Myosin motor function: structural and mutagenic approaches.

Recent advances in three areas of myosin research--structural biology, in vitro motility assays, and mutagenesis--are leading to a new understanding of the molecular mechanism of chemomechanical transduction by this motor protein. Highlights include rational design of mutants using the crystal structure of subfragment 1, combined in vivo and in vitro mutant analyses using Dictyostelium, and the emergence of baculovirus as an in vitro system for expression of mutated mammalian myosins.

A myosin II mutation uncouples ATPase activity from motility and shortens step size.

It is thought that Switch II of myosin, kinesin and G proteins has an important function in relating nucleotide state to protein conformation. Here we examine a myosin mutant containing an S456L substitution in the Switch II region. In this protein, mechanical activity is uncoupled from the chemical energy of ATP hydrolysis so that its gliding velocity on actin filaments is only one-tenth of that of the wild type. The mutant spends longer in the strongly bound state and exhibits a shorter step size, which together account for the reduction in in vitro velocity. This is the first single point mutation in myosin that has been found to affect step size.

Combining single-molecule optical trapping and small-angle x-ray scattering measurements to compute the persistence length of a protein ER/K alpha-helix.

A relatively unknown protein structure motif forms stable isolated single alpha-helices, termed ER/K alpha-helices, in a wide variety of proteins and has been shown to be essential for the function of some molecular motors. The flexibility of the ER/K alpha-helix determines whether it behaves as a force transducer, rigid spacer, or flexible linker in proteins. In this study, we quantify this flexibility in terms of persistence length, namely the length scale over which it is rigid. We use single-molecule optical trapping and small-angle x-ray scattering, combined with Monte Carlo simulations to demonstrate that the Kelch ER/K alpha-helix behaves as a wormlike chain with a persistence length of approximately 15 nm or approximately 28 turns of alpha-helix. The ER/K alpha-helix length in proteins varies from 3 to 60 nm, with a median length of approximately 5 nm. Knowledge of its persistence length enables us to define its function as a rigid spacer in a translation initiation factor, as a force transducer in the mechanoenzyme myosin VI, and as a flexible spacer in the Kelch-motif-containing protein.

Towards a molecular understanding of cytokinesis.

In this review, we focus on recent discoveries regarding the molecular basis of cleavage furrow positioning and contractile ring assembly and contraction during cytokinesis. However, some of these mechanisms might have different degrees of importance in different organisms. This synthesis attempts to uncover common themes and to reveal potential relationships that might contribute to the biochemical and mechanical aspects of cytokinesis. Because the information about cytokinesis is still fairly rudimentary, our goal is not to present a definitive model but to present testable hypotheses that might lead to a better mechanistic understanding of the process.

Actin in triton-treated cortical preparations of unfertilized and fertilized sea urchin eggs.

Triton-treated cortical fragments of unfertilized and fertilized sea urchin eggs prepared in the presence of greater than or equal to 5 mM EGTA contain 15-30% of the total egg actin. However, actin filaments are not readily apparent by electron microscopy on the cortical fragments of unfertilized eggs but are numerous on those of fertilized eggs. The majority of the actin associated with cortical fragments of unfertilized eggs is solubilized by dialysis against a low ionic strength buffer at pH 7.5. This soluble actin preparation (less than 50% pure actin) does not form proper filaments in 0.1 M KCl and 3 mM MgCl2, whereas actin purified from this preparation does, as judged by electron microscopy. Optical diffraction analysis reveals that these purified actin filaments have helical parameters very similar to those of muscle actin. Furthermore, the properties of the purified actin with regard to activation of myosin ATPase are similar to those of actin from other cell types. The possibility that actin is maintained in a nonfilamentous form on the inner surface of the unfertilized egg plasma membrane and is induced to assemble upon fertilization is discussed.

Dynacortin, a genetic link between equatorial contractility and global shape control discovered by library complementation of a Dictyostelium discoideum cytokinesis mutant.

We have developed a system for performing interaction genetics in Dictyostelium discoideum that uses a cDNA library complementation/multicopy suppression strategy. Chemically mutagenized cells were screened for cytokinesis-deficient mutants and one mutant was subjected to library complementation. Isolates of four different genes were recovered as modifiers of this strain's cytokinesis defect. These include the cleavage furrow protein cortexillin I, a novel protein we named dynacortin, an ezrin-radixin-moesin-family protein, and coronin. The cortexillin I locus and transcript were found to be disrupted in the strain, identifying it as the affected gene. Dynacortin is localized partly to the cell cortex and becomes enriched in protrusive regions, a localization pattern that is similar to coronin and partly dependent on RacE. During cytokinesis, dynacortin is found in the cortex and is somewhat enriched at the poles. Furthermore, it appears to be reduced in the cleavage furrow. The genetic interactions and the cellular distributions of the proteins suggest a hypothesis for cytokinesis in which the contraction of the medial ring is a function of spatially restricted cortexillin I and myosin II and globally distributed dynacortin, coronin, and RacE.

Dictyostelium myosin: characterization of chymotryptic fragments and localization of the heavy-chain phosphorylation site.

Chymotrypsin cleaves Dictyostelium myosin in half, splitting the heavy chain (210,000 daltons) into two fragments of 105,000 daltons each. One of the two major fragments is soluble at low ionic strength and has a native molecular weight of 130,000. As judged by SDS polyacrylamide gel electrophoresis, this soluble fragment consists of the two intact myosin light chains of 18,000 and 16,000 daltons and a 105,000-dalton polypeptide derived from the myosin heavy chain. The soluble fragment retains actin-activated ATPase activity and the ability to bind to actin in an ATP-dissociable fashion. The maximal velocity of the actin-activated ATPase activity of the soluble fragment is 80% of that of uncleaved myosin, although its apparent Km for actin is 12-fold greater than that of myosin. In addition to the major soluble 105,000-dalton fragment discussed above, chymotryptic cleavage of the Dictyostelium myosin also generates fragments that are insoluble at low ionic strength. The major insoluble fragment is 105,000 daltons on an SDS polyacrylamide gel and forms thick filaments that are devoid of myosin heads. A less prevalent insoluble fragment has a molecular weight of 83,000 and is probably a subfragment of the insoluble 105,000-dalton fragment. The heavy chain of myosin is phosphorylated in vivo and the phosphorylation site has been localized to the insoluble fragments, which derive from the tail portion of the myosin molecule.

Nucleation of polar actin filament assembly by a positively charged surface.

Polylysine-coated polystyrene beads can nucleate polar assembly of monomeric actin into filamentous form. This nucleation has been demonstrated by a combination of biochemical and structural experiments. The polylysine-coated beads accelerate the rate of actin assembly as detected by two different biochemical assays. Subsequent examination of the beads by electron microscopy reveals numerous actin filaments of similar length radiating from the beads. ATP promotes this bead-induced acceleration of assembly. Decoration of the filaments with the myosin fragment S1 shows that these filaments all have the same polarity, with the arrowhead pattern pointing toward the bead. The relevance of the system to in vitro mechanisms and its usefulness in other studies are discussed.

Cytochalasin inhibits the rate of elongation of actin filament fragments.

Submicromolar concentrations of cytochalasin inhibit the rate of assembly of highly purified dictyostelium discoideum actin, using a cytochalasin concentration range in which the final extent of assembly is minimally affected. Cytochalasin D is a more effective inhibitor than cytochalasin B, which is in keeping with the effects that have been reported on cell motility and with binding to a class of high-affinity binding sites from human erythrocyte membranes (Lin and Lin. 1978. J. Biol. CHem. 253:1415; Lin and Lin. 1979. Proc. Natl. Acad. Sci. U.S.A. 76:2345); 5x10(-7) M cytochalasin B lowers it to 70 percent of the control value, whereas 10(-7) M cytochalasin B lowers the rate to 25 percent. Fragments of F-actin were used to increase the rate of assembly fivefold by providing more filament ends on to which monomers could add. Under these conditions, cytochalasin has an even more dramatic effect on the assembly rate; the concentrations of cytochalasin B and cytochalasin D required for half-maximal inhibition are 2x10(-7) M and 10(-8) M, respectively. The assembly rate is most sensitive to cytochalasin when actin assembly is carried out in the absence of ATP (with 3 mM ADP present to stabilize the actin). In this case, the concentrations of cytochalasin B and cytochalasin D required for half-maximal inhibition are 4x10(-8) M and 1x10(-9) M, respectively. A scatchard plot has been obtained using [(3)H]cytochalasin B binding to F-actin in the absence of ATP. The K(d) from this plot (approximately 4x10(-8) M) agrees well with the concentration of cytochalasin B required for half-maximal inhibition of the rate of assembly under these conditions. The number of cytochalasin binding sites is roughly one per F-actin filament, suggesting that cytochalasin has a specific action on actin filament ends.

Monoclonal antibodies against seven sites on the head and tail of Dictyostelium myosin.

Ten monoclonal antibodies (My1-10) against Dictyostelium discoideum myosin were prepared and characterized. Nine bound to the 210-kD heavy chain and one (My8) bound to the 18-kD light chain. They defined six topographically distinct antigenic sites of the heavy chain. Five binding sites (the My1, My5, My10 site, and the My2, My3, My4, and My9 sites) are located on the rod portion of the myosin molecule. The position of the sixth site (the My6 and My7 site) is less certain, but it appears to be near the junction of the globular heads and the rod. Three of the antibodies (My2, My3, and My6) bound to myosin filaments in solution and could be sedimented in stoichiometric amounts with the filamentous myosin. In contrast, My4, which recognized a site on the rod, inhibited the polymerization of monomeric myosin into filaments. A single antibody (My6) affected the actin-activated ATPase of myosin. The nature of the effect depended on the valency of the antibody and the myosin. Bivalent IgG and F(ab')2 fragments of My6 inhibited the actin-activated ATPase of filamentous myosin by 50% whereas univalent Fab' fragments increased the activity by 50%. The actin-activated ATPase activity of the soluble chymotryptic fragment of myosin was increased 80-90% by both F(ab')2 and Fab' of My6.

A 45,000-mol-wt protein from unfertilized sea urchin eggs severs actin filaments in a calcium-dependent manner and increases the steady-state concentration of nonfilamentous actin.

A 45,000-mol-wt protein has been purified from unfertilized sea urchin (Strongylocentrotus purpuratus) eggs. The isolation scheme includes DEAE cellulose ion-exchange chromatography, gel filtration, and hydroxylapatite chromatography. The homogeneity of the isolated protein is greater than 90% by SDS PAGE. The 45,000-mol-wt protein reduces the viscosity of actin filaments in a Ca2+-dependent manner. The free calcium concentration required for the activity of this protein is in the micromolar range. Electron microscopic studies reveal that the formation of short filaments parallels the decrease in viscosity. Energy transfer and sedimentation experiments indicate a net disassembly of actin filaments and an increase in the steady-state nonfilamentous actin concentration in the presence of Ca2+ ions and the 45,000-mol-wt protein. The increase in the steady-state nonfilamentous actin concentration is proportional to the amount of 45,000-mol-wt protein added. The actin molecules disassembled by the addition of the 45,000-mol-wt protein are capable of polymerization.

Monoclonal antibodies prepared against Dictyostelium actin: characterization and interactions with actin.

Three mouse monoclonal antibodies, Act I, Act II, and Act IV, against actin from the cellular slime mold Dictyostelium discoideum, have been made and characterized. All three antibodies are IgG1 and share the following properties: They form stable complexes with monomeric Dictyostelium actin, which prevents polymerization of the actin into filaments. On addition to preformed actin filaments, they cause a reduction in filament size and in the viscosity of the actin solution. They cross-react strongly with actins from the lower eucaryotes Physarum and Acanthamoeba, but not with alpha-actins from rabbit and human muscle or beta- and gamma-actins from human erythrocytes and a human B lymphoid cell line. Act II and Act IV recognize a similar antigenic determinant that is topographically distinct from that identified by Act I. In protein immunoblotting, only Act I bound strongly to Dictyostelium actin. Analysis of actin fragments with this technique showed that amino acids 13 to about 50 are required for Act I binding to actin. A comparison of the amino acid sequences of actins from lower eucaryotes and higher vertebrates implicates threonine 41 as a critical residue in the Act I antigenic site. The properties of Act II and Act IV suggest that they recognize antigenic sites involving the NH2-terminal six residues.

ATP-dependent movement of myosin in vitro: characterization of a quantitative assay.

Sheetz and Spudich (1983, Nature (Lond.), 303:31-35) showed that ATP-dependent movement of myosin along actin filaments can be measured in vitro using myosin-coated beads and oriented actin cables from Nitella. To establish this in vitro movement as a quantitative assay and to understand better the basis for the movement, we have defined the factors that affect the myosin-bead velocity. Beads coated with skeletal muscle myosin move at a rate of 2-6 micron/s, depending on the myosin preparation. This velocity is independent of myosin concentration on the bead surface for concentrations above a critical value (approximately 20 micrograms myosin/2.5 X 10(9) beads of 1 micron in diameter). Movement is optimal between pH 6.8 and 7.5, at KCl concentrations less than 70 mM, at ATP concentrations greater than 0.1 mM, and at Mg2+ concentrations between 2 and 6 mM. From the temperature dependence of bead velocity, we calculate activation energies of 90 kJ/mol below 22 degrees C and 40 kJ/mol above 22 degrees C. Different myosin species move at their own characteristic velocities, and these velocities are proportional to their actin-activated ATPase activities. Further, the velocities of beads coated with smooth or skeletal muscle myosin correlate well with the known in vivo rates of myosin movement along actin filaments in these muscles. This in vitro assay, therefore, provides a rapid, reproducible method for quantitating the ATP-dependent movement of myosin molecules on actin.

A 40,000-dalton protein from Dictyostelium discoideum affects assembly properties of actin in a Ca2+-dependent manner.

A 40,000-dalton protein that affects the assembly properties of actin in a Ca2+-dependent manner has been purified from Dictyostelium discoideum. Gel filtration chromatography indicates that the native form of this protein is a monomer. A major effect of this protein is to reduce the sedimentability of F-actin in a stoichiometric fashion. Nearly complete loss of sedimentability is observed at ratios of the 40,000-dalton protein to actin of greater than 1:10. At low stoichiometries, this protein can accelerate the rate of actin assembly under certain experimental conditions. These effects of the 40,000-dalton protein on the actin assembly properties described above require calcium ion. The 40,000-dalton protein does not exert its effects by proteolyzing actin. Furthermore, peptide maps demonstrate that this protein is not a proteolytic fragment of actin.

Mechanism of K+-induced actin assembly.

The assembly of highly purified actin from Dictyostelium discoideum amoebae and rabbit skeletal muscle by physiological concentrations of KCI proceeds through successive stages of (a) rapid formation of a distinct monomeric species referred to as KCI-monomer, (b) incorporation of KCI-monomers into an ATP-containing filament, and (c) ATP hydrolysis that occurs significantly after the incorporation event. KCI-monomer has a conformation which is distinct from that of either conventional G- or F-actin, as judged by UV spectroscopy at 210-220 nm and by changes in ATP affinity. ATP is not hydrolyzed during conversion of G-actin to KCI-monomer. KCI-monomer formation precedes filament formation and may be necessary for the assembly event. Although incorporation of KCI-monomers into filaments demonstrates lagphase kinetics by viscometry, both continuous absorbance monitoring at 232 nm and rapid sedimentation of filaments demonstrate hyperbolic assembly curves. ATP hydrolysis significantly lags the formation of actin filaments. When half of the actin has assembled, only 0.1 to 0.2 mole of ATP are hydrolyzed per mole of actin present as filaments.

Mechanism of action of cytochalasin: evidence that it binds to actin filament ends.

To test the idea that cytochalasin retards actin assembly by binding to filament ends, we have designed a new assay for cytochalasin binding in which the number of filament ends can be varied independently of the total actin concentration. Actin is reacted with polylysine-coated polystyrene beads to make filament ends (Brown and Spudich, 1979, J. Cell Biol. 80:499-504) and then reacted with [3H]cytochalasin B. We have found that cytochalasin B binds to beads in the presence of actin, and that the number of cytochalasin B binding sites can be varied as a function of the number of filament ends independent of the total actin concentration by varying the bead concentration.

A structural model for phosphorylation control of Dictyostelium myosin II thick filament assembly.

Myosin II thick filament assembly in Dictyostelium is regulated by phosphorylation at three threonines in the tail region of the molecule. Converting these three threonines to aspartates (3 x Asp myosin II), which mimics the phosphorylated state, inhibits filament assembly in vitro, and 3 x Asp myosin II fails to rescue myosin II-null phenotypes. Here we report a suppressor screen of Dictyostelium myosin II-null cells containing 3 x Asp myosin II, which reveals a 21-kD region in the tail that is critical for the phosphorylation control. These data, combined with new structural evidence from electron microscopy and sequence analyses, provide evidence that thick filament assembly control involves the folding of myosin II into a bent monomer, which is unable to incorporate into thick filaments. The data are consistent with a structural model for the bent monomer in which two specific regions of the tail interact to form an antiparallel tetrameric coiled-coil structure.

Structure and function of the cytoskeleton of a Dictyostelium myosin-defective mutant.

To study the role of conventional myosin in nonmuscle cells, we determined the cytoskeletal organization and physiological responses of a Dictyostelium myosin-defective mutant. Dictyostelium hmm cells were created by insertional mutagenesis of the myosin heavy chain gene (De Lozanne, A., and J. A. Spudich. 1987. Science (Wash. DC). 236: 1086-1091). Western blot analysis using different mAbs confirms that hmm cells express a truncated myosin fragment of 140 kD (HMM-140 protein) instead of the normal 243-kD myosin heavy chain (MHC). Spontaneous revertants appear at a frequency less than 4 x 10(-5), which synthesize normal myosin and are capable of forming thick filaments. In hmm cells, the HMM-140 protein is diffusely distributed in the cytoplasm, indicating that it cannot assemble into thick filaments. The actin distribution in these mutant cells appears similar to that of wild-type cells. However, there is a significant abnormality in the organization of cytoplasmic microtubules, which penetrate into lamellipodial regions. The microtubule networks consist of approximately 13 microtubules on average and their pattern is abnormal. Although hmm cells can form mitotic spindles, mitosis is not coordinated with normal furrow formation. The hmm cells are clearly defective in the contractile events that lead to normal cytokinesis. The retraction of different regions of the cell can result in the occasional pinching off of part of the cell. This process is not coupled with formation of mitotic spindles. There is no specific accumulation of HMM-140 in such constrictions, whereas 73% of such cells show actin concentrated in these regions. The mutant hmm cells are also deficient in capping of Con-A-bound surface receptors, but instead internalize this complex into the cytoplasm. The hmm cells display active phagocytosis of bacteria. Whereas actin is concentrated in the phagocytic cups, HMM-140 protein is not localized in these regions. cAMP, a chemoattractant that induces drastic rounding up and formation of surface blebs in wild type cells, does not induce rounding up in the hmm cells. A Triton-permeabilized cell model of the wild-type amebae contracts on reactivation with Mg-ATP, whereas a model of the hmm cell shows no detectable contraction. Our data demonstrate that the conventional myosin participates in the significant cortical motile activities of Dictyostelium cells, which include rounding up, constriction of cleavage furrows, capping surface receptors, and establishing cell polarity.

Spatial and temporal control of nonmuscle myosin localization: identification of a domain that is necessary for myosin filament disassembly in vivo.

Myosin null mutants of Dictyostelium are defective for cytokinesis, multicellular development, and capping of surface proteins. We have used these cells as transformation recipients for an altered myosin heavy chain gene that encodes a protein bearing a carboxy-terminal 34-kD truncation. This truncation eliminates threonine phosphorylation sites previously shown to control filament assembly in vitro. Despite restoration of growth in suspension, development, and ability to cap cell surface proteins, these delta C34-truncated myosin transformants display severe cytoskeletal abnormalities, including excessive localization of the truncated myosin to the cortical cytoskeleton, impaired cell shaped dynamics, and a temporal defect in myosin dissociation from beneath capped surface proteins. These data demonstrate that the carboxy-terminal domain of myosin plays a critical role in regulating the disassembly of the protein from contractile structures in vivo.

Myosin light chain kinase and myosin light chain phosphatase from Dictyostelium: effects of reversible phosphorylation on myosin structure and function.

We have partially purified myosin light chain kinase (MLCK) and myosin light chain phosphatase (MLCP) from Dictyostelium discoideum. MLCK was purified 4,700-fold with a yield of approximately 1 mg from 350 g of cells. The enzyme is very acidic as suggested by its tight binding to DEAE. Dictyostelium MLCK has an apparent native molecular mass on HPLC G3000SW of approximately 30,000 D. Mg2+ is required for enzyme activity. Ca2+ inhibits activity and this inhibition is not relieved by calmodulin. cAMP or cGMP have no effect on enzyme activity. Dictyostelium MLCK is very specific for the 18,000-D light chain of Dictyostelium myosin and does not phosphorylate the light chain of several other myosins tested. Myosin purified from log-phase amebas of Dictyostelium has approximately 0.3 mol Pi/mol 18,000-D light chain as assayed by glycerol-urea gel electrophoresis. Dictyostelium MLCK can phosphorylate this myosin to a stoichiometry approaching 1 mol Pi/mol 18,000-D light chain. MLCP, which was partially purified, selectively removes phosphate from the 18,000-D light chain but not from the heavy chain of Dictyostelium myosin. Phosphatase-treated Dictyostelium myosin has less than or equal to 0.01 mol Pi/mol 18,000-D light chain. Phosphatase-treated myosin could be rephosphorylated to greater than or equal to 0.96 mol Pi/mol 18,000-D light chain by incubation with MLCK and ATP. We found myosin thick filament assembly to be independent of the extent of 18,000-D light-chain phosphorylation when measured as a function of ionic strength. However, actin-activated Mg2+-ATPase activity of Dictyostelium myosin was found to be directly related to the extent of phosphorylation of the 18,000-D light chain. MLCK-treated myosin moved in an in vitro motility assay (Sheetz, M. P., and J. A. Spudich, 1983, Nature (Lond.), 305:31-35) at approximately 1.4 micron/s whereas phosphatase-treated myosin moved only slowly or not at all. The effects of phosphatase treatment on the movement were fully reversed by subsequent treatment with MLCK.