Brief Genetics Report

Linkage of the Metabolic Syndrome to 1q23-q31 in

Hispanic Families
The Insulin Resistance Atherosclerosis Study Family Study
Carl D. Langefeld,1 Lynne E. Wagenknecht,1 Jerome I. Rotter,2 Adrienne H. Williams,1
John E. Hokanson,3 Mohammad F. Saad,4 Donald W. Bowden,5 Stephen Haffner,6 Jill M. Norris,3
Stephen S. Rich,1 and Braxton D. Mitchell7

The metabolic syndrome is characterized by central

obesity, dyslipidemia, elevated blood pressure, and hyperglycemia. The Insulin Resistance Atherosclerosis
Study (IRAS) Family Study recruited extended pedigrees of Hispanic descent from San Antonio, TX (SA)
and San Luis Valley, CO (SLV). Thirty-five of these
pedigrees (27 SA and 8 SLV) had at least 2 individuals
with metabolic syndrome (216 affected individuals and
563 affected relative pairs). The prevalence of metabolic syndrome and component criteria in subjects from
these pedigrees were 35% metabolic syndrome, 43%
increased waist circumference, 31% hypertriglyceridemia, 69% low HDL cholesterol, 31% increased blood
pressure, and 25% either increased fasting glucose or
presence of diabetes. Nonparametric linkage analysis
provided evidence for linkage of metabolic syndrome to
1q23-q31 (D1S518; logarithm of odds [LOD] 1.6) with
significant site heterogeneity (SA LOD 2.6 and SLV LOD
0.0), and removing all individuals with diabetes reduced, but did not eliminate, the evidence for linkage to
this region (LOD 1.2). This heterogeneity may partially
be explained by phenotypic differences. Members in the
SA pedigrees were older, had greater central obesity,
had higher prevalence of the metabolic syndrome, and
were from a more urban environment than members of
the SLV pedigrees. These results contribute to the
growing evidence that chromosome 1q harbors at least
one locus related to the metabolic precursors of diabetes. Diabetes 53:1170 1174, 2004

From the 1Department of Public Health Sciences, Wake Forest University

Health Sciences, Winston-Salem, North Carolina; the 2Division of Medical
Genetics, Cedars-Sinai Medical Center, Los Angeles, California; the 3Department of Preventive Medicine, University of Colorado Health Sciences, Denver,
Colorado; the 4Division of Endocrinology, University of California, Los
Angeles, California; the 5Department of Biochemistry, Wake Forest University
Health Sciences, Winston-Salem, North Carolina; the 6Division of Clinical
Epidemiology, Department of Medicine, University of Texas Health Science
Center, San Antonio, Texas; and the 7Division of Endocrinology, Diabetes, and
Nutrition, University of Maryland School of Medicine, Baltimore, Maryland.
Address correspondence and reprint requests to Carl D. Langefeld, PhD,
Department of Public Health Sciences, Wake Forest University School of
Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1063. E-mail:
clangefe@wfubmc.edu.
Received for publication 16 September 2003 and accepted in revised form
14 January 2004.
ARP, affected relative pair; IRAS, Insulin Resistance Atherosclerosis Study;
LOD, logarithm of odds; NPL, nonparametric linkage.
2004 by the American Diabetes Association.
1170

he metabolic syndrome is a clinical entity consisting of central obesity, dyslipidemia, elevated

blood pressure, and hyperglycemia, and its presence is associated with an increased risk for type
2 diabetes and cardiovascular disease (1). A major etiologic factor underlying this syndrome is insulin resistance
(2). The metabolic syndrome was defined in the Third
Report of the National Cholesterol Education Program
Expert Panel Detection, Evaluation, and Treatment in
Adults (Adult Treatment Panel III) as the presence of three
or more of the following risk factors: waist circumference
88 cm in women and 102 in men, triglycerides 150
mg/dl, HDL cholesterol 50 mg/dl in women and 40 in
men, blood pressure 130/85 mmHg, and fasting glucose
110 mg/dl or a diagnosis of diabetes (1). Recent estimates suggest that the adult prevalence of the metabolic
syndrome is 32% in Hispanic Americans, 22% in African
Americans, and 24% in Caucasian Americans (3).
It is believed that there is substantial overlap in the
environmental (e.g., sedentary lifestyle and excessive caloric intake) and genetic risk factors associated with the
development of metabolic syndrome and those factors
responsible for the development of type 2 diabetes (1).
While efforts to localize genes for the metabolic syndrome
per se are in their infancy, genome-wide scans for type 2
diabetes have been conducted in a number of populations.
A growing number of studies (4 10) have reported linkage
of diabetes or hyperglycemia to a region on human chromosome 1q21-q25. Further support for a chromosome 1q
diabetes susceptibility locus derives from the mapping of
the gk2 diabetes susceptibility locus in the Goto-Kakizaki
rat to a syntenic region (11).
In this report, the Insulin Resistance Atherosclerosis
Study (IRAS) Family Study provides evidence that the
metabolic syndrome links to chromosome 1q in 35 Hispanic pedigrees (216 individuals with the metabolic syndrome) ascertained from San Antonio, TX (SA), and San
Luis Valley, CO (SLV) (12). The IRAS Family Study is a
multicenter study designed to identify genes predisposing
to insulin resistance, adiposity, and related traits (12).
Sample summary statistics, including the numbers of
DIABETES, VOL. 53, APRIL 2004

C.D. LANGEFELD AND ASSOCIATES

TABLE 1
Characteristics of the 216 individuals meeting the criteria of at least three elements of the metabolic syndrome
Characteristic
Pedigrees (n)
Metabolic syndrome (n)
ARPs (n)
Full sibling
Avuncular
Cousin
Other
Total affected pairs
Age (years)
Sex (% female)
BMI
Women
Men
Diabetes (% in family)
Metabolic syndrome (% in family)
Risk factors for the metabolic syndrome
Waist circumference
Women
Men
Fasting plasma glucose (mg/dl)
Elevated blood pressure (mmHg)
Systolic
Diastolic
Serum triglycerides (mg/dl)
HDL cholesterol (mg/dl)
Women
Men

SA

SLV

Combined

27
159

8
57

97
149
94
40
380
50.6 13.2 (50)
57.9

41
83
58
1
183
45.4 13.4 (44)
56.1

138
232
152
41
563
49.2 13.4 (48)
57.4

34.9 6.6 (34.4)

33.8 5.0 (33.9)
38.2
38.6

33.4 5.8 (34.3)

31.3 4.4 (32.6)
33.3
27.9

34.5 6.4 (34.3)

33.2 4.9 (33.1)
36.9
35.1

100.9 12.7 (98.7)

108.5 10.4 (108.1)
122.8 39.2 (110.3)

98.8 12.0 (99.2)

101.0 8.5 (103.0)
132.4 55.6 (112.5)

100.4 12.5 (98.8)

106.5 10.4 (106.7)
125.4 44.3 (111.5)

128.3 16.5 (128.5)

78.3 10.6 (79.0)
201.4 116.0 (181)

125.4 16.3 (123)

83.5 9.0 (84)
206.5 106.1 (182.5)

127.5 16.4 (126)

79.7 10.4 (80)
202.7 113.3 (181)

39.4 10.7 (38)

31.8 6.5 (31)

35
216

37.4 8.0 (37)

30.6 4.8 (32)

38.9 10.1 (38)

31.5 6.1 (32)

Data are means SD (median) unless noted otherwise.

affected relative pairs (ARPs), are shown in Table 1. The

overall prevalence of metabolic syndrome in these families
was 35%, and the proportion of subjects having risk factor
values meeting Adult Treatment Panel III criteria for the
definition of metabolic syndrome was 43% for waist circumference, 31% for triglycerides, 69% for HDL choles-

terol, 31% for blood pressure, and 25% for fasting glucose
or diabetes.
Nonparametric linkage (NPL) analysis provided strong
evidence for linkage of the metabolic syndrome to chromosome 1q between markers D1S1589 and D1S518 (Table
2) (Fig. 1). In the SA pedigrees, the evidence for linkage

TABLE 2
Summary of the NPL analyses on chromosome 1

Metabolic syndrome
Combined Hispanic
SA
SLV
Metabolic syndrome removing individuals with
only two risk factors plus diabetes
Combined Hispanic
SA
SLV
Metabolic syndrome removing all diabetic subjects
Combined Hispanic
SA
SLV
Diabetes
Combined Hispanic
SA
SLV
DIABETES, VOL. 53, APRIL 2004

Position (cM)

LOD

LOD-1 interval

P value

Flanking markers

140.5
205.0
139.5
199.0
286.5

1.32
1.58
1.16
2.59
0.56

114.5222.5
131.0220.0
58.5223.0
180.5209.0
207.0

0.0138
0.0070
0.0207
0.0006
0.1078

ATA42G12/
D1S518/
ATA42G12/
D1S1589/D1S518
/TTTA049

146.5
204.0
150.0
200.0
282.0

1.31
1.49
1.03
2.23
0.67

129.5220.5
131.5218.0
52.5221.0
179.5208.0
204.5

0.0139
0.0089
0.0294
0.0013
0.0789

/D1S534
D1S518/
/D1S534
D1S518/
/TTTA049

205.0
202.0
139.0

1.23
1.88
0.43

131.5224.0
193.5212.5
131.5151.0

0.0172
0.0033
0.1599

D1S518/
D1S518/
/D1S1627

211.0
210.5
216.5

1.76
0.86
1.50

190.5220.5
167.5232.0
192.0221.5

0.0045
0.0469
0.0085

/D1S1660
/D1S1660
D1S1660/
1171

FIG. 1. Result of NPL analysis on chromosome 1 for four phenotypes: metabolic syndrome (A), metabolic syndrome removing individuals who had only two of the
five criteria plus a diagnosis of diabetes (B), metabolic syndrome removing all individuals with a diagnosis of diabetes regardless of the number of other criteria
met (C), and type 2 diabetes (D).

METABOLIC SYNDROME LINKS TO 1q23-q31

1172

DIABETES, VOL. 53, APRIL 2004

C.D. LANGEFELD AND ASSOCIATES

maximized (logarithm of odds [LOD] 2.59) at 1q31 near

D1S1589/D1S518 (199 cM). The linkage support interval
defined by the region in which the LOD score is 1.59
(LOD-1 interval) was 28.5 cM in the SA pedigrees and
was bounded by the nearest markers, TATC028 (1q23) and
D1S1660 (1q31). There was no evidence of linkage to this
region within the SLV pedigrees (LOD 0.0). Combining the
two Hispanic pedigree collections yielded LOD 1.58 near
D1S1589/D1S518 (Table 2).
To determine whether the evidence for linkage to 1q
might reflect linkage to diabetes, three additional analyses
were performed (Table 2). First, removing individuals
having diabetes and only two metabolic syndrome risk
factors from the analysis yielded a reduced sample containing 415 ARPs from the same 35 pedigrees. In the
combined collection of Hispanic pedigrees, there was
modest evidence for linkage to 1p12 near D1S534 (147 cM,
LOD 1.31) and 1q31 near D1S518 (204 cM, LOD 1.49).
These results were dominated by the SA pedigrees since
the SLV pedigrees provided very modest evidence for
linkage near the D1S534 locus (LOD 0.17) and no evidence
for linkage near the D1S518 locus (LOD 0.0).
In the second ancillary analysis, all individuals with a
diagnosis of diabetes were removed, providing a total of
231 ARPs from 31 pedigrees for analysis (Table 2). The
combined Hispanic sample provided modest evidence for
linkage to a single region near D1S518 (1q31, 205 cM, LOD
1.23). The SLV pedigrees did not contribute evidence for
linkage at this locus (LOD 0.0).
In the third ancillary analysis, only pedigrees with at
least two individuals diagnosed with type 2 diabetes were
included (30 pedigrees containing 139 ARPs) and a linkage
analysis for type 2 diabetes was performed. In both the SA
and SLV pedigrees, the strongest evidence for linkage to
type 2 diabetes was near D1S1660 (SA LOD 0.86, 211 cM
and SLV LOD 1.50, 217 cM), providing a maximum LOD
score of 1.76 near D1S1660 (211 cM, 1q31). The LOD-1
support interval spanned from 191 (D1S1589, 1q25) to 221
cM (D1S1663, 1q32).
As an exploratory analysis, a series of ordered subset
analyses were computed to investigate whether any of the
five specific risk factors influenced the magnitude of the
evidence for linkage within the 1q region. Subsetting on the
levels of the five individual risk factors defining the metabolic
syndrome did not yield a statistically significant increase in
the evidence for linkage within any of the three combinations
of pedigrees (i.e., SA, SLV, or combined Hispanic).
There are several possible factors that could contribute
to the stronger evidence for linkage detected in the 27 SA
Hispanic pedigrees collected relative to the 8 SLV Hispanic
pedigrees. One might posit that the primary difference in
the linkage results reflects a difference in the statistical
power of the respective samples. However, there are also
substantive differences in the characteristics of these two
samples (Table 1). First, the median age of individuals
from the SA pedigrees was 6 years older than that from the
SLV pedigrees (P 0.0153). Second, men from San
Antonio tended to have a greater BMI (P 0.0323) and
waist circumference (P 0.0005) than men from San Luis
Valley; a similar but not statistically significant difference
was observed among women. Third, diastolic blood pressure tended to be higher in the SLV pedigrees (P 0.0007).
DIABETES, VOL. 53, APRIL 2004

Fourth, San Antonio is a more urban environment relative

to San Luis Valley. In summary, the SA pedigrees tended to
be older, have greater central adiposity, have a higher
prevalence of the metabolic syndrome, and be from a
more urban environment.
These results contribute to the growing evidence that
chromosome 1q harbors a diabetes-related susceptibility
locus. Evidence for linkage in this population does not
appear to be restricted to subjects with diabetes alone,
thus suggesting that the putative disease-predisposing loci
may influence susceptibility to a broader spectrum of
metabolic disorders than just diabetes. Because of the
relatively high prevalence of the metabolic syndrome in
the population, even a modest increase in risk associated
with the putative mutation will be associated with a
relatively large burden of disease in the population.
Linkage of diabetes to the 1q21-25 regions has previously been reported (4 10) in European-Caucasian, Amerindian, and Chinese populations. The current results
suggest that the putative mutation(s) may also be present
in Hispanic Americans. Although the position that maximizes the LOD score for metabolic syndrome in our
sample is the most telomeric of these seven studies, it is
within the LOD-1 support interval of the majority of these
studies. Conversely, our LOD-1 support interval contains
the loci that maximize the evidence for linkage in the
majority of these studies. The apparent presence of linkage across such diverse populations is striking and is
consistent with an ancient source of the mutation.
RESEARCH DESIGN AND METHODS
Study design and data collection. The IRAS Family Study was designed to
map genes that predispose to insulin resistance, cardiovascular disease, and
obesity. Details of the study design, patient recruitment, and phenotyping are
described in detail elsewhere (12). Briefly, probands of Hispanic-American
heritage who self-reported a large family structure were recruited from San
Antonio, TX, and San Luis Valley, CO. Probands were identified from the
original IRAS cohort study (13), which was designed to recruit approximately
equal numbers of diabetic, impaired glucose tolerant, and normal individuals.
To meet recruitment goals in the IRAS Family Study, additional probands
were recruited from the general population at the San Antonio clinic. From
this nonrandom collection of probands, ascertainment of families was based
on large family structure and not on any clinical characteristics related to
elements of the metabolic syndrome, diabetes, cardiovascular disease, or
obesity. The target family structure criteria included DNA collectible on at
least 12 and full phenotyping possible on at least 9 family members. We did not
specifically screen for maturity-onset diabetes of the youngassociated mutations or GAD antibodies for type 1 diabetes. However, our sample contained
only 22 insulin-taking diabetic subjects; of these, all had an age of diabetes
onset 30 years and all but 1 had a BMI 27 kg/m2.
Medical history, health behaviors, and demographic data were collected by
interview. The medical history interview focused on the assessment of current
health status and clinical conditions, particularly self-reported type 2 diabetes
and its complications, hypertension, and cardiovascular disease events and
procedures. At the time of the clinic visit, a fasting blood draw was obtained to
determine plasma glucose, insulin, and lipid levels. Diabetes was defined using
the American Diabetes Association criteria. The systolic and diastolic blood
pressures analyzed represent the average of three measures, respectively. BMI (in
kilograms per square meter) was obtained from height and weight measurements.
The five individual risk factors defining the metabolic syndrome were assessed,
and the individuals metabolic syndrome status was determined independent of medications used. Linkage results were qualitatively similar when we
repeated the analysis using a slightly expanded definition that included antihypertensive or antilipid medications as affected for that trait.
Whole blood obtained from each IRAS Family Study participant was frozen
and shipped to the Molecular Genetics Laboratory at Wake Forest University
School of Medicine. DNA from these 35 families, who completed a 10-cM
genome scan (marker set 11), was shipped to the Mammalian Genotyping
Center in Marshfield, WI (http://research.marshfieldclinic.org/genetics). Chro1173

METABOLIC SYNDROME LINKS TO 1q23-q31

mosomal maps were constructed using the marker order and distances
available from the Mammalian Genotyping Center.
Statistical methods
Preliminary analyses. Center-specific maximum likelihood estimates of
allele frequencies were computed using the Recode software (D. Weeks,
personal communication). Each pedigree was examined for potentially incorrectly self-reported familial relationships using the entire genome scan data
(383 markers) and the Prest software (14). Inconsistent relationships were
modified when the data suggested a clear alternative (e.g., full sibling to half
sibling) or the genotype data were converted to missing. Each marker was
examined for Mendelian inconsistencies using the Pedcheck software (15),
and probable genotyping errors were converted to missing.
Linkage analysis. Multipoint NPL regression analysis using the NPLpairs
statistic was computed for the markers on chromosome 1. The NPL regression
approach is a conditional logistic regression analysis in which the familyspecific NPL statistics at one or more loci are the predictor variables (16).
Analyses were repeated using the exponential allele-sharing model (17).
Linkage analyses based on ARPs were selected to minimize the influence of
misclassification errors (e.g., unaffected subjects who develop metabolic
syndrome in the future).
A series of ordered subset analyses (18) were computed to investigate the
influence of each of the metabolic syndrome risk factors on the evidence for
linkage. First, the mean of each risk factor was calculated for each pedigree
and ranked from smallest to largest. For a specific risk factor (e.g., triglycerides), the family with the largest risk factor mean was entered into the analysis
and the corresponding LOD score was computed on chromosome 1 for that
family. The ith ordered subset analysis proceeds by computing a linkage
analysis on chromosome 1 using the subset of families with the ith largest risk
factor means. This process is repeated until all families have been added to the
linkage analysis. The subset of families that yield the largest LOD score on
chromosome 1 is taken as the LOD score of interest. The statistical significance of the change in the LOD score was computed by a permutation test
under the null hypothesis that the familys ranking for a specific risk factor is
independent of their evidence for linkage, resulting in a chromosome-wide P
value. For each risk factor, the ordered subset analysis was repeated, ranking
families from smallest to largest.

REFERENCES
1. National Institutes of Health: Third Report of the National Cholesterol
Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III).
Bethesda, MD, National Institutes of Health, 2001 (NIH publ. no. 01-3670)
2. Grundy SM: Hypertriglyceridemia, insulin resistance, and the metabolic
syndrome. Am J Cardiol 83:25F29F, 1999
3. Ford ES, Giles WH, Dietz WH: Prevalence of the metabolic syndrome
among US adults: findings from the Third National Health and Nutrition
Examination Survey. JAMA 287:356 359, 2002
4. Hanson RL, Ehm MG, Pettitt DJ, Prochazka M, Thompson DB, Timberlake
D, Foroud T, Kobes S, Baier L, Burns DK, Almasy L, Blangero J, Garvey
WT, Bennett PH, Knowler WC: An autosomal genomic scan for loci linked
to type II diabetes mellitus and body-mass index in Pima Indians. Am J
Hum Genet 63:1130 1138, 1998
5. Elbein SC, Hoffman MD, Teng K, Leppert MF, Hasstedt SJ: A genome-wide

1174

search for type 2 diabetes susceptibility genes in Utah Caucasians.

Diabetes 48:11751182, 1999
6. Vionnet N, Hani E, Dupont S, Gallina S, Francke S, Dotte S, De Matos F,
Durand E, Lepretre F, Lecoeur C, Gallina P, Zekiri L, Dina C, Froguel P:
Genomewide search for type 2 diabetes-susceptibility genes in French
whites: evidence for a novel susceptibility locus for early-onset diabetes on
chromosome 3q27-qter and independent replication of a type 2-diabetes
locus on chromosome 1q21 q24. Am J Hum Genet 67:1470 1480, 2000
7. Wiltshire S, Hattersley AT, Hitman GA, Walker M, Levy JC, Sampson M,
ORahilly S, Frayling TM, Bell JI, Lathrop GM, Bennett A, Dhillon R,
Fletcher C, Groves CJ, Jones E, Prestwich P, Simecek N, Rao PV, Wishart
M, Foxon R, Howell S, Smedley D, Cardon LR, Menzel S, McCarthy MI: A
genomewide scan for loci predisposing to type 2 diabetes in a U.K.
population (the Diabetes U.K. Warren 2 Repository): analysis of 573
pedigrees provides independent replication of a susceptibility locus on
chromosome 1q. Am J Hum Genet 69:553569, 2001
8. Hsueh W-C, St Jean PL, Mitchell BD, Pollin TI, Knowler WC, Ehm MG, Bell CJ,
Sakul H, Wagner MJ, Burns DK, Shuldiner AR: Genome-wide and finemapping linkage studies of type 2 diabetes and glucose traits in the Old Order
Amish: evidence for a new diabetes locus on chromosome 14q11 and
confirmation of a locus on chromosome 1q21q24. Diabetes 52:550557, 2003
9. Meigs JB, Panhuysen CI, Myers RH, Wilson PW, Cupples LA: A genomewide 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
10. Xiang K, Wang Y, Zheng T, Shen K, Jia W, Li J, Lin X, Wu S, Zhang G, Wang
S, Lu H: Genome-wide scan search for type 2 diabetes susceptibility loci in
Chinese (Abstract). Diabetes 51 (Suppl. 2):A262, 2002
11. Gauguier D, Froguel P, Parent V, Bernard C, Bihoreau MT, Portha B, James
MR, Penicaud L, Lathrop M, Ktorza A: Chromosomal mapping of genetic
loci associated with non-insulin dependent diabetes in the GK rat. Nat
Genet 12:38 43, 1996
12. Henkin L, Bergman RN, Bowden DW, Ellsworth DL, Haffner SM, Mitchell
BD, Norris JM.,: Rewers M, Saad MF, Wagenknecht LE, Rich SS: Genetic
epidemiology of insulin resistance and visceral adiposity: the Family IRAS
Study design and methods. Ann Epidemiol 13:211217, 2002
13. Wagenknecht LE, Mayer EJ, Rewers M, Haffner S, Selby J, Borok GM,
Henkin L, Howard G, Savage PJ, Saad MF: The Insulin Resistance
Atherosclerosis Study (IRAS) objectives, design, and recruitment results.
Ann Epidemiol 5:464 472, 1995
14. McPeek MS, Sun L: Statistical tests for detection of misspecified relationships by use of genome scan data. Am J Hum Genet 66:1070 1094, 2000
15. OConnell JR, Weeks DE: PedCheck: a program for identification of
genotype incompatibilities in linkage analysis. Am J Hum Genet 63:259
266, 1998
16. Langefeld CD, Davis CC, Brown WM: Nonparametric linkage regression. I:
combined European-American CSGA and German genome scans for
asthma. Genet Epid 21 (Suppl. 1):S136 S141, 2001
17. Kong A, Cox NJ: Allele-sharing models: LOD scores and accurate linkage
tests. Am J Hum Genet 61:1179 1188, 1997
18. Hauser ER, Watanabe RM, Duren WL, Langefeld CD, Boehnke M: Ordered
subset analysis in genetic linkage mapping of complex traits. Genet Epid.
In press

DIABETES, VOL. 53, APRIL 2004