Single nucleotide polymorphisms (SNPs) rs7903146 and rs12255372 located within TCF7L2 gene have been identified as the strongest common genetic risk factors for development of type 2 diabetes (T2D). We hypothesized that these genetic variants might increase the risk of T2D through regulation of alternative splicing or expression level of TCF7L2 in human adipose tissue.
Expression of 13 assays detecting alternatively spliced forms of TCF7L2 was measured by quantitative reverse-transcriptase PCR (qRT-PCR) in paired biopsies of omental and subcutaneous adipose tissue from 159 obese individuals (BMI 54.6+/−12.2 kg/m2). TCF7L2 expression in both types of adipose tissue was not associated with SNPs rs7903146 and rs12255372, T2D status and blood levels of glucose or glycosylated hemoglobin (HbA1c). Expression of assays “ex12-13”, “ex12-14” and “ex13-13a” detecting C-terminal alternative exons of TCF7L2 was higher in subcutaneous compared to omental adipose tissue by 1.46 fold (p = 6.5×10−15), 1.41 fold (p = 1.4×10−9) and 1.26 fold (p = 4.7×10−6) in the control group and by 1.86 fold (p = 1.7×10−4), 1.77 fold (p = 7.3×10−4) and 1.58 fold (p = 6.1×10−4) in the T2D group. A pathway enrichment analysis on transcripts significantly co-expressed with TCF7L2 in a microarray set combined with individual expression assays, suggested tissue-specific roles of TCF7L2 splicing forms in regulation of transcription, signal transduction and cell adhesion.
Expression of TCF7L2 alternatively spliced forms may have different functional roles in omental and subcutaneous adipose tissue but is not associated with SNPs rs7903146 and rs12255372 or T2D status.
Citation: Prokunina-Olsson L, Kaplan LM, Schadt EE, Collins FS (2009) Alternative Splicing of TCF7L2 Gene in Omental and Subcutaneous Adipose Tissue and Risk of Type 2 Diabetes. PLoS ONE 4(9): e7231. doi:10.1371/journal.pone.0007231
Editor: Per Westermark, Uppsala University, Sweden
Received: May 8, 2009; Accepted: September 1, 2009; Published: September 30, 2009
This is an open-access article distributed under the terms of the Creative Commons Public Domain declaration which stipulates that, once placed in the public domain, this work may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose.
Funding: The study was supported by the Intramural Research Programs of the NHGRI and NCI/NIH. One of the authors (EES) is employed by Pacific Biosciences. The company did not have a role in funding or oversight of this study, data collection and analysis, decision to publish, or preparation of the manuscript. No patent can be claimed based on this study.
Competing interests: The authors have declared that no competing interests exist.
Common intronic single nucleotide polymorphisms (SNPs) within the transcription factor 7 - like 2 gene (TCF7L2) have been identified as genetic factors that significantly increase risk of type 2 diabetes (T2D) , , , . TCF7L2 belongs to a family of TCF/LEF transcription factors that interact with β-catenin and regulate the WNT pathway . Activation of the WNT pathway leads to increased cell proliferation due to effects of downstream targets of TCF7L2 such as MYC ,  and CCND1 (Cyclin D1) . A complex interplay of activation and repression of the WNT pathway, orchestrated by different protein isoforms of TCF/LEF transcription factors, is required for tissue-specific differentiation of stem cells. For example, differentiation of skin stem cells into either hair follicle or sebum-producing cells is regulated by expression of alternatively spliced forms of the LEF1 transcription factor . Similarly, an active WNT pathway is required for myogenesis, while inactivation of the pathway by a dominant-negative form of TCF7L2 promotes adipogenesis . Increased adiposity, as measured by body mass index (BMI), is a strong risk factor for development of insulin resistance, T2D and cardiovascular disease . Several studies have reported that patients carrying risk alleles of the associated SNPs rs7903146 and rs12255372 of TCF7L2 have lower BMI compared to carriers of non-risk alleles , , , . Potentially, risk alleles of TCF7L2 might increase risk of T2D even in lean individuals, or affect diabetes and adiposity through independent mechanisms.
Non-coding genetic variants can affect mRNA expression and splicing , . Several studies attempted to correlate genotypes of T2D-associated variants of TCF7L2 with mRNA expression of TCF7L2 in adipose tissue ,  skeletal muscle , lymphoblastoid cell lines  and pancreatic islets , , but no consistent associations have been reported. One study detected a significant decrease in TCF7L2 expression in obese individuals with T2D compared to obese controls, but this study was based only on six samples . We previously catalogued and evaluated expression of multiple splicing forms of TCF7L2 in several types of human tissue . We observed tentative association between expression of several assays for C-terminal exons of TCF7L2 and genotypes of SNPs rs7903146 and rs12255372 in pancreatic islets but not in a small set of samples of subcutaneous adipose tissue .
Here, we used 13 assays detecting all known splicing forms of TCF7L2 to evaluate gene expression in paired biopsies of subcutaneous and omental adipose tissue from 159 obese individuals. We evaluated the association between expression of these assays and genotypes of T2D-associated variants rs7903146 and rs12255372, T2D status, type of adipose tissue, BMI (37.6–89.6 kg/m2) and blood levels of glucose and glycosylated hemoglobin (HbA1c). We show that expression of alternatively spliced forms of TCF7L2 may have different functional roles in omental and subcutaneous adipose tissue but is not associated with SNPs rs7903146 and rs12255372 or T2D status.
Characteristics of the T2D and control groups
SNPs rs7903146 and rs12255372 are located in introns 3 and 4 of TCF7L2 gene within the associated linkage disequilibrium (LD) block and 50 kb apart from each other , ,  (Fig. 1). In our set of 159 Caucasian individuals, the frequencies of risk alleles of both SNPs were higher in the T2D group (n = 16) than in the control group (n = 143), 0.41 in T2D vs. 0.26 in controls for rs7903146 and 0.38 in T2D vs. 0.25 in controls for rs12255372 (Table 1). Similarly to other European sets , ,  and to the European set (CEU) of the HapMap , these two SNPs were in high linkage disequilibrium (LD) with each other (D′ = 0.88, r2 = 0.73). Samples in the T2D and control groups were matched by age, gender and BMI by design but several T2D-related traits were significantly different between these groups: levels of blood glucose, HbA1c and homeostasis model of insulin resistance (Homa-IR) were increased, while the level of HDL cholesterol was decreased in the T2D group (Table 1).
The scheme shows constitutive exons (black rectangles), alternative exons (white rectangles); linkage disequilibrium (LD) block with associated SNPs rs7903146 and rs12255372; protein domains: ß-catenin-interacting domain, DNA-binding HMG domain and C-terminal Binding Protein (CtBP)-binding domain; expression assays for detection of alternative splicing forms of TCF7L2 are indicated by connected arrows: assays TSS1, TSS2 and TSS3 target transcripts produced from alternative transcription start sites while other assays target alternatively spliced forms with different combinations of exons, * - expression of assay “ex13-13b” was tested but not detected in adipose tissue.
Test for association between TCF7L2 expression and genotypes of rs7903146 and rs12255372, T2D status and blood levels of glucose and HbA1c
Levels of expression of 13 TCF7L2 assays in both types of adipose tissue were adjusted for age, sex, BMI and blood levels of glucose and HbA1c. Expression of these assays in each individual tissue and the ratio of subcutaneous: omental expression was similar in the control and T2D groups and in carriers of different genotypes of SNPs rs7903146 and rs1225532 (Table 2 and 3). Expression of assay “ex13-13b” (Fig. 1) previously studied in human pancreatic islets  was tested but not detected in adipose tissue (data not shown). Based on the number of samples and the observed standard deviation in expression of each assay, we had 80% power and 95% confidence to detect a >1.2-fold difference in expression between groups with risk and non-risk alleles for each SNP and between the T2D and control groups.
Difference in TCF7L2 expression between omental and subcutaneous adipose tissues
Expression of several assays was higher in subcutaneous compared to omental adipose tissue both in the control and T2D groups: 1.46 and 1.86-fold for assay “ex12-13”, 1.41 and 1.77-fold for assay “ex12-14” and 1.26 and 1.58-fold for assay “ex13-13a” (Table 4). Only expression of alternative exon 3a (assay “ex3a-4”) was decreased in subcutaneous compared to adipose tissue (0.81 fold) in the control group, but not in the T2D group (0.91 fold) (Table 4). The results for assays “ex12-13”, “13-13a” and “12-14” will remain significant even after adjustment for multiple tests. Different ways of normalization of TCF7L2 expression (by expression levels of endogenous controls B2M, GAPDH or both genes together) did not significantly affect the conclusions (data not shown). The assays “ex12-13”, “13-13a” and “12-14” detect two distinct splicing forms that include alternative exon 12 (Figure 2). The first form (GenBank FJ010174) includes C-terminal exons 11-12-13-13a and encodes a protein with a short reading frame terminated by an alternative stop codon within exon 13a. The second form (GenBank FJ010170) includes exons 11-12-14 and encodes a protein with a medium reading frame terminated by an alternative stop codon in the beginning of exon 14 (Fig. 2).
A. Constitutive exons 11 and 14 are marked as black rectangles, alternative exons 12, 13, 13a, 13b are marked as white rectangles, black triangles above exons indicate location of alternative stop codons that define protein reading frames: short (S), medium (M) or long (L); B. Alternative splicing form with exons 11-12-13-13a (GenBank FJ010174) has an alternative stop codon within exon 13a and encodes a protein with short reading frame; C. Alternative splicing form with exons 11-12-14 (GenBank FJ010170) has an alternative stop codon in the beginning of exon 14 and encodes a protein with medium reading frame. Positions of expression assays “ex12-13”, “ex13-13a” and “12-14” used in this study are indicated above corresponding exons. Assays “ex12-13”, “ex13-13a” and “12-14” detect all splicing forms of TCF7L2 that include alternative exon 12.
Pathway enrichment analysis of TCF7L2 expression
To better understand the function of TCF7L2 in human adipose tissue we searched for transcripts that positively or negatively correlated with expression of TCF7L2. We combined global expression data measured by microarrays  with data for individual expression assays “ex7-8”, “ex3a-4”, “ex12-13” and “ex13-13a”, generated in the same set of 159 samples of omental and subcutaneous adipose tissue. On the microarrays, TCF7L2 was represented by 2 probes located within the 3′ untranslated region (UTR) of RefSeq transcript NM_030756. The microarray probes and the assay “ex7-8” detect a mix of all splicing forms of TCF7L2, while the assays “ex12-13”, “ex13-13a” and “ex3a-4” detect distinct splicing forms that showed difference in expression between the two types of adipose tissue. There was a significant correlation between the microarray and TaqMan expression data in omental adipose tissue: r2 = 0.51 for assay “ex7-8”, r2 = 0.49 for assay “ex13-13a”, r2 = 0.42 for assay “ex12-13” and r2 = 0.31 for “ex3a-4” but in subcutaneous adipose tissue the correlation for the same assays was not significant (r2<0.25). Using the joint expression set (microarrays and TaqMan data), we identified sets of transcripts that correlated with expression of these assays at a cutoff of r2>+/−0.25, corresponding to False Discovery Rate (FDR) p<0.0015, and analyzed each set of transcripts for enrichment of Gene Ontology categories (GO). In omental adipose tissue TCF7L2 expression positively correlated with activation of transcription through KRAB box Zn finger transcription factors but negatively correlated with protein biosynthesis, signal transduction, oxidative phosphorylation and electron transport (Table 5). In subcutaneous adipose tissue the strongest positive correlation was between expression of assay “ex13-13a” and the cadherin signaling pathway and cell adhesion, while negative correlation was detected with protein biosynthesis and receptor function (Table 5). Full list of transcripts with positive and negative correlations with TCF7L2 assays “ex12-13” and “ex13-13a” are presented in Tables S2, S3, S4, S5.
Our study provides a detailed analysis of expression of multiple TCF7L2 splicing forms in paired biopsies of omental and subcutaneous adipose tissue from a large set of obese individuals. The first goal of this study was to evaluate the association between the expression of TCF7L2 splicing forms and T2D-associated SNPs rs7903146 and rs12255372, T2D status or blood levels of glucose or HbA1c. Despite 80% of power to detect >1.2 fold differences in expression between different groups, we did not observe any significant association between these factors. Our results are similar to a recently published data on 235 samples of subcutaneous adipose tissue where expression of TCF7L2 (measured by one assay corresponding to our “ex7-8” assay) was not affected by T2D status, sex, and genotypes of rs7903146 .
The second goal of this study was to compare the expression of TCF7L2 splicing forms in two types of adipose tissue. We observed that expression of assays “ex12-13”, “ex12-14” and “ex13-13a” was higher in subcutaneous compared to omental tissue and this difference could not be explained by genotypes of the T2D-associated SNPs, sex, BMI and blood levels of glucose and HbA1c. These findings were similar to the results of a study that did not detect association between expression of assay “ex9-10” of TCF7L2 and genotypes of rs7903146, rs12255372, age, sex and BMI in paired biopsies of adipose tissue from 49 individuals . However, the expression of this assay was 4.8-fold higher in subcutaneous compared to omental tissue in controls and 6-fold higher in the T2D group . The effects observed in our study were weaker, 1.46-fold in the control group and 1.86-fold in the T2D group and the effect was not detectable for assay “ex7-8” that measures the same splicing forms as assay “ex9-10”. This could potentially be attributed to the difference in sample size (159 samples in our study and 49 samples in Kovacs et al.), use of different custom-designed expression assays and reference genes (one reference gene at Kovacs et al. and two reference genes in our study). Additionally, individuals in our study were more obese (BMI 54.6+/−12.2) compared to less obese individuals in Kovacs et al (BMI 29.7+/−0.8 kg/m2) .
TCF7L2 is a ubiquitously expressed transcription factor that mediates signals of the WNT pathway. The function of TCF7L2 and, particularly, in different types of adipose tissue is not clear. A pathway enrichment analysis on transcripts significantly coexpressed with TCF7L2 showed that the main role of TCF7L2 in omental adipose tissue is associated with activation of transcription and inhibition of protein biosynthesis and signal transduction. In contrast, in subcutaneous adipose tissue we did not observe any evidence for correlation with activation of transcription, but with increased cell adhesion through cadherin protein family (for assay “ex13-13a”). Adipose consists of several types of cells such as adipocytes, endothelial cells, macrophages that can differently express TCF7L2. For example, assays “ex12-13” and “13-13a” specifically detect a unique splicing form that includes exons 12-13-13a, expressed in multiple tissues and is the major TCF7L2 splicing form in peripheral blood monocytes (, GenBank FJ010174, Figure 2). Adhesion of monocytes to endothelial lining of capillaries, infiltration into adipose tissue and differentiation into macrophages contributes to low-grade chronic inflammation in adipose tissue and development of insulin resistance –. Therefore, increased expression of assays “ex12-13” and “ex13-13a” in subcutaneous adipose tissue and co-expression with genes involved in cell adhesion might indicate increased monocytes influx into subcutaneous adipose tissue or other functional differences between these two tissue types. All assays of TCF7L2 with significantly increased expression in subcutaneous compared to omental adipose (“ex12-13”, “ex13-13a” and “ex12-14”) detect splicing forms that encode truncated TCF7L2 protein with short or medium reading frames and include an alternative exon 12 (Figure 2). The truncated protein isoforms of TCF7L2 lack binding sites for the CTBP protein are not post-translationally regulated and may provide alternative regulation of the WNT pathway .
In conclusion, we did not observe association between expression of TCF7L2 splicing forms, T2D status and genotypes of T2D-associated SNPs rs7903146 and rs12255372. Differential expression of TCF7L2 splicing forms with alternative exon 12 encoding for truncated protein isoforms of TCF7L2 in omental and subcutaneous adipose tissues deserves future studies.
Materials and Methods
The study on anonymized samples was approved by the Internal Review Board (IRB) of Massachusetts General Hospital, protocol #2001-P-001942/22 and exempted from IRB approval at the NHGRI/NIH.
Paired samples of omental and subcutaneous adipose tissue were obtained from bariatric patients that underwent weight-reduction surgery and signed informed consent forms . Following traits were available for all or subsets of samples: age, gender, BMI, white blood cells counts (WBC), fasting levels of glucose, insulin, total cholesterol, HDL and LDL cholesterol, triglycerides, homeostasis model of insulin resistance (Homa-IR) and glycosylated hemoglobin (HbA1c). The T2D status was assigned to individuals with fasting blood glucose higher than 140 mg/dl (7.8 mmol/L). Total RNA from flash-frozen adipose samples was extracted using Trizol reagent (Invitrogen) and DNA was isolated from liver samples of the same patients with DNeasy kit (Qiagen).
Genotyping was performed with pre-developed TaqMan allelic discrimination assays for rs7903146 and rs12255372 (Applied Biosystems). The genotyping success was above 99% and only samples genotyped for both markers were used.
Quantitative reverse-transcriptase PCR (qRT-PCR) expression studies
cDNA was prepared from total RNA of selected samples of omental and subcutaneous adipose tissue used for the microarray expression studies . For each sample, 100 ng of total RNA was convereted to cDNA with SuperScript III reverse transcriptase and random hexamers (Invitrogen). Expression assays for TCF7L2 were custom designed for each splicing form ( and Table S1). Primers for SYBR Green assays were purchased from Integrated DNA Technologies and TaqMan assays were manufactured on demand by Applied Biosystems. qRT-PCR was performed in a 10 ul reaction volume in 384 well plates with Power SYBR Green master mix (Applied Biosystems) or Expression Master Mix for TaqMan assays (Applied Biosystems) on the Sequence Detection System 7900 (Applied Biosystems). Each expression assay was run in technical duplicates and the average values for each sample were normalized to a geometric mean of mRNA levels of reference genes beta 2 microglobulin (B2M, assay Hs00187842_m1, Applied Biosystems) and Glyceraldehyde 3-phosphate dehydrogenase (GAPDH, assay 4333764F, Applied Biosystems) run in separate reactions from the same cDNA preparations.
qRT-PCR expression was measured in Ct values (PCR cycle at which expression was detected). For each sample and assay, the average of technical replicates was first normalized to a geometric mean values of reference genes, B2M and GAPDH according to the formula: Ct(geometric) = (Ct(B2M)+Ct(GAPDH))/2. Relative expression of each assay was calculated as: dCt = Ct(geometric)−Ct(assay). For each assay, the dCt values were first tested for normality of distribution. The association between expression of TCF7L2 assays and the counts of risk alleles (0, 1, 2) of the T2D-associated SNPs rs7903145 and rs12255372 was tested with univariate linear regression model with adjustment for covariates (age, sex, and log-transformed values of BMI and blood levels of glucose and HbA1c). The analyses were performed with SPSS 16.0 (SPSS Inc.). Microarray expression was analyzed with R statistical programming package. Power analysis was performed with StatMate 2.0 (GraphPad) and was based on standard deviations in expression of each of assays, number of samples with 0 or 1 and 2 risk alleles at each SNP and 80% power to detect the difference with 5% of type I error.
Joint analysis of the microarray and TaqMan data, pathway enrichment analysis
The microarray data for 159 paired samples of omental and subcutaneous adipose tissue was a part of a previously described study . The microarray data was available to all authors of the article and is available upon request. Each of the RNA samples was profiled with custom-designed microarrays that included 39,280 oligonucleotide probes representing 34,266 known and predicted genes (Agilent Technologies, Palo Alto, CA). We used assays “ex3a-4”, “ex7-8”, ex”12-13” and “ex13-13a” of TCF7L2 for joint analysis of expression. Expression of B2M and GAPDH was used as an endogenous control for TaqMan TCF7L2 expression. However, when we combined the TaqMan and microarray data, we observed that the sets of transcripts that correlated with TCF7L2 expression significantly overlapped with the set of transcripts that correlated with GAPDH and B2M expression. To remove this biologically relevant variation component from the GAPDH and B2M expression, we identified a set of transcripts that correlated with the geometric mean of B2M and GAPDH expression at a 5% FDR (false-discovery rate). We then performed Principal Component Analysis on this set of transcripts to identify those factors (eigenvectors) that explained greater than 90% of the variation in the GAPDH/B2M geometric mean and adjusted the GAPDH/B2M expression by each of these eigenvectors. The adjusted B2M/GAPDH expression was then used to normalize TCF7L2 expression using standard linear regression methods. For co-expression analysis, the normalized expression values for all transcripts were adjusted for age and sex and the residuals were computed using the rlm function from the R statistical package (M-estimation with Tukey's bisquare weights). The Spearman correlation coefficients for expression of TCF7L2 and all other transcripts were calculated. The FDR for these correlations was calculated by performing 1000 permutations of sample IDs while preserving the correlation structure among the gene expression values. The threshold for Spearman coefficient was set up at r = +/−0.25, corresponding to FDR<1% (p-value<0.0005). The pathway enrichment analysis on sets of transcripts with significant correlations was performed using the Panther Classification System . For each GO category, we compared expected and observed frequencies between a set of 20,316 annotated transcripts presented on the array and the sets of transcripts significantly correlated with TCF7L2 expression. The p-value for significance of enrichment in each group was adjusted for the number of GO categories using Bonferroni method implemented in the Panther analysis .
Primers and probes for expression assays.
(0.04 MB RTF)
List of significant correlations between TCF7L2 assay.
(0.05 MB XLS)
List of significant correlations between TCF7L2 assay.
(0.27 MB XLS)
List of significant correlations between TCF7L2 assay.
(0.17 MB XLS)
List of significant correlations between TCF7L2 assay.
(0.11 MB XLS)
We are thankful to all patients participated in this study, Victor Castro at MGH for assistance in sample collection, Christine Suver and the Rosetta Gene Expression Laboratory for sample preparation.
Conceived and designed the experiments: LPO LMK EES FSC. Performed the experiments: LPO. Analyzed the data: LPO EES. Contributed reagents/materials/analysis tools: LMK EES. Wrote the paper: LPO EES FSC.
- 1. Grant SF, Thorleifsson G, Reynisdottir I, Benediktsson R, Manolescu A, et al. (2006) Variant of transcription factor 7-like 2 (TCF7L2) gene confers risk of type 2 diabetes. Nat Genet 38: 320–323.
- 2. Scott LJ, Bonnycastle LL, Willer CJ, Sprau AG, Jackson AU, et al. (2006) Association of transcription factor 7-like 2 (TCF7L2) variants with type 2 diabetes in a Finnish sample. Diabetes 55: 2649–2653.
- 3. Saxena R, Gianniny L, Burtt NP, Lyssenko V, Giuducci C, et al. (2006) 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.
- 4. Groves CJ, Zeggini E, Minton J, Frayling TM, Weedon MN, et al. (2006) Association analysis of 6,736 U.K. subjects provides replication and confirms TCF7L2 as a type 2 diabetes susceptibility gene with a substantial effect on individual risk. Diabetes 55: 2640–2644.
- 5. Reya T, Clevers H (2005) Wnt signalling in stem cells and cancer. Nature 434: 843–850.
- 6. He TC, Sparks AB, Rago C, Hermeking H, Zawel L, et al. (1998) Identification of c-MYC as a target of the APC pathway. Science 281: 1509–1512.
- 7. Muncan V, Sansom OJ, Tertoolen L, Phesse TJ, Begthel H, et al. (2006) Rapid loss of intestinal crypts upon conditional deletion of the Wnt/Tcf-4 target gene c-Myc. Mol Cell Biol 26: 8418–8426.
- 8. Tetsu O, McCormick F (1999) Beta-catenin regulates expression of cyclin D1 in colon carcinoma cells. Nature 398: 422–426.
- 9. Merrill BJ, Gat U, DasGupta R, Fuchs E (2001) Tcf3 and Lef1 regulate lineage differentiation of multipotent stem cells in skin. Genes Dev 15: 1688–1705.
- 10. Ross SE, Hemati N, Longo KA, Bennett CN, Lucas PC, et al. (2000) Inhibition of adipogenesis by Wnt signaling. Science 289: 950–953.
- 11. Miranda PJ, DeFronzo RA, Califf RM, Guyton JR (2005) Metabolic syndrome: definition, pathophysiology, and mechanisms. Am Heart J 149: 33–45.
- 12. Florez JC, Jablonski KA, Bayley N, Pollin TI, de Bakker PI, et al. (2006) TCF7L2 polymorphisms and progression to diabetes in the Diabetes Prevention Program. N Engl J Med 355: 241–250.
- 13. Cauchi S, Meyre D, Dina C, Choquet H, Samson C, et al. (2006) Transcription factor TCF7L2 genetic study in the French population: expression in human beta-cells and adipose tissue and strong association with type 2 diabetes. Diabetes 55: 2903–2908.
- 14. Cauchi S, Meyre D, Choquet H, Dina C, Born C, et al. (2006) TCF7L2 Variation Predicts Hyperglycemia Incidence in a French General Population: The Data From an Epidemiological Study on the Insulin Resistance Syndrome (DESIR) Study. Diabetes 55: 3189–3192.
- 15. Cauchi S, Choquet H, Gutierrez-Aguilar R, Capel F, Grau K, et al. (2008) Effects of TCF7L2 polymorphisms on obesity in European populations. Obesity (Silver Spring) 16: 476–482.
- 16. Zhang W, Duan S, Bleibel WK, Wisel SA, Huang RS, et al. (2008) Identification of common genetic variants that account for transcript isoform variation between human populations. Hum Genet.
- 17. Graham RR, Kozyrev SV, Baechler EC, Reddy MV, Plenge RM, et al. (2006) A common haplotype of interferon regulatory factor 5 (IRF5) regulates splicing and expression and is associated with increased risk of systemic lupus erythematosus. Nat Genet 38: 550–555.
- 18. Elbein SC, Chu WS, Das SK, Yao-Borengasser A, Hasstedt SJ, et al. (2007) Transcription factor 7-like 2 polymorphisms and type 2 diabetes, glucose homeostasis traits and gene expression in US participants of European and African descent. Diabetologia 50: 1621–1630.
- 19. Lyssenko V, Lupi R, Marchetti P, Del Guerra S, Orho-Melander M, et al. (2007) Mechanisms by which common variants in the TCF7L2 gene increase risk of type 2 diabetes. J Clin Invest 117: 2155–2163.
- 20. Osmark P, Hansson O, Jonsson A, Ronn T, Groop L, et al. (2009) Unique splicing pattern of the TCF7L2 gene in human pancreatic islets. Diabetologia.
- 21. Prokunina-Olsson L, Welch C, Hansson O, Adhikari N, Scott L, et al. (2009) Tissue-specific alternative splicing of TCF7L2. Hum Mol Genet.
- 22. http://hapmap.org/.
- 23. Zhong H, Beaulaurier J, Lum PY, Molony C, Yang X, et al. (2009) Elucidating Networks of eSNPs Associated with Type 2 Diabetes. PLOS Biology. in press.
- 24. Wegner L, Hussain MS, Pilgaard K, Hansen T, Pedersen O, et al. (2008) Impact of TCF7L2 rs7903146 on Insulin Secretion and Action in Young and Elderly Danish Twins. J Clin Endocrinol Metab.
- 25. Kovacs P, Berndt J, Ruschke K, Kloting N, Schon MR, et al. (2008) TCF7L2 gene expression in human visceral and subcutaneous adipose tissue is differentially regulated but not associated with type 2 diabetes mellitus. Metabolism 57: 1227–1231.
- 26. Fang M, Li J, Blauwkamp T, Bhambhani C, Campbell N, et al. (2006) C-terminal-binding protein directly activates and represses Wnt transcriptional targets in Drosophila. EMBO J 25: 2735–2745.
- 27. Thomas PD, Kejariwal A, Guo N, Mi H, Campbell MJ, et al. (2006) Applications for protein sequence-function evolution data: mRNA/protein expression analysis and coding SNP scoring tools. Nucleic Acids Res 34: W645–650.