Biochemical and Biophysical Research Communications 254, 430 – 435 (1999) Article ID bbrc.1998.9957, available online at http://www.idealibrary.com on
Pyruvate Kinase as a Microtubule Destabilizing Factor in Vitro Bea´ta G. Ve´rtessy,* ,1 Do´ra Ba´nkfalvi,* Ja´nos Kova´cs,† Pe´ter Lo¨w,† Attila Lehotzky,‡ and Judit Ova´di* *Institute of Enzymology, Biological Research Center, Hungarian Academy of Sciences, POB 7, H-1518, Budapest, Hungary; †Department of General Zoology, University of Eo¨tvo¨s Lora´nd, Budapest, Hungary; and ‡Pharmaceutical Works of Gedeon Richter, Budapest, Hungary
Received December 3, 1998
Endogenous control of microtubule dynamism is essential in many cell types. Numerous microtubuleadhering proteins stabilize the polymer status, while very few protein factors are described with opposite effects. The brain- and muscle-specific M1 isoform of the enzyme pyruvate kinase is investigated here in this respect. Three pieces of evidence indicate antimicrotubular effects of this protein. (1) Pyruvate kinase inhibits taxol-induced tubulin polymerization into microtubules as revealed by turbidimetry. (2) Pelleting experiments show that pyruvate kinase partially disassembles taxol-stabilized microtubules into less sedimentable oligomers leading to the appearance of tubulin in the supernatant fractions. (3) Electron microscopy reveals the kinase-induced formation of great amounts of thread-like tubulin oligomers which tend to accumulate in a light/less sedimentable fraction. Immunoelectron micrographs using labeled antibody against pyruvate kinase provide evidence for the binding of pyruvate kinase to the thread-like oligomeric forms. The present data allow the assumption that pyruvate kinase may display multiple regulatory functions as a glycolytic control enzyme and as a modulator of microtubule dynamism. © 1999 Academic Press Key Words: microtubule antagonism; pyruvate kinase; microtubule dynamism; turbidimetry; differential pelleting; microtubule ultrastructure.
Microtubules (MTs), polymers of the a-b tubulin heterodimer, are central to numerous physiological phenomena, e.g., intracellular transport, cell shape formation and mitosis (1). The dynamic nature of the tubulin-MT equilibrium is indispensable for normal To whom correspondence should be addressed. Fax: 1361 4 665465. E-mail:
[email protected]. Abbreviations used: MT, microtubule; MAP, microtubuleassociated protein; PK, pyruvate kinase; SDS/PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis. 1
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
cell function as for example in neuronal plasticity and mitotic spindle formation/disassembly (2, 3, and references therein). The nondividing neurons of the adult mammalian brain require effective control of tubulin/ MT dynamism especially in the process of neurite extension (4, 5). Endogenous factors responsible for regulation of MT dynamism are expressed in a tissue- and cell cycle-specific manner so that the physiological status of the cell is imprinted in the pattern of expression of the regulatory factors (6). The numerous MTassociated proteins (MAPs) stimulate MT assembly by preferentially binding to and stabilizing the tubules (1). MAP-MT interactions are usually regulated by MAP phosphorylation (7). Lack of the MAP-stabilizing effects indirectly enhances MT disassembly, however, recently some proteins were identified at the molecular level which directly act in an antimicrotubular way (8 –10). One of these factors is stathmin/op18, a small phosphoprotein present in high amounts in different tumour cells (11), while other factors belong to the kinesin family. These factors seem to employ different antagonistic mechanisms of action (10, 12, 13). Continued search for endogenous regulators of MT dynamism is urged by the essential need for understanding the tubulin assembly/disassembly process under different physiological circumstances and in different tissues. Several enzymes of glycolysis, indispensable in brain as the first step in the catabolism of the most important neuronal energy source glucose, are present at concentrations significantly exceeding the requirement as simple catalysts (14). In vivo evidence from axonal transport indicates at least transient proximity of glycolytic enzymes and MTs (15, 16). In vitro studies documented significant binding of some enzymes (e.g., phosphofructokinase, glyceraldehyde-3-phosphate dehydrogenase, aldolase, pyruvate kinase (PK)) to tubulin and MTs (17, 18). Enzyme binding modulated microtubular ultrastructure in the case of glyceraldehyde-3-phosphate dehydrogenase (19, 20) and
430
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
phosphofructokinase (18, 21, 22) but not aldolase (23). In the present study we decided to test the effects of the brain- and muscle-specific M1 isoform of PK (24) on tubulin assembly and MT ultrastructure by turbidimetry, differential pelleting, and morphological electron microscopic investigation. MATERIALS AND METHODS Materials. PK from rabbit muscle was obtained from Boehringer, buffer substances and other chemicals were obtained from Sigma. The crystalline suspension of the enzyme was centrifuged at 10,000g for 5 min. The pellet was resuspended in standard buffer (50 mM 2-[N-morpholino]ethanesulfonic acid (Mes), pH 6.6, containing 100 mM KCl, 2 mM dithioerythritol, 1 mM EGTA and 5 mM MgCl 2), and then the enzyme was dialyzed against the same buffer. Protein concentrations were determined spectrophotometrically, using the absorption coefficients of A 0.1% 5 1.03 cm 21 at 276 nm for tubulin (25) and A 0.1% 5 0.54 cm 21 at 280 nm for PK (26). Molar concentrations of tubulin dimer and PK tetramer were used in all calculations based on molecular weights of 100,000 (27) and 232,000 (28), respectively. Protein concentration of MTs was determined by Bradford’s assay (29) using bovine serum albumin as standard, it is expressed in tubulin dimers. Protein purity and the composition of the pellets and supernatants was determined by discontinuous sodium dodecyl sulfate–polyacrylamide gel electrophoresis in 9% polyacrylamide gels (SDS/PAGE) (30). No proteolytic degradation of either PK or tubulin was observed throughout the experimental procedures. PK enzymatic activity was determined by the lactate dehydrogenase coupled assay (28) at 25°C. Assay buffer was 100 mM N-[2hydroxyethyl]piperazine-N9-[2-ethane sulfonic acid] (Hepes), pH 7.5, containing 100 mM KCl, 5 mM MgCl 2 and 2 mM EGTA. The reaction mixture contained 1 mM phosphoenolpyruvate, 1 mM ADP, 0.3 mM NADH, 13 mg/ml lactate dehydrogenase and 1-2 nM PK in the assay buffer. Reaction was started by the addition of ADP, and NADH consumption was followed at 340 nm in thermostatted cuvettes using a JASCO V500 spectrophotometer. The specific activity of PK was determined to be 350 U/mg under these conditions, in good accordance with previous data (28). Control experiments showed that PK specific activity did not diminish during the experimental procedures of dialysis, incubation and pelleting, therefore protein structure most probably remained intact. Activity assays were used to quantify PK in the pellet and supernatant fractions. Data given in the text refer to mean 6 standard deviation for triplicate measurements. Tubulin and MT preparation. MAP-free tubulin was prepared from bovine brain as described (25). Purified tubulin showed no contamination with MAP when run on overloaded 9% polyacrylamide gels using SDS/PAGE. Tubulin was stored in 10 mM sodium phosphate buffer pH 7.0 containing 1 M sucrose, 0.5 mM MgCl 2 and 0.1 mM GTP at 280°C and dialyzed overnight against 50 mM Mes pH 6.0 buffer containing 2 mM dithioerythritol before use. Tubulin solution at a final concentration of 0.1– 0.2 mM in standard buffer was warmed to 37°C, and taxol was added to a final concentration of 20 mM. After a 30-min/37°C incubation, MTs were pelleted by 25 min 3 100,000g 3 35°C centrifugation and resuspended in standard buffer. Turbidity measurements. Tubulin at a final concentration of 12.5 mM was incubated for 3 min with or without PK in standard buffer at 37°C. The assembly was started by addition of taxol to the samples in a final concentration of 20 mM. Absorbance change was monitored at 350 nm in a JASCO V500 spectrophotometer. Pelleting measurements were performed at 30°C. MTs at 10 mM concentration were incubated for 30 min without or with PK at the concentrations indicated in the text. Following the incubation, aliquots were centrifuged for 25 min at different speed as indicated in
the text. Supernatant and pellet fractions were separated and pellets were resuspended to the original volume of the incubation mixture. Aliquots of each fraction was run on SDS/PAGE followed by densitometric analysis using the Bio-Rad GelDoc 1000 densitometer with the Molecular Analyst software as described previously (22). Densitometric data in the text are given as mean 6 standard deviation for triplicate measurements. Rat anti-PK was produced by immunizing animals with rabbit muscle PK. Twenty-five micrograms of protein was dissolved in 250 ml sterile phosphate buffered saline, the solution was homogenized and emulsified with 250 ml Freund’s complete adjuvant. Two subcutaneous injections, 250 ml each, were given in the back of the animal. After one, two and three weeks the same injections were repeated, but with incomplete adjuvant. The animal was bled on the fourth week, serum was used in the experiments at 1:50 dilution. Electron microscopy. For routine transmission electron microscopy MT samples were prepared as described previously (22). Briefly, MT-containing samples were pelleted, and the pellets were fixed with glutaraldehyde/tannic acid, postfixed with OsO 4, stained with uranyl acetate and embedded in Durcupan (Fluka, Switzerland). Thin sections were contrasted with uranyl acetate and lead citrate. Specimens were examined and photographed in a JEOL CX 100 electron microscope operated on an accelerating voltage of 80 kV. Magnification was calibrated with a diffraction grating replica (2160 l/mm, Balzers). For immunogold electron microscopy a small pellet of MTs was fixed in a mixture of freshly prepared 1% formaldehyde and 0.2% glutaraldehyde and embedded in Durcupan. Thin sections were cut and mounted on nickel grids. Immunogold labeling was carried out by a postembedding method. Sections on grids were etched in 5% H 2O 2 then washed thrice in double distilled water. The grids were blocked with non-fat dry milk and bovine serum albumin, then incubated on a drop of rat anti-PK IgG serum in a humid chamber overnight at 4°C. After washing in 1% bovine serum albumin in phosphate buffered saline (pH 7.4), the buffer was exchanged to 0.1 M Tris–HCl (pH 8.3). Then the grids were floated on a drop of 5 nm colloidal gold conjugated goat anti-rat IgG (Sigma) diluted 1:100 in 1% bovine serum albumin, 0.25% Tween 20 in 0.1 M Tris–HCl in a humid chamber for 5 h at 4°C. Finally grids were washed and dried. In control experiments the anti-PK antibody was omitted in order to check the specificity of the staining. Sections were counterstained with 2% uranylacetate prior to examination as above.
RESULTS PK effect on tubulin polymerization as followed by turbidimetry. Figure 1A presents turbidimetric measurements of taxol-induced tubulin polymerization at 12.5 mM tubulin concentration in the absence and in the presence of PK (0.16 – 4.5 mM). The initial rate of the turbidity increase is efficiently reduced at even relatively low, substoichiometric PK amounts. Dependence of the initial rate on kinase concentration is presented in Fig. 1B. There is a sharp decline in the rate at low kinase concentrations, at 0.81 mM kinase the rate is only 44% of the control. At higher PK concentrations the rate tends to reach a limiting value of 18 –20% of the control. Half-maximal effect is reached at approximately 1.6 mM PK. Despite the significant decrease of the initial rate, turbidity values tend to reach the same equilibrium level after 60 min of polymerization either without or with PK (Fig. 1B). The effects described above could not be induced by replacing PK with a considerably inert protein, bovine
431
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Effect of PK on turbidimetric parameters of tubulin polymerization. (A) Assembly of 12.5 mM tubulin dimers in the absence or presence of PK (at concentrations in mM indicated on the experimental tracings) was followed. (B) Initial rate of turbidity increase estimated from the highest slope of the turbidity curves (left axis, closed squares) and total DOD calculated from the OD values before addition of taxol and after 60 min from taxol addition (right axis, open squares) were plotted in function of PK concentration.
serum albumin (not shown). The glycolytic enzymes phosphofructokinase and aldolase were also tested in this respect. In accordance with previously reported data (22, 23, 31), although they bind to both tubulin and MTs, neither of these two enzymes altered the initial rate of taxol-induced tubulin polymerization. The apparently substoichiometric inhibition of tubulin assembly by PK suggested that kinase preferentially interacts with tubulin oligomers. To test this suggestion and to check whether PK could exert antimicrotubular effects on preformed MTs, pelleting experiments were carried out. Pelleting of PK with MTs. MT samples were incubated in the absence or presence of PK, then aliquots were spun at high (100,000g) or low (15,000g) centrifugal speed. This differential centrifugation was expected to achieve size-dependent separation of tubulin oligomers and MTs. Based on a similar rationale, bundled and single MTs were successfully separated in an earlier study (31). The enzyme in the absence of MTs did not sediment at all even at the highest speed (100,000g) and at the highest concentration used (4.3 mM) (Fig. 2A). MTs in the absence of kinase were quantitatively sedimented already at 15,000g (Fig. 2B). Figures 2C and 2D show the composition of the supernatant and pellet fractions of MT-PK mixtures after high or low speed centrifugation. There is a considerable amount of tubulin in the supernatant of the low speed spin if the sample contained PK. Therefore, kinase induces the appearance of less sedimentable tubulin oligomers of presumably lower molecular weight compared to MTs. No tubulin is present in the
supernatant of the high speed spin which pellets tubulin oligomers together with MTs. The absence of tubulin from the high speed supernatant indicates that the addition of PK to MTs does not induce the formation of small, non-sedimentable tubulin-PK heterocomplexes. This observation argues against the presence of the tubulin dimer-PK complex under the experimental conditions. PK is present predominantly in the supernatant of the low speed spin, while it is distributed between pellet and supernatant of the high speed spin. The presence of kinase in the high speed pellet indicates its binding to the tubulin oligomer/MT fraction as control experiments showed that the enzyme alone did not sediment under the experimental conditions (Fig. 2A). Kinase is suggested to preferentially bind to the tubulin oligomeric fraction as opposed to MTs since it practically cannot be pelleted at low speed. Binding of PK to the tubulin oligomeric fraction, present in the supernatant of the low speed spin, was directly demonstrated by two-step centrifugation experiments. Samples of MT and PK were first spun at 15,000g. The low speed supernatant, containing the presumed tubulin oligomers together with bound and unbound PK, was further centrifuged at 100,000g. Gel electropherograms of the pellet fraction of this second step is shown in Fig. 2E. The fractionated pellet contains mainly tubulin with a small amount of bound PK. Quantitative densitometry of the composition of pellet and supernatant fractions reinforces the qualitative results. Both the amount of tubulin in the supernatant of the high speed spin and the amount of PK in the
432
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
where the IC 50 value of PK was determined to be substoichiometric (1.6 mM at 12.5 mM tubulin in the turbidimetric assay).
FIG. 2. Differential pelleting of MT-PK mixtures. Electropherograms of pelleted and supernatant fractions. PK and TUB inscriptions mark the position of the respective protein bands on the gel. First row indicates S and P standing for supernatant and pellet fractions, the latter being resuspended to the original volume of the mixture. 5-5 ml of each fraction were applied to SDS–PAGE gels, unless stated otherwise. Second row gives the concentration of PK (in mM) in the incubation mixtures before spin. (A) PK was incubated (30 min, 30°C) without MT, then centrifuged at 100,000g (high speed). Although no sedimentable material was visible after removing the supernatant, the centrifuge tube was thoroughly rinsed with 80 ml standard buffer (the original volume of the mixture) to gain the pellet fraction. Note the absence of PK from the pellet. (B) 10 mM MT was incubated (30 min, 30°C) without PK, then centrifuged at 15,000g (low speed). Note the absence of tubulin from the supernatant. (C and D) (10 mM MT)–(2.1– 4.3 mM PK) mixtures were incubated (30 min, 30°C), then spun at 100,000g (high speed, C) or 15,000g (low speed, D). Supernatant and pellet fractions were investigated for protein content. (E) Final pellet fractions of MT-PK mixtures after the two-step centrifugation protocol. Samples were first spun at 15,000g, then supernatants were further spun at 100,000g. To enhance visualization of the faint bands in this final pellet, 10-ml aliquots were applied to the gel.
pellet of the low speed spin are at the detection limit. The amount of tubulin in the low speed spin supernatant, containing the less sedimentable tubulin oligomers, is (2 6 0.5) mM in the sample of 10 mM MT and 4.3 mM PK. The amount of PK bound to this oligomeric tubulin fraction is (0.25 6 0.05) mM, as calculated from electropherograms such as shown in Fig. 2E. Independent quantitative determination of PK distribution in the pellet and supernatant fractions was performed by activity measurements. Samples of 10 mM MT and 4.3 mM PK were incubated (30 min, 30°C) then subjected to the two-step spin protocol as above. PK activity was measured in the pellet and supernatant fractions. Data from the activity assays indicated that the amount of PK bound to the oligomeric tubulin fraction is (0.35 6 0.08) mM. This value is in reasonable accordance with the densitometric determination. Both methods therefore show that PK binds in substoichiometric amounts to the tubulin oligomeric fraction. The quantitative data support the two main statements of this section: (i) PK induces the formation of less sedimentable tubulin oligomers and (ii) it binds preferentially and substoichiometrically to this less sedimentable tubulin oligomer fraction. The substoichiometric binding is in agreement with the results from the turbidity assays
Morphological studies of the kinase induced alterations in MT structure. MT-PK mixtures prepared as described in the previous section were investigated by thin section electron microscopy (Fig. 3). MT (10 mM) was incubated in the absence of kinase, or in the presence of 2.1 mM kinase, then aliquots from the samples were pelleted. Control experiments showed that pellets prepared without the addition of PK contain intact MTs with minute amounts of oligomeric threads (Fig. 3A). However, if pellets were prepared from samples containing PK, then the morphology is drastically altered. Large amounts of oligomeric thread-like aggregates are dispersed within some seemingly intact MTs (Fig. 3B). The aggregates tend to accumulate on the top of the pellet suggesting their less sedimentable behavior. Therefore, the two-step differential centrifugation protocol was attempted to gain a fraction likely enriched in the oligomeric threads. Final pellet fractions prepared by the two-step centrifugation protocol (15,000g supernatants pelleted at 100,000g) were embedded and subjected to thin section electron microscopy. Pellets sedimented from the 15,000g supernatants of samples prepared without addition of PK are very small and contain loosely arranged MTs and trace amounts of oligomers (data not shown). On the other hand, pellets from supernatants of PK-containing samples can easily be collected, in agreement with the results from the previous section. These pellets predominantly contain clumps of thread-like oligomers (Fig. 3C). The threads are about 14 nm thick, contain knob-like densities and may form short helices and rings. Protein composition of these pellets (shown on Fig. 2E) demonstrated approximately 90% tubulin content. Therefore, the thread-like structures can probably be identified as tubulin oligomers. To study the distribution of PK between MTs and oligomeric threads, anti-PK immunogold labeling was applied to sections prepared from high speed (100,000g) spun samples containing 10 mM MT and 2.1 mM PK. As shown on Fig. 3D, black dots representing the immunogold conjugates tend to be dispersed on the thread-like oligomeric structures and are not present on intact MTs. Control sections stained without the application of primary antibody are practically free of gold label (Fig. 3E). Predominant localization of PK on the oligomeric threads supports the conclusion of the pelleting results, namely that PK preferentially interacts with tubulin oligomers compared to intact MTs. Taken together, electron microscopic data indicate that addition of pyruvate kinase to MTs induces the formation of thread-like oligomers which are the main constituents of the light/less sedimentable fraction of enzyme-containing samples.
433
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Pyruvate-kinase-induced changes in the ultrastructure of MTs. (A and B) Thin sections were prepared from samples of 10 mM MT in the absence (A) or in the presence of 2.1 mM kinase (B). Samples were incubated (30 min, 30°C) , then centrifuged at 15,000g. Pellets were fixed, embedded and sectioned for electron microscopy. (C) Thread-like oligomeric aggregates formed in the presence of PK. The sample was assembled from a mixture of MTs (10 mM) and PK (2.1 mM), incubated (30 min, 30°C) and centrifuged at 15,000g. Then the supernatant was further pelleted at 100,000g, fixed, embedded and sectioned for electron microscopy. (D) Immunostaining for PK. The section was prepared from the 100,000g pellet of a 10 mM MT-2.1 mM PK sample. 5-nm immunogold conjugates (black dots) are seen mainly on the thread-like oligomers. (E) Immunostaining was performed in a manner identical to the sample presented in D, but the primary antiPK antiserum was omitted. Note the absence of immunogold labeling. Bars for A, B, and C, 200 nm; for D and E, 100 nm.
DISCUSSION We identified the M1 isoform of PK as a MT destabilizing factor in vitro by three independent approaches: (i) turbidimetry, (ii) differential pelleting, and (iii) electron microscopy. In the taxol-stabilized system used in the present experiments, kinase is suggested to interact with tubulin oligomers that scatter light (Fig. 1) but are less sedimentable than MTs (Figs. 2 and 3). These tubulin oligomers are tentatively identified as the thread-like structures seen in electron micrographs of kinase-containing samples (Fig. 3). In this respect, kinase is different from the antimicro-
tubular stathmin, which sequesters tubulin dimers thereby inhibiting MT formation (12, 13). PK, as evidenced in the present results, induces morphological alterations of MTs if it is added to taxol-stabilized MTs practically free of nonsedimenting tubulin dimers. Under the experimental conditions, we could not detect non-sedimentable (a,b-tubulin dimer)-PK heterocomplexes in the pelleting experiments. These phenomena argue against the hypothesis that the tubulin dimer is the main interacting partner of PK. More probably, kinase targets MTs and degrades them by inducing partial depolymerization. Elucidation of molecular details of the kinase-induced MT destabilization requires
434
Vol. 254, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
further studies. However, the present data, gained by the different experimental approaches, consonantly argue for the interpretation that PK-induced morphological effects do not lead to complete depolymerization into tubulin dimers. The present experiments were performed at protein concentrations commensurable to in vivo levels in mammalian brain. PK was present at 0.16 – 4.5 mM concentration which is to be compared to the value of 1.5 mM (0.35 mg/g weight of tissue) found in mammalian brain (14). The brain-specific M1 isoform of rabbit PK, also present in muscle, was used with bovine brain tubulin. Although the sequence of bovine M1 PK is not published, comparison of rabbit M1 PK with other mammalian (human, rat, cat) M1 PK sequences show about 92% identity and 94% homology (32, 33). The bovine sequence would most probably give similar high homology. These slight alterations in the PK polypeptide are not expected to change the behavior of the MT-kinase system, allowing the assumption that endogenous M1 PK in bovine brain may have similar effects on cellular MTs. Experiments addressing this hypothesis are in progress in our laboratory. As the M1 isoform is also present in muscle, it will be worthwhile to investigate its interactions with muscle-specific tubulin isotypes as well. Mechanistically, the inhibition of MT formation by PK seems to be different from the effects of stathmin. It is therefore an additional useful tool for studying MT depolymerization in vitro. The present results taken together with earlier reports indicate that the surface of MTs contains specific recognition sites for MT stabilizer (MAPs, phosphofructokinase) (1, 22) and destabilizer (Kar3, PK) (10), present data) protein factors. Fine adjustment of these interactions may participate in modulation of MT dynamics. The glycolytic kinases PFK and PK, ancient proteins appearing early in evolution, may then perform a dual role as controllers of glucose catabolism and effectors of MT ultrastructure. ACKNOWLEDGMENTS The excellent technical assistance of Emma Hlavanda is gratefully acknowledged. We thank Sarolta Sipos for help with the electron microscopic studies. This work was supported by grants from the Hungarian National Science Foundation OTKA (F-17392 and F-020862 to B.G.V., T-25291 and T-17830 to J.O.), from the Ministry of Education (FKFP 0158/97 to J.K., FKFP1023/97), and from the European Community (Copernicus ERBIC15CT960307). B.G.V. and P.L. are holders of Ja´nos Bolyai Research Fellowships.
REFERENCES 1. Hyams, J. S., and Lloyd, C. W. (Eds.) (1994) Microtubules. WileyLiss, New York. 2. Joshi, H. C. (1998) Curr. Opin. Cell Biol. 10(1), 35– 44.
3. Desai, A., and Mitchison, T. J. (1997) Annu. Rev. Cell. Dev. Biol. 13, 83–117. 4. Himi, T., Okazaki, T., Wang, H., McNeill, T. H., and Mori, N. (1994) Neuroscience 60(4), 907–926. 5. Sugiura, Y., and Mori, N. (1996) Dev. Brain Res. 90, 73–91. 6. McNally, F. J. (1996) Curr. Opin. Cell Biol. 8(1), 23–29. 7. Diaz-Nido, J., Serrano, L., Hernandez, M. A., and Avila, J. (1990) J. Neurochem. 54(1), 211–222. 8. Belmont, L. D., and Mitchison, T. J. (1996) Cell 84(4), 623– 631. 9. Walczak, C. E., Mitchison, T. J., and Desai, A. (1996) Cell 84(1), 37– 47. 10. Endow, S. A., Kang, S. J., Satterwhite, L. L., Rose, M. D., Skeen, V. P., and Salmon, E. D. (1994) EMBO J. 13(11), 2708 –2713. 11. Sobel, A. (1991) Trends Biochem Sci. 16(8), 301–305. 12. Curmi, P. A., Andersen, S. S., Lachkar, S., Gavet, O., Karsenti, E., Knossow, M., and Sobel, A. (1997) J. Biol. Chem. 272(40), 25029 –25036. 13. Jourdain, L., Curmi, P., Sobel, A., Pantaloni, D., and Carlier, M. F. (1997) Biochemistry 36(36), 10817–10821. 14. Ottaway, J. H., and Mowbray, J. (1977) Curr. Topics Cell Regul. 12, 107–208. 15. Oblinger, M. M., Foe, L. G., Kwiatkowska, D., and Kemp, R. G. (1988) J. Neurosci. Res. 21(1), 25–34. 16. Brady, S. T., and Lasek, R. J. (1982) Methods Cell Biol. 25(Pt. B), 365–398. 17. Volker, K. W., and Knull, H. (1997) Arch. Biochem. Biophys. 338, 237–243. 18. Lehotzky, A., Pa´lfia, Z., Kova´cs, J., Molna´r, A., and Ova´di, J. (1994) Biochem. Biophys. Res. Commun. 204, 585–591. 19. Huitorel, P., and Pantaloni, D. (1985) Eur. J. Biochem. 150(2), 265–269. 20. Somers, M., Engelborghs, Y., and Baert, J. (1990) Eur. J. Biochem. 193(2), 437– 444. 21. Vertessy, B. G., Kova´cs, J., and Ova´di, J. (1996) FEBS Lett. 379(2), 191–195. 22. Vertessy, B. G., Kova´cs, J., Lo¨w, P., Lehotzky, A., Molna´r, A., Orosz, F., and Ova´di, J. (1997) Biochemistry 36(8), 2051–2062. 23. Vertessy, B. G., Orosz, F., Kova´cs, J., and Ova´di, J. (1997) J. Biol. Chem. 272(41), 25542–25546. 24. Noguchi, T., Inoue, H., and Tanaka, T. (1986) J. Biol. Chem. 261(29), 13807–13812. 25. Na, G. C., and Timasheff, S. N. (1986) Biochemistry 25(20), 6214 – 6422. 26. Boyer, P. D. (1962) in The Enzymes (Boyer, P. D., Lardy, H., and Myrback, K., Eds.), Vol. 6, pp. 95–113, Academic Press, New York. 27. Little, M., and Seehaus, T. (1988) Comp. Biochem. Physiol. [B] 90(4), 655– 670. 28. Cheng, X., Friesen, R. H., and Lee, J. C. (1996) J. Biol. Chem. 271(11), 6313– 6321. 29. Bradford, M. M. (1976) Anal. Biochem. 72, 248 –254. 30. Laemmli, U. K. (1970) Nature 227(259), 680 – 685. 31. Lehotzky, A., Telegdi, M., Liliom, K., and Ova´di, J. (1993) J. Biol. Chem. 268(15), 10888 –10894. 32. Larsen, T. M., Laughlin, L. T., Holden, H. M., Rayment, I., and Reed, G. H. (1994) Biochemistry 33(20), 6301– 6309. 33. Jurica, M. S., Mesecar, A., Heath, P. J., Shi, W., Nowak, T., and Stoddard, B. L. (1998) Structure 6(2), 195–210.
435
