Fragmentation of actin filaments

104

Under conditions where microtubule nucleation and growth are fast (i.e., high magnesium ion and tubulin concentrations and absence of glycerol), microtubule assembly in vitro exhibits an oscillatory regime preceding the establishment of steady state. The amplitude of the oscillations can represent >50% of the maximum turbidity change and oscillations persist for up to 20 periods of 80 s each. Oscillations are accompanied by extensive length redistribution of microtubules. Preliminary work suggests that the oscillatory kinetics can be simulated using a model in which many microtubules undergo synchronous transitions between growing and rapidly depolymerizing phases, complicated by the kinetically limiting rate of nucleotide exchange on free tubulin.Conditions have been found such that tubulin polymerization into microtubules in vitro exhibits rather persistent oscillations of large amplitude that dampen in the approach to steady state. These oscillations are due to periodic depolymerization and repolymerization processes of the bulk population of microtubules, acting at least partially in synchrony. The oscillations can be conveniently monitored turbidimetrically. This behavior differs strikingly from the classical monotonic growth up to steady state. This phenomenon is consistent with the known nonlinearity in microtubule growth (1, 2) and can be accounted for by synchronous transitions between the GTP-capped and uncapped phases of microtubule ends (3-5). It is worth noting that modest damped oscillations have already been reported, in the approach to steady state, in Monte Carlo simulations of the time course of microtubule growth (5). The present results document another aspect of the "dynamic instability" behavior of microtubules, now widely acknowledged in a variety of in vitro and cellular systems (6-12).The possible biological importance of large periodic changes in the concentrations of both microtubules and free tubulin in the cell is discussed. MATERIALS AND METHODSTubulin was purified through three assembly-disassembly cycles (13) followed by phosphocellulose chromatography (14), and kept at -80°C in 0.05 M Mes, pH 6.8/0.25 mM MgCl2/0.5 mM EGTA/3.4 M glycerol/200 ,uM GTP at a concentration of 3-8 mg/ml. Before each experiment, tubulin was polymerized for 30 min at 37°C in the presence of 0.4 mM GTP/6 mM MgCl2. Microtubules were pelleted and then suspended in ice-cold buffer P, which routinely consisted of 0.1 M Mes, pH 6.8/1 mM EGTA/12 mM MgCl2/2 mM GTP. After 20 min of depolymerization at 0WC, the solution was clarified by centrifugation at 25,000 x g, 4°C, for 20 min and used as such for polymerization. At this stage, tubulin could also be rapidly frozen in liquid nitrogen and used without any loss in polymerization properties for up to at least 3 weeks. Polymerization was carried out at 370C in a 120-Al thermostated cuvette (optical path, 0.5 cm) and monitored turbidimetrically at 350 nm. The reaction was started by switching the temperature from 80C to 370C (rising time, 15 s). Quantitative...

Section: Model and Discussionmentioning

confidence: 99%

Synchronous oscillations in microtubule polymerization.

Carlier¹,

Melki²,

Pantaloni³

et al. 1987

104

“…Detailed analysis using sophisticated computer programs have modified the model for actin assembly to include (i) a first-order Mg2" activation step, (ii) nuclei composed of actin trimers, (iii) different rates of assembly at the ends of the filament, and (iv) enhanced nucleation by spontaneous filament breakage. Although a variety of methods, including viscometry (4,5), flow birefringence (6), light scattering (7)(8)(9)(10)(11), and fluorimetry (11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21), have provided indirect evidence that nucleation is an important feature of actin assembly in vitro, no structural study has directly identified the existence or the structure of the nucleating species.…”

mentioning

confidence: 99%

Synchrotron x-ray diffraction studies of actin structure during polymerization.

Matsudaira¹,

Bordas²,

Koch³

1987

Synchrotron x-ray diffraction was used to identify the oligomers that formed during the earliest stages of actin polymerization. Solution diffraction patterns from Gactin (monomer) and from F-actin (polymer) contain information about the size and shape of actin monomers and the length, width, and subunit organization of filaments. Comparison of patterns collected during polymerization reveals an increase in scatter at spacings >9.0 nm; formation of scattering bands at 5.4, 4.9, and 3.4 nm; formation of a scattering minimum at 6.5 nm; and the presence of an isosbestic point at 9.0 nm. These scattering bands arise from the formation of, and organization of subunits in, framents. At various actin concentrations (0.37-5 mg/ml), the change in scatter in these regions follows simple exponential kinetics with no detectable lag. Our analysis of the x-ray patterns shows that by 0.4 sec after mixing, most of the actin has formed dimers, which then rapidly incorporate into oligomers. At 4 mg/ml the early oligomers increase in length to >30.0 nm within 10 sec. These results suggest that under our conditions actin molecules condense into filaments without the rate-limiting formation of nuclei.Studies on the mechanism of actin polymerization in vitro have influenced our ideas on the regulation of actin function in vivo. Central to these studies, and based partly on the kinetics of helical polymerization, is the proposal by Oosawa and colleagues (1-3) that actin polymerization is a nucleation-condensation process. During the initial phase of assembly, three to four actin monomers cooperatively associate into a short oligomer that nucleates rapid filament elongation. Nucleation is the rate-limiting step in assembly and is detected in kinetic studies as a lag in the time course of polymerization. The existence and size of nuclei is deduced from modeling the kinetics of the elongation phase, assuming that nucleation is an obligatory step in the reaction, and from the presence of a critical concentration for assembly. Detailed analysis using sophisticated computer programs have modified the model for actin assembly to include (i) a first-order Mg2" activation step, (ii) nuclei composed of actin trimers, (iii) different rates of assembly at the ends of the filament, and (iv) enhanced nucleation by spontaneous filament breakage. Although a variety of methods, including viscometry (4,5), flow birefringence (6), light scattering (7-11), and fluorimetry (11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21), have provided indirect evidence that nucleation is an important feature of actin assembly in vitro, no structural study has directly identified the existence or the structure of the nucleating species.Time-resolved x-ray diffraction using high-intensity synchrotron radiation sources has proved a powerful technique for describing the structure and assembly of proteins into multisubunit complexes. Studies on microtubule and collagen fiber assembly have shown that time-resolved x-ray scattering methods can identify the structures forme...

“…In the absence of Ca2', the mechanism of the Mg2+-induced polymerization is as follows: Mg2+ binds to a metal-binding site on G-actin and induces a conformational change, which is required for eventual polymerization. The overall dissociation constant for this binding is (6,7,11,12). Because of the nature of the equations that describe the polymerization process, investigators have had to assume that formation of each species prior to that of the nucleus that undergoes elongation is equally unfavorable.…”

mentioning

confidence: 99%

“…Mg2+ binding is the model used, there has been disagreement about the size of the nucleus, with values ranging from two to four monomer units (6,7,11,12). Because of the nature of the equations that describe the polymerization process, investigators have had to assume that formation of each species prior to that of the nucleus that undergoes elongation is equally unfavorable.…”

mentioning

confidence: 99%

Polymerization of actin: mechanism of the Mg2+-induced process at pH 8 and 20 degrees C.

Frieden¹

1983

124

118

A detailed mechanism that fully accounts for the Mg" -induced polymerization of actin in the presence or absence of Ca2' at 20oC and pH 8 is presented. In the absence of Ca2', the mechanism of the Mg2+-induced polymerization is as follows: Mg2+ binds to a metal-binding site on G-actin and induces a conformational change, which is required for eventual polymerization. The overall dissociation constant for this binding is (6,7,11,12). Because of the nature of the equations that describe the polymerization process, investigators have had to assume that formation of each species prior to that of the nucleus that undergoes elongation is equally unfavorable. This paper describes the time course of polymerization by using a computer simulation system in which the assumption is not a necessary one. The data suggest that, although the species that elongates is indeed a trimer, the formation of trimer is much more favorable than the formation of dimer. I provide here a mechanism that quantitatively describes the full time course of actin polymerization as a function of actin concentration, Mg2+ concentration, and Ca2" concentration.MATERIALS AND METHODS Actin. Rabbit skeletal muscle G-actin was isolated and purified according to Spudich and Watt (13) with gel filtration (Sephadex G-150) as described (14). Unless used immediately after gel filtration, actin was stored at -20oC after lyophilization in the presence of sucrose (2 mg per mg of actin). Lyophilized actin was dialyzed at 40C for 1-2 days either against 2 mM Tris-HCl, pH 8/200 puM ATP/200 AM CaCl2/1.5 mM NaN3 or against 2 mM Tris-HCl, pH 8/100 ,AM ATP/50 AM MgSO4/1.5 mM NaN3/200 ,uM dithiothreitol for Ca2+-free actin (6). The protein concentration was determined spectrophotometrically by using a value of Almg/mI = 0.63 at 290 nm (15) or by Bradford determination (16) for pyrene-labeled actin with G-actin as a standard.Polymerization Kinetics. All actin solutions were centrifuged before use at 180,000 X g for 30 min or 100,000 x g for 60 min. The polymerization process was followed continuously by fluorescence, using trace amounts of pyrene-labeled actin (excitation 365 nm, emission 386 nm) as described (17). Because the fluorescence change (25-fold) on polymerization is independent of the pyrene-labeled actin concentration (17), the amount of pyrene-labeled actin was kept constant (5 tug/ml) even in experiments in which the total actin concentration was changed. The advantages of this assay for quantitative measurement of actin polymerization have been discussed elsewhere (6,7,17). Solutions were stirred with a magnetic stirrer in the bottom of the fluorescence cuvette for only 10 sec after addition of the amount of Mg2+ required to start the polymerization. All polymerization reactions were performed at 20'C. Data were stored in a mode that could be used as a real data curve in the simulation program described below.Simulation of the Full Time Course. A general system of computer simulation of kinetic mechanisms by numerical integration has been descr...