According to the World Health Organization, cardiovascular diseases are one of the three leading causes of death worldwide (World Health Organization, 2020). The leading risk factor for cardiovascular diseases is high blood pressure through increased strain on the heart and blood vessels, which can lead to conditions such as heart attacks, strokes, and heart failure. Hypertension (HTN) contributes to the development of atherosclerosis, further increasing the risk of cardiovascular events (Whelton et al., 2018). Evidence suggests that genetic predispositions for developing elevated blood pressure exist (Garcia et al., 2003). One of the several genes implicated in HTN risk is CYP3A5 (Xi et al., 2011). CYP3A5 is a member of the superfamily of cytochrome P450 enzymes that plays a crucial role in the metabolism of a wide variety of endogenous and exogenous substances, including drugs, steroids, and toxins. This enzyme group is particularly important as it contributes to the metabolism of many clinically used drugs such as tacrolimus, cyclosporin, and midazolam (Klyushova et al., 2022). The expression of the CYP3A5 gene and CYP3A5 enzyme activity are highly variable among populations. While the CYP3A5*1 allele is functional, the CYP3A5*3 variant, which is most frequent in European ancestry, leads to incorrect splicing and a non-functional CYP3A5 enzyme. The CYP3A5*6 and *7 variants, which are overall rare but are relatively more frequent in African ancestry, also lead to a non-functional CYP3A5 enzyme. The CYP3A5*6 variant causes alternative splicing and deletion, while CYP3A5*7 results in a premature stop codon (Suarez-Kurtz et al., 2014).

Increased expression of CYP3A5 in the kidney is thought to lead to elevated blood pressure, and ultimately HTN, through increased metabolism of cortisol to 6-hydroxy-cortisol, leading to increased sodium and water retention. Since the CYP3A5*3, *6, and *7 variants lead to a non-functional CYP3A5 enzyme, individuals carrying these variants may have reduced cortisol metabolism and sodium retention, potentially reducing their risk of hypertension compared to CYP3A5*1 carriers (Lamba et al., 2002; Wojnowski, 2004; Xi et al., 2011; Zhang et al., 2010). Previous studies investigating the association between CYP3A5 polymorphisms and elevated blood pressure or HTN have shown conflicting results. Many of these studies were limited by small sample sizes, which constrained their statistical power to detect modest effect sizes (Xi et al., 2011; Galaviz- Hernández et al., 2020). The UK Biobank (UKB), with its large, well-characterized cohort and standardized recruitment and measurement protocols, provides a unique opportunity to overcome these limitations and improve the robustness of genetic association analyses. Therefore, we aimed to analyze associations between CYP3A5 variants and blood pressure or risk of HTN development in a study cohort from the UKB to better characterize the cardiovascular relationships of CYP3A5.

In this study, we aimed to assess potential associations between CYP3A5 variants, CYP3A5 activity and multiple blood pressure-related phenotypes in the UKB population. CYP3A5 is an important pharmacogene involved in the metabolism of several drugs. Our findings suggest that there is no meaningful association between CYP3A5 polymorphism and blood pressure and HTN risk. While we observed a statistically significant association between CYP3A5*3 and higher DBP and CYP3A5 activity and lower DBP in the White participants subgroup, the very small effect sizes (approximately 0.1 mmHg) suggest little clinical relevance. Furthermore, the absence of significant findings in non-White subgroups reduces the significance of these associations, suggesting that the associations in White participants may reflect other influences like population specific factors instead of a general genetic association.

Prior research on the relationship between CYP3A5 variants and blood pressure has produced mixed results. Some studies have linked the loss-of-function variant CYP3A5*3 with lower blood pressure (Bochud et al., 2006; Givens et al., 2003; Jin et al., 2007) and lower HTN risk (Ho et al., 2005; Kivistö et al., 2005) suggesting that, compared to a functional CYP3A5*1 allele, the loss-of-function variants lead to lower blood pressure. Conversely, other studies found the CYP3A5*3 allele to be associated with higher blood pressure (Galaviz-Hernández et al., 2020; Fromm et al., 2005; Ho et al., 2005; Kreutz et al., 2005; Zhang et al., 2010) and increased HTN risk (Galaviz-Hernández et al., 2020). The effect sizes observed in previous studies were variable, but most of them were small (Xi et al., 2011). A meta-analysis by Xi et al. found that overall, there is no significant association between CYP3A5 variants and blood pressure, or HTN based on the previous literature (Xi et al., 2011).

One potential reason for the conflicting findings in earlier research is that some of these studies were underpowered to detect such small effect sizes, with sample sizes of less than 500 individuals (Bochud et al., 2006; Fromm et al., 2005; Ho et al., 2005; Galaviz-Hernández et al., 2020; Givens et al., 2003; Jin et al., 2007; Kivistö et al., 2005; Zhang et al., 2010). Moreover, most studies did not account for all relevant non-functional variants, such as CYP3A5*6 and *7, which are common in some of the ancestral populations that they studied (e.g., African and Native American), potentially introducing confounding by misclassifying variant carriers as non-carriers (Galaviz-Hernández et al., 2020; Givens et al., 2003; Fromm et al., 2005; Ho et al., 2005). Notably, two studies that did consider CYP3A5 activity scores by incorporating two loss of function variants, CYP3A5*3 and *6, found no association with blood pressure rates (Fisher et al., 2016; Langaee et al., 2007). Consistent with this, our analysis of genotype-estimated CYP3A5 activity score was conducted with consideration of the *3, *6, and *7 alleles and similarly showed no clinically appreciable association with blood pressure or HTN. By including all three major functional variants, this approach offers a more comprehensive assessment, as it accounts for the participant’s haplotype profile and consequently reflects their CYP3A5 activity more accurately.

In subgroup analyses, we observed statistically significant associations between CYP3A5*3 and higher DBP as well as between CYP3A5 activity and lower DBP in White individuals (by far the largest in our population). The Black, South Asian and East Asian subgroups showed no significant associations. However, the very small effect sizes observed in the White participants subgroup suggest that each CYP3A5*3 allele is associated with only 0.1 mmHg increase in blood pressure. Likewise, each one-point increase in CYP3A5 activity score was associated with only a 0.1 mmHg decrease in blood pressure. These results underscore that statistical significance in large-scale studies does not inherently translate to clinical relevance, a critical consideration for interpreting pharmacogenomic findings. Therefore, even though these findings were statistically significant (likely only due to our large power to detect very small effect sizes), the clinical significance of these findings seems negligible, as supported by the lack of association with HTN risk. Moreover, in our study, CYP3A5*6 or *7 alone were not associated with blood pressure or HTN, which contrasts the theory that loss-of-function variants result in lower blood pressure. The results from the White participants subgroup could not be replicated in any other racial subgroup, which diminishes the generalizability and significance of the findings within the White population. Considering that our primary results did not show any associations between CYP3A5 variants or CYP3A5 activity and blood pressure-related phenotypes and given that the significant findings in the subgroups are restricted to the White population, this suggests the presence of other factors influencing blood pressure and HTN, which may account for the differences observed between White participants and other ancestral populations, as well as the lack of findings within the whole population. These factors, rather than CYP3A5 polymorphism, could potentially explain the significant findings in the White population.

Our study has several strengths. First, the large sample size of nearly half a million participants provides substantial power to detect subtle genetic effects if they exist. Additionally, this study included all known common genetic variants associated with loss of CYP3A5 function. This inclusive approach enables a nuanced assessment of genetic variability and its influence on blood pressure and HTN. Moreover, we adjusted for clinical factors and medications that influence blood pressure and we also conducted a subgroup analysis by race, which is particularly important given the different frequencies of CYP3A5 variants across populations.

Our study also has some limitations. The drug usage data was self-reported, which could introduce inaccuracies, and the diagnosis codes for HTN may not be the most reliable method for determining true HTN status. Moreover, while our study included a large overall sample, the smaller sizes of some racial subgroups limit the ability to detect small effect sizes within those populations. Considering that the cohort sizes still exceeded 1,500 individuals in each subgroup, the sample sizes remain substantial enough to detect clinically significant associations. Despite including over 1,500 non-White participants, we did not observe significant associations in these subgroups. This may suggest that the associations observed in White participants might be due to environmental factors rather than a generalizable genetic effect or, given the very small observed effect size, simply chance findings. Furthermore, differences in allele frequency and limited statistical power to detect such a small effect size in non-White groups may have contributed to the lack of replication. Future studies with larger and more diverse cohorts are needed to clarify these associations. Additionally, epigenetic modifications and environmental factors may play a role in gene regulation and hypertension risk. Future research integrating more refined environmental measures may provide additional insights into gene-environment interactions in hypertension.

In conclusion, our study provides evidence of a lack of associations between CYP3A5 variants with known functional consequences on enzyme activity, or the CYP3A5 activity score, and blood pressure and HTN. These findings suggest that common functional polymorphisms in CYP3A5 may not play a significant role in blood pressure regulation or the risk of developing HTN. Future research should continue to explore other factors that may influence blood pressure and HTN, including potential interactions between genetic variants and environmental or lifestyle factors.