Rhizobium Lipo-chitooligosaccharide Signaling Triggers Accumulation of Cytokinins in Medicago truncatula Roots

Accepted Manuscript Rhizobium Lipo-Chitooligosaccharide Signaling Triggers Accumulation of Cytokinins in Medicago Truncatula Roots Arjan van Zeijl, Rik H.M. Op den Camp, Eva E. Deinum, Tatsiana Charnikhova, Henk Franssen, Huub J.M. Op den Camp, Harro Bouwmeester, Wouter Kohlen, Ton Bisseling, René Geurts PII:

S1674-2052(15)00178-1

DOI:

10.1016/j.molp.2015.03.010

Reference:

MOLP 113

To appear in:

MOLECULAR PLANT

Received Date: 10 October 2014 Revised Date:

12 March 2015

Accepted Date: 15 March 2015

Please cite this article as: van Zeijl A., Op den Camp R.H.M., Deinum E.E., Charnikhova T., Franssen H., Op den Camp H.J.M., Bouwmeester H., Kohlen W., Bisseling T., and Geurts R. (2015). Rhizobium Lipo-Chitooligosaccharide Signaling Triggers Accumulation of Cytokinins in Medicago Truncatula Roots. Mol. Plant. doi: 10.1016/j.molp.2015.03.010. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT RUNNING TITLE: Cytokinin biosynthesis is rhizobium LCO induced

TITLE:

RI PT

RHIZOBIUM LIPO-CHITOOLIGOSACCHARIDE SIGNALING TRIGGERS ACCUMULATION OF CYTOKININS IN MEDICAGO TRUNCATULA ROOTS

Arjan van Zeijla, Rik H.M. Op den Campa, Eva E.Deinuma,b, Tatsiana Charnikhovac, Henk Franssena, Huub J.M. Op den Campd, Harro Bouwmeesterc, Wouter Kohlena, Ton Bisselinga,e

Department of Plant Sciences, Laboratory of Molecular Biology, Wageningen University,

M AN U

a

SC

and René Geurtsa,1

Droevendaalsesteeg 1, 6708 PB, Wageningen, The Netherlands. b

Department of Systems Biophysics, FOM institute AMOLF, Science Park 104, 1098 XG

Amsterdam, The Netherlands. c

Department of Plant Sciences, Laboratory of Plant Physiology, Wageningen University,

d

TE D

Droevendaalsesteeg 1, 6708 PB, Wageningen, The Netherlands.

Department of Microbiology, IWWR, Radboud University Nijmegen, Heyendaalseweg 135,

6525 AJ, Nijmegen, The Netherlands. e

College of Science, King Saud University, Post Office Box 2455, Riyadh 11451, Saudi

Address correspondence to René Geurts ([email protected]).

AC C

1

EP

Arabia.

SHORT SUMMARY:

In legumes, cytokinin signalling is both required as well as sufficient for initiation of rhizobium root nodules. Here, we show that the majority of transcriptional changes induced by rhizobium LCO signals are dependent on the cytokinin receptor MtCRE1. Additionally, we show that exposure to rhizobium LCO signals induces rapid accumulation of cytokinin in Medicago truncatularoots.

1

ACCEPTED MANUSCRIPT 1

ABSTRACT

2

The legume rhizobium symbiosis is initiated uponperception of bacterial secreted lipo-chito

4

oligosaccharides (LCOs). Perception of these signals by the plant initiates a signalling

5

cascade that leads to nodule formation. Several studies have implicated a function for

6

cytokinin in this process. However, whether cytokinin accumulation and subsequent

7

signalling are an integral part of rhizobium LCO signalling remains elusive. Here we show

8

that cytokinin signalling is required for the majority of transcriptional changes induced by

9

rhizobium LCOs. Additionally, we demonstrate that several cytokinins accumulate in the root

10

susceptible zone three hours post rhizobium LCO application, including the biologically most

11

active cytokinins trans-zeatin and isopentenyl adenine. These responses are dependent on

12

CCaMK; a key protein in rhizobial LCO induced signalling. Analysis of the ethylene

13

insensitive Mtein2/Mtsickle mutant showed that LCO-induced cytokinin accumulation is

14

negatively regulated by ethylene. Together with transcriptional induction of ethylene

15

biosynthesis genes, it suggests a feedback loop negatively regulating LCO signalling and

16

subsequent cytokinin accumulation. We argue that cytokinin accumulation is a key step in the

17

pathway leading to nodule organogenesis and that this is tightly controlled by feedback loops.

M AN U

SC

RI PT

3

EP

20

Keywords: Medicago truncatula, cytokinin, ethylene, rhizobium, nodule, lipo-chitooligo saccharides, Nod factor, CRE1

AC C

19

TE D

18

2

ACCEPTED MANUSCRIPT 21

INTRODUCTION

22

Legumes can engage endosymbiotically with different nitrogen fixing rhizobial species. Upon

24

signal exchange between host and microsymbiont, a developmental program is initiated that

25

gives rise to a new organ; the root nodule (Crespi and Frugier, 2008; Kouchi et al., 2010;

26

Murray, 2011; Yokota and Hayashi, 2011). These nodules are colonized intracellularly,

27

providing a niche to the bacteria optimal for the fixation of atmospheric nitrogen. The key

28

signal molecules that initiate nodule organogenesis are bacterial secreted lipo-chitooligo

29

saccharides (LCOs), also known as Nodulation (Nod) factors (Lerouge et al., 1990).In some

30

legumes, these bacterial signals are even sufficient to trigger the complete developmental

31

program resulting in formation of (empty) nodules (Mergaert et al., 1993; Relić et al., 1994;

32

Stokkermans and Peters, 1994; Truchet et al., 1991). Rhizobium LCOs are perceived by

33

LysM domain receptor kinases at the root epidermis (Arrighi et al., 2006; Limpens et al.,

34

2003; Radutoiu et al., 2003; 2007).This activates a signalling cascade consisting of a plasma

35

membrane localized LRR-type receptor kinase (named MtDMI2 in Medicago truncatula), a

36

cation ion channel (named MtDMI1 in M. truncatula) and several components of the nuclear

37

pore complex (Oldroyd, 2013). Activation of this signalling cascade results in nuclear calcium

38

oscillations, which are interpreted by a nuclear localized Calcium Calmodulin Kinase

39

(CCaMK named MtDMI3 in M. truncatula) (Levy et al., 2004; Mitra et al., 2004). CCaMK

40

activates the transcriptional regulator CYCLOPS, which subsequently activates transcription

41

of downstream targets, including the transcription factor NODULE INCEPTION (NIN)

42

(Singh et al., 2014).The rhizobium LCO signalling pathway interacts with plant hormone

43

homeostasis at multiple levels. For example, activation of cytokinin phosphorelay signalling

44

is essential for nodule formation, whereas ethylene signalling is known to inhibit this process

45

(Murray et al., 2007; Oldroyd et al., 2001a; Tirichine et al., 2007). Unravelling how these

46

hormonal pathways are integrated in symbiotic LCO signalling is key to understand how

47

legume plants establish symbiosis with rhizobium.

SC

M AN U

TE D

EP

AC C

48

RI PT

23

49

Ethylene negatively regulates rhizobium LCO-induced signalling and subsequent nodule

50

formation. The first and rate-limiting step in ethylene biosynthesis is conversion of S-

51

adenosyl-L-methionine (S-AdoMet) into 1-aminocyclopropane-1-carboxylic acid (ACC) by

52

ACC synthases (ACS) (Lin et al., 2009). Subsequently, ACC is converted into ethylene by

53

ACC oxidase (ACO) activity (Lin et al., 2009). Interestingly, in pea (Pisum sativum)roots,

54

ACO genes are expressed in pericycle, endodermal and inner cortical cells opposite phloem 3

ACCEPTED MANUSCRIPT poles, limiting nodule initiation to opposite protoxylem poles(Heidstra et al., 1997).Ethylene

56

is perceived by a set of ER-localized receptors that control activity of the ER localized EIN2

57

protein. EIN2 forms a central component of the ethylene signalling pathway(Alonso et al.,

58

1999; Lin et al., 2009; Roman et al., 1995).Upon activation EIN2 is cleaved, by which its C-

59

terminal domain is released and can move to the nucleus to modulate transcription

60

(Merchante et al., 2013). The Mtein2 mutant in M. truncatula, named Mtsickle, is ethylene

61

insensitive and displays hyper nodulation upon rhizobium inoculation, forming up to ten

62

times more nodules in a distinct zone (Penmetsa et al., 2003; 2008). Ethylene has been shown

63

to affect LCO signalling at an early step in the signalling cascade, most likely upstream or at

64

the point of calcium oscillation. This is based on the observation that in presence of ethylene

65

an increased LCO concentration is required to initiate this cellular calcium response.

66

Furthermore, a decrease in oscillation frequency is observed in Mtein2/Mtsickle compared to

67

wild type(Oldroyd et al., 2001a).

M AN U

68

SC

RI PT

55

In contrast to ethylene, cytokinin acts as an important positive regulator of nodule

70

organogenesis. This was first illustrated by physiological studies, which showed that external

71

application of cytokinin triggers the formation of nodule-like structures on the roots of several

72

legume plants (Cooper and Long, 1994; Heckmann et al., 2011; Mathesius et al., 2000;

73

Torrey, 1961).Genetic studies revealed involvement of a specific cytokinin receptor in

74

nodulation;MtCRE1 and LjLHK1 in M. truncatula and Lotus japonicas, respectively. Both

75

receptors

76

AtAHK4/AtCRE1 cytokinin receptor, which functions redundantly to cytokinin receptors

77

AtAHK2 and AtAHK3(de León et al., 2004; Franco-Zorrilla et al., 2002; Gonzalez-Rizzo et

78

al., 2006; Higuchi et al., 2004; Op den Camp et al., 2011; Plet et al., 2011).Not only are

79

MtCRE1 and LjLHK1 essential for nodule organogenesis, gain-of-function mutations that

80

make these receptors hypersensitive to cytokinin lead to spontaneous nodule formation

81

(Ovchinnikova et al., 2011; Tirichine et al., 2007). Several symbiotic genes have been

82

reported to be responsive to cytokinin signalling, including the transcriptional regulators

83

NSP1, NSP2, ERN1 and NIN(Plet et al., 2011). Rhizobium LCO signalling also induces

84

expression of typical cytokinin responsive genes like type-A Response Regulators (Gonzalez-

85

Rizzo et al., 2006; Op den Camp et al., 2011; Plet et al., 2011; Vernie et al., 2008). This

86

further underlines integration of cytokinin phosphorelay signalling in the LCO signalling

87

pathway.

putatively

orthologous

to

the

arabidopsis

(Arabidopsis

thaliana)

AC C

EP

are

TE D

69

88

4

ACCEPTED MANUSCRIPT Naturally occurring cytokinins are N6-substituted adenine derivatives that can be classified as

90

either aromatic or isoprenoid, depending on their side chain. Of both types, the isoprenoid-

91

type cytokinins are most abundant in plants, and their biosynthesis is well studied(Kamada-

92

Nobusada and Sakakibara, 2009; Sakakibara, 2006). The rate-limiting step in the biosynthesis

93

of isoprenoid-type cytokininsis catalysed by the isopentenyl transferases (IPTs), which can be

94

classified into two groups depending on their substrate. Adenylate-IPTs use adenosine 5’-

95

phosphate (AMP, ADP or ATP) as substrate and are involved in the production of isopentenyl

96

adenine (iP), trans-zeatin (tZ) and dihydrozeatin (DZ), whereas tRNA-IPTs use tRNAs as

97

substrate and are involved in biosynthesis of cis-zeatin (cZ). Biosynthesis of tZ-type

98

cytokinins involves hydroxylation of iP-nucleotides by a group of cytochrome P450 mono

99

oxygenases, which in arabidopsis are named AtCYP735A1 and AtCYP735A2(Takei et al.,

100

2004). Release of bioactive cytokinins from their nucleotide precursors is catalysed by a

101

group of 5’-monophosphate phosphoribohydrolases, referred to as LONELY GUYs (LOGs)

102

(Kurakawa et al., 2007; Kuroha et al., 2009). Cytokinin oxidases/dehydrogenases (CKX) are

103

involved in cytokinin degradation (Schmülling et al., 2003), whereas adenine phosphoribosyl

104

transferases (APTs) inactivate cytokinin by converting them back to their nucleotide-form

105

(Zhang et al., 2013). Recently, several studies reported a link between cytokinin biosynthesis

106

genes

107

L.japonicusLjIPT3expression increases within three hours after rhizobial inoculation, and in

108

M. truncatula both MtLOG1 and MtLOG2 are up-regulated during later stages of nodule

109

formation. Moreover, RNA interference-mediated knockdown of LjIPT3 and MtLOG1 leads

110

to a reduction of the total number of nodules formed per plant (Chen et al., 2014; Mortier et

111

al., 2014). However, what remains unknown is whether cytokinins are produced as a

112

signalling intermediate upon rhizobium LCO perception, and to what extent this contributes

113

to rhizobium LCO-induced signalling.

SC

formation

M AN U

nodule

(Chen

et

al.,

2014;

Mortier

et

al.,

2014).

In

EP

TE D

and

AC C

114

RI PT

89

115

In this study we used RNA sequencing to determine the role of cytokinin in rhizobium LCO-

116

induced signalling in M. truncatula. This revealed that the majority of transcriptional changes

117

induced three hours post rhizobium LCO application are dependent on the cytokinin receptor

118

MtCRE1. Furthermore, we show that LCO treatment induces expression of cytokinin as well

119

as ethylene biosynthesis genes. Measurements of the cytokinin concentration in M. truncatula

120

roots showed an accumulation of bioactive cytokinins 3 h post rhizobium LCO application.

121

The latter response is inhibited by ethylene, suggesting presence of a negative feedback loop.

122

5

ACCEPTED MANUSCRIPT 123

RESULTS

124 125

Majority of Nod factor-induced transcriptional changes are CRE1-dependent

126

Activation of the cytokinin receptor MtCRE1/LjLHK1 is both sufficient and essential to

128

trigger nodule organogenesis(Murray et al., 2007; Tirichine et al., 2007). However, the

129

temporal and mechanistic regulation of the cytokinin signalling pathway upon rhizobium

130

LCO signalling remains unknown. To gain insight in the regulation of this signalling

131

pathway, we studied the role of MtCRE1 in rhizobium LCO induced gene expression by RNA

132

sequencing (RNA-seq). Three biological replicates of M. truncatula wild type and Mtcre1

133

mutant were grown vertically on agar-solidified Fåhraeus medium (0.2 mM Ca(NO3)2)

134

supplemented with 1 µM amino ethoxyvinylglycine (AVG), commonly used in in vitro

135

nodulation assays to increase nodulation ability(e.g. Berrabah et al., 2014; Domonkos et al.,

136

2013; Gourion et al., 2013; Haney et al., 2011; Horváth et al., 2011). Seedling roots were

137

treated with Sinorhizobium meliloti LCOs (~10-9 M) for 3 h, and RNA was isolated from zone

138

of the root susceptible to rhizobium infection. This zone, hereafter referred to as the

139

susceptible zone, encompasses a region of ~5 mm and is analogous to the distal region of the

140

elongation zone and the entire differentiation zone of the root. We chose to sample after three

141

hours, as this time point represents a stage in the interaction at which cytokinin-dependent

142

transcriptional changes are reported, whereas mitotic activation of pericycle and cortical cells

143

is not yet occurring(Op den Camp et al., 2011; Plet et al., 2011; Xiao et al., 2014). As

144

cytokinin signalling is important for root development(Bianco et al., 2013), we first compared

145

the transcriptomes of mock treated wild-type and Mtcre1 mutant plants. This revealed 69

146

genes that display a significant transcriptional difference using a minimal fold change (FC)

147

larger than two and a Bonferroni-corrected p-value smaller than 0.001 (Table S1). This

148

limited number of differentially expressed genes between wild type and Mtcre1 is consistent

149

with the absence of anyobvious root phenotype in Mtcre1(Plet et al., 2011). Using the same

150

criteria, we identified 814 genes in M. truncatula wild-type roots that display differential

151

expression upon rhizobium LCO signalling. Of these genes, 609 were up- and 205 were

152

down-regulated (Table S2). Among the up-regulated genes are MtENOD11 and MtERN1 as

153

well as the cytokinin responsive genes MtNIN, MtCLE13, MtRR4, MtRR5, MtRR8, MtRR9

154

and MtRR11 that were found in previous studies(Figure S1, Ariel et al., 2012; Gonzalez-Rizzo

155

et al., 2006; Journet et al., 2001; Mortier et al., 2010; Op den Camp et al., 2011). In case of

156

the Mtcre1 mutant only 243 genes were differentially expressed upon LCO application (233

AC C

EP

TE D

M AN U

SC

RI PT

127

6

ACCEPTED MANUSCRIPT up-regulated and 10 down-regulated). Comparing the transcriptional changes observed in wild

158

type and Mtcre1 mutant after rhizobium LCO application demonstrates that a large set of 596

159

genes, which corresponds to ~73% of the genes differentially expressed in wild type, are

160

dependent on MtCRE1-mediated signalling (FC>2, p<0.001) (Figure 1). Twenty-five genes

161

showed significant differential expression upon LCO treatment in the Mtcre1 mutant

162

specifically (FC>2, p<0.001) (Table S3).

RI PT

157

163

To validate our data, the expression of the set of differentially regulated genes is compared to

165

that on the Medicago Gene Expression Atlas (Benedito et al., 2008). To this end, an averaged

166

expression was calculated over several studies comparable to ours, consisting of samples

167

taken at 6 and 24 h post LCO treatment and 1, 3 and 5 days post rhizobium inoculation

168

(Benedito et al., 2008; Breakspear et al., 2014; Czaja et al., 2012). Of the 814 differentially

169

regulated genes identified in wild-type M. truncatula, 623 genes are represented by at least

170

one unique probe on the array. Of these, ~82% (508 genes) behaves similar in the reported

171

microarray studies (Figures S2 and S3). Approximately 10% (63 genes) of the genes

172

responded exclusively in our experiment, whereas ~8% (52 genes) responds differently

173

between both sets (Figure S2; Table S5). Overall, this comparison suggests that the RNA-seq

174

data are in line with previous studies.

M AN U

TE D

175

SC

164

Gene Ontology (GO) enrichment analysis identified a significant overrepresentation of 48 GO

177

terms associated with genes differentially regulated in wild-type M. truncatula specifically

178

(p<0.05). These included phosphorelay response regulator activity, adenine phosphoribosyl

179

transferase activity and 1-aminocyclopropane-1-carboxylate synthase activity (Table S6). This

180

points at an induction of cytokinin signalling, cytokinin turnover and ethylene biosynthesis,

181

respectively. The same analysis on genes differentially expressed in both wild type and

182

Mtcre1 mutant after rhizobium LCO exposure revealed 27 significantly enriched GO terms

183

(p<0.05) (Table S7). Among these are GO terms associated with chitin breakdown and

184

phosphorylation of inositol phosphates. Altogether, these data show that rhizobium LCO

185

signalling induces a rapid transcriptional reprogramming, which to a large extent is dependent

186

on a functional MtCRE1 cytokinin receptor. This implies a key function for cytokinin at the

187

preinfection stage of the rhizobium-legume interaction, 3 h post rhizobium LCO perception.

AC C

EP

176

188 189

Spatiotemporal localization of cytokinin response

190

7

ACCEPTED MANUSCRIPT Since cytokinin signalling is an essential step in nodule formation, we questioned in which

192

cell layers cytokinin signalling is induced upon rhizobium LCO perception. To this end, we

193

used the synthetic cytokinin reporter TCS driving β-glucuronidase (TCS:GUS)(Muller and

194

Sheen, 2008). This chimeric cytokinin reporter construct was introduced in M. truncatula

195

roots using Agrobacterium rhizogenes-mediated transformation. Spatiotemporal expression

196

analysis revealed a moderate level of variation between individual transgenic roots carrying

197

TCS:GUS. Despite this level of variation, faint expression of the TCS:GUS reporter is

198

observed in some, but not all, pericycle cells in mock treated roots. This is consistent with

199

observations in arabidopsis, where expression of the TCS reporter was observed in xylem

200

pole pericycle cells specifically(Bielach et al., 2012). Application of rhizobium LCOs (~10-9

201

M) increased expression of the TCS:GUS reporter in the susceptible zone specifically in

202

~30% of the transgenic roots (n=12/38). Sectioning of these root segments revealed induction

203

of TCS:GUS in the cortex, endodermis and pericycle (Figure 2). This suggests that cytokinin

204

acts as a non-cell-autonomous signal, during the early stages of the legume-rhizobium

205

interaction.

M AN U

SC

RI PT

191

206 207

Rhizobium LCO signalling induces a change in the bioactive cytokinin pool

TE D

208

Mining the transcriptome data we noted differential expression of genes involved in cytokinin

210

biosynthesis and/or activation. A putative IPT (Medtr2g022140), CYP735A (Medtr6g017325)

211

and LOG (Medtr1g057020) homolog were induced 3 h post LCO application (Figure 3A). We

212

refer to these genes as MtIPT1, MtCYP735A1 and MtLOG3 based on consecutive numbering

213

in relation to a previous study(Mortier et al., 2014). Expression of MtCYP735A1 is induced

214

~6-fold in wild type and ~2.5-fold in Mtcre1 post LCO application (Figure 3A),although

215

induction in Mtcre1was not found to be significant when using the stringent correction for

216

multiple testing. This suggests that LCO-induced expression of MtCYP735A1 may be partly

217

dependent on MtCRE1. In contrast, MtIPT1 expression is induced ~4-fold in both wild-type

218

M. truncatula and Mtcre1 mutant, indicating that induction of this gene by rhizobium LCOs is

219

MtCRE1 independent. This suggests that rhizobium LCO induced signalling leads to (local)

220

cytokinin biosynthesis. Besides cytokinin synthesis genes, also genes involved in cytokinin

221

metabolism are induced by rhizobium LCOs (Figure 3B). These are two putative APT

222

(Medtr3g106780, Medtr5g012210) and two putative CXK (Medtr4g126150, Medtr2g039410)

223

homologs, referred to as MtAPT1, MtAPT2 and MtCKX2, MtCKX3, respectively(Ariel et al.,

224

2012). These genes are induced in wild-type M. truncatula, but not Mtcre1. As also MtLOG3

AC C

EP

209

8

ACCEPTED MANUSCRIPT 225

is induced in an MtCRE1 dependent manner, this suggests a set of feedback loops tightly

226

regulating LCO induced cytokinin levels. Therefore, these data indicate that LCO signalling

227

may induce a rapid change in the cytokinin concentration in the M. truncatula root susceptible

228

zone.

229

To obtain better insight in the regulation of cytokinin biosynthesis genes by rhizobium LCOs,

231

we studied their expression at different time points (30 min, 1 h and 3 h post LCO exposure),

232

and in absence or presence of the ethylene synthesis inhibitor AVG, using quantitative RT-

233

PCR (qRT-PCR). LCO responsiveness of these samples was confirmed by strong induction of

234

MtNIN(Figures4A and 5A).Analysis of MtIPT1 expression showed this gene is first induced

235

at 1 h post LCO treatment, whereas expression of MtCYP735A1 and MtLOG3 is only induced

236

at 3 h post LCO exposure (Figure 4B).Expression ofMtIPT1and MtCYP735A1is induced in

237

both presence and absence of AVG (Figure 5B), which excludes that induction of both

238

cytokinin biosynthesis genes by rhizobium LCOs results from low ethylene levels. Analysis

239

of the expression of the cytokinin oxidases MtCKX2 and MtCKX3 and the cytokinin response

240

regulators MtRR4 and MtRR9shows that cytokinin responses are first induced at 3 h post LCO

241

exposure (Figure 4A and 4C). Furthermore, our data indicate that induction of cytokinin

242

signalling by rhizobium LCOs might be inhibited by ethylene, as in absence of AVG only

243

MtRR9, but not MtRR4 expression is induced, and this response is less in absence of AVG

244

(~3.5 vs 10-fold, respectively) (Figure 5A). Altogether, these data suggest rapid activation of

245

cytokinin biosynthesis upon rhizobium LCO perception.

SC

M AN U

TE D

EP

246

RI PT

230

To test whether rhizobium LCO perception induces a change in the cytokinin concentration,

248

we measured cytokinin levels in the M. truncatula root susceptible zone of wild type and the

249

Mtccamk/Mtdmi3mutant at 3 h post LCO exposure using ultra-performance liquid

250

chromatography coupled to tandem mass spectrometry (UPLC-MS/MS). For this, we focused

251

on isoprenoid-type cytokinins, as these are the predominant form found in plants(Sakakibara,

252

2006). Of all cytokinins measured isopentenyl adenine (iP), iP-riboside (iPR), trans-zeatin

253

(tZ), cis-zeatin (cZ), tZ-9-glycosylated (tZ9G) and cZ-riboside (cZR) could be detected and

254

quantified. No change in the levels of cZ, tZ9G and cZR were detected post LCO exposure

255

(Figure S4). In contrast, the amounts of iP, iPR and tZ did increase significantly ~15, 2.5 and

256

8 times, respectively, in wild type upon LCO application (p<0.05, Figure 6A-C).This

257

response is dependent on a functional LCO signaling cascade, as it was not observed in the

258

Mtccamk/Mtdmi3 mutant (Figure 6A-C). These results demonstrate that a number of

AC C

247

9

ACCEPTED MANUSCRIPT 259

cytokinins accumulate within three hours post rhizobium LCO perception, including the

260

biologically most active cytokinins iP and tZ(Kamada-Nobusada and Sakakibara, 2009).

261

To position the cytokinin accumulation in the rhizobium LCO signalling pathway, we

263

analysed this response in the Mtnsp1, Mtnsp2 and Mtnsp1Mtnsp2 double mutant. In contrast

264

to MtNSP2,expression of MtNSP1 in the M. truncatula wild-type root is induced ~5-fold after

265

exposure to rhizobium LCOs, whereas only a ~2.5-fold induction is observed in the Mtcre1

266

mutant (Figure S5). This indicates that LCO-induced MtNSP1 expression is partly dependent

267

on cytokinin signalling. Next, we measured the cytokinin levels in the Mtnsp1 and

268

Mtnsp2mutants and compared it to wild type. Again we observed accumulation of iP, tZ and

269

iPR in wild-type M. truncatula(Figure 6). Measurements in theMtnsp1, Mtnsp2 and

270

Mtnsp1Mtnsp2mutant roots showed basal cytokinin levels are similar to those in wild type

271

(Figure 6D-F).Like in wild type, LCO treatment induced an accumulation of cytokinin

272

inMtnsp1, Mtnsp2 and Mtnps1Mtnsp2.However, in these mutants accumulation of iP and tZ is

273

approximately 60-75% and 30-40% of that in the wild type, respectively (Figure 6D-F).

274

Taken together, we demonstrated that rhizobium LCO signalling triggers rapid accumulation

275

of bioactive cytokinin in a CCaMK dependent manner. This response is controlled only

276

partially by the GRAS-type regulators NSP1 and/or NSP2, possibly in a positive feedback

277

loop.

280

SC

M AN U

TE D

279

Ethylene inhibits symbiotic cytokinin accumulation

EP

278

RI PT

262

Previous work has shown that cytokinin can induce ethylene production (Chae et al., 2003;

282

Fuchs and Lieberman, 1968; Lorteau et al., 2001). Based on this, we hypothesized that LCO-

283

induced cytokinin accumulation might also induce ethylene production. To test this

284

hypothesis, we checked for ethylene biosynthesis genes amongst the differentially expressed

285

genes identified through RNA-seq. This revealed an MtCRE1-dependenttranscriptional

286

induction of three ACS genes (Medtr8g101750, Medtr8g101820, Medtr4g097540)by

287

rhizobium LCOs. However, close inspection ofMedtr8g101820revealed that this gene model

288

either represents a pseudogene, or an error in the current genome assembly (Mt4.0; see Figure

289

S6). For both other ACS genes -Medtr8g101750 (MtACS1) and Medtr4g097540 (MtACS2)-

290

transcriptional induction of by rhizobium LCOs is strongly supported (Figure 3C). A detailed

291

look at the expression of all putative ethylene biosynthesis genes revealed induction of an

AC C

281

10

ACCEPTED MANUSCRIPT 292

additional ACS gene (Medtr6g091760, MtACS3) just below the 2-fold threshold (p<0.001,

293

FC=1.95) (Figure3C).

294

qRT-PCR reactions on the root samples taken at different time points after LCO treatment,

296

showed that all three ACS genes are induced already at 30 min after LCO exposure (Figure

297

4D). At 3 h after LCO exposure, expression of all three ACS genes is induced at >7-fold,

298

however, we also observed dampening of expression at 1 h post LCO treatment (Figure 4D).

299

This may suggest a fast induction of ACS genes that may occur in a cytokinin-independent

300

fashion, which is followed by a strong MtCRE1-dependent induction at 3 h post LCO

301

exposure. Induction of MtACS1occurs in both presence and absence of AVG, though

302

induction is much stronger if AVG is added to the growth medium (~3.5 vs ~40-fold) (Figure

303

5C). Expression of MtACS2 and MtACS3 is induced only in presence of AVG, suggesting that

304

induction of these genes requires low ethylene levels (Figure 5C). Altogether, these results

305

indicate anMtCRE1-dependenttranscriptional induction of at least one ethylene biosynthesis

306

gene by rhizobium LCOs. This suggests that LCO-induced cytokinin accumulation promotes

307

ethylene biosynthesis.

M AN U

SC

RI PT

295

308

Ethylene is a known inhibitor of rhizobium LCO signalling and subsequent nodule formation

310

(Oldroyd et al., 2001a; Penmetsa et al., 2008). Therefore, cytokinin-dependent induction of

311

ethylene biosynthesis genes by rhizobium LCOs may restrict LCO-induced cytokinin

312

accumulation. The latter is supported by the effect of AVG on the responsiveness of

313

cytokinin response regulators to rhizobium LCOs (Figure 5A). To determine the effect of

314

ethylene on cytokinin accumulation, we exploited the M. truncatula ethylene insensitive

315

mutant Mtein2/Mtsickle. To this end, M. truncatula wild type and Mtein2/Mtsickle plants

316

were grown on medium without AVG and treated with rhizobium LCOs for 3 h. Under these

317

conditions, wild type accumulated 2-3 fold higher concentrations of iP, iPR and tZ post LCO

318

treatment (Figure7). This response seems weaker than compared to the accumulation detected

319

in plants grown on regular plant growth medium that contains 1 µM AVG (Figure 6A-C),

320

indicating an inhibitory effect of ethylene. Cytokinin measurements in Mtein2/Mtsickle

321

mutant also indicate negative regulation by ethylene. In this mutant, LCO treatment induced a

322

15-fold increase in iP concentration, a response five times stronger compared to wild type

323

(Figure 7A). Taken together, we conclude that rhizobium LCO-induced cytokinin

324

accumulation is negatively affected by ethylene.

AC C

EP

TE D

309

325

11

ACCEPTED MANUSCRIPT 326

DISCUSSION

327

Cytokinin signalling is an integral part of the signalling pathway leading to nodule

329

organogenesis. However, whether cytokinin accumulates as signalling intermediate and to

330

what extent this is required for rhizobium LCO-induced signalling remained elusive. Here, we

331

show that in M. truncatula cytokinin rapidly accumulates in the root susceptible zone upon

332

application of rhizobium LCOs. The accumulating cytokinins include iP and tZ, which are

333

among the biologically most active cytokinins(Kamada-Nobusada and Sakakibara, 2009).

334

Furthermore, we show that the majority of transcriptional changes induced three hours post

335

rhizobium LCO treatment are dependent on the cytokinin receptor MtCRE1. Therefore, we

336

conclude that cytokinin accumulation is a key step in the pathway leading to rhizobium root

337

nodule organogenesis.

SC

RI PT

328

M AN U

338

Using RNA-seq we determined the extent of LCO-induced signalling dependent on MtCRE1.

340

This revealed that the vast majority (~73%) of transcriptional changes induced 3 h post LCO

341

application are dependent on this symbiotic cytokinin receptor. Genes that are differentially

342

expressed in both wild type and Mtcre1 probably function upstream of MtCRE1 or in a

343

cytokinin-independent pathway involved in rhizobial infection(Madsen et al., 2010).

344

Examples of these are the cytokinin biosynthesis gene MtIPT1 and genes involved in chitin

345

breakdown. The latter could be involved in removing excess rhizobium LCO signals or could

346

be part of a general defence response induced by these bacterial signal molecules(Nakagawa

347

et al., 2011; Serna-Sanz et al., 2011). In addition, some level of redundancy may contribute to

348

the transcriptional changes observed in Mtcre1 upon rhizobium LCO exposure. The Mtcre1-1

349

mutant used in this study is considered a full knock-out as it has a mutation that creates a

350

premature stop codon in the middle of the kinase domain(Plet et al., 2011). Still it is reported

351

that in rare cases this mutant can develop a few nodules(Plet et al., 2011), even though we

352

didn’t observe nodules on Mtcre1-1 mutant plants in any of our experiments (data not shown).

353

In case of L. japonicus, a similar phenotype of the Ljlhk1-1 mutant could be explained by

354

redundant functioning of the paralogous genes LjLHK1A and LjLHK3, indicating existence of

355

partial redundancy for symbiotic cytokinin perception(Held et al., 2014). However, in the M.

356

truncatulaMtcre1-1 mutant redundancy by additional histidine receptors in symbiotic

357

cytokinin signalling is most likely limited, as the expression of typical cytokinin responsive

358

genes (e.g. the A-type Response Regulators) is not or only moderately induced in this mutant

359

upon LCO application (Figures 5A and S1). Therefore, we conclude that most genes induced

AC C

EP

TE D

339

12

ACCEPTED MANUSCRIPT 360

3 h post rhizobium LCO application function in the MtCRE1-dependent nodule

361

organogenesis pathway.

362

Measurements of the cytokinin concentration in M. truncatula root susceptible zones showed

364

accumulation of three cytokinins 3 h post rhizobium LCO application. This response is

365

dependent on rhizobium LCO signalling, as it was not observed in the Mtccamk/Mtdmi3

366

mutant. Our observation that both iP and its biosynthetic precursor iPR accumulate upon

367

rhizobium LCO exposure suggests that local biosynthesis is the primary source of these

368

cytokinins. This is supported by induction of the cytokinin biosynthesis gene MtIPT1 at 1 h

369

post LCO exposure in both wild type and Mtcre1 mutant. However, considering the time

370

frame in which cytokinin accumulates upon LCO signalling, we cannot exclude that the

371

increase in cytokinin concentration is regulated primarily at the protein level.

SC

RI PT

363

M AN U

372

We also determined the role of the GRAS-type transcriptional regulators NSP1 and NSP2 in

374

LCO induced cytokinin accumulation, and showed that iP and tZ accumulate to intermediate

375

levels in both single and double mutants compared to wild type. Initially it was argued that

376

these GRAS-type proteins may act as primary rhizobium LCO response factors(Smit et al.,

377

2005), for which indeed some experimental evidence was found(Hirsch et al., 2009; Mitra et

378

al., 2004). However, more in-depth genetic dissection of the rhizobium LCO signalling

379

pathway suggests a more complex role for both proteins; downstream of as well as parallel to

380

symbiotic cytokinin signalling(Madsen et al., 2010). Our finding shows that both NSP1 and

381

NSP2 may function downstream of symbiotic cytokinin biosynthesis in an autoactivation

382

loop. Such hypothesis is supported by the finding that MtNSP1 expression is induced 3 h post

383

LCO treatment, and this induction is in part dependent on MtCRE1.

EP

AC C

384

TE D

373

385

We uncovered a negative feedback relation between cytokinin and ethylene. Cytokinin

386

measurements showed that upon LCO signalling the M. truncatula Mtein2/Mtsickle knockout

387

mutant accumulates more cytokinin in the root susceptible zone than wild-type plants. This

388

implies that ethylene signalling inhibits symbiotic cytokinin accumulation. This finding is in

389

line with earlier studies that showed that ethylene interferes with LCO induced calcium

390

oscillations in and around the nuclei of activated cells(Oldroyd et al., 2001a; 2001b), and the

391

absence of cytokinin accumulation in the Mtccamk/Mtdmi3 mutant (Figure 6A-C). Our data

392

also indicate a cytokinin-dependent transcriptional induction of (at least) one ACC synthase

393

by rhizobium LCOs. This is in line with results in L. japonicus, which show that the ethylene 13

ACCEPTED MANUSCRIPT producing ACC oxidase LjACO2 is also inducedby rhizobium LCOs(Miyata et al., 2013).

395

Furthermore, previous work has shown that cytokinin can induce ethylene production (Chae

396

et al., 2003; Fuchs and Lieberman, 1968; Lorteau et al., 2001). Therefore, it is conceivable

397

that LCO signalling, through an induction of cytokinin accumulation, might induce ethylene

398

biosynthesis. This response could be part of a negative feedback signal to inhibit further LCO

399

signalling and subsequent cytokinin accumulation (Figure 8). Such a feedback mechanism

400

might be important to restrict LCO induced cytokinin accumulation to a limited number of

401

cells. This to prevent initiation of multiple nodule primordia, as can be observed in the

402

Mtein2/Mtsickle mutant. As such, this response would resemble the mechanism through

403

which nodule positioning is regulated, where ethylene biosynthesis genes are expressed

404

opposite phloem poles, delineating nodule initiation(Heidstra et al., 1997). Additionally,

405

cytokinin-induced ethylene production might be required to restrict rhizobial infections. The

406

latter is suggested by the phenotype of the L. japonicusLjlhk1 mutant, which displays

407

excessive infection thread formation upon rhizobial inoculation(Murray et al., 2007). This

408

phenotype suggests absence of a negative feedback signal to inhibit rhizobium LCO-induced

409

signalling. It is conceivable that this signal would be ethylene.

M AN U

SC

RI PT

394

410

Measurements of the cytokinin concentration in wild-type M. truncatula showed that both iP

412

and tZ accumulate post LCO exposure. Interestingly, accumulation of iP post LCO treatment

413

was elevated in the Mtein2/Mtsickle mutant compared to wild type, whereas accumulation of

414

tZ was comparable between both genotypes. This suggests that LCO-induced accumulation of

415

iP is more sensitive to negative regulation by ethylene, as compared to accumulation of tZ.

416

However, the mechanistic basis for this difference is currently unclear. Sensitivity to different

417

cytokinins differs between individual cytokinin receptors(Lomin et al., 2015; Stolz et al.,

418

2011). The arabidopsis AtAHK4/AtCRE1 cytokinin receptor possesses similar sensitivity to

419

both iP and tZ(Lomin et al., 2015; Stolz et al., 2011). AtAHK4/MtCRE1 represents that

420

closest ortholog of MtCRE1 and LjLHK1 and was shown to functionally complement the

421

nodulation defect of the Ljlhk1 mutant (Gonzalez-Rizzo et al., 2006; Held et al., 2014; Op den

422

Camp et al., 2011; Plet et al., 2011). Therefore, it seems plausible that MtCRE1 and LjLHK1

423

also display equal sensitivity to iP and tZ, leaving the individual contributions of these

424

cytokinins to nodulation undetermined.

AC C

EP

TE D

411

425 426

Recent work on L. japonicus showed that cytokinins might function not only to induce

427

nodulation, but also suppress nodule formation as part of autoregulation of nodulation 14

ACCEPTED MANUSCRIPT (AON)(Sasaki et al., 2014). Activation of CLE-RS1/2-HAR1 signalling increased levels of

429

the iPRP cytokinin precursors in shoots, suggesting a role for cytokinins in AON. Increased

430

iPRP levels in shoots was linked to an induction of LjIPT3 in shoots at 3 days post rhizobium

431

inoculation (Sasaki et al., 2014). This response is much slower compared to accumulation of

432

cytokinins that we observed, which occurred at 3 h post LCO application. Therefore, it is

433

highly unlikely that the increase in cytokinin concentrations that we observed is caused by

434

activation of AON. Furthermore, Sasaki et al. (2014) did not report increased levels of

435

cytokinin in roots after activation of AON. Therefore, it remains undetermined whether

436

cytokinins function in roots to repress nodule formation upon activation of AON. Cytokinin

437

as part of AON seems to function upstream of TML, a Kelch-repeat containing F-box protein

438

(Sasaki et al., 2014). TML appears to inhibit nodule formation downstream of cytokinin

439

signalling, as spontaneous nodule formation induced by snf2, a gain-of-function allele of the

440

cytokinin receptor LjLHK1, is also suppressed by TML(Takahara et al., 2013). Identification

441

of the targets of TML might show if and how cytokinins might function to both induce and

442

suppress nodule initiation.

M AN U

SC

RI PT

428

443

Nodule primordium formation results from mitotic activation of root cortical and pericycle

445

cells, and is associated with the formation of a local auxin maximum(Huo et al., 2006;

446

Mathesius et al., 1998; Pacios-Bras et al., 2003; Takanashi et al., 2011; van Noorden et al.,

447

2007). Mathematical modelling revealed that creation of such auxin maximum is most likely

448

resulting from a reduction in auxin transport(Deinum et al., 2012). In arabidopsis, cytokinin

449

has been shown to affect auxin transport by targeting the PIN auxin efflux carriers at both the

450

gene expression and protein level (Dello Ioio et al., 2008; Marhavy et al., 2011; 2014). Also

451

in legumes, cytokinin seems to affect auxin transport. In Mtcre1 mutant roots expression of

452

several PIN genes is affected, and the polar auxin transport rate is increased(Plet et al., 2011).

453

Furthermore, reduction of the polar auxin transport rate observed upon inoculation with

454

rhizobia, does not occur in Mtcre1 mutant roots (Plet et al., 2011). Based on this, it is

455

hypothesized that cytokinin is the signal that induces auxin accumulation in the root cortex

456

and pericycle upon epidermal perception of rhizobium LCOs (Suzaki et al., 2013). Rapid

457

accumulation of cytokinin and induction of cytokinin signalling in root cortex, endodermis

458

and pericycle, as observed by TCS:GUS expression, suggests that cytokinin functions non-

459

cell-autonomously. We did not observe activation of TCS:GUS expression in the root

460

epidermis post rhizobium LCO treatment. This would suggest that cytokinin signalling and

461

possibly cytokinin biosynthesis would take place in the root cortex. However, studies on

AC C

EP

TE D

444

15

ACCEPTED MANUSCRIPT cytokinin response regulators did show activation of cytokinin responses in the root epidermis

463

post rhizobium LCO-induced signalling(Ariel et al., 2012; Lohar et al., 2004; Op den Camp et

464

al., 2011). A recent study on redundant function of cytokinin receptors in L. japonicus also

465

suggests a role for epidermal cytokinin signalling in root nodule initiation(Held et al., 2014).

466

The authors show that cytokinin receptors LjLHK1A and LjLHK3 can partially rescue the

467

Ljlhk1 mutant phenotype. However, in this case nodule formation is only initiated after

468

rhizobium bacteria reach the root cortex. As only LjLHK1 is expressed in the root epidermis

469

this suggests that epidermal cytokinin signalling is required for nodule formation(Held et al.,

470

2014). This is consistent with a function for cytokinin as systemic signal and suggests that

471

cytokinin might bridge the gap between epidermal perception of rhizobium LCOs and

472

initiation of nodule primordia in the interior cell layers of the legume root.

SC

RI PT

462

473

Taken together, we demonstrated that rhizobium LCOs cause an increase in the abundance of

475

various cytokinins prior to the first mitotic divisions. It is now a challenge to unveil the

476

molecular mechanism behind the accumulation of this hormone. This will provide insight into

477

how existing developmental modules have been co-opted during the evolution of the legume-

478

rhizobium root nodule symbiosis.

AC C

EP

TE D

M AN U

474

16

ACCEPTED MANUSCRIPT 479

METHODS

480 481

Plant materials and treatments

482

Medicago truncatula Jemalong A17, Mtccamk/Mtdmi3 (TRV25)(Catoira et al., 2000),

484

Mtnsp1-1 (B85)(Catoira et al., 2000; Smit et al., 2005),Mtnsp2-2 (0-4)(Kalo et al., 2005;

485

Oldroyd and Long, 2003), Mtnsp1Mtnsp2(Liu et al., 2011), Mtcre1-1(Plet et al., 2011)and

486

Mtein2/Mtsickle-1 mutant(Penmetsa and Cook, 1997) seedlings were grown vertically for

487

three days on modified Fåhraeus medium agar plates with low nitrate (including 0.2 mM

488

Ca(NO3)2) supplemented with 1 µM AVG, unless stated otherwise(Fåhraeus, 1957). Then,

489

Sinorhizobium meliloti 2011 Nod factors (~10-9 M) or water as a control was pipetted on top

490

of the root susceptible zone. Roots were exposed for three hours and subsequently 1 cm root

491

pieces were cut just above the root tip and snap-frozen in liquid nitrogen (~n=50 for cytokinin

492

extractions, ~n=15 for RNA isolation). For all experiments plants were grown in an

493

environmentally controlled growth chamber at 20°C with a 16h-light/8h-dark cycle and 70%

494

relative humidity.

M AN U

SC

RI PT

483

495

497

Vectors and constructs

TE D

496

The synthetic cytokinin reporter element TCS was recombined into a pENTR-D-

499

Topo(Invitrogen, Carlsbad, USA) thereby creating pENTR1-2_DR5 and pENTR1-2_TCS,

500

respectively (Muller and Sheen, 2008; Ulmasov et al., 1997). Subsequently both constructs

501

were recombined into pKGWFS7-RR containing a GUS-GFP fusion reporter by a gateway

502

reaction (Invitrogen, Carlsbad, USA). All binary destination vectors used in this study contain

503

AtUBQ10pro:DsRed1 as selectable marker(Karimi et al., 2002). All cloning vectors and

504

constructs are available upon request from our laboratory or via the Functional Genomics unit

505

of the Department of Plant Systems Biology (VIB-Ghent University).

507

AC C

506

EP

498

Plant transformation and histology

508 509

Agrobacterium rhizogenes-mediated root transformation was used to transform M. truncatula

510

(Jemalong A17) as described by Limpens et al. (2004). Transgenic roots were selected based

511

on DsRED1 expression. Transgenic roots were transferred to low nitrate Fåhraeus plates

512

(including 0.2mM Ca2(NO3)2) supplemented with 1 µM AVG three weeks after 17

ACCEPTED MANUSCRIPT transformation. After five days on these plates, Sinorhizobiummeliloti2011 Nod factors (~10-9

514

M) or water as a control was pipetted on top of the root zone susceptible to rhizobium

515

infection. Roots were GUS stained three hours post treatment and fixed as described in

516

Limpens et al. (2005). Microtome sections of 9 µm were stained with ruthenium red and

517

photographed using a Leica DM5500B microscope equipped with a DFC425C camera (Leica

518

Microsystems, Wetzlar, Germany). Images were digitally processed using Photoshop CS6

519

(Adobe Systems, San Jose, California).

RI PT

513

520 521

RNA sequencing

SC

522

RNA was isolated from snap-frozen roots samples using the plant RNA kit (E.Z.N.A, Omega

524

Biotek, Norcross, USA) as described in the manufacturer’s protocol. RNA was sequenced at

525

BGI Tech Solutions (Beijing Genomics Institute, HongKong, China) using the Illumina

526

Truseq (Transcriptome) protocol utilizing an Illumina Hiseq 2000 instrument. This generated

527

90 bp paired-end reads with an average insert size of 130-160 bp. In total 24-26 million clean

528

reads were generated for each sample (Table S8). Sequencing data were analysed by mapping

529

RNA-seq reads against the M. truncatula genome (Mt4.0) using the RNA-seq module

530

implemented in CLC Genomics Workbench v7.5 (CLC bio, Aarhus, Denmark) with default

531

settings, with the exception of the similarity fraction, which was changed to 0.95. As such,

532

>95% of the reads generated for each sample were successfully mapped against the M.

533

truncatula genome (Mt4.0), of which >90% mapped in pairs (Table S8). Differentially

534

expressed genes were selected based on a minimal fold change of 2 and a Bonferroni-

535

corrected p-value<0.001, as determined using the EdgeR test implemented in CLC(Robinson

536

et al., 2010). GO enrichment analysis was performed using the Hypergeometric tests on

537

annotations method implemented in CLC.

539 540

TE D

EP

AC C

538

M AN U

523

Quantitative RT-PCR

541

RNA was isolated from snap-frozen roots samples using the plant RNA kit (E.Z.N.A, Omega

542

Biotek, Norcross, USA) as described in the manufacturer’s protocol. cDNA was synthesized

543

from 1 µg total RNA using i-script cDNA synthesis kit (Bio-Rad, Hercules, USA) as

544

described in the manufacturer’s protocol. Real time qRT-PCR was set up in a 10µl reaction

545

system with 2× iQ SYBR Green Super-mix (Bio-Rad, Hercules, USA). Experimental setup

546

and execution have been conducted using a CFX Connect optical cycler, according to the 18

ACCEPTED MANUSCRIPT protocol provided by the manufacturer (Bio-Rad, Hercules, USA). All primers including the

548

genes used for normalization (MtUBQ10 and MtPTB) are given in Table S9. Data analysis

549

was performed using CFX Manager 3.0 software (Bio-Rad, Hercules, USA). Cq values of 32

550

and higher were excluded from the analysis, though still checked for transcriptional induction

551

(see Table S9). Statistical significance was determined based on student’s t-test (p<0.01), as

552

implemented in CFX Manager 3.0 software (Bio-Rad, Hercules, USA).

RI PT

547

553 554

Cytokinin extraction

555

For extraction of cytokinins from M. truncatula root material, ~100 mg of snap-frozen plant

557

material (root susceptible zones) was used. Tissue was ground to a fine powder using 3-mm

558

stainless steel beads at 50 Hz for one minute in a TissueLyser LT (Qiagen, Germantown,

559

USA). Ground root samples were extracted with 1 ml of 100% methanol (MeOH) containing

560

2.5 mM of diethyldithiocarbamic acid and stable isotope-labeled cytokinin internal standards

561

(IS, Table S10 numbers 1-7)were used at a concentration of 100 pmolper compound per

562

sample. Samples were extracted by short vortexing and ultrasonication during 10 min.

563

Subsequently, samples were centrifuged at 2000 rpm for 10 min in a tabletop centrifuge at

564

room temperature (RT). Supernatants were transferred to 4 ml glass vials. Pellets were re-

565

extracted with 1 ml 100% MeOH for one hour on a shaker at 4°C. After centrifugation as

566

above, both supernatants were pooled and MeOH evaporated in a speed vacuum system

567

(SPD121P, ThermoSavant, Hastings, UK) at RT. Residues were resuspended in 50 µl MeOH

568

and then diluted in 3 ml of water before loading on a 50 mg GracePure SPE C18-max

569

cartridge (Grace, Columbia, USA). The cartridge was equilibrated with 2 ml of water prior to

570

sample loading. Subsequently the cartridge was washed with 1 ml of water and eluted with 1

571

ml of 100% acetone. The acetone was evaporated in a speed vacuum system at RT and the

572

residue resuspended in 200 µl acetonitrile:water:formic acid (10:90:0.1, v/v/v). The sample

573

was filtered through a 0.45 µm Minisart SRP4 filter (Sartorius, Goettingen, Germany) and

574

stored at -20°C.

AC C

EP

TE D

M AN U

SC

556

575 576

Cytokinin detection and quantification by liquid chromatography-tandem mass

577

spectrometry

578 579

Analyses of cytokinins in M. truncatula root extracts were performed by comparing retention

580

times and mass transitions with those of cytokinin standards (Table S10) using a Waters 19

ACCEPTED MANUSCRIPT XevoTQ mass spectrometer equipped with an electrospray ionization source coupled to an

582

Acquity UPLC system (Waters, Milford, USA) using settings as previously described

583

(Kohlen et al., 2011; 2012). Chromatographic separations have been conducted on an Acquity

584

UPLC BEH C18 column (100 mm, 2.1 mm, 1.7 mm; Waters, USA) by applying a

585

water/acetonitrile gradient. The water/acetonitrile gradient started from 0.2% (v/v) of

586

acetonitrile for 1.5 min and a rise to 20% (v/v) of acetonitrile in 8.5 min. To wash the column

587

the water/acetonitrile gradient was increased to 70% (v/v) acetonitrile in a 1.0 min gradient,

588

which was maintained for 0.7 min before going back to 0.2% acetonitrile using a 0.3 min

589

gradient, prior to the next run. Finally, the column was equilibrated for 2.5 min using this

590

solvent composition. The column was operated at 50ºC with a flow-rate of 0.6 ml min-1.

591

Sample injection volume was 20 µl. The mass spectrometer was operated in positive ESI

592

mode. Cone and desolvation gas flows were set to 50 and 1,000 l h-1, respectively. The

593

capillary voltage was set at 3.0 kV, the source temperature at 150ºC and the desolvation

594

temperature at 650ºC. The cone voltage was optimized for each standard compound using the

595

IntelliStart MS Console (Waters, Milford, USA). Argon was used for fragmentation by

596

collision induced dissociation (CID). Multiple reaction monitoring (MRM) was used for

597

quantification. Parent-daughter transitions for the different cytokinins and deuterium labelled

598

cytokinins were set using the IntelliStart MS Console. MRM-transitions selected for cytokinin

599

identification and quantification are shown in Table S10. Cone voltage was set to 18 eV.

TE D

M AN U

SC

RI PT

581

600

Cytokinins were quantified using a calibration curve with known amount of standards and

602

based on the ratio of the peak areas of the MRM-transition for standards to the MRM

603

transition for the corresponding deuterium labelled cytokinins (Table S10 numbers 1-7). Data

604

acquisition and analysis were performed using MassLynx 4.1 (TargetLynx) software (Waters,

605

Milford, USA). The summed area of all corresponding MRM transitions was used for

606

statistical analysis.

AC C

607

EP

601

608

Accession numbers

609

RNA-seq reads have been deposited in the Array-Express (http://www.ebi.ac.uk/arrayexpress)

610

database under accession number E-MTAB-3007.

20

ACCEPTED MANUSCRIPT 611

SUPPLEMENTARY DATA

612

Supplementary data are available at Molecular Plant Online.

613

FUNDING

615

This work was supported by European Research Council (ERC-2011-AdG294790), NWO-

616

NSFC Joined Research project (846.11.005) and NWO VICI (865.13.001).

617

RI PT

614

ACKNOWLEDGMENTS

619

We thank Bruno Müller for providing the TCS reporter, Florian Frugier for Mtcre1-1 seeds,

620

and Robin van Velzen for help with RNA-seq analysis.

SC

618

AC C

EP

TE D

M AN U

621

21

ACCEPTED MANUSCRIPT 622

Figure legends

623

Figure 1. The majority of transcriptional changes induced three hours post rhizobium LCO

625

treatment depend on MtCRE1.

626

Venn diagram showing the number of genes differentially expressed (FC>2, p<0.001) in the

627

root susceptible zone of wild-type M. truncatula and Mtcre1 mutant upon rhizobium LCO

628

treatment (10-9 M, 3 h). Gene expression was determined by RNA-seq of three biological

629

replicates per treatment/genotype combination. Yellow, orange and red indicate the number of

630

genes differentially expressed in wild type only, both wild type and Mtcre1 and Mtcre1 only,

631

respectively.

SC

RI PT

624

632

Figure 2. Expression of the synthetic cytokinin responsive reporter TCS:GUS in M.

634

truncatula transgenic roots upon application of rhizobium LCOs. Transgenic roots were

635

generated by Agrobacterium rhizogenes-mediated transformation.

636

(A) and (B) Whole mount images of mock treated (A) or rhizobium LCOs (10-9 M) treated (3

637

h) (B) transgenic roots stained for GUS activity.

638

(C) and (D) Sections through the susceptible zone of roots shown in (A) and (B), respectively.

639

Sections are counter stained with ruthenium red. Scale bars represent 50 µm.

640

TE D

M AN U

633

Figure 3. Rhizobium LCO signalling induces expression of genes involved in cytokinin and

642

ethylene biosynthesis.

643

(A)Expressionof genes involved incytokinin biosynthesis in wild-type M. truncatula and

644

Mtcre1 mutant.

645

(B) Expression of genes involved in cytokinin turnover in wild-type M. truncatula and Mtcre1

646

mutant.

647

(C) Expression of ACC synthases (ACS) in wild-type M. truncatula and Mtcre1 mutant.

648

Expression was determined at 3 h post rhizobium LCO treatment (10-9 M) (or mock) using

649

RNA-seq and normalized to that in the mock treated wild type. Expression is shown for genes

650

that were found to be differentially expressed (FC>2, p<0.001) in wild type post LCO

651

treatment. Data shown represent means of 3 biological replicates ± SEM. Different letters

652

above bars indicate significant difference (p<0.001).

AC C

EP

641

653 654

Figure 4. Time course experiment of LCO responsive genes.

655

(A) Expression of MtNIN and cytokinin response regulators MtRR4 and MtRR9. 22

ACCEPTED MANUSCRIPT (B) Expression of cytokinin biosynthesis genes MtIPT1, MtCYP735A1 and MtLOG3.

657

(C) Expression of cytokinin oxidase/dehydrogenases MtCKX2 and MtCKX3.

658

(D) Expression of ethylene biosynthesis genes MtACS1-3.

659

Expression was determined at 30 min, 1 h and 3 h post rhizobium LCO treatment (10-9 M) (or

660

mock) using qRT-PCR and normalized to the 3 h mock sample. ne indicates genes not

661

expressed under the conditions specified. Data shown represent means of 3 biological

662

replicates ± SEM. Different letters above bars indicate significant difference (p<0.01).

RI PT

656

663

Figure 5. Expression of LCO responsive genes in absence or presence of the ethylene

665

synthesis inhibitor AVG.

666

(A) Expression of MtNIN and cytokinin response regulators MtRR4 and MtRR9.

667

(B) Expression of cytokinin biosynthesis genes MtIPT1, MtCYP735A1 and MtLOG3.

668

(C) Expression of ethylene biosynthesis genes MtACS1-3.

669

M. truncatulawild-type and Mtcre1mutant plants were grown on agar-solidified Fahraeus

670

medium with or without 1 µM AVG. Expression was determined at 3 h post rhizobium LCO

671

treatment (10-9 M) (or mock) using qRT-PCR and normalized to that in the mock treated wild

672

type grown in absence of AVG. Data shown represent means of 3 biological replicates ±

673

SEM. Different letters above bars indicate significant difference (p<0.01).

674

TE D

M AN U

SC

664

675

Figure 6. Cytokinins accumulate in the M. truncatularoot susceptible zone upon S. meliloti

676

LCO signalling.

677

(A-C)

678

isopentenyladenineriboside (iPR) (B) and trans-zeatin (tZ) (C) in root susceptible zones of the

679

M. truncatulawild type and Mtccamk/Mtdmi3 mutant treated with mock or rhizobium LCOs

680

(10-9 M, 3 h). Data represent means of 4 biological replicates (2 for Mtccamk/Mtdmi3 mock

681

trans-zeatin) ± SEM.

682

(D-F)

683

isopentenyladenineriboside (iPR) (E) and trans-zeatin (tZ) (F) in root susceptible zones of the

684

M. truncatula wild type and Mtnsp1, Mtnsp2 and Mtnsp1Mtnsp2 mutants treated with mock

685

or rhizobium LCOs (10-9 M, 3 h). Plants were grown on agar-solidified Fahraeus medium

686

containing 1 µM AVG. Data represent means of 5-6 biological replicates ± SEM.

687

Different letters above bars indicate statistical difference (p<0.05). Statistical significance was

688

assessed based on ANOVA.

of

the

cytokininsisopentenyladenine

(iP)

(A),

AC C

EP

Concentrations

Concentrations

of

the

cytokininsisopentenyladenine

(iP)

(D),

689

23

ACCEPTED MANUSCRIPT 690

Figure 7. Ethylene negatively regulates LCO-induced cytokinin accumulation.

691

(A-C)

692

isopentenyladenineriboside (iPR) (B) and trans-zeatin (tZ) (C) in root susceptible zones of the

693

M. truncatulawild type and Mtein2/Mtsickle mutant treated with mock or rhizobium LCOs

694

(10-9 M, 3 h). Plants were grown on agar-solidified Fahraeus medium without AVG. Data

695

represent means of 6 biological replicates ± SEM. Different letters above bars indicate

696

statistical difference (p<0.05). Statistical significance was assessed based on ANOVA.

of

the

cytokininsisopentenyladenine

(iP)

(A),

RI PT

Concentrations

697

Figure 8. Proposed model for the position of cytokinin and ethylene in the rhizobium LCO

699

signalling pathway.

AC C

EP

TE D

M AN U

SC

698

24

ACCEPTED MANUSCRIPT REFERENCES

AC C

EP

TE D

M AN U

SC

RI PT

Alonso, J.M., Hirayama, T., Roman, G., Nourizadeh, S., and Ecker, J.R. (1999). EIN2, a bifunctional transducer of ethylene and stress responses in Arabidopsis. Science 284:2148-2152. Ariel, F., Brault-Hernandez, M., Laffont, C., Huault, E., Brault, M., Plet, J., Moison, M., Blanchet, S., Ichante, J.L., Chabaud, M., et al. (2012). Two direct targets of cytokinin signaling regulate symbiotic nodulation in Medicago truncatula. Plant Cell 24:3838-3852. Arrighi, J.F., Barre, A., Ben Amor, B., Bersoult, A., Soriano, L.C., Mirabella, R., de Carvalho-Niebel, F., Journet, E.P., Gherardi, M., Huguet, T., et al. (2006). The Medicago truncatula lysine motif-receptorlike kinase gene family includes NFP and new nodule-expressed genes. Plant Physiology 142:265-279. Benedito, V.A., Torres‐Jerez, I., Murray, J.D., Andriankaja, A., Allen, S., Kakar, K., Wandrey, M., Verdier, J., Zuber, H., and Ott, T. (2008). A gene expression atlas of the model legume Medicago truncatula. The Plant Journal 55:504-513. Berrabah, F., Bourcy, M., Cayrel, A., Eschstruth, A., Mondy, S., Ratet, P., and Gourion, B. (2014). Growth conditions determine the DNF2 requirement for symbiosis. PLoS ONE 9:e91866. Bianco, M.D., Giustini, L., and Sabatini, S. (2013). Spatiotemporal changes in the role of cytokinin during root development. New Phytologist 199:324-338. Bielach, a., Podlesakova, K., Marhavy, P., Duclercq, J., Cuesta, C., Muller, B., Grunewald, W., Tarkowski, P., and Benkova, E. (2012). Spatiotemporal regulation of lateral root organogenesis in Arabidopsis by cytokinin. Plant Cell 24:3967-3981. Breakspear, A., Liu, C., Roy, S., Stacey, N., Rogers, C., Trick, M., Morieri, G., Mysore, K.S., Wen, J., and Oldroyd, G.E. (2014). The Root Hair “Infectome” of Medicago truncatula Uncovers Changes in Cell Cycle Genes and Reveals a Requirement for Auxin Signaling in Rhizobial Infection. The Plant Cell Online:tpc. 114.133496. 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 12:1647-1666. Chae, H.S., Faure, F., and Kieber, J.J. (2003). The eto1, eto2, and eto3 mutations and cytokinin treatment increase ethylene biosynthesis in Arabidopsis by increasing the stability of ACS protein. Plant Cell 15:545-559. Chen, Y., Chen, W., Li, X., Jiang, H., Wu, P., Xia, K., Yang, Y., and Wu, G. (2014). Knockdown of LjIPT3 influences nodule development in Lotus japonicus. Plant & Cell Physiology 55:183-193. Cooper, J., and Long, Sr. (1994). Morphogenetic rescue of Rhizobium meliloti nodulation mutants by transzeatin secretion. Plant Cell 6:215-225. Crespi, M., and Frugier, F. (2008). De novo organ formation from differentiated cells: root nodule organogenesis. Science Signaling 1:re11. Czaja, L.F., Hogekamp, C., Lamm, P., Maillet, F., Martinez, E.A., Samain, E., Dénarié, J., Küster, H., and Hohnjec, N. (2012). Transcriptional responses toward diffusible signals from symbiotic microbes reveal MtNFP-and MtDMI3-dependent reprogramming of host gene expression by arbuscular mycorrhizal fungal lipochitooligosaccharides. Plant physiology 159:1671-1685. de León, B.G.-P., Zorrilla, J.M.F., Rubio, V., Dahiya, P., Paz-Ares, J., and Leyva, A. (2004). Interallelic complementation at the Arabidopsis CRE1 locus uncovers independent pathways for the proliferation of vascular initials and canonical cytokinin signalling. Plant Journal 38:70-79. Deinum, E.E., Geurts, R., Bisseling, T., and Mulder, B.M. (2012). Modeling a cortical auxin maximum for nodulation: Different signatures of potential strategies. Frontiers in Plant Science 3:96. Dello Ioio, R., Nakamura, K., Moubayidin, L., Perilli, S., Taniguchi, M., Morita, M.T., Aoyama, T., Costantino, P., and Sabatini, S. (2008). A genetic framework for the control of cell division and differentiation in the root meristem. Science 322:1380-1384. Domonkos, A., Horvath, B., Marsh, J.F., Halasz, G., Ayaydin, F., Oldroyd, G.E.D., and Kalo, P. (2013). The identification of novel loci required for appropriate nodule development in Medicago truncatula. BMC Plant Biology 13:157. Fåhraeus, G. (1957). The infection of clover root hairs by nodule bacteria studied by a simple glass slide technique. Journal of General Microbiology 16:374-381. Franco-Zorrilla, J.M., Martin, A.C., Solano, R., Rubio, V., Leyva, A., and Paz-Ares, J. (2002). Mutations at CRE1 impair cytokinin-induced repression of phosphate starvation responses in Arabidopsis. Plant Journal 32:353-360. Fuchs, Y., and Lieberman, M. (1968). Effects of kinetin, IAA, and gibberellin on ethylene production, and their interactions in growth of seedlings. Plant physiology 43:2029-2036.

25

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Gonzalez-Rizzo, S., Crespi, M., and 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. Gourion, B., Bourcy, M., Cosson, V., and Ratet, P. (2013). Protocols for growing plant symbioses; rhizobia. In: Plant Mineral Nutrients: Methods and Protocols, Methods in Molecular Biology--Maathuis, F.J.M., ed. 61-75. Haney, C.H., Riely, B.K., Tricoli, D.M., Cook, D.R., Ehrhardt, D.W., and Long, S.R. (2011). Symbiotic rhizobia bacteria trigger a change in localization and dynamics of the Medicago truncatula receptor kinase LYK3. Plant Cell 23:2774-2787. Heckmann, A.B., Sandal, N., Bek, A.S., Madsen, L.H., Jurkiewicz, A., Nielsen, M.W., Tirichine, L., and Stougaard, J. (2011). Cytokinin induction of root nodule primordia in Lotus japonicus is regulated by a mechanism operating in the root cortex. Molecular Plant-Microbe Interactions 24:1385-1395. Heidstra, R., Yang, W.C., Yalcin, Y., Peck, S., Emons, A.M., Kammen van, A., and Bisseling, T. (1997). Ethylene provides positional information on cortical cell division but is not involved in Nod factorinduced root hair tip growth in rhizobium-legume interaction. Development 124:1781-1787. Held, M., Hou, H., Miri, M., Huynh, C., Ross, L., Hossain, M.S., Sato, S., Tabata, S., Perry, J., Wang, T.L., et al. (2014). Lotus japonicus cytokinin receptors work partially redundantly to mediate nodule formation. Plant Cell 26:678-694. Higuchi, M., Pischke, M.S., Mähönen, A.P., Miyawaki, K., Hashimoto, Y., Seki, M., Kobayashi, M., Shinozaki, K., Kato, T., Tabata, S., et al. (2004). In planta functions of the Arabidopsis cytokinin receptor family. Proceedings of the National Academy of Sciences of the United States of America 101:8821-8826. Hirsch, S., Kim, J., Muñoz, A., Heckmann, A.B., Downie, J.A., and Oldroyd, G.E. (2009). GRAS proteins form a DNA binding complex to induce gene expression during nodulation signaling in Medicago truncatula. Plant Cell 21:545-557. Horváth, B., Yeun, L.H., Domonkos, A., Halász, G., Gobbato, E., Ayaydin, F., Miró, K., Hirsch, S., Sun, J., Tadege, M., et al. (2011). Medicago truncatulaIPD3 is a member of the common symbiotic signaling pathway required for rhizobial and mycorrhizal symbioses. Molecular Plant-Microbe Interactions 24:1345-1358. Huo, X., Schnabel, E., Hughes, K., and Frugoli, J. (2006). RNAi phenotypes and the localization of a protein::GUS fusion imply a role for Medicago truncatulaPIN genes in nodulation. Journal of Plant Growth Regulation 25:156-165. 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 truncatulaENOD11: a novel RPRPencoding early nodulin gene expressed during mycorrhization in arbuscule-containing cells. Molecular Plant-Microbe Interactions 14:737-748. Kalo, P., Gleason, C., Edwards, A., Marsh, J., Mitra, R.M., Hirsch, S., Jakab, J., Sims, S., Long, S.R., Rogers, J., et al. (2005). Nodulation signaling in legumes requires NSP2, a member of the GRAS family of transcriptional regulators. Science 308:1786-1789. Kamada-Nobusada, T., and Sakakibara, H. (2009). Molecular basis for cytokinin biosynthesis. Phytochemistry 70:444-449. Karimi, M., Inzé, D., and Depicker, A. (2002). GATEWAY™ vectors for Agrobacterium-mediated plant transformation. Trends Plant Sci 7:193-195. Kohlen, W., Charnikhova, T., Liu, Q., Bours, R., Domagalska, M.a., Beguerie, S., Verstappen, F., Leyser, O., Bouwmeester, H., and Ruyter-Spira, C. (2011). Strigolactones are transported through the xylem and play a key role in shoot architectural response to phosphate deficiency in nonarbuscular mycorrhizal host Arabidopsis. Plant Physiology 155:974-987. Kohlen, W., Charnikhova, T., Lammers, M., Pollina, T., Tóth, P., Haider, I., Pozo, M.J., Maagd, R.A., Ruyter‐Spira, C., and Bouwmeester, H.J. (2012). The tomato CAROTENOID CLEAVAGE DIOXYGENASE8 (SlCCD8) regulates rhizosphere signaling, plant architecture and affects reproductive development through strigolactone biosynthesis. New Phytologist 196:535-547. Kouchi, H., Imaizumi-anraku, H., Hayashi, M., Hakoyama, T., Nakagawa, T., Umehara, Y., Suganuma, N., and Kawaguchi, M. (2010). How many peas in a pod? Legume genes responsible for mutualistic symbioses underground. Plant & Cell Physiology 51:1381-1397. Kurakawa, T., Ueda, N., Maekawa, M., Kobayashi, K., Kojima, M., Nagato, Y., Sakakibara, H., and Kyozuka, J. (2007). Direct control of shoot meristem activity by a cytokinin-activating enzyme. Nature 445:652655. Kuroha, T., Tokunaga, H., Kojima, M., Ueda, N., Ishida, T., Nagawa, S., Fukuda, H., Sugimoto, K., and Sakakibara, H. (2009). Functional analyses of LONELY GUY cytokinin-activating enzymes reveal the importance of the direct activation pathway in Arabidopsis. Plant Cell 21:3152-3169.

26

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Lerouge, P., Roche, P., Faucher, C., Maillet, F., Truchet, G., Prome, J.C., and Denarie, J. (1990). Symbiotic hostspecificity of Rhizobium meliloti is determined by a sulphated and acylated glucosamine oligosaccharide signal. Nature 344:781-784. Levy, J., Bres, C., Geurts, R., Chalhoub, B., Kulikova, O., Duc, G., Journet, E.P., Ane, J.M., Lauber, E., Bisseling, T., 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. Limpens, E., Ramos, J., Franken, C., Raz, V., Compaan, B., Franssen, H., Bisseling, T., and Geurts, R. (2004). RNA interference in Agrobacterium rhizogenes-transformed roots of Arabidopsis and Medicago truncatula. Journal of Experimental Botany 55:983-992. Limpens, E., Mirabella, R., Fedorova, E., Franken, C., Franssen, H., Bisseling, T., and Geurts, R. (2005). Formation of organelle-like N2-fixing symbiosomes in legume root nodules is controlled by DMI2. Proceedings of the National Academy of Sciences of the United States of America 102:10375-10380. Lin, Z., Zhong, S., and Grierson, D. (2009). Recent advances in ethylene research. Journal of Experimental Botany 60:3311-3336. Liu, W., Kohlen, W., Lillo, A., Op den Camp, R., Ivanov, S., Hartog, M., Limpens, E., Jamil, M., Smaczniak, C., Kaufmann, K., et al. (2011). Strigolactone biosynthesis in Medicago truncatula and rice requires the symbiotic GRAS-type transcription factors NSP1 and NSP2. Plant Cell 23:3853-3865. Lohar, D.P., Schaff, J.E., Laskey, J.G., Kieber, J.J., Bilyeu, K.D., and Bird, D.M. (2004). Cytokinins play opposite roles in lateral root formation, and nematode and Rhizobial symbioses. Plant Journal 38:203214. Lomin, S.N., Krivosheev, D.M., Steklov, M.Y., Arkhipov, D.V., Osolodkin, D.I., Schmülling, T., and Romanov, G.A. (2015). Plant membrane assays with cytokinin receptors underpin the unique role of free cytokinin bases as biologically active ligands. Journal of Experimental Botany:eru522. Lorteau, M.A., Ferguson, B.J., and Guinel, F.C. (2001). Effects of cytokinin on ethylene production and nodulation in pea (Pisum sativum) cv. Sparkle. Physiologia Plantarum 112:421-428. Madsen, L.H., Tirichine, L., Jurkiewicz, A., Sullivan, J.T., Heckmann, A.B., Bek, A.S., Ronson, C.W., James, E.K., and Stougaard, J. (2010). The molecular network governing nodule organogenesis and infection in the model legume Lotus japonicus. Nature Communications 1:10. Marhavy, P., Bielach, A., Abas, L., Abuzeineh, A., Duclercq, J., Tanaka, H., Parezova, M., Petrasek, J., Friml, J., Kleine-Vehn, J., et al. (2011). Cytokinin modulates endocytic trafficking of PIN1 auxin efflux carrier to control plant organogenesis. Developmental Cell 21:796-804. Marhavy, P., Duclercq, J., Weller, B., Feraru, E., Bielach, A., Offringa, R., Friml, J., Schwechheimer, C., Murphy, A., and Benková, E. (2014). Cytokinin controls polarity of PIN1-dependent auxin transport during lateral root organogenesis. Current Biology 24:1031-1037. Mathesius, U., Schlaman, H.R., Spaink, H.P., Of Sautter, C., Rolfe, B.G., and Djordjevic, M.A. (1998). Auxin transport inhibition precedes root nodule formation in white clover roots and is regulated by flavonoids and derivatives of chitin oligosaccharides. Plant Journal 14:23-34. Mathesius, U., Charon, C., Rolfe, B.G., Kondorosi, A., and Crespi, M. (2000). Temporal and spatial order of events during the induction of cortical cell divisions in white clover by Rhizobium leguminosarum bv. trifolii inoculation or localized cytokinin addition. Molecular Plant-Microbe Interactions 13:617-628. Merchante, C., Alonso, J.M., and Stepanova, A.N. (2013). Ethylene signaling: Simple ligand, complex regulation. Current Opinion in Plant Biology 16:554-560. Mergaert, P., Van Montagu, M., Promé, J.-C., and Holsters, M. (1993). Three unusual modifications, a Darabinosyl, an N-methyl, and a carbamoyl group, are present on the Nod factors of Azorhizobium caulinodans strain ORS571. Proceedings of the National Academy of Sciences 90:1551-1555. Mitra, R.M., Gleason, C.A., Edwards, A., Hadfield, J., Downie, J.A., Oldroyd, G.E., and Long, S.R. (2004). A Ca2+/calmodulin-dependent protein kinase required for symbiotic nodule development: Gene identification by transcript-based cloning. Proceedings of the National Academy of Sciences of the United States of America 101:4701-4705. Miyata, K., Kawaguchi, M., and Nakagawa, T. (2013). Two distinct EIN2 genes cooperatively regulate ethylene signaling in Lotus japonicus. Plant & Cell Physiology 54:1469-1477. Mortier, V., Den Herder, G., Whitford, R., Van de Velde, W., Rombauts, S., D'Haeseleer, K., Holsters, M., and Goormachtig, S. (2010). CLE peptides control Medicago truncatula nodulation locally and systemically. Plant Physiology 153:222-237. Mortier, V., Wasson, A., Jaworek, P., De Keyser, A., Decroos, M., Holsters, M., Tarkowski, P., Mathesius, U., and Goormachtig, S. (2014). Role of LONELY GUY genes in indeterminate nodulation on Medicago truncatula. New Phytologist 202:582-593.

27

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Muller, B., and Sheen, J. (2008). Cytokinin and auxin interaction in root stem-cell specification during early embryogenesis. Nature 453:1094-1098. Murray, J.D., Karas, B.J., Sato, S., Tabata, S., Amyot, L., and Szczyglowski, K. (2007). A cytokinin perception mutant colonized by Rhizobium in the absence of nodule organogenesis. Science 315:101-104. Murray, J.D. (2011). Invasion by invitation: Rhizobial infection in legumes. molecular Plant-Microbe Interactions 24:631-639. Nakagawa, T., Kaku, H., Shimoda, Y., Sugiyama, A., Shimamura, M., Takanashi, K., Yazaki, K., Aoki, T., Shibuya, N., and Kouchi, H. (2011). From defense to symbiosis: limited alterations in the kinase domain of LysM receptor-like kinases are crucial for evolution of legume-Rhizobium symbiosis. Plant Journal 65:169-180. Oldroyd, G.E., Engstrom, E.M., and Long, S.R. (2001a). Ethylene inhibits the Nod factor signal transduction pathway of Medicago truncatula. Plant Cell 13:1835-1849. Oldroyd, G.E., and Long, S.R. (2003). Identification and characterization of Nodulation-Signaling Pathway 2, a gene of Medicago truncatula involved in Nod factor signaling. Plant Physiology 131:1027-1032. Oldroyd, G.E.D., Mitra, R.M., Wais, R.J., and Long, S.R. (2001b). Evidence for structurally specific negative feedback in the Nod factor signal transduction pathway. Plant Journal 28:191-199. Oldroyd, G.E.D. (2013). Speak, friend, and enter: signalling systems that promote beneficial symbiotic associations in plants. Nature Reviews. Microbiology 11:252-263. Op den Camp, R.H., De Mita, S., Lillo, A., Cao, Q., Limpens, E., Bisseling, T., and Geurts, R. (2011). A phylogenetic strategy based on a legume-specific whole genome duplication yields symbiotic cytokinin type-A response regulators. Plant Physiology 157:2013-2022. Ovchinnikova, E., Journet, E.P., Chabaud, M., Cosson, V., Ratet, P., Duc, G., Fedorova, E., Liu, W., den Camp, R.O., Zhukov, V., et al. (2011). IPD3 controls the formation of nitrogen-fixing symbiosomes in pea and Medicago spp. Molecular Plant-Microbe Interactions 24:1333-1344. Pacios-Bras, C., Schlaman, H.R., Boot, K., Admiraal, P., Langerak, J.M., Stougaard, J., and Spaink, H.P. (2003). Auxin distribution in Lotus japonicus during root nodule development. Plant Molecular Biology 52:1169-1180. Penmetsa, R.V., and Cook, D.R. (1997). A legume ethylene-insensitive mutant hyperinfected by its rhizobial symbiont. Science 275:527-530. Penmetsa, R.V., Frugoli, J.A., Smith, L.S., Long, S.R., and Cook, D.R. (2003). Dual genetic pathways controlling nodule number in Medicago truncatula. Plant Physiology 131:998-1008. Penmetsa, R.V., Uribe, P., Anderson, J., Lichtenzveig, J., Gish, J.C., Nam, Y.W., Engstrom, E., Xu, K., Sckisel, G., Pereira, M., et al. (2008). The Medicago truncatula ortholog of Arabidopsis EIN2, sickle, is a negative regulator of symbiotic and pathogenic microbial associations. Plant Journal 55:580-595. Plet, J., Wasson, A., Ariel, F., Le Signor, C., Baker, D., Mathesius, U., Crespi, M., and Frugier, F. (2011). MtCRE1-dependent cytokinin signaling integrates bacterial and plant cues to coordinate symbiotic nodule organogenesis in Medicago truncatula. Plant Journal 65:622-633. Radutoiu, S., Madsen, L.H., Madsen, E.B., Felle, H.H., Umehara, Y., Gronlund, M., Sato, S., Nakamura, Y., Tabata, S., Sandal, N., et al. (2003). Plant recognition of symbiotic bacteria requires two LysM receptor-like kinases. Nature 425:585-592. Radutoiu, S., Madsen, L.H., Madsen, E.B., Jurkiewicz, A., Fukai, E., Quistgaard, E.M., Albrektsen, A.S., James, E.K., Thirup, S., and Stougaard, J. (2007). LysM domains mediate lipochitin-oligosaccharide recognition and Nfr genes extend the symbiotic host range. EMBO Journal 26:3923-3935. Relić, B., Perret, X., Estrada‐García, M.T., Kopcinska, J., Golinowski, W., Krishnan, H., Pueppke, S., and Broughton, W. (1994). Nod factors of Rhizobium are a key to the legume door. Molecular microbiology 13:171-179. Robinson, M.D., McCarthy, D.J., and Smyth, G.K. (2010). edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26:139-140. Roman, G., Lubarsky, B., Kieber, J.J., Rothenberg, M., and Ecker, J.R. (1995). Genetic analysis of ethylene signal transduction in Arabidopsis thaliana: five novel mutant loci integrated into a stress response pathway. Genetics 139:1393-1409. Sakakibara, H. (2006). Cytokinins: Activity, biosynthesis, and translocation. Annual Review of Plant Biology 57:431-449. Sasaki, T., Suzaki, T., Soyano, T., Kojima, M., Sakakibara, H., and Kawaguchi, M. (2014). Shoot-derived cytokinins systemically regulate root nodulation. Nature communications 5:4983. Schmülling, T., Werner, T., Riefler, M., Krupková, E., and Bartrina Y Manns, I. (2003). Structure and function of cytokinin oxidase/dehydrogenase genes of maize, rice, Arabidopsis and other species. Journal of Plant Research 116:241-252.

28

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Serna-Sanz, A., Parniske, M., and Peck, S.C. (2011). Phosphoproteome analysis of Lotus japonicus roots reveals shared and distinct components of symbiosis and defense. Molecular Plant-Microbe Interactions 24:932-937. Singh, S., Katzer, K., Lambert, J., Cerri, M., and Parniske, M. (2014). CYCLOPS, a DNA-binding transcriptional activator, orchestrates symbiotic root nodule development. Cell Host & Microbe 15:139152. Smit, P., Raedts, J., Portyanko, V., Debelle, F., Gough, C., Bisseling, T., and Geurts, R. (2005). NSP1 of the GRAS protein family is essential for rhizobial Nod factor-induced transcription. Science 308:17891791. Stokkermans, T.J., and Peters, N.K. (1994). Bradyrhizobium elkanii lipo-oligosaccharide signals induce complete nodule structures on Glycine soja Siebold et Zucc. Planta 193:413-420. Stolz, A., Riefler, M., Lomin, S.N., Achazi, K., Romanov, G.A., and Schmülling, T. (2011). The specificity of cytokinin signalling in Arabidopsis thaliana is mediated by differing ligand affinities and expression profiles of the receptors. The Plant Journal 67:157-168. Suzaki, T., Ito, M., and Kawaguchi, M. (2013). Genetic basis of cytokinin and auxin functions during root nodule development. Frontiers in Plant Science 4:42. Takahara, M., Magori, S., Soyano, T., Okamoto, S., Yoshida, C., Yano, K., Sato, S., Tabata, S., Yamaguchi, K., and Shigenobu, S. (2013). TOO MUCH LOVE, a novel Kelch repeat-containing F-box protein, functions in the long-distance regulation of the legume-Rhizobium symbiosis. Plant and cell physiology:pct022. Takanashi, K., Sugiyama, A., and Yazaki, K. (2011). Involvement of auxin distribution in root nodule development of Lotus japonicus. Planta 234:73-81. Takei, K., Yamaya, T., and Sakakibara, H. (2004). Arabidopsis CYP735A1 and CYP735A2 encode cytokinin hydroxylases that catalyze the biosynthesis of trans-zeatin. Journal of Biological Chemistry 279:4186641872. Tirichine, L., Sandal, N., Madsen, L.H., Radutoiu, S., Albrektsen, A.S., Sato, S., Asamizu, E., Tabata, S., and Stougaard, J. (2007). A gain-of-function mutation in a cytokinin receptor triggers spontaneous root nodule organogenesis. Science 315:104-107. Torrey, J.G. (1961). Kinetin as trigger for mitosis in mature endomitotic plant cells. Experimental Cell Research 23:281-299. Truchet, G., Roche, P., Lerouge, P., Vasse, J., Camut, S., de Billy, F., Promé, J.-C., and Dénarié, J. (1991). Sulphated lipo-oligosaccharide signals of Rhizobium meliloti elicit root nodule organogenesis in alfalfa. Ulmasov, T., Murfett, J., Hagen, G., and Guilfoyle, T.J. (1997). Aux/IAA proteins repress expression of reporter genes containing natural and highly active synthetic auxin response elements. Plant Cell 9:1963-1971. van Noorden, G.E., Kerim, T., Goffard, N., Wiblin, R., Pellerone, F.I., Rolfe, B.G., and Mathesius, U. (2007). Overlap of proteome changes in Medicago truncatula in response to auxin and Sinorhizobium meliloti. Plant Physiology 144:1115-1131. Vernie, T., Moreau, S., de Billy, F., Plet, J., Combier, J.P., Rogers, C., Oldroyd, G., Frugier, F., Niebel, A., and 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. Xiao, T.T., Schilderink, S., Moling, S., Deinum, E.E., Kondorosi, E., and Franssen, H. (2014). Fate map of Medicago truncatula root nodules. Development 141:3517-3528. Yokota, K., and Hayashi, M. (2011). Function and evolution of nodulation genes in legumes. Cellular and Molecular Life Sciences 68:1341-1351. Zhang, X., Chen, Y., Lin, X., Hong, X., Zhu, Y., Li, W., He, W., An, F., and Guo, H. (2013). Adenine phosphoribosyl transferase 1 is a key enzyme catalyzing cytokinin conversion from nucleobases to nucleotides in Arabidopsis. Molecular plant 6:1661-1672.

29

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT