Characterization of a protein kinase (FsPK4) with an acidic domain, regulated by abscisic acid and specifically located in Fagus sylvatica L. seeds

ARTICLE IN PRESS Journal of Plant Physiology 163 (2006) 761—769

www.elsevier.de/jplph

Characterization of a protein kinase (FsPK4) with an acidic domain, regulated by abscisic acid and specifically located in Fagus sylvatica L. seeds Jesu `s A. Jime ´nez, Dolores Rodrı´guez, Oscar Lorenzo, Gregorio Nicola ´s,  Carlos Nicola ´s Departamento de Fisiologı´a Vegetal, Centro Hispano-Luso de Investigaciones Agrarias, Facultad de Biologı´a, Universidad de Salamanca, Plaza de los Doctores de la Reina s/n, 37007 Salamanca, Spain Received 3 May 2005; accepted 7 July 2005

KEYWORDS Abscisic acid; Fagus sylvatica; Protein kinase; Seed dormancy

Summary An abscisic acid (ABA)-induced cDNA fragment encoding a putative serine/threonine protein kinase (PK) was obtained by means of differential reverse transcriptasepolymerase chain reaction (RT-PCR). The full-length clone (FsPK4) was isolated from a cDNA library constructed using mRNA from ABA-treated Fagus sylvatica L. seeds. This clone contained the 11 catalytic domains present in all PKs and a highly acidic domain in the C-terminus. By expressing FsPK4 in Escherichia coli as a His tag fusion protein, we obtained direct biochemical evidence supporting Ca2+-dependent kinase activity of this protein. The expression of FsPK4 increased after ABA treatment or warm pretreatment, when seeds are maintained dormant, but decreased and tended to disappear when dormancy was released by stratification or under gibberellic acid (GA3) treatment, and when seeds were artificially dried. Further, FsPK4 transcript expression is tissue specific, and was found to accumulate in ABA-treated seeds rather than in other ABA-treated vegetative tissues examined. These results suggest that the expression of the corresponding protein could be more closely related with the maintenance of seed dormancy than with responses to drought stress mediated by ABA. & 2005 Elsevier GmbH. All rights reserved.

Introduction Abbreviations: FW, fresh weight; IPTG, isopropyl b-thiogalactopyranoside; mc, moisture content; PK, protein kinase; RT-PCR, reverse transcriptase-polymerase chain reaction; SNF, sucrose non-fermenting Corresponding author. Fax: +34 923294682. E-mail address: [email protected] (C. Nicola ´s).

Abscisic acid (ABA) plays a crucial role in the regulation of seed development, maturation and germination, by inducing reserve accumulation, desiccation tolerance, development of dormancy

0176-1617/$ - see front matter & 2005 Elsevier GmbH. All rights reserved. doi:10.1016/j.jplph.2005.07.010

ARTICLE IN PRESS 762 and adaptation to environmental stresses (Bewley, 1997; Leung and Giraudat, 1998). Genetic analysis has identified the key role of ABA in seed dormancy (Koornneef and Karssen, 1994), primarily using Arabidopsis because of its excellent suitability for genetic and molecular studies (Koornneef et al., 1984). However, Fagus sylvatica L. seeds represent a suitable model to study seed dormancy of woody plants exhibiting an especially deep degree of dormancy maintained by ABA and overcome by stratification, gibberellic acid (GA3) or ethephon (compound that releases ethylene) treatment (Nicola ´s et al., 1996; Calvo et al., 2004). Most physiological responses mediated by ABA include changes in gene expression, and many genes and proteins involved in ABA signalling have been identified, although the signal transduction cascades are not yet clearly established (Leung and Giraudat, 1998). However, substantial progress has been made in the characterization of several ABAsignalling molecules including RNA-binding proteins such as HYL1, ABH1 or SAD1 (Lu and Fedoroff, 2000; Hugouvieux et al., 2001; Xiong et al., 2001) and a complex network of positive and negative regulators, including kinases, phosphatases and transcriptional regulators (reviewed by Finkelstein et al., 2002; Abe et al., 2003). In plants, two major classes of protein kinases (PK) have been identified: serine/threonine (ser/thr) and tyrosine PKs (Hanks et al., 1988) although kinases with dual specificity for both, ser/thr and tyrosine have also been described in plants (Stone and Walker, 1995; Lorenzo et al., 2003). Protein phosphorylation controlled by kinases is a reversible process with the capability to respond to different signals, and evidence suggests its involvement in the hormonal regulation of seed dormancy (Trewavas, 1988; Walker-Simmons, 1998; Wagner and Walker-Simmons, 2004). Several ABA and stress-responsive PKs have been described, including the guard cell specific PK, AAPK, essential for ABA-induced stomatal closing (Li et al., 2000), OST1 as a key element mediating stomatal regulation in response to drought (Mustilli et al., 2002), FsPK1 a dual PK up-regulated by ABA and calcium in beech seeds (Lorenzo et al., 2003) or the so-called PKABA1, a PK with an acidic domain (Anderberg and Walker-Simmons, 1992), whose expression is induced by ABA and various environmental stresses in dormant embryos (Holappa and Walker-Simmons, 1995) and suppresses GA-inducible gene expression in barley (Go ´mez-Cadenas et al., 1999, 2001; Yamauchi et al., 2002). Our work is focused on the study of dormancy in F. sylvatica L. seeds and the role of ABA in the expression of specific genes involved. In previous

J.A. Jime´nez et al. reports, we isolated several dormancy-related genes (Nicola ´s et al., 1997, 1998; Lorenzo et al., 2001, 2002, 2003), and in the current experiment, we isolated and characterized a ser/thr PK (FsPK4) induced by ABA, and specifically expressed in ABAtreated seeds. Additionally, we analysed the relationship between loss of dormancy and responses to drought stress in the expression of FsPK4, showing evidence that its function is more closely related with the level of seed dormancy than with responses to stress.

Materials and methods Plant material F. sylvatica L. seeds (beechnuts) were obtained from the Danish State Forestry Tree Improvement Station. Seeds were dried to a moisture content (mc) of 10% and stored at
4 1C in sealed jars. The pericarp was manually removed and seeds were sterilized in 1% sodium hypochlorite before imbibition in sterile water or solutions containing 100 mM ABA or 100 mM GA3. Seeds were maintained in the different media at 4 1C in the dark from 1 to 6 weeks. For the studies concerning loss of seed dormancy and responses to drought stress, seeds with an initial mc of 10% were moistened to a target mc of 34% by adding a known volume of distilled deionized water directly to a known weight of seeds in a plastic bag (Mortensen et al., 2004). Whole seeds (including the pericarp) were pre-chilled or warm pre-treated for 0, 5, 8, 13, 17, 19 and 26 weeks. After each period, one portion of the seeds was fully imbibed and germinated at either 5 or 15 1C directly, while another seed lot received a drying treatment as described previously (Mortensen et al., 2004). Seedlings were obtained from 4-week-stratified seeds sown in a controlled environment chamber under a 12-h-light and 12-h-dark cycle at 15 1C in moist vermiculite, and harvested after 6 weeks. ABA-treated seedlings were watered every 2 d and misted daily with a solution of 100 mM ABA; the corresponding tissues were collected 6 d afterwards. Then, treated or untreated seedlings were separated into roots, leaves and stems. All collected tissues were frozen in liquid nitrogen and stored at
80 1C (Lorenzo et al., 2001).

Differential RT-PCR approach Total RNA from either ABA- or GA3-treated seeds was extracted using the Qiagen pack-500 cartridge (Qiagen, Valencia, CA, USA) following the manufacturer’s protocol. Poly(A+) RNA was purified from

ARTICLE IN PRESS Protein kinase induced by ABA in Fagus each preparation of total RNA by affinity chromatography in oligo(dT)-cellulose columns using the mRNA Purification Kit (Amersham Pharmacia Biotech, Uppsala, Sweden). cDNA was synthesized from 1 mg of poly(A+) RNA prepared either from ABA-treated or GA3-treated seeds using the 1st Strand cDNA Synthesis kit for reverse transcriptasepolymerase chain reaction (RT-PCR) AMV (Roche Diagnostics, Mannheim, Germany) with oligo-p(dT) used as a primer and following the manufacturer’s instruction. Each cDNA was used as a template for a PCR reaction with degenerate oligonucleotides corresponding to two subdomains conserved among the ser/thr PKs (Hanks et al., 1988). The forward primer consisted of a 17-mer of the sequence 50 -GA(T,C)CT(G,T)AA(A,G)CCNGA(A,G)AA-30 corresponding to subdomain VI, and the reverse primer was a 26-mer of the sequence 50 -TC(A,G)GG(A,G) GC(A,G)TAGTACTC(A,T)GG(A,G)GTNCC-30 corresponding to subdomain VIII. The conditions of PCR were as follows: 1 min 94 1C, 2 min 45 1C, 2 min 72 1C for 30 cycles and 10 min 72 1C, using [a-35S] ATP. PCR products were fractionated on a 6% acrylamide gel, dried and exposed to autoradiographic films. A DNA band of 139 bp was amplified using as a template cDNA prepared from ABAtreated seeds. In contrast, this DNA band was absent when cDNA prepared from GA3-treated seeds was used as a template. The 139 bp DNA band was excised from the gel, re-amplified under the same PCR conditions, cloned into the pCR 2.1 vector (Original TA Cloning kit, Invitrogen) and sequenced. As the predicted gene product encoded by this clone revealed homology to ser/thr PKs, we named it FsPK4.

Isolation of the full-length cDNA clone The full-length cDNA clone was isolated from a cDNA library constructed using poly (A+) RNA from seeds imbibed in 100 mM ABA for 2 weeks as a template (Nicola ´s et al., 1997) with FsPK4 PCR fragment as a probe. Double stranded cDNA was synthesized using a cDNA kit (Stratagene, La Jolla, CA, USA) following the manufacturer’s instructions and inserted between the EcoRI and XhoI sites of the Uni-ZAP XR vector (Stratagene). The recombinant cDNA of the clones selected for further analysis (FsPK4) was excised from the phage in pBluescript SK(+) using the biological rescue recommended by the supplier (Stratagene).

DNA sequencing Plasmid DNA templates were isolated by the Wizard Plus Minipreps DNA Purification System

763 (Promega Corporation, Madison, WI, USA). Determination of the nucleotide sequence of the cDNA clone was performed on a ABI 377 sequencer (Applied Biosystems Inc., Foster City, CA, USA) using the Taq DyeDeoxyTM Terminator Cycle Sequencing kit. The DNA and deduced protein sequences were compared to other sequences in the EMBL (GenBank and SwissProt databases, respectively), using the FASTA algorithm.

Heterologous expression of FsPK4 in E. coli A cDNA fragment spanning the entire FsPK4 open reading frame was PCR-generated using the following primers: 50 -primer containing the Nde I site (50 -CATATGGACAAGTACGAGGTG-30 ; the Nde I site is underlined and the initial ATG is bold) and 30 -primer containing the Xho I site (50 -CTCGAGTTAACTGATATGAAA-30 ; the Xho I site is underlined and the stop codon is bold). The PCR product was digested with Nde I and Xho I, and the fragment was inserted in frame into the Nde I and Xho I sites of the pET28a(+) vector (Novagen, Inc., Madison, WI, USA) and verified by DNA sequencing. FsPK4 protein was expressed in E. coli BL21(DE3) as histidine tag fusion protein, cells carrying the recombinant plasmid were grown at 30 1C in 2xYT until A600 reached 0.6 units, and recombinant protein was induced by the addition of isopropyl b-D-thiogalactopyranoside (IPTG) to a final concentration of 1 mM. Cells were harvested by centrifugation at 5000g for 5 min after 5 h of induction and lysed using Bug Buster protein extraction reagent (Novagen). Purification of the fusion protein was carried out using Ni2+-resin (Novagen) according to the manufacturer’s instructions. Proteins were viewed by SDS-PAGE and Coomassie Blue staining.

Assay for protein kinase activity For the PK activity assays, 500 ng of the fusion protein were used in a 20 mL reaction mixture containing 50 mM Tris–HCl pH 7.6, 5 mM MgCl2, 1 mM CaCl2, 4 mCi g-32P dATP, and 10 mg of a-casein (Ali et al., 1994). The reactions were incubated for 15 min at 25 1C. In some assays, staurosporine (10 mM), a ser/thr PK inhibitor was used. Aliquots (2 mL) were removed after the indicated time period and placed on Whatman GF/C filters (Whatman, Kent, UK). The reaction was stopped by washing the filters three times with 200 mL of 20% (w/v) TCA containing 10 mM sodium pyrophosphate and once with 95% ethanol (v/v). Filters were dried and mixed with 1 mL of Ready SafeTM Liquid Scintillation Cocktail (Beckman Instruments, Inc.,

ARTICLE IN PRESS 764 CA, USA) and radioactivity determined in a TriCarbs 2100 TR Liquid Scintillation Analyzer (Packard Instrument Company Inc., Bedford, MA, USA).

Nucleic acid analysis Total RNA from seeds was RNA blot analysed following the same procedures described in Nicola ´s et al. (1997). Blotted membranes were hybridized with a 32P-labelled probe of FsPK4 cDNA clone and membranes were processed as described in Lorenzo et al. (2001) and exposed to X-Omat films (Kodak, Rochester, NY, USA).

J.A. Jime´nez et al. FsPK4 protein includes the 11 conserved catalytic subdomains present in all PKs (Hanks et al., 1988), and an acidic domain in the C-terminus. Comparison of the deduced amino acid sequence of FsPK4 with the data bases (GENEMBL and SWISSPIRALL) showed the 11 catalytic subdomains described in all PKs (Hanks et al., 1988) (Fig. 1A). FsPK4 shares homology with ser/thr PKs with highly acidic domains, such as SPK-3 (88% identity) of Glycine max, OSAK and WAPK (83% and 74% identity, respectively) of Nicotiana tabacum, ASK-1 of Arabidopsis thaliana (82 % identity) and AAPK (74% identity) of Vicia faba (Fig. 1B).

Kinase activity of recombinant FsPK4 protein

Results Isolation and characterization of a cDNA clone from F. sylvatica seeds, coding for a ser/thr PK In a previous report by Nicola ´s et al. (1996) we found that ABA is involved in beechnut dormancy while GA3 is able to release it. As PKs are key elements in ABA signal transduction and play a crucial role as regulators of seed dormancy (Leung and Giraudat, 1998), we attempted to identify PKs induced by ABA, putatively involved in ABA signal transduction and related to seed dormancy. A differential RT-PCR approach was used starting from mRNA extracted of either ABA- or GA3-treated seeds, and using degenerate oligonucleotides corresponding to conserved motifs VI and VIII present in all ser/thr PKs (Hanks et al., 1988). Some cDNA fragments, encoding partial gene products with homology to PKs, were selected and amplified from cDNA of ABA-treated seeds as template, but absent when the template used was cDNA of GA3-treated seeds. One of them was used as a probe to screen a cDNA library constructed from mRNA of ABAtreated seeds (Nicola ´s et al., 1997) and the corresponding full-length clone, named FsPK4, was found and sequenced (accession number AJ586509). It had 1698 bp and contained an ORF of 1077 bp-long. The length of FsPK4 cDNA correlates well with the size of the mRNA determined by RNA blot analysis (1.7 kb approximately). The deduced sequence of the corresponding protein contains 401 amino acids and has a predicted molecular mass of 41.3 kD. As shown in Fig. 1A,

To confirm that FsPK4 protein does indeed show kinase activity, the coding fragment of the clone was expressed in E. coli as His tag fusion protein. The in vitro kinase activity of the FsPK1 fusion protein was assayed in the presence of [g-32P] dATP using a-casein as substrate, and thus, we obtained direct biochemical evidence that FsPK4 encoded a functional PK, showing Ca2+-dependent kinase activity. In the absence of the divalent cation Ca2+, little detectable activity was present, and it was greatly reduced in the presence of 5 mM EGTA (a calcium chelating agent). This kinase activity was almost undetectable when staurosporine (ser/ thr PK inhibitor) was added to the in vitro assay (Fig. 2).

Regulatory effect of ABA on the expression of FsPK4 in dormant beechnuts The effect of ABA on the accumulation of FsPK4 mRNA in F. sylvatica seeds was tested by RNA blot analysis with an specific probe from this cDNA clone (DNA blot analysis indicates that FsPK4 is encoded by a single gene; data not shown). Expression of FsPK4 (Fig. 3B) was inversely correlated with the germination after the indicated treatments (Fig. 3A). Transcript levels are initially low in dormant seeds and increase after 2 weeks imbibition (except in GA treatment). During stratification at 4 1C in water, treatment that produces a slow release of dormancy, and in the presence of GA3, treatment which quickly induces germination of dormant Fagus seeds (Nicola ´s et al., 1996) no expression was detected. The addition of ABA

Figure 1. (A) Schematic diagram of FsPK4 showing catalytic domain with the 11 PK subdomains. The highly acidic domain is boxed. (B) Alignment of the derived amino acid sequences from FsPK4 and SPK-3 (accession number L19361) from Glycine max; OSAK (accession number AY081175) and WAPK (accession number U73938) from Nicotiana tabacum; ASK-1 (accession number M91548) from Arabidopsis thaliana and AAPK (accession number AF186020) from Vicia faba.

ARTICLE IN PRESS Protein kinase induced by ABA in Fagus

765

ARTICLE IN PRESS 766

J.A. Jime´nez et al. maintains the dormant state in F. sylvatica, decreasing the percentages of germination to 10% compared with the control in water. This hormone induces the expression of the FsPK4 transcript to higher levels than the rest of treatments. Although calcium is important for the expression of other F. sylvatica PK (Lorenzo et al., 2003), the addition of calcium does not seem to have any effect on the levels of expression of this clone (data not shown).

Transcript accumulation of FsPK4 in seeds with restricted water content

Figure 2. In vitro kinase activity of FsPK4: (1) Denatured protein used as control; (2) FsPK1 activity with Ca2+ (1 mM) in the in vitro assay; (3) without Ca2+; (4) with EGTA (5 mM)+Ca2+ (1 mM); (5) with staurosporine (10 mM)+Ca2+ (1 mM). Each value represents the average of duplicate assays, and kinase activity is expressed as cpm of 32Pi incorporated per mg of protein kinase.

Figure 3. (A) Germination percentages of beechnut seeds obtained after the indicated treatments. The numbers indicate weeks of imbibition. (B) RNA blot analysis of total RNA isolated from F. sylvatica dormant seeds (D) and seeds imbibed from 1 to 6 weeks at 4 1C in 100 mM ABA, 100 mM GA3 or water. Ten micrograms of RNA were used per lane and hybridized with FsPK4 cDNA probe. Top panel: stained gel showing rRNAs.

We also analysed the expression of FsPK4 in seeds pre-treated under moist pre-chilling (5 1C) or warm (15 1C) conditions at a restricted water content of 34% fresh weight (FW) (Fig. 4). After these pretreatments, some of the seeds were fully imbibed and germinated, and some were dried for 7 d to around 13% mc (Mortensen et al., 2004). FsPK4 mRNA, analysed by RNA blot, declined continuously during the 26 weeks of pre-chilling, rendering the transcript almost undetectable (Fig. 4, lanes 1–5). The warm pre-treatment, which had no effect on dormancy release, did not reduce the expression of FsPK4 (lanes 6–9). The decrease of transcript levels during cold treatment is accompanied by an even higher large decrease in the FsPK4 transcript in that specific time period in seeds after the drying treatment (lanes 10–15). Full germination, also induced a reduction in the transcript (lanes 16–18). In summary, Fig 4 shows that the transcripts were abundant in seeds maintained dormant; they were highly reduced during stratification and after imbibition and germination of seeds and finally no accumulation in the FsPK4 transcript was observed upon drying.

Figure 4. RNA blot analysis of total RNAs isolated from F. sylvatica seeds; subsequent to 0, 5, 13, 17 and 26 weeks of prechilling (5 1C) at restricted water content of 34% mc (FW) (lanes 1–5); in warm pretreated (15 1C) seeds after 8, 13, 17 and 26 weeks (lanes 6–9); in seeds that were artificially dried for 7 d to 13% mc after having been prechilled for 0–26 weeks (lanes 10–15); in imbibed and germinated seeds after having been prechilled for 5, 11 and 17 weeks (lanes 16–18). Ten mg RNA were used per lane and hybridized with FsERF1 cDNA probe. Top panel: stained gel showing RNAs.

ARTICLE IN PRESS Protein kinase induced by ABA in Fagus

Figure 5. Expression pattern of FsPK4 transcripts in different F. sylvatica seed (2 weeks imbibed in ABA) and seedling (6 weeks old). ABA-treated tissues: roots (Ra), stems (Sa), leaves (La), embryonic axes (Aa) and cotyledons (Ca) and untreated tissues: roots (R), stems (St), leaves (L). RNA (10 mg per lane) was used and hybridized with FsPK4 cDNA probe. Top panel: ethidium bromide-stained gel showing rRNAs.

Tissue transcript specificity Expression of the FsPK4 mRNA in different parts of F. sylvatica seeds and seedlings (6 weeks old) was analysed (Fig. 5). The FsPK4 transcript accumulated in the ABA-treated seeds in the cotyledons and embryonic axes. No expression was detected in ABA-treated or untreated seedling tissues (leaves, stems and roots from fully germinated seedlings), indicating that this ABA-induced gene is specifically expressed in embryonic tissue.

Discussion ABA is involved in the control of plant growth and development, including seed dormancy and germination (Bewley, 1997). The mechanism of ABA action is still unknown, but it now seems clear that phosphorylation/dephosphorylation events mediated by specific PKs and protein phosphatases (PPs) play an important role in ABA-regulated processes (Finkelstein et al., 2002). Seed dormancy constitutes an intrinsic impediment to germination. In our own studies, we have been using F. sylvatica as a model to study seed dormancy of woody plants, and have previously reported that these seeds display embryo dormancy, maintained by ABA and overcome by stratification at 4 1C or GA3 treatment. These treatments regulate the expression of a group of seed-expressed genes (Nicola ´s et al., 1996), including a glycine rich-protein (Nicola ´s et al., 1997; Mortensen et al., 2004), a GTP-binding protein (Nicola ´s et al., 1998), two PPs 2C (PP2C) (Lorenzo

767 et al., 2001, 2002; Gonza ´lez-Garcı´a et al., 2003) and a dual PK (Lorenzo et al., 2003), which are upregulated by ABA in hydrated, growth-arrested seeds and seem to be related to dormancy. In the present study, we isolated and characterized an FsPK4 clone using a differential RT-PCR approach to identify PKs whose levels of expression increased after ABA treatment. Furthermore, we also provide evidence that it belongs to the ser/thr PK family. The FsPK4 predicted gene product contains the 11 catalytic subdomains present in all PKs (Fig. 1A) (Hanks et al., 1988), and shows high similarity to different plant PKs (Fig. 1B). The full-length sequence of FsPK4, expressed in E. coli as histidine fusion protein exhibits Ca2+-dependent kinase activity (Fig. 2) and a response to staurosporine, a ser/thr PK inhibitor. An important characteristic of this PK is the presence of a 26 aa region highly acidic. This unusual domain in the C-terminus is characteristic of SNF1 (Sucrose NonFermenting) family of yeast and mammalian PKs, named SNRK1 (SNF-Related Kinases) in plants (Walker-Simmons, 1998). Additionally, these kind of ser/thr PKs, including PKABA1 from wheat (Anderberg and Walker-Simmons, 1992), SPK-3 and SPK-4 from soybean (Yoon et al., 1997) or the PKs isolated in wheat (Hotta et al., 1998) or bean (Li et al., 2000), usually are ABA-induced and may play an important role in ABA and/or stress signalling. Additionally, another member of this PK family, LeSNF4, isolated from tomato (Bradford et al., 2003) could act as an intermediary in ABA and sugar signal transduction, controlling several aspects of plant development, including seed dormancy and germination, indicating the key role of SNRK1 family of kinases, where FsPK4 would be included, in these developmental processes. The expression of FsPK4 was determined in the seed over 6 weeks of imbibition under the different treatments that maintain or eliminate dormancy (Fig. 3), as well as in different tissues of the young seedling (Fig. 5). FsPK4 expression is specifically induced in seeds upon ABA treatment, while stratification or GA treatment decreases the level of transcripts (Fig. 3B). Therefore, expression of FsPK4 negatively correlates with germination and is abolished by treatments that break seed dormancy (Fig. 3A). Moreover, this gene is specifically expressed in ABA-treated dormant seeds (Fig. 5). This data is very interesting since Arabidopsis and soybean orthologues are not seed specific since they are expressed in different tissues (Park et al., 1993; Yoon et al., 1997) Fagus dormant embryos contain high levels of ABA conjugates that are hydrolysed upon imbibition to constitute a potential source of active ABA, and

ARTICLE IN PRESS 768 cold stratification reduces the ABA levels in these seeds by both decreasing ABA synthesis and inducing ABA catabolism (Le Page-Degivry et al., 1997). These data correlate with the absence of FsPK4 in the dormant dry seed, its increase during the first weeks of imbibition in water and its reduction along stratification and disappearance in GA treatment (Fig. 3B), which together with its induction by ABA, indicate that this hormone regulates FsPK4 expression. Additionally, we analyse the relationship between loss of dormancy and responses to drought stress, another important process of plant physiology regulated by ABA, in the expression of FsPK4 (Fig. 4). If the FsPK4 protein is involved in desiccation tolerance, potential accumulation could be regarded as an adaptive response to drying, as seen for dehydration protectants in a number of species (Close, 1996; Greggains et al., 2000); however, no accumulation in the FsPK4 transcript was observed upon drying. The transcript levels are maintained during warm pre-treatment (treatment ineffective in breaking dormancy) and gradually decreases during pre-chilling, which produces a progressive release of seed dormancy (Mortensen et al., 2004). All of our results, taken together, are consistent with an important role for FsPK4 in ABA-regulated seed dormancy and its function is more closely related with the level of seed dormancy than with responses to stress, where ABA also plays an important role.

Acknowledgements This work was supported by grants BFI2003-01755 from the Ministerio de Ciencia y Tecnologı´a (Spain) and SA046A05 from Junta de Castilla y Leo ´n. OL is supported by a ‘‘Ramo ´n y Cajal’’ research contracts.

References Abe H, Urao T, Seki M, Shinozaki K, Yamaguchi-Shinozaki K. Arabidopsis AtMYC2 (bHLH) and AtMYB2 (MYB) function as transcriptinal activators in abscisic acid signaling. Plant Cell 2003;15:63–78. Ali N, Halfter U, Chua NH. Cloning and biochemical characterization of a plant protein kinase that phosphorylates serine, threonine and tyrosine. J Biol Chem 1994;269:31626–9. Anderberg RJ, Walker-Simmons MK. Isolation of a wheat cDNA clone for an abscisic acid-inducible transcript with homology to protein kinases. Proc Natl Acad Sci USA 1992;89:10183–7.

J.A. Jime´nez et al. Bewley JD. Seed germination and dormancy. Plant Cell 1997;9:1055–66. Bradford KJ, Downie AB, Gee OH, Alvarado V, Yang H, Dahal P. Abscisic acid and gibberellin differentially regulate expression of genes of the SNF1-related kinase complex in tomato seeds. Plant Physiol 2003; 132:1560–76. Calvo AP, Nicola ´s C, Nicola ´s G, Rodrı´guez D. Evidences of a cross-talk regulation of a GA 20-oxidase (FsGA20ox1) by gibberellins and ethylene during the breaking of dormancy in Fagus sylvatica L. seeds. Physiol Plant 2004;120:623–30. Close TJ. Dehydrins: emergence of a biochemical role of a family of plant dehydration proteins. Physiol Plant 1996;97:795–803. Finkelstein RR, Gampala SS, Rock CD. Abscisic acid signaling in seeds and seedlings. Plant Cell 2002; 14(Suppl.):S15–45. Go ´mez-Cadenas A, Verhey SD, Holappa LD, Shen Q, Ho THD, Walker-Simmons MK. An abscisic acid induced protein kinase, PKABA1, mediates abscisic acid-suppressed gene expresion in barley aleurone layers. Proc Natl Acad Sci USA 1999;96:1767–72. Go ´mez-Cadenas A, Zentella R, Walker-Simmons MK, Ho THD. Gibberellin/abscisic acid antagonism in barley aleurone cells: site of action of the protein kinase PKABA1 in relation to gibberellin signaling molecules. Plant Cell 2001;13:667–79. Gonza ´lez-Garcı´a MP, Rodrı´guez D, Nicola ´s C, Rodrı´guez PL, Nicola ´s G, Lorenzo O. Negative regulation of abscisic acid signaling by the Fagus sylvatica FsPP2C1 plays a role in seed dormancy regulation and promotion of seed germination. Plant Physiol 2003;133: 135–44. Greggains V, Finch-Savage WE, Quick WP, Atherton NM. Putative desiccation tolerance mechanisms in orthodox and recalcitrant seeds of the genus Acer. Seed Sci Res 2000;10:317–27. Hanks SK, Quinn AM, Hunter T. The protein kinase family: conserved features and deduced phylogeny of the catalytic domains. Science 1988;241:42–52. Holappa LD, Walker-Simmons MK. The wheat abscisic acid-responsive protein kinase mRNA, PKABA1, is upregulated by dehydration, cold temperature, and osmotic stress. Plant Physiol 1995;108:1203–10. Hotta H, Aoki N, Matsuda T, Adachi T. Molecular analysis of a novel protein kinase in maturing rice seed. Gene 1998;213:47–54. Hugouvieux V, Kwak JM, Schroeder JJ. An mRNA cap binding protein, ABH1, modulates early abscisic acid signal transduction in Arabidopsis. Cell 2001;106: 477–87. Koornneef M, Reuling G, Karssen CM. The isolation and characterization of abscisic acid-insensitive mutants of Arabidopsis thaliana. Physiol Plant 1984;61: 377–83. Koornneef M, Karssen CM. Seed dormancy and germination. In: Somerville CR, Meyerowitz EM, editors. Arabidopsis. New York: Cold Spring Harbour Laboratory Press; 1994. p. 313–34.

ARTICLE IN PRESS Protein kinase induced by ABA in Fagus Le Page-Degivry MT, Garello G, Barthe P. Changes in abscisic acid biosynthesis and catabolism during dormancy breaking in Fagus sylvatica embryo. J Plant Growth Reg 1997;16:57–61. Leung J, Giraudat J. Abscisic acid signal transduction. Ann Rev Plant Physiol Plant Mol Biol 1998;49:199–222. Li J, Wang XQ, Watson MB, Assmann SM. Regulation of abscisic acid-induced stomatal closure and anion channels by guard cell AAPK kinase. Science 2000; 287:300–3. Lorenzo O, Rodrı´guez D, Nicola ´s G, Rodrı´guez PL, Nicola ´s C. A new protein phosphatase 2C (FsPP2C1) induced by ABA is specifically expressed in dormant seeds of Fagus sylvatica L. Plant Physiol 2001;125:1949–56. Lorenzo O, Nicola ´s C, Nicola ´s G, Rodrı´guez D. Molecular cloning of a functional protein phosphatase 2C (FsPP2C2) with unusual features and synergistically up-regulated by ABA and calcium in dormant seeds of Fagus sylvatica. Physiol Plant 2002;114:482–90. Lorenzo O, Nicola ´s C, Nicola ´s G, Rodrı´guez D. Characterization of a dual protein kinase (FsPK1) up-regulated by abscisic acid and calcium and specifically expressed in dormant seeds of Fagus sylvatica L. Seed Sci Res 2003;13:261–71. Lu C, Fedoroff N. A mutation in the Arabidopsis HYL gene encoding a ds RNA binding protein affects responses to abscisic acid, auxin and cytokinin. Plant Cell 2000; 12:2351–66. Mortensen LC, Rodrı´guez D, Nicola ´s G, Eriksen EN, Nicola ´s C. Decline in a seed-specific abscisic acid-responsive glycine rich protein (GRPF1) mRNA may reflect the release of seed dormancy in Fagus sylvatica during moist prechilling. Seed Sci Res 2004;14:27–34. Mustilli AC, Merlot S, Vavasseur A, Fenzi F, Giraudat J. Arabidopsis OST1 protein kinase mediates the regulation of stomatal aperture by abscisic acid and acts upstream of reactive oxygen species production. Plant Cell 2002;14:3089–99. Nicola ´s C, Nicola ´s G, Rodrı´guez D. Antagonistic effects of abscisic acid and gibberellic acid on the breaking of

769 dormancy of Fagus sylvatica seeds. Physiol Plant 1996;96:244–50. Nicola ´s C, Rodrı´guez D, Poulsen F, Eriksen EN, Nicola ´s G. The expression of an abscisic acid responsive glycinerich protein coincides with the level of seed dormancy in Fagus sylvatica. Plant Cell Physiol 1997;38:1303–10. Nicola ´s C, Nicola ´s G, Rodrı´guez D. Transcripts of a gene, encoding a small GTP-binding protein from Fagus sylvatica, are induced by ABA and accumulated in the embryonic axis of dormant seeds. Plant Mol Biol 1998;36:487–91. Park YS, Hong SW, Oh SA, Kwak JM, Lee HH, Nam HG. Two putative protein kinases from Arabidopsis thaliana contain highly acidic domains. Plan Mol Biol 1993; 22:615–24. Stone JM, Walker JC. Plant protein kinase families and signal transduction. Plant Physiol 1995;108:451–7. Trewavas AJ. Timing and memory processes in seed embryo dormancy. A conceptual paradigm for plant development questions. Bioessays 1988;6:87–92. Walker-Simmons MK. Protein kinases in seeds. Seed Sci Res 1998;8:193–200. Wagner RL, Walker-Simmons MK. Protein phosphorylation is suppressed when wheat embryos are hydrated and remain growth arrested. Seed Sci Res 2004;14:287–96. Xiong L, Gong Z, Rock CD, Subramanian S, Guo Y, Xu W, Galbraith D, Zhu JK. Modulation of abscisic acid signal transduction and biosynthesis by an Sm-like protein in Arabidopsis. Dev Cell 2001;1:771–81. Yamauchi D, Zentella R, Ho THD. Molecular analysis of the barley (Hordeum vulgare L.) gene encoding the protein kinase PKABA1 capable of suppressing gibberellin action in aleurone layers. Planta 2002;215: 319–26. Yoon HW, Kim MC, Shin PG, Kim JS, Kim CY, Lee SY, Hwang I, Bahk JD, Hong JC, Han C, Cho MJ. Differential expression of two functional serine/threonine protein kinases from soybean that have an unusual acidic domain at the carboxy terminus. Mol Gen Genet 1997;255:359–71.