Journal of Muscle Research and Cell Motility 4, 253-262 (1983)

Pyrene actin" documentation of the validity of a sensitive assay for actin polymerization
J O H N A. C O O P E R , S I M O N T H O M A S D. P O L L A R D B. W A L K E R and

Department of Cell Biology and Anatomy, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, U.S.A.
Received 13 September 1982 and in revised form 26 November 1982

Summary
The fluorescence of pyrene-labelled actin is much higher after polymerization. We have characterized in detail the polymerization properties of pyrene actin and report that native and pyrene actin are identical using the following criteria: (1) the time course of polymerization; (2) the elongation rate constants; (3) the intrinsic viscosity; and (4) the critical concentration. Native and pyrene actin copolymerize. Fluorescence of polymerized pyrene actin is 7-10 times higher than monomer. The fluorescent signal is proportional to polymer weight concentration and is insensitive to filament length distribution. Bleaching can be minimized by appropriate filters to allow continuous monitoring of signal. Measurements do not influence polymerization kinetics. This establishes that pyrene actin fluorescence is a valid assay for actin polymerization that is more sensitive than any other current assay. Introduction Fluorescence assays of actin polymerization are highly useful because the signal is generally proportional only to polymer weight concentration and the sample need not be disturbed by flow. Other commonly employed assays, such as viscosity and flow birefringence, are sensitive to polymer length distribution but subject the sample to shear, which breaks filaments. In 1981, Kouyama & Mihashi reported the preparation of pyrene actin by the reaction of actin with N-(1-pyrenyl)iodoacetamide, a thiol-specific reagent. The stoichiometry was 1.0 mol mol ~, and the pyrene was probably located on Cys 374, under the new assignments (Korn, 1982), the fast reacting of the five Cys residues of muscle actin. At appropriate wavelengths the fluorescence of the polymer was 20 times higher than the monomer. This was an important finding because other assays which do not disturb the sample, such as OD 232, light scattering and other fluorescent probes have limited sensitivity (reviewed in Cooper & Pollard, 1982). Before a modified protein, such as pyrene actin, can be used as a probe in a detailed 0142-4319/83 $03.00 + .12 1983 Chapman and Hall Ltd.

254

COOPER, WALKER and POLLARB

quantitative analysis of the p r o p e r t i e s of the native protein, it m u s t be rigorously e s t a b lished that the p r o p e r t i e s are not altered by the Covalent modification. Previously, K o u y a m a & Mihashi (1981) m e a s u r e d the critical concentration of p y r e n e actin. Their result w a s a r e a s o n a b l e value. A l t h o u g h promising, this e x p e r i m e n t w a s not sufficient to e s t a b l i s h the validity of p y r e n e actin as an assay for actin polymerization, so we have p e r f o r m e d additional e x p e r i m e n t s to characterize p y r e n e actin. I n d e p e n d e n t l y , Tellam & Frieden (1982) h a v e carried out variations of t w o of the e x p e r i m e n t s w h i c h w e describe here.

Materials and methods

Preparation of pyrene actin

We made several minor modifications in the procedure of Kouyama & Mihashi (1981). We prepared a chloroform-washed acetone powder of rabbit skeletal muscle and extracted the powder with Buffer A (2mM Tris-HC1, pH 8.0, 0.2mM CaC12, 0.2raM ATP, 0.5raM dithiothreitol). The actin was polymerized and collected by centrifugation. The pellet was homogenized in Buffer P (1 mM NaHCO3, pH 7.6, 0.1 mM CaC12, 0.2 mM ATP) and dialysed against the same buffer for 48 h. The sample was clarified by centrifugation at 100 000g for 2 h. The supernatant was diluted to 1 mg m1-1 with Buffer P, made 0.1 M in KC1 and 1 mM in MgCI~, and incubated at room temperature for 30 min to allow polymerization. N-(1-Pyrenyl)iodoacetamide (Molecular Probes, Junction City, Oregon, U.S.A.) is stored desiccated in the dark at - 1 0 C. 7.5 molN-(1-pyrenyl)iodoacetamide (mol. wt 385) were added per mol actin. The N-(1-pyrenyl)iodoacetamide was first dissolved in dimethylformamide at a concentration of 6 mg m1-1. Although the N-(1-pyrenyl)iodoacetamide immediately precipitates from solution upon addition to the aqueous actin solution, this approach gave more efficient labelling than the addition of solidN-(1-pyrenyl)iodoacetamide. The mixture was rotated at room temperature in the dark for 16 to 24 h. The actin filaments were collected by centrifugation [a sizeable pellet of precipitated N-(1-pyrenyl)iodoacetamide also sediments], the pellet was homogenized in Buffer A, and the mixture dialysed against Buffer A for 48 h. After clarification (100 000g x 2 h) the upper 3/4 of the supernatant was chromatographed on Sephadex G-150 in Buffer A to yield monomeric pyrene-labelled actin.

Extinction coefficient for pyrene actin

The protein concentration of pyrene actin was measured using both Lowry et al. (1951) and Bradford (1976) protein assays with native actin as the standard. Kouyama & Mihashi (1:981) determined an extinction coefficient of 22 000 M-1 c m - I at 344 nm for the pyrene moiety of pyrene actin. From our protein determinations and the extinction coefficient of native actin at 290 nm (26 600 M-1 cm-1; Houk & Ue, 1974) we determined the extinction coefficient of pyrene actin at 290 nm to be 29 400 M-1 cm -1. Therefore, in a mixture of pyrene and native actin one can correct the absorbance at 290 nm by subtracting 0.127 times the absorbance at 344 nm. Total actin equals the quotient of corrected absorbance at 290 nm divided by 26 600 M-1 cm -1. Our labelling ratios were regularly greater than 0.95 but never greater than 1.00.

Fluorescence and light scattering

Measurements of fluorescence and light scattering were made in a Perkin-Elmer Model 650-10S spectrofluorometer. The instrument was equipped with a brass, water-jacketed cuvette holder

P y r e n e actin

255

and a magnetic stirrer, designed and built with the advice and assistance of Daniel Kiehart and Ed Horn. The cuvette holder supplied with the instrument would not accurately maintain the temperature of the sample. The temperature was 25 C, controlled by a circulating water bath. The 'normal' mode, which corrects for fluctuations in lamp intensity, was employed. To minimize fluorescent bleaching, an excitation siit width of 3 n m was used and a translucent glass filter was inserted between the lamp and the sample: Under these conditions bleaching of an actin polymer solution was 1% per hour, which allowed us to make continuous measurements. The emission slit width was 10 nm. The excitation and emission wavelengths were 365 and 407 nm, respectively, for fluorescence measurements and 450 and 450 nm for light scattering. The cuvette was 10 mm by 10 ram, constructed of glass. The sample size was 1-1.5 ml. For fluorescence measurements we used imidazole Grade 111 (Sigma, St Louis, Missouri, U.S.A.) as a pH buffer.

Ostwald viscometry
Ostwald capillary viscometry was performed as described previously (Cooper & Pollard, 1982). Cannon viscometers no. 150 (State College, Pennsylvania, U.S.A.) immersed in a 25 C water bath were used. The sample size was 0.6 ml.

Elongation rate constants

The elongation rate constants were measured in a manner similar to Pollard & Mooseker (1981). Low concentrations of monomeric actin from 1 to 7/~M in the experimental buffer were incubated at time zero with morphologically identifiable nuclei. For native actin these nuclei were glutaraldehyde-fixed Sl-decorated actin filament (developed by Marschall S. Runge, described in Tseng & Pollard, 1982) and for pyrene actin they were Limulus sperm acrosomal processes (Tilneyet al., 1981). The two types of nuclei give results similar to each other and similar to those for microvillar cores (Pollard & Mooseker, 1981). For example, with Limulus processes Bonder & Mooseker (1981) found k+ = 1 0 7 M-1 S 1, k+P = 1.5 x 1 0 6 M - 1 S- ~ , k B = 1.5 s -1, and k_ = 0.5 s 1 for native actin. B P These results agree with those reported here for fixed Sl-actin filaments, for microvillar cores (Pollard & Mooseker, 1981) and, from our laboratory, for fixed Sl-actin filaments (Runge & Pollard, unpublished results). At various times from 20 to 480 s after time zero, the reaction was stopped by the addition of buffer to dilute the monomeric actin to a concentration near its critical concentration, and the diluted samples were quickly applied to a glow-discharged, carbon and Parlodion-coated grid. The grids were negatively stained with uranyl acetate or uranyl formate. Electron micrographs were taken, and the length of undecorated actin filament growth at each end of a number (10-60) of nuclei was measured. The length was converted to actin protomers and divided by the reaction time to yield rate in monomers per second. The mean and standard deviation of the rates at each of four actin m o n o m e r concentrations were determined. These values were employed in a least-squares regression to a single line with weightedy values to yield the slope (association rate constant, k+) and y intercept (-k_, negative dissociation rate constant) and the standard deviation for each value.

Results

Copolymerization at steady-state equilibrium

A c o n v e n i e n t test for c o p o l y m e r i z a t i o n o f labelled a n d native actin critical c o n c e n t r a t i o n o f actin m i x t u r e s w i t h v a r i e d ratios o f labelled the a b s e n c e o f c o p o l y m e r i z a t i o n , the critical c o n c e n t r a t i o n , fluorescence, will d e p e n d o n l y o n the p y r e n e actin c o n c e n t r a t i o n a n d is to as be to m e a s u r e the n a t i v e actin. In measured by independent of

256

C O O P E R , W A L K E R and P O L L A R D

5-

4-

~3~_21'
0 0
! I I I I
0

2 Actin

(~M)

Fig. 1. Copolymerization of native and pyrene actin. Fluorescence signal in arbitrary units is plotted versus total actin concentration. The percentage of pyrene label was 30 (ll) and 6 ((9). The signal value for 6% were multiplied by five before the graph was made. The critical concentration is 0.6/~M for each set of data, which implies that copolymerization of pyrene and native actin has occurred. Conditions: 0.1 M KCI, 1 mM MgC12, 0.2 mM ATP, 10 mM potassium phosphate, pH 7.0, and 21.5% (v/v) Buffer A from the actin. This experiment was repeated for three other ionic conditions with similar actin concentrations, and copolymerization was found in each case. Those conditions were the following: (1) 0.1 M KC1, 1 mM MgC12, 20 mM imidazole-HC1, pH 7.0, and 90% (v/v) Buffer A without Ca2+; (2) 0.1 M KC1, 1 mM MgC12, 20 mM imidazole-HC1, pH 7.0, 90% (v/v) Buffer A; and (3) 0.1 M KC1, 20 mM imidazole-HC1, pH 7.0, 90% (v/v) Buffer A.

native actin concentration. If copolymerization occurs, the critical concentration in terms of total actin will b e the same for all ratios of p y r e n e to native actin. A representative e x p e r i m e n t is s h o w n in Fig. 1. The critical concentrations are the same for two mixtures in which the fraction of p y r e n e actin differs by a factor of 5. By this criterion, copolymerization also occurs u n d e r the several o t h e r ionic conditions given in the caption to Fig. 1. The slopes of the lines for the m o n o m e r and polymer sections of the curve s h o w that the fluorescent signal from p o l y m e r was 7-10-fold greater than the signal from m o n o m e r .

Elongation rate constants

Within the limits of experimental error the elongation rate constants for p y r e n e actin are the same as for native actin (Table 1).

Polymerization kinetics
The time course of polymerization was m e a s u r e d using varied ratios of p y r e n e to native actin. In one ionic condition, with Mg 2+ present, the early slow phase of polymerization

Pyrene actin Table 1. Elongation rate constants for pyrene actin. Error shown is one standard deviation, calculated as in Materials and methods. Conditions: 0.1 M KC1, 20 mM imidazole-HCl, pH 7.0, 25%(v/v) Buffer A, 25 C.
Association Dissociation

257

(x 10 -6 s -I M 1) Barbed end Pyrene Native Pointed end Pyrene Native

(s-l)

9.2 1.4 9.5 2.4

6.8 2.0 4.2 3.7

0.7 0.3 1.1 0.2

-0.1 0.5 0.8 0.3

is relatively fast a n d curves for p o l y m e r i z a t i o n of 100%, 10% a n d 1% p y r e n e actin w e r e a l w a y s s u p e r i m p o s a b l e , as s h o w n in Fig. 2a. In the p r e s e n c e of Ca 2+ a n d the a b s e n c e of M g 2+, the early slow p h a s e is relatively slow a n d such c u r v e s are rarely s u p e r i m p o s e d exactly. S o m e t i m e s the higher p e r c e n t a g e of p y r e n e p o l y m e r i z e d faster but s o m e t i m e s it w a s slower. W e suspect this difference w a s due to nucleating species w h i c h d e v e l o p e d in o u r p r e p a r a t i o n s of p y r e n e a n d native actins. Fig. 2b is a representative e x p e r i m e n t w h i c h s h o w s that the differences b e t w e e n the c u r v e s are small. Tellam & Frieden (1982) p e r f o r m e d one e x p e r i m e n t similar to those s h o w n here. T h e y u s e d a n a r r o w e r r a n g e of p y r e n e actin p e r c e n t a g e t h a n w e did, a n d t h e y m a d e d i s c o n t i n u o u s instead of continuous measurements. 10

///
E
O0 (a) 500 1000 1500 Time (s) 2000 2500

(b)

i!

800

1~00'24'00' s~00'4000
Time (s)

Fig. 2. Polymerization kinetics of varied ratios of pyrene to native actin. Fluorescence signal in normalized units is plotted versus time. (a) The curves represent, from left to right, 100%, 10% and 1% pyrene actin. The curves are displaced 250 seconds from each other for clarity. Conditions: 7.1/~M actin, 1 mM MgC12, 0.1 M KC1, 20 mM imidazole-HC1, pH 7.0, 90% (v/v) Buffer A. (b) The curves represent, from left to right, 20%, 10%, 1% and 0.2% pyrene actin. The curves are displaced 400 seconds from each other for clarity. Conditions: 18.9/2M actin, 0.1 M KC1, 20 mM imidazole-HCI, pH 7.0, 90% (v/v) Buffer A.

258

COOPER, WALKER and POLLARD

Continuous and discontinuous measurement of polymerization kinetics were compared to determine whether the measurement affects polymerization. The time course of polymerization is the same for continuous measurement and for discontinuous measurement where we have the sample illuminated for 5 s and then not illuminated for 25 s.

Comparison of fluorescence and light scattering

Ninety degrees light scattering has been s h o w n to be a useful assay for polymer weight concentration under these conditions where the filaments are long (Wegner & Engel, 1975). We measured the time course o f actin polymerization by fluorescence and by light scattering. The results, shown in Fig. 3, are superimposable for the two assays. The curves have the sigmoidal shape which we have always observed for the kinetics of polymerization of flesh gel-filtered actin. Tellam & Frieden (1982) performed a similar experiment comparing pyrene actin fluorescence to rapid sedimentation.
~10

3 4

,--r

800

1600 2400 Time (s)

3200

4000

Fig. 3. Comparison of polymerization kinetics assayed by pyrene actin fluorescence and light scattering. Light signal in normalized units is plotted on the ordinate versus time. Four smooth sigmoidal curves, generated by continuous monitoring, are shown. The two on the left are fluorescence, and the two on the right are light scattering. The curves are superimposable but are separated by 200 seconds for clarity.

Ostwald viscosity
Measurements of Ostwald capillary viscosity are a function of both polymer length distribution (Craig & Powell, 1980) and polymer weight concentration. The length distribution is probably a function of the relative rates of breaking and annealing of filaments. Table 2 s h o w s measurements o f intrinsic viscosity for native and pyrene actin filaments. The results are identical within the limits of error.

Fluorescence and filament length

To determine whether fluorescence is sensitive to filament length distribution the fluorescence of an unstirred sample was compared to that of the sample when it was stirred. A stir bar in the cuvette stirred the solution. Stirring creates shear, which breaks

Pyrene actin
Table 2. Intrinsic viscosity (cm3 g 1) of pyrene and native actin polymer. The values shown are the average from two experiments. Error shown is calculated by propagating error in time measurements through the viscosity calculations. Intrinsic viscosity was calculated as the slope of a plot of relative viscosity versus actin concentration at one shear rate. Conditions: 0.1 M KC1, 20 mM imidazole-HC1, pH 7.0, 90% (v/v) Buffer A, and magnesium chloride as shown in headings. Samples were incubated at 25 C overnight before measurements were made.
Pyrene actin Native actin

259

No Mg2+ 1 mM Mg2+

880 -+ 30 870 -+-30

890 + 30 870 + 30

actin filaments; to show that stirring actually broke filaments we followed the kinetics of polymerization versus time. Initially the sample was not stirred. At about 50% polymer formation the stirrer was turned on for 30 s. The polymerization rate after stirring was 50 times larger than the rate before stirring, which indicated that the filament number had increased by 50 times and therefore the average filament was broken into 50 pieces. When a sample in which polymerization was complete was stirred, the value of the fluorescence did not change. Even though the filaments were 50 times shorter, the fluorescence was the same. Therefore, fluorescence was independent of filament length distribution in this range of length.
Discussion

In 1981 Kouyama & Mihashi described the preparation of pyrene actin and pointed to its usefulness as a probe of actin polymerization and its high sensitivity. We feel it is of the utmost importance to document rigorously the validity of any assay for actin polymerization in the light of previous difficulties in the design and interpretation of such assays (Cooper & Pollard, 1982). This is particularly important for the pyrene actin fluorescence assay because the simplicity and straightforward interpretation of the assay make it likely that it will receive wide use. We have performed several key experiments which establish pyrene actin fluorescence as a valid assay for polymerization. Pyrene actin is identical to native actin in its critical concentration, elongation rate constants, kinetics of polymerization and Ostwald viscosity. Experiments with different ratios of pyrene to native actin show that the two copolymerize both kinetically and at steady state. At steady state our results show that two different mixtures of native actin and pyrene actin which differ by five-fold in their pyrene percentage have the same critical concentration. This finding unequivocally demonstrates copolymerization. A critical concentration measurement for 0 or 100% pyrene actin is neither necessary nor sufficient to document copolymerization. The signal for the mixture of lower pyrene actin concentration was

260

COOPER, WALKER and POLLARD

multiplied by 5 to make the graph (Fig. 1). The agreement between the two sets of values shows that the signal does not depend on interactions from one pyrene actin to another. That is, a pyrene actin protomer in a filament gives the same signal whether it is adjacent to pyrene or native actin. Kinetically, the assembly of actin monomers into polymer involves steps other than elongation. These steps include nucleation and possibly activation and fragmentation(Pollard & Craig, 1982). Our results show that various mixtures of pyrene actin and native actin polymerize with similar kinetics. The critical concentration, ~, is determined by the ratio of the elongation rate constants, as shown in Equation 1 (Oosawa & Asakura, 1975), where k+ is the sum of the association rate constants for the barbed and pointed ends, andk_ is a similar sum of the dissociation rate constants
e~ = k_/k+

(1)

Identity of the critical concentration for native and pyrene actin implies that the ratio of their elongation rate constants is also the same, but rigorous characterization of the modified actin requires that the elongation rate constants must be measured directly for several reasons. First, the identity of the critical concentration only implies the identity of the ratio of the elongation rate constants, and their actual values may differ. Second, the critical concentration measurement is at steady state, reflecting a combination of the critical concentration at the barbed and pointed ends, which may not be the same in the presence of ATP hydrolysis (Wegner, 1976). Third, since the critical concentration measurement is made at steady state in the presence of ATP hydrolysis, the measurement reflects an uncertain combination of interaction between actin species with bound ADP or ATP. Therefore, we have measured the elongation rate constants of pyrene actin directly using an electron microscopic assay (Pollard & Mooseker, 1981) with sheared acrosomal processes from Limulus sperm as the morphologically identifiable nuclei. This assay separates barbed and pointed ends and is performed rapidly in the presence of excess ATP. The ability of acrosomal processes to nucleate polymerization was described by Tilney et al. (1981). Bonder & Mooseker (t981) and ourselves (unpublished results) have found that the acrosomal processes give results similar to those using microvillar cores (Pollard & Mooseker, 1981) or fixed Sl-decorated actin filaments (unpublished results, Runge & Pollard) as nuclei in quantitative assays for elongation rates. Within the limits of experimental error, the elongation rate constants for pyrene actin are the same as for native actin. The comparison of pyrene actin fluorescence with light scattering confirms that the fluorescent signal is proportional to polymer weight concentration. We have shown, furthermore, that the fluorescent signal from a stirred solution of actin filaments is the same as that from an unstirred solution. This confirms that the assay is not sensitive to filament length. We have confirmed the high sensitivity of the assay. We measured an increase of 7-10-fold in the fluorescence between polymer and monomer, with excitation at 365 nm and emission at 407 nm. Kouyama & Mihashi (1981) estimated a

Pyrene actin

261

20-25-fold fluorescence increase from their observation that the quantum yield with excitation at 342 nm increased five-fold. We have shown that with appropriate filters bleaching can be minimized to allow continuous monitoring of the signal, a significant improvement over discontinuous measurement, which has theoretical and experimental difficulties. Under the same conditions, the polymerization rate was not affected by the excitation light employed in the fluorometer. We feel this assay is superior to other available assays for several reasons. First, the signal is proportional only to polymer weight concentration and not to length distribution, a drawback of capillary viscometry. Second, the sample is not disturbed, as in flow birefringence or capillary viscometry. Third, the signal is high, both relative to the monomer signal and absolutely on a standard fluorometer. To consider other similar assays, 7-chloro-4-nitrobenzeno-2-oxa-l,3-diazole actin (NBD-NEM-actin), another fluorescent derivative, increases its fluorescence by only 2.2-fold on polymerization (Detmers et al., 1981). Resonance energy transfer requires the synthesis of two fluorescent actin derivatives, and its sensitivity has not been compared to other methods (Taylor et al.,1981). A232 m e a s u r e s a small change (1-2%) in the total absorbance (Higashi & Oosawa, 1965), which limits the quantities of other substances, such as proteins and nucleotides, which can be present in the sample. The assay also requires a first-rate spectrophotometer. Light scattering also has a high signal, but the background is orders of magnitude higher than that of pyrene actin fluorescence, leading to a lower signal-to-noise ratio. Light scattering samples must be meticulously free of dust, debris and bubbles. Light scattering also might not be proportional to polymer weight concentration if the filaments are very short, but pyrene actin fluorescence should be unaffected. Light scattering is affected by filament aggregation and filament binding proteins, such as tropomyosin (Walsh & Wegner, 1980). In summary, we are pleased to extend the work of Kouyama & Mihashi (1981) to show that pyrene actin is a useful probe for actin polymerization. Its strong points are the following: (1) the signal is proportional to polymer weight concentration only; (2) the sample is not sheared; and (3) the sensitivity is high. One potential drawback is the effect of the pyrene label on the interaction of actin with other proteins. Kouyama & Mihashi found that heavy meromyosin interacted differently with native as opposed to pyrene actin. In preliminary results we have found (Lee et al., 1982) that Acanthamoeba profilin binds similarly to pyrene and native Acanthamoeba actin.

Acknowledgements
We are grateful to T. Y. Tsong for helpful discussions, and to Barry Knox for help with the error analysis for the elongation rate constants. J.A.C. was supported by the Medical Scientist Training Program, NIH GM 7039. The work was supported by research grants from the NIH (GM-26338) and the Muscular Dystrophy Association of America.

262

COOPER, WALKER and POLLARD

References
BONDER, E. M. & MOOSEKER, M. S. (1.981) The acrosomal process: a new tool in studying actin assembly a n d structure. J. Cell Biol. 91, 306a. BRAD F O RD, M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of p r o t e i n - d y e binding. Analyt. Biochem. 72, 248-54. COOPER, J. A. & POLLARD, T. D. (1982) Methods to measure actin polymerization. Meth. Enzymol. 85, 182-210. CRAIG, S. W. & POWELL, L. D. (1980) Regulation of actin polymerization by villin, a 95 000 dalton cytoskeleton c o m p o n e n t of intestinal brush border. Cell 22, 739-46. DETMERS, P., WEBER, A. ELZINGA, M. & STEPHENS, R. E. (1981) 7-Chloro4-nitrobenzeno-2-oxa-l,3-diazole actin as a probe for actin polymerization. J. biol. Chem. 256, 99-105. HIGASH], S. & OOSAWA, F. (1965)Conformational changes associated with polymerization and nucleotide binding in actin molecules. ]. molec. Biol. 12, 843-65. HOUK, T. W., JR &'UE, K. (1974) The m e a s u r e m e n t of actin concentration in solution: a comparison of methods. Analyt. Biochem. 62, 66-74. KOUYAMA, T. & MIHASHI, K. (1981) Fluorimetry study of N-(1-pyrenyl)iodoacetamide-labelled F-actin. Eur. J. Biochem. 114, 33-8. KORN, E. D. (1982) Actin polymerization and its regulation by proteins from non-muscle cells. Physiol. Rev. 62, 672-737. LEE, S., COOPER, I. A. & POLLARD, T. D. (1982) Fluorescence measurement of Acanthamoeba profilin b i n d i n g to pyrene-labeled actin monomers. J. Cell Biol. 95, 297a. LOWRY, O. H., ROSENBROUGH, N. J., FARR, A. L. & RANDALL, R. J. (1951) Protein m e a s u r e m e n t with the Folin phenol reagent. ]. biol. Chem. 193, 265-75. O O S A W A , F. & ASAKURA, S. (1975)Thermodynamics of the Polymerization of Protein. New York: Academic Press. POLLARD, T. D. & CRAIG, S. W. (1982) Mechanism of actin polymerization. TIBS 7, 55-8. POLLARD, T. D. & MOOSEKER, M. S. (1981) Direct m e a s u r e m e n t of actin polymerization rate constants by electron microscopy of actin filaments nucleated by isolated microvillus cores. J. Cell Biol. 88, 654-9. TAYLOR, D. L., REIDLER, J. SPUDICH, J. & STRYER, L. (1981) Detection of actin assembly by fluorescence energy transfer. ]. Cell Biol. 89, 362-7. TELLAM, R. & FRIEDEN, C. (1982)Cytochalasin D and platelet gelsolin accelerate actin polymer formation. A model for regulation of the extent of actin polymer formation in vivo. Biochemistry 21, 3207-14. TILNEY, L. G., BONDER, E. M. & DEROSIER, D. J. (1981) Actin filaments elongate from their membrane-associated ends. J. Cell Biol. 90, 485-94. TSENG, P. & POLLARD, T. D. (1982)Mechanism of action ofAcanthamoeba profilin. J. Cell Biol. 94, 213-8. WALSH, T. P. & WEGNER, A. (1980) Effect of the state of oxidation of cysteine 190 of tropomyosin on the assembly of the actin-tropomyosin complex. Biochim. Biophys. Acta 626, 79-87. WEGNER, A. (1976) Head to tail polymerization of actin. J. molec. Biol. 108, 139-50. WEGNER, A. & ENGEL, J. (1975) Kinetics of the cooperative association of actin to actin filaments. Biophys. Chem. 3, 215-25.