AGT M235T variant shows no link to hypertension in elderly South Africans

A genetic variant that has long been suspected of raising blood pressure through the renin–angiotensin–aldosterone system (RAAS) failed to show any measurable connection to the system’s key parameters in a small, predominantly hypertensive cohort of elderly black South African adults. Researchers genotyped 48 participants aged 60 and older from Sharpeville, South Africa, for the AGT rs699 single nucleotide polymorphism, commonly known as M235T, and then tested whether the variant’s genotypes were associated with plasma concentrations of angiotensin-converting enzyme (ACE), angiotensin I (Ang I) and angiotensin II (Ang II), as well as with blood pressure, body mass index (BMI) and gender. Across univariate, multivariable and Firth-penalized logistic regression models, none of these parameters reached statistical significance. The authors describe the work as a hypothesis-generating descriptive case series and stress that the findings require validation in larger and more diverse populations before any firm conclusions can be drawn.

The RAAS contributes to long-term blood pressure regulation, controlling vascular resistance and volume homeostasis through the actions of renin, angiotensin II and aldosterone. The system also involves angiotensin-converting enzyme and angiotensin I alongside angiotensin II. Because RAAS activation is clinically relevant to hypertension and renal disease, therapies targeting this pathway remain central to blood pressure management. Hypertension itself is a leading global risk factor for morbidity and mortality, affecting more than one billion adults worldwide, yet its underlying mechanisms remain incompletely understood. Any genetic influence on the RAAS could therefore shape cardiovascular risk, which is why the AGT gene, located on chromosome 1q42–q43, has attracted sustained attention from hypertension researchers. The effect of rs699 may also vary with age, because age-related RAAS changes, cumulative environmental exposures and comorbidities can modify genotype–phenotype relationships, and evaluating the variant in older adults may improve cardiovascular risk stratification.

The rs699 variant is a missense change that causes a methionine-to-threonine substitution. Reported effects on plasma angiotensinogen concentrations and RAAS activity suggest a potential contribution to elevated blood pressure and vascular remodelling, but associations with primary hypertension have remained inconsistent across populations and study designs. This heterogeneity may reflect population-specific genetic backgrounds, environmental exposures and interactions with other variants. The inconsistency is particularly relevant in under-studied groups, including older black South Africans, among whom hypertension and uncontrolled blood pressure contribute substantially to cardiovascular disease burden. ClinVar entries in the NCBI database largely classify the variant as having uncertain significance or conflicting interpretations of pathogenicity, which the authors cite as justification for further evaluation in vulnerable elderly populations. Prior consortium-based analyses have also demonstrated age-dependent genetic influences on systolic, diastolic, mean arterial and pulse pressure, strengthening the rationale for examining genetic variation in ageing populations.

The study was embedded within an ongoing longitudinal investigation of an older adult community in Sharpeville that the principal investigators have followed periodically since 2008. Participants were recruited from the only day-care centre in Sharpeville, and inclusion required an age of at least 60 years, voluntary attendance at the centre, residence in Sharpeville and provision of informed consent. Individuals who could not complete the informed-consent process, including those with dementia, or who were unable to complete the required study procedures were excluded. The final sample consisted of 48 black South African participants who were either pensioners or unemployed, with a mean age of 73 years and a predominantly female composition of 81.2 percent. Sample size was constrained by available funding, the cost of laboratory analyses, and attrition related to advanced age, frailty, illness or incomplete study procedures. Medication-intake information could not be obtained with sufficient reliability and was therefore not included in the analysis.

In the laboratory, the team combined genotyping with quantitative protein assays. Genomic DNA was extracted from fasting peripheral blood samples using a commercial purification kit, quantified with spectrophotometry, and assessed for purity using the A260/A280 ratio and agarose gel electrophoresis, with a ratio of approximately 1.8 considered acceptable. Genotyping was performed on the Agena Bioscience MassARRAY system, which applies single-base extension chemistry coupled with matrix-assisted laser desorption/ionization-time-of-flight mass spectrometry. The RAAS parameters were measured with standardized sandwich enzyme-linked immunosorbent assays (ELISAs), in which plasma proteins bound to antibody-coated microplate wells, a biotin-conjugated antibody formed a sandwich complex, and streptavidin-horseradish peroxidase with a tetramethylbenzidine substrate generated a colour reaction proportional to protein quantity, read at 450 nm. Across the ACE, Ang I and Ang II assays, intra-assay coefficients of variation remained below 8 percent and inter-assay coefficients below 10 percent, indicating consistently high within-run and between-run precision.

Blood pressure was measured with an automated upper-arm monitor that combines oscillometric measurement with auscultatory detection of Korotkoff sounds. Participants sat quietly for at least five minutes before readings, and two measurements were obtained on each of two separate days, with mean systolic and diastolic values calculated for analysis. The device achieved an A/A grade under the British Hypertension Society protocol and met the European Society of Hypertension International Protocol criteria, supporting its use in clinical, home and field settings. Hypertension was defined according to the South African Hypertension Society guidelines at 140/90 mmHg, which further divided the cohort into normal and hypertension grade 1 to 3 categories. BMI was calculated as weight divided by height squared, with participants classified according to World Health Organization categories ranging from severely underweight through normal weight and overweight to obesity classes I, II and III.

The genotyping results revealed a markedly skewed distribution. Most participants, 93.8 percent, carried the GG genotype, while only 4.2 percent carried the AG heterozygous genotype, and the SNP was not detected in one participant. Genotyping was successful in 97.92 percent of samples, indicating a low proportion of missing genotype data, and the average DNA yield was 32.81 nanograms per microlitre. The observed genotype distribution conformed to Hardy–Weinberg equilibrium, with no significant deviation detected by either the chi-square goodness-of-fit test (χ2 = 0.0222, df = 1, p = 0.8815) or Haldane’s exact test (p = 1.0000). The authors interpret these quality metrics as evidence of robust genotyping performance supporting the inclusion of rs699 in subsequent analyses. Quality control included the exclusion of samples with call rates below 95 percent, and variants with a minor allele frequency below 0.05 would have been excluded because low-frequency alleles yield unstable estimates and reduced statistical power.

The clinical picture of the cohort was one of substantial cardiovascular risk. Mean systolic blood pressure was approximately 158 mmHg, and roughly 75 percent of participants had grade 1 to 3 hypertension, including nearly 44 percent with grade 2 to 3 hypertension. Adiposity was equally prominent: 17 participants (35.4 percent) were classified as overweight, 13 (27.1 percent) as obesity class I, 8 (16.7 percent) as obesity class II and 1 (2.1 percent) as obesity class III, giving a mean BMI of approximately 29.9 kg/m2. Overweight or obesity therefore affected about 81 percent of the cohort, with nearly 46 percent classified as obese. On the RAAS side, ACE concentrations were below the reference range in 64.6 percent of participants, although one markedly elevated value influenced the overall distribution, while Ang II was elevated in the same percentage. Ang I concentrations remained within the stated reference range among those measured.

When the associations were modelled statistically, the results were uniformly negative. Genotype was treated as a three-level categorical variable, and Firth’s penalized logistic regression was chosen specifically to reduce small-sample bias arising from the low prevalence of the AG genotype and the limited sample size. None of the evaluated parameters—ACE, Ang I, Ang II, blood pressure, BMI or gender—was significantly associated with the rs699 SNP in the univariate, multivariable or Firth-penalized models. To assess robustness under sparse genotype distributions, the team ran four sensitivity specifications: standard maximum-likelihood logistic regression, a baseline Firth model, a Firth model after outlier exclusion, and a Firth model fitted to an imputed dataset. All four yielded consistent findings, with odds-ratio point estimates remaining close to the null value of 1.0, in a range spanning 0.9691 to 1.0957, and all p-values exceeding 0.05. The predominance of the GG genotype may have reduced the statistical power to detect modest effect sizes.

The authors interpret the negative result cautiously rather than as definitive evidence of absence. They note that the rs699 variant alone is unlikely to be a universal predictor of hypertension or RAAS activity, and that its effects depend on broader genetic and environmental contexts. The low ACE concentrations may reflect age-related changes in RAAS activity, genetic variation or antihypertensive treatment, particularly ACE inhibitor use, but because medication use could not be assessed, they should not be attributed solely to pathological dysfunction. Ang II concentrations varied considerably among participants, and some female participants had higher mean Ang II values than males, although substantial variability and the small male sample preclude firm interpretation. The authors also caution that circulating ACE, Ang I and Ang II may not reliably represent RAAS activity, because systemic concentrations do not necessarily reflect tissue-level signalling. Interpretation is further limited by the small sample, gender and genotype imbalances, analysis of a single SNP, the absence of a control group, and the fact that the ELISA kits provided analytical detection ranges rather than universal clinical reference intervals. The team concludes that the findings are hypothesis-generating and do not exclude potentially meaningful effects, warranting adequately powered, genotype-balanced studies in diverse African populations.

Subject of Research: Genetics and cardiovascular research

Article Title: The AGT M235T variant and renin–angiotensin–aldosterone system parameters in a predominantly hypertensive elderly black South African cohort: a hypothesis-generating descriptive case series

Article References: Grobler, C. J., Oldewage-Theron, W., & Chalwe, J. M. (2026). The AGT M235T variant and renin–angiotensin–aldosterone system parameters in a predominantly hypertensive elderly black South African cohort: a hypothesis-generating descriptive case series. Molecular Biology Reports, 53(1), Article 1790. https://doi.org/10.1007/s11033-026-12994-2

Image Credits: AI Generated

DOI: 10.1007/s11033-026-12994-2

Keywords: AGT, M235T, RAAS, hypertension, genetics, South Africa, elderly, ACE, angiotensin, variant, renin, aldosterone