J. Biochem. 87, 785-802 (1980)

Purification and Properties of Soluble Actin from Sea Urchin Eggs1

Issei MABUCHI* and James A. SPUDICH**

*Department of Biology , College of General Education, The University of Tokyo, Komaba, Meguro-ku, Tokyo 153, and **Department of Structural Biology, Sherman Fairchild Center, Stanford University School of Medicine, Stanford, Ca. 94305, U.S.A.

Received for publication, September 8, 1979

Unfertilized eggs of the sea urchin, Strongylocentrotus purpuratus, were homogenized in a buffer containing 0.1 M KCl and 2 mM MgCl2 at pH 6.85. About 50 % of the actin was recovered in the high-speed supernate of the homogenate. More than 80% of the actin in this supernate was found to be monomeric upon gel filtration chromatography through a Sephadex G-150 column or by a DNase I inhibition assay. The critical concentration for polymerization of this actin prior to further purification was 0.3-0.9 mg/ml under various conditions. Actin was purified to near homogeneity from the Sephadex G-150 pool with a high yield. The purified actin had a critical concentration for polymerization of 0.02-0.03 mg/ml. The isoelectric point of the crude actin and the purified actin was the same. Indeed, we found that there is only one isoelectric focusing species of actin in the sea urchin egg, and it has an isoelectric point more basic than rabbit skeletal muscle actin. The discrepancy between the polymerizability of the crude and purified actin may be due to the presence of factors in the crude fraction which inhibit the polymerization of actin.

Actin is an ubiquitous protein in muscle and to occur. Cytokinesis is one such example, in nonmuscle cells and is considered to contribute to which a contractile ring composed of actin filaments many motile activities of the cell, such as cyto formed at the equatorial cell cortex may function to kinesis, phagocytosis, migration, shape changes, cleave the cell (3). Just before cytokinesis numerous and cytoplasmic streaming. Contrary to the case parallel actin filaments, organized in the form of in muscle, the movements of nonmuscle cells may a bundle, become apparent in the region of the often involve reorganizations of contractile pro cleavage furrow. The contractile ring becomes teins. In other words, a temporary contractile smaller as cleavage proceeds, and filaments are not apparatus may be formed at the correct time and readily apparent after completion of cleavage (4). in the correct place for a particular cell movement In sea urchin or starfish eggs it generally takes about five minutes for the cell to cleave from the 1 This research was supported by grants to one of the beginning of the furrowing. Thus, there should be rapid reorganization of the contractile ring authors (J.A.S.) from the American Cancer Society actin during this period. (VC 121E) and the National Institutes of Health (GM- In the sea urchin egg, another type of organi 25240). Parts of this work have been reported earlier zation of actin occurs in the elongation of micro- (1, 2).

Vol. 87, No. 3, 1980 785 786 I. MABUCHI and J.A. SPUDICH

villi after fertilization (5). These microvilli con centrotus purpuratus, were washed twice with sea tain a bundle of actin filaments, all of which have water. Jelly layers were removed by bringing the

the same polarity; the arrowheads formed upon pH to 5.0 for 1 min. The pH was then brought decoration with heavy meromyosin point away back to 8.2. Eggs were further washed once with from the cell membrane (5). There is no evidence sea water and once with a cold solution of 0.5 M

that these actin bundles arise from preexistent glycerol, 0.2 M NaCl, and 10 mM NaHCO3, followed actin filaments, and it is possible that they are by centrifugation at 6,000 rpm for 10 min in a formed as a result of local, oriented polymerization Beckman JS 13 rotor. To the 17 ml of packed of monomeric actin. Thus, we have investigated eggs, an equal volume of F-buffer 1 was added and the amount of monomeric actin in the sea urchin the egg suspension was homogenized with a motor- egg and the polymerizability of this actin. We driven Teflon-glass homogenizer for ten strokes. have found in the unfertilized sea urchin egg, similar The homogenate was centrifuged at 260,000 •~ g to the situation in a number of other cell types for 2 h (under this condition, materials of greater (6-8), that a considerable amount of the actin is than 20S sediment) in a Beckman model L5 monomeric and the critical concentration of this ultracentrifuge. The supernate (Sl ; 20 ml) was monomeric actin for polymerization is unusually carefully separated from the pellet, the top of high. which was viscous. Inclusion of glucose and EGTA in the homogenizing buffer solution mini-

MATERIALS AND METHODS mized the appearance of this viscous material. The presence of EGTA was also important to

Buffer Solutions-"F-buffer" is a buffer solu suppress proteases which cleave actin. For this

tion which normally favors actin polymerization reason, EGTA was always present in the buffer

and "G-buffer" is one which favors actin depoly solutions used in the purification of egg actin.

merization. The compositions of the buffer S1 was sonicated with a Kontes sonicator solutions used were: F-buffer 1, 0.1 M KCl-2 for I min to reduce the viscosity and centrifuged

mm MgC12 5 mM ethyleneglycol-bis(ƒÀ-aminoethyl again at 260,000 •~ g for 2 h to sediment materials

ether)-N,N•Œ-tetraacetic acid (EGTA)2-20 mM 2(N- which did not sediment the first time because of

morpholino) ethane sulfonic acid (MES), pH 6.85- the viscosity of the solution. Pellets from the

1 mM ATP-1 mM dithiothreitol (DTT)-10 mM p- first and second centrifugation were designated as

tosyl-L-arginine methyl ester.HCl (TAME)-0.7m P1 and P2. The final supernate was designated

D-glucose; F-buffer 2, 0.1m KCl-2 mM MgC12- as S2 (19 ml, 31 mg/ml). In some experiments 1 mM EGTA-20 mM MES, pH 6.85-0.2 mM ATP- the sonicated S1 was dialyzed overnight against

0.5 mM DTT-1 mM TAME; F-buffer 3, 0.1 M F-buffer 2 to encourage any depolymerized actin

KCl-2 mM MgCl2-1 mM EGTA-20 mM MES, pH to polymerize. The G-actin concentration in the

6.85-0.5 mM ATP-0.5 mM DTT-1 mM TAME; S2 obtained from this dialyzed S1 after the second

G-buffer, 1 mM N-tris(hydroxymethyl)methyl-2- centrifugation was the same as that in the S2 of aminoethane sulfonic acid (TES), pH 7.5-0.5 mM the routine procedure (about I mg/ml). ATP-0.5 mM DTT-0.05 mM MgCl2 0.1 mM TAME; Purification of Egg Actin-In general, purifica KI-buffer, 0.6 M KI-0.1 M KCI-1 mM ATP-5 mM tion steps that have proven useful for muscle actin

DTT-20 mM MES, pH 6.85. (9), Acanthamoeba actin (10), and Dictyostelium Preparation of Egg Extract-Temperatures actin (11) proved useful for the egg actin as well. were between 0 and 4•Ž unless otherwise specified. S2 was applied to a Sephadex G-150 column

Unfertilized eggs of the sea urchin, Strongylo- (4 •~ 90 cm) which had been equilibrated with F-buffer 2. As described in the "RESULTS"

2 Abbreviations used are: EGTA section, a major part of the actin in S2 eluted as , ethyleneglycol-bis(ƒÀ- monomers. This monomeric actin was pooled aminoethyl ether)-N,N•Œ-tetraacetic acid; MES, 2(N- morpholino) ethane sulfonic acid; DTT, dithiothreitol; (G-150 actin; 200 ml, 0.68 mg/ml) and passed TAME, p-tosyl-L-arginine methyl ester.HCl; TES, N- through a DEAE-cellulose column (Whatman tris(hydroxymethyl)methyl-2-aminoethane sulfonic acid; DE-52; bed volume, 40 ml) which had been washed SDS, sodium dodecyl sulfate. with F-buffer 3. After application of the sample,

J. Biochem. SOLUBLE ACTIN FROM SEA URCHIN EGGS 787

the column was washed with 100 ml of F-buffer modifications. The temperature of the solutions 3 and then proteins were eluted with 400 ml of a was kept at 22•Ž. Ten ƒÊl of 0.2 mg/ml DNase I

linear KCl gradient of 0.1-0.4 M. Actin eluted (Worthington Biochem. Corp., 2,000-3,000 units/ at about 0.19 M KCl. mg) in 20 mM MES buffer, pH 6.85 was placed in To this actin fraction (DEAE actin; 68 ml, a corner of a microcuvette. Sample solution 0.37 mg/ml) solid ammonium sulfate was added containing actin (1-25 ƒÊl) was added to the drop to 600% saturation, and the solution was incubated of DNase I solution, followed by mixing by invert overnight. Precipitates were collected by cen ing the cuvette three times. The cuvette was

trifugation, washed once with 20 ml of 45 quickly placed in the spectrophotometer (Beckman ammonium sulfate-containing F-buffer 3 and once model 35) and 1 ml of 40 ƒÊg/ml calf thymus DNA with 10 ml of 35 % ammonium sulfate-containing (Sigma Chem. Co., grade I) dissolved in 2 mM F-buffer 3, suspended in 2 ml of F-buffer 3 and MgCl2, 1 mM CaCl2, and 20 mM MES buffer was then dialyzed against F-buffer 3 overnight. added. The solution was mixed by pipetting out

The dialyzed actin fraction, which contained and in three times. About 5 s elapsed from the

mostly polymerized actin was clarified by centrifu time of addition of the actin sample to the addition

gation at 10,000 •~ g for 10 min and actin was of the DNA solution. This was short enough to collected by centrifugation at 200,000 •~ g for 1 h. allow the selective binding of DNase I to G-actin

The pelleted actin was resuspended in 2 ml of (13).

G-buffer by sonication and dialyzed against G- Figure 1 shows the standard curve of the actin

buffer for 15 h. The G-actin solution was clarified determination using DNase I and rabbit skeletal

by centrifugation at 150,000 •~ g for 30 min. This muscle G-actin in G-buffer. This line deviated actin was subjected to a cycle of polymerization- from the theoretical line which was drawn suppos-

depolymerization. Polymerization was induced ing a 1 to 1 binding of actin and DNase I, the

by addition of 3 M KCI, 0.1 M MgCl2, and 0.5 M molecular weights of which are 42,000 and 31,000,

MES buffer, pH 6.85 to give final concentrations respectively. We used the linear portion of this

of 75 mM, 2 mM, and 10 mM, respectively. After standard curve to estimate the amount of actin

standing overnight at 0•Ž and for 1 h at 25•Ž, in egg samples. The total actin concentration was determined after depolymerization of poly polymerized actin was collected and depolymerized as above. The final G-actin is called purified egg merized actin in 0.5 M guanidine • HCl (13) at 22•Ž actin (2 ml, 5.5 mg/ml). for 5 to 10 min. Blikstad et al. (13) used 0.75 M

Quantitative Determination of Actin-The first guanidine•HCl instead because they found that method we employed to quantify actin was densito more than 0.3 mg/ml of F-actin was not com-

metry of a Coomassie brilliant blue-stained sodium pletely depolymerized with 0.5m guanidine•HCl.

dodecyl sulfate (SDS)-polyacrylamide gel. SDS- For us, however, 0.5 M guanidine•HCl was enough to depolymerize all of the polymerized actin in a polyacrylamide gel electrophoresis (SDS-PAGE) was carried out using the discontinuous buffer crude actin fraction (about 1.5 mg/ml actin) or in

system of Laemmli (12). The densitometry was a pure egg F-actin preparation (about 0.5 mg/ml

carried out with a Transidyne gel scanner (Biomed actin). Instruments Co.). Eight % and 12% o acrylamide Determination of the Critical Concentration of Actin for Polymerization-For purified actin, salt gels were used. A sample solution was diluted with a solution of 8 M urea, 0.5 % SDS, 5 mM solutions and MES buffer, pH 6.85, were added

EDTA, 0.1 M ƒÀ-mercaptoethanol, and 20 mM to aliquots to give 75 mM KCl and/or 2 mM MgCl2,

Tris•HCl, pH 8.5, boiled for 3 min, and 401 ƒÊl and 10 mM MES, after dilution of the G-actin

was applied to each channel. When quantifying solution with G-buffer. For the actin solution

actin from samples containing many other pro which was recovered from Sephadex G-150 column teins, the base line of the peak was selected to chromatography, the actin was first concentrated

account for the contribution of other staining in a dialysis bag with Aquacide II-A (Calbiochem).

materials (see Fig. 4a). The concentrated solution was then dialyzed The second method was the DNase I inhibition against F-buffer 3 for 5 h and diluted with F-buffer assay devised by Blikstad et al. (13), with minor 3 to give appropriate actin concentrations. Both

Vol. 87, No. 3, 1980 788 1. MABUCHI and J.A. SPUDICH

Fig. 1. Determination of actin by the DNase I inhibition assay. The assay was carried out according to Blikstad et al. (13) with minor modification (see " MATERIALS AND METHODS") . Ten ƒÊl of 0.2 mg/ml DNase I dissolved in 20 mM MES buffer, pH 6.85, and 1-25 ƒÊl of rabbit skeletal muscle G-actin in G-buffer were mixed in a microcuvette. Immediately after the addition of 1 ml of a 40 ƒÊg/ml DNA solution in 2 mM MgCl2, 1 mM CaCl2, and 20 mM MES buffer, the change in A260nm was recorded. The solid line represents the standard line obtained experimentally and the broken line represents the theoretical one, assuming that one mol of actin interacts with one mol of DNase I.

the purified actin and the concentrated G-150 ington Biochem. Corp., 1,000 units/ml), which had

actin were further incubated at 0•Ž for 48 h to been washed with G-buffer containing 3 M guani

reach monomer-polymer equilibrium. To deter- dine•EHCl and equilibrated with F-buffer 2 or K1-

mine the critical concentration at 25•Ž, an addi buffer. The column was washed first with the

tional incubation was carried out at 25•Ž for 2 h. equilibration buffer and then with G-buffer con The critical concentration was determined by taining 0.75 M guanidine.HCl, and then proteins

viscometry, sedimentation and the DNase I inhibi were eluted with G-buffer containing 3 M guanidine.

tion assay. Ostwald-type semi-microviscometers HCl. Recovered proteins were freed from guani

were used; a Cannon 75 viscometer which had an dine.HCl by dialysis against G-buffer and then

outflow time for water of 220 s was used for 0•Ž dialyzed against a solution of 8 M urea, 0.5 % SDS, measurements and a Cannon 50 viscometer which 5 mM EDTA, 0.1 M ƒÀ-mercaptoethanol, and 20

had an outflow time for water of 208 s was used mM Tris.HCl, pH 8.5 for SDS-PAGE or against

for 25•Ž measurements. Sedimentation was carried the lysis buffer (see below) for isoelectric focusing.

out in a Beckman 65 rotor. The centrifugation Isoelectric Focusing-Isoelectric focusing of

was done so that materials which had a sedimenta actin was carried out by the method described by

tion coefficient greater than 14S would sediment. O'Farrell (15) as modified by Rubenstein and For the crude actin fraction, the amount of actin Spudich (16). Ampholines of which 80% were

in the pellet and in the supernate was determined pH 5-7 and 20 % were pH 3.5-10 were used and the

by densitometry after SDS-PAGE. For purified gel was run at 8,800 volts •~ h. To prevent pos actin, protein determination (see below) was used sible proteolysis, 0.1 mM EGTA and 0.1 mM TAME

to quantify the actin in the pellet and the supernate. were included in the lysis buffer described by

DNase I-Agarose Affinity Chromatography- O'Farrell.

This was performed according to Lindberg (14). Preparation of Actin from Other Sources-

A monomeric actin fraction in F-buffer or KI- Rabbit skeletal muscle actin was prepared accord

buffer, which contained less than 0.4 mg actin, ing to Spudich and Watt (17). Dictyostelium was applied to a column (1 ml disposable syringe) actin was prepared from amoebae of Dictyostelium containing 0.3 ml DNase I-bound agarose (Worth- discoideum by Dr. Susan S. Brown (11).

J. Biochem. SOLUBLE ACTIN FROM SEA URCHIN EGGS 789

Electron Microscopy-Samples were mounted

on a carbon-coated formvar grid, negatively stained RESULTS with 1 % aqueous uranyl acetate (18) and viewed with a Siemens Elmiskop I electron microscope. Actin Distribution in the Egg Fractions-The Protein Determination-Protein was deter actin concentration in the sea urchin egg was mined by the method of Lowry et al. (19) using calculated from the summation of the actin con bovine serum albumin as a standard. Prior to tents in P1, P2, and S2 to be about 3 mg/ml. As the determination, the pelleted crude actin fraction shown in Table I, about 50% of the actin was was dissolved in a solution of 1 % SDS, 5 mM recovered in S2, a high-speed supernate of an EGTA, 5 mM TAME, and 10 mM TES, pH 7.5. extract prepared in F-buffer 1. The G-actin/total Pelleted pure actin was dissolved in G-buffer by actin ratio in S2 was estimated by means of the sonication. DNase I inhibition assay to be 80-90% o (Table

‡U). This ratio did not change dramatically when the pH of S2 was changed to 7.4 or to 8.0, when

TABLE I. Purification of actin from S. purpuratus eggs.

Initial amount of packed eggs: 16.6 ml. Total protein concentration in an egg: 156 mg/ml. Total actin concentra tion in an egg: 3.1 mg/ml. a Actin pelleted after the ammonium sulfate fractionation and then depolymerized. b Purified actin obtained by one cycle of polymerization-depolymerization of G-actin 1.

TABLE ‡U. G-actin concentration in the crude supernate of S. purpuratus eggs. The high-speed sup (S2) was kept at 0•Ž at the indicated pH (adjusted by addition of 1 M TES, pH 9.0) overnight, or dialyzed against F-buffer 2 overnight, or incubated at 25•Ž for 1 h after standing overnight at 0•Ž at the indicated pH. The Ca2+ con centration is the final free Ca2+ concentration. Determination of the G-actin and total actin concentrations was carried out by DNase I inhibition assay (7) with slight modifications (see the legend of Fig. 1).

Vol. 87, No. 3, 1980 790 I. MABUCHI and J.A. SPUDICH

Ca2+ was added to give a final free Ca2+ concen speed. The critical concentrations was obtained tration of 0.5 mm, when S2 was incubated at 25•Ž by densitometry of SDS-polyacrylamide gels of

for 1 h at pH 6.8 or 7.4 with or without Ca2+, or the supernate and the pellets (Fig. 3). The values

when S2 was dialyzed against F-buffer 3. were 0.34 mg/ml at 25•Ž and 0.68 mg/ml at 0•Ž.

When S2 was fractionated with a Sephadex It should be noted that the G-actin concentration G-150 column, about 80% of the actin eluted as was not constant but increased with increasing

monomers of a molecular weight of about 44,000 protein concentration. (Fig. 2), which is in good agreement with the value The determination of critical concentrations from the DNase I inhibition assay (Table II). for assembly by means of the DNase I inhibition Low molecular weight proteins were analyzed on a 12 % gel and we found little evidence of a protein of a molecular weight of about 16,000 like spleen profilin (20) co-eluting with actin (Fig. 2). Critical Concentration for Polymerization of

Crude Actin-The fact that monomeric actin is

present in S2 at a concentration of more than 1 mg/ml (Table ‡U) is similar to observations for

a number of cell types (see ref. 8). This is of interest because the critical concentrations of

purified muscle or nonmuscle actins are less than 0.05 mg/ml under these polymerizing conditions

(see refs. 7 and 8). Therefore, we investigated the critical concentration of the G-150 actin in detail.

Since the actin content in this fraction was only

about 10% (Fig. 4, Table I), viscometry or spectro

photometry was not adequate. Thus we used sedimentation and the DNase I inhibition assays.

Only a small amount of precipitates was visible when the G-150 actin was concentrated to about

2 mg/ml and incubated in F-buffer 2 for 48 h at

0•Ž. However, upon incubation at 25•Ž the

solution became cloudy. Electron microscopy

revealed that amorphous aggregates had formed. Fig. 2. An elution profile of a high-speed supernate Among these aggregates, some single actin filaments (S2) of S. purpuratus eggs from a Sephadex G-150 and paracrystalline actin bundles were observed column. S2 (20 ml) was applied to the column (4 by electron microscopy. These bundles looked •~ 90 cm) which had been equilibrated with F-buffer 2. very similar to those found in a gel formed in a (a) •œ, A29onm; •›, relative actin amount obtained by soluble extract of Hawaiian sea urchin egg upon SDS-PAGE-densitometry. The insert shows the deter warming to 37•Ž (21). These visible aggregates mination of the molecular weight of the actin eluted

sedimented at 20,000 •~ g for 15 min; about half of from this column. Marker proteins are ƒÁ-G: human ƒÁ-globulin (molecular weight 160 the polymerized actin was found in these low ,000), BSA: bovine speed pellets. The other half sedimented at high serum albumin (molecular weight 68,000), OA: oval bumin (molecular weight 43,000), and STI: soybean 3 The concentration below which a solute does not trypsin inhibitor (molecular weight 20,000). The arrow crystallize or polymerize is called the critical concen indicates the position of the eluted actin peak. (b) tration and this concentration is constant being in SDS-PAGE (12% gel) pattern of the eluted fractions. dependent of the total solute concentration. This term Marker proteins (the right and left channels) are rabbit will be used here as a matter of convenience although skeletal muscle actin (molecular weight 42,000), carbonic it appeared that the G-actin concentration in the G-150 anhydrase (molecular weight 30,000), soybean trypsin actin fraction was not constant when the total actin inhibitor (molecular weight 20,000), and cytochrome c

concentration was varied. (molecular weight 13,000).

J. Biochem. SOLUBLE ACTIN FROM SEA URCHIN EGGS 791

Vol. 87, No. 3, 1980 792 1. MABUCHI and J.A. SPUDICH

Fig. 3. Determination of the critical concentration of crude egg actin. The G-150 actin fraction was concentrated with Aquacide ‡U-A to obtain appropriate actin concen trations and kept at 0•Ž for 48 h. Aliquots were further incubated at 25•Ž for 2 h for the 25•Ž determinations. (a) Determined by sedimentation followed by SDS-PAGE- densitometry. Centrifugation was done so that materials of greater than 14S would

sediment. Both the supernates and pellets were electrophoresed, and actins in these fractions were calculated after scanning the gel. The symbols show the concentrations of actin larger than 14S at 25•Ž (•œ), in the supernate at 25•Ž (•ü), larger than 14S at 0•Ž

(A), and in the supernate at 0•Ž (•¢). (b) Determined by the DNase I inhibition assay. See the legend of Fig. 1 for the procedure. •œ, Total actin concentration determined in the presence of guanidine .HCI; •ü, G-actin concentration at 25•Ž; •¢, G-actin concen tration at 0•Ž.

assay gave similar values, i.e., 0.45 and 0.9 mg/ml similar to the case of Dictyostelium actin (11), at 25 and 0•Ž, respectively (Fig. 3). The G-actin addition of ammonium sulfate up to 35% satura concentration was again observed to increase with tion left a considerable amount of actin in the increasing protein concentration. supernate. However, precipitation at 60% ammo Purification of Egg Actin-When the G-150 nium sulfate saturation followed by backwashes actin fraction (fractions 61-82 in Fig. 2) was applied was found to be efficient for egg actin recovery to a DEAE-cellulose column, about 40% of the as in the case of Dictyosteliu,n (11).

protein was adsorbed to the column and a protein Egg actin was recovered as F-actin after the

peak eluted at about 0.18 rat KCl. Actin was ammonium sulfate fractionation. After depoly found in the latter half of the peak. The critical merization, this actin was repolymerized into concentration of this DEAE actin was determined filaments. However, although the purity of this by the DNase I inhibition assay to be 0.03 mg/ml actin was as high as 95 % (Table I), the poly at 25•Ž and 0.12 mg/ml at 0•Ž. In this case, merization induced by 2 mM MgCl2 was slower

J. Biochem. SOLUBLE ACTIN FROM SEA URCHIN EGGS 793

Fig. 4. Purification of egg actin. SDS-PAGE gels of the fractions during the puri fication were scanned with a Transidyne gel scanner. The actin content was esti mated from the shaded area. (a) Crude supernate, S2. (b) G-150 actin. (c) DEAE actin. (d) Ammonium sulfate-fractionated actin. (e) Purified actin.

Vol. 87, No. 3, 1980 794 I. MABUCHI and J.A. SPUDICH

Fig. 5.

J. Biochem. SOLUBLE ACTIN FROM SEA URCHIN EGGS 795

Fig. 6.

Vol. 87, No. 3, 1980 796 I. MABUCHI and J.A. SPUDICH

Fig. 7-8.

Figs. 5-8. Electron micrographs of polymerized egg actin. Fig. 5: Actin polymerized in 75 mm KCl. Magnifi

cation, •~ 114,000. Fig. 6: Polymerized in 10 mat CaCl2, •~ 19,800. Fig. 7: Polymerized in 10 mm CaCl2, •~ 100,000. Fig. 8: Polymerized in 50 mm CaCl2, •~ 100,000.

J. Biochem. SOLUBLE ACTIN FROM SEA URCHIN EGGS 797 and the viscosity was lower than that induced by induced one, and the final viscosities were almost 75 mM KCl, contrary to the cases of other non- the same (0.5-0.8 liter/g). The purity of this "purified actin" was more than 96% (Fig muscle and muscle actins. After one cycle of . 4, Table polymerization-depolymerization, the Mg2+-in- I). Electron microscopy revealed very long fila duced polymerization was faster than the KCl- ments-(Fig. 5).

Fig. 9. Determination of the critical concentration of purified egg actin. (a) Determined by sedi mentation at 25•Ž. Actin was polymerized with 75 mM KCl, 2 mM MgCl2, and 10 mM M ES buffer at pH 6.85. (b) Determined by the DNase I inhibition assay. For the procedures and symbols, see the legend of Fig. 3.

TABLE ‡V. Critical concentration of S. purpuratus egg actin.

a G-actin concentration increased with increasing protein concentration.

Vol. 87, No. 3, 1980 798 I. MABUCHI and J.A. SPUDICH

The yield of the actin from the egg homog When 10 mM MgCl2 or CaCl2 was added enates was more than 20% and that from the G-150 instead of 2 mM MgCl2 or 75 mM KCl, two dimen monomeric actin was usually greater than 40 sional nets were formed (Figs. 6 and 7). The and it was 62 % in the example in Table I. filament overlapping periodicity of these nets was

Polymerization of Purified Actin-The time about 380A and the angle between overlapped course of polymerization was measured by the filaments was about 30•‹ A similar net formation

DNase I inhibition assay or by viscometry at an was reported with rabbit skeletal muscle actin actin concentration of 0.5 mg/ml. The half time either at low pH (22, 23) or with 8-12 mM divalent for polymerization was 1 min with 2 mM MgCl2 cations (24), or with Dictyostelium actin in 10 mM at 25•Ž, 2 min with 75 mM KCl at 25•Ž, and 13 CaCl2 (25). min with both the salts at 0•Ž.

Fig. 10. SDS-polyacrylamide gel elec trophoresis of DNase I-affinity chro matographed actins. The actins were

purified through the DNase I-agarose column as described in " MATE- RIALS AND METHODS" and elec trophoresed in a 12% polyacrylamide

gel. Eight to 9 ƒÊg of protein was run in each column. a, actin fraction from Pl ; b, from P2; c, from S2; d, from the oligomeric actin fraction from the Sephadex G-150 column; e, from the monomeric actin fraction from the Sephadex G-150 column; f, marker proteins, the same as de scribed in the legend of Fig. 2b. The arrowheads indicate low molec ular weight protein bands referred to in the text.

J. Biochem. SOLUBLE ACTIN FROM SEA URCHIN EGGS 799

Fig. 11. Isoelectric focusing of actin from various fractions. DNase I- affinity chromatographed actins from the fractions indicated below were focused in gels according to the method of O'Farrell (15) as modified by Rubenstein and Spudich (16). a, rabbit skeletal muscle actin; b, Dictyostelium actin; c, egg actin purified by the routine procedure described in " MATERIALS AND METHODS;" d, a mixture of a-c; e-i, egg actins recovered from the DNase I-agarose column; e, Se phadex G-150 monomeric actin; f, Sephadex G-150 oligomeric actin; g, actin from S2; h, actin from P1 ; i, actin from P2.

Vol. 87, No. 3, 1980 800 I. MABUCHI and J.A. SPUDICH

With 50 mm divalent cations, egg actin formed was 0.1 for PI or P2 and 0.15 for G-150-I. A paracrystalline rods (Fig. 8) similar to rabbit 16,000 molecular weight component co-eluted with skeletal muscle actin (26), human platelet actin actin from G-150-II. The molar ratio of this (27), and Dictyostelium actin (28). component to actin was as low as 0.015 (Fig. 10). The critical concentration of purified egg We found that all of these actins migrated as

actin for polymerization was determined by three a single band with the same isoelectric point (Fig.

methods as described in " MATERIALS AND 11). Therefore, with respect to the isoelectric

METHODS." All methods gave similar values, point all the actins in the sea urchin egg were the 0.02-0.03 mg/ml (Table III). The G-actin con same. Egg actin showed an isoelectric point more

centration of the purified egg actin remained basic than rabbit skeletal muscle actin or Dictyo constant being independent of the total actin stelium actin; it comigrates with ƒÀ-actin from

concentration up to 1 mg/ml (Fig. 9). A little chick embryo fibroblasts (31).

increase in the G-actin concentration with increas

ing total actin concentration was observed when DISCUSSION the determination was carried out by sedimentation

at 0•Ž (Table III). This may be due to the presence The method used here for the purification of sea of short actin filaments which did not sediment urchin egg actin gave a good yield of highly purified under the conditions employed here. The critical actin. Methods previously used for the isolation concentration of rabbit skeletal muscle actin deter of actin from echinoderm eggs (32-34) were based mined by the DNase I inhibition assay was 0.025 on the method for purification of actin from

mg/ml, in good agreement with earlier values Physarum plasmodia developed by Hatano and

obtained by other methods (11, 25, 29, 30). Oosawa (35) which utilizes acetone treatment of Isoelectric Focusing of Egg Actin-It is possible the cells and specific binding of actin to exogenous that we purified an actin species present in the myosin. Although the purity was not described G-150 actin fraction which was competent to in detail in these reports, it is unlikely that the polymerize and discarded actin species which were purity exceeded that described here. In addition, unable to polymerize. One piece of evidence the present egg actin showed very high polymeri against this is that the yield of the purified actin zability as judged from the viscosity and the was more than 40% of that present in the G-150 electron microscope images of the polymerized fraction (Table I). To investigate this possibility form. In contrast, previous preparations showed further, we examined various pools of actin in relatively low viscosity (33) or short filaments the sea urchin egg by isoelectric focusing. (34, 36) which might be due to contamination by Actins were purified from various egg frac substances which influence actin polymerization.

tions [i.e., P1, P2, S2, the oligomeric actin fraction The egg actin showed an isoelectric point

from the Sephadex G-150 column (fractions 45- more basic than Dictyostelium or rabbit skeletal 59 in Fig. 2, G-150-I), and the monomeric actin muscle actin. Spudich and Spudich (31) showed

fraction from the Sephadex G-150 column (the that the actin from cortical fragments of unfertilized

same as the G-150 actin mentioned above, G-150- S. purpuratus eggs comigrates with ƒÀ-actin of chick

‡U)] with DNase I-bound agarose columns (Worth- embryo fibroblasts in an isoelectric focusing gel. ington Biochem. Corp.) (14). Actins in the first It is interesting that the sea urchin egg has only

four fractions were depolymerized by dialysis one isoelectric actin species. Nonmuscle cells

against KI-buffer before application to the DNase from mammalian tissues generally have both ƒÀ-

I column. Almost all the actin bound to the and ƒÁ-actin (16, 37, 38) while lower organisms column and was eluted with 3 M guanidine.HCl appear to have only one isoelectric actin species after a 0.75 M guanidine•HCl wash. The purities (11, 30). of actins in the 3 M guanidine•HCl-eluates from The polymerization of egg actin with divalent the five fractions were 72 %, 72 %, 55 %, 58 %, and cations showed similarities to the cases of other

89%, respectively. An 18,000 molecular weight actins such as Dictyostelium or rabbit skeletal component co-eluted with actin from P1, P2, and muscle actin. With 10 mM Ca2+ or Mg2+, egg

G-150-I. The molar ratio of this protein to actin actin formed two dimensional nets which looked

J. Biochem. OLUBLE ACTIN FROM SEA URCHIN EGGS 801

like the skeletal muscle actin nets made at low spleen profilin (20) co-eluting with actin from pH (22, 23) or the Dictyostelium actin nets formed this column. A protein of this molecular weight with 10 mM Call (22). Recently, it was shown was detected in this fraction with the DNase I that skeletal muscle actin formed similar nets affinity chromatography; however, the molar ratio with 8-12 mM divalent cations at neutral pH (24). of this protein to the actin was less than 0.02. The critical concentrations of egg actin for Thus, the controls operating on the assembly of

polymerization are summarized in Table III. The the actin in sea urchin eggs may not involve a critical concentration of the purified egg actin was profilin-like molecule. We are currently examining

similar to that of muscle or other nonmuscle actins. the mechanism whereby the assembly of the actin Spudich and Cooke (25) and Uyemura et al. (11) in the G-150 fraction is inhibited.

reported that skeletal muscle actin and Dictyo

stelium actin have a critical concentration for REFERENCES Sassembly of 0.02-0.05 mg/ml, determined by a 1. Mabuchi, I. & Spudich, J.A. (1978) Sixth Int. change in absorbance at 232 urn or by viscometry. Biophys. Congr. Abstr. (Kyoto) p. 304 These values are in very good agreement with the 2. Mabuchi, 1. (1979) in Cell Motility: Molecules and values for sea urchin egg actin and skeletal muscle Organization (Hatano, S., Ishikawa, H., & Sato, H., actin described here using viscometry, sedimenta eds.), pp. 147-163, University of Tokyo Press, Tokyo

tion, or the DNase I inhibition assay. Gordon 3. Schroeder, T.E. (1975) in Molecules and Cell Move

et al. (30) reported that skeletal muscle and a ment (Inoue, S. & Stephens, R.E., eds.) pp. 334-352, number of nonmuscle actins (i.e., human platelet, Raven Press, New York 4. Schroeder, T.E. (1972) J. Cell Biol. 53, 419-434 embryonic chick brain, rat liver, and Acantha 5. Burgess, D.R. & Schroeder, T.E. (1977) J. Cell moeba actins) had a critical concentration of Biol. 74, 1032-1037 0.02-0.09 mg/ml in 2 mM MgCl2 at either 25 or 6. Bray, D. & Thomas, C. (1976) in Cell Motility 5•Ž or in 0.1 M KCl at 25•Ž. (Goldman, R., Pollard, T., & Rosenbaum, J., eds.) The G-actin concentration in the crude egg pp. 461-473, Cold Spring Harbor Laboratory, Cold fraction (S2) was very high. This phenomenon Spring Harbor, New York seemed to be independent of pH, the presence of 7. Clarke, M. & Spudich, J.A. (1977) Ann. Rev. Bio Ca2+, or the presence of unknown dialyzable com chem. 46, 797-822

ponents, although some polymerization, that is, 8. Korn, E.D. (1978) Proc. Nat!. Acad. Sci. U.S. 75, about 20%. of the total actin present (see Table II), 588-599 9. Rees, M.K. & Young, M. (1967) J. Biol. Chem. occurred upon raising the temperature from 0 242,4449-4458 to 25•Ž. 10. Gordon, D.J., Eisenberg, E., & Korn, E.D. (1976) Two characteristic features were observed with J. Biol. Chem. 251, 4778-4786 the G-150 actin. The critical concentration was 11. Uyemura, D.G., Brown, S.S., & Spudich, J.A. (1978) high and the G-actin concentration increased with J. Biol. Chem. 253, 9088-9096 increasing protein concentration in contrast to 12. Laemmli, U.K. (1970) Nature 227, 680-685 the purified actin. On the other hand, it was 13. Blikstad, I., Markey, F., Carlsson, L., Persson, T., strongly suggested that the G-150 actin and the & Lindberg, U. (1978) Cell 15, 935-943 purified actin were the same protein because of 14. Lindberg, U. (1974) in Methods in Enzymology the same isoelectric point and the high yield in (Jacoby, W.B. & Wilcheck, M., eds.) Vol. 34, pp. the purification. Therefore these features of the 517-520, Academic Press, New York G-150 actin may indicate the presence of factors 15. O'Farrell, P. (1975) J. Biol. Chem. 250, 4007-4021 16. Rubenstein, P.A. & Spudich, J.A. (1977) Proc. which inhibit actin polymerization, although the Natl. Acad. Sci. U.S. 74,120-123 mechanism of the inhibition is not clear. We 17. Spudich, J.A. & Watt, S. (1971) J. Biol. Chem. 246, looked for evidence of a profilin-like molecule, 4866-4871 such as that described by Carlsson et al. (20). 18. Huxley, H.E. (1963) J. Mot. Biol. 7, 281-301 The actin in S2 eluted as monomers of a molecular 19. Lowry, O.H., Rosebrough, N.J., Farr, A.L., & weight of 44,000 from the Sephadex G-150 column. Randall, R.3. (1951) J. Biol. Chem. 193, 265-275 We did not find a stoichiometric amount of a 20. Carlsson, L., Nystrom, L.-E., Sundkvist, I., Markey, protein of a molecular weight of 16,000, such as F., & Lindberg, U. (1977) J. Mot. Biol. 115,465-483

Vol. 87, No. 3, 1980 802 I. MABUCHI and J.A. SPUDICH

21. Kane, R.E. (1975) J. Cell Biol. 66, 305-315 30. Gordon, D.J., Boyer, J.L., & Korn, E.D. (1977) 22. Kawamura, M. & Maruyama, K. (1970) J. Biochem. J. Biol. Chem. 252, 8300-8309 68,885-899 31. Spudich, A. & Spudich, J.A. (1979) J. Cell Biol. 23. Yamamoto, K., Yanagida, M., Kawamura, M., 82,212-226 Maruyama, K., & Noda, H. (1975) J. Mol. Biol. 91, 32. Hatano, S., Kondo, H., & Miki-Noumura, T. (1969) 463-469 Exptl. Cell Res. 55, 275-277 24. Strzelecka-Golaszewska, H., Pr6chniewicz, E., & 33. Miki-Noumura, T. & Oosawa, F. (1969) Exptl. Drabikowski, W. (1978) Eur. J. Biochem. 88, 219- Cell Res. 56, 224-232 227 34. Mabuchi, I. (1976) J. Mol. Biol. 100, 569-582 25. Spudich, J.A. & Cooke, R. (1975) J. Biol. Chem. 35. Hatano, S. & Oosawa, F. (1966) Biochim. Biophys. 250,7485-7491 Acta 127, 488-498 26. Hanson, J. (1968) in Symposia Biologica Hungarica 36. Miki-Noumura, T. & Kondo, H. (1970) Exptl. (Ernst, E. & Straub, F.B., eds.) Vol. 8, pp. 99-103, Cell Res: 61,31-41 Academiai Kiado, Budapest 37. Whalen, R.G., Butler-Browne, G.S., & Gros, F. 27. Spudich, J.A. (1972) Cold Spring Harbor Symp. (1976) Proc. Natl. Acad. Sci. U.S. 73, 2018-2022 Quant. Biol. 37, 585-593 38. Garrels, J.I. & Gibson, W. (1976) Cell 9, 793-805 28. Spudich, J.A. (1974) J. Biol. Chem. 249, 6013-6020 29. Kasai, M., Asakura, S., & Oosawa, F. (1962) Bio chim. Biophys. Acta 57,13-21

J. Biochem.