Testing the Gene or Testing a Variant?
The Case of TCF7L2
Mark O. Goodarzi

1,2,3
and Jerome I. Rotter

2,3,4,5

G
iven that susceptibility to type 2 diabetes ap-
pears in large measure due to genetic makeup,
investigators have spent years of effort trying to
identify genes that influence type 2 diabetes

risk. An important goal of gene identification is to improve
our understanding of the pathophysiology of diabetes,
leading to new measures of diagnosis, prevention, and
treatment. A diabetes gene is considered identified when
variants in that gene (more specifically, variation in DNA
sequence between individuals) are found to be associated
with type 2 diabetes and/or its pathophysiologic abnormal-
ities such as insulin resistance or secretion.

TCF7L2 (transcription factor 7-like 2) was identified as a
gene for type 2 diabetes in 2006 (1). Notably, its effect on
diabetes (relative risk 1.5–1.6) is substantially larger than
previously established diabetes genes (e.g., peroxisome
proliferator–activated receptor � [PPARG] and �-cell in-
wardly rectifying K� channel KIR6.2 [KCNJ11], relative
risk �1.2 each). Most studies of TCF7L2 have focused on
the genetic variants that were implicated in the original
report (1), largely ignoring the remainder of the gene. In
this issue of Diabetes, investigators took the alternative
approach of examining variants across the entire gene,
which allowed them to discover a completely novel variant
in TCF7L2 that affects diabetes risk (2). We believe this
approach to genetic association studies is of great merit,
as described below.

In the report that first identified TCF7L2 as a diabetes
gene, a microsatellite (DG10S478) was highly associated
with type 2 diabetes (in three Caucasian cohorts), as were
five single nucleotide polymorphisms (SNPs) in linkage
disequilibrium (LD). Of those five SNPs, two (rs12255372
and rs7903146) were identified as most strongly associated
with type 2 diabetes; subsequent reports determined SNP
rs7903146 had the greatest effect (3,4). Deep resequencing

of exons and the surrounding region has not identified any
variants with stronger effect (4). Over 50 articles subse-
quently described the role of TCF7L2 in multiple cohorts,
firmly establishing the gene’s place among diabetes genes
(5). Quantitative phenotypic associations (3,6,7) and its
role in proglucagon gene expression (8) suggest that
TCF7L2 modulates diabetes risk via impaired insulin se-
cretion. Further support that TCF7L2 influences insulin
secretion comes from an article recently published in
Diabetes, wherein genetic variation in TCF7L2 was shown
to influence efficacy of sulfonylureas (agents that promote
insulin secretion) but not efficacy of metformin (insulin
sensitizer) (9).

The majority of the subsequent studies followed the lead
of the original study, genotyping either the one or two
most associated SNPs, or all five, with or without the
microsatellite DG10S478. These variants are in a haplotype
block that encompasses DG10S478 and includes part of
intron 3, all of exon 4, and part of intron 4 (1). In the few
instances where more extensive SNP genotyping was
performed, it was usually centered around DG10S478
(3,10,11). As a result, SNP rs7903146 (specifically the
T-allele) has been firmly established as increasing risk of
type 2 diabetes in multiple European populations and in
Caucasians, in general, and including diverse groups such
as Mexicans, Amish, Indian Asians, and Moroccans (5).
Some studies of individuals of African descent were not
able to document association of TCF7L2 with diabetes
(12,13), while others, including a study in this issue of
Diabetes, were positive for rs7903146 (4,14). Unlike in
Caucasians, the two most associated SNPs exhibit weak
LD in Africans, allowing determination of the greater role
of rs7903146 than rs12255372 in diabetes susceptibility.
The functional role that this intronic SNP may play is still
unknown. Furthermore, it does not explain the linkage
signal that originally led investigators to chromosome 10q
(1), raising the possibility that other variants in the region
(in TCF7L2 or other genes) may predispose to type 2
diabetes. Other plausible candidate genes do exist on
chromosome 10q (15,16).

The near-exclusive focus on rs7903146 and SNPs in LD
with it has probably delayed the discovery of other vari-
ants in TCF7L2 that may affect risk of type 2 diabetes. In
Asian populations, the frequencies of SNPs rs7903146 and
rs12255372 are quite low. Nevertheless, association with
these SNPs was identified in two large Japanese cohorts
(�1,000 cases each), where the minor allele frequency
(MAF) ranged from 3 to 5% (17,18). The large size of these
cohorts provided adequate power for discovery despite
the rarity of the variants examined. This was fortuitous;
had the allele frequencies or sample sizes been slightly
lower, false-negative studies would have been the likely
result, even though the gene and those variants are pro-
ducing risk in these populations.

From the 1Division of Endocrinology, Diabetes and Metabolism, Cedars-Sinai
Medical Center, Los Angeles, California; the 2Medical Genetics Institute,
Cedars-Sinai Medical Center, Los Angeles, California; the 3Department of
Medicine, University of California, Los Angeles, Los Angeles, California; the
4Department of Pediatrics, University of California, Los Angeles, Los Angeles,
California; and the 5Department of Human Genetics, University of California,
Los Angeles, Los Angeles, California.

Address correspondence and reprint requests to Mark O. Goodarzi, MD,
PhD, Cedars-Sinai Medical Center, Division of Endocrinology, Diabetes and
Metabolism, 8700 Beverly Blvd., Becker B-131, Los Angeles, CA 90048. E-mail:
mark.goodarzi@cshs.org.

Received for publication 5 July 2007 and accepted in revised form 6 July
2007.

LD, linkage disequilibrium; MAF, minor allele frequency; SNP, single
nucleotide polymorphism.

DOI: 10.2337/db07-0923
© 2007 by the American Diabetes Association.
The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked “advertisement” in accordance

with 18 U.S.C. Section 1734 solely to indicate this fact.

COMMENTARY

DIABETES, VOL. 56, OCTOBER 2007 2417



In the study of Chinese Han (760 case and 760 control
subjects) reported in this issue (2), if the investigators had
followed the usual strategy of looking only at the original
associated SNPs, they would have found no associations.
In Chinese, the rarity of rs7903146 (MAF 2–2.5%) would
require 1,700 cases to detect an association with an effect
size similar to that in Caucasians (2). Fortunately, these
investigators tested the entire TCF7L2 gene rather than
only particular variants of prior interest. In addition to
genotyping SNPs previously associated with type 2 diabe-
tes, they used information from HapMap (19) to select a
set of 13 SNPs to capture the majority of SNPs in the gene
with MAF �20%. The “classic” SNPs showed no associa-
tion; however, SNP rs290487 at the 3� end of the gene, as
well as haplotypes including this SNP, were associated
with type 2 diabetes. The comprehensive approach of this
study led to this identification of a TCF7L2 SNP not in LD
with rs7903146 that may affect type 2 diabetes risk. This
demonstrates the utility of the global gene approach.

Given the narrow focus of prior studies of TCF7L2, no
other study to date has examined this recently identified 3�
variant. Those groups with large diabetic cohorts should
go back and examine this variant and its neighbors for a
role in type 2 diabetes, which is feasible given that
rs290487 occurs with appreciable frequency in all HapMap
populations except Yorubans. HapMap also reveals that
this SNP is not well captured by other SNPs (at r2 � 0.8),
indicating that its effect is likely to be missed if the SNP
itself is not genotyped. The only other study to take a
global approach to TCF7L2 (challenging with this large
[�216 kb] gene) was carried out in Mexican Americans
(20). This study did not genotype rs290487 but did geno-
type rs290483, also in the 3� end, with negative results.
Notably, one of the recent genome-wide association stud-
ies for type 2 diabetes found that rs290483 was associated
with type 2 diabetes (a finding eclipsed by the stronger
association of rs7903146 LD group SNPs) (21). This latter
observation and the data of Chang et al. (2) suggest the
potential functional importance of variation in the 3� end
of the gene, which may influence alternative splicing of
TCF7L2, in which there exist many alternative splice sites
(22). Clearly this end of the gene deserves the kind of
attention that the upstream region of rs7903149 has received.

When follow-up gene marker studies focus only on the
original associated variants, they amount only to replica-
tion studies that will not result in novel discoveries unless
they examine a related but different phenotype, e.g.,
response to treatment (9), or extend the result to addi-
tional populations (14). To test the entire gene, compre-
hensive genotyping such as that carried out by Chang et al.
(2) is necessary. This is particularly important if the
replication cohort is not of the same ethnic group as the
original cohort. Comprehensive global genotyping leads to
a complete understanding of the SNP frequencies and
haplotype structure, which may account for differences in
association results. Given the availability of the HapMap
database, and the falling price of genotyping, investigators
now more readily have the option to test the entire gene
(or region) rather than only a particular variant.

REFERENCES

1. Grant SF, Thorleifsson G, Reynisdottir I, Benediktsson R, Manolescu A,
Sainz J, Helgason A, Stefansson H, Emilsson V, Helgadottir A, Styrkars-
dottir U, Magnusson KP, Walters GB, Palsdottir E, Jonsdottir T, Gud-
mundsdottir T, Gylfason A, Saemundsdottir J, Wilensky RL, Reilly MP,
Rader DJ, Bagger Y, Christiansen C, Gudnason V, Sigurdsson G, Thor-

steinsdottir U, Gulcher JR, Kong A, Stefansson K: Variant of transcription
factor 7-like 2 (TCF7L2) gene confers risk of type 2 diabetes. Nat Genet
38:320 –323, 2006

2. Chang YC, Chang TJ, Jiang YD, Kuo SS, Lee KC, Chiu KC, Chuang LM:
Association study of the genetic polymorphisms of the transcription factor
7-like 2 (TCF7L2) gene and type 2 diabetes in the Chinese population.
Diabetes 56:2631–2637, 2007

3. Saxena R, Gianniny L, Burtt NP, Lyssenko V, Giuducci C, Sjogren M, Florez
JC, Almgren P, Isomaa B, Orho-Melander M, Lindblad U, Daly MJ, Tuomi
T, Hirschhorn JN, Ardlie KG, Groop LC, Altshuler D: Common single
nucleotide polymorphisms in TCF7L2 are reproducibly associated with
type 2 diabetes and reduce the insulin response to glucose in nondiabetic
individuals. Diabetes 55:2890 –2895, 2006

4. Helgason A, Palsson S, Thorleifsson G, Grant SF, Emilsson V, Gunnars-
dottir S, Adeyemo A, Chen Y, Chen G, Reynisdottir I, Benediktsson R,
Hinney A, Hansen T, Andersen G, Borch-Johnsen K, Jorgensen T, Schafer
H, Faruque M, Doumatey A, Zhou J, Wilensky RL, Reilly MP, Rader DJ,
Bagger Y, Christiansen C, Sigurdsson G, Hebebrand J, Pedersen O,
Thorsteinsdottir U, Gulcher JR, Kong A, Rotimi C, Stefansson K: Refining
the impact of TCF7L2 gene variants on type 2 diabetes and adaptive
evolution. Nat Genet 39:218 –225, 2007

5. Florez JC: The new type 2 diabetes gene TCF7L2. Curr Opin Clin Nutr
Metab Care 10:391–396, 2007

6. Damcott CM, Pollin TI, Reinhart LJ, Ott SH, Shen H, Silver KD, Mitchell
BD, Shuldiner AR: Polymorphisms in the transcription factor 7-like 2
(TCF7L2) gene are associated with type 2 diabetes in the Amish: replica-
tion and evidence for a role in both insulin secretion and insulin resistance.
Diabetes 55:2654 –2659, 2006

7. Florez JC, Jablonski KA, Bayley N, Pollin TI, de Bakker PI, Shuldiner AR,
Knowler WC, Nathan DM, Altshuler D: TCF7L2 polymorphisms and
progression to diabetes in the Diabetes Prevention Program. N Engl J Med
355:241–250, 2006

8. Yi F, Brubaker PL, Jin T: TCF-4 mediates cell type-specific regulation of
proglucagon gene expression by beta-catenin and glycogen synthase
kinase-3beta. J Biol Chem 280:1457–1464, 2005

9. Pearson ER, Donnelly LA, Kimber C, Whitley A, Doney AS, McCarthy MI,
Hattersley AT, Morris AD, Palmer CN: Variation in TCF7L2 influences
therapeutic response to sulfonylureas: a GoDARTs study. Diabetes 56:
2178 –2182, 2007

10. Watanabe RM, Allayee H, Xiang AH, Trigo E, Hartiala J, Lawrence JM,
Buchanan TA: Transcription factor 7-like 2 (TCF7L2) is associated with
gestational diabetes mellitus and interacts with adiposity to alter insulin
secretion in Mexican Americans. Diabetes 56:1481–1485, 2007

11. Scott LJ, Bonnycastle LL, Willer CJ, Sprau AG, Jackson AU, Narisu N,
Duren WL, Chines PS, Stringham HM, Erdos MR, Valle TT, Tuomilehto J,
Bergman RN, Mohlke KL, Collins FS, Boehnke M: Association of transcrip-
tion factor 7-like 2 (TCF7L2) variants with type 2 diabetes in a Finnish
sample. Diabetes 55:2649 –2653, 2006

12. Humphries SE, Gable D, Cooper JA, Ireland H, Stephens JW, Hurel SJ, Li
KW, Palmen J, Miller MA, Cappuccio FP, Elkeles R, Godsland I, Miller GJ,
Talmud PJ: Common variants in the TCF7L2 gene and predisposition to
type 2 diabetes in UK European Whites, Indian Asians and Afro-Caribbean
men and women. J Mol Med 84:1–10, 2006

13. Elbein SC, Chu WS, Das SK, Yao-Borengasser A, Hasstedt SJ, Wang H,
Rasouli N, Kern PA: Transcription factor 7-like 2 polymorphisms and type
2 diabetes, glucose homeostasis traits and gene expression in US partici-
pants of European and African descent. Diabetologia 50:1621–1630, 2007

14. Sale MM, Smith SG, Mychaleckyj JC, Keene KL, Langefeld CD, Leak TS,
Hicks PJ, Bowden DW, Rich SS, Freedman BI: Variants of the transcription
factor 7-like 2 (TCF7L2) gene are associated with type 2 diabetes in an
African-American population enriched for nephropathy. Diabetes 56:2638 –
2642, 2007

15. Goodarzi MO, Lehman DM, Taylor KD, Guo X, Cui J, Quinones MJ, Clee
SM, Yandell BS, Blangero J, Hsueh WA, Attie AD, Stern MP, Rotter JI:
SORCS1: a novel human type 2 diabetes susceptibility gene suggested by
the mouse. Diabetes 56:1922–1929, 2007

16. Lehman DM, Fu DJ, Freeman AB, Hunt KJ, Leach RJ, Johnson-Pais T,
Hamlington J, Dyer TD, Arya R, Abboud H, Goring HH, Duggirala R,
Blangero J, Konrad RJ, Stern MP: A single nucleotide polymorphism in
MGEA5 encoding O-GlcNAc-selective N-acetyl-�-D glucosaminidase is as-
sociated with type 2 diabetes in Mexican Americans. Diabetes 54:1214 –
1221, 2005

17. Hayashi T, Iwamoto Y, Kaku K, Hirose H, Maeda S: Replication study for
the association of TCF7L2 with susceptibility to type 2 diabetes in a
Japanese population. Diabetologia 50:980 –984, 2007

18. Horikoshi M, Hara K, Ito C, Nagai R, Froguel P, Kadowaki T: A genetic

TCF7L2: GENE OR VARIANT

2418 DIABETES, VOL. 56, OCTOBER 2007



variation of the transcription factor 7-like 2 gene is associated with risk of
type 2 diabetes in the Japanese population. Diabetologia 50:747–751, 2007

19. The International HapMap Consortium: The International HapMap Project.
Nature 426:789 –796, 2003

20. Lehman DM, Hunt KJ, Leach RJ, Hamlington J, Arya R, Abboud HE,
Duggirala R, Blangero J, Goring HH, Stern MP: Haplotypes of transcription
factor 7-like 2 (TCF7L2) gene and its upstream region are associated with
type 2 diabetes and age of onset in Mexican Americans. Diabetes 56:389 –
393, 2007

21. Sladek R, Rocheleau G, Rung J, Dina C, Shen L, Serre D, Boutin P, Vincent
D, Belisle A, Hadjadj S, Balkau B, Heude B, Charpentier G, Hudson TJ,
Montpetit A, Pshezhetsky AV, Prentki M, Posner BI, Balding DJ, Meyre D,
Polychronakos C, Froguel P: A genome-wide association study identifies
novel risk loci for type 2 diabetes. Nature 445:881– 885, 2007

22. Duval A, Rolland S, Tubacher E, Bui H, Thomas G, Hamelin R: The human
T-cell transcription factor-4 gene: structure, extensive characterization of
alternative splicings, and mutational analysis in colorectal cancer cell
lines. Cancer Res 60:3872–3879, 2000

M.O. GOODARZI AND J.I. ROTTER

DIABETES, VOL. 56, OCTOBER 2007 2419