Bempedoic acid reduces LDL-c through liver cell-specific cholesterol synthesis inhibition

SL Pinkosky, RS. Newton, EA. Day, et al
Nat Commun. 2016 Nov 28;7:13457

Background

The prodrug bempedoic acid reduces LDL-c and attenuates atherosclerosis after catalytic activation. This paper systematically outlines the experiments and analyses undertaken to fully understand the mechanism of action for how bempedoic acid reduces LDL-c.

Elevated LDL-c is a significant risk factor for atherosclerosis. Alternative therapies to further reduce LDL-c levels if goals are not met with statins, are being developed, including ETC-1002 (bempedoic acid), an oral, small molecule cholesterol synthesis inhibitor. Statins directly target hepatic 3-hydroxy-3-methyl-glutaryl-CoA (HMG-CoA) reductase and thereby reduce cholesterol biosynthesis. ETC-1002 inhibits ATP-citrate lyase (ACL), which acts upstream of HMG-CoA. ACL cleaves citrate into oxaloacetate and acetyl-CoA, the latter serving as common substrate for de novo cholesterol and fatty acid synthesis. As ETC-1002 is a prodrug, it first needs modulation by endogenous acyl-CoA synthetase (ACS), yielding ETC-1002-CoA, which can inhibit ACL.

In addition to significant lowering of LDL-c levels by ETC-1002 in hypercholesterolaemic patients by 30% as monotherapy, up to an additional 24% when added to stable statin therapy and up to 50% when combined with ezetimibe, ECT-1002 also increases AMP-activated protein kinase (AMPK)^1-5. This molecule is a metabolic sensor capable of catalysing regulatory phosphorylation of numerous substrates that affect inflammatory signalling and lipid metabolism. However, the importance of  AMPK elevation by ETC-1002, as well as ACL modulation, remains undefined. Furthermore, it is currently unknown whether LDL-c-lowering by ETC-1002 is sufficient to reduce the progression of atherosclerosis.

To elucidate this, in vitro and in vivo experiments were performed using the ETC-1002 prodrug that requires the activity of very-long-chain acyl-CoA synthease-1 (ACSVL1) for the conversion to an active modulator of ACL and AMPK activity. It was also explored why ETC-1002 does not co-exist with myotoxic effects, as statins do.

Main results

First, kinetic analyses confirmed that ETC-1002-CoA indeed competes with ACL for CoA binding, but not for binding with citrate and ATP, while the prodrug, ETC-1002 free acid, did not. Thereafter, in vitro models showed that ETC-1002-CoA is a competitive ACL inhibitor with respect to CoA and directly interacts with AMPKβ1 to increase AMPK activity. This is only possible after catalysing the CoA activation of ETC-1002. ETC-1002-CoA was not able to activate AMPKα1β2ϒ1.

To identify the specific ACS enzyme isoform that is required to catalyse the reaction of CoA to the ETC-1002 prodrug, the natural substrate profile of ACS was defined. Fatty acids with C12 and C20 carbon chain lengths were most competitive for ETC-1002. Near complete inhibition of ETC-1002-CoA synthesis was seen with C16 and C18 carbon lengths. All substrate binding data together indicate a profile most consistent with the activities described for ACSVL1. Using siRNA and reducing ACSVL1 protein levels by 80%, it was established that ACSVL1 was indeed the specific ACS isoform that catalyses the CoA activation of ETC-1002.

These data suggest that ETC-1002 only modulates ACL and AMPK activities in cell types that express ACSVL1. Tissue expression of ACSVL1 in mice is restricted to liver and kidney, while absent in other tissue, including skeletal muscle⁶. Importantly, ETC-1002-CoA and ACSLV1 were detected in liver, but not in skeletal muscle of ETC-1002-treated mice. ETC-1002 did not suppress cholesterol synthesis or induce signs of muscle apoptosis or cytotoxicity in primary human myotubes, demonstrating that the CoA activation of ETC-1002 and subsequent suppression of cholesterol synthesis, requires ACSVL1. Since ACSVL1 is not expressed in skeletal muscle, ETC-1002 is unlikely to cause the myotoxicity associated with statins.

It was next examined whether modulation of ACL and AMPK by ETC-1002 mediates the cholesterol-lowering effects of ETC-1002 and whether these effects lead to reduced vascular lesion development. Double-knockout (DKO) mice deficient for APOE and AMPKβ1 fed a high-fat-high cholesterol (HFHC) diet, displayed significant increases in body weight, adiposity, fasting glucose and diminished glucose tolerance, compared with chow-fed mice. This was, however, not observed in ETC-1002-treated mice. Focussing on hepatic lipids specifically, HFHC-feeding increased hepatic cholesterol mass by ~2 fold in both mice, an effect that was almost completely blocked by ETC-1002 treatment. Remarkably, ETC-1002 treatment also prevented HFHC-induced increase in triglycerides in both mice.

Changes in the rates of liver lipogenesis and fatty acid oxidation are important determinants in controlling liver lipid content. Consistent with decreased lipogenesis and increased fat oxidation, ETC-1002 treatment suppressed the respiratory exchange ratio (RER) during the dark (feeding) cycle in both mice, while no treatment effect was observed during the light (fasted) cycle. Moreover, ETC-1002 suppressed total lipid synthesis in both wild-type and AMPKβ1 knockout hepatocytes, indicating that ETC-1002 reduces LDL-c and liver triglycerides and cholesterol, through AMPK-independent pathways.

Statins reduce LDL-c by triggering a feedback mechanism whereby inhibition of cholesterol synthesis results in reduced cellular cholesterol levels, which activates sterol response element-binding protein-2 (SREBP2)-dependent LDL receptor transcription. HFHC-feeding significantly suppressed the expression of numerous transcriptionally SREBP2-regulated genes in APOE-deficient mice, including Srebf2, Ldlr (LDL-receptor), Pcsk9 and Hmgr. ETC-1002 treatment increased Srebf2 and also Ldlr expression in both APOE-deficient and DKO mice, and an increase in plasma membrane-associated LDL receptor was seen in liver sections from ETC-1002-treated mice. This suggests that ETC-1002 appears to lower LDL-c via compensatory SREBP2-dependent LDL receptor upregulation in response to AMPK-independent suppression of cholesterol synthesis.

Morphological assessments of lesions from sections from the aortic sinus of the mice revealed attenuated lesion development in chow-fed APOE-deficient mice, whereas HFHC-fed APOE-deficient and DKO mice developed significantly larger lesions. No difference in lesion size was observed between APOE-deficient mice and DKO mice, suggesting that absence of AMPKβ1 did not accelerate atherosclerosis. Furthermore, despite modest reductions in total plasma cholesterol by ETC-1002 treatment, there was a marked reduction in lesion size. Moreover, plasma serum amyloid A (SAA) levels were reduced by ETC-1002 treatment in both mice, indicating a reduction in diet-induced low-grade inflammation.

To assess the effect on humans, primary human hepatocytes were treated with ETC-1002, which caused reduced cholesterol synthesis. After 36 hours of treatment, ETC-1002 and atorvastatin treatment reduced total intracellular cholesterol mass by 21% and 42% respectively and reduced media apoB concentrations by 31% and 32%, which were not due to affected secretion rates. Also in these cells, ETC-1002 increased the expression of SREBF2-regulated genes. Furthermore, ETC-1002 and atorvastatin treatment also upregulates LDL receptor activity in human liver cells. Moreover, the addition of ETC-1002 to atorvastatin increased LDL receptor activity above atorvastatin treatment alone, supporting that the co-suppression of ACL and HMG-CoA reductase activity is complementary.

Conclusion

The mechanisms for ACL inhibition and AMPK signalling promotion by ETC-1002 have been investigated. This showed the requirement for CoA activation of ETC-1002 to directly modulate the activities of ACL and AMPK. Also, ACSVL1 was identified as specific ACS isoform responsible for catalysing the CoA activation of ETC-1002, which is required for ACL inhibition and mediation of β1-dependent AMPK activation. The activity of ETC-1002-CoA is almost exclusively restricted to liver. As ACSVL1 is also absent in human skeletal muscle, this suggest that the muscle cell function is maintained with ETC-1002 treatment, in contrast to statin treatment. Using mouse models, AMPK signalling involvement in LDL-lowering by ETC-1002 was excluded and ACL as a target was confirmed. Inhibition of ACL by ETC-1002 resulted in reduced cholesterol biosynthesis and upregulation of the LDL receptor. These LDL-c reductions were associated with reduction in atherosclerosis and it is suggested that this anti-atherosclerotic activity is primarily driven by reduced systemic inflammation. All these data suggest that ACL is a suitable target for therapeutic intervention.

Find this article online at Nat Commun

References

1. Pinkosky, S. L. et al. AMP-activated protein kinase and ATP-citrate lyase are two distinct molecular targets for ETC-1002, a novel small molecule regulator of lipid and carbohydrate metabolism. J. Lipid Res. 54, 134–151 (2013).
2. Ballantyne, C. M. et al. Efficacy and safety of a novel dual modulator of adenosine triphosphate-citrate lyase and adenosine monophosphate-activated protein kinase in patients with  hypercholesterolemia: results of a multicenter, randomized, double-blind, placebo-controlled, parallel-group trial. J. Am. Coll. Cardiol. 62, 1154–1162 (2013).
3. Gutierrez, M. J. et al. Efficacy and safety of ETC-1002, a novel investigational low-density lipoprotein-cholesterol-lowering therapy for the treatment of patients with hypercholesterolemia and type 2 diabetes mellitus. Arterioscler. Thromb. Vasc. Biol. 34, 676–683 (2014).
4. Ballantyne, C. M., MacDougall, D. E. & Margulies, J. R. ETC-1002 incrementally lowers low-density lipoprotein cholesterol in patients with hypercholesterolemia receiving stable statin therapy. Circulation 132, A17499 (2015).
5. Thompson, P. D. et al. Use of ETC-1002 to treat hypercholesterolemia in patients with statin intolerance. J. Clin. Lipidol. 9, 295–304 (2015).
6. Steinberg, S. J., Wang, S. J., Kim, D. G., Mihalik, S. J. & Watkins, P. A. Human very-long-chain acyl-CoA synthetase: cloning, topography, and relevance to branched-chain fatty acid metabolism. Biochem. Biophys. Res. Commun. 257, 615–621 (1999).