Activation of MAPK Kinase 9 Induces Ethylene and Camalexin Biosynthesis and Enhances Sensitivity to Salt Stress in Arabidopsis*

Mitogen-activated protein kinase (MAPK)³ cascades are highly conserved signaling modules in eukaryotes. MAPK cascades are minimally composed of three kinase modules, MAPKKK, MAPKK, and MAPK (called MKKK, MKK, and MPK in Arabidopsis, according to systemic nomenclature), which are linked in various ways to upstream receptors and downstream targets. MAPK cascades play important roles in coordinating stress responses and plant growth and development (1–4). The Arabidopsis genome encodes 60 MKKKs, 10 MKKs, and 20 MPKs (5). The relatively few MKKs imply that various signal transduction pathways converge at the MKK level in MAPK cascades. The 10 MKK genes are classified into four groups according to protein sequence alignments. The MKKs of group A include MKK1, MKK2, and MKK6. MKK1 participates in defense responses (6), whereas MKK2 mediates the regulation of cold and salt stress signaling (7). MKK6 is involved in cell division (8). MKK3, the only member of group B, has roles in jasmonate-mediated developmental signaling and in pathogen defense response signaling (9, 10). MKK4 and MKK5 of group C MKKs mainly function in the stomatal development pathway (11) and defense response signaling pathway (12, 13). Of the four MKKs in group D, MKK7 functions as an inhibitor of polar auxin transport (14) and is also involved in generating the mobile signal of systemic acquired resistance (15). So far, no transcripts have been identified for MKK8 and MKK10 (5, 16). MKK9 is a negative regulator of seed germination under stress (16). Recently, MKK9 was found to activate MPK3/MPK6, which in turn phosphorylates and positively modulates EIN3 stability, leading to transcription of early ERF (ethylene response factor) genes (17).

Phytoalexins, small molecules with antimicrobial activity, are synthesized by plants in response to pathogen attack (18). Camalexin (3-thiazol-2′-yl-indole), the main phytoalexin in Arabidopsis, is induced by a variety of microorganisms (19, 20) as well as abiotic factors (21).

Among the phytoalexin-deficient mutants (PAD1–PAD5), PAD3 is involved in camalexin biosynthesis (22–24). PAD3 encodes a cytochrome P450 monooxygenase 71A15 (CYP71A15) that catalyzes the conversion of dihydrocamalexic acid to camalexin (25, 26). CYP71A13, another cytochrome P450 monooxygenase, has been reported to catalyze the conversion of indole-3-acetaldoxime to indole-3-acetonitrile, the first committed step in camalexin biosynthesis (27). Although these mutants highlight the biochemical steps required for camalexin synthesis, much less is known about the signaling components upstream of the camalexin biosynthesis pathway. Most recently, we have shown that camalexin biosynthesis is regulated by MPK3/MPK6. That result directly linked MAPK signaling cascades to camalexin biosynthesis. However, the upstream MKKs responsible for the MPK3/MPK6-mediated camalexin biosynthesis pathway still need to be clarified. For example, it is unclear why activation of MKK4/MKK5 is less efficient in inducing camalexin biosynthesis in Arabidopsis compared with the tobacco ortholog NtMEK2 (28).

The phytohormone ethylene is also an important component of plant responses to biotic and abiotic stresses (29, 30). The conversion of S-adenosyl-l-methionine to 1-aminocyclopropane-1-carboxylic acid is catalyzed by the enzyme 1-aminocyclopropane-1-carboxylic acid synthase (ACS) and is generally a rate-limiting step in ethylene biosynthesis pathway (31, 32). There is evidence that ACS phosphorylation and dephosphorylation are involved in the regulation of ethylene biosynthesis (33–36). Activation of MPK6, the Arabidopsis ortholog of tobacco salicylic acid-induced protein kinase, has been proved to phosphorylate ACS2/ACS6 and to induce ethylene production in transgenic plants (37).

Here we report the identification of MKK9 as the upstream kinase of both MPK3 and MPK6 in planta. Activation of MKK9 leads to activation of the endogenous MPK3 and MPK6, the subsequent up-regulation of multiple genes in ethylene and camalexin biosynthetic pathways, and induces high levels of ethylene and camalexin accumulation. Mutation of either MPK3 or MPK6 compromises MKK9-induced ethylene and camalexin accumulation. In addition, activation of MKK9 enhances the sensitivity of transgenic seedling to salt stress. Our results suggest that MKK9 plays an important role in ethylene and camalexin biosynthesis and salt stress response in Arabidopsis.

EXPERIMENTAL PROCEDURES

Plant Materials—Arabidopsis thaliana (ecotype “Columbia-0”) wild type, mutants, and transgenic seeds were surface-sterilized. After cold treatment at 4 °C for 2 days, the seeds were germinated and grown on 0.5× MS, 1% agar, 1% sucrose plates.

For transformation and mutants screens, Col-0 wild type seedlings were transferred from plates to soil and grown at 22 °C in a growth room with a 12-h photoperiod at a photon flux density of 100 μE/m^–2 s^–1. To test the expression of transgenes, the fully expended leaves from 4-week-old plants were taken at the indicated times after treatment with 15 μm dexamethasone (DEX) (Sigma). For ethylene and camalexin assays, seedlings from plates were transferred to GC vials with liquid culture medium (0.5× MS medium, 0.025% Mes and 0.25% sucrose, pH 5.6) and grown in a growth room at 22 °C under continuous light at a photon flux density of 70 μE/m^–2 s^–1. Two-week-old seedlings were treated with DEX added directly to the medium at a final concentration of 1 μm. The samples were collected at indicated times, quick frozen in liquid nitrogen, and stored at –80 °C until use.

Tobacco (Nicotiana tabacum cv. Xanthi-nc) plants were grown in soil under the same growth condition as Arabidopsis plants. Six- to 7-week-old plants were used for transient transformation experiments.

Salt Treatment and Pathogen Inoculations—For salt-response assays, 4 days after the seeds were germinated and grown on 0.5× MS, 1% agar, 1% sucrose plates, seedlings were transferred to the 0.5× MS, 1% agar, 1% sucrose plate with 100 mm NaCl in the presence or absence of 0.02 μm of DEX. After an additional 15 days of growth, the photographs were taken, and samples were collected, as described above. Three-week-old soil-grown seedlings were sprayed with 0.02 μm DEX. One hour after DEX treatment, seedlings were inoculated with Pseudomonas syringae pv. maculicola ES4326, and the determination of bacterial growth was as described previously (23).

Cloning of MKKs and MPKs—Total RNA was isolated from samples with TRIzol reagent (Invitrogen). Reverse transcription was performed using (dT)₁₆ as primer and total RNA as template. Moloney murine leukemia virus reverse transcriptase (Promega) was used to reverse transcript the poly(A⁺) mRNAs. The coding regions of MKK9 and MPKs were obtained by PCR. MKK9, MPK3, and MPK6 were cloned into the pEGM-T easy vector, and MPK4 and MPK12 were cloned into pBlueScript vector. Point mutations were introduced with the Quick-Change site-directed mutagenesis kit (Stratagene) and confirmed by sequencing. Primers were as follows: MKK9 (At1g73500), 5′-CATATGGCTTTAGTACGTGAACGTCGTC-3′ and 5′-GTCGACTCAAAGATCTTCCCGGAGAAAAG-3′; MPK3 (At3g45640), 5′-CATATGAACACCGGCGGTGG-3′ and 5′-ACCGTATGTTGGATTGAGTG-3′; MPK4 (At4g01370), 5′-GGATCCATGTCGGCGGAGAGT-3′ and 5′-GTCGACCACTGAGTCTTGAGGATT-3′; MPK6 (At2g43790), 5′-CATATGGACGGTGGTCAGGTCAA-3′ and 5′-TTGCTGATATTCTGGATTGAAA-3′; and MPK12 (At2g46070) 5′-GGATCCATGTCTGGAGAATCAAGC-3′ and 5′-GTCGACGTGGTCAGGATTGAATTTG-3′.

MKK9 gene mutants with an NdeI site added before the first ATG were cloned into a modified pBlueScript vector with a FLAG epitope tag coding sequence at the 5′-end. The NcoI/SalI fragment of FLAG-MKK9 mutants from a modified pBlueScript vector were then ligated in-frame into the pET28a (+) vector. The NdeI/SalI fragment of MPK3 and MPK6 from the pBlueScript vector and the BamHI/SalI fragment of MPK4 and MPK12 from the pGEM T Easy vector were ligated into pET30a(+). All the resultant constructs were transformed into Escherichia coli strain BL21.

The MKK9 mutants with a FLAG epitope tag coding sequence were inserted into the SpeI/XhoI sites of the steroid-inducible pTA7002 binary vector (38). The Ω sequence from tobacco mosaic virus was placed before the FLAG epitope tag coding sequence. All the resultant constructs were electroporated into Agrobacterium tumefaciens strain C58C1.

Preparation of Recombinant Protein—E. coli cells (BL21 strain) transformed with pET28a(+) or pET30a(+) constructs were inoculated in LB broth containing 50 mg/liter kanamycin with 200 rpm shaking until the A₆₀₀ reached 0.6. The culture was incubated at 19 °C for an additional 3 h after the addition of isopropyl d-thiogalactopyranoside to a final concentration of 0.1 mm. The recombinant MPK proteins were affinity-purified using a Ni²⁺-chelating Sepharose Fast Flow (Amersham Biosciences) column, and MKK9 mutant proteins were purified on an anti-FLAG M2 affinity gel column (Sigma).

Kinase Assays—An autophosphorylation assay mixture (30 μl), containing 25 mm Tris-HCl, pH 7.5, 10 mm MgCl₂ or MnCl₂,1 μCi of [γ-³²P]ATP, and 2 μg of purified recombinant FLAG-MKK9 or its mutants derivatives, was incubated at 30 °C for 30 min. The reaction was stopped by the addition of 10 μl of 4× SDS loading buffer. The reaction mixture was heated at 95 °C for 2 min and electrophoresed on 10% SDS-polyacrylamide gels. After electrophoresis, the gel was dried, and the kinase autophosphorylation activities were detected by autoradiography.

Phosphorylation of MKK9 wild type and its mutants on MPKs were determined using His-tagged MPKs (1 μg) as substrates. The reaction was carried out under the same conditions as the autophosphorylation assay, except 0.1 μg of FLAG-MKK9 was used in each reaction.

Immunopreciptation experiments were performed, as described previously (39), with anti-p48N and anti-p44N antibodies. Kinase activities of the immunocomplex were determined by an in-gel kinase activity assay as described previously (12).

Agrobacterium-mediated Transformation—Agrobacterium-mediated transient transformation experiments in tobacco were performed, as described previously (39). The stable transgenic Arabidopsis plants were generated using the flower-dipping method (40). Transgenic plants were selected on 0.5× MS plates with 15 mg/liter hygromycin. The T2 and T3 generations of transgenic plants were used for experiments.

T-DNA Insertional Mutants and Crosses—T-DNA insert mutants of MPK3, MPK6, MKK9, and PAD3 were obtained from the Arabidopsis Biological Resource Center (mpk3, SALK_100651; mpk6, SALK_127507; pad3, SALK_026585; and mkk9, SAIL_060_H06). Homozygous null mutants were screened by genomic PCR and confirmed by RT-PCR. mpk3, mpk6, and pad3 were crossed into the MKK9^DD transgenic background. The MKK9^DD allele was selected for hygromycin resistance and transgene expression. mpk3, mpk6, and pad3 were identified by PCR of genomic DNA. Homozygous MKK9^DD/mpk3, MKK9^DD/mpk6, and MKK9^DD/pad3 F3 plants were used for experiments.

Protein Extraction—Total protein was extracted from samples by grinding in extraction buffer (100 mm HEPES, pH 7.5, 5 mm EDTA, 5 mm EGTA, 10 mm Na₃VO₄, 10 mm NaF, 50 mm β-glycerophosphate, 10 mm dithiothreitol, 1 mm phenylmethylsulfonyl fluoride, 5 μgml^–1 leupeptin, 5 μgml^–1 aprotinin, 5% glycerol). After centrifugation at 18,000 × g for 30 min, supernatants were transferred to clean tubes and stored at –80 °C until needed. For total protein extractions from tobacco leaf tissues, the supernatants from the first centrifugation were quickly frozen in liquid nitrogen and centrifuged at 18,000 × g for another 20 min. The concentrations of protein extracts were determined using the Bio-Rad protein assay kit (Bio-Rad) with bovine serum albumin as a standard.

Immunoblot Assay—For immunoblot assays, 7.5 μg of total protein or 0.25 μg of recombinant protein per lane was separated on 10% SDS-polyacrylamide gels, and the proteins were transferred to nitrocellulose membranes. The membranes were blocked for 1 h in TBST buffer (20 mm Tris-HCl, pH 7.5, 100 mm NaCl, and 0.05% Tween 20) with 5% no-fat dried milk at room temperature. The membranes were then incubated with the primary antibody (1:10,000). After three washes with TBST buffer, the membranes were incubated with horseradish peroxidase-conjugated goat anti-mouse antibody (1:10,000). The membranes were visualized by using an enhanced chemiluminescence kit (Roche Applied System). The primary antibodies used were anti-FLAG M2 antibody (Sigma) and anti-6× His tag antibodies (R & D Systems).

LC-MS Analysis—Arabidopsis seedlings in GC vials were treated with DEX. Eight hours after treatment, the culture medium was collected. The culture media were extracted with 2 volumes of chloroform, and then the chloroform was completely evaporated. The remaining residue was dissolved in a small volume of methanol. LC-MS was performed using a 1100 Series LC/MSD Trap VL (Agilent Technologies). An XDB C18 column (Agilent 5 μm, 4.6 × 150 mm) was used at a flow rate of 1 ml min^–1. The mobile phase was methanol/H₂O (9:1, v/v), and the column temperature was kept at 20 °C. The mass spectrometer was run in electrospray ionization mode. A camalexin standard was used as a control.

Measurement of Ethylene and Camalexin Production—The GC vials containing Arabidopsis seedlings were flushed and capped immediately after treatment with 1 μm DEX. At the indicated times, ethylene levels in the GC vials were determined by gas chromatography as previously described (41). For inhibitor treatments, inhibitors (AVG in 2 μm, CoCl₂ in 100 μm, and Ag⁺ in 100 μm) were added to GC vials 1 h before the DEX addition, respectively. The camalexin in the culture medium was measured using an HITACHI F4500 spectrofluorometer (315 nm excitation and 385 nm emission), and the concentration of camalexin was determined by comparison with a camalexin standard curve.

Growth Response Assays—For etiolated seedling growth assays, seedlings were grown for 3 days in the dark at 23 °C, on 0.5× MS plates supplemented with 0.02 μm DEX alone or plus 2 μm AVG or 100 μm Ag⁺ as indicated. To synchronize germination, surface-sterilized seeds were cold-treated at 4 °C for 4 days in the dark.

Quantitative RT-PCR Analysis—RT was performed with 2.4 μg of total RNA and 0.6 μg of (dT)₁₆ primer and Moloney murine leukemia virus reverse transcriptase (Promega). Aliquots of the products of RT reaction were used as templates for quantitative PCR (Q-PCR). Q-PCR was run on a DNA Engine Optican II system (MJ Research) as described previously (42). After normalization to a Ubiquitin5 control, the relative levels of gene expression were calculated. Primers used for Q-PCR were as follows: UBQ5 (At3g62250), 5′-CTCCTTCTTTCTGGTAAACGT-3′ and 5′-GGTGCTAAGAAGAGGAAGAAT-3′; ACS2 (At1g01480), 5′-ACATCGCTAATTTCCAAGACTACCA-3′ and 5′-CTCTATCAAATGCGGCATAGTACG-3′; ACS6 (At4g11280), 5′-CCGGGAATGTTTGAAGTCTCTTG-3′ and 5′-CGGTCTTAAGTCTGTGCACGG-3′; ERF1 (At3g23240) 5′-CGGCGGAGAGAGTTCAAGAGTC-3′ and 5′-TCCCACTATTTTCAGAAGACCCC-3′; ERF2 (At5g47220), 5′-CGGACTCCTCAAAGATGCCT-3′ and 5′CTCTGCCTCACTCCTCTGTAATG-3′; ERF5 (At5g47230) 5′-GTCTCAAATTCGTAAACCGCCA-3′ and 5′-CCCCACGGTCTTTGTCTTACTC-3′; ERF6 (At4g17490), 5′-GGAGAAGAGGCATTACAGAGGAGTG-3′ and 5′-TCGTAGTCTAAACGCTTCTTTGTCG-3′; CYP79B2 (At4g39950), 5′-ACCCACCGTAGAAGATGTAGAGCAC-3′ and 5′-CATTACAAGCTCCTTAATGGTGGG-3′; CYP79B3 (At2g22330), 5′-CTTACGGGATTGGATCTAAACGGAC-3′ and 5′-GCCATTACAAGTTCCTTAATGGTTGG-3′; CYP71A13 (At2g30770) 5′-TAAAGAGGTGCTTCGGTTGC-3′ and 5′-TATCGCAGTGTCTCGTTGGA-3′; PAD3 (At3g26830), 5′-CTAAAGGCTGAAGCGGTCATAAGAG-3′ and 5′-GCTACAAATATGTCCGAGATCATTCC-3′; COR47 (At1g20440), 5′-GAAACCTCAAGAGACAACGACGC-3′ and 5′-CATCGCTCGAAGAGGAAGAAGAG-3′; RD29A (At5g52310), 5′-GCAGCACCCAGAAGAAGTTGAAC-3′ and 5′-GTGTAATCGGAAGACACGACAGG-3′; MKK4 (At1g51660), 5′-ATGAGACCGATTCAATCGCCTC-3′ and 5′-CTATGTGGTTGGAGAAGAAGACGAG-3′; and MKK5 (At3g21220), 5′-GAAACCGATTCAATCTCCTTCTGG-3′ and 5′-CTAAGAGGCAGAAGGAAGAGGACG-3′.

RESULTS

MKK9 Is an Upstream MAPKK of MPK3 and MPK6—MAPKKs are activated through the phosphorylation of two Ser/Thr residues in a conserved (S/T)XXX(S/T) motif between kinase subdomains VII and VIII by MAPKK kinases (43). Phospho-mimetic mutation of the two S/T residues in the kinase catalytic domain to Glu or Asp results in a constitutively active MAPKK (44, 45). Conversely, an Lys to Arg mutation of a conserved residue in the ATP-binding site results in an MAPKK that is constitutively inactive. To facilitate the study of MKK9, we created an active allele (MKK9^DD) by mutating Thr-201 and Ser-205 to Asp and an inactive allele (MKK9^KR) by mutating Lys-76 to Arg. The activities of FLAG-tagged recombinant proteins were then determined in autophosphorylation assays. As expected, MKK9^KR had no detectable autophosphorylation activity. Both MKK9^DD and MKK9^WT were found to have autophosphorylation activity in the presence of Mg²⁺ or Mn²⁺ ions (Fig. 1A). The autophosphorylation activity of MKK9^WT places MKK9 in the autoactive class of MAPKKs (46, 47).

To identify possible downstream targets of MKK9, we performed in vitro phosphorylation assays using various His-tagged MPKs as substrates. MKK9^DD and MKK9^WT phosphorylate HisMPK3 and HisMPK6 but not HisMPK4 and HisMPK12. The phosphorylation of HisMPK3 and HisMPK6 by MKK9^DD and MKK9^WT enhanced the kinase activity of the MPKs as detected in a subsequent in-gel kinase assay (Fig. 1B).

To test whether MPK3 and MPK6 are substrates of MKK9 in planta, we assayed MAPK activity in the leaf extracts of tobacco plants transient-transformed with MKK9 mutant genes, and in Arabidopsis transgenic plants carrying MKK9 mutant genes under the control of a steroid-inducible promoter (12, 28). Expression of either MKK9^DD or MKK9^WT in tobacco leaves activated the endogenous WIPK (wounding-induced protein kinase) and SIPK (salicylic acid-induced protein kinase) proteins, as proved by pulldown experiments with SIPK- and WIPK-specific antibodies (Fig. 1C). On the other hand, the expression of the inactive mutant, MKK9^KR, resulted in very low SIPK activity. Stable expression of either MKK9^DD or MKK9^WT in Arabidopsis transgenic plants activated both MPK3 and MPK6, as demonstrated in kinase activity assays with the double mutants of MKK9^DD/mpk3 and MKK9^DD/mpk6 (Fig. 1D). These results suggest that MKK9 is an upstream MAPKK for MPK3 and MPK6 in planta. The results are consistent with the conclusions of Yoo et al. (17) obtained from protoplast assays.

Activation of MKK9-induced Ethylene Biosynthesis in Arabidopsis—Because activation of MPK6 is known to increase the biosynthesis of ethylene through the phosphorylation and stabilization of ACS2/ACS6, we assayed ethylene production in MKK9 transgenic plants. Five independent transgenic lines for each MKK9 mutant construct were used for ethylene measurement. As shown in Fig. 2A, MKK9^DD and MKK9^WT plants produced high levels of ethylene 12 h after the induction of transgene expression, whereas MKK9^KR and vector control plants did not produce any detectable ethylene. Activation of MKK4 and MKK5, two other MKKs in Arabidopsis, has been reported to activate endogenous MPK3/MPK6 (12, 17). Therefore, we measured ethylene levels in MKK4^DD and MKK5^DD (active forms of the two MKKs) plants. As shown in Fig. 2A, activation of the MKK4 or MKK5 leads to lower levels of ethylene production than does activated MKK9.

Activation of MKK9 Induces Camalexin Biosynthesis in Arabidopsis—After measurement of ethylene concentrations, the culture media were collected and extracted with chloroform. After evaporation of chloroform, the residue was redissolved in methanol and used to perform an LC-MS. The supplemental Fig. S1 shows that a chemical with molecular mass of 200.6 daltons was present in the methanol extract at 92% purity. The molecular mass of this compound is identical with that of the camalexin standard. Further analysis revealed that the chemical has the exact same fluorescence excitation and emission spectra as the standard camalexin (data not shown). Based on these results, we conclude that the chemical that accumulates in the culture media of MKK-activated transgenic seedlings is camalexin.

To compare camalexin production among different MKK9 mutant transgenic plants, five independent transgenic lines for each DEX-inducible construct were used for camalexin measurement. MKK9^DD and MKK9^WT plants produced high levels of camalexin after the induction of transgene expression, whereas MKK9^KR and vector control plants did not produce detectable camalexin. MKK4^DD and MKK5^DD plants produced lower levels of camalexin after the transgene induction (Fig. 2B). Camalexin has been shown to inhibit the growth of some plant pathogens (20, 48–50). Here we did not observe changes in resistance to P. syringae in MKK9 mutant transgenic plants (supplemental Fig. S2).

MKK9-induced Ethylene and Camalexin Biosynthesis Requires the Activation of Both MPK3 and MPK6—To determine whether the endogenous MAPKs activities are required for the ethylene and camalexin induction, we crossed MKK9^DD into mpk3 and mpk6 T-DNA insertion mutants. In MKK9^DD/mpk6 plants, the production of ethylene was dramatically reduced after MKK9^DD protein induction, and in MKK9^DD/mpk3 plants ethylene production was reduced to one-third the amount as compared with the MKK9^DD plants (Fig. 3A). Camalexin production in MKK9^DD/mpk3 and MKK9^DD/mpk6 plants was also greatly reduced (Fig. 3B). In-gel kinase assays showed undetectable MPK3 activity in MKK9^DD/mpk3 plants and MPK6 activity in MKK9^DD/mpk6 plants, respectively (Fig. 3C). These results suggest that the full activation of both MPK3 and MPK6 is required for MKK9-induced ethylene and camalexin biosynthesis.

Multiple Genes in the Ethylene Biosynthesis Pathway Are Induced by the Activation of MKK9-MPK3/MPK6—The rate-limiting step of ethylene biosynthesis is controlled by ACS (31, 51). The basal activity of ACS is very low in healthy plants, which do not produce significant amounts of ethylene under normal conditions. Therefore, the enhanced ethylene biosynthesis is thought to be associated with an increase in the activity of ACS proteins caused by the post-transcriptional regulation or the activation of ACS genes (31, 37, 51, 52). To reveal the correlation between the MKK9-induced ethylene production and the transcription of ACS genes, transcript levels of ACS genes were monitored in the MKK9 transgenic seedlings. We chose to monitor ACS2 and ACS6 expression by Q-PCR. The transcription of ACS2 and ACS6 was strongly induced in MKK9^DD plants after DEX treatment. Transcript levels of ACS6 began to increase 2 h after DEX treatment and peaked at 4 h. Transcript levels of ACS2 began to increase 4 h after DEX treatment and peaked at 8 h. The relative levels of induction were over 190-fold for ACS6 within 4 h and over 30-fold for ACS2 within 8 h of DEX treatment (Fig. 4). These results imply that induction of ACS genes contribute to the increased ethylene production caused by activation of MKK9. The induction of ACS gene expression was partially compromised in MKK9^DD/mpk3 and MKK9^DD/mpk6 seedlings.

Ethylene Response Factor Genes Are Induced by the Activation of MKK9—Recently, Yoo et al. (17) reported the unexpected result that activation of MKK9-MPK3/MPK6 promotes ethylene-insensitive 3 (EIN3)-mediated ethylene signaling through the direct phosphorylation and stabilization modulates of EIN3. However, these authors did not report a strong increase in ethylene production, as described above. To clarify the contribution of the elevated ethylene to ethylene signaling, we tested the ethylene response in MKK9 mutant transgenic plants. We show that the etiolated MKK9^DD seedlings display inhibition of hypocotyl and root elongation in the presence of DEX, whereas etiolated MKK9^KR seedlings grow normally (Fig. 5, +DEX). However, the apical hook did not display exaggerated curvature typical of an ethylene response. On the other hand, when DEX was combined with the ethylene biosynthesis inhibitor, AVG (+ AVG + DEX) or the receptor antagonist Ag⁺ (+Ag⁺ + DEX), hypocotyls of the etiolated seedlings grew as well as the untreated controls (–DEX) and MKK9^KR controls (Fig. 5, bottom panel). These results suggest that the ethylene response in etiolated MKK9^DD seedlings after induction of transgene expression is most likely because of the elevated ethylene, as the ethylene response could be blocked by ethylene receptor agonist.

ERFs (ethylene response factors) are reported to be downstream responsive genes of ethylene signaling (53, 54). The transcript levels of ERF1, ERF2, ERF5, and ERF6 were selected to monitor the ethylene response in MKK9^DD transgenic plants. All four ERFs were significantly induced in MKK9^DD plants after DEX treatment (Fig. 6). The relative levels of induction were over 55-fold for ERF1, over 15-fold for ERF2, and 50-fold for ERF5 within 8 h, and over 65-fold for ERF6 within 2 h of DEX treatment. AVG reduced but did not abolish the ERFs induction, which may be due to residual levels of ethylene. However, the induction of all four tested ERFs was completely blocked by the addition of Ag⁺. These results are consistent with those of Fig. 5 and suggest that the MKK9-induced ethylene response is caused by increased ethylene signaling through the ethylene receptor.

Activation of MKK9 Induces the Coordinated Transcription of Genes in the Camalexin Biosynthetic Pathway—We used Q-PCR to detected the mRNA transcript levels of genes, including CYP79B2, CYP79B3, CYP71A13, and PAD3, which are involved in the camalexin biosynthesis pathway (26, 27, 55). In MKK9^DD plants, transcript levels of all four genes that were coordinately up-regulated peaked 8 h after DEX treatment (Fig. 7). The relative levels were over 800-fold for CYP79B2, 10-fold for CYP79B3, over 2,800-fold for PAD3, and over 3,500-fold for CYP71A13 within 8 h of DEX treatment. The strong induction of PAD3 and CYP71A13 suggests that these genes are involved in the camalexin biosynthesis induced by MKK9. The induction of these genes was partially compromised in MKK9^DD/mpk3 and MKK9^DD/mpk6 plants.

Activation of MKK9-MPK3/MPK6 Independently Activates Camalexin and Ethylene Biosynthesis—Because both ethylene and camalexin biosynthesis pathways are activated by the MKK9-MPK3/MPK6 cascade, we were interested to see if the ethylene is required for camalexin biosynthesis. Application of CoCl₂ and AVG, two commonly used ethylene biosynthesis inhibitors, effectively inhibited MKK9^DD-induced ethylene biosynthesis (Fig. 8A). However, the kinetics of camalexin production after treatment with CoCl₂ and AVG were comparable with untreated controls (Fig. 8B). The product PAD3 catalyzes the last step of camalexin biosynthesis. We crossed the MKK9^DD transgene into the pad3 mutant plant, which produces a small amount camalexin in response to pathogen infection. Fig. 9 shows that the MKK9^DD/pad3 plants produce very low levels of camalexin but normal levels of ethylene, as by compared with MKK9^DD plants. These results agree with the conclusion that there is no cross-talk between the camalexin and ethylene pathways induced by pathogen infection (20).

MKK9 Regulates the Salt Stress Response in Seedlings—Because MAPK pathways have associated responses to various stimuli, the sensitivity of mkk9 and MKK9 mutant transgenic seedlings to NaCl was tested. As shown in Fig. 10, MKK9^DD and MKK9^WT seedlings had enhanced sensitivity to NaCl compared with MKK9^KR seedlings after transgene induction, whereas mkk9 seedlings were more tolerant to NaCl than the Col-0 wild type control.

Furthermore, we measured the transcription of RD29A and COR47, two marker genes for multiple stress responses, in MKK9 transgenic seedlings 8 h after the transgene induction and in mkk9 seedlings grown on the 100 mm NaCl plate for 15 days. The results showed that expression of both MKK9^DD and MKK9^WT leads to a reduction in RD29A and COR47 transcription, whereas, interestingly, expression of MKK9^KR stimulates RD29A and COR47 transcription (supplemental Fig. S3). mkk9 mutant seedlings grown in salt medium for 15 days maintained higher transcription levels of RD29A and COR47 genes than did the Col-0 wild type seedlings (supplemental Fig. S3). These results suggest that MKK9 also functions as a negative regulator of the response to salt stress in Arabidopsis seedlings and that RD29A and COR47 may be involved in the response.

DISCUSSION

Based on sequence alignments, plant MKKs are classified into four groups (A, B, C and D) (5). Arabidopsis MKK9 is a member of group D. MKK9 has been reported to be a negative regulator of the seed germination in response to various stresses (16). Recently, an Arabidopsis protoplast transient assay was used to show that MKK9 activates MPK3/MPK6, which in turn phosphorylates and stabilizes EIN3 (17). However, the role of the MKK9-MPK3/MPK6 cascade in the abiotic stress response is still far from clear. Here, kinase activity assays were used to show that MKK9 is an autoactivated MKK. The native protein (MKK9^WT) and the constitutively active mutant protein (MKK9^DD) of MKK9 were proven to phosphorylate and activate MPK3 and MPK6. Induction of MKK9^WT and MKK9^DD expression in stable transgenic plants activated two MPKs with the molecular masses of 44 and 48 kDa. The two MPKs were subsequently identified as MPK3 and MPK6 by crossing the transgenes into mpk3 and mpk6 plant backgrounds (Fig. 1 only showing MKK9^DD data). By combining biochemical and genetic tools, we have demonstrated that MPK3 and MPK6 are downstream MPKs of MKK9 both in vitro and in planta.

Yoo et al. (17) have reported that the MKK9-MPK3/MPK6 cascade activates the ethylene signaling pathway directly through EIN3; however, they did not report a strong induction of ethylene production in MKK9^DD transgenic plants, as described by our experiments (Fig. 2). Several ethylene overproducer (eto) mutants were shown to display the ethylene response phenotype (such as the triple response and activation of ethylene responsive genes) (56). The strong induction of ethylene production by the MKK9-MPK3/MPK6 cascade led us to speculate that the increase in ethylene may also contribute to MKK9-MPK3/MPK6-induced ethylene responses as in eto mutants. To test this hypothesis, we analyzed the etiolated seedling growth in the presence or absence of AVG or Ag⁺. We did observe inhibition of hypocotyl elongation in the etiolated MKK9^DD seedlings grown in the presence of DEX (Fig. 5), as reported by Yoo et al. (17). This confirmed that activation of MKK9-MPK3/MPK6 can induce ethylene responses in transgenic seedlings. However, the hypocotyls of the etiolated seedlings displayed normal growth when DEX was combined with either AVG or Ag⁺ (Fig. 5). These results suggest that both the ethylene biosynthesis inhibitor, AVG, and ethylene receptor antagonist, Ag⁺, could block the MKK9-MPK3/MPK6-induced ethylene signaling. Q-PCR results showed that induction of ERF gene transcripts was greatly inhibited by AVG and abolished by Ag⁺ (Fig. 6). Based on our results, we concluded that the MKK9-induced ethylene response is caused by increased ethylene signaling through ethylene receptors. This conclusion contrasts with that of the previous report (17).

Camalexin is synthesized from tryptophan via indole-3-acetaldoxime, a key branch point between primary and secondary metabolism in Arabidopsis (55). Camalexin is known to be induced by pathogen infection in Arabidopsis (19, 20). Analysis of the camalexin biosynthesis pathway shows that multiple genes for Trp and camalexin biosynthesis are coordinately and strongly induced by activation of MKK9-MPK3/MPK6 (Fig. 7, only data for dramatically induced genes, including CYP79B2, CYP71A13, and PAD3, are shown). As a consequence, camalexin accumulates in MKK9^WT and MKK9^DD plants. However, the overexpression of any of these genes, CYP79B2, CYP71A13, or PAD3, in Arabidopsis plants does not increase camalexin accumulation (26, 27, 55). These results imply two possibilities as follow: one is that the rate-limiting step in camalexin biosynthesis is not controlled by these genes; the other is that coordinate activation of multiple genes in the camalexin biosynthesis pathway, even those in Trp biosynthesis pathway, is needed for the camalexin accumulation. The high level of camalexin accumulation and coordinate induction of multiple genes by the MKK9-MPK3/MPK6 cascade suggest that MKK9-MPK3/MPK6 may phosphorylate transcription factor(s). The phosphorylation of transcription factor(s) is a common mechanism for stimulating the transcription of multiple genes.

It has been previously reported that the germination of mkk9 seeds is insensitive to salt (16), and the growth of mkk9 seedlings is hypersensitive to salt stress (17). This may imply that MKK9 performs different functions in response to salt at different stages of Arabidopsis development. These results led us to test the salt response of mkk9 and MKK9 transgenic plants. In our experiments, we treated the seedlings with 100 mm NaCl, the concentration used in both of the previous studies (16, 17). We found that, after transgene induction, MKK9^DD and MKK9^WT transgenic seedlings are more sensitive to salt stress compared with MKK9^KR seedlings, whereas mkk9 seedlings are less sensitive to salt compared with wild type seedlings (Fig. 10). The results of the salt response test for mkk9 seedlings are inconsistent with those of Yoo et al. (17). The possible reason for this controversy is that our seedlings were grown 4 days before they were transferred onto salt-containing media. However, the specific growth conditions were not mentioned in detail in the other report (17).

Several marker genes are induced in Arabidopsis plants in response to various stresses (such as cold, salt, drought, and osmotic stress) (57). We chose to monitor the transcription of RD29A and COR47 in MKK9 transgenic seedlings and mkk9 seedlings by Q-PCR. Our results showed that the expression of RD29A and COR47 genes was greatly reduced in MKK9^WT and MKK9^DD seedlings and increased in MKK9^KR seedlings after the transgene induction (supplemental Fig. S3A). The mkk9 seedlings growing in salt stress conditions had higher levels of RD29A and COR47 transcription (supplemental Fig. S3B). These results suggest that MKK9 negatively regulates seedling growth under salt stress conditions. However, the mechanism used by MKK9 to regulate the transcription of RD29A and COR47 is currently unknown.

There is ample evidence that activation of MPK3 and MPK6 by different MKKs mediates signaling in different biological responses. The MKK4/MKK5-MPK3/MPK6 axis has been reported to mediate the signaling of H₂O₂ generation and cell death (12), ethylene (37) and camalexin biosynthesis (28), and plant innate immunity (13). MKK3-MPK6 mediates signaling of jasmonic acid (9), whereas MKK2-MPK4/MPK6 mediates signaling of cold and salt stress (7). MKK9 also activates MPK3/MPK6 and induces ethylene and camalexin biosynthesis, suggesting that functions of MKK4, MKK5, and MKK9 in some pathways, at least at the seedling stage, are partially overlapping. We also checked the gene expression levels of MKK4 and MKK5 in Col-0 wild type and mkk9 seedlings (supplemental Fig. S4). Up-regulation of MKK4 and MKK5 expression in mkk9 plants suggests that loss of function in mkk9 mutant could be, at least partially, complemented by the other two MKKs. Therefore, it is not surprising that mkk9 mutant plants produce the same levels of camalexin as wild type plants when treated with the bacterial pathogen, P. syringae pv. maculicola strain ES4326 (data not shown). However, the high levels of ethylene and camalexin production in transgenic plants may imply that MKK9 plays more important roles than MKK4/MKK5 in regulating camalexin and ethylene biosynthesis. Furthermore, the sensitivity to salt stress in transgenic seedlings after MKK9 activation and inactivation and in mkk9 mutant seedlings indicates that MKK9 also functions in the signaling of the salt stress response. To distinguish the specific functions of MKK4, MKK5, and MKK9, the double or triple mutants will need to be analyzed.