with these advanced microbial strategies using
effectors, plants evolved the second layer of
defense system. Plant nucleotide-binding leucine-
rich repeat (NB-LRR) receptors constantly moni-
tor either the existence of microbial effectors or the
condition of key signaling components of MTI.
Detection of effectors or abnormal modification of
signaling components targeted by effectors results
in the induction of robust defense responses
including localized programmed cell death at
infection sites, called effector-triggered immunity
(ETI). Thus, these NB-LRR receptors, which have
been known as a major group of“resistance pro-
teins”(R proteins), play crucial roles for the spe-
cific recognition of pathogens (Heidrich et al.
2012 ). These highly sophisticated defense mech-
anisms effectively block the vast majority of hos-
tile microbes; however, a small number of
successful pathogens can overcome both of the
two layers of the defense system and cause serious
damage to the plant growth. Therefore, nature is a
grueling battlefield between plants and microbes
and it is surprising that mutual plant–microbe
symbioses are achieved in such an environment.
15.2 MAMPs-Triggered Immunity
inL. japonicus
Although legumes accept the infection of rhizo-
bia and establish an endosymbiosis, these plants
are assumed to be equipped with the defense
mechanism, protecting them from the pathogen’s
attacks. In L. japonicus, treatments with
MAMPs, such as chitin oligosaccharides orfla-
gellin epitope,flg22, upregulate the expression of
defense-related genes including PR gene homo-
logs, peroxidases, chitinases, ERF and WRKY
transcription factors, and the genes involved in
the biosynthesis of pterocarpans, which are
known to be legume-specific phytoalexins
(Shimada et al. 2007 ; Nakagawa et al. 2011 ).
These results clearly indicate the existence of
MTI in legumes. Interestingly, rhizobial symbi-
otic signal molecules, Nod factors (NFs), consist
of a chitin oligosaccharide backbone, in which
the non-reducing end is N-acylated and the
reducing end is decorated with various molecules
(Cullimore et al. 2001 ). Thus, the structure of
NFs is closely related to a typical MAMP, chitin
oligosaccharide; however, their physiological
effects are opposite, i.e., friendly acceptance or
rejection by host plants.
Recognition of MAMPs is the crucial step for
MTI, and plants have evolved the corresponding
receptors for each MAMP molecule. For exam-
ple,flg22 and chitin oligosaccharides are per-
ceived by FLS2 and CERK1 in A. thaliana,
respectively (Gomez-Gomez and Boller 2000 ;
Miya et al. 2007 ). A homologous gene of FLS2,
LjFLS2, is also found inL. japonicus. Expression
ofLjFLS2is observed in both leaves and roots
but decreased in nodules (Lopez-Gomez et al.
2012 ). The latter observation may suggest the
presence of downregulation of defense mecha-
nism in nodules to favor the entry of symbiotic
partners. On the other hand, the peptide
sequences offlg22 in Rhizobium or a highly
related pathogen,Agrobacterium, are very dif-
ferent and the Rhizobium peptide could not
trigger MTI in both host and non-host plants
(Felix et al. 1999 ; Lopez-Gomez et al. 2012 ).
These results suggested the presence of recipro-
cal adaptation between host and symbiotic
partner.
NF receptor, NFR1, has a domain structure
composed of an extracellular LysM domain, a
single-pass transmembrane domain and an
intracellular kinase domain, which is exactly the
same as that of CERK1 inArabidopsis(Radutoiu
et al. 2003 ; Miya et al. 2007 ; Shimizu et al.
2010 ). Among the LysM domain-containing
receptor-like kinases found inArabidopsisgen-
ome, CERK1 is a single best match homolog of
NFR1 (Zhang et al. 2007 ; Zhu et al. 2006 ). In
addition, genomic structures aroundArabidopsis
CERK1 (AtCERK1) andLotusNFR1 showed
limited but significant synteny, indicating that
these genes are the descendants of a common
ancestor (Zhu et al. 2006 ). Indeed, NFs not only
activate symbiosis genes but also transiently
activate defense-related genes through NFR1 in
L. japonicus(Nakagawa et al. 2011 ). On the
other hand, the kinase domain of AtCERK1 is164 T. Nakagawa et al.