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

J. Biol. Chem., Vol. 277, Issue 6, 4206-4214, February 8, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/6/4206    most recent M104595200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Reddy, A. S. N. Articles by Harnly, M. J. Search for Related Content PubMed PubMed Citation Articles by Reddy, A. S. N. Articles by Harnly, M. J. Social Bookmarking                      What's this?

Isolation and Characterization of a Novel Calmodulin-binding Protein from Potato^*

Anireddy S. N. Reddy, Irene S. Day, S. B. Narasimhulu, Farida Safadi, Vaka S. Reddy, Maxim Golovkin, and Melissa J. Harnly

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

Received for publication, May 20, 2001, and in revised form, September 24, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Tuberization in potato is controlled by hormonal and environmental signals. Ca²⁺, an important intracellular messenger, and calmodulin (CaM), one of the primary Ca²⁺ sensors, have been implicated in controlling diverse cellular processes in plants including tuberization. The regulation of cellular processes by CaM involves its interaction with other proteins. To understand the role of Ca²⁺/CaM in tuberization, we have screened an expression library prepared from developing tubers with biotinylated CaM. This screening resulted in isolation of a cDNA encoding a novel CaM-binding protein (potato calmodulin-binding protein (PCBP)). Ca²⁺-dependent binding of the cDNA-encoded protein to CaM is confirmed by ³⁵S-labeled CaM. The full-length cDNA is 5 kb long and encodes a protein of 1309 amino acids. The deduced amino acid sequence showed significant similarity with a hypothetical protein from another plant, Arabidopsis. However, no homologs of PCBP are found in nonplant systems, suggesting that it is likely to be specific to plants. Using truncated versions of the protein and a synthetic peptide in CaM binding assays we mapped the CaM-binding region to a 20-amino acid stretch (residues 1216-1237). The bacterially expressed protein containing the CaM-binding domain interacted with three CaM isoforms (CaM2, CaM4, and CaM6). PCBP is encoded by a single gene and is expressed differentially in the tissues tested. The expression of CaM, PCBP, and another CaM-binding protein is similar in different tissues and organs. The predicted protein contained seven putative nuclear localization signals and several strong PEST motifs. Fusion of the N-terminal region of the protein containing six of the seven nuclear localization signals to the reporter gene -glucuronidase targeted the reporter gene to the nucleus, suggesting a nuclear role for PCBP.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In plants, hormonal and environmental signals control many aspects of growth and development. In recent years, Ca²⁺ has emerged as an important messenger in controlling many normal growth and developmental processes and in eliciting responses to biotic and abiotic stresses (1-5). A tip-focused Ca²⁺ gradient has been shown to be important for the growth of root hairs and pollen tubes (6-8). Furthermore, diverse signals including hormones, light, touch, wind, cold, drought, oxidative stress, nodulation factors, and fungal elicitors have been shown to cause a transient increase in free Ca²⁺ in the cytoplasm (2, 5, 9, 10). Ca²⁺ sensors then transmit the changes in cytosolic Ca²⁺ level to cellular metabolic processes. Plants have at least four major families of Ca²⁺ sensors: (i) calmodulin (CaM)¹; (ii) CaM-like and other EF-hand containing Ca²⁺-binding proteins; (iii) Ca²⁺-regulated protein kinases; and (iv) Ca²⁺-binding proteins without EF-hand motifs (2). An EF-hand is a helix-loop-helix that binds Ca²⁺.

CaM, a highly conserved well characterized Ca²⁺ sensor in eukaryotes, is a small protein with four EF-hands that each bind Ca²⁺ (11). The EF-hands are paired in two globular domains connected by a central helix (12). In plants, CaM is encoded by multiple genes that have been shown to be expressed differentially (11, 13-17). In addition, different CaM isoforms are known to interact differentially with the target proteins (18-22). Binding of Ca²⁺ to CaM results in a conformational change that exposes hydrophobic pockets that can then interact with target proteins (23, 24). CaM mediates intracellular Ca²⁺ by regulating the activity/function of diverse proteins in a Ca²⁺-dependent manner. The physiological response that is elicited by the elevated cytosolic Ca²⁺ signal is derived, to some extent, from the expression patterns and activities of the proteins regulated by CaM (11). In a few cases CaM can interact with its target proteins in the absence of Ca²⁺ (2).

Identification and characterization of calmodulin-binding proteins (CBPs) is critical to the understanding of Ca²⁺/CaM-mediated signaling pathways. The CaM-binding domain of CBPs is not conserved; hence, biochemical and molecular methods based on protein-protein interaction have been used to identify CBPs (25-28). These methods have allowed identification of several plant CBPs that are involved in morphogenesis, cell division, cell elongation, ion transport, gene regulation, cytoskeletal organization, cytoplasmic streaming, pollen function, and stress tolerance (2, 11, 29).

Potato is the fourth most important crop, and understanding of molecular events in potato tuber initiation and development is of importance for manipulation of these processes to improve the yield and quality of potato (30). Both environmental and hormonal factors affect tuberization. Three factors have a major effect on tuberization: nitrogen levels, temperature, and light. High nitrogen levels were found to prevent tuberization, but how nitrogen levels control tuberization is not known (31). Some evidence shows that a withdrawal of nitrogen affects phytohormone levels, causing a reduction in gibberellic acid (GA) and an increase in abscisic acid, whereas other evidence indicates that the nitrogen/carbohydrate ratio is involved (31, 32). High temperature is also an inhibitor of tuberization and may also be mediated by GA levels. In one study, an inhibitor of GA biosynthesis overcame the inhibition of tuberization caused by high temperature (33). Light intensity as well as day length affects tuberization. Potato is a short day plant (the length of the dark period is the important factor) (34). Low light intensity delays tuberization even in short day conditions, and as is the case with nitrogen levels, both GA and carbohydrate have been implicated as mediators (30). The photoperiodism response of tuberization has to be transmitted from the leaf to the underground stolons, and GA has been implicated in this response also (30).

There is some evidence indicating the involvement of Ca²⁺/CaM in tuberization (35, 36). Potato leaf cuttings could be induced to produce tubers under short day conditions. Ca²⁺ depletion in the leaf cuttings prevented tuberization, and subsequent to Ca²⁺ depletion, if the leaf cuttings were transferred to CaCl₂ solution, tuber development occurred (35). In addition, treatment of leaf cuttings with CaM antagonists was found to inhibit tuberization (35). Transfer of the leaf cuttings that were pretreated with CaM antagonists to medium with no antagonists reversed the inhibition. Furthermore, transgenic potato plants with a moderate increase in expression of a CaM gene (PCM1) produced elongated tubers, whereas a large increase of PCM1 inhibited tuber formation and increased stolon elongation (36). Eight CaM (PCM1-PCM8) genes have been isolated and characterized from potato (16, 37). The expression patterns of these genes differed in various tissues (16, 37). These studies indicate regulation of tuberization by Ca²⁺/CaM.

To identify CaM-binding proteins that mediate CaM action in tuberization, we have used labeled CaM to screen an expression library prepared from developing tubers. A full-length clone of a novel CaM-binding protein (potato calmodulin-binding protein (PCBP)) was isolated in this screen. Ca²⁺-dependent binding of PCBP to CaM was confirmed using both biotinylated and ³⁵S-labeled CaM. The CaM-binding site was mapped to a stretch of 22 amino acids. Bacterially expressed PCBP interacted with three CaM isoforms. Fusion of the N-terminal region of PCBP to a cytosolic reporter gene targeted the fusion protein to the nucleus, indicating a nuclear role for PCBP.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Easy tag [³⁵S] isotope labeling mix was obtained from PerkinElmer Life Sciences. Exassist helper phage and Escherichia coli SOLR cells were from Stratagene. Triton X-100 free nitrocellulose filter discs were obtained from Millipore. Nitroblue tetrazoleum, 5-bromo-4-chloro-3-indolyl phosphate, and IPTG were obtained from Invitrogen. pET vectors and E. coli strain BL21 (DE3) were purchased from Novagen. Gelatin and diaminobenzidene were obtained from Sigma. Biotinylated CaM and bovine CaM were procured from Calbiochem. Avidin/biotin blocking reagents and Vectastain ABC horseradish peroxidase kit were obtained from Vector Laboratories. The complete protease inhibitor mix was from Roche Molecular Biochemicals. Bovine CaM Sepharose-4B was obtained from Amersham Biosciences, Inc. Immobilon membrane was obtained from Millipore. Potato plants were grown in the greenhouse at 26 °C (day)/18 °C (night).

Screening of Potato Expression Library with Biotinylated CaM-- A potato cDNA expression library was screened using biotinylated CaM following the procedure described previously (26, 27). The cDNA library in Zap II was made from RNA isolated from swelling stolon tips of potato (38). About 5 × 10⁴ plaque-forming units were plated on each plate using E. coli XL1-blue MRA as the host strain. The plates were incubated at 42 °C until the plaques appeared. Then the plates were overlaid with nitrocellulose filters that were presoaked in 10 mM IPTG. The plaques were allowed to grow overnight at 37 °C, the plates were cooled to 4 °C, and the filters were removed and rinsed briefly in TNMC buffer (50 mM Tris, pH 7.5, 0.2 M NaCl, 50 mM MgCl_2, 0.5 mM CaCl₂) at room temperature. The filters were blocked for 1 h in the above buffer containing 3% (w/v) gelatin at 30 °C, rinsed, and blocked with avidin and biotin as described earlier (26). The filters were then incubated in TNMC buffer containing biotinylated CaM (1 µg/ml) and 1% gelatin for 3 h at 30 °C. The filters were washed and transferred to TMNC buffer containing Vectastain ABC-horseradish peroxidase reagent (avidin DH-biotinylated horseradish peroxidase H complex) at 30 °C. The filters were washed twice, and the positive plaques were identified by incubating them in a substrate solution (0.8 mg/ml diaminobenzidine, 0.4 mg/ml NiCl₂, and 0.009% H₂O₂ in 100 mM Tris-HCl, pH 7.5). Screening of about a million recombinants resulted in the isolation of 10 putative CaM-binding clones, which were purified by two additional rounds of screening. The cDNA clones were excised in vivo in a plasmid (pBluescript) form following standard procedures.

Confirmation of Positive Clones with ³⁵S-Labeled CaM-- Ca²⁺-dependent binding of the isolated clones to CaM was further confirmed with ³⁵S-labeled CaM. An E. coli strain UT481 containing a CaM gene in an expression vector (pVUCH-1) was used to label CaM with [³⁵S]methionine (26, 39). Preparation of ³⁵S-CaM was described earlier (26). A cDNA (ICM-1) encoding a Ca²⁺/CaM-dependent protein kinase from mouse was used as a positive control (40).

DNA Sequencing and Analysis-- The cDNA insert from each clone was sequenced by a dideoxynucleotide chain termination method using double-stranded DNA as a template. Both strands of cDNAs were sequenced. Nucleotide sequences were assembled and analyzed using MacVector and Sequencher. BLAST searches were performed at the websites of the National Center for Biotechnology Information and the Arabidopsis Information Resource. Nuclear localization and PEST signals were identified using ExPASy Proteomics tools (www.expasy. ch/tools).

Construction of Recombinant Plasmids to Express Fusion Proteins-- Two constructs expressing truncated fusion proteins (pET 1.1 and pET 0.3) were prepared in pET28 expression vector. A 1.1-kb SacI/SalI cDNA fragment corresponding to the C-terminal end of the protein (amino acids 976-1309) was cloned into a pET28c vector that is digested with the same enzymes to generate pET 1.1 construct. The second construct (pET 0.3) containing the coding region for residues 976-1098 was made by deleting a 0.9-kb XhoI fragment from the pET 1.1 construct.

Expression of Fusion Proteins in E. coli-- pET 1.1 and pET 0.3 constructs were introduced into the E. coli BL21 (DE3) strain. Overnight cultures containing these constructs were used to inoculate fresh cultures that were grown to an approximate optical density of 0.6. The fusion protein was induced by adding 0.2 mM IPTG and growing the culture overnight at room temperature. The pellets were resuspended in 50 mM Tris, pH 8.0, 2 mM EDTA, 200 µg/ml lysozyme, and complete proteinase inhibitor mix. Lysed cells were sonicated six times for 10-15 s each and then centrifuged at 15,000 rpm (Sorvall) for 30 min to obtain soluble proteins.

Calmodulin-Sepharose Column Chromatography-- A calmodulin-Sepharose-4B column was prepared and equilibrated according to the instructions provided by Amersham Biosciences, Inc. The bacterial pellet from a 50-ml culture was suspended in 5 ml of buffer containing 50 mM Tris, pH 7.5, 150 mM NaCl, 10% glycerol, and complete proteinase inhibitor mix. The suspension was lysed by incubating on ice for a period of 1 h in the presence of 200 µg/ml lysozyme followed by five to six pulses of sonication on ice. The supernatant obtained after high speed centrifugation was passed through a CaM-Sepharose column. The unbound protein was washed with several volumes of binding buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 5 mM CaCl₂), and the bound protein was eluted with binding buffer except that CaCl₂ was replaced with 2 mM EGTA.

Calmodulin Interaction with Fusion Proteins-- Fusion proteins were separated on SDS-PAGE, blotted onto a nitrocellulose membrane, blocked overnight in 3% gelatin, and incubated with biotinylated CaM (26). The blots were then incubated in Vectastain ABC-horseradish peroxidase in TBS/calcium/magnesium or TBS/EGTA/magnesium for 30 min and washed three times for 10 min each time with the above buffer, and the CaM-binding proteins were detected colorimetrically as described above.

CaM Binding to a Synthetic Peptide-- A synthetic peptide (SKLKKLILLKRSIKALEKARKF) corresponding to amino acids 1216-1237 was synthesized in the Macromolecular Resource facility at Colorado State University. The purity of the peptides is over 95%. The interaction of bovine CaM and Arabidopsis CaM isoforms with the synthetic peptide was analyzed using electrophoretic mobility shift of CaM in the presence of the synthetic peptide (27, 41). Each of the Arabidopsis CaM isoforms and the bovine CaM (178 pmol) was incubated with increasing concentrations of the synthetic peptide (89, 178, and 356 pmol) in the presence of 4 M urea in a buffer containing 100 mM Tris-HCl, pH 7.5, and 1 mM CaCl₂ at room temperature for 1 h in a total volume of 20 µl. Ten µl of sample buffer (0.375 M Tris-HCl, pH 6.8, 30% glycerol, and 0.023% bromphenol blue) was added to each sample, and the mixture was electrophoresed in 14% polyacrylamide gels containing 7.5% glycerol, 0.375 M Tris-HCl, pH 8.8, and 1 mM CaCl₂. The gels were run, stained, and destained essentially as described previously (42).

Binding of PCBP to CaM Isoforms-- Soluble and insoluble proteins from the induced and uninduced cultures containing the pET 1.1 construct and the pET 1.1 protein that was purified on CaM-Sepharose column were separated on SDS-polyacrylamide gels and blotted as described above. To detect the binding of the fusion protein to ³⁵S-labeled CaM isoforms, blots were incubated for 15 min in a blocking buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1% nonfat dry milk) containing 5 mM of either CaCl₂ or EGTA. The blots were then incubated for 12 h in ³⁵S-labeled CaM2, CaM4, or CaM6 isoforms at 1 µg/ml in the same buffers as above. The filters were then washed with the corresponding buffer, dried, and exposed to x-ray film.

Southern and Northern Blots-- Genomic DNA was isolated from potato leaves following standard procedures and purified by CsCl density gradient (43). About 8 µg of genomic DNA was digested with different restriction enzymes, electrophoresed in a 1.0% agarose gel, and transferred onto a Hybond N nylon membrane. The DNA was fixed to the membrane by UV cross-linking. The blot was hybridized to radiolabeled cDNA fragment at 65 °C and washed under high stringency conditions (43). Total RNA was isolated by homogenizing the tissue in guanidinium thiocyanate and pelleting the RNA through a cesium chloride cushion (43). The RNA was separated in a 1.2% agarose gel containing formaldehyde and blotted onto a nylon membrane. The RNA blots were prehybridized and hybridized as above using the full-length cDNA as a probe. Prior to blotting RNA gels were tested for equal loading by staining with ethidium bromide.

Transient Expression of the N-terminal Region of PCBP Fusion in Onion Epidermal Cells-- A recombinant plasmid with a 2-kb BglII/BamHI fragment from PCBP containing multiple nuclear localization sequences was cloned in-frame following a GUS gene driven by a cauliflower mosaic virus 35S promoter in pRTL 2-GUS. A pRTL 2-GUS-NIa construct (44) was used to make the pRTL 2-GUS-PCBP construct. The BglII/BamHI fragment in pRTL 2-GUS/NIa was deleted and replaced with the BglII/BamHI fragment from PCBP to create pRTL 2-GUS-PCBP. The plasmids pBI-GUS, in which GUS is driven by the same promoter, and pRTL2-GUS-NIa, in which GUS is fused to a known nuclear protein, were used as controls. Cells in the epidermal layer of onion bulbs were transformed with the pRTL 2-GUS-PCBP, pRTL 2-GUS-NIa, or pBI-GUS using the helium biolistic particle delivery system (Bio-Rad). Onion epidermal layers were peeled and placed on solidified phytagar (0.4%) in Petri dishes. Plasmid DNA was precipitated on to sterilized tungsten particles (1.1 µm) by mixing 50 µl (3 mg) of particles, 5 µl of DNA (1 µg/µl), 50 µl of 2.5 M calcium chloride, and 20 µl of 0.1 M spermidine. The mixture was vortexed, incubated for 10 min at room temperature, and centrifuged briefly. The beads were washed first with 70% ethanol and then with 100% ethanol. Tungsten particles coated with DNA were resuspended in 48 µl of 100% ethanol, and 12 µl of this was used for each bombardment. 1100 p.s.i. rupture discs were used. After bombardment, Petri plates containing the epidermal layers were incubated overnight at 22 °C. The cell layers were then stained for GUS activity as described earlier (45) and mounted in phosphate-buffered saline containing 1 µg/ml 4',6'-diamidino-2-phenylindole (DAPI). The GUS-stained cells were visualized using bright field microscopy, and the nuclei were visualized under UV. The pictures were captured using a digital camera (Kodak).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Isolation of cDNAs Encoding a CaM-binding Protein-- To isolate CBPs from developing tubers, a cDNA library from swelling stolon tips of potato was screened with biotinylated CaM (26). Screening of about one million recombinants resulted in isolation of 10 positives. Rescreening of these positive clones by excluding the incubation step with biotinylated CaM did not yield positive signals, indicating that the isolated clones are true positives. Furthermore, all 10 clones bound biotinylated CaM only in the presence of Ca²⁺ (data not shown). To further confirm Ca²⁺-dependent binding of proteins encoded by isolated clones, plaques from positive clones were probed with ³⁵S-CaM in the presence of CaCl₂ or EGTA. A cDNA (ICM-1), which encodes a CaM-binding protein from mouse, was used as a positive control (40). Positive clones were detected in the presence of CaCl₂ but not in the presence of EGTA (Fig. 1). Sequence analysis of all the clones has revealed that the isolated cDNAs fall into two groups. One group of clones was represented by three independent clones (PCBP4, PCBP7, and PCBP25) of the same gene but differing in their length. One of the three clones (PCBP25) is a full-length cDNA, whereas the other two clones contain partial cDNAs (see below).

View larger version (33K): [in this window] [in a new window]   Fig. 1.   Ca2+-dependent binding of 35S-labeled CaM to PCBP. Recombinant phages containing full-length PCBP (upper panel) or ICM-1 (lower panel) were lifted from plates onto filters and probed with 35S-labeled CaM. One half of each filter was incubated in buffer containing CaCl2, and the other half was incubated in buffer containing EGTA. ICM-1 encodes a CaM-binding protein from mouse (40).

Sequence Analysis of the Potato CaM-binding Protein-- The longest cDNA is about 5 kb and has an open reading frame starting at nucleotide 70 and terminating at 3997. The deduced protein is 1309 amino acids long. PCBP7 and PCBP4 contain the coding region from amino acids 18 and 631, respectively. In addition, these three clones differed in the length of the 3'-untranslated region with the longest 3'-untranslated region in PCBP25. This variation in the 3'-untranslated region is most likely due to differential polyadenylation, which is prevalent in plants (46). The deduced protein from the full-length gene has a molecular mass of about 146 kDa with an isoelectric point of 6.06. The predicted protein is rich in serine (13%), acidic (16%), and basic (17%) amino acids. 46% of the total amino acids are represented by six amino acids (Ser, Asp, Glu, Lys, Arg, and His).

Similarity searches with PCBP were done using BLAST at the National Center for Biotechnology Information and the Arabidopsis Information Resource (www.arabidopsis.org). A hypothetical protein in the recently completed Arabidopsis genome sequence data base (AT5g04020) (47) showed significant similarity to PCBP. Overall there is 23% identity and 52.4% similarity between PCBP and the Arabidopsis protein (Fig. 2A). PCBP has two sequence stretches (PCBP/D1 and PCBP/D2) that are very similar, having 58% identity and 88% similarity (Fig. 3). PCBP/D1 consists of amino acids 750-838, and PCBP/D2 consists of amino acids 1215-1303. The CaM-binding domain is in PCBP/D2 (see below). AT5g04020 has four sequence regions (Fig. 3, At1/D1-At1/D4) that show very high amino acid identity (41-60%) and similarity (77-89%) to the two repeat regions of PCBP and to each other. Two other Arabidopsis hypothetical proteins (At2g38800 and At3g54570) have a small region that shows similarity to the PCBP repeat domains (41-49% identity, 75-82% similarity). The CaM-binding domain in PCBP (see below) is conserved in At5g04020 (At1) but not in At2g38800 and At3g54570 (Fig. 3, At2 and At3).

View larger version (84K): [in this window] [in a new window]   Fig. 2.   Analysis of PCBP sequence. A, alignment of deduced amino acid sequence of PCBP with a hypothetical from Arabidopsis (AT5g04020). Sequences were aligned using the FASTA program. Identical amino acids are in white on a black background, and similar amino acids are shaded gray. The numbers on the right correspond to the amino acid number of each protein. The AT5g04020 protein sequence starts at amino acid 138. PCBP is shorter than AT5g04020 by 131 amino acids. Dashes indicate gaps introduced to give the best alignment. The CaM-binding domain of PCBP is overlined. B, schematic diagram showing NLS and PEST sequences. The NLS are clustered in the N terminus except for one near the C terminus. Boxes labeled B, bipartite NLS; boxes labeled S, SV 40-like NLS (the number next to B or S indicates the beginning of the NLS). The range of PEST scores is 0.37 to 4.33 for weak PEST sequences and 6.31-14.99 for strong PEST sequences (weak PEST Scores <5; strong PEST scores >5).

View larger version (52K): [in this window] [in a new window]   Fig. 3.   Alignment of two PCBP repeats with similar regions in Arabidopsis hypothetical proteins. Two repeats in PCBP, four repeats (At1/D1-At1/D4) in Arabidopsis protein AT5g04020, and one domain of Arabidopsis proteins At2g38800 (At2) and At3g54570 (At3) have strong sequence similarity. Identical amino acids are in white on a black background. The numbers on the left correspond to the amino acid numbers of each protein. The asterisks mark residues conserved in all sequences. The CaM-binding region is overlined.

The PCBP sequence was analyzed using programs found on the ExPASy Proteomics tools web page (www.expasy.ch/tools). Using PSORT (psort.nibb.ac.jp/), seven potential nuclear localization signals were identified (Fig. 2B). Four SV40 large T antigen nuclear targeting signals (4-residue pattern composed of all basic amino acids or three basic amino acids (and a Pro) and three bipartite nuclear targeting signals were identified (48). The final certainty number reported by PSORT for PCBP being a nuclear protein is 0.980. Proteins containing PEST regions are thought to be rapidly degraded proteins (49, 50). PEST sequences are defined as hydrophilic stretches of amino acids greater than or equal to 12 residues in length. Such regions contain at least one Pro, one Glu or Asp, and one Ser or Thr. They are flanked by lysine, arginine, or histidine, but positively charged residues are disallowed within the PEST sequence (49, 50). Fig. 2B shows the location of several potential strong (pest scores, 6.31-14.99) and weak (pest scores, 0.37-4.33) PEST sequences in PCBP (PESTfind, www.at.embnet.org/embnet/tools/bio/PESTfind/). A SMART (Simple Modular Architecture Research Tool) search that identifies the presence of known domains in a protein did not result in any identifiable domains in PCBP (51, 52). Prosite scan revealed several potential phosphorylation sites in PCBP indicating that post-translation changes may be involved in regulating the activity/stability of this protein.

Mapping of the Calmodulin-binding Domain-- Both the full-length cDNA and the two shorter clones isolated in the original library screen were positive for binding CaM (Fig. 4A). Expressed full-length and truncated proteins from all three clones showed CaM-binding (data not shown). Because the shorter clone begins at amino acid 631 (Fig. 4A), the CaM-binding domain was first narrowed down to the region C-terminal to 631. To map the location of the CaM-binding domain, two truncated versions of the cDNA were made and expressed in E. coli as T-7 tag fusions. As shown in Fig. 4A, one truncated clone contained amino acids 976-1309 (pET1.1) and the other 976-1098 (pET 0.3). Protein from uninduced and induced E. coli containing pET 1.1 was isolated, and the induced protein was purified on a CaM-Sepharose column. Protein was loaded on two gels, and one gel was stained, and the other was blotted and probed with biotinylated CaM. As shown in Fig. 4B, the protein was purified to near homogeneity and was bound by biotinylated CaM both in the pure form and in the induced crude protein. Protein from pET 0.3 was run on a gel with uninduced bacterial protein and with crude extract from a positive control containing KCBP (expressed as T7 fusion) (53). Two identical blots were prepared and probed with either biotinylated CaM or T-7 antibody. Fig. 4C shows the stained gels and the two blots. The CaM-binding protein KCBP binds CAM, but the pET 0.3 clone does not. However, the T-7 antibody recognized both KCBP and the pET 0.3 protein product as expected. These results map the CaM-binding domain to the region between amino acids 1098 and 1309. 

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Mapping of CaM-binding domain. A, three cDNAs of PCBP (PCBP4, PCBP7, and PCBP25) were isolated from the library screen. pET 1.1 (amino acids 976-1309) and pET 0.3 (amino acids 976-1098) are constructs made in pET28 in which the proteins are expressed as a T7 tag fusion. CaM binding ability is shown in the right-hand column. +, binds CaM; , does not bind CaM. B, E. coli cells containing pET1.1 were induced with IPTG to express the truncated PCBP. Crude uninduced protein (u), induced protein (i), and purified protein (p) were run on duplicate gels. One gel was stained (Stain), and the other was blotted and probed with biotinylated CaM (BCaM). C, E. coli cells containing pET 0.3 were induced with IPTG to express the truncated PCBP. Crude uninduced (u), induced (i), and a known CaM-binding T7 tag fusion protein called KCBP (+) were run on triplicate gels. One was stained (Stain) and two were blotted and probed with biotinylated CaM (BCaM) or T7 antibody (T7). The arrows indicate PCBP fusion protein, and the arrowheads indicate truncated KCBP, a positive control. Molecular mass markers are shown to the left in kDa.

Binding of PCBP to CaM Isoforms and Mapping of Its CaM-binding Domain-- CaM isoforms have been shown to differentially regulate their target proteins, and some CaM isoforms have been shown to act as competitive antagonists in activating target enzymes (18-22). To test whether different isoforms of plant CaM interact with PCBP, bacterially produced truncated protein (crude and CaM-Sepharose purified) from the pET 1.1 construct was separated on a gel, blotted, and probed with three ³⁵S-labeled CaM isoforms. All three CaM isoforms (CaM2, CaM4, and CaM6) bound to purified PCBP and specifically to PCBP in the induced crude protein (Fig. 5A). No binding was seen with crude uninduced bacterial protein (Fig. 5A) or in the presence of calcium chelator, EGTA, in the buffer (data not shown).

View larger version (48K): [in this window] [in a new window]   Fig. 5.   PCBP binding to plant CaM isoforms. A, truncated PCBP binding to different CaMs. Bacterially expressed protein containing truncated PCBP (pET 1.1) consisting of amino acids 631-1309 (including the CaM-binding domain) was purified on a CaM-Sepharose column. Crude bacterial protein from E. coli containing the vector only (lanes 1), crude bacterial protein containing truncated PCBP (lanes 2), and purified truncated PCBP (lanes 3) were electrophoresed, and the gels were either stained or blotted to a membrane. The blots were probed with Arabidopsis CaM2, CaM4, or CaM6. The markers are indicated on the right in kDa. B, synthetic peptide binding to CaMs. Binding of a synthetic peptide corresponding to amino acids 1216-1237 of PCBP (Fig. 2) was analyzed by electrophoretic mobility shift. Bovine CaM or Arabidopsis CaM isoforms CaM2, CaM4, or CaM6 were mixed with the peptide in a 1:0 (CaM to peptide), 1:0.5, 1:1, or 1:2 molar ratio. The lower bands are unbound CaM, and the upper bands are bound CaM.

CaM-binding proteins do not have a conserved amino acid sequence that can be identified as the CaM-binding domain. However, the binding region is known to form a basic, amphiphilic -helix in which hydrophobic residues are segregated from hydrophilic residues along the helix (2, 11, 23, 29, 54, 55). The PCBP region that showed CaM binding (1098-1309) was analyzed using a helical-wheel program, and a sequence was identified that could form a basic, amphiphilic -helix. The sequence SKLKKLILLKRSIKALEKARKF from amino acids 1216-1237 (Fig. 2) was predicted to form a basic, amphiphilic -helix.

A synthetic peptide containing the predicted sequence was made and used for CaM binding studies. The binding of the synthetic peptide was detected by gel mobility shift assays (Fig. 5B). A bovine CaM and three plant CaM isoforms (CaM2, CaM4, and CaM6) were used in the binding assay with the synthetic peptide in three different ratios of CaM:peptide: 1:0.5, 1:1, and 1:2. Calmodulin alone (1:0 in Fig. 5B) migrates according to its molecular weight. With the addition of 1:0.5 synthetic peptide, a small amount of CaM is shifted up indicating binding by the synthetic peptide and formation of a peptide-CaM complex. An increase to 1:1 increases the amount of CaM bound to about 50%, whereas a 1:2 ratio of peptide bound almost all of the CaM protein. These binding studies indicate that PCBP binds three CaM isoforms and that the CaM-binding domain is present in a 22-amino acid stretch (from 1216-1237) at the C terminus.

PCBP Is Coded by a Single Gene-- To determine whether PCBP is coded by a single gene, we digested the potato genomic DNA with different restriction endonucleases and probed the DNA blot with a cDNA fragment that is expected to hybridize to a single band in each digestion. As shown in Fig. 6, Southern analysis revealed a single hybridizing band with each of the restriction enzyme digestions, indicating that PCBP is coded by a single gene. Low stringency washes did not reveal additional bands (data not shown), suggesting that there are no other PCBP-related genes in the potato genome. In the recently completed Arabidopsis genome sequence data base (www.arabidopsis.org), there is only one gene that is highly similar to PCBP, suggesting that the PCBP ortholog in Arabidopsis is also encoded by a single gene.

View larger version (43K): [in this window] [in a new window]   Fig. 6.   Southern blot analysis. 8 µg of genomic DNA was digested with EcoRI (E), HindIII (H), XhoI (X), or BamHI (B) and electrophoresed through a 0.8% agarose gel, transferred onto a Hybond nylon membrane, and probed with a 32P-labeled cDNA fragment. Size markers are shown on the right in kb.

Expression of PCBP, CaM, and Another Potato CBP in Different Tissues-- To determine whether the longest cDNA (about 5 kb) represents the full-length transcript and to determine the expression of PCBP in different tissues, RNA was isolated from the apical bud, leaf, root, stem, stolon tip, tuber, and flower tissues. Using labeled full-length cDNA as a probe, a transcript of about 5 kb was identified in Northern hybridization analysis (Fig. 7). High levels of PCBP transcript were found in the stem, leaf, and apical bud with the highest expression in the stem. The expression of PCBP in tubers and stolons is low. Longer exposure of blots clearly showed PCBP expression in tubers and stolons (data not shown).

View larger version (58K): [in this window] [in a new window]   Fig. 7.   Northern blot analysis. Total RNA from plant tissues as labeled was electrophoresed on formaldehyde-containing agarose gels, transferred to Hybond N+ membranes, and hybridized with 32P-labeled full-length cDNAs of PCBP, KCBP, CaM, or ubiquitin. The stained gel as well as the expression of a constitutively expressed ubiquitin show that the amount of RNA is loaded equally in each lane. A, expression in vegetative tissues. B, expression in reproductive tissues. Transcript sizes are shown on the right in kb.

Using duplicate blots, the RNA was also probed with KCBP, another known CaM-binding protein from potato, a potato CaM, and ubiquitin, a gene expressed equally in all tissues (Fig. 7) (37, 53). KCBP and CaM were expressed in a similar pattern to PCBP with most expression in the apical bud, leaf, and stem tissues. Ubiquitin was expressed equally in all tissues tested (Fig. 7). The expression of KCBP in all the tissues shown in Fig. 7A was demonstrated previously by reverse transcription PCR (53). PCBP, KCBP, and CaM were all also expressed in flower and berry (Fig. 7B). Expression in flowers was further analyzed by isolating RNA from the sepals, petals, stamens, and carpels separately. Expression of PCBP in petals, stamens, and carpels was less than in sepals (Fig. 7B). KCBP and CaM were present in a similar pattern except KCBP is expressed more strongly in carpels. These results indicate that CaM and its target proteins are coordinately expressed.

PCBP Localization-- The presence of multiple nuclear localization signals (NLS) in PCBP suggests that it is likely to be a nuclear protein. Six of the seven NLS sequences are in the first 600 amino acids of PCBP (Fig. 2B). GUS is a cytosolic protein, and fusion of this reporter gene to putative NLS sequences has been used to redirect GUS to the nucleus to verify the NLS sequences (56). The N-terminal region of PCBP (amino acids 18-716) containing six of the seven NLS sequences was fused to a GUS reporter gene under the control of a constitutive promoter (cauliflower mosaic virus 35S promoter). The GUS-PCBP construct was used to analyze the cellular localization of PCBP. A construct containing only the reporter (GUS) gene under the control of the same promoter was used as a negative control. A construct containing a known nuclear localized protein NIa (GUS-NIa) was used as a positive control (44). GUS, GUS-NIa, and GUS-PCBP constructs were introduced into onion epidermal cells by using particle bombardment and stained for GUS activity following overnight incubation. As shown in Fig. 8, for the construct containing the GUS reporter gene only, GUS activity is distributed more or less evenly in the cytoplasm, and the nucleus is not distinguishable (Fig. 8). However, the GUS activity from the fusion of PCBP to GUS was found mostly in the nucleus with some GUS staining in the cytoplasm (Fig. 8, GUS-PCBP). The distribution of PCBP-GUS is very similar to NIa-GUS, where GUS is fused to a known nuclear protein (Fig. 8, GUS-NIa). These results suggest that PCBP is a nuclear protein.

View larger version (52K): [in this window] [in a new window]   Fig. 8.   Nuclear localization of GUS-PCBP. Plasmids containing GUS, GUS-NIa, or GUS-PCBP fusions were introduced into onion bulb epidermal peels by biolistic transformation. Following bombardment, the epidermal peels were incubated overnight at 22 °C and stained with GUS. DAPI was used to stain the nucleus. A and B show two representative samples of GUS staining with each construct. The upper panels show GUS activity, and the lower panels show DAPI stain. The white arrows point to the DAPI stained nuclei, and the black arrows point to the corresponding nuclei showing high GUS activity.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We isolated a full-length cDNA encoding a novel CaM-binding protein (PCBP) using a protein-protein interaction-based screening. With ³⁵S-labeled CaM we confirmed that PCBP binds CaM in the presence of Ca²⁺ but not in the presence of EGTA (Fig. 1). CaM binding studies with truncated proteins have also shown that the CaM-binding domain is located near the C terminus of the protein (Figs. 2, 4, and 5). A truncated version of PCBP was purified on a CaM-Sepharose column, further confirming that the protein is a CaM-binding protein that binds CaM in the presence of CaCl₂ but not in the presence of EGTA (Fig. 5A). A peptide that is predicted to form a basic, amphiphilic -helix was identified as a possible CaM-binding domain, and this was verified by CaM mobility shift assays (Fig. 5B). Binding of CaM by peptide increased as the ratio of peptide to CaM increased until most of the CaM was bound at a ratio of 1:2 (CaM:peptide). The synthetic peptide retarded the mobility of plant and animal CaM in a similar manner. There are two regions in PCBP (PCBP/D1 and PCBP/D2) that share a very high sequence similarity (Fig. 3). The CaM-binding domain that we mapped is in the D2 region, and there is a high degree of similarity between PCBP/D1 and PCBP/D2 in the CaM-binding domain; hence it is possible that there is a second CaM-binding domain. Proteins with more than one CaM-binding site have been reported (57, 58). Although CaM-binding sequences in different proteins are not conserved, orthologs of a CBP in different species and members of CaM-binding protein gene families have a similar CaM-binding sequence (59-63).

Comparison of the sequence using Blast revealed significant similarity to a hypothetical protein from Arabidopsis (At5g04020), suggesting that it is an Arabidopsis ortholog of PCBP (Fig. 2). Because the Arabidopsis genome is completely sequenced (47), it is not likely that there is a more closely related protein in Arabidopsis. As shown in Fig. 3, the homologous regions of PCBP show significant similarity to each other and to four regions in the PCBP-related Arabidopsis protein. The CaM-binding domain (1216-1237) in the PCBP/D2 region (1215-1303) is highly conserved in the four At1 domains, suggesting that the Arabidopsis protein is also likely to be a CaM-binding protein and may contain more than one CaM-binding domain. In addition, two other Arabidopsis proteins showed limited sequence similarity to the repeated regions (D1 and D2) in PCBP (Fig. 3), suggesting that this region may represent a domain of unknown function that is present in other plant proteins. However, the CaM-binding domain is not conserved in these two other hypothetical proteins from Arabidopsis (Fig. 3, At2 and At3), and hence they may not be CaM-binding proteins. The highly conserved region between the four proteins, however, indicates that they are likely to be involved in some conserved function. No other proteins in the data bases, neither plant nor animal, showed any significant similarity to PCBP. The absence of PCBP in the completely sequenced genomes of yeast (Saccharomyces cerevisiae and Schizosaccharomyces pombe), Caenorhabditis elegans, Drosophila melanogaster, and Homo sapiens suggests that PCBP is likely to be a plant-specific CaM-binding protein.

So far, only two other CaM-binding proteins have been isolated from potato. These include KCBP (53) and a homolog of mammalian multidrug-resistant P-glycoprotein (PMDR1) (64). Genetic studies have shown that KCBP is essential for trichome morphogenesis (65, 66). In addition, several lines of evidence indicates that KCBP, a minus-end microtubule motor protein, is involved in cell division (67-72). PMDR1 was isolated from a stolon tip library and is expressed in all organs studied. Expression of PMDR1 is higher in the stem and the stolon tip with highest expression during tuber initiation and decreased expression during tuber development (64). The homologs of PMDR1 are involved in transport or secretion to keep toxic metabolites and xenobiotics out of normal tissues. PMDR1 may be involved in some transport, but its function has not been established.

Both of these potato CaM-binding proteins (KCBP and PMDR1) have homologs in Arabidopsis. The Arabidopsis KCBP (AtKCBP) is 80% identical and 91% similar to potato KCBP at the amino acid level (53). The Arabidopsis homolog of PMDR1 (ATPGP1) is 85.7% identical and 92.4% similar at the amino acid level (64). However, the similarity between PCBP and its ortholog in Arabidopsis is only 52.4%, suggesting that this protein is not as highly conserved as KCBP or PMDR1 between potato and Arabidopsis.

The presence of a number of PEST signals suggests that this protein is rapidly turned over. PEST signals are found in rapidly degraded enzymes, transcription factors, and components of receptor signaling pathways. They are, by contrast, rarely present among long lived cellular proteins (73). Multiple PEST sites are common among proteins in signal pathways (74). Because there are multiple PEST sites throughout the protein, PCBP is likely to be involved in a signal pathway (Fig. 2). The fact that PCBP binds CaM in a Ca²⁺-dependent manner is also indicative of its involvement in a Ca²⁺ signaling pathway.

Because PCBP was isolated from a swelling stolon tip library, logically it could be involved in stolonization, tuberization, or both. However, expression studies indicate that the gene encoding PCBP is expressed in all the tested tissues including vegetative and reproductive tissues. A high level of expression was observed in stems, flowers, and fruits, whereas its expression is very low in stolon tips and young tubers. Although the PCBP cDNAs are isolated from a library prepared from developing tubers (38), the expression patterns indicates a role for PCBP in other tissues also. Similar expression patterns of PCBP, CaM, and KCBP suggest a coordinate regulation of transcription of CaM and its targets. It is likely that CaM and CaM-binding proteins that are studied here may have some common cis-elements in their promoter regions.

Another feature of the amino acid sequence of PCBP is the presence of nuclear localization signals. Both SV40-type and bipartite NLS sequences are present in PCBP. The bipartite pattern is: 2 basic residues, an ~10-residue spacer, and another basic region consisting of at least 3 basic residues of 5 (48). Seven putative nuclear localization signals with high certainty were identified in PCBP using PSORT, suggesting that it is most likely a nuclear protein. Fusion of the PCBP N-terminal region containing six of seven nuclear localization signals to a cytosolic reporter gene targeted the reporter gene to the nucleus, giving more evidence that PCBP may be a nuclear protein.

The presence of CaM in the nucleus has been documented in different cell types (75), and nuclear CaM-binding proteins in animals have been identified that include protein kinases, calcineurin, transcription factors, RNA-binding proteins, nucleolar-ribosomal proteins, chaperones, and others (76). Nuclear CBPs have also been identified in plants including a chromatin-associated NTPase in pea (77) and a DNA-binding protein from cauliflower (78). A CaM-binding protein with a putative DNA-binding domain was also isolated recently from Arabidopsis (61). However, the localization of this protein to the nucleus has not been demonstrated. The localization of PCBP to the nucleus indicates that nuclear Ca²⁺ levels are likely to regulate the activity/function of this protein. Recently, it has been shown that the nuclear membrane is not passively permeable to Ca²⁺ and that a Ca²⁺ gradient exists between the nucleus and cytoplasm, suggesting the existence of regulatory mechanisms that control movement into and out of the nucleus (79, 80). Furthermore, the transport of Ca²⁺ across the nuclear membrane in plant and animal cells is dependent on ATP (79, 81). Also, changes in nuclear and cytoplasmic Ca²⁺ appear to be controlled independently (80). Certain signals in plants have been shown to differentially alter free Ca²⁺ in the cytoplasm and nuclear compartment (82). Nuclear CaM-binding proteins such as PCBP are likely to be regulated by the changes in free Ca²⁺ in the nucleus.

In summary, using a protein-protein interaction-based screening we have isolated a novel plant-specific CaM-binding protein and mapped the CaM-binding region to a short stretch of amino acids at the C terminus. The binding of PCBP to three CaM isoforms suggests its interaction with different CaM isoforms. Ca²⁺-dependent binding of CaM to PCBP suggests the involvement of PCBP in Ca²⁺ signaling pathway and likely regulation of its activity/function by Ca²⁺. The expression pattern of PCBP indicates its role in various tissues. Localization of PCBP to the nucleus strongly suggests a nuclear role for PCBP.

	ACKNOWLEDGEMENTS

We thank Dr. Mark A. Taylor of Scottish Crop Research Institute (Dundee, UK) for providing the cDNA library, Dr. D. M. Watterson (Vanderbilt University, Nashville, TN) for the pVUCH-1 clone, Dr. R. E. Zielinski (University of Illinois) for Arabidopsis CaM constructs, and Dr. James Carrington (Washington State University) for pRTL2-GUS vector. Nucleotide and protein sequence analyses were performed at the National Center for Biotechnology Information, the Arabidopsis Information Resource, and ExPASy.

	FOOTNOTES

^* This work was supported in part by Agricultural Experiment Station Project 702 and grants from the National Science Foundation and the National Aeronautic and Space Administration (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/EBI Data Bank with accession number(s) AF378084.

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

Published, JBC Papers in Press, October 29, 2001, DOI 10.1074/jbc.M104595200

	ABBREVIATIONS

The abbreviations used are: CaM, calmodulin; PCBP, potato calmodulin-binding protein; CBP, calmodulin-binding protein; KCBP, kinesin-like calmodulin-binding protein; GUS, -glucuronidase; GA, gibberellic acid; IPTG, isopropyl-1-thio--D-galactopyranoside; DAPI, diamidinophenylindole; NLS, nuclear localization signal(s).

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. C. Kim, W. S. Chung, D.-J. Yun, and M. J. Cho Calcium and Calmodulin-Mediated Regulation of Gene Expression in Plants Mol Plant, January 6, 2009; (2009) ssn091v1. [Abstract] [Full Text] [PDF]

				L. Ma, S. Liang, R. L. Jones, and Y.-T. Lu Characterization of a Novel Calcium/Calmodulin-Dependent Protein Kinase from Tobacco Plant Physiology, July 1, 2004; 135(3): 1280 - 1293. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/6/4206    most recent M104595200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Reddy, A. S. N. Articles by Harnly, M. J. Search for Related Content PubMed PubMed Citation Articles by Reddy, A. S. N. Articles by Harnly, M. J. Social Bookmarking                      What's this?