Signaling specificity and complexity of MAPK cascades in plant innate immunity Jen Sheen, Ping He, Libo Shan, Yan Xiong, Guillaume Tena, Sangdong Yoo, Younghee Cho, Marie Boudsocq, and Horim Lee Department of Molecular Biology, Massachusetts General Hospital & Department of Genetics, Harvard Medical School, Boston, 185 Cambridge Street, Simches Research Center, Boston, Massachusetts 02114, USA. sheen@molbio.mgh.harvard.edu The desire to understand and control plant diseases has been a major drive to study host-microbe interactions. In addition to the molecular basis of diseases and disease resistance, another fundamental question is how plants manage to stay healthy and survive constant exposure to a wide range of microbes. The innate immune response is the first line of defense, and is likely critical for the survival and fitness of plants in the ubiquitous presence of microbes. Research in the past decade has revealed remarkable convergent evolution in the recognition of pathogen- or microbe-associated molecular patterns (PAMPs or MAMPs) and the activation of innate immune responses in plants, insects and mammals (Nürnberger et al., 2004; Ausubel, 2005; Zipfel and Felix, 2005; Akira et al., 2006; Ferrandon et al., 2007). We are interested in dissecting the plant signal transduction networks that are responding to or manipulated by microbial signals and effectors. For a robust innate immune system, both signaling specificity and complexity are required. Currently, the sequenced plant model Arabidopsis thaliana provides the most advanced and sophisticated molecular, cellular, genetic, genomic, proteomic, and bioinformatic tools for studying signaling mechanisms and transcription controls in plant innate immunity. The primary cell assays using protoplasts freshly isolated from leaves support high- throughput and versatile analyses of diverse MAMP and microbial effector signal responses (Asai et al., 2002; He et al., 2006a, 2006b; Yoo et al., 2007). The virus-induced gene silencing (VIGS) method (Burch-Smith et al., 2006), inducible RNAi capability, saturating insertional mutant resources (Sessions et al., 2002; Alonso et al., 2003), and whole-genome microarray databases 1