THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 258. No. 2, Issue of January 25, pp. 1352-1356,1983 Printed in U.S.A.

The Periodic Synthesis of Tubulin in the Physarurn Cell Cycle CHARACTERIZATION OF PHYSARUM TUBULINS BY AFFINITY FOR MONOCLONAL ANTIBODIES AND BY PEPTIDE MAPPING*

(Received for publication, May 26, 1982)

Ming Tu Chang$, William F. Dove& and Thomas G. Lafflerl From the McArdle Laboratory for Cancer Research, University of Wisconsin, Madison, Wisconsin 53706

Polypeptides preferentially labeled in the G2 phase of the synchronous nuclear replication cycle of Physa- rum macroplasmodia were compared in electropho- retic mobility and peptide map with the tubulins en- riched from Physarum myxamoebae. One major and one minor fluorographic species match the myxamoe- bal a and fi chains, respectively. Thus, tubulins are among the proteins of Physarum selectively synthe- sized before nuclear division. A third species P, prom- inently labeled in premitotic plasmodia, is distinct from the two myxamoebal tubulins even though it co-po- lymerizes with microtubules. The nature of P remains unknown.

Two rat monoclonal antibodies directed against yeast tubulin were found to bind selectively to the a: tubulin of porcine brain. These served to confirm the assign- ment of the 50,000-dalton Physarum myxamoebal tu- bulin as an a-like polypeptide.

The natural synchrony of the syncytial macroplasmodium of Physarumpolycephalum offers an opportunity to study the regulation of macromolecular synthesis in the nuclear division cycle (1). Previous work from our laboratory has shown that several polypeptides capable of co-polymerization with por- cine brain microtubular protein are synthesized with pro- nounced periodicity (2). Where it has been measured, the rate of synthesis of these species is at least 30-fold greater before nuclear division than after. The molecular weights and isoe- lectric points of these periodic microtubular species resemble those of porcine brain tubulin in two-dimensional electro- pherograms.

Although it has not been possible to purify tubulin from Physarum plasmodia (3,4), Gull and his colleagues (5,6) have been able to purify by self-assembly two polypeptides from the myxamoebal form of Physarum. Defiied by self-assembly, one myxamoebal tubulin has been found to be partially d i k e and the other to be extensively P-like when compared to the mammalian tubulins (7).

In this report, we compare the periodic microtubular poly-

* This research has been supported in part by Program-Project Grant CA-23076 in Tumor Biology and Cancer Center Support Grant CA-07175 to the McArdle Laboratory from the National Cancer Institute. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

-$ Recipient of Postdoctoral Fellowship F32GM07481 from the Na- tional Institutes of Health. Present address, Food Research Institute, University of Wisconsin, Madison, WI 53706.

5 To whom to address correspondence. 1 Present address, Department of Microbiology and Immunology,

Northwestern University Medical School, Chicago, IL 60611.

peptides of plasmodia with the (Y and p tubulins from Physa- rum myxamoebae, using as criteria mobility on two-dimen- sional electropherograms, binding of monoclonal antitubulins (8), and peptide map by partial V8 proteolysis.

MATERIALS AND METHODS

Cultures-P. polycephalum myxamoebae (strain CLd-AXE (9)) were grown at 22 "C in a modified semidefined medium (10) in which 1% Oxoid mycological peptone substituted for casein hydrolysate. Microplasmodia and plasmodia (strain CL (ATCC 24112)) were grown as described by Laffler et al. (2).

Electrophoresis-SDS-PAGE' was carried out as described by Laemmli ( l l ) , utilizing technical grade SDS from Matheson, Coleman and Bell which allows separation of both porcine and Physarum tubulin subunits. Other reagents were from Bio-Rad. All slab gels were 1.5-mm thick. Molecular size markers, also from Bio-Rad, in- cluded phosphorylase b, 92,500, bovine serum albumin, 66,200, oval- bumin, 43,000; soybean trypsin inhibitor, 21,000; and lysozyme, 14,400. Gels were stained directly for protein with Coomassie brilliant blue (0.25% in 50% methanol, 10% acetic acid) and destained in 20% methanol with 10% acetic acid. Two-dimensional electrophoresis fol- lowed the O'Farrell procedure we have described (2), except that 0.4% SDS was included in the isoelectric focusing step (12). pH values were determined as described by O'Farrell (13) using degassed double- distilled water.

Protein Blot Transfer onto Nitrocellulose-After SDS-PAGE, the gel was equilibrated by gently shaking for 15-30 min in transfer buffer (50 lll~ NaCl, 2 mM Na-EDTA, 0.1 m~ dithiothreitol, 10 mM Tris- HCl, pH 7.0) supplemented with 4 M urea. The transfer of protein from the gels to nitrocellulose paper (BA85, Schleicher & Schuell) was then accomplished by forming a sandwich with two sheets of nitrocellulose and allowing diffusion for 12-16 h in 2 liters of transfer buffer (14). Of the two protein blot transfers, one was normally used for protein staining and the other for testing antibody binding. Protein was stained with amido black (0.1% in 45% methanol, 10% acetic acid) for 30 min and then destained with 90% methanol and 2% acetic acid.

Radioimmune Binding to Nitrocellulose Blot Transfers-The pre- pared filter was placed in a plastic container and immersed for 1-2 h a t 37 "C in at least 50 ml of blocking buffer (1% bovine serum albumin, 83 lll~ NaC1,8.5 mM KCl, 2.5 mM MgC12,0.87 m~ sodium phosphate, pH 7.4). Subsequent reactions and washings were performed in a reciprocal shaking water bath at 37 "C in blocking buffer to reduce background binding.

After decanting the blocking buffer, diluted rat antitubulin was added at 100 pl/cm2 of filter and incubated for 1 h. The filter was fust washed once briefly and then three times for 15 min each. The second antibody, rabbit anti-rat IgG (Miles), was added at a dilution of 500 in a volume of 100 pl/cm2 of filter and incubated for 1 h. The washing procedure was then repeated. Finally, the developing reagent, iodi- nated protein A (?IO7 cpm/pg), was added in a volume of 100 pl/cm2 of filter at a specific activity of lo5 cpm/ml. Protein A was obtained from Pharmacia and iodinated by the method of Nowinski et al. (15). After a 1-h incubation, the final washes were performed at least four times for 1-2 h each, using blocking buffer that lacked bovine serum albumin. The filter was dried and autoradiographed on Kodak NS- 2T frlm with an x-ray intensifying screen (Ilford, fast tungstate) at

' The abbreviation used is: SDS-PAGE, sodium dodecyl sulfate- polyacrylamide gel electrophoresis.

1352

Physarum

-70 OC for 16 h unless specified otherwise. Vinblastine Precipitation of Physarum Protein-Vinblastine pre-

cipitation was carried out as described by Kemphues et al. (16) modified as follows. One-tenth volume of vinblastine sulfate (Sigma) a t 22 mg/ml was added to the clarified sonicate. After 1-2 h on ice, the precipitate was collected by centrifugation at 39,000 X g for 30 min a t 4 “C. The pellet was dissolved in 1-2 ml of resuspension buffer (0.1 M 1,4-piperazinediethanesulfonic acid, pH 6.94, 1 m~ ethylene glycol bis (P-aminoethyl ether)-N,N,N’,N’-tetraacetic acid, 0.1 mM MgC12, 1% Trasylol) supplemented with 2 m~ GTP. The protease inhibitor Trasylol was obtained from Mobay Chemical Corp., FBA Pharmaceuticals. The milky white supernatant was collected after brief centrifugation and stored in aliquots at -70 “C.

Purification of Microtubular Protein from Porcine Brain a n d Physarum-Porcine brain microtubular protein and purified brain tubulin (17) were kindly provided by Dr. Robert Scheele (University of Wisconsin). Physarum myxamoebal microtubular proteins were purified by self-assembly (18).

Hybridoma Cell Lines a n d Antibodies-Kilmartin and his col- leagues (8) have prepared monoclonal antibodies to tubulin by fusing with rat myeloma cells the splenocytes of rats hyperimmunized with partially purified yeast tubulin. Among the antibodies that were positive for binding to yeast tubulin, a subset was also positive with chick brain tubulin. Two of these cross-reactive antibodies, YL1/2 (“2”) and YOL1/34 (“34”), were unique in also staining the mitotic spindle of yeast and the interphase microtubules of Chinese hamster ovary and 3T3 cells, as judged by indirect immunofluorescence. Hybridoma 2 emerged from a fusion employing the rat myeloma line Y3/Ag1.2.3, secreting its own light chain. By contrast, line 34 emerged from a fusion employing the nonsecreting rat myeloma line YB2/0. Both antibodies, 2 and 34, are of the IgG class (2). We confirmed that both antibodies were active in binding porcine tubulin using a solid phase microtiter well binding assay modified from that of Nowinski et al. (15). In addition, we also found that antibodies stained porcine microtubules polymerized in vitro by indirect immunofluorescence.

Tubulin Peptide Map from Two-dimensional Gels-A labeled plasmodial extract was mixed with myxamoebal tubulins (the pellet fraction from the third cycle of warm assembly, H:IP) and co-electro- phoresed on two-dimensional gels. Regions of the gel corresponding to the periodic microtubular proteins were detected by fluorography and cut out. Partial proteolysis by Staphylococcus aureus V8 protease (19) was analyzed on 15% polyacrylamide gels. The gels were stained with Coomassie brilliant blue, treated with EN’HANCE (New Eng- land Nuclear), dried, and exposed to Kodak XAH-5 film for 42-48 days.

RESULTS

The timed synthesis of microtubular proteins in the Phy- sarum plasmodium can be demonstrated by radiolabeling polypeptides that are resolved by two-dimensional electro- phoresis (2). When a surface plasmodium was incubated with [:‘“S]methionine for 2 h before nuclear division, the microtu- bular species were labeled maximally. The total plasmodial extract was then analyzed on a two-dimensional O’Farrell gel. A fluorogram of the region of interest is shown in Fig. 1. The G2-labeled polypeptides of the plasmodium lie in the molec- ular size range of 48,000-57,OOO and the PI range of 5.08-5.56. The intense set of spots to the lower left of the electrophero- gram may be actin (“A”; apparent mass 42,500 daltons; ap- parent PI 5.56). We have previously shown that co-polymeri- zation with porcine brain tubulin enriches the periodic species but not the presumptive actin (2).

Purification and Antigenic Characterization of Physarum Tubulins-We tested whether the interspecies reactivity of rat monoclonal antibodies raised against yeast tubulins could be used to identify tubulin-related polypeptides in the myxo- mycete P. polycephalum. Tubulin from myxamoebae of Phy- sarum was purified by self-assembly (6, 7) or enriched by vinblastine precipitation. Polypeptides were resolved by SDS- PAGE and a blot transfer was tested for reactivity to mono- clonal antitubulin antibodies.

We found that antibody 2 strongly labeled the a subunit of porcine tubulin of molecular weight 56,500 (Fig. 2, lane b).

Tubulins 1353

ISoaECTRlc FOCUSlNG

FIG. 1. Periodic microtubular proteins in the plasmodium of P. polycephalum. One-half of a plasmodium was labeled with 200 pCi of [:‘“S]methionine for 2 h before mitosis I1 (1). The total plas- modial extract (2,000,000 dpm/gel) was resolved on a two-dimensional O’Farrell gel and detected by fluorography. Only the middle portion of the lower half of the gel is shown here. Regions are indicated that correspond to a tubulin, B tubulin, presumptive actin (A ) , and the periodic microtubular polypeptide P.

a b c d e f g h i j

92500- - -7 - 43.000-

i

FIG. 2. Immunostaining of Physarum microtubular proteins by antibody 2. Self-assembled myxamoebal protein a t H:IP (15 pg) was mixed with molecular weight markers and run in lanes a and j . Porcine brain tubulin (4 pg) was run in lane b; 200 pg of vinblastine- fractionated amoeba1 proteins in lane d (precipitate fraction) and in lane c (supernatant fraction); and 200 pg of vinblastine-fractionated microplasmodial proteins in lane f (precipitate fraction) and in lane e (supernatant fraction). Vinblastine-treated microplasmodial mate- rial was further concentrated by acetone precipitation and run in lanes g-i: lane g, 200 pg of vinblastine supernatant; lane h, 200 pg of vinblastine precipitate; and lane i, 400 pg of vinblastine precipitate. The top section is a nitrocellulose protein blot immunostained with anti-a tubulin antibody; the bottom section is a gel stained for total protein. Heavy arrows locate the polypeptide bands that bind anti-a antibody.

Physarum Tubulins 1354

v, v)

LL W

0 I a 0 LL I- o W

W J

W J

a v) 0 v)

ISOELECTRIC FOCUSING

OH" Hi

FIG. 3. Identification of Physarum myxamoebal tubulin on a two-dimensional gel. Physarum amoebal tubulin (H:,P) was resolved as shown in the central area. Included in the second dimen- sion SDS gel were size markers plus porcine brain tubulin (4 pg) in the left lane and vinblastine-precipitated myxamoebal protein in the right lane. Molecular masses are expressed in kilodaltons. The top section is a nitrocellulose protein blot stained for total protein; the bottom section is a second blot from the same gel immunostained with anti-a tubulin antibody. Heavy arrows depict the protein band that binds antibody specific for the a tubulin of porcine brain.

Antibody 34 exhibited the same reactivity as 2 (data not shown). The a-specific antibody 2 reacted also with the faster moving of the two polypeptides of self-assembled Physarum amoebal tubulin (Fig. 2, lanes a and 1). The electrophoretic mobility of the a-like polypeptide is similar to the 50,000- dalton band of the vinblastine-precipitated material (Fig. 2, lane d ) . Notice that antibody 2 binding shows that the a-like Physarum tubulin has been enriched in the vinblastine pre- cipitation compared with the supernatant (Fig. 2, lane d uersus c ) , whereas Coomassie blue staining did not reveal such a clear distinction.

We also examined whether the tubulin of Physarum micro- plasmodia could be enriched by vinblastine precipitation on the basis of similar criteria. We did not detect any material stainable by Coomassie blue in the tubulin regions covering both porcine and Physarum amoebal tubulins (Fig. 2, lanes e-i). Antibody 2 also failed to identify any a-like tubulin in the microplasmodial material. Surface plasmodia were found also to lack polypeptides in the tubulin region (data not shown).

A minor reaction of antibody 2 can be seen in Fig. 2, lanes c, d , and i . This 57,000-dalton band is not enriched by vin- blastine precipitation, is absent from self-assembled microtu- bules, and is not detected by antibody 34.

Comparison by Electrophoresis-Since plasmodia seem to contain very little tubulin, direct identification of the period- ically labeled polypeptides is not feasible. Therefore, we were obliged to compare the electrophoretic mobilities of both myxamoebal and porcine tubulin with those of radiolabeled periodic polypeptides. We found that the less acidic myxam- oebal tubulin subunit binds anti-a antibody (Fig. 3). This a-

like myxamoebal tubulin has an isoelectric point of 5.31; the second myxamoebal tubulin focuses at 5.08. As shown in Fig. 4, co-electrophoresis of labeled plasmodial polypeptides with myxamoebal tubulin revealed that two of the periodic species had electrophoretic mobilities similar to myxamoebal tubu- lins. The species that co-migrates with the myxamoebal p chain is minor in the fluorogram. However, in Fig. 5B, it can be seen as a prominent spot that co-migrates with the porcine p chain. This spot is enriched by co-polymerization. We could not, however, demonstrate identity of porcine a tubulin to any of the periodic plasmodial polypeptides even after co-polym- erization.

The major periodic polypeptide designated "P" by us (2) has an isoelectric point intermediate between those of a and /3 myxamoebal tubulins. I t was also enriched after co-polym- erization (Fig. 5B).

Comparison by Peptide Mapping-Further characteriza- tion of the periodic plasmodial molecules was achieved by peptide mapping after limited proteolysis (19). In Fig. 6A, myxamoebal tubulin is detected by protein staining, while the labeled plasmodial polypeptides from the co-electrophero- gram depicted in Fig. 4 are detected by fluorography. The quality of these peptide maps is limited by the very small amounts of material eluted from these two-dimensional gels. Myxamoebal CY tubulin (lane b) shows clear similarity to the periodic polypeptide with which it co-migrates (lane a). Myxamoebal p tubulin (lane d ) shows peptide map similarity

I

I

FIG. 4. Co-electrophoresis of periodic microtubular poly- peptides of the plasmodium with myxamoebal tubulin. A plas- modium was labeled with 200 pCi of [""Slmethionine for 2 h prior to the second synchronous mitosis and harvested around metaphase. The plasmodial extract (1.5 X 10" dpm/gel) alone (left) or mixed with 25 pg of H:IP myxamoebal tubulin (right) was resolved on a two- dimensional gel. Fluorograms were developed after 2 days of exposure. The dotted area outlines the region of myxamoebal a and p tubulin as stained by Coomassie blue. The autoradiographic images of labeled plasmodial species in regions corresponding to amoebal tubulim are evidently distorted by the carrier protein or quenched by the protein stain of the latter. The set of polypeptides designated P includes the region lying between a and p tubulin in PI and slightly higher in molecular weight.

FIG. 5. Enrichment of Physarum /I tubulin by co-polymeri- zation. A plasmodium was labeled before nuclear division. Proteins in the extract were mixed with carrier porcine brain tubulin and resolved by two-dimensional electrophoresis (A). The carrier was detected by staining and the labeled Physarum polypeptides by fluorography. Porcine tubulins are identified in the dotted area. After two rounds of co-polymerization, the pellet fraction was analyzed similarly (B) .

Physarum

'A". lrpr

4

a , . . .*:.

b C d e

a b C d FIG. 6. Peptide mapping of periodic microtubular proteins

from plasmodia and of tubulin from myxamoebae of Physarum and porcine brain. A, mixtures of trace amounts of ["'S]methionine- labeled plasmodial polypeptides and unlabeled myxamoebal tubulins were resolved as described in the legend to Fig. 4. The plasmodial peptides were detected by fluorography, and the corresponding re- gions of the gels were cut out and subjected to proteolysis by S. aureus V8 protease. Fluorograms showing the peptide maps of plas- modial species were developed after 42-48 days of exposure. Lane a, fluorogram of labeled plasmodial peptides that co-migrate with myxamoebal a tubulin; lane b, Coomassie blue staining pattern of peptides derived from myxamoebal a tubulin; lane c, fluorogram of the peptides derived from the labeled plasmodial polypeptides in the myxamoebal/3 tubulin region; lane d, Coomassie blue staining pattern of myxamoebal /3 tubulin; lane e, fluorogram of the labeled peptides derived from the major plasmodial microtubular polypeptide P. B, tubulin bands were eluted from a preparative SDS-PAGE gel and then digested with V8 protease. Proteolytic fragments were resolved

Tubulins 1355

to the periodic plasmodial polypeptide with which it co-mi- grates (lane c). The similarity of myxamoebal /3 tubulin to mammalian /3 tubulin is seen in Fig. 6B (lane d versus c). Some differences in myxamoebal a tubulin are seen in Fig. 6B, lane a. Finally, the prominent periodic species P (Fig. 6A, lane e ) resembles neither myxamoebal tubulin in peptide map. This unique cleavage pattern was seen over a range of digestion conditions for labeled material eluted from the P region of Fig. 4. It is unaffected by the inclusion of carrier a or /3 tubulin from porcine brain or from Physarum; it seems to be an intrinsic and distinctive property of this set of radiola- beled species.

DISCUSSION

A key element in our characterization of Physarum tubulins is the use of the broadly reactive rat monoclonal antibodies of Kilmartin et al. (8). We have extended the observations of Kilmartin and his colleagues in showing interspecies reactivity to porcine brain tubulin and to Physarum myxamoebal tu- bulin. Additionally, we have found that each antibody is selective for the a subunit of purified porcine brain tubulin and for the 50,000-dalton polypeptide purified from Physarum myxamoebae by self-assembly or by vinblastine precipitation. Thus, each antibody fulfius the necessary conditions for a reagent to detect the a tubulin chain. I t is plausible that the monoclonal antibody recognizes a very small portion of the peptide either in porcine or Physarum a tubulin and thus permits positive and specific identification of this subunit even when neither the peptide map nor the SDS-PAGE mobility is conserved. These specific antibodies will provide a useful tool in defining the (Y subunit in a broad range of species; this circumvents the difficulty experienced in identifying a tubulin on the basis of mobility in SDS-PAGE (7, 20).

Despite variation in electrophoretic mobility of the Physa- rum myxamoebal a tubulin, conservation in peptide mapping has been inferred by Little et al. (20) and recently confirmed by Roobol and Gull (21) using chymotrypsin rather than V8 protease as the proteolytic agent.

How are the periodically labeled plasmodial molecules of Physarum related to tubulin? Our previous labeling study showed that there is a strong periodicity in synthesis of polypeptides in the tubulin region (2). However, the charac- terization of plasmodial tubulin is limited by its low level even after vinblastine enrichment. To resolve this difficulty, we compared labeled periodic polypeptides with both myxamoe- bal and porcine tubulin. We have concluded that plasmodial /3 tubulin is similar to each reference type of /3 tubulin in two- dimensional electrophoretic mobility. This plasmodial species gives only a faint fluorographic signal and has not been quantified previously (2). Judged visually, this minor &like species seems to be periodic in its synthesis. This species is also highly enriched after two cycles of co-polymerization with porcine tubulin (Fig. 5).

The set of periodic polypeptides denoted P includes two major and one minor fluorographic species. They fail to cor- respond either to self-assembled myxamoebal tubulin or to porcine tubulin in electrophoretic properties and in peptide map. Yet set P efficiently co-polymerizes with carrier porcine tubulin (2) (Fig. 5). P may be either a plasmodium-specific tubulin or a microtubule-associated protein. Further study of co-polymerization of P with carrier myxamoebal tubulin may yield useful information as to the identity of P.

on 15% acrylamide gels and detected by Coomassie blue staining. The gel conditions used here are different from those of Clayton el al. (71, which may account for the difference in peptide patterns. lane a, Myxamoebal a tubulin; lane b, porcine a tubulin; lane c, porcine /3 tubulin; lane d, myxamoebal p tubulin.

1356 Physarum Tubulins

Normally, the a and p tubulin subunits function as a het- erodimer and would be found in stoichiometrically equivalent amounts. However, the p subunit that we have identified in the set of periodically synthesized polypeptides is apparently less abundant than the identified a subunit. Alternatively, this subunit is methionine-poor. Further work, including resolu- tion of the nature of P, will be required to understand the significance of this apparent lack of equivalence.

Acknowledgments-We thank Dr. J. V. Kilmartin for his generous sharing of information and monoclonal antibodies in advance of publication, Dr. R. Scheele for continued gifts of purified porcine brain tubulin, Dr. Robert Nowinski for demonstrating the salient methods for radioimmune assays, and Drs. G. Borisy, R. R. Burgess, T. G. Burland, and K. Gull for their comments.

REFERENCES

1. Dove, W. F., and Rusch, H. P. (eds) (1980) Growth and Differ- entiation in Physarum polycephalum, Princeton University Press, Princeton, NJ

2. Laffler, T. G., Chang, M. T., and Dove, W. F. (1981) Proc. Natl. Acad. Sci. U. S. A . 78, 5000-5004

3. Adelman, M. R., Borisy, G. G., Shelanski, M. L., Weisenberg, R. C., and Taylor, E. W. (1968) Fed. Proc. 27, 1186-1193

4. Jacobson, D. N., Johnke, R. M., and Adelman, M. R. (1976) Cold Spring Harbor Conf. Cell Proliferation 3, 749-770

5. Roobol, A., Pogson, C. I., and Gull, K. (1980) Exp. Cell Res. 130,

6.

7.

8.

9.

10.

11. 12. 13. 14.

15.

16.

17.

18.

19.

20.

21.

203-215 Roobol, A., Pogson, C. I., and Gull, K. (1980) Biochem. J. 189,

Clayton, L., Quinlan, R. A., Roobol, A., Pogson, C. I., and Gull, K.

Kilmartin, J . V., Wright, B., and Milstein, C. (1982) J. Cell Biol.

McCullough, C. H. R., Dee, J., and Foxon, J . L. (1978) J. Gen.

McCullough, C. H. R., and Dee, J. (1976) J. Gen. Microbiol. 95,

Laemmli, U. K. (1970) Nature (Lond.) 227,680-685 Ames, G . F., and Nikaido, K. (1976) Biochemistry 15,616-623 O’Farrell, P. H. (1975) J. Biol. Chem. 250,4007-4021 Bowen, B., Steinberg, J., Laemmli, U. K., and Weintraub, H.

Nowinski, R. C., Lostrom, M. E., Tam, M. R., Stone, M. R., and

Kemphues, K. J., Raff, R. A,, Kaufman, T. C., and Raff, E. C.

Borisy, G. G., Marcum, J . M., Olmsted, J. B., Murphy, D. B., and

Roobol, A., Pogson, C. I., and Gull, K. (1980) Exp. Cell Res. 130,

Cleveland, D. W., Fischer, S. G., Kirschner, M. W., and Laemmli,

Little, M., Ludueiia, R. F., Langford, G. M., Asnes, C. F., and

Roobol, A,, and Gull, K. (1982) in Microtubules in Microorga-

305-312

(1980) FEBS Lett. 115,301-305

93,576-582

Microbiol. 106,297-306

151-158

(1980) Nucleic Acids Res. 8, 1-20

Burnette, W. N. (1979) Virology 93, 111-126

(1979) Proc. Natl. Acad. Sci. U. S. A . 76, 3991-3995

Johnson, K. A. (1975) Ann. N. Y. Acad. Sci. 253, 107-132

203-215

U. K. (1977) J. Biol. Chem. 252, 1102-1106

Farrell, K. (1981) J. Mol. Biol. 149, 95-107

nisms (Morris, N. R., ed) Marcel Dekker, New York