Lack of association of the HMGA1 IVS5-13insC variant with type 2 diabetes in an ethnically diverse hypertensive case con...

MISSING IMAGE

Material Information

Title:
Lack of association of the HMGA1 IVS5-13insC variant with type 2 diabetes in an ethnically diverse hypertensive case control cohort
Physical Description:
Mixed Material
Language:
English
Creator:
Karnes, Jason H.
Langaee, Taimour
McDonough, Caitrin W.
Chang, Shin-Wen
Ramos, Miguel
Catlin, James R. Jr.
Publisher:
BioMed Central (Journal of Translational Medicine)
Publication Date:

Notes

Abstract:
Background: Recently, the high-mobility group A1 gene (HMGA1) variant IVS5-13insC has been associated with type 2 diabetes, but reported associations are inconsistent and data are lacking in Hispanic and African American populations. We sought to investigate the HMGA1-diabetes association and to characterize IVS5-13insC allele frequencies and linkage disequilibrium (LD) in 3,070 Caucasian, Hispanic, and African American patients from the INternational VErapamil SR-Trandolapril STudy (INVEST). Methods: INVEST was a randomized, multicenter trial comparing two antihypertensive treatment strategies in an ethnically diverse cohort of hypertensive, coronary artery disease patients. Controls, who were diabetes-free throughout the study, and type 2 diabetes cases, either prevalent or incident, were genotyped for IVS5-13insC using TaqmanW, confirmed with Pyrosequencing and Sanger sequencing. For LD analysis, genotyping for eight additional HMGA1 single nucleotide polymorphisms (SNPs) was performed using the IlluminaW HumanCVD BeadChip. We used logistic regression to test association of the HMGA1 IVS5-13insC and diabetes, adjusted for age, gender, body mass index, and percentage European, African, and Native American ancestry. Results: We observed IVS5-13insC minor allele frequencies consistent with previous literature in Caucasians and African Americans (0.03 in cases and 0.04 in controls for both race/ethnic groups), and higher frequencies in Hispanics (0.07 in cases and 0.07 in controls). The IVS5-13insC was not associated with type 2 diabetes overall (odds ratio 0.98 0.76-1.26, p=0.88) or in any race/ethnic group. Pairwise LD (r2) of IVS5-13insC and rs9394200, a SNP previously used as a tag SNP for IVS5-13insC, was low (r2=0.47 in Caucasians, r2=0.25 in Hispanics, and r2=0.06 in African Americans). Furthermore, in silico analysis suggested a lack of functional consequences for the IVS5-13insC variant. Conclusions: Our results suggest that IVS5-13insC is not a functional variant and not associated with type 2 diabetes in an ethnically diverse, hypertensive, coronary artery disease population. Larger, more adequately powered studies need to be performed to confirm our findings. Trial registration: clinicaltrials.gov (NCT00133692) Keywords: HMGA1, Type 2 diabetes, Genetics
General Note:
Publication of this article was funded in part by the University of Florida Open-Access publishing Fund. In addition, requestors receiving funding through the UFOAP project are expected to submit a post-review, final draft of the article to UF's institutional repository, IR@UF, (www.uflib.ufl.edu/UFir) at the time of funding. The institutional Repository at the University of Florida community, with research, news, outreach, and educational materials.
General Note:
Karnes et al. Journal of Translational Medicine 2013, 11:12 http://www.translational-medicine.com/content/11/1/12; Pages 1-6
General Note:
doi:10.1186/1479-5876-11-12 Cite this article as: Karnes et al.: Lack of association of the HMGA1 IVS5- 13insC variant with type 2 diabetes in an ethnically diverse hypertensive case control cohort. Journal of Translational Medicine 2013 11:12.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution.
System ID:
AA00014159:00001


This item is only available as the following downloads:


Full Text
!DOCTYPE art SYSTEM 'http:www.biomedcentral.comxmlarticle.dtd'
ui 1479-5876-11-12
ji 1479-5876
fm
dochead Research
bibl
title
p Lack of association of the it HMGA1 IVS5-13insC variant with type 2 diabetes in an ethnically diverse hypertensive case control cohort
aug
au id A1 snm Karnesmi Hfnm Jasoninsr iid I1 I2 email jason.h.karnes@vanderbilt.edu
A2 LangaeeYTaimourlangaee@cop.ufl.edu
A3 McDonoughWCaitrincaitrinmcdonough@ufl.edu
A4 ChangShin-Wenswchang58@gmail.com
A5 RamosMiguelmig29ram@ufl.edu
A6 Catlin JrRJamesjcatlin@ufl.edu
A7 CasanovaEOctavioocsacool@ufl.edu
A8 GongYangong@cop.ufl.edu
A9 PepineJCarlI3 Carl.Pepine@medicine.ufl.edu
A10 JohnsonAJuliejohnson@cop.ufl.edu
A11 ca yes Cooper-DeHoffMRhondadehoff@cop.ufl.edu
insg
ins Division of Clinical Pharmacology, Vanderbilt University, 1275 Medical Research Building IV, Nashville, TN, 37232-0575, USA
Department of Pharmacotherapy and Translational Research, University of Florida, HSC PO Box 100486, Gainesville, FL, 32610-0486, USA
Division of Cardiovascular Medicine, University of Florida, PO Box 100277, Gainesville, FL, 32610-0486, USA
source Journal of Translational Medicine
section Cardiovascular, Metabolic and Lipoprotein Translation issn 1479-5876
pubdate 2013
volume 11
issue 1
fpage 12
url http://www.translational-medicine.com/content/11/1/12
xrefbib pubidlist pubid idtype doi 10.1186/1479-5876-11-12pmpid 23302499
history rec date day 5month 10year 2012acc 712013pub 912013
cpyrt 2013collab Karnes et al.; licensee BioMed Central Ltd.note This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
kwdg
kwd HMGA1
Type 2 diabetes
Genetics
abs
sec
st
Abstract
Background
Recently, the high-mobility group A1 gene (HMGA1) variant IVS5-13insC has been associated with type 2 diabetes, but reported associations are inconsistent and data are lacking in Hispanic and African American populations. We sought to investigate the HMGA1-diabetes association and to characterize IVS5-13insC allele frequencies and linkage disequilibrium (LD) in 3,070 Caucasian, Hispanic, and African American patients from the INternational VErapamil SR-Trandolapril STudy (INVEST).
Methods
INVEST was a randomized, multicenter trial comparing two antihypertensive treatment strategies in an ethnically diverse cohort of hypertensive, coronary artery disease patients. Controls, who were diabetes-free throughout the study, and type 2 diabetes cases, either prevalent or incident, were genotyped for IVS5-13insC using Taqman®, confirmed with Pyrosequencing and Sanger sequencing. For LD analysis, genotyping for eight additional HMGA1 single nucleotide polymorphisms (SNPs) was performed using the Illumina® HumanCVD BeadChip. We used logistic regression to test association of the HMGA1 IVS5-13insC and diabetes, adjusted for age, gender, body mass index, and percentage European, African, and Native American ancestry.
Results
We observed IVS5-13insC minor allele frequencies consistent with previous literature in Caucasians and African Americans (0.03 in cases and 0.04 in controls for both race/ethnic groups), and higher frequencies in Hispanics (0.07 in cases and 0.07 in controls). The IVS5-13insC was not associated with type 2 diabetes overall (odds ratio 0.98 [0.76-1.26], p=0.88) or in any race/ethnic group. Pairwise LD (rsup 2) of IVS5-13insC and rs9394200, a SNP previously used as a tag SNP for IVS5-13insC, was low (r2=0.47 in Caucasians, r2=0.25 in Hispanics, and r2=0.06 in African Americans). Furthermore, in silico analysis suggested a lack of functional consequences for the IVS5-13insC variant.
Conclusions
Our results suggest that IVS5-13insC is not a functional variant and not associated with type 2 diabetes in an ethnically diverse, hypertensive, coronary artery disease population. Larger, more adequately powered studies need to be performed to confirm our findings.
Trial registration
clinicaltrials.gov (NCT00133692)
bdy
Background
Type 2 diabetes constitutes a major and growing health problem worldwide and is predicted to afflict 490 million by 2030 abbrgrp
abbr bid B1 1
. Type 2 diabetes has strong genetic influences and many polymorphisms have now been reproducibly associated with type 2 diabetes
B2 2
B3 3
. However, genome wide association studies (GWAS) explain only 10-15% of heritability and have not consistently improved diabetes risk prediction
B4 4
. Low frequency variation may account for much of the missing heritability in type 2 diabetes risk and may help translate genetic association study results into clinical type 2 diabetes risk prediction.Recently, the low frequency insertion polymorphism IVS5-13insC (c.136-14_136-13insC) in the high-mobility group A1 gene (HMGA1), a transcriptional regulator of the insulin receptor gene (INSR), was identified and associated with type 2 diabetes
B5 5
. Whereas two studies have observed a significant positive association between IVS5-13insC and type 2 diabetes in Caucasian and Chinese populations
5
B6 6
, another study in Caucasians observed no association
B7 7
. Data are lacking for IVS5-13insC in populations with African and Hispanic descent, which have disproportionately high type 2 diabetes prevalence
B8 8
. Conflicting results for an HMGA1 association with type 2 diabetes and the lack of data in diverse race/ethnic groups make clinical translation of the HMGA1 IVS5-13insC genotyping especially difficult.Evidence for the functional impact of the HMGA1 IVS5-13insC variant is also conflicting. One study observed that HMGA1 and INSR expression was decreased in diabetic carriers of IVS5-13insC versus wild type diabetic and non-diabetic patients
5
. Additionally, INSR protein expression and insulin-binding capacity was restored in lymphoblasts obtained from diabetic IVS5-13insC carriers by HMGA1 DNA transfection. Another study observed no effect of IVS5-13insC on HMGA1 or INSR expression in adipose tissue of normoglycemic patients
7
. IVS5-13insC occurs at position −13 of HMGA1 exon 6, but the direct mechanism of the variant’s effects on mRNA expression or amino acid sequence remains unclear.We tested the association of HMGA1 IVS5-13insC with type 2 diabetes in an ethnically diverse population from the INternational VErapamil SR-Trandolapril STudy (INVEST). INVEST compared CV outcomes and NOD in hypertensive coronary artery disease patients treated with two antihypertensive treatment strategies. We also determined minor allele frequencies (MAF) and linkage disequilibrium (LD) for HMGA1 variants and tested the functional impact of IVS5-13insC in silico.
Methods
Study design and participants
INVEST compared CV outcomes and incident diabetes in hypertensive, coronary artery disease patients at least 50 years of age during randomized treatment with either an atenolol-based or a verapamil sustained release (SR)-based antihypertensive treatment strategy. The design, primary outcome, and NOD results have been previously published in detail
B9 9
B10 10
B11 11
B12 12
. Briefly, the verapamil SR strategy consisted of stepped therapy with verapamil SR, trandolapril add-on, dose titration, then HCTZ add-on treatment for BP control and end organ protection as necessary. The atenolol-based strategy consisted of atenolol, HCTZ add-on, dose titration, then trandolapril add-on treatment as necessary. The INVEST GENEtic Substudy (INVEST-GENES) collected DNA samples from 5,979 INVEST patients at 187 sites in the United States and Puerto Rico.We conducted a nested case control study including cases with type 2 diabetes at baseline (prevalent diabetes) or type 2 diabetes that developed during a mean 2.8 years follow-up (incident diabetes). Type 2 diabetes was determined by patient report and by site investigators from a review of all available patient data, including use of diabetic medication and lab measures
11
. We identified age, gender, and race/ethnicity-matched controls who remained diabetes-free over a mean 2.8 years follow-up. Age matching was performed after stratification by decade and we attempted to match cases and controls in a 1:1 ratio. The institutional review boards of participating study centers approved the study protocol and all patients provided written informed consent for participation in INVEST and additional written informed consent for genetic studies. INVEST is registered at clinicaltrials.gov (NCT00133692).
Genotyping
Genotyping for HMGA1 IVS5-13insC was performed using TaqMan® (Applied Biosystems, Foster City, CA, USA) with PCR primers and probe for IVS5-13insC (PN4331349) purchased from Applied Biosystems. For IVS5-13insC genotyping quality control, 5% of samples were genotyped in duplicate on Taqman®. A total of 612 Taqman genotypes were confirmed using pyrosequencing (Biotage AB, Uppsala, Sweden), using the following PCR and sequencing primers respectively: forward-biotinylated-5′-GGGGTGGAAACAGGTGATG-3′, reverse-5′-CACTTCGCTGGGCTCCTT-3′, and reverse-5′-TTCTGTAAAGACAGAGG-3′. Sanger sequencing was used to genotype 58 samples that showed discrepancies between Taqman® and Pyrosequencing platforms.Genotyping for eight additional HMGA1 single nucleotide polymorphisms (SNPs) was performed using the HumanCVD BeadChip and Infinium II Assay (Illumina, San Diego, CA) on 1,489 INVEST patients to perform LD analyses. The HumanCVD BeadChip contains approximately 50,000 cosmopolitan tag SNPs for 2,100 CV and metabolic-related genes
B13 13
. HumanCVD BeadChip data quality was ensured in PLINK using genotype and sample call rates, concordance rates for blind duplicates, gender confirmation, cryptic relatedness using pairwise identity-by-descent, and estimation of heterozygosity using the inbreeding coefficient F
B14 14
. Individuals were excluded if call rates were below 90 percent and SNPs were excluded if call rates were below 95 percent. In addition, 87 ancestry informative markers were genotyped in 2,860 INVEST patients to estimate Caucasian, African, and Native American ancestry using STRUCTURE
B15 15
. Race/ethnic groups were determined by patient self-report with interaction by the study investigator
B16 16
and confirmed using principal components analysis generated from LD-pruned HumanCVD BeadChip data and ancestry informative markers.
Statistical analysis
All statistical analyses were performed using SAS version 9.2 (SAS, Cary, NC). Differences in patient characteristics comparing cases and controls at baseline were determined using t-tests and chi square tests, as appropriate. Deviations from Hardy Weinberg Equilibrium were assessed using a chi square test. Multi-variable logistic regressions were performed overall and by race/ethnic group to determine odds ratios (ORs) and 95% confidence intervals (95%CIs) for type 2 diabetes in IVS5-13insC variant carriers versus non-variant carriers. Variables for adjustment included age, gender, and body mass index (BMI) in order to maintain consistency with previously published HMGA1 analyses
5
7
. Percentage of Caucasian, African, and Native American ancestry, as estimated by ancestry informative markers, was included as a variable for adjustment in analyses where all race/ethnic groups were combined.The combined race/ethnic group analysis was considered primary with alpha=0.05, since IVS5-13insC was presumed to be a functional SNP with similar consequences across race/ethnic groups, based on previously published reports
5
7
. Assuming a MAF of 0.05 and OR of 1.40, we had 84% power to detect an association between the IVS5-13insC variant and type 2 diabetes in a dominant model in the overall population. Assuming a MAF of 0.05 and OR of 1.40, we had 61% power in Hispanics, 44% power in whites, and 18% power in blacks to detect an association by race/ethnicity. Pairwise LD values (r2) and LD plots were generated by race/ethnic group using Haploview
B17 17
. We predicted in silico functional consequences of the IVS5-13insC variant using SNPNexus
B18 18
and ESEfinder 3.0
B19 19
.
Results
We identified 446 incident type 2 diabetes cases over a mean 2.8 years of follow-up and genotyped an additional 1329 prevalent type 2 diabetes cases in INVEST-GENES. At baseline, patients with incident or prevalent type 2 diabetes had higher BMI and lower diastolic BP versus age, race/ethnicity, and gender-matched controls (Table tblr tid T1 1). Cases also had a higher prevalence of hypercholesterolemia, left ventricular hypertrophy, and congestive heart failure.
table
Table 1
caption
b Characteristics of type 2 diabetes cases and controls at baseline
tgroup align left cols 4
colspec colname c1 colnum 1 colwidth 1*
c2 2
c3 3
c4
thead valign top
row rowsep
entry
Characteristic*
Type 2 diabetes cases (n=1775)
Controls (n=1295)
p
value

tfoot
BMI indicates body mass index; LVH, left ventricular hypertrophy; CHF, congestive heart failure.*Values are mean ± standard deviation unless otherwise noted. †P values represent t-tests and chi square tests where appropriate. ‡History of or currently taking lipid-lowering medications. **New York Heart Association Class I-III.
tbody
Age (years)
65.8 (9.2)
65.7 (9.1)
0.93
Female, n (%)
1,014 (57)
731 (56)
0.71
BMI (kg/m2)
30.6 (5.6)
28.9 (5.4)
<0.0001
nameend namest
Race/ethnicity, n (%)
0.15
Caucasian
608 (34)
493 (38)
Hispanic
937 (53)
642 (50)
African American
216 (12)
153 (12)
Blood pressure (mm Hg)
Systolic
149 (19)
148 (18)
0.32
Diastolic
85 (11)
86 (10)
0.0008
Hypercholesterolemia, n (%)‡
1,017 (57)
673 (52)
0.003
History of LVH, n (%)
319 (18)
174 (13)
0.0007
History of CHF, n (%)**
84 (5)
25 (2)
<0.0001
History of smoking, n (%)
706 (40)
507 (39)
0.73
The IVS5-13insC variant did not deviate from HWE in any race/ethnic group (Table T2 2). All HMGA1 SNPs had call rates above 95 percent and no SNPs failed quality control based on call rate, Hardy Weinberg equilibrium testing, concordance with blind duplicates, or estimation of heterozygosity using the inbreeding coefficient F. The genomic inflation factor lambda was 1.03 for INVEST individuals genotyped for the HMGA1 IVS5-13insC variant, suggesting minimal population stratification in genotyped individuals. For the IVS5-13insC variant, concordance of duplicates on Taqman® was 99% and concordance between Pyrosequencing and Taqman® was 95%. In Caucasians, the MAF of IVS5-13insC was 0.03 in diabetic cases and 0.04 in controls. The frequency of IVS5-13insC was similar in African American individuals (0.03 in cases and 0.04 in controls) and highest in Hispanics (0.07 in cases and 0.07 in controls). No significant associations were observed between the IVS5-13insC variant and diabetes overall (OR 0.98 95%CI 0.76-1.26, p=0.88). The IVS5-13insC variant was not associated with diabetes in any race/ethnic group (Table 2).
Table 2
IVS5-13insC genotype frequencies and associations with diabetes overall and by race/ethnicity
8
c5 5
c6 6
c7 7
c8
Unadjusted
Adjusted
95%CI indicates 95% confidence interval; MAF, minor allele frequency, HWE, Hardy Weinberg Equilibrium.*Hardy Weinberg Equilibrium p value calculated using a chi square test by race/ethnicity in non-diabetic patients. †Generated using logistic regression in a dominant model adjusted for age, gender, body mass index, and Caucasian, African, and Native American ancestry as estimated by ancestry informative markers.
Race/ethnic group
MAF (cases)
MAF (controls)
HWE p value*
Odds ratio (95%CI)
p value
Odds ratio (95%CI)

p value

Overall
0.053
0.052
-
1.00 (0.79-1.26)
0.98
0.98 (0.76-1.26)
0.88
(n=3070)
Caucasian
0.028
0.036
0.40
0.75 (0.46-1.22)
0.25
0.95 (0.44-2.06)
0.90
(n=1101)
Hispanic
0.074
0.070
0.20
1.09 (0.82-1.45)
0.57
0.79 (0.49-1.25)
0.31
(n=1579)
African American (n=369)
0.030
0.039
0.10
0.83 (0.36-1.90)
0.65
1.51 (0.48-4.74)
0.48
Pairwise LD of HMGA1 SNPs by race/ethnicity is presented in Figure figr fid F1 1. Pairwise LD (r2) of IVS5-13insC and rs9394200, the SNP previously used as a tag SNP for IVS5-13insC
7
, was 0.47 in Caucasians, 0.25 in Hispanics, 0.06 in African Americans. In silico functional analysis of IVS5-13insC revealed no predicted changes to amino acid sequence, conserved transcription factor binding sites, CpG islands, or miRNA regulatory sites. IVS5-13insC was not predicted to cause the creation or disruption of splice sites. However, in silico evaluation of exonic splice enhancer (ESE) sites revealed that the C insertion of IVS5-13insC creates an additional ESE motif for SF2/ASF (IgM-BRCA1).
fig Figure 1Haploview-generated linkage disequilibrium (LD) plot of HMGA1 variation in INVEST Caucasians (A), Hispanics (B), and African Americans (C)text
Haploview-generated linkage disequilibrium (LD) plot of HMGA1 variation in INVEST Caucasians (A), Hispanics (B), and African Americans (C). Regions of higher LD are shaded darker according to higher r2 values and pairwise r2 values are indicated in each box.
graphic file 1479-5876-11-12-1
Discussion
In the present study, we observed that the HMGA1 IVS5-13insC variant was not associated with diabetes in the INVEST-GENES population overall or in any race/ethnic group. We observed a frequency of IVS5-13insC consistent with previously published studies in Caucasians
5
7
and observed an increased frequency of this variant in Hispanics. Furthermore, pairwise LD in our population suggests that IVS5-13insC is not effectively tagged by rs9394200 in any race/ethnic group and in silico analysis revealed minimal evidence of putative functional consequences of the IVS5-13insC variant.Our observations extend existing knowledge of genetic determinants of type 2 diabetes and are consistent with the findings of Marquez et al., which suggest a lack of association between IVS5-13insC and type 2 diabetes
7
. Marquez et al. also found that the IVS5-13insC variant is not functional, which is supported by our lack of observed association in multiple race/ethnic groups and our in silico functional results which predicted no important consequences from this variability. Our findings are inconsistent with two studies that observed a significant association between IVS5-13insC and type 2 diabetes
5
6
. This discrepancy may be due to differences in race/ethnicity of our population compared with other studies. However, a functional IVS5-13insC variant would be expected to have similar effects across multiple race/ethnicity groups.Although our association analysis is consistent with the findings of Marquez et al.
7
, our observation of low LD between IVS5-13insC and rs9394200 in all race/ethnic groups suggests that rs9394200 is not an appropriate tag SNP for IVS5-13insC. In Hapmap, the MAF of the T allele for rs9394200 is similar to the MAF for IVS5-13insC in Caucasians (0.03), but is 0.45 in Yorubans, further suggesting that rs9394200 does not adequately tag IVS5-13insC, especially in non-Caucasian populations. Furthermore, rs9394200 is 5000 base pairs downstream from the of HMGA1 3′ end and is represented on arrays utilized by type 2 diabetes GWAS
3
B20 20
. Therefore, if rs9394200 were a significant contributor to risk for diabetes, it likely would have been identified in the GWAS analyses.Our observation of a lack of association in Hispanics and African Americans suggests that IVS5-13insC may be associated with diabetes only in Caucasian individuals. While we did not observe even a trend towards association in our Caucasian population, we acknowledge that our power to observe an association in Caucasians is lower than in the previously published studies. Although we had 84 percent power to detect an association in our overall population, we had inadequate power to definitively conclude an association in by race/ethnicity analyses.Our in silico analysis did not reveal any direct mechanism of the IVS5-13insC variant’s effects on mRNA expression or amino acid sequence, suggesting that the variant is not likely to be functional. Although the results of Chiefari et al. suggested a functional effect of the IVS5-13insC variant on HMGA1 and INSR mRNA and protein expression in monocytes
5
, Marquez et al. found no effect of the variant on mRNA expression in adipose tissue. The differences in observed effect on mRNA expression may potentially be explained by differential effects of the variant on transcription in monocytes and adipose tissue. The apparent functional effect of IVS5-13insC observed by Chiefari et al. may also be confounded by the lack of evaluation of IVS5-13insC variant carriers versus non-carriers among non-diabetic controls
7
or potential treatment with glucose-lowering medications in diabetic patients from whom monocytes were collected
B21 21
. Finally, IVS5-13insC may have an unknown functional mechanism not identified by in silico tools used in this study. Our study has several limitations worthy of mention. We recognize the potential for false negative results in our analyses, especially in African Americans, considering the low frequency of the variant and the limited power to detect associations within each race/ethic group. Our observations require replication in independent populations of similar race/ethnic makeup. In INVEST, incident and prevalent diabetes diagnosis was based on investigator reports, but the diabetes phenotype is well described in a previous publication
11
, the accuracy of such reporting has been verified by others
B22 22
, and has been used in other trials
B23 23
B24 24
B25 25
. INVEST investigator-reported diabetes phenotypes have also been used in a large scale gene-centric meta-analysis with HumanCVD BeadChip data
B26 26
, suggesting validity with regard to the diabetes phenotype and genetic association analysis. Although our HMGA1 association analysis may be confounded if control patients eventually develop diabetes after study follow-up, the high mean age and lower mean BMI suggests that our control patients are less likely to develop type 2 diabetes. In addition, although concordance within the Taqman® genotyping platform was high, our concordance between Taqman® and Pyrosequencing was 95%, suggesting some disagreement between platforms. However, Hardy Weinberg tests did not indicate genotype error and genotype error was minimized by utilization of Sanger sequencing for discrepancy confirmation.
Conclusions
Our results suggest that the HMGA1 IVS5-13insC is not associated with type 2 diabetes and may not have an important functional role in diabetes pathogenesis. We also provide frequency and LD data for Hispanic and African American populations, which have higher prevalence of type 2 diabetes. Our results also suggest that rs9394200 is not an effective tag SNP for HMGA1 IVS5-13insC, especially in non-Caucasian populations. Functional studies and replication of these associations are needed to better define the potential role of HMGA1 variants in predicting type 2 diabetes development. Although the current study suggests the lack of a functional role for IVS5-13insC, further study of HMGA1 is warranted to clarify the role of this gene in diabetes pathogenesis. Larger, more adequately powered studies need to be performed to confirm our findings.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
JHK drafted the manuscript and performed statistical analysis. TYL, JRC, MR, OEC, and SWC performed genotyping and helped draft the manuscript. CWM and YG performed HumanCVD BeadChip quality control procedures, contributed to statistical analysis and helped draft the manuscript. CJP, JAJ, and RCD conceived of the study, participated in its design and coordination, and helped draft the manuscript. All authors read and approved the final manuscript.
bm
ack
Acknowledgements
This study was funded by NIH grant U01 GM074492, R01 HL74730, HL086558 (RCD), and TL1RR029888 (JHK), and grants from Abbott Laboratories and the University of Florida Opportunity Fund. We thank Ben Burkley, Lynda Stauffer, and Cheryl Galloway for processing and genotyping samples.
refgrp Global healthcare expenditure on diabetes for 2010 and 2030ZhangPZhangXBrownJVistisenDSicreeRShawJNicholsGDiabetes Res Clin Pract201087293lpage 30110.1016/j.diabres.2010.01.026link fulltext 20171754Genetics of type 2 diabetes: pathophysiologic and clinical relevanceHerderCRodenMEur J Clin Invest20114167969210.1111/j.1365-2362.2010.02454.x21198561Twelve type 2 diabetes susceptibility loci identified through large-scale association analysisVoightBFScottLJSteinthorsdottirVMorrisAPDinaCWelchRPZegginiEHuthCAulchenkoYSThorleifssonGetal Nat Genet20104257958910.1038/ng.609pmcid 308065820581827Utility of genetic and non-genetic risk factors in prediction of type 2 diabetes: Whitehall II prospective cohort studyTalmudPJHingoraniADCooperJAMarmotMGBrunnerEJKumariMKivimakiMHumphriesSEBMJ2010340b483810.1136/bmj.b4838280694520075150Functional variants of the HMGA1 gene and type 2 diabetes mellitusChiefariETanyolacSPaonessaFPullingerCRCapulaCIiritanoSMazzaTForlinMFuscoADurlachVJAMA201130590391210.1001/jama.2011.20721364139Polymorphism of HMGA1 is associated with increased risk of type 2 diabetes among Chinese individualsLiuLDingHWangHRXuYJCuiGLWangPHYuanGYuXFWangDWDiabetologia2012551685168810.1007/s00125-012-2518-022411136Low-frequency variants in HMGA1 are not associated with type 2 diabetes riskMarquezMHuyvaertMPerryJRPearsonRDFalchiMMorrisAPVivequinSLobbensSYengoLGagetSDiabetes20126152453010.2337/db11-072822210315Executive Summary: Heart Disease and Stroke Statistics–2012 Update: A Report From the American Heart AssociationRogerVLGoASLloyd-JonesDMBenjaminEJBerryJDBordenWBBravataDMDaiSFordESFoxCSCirculation201212518819722215894Rationale and design of the International Verapamil SR/Trandolapril Study (INVEST): an Internet-based randomized trial in coronary artery disease patients with hypertensionPepineCJHandberg-ThurmondEMarksRGConlonMCooper-DeHoffRVolkersPZelligPJ Am Coll Cardiol1998321228123710.1016/S0735-1097(98)00423-99809930A calcium antagonist vs a non-calcium antagonist hypertension treatment strategy for patients with coronary artery disease. The International Verapamil-Trandolapril Study (INVEST): a randomized controlled trialPepineCJHandbergEMCooper-DeHoffRMMarksRGKoweyPMesserliFHManciaGCangianoJLGarcia-BarretoDKeltaiMJAMA20032902805281610.1001/jama.290.21.280514657064Predictors of development of diabetes mellitus in patients with coronary artery disease taking antihypertensive medications (findings from the INternational VErapamil SR-Trandolapril STudy [INVEST])Cooper-DehoffRCohenJDBakrisGLMesserliFHErdineSHewkinACKupferSPepineCJAm J Cardiol20069889089410.1016/j.amjcard.2006.04.03016996868INVEST revisited: review of findings from the International Verapamil SR-Trandolapril StudyCooper-DeHoffRMHandbergEMManciaGZhouQChampionALeglerUFPepineCJExpert Rev Cardiovasc Ther200971329134010.1586/erc.09.102280079019900016Concept, design and implementation of a cardiovascular gene-centric 50 k SNP array for large-scale genomic association studiesKeatingBJTischfieldSMurraySSBhangaleTPriceTSGlessnerJTGalverLBarrettJCGrantSFFarlowDNPLoS One20083e358310.1371/journal.pone.0003583257199518974833PLINK: a tool set for whole-genome association and population-based linkage analysesPurcellSNealeBTodd-BrownKThomasLFerreiraMABenderDMallerJSklarPde BakkerPIDalyMJShamPCAm J Hum Genet20078155957510.1086/519795195083817701901Inference of population structure using multilocus genotype dataPritchardJKStephensMDonnellyPGenetics2000155945959146109610835412Blood pressure control and cardiovascular outcomes in high-risk Hispanic patients–findings from the International Verapamil SR/Trandolapril Study (INVEST)Cooper-DeHoffRMArandaJMsuf JrGaxiolaECangianoJLGarcia-BarretoDContiCRHewkinAPepineCJAm Heart J20061511072107910.1016/j.ahj.2005.05.02416644338Haploview: analysis and visualization of LD and haplotype mapsBarrettJCFryBMallerJDalyMJBioinformatics20052126326510.1093/bioinformatics/bth45715297300SNPnexus: a web database for functional annotation of newly discovered and public domain single nucleotide polymorphismsChelalaCKhanALemoineNRBioinformatics20092565566110.1093/bioinformatics/btn653264783019098027ESEfinder: A web resource to identify exonic splicing enhancersCartegniLWangJZhuZZhangMQKrainerARNucleic Acids Res2003313568357110.1093/nar/gkg61616902212824367Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controlsBurtonPRClaytonDGCardonLRCraddockNDeloukasPDuncansonAKwiatkowskiDPMcCarthyMIOuwehandWHSamaniNJNature200744766167810.1038/nature05911271928817554300Response to comment on: Marquez et al. Low-frequency variants in HMGA1 are not associated with type 2 diabetes risk. Diabetes 2012;61:524–530FroguelPMarquezMCauchiSDiabetes201261e1510.2337/db12-080022923658Evaluating the quality of self-reports of hypertension and diabetesGoldmanNLinIFWeinsteinMLinYHJ Clin Epidemiol20035614815410.1016/S0895-4356(02)00580-212654409Newly diagnosed and previously known diabetes mellitus and 1-year outcomes of acute myocardial infarction: the VALsartan In Acute myocardial iNfarcTion (VALIANT) trialAguilarDSolomonSDKoberLRouleauJLSkaliHMcMurrayJJFrancisGSHenisMO’ConnorCMDiazRCirculation20041101572157810.1161/01.CIR.0000142047.28024.F215364810Long-term effects of ramipril on cardiovascular events and on diabetes: results of the HOPE study extensionBoschJLonnEPogueJArnoldJMDagenaisGRYusufSCirculation20051121339134616129815Ramipril and the development of diabetesYusufSGersteinHHoogwerfBPogueJBoschJWolffenbuttelBHZinmanBJAMA20012861882188510.1001/jama.286.15.188211597291Large-scale gene-centric meta-analysis across 39 studies identifies type 2 diabetes lociSaxenaRElbersCCGuoYPeterIGauntTRMegaJLLanktreeMBTareACastilloBALiYRAm J Hum Genet20129041042510.1016/j.ajhg.2011.12.022330918522325160



PAGE 1

RESEARCHOpenAccessLackofassociationofthe HMGA1 IVS5-13insC variantwithtype2diabetesinanethnically diversehypertensivecasecontrolcohortJasonHKarnes1,2,TaimourYLangaee2,CaitrinWMcDonough2,Shin-WenChang2,MiguelRamos2, JamesRCatlinJr2,OctavioECasanova2,YanGong2,CarlJPepine3,JulieAJohnson2,3and RhondaMCooper-DeHoff2,3*AbstractBackground: Recently,thehigh-mobilitygroupA1gene( HMGA1 )variantIVS5-13insChasbeenassociatedwith type2diabetes,butreportedassociationsareinconsistentanddataarelackinginHispanicandAfricanAmerican populations.Wesoughttoinvestigatethe HMGA1 -diabetesassociationandtocharacterizeIVS5-13insCallele frequenciesandlinkagedisequilibrium(LD)in3,070Caucasian,Hispanic,andAfricanAmericanpatientsfromthe INternationalVErapamilSR-TrandolaprilSTudy(INVEST). Methods: INVESTwasarandomized,multicentertrialcomparingtwoantihypertensivetreatmentstrategiesinan ethnicallydiversecohortofhypertensive,coronaryarterydiseasepatients.Controls,whowerediabetes-free throughoutthestudy,andtype2diabetescases,eitherprevalentorincident,weregenotypedforIVS5-13insC usingTaqmanW,confirmedwithPyrosequencingandSangersequencing.ForLDanalysis,genotypingforeight additional HMGA1 singlenucleotidepolymorphisms(SNPs)wasperformedusingtheIlluminaWHumanCVD BeadChip.Weusedlogisticregressiontotestassociationofthe HMGA1 IVS5-13insCanddiabetes,adjustedforage, gender,bodymassindex,andpercentageEuropean,African,andNativeAmericanancestry. Results: WeobservedIVS5-13insCminorallelefrequenciesconsistentwithpreviousliteratureinCaucasiansand AfricanAmericans(0.03incasesand0.04incontrolsforbothrace/ethnicgroups),andhigherfrequenciesinHispanics (0.07incasesand0.07incontrols).TheIVS5-13insCwasnotassociatedwithtype2diabetesoverall(oddsratio0.98 [0.76-1.26],p=0.88)orinanyrace/ethnicgroup.PairwiseLD(r2)ofIVS5-13insCandrs9394200,aSNPpreviouslyusedas atagSNPforIVS5-13insC,waslow(r2=0.47inCaucasians,r2=0.25inHispanics,andr2=0.06inAfricanAmericans). Furthermore, insilico analysissuggestedalackoffunctionalconsequencesfortheIVS5-13insCvariant. Conclusions: OurresultssuggestthatIVS5-13insCisnotafunctionalvariantandnotassociatedwithtype2diabetesin anethnicallydiverse,hypertensive,coronaryarterydiseasepopulation.Larger,moreadequatelypoweredstudiesneed tobeperformedtoconfirmourfindings. Trialregistration: clinicaltrials.gov(NCT00133692) Keywords: HMGA1,Type2diabetes,Genetics *Correspondence: dehoff@cop.ufl.edu2DepartmentofPharmacotherapyandTranslationalResearch,Universityof Florida,HSCPOBox100486,Gainesville,FL32610-0486,USA3DivisionofCardiovascularMedicine,UniversityofFlorida,POBox100277, Gainesville,FL32610-0486,USA Fulllistofauthorinformationisavailableattheendofthearticle 2013Karnesetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and reproductioninanymedium,providedtheoriginalworkisproperlycited.Karnes etal.JournalofTranslationalMedicine 2013, 11 :12 http://www.translational-medicine.com/content/11/1/12

PAGE 2

BackgroundType2diabetesconstitutesamajorandgrowinghealth problemworldwideandispredictedtoafflict490million by2030[1].Type2diabeteshasstronggeneticinfluences andmanypolymorphismshavenowbeenreproducibly associatedwithtype2diabetes[2,3].However,genome wideassociationstudies(GWAS)explainonly10-15%of heritabilityandhavenotconsistentlyimproveddiabetes riskprediction[4].Lowfrequencyvariationmayaccount formuchofthemissingheritabilityintype2diabetesrisk andmayhelptranslategeneticassociationstudyresults intoclinicaltype2diabetesriskprediction. Recently,thelowfrequencyinsertionpolymorphism IVS5-13insC(c.136-14_136-13insC)inthehigh-mobility groupA1gene( HMGA1 ),atranscriptionalregulatorof theinsulinreceptorgene( INSR ),wasidentifiedandassociatedwithtype2diabetes[5].Whereastwostudieshave observedasignificantpositiveassociationbetweenIVS513insCandtype2diabetesinCaucasianandChinese populations[5,6],anotherstudyinCaucasiansobserved noassociation[7].DataarelackingforIVS5-13insCin populationswithAfricanandHispanicdescent,which havedisproportionatelyhightype2diabetesprevalence [8].Conflictingresultsforan HMGA1 associationwith type2diabetesandthelackofdataindiverserace/ethnic groupsmakeclinicaltranslationofthe HMGA1 IVS513insCgenotypingespeciallydifficult. Evidenceforthefunctionalimpactofthe HMGA1 IVS513insCvariantisalsoconflicting.Onestudyobservedthat HMGA1 and INSR expressionwasdecreasedindiabetic carriersofIVS5-13insCversuswildtypediabeticandnondiabeticpatients[5].Additionally, INSR proteinexpression andinsulin-bindingcapacitywasrestoredinlymphoblasts obtainedfromdiabeticIVS5-13insCcarriersby HMGA1 DNAtransfection.Anotherstudyobservednoeffectof IVS5-13insCon HMGA1 or INSR expressioninadipose tissueofnormoglycemicpatients[7].IVS5-13insCoccurs atposition 13of HMGA1 exon6,butthedirectmechanismofthevariant ’ seffectsonmRNAexpressionoramino acidsequenceremainsunclear. Wetestedtheassociationof HMGA1 IVS5-13insCwith type2diabetesinanethnicallydiversepopulationfrom theINternationalVErapamilSR-TrandolaprilSTudy(INVEST).INVESTcomparedCVoutcomesandNODin hypertensivecoronaryarterydiseasepatientstreatedwith twoantihypertensivetreatmentstrategies.Wealsodeterminedminorallelefrequencies(MAF)andlinkagedisequilibrium(LD)for HMGA1 variantsandtestedthe functionalimpactofIVS5-13insC insilico .MethodsStudydesignandparticipantsINVESTcomparedCVoutcomesandincidentdiabetes inhypertensive,coronaryarterydiseasepatientsatleast 50yearsofageduringrandomizedtreatmentwitheither anatenolol-basedoraverapamilsustainedrelease(SR)basedantihypertensivetreatmentstrategy.Thedesign, primaryoutcome,andNODresultshavebeenpreviously publishedindetail[9-12].Briefly,theverapamilSRstrategyconsistedofsteppedtherapywithverapamilSR, trandolapriladd-on,dosetitration,thenHCTZadd-on treatmentforBPcontrolandendorganprotectionas necessary.Theatenolol-basedstrategyconsistedofatenolol,HCTZadd-on,dosetitration,thentrandolapril add-ontreatmentasnecessary.TheINVESTGENEtic Substudy(INVEST-GENES)collectedDNAsamples from5,979INVESTpatientsat187sitesintheUnited StatesandPuertoRico. Weconductedanestedcasecontrolstudyincluding caseswithtype2diabetesatbaseline(prevalentdiabetes)ortype2diabetesthatdevelopedduringamean 2.8yearsfollow-up(incidentdiabetes).Type2diabetes wasdeterminedbypatientreportandbysiteinvestigatorsfromareviewofallavailablepatientdata,including useofdiabeticmedicationandlabmeasures[11].We identifiedage,gender,andrace/ethnicity-matchedcontrolswhoremaineddiabetes-freeoveramean2.8years follow-up.Agematchingwasperformedafterstratificationbydecadeandweattemptedtomatchcasesand controlsina1:1ratio.Theinstitutionalreviewboardsof participatingstudycentersapprovedthestudyprotocol andallpatientsprovidedwritteninformedconsentfor participationinINVESTandadditionalwritteninformed consentforgeneticstudies.INVESTisregisteredat clinicaltrials.gov(NCT00133692).GenotypingGenotypingfor HMGA1 IVS5-13insCwasperformed usingTaqManW(AppliedBiosystems,FosterCity,CA, USA)withPCRprimersandprobeforIVS5-13insC (PN4331349)purchasedfromAppliedBiosystems.For IVS5-13insCgenotypingqualitycontrol,5%ofsamples weregenotypedinduplicateonTaqmanW.Atotalof612 Taqmangenotypeswereconfirmedusingpyrosequencing (BiotageAB,Uppsala,Sweden),usingthefollowingPCR andsequencingprimersrespectively:forward-biotinylated-5 -GGGGTGGAAACAGGTGATG-3 ,reverse-5 -C ACTTCGCTGGGCTCCTT-3 ,andreverse-5 -TTCTGT AAAGACAGAGG-3 .Sangersequencingwasusedto genotype58samplesthatshoweddiscrepanciesbetween TaqmanWandPyrosequencingplatforms. Genotypingforeightadditional HMGA1 singlenucleotidepolymorphisms(SNPs)wasperformedusingthe HumanCVDBeadChipandInfiniumIIAssay(Illumina, SanDiego,CA)on1,489INVESTpatientstoperform LDanalyses.TheHumanCVDBeadChipcontainsapproximately50,000cosmopolitantagSNPsfor2,100CV andmetabolic-relatedgenes[13].HumanCVDBeadChipKarnes etal.JournalofTranslationalMedicine 2013, 11 :12Page2of6 http://www.translational-medicine.com/content/11/1/12

PAGE 3

dataqualitywasensuredinPLINKusinggenotypeand samplecallrates,concordanceratesforblindduplicates, genderconfirmation,crypticrelatednessusingpairwise identity-by-descent,andestimationofheterozygosity usingtheinbreedingcoefficientF[14].Individualswere excludedifcallrateswerebelow90percentandSNPs wereexcludedifcallrateswerebelow95percent.In addition,87ancestryinformativemarkersweregenotyped in2,860INVESTpatientstoestimateCaucasian,African, andNativeAmericanancestryusingSTRUCTURE[15]. Race/ethnicgroupsweredeterminedbypatientself-report withinteractionbythestudyinvestigator[16]andconfirmedusingprincipalcomponentsanalysisgenerated fromLD-prunedHumanCVDBeadChipdataandancestry informativemarkers.StatisticalanalysisAllstatisticalanalyseswereperformedusingSASversion 9.2(SAS,Cary,NC).Differencesinpatientcharacteristicscomparingcasesandcontrolsatbaselineweredeterminedusingt-testsandchisquaretests,asappropriate. DeviationsfromHardyWeinbergEquilibriumwere assessedusingachisquaretest.Multi-variablelogistic regressionswereperformedoverallandbyrace/ethnic grouptodetermineoddsratios(ORs)and95%confidenceintervals(95%CIs)fortype2diabetesinIVS513insCvariantcarriersversusnon-variantcarriers.Variablesforadjustmentincludedage,gender,andbody massindex(BMI)inordertomaintainconsistencywith previouslypublished HMGA1 analyses[5,7].Percentage ofCaucasian,African,andNativeAmericanancestry,as estimatedbyancestryinformativemarkers,wasincluded asavariableforadjustmentinanalyseswhereallrace/ ethnicgroupswerecombined. Thecombinedrace/ethnicgroupanalysiswasconsideredprimarywithalpha=0.05,sinceIVS5-13insCwas presumedtobeafunctionalSNPwithsimilarconsequencesacrossrace/ethnicgroups,basedonpreviously publishedreports[5,7].AssumingaMAFof0.05and ORof1.40,wehad84%powertodetectanassociation betweentheIVS5-13insCvariantandtype2diabetesin adominantmodelintheoverallpopulation.Assuminga MAFof0.05andORof1.40,wehad61%powerinHispanics,44%powerinwhites,and18%powerinblacks todetectanassociationbyrace/ethnicity.PairwiseLD values(r2)andLDplotsweregeneratedbyrace/ethnic groupusingHaploview[17].Wepredicted insilico functionalconsequencesoftheIVS5-13insCvariantusing SNPNexus[18]andESEfinder3.0[19].ResultsWeidentified446incidenttype2diabetescasesovera mean2.8yearsoffollow-upandgenotypedanadditional 1329prevalenttype2diabetescasesinINVEST-GENES. Atbaseline,patientswithincidentorprevalenttype2diabeteshadhigherBMIandlowerdiastolicBP versus age, race/ethnicity,andgender-matchedcontrols(Table1). Casesalsohadahigherprevalenceofhypercholesterolemia, leftventricularhypertrophy,andcongestiveheartfailure. TheIVS5-13insCvariantdidnotdeviatefromHWEin anyrace/ethnicgroup(Table2).All HMGA1 SNPshad callratesabove95percentandnoSNPsfailedqualitycontrolbasedoncallrate,HardyWeinbergequilibriumtesting,concordancewithblindduplicates,orestimationof heterozygosityusingtheinbreedingcoefficientF.Thegenomicinflationfactorlambdawas1.03forINVESTindividualsgenotypedforthe HMGA1 IVS5-13insCvariant, suggestingminimalpopulationstratificationingenotyped individuals.FortheIVS5-13insCvariant,concordanceof duplicatesonTaqmanWwas99%andconcordancebetweenPyrosequencingandTaqmanWwas95%.InCaucasians,theMAFofIVS5-13insCwas0.03indiabeticcases and0.04incontrols.ThefrequencyofIVS5-13insCwas similarinAfricanAmericanindividuals(0.03incasesand 0.04incontrols)andhighestinHispanics(0.07incases and0.07incontrols).Nosignificantassociationswere observedbetweentheIVS5-13insCvariantanddiabetes overall(OR0.9895%CI0.76-1.26,p=0.88).TheIVS513insCvariantwasnotassociatedwithdiabetesinany race/ethnicgroup(Table2). PairwiseLDof HMGA1 SNPsbyrace/ethnicityispresentedinFigure1.PairwiseLD(r2)ofIVS5-13insCand rs9394200,theSNPpreviouslyusedasatagSNPfor IVS5-13insC[7],was0.47inCaucasians,0.25inHispanics,0.06inAfricanAmericans. Insilico functionalanalysisofIVS5-13insCrevealednopredictedchangesto aminoacidsequence,conservedtranscriptionfactor bindingsites,CpGislands,ormiRNAregulatorysites. IVS5-13insCwasnotpredictedtocausethecreationor disruptionofsplicesites.However, insilico evaluationof exonicspliceenhancer(ESE)sitesrevealedthattheCinsertionofIVS5-13insCcreatesanadditionalESEmotif forSF2/ASF(IgM-BRCA1).DiscussionInthepresentstudy,weobservedthatthe HMGA1 IVS5-13insCvariantwasnotassociatedwithdiabetesin theINVEST-GENESpopulationoverallorinanyrace/ ethnicgroup.WeobservedafrequencyofIVS5-13insC consistentwithpreviouslypublishedstudiesinCaucasians[5,7]andobservedanincreasedfrequencyofthis variantinHispanics.Furthermore,pairwiseLDinour populationsuggeststhatIVS5-13insCisnoteffectively taggedbyrs9394200inanyrace/ethnicgroupand in silico analysisrevealedminimalevidenceofputative functionalconsequencesoftheIVS5-13insCvariant. Ourobservationsextendexistingknowledgeofgenetic determinantsoftype2diabetesandareconsistentwithKarnes etal.JournalofTranslationalMedicine 2013, 11 :12Page3of6 http://www.translational-medicine.com/content/11/1/12

PAGE 4

thefindingsofMarquezetal.,whichsuggestalackof associationbetweenIVS5-13insCandtype2diabetes [7].Marquezetal.alsofoundthattheIVS5-13insCvariantisnotfunctional,whichissupportedbyourlackof observedassociationinmultiplerace/ethnicgroupsand our insilico functionalresultswhichpredictednoimportantconsequencesfromthisvariability Ourfindings areinconsistentwithtwostudiesthatobservedasignificantassociationbetweenIVS5-13insCandtype2diabetes[5,6].Thisdiscrepancymaybeduetodifferences inrace/ethnicityofourpopulationcomparedwithother studies.However,afunctionalIVS5-13insCvariant wouldbeexpectedtohavesimilareffectsacrossmultiple race/ethnicitygroups. Althoughourassociationanalysisisconsistentwith thefindingsofMarquezetal.[7],ourobservationoflow LDbetweenIVS5-13insCandrs9394200inallrace/ ethnicgroupssuggeststhatrs9394200isnotanappropriatetagSNPforIVS5-13insC.InHapmap,theMAFof theTalleleforrs9394200issimilartotheMAFfor IVS5-13insCinCaucasians(0.03),butis0.45inYorubans,furthersuggestingthatrs9394200doesnotadequatelytagIVS5-13insC,especiallyinnon-Caucasian populations.Furthermore,rs9394200is5000basepairs downstreamfromtheof HMGA1 3 endandisrepresentedonarraysutilizedbytype2diabetesGWAS [3,20].Therefore,ifrs9394200wereasignificantcontributortoriskfordiabetes,itlikelywouldhavebeen identifiedintheGWASanalyses. OurobservationofalackofassociationinHispanics andAfricanAmericanssuggeststhatIVS5-13insCmaybe associatedwithdiabetesonlyinCaucasianindividuals. WhilewedidnotobserveevenatrendtowardsassociationinourCaucasianpopulation,weacknowledgethat Table1Characteristicsoftype2diabetescasesandcontrolsatbaselineCharacteristic*Type2diabetescases(n=1775)Controls(n=1295) p value†Age(years)65.8(9.2)65.7(9.1)0.93 Female,n(%)1,014(57)731(56)0.71 BMI(kg/m2)30.6(5.6)28.9(5.4)<0.0001 Race/ethnicity,n(%) 0.15 Caucasian608(34)493(38) Hispanic937(53)642(50) AfricanAmerican216(12)153(12) Bloodpressure(mmHg) Systolic149(19)148(18)0.32 Diastolic85(11)86(10)0.0008 Hypercholesterolemia,n(%)‡1,017(57)673(52)0.003 HistoryofLVH,n(%)319(18)174(13)0.0007 HistoryofCHF,n(%)**84(5)25(2)<0.0001 Historyofsmoking,n(%)706(40)507(39)0.73BMIindicatesbodymassindex;LVH,leftventricularhypertrophy;CHF,congestiveheartfailure. *Valuesaremeanstandarddeviationunlessotherwisenoted. † Pvaluesrepresentt-testsandchisquaretestswhereappropriate. ‡ Historyoforcurrentlytaking lipid-loweringmedications.**NewYorkHeartAssociationClassI-III. Table2IVS5-13insCgenotypefrequenciesandassociationswithdiabetesoverallandbyrace/ethnicityUnadjustedAdjusted Race/ethnicgroupMAF(cases)MAF(controls)HWEpvalue*Oddsratio(95%CI)pvalueOddsratio(95%CI)†pvalue†Overall0.0530.052-1.00(0.79-1.26)0.980.98(0.76-1.26)0.88 (n=3070) Caucasian0.0280.0360.400.75(0.46-1.22)0.250.95(0.44-2.06)0.90 (n=1101) Hispanic0.0740.0700.201.09(0.82-1.45)0.570.79(0.49-1.25)0.31 (n=1579) AfricanAmerican(n=369)0.0300.0390.100.83(0.36-1.90)0.651.51(0.48-4.74)0.4895%CIindicates95%confidenceinterval;MAF,minorallelefrequency,HWE,HardyWeinbergEquilibrium. *HardyWeinbergEquilibriumpvaluecalculatedusingachisquaretestbyrace/ethnicityinnon-diabeticpatients. † Generatedusinglogisticregressionina dominantmodeladjustedforage,gender,bodymassindex,andCaucasian,African,andNativeAmericanancestryasestimatedbyancestryinformative markers.Karnes etal.JournalofTranslationalMedicine 2013, 11 :12Page4of6 http://www.translational-medicine.com/content/11/1/12

PAGE 5

ourpowertoobserveanassociationinCaucasiansis lowerthaninthepreviouslypublishedstudies.Although wehad84percentpowertodetectanassociationin ouroverallpopulation,wehadinadequatepowerto definitivelyconcludeanassociationinbyrace/ethnicity analyses. Our insilico analysisdidnotrevealanydirectmechanismoftheIVS5-13insCvariant ’ seffectsonmRNAexpressionoraminoacidsequence,suggestingthatthevariantis notlikelytobefunctional.AlthoughtheresultsofChiefari etal.suggestedafunctionaleffectoftheIVS5-13insCvarianton HMGA1 and INSR mRNAandproteinexpression inmonocytes[5],Marquezetal.foundnoeffectofthe variantonmRNAexpressioninadiposetissue.ThedifferencesinobservedeffectonmRNAexpressionmaypotentiallybeexplainedbydifferentialeffectsofthevarianton transcriptioninmonocytesandadiposetissue.TheapparentfunctionaleffectofIVS5-13insCobservedbyChiefari etal.mayalsobeconfoundedbythelackofevaluationof IVS5-13insCvariantcarriersversusnon-carriersamong non-diabeticcontrols[7]orpotentialtreatmentwith glucose-loweringmedicationsindiabeticpatientsfrom whommonocyteswerecollected[21].Finally,IVS513insCmayhaveanunknownfunctionalmechanismnot identifiedby insilico toolsusedinthisstudy.Ourstudy hasseverallimitationsworthyofmention.Werecognize thepotentialforfalsenegativeresultsinouranalyses,especiallyinAfricanAmericans,consideringthelowfrequencyofthevariantandthelimitedpowertodetect associationswithineachrace/ethicgroup.Ourobservationsrequirereplicationinindependentpopulationsof similarrace/ethnicmakeup.InINVEST,incidentand prevalentdiabetesdiagnosiswasbasedoninvestigator reports,butthediabetesphenotypeiswelldescribedina previouspublication[11],theaccuracyofsuchreporting hasbeenverifiedbyothers[22],andhasbeenusedin othertrials[23-25].INVESTinvestigator-reporteddiabetesphenotypeshavealsobeenusedinalargescale gene-centricmeta-analysiswithHumanCVDBeadChip data[26],suggestingvaliditywithregardtothediabetes phenotypeandgeneticassociationanalysis.Althoughour HMGA1 associationanalysismaybeconfoundedifcontrolpatientseventuallydevelopdiabetesafterstudyfollow-up,thehighmeanageandlowermeanBMIsuggests thatourcontrolpatientsarelesslikelytodeveloptype2 diabetes.Inaddition,althoughconcordancewithinthe TaqmanWgenotypingplatformwashigh,ourconcordance betweenTaqmanWandPyrosequencingwas95%,suggestingsomedisagreementbetweenplatforms.However, HardyWeinbergtestsdidnotindicategenotypeerrorand genotypeerrorwasminimizedbyutilizationofSangersequencingfordiscrepancyconfirmation.ConclusionsOurresultssuggestthatthe HMGA1 IVS5-13insCisnot associatedwithtype2diabetesandmaynothaveanimportantfunctionalroleindiabetespathogenesis.Wealso providefrequencyandLDdataforHispanicandAfrican Americanpopulations,whichhavehigherprevalenceof type2diabetes.Ourresultsalsosuggestthatrs9394200 isnotaneffectivetagSNPfor HMGA1 IVS5-13insC,especiallyinnon-Caucasianpopulations.Functionalstudiesandreplicationoftheseassociationsareneededto betterdefinethepotentialroleof HMGA1 variantsin predictingtype2diabetesdevelopment.Althoughthe currentstudysuggeststhelackofafunctionalrolefor IVS5-13insC,furtherstudyof HMGA1 iswarrantedto clarifytheroleofthisgeneindiabetespathogenesis.Larger,moreadequatelypoweredstudiesneedtobeperformedtoconfirmourfindings.Competinginterests Theauthorsdeclarethattheyhavenocompetinginterests. Authors ’ contributions JHKdraftedthemanuscriptandperformedstatisticalanalysis.TYL,JRC,MR, OEC,andSWCperformedgenotypingandhelpeddraftthemanuscript. CWMandYGperformedHumanCVDBeadChipqualitycontrolprocedures, Figure1 Haploview-generatedlinkagedisequilibrium(LD)plotof HMGA1 variationinINVESTCaucasians(A),Hispanics(B),andAfrican Americans(C). RegionsofhigherLDareshadeddarkeraccordingtohigherr2valuesandpairwiser2valuesareindicatedineachbox. Karnes etal.JournalofTranslationalMedicine 2013, 11 :12Page5of6 http://www.translational-medicine.com/content/11/1/12

PAGE 6

contributedtostatisticalanalysisandhelpeddraftthemanuscript.CJP,JAJ, andRCDconceivedofthestudy,participatedinitsdesignandcoordination, andhelpeddraftthemanuscript.Allauthorsreadandapprovedthefinal manuscript. Acknowledgements ThisstudywasfundedbyNIHgrantU01GM074492,R01HL74730,HL086558 (RCD),andTL1RR029888(JHK),andgrantsfromAbbottLaboratoriesandthe UniversityofFloridaOpportunityFund.WethankBenBurkley,LyndaStauffer, andCherylGallowayforprocessingandgenotypingsamples. Authordetails1DivisionofClinicalPharmacology,VanderbiltUniversity,1275Medical ResearchBuildingIV,Nashville,TN37232-0575,USA.2Departmentof PharmacotherapyandTranslationalResearch,UniversityofFlorida,HSCPO Box100486,Gainesville,FL32610-0486,USA.3DivisionofCardiovascular Medicine,UniversityofFlorida,POBox100277,Gainesville,FL32610-0486, USA. Received:5October2012Accepted:7January2013 Published:9January2013 References1.ZhangP,ZhangX,BrownJ,VistisenD,SicreeR,ShawJ,NicholsG: Global healthcareexpenditureondiabetesfor2010and2030. DiabetesResClin Pract 2010, 87: 293 – 301. 2.HerderC,RodenM: Geneticsoftype2diabetes:pathophysiologicand clinicalrelevance. EurJClinInvest 2011, 41: 679 – 692. 3.VoightBF,ScottLJ,SteinthorsdottirV,MorrisAP,DinaC,WelchRP,Zeggini E,HuthC,AulchenkoYS,ThorleifssonG, etal : Twelvetype2diabetes susceptibilitylociidentifiedthroughlarge-scaleassociationanalysis. Nat Genet 2010, 42: 579 – 589. 4.TalmudPJ,HingoraniAD,CooperJA,MarmotMG,BrunnerEJ,KumariM, KivimakiM,HumphriesSE: Utilityofgeneticandnon-geneticriskfactors inpredictionoftype2diabetes:WhitehallIIprospectivecohortstudy. BMJ 2010, 340: b4838. 5.ChiefariE,TanyolacS,PaonessaF,PullingerCR,CapulaC,IiritanoS,MazzaT, ForlinM,FuscoA,DurlachV, etal : FunctionalvariantsoftheHMGA1gene andtype2diabetesmellitus. JAMA 2011, 305: 903 – 912. 6.LiuL,DingH,WangHR,XuYJ,CuiGL,WangPH,YuanG,YuXF,WangDW: PolymorphismofHMGA1isassociatedwithincreasedriskoftype2 diabetesamongChineseindividuals. Diabetologia 2012, 55: 1685 – 1688. 7.MarquezM,HuyvaertM,PerryJR,PearsonRD,FalchiM,MorrisAP,Vivequin S,LobbensS,YengoL,GagetS, etal : Low-frequencyvariantsinHMGA1 arenotassociatedwithtype2diabetesrisk. Diabetes 2012, 61: 524 – 530. 8.RogerVL,GoAS,Lloyd-JonesDM,BenjaminEJ,BerryJD,BordenWB, BravataDM,DaiS,FordES,FoxCS, etal : ExecutiveSummary:Heart DiseaseandStrokeStatistics – 2012Update:AReportFromtheAmerican HeartAssociation. Circulation 2012, 125: 188 – 197. 9.PepineCJ,Handberg-ThurmondE,MarksRG,ConlonM,Cooper-DeHoffR, VolkersP,ZelligP: RationaleanddesignoftheInternationalVerapamil SR/TrandolaprilStudy(INVEST):anInternet-basedrandomizedtrialin coronaryarterydiseasepatientswithhypertension. JAmCollCardiol 1998, 32: 1228 – 1237. 10.PepineCJ,HandbergEM,Cooper-DeHoffRM,MarksRG,KoweyP,Messerli FH,ManciaG,CangianoJL,Garcia-BarretoD,KeltaiM, etal: Acalcium antagonistvsanon-calciumantagonisthypertensiontreatmentstrategy forpatientswithcoronaryarterydisease.TheInternationalVerapamilTrandolaprilStudy(INVEST):arandomizedcontrolledtrial. JAMA 2003, 290: 2805 – 2816. 11.Cooper-DehoffR,CohenJD,BakrisGL,MesserliFH,ErdineS,HewkinAC, KupferS,PepineCJ: Predictorsofdevelopmentofdiabetesmellitusin patientswithcoronaryarterydiseasetakingantihypertensive medications(findingsfromtheINternationalVErapamilSR-Trandolapril STudy[INVEST]). AmJCardiol 2006, 98: 890 – 894. 12.Cooper-DeHoffRM,HandbergEM,ManciaG,ZhouQ,ChampionA,LeglerUF, PepineCJ: INVESTrevisited:reviewoffi ndingsfromtheInternational VerapamilSR-TrandolaprilStudy. ExpertRevCardiovascTher 2009, 7: 1329 – 1340. 13.KeatingBJ,TischfieldS,MurraySS,BhangaleT,PriceTS,GlessnerJT,Galver L,BarrettJC,GrantSF,FarlowDN, etal : Concept,designand implementationofacardiovasculargene-centric50kSNParrayfor large-scalegenomicassociationstudies. PLoSOne 2008, 3: e3583. 14.PurcellS,NealeB,Todd-BrownK,ThomasL,FerreiraMA,BenderD,MallerJ, SklarP,deBakkerPI,DalyMJ,ShamPC: PLINK:atoolsetforwholegenomeassociationandpopulation-basedlinkageanalyses. AmJHum Genet 2007, 81: 559 – 575. 15.PritchardJK,StephensM,DonnellyP: Inferenceofpopulationstructure usingmultilocusgenotypedata. Genetics 2000, 155: 945 – 959. 16.Cooper-DeHoffRM,ArandaJMJr,GaxiolaE,CangianoJL,Garcia-BarretoD, ContiCR,HewkinA,PepineCJ: Bloodpressurecontrolandcardiovascular outcomesinhigh-riskHispanicpatients – findingsfromtheInternational VerapamilSR/TrandolaprilStudy(INVEST). AmHeartJ 2006, 151: 1072 – 1079. 17.BarrettJC,FryB,MallerJ,DalyMJ: Haploview:analysisandvisualizationof LDandhaplotypemaps. Bioinformatics 2005, 21: 263 – 265. 18.ChelalaC,KhanA,LemoineNR: SNPnexus:awebdatabaseforfunctional annotationofnewlydiscoveredandpublicdomainsinglenucleotide polymorphisms. Bioinformatics 2009, 25: 655 – 661. 19.CartegniL,WangJ,ZhuZ,ZhangMQ,KrainerAR: ESEfinder:Aweb resourcetoidentifyexonicsplicingenhancers. NucleicAcidsRes 2003, 31: 3568– 3571. 20.BurtonPR,ClaytonDG,CardonLR,CraddockN,DeloukasP,DuncansonA, KwiatkowskiDP,McCarthyMI,OuwehandWH,SamaniNJ, etal : Genomewideassociationstudyof14,000casesofsevencommondiseasesand 3,000sharedcontrols. Nature 2007, 447: 661 – 678. 21.FroguelP,MarquezM,CauchiS: Responsetocommenton:Marquezetal. Low-frequencyvariantsinHMGA1arenotassociatedwithtype2 diabetesrisk.Diabetes2012;61:524 – 530. Diabetes 2012, 61: e15. 22.GoldmanN,LinIF,WeinsteinM,LinYH: Evaluatingthequalityofselfreportsofhypertensionanddiabetes. JClinEpidemiol 2003, 56: 148 – 154. 23.AguilarD,SolomonSD,KoberL,RouleauJL,SkaliH,McMurrayJJ,Francis GS,HenisM,O ’ ConnorCM,DiazR, etal : Newlydiagnosedandpreviously knowndiabetesmellitusand1-yearoutcomesofacutemyocardial infarction:theVALsartanInAcutemyocardialiNfarcTion(VALIANT)trial. Circulation 2004, 110: 1572 – 1578. 24.BoschJ,LonnE,PogueJ,ArnoldJM,DagenaisGR,YusufS: Long-term effectsoframipriloncardiovasculareventsandondiabetes:resultsof theHOPEstudyextension. Circulation 2005, 112: 1339 – 1346. 25.YusufS,GersteinH,HoogwerfB,PogueJ,BoschJ,WolffenbuttelBH, ZinmanB: Ramiprilandthedevelopmentofdiabetes. JAMA 2001, 286: 1882 – 1885. 26.SaxenaR,ElbersCC,GuoY,PeterI,GauntTR,MegaJL,LanktreeMB,TareA, CastilloBA,LiYR, etal : Large-scalegene-centricmeta-analysisacross39 studiesidentifiestype2diabetesloci. AmJHumGenet 2012, 90: 410 – 425.doi:10.1186/1479-5876-11-12 Citethisarticleas: Karnes etal. : Lackofassociationofthe HMGA1 IVS513insCvariantwithtype2diabetesinanethnicallydiverse hypertensivecasecontrolcohort. JournalofTranslationalMedicine 2013 11 :12. Submit your next manuscript to BioMed Central and take full advantage of: € Convenient online submission € Thorough peer review € No space constraints or color “gure charges € Immediate publication on acceptance € Inclusion in PubMed, CAS, Scopus and Google Scholar € Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Karnes etal.JournalofTranslationalMedicine 2013, 11 :12Page6of6 http://www.translational-medicine.com/content/11/1/12


xml version 1.0 encoding utf-8 standalone no
mets ID sort-mets_mets OBJID sword-mets LABEL DSpace SWORD Item PROFILE METS SIP Profile xmlns http:www.loc.govMETS
xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
xsi:schemaLocation http:www.loc.govstandardsmetsmets.xsd
metsHdr CREATEDATE 2013-01-29T20:07:44
agent ROLE CUSTODIAN TYPE ORGANIZATION
name BioMed Central
dmdSec sword-mets-dmd-1 GROUPID sword-mets-dmd-1_group-1
mdWrap SWAP Metadata MDTYPE OTHER OTHERMDTYPE EPDCX MIMETYPE textxml
xmlData
epdcx:descriptionSet xmlns:epdcx http:purl.orgeprintepdcx2006-11-16 xmlns:MIOJAVI
http:purl.orgeprintepdcxxsd2006-11-16epdcx.xsd
epdcx:description epdcx:resourceId sword-mets-epdcx-1
epdcx:statement epdcx:propertyURI http:purl.orgdcelements1.1type epdcx:valueURI http:purl.orgeprintentityTypeScholarlyWork
http:purl.orgdcelements1.1title
epdcx:valueString Lack of association of the HMGA1 IVS5-13insC variant with type 2 diabetes in an ethnically diverse hypertensive case control cohort
http:purl.orgdctermsabstract
Abstract
Background
Recently, the high-mobility group A1 gene (HMGA1) variant IVS5-13insC has been associated with type 2 diabetes, but reported associations are inconsistent and data are lacking in Hispanic and African American populations. We sought to investigate the HMGA1-diabetes association and to characterize IVS5-13insC allele frequencies and linkage disequilibrium (LD) in 3,070 Caucasian, Hispanic, and African American patients from the INternational VErapamil SR-Trandolapril STudy (INVEST).
Methods
INVEST was a randomized, multicenter trial comparing two antihypertensive treatment strategies in an ethnically diverse cohort of hypertensive, coronary artery disease patients. Controls, who were diabetes-free throughout the study, and type 2 diabetes cases, either prevalent or incident, were genotyped for IVS5-13insC using Taqman®, confirmed with Pyrosequencing and Sanger sequencing. For LD analysis, genotyping for eight additional HMGA1 single nucleotide polymorphisms (SNPs) was performed using the Illumina® HumanCVD BeadChip. We used logistic regression to test association of the HMGA1 IVS5-13insC and diabetes, adjusted for age, gender, body mass index, and percentage European, African, and Native American ancestry.
Results
We observed IVS5-13insC minor allele frequencies consistent with previous literature in Caucasians and African Americans (0.03 in cases and 0.04 in controls for both race/ethnic groups), and higher frequencies in Hispanics (0.07 in cases and 0.07 in controls). The IVS5-13insC was not associated with type 2 diabetes overall (odds ratio 0.98 [0.76-1.26], p=0.88) or in any race/ethnic group. Pairwise LD (r2) of IVS5-13insC and rs9394200, a SNP previously used as a tag SNP for IVS5-13insC, was low (r2=0.47 in Caucasians, r2=0.25 in Hispanics, and r2=0.06 in African Americans). Furthermore, in silico analysis suggested a lack of functional consequences for the IVS5-13insC variant.
Conclusions
Our results suggest that IVS5-13insC is not a functional variant and not associated with type 2 diabetes in an ethnically diverse, hypertensive, coronary artery disease population. Larger, more adequately powered studies need to be performed to confirm our findings.
Trial registration
clinicaltrials.gov (NCT00133692)
http:purl.orgdcelements1.1creator
Karnes, Jason H
Langaee, Taimour Y
McDonough, Caitrin W
Chang, Shin-Wen
Ramos, Miguel
Catlin Jr, James R
Casanova, Octavio E
Gong, Yan
Pepine, Carl J
Johnson, Julie A
Cooper-DeHoff, Rhonda M
http:purl.orgeprinttermsisExpressedAs epdcx:valueRef sword-mets-expr-1
http:purl.orgeprintentityTypeExpression
http:purl.orgdcelements1.1language epdcx:vesURI http:purl.orgdctermsRFC3066
en
http:purl.orgeprinttermsType
http:purl.orgeprinttypeJournalArticle
http:purl.orgdctermsavailable
epdcx:sesURI http:purl.orgdctermsW3CDTF 2013-01-09
http:purl.orgdcelements1.1publisher
BioMed Central Ltd
http:purl.orgeprinttermsstatus http:purl.orgeprinttermsStatus
http:purl.orgeprintstatusPeerReviewed
http:purl.orgeprinttermscopyrightHolder
Jason H Karnes et al.; licensee BioMed Central Ltd.
http:purl.orgdctermslicense
http://creativecommons.org/licenses/by/2.0
http:purl.orgdctermsaccessRights http:purl.orgeprinttermsAccessRights
http:purl.orgeprintaccessRightsOpenAccess
http:purl.orgeprinttermsbibliographicCitation
Journal of Translational Medicine. 2013 Jan 09;11(1):12
http:purl.orgdcelements1.1identifier
http:purl.orgdctermsURI http://dx.doi.org/10.1186/1479-5876-11-12
fileSec
fileGrp sword-mets-fgrp-1 USE CONTENT
file sword-mets-fgid-0 sword-mets-file-1
FLocat LOCTYPE URL xlink:href 1479-5876-11-12.xml
sword-mets-fgid-1 sword-mets-file-2 applicationpdf
1479-5876-11-12.pdf
structMap sword-mets-struct-1 structure LOGICAL
div sword-mets-div-1 DMDID Object
sword-mets-div-2 File
fptr FILEID
sword-mets-div-3