Вы находитесь на странице: 1из 48

Plant Physiology Preview. Published on December 5, 2018, as DOI:10.1104/pp.18.

01408

1 Running head:

2 Phosphatidylcholine biosynthesis in Arabidopsis

4 Research Area:

5 Biochemistry and Metabolism

7 Corresponding Author:

9 Yuki Nakamura

10 Institute of Plant and Microbial Biology, Academia Sinica,

11 128 sec.2 Academia Rd., Nankang, Tapei 11529, Taiwan R.O.C.

12 +886 2 2787 1129

13 nakamura@gate.sinica.edu.tw.

1
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
Copyright 2018 by the American Society of Plant Biologists
14

15

16 A methyltransferase trio essential for phosphatidylcholine biosynthesis and growth

17

18 Yu-chi Liu, Ying-Chen Lin, Kazue Kanehara, and Yuki Nakamura*

19

20 Institute of Plant and Microbial Biology, Academia Sinica, Taipei 11529, Taiwan (Y-c.L.,

21 Y-C.L., K.K., Y.N.); Molecular and Biological Agricultural Sciences Program,

22 Academia Sinica, Taiwan International Graduate Program, Taipei 11529, Taiwan

23 (Y-C.L., K.K., Y.N.); 3Graduate Institute of Biotechnology, National Chung-Hsing

24 University, Taichung 402, Taiwan (Y-C.L.); Biotechnology Center, National

25 Chung-Hsing University, Taichung 402, Taiwan (K.K., Y.N.)

26

27 One-sentence summary:

28 Three phospho-base N-methyltransferases have tissue-specific functions in

29 phosphatidylcholine biosynthesis, which is essential for plant growth.

2
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
30 Footnotes:
1
31 Y-c.L. and Y-C.L. performed experiments and analyzed data; K.K. provided technical

32 support; Y.N. conceived research plan, performed transgenic plant management, genetic

33 crossing, supervised experiments, and wrote the article. All authors commented on the

34 article and approved the contents.


2
35 The research was supported by the core budget of the Institute of Plant and Microbial

36 Biology, Academia Sinica (Y.N. and K.K.), and the Ministry of Science and Technology,

37 Taiwan (No. 106-2628-B-001-002 to Y.N).

38 *E-mail address of Author for Contact: nakamura@gate.sinica.edu.tw. (YN)

39

3
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
40 ABSTRACT

41 Phosphatidylcholine (PC) is a primary class of membrane lipids in most eukaryotes. In

42 plants, the primary PC biosynthetic pathway and its role in plant growth and

43 development remain elusive due to lack of a mutant model with substantially decreased

44 PC content. Recently, a double mutant of Arabidopsis (Arabidopsis thaliana)

45 PHOSPHO-BASE N-METHYLTRANSFERASE 1 (PMT1) and PMT3 was reported

46 with reduced PC content and defective plant growth. However, residual PC content as

47 well as the non-lethal phenotype of the mutant suggests an additional enzyme

48 contributes to PC biosynthesis. Here, we report on the role of three PMTs in PC

49 biosynthesis and plant development, with a focus on PMT2. PMT2 had the highest

50 expression level among the three PMTs, and was highly expressed in roots. The pmt1

51 pmt2 double mutant enhanced the defects in root growth, cell viability, and PC content

52 of pmt1, suggesting that PMT2 functions together with PMT1 in roots. Chemical

53 inhibition of PMT activity in wild-type roots reproduced the short root phenotype

54 observed in pmt1 pmt2, suggesting that PMT1 and PMT2 are the major PMT isoforms

55 in roots. In shoots, pmt1 pmt2 pmt3 enhanced the phenotype of pmt1 pmt3, showing

56 seedling lethality and further reduced PC content without detectable de novo PC

57 biosynthesis. These results suggest that PMTs catalyze an essential reaction step in PC

58 biosynthesis and that the three PMTs have differential tissue-specific functions in PC

59 biosynthesis and plant growth.

60

4
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
61 INTRODUCTION

62 Phosphatidylcholine (PC) is an important class of phospholipid in eukaryotes. In

63 Arabidopsis (Arabidopsis thaliana), PC represents an abundant lipid class in

64 extraplastidic membranes (Devaiah et al., 2006). PC is synthesized by incorporating the

65 phosphocholine (PCho) polar head group into the sn-1,2-diacylglycerol (DAG)

66 backbone, which is catalyzed by aminoalcohol aminophosphotransferase (AAPT) (Liu

67 et al., 2015). DAG is a common precursor for the biosynthesis of different glycerolipid

68 classes, so its biosynthetic pathway has been studied extensively. On the other hand, the

69 biosynthesis of polar head groups, such as PCho, has been less well investigated. It is

70 known that biosynthesis of PCho involves three sequential methylations of

71 phosphoethanolamine (PEtn) catalyzed by phospho-base N-methyltransferase (PMT)

72 (Bolognese and McGraw, 2000). The product, PCho, is converted to cytidine

73 diphosphocholine (CDP-Cho) by CTP:phosphorylcholine cytidylyltransferase (CCT)

74 (Inatsugi et al., 2002, 2009), then is incorporated into the DAG backbone by AAPT to

75 produce PC (Liu et al., 2015). Because PEtn is a common precursor for the biosynthesis

76 of PC and phosphatidylethanolamine (PE) (Mizoi et al., 2006), PMT catalyzes the first

77 reaction step of the PC biosynthetic pathway and thus may have a critical role in PC

78 biosynthesis.

79 Methylation of the ethanolamine group occurs either at the phospho base or

80 phosphatidyl base. For example, in Arabidopsis and other seed plants, the first

81 methylation in the synthesis of choline moieties occurs exclusively at the level of PEtn

82 (Hanson and Rhodes., 1983; Datko and Mudd., 1988; Summers and Weretilnyk., 1993;

83 Nuccio et al., 2000), while the second and third methylation occur either at the phospho

84 base or phosphatidyl base. In contrast, the methyltransferases in Saccharomyces

5
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
85 cerevisiae take only phosphatidyl base as a substrate, so they convert PE to PC but not

86 PEtn to PCho (Kodaki and Yamashita., 1989). Arabidopsis has three PMTs and one

87 phosphatidyl-base N-methyltransferase (PLMT). PLMT1 is involved in the second and

88 third but not the first methylation of PE (Keogh et al., 2009). Knocking out PLMT1

89 does not alter PC level significantly (Keogh et al., 2009), suggesting that the

90 PLMT1-mediated pathway is not the primary PC biosynthesis pathway in Arabidopsis.

91 The three PMTs, PMT1 (or NMT1, PEAMT1, XPL1; At3g18000), PMT2 (or PMEAMT;

92 At1g48600), and PMT3 (or NMT3, At1g73600), are shown to catalyze all of the three

93 methylation steps in vitro (Lee and Jez., 2017). Knocking out of PMT1 causes root

94 growth defects and slight reduction in PC content (Cruz-Ramírez et al., 2004).

95 Disruption of PMT2 by T-DNA has no effect on PC level and plant growth (BeGora et

96 al., 2010). Gene knockout of PMT3 did not cause major defects in plant growth or PC

97 content (Lee and Jez., 2017; Chen et al., 2018); however, the pmt1 pmt3 double mutant

98 significantly affects PC content (Chen et al., 2018). Although this result indicates that

99 PMT1 and PMT3 are the major PMT isoforms in PC biosynthesis and plant growth, the

100 residual PC content and non-lethal phenotype of pmt1 pmt3 suggests the involvement of

101 another enzyme in PC biosynthesis.

102 To provide a comprehensive understanding of the function of the three PMT

103 isoforms in PC biosynthesis and plant growth, we herein focused on PMT2, a

104 less-characterized isoform with regard to its function in vivo. We showed that PMT2 has

105 a distinct expression profile with the highest expression in roots. Indeed, the pmt1 pmt2

106 double mutant had enhanced defective root growth, cell viability, and PC content

107 compared to the pmt1. Moreover, the pmt1 pmt2 pmt3 triple mutant enhanced the severe

108 growth defect of pmt1 pmt3. The triple mutant had no detectable de novo PC

6
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
109 biosynthesis activity, resulting in about 20% the PC content of wild type and seedling

110 lethality. These results suggest that the three PMTs play critical roles in PC biosynthesis,

111 providing the first evidence that PCs are an essential phospholipid class in Arabidopsis.

112

113

7
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
114 RESULTS

115

116 Expression of PMT2

117 To investigate the in vivo function of PMT2, we first referred to a public gene

118 expression database to analyze the tissue- and developmental stage-specific expression

119 patterns of the three PMTs using GENEVESTIGATOR (Hruz et al., 2008). As shown in

120 Fig. 1A, PMT2 showed a distinct tissue-specific expression pattern. Compared to PMT1

121 and PMT3, PMT2 showed higher expression in roots, including the primary and lateral

122 roots, elongation and maturation zones, and the stele. Next, we compared the

123 developmental stage-specific expression profiles (Fig. 1B). Among the three PMTs,

124 PMT2 showed the highest expression level in all developmental stages analyzed. The

125 expression level was lower in germinating seeds, mature siliques and during senescence

126 compared to the other stages. These data suggest that PMT2 is the major isoform of the

127 three PMTs and is consistently expressed in most developmental stages, with a

128 root-specific expression pattern.

129 To confirm the results of the database analysis, we produced a construct of the

130 genomic sequence of PMT2 including the 5’ promoter region and open reading frame

131 followed by a GUS reporter, and transduced it into the pmt1-1 pmt2-1 mutant

132 background (ProPMT2:PMT2-GUS pmt1-1 pmt2-1). In the T2 generation, we isolated

133 34 independent transgenic lines that harbor the transgene. We selected three

134 representative lines (Lines No. 7, 17, and 23) for histochemical GUS staining. During

135 vegetative growth of line No. 7, PMT2-GUS was highly expressed in leaf veins and in

136 both mature root tissue and root tips (Fig. 1C). In floral organs, PMT2-GUS expression

137 was observed clearly in young stamens (Fig. 1D, E), but faintly in mature flowers (Fig.

8
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
138 1F, G). This expression pattern was consistent with that of the other two lines. Moreover,

139 ProPMT2:PMT2-GUS pmt1-1 pmt2-1 complemented the short root phenotype of

9
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
140 pmt1-1 pmt2-1 (described later), suggesting that PMT2-GUS is functional in vivo (Fig.

141 S1). Thus, GUS staining supported the higher expression level of PMT2 in roots, and

142 showed leaf vasculature as an additional site where PMT2 is highly expressed.

143

144 Mutation of PMT2 enhanced the defective root growth of pmt1-1

145 Expression of PMT2 was highest in the roots (Fig. 1A). However, in agreement with the

146 previous report (BeGora et al., 2010), no root phenotype was found in the pmt2 mutants.

147 Because PMT1, but not PMT3, was also highly expressed in roots (Fig. 1A) and the

148 pmt1 mutant shows a short root phenotype (Cruz-Ramírez et al., 2004), we anticipated a

149 possible overlap in function between PMT1 and PMT2 in root growth. We therefore

150 created the pmt1-1 pmt2-1 double mutant to investigate whether the root phenotype in

151 pmt1-1 is enhanced by pmt2-1. As shown in Fig. 2A, 20-day-old pmt1-1 and pmt1-2

152 seedlings showed significantly reduced root length, whereas pmt2-1 and pmt2-2 did not.

153 Here, we found that pmt1-1 pmt2-1 had further reduced root length as compared to

154 pmt1-1 and pmt1-2. Indeed, quantitative measurement of the root length showed that the

155 primary root length of pmt1-1 pmt2-1 was about 25% that of the wild type and about

156 50% that of pmt1-1 (Fig. 2B). A closer look at the root morphology during the first 6

157 days after germination showed that pmt1-1 pmt1-2 showed reduced growth of the main

158 root as compared to pmt1-1 or WT (Fig. 2C to E). At 3 days after germination, the

159 mutants showed significantly retarded growth in the main root but generated precocious

160 lateral branches, whereas the wild-type roots still had no lateral branches. From 4 to 6

161 days after germination, the mutants produced more extensive lateral branches at the

162 expense of primary root elongation. The pmt1-1 pmt2-1 mutant showed swelling of the

163 root in addition to entirely reduced growth of lateral roots (Fig. 2D), whereas pmt1-1

10
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
164 showed relatively competent lateral root growth (Fig. 2C).

165 To obtain further evidence of the enhanced root growth defect in the pmt1-1

11
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
166 pmt2-1 mutant, we observed the cellular architecture of the branching part of the root

167 (Fig. 2F to H). In the wild type, epidermal cells were orderly shaped, for a flat root

168 surface (Fig. 2F). In pmt1-1, the surface of the main root was swollen, despite the

169 normal shape of the epidermal cells at the base of the lateral branch (Fig. 2G). However,

170 in pmt1-1 pmt2-1, the main root surface was swollen, and epidermal cells at the base of

171 the lateral branch bulged abnormally (Fig. 2H). Thus, in pmt1-1 pmt2-1, cell growth

172 may be impaired at the branch points of roots, causing abnormal lateral root

173 development.

174 To determine the viability of root cells in these mutants, we performed double

175 fluorescent staining with propidium iodide (PI) and fluorescein diacetate (FDA) as

176 previously used (Cruz-Ramírez et al., 2004). PI staining indicates dead cells and FDA

177 staining reveals viable cells (Celenza et al., 1995). PI stains cell walls independently of

178 cell viability (Kirik et al., 2001; Chaves et al., 2002). Wild-type root cells showed

179 extensive FDA staining but PI staining was limited to the external surface (Fig. 2I and J).

180 Root cells of pmt1-1 showed substantial PI staining but diminished FDA staining (Fig.

181 2K and L), which suggests that viable cells of pmt1-1 were limited to the tip of roots

182 (Fig. 2K). In pmt1-1 pmt2-1 roots, PI staining was further enhanced and FDA was

183 barely detectable, even at the root tips (Fig. 2M and N), which suggest that cell viability

184 was further impaired in pmt1-1 pmt2-1 as compared to pmt1-1. Taken together, we

185 concluded that mutation in PMT2 enhanced the defective root growth of pmt1-1.

186

187 Genetic and chemical complementation of the root phenotype in pmt1-1 and pmt1-1

188 pmt2-1

189 To determine whether the enhanced root growth defects in pmt1-1 pmt2-1 were due to

12
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
190 mutations in PMT1 and PMT2, we conducted genetic complementation of the

191 phenotype by introducing transacting genomic sequences of the respective genes

13
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
192 (ProPMT1:PMT1 and ProPMT2:PMT2) into pmt1-1 pmt2-1. We isolated 10

193 independent transgenic lines of each. In contrast to the root length of pmt1-1 pmt1-2,

194 root length of ProPMT1:PMT1 pmt1-1 pmt2-1 was indistinguishable from that of the

195 wild type, and that of ProPMT2:PMT2 pmt1-1 pmt2-1 was similar to pmt1-1 (Fig. 3A

196 and B). We confirmed this result by measuring the root length of all transgenic lines we

197 isolated (Fig. S2). Thus, the root growth defect in pmt1-1 pmt2-1 was caused by

198 mutations in PMT1 and PMT2.

199 To examine whether the short root phenotype in pmt1-1 pmt2-1 is due to

200 defective PCho biosynthesis, we conducted a chemical complementation experiment.

201 The short root phenotype of pmt1 is rescued by exogenous supplementation of Cho to

202 the MS-agar plate, which is readily taken up and is phosphorylated by endogenous Cho

203 kinase activity to produce PCho (Cruz-Ramírez et al., 2004). As shown in Fig. 3C,

204 exogenous supplementation of 100 μM Cho to the MS medium rescued the short root

205 phenotype of pmt1-1 pmt2-1 and pmt1-1 (Fig. 3D), whereas Etn supplementation did

206 not complement the phenotype (Fig. 3C and D). Interestingly, we found that

207 dimethylethanolamine (DMEtn) but not monomethylethanolamine (MMEtn)

208 complemented the root phenotype (Fig. 3C, D), suggesting that demethylation is

209 sufficient for functional complementation of the root phenotype in pmt1-1 pmt2-1 and

210 pmt1-1.

211 To further examine the link between PMT activity and root growth, we

212 employed chemical inhibitors against PMT activity. Hexadecylphosphocholine (HePC)

213 and hexadecyltrimethylammonium bromide (HDTA) are known to inhibit PMT activity

214 (Bobenchik et al., 2010), so we treated WT seedlings with these compounds and

215 observed the effect on root growth. As shown in Fig. 3E, 10-day-old WT roots treated

14
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
216 with HePC and HDTA were significantly shorter compared to mock-treated roots, and

217 were indistinguishable from those of pmt1-1 pmt2-1 with mock treatment. Thus,

218 chemical inhibition of PMT in WT reproduced the pmt1-1 pmt2-1 root growth defect,

219 suggesting that PMT1 and PMT2 are responsible for the PMT activity required for root

220 growth.

221

222 Polar glycerolipid contents were differently affected in the shoot and root of pmt1-1

223 pmt2-1 seedlings

224 Because PCho is a precursor for the biosynthesis of PC, a predominant membrane

225 phospholipid class, we analyzed the contents of polar glycerolipid classes in the shoots

226 and roots of 20-day-old seedlings of WT, pmt1-1, and pmt1-1 pmt2-1. In shoots, no

227 significant difference was found in the lipid contents among these lines (Fig. 4A).

228 However, in roots, pmt1-1 pmt2-1 showed a significant decrease in PC content and

229 increase in PE content, whereas the other lipid classes did not show any significant

230 changes (Fig. 4B). These results suggest that PMT1 and PMT2 play roles in PC

231 biosynthesis in roots, and increased PE content may be due to reduced PMT activity,

232 which altered the metabolic flux toward PE biosynthesis.

233

234 Three PMTs differently affected PMT activity in shoot and root

235 To dissect the differential contribution of the three PMT isoforms in shoot and root, we
14
236 performed a pulse-chase assay using C-labeled Etn in 14-d-old seedlings of WT,

237 pmt1-2, pmt2-1, and pmt3-1. Following 15 min of labeling, labeled PEtn and PCho were

238 chased at 0, 4, and 6 h in shoots and roots. As shown in Fig. 5, no significant difference

239 was observed in the levels of PEtn at 4 h and 6 h in shoot and root, although a slight

15
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
240 difference was observed at 0 h, possibly due to the short period of time for chasing.

241 However, profiles of labeled PCho showed some marked differences among the mutants.

16
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
242 In shoots, pmt3-1 showed significantly reduced levels of labeled PCho at 4 h and 6 h

243 compared to WT. At 6 h, pmt2-1 also showed a significant reduction of labeled PCho

17
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
244 compared to WT, albeit to a lesser extent compared to pmt3-1. In roots, all of the pmt

245 mutants showed a significantly reduced level of labeled PCho at 4 h. These results

246 suggest that all PMT isoforms contribute to the PMT activity in roots, whereas PMT3 is

247 the primary contributor in shoots with minor contribution from PMT2.

248

249 The PMT triple mutant was seedling lethal

250 We next focused on the contribution of PMT2 in shoot growth. Recently, the pmt1 pmt3

251 double mutant was shown to severely affect both vegetative and reproductive growth

252 (Chen et al., 2018; Liu et al., 2018a). However, PC biosynthesis is not completely

253 blocked and the genotype is non-lethal. We hypothesized that the remaining PC

254 biosynthesis is owed to PMT2, thus we created the triple mutant pmt1 pmt2 pmt3 and

255 examined its growth and lipid phenotypes. Because pmt1 pmt3 severely reduces fertility,

256 we crossed pmt1 pmt2 with pmt3 to obtain the triple mutant. About 25% of germinating

257 seedlings from pmt1-2 pmt2-2 pmt3-1/+ offspring showed a chlorotic phenotype (Fig.

258 6A) and these were confirmed by PCR-based genotyping to be the triple homozygous

259 mutants. As compared to WT, 10-day-old seedlings of pmt1-2 pmt2-2 pmt3-1 showed

260 pale cotyledons without visible emerging true leaves (Fig. 6B). To observe whether the

261 severe seedling phenotype of pmt1-2 pmt2-2 pmt3-1 is enhanced as compared to the

262 pmt1-2 pmt3-1, we performed time-course observation of seedling growth from 11 to 37

263 days after germination (Fig. 6C). We produced a second triple mutant (pmt1-1 pmt2-1

264 pmt3-2) to confirm the enhancement of the phenotype. Eleven days after germination,

265 pmt1-2 pmt3-1 and pmt1-2 pmt2-2 pmt3-1 showed no clear morphological differences,

266 except that the cotyledons of pmt1-2 pmt3-1 were more greenish. By 17 days after

267 germination, the shoot apices of pmt1-2 pmt3-1 remained greenish and then started to

18
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
268 produce underdeveloped true leaves at 21 days. However, pmt1-2 pmt2-2 pmt3-1 had

269 lost all green tissue by day 14 and no further development occurred at subsequent days.

19
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
270 At the 37-day-old growth stage, pmt1-2 pmt3-1 was dwarfed but alive, whereas pmt1-2

271 pmt2-2 pmt3-1 appeared dead since no morphological change was observed after 14

272 days. The pmt1-1 pmt2-1 pmt3-2 mutant also showed a similar growth profile to pmt1-2

273 pmt2-2 pmt3-1, although pmt1-1 pmt3-2 plants were slightly healthier than pmt1-2

274 pmt3-1 in their seedling growth. We performed chemical complementation of the

275 seedling growth. Seeds obtained from a pmt1-1 pmt2-1 pmt3-2/+ mutant were planted

276 on an MS agar plate with (+PCho) or without (-PCho) 100 μM of PCho and seedlings

277 were observed at 14-d-old. As shown in Fig. S3, seedlings of the triple mutant

278 (indicated in red arrows) in the presence of PCho recovered the cotyledons compared to

279 those without PCho. However, true leaf formation was not fully rescued as they

280 produced narrow leaves even in the presence of PCho. Thus, these observations

281 demonstrate that mutation in PMT2 enhances the severe growth phenotype of pmt1

282 pmt3, and that pmt1 pmt2 pmt3 is lethal at postembryonic seedling growth by 14 days

283 after germination.

284

285 The pmt1 pmt2 pmt3 triple mutant was devoid of de novo PC biosynthesis

286 To investigate PC profiles associated with the lethal phenotype of pmt1 pmt2 pmt3, we

287 first quantified the amount of PC in 14-day-old seedlings of WT, pmt1-2 pmt3-1, and

288 pmt1-2 pmt2-2 pmt3-1. As shown in Fig. 7A, PC content in pmt1-2 pmt3-1 was about

289 half that of WT. In the triple mutant, a further significant reduction in PC content was

290 observed, which was about 20% that of WT. We also analyzed the fatty acid

291 composition of PC (Fig. 7B). Compared to WT, the fatty acid composition of PC in

292 pmt1-2 pmt2-2 pmt3-1 showed a significant increase in 16:0, 16:2, and 18:0 but a

293 decrease in 18:1, 18:2, and 18:3 contents. Thus, PC in pmt1-2 pmt2-2 pmt3-1 possessed

20
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
294 considerably saturated fatty acid profiles with increased C16/C18 ratio. We further

295 analyzed the abundance of four major membrane glycerolipid classes, namely

21
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
296 monogalactosyldiacylglycerol (MGDG), digalactosyldiacylglycerol (DGDG), PE, and

297 PC (Fig. 7C). Compared to WT, pmt1-2 pmt2-2 pmt3-1 showed considerable reduction

298 in MGDG but not DGDG content. Besides, PE content increased markedly while the PC

299 level decreased. Thus, pmt1-2 pmt2-2 pmt3-1 showed an unusual membrane lipid

300 composition, with PE as the predominant lipid class at the expense of MGDG and PC.

301 To examine whether pmt1-2 pmt2-2 pmt3-1 is capable of de novo PC

302 biosynthesis, we labeled 14-d-old WT and pmt1-2 pmt2-2 pmt3-1 seedlings with

303 [14C]Etn and quantified the amount of labeled PCho. As shown in Fig. 8A and Fig. S4,

304 no labeled PCho was detected in the extract of pmt1-2 pmt2-2 pmt3-1, whereas

305 significant amounts were found in WT extracts. To test if PC biosynthesis was affected,

306 we quantified the amount of labeled PC, which again showed no detectable amount in

307 pmt1-2 pmt2-2 pmt3-1 (Fig. 8A). This suggests that pmt1-2 pmt2-2 pmt3-1 is devoid of

308 de novo PC biosynthesis and the remaining PC is derived from a carry-over of

309 choline-containing compounds from embryonic growth. Indeed, seedlings of pmt1-1

310 pmt2-1 pmt3-2 and pmt1-2 pmt2-2 pmt3-1 whose seed coat and aleurone were removed

311 before germination were smaller at the time of growth arrest (Fig. 8B).

312

313

22
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
314 DISCUSSION

315

23
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
316 Role of PMTs in PC biosynthesis and plant growth

317 We found that PMT2 is highly expressed in roots. Mutation of PMT2 did not affect root

318 growth; however, the double mutant pmt1-1 pmt2-1 enhanced the growth defect and

319 reduced PC content in roots of pmt1-1. Although pmt3-1 showed reduced PMT activity

320 in roots (Fig. 5), neither single mutation of PMT3 nor double mutation of PMT2 and

321 PMT3 showed a seedling phenotype, including in roots (Lee and Jez., 2017; Liu et al.,

322 2018). Chemical inhibition of PMT activity reduced the root length in WT as short as

323 that of the pmt1-1 pmt2-1. This suggests that PMT activity is required for root growth,

324 and that PMT1 and PMT2 are contribute to this activity. The pmt1 pmt3 mutant has

325 reduced root growth (Chen et al., 2018). Although PMT3 transcript was hardly

326 detectable in roots (Fig. 1A) and the pmt3 single mutant does not affect root growth

327 (Lee and Jez., 2017), our pulse-chase experiment showed that pmt3-1 roots have

328 reduced PMT activity (Fig. 5). It is possible that reduced root growth in pmt1 pmt3 may

329 be due to further reduction of the PMT activity rather than a secondary effect of

330 defective shoot growth. The pmt1-1 pmt2-1 mutant had further reduced epidermal cell

331 viability (Fig. 2F-N) and main root length (Fig. 2A,B) compared to pmt1-1. We have

332 currently no further explanation as to why defective PMT activity causes such

333 phenotypes. Results of a chemical complementation experiment suggest that at least

334 dimethylated ethanolamine is required to fully rescue root growth in the PMT mutants.

335 Exogenous supplementation of choline or PCho also rescues the root growth in pmt1

336 (Cruz-Ramírez et al., 2004); however, whether these compounds or some derivatives

337 thereof rescued the phenotype remains open to discussion.

338 The pmt1-1 pmt2-1 mutant showed a significant decrease in root PC content

339 (Fig. 4B), which was in line with the result of pulse-chase labeling (Fig. 5). However, a

24
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
340 substantial amount of PC remained in the roots. Remaining PC biosynthetic activity can

341 be explained by PMT3 or PLMT1. PMT3 is hardly expressed in roots (Fig. 1A), but our

342 pulse-chase experiment showed that pmt3-1 roots have reduced PMT activity (Fig. 5).

343 The pmt triple mutant is devoid of de novo PC biosynthesis (Fig. 8A) and is seedling

344 lethal. Since PCho is abundant in the vasculature, shoot-derived PCho can compensate

345 for its loss in roots, possibly through vascular transport. Indeed, a mutant of CHOLINE

346 TRANSPORTER-LIKE 1 (CTL1) affects root growth (Dettmer et al., 2014).

347 Involvement of PLMT1 postulates that the first methylation occurs at phospho-base

348 level, because PLMT1 does not catalyze the first methylation step at the phosphatidyl

349 base (Keogh et al., 2009). The lethal phenotype of the pmt triple mutant demonstrates an

350 essential role of PMTs in post-embryonic growth. However, considering that knockout

351 of most other phospholipid enzymes causes embryonic or gametophytic lethal

352 phenotypes, this phenotype is somewhat mild. Although the triple mutant is devoid of

353 de novo PC biosynthesis (Fig. 7), the mutant still possesses a detectable amount of PC

354 (Fig. 7A). It is possible that the remaining PC is derived from a carry-over of

355 choline-containing compounds from embryonic growth or from an as-yet-unknown PC

356 biosynthetic pathway independent of PMT. Indeed, we observed that removal of the

357 seed coat before germination considerably affects the seedling growth of the triple

358 mutants (Fig. 8B). Thus, our data clearly demonstrate that three PMTs are essential in de

359 novo PC biosynthesis and plant growth, as shown by their different contributions to

360 PMT activity in shoot and root (Fig. 5). This highlights the reaction catalyzed by the

361 phospho-base methyltransferase as an essential reaction step in the primary PC

362 biosynthesis pathway.

363

25
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
364 An updated metabolic map of PC biosynthesis

365 Since the triple mutant pmt1 pmt2 pmt3 has been characterized in this work, we

366 propose an updated metabolic map for the biosynthesis of PC and PE in Arabidopsis

367 (Fig. 9). The initial reaction step is the conversion of serine to ethanolamine by SERINE

368 DECARBOXYLASE 1 (SDC1; Yunus et al, 2016; Liu et al, 2018b). The product

369 ethanolamine is then phosphorylated to PEtn by CHOLINE/ETHANOLAMINE

370 KINASE 4 (CEK4; Lin et al, 2015). Since overexpression of SDC1 or CEK4 increases

371 PC content (Yunus et al, 2016; Lin et al, 2015), the pathway catalyzed by these enzymes

372 contributes to PC biosynthesis. The product PEtn has two alternative metabolic fates. To

373 synthesize PE, PEtn is converted to CDP-Etn by

374 CTP:PHOSPHORYLETHANOLAMINE CYTIDYLYLTRANSFERASE 1 (PECT1)

375 (Mizoi et al 2006), then CDP-Etn is converted to PE by AAPT1 and AAPT2 (Liu et al

376 2015). To synthesize PC, PMT1, PMT2 and PMT3 redundantly function to produce

377 PCho (shown in this work). PCho is converted to CDP-Cho by CCT1 and CCT2

378 (Inatsugi et al 2002; 2009), then CDP-Cho is converted to PC by AAPT1 and AAPT2

379 (Liu et al 2015). In animal cells, PCho biosynthesis from choline by choline kinase

380 contributes significantly to PC biosynthesis because choline is provided through the

381 food intake (Wu and Vance, 2010). In plants, choline is produced from ethanolamine

382 through PMT activity (Rhodes and Hanson., 1993), thus phosphorylation of choline

383 may not be a major pathway for PC biosynthesis. Indeed, none of the putative choline

384 kinases (CEK1, 2, and 3) affect PC contents when knocked out (Lin et al., 2015). In

385 addition, double knock out of two PCho phosphatases (PECP1 and PS2) does not affect

386 PC (Hanchi et al., 2018). Gene knockout studies have shown that SDC1, CEK4, and

387 PECT1 are essential in embryogenesis because their null mutants show an

26
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
388 embryonic-lethal phenotype (Yunus et al, 2016; Lin et al, 2015; Mizoi et al 2006).

389 Besides, the double knockout mutant of AAPT1 and AAPT2 cannot be isolated, possibly

27
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
390 due to a reproductive defect (Liu et al 2015). Considering the similar embryonic defects

391 in these mutants, it is possible that PE biosynthesis is required for embryogenesis and

392 thus is an essential phospholipid class. On the other hand, the requirement of PC in

393 plant development has remained obscure to date. Single mutants of cct1 and cct2

394 showed no defective growth (Inatsugi et al 2002; 2009), and knock-down mutants of

395 AAPT1 and AAPT2 do not affect PC content (Liu et al 2015). Because our study on the

396 triple mutant demonstrated an essential role of PC biosynthesis in plant growth, further

397 investigation is required with regard to the specific developmental defects caused by PC

398 deficiency.

399 In conclusion, our results demonstrated that PMT activity plays an essential

400 role in PC biosynthesis and that three PMTs have tissue-specific functions in PC

401 biosynthesis and plant growth.

402

403 MATERIALS AND METHODS

404 Plant materials and growth conditions

405 Arabidopsis (Arabidopsis thaliana) plants (Columbia-0 ecotype) were grown under

406 16-h light/8-h dark conditions at 22oC. Murashige and Skoog (MS) medium was used at

407 half-strength concentration for plant culture (Murashige and Skoog., 1962). Chemical

408 supplementation to seedlings grown on MS agar medium involved commercially

409 available Etn (E9508, Sigma-Aldrich), N-monomethylethanolamine (MMEtn, 192171,

410 Sigma-Aldrich), N-dimethylethanolamine (DMEtn, 391263, Sigma-Aldrich), and Cho

411 (C7527, Sigma-Aldrich) at a final concentration of 100 μM, and

412 hexadecylphosphocholine (HePC/miltefosine, 63280, Cayman) and

413 hexadecyltrimethylammonium bromide (HDTA, AM-0833, Amersco) at a final

28
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
414 concentration of 100 μM. The T-DNA–tagged mutants, pmt1-1 (CS801444), pmt1-2

415 (SALK_036458), pmt3-1 (SALK_016929), and pmt3-2 (CS827717) were obtained from

416 The Arabidopsis Information Resource (TAIR). The pmt2-1 (FLAG_115A04) and

417 pmt2-2 (FLAG_423E12) mutants were obtained from the Institut National de la

418 Recherche Agronomique (INRA). The original pmt2-1 and pmt2-2 lines were in the

419 Wassilewskija (WS) ecotype, so they were backcrossed 6 times with the Columbia-0

420 ecotype to replace the ecotype background. These mutants were shown to be null (Liu et

421 al., 2018). The double and triple mutants were created by genetic crossing of the

422 respective single mutants. Homozygous plants were isolated by PCR-based genotyping

423 with gene-specific primers and T-DNA-specific primers (Table S1). The combination of

424 primers used to isolate each line was; pmt1-1 (YN1051/YN1052 and YN1051/YN144),

425 pmt1-2 (YN1051/JL71 and YN902/KK8), pmt2-1 (YN904/YN905 and YN904/YN795),

426 pmt2-2 (YN906/YN907 and YN795/YN907), pmt3-1 (JL73/JL74 and KK8/YN909),

427 and pmt3-2 (YN912/YN1283 and YN912/YN144). The location of T-DNA was

428 confirmed by DNA sequencing following information available on TAIR’s official

429 website (www.arabidopsis.org). None of these T-DNA lines were leaky mutants since

430 we did not detect any full-length transcripts of the tagged genes by RT-PCR.

431

432 Vector construction and plant transformation

433 ProPMT1:PMT1 pmt1-1 pmt2-1 and ProPMT2:PMT2 pmt1-1 pmt2-1: genomic

434 sequences for PMT1 (ProPMT1:PMT1) and PMT2 (ProPMT2:PMT2) were amplified

435 by PCR with the primers YN1286 and YN1287, YN1288 and YN1289, respectively.

436 The amplicon was cloned into the pENTR_D_TOPO plasmid vector (Invitrogen) to

437 obtain pCC15 and pCC16, respectively, which was then recombined into the pHGW

29
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
438 destination vector with use of LR Clonase (Invitrogen) (Karimi et al., 2002) to obtain

439 pCC21 and pCC22, respectively. The plasmids were transduced into the pmt1-1 pmt2-1

440 background via Agrobacterium tumefasiens GV3101. Transformants were screened on

441 MS media containing hygromycin. Ten independent transformants were screened for

442 each construct, all of which complemented the short root phenotype (Fig. S2). Lines No.

443 2 and No. 3, respectively, were selected as representative lines.

444 ProPMT2:PMT2-GUS pmt1-1 pmt2-1: the genomic sequence of PMT2 without

445 the stop codon and 3’-untranslated region was amplified by PCR with oligonucleotide

446 primers YN1288 and YN1388. The amplicons were cloned into the pENTR_D_TOPO

447 plasmid vector (Invitrogen) to obtain pYL20, which was recombined with the

448 pGWB533 (Nakagawa et al., 2007) by LR Clonase to obtain pYL24. The plasmid was

449 transduced into the pmt1-1 pmt2-1 background by Agrobacterium tumefasiens GV3101.

450 Transformants were screened on MS media containing hygromycin. Twenty-six

451 independent transformants were screened, 21 of which complemented the short root

452 phenotype with detectable GUS staining. We confirmed the consistent GUS staining

453 pattern in 3 independent transgenic lines (No.7, 17, 23), and line No. 7 was selected as a

454 representative line.

455

456 Extraction and analysis of polar glycerolipids

457 Analysis of polar glycerolipid contents was conducted as previously described

458 (Nakamura et al., 2003). Total lipid was extracted as described (Bligh and Dyer., 1959),

459 and each glycerolipid class was separated on a silica gel thin layer chromatography

460 (TLC) plate using the solvent system of chloroform:methanol:aqueous ammonia

461 (120:80:8 by vol) for the first dimension and chloroform:methanol:acetic acid:water

30
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
462 (170:20:15:3 by vol) for the second dimension. Each lipid spot was detected under UV

463 light using purimuline, scraped off, and collected for hydrolysis and methylesterification

464 of the acyl moieties with HCl-MeOH containing pentadecanoic acid (15:0) as an

465 internal standard. Obtained fatty acid methyl esters were quantified by gas

466 chromatography (GC-2010, Shimadzu, Japan) equipped with a ULBON HR-SS-10

467 column (Shinwa Chemical Industries, Japan). Data shown are means ± SD from three

468 biological replicates.

469

470 Radiolabeling experiment

471 For the pulse-chase experiment, 14-d-old seedlings of WT, pmt1-2, pmt2-1, and

472 pmt3-1 were submerged into 4 ml of ½ MS containing 220 pmol of [14C]Etn

473 (55mCi/mmol) for 15 min. After washing the seedlings with ½ MS, samples were

474 incubated under light and harvested at 0, 4, and 6 h after the start of chasing. Tissues

475 were ground under liquid N2, and extraction was carried out on ice by 90 μl of

476 dichloromethane:acetone (3:1, v/v), and 20 μl of 50 mM ammonium formate (pH 3)

477 (Tannert et al., 2018). The aqueous phase (containing choline, phosphocholine,

478 ethanolamine, and phosphoethanolamine) was separated by centrifugation at 13,000 x g,

479 4 oC for 5 min. The extracts of the upper aqueous phase for each of the above assays

480 were spotted on Silica gel 60G TLC plates (Merck) and resolved by using a developing

481 solvent of 95% ethanol and 2% NH4OH (1:1, v/v) (Monks et al., 1996). Radioactive

482 spots were visualized by Imaging Plate (Fuji Film) and quantified by image analyzer

483 (BAS-2500, GE Healthcare). For radiolabeling of WT and pmt1-1 pmt2-1 pmt3-2,

484 14-d-old seedlings were labeled under the above-mentioned conditions for 3 h and

485 harvested immediately for the analysis as described above.

31
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
486

487 Histochemical GUS staining

488 GUS staining was performed as described previously (Lin et al., 2015).

489

490 Observation of roots

491 Root cells stained with propidium iodide (PI) and fluorescein diacetate (FDA) were

492 observed with a confocal microscope (LSM 510 Meta, Zeiss) as described previously

493 (Cruz-Ramírez et al., 2004), except that FDA concentration was 5 μg/ml and staining

494 time was 5 min.

495

496 Removal of seed coat

497 Seed coat was removed as reported previously (Lee and Lopez-Molina, 2013).

498

499 Nomenclature

500 The phospho-base N-methyltransferases have been abbreviated differently in

501 Arabidopsis. NMT: This abbreviation has been used in a few reports e.g. (Bolognese

502 and McGraw, 2000; Chen et al., 2018). However, NMT also stands for the

503 N-myristoyltransferases (Pierre et al., 2007), thus is not a unique abbreviation.

504 XPL: This abbreviation was given to At3g18000 because a forward-genetic screen

505 found that mutation in this gene causes swelling in epidermal cells (XPL stands for

506 XIPOTL, which means swelling or tumefaction in Nahuatl language; Cruz-Ramírez et

507 al., 2004). However, knockout mutants of the second and third isoforms (At1g48600

508 and At1g73600) do not cause any swelling phenotype, which suggests that XPL may not

509 be a suitable name for this gene family.

32
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
510 PEAMT: This abbreviation stands for S-adenosylmethionine:phospho-ethanolamine

511 N-methyltransferase, and was used in a few papers e.g. (Mou et al., 2002; Zhang et al.,

512 2010). However, in vitro enzyme activity assay and heterologous complementation

513 assay have shown that this enzyme accepts mono- and di-methylethanolamine as

514 substrates (Bolognese and McGraw., 2000; Chen et al., 2018), so this abbreviation does

515 not accurately reflect the enzymatic feature.

516 PMEAMT: This abbreviation stands for phospho-methylethanolamine

517 N-methyltransferase (PMEAMT) and was used to name At1g48600 because yeast

518 heterologous complementation assay showed that this isozyme does not catalyze the

519 first methylation reaction (BeGora et al., 2010). However, later, this isozyme was shown

520 to catalyze the first methylation reaction as well in vitro (Lee and Jez., 2017). Since the

521 other two isozymes also catalyze all the methylation reactions, this abbreviation does

522 not accurately reflect the enzymatic feature.

523 PMT: This abbreviation stands for phospho-base N-methyltransferase (PMT) and has

524 been used in Arabidopsis (Lee and Jez., 2017) and other models including Plasmodium

525 falciparum (a human malaria parasite) (Bobenchik et al., 2011). Since this abbreviation

526 describes features of this enzyme family without problems and also is in accordance

527 with the community standard of nomenclature (Meinke and Koornneef, 1997), we have

528 employed this abbreviation in this article.

529

530 ACCESSION NUMBERS

531 Sequence data from this article can be found under the following Arabidopsis Genome

532 Initiative accession numbers: PMT1 (At3g18000), PMT2 (At1g48600), PMT3

533 (At1g73600)

33
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
534

535 SUPPLEMENTAL DATA

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

537 Supplemental Figure S1. Measurement of root length in 7-day-old seedlings of

538 ProPMT2:PMT2-GUS pmt1-1 pmt2-1 (Line 7) in comparison with WT, pmt1-1, and

539 pmt1-1 pmt2-1.

540 Supplemental Figure S2. Measurement of root length in 7-day-old seedlings of 10

541 independent T2 lines of ProPMT1:PMT1 pmt1-1 pmt2-1 and ProPMT2:PMT2 pmt1-1

542 pmt2-1 compared to WT, pmt1-1, and pmt1-1 pmt2-1.

543 Supplemental Figure S3. Chemical complementation of the seedling phenotype of

544 pmt1-1 pmt2-1 pmt3-2.

545 Supplemental Figure S4. A TLC image showing spots of radioactive compounds

546 produced by labelling the 14-d-old seedlings of wild type and pmt1-1 pmt2-1 pmt3-2

547 with [14C]Etn.

548 Supplemental Table S1. List of oligonucleotide sequences used in this study.

549

550 ACKNOWLEDGEMENTS

551 We thank Chia-En Chen (Plant and Microbial Biology, Academia Sinica) for molecular

552 construction. The research was supported by the core budget of the Institute of Plant and

553 Microbial Biology, Academia Sinica (Y.N. and K.K.), and the Ministry of Science and

554 Technology, Taiwan (No. 106-2628-B-001-002 to Y.N). The authors have no conflict of

555 interest.

556

557 FIGURE LEGENDS

34
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
558 Figure 1. Expression pattern of PMT2. A, A heatmap of the tissue-specific gene

559 expression patterns of PMT1, PMT2, and PMT3. Data shown were obtained from

560 GENEVESTIGATOR. B, Expression patterns of PMT1, PMT2, and PMT3 at different

561 developmental stages. Stage of development (left to right); germinating seed, seedling,

562 young rosette, developed rosette, bolting rosette, young flower, developed flower,

563 flowers and siliques, mature siliques, and senescence. “High”, “Medium” and “Low”

564 expression were calculated by microarray assay. The number of samples indicates

565 microarray gene expression data collected by GENEVESTIGATOR. C to G,

566 Tissue-specific expression of PMT2-GUS in Arabidopsis ProPMT2:PMT2-GUS pmt1-1

567 pmt2-1 in a 14-day-old seedling (C), flowers at different developmental stages (D), a

568 young floral bud (E), a flower before anther dehiscence (F), and a mature flower (G).

569 Bars; 10 mm (C), 500 μm (D), 50 μm (E), 100 μm (F), and 200 μm (G).

570

571 Figure 2. Mutation of PMT2 enhanced the root phenotype of the pmt1 mutant. A,

572 Phenotype of 20-day-old seedlings of the pmt1-1, pmt1-2, pmt2-1, pmt2-2, and pmt1-1

573 pmt2-1 mutants as compared to WT. B, Quantitative measurement of root length

574 observed in (A). Data are mean ± SD from 16 seedlings with three biological replicates.

575 Statistical significance was analyzed by Student’s t-test (P<0.01, **; P<0.001, ***). C,

576 Daily observation of seedling root development in pmt1-1 (C), pmt1-1 pmt2-1 (D), and

577 WT (E) from 1 to 6 days after germination (from left to right). Bars = 5 mm. F to H,

578 Epidermal morphology of main roots producing lateral branches in wild type (F),

579 pmt1-1 (G), and pmt1-1 pmt2-1 (H). Bars = 50 μm. I to N, Viability of root cells by

580 double staining with propidium iodide (PI; red) and fluorescein diacetate (FDA; green).

581 Fluorescent (I, K, M) and bright-field (J, L, N) images of wild type (I and J), pmt1-1 (K

35
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
582 and L) and pmt1-1 pmt2-1 (M and N) are shown. Bars = 0.1 mm.

583

584 Figure 3. Genetic and chemical complementation of root growth in pmt1-1 pmt2-1. A

585 and B, Overall seedling phenotype (A) and root length (B) of the 7-day-old seedlings of

586 WT, pmt1-1 pmt2-1, ProPMT1:PMT1 pmt1-1 pmt2-1, and ProPMT2:PMT2 pmt1-1

587 pmt2-1. C and D, Chemical complementation of root phenotype by 100 μM of

588 ethanolamine (Etn), N-monomethylethanolamine (MMEtn), N-dimethylethanolamine

589 (DMEtn), and choline (Cho). Overall seedling phenotype (C) and root length (D) of

590 14-day-old pmt1-1, pmt1-1 pmt2-1, and WT seedlings. E, Chemical inhibition of PMT

591 activity reproduced the defective root growth phenotype of pmt1-1 pmt2-1. Root length

592 of 10-day-old WT seedlings treated with 100 μM of hexadecylphosphocholine (HePC)

593 or hexadecyltrimethylammonium bromide (HDTA) as compared to mock treatment

594 (Mock) of WT, pmt1-1, and pmt1-1 pmt2-1. Data for the measurement of root length are

595 mean ± SD from 16 seedlings and 3 biologically independent experiments. Statistical

596 significance was analyzed by Student’s t-test (P<0.01, **; P<0.001, ***).

597

598 Figure 4. Polar glycerolipid profiles in the shoot (A) and root (B) of 14-day-old

599 seedlings of pmt1-1 pmt2-1 as compared to WT and pmt1-1. Seedlings were grown on

600 MS medium at half-strength concentration. Data are mean ± SD from 3 biological

601 replicates. Statistical significance between WT and each mutant was analyzed by

602 Student’s t-test (P<0.05, *). MGDG, monogalactosyldiacylglycerol; DGDG,

603 digalactosyldiacylglycerol; SQDG, sulfoquinovosyldiacylglycerol; PE,

604 phosphatidylethanolamine; PC, phosphatidylcholine; PG, phosphatidylglycerol; PI,

605 phosphatidylinositol.

36
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
606

607 Figure 5. Pulse-chase assay of 14-d-old seedlings of WT, pmt1-2, pmt2-1, and pmt3-1
14
608 using C-labeled Etn. Following 15 min of labeling, seedlings were harvested at 0, 4,

609 and 6 h and shoot and root were separated for the analysis of labeled compounds by

610 TLC. Radioactive intensity of [14C]PEtn and [14C]PCho were quantified and plotted by

611 % of total intensity of labeled compounds. Data are mean ± SD from 3 biological

612 replicates. Statistical significance was analyzed by Student’s t-test (P<0.05, *; P<0.01,

613 **; P<0.001, ***).

614

615 Figure 6. Isolation and observation of the pmt1 pmt2 pmt3 triple mutants. A, Isolation

616 of pale seedlings from the germinating offspring of a pmt1-2 pmt2-2 pmt3-1/+ plant.

617 Pale seedlings were found at a frequency of about 25% in the population of pmt1-2

618 pmt2-2 pmt3-1/+ offspring. B, A magnified image of 10-day-old WT and pmt1-2 pmt2-2

619 pmt3-1 seedlings. C, Time-course observation from 11 to 37 days after germination of

620 defective seedling growth for 2 different alleles of pmt1 pmt2 pmt3 triple mutants

621 compared to the respective alleles of pmt1 pmt3 double mutants.

622

623 Figure 7. Analysis of PC and other primary membrane glycerolipid classes (MGDG,

624 DGDG, and PE) in the 14-day-old seedlings of WT and pmt1-2 pmt2-2 pmt3-1. A,

625 Amount of PC in WT, pmt1-2 pmt3-1, and pmt1-2 pmt2-2 pmt3-1 per dry tissue weight

626 (DW) in 14-day-old seedlings. B, Fatty acid composition (mol %) of PC analyzed in (A).

627 C, Contents of MGDG, DGDG, PE, and PC shown by mol % in the 14-day-old

628 seedlings of WT and pmt1-2 pmt2-2 pmt3-1. Data are mean ± SD from 3 biological

629 replicates. Statistical significance was analyzed by Student’s t-test (P<0.05, *; P<0.01,

37
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
630 **; P<0.001, ***). In panels B and C, significance was analyzed as compared to WT.

631

632 Figure 8. Characterization of pmt1 pmt2 pmt3 triple mutants. A, Levels of radiolabeled

633 PCho and PC in the 14-day-old seedlings of WT and pmt1-1 pmt2-1 pmt3-2 following

634 [14C]Etn labeling. Radioactive intensity (arbitrary unit) was normalized to the fresh

635 weight of seedlings. Data are mean ± SD from 3 biological replicates. N.D, not

636 detectable. B, Phenotype of the 14-day-old seedlings of WT, pmt1-1 pmt2-1 pmt3-2 and

637 pmt1-2 pmt2-2 pmt3-1 with or without removing the seed coat before germination. A

638 representative image was shown for each condition (n>30). Bars, 2mm.

639

640 Figure 9. An updated metabolic pathway for the biosynthesis of PE and PC in

641 Arabidopsis. Methylation of PEtn to produce PCho, catalyzed by PMT1, PMT2, and

642 PMT3, is a critical step in PC biosynthesis. AAPT, aminoalcohol

643 aminophosphotransferase; CCT, CTP:phosphorylcholine cytidylyltransferase; CDP-Cho,

644 cytidine diphosphocholine; CDP-Etn, cytidine diphosphoethanolamine; CEK,

645 choline/ethanolamine kinase; Cho, choline; Etn, ethanolamine; PC,

646 phosphatidylcholine; PCho, phosphocholine; PE, phosphatidylethanolamine; PECP1,

647 phosphoethanolamine/phosphocholine phosphatase 1; PECT1,

648 CTP:phosphorylethanolamine cytidylyltransferase 1; PEtn, phosphoethanolamine; PMT,

649 S-adenosylmethionine:phospho-base N-methyltransferase; PS2, phosphate

650 starvation-induced gene 2; Ser, serine; SDC1, serine decarboxylase 1.

651

652 Supplemental Figure S1. Measurement of root length in 7-day-old seedlings of

653 ProPMT2:PMT2-GUS pmt1-1 pmt2-1 (Line 7) in comparison with WT, pmt1-1, and

38
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
654 pmt1-1 pmt2-1. Data are mean ± SD from 8 seedlings. Statistical significance was

655 analyzed by Student’s t-test (P<0.05, *; P<0.01, **).

656

657 Supplemental Figure S2. Measurement of root length in 7-day-old seedlings of 10

658 independent T2 lines of ProPMT1:PMT1 pmt1-1 pmt2-1 and ProPMT2:PMT2 pmt1-1

659 pmt2-1 compared to WT, pmt1-1, and pmt1-1 pmt2-1. (A) ProPMT1:PMT1 pmt1-1

660 pmt2-1 lines compared to pmt1-1 pmt2-1 and WT. (B) ProPMT2:PMT2 pmt1-1 pmt2-1

661 compared to pmt1-1 pmt2-1 and pmt1-1. Data are mean ± SD from 8 seedlings.

662 Statistical significance was analyzed by Student’s t-test (P<0.05, *; P<0.01, **).

663

664 Supplemental Figure S3. Chemical complementation of the seedling phenotype of

665 pmt1-1 pmt2-1 pmt3-2. Seeds obtained from a pmt1-1 pmt2-1 pmt3-2/+ plant were

666 planted on a ½ MS agar plate with (+PCho) or without (-PCho) 100 μM of PCho and

667 seedlings were observed at 14-d-old. Indicated by red arrows are the triple homozygous

668 seedlings (25% of total seedlings). Bars, 2mm.

669

670 Supplemental Figure S4. A TLC image showing spots of radioactive compounds

671 produced by labelling the 14-d-old seedlings of wild type and pmt1-1 pmt2-1 pmt3-2

672 with [14C]Etn. Only PCho spots were undetectable in the extracts of pmt1-1 pmt2-1

673 pmt3-2. Samples #1 to #3 are the 3 biological replicates.

674

675 Table. S1 List of oligonucleotide sequences used in this study.

39
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
676 LITERATURE CITED

677 BeGora MD, Macleod MJ, McCarry BE, Summers PS, Weretilnyk EA (2010)

678 Identification of phosphomethylethanolamine N-methyltransferase from Arabidopsis

679 and its role in choline and phospholipid metabolism. J Biol Chem 285: 29147-55

680 Bligh EG, Dyer WJ (1959) A rapid method of total lipid extraction and purification.

681 Can J Biochem Physiol 37: 911-7

682 Bobenchik AM, Choi J-Y, Mishra A, Rujan IN, Hao B, Voelker DR, Hoch JC,

683 Mamoun CB (2010) Identification of inhibitors of Plasmodium falciparum

684 phosphoethanolamine methyltransferase using an enzyme-coupled transmethylation

685 assay. BMC Biochem 11: 4

686 Bobenchik AM, Augagneur Y, Hao B, Hoch JC, Mamoun CB (2011)

687 Phosphoethanolamine methyltransferases in phosphocholine biosynthesis: functions and

688 potential for antiparasite therapy. FEMS Microbiol Rev 35: 609-19

689 Bolognese CP, McGraw P (2000) The isolation and characterization in yeast of a gene

690 for Arabidopsis S-adenosylmethionine:phospho-ethanolamine N-methyltransferase.

691 Plant Physiol 124: 1800-13

692 Celenza JL Jr, Grisafi PL, Fink GR (1995) A pathway for lateral root formation in

693 Arabidopsis thaliana. Genes Dev 9: 2131-42

694 Chaves I, Regalado AP, Chen M, Ricardo CP, Showalter AM (2002) Programmed

695 cell death induced by (β-d-galactosyl)3 Yariv reagent in Nicotiana tabacum BY-2

696 suspension-cultured cells. Physiol Plant 116: 548–53

697 Chen W, Salari H, Taylor MC, Jost R, Berkowitz O, Barrow R, Qiu D, Branco R,

698 Masle J. (2018) NMT1 and NMT3 N-methyltransferase activity is critical to lipid

699 homeostasis, morphogenesis and reproduction. Plant Physiol doi: 10.1104/pp.18.00457

40
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
700 Cruz-Ramírez A, López-Bucio J, Ramírez-Pimentel G, Zurita-Silva A,

701 Sánchez-Calderon L, Ramírez-Chávez E, González-Ortega E, Herrera-Estrella L.

702 (2004) The xipotl mutant of Arabidopsis reveals a critical role for phospholipid

703 metabolism in root system development and epidermal cell integrity. Plant Cell 16:

704 2020-34

705 Datko AH, Mudd SH (1988) Phosphatidylcholine synthesis: differing patterns in

706 soybean and carrot. Plant Physiol 88: 854-61

707 Dettmer J, Ursache R, Campilho A, Miyashima S, Belevich I, O'Regan S,

708 Mullendore DL, Yadav SR, Lanz C, Beverina L, et al. (2014) CHOLINE

709 TRANSPORTER-LIKE1 is required for sieve plate development to mediate

710 long-distance cell-to-cell communication. Nat Commun 5:4276.

711 Devaiah SP, Roth MR, Baughman E, Li M, Tamura P, Jeannotte R, Welti R,

712 Wang X (2006) Quantitative profiling of polar glycerolipid species from organs of

713 wild-type Arabidopsis and a PHOSPHOLIPASE Da1 knockout mutant. Phytochemistry

714 67: 1907-24

715 Hanchi M, Thibaud MC, Légeret B, Kuwata K, Pochon N, Beisson F, Cao A,

716 Cuyas L, David P, Doerner P, et al (2018) The phosphate fast-responsive genes

717 PECP1 and PPsPase1 affect phosphocholine and phosphoethanolamine content. Plant

718 Physiol 176: 2943-62

719 Hanson AD, Rhodes D (1983) 14C tracer evidence for synthesis of choline and betaine

720 via phosphoryl base intermediates in salinized sugarbeet leaves. Plant Physiol 71:

721 692-700

722 Hruz T, Laule O, Szabo G, Wessendorp F, Bleuler S, Oertle L, Widmayer P,

723 Gruissem W, P Zimmermann (2008) Genevestigator V3: a reference expression

41
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
724 database for the meta-analysis of transcriptomes. Advances in Bioinformatics 2008,

725 420747

726 Inatsugi R, Nakamura M, Nishida I (2002) Phosphatidylcholine biosynthesis at low

727 temperature: differential expression of CTP: phosphorylcholine cytidylyltransferase

728 isogenes in Arabidopsis thaliana. Plant Cell Physiol 43: 1342-50.

729 Inatsugi R, Kawai H, Yamaoka Y, Yu Y, Sekiguchi A, Nakamura M, Nishida I

730 (2009) Isozyme-specific modes of activation of CTP:phosphorylcholine

731 cytidylyltransferase in Arabidopsis thaliana at low temperature. Plant Cell Physiol 50:

732 1727-35

733 Karimi M, Inzé D, Depicker A (2002) GATEWAYTM vectors for

734 Agrobacterium-mediated plant transformation. Trends Plant Sci 7: 193-5

735 Keogh MR, Courtney PD, Kinney AJ, Dewey RE (2009) Functional characterization

736 of phospholipid N-methyltransferases from Arabidopsis and soybean. J Biol Chem 284:

737 15439-47.

738 Kirik V, Bouyer D, Schöbinger U, Bechtold N, Herzog M, Bonneville JM,

739 Hülskamp M (2001) CPR5 is involved in cell proliferation and cell death control and

740 encodes a novel transmembrane protein. Curr Biol 11: 1891-5

741 Kodaki T, Yamashita S (1989) Characterization of the methyltransferases in the yeast

742 phosphatidylethanolamine methylation pathway by selective gene disruption. Eur J

743 Biochem 185: 243-51

744 Lee KP, Lopez-Molina L (2013) A seed coat bedding assay to genetically explore in

745 vitro how the endosperm controls seed germination in Arabidopsis thaliana. J Vis Exp

746 e50732

42
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
747 Lee SG, Jez JM (2017) Conformational changes in the di-domain structure of

748 Arabidopsis phosphoethanolamine methyltransferase leads to active-site formation. J

749 Biol Chem 292: 21690-702

750 Lin Y-C, Liu Y-c, Nakamura Y (2015) The choline/ethanolamine kinase family in

751 Arabidopsis: essential role of CEK4 in phospholipid biosynthesis and embryo

752 development. Plant Cell 27: 1497-511

753 Liu Y, Wang G, Wang X (2015) Role of aminoalcoholphosphotransferases 1 and 2 in

754 phospholipid homeostasis in Arabidopsis. Plant Cell 27: 1512-28

755 Liu Y-c, Gunawan F, Yunus IS, Nakamura Y (2018b) Arabidopsis serine

756 decarboxylase 1 (SDC1) in phospholipid and amino acid metabolism. Front Plant Sci

757 doi: 10.3389/fpls.2018.00972

758 Liu Y-c, Lin Y-C, Kanehara K, Nakamura Y (2018a) A pair of phospho-base

759 methyltransferases important for phosphatidylcholine biosynthesis in Arabidopsis. Plant

760 J doi:10.1111/tpj.14090

761 Meinke D, Koornneef M (1997) Community standards for Arabidopsis genetics. Plant

762 J 12: 247-53

763 Mizoi J, Nakamura M, Nishida I (2006) Defects in

764 CTP:PHOSPHORYLETHANOLAMINE CYTIDYLYLTRANSFERASE affect

765 embryonic and postembryonic development in Arabidopsis. Plant Cell 18: 3370-3385

766 Mou Z, Wang X, Fu Z, Dai Y, Han C, Ouyang J, Bao F, Hu Y, Li J (2002).

767 Silencing of phosphoethanolamine N-methyltransferase results in temperature-sensitive

768 male sterility and salt hypersensitivity in Arabidopsis. Plant Cell 14: 2031-43

769 Murashige T, Skoog F (1962) A revised medium for rapid growth and bio assays with

770 tobacco tissue cultures. Physiol Plant 15: 473-97

43
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
771 Nakagawa T, Kurose T, Hino T, Tanaka K, Kawamukai M, Niwa Y, Toyooka K,

772 Matsuoka K, Jinbo T, Kimura T (2007) Development of series of gateway binary

773 vectors, pGWBs, for realizing efficient construction of fusion genes for plant

774 transformation. J Biosci Bioeng 104: 34-41

775 Monks DE, Goode JH, Dewey RE (1996). Characterization of soybean choline kinase

776 cDNAs and their expression in yeast and Escherichia coli. Plant Physiol 110:

777 1197-1205.

778 Nakamura Y, Arimitsu H, Yamaryo Y, Awai K, Masuda T, Shimada H, Takamiya

779 K, Ohta H (2003) Digalactosyldiacylglycerol is a major glycolipid in floral organs of

780 Petunia hybrida. Lipids 38: 1107-12

781 Nuccio ML, Ziemak MJ, Henry SA, Weretilnyk EA, Hanson AD (2000) cDNA

782 cloning of phosphoethanolamine N-methyltransferase from spinach by complementation

783 in Schizosaccharomyces pombe and characterization of the recombinant enzyme. J Biol

784 Chem 275: 14095-101

785 Pierre M, Traverso JA, Boisson B, Domenichini S, Bouchez D, Giglione C, Meinnel

786 T (2007) N-Myristoylation regulates the SnRK1 pathway in Arabidopsis. Plant Cell 19:

787 2804-21

788 Rhodes D, Hanson AD (1993) Quaternary ammonium and tertiary sulfonium

789 compounds in higher plants. Annu Rev Plant Physiol Plant Mol Biol 44: 357-84.

790 Summers PS, Weretilnyk EA (1993) Choline synthesis in spinach in relation to salt

791 stress. Plant Physiol 103: 1269-76

792 Tannert M, May A, Ditfe D, Berger S, Balcke GU, Tissier A, et al. (2018) Pi

793 starvation-dependent regulation of ethanolamine metabolism by phosphoethanolamine

794 phosphatase PECP1 in Arabidopsis roots. J Exp Bot 69: 467-481

44
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
795 Wu G, Vance DE (2010) Choline kinase and its function. Biochem Cell Biol 88:

796 559-64

797 Yunus IS, Liu Y-c, Nakamura Y (2016) The importance of SERINE

798 DECARBOXYLASE1 (SDC1) and ethanolamine biosynthesis during embryogenesis of

799 Arabidopsis thaliana. Plant J 88: 559-69

800 Zhang H, Murzello C, Sun Y, Kim M-S, Xie X, Jeter RM, Zak JC, Dowd SE, Paré

801 PW (2010) Choline and osmotic-stress tolerance induced in Arabidopsis by the soil

802 microbe Bacillus subtilis (GB03). Mol Plant Microbe Interact 23: 1097-104

45
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
Parsed Citations
BeGora MD, Macleod MJ, McCarry BE, Summers PS, Weretilnyk EA (2010) Identification of phosphomethylethanolamine N-
methyltransferase from Arabidopsis and its role in choline and phospholipid metabolism. J Biol Chem 285: 29147-55
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Bligh EG, Dyer WJ (1959) A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37: 911-7
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Bobenchik AM, Choi J-Y, Mishra A, Rujan IN, Hao B, Voelker DR, Hoch JC, Mamoun CB (2010) Identification of inhibitors of Plasmodium
falciparum phosphoethanolamine methyltransferase using an enzyme-coupled transmethylation assay. BMC Biochem 11: 4
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Bobenchik AM, Augagneur Y, Hao B, Hoch JC, Mamoun CB (2011) Phosphoethanolamine methyltransferases in phosphocholine
biosynthesis: functions and potential for antiparasite therapy. FEMS Microbiol Rev 35: 609-19
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Bolognese CP, McGraw P (2000) The isolation and characterization in yeast of a gene for Arabidopsis S-adenosylmethionine:phospho-
ethanolamine N-methyltransferase. Plant Physiol 124: 1800-13
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Celenza JL Jr, Grisafi PL, Fink GR (1995) A pathway for lateral root formation in Arabidopsis thaliana. Genes Dev 9: 2131-42
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Chaves I, Regalado AP, Chen M, Ricardo CP, Showalter AM (2002) Programmed cell death induced by (β-d-galactosyl)3 Yariv reagent in
Nicotiana tabacum BY-2 suspension-cultured cells. Physiol Plant 116: 548–53
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Chen W, Salari H, Taylor MC, Jost R, Berkowitz O, Barrow R, Qiu D, Branco R, Masle J. (2018) NMT1 and NMT3 N-methyltransferase
activity is critical to lipid homeostasis, morphogenesis and reproduction. Plant Physiol doi: 10.1104/pp.18.00457
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Cruz-Ramírez A, López-Bucio J, Ramírez-Pimentel G, Zurita-Silva A, Sánchez-Calderon L, Ramírez-Chávez E, González-Ortega E,


Herrera-Estrella L. (2004) The xipotl mutant of Arabidopsis reveals a critical role for phospholipid metabolism in root system
development and epidermal cell integrity. Plant Cell 16: 2020-34
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Datko AH, Mudd SH (1988) Phosphatidylcholine synthesis: differing patterns in soybean and carrot. Plant Physiol 88: 854-61
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Dettmer J, Ursache R, Campilho A, Miyashima S, Belevich I, O'Regan S, Mullendore DL, Yadav SR, Lanz C, Beverina L, et al. (2014)
CHOLINE TRANSPORTER-LIKE1 is required for sieve plate development to mediate long-distance cell-to-cell communication. Nat
Commun 5:4276.
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Devaiah SP, Roth MR, Baughman E, Li M, Tamura P, Jeannotte R, Welti R, Wang X (2006) Quantitative profiling of polar glycerolipid
species from organs of wild-type Arabidopsis and a PHOSPHOLIPASE Da1 knockout mutant. Phytochemistry 67: 1907-24
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Hanchi M, Thibaud MC, Légeret B, Kuwata K, Pochon N, Beisson F, Cao A, Cuyas L, David P, Doerner P, et al (2018) The phosphate
fast-responsive genes PECP1 and PPsPase1 affect phosphocholine and phosphoethanolamine content. Plant Physiol 176: 2943-62
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Hanson AD, Rhodes D (1983) 14C tracer evidence for synthesis of choline and betaine via phosphoryl base intermediates in salinized
sugarbeet leaves. Plant Physiol 71: 692-700
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Hruz T, Laule O, Szabo G, Wessendorp F, Bleuler S, Oertle L, Widmayer P, Gruissem W, P Zimmermann (2008) Genevestigator V3: a
reference expression database for the meta-analysis of transcriptomes. Advances in Bioinformatics 2008, 420747
Pubmed: Author and Title Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
Google Scholar: Author Only Title Only Author and Title

Inatsugi R, Nakamura M, Nishida I (2002) Phosphatidylcholine biosynthesis at low temperature: differential expression of CTP:
phosphorylcholine cytidylyltransferase isogenes in Arabidopsis thaliana. Plant Cell Physiol 43: 1342-50.
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Inatsugi R, Kawai H, Yamaoka Y, Yu Y, Sekiguchi A, Nakamura M, Nishida I (2009) Isozyme-specific modes of activation of
CTP:phosphorylcholine cytidylyltransferase in Arabidopsis thaliana at low temperature. Plant Cell Physiol 50: 1727-35
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Karimi M, Inzé D, Depicker A (2002) GATEWAYTM vectors for Agrobacterium-mediated plant transformation. Trends Plant Sci 7: 193-5
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Keogh MR, Courtney PD, Kinney AJ, Dewey RE (2009) Functional characterization of phospholipid N-methyltransferases from
Arabidopsis and soybean. J Biol Chem 284: 15439-47.
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Kirik V, Bouyer D, Schöbinger U, Bechtold N, Herzog M, Bonneville JM, Hülskamp M (2001) CPR5 is involved in cell proliferation and
cell death control and encodes a novel transmembrane protein. Curr Biol 11: 1891-5
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Kodaki T, Yamashita S (1989) Characterization of the methyltransferases in the yeast phosphatidylethanolamine methylation pathway by
selective gene disruption. Eur J Biochem 185: 243-51
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Lee KP, Lopez-Molina L (2013) A seed coat bedding assay to genetically explore in vitro how the endosperm controls seed
germination in Arabidopsis thaliana. J Vis Exp e50732
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Lee SG, Jez JM (2017) Conformational changes in the di-domain structure of Arabidopsis phosphoethanolamine methyltransferase
leads to active-site formation. J Biol Chem 292: 21690-702
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Lin Y-C, Liu Y-c, Nakamura Y (2015) The choline/ethanolamine kinase family in Arabidopsis: essential role of CEK4 in phospholipid
biosynthesis and embryo development. Plant Cell 27: 1497-511
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Liu Y, Wang G, Wang X (2015) Role of aminoalcoholphosphotransferases 1 and 2 in phospholipid homeostasis in Arabidopsis. Plant
Cell 27: 1512-28
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Liu Y-c, Gunawan F, Yunus IS, Nakamura Y (2018b) Arabidopsis serine decarboxylase 1 (SDC1) in phospholipid and amino acid
metabolism. Front Plant Sci doi: 10.3389/fpls.2018.00972
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Liu Y-c, Lin Y-C, Kanehara K, Nakamura Y (2018a) A pair of phospho-base methyltransferases important for phosphatidylcholine
biosynthesis in Arabidopsis. Plant J doi:10.1111/tpj.14090
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Meinke D, Koornneef M (1997) Community standards for Arabidopsis genetics. Plant J 12: 247-53
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Mizoi J, Nakamura M, Nishida I (2006) Defects in CTP:PHOSPHORYLETHANOLAMINE CYTIDYLYLTRANSFERASE affect embryonic
and postembryonic development in Arabidopsis. Plant Cell 18: 3370-3385
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Mou Z, Wang X, Fu Z, Dai Y, Han C, Ouyang J, Bao F, Hu Y, Li J (2002). Silencing of phosphoethanolamine N-methyltransferase results
in temperature-sensitive male sterility and salt hypersensitivity in Arabidopsis. Plant Cell 14: 2031-43
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Downloaded from on December 6, 2018 - Published by www.plantphysiol.org
Copyright © 2018 American Society of Plant Biologists. All rights reserved.
Murashige T, Skoog F (1962) A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiol Plant 15: 473-97
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Nakagawa T, Kurose T, Hino T, Tanaka K, Kawamukai M, Niwa Y, Toyooka K, Matsuoka K, Jinbo T, Kimura T (2007) Development of
series of gateway binary vectors, pGWBs, for realizing efficient construction of fusion genes for plant transformation. J Biosci Bioeng
104: 34-41
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Monks DE, Goode JH, Dewey RE (1996). Characterization of soybean choline kinase cDNAs and their expression in yeast and
Escherichia coli. Plant Physiol 110: 1197-1205.
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Nakamura Y, Arimitsu H, Yamaryo Y, Awai K, Masuda T, Shimada H, Takamiya K, Ohta H (2003) Digalactosyldiacylglycerol is a major
glycolipid in floral organs of Petunia hybrida. Lipids 38: 1107-12
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Nuccio ML, Ziemak MJ, Henry SA, Weretilnyk EA, Hanson AD (2000) cDNA cloning of phosphoethanolamine N-methyltransferase from
spinach by complementation in Schizosaccharomyces pombe and characterization of the recombinant enzyme. J Biol Chem 275: 14095-
101
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Pierre M, Traverso JA, Boisson B, Domenichini S, Bouchez D, Giglione C, Meinnel T (2007) N-Myristoylation regulates the SnRK1
pathway in Arabidopsis. Plant Cell 19: 2804-21
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Rhodes D, Hanson AD (1993) Quaternary ammonium and tertiary sulfonium compounds in higher plants. Annu Rev Plant Physiol Plant
Mol Biol 44: 357-84.
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Summers PS, Weretilnyk EA (1993) Choline synthesis in spinach in relation to salt stress. Plant Physiol 103: 1269-76
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Tannert M, May A, Ditfe D, Berger S, Balcke GU, Tissier A, et al. (2018) Pi starvation-dependent regulation of ethanolamine metabolism
by phosphoethanolamine phosphatase PECP1 in Arabidopsis roots. J Exp Bot 69: 467-481
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Wu G, Vance DE (2010) Choline kinase and its function. Biochem Cell Biol 88: 559-64
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title
Yunus IS, Liu Y-c, Nakamura Y (2016) The importance of SERINE DECARBOXYLASE1 (SDC1) and ethanolamine biosynthesis during
embryogenesis of Arabidopsis thaliana. Plant J 88: 559-69
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Zhang H, Murzello C, Sun Y, Kim M-S, Xie X, Jeter RM, Zak JC, Dowd SE, Paré PW (2010) Choline and osmotic-stress tolerance induced
in Arabidopsis by the soil microbe Bacillus subtilis (GB03). Mol Plant Microbe Interact 23: 1097-104
Pubmed: Author and Title
Google Scholar: Author Only Title Only Author and Title

Downloaded from on December 6, 2018 - Published by www.plantphysiol.org


Copyright © 2018 American Society of Plant Biologists. All rights reserved.