identification enzymatically active ca2+/calmodulin-dependent proteon kinase in centrosomes of...

S.F. Pietromonaco, et al. Blood Cells, Molecules, & Diseases (1995) 21 (4) Feb. 28: 34-41

Abbreviations used: CaM, calmodulin; ECL, enhanced chemiluminescence; EDTA, ethylene diamine-N,N,N',N'-tetraacetic acid; EGTA,ethylene glycol-bis-(2-aminoethyl ether)-N,N,N',N'-tetraacetic acid; HEPES, N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid; PBS,phosphate-buffered saline; PIPES, piperazine-N,N'-bis[2-ethanesulfonicacid]; SDG, sucrose density gradient; SDS-PAGE, sodium dodecylsulfate polyacrylamide gel electrophoresis; TBS, Tris-buffered saline; TCA, trichloroacetic acid; W7, N-(6-aminohexyl)-5-chloro-1-naphthalene-sulfonamide.

Reprint requests to: Salvatore F. Pietromonaco, Ph.D., University of New Mexico Cancer Center, 900 Camino de Salud, Albuquerque, NM, USA1

87131. phone (505) 277-8532, fax (505) 277-2841.

34 Published by the Blood Cells Foundation

IDENTIFICATION OF ENZYMATICALLY ACTIVECa /CALMODULIN-DEPENDENT PROTEIN KINASE2+

IN CENTROSOMES OF HEMOPOIETIC CELLS

Salvatore F. Pietromonaco , Gustavo A. Seluja, Laurence Elias1

Department of Medicine and the Cancer Center, University of New Mexico School of Medicine,Albuquerque, New Mexico, USA 87131.Submitted on 02/13/95(Communicated by P.W. Majerus, M.D., 02/21/95)

ABSTRACT. In this study, we report the identification of enzymatically active, multifunctionalcalcium/calmodulin-dependent protein kinase in centrosomes of FDCP1 cells using subcellular fractionation andimmunofluorescence techniques. Centrosomes were isolated from detergent lysates of FDCP cells by sucrose densitygradient centrifugation and contain tubulin (M = 58 kDa) and centrin (M = 20 kDa) by immunoblotting. Analysisr rof these fractions with anti-calcium/calmodulin kinase II antibody revealed the presence of the 52 kDa and 56 kDadoublet corresponding to the " and the $/$' subunits of the enzyme complex. In vitro kinase reactions with isolatedcentrosomes and in the presence of calcium and calmodulin results in the phosphorylation of several centrosomalproteins.Keywords: hemopoeitic cells, protein kinases, centrosomes, calcium, calmodulin

INTRODUCTION

The multifunctional calcium/calmodulin- transient rise in intracellular Ca activatesdependent protein kinase (Ca /CaM kinase II) is Ca /CaM kinase II, suggesting that the enzyme2+

an important effector of Ca/calmodulin mediates the action of growth factors (3).2+

signalling in cells. The enzyme complex is a Transient increases in Ca are required for G -Smultimeric structure (550-650 kDa) consisting of and G -M phase progression (4). In afour (", $, (, *) closely related gene products of hemopoietic system in which buoyant tonsillar B56-60 kDa that become autophosphorylated in cells are transitting the cell cycle, Ca /CaMresponse to increases in intracellular Ca . kinase II is more active than in resting, dense2+

Ca /CaM kinase II is most abundant in brain, tonsillar cells (5). In a hairy cell leukemic cell2+

where it functions to regulate neurotransmittor line, which contains unusually high levels ofrelease and in the induction of long-term intracytoplasmic Ca , high levels of Ca /CaMpotentiation (1). kinase II activity result in hyperphosphorylation of

It is well documented that Ca is required for CD20, a membrane protein involved in B cell2+

cell cycle progression and mediates its actions via activation (6).CaM, the primary intracellular receptor (2). In A role for Ca /CaM kinase II in cell cycle

quiescent rat embryo fibroblasts that arestimulated with serum to re-enter the cell cycle, a

2+

2+

2+1

2

2+

2+ 2+

2+

Blood Cells, Molecules, and Diseases (1995) 21(4) Feb. 28: 34-41 S.F. Pietromonaco, et al.

35

progression is further supported byimmunofluorescence studies localizing the enzyme(7) as well as CaM (8) in the mitotic apparatusand centrosomes of mammalian cells. A putativesubstrate of Ca /CaM kinase II has been2+

localized to the mitotic apparatus and becomesphosphorylated in response to Ca and CaM (9).2+

Furthermore, a 20 kDa Ca-binding protein2+

named centrin or caltractin has extensivestructural homology with CaM and is acomponent of the pericentriolar material incentrosomes from divergent species (10, 11).

Mouse leukemic FDCP1 cells are a growthfactor-dependent cell line in which cell cycleprogression can be studied. We becameinterested in studying Ca /CaM kinase II as part2+

of our studies on the regulation of centrosomefunction by signal transduction pathways inhemopoietic cells. No studies to date havedirectly demonstrated the presence ofenzymatically active Ca /CaM kinase II in2+

centrosomes. In this report, we demonstrate thatin FDCP1 cells Ca /CaM kinase II is associated2+

with centrosomes using both immunofluorescencelocalization and subcellular fractionation.Furthermore, we demonstrate that thecentrosomal Ca /CaM kinase II can be activated2+

by Ca and CaM.2+

METHODS Subcellular fractions consisting of Triton-soluble

Antibodies. Affinity-purified antibody against lysed in 1% (v/v) NP-40. The lysate wasCa /CaM kinase II (RU-16) was obtained from centrifuged at 500 x g to pellet nuclei, and the2+

A. Czernik, Rockefeller University, New York, post-nuclear supernatant was spun at 100,000 x gNY. RU-16 was raised to a peptide correspond- for 60 min. The resulting supernatant is referreding to residues 501-531 of the $ subunit that is to as the Triton-soluble fraction, and the pellet ashighly conserved in all known subunits and the Triton-insoluble fraction. Centrosomes wererecognizes the " and $/$' subunits (12). isolated from FDCP1 cells essentially as describedAntiserum 26/14-1 against the pericentriolar for human KE37 T-lymphoblastic cells (13). Cellsprotein centrin was obtained from J. Salisbury, (2 x 10 ) were treated with 2 x 10 M nocodazoleMayo Clinic, Rochester, MN. This antibody was and 1 µg/ml cytochalasin B for 90 min at 37EC.raised against a trpE-centrin fusion protein Cells were harvested by centrifugation at 400 x gexpressed from a cDNA encoding for 5 min, washed twice with PBS, and once withChlamydomonas reinhardtii centrin and reacts 0.1x PBS containing 8% (w/v) sucrose. Cellswith mammalian centrin (10). Anti-" tubulin was were lysed in 0.5% (v/v) NP-40, 1mM Tris-HCl,obtained from Amersham (Arlington Heights, IL). pH 8.0, 0.01 M 2-mercaptoethanol, 0.5 mM

Immunofluorescence. FDCP1 cells (1 x 10 )6

were pelleted from culture medium and washedtwice with PBS. The cells were resuspended in0.05 ml PBS and applied to poly-L-lysine-coatedglass microscope slides. The cells were allowedto adhere for 15 min at 37EC, rinsed in PBS andthen fixed in acetone for 3 min. The cells wererehydrated with PBS containing 1% BSA andincubated for 30 min at 37EC with rabbit anti-Ca /CaM kinase II (2 µg/ml) and mouse anti-2+

centrin (1:500) antibodies diluted in the samebuffer. The slides were rinsed in PBS to removeunbound antibodies and incubated with goat anti-rabbit IgG (Fab') -FITC (Zymed, South San2Francisco, CA) and sheep anti-mouse IgG (Fab') -2Texas Red (Cappel, West Chester, PA) eachdiluted 1:400 in PBS/1% BSA. After additionalwashes in PBS, the slides were mounted in 2.5mg/ml DABCO [1,4-diazobicyclo-(2,2,2)-octane]in 90 % glycerol in PBS, pH 8.5 to reduce fadingof fluorescence. The cells were observed using aLeitz microscope equipped for epifluorescencemicroscopy and photographed using T-Max 400film.

Cell fractionation and centrosome isolation.FDCP1 cells were grown in RPMI mediumcontaining 5-10% WEHI-conditioned medium (asource of IL-3), and 10% fetal calf serum.

and insoluble fractions were prepared from cells

9 -7


36

MgCl , 1 mM phenylmethanesulfonyl fluoride, (v/v) Tween 20 and incubated with 1:50002and 10 µg/ml each of aprotinin, leupeptin, and dilution in blocking buffer of either goat anti-pepstatin A, and centrifuged at 1200 x g for 10rabbit or anti-mouse IgG-peroxidase conjugatemin to pellet nuclei. The post-nuclear supernatant (Amersham) for 1 hr at room temperature. Thewas adjusted to 0.01 M PIPES, pH 7.2 and blots were washed as described above and bounddigested with 600 units of DNase I. The treatedantibodies were visualized by ECL detectionsupernatant was loaded onto a discontinuous (Amersham). The molecular weight markers usedsucrose gradient consisting of 70% (w/v) sucrose for electrophoresis include bovine serum albuminat the bottom, followed by 50% and 40% layers (66 kDa), ovalbumin (43 kDa), carbonicmade in 0.01 M PIPES, pH 7.2, 0.1% (v/v) Triton anhydrase (30 kDa), soybean trypsin inhibitor (22X-100, 0.01 M 2-mercaptoethanol. The tubes kDa), and lysozyme (14.4 kDa).were centrifuged in a SW28 rotor at 100,000 x gfor 60 min and 0.5 ml fractions were collectedfrom the bottom of the tubes. Centrosomes wereconcentrated by diluting each fraction up to 5 mlwith 0.01 M PIPES, pH 7.2 and centrifuged at50,000 x g in a SW 50.1 rotor for 60 min. Thepelleted centrosomes were resuspended in 0.05 ml0.01 M PIPES pH 6.8 and stored at -70EC.Protein determinations were performed usingcolloidal gold (Quantigold, Diversified Biotech,Newton Centre, MA) to quantitate submicrogramamounts of protein.

SDS-PAGE and immunoblotting. SDGfractions were fractionated by SDS-PAGE (14)and transferred to Immobilon membranes(Millipore, Bedford, MA) using the modificationof Otter (15). Gels were transferred in 2x transferbuffer [50 mM Tris-base, 384 mM glycine, 0.01% (w/v) SDS, 20 % (v/v) methanol] at 80 mA for60 min, followed by 280 mA for 30 min. Aftertransfer, the membranes were stained with 0.2 %Ponceau S in 1 % acetic acid to visualize themolecular weight standards, and destained withTBS (0.02 M Tris-HCl, pH 7.5, 0.15 M NaCl).The blots were then treated with 0.2% (w/v)glutaraldehyde in PBS for 5 min and washed inTBS. The blots were incubated with 5% (w/v)non-fat dry milk in TBS containing 0.1% Tween-20 (blocking buffer) for 2 hrs at ambienttemperature. Antibodies against "-tubulin,Ca /CaM kinase II, and centrin were diluted in2+

blocking buffer containing 1 mM CaCl and2incubated overnight at 4EC. The blots wererinsed 4 x 30 min each with TBS containing 0.1%

In vitro phosphorylation. Endogenous proteinphosphorylation was performed in a 10 µlreaction volume containing centrosomes (100 ngprotein) in 25 mM HEPES, pH 7.4, 10 mMMgCl , 1 mM EGTA, 1.5 mM Ca , 30 µg/ml2

2+

CaM (Sigma, St. Louis, MO) and 5 µCi [(-P]ATP (3000 Ci/mmol, Amersham) at 37EC for32

20 min (12). The reactions were terminated bythe addition of 2x Laemmli sample buffer, boiledfor 3 min, and analyzed by SDS-PAGE/autoradiography.

Figure 1. Enrichment ofCa /CaM kinase II in the Triton-2+

insoluble fraction of mouseleukemic FDCP1 cells. Aliquots(1 µg each) of the Triton-soluble(A) and -insoluble (B) fractionswere separated by SDS-PAGE in12.5% gels and immunoblottedwith anti-Ca /CaM kinase II2+

antibody (1 µg/ml). The positionsof the " (52 kDa) and $/$' (56kDa) subunits of the kinase areindicated by arrows.

RESULTS

In order to determine whether Ca /CaM2+

kinase II is present in FDCP cells and isassociated with the cytoskeleton, cells were lysedin 0.5% NP-40 and the post-nuclear supernatant


37

was centrifuged to obtain Triton-soluble and - kinase II could be detected in centrosomes ofinsoluble fractions. As shown in Figure 1, FDCP1 cells using immunofluorescenceCa /CaM kinase II antibody recognizes both the microscopy. Acetone-fixed FDCP1 cells were2+

" (52 kDa) and $/$' (56 kDa) subunits in the incubated with affinity-purified rabbit anti-Triton-soluble (lane A) and -insoluble (lane B) Ca /CaM kinase II antibody in combination withfractions. The Triton-insoluble fraction, which mouse monoclonal anti-centrin as a marker forrepresents 14% of the total protein in the post- centrosomes. As shown in Figure 2, anti-nuclear supernatant, contains -2-fold greater " Ca /CaM kinase II antibody is concentrated in aand -4-fold greater $/$' subunits than the soluble highly localized, perinuclear region (A), whichfraction, which contains 86% of the total protein. co-localizes with centrosomes, as revealed byAccounting for this difference in protein levels in anti-centrin (B). In prometaphase/metaphase cells,the Triton-soluble and insoluble fractions, the " where centrosomes occupy opposite poles of theand the $/$' subunits are enriched in the cell, anti-Ca /CaM kinase II and anti-centrin co-detergent-insoluble fraction by factors of 12 and localize in the duplicated centrosomes (Fig. 2 A24 fold, respectively. The antibody also detects a and B, arrows). Cells incubated with the80 kDa protein in the Triton-soluble fraction secondary antibodies alone show no labeling of(Figure 1, lane A) that has been found in rat aortic centrosomes or any other cellular structure (Fig.vascular smooth muscle cells (16) and in rat heart 2 C and D).(17).

We next determined whether Ca /CaM2+

2+

2+

2+

Figure 2. Immunofluorescence localization of Ca /CaM2+

kinase II in centrosomes. FDCP1 cells were attached topoly-L-lysine glass slides and fixed in acetone. Cells wereincubated with affinity-purified rabbit anti-Ca /CaM2+

kinase II and mouse monoclonal anti-centrin antibodies (A,B) or buffer alone (C, D). The slides were then incubatedwith goat anti-rabbit IgG (Fab') -FITC and sheep anti-2mouse IgG (Fab') -Texas Red. Bar = 10 µm.2

We next determined whether Ca /CaM (13). Cells were lysed in 0.5% NP-40 and the2+

kinase II observed in centrosomes by immuno- post-nuclear supernatant was fractionated in afluorescence could be identified in centrosomes sucrose step gradient to isolate fractions enrichedpurified from FDCP1 cells. Centrosomes were in centrosomes. In order to identify fractionsisolated from FDCP1 cells treated with containing centrosomes, aliquots of each fractionnocodazole and cytochalasin B using low were subjected to immunoblotting with antibodyconcentrations of each drug, which are sufficient to centrin, a pericentriolar protein foundto depolymerize the microtubule-microfilament exclusively in centrosomes (10). These fractions

system, but maintain the integrity of centrosomes


38

were also immunoblotted with "-tubulin antibody 5000-fold purification. The purification attainedto detect tubulin in the centriolar microtubules. is comparable to estimates made by otherAs shown in Figure 3, both tubulin at 58 kDa investigators (13, 19).(panel A) and centrin at 20 kDa (panel B) are Immunoblot analysis of the gradient fractionsdetected in fractions 4-8 corresponding to 50% containing both tubulin and centrin (Fig. 3, A andsucrose. Fractions 10 and 11 containing soluble B) with anti-Ca /CaM kinase II antibody revealsprotein at the top of the gradient have little the " and $/$' subunits at 52 kDa and 56 kDa,detectable centrin, indicating that centrosomes are respectively, indicating that Ca /CaM kinase IIenriched in the 50% sucrose region of theco-purifies with centrosome-containing fractionsgradient. Similar results were reported using (Fig. 3, panel C). Interestingly, the centrosomeKE37 cells (13). When centrosomes were isolated fractions contain a greater proportion of " thanin buffer containing 1 mM EDTA (18), centrin $/$' subunits (Fig. 3, panel C). Furthermore,was not detected in fractions that would contain immunodetection of these subunits in centrosomecentrosomes (50% sucrose). The purity of the fractions corresponds to the antigen detected bycentrosome preparation was assessed byimmunofluorescence. It is of interest to note thatimmunofluorescence with anti-tubulin antibody, the centrosome fractions do not contain the 80which indicated that all the tubulin was associated kDa protein found in the Triton-soluble fractionwith centrosomes (data not shown). of total cell lysates, making it unlikely that this isFurthermore, the yield of total centrosome protein the antigen detected by immunofluorescence.(2 µg, fractions 4-7) from 2 x 10 cells represents9

0.02 % of total cell protein and corresponds to a

2+

2+

Figure 3. Isolation of centrosomes from FDCP1 cells by sucrosedensity gradient fractionation. A post-nuclear supernatant from FDCPcells was fractionated in a sucrose step gradient as described in"Methods". Aliquots (100 ng protein) of each fraction were analyzedby SDS-PAGE in 12.5% gels and immunoblotted with anti-"-tubulin(1:1000 dilution, panel A), anti-centrin (1:2000 dilution, panel B), andanti-Ca /CaM kinase II (1 µg/ml, panel C) antibodies.2+

The centrosome preparation corresponding to CaM results in phosphorylation of severalfraction 6 that contains the peak of centrin and proteins, the most predominant having molecularCa /CaM kinase II immunoreactivity was masses of 47, 52, 56, 80, and >200 kDa (lane 3).2+

assayed for endogenous Ca /CaM kinase II The addition of the CaM antagonist W7, which2+

enzymatic activity. As shown in Figure 4, basal binds CaM (20), abolishes the phosphorylation ofphosphorylation without added Ca or CaM is these proteins (lane 4). The inclusion of 1 mM2+

two-fold lower (lane 1) than in the presence of each of EDTA and EGTA, which would reduceCa and CaM (lane 3). The addition of Ca and [Ca ] to submicromolar amounts while leaving2+ 2+ 2+


39

-8.5 mM Mg in the reaction, abolished Ca and CaM was observed in centrosomes2+

endogenous phosphorylation (lane 2). The isolated in the absence of EDTA, suggesting thatproteins migrating at 52 and 56 kDa most likely sufficient Ca is bound to centrosomes, perhapsrepresent the " and $/$' subunits, respectively, of via CaM or centrin/caltractin, to activate theCa /CaM kinase that are detected by kinase. Furthermore, the results presented in this2+

immunoblotting. study underscore the importance of previous

Figure 4. Protein kinase assay of gradient-purified centrosomes.Autoradiogram of in vitro phosphorylation reactions performed as describedin "Methods". Code at top shows the composition of each reaction. Theconcentrations of EDTA and EGTA were 1 mM each, and W7 was 250 mM.The positions the " and $/$' subunits of Ca /CaM kinase II are indicated by2+

the lines. The dried gel was exposed to film for 4 hours.

DISCUSSION

This study demonstrates that centrosomespurified from FDCP1 cells contain Ca /CaM2+

kinase II. This observation is supported by the insitu immunofluorescence localization ofCa /CaM kinase II in centrosomes of FDCP12+

cells. Our findings are the first to demonstratethat centrosome-associated Ca /CaM kinase II is2+

activated by Ca and CaM, resulting in the2+

autophosphorylation of the " and $/$' subunits aswell as several other unidentified centrosomalproteins. The specificity of the phosphorylationwas assessed using the CaM antagonist W7 andthe Ca -specific chelator EGTA, both of which2+

inhibited the phosphorylation. Surprisingly, abasal level of phosphorylation without added

2+

2+

studies localizing CaM in centrosomes and thespindle apparatus (8).

The centrosome is a microtubule-organizingcenter, which nucleates polymerization of "/$tubulin to form a microtubular network that isrequired for cell division, the maintenance of cellshape, and supports intracellular vesicle andorganelle transport (21). Microtubule-associatedproteins such as MAP-2 and tau are required formicrotubule assembly and are phosphorylated byCa /CaM kinase II (22, 23). Our results2+

demonstrating the presence of primarily the " andsome $/$' subunits of Ca /CaM kinase II in the2+

Triton-insoluble fraction of FDCP1 cells isconsistent with previous studies in neuronal cellsdescribing the association of the enzyme with thecytoskeleton (24). The finding that centrosomescontain primarily the " isozymic form ofCa /CaM kinase II, while cells contain the $/$'2+

subunits as well, raises the possibility that a subsetof total intracellular " polypeptide is targeted tothe centrosome. Selective targeting of Ca /CaM2+

kinase II polypeptides has been demonstrated forthe * polypeptide, which acquires a nuclearlocalization signal by alternative splicing (25).However, there is at present a complete lack ofinformation on how intracellular proteins aretargeted to the centrosome.

The results presented here have importantimplications for the role of centrosomalCa /CaM kinase II. It is now generally accepted2+

that protein phosphorylation is essential for thecell cycle. Several protein kinases, includingp34 (26), cAMP-dependent protein kinasecdc2

(27), and Ca /CaM kinase II (28) are active2+

during mitosis and are associated with thecentrosome and/or the spindle apparatus. Withrespect to Ca /CaM kinase II, recent studies2+

suggest that only a transient activation of thekinase is necessary for G -M progression (28).2


40

The expression of a constitutively active, Ca -2+

independent form of the enzyme in mammaliancells resulted in cell cycle arrest in G , suggesting2that G -M progression is regulated by a2phosphorylation and dephosphorylation cycle.The mechanism by which this signalling pathwayaffects centrosome function and/or replication ispresently not known. However, several studieshave shown that epidermal growth factor (EGF) dependent phosphorylation of a 62 kDa protein induces

is required for centrosome separation during thecell cycle (29, 30). Extracellular EGF is capableof activating an intracellular Ca /CaM pathway2+

linking the extracellular milieu with thecentrosome replicative cycle and the cell cycle.We are currently studying the activation ofCa /CaM kinase II in the centrosome by growth2+

factors and the cell cycle-dependent phos-phorylation of Ca /CaM kinase II substrates in2+

the centrosome.

ACKNOWLEDGEMENTS

We thank Dr. A. Czernik (Laboratory of Neuroscience,Rockefeller University, New York, NY) for supplying anti-Ca /CaM kinase II. We also thank Dr. J. Salisbury2+

(Department of Biochemistry and Molecular Biology, MayoClinic, Rochester, MN) for providing centrin antibody andcommunicating to us the divalent metal ion-dependentretention of centrin in centrosomes before publication. Thisresearch was partially supported by grant # IRG 192 from theAmerican Cancer Society, grant # RO1 CA42520 from NIH,and the University of New Mexico Cancer Center.

REFERENCES

1. Hanson, P.I., and Schulman, H. Neuronal Ca/calmodulin-2+

dependent protein kinases. Annu. Rev. Biochem. 61: 559-601,1992.

2. Means, A.R., and Dedman, J.R. Calmodulin-an intracellularreceptor. Nature 285: 73-77, 1980.

3. Ohta, Y., Ohba, T., Fukunaga, K., and Miyamoto, E. Serumand growth factors rapidly elicit phosphorylation of theCa /calmodulin-dependent protein kinase II in intact quiescent2+

rat 3Y1 cells. J. Biol. Chem. 263: 11540-11547, 1988.4. Whitaker, M., and Patel, R. Calcium and cell cycle control.

Development 108: 525-542, 1990.5. Valentine, M.A., Leprince, C., Genot, E., Meier, K.E., and

Clark, E.A. In Protein kinases in Blood Cell Function.Huang,C. and Sha'afi, R., eds., CRC Press Inc., London, 1-20,1992.

6. Genot, E.M., Meier, K.E., Licciardi, K.A. et al.Phosphorylation of CD20 in cells from a hairy cell leukemia

cell line: Evidence for involvement of calcium/calmodulin-dependent protein linase II. J. Immunol. 151: 71-82, 1993.

7. Ohta, Y., Ohba, T., and Miyamoto, E. Ca /calmodulin-2+

dependent protein kinase II: localization in the interphasenucleus and the mitotic apparatus of mammalian cells. Proc.Natl. Acad. Sci USA 87: 5341-5345, 1990.

8. Welsh, M.J., Dedman, J.R., Brinkley, B.R., and Means, A.R.Tubulin and calmodulin: effects of microtubule andmicrofilament inhibitors on localization in the mitoticapparatus. J. Cell Biol. 81: 624-634, 1979.

9. Dinsmore, J.H., and Sloboda, R.D. Calcium and calmodulin

microtubule depolymerization in sea urchin mitoticapparatuses. Cell 53: 769-780, 1988.

10. Baron, A.T., Greenwood, T.M., Bazinet, C.W., and Salisbury,J.L. Centrin is a component of the pericentriolar lattice. Biol.Cell 76: 383-388, 1992.

11. Lee, V.D., and Huang, B. Molecular cloning and centrosomallocalization of human caltractin. Proc. Natl. Acad. Sci. (USA)90: 11039-11043, 1993.

12. Matovicik, L.M., Haimowitz, B., Goldenring, J.R., Czernik,A.J., and Gorelick, F.S. Distribution of calcium/calmodulin-dependent protein kinase II in rat ileal enterocytes. Am. J.Physiol. 264: C1029-C1036, 1993.

13. Bornens, M., Paintrand, M., Berges, J., Marty, M.C., andKarsenti, E. Structural and chemical characterization ofisolated centrosomes. Cell Motil. Cytoskel. 8: 238-249, 1987.

14. Laemmli, U.K. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature 227: 680-685, 1970.

15. Otter, T., King, S.M., and Witman, G. A two-step procedurefor efficient electrotransfer of both high-molecular weight(>400,000) and low-molcular weight (<20,000) proteins.Anal. Biochem. 162: 370-377, 1987.

16. Schworer, C.M., Rothblum L.I., Thekkumkara, T.J., andSinger, H.A. Identification of novel isoforms of the * subunitof Ca /calmodulin-dependent protein kinase II. J. Biol. Chem.2+

268: 14443-14449, 1993.17. Edman, C.F. and Schulman, H. Identification and

characterization of * -CaM kinase and * -CaM kinase fromB Crat heart, two new multifunctional Ca2 /calmodulin-dependent+

protein kinase isoforms. Biochem. Biophys. Acta 1221: 89-101, 1994.

18. Sanders, M.A., and Salisbury, J.L. Centrin plays an essentialrole in microtubule severing during flagellar excision inChlamydomonas reinhardtii. J. Cell Biol. 124: 795-805, 1994.

19. Mitchison, T.J., and Kirschner, M.W. Isolation of mammaliancentrosomes. Meth. Enzymol. 134: 261-268, 1986.

20. Tanaka, T., Ohmura, T., Yamakado, T., and Hidaka, H. Twotypes of calcium-dependent protein phosphorylationsmodulated by calmodulin antagonists: naphthalenesulfonamidederivatives. Mol. Pharmacol. 22: 408-415, 1982.

21. Vale, R.D. Intracellular transport using microtubule motors.Ann. Rev. Cell Biol. 3: 347-378, 1987.

22. Matus, A. In Microtubules. Hyams, J.S., and Lloyd C.W., eds.,Wiley-Liss, New York, 155-166, 1994.

23. Goedert, M., Jakes, R., Spillantini, M.G., and Crowther, R.A.In Microtubules. Hyams, J.S., and Lloyd, C.W., eds., Wiley-Liss, New York, 183-200, 1994.

24. Larson, R.E., Espindola, F.S., and Espreafico, E.M.Calmodulin-binding proteins and calcium/calmodulin regulated


41

enzyme activities associated with brain actomyosin. J. Golgi complex and with centrosomes. Cell 41: 1039-1051,Neurochem. 54: 1288-1294, 1990. 1985.

25. Srinivasan, M., Edman, C.F., Schulman, H. Alternative 28. Planas-Silva, M.D., and Means, A.R. Expression of asplicing introduces a nuclear localization signal that targets consititutive form of calcium/calmodulin dependent proteinmultifunctional CaM kinase to the nucleus. J. Cell Biol. 126: kinase II leads to cell cycle arrest in G2. EMBO J. 11: 507-839-852, 1994. 517, 1992.

26. Bailly, E., Pines, J., Hunter, T., and Bornens, M. Cytoplasmic 29. Sherline, P. and Mascardo, RN. Epidermal growth factoraccumulation of cyclin B1 in human cells: association with a induces rapid centrosomal separation in HeLa and 3T3 cells.detergent-resistant compartment and with the centrosome. J. J. Cell Biol. 93: 507-511, 1982.Cell Sci. 101: 529-545, 1992. 30. Sherline, P. and Mascardo, RN. Epidermal growth factor-

27. Nigg, E.A., Schafer, G., Hilz, H., and Eppenberger, H.M. induced centrosomal separation: mechanism and relationshipCyclic AMP-dependent protein kinase II is associated with the to mitogenesis. J. Cell Biol. 95: 316-322, 1982.

identification enzymatically active ca2+/calmodulin-dependent proteon kinase in centrosomes of...

Documents