Comp. Biochem. Physiol. Vol. 75B, No. 4, pp. 625-634, 1983 0305-0491/83 $3.00+0.00 Printed in Great Britain © 1983 Pergamon Press Ltd

CHARACTERIZATION OF MICROTUBULES ISOLATED FROM DOGFISH (SQUALUS ACANTHIAS A N D

SCYLIORHINUS CANICULA) BRAIN IN THE ABSENCE OF GLYCEROL

MARGARETA WALLIN and ANN-CATHRINE JONSSON University of G6teborg, Department of Zoophysiology, Box 250 59, S-400 31 G6teborg, Sweden (Tel:

031-853674)

(Received 23 December 1982)

Abstract---1. Microtubules were isolated by two cycles of assembly~tisassembly from dogfish brain in the absence of glycerol.

2. The microtubule protein preparation consist mainly of tubulin as characterized by SDS- polyacrylamide gel electrophoresis, but tau and high molecular weight microtubule associated proteins (MAPs) were also identified, in addition to several other MAPs.

3. The microtubules were cold-sensitive and had an assembly temperature optimum at 21-25°C. 4. Ca 2÷, at a concentration of 1 mM or higher, induced spirals of several protofilaments, sheets and

"macrotubules". 5. In the presence of colchicine (0.1-1.0 mM) spirals as well as microtubules were seen. These structures

were often found to be clustered.

INTRODUCTION

Microtubules isolated from mammalian brains have a temperature opt imum of 37°C and are disassembled when the temperature is lowered under 20°C (John- son and Borisy, 1979). Little attention has been focused on microtubules from non-mammalian verte- brates, although several of these animals live at low temperatures. However, recently microtubules have been isolated from dogfish brain (Langford, 1978) and a temperature opt imum of assembly was found around 21°C.

Microtubules isolated from bovine brain contain 75-85~ of tubulin and 15-25~ of microtubule- associated proteins (MAPs) (for a review see Scheele and Borisy, 1979). MAPs are known to stimulate assembly of tubulin, as tubulin itself is unable to assemble except in the presence of very high concen- trations of Mg 2+ and glycerol or in the presence of DMSO. Langford (1978) was however able to isolate assembly-competent microtubules from dogfish brain, microtubules which almost lacked MAPs, but only in the presence of glycerol during the first purification steps. Only trace amounts of two high- molecular weight MAPs, MAP~ and MAP> could be seen in a few preparations. Mammal ian brain glyc- erol microtubule protein preparations usually contain less MAPs, but MAPs are not excluded from the preparations. It was therefore of interest to study if isolated dogfish brain microtubules lacks MAPs and has assembly competent tubulin, or if the MAPs are excluded from the preparation. We used another protocol for purification of microtubules (in the absence of glycerol) to determine if the purification method was of importance.

In the present study we show that it is possible to isolate microtubules from dogfish brain without

glycerol and that the microtubule protein consist of both tubulin and MAPs. Several differences between these microtubules and microtubules isolated from bovine brain were found.

MATERIALS AND METHODS

Microtubule proteins

Microtubules were isolated from dogfish (Squalus acan- thias and Scyliorhinus canicula) and bovine brains by two cycles of assembly-disassembly in the absence of glycerol and presence of 0.5 mM MgSO4, according to Borisy et al. (1974), slightly modified by Larsson et al. (1976). The dogfish brains were removed immediately after decapitation and put on ice. All assembly steps were performed at 21-25°C, unless stated, for the dogfish brain microtubule preparation. The resulting pellets were frozen at -80°C. Prior to use, the pellet was re-suspended in assembly buffer (100mM Pipes, 0.5mM MgSO 4 and 1.0mM GTP at pH 6.8). After incubation at 4°C for 30min the sample was centrifuged at 35,000 g (ra~ , 7.1 cm) for 30 min at 4°C and the resulting supernatant used for experiments. In contrast to bovine brain, intact dogfish brains could be frozen and thereafter used for purification of microtubules. The mate- rial could also be frozen during the different purification steps.

Bovine brain tubulin was separated from microtubule- associated proteins by ion-exchange chromatography on a column containing from top to bottom phosphocellulose (Whatman PI 1) and Sephadex G-25 Fine (Pharmacia) in 20 mM Pipes and 0.5 mM MgSO 4 at pH 6.8, essentially as described by Weingarten et al. (1975). The void volume consisted of tubulin, and MAPs were eluted with 0.6 M NaC1 in the same buffer.

Assembly

Assembly of microtubule proteins was started by in- creasing the temperature from 4°C to the desired tem- perature in a series of 9, 13, 17, 21, 25, 29, 33 and 37°C in crude extracts as well as in purified samples. A temperature

625

626 MARGARETA WALLIN and ANN-CATHRINE JONSSON

optimum of 21-25C was found and used for further studies. Assembly was monitored continuously by the change of absorbance at 350 nm. All drugs were added from stock solutions.

Protein concentration

Protein concentration was determined according to Lowry et al. (1951) using bovine serum albumin as a standard.

SDS-polyacrvlamide gel electrophoresis The microtubule proteins were separated and analyzed by

SDS-polyacrylamide gel electrophoresis in a 1.5 mm thick vertical slab gel (LKB, Sweden) using a linear gradient (5-15%,) of polyacrylamide (O'Farrell, 1975). The proteins were stained in 0.25% Coomassie Brilliant Blue in meth- anol:acetic acid:water (5:1:5) and destained in 7'~/~, acetic acid and 5g.,; methanol.

Electron microscopy Negatively stained specimens for electron microscopy

were prepared from 5/~1 of microtubule protein samples. Fixation was performed with one drop of Karnovsky solu- tion (Karnovsky, 1965). The specimen was washed with distilled water and stained with 1°/o uranyl acetate.

Dogfish brain and bovine brain microtubule samples were also centrifuged for 30 min at 25 and 3TC, respectively. The pellets were fixed in Karnovsky solution, followed by l'~ osmium-tetroxide in 0.1 M cacodylate buffer. The pellets were dehydrated and embedded in Epon. Thin sections were made on a LKB Ultramicrotome, double-stained with ura- nylacetate and lead citrate, and were viewed in a Zeiss 109 electron microscope.

Map

Map

q-G

/ i ¸ ! / i ¸

TubuLin

RESULTS

Temperature optimum

Little assembly was seen at temperatures lower than 17"C. At 17"C microtubules assembled, but at a very slow rate and to a lower final level that at higher temperatures. At 21 and 2 5 C a rapid assem- bly reaching a steady state was found. Assembled microtubules were disassembled by cold (4~C). At temperatures higher than 29~C, and sometimes even at 29"~'C, the microtubule proteins probably partially denatured, as the absorbance slowly but continuously increased and did not reach a stable level (results not shown).

Microtubule proteins

As judged from SDS-polyacrylamide gel electro- phoresis tubulin was the dominating protein (Fig. 1). MAPs were also present, both tau and the high molecular weight proteins were identified, as well as several other proteins of molecular weight lower and higher than tubulin. A comparison of dogfish brain microtubule proteins (right lane) and bovine brain MAPs (left lane) is seen in Fig. 1. M A P 2 seem to correlate well with bovine brain MAP2, but MAPt to consist of one band with a higher molecular weight and one band with a lower molecular weight than bovine brain MAP~.

Mierotubule morphology

Bovine brain microtubules have arm-like projections extending from their surface, which clearly can be visualized on electron micrographs of embedded specimens (Fig. 2a). In contrast, micro- tubules assembled from pure tubulin in the presence

Fig. I. SDS-polyacrylamide gel electrophoresis of micro- tubule proteins from dogfish and bovine brains. Micro- tubule proteins were isolated by two cycles of assembly-disassembly from dogfish brain and were run on SDS-polyacrylamide gel electrophoresis (linear gradient of 5 15°i,) (right lane). As a comparison a bovine brain phosphocellulose chromatography 0.6 M NaCI MAPs sam-

ple was run simultaneously (left lane).

of 8{~J 0 D M S O do not have such arms (Fig. 2b). Electron micrographs of dogfish brain microtubules, which consist of tubulin and MAPs, does not reveal any clear arm-like projections, but the surface does not seem to be as smooth as the bovine brain microtubules assembled from pure tubulin (Figs 2c and d).

Elects of Ca 2~

Assembly was not inhibited at Ca 2 + concentrations of l mM, which inhibits assembly of bovine brain microtubules (not shown). However, the morphology of the microtubules was altered as spirals and sheets of several protofilaments were seen in addition to normal microtubules (Fig. 3a). At 2 mM Ca 2+, a higher ratio of spirals to normal microtubules were found (Fig. 3b) and at 4 mM Ca 24 the spirals were dominating, but "macromolecules" were also present (Fig. 3c).

Effects q/" colchieine

Colchicine at 0.1 mM decreased the rate as well as the final level of assembly. Electron micrographs revealed the presence of microtubules, which often were clustered together in addition to spirals of several

Dogfish brain microtubules 627

Fig. 2(a-b).

628 MARGARETA WALLIN and ANN-CATHRINE JI3NSSON

Fig. 2(c~1).

Fig. 2. Morphology of dogfish and bovine brain microtubules. Bovine brain microtubules were assembled from microtubule proteins or pure tubulin (in the presence of 8~o DMSO) at 37°C and centrifuged at 35,000g at 37°C for 30 min. Dogfish brain microtubule proteins were assembled at 25°C and centrifuged at 35,000g at 25c'C for 30 min. The pellets were thereafter fixed in Karnovsky solution, followed by 1~ osmiumtetroxide in 0.1 M cacodylate buffer. The pellets were dehydrated, embedded in Epon, thin sectioned and double-stained. (a) Bovine brain microtubules assembled from microtubule proteins, (b) bovine brain microtubules assembled from pure tubulin, and dogfish brain microtubules assembled from

microtubule proteins in (c) longitudinal section and (d) cross-section.

Dogfish brain microtubules 629

Fig. 3(a-b).

630 MARGARETA WALLIN and ANN-CATHRINE JONSSON

Fig. 3(c).

Fig. 3. Effects of Ca ~+ on dogfish brain microtubules. Microtubules were assembled at 21-25°C in the presence of (a) 1 mM Ca 2+, (b) 2 mM Ca 2+ and (c) 4 mM Ca 2÷. Negatively stained specimens were fixed

with Karnovsky solution and stained with 1~o uranyl acetate.

Dogfish brain microtubules 631

Fig. 4(a).

632 MARGARETA WALL1N and ANN-CATHRINE J()NSSON

Fig. 4(~c).

Fig. 4. Effects of colchicine on dogfish brain microtubule assembly. Microtubules were assembled at 21-25°C in the presence of (a) 0.1 mM colchicine, (b) and (c) 1.0mM colchicine. Negatively stained

specimens were prepared as described in the legend to Fig. 3.

Dogfish brain

protofilaments (Fig. 4a). At a higher colchicine con- centration (1 mM), the absorbance increased con- tinuously and did not reach a stable steady state in 60 min. In these samples spirals were dominating (Fig. 4b) which often were clustered (Fig. 4c).

DISCUSSION

In the present study we show that microtubules can be isolated from dogfish brain in the absence of glycerol. At temperatures lower than 17°C, no or very few microtubules assembled, although the dogfish can live at such low temperatures as 2°C. The micro- tubules were found to have an assembly temperature optimum of 21-25°C, a finding in agreement with Langford (1978). At temperatures higher than 29°C (and sometimes even at 29°C) a real steady state was never achieved, probably because of a partial dena- turation.

In contrast to microtubules isolated from dogfish brains in the presence of glycerol by Langford (1978), the present microtubule protein preparation con- tained both tubulin and MAPs. Glycerol is known to decrease the ratio of MAPs to tubulin in mammalian preparations, but both tau and the high molecular weight proteins are usually present, in addition to several other proteins (for a review see Scheele and Borisy, 1979). The reason for this discrepance be- tween the preparation of Langford and ours is un- known. However, Langford found small amounts of the high molecular weight proteins MAP, and MAP2 in some preparations. He suggested that they could be degraded during the purification process, since they can be seen in the supernatant and pellet frac- tions derived from the early stages of purification.

Of the high molecular weight MAPs found in mammalian brain microtubule protein preparations, MAP2 is bound in a regular helical pattern around microtubules as extending arm-like projections (Amos, 1979). Although high molecular weight MAPs are present in the dogfish brain microtubule protein preparation, we did not find any clear evi- dence for such arm-like structures. However, the surface was not as smooth as the surface of mam- malian brain microtubules assembled from phos- phocellulose purified tubulin. Whether the dogfish brain microtubule arms are less stable than bovine brain microtubule arms or whether these micro- tubules do not possess arm-like structures is under investigation.

Dogfish brain microtubules were cold-sensitive, but was affected very differently from mammalian brain microtubules by Ca 2÷ and colchicine in the pH range used for preparations of microtubules. Ca 2÷ induced spirals of several protofilaments, sheets and "macromolecules" as well as microtubules, when added initially in concentrations of 1-4 mM. Similar structures were found by Langford (1978) on addi- tion of 10 mM Ca 2÷ at steady state. The presence of MAPs therefore does not seem to alter this property. In a microtubule protein preparation from bovine brain prepared according to Kuriyama (1975), Ca 2÷ inhibited assembly at a pH over 6.4. However, in a lower pH range (5.8-6.4) twisted ribbons and "mac- rotubules", which show similarities with Ca2÷-induced structures in dogfish brain microtubule

c.sP. 75/4a--~-

microtubules 633

protein preparations was found (Matsumura and Hayashi, 1976). These microtubules also contained MAPs.

In the presence of colchicine at concentrations that inhibit mammalian brain microtubule assembly (0.1-1.0mM), an assembly of dogfish brain micro- tubule proteins were found. However, in addition to normal microtubules, spirals of several protofilaments and clustered microtubules were seen. Another altered morphological structure of micro- tubules, thin filaments, have recently been found assembled from pure mammalian tubulin in the pres- ence of colchicine, but no spirals were seen (Saltarelli and Pantaloni, 1982). Microtubule proteins isolated from carp brain (Maccioni and Mellado, 1981) binds colchicine at 4°C, another difference between mam- malian and non-mammalian animals.

The different characteristics between microtubules isolated from mammalians and fish are very inter- esting. It is known that the proteolytic cleavage pattern of tubulin from dogfish brain and mam- malian brain differs very little (Little et al., 1981). Can it be this small difference, the composition or ratio of MAPs to tubulin or perhaps post-translational events that can change the properties of microtubules from different species? We are currently investigating these matters.

Acknowledgements--The present work was supported by grants from Naturvetenskapliga Forskningsr~.det (Nos 2535-112 and 3635-106), Anna Ahrenbergs fond, Hierta Retzius' Stiftelse, Stiftelsen Lars Hiertas Minne, Magnus Bergvalls Stiftelse, Wilhelm och Martina Lundgrens Vet- enskapsfond and Kungliga Hvitfeldska Overskottsfonden. We are indepted to Dr Stefan Nilson for critical reading of the manuscript, to Mrs Inger Holmqvist for helpful assis- tance and to the Marine Laboratory, Citadel Hill, Plymouth, England, where part of the work was performed. We are also grateful to the Staff of the Kristineberg Marine Biological Station, Fiskebfickskil, Sweden, for providing us with dogfish.

REFERENCES

Amos L. A. (1979) Structure of microtubules. Microtules (Edited by Roberts K. and Hyams J. S.), pp. 1-64. Academic Press, New York.

Johnson K. A. and Borisy G. G. (1979) Thermodynamic analysis of microtubule self-assembly in vitro. J. molec. Biol. 133, 199-216.

Karnovsky N. J. (1965) A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. J. Cell Biol. 27, 137-138.

Kuriyama R. (1975) Further studies on tubulin poly- merization in vitro. J. Biochem. 77, 23-31.

Langford G. M. (1978) In vitro assembly of dogfish brain tubulin and the induction of coiled ribbon polymers by calcium. Expl. Cell Res. 111, 139-151.

Larsson H., Wallin M. and Edstr6m A. (1976) Induction of a sheet polymer of tubulin by Zn 2÷ . Expl. Cell Res. 100, 104-110.

Little M., Luduena R. F., Langford G. M., Asnes C. F. and Farrell K. (1981) Comparison of proteolytic cleavage patterns of ct-tubulin from taxonomically distant species. J. molec. Biol. 149, 95-107.

Lowry O. H., Rosebrough N. J., Farr A. L. and Randall R. J. (1951) Protein measurements with Folin phenol re- agent. J. bioL Chem. 193, 265-275.

Maccioni R. B. and Mellado W. (1981) Characteristics of the in vitro assembly of brain tubulin of Cyprinus carpio. Comp. Biochem. Physiol. 70B, 375-380.

634 MARGARETA WALLIN and ANN-CATHRINE J6NSSON

Matsumura F. and Hayashi M. (1976) Polymorphism of tubulin assembly. In vitro formation of sheet, twisted ribbon and microtubule. Biochim. biophys. Acta 453, 162-175.

O'Farrell P. H. (1975) High-resolution two-dimensional electrophoresis of proteins. J. biol. Chem. 250, 40074021.

Saltarelli D. and Pantaloni D. (1982) Polymerization of the tubulin-colchicine complex and guanosine 5'-triphosphate. Biochemistry 21, 2996-3006.

Scheele R. B. and Borisy G. G. (1979) In vitro assembly of microtubles. Microtubules (Edited by Roberts K. and Hyams J. S.), pp. 175-254. Academic Press, New York.