Prox1 gene variant is associated with fasting glucose change after antihypertensive treatment

PROX1 Gene Variant is Associated with Fasting Glucose Yan Gong,1* Caitrin W. McDonough,1 Amber L. Beitelshees,2 Jason H. Karnes,3 Jeffrey R. O’Connell,2 Stephen T. Turner,4 Arlene B. Chapman,5 John G. Gums,1,6 Kent R. Bailey,4 Eric Boerwinkle,7 Julie A.
Johnson,1,6 and Rhonda M. Cooper-DeHoff,1,6 1Department of Pharmacotherapy and Translational Research and Center for Pharmacogenomics, University of Florida, Gainesville, Florida; 2College of Medicine, University of Maryland, Baltimore, Maryland; 3Division of Clinical Pharmacology, Vanderbilt University, Nashville, Tennessee; 4College of Medicine, Mayo Clinic Rochester, Rochester, Minnesota; 5School of Medicine, Emory University, Atlanta, Georgia; 6College of Medicine, University of Florida, Gainesville, Florida; 7Center for Human Genetics, University of Texas at Houston, STUDY OBJECTIVE To assess the relationship of the 33 single nucleotide polymorphisms (SNPs) previously associated with fasting glucose in Caucasians in genome-wide association studies (GWAS) with glucoseresponse to antihypertensive drugs shown to increase risk for hyperglycemia and diabetes.
DESIGN Randomized, multicenter clinical trial.
PATIENTS A total of 456 Caucasian men and women with uncomplicated hypertension.
MEASUREMENTS AND MAIN RESULTS The Pharmacogenomic Evaluation of Antihypertensives Responses study evaluated blood pressure and glucose response in uncomplicated hypertensive patients random-ized to either atenolol or hydrochlorothiazide (HCTZ) monotherapy, followed by combination therapywith both agents. Association of these SNPs with atenolol- or HCTZ-induced glucose response wasevaluated in 456 Caucasian patients using linear regression adjusting for age, sex, body mass index,baseline glucose, baseline insulin, and principal component for ancestry. The SNP rs340874 in the 5′region of PROX1 gene was significantly associated with atenolol-induced glucose change (p=0.0013).
Participants harboring the C allele of this SNP had greater glucose elevation after approximately9 weeks of atenolol monotherapy (b = +2.39 mg/dl per C allele), consistent with the direction ofeffect in fasting glucose GWAS, that showed the C allele is associated with higher fasting glucose.
CONCLUSION These data suggest that PROX1 SNP rs340874, discovered in fasting glucose GWAS, may also be a pharmacogenetic risk factor for antihypertensive-induced hyperglycemia. b-blockers andthiazides may interact with genetic risk factors to increase risk for dysglycemia and diabetes.
KEY WORDS pharmacogenomics, dysglycemia, adverse metabolic effects, atenolol, Hydrochlorothiazide,hypertension, antihypertensives.
(Pharmacotherapy 2014;34(2):123–130) doi: 10.1002/phar.1355 PEAR was supported by the National Institute of Health Pharmacogenetics Research Network grant U01-GM074492 and the National Center for Advancing Translational Sciences under the award number UL1 TR000064 (University of Florida);UL1 TR000454 (Emory University) and UL1 TR000135 (Mayo Clinic) and funds from the Mayo Foundation. This researchwas also supported by K23 HL091120 (ALB) and K23HL086558 (RCD).
RMC, ALB, ABC, JGG, EB, STT and JAJ received funding from NIH. RMC also received funding from the NIH Women’s Health Initiative. JGG received funding from Janssen Pharmaceuticals, Inc., has Speaker’s Bureau appointment from Boehrin-ger-Ingelheim and is a consultant for Forest Pharmaceuticals and Boehringer-Ingelheim. EB received honoraria from the Foun-dation of Rome.
*Address for correspondence: Yan Gong, Department of Pharmacotherapy and Translational Research, University of Flor- ida, PO Box 100486, 1600 South West Archer Road, Gainesville, FL 32610-0486; e-mail: gong@cop.ufl.edu.
Ó 2013 Pharmacotherapy Publications, Inc.
PHARMACOTHERAPY Volume 34, Number 2, 2014 the association of these loci with glucose disease for which drugs are prescribed and response to atenolol (a b-blocker) and hydro- places over one-third of Americans at substan- tially increased risk for stroke, coronary heart Identifying genetic predictors of b-blocker and disease (including myocardial infarction), renal failure, and heart failure.1 Two of the widely guide clinicians to prescribe drugs that mitigate the development of hyperglycemia and diabetes b-blockers and thiazide diuretics,2 are associated with decreased insulin sensitivity and hypergly-cemia.3 Short-term use of these drugs is associ- ated with an increased incidence of impairedfasting glucose,4 and in numerous clinical trials, long-term use has been shown to increase therisk of new-onset diabetes.5–7 The PEAR study (clinicaltrials.gov Identifier There is considerable variability in glucose NCT00246519) was a randomized, multicenter changes after exposure to b-blockers and thia- clinical trial examining the role of genetic vari- zide diuretics.4 We have previously reported that ability on blood pressure and adverse metabolic reductions in blood pressure were not correlated responses to HCTZ and atenolol.17 Men and with changes in glucose elevation after b-blocker women of any race between the ages of 17 and and thiazide treatment.8 Although genetic poly- morphisms likely explain a portion of this vari- recruited to participate at University of Florida ability, the few studies that have investigated the (Gainesville, FL), Mayo Clinic (Rochester, MN), pharmacogenomics of b-blocker and thiazide- and Emory University (Atlanta, GA). Patients induced glucose change have largely focused on with diabetes were excluded from the study. The glucose response to thiazide diuretics.9, 10 institutional review boards at each institution Studies have shown that even for individuals approved the protocol and each participant pro- with normal fasting glucose, the diabetes risk vided informed written consent before entry into increases as fasting glucose levels increase.11, 12 In a study of more than 45,000 individuals with a mean follow-up of 81 months, each mg/ 4 weeks of washout of antihypertensive drugs.
dl increase in fasting glucose was associated Participants were randomized to receive either with a 6% higher risk for diabetes (hazard ratio HCTZ 12.5 mg/day or atenolol 50 mg/day for 1.06; 95% confidence interval 1.05–1.07).11 3 weeks, followed by dose titration to 25 mg There is evidence that drug-associated new- and 100 mg/day, respectively, for systolic blood onset diabetes and diabetes of other etiologies pressure (SBP) greater than 120 mm Hg or dia- may share common mechanisms.13, 14 Antihy- stolic blood pressure greater than 70 mm Hg.
pertensive drugs such as b-blockers and thia- In the atenolol treatment arm, 84% of the par- zide diuretics could act as an environmental ticipants underwent dose titration to atenolol trigger to accelerate the onset of diabetes.5–7 100 mg and 99% of the participants in the We believe that for nondiabetic individuals HCTZ arm received titration to HCTZ 25 mg/ with genetic risk for higher fasting glucose, tak- day. Blood pressure and metabolic responses to ing antihypertensives with glucose-increasing monotherapy were assessed after 9 weeks of side effects may further increase their glucose treatment. In the second part of the study, the and put them at a greater risk for developing alternate drug was added in those patients diabetes. More specifically, we hypothesize that with SBP above 120 mm Hg or diastolic blood genetic variants associated with higher fasting pressure above 70 mm Hg (> 90% for both randomization arms), with dose titration after increases induced by b-blockers and thiazide diuretics. To date, 36 loci have been associated 9 weeks, as in the first portion of the study.
with fasting glucose in genome-wide association Participants were asked to maintain their cur- studies (GWAS) of nondiabetic Caucasians.15, 16 rent lifestyle behaviors throughout the study To test our hypothesis, we analyzed data from period. The current analysis focuses on the the Pharmacogenomic Evaluation of Antihy- glucose change after atenolol or HCTZ mono- pertensive Responses (PEAR) study to assess PROX1 VARIANT AND ANTIHYPERTENSIVE-INDUCED GLUCOSE ELEVATION Gong et al rate in the remaining individuals in these lociwas 99.95%.
The phenotype of interest for this subgroup analysis was glucose change after monotherapy,calculated as glucose after 9 weeks of monothera- py minus glucose at baseline (untreated). Fasting blood samples were collected for glucose and insu- the Infinium II assay and genotypes were called lin before and after completion of atenolol or using BeadStudio software and GenTrain2 call- HCTZ monotherapy. Plasma glucose was mea- ing algorithm (Illumina, San Diego, CA). Quality sured using a Hitachi 911 Chemistry Analyzer control procedures were performed as with (Roche Diagnostics, Indianapolis, IN) at the central Human CVD Beadchip data. After the quality control procedures, the total SNP call rate in the remaining individuals and SNPs was 99.86%.
automated enzymatic assay. Plasma insulin was Principal component analysis was performed measured using the Access Ultrasensitive Insulin using Omni1M Quad GWAS data to assess the immunoassay system (Beckman Coulter, Brea, ancestral background. Participant’s self-identified CA). Insulin sensitivity status was calculated using race information was confirmed with principal the homeostatic model assessment – insulin resis- component analysis of the Omni1M Quad data tance.18 All samples were tested in duplicate, and data reported are means of the duplicate samples.
We limited our analysis in this study to the We were able to obtain genotypes on 33 sin- Caucasian patients enrolled in PEAR since SNPs gle nucleotide polymorphisms (SNPs) from the were selected from GWAS in Caucasians. To 36 loci associated with fasting glucose level in exclude patients that were potentially nonfasting at either visit, we regressed glucose response against nongenetic covariates (age, sex, body mass index , baseline glucose, and baseline insu- (rs16913693 in IKBKAP, rs10747083 in P2RX2 lin) and then standardized the residuals into a and rs2302593 in GIPR) or their proxies (SNPs distribution with mean of 0 and standard devia- with r2 > 0.8 with the index SNPs) were not tion of 1. Participants with standardized residu- als that were outside four standard deviationswere excluded from analysis (n=2). In theremaining 456 Caucasian participants, associa- tions of the 33 SNPs with glucose response after atenolol or HCTZ monotherapy were evaluated ized gene-centric array including approximately using linear regression that adjusted for signifi- cant nongenetic predictors of glucose response, genotyped using the Infinium II Assay (Illumina, including baseline glucose, baseline insulin, age, San Diego, CA). Genotypes were called using sex and body mass index. The analyses also GenomeStudio software version 2011.1 and the adjusted for the first principal component for Genotyping Module version 1.9 calling algo- ancestry, which corresponds to European ances- rithm (Illumina, San Diego, CA). Participants try in PEAR individuals. Additive mode of inher- were excluded if sample genotype call rates were itance was assumed where the SNPs were coded below 95% and SNPs were excluded if genotype as 0, 1, and 2 in the linear regression model.
call rates were below 95%. Sample contamina- To correct for multiple testing of 33 SNPs, we tion was detected by checking sex mismatches lowered the a level to 0.0015 (0.05/33) so that using X chromosome genotype data and cryptic the type I error (false positive) of the study is relatedness was estimated by pairwise identity- less than 0.05. Therefore SNPs with p values of by-descent analysis implemented using PLINK less than 0.0015 were considered statistically (http://pngu.mgh.harvard.edu/purcell/plink/).
significant. For nominally significant (p<0.05) Hardy–Weinberg Equilibrium was assessed with SNPs with lower minor allele frequencies, we a v2 test with one degree of freedom. After the also explored a dominant model whereby het- quality control procedures, the total SNP call erozygotes and minor allele homozygotes were PHARMACOTHERAPY Volume 34, Number 2, 2014 combined and compared with the common allele + 2.8 mg/dl (interquartile range: À 3.8 to homozygotes. To evaluate the overall risk and + 7.0 mg/dl). The median glucose change after benefit of antihypertensive treatments, we also HCTZ monotherapy was + 2.0 mg/dl (interquar- assessed the systolic blood pressure response for tile range: À4.3 to + 6.8 mg/dl). Figure 1 dem- onstrates the large interindividual variability in response using linear regression, adjusting for glucose response to atenolol and HCTZ mono- baseline systolic blood pressure, age, sex, and therapy in Caucasian hypertensive participants.
first principal component. All single SNP linearregression SNPs associated with atenolol-induced glucose PLINK.20 Other analyses were performed in SASversion 9.3 (Cary, NC).
Genotypes for the 33 previously identified fasting glucose GWAS SNPs were available forPEAR participants from the two genotyping plat- The baseline characteristics of the 456 Cauca- sian PEAR participants assigned to atenolol Hardy–Weinberg Equilibrium in Caucasians.
(n=232) and HCTZ treatment (n=224) are pre- An SNP in the 5′ untranslated region of the sented in Table 1. The participants had a mean prospero homeobox 1 gene (PROX1), rs340874, was significantly associated with glucose change 44% were overweight and obese, respectively.
The 145.0 Æ 9.4/92.9 Æ 5.5 mm Hg. The mean base- Atenolol Monotherapy
line glucose in both treatment groups was in the normal range, with a median of 90 mg/dl and interquartile range of 84–96 mg/dl.
After an average of 9 weeks of atenolol mono- Table 1. Baseline Characteristics of Caucasian Partici- glucose change (mg/dL)
HCTZ Monotherapy
BMI = body mass index; HOMA-IR = homeostatic model assess- glucose change (mg/dL)
aNumeric characteristics were presented as mean Æ standard devia-tion and categorical variables were presented as number and per- Figure 1. Distribution of glucose change (mg/dl) after atenolol (A) and hydrochlorothiazide monotherapy (B).
PROX1 VARIANT AND ANTIHYPERTENSIVE-INDUCED GLUCOSE ELEVATION Gong et al Table 2. Fasting Glucose SNPs Associated with Glucose Response to Atenolol or HCTZ Monotherapy in PEAR CaucasianHypertensive Patients (with p<0.05) MAF = minor allele frequency; SE = standard errorLocation: NCBI build 36 base pair position. Beta indicates the glucose response (in mg/dl) for each glucose increasing allele. p values werelinear regression p adjusted for baseline glucose, baseline insulin, age, sex, body mass index and principal components for ancestry.
aAlleles were presented as major/minor alleles.
bp value that reached Bonferroni-corrected significance level.
PROX1 rs340874
b = +2.4 mg/dl; Table 2). Participants with T/T, T/C and C/C genotypes had mean glucosechanges of À 0.46, + 1.77, and + 3.19 mg/dl, respectively (Figure 2). The systolic blood pres- sure reduction was not statistically different among the three genotype groups (À12.6 mm -induced
À9.3 mm Hg for C/C individuals, respectively, p=0.15), although the trend was such that those with the greatest glucose increase also had a smaller reduction in blood pressure.
An intronic SNP in ARAP1 gene (ArfGAP with RhoGAP domain, ankyrin repeat and PH domain1), rs11603334, was nominally associated with Figure 2. Fasting glucose SNP PROX1 rs340874 associated b = + 2.7 mg/dl) (Table 2). Participants with with glucose response to atenolol monotherapy among the A allele had a higher glucose increase after Caucasian hypertensive patients. Error bars represent + 0.7 mg/dl for A/A, A/G, and G/G individuals,respectively (Figure S1). There was no statisti- therapy, patients with A/A, A/T, and T/T geno- cally significant difference in reduction of sys- tolic blood pressure during the same treatment response of + 9.1, + 3.1, and + 0.6 mg/dl, period: A/A: À9.5 mm Hg, A/G: À9.9 mm Hg, respectively (Figure S2). There was no signifi- cant difference in systolic blood pressure reduc- Two other SNPs showed a trend toward asso- tion among the genotypes: À9.7, À6.7, and ciation with glucose change after atenolol mono- À8.0 mm Hg for A/A, A/T, and T/T, respectively therapy. These SNPs included rs11039149, an intronic SNP in NR1H3 (p=0.057, b = + 1.6 mg/ rs10830963 and ADCY5 rs11708067) showed a dl) and rs4869272, an intergenic SNP between trend toward association with glucose change PCSK1 and MIR583 (p=0.058, b = + 1.5 mg/dl) SNPs associated with HCTZ-induced glucose To our knowledge, this is the first study to test fasting glucose GWAS loci on drug-induced An intronic SNP rs11920090 in SLC2A2 (sol- glucose change. Among 33 SNPs previously doc- ute carrier family 2, member 2) gene was the umented to be associated with fasting glucose levels in Caucasians, one SNP achieved statistical significance for association with glucose change b = + 3.1 mg/dl) (Table 2). After HCTZ mono- after approximately 9 weeks of atenolol mono- PHARMACOTHERAPY Volume 34, Number 2, 2014 therapy. Two other SNPs were nominally associ- direction of effects appeared to be opposite of ated with glucose change after exposure to ate- the fasting glucose GWAS effects. In the prior fasting glucose GWAS, individuals with the G The effect sizes of these SNPs on the atenolol allele of ARAP1 rs11603334 had higher fasting or HCTZ-associated glucose change were in the glucose levels (b = 0.019 mmol/L or 0.34 mg/dl range of 2–3 mg/dl per allele, which is approxi- per allele).15 In this study, however, patients mately 10 times greater than the effect sizes with the G allele had a smaller glucose increase observed in the fasting glucose GWAS.15, 16 This after atenolol monotherapy. In fasting glucose is consistent with numerous studies indicating GWAS, the T allele of SLC2A2 rs11920090 was that pharmacogenetic effect sizes are larger than associated with higher glucose levels, with a b of + 0.02 mmol/L or + 0.36 mg/dl.16 In this study, however, the T allele was associated with a is an early specific marker for developing liver smaller increase in glucose after HCTZ mono- and pancreas in foregut endoderm. In an in vitro therapy. The associations of these two SNPs were not consistent with our original hypothesis.
repressor of hepatocyte nuclear factor 4a (HNF4 One possible explanation is that the patients a) that may play a key role in the regulation of may have other variants (for which we did not bile acid synthesis and gluconeogenesis in the test) that might influence the glucose response liver.25 HNF4A, which is the gene that encodes in the opposite direction from these two SNPs.
HNF4 a, is responsible for a type of diabetes For this study, we evaluated only the associa- called maturity-onset diabetes of the young tions of single SNPs (previously identified in (MODY).26 A genetically heterogeneous mono- fasting glucose-associated GWAS) with glucose genic form of noninsulin-dependent diabetes change to antihypertensive treatment. It is our mellitus, MODY is characterized by onset usu- long-term goal to evaluate combined effect of ally before age 25 and often in adolescence or multiple variants in a model to predict which childhood and by autosomal dominant inheri- patient population would benefit from antihy- tance. In GWAS studies and meta-analyses of pertensive treatment without adverse metabolic fasting glucose homeostasis and risk of type 2 side effects, including increased glucose.
The mechanisms by which atenolol and HCTZ rs340874 C allele had higher fasting glucose increase glucose are not well understood. This level (b = + 0.013 mmol/L or + 0.23 mg/dl per study identifies one SNP (PROX1 rs340874 C allele, p=6.6*10À6) and were at higher risk of allele) that is associated with higher glucose developing type 2 diabetes (odds ratio 1.07, increase after short term exposure (approxi- p=7.2*10À10).16 In our study, hypertensive mately 9 weeks) to atenolol monotherapy; this Caucasians harboring the C allele of this SNP same SNP was also previously associated with had greater glucose elevation after approximately These data suggest that administration of ate- 2.39 mg/dl per allele) and a trend for less pro- nolol in individuals already predisposed to nounced blood pressure reduction. If replicated, higher fasting glucose by their PROX1 genotype these data suggest that hypertensive Caucasians might provoke an exaggerated glucose elevation.
with the C allele of this SNP should avoid treat- Due to the long-term nature of antihypertensive ment with atenolol where possible, given that therapy and the wide use of b-blockers, it is even modest elevation in glucose could increase imperative to further investigate the underlying the long-term risk for diabetes (6% higher risk mechanisms of the adverse metabolic effects of diabetes for each mg/dl increase),11 the long- associated with these agents. We anticipate that term nature of antihypertensive therapy and the studies of larger sample size will enable us to high allele frequency of this SNP (48% minor find additional genetic variants. Hence, our goal is to identify multiple genetic variants that could Two SNPs achieved nominal p values of less be considered in a risk model for increased glu- than 0.05 but did not meet the requirement for cose when considering the initiation of antihy- statistical significance after considering multiple This study is not without some limitations.
associated with the atenolol-induced glucose The PEAR study was designed primarily as a response, and SLC2A2 rs11920090, associated blood pressure response study. For blood pres- with the HCTZ-induced glucose response, the sure lowering, 9 weeks of treatment is sufficient PROX1 VARIANT AND ANTIHYPERTENSIVE-INDUCED GLUCOSE ELEVATION Gong et al to observe maximal blood pressure lowering.
therapy in hypertension. Circulation 2008;117:2706–15; dis- Regarding the glucose response, data from the 7. Barzilay JI, Davis BR, Cutler JA, et al. Fasting glucose levels large hypertension study ALLHAT27 provides and incident diabetes mellitus in older nondiabetic adults ran- evidence that glucose continues to rise over the domized to receive 3 different classes of antihypertensive treat- long term, particularly in patients without diabe- ment: a report from the antihypertensive and lipid-loweringtreatment to prevent heart attack trial (ALLHAT). Arch Intern tes when antihypertensive treatment begins.
Importantly, fasting glucose and the incidence of 8. Smith SM, Gong Y, Turner ST, et al. Blood pressure responses new onset diabetes continued to rise over the and metabolic effects of hydrochlorothiazide and atenolol. AmJ Hypertens 2012;25:359–65.
entire 5-year study period. This suggests that the 9. Maitland-van der Zee AH, Turner ST, Schwartz GL, Chapman duration of exposure is very important regarding AB, Klungel OH, Boerwinkle E. Demographic, environmental, the dysglycemic effect and is likely more impor- and genetic predictors of metabolic side effects of hydrochloro-thiazide treatment in hypertensive subjects. Am J Hypertens tant than dose, which is titrated to blood pres- sure response early in therapy. Therefore, the 10. Karnes JH, McDonough CW, Gong Y, et al. Association of PEAR study likely underestimated the glucose kcnj1 variation with change in fasting glucose and new onsetdiabetes during hctz treatment. Pharmacogenomics J 2012; increase caused by the two antihypertensives.
[Epub ahead of print] doi: 10.1038/tpj.2012.34.
11. Nichols GA, Hillier TA, Brown JB. Normal fasting plasma glu- cose and risk of type 2 diabetes diagnosis. Am J Med 12. Tirosh A, Shai I, Tekes-Manova D, et al. Normal fasting These data suggest that variants discovered in plasma glucose levels and type 2 diabetes in young men. N fasting glucose GWAS provide pharmacogenetic 13. Wang TJ, Larson MG, Vasan RS, et al. Metabolite profiles risk factors for atenolol or HCTZ-induced hyper- and the risk of developing diabetes. Nat Med 2011;17:448–53.
glycemia. These drugs may interact with genetic 14. Cooper-DeHoff RM, Hou W, Baillie R, et al. Pharmacometa- risk factors to increase the risk for dysglycemia bolomics reveals a novel signature for predicting beta blockerassociated impaired fasting glucose. Circulation 2012;126: and diabetes. Functional studies and replication of these associations are needed. After validation 15. Scott RA, Lagou V, Welch RP, et al. Consortium DGRaM-aD.
in other independent populations, the clinical Large-scale association analyses identify new loci influencingglycemic traits and provide insight into the underlying biologi- goal would be to identify a priori those hyper- cal pathways. Nat Genet 2012;44:991–1005.
tensive individuals who are likely to have high 16. Dupuis J, Langenberg C, Prokopenko I, et al. Consortium risk for elevated glucose after treatment with at- AHoboP, investigators M. New genetic loci implicated in fast-ing glucose homeostasis and their impact on type 2 diabetes enolol or HCTZ and provide alternative treat- 17. Johnson JA, Boerwinkle E, Zineh I, et al. Pharmacogenomics of antihypertensive drugs: rationale and design of the phar-macogenomic evaluation of antihypertensive responses (PEAR) 18. Wallace TM, Levy JC, Matthews DR. Use and abuse of homa We acknowledge and thank the valuable contribu- modeling. Diabetes Care 2004;27:1487–95.
tions of the PEAR study participants, support staff, 19. Keating BJ, Tischfield S, Murray SS, et al. Concept, design and study physicians: R. Whit Curry, Karen Hall, and implementation of a cardiovascular gene-centric 50 k snp Frederic Rabari-Oskoui, Dan Rubin, and Siegfried array for large-scale genomic association studies. PLoS ONE2008;3:e3583.
20. Purcell S, Neale B, Todd-Brown K, et al. Plink: a tool set for whole-genome association and population-based linkage analy- ses. Am J Hum Genet 2007;81:559–75.
21. Daly AK. Genome-wide association studies in pharmacoge- 1. Go AS, Mozaffarian D, Roger VL, et al. Subcommittee obotA- nomics. Nat Rev Genet 2010;11:241–6.
HASCaSS. Heart disease and stroke statistics–2013 update: a 22. Gong Y, McDonough CW, Wang Z, et al. Hypertension sus- report from the american heart association. Circulation ceptibility loci and blood pressure response to antihyperten- sives: results from the pharmacogenomic evaluation of 2. Chobanian AV, Bakris GL, Black HR, et al. The seventh antihypertensive responses study. Circ Cardiovasc Genet report of the joint national committee on prevention, detec- tion, evaluation, and treatment of high blood pressure: the jnc 23. Del-Aguila JL, Beitelshees AL, Cooper-Dehoff RM, et al. Gen- ome-wide association analyses suggest nell1 influences adverse 3. Lithell HO. Effect of antihypertensive drugs on insulin, glu- metabolic response to hctz in African Americans. Pharmacoge- cose, and lipid metabolism. Diabetes Care 1991;14:203–9.
nomics J 2013;[Epub ahead of print] doi: 10.1038/tpj.2013.3.
4. Cooper-DeHoff RM, Wen S, Beitelshees AL, et al. Impact of 24. Turner ST, Bailey KR, Schwartz GL, Chapman AB, Chai abdominal obesity on incidence of adverse metabolic effects HS, Boerwinkle E. Genomic association analysis identifies associated with antihypertensive medications. Hypertension multiple loci influencing antihypertensive response to an angiotensin ii receptor blocker. Hypertension 2012;59:1204– 5. Elliott WJ, Meyer PM. Incident diabetes in clinical trials of 25. Song KH, Li T, Chiang JY. A prospero-related homeodomain protein is a novel co-regulator of hepatocyte nuclear factor 6. Messerli FH, Bangalore S, Julius S. Risk/benefit assessment of 4alpha that regulates the cholesterol 7alpha-hydroxylase gene.
beta-blockers and diuretics precludes their use for first-line PHARMACOTHERAPY Volume 34, Number 2, 2014 26. Yamagata K, Furuta H, Oda N, et al. Mutations in the hepato- Figure S1. Fasting glucose SNP ARAP1 rs11603334 nominally asso- cyte nuclear factor-4alpha gene in maturity-onset diabetes of ciated with glucose response to atenolol monotherapy among Cau- the young (mody1). Nature 1996;384:458–60.
casian hypertensive patients. Error bars represent standard errors 27. The ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group. Major outcomes in high-risk Figure S2. Fasting glucose SNP SLC2A2 rs11920090 nominally hypertensive patients randomized to angiotensin-converting associated with glucose response to hydrochlorothiazide monother- enzyme inhibitor or calcium channel blocker vs diuretic: the apy among Caucasian hypertensive patients. Error bars represent antihypertensive and lipid-lowering treatment to prevent heart attack trial (ALLHAT). JAMA 2002;288:2981–97.
Table S1. Fasting glucose GWAS SNPs included in this analysis.
Table S2. Other fasting glucose SNPs and association with ateno-lol-induced glucose change.
Table S3. Other fasting glucose SNPs and association with HCTZ-induced glucose change.
The following supporting information is available in the online

Source: https://www.pages.pharmacy.pitt.edu/dailydose2014/wp-content/blogs.dir/28/files/sites/28/2014/02/Prox1-Study.pdf