F1000Research F1000Research 2046-1402 F1000 Research Limited London, UK 10.12688/f1000research.109644.2 Research Article Articles Mitochondrial haplogroup association with fasting glucose response in African Americans treated with a thiazide diuretic

[version 2; peer review: awaiting peer review]

 Minniefield Bre Conceptualization Data Curation Formal Analysis Funding Acquisition Investigation Methodology Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0002-3008-6664 a 1 Armstrong Nicole Conceptualization Data Curation Resources Validation Writing – Review & Editing 1 Srinivasasainagendra Vinodh Conceptualization Data Curation Formal Analysis Resources Software Writing – Review & Editing 1 Tiwari Hemant Conceptualization Formal Analysis Project Administration Resources Supervision Validation Writing – Review & Editing 1 Ballinger Scott Resources Supervision Validation Writing – Review & Editing 1 Chong Zechen Resources Supervision Validation Writing – Review & Editing 1 Aslibekyan Stella Resources Supervision Validation Writing – Review & Editing 2 Arnett Donna Conceptualization Project Administration Resources Supervision Validation Writing – Review & Editing 3 Irvin Marguerite Conceptualization Funding Acquisition Project Administration Resources Supervision Validation Writing – Review & Editing https://orcid.org/0000-0002-1442-2023 1 University of Alabama at Birmingham, Birmingham, AL, 35294, USA 23andMe, Sunnyvale, CA, 94080, USA University of Kentucky, Lexington, KY, 40506, USA bd72m0w@uab.edu

Competing interests: Stella Aslibekyan is employed by 23 and Me.

 26 8 2022 2022 11 758 24 8 2022 Copyright: © 2022 Minniefield B et al. 2022 This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background: Hypertensive African Americans have a ~50% response rate to thiazide diuretic treatment. This contributes to a high prevalence of uncontrolled high blood pressure. Here, we examined the role of the mitochondrial genome on thiazide diuretic treatment response in hypertensive African Americans enrolled in a clinical trial.

Methods: Participants from the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT, n= 4279) were genotyped using the Illumina Infinium Multi-Ethnic Beadchip. Haplotype groups were called using HaploGrep. We used a linear regression analysis to examine the association between mitochondrial haplogroups (L, M, and N) and changes in blood pressure and fasting glucose over six months and two years, respectively.

Results: The analysis revealed a null association between mitochondrial haplogroups M and N versus L for each of the outcomes. In subgroup analysis, the L subclades L1, L2, and L3/L4 (versus L0) were each inversely associated with fasting glucose response (p < 0.05).

Conclusions: This discovery analysis suggests the mitochondrial genome has a small effect on fasting glucose response, but not that of blood pressure, to thiazide diuretic treatment in African Americans.

 Mitochondria Blood Pressure Hypertension Fasting Glucose Haplogroups Drug Response Thiazide Diuretic National Institutes of Health NATIONAL HEART, LUNG, AND BLOOD INSTITUTE 3R01HL123782-02S1 This study was supported by the National Institute of Health (NIH)/ National Heart, Lung, and Blood Institute (R01HL123782 and 3R01HL123782-02S1). The genotyping was funded by R01HL123782. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Amendments from Version 1

This version is not different from the previous, we just update the article type (from case study to research article) and corresponding author (now Marguerite Irvin).

 Introduction

Hypertension (HTN) is a chronic condition leading to various deleterious health concerns including cardiovascular disease (CVD). 1 Reports state African Americans (AAs) have the highest prevalence of HTN (54%) compared to all other American populations (European Americans (46%), Asian Americans (39%), Hispanic Americans (36%), and African-born immigrants (<40%)). 2 They also have the lowest prevalence of controlled HTN among hypertensive patients receiving medication. 2

These disparities may in part be due to differences in response to first-line antihypertensive (AHT) medications. It has been established that AAs respond better to diuretics and calcium channel blockers as opposed to other classes of AHT pharmaceuticals. 3 , 4 Still, a wide distribution of blood pressure (BP) change associated with diuretic treatment has been observed, demonstrating the need to understand factors associated with treatment responses. 5 In particular, genetic factors associated with BP response to thiazide diuretics in AA populations are understudied. Moreover, thiazide diuretics are significantly associated with metabolic side effects, like impaired fasting glucose (FG). 5 Pharmacogenomic studies can offer a solution to some of these problems by identifying genetic contributors to both treatment response and adverse effects. Remarkably, genetic risk factors for BP response to thiazide diuretics are scarce overall, especially in the AA population. 6

Mitochondria are cellular energy production organelles so important to cell function that they have their own independent genome. 7 Importantly, mitochondrial DNA (mtDNA) variation can contribute to an imbalance in cellular energy homeostasis and increase vulnerability to oxidative stress, which has been linked to cardiometabolic diseases like HTN. 8 – 11 The mitochondrial genome shows noteworthy diversity, due to a higher mutation rate than the nuclear genome, where most variants induce small changes in mitochondrial function that may contribute to disease. 12 , 13 Relevant to our study, previous reports have found several mitochondrial genetic variants associated with systolic BP, FG, and other cardiometabolic traits. 13 – 15

Another unique aspect of the mitochondrial genome is that, due to maternal inheritance and lack of recombination events, it displays distinct regional variation (potentially due to natural selection pressures) defining distinct lineages associated with the significant global ethnic groups. 16 – 20 Particularly, macrohaplogroup L is specific to sub-Saharan Africa, with the most ancient mtDNA haplogroups being L0 and L1 which form the basal lineages of the human mtDNA tree, followed by L2 and L3. 20 Previous studies suggest that specific mtDNA haplogroups, especially those specific to sub-Saharan Africa, may be associated with diseases of metabolism in modern times such as CVD, type 2 diabetes (T2D) , and HTN. 8 , 20 , 21 Further, cybrid studies comparing European-origin haplogroups (H and J) and African-origin haplogroups (L) have shown differences in cell growth, oxygen consumption rates, expression of non-energy-related genes, responses to stressors, and rates of glycolysis. 22 , 23

While mitochondrial dysfunction due to mtDNA variation has been consistently associated with cardiometabolic diseases, few studies have examined how variants are associated with responses to common treatments for these diseases (including AHTs) despite mounting evidence that the mitochondrial genome can play a role in drug response. 24 Additionally, most drugs used in cardiology have some degree of mitochondrial toxicity which could be exacerbated by genomic variations. 25 For example, the use of thiazide diuretics was found to decrease mitochondrial function by inducing apoptosis. 21 , 26 Also, a recent study observed links between nuclear variants with mitochondrial function and drug response to other AHT classes. 27 Given established racial and inter-individual differences in response to thiazide diuretics, we investigated whether mtDNA variants and haplogroups give additional insight into BP and FG response to thiazide diuretics treatment in AAs.

 Methods Sample population

All data used in this study were previously acquired from the Genetics of Hypertension Associated Treatments (GenHAT) study, an ancillary pharmacogenetics study of the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). 28 A total of 4526 GenHAT self-reported AAs were randomized to chlorthalidone, a thiazide diuretic, and had sufficient DNA for genotyping (genotyping was conducted during this study using samples from the previous studies described above). Ethical use of all data, study and protocol was fully approved by the University of Alabama at Birmingham Institutional Review Board (Approval No. IRB-160415008). Previous consent from each participating patient was previously acquired during previous studies ( Arnett et al., 2002).

 Clinical data

BP and FG measurements in the ALLHAT study have been described previously. 29 For the current study, BP response was defined as the change in systolic BP (SBP) and the change in diastolic BP (DBP) from baseline to six months post-thiazide diuretic treatment initiation in the trial. FG response was defined as the change in FG from baseline to two years of a treatment since initiation. Other clinical data including body mass index and T2D status were measured at baseline. FG was not a primary outcome measured in ALLHAT and was missing for 71.9% of the sample at the two-year visit. In total, 317 were missing data on change in DBP, 320 were missing data on change in SBP, and 3156 were missing data on change in FG.

 Genotyping and quality control (QC)

GenHAT samples were genotyped using the Expanded Illumina Infinium Multi-Ethnic AMR/AFR Beadchip, using the manufacturer’s protocol (Illumina, Inc., San Diego, CA). Genotypes were called using the Illumina GenomeStudio software. In total, 4296 samples were successfully genotyped, including 1559 mtDNA single nucleotide polymorphisms (SNPs). We set both the missing rate for samples and SNPs to 5% (removing 17 samples and 13 SNPs), and we removed 944 SNP duplicates, 261 singletons, four SNPs with potential heteroplasmic issues, and 26 SNPs with no dbSNP formally assigned rsID. We retained 4279 samples and 311 mtDNA SNPs to conduct a single SNP and haplogroup analysis. After these data were merged with the complete clinical data described above, there were 3962 (317 missing), 3959 (320 missing), and 1123 (3156 missing) samples for the analysis of DBP response, SBP response, and FG response with genotypes, respectively.

 Haplogroup discovery

Haplogroups were obtained for each individual using HaploGrep, 30 , 31 an open-source maternal haplogroup classifier. According to the recommendation from HaploGrep’s developer, we removed a total of 704 samples due to call discrepancy with the classified haplogroup HV, H2, and Z7. After merging this data with the complete clinical data described above, there were 3310, 3308, and 959 samples for the analysis of DBP response, SBP response, and FG response with haplogroups, respectively.

 Statistical methods

We standardized changes in SBP and DBP outcomes by Z-transformation. FG was transformed using an inverse rank-based transformation. We used linear regression models to examine the association between individual mtDNA haplogroup clades L, M, and N and SBP, DBP, and FG response. Next, we examined the association of haplogroup L subclades with each response (DBP n = 2803; SBP n = 2801; FG n = 786) (L subclades sample count was a subset of the original Haplogroup sample set, stated above (n= 3310)). The L3/L4 groups were combined due to small sample sizes ( Figure 1). All models were adjusted for baseline measures (DBP, SBP, or FG), age, and sex. Using the Bonferroni correction, we calculated the level of significance for each outcome (DBP, SNP, and FG response against clades and subclades) to 0.05/ (3 outcomes) = 0.0167. Finally, we conducted a linear regression analysis to examine the association between individual mtDNA variants and SBP, DBP, and FG response under an additive genetic model adjusting for baseline measure (DBP, SBP, or FG), age, sex, and principal components to control for global ancestry (estimated using EIGENSTRATv6.1.4). 32 Using the Bonferroni correction, we set the nominal significance to 0.000053 (=0.05/ (311 SNPs*3 outcomes)).

Sample frequency distribution of mtDNA haplogroup clades L, M, and N. Results Clinical data

Figure 1 shows the distribution of the mtDNA haplogroup clades L, M, and N in the GenHAT sample with clade L being the most common (84.68%). Table 1 presents the cohort baseline characteristics for the haplogroup analysis. The participants had a history of cigarette smoking (27.63%), ~41% had T2D, and ~14% were diagnosed with stage 3 or greater chronic kidney disease ( Table 1). Mean (± standard deviation, SD) baseline SBP was 145.6 mmHg (±15.71 mmHg); DBP was 84.48 mmHg (±10.03 mmHg), and FG was 127.6 mg/dL (±65.93 mg/dL). Mean (±SD) SBP response was -6.51mmHg (±19.13 mmHg); DBP response was -3.01 mmHg (±10.78 mmHg), and FG response was 6.95 mg/dL (±56.83 mg/dL) ( Table 1).

Baseline characteristics of GenHAT sample subset for Haplogroup Analysis.
Characteristics Cases
Age, years 66.1 ± 7.73
Female sex, n (%) 1836 (55.47)
BMI, kg/m 2 30.4 ± 6.75
T2D, n (%) 1351 (40.82)
CKD a, n (%) 459 (14.40)
Cigarette smoker a, yes (%) 772 (27.63)
Baseline SBP, mmHg 145.6 ± 15.71
Baseline DBP, mmHg 84.48 ± 10.03
Baseline FG, mg/dL 127.4.0 ± 65.93
Baseline serum potassium (K+), mmol/L 4.22 ± 0.54
SBP response, mm Hg -6.51 ± 19.13
DBP response, mm Hg -3.01 ± 10.78
FG response, mg/dL 6.95 ± 56.83
BP: time between baseline and follow-up, months 6
FG: time between baseline and follow-up, months 24

BMI, body mass index; T2D, type 2 diabetes; CKD, chronic kidney disease; SBP, systolic blood pressure; DBP, diastolic blood pressure; FG, fasting glucose; BP, meaning both SBP and DBP a CKD was calculated as baseline eGFR ≤ 60 mL/min/1.73 m 2.

Continuous variables are shown as mean ± standard deviation and dichotomous traits are shown as n (%).

 Haplogroup association analysis

There were no significant associations between haplogroup clades (L, M, N) and change in SBP, DBP, or FG ( Table 2). When examining the L subclades, we found no significant association of subclades with change in SBP and DBP. However, the 3 subclades (L1, L2, and L3/L4) compared to reference subclade L0 were significantly associated with change in FG (β = -0.405; -0.354; -0.365, P-value < 0.017), respectively ( Table 2).

Regression analysis Results of haplogroup clades and subclades.
D FG a D DBP b D SBP b
β (SE) P-value β (SE) P-value β (SE) P-value
Clades M and N vs. reference Clade L (DBP n = 3310; SBP n = 3308; FG n = 959)
M 0.226 (0.163) 0.166 -0.059 (0.083) 0.480 -0.110 (0.087) 0.207
N 0.020 (0.088) 0.822 -0.072 (0.043) 0.092 -0.019 (0.045) 0.669
Subclades L1, L2, L3/L4 vs. L0 (DBP n = 2803; SBP n = 2801; FG n = 786)
L1 -0.405 (0.134) 0.003 -0.004 (0.064) 0.951 0.071 (0.669) 0.286
L2 -0.354 (0.133) 0.008 0.047 (0.062) 0.448 0.107 (0.653) 0.102
L3/L4 -0.365 (0.135) 0.007 0.015 (0.064) 0.816 0.064 (0.671) 0.342

DBP: diastolic blood pressure; SBP: systolic blood pressure; SE: standard error.

All models were adjusted for age, sex, and baseline measure that matched the outcome (baseline FG, baseline DBP, and baseline SBP).

Outcome variable D FG was rank-based inverse transformed.

Outcome variables D DBP and D SBP were standardized.

 Single variant association analysis

No mtDNA SNP was associated with SBP, DBP, or FG response after correction for multiple testing. SNPs with marginal significance are presented in Table 3. In total, we observed one mtDNA SNP that was marginally associated with a change in SBP, six SNPs with a change in DBP, and 11 SNPs with a change in FG with P-value < 0.05.

Association analyses of mtDNA SNPs with baseline DBP and FG.
ID Trait dbSNP ID BP MAF BETA SE P Genes *
exm2216342BOT-1 SBP rs1116906 8460 0.015 -0.109 0.056 0.05204 ATP8
Seq-rs28358578BOT-2 DBP rs28358578 2332 0.111 0.061 0.021 0.0028 ND1
exm2216198 DBP rs28358570 921 0.059 -0.076 0.027 0.0054 -
Seq-rs28359170BOT-1 DBP rs28359170 12236 0.101 0.047 0.021 0.0297 ND4 - ND5
Seq-rs28358573BOT-2 DBP rs28358573 1442 0.101 0.0465 0.021 0.0300 ND1
Seq-rs41352249BOT-2 DBP rs41352249 6680 0.041 -0.069 0.032 0.0314 CO1
Seqrs202123618-TOP-1 DBP rs202123618 3434 0.018 0.101 0.048 0.0339 ND1
Seq-rs2853502BOT-2 FG rs2853502 13276 0.051 0.171 0.063 0.0096 ND5
Seq-rs2853487TOP-3 FG rs2853487 10589 0.054 0.169 0.062 0.0106 ND4L
rs373606184 FG rs3020601 5442 0.055 0.153 0.061 0.0145 ND2
Seqrs200931747-BOT-2 FG rs193303033 4312 0.049 0.168 0.064 0.0151 ND1 - ND2
Seq-rs2853489TOP-1 FG rs2853489 1122 0.012 0.340 0.140 0.0158 ND4
exm2216342BOT-1 FG rs1116906 8460 0.015 0.304 0.121 0.0213 ATP8
Seq-rs28357376BOT-2 FG rs28357376 15824 0.068 -0.125 0.053 0.0336 CYB
Seq-rs28445203BOT-2 FG rs28445203 1123 0.016 0.246 .118 0.0372 ND1 - ND2
exm2216469BOT-1 FG rs35070048 1109 0.059 -0.116 0.057 0.0428 CYB
exm2216384 FG rs28358274 10086 0.068 -0.116 0.053 0.0460 ND3
Seq-rs2854125BOT-2 FG rs201893071 1095 0.068 0.113 0.057 0.0479 CYB

Gene – Gene: SNP is located between the 2 Genes.

DBP, diastolic blood pressure; FG, fasting glucose; SNP, single nucleotide polymorphism; BP, base pair; MAF, minor allele frequency; BETA, parameter estimate from regression model; SE, standard error of BETA; P, SNP p-value.

Genes were annotated from Database of Single Nucleotide Polymorphisms (dbSNP [Sherry ST, et al. dbSNP: the NCBI database of genetic variation. Nucleic Acids Res, 2001; 29:308-311.])

 Discussion

This study examined the association between mtDNA genetic variants and haplogroups and BP and FG response to a thiazide diuretic within a large sample of hypertensive AAs. Results from the single variant analysis did not remain statistically significant after correcting for multiple testing, however the haplogroup analysis suggested FG response may differ within the L subclade. The discoveries obtained here complement previous research showing association of mtDNA variation with cardiometabolic traits and importantly builds on pharmacogenetics research focused on first-line AHT response in AAs. 13 , 33

Consistent with other studies of this population, the majority of GenHAT samples were classified into the mtDNA haplogroup L clade (over 80% of the sample, see Figure 1). 34 , 35 When considering the major clades (L, M, N) in the GenHAT population, we did not observe an association with any of the response outcomes. However, within the L clade, subclades L1, L2, and L3/L4 (versus L0) had an inverse association with change in FG, suggesting a potential protective effect. While few studies have reported on the association of mtDNA haplogroups with FG changes during AHT treatment, some research suggests haplogroups are associated with T2D. 33 , 36 For instance, Sun and colleagues reported women living with HIV possessing haplogroup L2 (versus non-L2 groups of L0/L3/other) had a 49% lower risk of developing T2D, suggesting haplogroup L2 may potentially be a protective factor. 35 Other reports of haplogroup association with T2D are mostly in other race groups. The haplogroups J/T and T were found to be associated with an increased risk of diabetes in Europeans. 37 The M8a, 38 B4, and D4 39 haplogroups have an association with T2D in East Asians. However, the haplogroup N9a 40 has been suggested to confer resistance against T2D in the Japanese and Korean populations and to be a protective factor against metabolic syndrome in Japanese women. 41 With multiple studies suggesting haplogroups may be related to T2D within race groups, more research on L subclades and FG levels during AHT treatment is warranted.

In the single variant analysis, we observed SNPs with marginal association to SBP, DBP, and FG response to chlorthalidone. These SNPs were in or near NADH-ubiquinone oxidoreductase chain 1 ( ND1), NADH-ubiquinone oxidoreductase chain 2 ( ND2), NADH-ubiquinone oxidoreductase chain 4 ( ND4), NADH-ubiquinone oxidoreductase chain 5 ( ND5), and cytochrome B ( CYB). Though few studies report on mtDNA SNPs and response to AHT treatment, researchers continue to discover significant associations between mtDNA variants and cardiometabolic traits across different ancestries (e.g . European, 13 , 14 , 42 and African 13 , 42 ). For example, Liu and colleagues reported a strong association between a nonsynonymous mtDNA variant (m.5913G>A in the cytochrome c oxidase subunit 1 of respiratory complex IV) and SBP and FG in a sample of 7000 European Americans from the Framingham Heart Study (minor allele being associated with a > 5 mm Hg and > 15 mg/dL higher SBP and FG level on average, respectively). 14 Similarly, Buford and colleagues examined the association between mtDNA variants measured by sequencing and SBP and mean arterial pressure across several ancestries (European ( n =1739), African ( n = 683), and others ( n = 121)). 42 They detected a statistically significant association of mtDNA variants with higher SBP (e.g. m.93 A>G in hypervariable segment II (HVII); P-value = 0.0001) and mean arterial pressure (e.g. m.16172T>C in a hypervariable segment I (HVI); P-value = 0.0002), respectively within African-ancestry samples. 42 Interestingly, these specific variants were low-frequency in the other race groups and not associated with these traits. Overall, these studies are supportive of a role for mtDNA variants in BP and FG; however, similar associations were not extrapolated to response to AHT treatment in our study.

This study is not without limitations. First, the coverage of mtDNA variants was restricted to GWAS data. Overall, there are no “gold standard” methods for calling mtDNA haplogroup with GWAS data and we used standard methodology adopted by another study for the same GWAS array. 43 Ultimately, deeper genomic coverage may enhance SNP and haplogroup detection using available software. Further, FG was missing on a considerable number of samples in our study. This is a limitation of the ALLHAT study since BP and cardiovascular disease outcomes were the main focus. Still, several published studies have made use of the available glucose data in this large, NIH-funded clinical trial. 5 , 44 – 46 Despite these limitations, this study remains the largest AA pharmacogenomics study to examine mtDNA variants and haplogroups with BP and FG response.

In summary, this study represents the most extensive investigation of the role of the mitochondrial genome in BP and FG response to chlorthalidone (a thiazide diuretic) within AAs. The overwhelming percentage of AAs suffering from HTN reveals the urgency to discover causal determinants of better BP response to treatment. Despite the urgency, there remains a lack of pharmacogenomic studies within hypertensive AAs. Hence, this study provides the inclusive research that is required to aid understanding of AHT treatment response and eventually facilitate better blood pressure control. The discoveries presented in this study support future investigations of mitochondrial genomics in metabolic response to chlorthalidone among AAs.

 Data availability

dbGaP: Genetics of Hypertension Associated Treatment (GenHAT) Study, Genomic Background of Blood Pressure Response to Treatment in African Americans, accession number: phs002716; https://identifiers.org/dbgap:phs002716 (contact: Marguerite R Irvin, irvinr@uab.edu).

 Acknowledgements

The authors are thankful to both ALLHAT and GenHAT for the use of their datasets.

 References Merai R Siegel C Rakotz M : CDC Grand Rounds: A Public Health Approach to Detect and Control Hypertension. MMWR Morb. Mortal. Wkly Rep. 2016;65:1261–1264. 27855138 10.15585/mmwr.mm6545a3 Centers for Disease Control and Prevention (CDC): Hypertension Cascade: Hypertension Prevalence, Treatment and Control Estimates Among US Adults Aged 18 Years and Older Applying the Criteria From the American College of Cardiology and American Heart Association’s 2017 Hypertension Guideline—NHANES 2013–2016.external icon. Atlanta, GA: US Department of Health and Human Services;2019. Williams SK Ravenell J Seyedali S : Hypertension Treatment in Blacks: Discussion of the U.S. Clinical Practice Guidelines. Prog. Cardiovasc. Dis. 2016;59:282–288. 27693861 10.1016/j.pcad.2016.09.004 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J. Am. Coll. Cardiol. 2018;71:127–248. 10.1161/HYP.0000000000000076 Barzilay JI Davis BR Cutler JA : Fasting glucose levels and incident diabetes mellitus in older nondiabetic adults randomized to receive 3 different classes of antihypertensive treatment: a report from the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Arch. Intern. Med. 2006;166:2191–2201. 17101936 Franceschini N Chasman DI Cooper-DeHoff RM : Genetics, ancestry, and hypertension: implications for targeted antihypertensive therapies. Curr. Hypertens. Rep. 2014;16:461. 24903233 10.1007/s11906-014-0461-9 Taanman WJT : The mitochondrial genome: structure, transcription, translation, and replication. Bioenergetics. 1999;1410:103–123. 10076021 10.1016/S0005-2728(98)00161-3 Watson B Jr Khan MA Desmond RA : Mitochondrial DNA mutations in black Americans with hypertension-associated end-stage renal disease. Am. J. Kidney Dis. 2001;38:529–536. 11532685 10.1053/ajkd.2001.26848 Pravenec M Hyakukoku M Houstek J : Direct linkage of mitochondrial genome variation to risk factors for type 2 diabetes in conplastic strains. Genome Res. 2007 Sep;17(9):1319–1326. 17693571 10.1101/gr.6548207 Houštek J Hejzlarová K Vrbacký M : Nonsynonymous variants in mt-Nd2, mt-Nd4, and mt-Nd5 are linked to effects on oxidative phosphorylation and insulin sensitivity in rat conplastic strains. Physiol. Genomics. 2012 May 1;44(9):487–494. 22414913 10.1152/physiolgenomics.00156.2011 Houštěk J Vrbacký M Hejzlarová K : Effects of mtDNA in SHR-mtF344 versus SHR conplastic strains on reduced OXPHOS enzyme levels, insulin resistance, cardiac hypertrophy, and systolic dysfunction. Physiol. Genomics. 2014;46(18):671–678. 25073601 10.1152/physiolgenomics.00069.2014 Neiman M Taylor DR : The causes of mutation accumulation in mitochondrial genomes. Proc. Biol. Sci. 2009;276:1201–1209. 10.1098/rspb.2008.1758 Kraja AT Liu C Fetterman JL : Associations of Mitochondrial and Nuclear Mitochondrial Variants and Genes with Seven Metabolic Traits. Am. J. Hum. Genet. 2019;104:112–138. 30595373 10.1016/j.ajhg.2018.12.001 Liu C Yang Q Hwang SJ : Association of genetic variation in the mitochondrial genome with blood pressure and metabolic traits. Hypertension. 2012;60:949–956. 22949535 10.1161/HYPERTENSIONAHA.112.196519 Rysz J Franczyk B Rysz-Górzyńska M : Pharmacogenomics of Hypertension Treatment. Int. J. Mol. Sci. 2020;21(13):4709. 32630286 10.3390/ijms21134709 Mishmar D Ruiz-Pesini E Golik P : Natural selection shaped regional mtDNA variation in humans. Proc. Natl. Acad. Sci. U. S. A. 2003;100(1):171–176. 12509511 10.1073/pnas.0136972100 Ruiz-Pesini E Mishmar D Brandon M : Effects of purifying and adaptive selection on regional variation in human mtDNA. Science. 2004;303(5655):223–226. 14716012 10.1126/science.1088434 Wallace DC : A mitochondrial paradigm of metabolic and degenerative diseases, aging, and cancer: a dawn for evolutionary medicine. Annu. Rev. Genet. 2005;39:359–407. 16285865 10.1146/annurev.genet.39.110304.095751 Fonseca RR da : The adaptive evolution of the mammalian mitochondrial genome. BMC Genomcis. 2008;9:119. 18318906 10.1186/1471-2164-9-119 Soares P Ermini L Thomson N : Correcting for purifying selection: an improved human mitochondrial molecular clock. Am. J. Hum. Genet. 2009;84:740–759. 19500773 10.1016/j.ajhg.2009.05.001 Bregman JA : Mitochondrial Haplogroups Affect Severity But Not Prevalence of Diabetic Retinopathy. Invest. Ophthalmol. Vis. Sci. 2017;58(2):1346–1351. 28245487 10.1167/iovs.16-20616 Kenney MC Chwa M Atilano SR : Molecular and bioenergetic differences between cells with African versus European inherited mitochondrial DNA haplogroups: implications for population susceptibility to diseases. Biochim. Biophys. Acta. 2014;1842(2):208–219. 24200652 10.1016/j.bbadis.2013.10.016 Kenny MC Falatoonzadeh P Atilano SR : African-origin Mitochondrial DNA Variants as a Contributing Factor to Suscptibilities for Diabetes and Age-related Diseas. Int. J. Diabetes Clin. Res. 2016;3:053. 10.23937/2377-3634/1410053 Jones SW Ball AL Chadwick AE : The Role of Mitochondrial DNA Variation in Drug Response: A Systematic Review. Front. Genet. 2021;12:1430. 10.3389/fgene.2021.698825 Finsterer J Zarrouk-Mahjoub S : Mitochondrial toxicity of cardiac drugs and its relevance to mitochondrial disorders. Expert Opin. Drug Metab. Toxicol. 2015;11(1):15–24. 25331002 10.1517/17425255.2015.973401 Schirris TJ Renkema GH Ritschel T : Statin-induced myopathy is associated with mitochondrial complex III inhibition. Cell Metab. 2015;22:399–407. 26331605 10.1016/j.cmet.2015.08.002 Hiltunen TP Donner KM Sarin AP : Pharmacogenomics of hypertension: A genome-wide; placebo-controlled cross-over study; using four classes of antihypertensive drugs. J. Am. Heart Assoc. 2015;4. 10.1161/JAHA.114.001521 Lynch AI Boerwinkle E Davis BR : Antihypertensive pharmacogenetic effect of fibrinogen-beta variant -455G>A on cardiovascular disease, end-stage renal disease, and mortality: the GenHAT study. Pharmacogenet. Genomics. 2009;19:415–421. 19352213 10.1097/FPC.0b013e32832a8e81 ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group: The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA. 2002;288:2981–2997. 10.1001/jama.288.23.2981 Weissensteiner H Pacher D Kloss-Brandstätter A : HaploGrep 2: mitochondrial haplogroup classification in the era of high-throughput sequencing. Nucleic Acids Res. 2016;44:W58–W63. 27084951 10.1093/nar/gkw233 Gudiseva HV Pistilli M Salowe R : The association of mitochondrial DNA haplogroups with POAG in African Americans. Exp. Eye Res. 2019;181:85–89. 30653964 10.1016/j.exer.2019.01.015 Price AL Patterson NJ Plenge RM : Principal components analysis corrects for stratification in genome-wide association studies. Nat. Genet. 2006 Aug;38(8):904–909. 16862161 10.1038/ng1847 Fang H Hu N Zhao Q : mtDNA Haplogroup N9a Increases the Risk of Type 2 Diabetes by Altering Mitochondrial Function and Intracellular Mitochondrial Signals. Diabetes. 2018;67:1441–1453. 29735607 10.2337/db17-0974 Johnson DC Shrestha S Wiener HW : Mitochondrial DNA diversity in the African American population. Mitochondrial DNA. 2015;26(3):445–451. 24102597 10.3109/19401736.2013.840591 Sun J Brown TT Tong W : African Mitochondrial DNA Haplogroup L2 Is Associated With Slower Decline of β-cell Function and Lower Incidence of Diabetes Mellitus in Non-Hispanic, Black Women Living With Human Immunodeficiency Virus. Clin. Infect. Dis. 2020;71:e218–e225. 10.1093/cid/ciaa026 Alwehaidah MS Bakhiet M AlFadhli S : Mitochondrial Haplogroup Reveals the Genetic Basis of Diabetes Mellitus Type 2 Comorbidity in Psoriasis. Med. Princ. Pract. 2021;30(1):62–68. 32629455 10.1159/000509937. Epub. 2020 PMC7923845 Crispim D Canani LH Gross JL : The European-specific mitochondrial cluster J/T could confer an increased risk of insulin-resistance and type 2 diabetes: An analysis of the m.4216T >C and m.4917A >G variants. Ann. Hum. Genet. 2006;70:488–495. 10.1111/j.1469-1809.2005.00249.x Guo LJ Oshida Y Fuku N : Mitochondrial genome polymorphisms associated with type-2 diabetes or obesity. Mitochondrion. 2005;5:15–33. 10.1016/j.mito.2004.09.001 Liou CW Chen JB Tiao MM : Mitochondrial DNA coding and control region variants as genetic risk factors for type 2 diabetes. Diabetes. 2012;61:2642–2651. 22891220 10.2337/db11-1369 Fuku N Park KS Yamada Y : Mitochondrial haplogroup N9a confers resistance against type 2 diabetes in Asians. Am. J. Hum. Genet. 2007;80:407–415. 17273962 10.1086/512202 Tanaka M Fuku N Nishigaki Y : Women with mitochondrial haplogroup N9a are protected against metabolic syndrome. Diabetes. 2007;56:518–521. 10.2337/db06-1105 Buford TW Manini TM Kairalla JA : Mitochondrial DNA Sequence Variants Associated With Blood Pressure Among 2 Cohorts of Older Adults. J. Am. Heart Assoc. 2018;7(18):e010009. 30371200 10.1161/JAHA.118.010009 Cui QN Ramakrishnan MS Gudiseva HV : Mitochondrial haplogroup L1c2 is associated with increased disease severity in African American patients with primary open-angle glaucoma. J. Clin. Exp. Ophthalmol. 2019;10(2):799. Whelton PK Barzilay J Cushman WC : Clinical outcomes in antihypertensive treatment of type 2 diabetes, impaired fasting glucose concentration, and normoglycemia: Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Arch. Intern. Med. 2005;165(12):1401–1409. 15983290 10.1001/archinte.165.12.1401 Black HR Davis B Barzilay J : Metabolic and clinical outcomes in nondiabetic individuals with the metabolic syndrome assigned to chlorthalidone, amlodipine, or lisinopril as initial treatment for hypertension: a report from the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Diabetes Care. 2008;31(2):353–360. 18000186 Irvin MR Lynch AI Kabagambe EK : Pharmacogenetic association of hypertension candidate genes with fasting glucose in the GenHAT Study. J. Hypertens. 2010;28(10):2076–2083. 20577119 10.1097/HJH.0b013e32833c7a4d