Functional characterization of vertebrate nonmuscle myosin IIB isoforms using Dictyostelium chimeric myosin II

Title Functional characterization of vertebrate nonmuscle myosin IIB isoforms using Dictyostelium chimeric myosin II. Author(s) Takahashi, M.; Takahashi, K.; Hiratsuka, Y.; Uchida, K.; Yamagishi, A.; Uyeda, T Q; Yazawa, M. Citation The Journal of biological chemistry, 276(2): 1034-1040 Issue Date 2001-01-12 Doc URL http://hdl.handle.net/2115/52172 Rights This research was originally published in Journal Name. Author(s). Title. Journal Name. Year; Vol:pp-pp. © the American Society for Biochemistry and Molecular Biology. Type article File Information JBC276-2 1034-1040.pdf


INTRODUCTION
Myosin is a member of a diverse superfamily of mechanochemical proteins (1,2). It produces motor activity together with actin filaments coupled with ATP hydrolysis.
Myosin II, simply myosin hereafter, molecules are the best studied members of the superfamily and are composed of a pair of heavy chains and two pairs of light chains. The amino-terminal half of the heavy chain forms the head region, termed subfragment 1 (S1), containing both ATP and actin binding sites.
It is well known that two proteolytically susceptible areas are present in the head region of skeletal muscle myosin, and the proteolytic cleavage of the myosin heavy chain (MHC) with trypsin produces fragments of 25 kDa, 50 kDa and 20 kDa (see Fig.1

) (3).
Two regions corresponding to the 25/50-kDa and 50/20-kDa junctions were not resolved in the crystal structure of chicken skeletal myosin S1, suggesting that they might exist as flexible surface loops (4). The locations of these two loops are of interest, as the 25/50-kDa loop is near the ATP binding pocket, while the 50/20-kDa loop is near the actin binding site. The amino acid sequence and the length of these two loops vary among different kinds of myosin molecules (5). Based on these observations, Spudich proposed that these regions (named loop 1 and loop 2 for 25/50-kDa and 50/20-kDa junctions, respectively) would play important roles in the tuning of motor activity of myosin (6).
Recently it was demonstrated that the amino acid sequences of these loop regions appeared to be more conserved than those of the rest of the myosin molecule among myosins with kinetically or developmentally similar properties, suggesting their functional roles. (7).
Nonmuscle myosin plays a role in cell motile processes such as cytokinesis, migration, and shape change (for a review, see Ref. 8). To date, two different isoforms of the MHC have been identified in nonmuscle cells of vertebrates (9,10). They were referred to as MHC-A and MHC-B or MHC-IIA and MHC-IIB. These two isoforms are expressed in a tissue-dependent manner. For example, MHC-IIA is abundant in spleen and intestines, while MHC-IIB is abundant in brain and testis (9)(10)(11)(12).
It has been demonstrated that the two loops serve as sites for alternative splicing of mRNA to produce inserted isoforms of MHC-IIB (13)(14)(15). One insert of 10 amino acid residues is located at loop 1 and another insert of 21 amino acid residues is located at loop 2. These inserts are referred to as B1 and B2, respectively. These inserted isoforms are expressed specifically in the brain and the spinal cord (12)(13)(14), and the expression of these inserted isoforms is regulated developmentally in brain (12,14,16,17). Pato et al. characterized the B1-inserted isoform, myosin IIB(B1), using the baculovirus expression system (18). However, to date, there has been no biochemical characterization of myosin IIB(B2) consisting of the B2-inserted MHC. This has mainly been due to the inability to purify sufficient quantities of pure myosin IIB(B2) from brain tissue.
The importance of loop 2 for myosin function was first suggested by proteolytic cleavage studies (19)(20)(21)(22). The actin-activated ATPase activity was decreased by proteolytic cleavage in the loop 2 region (19,21). The proteolytic cleavage of loop 2 was inhibited in the presence of F-actin (19,20) and it reduced the affinity of myosin for Factin (22). The importance of loop 2 was also indicated by molecular genetic studies. It was demonstrated that the substitution of loop 2 of Dictyostelium myosin with that of 5 other myosins caused a change in the actin-activated ATPase to values correlating with the activity of the donor myosins (23). A further detailed study by Murphy and Spudich showed that the V max of actin-activated ATPase activity and the affinity of myosin for actin are both affected by substitutions with loop 2 sequence (24). To examine the role of the B2 insert in the motor activity of the myosin molecule, we adopted a similar strategy.
We expressed chimeric heavy chains of Dictyostelium myosin and S1 in which the loop 2 sequence was replaced with either the non-inserted form or the B2-inserted form of human MHC-IIB (see Fig.1), and assessed the function of these chimeras using an in vitro motility assay system, and by measuring steady state ATPase activities and interaction with F-actin.
Our work suggests that the motor activity of myosin is reduced by the insertion of the B2 sequence, with a reduction of V max and a decrease of the affinity for actin. In addition, we demonstrate that the native loop 2 sequence of Dictyostelium myosin is required for the proper regulation of the actin-activated ATPase activity by phosphorylation of the regulatory light chain.

MATERIALS AND METHODS
Plasmid Construction. All DNA manipulations were done using standard procedures (25). The template for mutagenesis was pMyDAP (26) The plasmids for the expression of chimeric S1 fragments of Dic-B and Dic-B2 were constructed by replacing the BglII-NcoI fragments of the pTIKL_OE_S1-His 6 (unpublished data) with each of the BglII-NcoI fragments as described above.
Manipulation of Dictyostelium Cells. Dictyostelium cells were grown in HL5 medium (28) supplemented with 60 µg each of streptomycin and ampicilin per ml at 23°C . The plasmids carrying either mutant or wild-type MHC gene were transformed into 8 HS1, a MHC null strain (29) by electroporation (30). The plasmids carrying either mutant or wild-type S1 were transformed into HS1 or Ax2 cells. Transformants were selected in a medium supplemented with 12 µg/ml G418 (Roche Diagnostics) and maintained with 8 µg/ml G418 at 23 °C. For the isolation of myosin or S1-His 6  36,000 g for 20 min. The pellet was resuspended with 1.5 volume of lysis buffer, and made 5 mM with respect to ATP, and immediately centrifuged at 265,000 g for 10 min. S1-His 6 proteins were purified by using a Ni 2+ -affinity resin (His-Bind; Novagen) according to the manufacturer's procedure. The eluted S1-His 6 proteins were dialyzed against a buffer containing 50 mM KCl, 20 mM Tris-HCl (pH 7.5) overnight. The sample was finally centrifuged at 14,000 g for 10 min to remove insoluble materials and made 2 mM with respect to DTT.
Rabbit skeletal muscle actin was purified by the method of Pardee and Spudich (33), and its concentration was determined from the absorbance at 280 nm using an absorption coefficient of 1.1 for a 1 mg/ml solution. The concentration of the purified myosins, S1-His 6 proteins, and myosin light chain kinase were measured by the method of Bradford mixtures were centrifuged at 435,000 g for 10 min at 4 immediately after the addition of 2 mM ATP. The resulting supernatant and pellets were run on SDS-10 % polyacrylamide gels (39). The original uncentrifuged samples were also run on SDS-PAGE gels. The concentration of S1 was determined by densitometry of the Coomassie-Brilliant-Bluestained bands of the gels. The expression of the full-length chimeric or wild-type MHCs was confirmed by immunoblot analysis (data not shown). We also confirmed that each transformant expressed MHC at levels comparable with the parental wild type strain, Ax2.

Phenotypic Analysis of Cells Expressing Chimeric Myosins-
In order to assess chimeric myosin function in vivo, we analyzed the ability of the transformants to form fruiting bodies, a process known to depend on myosin functions (40,41). The transformants expressing the wild-type myosin and the chimeric myosin Solution ATPase Analysis of Chimeric S1s and Myosins-We constructed an expression system for Dictyostelium S1 and the chimeric S1s derived from Dic-B and Dic-B2 myosins to analyze the ATPase activity in solution. The high-salt Ca 2+ -ATPase activities of both Dic-B and Dic-B2 chimeric S1 were almost identical, though they were approximately 1.3-fold higher than the value of wild-type S1 (Fig. 4A). The Mg 2+ -ATPase activities of both chimeric S1s showed comparable values, though they were also 1.4-fold higher than that of wild-type S1 (Fig. 4B). These results suggest that the core structure of the motor domain is only slightly affected by replacement of the loop 2 sequence of Dictyostelium wild-type with that of human nonmuscle myosin IIB, and that the insertion of the B2 sequence does not have a further effect.
We then measured the actin-activated ATPase of the chimeric S1s as a function of actin concentration (Fig. 5A). The V max and the apparent K m for actin of the wild-type S1 were 2.37 sec -1 and 147 µM, respectively. The activities of Dic-B S1 were approximately 60 % of those of the wild-type S1 at all actin concentration. Dic-B2 S1 showed much lower activities than those of Dic-B S1. However, in the case of the chimeric S1s, the activities hardly reached the saturation level within the available actin concentrations of our experimental conditions, so that the values for V max and the apparent K m obtained by data fitting were uncertain. Actin Binding Affinity of Chimeric S1s-To examine the affinity of chimeric S1s to Factin, we performed cosedimentation assays in the presence of ATP (Fig. 6). Both Dic-B S1 and Dic-B2 S1 sedimented with F-actin in the absence of ATP (data not shown). In the presence of 2 mM ATP, Dic-B S1 showed a weaker affinity (K d = 21.2 µM) compared to wild-type S1 (6.3 µM). Dic-B2 S1 showed a slightly weaker affinity (28.7 µM) than Dic-B S1. These results indicate that the affinity of S1 for F-actin is affected by the insertion of B2 amino acid residues into loop 2. 16

DISCUSSION
In this report we describe the biochemical characterization of chimeric proteins of Dictyostelium myosin and S1, in which the native loop 2 is replaced with either human nonmuscle MHC-IIB or its B2-inserted isoform. We showed that the high-salt Ca 2+ -ATPase activities and the basal level of Mg 2+ -ATPase activities of chimeric S1s are slightly increased compared to those of the wild-type in agreement with the previous observation (23). It has been demonstrated that the loop 2 of vertebrate smooth muscle myosin is important for optimal regulation mediated by the phosphorylation of the regulatory light chain (45,55). With respect to the regulatory property of the loop 2, it is notable that the B2 sequence is inserted at the inhibitory domain in the loop 2 sequence as proposed by Rovner (55). The activities of vertebrate nonmuscle myosin are also regulated by the phosphorylation of its regulatory light chain (56), and the B2 insertion may modify the regulatory mechanism. This possibility can be examined with use of the baculovirus expression system for nonmuscle MHC-IIB sequence with or without the B2 insert. As the native loop 2 sequence is involved in the optimal regulation in Dictyostelium myosin in the same manner as vertebrate smooth muscle myosin (Figure 7), this could be a general property among the myosins regulated by the phosphorylation of the regulatory light chains.
Another flexible loop, loop 1, which is located at the 25/50-kDa junction was proposed to be important for the ATPase mechanism of the myosin molecule also (6).
Recently it was demonstrated that loop 1 modulates the rate of ADP release from the nucleotide binding pocket of myosin molecule (57)(58)(59). Loop 1 is also a site for tissue specific alternative splicing of mRNA to produce inserted isoforms of nonmuscle MHC-IIB (13)(14)(15)). An isoform containing a 10 amino acid insertion in loop 1, referred to as B1 insert, is specifically expressed in the central nervous system tissues, as is the isoform containing the B2 insert (13,14). Pato  For example, other systems such as vertebrate smooth muscle (60,61), Drosophila flight muscle (62) and scallop adductor muscle (63), adopt this alternative splicing at these loops to produce the diversity of myosin molecule.
The expression of the B1-inserted isoform (14) or the B2-inserted isoform (12,14,16,17) and probably the isoform containing both inserts is regulated developmentally in the brain. The B2-inserted isoform becomes apparent with a different timing in distinct regions during postnatal development of the rat brain (17). In the cerebellum, B2-inserted isoform is highly expressed in cell bodies and dendrites of Purkinje cells (17). The emergence of the B2-inserted isoform in Purkinje cells corresponds to the time when dendritic elongation and synaptogenesis occur actively in the cerebellum (64). This expression level is maintained throughout life of the rat (unpublished result). It is probable that these inserted myosin IIB isoforms play specific roles in the brain by tuning the functional properties in a distinct temporal and spatial manner. We demonstrate here that the B2-inserted chimeric myosin exhibits lower motor activity than the non-inserted one. Based on these results, we speculate that one of the role of B2-inserted myosin IIB is related to maintaining cell morphology particular in a mature brain caused by slowing down the motile events with which the non-B2 inserted isoform is concerned.      The curves are the best fit to the data using the Michaelis-Menten equation.