Page 1
en
n
uil
Salvador Ram�ırez-Jim�enez,b Maribel Rodr�ıguez-Torres,b Mois�es Aur�on-G�omez,b
Variation in the calpain-10 gene (CAPN10) has been associated with risk of type 2 diabetes in the Mexican American population
(band 2q37.3) [1]. This locus was estimated to have a
sibling recurrence risk (ks) of 1.37 (l-lod confidence in-
terval for ks, 1.13–1.74) for an additive model and 1.36
(1.14–1.67) for a non-additive model. It could account
pressed cysteine protease calpain-10, a finding that
identified a novel biochemical pathway involved in the
regulation of blood glucose levels [2]. The association of
variants in the calpain-10 gene (CAPN10)1 with risk of
1
bolis*of Starr County, Texas. We typed five polymorphisms in the calpain-10 gene (SNP-43, -43, -63, and -110 and Indel-19) to test for
association with type 2 diabetes in 248 individuals representative of the mestizo population of Mexico City and Orizaba, Mexico
including 134 patients with type 2 diabetes and 114 subjects with normal fasting blood glucose levels. We found a significant dif-
ference in SNP-44 allele and genotype frequencies between type 2 diabetic and non-diabetic subjects. The rare allele at SNP-44 was
associated with increased risk of type 2 diabetes (odds ratio (OR)¼ 2.72, 95% confidence interval (CI)¼ 1.16–6.35, P ¼ 0:017). SNP-
110, which is in perfect linkage disequilibrium with SNP-44, was also associated with type 2 diabetes. The SNP-43, Indel-19, and
SNP-63 haplogenotype 112/121 associated with significantly increased risk (OR¼ 2.16, 95% CI¼ 1.31–3.57) of type 2 diabetes in
Mexican Americans was not associated with significantly increased in risk in Mexicans (OR¼ 1.15, 95% CI¼ 0.57–2.34). The results
suggest that variation in CAPN10 affects risk of type 2 diabetes in the mestizo population of central Mexico (Mexico City and
Orizaba) and in Mexican Americans (Starr County, Texas).
� 2003 Elsevier Inc. All rights reserved.
Introduction
A genome-wide search for type 2 diabetes-suscepti-
bility genes in Mexican Americans from Starr County,
Texas localized a gene affecting risk, designated
NIDDM1, to the distal long arm of chromosome 2
for 21–30% of the familial clustering of type 2 diabetes
in this Mexican American population depending on the
model for the interaction between NIDDM1 and other
susceptibility loci, additive or multiplicative, respec-
tively. Subsequent studies suggested that NIDDM1 was
allelic with the gene encoding the ubiquitously ex-Erika Ram�ırez,b Mar�ıa Luisa Velasco-P�erez,b Alfredo Ram�ırez-Silva,c
Francisco G�omez-P�erez,b Craig L. Hanis,d Takafumi Tsuchiya,a Issei Yoshiuchi,a
Nancy J. Cox,a and Graeme I. Bella
a Departments of Biochemistry and Molecular Biology, Medicine and Human Genetics, The University of Chicago, 5841 S. Maryland Ave.,
MC1028, Chicago, IL 60637, USA
b Departamento de Endocrinolg�ıa y Metabolismo de L�ıpidos, and Unidad de Biolog�ıa Molecular y Medicina Gen�omica,
Instituto Nacional de Ciencias M�edicas y Nutrici�on ‘‘Salvador Zubir�an,’’ Universidad Nacional Aut�onoma de M�exico, M�exico City, Mexico
c Instituto Mexicano del Seguro Social, Ciudad Valles, San Luis Potos�ı, Mexico
d Human Genetics Center, The University of Texas Health Science Center at Houston, Houston, TX, USA
Received 8 September 2003; received in revised form 9 October 2003; accepted 9 October 2003
AbstractAssociation of the calpain-10 g
in a Mexica
Laura del Bosque-Plata,a,* Carlos A. Ag
Molecular Genetics and MetaCorresponding author. Fax: 1-773-702-9237.
E-mail address: ldelbosq@uchicago.edu (L. del Bosque-Plata).
1096-7192/$ - see front matter � 2003 Elsevier Inc. All rights reserved.
doi:10.1016/j.ymgme.2003.10.005e with type 2 diabetes mellitus
population
ar-Salinas,b Mar�ıa Teresa Tusi�e-Luna,c
m 81 (2004) 122–126
www.elsevier.com/locate/ymgmeAbbreviations used: CAPN10, calpain-10 gene; CI, confidence
interval; OR, odds ratio; SNP, single-nucleotide polymorphism.
Page 2
type 2 diabetes in Mexican Americans (an admixed
population primarily of European and Native Ameri-
can ancestry) especially the high-risk single-nucleotide
polymorphisms (SNP)-43, -Indel19, and SNP-63 hap-
logenotype called 112/121 has not been consistently
replicated in other populations [3,4]. However, another
variant SNP-44 has been associated with type 2 dia-
betes in individuals of British ancestry as well as in
Mexican Americans [3]. Here, we have tested five
polymorphisms in the calpain-10 previously associated
with type 2 diabetes in other studies for association
with type 2 diabetes in the mestizo population of
central Mexico.
Materials and methods
Subjects
The study population consisted of 134 subjects with
Genotyping and haplotype assignment
We genotyped five polymorphisms in CAPN10 as de-
scribed previously [2]: SNP-44 (g.4841T>C), SNP-43
(g.4852G>A), Indel-19 (g.7920in/del32bp), SNP-63
(g.16378C>T), and SNP-110 (g.9803A>G) which is a
missense mutation T504A in perfect linkage disequilib-
rium with the diabetes-associated C allele at SNP-44.
Alleles were designated as described inHorikawa et al. [2].
Previous studies have shown that the haplotype-tagging
polymorphisms SNP-43, Indel-19, and SNP-63 lead to
four haplotypes described as 111, 112, 121, and 221 (allele
1 or 2 at SNP-43, Indel-19, andSNP-63, respectively). The
haplogenotypes were assigned by inspection of the ge-
notypes at SNP-43, Indel-19, and SNP-63.
Statistical analyses
The genotypic distributions were tested for deviation
from Hardy–Weinberg equilibrium and genotypes and
by v2 tests of independence using the program CON-
C
n (M/F) 116 (45/71) 18 (11/7) 1
Age-at-study (years) 56.2� 9.4 48.6� 10.2 5
2
ifican
femal
by d
rable.
L. del Bosque-Plata et al. / Molecular Genetics and Metabolism 81 (2004) 122–126 123BMI (kg/m2) 27.0� 3.4 29.4� 6.5
Age-at-diagnosis (years) 44.8� 7.5 NA
Fasting plasma glucose at time
of diagnosis or study (mg/dl)
285.6� 89.0 NA
Fasting glucose after treatment
(mg/dl)
154.5� 51.8 168.1� 76.9
Treatment (n)
Diet 22 NA
OHA 69 NA
Insulin 24 NA
Data are means�SD.
In the type 2 diabetic subjects recruited in Mexico City, there is a sign
(females, 46.1� 7.2 years; males, 42.8� 7.6 years; P ¼ 0:03) and BMI (
levels in the samples from Mexico City and Orizaba were determined
(Accutrend GC, Roche), respectively) and thus are not directly compatype 2 diabetes and 114 with normal fasting blood glu-
cose (80–110mg/dl). All subjects were of mixed Euro-
pean and Native American ancestry (mestizo) based on
self report. The type 2 diabetic subjects were recruited
from the outpatient clinic of Instituto Nacional de
Ciencias M�edicas y Nutrici�on ‘‘Salvador Zubir�an,’’
Mexico City (n ¼ 116) and at the Comisi�on Nacional de
Electricidad in Orizaba (n ¼ 18). The non-diabetic
control subjects were recruited from the outpatient
clinics of Instituto Nacional de Ciencias M�edicas y
Nutrici�on ‘‘Salvador Zubir�an,’’ Mexico City (n ¼ 58)
and at the Comisi�on Federal de Electricidad in Orizaba
(n ¼ 56). Written informed consent was obtained from
all participants.
Table 1
Clinical characteristics of subjects
Type 2 diabetes
Mexico City OrizabaAbbreviations: BMI, body mass index; OHA, oral hypoglycemic agent; NTING (Linkage Utility Programs; http://linkage.rocke-
feller.edu/ott/linkutil.htm) and all P values are two-sided.
Results
We genotyped five polymorphisms, SNP-43, -44, -63,
and -110 and Indel-19, in groups of type 2 diabetic and
Non-diabetic
ombined Mexico City Orizaba Combined
34 (56/78) 58 (17/41) 56 (40/16) 114 (57/57)
4.0� 9.8 57.9� 12.4 42.1� 11.9 50.0� 14.5
7.4� 4.1 26.1� 3.7 27.1� 4.3 26.7� 4.1
83.5� 18.3 87.6� 8.3
t difference between males and females with respect to age-at-diagnosis
es, 27.6� 3.7; males, 26.2� 2.6; P ¼ 0:03). Note that the blood glucose
ifferent methods (glucose oxidase method (Beckman) and glucometeralleles were tested for association with type 2 diabetes
using the online resource at the Institute for Human Ge-
netics, Munich, Germany (http://ihg.gsf.de/cgi-bin/hw/
hwa1.pl). Differences in allele, genotype, haplotype, and
haplogenotype frequencies between groups were assessedA, not available.
Page 3
non-diabetic mestizo subjects recruited in Mexico City
and in Orizaba, a city 280 km east of Mexico City. The
clinical features of the study groups are described in
Table 1. There was a significant difference in genotype
and allele frequencies for SNP-44 and -110 (these poly-
morphisms are in perfect linkage disequilibrium) be-
tween the type 2 diabetic and non-diabetic groups
(Table 2). Individuals who inherit the rare allele at these
polymorphisms have an increased risk of developing
type 2 diabetes (odds ratio (OR) for genotypes with the
C-allele at SNP-44¼ 2.72 (95% confidence interval
(CI)¼ 1.16–6.35, P ¼ 0:017). The SNP-44 T/C and C/C
genotypes have also been associated with increased risk
type 2 diabetes in Mexican Americans from Starr
County, Texas (OR¼ 1.83, 95% CI¼ 0.90–3.72,
P ¼ 0:09) and in white subjects of British/Irish ancestry
in the United Kingdom (OR¼ 1.25, 95% CI¼ 1.01–1.56,
P ¼ 0:041) [2,3].
There were no significant differences in SNP-43, In-
del-19, and SNP-63 haplotype or haplogenotype fre-
quencies between Mexican subjects with type 2 diabetes
and non-diabetic controls (Tables 3 and 4). The 112/121
haplogenotype associated with significantly increased
risk in Mexican Americans (OR¼ 2.16, 95% CI¼ 1.31–
3.57) was not associated with significantly increased risk
in Mexicans (OR¼ 1.15, 95% CI¼ 0.57–2.34).
Table 2
Genotype and allele frequencies of calpain-10 gene polymorphisms in Mexicans
Polymorphism Genotype Type 2 diabetes ðnÞ Non-diabetic ðnÞ P
SNP-44 T/T 111 105 0.017
T/C 23 8
C/C 0 0
Allele frequency ðnÞ T: 0.914 (245) T: 0.965 (218) 0.021
C: 0.086 (23) C: 0.035 (8)
SNP-43 G/G 61 48 0.672
G/A 59 56
A/A 12 8
Allele frequency ðnÞ G: 0.688 (181) G: 0.679 (152) 0.868
A: 0.314 (83) A: 0.321 (72)
SNP-19 2R/2R 17 12 0.835
2R/3R 69 62
3R/3R 46 38
Allele frequency ðnÞ 2R: 0.390 (103) 2R: 0.384 (86) 0.888
3R: 0.610 (161) 3R: 0.616 (138)
SNP-63 C/C 78 59 0.447
C/T 48 50
T/T 6 4
Allele frequency ðnÞ C: 0.773 (204) C: 0.743 (168) 0.449
T: 0.227 (60) T: 0.257 (58)
SNP-110 (T504A) A/A 107 102 0.044
A/G 20 8
G/G 0 0
Allele frequency ðnÞ A: 0.921 (234) A: 0.964 (212) 0.051
G: 0.079 (20) G: 0.036 (8)
both
ts (R
non-d
ined
40)
62)
76)
80)
ifican
124 L. del Bosque-Plata et al. / Molecular Genetics and Metabolism 81 (2004) 122–126All of the polymorphisms are in Hardy–Weinberg equilibrium in
successfully typed in all samples. SNP-44/T, SNP-43/G, Indel-19/2 repea
as described in [2].
Table 3
SNP-43, -19, and -63 haplotype frequencies ðnÞ in type 2 diabetic and
Haplotype Type 2 diabetes
Mexico City Orizaba Comb
111 0.15 (33) 0.21 (7) 0.16 (
121 0.24 (53) 0.26 (9) 0.24 (
112 0.30 (68) 0.24 (8) 0.29 (
221 0.31 (70) 0.29 (10) 0.31 (
Haplotype numbers were obtained by counting. There are no sign
diabetic or non-diabetic groups from Mexico City and Orizaba or betweenthe cases and the controls. Note that not all polymorphisms were
), SNP-63/C and SNP-110/A represent allele 1 of these polymorphisms
iabetic subjects from Mexico City and Orizaba
Non-diabetic
Mexico City Orizaba Combined
0.15 (17) 0.09 (10) 0.12 (27)
0.24 (26) 0.29 (32) 0.26 (58)
0.29 (32) 0.30 (34) 0.30 (66)
0.32 (35) 0.32 (36) 0.32 (71)
t differences in the haplotype frequency distributions between type 2
the combined type 2 diabetic and non-diabetic groups.
Page 4
tions
nty, T
icans
iabeti
tarr C
exica
expan
rom t
L. del Bosque-Plata et al. / Molecular Genetics and Metabolism 81 (2004) 122–126 125Table 5
Comparison of four core polymorphisms in calpain-10 gene in popula
Polymorphism (allele) Mexican ancestry
Mexico City/Orizaba
ðn ¼ 226Þ
Starr Cou
ðn ¼ 206Þ
Table 4
SNP-43, -19, and -63 haplogenotypes and risk of type 2 diabetes in Mex
Haplogenotype n—Type 2 diabetes
(Mexico/Starr County)
n—Non-d
(Mexico/S
111/111 4/10 1/17
111/121 13/37 13/38
111/112 6/21 6/20
111/221 13/20 6/26
112/112 8/10 5/16
112/121 21/48 16/28
112/221 19/18 26/22
121/121 8/27 7/47
121/221 26/41 23/41
221/221 11/9 8/16
The number of subjects with each haplogenotype is shown for the M
group includes both patient groups described in [2] ðn ¼ 241Þ and an
community. Note that the data for the Starr County sample differ f
population.Discussion
Genetic variation in CAPN10 appears to affect risk of
type 2 diabetes in individuals of Mexican ancestry from
both central Mexico and the border region between
Mexico and the United States (Starr County, Texas).
SNP-44 was associated increased risk of type 2 diabetes
in both populations. This polymorphism has also been
associated with increased risk in the white population of
the United Kingdom [3] and with the shape of the glu-
cose curve during an oral glucose tolerance test in non-
diabetic German subjects [5]. Since SNP-44 is in perfect
linkage disequilibrium with the missense mutation
T504A (SNP-110) as well as two polymorphisms in the
50-untranslated region of calpain-10 mRNA, we cannot
say which is responsible for the increased risk. However,
it is tempting to speculate that the missense mutation
may be causal because it could potentially affect calpain-
10 function.
The 112/121 haplogenotype which has an OR of 2.16-
fold in the Mexican American population of Starr
County has an OR of 1.15 in the Mexico City/Orizaba
population studied here. It is possible that the initial
estimate of the genetic effect of the 112/121 haplogeno-
SNP-44 (C) 0.035 (0.011–0.060) 0.058 (0.026–0.
SNP-43 (G) 0.674 (0.612–0.736) 0.733 (0.673–0.
SNP-19 (2R) 0.384 (0.320–0.448) 0.435 (0.369–0.
SNP-63 (C) 0.712 (0.653–0.771) 0.757 (0.698–0.
The frequency and the 95% confidence individual on the proportion are s
Spanish subjects [9], and random samples from Starr County. The Nahua s
Zacatipan, Tamazunchale, San Luis Potos�ı.
n—The number of chromosomes typed although some markers were notof Mexican, native American, and European ancestry
Native American European
exas Nahua ðn ¼ 92Þ Spanish ðn ¼ 186Þ
(Mexico City/Orizaba) and Mexican Americans (Starr County, Texas)
c
ounty)
OR (95% CI)
Mexico City/Orizaba
OR (95% CI)
Starr County
3.52 (0.39–31.97) 0.65 (0.29–1.44)
0.78 (0.34–1.79) 1.11 (0.68–1.82)
0.85 (0.27–2.73) 1.20 (0.63–2.27)
1.96 (0.72–5.35) 0.85 (0.46–1.57)
1.40 (0.44–4.42) 0.69 (0.31–1.55)
1.15 (0.57–2.34) 2.16 (1.31–3.57)
0.56 (0.29–1.09) 0.91 (0.48–1.75)
0.98 (0.34–2.80) 0.60 (0.36–1.00)
0.97 (0.51–1.81) 1.15 (0.72–1.84)
1.20 (0.46–3.10) 0.62 (0.27–1.43)
n and Mexican–American study populations. The Mexican–American
ded random sample of 271 individuals drawn from the Starr County
hose reported in [2] and reflect our current estimates of risk in thistype may have been overestimated [6,7]. The difference
in risk may also be a consequence of study design. The
cases in the Starr County study were drawn from fam-
ilies with at least two affected siblings and the sample
itself provided strong evidence for linkage with markers
in chromosome band 2q37 whereas the cases in the
Mexico City/Orizaba sample did not uniformly have a
positive family history of diabetes. These populations
may also differ genetically due to differences in degree of
admixture. The frequencies of SNP-43, -44, and -63 and
Indel-19 are not significantly different between the two
populations of Mexican ancestry or between these
populations and the Nahua, a Mexican Native Ameri-
can population (Table 5). However, there are significant
differences in SNP-44 and -63 allele frequencies between
these populations and Europeans (Spanish). Further
studies are needed to determine the genetic relationships
between various populations of Mexican ancestry.
Studies of the geographic and haplotype structure of
the calpain-10 gene have suggested a history of natural
selection acting on this locus [8]. However, the nature of
the selective forces acting on this gene or the normal
function(s) calpain-10 in the regulation of blood glucose
levels are presently unknown. Ongoing studies may
090) 0.054 (0.008–0.101) 0.113 (0.067–0.158)
793) 0.706 (0.613–0.800) 0.753 (0.691–0.815)
501) 0.424 (0.323–0.525) 0.383 (0.279–0.417)
816) 0.641 (0.543–0.739) 0.918 (0.878–0.958)
hown. The data are for non-diabetic Mexico City/Orizaba, Nahua, and
ubjects were non-diabetic relatives of patients from Hospital Rural 44
typed in all chromosomes.
Page 5
reveal how natural selection has shaped this gene and
converted variants in this gene which may have been
adaptive at one point in our evolution into risk factors
for a common disease.
Acknowledgments
This study was supported by US Public Health Service
Grants DK-20595, DK-47486, DK-47487, and DK-
55889, andGrant 40400-M from theConsejoNacional de
Ciencia y Tecnolog�ıa (CONACyT) M�exico. G.I.B. is an
Investigator of the Howard Hughes Medical Institute.
References
[1] C.L Hanis, E. Boerwinkle, R. Chakraborty, D.L. Ellsworth, P.
Concannon, B. Stirling, V.A. Morrison, B. Wapelhorst, R.S.
Spielman, K.J. Gogolin-Ewens, J.M. Shephard, S.R. Williams, N.
Risch, D. Hinds, N. Iwasaki, M. Ogata, Y. Omori, C. Petzold, H.
Rietzsch, H.-E. Schr€oder, J. Schulze, N.J. Cox, S. Menzel, V.V.
Boriraj, X. Chen, L.R. Lim, T. Lindner, L.E. Mereu, Y.-Q. Wang,
K. Xiang, K. Yamagata, Y. Yang, G.I. Bell, A genome-wide search
for human non-insulin-dependent (type 2) diabetes genes reveals a
Bell, Genetic variation in the gene encoding calpain-10 is associated
with type 2 diabetes mellitus, Nat. Genet. 26 (2000) 163–175.
[3] J.C. Evans, T.M. Frayling, P.G. Cassell, P.J. Saker, G.A. Hitman,
M. Walker, J.C. Levy, S. O�Rahilly, P.V.S. Rao, A.J. Bennett, E.C.
Jones, S. Menzel, P. Prestwich, N. Simecek, M. Wishart, R.
Dhillon, C. Fletcher, A. Millward, A. Demaine, T. Wilkin, Y.
Horikawa, N.J. Cox, G.I. Bell, S. Ellard, M.I. McCarthy, A.T.
Hattersley, Studies of the association between the gene for calpain-
10 and type 2 diabetes mellitus in the United Kingdom, Am. J.
Hum. Genet. 69 (2001) 544–552.
[4] S.K. Rasmussen, S.A. Urhammer, L. Berglund, J.N. Jensen, L.
Hansen, S.M. Echwald, K. Borch-Johnsen, Y. Horikawa, H.
Mashima, H. Lithell, N.J. Cox, T. Hansen, G.I. Bell, O. Pedersen,
Variants within the calpain-10 gene on chromosome 2q37
(NIDDM1) and relationships to type 2 diabetes, insulin resistance,
and impaired insulin secretion among Scandinavian Caucasians,
Diabetes 51 (2002) 3561–3567.
[5] O. Tschritter, A. Fritsche, F. Shirkavand, F. Machicao, H. Haring,
M. Stumvoll, Assessing the shape of the glucose curve during an
oral glucose tolerance test, Diabetes Care 26 (2003) 1026–1033.
[6] J.P.A. Ioannidis, E.E. Ntzani, T.A. Trikalinos, D.G. Contopoulos-
Ioannidis, Replication validity of genetic association studies, Nat.
Genet. 29 (2001) 306–309.
[7] K.E. Lohmueller, C.L. Pearce, M. Pike, E.S. Lander, J.N.
Hirschhorn, Meta-analysis of genetic association studies supports
a contribution of common variants to susceptibility to common
disease, Nat. Genet. 33 (2003) 177–182.
[8] S.M. Fullerton, A. Bartoszewicz, G. Ybazeta, Y. Horikawa, G.I.
Bell, K.K. Kidd, N.J. Cox, R.R. Hudson, A. Di Rienzo,
126 L. del Bosque-Plata et al. / Molecular Genetics and Metabolism 81 (2004) 122–126major susceptibility locus on chromosome 2, Nat. Genet. 13 (1996)
161–166.
[2] Y. Horikawa, N. Oda, N.J. Cox, X. Li, M. Orho-Melander, M.
Hara, Y. Hinokio, T.H. Lindner, H. Mashima, P.E.H. Schwarz, L.
del Bosque-Plata, Yo. Horikawa, Y. Oda, I. Yoshiuchi, S. Colilla,
K.S. Polonsky, S. Wei, P. Concannon, N. Iwasaki, J. Schulze, L.J.
Baier, C. Bogardus, L. Groop, E. Boerwinkle, C.L. Hanis, G.I.Geographic and haplotype structure of candidate type 2 diabetes-
susceptibility variants at the calpain-10 locus, Am. J. Hum. Genet.
70 (2002) 1096–1106.
[9] A. Gonzalez, E. Abril, A. Roca, M.J. Arag�on, M.J. Figueroa, P.
Velarde, J.L. Royo, L.M. Real, A. Ruiz, CAPN10 alleles are
associated with polycystic ovary syndrome, J. Clin. Endocrinol.
Metab. 87 (2002) 3971–3976.
