Mechanisms of Action

The renin-angiotensin-aldosterone system regulates blood pressure via release of angiotensinogen from the liver into the bloodstream (Fig. 1, no. 1). Angiotensinogen is cleaved by renin to form angiotensin I (Fig. 1, no. 2) which is then converted to angiotensin II via angiotensin converting enzyme, ACE (Fig. 1, no. 3). Angiotensin II promotes constriction of blood vessels leading to an increase in blood pressure (Fig. 1, no. 4). In addition, angiotensin II leads to the formation of aldosterone (Fig. 1, nos. 5 and 6), which increases reabsorption of sodium and water into the blood, resulting in increased blood pressure (Fig. 1, no. 7). ACEIs inhibit ACE from converting angiotensin I to angiotensin II (Fig. 1, no. 8) thus lowering blood pressure by preventing smooth muscle constriction in vasculature and by reducing the release of aldosterone.

Open in a separate window

Fig. 1

ACE inhibitor interaction with the renin-angiotensin-aldosterone system. Angiotensinogen is released from the liver (no. 1) into the blood where it is converted to angiotensin I by renin (no. 2). Angiotensin I is converted to angiotensin II (ang II) by angiotensin converting enzyme (ACE, no. 3). Ang II binds angiotensin receptors which results in increased blood pressure (no. 4). Ang II also stimulates the production and release of aldosterone (nos. 5 and 6) which leads to increased blood pressure (no. 7). ACE inhibitors prevent the conversion of angiotensin I to ang II (no. 8). Angiotensin receptor blockers (ARBs) compete with Ang II at angiotensin receptor binding sites and result in decreased blood pressure (no. 9). ACE is an important enzyme for the breakdown of bradykinin (no. 12). In the presence of ACE inhibition, bradykinin accumulates and can lead to adverse effects such as cough and angioedema (nos. 10 and 11)

Conversion of angiotensin I to angiotensin II is not the sole pathway that produces angiotensin II; trypsin, cathepsin, or the heart chymase all produce angiotensin II leading to inevitable ceiling effect of blood pressure decline [35]. ACEIs are competitive inhibitors and hormone production can be overcome by increasing angiotensin I substrate in vitro [36, 37]. In addition, ACEIs inhibit bradykinin metabolism leading to side effects such as dry cough and angioedema (Fig. 1, nos. 10 and 11).

ARBs target the angiotensin pathway by blocking the angiotensin II receptor (Fig. 1, no. 9). ARBs lower blood pressure with fewer side effects than ACEIs due to their intervening role further downstream in the angiotensin cascade. Angiotensin II type 1 and type 2 (AT1 and AT2) receptors have been described; however, ARBs relax smooth muscle by blocking the interaction of angiotensin II at the AT1 receptor and AT2 plays only a minor role in the antihypertensive effect of the drug class [35]. AT1 receptors are coupled to Gq and IP3 signal transduction pathways which, when stimulated, lead to smooth muscle contraction and increase in blood pressure (Fig. 1, no. 4). ARBs, like ACEIs, are competitive inhibitors; however, the dissociation of ARB from the receptor is exceptionally slow leading to insurmountable antagonism [36, 37]. In addition, ARBs increase salt and water excretion, thus reducing the plasma volume and subsequently, blood pressure (Fig. 1, no. 9) [38].

Demographic and Clinical Associations

Age and race are two clinical factors physicians use to guide prescribing practices. An ACEI or ARB is recommended as first-line therapy for the general population and patients with diabetes or chronic kidney disease [4]. In the general adult black population, ACEIs and ARBS are a second-line treatment due to this population’s increased risk of cerebrovascular events [39]. Initial treatment in the black population should include a thiazide-type diuretic or a calcium channel blocker (CCB) [4]. Diuretics and CCBs are generally more effective in the elderly population [40].

Patients with chronic kidney disease, regardless of race or diabetes status, are recommended to begin HTN treatment with an ACEI or ARB to improve kidney outcomes [4]. The FDA classifies ACEIs [41] and ARBs [42] as pregnancy Category C in the first trimester and Category D in the subsequent trimesters.

Genetic Variants Associated with Effectiveness

Several investigators have examined genetic variants associated with ACEI and ARB effectiveness (Table 2). The ACE gene has a variant characterized as insertion (I) or deletion (D) of an Alu repeat in intron 16 (17q23.3, rs1799752). Variations at this locus have been associated with myocardial infarction [61, 62] and HTN [63, 64]. In a study of Malay male hypertensive subjects, the D/D genotype conferred a greater blood pressure lowering response to ACEIs (lisinopril and enalapril) [43]. It is hypothesized that ACEIs are particularly effective in patients with the D/D genotype since this genotype has been associated with increased ACE serum levels and higher ACE activity thus leading to more potential for inhibition [44, 45]. There are mixed data regarding the association of angiotensin II levels and ACE I/I, I/D, and D/D genotypes. Studies investigating associations of ACE I/D polymorphism have concluded that although the polymorphism influences the plasma ACE levels, it does not affect angiotensin II or aldosterone production in subjects [65, 66]. However, another study demonstrated that higher plasma levels of angiotensin II were generated in normotensive men with the D/D genotype [67]. A final study concluded that there is a higher level of vascular conversion of angiotensin I to angiotensin II in subjects with the D allele [68]. In the lungs where there is significant excess of ACE, certain ACE genotypes may not affect the net production of angiotensin II.

The I and D polymorphisms have also been studied in the context of ARBs. The Swedish Irbesartan Left Ventricular Hypertrophy Investigation versus Atenolol (SILVHIA) trial made significant advances in identifying single nucleotide variants (SNVs) associated with blood pressure response to ARBs. This trial showed the ARB irbesartan conferred a larger decrease in diastolic blood pressure in I/I genotype compared with D/D or I/D [46], whereas the D/D genotype conferred better response to ACE inhibitors [43].

Renin is the rate-limiting enzyme for the renin-angiotensin-aldosterone system and therefore an important gene to study when considering response to ACEIs and ARBs. A study investigating polymorphisms and response to valsartan demonstrated that CC homozygotes of the REN C5312T polymorphism had an increased response to ARBs [60]. The C5312T polymorphism is located in the distal enhancer region of the human renin gene [69] and this polymorphism is hypothesized to increase transcription of renin. Measuring plasma renin can give insight into the etiology of HTN in patients; however, the predictive value of renin status and response to ACEIs and ARBs remains controversial. Renin is known to decrease with age, thus complicating its predictive value when treating HTN [70–73].

Variants in the CYP11B2 gene (encodes aldosterone synthase) have been associated with variable responses to ARBs. The SILVHIA trial found that subjects with the TT variant at position 344 of CYP11B2 had a more pronounced response to irbesartan than individuals with TC or CC variants (rs1799998) [48]. The variants at position 344 were not related to the baseline serum aldosterone levels [48]. Another independent study found that the CC variant of CYP11B2 conferred a greater response to candesartan than the TC or TT variants [49]. Both of these studies were small and require further investigation regarding this variant and response to ARBs.

In addition to polymorphisms within elements of the angiotensin hormone system itself, variants in the drug metabolizing enzymes that interact with ACEIs and ARBs are associated with drug effectiveness. The SILVHIA trial showed the CYP2C9*2 variant (C430T rs1799853), which predicts less enzymatic activity compared to the wild type CYP2C9*1, reduced the metabolism of losartan and thus improved its treatment effect [50]. Enzyme CYP3A4, in addition to CYP2C9, contributes to losartan metabolism and variants of this enzyme may alter ARB effectiveness [74, 75]. Additional variation in drug metabolizing enzymes is likely, such as CES1 which is responsible for conversion of enalapril and ramipril into their active forms, though CES1 gene variants affecting conversion of ACEIs have not yet been determined.

ARBs conventionally were thought to have shallow dose-response curves. However, these curves were generated using data from both ARB responders and non-responders [76, 77]. For example, patients with the wild type CYP2C9*1 variant which confers increased enzymatic activity metabolize losartan faster than patients with CYP2C9*2 variant which confers less enzymatic activity [50]. Thus, the patient with increased enzymatic activity will demonstrate a flat dose-response curve to an ARB, whereas the patient with decreased enzymatic activity may demonstrate a linear dose-response. Further studies investigating the dose-response curve of ARBs by genotype are needed.

SNVs in ApoA (A1449G), EDNRB (G40A), and eNOS (A498G) were predictors of blood pressure reduction with ARB therapy in a group of 97 Swedish patients with HTN [78]. In particular, the association of the SNVA498G in eNOS and reduction in blood pressure on ARB therapy suggests the endothelium as a site of action for ARBs. This research suggests that the SNVs studied account for about 50% of the reduction in systolic and diastolic blood pressures in their subjects [78].

The SNV resulting in a missense mutation M235T in the AGT gene (coding for angiotensinogen) on chromosome 1 (rs699) may lead to higher plasma levels of angiotensinogen [79]. Carriers of the threonine amino acid change in the angiotensinogen protein had better response to ACEIs in one study [51] but this finding was not successfully replicated in subsequent studies [52].

Finally, ACE gene variant rs4329 has been hypothesized to predict effectiveness of ACE inhibition. Patients homozygous with the major allele (AA) at this locus who were treated with ACEIs showed lower levels of aspartylphenylalanine (Asp-Phe) (a by-product of ACE activity on cholecystokinin) when compared to patients homozygous for the minor allele (GG). This finding suggests that ACE is more fully inhibited by ACEIs in patients homozygous for the major ACE allele [53].

Genetic Variants Associated with Safety

There are several polymorphisms that have been associated with increased or decreased incidence of common side effects of ACEIs; however, many of these associations are weak and further data are needed to determine their validity. Identifying genomic markers that predict a patient’s susceptibility to certain side effects would allow for safer prescribing of ACEIs to hypertensive patients.

As detailed in the previous section, the ACE gene I/D polymorphism has been associated with ACEI and ARB effectiveness, however, a meta-analysis found that ACEI-induced angioedema was not significantly associated with the ACE gene I/D polymorphism [47].

Cough is the most common side effect of ACEIs. The SLCO1B1 gene variant T521C (Val174Ala, rs4149056) is known to have a strong association with increased risk of enalapril-induced cough [54]. SLCO1B1 encodes an organic anion transporter protein, OATP1B1, which is involved in hepatic drug clearance of enalapril [80]. Carriers of T521C variant showed increased plasma concentrations of enalapril. Both ACEIs [80] and ARBs [81, 82] are partly transported into bile and cleared from the plasma by the OATP1B1 transporter. Since enalapril and other ACEIs are cleared less efficiently in individuals with a mutated OATP1B1 transporter, the plasma levels of the drug remain high. High plasma levels of ACEIs lead to increased bradykinin in the blood since bradykinin is broken down into metabolites via ACE (Fig. 1, no. 12). Thus, patients with the T521C variant who are treated with ACEIs are at risk of bradykinin accumulation in plasma and increased incidence of cough (Fig. 1, nos. 10 and 11).

Polymorphisms in the bradykinin receptor BDKRB2 gene (rs8016905) and ABO gene (rs495828) have been associated with ACEI adverse effects [55]. The rs8016905 variant, located in intron 2 of BDKRB2, is associated with ACEI-induced cough and thus may implicate this intron in regulatory functions. Alternately, the rs8016905 polymorphism may be in linkage disequilibrium with a yet undescribed functional variant [55].

The ABO gene polymorphism rs495828 has been identified as a predictor of ACE activity. ACE activity is a possible predictor of ACEI response [56]. It is hypothesized that this variant in the ABO gene may affect aqueous solubility of ACE as well as ACE susceptibility to proteases. ACE molecules carry ABO antigens and the ABO gene encodes glycosyltranferases that transfer monosaccharides to cell surface H antigens. The oligosaccharide moieties are bound to ACE molecules thus affecting the solubility of ACE and ACE susceptibility to proteases. Increased ACE susceptibility to proteases may lead to lower baseline blood pressure or a more effective response to ACEIs at lower doses.

ACE gene variants, rs4459610 and rs4267385, were shown to be protective against ACEI-induced cough [55]. These variants are mapped to the 3′ end of the ACE gene in intronic regions that are hypothesized to have regulatory functions.

A genome-wide association study (GWAS) has identified additional variants associated with ACEI-induced cough. These include variants in the intron of KCNIP4 on chromosome 4 (rs145489027, rs16870989, rs1495509, rs1495509) [57]. A major function of the protein KCNIP4 is regulation of Kv4 potassium channels. Problems regulating Kv4 channels due to a mutation in KCNIP4 gene may lead to stimulation of sensory nerve afferents within the lung that result from accumulation of bradykinin (Fig. 1, no. 11) [83]. Patients with KCNIP4 variants that induce problems regulating Kv4 channels are thus at higher risk of adverse events attributable to ACEIs since ACEIs prevent the breakdown of bradykinin (Fig. 1, no. 10).

Angioedema is a potentially life-threatening side effect of ACEIs. A GWAS analysis in 2012 found that the variant allele with a T substitution (rs500766) in PRKCQ was associated with reduced risk of angioedema, whereas the G allele in ETV6 (rs2724635) was associated with increased risk of ACEI-associated angioedema [58]. PRKCQ and ETV6 genes are involved in regulation of immune function. An increased risk of angioedema has been noted in specific populations including African Americans [84–86], women [86], and older adults [85]. SNVs associated with angioedema vary between African Americans and European Americans. In African Americans, variant rs989692 in the MME gene (which encodes neprilysin) was significantly associated with angioedema [58]. Neprilysin is a secondary mechanism that works with ACE to degrade vasoactive pepeptides bradykinin and substance P (Fig. 2). Variants in the MME gene may result in reduced function of neprilysin which then leads to increased peptide mediators in the plasma in the presence of ACE inhibition. Ultimately, the accumulation of these peptides can result in increased incidence of cough and angioedema (Fig. 2). Patients with decreased neprilysin function are thus at increased risk of ACEI adverse events [58].

Open in a separate window

Fig. 2

Vasoactive peptide breakdown by neprilysin. Neprilysin breaks down bradykinin, substance P, endothelin, and atrial natriuretic peptide (ANP). Mutations in MME gene which encodes neprilysin can lead to decreased neprilysin function and accumulation of vasoactive peptides, especially in the presence of ACE inhibition, thus increasing the risk of cough and angioedema

Angioedema rates are estimated to be 50% higher in women than those in men [86, 87]. XPNPEP2 encodes membranous aminopeptidase P (APP) which contributes to inactivation of bradykinin. The C2399A variant (rs3788853) in XPNPEP2 was shown to be associated with ACEI-induced angioedema in men but not women [59]. While angioedema is more common in women, men may be at higher risk for angioedema when they carry certain variants in the XPNPEP2 gene since men have only one copy of the gene coding for APP. In contrast, women who carry a variant on XPNPEP2 may be protected since they carry two copies of the X chromosome (Fig. 3). Race and sex-specific polymorphisms alter the risk of these common and life-threatening adverse drug events.

Open in a separate window

Fig. 3

XPNPEP2 gene which encodes aminopeptidase P (APP) is located on the long arm of the X chromosome, marked above. APP is an inactivator of bradykinin, and is part of an important alternative bradykinin breakdown pathway in the presence of ACE inhibition. Females have redundancy in the XPENPEP2 gene due to the possession of two X chromosomes, while males only have one copy of the X chromosome and thus one copy of the XPNPEP2 gene. Males are likely more susceptible to adverse events during ACE inhibition due to variation in XPNPEP2 gene than are females

Metabolomic Associations with Effectiveness and Safety

Metabolomics have the potential to more fully elucidate cellular processes in the human body. ACEI therapy has been associated with increased levels of des-Arg(9)-bradykinin, the active metabolite of bradykinin, and increased HWESASXX, a peptide derived from C3 complement degradation (Table 3) [53, 88]. Bradykinin causes vasodilation and the increased level of bradykinin in patients taking ACEIs is supported by the mechanism of action of the drug class. Excess bradykinin accumulation is associated with ACEI-induced cough and angioedema (Fig. 1, no. 11).

ACEI therapy has been associated with decreased levels of Asp-Phe, the product of cholecystokinin (CCK-8) cleavage by ACE [53] (Fig. 4), and decreased levels of phenylalanylphenylalanine, which is hypothesized to be a product of the cleaving action of ACE [88]. In a study of 448 individuals on ACEIs, investigators found decreased levels of Asp-Phe in patients homozygous for the major ACE allele (rs4329, major homozygote AA) and to a lesser extent, in patients heterozygous for the major allele (AG). The Asp-Phe levels did not change in homozygotes for the minor ACE allele (GG) on ACEI therapy, which suggests patients with the minor ACE allele may not benefit fully from ACEI treatment [53]. In addition Asp-Phe, Leu-Ala, alpha-Glu-Tyr, Phe-Ser, and the unknown metabolite {"type":"entrez-nucleotide","attrs":{"text":"X14086","term_id":"56051","term_text":"X14086"}}X14086 were present in different concentrations varying with patient ACE genotype [53].

Open in a separate window

Fig. 4

ACE cleavage of antidiabetogenic cholecystokinin (CCK-8) with by-product aspartylphenylalanine (Asp-Phe). ACEI therapy leads to decreased levels of Asp-Phe. Patients with major ACE allele AA at rs4329 have lower levels of Asp-Phe indicating ACEI therapy is more effective in patients with this variant

Researchers have only begun to examine metabolomic markers of ACEI/ARB drug effectiveness and safety (Table 3) though this approach shows promise for improving drug therapy.

Epigenomics and Proteomics Associated with ACEIs and ARBs

Epigenomics and proteomics are techniques that hold potential for precision medicine. While genomics and metabolomics related to ACEIs and ARBs are emerging in humans, epigenomic and proteomic studies regarding ACEIs and ARBs have advanced further in murine models and may guide precision medicine efforts in humans in the near future.

Stress in utero may increase occurrence of HTN later in life through fetal programming. A murine model showed that maternal low protein diet results in HTN in the offspring. The hypertensive offspring have upregulated expression of AT1b angiotensin receptor gene. Additionally, the AT1b proximal promoter was undermethylated in the offspring. Finally, the hypertensive rats responded well to ACEIs and ARBs indicating the renin-angiotensin-aldosterone system as responsible for the increase in blood pressure [89]. These findings are further supported in humans. Low birth weight children have lower methylation levels of the ACE gene promoter, higher ACE activity, and higher systolic blood pressure than normal birth weight children [90].

HTN leads to cardiovascular protein aggregates similar to aggregates found in cardiac aging. Angiotensin II-infused mice showed more complex and diverse patterns of aggregated cardiac proteins when compared to controls. Notably, hypertensive rats and older rats showed alterations in proteins associated in neurodegenerative diseases. These proteins include ApoE, clusterin, complement C3, H1 histones, HSP90 α and β, catalase, and laminin γ-1. Mitochondrial HSP75 was absent in aged and hypertensive rat hearts [91]. Mitochondrial HSP75 is protective against cardiac hypertrophy and fibrosis [92].

These studies are experimental and researchers can only hypothesize what the findings may mean for humans. Further epigenomic and proteomic exploration is needed in order to develop a systemic biology approach to treating patients.

Funding Dr. Monte receives support from NIH 1 K23 GM110516 and NIH CTSI UL1 TR001082.

Dr. Monte and Hania Flaten receive support from the American Medical Association Foundation.