- Email: [email protected]
Evidence that S100 proteins regulate microtubule assembly and stability in rat brain extracts
Int. J. Biochem. Vol. 18, No. 8, pp. 691qi95, 1986 Printed in Great Britain
0020-711X/86 $3.00+ 0.00 Pergamon Journals Ltd
EVIDENCE THAT S100 PROTEINS REGULATE MICROTUBULE ASSEMBLY A N D STABILITY IN RAT BRAIN EXTRACTS JOHN HESKETHl* and JACQUES BAUDIER2 tRowett Research Institute, Bucksburn, Aberdeen AB2 9SB, U.K. 2Laboratoire de Physique, UA CNRS 491, UER des Sciences Pharmacetiques, B.P. No 10, 67048 Strasbourg, France (Received 25 November 1985)
Abstract--l. Microtubule re-assembly in rat brain extracts was inhibited by antibodies to SI00 proteins. 2. Anti-S100 antibodies caused an increase in the cold-stability of microtubules and this effect was abolished by the presence of short lengths of microtubules formed under control conditions. 3. Anti-S100 antibodies had no effect on the stimulation of assembly or the increase in microtubule stability caused by low zinc concentrations. 4. Addition of exogenous Sl00a and S100b to brain extracts had different effects on assembly; S100a caused an inhibition of assembly while S100b stimulated the early phase of assembly. 5. The data suggest that endogenous Sl00b is involved in the regulation of microtubule assembly in brain extracts. INTRODUCTION Microtubules comprise mostly tubulin together with smaller amounts of specific microtubule-associated proteins (MAPs) such as tau, MAP 1 and MAP 2. The regulation of microtubule assembly in the cell is poorly understood. Tubulin, purified or in tissue extracts, can polymerize to form microtubules and this assembly in vitro can be modified by a variety of factors (Timasheff and Grisham, 1980; Hesketh, 1985). However, many of these effects occur under non-physiological conditions and others are difficult to extrapolate to the situation in vivo. The association of tau and MAP 2 with microtubule arrays in situ, together with their ability to modify assembly in vitro, suggest however, that these proteins at least may be physiological regulators of assembly. It is also probable that divalent cations are important; calcium has been shown to induce microtubule disassembly and to inhibit assembly (Weisenberg, 1972) and small alterations in the zinc concentration of rat brain extracts affect the assembly and stability of microtubules (Hesketh, 1981, 1984). The effects of calcium appear to be mediated by calcium-binding proteins such as calmodulin (Klee and Vanaman, 1982) and S100 (Baudier et al., 1982a; Donato, 1983). Calcium and calmodulin modify the phosphorylation of tau (Kakiuchi and Sobue, 1981) and have also been implicated in the control of microtubule cold-stability and so in the more general processes of controlling microtubule length and orientation by stabilization mechanisms (Margolis and Rauch, 1981; Job et al., 1983). SI00 protein, which represents up to 0.2% of total soluble brain protein, is a mixture of two acidic dimeric proteins, the S100a and 100b proteins (Isobe
et al., 1977). S100a and Sl00b have ~//and ~ subunit composition respectively (Isobe et al., 1977); they are present in approximately equal amounts in bovine brain but in tissue of rat origin S100b predominates (Isobe et al., 1983; Baudier et al., 1985). Both S100a and S100b interact specifically with calcium ions (Baudier and Gerard, 1983; Mani and Kay, 1983; Baudier et al., 1985); however, S100b has been shown to bind zinc more strongly than calcium (Baudier and Gerard, 1983; Baudier et al., 1985) and it therefore may also function as a zinc binding protein. The biological function of the proteins is unknown but since S100 has been shown to mediate effects of calcium and zinc on assembly of purified microtubule protein (Baudier et al., 1982a; Deinum et al., 1983; Donato, 1983), it is possible that S100 proteins function in vivo as regulators of microtubule assembly. However, there is at present no evidence for an effect of S100 on microtubule assembly in vivo. The aim of the present work was to study the effects of antibodies raised against S100 proteins in order to assess the role of endogenous S100 protein in the assembly of brain microtubules. Low concentrations of zinc stimulate microtubule assembly in brain extracts and it was also possible to investigate if SI00 proteins are involved in this effect.
*Author to whom correspondence and reprint requests should be addressed. 691
MATERIALS AND M E T H O D S
Preparation of proteins and antiserum Purified S100 protein, Sl00a and Sl00b were prepared from bovine brain as described previously (Baudier et al., 1982b, 1983) and subsequently lyophilised. Throughout purification, 2mM 2-mercaptoethanol was present to maintain sulphydryl groups in the reduced state. Calmodulin was prepared from bovine brain as described by Isobe et al. (1977). Prior to the experiment the proteins were resuspended in 100mM Pipes (Piperazine-N-N-bis[2ethane]sulphonic acid) buffer, pH 6.95. Antibodies were raised in rabbits by repeated injections of bovine SI00 protein (mixture of Sl00a and b) complexed to methylated
jiCon*r°
692
JOHN HESKETH a n d JACQUES BAUDIER
O. 10
i+non_immun e
serum)
E 0 tO ¢0
+Anti-SlO0 o
•
c 0 . 0 .~ t~
O
I
5
10
20 Time
!
|
30
45
(min)
Fig. 1. Effect of anti-SlO0 antiserum on microtubule assembly in rat brain extracts. Brain extracts were
pre-incubated with either anti-Sl00 antiserum ( o ) or non-immune serum ( o ) and then assembly was initiated by adding I mM GTP and raising the temperature to 37°C. Subsequent assembly was monitored by the increase in absorbance at 350 nm. Sera were diluted 1--.50.
bovine serum albumin as described in Labourdette and Marks (1975). The antiserum obtained reacted against both S100a and S100b proteins of both rat and bovine origin (Ghandour et al., 1981).
Microtubule assembly Rat brain was homogenized in assembly buffer (I00 mM Pipes, 1 mM EGTA, pH 6.95) and maintained at 4°C for 30min. After centrifugation at 100,000g for 30 min, the supernatant was divided into aliquots for assessment of microtubule assembly. Antisera, used at a dilution of 1---,50, were pre-incubated with the brain extract for 60 min at 4°C prior to initiation of the assembly reaction; purified S100a and S100b were pre-incubated for 5 min. Assembly was started by the addition of guanosine-5-triphosphate (GTP) to a final concentration of I mM and by raising the temperature to 37°C. The extent of assembly was followed turbidometrically by measuring the optical density at 350 nM and the results expressed as the increase in optical density from the beginning of the incubation. After incubation at 37°C for 45 min the microtubules formed were then exposed to cold (4°C) for 30 min and the remaining optical density increase measured. Cold-stability was taken as the ratio of the optical density after cold exposure to that at the end of the assembly reaction at 37°C (Hesketh, 1984). Stimulation of assembly by zinc was carried out using 500#M ZnCI2 in the presence of 1 mM EGTA, giving an approximate free zinc concentration of 1 #M (Hesketh, 1984). Microtubule "seeds" were prepared by shearing (3 passages through a gauge 25 needle) assembled microtubules from normal rat brain extracts. RESULTS
Effects of antibodies Previous work has shown that microtubule assembly in rat brain extracts is optimal in the absence of exogenous GTP (Margolis and Rauch, 1981; Hesketh, 1984). However, in the present experiments the long pre-incubation period led to assembly being dependant on exogenous GTP and therefore all assembly reactions were carried out in the presence of 1 mM GTP. Under these conditions anti-S100 antibodies caused a marked inhibition of microtubule
assembly (Fig. 1); this was characterized by a longer lag phase at the beginning of the incubation, a small decrease in the rate of turbidity change and a reduction in the turbidity attained at the end of the incubation. The major effect was on the lag phase. As found previously low zinc concentrations caused a stimulation of microtubule assembly (Hesketh, 1984) and in addition this stimulation occurred in the presence of anti-Sl00 antibodies (Fig. 2). Anti-S100 antibodies inhibited normal and zincstimulated assembly to approximately the same extent, as judged by the reduction in the turbidity increases after 45 min incubation; again the effects were largely on the lag phase. Pre-incubation with anti-S100 antibodies led to the formation of a higher proportion of microtubules which were stable to cold (Table 1). This increase in the cold stability of the microtubules formed during the assembly reaction was additive with the increase in cold-stability due to zinc and was abolished by the presence of short lengths of microtubules which were used to "seed" the reaction. The increase in coldstability due to zinc-stimulation of assembly (Hesketh, 1984) was not reduced by the anti-Sl00 antibodies.
Effects of purified SlOOa and SlOOb As a complement to the antibody experiments, the effects of purified Sl00a and S100b on microtubule assembly in rat brain extracts was investigated. The two proteins had different effects (Fig. 3). S 100a caused an inhibition of assembly as measured by the final turbidity attained at the end of the incubation but had no effect on the early phase of assembly. On the other hand, S100b had a marked stimulatory effect on the early part of the assembly reaction such that the turbidity increased faster over the first 10-15 min of the reaction; thereafter the rate of assembly slowed and Sl00b had no consistent effect on the final turbidity attained at the end of the reaction. Calmodulin had no effect on assembly kinetics and neither
Sl00 and microtubule assembly
693
0.15 t.o
tO
•
Control
0.1
J
--
0¢ -
b. 0 if)
e-
~
~
m
--
+Zn
+Zn+anti-S iO0 j
• +anti-S I00
0.05
h. O e--
0
Ii 5
I
A
t
I
I
I
I
I
10
15
20
25
30
35
40
45
Time ( r a i n )
Fig. 2. Effect of anti-Sl00 antiserum on the stimulation of brain extract microtubule assembly by zinc. Assembly was studied in the presence of 500 #M added zinc (estimated free zinc, 1 #M) and anti-Sl00 antiserum (--~k--), in the presence of zinc and non-immune serum (--~) and without added zinc either in the presence of non-immune serum ( I ) or anti-Sl00 antiserum (e). Sera were diluted 1-+50.
did a equimolar mixture of Sl00a and b (results not shown). S100b had a small effect on the proportion cold-stable microtubules, reducing the percentage cold-stability from 64 to 59% but this was not statistically significant (Table 2). S100a had no effect on the microtubule cold-stability. DISCUSSION
The experiments using anti-SlO0 antibodies demonstrate that endogenous SIO0 in rat brain extracts can influence assembly of microtubules. This [extends the earlier findings that in the presence of calcium or zinc S100 can modulate the assembly of purified tubulin or microtubule protein (Baudier et al., 1982a; Deinum et al., 1983). The effectiveness of endogenous S100 in regulating assembly in such extracts suggest that S100 proteins may contribute to the regulation of brain microtubule assembly in vivo. Since S100 protein in rat brain is very largely (90%) or wholly S100b (Isobe et al., 1983; Baudier et al., 1985) it is probable that it is endogenous S100b which is the effective component. This view is supported by the observation that in the present experiments antiS100 antibodies and exogenous Sl00b had opposing effects on assembly kinetics and cold-stability; for example, Sl00b stimulated the early phase of assembly whereas the anti-S100 antibodies lengthened the lag phase. Changes in the lag phase of assembly are thought to reflect changes in the nucleation of the assembly reaction, that is the formation of some structure upon which microtubule elongation can occur. Thus the effect of anti-S100 antibodies is compatible with the data from experiments on purified microtubule protein (Donato, 1984) which have shown that S100 modulates the nucleation phase of assembly. Furthermore the effect of antiB.C. 18/S--B
SI00 antibodies was very largely on the lag phase rather than the rate of subsequent assembly and this suggests that the rate of nucleation was inhibited more than its extent. In addition to the effects on the kinetics of assembly S100b also changed the properties of the microtubules formed during incubation in vitro; inhibition of endogenous S100 with antibodies increased the microtubule cold-stability while addition of exogenous S100b caused a small, but statistically not significant, decrease. Stability to cold per se may not be physiologically significant but it may represent a more general form of stability which is involved in microtubule regulation (Job et al., 1983; Hesketh, 1985). Cold-stability has been found previously to be regulated by calcium and calmodulin (Margolis and Rauch, 1981; Job et al., 1983) and has been suggested to be due to the presence of specific proteins spaced periodically along the microtubule. The present results show that another cation-binding protein, Sl00b, can also modulate such stability and that this property is not unique to calmodulin. Zinc has also
Table 1. Effect of anti-SI00 antibodies on microtubule cold stability Incubation conditions Control ( + n o n - i m m u n e serum) + Anti-Sl00 (1 --~50 dilution) + Z n (500/aM) + A n t i - S l 0 0 + Zn + "'Seeds" + Anti-Sl00 + seeds Values given Students group; cp cp < 0.05
% Cold stability 63 _+ 3 (n = 4) 74 _+ 3 (n 76 + 2 (n 88 + 4 (n 60 + 5 (n 65 + 1 (n
= = = = =
4) a 4) b 4) c 4) 4) d
are means 5:SEM. G r o u p s were c o m p a r e d using t-test. , p <0.05; bp <0.01 c o m p a r e d to control < 0.05 compared to Zn only or anti-Sl00 only group; compared to anti-S100 group.
694
JOHN HESKETH and JACQUES BAUDIER
0.15 ~ a
-b
0.1
--
E t-
LIO O
ID
C°nt~+Sii0 b
e.,Q
s/
0 ffl
/
I
+//S~O0 a
e/
t~
/ . ='/
e-
0.05
_
//
,/~ I / 0
10
/,. 20
30
40
50
Time
I
I
I
i
J
10
20
30
40
50
(min)
Fig. 3. Effects of purified SI00 proteins on microtubule assembly in brain extracts. (a) Effect of 10/tM Sl00a. (b) Effect of 10 #M S100b. Both proteins were pre-incubated with the extracts for 5 min prior to initiation of assembly. Controls were run in the absence of any added proteins but with an equal volume of Pipes buffer added.
been found to increase microtubule stability in rat brain extracts (Hesketh, 1984) but this effect appears from the present data to be independant of the S100-induced increase. The finding that the addition of short microtubule fragments from control preparations abolishes the effect of anti-S100 antibodies on cold-stability suggests that cold-stability is controlled by the properties of an initial short length of microtubule on which subsequent polymerization occurs. The experiments with exogenous S100 proteins showed that the effects of the two SI00 components on the microtubule system were different; Sl00b stimulated the early part of the assembly reactions while S100a inhibited assembly. Different effects of the two S 100 components on the assembly of purified microtubule proteins have been observed previously (Deinum et al., 1983) although while this work was in progress further experiments with purified microtubule protein have shown the effects of Sl00a and b on microtubule assembly to be similar (Donato et al., 1985). This discrepancy cannot be explained at present. The disparate effects of the two proteins emphasises the need to carry out experiments with purified S100a and S100b rather than total unfractionated SI00 proteins. Small changes in the free zinc concentration in rat brain extracts can modulate microtubule assembly and stability (Hesketh, 1981, 1984) and one possible mechanism for such effects is mediation by SI00 Table 2. Effects of S100a and Sl00b on microtubule cold stability Incubation conditions
% Cold stability
Control + 10#M Sl00a + 10/tM Sl00b
64 + 4 (n = 4) 62 + 3 (n = 3) 59-f- 2(n = 3 )
Values given are means + SEM.
proteins. However, the effects of zinc were unaffected by anti-S100 antibodies, thus suggesting that zinc does not act by this mechanistri. Experiments with microtubule protein (Deinum et al., 1983) have suggested that Sl00a can chelate zinc and so prevent high zinc concentrations from inducing microtubule sheet formation. Such a chelation of zinc in brain extracts would explain the inhibition of assembly by Sl00a and would support the view (Hesketh, 1981, 1985) that the free zinc concentration in brain extracts is a critical determinant of microtubule assembly. It remains to be investigated if such a chelating effect of S100a is of physiological significance. In conclusion, the study of assembly in brain extracts has allowed demonstration that endogenous Sl00b regulates microtubule assembly and stability. However, the experimental system did not allow elucidation of the detailed mechanism and investigation of the complex relationship between calcium, zinc, S100 proteins and microtubules will require further reconstitution experiments with purified proteins. REFERENCES
Baudier J. and Gerard D. (1983) Ion binding to S100 proteins: structural changes induced by calcium and zinc on S100a and S100b proteins. Biochemistry 22, 3360-3369. Baudier J., Briving C., Deinum J., Haglid K., Sorskog L. and Wallin M. (1982a) Effect of S100 proteins and calmodulin on Ca 2+ induced disassembly of brain microtubule proteins/n vitro. F E B S Lett. 147, 165-167. Baudier J., Holtzcherer C. and Gerard D. (1982b) Zincdependant affinity chromatography of the S100b protein on phenyl-Sepharose. A rapid purification method. FEBS Lett. 148, 231-234. Baudier J., Mandel P. and Gerard D. (1983b) Bovine brain
S100 and microtubule assembly SI00 proteins: separation and characterization of a new S100 protein species. J. Neurochem. 40, 145-152. Baudier J., Labourdette G. and Gerard D. (1985) Rat brain S100b protein: purification, characterization and ionbinding properties. A comparison with bovine S100b protein. J. Neurochem. 44, 76-84. Deinum J., Baudier J., Briving C., Rosengren L., Wallin M., Gerard D. and Haglid K. (1983) The effect of S-100a and S-100b and Zn ~÷ on the assembly of brain microtubule proteins in vitro. FEBS Lett. 163, 287 291. Donato R. (1983) Effect of S-100 protein on assembly of brain microtubule proteins in vitro. FEBS Lett. 162, 310-313. Donato R. (1984) Mechanism of action of S-100 protein(s) on brain microtubule protein assembly. Biochem. biophys. Res. Commun. 124, 850-856. Donato R., Isobe T. and Okuyama T. (1985) S100 proteins and microtubules: analysis of the effects of rat brain S-100 (S100b) and ox brain Sl00a 0, Sl00a and S100b on microtubule assembly~lisassembly. FEBS Lett. 186, 65-69. Ghandour M. S., Labourdette G., Vincendon G. and Gombos G. (1981) A biochemical and immunohistological study of SI00 protein in developing rat cerebellum. Dev. Neurosci. 4, 98 109. Hesketh J. E. (1981) Impaired microtubule assembly in brain from zinc-deficient pigs and rats. Int. J. Biochem. 13, 921-926. Hesketh J. E. (1984) Microtubule assembly in rat brain extracts. Further characterization of the effects of zinc on assembly and cold-stability. Int. J. Biochem. 16, 1331-1339. Hesketh J. E. (1985) Regulation of microtubule assembly. Int. J. Biochem. 17, 761-766.
695
Isobe T., Nakajima T. and Okuyama T. (1977) Reinvestigation of extremely acidic proteins in bovine brain. Biochem. biophys. Acta 494, 222-232. Isobe T., Ishioka N., Masuda T., Takahashi Y., Ganno S. and Okuyama T. (1983) A rapid separation of S100 subunits by high performance liquid chromatography: the subunit compositions of S100 proteins. Biochem. Int. 6, 419-426. Job D., Rauch C. T., Fischer E. H. and Margolis R. L. (1983) Regulation of microtubule cold-stability by calmodulin-dependant and independant phosphorylation. Proc. natn. Acad. Sci. U.S.A. 80, 3894-3898. Kakiuchi S. and Sobue K. (1981) Ca ++ and calmodulin dependant flip-flop mechanism in microtubule assemblydisassembly. FEBS Lett. 132, 141-143. Klee C. B. and Vanaman T. C. (1982) Calmodulin. Adv. Prot. Chem. 35, 213-221. Labourdette G. and Marks A. (1975) Synthesis of S-100 protein in monolayer cultures of rat glial cells. Eur. J. Biochem. 58, 73-79. Mani R. S. and Kay C. M. (1981) Isolation and spectral studies on the calcium binding properties of bovine brain S-100a protein. Biochemistry 22, 3902-3907. Margolis R. L. and Rauch C. T. (1981) Characterization of rat brain crude extract microtubule assembly: correlation of cold stability with the phosphorylation state of a microtubule-associated 64-K protein. Biochemistry 20, 4451-4458. Timasheff S. N. and Grisham L. (1980) In vitro assembly of cytoplasmic microtubules. A. Rev. Biochem. 49, 565-591. Weisenberg R. C. (1972) Microtubule formation in solutions containing low calcium concentrations. Science 177, 1104-1105.
0020-711X/86 $3.00+ 0.00 Pergamon Journals Ltd
EVIDENCE THAT S100 PROTEINS REGULATE MICROTUBULE ASSEMBLY A N D STABILITY IN RAT BRAIN EXTRACTS JOHN HESKETHl* and JACQUES BAUDIER2 tRowett Research Institute, Bucksburn, Aberdeen AB2 9SB, U.K. 2Laboratoire de Physique, UA CNRS 491, UER des Sciences Pharmacetiques, B.P. No 10, 67048 Strasbourg, France (Received 25 November 1985)
Abstract--l. Microtubule re-assembly in rat brain extracts was inhibited by antibodies to SI00 proteins. 2. Anti-S100 antibodies caused an increase in the cold-stability of microtubules and this effect was abolished by the presence of short lengths of microtubules formed under control conditions. 3. Anti-S100 antibodies had no effect on the stimulation of assembly or the increase in microtubule stability caused by low zinc concentrations. 4. Addition of exogenous Sl00a and S100b to brain extracts had different effects on assembly; S100a caused an inhibition of assembly while S100b stimulated the early phase of assembly. 5. The data suggest that endogenous Sl00b is involved in the regulation of microtubule assembly in brain extracts. INTRODUCTION Microtubules comprise mostly tubulin together with smaller amounts of specific microtubule-associated proteins (MAPs) such as tau, MAP 1 and MAP 2. The regulation of microtubule assembly in the cell is poorly understood. Tubulin, purified or in tissue extracts, can polymerize to form microtubules and this assembly in vitro can be modified by a variety of factors (Timasheff and Grisham, 1980; Hesketh, 1985). However, many of these effects occur under non-physiological conditions and others are difficult to extrapolate to the situation in vivo. The association of tau and MAP 2 with microtubule arrays in situ, together with their ability to modify assembly in vitro, suggest however, that these proteins at least may be physiological regulators of assembly. It is also probable that divalent cations are important; calcium has been shown to induce microtubule disassembly and to inhibit assembly (Weisenberg, 1972) and small alterations in the zinc concentration of rat brain extracts affect the assembly and stability of microtubules (Hesketh, 1981, 1984). The effects of calcium appear to be mediated by calcium-binding proteins such as calmodulin (Klee and Vanaman, 1982) and S100 (Baudier et al., 1982a; Donato, 1983). Calcium and calmodulin modify the phosphorylation of tau (Kakiuchi and Sobue, 1981) and have also been implicated in the control of microtubule cold-stability and so in the more general processes of controlling microtubule length and orientation by stabilization mechanisms (Margolis and Rauch, 1981; Job et al., 1983). SI00 protein, which represents up to 0.2% of total soluble brain protein, is a mixture of two acidic dimeric proteins, the S100a and 100b proteins (Isobe
et al., 1977). S100a and Sl00b have ~//and ~ subunit composition respectively (Isobe et al., 1977); they are present in approximately equal amounts in bovine brain but in tissue of rat origin S100b predominates (Isobe et al., 1983; Baudier et al., 1985). Both S100a and S100b interact specifically with calcium ions (Baudier and Gerard, 1983; Mani and Kay, 1983; Baudier et al., 1985); however, S100b has been shown to bind zinc more strongly than calcium (Baudier and Gerard, 1983; Baudier et al., 1985) and it therefore may also function as a zinc binding protein. The biological function of the proteins is unknown but since S100 has been shown to mediate effects of calcium and zinc on assembly of purified microtubule protein (Baudier et al., 1982a; Deinum et al., 1983; Donato, 1983), it is possible that S100 proteins function in vivo as regulators of microtubule assembly. However, there is at present no evidence for an effect of S100 on microtubule assembly in vivo. The aim of the present work was to study the effects of antibodies raised against S100 proteins in order to assess the role of endogenous S100 protein in the assembly of brain microtubules. Low concentrations of zinc stimulate microtubule assembly in brain extracts and it was also possible to investigate if SI00 proteins are involved in this effect.
*Author to whom correspondence and reprint requests should be addressed. 691
MATERIALS AND M E T H O D S
Preparation of proteins and antiserum Purified S100 protein, Sl00a and Sl00b were prepared from bovine brain as described previously (Baudier et al., 1982b, 1983) and subsequently lyophilised. Throughout purification, 2mM 2-mercaptoethanol was present to maintain sulphydryl groups in the reduced state. Calmodulin was prepared from bovine brain as described by Isobe et al. (1977). Prior to the experiment the proteins were resuspended in 100mM Pipes (Piperazine-N-N-bis[2ethane]sulphonic acid) buffer, pH 6.95. Antibodies were raised in rabbits by repeated injections of bovine SI00 protein (mixture of Sl00a and b) complexed to methylated
jiCon*r°
692
JOHN HESKETH a n d JACQUES BAUDIER
O. 10
i+non_immun e
serum)
E 0 tO ¢0
+Anti-SlO0 o
•
c 0 . 0 .~ t~
O
I
5
10
20 Time
!
|
30
45
(min)
Fig. 1. Effect of anti-SlO0 antiserum on microtubule assembly in rat brain extracts. Brain extracts were
pre-incubated with either anti-Sl00 antiserum ( o ) or non-immune serum ( o ) and then assembly was initiated by adding I mM GTP and raising the temperature to 37°C. Subsequent assembly was monitored by the increase in absorbance at 350 nm. Sera were diluted 1--.50.
bovine serum albumin as described in Labourdette and Marks (1975). The antiserum obtained reacted against both S100a and S100b proteins of both rat and bovine origin (Ghandour et al., 1981).
Microtubule assembly Rat brain was homogenized in assembly buffer (I00 mM Pipes, 1 mM EGTA, pH 6.95) and maintained at 4°C for 30min. After centrifugation at 100,000g for 30 min, the supernatant was divided into aliquots for assessment of microtubule assembly. Antisera, used at a dilution of 1---,50, were pre-incubated with the brain extract for 60 min at 4°C prior to initiation of the assembly reaction; purified S100a and S100b were pre-incubated for 5 min. Assembly was started by the addition of guanosine-5-triphosphate (GTP) to a final concentration of I mM and by raising the temperature to 37°C. The extent of assembly was followed turbidometrically by measuring the optical density at 350 nM and the results expressed as the increase in optical density from the beginning of the incubation. After incubation at 37°C for 45 min the microtubules formed were then exposed to cold (4°C) for 30 min and the remaining optical density increase measured. Cold-stability was taken as the ratio of the optical density after cold exposure to that at the end of the assembly reaction at 37°C (Hesketh, 1984). Stimulation of assembly by zinc was carried out using 500#M ZnCI2 in the presence of 1 mM EGTA, giving an approximate free zinc concentration of 1 #M (Hesketh, 1984). Microtubule "seeds" were prepared by shearing (3 passages through a gauge 25 needle) assembled microtubules from normal rat brain extracts. RESULTS
Effects of antibodies Previous work has shown that microtubule assembly in rat brain extracts is optimal in the absence of exogenous GTP (Margolis and Rauch, 1981; Hesketh, 1984). However, in the present experiments the long pre-incubation period led to assembly being dependant on exogenous GTP and therefore all assembly reactions were carried out in the presence of 1 mM GTP. Under these conditions anti-S100 antibodies caused a marked inhibition of microtubule
assembly (Fig. 1); this was characterized by a longer lag phase at the beginning of the incubation, a small decrease in the rate of turbidity change and a reduction in the turbidity attained at the end of the incubation. The major effect was on the lag phase. As found previously low zinc concentrations caused a stimulation of microtubule assembly (Hesketh, 1984) and in addition this stimulation occurred in the presence of anti-Sl00 antibodies (Fig. 2). Anti-S100 antibodies inhibited normal and zincstimulated assembly to approximately the same extent, as judged by the reduction in the turbidity increases after 45 min incubation; again the effects were largely on the lag phase. Pre-incubation with anti-S100 antibodies led to the formation of a higher proportion of microtubules which were stable to cold (Table 1). This increase in the cold stability of the microtubules formed during the assembly reaction was additive with the increase in cold-stability due to zinc and was abolished by the presence of short lengths of microtubules which were used to "seed" the reaction. The increase in coldstability due to zinc-stimulation of assembly (Hesketh, 1984) was not reduced by the anti-Sl00 antibodies.
Effects of purified SlOOa and SlOOb As a complement to the antibody experiments, the effects of purified Sl00a and S100b on microtubule assembly in rat brain extracts was investigated. The two proteins had different effects (Fig. 3). S 100a caused an inhibition of assembly as measured by the final turbidity attained at the end of the incubation but had no effect on the early phase of assembly. On the other hand, S100b had a marked stimulatory effect on the early part of the assembly reaction such that the turbidity increased faster over the first 10-15 min of the reaction; thereafter the rate of assembly slowed and Sl00b had no consistent effect on the final turbidity attained at the end of the reaction. Calmodulin had no effect on assembly kinetics and neither
Sl00 and microtubule assembly
693
0.15 t.o
tO
•
Control
0.1
J
--
0¢ -
b. 0 if)
e-
~
~
m
--
+Zn
+Zn+anti-S iO0 j
• +anti-S I00
0.05
h. O e--
0
Ii 5
I
A
t
I
I
I
I
I
10
15
20
25
30
35
40
45
Time ( r a i n )
Fig. 2. Effect of anti-Sl00 antiserum on the stimulation of brain extract microtubule assembly by zinc. Assembly was studied in the presence of 500 #M added zinc (estimated free zinc, 1 #M) and anti-Sl00 antiserum (--~k--), in the presence of zinc and non-immune serum (--~) and without added zinc either in the presence of non-immune serum ( I ) or anti-Sl00 antiserum (e). Sera were diluted 1-+50.
did a equimolar mixture of Sl00a and b (results not shown). S100b had a small effect on the proportion cold-stable microtubules, reducing the percentage cold-stability from 64 to 59% but this was not statistically significant (Table 2). S100a had no effect on the microtubule cold-stability. DISCUSSION
The experiments using anti-SlO0 antibodies demonstrate that endogenous SIO0 in rat brain extracts can influence assembly of microtubules. This [extends the earlier findings that in the presence of calcium or zinc S100 can modulate the assembly of purified tubulin or microtubule protein (Baudier et al., 1982a; Deinum et al., 1983). The effectiveness of endogenous S100 in regulating assembly in such extracts suggest that S100 proteins may contribute to the regulation of brain microtubule assembly in vivo. Since S100 protein in rat brain is very largely (90%) or wholly S100b (Isobe et al., 1983; Baudier et al., 1985) it is probable that it is endogenous S100b which is the effective component. This view is supported by the observation that in the present experiments antiS100 antibodies and exogenous Sl00b had opposing effects on assembly kinetics and cold-stability; for example, Sl00b stimulated the early phase of assembly whereas the anti-S100 antibodies lengthened the lag phase. Changes in the lag phase of assembly are thought to reflect changes in the nucleation of the assembly reaction, that is the formation of some structure upon which microtubule elongation can occur. Thus the effect of anti-S100 antibodies is compatible with the data from experiments on purified microtubule protein (Donato, 1984) which have shown that S100 modulates the nucleation phase of assembly. Furthermore the effect of antiB.C. 18/S--B
SI00 antibodies was very largely on the lag phase rather than the rate of subsequent assembly and this suggests that the rate of nucleation was inhibited more than its extent. In addition to the effects on the kinetics of assembly S100b also changed the properties of the microtubules formed during incubation in vitro; inhibition of endogenous S100 with antibodies increased the microtubule cold-stability while addition of exogenous S100b caused a small, but statistically not significant, decrease. Stability to cold per se may not be physiologically significant but it may represent a more general form of stability which is involved in microtubule regulation (Job et al., 1983; Hesketh, 1985). Cold-stability has been found previously to be regulated by calcium and calmodulin (Margolis and Rauch, 1981; Job et al., 1983) and has been suggested to be due to the presence of specific proteins spaced periodically along the microtubule. The present results show that another cation-binding protein, Sl00b, can also modulate such stability and that this property is not unique to calmodulin. Zinc has also
Table 1. Effect of anti-SI00 antibodies on microtubule cold stability Incubation conditions Control ( + n o n - i m m u n e serum) + Anti-Sl00 (1 --~50 dilution) + Z n (500/aM) + A n t i - S l 0 0 + Zn + "'Seeds" + Anti-Sl00 + seeds Values given Students group; cp cp < 0.05
% Cold stability 63 _+ 3 (n = 4) 74 _+ 3 (n 76 + 2 (n 88 + 4 (n 60 + 5 (n 65 + 1 (n
= = = = =
4) a 4) b 4) c 4) 4) d
are means 5:SEM. G r o u p s were c o m p a r e d using t-test. , p <0.05; bp <0.01 c o m p a r e d to control < 0.05 compared to Zn only or anti-Sl00 only group; compared to anti-S100 group.
694
JOHN HESKETH and JACQUES BAUDIER
0.15 ~ a
-b
0.1
--
E t-
LIO O
ID
C°nt~+Sii0 b
e.,Q
s/
0 ffl
/
I
+//S~O0 a
e/
t~
/ . ='/
e-
0.05
_
//
,/~ I / 0
10
/,. 20
30
40
50
Time
I
I
I
i
J
10
20
30
40
50
(min)
Fig. 3. Effects of purified SI00 proteins on microtubule assembly in brain extracts. (a) Effect of 10/tM Sl00a. (b) Effect of 10 #M S100b. Both proteins were pre-incubated with the extracts for 5 min prior to initiation of assembly. Controls were run in the absence of any added proteins but with an equal volume of Pipes buffer added.
been found to increase microtubule stability in rat brain extracts (Hesketh, 1984) but this effect appears from the present data to be independant of the S100-induced increase. The finding that the addition of short microtubule fragments from control preparations abolishes the effect of anti-S100 antibodies on cold-stability suggests that cold-stability is controlled by the properties of an initial short length of microtubule on which subsequent polymerization occurs. The experiments with exogenous S100 proteins showed that the effects of the two SI00 components on the microtubule system were different; Sl00b stimulated the early part of the assembly reactions while S100a inhibited assembly. Different effects of the two S 100 components on the assembly of purified microtubule proteins have been observed previously (Deinum et al., 1983) although while this work was in progress further experiments with purified microtubule protein have shown the effects of Sl00a and b on microtubule assembly to be similar (Donato et al., 1985). This discrepancy cannot be explained at present. The disparate effects of the two proteins emphasises the need to carry out experiments with purified S100a and S100b rather than total unfractionated SI00 proteins. Small changes in the free zinc concentration in rat brain extracts can modulate microtubule assembly and stability (Hesketh, 1981, 1984) and one possible mechanism for such effects is mediation by SI00 Table 2. Effects of S100a and Sl00b on microtubule cold stability Incubation conditions
% Cold stability
Control + 10#M Sl00a + 10/tM Sl00b
64 + 4 (n = 4) 62 + 3 (n = 3) 59-f- 2(n = 3 )
Values given are means + SEM.
proteins. However, the effects of zinc were unaffected by anti-S100 antibodies, thus suggesting that zinc does not act by this mechanistri. Experiments with microtubule protein (Deinum et al., 1983) have suggested that Sl00a can chelate zinc and so prevent high zinc concentrations from inducing microtubule sheet formation. Such a chelation of zinc in brain extracts would explain the inhibition of assembly by Sl00a and would support the view (Hesketh, 1981, 1985) that the free zinc concentration in brain extracts is a critical determinant of microtubule assembly. It remains to be investigated if such a chelating effect of S100a is of physiological significance. In conclusion, the study of assembly in brain extracts has allowed demonstration that endogenous Sl00b regulates microtubule assembly and stability. However, the experimental system did not allow elucidation of the detailed mechanism and investigation of the complex relationship between calcium, zinc, S100 proteins and microtubules will require further reconstitution experiments with purified proteins. REFERENCES
Baudier J. and Gerard D. (1983) Ion binding to S100 proteins: structural changes induced by calcium and zinc on S100a and S100b proteins. Biochemistry 22, 3360-3369. Baudier J., Briving C., Deinum J., Haglid K., Sorskog L. and Wallin M. (1982a) Effect of S100 proteins and calmodulin on Ca 2+ induced disassembly of brain microtubule proteins/n vitro. F E B S Lett. 147, 165-167. Baudier J., Holtzcherer C. and Gerard D. (1982b) Zincdependant affinity chromatography of the S100b protein on phenyl-Sepharose. A rapid purification method. FEBS Lett. 148, 231-234. Baudier J., Mandel P. and Gerard D. (1983b) Bovine brain
S100 and microtubule assembly SI00 proteins: separation and characterization of a new S100 protein species. J. Neurochem. 40, 145-152. Baudier J., Labourdette G. and Gerard D. (1985) Rat brain S100b protein: purification, characterization and ionbinding properties. A comparison with bovine S100b protein. J. Neurochem. 44, 76-84. Deinum J., Baudier J., Briving C., Rosengren L., Wallin M., Gerard D. and Haglid K. (1983) The effect of S-100a and S-100b and Zn ~÷ on the assembly of brain microtubule proteins in vitro. FEBS Lett. 163, 287 291. Donato R. (1983) Effect of S-100 protein on assembly of brain microtubule proteins in vitro. FEBS Lett. 162, 310-313. Donato R. (1984) Mechanism of action of S-100 protein(s) on brain microtubule protein assembly. Biochem. biophys. Res. Commun. 124, 850-856. Donato R., Isobe T. and Okuyama T. (1985) S100 proteins and microtubules: analysis of the effects of rat brain S-100 (S100b) and ox brain Sl00a 0, Sl00a and S100b on microtubule assembly~lisassembly. FEBS Lett. 186, 65-69. Ghandour M. S., Labourdette G., Vincendon G. and Gombos G. (1981) A biochemical and immunohistological study of SI00 protein in developing rat cerebellum. Dev. Neurosci. 4, 98 109. Hesketh J. E. (1981) Impaired microtubule assembly in brain from zinc-deficient pigs and rats. Int. J. Biochem. 13, 921-926. Hesketh J. E. (1984) Microtubule assembly in rat brain extracts. Further characterization of the effects of zinc on assembly and cold-stability. Int. J. Biochem. 16, 1331-1339. Hesketh J. E. (1985) Regulation of microtubule assembly. Int. J. Biochem. 17, 761-766.
695
Isobe T., Nakajima T. and Okuyama T. (1977) Reinvestigation of extremely acidic proteins in bovine brain. Biochem. biophys. Acta 494, 222-232. Isobe T., Ishioka N., Masuda T., Takahashi Y., Ganno S. and Okuyama T. (1983) A rapid separation of S100 subunits by high performance liquid chromatography: the subunit compositions of S100 proteins. Biochem. Int. 6, 419-426. Job D., Rauch C. T., Fischer E. H. and Margolis R. L. (1983) Regulation of microtubule cold-stability by calmodulin-dependant and independant phosphorylation. Proc. natn. Acad. Sci. U.S.A. 80, 3894-3898. Kakiuchi S. and Sobue K. (1981) Ca ++ and calmodulin dependant flip-flop mechanism in microtubule assemblydisassembly. FEBS Lett. 132, 141-143. Klee C. B. and Vanaman T. C. (1982) Calmodulin. Adv. Prot. Chem. 35, 213-221. Labourdette G. and Marks A. (1975) Synthesis of S-100 protein in monolayer cultures of rat glial cells. Eur. J. Biochem. 58, 73-79. Mani R. S. and Kay C. M. (1981) Isolation and spectral studies on the calcium binding properties of bovine brain S-100a protein. Biochemistry 22, 3902-3907. Margolis R. L. and Rauch C. T. (1981) Characterization of rat brain crude extract microtubule assembly: correlation of cold stability with the phosphorylation state of a microtubule-associated 64-K protein. Biochemistry 20, 4451-4458. Timasheff S. N. and Grisham L. (1980) In vitro assembly of cytoplasmic microtubules. A. Rev. Biochem. 49, 565-591. Weisenberg R. C. (1972) Microtubule formation in solutions containing low calcium concentrations. Science 177, 1104-1105.