PsCCaMK is tightly associated with its substrate, a 40-kDa protein p40 that binds to specific sequence elements in the promoter of the Arabidopsis CaM5 gene AtCaM5 gene in a phosphorylati
Trang 1A Ca2+/CaM-dependent kinase from pea is stress regulated
Sona Pandey*, Shiv B Tiwari†, Wricha Tyagi, Mali K Reddy, Kailash C Upadhyaya and Sudhir K Sopory
School of Life Sciences, Jawaharlal Nehru University, New Delhi, India and International Center for Genetic Engineering and Biotechnology, New Delhi, India
An immuno-homologue of maize Ca2+/calmodulin
(CaM)-dependent protein kinase with a molecular mass of 72 kDa
was identified in pea The pea kinase (PsCCaMK) was
upregulated in roots in response to low temperature and
increased salinity Exogenous Ca2+ application increased
the kinase level and the response was faster than that
obtained following stress application Low
temperature-mediated, but not salinity-mediated stress kinase increase
was inhibited by the application of EGTA and W7, a CaM
inhibitor The purification of PsCCaMK using
immuno-affinity chromatography resulted in coelution of the kinase
with another polypeptide of molecular mass 40 kDa (p40)
Western blot revealed the presence of PsCCaMK in nuclear protein extracts and was found to phosphorylate p40 in vitro Gel mobility shift and South-Western analysis showed that p40 is a DNA-binding protein and it interacted specifically with one of the cis acting elements of the Arabidopsis CaM5 gene (AtCaM5) promoter The binding of p40 to the specific elements in the AtCaM5 promoter was dependent of its dephosphorylated state Our results suggest that p40 could
be an upstream signal component of the stress responses Keywords: calmodulin; DNA–protein interaction; plant protein kinase; protein phosphorylation; stress signaling
Plants perceive a variety of signals from the external
environment as well as from the internal cellular milieu
generated during various developmental processes Signals,
such as light, nutrients and various environmental stresses,
etc are perceived by specific receptors Following
percep-tion, a number of second messengers are generated that
regulate the activity of other proteins such as kinases and
phosphatases to transduce the signal downstream Towards
the end of the signal transduction pathway, these second
messengers and/or other accessory protein(s) affected by
them modulate the activity of the transcription factors,
which regulate the expression of specific gene(s) leading to
the final response The events related to perception of a
signal and corresponding changes in the activity and
concentration of various second messengers have been
studied in great detail However, the downstream pathways
leading to the final control of gene expression have been
elucidated in few cases only [1–3]
Calcium ions are the most important second messengers, controlling a variety of cellular and physiological responses [4,5] Cytosolic concentration of calcium ([Ca2+]cyt) chan-ges in response to a number of external stimuli and internal physiological developments [6,7] Besides, a num-ber of other second messenger such as cyclic ADP-ribose, inositol-3-phosphate and various proteins such as calmo-dulin (CaM) and calcium-dependent protein kinases (CDPKs) are also affected by changes in ([Ca2+]cyt) [6,8] The information through ([Ca2+]cyt) is transduced by two main pathways, one involving CaM and CaM-related proteins, and the other involving CDPKs Both of these pathways cross-talk at various points in the signaling cascade [9] The presence of CaM has been detected in different plant cell compartments especially in the nuclei [10] In animal systems, it has been shown conclusively that nuclear CaM, in combination with nuclear Ca2+changes, regulates the expression of various genes either by inter-acting with the transcription factors directly [11] or via specific calmodulin kinases (CaMKs) [12–14] In plant systems too, recent studies have shown that CaM is involved in the regulation of gene expression Szymanski
et al [15] have shown that CaM affects the binding of TGA3 to the Arabidopsis CaM3 promoter A study by van der Luit et al [16] showed that distinct calcium signaling pathways that operate during cold (predominantly cyto-plasmic) and wind (predominantly nucleus) signaling are regulated via CaM gene expression They have identified two different CaM isoforms, having different nucleotide sequences, but coding for the same polypeptide, and only one of these (NpCaM-1) is affected by cold and wind signaling
CaMKs are important junctions in signal transduction where Ca2+-dependent and CaM-dependent signaling pathways converge and these kinases are potential candi-dates in regulating gene expression Reports on the existence
Correspondence to S Pandey, 208 Mueller Laboratory,
Biology Department, Pennsylvania State University,
University Park, PA 16802, USA.
Fax: + 1 814 865 9131, E-mail: sxp49@psu.edu
Abbreviations: CaM, calmodulin; CaMK, calmodulin kinase;
CDPK, calcium-dependent protein kinase; GMSA, gel mobility shift
analysis; W7, N-(6-aminohexyl)-5-chloro-1-napthalenesulphonamide
hydrochloride; HMG, high mobility group.
Note: S Pandey and S B Tiwari contributed equally to this work.
*Present address: 208 Mueller Laboratory, Biology Department, The
Pennsylvania State University, University Park, PA 16802, USA.
Present address: Biochemistry Department, 117 Schweitzer Hall,
University of Missouri, Columbia, MO 65211, USA.
(Received 15 February 2002, revised 8 May 2002,
accepted 13 May 2002)
Trang 2of this class of kinase are limited in plants [17–21] The
Ca2+/CaM kinase isolated from maize roots has been
shown to be involved in gravitropism [19] and the Ca2+/
CaM kinase isolated from lily anthers binds with the
translational elongation factor II [22]
We reported previously the purification and
characteri-zation of a novel Ca2+/CaM-dependent protein kinase,
ZmCCaMK from etiolated maize coleoptiles [21,23] We
have now identified an immuno-homologue of this kinase in
pea (PsCCaMK) and in this study we show that PsCCaMK
is involved in Ca2+/CaM-regulated stress signaling in
plants PsCCaMK is tightly associated with its substrate,
a 40-kDa protein (p40) that binds to specific sequence
elements in the promoter of the Arabidopsis CaM5 gene
(AtCaM5) gene in a
phosphorylation/dephosphorylation-dependent manner
E X P E R I M E N T A L P R O C E D U R E S
Plant material, growth conditions and stress
treatments
Pea (Pisum sativum) plants were grown in moist vermiculite
(16-h light/8-h dark) at 25C for 5 days For salinity stress,
5-day-old plants were treated in 50, 100, 200, 300 or 500 mM
NaCl solution for specified time periods For osmotic stress,
plants were similarly treated with 300 mMmannitol for 24,
48 or 96 h For temperature stress, 5-day-old plants were
transferred either to 4C (cold stress) or 40 C (heat stress)
for specified time periods Calcium treatment (10, 25, 50 or
100 mM) was given to the 5-day-old plants for 3, 6, 12 or
24 h Plants grown under normal conditions for the same
time period served as controls
For pharmacological experiments, 5-day-old plants were
treated with EGTA (10 mM), lanthanum (10 mM) or W7
(60 lM) for 12 h These plants were further subjected to
different stresses as described earlier Plants treated with
these compounds but without stress conditions served as
control for these experiments After the treatments roots
and shoots of the treated as well as untreated (control)
plants were harvested separately, frozen in liquid N2 and
stored at)80 C until use
Protein extraction, SDS/PAGE and immunoblotting
Frozen tissue was ground to a fine powder in liquid N2and
extracted with 3 vols extraction buffer [20 mM Hepes
pH 7.5, 2 mM EDTA, 5 mM EGTA, 2 mM
phenyl-methanesulfonyl fluoride, 5 mM dithiothreitol and 10%
glycerol (v/v)] on ice The slurry obtained was centrifuged at
12 000 g for 30 min and the supernatant was used for SDS/
PAGE and immunoblotting studies Protein estimation was
carried out according to Bradford [24] and equal amounts
of proteins were resolved on SDS/PAGE according to
Laemmli [25] Gels were run in duplicate; one part was used
for staining with Coomassie brilliant blue R250 to ascertain
equal loading of proteins, and the second part was
electrophoretically transferred to nitrocellulose membrane
Equal amount of the proteins on blots was further
confirmed by staining with Ponceau S (Sigma Chemical
Co.) The blots were probed with antibodies raised in rabbit
against purified ZmCCaMK [21] at 1 : 25000 dilution Goat
anti-(rabbit IgG) Ig conjugated with alkaline phosphatase
(Sigma Chemical Co.) was used as secondary antibody and the antigen–antibody complex was visualized by the reac-tion of 5-bromo-4-chloro-3-indocyl phosphate/nitroblue tetrazolium (Sigma Chemical Co.) as described by Harlow
& Lane [26]
Protein purification using immuno-affinity chromatography
For purification of the pea immuno-homologue of ZmC-CaMK, 5-day-old pea plants were treated overnight with CaCl2 (100 mM); roots were harvested and washed exten-sively with water Further steps were performed at 4C unless stated otherwise Tissue was ground in presence of liquid N2, extracted with 3 vol extraction buffer as described earlier and centrifuged at 15000 g for 45 min The crude protein extract obtained was precipitated with 0–40%, 40–50% and 50–80% ammonium sulfate and dialyzed extensively against 50 vols extraction buffer To detect the presence of the ZmCCaMK homologue, all the three fractions were immunoblotted with anti-ZmCCaMK
Ig The 40–50% ammonium sulfate precipitated proteins containing the kinase homologue were utilized for immuno-affinity chromatography
Protein A sepharose purified anti-ZmCCaMK Ig were linked to CNBr-activated Sepharose 6B (Pharmacia Bio-tech) according to the manufacturer’s instructions and the matrix was packed in a 4-mL column The column was equilibrated with buffer containing 50 mM Tris pH 7.5,
10 mM MgSO4, 1 mM phenylmethanesulfonyl fluoride,
1 mMdithiothreitol, 0.1% Triton-X 100, 10% glycerol (v/v) The proteins were bound to the column by recirculating twice through it The bound proteins were washed exten-sively with the same buffer containing 50 mMNaCl until the
A280of the column flow-through reached zero Specifically bound proteins were eluted either with a salt gradient of 0.05–1M in the presence of 50 mM Tris pH 7.5, 10 mM
MgSO4, 1 mM phenylmethanesulfonyl fluoride, 1 mM
dithiothreitol, 10 mM 2-mercaptoethanol, 0.5% Triton-X
100 and 10% (v/v) glycerol or with 2.9 pH glycine buffer [26] When low pH glycine buffer was used for elution, the samples were neutralized immediately using a calibrated amount of Tris pH 7.5
In vitro phosphorylation assays
In vitrophosphorylation assays were performed as described earlier [21] Briefly, protein eluted from the affinity column (1 lg) was incubated in phosphorylation buffer (30 mM
Hepes pH 7.5, 5 mMMgCl2, 0.5 mMdithiothreitol, 25 mM
NaCl) in the absence (control) or presence of 100 lM
calcium and 110 nM CaM The reaction was started by the addition of 100 lM[c-32P]ATP (Amersham Biosciences Corp.) to the reaction mix in a total volume of 50 lL and incubated at 30C for 5 min An equal volume of SDS sample buffer was added to stop the reaction The reaction mix was boiled at 100C for 5 min and resolved by SDS/ PAGE The gel was dried and exposed for autoradiography
To test the effect of KN-62, 50 lMof the compound was also included in the reaction mix To determine if anti-ZmCCaMK Ig blocked the phosphorylation reaction, the protein fraction was first incubated with 1 lg of antibodies
at room temperature for 1 h with shaking The antigen–
Trang 3antibody complex was then precipitated out using protein A
sepharose beads and the resulting supernatant was used for
phosphorylation assay
RT-PCR
Approximately 0.5 g plant tissue (control and stress treated)
was ground in liquid nitrogen and total RNA was extracted
with TRIzol reagent (Life Technologies) according to the
manufacturer’s protocol Total RNA concentration was
determined by UV absorbance at 260 nm For each sample
5 lg total RNA was reverse transcribed with an oligo dT
primer and Superscript II (Life Technologies) according to
manufacturer’s instructions One ll of reaction product was
used as template in PCR reaction with AtCaM5 primers
(forward primer: 5¢-GATGTTGATGGTGATGGTCA-3¢;
reverse primer: 5¢-AAACCAGCCATGAATGAAAT-3¢)
and with actin primers (forward primer: 5¢-GTTGGGAT
GAACCAGAAGGA-3¢; reverse primer: 5¢-GAACCA
CCGATCCAGACACT-3¢) as a control Reactions with
no DNA added served as a negative control The PCR
cycling profile was: denaturation at 92C for 30 s, annealing
at 58C for 1 min and extension at 72 C for 1.5 min for 25
cycles PCR products were analyzed on 1% agarose gels
Preparation of nuclear protein extract, heparin–agarose
chromatography and gel mobility shift analysis
Pea nuclei were isolated as described previously [27] and
purified on a discontinuous percoll gradient For
prepar-ation of nuclear protein extract, the nuclei were washed with
buffer containing 50 mMTris pH 7.8, 5 mMMgCl2, 1 mM
dithiothreitol, 20% glycerol (v/v) and collected by gentle
centrifugation Nuclei were resuspended in washing buffer
containing 110 mM KCl, 10 mM phenylmethanesulfonyl
fluoride, and 5 lgÆmL)1each of antipain and leupeptin and
the suspension was brought to 40% ammonium sulfate
saturation The suspension was centrifuged at 100 000 g for
1 h in a Beckman Ti 75 rotor The supernatant obtained
was brought to 70% saturation and the nuclear proteins
were obtained by centrifugation at 100 000 g for 1 h The
proteins were finally resuspended in the same buffer without
MgCl2and stored frozen in small aliquots at)80 C
A pre-packed 2.5-mL heparin–agarose column (Sigma
Chemical Co.) was equilibrated with 10 column vols of
binding buffer (50 mMTris pH 7.2, 10 mMMgSO4, 1 mM
dithiothreitol, 1 mM phenylmethanesulfonyl fluoride and
10% glycerol, v/v) Ten to 20 mg proteins were loaded on to
the column and washed with binding buffer containing
50 mM NaCl until the A260 of the flow through reached
zero The proteins specifically bound to the column were
eluted with a 0.05 to 1MNaCl gradient in binding buffer
and eluted proteins were tested for binding with the
AtCaM5promoter fragment Active fractions were pooled,
dialyzed and stored in small aliquots at)80 C
Gel mobility shift analysis (GMSA) were performed
according to Ausubel et al [28] either with the labeled
AtCaM5promoter fragment ()588 to )339) or with specific
oligonucleotides designed from the same promoter
frag-ment The sequences of the oligonucleotides used for these
experiments are: Oligo I, 5¢-CAAGGACGTTCGATGCA
CTTCCAAAAAACATATAAT-3¢; Oligo II, 5¢-CAAT
GTAGTATTAAAAAGTAGTAGTTAAAAGC-3¢; Oligo
III, 5¢-GTTTTTATCCGATGCAAATTTTTGCTTTGT GATTG-3¢
The reaction was performed in 20 lL DNA-binding buffer containing (50 mMTris pH 7.4, 50 mMKCl, 1 mM
dithiothreitol, 6% glycerol, v/v) supplemented with
1 lg sonicated calf thymus DNA Labeled probe ( 10 000 c.p.m.) was incubated with the required amount
of protein at room temperature for 10 min DNA–protein complexes formed were fractionated by 5% nondenaturing PAGE and autoradiographed To abolish any protein– protein interaction, 0.5% deoxycholate (Sigma Chemical Co) was incubated in the reaction mixture For supershift analysis, the required dilution of anti-ZmCCaMK Ig was included in the reaction mixture with or without deoxy-cholate Competition analyses were performed by including
1000· concentration of self or nonself oligonucleotides in the reaction mix To test the affect of phosphorylation on binding of p40 with DNA, p40 was phosphorylated using cold ATP as described above and used for assays An identical experiment performed with labeled ATP was run
on a gel to verify the phosphorylation and to confirm the integrity of protein (data not shown) For dephosphoryla-tion 10 lg of phosphorylated protein was incubated in buffer containing 50 mM Hepes pH 7.5, 1 mM MgCl2, 0.5 mMdithiothreitol and calf intestinal phosphatase (10 U)
in a total volume of 50 lL, at 30C for 10 min Reactions were stopped by the addition 5 lL 100 mM sodium pyrophosphate Protein was precipitated using ice-cold acetone and resuspended in DNA-binding buffer to study DNA–protein interaction
For South-Western analysis, proteins separated on 10% SDS/PAGE were electro-blotted on nitrocellulose mem-brane The membrane was blocked with the binding buffer (described above) containing 3% BSA at room temperature with gentle shaking The membrane was washed twice with same buffer containing 0.25% BSA Hybridization was carried out in the presence of 50 lgÆmL)1 sonicated calf thymus DNA and labeled probe at room temperature for
1 h The membrane was washed, dried and exposed for autoradiography
R E S U L T S Immuno-homologue of maize ZmCCaMK is upregulated
by low temperature and salinity stresses in pea roots Western blot analysis using anti-ZmCCaMK Ig of total protein extracts from pea shoots and roots showed the presence immuno-homologue of maize kinase in pea (PsCCaMK) PsCCaMK showed a development-depend-ent and tissue-specific expression (S Pandey & S K Sopory, unpublished data) The level of PsCCaMK was very low in roots as compared to the shoots (Fig 1, control lanes) As some protein kinases are involved in stress signaling pathways [29] and recent work points towards the involvement of Ca2+/CaM-dependent protein kinases in stress signaling [16], changes in the level of PsCCaMK was evaluated under various stress conditions Five-day-old pea seedlings were subjected to temperature, salinity and osmotic stress (as described in Experimental procedures) and the level of PsCCaMK was monitored by Western blotting in both roots and shoots using anti-ZmCCaMK Ig It was seen that PsCCaMK level remained
Trang 4unchanged under all the conditions tested in shoots
(Fig 1) However, a strong upregulation of the protein
level was observed in roots when salt (0.3MNaCl) or low
temperature (4C) treatment was given to the plants
Osmotic stress or heat shock had no effect on the level of
PsCCaMK suggesting that this kinase is not a general
stress-regulated kinase but may specifically be involved in a
signaling pathway associated with salinity and low
tem-perature stress
As both NaCl and low temperature upregulated the
kinase level, time-kinetics experiments were performed for
these stress treatments The optimum concentration of
NaCl required to upregulate the kinase level was also
determined As shown in Fig 2, the level of PsCCaMK
started increasing in response to 50 mM NaCl, reached
maxima at 300 mM and then remained constant up to
500 mMNaCl The kinase level started increasing after 6 h
of treatment of plants with NaCl as well as low temperature
and the maximum level was observed following 24 h of
treatment
Calcium upregulates the PsCCaMK level in a
time-and concentration-dependent manner
Because PsCCaMK showed a strong upregulation in
response to low temperature and salinity stress in pea roots
under in vivo conditions, and the signaling pathways for
both of these stresses are often mediated by Ca2+, the effect
of exogenous Ca2+ was analyzed on the protein level of
PsCCaMK As shown in Fig 3, the level of the kinase was
strongly upregulated by Ca2+ The kinase level started
increasing at 10–25 mMexogenous Ca2+and the maximum level was observed at 100 mM The time-kinetics data showed that the appearance of PsCCaMK after Ca2+
Fig 2 Western blots showing the kinetics of induction of PsCCaMK following NaCl and cold treatment Five-day-old pea plants were given NaCl or cold treatment for indicated time periods and concentrations Roots were harvested and immediately frozen Twenty-five lg total protein extracts were separated by SDS/PAGE, and Western blotting was performed using anti-ZmCCaMK Ig Equal loading of proteins per lane was confirmed by Ponceau S staining of the Western blot Lane C denotes protein isolated from control plants that were not given any stress treatment.
Fig 1 Western blot analysis of level of PsCCaMK in response to
dif-ferent stresses in roots and shoots of pea Five-day-old pea plants were
given heat stress (42 C), low temperature stress (4 C), salt stress
(0.3 M NaCl) and osmotic stress (0.3 M mannitol) for 24 h and roots
and shoots were harvested separately Extracted proteins were
separ-ated by SDS/PAGE (25 lg per lane) and probed with
anti-ZmC-CaMK Ig Roots and shoots from normal vermiculite-grown plants
served as controls Numbers on the left indicate molecular weight
markers in kDa.
Fig 3 Effect of exogenous calcium on the level of PsCCaMK Calcium treatment was given to the 5-day-old pea plants for the indicated time periods and concentration and roots were harvested Total protein extracted (25 lg per lane) was Western blotted and probed with anti-ZmCCaMK Ig Mg denotes plants treated with 50 m M MgCl 2 instead
of CaCl 2 Lane C denotes protein isolated from control plants not given any calcium treatment.
Trang 5treatment was earlier (at 3 h) than that obtained following
NaCl and low temperature stress treatment (at 6 h) To
exclude the possibility of this upregulation being mediated
via a divalent cation in general, the effect of Mg2+was also
tested Plants treated with 50 mM Mg2+for 24 h did not
show any upregulation of the PsCCaMK level
Low temperature, but not salinity-stimulated kinase
level is mediated via a Ca2+/CaM pathway
To further confirm that the upregulation of PsCCaMK by
NaCl and low temperature is mediated via a Ca2+/CaM
signaling pathway, the plants were pretreated with EGTA
(10 mM), lanthanum (10 mM) or W7 (60 lM), before
exposure to low temperature, NaCl or Ca2+ Plants
treated with these compounds but not given any further
stress treatments served as controls along with normal
vermiculite-grown plants The Western blot analysis
showed that EGTA as well as W7 almost completely
blocked low temperature- and Ca2+-induced upregulation
of the kinase but that the NaCl-stimulated kinase level
was unaffected by these treatments (Fig 4) These results
suggest that the low temperature-induced response may be
mediated by Ca2+/CaM whereas the salt-induced
upreg-ulation might be mediated via some other pathway
Lanthanum had no effect on the expression level of the
kinase
Purification of PsCCaMK by immuno-affinity chromatography: a 40-kDa protein always coelutes
To purify PsCCaMK, 5-day-old pea plants were treated overnight with 100 mMCa2+and then roots were excised for purification of protein The total soluble protein extract was fractionated with 40–50% ammonium sulfate and loaded on to an immuno-affinity column prepared using anti-ZmCCaMK Ig The protein was bound in the presence of 50 mM NaCl and eluted either using a salt gradient of 0.05–1M containing 0.5% TritonX-100 and
10 mM 2-mercaptoethanol or with low pH glycine buffer (pH 2.9) Under both of these sets of conditions a 40-kDa protein eluted first from the column followed by elution of the 72-kDa protein corresponding to the PsCCaMK (Fig 5) Under all elution conditions tested, the 72-kDa protein could not be eluted independently of the 40-kDa protein Western blot analysis of the eluted fractions with the same antibodies that were used for making the immuno-affinity column showed cross-reactivity with the 72-kDa PsCCaMK protein only This suggests that the 40-kDa protein does not bind to the antibodies on the column directly but possibly it is very tightly associated with PsCCaMK
PsCCaMK is present in nuclear protein extracts and possibly interacts with DNA
A number of CaM kinases from animal systems have been reported to be present in nuclei and to regulate gene expression [30] Our studies with the AtCaM5 promoter, which is induced under different stress conditions (S B Tiwari & K C Upadhyaya, unpublished data), gave indications that the anti-ZmCCaMK Ig affected binding of the AtCaM5 promoter with specific proteins eluted from
Fig 4 Effect of various pharmacological compounds on the expression
level of PsCCaMK Five-day-old pea plants were pretreated with
EGTA (10 m M ), lanthanum (La, 10 m M ) or W7 (60 l M ) Pre-treated
plants were given low temperature (4 C), NaCl (0.3 M ) or calcium
(100 m M ) treatment for 24 h Pre-treated plants, not given further
treatment, as well as normal vermiculite-grown plants (C) served as
different controls Twenty-five lg total proteins extracted from roots
were separated by SDS/PAGE, Western blotted and probed with
anti-ZmCCaMK Ig.
Fig 5 Purification of PsCCaMK from immuno-affinitycolumns Roots of 5-day-old plants treated with calcium (100 m M ) for 24 h were used for the purification of kinase using the immuno-affinity column prepared using anti-ZmCCaMK Ig Elution was with a 0.05–1 M NaCl gradient The left panel shows the SDS/PAGE profile of proteins from fractions 7, 9 and 11 after silver staining The right panel shows the Western blot of the same fractions using anti-ZmCCaMK Ig.
Trang 6the heparin–agarose column Moreover, expression of
AtCaM5 gene is strongly upregulated in response to
identical conditions of low temperature and salinity stress
as analyzed by RT-PCR (Fig 6A) Taking clues from these
observations we tested for the presence of PsCCaMK in
nuclear protein fractions as well as in the proteins
fractionated on the heparin–agarose column As shown in
Fig 6B, the antibodies cross-reacted with a 72-kDa protein
in the total nuclear protein fractions indicating its possible
nuclear localization The antibodies also cross-reacted with
the nuclear proteins fractionated on the heparin–agarose
column The 72-kDa kinase band could be detected
specifically in the 0.2–0.4M salt-eluted protein fractions
The same fractions also showed binding to the AtCaM5
promoter under different physiological conditions
(unpub-lished data) As the salt concentration used was high
enough not to let nonspecific proteins interact with the
heparin column, it was predicted that this protein might be
interacting with DNA directly To confirm this
observa-tion, we performed GMSA of total pea nuclear protein
extract with a 249-bp labeled AtCaM5 promoter region
()588 to )339) in the presence and absence of
anti-ZmCCaMK Ig This particular region was selected as it
showed maximum protection and structural changes when
foot-printing analysis was performed (S B Tiwari & K C
Upadhyaya, unpublished data) As shown in Fig 6B, a
strong DNA–protein complex was formed with the
AtCaM5promoter fragment and total pea nuclear protein
extract, which showed a supershift with the anti-ZmC-CaMK Ig Addition of 0.5% deoxycholate to abolish any ionic protein–protein interactions showed the presence of two loose complexes Addition of antibodies along with deoxycholate also showed a supershift, giving further indications that the kinase interacts with DNA either directly or it is strongly associated with some DNA binding protein Antibodies alone showed no interaction with DNA We could not determine the direct binding of the purified kinase with DNA as we could not obtain the kinase preparation without p40 under any conditions and attempts to purify the kinase by gel elution gave a very low yield and the eluted protein was highly labile
The 40-kDa protein is phosphorylated by PsCCaMK and binds directly with theAtCaM5 promoter
As p40 remains tightly bound to the PsCCaMK during the purification process, the possibility of it being a substrate for PsCCaMK was tested Further it was also examined if p40 binds with the specific regions of AtCaM5 promoter To ascertain these facts, proteins eluted from the immuno-affinity column were pooled in two different fractions, one containing pure p40 (fraction A) and other containing both the p72 and p40 (fraction B) Both of these fractions were used for in vitro phosphorylation experiments As shown in Fig 7A, no phosphorylation could be detected in fraction A under any of the
Fig 6 RT-PCR, Western blot and supershift analyses (A) RT-PCR of AtCaM5 gene RNA isolated from roots of control, low temperature (4 C) and salinity (0.3 M NaCl) stressed plants was reverse transcribed and amplified with AtCaM5 gene-specific primers PCR with actin primers is included as control (B) Western blot analysis of pea nuclear extracts Five lg pea nuclear proteins (NP) and 2 lg pea nuclear proteins fractionated
on heparin–agarose column (HAFr) were separated by SDS/PAGE in duplicate One set was silver stained while the other set was Western blotted and probed with anti-ZmCCaMK Ig (C) Interaction of pea nuclear proteins with AtCaM5 promoter and supershift analysis The nuclear proteins were used in GMSAs to analyze their interaction with AtCaM5 promoter fragment Analysis was also performed in the presence or absence of 0.5% deoxycholate and anti-ZmCCaMK Ig to determine supershift Antibodies (Abs) alone were included as control.
Trang 7conditions tested, but a Ca2+-dependent, CaM-stimulated
phosphorylation of p40 could be seen in fraction B This
phosphorylation could be specifically blocked by KN-62, a
Ca2+/CaM kinase inhibitor as well as with
anti-ZmC-CaMK Ig Addition of 50 ng fraction B proteins into
fraction A led to phosphorylation of p40 in fraction A
also On longer exposure of the blots a faint signal could
be detected at the 72-kDa position, corresponding to the
autophosphorylated PsCCaMK in fraction B, but no such
signal was observed with fraction A (data not shown),
confirming that p40 has no phosphorylation activity of its
own These results clearly established that the PsCCaMK
has Ca2+/CaM kinase activity and uses p40 as its in vitro
substrate
To ensure the DNA-binding property of p40, both
fractions A and B were used for South-Western analysis
with the AtCaM5 promoter fragment ()588 to )339) that
was used earlier for GMSA A strong signal was
observed at the 40-kDa position with both the fractions,
showing that p40 binds to this promoter fragment
(Fig 7B) To confirm this observation GMSA was
performed with the AtCaM5 promoter fragment and
fraction B, in the presence of excess of calf thymus DNA
As shown in Fig 7C two specific DNA–protein
com-plexes were formed These data showed that p40 binds
directly to the AtCaM5 promoter, but whether
PsC-CaMK binds with DNA directly or through p40 remains
inconclusive
p40 binds to the specificcis-elements in the AtCaM5 promoter region and the binding is affected
by phosphorylation
To further analyze the specific binding of p40 to DNA, GMSAs were performed with fraction A (pure p40) and three specific oligonucleotides (see Experimental proce-dures) These oligonucleotides were designed based on the protected regions of the AtCaM5 promoter fragment ()588
to)339) in the foot-printing experiments (data not shown) p40 showed binding with two of the sequences, Oligo I and Oligo III but not with Oligo II (data for Oligos I and II are shown in Fig 8) The specificity of the binding was further confirmed by competition assays where excess concentra-tions (1000·) of the self-oligonucleotides as well as nonself oligonucleotides were used As shown in the autoradiogram (Fig 8), the DNA–protein complex could not be detected when the self-oligonucleotides were used at higher concen-tration, whereas no such effect could be seen with the nonself oligonucletides
As p40 was found to be an in vitro substrate for the PsCCaMK, the effect of phosphorylation on its DNA-binding property was tested The protein was in vitro phosphorylated using cold ATP and used for DNA binding studies No binding was detected when prephosphorylated protein was used for the DNA binding reaction To confirm the reversibility of the phosphorylation reaction and the associated DNA-binding activity, phosphorylated p40 was
Fig 7 In vitro protein phosphorylation, South-Western and DNA GMSA with pea nuclear proteins fractionated byimmuno-affinitychromatography (A) Protein fractions eluted from kinase antibody affinity column containing 1 lg purified p40 (fraction A) and those containing 1 lg of both p40 and p72 PsCCaMK kinase (fraction B) were used for in vitro phosphorylation in the absence or presence of Ca2+, CaM, kinase antibodies and
KN-62 In one set 50 ng fraction B was added to fraction A (B) Proteins from fraction A and B were separated by PAGE and probed with labeled AtCaM5 promoter fragment in South-Western analysis (C) GMSA was performed in the absence (control) or presence of fraction B proteins with labeled AtCaM5 promoter fragment.
Trang 8treated with calf intestinal phosphatase A DNA–protein
complex could be detected using the dephosphorylated
protein, confirming that p40 could bind to the AtCaM5
promoter only in the dephosphorylated form
D I S C U S S I O N
Role of PsCCaMK in stress signaling
Protein kinases and phosphatases are important
compo-nents of signaling cascades, which by changing the
phosphorylation status of the target proteins transduce
the signal to elicit the final response [31] A number of
studies in recent years have linked the stress signaling with
changes in the calcium level and the upregulation of
various protein kinase transcripts [1–3,32–36] A receptor
protein kinase RPK1 is regulated in response to multiple
stresses and probably has a role at the very beginning of
multiple stress signaling pathways [37] Involvement
of CDPKs in stress signaling has also been shown by
using chimeric gene constructs containing
abscisic-acid-responsive elements and the GFP reporter gene [29] In this
system, transient expression of CDPKs could be observed
in response to various stresses as well as exogenous
calcium A number of kinases of the mitogen activated
protein signaling cascade have also been reported to be
involved in stress signaling pathways [38]
Upregulation of protein kinase transcripts is a complex process, as multiple signals (stress, light as well as phyto-hormones) affect the level of same kinase and in turn different kinases are modulated by the same signal, depending on the specificity of calcium signal and position
of the kinase in the signaling cascade Though the role of some protein kinases is established in the stress signaling pathways, the exact sequence of events that leads to the final gene expression in response to a particular signal is not very well elucidated
We have reported earlier the purification and character-ization of ZmCCaMK and the presence of its immuno-homologues in a variety of other plants [21] including Arabidopsis(data not shown) We now show that the kinase homologue from pea is involved in stress signaling Con-ditions such as 0.3MNaCl and 4C are widely reported to cause severe stress to plants Under these conditions, pea plants showed visible effect of stress in both roots and shoots However, the upregulation of kinase level was observed in roots only Shoots had higher kinase levels compared to roots to begin with, and it was not altered by stress treatment This is similar to the Nicotiana tobacum CDPKl transcript that could not be detected in leaves of the normal plants, but shows a strong upregulation in response
to different stresses [39]
The kinase level also increases in response to exogenous
Ca2+and even though a maximum increase in the kinase
Fig 8 GMSA with p40 with specific cis elements (oligonucleotides) of AtCaM5 promoter Oligonucleotides I and II representing specific cis-regions
of the AtCaM5 promoter were used to study the specific interaction of p40 Proteins from fraction A (see Fig 7) interacted specifically with Oligo I (A) but not with Oligo II (B) For competition a 1000 · excess of self or nonself oligonucleotides were included in the reaction mix Addition of ATP abolished the DNA–protein complex formation with Oligo I The binding with Oligo I was also studied following phosphorylation/dephospho-rylation of p40 (C).
Trang 9level was obtained with 100 mMCa2+, stimulatory effects
are seen at much lower concentrations (10–25 mM) as well
As these treatments were given to whole plants (not to
isolated protoplasts or proteins) for very short duration, the
actual Ca2+uptake would be much lower Moreover, the
low temperature- and Ca2+-induced increase in kinase level
could be blocked by EGTA and W7 (a CaM inhibitor) and
the high Mg2+treatment given for 24 h has no effect on the
kinase level; it is therefore very likely that Ca2+is acting as a
signal molecule However, the possibility that at such high
concentrations Ca2+ might be causing some pleiotropic
effects to cell physiology cannot be ruled out totally
It has been shown earlier that both salt- and low
temperature stress-mediated signaling pathways are
modu-lated via Ca2+[2,40–42] In most of the cases the same
protein is affected in response to both of these stresses, but
proteins affected by low temperature but not by salt or
dehydration and vice versa are also known [34,43,44]
Recently, the role of a novel kinase SOS2 in salt tolerance
has been suggested [36,45] Specific sequence elements have
been identified (e.g DRE) that are important for cold and/
or dehydration responses [46] In some cases different
proteins bind to the same sequence under different
physio-logical conditions to give specific responses [43,47,48]
PsCCaMK does not appear to be a general
stress-responsive kinase, as its expression does not change in
response to mannitol or heat shock As dehydration or
osmotic shock are known to affect the level of proteins
regulated by salt stress, this kinase falls into a different
category The response of PsCCaMK to NaCl is similar to
that of the SOS3 gene [42] and Ca2+appears to be involved
in its regulation by upregulating the protein level; also the
response is faster than that to either NaCl or to low
temperature treatment The differential effects of EGTA
and W7 on low temperature- and salinity-mediated
expres-sion of PsCCaMK indicate that their responses are
medi-ated by different signaling pathways and that Ca2+/CaM
signaling is involved only in the upregulation of PsCCaMK
in response to low temperature To ascertain whether this
kinase acts as a junction point of two different signaling
pathways needs further work
Purification of PsCCaMK and its substrate protein
and their functional characterization
PsCCaMK is present in the nuclear protein fractions as well
as in the protein fractions eluted from the heparin–agarose
column and possibly interacts with the AtCaM5 promoter
In animal systems, a number of CaM kinases have been
identified in the nuclei that are shown to affect the
expression of specific genes by changing the
phosphoryla-tion status of transcripphosphoryla-tion factors [30,49] It has been shown
that plant nuclear extracts could be phosphorylated [50],
showing the presence of kinase(s) in the nuclei In addition,
different signaling pathways point towards the possibility of
Ca2+and CaM acting through the nucleus via specific CaM
kinases [16] However, our study provides strong evidence
of the presence of a CaM kinase and its possible function in
plant nuclei
It has been observed that during purification of
PsC-CaMK using an immuno-affinity column, p40 always
coelutes with the PsCCaMK, even under very stringent
conditions Moreover, p40 does not cross-react with the
kinase antibodies used to make the column, showing that it
is not interacting directly with the antibodies linked with the column but is possibly very tightly associated with the PsCCaMK There are other examples where substrate or interacting protein has been eluted from the affinity column along with the relevant protein [45] In vitro phosphoryla-tion experiments show that p40 could not be phosphoryl-ated on its own, but only in the presence of PsCCaMK, in a
Ca2+-dependent, CaM-stimulated manner The phos-phorylation of p40 could be blocked by KN-62 (a specific CaM kinase inhibitor) as well as the anti-ZmCCaMK Ig (Fig 7) This shows that PsCCaMK properties are similar
to those of ZmCCaMK and p40 is one of its in vitro substrates
To check the interaction of these proteins with DNA, we selected the promoter of the Arabidopsis CaM5 gene (AtCaM5) The reason to use this promoter was that the AtCaM5 gene was strongly upregulated in response to identical conditions of salinity and low temperature stress Besides, PsCCaMK was specifically present in the same fractions of heparin–agarose eluted proteins that show binding with the AtCaM5 gene promoter We also had preliminary data for this promoter, using GMSA and DNase I foot-printing, about the regions of the promoter showing structural changes under different physiological conditions Based on these studies we used the)588 to )399 fragment of this promoter (that showed maximum protec-tion and changes in footprinting analysis) for our studies Detection of DNA–protein complexes by South-Western analysis, using the labeled fragment and the affinity purified protein fraction, shows the binding of the promoter with the p40 protein and not with PsCCaMK (Fig 7) The absence
of binding with PsCCaMK could be due either to the fact that it does not bind to DNA at all, to a weak interaction, or
to a very much lower amount of protein that could not be detected However, GMSAs with the nuclear protein extract and anti-ZmCCaMK Ig showed a supershift though weak, even in presence of high concentration of deoxycholate, indicating the possible interaction of this kinase directly with the DNA (Fig 6B) On the other hand, pure p40 protein showed direct binding with the AtCaM5 gene promoter fragment by GMSA Binding experiments with specific oligonucleotides confirm that the p40 interacts with defined sequence elements of the promoter, and that the binding is highly specific The same oligonucleotides, which do not show binding with p40 when tested for binding with total pea nuclear extract, show a strong binding (data not shown) which further confirms that p40 binds to some specific cis sequences only The question of whether PsCCaMK binds directly or via its association with p40 has not been fully resolved During purification of the protein the association
of the protein kinase and its substrate was not affected even though stringent conditions were used It is possible therefore, that these two proteins physically interact and even in the presence of 0.5% deoxycholate this interaction remains intact In this case the supershift could be a result of antibody interacting with the PsCCaMK bound to p40, which in turn interacts directly with DNA
Of the several strategies that modulate the binding of a protein with its target DNA, to effect transcription, phosphorylation is regarded as one of the major mecha-nisms [51,52] In plants too, a number of studies have shown that phosphorylating the transcription factors affects their
Trang 10binding with DNA [27,53–55] We have found that the
binding of p40 is dependent on the dephosphorylation
status of the protein and the binding is fully abolished once
the protein is phosphorylated
Analysis of the specific sequence elements, with which
the p40 interacts, shows that these elements are not yet
reported in any of the stress responsive genes On
computational analysis, both of the sequences with which
p40 interacts show the presence of a binding site for high
mobility group proteins (HMG boxes) HMG box
proteins have been identified from many plant species
and it has been shown that their binding with DNA is
affected by the phosphorylation status of the proteins [56–
58] In animal systems it has been shown conclusively that
these proteins usually interact with the AT-rich sequences
that show high structural changes [59–62], which is the
case with the oligonucleotides that we used It could
therefore be speculated that p40 is a protein similar to
HMG box proteins; this, however, requires further
char-acterization
Proposed mechanism of action
Based on all these results we propose that p40 is a negative
regulator of the CaM5 gene promoter that may act in a
similar way to the DREAM protein, a negative
transcrip-tional regulator that acts in a calcium-dependent manner
[63] During normal growth and development, p40 is
bound to the promoter When the plant is under salinity or
low temperature stress, the calcium level inside the cell
increases This elevated calcium could then affect the level
and activity of the Ca2+/CaM-dependent protein kinase
homologue, which is either always present in the nuclei or
it gets translocated to the nuclei [55,64] Once activated,
the kinase phosphorylates p40 and as a result its binding
to DNA is abolished and the protein is released from the
DNA We also have preliminary data (not shown) which
show that some other protein of unknown identity binds
to the same sequence elements, after p40 dissociates from
it following phosphorylation There are earlier reports
which show that different proteins, though unrelated, bind
to the same sequence elements, as in case of DREB1A and
DREB2A [48] The present study thus proposes that
protein kinases that are upregulated in response to specific
stress function in the nuclei, might use DNA binding
proteins as substrate(s) and affect their binding property
thereby regulating the expression of stress-induced genes
To be able to apply this statement in a broader
perspec-tive, it would be essential to look for genes for both
CCaMK and p40 to be able to functionally dissect the
underlying mechanistic pathways It would also be
important to look for true functional orthologues of the
kinase and p40 in other plant species, especially in
Arabidopsis where powerful genetic tools are available
for further studies
A C K N O W L E D G E M E N T S
This work was supported partly by internal grants from the
International Center for Genetic Engineering and Biotechnology and
funds from Department of Science and Technology, Government of
India We thank Prof S Assmann for critical reading of the manuscript
and constructive suggestions and B Yadav for technical help.
R E F E R E N C E S
1 Romeis, T., Ludwig, A.A., Martin, R & Jones, J.D (2001) Cal-cium-dependent protein kinases play an essential role in plant defense response EMBO J 20, 5556–5567.
2 Knight, H., Trewavas, A.J & Knight, M.R (1996) Cold calcium signalling in Arabidopsis involves two cellular pools and a change
in calcium signature after acclimation Plant Cell 8, 489–503.
3 Knight, H., Brandt, S & Knight, M.R (1998) A history of stress alters draught calcium signalling pathway in Arabidopsis Plant J.
16, 681–687.
4 Berridge, M.J., Bootman, M.D & Lipp, P (1998) Calcium – a life and death signal Nature 395, 645–648.
5 Sanders, D., Brownlee, C & Harper, J (1999) Communicating with calcium Plant Cell 11, 691–706.
6 Trewavas, A.J & Malho, R (1997) Signal perception and trans-duction: the origin of the phenotype Plant Cell 9, 1181–1195.
7 Trewavas, A.J (1999) Le calcium, c’est la vie: calcium makes waves Plant Physiol 120, 1–6.
8 Pandey, S., Tiwari, S.B., Upadhyaya, K.C & Sopory, S.K (2000) Calcium signalling: Linking environmental signals to cellular functions Crit Rev Plant Sci 19, 291–318.
9 Trewavas, A.J & Malho, R (1998) Ca2+signalling in plant cells: the big network Curr Opin Plant Biol 1, 428–433.
10 Collinge, M & Trewavas, A.J (1989) The location of calmodulin
in the pea plasma membrane J Biol Chem 264, 8865–8872.
11 Corneliussen, B., Holm, M., Watersson, Y., Onions, J., Hallberg, B., Thornell, A & Grrundstrom, T (1994) Calcium/calmodulin inhibition of basic helix-loop-helix transcription factor domains Nature 368, 760–764.
12 Bachs, O., Agell, N & Carafoli, E (1992) Calcium and calmodulin function in the cell nucleus Biochim Biophys Acta 1113, 259–270.
13 Gilchrist, J.S.C., Czubryt, M.P & Pierce, G.N (1994) Calcium and calmodulin binding proteins in the nucleus Mol Cell Bio-chem 135, 79–88.
14 Kocsis, R.D., Rand, M.N., Lankford, K.L & Waxman, S.G (1994) Intracellular calcium mobilization and neurite outgrowth in mammalian neurons J Neurobiol 25, 252–265.
15 Szymanski, D.B., Liao, B & Zeilinsky, R.E (1996) Calmodulin isoforms differentially enhance the binding of cauliflower nuclear proteins and recombinant TGA3 to a region derived from the Arabidopsis CaM-3 promoter Plant Cell 8, 1069–1077.
16 van der Luit, A.H., Olivari, C., Haley, A., Knight, M.R & Tre-wavas, A.J (1999) Distinct calcium signalling pathways regulate calmodulin gene expression in tobacco Plant Physiol 121, 705– 714.
17 Watillion, B., Kettmann, R., Boxus, P & Burny, A (1993) A calcium/calmodulin-binding serine/threonine protein kinase homologs to the mammalian type II calcium/calmodulin-dependent protein kinase is expressed in plant cells Plant Physiol.
101, 1381–1384.
18 Patil, S., Takezawa, D & Poovaiah, B.W (1995) Chimeric plant calcium/calmodulin dependent protein kinase gene with a neural visinin like calcium binding domain Proc Natl Acad Sci USA 92, 4897–4901.
19 Lu, Y.T., Hidaka, H & Feldman, L.J (1996) Characterization of
a calcium/calmodulin protein kinase homologue from maize roots showing light-regulated gravitropism Planta 199, 18–24.
20 Takezawa, D., Ramachandiran, S., Paranjape, V & Poovaiah, B.W (1996) Dual regulation of a chimeric plant serine/threonine kinase by calcium and calcium/calmodulin J Biol Chem 271, 8126–8132.
21 Pandey, S & Sopory, S.K (1998) Biochemical evidence for a calmodulin-stimulated calcium-dependent protein kinase in maize Eur J Biochem 255, 718–726.
22 Wang, W & Poovaiah, B.W (1999) Interaction of plant chimeric calcium/calmodulin-dependent protein kinase with a homologue