Emerging topics in the cell biology of mitogen-activated protein kinases Olga S ˇ amajova ´ * , George Komis * , and Jozef S ˇ amaj Centre of the Region Hana ´ for Biotechnological and Agricultural Research, Faculty of Science, Palacky ´ University, Department of Cell Biology, S ˇ lechtitelu ˚ 11, CZ-783 71 Olomouc, Czech Republic Signaling through mitogen-activated protein kinase (MAPK) cascades is organized in complex intercon- nected subcellular networks. Upon MAPK activation, signals are transferred to targets in different subcellular compartments able to regulate various cellular process- es. Therefore, subcellular dissection of individual MAPK modules is vital to understand how a single MAPK can simultaneously mediate many tasks and how a single stimulus can direct different MAPK modules to separat- ed tasks. In this opinion article, we present a subcellular localization prediction of all members of Arabidopsis thaliana MAPK modules validated wherever possible with experimental data. Furthermore, we propose, that at least in part, the complexity of plant MAPK signaling can be explained by unique strategies of subcellular targeting, which will be worth investigating in the near future. Complexity, crosstalk, redundancy, and subcellular diversification in plant MAPK signaling MAPK signaling is a universal extracellular stimulus de- coder throughout the eukaryotic kingdom. MAPKs are able to translate extracellular signals (e.g., those induced by biotic and abiotic stresses), to a temporally hierarchical cell response, ranging from early subcellular remodeling to late gene expression transactivation, which can take between a few minutes up to a few hours [1–4]. Early signaling due to MAPK activation may lead to direct subcellular rearrangements, such as cytoskeletal remodeling by phosphorylated substrates associated with the cytoskeleton [5,6]. However, such changes may also occur secondarily over a longer timeframe following MAPK activation due to gene expression transactivation [7]. For example, transcriptome- and (phospho)-proteome- wide studies on stimulus-activated MAPKs or on the MAPK-deficient and constitutively active mutants have shown either the MAPK-dependent phosphorylation of cytoskeletal proteins or their transcriptional up- or down- regulation in the absence of phosphorylation [8,9]. More- over, high-throughput interaction studies revealed a diverse array of MAPK interacting partners, ranging from plasma membrane proteins to transcriptional factors [10,11]. The core of MAPK signaling in all eukaryotes is orga- nized in three-tiered modules comprising a MAPK kinase kinase (MAPKKK), a dual-specificity MAPKK, and a MAPK, within which phosphorylation signals are trans- duced linearly from the MAPKKK to the MAPK [12]. In general, upstream regulators of MAPK modules include MAPKKK kinases (MAPKKKKs), small and heterotri- meric GTPases and membrane-confined receptor-like kinases, all of which are found close to the initiation of signaling cascades [3]. The MAPK becomes activated by dual phosphorylation of a threonine–X–tyrosine (TXY) motif (see Glossary) within its activation loop by the MAPKK, which has dual specificity [12]. In principle, the dual phosphorylation of MAPK serves three purposes. At first, it disrupts the association with its cognate MAPKK, allowing the interaction with the sub- strate, because D-motifs in both substrates and MAPKKs antagonize for the CD site of the MAPK [13]. It then renders the structure of the catalytic domain highly active [3]. Subsequently, activated MAPKs phosphorylate plasma membrane, cytoplasmic and nuclear targets, such as ion transporters, cytoskeletal proteins, and transcription fac- tors [14–16]. Finally, the liberated, active MAPK is able to translocate to subcellular compartments relevant to its function [3]. MAPK modules have been conserved during the evolu- tion of higher plants. Moreover, they have been expanded to an unprecedented complexity, with at least 20 MAPKs, ten MAPKKs and 60–80 MAPKKKs in the annotated A. thaliana genome [1], and with similar complexity in rice [17]. Individual plant MAPKs can be activated by a wide array of stimuli; for example, AtMPK6 responds to patho- gen-associated molecular patterns (PAMPs), hyperosmo- tic, and genotoxic stress [18–20]. Additionally, the same MAPK may be activated under diverse circumstances with a different outcome; for example, MPK4 is independently involved in oxidative stress, the downregulation of innate immune responses, and also the regulation of cortical microtubule organization and cytokinesis [21–25]. Finally, two or more MAPKs may operate, often redundantly, in the same pathway; for example, AtMPK3 and AtMPK6 are involved in stomatal ontogenesis during leaf epidermal tissue patterning in A. thaliana [26]. To this we may add the spatial organization of MAPK modules by scaffold proteins, which represent major means for localizing MAPK cascades to certain subcellular compartments [27]. Localization of activated MAPK may also be dictated by the availability of substrates confined to such compartments [3], and their facilitated localization or activation-dependent relocalization with the aid of binding partners, including MAPKK(K)s [28] and MAPK Opinion Corresponding author: S ˇ amaj, J. (jozef.samaj@upol.cz) * These authors contributed equally to this work. TRPLSC-1030; No. of Pages 9 1360-1385/$ – see front matter ß 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tplants.2012.11.004 Trends in Plant Science xx (2012) 1–9 1