Interaction of microtubule-associated protein 2 with actin filaments

The interaction of unphosphorylated and phosphorylated microtubule-associated protein 2 (MAP-2) with actin filaments was examined by electron microscopic, electrophoretic, and dark-field light microscopic techniques. Unphosphorylated MAP-2 was observed to cross-link and bundle individual actin filaments. Chymotryptic fragments of MAP-2 protein were produced which bound to, but could not cross-link, actin polymer; these fragments encompassed the tubulin binding domain of MAP-2. The phosphorylation of intact MAP-2, by means of endogenous protein kinases, inhibited the ability of this molecule to cross-link and bundle actin filaments. Phosphorylation did not, however, inhibit the binding of MAP-2 to F-actin. The chymotryptic fragments of phosphorylated MAP-2 that retained their ability to bind to actin and promote microtubule assembly also encompassed the tubulin binding domain of this molecule. An analysis of MAP-2 fragments by nonequilibrium pH gradient electrophoresis indicated that most of the polypeptide backbone is relatively acidic with the exception of the tubulin binding domain. This region was determined to be the most basic (positively charged) region of the MAP-2 molecule. Biochemical and morphological evidence is presented to demonstrate that both unphosphorylated MAP-2 and phosphorylated MAP-2 have the capacity to use actin, in addition to microtubules, as a separate anchoring substrate. The presence of tubulin, however, strongly inhibits the interaction of MAP-2 with actin filaments.     

The interaction of actin filaments with microtubules and microtubule-associated proteins

Purified actin and microtubule proteins polymerized together form a gel, while mixtures of actin with tubulin polymers lacking microtubule-associated proteins (MAPs) have low viscosities close to the sum of the viscosities of the constituents. Mixtures of actin and MAPs also have high viscosities. Our interpretation of these observations was that there is interaction of actin filaments and microtubules which is mediated by MAPs (Griffith, L. M., and Pollard, T. D. (1978) J. Cell Biol. 78, 958-965). We report here further evidence for this interaction. 1) Actin filaments and microtubules can form gels at physiological ionic strength providing the anion is glutamate rather than chloride. Both glutamate and chloride inhibit actin-MAPs interaction, but this is compensated for in glutamate where the microtubules are longer than in chloride. 2) The low shear viscosity of mixtures of isolated MAPs and actin filaments is enhanced by acidic pH and inhibited by high ionic strength. 3) MAPs can be fractionated to yield four different fractions with actin cross-linking activity: a subset of high molecular weight MAPs, purified "MAP-2" and two different fractions of tau polypeptides. 4) We have reconstituted a gel from actin, purified tubulin, and whole MAPs, but have not yet been successful with actin, purified tubulin, and any single purified MAP.     

Phosphorylation of microtubule-associated proteins regulates their interaction with actin filaments

We have determined the absolute phosphate content of microtubule-associated proteins (MAPs) and established that phosphorylation inhibits the actin filament cross-linking activity of MAPs and both of the major MAP components, MAP-2 and tau. Similar results were obtained with actin from rabbit muscle, hog brain, and Acanthamoeba castellanii. We used the endogenous phosphatases and kinases in hog brain microtubule protein to modulate MAP phosphate level before isolating heat-stable MAPs. MAPs isolated directly from twice-cycled microtubule protein contain 7.1 +/- 0.1 (S.E.) mol of phosphate/300,000 g protein. After incubating microtubule protein without ATP, MAPs, had 4.9 +/- 0.6 phosphates. After incubating microtubule protein with 1 mM ATP and 5 microM cAMP in 2 mM EGTA, MAPs had 8.6 +/- 0.5 phosphates but there was also exchange of three more [32P]phosphates from gamma-labeled ATP for preexisting MAP phosphate. Incubation of microtubule protein with ATP and cAMP in 5 mM CaCl2 resulted in exchange but no net addition of phosphate to MAPs. We fractionated the MAP preparations by gel filtration and obtained MAP-2 with 4.3 to 7.5 and tau with 1.5 to 2.2 mol of phosphate/mol of protein depending on how we treated the microtubule protein prior to MAP isolation. The actin filament cross-linking activity of whole MAPs, MAP-2, and tau depended on the MAP-phosphate content. In all cases, phosphorylation of MAPs inhibited actin filament cross-linking activity. The concentration of high phosphate MAPs required to form a high viscosity solution with actin filaments was 2 to 4 times more than that of low phosphate. MAPs. During incubation of microtubule protein with [gamma-32P]ATP, only MAP peptides are labeled. Treatment of these MAPs with either acid or alkaline phosphatase removes phosphate mainly from MAP-2, with an increase in actin filament cross-linking activity. Thus, both MAP phosphorylation and the effect of phosphorylation on actin cross-linking activity of MAPs are reversible.     

Interaction of tropomyosin-troponin with actin filaments

The assembly of actin filaments with tropomyosin-troponin was investigated by means of light scattering. Binding curves of tropomyosin-troponin [consisting of all three subunits (holotroponin)] and of tropomyosin-troponin-T-I to actin filaments were analyzed by separating the affinity of tropomyosin-troponin for actin filaments and the affinity for the end-to-end contact of tropomyosin molecules. Under the experimental conditions (42.4 degrees C, 300 mM KCl), tropomyosin-holotroponin in the absence of calcium and tropomyosin-troponin-T-I had similar affinities for actin filaments whereas tropomyosin-holotroponin in the presence of calcium was found to bind more weakly. Tropomyosin-holotroponin and tropomyosin-troponin-T-I bound about 200-300-fold more strongly to binding sites with adjacent tropomyosin-troponin units than to isolated sites on actin filaments. The equilibrium constant for isolated association with actin filaments was more than 2-fold higher for tropomyosin-holotroponin in the absence of calcium (15 400 M-1) and tropomyosin-troponin-T-I (17 500 M-1) than for tropomyosin-holotroponin in the presence of calcium (6600 M-1). Binding curves of mixtures of tropomyosin-holotroponin in the presence of calcium and of tropomyosin-troponin-T-I were measured and analyzed on the basis of a model of cooperative binding of two types of large ligands to a one-dimensional homogeneous lattice. The results provided information on the strength of the end-to-end contacts of tropomyosin-troponin units in different positions on an actin filament. It was found that a tropomyosin-troponin unit binds adjacently to another unit in a different position on an actin filament about 2-fold more weakly than adjacent to a unit in the same position. With the aid of these results, it was possible to obtain information of the equilibrium distribution of tropomyosin-troponin in the two positions on actin filaments. Generation of a sequence of tropomyosin-troponin units in a different position on actin filaments was found to be 4-fold less favored than elongation of an existing sequence (cooperativity parameter sigma = 1/4). Shifting of tropomyosin-troponin on actin filaments appears to be accompanied by small free-energy changes in the various interactions of the components of actin-tropomyosin-troponin filaments and not to be an all-or-none reaction     

Interaction of ADP-ribosylated actin with actin binding proteins.

Actin ADP-ribosylated at Arg177 was previously shown not to polymerise after increasing the ionic strength, but to cap the barbed ends of filaments. Here we confirm that the polymerisation of ADP-ribosylated actin is inhibited, however, under specific conditions the modified actin copolymerises with native actin, indicating that its ability to take part in normal subunit interactions within filaments is not fully eliminated. We also show that ADP-ribosylated actin forms antiparallel but not parallel dimers: the former are not able to form filaments. ADP-ribosylated actin interacts with deoxyribonuclease I, vitamin D binding protein, thymosin beta(4), cofilin and gelsolin segment 1 like native actin. Interaction with myosin subfragment 1 revealed that the potential of the modified actin to aggregate into oligomers or short filaments is not fully eliminated.     

Interaction of estramustine phosphate with microtubule-associated proteins

We have reported [(1984) Cancer Res., in press] that estramustine phosphate inhibits microtubule assembly and disassembled preformed microtubules. We now present evidence that estramustine phosphate inhibits microtubule assembly by binding to the microtubule-associated proteins. We have found that: additional microtubule-associated proteins relieved the inhibition of assembly by estramustine phosphate; 3H-labelled estramustine phosphate bound predominantly to the microtubule-associated proteins; and the content of the microtubule-associated proteins was reduced in taxol reversed estramustine phosphate-inhibited microtubules.     

Ca2+ control of actin gelation. Interaction of gelsolin with actin filaments and regulation of actin gelation

We elucidated the mechanism by which gelsolin, a Ca2+-dependent regulatory protein from lung macrophages, controls the network structure of actin filaments. In the presence of micromolar Ca2+, gelsolin bound Ca2+. The Ca2+-gelsolin complex reduced the apparent viscosity and flow birefringence of F-actin and the lengths of actin filaments viewed in the electron microscope. However, concentrations of gelsolin causing these alterations did not effect proportionate changes in the turbidity of actin filament solutions or in the quantity of nonsedimentable actin as determined by a radioassay. From these findings, we conclude that gelsolin shortens actin filaments without net depolymerization. Such an effect on the distribution of actin filament lengths led to the prediction that low concentrations of gelsolin would increase the critical concentration of actin-binding protein required for incipient gelation of actin filaments in the presence of Ca2+, providing an efficient mechanism for controlling actin network structure. We verified the prediction experimentally, and we estimated that the Ca2+-gelsolin complex effectively breaks the bond between actin monomers in filaments with a stoichiometry of 1:1. The effect of Ca2+-gelsolin complex on actin solation was rapid, independent of temperature between 0 degrees and 37 degrees C, and reversed by reducing the free Ca2+ concentration.     

Regulation of the interaction of actin filaments with microtubule-associated protein 2 by calmodulin-dependent protein kinase II

Phosphorylation of microtubule-associated protein 2 (MAP 2) by Ca²⁺-, calmodulin-dependent protein kinase II (protein kinase II) inhibited the actin filament cross-linking activity of MAP 2. This inhibition required the presence of ATP, Mg²⁺, Ca²⁺ and calmodulin. The minimal concentration of MAP 2 required for gel formation of actin filaments was increased with increasing amounts of phosphate incorporated into MAP 2, and the phosphorylated MAP 2, into which 10.3 mol of phosphate/mol of protein had been incorporated, did not cause actin filaments to gel under the experimental conditions used. The phosphorylation of MAP 2 by Ca²⁺-, phospholipid-dependent protein kinase (protein kinase C) and cAMP-dependent protein kinase also inhibited the actin filament cross-linking activity of MAP 2. The extent and rate of phosphorylation of MAP 2 by protein kinase II were higher than those of the phosphorylation by protein kinase C and cAMP-dependent protein kinase. The interaction of actin filaments with MAP 2 was inhibited more by the actions of protein kinase II and protein kinase C than by cAMP-dependent protein kinase. The actin filament cross-linking activity of MAP 2 phosphorylated either by protein kinase II, cAMP-dependent protein kinase or protein kinase C was retrieved when phosphorylated MAP 2 was treated by protein phosphatase. These results indicate that the interaction of actin filaments with MAP 2 is regulated by the phosphorylation-dephosphorylation of MAP 2.     

Evidence for actin filament-microtubule interaction mediated by microtubule-associated proteins

We have used low shear viscometry and electron microscopy to study the interaction between pure actin filaments and microtubules. Mixtures of microtubules having microtubule-associated proteins (MAPs) with actin filament have very high viscosities compared with the viscosities of the separate components. MAPs themselves also cause a large increase in the viscosity of actin filaments. In contrast, mixtures of actin filaments with tubulin polymers lacking MAPs have low viscosities, close to the sum of the viscosities of the separate components. Our interpretation of these observations is that there is an interaction between actin filaments and microtubules which requires MAPs. This interaction is inhibited by ATP and some related compounds. Electron micrographs of thin sections through mixtures of actin and microtubules show numerous close associations between the two polymers which may be responsible for their high viscosity.     

Interaction of lipid-bound myelin basic protein with actin filaments and calmodulin

Myelin basic protein (MBP) binds to negatively charged lipids on the cytosolic surface of oligodendrocytes (OLs) and is believed to be responsible for adhesion of these surfaces in the multilayered myelin sheath. MBP in solution has been shown by others to bind to both G- and F-actin, to bundle F-actin filaments, and to induce polymerization of G-actin. Here we show that MBP bound to acidic lipids can also bind to both G- and F-actin and cause their sedimentation together with MBP-lipid vesicles. Thus it can simultaneously utilize some of its basic residues to bind to the lipid bilayer and some to bind to actin. The amount of actin bound to the MBP-lipid vesicles decreased with increasing net negative surface charge of the lipid vesicles. It was also less for vesicles containing the lipid composition predicted for the cytosolic surface of myelin than for PC vesicles containing a similar amount of an acidic lipid. Calmodulin caused dissociation of actin from MBP and of the MBP-actin complex from the vesicles. However, it did not cause dissociation of bundles of actin filaments once these had formed as long as some MBP was still present. These results suggest that MBP could be a membrane actin-binding protein in OLs/myelin and its actin binding can be regulated by calmodulin and by the lipid composition of the membrane. Actin binding to MBP decreased the labeling of MBP by the hydrophobic photolabel 3-(trifluoromethyl)-3-(m-[(125)I]iodophenyl)diazirine (TID), indicating that it decreased the hydrophobic interactions of MBP with the bilayer. This change in interaction of MBP with the bilayer could then create a cytosol to membrane signal caused by changes in interaction of the cytoskeleton with the membrane.     

Multiscale simulation of actin filaments and actin-associated proteins

Actin is an important cytoskeletal protein that serves as a building block to form filament networks that span across the cell. These networks are orchestrated by a myriad of other cytoskeletal entities including the unbranched filament–forming protein formin and branched network–forming protein complex Arp2/3. Computational models have been able to provide insights into many important structural transitions that are involved in forming these networks, and into the nature of interactions essential for actin filament formation and for regulating the behavior of actin-associated proteins. In this review, we summarize a subset of such models that focus on the atomistic features and those that can integrate atomistic features into a larger picture in a multiscale fashion.     

Microtubule-binding proteins CLASP1 and CLASP2 interact with actin filaments

Cell morphogenesis requires dynamic communication between actin filaments and microtubules which is mediated, at least in part, by direct structural links between the two cytoskeletal systems. Here, we examined interaction between the CLIP-associated proteins (CLASP) CLASP1 and CLASP2, and actin filaments. We demonstrate that, in addition to a well-established association with the distal ends of microtubules, CLASP2alpha co-localizes with stress fibers, and that both CLASP1alpha and CLASP2alpha co-immunoprecipitate with actin. GFP-CLASP2alpha exhibits retrograde flow in the lamellipodia of Xenopus primary fibroblasts and in the filopodia of Xenopus spinal cord neurons. A deletion mapping analysis reveals that both the microtubule-binding domain of CLASP2 (which is homologous between all CLASPs) and the N-terminal dis1/TOG domain of CLASP2alpha (which is homologous between alpha isoforms) possess actin-binding activity. Fluorescence resonance energy transfer experiments demonstrate significant energy transfer between YFP-CLASP2alpha and CFP-actin. Our results indicate that CLASPs function as actin/microtubule crosslinkers in interphase cells. We propose that CLASPs facilitate recognition of actin filaments by the plus ends of growing microtubules at the initial stages of actin-microtubule interaction. Cell Motil.     

Interaction of cytochalasin D with actin filaments in the presence of ADP and ATP

Cytochalasin D strongly inhibits the faster components in the reactions of actin filament depolymerization and elongation in the presence of 10 mM Tris-Cl-, pH 7.8, 0.2 mM dithiothreitol, 1 mM MgCl2, 0.1 mM CaCl2, and 0.2 mM ATP or ADP. Assuming an exclusive and total capping of the barbed end by the drug, the kinetic parameters derived at saturation by cytochalasin D refer to the pointed end and are 10-15-fold lower than at the barbed end. In ATP, the critical concentration increases with cytochalasin D up to 12-fold its value when both ends are free; as a result of the lowering of the free energy of nucleation by cytochalasin D, short oligomers of F-actin exist just above and below the critical concentration. Cytochalasin D interacts strongly with the barbed ends independently of the ADP-G-actin concentration (K = 0.5 nM-1). In contrast, the affinity of cytochalasin D decreases cooperatively with increasing ATP-G-actin concentration. These data are equally well accounted for by two different models: either cytochalasin D binds very poorly to ATP-capped filament ends whose proportion increases with actin concentration, or cytochalasin D binds equally well to ATP-ends and ADP-ends and also binds to actin dimers in ATP but not in ADP. A linear actin concentration dependence of the rate of growth was found at the pointed end, consistent with the virtual absence of an ATP cap at that end.     

Non-motor microtubule-associated proteins.

Cloning of primary sequences has generated information on the structures of the non-motor microtubule-associated proteins and their relationship to one another. Questions about how classes of microtubule-associated proteins interact are starting to be addressed in vitro and, in vivo, tests of function are being pursued using a variety of cellular and molecular biological strategies.     

Polyphosphoinositides inhibit the interaction of vinculin with actin filaments.

Binding of vinculin to adhesion plaque proteins is restricted by an intramolecular association of vinculin's head and tail regions. Results of previous work suggest that polyphosphoinositides disrupt this interaction and thereby promote binding of vinculin to both talin and actin. However, data presented here show that phosphatidylinositol 4,5-bisphosphate (PI4,5P2) inhibits the interaction of purified tail domain with F-actin. Upon re-examining the effect of PI4,5P2 on the actin and talin-binding activities of intact vinculin, we find that when the experimental design controls for the effect of magnesium on aggregation of PI4,5P2 micelles, polyphosphoinositides promote interactions with the talin-binding domain, but block interactions of the actin-binding domain. In contrast, if vinculin is trapped in an open confirmation by a peptide specific for the talin-binding domain of vinculin, actin binding is allowed. These results demonstrate that activation of the actin-binding activity of vinculin requires steps other than or in addition to the binding of PI4,5P2.     

Mechanically tuning actin filaments to modulate the action of actin-binding proteins

In cells, the actin cytoskeleton is regulated by an interplay between mechanics and biochemistry. A key mechanism, which has emerged based on converging indications from structural, cellular, and biophysical data, depicts the actin filament as a mechanically tunable substrate: mechanical stress applied to an actin filament induces conformational changes, which modify the binding and the regulatory action of actin-binding proteins. For a long time, however, direct evidence of this mechanotransductive mechanism was very scarce. This situation is changing rapidly, and recent in vitro single-filament studies using different techniques have revealed that several actin-binding proteins are able to sense tension, curvature, and/or torsion, applied to actin filaments. Here, we discuss these recent advances and their possible implications.     

Kinetics of the cooperative association of actin to actin filaments.

The cooperative formation of actin filaments from monomers was followed by light scattering and electron microscopy. The results are well described by a simple model mechanism in which the growth and destruction of filaments occurs by stepwise addition or dissociation of protomers. All steps except the dimerisation step are assumed to have identical rate constants. These were found to be 5 X 10(3) M-1 - sec-1 and 3 X 10(-2) sec-1 for the association and dissociation, respectively (at pH 7.5 and in the presence of 10(-3) M calcium chloride). The equilibrium constant of elongation as obtained from the critical concentration is 1.7 X 10(5) M-1. The corresponding equilibrium constant of dimerisation is about 10 million times smaller (cooperativity parameter sigma = 2 X 10(-7)). This makes the nucleation extremely difficult and cooperativity very high. A best fit of the model to the experimental data is achieved when the destruction of a dimer is much faster than the addition of a third protomer (fast monomer- dimer pre-equilibrium). The size of the nucleus from which propagation becomes faster than dissociation is 3.     

Interaction of a Dictyostelium member of the plastin/fimbrin family with actin filaments and actin-myosin complexes.

A protein purified from cytoskeletal fractions of Dictyostelium discoideum proved to be a member of the fimbrin/plastin family of actin-bundling proteins. Like other family members, this Ca(2+)-inhibited 67-kDa protein contains two EF hands followed by two actin-binding sites of the alpha-actinin/beta-spectrin type. Dd plastin interacted selectively with actin isoforms: it bound to D. discoideum actin and to beta/gamma-actin from bovine spleen but not to alpha-actin from rabbit skeletal muscle. Immunofluorescence labeling of growth phase cells showed accumulation of Dd plastin in cortical structures associated with cell surface extensions. In the elongated, streaming cells of the early aggregation stage, Dd plastin was enriched in the front regions. To examine how the bundled actin filaments behave in myosin II-driven motility, complexes of F-actin and Dd plastin were bound to immobilized heavy meromyosin, and motility was started by photoactivating caged ATP. Actin filaments were immediately propelled out of bundles or even larger aggregates and moved on the myosin as separate filaments. This result shows that myosin can disperse an actin network when it acts as a motor and sheds light on the dynamics of protein-protein interactions in the cortex of a motile cell where myosin II and Dd plastin are simultaneously present.     

Formation and destabilization of actin filaments with tetramethylrhodamine-modified actin

Actin labeling at Cys(374) with tethramethylrhodamine derivatives (TMR-actin) has been widely used for direct observation of the in vitro filaments growth, branching, and treadmilling, as well as for the in vivo visualization of actin cytoskeleton. The advantage of TMR-actin is that it does not lock actin in filaments (as rhodamine-phalloidin does), possibly allowing for its use in investigating the dynamic assembly behavior of actin polymers. Although it is established that TMR-actin alone is polymerization incompetent, the impact of its copolymerization with unlabeled actin on filament structure and dynamics has not been tested yet. In this study, we show that TMR-actin perturbs the filaments structure when copolymerized with unlabeled actin; the resulting filaments are more fragile and shorter than the control filaments. Due to the increased severing of copolymer filaments, TMR-actin accelerates the polymerization of unlabeled actin in solution also at mole ratios lower than those used in most fluorescence microscopy experiments. The destabilizing and severing effect of TMR-actin is countered by filament stabilizing factors, phalloidin, S1, and tropomyosin. These results point to an analogy between the effects of TMR-actin and severing proteins on F-actin, and imply that TMR-actin may be inappropriate for investigations of actin filaments dynamics.