Isoprenoid Metabolism and the Pleiotropic Effects of Statins

Abstract

Convincing evidence from basic research and animal studies shows that HMG CoA reductase inhibitors or statins, exert cardiovascular protective effects beyond cholesterol-lowering. Because of the central role of LDL-C in mediating vascular pathology and the efficacy of statins for lowering LDL-C, the clinical importance of these additional non-lipid effects remains to be determined. Nevertheless, there is growing evidence from recent clinical trials, which suggests that some of the beneficial effects of statins may be unrelated to changes in LDL-C. Indeed, in animal studies, many of the cholesterol-independent or “pleiotropic” effects of statins are due predominantly to inhibition of isoprenoid, but not cholesterol synthesis. Thus, with the recent findings of the HPS and ASCOT-LLA, the potential cholesterol-independent effects of statins have shifted the treatment strategy from numerical lipid parameters to the global assessment of cardiovascular risks.

Introduction

At the present time, there is on-going debate as to whether there are cholesterol-independent effects of statins, which contribute to the clinical outcome in vascular disease, or whether all the beneficial effects of statins can be attributed to lipid lowering. The word “pleiotropic” is derived from a Greek word, pleion, meaning ‘more’ and trepein, meaning ‘to turn’, and is usually applied to genetics and the multiple actions of a single gene. However, in the context of a drug, it can be interpreted as that drug having more than one effect or action. Such is the case with statin therapy.

Strong evidence from animal experiments indicates that inhibition of HMG CoA reductase leads to more than one beneficial effect. However, whether stains exert significant effects beyond cholesterol lowering in humans is much less clear. Given the great importance of lipids in the development and progression of atherosclerotic diseases, this question is of great clinical relevance. For example, should all patients with vascular disease, regardless of their cholesterol levels, receive statin therapy? Should there be a threshold or target for cholesterol level for the initiation and titration of statins? Or should we just fire away and give statin therapy to anyone who is at risks for cardiovascular disease? These and other questions regarding statin pleiotropy will need to be addressed so that a rational approach can be developed to target patients who will benefit from these agents.

Evidence from basic research and animal studies

Over the past decade, cell culture and animal studies have provided convincing evidence for cholesterol-independent effects of statins. It is believed that some or all of these pleiotropic effects may contribute to the reduction in cardiovascular risks with statin therapy. These effects include, but are not limited to: improvement in endothelial function through upregulation of NO synthesis and bioavailability or decrease in endothelin-1 expression; reduction in vascular inflammation through inhibition of macrophage activation, proliferation, and secretion of matrix metalloproteinases; inhibition of vascular smooth muscle proliferation and platelet activation; enhancement of the fibrinolytic system through induction of tissue-type plasminogen activator (t-PA) and downregulation of plasminogen activator-1 (PAI-1) expression; and promotion of neovascularization through direct angiogenesis or mobilization of endothelial progenitor cells (EPCs) ^1–5 (Fig. 1). In addition, statins have been shown to exert antioxidant effect through downregulation of AT1R expression and inhibition of NADPH oxidase activity ^6;7. Some of these antioxidant effects of statins may contribute to their beneficial effects on myocardial hypertrophy, fibrosis, and possibly, heart failure ^8;9.

The biological mechanism underlying most of these pleiotropic effects of statins is inhibition of isoprenoid metabolism in non-hepatic cells (Table 1). Inhibition of mevalonate synthesis reduces cholesterol synthesis as well as synthesis of the isoprenoid intermediates of the cholesterol synthesis pathway, such as farnesol and geranylgeraniol ¹⁰. Isoprenoids are necessary for the posttranslational modification and subsequent trafficking of intracellular signalling molecules. In experimental systems, inhibition of isoprenylation of Rho GTPases mediates some of the cholesterol-independent effects of statins ¹¹ (Fig. 2). For example, the RhoA-dependent cytoskeleton is a negative regulator of eNOS mRNA stability, and statins upregulate eNOS expression by inhibition of RhoA geranylgeranylation ¹². Hence, is it is now possible to explain some important downstream effects of HMG CoA reduction at a cellular level and how they modify cellular phenotype through non-lipid-dependent mechanisms. In addition, many of these effects can be experimentally reproduced in much shorter time course than would be expected by effects mediated by lipid lowering alone. However, the question remains, do these effects occur in humans - and are they quantitatively relevant?

Table 1.

Effects of Statins Unrelated to Lipid Lowering

Endothelial Cells	↑Endothelial nitric oxide synthase (eNOS) expression and activity
	↑Tissue-type plasmiogen activator (t-PA) expression
	↓Plasminogen activator inhibitor-1 (PAI-1) expression
	↓Endothelin-1 synthesis and expression
	↓Reactive oxygen species
	↑Peroxisome proliferator activated receptor (PPAR)-α expression
	↓Proinflammatory cytokines (IL-1β, IL-6, cyclooxygenase-2) expression
	↓Major histocompatiblility (MHC) class II antigen expression
Smooth Muscle Cells	↓Migration and proliferation
	↓Reactive oxygen species
	↓Rac1-mediated NADH oxidase act
	↓Angiotensin AT1 receptor expression
	↑Apoptosis
Platelet	↓Platelet reactivity
	↓Thromboxane A2 biosynthesis
Macrophage	↓Macrophage growth
	↓Matrix metalloproteinase (MMP) expression and secretion
	↓Tissue factor expression and activity
	↓Inducible nitric oxide syntase (iNOS) expression
	↓Proinflammatory cytokines (TNF-α, IL-1β, IL-6) expression
	↓monocytochemoattractant protein (MCP)-1 secretion
	↓Interleukin (IL)-8 secretion
	↓Major histocompatiblility (MHC) class II antigen expression
Vascular inflammation	↓High-sensitivity C-reactive protein (hs-CRP) level
	↓Leukocyte-endothelial cell adhesion
	↓Intracellular adhesion molecule-1 expression
	↓P-selectin expression
LDL cholesterol	↑Uptake and degradation
	↓Oxidation
	↓Scavenger receptor expression
	↓Endocytosis
Other effects	↓Tumor growth
	↑Bone formation
	↓Intracellular Ca2+ mobilization

Evidence for pleiotropic effects from clinical trials

There is close association between elevated plasma cholesterol levels, endothelial function and atherosclerosis ¹³. Statins are the most effective agents currently available for lowering plasma levels of LDL-C, and statin will lower LDL-C to some extent in all humans. Therefore, clinical studies that are designed to distinguish between potential "pleiotropic" effects from cholesterol lowering are difficult, if not impossible, to perform. Nevertheless, there is indirect evidence from clinical trials with statins, which suggest that cholesterol reduction may not be the only contributor to the beneficial effects of these agents.

1) Effects of statins on patients with relatively low cholesterol

The LDL-C treatment goal for secondary prevention is currently set at 100 mg/dl ^14;15. Available evidence provides conflicting information regarding effects of statins below this level. Subgroup analyses of the West of Scotland Coronary Prevention Study (WOSCOPS) and Cholesterol and Recurrent Events (CARE) trial do not support the view that further reduction of LDL-C is beneficial ^16;17. This may support the existence of a threshold for reducing LDL-C below which no further reduction in the risk of coronary events is evident. Similarly, analysis of the CARE study showed no additional event reduction below levels of 125 mg/dl. CARE showed that the number of CHD cases fell when LDL-C was reduced from 174 to 125 mg/dl, but CHD did not decline further when the levels ranged from 71–125 mg/dl. It should be noted that only 20% of patients in the treatment group had LDL-C level below 125 mg/dl ¹⁷.

In contrast, the Scandinavian Simvastatin Survival Study (4S) showed the greatest benefit in the patients with the lowest LDL-C levels, and the Post Coronary Artery Bypass Graft (Post-CABG) study demonstrated a 30% reduction in revascularisation procedures and a 24% reduction in clinical end points in patients assigned to the aggressive strategy (less than 100 mg/dl) compared with patients assigned to the moderate strategy (132–136 mg/dl) ^18;19. The recent Heart Protection Study (HPS) showed no difference in the relative risk reduction of 24% between patients with LDL-C >135 mg/dl and LDL-C <116 mg/dl. The authors of HPS conclude that the protective effect of statins is independent of baseline lipid levels ²⁰. According to HPS, the absolute benefit relates directly to the individual vascular risk of the patients. For the majority of patients, reduction in LDL-C levels to very low values continues to lower risk, but with diminishing returns. However, patients with low LDL-C and high vascular risk (e.g. patients with diabetes) may significantly benefit from statin treatment, possibly through a non-lipid effect. In WOSCOPS, using Cox regression analyses, the percentage decrease in LDL cholesterol in the pravastatin-treated group did not correlate with CHD risk reduction and a decrease in the range of 24% was sufficient to produce the full benefits of therapy, with no further incremental benefits demonstrated for LDL cholesterol reductions up to 39% ¹⁶. This point of view is further supported by the lipid lowering arm of the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT-LLA) that randomised 10,305 hypertensive patients with at least three cardiovascular risk factors and total cholesterol levels below 250 mg/dl to a fixed dose of statin, atorvastatin 10 mg, or placebo ²¹. Treatment was stopped after a median follow-up of 3.3 years because of a 36% risk reduction of myocardial infarction and fatal CHD in the statin group. Similar to HPS, the investigators conclude "benefits of lipid lowering are apparent across the whole range of serum cholesterol concentrations" ²¹.

2) Effects of statins in addition to cholesterol lowering

Another important question relates to potential beneficial effects of statin treatment in addition to cholesterol lowering. Interestingly, post hoc subgroup analyses of the WOSCOP and CARE trials suggest that despite comparable serum cholesterol levels, statin-treated individuals have significantly lower risks for coronary heart disease compared to age-matched placebo-controlled individuals. For example, patients with LDL-C between 140–180 mg/dl on statin showed a 36% lower relative risk compared to patients on placebo with the same cholesterol levels ^3;16;22. Meta-analyses of past clinical trials suggest that the risk of adverse vascular events in patients treated with statins is significantly lower compared to individuals treated with other cholesterol-lowering agents or modalities despite comparable reduction in serum cholesterol levels in both groups ²³. For example, application of the Framingham risk score to WOSCOPS produced a coincidence between predicted and observed risk in the placebo group, but underestimated the benefit of the pravastatin group by 31% ^3;16;22.

3) Rapid effects of statins

On the most impressive results of the statin trials is the rapidity by which the beneficial effects occur. Compared to the POSCH study where survival curves do not separate for more than 4–5 years after surgery ²⁴, most of the statins trials demonstrate a survival advantage within 2 years of therapy and in some cases, within 6 months, despite comparable reduction in serum cholesterol levels ²¹. Indeed, in the Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering (MIRACL) trial, statins were found to be effective in reducing recurrent ischemic events as early as 16 weeks following acute coronary ischemia ²⁵. Although the serum LDL-cholesterol was reduced by 40%, this time frame was probably too short for appreciable changes in lesion size and plaque stability. Therefore, it is widely believed that some other actions of statins, particularly the improvement of endothelial function, may have contributed to these early benefits. Taken together, these findings suggest that the clinical benefits of statin therapy may extend beyond cholesterol lowering to include direct cellular effects on the vascular wall.

As mentioned, the effects of statins on vascular function is more rapid than the effects of lipid lowering, e.g. statins improve impaired endothelial function in diabetic patients within 1–3 days without affecting lipid levels ^26–28. Statins have been shown to increase NO production in humans ²⁹. For example, fluvastatin increased the bioavailability of NO in a randomised, double-blind, placebo-controlled trial of 29 hypercholesterolaemic patients ²⁶. Similarly, simvastatin has also been shown to improve endothelial function, augment basal and acetylcholine-stimulated vasodilation, over 4 weeks in a double-blind, placebo-controlled, crossover trial ³⁰. A short-term study with pravastatin has demonstrated an NO-mediated improvement in forearm blood flow and an increase in urinary nitrite/nitrate production, indicative of an increase in NO production, which occurred independently of changes in cholesterol levels in normocholesterolaemic patients with other risk factors for CHD ³¹. In another study, short-term (2 weeks) lipid-lowering therapy with cerivastatin improved endothelial function and NO bioavailability in patients with hypercholesterolaemia ²⁶. This improvement in endothelium-dependent vasodilation was no longer observed when the nitric oxide synthase inhibitor N^G-monomethyl-L-arginine was co-infused.

Similarly, in normocholesterolaemic men with normal vascular function, treatment with high-dose atorvastatin (80 mg) increased forearm blood flow within 24 h, whereas serum cholesterol and high-sensitivity C reactive protein (hs-CRP) were not decreased until 2 days of treatment. Cessation of statin treatment after 30 days, resulted in a rebound of forearm blood flow within 24 h of withdrawal, whereas cholesterol and hs-CRP slowly and steadily returned to baseline ²⁸. This is supported by other studies, involving withdrawal of treatment (e.g., PRISM), where the benefits of statin treatment have been lost sooner than any changes in lipid levels ^32;33. Therefore, the potential importance of cholesterol-independent effects of statins in humans may also relate to the time-course of their effects, for example, after withdrawal of treatment.

4) Plaque stabilizing and anti-inflammatory effects of statins

Lipid lowering by statins may contribute to atherosclerotic plaque stability by reducing plaque size or by modifying the physiochemical properties of the lipid core ⁴. However, changes in plaque size by lipid lowering tend to occur over extended time and are quite minimal as assessed by angiography ². Rather, the clinical benefits from lipid lowering are probably due to decrease in vascular inflammation. For example, lipid lowering alone reduces macrophage accumulation in atherosclerotic lesions and inhibition of MMP production by activated macrophages. Indeed, statins inhibit the expression of MMPs and tissue factor by cholesterol-dependent and –independent mechanisms ^4;34;35, with the cholesterol-independent or direct macrophage effects occurring at a much earlier time point. The plaque stabilizing properties of statins, therefore, are mediated through a combined reduction in lipids, macrophages, and MMPs ⁴. These effects of statins may reduce the incidence of acute coronary syndromes by lessening the propensity for plaque to rupture.

High sensitivity C-reactive protein (CRP) is a non-specific marker of inflammatory disease, which has been shown to be associated with vascular risk independent of cholesterol levels. Treatment with statins lowers CRP levels, and, interestingly, this effect does not correlate with the lowering of LDL-C ^36–38. For example, retrospective analysis of the CARE trial suggests that elevated CRP predicted risk independent of the lipid profile and patients with high CRP benefited the most from pravastatin treatment, supporting an anti-inflammatory and cholesterol-independent effect of statins in humans ³⁹. However, the relation between CRP and atherogenesis needs to be further investigated. In concert with these findings, statins have been shown to reduce graft vessel disease after heart and renal transplantations and to inhibit pro-inflammatory cytokine release ^40–43.

5) Effects of statins on stroke

In contrast to CHD, serum cholesterol levels are poorly correlated with the risk for ischemic stroke. Therefore, the ability of statins to decrease the incidence of ischemic stroke highlights some of their cholesterol-independent effects. HPS and ASCOT-LLA showed unequivocally their benefits in stroke prevention regardless of baseline cholesterol levels ^{20;21;25;44;45}. Indeed, previous non-statin lipid lowering trials do not show any significant reduction in ischemic stroke ⁴⁶. Interestingly, animal studies show a reduction of the size of ischemic cerebral infarcts by a lipid-independent NO-mediated mechanism ^47;48, a beneficial effect that could contribute to an overall decrease in the incidence of ischemic stroke. For example, it is possible that statins may have contributed to the decrease in the incidence of ischemic strokes in clinical trials, in part, by reducing cerebral infarcts size to levels, which are clinically unappreciated.

6) Cholesterol-independent effects of statins on atherosclerosis

The clinical importance of non-cholesterol effects of statins in humans is difficult to determine in a prospective clinical trial because statins effectively reduce cholesterol levels even in individuals with low baseline cholesterol levels. In cynomolgus monkeys, cholesterol-independent effects of statins were demonstrated using a "clamped-cholesterol" design, where dietary cholesterol was regulated to maintain equal changes in lipoprotein levels in control and pravastatin-treated animals. Compared with control monkeys, the arteries of statin-treated monkeys had better dilator function than those of monkeys not receiving statin. Drug treatment reduced inflammation and features of plaque vulnerability. Lesional smooth muscle cell and collagen content was greater in the pravastatin and simvastatin treated groups than in the control group suggesting a more stable plaque phenotype ^35;49. These beneficial vascular effects of statins occurred independently of plasma lipoprotein concentrations.

Role of mevalonate and isoprenoids

The predominant pharmacological site of action of statins in terms of lipid lowering is the liver. The established mechanism of statins is via SREBP activation due to reduced hepatic cholesterol synthesis and secondary upregulation of the LDL-receptor, leading to enhanced clearance of circulating cholesterol and lipids ^10;50. The effect of statins on extra-hepatic cells is much less clear, since circulating levels of the hydrophilic statins are very low ⁵¹. Little has been published on drug levels within peripheral tissues such as the endothelium. Similarly, it is not clear whether effects demonstrated in animals, such as inhibition of Rho isoprenylation, could be observed in cardiovascular cells after oral statin treatment in humans.

Most pleiotropic effects of statins in cell cultures can be reversed by the addition of mevalonate. Effects that are mediated by Rho GTPases are attributed to the isoprenoid, geranylgeranlpyrophosphate (GGPP). Therefore, increasing mevalonate or GGPP concentrations within extra hepatic cells could limit the benefits of statins. Intracellular isoprenoid concentration may be dependent on extracellular sources or intracellular production. It seems most likely that the liver, being involved in the synthesis of cholesterol, may be a major source of mevalonate and isoprenoids. Hence, it may be hypothesised that circulating mevalonate levels are regulated by inhibiting hepatocyte HMG CoA reductase. Unfortunately, there is insufficient data currently available on the role of circulating mevalonate and specific isoprenoids at limiting the extra-hepatic effects of statins.

Another source of mevalonate that would influence the pleiotropic activity of statins is the extra hepatic cells such as vascular wall cells. Intracellular mevalonate production in these cells is likely to be dependent upon a statin’s permeability and its potency for inhibiting the cell-type specific HMG CoA reductase. Consequently, varying lipophilicity of statins affect may their uptake into target organs and, therefore, their ability to elicit pleiotropic effects ³. Lipophilic statins such as lovastatin and cerivastatin are considered more likely to enter endothelial cells by passive diffusion than hydrophilic statins such as pravastatin and rosuvastatin, which are targeted to the liver ^3;52. However, in bovine endothelial cells in culture, where external sources of cholesterol and mevalonate are absent, pravastatin was at least as effective as more lipophilic statins such as atorvastatin, lovastatin and simvastatin at stimulating the release of NO from cultured bovine endothelial cells ⁵³.

It would also seem that pravastatin is the least potent statin at inhibiting HMG CoA reductase from human endothelial cells in culture ⁵⁰. Similarly, despite being less lipophilic and less potent at inhibiting endothelial HMG-CoA reductase than atorvastatin and simvastatin, rosuvastatin is at least as effective and potent as these agents at increasing eNOS activity in mouse aorta ⁵⁴. Thus, the lipophilicity and potency within endothelial cells does not entirely predict the ability of statins to improve NO production/release, and so other unidentified factors may play a role. It is possible that there may exist other specific mechanisms for hydrophilic statins to enter endothelial cells. Such a mechanism is present in the liver, where the hepatic organic anion transporter (OATP-C) helps hydrophilic statins enter hepatocytes ^55;56, and a similar process in endothelial cells would explain why the lipophilicity of a statin does not always predict its pleiotropic efficacy. More data on how different statins enter extra-hepatic cells are needed to address this issue.

Interestingly, an unwanted negative effect of statins, myotoxicity, does suggest that statins exert effects on extra-hepatic cells. Recent studies report that inhibition of geranylgeranylation of low molecular weight proteins in skeletal muscle may be involved ⁵⁷. However, the exact mechanism leading to rhabdomyolysis is still not know or has not been published.

Conclusion for clinical practice: Focus on cardiovascular risk

If statins reduce vascular risk independent of cholesterol levels, the most important clinical task would be to select those patients who would benefit the most from the treatment independent of their lipid profile. The recent HPS and ASCOT-LLA provide compelling evidence for the concept that treatment strategies to reduce cardiovascular disease should depend on global assessment of risks rather than on numerical values of lipid levels ^20;21. Current lipid guidelines adopt the concept that high risk patients, such as diabetics, patients with aortic aneurysms or significant peripheral artery disease should be treated to the same LDL-C target of 100 mg/dl as patients with symptomatic CHD ^14;15. A 10-year Framingham coronary risk of 20% is currently regarded "high risk" requiring aggressive intervention including lipid treatment ⁵⁸. However, the placebo group in ASCOT-LLA had a 10-year coronary risk of 9.4%, suggesting that even patients with a global risk > 10% will still benefit from statin therapy ²¹. Indeed, in the 26-year follow up of the Framingham Heart Study, the cholesterol profile of individuals who develop or do not develop coronary heart disease are almost indistinguishable. Thus, in the future, serum markers of inflammation which may not necessarily be related to cholesterol levels, such as CD40-ligand or CRP, may be useful in targeting patients who would benefit from statin therapy. However, prospective research is required before these markers can be routinely used for clinical practice.