 4.0 BETA 6 package ( We also estimated heritability of the leptin and leptin resistance measures using the computer program package SOLAR ( Adjustment for covariates was completed as described above, ie outside the SOLAR program. In the heritability analyses using SOLAR, we corrected for ascertainment of the primary proband.

Descriptive, linear and logistic regression analyses were computed using the SPSS software (SPSS 9.0 for WINDOWS 2000).


Descriptive and demographic statistics

Mean leptin levels differed by family relationship and by sex within each family relationship (Table 1). As expected, the mean leptin level was highest for probands (59.0 ng/ml) and lowest for fathers (11.9 ng/ml) and brothers (11.7 ng/ml), who also had the lowest BMI. Plasma leptin concentration for females was much higher than for males, in part because most extremely obese individuals were women. No signi-ficant difference in leptin levels were observed between European-American and African-American individuals.

Linear regression analysis between leptin and other obesity characteristics

In univariate analyses, BMI, hip circumference, and percentage fat accounted for 55, 53 and 49% of the phenotypic variance, respectively. Sex and waist circumference were also predictors of leptin level, accounting for 23 and 32% of the variance, respectively. Although WHR and age were statistically related to leptin values, they explained little phenotypic variance (8 and 1%, respectively). Race was not a significant predictor of plasma leptin levels, accounting for 0% of the variance.

We then applied multivariate models using variables significant in the univariate analysis. A bivariate model including sex and BMI accounted for the most variance (r2=0.63). Other models did not account for significant amounts of additional variance. Thus, the model including BMI and sex was the best predictive model.

Risk for leptin resistance

Extreme obesity (BMI40) predicted leptin resistance (LEP/BMI90th percentile) with odds ratio of 4.12 (CI=3.29-5.16). Normal weight (BMI<27) predicted the absence of leptin resistance with an odds ratio of 0.13 (CI=0.01-0.20). Sex did not predict leptin resistance (see Table 2).

Family resemblance

Parent-offspring and sibling correlations were highest for leptin residualized for BMI, 0.12 (P<0.05) and 0.18 (P<0.05), respectively. Parent-offspring and sibling correlations were positive but low for plasma leptin, 0.07 and 0.06, respectively, and neither differed from zero. Parent-offspring correlations were similarly low for leptin to BMI ratio, 0.05. However, the sibling correlation for leptin/BMI was higher (0.12) and reached statistical significance (P<0.05). Polygenic analysis which controlled for the ascertainment of the primary proband gave significant heritabilities for plasma leptin (0.18), leptin/BMI (0.15), and leptin residualized for BMI (0.16). All heritabilities were significantly different from zero.


We found BMI and sex accounted for 63% of the variance in plasma leptin levels, while race, age and indices of body fat distribution accounted for almost none of the remaining variability. The subjects with highest leptin levels were examined to determine the factors associated with this leptin resistance. Leptin resistance appears to be common in the extremely obese but rare in normal weight subjects.

Plasma leptin concentration aggregated in families. Because plasma leptin concentrations are highly correlated with BMI and BMI is heritable, this observation is expected. However, the families studied here were selected for extreme discordance in body weight,7 and so the familial aggregation of plasma leptin concentration was surprising. Other studies using families unselected for obesity variables8,9,10 found additive genetic effects ranging from 39 to 73% for plasma leptin, but the familial similarity for leptin was largely due to the familial aggregation of BMI. When our families discordant for weight were examined, additive genetic effects accounted for 18% of the variance in plasma leptin and 15-16% of the variance in leptin resistance. Because our families were ascertained for extreme obesity and normal weight, family correlations appear to have been suppressed and the true heritabilities may be higher.

In summary, we found that both plasma leptin level and leptin resistance are heritable traits. Plasma leptin levels and an index of leptin resistance were related to extreme obesity.


We thank the participating families for their generous cooperation. This research was supported in part by NIH grants R01DK44073, R01DK48095 and R01DK56210 to RAP. Some of the results of this paper were obtained by using the program package SAGE, which is supported by a US Public Health Service Resource grant (1P41RR03655) from the National Center for Research Resources.


1 Friedman JM, Halaas JL. Leptin and the regulation of body weight in mammals. Nature 1998; 395: 763-770, Article MEDLINE

2 Chehab FF, Mounzih K, Lu R, Lim ME. Early onset of reproductive function in normal female mice treated with leptin. Science 1997; 275: 88-90, Article MEDLINE

3 Allison D, Heo M. Meta-analysis of linkage data under worst-case conditions. A demonstration using the human ob region. Genetics 1998; 148: 859-866, MEDLINE

4 Hager J et al. A polymorphism in the 5' untranslated region of the human ob gene is associated with low leptin levels. Int J Obes Relat Metab Disord 1998; 22: 200-205, MEDLINE

5 Mammes O et al. Novel polymorphisms in the 5' region of the LEP gene: association with leptin levels and response to low-calorie diet in human obesity. (Published erratum appears in Diabetes 2000; 49: 1617.), 1998, Diabetes. 47: 487-489.

6 Li W et al. Sequence variants in the 5' flanking region of the leptin gene are associated with obesity in women. Ann Hum Genet 1999; 63: 227-234, Article MEDLINE

7 Price RA, Reed DR, Lee JH. Obesity related phenotypes in families selected for extreme obesity and leanness. Int J Obes Relat Metab Disord 1998; 22: 406-413, MEDLINE

8 Narkiewicz K et al. Heritability of plasma leptin levels: a twin study. (In Process Citation.). J Hypertens 1999; 17: 27-31, MEDLINE

9 Comuzzie A et al. A major quantitative trait locus determining serum leptin levels and fat mass is located on human chromosome 2. Nature Genet 1997; 15: 273-276, MEDLINE

10 Rotimi C et al. Heritability of plasma leptin in a population sample of African-American families. Genet Epidemiol 1997; 14: 255-263, MEDLINE


Table 1 Mean values of age, body mass index (BMI), and leptin stratified by family relationship and subject characteristics

Table 2 Risk of leptin resistance (LEP/BMI) for family members

Received 12 January 2000; revised 31 October 2000; accepted 20 December 2000
October 2001, Volume 25, Number 10, Pages 1471-1473
Table of contents    Previous  Article  Next    [PDF]