Search

Neurological complications as well as movement disorders are relevant symptoms in animals and humans chronically exposed to dithiocarbamates. Using rat pheochromocytoma cells differentiated by NGF (PC12), we investigated whether propineb affects acetylcholine (Ach) release and the molecular mechanisms involved. Propineb (0.001-100 nM) dose-dependently increased Ach release from PC12. Thus, 0.001-1 nM propineb-induced Ach release, reaching a maximal effect ( approximately 50%) at 0.1-1 nM. Higher concentrations of propineb (10-100 nM) caused a progressive disappearance of the effect. Chelation of extra- and intracellular Ca(2+) did not affect Ach release by propineb, which was prevented by the actin stabilizer jasplakinolide (500 nM). Accordingly, actin depolymerization was observed after exposure of differentiated PC12 to 0.1-1 nM propineb, a loss of effect was evident at higher concentrations (100 nM), and the effect was Ca(2+)-independent. Disulfiram, a related dithiocarbamate not coordinated with Zn(2+), also depolymerized actin, suggesting the involvement of the organic structure of dithiocarbamates rather than the leakage of Zn(2+). Nevertheless, propineb did not depolymerize actin in a cell-free system. These data suggest that dithiocarbamates, through the activation of intracellular cascade(s), impair cytoskeletal actin. This effect may contribute to affect synaptic vesicles processing resulting in an impaired cholinergic transmission.

The properties of some models of the actin filament are compared with those of the thin filament in muscle. The greater stiffness of thin filaments ex vivo with respect to F-actin in vitro is attributed to the effect of both protein osmotic pressure and the attached cross-bridges. By comparing the stiffness of thin filaments in vitro and in isometric and rigor muscles the stiffness of thin filaments in relaxed muscle is computed. The upper limit of thin filament stretching is deduced to approach approximately 10 nm microm(-1). It is also calculated that, on stretching by 2.02 nm of the fully non-overlapped thin filament or by 1.59 nm of the thin filament on isometric contraction, the energy released on the hydrolysis of one molecule of ATP is fully used up.

Comparison of the behaviour of actin filaments either modified with tetramethylrhodamine iodoacetamide or decorated with tetramethylrhodamine-phalloidin or with tropomyosin or with myosin subfragment 1 shows that, in all the cases, yield strength is linearly related to stiffness.

Natural, phalloidin-free, actin filaments were decorated with tropomyosin made fluorescent by reaction with alexa fluor (R) 488 C(5) maleimide. The elastic modulus by stretching of these filaments was then determined and found to span between 38.2 MPa and 61.48 MPa. We tried also to determine the yield strength of the same filaments in the laser light trap operated at 920 mW, the maximum power of the apparatus. Only two out of the 10 filaments tested were broken under these conditions, yield strength being 50.5 and 55 pN, respectively.

The effects of coupling with tetramethylrhodamine-5-iodoacetamide and of the decoration with tropomyosin and with myosin subfragment 1 on the elastic properties of F-actin filament are investigated. At 22 degrees C, in 15 mM orthophosphate and 3 mM MgCl2, tetramethylrhodamine F-actin displays a yield strength of 3.69 +/- 0.213 pN and an elastic modulus by stretching of 0.91 MPa. Decoration with tropomyosin increases the yield strength of tetramethylrhodamine F-actin to 10.51 +/- 0.24 pN and the elastic modulus by stretching to 23-75 MPa. Mixtures of myosin subfragment 1 and tetramethylrhodamine F-actin at the 0.2:1, 0.4:1, 0.6:1, 0.8:1, and 1:1 molar ratios are also studied. Both yield strength and the elastic modulus by stretching are found to increase progressively with the ratio. At the 1:1 molar ratio, the yield strength is 15.81 +/- 0.26 pN and the elastic modulus by stretching is 13.45 to 40 MPa. Decoration of tetramethylrhodamine F-actin with both tropomyosin and myosin subfragment 1, at the 1:1 molar ratio with the actin monomer, produces filaments with an yield strength of 22.3 +/- 0.48 pN.

Fast muscle myosin responds in similar way to F-actin and to phalloidin F-actin. It is activated 7.5 fold at infinite F-actin concentration and 6.8 fold at infinite phalloidin F-actin. The actomyosin dissociation constants are 0.89 +/- 0.34 microM with F-actin and 0.90 +/- 0.71 microM with phalloidin F-actin. Slow muscle myosin responds differently to F-actin and to phalloidin F-actin. It is activated 3.76 fold at infinite F-actin concentration and only 2.27 fold at infinite phalloidin F-actin concentration. The actomyosin dissociation constants are 1.95 +/- 1.27 microM with F-actin and 0.27 +/- 0.16 microM with phalloidin F-actin. At first glance this means that substitution of F-actin with phalloidin F-actin magnifies the difference between fast muscle and slow muscle myosins. Furthermore the change of the dissociation constants may affect the contractile force of the attached crossbridge.

(Iodoacetamido)tetramethylrhodamine disrupts F-actin. At the 1:1 fluorophore to actin (as monomer) ratio approximately 80% of the protein becomes non-sedimentable. The fluorescent, non-sedimentable actin copolymerizes with G-actin to yield fluorescent filaments. The tensile strength of these filaments changes with the ratio of the fluorescent non-sedimentable actin to the G-actin, being 1.6 pN, 2.9 pN and 3.6 pN at the 1/4, 2/3 and 1/1 ratios, respectively. These tensile strengths are approximately two orders of magnitude lower than those obtained by decoration of F-actin with phalloidin.

The behaviour of solutions of pure myosin, of pure F-actin and of the equimolar mixture of myosin and of F-actin is studied. It is found that the chemical potential of the two proteins, in separate solutions, increases monotonically with the increase of protein osmotic pressure. A method is presented to determine the chemical potential of the 1:1 actin-myosin complex formed from equimolar solutions of myosin and of F-actin (as monomer). This is the first evaluation of the chemical potential of actomyosin under conditions similar to those of skeletal muscle. It is found that the filament suspensions of myosin and of the 1:1 actin-myosin complex display a high non-ideal behavior as well as distinctly different energy profiles as a function of protein osmotic pressure. This supports the hypothesis that, in muscle: (a) detached cross-bridge change significantly their free energy when sarcomere is shifting from the relaxed to the active or to the rigor state; and (b) the cross-bridge attachment-detachment process is accompanied by changes of muscle protein osmotic pressure.

The mechanic and elastic properties of rhodamine phalloidin F-actin were investigated as a function of the ionic strength and in the absence of Mg2+. By increasing ionic strength from 3 to 19 mM, critical concentration decreased from 146 to 36 nM and the yield strength increased from 5.6 pN to 28.6 pN. At the ionic strength of 12-13 mM, the elastic modulus by stretching increased by 330-430 kP. nm-1 up to the break point, where it was 38-44.2 MP. The work required to break the filament, 403-439 kJ.M-1 provides an estimate of the free energy of annealing of rhodamine phalloidin F-actin, the annealing constant being 2.8 x 1074 M-1.

The limits of the most recent models of the actin filament are discussed. These model are generated without taking into account the effect of protein osmotic pressure and, in general, of the solvent conditions. As a result they do not provide a bona fide representation of the actin filament in vivo. A new 'fluttering wing' model is proposed which predicts that orientation of the monomers, intermonomer contacts and diameter of the actin filament are sensitive to protein osmotic pressure and to interaction with regulatory proteins.

A model is presented that makes it possible to determine the stiffness of the crossbridge from protein osmotic stress experiments. The model was elaborated while studying the osmotic properties of F-actin and of myosin subfragment-1 F-actin. These studies showed that the elastic modulus by bending of the monomer is directly related to the intrinsic protein osmotic pressure of the system. At a protein osmotic pressure of 1.8 x 10(5) dynes/cm2, the physiological protein osmotic pressure of frog skeletal muscle, it was found that the elastic moduli by bending of the monomer in F-actin and in the myosin subfragment-1 decorated F-actin are 6.5 X 10(7) and 3.3 X 10(8) dynes/cm2, respectively. The value of the elastic modulus by bending of the monomer in the myosin subfragment-1 decorated F-actin compares favorably with the values of the elastic modulus by stretching determined in skeletal muscle fibres.

We have compared the osmotic properties of the hydrated, native actin filament and of hydrated phalloidin-F-actin. We have found that phalloidin-F-actin interacts much more strongly with water than native F-actin. It is therefore very likely that the interaction with myosin (that requires the expulsion of the protein solvation water) is more problematic for phalloidin-F-actin that for native F-actin. We conclude that phalloidin-F-actin is not a bona fide substitute for native F-actin in the "in vitro motility assay".

The diameter of the actin filament decreases with an increase of the protein osmotic pressure. This phenomenon is accompanied by a decrease of the angle (alpha) formed between the long axis of the actin monomer and the pointed end of the filament axis. At 1.8 x 10(5) dyn/cm2 (the protein osmotic pressure in frog muscle) the diameter is 8.34 nm and the angle (alpha) is 61.5 degrees. The interfilament distance of tropomyosin-decorated actin filaments, at a set of different osmotic pressures, is larger than that of F-actin filaments. This suggests that the two tropomyosin helices project out of the contour of the actin filament. The tropomyosin-decorated actin filament is more rigid than F-actin. At 1.8 x 10(5) dyn/cm2, the angle (alpha) is 76.4 degrees, as compared to the value of 61.5 degrees for F-actin. The interfilament distance of troponin-tropomyosin-decorated actin filaments is sensitive to Ca2+: in the physiological range of protein osmotic pressure it decreases from 13.3 nm, in the presence of 2 mM EGTA, to 12.2 nm in the presence of 0.2 mM CaCl2. Two alternative models are proposed to explain the decrease in interfilament distance. (a) Calcium shifts tropomyosin along the actin monomer, toward the filament axis (the classical model). (b) Calcium releases the rigidity of the tropomyosin-decorated filament and restores the original plasticity of F-actin. The consequent decrease of the angle (alpha) brings the tropomyosin helices nearer to the filament axis, without any real movement of tropomyosin along the actin monomer.

At the physiological protein osmotic pressure, the angle formed between the long axis of the actin monomer and the pointed end of the filament axis is roughly 61 degrees in F-actin and about 90 degrees in the myosin subfragment 1--decorated F-actin. This implies that, in the course of the contractile cycle, actin itself contributes, by about 4 nm, to the displacement of the actin filament toward the center of the sarcomer.

The effect of alpha-actinin from chicken gizzard on the properties of F-actin solutions at 37 degrees C and at 4 degrees C was investigated. Beside the well-known increase of the gelling activity of alpha-actinin, it was found that lowering temperature to 4 degrees C: (a) modifies the shape of the alpha-actinin-F-actin binding isotherm; (b) increases the light scattering of the alpha-actinin-F-actin mixtures; (c) induces the formation of ribbons and bundles of F-actin. It was also observed that, by warming to 37 degrees C, the bundles of F-actin formed at 4 degrees C were dissociated into quasi-parallel actin filaments running at a distance of 25-42 nm from each other. On subsequent cooling to 4 degrees C, these parallel filaments were rapidly assembled into bundles. As at 37 degrees C, alpha-actinin displays a potent gelling activity on bundles but not on filaments of actin, and as the gelling activity at 4 degrees C is accompanied by the formation of actin bundles, it is concluded that actin bundling is a necessary condition to promote the actin-gelling activity.

The diameter of the actin filament is influenced by osmotic stress, being 9.0 nm at 1 x 10(5) dynes/cm2 and 6.8 nm at 9.00 x 10(6) dynes/cm2. At 1.81 x 10(5) dynes/cm2, the protein osmotic pressure in frog muscle, the diameter is 7.95 nm. The diameter of the tropomyosin-decorated actin filament is also influenced by osmotic stress even though, at the low pressures (up to 2 x 10(5) dynes/cm2), the decorated filament is significantly more resistant to compression than the undecorated actin filament.

The interaction of alpha-actinin from chicken gizzard with F-actin is quite complex. The apparent dissociation constant, C, increases with the increase of actin concentration according to the following expression: C = Ko + a[actin] - c[actin]5/2. At pH 7.5 and 37 degrees C, in the presence of 0.1 M KCl and 2 mM MgCl2, the dissociation constant at infinite actin dilution, Ko, is 2.17 microM. The binding of alpha-actinin to actin is related by the term a[actin] to the diffusion of actin filaments and by the term c[actin]5/2 to the crossing number concentration of the F-actin network. Especially at low actin concentration, the binding of alpha-actinin to actin is increased by gelsolin, which fragments actin filaments and increases their diffusion. The different binding isotherms of alpha-actinin to actin filaments and to actin bundles are discussed.

At 37 degrees C, the alpha-actin-F-actin binding isotherm is anomalous. In 6.7% polyethylene glycol 6000, concomitantly with the formation of actin bundles, the binding isotherm becomes hyperbolic (Kdiss. = 11.3 microM). alpha-Actinin increases the rigidity of the networks formed by actin bundles in polyethylene glycol and by paracrystalline actin in 16 mM MgCl2 but not by F-actin. It is proposed that in the cell alpha-actinin functions are mostly carried on by interaction with actin bundles.

We propose that, in the cell, the reversible conversion of actin filaments into actin bundles is controlled by the concentration of the macromolecules [we have employed poly(ethylene glycol) 6000 to mimic the macromolecules of the cell] as well as by the nature of the ancillary cytoskeletal proteins that decorate actin filaments. The proposal is based on the following evidence. (1) Under our experimental conditions the transition from filaments into bundles occurs at increasing concentrations of poly(ethylene glycol), with the following sequence: caldesmon-actin, 3%; filamin-actin, 4-5%; caldesmon-tropomyosin-actin, 5-7%; actin, 6-7%; tropomyosin-actin, 9-10%. (2) Under conditions of low osmoelastic stress [3% poly(ethylene glycol)], preformed caldesmon-actin bundles are dissociated by the addition of either tropomyosin or tropomyosin-decorated actin. The dissociation of the bundles promoted by the addition of tropomyosin-decorated actin is faster than that promoted by the addition of tropomyosin.

Solation of actin gel by gelsolin is much less efficient in the presence of a high concentration of macromolecular solutes. The rigidity of the gel formed by 12 microM actin is lowered from 4 to 0.33 dynes/cm2 by 15 nM gelsolin, while in 6% (w/v) polyethylene glycol, rigidity is lowered only from 20 to 11 dynes/cm2 by 64 nM gelsolin. Owing to the large concentration of protein, transitions in the fluid- and gel-like properties of the cytoplasm are expected to be problematic when promoted by gelsolin alone.

We have studied by electron microscopy as well as by measurements of low shear viscosity, rigidity and binding, the effect of alpha-actinin on the gel formed at 37 degrees C with F-actin and with tropomyosin-decorated F-actin. Contrary to previous reports in the literature, alpha-actinin at nanomolar concentrations is an efficient actin gelling protein, even at 37 degrees C, provided that the concentration of actin (or of tropomyosin-decorated F-actin) is low (1.2-2.4 microM). The binding of alpha-actinin to F-actin, as a function of actin concentration, is anomalous. The amount of bound alpha-actinin increases when actin concentration increases from 0 to 1.2 microM but does not change significantly when actin concentration is further increased up to 48 microM. A similar result is obtained with tropomyosin-decorated F-actin. These observations can be explained by an hypothesis that binding is a function of the alpha-actinin - F-actin association constant as well as of the rigidity of the gel. When the concentration of actin increases, the rigidity of the gel also increases and more work is required to bring two actin filaments to the reaction distance with alpha-actinin and, consequently, a larger alpha-actinin concentration is required to attain the same ratio of bound alpha-actinin to actin monomers in the filaments.

At 37 degrees C, in the presence of 6% (w/v) polyethylene glycol 6000, 30 nM alpha-actinin from chicken gizzard induces the gelation of 12 microM actin. Static measurement shows that the addition of 30 nM alpha-actinin increases the rigidity of the system from 23.5 to 54 dynes/cm2. According to the theory of osmoelastic coupling, also large additives, such as the proteins of the cell sap, are able to cause an osmotic stress equivalent to that caused by polyethylene glycol. We thus conclude that, in vivo, alpha-actinin acts as an actin gelling protein.

Filamin increases and tropomyosin decreases the susceptibility of F-actin to form bundles of filaments in the presence of polyethylene glycol 6000. The two proteins, which are located in the leading edge and in the internal part of the cell, respectively, are thus likely to display divergent effects on the microfilaments into bundles transition in these two areas of the cell.

When pyrenyl-labelled actin, at intermediate stages of polymerization, is diluted in the polymerization buffer, the decrease of fluorescence takes place stepwise through pseudo zero order reactions of decreasing rate. It is shown that the analysis of the kinetic course of the reaction allows the evaluation of the length of the actin filaments.

A review of the literature concerning "treadmilling" and fragmentation of actin filaments indicates that these two processes take place under similar conditions and that the mere difference of critical concentration at the two ends of the actin filaments is not a sufficient prerequisite for the exchange of actin monomers with polymer subunits. The hypothesis is made that fragmentation is the main prerequisite for the exchange. A repair mechanism for actin filaments is also presented.

[gamma-32P]ATP-G-actin was polymerized in 4 mM-CaCl2, and the distribution of the radioactive nucleotide among the oligomeric and the polymeric species was studied. The results obtained are best explained by assuming spontaneous fragmentation.

We have studied the effect of sonication on the fluorescence of N-(1-pyrenyl)iodoacetamide-labeled F-actin as well as of native actin-pyrenyl-actin mixed oligomers in which the subunits were covalently attached to each other by phenylenebismaleimide. In both cases the fluorescence of the solution was largely decreased by sonication. We have found that this effect is due (a) to a 20-30% decrease of the specific fluorescence of the polymers. These results question the validity of the novel mechanism for the polymerization of actin recently proposed (D. Pantaloni et al. (1984) J. Biol. Chem. 259, 6274-6283). In these studies, in fact, the implicit assumption was made that the quenching of the fluorescence of the solution under sonication was due exclusively to the conversion of F-actin into G-actin.

In the course of the polymerization of N-(1-pyrenyl) iodoacetamide-labelled actin the increase of the fluorescence is not directly proportional to the amount of the polymer mass formed. This precludes the use of the fluorescence signal as a tool for the quantitative description of the polymerization process.

The interaction of actin with Tb3+ was studied by following the fluorescence emitted at 545 nm when the protein was excited at 285 nm in the presence of Tb3+. It was shown that, at low ionic strength, each actin monomer binds, at saturation, six Tb3+ with an association constant of 0.8 microM-1. In the presence of 0.1 M KCl the association constant decreases to 0.15 and 0.24 microM-1 at subcritical and overcritical actin concentrations, respectively; the number of the binding sites remains six. When polymeric actin is formed by the addition of 2 mM MgCl2, the association constant drops to 0.008 micro M-1 and the number of the binding sites to four. The lower number of the Tb3+ binding sites (four) in the actin polymerized by MgCl2 as compared to the number of binding sites (six) of the actin polymerized by KCl is taken as evidence of a looser structure of this latter polymer. We have also shown that Tb3+ does not replace 45Ca2+ at the single, "high-affinity" site of G-actin. Removal of this Ca2+, in the presence of Tb3+, destroys the typical G- and F-actin structures.

At 3 degrees C and pH 7.0, the addition of 40 nM ATP-G-actin to F-actin (12 microM as the monomer), polymerized in the presence of 4 mM CaCl2, determines a substantial and rapid increase of the viscosity of the solution, which is accompanied by the incorporation of the ATP-G-actin added into the polymer. The hypothesis that the presence of ATP-actin at the filament end(s) promotes the annealing reaction is substantiated by the finding that, after the addition of ATP-G-actin, the average filament length is increased. This finding is relevant, not only because it provides evidences in favour of the existence of annealing but also because it shows that the concentration of ATP-G-actin influences the filaments length distribution through a mechanism different from the elongation reaction.

We compared the properties of the nuclei that accumulate in 7.5 mM-KCl in ATP-G-actin solutions and of the oligomers that are formed by sonication of either G-actin or F-actin. We found that the ability of the above species to prime the polymerization of actin decays with different rates. The nuclei are stable in 7.5 mM-KCl (they decay with a rate constant of 1.5 X 10(-3) s -1 at pH 7.8 at 22 degrees C in the absence of KCl). The oligomers formed by sonication of either G-actin or F-actin, once the sonication is stopped, revert to simpler structures or evolve into F-actin, depending on the KCl concentration in which they are kept. In 10.5 mM-KCl at pH 7.8 at 22 degrees C their priming ability decays with a rate constant of 6 X 10(-3) s -1. We propose that the nuclei that form spontaneously in 7.5 mM-KCl are not directly susceptible to elongation. They must first be converted into activated nuclei, which exist in very low concentration at the steady state. The activated nuclei are directly susceptible to elongation, they have a short life and they decay rapidly into the ground state unless the elongation reaction occurs. Sonication displaces the steady-state concentration in favour of the activated state.

Mechanical stimulation (forced circulation in narrow tubing) accelerates as much as 10-fold the rate of polymerization of actin. The increase in the rate is proportional to the intensity of the stimulation for flow rates between 0 and 3 cm/s. This supports the hypothesis that a statistical factor (the orientation of the flowing particles) is influenced by the flow. Comparison of the kinetics of the polymerization of resting and of mechanically stimulated actin solutions shows that both the nucleation and the elongation steps are accelerated. It is thus concluded that flow orients not only the oligomeric structures but also the actin monomers. The elongation reaction, also in the flow-stimulated samples, occurs always by the addition of ATP--G-actin (or ATP-containing oligomers) and not by the fusion of ADP-containing oligomeric structures.

The cold-induced depolymerization of F(ADP) actin, coupled with the repolymerization by ATP, was employed to study the formation of F(ATP) actin. It is proposed that the cold-induced lability of F(ADP) actin, together with the spontaneous fragmentation of the filament, provides a mechanism for the selective preservation of homogeneous F(ATP) actin.

When protamine is added to actin, different supramolecular structures are formed depending on the molar ratio of the two proteins and of the ionic strength of the medium. At low ionic strength, and going from a molar ratio of protamine to G-actin of 4:1, 2:1 and 1:1, globular aggregates are first converted into extended structures and then to long threads in which the constituent ATP-G-actin is rapidly exchangeable with the actin of the medium. At high ionic strength {Tyrode [(1910) Arch. Int. Pharmacodyn. Ther.20, 205-212] solution}, starting from G-actin and protamine in the 1:1 molar ratio, long ropes are formed that can be resolved into intertwining filaments of 4-5nm diameter. The addition of protamine in a 1:1 molar ratio to a solution of F-actin in Tyrode solution causes the breakage of the actin filaments, which is also revealed by the decrease of the viscosity of the solution and the formation of ordered latero-lateral aggregates. The structures formed by reaction of protamine with G-actin can be separated from free G-actin and protamine by filtration through 0.45mum-pore-size Millipore filters. This technique has been exploited to study the exchange reaction between free actin and the actin-protamine complexes. For these studies the 1:1 actin-protamine complex formed at low ionic strength and the 2:1 actin-protamine complex formed in the presence of 23nm-free Mg(2+) have been selected. In the first case the exchange reaction is practically complete in the dead time of the experiment (20s). In the second case, where the complex operates like a true ATPase, the rate of the exchange is initially comparable with the rate of the ATP cleavage. Later on, however, the complex undergoes a change and the rate of the exchange between free actin and the actin bound to protamine becomes lower than the rate of the ATPase reaction. It is proposed that the ATP exchanges for ADP directly on the G-actin bound in the complex.

We have found that sarcoplasmic reticulum from rabbit muscle contains an ADPase which cleaves ADP bound to F-actin. The interaction is not of the simple Michaelis-Menten type, the order of the reaction being larger than the first. A possible explanation of this behaviour could be that ADPase binds to two adjacent actin monomers with a preferred orientation thus cleaving preferentially the nucleotide of one of the two monomers.

Xanthine oxidase increases the rate of actin polymerization. This occurs at oxidase concentrations as low as 40 nM provided the concentration of the polymerizing agent is low (0.5 mM MgCl2). In the presence of 0.1 M KCl plus 1 mM MgCl2 as the polymerizing agents, xanthine oxidase does not affect the rate of the polymerization but increases significantly the rate of the conversion of F(ATP)actin into F(ADP.Pi)actin and probably also the rate of the orthophosphate release.

In the presence of Mg2+, the formation of actin filaments is hindered by glyceraldehyde-3-phosphate dehydrogenase. This effect, which increases with the square of Mg2+ concentration, is counteracted by 0.15 M KCl. Thus KCl, at concentrations found in the intracellular compartment, appears to be strictly required for the correct formation of actin filaments in all tissues in which the glyceraldehyde-phosphate dehydrogenase concentration is high.

The ATP-G-actin aggregates formed by incubation of ATP-G-actin in 7.5 mM-KCl were characterized by electron-microscopical observation, by high-pressure liquid chromatography and by the study of the 1,N6-etheno-ATP-ATP exchange reaction between the free and the actin-bound nucleotide. In 30 mM-KCl the initial rate of the reduced-viscosity increase is found to be directly related to the amount of the aggregates formed in the course of the preincubation in 7.5 mM-KCl.

The release of Pi from the Pi-G-actin-ADP complex is the rate-limiting step in the ATPase activity that is shown by ATP-G-actin in the presence of protamine.

We have shown that F-actin, polymerized in 50 mM-KCl at 20 degrees C and pH 8.0, can be resolved by centrifugation into two polymer populations, which differ morphologically as well as kinetically. The first population represents about 10% of the overall polymer and is composed of small amorphous aggregates. It rapidly exchanges the bound nucleotide with free ATP in the medium, either directly or through the monomers. The second population is composed of long actin filaments. These are labelled by free ATP in the medium only through condensation with labelled small amorphous aggregates.

Mg2+, at submicromolar concentrations, is needed for the nucleation of actin [Maruyama (1981) J. Biol. Chem. 256, 1060-1062]. I show here that Mn2+ fulfils the same function. It binds to oligomers present in the ATP-G-actin solutions with a ratio of 2-3 Mn2+ ions per 100 actin monomers and with an association constant of 0.66 X 10(10) M-1 at pH 8.2 at 25 degrees C. The time course of the binding of Mn2+ to polymerizing actin is not affected by the initial concentration of the protein. Analysis of the distribution of the binding shows that, both in the large oligomeric species and in the polymers, 1 Mn2+ ion is bound for every 14-25 actin monomers, whereas in the smaller oligomeric species 1 Mn2+ ion is bound for every 4 actin monomers. The proposal is made that Mn2+ stabilizes actin nuclei and decreases the concentration of the monomers at the steady state. It is also proposed that, at least in some experimental conditions, the direct condensation of oligomers of intermediate length is an effective mechanism of F-actin formation.

In 3 mM KCl, 2 mM Tris/HCl pH 7.5, 22 degrees C, 0.38 microM myosin subfragment 1 delays the depolymerization of F-actin (7.2 microM measured as monomer). The depolymerization proceeds rapidly for a few minutes and then slows down suddenly when the ratio between the monomers in the actin filaments and myosin subfragment 1 reaches the value of 11. At this time myosin subfragment 1 is substantially all bound to the actin polymers which form an irregular and discontinuous network of filaments running in doublets and in triplets, perhaps cross-linked by myosin subfragment 1. Depolymerization proceeds then for several hours, apparently ending up with the formation of the 1:1 actin-S1 heteropolymer. The ratio between the monomers in the actin filaments and myosin subfragment 1 at the end of the rapid depolymerization process is different for different protein preparations and may be as low as 5.5. In 2 mM Tris/HCl pH 7.5, 25 degrees C, 1 microM myosin subfragment 1 is able to induce the formation of undecorated actin filaments from 12 microM ATP--G-actin. These filaments probably originate by redistribution of myosin subfragment 1 between the newly formed 1/1 actin-S1 heteropolymer and G-actin in the medium, a process which allows the transient formation of undecorated actin filaments.

When the temperature is lowered from 20 to 4 degrees C, the specific viscosity of actin polymerized in the presence of either 4 mM-CaCl2 or 2 mM-MgCl2, but not of actin polymerized in the presence of 90 mM-KCl, is decreased by 50% in the absence of free ATP. Addition of ATP restores the viscosity of the actin polymerized by Mg2+, but not that of actin polymerized by Ca2+, to the original value. The effect of temperature on actin polymerized in the presence of Ca2+ is due to (a) polymer-into-monomer conversion, (b) latero-lateral aggregation of filaments, and (c) fragmentation of the filaments. Fragmentation, as demonstrated by fractional centrifugation and electron microscopy, was the most important of these.