Abstract

We report the design, synthesis, and optimization of the first, selective activators of cardiac myosin. Starting with a poorly soluble, nitro-aromatic hit compound (1), potent, selective, and soluble myosin activators were designed culminating in the discovery of omecamtiv mecarbil (24). Compound 24 is currently in clinical trials for the treatment of systolic heart failure.
Keywords: Omecamtiv mecarbil, small molecule enzyme activator, cardiac myosin, systolic heart failure
A strategy to treat systolic heart failure (HF) is to increase cardiac contractility. Current drugs, such as dobutamine, milrinone, and digoxin, activate second messenger pathways that raise cardiac myocyte calcium concentrations and subsequently increase cardiac contractility. Unfortunately, these drugs also increase heart rate and myocardial oxygen consumption and can produce clinically significant arrhythmias and hypotension, and their use is associated with higher mortality.1,2
Our approach is to selectively increase cardiac contractility by exploiting a novel mechanism of action that directly targets cardiac myosin, acting downstream of second messenger activation and calcium regulation.3 Direct activation of myosin, the molecular motor that drives muscle contraction, may avoid the mechanism-related toxicities of current inotropes and offer a novel and potentially useful treatment for HF.4 This report describes the discovery and optimization program that led to selective activators of the cardiac sarcomere culminating in the discovery of omecamtiv mecarbil (24), a compound with the potential to treat systolic HF in an acute situation through iv administration and acute and chronic systolic HF through oral administration.
In the cardiac sarcomere, force generation is directly coupled to adenosine triphosphate (ATP) hydrolysis. Compounds that activate the cardiac sarcomere were identified by measuring increases in myosin ATPase activity in a high-throughput, fully soluble, calcium-responsive, reconstituted sarcomere or myofibril assay.6−8 Data were reported as an AC40, the compound concentration resulting in a 40% increase in the cardiac sarcomere ATPase activity at the calcium concentrations that produced 25−50% of maximum calcium-dependent activation.9
A high-throughput screen (HTS) identified compound 1 as a sarcomere activator having an AC40 value in the low micromolar range. There were several interesting biological characteristics of 1 that made this series of compounds worthy of pursuit. First, 1 demonstrated excellent selectivity for the cardiac sarcomere over the skeletal sarcomere and smooth muscle myosin.10 Second, 1 demonstrated increased contractility in freshly isolated, adult rat myocytes as measured by an increase in fractional shortening (FS).11 Importantly, no effects on the myocyte calcium transient were observed. The degree of FS observed (ΔFS % = 135% at 2 μM) was substantial and was believed to be of sufficient potency to elicit a biological response in a whole animal at therapeutically relevant doses. However, the poor physical properties of 1 did not allow this hypothesis to be tested in vivo.
Although compound 1 was viewed as a promising lead based on its interesting biological properties, 1 and various analogs presented several formidable challenges to obtain agents suitable for both acute iv administration and chronic oral therapy. These challenges included an undesired structural moiety (nitro group), hydrolytic instability of the diaryl amide, poor aqueous solubility, low free fraction, modest intrinsic potency, off-target activities such as direct vasodilation through KATP channel activity12 and CYP 1A2 inhibition, and high intrinsic clearance. Our goal was to address these challenges systematically while maintaining the favorable properties for compound 1 such as selectivity for cardiac myosin over other type II myosins, lack of effect on the calcium transient or other off-target cellular effects, lack of additional off-target effects as measured by general screening,13 and good permeability.
To achieve this goal, we designed flexible synthetic routes to allow variation at each part of the molecule as shown in Schemes 1 and 2. Nucleophilic aromatic substitution of nitroarene 15 (Scheme 1) with the appropriate deprotonated hydroxypyridine or amino alcohol provided nitro-substituted ether 16.14 The nitro group was reduced to aniline 17 using standard protocols (e.g., H2, Pd/C, or Pd/C, NH4CO3, or SnCl2). In situ generation of the isocyanate of R2NH2 using 0.4 equiv of triphosgene and 2.1 equiv of diisopropylethylamine (DIPEA),15 followed by reaction with a substituted amino pyridine, afforded desired analogues 4−14. Reductive amination16 of 18 with the appropriately substituted benzaldehyde 19 (Scheme 2) yielded desired benzyl piperazines 21 (after reduction of the nitro moiety for R′ = NO2). Alternatively, 21 was attained by alkylation of piperazine 18 with the benzyl bromide or mesylate of the nitroaromatic compound 20 in the presence of DIPEA as base followed by reduction of the nitro group as described above. Urea formation as described above or by direct addition of the isocyanatopyridine to 21 furnished the desired ureas 22−24 in good yield.
Scheme 1. General Synthesis of Aryl Ether Ureas.
Reagents and conditions: (a) NaH or K2CO3, R1OH, DMF, room temperature−70 °C, 30−85%. (b) H2, Pd/C, MeOH, room temperature; or Pd/C, NH4CO3, MeOH, reflux or SnCl2, EtOH, 50 → 95%. (c) Triphosgene, DIPEA, R2NH2, THF or CH2Cl2, room temperature, 60−85%.
Scheme 2. General Synthesis of Benzylpiperazine Ureas.
Reagents and conditions: X = Br or OMs; Y = F or H; R = methoxycarbonyl or t-butoxycarbonyl (t-Boc). When t-Boc was used as a protecting group, it was converted to the methyl carbamate in the following manner: (1) Trifluoroacetic acid, CH2Cl2, room temperature. (2) Methyl chloroformate, DIPEA, THF or CH2Cl2, 0 °C. (a) NaHB(OAc)3, CH2Cl2, room temperature. (b) H2, Pd/C, MeOH, room temperature. (c) DIPEA, CH2Cl2, room temperature, 85%. (d) H2, Pd/C, K2CO3, MeOH, room temperature, 80%. (e) Triphosgene, DIPEA, 2-methyl-5-aminopyridine, THF or CH2Cl2, room temperature; or 2-methyl-5-isocyanatopyridine, acetone, room temperature, 65−85%.
The first level of optimization served to remove the undesirable nitro group of 1 (Table 1). A first principle driving this structure−activity relationship (SAR) was to maintain an electron-withdrawing group at this position to reduce the potential for oxidative metabolism of the pendant aromatic ring. A fluorine atom replacement (2) maintained the biochemical activity of the parent nitro compound. Hydrogen (3) substitution dramatically attenuated the activity.
Table 1. HTS Hit (1) and Replacement of the Nitro Moiety.

| no. | X | AC40 (μM) | kinetic solubility (μg/mL) |
|---|---|---|---|
| 1 | −NO2 | 1.4 | <1.0 |
| 2 | −F | 2.5 | 5.3 |
| 3 | −H | 32 | 17 |
The next level of optimization addressed the low hydrolytic stability of the diaryl amide moiety and the poor aqueous solubility. Substitution of the amide of 2 with a more hydrolytically stable urea moiety17 (4) maintained biochemical potency and gave measurable improvements in cardiomyocte activity (Table 2). Substitution of the 4-fluoro moiety on the terminal phenyl ring was tolerated as exemplified by compounds 4 and 5; however, the compounds had poor aqueous solubility. The addition of a nitrogen atom into the terminal phenyl ring (6) began to improve solubility without affecting biochemical or cellular potency. Removal of the methoxy adjacent to the pyridine nitrogen of 6 to form the more basic pyridine 7 resulted in additional modest improvements in solubility without affecting biochemical or cellular potency.
Table 2. Improved Hydrolytic Stability and Modest Improvements in Solubilitya.

| no. | Y | A | Z | AC40 (μΜ) | CM (ΔFS % at μΜ) | KSol (μg/mL) | PB (% bound) |
|---|---|---|---|---|---|---|---|
| 2 | −NHCO− | CH | F | 2.5 | 104 at 2.0 | 5.3 | 92.1 |
| 4 | −NHCONH− | CH | F | 1.8 | 121 at 2.0 | 1.1 | NA |
| 5 | −NHCONH− | CH | OMe | 0.7 | 139 at 2.0 | <1.0 | 95.1 |
| 6 | −NHCONH− | N | OMe | 1.4 | 134 at 2.0 | 28 | 92.5 |
| 7 | −NHCONH− | N | H | 1.1 | 137 at 2.0 | 46 | 92.9 |
CM, cardiomyocytes; KSol, kinetic solubility; and PB, plasma protein binding.
Compounds were periodically assessed for increases in FS (ΔFS %)18 in normal rats using a bolus iv injection at typical screening doses of 5 and 10 mg/kg to provide a yes/no answer for in vivo activity. A common feature for all compounds at this stage of optimization was high clearance in rats. When high clearance is combined with a moderate affinity for the target, binding to plasma protein generally plays a role in in vivo potency.19 The relationship of plasma protein binding20−23 versus the FS (in vivo) observed at bolus doses of 5 or 10 mg/kg for 20 compounds of similar biochemical and cellular potencies is shown in Figure 1. Compounds with ΔFS % < 120% in vivo span a range of protein binding from approximately 80 to >95% with several compounds clustering above the 90% bound. However, compounds with ΔFS % > 120% contain only the set of molecules that have ≤90% of the drug bound to plasma proteins.
Figure 1.

Plasma protein binding (% bound HPLC) vs increased FS (ΔFS %) in vivo (rats) from an iv injection of test article where measured ΔFS % > 120% is considered more active and ΔFS % < 120% is less active.
Table 3 outlines representative compounds that demonstrated reduced protein binding, removed off-target activities, improved cardiomyocyte potency, and improved intrinsic clearance. Saturation of the “left-side” aryl ring while maintaining an H-bond acceptor maintained good biochemical and cellular potencies and reduced the protein binding (8). CYP 1A2 inhibition was removed by sterically blocking the “right-side” pyridyl nitrogen with an adjacent methyl group (e.g., 8 versus 10).25,26 The vasodilatory activity mediated through KATP channel opening27 was very sensitive to the “left-side” moiety. For example, substitution of the pendant acetyl moiety of 8 as a carbamate 11 or N,N-dimethyl sulfonyl 12 attenuated direct vasodilation, as measured by the aortic ring R50, and inverting the stereochemistry of 8 to 9 removed any detectable vasodilation. Additional “left-side” modifications resulted in removal of any detectable vasodilation and improved cardiomyocyte potency (8, 12, and 13). Modest improvement in oxidative metabolism, as shown by the human extraction ratio (Re) in microsomes, was observed by shifting the connectivity of the piperidine ring to the core phenyl substituent, thus removing the stereocenter (11 vs 14).
Table 3. Reduction in Protein Binding, Removal of Off-Target Activities, Improvements in Cardiomyocyte Potency, and Improvements in Intrinsic Clearancea.

![]() |
CM, cardiomyocytes; Shake Sol, shake flask solubility; PB, plasma protein binding; Re, ratio of extraction in microsomes; and NA, not available.
With an understanding of how to remove the off-target liabilities of our compounds, additional optimization focused on further reduction in oxidative turnover to improve the in vivo pharmacokinetic (PK) parameters, improved solubility for the iv administration, and improved potency in cardiomyocytes that should result in better dose efficacy. As shown in Table 4, substitution of the O-linked left side piperidine moiety of 14 with the methylene linked piperazine ring of CK-1316189 (22) reduced the extraction ratio in human microsomes, dramatically improved the solubility, and improved cardiomyocyte potency. Unlike the initial SAR, removal of the fluorine from the 5-position of the central ring did not attenuate potency (23 versus 22). Finally, reinstallation of the fluorine in the 2-position of the central ring resulted in improved cardiomyocyte potency with no measurable off-target liabilities (24). Introduction of the basic nitrogen of the piperazine improved the physical properties of the molecule (e.g., cLog P values: 14 = 3.0, and 24 = 1.4), and the weak basicity of this nitrogen (pKa = 6.1) did not introduce additional off-target interactions.
Table 4. Additional Improvements in Oxidative Turnover, Increased Solubility, and Improved Potency in Cardiomyocytesa.

| no. | X | W | AC40 (μM) | CM (ΔFS % at μM) | Shake Sol (mg/mL at pH 5) | PB (% bound) | aortic ring R50 (μM) | CYP 1A2 IC50 (μM) | human Re |
|---|---|---|---|---|---|---|---|---|---|
| CK-1316189 (22) | F | H | 1.7 | 131 at 0.8 | 4.6 | 89.7 | >25 | >10 | <0.2 |
| CK-1317138 (23) | H | H | 4.5 | 138 at 0.8 | 4.0 | 87.2 | 15 | >10 | <0.2 |
| omecamtiv mecarbil (24) | H | F | 0.58 | 123 at 0.2 | 2.2 | 79.6 | >25 | >10 | <0.2 |
CM, cardiomyocytes; Shake Sol, shake flask solubility; PB, plasma protein binding; and Re, ratio of extraction in microsomes.
Improved cardiac contractility as measured by increased FS in normal, Sprague−Dawley rats (Figure 2) confirmed the desired effect of cardiac sarcomere activation on systolic function.28 Compounds 23 and 24 demonstrated concentration-dependent increases in FS in this model. As expected from its improved in vitro potency and greater free fraction, 24 showed greater potency in vivo as compared to 23. In dog models of HF, 24 showed increases in cardiac contractility as measured by increased FS, increased stroke volume, and increased cardiac output.29,30,33
Figure 2.

Concentration-dependent increase in placebo-corrected absolute fraction shortening in rats by CK-1317138 and omecamtiv mecarbil. N = 5−6 per data point. Points within the dashed lines were significantly different from placebo (p < 0. 05).
As predicted from in vitro measures of drug metabolism, 24 demonstrated good PK parameters in both rats (Sprague−Dawley) and dogs (Beagle) with clearances of 22 and 7.2 mL/min/kg, volumes of 3.5 and 3.6 L/kg, and bioavailabilities (F%) of 100 and 80%, respectively.
In conclusion, omecamtiv mecarbil (24) is a potent, selective activator of cardiac myosin that leads to increases in cardiac function in preclinical models of HF and now in patients with HF due to systolic dysfunction.31 Its physical and PK properties make 24 suitable for either intravenous or oral administration. With the support of these results and other promising data, 24 is currently in clinical trials for the treatment of systolic HF.32
Acknowledgments
We thank all members of the much larger Cardiovascular Program Team at Cytokinetics for their contributions.
Abbreviations
ATP, adenosine triphosphate; DIPEA, diisopropylethylamine; PK, pharmacokinetic; SAR, structure−activity relationship.
Supporting Information Available
Experimental procedures and analytical data. This material is available free of charge via the Internet at http://pubs.acs.org.
Supplementary Material
References
- Cohn J. N.; Goldstein S. O.; Greenberg B. H.; Lorell B. H.; Bourge R. C.; Jaski B. E.; Gottlieb S. O.; McGrew F.; DeMets D. L.; White B. G. A dose-dependent increase in mortality with vesnarinone among patents with severe heart failure. Vesnarinone Trial Investigators. N. Engl. J. Med. 1998, 339251810–1816. [DOI] [PubMed] [Google Scholar]
- Packer M.; Carver J. R.; Rodeheffer R. J.; Ivanhoe R. J.; DiBianco R.; Zeldis S. M.; Hendrix G. H.; Bommer W. J.; Elkayam U.; Kukin M. L. Effect of oral milrinone on mortality in severe chronic heart failure. The PROMISE Study Research Group. N. Engl. J. Med. 1991, 325211468–1475. [DOI] [PubMed] [Google Scholar]
- Spirito P.; Seidman C. E.; McKenna W. J.; Maron B. J. The management of hypertrophic cardiomyopathy. N. Engl. J. Med. 1997, 33611775–785. [DOI] [PubMed] [Google Scholar]
- Teerlink J. R. A novel approach to improve cardiac performance: cardiac myosin activators. Heart Failure Rev. 2009, 14, 289–298 and references therein. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Finer J. T.; Malik F.; Sakowicz R.; Shumate C.; Wood K.. Compositions and assays utilizing ADP or phosphate for detecting protein modulators. U.S. Patent 6410254 B1, June 25, 2002; 13 pp.
- Hartman J. J.; Malik F.; Sakowicz R.; Finer J. T.. High throughput sarcomeric assay. U.S. Patent 6495337 B1, December 17, 2002; 12 pp.
- Hartman J. J.; Malik F.; Sakowicz R.; Finer J. T.. High throughput sarcomeric assay. U.S. Patent 6509167 B1, January 21, 2003; 12 pp.
- Select compounds were periodically examined using bovine, rat, and human sources of an active cardiac myosin fragment to assess variability across species. No difference in activity was observed across species.
- Compound specificity with respect to muscle type was evaluated by comparing the effect of the compound on the actin-stimulated ATPase activity of a panel of myosin isoforms [cardiac (bovine), skeletal (rabbit), and smooth muscle (bovine uterus or chicken gizzard)] at a single dose of 40, 50, or 100 μM compound. The data for compound 1 are shown in Figure S1 in the Supporting Information.
- FS is the change in myocyte length divided by the resting length of the cell. ΔFS is the percent change in FS after treatment relative to FS at baseline, which is defined as 100%.
- Direct vasodilatation effects were measured by the ability of compounds to relax a potassium chloride (KCl)- or a phenylephrine-induced contraction of a rat aortic ring. The vasodilatation effect of 8 in rat aortic rings was shown to be independent of endothelium, protein kinase G, and protein kinase A pathways for vascular relaxation. However, a dose-related inhibition of the vasodilatation effect of 8 was observed when the aortic ring preparations were pretreated with the KATP channel inhibitor glybenclamide. Therefore, the vasodilatation effect of 8 was determined to be due to its ability to activate KATP channels.
- https://pharmacology.ricerca.com/Catalog/AssayCatalog/AssayCatalog.aspx?path=128&leaf=128.
- Chen N.; Hale C; Karbon W.; Norman M. H. Synthesis and structure-activity relationships of trisubstituted phenyl urea derivatives as neuropeptide Y5 receptor antagonists. J. Med. Chem. 2001, 44, 2344–2356. [DOI] [PubMed] [Google Scholar]
- Majer P.; Randad R. S. A Safe and Efficient Method for Preparation of N,N′-Unsymmetrically Disubstituted Ureas Utilizing Triphosgene. J. Org. Chem. 1994, 59, 1937–1938. [Google Scholar]
- Abdel-Magid A. F.; Carson K. G.; Harris B. D.; Maryanoff C. A.; Shah R. D. Reductive Amination of Aldehydes and Ketones with Sodium Triacetoxyborohydride. Studies on Direct and Indirect Reductive Amination Procedures. J. Org. Chem. 1996, 61, 3849–3862. [DOI] [PubMed] [Google Scholar]
- Estiu G.; Merz K. M. The Hydrolysis of Amides and the Proficiency of Amidohydrolases. The Burden Borne by kw. J. Phys. Chem. B 2007, 111, 6507–6519 and references therein. [DOI] [PubMed] [Google Scholar]
- FS, measured by echocardiographic imaging, is an assessment of cardiac contractility. FS = (end diastolic diameter − end systolic diameter)/end diastolic diameter × 100 as determined by standard analysis of the M-mode images.
- Benet L. Z.; Hoener B. Changes in plasma protein binding have little clinical relevance. Clin. Pharm. Ther. 2002, 713115–121. [DOI] [PubMed] [Google Scholar]
- Kim H. S.; Mallic R.; Hage D. S. Chromatographic analysis of carbamazepine binding to human serum albumin. II. Comparison of the Schiff base and N-hydroxysuccinimide immobilization methods. J. Chromatogr. 2006, B837, 138–146 and references therein. [DOI] [PubMed] [Google Scholar]
- Hage D. S. High-performance affinity chromatography: A powerful tool for studying serum protein binding. J. Chromatogr. 2002, B768, 3–30. [DOI] [PubMed] [Google Scholar]
- Noctor T. A.; Pham C. D.; Kaliszan R.; Wainer I. W. Stereochemical aspects of benzodiazepine binding to human serum albumin. I. Enantioselective high performance liquid affinity chromatographic examination of chiral and achiral binding interactions between 1,4-benzodiazepines and human serum albumin. Mol. Pharmacol. 1992, 42, 506–511. [PubMed] [Google Scholar]
- Kim H. S.; Hage D. S. Chromatographic analysis of carbamazepine binding to human serum albumin. J. Chromatogr. 2005, B816, 57–66. [DOI] [PubMed] [Google Scholar]
- Obach R. S. Prediction of human clearance of twenty-nine drugs from hepatic microsomal intrinsic clearance data: An examination of in vitro half-life approach and nonspecific binding to microsomes. Drug Metab. Dispos. 1999, 27, 1350–1359. [PubMed] [Google Scholar]
- Bromidge S. M.; Duckworth M.; Forbes I. T.; Ham P.; King F. D.; Thewlas K. M.; Blaney F. E.; Naylor C. B.; Blackburn T. P.; Kennett G. A.; Wood M. D.; Clarke S. E. 6-Chloro-5-methyl-1-[[2-[(2-methyl-3-pyridyl)oxy]-5-pyridyl]carbamoyl]-indoline (SB-242084): The first selective and brain penetrant 5-HT2C receptor antagonist. J. Med. Chem. 1997, 40, 3494–3496. [DOI] [PubMed] [Google Scholar]
- Bromidge S. M.; Dabbs S.; Davies D. T.; Davies S.; Duckworth M.; Forbes I. T.; Gaster L. M.; Ham P.; Jones G. E.; King F. D.; Mulholland K. R.; Saunders D. V.; Wyman P. A.; Blaney F. E.; Clarke S. E.; Blackburn T. P.; Holland V.; Kennett G. A.; Lightowler S.; Middlemiss D. N.; Trail B.; Riley G. J.; Wood M. D. Biarylcarbamoylindolines Are Novel and Selective 5-HT2C Receptor Inverse Agonists: Identification of 5-Methyl-1-[[2-[(2-methyl-3-pyridyl)oxy]- 5-pyridyl]carbamoyl]-6-trifluoromethylindoline (SB-243213) as a Potential Antidepressant/Anxiolytic Agent. J. Med. Chem. 2000, 43, 1123–1134. [DOI] [PubMed] [Google Scholar]
- Measured by the concentration at which a compound exhibited a 50% relaxation to baseline levels of a precontracted rat aortic ring (R50 in Table 3).
- Anderson R. L.; Sueoka S. H.; Rodriguez H. M.; Lee K. H.; Cox D. R.; Kawas R.; Morgan B. P.; Sakowicz R.; Morgans D. J. Jr.; Malik F.; Elias K. A.. In Vitro and In Vivo Efficacy of the Cardiac Myosin Activator CK-1827452. American Society of Cell Biology (ASCB), December, 2005. [Google Scholar]
- Shen Y.; Zhang Y.; Morgans D. J.; Vatner S. F.; Malik F. . A Novel Inotropic Agent That Activates Cardiac Myosin and Increases Cardiac Contractility Without Increasing MVO2 in Heart Failure with Left Ventricular Hypertrophy. American College of Cardiology Conference, Chicago, IL, March, 2008. [Google Scholar]
- Malik F.; Elias K. A.; Finer J. T.; Morgan B. P.; Sakowicz R.; Anderson R.; Baliga R.; Conn L.; Cox D.; Garard M.; Hartman J;. Irvine J.; Kawas R.; Kraynack E.; Kuklov A.; Lee K. H.; Lu P.; Muci A.; Pierce D. W.; Rodriguez H.; Suehiro I.; Sueoka S. H.; Sylvester S.; Tochimoto T.; Valdez C.; Wang W.; Katori T.; Kass D. A.; Shen Y. T.; Vatner S. F.; Morgans D. J.. Direct Activation of Cardiac Myosin by CK-1827452 Improves Cardiac Function in a Dog Heart Failure Model. Heart Failure Society of America Meeting, September, 2005.
- Shen Y. T.; Malik F. I; Zhao X.; Depre C.; Dhar S. K.; Abarzúa P.; Morgans D. J.; Vatner S. F. Improvement of Cardiac Function by a Cardiac Myosin Activator in Conscious Dogs with Systolic Heart Failure. Circulation: Heart Failure, 2010, 34522–527. [DOI] [PubMed] [Google Scholar]
- Senior R.; Malik F. I.; Saikali K. G.; Lee J. H.; Chen M. M.; Brand G.; Wolff A. A., and the CY 1121 Investigators. The Selective Cardiac Myosin Activator, CK-1827452, Increases Systolic Function in a Concentration Dependent Manner in Patients with Sable Heart Failure. Annual American College of Cardiology (ACC), Orlando, FL, March 2009.
- http://clinicaltrials.gov.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.



