BIOLOGICAL OVERVIEW

Drosophila Myosin-binding substrate (Mbs), the homolog of mammalian MBS, was identified to study the roles of myosin phosphatase in morphogenesis. Myosin phosphatase negatively regulates nonmuscle myosin II through dephosphorylation of the myosin regulatory light chain (MRLC: Spaghetti squash). Myosin phosphatase's regulatory myosin-binding subunit, Mbs, is responsible for regulating the myosin phosphatase catalytic subunit in response to upstream signals and for determining myosin phosphatase's substrate specificity (Mizuno, 2002).

Embryos defective for both maternal and zygotic Mbs demonstrate a failure in dorsal closure. In the mutant embryos, the defects are mainly confined to the leading edge cells, which fail to fully elongate. Ectopic accumulation of phosphorylated MRLC is detected in the lateral region of the leading edge cells, suggesting that the role of Mbs is to repress the activation of nonmuscle myosin II at the subcellular location for coordinated cell shape change. Aberrant accumulation of F-actin within the leading edge cells may correspond to the morphological aberrations of such cells. Similar defects were seen in embryos overexpressing Rho-associated kinase, suggesting that myosin phosphatase and Rho-kinase function antagonistically. The genetic interaction of Mbs with mutations in the components of the Rho signaling cascade also indicates that Mbs functions antagonistically to the Rho signal transduction pathway. The results indicate an important role for myosin phosphatase in morphogenesis (Mizuno, 2002).

Dorsal closure is a morphogenetic event that takes place during the late stages of embryogenesis in Drosophila. Halfway through embryogenesis, the dorsal surface of the embryo is covered by an extraembryonic membrane, the amnioserosa. Later on, the lateral epidermis stretches dorsally and spreads over the amnioserosa. The two edges of lateral epidermis meet at the dorsal midline and fuse to close the dorsal surface of the embryo. This process is carried out by elongation of the epidermal cells without proliferation or cell recruitment (Mizuno, 2002).

Dorsal closure is a process well suited to studies on the molecular and cellular basis of morphogenesis, and a number of loci involved in this process have been identified from their 'dorsal open' or 'dorsal hole' phenotypes, which are characterized by large holes in their dorsal cuticle. They can be grouped into at least four classes:
(1) genes involved in the Jun amino-terminal kinase (JNK) signaling cascade
(2) genes encoding the components of the Decapentaplegic (Dpp)-mediated signal transduction pathway
(3) genes involved in the Rho GTPase-mediated signaling pathway
(4) genes encoding cytoskeletal proteins and membrane-associated molecules for cell adhesion.
Activation of the JNK signaling cascade is required in the dorsal-most cells of the lateral epidermis, the leading edge cells, to induce the expression of Dpp. One of the target genes of Dpp signaling is zipper (zip), which encodes the heavy chain of nonmuscle myosin II, and Dpp signaling in the leading edge cells activates the transcription (Ariquier, 2001) of zip (Mizuno, 2002).

At the leading edge of the lateral epidermis, filamentous actin (F-actin) and nonmuscle myosin II are prominently accumulated. The supracellular purse-string composed of the actomyosin contractile apparatus provides one of the major forces for promoting dorsal closure. An analysis of the zip mutations has demonstrated an embryonic lethality due to defects in dorsal closure, indicating that nonmuscle myosin II is required for the morphogenetic processes in dorsal closure. Genetic interactions between zip and mutations in the components of the Rho signaling pathway suggest that nonmuscle myosin II is regulated by the Rho signals (Mizuno, 2002).

Nonmuscle myosin II is a hexamer composed of two of each of three subunits; the heavy chain, the regulatory light chain (MRLC) and the essential light chain. The force-generating activity of actomyosin is mainly regulated by phosphorylation and dephosphorylation of MRLC. Ca²⁺/calmodulin-dependent myosin light chain kinase (MLCK) and Rho-kinase/Rokalpha, one of the effectors of the Rho GTPase, are responsible for the phosphorylation of MRLC. However, myosin phosphatase dephosphorylates MRLC, leading to the inactivation of nonmuscle myosin II. Myosin phosphatase is a heterotrimer composed of a catalytic subunit (Myosin phosphatase) belonging to protein phosphatase 1c (PP1c), the myosin-binding subunit (Mbs) and M20 (Alessi, 1992; Hartshorne, 1998). Mbs plays the regulatory roles of myosin phosphatase as a target of the upstream signals and as a determinant of substrate specificity. Myosin phosphatase is negatively regulated through phosphorylation of Mbs by Rho-kinase/Rokalpha (Kimura, 1996; Kawano, 1999). Thus, Rho-kinase/Rokalpha doubly activates nonmuscle myosin II (Kaibuchi, 1999) through direct phosphorylation of MRLC and inactivation of myosin phosphatase by phosphorylating Mbs (Mizuno, 2002).

To elucidate the roles of Mbs during dorsal closure, the distribution of nonmuscle myosin II and its activation were analyzed during the course of dorsal closure. The process can be divided into three phases. It starts with elongation of the dorsal-most, leading edge cells of the lateral epidermis along the dorsoventral axis. This is followed by elongation of the other lateral epidermal cells with the amnioserosa becoming covered by the elongating epidermis. Finally, the two lateral edges of the epidermis meet at the dorsal midline and fuse to close the epidermis (Mizuno, 2002).

Before the onset of the lateral epidermis elongation, the heavy chain of nonmuscle myosin II and MRLC outline the inner surface of the plasma membrane of the lateral epidermal cells at low levels, and they are concentrated at moderate levels along the leading edge of the lateral epidermis. The phosphorylated form of MRLC is localized similarly. During the course of extensive elongation, the heavy chain and MRLC accumulate at high levels along the leading edge. Phosphorylated MRLC is also detected at a high level along the leading edge. After the meeting of the two lateral epidermal sheets, both the heavy chain and MRLC at the leading edge become diffuse, and the phosphorylated form of MRLC disappears at the site of fusion. These observations indicate that the distribution and phosphorylation of nonmuscle myosin II are dynamically regulated during dorsal closure (Mizuno, 2002).

Next the distribution and activation of nonmuscle myosin II and F-actin were examined in Mbs mutant embryos. Examination of the cell shape with anti-phosphotyrosine antibody has revealed that the leading edge cells fail to fully elongate, and some of them remain polygonal. In contrast, the epidermal cells located more ventrally elongate nearly normally. A significant amount of phosphorylated MRLC is detected along the dorsal side of the boundaries between the leading edge cells in the mutant embryos, while the distribution of the heavy chain of nonmuscle myosin II is essentially not affected. F-actin is highly accumulated along the leading edge of the lateral epidermis. In the mutant embryos, an aberrant accumulation of F-actin within the leading edge cells is observed, while its distribution along the leading edge is essentially unaffected. This may correspond to the morphological aberrations of the leading edge cells (Mizuno, 2002).

The phenotype produced by overexpression of Rho-kinase resembles that of Mbs mutant embryos, and both phenotypes can be suppressed by reducing the gene dosage of zip⁺. The results suggest that these phenotypes are the result of the hyperactivation of nonmuscle myosin II, and that myosin phosphatase and Rho-kinase function antagonistically toward each other in regulating nonmuscle myosin II. A similar observation has been reported for let-502 and mel-11 of C. elegans, which encode the homolog of Rho-kinase and Mbs, respectively. The genes have been demonstrated to function in the hypodermal cell-shape change associated with the elongation of the embryo in C. elegans (Wissmann, 1997), thus indicating the conservation of the regulatory mechanisms of nonmuscle myosin II in morphogenesis. The genetic link between zip and the mutations in the Rho signaling pathway has been demonstrated, and this indicates the regulation of nonmuscle myosin II by the Rho signaling pathway. The suppression of both phenotypes generated by double heterozygosity for zip and the mutations in the components of the Rho-signaling pathway by Mbs demonstrates that Mbs functions antagonistically toward the Rho-signaling pathway in the regulation of nonmuscle myosin II (Mizuno, 2002).

During dorsal closure, an extensive cell shape change in the lateral epidermis takes place. One of the major forces underlying the morphogenetic process is provided by the constriction of a supracellular purse-string revealed by the high level accumulation of F-actin and the heavy chain of nonmuscle myosin II along the leading edge of the lateral epidermis. MRLC is also highly accumulated along the leading edge. Small quantities of these components of actomyosin are also detected along the inner surface of the plasma membrane. Phosphorylated MRLC was detected in significant amounts along the leading edge of the lateral epidermis, indicating that nonmuscle myosin II is persistently activated along the leading edge during extensive epidermal spreading (Mizuno, 2002).

In Mbs mutant embryos, the defects in dorsal closure seem to be confined to the leading edge cells, and these cells fail to fully elongate. In contrast, the lateral epidermal cells located more ventrally elongate more or less normally. It has been reported that dorsal closure is driven by multiple forces and that it can proceed in the absence of an intact contractile purse-string at the leading edge. These lateral epidermal cells are under tension during dorsal closure, and they themselves may produce the forces to elongate (Mizuno, 2002).

It should be noted that the accumulation of nonmuscle myosin II in the leading edge cells was essentially not affected. Activation of nonmuscle myosin II takes place along the leading edge as in normal embryos, since the phosphorylated MRLC was detected there in the mutant embryos. In addition to the distribution along the leading edge, a significant accumulation of phosphorylated MRLC was detected also on the dorsal side of the boundaries between the leading edge cells in the mutant embryos. This may indicate the role of myosin phosphatase in inactivating nonmuscle myosin II in this subcellular location to coordinate elongation of the leading edge cells (Mizuno, 2002).

The results suggest the localized activation of myosin phosphatase during the normal course of dorsal closure. One possible explanation for this localized activation is that there is a subcellular distribution of myosin phosphatase itself in this region. However, a suitable antibody would have to be raised against Mbs to determine if this was so. Another possible explanation is the localization of an activator of myosin phosphatase in this region. Thus, there must be a subcellular localization of mechanisms for the regulation of nonmuscle myosin II. It has been demonstrated that the cellular polarity of the leading edge cells is altered from basolateral to apical in the leading edge during elongation. It would be of interest to learn how cellular polarity affects the pattern of regulation of the nonmuscle myosin II in the leading edge cells. The results obtained in this study demonstrate the importance of both positive and negative regulation of nonmuscle myosin II in morphogenesis (Mizuno, 2002).

PROTEIN STRUCTURE

Amino Acids - 927 and 797: DMBS-L and DMBS-S respectively

Structural Domains

To identify the Drosophila homolog of Mbs, a search of the Drosophila genome database was conducted for sequences similar to vertebrate MBSs. The vertebrate MBSs share three conserved domains or motifs: amino-terminal ankyrin-repeats, a centrally located major phosphorylation site, and a carboxy-terminal leucine-zipper motif. A BLAST search for the putative open reading frame encoding a sequence similar to the phosphorylation site revealed a genomic fragment flanked by sequences encoding ankyrin repeats and a leucine zipper motif (Mizuno, 2002).

Analysis of the corresponding EST clones generated by the Berkeley Drosophila Genome Project revealed that none of them contained all three conserved domains. Therefore, the 5'- and 3'-UTR sequences were predicted from the EST and genomic sequences, and the cDNA fragments containing the complete coding sequence were amplified by PCR using cDNA libraries for templates. Two types of cDNAs were thus obtained from different cDNA libraries, each encoding a polypeptide of 927 and 797 amino acid residues with the predicted relative molecular mass of 101,409 and 87,509, respectively. The two sequences are identical except for an insertion of a 129 amino acid sequence upstream of the putative phosphorylation site in the longer polypeptide (Mizuno, 2002).

They contain the three conserved motifs, and the amino acid identities to their corresponding regions of human MYPT2 (Fujioka, 1998) are 57.8%, 55.6% and 77.8%, respectively, from the amino terminus. The similarity to vertebrate MBSs is restricted to these regions. Vertebrate MBSs and the homolog of Caenorhabditis elegans, MEL-11, contain seven ankyrin repeats (Fujioka, 1998; Wissmann, 1997), but the sequences in the fourth and seventh repeats diverge in Drosophila. In the Drosophila Genome Database, no other sequence similar to Mbs was found, and these sequences are now considered as Drosophila homologs of Mbs. The longer and shorter forms are referred to as Drosophila Mbs-long (DMBS-L) and Drosophila MBS-short (DMBS-S), respectively (Mizuno, 2002).

The two cDNAs derive from a single gene by alternative splicing. The DMBS-L-specific, 7th exon encodes a sequence of 129 amino acid residues. Furthermore, two in-frame consecutive splicing acceptor sites are present 5' of the 4th exon, and splicing variations at this site have added one more amino acid residue in DMBS-L as compared to DMBS-S. Several partial cDNA fragments different from DMBS-L and DMBS-S were obtained. Thus, the Mbs gene encodes multiple forms of Mbs through differential splicing. Splicing variants have been reported also for the vertebrate MBSs and MEL-11 of C. elegans (Mizuno, 2002).

date revised: 10 May 2002

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

Please e-mail comments/corrections to brodyt@codon.nih.gov