doi: 10.1161/CIRCRESAHA.107.165779
© 2007 American Heart Association, Inc.
Editorials |
cGMP-Dependent Protein Kinase I and Smooth Muscle Relaxation
A Tale of Two Isoforms
From the Tufts-New England Medical Center, Boston, Mass.
Correspondence to Howard K. Surks, Tufts-New England Med Center, Molecular Cardiology Research Institute, 750 Washington St, Box 80, Boston, MA 02111. E-mail hsurks{at}tufts-nemc.org
See related article, pages 1096–1103
Key Words: myosin phosphatase • RhoA • smooth muscle • GMP-dependent protein kinase I • inositol 1,4,5 triphosphate receptor-associated cGMP kinase substrate • myosin phosphatase-rho interacting protein • regulator of G protein signaling 2
The maintenance of vascular tone is central to the regulation of blood pressure and tissue perfusion and plays a role in the pathogenesis of hypertension and atherosclerosis. Vascular tone is determined by the balance of vasodilator and vasoconstrictor stimuli. After several decades of research, the NO/cGMP/cGMP-dependent protein kinase (cGK) pathway is now recognized as an important mediator of vasodilation. However, the mechanisms by which cGK causes smooth muscle relaxation continue to be an important question.
Smooth muscle contraction and relaxation are tightly coupled to the phosphorylation and dephosphorylation, respectively, of the regulatory myosin light chain.1 Myosin light chain phosphorylation state is determined by the relative activities of myosin light chain kinase (MLCK) and myosin light chain phosphatase (MLCP). MLCK phosphorylates MLC leading to contraction,2 and MLCP dephosphorylates MLC, leading to relaxation3 (Figure). Both MLCK and MLCP activities are highly regulated. MLCK activity is activated by the binding of calcium/calmodulin and thus is the primary mechanism linking intracellular calcium concentration to smooth muscle contractility.4 MLCP activity is regulated by both vasodilator and vasoconstrictor stimuli, and is therefore responsible for much of the calcium-independent regulation of contractility (reviewed in5).
|
cGK mediates vasorelaxation by both calcium-dependent and calcium-independent pathways. Initial experiments found that cGMP inhibited calcium elevations in freshly isolated aortic smooth muscle cells. However, in passaged aortic smooth muscle cells in which cGK expression is downregulated, cGMP failed to inhibit calcium elevations, supporting that cGK mediated smooth muscle relaxation, in part, results from the inhibition of intracellular calcium mobilization.6 In permeabilized smooth muscle strips in which transmembrane calcium concentrations can be clamped without loss of intracellular proteins, cGMP induces smooth muscle relaxation without a change in intracellular calcium, suggesting that cGK could mediate calcium-independent relaxation via activation of MLCP.7,8 Perhaps the most definitive evidence to date comes from the complete cGKI knockout mouse model. These mice have multiple cardiovascular, gastrointestinal, hematopoietic, and neurological abnormalities, resulting in early postnatal mortality and making them difficult to study. Nevertheless, the aortas from these mice failed to relax after stimulation of the NO/cGMP/cGK pathway, supporting a critical role for cGKI in vascular relaxation.9
The cGK family includes two distinct genes, prkg1 and prkg2, that encode cGKI and cGKII, respectively.10 cGKI is the isoform expressed in smooth muscle tissues.11 The amino terminus of cGKI is encoded by two alternatively spliced exons, resulting in the two isoforms cGKI
and cGKIß, that differ only in the amino-terminal leucine zipper domains.10 Several studies have shown that both cGKI isoforms are expressed in smooth muscle, raising the question whether one or both isoforms are responsible for smooth muscle relaxation, which is addressed by Weber et al in this issue of Circulation Research.12
In recent years, molecular studies have contributed to the hypothesis that both cGKI isoforms contribute to smooth muscle relaxation and are targeted to specific substrates via their amino-terminal leucine zipper domains. In a screen for cGKI interacting proteins, our laboratory found that cGK1 interacts directly with the regulatory myosin binding subunit (MYPT1) of MLCP. This interaction targets cGKI to MLCP and is critical for cGKI-mediated activation of MLCP.13 Although MYPT1 was originally thought to target MLCP to contractile fibers, we found that targeting of MLCP is more complex and involves a molecular scaffold, myosin phosphatase-rho interacting protein, that anchors MLCP to actin and colocalizes RhoA to regulate MLCP.14–16 Tang et al demonstrated that cGKI inhibits thrombin receptor-mediated calcium release through a cGKI-regulator of G protein signaling (RGS)2 interaction. This leucine zipper-mediated interaction targets cGKI to phosphorylate and activate RGS2, which in turn increases the GTPase activity of Gq, terminating thrombin receptor signaling.17 cGKIß was found to regulate inositol 1,4,5 triphosphate (IP3)-mediated calcium release by binding to and phosphorylating the inositol 1,4,5-triphosphate receptor-associated cGMP kinase substrate (IRAG). The cGKIß-IRAG interaction leads to phosphorylation of IRAG and inhibition of calcium release via the inositol triphosphate type I receptor.18 These isoform-specific leucine zipper-mediated interactions have formed the basis of a targeting hypothesis in which both cGKI isoforms have distinct functional roles in smooth muscle relaxation by virtue of their specific targets (Figure).
Previous approaches have used cell culture models to test the targeting hypothesis. In one study, GFP-tagged cGKI or cGKIß were transfected into cGKI-null smooth muscle cells and tested for inhibition of calcium release. Only the cGKI-containing cells exhibited calcium regulation by cGMP.19 A potential limitation in this study is the possibility that the GFP moiety interfered with cGKI targeting. In a recent study by Christensen and Mendelsohn, the role of PKGI isoforms in thrombin receptor-mediated calcium mobilization was studied in both CHO cells stably expressing either cGKI isoform and human aortic smooth muscle cells expressing primarily cGKI or cGKIß. In CHO cells, cGKI had a significantly greater calcium-lowering effect, whereas in cultured human aorta smooth muscle cells, only cGKI lowered calcium in response to cGMP.20 Thus neither of these cell culture-based studies that manipulated the expression of cGKI isoforms confirmed the expected isoform-specific functions based on cGKI targeting.
In the current study, Weber et al took the interesting approach of creating mouse lines that express either cGKI or cGKIß on a cGKI null background. Expression of the cGKI genes was driven by the endogenous SM22 promoter, resulting in smooth muscle–specific expression. The cGKI isoforms were expressed at levels and activity 1.5- to 2.0-fold greater than control mice. The transgenic rescue mice had a life expectancy greater than the cGKI null mice, but less than the wild-type control mice. Interestingly, all of the tested smooth muscle functions in the cGKI transgenics were equivalent to those in the wild-type mice. These included intestinal transit of barium, jejunal, and aortic smooth muscle relaxation and inhibition of norepinephrine-induced calcium transients by cGMP. Although surprising, these data could still be explained by the known interactions between cGKI and cGKIß with RGS2 and IRAG, respectively.
Weber et al performed a thorough examination of blood pressure in the null, wild-type, and the cGKI and cGKIß rescue mice. The basal blood pressure was not different between wild-type and rescue mice, but was elevated as expected in the cGKI-null mice. Moreover, the hypotensive effect of nitrovasodilators, carbachol, and the ROCK inhibitor Y27632 were all preserved in the cGKI rescue mice. The latter manipulation is particularly interesting because cGKI opposes the inhibitory effect of RhoA/ROCK on MLCP activity and inhibits RhoA directly by phosphorylation at Ser188.21 One would therefore expect that the cGKI rescue mice would have less RhoA/ROCK activity and therefore less hypotension from Y27632.
The elegant approach by Weber et al avoids many of the pitfalls of the earlier studies of cGKI isoform-specific functions in the vasculature, including reliance on cell culture models and use of epitope-tagged cGKI isoforms. How then can we explain the apparently equivalent physiologic effect of cGKI and cGKIß rescue? The authors provide two hypotheses. First, it is possible that the specificity for each isoform for their respective targets is less pronounced in vivo. This is possible because most of the experiments that characterized the cGKI isoform specific targets were performed in cell culture systems and with purified proteins. If there were more overlap between cGKI isoforms and their target interactions, then expression of individual cGKI isoforms might exhibit subtle, if any, differences from wild-type mice. Second, each cGKI isoform alone is sufficient to maintain circulatory physiology based on its known interactions. For example, cGKI rescue can mediate cGMP-mediated vasodilation because it can lower calcium via its interaction with RGS2, and activate MLCP via its interaction with MYPT1. It is more difficult to reconcile how cGKIß rescue can fully reconstitute cGMP-mediated vasodilation without regulating MLCP.
There are several additional possibilities that may explain the apparent functional equivalence of cGKI and cGKIß to rescue vascular function. The cGKI targets discussed here have well-described isoform-specific interactions, yet there are additional cGKI targets that do not bind in an isoform specific manner, whose role in vascular physiology is less clear, and there are likely more targets that are undiscovered.22,23 These cGKI targets may allow functional overlap between the cGKI isoforms. Furthermore, although Weber et al observed similar expression of the cGKI isoform targets, IRAG, MYPT1, and RhoA, other critical proteins within these signaling pathways may be upregulated. Moreover, posttranslational modification (eg, phosphorylation) of these isoform-specific targets or splice variation may also contribute to altered activity of these signaling pathways without apparent differences in their protein expression levels. A third possibility represents a limitation to any transgenic approach when used to explore pathways regulated by differential targeting. Even modestly increased levels of protein overexpression may be adequate to obscure the specificity of protein targeting, particularly if the protein in question is present in excess of the targeting protein.
Future experiments with this model may provide important insights into cGKI isoform specificity. Biochemical experiments comparing wild-type, null, and rescue mouse vascular tissue for isoform-specific binding to respective targets and regulation of these pathways would be revealing. For example, is there cGMP-mediated activation of MLCP and phosphorylation/inhibition of RhoA in both cGKI isoform rescue mice? Can cGMP induce cGKI interaction with and phosphorylation of IRAG in both rescue mice?
Ultimately, additional genetic models in mice may be the best approach. Among these, inducible, tissue-specific deletion of each cGKI isoform will provide the opportunity to dissect their roles in vascular function while avoiding any concerns about loss of specificity in cGKI overexpression mice. Finally, specific knock-in mutations to disrupt cGKI isoform targeting will also be informative. Our laboratory has created a mouse in which the leucines and isoleucines in the cGKI leucine zipper domain have been mutated to alanine, disrupting its interaction with MLCP without affecting kinase activity. Preliminary studies show that these mice are hypertensive and have abnormalities of vascular relaxation, supporting the isoform specific targeting hypothesis and the importance of cGKI in the regulation of vascular tone.24
cGKI is critical for the maintenance of vascular tone by mediating vasodilation in response to nitrovasodilators. The two cGKI isoforms interact with specific targets in smooth muscle via their amino-terminal leucine zipper domains, and there is debate about their relative roles in smooth muscle relaxation. This carefully performed, elegant study by Weber et al adds an interesting wrinkle to the targeting hypothesis by showing that both cGKI and cGKIß can rescue vascular relaxation and intestinal motility in cGKI null mice. Future biochemical studies with this model, as well as additional genetic targeting approaches to delete or mutate specific cGKI isoforms, will provide a wealth of information about these pathways in the near future.
 |
Acknowledgments |
|---|
The author thanks Richard Patten and Akiko Hata for critical review of the manuscript.
Sources of Funding
The author is supported by National Institutes of Health grants HL074069 and HL077378.
Disclosures
None.
|
Footnotes |
|---|
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
|
References |
|---|
 Top References |
|---|
1. Hartshorne DJ. Biochemistry of the contractile process in smooth muscle. In: Physiology of the Gastrointestinal Tract. Johnson DR, ed. 1987. Raven, New York.
2. Kamm KE, Stull JT. The function of myosin and myosin light chain kinase phosphorylation in smooth muscle. Ann Rev Pharmacol Toxicol. 1985; 25: 593.[CrossRef][Medline] [Order article via Infotrieve]
3. Alessi D, MacDougall LK, Sola MM, Ikebe M, Cohen P. The control of protein phosphatase-1 by targetting subunits. Eur J Biochem. 1992; 210: 1023–1035.[Medline] [Order article via Infotrieve]
4. Taylor DA, Stull JT. Calcium Dependence of Myosin Light Chain Phosphorylation in Smooth Muscle Cells. J Biol Chem. 1988; 263: 14456–14462.
5. Hartshorne DJ, Ito M, Erdodi F. Role of protein phosphatase type 1 in contractile functions: myosin phosphatase. J Biol Chem. 2004; 279 (36): 37211–37214.
6. Cornwell TL, Lincoln TM. Regulation of intracellular Ca2+ levels in cultured vascular smooth muscle cells: reduction of Ca2+ by atriopeptin and 8-bromo-cGMP is mediated by cGMP-dependent protein kinase. J Biol Chem. 1989; 264: 1146–1155.
7. Lee MR, Li L, Kitazawa T. Cyclic GMP causes Ca+2 desensitization in vascular smooth muscle by activating the myosin light chain phosphatase. J Biol Chem. 1997; 272: 5063–5068.
8. Wu X, Somlyo AV, Somlyo AP. Cyclic GMP-dependent stimulation reverses G-protein-coupled inhibition of smooth muscle myosin light chain phosphatase. Biochem Biophys Res Com. 1996; 220: 658–663.[CrossRef][Medline] [Order article via Infotrieve]
9. Pfeifer A, Klatt P, Massberg S, Ny L, Sausbier M, Hirneil C, Wang G-X, Korth M, Aszodi A, Andersson K-E, Krombach F, Mayerhofer A, Ruth P, Fassler R, Hofmann F. Defective smooth muscle regulation in cGMP kinase 1-deficient mice. EMBO J. 1998; 17: 3045–3051.[CrossRef][Medline] [Order article via Infotrieve]
10. Orstavik S, Natarajan V, Tasken K, Jahnsen T, Sandberg M. Characterization of the human gene encoding the type 1alpha and type 1beta cGMP-dependent protein kinase (PRKG1). Genomics. 1997; 42: 311–318.[CrossRef][Medline] [Order article via Infotrieve]
11. Tamura N, Itoh H, Ogawa Y, Nakagawa O, Harada M, Chun T-H, Suga S, Yoshimasa T, Nakao K. cDNA cloning and gene expression of human type 1 alpha cGMP-dependent protein kinase. Hypertension. 1996; 27: 552–557.
12. Weber S, Bernhard D, Lukowski R, Weinmeister P, Worner R, Wegener JW, Valtcheva N, Feil S, Schlossmann J, Hofmann F, Feil R. Rescue of cGMP kinase I knockout mice by smooth muscle-specific expression of either isozyme. Circ Res. 2007; 101: 1096–1103.
13. Surks HK, Mochizuki N, Kasai Y, Georgescu SP, Tang KM, Ito M, Lincoln TM, Mendelsohn ME. Regulation of myosin phosphatase by a specific interaction with cGMP-dependent protein kinase 1 alpha. Science. 1999; 286: 1583–1587.
14. Surks HK, Richards CT, Mendelsohn ME. Myosin phosphatase-Rho interacting protein: a new member of the myosin phosphatase complex that directly binds RhoA. J Biol Chem. 2003; 278: 51484–51493.
15. Surks HK, Riddick N, Ohtani K. M-RIP targets myosin phosphatase to stress fibers to regulate myosin light chain phosphorylation in vascular smooth muscle cells. J Biol Chem. 2005; 280: 42543–42551.
16. Riddick N, Ohtani K, Surks HK. Targeting by myosin phosphatase-Rho interacting protein mediates RhoA/ROCK regulation of myosin phosphatase. J Cell Biochem. In press.
17. Tang KM, Wang G-R, Lu P, Karas RH, Aronovitz M, Heximer SP, Kaltenbronn KM, Blumer KJ, Siderovski DP, Zhu Y, Mendelsohn ME. Regulator of G-protein signaling 2 mediates vascular smooth muscle relaxation and blood pressure. Nat Med. 2003; 9 (12): 1506–1512.[CrossRef][Medline] [Order article via Infotrieve]
18. Schlossmann J, Ammendola A, Ashman K, Zong X, Huber A, Neubauer G, Wang G-X, Allescher H-D, Korth M, Wilm M, Hofmann F, Ruth P. Regulation of intracellular calcium by a signalling complex of IRAG, IP3 receptor and cGMP kinase 1B. Nature. 2000; 404: 197–201.[CrossRef][Medline] [Order article via Infotrieve]
19. Feil R, Gappa N, Rutz M, Schlossmann J, Rose CR, Konnerth A, Brummer S, Kuhbandner S, Hofmann F. Functional reconstitution of vascular smooth muscle cells with cGMP-dependent protein kinase I isoforms. Circ Res. 2002; 90: 1080–1086.
20. Christensen EN, Mendelsohn ME. Cyclic GMP-dependent protein kinase I alpha Inhibits Thrombin Receptor-mediated calcium mobilization in vascular smooth muscle cells. J Biol Chem. 2006; 281: 8409–8416.
21. Sauzeau V, Le Jeune H, Cario-Toumaniantz C, Smolenski A, Lohmann S, Bertoglio J, Chardin P, Pacaud P, Loirand G. Cyclic GMP-dependent protein kinase signaling pathway inhibits RhoA-induced calcium sensitization of contraction in vascular smooth muscle. J Biol Chem. 2000; 275: 21722–21729.
22. Zhang T, Zhuang S, Casteel DE, Looney DJ, Boss GR, Pilz RB. A cysteine-rich LIM-only protein mediates regulation of smooth muscle-specific gene expression by cGMP-dependent protein kinase. J Biol Chem. 2007; Papers In Press Sept. 18.
23. Yuasa K, Michibata H, Omori K, Yanaka N. A novel interaction of cGMP-dependent protein kinase 1 with troponin T. J Biol Chem. 1999; 274: 37429–37434.
24. Mendelsohn ME. Regulation of vascular smooth muscle cell phenotype, vascular function and blood pressure by cyclic GMP-dependent protein Kinase 1 alpha. Keystone Symposia, Molecular Biology of the Vasculature, April 5, 2006.
Related Article:
Circ. Res. 2007 101: 1096-1103.
This article has been cited by other articles:
 |
|
 |
|
  A. K. Sharma, G.-P. Zhou, J. Kupferman, H. K. Surks, E. N. Christensen, J. J. Chou, M. E. Mendelsohn, and A. C. Rigby Probing the Interaction between the Coiled Coil Leucine Zipper of cGMP-dependent Protein Kinase I{alpha} and the C Terminus of the Myosin Binding Subunit of the Myosin Light Chain Phosphatase J. Biol. Chem., November 21, 2008; 283(47): 32860 - 32869. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Sugiura, H. Nakanishi, and J. D. Roberts Jr Proteolytic Processing of cGMP-Dependent Protein Kinase I Mediates Nuclear cGMP Signaling in Vascular Smooth Muscle Cells Circ. Res., July 3, 2008; 103(1): 53 - 60. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||