GRAS Proteins Form a DNA Binding Complex to InduceGene Expression during Nodulation Signaling inMedicago truncatula W

Sibylle Hirsch,a,1 Jiyoung Kim,a,1 Alfonso Munoz,a,b,1 Anne B. Heckmann,c,d J. Allan Downie,c

and Giles E.D. Oldroyda,2

a Department of Disease and Stress Biology, John Innes Centre, Norwich NR4 7UH, United KingdombCentro Nacional de Biotecnologia, Darwin 3, 28049 Madrid, Spainc Department of Molecular Microbiology, John Innes Centre, Norwich NR4 7UH, United KingdomdDepartment of Molecular Biology, University of Aarhus, 8000 Aarhus C, Denmark

The symbiotic association of legumes with rhizobia involves bacterially derived Nod factor, which is sufficient to activate the

formation of nodules on the roots of the host plant. Perception ofNod factor by root hair cells induces calciumoscillations that

are a component of the Nod factor signal transduction pathway. Perception of the calcium oscillations is a function of a

calcium- and calmodulin-dependent protein kinase, and this activates nodulation gene expression via two GRAS domain

transcriptional regulators, Nodulation Signaling Pathway1 (NSP1) and NSP2, and an ERF transcription factor required for

nodulation. Here, we show that NSP1 and NSP2 form a complex that is associated with the promoters of early nodulin genes.

We show that NSP1 binds directly toENOD promoters through the novel cis-element AATTT.While NSP1 shows direct binding

to the ENOD11 promoter in vitro, this association in vivo requires NSP2. The NSP1-NSP2 association with the ENOD11

promoter is enhanced following Nod factor elicitation. Mutations in the domain of NSP2 responsible for its interaction with

NSP1 highlight the significance of theNSP1-NSP2 heteropolymer for nodulation signaling. Ourwork reveals direct binding of a

GRAS protein complex to DNA and highlights the importance of the NSP1-NSP2 complex for efficient nodulation in the model

legume Medicago truncatula.

INTRODUCTION

Legumes enter symbiotic interactions with rhizobial bacteria that

help the plant in the acquisition of nitrogen. The interaction is

newly established in every plant and involves recognition of

rhizobially made Nod factor signal (Denarie et al., 1996; Downie

and Walker, 1999). Perception of Nod factor leads to the induc-

tion of a unique organ, the root nodule that accommodates the

bacteria and provides suitable conditions for nitrogen fixation.

Nod factor perception by legume root hair cells leads to oscil-

lations in calcium levels, termed calcium spiking (Ehrhardt et al.,

1996). Genetic and pharmacological studies indicate that cal-

cium spiking plays a pivotal role in the Nod factor signal trans-

duction pathway that activates transcriptional changes in

response to the symbiotic signal (Wais et al., 2000; Walker

et al., 2000; Charron et al., 2004; Miwa et al., 2006).

Genetic dissection of nodulation in Medicago truncatula and

Lotus japonicus revealed a signaling pathway necessary for Nod

factor signal transduction (Oldroyd and Downie, 2004, 2006). A

suite of receptor-like kinases are necessary for Nod factor

perception (Limpens et al., 2003; Madsen et al., 2003; Radutoiu

et al., 2003). Following this perception presumably at the plasma

membrane, all remaining components of theNod factor signaling

pathway are associated with the nucleus: a number of compo-

nents at the core of the nuclear pore (Kanamori et al., 2006; Saito

et al., 2007), as well as two cation channels (Ane et al., 2004;

Imaizumi-Anraku et al., 2005), at least one of which is located on

the nuclear membrane (Riely et al., 2007), are necessary for the

Nod factor induction of calcium spiking. The calcium response

itself is strongly associated with the nucleus (Ehrhardt et al.,

1996). Downstream of the calcium response is a nuclear-

localized calcium- and calmodulin-dependent protein kinase

(CCaMK) (Levy et al., 2004; Mitra et al., 2004; Kalo et al., 2005;

Smit et al., 2005; Gleason et al., 2006; Tirichine et al., 2006), three

nuclear-associated transcriptional regulators, two GRAS family

proteins, Nodulation Signaling Pathway1 (NSP1) and NSP2 (Kalo

et al., 2005; Smit et al., 2005; Heckmann et al., 2006), and an ERF

transcription factor, ERN1 (Middleton et al., 2007). CCaMK is

regulated by autoinhibition that is relieved upon calmodulin

binding. The removal of the autoinhibition domain leads to

constitutive calcium-independent kinase activity in vitro and

spontaneous nodule formation in vivo (Gleason et al., 2006).

In M. truncatula, several genes have been identified that are

rapidly induced upon Nod factor application. The best studied of

these early nodulins (ENODs) is ENOD11; however, a number

of the genetic components identified as being essential for

1 These authors contributed equally to this work.2 Address correspondence to giles.oldroyd@bbsrc.ac.uk.The author responsible for distribution of materials integral to thefindings presented in this article in accordance with the policy describedin the Instructions for Authors (www.plantcell.org) is: Giles E.D. Oldroyd(giles.oldroyd@bbsrc.ac.uk).WOnline version contains Web-only data.www.plantcell.org/cgi/doi/10.1105/tpc.108.064501

The Plant Cell, Vol. 21: 545–557, February 2009, www.plantcell.org ã 2009 American Society of Plant Biologists

nodulation are themselves induced following Nod factor elicita-

tion. The autoactivated CCaMK induces ENOD11 expression,

and this requires NSP1, NSP2, and ERN1 (Gleason et al., 2006;

Tirichine et al., 2006; Middleton et al., 2007), indicating that the

calcium signal is transduced by CCaMK through these transcrip-

tional regulators to activate gene expression. ERN1 and two

close homologs, ERN2 and ERN3, have been shown to directly

associate with the ENOD11 promoter (Andriankaja et al., 2007).

While ERN1 and ERN2 function as transcriptional activators,

ERN3 represses ERN1/ERN2-dependent transcriptional activa-

tion of ENOD11 (Andriankaja et al., 2007). Nod factor induction of

ENOD11 is confined to a zoneof the root epidermis (Journet et al.,

2001), and this restriction of ENOD11 is a function of Nodule

Inception (NIN), which encodes a protein with transmembrane

domains and a putative DNA binding domain (Schauser et al.,

1999; Marsh et al., 2007).

The GRAS family of putative transcriptional regulators is found

throughout the plant kingdom, and these proteins have diverse

roles in plant development, including root development, axillary

shoot development, and maintenance of the shoot apical mer-

istem (Bolle, 2004). GRAS proteins show conserved residues in

the C terminus but contain a variable N terminus with homopol-

ymeric stretches of certain amino acids. It has recently been

shown that two GRAS proteins that regulate root growth,

SCARECROW (SCR) and SHORTROOT (SHR), interact with

each other (Cui et al., 2007), while a class of GRAS proteins

involved in regulating plant growth, the DELLA proteins, interact

with a transcription factor involved in phytochrome signaling (de

Lucas et al., 2008; Feng et al., 2008)

Although NSP1 and NSP2 both encode GRAS family mem-

bers, they are not very similar (20.1% identity and 32.9% sim-

ilarity), suggesting that they have different functions in the Nod

factor signaling pathway. Here, we show that NSP1 is a DNA

binding protein that binds to the promoter of the Nod factor–

inducible genes ENOD11, ERN1, and NIN. The in vivo associa-

tion between NSP1 and the ENOD11 promoter is enhanced by

Nod factor elicitation. NSP1 and NSP2 form homopolymers and

heteropolymers, and the binding of NSP1 to the ENOD11 pro-

moter requires the action of NSP2. A pointmutation in the leucine

heptad repeat (LHRI) domain of NSP2, which is necessary for the

heteropolymerization, reveals a function for heteropolymeriza-

tion in nodulation signaling. Our data reveal that NSP1 and NSP2

form a complex at the DNA to induce specific gene expression

changes essential for the root nodule symbiosis.

RESULTS

NSP1 and NSP2 Form Heterodimers in

Saccharomyces cerevisiae

The similarity in the nsp1 and nsp2 mutant phenotypes (Catoira

et al., 2000; Oldroyd and Long, 2003) and the fact that they both

encode GRAS domain proteins suggested the possibility that

NSP1 andNSP2may form a complex. To test this hypothesis, we

assessed interactions betweenNSP1 andNSP2 in theGal4 yeast

two-hybrid system. Fusions of NSP1 or NSP2 to the Gal4 DNA

binding domain caused autoactivation of the Gal4 system, indi-

cating a function for NSP1andNSP2as transcriptional regulators

(Figures 1A and 1C). Autoactivation is a function of theN-terminal

domains ofbothNSP1andNSP2, ashasbeenpreviously reported

for a GRAS protein from lily (Lilium longiflorum; Morohashi et al.,

2003), since removal of the N-terminal domains of NSP1 (NSP1

156-554) andNSP2 (NSP2 113-508) abolished the autoactivation

in S. cerevisiae, and the N-terminal domains alone (NSP1 1-155

and NSP2 1-112) were sufficient for autoactivation (Figures 1A

and 1C). However, the autoactivation of NSP2 by the full-length

protein is also regulated by the VHIID domain as a specific

deletion of the VHIID domain (NSP2D208-276) abolishes auto-

activation (Figure 1C), andapointmutant in thismotif (NSP2E232K)

also inhibits autoactivation (Figure 1A).NSP2E232K was identified

in the nsp2-3 mutant of M. truncatula, a weak nsp2 mutant

allele showing limited nodule formation in contrast with the null

allele nsp2-2 that shows no nodule formation (Kalo et al., 2005).

The fact that nsp2-3 showed defects in nodulation and this

mutation abolishes autoactivation in S. cerevisiae indicates that

the transcriptional activity by NSP2 is relevant for its function in

the plant.

S. cerevisiae containing NSP2E232K fused to the Gal4 DNA

binding domain and NSP1 fused to the Gal4 activation domain

showed increased reporter gene expression (Figure 1A), indicat-

ing an interaction between NSP1 and NSP2. The autoactivation

of NSP1 and NSP2 limited this yeast assay, but by quantifying

b-galactosidase activity, we found a significant enhancement in

S. cerevisiae containing NSP1 and wild-type NSP2 compared

with the autoactivation by NSP1 or NSP2 alone (Figure 1B).

Assays of b-galactosidase indicated that NSP1 and NSP2 inter-

actwhenNSP1orNSP2was fused to either theGal4DNAbinding

or activation domains (Figure 1B).

GRAS proteins contain two leucine heptad repeat domains

(LHRI and LHRII) and three domains, VHIID, PFYRE, and SAW,

that have residues conserved across all GRAS proteins (Bolle,

2004; Kalo et al., 2005). By generating both N- and C-terminal

deletions of NSP1 and NSP2, we could identify the domains of

these proteins that were responsible for the interaction. Most

N- and C-terminal deletions of NSP1 abolished its interaction

with NSP2 (Figure 1C; see Supplemental Table 1 online). Indeed,

the only deleted version of NSP1 that still showed an interaction

with NSP2 was an N-terminal deletion that removes the first 155

amino acids (NSP2 156-554). All the NSP1 deletions were stable

in S. cerevisiae, as shown by antibody staining (see Supplemen-

tal Figure 1A online). This work suggests that more than one

region of NSP1 is required for its interaction with NSP2. By

contrast, C-terminal deletions of NSP2 (NSP2 1-276 and NSP2

1-180) had no effect on the interaction with NSP1, whereas

N-terminal deletions completely abolished the interaction (NSP2

208-508 andNSP2 287-508) (Figure 1C; see Supplemental Table

1 online). Immunostaining showed that these NSP2 deletions

were also stable in S. cerevisiae (see Supplemental Figure 1B

online). From these studies we deduced that either the

N-terminal region or the LHRI domain of NSP2 may be respon-

sible for its interaction with NSP1. The LHRI domain of NSP2

(NSP2 113-180)was sufficient to interact withNSP1, and specific

deletion of the LHRI domain (NSP2D113-508) abolished the

interaction with NSP1 (Figure 1C; see Supplemental Table 1 on-

line). This indicates that the LHRI domain of NSP2 is necessary

546 The Plant Cell

and sufficient for the interaction with NSP1. However, two NSP2

constructs (NSP2 113-508 and NSP2D208-276) do not interact

withNSP1, but contain the LHRI domain. At this stage, we cannot

explain this inconsistency, but it seems likely that these two

constructs are inappropriately folded to allow LHRI activity.

These studies suggest that the LHRI domain of NSP2 interacts

with multiple domains of the NSP1 protein.

NSP1 and NSP2 Form Homopolymers and Heteropolymers

in Plant Cells

To validate theNSP1 andNSP2 interactionwehadobserved inS.

cerevisiae, we assessed the interaction of these proteins inside

plant cells. We tested the interaction of NSP1 and NSP2 in

leaves ofNicotiana benthamiana using bimolecular fluorescence

Figure 1. NSP1 and NSP2 Form Heteropolymers in Yeast.

(A)NSP1 and NSP2 Gal4 DNA binding domain fusions autoactivate the Gal4 system in S. cerevisiae. S. cerevisiaewas grown onmedia with selection for

growth and b-galactosidase activity. The N-terminal domains of NSP1 (1-155) and NSP2 (1-112) are sufficient for autoactivation, and if this domain is

removed from NSP1 (NSP1 156-554), no autoactivation was observed. Autoactivation of NSP2 is abolished by the NSP2E232K mutation. NSP2E232Kfused to the Gal4 DNA binding domain interacts with NSP1 fused to the activation domain.

(B) Quantification of b-galactosidase activity in S. cerevisiae revealed interactions between wild-type NSP1 and NSP2. One unit of b-galactosidase is

defined as the amount that hydrolyzes 1 mmol o-nitrophenyl b-D-galactopyranoside (ONPG) per minute. Each value is the mean of three technical and

three biological replicates. Error bars indicate the SE of the three biological replicates. AD, Gal4 activation domain; BD, Gal4 DNA binding domain.

(C) The GRAS proteins are defined by two leucine heptad repeats (LHRI and LHRII) and three domains with highly conserved residues (VHIID, PYFRE,

and SAW). Several N- and C-terminal deletions of NSP1 and NSP2 were fused to the Gal4 DNA binding domain and tested for interaction with NSP2 or

NSP1, respectively. b-Galactosidase activity is shown. Please note the change in scale. Positive interactions are indicated with +, and – indicates no

interaction.

GRAS Protein/ DNA Complex Formation 547

complementation (BiFC) and coimmunoprecipitation. We gener-

ated different fusions of the yellow fluorescent protein (YFP)

domains to NSP1 and NSP2 and transiently expressed these

proteins in N. benthamiana leaves using Agrobacterium tumefa-

ciens infiltrations. Clear indications for an in vivo interaction

between NSP1 and NSP2 were observed, with strong yellow

fluorescence in the nuclei of N. benthamiana cells (Figure 2A).

This nuclear localization of the NSP1 and NSP2 interaction in N.

benthamiana is in agreement with earlier observations indicating

a nuclear localization of NSP1 and NSP2 (Kalo et al., 2005; Smit

et al., 2005; Murakami et al., 2006). Consistent with the analyses

inS. cerevisiae, in planta BiFC experiments showed that the LHRI

domain of NSP2 is sufficient for the interaction with NSP1 (Figure

2B), and removal of the LHRI domain abolishes the interaction

withNSP1 (Figure 2C). TheNSP2 construct lacking LHRI is stable

in N. benthamiana (Figure 2C, inset).

Coimmunoprecipitation experiments also revealed an interac-

tion between NSP1 and NSP2 in N. benthamiana. For these

experiments, we generated constructs for the expression of

NSP1 with a C-terminal triple HA tag and NSP2 with a C-terminal

FLAG tag. These fusions are functional since they are able to

complement M. truncatula nsp1 or nsp2 mutants, respectively

(see Supplemental Table 2 online). Using the FLAG epitope, we

immunoprecipitated NSP2 from leaf extracts and NSP1 copre-

cipitated as revealed by the presence of the HA epitope (Figure

2D). NSP1-HA was not precipitated from N. benthamiana leaves

that lacked NSP2-FLAG (Figure 2D). Therefore, we conclude that

NSP1 and NSP2 are able to interact in both S. cerevisiae and in

plant cells.

To assess the possibility of self-interaction of NSP1 or NSP2,

we coexpressed HA, green fluorescent protein (GFP), and FLAG-

tagged NSP2 or HA and FLAG-tagged NSP1 in N. benthamiana

leaves. All epitope fusions used in this study were able to

complement their reciprocal M. truncatula mutant, indicating

functionality. Using the FLAG epitope to isolate NSP1 or NSP2

from leaf extracts, we found the HA epitope coprecipitating in

both cases (Figures 3A and 3B), indicating homopolymerization.

This could be validated with BiFC for NSP1 (Figure 3C). We did

not observe NSP2–NSP2 interaction using BiFC, but we could

coprecipitate NSP2-GFP along with NSP2-FLAG using a FLAG

epitope immunoprecipitation, confirming the previous result

using NSP2-HA. This work indicates that NSP1 and NSP2 form

homodimers or higher order homopolymers, as well as hetero-

dimers or higher-order heteropolymers.

Figure 2. NSP1 and NSP2 Form Heteropolymers in Planta.

(A) to (C) N. benthamiana leaves were cotransformed with YFPN-NSP1

and YFPC-NSP2 (A), YFPN-NSP1 and YFPC-NSP2-LHRI (B), and YFPN-

NSP1 and YFPC-NSP2DLHRI (C). Overlays of confocal YFP and bright-

field images of epidermal N. benthamiana leaf cells are shown. (A) and

(B) show the nuclear localization of the NSP1 and NSP2 interaction.

Removal of the LHRI domain of NSP2 abolished the interaction with

NSP1 (C). The inset of (C) confirms protein stability of (a) YFPC-NSP2 (65

kD) and (b) YFPC-NSP2DLHRI (58 kD) in these N. benthamiana transient

expression studies. Bars = 40 mm.

(D)NSP1-3xHA alone and NSP2-FLAG with NSP1-3xHA were transiently

expressed in N. benthamiana leaves. Leaves were harvested 3 d after

infiltration, and equal amounts of total protein extracts (input) were

subjected to immunoprecipitation with a-FLAG antibodies. Analysis of

the elution revealed coimmunoprecipitation of NSP1-3xHA with NSP2-

FLAG that was not observed in the leaves transformed with NSP1-3xHA

alone.

548 The Plant Cell

The NSP1–NSP2 Interaction Is Required for

Nodule Formation

To investigate the function of the heteropolymerization between

NSP1 and NSP2 in M. truncatula, we transformed nsp2-2 mu-

tants, which are unable to form nodules, with the NSP2 deletion

construct lacking the LHRI domain responsible for its interaction

with NSP1. Five weeks after inoculation with Sinorhizobiummeliloti,

no nodules were detected on the nsp2-2 roots transformed with

NSP2DLHR1 (Figure 4B), in contrast with nsp2-2 roots trans-

formed with full-length NSP2 that showed extensive nodulation

(Figure 4A). This result suggests that the interaction that occurs

through the LHRI domain is required for NSP2 action during

nodule formation.

Point mutations are likely to cause less deleterious effects on

protein function than deletions. To further assess the relevance

of the interaction between NSP1 and NSP2, we identified a point

mutation in the LHRI domain of NSP2 that reduced the interac-

tion with NSP1 by approximately threefold (Figure 4C). This

mutation results in the amino acid substitution A168V. We

assessed the ability of NSP2A168V to complement the nsp2-2

mutant. Nodule numbers were reduced by approximately three-

fold in transgenic nsp2-2 roots harboring NSP2A168V compared

with roots transformed with wild-type NSP2 (Figure 4D). To test

whether the few nodules that formed on the NSP2A168V roots

were functional, the rate of nitrogenase activity within the nod-

ules was measured per milligram of nodule tissue. Acetylene

reductionwas impaired in the nodules of the nsp2-2mutant roots

transformed with NSP2A168V compared with nodules on nsp2-2

transformed with wild-type NSP2 (Figure 4E). Taken together,

these results reveal that in the absence of the NSP1–NSP2

interaction, there is no nodule formation, while when the inter-

action is reduced, there is a reduced level of nodulation. These

correlations imply that the NSP1–NSP2 interaction is important

for nodulation signaling in M. truncatula.

NSP1 Directly Associates with ENOD Promoters

GRAS proteins have recently been shown to be associated with

the promoters of target genes in vivo (Cui et al., 2007), although

no direct binding of GRAS proteins to DNA has been observed.

ENOD11 is rapidly induced by Nod factor treatment in M.

truncatula roots, and this requires bothNSP1 andNSP2 (Journet

et al., 2001; Boisson-Dernier et al., 2005; Middleton et al., 2007).

As ENOD11 is a potential target for inducibility by NSP1 and

NSP2, we assessed whether these GRAS proteins could directly

bind to the ENOD11 promoter. Using purified NSP1 and NSP2

from Escherichia coli, we tested if these proteins could associate

with a radiolabeled 1049-bp fragment containing the start codon

at the 39 end of the ENOD11 promoter that shows Nod factor

inducibility (Boisson-Dernier et al., 2005). We saw a clear band

shift in the electrophoretic mobility shift analysis (EMSA) when

the ENOD11 promoter was incubated with NSP1, but not with

either NSP2 or any of the controls (Figure 5A). The binding of

NSP1 to the ENOD11 promoter can be outcompeted by the

addition of unlabeledENOD11 promoter (Figure 5A). Considering

the interaction we have observed between NSP1 and NSP2, we

assessed whether incubation with both proteins would cause a

supershift in this assay but saw no evidence for this. Our work

indicates that NSP1, but not NSP2, can directly bind to the

ENOD11 promoter.

NIN and ERN1 are both essential for nodulation and are

induced following rhizobia or Nod factor treatment (Schauser

Figure 3. NSP1 and NSP2 Form Homopolymers.

(A) NSP1-3xHA alone or NSP1-FLAG and NSP1-3xHA were transiently

expressed under the control of the 35S promoter in N. benthamiana

leaves. Leaves were harvested 3 d after infiltration and NSP1-FLAG

immunoprecipitated (elution) from total protein extracts (input). Coimmu-

noprecipitation ofNSP1-3xHA indicated an interaction ofNSP1with itself.

(B) In a similar manner, N. benthamiana leaves were transformed with

NSP2-3xHA or NSP2::GFP alone, NSP2::3xHA and NSP2-FLAG, or

NSP2-GFP and NSP2::FLAG. NSP2-FLAG was immunoprecipitated

(elution) from total protein extracts (input) 3 d after infiltration. Coimmu-

noprecipitation of NSP2-GFP and NSP2-3xHA with NSP2-FLAG re-

vealed homopolymerization of this protein. The input shown is from the

NSP2-FLAG/NSP2-3xHA experiment. Similar inputs were observed with

the NSP2-FLAG/NSP2-GFP experiment.

(C) Leaves of N. benthamiana were transformed with YFPN-NSP1 and

YFPC-NSP1. Yellow fluorescence in the nuclei indicates homopolyme-

rization of NSP1. Bar = 40 mm.

GRAS Protein/ DNA Complex Formation 549

et al., 1999; Marsh et al., 2007; Middleton et al., 2007). The

induction of NIN and ERN1 by rhizobia requires both NSP1 and

NSP2 (Figure 5D) (Murakami et al., 2006). We tested whether

NSP1 could bind to the NIN and ERN1 promoters in vitro by

EMSA.Weobserved thatpurifiedNSP1associateswith the2892

to213 region of theNIN promoter and the2862 to229 region of

the ERN1 promoter (Figures 5B and 5C). This work indicates that

NSP1 binds directly to multiple early nodulin promoters.

We assessedwhich domains of NSP1were responsible for the

DNA binding. Most of the N- and C-terminal deleted derivatives

of NSP1 still associated with the ENOD11 promoter (see Sup-

plemental Figures 2A and 2B online), and the only derivatives that

lost DNA binding capability were deletions that removed both

LHRI and LHRII domains. The isolated LHRI and LHRII domains

of NSP1 retarded the ENOD11 promoter fragment (see Supple-

mental Figure 2B online), indicating that NSP1 associates with

the ENOD11 promoter through the LHR domains.

The NSP1 Binding cis-Element

We next attempted to define the NSP1 targeted cis-elements.

Using random binding site selection to characterize the se-

quence binding specificity of NSP1, we identified several ran-

dom oligonucleotides that bind NSP1; all contained the

consensus motif AATTT (Figure 6A), indicating that this is likely

to represent the NSP1 targeted cis-element. The ENOD11,NIN,

and ERN1 promoters contain this AATTT motif. The domains of

the ENOD11 promoter have been well characterized, and the

2411 to2257 region has been shown to be responsible for Nod

factor inducibility (Boisson-Dernier et al., 2005). This region

contains two AATTT motifs (Figure 6B). When we specifically

isolated these two motifs, embedded within a 27-bp region of

the surrounding promoter sequence, we were able to see

binding of NSP1 to both cis-elements (Figure 6C). The motifs

were called NODULATION RESPONSIVE ELEMENT1 (NRE1)

Figure 4. NSP1-NSP2 Heteropolymerization Is Relevant for Nodulation Signaling.

(A) and (B) M. truncatula nsp2-2 roots were transformed with NSP2 (A) and NSP2DLHRI (B) under control of the 35S promoter. Root systems were

inoculated with S. meliloti and nodule formation monitored for up to 5 weeks. Transgenic roots were identified via fluorescence of GFP encoded on the

binary vector. Transgenic Medicago roots harboring NSP2DLHRI did not develop nodules (B), while roots harboring NSP2 form wild-type-like nodules

(A). The inset of (B) confirms the presence of (a) NSP2 and (b) NSP2DLHRI by PCR: the lower band of 522 bp represents the deleted form of NSP2 in the

nsp2-2 mutant, while the upper represents either full-length NSP2 (984 bp) or NSP2DLHRI (780 bp). Bars = 5 mm.

(C) Quantification of b-galactosidase activity in S. cerevisiae revealed a threefold reduction in the heteropolymerization between NSP1 and NSP2 when

a single amino acid change (A168V) is present in the LHRI domain of NSP2. One unit of b-galactosidase is defined as the amount that hydrolyzes 1 mmol

o-nitrophenyl b-D-galactopyranoside per min. Each value is the mean of three technical and three biological replicates. Error bars represent SE.

(D)M. truncatula nsp2-2 rootswere transformedwithNSP2A168VandNSP2ascontrol. Thenodule number of transgenic roots that showGFPfluorescence

was determined 35 d after inoculation with S. meliloti. Independent transformed roots examined were n = 75 (NSP2A168V) and n = 66 (NSP2).

(E) Nodules obtained from roots transformedwithNSP2A168V andNSP2 as control were tested for their ability to fix nitrogen. The introduction of the amino

acid change led to a severe decrease of nitrogen fixation asmeasured by rates of acetylene reduction (calculated as the amount of ethylene produced per

hour and per milligram of nodule fresh weight). Each value is the mean of three biological replicates. Error bars indicate the SE of the replicates.

550 The Plant Cell

and NRE2. Mutation of the AATTT motif to CCCCC in NRE1

(NRE1-C5) and NRE2 (NRE2-C5) greatly reduced the binding of

NSP1 to these promoter fragments (Figure 6C). Deletion deriv-

atives of NSP1 that contain either the LHRI or LHRII domains

bound to NRE1 andNRE2, but not to themutant formsNRE1-C5

or NRE2-C5 (see Supplemental Figure 2C online), indicating that

the LHR domains are responsible for the specific recognition of

the AATTT motif.

To test the relative importance of the AATTT cis-elements for

Nod factor induction, we generated a fusion construct where a

single copy of the 27-bp NRE2 region containing the AATTT

motif of ENOD11 was fused to the minimal ENOD11 promoter,

2257 to 21, that lacks Nod factor inducibility. In addition, we

generated a construct where the AATTT motif of NRE2 was

mutated to CCCCC. We tested the ability of these promoter

fusions to drive the expression of b-glucuronidase (GUS) in M.

truncatula roots. We found a small but significant induction of

GUS following Nod factor treatment in the NRE2 fusion (Figure

6D). This induction is absent in an nsp1 mutant and is also

absent in the mNRE2 fusion (Figure 6D). This work shows that

NRE2 plays a role in Nod factor inducibility of the ENOD11

promoter, and this highlights the significance of NSP1 for Nod

factor gene induction.

NSP1 andNSP2 Associatewith theENOD11Promoter in Vivo

To test the NSP1 binding to the ENOD11 promoter in plants, we

undertook chromatin immunoprecipitation (ChIP) to screen for

theNSP1-ENOD11 promoter interaction in roots ofM. truncatula.

We generated antibodies against NSP1 and NSP2 to isolate the

GRAS proteins from M. truncatula root extracts. These anti-

bodies recognize their target proteins specifically and do not

cross reactwith either NSP1 or NSP2 (see Supplemental Figure 3

online). Immunoprecipitation of NSP1 from M. truncatula roots

coprecipitated PCR-amplifiable ENOD11 promoter, and this was

absent in the negative control (c-Myc). The amount of PCR-

amplifiable ENOD11 promoter was enhanced following Nod

factor treatment (Figure 7), suggesting increased NSP1 associ-

ation with ENOD11. Even though we did not observe direct

binding of NSP2 to the ENOD11 promoter in vitro, we did see

evidence for the association of NSP2 with the ENOD11 promoter

in vivo. This association was only observed following Nod factor

treatment and was absent in the nsp1-1 mutant (Figure 7). We

conclude that NSP2 associates with the ENOD11 promoter

through its interaction with NSP1. The reciprocal case was also

true; there was no NSP1 association with the ENOD11 promoter

in the nsp2-2mutant. One possible explanation for this would be

Figure 5. NSP1 Associates with ENOD Promoters in Vitro.

(A) to (C) EMSA with NSP1 and NSP2 proteins using radiolabeled ENOD promoter probes. The promoter regions of ENOD11 (�1046 to +3) (A), NIN

(�892 to �13) (B), and ERN1 (�862 to �29) (C) were used. The free ENOD promoter probe is present at the bottom of each image. The band shift,

indicated with an arrow, revealed NSP1 association with the promoter of ENOD11 (A), NIN (B), and ERN1 (C) that is absent in the GST control and in

NSP2. In (A), unlabeled ENOD11 fragment was used as competitor DNA. c10x and c50x, unlabeled competitor DNA in 10- and 50-fold excess,

respectively.

(D) RT-PCR showing that the S. meliloti induction of NIN and ERN1 is absent in nsp1-1 and nsp2-2mutants. Actin was amplified as a control. dpi, days

post inoculation.

GRAS Protein/ DNA Complex Formation 551

a mutual requirement of NSP1 and NSP2 for appropriate gene

expression. However, the interactions we have observed be-

tween NSP1 and NSP2 suggests an interdependence of NSP1

and NSP2 for appropriate activity.

DISCUSSION

Much evidence exists for GRAS proteins functioning in tran-

scriptional regulation: they are generally located in the nucleus

(Bolle, 2004), they show association with promoters in vivo (Cui

et al., 2007), and a GRAS protein has been shown to have an

innate capability to activate RNA polymerase in both S. cerevi-

siae and in plants (Morohashi et al., 2003). However, transcription

factor classification generally requires direct DNA association,

and the emerging evidence suggests that GRAS proteins may

function to regulate the action of other transcription factors (de

Lucas et al., 2008; Feng et al., 2008). Here, we show that the

GRAS protein NSP1 binds directly to DNA and does so in a

complex with a second GRAS protein, NSP2. NSP1 can activate

RNA polymerase in S. cerevisiae, and this transcriptional activity

of GRAS proteins has been previously validated in plant cells

using L. longiflorum SCL, a GRAS protein from lily (Morohashi

et al., 2003). The association of NSP1with the ENOD11 promoter

through a novel cis-element is relevant for Nod factor induction of

this promoter inM. truncatula. Our work shows the GRAS protein

NSP1 functioning as a classic transcription factor, with direct

binding to a promoter element. By contrast, we have no evidence

Figure 6. NSP1 Binds Specifically to an AATTT Motif Present in the ENOD11 Promoter.

(A) Identification of the NSP1 cis-element by random binding site selection (RBSS). NSP1 was found to bind to the indicated 15 oligonucleotides from a

pool of 420 random-sequence oligonucleotides. All oligonucleotides bound to NSP1 contained the AATTT motif.

(B) The �411 to �257 region of the ENOD11 promoter contains two AATTT motifs (underlined). Regions around these motifs were synthesized and

named NRE1 and NRE2.

(C) EMSA analysis showed that NSP1 binds to NRE1 and NRE2. Mutation of the AATTT motif to CCCCC in both NRE1 and NRE2 (indicated as NRE1-C5

and NRE2-C5) caused a dramatic reduction in NSP1 binding.

(D) A synthetic promoter was generated by fusing the minimal ENOD11 promoter lacking Nod factor inducibility to the NRE2 element and used to drive

expression of GUS. This construct was transformed into plants and assessed for Nod factor inducibility, measured by a fluorimetric GUS assay. Nod

factor inducibility was observed that was absent in nsp1-1mutants and when using NRE2-C5. The asterisk indicates a significant difference to all other

data points, measured with a t test at 95% confidence. NF, Nod factor. Values are means 6 SD (n = 15 plants for each treatment).

552 The Plant Cell

that NSP2 can bind DNA. Indeed, our data suggest that these

two GRAS proteins have divergent biochemical functions during

Nod factor signaling.

It has been previously shown that the GRAS domain proteins

SHR and SCR are able to interact with each other and that this

complex formation is indispensable for their function (Cui et al.,

2007). In a similar fashion, NSP1 and NSP2 form homopolymers

and heteropolymers. This suggests the formation of a complex

consisting of either dimers or higher-order polymers of NSP1 and

NSP2. However, we cannot exclude that competition between

the formation of homopolymers and heteropolymers might exist.

While NSP2 shows no evidence for direct binding to the ENOD11

promoter in vitro, we did find evidence for NSP2 association with

the ENOD11 promoter in vivo. We propose that this association

is as part of a complex with NSP1. While NSP1 is capable of

binding the ENOD11 promoter in vitro, its association with the

ENOD11 promoter in vivo was dependent on NSP2. This com-

plex of NSP2 and NSP1 is directly associated with the promoters

of Nod factor–inducible genes.

Our data reveal that the domains of NSP1 and NSP2 fulfill

different roles: the LHRI andLHRII domains ofNSP1are sufficient

for DNA binding, while the LHRI domain of NSP2 is responsible

for the association with NSP1. The LHRI domains of NSP1 and

NSP2 show 36.4% similarity and 31% identity, highlighting the

differential roles these two domains play. It has been shown that

the LHRI domain of theGRASproteinSolanum tuberosumRGA is

necessary for its interaction with the PIF4 basic helix-loop-helix

(bHLH) transcription factor (de Lucas et al., 2008), revealing a

similarity with the protein–protein interaction function we have

observed for NSP2-LHRI. Specific removal of the LHRI domain of

NSP2 results in the loss of interaction in plant cells and in nodule

formation. In addition, a single point mutation located in the LHRI

domain of NSP2 reduced its interaction with NSP1 and its ability

to function in nodulation signaling in M. truncatula. This correla-

tion between the ability of NSP2 to interact with NSP1 and

its ability to allow nodule formation implicates relevance for

NSP1-NSP2 heteropolymerization during nodulation signaling.

However, we cannot exclude that the inability of these NSP2

mutations to function in nodulation signaling is the result of

defects in alternative aspects of NSP2 action.

ERN1 encodes an ERF transcription factor that is necessary

for Nod factor–induced gene expression (Middleton et al., 2007).

Like NSP1 and NSP2, ERN1 is necessary for CCaMK-induced

spontaneous nodulation (Middleton et al., 2007). We show that

NSP1 associates directly with the promoter of ERN1 and

that ERN1 expression is abolished in an nsp1 mutant, indicating

that NSP1 is a direct regulator of ERN1 induction. It has recently

been shown that ERN1 and two close homologs, ERN2 and

ERN3, bind to the ENOD11 promoter (Andriankaja et al., 2007).

The binding sites for the ERN proteins and NSP1 in the ENOD11

promoter are separated by 26 bp. This close proximity in DNA

binding sites may indicate an interaction between NSP1 and

ERN1, although we have not seen evidence for such an interac-

tion in either S. cerevisiae or N. benthamiana (data not shown).

ERN1 and ERN2 alone are capable of activating ENOD11 induc-

tion in N. benthamiana leaves (Andriankaja et al., 2007). N.

benthamiana contains a functional NSP1 gene that is able to

complement an nsp1 legume mutant (Heckmann et al., 2006).

Therefore, it is possible that the ENOD11 induction by ERN1 and

ERN2 in N. benthamiana requires NSP1.

The Nod factor signaling pathway is regulated by a number of

components, including the putative transcriptional regulator

NIN. We have previously shown that nin mutants show exces-

sive Nod factor–induced ENOD11 expression in epidermal cells

(Marsh et al., 2007). NIN expression is strongly induced follow-

ing Nod factor treatment, analogous to the expression of early

nodulins, such as ENOD11 (Schauser et al., 1999; Marsh et al.,

2007). Here, we show that NSP1 binds directly to the NIN

promoter and is necessary forNIN induction. This suggests that

NSP1 is likely to induce a negative feedback loop through

the induction of NIN that regulates the Nod factor signaling

pathway.

Nod factor signaling uses oscillations in calcium as a second-

ary messenger that links Nod factor perception at the plasma

membrane to gene expression changes in the nucleus. The

observed enhancement of NSP1 and NSP2 binding to the

ENOD11 promoter in vivo following Nod factor treatment indi-

cates that some modification of these GRAS proteins occurs

during Nod factor signaling. Considering that NSP1 and NSP2

are necessary for CCaMK-induced gene expression (Gleason

et al., 2006), CCaMK is a likely candidate for suchmodification. It

is possible that NSP1 or NSP2 are modified by CCaMK to

facilitate their interaction or their association with DNA. Further

investigations are necessary to define the role of calcium and

CCaMK in the activation of NSP1–NSP2 interaction, DNA asso-

ciation, and specific gene induction. Our work reveals that a

NSP1-NSP2 complex binds to the promoters of early nodulin

genes, and this binding is in close proximity to ERN1, ERN2, and

ERN3 binding sites.We propose that NSP1, NSP2, ERN1, ERN2,

ERN3, and NIN all act in combination to regulate the expression

of early nodulins with the appropriate spatial and temporal

pattern. The promotion of NSP1-NSP2 association with ENOD

Figure 7. NSP1 and NSP2 Associate with the ENOD11 Promoter in Vivo.

ChIP followed by PCR show in vivo association of both NSP1 and NSP2

with the ENOD11 promoter. NSP1 and NSP2 were purified from M.

truncatula root extracts using their native antibodies; c-Myc antibodies

were used as a negative control. Pretreatment with Nod factor enhanced

the binding of NSP1 to the ENOD11 promoter and revealed an NSP2

association that we propose is via the interaction with NSP1. An

equivalent analysis in nsp1-1 and nsp2-2 mutants revealed no associ-

ation of NSP1 or NSP2 with the ENOD11 promoter.

GRAS Protein/ DNA Complex Formation 553

promoters is likely to be an important target of Nod factor signal

transduction.

METHODS

Plant Material and Bacterial Strains

Seeds of Medicago truncatula cv Jemalong A17, nsp1-1, and nsp2-2

were scarified with sand paper, surface-sterilized in 100% bleach,

imbibed in sterile water, and plated on 1% deionized water agar plates.

Seeds were left for 24 to 48 h at 48C and germinated on inverted agar

plates at room temperature for;30 h. Seedlings were then transferred to

modified buffered nodulation medium containing 0.1 mM 2-aminoethox-

yvinylglycine (Sigma-Aldrich). For complementation tests, M. truncatula

seedlings were transformed withAgrobacterium rhizogenes strain ARqua

carrying the appropriate binary vector using standard protocols (Boisson-

Dernier et al., 2001). One week after the transfer of the plants to growth

pouches, rootswere inoculatedwithSinorhizobiummeliloti 1021 pXLGD4

(;107 colony-forming units/mL). Nodule formation was scored 2 to 5

weeks after inoculation. The presence of the transformed gene was

verified by DNA extraction and PCR. For transient expression, Nicotiana

benthamiana plants were grown in greenhouses or growth chambers at

218C and 16 h of light for 2 to 5 weeks, and leaves were infiltrated with

Agrobacterium tumefaciens C58C1 pCH32.

Plasmid Construction

The appropriate genes were cloned into the Gateway donor vector

pDONR207 by BP recombination reactions using appropriate primers

(see Supplemental Table 3 online), sequenced, and then recombined into

the appropriate Gateway destination vectors by LR reactions (Invitrogen).

The DNA binding and activation domain vectors, pGBKT7 and pGADT7

(Clontech), had been previously made Gateway compatible by introduc-

ing the Gateway cassette B into the SmaI/XmaI restriction site and called

pDEST-GBKT7 and pDEST-GADT7. To test the interaction in N. ben-

thamiana using the BiFC approach, the genes of interest were recom-

bined into the Gateway destination vectors pGPTVII.Bar.YN-GW,

pGPTVII.Bar.GW-YN, pGPTVII.Hyg.YC-GW, and pGPTVII.Hyg.GW-YC.

For Escherichia coli expression, the following destination vectors were

used: pDEST15, pDEST17, and pBAD-DEST49. For N. benthamiana

transient expression and M. truncatula hairy root transformation, the

following destination vectors were used: pGWB11, pGWB14, pGWB15,

pK7FWG2, and pK7WGF2.

Antibody Production and Purification

To raise polyclonal antisera against NSP1 and NSP2, both proteins

were expressed as tagged forms (N-terminal His for NSP1 and thiore-

doxin for NSP2) in E. coli. Purified protein from the inclusion bodies was

used to immunize rabbits for antisera production by Eurogentec. GST-

NSP1 and GST-NSP2 were expressed in E. coli and purified using

glutathione agarose (Novagen). Both protein preparations were dia-

lyzed against the coupling buffer (0.2 M NaHCO3 and 0.5 M NaCl, pH

8.3) and were coupled to HiTrap NHS-activated HP columns (GE

Healthcare) overnight. After deactivation of the columns, 10 mL of

each serum were circulated through the respective column for 6 h. The

columns were washed thoroughly with TBS until no protein was

detected. Elution of the purified antibodies was performed with 100

mM glycine, pH 2.5, and 100 mM triethylamine, pH 11.5, with a

neutralization wash in between with TBS. Purified fractions were col-

lected in one-tenth volume of 1 M Tris-HCl, pH 8.3, to neutralize the pH.

Both acid and alkaline collected fractions for each antibody were

pooled and dialyzed against TBS.

Yeast Two-Hybrid Analysis

The yeast strains AH109 and Y187 were transformed with the destination

vectors pDEST-GBKT7 and pDEST-GADT7 containing NSP1, NSP2, or

CCaMK according to the LiAc transformation method (Gietz and Woods,

2002). Diploid yeast cells containing both expression vectorswere formed

by standardmating procedures. The expressed proteinswere then tested

for interaction by dropping 5 mL of yeast suspension onminimal synthetic

dropout (SD) agar medium containing the dropout supplement (DO) His–,

Leu–, and Trp–, 5 mM 3-amino-1,2,4-triazole, and 25 mg/ mL 5-bromo-4-

chloro-3-indolyl-a-D-galactoside (Clontech). Growth and blue coloration

were monitored for up to 7 d. The liquid b-galactosidase assay was

performed according to the Yeast Protocols Handbook PT3024-1 (Clon-

tech). Immunoblot analysis was used to validate the stability of the protein

fusions in yeast with a-HA (HA tag fused to activation domain) or a-Myc

(c-Myc tag fused to DNA binding domain) antibodies.

BiFC Analysis

A. tumefaciens strain C58C1 carrying the helper plasmid pCH32 was

transformed with the BiFC constructs by electroporation. The strains of

interest and the C58C1 strain carrying the p19 silencing suppressor

plasmid (Voinnet et al., 2003) were brought to an optical density (OD600) of

0.2 to 0.5 with 10mMMgCl2 and 150mMacetosyringone (Sigma-Aldrich).

The strains were mixed and incubated at room temperature for;2 h. The

A. tumefaciens mixture was infiltrated into N. benthamiana leaves as

described previously (Voinnet et al., 2003; Witte et al., 2004). The fluo-

rescence was assayed 60 to 80 h after infiltration using a 340 1.25

numerical aperture oil-immersion objective on an inverted Leica DM IRB

confocal microscope. For monitoring YFP signals, an argon laser at 514-

nmwavelength was used for excitation. Immunoblot analysis was used to

validate the stability of the protein fusions in N. benthamiana with a-GFP

used to detect fusions to YFPC and a-NSP1/a-NSP2 used to detect

fusions to YFPN.

EMSA

Recombinant NSP1-His and GST-NSP2 were produced in E. coli strains

DH5a and BL21 and purified using the ProBond resin (Invitrogen) and

glutathione-agarose (Novagen). The amount of purified proteins was

estimated by the Bradford method using a protein assay kit (Bio-Rad).

EMSAswere performed as described previously (Kim andKim, 2006). The

oligonucleotide probes were labeled with 32P-g-ATP using T4 nucleotide

kinase (Invitrogen) or 32P-a-dTTP using Klenow fragment (Invitrogen). The

binding reactions were performed in 20mL binding buffer (25mMHEPES-

KOH, pH 8.0, 50 mM KCl, 1 mM DTT, 0.05% Triton X-100, and 15%

glycerol). After 20 min of incubation at room temperature, the reactions

were resolvedby6 to8%nativepolyacrylamidegelswith 0.53TBEbuffer.

ChIP

For each experiment, 5 g of root tissues from 10-d-old seedlings were

used for ChIP with a-NSP1 and a-NSP2 antibodies. Nuclei were isolated

as described previously (Delaney et al., 2006). Purified nuclei were fixed

with 1% formaldehyde for 20 min immediately after extraction. The

immunoprecipitation of purified chromatin was performed using a ChIP

assay kit (Upstate) according to the manufacturer’s instructions. The

presence of the ENOD11 promoter in the ChIP immunoprecipitate was

tested by PCR analysis (see Supplemental Table 3 online).

RBSS

The RBSS assay was performed essentially as described previously

(Wilson et al., 1993). Briefly, the selection was performed by mixing 100

554 The Plant Cell

ng of double-stranded random oligonucleotides, 20 bp in length (420

possible combinations) with 200 ng of NSP1-His fusion protein attached

to 40 mL of ProBond resins (Invitrogen). The supernatant was removed,

and the pellet was resuspended in 50 mL of water, boiled for 3 min, and

centrifuged rapidly. The supernatant (5mL) was used as the template for a

PCR. The PCR product was purified from 2%agarose gel. The procedure

was repeated 10 times. After the last PCR amplification step, the PCR

products were ligated to a T/A cloning vector (Promega), and individual

selected clones were sequenced.

Fluorimetric GUS Assay

Transgenic roots of theNRE2 element driving the expression ofGUSwere

generated using A. rhizogenes. GUS activity was measured following

1 nM Nod factor or buffer alone treatment using a fluorimetric assay as

described previously (Jefferson et al., 1987). The transgenic roots were

ground in liquid nitrogen and homogenized in GUS extraction buffer for

total protein extraction. Enzymatic reactions were performed using 1 mg

of total protein extract with 4-methylumbelliferyl-b-D-glucuronide as

substrate (Sigma-Aldrich). GUS activities were measured using a micro-

titer fluorimeter.

Coimmunoprecipitation

NSP1 and NSP2 fusions were transiently expressed in N. benthamiana

and infiltrated leaves harvested 3 d after infiltration. Two grams of the

corresponding N. benthamiana leaves were ground in liquid nitrogen and

incubated in extraction buffer (100 mM Tris-HCl, pH 7.5, 150 mM NaCl,

10% glycerol, 50 mM glutamate, 50 mM arginine, 0.1% Triton X-100, 5

mM DTT, 2% polyvinylpolypyrrolidone, and protease inhibitors [Sigma-

Aldrich]) with end over end shaking for 20 min. After centrifugation at

2500g for 20 min, the extracts were ultracentrifuged at 100,000g for 30

min. The supernatant was collected and its protein concentration was

measured by the Bradford method (Bio-Rad). Coimmunoprecipitation

mixtures were made containing the same amounts of total protein in the

same volume. These mixtures were precleared by incubation with 35 mL

of IgG agarose beads for 1 h. After centrifugation, the supernatant was

incubated with constant rotation for 1.5 h with 35 mL of a-FLAG agarose

beads (Sigma-Aldrich), previously blocked to avoid unspecific binding

with 1 mg/mL BSA in extraction buffer lacking polyvinylpolypyrrolidone

for 2 h. Subsequently, the beads were collected by centrifugation,

washed six times with 1 mL of the extraction buffer lacking polyvinylpo-

lypyrrolidone, and eluted with 70 mL of 0.2 mg/mL FLAG-peptide (Sigma-

Aldrich) at room temperature for 10 min. The eluted fractions and crude

extracts were run in SDS-PAGE and subjected to immunoblots with the

corresponding antibodies. For the immunoblots, the polyclonal a-FLAG

(Sigma-Aldrich), the polyclonal a-GFP (Santa Cruz), and the high-affinity

clone 3F-10 a-HA (Roche) antibodies were used with their respective

horseradish peroxidase secondary antibodies. Enhanced chemilumines-

cent reagent (Amersham) was used to detect the proteins.

RT-PCR

Total RNA was extracted from uninoculated and S. meliloti–inoculated

roots (2 and 4 d after inoculation) of 10-d-old seedlings. Onemicrogram of

total RNA was reverse transcribed and subjected to RT-PCR. The

following PCR conditions were used: 30 cycles at 968C for 30 s, 548C

for 30 s, and 728C for 2min, followed by 10min of a final extension at 728C.

Acetylene Reduction Assay

Ten to fifteen nodules were placed into a small sealed vial, and 5% volume

acetylenewas injected. The reactionwas incubated for 1 h and the amount

of producedethylenewasmeasured by gas chromatography using 1mLof

each sample. The fresh weight of the nodules was determined and the

concentration of ethylene calculated in nanomoles per hour and milligram.

Accession Numbers

Sequence data from this article can be found in the GenBank/EMBL

databases under the following accession numbers:NSP1, AJ972473 and

AJ972478; NIN, FJ719774; NSP2, AJ832138; ERN1, EU038802; and

ENOD11, AJ297721.

Supplemental Data

The following materials are available in the online version of this article.

Supplemental Figure 1. Stability of NSP1 and NSP2 Deletions.

Supplemental Figure 2. Defining the Domains of NSP1 Required for

DNA Binding.

Supplemental Figure 3. Specificity of a-NSP1 and a-NSP2 Antibodies.

Supplemental Table 1. Quantification of NSP1/2 Autoactivation and

Interaction.

Supplemental Table 2. Complemention Tests of the C-Terminal

Tagged Versions of NSP1 and NSP2 Used in the Coimmunoprecipi-

tation Experiments.

Supplemental Table 3. Primers Used for Plasmid Construction (GW,

Gateway), ChIP-PCR, and RT-PCR Analysis.

ACKNOWLEDGMENTS

We thank Carole Thomas for help with yeast two-hybrid assays and

Katia Marrocco and Thomas Kretsch for supplying the gateway-

compatible plasmids necessary for BiFC. We also thank Verity Bonnell

for constructing the point mutation in NSP2. This work was supported

by the Biotechnology and Biological Science Research Council as grant

BB/D521749/1, John Innes Foundation funding to S.H., a Marie Curie

training fellowship to J.K., and a postdoctoral fellowship to A.M. from

the Ministerio de Educacion y Ciencia, Spain.

Received November 19, 2008; revised January 29, 2009; accepted

February 11, 2009; published February 27, 2009.

REFERENCES

Andriankaja, A., Boisson-Dernier, A., Frances, L., Sauviac, L.,

Jauneau, A., Barker, D.G., and de Carvalho-Niebel, F. (2007).

AP2-ERF transcription factors mediate nod factor dependent Mt

ENOD11 activation in root hairs via a novel cis-regulatory motif. Plant

Cell 19: 2866–2885.

Ane, J.M., et al. (2004). Medicago truncatula DMI1 required for bacterial

and fungal symbioses in legumes. Science 303: 1364–1367.

Boisson-Dernier, A., Andriankaja, A., Chabaud, M., Niebel, A.,

Journet, E.P., Barker, D.G., and de Carvalho-Niebel, F. (2005).

MtENOD11 gene activation during rhizobial infection and mycorrhizal

arbuscule development requires a common AT-rich-containing regu-

latory sequence. Mol. Plant Microbe Interact. 18: 1269–1276.

Boisson-Dernier, A., Chabaud, M., Garcia, F., Becard, G.,

Rosenberg, C., and Barker, D.G. (2001). Agrobacterium rhizogenes-

transformed roots of Medicago truncatula for the study of nitrogen-

fixing and endomycorrhizal symbiotic associations. Mol. Plant Microbe

Interact. 14: 695–700.

GRAS Protein/ DNA Complex Formation 555

Bolle, C. (2004). The role of GRAS proteins in plant signal transduction

and development. Planta 218: 683–692.

Catoira, R., Galera, C., de Billy, F., Penmetsa, R.V., Journet, E.P.,

Maillet, F., Rosenberg, C., Cook, D., Gough, C., and Denarie, J.

(2000). Four genes of Medicago truncatula controlling components of

a nod factor transduction pathway. Plant Cell 12: 1647–1665.

Charron, D., Pingret, J.L., Chabaud, M., Journet, E.P., and Barker,

D.G. (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.

Cui, H., Levesque, M.P., Vernoux, T., Jung, J.W., Paquette, A.J.,

Gallagher, K.L., Wang, J.Y., Blilou, I., Scheres, B., and Benfey,

P.N. (2007). An evolutionarily conserved mechanism delimiting SHR

movement defines a single layer of endodermis in plants. Science

316: 421–425.

Delaney, K.J., Xu, R., Zhang, J., Li, Q.Q., Yun, K.Y., Falcone, D.L.,

and Hunt, A.G. (2006). Calmodulin interacts with and regulates the

RNA-binding activity of an Arabidopsis polyadenylation factor subunit.

Plant Physiol. 140: 1507–1521.

de Lucas, M., Daviere, J.M., Rodriguez-Falcon, M., Pontin, M.,

Iglesias-Pedraz, J.M., Lorrain, S., Fankhauser, C., Blazquez,

M.A., Titarenko, E., and Prat, S. (2008). A molecular framework for

light and gibberellin control of cell elongation. Nature 451: 480–484.

Denarie, J., Debelle, F., and Prome, J.C. (1996). Rhizobium lipo-

chitooligosaccharide nodulation factors: Signaling molecules mediat-

ing recognition and morphogenesis. Annu. Rev. Biochem. 65: 503–535.

Downie, J.A., and Walker, S.A. (1999). Plant responses to nodulation

factors. Curr. Opin. Plant Biol. 2: 483–489.

Ehrhardt, D.W., Wais, R., Long, S.R. (1996). Calcium spiking in plant

root hairs responding to rhizobium nodulation signals. Cell. 85:

673–681.

Feng, S., et al. (2008). Coordinated regulation of Arabidopsis thaliana

development by light and gibberellins. Nature 451: 475–479.

Gietz, R.D., and Woods, R.A. (2002). Transformation of yeast by lithium

acetate/single-stranded carrier DNA/polyethylene glycol method.

Methods Enzymol. 350: 87–96.

Gleason, C., Chaudhuri, S., Yang, T.B., Munoz, A., Poovaiah, B.W.,

and Oldroyd, G.E.D. (2006). Nodulation independent of rhizobia

induced by a calcium-activated kinase lacking autoinhibition. Nature

441: 1149–1152.

Heckmann, A.B., Lombardo, F., Miwa, H., Perry, J.A., Bunnewell, S.,

Parniske, M., Wang, T.L., and Downie, J.A. (2006). Lotus japonicus

nodulation requires two GRAS-domain regulators, one of which is

functionally conserved in a non-legume. Plant Physiol. 142: 1739–

1750.

Imaizumi-Anraku, H., et al. (2005). Plastid proteins crucial for symbi-

otic fungal and bacterial entry into plant roots. Nature 433: 527–531.

Jefferson, R.A., Kavanagh, T.A., and Bevan, M.W. (1987). GUS

fusions: Beta-glucuronidase as a sensitive and versatile gene fusion

marker in higher plants. EMBO J. 6: 3901–3907.

Journet, E.P., El-Gachtouli, N., Vernoud, V., de Billy, F., Pichon, M.,

Dedieu, A., Arnould, C., Morandi, D., Barker, D.G., and Gianinazzi-

Pearson, V. (2001). Medicago truncatula ENOD11: A novel RPRP-

encoding early nodulin gene expressed during mycorrhization in

arbuscule-containing cells. Mol. Plant Microbe Interact. 14: 737–748.

Kalo, P., et al. (2005). Nodulation signaling in legumes requires NSP2, a

member of the GRAS family of transcriptional regulators. Science 308:

1786–1789.

Kanamori, N., et al. (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.

Kim, J., and Kim, H.Y. (2006). Functional analysis of a calcium-binding

transcription factor involved in plant salt stress signaling. FEBS Lett.

580: 5251–5256.

Levy, J., et al. (2004). A putative Ca2+ and calmodulin-dependent

protein kinase required for bacterial and fungal symbioses. Science

303: 1361–1364.

Limpens, E., Franken, C., Smit, P., Willemse, J., Bisseling, T., and

Geurts, R. (2003). LysM domain receptor kinases regulating rhizobial

Nod factor-induced infection. Science 302: 630–633.

Madsen, E.B., Madsen, L.H., Radutoiu, S., Olbryt, M., Rakwalska,

M., Szczyglowski, K., Sato, S., Kaneko, T., Tabata, S., Sandal, N.,

and Stougaard, J. (2003). A receptor kinase gene of the LysM type

is involved in legume perception of rhizobial signals. Nature 425:

637–640.

Marsh, J.F., Rakocevic, A., Mitra, R.M., Brocard, L., Sun, J.,

Eschstruth, A., Long, S.R., Schultze, M., Ratet, P., and Oldroyd,

G.E. (2007). Medicago truncatula NIN is essential for rhizobium-

independent nodule organogenesis induced by autoactive CCaMK.

Plant Physiol. 144: 324–335.

Middleton, P.H., et al. (2007). An ERF transcription factor in Medicago

truncatula that is essential for nod factor signal transduction. Plant

Cell 19: 1221–1234.

Mitra, R.M., Gleason, C.A., Edwards, A., Hadfield, J., Downie, J.A.,

Oldroyd, G.E.D., and Long, S.R. (2004). A Ca2+/calmodulin-

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, G.E., and Downie, J.A. (2006). Analysis of

Nod-factor-induced calcium signaling in root hairs of symbiotically

defective mutants of Lotus japonicus. Mol. Plant Microbe Interact. 19:

914–923.

Morohashi, K., Minami, M., Takase, H., Hotta, Y., and Hiratsuka, K.

(2003). Isolation and characterization of a novel GRAS gene that

regulates meiosis-associated gene expression. J. Biol. Chem. 278:

20865–20873.

Murakami, Y., Miwa, H., Imaizumi-Anraku, H., Kouchi, H., Downie,

J.A., Kawaguchi, M., and 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.

Oldroyd, G.E., and Downie, J.A. (2006). Nuclear calcium changes at

the core of symbiosis signalling. Curr. Opin. Plant Biol. 9: 351–357.

Oldroyd, G.E.D., and Downie, J.A. (2004). Calcium, kinases and

nodulation signalling in legumes. Nat. Rev. Mol. Cell Biol. 5: 566–576.

Oldroyd, G.E.D., and Long, S.R. (2003). Identification and char-

acterization of nodulation-signaling pathway 2, a gene of Medicago

truncatula involved in Nod factor signaling. Plant Physiol. 131:

1027–1032.

Radutoiu, S., Madsen, L.H., Madsen, E.B., Felle, H.H., Umehara, Y.,

Gronlund, M., Sato, S., Nakamura, Y., Tabata, S., Sandal, N., and

Stougaard, J. (2003). Plant recognition of symbiotic bacteria requires

two LysM receptor-like kinases. Nature 425: 585–592.

Riely, B.K., Lougnon, G., Ane, J.M., and Cook, D.R. (2007). The

symbiotic ion channel homolog DMI1 is localized in the nuclear

membrane of Medicago truncatula roots. Plant J. 49: 208–216.

Saito, K., et al. (2007). NUCLEOPORIN85 is required for calcium

spiking, fungal and bacterial symbioses, and seed production in

Lotus japonicus. Plant Cell 19: 610–624.

Schauser, L., Roussis, A., Stiller, J., and Stougaard, J. (1999). A plant

regulator controlling development of symbiotic root nodules. Nature

402: 191–195.

Smit, P., Raedts, J., Portyanko, V., Debelle, F., Gough, C., Bisseling,

T., and Geurts, R. (2005). NSP1 of the GRAS protein family is

556 The Plant Cell

essential for rhizobial Nod factor-induced transcription. Science 308:

1789–1791.

Tirichine, L., et al. (2006). Deregulation of a Ca2+/calmodulin-

dependent kinase leads to spontaneous nodule development. Nature

441: 1153–1156.

Voinnet, O., Rivas, S., Mestre, P., and Baulcombe, D. (2003). An

enhanced transient expression system in plants based on suppres-

sion of gene silencing by the p19 protein of tomato bushy stunt virus.

Plant J. 33: 949–956.

Wais, R.J., Galera, C., Oldroyd, G., Catoira, R., Penmetsa, R.V.,

Cook, D., Gough, C., Denarie, J., and Long, S.R. (2000). Genetic

analysis of calcium spiking responses in nodulation mutants

of Medicago truncatula. Proc. Natl. Acad. Sci. USA 97: 13407–

13412.

Walker, S.A., Viprey, V., and Downie, J.A. (2000). Dissection of

nodulation signaling using pea mutants defective for calcium spiking

induced by Nod factors and chitin oligomers. Proc. Natl. Acad. Sci.

USA 97: 13413–13418.

Wilson, D., Sheng, G., Lecuit, T., Dostatni, N., and Desplan, C.

(1993). Cooperative dimerization of paired class homeo domains on

DNA. Genes Dev. 7: 2120–2134.

Witte, C.P., Noel, L.D., Gielbert, J., Parker, J.E., and Romeis, T.

(2004). Rapid one-step protein purification from plant material using

the eight-amino acid StrepII epitope. Plant Mol. Biol. 55: 135–147.

GRAS Protein/ DNA Complex Formation 557

DOI 10.1105/tpc.108.064501; originally published online February 27, 2009; 2009;21;545-557Plant Cell

OldroydSibylle Hirsch, Jiyoung Kim, Alfonso Muñoz, Anne B. Heckmann, J. Allan Downie and Giles E.D.

Medicago truncatulaSignaling in GRAS Proteins Form a DNA Binding Complex to Induce Gene Expression during Nodulation

 This information is current as of December 5, 2015

 

Supplemental Data http://www.plantcell.org/content/suppl/2009/02/23/tpc.108.064501.DC2.html http://www.plantcell.org/content/suppl/2009/02/12/tpc.108.064501.DC1.html

References http://www.plantcell.org/content/21/2/545.full.html#ref-list-1

This article cites 46 articles, 22 of which can be accessed free at:

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists