Direct regulation of microtubule dynamics by protein kinase CK2.

Microtubule dynamics is essential for many vital cellular processes such as morphogenesis and motility. Protein kinase CK2 is a ubiquitous protein kinase that is involved in diverse cellular functions. CK2 holoenzyme is composed of two catalytic alpha or alpha' subunits and two regulatory beta subunits. We show that the alpha subunit of CK2 binds directly to both microtubules and tubulin heterodimers. CK2 holoenzyme but neither of its individual subunits exhibited a potent effect of inducing microtubule assembly and bundling. Moreover, the polymerized microtubules were strongly stabilized by CK2 against cold-induced depolymerization. Interestingly, the kinase activity of CK2 is not required for its microtubule-assembling and stabilizing function because a kinase-inactive mutant of CK2 displayed the same microtubule-assembling activity as the wild-type protein. Knockdown of CK2alpha/alpha' in cultured cells by RNA interference dramatically destabilized their microtubule networks, and the destabilized microtubules were readily destructed by colchicine at a very low concentration. Further, over-expression of chicken CK2alpha or its kinaseinactive mutant in the endogenous CK2alpha/alpha'-depleted cells fully restored the microtubule resistance to the low dose of colchicine. Taken together, CK2 is a microtubule-associated protein that confers microtubule stability in a phosphorylation-independent manner.

Protein kinase CK2 (formerly known as casein kinase 2) is ubiquitously expressed and highly conserved in eukaryotic cells (1)(2)(3)(4). It comprises two catalytic ␣ or ␣Ј subunits and two regulatory ␤ subunits to form a heterotetrameric structure in which the two ␤ subunits dimerize to link the two ␣ or ␣Ј subunits (5). As a protein serine/threonine kinase, CK2 has a very broad phosphorylation spectrum, and over 300 protein substrates of CK2 have been identified to date (6). A number of studies have indicated that CK2 is involved in a wide variety of cellular processes including cell cycle, apoptosis, transcriptional regulation, and signal transduction (1,3,6). CK2 is instrumental and necessary for promoting cell survival (3,7). Disruption of genes encoding both of the catalytic subunits of CK2 is synthetic lethal in fission yeast (8,9). Similarly, it is embryonic lethal when CK2␤ is knocked down in Caenorhabditis elegans by RNA interference or in mice by gene disruption, reminiscent of an essential role of CK2␤ during embryonic development and organogenesis (10,11). Hence, production of both the ␣ and ␤ subunits of CK2 appears to be mandatory for cell viability.
A few lines of evidence have lead to implication that CK2 might be involved in the regulation of microtubule cytoskeleton reorganization (12)(13)(14). CK2 was localized to microtubule structures such as the mitotic spindle of dividing cells and was found to associate with the cold-stable fraction of microtubules from the rat brain (14,15). More recently, the ␣ and ␣Ј subunits were shown to bind tubulin in a far Western assay (16). Further, CK2 is able to phosphorylate a number of microtubule elements, including MAP1B and a neuron-specific ␤-tubulin isotype (6). The phosphorylation of MAP1B was proposed to facilitate the microtubule association of MAP1B and thereby microtubule assembly, whereas the physiological role of the ␤-tubulin isotype phosphorylation is still unclear (12,17). Despite these findings, the direct correlation of CK2 and microtubule stability has not been established.
In the present study, we have investigated the physical association of CK2 with microtubules and the direct effect of CK2 on microtubule dynamics. Our results show that CK2 is a microtubule-associated protein (MAP) 1 that induces microtubule assembly and bundling in vitro. CK2-polymerized microtubules appear stable under cold treatment. In cultured cells, knockdown of CK2␣/␣Ј has a severe effect on microtubule stability, which implies that CK2 mediates microtubule integrity in vivo. Moreover, a kinase-inactive mutant of CK2 displayed the same microtubule polymerizing and stabilizing activity in vitro and in vivo. Thus, the microtubule assembling and stabilizing action of CK2 is independent of its kinase function.
Protein Binding Assay-Proteins tagged with GST or His 6 were bacterially expressed and prepared as described previously (19). To test tubulin binding, GSH-Sepharose beads (Amersham Biosciences) prebound with GST, GST-CK2␣, or the complex of GST-CK2␣/His-CK2␤ were incubated with purified tubulin (Ͼ99% pure and MAP-free, Cytoskeleton) for 1 h at 4°C. After being extensively washed with binding buffer (20 mM Tris-HCl, pH 7.4, 50 mM NaCl, 20 mM MgCl 2 , 1 mM dithiothreitol, and 0.1% Nonidet P-40), the beads were boiled in SDS-PAGE sample buffer and analyzed by immunoblotting. Antibodies against ␣and ␤-tubulin were from Sigma. The binding of His-tagged proteins with tubulin was performed with nickel-nitrilotriacetic acid beads (Ni-NTA, Qiagen) in binding buffer without dithiothreitol. In the microtubule binding assay, microtubules, which were pre-assembled using taxol in PEM buffer (80 mM PIPES, pH 6.8, 1 mM MgCl 2 , 1 mM EGTA) supplemented with 1 mM GTP, were incubated with the indicated proteins. The samples were subsequently loaded onto a buffered cushion (50% glycerol in PEM buffer) and centrifuged to spin down the microtubules and associated proteins. The pellet and the supernatant were analyzed by immunoblotting.
Microtubule Assembly-Microtubules were assembled in vitro from the purified MAP-free tubulin at 2 mg/ml in PEM buffer supplemented with 1 mM GTP at 35°C, and the turbidity of the solutions was monitored at 340 nm (20). CK2 was added at various amounts as indicated to promote the assembly. To visualize assembled microtubules, tubulin and rhodamine-labeled tubulin (Cytoskeleton) at the ratio of 7:1 were used in the polymerization (21). Microtubules were fixed with 0.5% gluteraldehyde and visualized by fluorescence microscopy.
Differential Tubulin Extraction from Intact Cells-Differential extraction of tubulin heterodimers and polymers from cells was performed using a protocol described previously (22). Briefly, cultured cells were lysed with the microtubule-stabilizing buffer (80 mM PIPES, pH 6.8, 1 mM MgCl 2 , 1 mM EGTA, 0.5% Triton X-100, 10% glycerol, and Roche protease inhibitor mixture), which was prewarmed to 35°C, to extract cytosolic soluble tubulin heterodimers and preserve microtubules (assembled insoluble tubulin polymers). The extract was cleared by centrifugation and the supernatant designated as the free tubulin fraction. After a brief washing with the microtubule-stabilizing buffer, the pellet was extracted in the microtubule-destabilizing buffer (20 mM Tris, pH 7.4, 150 mM NaCl, 1% Triton X-100, 10 mM CaCl 2 , and Roche protease inhibitor mixture). The extract was clarified by centrifugation to yield the polymerized tubulin fraction. Both fractions were analyzed by immunoblotting, and each band on the blots was quantitated using a Bio-Rad GS-700 imaging densitometer and analyzed with the Multi-Analyst, version 1.0.1, program (Bio-Rad).
Cell Culture, Transfection, and Immunofluorescence-COS-7, HeLa and 293T cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. The siRNA sequence designed for human CK2␣/␣Ј is 5Ј-CCAGCUGGUAGUCAUCUUGUU-3Ј, which has a few discrepancies with the corresponding sequence of chicken CK2␣/␣Ј. 20 M CK2␣/␣Ј siRNA or a scrambled siRNA sequence was applied into the transfection using a TransIT-TKO transfection reagent (Mirus). Simultaneous transfection of siRNA and plasmid DNA was done using TransIT-TKO and LipofectAMINE (Invitrogen) concurrently. After transfection, the cells were cultured for 24 h before treatment with 0.2 M colchicine (Sigma) for 3 h. The cells were subjected to differential extraction of free and polymerized tubulin or to immunostaining. For immunofluorescence, the cells were fixed in PBS containing 4% paraformaldehyde and then permeabilized in PBS containing 0.2% Triton X-100. After a blocking wash with 10% goat serum and 0.1% Triton X-100 in PBS, immunostaining was performed with antibodies as indicated. CK2␣and CK2␤-specific antibodies were from Santa Cruz Biotechnology. The secondary antibodies are Fluor594 goat anti-mouse IgG and Fluor488 donkey anti-goat IgG (Molecular Probes). The cells were then washed in PBS, mounted, and photographed on an MRC-1024 laser scanning confocal microscope (Bio-Rad).

CK2 Forms a Direct Complex with Microtubules-
The direct association of CK2 and microtubules was probed by a series of binding assays using recombinant CK2 and purified MAP-free tubulin as well as pre-assembled microtubules. The ␣/␤ heterodimer of tubulin was found to associate with the catalytic ␣ subunit as well as the holoenzyme of CK2 (Fig. 1A). CK2␤ alone did not result in the pull-down of any tubulin (Fig. 1B), which is in agreement with a previous observation using far Western blotting (16). To verify the microtubule association of CK2, taxol-assembled microtubules were incubated with CK2 holoenzyme or its individual subunit proteins. The microtubules were then spun down to test whether these proteins co-precipitated with the microtubules. Consistently, CK2␣ and the holoenzyme of CK2 were found to associate with the microtubule pellet, whereas CK2␤ and GST, as a control protein, failed to co-precipitate with the microtubules (Fig. 1C), indicating that the CK2 holoenzyme associates with microtubules at a high affinity through CK2␣.
Cellular localization of CK2 to microtubule networks was revealed by immunofluorescent staining of cultured COS-7 cells. Microscopic imaging of endogenous CK2␣ and CK2␤ displayed a clearly defined positioning with the microtubule network, particularly in the cell periphery ( Fig. 2A). As confirma-tion, pools of tubulin existing as free heterodimers or polymers (microtubules) were differentially extracted from the cultured cells to examine the distribution of CK2 (22). Both CK2␣ and CK2␤ appeared in the microtubule fraction as well as the fraction of free tubulin heterodimers, although there appeared to be more CK2␤ in the microtubule fraction (Fig. 2B). Taken together with the results from the in vitro binding assays, this provides evidence of the direct association of CK2 with cellular microtubules.
CK2 Induces Microtubule Polymerization-We next investigated whether CK2 has any effect on microtubule dynamics by using an in vitro assay of microtubule assembly from purified MAP-free tubulin (20). During the assay, the turbidity change of the solution was measured as tubulin polymerizes or depolymerizes. In the absence of CK2, there was minimal polymerization of tubulin even after a prolonged incubation (Fig. 3, A  and B). The addition of CK2 at a ratio of 1:240 to tubulin resulted in substantial polymerization of tubulin into microtubules (Fig. 3, A and B). Clearly, both the rate and extent of polymerization were dramatically enhanced by CK2. When the amount of CK2 was increased, tubulin polymerization was increased in a dose-dependent manner (Fig. 3, A and B). To verify the microtubule formation, rhodamine-labeled tubulin was applied into the polymerization experiments for direct visualization of the assembled microtubules by fluorescence microscopy (21). As shown in Fig. 3C, microtubule filaments and bundles were readily observed with the CK2-incubated tubulin, whereas the incubation of tubulin without CK2 showed no obvious microtubule formation. Therefore, we have found that CK2, in addition to showing high affinity binding to tubulin and microtubules, induces the assembly of tubulin into microtubules. Moreover, CK2 appeared to cause microtubule bundling, suggesting a strong stabilizing effect on the microtubules.
CK2 holoenzyme is a tetrameric complex of two ␣ or ␣Ј subunits and two ␤ subunits (5). Given that observation that CK2␣ of the holoenzyme interacts with microtubules, we explored whether the microtubule assembling function of CK2 is restricted to the holoenzyme by applying the ␣ and ␤ subunits of CK2 individually into the microtubule assembly assay. In contrast to the holoenzyme, when either the ␣ or ␤ subunit was tested, there was minimal polymerization of tubulin even after a prolonged incubation (Fig. 4). The CK2␣-and CK2␤-polymerized samples had no marked difference from the background tubulin polymerization, which was shown in the GST-incubated sample. Thus, only CK2 holoenzyme, but not each of the individual subunits, has the ability to induce microtubule assembly even though CK2␣ has shown microtubule binding activity.
CK2 has been known to catalyze phosphorylation of a neural isoform of ␤-tubulin and some of the MAPs, raising the possibility that it may affect microtubule dynamics through a kinase reaction (12,17). Although ATP was not present in the in vitro microtubule assembly assay, CK2 is capable of utilizing either ATP or GTP as the phosphate donor in its phosphorylating reactions (23). We designed an experiment to assess the role of CK2 kinase activity in microtubule assembly. A kinase-inactive holoenzyme of CK2, in which CK2␣ was replaced with the kinase-inactive mutant CK2␣⌲68A, was tested in the microtubule assembly assay. Fig. 5 shows that the kinase-inactive CK2 conferred the same microtubule polymerizing activity as the wild-type enzyme, indicating that the microtubule assembly entity of CK2 is independent of its kinase activity and phosphorylation of any microtubule proteins.
Microtubules from brains can be separated into two pools, namely "cold labile" and "cold stable," according to whether they are resistant to cold treatment for microtubule disassembly (24). It has been found that CK2 is enriched in the coldstable fraction of the microtubule preparation from rat brain (14). This observation, together with our findings that CK2 associates with microtubules to promote microtubule assembly, prompted us to explore the possibility that CK2 may contribute to the cold stability of microtubules. To test this likelihood, CK2-polymerized microtubules were incubated on ice, and the FIG. 1. Microtubule association of  CK2. A, direct interaction of CK2 and tubulin heterodimers. 5 g of GST, GST-CK2␣, or a complex of GST-CK2␣/His-CK2␤ were incubated with 2.5 g of purified tubulin. The GST fusion proteins were then retrieved using GSH beads, and bound proteins were analyzed by immunoblotting with antibodies recognizing ␣and ␤-tubulin. The protein input column was visualized by Coomassie Blue staining. B, CK2␤ does not interact physically with tubulin. 5 g of His-CK2␣ or His-CK2␤ were incubated with 2.5 g of purified tubulin. There was no His-CK2␣ or His-CK2␤ in the beads control sample. After pull-down with nickel-nitrilotriacetic acid beads, bound proteins were analyzed by anti-␤-tubulin immunoblotting. C, direct association of CK2 with microtubules. 1 g of GST, CK2␣, CK2␤, or CK2 holoenzyme was incubated with microtubules pre-assembled using taxol from 10 g of purified tubulin. After precipitation of the microtubules, proteins in the supernatant and the microtubule pellet were analyzed by immunoblotting using an antibody mixture recognizing GST, CK2␣, and CK2␤. turbidity change was monitored. As a comparison, tau-polymerized microtubules were treated under the same condition, given the fact that tau does not confer cold stability to microtubules (25). As expected, the tau-polymerized sample was depolymerized almost completely within a few minutes (Fig. 6). However, the turbidity of the CK2-polymerized sample was only marginally reduced even after a prolonged cold incubation (Fig. 6), indicating that CK2 functions to stabilize microtubules against cold-induced disassembly.
CK2 Stabilizes Microtubules in Vivo-To evaluate the role of CK2 in microtubule dynamics in vivo, we knocked down CK2␣/␣Ј in HeLa cells by gene silencing using a siRNA duplex derived from the human sequence of CK2␣/␣Ј (26,27). As shown by the CK2␣ immunoblot, the introduction of CK2␣/␣Ј siRNA into the cells led to a dramatic decrease of the CK2␣/␣Ј proteins to a minimal cellular level (Fig. 7A). To assess the knockdown effect on microtubules, the amount of cellular mi-crotubules (assembled insoluble tubulin polymers) was determined using the differential extraction method and immunoblotting (22). In addition, the integrity of the cellular microtubule network was examined by immunofluorescent staining and confocal microscopy. The knockdown of CK2␣/␣Ј significantly reduced the cellular content of microtubules (Fig.  7, A and B), suggesting CK2 as one of the factors in stabilizing microtubules in vivo. We further assessed microtubule stability using colchicine, which is a microtubule-disrupting agent. When colchicine was applied at a low concentration (0.2 M) onto the cells that were transfected with a scrambled siRNA sequence, most of the microtubule structure remained intact (Fig. 7, A and B). However, such a low dose of colchicine caused severe disruption of the microtubule structure in the CK2␣/␣Јdepleted cells where the microtubule networks were collapsing toward the perinuclear membrane (Fig. 7B); almost negligible amount of microtubules was extracted from these cells (Fig.   FIG. 2. Cellular localization of CK2␣ and CK2␤ to microtubules. A, COS-7 cells were immunostained for confocal microscopic analysis. Top row, double staining of CK2␣ and ␤-tubulin; bottom row, CK2␤ and ␤-tubulin. B, soluble tubulin heterodimers (free tubulin) and microtubules (polymerized tubulin) were differentially extracted from HeLa cells. Both fractions as well as the total cell lysate (TCL) were analyzed by immunoblotting using antibodies as indicated. The histogram shows the relative amounts of CK2␣ and CK2␤ in the free and polymerized tubulin fractions. These data are representative of three independent experiments. 7A). Apparently, the removal of CK2␣ had a strong effect on cellular microtubule architecture, rendering it very unstable. As a result, it was readily destructed by colchicine at a very low concentration.
To further substantiate the microtubule stabilizing function of CK2, we tested whether microtubule stability could be restored by expression of chicken CK2␣ in endogenous CK2␣/␣Јdepleted cells. As observed with the HeLa cells, knockdown of CK2␣/␣Ј in cultured human 293T fibroblasts using siRNA strongly destabilized the microtubule network, resulting in almost complete disruption of the microtubules by colchicine at 0.2 M (Fig. 7C). When chicken CK2␣ was expressed in the 293T cells in which endogenous CK2␣/␣Ј was knocked down, the cellular microtubules completely retained their integrity against the colchicine-induced disruption (Fig. 7C). More interestingly, when the expression was performed using the kinaseinactive mutant CK2␣K68A, it exhibited the same effect as wild-type CK2␣ in rescuing microtubules from colchicine treatment (Fig. 7C). These data demonstrate that CK2 is an important mediator of cellular microtubule stability and exerts its effect in a phosphorylation-independent manner. DISCUSSION Microtubules are a major cytoskeletal constituent in all eukaryotes. In living cells, the microtubule architecture is stabilized by structural MAPs, which associate with microtubules FIG. 3. Effect of CK2 on microtubule assembly. A, the turbidimetric assay of tubulin polymerization. Microtubule assembly from purified MAP-free tubulin was carried out in the presence of the CK2 holoenzyme at various concentrations (molar ratios to tubulin). The concentration of tubulin was constant in each assay at 2 mg/ml. B, histogram of the microtubule assembly at various amounts of CK2. The assembly assay was performed as described in A for 30 min. The data shown are representative of three separate experiments. C, fluorescent imaging of microtubules polymerized from a mixture of rhodaminelabeled and unlabeled tubulin (7:1). The tubulin concentration is 2 mg/ml, and the CK2 concentration is 62 g/ml. The arrows point to microtubule bundles in the CK2-polymerized sample.
FIG. 4. Microtubule assembly can be induced by the CK2 holoenzyme but not its individual subunits. GST, GST-CK2␣, and His-CK2␤ were applied as indicated at 0.1 mg/ml in the microtubule assembly assay. As a control, the CK2 holoenzyme reconstituted from the same amount of GST-CK2␣ and His-CK2␤ as described under "Experimental Procedures" was applied. Microtubule assembly was performed at 2 mg/ml tubulin as described under ''Experimental Procedures.'' FIG. 5. The kinase activity of CK2 is not required for its function to induce microtubule assembly. The wild-type CK2 enzyme (GST-CK2␣/His-CK2␤) and the kinase-inactive enzyme (GST-CK2␣K68A/His-CK2␤) were applied as indicated at 0.1 mg/ml in the microtubule assembly assay. GST-CK2␣K68A and GST were also tested at the same amount. Microtubule assembly was performed with 2 mg/ml tubulin as described under ''Experimental Procedures.'' FIG. 6. CK2 confers cold stability to microtubules. Microtubules were polymerized with 0.05 mg/ml CK2 or 0.16 mg/ml tau protein for 30 min at 35°C, where they attained similar turbidity measurements. The microtubule samples were then incubated on ice, and the turbidity measurement was begun. Absorbance was expressed as a percentage of the measurement when ice incubation was started. and promote in vitro microtubule assembly (28,29). The evidence presented here identifies CK2 as a structural MAP that mediates microtubule dynamics. We have conducted experiments showing that CK2 is localized to and co-extracted with microtubules. The in vitro binding assays demonstrate the direct interaction of CK2 with microtubules as well as tubulin heterodimers, and the binding affinity is comparable with that of known MAPs. Microtubule binding sequences are often found in MAPs as repeated sequence stretches rich in basic amino acids. Although the sequence of CK2␣ contains some basic regions, it is not found to have any typical microtubulebinding motif in CK2␣. Thus, the microtubule association of CK2␣ may suggest new microtubule-binding domains.
Structural MAPs such as MAP2 and tau are known to stimulate microtubule assembly from tubulin heterodimers. In our microtubule assembly assays, CK2 exhibited a potent activity of inducing microtubule assembly and bundling from purified tubulin. The physical association of CK2 to microtubules and tubulin heterodimers stimulates both the rate and the extent of microtubule growth. Although CK2␣ can bind microtubules, the microtubule assembling and stabilizing function is solely a property of the holoenzyme. In addition, CK2-polymerized microtubules display stability against cold treatment, suggesting that CK2 is a strong stabilizer of microtubules. Taken together with the observation that a substantial amount of CK2 exists in the cold-stable microtubules of rat brain (14), our findings suggest that CK2 is a new factor endowing the cold stability of microtubules. To date, the STOP proteins, double-cortin and BPAG1n3, are the only known MAPs that confer cold stability on microtubules (30 -34).
Structural MAPs are known to contribute to microtubule stability and distribution within cells (35). The finding of CK2 as a structural MAP stimulated our interest in evaluating the regulatory role of CK2 in vivo in microtubule cytoskeleton. The knockdown of CK2␣/␣Ј from cells has a strong destabilizing effect on the microtubule architecture. As a result, the microtubule network is very vulnerable and can be readily destroyed by colchicine insult at 0.2 M, whereas such a low concentration of colchicine does not have any significant effect on microtubules of cells with intact CK2. Thus, CK2 has an indispensable role in stabilizing cellular microtubules. This is substantiated by the introduction of chicken CK2␣ into the cells to compensate for the loss of endogenous CK2␣/␣Ј. The microtubule instability caused by the deficit of CK2␣/␣Ј can be rectified completely by the expression of chicken CK2␣, which assures that CK2 is a vital structural MAP conferring microtubule stability in vivo. It is noteworthy that the removal of CK2␣/␣Ј did not cause severe microtubule disruption in the cells, possibly because of the existence of multiple MAPs other than CK2 in the cells, to support the microtubule network.
CK2 is a Ser/Thr protein kinase with a broad substrate spectrum that includes MAP1B and a neural-specific isoform of amounts of microtubules extracted from the cells as compared with the control, which is the sample transfected with the scrambled siRNA sequence and treated without colchicine. The data are representative of three separate experiments. B, cells in the experiments described in A were fixed and stained with the ␤-tubulin antibody for confocal microscopic imaging. C, expression of the wild-type or the kinase-inactive mutant of chicken CK2␣ restored microtubule stability against colchicine treatment in CK2␣/␣Ј-depleted cells. Prior to treatment with colchicine (0.2 M), 293T cells were double transfected with siRNA of human CK2␣/␣Ј and one of the following expression constructs: chicken CK2␣, the kinase-inactive mutant of chicken CK2␣ (CK2␣K68A), or the empty vector. Expression of Myc-tagged chicken CK2␣ and CK2␣K68A was detected by anti-Myc immunoblotting of the cell lysates. Microtubules were extracted using the differential extraction method (see ''Experimental Procedures'') for anti-␤-tubulin immunoblotting. Representative results of three separate experiments are shown. FIG. 7. CK2 stabilizes microtubules in vivo. A, HeLa cells were introduced with siRNA of human CK2␣/␣Ј or a scrambled sequence. Knockdown of CK2␣ was monitored by anti-CK2␣ immunoblotting. The cells were subsequently treated with 0.2 M colchicine or its solvent. Tubulin in the form of polymers (microtubules) was extracted from the cells for ␤-tubulin immunoblotting. The histogram reflects the relative ␤-tubulin. We examined whether the kinase activity of CK2 is involved in the microtubule assembly stimulated by CK2. Our in vitro assays of microtubule assembly using the kinase-inactive mutant of CK2 indicate that the microtubule-assembling activity of CK2 is independent of its kinase activity. This was corroborated by the experiment of expressing the kinase-inactive mutant of chicken CK2␣ in the CK2␣/␣Ј-knock-down cells, which completely compensated for the lost of endogenous CK2␣/␣Ј, rendering the microtubules resistant to colchicine attack. With these results, it becomes clear that CK2 imparts a direct regulation of microtubule organization through its physical association with microtubules but not through any enzymatic action. As a multifunctional enzyme, CK2 has been thought to execute its functions through its phosphorylation of a wide range of substrates. The results presented here reveal a novel CK2 function that is dissociated from its intrinsic kinase property.
It has been proposed that CK2 plays an important role in the maintenance of cell morphology and polarity. Depletion of the catalytic subunits of CK2 in neuroblastoma cells using an antisense approach blocks neuritogenesis (27,36). Pertinent observations also came from yeasts, of which the temperaturesensitive mutants of CK2␣ and CK2␤ demonstrated their importance in cell morphogenesis (9,37,38). The function described here for CK2 in microtubule dynamics may provide a mechanistic explanation of its role in cell shape control.