Endothelial Dysfunction: Do Angiotensin Receptor Blockers Have a Role?

Abstract

Endothelial dysfunction, oxidative stress, and inflammation are several destructive mechanisms of disease, including hypertension, atherosclerosis, and type 2 diabetes.  There is clear evidence linking endothelial dysfunction to abnormal vasodilation with these well‐known risk factors.  This paper will discuss the understanding of endothelial dysfunction as it applies to both hypertension and dyslipidemia models and the pharmacologic approaches available to improve endothelial dysfunction.

Risk factors, whether hypertension, dyslipidemia, diabetes, or smoking, contribute to common and destructive mechanisms of disease, including endothelial dysfunction, oxidative stress, and inflammation. Ultimately, these factors conspire to cause cardiovascular disease and its clinical manifestations. The reason for treating these risk factors is because it is the best way of slowing and even preventing the underlying disease.

Understanding Endothelial Dysfunction: Hypertensive Models

Our group has recently investigated endothelial dysfunction in high‐risk populations, including African Americans and Mexican Americans, trying to understand the cellular basis for different responsiveness to stimuli of endothelial function in these populations. Much interest has focused on the early stages of atherosclerosis, specifically endothelial dysfunction. Normal endothelial cells (ECs) produce a smooth nonthrombogenic surface, which generate nitric oxide (NO) and other important mediators of vasodilation and vascular health. In endothelial dysfunction, these cells transform into a disease state that is first evidenced by the expression of adhesion molecules, which recruit circulating white blood cells into the vessel wall.

With more advanced plaque, endothelial dysfunction precedes a conspicuous accumulation of neutral lipid including cholesterol, esterified cholesterol, and phospholipid. These lipids are subjected to oxidative modification, which leads to further inflammation, including foam cell formation. With activation of T lymphocytes and other inflammatory processes, proteolytic enzymes are released that degrade the fibrous cap, ultimately leading to plaque rupture and thrombus formation. The important point is that risk factors are driving these atherogenic pathways in a direct and even synergistic fashion.

There is clear evidence linking endothelial dysfunction to abnormal vasodilation with well‐known risk factors, including hypertension. Patients with hypertension have decreased responsiveness to a biochemical signal for mediating NO release and vasodilation. There may be genetic contributions to this condition as normotensive offspring of hypertensive individuals already show evidence of abnormal responsiveness to stimuli of NO. ¹ , ² Patients with abnormalities in vasodilation mediated by NO have clinical outcomes far worse than patients with evidence of normal vasodilation. This can be demonstrated in the catheter laboratory by looking at changes in vessel diameter in response to acetylcholine. ³

A paradoxical constriction with acetylcholine is clear evidence of endothelial dysfunction. Nanotechnology can measure NO release in isolated vessels or in cultured cells. But because NO has an exceedingly short half‐life, it is difficult to measure in circulation. In addition, due to rapid conversion to other nitrogen‐containing compounds, circulating levels of NO may be confused for other nitrous species, including peroxynitrite. However, when nanosensors are placed in close proximity to the cell surface (<5 μm), molecules with distinct chemical properties, such as NO, superoxide, and peroxynitrite, can be measured separately. ⁴

An example of this is shown in a study performed in ECs from various vascular beds, including iliac arteries. In white and black American populations, differences were seen in NO release. ⁵ Nebivolol was shown to reduce nitro‐oxidative stress and enhance NO bioavailability in the endothelium of black Americans. If an inhibitor of a free radical–generating enzyme, such as nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, is used in this population, the production of NO can be enhanced, especially in cells from these donors. ⁵ This is because superoxide reacts with NO to form peroxynitrite, thereby reducing the amount of bioavailable NO.

The nanosensor measures the amount of NO released from the cell; it is the bioavailable NO that is released into the subendothelial space to cause vasodilation. Beyond NO, other radicals can simultaneously be measured with the nanosensor. Superoxide is produced at very low levels in normal tissue, but with endothelial dysfunction, there is significant release of superoxide relative to NO. This reaction can be reversed with an inhibitor of NADPH oxidase and other inhibitors of oxidative stress.

In both high‐risk human populations and animal disease models, a significant loss in NO production is observed with either a receptor‐independent agent such as calcium ionophore or receptor‐dependent agent such as acetylcholine. With both stimuli, there is a significant loss in NO bioavailability. ⁶ This may lead one to believe that there are lower NO synthase (NOS) levels. Paradoxically, there are excess NOS levels in these animals. This has also been observed in vessels from hypertensive animals where a loss in NO was observed concomitantly with excessive peroxynitrite production. ⁶ Thus, normal conditions are characterized by low superoxide production within the cell, adequate levels of substrate and cofactors, and normal NO bioavailability. By contrast, in disease or in high‐risk patients, NOS is uncoupled, resulting in a disproportionate amount of superoxide relative to NO. This excessive superoxide reacts with NO to form peroxynitrite, a toxic radical, which directly contributes to disease (Figure 1).

Figure 1.

 Role of endothelium‐dependent nitric oxide synthase (eNOS) uncoupling in endothelial dysfunction. NADPH indicates nicotinamide adenine dinucleotide phosphate; , superoxide; ONOO ⁻ , peroxynitrite; NO, nitric oxide. Adapted from Mason et al. ⁵

Understanding Endothelial Dysfunction: Dyslipidemia Models

Hypertensive models are not alone in displaying NOS uncoupling and endothelial dysfunction. This condition is present with other risk factors, such as hyperlipidemia. With increasing levels of low‐density lipoprotein cholesterol (LDL‐C), the reduction in NO is very evident. ⁷ When LDL‐C becomes oxidized (oxLDL), a more pronounced reduction in NO levels will result. Thus, high LDL‐C levels have an adverse effect on endothelial function and NO bioavailability.

At the subcellular level, cholesterol promotes the formation of domains within the EC plasma membrane, which can reduce NO production. ⁸ Certain domains are flask‐shaped invaginations that sequester proteins such as caveolin‐1, which bind to NOS and thereby prevent its activation. Cholesterol enrichment is a signal to produce caveolae, particularly caveolin‐1, since these domains are enriched with free cholesterol, along with sphingolipids. Cholesterol itself will form domains that have been detected with x‐ray diffraction. These cholesterol crystalline domains precipitate the formation of extracellular crystals, which are highly toxic within the vessel wall. Therefore, there is a link between risk factors, such as hyperlipidemia and hypertension, due to a change in NOS levels or activation, leading to a decrease in NO production.

Oxidized LDL‐C especially reduces endothelial function and yet oxLDL‐C levels are not routinely measured in patients. Our group has looked at different markers of oxLDL and their prognostic values in patients with documented coronary artery disease. This includes an early marker of oxLDL‐C, known as lipid hydroperoxides. This molecule is characterized by the addition of a hydroxyl group with the lipid acyl chain.

In a study of more than 600 patients with stable coronary disease, oxLDL‐C was found to be highly predictive of events, ⁹ independent of overall LDL‐C, high‐density lipoprotein cholesterol (HDL‐C), blood pressure, and a number of other conventional risk factors. This is consistent with other oxLDL‐C markers that have been evaluated. ¹⁰

Pharmacologic Approaches to Improving Endothelial Function

A number of pharmacologic agents used to reduce blood pressure or LDL‐C levels also have additional beneficial effects. With respect to endothelial function, angiotensin‐converting enzyme (ACE) inhibitors prevent the break down of bradykinin, a stimulus of NOS. Statins stabilize the mRNA message for NOS and thereby enhance NOS levels in the cell.

Angiotensin receptor blockers (ARBs) and ACE inhibitors can improve endothelial function in a way that is directly related to NOS and NO bioavailability. A study with ramipril and losartan demonstrated an enhancement in the ability of the vessels to dilate the response to increased blood flow. ¹¹ This process was specifically inhibited with an inhibitor of NOS in patients with coronary disease. The key finding was that the benefit of these renin‐angiotensin‐aldosterone system (RAAS) inhibitors could be related to a reduction in oxidative stress within the vessel wall via inhibition of NADPH oxidase as opposed to specific inhibition of the angiotensin type 2 receptor (AT₂R). Additionally, ARBs block AT₁Rs, leaving the AT₂R available to stimulation by angiotensin II with subsequent stimulation of NOS. This can be considered perhaps an indirect way by which ARBs work. The question is: are there intrinsic effects of ARBs on NO release similar to other classes of agents that may be unrelated to their conventional pharmacology? The Vascular Improvement With Olmesartan Medoxomil Study (VIOS) trial provides some answers.

In this study, olmesartan was compared with atenolol in individuals with stage 1 or 2 hypertension. Using biopsies of gluteal subcutaneous resistance vessels, olmesartan showed a favorable effect on the augmentation index, while there was no effect with atenolol. In addition, interesting structural changes at the wall‐to‐lumen ratio in these vessels was seen, suggesting an effect on vascular remodeling and atherosclerosis. ¹²

In a preliminary, unpublished study by our group using tissue from African Americans and Mexican Americans, the effect of olmesartan was compared with atenolol with respect to NO release from ECs. Even at very high doses, little activity was seen with atenolol; however, olmesartan generated a strong signal even at low concentrations.

Interestingly, however, not all ARBs work the same. When telmisartan was tested in the same model, there was a small decrease in NO release, while other ARBs demonstrated intermediate effects. Therefore, there are clear differences among ARBs with respect to NO bioavailability that may be related to specific physiochemical properties. In particular, olmesartan contains a hydroxyl group that may donate protons capable of diminishing free radical reactions. There is also the presence of conjugated ring structures that may trap and stabilize unpaired electrons (free radicals) in various resonance configurations. (Figure 2).

Figure 2.

 Chemical structures of representative angiotensin receptor blockers.

There is a logical interest in what effect various drug combinations have on endothelial dysfunction, especially when a RAAS inhibitor is combined with amlodipine, which works through an entirely different mechanism, but positively affects NO. ¹³ Amlodipine has unique properties compared with other calcium channel blockers at least with respect to NO. Even the inactive enantiomer of this drug has very potent effects on stimulating NO from ECs. Amlodipine stimulates NOS by binding to a yet‐unidentified angiotensin type receptor that increases the sensitivity of cells to existing levels of bradykinin. Thus, there are multiple pharmacologic ways of stimulating NOS.

Conclusions

What are the unanswered questions? Are there effects of ARBs beyond AT₁R inhibition in terms of stimulating NOS, and might there be differences among the various agents? Are the effects of ARBs on endothelial function more apparent under disease conditions, especially with regard to differences in redox potential and NADPH oxidase? Do these effects play a more important role in certain high‐risk populations, where less NOS is available, but existing in an uncoupled state, or where there are differences in certain cofactors and redox states? The effects of olmesartan and other ARBs combined with other agents such as amlodipine on endothelial function need further study to fully appreciate the potential pharmacologic benefits.

Disclosure:  Dr Mason has received honoraria for speaking and/or serving on advisory boards for Pfizer, Sanofi Aventis, Forest Laboratories, Novartis, and Cardax Pharmaceuticals. He has also received independent investigation initiated grant/research support from Pfizer, Sanofi Aventis, Forest Laboratories, Arca Discovery, and Bristol‐Myers Squibb. The author acknowledges the assistance of Practicum Educational Services in preparing this article and styling the paper for journal submission. Editorial support was provided by Ronald K Miller, PhD, and funded by Daiichi Sankyo, Inc. The author received an honorarium from Daiichi Sankyo, Inc, for time and effort spent preparing this article.