Plant-Microbe Communications for Symbiosis

Symbiosis is a biological phenomenon involving dynamic changes in the genome, metabolism and signaling network, and a multidirectional comprehension of these interactions is required when studying symbiotic organisms. In plant–microbe interactions, two symbiotic systems have been actively studied for many years. One is arbuscular mycorrhizal (AM) symbiosis and the other is root nodule (RN) symbiosis. AM symbiosis is probably the most widespread interaction between plants and microbes, in the context of phylogeny and ecology (Kistner and Parniske 2002, Bonfante and Genre 2010). More than 80% of all land plant families are thought to have a symbiotic relationship with AM fungi that belong to the Glomeromycota. The origin of AM symbiosis is thought to be in the early Devonian period, approximately 400 million years ago. Thus, AM symbiosis is also called the mother of plant root endosymbioses (Parniske 2008). On the other hand, RN symbiosis involves morphogenesis and is formed by communication between plants and nitrogen-fixing bacteria. AM fungi absorb minerals, including phosphate and nitrogen, via extraradical hyphae and supply them to the plant, possibly via highly branched structures inside root cells known as arbuscules (Harrison 1999, Parsnike 2008). The arbuscules are enveloped in plant-derived membranes, the periarbuscular membrane. To absorb phosphate, mycorrhiza-induced phosphate transporter genes, such as MtPT4, are predominantly or exclusively up-regulated in plant root cells containing arbuscules. Some of these genes have been shown to be required for AM symbiosis as well as the acquisition of phosphate delivered by the AM fungus (Javot et al. 2007). Interestingly, in addition to phosphate, AM fungi transfer substantial amounts of nitrogen to the host plants (Javot et al. 2007). Nitrogen sources such as ammonium and amino acids are translocated to the plant via fungal hyphae. Several ammonium transporters (AMTs) have been identified in plants, and it will be interesting to determine their precise expression patterns and subcellular location. Kobae et al. report in this issue that GmAMT4.1 identified from the soybean genome database shows specific expression in arbusculated cortical cells. Moreover, they show that GmAMT4.1 is localized on the branch domains of periarbuscular membranes, indicating that arbuscule branches are active sites for ammonium transport (Kobae et al. 2010). RN symbiosis in legumes involves host-specific recognition and post-embryonic development of a nitrogen-fixing organ, the root nodule. Compared with our extensive knowledge of the bacterial factors required for RN symbiosis, such as Nod factors, lipo- and exopolysaccharides (Jones et al. 2007), little is known about the symbiotic factors in terms of the plant. Molecular genetics and genomics of two model legumes, Lotus japonicus and Medicago truncatula, have made a significant contribution to the identification of a number of plant factors and to an understanding of the molecular basis of nodule initiation, rhizobial infection, organogenesis, nitrogen fixation, senescence and feedback regulation (Sandal et al. 2006, Oldroyd and Downie 2008, Kouchi et al. 2010). Among these phenomena, nodule senescence, namely regulation of nodule lifespan, is uncharted territory but would gain in importance when our aim is to achieve long-term nitrogen-fixing activity in legumes. D’haeseleer et al. analyzed the nodule senescence up-regulated gene M. truncatula ATB2. The MtATB2 gene encodes a bZIP transcription factor and is regulated by sucrose and light conditions as well as during nodule senescence. For nodule function, carbon is provided mainly as sucrose derived from photosynthesis and transported via the phloem. These authors show that the MtATB2 transcripts occur in and around the vascular tissue of nodules and roots and in the nodule apex (D’haeseleer et al. 2010). Leguminous plants strictly control nodule numbers, because nodulation and nitrogen fixation are an energy drain on the host. To maintain the symbiotic balance with rhizobia, plants have evolved negative feedback systems known as autoregulation of nodulation (AON). AON involves long-distance signaling via shoot–root communication and is mediated by CLAVATA1-like receptor kinases such as L. japonicus HAR1, Glycine max NARK and M. truncatula SUNN (Oka-Kira and Kawaguchi 2006, Ferguson et al. 2010). In L. japonicus, grafting experiments have shown that HAR1 and KLAVIER function in the shoot to restrict the nodule number while TML functions in the root. In this issue, Yoshida et al. report the characteristics of a novel root-regulated hypernodulating mutant in L. japonicus named plenty. Unlike TML, PLENTY appears to act independently of HAR1-mediated long-distance control of nodulation and mediates nitrate signaling (Yoshida et al. 2010). Thus, plenty is invaluable for dissecting the complex web of negative regulatory systems in nodulation. NSP1 and NSP2, which encode a GRAS family transcription factor, are required for nodule initiation and act as components of the Nod factor signaling pathway downstream of Ca2+ spiking (Oldroyd and Downie 2008, Kouchi et al. 2010). The homologs are widely conserved in non-leguminous plants but their functions are unknown. Yokota et al. isolated homologs of NSP1/2 from the monocot rice, OsNSP1 and OsNSP2, and introduced these genes into nsp1 and nsp2 non-nodulating mutants in L. japonicus, respectively. Transformation with OsNSP1 and OsNSP2 fully rescued the mutant phenotypes such as nodule development and nitrogenase activity, indicating that these rice transcription factors can potentially mediate Nod factor signaling in L. japonicus (Yokota et al. 2010). Precise expression and functional analyses of NSP1/2 homologs in rice would shed light on their original function in plants. In response to stimulation by flavonoids exuded from legume roots into soil, rhizobia synthesize signaling molecules that are responsible for nodule formation. These signaling molecules, named Nod factors, have been identified as lipochito- oligosaccharides decorated with diverse chemical substitutions (Spaink 1995). Nod factors are certainly a key trigger for legume symbiotic signaling and nodule organogenesis (Kouchi et al. 2010). However, other rhizobial systems such as exopolysaccharide excretion (Skorupska et al. 2006), ethylene biosynthesis regulation (Okazaki et al. 2004, Gresshoff et al. 2009), protein secretion systems (Deakin and Broughton 2009) and BacA (LeVier et al. 2000) are often required for the establishment of symbiosis with legumes, probably because they are involved in bacterial release into the host cytoplasm and bacteroid development. BacA protein, a cytoplasmic membrane protein conferring resistance to peptide antibiotics, was the first bacterial factor identified as essential for bacteroid development in Sinorhizobium meliloti (LeVier et al. 2000). Interestingly, a homolog of BacA in Brucella abortus, an animal pathogen, is required for effective survival in murine macrophages (LeVier et al. 2000). In this special issue, Maruya and Saeki (2010) examine the physiological functions of the BacA homolog in Mesorhizobium loti. From study of a bacA mutant, they found that BacA is dispensable for M. loti symbiosis with L. japonicus. However, the M. loti bacA gene partially restored the symbiosis-defective phenotype of a S. meliloti bacA mutant by genetic complementation. These results raise the question of why BacA is not absolutely required for symbiosis with M. loti but it is required for that with S. meliloti. Recently, other legume–rhizobium partners were examined for their BacA requirement for symbiosis (Karumakaran et al. 2010). One fascinating explanation is that BacA is exclusively required in galegoid legumes producing defensin-type antimicrobial peptides (NCR peptides). Our knowledge of the molecular interplay between plant and AM and RN symbionts has developed over the past decade (Oldroyd and Downie 2008, Parniske 2008, Kouchi et al. 2010). Diverse microbes are able to associate with plants in natural habitats as endophytes and epiphytes (Mano and Morisaki 2008, Saito et al. 2007). The cellular signaling networks that determine RN and AM symbioses overlap in legumes, suggesting that the molecular components of the legume-specific networks are shared with other plant-associated microbes. In this special issue, Ikeda et al. (2010) report the microbial community shifts of field-grown soybeans with different nodulation genotypes and nitrogen levels, suggesting that soybean accommodates a taxonomically characteristic microbial community by a system of autoregulation of nodulation involving symbiotic signal networks. A unique procedure of bacterial enrichment from plant tissues assisted this study significantly, although profiling techniques currently used to assess the microbial community structure in environmental microbiology required further development (Saito et al. 2007). These efforts could provide platforms for dissecting the organization and functions of rhizosphere and phyllosphere microbial communities, and for identifying the plant loci that contribute to their formation (Bisseling 2009). In addition, a deeper understanding of plant-associated microbial communities offers exciting opportunities for controlling crop growth in sustainable agricultural settings. The genomics of both partners (legumes and rhizobia) and their bioresources have been crucial in the facilitation of recent progress in the study of legume and microbe interactions. In particular, the activities of the Kazusa DNA Research Institute and the M. truncatula genome sequence consortium have contributed significantly to our understanding of plant–microbe interactions, which may be summarized as follows: Information has been generated on the majority of genome sequence from euchromatic regions and corresponding genetic markers for model legumes (Young et al. 2005, Sato et al. 2008). Web databases have been constructed for the L. japonicus genome database, miyakogusa.jp (http://www.kazusa.or.jp/lotus/) and M. truncatula sequencing resources (http://www.medicago.org/genome/), to provide detailed information on L. japonicus and M. truncatula genomes and markers (Young et al. 2005, Sato et al. 2008). This greatly facilitates the identification of causal genes of symbiotic mutants in the model legumes (Sandal et al. 2006, Oldroyd and Downie 2008, Parniske 2008, Kouchi et al. 2010). Complete genome sequences have been obtained of the endosymbionts Mesorhizobium loti MAFF303099 (Kaneko et al. 2000), Sinorhizobium meliloti 1021 (Galibert et al. 2001) and Bradyrhizobium japonicum USDA110 (Kaneko et al. 2002), and the endophyte Azospirillum sp. B510 (Kaneko et al. 2010), and comprehensive web databases for these bacteria have been constructed: RhizoBase (http://genome .kazusa.or.jp/rhizobase/) and RhizoGATE (http://www .cebitec.uni-bielefeld.de/CeBiTec/rhizogate/). Several analysis tools and post-genomic data for S. meliloti have been well integrated in RhizoGATE (http://www .cebitec.uni-bielefeld.de/CeBiTec/rhizogate/). A large-scale protein–protein interaction database for M. loti using the yeast two-hybrid system is available (Shimoda et al. 2008a). In addition, a mutant library of M. loti has been constructed using the signature-tagged mutagenesis technique, covering 3,680 non-redundant M. loti genes (51% of the total genes) (Shimoda et al. 2008b), which is distributed worldwide via the NBRP (http://www.legumebase .brc.miyazaki-u.ac.jp/). This information together with the mutant material will enable significant progress in this field in the next decade. This special collection of articles highlights some of the recent progress made in understanding plant–microbe symbiosis and its contribution to the plant and microbial worlds.