ROCK inhibitor: Focus on recent updates

Yaodong You Kun Zhu Jie Wang Qi Liang Wen Li Lin Wang Baojun Guo Jing Zhou Xuanlin Feng Jianyou Shi

Citation:  Yaodong You, Kun Zhu, Jie Wang, Qi Liang, Wen Li, Lin Wang, Baojun Guo, Jing Zhou, Xuanlin Feng, Jianyou Shi. ROCK inhibitor: Focus on recent updates[J]. Chinese Chemical Letters, 2023, 34(12): 108336. doi: 10.1016/j.cclet.2023.108336 shu

ROCK inhibitor: Focus on recent updates

English

  • Rho is a Ras-related small guanosine triphosphate-binding protein (GTP-binding protein), which can be activated by guanine nucleotide exchange factors (GEF) as an active GTP-bound state. Rho-associated coiled-coil-containing protein kinase (ROCK) acts as downstream effectors of the small GTP-binding protein Rho [1,2].

    GTP-bound form of Rho activates ROCK which further phosphorylate several downstream targets include myosin light chain (MLC), Rho-kinase inactivates myosin light chain (MLC) phosphatase (MLCP) through phosphorylating the myosin phosphatase targeting subunit 1 (MYPT1) at Ser696/Thr697/Ser854/Thr855, which leads to the enhancement of MLC phosphorylation. And some ROCK targets regulate in particular tissues, like formin homology 2 domain containing 3 (PHOD-3), which plays a vital role in sarcomere organization, myofibrillogenesis and maintenance of the contractile apparatus in cardiomyocytes [3]; the tubulin-associated unit has been associated with more than 25 neurological disorders, hyperphosphorylated and aggregated tau have been shown to increase the likelihood of cognitive impairments [4,5]. All these mechanisms have been summarized in Fig. S1 (Supporting information). ROCK mediates a wide range of biological processes including cell migration, adhesion, cytokinesis, cell proliferation, differentiation and apoptosis. Besides, ROCK also plays an essential role in neuropathies, cardiovascular disease, cancer development and inflammation. For example, ROCK directly phosphorylate phosphatase and tensin homolog (PTEN), therefore inhibit the growth and survival of tumor cells [6]. In addition, ROCK enhances endothelial function and neuroprotection through mediating elongation factor 1 alpha (EF-1α) [7]. The role of ROCK in controlling multiple biological events have made it an essential target for a variety of diseases. Hence ROCK inhibitors have received much attention in recent years. This review discusses some important structure–activity relationships (SAR), the binding patterns and design methods of ROCK inhibitors.

    The structural and expressive characteristics of ROCK are widely acknowledged (Text S1 in Supporting information). Given the high similarity of the structural domains between ROCK-Ⅰ and ROCK-Ⅱ kinases, both isoforms phosphorylate common substrates such as MLC, MYPT1 and LIMK to further regulate the actin-myosin cytoskeletal organization. In addition, both ROCK isoforms participate in the physiological processes of a variety of different systems including nervous, immune, and cardiovascular systems [8]. Besides modulate the cellular biomechanics through phosphorylation, ROCK also mediates gene regulatory signals. Combined with phosphorylated-STAT3 and regulating TH17/TFH gene transcription, ROCK-Ⅱ regulates autoimmunity and inflammation [9]. ROCK also mediates the release of transcription factors, including myocardin-related transcription factor (MRTF), which indirectly modulates myocardial physiology [10]. Abnormal ROCK expression has been reported in multiple pathological conditions, including glaucoma [11], pancreatic cancer [12,13], gastric cancer [14], renal cell carcinoma [15], osteosarcoma [16], myocardial remodeling [17], aortic sclerosis [10,18], pulmonary arterial hypertension (PAH) [19], cognitive deficits [20], Parkinson's disease (PD), Alzheimer's disease (AD) [21,22], inflammation [18] and allergies [23]. Considering the role of the ROCK signaling pathway in oculopathy, neuropathies, cardiovascular disease, cancer, inflammation, human pluripotent stem cells, etc. ROCK targeting agents may potentially be exploited for treating the aforementioned diseases. Several drugs that specifically target ROCK are under development right now.

    The primary approach to inhibiting the ROCK activity adopted by many investigators in the field is competing with type Ⅰ adenosine triphosphate (ATP) in ROCK active conformation. Therefore, the ATP-binding pocket of the ROCK catalytic domain has been used as a basis for the development of small molecular inhibitors against ROCK [24].

    The critical step in the development of ROCK inhibitors is to improve the binding affinity of the small molecule to the ATP-binding pocket of the ROCK catalytic domain. Considering the high similarity between ROCK-Ⅰ and ROCK-Ⅱ [25] and the similarity between ROCK and the closely homologous protein kinase A, G and C (AGC) kinase family, the inability of most ROCK inhibitors to specifically inhibit their targets may cause severe off-target effects [26-29]. Therefore, improving selectivity is another vital point for developing ROCK inhibitors.

    Generally, the hit compounds are obtained via high-throughput screening (HTS) step, and then the functional groups are screened by SAR analysis for further optimization of the original compounds. Furthermore, in recent years, computer-aided drug design (CADD) has also been applied to predict the new inhibitor candidates [30]. The parameters CADD approach adopted include compound selectivity, toxicity and biological activity [31-34]. In particular, Sadybekov et al. [35] present the virtual synthon hierarchical enumeration screening (V-SYNTHES) approach, which rapidly detects the gigascale chemical spaces scoring the best-scoring lead. The team also apply V-SYNTHES on ROCK-Ⅰ to predict small molecule compounds targeting ROCK-Ⅰ, which yields a 28.5% hit rate, including ligands with nanomolar affinity and potency. The new method provides a practical alternative for rapidly screening the growing gigascale modular virtual libraries, helping to identify leads suitable for rapid optimization in the same REAL space.

    Feng et al. [27] have reported more than 170 ROCK inhibitors and discussed their therapeutic potential before 2015. This review mainly summarized the new ROCK inhibitors identified in research papers in the last seven years (2016–2022). Here we categorize the ROCK inhibitors based on their hinge binding moieties. This review mainly focuses on SAR, selectivity, inhibitory efficacy, pharmacological mechanism and clinical research of small molecule compounds. We will also discuss classes of inhibitors that have not been reviewed before, such as natural plant extracts and peptides.

    Fasudil (1) (Fig. 1) was the first ROCK inhibitor approved for clinical use in Japan and China to treat vasospasms following subarachnoid hemorrhages. Since then, fasudil has been subsequently evaluated in acute ischemic stroke, angina pectoris, pulmonary hypertension, Raynaud's syndrome, arterial hypertension, neurodegeneration and glaucoma [36]. In 2014, ripasudil (2) (Fig. 1) has been approved for clinical use in Japan for the treatment of increased intraocular pressure in patients with glaucoma. In recent years, two new ROCK inhibitors were approved in the United States, netarsudil (3) (Fig. 1) in 2017 for treating hypertension and open-angle glaucoma, and belumosudil (4) (Fig. 1) in 2021 for the treatment of graft-versus-host disease, idiopathic pulmonary fibrosis, liver disease and scleroderma. Ripasudil was developed by Kowa Co., Ltd., and fasudil was developed by Asahi Kasei, which were the first classical ROCK inhibitors, and both compounds are classical isoquinoline-based inhibitors. Another classical ROCK inhibitor is represented by Y27632, based on the 4-aminopyridine (and 4-aminopyrrolidine) scaffold, which is mentioned in the paper by Feng et al. [27].

    Figure 1

    Figure 1.  ROCK inhibitors based on isoquinoline and pyridine/pyrrolopyridine derivatives.

    Several derivative compounds based on the above classical inhibitor scaffolds have emerged recently. Some researchers have attempted to modify the substituent at the 5-position of isoquinoline to improve its inhibitory efficacy against ROCK. Chong et al. screened 1,3-benzodioxol-5-yl-[1-(isoquinolin-5-ylmethyl)piperidin-3-yl]methanone (BIPM) (5) (Fig. 1) as a potential ROCK-Ⅱ inhibitor by docking and high-content drug screening [37]. The model analysis, kinase activity assay, and cell-based assays studied coherently confirmed that BIPM has ROCK-Ⅱ inhibitory effects and is selective for focal adhesion kinase (FAK). The BIPM could promote neurite outgrowth; In SH-SY5Y cells (a human neuroblastoma cell line), BIPM could suppress cell migration, wound healing and stress fibers formation. Given that BIPM has no cytotoxicity to SH-SY5Y cells at concentrations ranging from 25 µmol/L to 100 µmol/L, it has been recognized as a promising inhibitor for anti-cancer treatments. Some investigators have made progress in molecular hybridization between fasudil and different scaffolds. The 5-substituted isoquinoline ring is essential for molecular recognition of the active site and is often preserved. BA-1049 (also known as NRL1049) is a racemic mixture designed by BioAxone BioSciences Inc (BBI) based upon fasudil structure [38]. The R enantiomer BA-1049 (R) (6) (Fig. 1) showed better selectivity for ROCK-Ⅱ. BA-1049 (R) is 20–80 times more potent against ROCK-Ⅱ [half-maximal inhibitory concentration (IC50) = 0.24 µmol/L] than ROCK-Ⅰ (IC50 = 3.9 µmol/L), and its metabolite 1-hydroxy-BA-1049 (R) showed better potency for ROCK-Ⅱ. Both cellular and animal studies have demonstrated that BA-1049 (R) alleviates the symptoms of neurological diseases, disorders, and injuries in the tested models through oral or intravenous administration. In a follow-up study, the Ccm1+/− Msh2−/− and Ccm3+/− Trp53−/− mouse models with cavernous angioma (CA) lesions were selected for oral administration of the ROCK-Ⅱ selective inhibitor BA-1049. The results showed BA-1049 slowed CA disease progression by decreasing non-hemoglobin deposition near lesions and reducing lesion growth and maturation. Besides, BA-1049 also exhibited little adverse reactions in the tested mice models that no dose-dependent wear and tear behaviors were observed even after four months of oral administration of 100 mg kg−1 day−1 [39]. BA-1049 is a promising selective inhibitor of ROCK-Ⅱ for CA treatment.

    Liang et al. [40] synthesized a series of fasudil-substituted thiazole compounds and screened out the ROCK-Ⅱ inhibitor compound 7 (Fig. 1). SAR indicate that the pyridine-containing molecules showed higher ROCK inhibitory activity than other heterocyclic substituents, with the fluorine-substituted pyridine showing higher activity than the chlorine-substituted derivatives. Compound 7 showed significant cytotoxic activity against human breast cancer (MCF-7) cells (IC50 = 3.21 µmol/L) and was found more potent than fasudil (IC50 = 10.35 µmol/L). Cellular assays showed the inhibitory properties of compound 7 on the migration and invasion of MCF-7 cells. The in vivo pharmacological experiments results showed that compound 7 improved the body weight loss, reduced the oxidative stress in liver and mammary tissues, and improved intestinal function in the rat model with breast cancer.

    Oliveira et al. designed a new family of N-sulfonyl hydrazide congeners by molecular hybridizing fasudil and the IKK-β inhibitor LASSBio-1524 (Fig. 2) [41]. The SAR indicate that the introduction of any substituent on the terminal benzene ring of N-sulphonylhydrazone derivatives leads to a reduction in the activity of the compound. Only the unsubstituted derivative LASSBio-2020 (8) (Fig. 1) showed better activity in inhibiting ROCK compare to the original two molecules with the IC50 values of 13 and 7.1 µmol/L for ROCK-Ⅰ and ROCK-Ⅱ, respectively [42]. Substituting an N-acylhydrazone group for the sulphonyl hydrazone group in LASSBio-2020 would result in an ineffective ROCK inhibitor. However, the N-alkylation of the N-sulphonylhydrazone moiety contributes to the bioactivity. The N-methylated analog LASSBio-2065 (9) (Fig. 1) was a slightly more potent ROCK inhibitor, which exhibited IC50 values of 3.1 and 3.8 µmol/L for ROCK-Ⅰ and ROCK-Ⅱ, respectively. The small difference in potency between these compounds may be because the N-methyl group in LASSBio-2065 rotates the N-S bond, which allows the compound to have extra hydrophobic interaction with Leu205. This interaction was not seen in LASSBio-2020. Neither of these compounds inhibited IKK-β at a screening concentration of 10 µmol/L, and both derivatives LASSBio-2020 and LASSBio-2065 effectively inhibited cell migration in cell-based assays. In addition, no cytotoxicity was observed for these compounds in serum lactate dehydrogenase (LDH) and 3-(4,5-dimethyl-2-thiazyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assays.

    Figure 2

    Figure 2.  Molecular hybridization to design ROCK inhibitors.

    Another class of classical ROCK inhibitors is compounds designed based upon aminopyridine scaffolds by Yoshitomi Pharmaceutical [43,44]. As an antagonist of voltage-gated potassium (Kv) channels, 4-aminopyridine (4-AP) is used as symptomatic therapy in several neurologic disorders [45]. Based on the mother nucleus 4-AP, Li et al. synthesized five novel 4-AP derivatives as neuroprotective agents. Three compounds exhibited ROCK inhibitory activity [46].

    Among these, compound 10 (Fig. 1) exhibited the most favorable properties compared to the other two. The docking result shows that the hydrogen bond formed between the pyridine ring and hydroxyl in compound 10 and Met156 in ROCK-Ⅱ, and Met172 in ROCK-Ⅱ is the key to the binding possibility. The docking result also indicates that the affinity of this 4-AP derivative for ROCK is similar to that of fasudil. The hydroxyl group on compound 10 effectively enhances the expression of heme oxygenase-1 (HO-1) (a stress-inducible and redox-sensitive protein) and also enhances the ROCK potency of compound 10. Cellular assays have shown that compound 10 effectively inhibits the expression of α-synuclein, exhibiting potential neuroprotective effects, and it also shows some antioxidant and anti-inflammatory properties [46].

    The indazole derivatives developed as ROCK inhibitors in the late 1990s and early 2000s. The indazole portion of these compounds serves as a hinge binder. Yao et al. reported the design, synthesis and SAR studies of a series of novel N-substituted prolinamido indazoles as ROCK-Ⅰ inhibitors [47]. The investigators reported a ROCK-Ⅰ inhibitor, 5-nitro-1H-indazole-3-carbonitrile (DL0805) (Fig. S3 in Supporting information), in the previous literature [48]. The series Ⅲ (Fig. S3) compounds were synthesized by selecting the flexible N-substituted proline amino group containing the benzyl substituent to replace the NO2 group at the 5-position of the indazole of DL0805, and this series exhibited the best inhibitory activity. The 4-position substituent of the terminal benzene ring affects the activity of the compounds in the following order: CH3 > H > Br > OCH3 > F > NO2, CN (Fig. S3). Compounds 11 and 12 (Fig. 3) exhibited potent ROCK-Ⅰ inhibitory activity with IC50 values of 0.27 and 0.17 µmol/L respectively. The in vivo experiments results demonstrated that 11 and 12 have vasodilatory activity in rat aortic rings comparable to that of the clinically approved drug fasudil. The potent active compound 12 (also named DL0805-2) with superior activity was further evaluated to prevent angiotensin Ⅱ (Ang Ⅱ)-induced vasoconstriction by inhibiting Ang Ⅱ-induced intracellular Ca2+ elevation and ROCK pathway activation under in vitro conditions [49].

    Figure 3

    Figure 3.  Indazole, pyridine, and pyrimidine based ROCK inhibitors.

    This class of compounds form hydrogen bonds with ROCK mainly through the nitrogen in the pyridine and the amide carbonyl, respectively. Based on the HTS results, Hobson et al. [50] identified a series of potent and kinome-selective dual ROCK-Ⅰ and ROCK-Ⅱ inhibitors. The experiment results demonstrate the compounds 13 and 14 (Fig. 3) are excellent dual ROCK-Ⅰ and ROCK-Ⅱ inhibitors. The interrogation of the co-crystal structures indicated that, compounds with the (R) configuration at the benzylic carbon were more effective than those with the (S) configuration as ROCK inhibitor. Compound 13 was shown to be a potent dual inhibitor of ROCK-Ⅰ and ROCK-Ⅱ in enzymatic and cellular assays, with an IC50 value of 1 nmol/L for ROCK-Ⅰ/Ⅱ. However, extensive kinome analysis showed compound 13 inhibited two other kinases at the low nanomolar level, PKG1A (IC50 = 1 nmol/L) and PKA (IC50 = 47 nmol/L). Furthermore, compound 13 protects the axons in the retinal nerve fiber layer after oral dosing. Compound 13 may be considered a mother molecule for further studies of developing bi-functional inhibitors. As shown in Fig. S4 (Supporting information), the substitution of the hinge part from pyridine to pyrimidine and the introduction of piperidine-3-ylmethanamine from the meta position of the central aryl ring (R1 group) improved the selectivity profile for ROCK-Ⅰ and ROCK-Ⅱ inhibitors, which led to compound 14. The substitution of R2 with a propyloxy group was not conducive to identifying ROCK-Ⅰ and ROCK-Ⅱ inhibitors, but that of the methoxy group resulted in increased activity. The pyrimidine substitution gave compounds afforded a clean cytochrome P450 (CYP) inhibition profile. In addition, the group explored the effects of different hinge binding motifs on polar surface area (PSA) and fraction sp3 carbon (Fsp3). Piperidine increased the PSA and Fsp3 values associated with a safer predicted toxicity profile rather than 2-aminopyridine, 2-fluoropyridine, and 3-fluoropyridine. Compound 14 was shown to be an excellent and selective dual ROCK-Ⅰ and ROCK-Ⅱ inhibitor with IC50 value of 0.094 µmol/L in the pMYPT1 enzyme-linked immunosorbent assay (ELISA) cell assay. This compound has excellent predicted drug-like properties, including a significantly reduced CYP inhibition profile [50].

    CADD also contributes to the development of new ROCK inhibitors [32,34,35]. Ghosh et al. [31] analyzed the ROCK inhibitors proposed by Hobson et al. [50] using a prediction models established by approaches include docking, molecular dynamics, and three-dimensional quantitative structure–activity relationships (3D-QSAR). Based upon the aforementioned analysis, 40 pyridinyl-benzamide-based ROCK-Ⅰ inhibitors were designed in silico. The biological activities of the designed molecules were also analyzed using the same prediction models and seven lead compounds of ROCK-Ⅰ inhibitors (D02, D03, D06, D31, D32, D33, D35) were selected from them. In this study, the IC50 values, which were in µmol/L units, were converted to log IC50 (pIC50) values. The predicted pIC50 values of all seven compounds were around 8; for example, compound D06 (15) (Fig. 3) with pIC50 value of 8.58. Although absorption distribution metabolism excretion toxicity (ADME/Tox) and synthetic accessibility (SA) score analysis indicate that these compounds have desirable pharmacokinetics (PK) property and bioavailability, investigators still need to synthesize the compounds and verify the relevant parameters with both in vitro and in vivo experiments.

    Through kinase screening, Bristol-Myers Squibb (BMS) reported a series of 5H-chromeno[3,4-c]pyridine, 6H-isochromeno[3,4-c]pyridine, and 6H-isochromeno[4,3-d]pyrimidine based dual ROCK-Ⅰ and ROCK-Ⅱ inhibitors and compound 16 (Fig. 3) was screened among them [51]. The SAR analysis indicate that the replacement of the ether with a benzamide linker lead to moderate improvement in ROCK inhibition potency. The introduction of a meta-methoxy substitution on the phenyl or a benzylic methyl substitution improved the molecule's ROCK-Ⅱ inhibitory potency (IC50 value) further to single-digit nanomolar level. Compared to C5-methyl substitutions, C6-methyl/ethyl substitutions on the central ring could potentially improve its ROCK-Ⅱ inhibitory potency (Fig. S5 in Supporting information). The diastereomer compound 16 exhibits sub-nanomolar ROCK-Ⅰ and ROCK-Ⅱ IC50 potencies (0.67 and 0.18 nmol/L, respectively) with outstanding kinome selectivity.

    In some structures, the amido nitrogen is ring-closure to the central phenyl ring. Sun et al. performed structural optimization and SAR analysis on three regions of the scaffold 3,4-dihydroisoquinolin-1(2H)-one and reported a series of 3,4-dihydrobenzo[f][1,4]oxazepin-5(2H)-one derivative [52]. The most active compound, 17 (Fig. 3), was obtained by replacing 3,4-dihydroisoquinolin-1(2H)-one moiety, 1H-indazole molecule with 3,4-dihydrobenzo[f][1,4]oxazepin-5(2H)-one, and 4-pyridine groups, respectively, which had the highest activity against ROCK-Ⅰ and ROCK-Ⅱ (IC50 = 0.093 and 0.003 µmol/L, respectively). The crystal structural analysis of the compound 17 and ROCK-Ⅰ/Ⅱ complexes confirmed that there are two hydrogen bonds formed between 17 and ROCK-Ⅰ: the nitrogen of pyridine and Met156 and the amide carbonyl with Lys105. Three hydrogen bonds formed between 17 and ROCK-Ⅱ: the nitrogen of pyridine and Met172, the amide carbonyl and Lys121, and the 3-methoxy oxygen atom with Lys216. This explained why 17 exhibits a higher inhibitory activity against ROCK-Ⅱ than ROCK-Ⅰ. At a fixed concentration of 10 µmol/L, 17 exhibited higher inhibition only against LIMK2. What is more, the in vivo experiment results showed 17 rapidly and significantly lowers the intraocular pressure (IOP), and no evident hyperemia was observed during the experiment in the normotensive New Zealand rabbits. But as a lead compound, further refined and in-depth studies are necessary to obtain more effective ROCK inhibitors.

    For pyridine derivatives, 4-pyridine derivatives are usually more effective as ROCK inhibitors than 3-pyridine derivatives. Usually, the introduction of side chain solubilizing groups not only enhances potency and selectivity but also increases the solubility of the compounds [53-55].

    Recently, a series of potent and selective ROCK-Ⅱ inhibitors based on 4-aryl-5-aminoalkyl-thiazole-2-amines was reported by Ma and Wang et al. [56,57]. SAR of this series of compounds have been summarized as follows: Firstly, the substitution of the acid moiety (R1 group) with a pyridine group was not conducive to identifying target inhibitors, but that of 2,3-dihydrobenzo[b][1,4] dioxin group substitution resulted in higher activity. Secondly, compounds with a morpholine group substituent at the aminoalkyl side chain (R2 group) enhanced anti-ROCK-Ⅱ enzyme activity more than compounds with other groups, such as dimethylamine, N-methyl piperazine, N-methyl-N-iso-propylamine, piperidine, and pyrrolidine. Thirdly, it seems that the basic tertiary amine group in the side chain was beneficial to the ROCK-Ⅱ inhibitory activity (Fig. S6 in Supporting information). The most effective inhibitor against ROCK-Ⅱ is 18 (Fig. 4) (IC50 = 0.02 µmol/L). Molecular docking of compound 18 with ROCK-Ⅱ indicated two key hydrogen-bond interactions between the nitrogen in the pyridine ring with Met172 and Lys121. More experiments are needed to further verify and disclose more SAR features of the pyridine and pyrrolopyridine based ROCK inhibitors.

    Figure 4

    Figure 4.  Pyridine and pyrrolopyridine based ROCK inhibitors.

    Based on the compound 18 [56], to improve its inhibitory potency and selectivity, Wang et al. designed 24 new chemical entities as the lead compounds of ROCK inhibitors. They found some active urea-like compounds using the splicing and bioisosteric isometric method [58]. Among them, compound 19 (Fig. 4) is the most potent inhibitor of ROCK-Ⅰ and ROCK-Ⅱ, with IC50 values of 1.1 and 0.03 µmol/L, respectively. The SAR analysis indicated that the pyrrolidine groups in compound 19 enhanced the ROCK inhibition. The previous studies have shown that the hinge binder of pyridinethiazole urea-based ROCK inhibitors is the pyridine ring nitrogen. The nitrogen atom of pyridine formed one H-bonding interaction with the main chain amide NH of Met156 [59]. However, considering the small molecular PSA of compound 19 and the short dialkyl aminomethyl side chain, it only forms an H-bond with Lys216 through the pyridine ring nitrogen. In addition, the authors disclosed the azaindole group scaffold 20 (Fig. 4), which demonstrated a selectivity ratio of 118 for ROCK-Ⅱ inhibition, making it the most selective inhibitor among the compounds studied [58].

    Based on the previous well-known ROCK inhibitor benzodioxane 3′-carboxamide SR3677 [60,61], the Scripps Research Institute has developed a potent and enantiomer-selective ROCK-Ⅱ inhibitor [62]. SAR analysis indicated that substituting the pyridine group with a hinge-binding molecule and substituting the C6 position (R group) of the tryptophan ring with with a 6-methoxy group reduces inhibition of ROCK-Ⅰ while maintaining activity against ROCK-Ⅱ. This compound exists in both R and S conformations, and (S)-7c (S-configuration of this compound) (21) (Fig. 4) showed better ROCK inhibitory activity (IC50 = 170 nmol/L against ROCK-Ⅰ and 9 nmol/L against ROCK-Ⅱ). Molecular docking analysis showed that both the R and S conformations form the same hydrogen bond with ROCK-Ⅱ: the amide carbonyl with Lys121 (critical for isomer selectivity), protonated (CH3)2N with Asp176, pyridine nitrogen with Met172. The seven hydrophobic rings in the 21-ROCK-Ⅱ complex (Met169, Leu221, Val106-C···pyridine ring; Lys121-C-pyran ring; Leu123-C, Phe136···phenyl ring of chroman) explain why 21 showed better ROCK-Ⅱ inhibition and isomer selectivity (Fig. 5). However, the authors did not report the PK parameters of compound 21 in vitro or in vivo.

    Figure 5

    Figure 5.  Crystal structure of 21 complexed with the kinase domain of ROCK-Ⅱ (PDB ID: 4L6Q). The crystal structure of 21 (green) bound to the active site of ROCK-Ⅱ. Hydrogen bonds are shown as dashed lines.

    Vertex Pharmaceuticals Incorporated (VPI) has synthesized a series of N-acylated aminothiazoles and N-acylated aminothiophene and evaluated their biological activity against Rho kinase [63]. The X-ray crystal structure indicated that the substitution of the central thiazole or thiophene ring is the most suitable of all five- or six-membered rings to increase the inhibitory potency of ROCK. Extending the substituent into the pocket defined by the glycine-rich loop enhanced ROCK-Ⅰ inhibition potency. Compound bearing a 3-methane-sulfonyl amino group provided the best overall combination of potency and selectivity. In addition, the 2-fluoro-substituted pyridine effectively reduces CYP inhibition while maintaining good ROCK-Ⅰ potency and PKA selectivity (Fig. 6). Compound 22 (Fig. 4) was able to inhibit the enzymatic activities of ROCK-Ⅰ (IC50 = 10 nmol/L) with 300-fold selectivity against PKA, desirable selectivity against a panel of kinases, good cellular potency, and excellent PK profiles [63].

    Figure 6

    Figure 6.  Structure–activity diagram of 2230.

    To enhance the solubility and target potency and selectivity against PKA of ROCK inhibitors, VPI introduced a solubilizing side chain at the 3-position of phenylethylamide in the previously reported N-acylated aminothiazole compound a (Fig. 6) [53,63]. SAR analysis indicated that the best solubilizing group was solubilized piperidine or piperazine linked by a 3-carbon linker, which resulted in solubility values > 200 µmol/L for this class of compounds (Fig. 6). Modeling and X-ray crystallographic analysis showed that the inhibitors with additional solubilizing groups bind to Asp117 of ROCK, providing additional interactions [53]. Piperidine derivatives with nitrogen atoms at the end of the molecule, such as compounds 23 and 24 (Fig. 4), exhibited the best ROCK inhibition. In contrast, compound 23 showed no significant inhibition towards the 20 kinases tested (all IC50 values > 1 µmol/L) [53].

    VPI modified the 2-amino substituent of N-acylated aminothiazoles compound b (Fig. 6) to a 4-substituted 7-azaindole ring and reported a highly potent and PKA selective series 7-azaindole-based ROCK inhibitors [54]. The best compounds reported were shown to have a high ROCK potency with Ki values in the single-digit nanomolar level; for example, compound 25 (Fig. 4) demonstrated a greater ROCK-Ⅰ inhibition with Ki < 1 nmol/L [54]. SAR indicated that 3-substituted 7-azaindole substitution enhances the ROCK potency (Fig. 6). X-ray crystal structure indicated that the lowest energy state for compounds occurs when the biaryl from the 4-position is shifted to the 3-position of 7-azaindole, and thiazole and azaindole rings are coplanar. The 3-substituted 7-azaindoles maintain the same ROCK inhibitory effect and the corresponding hydrogen bond, indicating that the 3-substituted 7-azaindoles are bioisosteric replacements for 4-substituted pyridines. The 3-substituted 7-azaindole compounds form three hydrogen bonds between the 7-azaindole nitrogen and Met156, azaindole pyrrole NH and backbone carbonyl of Glu154, the backbone NH and Lys105. Furthermore, the addition of the solubilizing group piperazine N-Me, which interacts with Asp117, improves the ROCK inhibition potency and selectivity for PKA and CYP (Fig. 7).

    Figure 7

    Figure 7.  Crystal structure of compound 25 complexed with the kinase domain of ROCK-Ⅰ (PDB ID: 5KKT). The crystal structure of compound 25 (green) bound to the active site of ROCK-Ⅰ. Hydrogen bonds are shown as dashed lines.

    VPI next explored 7-azaindole substituents at the 4- and 5-positions with fluorine, chlorine, bromine, and hydroxyl. Moreover, VPI designed a series of ROCK-Ⅰ inhibitors (compounds 2630) (Fig. 4) that exhibited excellent ROCK potency. Compound 30 showed the highest PKA selectivity (> 480-fold) while maintaining excellent ROCK-Ⅰ potency (Ki = 0.001 nmol/L) [55]. SAR analysis indicated that the central thiazole ring substitution was more effective than substitutions with other five- or six-membered rings in enhancing ROCK inhibition potency. The 4- or 5-position fluorine substitution resulted in better ROCK potency, human monocytic leukemia cell (THP-1) migrating cell potency, and PKA selectivity (Fig. 6). In vivo rat PK profiling indicated that compounds substituted with solubilizing chains for the R group exhibited higher plasma clearance (> 65), higher volume of distribution (> 10), and longer T1/2 (> 3 h).

    BMS scientists have identified a series of novel phenylpyrazole amide derivatives through internal kinase data mining, rational design, and SAR analysis [64]. The pyrazole part of the molecule is considered to be the hinge binder. Compound 31 (Fig. 8) was discovered to be a potent dual ROCK-Ⅰ/Ⅱ inhibitor. The introduction of butylbenzene with a polar group (amino) to the aniline amide structure effectively improved metabolic stability, solubility, and inhibition of ROCK. Among the identified phenylpyrazole amide derivatives, the compound 31 has a balanced kinome selectivity (value 6/261) and a half-life value of 41 min for human liver microsome (HML) metabolic stability. Furthermore, compound 31 demonstrated good PK profiles with Cmax 28 nmol/L, AUC of 140 nmol/L*h. The Parallel artificial membrane permeation assay (PAMPA) value was 450 nm/740 sec at pH 5.5 and 7.4. Also, the in vivo experiments results indicate that the antihypertensive effect of compound 31 is dose-dependent.

    Figure 8

    Figure 8.  Pyrazole, thiophene substituted pyrrolopyridine, and benzothiophene substituted pyridine based ROCK inhibitors.

    Miao et al. discovered a hit compound based on molecular docking virtual screening (VS) approach [65]. The relevant structural optimization and SAR evaluation were performed based on the hit compound (Fig. S7 in Supporting information). The SAR analysis demonstrate the thieno[2,3-d]pyrimidin-4(3H)-one moiety is the optimal fragment in region Ⅰ, which made the compound exhibit a significant improvement in ROCK inhibitory potency. Region Ⅲ was fixed in its original form, as the other substituted phenyl groups of region Ⅲ could not improve the compound's activity. The substitution of region Ⅱ showed that the introduction of either the pyridine group or 7-azaindole enhanced the ROCK inhibitory activity, but 7-azaindole was the best choice for region Ⅱ. The most representative compound 32 (Fig. 8) displayed IC50 values against ROCK-Ⅰ and ROCK-Ⅱ of 0.004 and 0.001 µmol/L, respectively. The in vitro experiments showed compound 32 effectively changed cell shape, prevented cell migration and inhibited ROCK activity in intact cells.

    Benzothiazole-substituted pyridine derivatives as ROCK inhibitor scaffolds are more classical [66]. Judge et al. synthesized a series of 2-aminobenzothiazole-like inhibitors of ROCK-Ⅰ/Ⅱ based on molecular docking and SAR approaches [67]. The investigators tried to introduce the benzylamine group to benzothiazole. The crystal structures of compound-ROCK complexes indicated that (R)-configuration benzyl moiety in these compounds is sandwiched by hydrophobic interactions between phenylalanine 55, lysine 73, and the flexible glycine-rich loop of the kinase. Such interaction often plays a large role in potency and kinase selectivity. Therefore, the (R)-configuration is the preferred stereochemistry at the benzyl carbon of these compounds. As a result, compound 33 (Fig. 8) was obtained. The phenyl group at the end of 33 contributes to its activity, and the introduction of N-methyl-N-(3-methylbutyl)methanamide at the phenyl 3-position afforded compound 34 (Fig. 8) and a slight decrease in ROCK inhibition was observed, but the selectivity was greatly enhanced. Although some compounds exhibited higher ROCK inhibition by compound modification, only compounds 33 and 34 showed the cleanest kinome profiles. In cellular assays, both compounds showed submicromolar activity and exhibited a role as ROCK inhibitors in pain management. In a model of chronic constriction injury of the sciatic nerve, compound 33 exhibited the same efficacy as fasudil.

    Some of the ROCK inhibitors are derivatives based on the bicyclic aromatic ring ([6,5]- and [6,6]-fused rings) scaffolds. Previous studies showed that the potent ROCK inhibitors with good enzymatic and cellular potency can be developed by attaching aryl or heterocyclic portions (e.g., pyrazoles, amino pyrimidines) to bicyclic aromatic ring scaffolds using appropriate linkers [65,67], such as the benzimidazole-substituted pyrazole scaffolds [68,69]. Unlike the pyrazole or aminopyrimidine ROCK inhibitors with a bicyclic aromatic ring as the central part, the following compounds are all bicyclic aromatic ring derivatives.

    AR-12286 is a ROCK inhibitor developed by Aerie Pharmaceuticals, but it failed to pass the phase Ⅱ clinical trial [70,71]. Aerie described the discovery of novel amino-isoquinoline amide ROCK inhibitors that provide a longer duration of IOP-lowering effect in animal models than AR-12286 [72]. The 2,4-dimethyl benzoate analog 35 (Fig. 9) is the most effective compound obtained by a series of modifications based on AR-12286. First, the introduction of the α-aryl-β-amino group at the aminoalkyl arm of AR-12286 then substituted with hydroxymethyl at the 4-position of the α-aryl ring. To reduce the hyperemia and improve corneal penetration, the group prepared esters of the 4-hydroxymethyl fraction. These modifications greatly enhanced the inhibitory effect on ROCK. Specifically, the α-aryl-β-amino derivative exhibited significant improvement of ROCK potency in both human trabecular meshwork (HTM) and primary porcine trabecular meshwork (PTM) cell assays. Compound 35 significantly lowers its IOP sustainably and producing only mild hyperemia with no chemosis. The introduction of the α-aryl-β-amino portion mentioned earlier resulted in compound 35 have both (S) and (R) enantiomers. The S-enantiomer has better activity, and continued studies on compound 35 led to the discovery of the S-enantiomer, netarsudil [73].

    Figure 9

    Figure 9.  6- and 7-substituted isoquinoline, 4-chromanone, benzimidazole, purine derivatives as ROCK inhibitors.

    Hsu et al. reported an aminoisoquinoline derivative, ITRI-E-212 (structure not disclosed), could reduce IOP by increasing the water flow through the trabecular meshwork [74]. ITRI-E-212 exhibited an IC50 of 0.25 µmol/L for ROCK-Ⅱ. ITRI-E-212 effectively penetrates the cornea and maintains ROCK-Ⅱ inhibitory activity in the aqueous humor. In a normotensive rabbit model, nearly 25% of the peak percentage change in IOP was observed within 4 h after administration of 1.0% ITRI-E-212 eye drops. In a hypertensive rabbit model, IOP was reduced by 28.6% after 6 h of administration of 1.0% ITRI-E-212 drops. In this study, 1.0% ITRI-E-212 maintained low IOP for at least 6 h with only transient mild conjunctival congestion following a once-daily dosing regimen. The drug still needs further human trials.

    ITRI-E-247 (structure not disclosed) is a highly selective and potent ROCK inhibitor reported by Tu et al. in 2019. This agent inhibits MLC phosphorylation in a concentration-dependent manner and causes cell relaxation in vascular smooth muscle cells (A7r5 cells). In rabbits, 0.5% ITRI-E-247 drops significantly reduced IOP by 32% at the time point of 6 hours post topical drops with no severe ocular toxicity observed. No information is available regarding its use in clinical trials [75].

    Kesar et al. reported a series of ROCK inhibitors developed based on pharmacological models [76]. This pharmacological model contains four basic chemical features: hydrogen-bond acceptor (HBA), hydrogen bond donor (HBD) and hydrophobic agent (HY). NSC 2488 (36) (Fig. 9) exhibited the highest inhibitory potential (IC50 = 8.02 nmol/L) and showed comparable potency to Y-27632. The molecular interactions between the functional groups of NSC 2488 and the active site amino acids of the ROCK-Ⅱ protein structure are summarized as follows. The nitro group of 6-methoxyquinolin-4-yl showed hydrogen bonding with GLY104. The methyl group interacts with Lys121 and Phe103 through van der Waals force and the hydroxyl group of 2-butanol interacts with Gly234 through van der Waals force too. In addition, Asn219, Asp232, Met169, Ala119 and Val106 formed a hydrophobic environment around NSC 2488. Kesar et al. synthesized two new ROCK inhibitors NSC 4231 (37) and NSC 2888 (38) (Fig. 9) with IC50 values of 9.01 and 8.4 nmol/L which are based upon the same pharmacophore as NSC 2488. Ex vivo experiments results revealed that all three compounds exhibited concentration-dependent relaxation and vasodilatory activity similar to standard Y-27632. The Tanimoto similarity index for the three compounds was 0.08, 0.12 and 0.13 (< 0.5), respectively, indicating all three are novel compounds.

    Different from 6-amino-substituted isoquinoline derivatives, Domokos et al. synthesized the ROCK inhibitor FF 852 (39) (Fig. 9) through linkage dialkoxy to the 6- and 7-positions of the tetrahydroisoquinoline core. The replacement of the tetrahydroisoquinoline with a benzene ring resulted in the synthesis of FF 218 (40) (Fig. 9). FF 852 and FF 218 relax uterine smooth muscle in pregnant rats. Furthermore, FF 852 and FF 218 relax the uterus during parturition (when ROCK expression was high) in rats with greater potency than fasudil [77].

    Zhao et al. reported a series of new ROCK inhibitors based on the 4H-chromen-4-one core. [78]. Among which, the compound 41 (Fig. 9) is the most representative potent and kinome-selective dual ROCK-Ⅰ and ROCK-Ⅱ inhibitor. Compound 41 form four hydrogen bonds with ROCK-Ⅱ. Among them, the carbamoyl group flanked by the intermediate group formed three hydrogen bonds with ROCK-Ⅱ residues Phe103, Asp232 and Lys121. In addition, the carbonyl of the chromone forms a hydrogen bond with residue Met172, and this bond is very important for stabilizing the orientation and conformation of 41. The in vitro assays results showed compound 41 prevents endothelial cell migration and rescues neurons from high glucose-induced injury. In retinal explants, compound 41 effectively reduced high glucose-induced Müller glial growth. In general, compound 41 is promising for diabetic retinopathy (DR) treatment.

    Several studies have demonstrated that benzimidazole substituted pyrazole and piperazine groups are the active structural components of many ROCK inhibitors [68,69,79]. By linking a piperazine moiety to the benzimidazole scaffold with a hydrophobic linker, many hydrophobic interactions arise with the ROCK hydrophobic P-loop. A series of benzimidazole derivatives containing acetamide-linked substituted piperazines were designed and synthesized by Abbhi et al. [80]. Compound 42 (Fig. 9) containing p-nitrophenyl substituted piperazine showed good ROCK-Ⅱ inhibitory activity. Compound 42 inhibited ROCK-Ⅱ enzymes at a rate similar to that of fasudil. The inhibition of ROCK-Ⅱ by compound 42 at a concentration of 0.5 and 1 mmol/L was 57.25% and 77.92%, respectively. Molecular docking analysis revealed there are two hydrogen bonding links between ROCK-Ⅱ and compound 42: the carbonyl oxygen of the acetamide and Lys121, the nitrogen of the acetamide linker, and Ala231. The acetamido portion of the compound increased its lipophilicity and promoted corneal permeation. The percentage reduction in IOP at peak effect (2.5 h after administration) of compound 42 was 51.56%. In the 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging assay, compound 42 had an IC50 value of 95.49 mg/mL, indicating that it has significant antioxidant potential and may protect neurons from oxidative stress. Because oxidative free radicals damage neuronal cells and cause optic nerve damage, ROCK inhibition combined with the targeting oxidative free radicals is an effective therapeutic route for glaucoma [81,82].

    A series of C6-substituted purine derivatives were designed by Voller et al. [83]. The most active compounds, 43 and 44 (Fig. 9), have IC50 values of 2.9 and 1.5 µmol/L, respectively. The SAR analysis revealed that the anti-ROCK activity could be further improved by substituting the purine C2 atom or by extension of the purine C6 side chain. Cellular assays show that this compound reduces downstream MLC phosphorylation by inhibiting ROCK and inhibits the invasion of ROCK-dependent melanoma cells into the collagen matrix. Overall, the increase in activity was modest, and all were lower than that of the classical inhibitor fasudil.

    Considering the ability of piperazine to form effective hydrogen bonds with ROCK, Li et al. introduced phenyl piperazine in the compound. A ROCK-Ⅰ inhibitor, FPND (45) (Fig. 10), was recently reported by Li et al. [84]. The structure of 45 consists mainly of naphthalene, triazine and phenylpiperazine. SAR analysis revealed that the fragments of the naphthalene ring are essential for maintaining the activity of 45; phenylpiperazine is also required for anti-hemorrhagic activity, particularly the fluorine atom of phenylpiperazine, predominantly controlled 45 activity independently. Studies showed 45 interacts with ROCK-Ⅰ mainly through the following nonpolar links: Val90 and Lys105 with the naphthalene ring and fluorophenyl ring with residues Met153 and Met156. Further studies demonstrate 45 prevents statin-induced cerebral hemorrhage in a zebrafish model through inducing cytoskeletal rearrangement and enhancement of cell–cell junctions in endothelial cells (ECs) via ROCK-Ⅰ and VE-cadherin (VEC) signaling pathways [84-86]. In addition, 45 exhibits very low toxicity in mice through oral administration. Further investigation on the interactions between 45 with ROCK-Ⅱ and VECs is needed. Ladduwahetty et al. [87] designed and synthesized a series of ROCK inhibitors with piperazine core based on SAR and molecular modeling analysis for identifying the small molecules with better central nervous system-penetrant and target specificity for treating the symptoms associated with Huntington's disease (HD). The lead compound c (Fig. S8 in Supporting information) showed good potency but poor kinase selectivity [60,63,88]. Fluorine substitution at the 3-position of pyridine greatly improved potency compared to the 2-position substitution. Substitution of the core benzene ring on the lead compound c with 2R-methylpiperazine resulted in a small increase in kinase activity and a significant increase in cellular potency (Fig. S8). The representative compound 46 (Fig. 10) exhibited good solubility, permeability, and kinome selectivity. The binding kinetics for compound 46 demonstrated a slow dissociation rate (kon: 4.06 × 105 s−1, koff: 1.31 × 10−3 s−1, residence time = 763 s). The relevant in vitro absorption, distribution, metabolism and excretion (ADME) assays indicated that incorporating fluorine atom on the methoxy group that on the electron-rich phenyl ring at the positions of o-, m-, and p- decreases the compound stability and leads to high efflux. The presence of the 3-fluoro substituent on the pyridine mitigates the risk of CYP3A4 inhibition compared to the 2-fluoro substituent, and compound 46 inhibited CYP3A4 with an IC50 value of 1.1 µmol/L. Comparing the PK parameters of compound 46 given in mice with the dose of 2 mg/kg through intravenous (i.v.) and oral administration revealed it has a high oral bioavailability [87]. Furthermore, compound 46 showed excellent properties of penetrating the blood–brain barrier and showed highest brain area under the concentration versus time curve (AUC). Compound 46 dose-dependently leads to an increase in protein kinase B (PKB, also known as AKT) phosphorylation (a distal substrate of the ROCK pathway) and a decrease in MYPT1 phosphorylation. Compound 46 showed a superior profile in all in vitro cellular readouts. It has a potent pharmacodynamic effect after oral administration. In the chemoproteomics KiNativ technology (ActivX Biosciences), dose- and time-dependent ROCK-Ⅰ and ROCK-Ⅱ target engagement were observed at a 10 mg/kg dose showing equipotency toward each isoform, with a free brain KiNativ ROCK-Ⅰ and ROCK-Ⅱ with the IC50 values at around 6 nmol/L. Administration of the pi-perazine analogue 46 in the heterozygote Q175DN KI mouse model of HD through chronic dosing (10 mg/kg BID for 90 days) resulted in brain concentrations that were lower than anticipated when compared to a single dose. Notably, neurological index scoring of the mice remained normal at all doses administered.

    Figure 10

    Figure 10.  Piperazine, boron, and ligooxazole and pyridyl oligomeric derivatives as ROCK inhibitors.

    In recent years, more and more studies have shown that the boronic acid part of the benzoxaborane compounds serves as the binding hinge to the residue part of various kinases. Dayal et al. synthesized a series of 3H-pyrazolo[4,3-f]quinolone ROCK inhibitors containing boronic acids [89]. Among which, the compound HSD1590 (47) (Fig. 10) showed greater than 90% inhibition at the dose of 50 nmol/L, with the IC50 values of 1.22 ± 0.19 nmol/L and 0.51 ± 0.05 nmol/L for ROCK-Ⅰ and ROCK-Ⅱ, respectively. SAR analysis showed that the compounds containing phenylboronic acids substituted at the 3-position exhibited better ROCK inhibitory activity than furanyl or phenylthio boronic acids, while the introduction of a methoxy at the 2-position in the phenyl group significantly increased the inhibition activity. The in vitro assay showed 47 inhibited the migration of breast cancer cell line MDA-MB-231 at 1 µmol/L with low cytotoxicity. Boron derivatives have been less studied as ROCK inhibitors, and the previously discovered boron derivatives have a poor ROCK inhibitory effect. 47 is one of the most effective ROCK inhibitors, but more investigations are needed.

    G-quadruplexes (G4s) are four-stranded nucleic acid secondary structures [90]. Given that G4s are frequently associated with breakpoint regions during DNA replication, telomeric dysfunctions and repression of oncogene expression, G4-binding molecules may have the potential to be developed as one approach for treating cancer [91]. Verga et al. identified some oligooxazole/pyridyl oligomeric derivatives as ROCK-Ⅱ inhibitors in synthesizing compounds that target G4 [92]. The in vitro experiments have shown that the oligooxazole/pyridyloligomeric compounds all possess G4 binding activity. Among these compounds, iso-TOXaPy (48) (Fig. 10) exhibited excellent ROCK-Ⅱ inhibitory effects with an IC50 value of 29 nmol/L. Further studies revealed that the 48 specifically targets ROCK-Ⅱ under both in vitro conditions and in cells. Besides, the cellular experiment results showed 48 also exhibits high antiproliferative effects on multiple cancer cell lines including glioblastoma, HeLa cells, epidermal cells and melanoma cells. Unlike the common water-soluble ROCK inhibitors, these oligoheteroaryl derivatives have poor water solubility. They are difficult to enter the organism's circulation directly. However, their flexible oligomeric scaffolds adopt multiple conformations that may facilitate intra-organoid circulation and influence their expansion, suggesting their potential applications in antitumor treatment.

    Currently, developing small molecular bispecific ROCK inhibitors is the hot topic in the field. This type of inhibitor includes ROCK inhibition plus another therapeutically relevant non-kinase target, as simultaneous inhibition of two or more synergistic targets provides better therapeutic efficacy.

    Poly(ADP-ribose) polymerase (PARP) inhibitors (PARPi) have therapeutic potential for ovarian cancer, thymic cancer, pancreatic cancer and prostate cancer [93]. Antolín et al. screened out the previously published commercial chemical probe UPF 1069 (49) (Fig. 11), a 2H-isoquinoline 1-one scaffold substituted at position 7 [94,95]. 49 has the IC50 values of 0.1 and 1.0 µmol/L for inhibiting PARP1 and PARP2 under in vitro conditions, respectively. 49 and hydroxyfasudil have benzamide pharmacodynamic groups in their structures, and both compounds have similar structural properties with micromolar PARP inhibitor CHEMBL1767049 [96,97]. Previous studies demonstrated there are cross-pharmacological properties between ROCK and PARP, and a query of open targets mining revealed that the two protein families share seven diseases areas. The team, therefore, suspected that 49 also has ROCK inhibitory activity and subsequently found that 49 also showed 38% ROCK-Ⅰ and 20.5% ROCK-Ⅱ inhibition at 10 µmol/L, with an IC50 value of 9.3 µmol/L for ROCK-Ⅰ. Dual PARP-ROCK inhibitors may play an important role in cancer, DR, AD, cardiovascular disease, skin and skeletal system disorders, and therefore 49 is an excellent lead compound for further study.

    Figure 11

    Figure 11.  Bis-functional ROCK inhibitors.

    The Food and Drug Administration (FDA) has approved three rheumatoid arthritis (RA) indications drugs: baricitinib, tofacitinib and upadacitinib. Early clinical trials have demonstrated the application of Janus kinase (JAK) inhibitors as an effective treatment for a variety of autoimmune diseases [98-100]. Celon Pharma S.A. reported CPL409116 (50) (Fig. 11), a potent and selective inhibitor of JAK and ROCK as an alternative therapeutic agent for the systemic lupus erythematosustreatment (SLE) [101]. JAK is a family of non-receptor tyrosine kinases and co-mediates signals from growth factors with tyrosine kinase 2 (Tyk2). Upon activation, JAK leads to downstream phosphorylation of signal transducers and activates the transcription (STATs) of genes associated with the development of SLE symptoms. Since ROCK plays an important role in immune cell regulation and inflammatory disease development, inhibition of ROCK may also be exploited for treating SLE [102,103]. Surprisingly, the ADP-Glo data showed 50 is over 40 times more potent in blocking ROCK activity than fasudil. Besides, 50 also displayed high affinity towards JAK1/2/3 kinases (IC50 0.95/5.36/0.87 nmol/L, respectively) and moderate towards ROCK-Ⅰ/Ⅱ kinases (IC50 110.6/106.7 nmol/L, respectively). This dual inhibitor is competitive but binds to additional enzymes at higher concentrations. For example, Tyk (IC50 = 63.21 nmol/L). The cellular experiment results showed 50 effectively blocked the proliferation of CD4 T cells and the activation of B cells and reduced the expression of multiple cytokines [including interleukin-1β, 4, 10, 17, 21 and 22, interferon gamma (IFNG) and tumor necrosis factor alpha (TNFα)] by mediating ROCK and JAK. Moreover, the oral administration of 20 mg/kg compound 50 twice a day to SLE mice effectively alleviated SLE-associated organ damage (predominantly in kidneys) and improved the survival rate of SLE mice in a dose-dependent manner. Currently, 50 is going through phase Ⅰ or Ⅱ clinical trials and exhibited excellent activity in treating active RA (Table 1).

    Table 1

    Table 1.  Overview of ROCK inhibitors in clinical trials.
    DownLoad: CSV

    The (4Z)-2-(4-chloro-3-nitrophenyl)-4-(pyridin-3-ylmethylidene)-1,3-oxazol-5-one (DI) (51) (Fig. 11), molecule was originally discovered as a potent inhibitor of Zipper-interacting protein kinase (ZIPK) (IC50 = 69 nmol/L) and death-associated protein kinase 3 (DAPK3) (IC50 = 225 nmol/L) [104].

    AI-Ghabkari et al. [104] re-examined the 51 and found that ROCK-Ⅱ is prominently off-target for 51. The application of 51 to human coronary artery vascular smooth muscle cells (CASMCs) attenuated myosin regulatory light chains (LC20), Par-4, LIMK and cofilin phosphorylations. However, phosphorylation of cofilin was not significantly reduced with the ZIPK/DAPK-selective inhibitor (HS-38). Cofilin has been identified as a downstream target of ROCK signaling, suggesting that 51 has a ROCK inhibitory effect. Enzyme kinetic experiments determined the potency of compound 51 (Ki = 132 nmol/L) toward ROCK-Ⅱ to be similar to that of Y27632 (Ki = 140 nmol/L) and fasudil (Ki = 158 nmol/L). Previous studies have shown that administration of compound 51 in vivo in spontaneously hypertensive rats effectively reduces blood pressure, reactive oxygen species (ROS) production, and inflammatory responses. 51 also inhibits TNFα-induced inflammatory responses through a ROS-dependent mechanism. All these results suggest that 51 is a promising ROCK/ZIPK/DAPK inhibitor [105].

    Chen et al. reported a novel, highly specific ROCK-Ⅰ and MYLK4 inhibitor, ITRI-E-(S)4046 (52) (Fig. 11) [106]. With strong synergistic inhibition of ROCK-Ⅰ and MYLK4, 52 reduced IOP in normotensive and hypertensive animal models with minimal hyperemia. Topical application of 0.1% 52 resulted in a peak active drug concentration in plasma at 2 h (3.0 ng/mL), while 52 has a short half-life and disappears within 60 min after topical administration to rabbit eyes. From hepatic microsomal stability studies, 52 gave a mean intrinsic clearance of 8.56 µL min−1 mg−1 in humans [107]. Further studies are needed to determine whether 52 interferes with other kinases and whether it has cytotoxic effects on ocular tissues.

    In a recent patent application, Aerie et al. reported synthesizing a series of ROCK and monoamine transporters (MAT) bi-functional inhibitors for targeting glaucoma [108]. The 6-substituted isoquinoline amide scaffold series is structurally similar to netarsudil and may act as rho kinase inhibitors. The chemistry section of the patent exhaustively described the synthesis of a diverse range of compounds prior to their potential applications, with a focus on anti-glaucoma activity. Unfortunately, patents do not report ROCK and MAT inhibition data, such as positive results related to kinase selectivity and hyperemia [109].

    Over the past few decades, given the high attrition rates, substantial costs, and slow pace of new drug discovery and development, repurposing drugs approved for the market has emerged as a potential strategy to reduce development costs. For example, thalidomide was initially marketed as a sedative in 1957 and later used to treat erythema nodosum leprosum (ENL) (1964) [110] and multiple myeloma (1999) [111]. Typically, the first step in drug repurposing is the identification of candidate molecules for a given indication (hypothesis generation), which can be subdivided into computational and experimental approaches. Computational methods include signature matching, computational molecular docking, genome-wide association studies, pathway or network mapping, retrospective clinical analysis, and the use of electronic health records. Experimental methods include binding assays to identify target interactions [112]. This approach can also develop ROCK inhibitors by matching the transcriptome signature of fasudil with other drugs. For example, Iorio et al. found that fasudil could be applied in amyotrophic lateral sclerosis (ALS) [113-115].

    Ansar et al. virtually screened 1852 ligands approved by the FDA to dock with binding mode of ROCK (PDB ID: pdb: 3NCZ). Conivaptan (53) and polydatin (54) (Fig. 12) from the FDA-approved Drugbank dataset were potentially repurposable inhibitors targeting ROCK-Ⅰ. By the same approach, the team also identified WWWF and WWVW as potential and novel tetrapeptide inhibitors targeting the ATP binding domain of ROCK-Ⅰ. Regrettably, in order to conduct therapeutic studies, it is necessary to experimentally validate these molecules or peptides. However, the authors did not disclose the precise structures and biological activity information of WWVW and WWWF [34].

    Figure 12

    Figure 12.  Structures of ROCK inhibitors discovered from drug repurposing and natural compounds.

    Some of the classic drugs with newly identified ROCK inhibitory effects are summarized below. ÖZDEMİR et al. found that the FDA-approved anti-type Ⅱ diabetes drug metformin inhibits ROCK through the connectivity map program. This study further revealed the mechanisms of metformin, which has a guiding significance for the treatment of various pathological conditions involving ROCK, such as asthma, cancer, erectile dysfunction and glaucoma [116]. Recently, Borin et al. found that melatonin delays cancer progression not only via inhibiting the tumor cells proliferation but also through a direct inhibition of ROCK-Ⅰ which results in the anti-metastatic effects in the tested cancer cell lines [117]. Nicotinamide is an inhibitor of multiple kinases, and Meng's team recently discovered that it has ROCK inhibitory properties [118]. The in vitro experiments results demonstrated that nicotinamide affects the pluripotency and differentiation of human embryonic stem cells. Also, nicotinamide promotes pancreatic progenitor cell differentiation through dual inhibition of ROCK and casein kinases [119]. Discovering more ROCK inhibitors through drug repurposing methods at a lower cost is a feasible research direction.

    The extraction of compounds from natural products represents a powerful strategy for the discovery of new bioactive molecules [120-122]. There are a number of research groups that did tremendous work to further optimize the natural origin ROCK inhibitors, and their recent progress has been summarized below.

    Arya et al. found that three plant constituents: acteoside, martinoside and osmanthuside β6 from Clerodendrum colebrookianum Walpers, made critical interactions with the drug targets ROCK-Ⅰ/Ⅱ. Specifically, the phenolic hydroxyl group from the aglycone portion of acteoside and osmanthuside β6 was found interact with the hinge region residue Met156/Met157 of ROCK-Ⅰ/Ⅱ [123]. Arya et al. designed a series of hybrid molecules based on caffeic acid, ferulic acid and p-coumaric acid structures (the aglycone portion of the three plant constituents) with the homopiperazine ring of fasudil [124]. Among which, the thiourea analogue 55 (Fig. 12) (IC50 = 25 µmol/L) showed more than 3-fold anti-proliferative activity than fasudil (IC50 = 87 µmol/L) in colon cancer cells (SW480). The molecular docking and molecular dynamics simulation results showed compound 55 forms three hydrogen bonds with ROCK-Ⅰ (Glu154, Met156, Asp202) and ROCK-Ⅱ (Lys106, Met157, Asp203), respectively. Cellular experiments results showed compound 55 has excellent anti-proliferation and anti-migration activities against tested cancer cell lines [124]. Besides, Arya et al. synthesized a series of ROCK inhibitors based on caffeic acid scaffolds. Basic medicinal chemistry rules, coupled with structural insights generated by docking, led to the design of compounds 56, 57, 58 and 59 (Fig. 12) were selected as potential ROCK-Ⅰ inhibitors. To further explore the inhibition mechanisms of the four lead compounds, approaches including Lipinski's rule of five (Ro5), density functional theory (DFT) calculations, molecular docking and molecular dynamics simulations were taken and the results suggest the compounds bind to ROCK-Ⅰ with a lower binding free energy than fasudil does. However, further cellular experiments are needed to validate the aforementioned mechanisms [125].

    Amen et al. screened 35 compounds from the total ethanol extract of medicinal mushroom Ganoderma (lingzhi), and all compounds showed ROCK inhibitory activities. Further studies revealed that the extract containing lanostane triterpenes exhibit highest ROCK inhibitory activity. Thus, lanostane triterpenes can be potential lead compounds for further investigation as ROCK inhibitors [126].

    Liu et al. elucidated that six isoquinoline alkaloids of the genus Lotus are potential ROCK-Ⅰ inhibitors. Compared with other alkaloids and fasudil, neoliensinine exhibits a higher ROCK-Ⅰ binding energy which makes it an irreversible ROCK-Ⅰ inhibitor [127]. In vitro cellular experiment and animal trial results showed that neoliensinine inhibits mesenteric vascular smooth muscle contraction [127,128].

    Wang et al. isolated nine new indolocarbazoles from marine-derived Streptomyces sp. DT-A61. This class of compounds showed significant activity against human prostate PC-3 cancer cells. Among which, the compound 60 (Fig. 12) showed significant inhibitory effects of ROCK-Ⅱ with an IC50 value of 5.7 nmol/L [129]. Similarly, Jiang et al. identified four new Medermycin-type naphthoquinones from Streptomyces, of which compound 61 (Fig. 12) exhibited moderate inhibition of ROCK-Ⅱ with an IC50 value of 9.0 ± 0.6 µmol/L [130].

    Ezzat et al. reported the E. longifolia aqueous extract inhibits ROCK-Ⅱ under in vitro conditions. The maximum percentage inhibition of ROCK-Ⅱ by transconiferyl aldehyde isolated from this plant was 82.1% ± 0.63% with an IC50 value of 153 ± 14.1 ng/mL, which was better than the IC50 value of standard Y-27632 (192 ± 8.37 ng/mL) [131].

    Nimgampalle's study revealed that plant extracts, such as quadrangularin, luteolin and laetevirenol, have ROCK-Ⅱ inhibitory effects [132]. Amen et al. found that C. lacryma-jobi seeds extracts β-sitosterol, stigmasterol, chlorogenic acid, 2-O-β-glucopyranosyl-7-methoxy-2H-1,4-benzoxazin-3(4H)-one have higher ROCK inhibitory activities compare to the control Y-27632 [133]. In addition, natural extracts such as glycyrrhizin [134], curcumin [135], sodium aescinate [136], catalpol [137], terpinen-4-ol [138], total flavone from Rhododendron simsii Planch flower [139,140], loganin [141], ferulic acid [142], tetrandrine [143], oxymatrine [144], baicalein [145], sanggenon c [146] and hirudin [147], were found to be potential ROCK inhibitors as well.

    In addition to the tetrapeptides mentioned in the previous section on drug repurposing, other ROCK inhibitory peptides are summarized below. Considering that the dimerization domains (coiled-coil domains) of two ROCK isoforms share relatively low conservation (55%–60%) and ROCK dimerization is a critical biological event in activating the Rho signaling pathway, developing dimerization-disrupting peptides to selectively target the dimerization domains of the two isoforms is also an effective strategy for ROCK inhibitor development [148].

    By isolating three helical peptides: H1, H2 and H3 from the kinase dimerization domain, Gu et al. found that the ROCK-Ⅰ H2 and ROCK-Ⅱ H2 peptides could bind to the active pocket of the kinase dimerization domain [149]. Through hundreds of parallel simulated annealing runs, two peptides were obtained that possess optimal selectivity for ROCK and can effectively inhibit its activity. IoII-p5 (ROCK-Ⅰ over ROCK-Ⅱ) and IIoI-p1 (ROCK-Ⅰ over ROCK-Ⅱ) were designed to have strong IoII and IIoI selectivity for ROCK-Ⅰ over ROCK-Ⅱ (83.9-fold) and for ROCK-Ⅱ over ROCK-Ⅰ (> 21.2-fold), respectively (Table 1). Structural modeling results revealed that the non-conserved residues (including T1, D9, V10 and Y13) play a crucial role in defining the peptide selectivity, while conserved residues confer stability to kinase peptide binding [149]. The above results identify the crucial residues of such peptides and their functions, which are important for developing ROCK inhibitors.

    Abbasgholizadeh et al. screened a phage display library at high levels of ATP (1 mmol/L) and detected peptides that bind to the ROCK N' terminal catalytic domain of ROCK [24]. The investigators identified peptide7, a highly selective allosteric ROCK inhibitory peptide. Considering the long sequence of peptide7, the investigators identified the critical residues of peptide7. They found that tyrosine-serine-proline-serine (YSPS) is the critical amino acid residue that determines the inhibitory properties of the peptide, and obtained peptide22, a highly selective allosteric ROCK inhibitory peptide, with an IC50 value of 44.4 µmol/L for ROCK-Ⅰ (Table 2). In the wire myography assay, peptide22 effectively relaxed the contracted aortic rings, and this effect was in a dose-dependent manner with the maximum inhibitory effect observed at 250 µmol/L. Furthermore, In the presence of 0.1% DMSO, peptide22 was shown to be effective in the corneal neovascularization model. Peptide22 is a very potential ROCK inhibitory peptide. It has shown good activity in tissue and organ assays, but its in vivo activity, efficacy, safety and specific mechanisms remain to be investigated. The highlight is that this study reveals that tthese peptides effectively inhibit ROCK, which could guide the development of novel ROCK inhibitory peptides [24]. In the future, peptide drug delivery systems (DDS) capable of binding to peptide inhibitors to enhance their bioavailability may play a significant role in the further development of ROCK inhibitors [150].

    Table 2

    Table 2.  Peptides reported by Gu and Abbasgholizadeh [24,149].
    DownLoad: CSV

    ROCK is an attractive target for various pathological conditions as a critical mediator of the cellular response to environmental signals. ROCK is distributed throughout the body, with ROCK-Ⅰ mainly in the lung, liver, testis, blood and immune system, and ROCK-Ⅱ more significantly expressed in the brain, heart and smooth muscle cells. ROCK signaling is involved in various processes, such as metastasis, invasion, tumor growth and angiogenesis in various cancers. ROCK also plays a vital role in regulating cellular morphology and cardiovascular function. Therefore, research targeting ROCK is rapidly developing in treating various diseases such as cancer, neurological diseases, inflammation, cardiovascular diseases and glaucoma. Although several ROCK inhibitors have been developed, only four have been approved for clinical application, including fasudil, ripasudil, netarsudil and belumosudil. In this review, we summarized more than 60 ROCK inhibitors with different chemical families, including isoquinoline, pyridine, indazole, pyrimidine, pyrroloxypyridine, pyrazole, thieno[2,3-d]pyrimidin-4(3H)-one, benzothiophene, 4-chromanone, benzimidazole, naphthalene, purine, boron, ligooxazole/pyridyl oligomeric derivatives and other small molecule inhibitors. These small molecule inhibitors tend to exhibit more potent ROCK inhibition. However, the development of these compounds is still in a relatively early stage, and most of the ROCK inhibitors reported to be effective are ATP-competitive inhibitors. The selectivity of these compounds may be limited. Due to the pervasive distribution of ROCK in the human body, its inhibition may cause some adverse reactions caused by the off-target effects observed in both preclinical studies and clinical applications, such as renal insufficiency, intestinal obstruction and systemic hypotension. Therefore, the screening and design of ROCK-targeted drugs require more consideration of their protein structure, and the selectivity, inhibitory efficacy and safety must be balanced. Scientists have observed that DDS possess the capability to administer medications for diverse ailments, surmounting issues related to drug inefficacy and delivery hurdles. The potential of DDS is not confined to a particular disease but can be employed for a multitude of maladies. For instance, nanomaterials and liposomes as biological drug carriers have shown promise in delivering drugs for cancer and Alzheimer's disease treatment [151-154]. Thus, the development of new DDS could potentially optimize the pharmaceutical profiles of ROCK inhibitors [155-158], with articles reporting VA-coupled liposomal ROCK inhibitors that target hepatic stellate cells to inhibit hepatic fibrosis without causing systemic side effects [156,159]. In addition, several studies reported the selection of macrocyclic host molecules or peptides as molecular carriers for ROCK inhibitors, which improved drug stability and bioavailability [160,161]. The potential approaches include developing bioreductive prodrugs and photo-controllable drugs [162,163]. The development of stable and controllable ROCK inhibitors is the future direction of solving the adverse reaction problem.

    In addition to the development of novel small molecule inhibitors for ROCK, the identification of ROCK inhibitors from existing marketed drugs holds promise for the discovery of multi-target inhibitors with enhanced efficacy. For example, compound 50, a multi-target inhibitor of JAK and ROCK, has been approved for phase Ⅱ clinical trials in active RA. Multi-target ROCK inhibitors have a higher potential to simultaneously inhibit two or more multiple targets than single-target ROCK inhibitors. Furthermore, the natural extract derivatives of ROCK inhibitors have been reported and are in the early stages of preclinical studies. Further biophysical studies of natural ROCK inhibitors and derivatives are warranted, which may contribute to developing effective ROCK inhibitors in the future. In addition to the standard small molecule compounds as ROCK inhibitors, we noted that peptide-based agents that target active pockets of the kinase dimerization domain might be the basis for further work.

    The recent years have been characterized by an impressive surge in developmental activities in the field of ROCK inhibitors focusing on improved target selectivity and refined pharmacological properties to reduce unwanted side effects. This review discusses some important SAR and ROCK inhibitors binding modes, design approaches, and clinical studies. By utilizing these structures and drug design approaches, the development of more effective and selective ROCK inhibitors could provide a stronger scientific basis for treating various diseases.

    The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

    This work is supported by the National Natural Science Foundation of China (Nos. 82073311 and 81973866), Natural Science Foundation of Sichuan Province (No. 2022JDTD0025), Science and Technology Research Project of Sichuan Traditional Chinese Medicine Administration (No. 2021ZD016).

    Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2023.108336.


    1. [1]

      J. Hu, P. Zhu, Y. Li, Y. Chen, Chin. Chem. Lett. 29 (2018) 1043-1050. doi: 10.1016/j.cclet.2018.05.035

    2. [2]

      J. Shi, L. Wei, Arch. Immunol. Ther. Exp. 70 (2022) 4. doi: 10.1007/s00005-022-00642-z

    3. [3]

      T. Ushijima, N. Fujimoto, S. Matsuyama, et al., J. Biol. Chem. 293 (2018) 148-162. doi: 10.1074/jbc.m117.813931

    4. [4]

      N. Karimi, F. Bayram Catak, E. Arslan, et al., Int. Immunopharmacol. 113 (2022) 109445. doi: 10.1016/j.intimp.2022.109445

    5. [5]

      D.R. Thal, S.O. Tome, Brain Res. Bull. 190 (2022) 204-217. doi: 10.1016/j.brainresbull.2022.10.006

    6. [6]

      C. Hu, H. Zhou, Y. Liu, et al., Int. J. Oncol. 55 (2019) 833-844.

    7. [7]

      Y. Hiroi, K. Noma, H.H. Kim, et al., Circ. J. 82 (2018) 1195-1204. doi: 10.1253/circj.CJ-17-0732

    8. [8]

      L. Julian, M.F. Olson, Small GTPases 5 (2014) e29846. doi: 10.4161/sgtp.29846

    9. [9]

      W. Chen, M.S. Nyuydzefe, J.M. Weiss, et al., Sci. Rep. 8 (2018) 16636. doi: 10.1038/s41598-018-35109-9

    10. [10]

      N. Zhou, J.J. Lee, S. Stoll, et al., Cell. Physiol. Biochem. 44 (2017) 701-715. doi: 10.1159/000485284

    11. [11]

      E. Berrino, C.T. Supuran, Expert Opin. Ther. Pat. 29 (2019) 817-827. doi: 10.1080/13543776.2019.1670812

    12. [12]

      C. Vennin, V.T. Chin, S.C. Warren, et al., Sci. Transl. Med. 9 (2017) eaai8504. doi: 10.1126/scitranslmed.aai8504

    13. [13]

      L. Deng, G. Li, R. Li, et al., Cancer Biol. Ther. 9 (2010) 875-884. doi: 10.4161/cbt.9.11.11634

    14. [14]

      I. Hinsenkamp, S. Schulz, M. Roscher, et al., Neoplasia 18 (2016) 500-511. doi: 10.1016/j.neo.2016.07.002

    15. [15]

      J.M. Thompson, Q.H. Nguyen, M. Singh, et al., Oncogene 36 (2017) 1080-1089. doi: 10.1038/onc.2016.272

    16. [16]

      Y. Xia, X. Cai, J. Fan, et al., Anticancer Drugs 28 (2017) 514-521. doi: 10.1097/CAD.0000000000000490

    17. [17]

      Z.H. Wang, D. Zhu, S. Xie, et al., J. Cardiovasc. Pharmacol. 70 (2017) 102-109. doi: 10.1097/FJC.0000000000000496

    18. [18]

      N.T. Abdali, A.H. Yaseen, E. Said, T.M. Ibrahim, Naunyn. Schmiedebergs Arch. Pharmacol. 390 (2017) 409-422. doi: 10.1007/s00210-017-1343-x

    19. [19]

      T.L. Montagnoli, J.S. da Silva, S.Z. Sudo, et al., Cells 10 (2021) 1648. doi: 10.3390/cells10071648

    20. [20]

      D. Xueyang, M. Zhanqiang, M. Chunhua, H. Kun, Oncotarget 7 (2016) 78764-78772. doi: 10.18632/oncotarget.12853

    21. [21]

      L. Qi, Y.G. Tang, L. Wang, et al., J. Neurol. Sci. 370 (2016) 21-26. doi: 10.1016/j.jns.2016.08.061

    22. [22]

      J.L. Labandeira-Garcia, A.I. Rodriguez-Perez, B. Villar-Cheda, et al., Neuroscientist 21 (2015) 616-629. doi: 10.1177/1073858414554954

    23. [23]

      J. Han, Y. Zhao, Y. Zhang, et al., Front. Pharmacol. 9 (2018) 87. doi: 10.3389/fphar.2018.00087

    24. [24]

      R. Abbasgholizadeh, H. Zhang, J.W. Craft Jr., et al., Exp. Biol. Med. 244 (2019) 940-951. doi: 10.1177/1535370219849581

    25. [25]

      L. Song, C. Zhu, W. Zheng, et al., Comput. Biol. Chem. 87 (2020) 107314. doi: 10.1016/j.compbiolchem.2020.107314

    26. [26]

      S.P. Davies, H. Reddy, M. Caivano, P. Cohen, Biochem. J. 351 (2000) 95-105. doi: 10.1042/bj3510095

    27. [27]

      Y. Feng, P.V. LoGrasso, O. Defert, R. Li, J. Med. Chem. 59 (2016) 2269-2300. doi: 10.1021/acs.jmedchem.5b00683

    28. [28]

      T. Landry, D. Shookster, H. Huang, Front. Endocrinol. 11 (2020) 622581.

    29. [29]

      L. Huang, F. Dai, L. Tang, et al., Cell. Physiol. Biochem. 49 (2018) 565-577. doi: 10.1159/000492994

    30. [30]

      J. Du, J. Guo, D. Kang, et al., Chin. Chem. Lett. 31 (2020) 1695-1708. doi: 10.1016/j.cclet.2020.03.028

    31. [31]

      S. Ghosh, S. Keretsu, S.J. Cho, PeerJ 9 (2021) e11951. doi: 10.7717/peerj.11951

    32. [32]

      B. Bayel Secinti, G. Tatar, T. Taskin Tok, J. Biomol. Struct. Dyn. 37 (2019) 2457-2463. doi: 10.1080/07391102.2018.1491420

    33. [33]

      H. Chen, S. Li, Y. Hu, et al., Comb. Chem. High Throughput Screen. 19 (2016) 36-50. doi: 10.2174/1386207319666151203001946

    34. [34]

      S. Ansar, U. Vetrivel, J. Biomol. Struct. Dyn. 40 (2022) 7450-7468. doi: 10.1080/07391102.2021.1898470

    35. [35]

      A.A. Sadybekov, A.V. Sadybekov, Y. Liu, et al., Nature 601 (2022) 452-459. doi: 10.1038/s41586-021-04220-9

    36. [36]

      J.C. Koch, L. Tatenhorst, A.E. Roser, et al., Pharmacol. Ther. 189 (2018) 1-21. doi: 10.1016/j.pharmthera.2018.03.008

    37. [37]

      C.M. Chong, M.T. Kou, P. Pan, et al., Mol. Biosyst. 12 (2016) 2713-2721. doi: 10.1039/C6MB00343E

    38. [38]

      M. Rosen Kenneth, D. Abbinanti Matthew, J. Ruschel, et al., BioAxone BioSciences Inc. Patent, US2020199105 (A1), 2020.

    39. [39]

      L. McKerracher, R. Shenkar, M. Abbinanti, et al., Transl. Stroke Res. 11 (2020) 365-376. doi: 10.1007/s12975-019-00725-8

    40. [40]

      J. Liang, M. Tang, L. Wang, et al., Chem. Biol. Drug Des. 98 (2021) 1065-1078. doi: 10.1111/cbdd.13963

    41. [41]

      C.M. Avila, A.B. Lopes, A.S. Goncalves, et al., Eur. J. Med. Chem. 46 (2011) 1245-1253. doi: 10.1016/j.ejmech.2011.01.045

    42. [42]

      R.G. Oliveira, F.S. Guerra, C.D.S. Mermelstein, et al., J. Enzyme Inhib. Med. Chem. 33 (2018) 1181-1193. doi: 10.1080/14756366.2018.1490732

    43. [43]

      M. Uehata, T. Ishizaki, H. Satoh, et al., Nature 389 (1997) 990-994. doi: 10.1038/40187

    44. [44]

      M. Arita, T. Saito, H. Okuda, et al., Yoshitomi Pharmaceutical Industries Ltd., Patent, US19940204211, 1992.

    45. [45]

      M. Dietrich, H.P. Hartung, P. Albrecht, Neurol. Neuroimmunol. Neuroinflamm. 8 (2021) e976. doi: 10.1212/nxi.0000000000000976

    46. [46]

      S. Li, D. Wei, Z. Mao, et al., Bioorg. Chem. 73 (2017) 63-75. doi: 10.1016/j.bioorg.2017.05.010

    47. [47]

      Y. Yao, R. Li, X. Liu, et al., Molecules 22 (2017) 1766. doi: 10.3390/molecules22101766

    48. [48]

      L. Gong, J. Peng, L. Fang, et al., Molecules 17 (2012) 5935-5944. doi: 10.3390/molecules17055935

    49. [49]

      T.Y. Yuan, Y.C. Chen, H.F. Zhang, et al., Acta Pharmacol. Sin. 37 (2016) 604-616. doi: 10.1038/aps.2015.161

    50. [50]

      A.D. Hobson, R.A. Judge, A.L. Aguirre, et al., J. Med. Chem. 61 (2018) 11074-11100. doi: 10.1021/acs.jmedchem.8b01098

    51. [51]

      Z. Hu, C. Wang, D. Sitkoff, et al., Bioorg. Med. Chem. Lett. 30 (2020) 127474. doi: 10.1016/j.bmcl.2020.127474

    52. [52]

      Y. Sun, Y. Li, Z. Miao, et al., Bioorg. Med. Chem. Lett. 45 (2021) 128138. doi: 10.1016/j.bmcl.2021.128138

    53. [53]

      H. Gao, C. Marhefka, M.D. Jacobs, et al., Bioorg. Med. Chem. Lett. 28 (2018) 2616-2621. doi: 10.1016/j.bmcl.2018.06.043

    54. [54]

      U.K. Bandarage, J. Cao, J.H. Come, et al., Bioorg. Med. Chem. Lett. 28 (2018) 2622-2626. doi: 10.1016/j.bmcl.2018.06.040

    55. [55]

      U.K. Bandarage, J. Court, H. Gao, et al., Bioorg. Med. Chem. Lett. 33 (2021) 127721. doi: 10.1016/j.bmcl.2020.127721

    56. [56]

      S. Ma, L. Wang, B. Ouyang, et al., Bioorg. Med. Chem. 28 (2020) 115683. doi: 10.1016/j.bmc.2020.115683

    57. [57]

      L. Wang, B. Ouyang, M. Fan, et al., Iran. J. Pharm. Res. 20 (2021) 121-131.

    58. [58]

      L. Wang, J. Qi, M. Fan, L. Yao, Chem. Biol. Drug Des. 98 (2021) 969-978. doi: 10.1111/cbdd.13961

    59. [59]

      R. Pireddu, K.D. Forinash, N.N. Sun, et al., Medchemcomm 3 (2012) 699-709. doi: 10.1039/c2md00320a

    60. [60]

      Y. Feng, Y. Yin, A. Weiser, et al., J. Med. Chem. 51 (2008) 6642-6645. doi: 10.1021/jm800986w

    61. [61]

      Y. Feng, P. Lograsso, T. Bannister, et al., Patent, WO2008US13844, 2009.

    62. [62]

      J. Pan, Y. Yin, L. Zhao, Y. Feng, Bioorg. Med. Chem. 27 (2019) 1382-1390. doi: 10.1016/j.bmc.2019.02.047

    63. [63]

      J. Green, J. Cao, U.K. Bandarage, et al., J. Med. Chem. 58 (2015) 5028-5037. doi: 10.1021/acs.jmedchem.5b00424

    64. [64]

      Z. Hu, C. Wang, P.W. Glunz, et al., Bioorg. Med. Chem. Lett. 30 (2020) 127495. doi: 10.1016/j.bmcl.2020.127495

    65. [65]

      Z. Miao, Y.M. Sun, L.Y. Zhao, et al., Bioorg. Med. Chem. Lett. 30 (2020) 126966. doi: 10.1016/j.bmcl.2020.126966

    66. [66]

      Y. Yin, L. Lin, C. Ruiz, et al., Bioorg. Med. Chem. Lett. 19 (2009) 6686-6690. doi: 10.1016/j.bmcl.2009.09.115

    67. [67]

      R.A. Judge, A. Vasudevan, V.E. Scott, et al., Chembiochem 19 (2018) 613-621. doi: 10.1002/cbic.201700547

    68. [68]

      E.H. Sessions, Y. Yin, T.D. Bannister, et al., Bioorg. Med. Chem. Lett. 18 (2008) 6390-6393. doi: 10.1016/j.bmcl.2008.10.095

    69. [69]

      Y. Feng, T. Chen Yen, H. Sessions, et al., Patent, WO2011050245 (A1), 2011.

    70. [70]

      M.A. Delong, J. Sturdivant, S.M. Royalty, et al., Invest. Ophthalmol. Vis. Sci. 50 (2009) 4058.

    71. [71]

      M.A. deLong, J. Yingling, C.W. Lin, et al., Invest. Ophthalmol. Vis. Sci. 53 (2012) 3867.

    72. [72]

      J.M. Sturdivant, S.M. Royalty, C.W. Lin, et al., Bioorg. Med. Chem. Lett. 26 (2016) 2475-2480. doi: 10.1016/j.bmcl.2016.03.104

    73. [73]

      S.M. Hoy, Drugs 78 (2018) 389-396. doi: 10.1007/s40265-018-0877-7

    74. [74]

      C.R. Hsu, Y.H. Chen, C.P. Liu, et al., Invest. Ophthalmol. Vis. Sci. 60 (2019) 624-633. doi: 10.1167/iovs.18-25252

    75. [75]

      C.M. Tu, C.P. Liu, C.H. Chen, et al., Invest. Ophthalmol. Vis. Sci. 60 (2019) 3768. doi: 10.1167/iovs.18-26286

    76. [76]

      S. Kesar, S. Paliwal, P. Mishra, et al., ACS Med. Chem. Lett. 11 (2020) 1694-1703. doi: 10.1021/acsmedchemlett.0c00126

    77. [77]

      D. Domokos, F. Fulop, G. Falkay, R. Gaspar, Bioorg. Med. Chem. Lett. 28 (2018) 466-469. doi: 10.1016/j.bmcl.2017.12.017

    78. [78]

      L. Zhao, Y. Li, Y. Wang, et al., J. Med. Chem. 62 (2019) 10691-10710. doi: 10.1021/acs.jmedchem.9b01143

    79. [79]

      K.S. Oh, B.K. Oh, C.H. Park, et al., Eur. J. Pharmacol. 702 (2013) 218-226. doi: 10.1016/j.ejphar.2013.01.027

    80. [80]

      V. Abbhi, L. Saini, S. Mishra, et al., Bioorg. Med. Chem. 25 (2017) 6071-6085. doi: 10.1016/j.bmc.2017.09.045

    81. [81]

      E. Oshida, Y. Matsumoto, K. Arai, Clin. Ophthalmol. 4 (2010) 653-660.

    82. [82]

      A. Izzotti, A. Bagnis, S.C. Sacca, Mutat. Res. 612 (2006) 105-114. doi: 10.1016/j.mrrev.2005.11.001

    83. [83]

      J. Voller, L. Zahajska, L. Plihalova, et al., Bioorg. Chem. 90 (2019) 103005. doi: 10.1016/j.bioorg.2019.103005

    84. [84]

      S. Li, N. Ai, M. Shen, et al., Cell Death Dis. 3 (2017) 17051. doi: 10.1038/cddiscovery.2017.51

    85. [85]

      T. Skaria, E. Bachli, G. Schoedon, Cell Adhes. Migr. 11 (2017) 24-38. doi: 10.1080/19336918.2016.1178449

    86. [86]

      M. Corada, M. Mariotti, G. Thurston, et al., Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 9815-9820. doi: 10.1073/pnas.96.17.9815

    87. [87]

      T. Ladduwahetty, M.R. Lee, M.C. Maillard, et al., J. Med. Chem. 65 (2022) 9819-9845. doi: 10.1021/acs.jmedchem.2c00474

    88. [88]

      Y.D. Zhao, L. Cai, M.K. Mirza, et al., Am. J. Pathol. 180 (2012) 2268-2275. doi: 10.1016/j.ajpath.2012.02.016

    89. [89]

      N. Dayal, C.G. Mikek, D. Hernandez, et al., Eur. J. Med. Chem. 180 (2019) 449-456. doi: 10.1016/j.ejmech.2019.06.089

    90. [90]

      W. Liu, Y. Xu, X. Li, et al., Chin. Chem. Lett. 32 (2021) 2322-2326. doi: 10.1016/j.cclet.2021.01.045

    91. [91]

      S. Neidle, J. Med. Chem. 59 (2016) 5987-6011. doi: 10.1021/acs.jmedchem.5b01835

    92. [92]

      D. Verga, C.H. N'Guyen, M. Dakir, et al., J. Med. Chem. 61 (2018) 10502-10518. doi: 10.1021/acs.jmedchem.8b01023

    93. [93]

      C.J. Lord, A. Ashworth, Science 355 (2017) 1152-1158. doi: 10.1126/science.aam7344

    94. [94]

      A.A. Antolin, J. Mestres, ACS Omega 3 (2018) 12707-12712. doi: 10.1021/acsomega.8b02337

    95. [95]

      R. Pellicciari, E. Camaioni, G. Costantino, et al., Chemmedchem 3 (2008) 914-923. doi: 10.1002/cmdc.200800010

    96. [96]

      H. Chen, Y. Lin, M. Han, et al., J. Pharm. Biomed. Anal. 52 (2010) 242-248. doi: 10.1016/j.jpba.2009.12.028

    97. [97]

      A. Gaulton, A. Hersey, M. Nowotka, et al., Nucleic Acids Res. 45 (2017) D945-D954. doi: 10.1093/nar/gkw1074

    98. [98]

      J.J. O'Shea, D.M. Schwartz, A.V. Villarino, et al., Annu. Rev. Med. 66 (2015) 311-328. doi: 10.1146/annurev-med-051113-024537

    99. [99]

      D.M. Schwartz, Y. Kanno, A. Villarino, et al., Nat. Rev. Drug Discov. 17 (2017) 78.

    100. [100]

      L.M. Mathias, W. Stohl, Expert Opin. Ther. Targets 24 (2020) 1283-1302. doi: 10.1080/14728222.2020.1832464

    101. [101]

      M. Dulak-Lis, A. Bujak, K. Gala, et al., J. Pharmacol. Sci. 145 (2021) 340-348. doi: 10.1016/j.jphs.2021.02.002

    102. [102]

      R.A. Stirzaker, P.S. Biswas, S. Gupta, et al., Lupus 21 (2012) 656-661. doi: 10.1177/0961203312436862

    103. [103]

      P.S. Biswas, S. Gupta, E. Chang, et al., J. Clin. Invest. 120 (2010) 3280-3295. doi: 10.1172/JCI42856

    104. [104]

      M. Okamoto, K. Takayama, T. Shimizu, et al., Bioorg. Med. Chem. 18 (2010) 2728-2734. doi: 10.1016/j.bmc.2010.02.018

    105. [105]

      T. Usui, M. Okada, Y. Hara, H. Yamawaki, Hypertension 60 (2012) 1031-1039. doi: 10.1161/HYPERTENSIONAHA.112.200337

    106. [106]

      Y.H. Chen, W.Y. Lin, Y.C. Huang, et al., Invest. Ophthalmol. Vis. Sci. 62 (2021) 12.

    107. [107]

      H. Alrabiah, A.A. Kadi, M.W. Attwa, G.A.E. Mostafa, Chem. Cent. J. 12 (2018) 47. doi: 10.1186/s13065-018-0414-5

    108. [108]

      A. Bonardi, C.T. Supuran, Expert Opin. Ther. Pat. 29 (2019) 753-759. doi: 10.1080/13543776.2019.1658743

    109. [109]

      A. Delong Mitchell, M. Sturdivant Jill, M. Royalty Susan, Aerie Pharmaceuticals Inc., Patent, US2018244666A1, 2018.

    110. [110]

      T.T. Ashburn, K.B. Thor, Nat. Rev. Drug Discov. 3 (2004) 673-683. doi: 10.1038/nrd1468

    111. [111]

      S. Singhal, J. Mehta, R. Desikan, et al., N. Engl. J. Med. 341 (1999) 1565-1571. doi: 10.1056/NEJM199911183412102

    112. [112]

      S. Pushpakom, F. Iorio, P.A. Eyers, et al., Nat. Rev. Drug Discov. 18 (2019) 41-58. doi: 10.1038/nrd.2018.168

    113. [113]

      F. Iorio, R. Bosotti, E. Scacheri, et al., Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 14621-14626. doi: 10.1073/pnas.1000138107

    114. [114]

      R. Gunther, A. Balck, J.C. Koch, et al., Front. Pharmacol. 8 (2017) 17.

    115. [115]

      L. Tonges, R. Gunther, M. Suhr, et al., Glia 62 (2014) 217-232. doi: 10.1002/glia.22601

    116. [116]

      A. OzdemIr, M. Ark, Turk. J. Biol. 45 (2021) 35-45. doi: 10.3906/biy-2004-12

    117. [117]

      T.F. Borin, A.S. Arbab, G.B. Gelaleti, et al., J. Pineal Res. 60 (2016) 3-15. doi: 10.1111/jpi.12270

    118. [118]

      Y. Meng, Z. Ren, F. Xu, et al., Stem Cell Rep. 11 (2018) 1347-1356. doi: 10.1016/j.stemcr.2018.10.023

    119. [119]

      Y. Zhang, J. Xu, Z. Ren, et al., Stem Cell Res. Ther. 12 (2021) 1-12. doi: 10.1186/s13287-020-02006-w

    120. [120]

      C. Long, W.C. Luo, H.Y. Zhou, et al., Chin. Chem. Lett. 27 (2016) 247-250. doi: 10.1016/j.cclet.2015.10.005

    121. [121]

      C.L. Xie, D. Zhang, K.Q. Guo, et al., Chin. Chem. Lett. 33 (2022) 2057-2059. doi: 10.1016/j.cclet.2021.09.073

    122. [122]

      H. Wang, X. Bai, Y. Huang, et al., Chin. Chem. Lett. 34 (2023) 107671. doi: 10.1016/j.cclet.2022.07.014

    123. [123]

      H. Arya, S.B. Syed, S.S. Singh, et al., Interdiscip. Sci. Comput. Life Sci. 10 (2018) 792-804. doi: 10.1007/s12539-017-0243-6

    124. [124]

      H. Arya, C.S. Yadav, S.Y. Lin, et al., J. Biomol. Struct. Dyn. 38 (2020) 3563-3577. doi: 10.1080/07391102.2019.1664326

    125. [125]

      H. Arya, M.S. Coumar, J. Mol. Model. 26 (2020) 249. doi: 10.5958/0974-4576.2020.00043.2

    126. [126]

      Y. Amen, Q. Zhu, H.B. Tran, et al., J. Nat. Med. 71 (2017) 380-388. doi: 10.1007/s11418-016-1069-y

    127. [127]

      J. Liu, Y. Lu, G. Li, et al., J. Biomol. Struct. Dyn. 39 (2021) 379-394. doi: 10.1080/07391102.2020.1714484

    128. [128]

      G.M. Yang, J. Sun, Y. Pan, et al., Fitoterapia 124 (2018) 58-65. doi: 10.1016/j.fitote.2017.10.020

    129. [129]

      J.N. Wang, H.J. Zhang, J.Q. Li, et al., J. Nat. Prod. 81 (2018) 949-956. doi: 10.1021/acs.jnatprod.7b01058

    130. [130]

      Y.J. Jiang, D.S. Zhang, H.J. Zhang, et al., J. Nat. Prod. 81 (2018) 2120-2124. doi: 10.1021/acs.jnatprod.8b00544

    131. [131]

      S.M. Ezzat, M.M. Okba, M.I. Ezzat, et al., Evid. Based Complement. Alternat. Med. 2019 (2019) 4341592.

    132. [132]

      M. Nimgampalle, H.N. Banavath, H. Chakravarthy, et al., J. Biomol. Struct. Dyn. 38 (2020) 4669-4686. doi: 10.1080/07391102.2019.1686066

    133. [133]

      Y. Amen, Q. Zhu, H.B. Tran, et al., Nat. Prod. Res. 32 (2018) 1955-1959. doi: 10.1080/14786419.2017.1354183

    134. [134]

      Z. Xu, Z. Hu, H. Xu, et al., Exp. Cell Res. 411 (2022) 113008. doi: 10.1016/j.yexcr.2022.113008

    135. [135]

      Y. Zhou, P.J. Little, S. Xu, D. Kamato, Molecules 26 (2021) 2320. doi: 10.3390/molecules26082320

    136. [136]

      W. Liu, F. Qin, F. Wu, et al., Phytomedicine 69 (2020) 153193. doi: 10.1016/j.phymed.2020.153193

    137. [137]

      A. Shu, Q. Du, J. Chen, et al., Chem. Biol. Interact. 348 (2021) 109625. doi: 10.1016/j.cbi.2021.109625

    138. [138]

      W. Cao, R. Tian, R. Pan, et al., Bioengineered 13 (2022) 8643-8656. doi: 10.1080/21655979.2022.2054205

    139. [139]

      X. Cheng, J. Zhang, Z. Chen, Evid. Based Complement. Alternat. Med. 2017 (2017) 5389272.

    140. [140]

      X.Q. Sun, S. Chen, L.F. Wang, Z.W. Chen, J. Pharm. Pharmacol. 70 (2018) 1713-1722. doi: 10.1111/jphp.13016

    141. [141]

      Y.T. Tseng, W.J. Lin, W.H. Chang, Y.C. Lo, Phytother. Res. 33 (2019) 690-701. doi: 10.1002/ptr.6259

    142. [142]

      Z. Wei, Y. Xue, Y. Xue, et al., J. Pharmacol. Sci. 147 (2021) 72-80. doi: 10.1016/j.jphs.2021.05.006

    143. [143]

      J. Yu, C. Zhu, J. Yin, et al., Drug Des. Devel. Ther. 14 (2020) 361-370. doi: 10.2147/dddt.s234262

    144. [144]

      Y. Wang, Z. Shou, H. Fan, et al., Biosci. Rep. 39 (2019) BSR20182297. doi: 10.1042/BSR20182297

    145. [145]

      Z. Zhang, L. Nong, M. Chen, et al., Acta Biochim. Biophys. Sin. 52 (2020) 1007-1015. doi: 10.1093/abbs/gmaa075

    146. [146]

      Y. Zhao, J. Xu, Inflammation 43 (2020) 1476-1487. doi: 10.1007/s10753-020-01225-w

    147. [147]

      J. Chen, W. Shi, Y. Xu, et al., Drug Des. Devel. Ther. 82 (2021) 553-561. doi: 10.1002/ddr.21773

    148. [148]

      X. Wang, D. Hou, W. Dai, et al., Mol. Inf. 35 (2016) 262-267. doi: 10.1002/minf.201501022

    149. [149]

      Z. Gu, T. Yan, F. Yan, J. Mol. Recognit. 33 (2020) e2835. doi: 10.1002/jmr.2835

    150. [150]

      S. Zhao, J. Li, F. Wang, et al., Chin. Chem. Lett. 31 (2020) 1147-1152. doi: 10.1109/ispec50848.2020.9351076

    151. [151]

      Q. Zhu, M. Saeed, R. Song, et al., Chin. Chem. Lett. 31 (2020) 1051-1059. doi: 10.1016/j.cclet.2019.12.002

    152. [152]

      H. Zeng, Y. Qi, Z. Zhang, et al., Chin. Chem. Lett. 32 (2021) 1857-1868. doi: 10.1016/j.cclet.2021.01.014

    153. [153]

      K. Thapa Magar, G.F. Boafo, X. Li, et al., Chin. Chem. Lett. 33 (2022) 587-596. doi: 10.1016/j.cclet.2021.08.020

    154. [154]

      P. Wang, Y. Wang, P. Li, et al., Chin. Chem. Lett. 34 (2023) 107691. doi: 10.1016/j.cclet.2022.07.034

    155. [155]

      N.D. Sheybani, H. Yang, Chin. Chem. Lett. 28 (2017) 1817-1821. doi: 10.1016/j.cclet.2017.07.022

    156. [156]

      H. Zhu, J. Yu, J. Ye, et al., Chin. Chem. Lett. 34 (2023) 107648. doi: 10.1016/j.cclet.2022.06.071

    157. [157]

      R. Liu, C. Luo, Z. Pang, et al., Chin. Chem. Lett. 34 (2023) 107518. doi: 10.1016/j.cclet.2022.05.032

    158. [158]

      Y. Gao, Y. Gao, Y. Ding, et al., Chin. Chem. Lett. 32 (2021) 949-953. doi: 10.1016/j.cclet.2020.08.010

    159. [159]

      S. Okimoto, S. Kuroda, H. Tashiro, et al., Hepatol. Res. 49 (2019) 663-675. doi: 10.1111/hepr.13317

    160. [160]

      S. Klein, F. Frohn, F. Magdaleno, et al., Sci. Rep. 9 (2019) 2256. doi: 10.1038/s41598-019-38678-5

    161. [161]

      H. Yin, L. Chen, B. Yang, et al., Org. Biomol. Chem. 15 (2017) 4336-4343. doi: 10.1039/C7OB00547D

    162. [162]

      T.A. Al-Hilal, M.A. Hossain, A. Alobaida, et al., J. Control. Release 334 (2021) 237-247. doi: 10.1016/j.jconrel.2021.04.030

    163. [163]

      K. Matsuo, S. Thayyil, M. Kawaguchi, et al., Chem. Commun. 57 (2021) 12500-12503. doi: 10.1039/d1cc04905d

  • Figure 1  ROCK inhibitors based on isoquinoline and pyridine/pyrrolopyridine derivatives.

    Figure 2  Molecular hybridization to design ROCK inhibitors.

    Figure 3  Indazole, pyridine, and pyrimidine based ROCK inhibitors.

    Figure 4  Pyridine and pyrrolopyridine based ROCK inhibitors.

    Figure 5  Crystal structure of 21 complexed with the kinase domain of ROCK-Ⅱ (PDB ID: 4L6Q). The crystal structure of 21 (green) bound to the active site of ROCK-Ⅱ. Hydrogen bonds are shown as dashed lines.

    Figure 6  Structure–activity diagram of 2230.

    Figure 7  Crystal structure of compound 25 complexed with the kinase domain of ROCK-Ⅰ (PDB ID: 5KKT). The crystal structure of compound 25 (green) bound to the active site of ROCK-Ⅰ. Hydrogen bonds are shown as dashed lines.

    Figure 8  Pyrazole, thiophene substituted pyrrolopyridine, and benzothiophene substituted pyridine based ROCK inhibitors.

    Figure 9  6- and 7-substituted isoquinoline, 4-chromanone, benzimidazole, purine derivatives as ROCK inhibitors.

    Figure 10  Piperazine, boron, and ligooxazole and pyridyl oligomeric derivatives as ROCK inhibitors.

    Figure 11  Bis-functional ROCK inhibitors.

    Figure 12  Structures of ROCK inhibitors discovered from drug repurposing and natural compounds.

    Table 1.  Overview of ROCK inhibitors in clinical trials.

    下载: 导出CSV

    Table 2.  Peptides reported by Gu and Abbasgholizadeh [24,149].

    下载: 导出CSV
  • 加载中
计量
  • PDF下载量:  10
  • 文章访问数:  676
  • HTML全文浏览量:  32
文章相关
  • 发布日期:  2023-12-15
  • 收稿日期:  2023-01-05
  • 接受日期:  2023-03-12
  • 修回日期:  2023-03-08
  • 网络出版日期:  2023-03-15
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

/

返回文章