uva-dare (digital academic repository) insights into nod ... · triggering root nodule...
TRANSCRIPT
UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
Insights into Nod factor signaling mediated by Medicago truncatula LysM receptor-likekinases, MtNFP and MtLYK3
Pietraszewska-Bogiel, A.
Link to publication
Citation for published version (APA):Pietraszewska-Bogiel, A. (2012). Insights into Nod factor signaling mediated by Medicago truncatula LysMreceptor-like kinases, MtNFP and MtLYK3.
General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.
Download date: 09 Apr 2020
Insights into Nod factor signaling
mediated by Medicago truncatula
LysM receptor-like kinases, MtNFP and MtLYK3
Anna Pietraszewska-Bogiel
University of Amsterdam
2012
Insi
gh
ts i
nto
Nod
fact
or
sign
ali
ng m
edia
ted
by M
edic
ago t
run
catu
la L
ysM
rec
epto
r-li
ke
kin
ase
s, M
tNF
P a
nd
MtL
YK
3
1
Insights into Nod factor signaling
mediated by Medicago truncatula
LysM receptor-like kinases, MtNFP and MtLYK3
ACADEMISCH PROEFSCHRIFT
ter verkrijging van de graad van doctor
aan de Universiteit van Amsterdam
op gezag van de Rector Magnificus
prof. dr. D.C. van den Boom
ten overstaan van een door het college voor promoties
ingestelde commissie,
in het openbaar te verdedigen in de Agnietenkapel
op woensdag 12 december 2012, te 12:00 uur
door
Anna Pietraszewska-Bogiel
geboren in Warschau, Polen
2
Promotiecommissie
Promotor: Prof. dr. Th.W.J. Gadella
Faculteit der Natuurwetenschappen, Wiskunde en Informatica
Overige Leden: Prof. Dr. A.H.J. Bisseling,
Prof. Dr. B.J.C. Cornelissen,
Prof. Dr. M.A. Haring,
Dr. J.V. Cullimore,
Dr. K. Jalink,
3
Rodzicom i Mężowi
Gdy człowiek wierzy w siebie,
to cała reszta to betka...
Jonasz Kofta
4
ISBN 978-94-6203-255-2
The research described in this thesis was carried out at the Laboratory of Molecular
Cytology, Swammerdam Institute for Life Sciences, Faculty of Science,
University of Amsterdam, The Netherlands
Printed by CPI-Wöhrmann Print Service – Zutphen
Picture of a Medicago truncatula nodule used on the cover of this Thesis was taken
from Klaus-Heisen et al. (2011), with the permission from Dr. Julie V. Cullimore.
5
Abbreviations
AM: arbuscular mycorrhiza
AS activation segment
CD: cell death
CO chitooligosaccharide
CSP: common symbiosis (signaling) pathway
ePK: eukaryotic protein kinase
ER: endoplasmic reticulum
ExR: extracellular region
FLIM: Fluorescence Lifetime Imaging Microscopy
FP: fluorescent protein
FRET: Förster/Fluorescence Resonance Energy Transfer
GlcNAc: N-acetylglucosamine
hai: hours after inoculation/infiltration
InR: intracellular region
IT: infection thread
KD: kinase domain
LCO: lipo-chitooligosaccharide
LysM: lysin motif
Myc factor: mycorrhization factor
NF: Nodulation factor
PAMP: pathogen-associated molecular pattern
PGN: peptidoglycan
PLC/PLD: phospholipase C / phospholipase D
PM: plasma membrane
RH: root hair
RL: Rhizobium-legume (symbiosis)
RLK: receptor-like kinase
ROS: reactive oxygen species
WT: wild-type
YFP: yellow fluorescent protein
6
7
Contents
Outline
Chapter 1 Introduction
Chapter 2 Functional interaction of Medicago truncatula NFP and
LYK3 symbiotic LysM receptor-like kinases results in a
defense(-like) response and cell death induction in
Nicotiana benthamiana
Chapter 3 Molecular interactions of Medicago truncatula putative
Nod factor receptors, NFP and LYK3, as demonstrated
with Fluorescence Resonance Energy Transfer-
Fluorescence Lifetime Imaging Microscopy
Chapter 4 Protein kinase domain is indispensible for biological
activity of Medicago truncatula NFP LysM receptor-like
kinase
Chapter 5 Structure-function analysis of Medicago truncatula
LYK3 LysM receptor-like kinase in Nicotiana
benthamiana
Chapter 6 Concluding remarks
Summary
References
Acknowledgements
Curriculum Vitae
9
11
41
77
107
129
157
169
163
187
189
8
9
Outline
Medicago truncatula can establish a mutually beneficial symbiosis
with a specific rhizobium bacterium, Sinorhizobium meliloti, ultimately leading to
the formation of specialized symbiotic organs (nodules) in which atmospheric nitrogen
is converted into ammonia by the bacteria in exchange for plant carbohydrates.
Lipo-chitooligosaccharides (LCOs), termed Nodulation (Nod) factors (NFs),
are secreted bacterial signals whose perception by the host plant is essential for
triggering root nodule organogenesis and for nodule infection by rhizobia. The goal
of the research presented in this thesis was to investigate signaling mechanisms
employed by two Lysin motif (LysM) domain-containing receptor-like kinases (RLKs),
MtNFP and MtLYK3, postulated to function as putative NF receptors in Medicago.
Attempts to visualize these proteins in situ have been unsuccessful. On the contrary,
Agrobacterium-mediated transient transformation of Nicotiana benthamiana leaf
has allowed production of both MtNFP and MtLYK3 fluorescent protein fusions
to the extent that they could be visualized with fluorescence microscopy. Chapters 2
and 3 present characterization of subcellular localization and oligomerization status
of these RLKs in vivo. Importantly, simultaneous accumulation of MtNFP and MtLYK3
but not the accumulation of either protein alone resulted in induction of cell death (CD)
and defense(-like) response in Nicotiana leaf, indicating their functional interaction.
Chapter 2 presents a detailed characterization of this specific response. Chapters 4 and 5
present a detailed comparison of the requirements of nodulation and CD induction
with respect to MtNFP and MtLYK3 structure.
10
11
CHAPTER 1
In Rhizobium-legume (RL) symbiosis, rhizobia can invade roots of legume
plants, ultimately leading to the formation of specialized symbiotic organs, termed
nodules, in which the bacteria differentiate and reduce atmospheric dinitrogen
into ammonia in exchange for plant carbohydrates (Oldroyd & Downie 2008; Oldroyd
et al., 2011). The onset of this interaction requires mutual signaling: plant flavonoids
(secondary metabolites) are exuded by the root under nitrogen starvation and,
in response, rhizobia secrete lipo-chitooligosaccharides, termed Nodulation (Nod)
factors (NFs) (Zhang et al., 2009a and refs therein). These NFs generally play a central
role in RL symbiosis (Masson-Boivin et al., 2009). They consist of three to five ß 1-4-
linked N-acetylglucosamine (GlcNAc) residues with an N-linked fatty acid moiety
attached to the non-reducing terminal sugar (D’Haeze & Holsters 2002). Additional
decorations of the GlcNAc backbone, and length and degree of saturation
of the fatty acid chain, differ between rhizobial strains and serve as a principle
determinant of the specificity of RL interaction. For example, the NF of S. meliloti
is N-acylated with an unsaturated C16 fatty acid (mostly C16:2), O-acetylated
at the non-reducing, and O-sulphated at the reducing terminal sugar. The latter
modification allows S. meliloti to nodulate Medicago spp., but not pea (Pisum sativum),
which requires non-sulphated NF produced by Rhizobium leguminosarum.
In the following summary, I will give a short description of developmental
steps during RL symbiosis, followed by a presentation of selected molecular
components involved in NF signaling. The main focus will be the NF-dependent
responses and putative NF receptors in Medicago truncatula (Medicago), Lotus
japonicus (Lotus) and Sesbania rostrata (Sesbania): three model legume species
12
that differ with respect to rhizobia strains that nodulate them, anatomy and physiology
of nodules formed, and mechanisms of rhizobia infection. Finally, other molecules
structurally related to the putative NF receptors but involved in different processes,
i.e. in the arbuscular mycorrhiza symbiosis and innate immunity, will be described.
1. Developmental steps in Medicago truncatula-Sinorhizobium meliloti root nodule
symbiosis
In many RL interactions that result in the formation of root nodules (including
that of Medicago - S. meliloti), infection of the host root starts with specific adhesion
of compatible rhizobia to root hairs (RHs), i.e. specialized extensions of root epidermal
cells (see Figure 1 step 1). This step involves both surface polysaccharides and secreted
proteins (Downie 2010). As NFs are accumulated and immobilized in cell walls
of the root epidermis (Goedhart et al., 2003 and refs therein), RHs are likely exposed to
a localized elevated concentration of NF which activates symbiotic signaling
(see below) and is postulated to serve as a positional cue for redirecting RH growth.
Deformation of RHs is restricted to a so-called susceptible zone characterized by
actively growing RHs (Gage 2004; Esseling et al., 2004), and is the first morphological
response to the perception of rhizobia or NFs (observed within 1-6h after application of
10-11
M NF). Continuous re-establishing of off-axis growth at the tip of the RH (Gage
2004) results in the entrapment of bacteria inside the curled RH tip within 24h after
inoculation (hai) with rhizobia (timing of subsequent events is presented after [Timmers
et al., 1999; Lohar et al., 2006]). NF is required and sufficient for this process (Esseling
et al., 2003 and refs therein).
Entrapped bacteria form a microcolony before initiating (within 48hai)
the epidermal infection program: a local cell wall degradation and invagination of
13
the plant cell membrane within the crook of the curl (Oldroyd et al., 2011).
Subsequently, the plasma membrane (PM) invagination is extended into a tubular
structure within the RH cytoplasm, termed the infection thread (IT) (see Figure 1
step 2). Its growth is realized via an inverted tip growth-like mechanism that involves
vesicular trafficking and cytoskeleton functions on the plant side, and continuous
division of bacteria at the tip of the growing IT (Brewin 2004; Gage 2004; Fournier
et al., 2008; Timmers 2008; Oldroyd et al., 2011). As a result, the IT grows from
the curl towards the RH cell base and further traverses cortical cell layers, guiding
the bacteria towards a nodule primordium (see below). The transcellular growth
of the IT is guided by a specific polarized arrangement of the cytoplasm (called pre-IT)
in the underlying cortical cells. Presumed high local concentrations of NF within
the RH curl are proposed to act as a switch for the IT initiation (Miwa et al., 2006b),
and continued production of NFs seems essential for IT growth (see below). Moreover,
bacterial mutants with defects in surface polysaccharides are impaired in the initiation
or progression of ITs, indicating that additional rhizobial polysaccharides play
an important role in infection (Gibson et al., 2008; Deakin & Broughton 2009; Downie
2010). Especially exopolysaccharide I (EPSI or succinoglycan) is proposed to have
a signalling function in Medicago, modulating a plant defense response to rhizobial
infection (see Chapter 6) and further defining host-specificity of the RL symbiosis
(Simsek et al., 2007; Jones & Walker 2008).
Also pericycle and cortical cells respond in a local manner to compatible
rhizobia or NF: activation of the pericycle cells opposite a protoxylem pole
(approximately 16-24hai) is followed by dedifferentiation of cells in the inner,
and subsequently middle cortex, which start dividing and form a nodule primordium
within 24-48hai (see Figure 1 step 2). An elaborate IT network (Monahan-Giovanelli
14
et al., 2006) spreads into a central region of the nodule primordium within 72hai
(see Figure 1 step 3) to provide bacteria for the NF-dependent (at least in legume
species forming indeterminate-type nodules, see below) endocytotic(-like) release
(Brewin 2004; Gibson et al., 2008; Ivanov et al., 2010). This results in the formation
of organelle-like structures, termed symbiosomes, containing a bacterium enclosed by
a host-derived symbiosome (or peribacteroid) membrane. Bacteria within symbiosomes
undergo morphological and metabolic differentiation into their symbiotic state (able to
reduce N2), termed bacteroid. This process is imposed and regulated by the host
(Kereszt et al., 2011) and seems to be NF-independent (Schlaman et al., 1991).
Simultaneously, the symbiosome membrane acquires specialized functions to regulate
metabolite exchange between bacteroids and cytoplasm of the host nodule cell (Gibson
et al., 2008; Den Herder & Parniske 2009; Oldroyd et al., 2011). Additional rhizobial
surface polysaccharides play important roles both in bacterial release from the IT
and in bacteroid persistence inside nodule cells (Gibson et al., 2008).
Figure 1. Developmental stages during formation of an indeterminate nodule.
Step 1) Emerging RHs in the susceptible root zone respond to NF secreted by rhizobia: the tip of the RH curls
around the attached rhizobia (in red), entrapping the dividing bacteria in a tight pocket (an infection focus).
Simultaneously, NF perception induces periclinal cell divisions in the underlying pericycle (opposite
a protoxylem pole), followed by inner cortical cell proliferation (marked with *).
Step 2) ITs (in red) emerge from the RH curl and grow transcellulary towards the dividing cortical cells.
Simultaneously, additional cortical cell layers divide, leading to the formation of a nodule primordium.
Step 3) Invasion of ITs into the nodule primordium is concomitant with the establishment of a nodule
primordium meristem (in blue). From this stage, development of the nodule primordium is driven by
the activity of this persistent meristem.
Step 4) Continuous cell division activity and growth of the ITs into cell layers underlying the meristem lead to
a zonation observed in fully mature nodules with: the meristem, the invasion zone, the interzone, the fixation
zone, and the senescence zone.
After Popp & Ott (2011), changed.
* Expression of MtNFP and MtLYK3 in specific cell layers during RL interaction is highlighted (after
Limpens et al. [2005], Arrighi et al. [2006], Mbengue et al. [2010], Haney et al. [2011]).
15
16
Cortical cell divisions must be provoked from a distance, given
the accumulation and immobilization of NFs in plant cell walls. Therefore, NF signaling
in the root epidermis is predicted to activate complex hormonal signaling
(with a prominent role of ethylene and auxin-to-cytokinin level) in the underlying cell
layers that determine nodule positioning and organogenesis (Crespi and Frugier 2008;
Desbrosses & Stougaard 2011; Oldroyd et al., 2011). Development and functioning
of nodules are host-determined and vary substantially among legume species (Den
Herder & Parniske 2009). Elongated indeterminate nodules produced e.g., by Medicago
spp. and pea, require the activation of inner cortex cell layers and are characterized by
the persistence of an apical meristem. At full maturation, they display a spatial
differentiation gradient visible as different developmental zones (see Figure 1 step 4):
the apical meristem (zone I); the invasion (prefixation) zone II with invading ITs
and release of rhizobia; the interzone II/III; the fixation zone III with infected
and uninfected cells and bacteroids actively reducing N2; and the senescence zone IV
(Maunory et al., 2010 and refs therein). Similar differentiation steps occur during
development of round-shaped determinate nodules produced e.g., by soybean (Glycine
max) and Lotus spp., but in a time-sequential manner. In addition, determinate nodules
arise from the outer or central cortex and have a transient meristem.
Successful formation of functional nodules requires an integration of two
seperate programs: the infection process initiated in the root epidermis,
and the development of a nodule primordium in the cortex (Frugier et al., 2008;
Oldroyd & Downie 2008). Identification of various mutants (Popp & Ott 2011) affected
in IT initiation/growth, in which cortical cell divisions are still present but the level
of nodule organogenesis is dictated by the extent of IT growth, confirms the close
synchronization between these two programs. However, the mechanism(s) responsible
17
for this coordination is currently not known. In addition, the total number of functional
nodules is controlled by autoregulatory mechanisms that limit the extent of nodulation
upon the formation of first nodules or in a situation of sufficient availability of ammonia
in the soil (Ferguson et al., 2011; Mortier et al., 2012).
2. Molecular events in Medicago truncatula-Sinorhizobium meliloti interaction
In the following paragraph, I will present molecular events that take place
in RHs upon NF perception (combining the findings in different legume species),
and then characterize putative receptors and molecular components required for NF
signaling in Medicago and Lotus (see Figure 2). Subsequently, the molecular
components required for initiation of the epidermal infection program, and signaling
events driving the organogenesis of a nodule will be presented.
2a) NF-induced responses in the root hair cell
The NF is required and sufficient for the induction of early host responses
in the RH (see Figure 2) and in the cortex. In many legume species, distinct patterns
of sustained oscillations of [Ca2+
] in (Sieberer et al., 2009; Krebs et al., 2012)
and around (Harris et al., 2003; Shaw & Long 2003 and refs therein; Kosuta et al.,
2008) the nucleus, termed Ca2+
spiking, can be induced with subnanomolar to picomolar
[NF]. Nanomolar to subnanomolar [NF] induces a rapid (den Hartog et al., 2003)
and transient activation of phospholipase C (PLC) and D, resulting in phospholipid
signaling and accumulation of phosphatidic acid (PA) and diacylglycerol pyrophosphate
in vetch (Vicia sativa) roots (den Hartog et al., 2001). PLD activation has also been
reported in Medicago and Lotus roots and soybean RHs (Charron et al., 2004; Wan
et al., 2005; Serna-Sanz et al., 2011 used submicromolar [NF]). In addition, a rapid
18
and transient increase in reactive oxygen species (ROS) production in RHs has been
shown in bean (Phaseolus vulgaris) (Cárdenas et al., 2008), and activation/translocation
of yet-unidentified small GTP-binding proteins has been reported in cowpea (Vigna
unguiculata) roots (Kelly & Irving 2003; Kelly-Skupek & Irving 2006). Induction
of many genes encoding components involved in Ca2+
transport/ binding or ROS
metabolism as early as 1hai in Medicago roots indicates the essential signaling role
of Ca2+
and ROS during the initiation of RL symbiosis (Lohar et al., 2006).
The most thoroughly studied of the RH responses to NF is Ca2+
spiking,
which can be elicited within 10 minutes with NF concentration down to 10-12
-10-13
M
(Oldroyd et al., 2001), and mimics RH response to compatible rhizobia (Wais et al.,
2002; Harris et al., 2003). The same range of [NF] is sufficient for triggering changes
in the host root transcriptome, as demonstrated by the induction of so-called early
nodulin (ENOD) gene expression (e.g. Charron et al., 2004 and refs therein; Miwa
et al., 2006a) in the root epidermis and cortex (see below). Impairment of the generation
or perception of Ca2+
spiking affects the induction of NF-responsive genes, implicating
nuclear calcium oscillations as the switch for NF-mediated gene regulation (Charron
et al., 2004 and refs therein; Oldroyd & Downie, 2006; Miwa et al., 2006a; see below).
Activation of PLD, and possibly of PLC is required for induction of expression
of several ENOD genes (Charron et al., 2004 and refs therein), and these enzymes
are postulated to generate (a) potential second messenger(s), such as PA, that would
provide the link between NF perception at the PM and regulation of gene expression
(see below). On the other hand, identification of genes whose transcriptional regulation
seems to be independent from Ca2+
spiking (Weidmann et al., 2004; Kistner et al.,
2005; Sanchez et al., 2005; Amiour et al., 2006; Lohar et al., 2007; Massoumou et al.,
2007; Siciliano et al., 2007; Kuhn et al., 2010; Murray et al., 2011; Takeda et al., 2011)
19
indicates that NF-responsive genes are regulated via more than one signaling pathway.
The transient ROS production in the root epidermis in response to NF is postulated
to regulate expression of some early NF-responsive genes (Ramu et al., 2002).
Another process that seems to be independent from Ca2+
spiking is NF-induced
rearrangement of the actin cytoskeleton (Esseling et al., 2004) observed in RHs of many
legume species and implicated in RH deformation response (Timmers 2008; Yokota
et al., 2009; Oldroyd et al., 2011). As the NF-induced RH curling is a form of polarized
tip growth (Lee & Yang 2008; Cárdenas 2009), this process is likely regulated by Ca2+
(but not in a form of Ca2+
spiking), phospholipids, and ROS signaling (Ivashuta et al.,
2005; Lohar et al., 2007; Pelleg-Grossman et al., 2007; Cárdenas et al., 2008). So far,
RH deformations and/or curling have been shown to require homologs of known small
GTPases from ROP and RAB families (Pelleg-Grossman et al., 2007; Blanco et al.,
2009; Liu et al., 2010; Riely et al., 2011), as well as legume-specific molecular
components (Lohar et al., 2007 and refs therein).
Finally, NFs induce Ca2+
influx at the RH tip, as well as cell membrane
depolarization associated with ion fluxes across the plasma membrane, but as these
responses require much higher (approximately 10-8
M) [NF] (Cárdenas et al., 2000;
Shaw & Long, 2003; Smit et al., 2005; Miwa et al., 2006b;), their function during
the establishment of RL symbiosis is currently not understood.
2b) Genetic components of the early NF signaling located at the plasma membrane
In model legume species, early responses to compatible rhizobia/NF depend on
the function of (a) receptor-like kinase(s) (RLKs): Medicago MtNFP (for Nod Factor
Perception), and Lotus LjNFR1 and LjNFR5 (for Nod Factor Receptor 1 and 5) (Ben
Amor et al., 2003; Madsen et al., 2003; Radutoiu et al., 2003, 2007; El Yahyaoui et al.,
20
2004; Mitra et al., 2004a; Miwa et al., 2006b; Høgslund et al., 2009; Nakagawa et al.,
2010; Serna-Sanz et al., 2011). MtNFP, LjNFR1, and LjNFR5 share a similar structure:
a putative ligand-binding extracellular region (ExR) predicted to contain three LysM
domains, a single-pass transmembrane (TM) helix, and an intracellular region (InR)
containing a Ser/Thr protein kinase domain (KD) (Madsen et al., 2003; Radutoiu et al.,
2003; Arrighi et al., 2006; Mulder et al., 2006). MtNFP promoter is active already
in the epidermis of uninoculated roots and is responsive to inoculation
with rhizobia (Arrighi et al., 2006; for MtNFP role at later steps of RL symbiosis,
see below). In silico docking of compatible NF structures to the modeled LysM
domain(s) of MtNFP and LjNFR5, and structure-function studies demonstrating
the dependence of the biological activity of these LysM-RLKs on the specific structure
of NF support their function as putative NF receptors (Mulder et al., 2006; Radutoiu
et al., 2007; Bek et al., 2010; Bensmihen et al., 2011). In addition, co-expression
of LjNFR1 and LjNFR5 genes in Medicago and Lotus filicaulis allows nodule
organogenesis and (inefficient) infection of these species with bacterial strains normally
nodulating only Lotus (Radutoiu et al., 2007). Therefore, both LjNFR1 and LjNFR5
are postulated to act as putative NF receptors and to co-function in the determination
of specificity of RL interaction. However, no direct binding of NFs to any of these
proteins has been shown so far.
Similarly, not much is known about the signaling mechanism employed by
these LysM-RLKs. The kinase domain of LjNFR1 is able to autophosphorylate
and transphosphorylate the LjNFR5 InR in vitro, and kinase activity seems to be
essential for LjNFR1 biological activity (Madsen et al., 2011; Tόth et al., 2012).
On the contrary, LjNFR5 and MtNFP possess inactive KDs and most likely do not show
or rely on the intrinsic phosphorylation activity to signal (Radutoiu et al., 2003; Arrighi
21
et al., 2006; Madsen et al., 2011; Lefebvre et al., 2012). Instead, they are predicted to
form a receptor complex: LjNFR5 with LjNFR1 and MtNFP with an unknown LysM-
RLK (Radutoiu et al., 2003, 2007; Arrighi et al., 2006; Smit et al., 2007). Recent
findings that MtNFP is not directly responsible for the recognition of a sulphate group
on the S. meliloti NF during the early steps of NF signaling (Bensmihen et al., 2011)
supports the hypothesis that another LysM-RLK-encoding gene in Medicago (Limpens
et al., 2003; Arrighi et al., 2006) may constitute a yet-unidentified NF receptor.
An additional RLK in Medicago, MtDMI2/MtNORK (for Does not Make Infections
2/Nodulation Receptor Kinase; from now on referred to as MtDMI2)
and Lotus, LjSymRK (for Symbiosis Receptor-like Kinase), is localized in the root
epidermis (Bersoult et al., 2005; Kevei et al., 2007; Den Herder et al., 2012).
It possesses an N-terminal domain of unknown homology and three Leucine-rich
repeats (LRRs) in its ExR, and an active protein KD in its InR (Endre et al., 2002;
Stracke et al., 2002; Yoshida & Parniske 2005; Kosuta et al., 2011; Tόth et al., 2012).
Nodulation defects of Mtdmi2 and Ljsymrk mutants can be rescued by their orthologs
from species within the Rosids clade, nodulated by different rhizobia strains
or by Frankia bacteria (Gherbi et al., 2008; Markmann et al., 2008), indicating that
this LRR-RLK does not contribute to specificity of RL interaction. MtDMI2/LjSymRK
is required for cortical cell divisions and efficient RH curling in response to the NF
(Catoira et al., 2000; Esseling et al., 2004 and refs therein; for MtDMI2 role at later
steps of RL symbiosis, see below). It also functions during arbuscular mycorrhiza (AM)
symbiosis, where it is required for intracellular passage of fungus through outer root
tissue layers (Kosuta et al., 2011 and refs therein; Demchenko et al., 2004 and refs
therein; Genre et al., 2005 and refs therein; Morandi et al., 2005).
22
ENODs
NSP2
Ca2
+
mV
K+
PLD
MtDMI1/LjPollux, LjCastor
Voltage gated
calcium channel?
Calcium ATPase
MtMCA8
Ca2+
NADPH oxidase
MtNFP/LjNFR5 MtLYK?/LjNFR1
NF MtDMI2/LjSymRK
MtIPD3/LjCYCLOPS
MtDMI3/LjCCaMK
NSP1
Ca2+
NucleusPM
PA
Ca2+ spiking
PLC
H2O2
Calcium
ATPase
Calcium
channel
nucleopore
complex:
LjNUP85
LjNUP133
LjNENA
PA
IP3
Figure 2. The NF signaling in a root hair cell in legumes.
The molecular components are labeled above.. Thin black arrows indicate direction of ion fluxes or products
of enzymatic activities. Predicted signaling events are indicated with thick arrows Hypothesized signaling
events and processes are indicated with dashed arrows.
NF is postulated to be perceived at the PM by two LysM-RLKs (in green) forming a complex (underlined):
MtNFP/LjNFR5 is postulated to interact with a kinase-active (marked with a red star) MtLYK/LjNFR1.
Signal transduction is likely to be mediated by the kinase activity of the latter protein, and is postulated
to result in activation of an LRR-RLK (in yellow), MtDMI2/LjSymRK. Additional components associated
with the PM and (postulated) to function in the NF signaling include: NADPH oxidase (in orange); PLC
and PLD (in purple); and Ca2+ channel and Ca2+ ATPase (in grey). NADPH oxidase, PLC, and PLD
are activated rapidly upon NF perception (via unknown mechanisms) and are postulated to generate secondary
signals for generation of Ca2+ spiking. These signals could be: H2O2, phosphatidic acid (PA) or inositol
phosphates (IP3).
NF-induced signaling at the PM activates the Ca2+ spiking (indicated in red) in and around the nucleus.
The Ca2+ spiking is generated via synchronous release and uptake of Ca2+ from the sarco/ER by an unknown
Ca2+ channel and a Ca2+ ATPase (MtMCA8), respectively, and requires function of (an) ion channel(s),
MtDMI1/LjPollux and LjCASTOR, and of nucleopore complex (in brown). Ca2+ spiking plays an essential
role in regulation of NF-responsive gene expression (indicated with a black arrow), including Early Nodulin
genes. A nuclear complex including a Ca2+/calmodulin-dependent kinase, MtDMI3/LjCCaMK, and MtIPD3/
LjCYCLOPS is postulated to interact with two main transcriptional regulators, Mt/LjNSP1 and Mt/LjNSP2,
which translocate to the nucleus (indicated with a blue arrow) and dimerize upon NF perception.
A set of genes is regulated independent of Ca2+ spiking, likely via a parallel signaling pathway activated by
the LysM-RLK complex (indicated with a green arrow).
23
Aberrant RH curling observed in the Mtdmi2/Ljsymrk mutants places
the encoded LRR-RLK downstream of MtNFP and LjNFR1/LjNFR5 in the NF
signaling pathway (Esseling et al., 2004). During both RL and AM symbioses,
MtDMI2/LjSymRK is required for the induction of Ca2+
spiking (Wais et al., 2000;
Oldroyd et al., 2001; Shaw & Long, 2003; Miwa et al., 2006b; Sieberer et al., 2009;
Chabaud et al., 2011 and refs therein; Sieberer et al., 2012) and transcriptional
reprogramming (see below) in response to, respectively, NF and diffusible signals
secreted by the AM fungus, termed Mycorrhization (Myc) factors. However, an exact
mechanism of the postulated MtDMI2/LjSymRK activation via the NF receptor
(complex) (Radutoiu et al., 2003), as well as downstream signaling from this LRR-RLK
remain largely unknown. LjSymRK is able to phosphorylate the LjNFR5 InR in vitro
(Madsen et al., 2011), although the effect of this phosphorylation on nodulation remains
to be demonstrated. MtDMI2/LjSymRK InR interacts in vivo, respectively, with
a 3-hydroxy-3-methylglutaryl coenzyme A reductase 1 (MtHMGR1) and an ARID
DNA-binding protein, termed SymRK-Interacting Protein 1 (LjSIP1) (Kevei et al.,
2007; Zhu et al., 2008). However, distinct isoforms of HMGRs display metabolic
specialization, and a specific branch of the mevalonate/isoprenoid pathway in which
MtHMGR1 is involved has to be identified before one can predict its function during
RL and/or AM symbiosis. LjSIP1 specifically binds the LjNIN promoter in vitro
and is proposed to control the induction of this putative transcriptional regulator
in response to rhizobia inoculation (see below) but it remains to be analyzed when
and where this activity is required for LjSymRK function in vivo. Interestingly, protein
encoded by LjSymRK homologs from monocots that lack the N-terminal and one LRR
domain in their ExR, are able to rescue only the AM phenotype of the Ljsymrk mutant
(Gherbi et al., 2008; Markmann et al., 2008), indicating that domain composition
24
of the LjSymRK ExR is important for its function in AM. In addition, a point mutation
within LjSymRK ExR uncouples the epidermal and cortical symbiotic programs in both
RL and AM symbioses (Kosuta et al., 2011). This suggests that LjSymRK (and possibly
also of MtDMI2) may perceive non-equivalent signals in the tissue-dependent manner,
possibly including activation by the putative NF/Myc factor receptors, and direct/
indirect perception of a yet-unknown external stimulus. In fact, RHs of the Mtdmi2/
Ljsymrk mutant show an increased susceptibility to mechanical stimuli, which may
provide a clue to understanding of MtDMI2/LjSymRK role during root endosymbioses
(Esseling et al., 2004; Holsters 2008).
2c) NF-induced Ca2+
spiking and gene expression in the root epidermis
Following NF perception at the PM, the majority of components implicated in
the NF signaling are linked to the nucleus. These include components involved in
the generation of Ca2+
spiking: (a) putative Ca2+
channel(s) and Ca2+
ATPase(s)/pump(s)
(MtMCA8) involved, respectively, in Ca2+
release from and sequestration in the sarco/
endoplasmic reticulum (Oldroyd & Downie 2006; Capoen et al., 2011); (a) predicted
cation channel(s) located in the nuclear envelope, termed MtDMI1 (for Does not Make
Infections 1)/LjPOLLUX and LjCASTOR (Ané et al., 2004; Peiter et al., 2007; Riely
et al., 2007; Charpentier et al., 2008 and refs therein; Capoen et al., 2011),
that are proposed to regulate activity of Ca2+
channel(s) located on the same membrane
system or function as counter ion channels during Ca2+
release; and three putative
nucleoporins (LjNUP85, LjNUP133 and LjNENA), which are probably involved in
transport or localization of the factor(s) needed for the induction of Ca2+
spiking
(Kanamori et al., 2006; Saito et al., 2007; Groth et al., 2010). A nuclear Ca2+
/
calmodulin-dependent kinase (CCaMK), termed MtDMI3 (for Does not Make
25
Infections 3)/LjCCaMK, and a coiled-coil protein, termed MtIPD3/LjCYCLOPS
(for Interacting Protein of DMI3), are proposed to form a complex whose activity
is regulated in response to Ca2+
spiking (Lévy et al., 2004; Mitra et al., 2004b; Smit
et al., 2005; Gleason et al., 2006; Tirichine et al., 2006; Messinese et al., 2007; Yano
et al., 2008; Horváth et al., 2011).
Components involved in the generation and perception of Ca2+
spiking
are required for the transcriptional reprogramming in response to the NF (see refs
for the respective components, and El Yahyaoui et al., 2004; Mitra et al., 2004a; Kistner
et al., 2005; Middleton et al., 2007; Høgslund et al., 2009; Murray et al., 2011).
They are also required for transcriptional reprogramming in response to Myc factors
(see refs for the respective components, and Harris et al., 2003; Kistner et al., 2005;
Sanchez et al., 2005; Amiour et al., 2006; Lohar et al., 2007; Siciliano et al., 2007
and refs therein; Kuhn et al., 2010; Takeda et al., 2011 and refs therein). Therefore,
they constitute a so-called common symbiosis (SYM) pathway (CSP), and the genes
encoding them are termed common symbiosis (SYM) genes (Kistner & Parniske 2002).
Constitutive active LjCCaMK variants (Tirichine et al., 2006; Yano et al., 2008) allow
the establishment of AM and RL symbioses independently of the Ca2+
spiking (Madsen
et al., 2010; Hayashi et al., 2010; the involvement of LjNENA has not been studied).
Therefore, the sole role of these components seems to be LjCCaMK activation.
On the contrary, LjNFR1 and LjNFR5 are required for RH curling and IT initiation
in Lotus, indicating that other signals, in addition to Ca2+
spiking, are required for
rhizobial infection via RHs (Held et al., 2010; Madsen et al., 2010; Hayashi et al.,
2010). It remains to be shown whether the same is true in Medicago.
The identification of CSP has raised questions on the mechanism that confers
specific host responses to NF or Myc factor(s). It has been suggested that the respective
26
receptors induce parallel signaling pathways, whose integration would modulate
the host responses according to the identity of the microsymbiont. Corroborating
this model, genes expressed specifically during AM symbiosis are not induced
in nodules formed spontaneously in Lotus (Takeda et al., 2011) due to the constitutive
activity of LjCCaMK. Therefore, induction of a single signaling pathway
(i.e. the CCaMK-mediated trascriptional reprogramming) does not seem to result in
switching on the entire and specific host symbiotic program, in agreement with
the identification of genes whose expression is independent of the CSP.
Transcriptional reprogramming during RL interaction depends on: two master
regulators from GRAS family, MtNSP1/LjNSP1 and MtNSP2/LjNSP2 (for Nodulation
Signaling Pathway 1 and 2) (Catoira et al., 2000; Oldroyd & Shaw 2003; Mitra et al.,
2004a; Kaló et al., 2005; Smit et al., 2005, Heckmann et al., 2006; Murakami et al.,
2006; Hirsch et al., 2009); various transcription factors (see especially El Yahyaoui
et al., 2004; Lohar et al., 2006; Udvardi et al., 2007), and one membrane-anchored
putative transcriptional regulator, MtNIN1/LjNIN (for Nodule Inception) (Schauser
et al., 1999; Marsh et al., 2007). The early transcriptional reprogramming in response to
rhizobia/NF involves genes related to signal transduction and gene regulation,
cytoskeleton structure and cell wall composition, and stress/defense-related genes
(see Lohar et al., 2006; El Yahyaoui et al., 2004; Kouchi et al., 2004; Høgslund et al.,
2009; and Chapters 2 and 6 for discussion).
2d) NF-dependent infection through root hair cells
The NF is the major bacterial signal required for the invasion of host roots
by rhizobia. There are stringent requirements towards the NF structure in species
forming indeterminate-type nodules: a significant reduction in IT numbers, as well as
27
arrest and impairment of IT growth and structure is a common response to rhizobia
mutants defective in the modification(s) of the NF GlcNAc backbone (Downie & Surin,
1990; Ardourel et al., 1994; Geurts et al., 1997; Walker & Downie, 2000; Catoira et al.,
2001; Wais et al., 2002; Limpens et al., 2003; Goormachtig et al., 2004a). In contrast,
these altered NFs are still capable of inducing RH deformations, Ca2+
spiking, gene
expression, and cortical cell divisions (see refs above and Wais et al., 2000; Oldroyd
et al., 2001; Charron et al., 2004). In order to explain different requirements of the host
responses for the modification(s) of the GlcNAc backbone, a single NF receptor
with different affinities for various NF structures has been suggested by Oldroyd et al.
(2001). In this model, induction of downstream signaling components essential for
the initiation of IT requires a higher level of NF receptor activation, which is achieved
only with the fully decorated NF structure. On the contrary, receptor activation
with suboptimal NF structures can still initiate RH deformations and cortical cell
divisions but is insufficient for IT formation. Corroborating this model, the efficiency
of Ca2+
spiking induction by differently modified chitin oligosaccharides (COs) depends
on their concentration used for root elicitation (Walker et al., 2000; Oldroyd et al.,
2001; Wais et al., 2002). Alternatively, Ardourel et al. (1994) have proposed
the existence of two different receptors in legume species forming indeterminate-type
nodules: the less stringent signaling receptor involved in the early, pre-infection step,
and the more stringent entry receptor essential for the infection.
MtNFP is proposed to constitute (part of) the signaling receptor, and (part of)
the entry receptor (Arrighi et al., 2006; Smit et al., 2007; Bensmihen et al., 2011).
In support of the latter hypothesis, a knock-down of MtNFP expression impairs the IT
growth (Arrighi et al., 2006), and Leu 154 (denoted as Leu 127 in Mulder et al., 2006),
positioned in the second LysM domain of MtNFP, is essential for efficient infection
28
of Medicago by S. meliloti (Bensmihen et al., 2011). As this residue is predicted
to interact with the fatty acid moiety of the S. meliloti NF (Mulder et al., 2006),
it is possible that MtNFP provides a stringent recognition of the NF lipo-chito-
oligosaccharidic backbone (Bensmihen et al., 2011).
In addition to MtNFP, MtLYK3/MtHCL (for LysM domain-containing RLK 3/
Root Hair Curling; from now on referred to as MtLYK3) and MtLYK4 are involved in
the infection process in a NF structure-dependent manner (Limpens et al., 2003; Smit
et al., 2007). The lack of IT formation or abnormal morphology of infrequently formed
ITs (and a reduction of nodule numbers in case of MtLYK3 knock-down) is observed
in plants with knocked-down MtLYK3 and MtLYK4 expression only upon their
inoculation with a S. meliloti mutant producing aberrant NF structure. MtLYK3
is expressed in the epidermis of uninoculated roots (Mbengue et al., 2010),
and the localization of MtLYK3-GFP (for Green Fluorescent Protein) fusion at the cell
periphery of RHs and along the IT membrane (Haney et al., 2011) agrees with
the proposed function of MtLYK3 in initiation and growth of ITs. In addition, Mtlyk3
loss-of-function mutant plants are strongly impaired in the formation of tight RH curls
containing rhizobial microcolonies, and pre-IT and IT formation, but show wild-type
RH deformations and Ca2+
spiking, reduced cortical cell divisions, and partial
transcriptional reprogramming in response to rhizobia/NF (Wais et al., 2000; Catoira
et al., 2001; El Yahyaoui et al., 2004; Mitra et al., 2004a; Middleton et al., 2007).
Both MtLYK3 and MtLYK4 encode LysM-RLKs with predicted three extracellular
LysM domains and an intracellular KD (Limpens et al., 2003; Arrighi et al., 2006).
In addition, MtLYK3 shows in vitro auto- and trans-phosphorylation activity,
and is negatively regulated by an E3 ubiquitin ligase, MtPUB1 (for Plant U-box Protein
1; Mbengue et al., 2010). As both MtNFP and MtLYK3 function during the infection
29
process in a NF-dependent manner, they are postulated to form a NF receptor complex
in Medicago (Arrighi et al., 2006; Smit et al., 2007; Bensmihen et al., 2011).
NFs continue to be produced by rhizobia within ITs (Schlaman et al., 1991;
Timmers et al., 1998; Den Herder et al., 2007). In addition, upon mixed inoculation
of Sesbania roots with two rhizobial strains: fast growing, defective in NF production
(nod-) and slow growing, defective in surface polysaccharide synthesis (nod+), distant
NF signaling provided by the nod+ strain impairs the IT progress through the cortex
(Den Herder et al., 2007). Therefore, NF signaling plays an essential role
in the regulation of IT growth. Initiation and growth of ITs in Medicago, Lotus
and Sesbania require the function of common SYM genes (Capoen et al., 2005, 2009;
Godfroy et al., 2006; Yano et al., 2008; Horváth et al., 2011; Ovchinnikova et al.,
2011), indicating that Ca2+
spiking is required also at the stage of rhizobial infection.
Corroborating this, outer cortical cells exhibit low frequency Ca2+
spiking during
the pre-infection stage that changes into high-frequency spiking concomitantly
with intracellular infection (Miwa et al., 2006a; Sieberer et al., 2012). In addition,
NF signaling displays a remarkable regulatory effect in Sesbania which can be infected
via two alternating mechanisms: epidermal ITs are formed in RH initials only during
non-flooded conditions, whereas upon water-lodging cortical ITs originate from
infection pockets formed in cracks at the sites of lateral root emergence (Capoen et al.,
2010; see below). In this species, a specific NF-induced Ca2+
spiking pattern in RH
initials correlates with and allow for the initiation/growth of IT (Capoen et al., 2009).
Additional components involved in the IT progression include: an E3 ubiquitin
ligase, MtLIN/LjCERBERUS (Kiss et al., 2009; Yano et al., 2009); a protein containing
a Major Sperm Protein domain and several ankyrin repeats, MtVAPYRIN, possibly
involved in membrane biogenesis or trafficking (Murray et al., 2011); a coiled-coil
30
protein, MtRPG (Arrighi et al., 2008); two flotillins (Haney & Long 2010)
and a remorin, Mt/LjSYMREM1 (Lefebvre et al., 2010; Tóth et al., 2012).
2e) NF role in the intracellular accommodation of bacteroids
In alfalfa (Medicago sativa), NFs have been localized in the infection zone
of nodules, particularly in association with the ITs and in the cytoplasm of infected cells
(Timmers et al., 1998). Additionally, in Medicago the expression of MtNFP, MtLYK3,
and common SYM genes is localized first in the apex of a young nodule, and later on,
in the invasion zone of a mature nodule (Bersoult et al., 2005; Limpens et al., 2005;
Arrighi et al., 2006; Riely et al., 2007; Messinese et al., 2007; Mbengue et al., 2010;
Haney et al., 2011; Ovchinnikova et al., 2011). Knock-down of MtDMI2 or its ortholog
from Sesbania, SrSymRK, specifically in the nodule results in an extensive IT growth
in the nodule and the lack of symbiosome formation (Capoen et al., 2005; Limpens
et al., 2005), indicating the role of MtDMI2/SrSymRK in IT growth-restriction
and switching on the endocytosis program. In addition, a function in symbiosome
formation has been demonstrated also for other common SYM genes of Medicago
(Godfroy et al., 2006; Heckmann et al., 2006; Chen et al., 2007, 2009; Banba et al.,
2008; Markmann et al., 2008; Horváth et al., 2011; Ovchinnikova et al., 2011).
On the contrary, the CSP involvement during IT growth and symbiosome formation
in Lotus is currently not clear. Loss-of-function of LjSymRK does not abolish infection
of nodules formed due to the constitutive activity of LjCCaMK (Madsen et al., 2010;
Hayashi et al., 2010), and a rice (Oryza sativa) ortholog, OsSymRK, is able to restore
the formation of functional nodules in Ljsymrk but not in Mtdmi2 mutants (Banba et al.,
2008). On the contrary, presumed reduction of LjSymRK accumulation mediated by
an E3 ubiquitin ligase, LjSINA4 (for Seven in Absentia 4), impairs IT and symbiosome
31
formation (Den Herder et al., 2012). In addition, rice and Nicotiana benthamiana
orthologs, OsNSP1 and NbNSP1, can fully complement the Ljnsp1 mutant (Heckmann
et al. 2006; Yokota et al., 2010). Furthermore, LjNFR1 and LjNFR5 are not expressed
in nodules (Madsen et al., 2003; Radutoiu et al., 2003), and local production of NF
is not required for the correct symbiosome development in the determinate-type nodules
formed by Sesbania (Den Herder et al., 2007). Therefore, NF signaling seems to be
required for intracellular accommodation of rhizobia only in the indeterminate-type
nodules (Limpens et al., 2005; Horváth et al., 2011; Ovchinnikova et al., 2011).
2f) Different requirements for the NF structure during epidermal vs cortical infection
In some tropical legume species, the epidermal barrier to rhizobial infection
can be overcome at sites of lateral root emergence (termed crack entry or intercellular
invasion). After colonization of the crack, bacteria gain access to the developing nodule
primordium intercellularly or by inducing the formation of intracellular ITs in the root
cortex cells (Goormachtig et al., 2004b; Sprent 2007). In Sesbania, the crack entry
by Azorhizobium caulinodans involves ethylene- and ROS-dependent formation
of infection pockets in the cortex, and intracellular ITs in the root cortex cells
(Goormachtig et al., 2004a and refs therein; Capoen et al., 2010). Cortical infection
is also a NF-dependent process (D’Haeze et al., 1998, 2000; Op den Camp et al., 2011)
but shows a less stringent requirement for the NF structure: e.g. A. caulinodans mutants
producing aberrantly modified NF display, although reduced, ability to invade Sesbania
via cortical infection, whereas they are completely incapable of intracellular infection
via RH initials (D’Haeze et al., 2000; Goormachtig et al., 2004a). Therefore,
the stringent NF recognition seems to be necessary only for the formation of ITs
in the epidermis but not in the cortex (Capoen et al., 2010).
32
Recently, cortical infection has been reported for various Lotus mutants
affected in the epidermal but not the cortical symbiotic program, i.e. capable of nodule
organogenesis (Karas et al., 2005; Yokota et al., 2009; Madsen et al., 2010; Groth et al.,
2010; Kosuta et al., 2011). Mesorhizobium loti invasion of Lotus root via cracks
at the site of nodule emergence or on the surface of developing nodules is stimulated by
ethylene (Groth et al., 2010), and bacterial accumulation in intercellular pockets
is followed by the initiation of cortical ITs in the NF-dependent manner (Madsen et al.,
2010). Remarkably, formation of the cortical ITs is independent of LjNFR1/LjNFR5,
suggesting that these putative receptors are specifically required for the infection
through RHs, whereas other NF receptors might be involved in the initiation of ITs
by the root/nodule cortex cells. In contrast, no evidence for the crack entry in Medicago
has been found to date (Miyahara et al., 2010). Even more surprising is an observation
of infrequent formation of rhizobia invaded nodules on Lotus roots inoculated with
M. loti nod- mutant (Madsen et al., 2010). In this case, rhizobial infection involves
the intercellular epidermal entry and an invasion mechanism of nodule cells
via formation of an intracellular infection peg, termed single-cell pegentry. Infection
threads (epidermal or cortical) are never observed upon inoculation with M. loti nod-
mutants, confirming the strict requirement for NF during IT formation. This invasion
mode resembles the direct uptake of rhizobia from intercellular infection pockets by
nodule primordium cells in several tropical legume species (Goormachtig et al., 2004b).
2g) CCaMK-dependent nodule organogenesis in the cortex
Constitutive activation of CCaMK allows complete nodule formation
independent of NF perception, suggesting that it plays a central role during nodule
organogenesis (Gleason et al., 2006; Tirichine et al., 2006; Yano et al., 2008).
33
In Sesbania, SrCCaMK is required for A. caulinodans infection via RH initials
but not for crack entry (Capoen et al., 2009). Therefore, the absence of nodule
primordia upon SrCCaMK knock-down has demonstrated SrCCaMK role in nodule
primordium formation in a cell autonomous manner, consistent with SrCCaMK
expression in the nodule primordium and in the infection zone of a mature nodule.
Spontaneous nodulation (i.e. without rhizobia or NF perception) requires NSP1, NSP2,
and NIN (Gleason et al., 2006; Marsh et al., 2007; Middleton et al., 2007; Madsen
et al., 2010; Hayashi et al., 2010), whereas MtIPD3/ LjCYCLOPS appears to be less
essential for this process (Kistner et al., 2005; Yano et al., 2008; Horváth et al., 2011;
Ovchinnikova et al., 2011). Transcriptional reprogramming associated with nodule
primordium formation involves genes implicated in cell cycle; transcription
and translation; cell wall biosynthesis and remodeling; and secondary and hormone
metabolism (El Yahyaoui et al., 2004; Kouchi et al., 2004; Høgslund et al., 2009).
The role of NF-dependent cytokinin and auxin signaling during initiation
and differentiation of a nodule, as well as characterization of other molecular
components involved in nodule organogenesis (Frugier et al., 2008; Vernié et al., 2008;
Zhang et al., 2009a and refs therein; Heckmann et al., 2011; Oldroyd et al., 2011; Plet
et al., 2011 and refs therein) is beyond the scope of this chapter. However, it should be
noted that a constitutive active version of a cytokinin receptor, MtCRE1/LjLHK1
(for Cytokinin Response 1/Lotus Histidine Kinase 1) (Gonzalez-Rizzo et al., 2006;
Murray et al., 2007; Plet et al., 2011), functioning downstream of MtDMI2/LjCCaMK
(Hayashi et al., 2010; Madsen et al., 2010), is able to restore nodule organogenesis
but not infection in the Ljccamk mutant (Madsen et al., 2010). Moreover, invasion
of nodules formed on roots of Ljnfr1 nfr5 double mutants via the cortical infection
is possible only if nodule organogenesis is driven by constitutive active LjCCaMK
34
and not via constitutive active LjLHK1. Therefore, LjCCaMK seems to regulate nodule
organogenesis via or in co-operation with cytokinin signaling and it seems to regulate
the infection process.
3. Signaling pathway(s) shared between Rhizobium-legume and arbuscular
mycorrhiza endosymbioses
AM constitutes an evolutionarily old symbiosis between fungi
of the Glomeromycota phylum and a wide range of plant species, including legumes
(Bonfante & Genre 2010; Humphreys et al., 2010). Ongoing identification of the role
of Medicago and Lotus common SYM genes in AM (see refs on page 21 and 25,
and Yokota et al., 2010; Horváth et al., 2011; Ovchinnikova et al., 2011; Murray et al.,
2011), and of their phylogenetic and functional conservation in non-nodulating species
capable of establishing AM (Charpentier & Oldroyd 2010; Gough & Cullimore, 2011)
supports the hypothesis of AM being „the mother of plant root symbioses“.
More specifically, it is postulated that certain molecular and cellular processes operating
during AM symbiosis have been recruited from the AM during later evolution of RL
symbiosis (Ivanov et al., 2012 and refs therein; see Chapter 6).
Because putative NF receptors are not essential for the establishment of AM
symbiosis (Wegel et al., 1998; Ben Amor et al., 2003; Radutoiu et al., 2003),
it is generally assumed that AM symbiosis is controlled by other receptor(s) specific
for Myc factor(s). Interestingly, a single MtNFP-like putative receptor of Parasponia
andersonii (Parasponia), PaNFP, is involved in interactions of this nonlegume species
with both Sinorhizobium sp. NGR234 and Glomus intraradices (Op den Camp et al.,
2011). Knock-down of PaNFP results in the lack of intracellular accommodation
of rhizobia and AM fungus. Findings in Parasponia indicate that during evolution,
35
a Myc factor receptor, as part of the CSP, was recruited to function in RL symbiosis
(Streng et al., 2011), and that structural diversification of present NF and Myc factor
receptors is a result of subsequent gene duplication and neofunctionalization for specific
ligand recognition (Zhang et al., 2009b; Young et al., 2011).
Recently, sulphated and nonsulphated tetrameric and pentameric LCOs
secreted by G. intraradices have been proposed as specific Myc factors that stimulate
mycorrhization both on legume and non-legume species (Maillet et al., 2011). Close
structural similarity of G. intraradices putative Myc factors to S. meliloti NFs raises
questions about the specificity of the signals used by both microsymbionts. (Maillet
et al., 2011; see Gough & Cullimore 2011 for discussion). In comparison with
the sulphated Myc LCO, the non-sulphated Myc factor enhances the number
of mycorrhizal infection sites but is less active in promoting lateral root formation
(Maillet et al., 2011). The higher activity of the sulphated Myc LCO for the latter
response is suggested to result from the (unspecific) induction of the NF signaling
pathway, which also shows a stimulatory effect in this respect in Medicago (Oláh et al.,
2005). However, whether or not MtNFP is involved in the stimulation of lateral root
formation also in response to non-sulphated putative Myc factor is unclear
at the moment (Ben Amor et al., 2003; Oláh et al., 2005; Maillet et al., 2011). Since
MtNFP is dispensable for perinuclear Ca2+
spiking in Medicago roots in response to
Gigaspora margarita spore exudate (Chabaud et al., 2011), the putative Myc factors are
likely to be perceived via (a) specific receptor(s) other than the putative NF receptors.
Accumulating evidence indicates the existence of at least two parallel signaling
pathways - dependent and independent of the CSP (see refs on page 26 and Genre et al.,
2005; Oláh et al., 2005) - are induced by AM fugal diffusible signal(s). Whether these
pathways operate in parallel or in series (i.e. one prior to and the second upon contact
36
with root epidermis) or are activated via (an) identical or different Myc factor(s),
remains to be identified.
4. Plant LysM domain-containing proteins involved in innate immunity
Plants have evolved a robust and inducible innate immune system to defend
against pathogens (Jones & Dangl 2006). First inducible responses are launched by
recognition of pathogen/microbe-associated molecular patterns (PAMPs/MAMPs)
or damage-associated molecular patterns (DAMPs) via surface receptors (Boller & Felix
2009; Thomma et al., 2011). Perception of the conserved molecular components
characteristic of a whole class of microbes (i.e. PAMPs/MAMPs) or effects
of the pathogenic action of microbes (i.e. DAMPs) on the surface of plant cells trigger
basal or PAMP-triggered immunity (PTI) resulting in the synthesis of many defense-
related proteins and secondary metabolites.
Insoluble chitin, a ß 1-4-linked GlcNAc homopolymer present in fungal cell
walls, is presumed to be digested by extracellular chitinases into shorter chitin
oligosaccharides (COs) that function as a PAMP. Their ability to activate innate
immunity in plants and animals (Ramonell et al., 2005 and refs therein; Miya et al.,
2007; Lee et al., 2008; Kishimoto et al., 2010; Lenardon et al., 2010) strongly depends
on the degree of polymerization and acetylation (Shibuya & Minami 2001; Hamel &
Beaudoin 2010). Plant cell responses to COs resemble those triggered by other PAMPs
(Garcia-Brugger et al., 2006; Nicaise et al., 2009; Boudsocq et al., 2010; Ranf et al.,
2011; Segonzac et al., 2011; Manzoor et al., 2012), and include: a burst in intracellular
[Ca2+
]; ROS production; induction of mitogen-activated protein kinase (MAPK)
cascade; activation of PLC; transcriptional reprogramming; increased chitinase activity;
synthesis of phytoalexins; and deposition of callose and cell wall lignification
37
(Ramonell et al., 2002; Zhang et al., 2002; den Hartog et al., 2003 and refs therein;
Wan et al., 2004; Libault et al., 2007; Miya et al., 2007; Wan et al., 2008; Gimenez-
Ibanez et al., 2009; Hamel & Beaudoin 2010; Kishi-Kaboshi et al., 2010; Shimizu
et al., 2010; Segonzac et al., 2011). A complex signaling cascade induced upon COs
stimulation is being revealed, with a recently demonstrated hierarchy of molecular
responses in N. benthamiana (Segonzac et al., 2011); interaction network between
MAPKs and TFs in A. thaliana (Arabidopsis) (Son et al., 2012); and a postulated
formation of a “defensome”, composed of a putative chitin receptor complex, a ROP
(OsRac1), and several (co-)chaperons in rice (Chen et al., 2010) (see Figure 3).
In Arabidopsis, chitin signaling is linked to a LjNFR1/MtLYK3 homolog,
AtCERK1/LysM-RLK1 (for Chitin Elicitor Receptor Kinase 1; from now on referred to
as AtCERK1) (Miya et al., 2007; Wan et al., 2008; Zhang et al., 2009b and refs
therein). AtCERK1 is predicted to posses three LysM domains in its ExR, and activity
of its intracellular KD is required for CO-dependent phosphorylation of AtCERK1
in vivo; ROS production; activation of MAPK cascade; and transcriptional
reprogramming (Miya et al., 2007; Wan et al., 2008; Gimenez-Ibanez et al., 2009;
Petutschnig et al., 2010). Knock-out of AtCERK1 abolishes Arabidopsis resistance
to Alternaria brassicicola, Botrytis cinerea, Erysiphe cichoracearum, and Pseudomonas
syringae pv. tomato DC3000 (Miya et al., 2007; Wan et al., 2008; Gimenez-Ibanez
et al., 2009; Willmann et al., 2011; Brotman et al., 2012), indicating that AtCERK1
functions in innate immunity against fungal and bacterial pathogens. Corroborating this,
attenuation of AtCERK1 function constitutes an important step for the bacterial
virulence (Gimenez-Ibanez et al., 2009; Wu et al., 2011). In addition, AtCERK1 has
recently been implicated in COs-induced abiotic (salinity and heavy metals) stress
signaling (Brotman et al., 2012).
38
In addition, AtCERK1/OsCERK1 has recently been shown to co-function
with two LysM domain-containing extracellular proteins, AtLYM1/AtLYM3
and OsLYP4/OsLYP6, respectively, in peptidoglycan (PGN) signaling (Willmann et al.,
2011; Liu et al., 2012a) (see Figure 3). AtCERK1/OsCERK1 does not bind PGN
but is involved in signal transduction upon binding of this PAMP/MAMP to AtLYM1
and AtLYM3 or OsLYP4 and OsLYP6, respectively (Petutschnig et al., 2010;
Willmann et al., 2011; Liu et al., 2012a).
Similarly, COs signaling and resistance to a fungal pathogen in rice
are mediated by an AtCERK1 ortholog, OsCERK1, and three extracellular LysM
domain-containing proteins, OsCEBiP (for Chitin Elicitor-Binding Protein), OsLYP4
(for Lysin domain-containing protein 4), and OsLYP6 (Kaku et al., 2006; Kishimoto
et al., 2010; Shimizu et al., 2010; Liu et al., 2012a). OsCEBiP, OsLYP4 and OsLYP6
are putative extracellular LysM proteins anchored to the PM, predicted to possess
two or three LysM domains (Kaku et al., 2006; Fliegmann et al., 2011; Liu et al.,
2012a). OsCEBiP and OsLYP4&6, but not OsCERK1, show high affinity for COs
and are postulated to activate OsCERK1 upon COs binding (Shimizu et al., 2010; Liu
et al., 2012a). On the contrary, AtCERK is able to bind chitin and COs (Iizasa et al.,
2010; Petutschnig et al., 2010; Liu et al., 2012b). Since Arabidopsis genome contains
three OsCEBiP homologs (Wan et al., 2008) of which at least one (AtLYM2) encodes a
protein able to bind chitin (Petutschnig et al., 2010), it remains to be shown whether
AtCERK1 activation is autonomous or requires (a) LysM protein(s). Identification of
chitin perception components in other plant species (Fliegmann et al., 2011; Zeng et al.,
2012) should reveal whether they rely on a similar or different mechanism of chitin/CO
perception.
39
MAPKKK
MAPKK
MAPK
TFs
Calcium
ATPase
NADPH
Calcium
channel
Ca2+
CDPK
OsCERK1
AtCERK1?
NbCERK1?
H2O2
NucleusPM
COs/PGN-responsive genes
COs
OsCEBiP,
OsLYP4
OsLYP6
PGN
AtLYM1
AtLYM3
OsLYP4
OsLYP6? PLC
PA
IP3
AtCERK1
OsCERK1
P
Figure 3. The CERK1-mediated signaling in a plant cell in response to peptidoglycan (PGN)
and chitooligosaccharides (COs).
The molecular components are labeled above. Thin black arrows indicate direction of ion fluxes or products
of enzymatic activities. Predicted signaling events and processes are indicated with thick black arrows.
Hypothesized signaling events and processes are indicated with dashed arrows.
To the left: two LysM proteins, AtLYM1 and AtLYM3 in Arabidopsis (and possibly OsLYP4 and OsLYP6
in rice), are postulated to form a complex (underlined) and bind peptidoglycan (PGN) at the PM.
At/OsCERK1 (in green) is postulated to form a complex (underlined) with AtLYM1&3 or OsLYP4&6,
respectively, and become activated upon their binding to PGN. To the right: OsCEBiP, OsLYP4 and OsLYP6
bind COs at the PM. They are postulated to form complexes with OsCERK1 (in green), and to activate it
upon COs binding. Similar complexes involving AtCERK1 or NbCERK1 remain to be shown. In both PGN-
and COs-induced signaling, kinase activity of At/OsCERK1 (marked with a red star) is postulated to trigger
(via an unknown mechanism) an influx of extracellular Ca2+ (Ca2+ channel and Ca2+ ATPase required for this
are in grey). The resulting increase of cytosolic [Ca2+] is postulated to activate PM-associated NADPH
oxidase (in orange) directly via Ca2+ binding or indirectly via phosphorylation by a Ca2+-activated calcium-
dependent protein kinase [CDPK]. Activation of PLC (in purple), and the resulting generation of secondary
signals: phosphatidic acid (PA) and inositol phosphates (IP3) is shown in response to COs perception.
Activation of mitogen-activated protein kinase (MAPK) cascade (boxed) requires an oxidative burst signal
(including H2O2), and results in activation (via phosphorylation) of multiple transcription factors (TFs).
Upon activation, TFs translocate (indicated with a blue arrow) to the nucleus and regulate expression of PGN-
and COs-responsive genes.
40
41
CHAPTER 2
Functional interaction of Medicago truncatula NFP and LYK3
symbiotic receptor-like kinases results in a defense(-like) response
and cell death induction in Nicotiana benthamiana.
Anna Pietraszewska-Bogiel1, Benoit Lefebvre
2,3, Frank L.W. Takken
5, René Geurts
4,
Julie V. Cullimore2,3
, and Theodorus W.J. Gadella1
1 Section of Molecular Cytology, Swammerdam Institute for Life Sciences, University of Amsterdam,
Science Park 904, 1098 XH Amsterdam, The Netherlands
2 INRA, Laboratoire des Interactions Plantes-Microorganismes (LIPM), UMR441, F-31326 Castanet-
Tolosan, France.
3 CNRS, Laboratoire des Interactions Plantes-Microorganismes (LIPM), UMR2594, F-31326 Castanet-
Tolosan, France.
4 Department of Plant Science, Laboratory of Molecular Biology, Wageningen University,
Droevendaalsesteeg 1, 6708 PB Wageningen, The Netherlands
5 Section of Plant Pathology, Swammerdam Institute for Life Sciences, University of Amsterdam, Science
Park 904, 109 XA Amsterdam, The Netherlands
This work was supported by the EC Marie Curie Research Training Network Programme through contract
MRTN-CT-2006-035546 “NODPERCEPTION”, by the French National Research Agency (ANR) through
contract NodBindsLysM, and by the Dutch Science Organisation NWO (VIDI 864.06.007). We thank
Dr. Giulia Morieri (John Innes Centre, Norwich, UK) for providing purified Sinorhizobium meliloti NF,
Dr. Dörte Klaus-Heisen (CNRS-INRA, Toulouse, France) for providing pBin+ 35S::MtLYK3 [G334E]-
3xFLAG vector, and Dr. Harold van den Burg (Univeristy of Amsterdam, The Netherlands) for helpful
discussion.
42
SUMMARY
Receptor-like kinases (RLKs) with Lysin Motif (LysM) domains in their
extracellular region (LysM-RLKs) play crucial roles during plant interactions with
various pathogenic and beneficial microorganisms. Two LysM-RLKs of Medicago
truncatula (Medicago), MtNFP and MtLYK3, are postulated to co-function as putative
receptors for lipo-chitooligosaccharides (LCOs), termed Nodulation (Nod) factors
(NFs), which are indispensiable for the establishment of symbiosis with nitrogen-fixing
Sinorhizobium meliloti. We aimed at investigating putative molecular interactions
of these RLKs after their heterologous production in Nicotiana benthamiana
(Nicotiana) leaf. Surprisingly, co-expression of MtNFP and MtLYK3 resulted in
a defense(-like) response within 48 hours that included induction of stress/defense-
related gene expression, accumulation of phenolic compounds, and cell death (CD).
Interestingly, heterologous expression of AtCERK1, implicated in Arabidopsis thaliana
(Arabidopsis) innate immunity, led to a phenotypically similar response. Importantly,
MtNFP intracellular region, and MtLYK3 and AtCERK1 kinase activity
were necessary for CD induction, thereby mimicking the requirements for NF-
and chitin-induced signaling, respectively. Expression of either MtNFP or MtLYK3
alone or together with other constructs encoding unrelated RLKs did not result in CD,
indicating the requirement for a functional interaction of these LysM-RLKs.
In contrast, a homologous MtLYK2 was able to co-function with MtNFP for CD
induction in Nicotiana, supporting the hypothesized functional redundancy of LYK
proteins. The implications of these findings for NF-mediated signaling and possible
parallels between symbiotic and defense signaling are discussed.
43
INTRODUCTION
Perception of compatible NFs by specialized root hair (RH) cells on roots
of legume plants triggers complex cellular and molecular responses (D’Haeze &
Holsters 2002; Gibson et al., 2008) required for both nodule organogenesis
and infection by rhizobia (Murray 2011; Oldroyd et al., 2011; Popp & Ott 2011).
Several NF-induced processes, such as a transient increase of reactive oxygen species
(ROS) production (Bueno et al., 2001; Lohar et al., 2007; Cárdenas et al., 2008);
activation of phospholipase C (PLC) and D (PLD), and a resulting transient
accumulation of phosphatidic acid (den Hartog et al., 2001, Charron et al., 2004; Wan
et al., 2005; Serna-Sanz et al., 2011); and prolonged oscillations of nuclear
and perinuclear [Ca2+
], termed Ca2+
spiking (Oldroyd & Downie 2006; Sieberer et al.,
2009; Krebs et al., 2012), are implicated in switching on the symbiotic program.
Loss-of-function mutations in Medicago Nod Factor Perception (MtNFP), and Lotus
japonicus (Lotus) Nod Factor Receptor 1 and 5 (LjNFR1 and LjNFR5) abolish virtually
all NF-dependent responses (Ben Amor et al., 2003; Madsen et al., 2003; Radutoiu
et al., 2003; El Yahyaoui et al., 2004; Mitra et al., 2004a; Miwa et al., 2006b; Høgslund
et al., 2009; Nakagawa et al., 2010). Additionally, LjNFR1 and LjNFR5 function
is indispensible for host root infection via RHs (Hayashi et al., 2010; Madsen et al.,
2010), which requires active entrapment of rhizobia inside RH curls and formation
of so-called infection threads (ITs) through which the bacteria penetrate nodule
primordia. In agreement, MtNFP has been shown to be required for IT growth
and successful infection by rhizobia (Arrighi et al., 2006; Bensmihen et al., 2011),
where it is thought to co-function with another Medicago gene, LysM domain-
containing RLK/Root Hair Curling (MtLYK3/MtHCL; from now on referred to
44
as MtLYK3) (Catoira et al., 2001; Limpens et al., 2003; Smit et al., 2007; Haney et al.,
2011). Curiously, despite being a putative LjNFR1 ortholog, MtLYK3 is not essential
for the early NF-induced responses (Wais et al., 2000; Catoira et al., 2001; El Yahyaoui
et al., 2004; Middleton et al., 2007; Smit et al., 2007). Finally, both MtNFP
and MtLYK3 might still function in an invasion zone of a nodule (Limpens et al., 2005;
Arrighi et al., 2006; Mbengue et al., 2010; Haney et al., 2011), although their exact role
in this organ still awaits description.
All four genes encode RLKs with three LysM domains predicted
in the extracellular region (ExR), a single-pass transmembrane helix, and a protein
kinase domain (KD) within the intracellular region (InR) (Madsen et al., 2003;
Radutoiu et al., 2003; Arrighi et al., 2006; Mulder et al., 2006). Structure-function
studies demonstrating the dependence of the biological activity of these LysM-RLKs
on the specific structure of NF support their function as putative NF receptors (Radutoiu
et al., 2007; Bek et al., 2010; Bensmihen et al., 2011). However, binding of a specific
NF to any of these putative NF receptors or the exact mechanism of the signal
transduction employed by these RLKs still await characterization. Remarkably,
in contrast to MtLYK3 and LjNFR1, which display kinase activity, MtNFP and LjNFR5
seem to function as pseudokinases that neither show nor rely on the intrinsic kinase
activity to signal (Arrighi et al., 2006; Mbengue et al., 2010; Klaus-Heisen et al., 2011;
Madsen et al., 2011; Lefebvre et al., 2012). LjNFR5 is hypothesized to form a receptor
complex with LjNFR1: a notion consistent with their demonstrated co-functioning
during the determination of RL specificity (Radutoiu et al., 2007). Similarly, a receptor
complex composed of MtNFP and, respectively, a yet-unidentified LysM-RLK
or MtLYK3 is predicted to initiate the early symbiotic signaling and the infection
process (Arrighi et al., 2006; Smit et al., 2007). Since mutagenesis studies in Medicago
45
have not identified lesions in other genes that would mimic the Mtnfp mutant
phenotype, a function of this additional NF receptor in the early symbiotic signaling
is most likely redundant.
Intriguingly, LysM-RLKs in non-legume species also govern plant
interactions with microorganisms. For example, an MtNFP/LjNFR5 homolog
in Parasponia andersonii, PaNFP, is involved in interactions of this species
with Sinorhizobium sp. NGR234 and Glomus intraradices, resulting in nitrogen-fixing
and arbuscular mycorrhiza (AM) symbiosis, respectively (Op den Camp et al., 2011).
Arabidopsis LysM-RLK1/CERK1 (for Chitin Elicitor Receptor Kinase 1; from now on
referred to as AtCERK1) and its orthologs from other plant species are essential
for pathogen/microbe-associated molecular pattern (PAMP)-triggered immunity.
PAMPs/MAMPs are specific molecules conserved in various classes of micro-
organisms that act as elicitors/agonists of receptor-mediated defense signaling (Boller
& Felix 2009; Thomma et al., 2011). CERK1-mediated innate immunity to fungal
pathogens (Miya et al., 2007; Wan et al., 2008; Shimizu et al., 2010; Brotman et al.,
2012) is a consequence of its activation upon direct (in case of AtCERK1) or indirect
(in case of OsCERK1 from rice [Oryza sativa]) perception of chitin or chitin
oligosaccharides (COs) (Iizasa et al., 2010; Petutschnig et al., 2010; Shimizu et al.,
2010; Liu et al., 2012b). In addition, CERK1 also mediates innate immunity
to bacterial pathogen, Pseudomonas syringae pv. tomato DC3000 (Gimenez-Ibanez
et al., 2009; Willmann et al., 2011). This results from At/OsCERK1 activation
by LysM domain containing protein(s), respectively AtLYM1 and AtLYM3
or OsLYP4 and OsLYP6, that are able to bind peptidoglycan (PGN) (Willman et al.,
2011; Liu et al., 2012a). Plant cell responses to COs and PGN include: a burst
in cytosolic [Ca2+
], elevated ROS production, termed oxidative burst; induction
46
of mitogen-activated protein kinases (MAPKs) and PLC; induction of stress/defense-
related gene expression; synthesis of phytoalexins; deposition of callose and cell wall
lignification (den Hartog et al., 2003; Gust et al., 2007; Erbs et al., 2008; Hamel &
Beaudoin 2010; Kishi-Kaboshi et al., 2010; Millet et al., 2010; Shimizu et al., 2010;
Segonzac et al., 2011; Frei dit Frey et al., 2012 and refs therein). An oxidative burst,
activation of MAPKs, and transcriptional reprogramming in Arabidopsis seedlings
stimulated with COs and/or PGN depends on AtCERK1 function (A. Gust, personal
communication; Miya et al., 2007; Wan et al., 2008; Gimenez-Ibanez et al., 2009;
Petutschnig et al., 2010; Willmann et al., 2011; Liu et al., 2012b).
We are interested in NF signaling mediated by MtNFP and MtLYK3.
For this purpose, we employed various microscopical techniques to characterize
the localization and possible molecular interaction of these proteins in situ. However,
our attempts to visualize these proteins in Medicago root have been unsuccessful,
presumably due to stringent regulation of their accumulation. Therefore, we employed
an Agrobacterium-mediated transient transformation (AgroTT) of Nicotiana leaf
(Nguyen et al., 2010), which allowed us to produce both proteins to the extent that they
could be visualized with fluorescence microscopy. Remarkably, we found that
simultaneous accumulation of MtNFP and MtLYK3 in Nicotiana leaf resulted in
the induction of CD and associated defense(-like) response. Here, we present a detailed
characterization of the Nicotiana response to production of AtCERK1
and (co-)production of LysM-RLKs from Medicago. Our results indicate functional
interaction between MtNFP and kinase-active LYK protein that appears to trigger
stress/defense signaling in Nicotiana leaf, culminating in CD.
47
RESULTS
Co-expression of MtNFP and MtLYK3 in N. benthamiana leaf results in cell death
MtNFP and MtLYK3 cDNA sequences were fused C-terminally
to the sequence of a fluorescent protein (FP), i.e. super yellow fluorescent protein 2
(sYFP2) or mCherry, and were heterologously expressed in Nicotiana leaf using
Agro TT and a constitutive 35S promoter of the cauliflower mosaic virus (CaMV).
These or similar MtNFP and MtLYK3 fusion constructs were shown to complement,
Mtnfp and Mtlyk3 mutants, respectively (Haney et al., 2011; Klaus-Heisen et al., 2011;
Lefebvre et al., 2012), and are therefore suitable for studying in vivo localization
and biological activity of the encoded LysM-RLKs. Both sYFP2 and mCherry protein
fusions of MtNFP and MtLYK3 were efficiently produced in Nicotiana leaf epidermal
cells. In order to characterize their subcellular localization, leaf regions co-expressing
MtNFP-sYFP2 or MtLYK3-sYFP2 with a construct encoding a plasma membrane
(PM) marker (the hyper-variable region [HVR] of maize [Zea mays] ROP7 fused
N-terminally to mCherry; Ivanchenko et al., 2000) were analyzed using confocal laser-
scanning microscopy analysis. Twenty four hours after infiltration (hai), a complex
subcellular localization was observed for MtNFP-sYFP2 protein fusion: a significant
fraction of it was still trafficking through the endoplasmic reticulum (ER) network
(yielding the “patchy” fluorescent signal at the boundary and inside of the cells),
indicating incomplete PM localization of MtNFP-sYFP2 proteinfusion 24hai (Fig. 1
upper middle panel). To confirm the localization of MtNFP protein fusion 24hai,
MtNFP-sYFP2 was co-produced with mCherry fused C-terminally to HDEL signal,
resulting in its retention in the ER. Indeed, we observed good co-localization of both
fluorescent protein fusions (Fig. 1 upper panel). Approximately 36 to 48hai, clear
48
co-localization of MtNFP-sYFP2 protein fusion with the PM marker was observed,
indicating that MtNFP was ultimately localized to the PM in Nicotiana leaf epidermal
cells (Fig. 1 lower middle panel). MtLYK3-sYFP2 co-localization with the PM marker
at the cell boundary was observed already 24hai (Fig. 1 lower panel). These results
agreed with previous reports on subcellular localization of MtNFP and MtLYK3
in Nicotiana leaf epidermal cells (Klaus-Heisen et al., 2011; Lefebvre et al., 2012;
Lefebvre et al., 2010; Mbengue et al., 2010).
Surprisingly, MtNFP-FP and MtLYK3-FP co-expression resulted in collapse
and desiccation of the transformed cells within 48hai (Fig. 2A, Table 1), regardless
of the Agrobacterium tumefaciens strain used (i.e. GV3101::pMP90 or LBA4404)
(data not shown). This response was not dependent on a tag attached to MtNFP
or MtLYK3, since an identical response was observed upon simultaneous accumulation
of FP-tagged, 3xFLAG-tagged and untagged MtNFP and MtLYK3 (Fig. 2A, Table 1).
Importantly, expression of the separate MtNFP or MtLYK3 constructs or mock
infiltration did not induce CD, as confirmed with an exclusion dye (Evans blue)
staining (Fig. 2A). Cell death induction upon MtNFP and MtLYK3 co-expression
was also observed on tobacco leaves (Nicotiana tabacum cv. Samsun; Fig. 2B).
To investigate whether a similar CD response could be triggered by other
RLKs, we tested Nicotiana response to the expression of Medicago DMI2 (for Doesn’t
Make Infection 2; Endre et al., 2002) and MtLRRII.1 (Lefebvre et al., 2010),
and Arabidopsis BRI1 (Brassinosteroid Insensitive 1; Li & Chory 1997) driven by
the CaMV 35S promoter. Importantly, none of these RLKs alone or combination with
either MtNFP or MtLYK3 induced CD (Fig. 2C, Table 1), despite efficient production
of all tested protein fusions in Nicotiana leaf (data not shown).
49
mCherry-HDEL
NFP-sYFP2
24hai
HVR-mCherry
NFP-sYFP2
24hai
NFP-sYFP2
48hai
LYK3-sYFP2
24hai
channel: YFP mCherry merged & DIC
Figure 1. MtNFP and MtLYK3 C-terminal fusions to a fluorescent protein localize to the PM
of N. benthamiana leaf epidermal cells.
HVR-mCherry (encoding a PM marker) or mCherry-HDEL (encoding an ER marker) was co-expressed with
MtNFP-sYFP2 or MtLYK3-sYFP2 in Nicotiana leaf via Agro TT, and fluorescence (viewed from abaxial
side) was imaged 24hai and 48hai using confocal laser scanning microscopy: From left to right: green
fluorescence of sYFP2; orange fluorescence of mCherry; superimposition of green, orange, and far red
(chlorophyll) fluorescence with the differential interference contrast (DIC) image. Upper and upper middle
panel: note the pronounced localization of MtNFP-sYFP2 protein fusion in the ER that is continuous with
the nuclear envelope (indicated with the arrowhead) 24hai. Lower middle and lower panels: note
the co-localization of the PM marker with MtNFP-sYFP2 or MtLYK3-sYFP2 protein fusions at the cell
boundary. Bars are 20 µm.
Therefore, the CD response upon MtNFP and MtLYK3 co-expression was not a general
response to a heterologous (co-)expression of (a) RLK-encoding gene(s).
50
1
2
34
5
6
7
1
2
34
5
6
7
NFP+LYK3
NFP
LYK3
DMI2
DMI2+NFP
DMI2+LYK3
Figure 2. Co-expression of MtNFP and MtLYK3 but not the expression of the separate constructs
induces cell death in N. benthamiana or N. tabacum leaves.
A, CD induction assay in N. benthamiana. Empty vector or MtNFP and MtLYK3, either untagged or fused
to sYFP2, were expressed alone or co-expressed in Nicotiana leaves via Agro TT, and the infiltrated regions
were marked: mock (1); MtNFP untagged + MtLYK3 untagged (2); MtNFP-sYFP2+MtLYK3-sYFP2 (3);
MtLYK3-sYFP2 (4); MtLYK3 untagged (5); MtNFP-sYFP2 (6); MtNFP untagged (7). Left panel presents the
macroscopic symptoms of CD observed 48hai. Right panel presents the same leaf stained with Evans blue.
B, CD induction assay in N. tabacum. MtNFP or MtLYK3 (untagged) constructs were expressed alone
(see white arrowheads) or co-expressed (see the red arrowhead) in N. tabacum leaves via Agro TT,
and the infiltrated regions were marked. Macroscopic symptoms of CD are presented 48hai.
C, CD induction assay in N. benthamiana. MtDMI2-sYFP2 construct was expressed alone or co-expressed
with untagged MtNFP or MtLYK3 in Nicotiana leaves via Agro TT, and the infiltrated regions were marked.
Left panels show leaf regions expressing indicated construct(s) 48hai. Right panels present the same leaf
regions stained with Evans blue. All depicted regions come from the same leaf.
A
B C
51
Table 1. Cell death induction upon (co-)expression of various RLK-encoding genes in N. benthamiana
leaves.
Construct
Cell death induction
Separate
expression
Co-expression with
MtNFP-FP #
Co-expression with
MtLYK3-FP #
MtNFP–sYFP2 0/20 Not applicable 48/52
MtNFP –3xFLAG 0/9 Not applicable 12/13 (with MtLYK3-
3xFLAG)
MtNFP 0/9 Not applicable 8/9 (with MtLYK3)
MtLYK3–sYFP2 0/20 48/52 Not applicable
MtLYK3–3xFLAG 0/9 12/13
(with MtNFP-
3xFLAG)
Not applicable
MtLYK3 0/9 8/9
(with MtNFP- YFPN) Not applicable
MtDMI2-sYFP2 0/12 0/12 0/12
MtLRRII.1-YFPC 0/9 0/9
(with MtNFP- YFPN)
0/9
(with MtLYK3-
YFPN) AtBRI1- YFPC 0/9
0/9
(with MtNFP- YFPN)
0/9
(with MtLYK3-
YFPN) AtCERK1–sYFP2 21/22 19/20 12/14
AtCERK1 [K350E] –sYFP2 0/10 NT NT
MtLYK3–sYFP2 0/20 19/20 Not applicable
MtLYK2–sYFP2 0/9 13/20 NT
# - unless stated differently; NT - not tested.
Indicated constructs were expressed alone or co-expressed with either MtNFP or MtLYK3 in Nicotiana
leaves via Agro TT, and the infiltrated regions were marked. Macroscopic symptoms of CD were scored
48hai: only infiltrations that resulted in confluent death of (nearly) the entire infiltrated region were scored
as a fraction of total independent infiltrations performed. In case of no macroscopic symptoms, three leaves
were stained with Evans blue to confirm the lack of CD (data not shown). No macroscopic symptoms of CD
were observed 72hai in the remaining infiltrated regions (data not shown), after which point a weak
unspecific chlorosis could be observed.
52
Cell death induced in N. benthamiana leaf upon MtNFP and MtLYK3
co-production is a NF-independent response
In Medicago, MtNFP and MtLYK3 are postulated to co-function
in the perception of NF produced by S. meliloti, ultimately leading to symbiosis.
In contrast, their co-expression in Nicotiana leaf apparently triggered
a CD response in the absence of NF. To investigate the effect of S. meliloti NF
on the CD response, we co-infiltrated Agrobacterium transformants carrying either
MtNFP-sYFP2 or MtLYK3-mCherry construct at varying concentrations (as measured
with OD600). Then, purified NF at 10-7
M concentration or DMSO diluted to the same
concentration was applied between 9 and 24hai to part of the region previously
infiltrated with Agrobacterium. Subsequently, CD development was monitored
between 24 and 72hai using Evans blue staining. For all bacterial concentrations
and time-points of NF/DMSO application tested, part of leaf regions co-expressing
MtNFP and MtLYK3 and treated with the NF showed compromised membrane
permeability at similar time as the untreated leaf regions or regions treated with DMSO
(Fig. 3). Therefore, we did not obtain evidence for any stimulatory or inhibitory effect
of NF on the CD development in response to MtNFP and MtLYK3 co-expression.
Intracellular region of MtNFP and MtLYK3 kinase activity are required for
cell death induction in N. benthamiana leaf
The serendipitous observation of CD induction in Nicotiana leaf upon MtNFP
and MtLYK3 co-expression, and its apparent independence from the NF perception
prompted us to analyze whether this response requires the same structural features
of MtNFP and/or MtLYK3 as for symbiotic signaling. The MtNFP mutated variant
encoded by the Mtnfp-2 allele possesses the S67F substitution in the first LysM domain
53
1
1
2
2
3
1
1
2
2
3
Figure 3. Cell death induction in N. benthamiana leaf upon MtNFP and MtLYK3 co-expression
does not require, nor is modulated by S. meliloti NF.
Effect of purified NF on the kinetics of CD development in response to MtNFP-3xFLAG and MtLYK3-
3xFLAG co-expression. Agrobacterium transformants carrying the individual constructs were mixed together
and infiltrated in Nicotiana leaves at a range of optical densities: final OD600 [MtNFP]=0.25
and [MtLYK3]=0.4 (1); final OD600 [MtNFP]=0.15 and [MtLYK3]=0.25 (2). 12 hai parts of the transformed
regions were syringe-infiltrated with 10-7mM NF (circled in red) or DMSO diluted
to the same concentration (circled in white). Macroscopic observation (left panel) and Evans blue staining
(right panel) are depicted 36hai. Mock infiltrated region (3) was also treated with 10-7mM NF as control.
that results in retention of the mutated protein in the endoplasmic reticulum (Lefebvre
et al., 2012). However, as the PM localization of MtNFP is postulated to be required
for its biological activity (Lefebvre et al., 2012), this mutant variant was not included
in our analyses in Nicotiana. Instead, we focused on MtNFP truncated variant
with almost the entire InR deleted, termed MtNFP [∆InR] (amino acids: 1-283),
as it is not able to rescue the nodulation phenotype of the Mtnfp-1 mutant (Lefebvre
et al., 2012). In regard to MtLYK3, we chose to test two mutated variants: one with
the G334E substitution in the KD (encoded by the Mtlyk3-1 allele), and the second one
with the P87S substitution in the first LysM-domain (encoded by the Mtlyk3-3 allele)
(Smit et al., 2007). The former mutation has been shown to abolish MtLYK3 kinase
54
activity (Klaus-Heisen et al., 2011), whereas the exact effect of the P87S mutation
on MtLYK3 biological activity remains to be revealed.
All three constructs were prepared as C-terminal fusions to sYFP2,
and we confirmed the efficient production and correct PM localization of the encoded
fusions in Nicotiana leaf epidermal cells (Fig. 4 and Klaus-Heisen et al., 2011;
Lefebvre et al., 2012). The co-expression of MtNFP [∆InR]-sYFP2 with MtLYK3-
mCherry or of MtLYK3 [G334E]-sYFP2 with MtNFP-mCherry did not result in CD
induction, as no macroscopic symptoms of tissue collapse and desiccation or increased
staining with Evans blue was observed 48hai (Table 2). To rule out a possibility that
the presence of wild-type (WT) MtNFP or MtLYK3 receptor might affect stability/
accumulation of the mutated protein fusions, we confirmed efficient production
and PM localization of MtNFP and MtLYK3 truncated/mutated variants in Nicotiana
leaf epidermal cells also in the presence of WT full-length MtLYK3-mCherry
and MtNFP-mCherry. Both MtNFP [∆InR]-sYFP2 and MtLYK3 [G334E]-sYFP2 were
efficiently produced in the presence of, respectively, MtLYK3-mCherry or MtNFP-
mCherry (data not shown). In contrast, co-expression of MtLYK3 [P87S]–sYFP2 with
MtNFP-mCherry induced confluent CD in 15 out of 15 infiltrated regions (Table 2).
Taken together, the structural alterations within the InRs of these LysM-RLKs, but not
within the MtLYK3 ExR, appeared to have an identical effect on their biological
activity in both plant systems.
Expression of AtCERK1 induces cell death in N. benthamiana leaf independently
from MtNFP co-expression
A rapid tissue collapse at the site of pathogen attack, termed hypersensitive
response (HR), is observed in various incompatible plant-pathogen interactions where
55
HVR-mCherry
NFP [∆InR ]-
sYFP2
LYK3 [P87S]-
sYFP2
LYK3 [G334E]-
sYFP2
LYK2-sYFP2
CERK1[K350E]
-sYFP2
channel: YFP mCherry merged & DIC
Figure 4. Subcellular localization of various protein fusions in N. benthamiana leaf epidermal cells.
HVR-mCherry was co-expressed with MtNFP [ΔInR]-sYFP2, MtLYK3 [P87S]-sYFP2, MtLYK3 [G334E]-
sYFP2, MtLYK2-sYFP2 or AtCERK1 [K350E]-sYFP2 in Nicotiana leaf epidermal cells via Agro TT,
and the fluorescence (viewed from abaxial side) was imaged 24hai using confocal laser scanning
microscopy. From left to right: green fluorescence of sYFP2; orange fluorescence of mCherry;
superimposition of green, orange, and far red (chlorophyll) fluorescence with the differential interference
contrast (DIC) image. Bars are 20 µm.
it is thought to contribute to pathogen restriction and to generate a signal that activates
plant defense mechanisms (Heath 2000; Mur et al., 2008). The apparent phenotypic
56
Table 2. Cell death induction in N. benthamiana leaf upon (co-)production of MtNFP and MtLYK3
mutated and truncated variants.
Construct
Cell death induction
Separate
expressio
n
Co-expression with
MtNFP-mCherry
Co-expression with
MtLYK3-mCherry
MtNFP [∆InR]-sYFP2 0/9 Not applicable
0/20
MtLYK3 [P87S]-sYFP2 0/9
15/15
Not applicable
MtLYK3 [G334E]-sYFP2 0/9
0/20
Not applicable
Indicated constructs were expressed alone or co-expressed with either full length MtNFP-mCherry
or MtLYK3-mCherry in Nicotiana leaves via Agro TT, and the infiltrated regions were marked. Macroscopic
symptoms of CD were scored 48hai: only infiltrations that resulted in confluent death of (nearly) the entire
infiltrated region were scored as a fraction of total independent infiltrations performed. In case of
no macroscopic symptoms, three leaves were stained with Evans blue to confirm the lack of CD.
No macroscopic symptoms of CD were observed 72hai in the remaining infiltrated regions (data not shown),
after which point a weak unspecific chlorosis could be observed. The insets present macroscopic observation
(left image) and subsequent Evans blue staining (right image) of leaf regions co-expressing the designated
constructs 48hai.
similarity of Nicotiana response to MtNFP and MtLYK3 co-expression with the HR
elicited by, e.g. Phytophthora infestans elicitin, INF1 (Huitema et al., 2005),prompted
us to investigate whether it might result from triggering of stress/defense signaling.
AtCERK1-mediated signaling participates in Arabidopsis innate immunity to fungal
and bacterial pathogens, although, to our knowledge, CD induction
57
in response to chitin/COs or PGN has not been reported so far. Hence, we investigated
the Nicotiana response to the production of WT AtCERK1 or AtCERK1 mutated
variant whose kinase activity was abolished by substitution of the conserved Lys 350
(Petutschnig et al., 2010). Again, in order to ensure efficient production of the encoded
proteins in Nicotiana leaf, both WT AtCERK1 and AtCERK1 [K350E] constructs
were generated as C-terminal fusions to sYFP2, and their expression was driven by
the CaMV 35S promoter.
Surprisingly, tissue collapse and desiccation in the entire infiltrated region
expressing AtCERK1-sYFP2 construct was observed 36hai in 20 out of 22 infiltrated
regions (Fig. 5, Table 1). This CD induction abolished our attempts of precisely
characterizing the subcellular localization of AtCERK1-sYFP2 protein fusion
in Nicotiana leaf epidermal cells, although we could detect sYFP2 fluorescence
at the cell boundary (data not shown). On the contrary, production of the kinase-
inactive AtCERK1 [K350E]-sYFP2 protein fusion did not result in CD induction,
as confirmed with Evans blue staining (Fig. 5, Table 1). Co-localization of AtCERK1
[K350E]-sYFP2 with the PM marker using confocal laser scanning microscopy
analysis indicated its PM localization in Nicotiana leaf epidermal cells (Fig. 4),
in agreement with the findings of Miya et al. (2007) in onion epidermal cells.
Encouraged by this apparent similarity of Nicotiana response to separate expression
of AtCERK1, and co-expression of MtNFP and MtLYK3, we tested a hypothesis
that simultaneous accumulation of the symbiotic LysM-RLKs in Nicotiana leaf results
in stress/defense signaling and defense(-like) response.
58
Cell death induction upon (co-)production of MtNFP and MtLYK3 or AtCERK1
in N. benthamiana leaf requires an influx of the extracellular Ca2+
An influx of extracellular Ca2+
causes a rapid increase in the cytosolic [Ca2+
]
that is required for the activation of a MAPK cascade, ROS production, and PAMP-
induced gene expression. Therefore, it is postulated to be located very early in the plant
stress/defense signaling pathway (Ranf et al., 2011; Segonzac et al., 2011), possibly
immediately downstream from PM receptors for PAMPs (Jeworutzki et al., 2010;
Frei dit Frey et al., 2012). We wanted to know whether an influx of extracellular Ca2+
might be similarly implicated in CD induction upon MtNFP and MtLYK3
co-expression or AtCERK1 expression. Agrobacterium transformants carrying MtNFP-
3xFLAG or MtLYK3-3xFLAG constructs were co-infiltrated in Nicotiana leaves
concomitantly with the infiltration of Agrobacterium transformants carrying AtCERK1-
3xFLAG. Twelve hours later, parts of the infiltrated regions were treated with 5mM
lanthanum chloride (Merck), an established inhibitor of the PM calcium channels,
or water, and the CD development was monitored between 24 and 72hai. In case of
MtNFP-3xFLAG and MtLYK3-3xFLAG co-expression, 42hai tissue collapse
and compromised membrane permeability (as visualizec with Evans blue staining)
was observed only/mostly outside the lanthanum chloride-treated infiltrated regions
in 24 out of 30 infiltrations (Fig. 6). On the contrary, control treatment with water
did not affect the confluent CD development in 18 out of 19 infiltrations. Notably,
60hai 26 out of 30 infiltrated regions treated with lanthanum chloride still developed
confluent death of the entire infiltrated region (data not shown). A similar delay
of the CD development was observed in parts of the leaf regions expressing AtCERK1-
3xFLAG that were treated with 5mM lanthanum chloride but not in those treated with
water (data not shown). Therefore, lanthanum chloride apparently delayed the CD
59
CERK1-sYFP2
CERK1[K350E]-
sYFP2
Figure 5. Heterologous production of AtCERK1 in N. benthamiana leaf induces cell death
that is dependent on AtCERK1 kinase activity.
CD induction assay in Nicotiana. Macroscopic observation (left panel) and subsequent Evans blue staining
(right panel) of the leaf region expressing AtCERK1-sYFP2 or AtCERK1 [K350]-sYFP2 (36hai).
1
2
3
3
4
1
2
3
3
4
Figure 6. Effect of lanthanum chloride on the kinetics of CD development induced by MtNFP
and MtLYK3 co-expression.
Agrobacterium transformants carrying MtNFP-3xFLAG or MtLYK3-3xFLAG construct were co-infiltrated
at a range of bacterial concentration: final OD600 [MtNFP]=0.25 and final OD600 [MtLYK3]=0.4 (1);
final OD600 [MtNFP]=0.19 and final OD600 [MtLYK3]=0.3 (2); final OD600 [MtNFP]=0.125 and final OD600
[MtLYK3]=0.2 (3). 12 hai parts of the infiltrated regions were syringe-infiltrated with 5mM lanthanum
chloride (circled in red) or water (circled in white). Macroscopic symptoms of CD (left panel) and Evans
blue staining (right panel) are depicted 42hai. Mock infiltrated region (4) was also treated with 5mM
lanthanum chloride as control.
60
development in response to MtNFP and MtLYK3 co-expression or AtCERK1
expression. Next, we wanted to know whether the observed CD induction
was associated with typical plant stress/defense-related processes.
MtNFP and MtLYK3-induced cell death in N. benthamiana leaf is associated
with induction of plant stress/defense-related responses
We focused on the accumulation of phenolic compounds and induction
of stress/defense-related gene expression, which are hallmarks of defense response
induction in various plant species, including Nicotiana spp. As a positive control,
we used a transient expression of INF1 that was reported to induce CD pathogenesis-
related (PR) gene expression and CD in Nicotiana spp. (Kamoun et al., 1997, 1998;
Huitema et al., 2005). In addition, several Nicotiana responses to COs stimulation,
presumably mediated by the AtCERK1 ortholog in Nicotiana, NbCERK1, have recently
been characterized (Gimenez-Ibanez et al., 2009; Segonzac et al., 2011). Therefore,
AtCERK1 expression was used as an additional control for CD induction and activation
of stress/defense-related gene expression. As the accumulation of phenolic compounds
was investigated through monitoring changes in the leaf tissue autofluorescence,
we generated MtNFP, MtLYK3, MtLYK3 [G334E] and AtCERK1 constructs fused
C-terminally to the sequence encoding 3xFLAG epitope tag.
We started by analysing the kinetics of CD development. To this end,
Agrobacterium transformants carrying MtNFP-3xFLAG or MtLYK3-3xFLAG
constructs were co-infiltrated at different time-points in adjacent circles in Nicotiana
leaves, and CD development was monitored between 24 and 48hai. Alternatively,
co-infiltration of Agrobacterium transformants carrying MtNFP-3xFLAG or MtLYK3-
3xFLAG constructs was done concomitantly with the infiltration of Agrobacterium
61
transformants carrying INF1 or AtCERK1-3xFLAG construct. In case of MtNFP-
3xFLAG and MtLYK3-3xFLAG co-expression, macroscopic symptoms of CD were first
observed around 33hai as a type of flaccidity that preceded tissue collapse and often
occurred over the entire infiltrated region approximately 36hai (Fig. 7A region 3)
and the appearance of small patches of collapsed tissue (these were more pronounced
on the abaxial side of the leaf). Approximately 42hai the collapsed zone encompassed
nearly the entire infiltrated region in 25 out of 31 infiltrations (Fig. 7A region 1),
and at 48hai 30 out of 31 infiltrated regions showed pronounced tissue desiccation
in the entire infiltrated region (Fig. 2A). Compromised membrane permeability,
as visualized by Evans blue staining, Evans blue staining of leaf regions expressing
AtCERK1-sYFP2 was observed already approximately 24hai (Fig. 7A region 1B),
in agreement with the faster development of macroscopic symptoms of CD (Fig. 4A).
In addition, MtNFP-3xFLAG and MtLYK3-3xFLAG co-expression resulted in a marked
decrease of (far-red) chlorophyll fluorescence (Fig. 7C left panel) and accumulation
of blue light-excited autofluorescence (Fig. 7C right panel) approximately 36hai.
This was not observed after separate expression of MtNFP-3xFLAG or MtLYK3-
3xFLAG, or after co-expression of MtNFP-3xFLAG and MtLYK3 [G334E]-3xFLAG
(data not shown). Ethanol/lactophenol-inextractable and UV-excited autofluorescence,
indicative of phenolic compounds, was detected approximately 36hai and 30hai in leaf
regions co-expressing MtNFP-3xFLAG and MtLYK3-3xFLAG or expressing AtCERK1-
3xFLAG, respectively (Fig. 7D). Similar UV-excited autofluorescence was
not observed with mock infiltration, upon separate expression of MtNFP-3xFLAG
or MtLYK3-3xFLAG, or upon co-expression of MtNFP-3xFLAG and MtLYK3
[G334E]-3xFLAG (Fig. 7D).
62
63
Figure 7. MtNFP and MtLYK3-induced cell death in N. benthamiana leaves is associated
with defense-like response.
A, Kinetics of CD development in Nicotiana. Agrobacterium transfomants carrying MtNFP-3xFLAG
or MtLYK3-3xFLAG constructs were mixed 1:1 and infiltrated into Nicotiana leaves at five different time
points (1-5). Macroscopic observations (left panel) and subsequent Evans blue staining (right panel)
were performed 42hai (region 1), 39hai (region 2), 36hai (region 3), 33hai (region 4) and 30hai (region 5).
Mock infiltration (region 6) was done in the same leaf concomitantly with the first infiltration
of Agrobacterium transformants carrying MtNFP-3xFLAG or MtLYK3-3xFLAG constructs.
B, Kinetics of CD development in Nicotiana. Agrobacterium transfomants carrying MtNFP-3xFLAG
or MtLYK3-3xFLAG constructs were mixed 1:1 and infiltrated into Nicotiana leaves concomitantly with the
mock infiltration (not shown) and infiltration of Agrobacterium transformants carrying AtCERK1-3xFLAG
construct. Macroscopic observations (left panel) and subsequent Evans blue staining (right panel)
were performed between 24 and 36hai (here depicted 30hai).
C, Changes in leaf autofluorescence upon MtNFP and MtLYK3 co-expression. Leaf regions co-expressing
MtNFP-3xFLAG and MtLYK3-3xFLAG were analyzed between 24 and 48hai (here depicted 36hai) using
a stereoscope. Note the decrease in chlorophyll content, as indicated by the decrease of far-red
autofluorescence of chlorophyll (left panel), and enhanced accumulation of blue light-excited
autofluorescence (right panel) within the infiltrated region.
D, Accumulation of phenolic compounds. Agrobacterium transformants carrying the indicated construct
were (co-)infiltrated in Nicotiana leaves, and the infiltrated regions were marked: MtNFP-3xFLAG (1);
MtLYK3-3xFLAG (2); MtNFP-3xFLAG + MtLYK3-3xFLAG (3); MtNFP-3xFLAG + MtLYK3 [G334E]-
3xFLAG (4); AtCERK1-3xFLAG (5); AtCERK1 [K350]-3xFLAG (6); INF1 (7). Macroscopic observations
(left panel) and subsequent UV-excited autofluorescence of ethanol/lactophenol-cleared leaf (right panel)
depicted 36 hai.
E, induction of NbHIN1, NbPR-1 basic, NbACRE31, and NbACRE132 expression in response
to (co)-expression of LysM-RLK-encoding genes or INF1 in Nicotiana leaves. MtNFP-3xFLAG (NFP),
MtLYK3-3xFLAG (LYK3), MtLYK3 [G334E]-3xFLAG (LYK3 [G334E]), AtCERK1-3xFLAG (CERK1)
or INF1 were expressed alone or co-expressed in Nicotiana leaves using Agro TT. Leaf samples
were collected 24hai and induction of gene expression was analysed using qRT-PCR. Histograms represent
induction of NbHIN1 (white columns), NbPR-1 basic (grey columns), NbACRE31 (hatched columns),
and NbACRE132 (black columns) normalized by one reference gene, MtEF1 . Induction of each gene
was normalized to the response to mock infiltration, and then calculated as % induction relative
to the induction observed upon MtNFP-3xFLAG and MtLYK3-3xFLAG co-production. Bars represent
standard deviations. At least two technical replicates from two biological replicates were analyzed.
Subsequently, we investigated the presumed induction of stress/defense-
related genes in Nicotiana leaf in response to: MtNFP-3xFLAG and MtLYK3-3xFLAG
co-expression, and AtCERK1-3xFLAG or INF1 expression. In addition, we investigated
the induction of the same stress/defense-related genes in mock-infiltrated leaves, leaves
64
expressing MtNFP-3xFLAG, MtLYK3-3xFLAG or MtLYK3 [G334E]-3xFLAG,
and leaves co-expressing MtNFP-3xFLAG and MtLYK3 [G334E]-3xFLAG constructs.
Induction of gene expression was analyzed 24hai (i.e. before the occurrence
of macroscopic symptoms of CD) using quantitative reverse-transcriptase polymerase
chain reaction (qRT-PCR). We focused on the expression of: NbHIN1 – the postulated
marker gene for HR (Gopalan et al., 1996; Taguchi et al., 2009); two PR-1 genes,
i.e. NbPR-1 acidic and NbPR-1 basic (Cornelissen 1987); and NbACRE31,
NbACRE132, and NbCYP71D20 – the postulated marker genes for PAMP-triggered
immunity (Segonzac et al., 2011 and refs therein). Co-production of MtNFP-3xFLAG
and MtLYK3-3xFLAG, and production of AtCERK1-3xFLAG resulted in induction
of NbHIN1, NbPR-1 basic, NbACRE31, and NbACRE132 gene expression that was
significantly higher than that following co-production of MtLYK3 [G334E]-3xFLAG
and MtNFP-3xFLAG (Fig. 7E). Induction of the above four genes upon separate
production of MtNFP-3xFLAG, MtLYK3-3xFLAG or MtLYK3 [G334E]-3xFLAG
was higher than that observed in mock-infiltrated leaves but was significantly lower
than that resulting from MtNFP-3xFLAG and MtLYK3-3xFLAG co-production
(Fig. 7E). Production of INF1 resulted in strong induction of NbPR-1 gene,
and somewhat weaker induction of NbHIN1, NbACRE31 and NbACRE132 genes
(Fig. 7E). The PR-1 acidic and NbCYP71D20 genes did not display significant
induction upon (co-)expression of any of the constructs tested (data not shown).
Taken together, the observed localized accumulation of phenolic compounds
and induction of stress/defense-related gene expression suggested that MtNFP
and MtLYK3 co-production triggered a defense(-like) response that was similar to that
resulting from the transient expression of AtCERK1 and INF1.
65
Co-expression of MtLYK2 and MtNFP in N. benthamiana leaf induces cell death
The family of Medicago LysM-RLK genes includes 17 members, among
which one (MtLYK5) is predicted to be a pseudogene (Limpens et al., 2003), and eight
(including MtLYK3) are predicted to encode RLKs possessing putatively active KDs
(Arrighi et al., 2006). MtLYK2 is phylogenetically most closely related to MtLYK3
(Zhang et al., 2009b and refs therein), resulting in 81% overall amino acid sequence
identity of their encoded proteins. Therefore, we decided to investigate whether
MtLYK2 was able to induce CD in Nicotiana in the presence of MtNFP. Expression
of MtLYK2-sYFP2 construct driven by the CaMV 35S promoter resulted in efficient
production and clear PM localization of the encoded protein fusion in Nicotiana leaf
epidermal cells (Fig. S1). Remarkably, co-expression of MtLYK2-sYFP2 and MtNFP-
mCherry resulted in collapse and desiccation of the entire infiltrated region in 13 out of
20 infiltrations (Table 1), whereas control co-expression of MtLYK3-sYFP2
and MtNFP-mCherry in the same leaves led to death of the entire infiltrated region
in 19 out of 20 infiltrated regions (Table 1). Therefore, MtLYK2 was capable
of CD induction in Nicotiana leaf upon co-production with MtNFP.
DISCUSSION
Co-expression of MtNFP and MtLYK3 induces a defense(-like) response
in N. benthamiana leaf
The use of Nicotiana allowed efficient production of both MtNFP
and MtLYK3 protein fusions in leaf tissue (Fig. 1), and led to the surprising
observation of CD induction upon their simultaneous accumulation (Fig. 2A and 6A).
The observed tissue collapse and desiccation resembled macroscopic symptoms
66
of the HR elicited in Nicotiana spp. by various pathogen-derived molecules
(e.g. Huitema et al., 2005; Gabriels et al., 2006), and the kinetics of CD development
(Fig. 7AB) agreed with the reported timing of HR triggered by transient
(co-)expression of certain defense-related plant genes (e.g. Mucyn et al., 2006 and refs
therein). Moreover, our results agree with the recently demonstrated CD induction
upon transient co-expression of LjNFR1 and LjNFR5 in Nicotiana leaf (Madsen et al.,
2011), although, in that case, putative induction of associated defense(-like) responses
has not been studied. We showed that MtNFP and MtLYK3 co-expression stimulated
local accumulation of phenolic compounds (Fig. 7CD), and significant induction
of 4 out of 6 tested stress/defense-related genes (Fig. 7E). We speculate that the lack
of induction of two other tested genes might resemble a natural variation
in the repertoire of stress/defense-related genes induced by various PAMPs (Navarro
et al., 2004; Zipfel et al., 2006; Wan et al., 2008) or might be caused by different
kinetics of gene induction (here analyzed only 24hai). Importantly, induction of PR-1,
ACRE31 (At4g20780), ACRE132 (At3g16720) and a member of a HIN1 gene family
in Arabidopsis and Nicotiana was shown to result from COs and/or PGN perception
(Gust et al., 2007; Segonzac et al., 2011) in AtCERK1-dependent manner (Wan et al.,
2008). Taken together, our results suggest that the heterologous co-producion of two
symbiotic LysM-RLKs in Nicotiana leaf activates stress/defense signaling that results
in a defense(-like) response. Interestingly, we showed that treatment with lanthanum
chloride delayed the CD development upon MtNFP and MtLYK3 co-production
(Fig. 6). Therefore, we speculate that simultaneous accumulation of these RLKs might
activate (a) putative component(s) located early (i.e. at or preceding the extracellular
Ca2+
influx step) in the Nicotiana stress/defense signaling pathway.
67
Heterologous expression of AtCERK1 induces a defense(-like) response
in N. benthamiana leaf
Interestingly, Nicotiana response to MtNFP and MtLYK3 co-expression
resembled phenotypically the outcome of the heterologous expression of AtCERK1,
implicated in defense signaling in Arabidopsis (Miya et al., 2007; Wan et al., 2008;
Gimenez-Ibanez et al., 2009; Petutschnig et al., 2010; Willmann et al., 2011; Liu et al.,
2012b). Accumulation of phenolic compounds, and induction of stress/defense-related
gene expression, but not CD induction, has been demonstrated in several plant species
in response to COs or PGN (Gust et al., 2007; Erbs et al., 2008; Hamel & Beaudoin
2010). However, misregulation of various stress/defense-related components
(e.g. Mucyn et al., 2006 and refs therein; Gao et al., 2009 and refs therein), including
rice MAPK kinase (OsMKK4) implicated in chitin/COs signaling (Kishi-Kaboshi
et al., 2010), has been reported to result in CD induction. Analogously, we hypothesize
that the heterologous production of AtCERK1 might result in the misregulation
of its kinase activity, and in turn – in CD induction. Importantly, the K350E mutation
similarly abolished AtCERK1 biological activity in Arabidopsis (Petutschnig et al.,
2010) and in Nicotiana (Fig. 4A), suggesting that AtCERK1 kinase activity
was similarly required for AtCERK1 signaling in both plant systems. In addition,
it is formally possible that AtCERK1 is activated in Nicotiana leaf by a putative signal
of Agrobacterium origins. Therefore, the outcome of the heterologous production
of AtCERK1 might be linked directly to the specific biological activity of this RLK,
rather than to a putative unspecific result of AtCERK1 accumulation. Interestingly,
MtLYK2/MtLYK3 (Fig. 2A, Table 1) but not the homologous AtCERK1 (Fig. 4A,
Table 1) required the presence of MtNFP for CD induction, although the involvement
of Nicotiana endogenous extracellular proteins (e.g. homologous to AtLYM1/
68
AtLYM3) in the AtCERK1-mediated CD induction remains to be investigated.
The apparent redundant function of MtLYK2 and MtLYK3 agrees with
the hypothesized functionalization of the putative NF receptors during
the co-evolution of legumes with rhizobia (Streng et al., 2011).
Similarities and differences between symbiotic and stress/defense signaling
The similar outcome of MtNFP and MtLYK3 co-expression and AtCERK1
and INF1 separate expression in Nicotiana leaf suggest a possible overlap between,
respectively, symbiotic and stress/defense signaling mediated by the encoded receptors.
Corroborating this, ROS production (Cárdenas et al., 2008), increase in cytosolic
[Ca2+
] (Cárdenas 2000; Shaw & Long 2003), and Ca2+
spiking (Oldroyd & Downie
2006; Sieberer et al., 2009; Krebs et al., 2012) induced by compatible NFs resemble
signaling events in response to various PAMPs, including chitin/COs and PGN
(Garcia-Brugger et al., 2006; Nicaise et al., 2009; Boudsocq et al., 2010; Ranf et al.,
2011; Segonzac et al., 2011; Manzoor et al., 2012). We speculate that similar
molecular processes can be regulated by homologous molecular components, thereby
allowing two Medicago LysM-RLKs to activate putative signaling components present
in Nicotiana leaf tissue. Remarkably, Nakagawa et al. (2010) demonstrated that
specific sequence alterations within the AtCERK1 InR allowed it to replace
the LjNFR1 InR during nodulation, indicating its competence for symbiotic signaling.
Conversely, our results demonstrate that symbiotic LysM-RLKs, when heterologously
produced in Nicotiana, are sufficient to trigger defense(-like) response. We speculate
that due to the absence of symbiosis-specific “decoders” or “modulators”, the signaling
induced in response to simultaneous accumulation of MtNFP and MtLYK3
69
in Nicotiana leaf might be differently interpreted in this heterologous system, resulting
in an induction of a defense(-like) response.
Curiously, MtNFP and MtLYK3 biological activity in Nicotiana seemed
to be independent from the NF perception (Fig. 3). However, at present we cannot
exclude a possibility that a ligand of Agrobacterium origin is responsible for (in)direct
activation of these RLKs, in agreement with the presumed activation of CERK1-
mediated signaling upon binding of bacterial PGN to LysM domain-containing
proteins, AtLYM1 and AtLYM3 or OsLYP4 and OsLYP6 (Willmann et al., 2011; Liu
et al., 2012a). Similarly, the P87S mutation abolished MtLYK3 biological activity
in Medicago (Smit et al., 2007) but not in Nicotiana (Table 2). However, further
mapping of crucial amino acid residues, and detailed characterization of their exact
role in NF signaling would be required to clarify whether or not nodulation and CD
induction indeed hold different requirements with respect to the MtNFP
and/or MtLYK3 ExRs. On the contrary, the ability of MtNFP and MtLYK3 to induce
CD in Nicotiana leaf was dependent on the presence of MtNFP InR and (putatively)
MtLYK3 kinase activity (Fig. 7DE, Table 2), thereby mimicking the requirements
for nodulation (Smit et al., 2007; Klaus-Heisen et al., 2011; Lefebvre et al., 2012;
and in agreement with the requirements reported for MtNFP orthologs in Lotus [Ljnfr5-
4] and pea [Pisum sativum] [P56 and RisFixG]; Madsen et al., 2003; Murray et al.,
2006). Further work will be required to characterize in more detail the hypothesis
of common requirements with respect to the MtNFP, MtLYK3, and MtLYK2 structure
for their biological activity in both Medicago and Nicotiana, as indicated by the results
presented in this report. If so, the Nicotiana leaf might present a relevant system
to analyze LysM-RLK function in early symbiotic signaling. This system presents
certain practical advantages over the legume root system, in terms of rapidity and ease
70
of expression of multiple constructs. Possible candidate signaling molecule(s)
functioning in co-operation with or downstream from the putative NF receptors,
if identified in such studies, should then be evaluated in legume root system
to confirm their involvement in symbiosis.
Symbiotic LysM-RLKs might be involved in activation of weak, partial
or modulated defense(-like) response upon NF perception
NFs are structurally similar to COs, potent elicitors of PAMP-triggered
immunity in many plant species (Hamel & Beaudoin 2010). The perception
of structurally similar ligands by homologous LysM-RLKs suggests that symbiosis
and innate immunity, two diametrically different outcomes of plant-microbe
interactions, are evolutionarily related (Stacey et al., 2006; Hamel & Beaudoin 2010).
In this respect, it is interesting to note that NF signaling seems to have a rather
complex, partially contradictory effect on the host responses. Increase of endogenous
salicylic acid (SA) level and ROS production upon legume root inoculation
with various rhizobia nod mutants (defective in NF biosynthesis) indicate NF role
as a suppressor of initial host defense response (Martinez-Abarca et al., 1998; Blilou
et al., 1999; Bueno et al., 2001). This attenuation or modulation of the initial defense
response could be achieved via PLD activity, reported in soybean (Glycine max)
and Lotus roots, and alfalfa (M. sativa) suspension cell cultures specifically in response
to the compatible NF (den Hartog et al., 2003; Wan et al., 2005; Serna-Sanz et al.,
2011). Moreover, as overexpression and silencing of MtNPR1 (for Non-expressor
of Pathogenesis-related genes 1) resulted in, respectively, suppressed and enhanced
NF-induced RH curling in Medicago (Peleg-Grossman et al., 2009), differential
71
induction of the symbiotic or defense response could be regulated quantitatively
via the NPR1-mediated signaling pathway.
On the other hand, NF perception is required and sufficient for the induction
of (at least some) stress/defense-related genes and phosphorylation of stress/defense-
related proteins reported in the initial stage of various RL interactions (Lange et al.,
1999; El Yahyaoui et al., 2004; Mitra et al., 2004a; Høgslund et al., 2009; Nakagawa
et al., 2010; Serna-Sanz et al., 2011). Even more striking example of NF-induced
defense(-like) response comes from an aquatic model legume, Sesbania rostrata.
NF produced by its microsymbiont, Azorhizobium caulinodans, triggers the production
of hydrogen peroxide and ethylene that eventually lead to local CD allowing
the bacteria to proliferate in cortical infection pockets (Capoen et al., 2010).
Our results agree with the hypothesized stimulation of limited host defense responses
via the putative NF receptors. Future identification of the NF receptors in S. rostrata,
and investigation of the involvement of ROS and/or localized CD during the recently
reported crack entry of M. loti into Lotus roots (Karas et al., 2005; Yokota et al., 2009;
Madsen et al., 2010; Groth et al., 2010; Kosuta et al., 2011), would be invaluable
in deciphering the mechanism of ROS signaling activation upon NF perception.
Possible co-function of MtNFP and MtLYK3 during symbiotic signaling
and the presumed functional redundancy of LYK proteins
In order to elucidate the signaling mechanism employed by the kinase-inactive
MtNFP, this LysM-RLK has been proposed to form a putative receptor complex
with MtLYK3 or another LYK protein during, respectively, the IT growth and during
early symbiotic signaling (Arrighi et al., 2006; Smit et al., 2007; Bensmihen et al.,
2011). However, to our knowledge, evidence of a physical interaction between
72
the LysM-RLKs in a receptor complex has not been provided yet. We demonstrated
that only the simultaneous accumulation of MtNFP with either MtLYK3 or MtLYK2,
but not with other RLKs, resulted in CD induction in Nicotiana leaf (Fig. 2C, Table 1).
We propose that this Nicotiana response reports on the functional interaction between
those LysM-RLKs, and therefore supports the previously hypothesized co-functioning
of MtNFP with MtLYK2 (Smit et al., 2007) and MtLYK3 (Arrighi et al., 2006;
Bensmihen et al., 2011). In addition, our results provide evidence for the functional
redundancy of the MtLYK proteins in vivo. In order to confirm the possible
involvement of MtLYK2 during RL symbiosis, putative Mtlyk2 mutants would have
to be first identified in the MtLYK3 knock-out (i.e. Mtlyk3-2) or loss-of-function
(i.e. Mtlyk3-1 or Mtlyk3-3) background (Smit et al., 2007). Alternatively, knock-down
of the MtLYK2 expression in those Mtlyk3 mutants could indicate MtLYK2 role
in nodulation. Further work is required to find out whether the observed functional
interaction of MtNFP and MtLYK2 or MtLYK3 indeed requires their direct molecular
interaction. Alternatively, they might independently activate different molecular
components, thereby triggering parallel signaling pathways that later convergence
or otherwise constructively interfere to generate the observed CD and associated stress
responses.
73
EXPERIMENTAL PROCEDURES
Constructs for plant expression
The coding sequence of MtNFP, MtLYK3, and MtDMI2 (in the latter case the first intron was included
in the cDNA sequence) or the genomic sequence of AtCERK1 was PCR amplified and in the CaMV
35Sp::sYFP2, CaMV 35Sp::mCherry or CaMV 35Sp::3xFLAG pMon999 vector (sYFP2 or mCherry;
Kremers et al., 2006; Shaner et al., 2004) in a way that the stop codons were removed from the MtNFP,
MtLYK3, MtDMI2, and AtCERK1 coding sequences to allow the translational fusion. Sequences
of the primers and linkers are given in Table 3. Full length untagged MtNFP and MtLYK3 constructs were
generated by cloning their coding sequences, including the stop codons, into CaMV p35S::35S terminator
pMon999 vector. All point mutations were introduced using the QuickChangeTM site-directed mutagenesis
kit (Stratagene) as described (Klaus-Heisen et al., 2011). MtNFP [∆InR] truncated construct was generated
by PCR amplification (see Table 3 for primer sequences) and cloned into CaMV 35Sp::sYFP2 pMon999
vector. All constructs were sequenced to verify the correct insert sequence. Constructs generated
in pMON999 vector were subsequently recloned into a pBin+ (all MtNFP constructs) or pCambia1390
(www.cambia.org) binary vectors using HindIII/SmaI sites (ligation of double fragments in case of MtLYK2).
MtNFP-YFPN, MtLYK3-YFPN, MtLRRII.1-YFPc, AtBRI1-YFPc (where YFPN or YFPc encode, respectively,
the N- and C-terminal part of split YFP sequence used in BiFluorescence Complementation assay [BiFC]),
and INF1 expression vectors are described (Huitema et al., 2005; Lefebvre et al., 2010; Mbengue et al.,
2010).
Plant transformations
Agrobacterium tumefaciens GV3101::pMP90 and LBA4404 strains were transformed with the respective
constructs via electroporation. Agrobacterium tumefaciens LBA4404 strain was used only in the experiments
that compared the effect of different A. tumefaciens strains on MtNFP and MtLYK3-mediated CD induction.
All results presented in Figures 1 to 5 and Table 1 were obtained with A. tumefaciens GV3101::pMP90
strain. Agrobacterium-mediated transient transformation of Nicotiana spp. was performed essentially
as described (van Ooijen et al., 2008), except that Agrobacterium cultures were grown in LB medium
supplemented with 25μg/mL of rifampicin and 50 μg/mL of kanamycin. Resuspended cells were incubated
at room temperature for at least 1h before being infiltrated into intact, fully expanded leaves of green house-
grown Nicotiana benthamiana or Nicotiana tabacum cv. Samsun using needleless syringes. Agrobacterium
transformants carrying the respective construct were resuspended in the infiltration medium to desired OD600:
all MtNFP constructs (untagged or C-terminally fused to 3xFLAG or FP sequence) and MtNFP [∆IR]-sYFP2
construct - OD600=0.4; MtLYK2-sYFP2, all MtLYK3 constructs (untagged or C-terminally fused to 3xFLAG
or FP sequence), and all AtCERK1 constructs (WT or carrying the K350E mutation; C-terminally fused to
3xFLAG or sYFP2 sequence) - OD600=0.7; MtDMI2-sYFP2 - OD600=; INF1 - OD600=1. Subsequently,
they were mixed 1:1 with Agrobacterium transformants carrying: an empty pCambia1390 vector
(for “separate expression”), MtNFP construct (untagged or C-terminally fused to 3xFLAG or FP sequence)
or MtLYK3 construct (untagged or C-terminally fused to 3xFLAG or FP sequence) before being infiltrated
into N. benthamiana leaf. All experiments included mock control (GV3101::pMP90 transformants carrying
empty pCambia1390 vector) and a positive control (co-expression of full length MtNFP-FP and MtLYK3-FP
74
constructs). Cell death induction upon separate expression or co-expression of each (pair of) constructs
were analyzed in at least two independent experiments, every time using three different plants. Macroscopic
observations were carried out between 24 and 72hai, and the results obtained from at least two independent
experiments were collated.
To confirm efficient accumulation of MtLRRII.1-YFPC and AtBRI1-YFPC and protein fusions
in N. benthamiana, MtLRRII.1-YFPC and AtBRI1-YFPC were co-expressed with either MtNFP-YFPN
or MtLYK3-YFPN. Agrobacterium tumefaciens GV3101::pMP90 transformants carrying the respective
constructs were co-infiltrated at high optical densities (final OD600=0.5 each), and the observed
complementation of YFP fluorescence proved efficient accumulation, and even unspecific oligomerization
of the respective encoded fusions due to 2-dimensional surface concentration.
Detection of phenolic compounds
Blue light-excited autofluorescence and far-red chlorophyll autofluorescence in intact
N. benthamiana leaves were imaged using 430/40 excitation and 485/50 emission BP filters, or 480/40 BP
excitation and 510 LP emission filters, respectively. Images were captured using CMOS USB DCC1645C
camera (THORLabs, Newton NJ, USA) implemented on a Leica MZ FLIII stereoscope. Evans blue staining
was performed as described (van Ooijen et al., 2008). Clearing of leaves was achieved by boiling in acidic
lactophenol:ethanol solution (10g phenol, 10ml lactic acid mixed 2:1 with 96% ethanol) until the complete
removal of chlorophyll (approximately 3min per leaf) ethanol-inextractable autofluorescence was stimulated
with 312nm UV excitation. Images were captured using a Cool Snap CF camera (Photometrix, Tucson AZ,
USA).
qRT-PCR analysis
RNA extraction and qRT-PCR were performed as described in Mbengue et al. (2010) except that cDNA
were prepared from 500 µg of total RNA. See Table 3 for primer sequences. Two technical replicates
from at least two biological replicates were analyzed.
Microscopic analysis
Was carried as described in Klaus-Heisen et al. (2011).
75
Table 3. Primer and linker sequences.
Name Type Sequence
MtNFP fw Cloning (NheI) GGGGCTAGCATGTCTGCCTTCTTTCTTC
MtNFP (no stop) rev Cloning (EcoRI) GGGAATTCACGAGCTATTACAGAAGTAA
MtNFP (stop) rev Cloning (EcoRI) GGGAATTCTCAACGAGCTATTACAGAAGTAA
MtNFP∆InR rev Cloning (EcoRI) GGGAATCCTTCTATTCAATCTCTTCATTTTG
MtLYK3 fw Cloning (NheI) GGGGCTAGCATGAATCTCAAAAATGGATTAC
MtLYK3 (no stop) rev Cloning (EcoRI) GGGAATTCTCTAGTTGACAACAGATTTATG
MtLYK3 (stop) rev Cloning (EcoRI) GGGAATTCTCATCTAGTTGACAACAGATTTATG
MtLYK2 fw Cloning (NheI) GGGCTAGCATGAAACTAAAAAATGGTTTAC
MtLYK2 rev Cloning (EcoRI) CCGAATTCTCTCACTGACAAGAGATTTA
AtCERK1 fw Cloning (NheI) GGGCTAGCATGAAGCTAAAGATTTCTC
AtCERK1 (no stop) rev Cloning (EcoRI) GGGAATTCCCGGCCGGACATAAGAC
MtDMI2 fw Cloning (NheI) GGGGCTAGCATGATGGAGTTACAAGTTATT
MtDMI2 rev Cloning (KpnI) GGGGGTACCTCTCGGCTGTGGGTGAG
Linker to FP GAATTC for all the constructs, except for DMI2: GGTACC
Linker to 3xFlag tag
GAATTCCGGGCTGACTACAAAGACCATGACGGTGATTATAA
AGA
TCATGACATC
NbPR1 basic fw qRT- PCR GTTGCTTGTTTCATTACCTTTGC
NbPR1 basic rev qRT- PCR TTCTCATCGACCCACATTTTTAC
NbHIN1 fw qRT- PCR GAGGGTCACAAGAATACTAGCAGC
NbHIN1 rev qRT- PCR CGCATGTAAAGCTTCACTTCCATCTC
NbACRE31 fw* qRT- PCR AAGGTCCCGTCTTCGTCGGATCTTCG
NbACRE31 rev* qRT- PCR AAGAATTCGGCCATCGTGATCTTGGTC
NbACRE132 fw* qRT- PCR AAGGTCCAGCGAAGTCTCTGAGGGTGA
NbACRE132 rev* qRT- PCR AAGAATTCCAATCCTAGCTCTGGCTCCTG
NbEF1 fw qRT- PCR GCTGCTGCAACAAGATGGATG
NbEF1 rev qRT- PCR CGAGCATGTTGTCACCTTCCA
Sequences for restriction sites are underlined, stop codons are in italics.
* - as described in Segonzac et al., (2011).
76
77
CHAPTER 3
Molecular interactions of Medicago truncatula putative Nod factor
receptors, NFP and MtLYK3, as demonstrated with Fluorescence
Resonance Energy Transfer-Fluorescence Lifetime Imaging
Microscopy.
Anna Pietraszewska-Bogiel1, Inge van’t Veer
1, Kevin C. Crosby
1, Laura van Weeren
1,
Joachim Goedhart1, and Theodorus W.J. Gadella
1
1 Section of Molecular Cytology, Swammerdam Institute for Life Sciences, University of Amsterdam, Science
Park 904, 1098 XH Amsterdam, The Netherlands
This work was supported by the EC Marie Curie Research Training Network Programme through contract
MRTN-CT-2006-035546 “NODPERCEPTION”. We thank Dr. Giulia Morieri (John Innes Centre, Norwich,
UK) for providing purified Sinorhizobium meliloti Nod factor.
78
SUMMARY
During symbiosis of legume plants with nitrogen-fixing rhizobia bacteria,
receptor-like kinases (RLKs) with Lysin Motif (LysM) domains in their extracellular
region (LysM-RLKs) are postulated to function as receptors for bacterial lipo-
chitooligosaccharidic signals, termed Nod factors (NFs). Despite substantial progress
in identifying molecular components essential for Rhizobium-legume (RL) symbiosis,
little is known about the exact activation and signaling mechanisms of the putative NF
receptors. Recent studies have demonstrated that plant RLKs, similarly to animal
receptors, rely on oligomerization to signal. Here, we present in vivo analysis
of oligomerization status of two Medicago truncatula (Medicago) putative NF
receptors, MtNFP and MtLYK3, heterologously produced in Nicotiana benthamiana
(Nicotiana) leaves. Using Fluorescence Resonance Energy Transfer-Fluorescence
Lifetime Imaging Microscopy (FRET-FLIM) we demonstrated that MtNFP
and MtLYK3 formed homo(di)mers at the plasma membrane (PM) of Nicotiana leaf
epidermal cells that was not dependent on the MtLYK3 kinase activity or the presence
of NF. Heteromerization of these LysM-RLKs was studied using a kinase-inactive
MtLYK3 variant, due to the development of cell death (CD) in response to
the co-production of wild-type (WT) MtLYK3 and MtNFP. Tendency of these proteins
to heteromerize was also independent of the presence of NF but was less pronounced
than their homomerization potential. Specific considerations for in vivo FRET studies
of dynamic (plant) protein-protein interactions in membranes are discussed.
79
INTRODUCTION
Many of the PM-spanning proteins that perceive and transduce extracellular
signals in eukaryotic cells rely on di- or oligo-merization to function. This can be
through homo-merization (i.e. interaction of identical receptor protomers)
or heteromerization (i.e. interaction of different, though often structurally related
receptor protomers). Di- or oligo-merization of receptor protomers is postulated to
modulate their affinity for a ligand. In addition, di- or oligo-merization is often required
to stimulate kinase activity of the cytosolic kinases associated with the intracellular
regions (InRs) of these receptor protomers (Heldin 1995). Similar kinase activation
via allostery or (intermolecular) transphosphorylation has been reported for cell surface
receptors possessing an intrinsic kinase activity, termed receptor kinases (RKs),
e.g. animal receptor Tyr kinases (RTKs) are generally displayed on the cell surface
as monomers and their di- or (less frequently) oligomerization is driven by ligand
binding (Lemmon & Schlessinger 2010; Ward & Lawrence 2012). In case of animal
TGF-β (for transforming growth factor β) receptor, two distinct homodimers have to
heterodimerize in order to transduce the signal. This is because type-I TGF-β receptors
(or transducers) are unable to bind ligands in the absence of type-II TGF-β receptors
(or primary receptors), whereas type-II receptors are unable to signal in the absence
of type-I receptors (Groppe et al., 2008). In both RTKs and TGF-β receptors,
dimerization results in stimulation of their kinase activity, and hence their signaling
output. Therefore, dynamic modulation of the oligomerization status is a general
mechanism for regulating activation of animal cell surface receptors.
Plant RLKs (termed RKs upon demonstration of the ligand binding)
are postulated to share a monophyletic origin with animal RTKs and TGF-β receptors
80
(Shiu & Bleecker 2003 and refs therein), and thereby are also believed to employ
similar mechanisms for activation. Corroborating this notion, three Arabidopsis thaliana
(Arabidopsis) RKs: Clavata 1 (AtCLV1), Brassinosteroid Insensitive 1 (AtBRI1),
and Flagellin Sensing 2 (AtFLS2) have been demonstrated to homomerize in planta
(Russinova et al., 2004; Hink et al., 2008; Bleckmann et al., 2009; Guo et al., 2010;
Sun et al., 2012 and refs therein), and this homomerization often increases upon ligand
stimulation (Wang et al., 2005). In addition, several plant RLKs, but not AtCLV1,
have been shown to heteromerize in a ligand-dependent manner with one or several
different members of Arabidopsis Somatic Embrogenesis Receptor Kinase (AtSERK)
subfamily (Wang et al., 2008 and refs therein; Karlova et al., 2006; Jeong et al., 2010;
Schulze et al., 2010; Jaillais et al., 2011; Roux et al., 2011; Schwessinger et al., 2011
and refs therein). Finally, heteromerization of AtBRI1 and AtBRI1-Associated Kinase 1
(AtBAK1)/AtSERK3 is postulated to result in sequential reciprocal trans-
phosphorylation of both proteins, which ultimately increases the kinase activity
downstream signaling output of AtBRI1 (Wang et al., 2008).
In several legume plant species, specific LysM-RLKs have been implicated
in NF perception during the RL interaction. In Medicago, Nod Factor Perception
(MtNFP) is required for early symbiotic signaling: a complex network of cellular
and molecular processes that control nodule organogenesis and prepare the host root
for infection by rhizobia (Ben Amor et al., 2003; El Yahyaoui et al., 2004; Mitra et al.,
2004a; Arrighi et al., 2006). The subsequent curling of root hair cells that initiates
the rhizobial infection requires Medicago LysM domain-containing RLK/Root Hair
Curling (MtLYK3/HCL, from now on referred to as MtLYK3) (Catoira et al., 2001;
Smit et al., 2007). In addition, formation of so-called infection threads (ITs) through
which the bacteria colonize the nodule primordium requires both MtNFP and MtLYK3
81
(Limpens et al., 2003; Arrighi et al., 2006; Smit et al., 2007; Bensmihen et al., 2011),
indicating their co-function at this step. Finally, detection of MtNFP and MtLYK3
transcripts/promoter activity in the nodule primordium and the invasion zone
of a mature nodule (Limpens et al., 2005; Arrighi et al., 2006; Mbengue et al., 2010;
Haney et al., 2011) suggests a role for the encoded LysM-RLKs in nodule development
and/or accommodation of rhizobia inside nodule cells.
Interestingly, in contrast to MtLYK3, which shows in vitro phosphorylation
activity, MtNFP appears to be a pseudokinase, i.e. it does not show nor rely on
the intrinsic kinase activity to signal (Arrighi et al., 2006; Mbengue et al., 2010; Klaus-
Heisen et al., 2011; Madsen et al., 2011; Lefebvre et al., 2012). Mechanisms through
which pseudokinases function only start to be revealed. The few functionally
characterized examples indicate that they can interact with and activate (via allostery)
other true kinases (Zegiraj & Aalten 2010). Alternatively, pseudokinases may function
as scaffolding proteins, facilitating molecular interactions among different proteins with
which they interact (Zegiraj & Aalten 2010). Likewise, the formation of a receptor
complex between MtNFP and a yet-unidentified LysM-RLK or MtLYK3 is postulated
to initiate the early symbiotic signaling and the infection process, respectively (Arrighi
et al., 2006; Smit et al., 2007; Bensmihen et al., 2011). However, direct proof
of the molecular interaction between MtNFP and any other member of the LysM-RLK
family of Medicago (Limpens et al., 2003; Arrighi et al., 2006), is currently lacking.
In our previous study, we obtained indications of a functional interaction
between MtNFP and MtLYK3 in a heterologous system of Nicotiana (see Chapter 2).
In the current study, we employed the FRET-FLIM technique to investigate a possible
molecular interaction between these LysM-RLKs produced in Nicotiana leaf epidermal
cells. FRET is the physical phenomenon of non-radiative transfer of energy from
82
an excited fluorophore, termed donor, to a nearby chromophore, termed acceptor.
This can occur only if the two molecules are sufficiently close to each other (generally
not further apart than 10 nm), and if other requirements are fulfilled (Jares-Erijman &
Jovin 2003; Pietraszewska-Bogiel & Gadella 2010). In the FRET-FLIM, changes
of the donor fluorescence lifetime ( ), i.e. of time the donor stays in the excited state,
are monitored:
decreases (is quenched) due to FRET and thus reports
on the proximity of the donor and the acceptor. Because the measurement
is independent of the intensity of the fluorescent signal (although this parameter
becomes of importance in samples containing multiple fluorescence species;
see Discussion), FRET-FLIM allows robust investigation of proximity between
molecules, and is the method of choice for interaction studies in many cell biology
applications (Pietraszewska-Bogiel & Gadella, 2010). We aimed at investigating
interactions between MtNFP and MtLYK3 proteins fused C-terminally to super yellow
fluorescent protein (sYFP2; Kremers et al., 2006) or mCherry (Shaner et al., 2004)
heterologously produced in Nicotiana leaf using frequency domain FLIM operating
on a wide field microscopy setup (van Munster & Gadella 2004ab). The sYFP2
and mCherry were chosen as donor and acceptor, respectively, because of the enhanced
Förster radius of this FRET pair as compared to other pairs available at the time
of experimental planning (Goedhart et al., 2007), and because of optimal compatibility
with the spectral characteristics of the leaf tissue autofluorescence.
Here, we demonstrate that the tendency of MtLYK3 and (to a lesser extent) MtNFP
to form homomers at the PM of Nicotiana leaf epidermal cells is more pronounced
than MtNFP tendency to heteromerize with a kinase-inactive MtLYK3 variant. Specific
considerations for in vivo FRET studies of dynamic (plant) protein-protein interactions
in membranes are discussed.
83
RESULTS
Establishment of robust conditions for FRET-FLIM experiments in Nicotiana leaf
Expression of all constructs in study was driven by a constitutive 35S promoter
of the cauliflower mosaic virus (CaMV) in order to ascertain efficient production
of the encoded protein fusions in Nicotiana leaf upon Agrobacterium-mediated transient
transformation (Agro TT). In order to avoid too high expression levels and bystander
FRET (FRET due to concentration effect and not due to specific protein-protein
interaction, see below), Agrobacterium transformants carrying the respective constructs
were infiltrated into Nicotiana leaves at low concentrations (measured as OD600),
and care was taken to measure only from cells with relatively low accumulation
of both donor- and acceptor-tagged proteins. In addition, was not measured
in those regions of interest (ROIs) that adjoined cells displaying high accumulation
of the donor-tagged protein (see Fig. 1 for description), as the fluorescence signal
from the latter could dominate the measurement. Frequency-domain FLIM provides two
independent estimates of the average : one calculated from the phase shift ( )
and one from the demodulation ( M ) of the fluorescence relative to the excitation.
An intensity threshold was applied post-acquisition in order to calculate and M
predominantly from the cell boundary region (Fig. 1), in agreement with the PM
localization of the RLKs in study (see Chapter 2). We used the square root
of the multiplied and M values (i.e. M ) to characterize in every
sample. The apparent (see below) FRET efficiency ( appE ) was calculated as:
D
DAappE
1 (eq. 1)
84
where DA is the average measured in samples containing both donor- and acceptor-
tagged proteins (so called FRET samples) and D is the average measured in
samples containing only the respective donor-tagged protein (the reference samples).
Immediately after acquiring the images of distribution, two images
recording the sYFP2 and mCherry fluorescence intensity (termed donor [ID]
and acceptor [IA] intensity, respectively) were acquired using direct excitation
(see Experimental Procedures) (Fig. 1). The ID and IA values were calculated from the
same region as that used to calculate the values, and were used for quality control.
This included check for possible artifacts in measurement due to autofluorescence,
unbalanced donor- to acceptor-tagged protein levels, and bystander FRET
that was carried out as follows.
The autofluorescence of Nicotiana leaf tissue (transformed with an empty
vector) excited with the 514 nm wavelength displayed a short lifetime (0.77±0.09 ns
for , 1.66±0.09 ns for M ; n = 10). In order to analyze its effect on the measurement
of sYFP2 , we analyzed several different leaf samples expressing only MtNFP-sYFP2
protein fusion, in which a range of measured ID values indicated varying levels
of accumulation of the encoded protein fusion. We noted that lower values
were measured in samples characterized by lower ID values (Fig. 2A). Due to relatively
constant amount of autofluorescence in each leaf sample (data not shown), this result
indicated that the value decreased proportionally to the relative contribution
of the autofluorescence in the overall recorded signal. Furthermore, the comparison
of values measured in samples with varying ID values demonstrated that the sheer
contribution of the autofluorescence could yield false positive estimation of appE
(see Fig. 2A for description). Therefore, care was taken to calculate appE only using
85
Intensity Thresholding
= 2.92 ± 0.08 [ns]
ID = 2100 [a.u.]
IA = 4300 [a.u.]
= 3.05 ± 0.08 [ns]
ID = 2500 [a.u.]
IA= 2400 [a.u.]
YFP intensity mCherry intensity
Figure 1. Post-acquisition processing of images of distribution.
The upper left panel presents a ROI from a distribution image of Nicotiana leaf epidermal cell producing
MtNFP-sYFP2 protein fusion acquired on the FRET-FLIM setup. An intensity threshold (depicted as red
and blue highlight in the upper middle panel) was applied post-acquisition in order to calculate
and M predominantly from the cell boundary region encompassing PM. Immediately after acquiring
the images of distribution, two images recording the sYFP2 fluorescence intensity ([ID], depicted
in the lower left panel) and mCherry fluorescence intensity ([IA], depicted in the lower middle panel)
were acquired using direct excitation. The ID and IA values were calculated from the same region as that used
for calculating and
M . Right panel present , ID and IA values calculated from red and blue regions
depicted in the upper middle panel. Please note that the region highlighted in blue encompasses both a part
of the upper cell boundary (producing both donor-tagged and acceptor-tagged proteins) and a part of the lower
cell boundary (producing high levels of donor- and low levels of acceptor-tagged protein) (as estimated
from the ID and IA images in the lower left and middle panel, respectively). The fluorescence signal
from the latter cell (displaying low FRET) would dominate the overall detected fluorescence signal,
and thereby was not measured in this ROI.
D and DA values that were characterized by the same or comparable ID values high
enough to exclude a large autofluorescence bias.
86
A
B
C
Figure 2. Optimization of conditions for FRET-FLIM experiments in Nicotiana leaf.
A, Estimating bias in measurement due to autofluorescence. Agrobacterium transformants carrying
MtNFP-sYFP2 construct were infiltrated at a range of concentrations into Nicotiana leaf (in adjoining circles).
The average (grey diamonds),
M (open diamonds), M (black diamonds) present
in the sample are plotted as a function of ID (in arbitrary units [a.u.]).
Please note that the lower values are characterized by the lower ID values, and that (in the presented
example) the sheer effect of autofluorescence on measurement could result in estimation of 2 (in M )
to 6 (in ) % of
appE when the values corresponding to the extreme ID values are compared. Therefore,
if the DA values calculated for a specific FRET sample were distributed within a certain range of ID values
87
(e.g. 200-400 [a. u.]), the reference D value used for the estimation of
appE was calculated using only
the measurements that falls within the same range of ID values.
B, Estimating the ratio of donor- to acceptor-tagged proteins. Agrobacterium transformants carrying MtLYK3-
sYFP2 construct were infiltrated separately (reference sample) or together (FRET sample)
with Agrobacterium transformants carrying MtLYK3-mCherry construct into Nicotiana leaf. The average D
(black diamonds) and DA (grey triangles) present in the sample are plotted as a function of fA. Please note
that the lower DA values are characterized by the higher fA values. Only the measurements from a defined
range of fA values (e.g. 0.5-0.6) were included in the calculation of appE .
C, Estimating putative presence of bystander FRET. The DA measurements for a selected range of fA values
(here the range 0.5-0.6 was used) were plotted as a function of IA (in a.u.). to check whether the lowest DA
values do not correspond exclusively with the highest IA values.
Secondly, the optimal FRET experiment requires the ratio of donor-tagged
to acceptor-tagged protein not lower than 1:1; otherwise the excited state donors
compete for acceptors. In order to evaluate the fraction of donor- to acceptor-tagged
proteins in each sample, the relative fraction of the latter (fA) can be calculated
by normalizing IA to the sum of ID and IA (i.e. fA=IA/(IA+ID), yielding a dimensionless
parameter ranging from 0 to 1. A value close to 0 reports on virtual absence
of the acceptor-tagged protein; a value approaching 1 reports on great over-
accumulation of acceptor-tagged protein; and a value approximating 0.5 - on balanced
production of donor- and acceptor-tagged protein fusion. To increase the sensitivity
of FRET detection, only the samples where this fraction was equal or greater than 0.5
were included in the calculations of appE (Fig. 2B). The samples with insufficient
production of acceptor-tagged protein (fA lower than 0.2) showed DA values
comparable with the D value, and were regarded as an internal negative control.
Thirdly, to exclude major artifacts due to bystander FRET caused
by overproduction of the acceptor, the appE
was plotted as a function of the IA
88
for samples with comparable relative fractions of donor- to acceptor-tagged protein
[fA value] (Fig.2C). In case of significant bystander FRET, samples with equal fA
but high IA values will display lower DA . For the results shown below, cells displaying
overproduction of acceptor-tagged protein giving rise to significant bystander FRET
were discarded for further analysis.
FRET measurements indicate the capacity of MtLYK3 and MtNFP to homomerize
in the plasma membrane of Nicotiana leaf epidermal cells
We started our FRET-FLIM analysis by investigating whether the RLKs
in study can homomerize. To this end, the respective constructs were co-expressed
in Nicotiana leaf via Agro TT, and was measured 24 hours after infiltration (hai)
or 36hai (see Experimental Procedures). The DA values measured in leaf samples
co-producing MtLYK3-sYFP2 and MtLYK3-mCherry constructs were significantly
lower (on average 2.86±0.02 ns) than the reference D (on average 3.06±0.02 ns),
yielding an appE = 6.3±0.8% (Fig. 3). In order to investigate whether kinase activity
is required for homomerization, we studied the MtLYK3 [G334E] mutated variant
(Klaus-Heisen et al., 2011 and Chapter 2). The DA values measured in leaf samples
co-expressing MtLYK3 [G334E]-sYFP2 and MtLYK3 [G334E]-mCherry constructs
were significantly lower (on average 2.88±0.02 ns) than the reference D (on average
3.03±0.02 ns), yielding an appE = 5.1±0.8% (Fig. 3). Finally, significantly lower values
(on average 2.90±0.02 ns) were measured in leaf samples co-expressing MtNFP-sYFP2
and MtNFP-mCherry constructs relative to the respective reference D (on average
3.04±0.01 ns), yielding an appE = 4.6±0.8% (Fig. 3).
89
Figure 3. Homomerization of MtLYK3, MtLYK3 [G334E], and MtNFP fluorescent protein fusions
at the plasma membrane of Nicotiana leaf epidermal cells.
Agrobacterium transformants carrying MtLYK3-sYFP2 (LYK3), MtLYK3 [G334E]-sYFP2 (LYK3 [G334E])
or MtNFP-sYFP2 (NFP) construct were infiltrated separately (reference sample) or together with
Agrobacterium transformants carrying the respective construct fused to mCherry (FRET sample). Columns
present the measured values normalized to the respective D value, and the bars present standard
deviation of the measurements (the values and st. dev. are also presented in the legend). Number of images
analyzed (n) is given below. In the legend, the identity of donor-tagged protein is specified as first, before
the acceptor-tagged protein.
Heteromerization between kinase-inactive MtLYK3 variant and MtNFP
in the plasma membrane of Nicotiana leaf epidermal cells is less pronounced
Co-expression of MtNFP and MtLYK3 in Nicotiana leaf induces cell death
(CD) within 48hai (see Chapter 2). Nevertheless, co-infiltration of Agrobacterium
transformants carrying either MtNFP-sYFP2 or MtLYK3-mCherry construct at higher
concentration allowed detection of both encoded protein fusions approximately
20-24hai at the boundary of single dispersed cells (data not shown). However, before
proceeding with measurements, we tested whether the autofluorescence associated
with CD could be excited with the 514 nm wavelength and detected with the emission
window (525-556 nm) used in our experiments. To this end, leaf samples co-expressing
untagged MtNFP and MtLYK3 constructs were imaged 24hai on the FRET-FLIM setup.
Bright patches of autofluorescence could be detected using 514 nm excitation
90
wavelength, and displayed short (0.76±0.07 ns for , 1.66±0.07 ns for M , n=10).
This indicated that autofluorescence associated with CD induced upon simultaneous
accumulation of MtNFP and MtLYK3 must be taken into consideration during FRET-
FLIM analysis of their putative heteromerization. Therefore, leaf samples
co-expressing MtNFP-sYFP2 and untagged MtLYK3 or MtLYK3-mCherry constructs
were imaged concomitantly. The appE
calculated for leaf samples co-expressing
MtNFP-sYFP2 and untagged MtLYK3 or MtLYK3-mCherry constructs were not
significantly different and accounted to 5.3±1.4% and 6.2±1.2%, respectively (Fig. 4A).
In contrast to WT MtLYK3, a kinase-inactive MtLYK3 variant carrying
the G334E substitution does not induce CD in Nicotiana leaf in the presence of MtNFP
(see Chapter 2). Therefore, we decided to investigate putative MtNFP and MtLYK3
heteromerization using this MtLYK3 variant. Leaf samples (co-)producing MtLYK3
[G334E]-sYFP2 alone or together with untagged MtNFP construct displayed
comparable values (see below), in agreement with the inability of this mutated
variant to induce accumulation of blue light/UV-excited fluorescence in Nicotiana leaf
in the presence of MtNFP (see Chapter 2). Confocal microscopy analysis of leaf
samples co-producing MtLYK3 [G334E]-sYFP2 and MtNFP-mCherry demonstrated
co-localization of the protein fusions at the cell boundary, in agreement with the PM
localization of both RLKs (see Chapter 2). The DA values measured in leaf samples
co-expressing MtLYK3 [G334E]-sYFP2 and MtNFP-mCherry constructs
were significantly lower (on average 2.95±0.02 ns) than D measured in leaf samples
co-expressing MtLYK3 [G334E]-sYFP2 and untagged MtNFP constructs or expressing
only MtLYK3 [G334E]-sYFP2 construct (on average 3.06±0.02 ns), and accounted to
91
3.6±0.8% (Fig. 4A). The appE
calculated for the reciprocal protein fusions
(i.e. MtNFP-sYFP2 and MtLYK3 [G334E]-mCherry) equaled 2.8±0.7% (Fig. 4A).
While these two estimated appE values were within their mutual error margins,
the estimated ratio of donor- to acceptor-tagged proteins in leaf samples co-expressing
MtLYK3 [G334E]-sYFP2 and MtNFP-mCherry constructs was 1:3, whereas it was 1:1
in leaf samples co-expressing MtNFP-sYFP2 and MtLYK3 [G334E]-mCherry.
Treatment with NF does not seem to affect the heteromerization between MtNFP
and kinase-inactive MtLYK3 variant
Subsequently, we studied the effect of purified NF of S. meliloti on MtNFP
and MtLYK3 heteromerization. To this end, was measured in leaf samples
co-expressing MtLYK3 [G334E]-sYFP2 and MtNFP-mCherry constructs before
and after they were treated with purified NF of S. meliloti at 10-7
M or with DMSO
(the solvent for NF) diluted to the same concentration. The DA values measured in leaf
samples co-expressing MtLYK3 [G334E]-sYFP2 and MtNFP-mCherry constructs
and treated with the NF were significantly lower (on average 2.92±0.02 ns) than
the reference D (on average 3.06±0.02 ns). However, the DA measured in leaf
samples co-expressing MtLYK3 [G334E]-sYFP2 and MtNFP-mCherry constructs that
were treated with a control DMSO dilution were similarly low (on average 2.92±0.02
ns). A similar decrease of value was observedin leaf samples producing only
MtLYK3-sYFP2 or only MtNFP-sYFP2 protein fusion and treated with the NF (data
not shown). In addition, we noted the decrease of ID and IA values upon treatment with
either NF or control DMSO dilution, as compared to untreated leaf samples, although
confocal microscopy analysis did not reveal any change in the subcellular
92
A
B
Figure 4. Heteromerization of MtLYK3 [G334E] and MtNFP fluorescent protein fusions at the plasma
membrane of Nicotiana leaf epidermal cells.
Columns present the measured values normalized to the respective D value, and the bars present
standard deviation of the measurements (the values and st. dev. are also presented in the legend). Number
of images analyzed (n) is given below. In the legend, the identity of donor-tagged protein is specified as first,
before the acceptor-tagged protein or the NF/DMSO treatment applied.
A, Agrobacterium transformants carrying MtNFP-sYFP2 (NFP) or MtLYK3 [G334E]-sYFP2 (LYK3
[G334E]) construct were infiltrated: separately (reference sample); together with Agrobacterium
transformants carrying MtLYK3 untagged or MtNFP untagged construct (control sample); co-infiltrated with
Agrobacterium transformants carrying MtLYK3 [G334E]-mCherry (LYK3) or MtNFP-mCherry (NFP)
construct (FRET samples).
B, Effect of S. meliloti NF. Agrobacterium transformants carrying MtLYK3 [G334E]-sYFP2 (LYK3 [G334E])
construct were infiltrated separately (reference sample) or together with Agrobacterium transformants
carrying MtNFP-mCherry (NFP) construct. The measurements were performed prior and immediately
after the application of purified NF (at 10-7 M) or DMSO (diluted to the same concentration).
93
localization of MtLYK3 [G334E]-sYFP2 or MtNFP-mCherry protein fusion upon either
treatment (data not shown). The FRET calculated using only the measurements
characterized by comparable ID values accounted to appE = 4.5%±1% and 4.7±0.8%
upon treatment with either NF or control DMSO dilution, respectively (Fig. 4B).
Therefore, application of specific NFs did not seem to affect the extent
of heteromerization between MtNFP and kinase-inactive variant of MtLYK3
in Nicotiana leaf epidermal cells.
Estimations of a fraction of homomerizing and heteromerizing MtNFP
and MtLYK3 protein fusion
As stated before, frequency-domain FLIM provides two independent estimates
of the average : one calculated from the phase shift ( ) and one from
the demodulation ( M ) of the fluorescence relative to the excitation. When assuming
monoexponential decay with lifetime D for an unquenched donor, the FLIM method
will yield one estimated lifetime, i.e. DM . However, in case of a mixed
population of FRETting donors displaying a quenched lifetime ( DA ) and donors
displaying unquenched lifetime ( D ), the two lifetime estimates differ due to different
weighing of lifetimes and fractional contributions. For this dual decay situation it can be
shown that the and M are given by:
G
S
1
111
22
GSM
(eq. 2 and 3)
94
where S and G are given by:
222 121 DA
DADA
D
DDS
2222 11 DA
DA
D
DG
αD and αDA are the fractional contribution to the steady-state fluorescence intensity
of the non-FRETting and FRETting donors, respectively, and are given by:
Ef
f
DA
DAD
1
1
Ef
Ef
DA
DADA
1
)1(
where is the angular frequency of modulation ( f2 ; 1.75f MHz for our
experiments); E is the FRET within a dimer; and DAf is the fraction of FRETting
donor molecules as compared to the total population of donor molecules (equal to
the fraction of FRETting dimers, i.e. composed of donor- and acceptor-tagged proteins).
Therefore, equations to calculate and m include the two variables: DAf
and E, a constant D , and DA that is equal to )1( ED . In order to understand
our results in terms of fraction of (oligo)merizing protein, we simulated every
combination of these two variables (Fig. 5A). To simplify the calculations, we assumed
formation of dimers rather than oligomers, and we assumed that, in case of homo-
merization, the affinity for dimerization is identical for donor- and acceptor-tagged
proteins. For simulated DAf and E values (both can be between 0 and 1), we calculated
corresponding S and G parameters (using eq. 4 and 5), then and M (using eq. 2
and 3), and subsequently the square root of their values, termed the apparent average
lifetime ( app ). Finally, assuming the unquenched donor lifetime of D = 3.05 ns
(eq. 4 and 5)
(eq. 6 and 7)
95
(the average D in our measurements), we calculated appE for each combination
of varying DAf and E values:
D
app
appE
1
Therefore, by assuming the two variables ( DAf and E) as known, we ultimately
were able to calculate the appE value, thus correlating it with the minimal (see below)
fraction of FRETting dimers ( DAf ) that was required to generate this value. Plotting
of the calculated appE (resulting from the simulated DAf values) as a function of E
(Fig. 5B) showed that for each simulated DAf value, the appE increased until it reached
a maximum at certain E value. In addition, the same appE value could be generated
by various DAf , depending on E in a given dimers. For example, appE =0.08 (8%)
could be obtained for DAf =1 (100%) displaying E<0.1 (<10%), as well as for lower
DAf values displaying E>0.1 (please, look for the intercept of appE = 0.08
with the curves representing various simulated DAf values). However, in order to obtain
appE = 0.08 in our example, the DAf value would have to be minimally 0.2 (please note
that the curve for DAf <0.02 do not intercept the appE = 0.08).
In order to estimate the minimal DAf value required to detect a given
appE value, we identified the maximal appE values in each simulated DAf (boxed
in Fig. 5A), and plotted them as a function of these DAf values (Fig. 5C). Please, note
that at the same time the identified maximal appE values could be found for higher
DAf values (e.g. appE = 0.4 was present in DAf = 0.4 and DAf = 0.8
but as a maximum was marked for DAf = 0.1). In other words, we identified
the minimal DAf values that could generate the given (maximal) appE values, and used
(eq. 8)
96
the resulting graph (Fig. 5C) to estimate the minimal DAf values that were required
to obtain the average appE values detected in our experiments. From this graph, it could
be estimated that in order to detect the appE = 2.8% (value calculated for MtLYK3
[G334E] and MtNFP heteromerization), the minimal DAf value would have to be 7%
(Fig. 5C). In case of MtLYK3, MtLYK3 [G334E] and MtNFP homomerization,
the average appE of 6.4%, 5% and 4.6%, respectively, resulted in estimation of at least
16% of MtLYK3, 12% of MtLYK3 [G334E]-sYFP2, and 11% of MtNFP-sYFP2
protein fusion present in FRETting homomers.
This brings us to another notion that only in FRETting homomers display
quenched donor lifetime values, whereas homomers composed of only donor-tagged
proteins display an unquenched value that is equal to D (please, note that
this situation does not apply to heteromers). In other words: for heterodimerization,
the fraction of dimerized molecules ( dimf ) is equal to DAf , but for homodimerization
DAf is lower than dimf (even in case of 100% homodimerization). In the latter
situation, the chance of finding: a dimer composed of two donor-tagged proteins
is proportional to2
Df (where Df is the fraction of donor-tagged protein); a dimer
composed of two donor-tagged proteins is proportional to
2)1( Df ; and of a FRET-
ting dimer is proportional to )1(2 DD ff (note that these factors add up to 1).
In case of 100% homodimerization, the chance of finding a donor-tagged protein
in a FRETting dimer equals D
DDD
DD ffff
ff
1
)1(22
)1(22
.
Therefore, in case of homo-merization, the experimentally determined DAf
underestimates dimf by a factor Df1 :
A
DA
D
DA
f
f
f
ff
1dim
(eq. 9)
97
98
0,00
0,02
0,04
0,06
0,08
0,10
0,12
0,14
0,16
0,18
0,20
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0
0,02
0,04
0,06
0,08
0,10
0,20
0,30
0,40
0,50
0,60
0,70
0,80
0,90
1,00
fDA (color coded)D
ete
cte
d E
ap
p
FRET within dimer (E)
Figure 5. Estimating a fraction of protein present in FRETting dimers based on the simulation ofDAf
and E values.
A, DAf and E values ranging from 0 to 1 were simulated in a matrix (on x and y axis, respectively),
and app
was calculated for all combinations of these two variables (not shown).
appE was calculated using
these app
values and
D value of 3.05 ns. The maximal appE values for each simulated
DAf are boxed.
B, appE calculated for each
DAf (see the color-coded legend) was plotted as a function of E. Please note
that a given appE value can be obtained for various
DAf , depending on E in a given dimer, e.g. appE = 0.08
(8%) can be obtained for allDAf values which curves intercept the value 0.08 on the y axis (indicated with
a dashed line). TheDAf values that do not intercept this line (e.g. DAf < 0.2) cannot display
appE = 0.08.
C, CalculatedappE was plotted as a function of minimal
DAf that can generate this value. The appE values
detected in MtLYK3 (6.4%), MtLYK3 [G334E] (5.1%), and MtNFP (4.6%) homomerization experiments,
as well as the appE value detected in MtNFP and MtLYK3 [G334E] (2.8%) heteromerization experiment
are indicated with dashed lines.
B
Eapp (detected/calculated)
Min
imal
fD
A
C C
99
As a consequence, FRETting homodimers in samples with roughly 1:1 ratio of donor-
and acceptor-tagged protein ( 5.0 AD ff ) yields a dilution of the FRET signal
by a factor of 2. This was the case in all presented MtLYK3 and MtLYK3 [G334E]
homomerization experiment, and hence the total fraction of homodimerizing MtLYK3
and MtLYK3 [G334E] protein fusions are twice the estimates of DAf values,
and account to at least 32% and at least 24% for MtLYK3 and MtLYK3 [G334E],
respectively. In case of the MtNFP homo-merization, the analyzed samples
were characterized by approximately 1:3 ratio of donor- to acceptor-tagged protein
( 75.0Af ). Therefore, the total fraction of homodimerizing MtNFP protein fusion
could be estimated as at least 15%.
DISCUSSION
Using FRET-FLIM, we were able to demonstrate specific homomerization
of MtNFP and MtLYK3 fluorescent protein fusions at the PM of Nicotiana leaf
epidermal cells (Fig. 3) that agreed with the demonstrated homomerization of their
orthologues (Zhang et al. 2009b and refs therein) from L. japonicus, LjNFR5
and LjNFR1 (Madsen et al., 2011), and MtLYK3/LjNFR1 homolog, CERK1, from rice
(Oryza sativa OsCERK1; Shimizu et al., 2010), and Arabidopsis (AtCERK1; Liu et al.,
2012b). Although not directly comparable, the appE values obtained for
the homomerization of LysM-RLKs in study were lower than those measured in vivo
for fluorescent protein fusions of AtBRI1 and AtSERK1 (20% and more; Russinova
et al., 2004). However, we cannot exclude that different treatment of the effect
of autofluorescence on measurements (Fig. 2A) accounted, at least partially,
100
for this difference. In fact, we noted that autofluorescence (in addition to bystander
FRET) was one of the main factor contributing to occasional measurements of low
values, and in turn – of high appE (data not shown). In order to calculate appE value
correctly, it was essential that both D and DA were characterized by the comparable
ID values. At the same time, we note that this is often not characterized in published
FRET-FLIM reports. Therefore, we propose that ID value should be included into
the characteristic of the FRET/reference samples. In addition, we demonstrate that
characterization of the FRET samples for IA helps to optimize FRET analysis,
as the knowledge of both ID and IA values allows to analyze only those measurements
with favorable donor- to acceptor-tagged proteins levels. This is similar to the recently
reported benefits of working with controlled donor- to acceptor-tagged protein levels
in FRET-FLIM analysis in mammalian cells (Goedhart et al., 2011).
In addition, complementation of FRET-FLIM studies with estimations
of the donor- and acceptor-tagged protein fractions present in the analyzed samples
allowed us to gain further insights into the oligomerization of these LysM-RLKs.
We systematically analyzed how the detected FRET signal is dependent on:
1) the extent of FRET in a FRETting dimer (E), and 2) the fraction of FRETting dimers
( DAf ). Assuming dimerization as the predominant mechanism of MtLYK3 homo-
merization, we estimated that at least 32% of MtLYK3 protein fusion forms
homodimers (see Fig. 5C and eq. 9). Although we cannot exclude a possibility
of multimeric (ternary and higher) complex formation by MtLYK3,
our assumptions are supported by the exclusively homodimer formation and lack
of evidence for higher-order complex formations reported for AtBRI1 and AtSERK1
(Hink et al., 2008). Reported estimations of the fraction of AtSERK1 (15%)
and AtBRI1 (20%) homodimers present at the PM of Arabidopsis protoplasts (Shah
101
et al., 2001; Hink et al., 2008) agree with our findings on MtLYK3 homo(di)merization.
In addition, we showed that homomerization of MtLYK3 did not depend on its kinase
activity, and we estimated that at least 24% of MtLYK3 [G334E] protein formed
homo(di)mers. In case of MtNFP, the lower estimated DAf value (15%) can indicate
a lower tendency of this LysM-RLK to homomerize, as well as less favorable spatial
orientation of donor and acceptor molecules within the formed (di)mer than that
in the MtLYK3 or MtLYK3 [G334E] (di)mer.
NF-independent interaction of MtNFP and MtLYK3 might be a result
of their heterologous production in Nicotiana leaf
We were not able to demonstrate a specific effect of S. meliloti NF
on the homo-merization of the LysM-RLKs in study, as the treatment with DMSO
dilution similarly decreased the measured . We hypothesize that this effect might
result from the cell response to filling in the apoplastic space with water that increases
the local refractive index and thereby lowers the value. Another method of NF
application will be required to study its effect on the oligomerization of LysM-RLKs
in study. On the other hand, the demonstrated homo(di)merization of MtNFP
and of MtLYK3 in the absence of S. meliloti NFs is consistent with ligand-independent
formation of receptor complexes in transient expression studies reported previously
(Bleckmann et al., 2009), and with chitin-independent oligomerization of AtCERK1
in vivo in a situation when its production was driven by the CaMV 35S promoter (Liu
et al., 2012b). Alternatively, an Agrobacterium-derived signal with sufficient structural
similarity to the postulated lipo-chitooligosaccharidic ligand of MtNFP and MtLYK3
could trigger their homomerization.
102
Heteromerization of MtNFP and MtLYK3 might require MtLYK3 kinase activity
Co-expression of MtNFP and MtLYK3 in Nicotiana leaves results in CD
and enhanced tissue autofluorescence which can be excited with 514 nm wavelength
and displays short . Imaging MtNFP and MtLYK3 fluorescent protein fusions in cells
undergoing death was very inefficient, and the measured was significantly affected
by the autofluorescence contribution. For this reason, we were not able to investigate
heteromerization between MtNFP and WT (kinase active) MtLYK3. Unfortunately,
all MtLYK3 mutated variants that retained its autophosphorylation activity displayed
the ability to induce CD in Nicotiana leaf in the presence of MtNFP, whereas tested
chemicals either had no effect or only delayed the development of CD (e.g. lanthanum
chloride) (see Chapter 2 and 5). Instead, we took advantage of the fact that CD
induction in Nicotiana requires the kinase activity of MtLYK3 (see Chapter 2),
and we investigated the heteromerization between MtNFP and kinase inactive MtLYK3
protein fusions. Assuming dimerization as the underlying mechanism, we estimated that
in leaf samples with approximately 1:1 ratio of MtNFP-sYFP2 and MtLYK3 [G334E]-
mCherry protein fusions, at least 7% protein formed heteromers. While this value
may be small as compared to the extent of estimated homomerization
of these proteins, its value being larger than 0 demonstrates that MtNFP and kinase-
inactive MtLYK3 are to a certain (limited) extent within the same molecular complex
at a distance less than 10 nm. This limited fraction may be the biological relevant
fraction of specifically interacting molecules or it may be caused by residual bystander
FRET (despite the fact that care was taken to exclude such samples from analysis)
considering the low detected lifetime reduction (of only 3% being very close
to the detection limit of the technique). Bystander FRET should especially
be considered in a situation when donor- and acceptor-tagged proteins are concentrated
103
to the same subcellular localization, such as a 2-dimensional membrane plane.
Therefore, the occurrence of bystander FRET should be taken into consideration,
especially with techniques that lock the transiently interacting proteins
in persistent complexes, such as BiFluorescence Complementation (BiFC) assay.
However, other explanations for the relatively low heteromerization tendency of MtNFP
and MtLYK3 [G334E] proteins are as likely.
First of all, in a situation of pronounced homomerization of MtLYK3 [G334E]
and MtNFP, there will be a significant dilution of FRET signal from the smaller fraction
of heteromerized donor-tagged protein. Our finding of less pronounced
heteromerization between MtNFP and MtLYK3 [G334E] at the PM of Nicotiana leaf
epidermal cells agrees with the findings of Madsen et al. (2011) on the oligomerization
potential of LjNFR5 and LjNFR1 that are postulated to form receptor complex in order
to transduce NF signal (Radutoiu et al., 2003, 2007; Miwa et al., 2006b; Høgslund
et al., 2009; Nakagawa et al., 2010). Using BiFC in Nicotiana transient expression
system, LjNFR1 and LjNFR5 fusions were demonstrated to form homomers, whereas
co-production of LjNFR1 and LjNFR5 proteins fused to different parts of the split YFP
did not result in complementation of YFP fluorescence (reporting on the lack of
heteromerization between these LysM-RLKs). Both our relatively low heteromerization
potential of MtNFP and kinase-inactive MtLYK3, and the negative result of Madsen
et al. (2011) could be explained by the pronounced homomerization of the proteins
(in the latter case causing sequestration of the protein in homomers with relatively little
split FP fusions left to complement the YFP fluorescence at a level sufficient enough
for detection). Another method capable of differentiating between the homomers
and heteromers would be better suited for the analysis of such complex samples.
104
Alternatively, it cannot be excluded that the MtLYK3/LjNFR1 kinase activity
is required for interaction with MtNFP/LjNFR5, similarly to the phosphorylation-
dependent heteromerization reported for AtBRI1 and AtBAK1 (Li et al., 2002; Wang
et al., 2008; Schulze et al., 2010; Schwessinger et al., 2011). In addition, as both
MtLYK3 and LjNFR1 have been shown to interact with PM-associated remorin
proteins (Lefebvre et al., 2010; Tóth et al., 2010), it is possible that these proteins
engage in the formation of multiprotein complexes including their presumed interacting
partner, MtNFP/LjNFR5. In such a putative multiprotein complex, MtNFP/LjNFR5
and MtLYK3/LjNFR1 might interact via discrete surface regions, and the C-terminally
fused sYFP2 and mCherry molecules might be separated or forced in an unfavorable
orientation for FRET by other proteins present in the complex.
In conclusion, our results indicate that MtLYK3 forms homomers at the PM
of a plant cell, which is in agreement with the oligomerization status of other plant
RLKs. Analogously, homomerization of MtLYK3 might be required for activating
of its KD via trans-phosphorylation. On the other hand, heteromerization of MtLYK3
and MtNFP might require kinase activity of the former protein or might occur
in a multiprotein complex.
105
EXPERIMENTAL PROCEDURES
Plant transformations
pBin+ CaMV 35Sp::MtNFP-sYFP2 or -mCherry, pCambia1390 CaMV 35Sp::MtLYK3-sYFP2
or -mCherry, and pCambia1390 CaMV 35Sp::MtLYK3 [G334E]-sYFP2 or -mCherry constructs are described
in Chapter 2. Transformation of Agrobacterium tumefaciens strain GV3101::pMP90 and Agrobacterium
infiltration of N. benthamiana leaf is described in Chapter 2. Agrobacterium transformants carrying
the respective construct were resuspended in the infiltration medium to desired OD600: MtNFP constructs -
OD600= between 0.1 and 0.3; WT MtLYK3 and MtLYK3 [G334E] constructs- OD600=between 0.5 and 0.7.
Subsequently, Agrobacterium transformants carrying a desired construct fused to sYFP2 were mixed 1:1 with
Agrobacterium transformants carrying an empty pCambia1390 vector (for the reference sample) or a desired
construct fused to mCherry (for the FRET sample) before being infiltrated into Nicotiana leaf. FRET-FLIM
analyses in leaf samples (co-)expressing MtNFP-sYFP2 and MtNFP–mCherry constructs or MtLYK3
[G334E]-sYFP2 and MtNFP–mCherry constructs were performed not sooner than 36hai; in those
(co-)expressing WT MtLYK3 or MtLYK3 [G334E] constructs - between 24-36hai.
FRET-FLIM imaging and data analysis
The images of distribution were obtained using a laser 514 nm excitation and 542/27
and 525LP nm emission filters. The images of ID were obtained using Argon 500/20 nm excitation,
and 593/40 nm emission filters. During each FRET-FLIM experiment, at least 10 independent
images were acquired, and each FRET experiment was repeated at least twice (generally three repetitions
were performed). During each FRET experiment, exposition times for the reference and FRET samples
were kept constant or similar, and generally were 100-200 ms per a image. Measurements of leaf tissue
autofluorescence (in the presence or absence of CD) required longer exposition times, approximately 300 ms
per a image. Post-acquisition, an intensity threshold was applied to each region of interest in order to
calculate 4 individual values: , M , ID, and IA. The difference in the excitation and sensitivity of sYFP2
and mCherry detection of the FRET-FLIM microscopy set-up was calibrated using 2µM purified sYFP2
and 2µM purified mCherry proteins in 20mM Tris-HCl (pH 8, 1mM EDTA) buffer. From the intensity images
of purified sYFP2 protein or Tris-HCl buffer we calculated that the leak-through of the YFP emission
into mCherry detection channel (using Argon lamp and 500/20 nm excitation and 593/40 nm emission filters)
accounted to approx. 6.5%. Therefore, IA corrected = IA apparent -0,065*ID apparent. From the intensity
106
images of 1:1 molar mixture of purified sYFP2 and mCherry we calculated that the ID apparent normalized
to the IA corrected equaled 0.5. Therefore, ID corrected = 2*ID apparent.
107
CHAPTER 4
Protein kinase domain is indispensible for biological activity
of Medicago truncatula NFP LysM receptor-like kinase.
Anna Pietraszewska-Bogiel1, Maria A. Koini
1, and Theodorus W.J. Gadella
1
1 Section of Molecular Cytology, Swammerdam Institute for Life Sciences, University of Amsterdam, Science
Park 904, 1098 XH Amsterdam, The Netherlands
2 INRA, Laboratoire des Interactions Plantes-Microorganismes (LIPM), UMR441, F-31326 Castanet-Tolosan,
France
3 CNRS, Laboratoire des Interactions Plantes-Microorganismes (LIPM), UMR2594, F-31326 Castanet-
Tolosan, France
This work was supported by the EC Marie Curie Research Training Network Programme through contract
MRTN-CT-2006-035546 “NODPERCEPTION”, by the French National Research Agency (ANR) through
contract NodBindsLysM. We thank Dr Dörte Klaus-Heisen (LIPM, CNRS-INRA, Toulouse, France)
for providing pBin+ 35S::MtNFP-3xFLAG constructs carrying some of the point mutations in study.
108
SUMMARY
In successful symbiosis between legume plants and nitrogen-fixing rhizobia,
the bacteria are accommodated inside de novo formed plant organs (nodules),
where they reduce dinitrogen for the plant’s use in exchange for carbohydrates.
Perception of bacterial lipo-chitooligosaccharidic signals, termed Nodulation (Nod)
Factors (NFs) is generally indispensable for establishing this interaction. However, little
is still known about the activation and signaling mechanisms of the putative NF
receptors. Our previous results (presented in Chapter 2) indicated a functional
interaction of two Medicago truncatula (Medicago) putative NF receptors, MtNFP
and MtLYK3, resulting in cell death (CD) induction upon their simultaneous production
in Nicotiana benthamiana (Nicotiana) leaf. Here, we present a detailed structure-
function study on MtNFP, using the ability of MtNFP to induce CD in the presence
of MtLYK3 as readout of its biological activity. The MtNFP intracellular region (InR)
seemed to be relatively tolerant to changes in the sequence. Nevertheless, substitution
of the conserved Lys (in a β-strand 3) and Gly (in an α-helix F), possibly required
for the correct tertiary structure of the kinase domain (KD), abolished MtNFP biological
activity. Dissection of the MtNFP InR demonstrated that the KD itself was necessary
for MtNFP signaling in Nicotiana, whereas the flanking sequences, i.e. a C-tail
and a juxtamembrane region were dispensable or not sufficient for CD induction,
respectively. Our results imply similar requirements for MtNFP biological activity
in Medicago and Nicotiana with respect to the structure of the MtNFP InR.
109
INTRODUCTION
Perception of NFs is generally indispensable for nodule organogenesis
and regulation of the infection process (Giles & Oldroyd 2008; Madsen et al., 2010).
Up to now, several plasma membrane (PM)-spanning receptor-like kinases (RLKs)
that possess three Lysin Motif (LysM) domains in their extracellular region (ExR),
and a Ser/Thr protein KD in their InR are postulated to function as putative NF
receptors. Among those, Nod Factor Perception (MtNFP) in Medicago and Nod Factor
Receptor 1 and 5 (LjNFR1 and LjNFR5) in Lotus japonicus (Lotus) are indispensable
for triggering of early symbiotic signaling upon perception of compatible NF,
and for the host root infection by rhizobia via root hairs (RHs) (Ben Amor et al., 2003;
Madsen et al., 2003; Radutoiu et al., 2003; El Yahyaoui et al., 2004; Mitra et al., 2004a;
Arrighi et al., 2006; Miwa et al., 2006b; Høgslund et al., 2009; Madsen et al., 2010;
Nakagawa et al., 2010; Bensmihen et al., 2011). In contrast, Medicago LysM domain-
containing RLK/Root Hair Curling (MtLYK3/HCL; from now on referred to
as MtLYK3) is specifically required at the infection step, i.e. an entrapment of rhizobia
in RH curls, and formation of specialized infection structures, termed infection threads
[ITs], through which the bacteria colonize the nodule primordium (Catoira et al., 2001;
Limpens et al., 2003; Smit et al., 2007). In addition, detection of MtNFP and MtLYK3
transcripts/ promoter activity in a nodule primordium and the invasion zone of a mature
nodule points out to a yet-unidentified role for these genes in nodule development
and/or accommodation of rhizobia inside nodule cells (Limpens et al.,2005; Arrighi
et al., 2006; Mbengue et al., 2010; Haney et al., 2011).
RLKs function as PM-localized protein kinases and signal transducers. Direct
or indirect perception of a cognate extracellular signal induces specific conformational
110
changes in the ligand-bound RLK (often accompanied by a modulation of its
oligomerization state) that, in turn, leads to enhanced activity of its KD and downstream
signaling (Wang et al., 2008; Lemmon & Schlessinger 2010). All protein kinases share
a conserved KD characterized by a distinctive tertiary fold composed of an N-terminal
and a C-terminal lobe. In the past 20 years, crystal structures of many eukaryotic protein
kinases (ePKs) have been solved, greatly contributing to our understanding of the exact
role of various conserved residues for either catalysis or maintenance of the KD
structure (Kornev & Taylor 2010; Klaus-Heisen et al., 2011; Taylor & Kornev 2011;
Yang et al., 2012)1. In short, the N-lobe contains three distinctive motifs/structural
elements crucial for kinase activity: a Gly-rich loop, an Ala-x-Lys (AxK, where x
is any amino acid) motif, and an α-helix C. Within the C-lobe, a catalytic loop
and an activation segment (AS) are crucial for catalysis and substrate binding,
and an α-helix F serves as an organizing element for the entire KD. In most ePKs,
ATP binds in the deep cleft between the two lobes through contacts with: amino acids
in the Gly-rich loop (G193 through to G198 in IRAK4) and the immediately proceeding
Val (V200 in IRAK4), the Lys (K213 in IRAK4) in the AxK motif; the conserved Lys
(K313 in IRAK4) in the catalytic loop; and the Asp (D329 in IRAK4) in an Asp-Phe-
Gly (DFG) motif at the start of the AS. Activation of ePKs typically involves
translocation of the α-helix C and of the activation segment resulting in a specific
1 In Figure S1 we present a tertiary structure of a human (Homo sapiens) Interleukin Receptor-Associated
Kinase 4 (HsIRAK-4) reported by Wang et al. (2006). HsIRAK-4 was chosen because it was used to generate
a homology model of the MtLYK3 KD (see Klaus-Heisen et al., 2011). The selected conserved residues
discussed in this Chapter are highlighted in the depicted structure of HsIRAK-4, in order to orient the reader
in their positions in a 3D structure of a KD.
111
positioning of the two lobes optimally for catalysis (Huse & Kuriyan 2002; Nolen et al.,
2004; Shi et al., 2006; Zhang et al., 2012). The recently postulated dynamic formation/
disruption of two non-consecutive hydrophobic structures, termed regulatory (R)
and catalytic (C) spines (Kornev & Taylor 2010; Taylor & Kornev 2011; see Fig. S1
for a detailed description), explains how the two previously established regulatory
elements, the α-helix C and the AS, govern the activity of many ePKs.
EPKs function as crucial regulatory components in a multitude of cellular
processes where they provide reversible phosphorylation of their protein substrates,
resulting in a rapid and specific modulation of protein catalytic activity, oligomerization
status, and/or subcellular localization (e.g. Wang et al., 2008; Lew et al., 2009; Jaillais
et al., 2011; Oh et al., 2012). In addition, the sequenced genomes of model organisms
(Boudeau et al., 2006; Castels & Casacuberta 2007) reveal a significant complement
of genes encoding atypical ePKs lacking one or more key catalytic residues.
Remarkably, some of these ePKs display phosphorylation activity, indicating that
the sheer absence of certain conserved residues does not preclude kinase activity
(Zegiraj & Aalten 2010). Other atypical ePKs are truly kinase-inactive and are classified
as pseudokinases. Interestingly, pseudokinases have recently been implicated
in the allosteric regulation of true kinases with which they interact (Zegiraj & Aalten
2010). Alternatively, they function as molecular scaffolds, facilitating the formation
of multiprotein complexes (Zegiraj & Aalten 2010).
In contrast to MtLYK3 and LjNFR1 that possess an active KD capable of auto-
and trans-phosphorylation (Arrighi et al., 2006; Mbengue et al., 2010; Klaus-Heisen
et al., 2011; Madsen et al., 2011), MtNFP and LjNFR5 possess an atypical KD
in which certain conserved motifs/substructures are either lacking or are significantly
modified (Madsen et al., 2003; Arrighi et al., 2006). For instance, the Gly-rich loop
112
is absent except for the first Gly (G323 in MtNFP) and the AS is abnormally short
and contains an Asn (N453 in MtNFP) in place of the conserved Asp in the DFG motif.
In addition, the conserved Lys (K339 in MtNFP) in the AxK motif and the conserved
Asp (D435 in MtNFP) in the catalytic loop are dispensable for MtNFP function
in nodulation (Lefebvre et al., 2012). Therefore, it seems that MtNFP and LjNFR5
do not display nor rely on intrinsic kinase activity to signal. Analogously to the function
of several known pseudokinases, a regulatory role of MtNFP over another kinase-active
RLK presents an intriguing possibility. However, such interaction of MtNFP with
MtLYK3 or any other RLK implicated in the RL symbiosis in Medicago (Endre et al.,
2002; Limpens et al., 2003; Arrighi et al., 2006) remains to be shown. Similarly,
LjNFR5 is hypothesized to interact and activate (via an unspecified mechanism)
LjNFR1 (Radutoiu et al., 2003, 2007; Madsen et al., 2011), although a convincing proof
for such mechanism is still lacking.
We recently revealed a functional interaction between MtNFP and MtLYK3
in a heterologous system of Nicotiana leaf (see Chapter 2), supporting their postulated
co-functioning during nodulation. Since the Nicotiana system provided rapid and robust
readout of the functional interaction between these proteins, we decided to further
characterize the similarity between MtNFP-mediated signaling in Medicago
and Nicotiana. Here, we present a detailed structure-function study using various
truncated and mutated variants of MtNFP. We demonstrate that the requirements
of nodulation and CD induction show significant similarities with respect to the MtNFP
InR sequence, supporting the relevance of the Nicotiana system for structure-function
studies on this and potentially other LysM-RLKs. In addition, our results indicate
that MtNFP signaling role is likely dependent on the conserved fold of its KD.
113
RESULTS
Lys 339 and Gly 474 are essential for MtNFP biological activity in Nicotiana
In our structure-function study on MtNFP, we first focused on the residues
characterized in the recent work of Lefebvre et al. (2012) (listed in Table 1).
These included the conserved Lys 339, Asp 435, and Gly 447 (located in the α-helix F),
as well as phosphorylation sites predicted in the MtNFP InR by NetPhos programme:
Thr 281, Ser 282 and Ser 283 in a juxtamembrane (JM) region (an intracellular
sequence that flanks the N-terminus of the KD); Thr 459, Ser 460 and Thr 461
in the AS; and Thr 578 and Ser 579 in a C-tail (an intracellular sequence that flanks
the C-terminus of the KD) (see Fig. 1 for their position in the MtNFP InR). All MtNFP
constructs carrying (a) mutation(s) were generated as C-terminal fusions to the sequence
encoding super yellow fluorescent protein (FP) 2 (sYFP2) (Kremers et al., 2006).
We used a constitutive 35S promoter of the cauliflower mosaic virus (CaMV) to drive
the expression of all constructs in study. In order to test biological activity of MtNFP-
sYFP2 mutated variants, they were produced alone or co-produced with MtLYK3-
mCherry (a C-terminal fusion of MtLYK3 to a monomeric red FP [mCherry]) (Shaner
et al., 2004) in Nicotiana leaves via Agrobacterium-mediated transient transformation
(Agro TT). Concomitant mock infiltration (with Agrobacterium strain carrying
an empty vector) and control co-expression of MtNFP-sYFP2 and MtLYK3-mCherry
were performed on every leaf. Development of CD was monitored between
36 and 72 hours after infiltration (hai), and in case of the absence of or weakly
pronounced macroscopic symptoms, the absence/presence of CD was further scrutinized
with an exclusion dye (Evans blue) staining.
114
T281 S282 S283 G323
MtNFP: 270 YCLKMKRLNRSTSSSETADKLLSGVSGYVSKPTMYEIDAIMEGTTNLSDN------CKIG
PsSYM10:269 YCLKMKRLNRSTSLAETADKLLSGVSGYVSKPTMYEMDAIMEATMNLSEN------CKIG
LjNFR5: 270 YCRRKKALNRTASSAETADKLLSGVSGYVSKPNVYEIDEIMEATKDFSDE------CKVG
MtLYK3: 310 FTYQELAKATNNFSLD------NKIG
HsIRAK4:168 MPFCDKDR---------TLMTPVQNLEQSYMPPD FSFYELKNVTNNFDERPISVGGNKMG
** * *** * **************** * * *V326 A337 K339 K340 E348 L352 L363
MtNFP: 324 ES----VYKANIDGRVLAVKKIKKD-----------ASEELKILQKVNHGNLVKLMGVSS
PsSYM10:323 ES----VYKANIDGRVLAVKKIKKD-----------ASEELKILQKVNHGNLVKLMGVSS
LjNFR5: 324 ES----VYKANIEGRVVAVKKIKEGG----------ANEELKILQKVNHGNLVKLMGVSS
MtLYK3: 330 QGGFGAVYYAELRGEKTAIKKMDVQASS-------EFLCELKVLTHVHHLNLVRLIGYC-
HsIRAK4:194 EGGFGVVYKGYVNNTTVAVKKLAAMVDITTEELKQQFDQEIKVMAKCQHENLVELLGFSS
*** : : * **::. *:*:. * * . : .* *** * *
Gly-rich loop α-helix C
MtNFP: 369 DNDGNCFLVYEYAENGSLEEWLFSESSKTSNSVVSLTWSQRITIAMDVAIGLQYMHEHTY
PsSYM10:368 DNDGNCFLVYEYAENGSLDEWLFSESSKTSNSVVSLTWSQRITVAVDVAVGLQYMHEHTY
LjNFR5: 370 GYDGNCFLVYEYAENGSLAEWLFSKSSGTPN---SLTWSQRISIAVDVAVGLQYMHEHTY
MtLYK3: 382 -VEGSLFLVYEHIDNGNLGQYLHGIG------TEPLPWSSRVQIALDSARGLEYIHEHTV
HsIRAK4:254 -DGDDLCLVYVYMPNGSLLDRLSCLDG-----TPPLSWHMRCKIAQGAANGINFLHEN--
***:: :** * * . ..* * * : * *.* : *
H433 D435 L442 N543 F454 T459 S460 T461
MtNFP: 429 PRIIHRDITTSNILLGSNFKAKIANFGMAR--------------TSTNSMM---------
PsSYM10:428 PRIIHRDITTSNILLDSNFKAKIANFSMAR--------------TSTNSMM---------
LjNFR5: 427 PRIIHRDITTSNILLDSNFKAKIANFAMAR--------------TSTNPMM---------
MtLYK3: 435 PVYIHRDVKSANILIDKNLRGKVADFGLTKL-IEVGNSTLHTRLVGTFGYMPPEYAQYGD
HsIRAK4:306 -HHIHRDIKSANILLDEAFTAKISDFGLARASEKFAQTVMTSRIVGTTAYMAPEALR-GE
***** :**** * *::::*: :.:: * *
catalytic loop activation segment
MtNFP: 466 --PKIDVFAFGVVLIELLTGKKAMTTKENGEVVILWKDFWKIFDLEGNREE---RLRKWM
PsSYM10:465 --PKIDVFAFGVVLIELLTGKKAITTMENGEVVILWKDFWKIFDLEGNREE---SLRKWM
LjNFR5: 464 --PKIDVFAFGVLLIELLTGRKAMTTKENGEVVMLWKDMWEIFDIEENREE---RIRKWM
MtLYK3: 494 VSPKIDVYAFGVVLYELITAKNAVLKT--GESVAESKGLVQLFEEALHRMDPLEGLRKLV
HsIRAK4:364 ITPKSDIYSFGVVLLEIITGLPAVDEHREPQLLLDIKEEIEDEEK---------TIEDYI
** *:::***** *::*. *: : : *
α-helix F T578 S579
MtNFP: 521 DPKLESFYPIDNALSLASLAVNCTADKSLSRPTIAEIVLCLSLLNQPSSEPMLERSLTS-
PsSYM10:520 DPKLENFYPIDNALSLASLAVNCTADKSLSRPSIAEIVLCLSLLNQSSSEPMLERSLTS-
LjNFR5: 519 DPNLESFYHIDNALSLASLAVNCTADKSLSRPSMAEIVLSLSFLTQQSSNPTLERSLTSS
MtLYK3: 552 DPRLKENYPIDSVLKMAQLGRACTRDNPLLRPSMRSIVVALMTLS
HsIRAK4:415 DKKMNDADSTS-VEAMYSVASQCLHEKKNKRPDIKKVQQLLQEMT
* *.::. . . : .:. * ** :: * : .:
MtNFP: 580 GLDAEA-THVVTSVIAR
PsSYM10:579 GLDVEA-THVVTSIVAR
LjNFR5: 579 GLDVEDDAHIVTSITAR
*** * * *** **
Figure. 1 Alignment of the InR sequences of MtNFP and its orthologs from pea (Pisum sativum),
PsSYM10, and Lotus japonicus, LjNFR5.
The KD sequences of MtLYK3 and human (Homo sapiens) IRAK4 are included in order to compare MtNFP
with sequences of active ePKs. Conserved residues are indicated in the underlying consensus sequence as (*).
The predicted start and end of the core KDs are indicated with black vertical lines. Motifs and structural
features of interest are boxed in black and named underneath. MtNFP residues analyzed for their role in CD
induction are highlighted in blue, and their positions are given above. Conserved residues with a putative role
in the assembly of “active” conformation of the MtNFP KD are highlighted in grey boxes, and their positions
are given above. In vitro phosphorylation sites found in the LjNFR5 InR are highlighted in red.
115
Twenty four hai, a complex subcellular localization in Nicotiana leaf epidermal
cells was observed for all MtNFP-sYFP2 mutated protein fusions that resembled
the reported subcellular localization of the wild-type (WT) MtNFP fusion protein
(see Lefebvre et al., 2012 and Chapter 2). Approximately 36 to 48hai, clear
co-localization of all but one mutated MtNFP-sYFP2 protein fusions with the PM
marker (HVR-mCherry, see Chapter 2) was observed, and indicated their efficient
production and PM localization in Nicotiana leaf epidermal cells (Fig. 2–demonstrated
for MtNFP [K339A]-sYFP2, rest not shown). In case of MtNFP [G474E]-sYFP2,
pronounced localization of the protein fusion in the endoplasmic reticulum (ER)
(indicated with the arrows in Fig. 2) was still observed 48hai, although some cells
showed a more uniform pattern of fluorescence at the cell boundary, indicating that
some protein fusion had reached the PM. In addition, strong fluorescent puncta
(see the arrowhead in Fig. 2 bottom panel) were visible at the cell boundary of many
cells (sometimes in association with nuclei). Such fluorescent puncta were not observed
in leaf regions co-producing WT MtNFP-sYFP2 and HVR-mCherry protein fusions.
The CD induction assay demonstrated that the G474E substitution abolished MtNFP-
sYFP2 biological activity in Nicotiana leaf (Table 1). On the contrary, MtNFP
[D435A]-sYFP2 and MtNFP-sYFP2 mutated protein fusions carrying Ala substitutions
of the predicted phosphorylation sites showed WT-like CD induction when co-produced
with MtLYK3-mCherry in Nicotiana leaf (Table 1). Co-production of MtNFP [K339A]-
sYFP2 and MtLYK3-mCherry resulted in confluent CD of (nearly) the entire infiltrated
region in only 3 out of 11 infiltrations, whereas the remaining regions displayed only
a patch or spot of dead tissue (Table 1). This surprising observation led us to investigate
the role of additional residues that might be important for MtNFP biological activity
in Nicotiana. We chose to test the effect of Ala substitution of the conserved Gly 323,
116
HVR-mCherry
NFP [K339A]-
sYFP2
NFP [K339A
K340A]-sYFP2
NFP [G474E]-
sYFP2
NFP [G474E]-
sYFP2
channel: YFP mCherry merged & DIC
Figure 2. Subcellular localization of various MtNFP-sYFP2 mutated protein fusions.
HVR-mCherry, encoding the PM marker, was co-expressed with (from top to bottom): MtNFP [K339A]-
sYFP2; MtNFP [K339A K340A]-sYFP2; MtNFP [G474E]-sYFP2 in Nicotiana leaf epidermal cells
via Agro TT, and the fluorescence (viewed from abaxial side) was imaged 48hai using confocal laser scanning
microscopy. YFP channel presents green fluorescence of sYFP2; mCherry channel present orange
fluorescence of mCherry; merged channel superimposes green, orange, red (chlorophyll autofluorescence),
and differential interference contrast (DIC) image. Note co-localization of the PM marker with MtNFP
[K339A]-sYFP2 or MtNFP [K339A K340A]-sYFP2. Localization of MtNFP [G474E]-sYFP2 in ER
is indicated with arrows, and in puncta structures – with an arrowhead. Bars are 20 µm.
the Asn 453, and the Lys 340 immediately downstream from the AxK motif (the latter
mutation was introduced into MtNFP [K339A]-sYFP2 construct). Again, all three
protein fusions were efficiently produced and correctly localized to the PM
in Nicotianaleaf epidermal cells (Fig. 2–demonstrated for MtNFP [K339A K340A]-
117
sYFP2, rest not shown) but only MtNFP [G332A]-sYFP2 and MtNFP [N453A]-sYFP2
were able to induce efficient CD upon co-production with MtLYK3-mCherry (Table 1).
On the contrary, MtLYK3-mCherry and MtNFP [K339A K340A]-sYFP2 co-production
resulted in very limited death in 3 out of 11 infiltrated regions, and a confluent CD
of the entire infiltrated region was not observed (Table 1). In addition, no CD
was observed after separate expression of any of the MtNFP-sYFP2 constructs carrying
the above mutations (Table 1). Taken together, these results indicated that many
residues were apparently dispensable, whereas the Lys 339 and Gly 474 were essential
for MtNFP biological activity in Nicotiana.
The kinase domain but not the C-tail is necessary for MtNFP biological activity
in Nicotiana
Previously tested MtNFP truncated variant with almost the entire InR deleted,
i.e. MtNFP [ΔInR] (see Lefebvre et al., 2012 and Chapter 2), still retained part
of the MtNFP JM region, including three predicted phosphorylation sites (i.e. Thr 281,
Ser 282 and Ser 283). Recently, the InR of MtNFP ortholog from Lotus, LjNFR5,
was found to be phosphorylated in vitro by LjSYMRK (for Symbiotic Receptor Kinase)
of Lotus on four sites located in the JM region (Madsen et al., 2011): the Thr 280,
Ser 282, Ser 292 and Ser 295 (the last two residues were wrongly annotated by Madsen
et al. [2003, 2011] as being located within the LjNFR5 KD). The same conserved
(Ser 282, Ser 292, Ser 295) or corresponding (Thr 281) residues are present in MtNFP
(see Fig. 1) but the Ser 292 and Ser 295 are missing from the MtNFP [ΔInR]-sYFP2
protein fusion. In order to study whether a truncated MtNFP variant possessing
the entire JM region was capable of CD induction in Nicotiana, we generated another
truncated protein fusion, MtNFP [full JM]-sYFP2 (amino acids [aa’s]: 1-312; prediction
118
Table 1. Cell death induction upon (co-)expression of MtNFP mutated and truncated constructs
and WT MtLYK3 in Nicotiana leaves.
MtNFP-sYFP2 construct Nodulation *
Cell death induction
Co-expression with
MtLYK3-mCherry
Separate
expression
WT + 48/52 0/9
[G323A] NT 10/11 0/9
[K339A]
(AxK motif)
+
(impaired)
3/11
0/9
[K339A K340A] NT 1/11
0/9
[D435A]
(catalytic loop)
+ 11/11
0/9
[N453A] (NFG motif) NT 11/11 0/9
[G474E]
(α-helix F)
-
0/15
0/9
T281A/S282A/S283A (JM) + 12/12 0/9
T459A/S460A/T461A (AS) + 12/12 0/9
T578A/S579A (C-tail) + 11/12 0/9
full JM NT 0/9 0/9
ΔC-tail NT 13/13 0/9
119
* see Lefebvre et al. (2012); NT - not tested, Δ - deleted.
The designated constructs were expressed alone or co-expressed with MtLYK3-mCherry in Nicotiana leaves
via Agro TT, and the infiltrated regions were marked. Macroscopic symptoms of CD were scored 48hai:
only infiltrations that resulted in confluent death of (nearly) the entire infiltrated region were scored
as a fraction of total independent infiltrations performed. In case of no or weak macroscopic symptoms for a
particular (pair of) construct(s), three leaves were stained with Evans blue to confirm the lack of CD (data not
shown), and macroscopic symptoms in the remaining infiltrated regions were scored again 72hai, after which
point a weak unspecific chlorosis could be observed. For selected pairs of constructs, macroscopic symptoms
of CD 48hai and subsequent Evans blue staining are presented.
of the MtNFP KD was done before the identification of an α-helix B as a possible
integral part of a Ser/Thr KD; see Klaus-Heisen et al., [2011] and Fig. 1). MtNFP
[full JM]-sYFP2 was not able to induce CD either in the absence or presence
of MtLYK3-mCherry (Table 1), despite being efficiently produced and correctly
localized to the PM of Nicotiana leaf epidermal cells (data not shown). Since this result
suggested that MtNFP KD or the entire MtNFP InR was required for MtNFP biological
activity in Nicotiana, we tested CD induction ability of MtNFP after removal
of its C-tail (MtNFP [ΔCtail], aa’s: 1-568). MtNFP [ΔCtail]-sYFP2 was efficiently
produced and localized to the PM of Nicotiana leaf epidermal cells (data not shown),
and displayed WT-like ability to induce CD when co-produced with MtLYK3-mCherry
(Table 1). Taken together, we showed that MtNFP KD played a crucial role in CD
induction, whereas the JM region alone was not sufficient, and the C-tail
was dispensable for MtNFP biological activity in Nicotiana.
DISCUSSION
Similarities between MtNFP signaling in Medicago and Nicotiana
Nicotiana has proved to be a useful model for heterologous production
and structure-function studies on multiple proteins, providing invaluable insights
120
into their biological activity that guided their subsequent analyses in the respective
homologous systems (e.g. dissection of tomato (Solanum lycopersicum) Pto
[for Pseudomonas syringae pv. tomato]-mediated signaling; see Oh & Martin 2010).
In a similar fashion, Nicotiana was useful for efficient production of MtNFP
and MtLYK3, facilitating characterization of their subcellular localization,
and homo(oligo)merization in vivo (see Chapter 2 and 3). The use of Nicotiana
as a possible system to investigate signaling mediated by the symbiotic LysM-RLKs
present important practical advantages over the legume root system, especially in terms
of rapidity and ease of protein production. Therefore, to further investigate the relevance
of Nicotiana system in this respect, we compared the requirements of nodulation
in Medicago and CD induction in Nicotiana in respect to the structure of MtNFP InR.
Remarkably, we showed that out of eleven residues tested, nine (Thr 281, Ser 282,
Ser 283, Asp 435, Thr 459, Ser 460, Thr 461, T578, and S579) were equally
dispensable, and one (Gly 474) was equally crucial for MtNFP biological activity
in Nicotiana. In case of the K339A substitution, the observed impairment of biological
activity of the MtNFP [K339A] in Nicotiana agreed with the partial impairment of this
mutated variant in nodulation (B. Lefebvre, personal communication). This significant
overlap indicates that requirements of CD induction and nodulation in respect to
the structure of MtNFP IR are similar, supporting our notion of the relevance
of Nicotiana system for studies on this, and potentially other symbiotic LysM-RLKs.
MtNFP might not require trans-phosphorylation by another kinase to signal
The LjNFR5 IR was shown to be trans-phosphorylated in vitro by the LjNFR1
and LjSYMRK (for Symbiotic Receptor Kinase) IRs, although the importance of these
phosphorylation sites for LjNFR5 function in nodulation has not been shown (Madsen
121
et al., 2011). On the contrary, evidence is lacking for trans-phosphorylation
of the MtNFP InR by MtLYK3 (Klaus-Heisen et al., 2011), despite the fact that
the phosphorylation sites found in the LjNFR5 JM region (Thr 280, Ser 282, Ser 292,
and Ser 295) are conserved in the MtNFP sequence (respectively, Ser 280, Ser 282,
Ser 292, and Ser 295). The effect of S280A, S292A or S295A substitution on MtNFP
biological activity has not been analyzed, and therefore it is formally possible that these
residues might be important for MtNFP signaling. However, Ala substitution of other
predicted phosphorylation sites, i.e. T281A, S283A, T459A, S460A, T461A, T578A,
and S579A did not affect MtNFP biological activity (Lefebvre et al., 2012 and Table 1).
Therefore, transphosphorylation of these sites, if it occurs in vivo, does not seem
to be crucial for MtNFP signaling in either Medicago or Nicotiana.
The conserved fold of kinase domain might be required for MtNFP signaling
KDs found in divergent ePK subfamilies display not only a significant
conservation of certain residues/motifs but also a remarkable similarity of their tertiary
structures in an active state. More specifically, three hydrophobic elements (the α-helix
F, the R spine, and the C spine) are postulated to constitute an “internal frame”
for assuming the active conformation by ePKs (Kornev & Taylor 2010; Taylor &
Kornev 2011). Interestingly, the α-helix F is conserved in the MtNFP sequence,
and the G474E substitution abolishes MtNFP biological activity in vivo (Lefebvre et al.,
2012 and Table 1). Kornev et al. (2008) postulated that this conserved Gly
(G225 in protein kinase A [PKA], G374 in IRAK4) facilitates tight packing of α-helix F
and α-helix H that is required for a correct tertiary structure of the KD of PKA,
as well as of other ePKs. Incorrect folding of the KD is a likely explanation
for the negative effect of various substitutions of this conserved Gly on in vitro
122
and/or in vivo activity reported for MtNFP and several other RLKs (Clark et al., 1997;
Gomez-Gomez & Boller 2000). In addition, an Asp substitution of the conserved Val
immediately downstream from this Gly (Gly 228 and Val 229) in the sequence
of tomato Pto decreased accumulation of Pto [V229D] mutated variant in vivo (Dong
et al., 2009). Therefore, we hypothesize that the impairment of the KD tertiary structure
in MtNFP [G474E] mutated protein may affect its trafficking to and/or its stability
at the PM. Both hypotheses agree with the recently demonstrated sensitivity of MtNFP
to the ER quality control system in Medicago and Nicotiana, and the postulated
requirement of the PM localization for MtNFP function in nodulation (Lefebvre et al.,
2012). Analogously, the decreased PM occupancy by MtNFP [G474E] mutated protein
might contribute to the lack of its biological activity in Nicotiana. However, as shown
in Medicago (Lefebvre et al., 2012) and (with less certainty) in Nicotiana (Fig. 2),
at least a fraction of MtNFP [G474E]-sYFP2 protein fusion is localized to the PM,
suggesting that the G474E substitution has also a more direct effect on MtNFP
biological activity, possibly via affecting the tertiary structure of its KD.
In this respect, we note the conservation of residues implicated
in the orientation of the α-helix C (K339, E348, and F454 in MtNFP; see Fig. 1)
in an active conformation and assembly of the R- (L352, L363, H433, and F454
in MtNFP) and C- (V326, A337, and L442 in MtNFP) spines in the MtNFP sequence
(see Fig. 1). Therefore, we hypothesize that the assembly of the MtNFP KD in an active
conformation, possibly driven by an analogous “internal frame” and stabilized
by interactions between additional residues, might be required for MtNFP interaction
with, and possible regulation of another (membrane-bound) protein. Different biological
activities of various MtNFP truncated variants in the CD induction assay support
the notion that the KD itself, rather than the JM region or the C-tail, is required
123
for MtNFP-mediated signaling in Nicotiana. This might also be true for MtNFP
function in Medicago, as MtNFP mutated variants carrying substitutions in the JM
region (T281A, S282A, and S283A) or C-tail (T578A and S579A) were active
in nodulation (Lefebvre et al., 2012). Active-like conformation (and analogous
formation of R- and C-spines) has been reported for various pseudokinases, despite
significant alterations of their sequence. In fact, it is thought to underlie their function
as scaffold proteins or allosteric regulators (Kornev & Taylor 2009; Zegiraj & Aalten
2010). A similar active conformation of the MtNFP KD is not unlikely, although
a crystal structure of an ePK with a similarly altered sequence that would confirm
or falsify this notion is currently lacking.
MtNFP binding of ATP is not excluded
MtNFP KD possesses all residues required for a putative assembly
of the R- and C- spines. However, whereas formation of the R-spine in the MtNFP KD
might be readily possible (providing favorable positioning of the H433 and F454;
see Fig. S1 for a detailed explanation), assembly of the C-spine is somewhat more
speculative. It is postulated to be completed by the adenine ring of ATP (Kornev &
Taylor 2010), whereas one of the structural elements implicated in ATP binding
(i.e. the Gly-rich loop) is missing in the MtNFP KD. Nevertheless, recent demonstration
of ATP binding or even kinase activity for several ePKs previously classified
as pseudokinases (Xu et al., 2000; Mukherjee et al., 2008; Eswaran et al., 2009; Zegiraj
et al., 2009 and refs therein; Shi et al., 2010) has revealed the existence of alternative
ways of coordinating the ATP molecule. In the light of these findings, the observed
impairment of MtNFP biological activity due to the K339A and K339A/K340A
substitutions (B. Lefebvre, personal communication, Table 1) is highly interesting.
124
For instance, it might indicate that ATP binding, involving Lys 339, modulates
the tertiary structure of the MtNFP KD, as demonstrated for other pseudokinases
(Zegiraj & Aalten 2010). Alternatively, the conserved Lys 339, might be required
for positioning of the α-helix C (through the conserved Lys 339 - Glu 348 salt bridge),
and in result, for adopting an active conformation by the MtNFP KD. At the moment,
MtNFP and LjNFR5 are proposed to signal via allosteric regulation of their putative
protein interactors (Madsen et al., 2011; Lefebvre et al., 2012). Detailed analysis
of ATP binding to MtNFP or any of its orthologs has not yet been investigated,
and thereby cannot be excluded at this moment.
In conclusion, an extensive correlation between the effect of various
substitutions on MtNFP function in nodulation (Lefebvre et al., 2012) and MtNFP role
in CD induction (this study) implies that the heterologous system of Nicotiana leaf
constitutes a relevant and convenient system for high-throughput structure-function
studies on this, and potentially other symbiotic LysM-RLKs. As substitutions
of the residues essential for correct folding of a KD are expected to abolish activity
of both true ePKs and pseudokinases, we propose that the CD induction assay
in Nicotiana leaf can be used for further high-throughput mutagenesis studies
on the MtNFP KD.
125
EXPERIMENTAL PROCEDURES
Constructs for plant expression
pBin+ CaMV 35S::MtNFP-sYFP2 and pCambia1390 CAMV 35S::MtLYK3-mCherry constructs are described
in Chapter 2. Point mutations were introduced in the pMon999 CaMV 35S::MtNFP-sYFP2 using
the QuickChangeTM site-directed mutagenesis kit (Stratagene) as described, and the HindIII-SmaI fragment
(containing the CAMV 35S::MtNFP-sYFP2 sequence) was subsequently recloned into pBin+ vector.
All truncated constructs were generated by PCR amplification (see Table 2 for primer sequences) and cloned
into PbIN+ 35S::sYFP2 vector. All pBin+ constructs were sequenced to verify the correct insert sequence.
Plant transformations
Transformation of Agrobacterium tumefaciens strain GV3101::pMP90 and Agrobacterium infiltration
of N. benthamiana leaf is described in Chapter 2. Agrobacterium transformants carrying the respective
constructs were resuspended in the infiltration medium to desired OD600: WT MtNFP, all MtNFP truncated
and mutated constructs (except for MtNFP [G474E]-sYFP2- OD600=0,5; MtNFP [G447E]-sYFP2-OD600=1;
MtLYK3-mCherry-OD600=0,7. Subsequently, they were mixed 1:1 with Agrobacterium transformants
carrying an empty pCambia1390 vector (for separate expression) or MtLYK3-mCherry construct before being
infiltrated into Nicotiana leaf. All experiments included mock control (GV3101::pMP90 transformants
carrying empty pCambia1390 vector) and a positive control (co-expression of full length MtNFP-sYFP2
and MtLYK3-mCherry constructs). All CD assays were performed at least three times, every time using three
different plants. Macroscopic observations were carried out between 24 and 72hai, and the results
were collated. In case of lack or inefficient CD induction by the (co-)expression of (a) certain construct(s),
three leaves were stained with Evans blue (as described in Chapter 2) 48hai to check for the presence
of infrequent CD.
Microscopic analysis
Was carried as described in Klaus-Heisen et al. (2011).
Table 2. Primer and linker sequences.
Name Type Sequence
MtNFP fw Cloning (NheI) GGGGCTAGCATGTCTGCCTTCTTTCTTC
MtNFP full JM rev Cloning (EcoRI) GGGAATTCACCTTCCATGATTGCATCAA
MtNFP ΔCtail rev Cloning (EcoRI) GGGAATTCTGATGGTTGGTTGAGAAGAG
Linker to FP GAATTC for all the constructs
Restriction sites are underlined.
126
Figure S1. Tertiary structure of human IRAK-4 kinase domain.
All protein kinases share a conserved KD characterized by a distinctive tertiary fold composed
of an N-terminal and a C-terminal lobe. The α-helix F (in red) serves as an organizing element
for the entire KD, i.e. provides a firm anchor for various structural elements located
in the N- and C-lobe. The orientation of two regulatory elements, the α-helix C and the activation segment,
governs the activity of many ePKs. Active conformation of the α-helix C is achieved mainly via two contacts
made by its conserved Glu (E91 in PKA, E232 in IRAK-4) residue. One contact is made to the conserved Lys
(K72 in PKA, K213 in IRAK-4) in the AxK motif (note the proximity of these two residues – highlighted
in purple), and the other to the conserved Phe (F185 in PKA, F330 in IRAK-4) in the DFG motif. Orientation
of the AS (green and lime) of many ePKs requires phosphorylation on a specific residue located
in an activation loop within. Upon phosphorylation, this so-called primary phosphorylation site connects
with two conserved basic residues: one in the catalytic loop (Arg165 in PKA, R310 in IRAK-4) and the other
downstream from the DFG motif (K189 in PKA, R334 in IRAK-4).
In addition, the dynamic assembly/disassembly of two non-consecutive hydrophobic spines is postulated
to regulate the activation of an ePK. The regulatory spine comprises of the following conserved residues
(in blue): Leu (L95 in PKA, Met 237 in IRAK-4) in the α-helix C, Leu (L106 in PKA, L248 in IRAK-4)
in a β-strand 4 (immediately proceeding the α-helix C), Phe (F185 in PKA, F330 in IRAK-4) in the DFG
motif, and a residue at the position -2 from the conserved Asp in the catalytic loop (Y164 in PKA, His 309
in IRAK-4). The catalytic spine comprises of the following conserved residues (in orange): Val (V57 in PKA,
V200 in IRAK-4) after the Gly-rich loop, Ala (A70 in PKA, A211 in IRAK-4) in the AxK motif,
127
and Leu (L173 in PKA, L318 in IRAK-4) in the catalytic loop, and is completed by the adenine ring of ATP.
The orientation of the α-helix C (via positioning of the L95/M237, the L106/L248 or the AxK motif)
and the activation segment (via positioning of the catalytic loop and the DFG motif) governs the dynamic
assembly/disassembly of both spines.
The structural elements that are either missing or highly modified in the MtNFP are highlighted:
the Gly-rich loop (in cyan), and the AS (in green and lime).
Note for a putative assembly of the R spine in MtNFP: in kinase-active ePKs, the favorable orientation
of the Phe 185/330 (F454 in MtNFP) and Tyr 164/His 309 (H433 in MtNFP) is dependent on the orientation
of the activation segment. Due to the lack of a crystal structure of an ePK with a similarly altered sequence
of the AS, homology modeling with respect to these residues and to the exact position of this element
in the MtNFP KD is currently not possible.
128
129
CHAPTER 5
Structure-function analysis of Medicago truncatula LYK3 LysM
receptor-like kinase in Nicotiana benthamiana.
Anna Pietraszewska-Bogiel1, Maria A. Koini
1, Linda A.C. Joosen
1, Dörte Klaus-
Heisen2,3
, Julie V. Cullimore2,3
and Theodorus W.J. Gadella1
1 Section of Molecular Cytology, Swammerdam Institute for Life Sciences, University of Amsterdam, Science
Park 904, 1098 XH Amsterdam, The Netherlands
2 INRA, Laboratoire des Interactions Plantes-Microorganismes (LIPM), UMR441, F-31326 Castanet-Tolosan,
France
3 CNRS, Laboratoire des Interactions Plantes-Microorganismes (LIPM), UMR2594, F-31326 Castanet-
Tolosan, France
This work was supported by the EC Marie Curie Research Training Network Programme through contract
MRTN-CT-2006-035546 “NODPERCEPTION”, and by the French National Research Agency (ANR)
through contract NodBindsLysM.
130
SUMMARY
During symbiosis between legume plants and rhizobia, perception of bacterial
lipo-chitooligosaccharidic signals, termed Nodulation (Nod) Factors (NFs), is generally
required for the formation of a novel plant organ (nodule) and host root infection
by rhizobia. Two putative NF receptors of Medicago truncatula (Medicago), MtNFP
and MtLYK3, are postulated to co-function during the latter process. In the previous
study (see Chapter 2) we demonstrated a functional interaction of MtNFP and MtLYK3
in Nicotiana benthamiana (Nicotiana) leaf, resulting in defence(-like) response
and cell death (CD) induction. Remarkably, symbiotic signaling in Medicago and CD
induction in Nicotiana displayed common requirements with respect to the intracellular
region (InR) of MtNFP and activity of MtLYK3 kinase (see Chapter 2 and 4).
Here, we report a detailed comparison of MtLYK3 structural features required for
its biological activity in Medicago and Nicotiana. We show that five out of seven
symbiotic-important phosphorylation sites found in the MtLYK3 InR are essential for
CD induction, indicating similar requirements for MtLYK3-mediated signaling in both
plant systems. In addition, our results suggest that the MtLYK3 extracellular region
(ExR) might exert an inhibitory role over MtLYK3 kinase, and that the plasma
membrane (PM) localization of the MtLYK3 InR is crucial for its biological activity
in Nicotiana. Finally, we report on in vitro and in vivo activity of MtLYK3 variants
in which a part of the MtLYK3 kinase domain, specifically an activation segment (AS),
was altered.
131
INTRODUCTION
Legume plants can form symbiosis with nitrogen-fixing bacteria, collectively
termed rhizobia, in which the bacteria are accommodated inside the cells of de novo
formed nodules where they reduce dinitrogen into ammonia in exchange for plant’s
carbohydrates (Masson-Boivin et al., 2009; Murray 2011; Oldroyd et al., 2011).
In Medicago, two receptor-like kinases (RLKs) (see below) with three Lysin Motif
(LysM) domains in their ExR are implicated in NF perception at various steps during
the Rhizobium-legume (RL) interaction. Nod Factor Perception (MtNFP) is crucial
for early symbiotic signaling: triggering of a complex network of cellular
and molecular processes implicated in nodule organogenesis and preparation
of the host root for a subsequent infection by rhizobia (Ben Amor et al., 2003;
El Yahyaoui et al., 2004; Mitra et al., 2004a; Arrighi et al., 2006). In addition, MtNFP
is required for growth of specialized infection structures, termed infection threads
(ITs), through which the rhizobia colonize the nodule primordium (Arrighi et al., 2006;
Bensmihen et al., 2011). LysM domain-containing RLK/Root Hair Curling
(MtLYK3/HCL, from now on referred to as MtLYK3) is not essential for early
signaling but is indispensable for curling of root hairs that entrap the rhizobia
in infection foci, and for the IT growth (Wais et al., 2000; Catoira et al., 2001;
Limpens et al., 2003; El Yahyaoui et al., 2004; Middleton et al., 2007; Smit et al.,
2007) where it is postulated to co-function with MtNFP (Arrighi et al., 2006; Smit
et al., 2007; Bensmihen et al., 2011). Finally, detection of MtNFP and MtLYK3
transcripts/promoter activity in the nodule primordium and the invasion zone
of a mature nodule (Limpens et al., 2005; Arrighi et al., 2006; Mbengue et al., 2010;
Haney et al., 2011) suggests a role for the encoded LysM-RLKs in nodule
132
development and/or accommodation of rhizobia inside nodule cells. However, little
is still known about the activation mechanisms of these putative NF receptors.
Plant RLKs constitute a family of single-spanning PM proteins that possess
various motifs in their ExR, and a Ser/Thr protein kinase domain (KD) within their InR
(Shiu & Bleecker 2003; Gish & Clark 2011). Importantly, a monophyletic origin
of structurally-related animal receptor kinases (RKs) and plant RLKs (Shiu & Bleecker
2003 and refs therein) implies that they might employ similar regulatory and signaling
mechanisms. As demonstrated for several animal RKs, direct or indirect perception
of a cognate extracellular ligand induces specific conformational changes in the ligand-
bound receptor. This is often accompanied by a change in its oligomerization state:
formation of homomers (homocomplexes) and/or heterocomplexes with other, often
related RKs (Heldin 1995; Groppe et al., 2008; Lemmon & Schlessinger 2010).
Moreover, in several animal receptor Tyr kinases (RTKs) a specific juxtaposition
of the InRs within the receptor dimer is a prerequisite for the activation of their KDs
via an allosteric and/or phosphorylation mechanism (Lemmon & Schlessinger 2010).
During activation, a precise and sequential ordered transphosphorylation (Lew et al.,
2009 and refs therein) on multiple sites within the KD (especially in a so-called
activation loop [AL]), and/or in the flanking N-terminal (termed juxtamembrane [JM])
and C-terminal (termed C-tail) regions occurs (Huse & Kuriyan 2002; Furdui et al.,
2006 and for refs therein). As a result, the kinase inhibition (intra- or inter-molecular)
is removed, the kinase activity is increased, and specific phosphorylation-dependent
interaction motifs for binding downstream signaling components are generated to allow
stringent regulation of the RK activity (Hubbard 2004; Lemmon & Schlessinger 2010).
Signal transduction mechanisms of plant RLKs are only beginning to be
revealed but already show a significant analogy with those of animal RKs, especially
133
in terms of the role of oligomerization and transphosphorylation in regulating kinase
activity (e.g. Wang et al., 2005; Guo et al., 2010; Schulze et al., 2010 and refs therein;
Jaillais et al., 2011ab; Schwessinger et al., 2011; Sun et al., 2012 and refs therein).
For example, heteromerization of Arabidopsis thaliana (Arabidopsis) Brassinosteroid
Insensitive 1 (AtBRI1) and BRI1-Associated Kinase 1/Somatic Embryogenesis
Receptor Kinase 3 (AtBAK1/SERK3; from now on referred to as AtBAK1)
is postulated to result in sequential transphosphorylation of both proteins, which
ultimately increases their kinase activity and signaling output (Wang et al., 2008).
MtLYK3 possesses an active KD capable of auto- and transphosphorylation
in vitro (Arrighi et al., 2006; Mbengue et al., 2010; Klaus-Heisen et al., 2011).
Moreover, impairment of MtLYK3 kinase activity abolishes its function in vivo
(see Klaus-Heisen et al., 2011 and Chapter 2), indicating that this RLK relies
on phosphorylation activity to signal. We demonstrated that upon heterologous
production in Nicotiana leaf, MtLYK3 showed tendency to form homomers, most likely
dimers, at the PM of leaf epidermal cells (see Chapter 3). In addition, our results
indicated that MtLYK3 was able to functionally interact with MtNFP in this
heterologous system, triggering CD and defense(-like) response (see Chapter 2).
A homologous Arabidopsis LysM-RLK1/CERK1 (for Chitin Elicitor Receptor Kinase
1; from now on referred to as AtCERK1) was able to trigger a similar Nicotiana
response (see Chapter 2), indicating that MtLYK3 but not AtCERK1 specifically
required MtNFP to signal in this plant species. Importantly, nodulation in legumes
and CD induction in Nicotiana showed similar requirements with respect to the MtNFP
InR structure (see Chapter 2 and 4), implying the relevance of the Nicotiana system
for structure-function studies on this and potentially other LysM-RLKs. Conveniently,
a recent structure-function study identified several in vitro auto-phosphorylation sites
134
and conserved residues important for MtLYK3 kinase activity in vitro, and MtLYK3
function in nodulation (Klaus-Heisen et al., 2011). In addition, a limited alteration
within the AtCERK1 KD and replacement of its ExR for that of Lotus japonicus (Lotus)
Nod Factor Receptor 1 (LjNFR1; the Lotus putative MtLYK3 ortholog) allowed
AtCERK1 chimeric receptor to function in nodulation (Nakagawa et al., 2010).
Here, we employed the CD induction assay in Nicotiana leaf as a rapid and robust
readout of the functional interaction between MtNFP and MtLYK3 to compare
the requirements for MtLYK3 biological activity in Medicago and Nicotiana,
and to gain further insights into possible mechanisms regulating its kinase activity.
RESULTS
Autophosphorylation activity of MtLYK3 kinase is essential for cell death
induction in Nicotiana in the presence of MtNFP
We started our structure-function study on MtLYK3 by testing the residues
essential for its function in nodulation (Klaus-Heisen et al., 2011; see Fig. 1 for their
position in the MtLYK3 InR). All MtLYK3 constructs carrying the mutations listed
in Table 1 were generated as C-terminal fusions to the sequence encoding super yellow
fluorescent protein 2 (sYFP2; Kremers et al., 2006). The assessment of their ability
to induce CD in Nicotiana leaf was performed as described for the MtNFP mutated
variants (see Chapter 4), except that MtLYK3-sYFP2 mutated variants
were co-produced with MtNFP-mCherrry. Before assessing biological activity
of MtLYK3 mutated variants, their efficient production and PM localization
in Nicotiana leaf epidermal cells was confirmed using confocal laser scanning
microscopy and co-localization with a PM marker (HVR-mCherry; see Chapter 2).
135
First, we focused on residues required for in vitro autophosphorylation activity
of MtLYK3 kinase that included the conserved: Thr 319 at the end of an α-helix B;
Lys 349 in an Ala-x-Lys (AxK, where x is any amino acid [aa]) motif; Glu 362
in an α-helix C; (partially) Arg 440 and Asp 441 in a catalytic loop; and Lys 464
in the AS, as well as a so-called primary phosphorylation site (Thr 475) and an Arg 476
in the AL (Klaus-Heisen et al., 2011; see Table 1)2. MtLYK3-sYFP2 protein fusions
carrying T319A, K349A, E362A, R440A, D441A, K464A, T475A or R476A
substitution were all efficiently produced and localized to the PM in Nicotiana leaf
epidermal cells (Klaus-Heisen et al., 2011) but were incapable of CD induction upon
their co-production with MtNFP-mCherry (Table 1). In contrast, Ala substitution
of the Tyr 390 did not abolish MtLYK3-sYFP2 ability to induce CD upon
its co-production with MtNFP-mCherry (Table 1), in agreement withits retained in vitro
autophosphorylation activity (Klaus-Heisen et al., 2011; see Table 1). Finally, no CD
induction was observed after production of any of MtLYK3-sYFP2 mutated variants
in the absence of MtNFP-mCherry (Table 1). Therefore, our results indicated that
abrogation of in vitro autophosphorylation activity of the MtLYK3 KD abolished
MtLYK3 ability to induce CD in the presence of MtNFP.
Thr 433, Thr 472, and Thr 512 are required for cell death induction in Nicotiana
In addition to phosphorylation sites required for the kinase activation,
phosphorylation of various other residues can modulate catalytic or biological activity
2 for the predicted role of these residues for kinase activity, see Klaus-Heisen et al. (2011) and Chapter 4.
136
of eukaryotic protein kinases (ePKs) (see Discussion). Klaus-Heisen et al. (2011)
identified one phosphorylation site (Thr 300) in the JM region, and four sites (Thr 433,
Ser 471, Thr 472, and Thr 512) in the KD as important for MtLYK3 function
in nodulation but not for its autophosphorylation activity (see Table 1). As additional
Thr and Ser residues in the JM region were identified as in vitro autophosphorylation
sites, we tested the effect of the T300A substitution in a triply mutated MtLYK3 protein
fusion, i.e. MtLYK3 [T285A S286A T300A]-sYFP2. All five MtLYK3-sYFP2 mutated
protein fusions were efficiently produced and correctly localized to the PM of Nicotiana
leaf epidermal cells (Fig. 2–demonstrated for MtLYK3 [T433A]-sYFP2 and MtLYK3
[T512A]-sYFP2, rest not shown). However, only MtLYK3 [T285A S286A T300A]-
sYFP2 and MtLYK3 [S471A]-sYFP2 protein fusions were as active as wild-type (WT)
MtLYK3-sYFP2 for CD induction upon co-production with MtNFP-mCherry (Fig. 3A,
Table 1). In case of co-production of MtLYK3 [T433A]-sYFP2, MtLYK3 [T472A]-
sYFP2 or MtLYK3 [T512A]-sYFP2 with MtNFP-mCherry, a confluent death
of (nearly) the entire infiltrated region was observed, respectively, in 10 out of 29,
5 out of 11, and 8 out of 15 infiltrations, whereas the remaining regions displayed only
a patch or spot of dead tissue (Fig. 3A, Table 1). Therefore, Ala substitution of three
out of five (putative) phosphorylation sites required for MtLYK3 function in nodulation
impaired MtLYK3 biological activity in Nicotiana.
Additionally, the Thr 481 positioned in the LjNFR1 AS was identified
as in vitro autophosphorylation site, and was shown to be essential for LjNFR1 activity
in vitro and in vivo (Madsen et al., 2011). MtLYK3 also contains a Thr conserved
in this position (Thr 480), although it has not been identified as an autophosphorylation
site in vitro (Klaus-Heisen et al., 2011; see Fig. 1). As the MtLYK3 and LjNFR1 InRs
show 89% identity in aa sequence, we decided to analyze the effect of the T480A
137
(4)
T300
MtLYK3: 253 KYFQKKEEEKTKLP-QTSRAFSTQDASGSAEYETSGSSGHATGSAAGLTGIMVAKSTEFT
LjNFR1: 254 -RYQKKEEEKAKLPTDISMALSTQDASSSAEYETSGSSGPGTASATGLTSIMVAKSMEFS
AtCERK1:255 YAYRKNKSKGDSFS--SSIPLSTKADHASSTSLQSGGLG-GAGVSPGIAAISVDKSVEFS
SlPto: 30 -----------------------------------------------------------P
T319 G334 K349 E362
MtLYK3: 312 YQELAKATNNFSLDNKIGQGGFGAVYYAELR-GEKTAIKKMDVQAS---SEFLCELKVLT
LjNFR1: 313 YQELAKATNNFSLDNKIGQGGFGAVYYAELR-GKKTAIKKMDVQAS---TEFLCELKVLT
AtCERK1:312 LEELAKATDNFNLSFKIGQGGFGAVYYAELR-GEKAAIKKMDMEAS---KQFLAELKVLT
SlPto: 31 LVDLEEATNNFDHKFLIGHGVFGKVYKGVLRDGAKVALKRRTPESSQGIEEFETEIETLS
::* :**:**. . **:******* . ** * * *:** : :: :* *: :*:
a-helix B
MtLYK3: 368 HVHHLNLVRLIGYCVEG-SLFLVYEHIDNGNLGQYLHGIGTE--PLPWSSRVQIALDSAR
LjNFR1: 369 HVHHLNLVRLIGYCVEG-SLFLVYEHIDNGNLGQYLHGSGKE--PLPWSSRVQIALDAAR
AtCERK1:368 RVHHVNLVRLIGYCVEG-SLFLVYEYVENGNLGQHLHGSGRE--PLPWTKRVQIALDSAR
SlPto: 91 FCRHPHLVSLIGFCDERNEMILIYKYMENGNLKRHLYGSDLPTMSMSWEQRLEICIGAAR
: * :** ***** * .:.*:*::::**** .:* * . .:.* .*::*.:.:****
T433 R440 D441 K464 S471 T472 T475 R476 T480 Y483
MtLYK3: 425 GLEYIHEHTVPVYIHRDVKSANILIDKNLRGKVADFGLTKLIEVGNSTLHTRL-VGTFGY
LjNFR1: 426 GLEYIHEHTVPVYIHRDVKSANILIDKNLRGKVADFGLTKLIEVGNSTLQTRL-VGTFGY
AtCERK1:425 GLEYIHEHTVPVYVHRDIKSANILIDQKFRAKVADFGLTKLTEVGGSATRG-A-MGTFGY
SlPto: 151 GLHYLHTRAI---IHRDVKSINILLDENFVPKITDFGISKKGTELDQTHLSTVVKGTLGY
**:* * : ***: **:***:*::: *::****:* *****
catalytic loop T433 activation segment
MtLYK3: 484 MPPEYAQYGDVSPKIDVYAFGVVLYELITAKNAVLKTGESVAESKGLVQLFEEALHRMDP
LjNFR1: 485 MPPEYAQYGDISPKIDVYAFGVVLFELISAKNAVLKTGELVAESKGLVALFEEALNKSDP
AtCERK1:483 MAPE-TVYGEVSAKVDVYAFGVVLYELISAKGAVVKMTEAVGEFRGLVGVFEESFKETDK
SlPto: 208 IDPEYFIKGRLTEKSDVYSFGVVLFEVLCARSAIVQS-----LPREMVNLAEWAVESHNN
** * :: * ***:*****:*:: *:.*::: : :* : * :.. :
a-helix F
MtLYK3: 544 LEGLRKLVDPRLKENYPIDSVLKMAQLGRACTRDNPLLRPSMRSIVVALMTLSSPTEDCD
LjNFR1: 545 CDALRKLVDPRLGENYPIDSVLKIAQLGRACTRDNPLLRPSMRSLVVALMTLSSLTEDCD
AtCERK1:542 EEALRKIIDPRLGDSYPFDSVYKMAELGKACTQENAQLRPSMRYIVVALSTLFSSTGNWD
SlPto: 263 -GQLEQIVDPNLADKIRPESLRKFGDTAVKCLALSSEDRPSMGDVLWKLEYALR------
*.:::**.**:. :*: **.: . .. **** :: *
MtLYK3: 604 DDSSYENQSLINLLSTR
LjNFR1: 605 DESSYESQTLINLLSVR
AtCERK1:601 VG-NFQNEDLVSLMSGR
SlPto: 316 -----LQESVI------
Fig. 1 Alignment of the InR sequences of MtLYK3, LjNFR1 and AtCERK1, and of Pto KD.
Conserved residues are indicated as (*) in the underlying consensus sequence. Motifs and structural features
of interest are boxed in black and named underneath, except for the activation loops (shaded in grey).
The predicted core KDs start with the α-helix B, and the predicted end of core KDs is indicated with the black
vertical line. MtLYK3 residues analyzed for their role in CD induction are highlighted in red ([putative]
phosphorylation sites) and blue (other), and their positions are given above In vitro phosphorylation sites
found in the MtLYK3 InR (but not tested for their role in CD) and LjNFR1 InR (Madsen et al., 2011)
are highlighted in black. In vivo phosphorylation sites found in the AtCERK1 InR (Petutschnig et al., 2010)
and in vitro phosphorylation sites found in SlPto (Sessa et al., 2000) are highlighted in green and orange,
respectively.
138
HVR-mCherry
LYK3 [mpInR]-
sYFP2
LYK3 [cyt InR]-
sYFP2
LYK3 [T433A]-
sYFP2
LYK3 [T480E]-
sYFP2
LYK3 [Y483D]-
sYFP2
LYK3 [T512A]-
sYFP2
LYK3
[CERK1]-
sYFP2
channel: YFP mCherry merged & DIC
139
Figure 2. Subcellular localization of various mutated and truncated MtLYK3-sYFP2 protein fusions in
Nicotiana leaf epidermal cells.
HVR-mCherry, encoding the PM marker, was co-expressed with (from top to bottom): MtLYK3 [T480A]-
sYFP2 (1); MtLYK3 [T480E]-sYFP2 (2); MtLYK3 [mp IR]-sYFP2 (3); or MtLYK3 [cyt IR]-sYFP2 (4)
in Nicotiana leaf epidermal cells via Agrobacterium-mediated transient transformation, and the fluorescence
(viewed from abaxial side) was imaged 24hai using confocal laser scanning microscopy. YFP channel
presents green fluorescence of sYFP2; mCherry channel presents orange fluorescence of mCherry; merged
channel superimposes green, orange, and red (chlorophyll fluorescence) with differential interference contrast
(DIC) image. Bars are 20 µm.
substitution on MtLYK3 activity in vitro and in vivo. MtLYK3 [T480A]-sYFP2 protein
fusion was efficiently produced and correctly localized to the PM of Nicotiana leaf
epidermal cells (Fig. 2) but was incapable of CD induction upon co-production with
MtNFP-mCherry (Fig. 3A, Table 1). In order to test its in vitro autophosphorylation
activity, recombinant InR (fused N-terminally to a glutathione S-transferase [GST])
of WT MtLYK3, kinase-inactive MtLYK3 [G334E] (Klaus-Heisen et al., 2011;
see Table 1), and MtLYK3 [T480A] mutated variant were produced in Escherichia coli,
and the purified GST protein fusions were incubated with radiolabeled ATP
(γ-32
P ATP). Phosphorimage analysis of the protein fusions separated on SDS-PAGE
gel demonstrated that the G334E and T480A substitutions similarly abolished in vitro
autophosphorylation activity of MtLYK3 kinase (Fig. 3B). Therefore, T480A
substitution abolished MtLYK3 activity in vitro and in vivo.
Removal of nearly the entire extracellular region of MtLYK3 results in
gain-of-function phenotype in Nicotiana leaf
In order to test the importance of the MtLYK3 ExR for CD induction,
we deleted all three LysM domains of MtLYK3 by fusing the signal peptide sequence
to the sequence encoding the transmembrane (TM) helix and the InR of the protein.
To our surprise, MtLYK3 ΔLysMs-sYFP2 (aa’s: 1-23 and 223-620) was able to induce
140
Table 1. Cell death induction upon (co-)production of various MtLYK3-sYFP2 mutated variants with
MtNFP-mCherry in Nicotiana leaves.
MtLYK3-sYFP2
construct
Auto-
phosphorylation
activity*
Nodulation
activity*
Cell death induction
Co-expression
with MtNFP-
mCherry
Separate
expression
WT + + 48/52 0/20
T319A (α-helix B) - - 0/11 0/9
K349A (AxK motif) - - 0/16 0/9
E362A (α-helix C) - - 0/15 0/9
Y390F (gatekeeper) Reduced + 11/12 0/9
R440A (catalytic loop) Reduced NT 0/12 0/9
D441A (catalytic loop) - - 0/16 0/9
K464A (activation segment) Highly reduced NT 0/12 0/9
S471A (AL) + Reduced 9/11 0/9
T472A (AL) + Reduced 5/11 0/9
T475A (AL) - - 0/13 0/9
R476A (AL) Highly reduced NT 0/12 0/9
T480A (after AL) - NT 0/18 0/9
T285A/S286A/T300A (JM) + Reduced with
T300A
25/26 0/9
T433A + Reduced 10/29 0/10
T512A (α-helix F) Reduced - 8/15 0/9
S471D T472E T472E (AL) + + 8/12# 0/9
T480E - NT 0/12 0/9
Y483D NT NT 0/12 0/9
[AtCERK1] Reduced -** 0/12 0/9
*-see Klaus-Heisen et al. (2011), **- B. Lefebvre, personal communication. NT-not tested
The designated constructs were expressed alone or co-expressed with MtNFP-mCherry in Nicotiana leaves
via Agro TT, and the infiltrated regions were marked. Macroscopic symptoms of CD were scored 48hai: only
infiltrations that resulted in confluent death of (nearly) the entire infiltrated region were scored as a fraction
of total independent infiltrations performed. # for MtLYK3 phosphomimic variant, fraction of infiltrated
regions showing limited (patch or spot of dead tissue) is presented. In case of no macroscopic symptoms,
three leaves were stained with Evans blue to confirm the lack of CD (data not shown), and macroscopic
symptoms in the remaining infiltrated regions were scored again 72hai (no further increase in macroscopic
symptoms was noted), after which point a weak unspecific chlorosis could be observed.
CD even in the absence of MtNFP-mCherry: confluent death of (nearly) the entire
infiltrated region was observed in 5 out of 20 infiltrations, whereas the remaining
141
1
4
3
2
1
4
3
2PI, γ-32P ATP
A
CB
B
Figure 3. Various (putative) phosphorylation sites are differentially required for MtLYK3 kinase
activity in vitro and its biological activity in Nicotiana.
A, CD induction assay in N. benthamiana. MtLYK3-sYFP2 constructs carrying point mutations
were co-expressed with MtNFP-mCherry in Nicotiana leaves via Agro TT, and the infiltrated regions
were marked: MtNFP-mCherry + MtLYK3 [T285A S286A T300A]-sYFP2 (1); MtNFP-mCherry + MtLYK3
[T433A]-sYFP2 (2); MtNFP-mCherry + MtLYK3 [T512A]-sYFP2 (3); MtNFP-mCherry + MtLYK3 [T480A]-
sYFP2 (4). Left panel presents the macroscopic symptoms of CD observed 48hai. Right panel presents
the same leaf half stained with Evans blue.
B, in vitro kinase activity assay. The purified GST protein fusion of WT MtLYK3, MtLYK3 [G334E],
MtLYK3 [T480A], MtLYK3 [T480E], and MtLYK3 [CERK1] InRs were analyzed for their
autophosphorylation activity in vitro using radiolabeled ATP (γ-32P ATP) and phosphorimaging (PI).
The coomassie blue staining (CB) shows the protein loading.
regions displayed a patch or spot of dead tissue (Table 2). Co-production of MtLYK3
[ΔLysMs]-sYFP2 and MtNFP-mCherry resulted in confluent death of (nearly) the entire
infiltrated region in 11 out of 12 infiltrations (Table 2). As full-length MtLYK3 required
its kinase activity to induce CD in Nicotiana leaf upon co-production with MtNFP,
we tested the effect of the G334E substitution on the biological activity of MtLYK3
[ΔLysMs] truncated variant. MtLYK3 [ΔLysMs G334E]-sYFP2 was still able to induce
142
a limited CD in the absence and presence of MtNFP-mCherry, as observed
in 8 out of 12 and 5 out of 12 infiltrated regions, respectively (Table 2). Therefore,
removal of the MtLYK3 LysM domains generated a gain-of-function truncated variant
capable of CD induction in the absence of MtNFP. This activity seemed to be enhanced
in the presence of MtNFP and was not abrogated by the G334E substitution.
Plasma membrane localization of the MtLYK3 intracellular region is required for
cell death induction in Nicotiana leaf
Deletion of the ExR has been reported to activate several RTKs in the ligand-
independent manner (Merlin et al., 2009 and refs therein), providing that the TM helix
and the InR were retained in these truncated receptors. In order to investigate whether
the TM helix is required for MtLYK3 ΔLysMs-sYFP2 biological activity, we generated
two constructs encoding only the MtLYK3 InR (aa’s 253-620) fused C-terminally
to sYFP2. Introduction of a start codon before the sequence for MtLYK3 InR resulted in
cytosolic localization of the encoded protein fusion in Nicotiana leaf epidermal cells
(Fig. 2), and therefore the construct was termed MtLYK3 [cyt InR]-sYFP2. Introduction
of an eight amino acid myristoylation palmitoylation signal before the sequence
for the MtLYK3 InR targeted MtLYK3 [mp InR]-sYFP2 protein fusion to the PM
of Nicotiana leaf epidermal cells (Fig. S1). Production of MtLYK3 [mp InR]-sYFP2
alone did not induce CD, as confirmed with Evans blue staining (Table 2) but its
co-production with MtNFP-mCherry resulted in confluent death of (nearly) the entire
infiltrated region in 19 out of 20 infiltrations (Table 2). Expression of MtLYK3
[cyt InR]-sYFP2 alone or its co-expression with MtNFP-mCherry did not result in CD
induction (Table 2). Therefore, the localization of the MtLYK3 InR at the PM
was required and sufficient for CD induction in the presence of MtNFP.
143
Table 2. Cell death induction upon (co-)production of various MtLYK3-sYFP2 truncated protein
fusions with MtNFP-mCherry in Nicotiana leaves.
MtLYK3-sYFP2
construct
Cell death induction
Separate expression Co-expression with
MtNFP-mCherry
[∆LysMs]
5/20 *
11/12 *
[∆LysMs G334E]
8/12 **
5/12 **
[mp InR]
0/9
19/20 *
[cyt InR]
0/9
0/20
The designated constructs were expressed alone or co-expressed with MtNFP-mCherry in Nicotiana leaves
via Agro TT. The infiltrated regions were marked, and the macroscopic symptoms of CD were scored 48hai.
Infiltrations that resulted in confluent death of (nearly) the entire infiltrated region are presented as fractions
of total independent infiltrations performed, and are marked with (*). Infiltrations that resulted in only a patch
of dead tissue are presented as fractions of total independent infiltrations performed, and are marked with
(**).In case of no or weak macroscopic symptoms for a particular (pair of) construct(s), three leaves
were stained with Evans blue to confirm the lack of CD (data not shown), and macroscopic symptoms
in the remaining infiltrated regions were scored again 72hai, after which point a weak unspecific chlorosis
could be observed. For selected pairs of constructs, macroscopic symptoms of CD 48hai and subsequent
Evans blue staining are presented.
144
Alterations of the MtLYK3 activation segment sequence does not result in
gain-of-function protein variants
Prior to ligand binding, KDs of RKs and RLKs are kept in an inactive state
via intra- or inter-molecular inhibition (e.g. Lemmon & Schlessinger 2010; Jaillais
et al., 2011b). The observed functional interaction between MtNFP and MtLYK3
in Nicotiana could imply that the putative inhibition of the MtLYK3 KD is alleviated
upon its direct or indirect interaction with MtNFP, and we hypothesized that this effect
could be mimicked via a specific yet unknown alteration of the MtLYK3 structure.
We focused on investigating the effect of altering the sequence of the activation
segment, as this structural element plays a regulatory role in many ePKs (Huse &
Kuriyan 2002; Nolen et al., 2004). Our aim was to find (a) “sensitized”
or gain-of-function MtLYK3 variant(s) that would show faster CD induction
in comparison to WT MtLYK3 or CD induction independently of the MtNFP presence.
MtLYK3 possesses an Arg immediately preceding the conserved Asp
in the catalytic loop (Arg 440 and Asp 441), and thereby it belongs to the so-called
RD kinases that are generally activated through phosphorylation within the AL (Huse &
Kuriyan 2002; Nolen et al., 2004). Acidic amino acids that mimic the negative charge
conferred by phosphorylation can in some instances lead to constitutive activation
of mutant kinases (Johnson et al., 1996). Therefore, we generated a MtLYK3 phospho-
mimic variant via substitution of all three phosphorylation residues in the AL with Asp
or Glu. MtLYK3 [S471D T472E T475E]-sYFP2 was efficiently produced and correctly
localized to the PM in Nicotiana leaf epidermal cells (Fig. S1) but was unable to induce
CD in the absence of MtNFP-mCherry, as confirmed with Evans blue staining (data
not shown). Co-production of the phosphomimic MtLYK3-sYFP2 variant with MtNFP-
mCherry resulted in limited CD in 9 out of 12 infiltrated regions (Table 1),
145
and confluent death of (nearly) the entire infiltrated region was not observed. Therefore,
the mimicking of the AL phosphorylation did not result in the MtNFP-independent CD
induction but instead somewhat decreased MtLYK3 biological activity in Nicotiana.
Then, we searched for other receptor(-like) kinases with significant sequence
homology to the MtLYK3 KD for which detailed phosphorylation and structure-
function studies were described. We focused on the close homolog of MtLYK3,
AtCERK1 (69% identity of MtLYK3 and AtCERK1 InR aa sequences), and a receptor
kinase of tomato (Solanum lycopersicum), Pto (for Pseudomonas syringae tomato)
(32% identity of MtLYK3 and Pto KD aa sequences) (see Fig. 1 for the alignment
of MtLYK3 and AtCERK1 InRs, and Pto KD aa sequences). In case of AtCERK1,
we focused on the recent work of Nakagawa et al. (2011) who demonstrated AtCERK1
ability to function as a putative NF receptor upon alteration of its ExR and InR.
More specifically, replacement of the ExR sequence and the sequence encoding the AL
until an α-helix F in AtCERK1 for the corresponding sequences from LjNFR1 allowed
the resulting LjNFR1/AtCERK1 chimeric receptor to rescue the nodulation phenotype
of the Ljnfr1 mutant. Conversely, we wanted to know whether a reciprocal chimeric
protein composed of the AtCERK1 ExR and the MtLYK3 TM and InR
(where the sequence encoding the AL until the α-helix F was replaced with that
of AtCERK1) would change MtLYK3 biological activity in Nicotiana.
In case of Pto, we noted a series of constitutive gain-of-function Pto variants
that were found in several structure-function studies (Rathjen et al., 1999; Wu et al.,
2004; Xing et al., 2007; Dong et al., 2009). Several of these constitutive Pto variants
resulted from disruption of the C-terminal part of the AS responsible for the recognition
and binding of Pto ligands, AvrPto and AvrPtoB (Xing et al., 2007; Dong et al., 2009),
and possibly of a Pto interacting protein, Prf (Mucyn et al., 2006). Importantly, four
146
of the Pto residues (i.e. Gly 203, Thr 204, Gly 206, and Tyr 207) whose substitution
with Asp resulted in ligand-independent activation of Pto, are conserved in the MtLYK3
sequence (Gly 479, Thr 480, Gly 482, and Tyr 483, respectively). Therefore,
we decided to test the effect of Glu or Asp substitution of two of these residues,
i.e. Thr 480 and Tyr 483, on MtLYK3 biological activity in Nicotiana.
Again, MtLYK3 [T480E], MtLYK3 [Y483D], and MtLYK3 construct where
the sequence encoding the AL until the α-helix F was replaced for the corresponding
sequence from AtCERK1, termed MtLYK3 [CERK1], were generated as C-terminal
fusions to sYFP2. All three protein fusions were efficiently produced and localized
to the PM in Nicotiana leaf epidermal cells (Fig. 2). However, no CD induction
was observed upon production of any of these protein fusions alone or together with
MtNFP-mCherry (Table 1), as confirmed with Evans blue staining (data not shown).
Therefore, we decided to test in vitro autophosphorylation activity of MtLYK3
[CERK1] chimeric and MtLYK3 [T480E] mutated proteins. Recombinant InRs of WT
MtLYK3, MtLYK3 [G334E], MtLYK3 [T480E], and MtLYK3 [CERK1] (fused
N-terminally to GST) were produced in E. coli, and the purified GST protein fusions
were incubated with radiolabeled ATP (γ-32
P ATP). Phosphorimage analysis
of the proteins separated on SDS-PAGE gel demonstrated that the T480E substitution
abolished autophosphorylation activity of the MtLYK3 kinase in vitro (Fig. 3B).
Recombinant MtLYK3 [AtCERK1] InR showed limited autophosphorylation activity
in vitro, which however was significantly lower than that of recombinant WT MtLYK3
InR (Fig. 3B). Therefore, the above substitutions (T480A, T480E, and Y483D)
or the replacement of a part of the activation segment (from the AL until α-helix F) with
the corresponding sequence from AtCERK1 impaired MtLYK3 autophosphorylation
activity in vitro and precluded MtLYK3 biological activity in Nicotiana.
147
DISCUSSION
Similarities between MtLYK3 signaling in Medicago and Nicotiana
We previously demonstrated the relevance of the heterologous system
of Nicotiana leaf for structure-function studies on the MtNFP InR (see Chapter 2 and 4).
Here, we compared the requirements with respect to MtLYK3 structure
for its biological activity in Medicago (Klaus-Heisen et al., 2011) and Nicotiana.
Remarkably, out of 16 residues whose role in MtLYK3 function in nodulation
was characterized, 10 residues (Thr 319, Gly 334 [see Chapter 2], Lys 349, Glu 362,
Thr 433, Arg 440, Asp 441, Lys 464, Thr 472, Thr 475) were found to be equally
important, and 3 residues (Thr 285, Ser 286, and Tyr 390) were equally dispensable for
MtLYK3 ability to induce CD in Nicotiana leaf in the presence of MtNFP (Table 1).
In addition, we showed that the T512A substitution partially impaired MtLYK3
biological activity in Nicotiana, similarly to its effect (although more severe)
on MtLYK3 function in nodulation (see Table 1). We demonstrated that various
mutations abrogating MtLYK3 in vitro auto-phosphorylation activity (i.e. T319A,
K349A, E362A, D441A, K464A, T475A, R476A, T480A, T480E) similarly abolished
MtLYK3 biological activity in Nicotiana (Table 1). As all kinase-inactive MtLYK3
variants were efficiently produced and correctly localized to the PM in Nicotiana leaf
epidermal cells (Fig. 2), their lack of biological activity can be attributed to the general
lack of kinase activity rather than to an individual effect of a particular mutation
(see discussion on the D164N substitution in Pto [Xing et al., 2007]).
In addition, as the effect of the R440A, K464A or R476A substitutions
on MtLYK3 function in nodulation has not been described so far, our results provide
a proof for the role of these residues for MtLYK3 biological activity. Moreover,
148
residues corresponding to the Thr 319, Thr 433, and Thr 475 in MtLYK3 sequence have
been identified as in vitro phosphorylation sites in the LjNFR1 InR (Thr 319, Thr 434,
and Thr 476, respectively) but their effect on LjNFR1 biological activity has not been
analyzed (Madsen et al., 2011). Our results in Nicotiana independently support
the importance of these residues for MtLYK3 function in vivo. Finally, we showed
that the Thr 480 was similarly important for MtLYK3 activity in vitro and in vivo
(Fig. 3, Table 1), as the corresponding Thr 481 for LjNFR1 (Madsen et al., 2011).
Future analysis of the effect of substitution of the Thr319, Thr 434, and Thr 476
on LjNFR1 activity in vitro and in vivo should reveal whether MtLYK3 and LjNFR1
share structural requirements also for these residues.
With respect to the MtLYK3 ExR, we previously reported that the P87S
substitution in the first LysM domain did not affect MtLYK3 biological activity
in Nicotiana (see Chapter 2). As this point mutation abolishes MtLYK3 function
in nodulation (Smit et al., 2007), we concluded that nodulation and CD induction
in Nicotiana might have different requirements with respect to the MtLYK3 ExR
sequence. Our results with MtLYK3 [mp InR] truncated variant (Table 2) confirmed
and expanded this observation. As expression of MtLYK3 [mp InR]-sYFP2 constructs
driven by a constitutive CaMV 35S promoter did not complement the Mtlyk3-1
loss-of-function mutant (B. Lefebvre, personal communication), we conclude that
MtLYK3 biological activities in Medicago and Nicotiana differ with respect to the role
of the MtLYK3 ExR. We speculate that the heterologous co-production of MtNFP
and MtLYK3 circumvents their activation, normally achieved upon (direct or indirect)
NF perception, and that their observed functional interaction in Nicotiana leaf reports
only on the subsequent activation of the presumed receptor complex, i.e. activation
of the MtLYK3 KD. The expandability of the MtLYK3 ExR (this study) and of NF
149
(see Chapter 2) for triggering CD in Nicotiana leaf in the presence of MtNFP is in sharp
contrast to their absolute requirement for triggering nodulation in Medicago. This might
suggests that a specific modulation of the interaction between MtNFP and MtLYK3
interaction by NF is required for nodulation, and is mediated at least by the MtLYK3
ExR, whereas it is apparently bypassed during CD induction in Nicotiana. In this view,
the triggering of CD might be caused by the absence of a modulatory component
that is present in Medicago roots and keeps MtLYK3 inactive in the absence of NFs.
Interestingly, co-localization of the MtLYK3 InR with MtNFP at the PM
was required and sufficient for triggering CD in Nicotiana, whereas co-production
of MtNFP and the MtLYK3 InR in the cytoplasm was not sufficient to trigger CD
(Table 2). Recently, formation of a stable receptor complex between Arabidopsis
Flagellin Sensing 2 (AtFLS2) and AtBAK1 was shown to require a membrane
environment (Schulze et al., 2010). We hypothesize that co-localization of MtNFP
and MtLYK3 InR at the PM might facilitate their (direct or indirect) interaction.
Alternatively, putative downstream signalling components (e.g. calcium channels
or NADPH oxidases), whose activation in response to MtNFP and MtLYK3
co-production ultimately leads to CD, are located at the PM.
The hypothesized role of (putative) autophosphorylation sites for MtLYK3
biological activity
Three Ala substitutions, i.e. T433A, T472A, and T512A impair or abolish
MtLYK3 biological activity in Medicago and Nicotiana without impairing MtLYK3
in vitro autophosphorylation activity (Klaus-Heisen et al., 2011; Fig. 3A and Table 1).
A sequential phosphorylation of certain residues, often located in the AL
(but not necessarily), is required for full activation of the KDs of several animal RKs
150
(Lew et al., 2009 and refs therein). We speculate that phosphorylation of Thr 472 might
be required to enhance catalytic activity of MtLYK3 kinase. In case of Thr 433
and Thr 512 an educated guess on their role in MtLYK3-mediated signaling
is hampered by the fact that the importance of the corresponding Ser or Thr residues
in other receptor(-like) kinases has not been described yet. At the moment,
we hypothesize that phosphorylation of these residues similarly regulate MtLYK3
kinase activity or is required for generation of binding motifs for putative downstream
signaling components. With respect to the latter (presumed) function, it is interesting
to note that a Thr residue corresponding to the Thr 433 in MtLYK3 is conserved
in several LysM-RLKs and Pto (data not shown) but not in Leucine-rich Repeat RLKs,
such as Arabidopsis BRI1, BAK1, Somatic Embryogenesis Receptor Kinase 1
(AtSERK1), and Lotus Symbiotic Receptor Kinase 1 (LjSYMRK). In the AtBAK1 KD,
the corresponding Cys 408 is required for AtBAK1 function as a co-receptor for
AtFLS2 or Arabidopsis EF-Tu Receptor Kinase (AtEFR) but not for its role in brassino-
steroid signaling and control of seedling lethality (Schwessinger et al., 2011).
Mutagenesis analysis of the corresponding residue in other LysM-RLKs or Pto should
provide valuable insights into a possible role of this conserved Thr, as it might be
implicated in establishing protein-protein interactions of these receptor kinases with
additional signaling components. As the phosphorylation substrate of MtLYK3,
MtPUB1 (for Plant U-box protein 1) has already been identified (Mbengue et al., 2010),
it would be interesting to test the effect of the T433A substitution on MtLYK3 binding
and transphosphorylation of this regulator.
In contrast, the T300A mutation reduced MtLYK3 biological activity
in Medicago (Klaus-Heisen et al., 2011; see Table 1) but not in Nicotiana (Fig. 3A,
Table 1). In many animal RKs and several RLKs, phosphorylation outside the AL
151
has been shown to generate docking sites for adaptor proteins or downstream signaling
components (e.g. Wang et al., 2008; Lemmon & Schlessinger 2010). Certain
phosphorylation sites are required for docking of (a) specific signaling component(s),
while they can be dispensable for others. For example, phosphorylation of Tyr 610
located in the C-tail of AtBAK1 was crucial for its function in brassinosteroid signaling
and control of seedling lethality but was dispensable for its role as a co-receptor
for Arabidopsis Flagellin Sensing 2 (AtFLS2) receptor (Oh et al., 2010 and refs
therein). In addition, phosphorylation of both animal RKs and plant RLKs on specific
residues has been implicated in the regulation of their kinase activity (e.g. Oh et al.,
2012). In analogy, it is not excluded that binding of (a) symbiosis-specific interactor(s)
of MtLYK3 requires the phosphorylation of Thr 300, while it is dispensable for
MtLYK3 signaling for CD induction. Alternatively, phosphorylation of this residue
might provide a specific regulation of MtLYK3 kinase activity required for efficient
nodulation, whereas this function is not required or might even be the cause of CD
induction in Nicotiana.
Various alterations in the MtLYK3 structure do not alleviate the hypothesized
activation of MtLYK3 kinase by MtNFP
Encouraged by the demonstrated significant similarity of nodulation and CD
induction requirements with respect to the MtNFP and MtLYK3 structure (see Chapter
2 and 4; this study), we set out to investigate regulatory mechanisms governing
MtLYK3 kinase activity in Nicotiana, using a series of mutated, truncated and chimeric
constructs. Our working hypothesis was that the putative (auto)inhibition of MtLYK3
kinase activity is alleviated upon its direct or indirect interaction with MtNFP,
and that this effect can be mimicked via a specific yet-unknown alteration
152
of the MtLYK3 structure. In this respect, removal of nearly the entire ExR of MtLYK3
resulted in a gain-of-function (Table 2): MtLYK3 [ΔLysMs] induced CD in the absence
of MtNFP and also independent of its intrinsic (MtNFP-induced) kinase activity.
As this MtNFP-independent CD induction was observed only in case of MtLYK3
[ΔLysMs] and MtLYK3 [ΔLysMs G334E] truncated protein fusions but not with
MtLYK3 [mp InR], this phenomenon might be attributed to the presence
of the MtLYK3 TM helix. Deletion of the ExR of several RTKs has been shown
to increase their signaling output (Merlin et al., 2009 and refs therein), presumably
as a result of co-operation between TM helices and JM regions in stabilization
of the KDs in a specific active position within the receptor dimer (Lemmon &
Schlessinger, 2010; Endres et al., 2011). However, as the biological activity
of MtLYK3 [ΔLysMs] truncated variant was not abrogated by the G334E mutation
(similarly to the biological activity of full length MtLYK3), it is not related to kinase
activity. Therefore, CD-induction by MtLYK3 [ΔLysMs] or MtLYK3 [ΔLysMs
G334E] truncated variant might involve another yet-unidentified protein (kinase)
in Nicotiana leaf. Future identification of (a) putative interactor(s) of MtNFP
and MtLYK3 in Nicotiana is required to understand the underlying mechanism of CD
induction upon their simultaneous accumulation, and to verify the mechanism
of apparent MtLYK3 activation due to the removal of its LysM domains.
In addition, our results suggest that alterations of the activation segment alone
are not sufficient to activate MtLYK3 kinase but instead have a negative effect
on its in vitro and/or in vivo activity. In this respect, MtLYK3 and possibly other RLKs
seem to differ from mitogen-activated protein kinases (MAPKs), for which mimicking
of the AL phosphorylation with acidic amino acids results in generation
of constitutively active variants (e.g. Asai et al., 2008 and refs therein). On the contrary,
153
Asp or Glu substitution of the phosphorylation sites located in the AL did not produce
a constitutively active AtBAK1 mutated variant (Wang et al., 2008) or even abolished
in vitro and/or in vivo activity of MtLYK3 (Fig. 2B, Table 1) and Pto (Sessa et al.,
2000). Moreover, alteration of the C-terminal part of the AS had a strikingly different
effect on MtLYK3 (Table 1) and Pto (Rathjen et al., 1999; Wu et al., 2004; Xing et al.,
2007; Dong et al., 2009) biological activities in Nicotiana. More specifically,
substitution of the conserved Thr (Thr 204 in Pto and Thr 480 in MtLYK3) or Tyr
(Tyr 207 in Pto and Tyr 483 in MtLYK3) resulted in ligand-independent induction
of the Pto-mediated signaling, whereas it abolished MtLYK3 biological activity
in Nicotiana. The lack of biological activity of MtLYK3 T480E mutated variant
is supported by the abrogation of its in vitro autophosphorylation activity (Fig. 3B),
and we speculate that the Y483D mutation has a similar negative effect on MtLYK3
kinase activity. In addition, our results agree with the effect of a corresponding mutation
on in vitro and/or in vivo activity of LjNFR1, AtBRI1, and AtBAK1 (T455D and T455E
in AtBAK1, T1049D and T1049E in AtBRI1: Wang et al., 2008; T481D in LjNFR1:
Madsen et al., 2011). Strikingly, a similar abrogation of in vitro autophosphorylation
activity of Pto was reported for the T204D, T204N, and Y207D mutations (Xing et al.,
2007). Therefore, the ligand-independent activation of Pto kinase resulting from
the alterations of the C-terminal part of its AS is not reflected by the phosphorylation
activity of Pto kinase but instead is likely achieved via a phosphorylation independent
mechanism. Mucyn et al. (2006) postulated that Pto is required to inhibit Prf activity
in vivo, and modulation of the interaction between these two proteins (in result of
mutating the C-terminal part of the Pto activation segment), rather than Pto catalytic
activity, is required to induce Prf-mediated signaling (although whether a similar
mechanism underlies the ligand-dependent activation of Pto remains to be investigated).
154
On the contrary, MtLYK3 biological activity seems to be stringently dependent
on its kinase activity (Table 1), thereby explaining the lack of biological activity
of MtLYK3 [T480E] or [Y483D] mutated variants. Similarly, we speculate that
the apparent lack of biological activity of MtLYK3 [AtCERK1] chimeric protein
(Table 1) could be attributed to its impaired kinase activity (Fig. 3B).
In conclusion, our results confirm that the heterologous system of Nicotiana
is relevant for structure-function studies on the kinase and/or biological activity
of MtNFP and MtLYK3, and possibly of other LysM-RLKs. This is especially
interesting, as the KDs of two extensively studied Arabidopsis receptor kinases, AtBRI1
and AtBAK1, display limited homology with the KD of LysM-RLKs (Shiu & Bleecker
2003), and seem to employ specific regulatory mechanisms not applicable for other
RLKs (Oh et al., 2012). Therefore, we propose that the observed CD induction
in Nicotiana leaf upon transient expression of AtCERK1 could be used to analyze
the effect of the recently identified in vivo phosphorylation sites in AtCERK1
(Petutschnig et al., 2010) on its biological activity. Such future studies would
potentially be of great importance for expanding our understanding of molecular
mechanisms that regulate kinase activity of various ePKs.
155
EXPERIMENTAL PROCEDURES
Constructs for plant expression
pBin+ CaMV 35Sp::MtNFP-mCherry and pCambia1390 CaMV 35Sp::MtLYK3-sYFP2 constructs
are described in Chapter 2. Point mutations were introduced in the pMon999 CaMV 35Sp::MtLYK3-sYFP2
vector using the QuickChangeTM site-directed mutagenesis kit (Stratagene), and the HindIII-SmaI fragment
(containing the 35Sp::MtLYK3-sYFP2 sequence) was subsequently recloned into pCambia1390 vector.
An XbaI restriction site within the MtLYK3 ExR sequence was used to generate MtLYK3 [ΔLysMs]
construct. All truncated constructs were generated by PCR amplification (see Table 4 for primer sequences)
and cloned into CaMV 35Sp::sYFP2 pCambia1390 vector. All pCambia1390 constructs were sequenced
to verify the correct insert sequence.
Plant transformations
Transformation of Agrobacterium tumefaciens strain GV3101::pMP90 and Agrobacterium infiltration
of N. benthamiana leaf is described in Chapter 2. Agrobacterium transformants carrying wild-type or mutated
MtLYK3-sYFP2 construct were resuspended to OD600=0.8 and mixed 1:1 with GV3101 transformants carrying
MtNFP-mCherry or empty pCambia1390 vector (resuspended to OD600=0.5) before infiltration.
Agrobacterium transformants carrying MtLYK3 truncated constructs were resuspended to OD600=0.5
and mixed 1:1 with GV3101 transformants carrying MtNFP-mCherry or empty pCambia1390 vector
(resuspended to OD600=0.5) before infiltration. All CD assays were performed at least three times, every time
using three different plants, and the results were collated. In case of lack or inefficient CD induction
by the (co-)expression of (a) certain construct(s), three leaves were stained with Evans blue (as described
in Chapter 2) 48hai to check for infrequent CD.
Microscopic analysis and kinase autophosphorylation assay
Were carried as described in Klaus-Heisen et al., (2011).
Table 4. Primer and linker sequences.
Name Type Sequence
MtLYK3 fw Cloning (NheI) GGGGCTAGCATGAATCTCAAAAATGGATTAC
MtLYK3 sp rev Cloning (XbaI) GTTCTAGAATACAAAGGAAAAA
mp fw# Cloning (NheI) GGGCTAGCATGGGAGGATGCTTCTCTAAGAAG
MtLYK3 mp InR fw Cloning GGATGCTTCTCTAAGAAGAAATACTTCCAAAAGAAGGAA
MtLYK3 cyt InR fw Cloning (NheI) GGGGCTAGCATGAAATACTTCCAAAAGAAGGAA
MtLYK3 rev Cloning (EcoRI) GGGAATTCTCTAGTTGACAACAGATTTATG
Linker to FP GAATTC for all the constructs
* Sequences for restriction sites are underlined, myristoylation and palmitoylation signal is in italics.
# - mp fw primer was mixed 2:1 with either MtLYK3 mp InR fw or MtLYK3 cyt InR fw primer before setting
up PCR reaction.
156
157
CHAPTER 6 CONCLUDING REMARKS
Symbiosis with nitrogen-fixing Rhizobium and Frankia bacteria provides
legume and actinorrhiza species with a nitrogen source so that they need no chemical
fertilizers for supporting growth. As the usage of chemical fertilizers has a significant
negative impact on the environment (Brentrup & Palliere 2008; Dawson & Hilton 2011)
and is costly, improvement of symbiotic performance in legume species is a remarkable
opportunity for sustainable agriculture. On the contrary, many economically important
plant species cannot establish symbiosis with rhizobia. The prospects of introducing this
trait into crops – the holy grail of the symbiosis community – requires understanding
of the origins of root nodule symbioses (with both Rhizobium and Frankia bacteria)
and of the underlying molecular processes (Charpentier & Oldroyd 2010).
The current consensus is that AM symbiosis has evolved at least 400 Million
years ago (Humphreys et al., 2010; Wang et al., 2010) and provided certain traits
for the more recent (approximately 70 Million years ago) evolution of the RL symbiosis
(Bonfante & Genre 2008; Parniske 2008; Doyle 2011). For example, it has been
postulated that a specific rearrangement of the actin cytoskeleton and cytoplasm
at the pre- and infection stage (Genre et al., 2005, 2008) and a specific exocytotic
pathway required for the formation of a membrane compartment for accommodation
of the microsymbiont (Ivanov et al., 2012) have been recruited from AM by the RL
symbiosis. In addition, recent findings indicate that the latter interaction likely co-opted
the complete signaling pathway, i.e. the receptor and downstream CSP,
for the perception of the microsymbiont and triggering of the symbiotic program
from the AM. First of all, Myc factors secreted by G. intraradices have shown to be
sulphated and nonsulphated tetrameric and pentameric LCOs, and hence resemble
158
the NF produced by rhizobia (Maillet et al., 2011). Secondly, a single LysM-RLK
in P. andersonii, PaNFP, has been implicated in interactions of this nonlegume species
with both Sinorhizobium sp. NGR234 and G. intraradices (Op den Camp et al., 2011).
Because Myc factors and NFs are structurally similar (with NFs showing higher
structural variation) and are putatively perceived by related receptors, LCO signaling
is not unique to legumes (as previously believed) but is likely widespread in plants,
and might be evolutionarily as old as AM symbiosis. The findings presented in this
Thesis indicate that besides the striking similarities between signaling in RL and AM
symbioses, there are also similarities with stress/defense signaling, which may be even
more ancient. Here, selected aspects of NF binding, formation of putative NF receptor
complex, and signal transduction will be revisited with an aim of pinpointing
similarities and differences in NF, Myc factor, and chitin/COs signaling.
Possible mechanisms of MtNFP and MtLYK3 interaction
Knock-down of MtNFP and MtLYK3 has a similar disrupting effect on IT
growth, and therefore the encoded LysM-RLKs are postulated to co-function during
regulation of rhizobial intracellular invasion (Limpens et al., 2003; Arrighi et al., 2006;
Smit et al., 2007; Bensmihen et al., 2011). Here, we showed that simultaneous
accumulation of MtNFP and MtLYK3 (or MtLYK2) in Nicotiana leaf triggers CD
and a defense(-like) response, indicating a functional interaction of these LysM-RLKs
in this heterologous plant system. One of the explanations for the apparent
co-functioning of these LysM-RLKs in both nodulation and CD induction is that they
interact with and activate different downstream signaling component(s)
and the triggered parallel pathways converge at the later point. Alternatively, MtNFP
and MtLYK3 might form a receptor complex to signal for nodulation and CD induction,
159
although this notion has not been confirmed directly in either Medicago or Nicotiana.
However, in situ molecular interaction studies between the putative NF receptors have
so far been limited to the heteromerization tendency of the kinase-inactive MtLYK3
(this study) or LjNFR1 (Madsen et al., 2011) with, respectively, MtNFP and LjNFR5
upon their transient production in Nicotiana leaf. While these studies have indicated
modest heteromerization potential of the kinase-inactive MtLYK3/LjNFR1 (discuss
ed in Chapter 3), physical interaction between WT NF receptors remains to be shown.
Therefore, at the moment both hypotheses can be true and even a combination
of the two alternatives cannot be ruled out.
In the light of recent findings on the evolutionary origin of RL symbiosis,
the postulated interaction of MtNFP/LjNFR5 and MtLYK3/LjNFR1 is highly
interesting. In short, gene duplication events underlie the significant expansion
of LysM-RLK-encoding genes in legumes (Shiu et al., 2004; Zhang et al., 2009b),
and the maintenance and neofunctionalization of duplicated gene copies is postulated
to play an essential role during the evolution of RL symbiosis (Doyle 2011; Young
et al., 2011). The recruitment of a duplicated copy of the LysM-RLK gene
in the MtNFP/LjNFR5 clade, presumably functioning upstream the CSP in AM
symbiosis, and its specialization due to co-evolution of legumes and rhizobia is thought
to result in the present structure of MtNFP/LjNFR5 receptor (Streng et al., 2011).
Sequence alterations, especially in the second LysM domain in this duplicated receptor,
are believed to result in its specialization for stringent recognition of rhizobial signal,
the NF (Gough & Cullimore 2011). However as there is currently no proof
for the involvement of a protein encoded by a gene in the MtLYK3/LjNFR1 clade
in AM symbiosis, it is not clear whether the ancestors of these putative NF receptors
have already co-functioned prior the recruitment by the RL symbiosis. As the recently
160
identified MtNFP homolog in P. andersonii, PaNFP, is a putative pseudokinase (Gough
& Cullimore 2011), the involvement of a yet-unidentified receptor component
is conceivable but remains to be shown. On the other hand, while MtLYK3/LjNFR1
and AtCERK1 are likely descendants of a common ancestor (Zhang et al., 2009b
and refs therein), no MtNFP/LjNFR5-like protein has so far been implicated
in AtCERK1-mediated chitin or PGN signaling. Instead, OsCERK1 is hypothesized
to interact with and become activated by (an) extracellular LysM protein(s), termed
LYMs or LYPs (Zhang et al., 2009b; Fliegmann et al., 2011 and refs therein),
upon its/their binding to COs/PGN (Shimizu et al., 2010; Liu et al.,2012a). A similar
activation of AtCERK1 by the LysM proteins, AtLYM1 and AtLYM3, upon their
binding to PGN has been suggested by Willmann et al. (2011). If the postulated
physical interaction between MtNFP/LjNFR5 and MtLYK3/LjNFR1 is confirmed
conclusively, it may indicate that the ancient forms of these LysM-RLKs already
co-functioned prior the recruitment by RL symbiosis or that the putative NF receptors
co-evolved after independent recruitment to co-function in NF signaling. Future
identification of Myc factor receptors in model legume species (which might include
MtLYR1 of Medicago; Gomez et al., 2009) and dissection of evolutionarily younger
symbiosis of P. andersonii with Sinorhizobium sp. NGR234 should help clarify
this point, and would surely provide a highly interesting case study of receptor
diversification and specialization.
NF binding might require extracellular proteins other than CEBiP-like
LYM proteins
The hypothesized similarity between NF and CO signaling and their common
evolutionary roots is of interest in respect to ligand binding-mechanisms. For example,
161
AtCERK1/OsCERK1 is incapable of PGN binding but instead is postulated to become
activated upon perception of this PAMP by two extracellular LysM proteins,
AtLYM1/AtLYM3 and OsLYP4/OsLYP6, respectively (Willmann et al., 2011; Liu
et al., 2012a). A similar mechanism is envisaged in case of COs signaling in rice,
involving OsCERK1 and (a) LYM protein(s), OsCEBiP, OsLYP4 and/or OsLYP6
(Shimizu et al., 2010; Liu et al., 2012a).
Analogously, MtNFP is indispensable for triggering virtually all early
symbiotic responses in NF-dependent manner (Ben Amor et al., 2003; El Yahyaoui
et al., 2004; Mitra et al., 2004a; Arrighi et al., 2006), and it is postulated to do so
in co-operation with a kinase-active LysM-RLK (Arrighi et al., 2006; Smit et al., 2007;
Bensmihen et al., 2011). However, direct binding of LCOs to this or other putative NF
receptor has not been shown to date, despite an enormous effort of many groups.
On the contrary, it has been postulated that a putative Nod Factor Binding Factor
(NFBF) is abundant in the RH cell walls (Goedhart et al., 2003), in agreement with
the identification of a NF binding site (NFBS1) in high-density fraction from Medicago
spp. root extracts (Bono et al., 1995). Therefore, NF perception might require
formation of a multiprotein complex in which binding and recognition of specific
decorations are achieved by different components. This is corroborated by the fact that
NFBS1 from Medicago requires both the NF chito-oligosaccharide backbone
and a fatty acid moiety but not the presence of the sulphate group for binding (Bono
et al., 1995), while the latter modification is essential for the Medicago-S. meliloti
symbiosis (Ardourel et al., 1994; Oldroyd et al., 2001; D’Haeze & Holsters 2002).
Furthermore, both sulphated and non-sulphated NFs show equal accumulation
and immobilization in RH cell walls (Goedhart et al., 2003 and refs therein).
The recent demonstration that recognition of S. meliloti-specific NF sulphation
162
is independent of the MtNFP ExR led to hypothesis that another LysM domain-
containing protein (possibly a yet-unidentified LysM-RLK) might provide this function
(Bensmihen et al., 2011).
What could be the identity of the Nod Factor Binding Site in the cell wall?
Drawing from the chitin/COs and PGN signaling (Kaku et al., 2006; Shimizu et al.,
2010; Willmann et al., 2011), it might be hypothesized that high affinity NF binding
might be achieved by LYMs/LYPs. One of the two LYM proteins of Medicago,
MtLYM2, has recentlybeen demonstrated to bind COs (Fliegmann et al., 2011),
in agreement with the demonstrated affinity of LYM2-type proteins in rice (CEBiP)
and Arabidopsis (AtLYM2) for chitin/COs (Kaku et al., 2006; Petutschnig et al.,
2010). Therefore, chitin/COs signaling in this plant system might resemble a molecular
mechanism operating in rice and possibly also in Arabidopsis (see Shimizu et al., 2010
for discussion). However, the promoter of theother gene encoding a LYM protein,
MtLYM1, is not active in root epidermis or RHs (Fliegmann et al., 2011),
and homologous AtLYM1 and AtLYM3 have recently been implicated in PGN
binding (Willmann et al., 2011), indicating that MtLYM1 is unlikely to function
as a putative NF binding component. Therefore, it is possible that binding
of extracellular GlcNAc-containing molecules by LYMs/LYPs evolved specifically
during PAMP signaling, whereas binding of NFs involves a different molecular
mechanism. This might have resulted from the modification of the chito-
oligosaccharide backbone of NFs, especially with a fatty acid moiety (which is thought
to contribute to the binding affinity of LCOs; Gressent et al., 2002),
and is corroborated by significant diversification of LysM domains associated with
LysM-RLKs and LYMs/LYPs (Zhang et al., 2007). As both NFs and Myc factors
possess an acyl chain at the nonreducing termini, it can be postulated that they might
163
be bound in a similar fashion by related proteins or even that a shared protein might
bind both bacterial and fungal signaling molecules. After Goedhart et al. (2003),
it can be hypothesized that Myc factors and NFs might be bound with high affinity
by these putative components, termed binding factors, and subsequently presented
to LysM-RLK complexes. Drawing from the finding of NF binding studies (Gressent
et al., 2002 and refs therein) and NF immobilization in the RH cell walls (Goedhart
et al., 2003), it can be envisaged that LCOs might be bound to these sites irrespectively
of the presence/absence of certain modifications of the chito-oligosaccharide backbone
(such as sulphate group). In this scenario, stringent recognition of strain-specific
modifications of the LCO structure would be the function of the LysM-RLK receptor
complexes, specific for NF or Myc factor. This putative mechanism would ensure
robust perception of potentially limited signals from the microsymbionts (as all
produced Myc factors/NFs would be efficiently bound), and could explain the lack
of significant affinity of identified putative NF receptors for the NF.
AM and afterwards RL symbiosis might have co-opted signaling pathways in
response to chitin/chitin oligosaccharides for perception of bacterial signals
Heterologous co-expression of MtNFP and MtLYK3 in Nicotiana leaf results
in induction of CD and defense(-like) response that resembles the outcome
of the transient expression of AtCERK1 in this plant system (see Chapter 2). Influx
of extracellular Ca2+
is required for CD development triggered by AtCERK1
or simultaneous accumulation of MtNFP and MtLYK3. In addition, similar induction
of NbHIN1, NbPR1, NbACRE31, and NbACRE132 gene expression observed upon
production of AtCERK1 or co-production of MtNFP and MtLYK3 agrees with
AtCERK1-dependent induction of these genes in response to COs and/or PGN (Wan
164
et al., 2008; Segonzac et al., 2011; A. Gust personal communication). Therefore,
simultaneous accumulation of MtNFP and MtLYK3 in Nicotiana leaf appears
to induce a specific stress/defense signaling, possibly resembling CO or PGN
signaling. How can this happen? Based on the structural similarity of NFs (D’Haeze &
Holsters 2002) to COs, and the phylogenetic relationship between MtLYK3/LjNFR1
and AtCERK1 (Zhang et al., 2009b and refs therein), it has been postulated that the NF
signaling evolved from a more ancient and widespread chitin/COs signaling (Stacey
et al., 2006). In this scenario, decrease of the length of COs and their modification with
a fatty acid moiety are postulated to have contributed to modulation of their PAMP
activity (Hamel & Beaudoin 2010). However, with the recent demonstration
of structural similarities of and putative G. intraradices Myc factors (Maillet et al.,
2011) to both NFs and COs, it can be hypothesized that signaling in response to COs
evolved early during the course of plant-fungus interactions, and possibly might have
even predated the ancient form of AM symbiosis. Corroborating this, evidence for
defense-like response to a fungal endophyte has been found in the same Rhynie chert
as the earliest fossils representing putative AM symbiosis (Krings et al., 2007).
Is there evidence for stress/defense signaling induction during RL and/or AM
symbioses? Indeed, similarities with plant-pathogen interactions have been reported
for both RL symbioses. For instance, a similar repositioning of the plant nucleus
and aggregation of cytoplasm, development of a specific membrane compartment
accommodating the fungus intracellularly, and a significant overlap in host
transcriptomes (>40%) in response to beneficial or pathogenic biotrophic fungi indicate
an existence of a conserved “core” plant program during interactions with compatible
fungi (Parniske 2000; Güimil et al., 2005; Genre et al., 2009). In case of RL symbiosis,
homologs of pathogenic bacteria secretion systems (especially type III and type IV)
165
used to modulate host responses are found in certain rhizobia strains (Deakin &
Broughton 2009; Downie 2010). In addition, accumulation of phenolic compounds,
and a localized hypersensitive-like CD is associated with abortion of excessive ITs
in alfalfa (Vasse et al., 1993), and a premature senescence of ineffective (Fix-) nodules
in glycine, pea, and Medicago (Parniske et al., 1990; Perotto et al., 1994; Maunoury
et al., 2010 and refs therein). Importantly, such conditional induction of a defense
response is postulated to participate in the host autoregulation that monitors either
the total number of nodules formed or the symbiotic performance of rhizobia inside
these nodules. Recently, a symbiosis-specific role has been reported for small secreted
antimicrobial nodule-specific cysteine rich (NCR) polypeptides that govern bacteroid
differentiation in legume species forming indeterminate-type nodules (Haag et al., 2011;
Kereszt et al., 2011). Therefore, the accumulating body of evidence indicates that
the symbiotic program of mycorrhizal plant species, including legumes,
is evolutionarily related to plant innate immunity, and that certain symbiotic processes
are, in fact, a modulated version of plant responses to biotic stress.
Remarkably, similarities with plant stress/defense response can already
be found at the stage of early symbiotic signaling upon NF perception (please, compare
Figure 2and 3 in Chapter 1) or in response to diffusible signals produced by AM fungi
(although it remains to be confirmed whether these responses are triggered specifically
by the recently identified Myc LCOs). Corroborating this, NF and flg22 (an N-terminal
epitope of bacterial flagellin that acts as a PAMP) have been shown to trigger
phosphorylation of overlapping sets of proteins (Serna-Sanz et a., 2011), whereas
MtDMI3 has been implicated in regulation of a set of genes in response to plant growth-
promoting Pseudomonas fluorescens (Sanchez et al., 2005). Moreover, the initial stage
of AM but not of RL (Lohar et al., 2006) symbiosis in Lotus and Medicago is associated
166
with MtDMI3-dependent (Weidmann et al., 2004; Siciliano et al., 2007) activation
of MAPKs (Liu et al., 2003; Deguchi et al., 2007). The involvement of two GRAS type
TFs in both NF (NSP1 and NSP2) and COs (CIGR 1 and CIGR 2 for chitin-inducible
gibberellin-responsive 1 and 2) signaling points to additional similarity between these
pathways (Day et al., 2003; Heckmann et al., 2006; Murakami et al., 2006; Hirsch
et al., 2009 and refs therein).
What could be the underlying mechanism of triggering a specific response
(symbiosis vs defense) despite the involvement of apparently similar signaling
components? In addition to the modification of the CO structure, integration
of additional molecular components or even entire parallel signaling pathways might
have resulted in differentiation of host responses to beneficial and pathogenic
microorganisms, as observed during the extant plant-microorganism interactions
(Hamel & Beaudoin 2010). For example, recognition of compatible NFs
is not accompanied by an oxidative burst observed in response to COs and PGN (Gust
et al., 2007; Miya et al., 2007; Erbs et al., 2008; Wan et al., 2008; Petutschnig et al.,
2010) but by a transient increase in the intracellular ROS production, followed
by a prolonged slight elevation of ROS levels (Baier et al., 1999; Bueno et al., 2001;
Cárdenas et al., 2008; Serna-Sanz et al., 2011). Decreased expression of NADPH
oxidase genes in Medicago roots (Lohar et al., 2007), and increased activity of various
ROS scavenging enzymes in alfalfa roots (Bueno et al., 2001) upon perception
of specific NF, suggest modulation of ROS production by legume roots
by a yet-unidentified mechanism in response to compatible rhizobium. This specific
modulation is postulated to preclude the activation of plant innate immunity,
concomitantly ensuring ROS-mediated induction of gene expression
(e.g. of a peroxidase gene, MtRip1; Ramu et al., 2002 and refs therein) and RH curling
167
(Lohar et al., 2007). In addition, the triggered responses, including stress/defense-like
process, might play an important signaling role between the host plant
and the microsymbiont (Seddas et al., 2009). In view of hypothesized evolutionary
origin of LCO signaling from the COs signaling, detailed comparison of the induced
responses might provide information on the yet-unidentified molecular components
involved in perception and transduction of NF, Myc factor and CO signals.
For example, Ca2+
spiking is essential in both RL and AM symbioses (Oldroyd &
Downie 2006; Miwa et al., 2006a; Chabaud et al., 2011 and refs therein; Sieberer
et al., 2012) but no Ca2+
spiking has been demonstrated yet that in response to COs
(although high COs concentrations can trigger this response in RHs of legume cells;
Walker et al., 2000; Oldroyd et al., 2001). However, it is interesting to note that unlike
flg22-induced defense responses, COs-elicited gene induction and ethylene-dependent
deposition of callose occurs only in mature parts of Arabidopsis roots (Millet et al.,
2010). Similar spatial limitation of root responsiveness to this PAMP, if demonstrated
in legume roots, could explain the lack of Ca2+
spiking and ROS production
in response to lower concentrations of COs (Ehrhardt et al., 1996; Walker et al., 2000;
Oldroyd et al., 2001; Cárdenas et al., 2008). Detailed analysis of subcellular responses
of the root epidermis cells in the elongation (NF-responsive) and mature zone
of legume roots is required to verify the hypothesized similarity of CO-
and NF-perception in plants. This could be of special interest as the spatial separation
of NF and CO signaling might underlie the specificity of the triggered response
(symbiosis vs defense). In addition, detailed characterization of MAPK cascades
operating during AM symbiosis and of presumed activation of MAPKs in response
to COs signaling in model mycorrhizal plant species could be valuable. Similarly,
it will be of interest to characterize isoforms of PLC and PLD (presumably) involved
168
in compatible interaction of model mycorrhizal species, including legumes,
with microsymbionts and root pathogens. Finally, future identification of AtCERK1
orthologs in model legume species, facilitated by the analysis of their postulated
induction in response to chitin/COs (Lohmann et al., 2010) and the conservation
of the part of their KDs downstream of the activation segment (Nakagawa et al., 2010)
would be invaluable for studies of receptor specialization.
These studies are likely to reveal the molecular mechanisms underlying
the relay of signaling towards the appropriate plant response, and will greatly enhance
our understanding of where the key switches were made in the evolution of RL
and AM symbioses and pathogenic signaling in plants. Perhaps, in future, this will
allow us to artificially introduce nitrogen fixation into non-nodulating plants possibly
solving a major environmental burden.
169
SUMMARY
During interaction of legume plants with nitrogen-fixing rhizobia, perception
of bacterial lipo-chitooligosaccharides (LCOs), termed Nodulation (Nod) factors (NFs),
is usually required for formation of specialized plant organs (nodules) and for the host
root infection. Two LysM-RLKs, MtNFP and MtLYK3, are postulated to co-function
as putative NF receptors in Medicago truncatula (Medicago). We are interested in NF
signaling mediated by MtNFP and MtLYK3, and for this reason we have aimed at
characterizing the localization and possible molecular interaction of these proteins
in situ. However, our attempts to visualize these proteins in Medicago root have been
unsuccessful, presumably due to stringent regulation of their accumulation. Therefore,
we employed an Agrobacterium-mediated transient transformation of Nicotiana
benthamiana (Nicotiana) leaf, which allowed us to produce both proteins to the extent
that they could be visualized with fluorescence microscopy (presented in Chapter 2).
We demonstrated PM localization of both MtNFP and MtLYK3 C-terminal fusions
to a FP (sYFP2 or mCherry) in Nicotiana leaf epidermal cells. In addition,
we discovered that co-production of these proteins induced CD within 48 hours
regardless of the Agrobacterium tumefaciens strain used (i.e. GV3101::pMP90
or LBA4404) or the identity or presence of a tag attached to the MtNFP or MtLYK3
C-terminus (FP-tagged, 3xFLAG-tagged, and untagged protein fusions gave the same
response). The observed CD induction required simultaneous accumulation of MtNFP
and MtLYK3 or a homologous MtLYK2, and neither protein could be replaced
by a structurally unrelated RLK (MtDMI2, MtLRRII.1 or AtBRI1) for triggering
the CD response. Therefore, our results indicated functional interaction of MtNFP
with MtLYK3 and MtLYK2 for CD induction in Nicotiana heterologous system,
170
in agreements with the postulated co-functioning of MtNFP and MtLYK3 during IT
growth and functional redundancy of LYK proteins. Interestingly, heterologous
production of homologous AtCERK1, implicated in Arabidopsis thaliana innate
immunity in response to chitin/COs and PGN perception, in Nicotiana leaf led
to a phenotypically similar response. A detailed characterization of the Nicotiana
response to heterologous production of AtCERK1 and (co-)production of LysM-RLKs
from Medicago demonstrated that in both situations the CD induction required influx
of extracellular Ca2+
(as demonstrated with lanthanum chloride), and was associated
with accumulation of phenolic compounds and induction of a set of stress/defense-
related genes (NbHIN1, NbPR-1 basic, NbACRE31, and NbACRE132). Therefore,
our results suggested that MtNFP and MtLYK3 co-production apparently triggered
stress/defense signaling in Nicotiana leaf that resembled AtCERK1-induced defense
response and that culminated in CD development. In addition, we compared selected
requirements of nodulation and chitin signaling with CD induction in order to elucidate
whether the signaling triggered in Nicotiana leaf resembled biological activities of these
RLKs. We showed that CD induction in response to co-production of the symbiotic
LysM-RLKs was apparently independent of NF. On the contrary, MtNFP InR,
and MtLYK3 and AtCERK1 kinase activity were necessary for CD induction, thereby
mimicking the requirements for NF- and chitin-induced signaling, respectively.
As the results presented in Chapter 2 indicated a functional interaction between
MtNFP and MtLYK3 in a heterologous system of Nicotiana leaf, we subsequently
employed the Förster/Fluorescence Resonance Energy Transfer - Fluorescence Lifetime
Imaging Microscopy (FRET-FLIM) to investigate possible molecular interaction
of these LysM-RLKs in this plant system. We showed the tendency of MtLYK3
and (to a lesser extent) MtNFP to form homomers at the PM of Nicotiana leaf epidermal
171
cells that was independent of NF perception, in agreement with the dispensibility of NF
application for CD induction. Heteromerization of MtNFP and WT MtLYK3 could not
be studied due to the negative effect of the CD progression and associated
autofluorescence on the measurements of fluorescence lifetime. In order to circumvent
this obstacle, heteromerization between MtNFP and kinase-inactive MtLYK3
was analyzed. Heteromerization tendency of these proteins was apparently
less pronounced than their capacity to form homomers, and again was not affected
by the application of NF. In addition, Chapter 3 presents important quality measures
that allow to minimize various potential artifacts that could hamper FRET-FLIM
studies, especially: the effect of autofluorescence, unbalanced donor- and acceptor-
tagged protein levels, and bystander FRET.
Finally, we took an advantage of the functional interaction of MtNFP
and MtLYK3 in Nicotiana leaf for structure-function studies in which the CD induction
assay was used as a robust readout of their biological activities. In Chapter 4, detailed
analysis of CD requirements with respect to MtNFP structure is presented.
We demonstrated similar requirements for MtNFP biological activity in Medicago and
Nicotiana with respect to the structure of MtNFP InR, with nine out of eleven residues
equally dispensable, and one residue equally crucial for MtNFP function in vivo.
In addition, dissection of the MtNFP InR showed that the KD itself was necessary
for MtNFP signaling in Nicotiana, whereas the flanking sequences, i.e. a C-tail
and a juxtamembrane region were dispensable or not sufficient for CD induction,
respectively. The results presented in Chapter 4 suggested that a conserved fold of a KD
might be essential for MtNFP function as a pseudokinase, and that binding of an ATP
molecule by the MtNFP KD cannot be ruled out.
172
Chapter 5 presents a detailed structure-function study that aimed at
characterizing requirements of nodulation and CD induction with respect to MtLYK3
structure. We show that ten out of sixteen residues were equally important, and three
were equally dispensable for MtLYK3 biological activity in Medicago and Nicotiana,
indicating a significant similarity of MtLYK3 function in both plant systems.
Importantly, we demonstrated that 5 out of 7 symbiotic-important phosphorylation sites
found in the MtLYK3 InR were essential for CD induction, indicating that Nicotiana
leaf might be a relevant system for studying MtLYK3-mediated signaling. In addition,
our results suggest that the MtLYK3 ExR might exert an inhibitory role over MtLYK3
kinase, and that the PM localization of the MtLYK3 InR was crucial for its biological
activity in Nicotiana. Finally, we analyzed in vitro and in vivo activity of MtLYK3
variants in which a part of the MtLYK3 KD, specifically an activation segment,
was altered. We confirmed that autophosphorylation activity was indispensable
for MtLYK3 biological activity, and that in this respect, MtLYK3 differed from Pto
kinase.
173
REFERENCES
Amiour N, Recorbet G, Robert F, Gianinazzi S, Dumas-Gaudot E. (2006) Mutations in DMI3 and SUNN modify the appressorium-responsive
root proteome in arbuscular mycorrhiza. Mol. Plant Microbe Interact. 19: 988-997
Ané JM, Kiss GB, Riely BK, Penmetsa RV, Oldroyd GED, Ayax C, Lévy J, Debellé F, Baek JM, Kalo P, Rosenberg C, Roe BA, Long SR,
Dénarié J, Cook DR. (2004) Medicago truncatula DMI1 required for bacterial and fungal symbioses in legumes. Science, 303: 1364-1367
Ardourel M, Demont N, Debellé F, Maillet F, de Billy F, Promé JC, Dénarie J, Truchet G. (1994) Rhizobium meliloti lipooligosaccharide
nodulation factors: different structural requirements for bacterial entry into target root hair cells and induction of plant symbiotic developmental
responses. Plant Cell, 6: 1357-1374.
Arrighi JF, Barre A, Ben Amor B, Bersoult A, Soriano LC, Mirabella R, de Carvalho-Niebel F, Journet EP, Ghérardi M, Huguet T,
Geurts R, Dénarié J, Rougé P, Gough C. (2006) The Medicago truncatula Lysin [corrected] motif-Receptor-like kinase gene family includes
NFP and new nodule-expressed genes. Plant Physiol. 142: 265-279
Arrighi JF, Godfroy O, de Billy F, Saurat O, Jauneau A, Gough C. (2008) The RPG gene of Medicago truncatula controls Rhizobium-directed
polar growth during infection. Proc. Natl. Acad. Sci. USA, 15: 9817-9822
Assai S, Ohta K, Yoshioka H. (2008) MAPK signaling regulates nitric oxide and NADPH oxidase-dependent oxidative bursts in Nicotiana
benthamiana. Plant Cell, 20: 1390-1406
Baier R, Schiene K, Kohring B, Flaschel E, Niehaus K. (1999) Alfalfa and tobacco cells react differently to chitin oligosaccharides
and Sinorhizobium meliloti nodulation factors. Planta, 210: 157-164.
Banba M, Gutjahr C, Miyao A, Hirochika H, Paszkowski U, Kouchi H, Imaizumi-Anraku H. (2008) Divergence of evolutionary ways among
common sym genes: CASTOR and CCaMK show functional conservation between two symbiosis systems and constitute the root of a common
signalling pathway. Plant Cell Physiol. 49: 1659-1671
Bek AS, Sauer J, Thygesen MB, Duus JØ, Petersen BO, Thirup S, James E, Jensen KJ, Stougaard J, Radutoiu S. (2010) Improved
characterization of Nod factors and genetically based variation in LysM receptor domains identify amino acids expendable for Nod factor
recognition in Lotus spp. Mol. Plant Microbe Interact. 23: 58-66
Ben Amor B, Shaw SL, Oldroyd GED, Maillet F, Penmetsa RV, Cook, Long SR, Dénarie J, Gough C. (2003) The NFP locus of Medicago
truncatula controls an early step of Nod factor signal transduction upstream of a rapid calcium flux and root hair deformation. Plant J. 34: 495-506
Bensmihen S, de Billy F, Gough C. (2011) Contribution of NFP LysM domains to the recognition of Nod Factors during the Medicago
truncatula/Sinorhizobium meliloti symbiosis. PloS ONE, 6: e26114
Bersoult A, Camut S, Perhald A, Kereszt A, Kiss GB, Cullimore JV. (2005) Expression of the Medicago truncatula DMI2 gene suggests roles
of the symbiotic nodulation receptor kinase in nodules and during early nodule development. Mol. Plant Microbe Interact. 18: 869-876
Blanco FA, Meschini EP, Zanetti ME, Aguilar OM. (2009) A small GTPase of the Rab family is required for root hair formation
and preinfection stages of the common bean-Rhizobium symbiotic association. Plant Cell, 21: 2797-2810
Bleckmann A, Weidtkamp-Peters S, Seidel CAM, Simon R. (2009) Stem cell signaling in Arabidopsis requires CRN to localize CLV2
to the plasma membrane. Plant Physiol. 152: 166-176
Blilou I, Ocampo JA, Garcia-Garrido JM. (1999) Resistance of pea roots to endomycorrhizal fungus or Rhizobium correlates with enhanced
levels of endogenous salicylic acid. J. Exp. Bot. 50: 1663-1668.
Boller T, Felix G. (2009) A renaissance of elicitors: perception of microbe-associated molecular patterns and danger signals by pattern-recognition
receptors. Annu. Rev. Plant Biol. 60: 379-406
Bonfante P, Genre A. (2010) Mechanisms underlying beneficial plant-fungus interactions in mycorrhizal symbiosis. Nat. Commun. 1:
doi: 10.1038/ncomms1046
Bono JJ, Riond J, Nicolaou KC, Bockovich NJ, Estevez VA, Cullimore JV, Ranjeva R. (1995) Characterization of a binding site
for chemically synthesized lipo-oligosaccharidic NodRm factors in particulate fractions prepared from roots. Plant J. 7:253-260
Boudeau J, Miranda-Saavedra D, Barton G, Alessi DR. (2006) Emerging roles of pseudokinases. Trends Cell Biol. 16: 443-452
Boudsocq M, Willmann MR, McCormack M, Lee H, Shan L, He P, Bush J, Cheng SH, Sheen J. (2010) Differential innate immune
signalling via Ca2+ sensor protein kinases. Nature, 464: 418-422
Brentrup F, Palliere C. (2008) Nitrogen use efficiency as an Agro-environmental indicator. OECD materials.
Brewin NJ. (2004) Plant cell wall remodelling in the rhizobium-legume symbiosis. Crit. Rev. Plant Sci. 23: 293-316
Brotman Y, Lanau U, Pnini S, Lisec J, Balazadeh S, Mueller-Roeber B, Zilberstein A, Willmitzer L, Chet I, Viterbo A. (2012) The LysM
receptor-like kinase LysM RLK1 is required to activate defense and abiotic-stress responses induced by overexpression of fungal chitinases
in Arabidopsis plants. Mol. Plant, 5:1113-1124
Bueno P, Soto M.J, Rodríguez-Rosales M.P, Sanjuan J, Olivares J, Donaire JP. (2001) Time-course of lipoxygenase, antioxidant enzyme
activities and H2O2 accumulation during the early stages of Rhizobium-legume symbiosis. New Phytol. 152: 91-96
Capoen W, Goormachtig S, De Rycke R, Schroeyers K, Holsters M. (2005) SrSymRK, a plant receptor essential for symbiosome formation.
Proc. Natl. Acad. Sci. USA, 102: 10369-10374
174
Capoen W, Den Harder J, Sun J, Verplancke C, De Keyser A, De Rycke R, Goormachtig S, Oldroyd GED, Holsters M. (2009) Calcium
spiking patterns and the role of the calcium/calmodulin-dependent kinase CCaMK in lateral root base nodulation of Sesbania rostrata. Plant Cell,
21: 1526-1540
Capoen W, Oldroyd G, Goormachtig S, Holsters M. (2010) Sesbania rostrata: a case study of natural variation in legume nodulation.
New Phytol. 186: 340-345
Capoen W, Sun J, Wysham D, Otegui MS, Venkateshwaran M, Hirsch S, Miwa H, Downie JA, Morris RJ, Ané JM, Oldroyd GED. (2011)
Nuclear membranes control symbiotic calcium signaling of legumes. Proc. Natl. Acad. Sci. USA, 23: 14348-14353
Castells E, Casacuberta JM. (2007) Signaling through kinase-defective domains: the prevalence of atypical receptor-like kinases in plants.
J. Exp. Bot. 58: 3503-3511
Catoira R, Galera C, de Billy F, Penmetsa RV, Journet EP, Maillet F, Rosenberg C, Cook D, Gough C, Dénarié J. (2000) Four genes
of Medicago truncatula controlling components of a Nod Factor transduction pathway. Plant Cell, 12: 1647-1665
Catoira R, Timmers ACJ, Maillet F, Galera C, Penmetsa RV, Cook D, Dénarié J, Gough C. (2001) The HCL gene of Medicago truncatula
controls Rhizobium-induced root hair curling. Development, 128: 1507-1518
Cárdenas L, Holdaway-Clarke TL, Sánchez F, Quinto C, Feijó JA, Kunkel JG, Hepler PK. (2000) Ion changes in legume root hairs
responding to Nod factors. Plant Physiol. 123: 443-451
Cárdenas L, Martinez A, Sánchez F, Quinto C. (2008) Fast, transient and specific intracellular ROS changes in living root hair cells responding
to Nod factors (NFs). Plant J. 56: 802-813
Cárdenas L. (2009) New findings in the mechanisms regulating polar growth in root hair cells. Plant Signal. Behav. 4: 4-8
Chabaud M, Genre A, Sieberer BJ, Faccio A, Fournier J, Novero M, Barker DG, Bonfante P. (2011) Arbuscular mycorrhizal hyphopodia
and germinated spore exudates trigger Ca2+ spiking in the legume and nonlegume root epidermis. New Phytol. 189: 347-355
Charpentier M, Bredemeier R, Wanner G, Takeda N, Schleiff E, Parniske M. (2008) Lotus japonicus CASTOR and POLLUX are ion
channels essential for perinuclear calcium spiking in legume root endosymbiosis. Plant Cell, 20: 3467-3479
Charpentier M, Oldroyd G. (2010) How close are we to nitrogen-fixing cereals? Curr. Opin. Plant Biol. 13: 556-564
Charron D, Pingret JL, Chabaud M, Journet EP, Barker DG. (2004) Pharmacological evidence that multiple phospholipid signaling pathways
link rhizobium nodulation factor perception in Medicago truncatula root hairs to intracellular responses, including Ca2+ spiking and specific ENOD
gene expression. Plant Physiol. 136: 3582-3593
Chen C, Gao M, Liu J, Zhu H. (2007) Fungal symbiosis in rice requires an ortholog of a legume common symbiosis gene encoding
a Ca2+/calmodulin-dependent protein kinase. Plant Physiol. 145: 1619-1628
Chen C, Fan C, Gao M, Zhu H. (2009) Antiquity and function of CASTOR and POLLUX, the twin ion channel-encoding genes key
to the evolution of root symbioses in plants. Plant Physiol. 149: 306-317
Chen L, Hamada S, Fujiwara M, Zhu T, Thao NP, Wong H.L, Krishna P, Ueda T, Kaku H, Shibuya N, Kawasaki T, Shimamoto K. (2010)
The Hop/Sti1-Hsp90 chaperone complex facilitates the maturation and transport of a PAMP receptor in rice innate immunity. Cell Host Microbe,
7: 185-196
Clark SE, Williams RW, Meyerowitz EM. (1997) The CLAVATA1 gene encodes a putative receptor kinase that controls shoot and floral
meristem size in Arabidopsis. Cell, 89: 575-585
Cornelissen B, Horowitz J, van Kan JA, Goldberg RB, Bol JF. (1987) Structure of tobacco genes encoding pathogenesis-related proteins
from the PR-1 group. Nucl. Acid Res. 15: 6799-6811
Crespi M, Frugier F. (2008) De novo organ formation from dedifferentiated cells: root nodule organogenesis. Sci. Signal. 1: 1-8
Dawson CJ, Hilton J. (2011) Fertiliser availability in a resource-limited world: production and recycling of nitrogen and phosphorus.Food Policy,
doi:10.1016/j.foodpol.2010.11.012
Day RB, Shibuya N, Minami E. (2003) Identification and characterization of two members of the GRAS gene family in rice responsive
to N-acetylchitooligosaccharide elicitor. Biochim. Biophys. Acta 1625: 261-268
Deakin WJ, Broughton WJ. (2009) Symbiotic use of pathogenic strategies: rhizobial protein secretion systems. Nature, 7: 312-320
Deguchi Y, Banba M, Shimoda Y, Chechetka SA, Suzuri R, Okusako Y, Ooki Y, Toyokura K, Suzuki A, Uchiumi T, Higashi S, Abe M,
Kouchi H, Izui K, Hata S. (2007) Transcriptome profiling of Lotus japonicus roots during arbuscular mycorrhiza development and comparison
with that of nodulation. DNA Res. 14: 117-133
Demchenko K, Winzer T, Stougaard J, Parniske M, Pawlowski K. (2004) Distinct roles of Lotus japonicus SYMRK and SYM15 in root
colonization and arbuscule formation. New Phytol. 163: 381-392
den Hartog M, Musgrave A, Munnik T. (2001) Nod factor-induced phosphatidic acid and diacylglycerol pyrophosphate formation: a role
for phospholipase C and D in root hair deformation. Plant J. 25: 55-65
den Hartog M, Verhoef N, Munnik T. (2003) Nod factor and elicitors activate different phospholipid signaling pathways in suspension-cultured
alfalfa cells. Plant Physiol. 132: 311-317
Den Herder J, Vanhee C, De Rycke R, Corich V, Holsters M, Goormachtig S. (2007) Nod factor perception during infection thread growth
fine-tunes nodulation. Mol. Plant Microbe Interact. 20: 129-137
Den Herder G, Parniske M. (2009) The unbearable naivety of legumes in symbiosis. Curr. Opin. Plant Biol. 12: 491-499
175
Den Herder G, Yoshida S, Antolin-Llovra M, Ried MK, Parniske M. (2012) Lotus japonicus E3 ligase SEVEN IN ABSENTIA4 destabilizes
the Symbiosis Receptor-like Kinase SYMRK and negatively regulates rhizobial infection. Plant Cell, 24: 1691-1707
Desbrosses GJ, Stougaard J. (2011) Root nodulation: a paradigm for how plant-microbe symbiosis influences host developmental pathways.
Cell Host Microbe, 10: 348-358
D’Haeze W, Gao M, De Rycke R, Van Montagu M, Engler G, Holsters M. (1998) Roles for Azorhizobial Nod factors and surface
polysaccharides in intercellular invasion and nodule penetration, respectively. Mol. Plant Microbe Interact. 11: 999-1008
D’Haeze W, Mergaert P, Promé JC, Holsters M. (2000) Nod factor requirements for efficient stem and root nodulation of the tropical legume
Sesbania rostrata. J. Bio. Chem. 275: 15676-15684
D’Haeze W, Holsters M. (2002) Nod factor structures, responses, and perception during initiation of nodule development. Glycobiology, 12:
79R-105R
Dong J, Xiao F, Fan F, Gu L, Cang H, Martin GB, Chai J. (2009) Crystal structure of the complex between Pseudomonas effector AvrPtoB
and the tomato Pto kinase reveals both a shared and a unique interface compared with AvrPto-Pto. Plant Cell, 21: 1846-1859
Downie JA. (2010) The roles of extracellular proteins, polysaccharides and signals in the interactions of rhizobia with legume roots. FEMS
Microbiol Rev. 34: 150-170.
Downie JA, Surin BP. (1990) Either of two nod gene loci can complement the nodulation defect of a nod deletion mutant of Rhizobium
leguminosarum bv. viciae. Mol. Gen. Genet. 222: 81-86
Doyle JJ. (2011) Phylogenetic perspectives on the origins of nodulation. Mol. Plant Microbe Interact. 24: 1289-1295
Ehrhardt DW, Wais R, LongS.R. (1996) Calcium spiking in plant root hairs responding to rhizobium nodulation signals. Cell, 85: 673-681
El Yahyaoui F, Küster H, Ben Amor B, Hohnjec N, Pühler A, Becker A, Gouzy J, Vernié T, Gough C, Niebel A, Godiard L, Gamas P.
(2004) Expression profiling in Medicago truncatula identifies more than 750 genes differentially expressed during nodulation, including many
potential regulators of the symbiotic program. Plant Physiol. 136: 3159-3176
Endre G, Kereszt A, Kevel Z, Mihacea S, Kaló P, Kiss GB. (2002) A receptor kinase gene regulating symbiotic nodule development. Nature,
417: 962-966
Endres NF, Engel K, Das R, Kovacs E, Kuriyan J. (2011) Regulation of the catalytic activity of the EGF receptor. Curr. Opin. Struct. Biol. 21:
777-784
Erbs G, Silipo A, Aslam S, De Castro C, Liparoti V, Flagiello A, Pucci P, Lanzetta R, Parrilli M, Molinaro A, Newman MA, Cooper RM.
(2008) Peptidoglycan and muropeptides from pathogens Agrobacterium and Xanthomonas elicit plant innate immunity: structure and activity.
Cell, 15: 438-448
Esseling JJ, Lhuissier FGP, Emons AMC. (2003) Nod factor-induced root hair curling: continuous polar growth towards the point of Nod factor
application. Plant Physiol. 132: 1982-1988
Esseling JJ, Lhuissier FGP, Emons AMC. (2004) A nonsymbiotic root hair tip growth phenotype in NORK-mutated legumes: implications for
nodulation factor-induced signaling and formation of a multifaceted root hair pocket for bacteria. Plant Cell, 16: 933-944
Eswaran J, Patnaik D, Filippakopoulos P, Wang F, Stein RL, Murray JW, Higgins JMG, Knapp S. (2009) Structure and functional
characterization of the atypical human kinase haspin. Proc. Natl. Acad. Sci. USA, 106: 20198-20203
Ferguson B.J, Indrasumunar A, Hayashi S, Lin MH, Lin YH, Reid DE, Gresshoff PM. (2011) Molecular analysis of legume nodule
development and autoregulation. J. Integ. Plant Biol. 52: 61-76
Fliegmann J, Uhlenbroich S, Shinya T, Martinez Y, Lefebvre B, Shibuya N, Bono JJ. (2011) Biochemical and phylogenetic analysis
of CEBiP-like LysM domain-containing extracellular proteins in higher plants. Plant Physiol. Biochem. 49: 709-720
Fournier J, Timmers ACJ, Sieberer BJ, Jauneau A, Chabaud M, Barker DG. (2008) Mechanism of infection thread elongation in root hairs
of Medicago truncatula and dynamic interplay with associated rhizobial colonization. Plant Physiol. 148: 1985-1995
Frei dit Frey N, Mbengue M, Kwaaitaal M, Nitsch L, Altenbach D, Häweker H, Lozano-Duran R, Njo MF, Beeckman T, Huettel B,
Borst JW, Panstruga R, Robatzek S. (2012) Plasma membrane calcium ATPases are important components of receptor-mediated signaling in
plant immune responses and development. Plant Physiol. 159: 798-809
Frugier F, Kosuta S, Murray JD, Crespi M, Szczyglowski K. (2008) Cytokinin: secret agent of symbiosis. Trends Plant Sci. 13: 115-120
Gabriëls SHEJ, Takken FLW, Vossen JH, de Jong CF, Liu Q, Turk SCHJ, Wachowski LK, Peters J, Witsenboer HMA, de Wit PJGM,
Joosten MHAJ. (2006) cDNA-AFLP combined with functional analysis reveals novel genes involved in the hypersensitive response. Mol. Plant
Microbe Interact. 19: 567-576
Gage DJ. (2004) Infection and invasion of roots by symbiotic, nitrogen-fixing rhizobia during nodulation of temperate legumes. Microbiol. Mol.
Biol. Rev. 68: 280-300
Gao M, Wang X, Wang D, Xu F, Ding X, Zhang Z, Bi D, Cheng YT, Chen S, Li X, Zhang Y. (2009) Regulation of cell death and innate
immunity by two receptor-like kinases in Arabidopsis. Cell, 6: 34-44
Garcia-Brugger A, Lamotte O, Vandelle E, Bourque S, Lecourieux D, Poinssot B, Wendehenne D, Pugin A. (2006) Early signaling events
induced by elicitors of plant defenses. Mol. Plant Microbe Interact. 19: 711-724
Genre A, Chabaud M, Timmers T, Bonfante P, Barker DG. (2005) Arbuscular mycorrhizal fungi elicit a novel intracellular apparatus
in Medicago truncatula root epidermal cells before infection. Plant Cell, 17: 3489-3499
176
Genre A, Chabaud M, Faccio A, Barker DG, Bonfante P. (2008) Prepenetration apparatus assembly precedes and predicts the colonization
patterns of arbuscular mycorrhizal fungi within the root cortex of both Medicago truncatula and Daucus carota. Plant Cell, 20: 1407-1420
Genre A, Ortu G, Bertoldo C, Martino E, Bonfante P. (2009) Biotic and abiotic stimulation of root epidermal cells reveals common
and specific responses to arbuscular mycorrhizal fungi. Plant Physiol. 149: 1424-1434
Geurts R, Heidstra R, Hadri AE, Downie JA, Franssen H, van Kammen A, Bisseling T. (1997) Sym2 of pea is involved in a Nodulation factor
perception mechanism that controls the infection process in the epidermis. Plant Physiol. 115: 351-359
Gherbi H, Marknamm K, Svistoonoff S, Estevan J, Autran D, Giczey G, Auguy F, Péret B, Laplaze L, Franche C, Parniske M, Bogusz D.
(2008) SymRK defines a common genetic basis for plant root endosymbioses with arbuscular mycorrhiza fungi, rhizobia, and Frankia bacteria.
Proc. Natl. Acad. Sci. USA, 105: 4928-4932
Gibson KE, Kobayashi H, Walker GC. (2008) Molecular determinants of a symbiotic chronic infection. Annu. Rev. Genet. 42: 413-441
Gimenez-Ibanez S, Hann D.R, Ntoukakis V, Petutschnig E, Lipka V, Rathjen JP. (2009) AvrPtoB targets the LysM receptor kinase CERK1
to promote bacterial virulence on plants. Curr. Biol. 19: 423-429
Gish LA, Clark SE. (2011) The RLK/Pelle family of kinases. Plant J. 66: 117-127
Gleason C, Chaudhuri S, Yang T, Muñoz A, Poovaiah BV, Oldroyd GED. (2006) Nodulation independent of rhizobia induced by a calcium-
activated kinase lacking autoinhibition. Nature, 441: 1149-1152
Godfroy O, Debellé F, Timmers T, Rosenberg C. (2006) A rice calcium- and calmodulin-dependent protein kinase restores nodulation
to a legume mutant. Mol. Plant Microbe Interact. 19: 495-501
Goedhart J, Bono JJ, Bisseling T, Gadella Jr. TWJ. (2003) Identical accumulation and immobilisation of sulfated and nonsulfated Nod factors
in host and nonhost root hair cell walls. Mol. Plant Microbe Interact. 16: 884-892
Goedhart J, Vermeer JEM, Adjobo-Hermans MJW, van Weeren L, Gadella TWJ. (2007) Sensitive detection of p65 homodimers using red-
shifted and fluorescent protein-based FRET couples. PLoS ONE, 10: e1011
Goedhart J, van Weeren L, Hink MA, Vischer NO, Jalink K, Gadella TWJ. (2010) Bright cyan fluorescent protein variants identified
by fluorescence lifetime screening. Nat. Methods, 7: 137-139
Goedhart J, van Weeren L, Adjobo-Hermans MJW, Elzenaar I, Hink MA, Gadella TWJ. (2011) Quantitative co-expression of proteins
at the single cell level – application to a multimeric FRET sensor. PloS ONE, 6: e27321
Gomez-Gomez L, Boller T. (2000) FLS2: an LRR receptor-like kinase involved in the perception of the bactrial elicitor flagellin in Arabidopsis.
Mol. Cell, 5: 1003-1011
Gomez SK, Javot H, Deewatthanawong P, Torres-Jerez I, Tang Y, Blancaflor EB, Udvardi MK, Harrison MJ. (2009) Medicago truncatula
and Glomus intraradices gene expression in cortical cells harboring arbuscules in the arbuscular mycorrhizal symbiosis. BMC Plant Biology, 9: 10
Gonzalez-Rizzo S, Crespi M, Frugier F. (2006) The Medicago truncatula CRE1 cytokinin receptor regulates lateral root development and early
symbiotic interaction with Sinorhizobium meliloti. Plant Cell, 18: 2680-2693
Goormachtig S, Capoen W, James EK, Holsters M. (2004a) Switch from intracellular to intercellular invasion during water stress-tolerant
legume nodulation. Proc. Natl. Acad. Sci. USA, 101: 6303-6308
Goormachtig S, Capoen W, Holsters M. (2004b) Rhizobium infection: lessons from the versatile nodulation behaviour of water-tolerant
legumes. Trends Plant Sci. 9: 518-522
Gopalan S, Wei W, He SY. (1996) Hrp gene-dependent induction of hin1: a plant gene activated rapidly by both harpins and the avrPto
gene-mediated signal. Plant J. 10: 591-600
Gough C, Cullimore J. (2011) Lipo-chitooligosaccharide signaling in endosymbiotic plant-microbe interactions. Mol. Plant Microbe Interact. 24:
867-878
Gressent F, Mantegazza N, Cullimore JV, Driguez H, Ranjeva R, Bono JJ. (2002) High-affinity Nod factor binding site from Phaseolus
vulgaris cell suspension cultures. Mol. Plant Microbe Interact. 15: 834-839
Groppe J, Hinck CS, Samavarchi-Tehrani P, Zubieta C, Schuemann JP, Taylor AB, Schwarz PM, Wrana JL, Hinck AP. (2008)
Cooperative assembly of TGF-β superfamily signaling complexes is mediated by two disparate mechanisms and distinct modes of receptor
binding. Mol. Cell, 29: 157-168
Groth M, Takeda N, Perry J, Uchida H, Dräxl S, Brachmann A, Sato S, Tabata S, Kawaguchi M, Wang TL, Parniske M. (2010) NENA,
a Lotus japonicus homolog of Sec13, is required for rhizodermal infection by arbuscular mycorrhiza fungi and rhizobia but dispensable for cortical
endosymbiotic development. Plant Cell, 22: 2509-2526
Guo Y, Han L, Hymes M, Denver R, Clark SE. (2010) CLAVATA2 forms a distinct CLE-binding receptor complex regulating Arabidopsis
stem cell specification. Plant J. 63: 889-900
Gust AA, Biswas R, Lenz H.D, Rauhut T, Ranf S, Kemmerling B, Götz F, Glawischnig E, Lee J, Felix G, Nürnberger T. (2007) Bacteria-
derived peptidoglycans constitute pathogen-associated molecular patterns triggering innate immunity in Arabidopsis. J. Biol. Chem. 282: 32338-
32348
Güimil S, Chang HS, Zhu T, Sesma A, Osbourn A, Roux C, Ioannidis V, Oakeley EJ, Docquier M, Descombes P, Briggs SP, Paszkowski
U. (2005) Comparative transcriptomics of rice reveals an ancient pattern of response to microbial colonization. Proc. Natl. Acad. Sci. USA,
102:8066-8070
177
Haag AF, Baloban M, Sani M, Kerscher B, Pierre O, Farkas A, Longhi R, Boncompagni E, Hérouart D, Dall’Angelo S, Kondorosi E,
Zanda M, Mergaert P, Ferguson GP. (2011) Protection of Sinorhizobium against host cysteine-rich antimicrobial peptides is critical for
symbiosis. PLoS Biol. 9: e1001169
Hamel LP, Beaudoin N. (2010) Chitooligosaccharide sensing and downstream signaling: contrasted outcomes in pathogenic and beneficial plant-
microbe interactions. Planta, 232: 787-806
Haney CH, Long SR. (2010) Plant flotillins are required for infection by nitrogen-fixing bacteria. Proc. Natl. Acad. Sci. USA, 5: 478-483
Haney CH, Riely BK, Tricoli DM, Cook DR, Ehrhardt DW, Long SR. (2011) Symbiotic rhizobia bacteria trigger a change in localization
and dynamics of the Medicago truncatula receptor kinase LYK3. Plant Cell, 23: 2774-2787
Harris JM, Wais R, Long SR. (2003) Rhizobium-induced calcium spiking in Lotus japonicus. Mol. Plant Microbe Interact. 16: 335-341
Hayashi T, Banba M, Shimoda Y, Kouchi H, Hayashi M, Imaizumi-Anraku H. (2010) A dominant function of CCaMK in intracellular
accommodation of bacterial and fungal endosymbionts. Plant J. 63: 141-154
He K, Gou X, Yuan T, Lin H, Asami T, Yoshida S, Russel SD, Li J. (2007) BAK1 and BKK1 regulate brassinosteroid-dependent growth
and brassinosteroid-independent cell-death pathways. Curr. Biol. 17: 1109-1115.
Heath MC. (2000) Hypersensitive response-related death. Plant Mol. Biol. 44: 321-334
Heckmann AB, Lombardo F, Miwa H, Perry JA, Bunnewell S, Parniske M, Wang TL, Downie JA. (2006) Lotus japonicus nodulation
requires two GRAS domain regulators, one of which is functionally conserved in a non-legume. Plant Physiol. 142: 1739-1750
Heckmann AB, Sandal N, Bek AS, Madsen LH, Jurkiewicz A, Nielsen MW, Tirichine L, Stougaard J. (2011) Cytokinin induction of root
nodule primordia in Lotus japonicus is regulated by a mechanism operating in the root cortex. Mol. Plant Microbe Interact. 24: 1385-1395
Held M, Hossain M.S, Yokota K, Bonfante P, Stougaard J, Szczyglowski K. (2010) Common and not so common symbiotic entry. Trends
Plant Sci. 15: 540-545
Heldin CH. (1995) Dimerization of cell surface receptors in signal transduction. Cell, 80: 213-223
Hink MA, Shah K, Russinova E, de Vries SC, Visser AJW. (2008) Fluorescence fluctuation analysis of Arabidopsis thaliana Somatic
Embryogenesis Receptor-like Kinase and Brassinosteroid Insensitive 1 Receptor oligomerization. Biophys. J. 94: 1052-1062
Hirsch S, Kim J, Muñoz A, Heckmann AB, Downie JA, Oldroyd GED. (2009) GRAS proteins form a DNA binding complex to induce gene
expression during nodulation signaling in Medicago truncatula. Plant Cell, 21: 545-557
Holsters M. (2008) SYMRK, an enigmatic receptor guarding and guiding microbial endosymbioses with plant roots. Proc. Natl. Acad. Sci. USA,
105: 4537-4538
Horváth B, Yeun L.H, Domonkos Á, Halász G, Gobbato E, Ayaydin F, Miró K, Hirsch S, Sun J, Tadege M, Ratet P, Mysore KS, Ané JM,
Oldroyd GED, Kaló P. (2011) Medicago truncatula IPD3 is a member of the common symbiotic signaling pathway required for rhizobial
and mycorrhizal symbioses. Mol. Plant Microbe Interact. 24: 1345-1358
Høgslund N, Radutoiu S, Krusell L, Voroshilova V, Hannah MA, Goffard N, Sanchez DH, Lippold F, Ott T, Sato S, Tabata S, Liboriussen
P, Lohmann GV, Schauser GF, Weiller GF, Udvardi MK, Stougaard J. (2009) Dissection of symbiosis and organ development by integrated
transcriptome analysis of Lotus japonicus mutant and wild-type plants. PLoS ONE, 4: e6556
Huitema E, Vleeshouwers VG, Cakir C, Kamoun S, Govers F. (2005) Differences in intensity and specificity of hypersensitive response
induction in Nicotiana spp. by INF1, INF2A, and INF2B of Phytophthora infestans. Mol. Plant Microbe Interact. 18: 183-193
Humphreys CP, Franks PJ, Rees M, Bidartondo MI, Leake JR, Beerling DJ. (2010) Mutualistic mycorrhiza-like symbiosis in the most
ancient group of land plants. Nat. Commun. 1: 103
Huse M, Kuriyan J. (2002) The conformational plasticity of protein kinases. Cell, 109: 275-282
Iizasa E, Mitsutomi M, Nagano Y. (2010) Direct binding of a plant LysM receptor-like kinase, LysM RLK1/CERK1, to chitin in vitro.
J. Bio. Chem. 285: 2996-3004
Ivanchenko M, Vejlupkova Z, Quatrano RS, Fowler JE. (2000) Maize ROP7 GTPase contains a unique, CaaX box-independent plasma
membrane targeting signal. Plant J. 24: 79-90
Ivanov S, Fedorova E, Bisseling T. (2010) Intracellular microbe associations: secretory pathways and the formation of perimicrobial
compartments. Curr. Opin. Plant Biol. 13: 372-377
Ivanov S, Fedorova EE, Limpens E, De Mita S, Genre A, Bonfante P, Bisseling T. (2012) Rhizobium-legume symbiosis shares an exocytotic
pathway required for arbuscule formation. Proc. Natl. Acad. Sci. USA, 109: 8316-8321
Ivashuta S, Liu J, Liu J, Lohar DP, Haridas S, Bucciarelli B, VandenBosch KA, Vance CP, Harrison MJ, Gantt JS. (2005)
RNA interference identifies a calcium-dependent protein kinase involved in Medicago truncatula root development. Plant Cell, 17: 2911-2921
Jaillais Y, Belkhadir Y, Balsemão-Pires E, Dangl JL, Chory J. (2011a) Extracellular leucine-rich repeats as a platform for receptor/coreceptor
complex formation. Proc. Natl. Acad. Sci. USA, 108: 8503-8507
Jaillais Y, Hothorn M, Belkhadir Y, Dai T, Nimchuk ZL, Meyerowitz EM, Chory J. (2011b) Tyrosine phosphorylation controls
brassinosteroid receptor activation by triggering membrane release of its kinase inhibitor. Genes Dev. 25: 232-237
Jares-Erijman EA, Jovin TM. (2003) FRET imaging. Nat. Biotech. 21: 1387-1395
178
Jeworutzki E, Roelfsema MRG, Anschütz U, Krol E, Elzenga JTM, Felix G, Boller T, Hendrich R, Becker D. (2010) Early signaling
through the Arabidopsis pattern recognition receptors FLS2 and EFR involves Ca2+-associated opening of plasma membrane anion channels.
Plant J. 62: 367-378
Jones JD, Dangl JL. (2006) The plant immune system. Nature, 444: 323-329
Jones KM, Walker GC. (2008) Responses of the model legume Medicago truncatula to the rhizobial exopolysaccharide succinoglycan.
Plant Signal. Behav. 3: 888-890
Kaku H, Nishizawa Y, Ishii-Minami N, Akimoto-Tomiyama C, Dohmae N, Takio K, Minami E, Shibuya N. (2006) Plant cells recognize
chitin fragments for defense signaling through a plasma membrane receptor. Proc. Natl. Acad. Sci. USA, 18: 11086-11091
Kaló P, Gleason C, Edwards A, Marsh J, Mitra RM, Hirsch S, Jakab J, Sims S, Long SR, Rogers J, Kiss GB, Downie JA. Oldroyd GED.
(2005) Nodulation signaling in legumes requires NSP2, a member of the GRAS family of transcriptional regulators. Science, 308: 1786-1789
Kamoun S, van Weest P, de Jong A.J, de Groot KE, Vleeshouwers VG, Govers F. (1997) A gene encoding a protein elicitor of Phytophthora
infestans is down-regulated during infection of potato. Mol. Plant Microbe Interact. 10: 13-20
Kamoun S, van Weest P, Vleeshouwers VG, de Groot KE, Govers F. (1998) Resistance of Nicotiana benthamiana to Phytophthora infestans
is mediated by the recognition of the elicitor protein INF1. Plant Cell, 10, 1413-1426
Kanamori N, Madsen LH, Radutoiu S, Frantescu M, Quistgaard EMH, Miwa H, Downie JA, James EK, Felle HH, Haaning LL, Jensen
TH, Sato S, Nakamura Y, Tabata S, Sandal N, Stougaard J. (2006) A nucleoporin is required for induction of Ca2+ spiking in legume nodule
development and essential for rhizobial and fungal symbiosis. Proc. Natl. Acad. Sci. USA, 103: 359-364
Karas B, Murray J, Gorzelak M, Smith A, Sato S, Tabata S, Szczyglowski K. (2005) Invasion of Lotus japonicus root hairless 1
by Mesorhizobium loti involves the nodulation factor-dependent induction of root hairs. Plant Physiol. 137: 1331-1344
Karlova R, Boeren S, Russinova E, Aker J, Vervoort J, de Vries S. (2006) The Arabidopsis Somatic Embryogenesis Receptor-like Kinase 1
protein complex includes Brassinosteroid Insensitive 1. Plant Cell, 18: 626-638.
Kelly MN, Irving HR. (2003) Nod factors activate both heteromeric and monomeric G-proteins in Vigna unguicuata (L.) Walp. Planta, 216:
674-685
Kereszt A, Mergaert P, Maróti G, Kondorosi E. (2011) Innate immunity effectors and virulence factors in symbiosis. Curr. Opin. Microbiol. 14:
76-81
Kevei Z, Lougnon G, Mergaert P, Horvath GV, Kereszt A, Jayaraman D, Zaman N, Marcel F, Regulski K, Kiss GB, Kondorosi A, Endre
G, Kondorosi E, Ané JM. (2007) 3-hydroxy-3-methylglutaryl coenzyme A reductase1 interacts with NORK and is crucial for nodulation
in Medicago truncatula. Plant Cell, 19: 3974-3989
Kishi-Kaboshi M, Okada K, Kurimoto L, Murakami S, Umezawa T, Shibuya N, Yamane H, Miyao A, Takatsuji H, Takahashi A,
Hirochika H. (2010) A rice fungal MAMP-responsive cascade regulates metabolic flow to antimicrobial metabolite synthesis. Plant J. 63: 599-612
Kishimoto K, Kouzai Y, Kaku H, Shibuya N, Minami E, Nishizawa Y. (2010) Perception of the chitin oligosaccharides contributes to disease
resistance to blast fungus Magnaporthe oryzae in rice. Plant J. 64: 343-354
Kiss E, Oláh B, Kaló P, Morales M, Heckmann AB, Borbola A, Lózsa A, Kontár K, Middleton P, Downie JA, Oldroyd GE, Endre G.
(2009) LIN, a novel type of U-box/WD40 protein, controls early infection by rhizobia in legumes. Plant Physiol. 151: 1239-1249
Kistner C, Parniske M. (2002) Evolution of signal transduction in intracellular symbiosis.Trends Plant Sci. 7: 511-518
Kistner C, Winzer T, Pitzschke A, Mulder L, Sato S, Kaneko T, Tabata S, Sandal N, Stougaard J, Webb KJ, Szczyglowski K, Parniske M.
(2005) Seven Lotus japonicus genes required for transcriptional reprogramming of the root during fungal and bacterial symbiosis. Plant Cell, 17:
2217-2229
Klaus-Heisen D, Nurisso A, Pietraszewska-Bogiel A, Mbengue M, Camut S, Timmers T, Pichereaux C, Rossignol M, Gadella TWJ,
Imberty A, Cullimore JV. (2011) Structure-function similarities between a plant receptor-like kinase and the human Interleukin-1 Receptor-
Associated Kinase-4. J. Biol. Chem. 286: 11202-11210
Kornev AP, Taylor SS, Eyck LFT. (2008) A helix scaffold for the assembly of active protein kinases. Proc. Natl. Acad. Sci. USA, 105:
14377-14382
Kornev AP, Taylor SS. (2009) Pseudokinases: functional insights gleaned from structure. Structure, 17: 5-7
Kornev AP, Taylor SS. (2010) Defining the conserved internal architecture of a protein kinase. Bioch. Biophys. Acta, 1840: 440-444
Kosuta S, Hazledine S, Sun J, Miwa H, Morris R.J, Downie JA, Oldroyd GED. (2008) Differential and chaotic calcium signatures
in the symbiosis signaling pathway of legumes. Proc. Natl. Acad. Sci. USA, 105: 9823-9828
Kosuta S, Held M, Hossain MS, Morieri G, MacGillivary A, Johansen C, Antolín-Llovera M, Parniske M, Oldroyd GED, Downie AJ,
Karas B, Szczyglowski K. (2011) Lotus japonicus symRK-14 uncouples the cortical and epidermal symbiotic program. Plant J. 67: 929-940
Kouchi H, Shimomura K, Hata S, Hirota A, Wu G-J, Kumagai H, Tajima S, Suganuma N, Suzuki A, Aoki T, Hayashi M, Yokoyama T,
Ohyama T, Asamizu E, Kuwata C, Shibata D, Tabata S. (2004) Large-scale analysis of gene expression profiles during early stages of root
nodule formation in a model legume, Lotus japonicus. DNA Res. 11: 263-274
Krebs M, Held K, Binder A, Hashimoto K, Den Herder G, Parniske M, Kudla J, Schumacher K . (2012) FRET-based genetically encoded
sensors allow high-resolution live cell imaging of Ca2+ dynamics. Plant J. 69: 181-192
179
Kremers G.J., Goedhart J., van Munster E.B., and Gadella T.W.Jr. (2006) Cyan and yellow super fluorescent proteins with improved
brightness, protein folding, and FRET Förster radius. Biochemistry, 45, 6570-6580
Kuhn H, Küster H, and Requena N. (2010) Membrane steroid-binding protein 1 induced by a diffusible fungal signal is critical for
mycorrhization in Medicago truncatula. New Phytol. 185: 716-733
Lange J, Xie Z.P, Broughton WJ, Vögeli-Lange R, Boller T. (1999) A gene encoding a receptor-like protein kinase in the roots of common
bean is differentially regulated in response to pathogens, symbionts and nodulation factors. Plant Sci. 142: 133-145
Lee CG, Da Silva CA, Lee JY, Hartl D, Elias JA. (2008) Chitin regulation of immune responses: an old molecule with new roles. Curr. Opin.
Immun. 20: 684-689
Lee YJ, Yang Z. (2008) Tip growth: signaling in the apical dome. Curr. Opin. Plant Biol. 11: 662-671
Lefebvre B, Timmers T, Mbengue M, Moreau S, Hervé C, Tóth K, Bittencourt-Silvestre J, Klaus D, Deslandes L, Godiard L, Murray JD,
Udvardi MK, Raffaele S, Mongrand S, Cullimore J, Gamas P, Niebel A, Ott T. (2010) A remorin protein interacts with symbiotic receptors
and regulates bacterial infection. Proc. Natl. Acad. Sci. USA, 2: 2343-2348
Lefebvre B, Klaus-Heisen D, Pietraszewska-Bogiel A, Hervé C, Camut S, Auriac M-C, Gasciolli V, Nurisso A, Gadella TWJ, Cullimore J.
(2012) Role of N-glycosylation sites and CXC motifs in trafficking of Medicago truncatula Nod Factor Perception protein to the plasma
membrane. J. Biol. Chem. 287: 10812-10823
Lemmon MA, Schlessinger J. (2010) Cell signaling by receptor tyrosine kinases. Cell, 141: 1117-1134
Lenardon MD, Munro CA, Gow NAR. (2010) Chitin synthesis and fungal pathogenesis. Curr. Opin. Microbiol. 13: 416-423
Lew ED, Furdui CM, Anderson KS, Schlessinger J. (2009) The precise sequence of FGF receptor autophosphorylation is kinetically driven
and is disrupted by oncogenic mutations. Sci: Signal. 2: ra6
Lévy J, Bres C, Geurts R, Chalhoub B, Kulikova O, Duc G, Journet EP, Ané JM, Lauber E, Bisseling T, Dénarié J, Rosenberg C, Debellé
F. (2004) A putative Ca2+ and calmodulin-dependent protein kinase required for bacterial and fungal symbioses. Science, 303: 1361-1364
Li J, Chory J. (1997) A putative leucine-rich repeat kinase involved in brassinosteroid signal transduction. Cell, 90, 929-938
Li J, Wen J, Lease KA, Doke JT, Tax FE, Walker JC. (2002) BAK1, an Arabidopsis LRR receptor-like protein kinase, interacts with BRI1
and modulates brassinosteroid signalling. Cell, 110:213-222
Libault M, Wan J, Czechowski T, Udvardi M, Stacey G. (2007) Identification of 118 Arabidopsis transcription factor and 30 ubiquitin-ligase
genes responding to chitin, a plant defense elicitor. Mol. Plant Microbe Interact. 20: 900-911
Limpens E, Franken C, Smit P, Willemse J, Bisseling T, Geurts R. (2003) LysM domain receptor kinases regulating rhizobial Nod factor-
induced infection. Science, 302: 630-633
Limpens E, Mirabella R, Fedorova E, Franken C, Franssen H, Bisseling T, Geurts R. (2005) Formation of organelle-like N2-fixing
symbiosomes in legume root nodules is controlled by DMI2. Proc. Natl. Acad. Sci. USA, 102: 10375-10380
Liu J, Blaylock LA, Endre G, Cho J, Town CD, VandenBosch KA, Harrison MJ. (2003) Transcript profiling coupled with spatial expression
analyses reveals genes involved in distinct developmental stages of an arbuscular mycorrhizal symbiosis. Plant Cell, 15:2106-2123
Liu W, Chen A.M, Luo L, Sun J, Cao L.P, Yu GQ, Zhu JB, Wang YZ. (2010) Characterization and expression analysis of Medicago
truncatula ROP GTPase family during the early stage of symbiosis. J. Integr. Plant Biol. 52:639-652
Liu B, Li JF, Ao Y, Qu J, Li Z, Su J, Zhang Y, Liu J, Feng D, Qi K, He Y, Wang J, Wang HB. (2012a) Lysin motif-containing proteins LYP4
and LYP6 play dual roles in peptidoglycan and chitin perception in rice innate immunity. Plant Cell, 24: 3406-3419
Liu T, Liu Z, Song C, Hu Y, Han Z, She J, Fan F, Wang J, Jin C, Chang J, Zhou JM, Chai J. (2012b) Chitin-induced dimerization activates
a plant immune receptor. Science, 336: 1160-1164
Lohar DP, Sharopova N, Endre G, Peñuela S, Samac D, Town C, Silverstein KAT, VandenBosch KA. (2006) Transcript analysis of early
nodulation events in Medicago truncatula. Plant Physiol. 140: 221-234
Lohar DP, Haridas S, Gantt JS, VandenBosch KA. (2007) A transient decrease in reactive oxygen species in roots leads to root hair
deformation in the legume-rhizobia symbiosis. New Phytol. 173: 39-49
Lohmann GV, Shimoda Y, Nielsen MW, Jørgensen FG, Grossmann C, Sandal N, Sørensen K, Thirup S, Madsen L.H, Tabata S, Sato S,
Stougaard J, Radutoiu S. (2010) Evolution and regulation of the Lotus japonicus LysM receptor gene family. Mol. Plant Microbe Interact. 23:
510-521
Madsen EB, Madsen LH, Radutoiu S, Olbryt M, Rakwalska M, Szczyglowski K, Sato S, Kaneko T, Tabata S, Sandal N. Stougaard J.
(2003) A receptor kinase gene of the LysM type is involved in legume perception of rhizobial signals. Nature, 425: 637-640
Madsen LH, Tirichine L, Jurkiewicz A, Sullivan JT, Heckmann AB, Bek AS, Ronson CW, James EK, Stougaard J. (2010) The molecular
network governing nodule organogenesis and infection in the model legume Lotus japonicus. Nature, 1: doi: 10.1038/ncomms1009
Madsen E.B, Antolin-Llovera M, Grossmann C, Ye J, Vieweg S, Broghammer A, Krussel L, Radutoiu S, Jensen O.N, Stougaard J,
Parniske M. (2011) Autophosphorylation is essential for the in vivo function of the Lotus japonicus Nod factor receptor 1 and receptor-mediated
signalling in cooperation with Nod factor receptor 5. Plant J. 65: 404-417
Maillet F, Poinsot V, André O, Peuch-Pagès V, Haouy A, Gueunier M, Cromer L, Giraudet D, Formey D, Niebel A, Martinez EA, Driguez
H, Bécard G, Dénarié J. (2011) Fungal lipochitooligosaccharide symbiotic signals in arbuscular mycorrhiza. Nature, 469:58-64
180
Manzoor H, Chiltz A, Madani S, Vatsa P, Schoefs B, Pugin A, Garcia-Brugger A. (2012) Calcium signatures and signaling in cytosol
and organelles of tobacco cells induced by plant defense elicitors. Cell Calcium, 51: 434-444.
Markmann K, Giczey G, Parniske M. (2008) Functional adaptation of a plant receptor-kinase paved the way for the evolution of intracellular
root symbioses with bacteria. PLoS Biol. 6: 497-506
Martinez-Abarca F, Herrera-Cervera J.A, Bueno P, Sanjuan J, Bisseling T, Olivares J. (1998) Involvement of salicylic acid
in the establishment of he Rhizobium meliloti-alfalfa symbiosis. Mol. Plant Microbe Interact. 11, 153-155
Marsh JF, Rakocevic A, Mitra R.M, Brocard L, Sun J, Eschstruth A, Long SR, Scchultze M, Ratet P, Oldroyd GED. (2007) Medicago
truncatula NIN is essential for rhizobial-independent nodule organogenesis induced by autoactive calcium/calmodulin-dependent protein kinase.
Plant Physiol. 144: 324-335
Masson-Boivin C, Giraud E, Perret X, Batut J. (2009) Establishing nitrogen-fixing symbiosis with legumes: how many rhizobium recipies?
Trends Microbiol. 17: 458-466
Massoumou M, van Tuinen D, Chatagnier O, Arnold C, Brechenmacher L, Sanchez L, Selim S, Gianinazzi S, Gianinazzi-Pearson V.
(2007) Medicago truncatula gene responses specific to arbuscular mycorrhiza interactions with different species and genera of Glomeromycota.
Mycorrhiza, 17: 223-234.
Maunory N, Redondo-Nieto M, Bourcy M, Van de Velde W, Alunni B, Laporte P, Durand P, Agier N, Marisa L, Vaubert D, Delacroix H,
Duc G, Ratet P, Aggerbeck L, Kondorosi E, Mergaert P. (2010) Differentiation of symbiotic cells and endosymbionts in Medicago truncatula
nodulation are coupled to two transcriptome-switches. PLoS ONE, 5: e9519
Mbengue M, Camut S, de Carvalho-Niebel F, Deslandes L, Froidure S, Klaus-Heisen D, Moreau S, Rivas S, Timmers T, Hervé C,
Cullimore J, Lefebvre B. (2010) The Medicago truncatula E3 ubiqutin ligase PUB1 interacts with the LYK3 symbiotic receptor and negatively
regulates infection and nodulation. Plant Cell, 22: 3474-3488
Merlin S, Pietronave S, Locarno D, Valente G, Follenzi A, Prat M. (2009) Deletion of the ectodomain unleashes the transforming, invasive,
and tumorigenic potential of the MET oncogene. Cancer Sci. 100: 633-638
Messinese E, Mun JH, Yeun L.H. Jayaraman D, Rougé P, Barre A, Lougnon G, Schornack S, Bono JJ, Cook DR, Ané JM. (2007) A novel
nuclear protein interacts with the symbiotic DMI3 calcium- and calmodulin-dependent proten kinase of Medicago truncatula. Mol. Plant Microbe
Interact. 20: 912-921
Middleton PH, Jakab J, Penmetsa RV, Starker CG, Doll J, Kaló P, Prabhu R, Marsh JF, Mitra RM, Kereszt A, Dudas B, VandenBosch
K, Long SR, Cook DR, Kiss GB, Oldroyd GED. (2007) An ERF transcription factor in Medicago truncatula that is essential for Nod Factor
signal transduction. Plant Cell, 19: 1221-1234
Millet YA, Danna CH, Clay NK, Songnuan W, Simon MD, Werck-Reichart D, Ausubel FM. (2010) Innate immune responses activated
in Arabidopsis roots by microbe-associated molecular patterns. Plant Cell, 22: 973-990
Mitra RM, Shaw SL, Long SR. (2004a) Six nonnodulating plant mutants defective for Nod factor-induced transcriptional changes associated
with the legume-rhizobia symbiosis. Proc. Natl. Acad. Sci. USA, 101: 10217-10222
Mitra R.M, Gleason C.A, Edwards A, Hadfield J, Downie J.A, Oldroyd G.E.D. & Long S.R. (2004b), A Ca2+/calmudulin-dependent protein
kinase required for symbiotic nodule development: gene identification by transcript-based cloning. Proc. Natl. Acad. Sci. USA, 101: 4701-4705
Miwa H, Sun J, Oldroyd GED, Downie JA. (2006a) Analysis of calcium spiking using cameleon calcium sensor reveals that nodulation gene
expression is regulated by calcium spike number and the developmental status of the cell. Plant J. 48: 883-894
Miwa H, Sun J, Oldroyd GE, Downie JA. (2006b) Analysis of Nod-factor-induced calcium signaling in root hairs of symbiotically defective
mutants of Lotus japonicus. Mol. Plant Microbe Interact. 19: 914-923
Miya A, Albert P, Shinya T, Desaki Y, Ichimura K, Shirasu K, Narasuka Y, Kawakami N, Kaku H, Shibuya N. (2007) CERK1, a LysM
receptor kinase, is essential for chitin elicitor signalling in Arabidopsis. Proc. Natl. Acad. Sci. USA, 104: 19613-19618
Miyahara A, Richens J, Starker C, Morieri G, Smith L, Long S, Downie JA, Oldroyd GED. (2010) Conservation in function
of a SCARE/WAVE component during infection thread and root hair growth in Medicago truncatula. Mol. Plant Microbe Interact. 23: 1553-1562
Monahan-Giovanelli H, Pinedo CA, Gage J. (2006) Architecture of infection thread networks in developing root nodules induced
by the symbiotic bacterium Sinorhizobium meliloti on Medicago truncatula. Plant Physiol. 140: 661-670
Morandi D, Prado E, Sagan M, Duc G. (2005) Characterization of new symbiotic Medicago truncatula (Gaertn.) mutants, and phenotypic
or genotypic complementary information on previously described mutants. Mycorrhiza, 15: 283-289
Mortier V, Holsters M, Goormachtig S. (2012) Never too many? How legumes control nodule numbers. Plant Cell Environ. 35: 245-258
Mucyn TS, Clemente A, Andriotis VME, Balmuth AL, Oldroyd GED, Staskawicz BJ, Rathjen JP. (2006) The tomato NBARC-LRR protein
Prf interacts with Pto kinase in vivo to regulate specific plant immunity. Plant Cell, 18, 2792-2806
Mukherjee K, Sharma M, Urlaub H, Bourenkov GP, Jahn R, Südhof TC, Wahl MC. (2008) CASK functions as a Mg2+ -independent
neurexin kinase. Cell, 133: 328-339
Mulder L, Lefebvre B, Cullimore J, Imberty A. (2006) LysM domains of Medicago truncatula NFP protein involved in Nod factor perception.
Glycosylation state, molecular modeling and docking of chitooligosaccharides and Nod factors. Glycobiology, 16: 801-809
Van Munster EB, Gadella TW. Jr. (2004a) Suppression of photobleaching-induced artifacts in frequency-domain FLIM by permutation
of the recording order. Cytometry, 58: 185-194.
181
Van Munster EB, Gadella TW. Jr. (2004b) phiFLIM: a new method to avoid aliasing in frequency-domain fluorescence lifetime imaging
microscopy. J. Microsc. 213: 29-38
Mur LA, Kenton P, Lloyd AJ, Ougham H., Prats E. (2008) The hypersensitive response; the centenary is upon us but how much do we know?
J. Exp. Bot. 59, 501-520
Murakami Y, Miwa H, Imaizumi-Anraku H, Kouchi H, Downie JA, Kawaguchi M, Kawasaki S. (2006) Positional cloning identifies Lotus
japonicus NSP2, a putative transcription factor of the GRAS family, required for NIN and ENOD40 gene expression in nodule initiation.
DNA Res. 13: 255-265
Murray JD, Karas BJ, Sato S, Tabata S, Amyot L, Szczyglowski K. (2007) A cytokinin perception mutant colonized by Rhizobium
in the absence of nodule organogenesis. Science, 315: 101-104
Murray JD. (2011) Invasion by invitation: rhizobial infection in legumes. Mol. Plant Microbe Interact. 24: 631-639
Murray JD, Muni RR, Torres-Jerez I, Tang Y, Allen S, Adriankaja M, Li G, Laxmi A, Cheng X, Wen J, Vaughan D, Schultze M, Sun J,
Charpentier M, Oldroyd G, Tadege M, Ratet P, Mysore KS, Chen R, Udvardi M.K. (2011) Vapyrin, a gene essential for intracellular
progression of arbuscular mycorrhizal symbiosis, is also essential for infection by rhizobia in the nodule symbiosis of Medicago truncatula.
Plant J. 65: 244-252
Nakagawa T, Kaku H, Shimoda Y, Sugiyama A, Shimamura M, Takanashi K, Yazaki K, Aoki T, Shibuya N, Kouchi H, (2010)
From defense to symbiosis: limited alterations in the kinase domain of LysM receptor-like kinases are crucial for evolution of legume-Rhizobium
symbiosis. Plant J. 65: 169-180
Navarro L, Zipfel C, Rowland O, Keller I, Robatzek S, Boller T, Jones JDG. (2004) The transcriptional innate immune response to flg22.
Interplay and overlap with Avr gene-dependent defense responses and bacterial pathogenesis. Plant Physiol. 135: 1113-1128
Nguyen HP, Chakravarthy S, Velásquez AC, McLane H.L, Zeng L, Nakayashiki H, Park DH, Collmer A, Martin GB. (2010) Methods
to study PAMP-triggered immunity using tomato and Nicotiana benthamiana. Mol. Plant Microbe Interact. 23: 991-999
Nicaise V, Roux M, Zipfel C. (2009) Recent advances in PAMP-triggered immnity against bacteria: pattern recognition receptors watch over
and raise the alarm. Plant Physiol. 150: 1638-1647
Nolen B, Taylor S, Ghosh G. (2004) Regulation of protein kinases: controlling activity through activation segment conformation. Mol. Cell, 15:
661-675
Oh CS, Martin GB. (2010) Effector-triggered immunity mediated by the Pto kinase. Trends Plant Sci. 16: 1360-1385
Oh M.H, Wang X, Wu X, Zhao Y, Clouse SD, Hubet SC. (2010) Autophosphorylation of Tyr-610 in the receptor kinase BAK1 plays a role
in brassinosteroid signaling and basal defense gene expression. Proc. Natl. Acad. Sci. USA, 107: 17827-17832
Oh MH, Wang X, Clouse SD, Huber SC. (2012) Deactivation of the Arabidopsis Brassinosteroid Insensitive 1 (BRI1) receptor kinase
by autophosphorylation within the glycine-rich loop. Proc. Natl. Acad. Sci. USA, 109: 327-332
Oláh B, Brière C, Bécard G, Dénerié J, Gough C. (2005) Nod factors and a diffusible factor from arbuscular mycorrhizal fungi stimulate lateral
root formation in Medicago truncatula via the DMI1/DMI2 signalling pathway. Plant J. 44: 195-207
Oldroyd GED, Mitra RM, Wais RJ, Long SR. (2001) Evidence for structurally specific negative feedback in the Nod factor signal transduction
pathway. Plant J. 28: 191-199
Oldroyd GED, Shaw SR. (2003) Identification and characterization of Nodulation-Signaling Pathway 2, a gene of Medicago truncatula involved
in Nod factor signaling. Plant Physiol. 131: 1027-1032
Oldroyd GED, Downie JA. (2006) Nuclear calcium changes at the core of symbiosis signalling. Curr. Opin. Plant Biol. 9: 351-357
Oldroyd GED, Downie JA. (2008) Coordinating nodule morphogenesis with rhizobial infection in legumes. Annu. Rev. Plant Biol. 59: 519-546
Oldroyd GED, Murray JD, Poole PS, Downie JA. (2011) The rules of engagement in legume-rhizobial symbiosis. Annu. Rev. Genet. 45: 119-
144
van Ooijen G, Mayr G, Kasiem MM, Albrecht M, Cornelissen BJ, Takken FLW. (2008) Structure-function analysis of the NB-ARC domain
of plant disease resistance proteins. J. Exp. Bot. 59: 1383-1397
Op den Camp R, Streng A, De Mita S, Cao Q, Polone E, Liu W, Ammiraju JSS, Kudrna D, Wing R, Untergasser A, Bisseling T, Geurts R.
(2011) LysM-type mycorrhizal receptor recruited for rhizobium symbiosis in nonlegume Parasponia. Science, 331: 909-912
Ovchinnikova E, Journet EP, Chabaud M, Cosson V, Ratet P, Duc G, Fedorova E, Liu W, Op den Camp R, Zhukov V, Tikhonovich I,
Borisov A, Bisseling T, Limpens E. (2011) IPD3 controls the formation of nitrogen-fixing symbiosomes in pea and Medicago. Mol. Plant
Microbe Interact. 24: 1333-1344
Parniske M, Zimmermann C, Cregan PB, Werner D. (1990) Hypersensitive reaction of nodule cells in the Glycine sp./Bradyrhizobium
japonicus-symbiosis occurs at the genotype-specific level. Bot. Acta, 103: 143-148
Parniske M. (2008) Arbuscular mycorrhiza: the mother of plant root endosymbioses. Nature, 6: 763-775
Peleg-Grossman S, Volpin H, Levine A. (2007) Root hair curling and Rhizobium infection in Medicago truncatula are mediated
by phosphatidylinositide-regulated endocytosis and reactive oxygen species. J. Exp. Bot. 58: 1637-1649
Peiter E, Sun J, Heckmann AB, Venkateshwaran M, Riely BK, Otegui MS, Edwards A, Freshour G, Hahn MG, Cook DR, Sanders D,
Oldroyd GED, Downie JA, Ané JM. (2007) The Medicago truncatula DMI1 protein modulates cytosolic calcium signaling. Plant Physiol. 145:
192-203
182
Peleg-Grosmann S., Golani Y., Kaye Y., Melamed-Book N., and Levine A. (2009) NPR1 protein regulates pathogenic and symbiotic
interactions between Rhizobium and legumes and non-legumes. PLoS ONE, 4: e8399
Perotto S, Brewin NJ, Kannenberg EL. (1994) Cytological evidence for a host defence response that reduces cell and tissue invasion in pea
nodules by lipopolysaccharide-defective mutants of Rhizobium leguminosarum strain 3841. Mol. Plant Microbe Interact. 7: 99-112
Petutschnig EK, Jones AME, Serazetdinova L, Lipka U, Lipka V. (2010) The Lysin Motif receptor-like kinase (LysM-RLK) CERK1
is a major chitin-binding protein in Arabidopsis thaliana and subject to chitin-induced phosphorylation. J. Bio. Chem. 285: 28902-28911
Pietraszewska-Bogiel A. Gadella TWJ. (2010) FRET microscopy: from principle to routine technology in cell biology. J. Microsc. 241: 111-118
Plet J, Wasson A, Ariel F, Le Signor C, Baker D, Mathesius U, Crespi M, Frugier F. (2011) MtCRE1-dependent cytokinin signalling
integrates bacterial and plant cues to coordinate symbiotic nodule organogenesis in Medicago truncatula. Plant J. 65: 622-633
Popp C, Ott T. (2011) Regulation of signal transduction and bacterial infection during root nodule symbiosis. Curr.Opin.Plant Biol. 14: 458-467.
Radutoiu S, Madsen LH, Madsen EB, Felle HH, Umehara Y, Grønlund M, Sato S, Nakamura Y, Tabata S, Sandal N, Stougaard J. (2003)
Plant recognition of symbiotic bacteria requires two LysM receptor-like kinases. Nature, 425: 585-592
Radutoiu S, Madsen LH, Madsen EB, Jurkiewicz A, Fukai E, Quistgaard EMH, Albrektsen AS, James EK, Thirup S, Stougaard J. (2007)
LysM domains mediate lipochitin-oligosaccharide recognition and NFR genes extend the symbiotic host range. EMBO J. 26: 3923-3935
Ramonell KM, Zhang B, Ewing RM, Chen Y, Xu D, Stacey G, Somerville S. (2002) Microarray analysis of chitin elicitation in Arabidopsis
thaliana. Mol. Plant Pathol. 3: 301-311
Ramonell K, Berrocal-Lobo M, Koh S, Wan J, Edwards H, Stacey G, Somerville S. (2005) Loss-of-function mutations in chitin responsive
genes show increased susceptibility to the powdery mildew pathogen Erysiphe cichoracearum. Plant Physiol. 138: 1027-1036
Ramu SK, Peng HM, Cook DR. (2002) Nod factor induction of reactive oxygen species production is correlated with expression of the early
nodulin gene rip1 in Medicago truncatula. Mol. Plant Microbe Interact. 15: 522-528
Ranf S, Eschen-Lippold L, Pecher P, Lee J, Scheel D. (2011) Interplay between calcium cignalling and early signaling elements during defence
responses to microbe- or damage-associated molecular patterns. Plant J. 68: 100-113
Rathjen JP, Chang JH, Staskawicz B, Michelmore RW. (1999) Constitutively active Pto induces a Prf-dependent hypersensitive response
in the absence of avrPto. EMBO J. 18:3232-3240
Riely BK, Lougnon G, Ané JM, Cook DR. (2007) The symbiotic ion channel homolog DMI1 is localized in the nuclear membrane of Medicago
truncatula roots. Plant J. 49: 208-216
Riely BK, He H, Venkateshwaran M, Sarma B, Schraiber J, Ané JM, Cook DR. (2011) Identification of legume RopGEF gene families
and characterization of a Medicago truncatula RopGEF mediating polar growth of root hairs. Plant J. 56: 230-243
Roux M, Schwessinger B, Albrecht C, Chinchilla D, Jonex A, Holton N, Malinovsky FG, Tör M, de Vries S, Zipfel C. (2011)
The Arabidopsis Leucine-rich repeat receptor-like kinases BAK1/SERK3 and BKK1/SERK4 are required for innate immunity to hemibiotrophic
and biotrophic pathogens. Plant Cell, 23: 2440-2455
Russinova E, Borst JW, Kwaaitaal M, Canõ-Delgado A, Yin Y, Chory J, de Vries SC. (2004) Heteromerization and endocytosis
of Arabidopsis brassinosteroid receptors BRI1 and AtSERK3 (BAK1). Plant Cell, 16: 3216-3229
Saito K, Yoshikawa M, Yano K, Miwa H, Uchida H, Asamizu E, Sato S, Tabata S, Imaizumi-Anraku H, Umehara Y, Kouchi H, Murooka
Y, Szczyglowski K, Downie JA, Parniske M, Hayashi M, Kawaguchi M. (2007) NUCLEOPORIN85 is required for calcium spiking, fungal
and bacterial symbioses, and seed production in Lotus japonicus. Plant Cell, 19: 610-624
Sanchez L, Weidmann S, Arnould C, Bernard AR, Gianinazzi S, Gianinazzi-Pearson V. (2005) Pseudomonas fluorescens and Glomus
mosseae trigger DMI3-dependent activation of genes related to a signal transduction pathway in roots of Medicago truncatula. Plant Physiol. 139:
1065-1077.
Schauser L, Roussis A, Stiller J, Stougaard J. (1999) A plant regulator controlling development of symbiotic root nodules. Nature, 402: 191-195
Schlaman HRM, Horvath B, Vijgenboom E, Okker RJH, Lugtenberg BJJ. (1991) Suppression of nodulation gene expression in bacteroids
of Rhizobium leguminosarum biovar viciae. J. Bacteriol. 173: 4277-4287
Schulze B, Mentzel T, Jehle AK, Mueller K, Beeler S, Boller T, Felix G, Chinchilla D. (2010) Rapid heteromerization and phosphorylation
of ligand-activated plant transmembrane receptors and their associated kinase BAK1. J. Biol. Chem. 285: 9444-9451
Schwessinger B, Roux M, Kadota Y, Ntoukakis V, Sklenar J, Jones A, Zipfel C. (2011) Phosphorylation-dependent differential regulation
of plant growth, cell death, and innate immunity by the regulatory receptor-like kinase BAK1. PLoS Genet. 7:e1002046
Seddas PMA, Arias CM, Arnould C, van Tuinen D, Godfroy O, Benhassou HA, Gouzy J, Morandi D, Dessaint F, Gianinazzi-Pearson V.
(2009) Symbiosis-related plant genes modulate molecular responses in an arbuscular mycorrhizal fungus during early root interactions. Mol. Plant
Microbe Interact. 22: 341-351.
Segonzac C, Feike D, Gimenez-Ibanez S, Hann DR, Zipfel C, Rathjen JP. (2011) Hierarchy and roles of pathogen-associated molecular
pattern-induced responses in Nicotiana benthamiana. Plant Physiol. 156: 687-699
Serna-Sanz A, Parniske M, Peck SC. (2011) Phosphoproteome analysis of Lotus japonicus roots reveals shared and distinct components
of symbiosis and defense. Mol. Plant Microbe Interact. 24: 932-937
Sessa G, D’Ascenzo M, Martin GB. (2000) Thr38 and Ser 198 are Pto autophosphorylation sites required for the AvrPto-Pto-mediated
hypersensitive response. EMBO J. 19, 2257-2269
183
Shah K, Gadella Jr. TWJ, van Erp H, Hecht V, de Vries SC. (2001) Subcellular localization and oligomerization of the Arabidopsis thaliana
Somatic Embryogenesis Receptor Kinase 1 protein. J. Mol. Biol. 309: 641-655
Shaner NC, Campbell RE, Steinbach PA, Giepmans BN, Palmer AE, Tsien RY. (2004) Improved monomeric red, orange and yellow
fluorescent proteins derived from Discosoma sp. red fluorescent protein. Nature, 12: 1567-1572
Shaw SL, Long SR (2003) Nod Factor elicits two separable calcium responses in Medicago truncatula root hair cells. Plant Physiol. 131: 976-984
Shi Z, Resing KA, Ahn NG. (2006) Networks for the allosteric control of protein kinases. Curr. Opin. Struct. Biol. 16: 686-692
Shi F, Telesco SE., Liu Y, Radhakrishanan R, Lemmon MA. (2010) ErbB3/HER3 intracellular domain is competent to bind ATP and catalyze
autophosphorylation. Proc. Natl. Acad. Sci. USA, 107: 7692-7697
Shibuya N, Minami E. (2001) Oligosaccharide signaling for defence responses in plant. Physiol. Mol. Plant Pathol. 59: 223-233
Shimizu T, Nakano T, Takamizawa D, Desaki Y, Ishii-Minami N, Nishizawa Y, Minami E, Okada K, Yamane H, Shibuya N. (2010)
Two LysM receptor molecules, CEBiP and OsCERK1, cooperatively regulate chitin elicitor signaling in rice. Plant J. 64: 204-214
Shiu SH, Bleecker AB. (2001) Receptor-like kinases from Arabidopsis form a monophyletic gene family related to animal receptor kinases.
Proc. Natl. Acad. Sci. USA, 98: 10763-10768
Shiu SH, Bleecker AB. (2003) Expansion of the Receptor-like kinase/Pelle gene family and Receptor-like proteins in Arabidopsis. Plant Physiol.
132: 530-543
Shiu SH, Karlowski WM, Pan R, Tzeng YH, Mayer KF, Li WH. (2004) Comparative analysis of the receptor-like kinase family in Arabidopsis
and rice. Plant Cell, 16: 1220-1234
Siciliano V, Genre A, Balestrini R, Cappellazzo G, de Wit PJGM, Bonfante P. (2007) Transcriptome analysis of arbuscular mycorrhizal roots
during development of the prepenetration apparatus. Plant Physiol. 144: 1455-1466
Sieberer BJ, Chabaud M, Timmers AC, Monin A, Fournier J, Barker DC. (2009) A nuclear-targeted cameleon demonstrates intranuclear Ca2+
spiking in Medicago truncatula root hairs in response to rhizobial Nodulation factors. Plant Physiol. 151: 1197-1206.
Sieberer BJ, Chabaud M, Fournier J, Timmers AC, Barker DC. (2012) A switch in Ca2+ spiking signature is concomitant with endosymbiotic
microbe entry into cortical root cells of Medicago truncatula. Plant J. 69: 822-830
Simsek S, Ojanen-Reuhs T, Stephens SB, Reuhs BL. (2007) Strain-ecotype specificity in Sinorhizobium meliloti-Medicago truncatula
symbiosis is correlated to succinoglycan oligosaccharide structure. J. Bacteriol. 189: 7733-7740
Smit P, Raedts J, Portyanko V, Debellé F, Gough C, Bisseling T, Geurts R. (2005) NSP1 of the GRAS protein family is essential for rhizobial
Nod Factor-induced transcription. Science, 308: 1789-1791
Smit P, Limpens E, Geurts R, Fedorova E, Dolgikh E, Gough C, Bisseling T. (2007) Medicago LYK3, an entry receptor in rhizobial
nodulation factor signaling. Plant Physiol. 145: 183-191
Son G.H, Wan J, Kim HJ, Nguyen XC, Chung WS, Hong JC, Stacey G. (2012) Ethylene-responsive Element-binding Factor 5, ERF5,
is involved in chitin-induced innate immunity response. Mol. Plant Microbe Interact. 25: 48-60
Sprent JI. (2007) Evolving ideas of legume evolution and diversity: a taxonomic perspective on the occurrence of nodulation. New Phytol. 174:
11-25
Stacey G, Libault M, Brechenmacher L, May GD. (2006) Genetics and functional genomics of legume nodulation. Curr. Opin. Plant Biol. 9:
110-121.
Stracke S, Kistner C, Yoshida S, Mulder L, Sato S, Kaneko T, Tabata S, Sandal N. Stougaard J, Szczyglowski K, Parnise M. (2002)
A plant receptor-like kinase required for both bacterial and fungal symbiosis. Nature, 417: 959-962
Streng A, Op den Camp R, Bisseling T, Geurts R. (2011) Evolutionary origin of Rhizobium Nod factor signaling. Plant Signal. Behav. 6:
1510-1514
Sun W., Cao Y., Labby K.J., Bittel P., Boller T., and Bent A.F. (2012) Probing the Arabidopsis Flagellin Receptor: FLS2-FLS2 association
and the contributions of specific domains to signaling function. Plant Cell, 24: 1096-1113
Taguchi F, Suzuki T, Takeuchi K, Inagaki Y, Toyoda K, Shiraishi T, Ichinose Y. (2009) Glycosylation of flagellin from Pseudomonas
syringae pv. tabaci 6605 contributes to evasion of host tobacco plant surveillance system. Physiol Mol. Plant Pathol. 74: 11-17
Takeda N, Haage K, Sato S, Parniske M. (2011) Activation of a Lotus japonicus subtilase gene during arbuscular mycorrhiza is dependent
on the common symbiosis genes and two cis-active promoter regions. Mol. Plant Microbe Interact. 24: 662-670
Taylor SS, Kornev AP. (2011) Protein kinases: evolution of dynamic regulatory proteins. Trends Biochem. Sci. 36, 65-77
Thomma BPHJ, Nürnberger T, Joosten MHAJ. (2011) Of PAMPs and effectors: the blurred PTI-ETI dichotomy. Plant Cell, 23: 4-15
Timmers ACJ, Auriac MC, de Billy F, Truchet G. (1998) Nod factor internalization and microtubular cytoskeleton changes occur
concomitantly during nodule differentiation in alfalfa. Development, 125: 339-349
Timmers ACJ, Auriac MC, Truchet G. (1999) Refined analysis of early symbiotic steps of the Rhizobium-Medicago interaction in relationship
with microtubular cytoskeleton rearrangements. Development, 126: 3617-3628
Timmers ACJ. (2008) The role of the plant cytoskeleton in the interaction between legumes and rhizobia. J. Microsc. 231: 247-256
Tirichine L, Imaizumi-Anraku H, Yoshida S, Murakami Y, Madsen LH, Miwa H, Nakagawa T, Sandal N, Albrektsen AS, Kawaguchi M,
Downie A, Sato S, Tabata S, Kouchi H, Parniske M, Kawasaki S, Stougaard J. (2006) Deregulation of a Ca2+/calmodulin-dependent kinase
leads to spontaneous nodule development. Nature, 441: 1153-1156
184
Tirichine L, Sandal N, Madsen LH, Radutoiu S, Albrektsen AS, Sato S, Asamizu E, Tabata S, Stougaard J. (2007) A gain-of-function
mutation in a cytokinin receptor triggers spontaneous root nodule organogenesis. Science, 315: 104-107
Tóth K, Stratil TF, Madsen EB, Ye J, Popp C, Antolín-Llovera M, Grossmann C, Jensen ON, Schüssler A, Parniske M, Ott T. (2012)
Functional domain analysis of the remorin protein LjSYMREM1 in Lotus japonicus. PloS ONE, 7:e30817
Udvardi MK, Kakar K, Wendrey M, Montanari O, Murray J, Adriankaja A, Zhang JY, Benedito V, Hofer JMI, Chueng F, Town CD.
(2007) Legume transcription factors: global regulators of plant development and response to the environment. Plant Physiol. 144: 538-549
Vasse J., de Billy F. and Truchet G. (1993) Abortion of infection during the Rhizobium meliloti-alfalfa symbiotic interaction is accompanied
by a hypersensitive reaction. Plant J. 4: 555-566
Vernié T, Moreau S, de Billy F, Plet J, Combier JP, Rogers C, Oldroyd G, Frugier F, Niebel A, Gamas P. (2008) EFD is an ERF
transcription factor involved in the control of nodule number and differentiation in Medicago truncatula. Plant Cell, 20: 2696-2713
Wais RJ, Galera C, Oldroyd G, Catoira R, Penmetsa RV, Cook D, Gough C, Dénarié J, Long SR. (2000) Genetic analysis of calcium spiking
responses in nodulation mutants of Medicago truncatula. Proc. Natl. Acad. Sci. USA, 97: 13407-13412
Wais RJ, Keating DH, Long SR. (2002) Structure-function analysis of Nod factor-induced root hair calcium spiking in Rhizobium-legume
symbiosis. Plant Physiol. 129: 211-224
Walker SA, Downie JA. (2000) Entry of Rhizobium leguminosarum bv. viciae into root hairs requires minimal Nod factor specificity,
but subsequent infection thread growth requires nodO or nodE. Mol. Plant Microbe Interact. 13: 754-762
Wan J, Zhang S, Stacey G. (2004) Activation of a mitogen-activated protein kinase pathway in Arabidopsis by chitin. Mol. Plant Pathol. 5:
125-135
Wan J, Torres M, Ganapathy A, Thelen J, DaGue BB, Mooney B, Xu D, Stacey G. (2005) Proteomic analysis of soybean root hairs after
infection by Bradyrhizobium japonicum. Mol. Plant Microbe Interact. 18: 458-467
Wan J, Zhang XC, Neece D, Ramonell KM, Clough S, Kim S, Stacey MG, Stacey G. (2008) A LysM receptor-like kinase plays a critical role
in chitin signalling and fungal resistance in Arabidopsis. Plant Cell, 20: 471-481
Wang X, Li X, Meisenhelder J, Hunter T, Yoshida S, Asami T, Chory J. (2005) Autoregulation and homomerization are involved
in the activation of the plant steroid receptor BRI1. Dev. Cell, 8: 855-865
Wang Z, Liu J, Sudom A, Ayres M, Li S, Wesche H, Powers JP, Walker NP. (2006) Crystal structures of IRAK-4 kinase in complex with
inhibitors: a serine/threonine kinase with tyrosine as a gatekeeper. Structure, 14: 1835-1844
Wang X, Kota U, He K, Blackburn K, Li J, Goshe MB, Huber SC, Clouse SD. (2008) Sequential transphosphoylation of the BRI1/BAK1
receptor kinase complex impacts early events in brassinosteroid signaling. Cell, 15: 220-235
Wang B, Yeun LH, Liu JY, Ané JM, Qiu YL. (2010) Presence of three mycorrhizal genes in the common ancestor of land plants suggests a key
role of mycorrhizas in the colonization of land by plants. New Phytol. 186: 514-525
Ward CW, Lawrence MC. (2012) Similar but different: ligand-induced activation of the insulin and epidermal growth factor receptor families.
Curr. Opin. Struct. Biol., 22: 1-7
Wegel E, Schauser L, Sandal N, Stougaard J, Parniske M. (1998) Mycorrhiza mutants of Lotus japonicus define genetically independent steps
during symbiotic infection. Mol. Plant Microbe Interact. 11: 933-936
Weidmann S, Sanchez L, Descombin J, Chatagnier O, Gianinazzi S, Gianinazzi-Pearson V. (2004) Fungal elicitation of signal transduction-
related plant genes precedes mycorrhiza establishment and requires the dmi3 gene in Medicago truncatula. Mol. Plant Microbe Interact. 17:
1385-1393
Willmann R, Lajunen HM, Erbs G, Newman MA, Kolb D, Tsuda K, Katagiri F, Fliegmann J, Bono JJ, Cullimore JV, Jehle AK, Götz F,
Kulik A, Molinaro A, Lipka V, Gust AA, Nürnberger T. (2011) Arabidopsis lysin-motif proteins LYM1 LYM3 and CERK1 mediate bacterial
peptidoglycan sensing and immunity to bacterial infection. Proc. Natl. Acad. Sci. USA, 108: 19824-19829
Wu AJ, Andriotis VME, Durrant MC, Rathjen JP. (2004) A patch of surface-exposed residues mediates negative regulation of immune
signaling by tomato Pto kinase. Plant Cell, 16: 2809-2821
Wu S, Lu D, Kabbage M, Wei HL, Swingle B, Records AR, Dickman M, He P, Shan L. (2011) Bacterial effector HopF2 suppresses
Arabidopsis innate immunity at the plasma membrane. Mol. Plant Microbe Interact. 24: 585-593
Xiang T, Zong N, Zou Y, Wu Y, Zhang J, Xing W, Li Y, Tang X, Zhu L, Chai J, Zhou JM. (2008) Pseudomonas syringae effector AvrPto
blocks innate immunity by targeting receptor kinases. Curr. Biol. 18: 74-80
Xing W, Zou Y, Liu Q, Liu J, Luo X, Huang Q, Chen S, Zhu L, Bi R, Hao Q, Wu JW, Zhou JM, Chai J. (2007) The structural basis
for activation of plant immunity by bacterial effector protein AvrPto. Nature, 449: 243-247
Xu B, English JM, Wilsbacher JL, Stippec S, Goldsmith EJ, Cobb MH. (2000) WNK1, a novel mammalian serine/threonine protein kinase
lacking the catalytic lysine in subdomain II. J. Biol. Chem. 275: 16795-16801
Yang J, Wu J., Steichen JM, Kornev AP, Deal MS, Li S, Sankaran B, Woods Jr. V.L., Taylor SS. (2012) A conserved Glu-Arg salt bridge
connects coevolved motifs that define the eukaryotic protein kinase fold. J. Mol. Biol. 415: 666-679
Yano K, Yoshida S, Müller J, Singh S, Banba M, Vickers K, Markmann K, White C, Schuller B, Sato S, Asamizu E, Tabata S, Murooka
Y, Perry J, Wang TL, Kawaguchi M, Imaizumi-Anraku H, Hayashi M, Parniske M. (2008) CYCLOPS, a mediator of symbiotic intracellular
accommodation. Proc. Natl. Acad. Sci. USA, 105: 20540-20545
185
Yano K, Shibata S, Chen WL, Sato S, Kaneko T, Jurkiewicz A, Sandal N, Banba M, Imaizumi-Anraku H, Kojima T, Ohtomo R,
Szczyglowski K, Stougaard J, Tabata S, Hayashi M, Kouchi H, Umehara Y. (2009) CERBERUS, a novel U-box protein containing WD-40
repeats, is required for formation of the infection thread and nodule development in the legume-Rhizobium symbiosis. Plant J. 60: 168-180
Yokota K, Fukai E, Madsen LH, Jurkiewicz A, Rueda P, Radutoiu S, Held M, Hossain MS, Szczyglowski K, Morieri G, Oldroyd GED,
Downie JA, Nielsen MW, Rusek AM, Sato S, Tabata S, James EK, Oyaizu H, Sandal N, Stougaard J. (2009) Rearrangement of actin
cytoskeleton mediates invasion of Lotus japonicus roots by Mesorhizobium loti. Plant Cell, 21: 267-284
Yokota K, Soyano T, Kouchi H, Hayashi M. (2010) Function of GRAS proteins in root nodule symbiosis is retained in homologs
of a non-legume, rice. Plant Cell Physiol. 51: 1436-1442
Yoshida S, Parniske M. (2005) Regulation of plant symbiosis receptor kinase (SymRK) through serine and threonine phosphorylation.
J. Bio. Chem. 275: 8779-8784
Young ND et al. (2011) The Medicago genome provides insight into the evolution of rhizobial symbioses. Nature, 480: 520-524
Zegiraj E, Filippi BM, Deak M, Alessi DR, van Aalten DMF. (2009) Structure of the LKB1-STRAD-MO25 complex reveals an allosteric
mechanism of kinase activation. Science, 326: 1707-1711
Zegiraj E, Aalten DMF. (2010) Pseudokinases-remnants of evolution or key allosteric regulators. Curr. Opin. Struct. Biol. 20: 772-781
Zeng L, Velásquez AC, Munkvold KR, Zhang J, Martin GB. (2012) A tomato LysM receptor-like kinase promotes immunity and its kinase
activity is inhibited by AvrPtoB. Plant J. 69: 92-103
Zhang B, Ramonell K, Somerville S, Stacey G. (2002) Characterization of early, chitin-induced gene expression in Arabidopsis. Mol. Plant
Microbe Interact. 15: 963-970
Zhang X, Gureasko J, Shen K, Cole PA, Kuriyan J. (2006) An allosteric mechanism for activation of the kinase domain of epidermal growth
factor receptor. Cell, 125, 1137-1149
Zhang XC, Wu X, Findley S, Wan J, Libault M, Nguyen HT, Cannon SB, Stacey G. (2007) Molecular evolution of lysine motif-type receptor-
like kinases in plants. Plant Physiol. 144: 623-636
Zhang J, Subramanian S, Stacey G, Yu O. (2009a) Flavones and flavonols play distinct critical roles during nodulation of Medicago truncatula
by Sinorhizobium meliloti. Plant J, 57: 171-183
Zhang XC, Cannon SB, Stacey G. (2009b) Evolutionary genomics of LysM genes in land plants. BMC Evol. Biol. 9: 183
Zhang P, Smith-Nguyen E.V, Keshwani MM, Deal MS, Kornev AP, Taylor SS. (2012) Structure and allostery of the PKA RIIβ tetrameric
holoenzyme. Science, 335: 712-716
Zhu H, Chen T, Zhu M, Fang Q, Kang H, Hong Z, Zhang Z. (2008) A novel ARID DNA-binding protein interacts with SymRK
and is expressed during early nodule development in Lotus japonicus. Plant Physiol. 148: 337-347
Zhu Y, Wang Y, Li R, Song X, Wang Q, Huang S, Jin J.B, Liu CM, Lin J. (2009) Analysis of interactions among the CLAVATA3 receptors
reveals a direct interaction between CLAVATA2 and CORYNE in Arabidopsis. Plant J. 61: 223-233
Zipfel C, Kunze G, Chinchilla D, Caniard A, Jones JDG, Boller T, Felix G. (2006) Perception of the bacterial PAMP EF-Tu by the receptor
EFR restricts Agrobacterium-mediated transformation. Cell, 125: 749-760
186
187
ACKNOWLEDGMENTS
Our attitude towards life and faith in personal capabilities are largely
dependent on the people closest to us. Therefore, I would first like to thank all my
family, and especially my parents and my husband. Dziękuję Wam za to, że zawsze
we mnie wierzycie i pozwalacie mi iść własną drogą – tym bardziej, że posiadanie
powsinogi za dziecko i pracoholika za partnera życiowego nie może być łatwe. Bratku –
doceniam wszystkie awantury o to, żebym już poszła spać (chociaż nadal czytam
po nocach). Kocham Was.
Secondly, I would like to show my gratitude to two people who made
the project presented in this thesis, and my Amsterdam adventure possible. Dorus, thank
you for giving me the opportunity to join your group and to work in this stimulating
environment. I have surely learnt a lot and enjoyed my work here, although maybe
sometimes your approach towards students was somewhat unorthodox. But it is usually
only when we look back that we see how much we have grown - and for this I thank
you. Julie, many thanks not only for the enormous work of co-ordinating the work
between various research groups, but also for the fruitful collaboration. And since
we are on a subject of the Marie Curie „NodPerception“ project, I would like to thank
all the people involved in it, officially or less officially. Special thanks go to Dr. René
Geurts, Dr. Dörte Klaus-Heisen, and Dr. Benoit Lefebvre for the fruitfull collaboration.
Two Ale’s, Edo, Giulia, Malick, Maria – you sparkled the network meetings not only
with inspiring discussions but also with a lot of fun! Thanks!
I own a big THANK YOU to all past and present members of the Molecular
Cytology group, whom I had a pleasure to work with but especially to: Laura – thanks
for your help in the lab, translator service, and great dining company, among other
188
things. Joachim – you always had time to answer my „quick questions“, and I have not
heard too many „well, what do you think?“ from you, so for both things – thank you!
Mark – thanks for your interest in the project, and an introduction to powerful
microscopy techniques. Eric – thanks for passing on some of your expertise
in microscopy. Merel and Joop – thanks for clearing the path, helpful tips,
and companionship. Ronald – thanks for the timely support and climbing opportunities
;) Jolanda – I appreciated the good vibes and stimulating beat in the lab. Linda – thanks
for the „while I am gone“ cloning projects, cookies, and company after hours. Fred –
your straightforwardness always amazes me but also sets a new standard to strive for.
Ania i Aneta – za dawke czarnego humoru każdego dnia i za wytyczanie nowych
szlaków. Inge – I am glad you were my first student and even more glad that you
survived ;) And last but not least, I would like to thank Maria and Kevin – guys, I do not
even know where to begin .....
Big thank you goes also to my „greener“ colleagues in Plant Phytopathology
and Plant Physiology groups, and especially to: Ewy i obu Magd –
the Polish connection ;) Lotje – for inspiring attitude. Frank – for setting a record
in correcting an article over night, and for many helpful discussions. Harrold and
Martijn – for helpful tips and discussions. Eleni – finishing thesis to the smoothing
tunes outside the window was just perfect. Big thanks to you and Mark! Alex –
you made mine and Thomas lives so much easier here last year, I will not forget it. Petra
B., Fleur, Rossana, Ale, and Carlos – for your interest, tips, and words of motivation.
Petra H. and Ringo – for helping out at the time of lab goodies crises. Moreover,
I would like to thank Ludek and Harold for taking care of plants in the greenhouse,
and Laura and Casper for helping with the administration matters.
THANK YOU
189
CURRICULUM VITAE
Anna Pietraszewska-Bogiel was born on second November 1981 in Warsaw,
Poland. In 2005, she finished a five-year Master course in Plant Biotechnology
at Warsaw Agricultural University in Poland. During the course of her studies,
she participated in the Erasmus Student Exchange Programme that allowed her to enrol
as a student at Gent University in Belgium. During this half a year, she contributed
to the research of Dr. L. de Veylder group, focusing on transcription factors implicated
in cell cycle regulation in plant cells. For her Master project, she joined Dr M. Filipecki
group at the Faculty of Biotechnology, Warsaw Agricultural University, where she
studied transcription factors essential for plant embryogenesis using a model of somatic
embryogenesis in cucumber. From September 2005 until December 2006,
she contributed as a research assistant to the work of Dr. C. Spillane group
at Biochemistry Department in College Cork, Ireland. During this time, she investigated
the possibility of triggering asexual reproduction in Arabidopsis thaliana. In December
2006, she was appointed a PhD student in Prof. Dr. Th.W.J. Gadella group at Molecular
Cytology, Swammerdam Institute for Life Sciences and Netherlands Institute
for Systems Biology, University of Amsterdam, The Netherlands, and an Early
Scientific Researcher in the EC Marie Curie Research Training Network Programme
(through „NodPerception“ contract). For her PhD project, she studied beneficial
symbiosis between a model legume plant, Medicago truncatula, and nitrogen-fixing
Sinorhizobium meliloti. More specifically, she focused on deciphering the signaling
mechanisms employed by two putative Nodulation factor receptors of Medicago using
various fluorescence microscopy techniques and structure-function assays.
190
List of publications
Pietraszewska-Bogiel A, Gadella TWJ. (2010) FRET microscopy: from principle
to routine technology in cell biology. J. Microsc. 241: 111-118
Crosby KC, Pietraszewska-Bogiel A, Gadella TW Jr, Winkel BS. (2011) Förster
resonance energy transfer demonstrates a flavonoid metabolon in living plant cells
that displays competitive interactions between enzymes. FEBS Lett. 585: 2193-2198
Klaus-Heisen D, Nurisso A, Pietraszewska-Bogiel A, Mbengue M, Camut S, Timmers
T, Pichereaux C, Rossignol M, Gadella TWJ, Imberty A, Cullimore JV. (2011)
Structure-function similarities between a plant receptor-like kinase and the human
Interleukin-1 Receptor-Associated Kinase-4. J. Biol. Chem. 286: 11202-11210
Lefebvre B, Klaus-Heisen D. Pietraszewska-Bogiel A, Hervé C, Camut S, Auriac M-C,
Gasciolli V, Nurisso A, Gadella TWJ, Cullimore J. (2012) Role of N-glycosylation sites
and CXC motifs in trafficking of Medicago truncatula Nod Factor Perception protein
to the plasma membrane. J. Biol. Chem. 287: 10812-10823
Wiśniewska A, Grabowska A, Pietraszewska-Bogiel A, Tagashira N, Zuzga S,
Wóycicki R, Przybecki Z, Malepszy S, Filipecki M. (2012) Identification of genes
up-regulated during somatic embryogenesis of cucumber. Plant Physiol. Biochem. 50:
54-64