Rho Kinase in Vascular Smooth Muscle (original) (raw)

Agonists of the G-protein-coupled receptors (GPCRs) such as phenylephrine, angiotensin II (AngII), and endothelin-1 (ET-1) stimulate phospholipase C and generate IP3 and DAG. IP3 activates receptors located on the sarcoplasmic reticulum, leading to Ca2+ release and a transient increase in [Ca2+]i. Agonists also stimulate Ca2+ entry through various types of Ca2+ channels leading to maintained increase in [Ca2+]i. Ca2+ then binds to CAM, and the resulting Ca2+–CAM complex activates MLCK, which phosphorylates MLC, promoting the interaction of myosin with actin and subsequent VSM contraction. This mechanism of VSM contraction is entirely Ca2+-dependent.

GPCR agonists, particularly those coupling to Ga12/13 proteins, can also activate the small G-protein RhoA. In its active GTP-bound form, RhoA activates Rho kinase (ROCK), which then phosphorylates and inactivates MLC phosphatase. This in turn increases the proportion of MLC in its phosphorylated form (MLC-P) and thereby promotes further VSM contraction. The ROCK-mediated enhancement of smooth muscle contraction occurs in the absence of significant changes in [Ca2+]i and is therefore considered a Ca2+ sensitization mechanism [158–161].

7.1. ROCK STRUCTURE AND ISOFORMS

ROCKs are serine/threonine kinases with a molecular mass of ≈160 kDa. Two ROCK isoforms encoded by two different genes have been identified: ROCK-1 (ROCK I, or ROKβ) and ROCK-2 (ROCK II, or ROKα) [162–164]. Human ROCK-1 and ROCK-2 genes are located on chromosome 18 (18q11.1) and chromosome 2 (2p24), respectively. ROCK structure comprises a kinase domain located at the amino terminus of the protein, a coiled-coil region containing the Rho-binding domain, and a pleckstrin-homology domain with a cysteine-rich domain (Figure 7.1). ROCK-1 and ROCK-2 are highly homologous, with an overall amino acid sequence identity of 65%. The identity in the Rho-binding domain is 58% and approximately 92% in the kinase domain [165].

Figure 7.1

Molecular structure of ROCKs. ROCK amino acid sequence comprises a kinase domain located at the amino terminus, a coiled-coil region containing the Rho-binding domain (Rho BD), and a pleckstrin-homology domain (PHD) with a cysteine-rich domain (CRD). (more...)

7.2. TISSUE EXPRESSION OF ROCK

Both ROCK-1 and ROCK-2 are ubiquitously expressed. ROCK-2 mRNA is highly expressed in the brain and skeletal muscle [165,166]. Both ROCK-1 and ROCK-2 are expressed in VSM and the heart [167].

Some studies suggest that the expression of ROCK is regulated. Both ROCK-1 and ROCK-2 mRNAs and proteins are up-regulated by AngII via angiotensin type 1 receptor stimulation and by interleukin-1β [168]. The up-regulation of ROCK may occur through PKC- and nuclear factor κB-dependent pathways. Also, chronic administration of AngII in mice is associated with up- regulation of ROCK in coronary artery [168].

7.3. SUBCELLULAR DISTRIBUTION OF ROCK

Similar to other GTPases, Rho cycles between an inactive guanosine diphosphate (GDP)-bound form and an active GTP-bound form. When bound to GTP, Rho recognizes and interacts with its downstream targets and initiates a cellular response. The subcellular localization of Rho also determines its activation state. In unstimulated cells, RhoA resides predominantly in the cytosol, bound to GDP. During receptor stimulation, RhoA undergoes translocation to the plasma membrane where GDP–GTP exchange takes place [161]. The translocation of RhoA to the plasma membrane is facilitated by the hydrophobic geranylgeranyl tail that is attached to the C-terminal of RhoA during posttranslational modification of the protein.

ROCKs are also essentially distributed in the cytoplasm but are partially translocated to peripheral membrane during RhoA activation [163,164]. The mechanisms responsible for the subcellular localization of ROCKs are unclear but could involve mechanisms involved in the translocation of other protein kinases such as PKC.

7.4. REGULATION OF ROCK ACTIVITY

The C-terminal of ROCK contains an autoinhibitory region [169], including the pleckstrin- homology and Rho-binding domains (Figure 7.2). Each of these domains binds independently to the N-terminal kinase domain and, in turn, inhibits the enzyme activity [170]. During ROCK activation, the interaction between the C-terminal autoinhibitory region and kinase domain is disrupted. This disruption could occur as a result of binding of RhoA or complete cleavage of the C-terminal, which yields a constitutively active kinase.

Figure 7.2

Mechanisms of ROCK activation. In the inactive form, the C terminus of ROCK is folded over the N-terminal region of the enzyme, forming an autoinhibitory loop. Binding of active GTP-bound RhoA causes unfolding and activation of ROCKs, and thereby exposes (more...)

The kinase domain of ROCKs is localized in the N-terminal region of the protein sequence. Truncated forms of ROCKs lacking the C-terminal portion of the protein, which contains the Rho-binding domain and the pleckstrin-homology domain, are constitutively active, whereas C-terminal portions of ROCKs expressed in cells act as dominant negatives [171]. This led to the suggestion that the C-terminal region of ROCKs is a negative regulatory region, responsible for autoinhibition of the kinase activity in resting cells, probably through interaction with the catalytic domain of ROCKs [172]. Oligomerization (dimerization) also influences the kinase activity of ROCKs and its affinity for ATP [173]. Binding of active GTP-bound form of RhoA to Rho-binding domain stimulates the phosphotransferase activity of ROCK by disrupting the interaction between the catalytic and the inhibitory C-terminal region of the enzyme. However, the stimulatory effect of GTP-RhoA on the enzyme activity of ROCKs is limited to a 1.5- to 2-fold increase [174]. Lipids such as arachidonic acid or sphingosine phosphorylcholine efficiently increase ROCK activity 5- to 6-fold independently of RhoA [174,175]. Arachidonic acid and sphingosine phosphorylcholine may interact with the regulatory region of ROCK, possibly the pleckstrin-homology domain, thus disrupting its inhibitory action on the catalytic activity of ROCK [172]. ROCKs are also activated by cleavage of the inhibitory C-terminal region, which results in the release of a truncated active form of the kinase.

Negative control of the kinase activity of ROCKs has also been described. The small G-protein RhoE binds to the N-terminal region of ROCK-1 (amino acids 1–420) containing the kinase domain [176]. RhoE binding to ROCK-1 inhibits its activity and prevents RhoA binding to Rho-binding domain [176]. Two other small G proteins, Gem and Rad, have been shown to bind and inhibit ROCK function, but their mechanism of action is not clear [177].

7.5. ROCK SUBSTRATES

The consensus sequence of ROCK phosphorylation site is RXXS/T or RXS/T [178–181]. ROCKs seem to require basic amino acids such as Arg (R) close to its phosphorylation site. More than 15 ROCK substrates have been identified. For many ROCK substrates, the functional consequence of ROCK-mediated phosphorylation is related to actin filament formation and organization and cytoskeleton rearrangements [182,183].

A major group of ROCK targets includes the myosin phosphatase target subunit (MYPT-1) [184], CPI-17 [185], the 20-kDa MLC [178], and calponin [186], which are known to modulate smooth muscle cell contraction. MYPT-1 is the major effector of ROCK-mediated Ca2+ sensitization pathway of smooth muscle contraction. Cardiac troponin is another ROCK substrate. Phosphorylation of troponin by ROCK causes reduction in tension generation in cardiac myocytes [187].

Phosphatase and tensin homologue (PTEN) is a newly identified ROCK substrate [188]. PTEN is a phosphatase that dephosphorylates both proteins and phosphoinositide substrates such as the phosphatidylinositol 3-kinase (PI3-kinase)/Akt pathway involved in the regulation of cell growth, protein synthesis, transcriptional regulation, and cell survival. Phosphorylation of PTEN by ROCK stimulates its phosphatase activity, and ROCK inhibitors could reduce ROCK- mediated PTEN phosphorylation and thereby enhances Akt signaling in endothelial cells [189]. Active ROCK also interacts with and phosphorylates the insulin receptor substrate-1 (IRS-1) in VSMCs, leading to inhibition of both insulin-induced IRS-1 tyrosine phosphorylation and PI3- kinase activation [190]. In VSMCs from hypertensive rats, the ROCK/IRS-1 association is increased, and insulin signaling is markedly reduced [190].

Most of ROCK substrates have been identified from in vitro experiments performed after either activation of endogenous ROCK or transfection of one of the two ROCK isoforms. Because the kinase domains of both isoforms are nearly identical, it has been thought that ROCK-1 and ROCK-2 share the same substrates. However, ROCK-1 but not ROCK-2 binds to and phosphorylates RhoE, providing evidence that ROCK-1 and ROCK-2 have different targets [191]. The N-terminal regions, upstream of the kinase domains of ROCKs, are involved in the interaction with the substrates [176] and could play a role in determining substrate specificity of the ROCK isoforms.

7.6. ROCK AND VSM FUNCTION

A large body of evidence suggests important functions of ROCK in VSMCs. The phosphorylation/dephosphorylation of MLC is a major regulatory mechanism of smooth muscle contraction. A rise in [Ca2+]i causes activation of MLCK and consequent phosphorylation of MLC and smooth muscle contraction. However, MLC phosphorylation and VSM contraction can be induced in the absence of significant increases in [Ca2+]i. RhoA-mediated ROCK activation phosphorylates MYPT-1, the regulatory subunit of MLC phosphatase, and inhibits its activity and thereby causes Ca2+ sensitization of the contractile proteins and enhances VSM contraction.

ROCK is not only a major regulator of VSM cell contraction but is also important in controlling cell migration, proliferation, apoptosis/survival, gene transcription, and differentiation.

To investigate the role of ROCK isoforms in vivo, ROCK-1 and ROCK-2 knockout mice have been generated. ROCK-1-deficient mice have open eyelids at birth [192]. ROCK-2-deficient mice develop placental dysfunction leading to intrauterine growth retardation and fetal death [193]. The cardiovascular phenotype of ROCK-1 and ROCK-2 knockout mice has not been clearly analyzed.

7.7. RHO KINASE INHIBITORS

ROCK activation involves RhoA translocation to the plasma membrane, RhoA binding to ROCK, and ATP-dependent phosphorylation of various substrates. Interference with any of these processes could inhibit ROCK activity, phosphorylation of the specific substrate, and inhibition of ROCK-stimulated cellular response. For example, disruption of the prenylation process, by agents such as 3-hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors (statins) or protein prenyltransferase inhibitors, can prevent the membrane translocation and activation of Rho and thereby impair the ability of Rho to activate its downstream target ROCK. Some pharmacological ROCK inhibitors exert their effect by competing with ATP for binding at the kinase domain, or “active site” of the enzyme. Some of the commonly known ROCK inhibitors include Y-27632 and fasudil.

Y-27632 is a synthetic pyridine derivative that inhibits ROCK by competing with ATP for the kinase active site. Y-27632 has an affinity for ROCK that is ~200- and ~2000-fold higher than for the structurally similar kinases PKC and MLCK, respectively. Y-27632 selectively lower elevated blood pressure in animal models of experimental hypertension, supporting a link between VSM ROCK and the development of hypertension [194]. Experiments with Y-27632 have also revealed that ROCK activity might be enhanced in the cerebral circulation under conditions of chronic hypertension [195–198]. These experimental findings made Y-27632 a promising candidate as a potential therapeutic antihypertensive agent. However, despite its utility in numerous experimental studies in vivo [194,199,200], the safety profile of Y-27632 has not yet been verified and its future use as a therapeutic agent is uncertain.

In addition to its ability to potently inhibit ROCKI and ROCKII, Y-27632 could inhibit PKC-related kinase (PRK2) with a similar potency to that of ROCKII [201], although the role of PRK2 in the vasculature is unknown. Also, at a high 10-µM concentration, Y-27632 inhibits PKC-dependent vasoconstriction [202]. Although these observations suggest that ROCK activation may occur downstream of PKC-mediated vasoconstriction as have been described in the coronary circulation [203], the possibility that Y-27632 causes nonselective inhibition of PKC cannot be ruled out.

Fasudil is an isoquinoline derivative that also inhibits ROCK by competing with ATP for the kinase active site. Fasudil has been used to characterize the role of ROCK in vascular function in small-scale clinical studies [204–206]. The good safety profile of fasudil has facilitated its approval for use in Japan for the treatment of cerebral vasospasm following subarachnoid hemorrhage. After oral administration, fasudil is metabolized to the more selective ROCK inhibitor hydroxyfasudil. Hence, fasudil is thought of as a prodrug whose therapeutic effects stem from the relatively potent and selective ROCK inhibitor activity of its metabolite.

It is important to note that the active site of ROCK is similar to that of protein kinase A (PKA), the target of cyclic adenosine monophosphate that mediates vasodilation and inhibits platelet aggregation. This structural similarity may account for the ability of fasudil to inhibit ROCK and PKA with equal potency [207] and raise the possibility of unwanted effects in the clinical use of fasudil as a result of PKA inhibition. Although the metabolite hydroxyfasudil is 15-fold more selective for ROCK than for PKA [207], the possibility that fasudil may exert unwanted vascular effects if its metabolism was compromised should be taken into account. Studies have revealed important sequence differences between ROCK and PKA, which could be used to further improve the selectivity of future ROCK inhibitors [207].