Convergence and divergence of stress-induced mitogen-activated protein kinase signaling pathways at the level of two distinct mitogen-activated protein kinase kinases

Abstract

Plants respond to biotic and abiotic stresses by inducing overlapping sets of mitogen-activated protein kinases (MAPKs) and response genes. To define the mechanisms of how different signals can activate a common signaling pathway, upstream activators of SIMK, a salt stress- and pathogen-induced alfalfa MAPK, were identified. Here, we compare the properties of SIMKK, a MAPK kinase (MAPKK) that mediates the activation of SIMK by salt stress, with those of PRKK, a distantly related novel MAPKK. Although both SIMKK and PRKK show strongest interaction with SIMK, SIMKK can activate SIMK without stimulation by upstream factors. In contrast, PRKK requires activation by an upstream activated MAPKK kinase. SIMKK mediates pathogen elicitor signaling and salt stress, but PRKK transmits only elicitor-induced MAPK activation. Of four tested MAPKs, PRKK activates three of them (SIMK, MMK3, and SAMK) upon elicitor treatment of cells. However, PRKK is unable to activate any MAPK upon salt stress. In contrast, SIMKK activates SIMK and MMK3 in response to elicitor, but it activates only SIMK upon salt stress. These data show that (1) MAPKKs function as convergence points for stress signals, (2) MAPKKs activate multiple MAPKs, and (3) signaling specificity is obtained not only through the inherent affinities of MAPKK-MAPK combinations but also through stress signal-dependent intracellular mechanisms.

Figure 1.

Primary Structure of PRKK, a Pathogen-Responsive Alfalfa MAPKK. The amino acid sequence of PRKK was aligned with its closest homologs belonging to the PMKK1 subfamily of plant MAPKKs (Ligterink and Hirt, 2001). PRKK showed 73% identity to tobacco SIPKK (Liu et al., 2000), 71% identity to tomato LeMEK1 (Hackett et al., 1998), and 68% identity to Arabidopsis AtMKK2 (Ichimura et al., 1998a). For comparison, SIMKK also was included (Kiegerl et al., 2000). Identical and conserved amino acids are shaded in black and gray, respectively; dashes represent gaps. The 11 catalytic subdomains are represented by roman numerals above the respective regions, and amino acids defining the putative MAPK docking site are underlined. The putative phosphorylation sites are indicated by asterisks; note that the consensus sequence for plant MAPKKs in this region is S/T-(X)5-S/T and not S/T-(X)3-S/T, as in yeast and animal MAPKKs (Alessi et al., 1994).

Figure 2.

Recombinant PRKK Is an Inactive MAPKK. Different abilities of GST-PRKK and GST-SIMKK to phosphorylate the kinase-negative GST fusion protein of SIMK (GST-SIMK[K84R]) in vitro. Autoradiogram (A) and Coomassie blue staining (B) of SDS-PAGE showing analyses of the products of in vitro kinase reactions between affinity-purified GST-PRKK, GST-SIMKK, or GST on GST-SIMK(K84R) as substrate. Proteins to be tested for MAPK phosphorylating activity (GST [1 μg], GST-PRKK [1 μg], or GST-SIMKK [0.5 μg]) were incubated for 30 min in kinase reaction buffer (50 mM Tris, pH 7.5, 10 mM MgCl₂, 1 mM DTT, 0.1 mM ATP, and 6 μCi of γ-³²P-ATP) alone or with 5 μg of MAPK (GST-SIMK[K84R]) as a substrate.

Figure 3.

Active MAPKKK Is Required for PRKK Activation. Approximately 10⁶ parsley protoplasts per treatment were transformed to transiently express the constitutively active MAPKKK ΔNPK1 (Kovtun et al., 2000) (lane 2), the HA-tagged version of PRKK (lane 3), or both (lane 4). Protoplasts transformed with the empty vector pRT101 were used as controls (lane 1). After protein extraction, samples were immunoprecipitated with HA antibody. In vitro kinase assays of immunoprecipitated PRKK-HA were performed on 4 μg of kinase-negative GST-SIMK as substrate. Reaction products were analyzed by SDS-PAGE followed by autoradiography (A) and Coomassie blue staining (B). (C) shows PRKK-HA expression. The specificity of immunoprecipitation was controlled by immunoblotting extracts with HA antibody.

Figure 4.

PRKK Activates SIMK, MMK3, and SAMK in Vivo. MBP kinase activities of immunoprecipitated SIMK-HA, MMK2, MMK3-HA, and SAMK when expressed transiently in protoplasts alone (lane 2), with active MAPKKK ΔNPK1 (lane 5), with SIMKK (lane 3), or with PRKK in the absence (lane 4) or presence (lane 6) of ΔNPK1. Protoplasts transformed with the empty vector pRT101 were used as a negative control (lane 1). MAPK activity was analyzed by MBP phosphorylation of immunoprecipitated SIMK-HA, MMK2, MMK3-HA, and SAMK. After SDS-PAGE, MAPK activities were determined by autoradiography (A) and MAPK expression levels were assessed by immunoblotting with HA, MMK2, or SAMK antibody (B).

Figure 5.

PRKK and SIMKK Enhance the Elicitor-Induced Activation of MAPKs. Parsley protoplasts were transformed transiently with a combination of the following expression plasmids: pRT101-SIMKK, pRT101-PRKK, pSH9-SIMK-HA, pRT101-MMK2, pSH9-MMK3-HA, and pRT101-SAMK. MBP kinase activity of immunoprecipitated MAPK was evaluated after transient expression in protoplasts with SIMKK or PRKK in the absence or presence of 50 nM Pep13 elicitor (E; lanes 5 and 8) or 250 mM NaCl (S; lanes 6 and 9). After harvesting of cells 10 min after treatment, proteins were extracted and MAPKs were immunoprecipitated. Untreated samples were used as controls for the basal activity of overexpressed MAPKs; treated samples transformed with the empty vector pRT101 are shown to exclude nonspecific immunoprecipitation of endogenous elicitor- or salt-activated MAPKs (lanes 10 and 11). (A) The activity of the MAPKs was analyzed by in vitro MBP kinase reactions. The phosphorylated products were analyzed by SDS-PAGE followed by autoradiography. (B) Immunoblotting of cell extracts used in (A) with anti-HA, anti-MMK2, or anti-SAMK antibody confirmed that equal expression levels of MAPKs were present in different protoplast expression assays.

Figure 6.

Convergence and Divergence of Salt- and Elicitor-Induced Signals at the Level of MAPKKs. Both stimuli seem to be mediated by SIMKK, which activates SIMK in the case of salt stress and activates SIMK and MMK3 upon elicitor treatment. Although unable to mediate salt stress signaling, PRKK is capable of distributing the elicitor signal onto SIMK, MMK3, and SAMK in vivo. The MAPKK responsible for elicitor-induced MMK2 activation remains to be identified.