HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Day, I. S. Articles by Reddy, A. S. N. Search for Related Content PubMed PubMed Citation Articles by Day, I. S. Articles by Reddy, A. S. N. Social Bookmarking                      What's this?

J Biol Chem, Vol. 275, Issue 18, 13737-13745, May 5, 2000

Interaction of a Kinesin-like Calmodulin-binding Protein with a Protein Kinase^*

Irene S. Day, Cindy Miller, Maxim Golovkin, and A. S. N. Reddy

From the Department of Biology and Program in Cell and Molecular Biology, Colorado State University, Fort Collins, Colorado 80523

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Kinesin-like calmodulin-binding protein (KCBP) is a novel member of the kinesin superfamily that is involved in cell division and trichome morphogenesis. KCBP is unique among all known kinesins in having a myosin tail homology-4 region in the N-terminal tail and a calmodulin-binding region following the motor domain. Calcium, through calmodulin, has been shown to negatively regulate the interaction of KCBP with microtubules. Here we have used the yeast two-hybrid system to identify the proteins that interact with the tail region of KCBP. A protein kinase (KCBP-interacting protein kinase (KIPK)) was found to interact specifically with the tail region of KCBP. KIPK is related to a group of protein kinases specific to plants that has an additional sequence between subdomains VII and VIII of the conserved C-terminal catalytic domain and an extensive N-terminal region. The catalytic domain alone of KIPK interacted weakly with the N-terminal KCBP protein but strongly with full-length KCBP, whereas the noncatalytic region did not interact with either protein. The interaction of KCBP with KIPK was confirmed using coprecipitation assays. Using bacterially expressed full-length and truncated proteins, we have shown that the catalytic domain is capable of phosphorylating itself. The association of KIPK with KCBP suggests regulation of KCBP or KCBP-associated proteins by phosphorylation and/or that KCBP is involved in targeting KIPK to its proper cellular location.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Kinesin and kinesin-like proteins (KLPs)¹ are microtubule-based transport motor proteins that perform a variety of important cellular and developmental functions such as controlling steady-state structural organization of cytoplasm; moving and positioning of intracellular macromolecules and organelles; and driving cell movement, cell division, and flagellar movement (1-3). The first kinesin was identified as a protein likely to be responsible for fast anterograde axonal transport (4). Kinesin is a tetramer consisting of two heavy chains (KHCs) and two light chains (KLCs) with the motor activity being in the heavy chains (5, 6) The heavy chain consists of three domains: a motor domain that contains conserved ATP and microtubule binding sites, a stalk domain that forms an -helical coiled-coil, and a globular tail that binds the light chains. KLPs show sequence similarity to the conserved motor domain of the KHC, which consists of 320-340 amino acids (7). In kinesin, the motor domain is in the amino-terminal region of the heavy chain, whereas in KLPs the motor domain is found in the amino-terminal, carboxyl-terminal, or middle region (3, 4). Outside the motor domain there is little homology between KLPs, but many contain an -helical region predicted to form a coiled-coil. The tail regions are not conserved with respect to kinesin heavy chain or among members of the same KLP subfamilies. It is widely believed that the tail regions of the KLPs may associate with accessory proteins important for function (7). The unique tail domains of the KLPs are thought to be responsible for the varied microtubule-based functions of the conserved motor domain (6). About 193 sequences containing the kinesin motor domain have been deposited into the data bases. Multiple KLPs have been identified in several organisms (Caenorhabditis elegans (19 KLPs), Drosophila melanogaster (14 KLPs), Homo sapiens (20 KLPs), Mus musculus (36 KLPs), and Arabidopsis thaliana (21 KLPs)). The challenge facing researchers now is to elucidate the function of these multiple microtubule motor proteins and their regulation.

Although it is widely believed that the tail region of KLPs performs its function(s) by interacting with other proteins, in most cases the proteins that interact with the tail region have not been identified. In the case of KHC, the tail region has been shown to interact with KLCs (8-10). Another protein called kinectin also interacts with KHC (11). Kinectin was isolated from chick embryo brain microsomes and was localized to the cytoplasmic face of the microsomes. A cDNA clone for kinectin was isolated from a chick embryo cDNA library by immunoscreening with mAbs to kinectin (12). Studies using antibodies to kinectin inhibited plus-end-directed vesicle motility by approximately 90% and reduced kinesin binding to vesicles (13). Using yeast two-hybrid methods, mitogen-activated protein kinase kinase kinases MLK2 and MLK3 were found to interact with the kinesin, KIF3 (14). Another yeast two-hybrid interaction study showed an interaction between an actin-based vesicle-transport motor, MyoVA, and a microtubule-based motor, KhcU (15).

Microtubule-based motility is known to be precisely regulated. This regulation could be due to the regulation of the motor protein, the microtubules or both components of the system (16). Studies on the regulation of kinesin and KLPs have focused mainly on phosphorylation of KHC, KLC, and kinesin-interacting proteins (7, 16-18). Phosphorylation of these proteins can affect their motor activity or their ability to interact with organelles or other proteins. Phosphorylation has been shown to be involved in cell cycle regulation of kinesin and KLPs (18, 19). The protein kinase or family of kinases responsible for phosphorylation has not been established in all cases. Two kinases have been shown to phosphorylate specific KLPs. A human KLP, HsEg5, has been shown to be phosphorylated by the cyclin-dependent protein kinase, p34^cdc2, specifically during mitosis (20). Phosphorylation controls the association of HsEg5 with the spindle apparatus. A murine protein kinase, PLK, which reaches its maximum activity at the onset of mitosis, appears to interact with and phosphorylate CHO1/mitotic kinesin-like protein 1 (21).

Recently a KLP, kinesin-like calmodulin-binding protein (KCBP), was isolated from Arabidopsis and other flowering plants using a protein-protein interaction-based screening with calmodulin as a probe (22-24). KCBP has two unique domains: a calmodulin-binding domain at the C terminus following the motor domain and a myosin tail homology domain in the tail (22, 23, 25-27). KCBP binds calmodulin in a calcium-dependent manner at physiological calcium concentration (22), and the binding of calmodulin inhibits KCBP from binding microtubules or dissociates the preformed KCBP-microtubule complex (28-30). Until recently, no other KLP had been known to bind calmodulin, and a KCBP homologue had not been found in animals including C. elegans whose genome has been completely sequenced. Using a PCR screen for kinesins in sea urchin, Rogers et al. (31) isolated a kinesin (kinesin-C) with a calmodulin-binding domain. Although it has sequence similarity in the motor and calmodulin-binding domains with KCBP, it does not contain the myosin tail homology-4 and talin-like regions present in the N-terminal portion of KCBP.

Many KLPs are involved in chromosome distribution and spindle function. Some of the functions include separation of microtubule organizing centers for spindle formation, spindle assembly, spindle elongation, and maintenance of spindle bipolarity and spindle pole integrity (1, 7, 32). Vernos and Karsenti (32, 33) proposed that kinetochore KLPs may tether the ends of growing and shrinking microtubules to kinetochores during chromosome movements. Immunolocalization studies indicated the involvement of KCBP in cell division (34, 35). It was also shown that KCBP is differentially active during the various phases of cell division in Tradescantia stamen hair cells (36). KCBP was found to be capable of bundling microtubules, suggesting the involvement of KCBP in establishing mitotic microtubule arrays during different stages of the cell cycle (37). Genetic studies have shown that KCBP is essential for trichome morphogenesis (38, 39). KCBP mutants (zwichel) had shorter and less branched trichomes (38). Three extragenic suppressors ("suppressor of zwichel-3"; suz1, suz2, and suz3) that rescued trichome branch number defect were isolated (39). Genetic studies have shown that SUZ2 may interact with not only ZWI (KCBP) but also with FURCAI, a protein involved in trichome branching. These suppressor studies suggest that KCBP interacts with several other proteins in the cell (27, 40).

Because the tail region of KLPs is implicated in interacting with other proteins as either cargo or regulator and because the genetic evidence suggests that KCBP interacts with other proteins (39), we used the yeast two-hybrid screen to isolate proteins that interact with KCBP. Here we report that the N-terminal region of KCBP interacts with a protein kinase, termed KCBP-interacting protein kinase (KIPK). It has homology to a group of protein kinases that appears to be unique to plants. We have confirmed the interaction in vitro and have demonstrated autophosphorylation of KIPK.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Plant Materials-- A. thaliana L. Heynh. ecotype Columbia was grown at 22 °C on a mixture of peat/perlite/vermiculite (1:1:1) under continuous light. Leaves, stems, and flowers were collected from 5-6-week-old plants. Roots were grown in liquid cultures as described earlier (41). Suspension-cultured cells from the same ecotype were grown in Murashige and Skogg salts supplemented with vitamins, 3% sucrose, and 0.5 µg/ml 2,4-dichlorophenoxyacetic acid.

Plasmid Construction-- The pAS1CYH2/N-terminal KCBP plasmid was constructed by cloning a SacI/NotI fragment (amino acids 12-1261) of Arabidopsis KCBP into pET32b (30). This plasmid was then cut with NcoI, releasing a 2.8-kilobase fragment (amino acids 12-924), which was ligated into pAS1CYH2 digested with NcoI. Full-length Arabidopsis KCBP/pAS1CYH2 was made by cutting the pAS1CYH2/N-terminal KCBP with AvrII and SalI and dropping the fragment. Full-length Arabidopsis KCBP in pET28 (29) was also cut with AvrII and SalI. The fragment was isolated and ligated into the pAS1CYH2 AvrII/SalI-cut plasmid.

The expression plasmid for KIPK was constructed by excising KIPK from pACT-KIPK with XhoI and ligating it into pET32a (Novagen) also cut with XhoI. Preparation of KCBP N-terminal construct in pET28 and its expression was described previously (29).

Subclones of KIPK were made using PCR-generated fragments. Primers were designed to amplify an N-terminal fragment from amino acids 1-347 and a C-terminal fragment from amino acids 323-744. A BamHI site was included in the 5' primers, and a XhoI site was included in the 3' primers. Primers used were KIPK1S (5'-CGGGATCCGATGCCTGTAAACGATAAACC-3'), KIPK1A (5'-CCGCTCGAGAGTGTCGTAAACCCAAAGAGCC-3'), KIPK2S (5'-CGGGATCCCATGTCGATGGATGTAGATGGG-3'), and KIPK2A (5'-CCGCTCGAGACTCAAGATGATCGCCTATGGC-3'). pET32b and the PCR-generated fragments were cut with BamHI and XhoI, ligated, and used to transform Escherichia coli BL21 (DE3). Subclones were cut from pET32b using NcoI and XhoI and ligated into NcoI/XhoI-cut pACT2. All clones were sequenced to confirm the accuracy of the PCR product and in-frame ligation.

A motor domain subclone of KCBP was constructed by PCR amplification of a C-terminal fragment coding for amino acids 860-1261, including the motor domain, calmodulin binding domain, and a small portion of the coiled-coil region (23). The primers included a site for BspHI, which when cut has ends compatible with NcoI. pAS1CYH2 was digested with NcoI, and the fragment was digested with BspHI; they were ligated and used to transform E. coli (DH5)

Yeast Transformation and -Galactosidase Filter Assays-- Competent cells were prepared by inoculating several colonies of yeast strain Y190 into YPD (12 g/liter peptone, 10 g/liter yeast extract, 2% dextrose) plus cycloheximide (10 µg/ml) or colonies of transformed Y190 in appropriate SD (6.7 g/liter yeast nitrogen base without amino acids (Difco), 2% dextrose, and 1× dropout solution minus Trp or minus Leu (CLONTECH)), vortexing vigorously, and incubating at 30 °C with shaking overnight. Cultures were then transferred to 50 ml of YPD/cycloheximide or SD lacking Trp or Leu and incubated at 30 °C overnight with shaking. Cultures were diluted with 250 ml of YPD and incubated at 30 °C with shaking at 250 rpm for 3 h. Cells were centrifuged at 1,000 × g for 5 min at room temperature, and the pellet was resuspended in 50 ml of sterile water and centrifuged at 1,000 × g for 5 min at room temperature. The pellet was then resuspended in 1.5 ml of freshly prepared, sterile 1× TE/LiAc (10 mM Tris-HCl, 1 mM EDTA, 100 mM lithium acetate, pH 7.5).

For library transformation of Y190-KCBP, 5 µg of an Arabidopsis cDNA library, 0.1 mg of salmon sperm carrier DNA, and 0.1 ml of Y190-KCBP-competent cells were added to 16 separate tubes; the contents were mixed well. For transformation with plasmid DNA from constructs, 0.1 µg of plasmid DNA was mixed with 0.1 mg of salmon sperm carrier DNA, and 0.1 ml of competent cells was added to each tube. For double transformations, 0.1 µg of each construct was added to 0.1 mg of salmon sperm carrier DNA, and 0.1 ml of competent cells was added and transformed as described (CLONTECH manual). Approximately 100 µl of transformed cells was spread on SD plates containing 20 g/liter agar and Trp dropout for pAS1CYH2 transformations, Leu dropout for pACT transformations, and Trp, Leu, His dropout plus 25 mM 3-aminotriazole for transformations with the library or both pACT and pAS1CYH2 plasmids. Plates were incubated 3-8 days at 30 °C.

Colonies were either patched to new plates or used as grown for assays. Sterile filters (Whatman number 5) were soaked in Z buffer/5-bromo-4-chloro-3-indolyl--D-galactopyranoside solution (prepared according to the CLONTECH manual) and placed over the surface of the plates containing transformants. Filters were carefully lifted, transferred (colonies facing up) into liquid nitrogen, completely submerged for 10 s, and then allowed to thaw at room temperature. Filters were then placed on top of filters freshly soaked in Z buffer/5-bromo-4-chloro-3-indolyl--D-galactopyranoside solution and incubated at room temperature until blue color appeared.

Plasmid and Protein Isolation from Yeast-- Yeast transformants were grown in 5 ml of the appropriate SD, and plasmids were isolated using a kit from Zymo Research according to the instructions provided by the manufacturer. The culture was incubated at 30 °C with shaking until saturated. For protein isolation, overnight cultures of transformant yeast colonies were used to inoculate 100 ml of appropriate SD, which were grown overnight at 30 °C. Cells were spun at 5,000 rpm for 5 min and resuspended in a small portion of the medium. The cells were transferred to flat-bottom, O-ring screw cap microcentrifuge tubes and spun. After aspiration of the medium, the cells were resuspended in 40 µl of Laemmli buffer and 400 µl of acid-washed beads and vortexed two times for 1 min each. An additional 460 µl of Laemmli buffer was added, and tubes were transferred to boiling water for 3 min. Following chilling on ice, the tubes were spun at high speed for 5 min at 4 °C, and the supernatant was transferred to a clean tube and stored at 20 °C.

Sequencing-- Sequences on either side of the multiple cloning site of pACT were used to design primers for sequencing. Primers were 5'-TACCACTACAATGGATGATG-3' (pACT5') and 5'-GTTGAAGTGAACTTGCGGGG-3', (pACT3'). The clone was sequenced by the dideoxynucleotide chain termination method using double-stranded plasmid DNA as a template. Sequence data from the initial sequencing was used to construct subsequent primers to sequence the clone from end to end in both directions. Subclones were sequenced to verify in frame insertion of sequence.

DNA Hybridization and Reverse Transcription PCR-- An Arabidopsis genomic DNA blot was prepared and probed with KIPK as described earlier (42). Total RNA from different organs and tissues was isolated and purified on a CsCl cushion (43). Poly(A)⁺ mRNA was isolated by an oligo(dT)-cellulose column. First strand cDNA was synthesized using Promega's Reverse Transcription System (44). The first strand cDNA was used in a PCR using primers specific to the C-terminal half of KIPK (KIPK2S and KIPK2A). The PCR product was separated electrophoretically on an 0.8% gel, transferred to a nylon membrane, and probed with ³²P-labeled KIPK sequence.

Protein Induction, Isolation, and Detection-- BL21(DE3) cells transformed with either pET28b-Arabidopsis KCBP or pET32a-Arabidopsis KIPK were cultured to an A₆₀₀ of approximately 0.6. Expression of protein was induced by the addition of isopropyl-1-thio--D-galactopyranoside to a concentration of 0.2 mM. Induced cells were grown overnight at room temperature with constant shaking. Bacteria were harvested and resuspended in ice-cold 1× T7 tag bind/wash buffer (10× T7 buffer: 42.9 mM Na₂HPO₄, 14.7 mM KH₂PO₄, 27 mM KCl, 1.37M NaCl, 1% Tween 20, 0.02% sodium azide, pH 7.3) for KCBP protein or ice-cold 1× S tag bind/wash buffer (10× S tag buffer: 200 mM Tris-HCl, pH 7.5, 1.5 mM NaCl, 1% Triton X-100) for KIPK protein. Lysozyme (200 µg/ml) and Complete Protease Inhibitor (Roche Molecular Biochemicals) were added to each sample, and samples were placed on ice for 1 h and sonicated on ice five times for 15 s. Lysate was centrifuged at 18,000 rpm for 30 min. Protein was stored at 4 °C.

In Vitro Interaction of KCBP and KIPK-- Crude extract (2 ml) of KCBP protein was added to 150 µl of T7 tag antibody-agarose (Novagen) and incubated with shaking at room temperature for 30 min. The agarose beads were washed three times with 1× T7 tag bind/wash buffer, and 2 ml of KIPK crude protein was added to the agarose beads and incubated as above. KIPK crude protein alone was also added to T7 tag antibody-agarose beads that were not incubated with KCBP. Both sets of agarose beads were washed three times. Following the final wash, 75 µl of 1× sample loading buffer was added to the beads. Twenty-five µl was loaded on each of three polyacrylamide gels and electrophoresed. One gel was stained, and two were blotted and probed with either T7 tag antibody or S tag protein.

Phosphorylation-- KIPK (full-length and subclone 2-catalytic domain) and amino-terminal KCBP proteins were expressed and isolated as above. The proteins were bound by S tag-agarose beads or T7 tag-agarose beads, respectively (Novagen). For one set of phosphorylation experiments, full-length KIPK protein bound to S tag beads was mixed with amino-terminal KCBP bound to T7 tag beads, subclone 1 (noncatalytic domain) protein bound to S tag beads, or water. Phosphorylation assays were performed in a buffer (20 mM Tris, pH 7.5, 6 mM MgCl₂) containing 5 mM -mercaptoethanol, 1 mM sodium orthovanadate, 0.1 mM ATP, and [-³²P]ATP at room temperature for 1 h. The reaction was stopped by adding protein sample buffer, and reaction mixtures were electrophoresed on acrylamide gels, blotted to a membrane, and exposed to a phosphor imaging screen. A second set of phosphorylation reactions was performed essentially as above except that the catalytic domain of KIPK bound to S tag beads was used.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Screening of an Arabidopsis Two-hybrid Library for KCBP-interacting Proteins-- To identify KCBP-interacting proteins, we employed the yeast two-hybrid system (45, 46). Previous studies have shown that the C-terminal region of KCBP interacts with microtubules, tubulin subunits, and calmodulin (22, 25, 28-30, 47). To avoid isolation of tubulin and calmodulin, the N-terminal end of Arabidopsis KCBP comprising the "tail" and the coiled-coil region but not the motor or calmodulin binding domain was used as the "bait" in the screen. An A. thaliana cDNA library consisting of clones in pACT as fusions to the activation domain of Gal4 was screened using the amino-terminal domain (amino acids 12-806) of KCBP in pAS1CYH2 as a fusion to the DNA binding domain of Gal4.

Approximately 10⁶ transformants were plated on selection plates (synthetic complete minus His, Leu, and Trp). Transformants growing on these plates were screened for -galactosidase activity using a filter assay (48). Colonies that were His⁺ and blue were selected for further analysis. Possible interacting clones were sequenced at the 5'- and 3'-ends and compared with the NCBI data base. One of the clones showed homology to a type of protein kinase found, to date, only in plants (see below). Because of the sequence similarities between the isolated clone and protein kinases, we designated this protein as KIPK. The pACT-KIPK plasmid was isolated and used to transform Y190 by itself. Yeast colonies expressing both the pAS1CYH2 amino-terminal KCBP and the pACT-KIPK plasmid grew on selection medium and showed positive -galactosidase activity, while Y190 colonies with neither plasmid or with either just the amino-terminal KCBP or KIPK plasmid did not (Fig. 1A).

View larger version (39K): [in this window] [in a new window]   Fig. 1.   Interaction of KCBP with KIPK. A, yeast containing no plasmids (Y190), the pAS1CYH2/N-terminal KCBP plasmid (Y190 + KCBP), the pACT-KIPK plasmid (Y190 + KIPK), or both plasmids (Y190 + KCBP + KIPK) were replica-plated on YPD; SC minus tryptophan (W); SC minus leucine (L); and SC minus tryptophan, leucine, and histidine (3). A -galactosidase assay (A) was carried out using a YPD plate replicate. B, the noncatalytic domain (subclone 1) and catalytic domain (subclone 2) were cloned into pACT2 and used in a two-hybrid interaction with pAS1CYH2/N-terminal KCBP. Yeast (Y190) cells containing the N-terminal region of KCBP and either the noncatalytic domain (Sub 1) or the catalytic domain (Sub 2) were grown on SC minus leucine and tryptophan (L,W) and assayed for -galactosidase activity (A). C, Y190 yeast containing pAS1CYH2/full-length KCBP were transformed with the noncatalytic region (Sub 1), the catalytic domain (Sub 2) or the full-length KIPK (Kipk) clone. Transformed yeast grew on SC minus leucine, minus tryptophan (L,W). A -galactosidase assay (A) was carried out using duplicate plates.

Interaction of Different Regions of KCPB and KIPK-- KIPK was isolated as a clone that interacted with an amino-terminal subclone of KCBP. The KIPK clone contained the kinase catalytic domain at the C terminus and an extensive noncatalytic N-terminal region (see below). To determine if certain regions of KIPK were responsible for the interaction, we constructed two subclones of KIPK (one containing the noncatalytic amino-terminal region and the second one with the catalytic carboxyl-terminal region) and tested their interaction with the N-terminal region of KCBP using the yeast two-hybrid assay. The N-terminal region of KIPK (subclone 1) showed no interaction with KCBP, whereas the catalytic domain (subclone 2) showed weak interaction (Fig. 1B). The blue color from the galactosidase assay took greater than 24 h to develop. To determine if KIPK interacts with the motor domain of KCBP, we constructed a motor domain subclone in pAS1CYH2 and used it to transform Y190 containing the pACT-KIPK full-length clone. No interaction was observed between KIPK and the motor domain of KCBP (data not shown). A full-length KCPB clone was constructed in pAS1CYH2 and used in an interaction assay with full-length KIPK and both subdomains. Both full-length KIPK and the catalytic domain showed strong interaction with full-length KCPB, but the N-terminal region showed no interaction (Fig. 1C). The presence of the plasmids in the transformants was confirmed by PCR using the primers specific to KCBP and KIPK. All transformants showed the presence of both plasmids (data not shown).

KIPK Belongs to a Group of Protein Kinases That Are Specific to Plants-- The KIPK sequence contains an open reading frame that encodes a protein of 744 amino acids with a calculated molecular mass of 86 kDa (Fig. 2). The putative start site is the first methionine in the clone. The size of the predicted protein is larger than the protein kinases of this type. Hence, it is unlikely that the native protein is larger than 744 amino acids. The KIPK protein contains all conserved domains and amino acids found in other protein kinases (9, 49-51). The catalytic domain starts at amino acid 348 (Fig. 2) and contains the 11 conserved subdomains typical of protein kinases. Highly conserved features include the glycine-rich stretch GCGDIG³⁵⁵, which forms the nucleotide binding site along with Lys³⁷⁷ and Glu³⁹⁶; the catalytic loop, which contains Asp⁴⁷³, Lys⁴⁷⁵, and Asn⁴⁷⁸ and which also participates in ATP binding; and the substrate binding site comprising APE⁵⁹⁷, Gly⁶¹⁶, and Arg⁶⁷³. Two sequences diagnostic of serine/threonine kinases, RDLKPEN⁴⁷² and GTHEYLAPE⁵⁹¹, are also present in KIPK.

View larger version (98K): [in this window] [in a new window]   Fig. 2.   Nucleotide and deduced amino acid sequence of KIPK. The putative initiation codon is in boldface type. The 11 protein kinase domains (49) are enclosed in brackets and are in italic type. Highly conserved amino acids are in boldface type. The catalytic domain that is diagnostic of serine/threonine kinases is underlined. Nucleotides are numbered on the left, and amino acids are numbered on the right.

Two features of the KIPK sequence are found in several other plant protein kinases. One is a change of the highly conserved tripeptide DFG to DFD⁴⁹¹. The other is the presence of an additional sequence between subdomains VII and VIII, and this insertion is found only in plant protein kinases (51-54). Fig. 3 shows an alignment of KIPK with two other protein kinases: PVPK-1, a protein kinase from Phaseolus vulgaris L., and STPK1, a protein kinase from Solanum tuberosum, that showed very high sequence similarity. The unique sequence between subdomains VII and VIII contains repeats of the motif CXX(X)P (where "XX(X)" indicates any 2 or 3 amino acids). In KIPK, there is one repeat at the beginning of the insertion sequence, and later in the insertion sequence three other copies in tandem are interrupted by a repeat lacking the proline. In STPK1 and PVPK-1, there are six repeats, one at the beginning and five in tandem later in the sequence (51). A BLAST search using the insertion sequence identified 18 different plant protein kinases with significant homology to that region: four other protein kinases in Arabidopsis; five in rice; three in tomato; and one each in bean, ice plant, maize, potato, pea, and soybean. The amino-terminal portion of the kinase shows more limited homology than the catalytic domain (Fig. 3).

View larger version (82K): [in this window] [in a new window]   Fig. 3.   Comparison of KIPK with two other plant kinases. The protein kinase domains are overlined, and the unique sequence between domains VII and VIII is double overlined. Identical amino acids between two or more sequences are indicated by reverse lettering (white letters on black background). The numbers on the left are the amino acid residue numbers for each protein. PVPK-1 is a protein kinase from P. vulgaris L. (GenBankTM accession no. J04555). STPK1 is a protein kinase from S. tuberosum (GenBankTM accession no. X90990).

Gene Copy Number and Expression of KIPK-- The gene copy number was determined using DNA gel blot analysis. As shown in Fig. 4A, a single hybridizing band was detected with HindIII and EcoRI, indicating that it is coded by a single gene. To study the expression of KIPK in various organs and tissues, RNA was isolated from fruits, leaves, roots, flowers, and cell suspension cultures. Reverse transcription followed by PCR using primers specific to the carboxyl-terminal portion of KIPK was used to amplify the message. An expected fragment of approximately 1.2 kilobases was amplified from the first strand cDNA (Fig. 4B). KIPK was expressed in all tissue examined but was more highly expressed in fruit, roots, and flowers. As a control, genomic DNA was also used for PCR. The fragment from the genomic DNA was slightly larger, indicating the presence of at least one intron. ³²P-Labeled KIPK hybridized to the PCR-generated fragments, verifying the KIPK product. As an internal control, primers were also included for the constitutively expressed cyclophilin gene (55).

View larger version (53K): [in this window] [in a new window]   Fig. 4.   Gene copy number and expression of KIPK. A, DNA gel blot. Arabidopsis genomic DNA was digested with HindIII or EcoRI, separated electrophoretically, and transferred to a nylon membrane. The blot was hybridized with 32P-labeled KIPK and exposed to x-ray film. B, expression of KIPK. First strand cDNA was made from poly(A)+ mRNA and used in PCR using primers specific to the C-terminal half of KIPK. As an internal control, primers to the constitutively expressed cyclophilin gene were included in the reaction. The PCR product was separated electrophoretically, transferred to a nylon membrane, and probed with 32P-labeled KIPK sequence. An arrow points to the expected KIPK fragment. An arrowhead points to the cyclophilin fragment. P, plasmid; G, genomic DNA; F, fruit, L, leaves; R, root; Fl, flower; S, cell suspension. Size markers are indicated on the left in kilobases.

Expression of KCBP and KIPK Protein in Yeast and Bacteria and in Vitro Interaction of KCBP and KIPK-- To demonstrate the presence of KCBP and KIPK protein in transformed yeast, protein was isolated from yeast containing the pAS1CYH2-KCBP and pACT-KIPK plasmids. Proteins blots were probed with GAL4 AD monoclonal antibody (pACT) or GAL4 DNA-BD monoclonal antibody (pAS1CYH2) (CLONTECH). The gene fusion product from amino-terminal KCBP was expected be approximately 109 kDa (87 kDa + 22 kDa from the binding domain). The gene fusion product from KIPK was expected to be 105 kDa (86 kDa + 19 kDa from the activation domain). Proteins of the expected molecular masses were detected for both proteins (Fig. 5A). These results further confirm the interaction data obtained with the two-hybrid interaction assay.

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Expression of KCBP and KIPK in yeast and bacteria. A, protein was isolated from yeast strain Y190 harboring the pAS1CYH2-KCBP and pACT-KIPK plasmids and Y190 with no plasmids. The blot on the left was probed with GAL4 binding domain antibody, and the blot on the right was probed with GAL4 activation domain antibody. B, protein was isolated from bacteria containing pET28-KCBP or pET32-KIPK. The blot on the left was probed with S tag protein, which binds to the S tag expressed as a fusion with KIPK in pET32. The blot on the right was probed with T7 tag antibody, which binds to the T7 tag expressed as a fusion to KCBP in pET28. The two lower bands are degraded forms of KCBP. Numbers on the left indicate the size of molecular mass markers in kDa. A thin arrow points to KIPK protein; a thick arrow points to KCBP protein.

Both amino-terminal KCBP and KIPK were cloned into bacterial expression vectors. KCBP was expressed using pET28a (Novagen) as a fusion to T7 tag, and KIPK was expressed using pET32a (Novagen) as a fusion to S tag. Protein blots containing the induced proteins were probed with T7 tag antibody for KCBP or S tag protein for KIPK. The gene fusion product from amino-terminal KCBP was expected be approximately 92 kDa (87 kDa + 5 kDa from the T7 tag). The gene fusion product from KIPK was expected to be 105 kDa (86 kDa + 19 kDa from the S tag). Both proteins of expected molecular mass were detected in the supernatant fraction (Fig. 5B).

To demonstrate the interaction of KIPK and KCBP proteins in vitro, a co-precipitation experiment was performed. T7 tag antibody-agarose beads (Novagen) were incubated with proteins isolated from bacteria expressing KCBP as a T7 tag fusion. The beads were washed and then incubated with protein isolated from bacteria expressing KIPK as a fusion to S tag. The beads were washed again, and protein was released using protein sample loading buffer. Proteins were separated on a gel and blotted to membranes and probed with either T7 tag antibody to identify KCBP or S tag protein to identify KIPK. As shown in Fig. 6, KIPK coprecipitated with N-terminal KCBP bound to the beads. As a control, crude extract containing KIPK protein was incubated with T7 tag antibody-agarose beads. No KIPK protein was precipitated by the beads alone (Fig. 6).

View larger version (81K): [in this window] [in a new window]   Fig. 6.   In vitro interaction of KCBP and KIPK. Agarose beads conjugated to T7 tag antibody were used to precipitate KCBP expressed as a T7 tag fusion. KIPK expressed as a fusion to S tag was precipitated from crude bacterial extract by incubation with the KCBP-bound T7 tag antibody beads. As a control, an equal amount of T7 tag antibody beads with no KCBP was also incubated with the crude extract containing the KIPK protein. The beads were washed, and equal portions were loaded on three gels. One gel was stained, and the other two were blotted to nitrocellulose membranes. One blot was probed with T7 tag antibody to detect KCBP, and one was probed with S tag protein to detect KIPK. Lane 1 in each case is from beads incubated with KCBP followed by KIPK crude extract. Lane 2 in each case is from beads incubated in KIPK crude extract alone. An arrow points to the band corresponding to the expected molecular weight of KIPK. The band in T7 tag, lane 2, and the corresponding band in lane 1 are from recognition of the T7 tag antibody from the beads by the secondary antibody. Numbers on the left indicate the size of molecular mass markers in kDa.

KIPK Undergoes Autophosphorylation-- Bacterially expressed KIPK protein was used in a series of phosphorylation experiments to determine if it could autophosphorylate and to determine if it phophorylates KCBP protein. The KIPK protein sequence contains several potential phosphorylation sites, one of which (Ser⁵⁸⁸) has been characterized in other protein kinases (51). The KCBP protein sequence also has a number of possible phosphorylation sites. Full-length KIPK protein and the catalytic domain (subclone 2) protein either bound to beads or free in solution were used in phosphorylation experiments. Both KIPK full-length protein and the catalytic domain alone autophosphorylated (Fig. 7, A, lanes 1 and 3; B, lanes 1 and 3). Duplicate phosphorylation experiments carried out with no magnesium were negative (data not shown). KCBP N-terminal protein bound to T7 tag beads was incubated with the full-length KIPK protein (Fig. 7A, lane 4) or the catalytic domain of KIPK protein (Fig. 7B, lane 4). Neither the full-length KIPK nor catalytic domain of KIPK phosphorylated KCBP under the conditions used (Fig. 7A, lane 4; B, lane 4). The N-terminal region of KIPK bound to S tag beads was also incubated with full-length KIPK (Fig. 7A, lane 5) or catalytic domain of KIPK (Fig. 7B, lane 5) bound to S tag beads to test if KIPK phosphorylates the N-terminal noncatalytic region. As shown in Fig. 7, phosphorylation of the N-terminal region was not observed under these conditions (Fig. 7A, lane 5; B lane 5). The presence of KCBP and different regions of KIPK proteins in the assay was verified using the appropriate protein probes (Fig. 7, middle panels).

View larger version (55K): [in this window] [in a new window]   Fig. 7.   Autophosphorylation of KIPK. Full-length KIPK protein or the KIPK catalytic domain protein (subclone 2) was incubated in phosphorylation buffer containing [-32P]ATP electrophoresed, blotted, and exposed to a phosphor imaging screen. The blot was then probed with anti-T7 tag and/or S tag protein conjugated to alkaline phosphatase to show the presence of protein in the phosphorylation reaction. The thin arrows point to the full-length KIPK. The arrowhead points to KCBP protein. The thick arrow points to the noncatalytic domain (subclone 1) protein, and asterisks denote catalytic domain protein (subclone 2). A, lane 1, crude KIPK protein; lane 2, crude BL21 protein; lane 3, KIPK protein bound to S tag beads alone; lane 4, KIPK protein bound to S tag beads incubated with KCBP amino-terminal protein bound to T7 tag beads; lane 5, KIPK protein bound to S tag beads incubated with subclone 1 bound to S tag beads. B, subclone 2 protein. Lane 1, crude subclone 2 protein; lane 2, crude BL21 protein; lane 3, subclone 2 protein bound to S tag beads alone; lane 4, subclone 2 protein bound to S tag beads incubated with KCBP amino-terminal protein bound to T7 tag beads; lane 5, subclone 2 protein bound to S tag beads incubated with subclone 1 bound to S tag beads. The numbers on the left represent kDa.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

A clone (KIPK) was isolated from a yeast two-hybrid screen using the amino-terminal region of the kinesin KCBP. Based on sequence similarities and the presence of specific diagnostic features, the protein product of the isolated clone was identified as a serine/threonine protein kinase. It contains the 11 subdomains (12 if domain VI is split into VIa and VIb) (49) characteristic of protein kinases. Protein kinases are a superfamily of proteins that have been divided into five main groups by Hanks and Hunter (56). A classification of plant protein kinases done by Stone and Walker (57) groups plant kinases also into these five families. Of these five groups, the "AGC" group consists of the cyclic nucleotide-dependent family (protein kinase A and protein kinase G), the protein kinase C family, and the ribosomal S6 kinase family. A common theme of the AGC group is regulation by second messenger (i.e. cAMP, cGMP, diacylglycerol, and Ca²⁺). Although protein kinases with properties similar to protein kinase C have been isolated (58-61) there is a lack of evidence for any of the major AGC kinases in plants. However, a group of plant protein kinases including the first plant protein kinase cloned, PVPK-1 (54), showed some similarity to protein kinase A (57). However, PVPK-1 and other related protein kinases, unlike other members of the AGC group, are characterized by a 70-90-amino acid insert between subdomains VII and VIII of the catalytic domain. This insert may be involved in regulation of the kinases. It is thought that the CXX(X)P repeats in this insertion may contribute to protein folding and thus to binding specificity of the insert (51). Also, the insertion is in the activation region of protein kinases, which is defined as the region spanning DFG (DFD⁴⁹¹ in KIPK) in subdomain VII and APE⁵⁹⁷ of subdomain VIII. This segment contains a residue (or residues) that are phosphorylated, and the phosphorylation plays a crucial role in the regulation of some protein kinases (9). KIPK showed highest similarity to five of the protein kinases in this group, a maize protein kinase (52), potato STPK1 (51), rice Gl1A and bean PVPK-1 (54), and another Arabidopsis protein kinase, ATPK5 (53). In a recent review of plant serine/threonine kinases, Hardie (62), using 89 A. thaliana protein kinases in a phylogenetic study, classified plant protein kinases into 12 groups. One of these groups included kinases homologous to KIPK. In discussing the functions of the different protein kinases, he makes the point that no function has been established for any kinase in this group. Knowing that KIPK interacts with KCBP should lead to the function of this kinase, which may then give insight to the functions of similar kinases.

The amino-terminal regions show less homology to other similar proteins and are thought to have some regulatory or protein interaction role (4, 63). The amino-terminal region of KIPK is highly rich in Ser and Thr (23%) over 347 amino acids. Other protein kinases in this plant-specific group are similarly rich in Ser and Thr, 20-30%. It is thought that like animal protein kinases C these upstream regions could represent regulatory domains (53).

Protein kinases are involved in many aspects of cellular regulation and metabolism. Protein phosphorylation has been implicated in response to many signals, including light, pathogen invasion, hormones, temperature stress, and nutrient deprivation (57). STPK1 expression is increased after infection with Phytophthora infestans, which indicates that STPK1 may participate in the signal transduction pathway triggered by the pathogen (51). The maize protein kinase isolated by Bierman et al. (52) was shown to be phylogenetically related to protein kinases involved in transduction of extracellular signals via regulation of their activity by second messengers (cAMP, cGMP, calcium, and diacylglycerol). A group of protein kinases in pea that show homology to these AGC group VIII kinases have the DFG modification and a small insert between subdomains VII and VIII (64, 65). Two of the genes, PsPK3 and PsPK5, were shown to be photoregulated. G11A and PVPK-1 are also most like protein kinases that transduce extracellular signals, but their noncatalytic sequences exhibit no homology to regulatory domains of other protein kinases (54). Since KIPK is most closely related to these plant kinases and the AGC animal protein kinases, it too may be involved in transduction of extracellular messages.

In this study, we have demonstrated that KIPK interacts with the kinesin KCBP. This conclusion is supported first by the yeast two-hybrid interaction results. Full-length KIPK interacted with the amino-terminal portion and full-length KCBP but did not interact with the motor domain of KCBP. The interaction of KCBP and KIPK is further supported by the results of the in vitro co-precipitation results. KCBP protein, bacterially expressed as a fusion to T7 tag, was purified on T7 tag antibody-agarose beads and used to co-precipitate bacterially expressed KIPK. KIPK was selectively precipitated from crude protein extract in the presence of KCBP. T7 tag antibody-agarose beads on the other hand did not bind to KIPK when incubated with crude protein extract from bacteria expressing KIPK.

Regulation of kinesin and KLPs by phosphorylation may be by direct phosphorylation of the KLP or by phosphorylation of a protein that interacts with the KLP. Evidence of both types of phosphorylation has been found (66, 67). Phosphorylation has been correlated with vesicle transport in several systems (21, 68, 69), and phosphorylation of KLPs during the cell cycle has also been shown (20, 21, 70-73). Both the full-length KIPK and the catalytic domain of KIPK show autophosphorylation. The N-terminal region of KCBP, on the other hand, was not phosphorylated by either full-length or catalytic domain KIPK protein under the conditions tested in our study. KIPK may need to be phosphorylated by another protein kinase or need some other form of post-translational modification in order to phosphorylate KCBP. The other possibility is that KIPK phosphorylates the full-length or native KCBP. We could not use the full-length KCBP in phosphorylation studies, since the expression of full-length cDNA always yielded proteolyzed N-terminal and C-terminal fragments.

Alternatively, KCBP may not be a target for the kinase itself but may be involved in the targeting of the kinase to a specific location or to a KCBP-associated protein that may be the substrate for the kinase. A calmodulin-binding class III myosin called NinaC is thought to transport calmodulin and control the subcellular distribution of calmodulin (40, 74, 75). Phosphorylation of kinesin-associated proteins with concomitant stimulation of kinesin activity has been shown (76). Three proteins that co-purified with kinesin were hyperphosphorylated under conditions of enhanced activity. Later, one of the proteins was identified as KLC (77). A 100-kDa kinase and a 150-kDa type 1 phosphatase regulated the phosphorylation of this KLC. The kinase and phosphatase both co-purify with the KHC, suggesting that kinesin exists in a large complex capable of self-regulation. MLK Ser/Thr kinase related to the mitogen-activated protein kinase kinase kinase family used in a yeast two-hybrid screen interacted with a KLP related to KIF3 and with a KIF3-interacting protein, KAP3A, the putative targeting component of KIF3 (78). KIF3 localized to structures along microtubules at the same location as MLK2 and phosphorylation-activated c-Jun N-terminal kinase, a stress-regulated protein kinase (14). Another complex of proteins containing a KLP and a protein kinase was isolated in Drosophila (79). Co-immunopreciptation experiments revealed a complex containing a protein-serine/threonine kinase (FU), a transcription factor (cubitus interruptus), and a KLP (COS2). A signaling protein HH (hedgehog gene) stimulated phosphorylation of COS2 and FU, but the phosphorylation of neither FU nor COS2 was by FU. Even when the catalytic domain of FU was mutated, both proteins were phosphorylated (79).

This study, together with previous biochemical (22, 25, 28-30) and genetic studies (38, 39), strongly suggests that KCBP interacts with several proteins and may function as a multiprotein complex. It is interesting that KLPs seem to be present in complexes of proteins that often contain protein kinases. Work in the future to elucidate the substrate(s) for KIPK will help to answer some of the questions raised by the presence of these complexes in cells.

	FOOTNOTES

^* This work was supported by National Science Foundation Grant MCB-9630782 (to A. S. N. R.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EMBL Data Bank with accession number(s) AF236104.

To whom correspondence should be addressed. Tel.: 970-491-5773; Fax: 970-491-0649; E-mail: reddy@lamar.colostate.edu.

	ABBREVIATIONS

The abbreviations used are: KLP, kinesin-like protein; KHC, kinesin heavy chain; KCBP, kinesin-like calmodulin-binding protein; KIPK, KCBP-interacting protein kinase; KLC, kinesin light chain; YPD, yeast extract-peptone-dextrose medium; SD, synthetic dropout medium; SC, synthetic complete medium.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES