Fasting plasma glucose is a multigenic trait related to both diabetes and obesity. We performed a genome scan for quantitative fasting plasma glucose levels in 320 families (1,514 subjects), segregating extreme obesity and normal weight using 382 autosomal microsatellite markers. We found significant linkages on 18q22-23 using family regression (logarithm of odds [LOD] 3.67, P = 0.00002, D18S1371 at 116 cM) and variance components (LOD 4.48, P < 0.00001) methods. Evidence for linkage remained strong when restricted to European Americans (260 families, 1,258 individuals). After an additional 60 families were added, the linkage signal strengthened (LOD 6.59). The result on 18q22-23 remained significant, even after controlling for both BMI and diabetes status. We also found suggestive linkages on chromosomes 2 (LOD 1.58, 216 cM) and 7 (LOD 1.78, 163 cM). Our results suggest that there is a quantitative trait locus in chromosome region 18q22-23 that influences fasting glucose levels and may play a role in the pathogenesis of type 2 diabetes. The strength of the serum glucose results after controlling for BMI suggests that this putative gene does not influence glucose levels merely through an effect on obesity.

Fasting plasma glucose is a key index of type 2 diabetes (NIDDM, MIM 125853) (1), and high levels of plasma glucose have been associated with obesity (2). Fasting glucose levels are moderately heritable, with heritabilities averaging around 30% (3,4). Previous genome scans have found linkage signals for fasting glucose in several chromosome regions (1p, 1q, 3p, 10q, and 17p [5–7]), a pattern that is consistent with a complex and multigenic mode of inheritance. Obesity is a major risk factor for type 2 diabetes, and nondiabetic or pre-diabetic obese individuals often have high plasma glucose levels. Studies of plasma glucose may help identify genes that influence the transition to insulin resistance in individuals at high risk for diabetes.

Our families were ascertained through obese probands (BMI >40 kg/m2) having obese and normal-weight siblings and parents (8). We selected 320 families (1,514 subjects), including European Americans (260 families, 1,258 individuals), African Americans (58 families, 247 individuals), and two families from other ethnic groups.

Most probands (301 of 320, 94.1%) and about two-thirds of all subjects (1,049 of 1,514, 69.3%) were women. Sibship size ranged from one to eight; 25 families had only one sibling, and most families (285 of 320, 89.1%) had two to six siblings (1–15 sibpairs), with a median sibship size of three. DNA was available from both parents for 194 families, one parent for 120 families, and no parent for 6 families.

Among these 1,514 subjects, the average fasting glucose level was 5.5 ± 2.2 mmol/l in the combined sample set and 5.5 ± 2.1 mmol/l in European Americans. The distribution of fasting glucose is shown in Table 1. Because several subjects have very high fasting glucose (>4 SD), log transformation was used to reduce skewness and kurtosis in the distribution.

Thirteen and one-half percent (204 of 1,514 individuals) had self-reported (type 1 or type 2) diabetes. The numbers of families with two, three, and four diabetic subjects were 27, 6, and 1, respectively. Thirty-six subjects had fasting glucose >7.0 mmol/l, but were without self-reported diabetes; those individuals could have undiagnosed diabetes or could be pre-diabetic. More than 62% (127 of 204 case subjects) of self-reported diabetic subjects had fasting glucose >7.0 mmol/l, and 44 of them had fasting glucose <5.6 mmol/l. For initial analyses of the genome scan, we deleted 44 diabetic individuals with glucose <5.6 mmol/l from our analyses. We assumed that those “diabetic” individuals with glucose <5.6 mmol/l (100 mg/dl, close to the mean of fasting glucose in our samples) were treated and that their plasma glucose levels were controlled. In subsequent analyses of linkage to chromosome 18, we conducted analyses that 1) included all self-reported diabetic subjects, 2) excluded all self-reported diabetic subjects, and 3) adjusted for self-reported diabetes status.

Genotyping was completed by Marshfield Genotyping Service using 382 autosomal microsatellite markers with an average interval of 8.9 cM (Marshfield set 11).

Two different quantitative analyses were used: family regression and variance components. The family regression analysis model developed by Sham et al. (9), based on squared differences and sums, appears to be largely insensitive to trait distribution or ascertainment thorough extreme phenotypes, an important consideration with our selected sample.

We found significant linkage on 18q22-23: logarithm of odds (LOD) 3.67 (P = 0.00002) at marker D18S1371 (116 cM) for log-transformed, age-adjusted fasting glucose levels by family regression and LOD 4.48 (P < 0.00001) by the variance components method (Table 2). In separate analyses for European Americans (260 families, 1,258 individuals), the linkage signal remained strong in chromosome region 18q22-23 (LOD 2.68, P = 0.0002) (Table 2).

To refine the chromosome 18 linkage results, we genotyped D18S1371 (116 cM) and ATA82B02 for an additional 53 European-American families (237 individuals) and 7 African-American families (37 individuals) accrued after genotyping for the genome scan began. The results, presented in Table 3 and Fig. 1, strengthened the overall findings, with a peak LOD score of 4.59 (P < 0.00001) using family regression analysis and 6.59 (P < 0.00001) using variance components analysis at 116 cM. Results are also presented for analyses after adjustment for BMI and also after conditioning on diabetes status. Even after controlling for both BMI and diabetes status, the LOD score was 3.43 (P < 0.00004) for the combined sample.

We were concerned about how treated and untreated diabetes might affect the outcome of the analyses. When variance components analyses were carried out using diabetes affection status as a covariate, the linkage results were not substantially different. The LOD scores changed from 6.59 to 5.09, 4.99 to 3.51, and 1.54 to 1.44 on D18S1371 (116 cM) for the combined sample, European Americans, and African Americans, respectively (Table 3). Exclusion of all self-reported diabetic patients substantially reduced the overall variance and LOD score for fasting plasma glucose. Separate linkage analyses using type 2 diabetes as the phenotype failed to provide suggestive linkage (LOD >1). Overall, the results suggest that the linkage result for serum glucose on chromosome 18 is not independent of diabetes status. Since glucose levels are diagnostic, this result is not surprising.

The 18q22-23 linked region spans ∼20 cM (1 LOD CI) (Fig. 1) and has >200 genes. However, several genes stand out as plausible candidates. These include the glucose-regulated zinc finger protein ZNF236 (MIM 604760) and the galanin receptor 1 (GALR1, MIM 600377). Both ZNF236 and GALR1 are within 2 Mb of marker D18S1371. ZNF236 can be induced by glucose (10). Galanin is a 29–amino acid peptide widely distributed in the peripheral and central nervous systems, where it acts as an important neuromodulator in the brain, gastrointestinal system, and hypothalamopituitary axis. GALR1 is a G-protein–coupled receptor expressed in various areas in the human brain (including hypothalamus, amygdale, and cortex cerebri) (11) and gastrointestinal tract (12).

In a separate study, we did not find linkage in the 18q22-23 region for obesity-related phenotypes (BMI, percentage of fat, and waist circumference) (13). The absence of linkage and the modest effect of using BMI as a covariate suggest that the putative gene(s) in 18q22-23 does not exert its influence on glucose levels through its effect on obesity.

We also found two other suggestive linkages on chromosome 18: D18S542, LOD 2.29, 41 cM, 18p11, and D18S858, LOD 2.62, 80 cM, 18q21 (Table 2 and Fig. 1). At least two research groups (14,15) have reported linkages for diabetic nephropathy on 18q. Lindgren et al. (16) found suggestive linkage on 18q12 for early-onset diabetes. A locus for type 1 diabetes (IDDM6, MIM 601941) also mapped to 18q21 (17). Melanocortin receptor 4 (MC4R, MIM 155541) gene mutations are relatively common in extremely obese individuals (18–20), and MC4R could be a candidate gene because of its proximity to the 18q21 peak (D18S858).

Only two suggestive linkages (LOD >1) were found for log-transformed, adjusted fasting glucose other than chromosome 18. One is on 2q35 (D2S434, LOD 1.58 by variance components method; LOD 1.32 by family regression). The linkage on 2q35 is >20 cM upstream of the calpain-10 gene (CAPN10, MIM 605286), but it may be difficult to rule out the relationship between the calpain-10 gene and the 2q linkage due to the coarse grained resolution of our 10-cM scan. The other is on 7q36 (D7S3070, LOD 1.78 by variance components and LOD 1.02 by family regression analyses), close to a quantitative trait locus for abdominal total fat reported by Rice et al. (21).

Although linkages for type 2 diabetes may overlap with quantitative trait loci for fasting glucose, some of the linkages identified in these analyses may be unique.

We used fasting plasma glucose as an index for insulin sensitivity (resistance) because our assessments took place in field settings. There are several more accurate methods for measuring insulin sensitivity, e.g., the euglycemic clamp and the oral glucose tolerance test. However, it was impractical to obtain these measures in a large-scale study such as ours.

In summary, we have found significant linkage for fasting plasma glucose on 18q22-23 in extremely obese families. This region may harbor a gene that influences plasma glucose level in obese individuals.

Three hundred twenty families (1,514 subjects) were chosen, as previously described (8). Briefly, all family probands (BMI ≥40 kg/m2) had at least one obese sibling (BMI ≥30 kg/m2) and at least one parent and one sibling who were of normal weight (BMI <27 kg/m2). Most of these families were European American (260 families, 1,258 individuals) or African American (58 families, 247 individuals); only two families (9 individuals) reported other ethnic origins. Phenotypes were available for most parents. Genome scan results for body weight–related phenotypes (BMI, percentage of fat, and waist circumferences) in 260 European-American families were reported elsewhere (13).

All subjects gave informed consent, and the protocol was approved by the Committee on Studies Involving Human Beings at the University of Pennsylvania.

Phenotypes.

All subjects fasted for at least 6 h before blood samples were drawn. Clinical coordinators interviewed all subjects and measured BMI and other obesity-related phenotypes in field settings. Medical history of obesity, type 1 and 2 diabetes, and treatment were also recorded. Two hundred four individuals had self-reported diabetes.

BMI was calculated based on measured height and weight: BMI = weight (in kilograms)/height (in meters)2. Fasting plasma glucose was measured by Quest Diagnostics (Philadelphia, PA). Log-transformed fasting glucose levels were adjusted for linear effects of age within generation, sex, and race using SPSS 11.0 (GLU_LRES) (Table 1), after which higher-order age effects were not significant. The heritability of log-transformed, adjusted fasting glucose in our samples was 0.24 (calculated by SOLAR [sequential oligogenic linkage analysis routines]).

DNA preparation and genotyping.

DNA was extracted using a high-salt method (22) and diluted to 10 ng/μl for genotyping. Three hundred eighty-two polymorphic Marshfield microsatellite markers from Marshfield screening set 11 were genotyped by the Marshfield Center for Medical Genetics. Map distances were taken from the Marshfield Database (http://research.marshfieldclinic.org/genetics). One family was duplicated (coded as different family) as an inner control for genotyping. Sex-specific PCR markers were amplified to verify sex. Mendel checks were performed by MERLIN, and all errors were corrected or dropped.

Statistical analyses.

Log-transformed adjusted glucose (GLU_LRES) was analyzed using the family regression (9,23) and variance components methods implemented in the computer program package MERLIN. For the family regression analyses, glucose values were standardized for age within race, sex, and generation. For the variance components analyses, sex, age, and generation were included as covariates. Follow-up analyses also used BMI and diabetes status as covariates. For the initial genome scan, we deleted 44 individuals from the study who had type 1 or type 2 diabetes but fasting plasma glucose levels <5.6 mmol/l. Analyses for combined samples, European Americans, and African Americans were carried out separately. Race-specific allele frequencies were used for European Americans and African Americans.

Separate variance components studies using diabetes affection status as covariates were also carried out in combined samples of European Americans and African Americans.

Additional genotyping and analyses.

To verify the linkage on 18q22-13, 53 European-American families (237 individuals) and 7 African-American families (37 individuals) were added after the genome scan finished. We genotyped markers D18S1371 and ATA82B02 for those 60 additional families. Both family regression and variance components analyses were conducted after new genotyping data were added. All self-reported diabetic subjects were included; the diabetes affection status was used as a covariate in the variance components study. We also adjusted fasting plasma glucose by BMI. For these later analyses, we either excluded all self-reported diabetic patients or included all individuals without regard for diabetes status.

FIG. 1.
FIG. 1. Family regression quantitative linkage using the computer program MERLIN for log-transformed, adjusted fasting plasma glucose on chromosome 18 in the combined sample (all), European Americans (EA), and African Americans (AA), after an additional 60 families were added. The 1-LOD CIs (∼20 cM) for combined samples and European Americans are labeled in the figure.
View largeDownload slide

Family regression quantitative linkage using the computer program MERLIN for log-transformed, adjusted fasting plasma glucose on chromosome 18 in the combined sample (all), European Americans (EA), and African Americans (AA), after an additional 60 families were added. The 1-LOD CIs (∼20 cM) for combined samples and European Americans are labeled in the figure.

FIG. 1.
FIG. 1. Family regression quantitative linkage using the computer program MERLIN for log-transformed, adjusted fasting plasma glucose on chromosome 18 in the combined sample (all), European Americans (EA), and African Americans (AA), after an additional 60 families were added. The 1-LOD CIs (∼20 cM) for combined samples and European Americans are labeled in the figure.
View largeDownload slide

Family regression quantitative linkage using the computer program MERLIN for log-transformed, adjusted fasting plasma glucose on chromosome 18 in the combined sample (all), European Americans (EA), and African Americans (AA), after an additional 60 families were added. The 1-LOD CIs (∼20 cM) for combined samples and European Americans are labeled in the figure.

Close modal
TABLE 1

Distribution of fasting glucose (original and log-transformed, adjusted) in the combined sample, European Americans, and African Americans

nMinimumMaximumMean ± SDSkewnessKurtosis
Combined sample       
    GLU 1,514 2.8 27.8 5.5 ± 2.2 4.35 26.41 
    GLU_LRES 1,497 −2.25 6.02 — 2.09 6.93 
European Americans       
    GLU 1,258 2.8 27.8 5.5 ± 2.1 4.35 27.82 
    GLU_LRES 1,255 −2.25 6.02 — 2.04 6.79 
African Americans       
    GLU 247 2.9 35.4 5.8 ± 3.4 4.91 30.98 
    GLU_LRES 236 −1.66 5.50 — 2.35 7.95 
nMinimumMaximumMean ± SDSkewnessKurtosis
Combined sample       
    GLU 1,514 2.8 27.8 5.5 ± 2.2 4.35 26.41 
    GLU_LRES 1,497 −2.25 6.02 — 2.09 6.93 
European Americans       
    GLU 1,258 2.8 27.8 5.5 ± 2.1 4.35 27.82 
    GLU_LRES 1,255 −2.25 6.02 — 2.04 6.79 
African Americans       
    GLU 247 2.9 35.4 5.8 ± 3.4 4.91 30.98 
    GLU_LRES 236 −1.66 5.50 — 2.35 7.95 

GLU, original fasting plasma glucose; GLU_LRES: log-transformed, adjusted fasting plasma glucose.

View Large
TABLE 2

Family regression (MERLIN_regress) and variance components (MERLIN_vc) analyses (chromosome 18) for log-transformed, adjusted fasting glucose

MERLIN_regress
MERLIN_vc
Location (cM)LODPLocation (cM)LODP
All samples       
 41 1.60 0.003 41 2.29 0.0006 
 80 2.14 0.0008 80 2.62 0.0003 
 116 3.67 0.00002 116 4.48 <0.00001 
European Americans       
 54 1.51 0.004 41 1.83 0.002 
 116 2.68 0.0002 116 3.02 0.00009 
African Americans       
 80 1.97 0.0013 99 2.21 0.0007 
 116 1.09 0.013 116 1.71 0.002 
MERLIN_regress
MERLIN_vc
Location (cM)LODPLocation (cM)LODP
All samples       
 41 1.60 0.003 41 2.29 0.0006 
 80 2.14 0.0008 80 2.62 0.0003 
 116 3.67 0.00002 116 4.48 <0.00001 
European Americans       
 54 1.51 0.004 41 1.83 0.002 
 116 2.68 0.0002 116 3.02 0.00009 
African Americans       
 80 1.97 0.0013 99 2.21 0.0007 
 116 1.09 0.013 116 1.71 0.002 

Diabetic subjects (type 1 and type 2 diabetes) with fasting plasma glucose <5.6 mmol/l were deleted. GLU_LRES, log-transformed, adjusted fasting plasma glucose.

View Large
TABLE 3

Linkages on 18q22-23 for fasting glucose after 60 new families were added

Subjects/phenotypesLODPLocation (cM)
Family regression    
    Combined sample*    
        Glucose† 4.59 <0.00001 116 
        Glucose_BMI‡ 2.69 0.0002 116 
    European Americans    
        Glucose 3.93 0.00001 116 
        Glucose_BMI 1.97 0.0013 116 
    African Americans    
        Glucose 1.28 0.008 89 
        Glucose_BMI 0.88 0.02 89 
Variance components    
    Combined sample    
        Glucose 6.59 <0.00001 116 
        Glucose_BMI 4.38 <0.00001 107 
    European Americans    
        Glucose 4.99 <0.00001 116 
        Glucose_BMI 3.33 0.00005 107 
    African Americans    
        Glucose 1.54 0.004 116 
        Glucose_BMI 1.19 0.01 116 
Diabetes as a covariate    
Combined sample    
        Glucose 5.09 <0.00001 116 
        Glucose_BMI 3.43 0.00004 107 
    European Americans    
        Glucose 3.51 0.00003 116 
        Glucose_BMI 2.32 0.0005 107 
    African Americans    
        Glucose 1.44 0.005 107 
        Glucose_BMI 1.08 0.013 116 
Subjects/phenotypesLODPLocation (cM)
Family regression    
    Combined sample*    
        Glucose† 4.59 <0.00001 116 
        Glucose_BMI‡ 2.69 0.0002 116 
    European Americans    
        Glucose 3.93 0.00001 116 
        Glucose_BMI 1.97 0.0013 116 
    African Americans    
        Glucose 1.28 0.008 89 
        Glucose_BMI 0.88 0.02 89 
Variance components    
    Combined sample    
        Glucose 6.59 <0.00001 116 
        Glucose_BMI 4.38 <0.00001 107 
    European Americans    
        Glucose 4.99 <0.00001 116 
        Glucose_BMI 3.33 0.00005 107 
    African Americans    
        Glucose 1.54 0.004 116 
        Glucose_BMI 1.19 0.01 116 
Diabetes as a covariate    
Combined sample    
        Glucose 5.09 <0.00001 116 
        Glucose_BMI 3.43 0.00004 107 
    European Americans    
        Glucose 3.51 0.00003 116 
        Glucose_BMI 2.32 0.0005 107 
    African Americans    
        Glucose 1.44 0.005 107 
        Glucose_BMI 1.08 0.013 116 
*

All samples, including self-reported diabetes, were included;

†

Glucose: log-transformed, adjusted fasting glucose was used in this study;

‡

Glucose_BMI: log-transformed fasting glucose adjusted by BMI and age within sex, race, and generation.

View Large

This work was supported by the National Institutes of Health (NIH R01DK44073, R01DK48095, and R01DK56210 to R.A.P.).

Except for five markers, genotypes were completed by the National Heart, Lung, and Blood Institute (NHLBI)-supported Marshfield Genotyping Service. We acknowledge the cooperation of our subjects. We also thank Quan Cao, Jeffrey Hannah, Dr. Balasahib Shinde, Elizabeth Joe, Jan Merideth, Cameron Braswell, and Kye Yun for technical assistance. We thank three anonymous reviewers for their comments and suggestions.

1
Sacks DB, Bruns DE, Goldstein DE, Maclaren NK, McDonald JM, Parrott M: Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus.
Clin Chem
 
48
:
436
–472,
2002
2
Reaven G, Abbasi F, McLaughlin T: Obesity, insulin resistance, and cardiovascular disease.
Recent Prog Horm Res
 
59
:
207
–223,
2004
3
Henkin L, Bergman RN, Bowden DW, Ellsworth DL, Haffner SM, Langefeld CD, Mitchell BD, Norris JM, Rewers M, Saad MF, Stamm E, Wagenknecht LE, Rich SS: Genetic epidemiology of insulin resistance and visceral adiposity: the IRAS Family Study design and methods.
Ann Epidemiol
 
13
:
211
–217,
2003
4
Freeman MS, Mansfield MW, Barrett JH, Grant PJ: Heritability of features of the insulin resistance syndrome in a community-based study of healthy families.
Diabet Med
 
19
:
994
–999,
2002
5
Meigs JB, Panhuysen CI, Myers RH, Wilson PW, Cupples LA: A genome-wide scan for loci linked to plasma levels of glucose and HbA1c in a community-based sample of Caucasian pedigrees: the Framingham Offspring Study.
Diabetes
 
51
:
833
–840,
2002
6
Watanabe RM, Ghosh S, Langefeld CD, Valle TT, Hauser ER, Magnuson VL, Mohlke KL, Silander K, Ally DS, Chines P, Blaschak-Harvan J, Douglas JA, Duren WL, Epstein MP, Fingerlin TE, Kaleta HS, Lange EM, Li C, McEachin RC, Stringham HM, Trager E, White PP, Balow Jr J, Birznieks G, Chang J, Eldridge W: The Finland-United States investigation of non-insulin-dependent diabetes mellitus genetics (FUSION) study. II. An autosomal genome scan for diabetes-related quantitative-trait loci.
Am J Hum Genet
 
67
:
1186
–1200,
2000
7
Bray MS, Boerwinkle E, Hanis CL: Linkage analysis of candidate obesity genes among the Mexican-American population of Starr County, Texas.
Genet Epidemiol
 
16
:
397
–411,
1999
8
Price RA, Reed DR, Lee JH: Obesity related phenotypes in families selected for extreme obesity and leanness.
Int J Obes Relat Metab Disord
 
22
:
406
–413,
1998
9
Sham PC, Purcell S, Cherny SS, Abecasis GR: Powerful regression-based quantitative-trait linkage analysis of general pedigrees.
Am J Hum Genet
 
71
:
238
–253,
2002
10
Holmes DI, Wahab NA, Mason RM: Cloning and characterization of ZNF236, a glucose-regulated Kruppel-like zinc-finger gene mapping to human chromosome 18q22–q23.
Genomics
 
60
:
105
–109,
1999
11
Walli R, Schafer H, Morys-Wortmann C, Paetzold G, Nustede R, Schmidt WE: Identification and biochemical characterization of the human brain galanin receptor.
J Mol Endocrinol
 
13
:
347
–356,
1994
12
Lorimer DD, Benya RV: Cloning and quantification of galanin-1 receptor expression by mucosal cells lining the human gastrointestinal tract.
Biochem Biophys Res Commun
 
222
:
379
–385,
1996
13
Li WD, Dong C, Li D, Zhao H, Price RA: An obesity-related locus in chromosome region 12q23-24.
Diabetes
 
53
:
812
–820,
2004
14
Imperatore G, Knowler WC, Nelson RG, Hanson RL: Genetics of diabetic nephropathy in the Pima Indians.
Curr Diab Rep
 
1
:
275
–281,
2001
15
Vardarli I, Baier LJ, Hanson RL, Akkoyun I, Fischer C, Rohmeiss P, Basci A, Bartram CR, Van Der Woude FJ, Janssen B: Gene for susceptibility to diabetic nephropathy in type 2 diabetes maps to 18q22.3-23.
Kidney Int
 
62
:
2176
–2183,
2002
16
Lindgren CM, Widen E, Tuomi T, Li H, Almgren P, Kanninen T, Melander O, Weng J, Lehto M, Groop LC: Contribution of known and unknown susceptibility genes to early-onset diabetes in Scandinavia: evidence for heterogeneity.
Diabetes
 
51
:
1609
–1617,
2002
17
Merriman TR, Eaves IA, Twells RC, Merriman ME, Danoy PA, Muxworthy CE, Hunter KM, Cox RD, Cucca F, McKinney PA, Shield JP, Baum JD, Tuomilehto J, Tuomilehto-Wolf E, Ionesco-Tirgoviste C, Joner G, Thorsby E, Undlien DE, Pociot F, Nerup J, Ronningen KS, Bain SC, Todd JA: Transmission of haplotypes of microsatellite markers rather than single marker alleles in the mapping of a putative type 1 diabetes susceptibility gene (IDDM6).
Hum Mol Genet
 
7
:
517
–524,
1998
18
Yeo GS, Farooqi IS, Aminian S, Halsall DJ, Stanhope RG, O’Rahilly S: A frameshift mutation in MC4R associated with dominantly inherited human obesity.
Nat Genet
 
20
:
111
–112,
1998
19
Vaisse C, Clement K, Guy-Grand B, Froguel P: A frameshift mutation in human MC4R is associated with a dominant form of obesity (Letter).
Nat Genet
 
20
:
113
–114,
1998
20
Hinney A, Schmidt A, Nottebom K, Heibult O, Becker I, Ziegler A, Gerber G, Sina M, Gorg T, Mayer H, Siegfried W, Fichter M, Remschmidt H, Hebebrand J: Several mutations in the melanocortin-4 receptor gene including a nonsense and a frameshift mutation associated with dominantly inherited obesity in humans.
J Clin Endocrinol Metab
 
84
:
1483
–1486,
1999
21
Rice T, Chagnon YC, Perusse L, Borecki IB, Ukkola O, Rankinen T, Gagnon J, Leon AS, Skinner JS, Wilmore JH, Bouchard C, Rao DC: A genome-wide linkage scan for abdominal subcutaneous and visceral fat in black and white families: the HERITAGE Family Study.
Diabetes
 
51
:
848
–855,
2002
22
Lahiri DK, Nurnberger JI Jr: A rapid non-enzymatic method for the preparation of HMW DNA from blood for RFLP studies.
Nucleic Acid Res
 
19
:
5444
,
1991
23
Abecasis GR, Cherny SS, Cookson WO, Cardon LR: Merlin: rapid analysis of dense genetic maps using sparse gene flow trees.
Nat Genet
 
30
:
97
–101,
2002
DIABETES
2004