The PsCIPK gene is intronless and encodes a protein that showed partial homology to the members of CIPK family.. Immu-noprecipitation and yeast two-hybrid analysis showed direct interact
Trang 1components from a legume (Pisum sativum)
Shilpi Mahajan, Sudhir K Sopory and Narendra Tuteja
Plant Molecular Biology, International Centre for Genetic Engineering and Biotechnology, New Delhi, India
In plants, calcium plays an important role in
regula-ting gene expression and many other processes
inclu-ding abiotic stress signalling However, the molecular
mechanisms underlying the role of calcium in cellular
functions are not well established Many external
stimuli including light and various stress factors can
bring out changes in cellular Ca2+ level, which can
affect plant growth and development [1,2] The Ca2+
serves as second messenger and its concentration is
delicately balanced by the presence of ‘Ca2+ stores’
such as vacuoles, endoplasmic reticulum,
mitochon-dria and cell wall Ca2+ signals exhibit a high degree
of specificity and are decoded by Ca2+ sensing
proteins known as Ca2+ sensors, which are small proteins interacting with their target proteins to relay the signal In plant cells many Ca2+ sensors have been identified which include calmodulin (CaM) and calmodulin-related proteins [3,4], Ca2+-dependent protein kinases (CDPKs) [5,6], and calcineurin B-like proteins (CBLs) [4] The first plant CBL to be iden-tified was from Arabidopsis thaliana, and is known
as both AtCBL and ScaBP (SOS3-like calcium-binding protein) [7,8] CBL proteins contain four
Ca2+-binding EF hand motifs [9] and functions by interacting and regulating a group of Ser⁄ Thr pro-tein kinases called CBL-interacting propro-tein kinases
Keywords
abscisic acid; abiotic stress; biotic stress;
calcium sensor CBL; CIPK
Correspondence
N Tuteja, Plant Molecular Biology,
International Centre for Genetic Engineering
and Biotechnology, Aruna Asaf Ali Marg,
New Delhi, 110067, India
Fax: +91 11 26162316
E-mail: narendra@icgeb.res.in
Note
The sequences reported in this paper have
been deposited in the General Bank
database (accession nos AY134619 (pea
CBL cDNA); AY883569 (pea CBL genomic
clone); AY191840 (pea CIPK cDNA).
(Received 7 October 2005, revised 11
December 2005, accepted 19 December
2005)
doi:10.1111/j.1742-4658.2006.05111.x
The studies on calcium sensor calcineurin B-like protein (CBL) and CBL interacting protein kinases (CIPK) are limited to Arabidopsis and rice and their functional role is only beginning to emerge Here, we present cloning and characterization of a protein kinase (PsCIPK) from a legume, pea, with novel properties The PsCIPK gene is intronless and encodes a protein that showed partial homology to the members of CIPK family The recom-binant PsCIPK protein was autophosphorylated at Thr residue(s) Immu-noprecipitation and yeast two-hybrid analysis showed direct interaction of PsCIPK with PsCBL, whose cDNA and genomic DNA were also cloned in this study PsCBL showed homology to AtCBL3 and contained calcium-binding activity We demonstrate for the first time that PsCBL is phos-phorylated at its Thr residue(s) by PsCIPK Immunofluorescence⁄ confocal microscopy showed that PsCBL is exclusively localized in the cytosol, whereas PsCIPK is localized in the cytosol and the outer membrane The exposure of plants to NaCl, cold and wounding co-ordinately upregulated the expression of PsCBL and PsCIPK genes The transcript levels of both genes were also coordinately stimulated in response to calcium and salicylic acid However, drought and abscisic acid had no effect on the expression
of these genes These studies show the ubiquitous presence of CBL⁄ CIPK
in higher plants and enhance our understanding of their role in abiotic and biotic stress signalling
Abbreviations
3-AT, 3-aminotrizole; ABA, abscisic acid; CaM, calmodulin; CBL, calcineurin B-like protein; CDPK, Ca2+-dependent protein kinase; CIPK, CBL interacting protein kinases; DAPI, diamidino-2phenylindole hydrochloride; DTT, dithiothreitol; IPTG, isopropyl thio-b- D -galactoside; SA, salicylic acid; SD, synthetic dextrose; UAS, upstream activating sequences; UTR, 5¢ untranslated region; YPD, yeast extract–peptone–dextrose.
Trang 2(CIPKs) [4,10,11] CIPKs most likely represent
tar-gets of Ca2+ signals sensed and transduced by CBL
proteins
CIPK consist of a catalytic domain (at the N
ter-minus) and a regulatory domain (at the C terter-minus)
that interact with each other to keep the enzyme
inactive (autoinhibition), presumably by preventing
substrate access to the catalytic site [12] CBL binds
to the FISL motif (or NAF domain)) an
autoinhib-itory domain present in the regulatory domain of the
CIPK – and thereby makes the enzyme active by
disrupting the intramolecular domain interaction of
CIPK [12–14] A database search revealed 10
AtCBLs and 25 CIPKs in the Arabidopsis genome
and 10 CBLs and 30 CIPKs in the rice genome
[4,12,15] An analysis of genome evolution suggested
that a large number of gene family members resulted
from segmental duplications [15] Furthermore,
dif-ferential affinities among different AtCBL–CIPK
members have been reported [11,12] For example,
AtCBL1 is known to interact only with a subset of
six CIPKs (AtCIPK 1, 7, 8, 17, 18 and 24) [15] The
multiple combinations of CBL–CIPK complexes
might provide a novel mechanism to integrate and
specifically decode signals in plants [12,13] Recent
studies in Arabidopsis indicated that several such
genes function in stress [12,14–21] Except in
Arabid-opsis and rice the CBL–CIPK pathways have not
been well studied in higher plants
In this report, we describe the cloning and
charac-terization of a novel CIPK and its interacting partner
CBL from Pisum sativum PsCIPK showed
auto-phosphorylation and could phosphorylate pea CBL
and other substrates such as casein The mRNA
lev-els of PsCIPK were coordinately upregulated along
with CBL, in response to various abiotic and biotic
stresses, and to calcium and salicylic acid, but not to
abscisic acid (ABA) or dehydration PsCIPK showed
dual localization (in the cytosol and the plasma
mem-brane) while CBL was localized exclusively in the
cytosol
Results Isolation and sequence analysis of PsCIPK and CBL cDNAs and genomic clones
For cDNA cloning, first partial fragments of 550 bp for PsCIPKand 335 bp for PsCBL were amplified by PCR using double-stranded cDNAs (prepared from mRNA isolated from NaCl-stressed pea seedlings) as template and the degenerate primers, designed from the conserved areas of AtCIPK and AtCBL of Arabidopsis, respect-ively (data not shown) The cDNA clones of CIPK (pBS-PsCIPK) and CBL (pBS-PsCBL) were obtained
by screening the pea cDNA library with respective par-tial DNA fragments as probes Sequence analysis of pBS-CIPK cDNA (Accession no AY191840) shows that
it encodes a full length cDNA, 1842 bp in size with an ORF of 1553 bp, a 5¢ untranslated region (UTR) of
47 bp and a 3¢ UTR of 242 bp including 39 bp of poly(A) tail The PsCIPK ORF encodes a protein of 516 amino acid residues with a predicted molecular mass of
57.9 kDa and pI 8.23 Sequence analysis of pBS-CBL
cDNA (Accession no AY134619) shows that it encodes
a full length cDNA, 972 bp in size with an ORF of
678 bp, a 5¢ UTR of 131 bp and a 3¢ UTR of 163 bp including 20 bp of poly(A) tail The PsCBL ORF encodes a protein of 225 amino acid residues with a calculated molecular mass of 25.9 kDa and pI 4.67 The amino acid sequence alignment of PsCIPK with AtCIPK12, AtCBL19, Gossypium hirsutum (Gh) kin-ase, and AtCIPK18 is shown in Fig 1A The N-ter-minal domain of PsCIPK contains an activation domain starting from the conserved DFG and ending
at APE; the C-terminal domain contains the NAF (FISL) motif (Fig 1A) Phylogenetic analysis indicated 67% sequence identity with AtCIPK12 (Accession
no NP_193605), 66% with AtCIPK24⁄ SOS2-like (AAK26847), and 66% with GhCIPK (AAT64036) (data not shown) The identity of PsCIPK with other AtCIPKs is: 64% with AtCIPK19 (NP199393), and 62% with AtCIPK18 (NP174217) (data not shown)
Fig 1 Multiple amino acid sequence alignment (A) Comparison of predicted amino acid sequences of PsCIPK with AtCIPK12 (Accession no.NP_193605), AtCIPK19 (NP199393), GhCIPK (AAT64036) and AtCIPK18 (NP174217) The activation and NAF domains are shown in the boxes (B) The deduced amino acid sequence of PsCBL is aligned with rice CBL (OsCBL, Accession no AAR01663) and AtCBL3 (AAM91280) The calcium binding domains (EF1–4) and calcineurin A binding domain are shown in the box The dot in the EF1 box repre-sents the modified amino acids alanine (A) as compared to the oxygen containing-calcium binding residue aspartate (D) The conserved dis-tances between EF hands are marked Multiple alignments were performed using CLUSTAL W The program recognizes a consensus residue and based on that residue other amino acids that fall in that consensus position are marked The most identical amino acids at each protein are dark shaded and similar ones are light shaded whereas nonsimilar ones are left unshaded The amino acids marked by red, blue, green and pink lines indicates the putative casein kinase II, protein kinase C, the cAMP- and cGMP-dependent protein kinase and putative tyrosine kinase phosphorylation sites, respectively.
Trang 3B
Trang 4The amino acid sequence alignment of PsCBL with
rice CBL (OsCBL) and Arabibopsis CBL (AtCBL3) is
shown in Fig 1B It lacks the myristoylation site in
the N-terminal sequence PsCBL contains four EF
hand Ca2+-binding domains (Fig 1B) The EF1 shows
variation from the canonical EF hand The amino acid
D at position 1, of EF1 is replaced by amino acid A
(Fig 1B) The EF1 and EF2 are 22 amino acids apart,
whereas EF2 and EF3, and EF3 and EF4 are 25 and
32 amino acids apart, respectively (Fig 1B) The
cal-cineurin A binding domain is also present between
positions 155 and 172 (Fig 1B) Phylogenetic analysis
indicated the identity of PsCBL with OsCBL
(Acces-sion no AAR01663), AtCBL3 (AAM91280), AtCBL2
(AAM65177), AtCBL6 (AAG28400), AtCBL4⁄
SOS3-like (BAD43952), and AtCBL1 (BAC43389) as 92, 90,
89, 71, 68, and 66%, respectively (data not shown)
Genomic organization of PsCIPK and PsCBL
For PsCIPK, a genomic fragment (1.84 Kb) was
amplified by PCR from the pea genomic DNA as a
template with the 5¢ UTR and 3¢ UTR specific primers
of PsCIPK gene As a control the primers were used
to amplify a cDNA fragment of expected size 1.84 Kb
using cDNA as a template To confirm the specificity
of the PCR products a nested PCR (2nd PCR) was
performed using PsCIPK gene-specific internal
prim-ers These fragments were then cloned and sequenced
(data not shown) The same size and sequence of the
genomic fragment of PsCIPK and the cDNA show
that PsCIPK is an intron-less gene
For the PsCBL gene, a genomic fragment (3.22 Kb)
was amplified by PCR using the pea genomic DNA as
a template with the 5¢ UTR and 3¢ UTR specific
prim-ers of the PsCBL gene As a control, the same set of
primers was used to amplify a cDNA fragment of expected size 0.97 Kb using cDNA as a template To confirm the specificity of the PCR products a nested PCR (2nd PCR) was performed using gene-specific internal primers As a result 2.547 Kb genomic and 0.67 Kb (expected size) cDNA fragments were obtained, which were cloned and sequenced (data not shown) The higher size of the genomic fragment of PsCBL as compared to the cDNA indicates that this gene contains introns Sequence analysis of the genomic clone reveals that the PsCBL genomic clone spans 2.547 Kb (Accession no AY883569) (from ATG to TAA) Alignment of the genomic sequence with the cDNA sequence identified eight exons (121, 82, 59, 108,
52, 80, 112, and 58 bp in size) and seven introns (331,
223, 682, 346, 80, 109, and 92 bp in size) (Fig 2A) Two introns of 401 and 81 bp were found localized in the 5¢ UTR region (Fig 2A) Most of the 3¢ and 5¢ splice junctions follow the typical canonical consensus dinucleotide sequence GU-AG found in other plant in-trons Figure 2B shows the genomic organization of AtCBL3 (Accession no AT4G265702) containing seven exons and six introns The sizes of all the exons except exon 5 were found to be mostly conserved between PsCBLand AtCBL3 (Fig 2A and B) The PsCBL gene has an additional splice site at the fifth exon Accord-ingly; there was one intron fewer in AtCBL3 as com-pared to PsCBL (Fig 2A and B) The sizes of introns are not conserved between the two species (Fig 2A and B)
Tissue distribution of PsCBL and CIPK and their copy number in pea genome
The transcript levels of PsCIPK and PsCBL in different tissues of pea were studied by northern
A
B
Fig 2 Genomic organization of PsCBL The schematic representation of the exon–intron organization of genomic PsCBL clone (A) and the Arabidopsis homologue (AtCBL3) clone (B) Closed boxes represent exons, and lines between closed boxes represent introns The dark boxes represent the UTRs The position of ATG and TAA are marked The numbers below the lines and the above boxes indicate the sizes (bp) of introns and exons, respectively.
Trang 5hybridization PsCIPK (1.8 Kb) and PsCBL (1.0 Kb)
were ubiquitously present in all the tissues examined
including root, shoot, tendril and flower, but at
relat-ively higher levels in leaves and roots as compared to
the other tissues (data not shown)
The pattern of Southern genomic hybridization
bands under low (data not shown) and high
strin-gency washing conditions suggests that both PsCIPK
and PsCBL exist as single-copy genes in the pea
genome (Fig 3A and B, respectively) Restriction
enzymes which either had a specific site in the gene
or which had no restriction site were used Some of
the enzymes such as SpeI and XbaI, which had no
recognition site in PsCIPK cDNA and genomic
DNA sequence gave a single band after
hybridiza-tion (Fig 3A, lanes 1 and 2), whereas enzymes such
as BglII and NdeI, which had a single specific site in
the gene gave two bands after hybridization (lanes 3
and 4) However, with HindIII, which has a single
site in the gene towards the 3¢ end (that would
result in 3127 and 94 bp fragments) gave single band
around the 5 Kb region (Fig 3B, lane 6) It is
poss-ible that the second fragment containing a very small
part of the gene did not hybridize under the
condi-tions used Enzymes such as EcoRI, BglII and NdeI,
which had no recognition site in the PsCBL cDNA
and genomic DNA sequence gave a single band after
hybridization (Fig 3B, lanes 3, 4 and 7), whereas
enzymes such as SpeI and SacI, which had a single
specific restriction site in the gene gave two bands after hybridization (lanes 2 and 5)
Expression and purification of PsCIPK and PsCBL The pea cDNA encoding CIPK and CBL were cloned into the expression vector pET28a and the recombin-ant proteins were expressed in Escherichia coli SDS⁄ PAGE analysis showed a highly expressed a
58 kDa additional polypeptide for PsCIPK (Fig 4A, lane 2) and a 26 kDa additional polypeptide for PsCBL (Fig 4G, lane 2) in isopropyl thio-b-d-gal-actoside (IPTG) induced fractions, respectively, as compared to uninduced (lane 1) The recombinant PsCIPK and PsCBL were present in the soluble frac-tions and therefore purified in the soluble form through a single Ni2+–NTA–agarose column chroma-tography step PsCIPK and PsCBL proteins, purified
to near homogeneity, showed a 58-kDa (Fig 4A, lane 3) and a 26-kDa band (Fig 4G, lane 3), respectively
In western blotting, the anti-PsCIPK and PsCBL antibodies detected PsCIPK as a single band of
58 kDa (Fig 4B, lane 2 and 3, respectively) and a single band 26 kDa of PsCBL (Fig 4H, lane 2 and 3)
in the IPTG-induced fraction and in the purified fraction There was no signal in the uninduced frac-tion of PsCIPK (Fig 4B, lane 1) or PsCBL (Fig 4H, lane 1) The purified PsCIPK and PsCBL proteins were also recognized by anti-His antibody (data not shown)
PsCBL encodes a functional Ca2+-binding protein The presence of conserved EF-hand motifs in the predicted protein sequence of PsCBL suggests that it may function as Ca2+-binding protein To check the
Ca2+-binding activity of PsCBL, the purified protein
in two different concentrations (3 and 4 lg) along with positive and negative controls were fractionated
by SDS⁄ PAGE (Fig 4I), electro-blotted onto mem-brane and incubated with radioactive 45CaCl2
(Fig 4J) The same sets of proteins were also spotted
on a membrane (dot blot) and treated as above (Fig 4K) The results show that PsCBL binds to
45Ca2+ (Fig 4J, lanes 2 and 3) The positive control Entamoeba histolytica calcium binding protein (EhCaBP) [22], showed binding to 45Ca2+ (Fig 4J, lane 1), while negative controls (glutathione S-trans-ferase and BSA) lack the binding (Fig 4J, lanes 4 and 5) Similar results were obtained with dot blot analysis (Fig 4K) In Fig 4K, spots 1 and 2 are PsCBL protein (3 and 4 lg), lanes 3 and 4 are the same negative controls and lane 5 is the positive
g 1ed
A CIPK
12.0
0.5
1.0
1.6
2.0
3.0
4.0
5.0
7.0
B CBL
kb kb
12.0
0.5 1.0 1.6 2.0 3.0 4.0 5.0 7.0
g 1ed
1ep
Fig 3 Southern blot analysis of PsCIPK and PsCBL to determine
copy number in the pea genome (A, B) Genomic DNA gel blots
analysis Pea genomic DNA (30 lg) was completely digested with
the enzyme indicated, separated by electrophoresis, blotted and
hybridized with the [a-32P]dCTP-labelled PsCIPK (1.2-Kb fragment
from the 3¢ end containing the 3¢ UTR) (A) and [a- 32 P]dCTP-labelled
PsCBL cDNA (0.97 Kb, full-length) (B), cDNAs as probes Un, Uncut
DNA The DNA size (Kb) is indicated at the left.
Trang 6control The CD spectrum of purified recombinant
PsCBL (1.2 mgÆmL)1) in the presence and absence of
Ca2+ was markedly different (Fig 4L) The spectrum
of PsCBL changed significantly when Ca2+ was either added or depleted by the addition of EGTA (Fig 4L) No significant change in the spectra of the
H
L
K G
Fig 4 Purification of PsCIPK and PsCBL proteins and their activities (A) Induction and purification of overexpressed PsCIPK in E coli is shown on SDS ⁄ PAGE Lane M, Molecular weight marker; lane 1, uninduced; lane 2, IPTG induced; lane 3, PsCIPK protein after Ni 2+ –NTA– agarose column chromatography The protein size markers are indicated at the left side of the gel (B) Western blot analysis of the same protein fractions of lanes 1–3 as shown in panel (A) using polyclonal anti-PsCIPK antiserum (C, D) Autophosphorylation of PsCIPK and phos-phorylation of PsCBL by PsCIPK PsCIPK protein in the presence of Mn 2+ (lane 1), Mg 2+ (lane 2), PsCBL plus Mn 2+ (lane 3), PsCBL plus
Mg2+(lane 4) and casein plus Mg2+(lane 5) incubated with [c-32P]ATP in kinase buffer, electrophoresed on SDS ⁄ PAGE and stained with Coomassie blue (C) followed by autoradiography (D) (E) Phosphoamino acid analysis of PsCIPK autophosphorylation (lane 2) and PsCBL phosphorylation by PsCIPK (lane 1) Positions of phosphoserine (P-Ser), phosphothreonine (P-Thr), and phosphotyrosine (P-Tyr) are marked at right side of autoradiogram (F) Immunodepletion of kinase activity of PsCIPK PsCIPK protein was immunodepleted using PsCIPK anti-bodies Lane 1, Phosphorylation of PsCBL by PsCIPK (control without any IgG); lane 2, PsCIPK pretreated with preimmune IgG; lane 3, PsCIPK pretreated with anti-PsCIPK IgG (G) Induction and purification of overexpressed PsCBL in E coli is shown on SDS ⁄ PAGE Lane M, Molecular weight marker; lane 1, uninduced; lane 2, IPTG induced; lane 3, PsCBL protein after Ni 2+ –NTA–agarose column chromatography The protein size markers are indicated at the left side of the gel (H) Western blot analysis of the same protein fractions of lanes 1–3 as shown in panel (G) using polyclonal anti-PsCBL antiserum (I, J, K) 45 Ca 2+ overlay assay showing that PsCBL is a functional Ca 2+ binding pro-tein (I) PsCBL along with the controls were run on 12% SDS ⁄ PAGE and stained with Coomassie blue Lane 1, EhCaBP protein (positive control); lanes 2 and 3, PsCBL (3 and 4 lg); lanes 4 and 5, GST and BSA (negative controls) Lane M, Pre-stained marker (J) The same sam-ples (as in panel I) transferred to nitrocellulose membrane and assayed by 45 Ca 2+ binding Only PsCBL (lane 2 and 3) and the positive control (lane 1) showed Ca 2+ -binding capability (K) Dot blot analysis of the same protein samples (as in panel I) followed by 45 Ca 2+ overlay assay to confirm the45Ca2+binding data Spots 1 and 2, PsCBL proteins (3 and 4 lg); lanes 3 and 4, negative controls; lane 5, positive control (L)
CD spectra of PsCBL, calcium-bound PsCBL and the calcium-bound PsCBL treated with 1.25 m M EGTA.
Trang 7protein was observed by the addition or depletion of
Mg2+ (data not shown) These results suggest that
PsCBL changes its conformation in a Ca2+
-depend-ent manner
PsCIPK phosphorylates PsCBL at Thr residue(s)
To determine whether PsCIPK is a functional protein
kinase, the autophosphorylation and substrate
phos-phorylation activities of the enzyme were checked by
incubating the enzyme with [c-32P]ATP in the absence
or presence of the substrates After incubation, the
phosphorylation of the proteins was examined by
SDS⁄ PAGE (Fig 4C) followed by autoradiography
(Fig 4D) The result shows that PsCIPK
autophos-phorylated (58 kDa) in the presence of Mn2+
(Fig 4D, lane 1) as well as Mg2+(Fig 4D, lane 2)
The sequence analysis of PsCBL revealed that it has
putative phosphorylation sites (Fig 1B) We therefore
tested whether PsCBL is a substrate for the PsCIPK
enzyme The result shows that PsCBL is
phosphorylat-ed strongly by PsCIPK in the presence of the divalent
cations Mn2+ and Mg2+ (Fig 4D lanes 3 and 4,
respectively) We have shown that PsCIPK also
phos-phorylates casein in the presence of Mg2+ (Fig 4D,
lane 5) CBL has no effect on the autophosphorylation
of CIPK (Fig 4D, lane 3 and 4) PsCIPK
phosphoryl-ated PsCBL even in the absence of exogenous Ca2+in
the reaction buffer (data not shown) This data is
sim-ilar to that reported earlier for AtCIPK1, where no
effect was noted on substrate phosphorylation (MBP
and casein) in the presence or absence of any
exogen-ously supplied Ca2+in the reaction buffer [10]
For phosphoamino acid analysis, the radioactive
autophosphorylated 58-kDa band of PsCIPK and the
26-kDa band of PsCBL from the above gel were
excised, acid hydrolysed and subjected to paper
chro-matography The results show that PsCIPK
phos-phorylates PsCBL at Thr residue(s) (Fig 4E, lane 1)
and also becomes autophosphorylated at its Thr
resi-due(s) (Fig 4E, lane 2) PsCBL did not show any
autophosphorylation, as without any kinase there was
no phosphorylation of CBL (data not shown)
To confirm the phosphorylation activity of
PsCIPK, an immunodepletion experiment was
per-formed as follows Purified PsCIPK was reacted
sepa-rately with IgG purified from the sera of preimmune
rabbit and a rabbit immunized with PsCIPK The
antigen–antibody complex was removed by protein
A-Sepharose The supernatant was analysed for PsCIPK
activity to phosphorylate PsCBL Results revealed
that immunodepletion of PsCIPK in the extract
decreased the phosphorylation of PsCBL significantly
(Fig 4F, lane 3),whereas there was no reduction of PsCIPK activity to phosphorylate PsCBL in the sam-ple treated with preimmune IgG (Fig 4F, lane 2) Lane 1 is the control reaction without the addition of IgG
Regulation of transcript levels of PsCIPK and PsCBL in response to stress
To analyse PsCIPK and PsCBL expression under various abiotic and biotic stresses, 7-day-old pea seedlings were stressed for different times The control plants were grown without any stress treatment Total RNAs were extracted from control and treated shoot tissues and hybridized with PsCIPK (1.2-Kb fragment from the 3¢ end containing the 3¢ UTR) and PsCBL (0.97 Kb, full-length) cDNA probes As shown in Fig 5, the transcript levels of both PsCIPK (panels A,
C, E, G, I, K, and N) and PsCBL (panels B, D, F, H,
J, L, and O) are coordinately regulated following a similar trend Following low temperature treatment the transcripts of both genes started increasing from
9 h, reaching a maximum at 12–24 h (Fig 5A and B) After NaCl treatment the levels increased after 12 h and were maintained high at least until 24 h (Fig 5C and D) In response to wounding stress, both the genes showed an early induction at 3 h; however, the levels decreased by 6 h (Fig 5E and F) The transcript levels in salicylic acid (SA) stress were increased after
8 h of treatment and then decreased at 12 h (Fig 5I and J) Interestingly, the transcript levels of both the genes did not alter in response to dehydration stress (Fig 5G and H) and after the exogenous application
of ABA hormone (Fig 5K and L) As a positive con-trol, an ABA responsive gene PDH45 (see Fig 5 leg-ends) was used The transcript level of PDH45 strongly increased from 12 to 24 h under similar experimental conditions (Fig 5M)
Calcium upregulates PsCIPK and PsCBL in a dose-dependent manner
As PsCIPK and PsCBL are strongly upregulated in response to various abiotic and biotic stresses and as the signalling pathway for these stresses are often mediated
by Ca2+, the effect of exogenous Ca2+was analysed on the transcript levels of both the genes As shown in Fig 5N the transcript level of PsCIPK was upregulated
in response to Ca2+, reaching a maximum at 10 mm and declined at higher Ca2+ concentrations (Fig 5N) The transcript level of PsCBL was strongly upregulated
by Ca2+ The level started increasing at 5 mm of exo-genously supplied Ca2+ and the maximum level was
Trang 8observed at 50 mm and remained constant thereafter
(Fig 5O) To exclude the possibility of this upregulation
being mediated via any divalent cation, the effect of
Mg2+was also tested Plants treated with 50 mm Mg2+ for 24 h did not show any upregulation of transcripts
of either of the genes (Fig 5N and O, second lane)
PsCBL
B
D
F
H
J
L
O
18 S
1.0 kb
18 S
1.0 kb
18 S
1.0 kb
18 S
1.0 kb
18 S
1.0 kb
18 S
1.0 kb
18 S
1.0 kb
PsCIPK
A
C
E
G
I
K
N
18 S
1.8 kb
18 S
1.8 kb
18 S
1.8 kb
18 S
1.8 kb
18 S
1.8 kb
18 S
1.8 kb
18 S
1.8
kb CaCl 2
SALICYLIC ACID
COLD
SALINITY
NaCl )
WOUNDING
DEHYDRATION
ABSICIC ACID
M
Fig 5 Expression pattern of PsCIPK and PsCBL genes in response to various abiotic and biotic stresses The total RNAs were extracted from leaf tissue after the stress treatment The various abiotic stresses used for treatment of pea seedlings were cold (A and B), salinity (C and D), wounding (E and F), drought (G and H), SA (I and J), ABA (K and L) and calcium (N and O) Panel M is the control for ABA responsive gene, PDH45 [35] The RNAs (50 lg) samples were separated by electrophoresis, blotted and hybridized with the [a- 32 P]dCTP-labelled PsCIPK (1.2-Kb fragment from 3¢ end containing the 3¢ UTR) (panels A, C, E, G, I, K and N), and [a- 32 P]dCTP-labelled PsCBL cDNA (0.97 Kb, full-length) probes (panels B, D, F, H, J, L, and O) For each stress examined the upper panel shows the autoradiograph of transcript (1.8 Kb for PsCIPK and 1 Kb for PsCBL), while the lower panel shows the hybridization of same blot with 18S rRNA gene (loading control) In each panel, lane 1 is the control (C) without any treatment while other lanes are the RNAs samples collected after stress treatments at the indicated time points.
Trang 9In vitro interaction of PsCBL with PsCIPK protein
by far-western blotting
As the two genes (PsCBL and PsCIPK) showed a
sim-ilar and synchronized transcript profile, we speculated
that these may interact with each other We studied
the interaction of PsCBL with PsCIPK by the
far-western method (see Experimental procedures) Briefly,
the two proteins and controls were separated by
SDS⁄ PAGE, transferred to nylon membrane and then
renatured on the membrane Next they were incubated
with the second protein PsCBL in the presence or
absence of CaCl2(1 mm), followed by western blotting
with anti-CBL IgG The results of far-western blotting
showed that PsCBL binds to PsCIPK, which was
recognized by anti-CBL IgG (Fig 6B, lanes 1 and 2)
This binding is calcium dependent as no signal was
observed when the experiment was performed in the
absence of calcium (data not shown) As a negative
control, 47-kDa pea helicase (lane 3) and 80-kDa pea
MCM7 protein (lane 4) were used; these produced no
signal Figure 6A, shows a Ponceau-S stained
mem-brane in which lane 1 contains prephosphorylated
CIPK which suggests that CBL can interact with both
phosphorylated and nonphosphorylated forms of
CIPK To further confirm binding, the same
experi-(-Leu,- Trp)
(-Leu,-Trp,-His +
15 mM 3AT.)
YPD MEDIA
β galactosidase assay
Pea CBL
in pGBKT7
1.6
Pea CIPK
in pGADT7
A
C
B
D
58 47 26 80
58 47 26 80
kDa
kDa
0.5 1.0
3
4 5
6
1.6
1.0
2.0 3.0 2.0
Fig 6 Direct interaction of PsCBL and PsCIPK proteins, in vitro, as
well as via a yeast two-hybrid system (A, B) PsCBL interacts with
PsCIPK in vitro PsCIPK prephosphorylated (2 lg, lane 1), PsCIPK
(2 lg, lane 2), pea helicase (PDH47) [36] (6 lg, lane 3, negative
control), pea MCM7 (3 lg, lane 4, negative control) and PsCBL
(5 lg, lane 5) were run on SDS ⁄ PAGE, transferred to PVDF
mem-branes, stained with Ponceau-S (A) The proteins on the same blot
were denatured ⁄ renatured, blocked with BSA, incubated with
1 lgÆmL)1 CBL protein followed by standard western using
anti-PsCBL antibodies (B) (C, D) PsCIPK interacts with anti-PsCBL in vitro.
The same set of proteins as (A) were stained with Ponceau-S (C)
treated as above until the BSA blocking step, and then incubated
with 1 lgÆmL)1of PsCIPK protein and detected with anti-PsCIPK
antibodies (D) (E–K) PsCIPK interacts with PsCBL in a yeast
two-hybrid system (E) The ORF of PsCBL was cloned in pGBKT7 and
checked by restriction (NcoI and EcoRI) to show the insert size of
0.67 Kb (lane 2), lane 1 is the DNA marker (F) The ORF of PsCIPK
was cloned in pGADT7 vector and checked by restriction (EcoRI
and XhoI) to give the insert size of 1.55 Kb (lane 2), lane 1 is DNA
marker (G) Template for panels H–K (H) Phenotype on YPD plate
showing uninhibited growth of all the above (I) Phenotype on
syn-thetic dextrose lacking Leucine and Trytophan (SD –Leu–Trp) plate;
this is selection medium for double transformants (J) Phenotype
on synthetic dextrose lacking Leucine, Trytophan, and Histidine
containing 15 m M 3-AT (SD–Leu–Trp–His+3AT) plate; here growth
represent the interaction of PsCBL with PsCIPK (K) b-galactosidase
filter assay further confirms the interaction The blue colour
repre-sents interaction of both the proteins (PsCBL-CIPK) resulting in the
expression of b-galactosidase reporter gene.
Trang 10ment was performed by incubating the same proteins
on the membrane with the PsCIPK followed by
West-ern blotting with anti-CIPK antibodies (Fig 6C and
D) The results show that PsCIPK can also bind to
PsCBL (Fig 6D, lane 5) PsCIPK did not bind to the
negative controls (Fig 6D) Figure 6C is a Ponceau-S
stained membrane
Interaction of PsCBL with PsCIPK via yeast
two-hybrid system
The complete ORF of PsCBL (678 bp) was cloned into
the NcoI and EcoRI sites of yeast two-hybrid binding
domain vector (pGBKT7) The resulting construct
(pGBKT7-PsCBL or BD-CBL) was verified by
sequen-cing and digestion with NcoI and EcoRI to give a band
of 678 bp (Fig 6E, lane 2) On the other hand the
com-plete ORF of PsCIPK (1.5 Kb) was cloned into the
EcoRI and XhoI sites of yeast two-hybrid activating
domain vector (pGADT7) The resulting construct
(pGADT7-PsCIPK or AD-CIPK) was verified by
sequencing and digestion with EcoRI and XhoI to give
a band of 1.5 Kb on gel electrophoresis (Fig 6F, lane
2) The Saccharomyces cervisiae AH109 cells were
co-transformed with both the constructs (BD-PsCBL plus
AD-PsCIPK) as well as with several combinations of
plasmids which served as controls for this experiment
Interactions between PsCBL and PsCIPK were
deter-mined by growth of the cotransformants on the
selec-tion media of synthetic dextrose (SD) lacking Leu, Trp,
and His (SD-Leu–Trp–His– and containing 15 mm
3-aminotrizole, 3-AT) 3-AT is a competitive inhibitor
of histidine and checks any leaky expression of
histi-dine Yeast cells could survive due to the activation of
the nutritional marker gene HIS3 Activation of the
sec-ond reporter gene (lacZ), was monitored by measuring
b-galactosidase activity
The results are shown in Fig 6G–K Figure 6G is a
template for panels H to K showing the clones
streaked: clone 1, Yeast (AH109) cells containing
co-transformants of BD-PsCBL plus AD-PsCIPK; clone
2, cotransformants of BD-PsCBL and AD vector
alone; clone 3, AD-PsCIPK and BD vector alone;
clone 4, cotransformants of empty AD and BD
vec-tors; clone 5, cotransformants of Pea Gb and Gc
served as a positive control (unpublished data); clone
6, yeast AH109 cells alone All these transformants
and AH109 cells grew on the yeast extract–peptone–
dextrose (YPD) plate (nonselective medium) (Fig 6H)
Except AH109 cells, all of the cotransformants
con-taining AD and BD vectors showed growth on
SD-Leu–Trp– medium (Fig 6I) In a selection medium
lacking Leu, Trp and His (SD-Leu–Trp–His–+ 15 mm
3-AT) only selected clones of cotransformants (BD-PsCBL plus AD-PsCIPK) and the positive control, in which the HIS3 gene was transactivated grew (Fig 6J) This confirmed the interaction of PsCBL and PsCIPK proteins The results from b-galactosidase filter assay
of colonies of cotransformants (BD-PsCBL plus AD-PsCIPK) further confirmed the interaction between PsCBL and PsCIPK (Fig 6K, blue colonies) Domain swapping was also performed in which PsCBL was cloned in pGADT7 and PsCIPK was cloned in pGBKT7 and similar interaction results were obtained (data not shown) The results show that PsCIPK inter-acts with PsCBL in a yeast two-hybrid system A PsCIPK mutant with a deletion in the autoinhibitory (NAF) motif failed to interact with PsCBL thus con-firming the authenticity of these proteins and emphasi-zing the importance NAF in the interaction between them (data not shown)
Localization by immunofluorescence labelling and confocal microscopy
Localization of PsCIPK and PsCBL was analysed by immunofluorescence labelling of tobacco BY2 cells fol-lowed by confocal microscopy Cell cultures were found to better for these studies than a whole-plant system Immunofluorescence labelling of tobacco BY2 cells with anti-PsCIPK (Fig 7B) and anti-PsCBL (Fig 7J) antibodies showed the localization of both proteins in the cytosol of all cells In addition, PsCIPK protein was also localized in the outer membrane (Fig 7B) A single enlarged cell showing PsCIPK localization is shown in Fig 7E, whereas Fig 7M is the single enlarged cell showing PsCBL localization Figure 7A, D, I and L are diamidino-2phenylindole hydrochloride (DAPI) stained cells showing the nuc-leus, and Fig 7C, F, K and N are the merged images
of A and B, D and E, I and J, and L and M, respect-ively
Discussion Nature has developed many pathways for combating and tolerating the the various stress signals that cross-talk with each other The CBL⁄ CIPK pathway is one
of these; it emerged as a novel pathway for deciphering calcium signatures and initiating a series of phosphory-lation cascades This ultimately results in the expres-sion and regulation of stress genes mediating plant adaptation in response to array of stresses The exist-ence of a large family of CIPK and CBL genes has been reported in Arabidopsis and rice [15] However, the details of the role of these proteins and the