Differential regulation of vertebrate myosins I and II

KATHLEEN COLLINS and PAUL MATSUDAIRA M. I. T. and Whitehead Institute for Biomedical Research, Cambridge, MA 02142, USA

Summary Cell motility events require movement of the cytoskeleton. Actin-based movement is catalyzed by the mechanoenzyme myosin, which translocates toward the barbed end of actin filaments in an ATPdependent fashion. There are two subclasses of myosin with different structures and functions: conventional filamentous myosin (myosin II) and monomeric myosin I. Vertebrate non-muscle myo­ sins I and II function as similar actin motors in vitro, catalyzing virtually identical actin-activated MgATP hydrolysis and motility. The functional diversifi­ cation of these two enzymes results from their differential regulation. Calcium and tropomyosin, which activate the MgATP hydrolysis and motility of vertebrate non-skeletal muscle myosin II proteins, inhibit vertebrate (brush border) myosin I. The activities and regulation of brush border myosin I provide insight into conserved and unique features of the myosin mechanoenzymes and suggest how the functions of myosins I and II are divided in ver­ tebrate cells. Brush border myosin I as an enzyme also contributes to our understanding of the molecu­ lar mechanism of motility. Key words: cell motility, myosin regulation, tropomyosin, calmodulin.

Introduction Myosin, isolated as a force-generating component of muscle, was the first cellular enzyme studied for its ability to produce mechanical force from the energy of ATP hydrolysis. In the presence of ATP and calcium, myosin in muscle bipolar thick filaments induces sliding of actincontaining thin filaments toward each other. This results in a net contraction, or shortening, of muscle. The discovery of molecules similar to skeletal muscle myosin in non-muscle cells suggested that myosin-mediated contraction might be involved in cell shape change, locomotion, cytokinesis, phagocytosis and intracellular transport. The requirement for conventional myosin, termed myosin II, for non-muscle cell contraction and cytokinesis has been shown by selective depletion exper­ iments using extraction (Keller et al. 1985), antibodies (Mabuchi and Okuno, 1977), antisense RNA (Knecht and Loomis, 1987) and gene disruption (De Lozanne and Spudich, 1987). These same experiments, however, demon­ strate that other actin-based motility events do not require myosin II. Cells lacking myosin II still extend pseudopods, phagocytose and chemotax. Journal of Cell Science, Supplement 14, 11-16 (1991) Printed in Great Britain © The Company of Biologists Limited 1991

Conventional myosins, or myosin II proteins, consist of a hexamer of two associated heavy chains (of about 200xl03Mr), each of which binds two light chains (162 0 x l0 sMr). The C-terminal ‘tail’ half of the heavy chain dimerizes to form a parallel a-helical coiled coil. This tail domain also directs the staggered assembly of heavy-chain dimers into the large, bipolar filaments characteristic of skeletal muscle. The globular, N-terminal ‘head’ domain of myosin binds actin, ATP and light chains, and by itself retains myosin’s two defining activities: actin-activated hydrolysis of MgATP and actin filament translocation (see Warrick and Spudich, 1987, for a review of myosin II). These two activities are thought to be coupled, with efficient ATP hydrolysis linked to productive mechanical work. The first actin-activated ATP-hydrolyzing enzyme not conforming to the structure of myosin II was discovered in the single-celled eukaryote Acanthamoeba (Pollard and Korn, 1973). This new myosin, because it lacks a tail capable of dimerization, has been termed myosin I. The generality of this myosin variant to other cells has only recently become clear. Myosin I proteins from Acan­ thamoeba, the slime mold Dictyostelium and vertebrate cells all share several defining characteristics. Structur­ ally, all myosin I proteins exist as monomers and have relatively shorter, non-helical tail domains with some unique and some shared features (Fig. 1). The tails of myosin I proteins from Acanthamoeba and Dictyostelium contain a glycine/proline/alanine or threonine-rich region (Jung et al. 1987; Jung et al. 1989) with a second (ATP-independent) actin-binding site (Lynch et al. 1986). The vertebrate (brush border) myosin I tail contains unique regions with homology to concensus calmodulin binding sites (Hoshimaru et al. 1989; unpublished obser­ vation). Acanthamoeba, Dictyostelium, and brush border myosin I tails all share a region which binds phospholipids in vitro (Adams and Pollard, 1989). Myosin I enzymes also share similar cellular localiz­ ations. Whereas myosin II is distributed in cytoplasm, actin bundles or ordered actin arrays, myosin I is associated predominantly with cellular membranes. Acan­ thamoeba myosin I localizes to and cofractionates with the plasma membrane and membrane vesicles (Gadasi and Korn, 1980; Adams and Pollard, 1986). Motile Dictyo­ stelium cells concentrate myosin I specifically at the leading edge, with myosin II distributed in the posterior cortex (Fukui et al. 1989). In vertebrate intestinal epithelial cells, myosin I is localized to apical and basolateral plasma membranes and to the vesicle-contain­ ing terminal web region (Coudrier et al. 1981; Fig. 2). Because of its monomeric nature and membrane localiz­ ation, myosin I is likely to be involved in events such as membrane transport, cortical flow and phagocytosis, events shown not to depend on myosin II. '

11

ATP binding A T P-dependent actin binding

I------------------1------------------ 1

0

100

200

A/rxl0-3 Fig. 1. Comparison of common and distinguishing structural features of myosins. All myosins share a conserved N-terminal domain which binds ATP and actin. The C-terminal domains of the myosins, however, have unique features (shaded). Myosin II proteins possess a tail domain which dimerizes to form «■-helical coiled coil. Acanthamoeba (Jung et al. 1987) and Dictyostelium (Jung et al. 1989) myosin I have a second, ATPindependent actin-binding site. The brush border myosin I tail (Hoshimaru and Nakanishi, 1987; Garcia et al. 1989) contains three sites which by sequence homology are likely to bind the three calmodulins of the complex. All myosin I proteins share a region which binds phospholipids in vitro (thin rod).

The activities of brush border m yosin I To study the properties of myosin I and the division of cellular labor between myosins I and II in a vertebrate system, we have characterized myosin I from intestinal epithelial cells. These cells have an apical domain specialized to absorb essential nutrients from the lumen of the gut, separated from the basolateral cell domain by membrane tight junctions and a circumferential belt of actin filaments. The apical membrane is highly increased in surface area by packed, finger-like protrusions known as microvilli, each stabilized by a central core bundle of 15-20 actin filaments extending into the cytosol. Micro­ villi and the actin-rich region beneath them, the terminal web, together constitute the brush border, which can be isolated intact after cell homogenization (see Louvard, 1989, and Mooseker, 1985, for reviews of brush border). Myosin I was discovered in brush border as the lateral linkage of the microvillus actin core to the plasma membrane by virtue of observations that ATP extracted the linkage from actin (Matsudaira and Burgess, 1979) and that the purified linker complex catalyzed a weakly actin-activated MgATPase activity (Collins and Borysenko, 1984). Brush border also contains myosin II (Mooseker et al. 1978). We purify myosin I by MgATP extraction of isolated chicken brush borders (Collins et al. 1990). The purified complex consists of a 1 1 0 x l0 3Mr heavy chain associated with 1 8 x l0 3Mr subunits. The 1 8 x l0 3Mr protein has been identified as calmodulin by a variety of criteria, including calcium-dependent apparent molecular weight shift on SDS gels, heat stability, and phosphodiesterase activation. We have confirmed the exact identity of the 1 8 x l0 3Mr polypeptide to be calmodulin by amino acid sequencing. Calmodulin, which serves as a regulatory subunit for several cellular enzymes, changes conformation with the cooperative binding of four calcium ions. The function of calmodulin as a subunit of brush border myosin I is not 12

K. Collins and P. Matsudaira

Fig. 2. Localization of brush border myosin I in intestinal epithelial cells. Sections of fixed, frozen adult mouse intestine were stained with affinity-purified chicken brush border myosin I antiserum and fluorescein-conjugated secondary antibody. A villus cross-section is shown, with columnar epithelial cells (a single, polarized sheet) exposed to the lumen (lumen is on the top, left, and right of the figure) at their apical surface. Most myosin I is localized apically in epithelial cells, to microvilli and the terminal web. Lighter staining of the cells’ basolateral membrane is also observed (appears as thin, parallel stripes).

known, but the association of the myosin I heavy chain with a common regulatory protein is in contrast to the association of specialized light chains with myosin II. The localization of myosin I in intestinal epithelial cells is shown by immunofluorescence microscopy in Fig. 2. Myosin I antiserum was raised and affinity-purified against chicken brush border myosin I and detects only a single llO x 103Mr protein in total epithelial cell homogenates (Collins et al. 1990). The apical region of the epithelial cells, containing microvilli and the terminal web, is labeled most strongly, but the basolateral membrane also contains myosin I. This localization is similar to a previous report (Coudrier et al. 1981). The distribution of ‘brush border’ myosin I in several regions of the cell suggests that although a subset of myosin I laterally links the microvillar actin core to the plasma membrane, it probably performs other functions as well. Purified brush border myosin I demonstrates the simplehyperbolic actin-activation of MgATPase activity charac­ teristic of myosin II (Collins et al. 1990). As for non-muscle myosin II, Vmax increases about 40-fold in the presence of actin. At 37 °C and a saturating actin concentration, brush border myosin I hydrolyzes MgATP at approximately 1 s , with a K m of actin-activation of 20-40 ,u m . Brush border myosin I also demonstrates properties of actin filament translocation similar to myosin II. Using the in

vitro sliding filament assay, in which fluorescently labeled actin filaments are translocated over a nitrocellulosecoated, myosin-bound coverslip (Kron and Spudich, 1986), we determine a myosin I-mediated rate of motility of 0.08,ams-1 at 37°C. This rate of motility is identical to that we measure for brush border myosin II in the same assay, but it differs from a previous report of 10-fold slower brush border myosin I motility (Mooseker and Coleman, 1989). In addition, brush border myosin I and rabbit skeletal myosin II (Sheetz et al. 1984) both have the same ATP-concentration dependence of motility, and as for all myosins, brush border myosin I translocates actin inde­ pendent of actin filament length or myosin concentration above a minimum threshhold. Myosin I motility is more stable than myosin II motility in vitro', the rate of translocation by brush border myosin I on a single coverslip is constant for at least two hours at room temperature, with virtually all filaments which attach to the coverslip in constant motion. Preparations of myosin I stored at 4°C show full activity for at least 1 -2 months, in contrast with the very rapid loss, in days, of observable myosin II-mediated motility. As described above, the actin-activated MgATPase activity and motility of vertebrate non-muscle cell myo­ sins I and II are very similar. Therefore, the myosins probably do not serve unique functions by virtue of unique enzymatic activities. This similarity of myosin I and II enzymes appears to be specific to the vertebrate system. Myosin I from Acanthamoeba, but not myosin II, demon­ strates complex kinetics of actin-activated MgATPase activity due to the additional, ATP-independent actinbinding site (Korn et al. 1988). Even without the extra binding site, however, Acanthamoeba myosin I retains unique activities including an actin-activated MgATPase activity 10-fold greater than myosin II (Lynch et al. 1986). The reason for the similarity of myosin I and II enzymes in vertebrate cells in particular is not clear. It is possible that vertebrate myosin I isoforms discovered in the future may possess activities unlike myosin II, and that Acan­ thamoeba myosin I isoforms not yet purified may more closely resemble myosin II.

Calcium regulation The striking similarity of the enzymatic activities of myosins I and II was unexpected, but is consistent with their structural homology. This identity does, however, leave unexplained the basis for division of cell functions. If myosins I and II are similar enzymes, what suits the two proteins to different motility functions in a cell? As one answer to this question, we find that myosins I and II are regulated distinctly: conditions which activate myosin I inhibit smooth and non-muscle myosin II, and vice versa. For smooth and non-muscle myosin II, calcium induces the phosphorylation of regulatory light chain by myosin light chain kinase. This light chain phosphorylation greatly stimulates myosin II actin-activated MgATPase activity and motility. Brush border myosin I is not phosphorylated by myosin light chain kinase (Keller and Mooseker, 1982; unpublished observation). In contrast with its activation of myosin II, we find that calcium inhibits brush border myosin I, and does so directly rather than through phosphorylation (Collins et al. 1990). One to 100 ,um calcium causes a gradual inhibition of myosin I actin-activated MgATPase activity, resulting from the gradual dissociation of a subset of calmodulin in vitro.

Fig. 3. Calcium regulation of brush border myosin I (data from Collins et al. 1990). Myosin I-mediated actin-activated MgATPase activity is measured as a function of increasing actin concentration in 1 mM EGTA (solid lines, filled points) or 1 mM calcium (dashed lines, open points). Actin-independent ATPase activity is subtracted from all points. Circles indicate myosin I alone; triangles indicate the presence of 5 -1 0 /¿M calmodulin (CAM). The actin-activation of myosin I MgATPase activity is inhibited in calcium, but inhibition is reversible with addition of exogenous calmodulin. Calmodulin has no effect on MgATPase activity in EGTA. The Km of actinactivation decreases in calcium, an effect separate from calmodulin dissociation. This second effect of calcium is responsible for the greater activity of myosin I at 20 [M actin in calcium versus EGTA.

Readdition of calmodulin (5-10 ,u m ) completely restores MgATPase activity even in buffers of high calcium (1 mM), demonstrating that inhibition results from calmodulin dissociation rather than calcium-binding (Fig. 3). By determining the stoichiometry of 110xl03Mr and 18xl03Mr myosin I subunits in calcium and EGTA, we find that three calmodulin molecules are bound per heavy chain in EGTA while only two remain associated after repurification of the complex in calcium. These molar ratios of 110xl03Mr to 18xl03Mr polypeptides were derived from comparison of integrated peak areas of subunits fractionated by gel filtration in guanidine, monitoring peptide bond absorbance (unpublished obser­ vations). The affinity of the dissociable calmodulin for the heavy chain can be determined from the calmodulinconcentration dependence of reactivation of myosin I MgATPase activity. The regulatory calmodulin binds to the myosin I heavy chain with an affinity in the nanomolar range in EGTA, but in calcium this affinity decreases to micromolar (unpublished observations). High calcium also completely inhibits brush border myosin I motility (Collins et al. 1990). In the coverslip motility assay, only a small number of active myosin heads are necessary to promote the maximum rate of motility if inactive heads do not act as a brake. Brush border myosin I in high calcium still binds actin but releases it in the presence of ATP. Therefore, myosin I-mediated motility is not inhibited in vitro until complete dissociation of the regulatory calmodulin subunit has occurred, at 100 ¡j m calcium. Movement ceases abruptly rather than gradually with increasing calcium. Inhibition of motility, as for MgATPase activity, is reversible with an excess of calmodulin. The direct calcium inhibition of brush border myosin I is significant in its contrast with the regulation of myosin II Regulation o f vertebrate myosins

13

and also in its mechanism. The reversible dissociation of a calmodulin from the complex provides a model system for studying the coupling of MgATP hydrolysis and translo­ cation in the myosin kinetic cycle. Brush border myosin I without regulatory calmodulin still binds actin in an ATPsensitive manner, but does not translocate filaments. It still binds and hydrolyzes MgATP, dependent on the association of actin, but at a 10-fold lower rate and independent of actin concentration between 0.5 and 60 /jm . Therefore, although the actin- and ATP-binding sites are still functional and active, the cycle of MgATP hydrolysis coupled to force production has been eliminated. Removal of scallop myosin II light chain has a similar effect: motility is reversibly inactivated and MgATPase activity becomes deregulated, in this case constitutively high (Vale et al. 1984). These induced deregulations of the myosin I and II mechanoenzymes suggest that even for myosins with divergent heavy- and light-chain structures, the association of regulatory light chain or its equivalent may coordinate heavy chain activities. Because the mechanism of energy coupling appears to be conserved between myosins, the more easily reversible inhibition of myosin I makes it an excellent model system to study this basic feature of motor proteins.

Tropomyosin regulation The mechanisms of regulation described above rely on modification of myosin structure and thus are termed myosin-linked forms of regulation. A second type of myosin regulation is mediated by actin structure, in particular by the actin-binding protein tropomyosin. Actin-linked regulation is most thoroughly characterized in skeletal muscle, where tropomyosin and the tropomyo­ sin-associated troponin complex saturate the actin of thin filaments. Troponin-tropomyosin inhibits myosin cross­ bridge cycling on actin in low calcium, maintaining the relaxed state. An increase in calcium concentration induces a conformational change in the troponin—tropo­ myosin complex which activates myosin-mediated ATP hydrolysis and muscle contraction. Isoforms of tropomyo­ sin also exist in smooth and non-muscle cells, but the significance of actin-linked regulation in these cells is less clear. These tropomyosin isoforms stimulate rather than inhibit smooth and non-muscle myosin II motility and MgATPase activity by 2- to 3-fold (Umemoto et al. 1989). To determine if actin-linked as well as myosin-linked regulation differentiates vertebrate myosins I and II, we tested the effect of smooth (turkey gizzard) and non­ muscle (chicken fibroblast, chicken brush border) isoforms of tropomyosin on the activities of chicken brush border myosin I (K. Collins, J. Sellers, P. Matsudaira, manuscript in preparation). Although these isoforms activate smooth and non-muscle myosin II actin-activated MgATPase activity and motility, they inhibit brush border myosin I. For smooth muscle tropomyosin (a high molecular weight, high actin-affinity tropomyosin isoform), the actin-acti­ vated MgATPase activity of myosin I decreases with increasing tropomyosin concentration until a saturating ratio of 1 tropomyosin dimer per 7 actin monomers is reached. Brush border tropomyosin, a low molecular weight tropomyosin with a lower affinity for actin under the assay conditions, also inhibits myosin I MgATPase activity, but a higher concentration of tropomyosin is required for the same effect. The rate of MgATP hydrolysis at maximal inhibition is approximately 15-20% of the uninhibited rate. This residual MgATPase activity, as 14

K. Collins and P. Matsudaira

with the MgATPase activity remaining after calcium inhibition, is not coupled to motility. In the motility assay, actin filaments which are bound to the myosin-coated coverslip under conditions of tropomyosin saturation do not move at all, to within the error of our measurements (0.004 /an s"1, 20-fold less than normal velocity). We also assayed the dependence of brush border myosin I actin-binding on tropomyosin. Saturating concentrations of smooth and non-muscle tropomyosin inhibit the affinity of myosin I for actin by 25-50 % in the presence of ATP. This binding inhibition is not sufficient to explain the inhibition of MgATPase activity and motility. Therefore, as is the case with inhibition of skeletal myosin by skeletal tropomyosin, smooth and non-muscle tropomyosins most likely inhibit myosin I MgATPase activity and motility by inhibiting a kinetic step of the cross-bridge cycle. Although the inhibition of myosin I actin-binding by tropomyosin is not responsible for the inhibition of myosin I MgATPase activity and motility, it may be relevant to myosin I localization in vivo. We find that the localization of myosin I in intestinal epithelial cells is consistent with a competition of myosin I and at least a low molecular weight isoform of tropomyosin for binding to actin (unpublished observations). In contrast, myosin II colocalizes with tropomyosin, perhaps with a high molecular weight tropomyosin isoform(s) in particular. Although myosin and tropomyosin localizations are consistent with an effect of tropomyosin on myosin distribution in vivo, myosin I also has membrane-binding activities not shared with myosin II which are equally as likely to contribute to differential localization.

Structural revelations Although vertebrate myosin I and smooth or non-muscle myosin IIs bind actin with similar affinity or display similar kinetics of MgATPase activity actin-activation, they are differentially regulated by tropomyosin. This suggests a divergence of fine structure in the otherwise conserved actin-binding site. The structure of the myosin II subfragment-1 (S-l)-tropomyosin-actin complex in the absence of ATP has been resolved to 2—3 nm by reconstruc­ tions from electron micrographs (Milligan and Flicker, 1987). Actin saturated with S-l in this fashion has a distinctive arrowhead appearance, due to the helical repeat of the complex. Brush border myosin I similarly decorates actin filaments (Coluccio and Bretscher, 1987; Fig. 4A). Reconstruction of the myosin I-actin complex will allow comparison of the myosin I and II actin-binding sites in the absence of ATP. In calcium, myosin I arrowheads have a different structure (Fig. 4B), half as wide and angled more sharply from the filament axis. This calcium-induced change could represent the loss of a calmodulin subunit or it could also reflect a different state of actin-myosin association, induced by the inhibition of force production. Reconstruction of the calcium and EGTA complexes should provide information about the location of the dissociable calmodulin and the extent of confor­ mational change in the rigor complex with inhibition of translocation.

Implications The regulation of brush border myosin I, as well as myosin I localization and structure, suggest that myosins I and II fulfill complementary functions in the cell. Due to the differential regulation of myosins I and II by calcium,

Fig. 4. Actin filaments decorated with brush border myosin I in EGTA (A) or calcium (B). An excess of myosin I was incubated with 0.5-1.0 /um actin in the standard myosin I purification buffer (Collins et al. 1990) to form decorated filaments. Calcium was added in excess of EGTA to the myosin-actin mixture in (B). Samples were stained with 2 % uranyl acetate on carboncoated 400-mesh grids and examined at 80 kV with a Phillips 410 electron microscope. Final magnification in (A) and (B) is approximately x 50 000.

myosin I would remain constitutively active in the low calcium concentrations predominant in cells in vivo, whereas myosin II, lacking light chain phosphorylation, would be predominantly inactive. Constitutive cellular motility processes include vesicle transport, membrane movement, cortical flow, and phagocytosis, events also implicated as myosin I functions on the basis of myosin I localization and the lack of requirement for myosin II. The increase in calcium concentration necessary to activate myosin II, occurring when a cell is stimulated to divide or contract, would inhibit myosin I. This inhibition of myosin I functions with myosin II activation would reduce the use of cell energy for housekeeping purposes during the period the cell is responding to a contractile stimulus. Calcium regulation partitions actin motility events between cell states; any cell state or compartment with functioning myosin I would not have active myosin II and vice versa. Differential regulation extends to actin-linked as well as myosin-linked mechanisms. The reciprocal effects of tropomyosin on myosins I and II suggests that even if both myosins could associate with the same actin population in vivo, the enzymes could not both be maximally active. Myosin I, however, does not demonstrate the apparent in vivo preference for actin structures containing tropomyo­ sin that myosin II does. As a consequence, myosin I would be associated with a less stable subset of actin filaments in the cell. Stabilized actin arrays may not be as necessary for myosin I function as for the contraction mediated by myosin II. Myosin I enzymes may serve a multiplicity of functions. The localization of brush border myosin I in the terminal web and at the apical and basolateral membranes of intestinal epithelial cells suggests that it could perform several different membrane-related tasks. Lower eukary­ otic myosin I also seems to be a diverse cellular workhorse: Dictyostelium and Acanthamoeba express a large number of myosin I isoforms but only one myosin II. The existence of multiple myosin I isoforms in vertebrates is hinted by cloning (Montell and Rubin, 1988) but deserves future experimental attention. Although the localization and in vitro activities of vertebrate myosin I suggest possible functions for this protein, definitive demonstration of the

role of myosin I in intact cells is difficult. Gene disruption and cell assays are more complex with vertebrate cells than with simpler organisms, but microinjection of antibodies which specifically inhibit myosin I-mediated motility in vitro (unpublished observations) might halt myosin I-requiring activities in the cell. A better under­ standing of the structure and composition of ‘motile’ cell regions will hopefully complement the investigation of purified motor proteins to determine the operating principles of cell motility.

C onclusions Vertebrate actin-based cell motility has diversified by the evolution of novel regulatory mechanisms for a single family of enzymes, the myosins. There is not yet evidence for a multiplicity of different classes of motor enzymes capable of translocating on actin, in contrast with the several different categories of microtubule-translocating enzymes that have been discovered (Gelfand, 1989). Myosin I motors may perform a diversity of actin-based motility events previously ascribed to myosin II, particu­ larly functions involving membranes or membraneassociated actin. The differential regulation of myosins I and II provides a mechanism for diversification of the two enzymes, and suggests a balance of myosin I-mediated constitutive motor activities with myosin II-mediated contractile events induced by specific stimuli or cell states. Future study of myosin I will provide new insights into both the cellular and molecular bases of motility. This work was funded by grants from the American Heart Association, The Pew Foundation, and the Markey Foundation.

References R. J. a n d P o l l a r d , T. D. (1986). Propulsion of organelles isolated from Acanthamoeba along actin filaments by myosin-I. Nature 322, 754-756. A d a m s , R. J. a n d P o l l a r d , T. D. (1989). Binding of myosin I to membrane lipids. Nature 340. 565-568.

A

d am s,

Regulation o f vertebrate myosins

15

J. H. a n d B o r y s e n k o , C . W. (1984). The 110,000-dalton actinand calmodulin-binding protein from intestinal brush border is a myosin-like ATPase. J. biol. Chem. 259, 14128-14135. C o l l i n s , K., S e l l e r s , J. R. a n d M a t s u d a i r a , P. (1990). Calmodulin dissociation regulates brush border myosin I (110-kD-calmodulin) mechanochemical activity in vitro. J. Cell Biol. 110, 1137—1147. C o l u c c i o , L . M . a n d B r e t s c h e r , A. (1987). Calcium-regulated cooperative binding of the microvillar 110K-calmodulin complex to Factin: formation of decorated filaments. J. Cell Biol. 105, 325-333. C o u d r i e r , E., R e g g i o , H. a n d L o u v a r d , D. (1981). Immunolocalization of the 110,000 molecular weight cytoskeletal protein of intestinal microvilli. J. molec. Biol. 152, 49-66. De L o z a n n e , A. a n d S p u d i c h , J. A. (1987). Disruption of the Dictyostelium myosin heavy chain gene by homologous recombination. Science 236, 1086-1091. F u k u i , Y., L y n c h , T. J., B r z e s k a , H. a n d K o r n , E. D. ( 1 9 8 9 ) . Myosin I is located at the leading edges of locomoting Dictyostelium amoebae. Nature 341, 3 2 8 - 3 3 1 . G a d a s i , H. a n d K o r n , E. D. (1980). Evidence for differential intracellular localization of the Acanthamoeba myosin isoenzymes. Nature 286, 452-456. G a r c i a , A . , C o u d r i e r , E., C a r b o n i , J., A n d e r s o n , J., V a n d e r k h o v e , J., M o o s e k e r , M . , L o u v a r d , D. a n d A r p i n , M . (1989). Partial deduced sequence of the 110-kD-calmodulin complex of the avian intestinal microvillus shows that this mechanoenzyme is a member of the myosin I family. J. Cell Biol. 109, 2895—2903. G e l f a n d , V. I. (1989). Cytoplasmic microtubular motors. Current Opinion in Cell Biology 1, 63-66. H o s h i m a r u , M., F u j i o , Y., S o b u e , K., S u g i m o t o , T. a n d N a k a n i s h i , S . (1989). Immunochemical evidence that myosin I heavy chain-like protein is identical to the 110-kilodalton brush-border protein. J. Biochem. 106, 455-459. H o s h i m a r u , M. a n d N a k a n i s h i , S. (1987). Identification of a new type of mammalian myosin heavy chain by molecular cloning: overlap of its m R N A with preprotachykinin B m R N A . J. biol. Chem. 262, 14625-14632. J u n g , G., K o r n , E. D. a n d H a m m e r , J . A. III. (1987). The heavy chain of Acanthamoeba myosin IB is a fusion of myosin-like and non­ myosin-like sequences. Proc. natn. Acad. Sci. U.S.A. 84, 6720—6724. J u n g , G., S a x e , C. L. Ill, K i m m e l , A. R. a n d H a m m e r , J . A. III. (1989). Dictyostelium discoideum contains a gene encoding a myosin I heavy chain. Proc. natn. Acad. Sci. U.S.A. 86, 6186-6190. K e l l e r , T. C . S . Ill, C o n z e l m a n , K . A . , C h a s a n , R. a n d M o o s e k e r . M . S . (1985). Role of myosin in terminal web contraction in isolated intestinal epithelial brush borders. J. Cell Biol. 100, 1647-1655. K e l l e r , T. C. S . Ill a n d M o o s e k e r , M . S . (1982). Ca+^calmodulindependent phosphorylation of myosin, and its role in brush border contraction in vitro. J. Cell Biol. 95, 943—959. K n e c h t , D. A. a n d L o o m i s , W. F. (1987). Antisense RNA inactivation of C o llin s ,

16

K. Collins and P. Matsudaira

myosin heavy chain gene expression in Dictyostelium discoideum. Science 236, 1081—1086. K o r n , E. D., A t k i n s o n , M . A . L ., B r z e s k a , H . , H a m m e r , J . A . I l l , J u n g , G . a n d L y n c h , T. J . (1988). Structure— function studies on Acanthamoeba myosins IA, IB, and II. J. Cell Biochem. 36, 37-50. K r o n , S . J. a n d J. A . S p u d i c h . (1986). Fluorescent actin filaments move on myosin fixed to a glass surface. Proc. natn. Acad. Sci. U.S.A. 83, 6272-6276. L o u v a r d , D. (1989). The function of the major cytoskeletal components of the brush border. Current Opinion in Cell Biology 1, 51—57. L y n c h , T. J., A l b a n e s i , J. P., K o r n , E. D., R o b i n s o n , E. A . , B o w e r s , B . a n d F u j i s a k i , H . (1986). ATPase activities and actin-binding properties of subfragments of Acanthamoeba myosin I A . J. biol. Chem. 261, 17 156-17162. M a b u c h i , I. a n d O k u n o , M . (1977). The effect of myosin antibody on the division of starfish blastomeres. J. Cell Biol. 74, 251-263. M a t s u d a i r a , P . T. a n d B u r g e s s , D. R . (1979). Identification and organization of the components in the isolated microvillus cytoskeleton. J. Cell Biol. 83, 667—673. M i l l i g a n , R . A . a n d F l i c k e r , P . F. (1987). Structural relationships of actin, myosin, and tropomyosin revealed by cryo-electron microscopy. J. Cell Biol 105, 29-39. M o n t e l l , C. a n d R u b i n , G . M . (1988). The Drosophila ninaC locus encodes two photoreceptor cell specific proteins with domains homologous to protein kinases and the myosin heavy chain head. Cell 52, 757-772. M o o s e k e r , M . S . (1985). Organization, chemistry, and assembly of the cytoskeletal apparatus of the intestinal brush border. A. Rev. Cell Biol. 1, 209-241. . M o o s e k e r , M . S. a n d C o l e m a n , T. R . (1989). The 110-kD proteincalmodulin complex of the intestinal microvillus (brush border myosin I) is a mechanoenzyme. J. Cell Biol. 108, 2395-2400. M o o s e k e r , M . S ., P o l l a r d , T. D. a n d F u j i w a r a , K . (1978). Characterization and localization of myosin in the brush border of intestinal epithelial cells. J. Cell Biol. 79, 444-453. P o l l a r d , T. D. a n d K o r n , E. D. (1973). Acanthamoeba myosin: isolation from Acanthamoeba castellanii of an enzyme similar to muscle myosin. J. biol. Chem. 248, 4682-4690. S h e e t z , M. P., C h a s a n , R . a n d S p u d i c h , J . A. (1984). ATP-dependent movement of myosin in vitro: characterization of a quantitative assay. J. Cell Biol. 99, 1867-1871. U m e m o t o , S ., B e n g u r , A. R . a n d S e l l e r s , J . R . (1989). Effect of multiple phosphorylations of smooth muscle and cytoplasmic myosins on movement in an in vitro motility assay. J. biol. Chem. 264, 1431-1436. V a l e , R . D., S z e n t - G y o r g y i , A. G . a n d S h e e t z , M. P . (1984). Movement of scallop myosin on Nitella actin filaments: regulation by calcium. Proc. natn. Acad. Sci. U.S.A. 81, 6775-6778. W a r r i c k , H . M. a n d S p u d i c h , J . A. (1987). Myosin structure and function in cell motility. A. Rev. Cell Biol. 3, 379-421.

Differential regulation of vertebrate myosins I and II.

Cell motility events require movement of the cytoskeleton. Actin-based movement is catalyzed by the mechanoenzyme myosin, which translocates toward th...
10MB Sizes 0 Downloads 0 Views