Voltar
Revisão Acesso aberto Revisado por pares
BIBTEX RIS

Role of Protein Phosphatase Type 1 in Contractile Functions: Myosin Phosphatase

2004; Elsevier BV; Volume: 279; Issue: 36 Linguagem: Inglês

10.1074/jbc.r400018200

ISSN

1083-351X

Autores

David J. Hartshorne, Masaaki Ito, Ferenc Erdődi,

Tópico(s)

Muscle Physiology and Disorders

Resumo

Protein phosphatase type 1 (PP1) 1The abbreviations used are: PP1, protein phosphatase type 1; cGK, cGMP-dependent kinase; CPI-17, C-kinase-dependent phosphatase inhibitor of 17 kDa; ILK, integrin-linked kinase; LZ, leucine zipper motifs; MLCK, myosin light chain kinase; MP, myosin phosphatase; MYPT, myosin phosphatase target subunit; PKA, cAMP-dependent kinase; PKC, protein kinase C; P-LC20, phosphorylated 20-kDa myosin light chain; P-myosin, phosphorylated myosin; PP1c, catalytic subunit of PP1; ROK, Rho-associated kinase; TIMAP, transforming growth factor-β-inhibited membrane-associated protein; ATPγS, adenosine 5′-3-O-(thio)triphosphate. is involved in a wide range of cell activities (1Cohen P.T.W. J. Cell Sci. 2002; 115: 241-256Crossref PubMed Google Scholar), and even within the more restricted theme of contractile activity in muscle several processes may be considered. Important areas include regulation of ion channels (2Herzig S. Neumann J. Physiol. Rev. 2002; 80: 173-210Crossref Scopus (239) Google Scholar), effect of phospholamban on Ca2+ uptake by the SR (3Asahi M. Nakayama H. Tada M. Otsu K. Trends Cardiovasc. Med. 2003; 13: 152-157Crossref PubMed Scopus (51) Google Scholar), and phosphorylation-dephosphorylation of myosin II. Phosphorylation of myosin light chains (located in the head-neck junction of the myosin molecule) by the Ca2+-calmodulin-dependent MLCK in all muscle types is established (4Kamm K.E. Stull J.T. J. Biol. Chem. 2001; 276: 4527-4530Abstract Full Text Full Text PDF PubMed Scopus (472) Google Scholar). Discovery of MLCK spurred numerous reports on the phosphatases involved. In smooth muscle, phosphorylation of myosin II increases actin-activated ATPase activity and is required for contraction (4Kamm K.E. Stull J.T. J. Biol. Chem. 2001; 276: 4527-4530Abstract Full Text Full Text PDF PubMed Scopus (472) Google Scholar). Much of the earlier work focused on smooth muscle myosin phosphatase (MP). An initial controversy was the type of catalytic subunit involved, i.e. PP1c, PP2Ac, etc. In smooth muscle the majority of MP activity is due to PP1c (5Erd[umlaut]odi F. Ito M. Hartshorne D.J. Barany M. Biochemistry of Smooth Muscle Contraction. Academic Press, San Diego, CA1996: 131-142Crossref Google Scholar), and this finding was extended to include skeletal and cardiac muscle. Three genes encode PP1c: α, γ, and δ (also called β). Five PP1c isoforms are expressed, where α1/α2 and γ1/γ2 are generated by alternative splicing (1Cohen P.T.W. J. Cell Sci. 2002; 115: 241-256Crossref PubMed Google Scholar). To accommodate specific functions of the limited number of PP1c isoforms with the multiple roles of PP1c the concept of target subunits was developed. Over 50 potential target subunits have been identified (1Cohen P.T.W. J. Cell Sci. 2002; 115: 241-256Crossref PubMed Google Scholar) that in complex with PP1c may designate specific substrates, regulate activity, and direct distinct cell localization. This review describes one of the PP1 holoenzymes, namely the myosin phosphatase of muscle. In smooth muscle the role of MP is to dephosphorylate Ser-19 and to a lesser extent Thr-18 in P-LC20. (The PKC site, Thr-9, can also be dephosphorylated by MP.) The current model for smooth muscle MP is based on the gizzard holoenzyme (6Alessi D. MacDougall L.K. Sola M.M. Ikebe M. Cohen P. Eur. J. Biochem. 1992; 210: 1023-1035Crossref PubMed Scopus (331) Google Scholar, 7Hartshorne D.J. Ito M. Erdödi F. J. Muscle Res. Cell Motil. 1998; 19: 325-341Crossref PubMed Scopus (345) Google Scholar). The holoenzyme is a trimer consisting of: a catalytic subunit, PP1cδ; a target subunit of ∼110 kDa; and a smaller subunit of ∼20 kDa (M20). A scheme of the MP holoenzyme is shown in Fig. 1. The function of M20 is not known, and the critical properties of MP can be ascribed to the large subunit, i.e. binding of PP1c and the substrate, P-myosin. Thus, it is termed myosin phosphatase target subunit, MYPT1. (Other terms include M110, myosin binding subunit (MBS), and M130/M133). Initially MYPT1 was cloned from chicken gizzard (M130/M133 (8Shimizu H. Ito M. Miyahara M. Ichikawa K. Okubo S. Konishi T. Naka M. Tanaka T. Hirano K. Hartshorne D.J. Nakano T. J. Biol. Chem. 1994; 269: 30407-30411Abstract Full Text PDF PubMed Google Scholar)) and also from rat aorta (rat3 isoform (9Chen Y.H. Chen M.X. Alessi D.R. Campbell D.G. Shanahan C. Cohen P. Cohen P.T.W. FEBS Lett. 1994; 356: 51-55Crossref PubMed Scopus (128) Google Scholar)). From the initial reports the basic features of the MYPT1 molecule were established. The molecule is hydrophilic, and no extensive hydrophobic patches are found. Plans of human and chicken MYPT1 are shown in Fig. 1. A striking feature of all MYPT isoforms is the presence of N-terminal ankyrin repeats. In MYPT1 each of the 7 or 8 repeats contains ∼33 residues with 20 residues conserved. The 171–197 region is less homologous but shows similarity to ankyrin repeats and was so considered in chicken M130/M133 (8Shimizu H. Ito M. Miyahara M. Ichikawa K. Okubo S. Konishi T. Naka M. Tanaka T. Hirano K. Hartshorne D.J. Nakano T. J. Biol. Chem. 1994; 269: 30407-30411Abstract Full Text PDF PubMed Google Scholar). It is suggested that the conserved sequence for ankyrin repeats is structure-based in that it forms a β-hairpin-helix-loop-helix (β2α2) structure (10Sedgewick S.G. Smerdon S.J. Trends Biochem. Sci. 1999; 24: 311-316Abstract Full Text Full Text PDF PubMed Scopus (664) Google Scholar). Based on several solved structures it is known that both the β-hairpins and the surface of the ankyrin groove (helical bundle) can be involved in binding to target proteins (10Sedgewick S.G. Smerdon S.J. Trends Biochem. Sci. 1999; 24: 311-316Abstract Full Text Full Text PDF PubMed Scopus (664) Google Scholar). Many proteins interact with ankyrin repeats (11Bennett V. Baines A.J. Physiol. Rev. 2001; 81: 1353-1392Crossref PubMed Scopus (795) Google Scholar), and thus the proposed role for these repeats in MYPT1 is to act as an interactive protein platform. Flanking the N-terminal edge of the first ankyrin repeat is the PP1c-binding motif, the "RVXF" (1Cohen P.T.W. J. Cell Sci. 2002; 115: 241-256Crossref PubMed Google Scholar) motif (residues 35–38). Some variation is allowed and a general consensus is (R/K)X1(V/I)X2(F/W), where X1 may be absent or be residues other than large hydrophobes and X2 is any residue except large hydrophobes, phosphoserine and probably aspartic acid (1Cohen P.T.W. J. Cell Sci. 2002; 115: 241-256Crossref PubMed Google Scholar). Residues flanking the motif, N-terminal basic residues, and C-terminal acidic residue(s) may contribute to binding (12Zhao S. Lee E.Y.C. J. Biol. Chem. 1997; 272: 28368-28372Abstract Full Text Full Text PDF PubMed Scopus (128) Google Scholar). In human MYPT1 the pertinent sequence is 30KRQKTKVKFDD. The RVXF motif is present in many target proteins and even occurs in proteins unlikely to bind PP1c (1Cohen P.T.W. J. Cell Sci. 2002; 115: 241-256Crossref PubMed Google Scholar). This motif interacts with PP1c in a hydrophobic groove involving residues Ile-169, Leu-243, Phe-257, Leu-289, Cys-291, and Phe-293. Important points are that the interaction site for RVXF on PP1c is within the invariant region for all PP1c isoforms (but not conserved in PP2A and PP2B) and that the site is distinct from the catalytic site. Peptides containing the RVXF motif may displace target subunits, but binding of the RVXF motif to PP1c does not directly influence activity (1Cohen P.T.W. J. Cell Sci. 2002; 115: 241-256Crossref PubMed Google Scholar, 13Bollen M. Trends Biochem. Sci. 2001; 26: 426-431Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar). In MYPT1 the RVXF motif (KVKF) acts as the primary interaction site (an anchoring site) for PP1cδ, but other interactions are involved (14Tóth A. Kiss E. Herberg F.W. Gergely P. Hartshorne D.J. Erdödi F. Eur. J. Biochem. 2000; 267: 1687-1697Crossref PubMed Scopus (66) Google Scholar). These include residues 1–22, the ankyrin repeats (possibly repeats 5–8, sequence 167–295 (7Hartshorne D.J. Ito M. Erdödi F. J. Muscle Res. Cell Motil. 1998; 19: 325-341Crossref PubMed Scopus (345) Google Scholar)), and a site within the sequence 304–511. Only the interaction with KVKF has high affinity, but the other secondary interactions are important in that they may modify PP1c properties. For example, interaction of PP1c or P-myosin with the N-terminal segment of MYPT1 could activate PP1c. These multiple and hierarchical interactions form a combinatorial control of PP1c (13Bollen M. Trends Biochem. Sci. 2001; 26: 426-431Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar), and considering these, it is likely that PP1c is pivoted via the RVXF motif and clasped by the N-terminal sequence of MYPT1 and the ankyrin repeats. Other structural features of MYPT1, including some phosphorylation sites, are shown in Fig. 1. A potentially important feature of MYPT1 as a target subunit is that it is a platform for multiple interactions. The binding of myosin to MYPT1 is an important but controversial point. One view is that P-myosin or P-LC20 binds to the ankyrin repeats, possibly repeats 6–8 (15Hirano K. Phan B.C. Hartshorne D.J. J. Biol. Chem. 1997; 272: 3683-3688Abstract Full Text Full Text PDF PubMed Scopus (84) Google Scholar). Interaction of the phosphorylated substrate with the catalytic site of PP1c is expected with an added contribution to binding by the ankyrin repeats. Dephosphorylated substrate binds less effectively, and in the presence of ATP only P-myosin or P-LC20 is bound (7Hartshorne D.J. Ito M. Erdödi F. J. Muscle Res. Cell Motil. 1998; 19: 325-341Crossref PubMed Scopus (345) Google Scholar). The opposing view is that dephosphorylated myosin binds to the C-terminal sequence of MYPT1 (excluding the C-terminal 72 residues (16Johnson D. Cohen P. Chen M.X. Chen Y.H. Cohen P.T.W. Eur. J. Biochem. 1997; 244: 931-939Crossref PubMed Scopus (83) Google Scholar) in the chicken isoforms). Binding to both the N-terminal and C-terminal regions of MYPT1 is feasible if P-LC20 (in the myosin S2 region) binds to the N-terminal sites and the rod portion of myosin binds to the C-terminal sites. It was suggested that phosphorylation of Thr-850 (chicken M133) by ROK reduced binding of MYPT1 to myosin (17Velasco G. Armstrong C. Morrice N. Frame S. Cohen P. FEBS Lett. 2002; 527: 101-104Crossref PubMed Scopus (181) Google Scholar). Adducin (α, β, γ (18Amano M. Fukata Y. Kaibuchi K. Exp. Cell Res. 2000; 261: 44-51Crossref PubMed Scopus (454) Google Scholar)) and Tau and MAP2 (19Amano M. Kaneko T. Maeda A. Nakayama M. Ito M. Yamauchi T. Goto H. Fukata Y. Oshiro N. Shinohara A. Iwamatsu A. Kaibuchi K. J. Neurochem. 2003; 87: 780-790Crossref PubMed Scopus (89) Google Scholar) also bind to the ankyrin repeats and are phosphorylated and dephosphorylated by ROK and MP, respectively. Multiple interactions with the ankyrin repeats are expected (11Bennett V. Baines A.J. Physiol. Rev. 2001; 81: 1353-1392Crossref PubMed Scopus (795) Google Scholar), but what is surprising is that the C-terminal half of MYPT1 also interacts with many molecules (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar). M20 binds to the C-terminal region of MYPT1, residues 934–1006 of human MYPT1, in an interaction not involving LZ repeats (15Hirano K. Phan B.C. Hartshorne D.J. J. Biol. Chem. 1997; 272: 3683-3688Abstract Full Text Full Text PDF PubMed Scopus (84) Google Scholar, 16Johnson D. Cohen P. Chen M.X. Chen Y.H. Cohen P.T.W. Eur. J. Biochem. 1997; 244: 931-939Crossref PubMed Scopus (83) Google Scholar). M20 also binds to myosin but not PP1c (16Johnson D. Cohen P. Chen M.X. Chen Y.H. Cohen P.T.W. Eur. J. Biochem. 1997; 244: 931-939Crossref PubMed Scopus (83) Google Scholar). GTP-RhoA (but not inactive GDP-RhoA) binds to the C terminus and could represent an alternative docking site for GTP-RhoA in addition to the plasmalemma. Acidic phospholipids target residues 667–1004 of chicken M133 (does not contain LZ sequences) and inhibit PP1c activity (7Hartshorne D.J. Ito M. Erdödi F. J. Muscle Res. Cell Motil. 1998; 19: 325-341Crossref PubMed Scopus (345) Google Scholar). This binding/inhibition of MP activity is reversed on phosphorylation of MYPT1 by PKA. Moesin also binds to MYPT1 via the C-terminal part (18Amano M. Fukata Y. Kaibuchi K. Exp. Cell Res. 2000; 261: 44-51Crossref PubMed Scopus (454) Google Scholar). Other interactions involve the LZ motifs in some isoforms of MYPT1 and include: 1) interaction of the LZ motifs of cGMP-dependent kinase (cGKIα) and MYPT1 (21Surks H.K. Mochizuki N. Kasai Y. Georgescu S.P. Tang K.M. Ito M. Lincoln T.M. Mendelsohn M.E. Science. 1999; 286: 1583-1587Crossref PubMed Scopus (444) Google Scholar); 2) binding of the PDZ2 domain of interleukin-16 precursor proteins to the C-terminal 30 residues of MYPT (22Bannert N. Vollhardt K. Asomuddinov B. Haag M. König H. Norley S. Kurth R. J. Biol. Chem. 2003; 278: 42190-42199Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar); and 3) interaction of a coiled-coil domain in RhoA-interacting protein (expressed in vascular smooth muscle) with the LZ motifs of MYPT1 (23Surks H.K. Richards C.T. Mendelsohn M.E. J. Biol. Chem. 2003; 278: 51484-51493Abstract Full Text Full Text PDF PubMed Scopus (74) Google Scholar). The many interactions involving MYPT1 suggest several substrates and a possible targeting function in other macromolecular complexes and thus a much broader role in cell function than only dephosphorylation of P-myosin. This complexity may reflect the varied cell localizations observed with MYPT1 on filaments and membranes (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar). In differentiated striated muscle cells MYPT may have a more restricted role. Several isoforms of MYPT1 2The Human Gene Nomenclature Committee has assigned the following: PPP1R12 A, B, or C for MYPT1, MYPT2, and MBS85, respectively; PPP1R16 A or B for MYPT3 and TIMAP, respectively; and PPP1R14A for CPI-17. Gene location is listed on www.gene.ucl.ac.uk/cgi-bin/nomenclature/searchgenes.pl using PPP1R* as query. are generated by cassette-type alternative splicing of single gene pre-mRNAs and involve the presence or absence of central inserts and the C-terminal LZ motifs. The M130/133 chicken gizzard MYPT1 isoforms differ by a 123-nucleotide insert (residues 512–552 of M133 (8Shimizu H. Ito M. Miyahara M. Ichikawa K. Okubo S. Konishi T. Naka M. Tanaka T. Hirano K. Hartshorne D.J. Nakano T. J. Biol. Chem. 1994; 269: 30407-30411Abstract Full Text PDF PubMed Google Scholar)) arising from a single exon (24Dirksen W.P. Vladic F. Fisher S.A. Am. J. Physiol. 2000; 278: C589-C600Crossref PubMed Google Scholar). In rat the situation is more complex, and 5 central insert isoforms are generated (24Dirksen W.P. Vladic F. Fisher S.A. Am. J. Physiol. 2000; 278: C589-C600Crossref PubMed Google Scholar) resulting from differing expression of 2 exons. Control of this central region (in chicken) is exerted by a cis-enhancer complex close to the alternative exon 5′-splice site (25Dirksen W.P. Mohamed S.A. Fisher S.A. J. Biol. Chem. 2003; 278: 9722-9732Abstract Full Text Full Text PDF PubMed Scopus (27) Google Scholar). In addition, isoforms are generated by cassette-type alternative splicing of a 3′-exon (31 nucleotides). Skipping of this exon codes for the LZ-positive MYPT1, and inclusion of the exon introduces a premature stop codon and codes for the LZ-negative MYPT1 (26Khatri J.J. Joyce K.M. Brozovich F.V. Fisher S.A. J. Biol. Chem. 2001; 276: 37250-37257Abstract Full Text Full Text PDF PubMed Scopus (111) Google Scholar). Thus many isoforms of MYPT1 can be generated and to some extent are expressed in a tissue-specific fashion (24Dirksen W.P. Vladic F. Fisher S.A. Am. J. Physiol. 2000; 278: C589-C600Crossref PubMed Google Scholar, 26Khatri J.J. Joyce K.M. Brozovich F.V. Fisher S.A. J. Biol. Chem. 2001; 276: 37250-37257Abstract Full Text Full Text PDF PubMed Scopus (111) Google Scholar). Recently several molecules similar to MYPT1 have been described (Fig. 2). Each has the N-terminal ankyrin repeats and the RVXF motif. The C-terminal half is more variable. As originally cloned (27Fujioka M. Takahashi N. Odai H. Araki S. Ichikawa K. Feng J. Nakamura M. Kaibuchi K. Hartshorne D.J. Nakano T. Ito M. Genomics. 1998; 49: 59-68Crossref PubMed Scopus (69) Google Scholar) MYPT2 (human chromosome 1q32) contained 982 residues (110 kDa), and subsequently a second isoform (MYPT2B) was recognized from the genomic organization of the MYPT2 gene (28Arimura T. Suematsu N. Zhou Y-B. Nishimura J. Satoh S. Takeshita A. Kanaide H. Kimura A. J. Biol. Chem. 2001; 276: 6073-6082Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). These arise from alternative splicing of exon 24, each isoform reflecting the inclusion of either exon 24 (MYPT2B) or exon 25 (MYPT2A). Both contain C-terminal LZ motifs. A comparison of MYPT1 and MYPT2A is shown in Fig. 2. Overall identity is 52% but several areas are more conserved (Fig. 2). The N-terminal sequences (1–57 of MYPT2 and 1–38 of MYPT1) are distinct (36% identity) although both contain the PP1c-binding motif. MYPT1 may be considered a housekeeping gene and is expressed in most tissues but is higher in smooth muscle ([MYPT1] in rabbit portal vein is ∼1.2 μm). 3T. M. Butler, and M. J. Siegman, personal communication. MYPT2 is more restricted with both A and B isoforms found in heart, skeletal muscle, and brain and predominantly MYPT2A in other tissues (28Arimura T. Suematsu N. Zhou Y-B. Nishimura J. Satoh S. Takeshita A. Kanaide H. Kimura A. J. Biol. Chem. 2001; 276: 6073-6082Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). Another member of the MYPT family is MBS85 (29Tan I. Ng C.H. Lim L. Leung T. J. Biol. Chem. 2001; 276: 21209-21216Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar). This is widely distributed with higher levels in the heart. MBS85 binds specifically to the PP1cδ isoform and also binds and dephosphorylates P-myosin (or P-LC20). Phosphorylation of Thr-560 by myotonic dystrophy kinase-related Cdc42-binding kinase or ROK is required for its binding to PP1cδ and inhibition of phosphatase activity (29Tan I. Ng C.H. Lim L. Leung T. J. Biol. Chem. 2001; 276: 21209-21216Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar). MYPT3 has some similarity to MYPT1/2 (Fig. 2) but only in its N-terminal half (30Skinner J.A. Saltiel A.R. Biochem. J. 2001; 356: 257-267Crossref PubMed Scopus (37) Google Scholar). Other features are shown in Fig. 2. A putative regulatory site is not present. MYPT3 inhibited the activity of PP1cγ with phosphorylase a and P-myosin in contrast to activation of PP1cδ by MYPT1 with P-myosin. MYPT3 is widely distributed (mouse) and appears in high amounts in heart, brain, and kidney. TIMAP is similar to MYPT3 (∼45% identity) and contains the "marker" N-terminal structure, conserved nuclear localization signal and a C-terminal CAAX box (31Cao W. Mattagajasingh S.N. Xu H. Kim K. Fierlbeck W. Deng J. Lowenstein C.J. Ballerman B.J. Am. J. Physiol. 2002; 283: C327-C337Crossref PubMed Scopus (41) Google Scholar). It shows high expression in endothelial and hematopoietic cells. Analysis of genomic data bases indicates that several gene products homologous to MYPT3 and TIMAP exist (30Skinner J.A. Saltiel A.R. Biochem. J. 2001; 356: 257-267Crossref PubMed Scopus (37) Google Scholar, 31Cao W. Mattagajasingh S.N. Xu H. Kim K. Fierlbeck W. Deng J. Lowenstein C.J. Ballerman B.J. Am. J. Physiol. 2002; 283: C327-C337Crossref PubMed Scopus (41) Google Scholar). Each family member binds PP1c via the RVXF motif, and the binding of PP1c and substrate may be augmented by the ankyrin repeats. The C-terminal region is adapted for a given function, e.g. regulation or binding of other ligands or attachment to membranes (via the prenylated C terminus). However, the function(s) of most MYPT family members is not known. Even with MYPT1 and MYPT2 it is possible that substrates other than P-myosin are implicated. M20 was cloned from chicken gizzard (9Chen Y.H. Chen M.X. Alessi D.R. Campbell D.G. Shanahan C. Cohen P. Cohen P.T.W. FEBS Lett. 1994; 356: 51-55Crossref PubMed Scopus (128) Google Scholar), and two splicing variants of 161 (M18) and 186 (M21) residues were detected (32Mabuchi K. Gong B.J. Langsetmo K. Ito M. Nakano T. Tao T. Biochim. Biophys. Acta. 1999; 1434: 296-303Crossref PubMed Scopus (8) Google Scholar). Only M21 contains C-terminal LZ motifs. Expression of each isoform is tissue-specific (32Mabuchi K. Gong B.J. Langsetmo K. Ito M. Nakano T. Tao T. Biochim. Biophys. Acta. 1999; 1434: 296-303Crossref PubMed Scopus (8) Google Scholar). Part of M20 arises from the MYPT2 gene (33Moorhead G. Johnson D. Morrice N. Cohen P. FEBS Lett. 1998; 438: 141-144Crossref PubMed Scopus (48) Google Scholar), and two heart-specific isoforms of M20 arise from the MYPT2A and MYPT2B genes involving exons 14–25 (28Arimura T. Suematsu N. Zhou Y-B. Nishimura J. Satoh S. Takeshita A. Kanaide H. Kimura A. J. Biol. Chem. 2001; 276: 6073-6082Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar). M20 was not detected in either brain or skeletal muscle (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar). The function of M20 is not established. Suggested roles (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar) include: modification of Ca2+ sensitivity in renal artery and cardiac myocytes; binding to the myosin dimer; and a role in microtubule dynamics. The binding of M20 to MYPT1 does not affect phosphatase activity (7Hartshorne D.J. Ito M. Erdödi F. J. Muscle Res. Cell Motil. 1998; 19: 325-341Crossref PubMed Scopus (345) Google Scholar). Early studies assumed that MP activity was unregulated, but more recently inhibition and activation have been documented. At fixed suboptimal [Ca2+] inhibition of MP would increase [P-myosin] and activation would decrease [P-myosin], resulting in Ca2+ sensitization and Ca2+ desensitization, respectively. More data are available for Ca2+ sensitization. Inhibition of MP was found to be linked to agonist stimulation via numerous trimeric G protein-coupled receptors (list given in Ref. 34Somlyo A.P. Somlyo A.V. Physiol. Rev. 2003; 83: 1325-1358Crossref PubMed Scopus (1696) Google Scholar), and RhoA and thus Rho-associated kinase, ROK, are important downstream links (7Hartshorne D.J. Ito M. Erdödi F. J. Muscle Res. Cell Motil. 1998; 19: 325-341Crossref PubMed Scopus (345) Google Scholar, 34Somlyo A.P. Somlyo A.V. Physiol. Rev. 2003; 83: 1325-1358Crossref PubMed Scopus (1696) Google Scholar). There are two isoforms of ROK (ROKα/ROCKII and ROKβ/ROCKI (18Amano M. Fukata Y. Kaibuchi K. Exp. Cell Res. 2000; 261: 44-51Crossref PubMed Scopus (454) Google Scholar) with 32% identity). ROKα was isolated from chicken gizzard smooth muscle (35Feng J. Ito M. Kureishi Y. Ichikawa K. Amano M. Isaka N. Okawa K. Iwamatsu A. Kaibuchi K. Hartshorne D.J. Nakano T. J. Biol. Chem. 1999; 274: 3744-3752Abstract Full Text Full Text PDF PubMed Scopus (225) Google Scholar). There are several mechanisms proposed for inhibition of MP (7Hartshorne D.J. Ito M. Erdödi F. J. Muscle Res. Cell Motil. 1998; 19: 325-341Crossref PubMed Scopus (345) Google Scholar, 20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar), i.e. phosphorylation of MYPT1; phosphorylation of an inhibitory protein, CPI-17; dissociation of the holoenzyme; and translocation and subsequent dissociation of the holoenzyme. The first two are cited more frequently. Initially it was found that incubation of permeabilized portal vein with ATPγS induced Ca2+ sensitization, inhibition of MP, and thiophosphorylation of MYPT1 (see Refs. 7Hartshorne D.J. Ito M. Erdödi F. J. Muscle Res. Cell Motil. 1998; 19: 325-341Crossref PubMed Scopus (345) Google Scholar and 20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar). Subsequently, it was shown that phosphorylation of MYPT1 at Thr-695 (Thr-696 in human MYPT1) by an endogenous kinase inhibited gizzard MP activity. ROK was the first known kinase shown to phosphorylate MYPT1 (7Hartshorne D.J. Ito M. Erdödi F. J. Muscle Res. Cell Motil. 1998; 19: 325-341Crossref PubMed Scopus (345) Google Scholar, 20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar), and two major sites, Thr-696 and Thr-853 (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar, 36Kawano Y. Fukota Y. Oshiro N. Amano M. Nakamura T. Ito M. Matsumura F. Inagaki M. Kaibuchi K. J. Cell Biol. 1999; 147: 1023-1037Crossref PubMed Scopus (477) Google Scholar), and several minor sites (36Kawano Y. Fukota Y. Oshiro N. Amano M. Nakamura T. Ito M. Matsumura F. Inagaki M. Kaibuchi K. J. Cell Biol. 1999; 147: 1023-1037Crossref PubMed Scopus (477) Google Scholar) were identified. Several kinases are now known to phosphorylate MYPT1 (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar), many at the inhibitory site. The endogenous kinase was identified and termed MYPT1 kinase (previously ZIP-like kinase (37Borman M.A. MacDonald J.A. Murányi A. Hartshorne D.J. Haystead T.A.J. J. Biol. Chem. 2002; 277: 23441-23446Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar)). Some of these kinases (ROK, MYPT1 kinase, and ILK) can also directly phosphorylate LC20 at Ser-19 and have been implicated in Ca2+-independent contractile events, particularly in non-muscle cells. The relevance of multiple kinases that phosphorylate the inhibitory site is not known but may represent the convergence of different pathways at MYPT1, e.g. Rac-1/myotonic dystrophy protein kinase, Rac or Cdc42/p21-activated kinase, and Ras/Raf-1. The molecular basis for inhibition of MP as a result of phosphorylation is not understood. The inhibition with several substrates is due largely to a decrease in Vmax. Dissociation of MP following phosphorylation is unlikely (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar). For intramolecular inhibition (autoinhibition) folding of the molecule is expected to allow interaction of Thr-696 with the N-terminally located PP1c. However, it should be pointed out that sequences around both Thr-696 and Thr-853 are similar to that around Ser-19 of LC20 (from +3 to –6). Intermolecular inhibition also could occur via formation of head-to-tail dimers of MYPT1 (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar). CPI-17 inhibits PP1. It is composed of 147 residues (17 kDa) and expressed in smooth muscle and brain (38Eto M. Senba S. Morita F. Yazawa M. FEBS Lett. 1997; 410: 356-360Crossref PubMed Scopus (226) Google Scholar). In human aorta a splicing variant is expressed that lacks sequence 68–94, encoded by exon 2 of the 4 exons (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar). Unlike other inhibitors of PP1, e.g. inhibitors 1 and 2, CPI-17 inhibits both the catalytic subunit and PP1 holoenzymes. Phosphorylation at Thr-38 enhances inhibitory potency about 1000-fold, and the first kinase implicated was PKC (α and δ isoforms). Subsequently several kinases, notably ROK, were found to phosphorylate Thr-38. Residues 35–120 are required for recognition of MP and Tyr-41 is necessary to reduce dephosphorylation of Thr-38 by MP (39Hayashi Y. Senba S. Yazawa M. Brautigan D.L. Eto M. J. Biol. Chem. 2001; 276: 39858-39863Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar). The solution NMR structure (40Ohki S-Y. Eto M. Kariya E. Hayano T. Hayashi Y. Yazawa M. Brautigan D. Kainosho M. J. Mol. Biol. 2001; 314: 839-849Crossref PubMed Scopus (36) Google Scholar) indicates that phosphorylation of Thr-38 induces a conformational change that promotes specific recognition of MP by CPI-17. Histamine stimulation of smooth muscle fibers caused phosphorylation of Thr-38 that was reduced by both ROK and PKC inhibitors, suggesting input from two signaling pathways (41Kitazawa T. Eto M. Woodsome T.P. Brautigan D.L. J. Biol. Chem. 2000; 275: 9897-9900Abstract Full Text Full Text PDF PubMed Scopus (290) Google Scholar). Whether CPI-17 or MYPT1 is dominant in regulation of MP is controversial and perhaps depends on the tissue/cell involved. General considerations are that MLCK and MP are lower in tonic than phasic smooth muscle and that CPI-17 is higher in vascular compared with visceral muscle (42Woodsome T.P. Eto M. Everett A. Brautigan D.L. Kitazawa T. J. Physiol. (Lond.). 2001; 535: 553-564Crossref Scopus (213) Google Scholar). Several reports have suggested that CPI-17 plays a dominant role in Ca2+ sensitization in various smooth muscles (42Woodsome T.P. Eto M. Everett A. Brautigan D.L. Kitazawa T. J. Physiol. (Lond.). 2001; 535: 553-564Crossref Scopus (213) Google Scholar, 43Kitazawa T. Eto M. Woodsome T.P. Khalequzzaman M. J. Physiol. (Lond.). 2003; 546: 879-889Crossref Scopus (204) Google Scholar, 44Niiro N. Koga Y. Ikebe M. Biochem. J. 2003; 369: 117-128Crossref PubMed Scopus (109) Google Scholar). Also, in NO-mediated vasorelaxation a transient decrease in CPI-17 phosphorylation (Thr-38) correlated to a transient activation of MP activity (45Etter E.F. Eto M. Wardle R.L. Brautigan D.L. Murphy R.A. J. Biol. Chem. 2001; 276: 34681-34685Abstract Full Text Full Text PDF PubMed Scopus (118) Google Scholar). Two other phosphorylation-dependent inhibitors of PP1 are the phosphatase holoenzyme inhibitors (PHI-1 and PHI-2 (46Eto M. Karginov A. Brautigan D.L. Biochemistry. 1999; 38: 16952-16957Crossref PubMed Scopus (90) Google Scholar)) and kinase-enhanced protein phosphatase type 1 inhibitor (KEPI (47Liu Q-R. Zhang P-W. Zhen Q. Walther D. Wang X-B. Uhl G.R. J. Biol. Chem. 2002; 277: 13312-13320Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar)). PHI-2 and KEPI are high in cardiac muscle. CPI-17 has sequence similarity to PHI-1 and also to the C-terminal domain of Lim kinase-2 (48Dubois T. Howell S. Zemlickova E. Learmonth M. Cronshaw A. Aitken A. Biochem. Biophys. Res. Commun. 2003; 302: 186-192Crossref PubMed Scopus (19) Google Scholar). Lipid messengers should be considered in inhibition of MP and Ca2+ sensitization. Arachidonic acid was proposed to dissociate the MP subunits (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar, 34Somlyo A.P. Somlyo A.V. Physiol. Rev. 2003; 83: 1325-1358Crossref PubMed Scopus (1696) Google Scholar) with inhibition reflecting the reduced activity of isolated PP1c. An alternative explanation is that arachidonic acid activates ROK, independent of RhoA, and this phosphorylates either MYPT1 or CPI-17 (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar). Sphingosine 1-phosphate, sphingo-sylphosphorylcholine, and lysophosphatidic acid also activate the ROK pathway (34Somlyo A.P. Somlyo A.V. Physiol. Rev. 2003; 83: 1325-1358Crossref PubMed Scopus (1696) Google Scholar). Because phosphorylation of myosin II is critical for smooth muscle function defects in the balance of phosphorylation may lead to disorders of smooth muscle. Several implicate inhibition of MP via the RhoA/ROK pathway, and the ROK inhibitors, Y-27632 and HA-1077, have been widely used. Examples include hypertension, coronary spasm, cerebral vasospasm, vasoplastic angina, and bronchial asthma (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar, 34Somlyo A.P. Somlyo A.V. Physiol. Rev. 2003; 83: 1325-1358Crossref PubMed Scopus (1696) Google Scholar). Less is known about activation of MP activity. Usually this is associated with relaxation of smooth muscle caused by increases in cAMP and cGMP. An exception is the activation of MP by mitosis-specific phosphorylation of MYPT1 at Thr-435 and/or Ser-432 (in human isoforms) thought to reflect an increase in binding to P-myosin (49Totsukawa G. Yamakita Y. Yamashiro S. Hosoya H. Hartshorne D.J. Matsumura F. J. Cell Biol. 1999; 144: 735-744Crossref PubMed Scopus (54) Google Scholar). The mechanism underlying activation of MP (initially shown in permeabilized smooth muscle preparations (34Somlyo A.P. Somlyo A.V. Physiol. Rev. 2003; 83: 1325-1358Crossref PubMed Scopus (1696) Google Scholar)) by PKA and cGK is not clear, but in general, this opposes the RhoA/ROK pathway. Direct phosphorylation of MYPT1 by either kinase does not activate MP (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar), and thus it is assumed that additional components are involved. One suggestion is that PKA and cGK inactivate the RhoA/ROK pathway. Under in vitro conditions RhoA is phosphorylated by PKA and cGK at Ser-188 and was reported to inactivate RhoA (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar, 34Somlyo A.P. Somlyo A.V. Physiol. Rev. 2003; 83: 1325-1358Crossref PubMed Scopus (1696) Google Scholar). A problem with this proposal is that phosphorylation of RhoA under in vivo conditions was difficult to detect although phosphorylation of non-prenylated RhoA was detected in cells (50Ellerbroek S.M. Wennerberg K. Burridge K. J. Biol. Chem. 2003; 278: 19023-19031Abstract Full Text Full Text PDF PubMed Scopus (260) Google Scholar). The mechanism of RhoA inactivation may be because of enhanced interaction of phosphorylated RhoA with GDI (50Ellerbroek S.M. Wennerberg K. Burridge K. J. Biol. Chem. 2003; 278: 19023-19031Abstract Full Text Full Text PDF PubMed Scopus (260) Google Scholar) and/or phosphorylation of the switch I region of Gα13 and reduced downstream signaling by Gα13 (51Manganello J.M. Huang J-S. Kozawa T. Voyno-Yasenetskaya T.A. LeBreton G.C. J. Biol. Chem. 2003; 278: 124-130Abstract Full Text Full Text PDF PubMed Scopus (61) Google Scholar). Any decrease in RhoA activation would not effect a net activation of MP but would reduce the extent of inhibition. Activation of MP by cGKIα is suggested to require interaction of the LZs of MYPT1 and cGKIα (21Surks H.K. Mochizuki N. Kasai Y. Georgescu S.P. Tang K.M. Ito M. Lincoln T.M. Mendelsohn M.E. Science. 1999; 286: 1583-1587Crossref PubMed Scopus (444) Google Scholar), and only those smooth muscles expressing the LZ-positive MYPT1 isoforms show cGMP-dependent Ca2+ desensitization (26Khatri J.J. Joyce K.M. Brozovich F.V. Fisher S.A. J. Biol. Chem. 2001; 276: 37250-37257Abstract Full Text Full Text PDF PubMed Scopus (111) Google Scholar). Telokin, an independent protein derived from the smooth muscle MLCK gene, contains the C-terminal domain of MLCK (20Ito M. Nakano T. Erdödi F. Hartshorne D.J. Mol. Cell. Biochem. 2004; 259: 197-209Crossref PubMed Scopus (382) Google Scholar) and also has been implicated in cyclic nucleotide-dependent relaxation of smooth muscle (34Somlyo A.P. Somlyo A.V. Physiol. Rev. 2003; 83: 1325-1358Crossref PubMed Scopus (1696) Google Scholar). It is expressed at relatively high levels (70–80 μm, i.e. about the same concentration as myosin II) only in phasic smooth muscle. Under in vivo conditions telokin is phosphorylated by PKA and cGK at Ser-13 (34Somlyo A.P. Somlyo A.V. Physiol. Rev. 2003; 83: 1325-1358Crossref PubMed Scopus (1696) Google Scholar). Telokin also is phosphorylated at Ser-19, a mitogen-activated protein kinase site, but the in vivo role for this is not known (34Somlyo A.P. Somlyo A.V. Physiol. Rev. 2003; 83: 1325-1358Crossref PubMed Scopus (1696) Google Scholar). The mechanism by which telokin or phosphorylated telokin activates MP is not established. In striated muscle the target for MP is Ser-15 on the P-regulatory light chain. The idea that MP in skeletal muscle requires a different regulatory subunit than that in smooth muscle was proposed by Cohen and co-workers (6Alessi D. MacDougall L.K. Sola M.M. Ikebe M. Cohen P. Eur. J. Biochem. 1992; 210: 1023-1035Crossref PubMed Scopus (331) Google Scholar). This was identified as MYPT2 based on tissue distribution (27Fujioka M. Takahashi N. Odai H. Araki S. Ichikawa K. Feng J. Nakamura M. Kaibuchi K. Hartshorne D.J. Nakano T. Ito M. Genomics. 1998; 49: 59-68Crossref PubMed Scopus (69) Google Scholar), matching of fragments isolated from skeletal muscle (33Moorhead G. Johnson D. Morrice N. Cohen P. FEBS Lett. 1998; 438: 141-144Crossref PubMed Scopus (48) Google Scholar, 52Damer C.K. Partridge J. Pearson W.R. Haystead T.A.J. J. Biol. Chem. 1998; 273: 24396-24405Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar), and the detection of the full-length MYPT2 in rat muscle (53Wu Y. Erdödi F. Murányi A. Nullmeyer K.D. Lynch R.M. Hartshorne D.J. J. Muscle Res. Cell Motil. 2003; 24: 499-511Crossref PubMed Scopus (23) Google Scholar). In cardiac muscle, MYPT2 also is the major targeting subunit of MP (27Fujioka M. Takahashi N. Odai H. Araki S. Ichikawa K. Feng J. Nakamura M. Kaibuchi K. Hartshorne D.J. Nakano T. Ito M. Genomics. 1998; 49: 59-68Crossref PubMed Scopus (69) Google Scholar). In the myoblast cell line, C2C12, a transition from MYPT1 to MYPT2 occurs as the nondifferentiated cells develop a sarcomeric phenotype (53Wu Y. Erdödi F. Murányi A. Nullmeyer K.D. Lynch R.M. Hartshorne D.J. J. Muscle Res. Cell Motil. 2003; 24: 499-511Crossref PubMed Scopus (23) Google Scholar). The function of myosin phosphorylation in striated muscle is not as pronounced as in smooth muscle. In general, phosphorylation of striated muscle myosin increases force at submaximal [Ca2+], i.e. an increase in Ca2+ sensitivity, and is most pronounced in fast-twitch muscle. It is proposed that phosphorylation of the regulatory light chains moves the myosin head closer to the thin filament and increases the transition from non-force- to force-generating states (54Sweeney H.L. Bowman B.F. Stull J.T. Am. J. Physiol. 1993; 264: C1085-C1095Crossref PubMed Google Scholar). This may be a general mechanism in all muscle types. Myosin phosphorylation (in fast-twitch muscle) is slower than the twitch contraction but considerably faster than calculated phosphatase rates (approximately 1 s–1 compared with 0.007 s–1 (54Sweeney H.L. Bowman B.F. Stull J.T. Am. J. Physiol. 1993; 264: C1085-C1095Crossref PubMed Google Scholar)). In cardiac muscle the rates of myosin phosphorylation and dephosphorylation are lower (1–4% of fast-twitch muscle (54Sweeney H.L. Bowman B.F. Stull J.T. Am. J. Physiol. 1993; 264: C1085-C1095Crossref PubMed Google Scholar)), and effects due to myosin phosphorylation thus are more difficult to detect. However, in rat hearts a positive correlation was found between myosin phosphorylation and left ventricular pressure development (54Sweeney H.L. Bowman B.F. Stull J.T. Am. J. Physiol. 1993; 264: C1085-C1095Crossref PubMed Google Scholar). Transgenic mice expressing a non-phosphorylatable regulatory light chain showed loss of Ca2+ sensitivity and longer term structural changes (55Sanbe A. Fewell J.G. Gulick J. Osinska H. Lorenz J. Hall D.G. Murray L.A. Kimball T.R. Witt S.A. Robbins J. J. Biol. Chem. 1999; 274: 21085-21094Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar). A more recent possibility is that the level of myosin phosphorylation may be important in differentiation or sarcomere organization. In C2C12 cells partial inhibition of non-muscle MLCK caused a decrease in the numbers of larger myotubules (53Wu Y. Erdödi F. Murányi A. Nullmeyer K.D. Lynch R.M. Hartshorne D.J. J. Muscle Res. Cell Motil. 2003; 24: 499-511Crossref PubMed Scopus (23) Google Scholar), and in cardiac myocytes MLCK mediates agonist-induced sarcomere organization during the early hypertrophic response (56Aoki H. Sadoshima J. Izumo S. Nat. Med. 2000; 6: 183-188Crossref PubMed Scopus (124) Google Scholar). In view of the importance of maintaining a certain level of myosin phosphorylation it is likely that striated muscle MP is regulated, although there are no data to indicate this. Although knowledge on MP is progressing, notably acceptance of the basic molecular structure, i.e. PP1cδ plus MYPT1 (in smooth muscle) and MYPT2 (striated muscle), several questions are unanswered. Details of MP regulation in smooth muscle are important to establish, both for a molecular appreciation of contractile functions and to facilitate pharmaceutical intervention for many disorders of smooth muscle function. In striated muscle there are no data on regulation of MP. Several components (proteins/lipids) bind to MYPT1, and the possibility is raised that MP is not dedicated to P-myosin but has alternative substrates and functions. Another intriguing area for future research is to establish the roles of the other members of the MYPT family.

Referência(s)