Endogenous isoflavone methylation correlates with the in vitro rooting phases of Spartium junceum L. (Leguminosae)

Accepted Manuscript Title: Endogenous isoflavone methylation correlates with the in vitro rooting phases of Spartium junceum L. (Leguminosae) Author: Francesca Clematis Serena Viglione Margherita Beruto Virginia Lanzotti Paola Dolci Christine Poncet Paolo Curir PII: DOI: Reference:

S0176-1617(14)00085-6 http://dx.doi.org/doi:10.1016/j.jplph.2014.03.013 JPLPH 51921

To appear in: Received date: Revised date: Accepted date:

3-1-2014 6-3-2014 8-3-2014

Please cite this article as: Clematis F, Viglione S, Beruto M, Lanzotti V, Dolci P, Poncet C, Curir P, Endogenous isoflavone methylation correlates with the in vitro rooting phases of Spartium junceum L. (Leguminosae), Journal of Plant Physiology (2014), http://dx.doi.org/10.1016/j.jplph.2014.03.013 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.

1

Endogenous isoflavone methylation correlates with the in vitro rooting phases of

2

Spartium junceum L. (Leguminosae)

3 Francesca Clematisa, Serena Viglioneb, Margherita Berutob, Virginia Lanzottic, Paola Dolcid,

5

Christine Poncete, Paolo Curira*

6

a

7

Ornamental Species, Corso Inglesi I-18038 Sanremo (IM) Italy. Ph.: +39-0184694845; Fax +39-

8

0184-694856; e-mail: [email protected]

9

b

Regional Institute for Floriculture, via Carducci 12 I-18038 Sanremo (IM) Italy

10

c

Department of Food Science, University of Napoli Federico II, via Università 100, Portici, (NA)

11

Italy

12

d

13

da Vinci 44, 10095, Grugliasco, (TO), Italy

14

e

15

06903 Sophia Antipolis Cedex, France

16

*

M

an

us

cr

Consiglio per la Ricerca e la Sperimentazione in Agricoltura - Research Unit for Floriculture and

d

DIVAPRA , Agriculture Microbiology and Food Technology, University of Turin, Via Leonardo

te

Sophia Agrobiotech Institute, UMR INRA, University of Nice, 400 Route des Chappes - BP 167 –

Ac ce p

17

ip t

4

Author for correspondence and reprint requests

18

Abbreviations

19

BA, 6-benzylaminopurine; BSA, bovine serum albumin; CEM, Channel Electron Multiplier; CPE,

20

crude protein extract; DTT, Dithiothreitol; DMSO, Dimethyl sulfoxide; EDTA,

21

Ethylenediaminetetraacetic acid; E.I., Ebullition Interval; MCE, Methanolic Crude Extract; HPLC,

22

High Performance Liquid Chromatography; mAU, Absorbance Unit; MeOH, Methanol; MS,

23

Murashige and Skoog; Rt, retention time; SAM, S-adenosyl-methionine; TMIF, 4’, 5, 7-

24

trimethoxyisoflavone.

25

Page 1 of 35

26 Abstract

28

Spartium junceum L. (Leguminosae) is a perennial shrub, native to the Mediterranean region in

29

southern Europe, widespread in all the Italian regions and, as a leguminous species, it has a high

30

isoflavone content. An in vitro culture protocol was developed for this species starting from stem

31

nodal sections of in vivo plants, and isoflavone components of the in vitro cultured tissues were

32

studied by means of High Performance Liquid Chromatography (HPLC) analytical techniques. Two

33

main isoflavones were detected in the S. junceum tissues during the in vitro propagation phases:

34

Genistein (4’,5,7-Trihydroxyisoflavone), already reported in this species, and its methylated form

35

4’, 5, 7-Trimethoxyisoflavone, detected for the first time in this plant species (0.750 ± 0.02 mg·g-1

36

dry tissue). The presence of both of these compounds in S. junceum tissues was consistently

37

detected during the in vitro multiplication phase. The absence of the methylated form within plant

38

tissues in the early phases of the in vitro adventitious root formation was correlated with its negative

39

effect displayed on root induction and initiation phases, while its presence in the final “root

40

manifestation” phase influenced positively the rooting process. The unmethylated form, although

41

detectable in tissues in the precocious rooting phases, was no longer present in the final rooting

42

phase. Its effect on rooting, however, proved always to be beneficial.

cr

us

an

M

d

te

Ac ce p

43

ip t

27

44

Keywords: 4’, 5, 7-Trimethoxyisoflavone, HPLC, in vitro rooting, isoflavone methylation,

45

Spartium junceum

46 47

Introduction

48

Spartium junceum L., known as Spanish Broom or Weaver’s Broom, is an economically important

49

leguminous shrub, evergreen, perennial, 2 to 4 meters tall, native to the Mediterranean region in

50

southern Europe, southwest Asia and northwest Africa (Oggiano et al., 1997). In addition to its use

51

in perfumery (Miraldi et al., 2004), S. junceum is utilized in Europe for cordage, in badlands Page 2 of 35

recovery and land restoration (Preti et al., 2009). It is a species characterized by an abundant

53

production of secondary metabolites, such as alkaloids, saponins, simple phenols and flavonoids

54

(Yesilada et al., 2000; Proestos et al 2006). These molecules have been studied from a

55

pharmacological point of view (Cerchiara et al., 2012) but they have not been associated in this

56

species to any particular physiological state of the plant. Phenolics, however, in addition to their

57

strictly metabolic roles, may be related to plant specific physiological phases, such as flowering,

58

rooting, juvenility, senescence, fruit production, response to biotic and abiotic stresses (Ozygit et

59

al., 2008) during which they may display quali-quantitative variations (Fernandez-Lorenzo et al.,

60

1999). Adventitious root formation in microcuttings, a fundamental step of the in vitro propagation

61

of plant species, may fall within the above-mentioned physiological states. Rhizogenesis is indeed a

62

very complex process in which, in addition to other physico-chemical factors (Da Rocha Correa et

63

al., 2012), endogenous flavonoids play important roles (De Klerk et al., 2011). In this context,

64

flavonoid quali-quantitative changes could therefore be expected during in vitro rooting of S.

65

junceum microcuttings. During preliminary investigations (unpublished data), we observed

66

important isoflavone variations in S. junceum L. tissues occurring when in vitro-cultured plants

67

were induced to root. Isoflavones, however, have rarely been associated with plant rooting events

68

and this class of flavonoids has been mainly studied in relation to symbiotic partnership

69

establishment (Larose 2002) and as plant defensive molecules active against various pathogens

70

(Treutter et al., 2005).

71

The lack of information concerning a possible involvement of isoflavones in plant rooting processes

72

prompted us to carry out the present work, with the aim of finding a possible correlation between

73

the previously observed isoflavone changes in S. junceum in vitro cultured tissues and the in vitro

74

rooting process.

Ac ce p

te

d

M

an

us

cr

ip t

52

75 76 77 Page 3 of 35

78 79

Material and methods

80 Plant Material

82

Stem and leafy branches 10 months old from S. junceum L., Leguminosae, used for the in vitro

83

culture experiments, were collected from a selected genotype grown in an open air cultivation near

84

Sanremo (IM, Italy). A plant specimen is conserved at the Botanical Garden of University of Turin,

85

Italy (Voucher nr. 4.04, Prof. Silvano Scannerini, Vegetal Biology Dept., University of Turin).

us

cr

ip t

81

an

86 Tissue culture and rooting stages

88

Softwood branches were used as a starting material for tissue culture. Branch pieces, cut in nodal

89

sections 4 cm long, 1 cm diameter and with a pair of axillary buds, were sterilized in a 1.5% free

90

chlorine solution for 10 min, rinsed three times with sterile distilled water, then aseptically

91

inoculated into MS basal medium (Murashige and Skoog, 1962) without growth regulators and left

92

on this medium for one month to induce their adaptation to the in vitro culture conditions. Explants

93

were then transferred onto a multiplication medium, consisting of MS macro and micro elements

94

and vitamins, 0.5 mg·L-1 BA supplemented with 100 mg·L-1 inositol, 30 g·L-1 sucrose, 8.8 g·L-1

95

agar. All media were brought to pH 5.8 before sterilization by autoclaving at 121°C at 1 atm for

96

20 min.

97

Each subculture had a duration of one month and the culture conditions were 20°C temperature and

98

16 h photoperiod. Illumination was provided by Philips 84 white fluorescent lamps, with an

99

irradiance of 22 μmol m-2 s-1 photosynthetically active radiation (PAR). This phase was considered

Ac ce p

te

d

M

87

100

as the multiplication phase.

101

Explants to be rooted (microcuttings 4 cm tall) were transferred onto a R M rooting medium

102

composed as the above mentioned multiplication medium, but without hormones. In particular, no

103

rooting hormones were applied to evaluate the tissue response under natural physiological Page 4 of 35

conditions and in the absence of exogenous stimuli. The rooting stages, based on observations on

105

free hand obtained histological sections, were conventionally divided as follows, according to Uribe

106

et al.(2008): (1) phase of induction, (0 to 7 days of culture), neoformation of parenchymatous

107

cells/vascular elements; (2) phase of initiation, (8 to 18 days of culture), differentiation of

108

meristematic areas within callus tissues; (3) phase of manifestation, (19 to 40 days of culture), well

109

developed primary and secondary roots. The in vitro explants were considered rooted when they

110

showed at least three adventitious roots, each with a length of 2 cm.

Phenol extraction and column chromatography procedures

an

112

us

111

cr

ip t

104

113

20 g fresh tissues from in vitro plants were harvested at different times during both explant

115

multiplication and each of the three rooting phases described above. Tissues were extracted with

116

MeOH:H2O (1:1) in a soxhlet apparatus two times for 1 hour. Extracts were filtered through paper

117

filters (Whatman 2V), put in a separation funnel, brought to pH 5.0 by HCOOH addition and then

118

partitioned in layers with Petroleum ether (E.I., 40 - 60°C); the upper phase, containing pigments,

119

was discarded while the hydro-alcoholic fraction, collected and neutralized with Na 2 CO 3 , was

120

evaporated to dryness through a rotary evaporator (Buchi, Rotovapor L - 200); the residue was then

121

re-dissolved in MeOH and stored as methanolic crude extract (MCE) at -20°C until needed.

122

Column chromatography separation and purification of the investigated compounds were performed

123

according to a previously published protocol (Ferracini et al., 2010).

124

Several mg of an uncommon isoflavone constituent of S. junceum, not commercially available, were

125

obtained from plant extract after repeated column chromatography over silica gel 100 C8 reverse

126

phase column (40-63 μm particle size, Fluka, Germany) with a linear gradient eluting profile

127

according to a previous protocol (Ferracini et al., 2010). After purification and identification, this

128

compound was dried under nitrogen reflux, stored as powder and used as a pure reference molecule

Ac ce p

te

d

M

114

Page 5 of 35

129

in the further analyses. All solvents and reagents from various suppliers were of the highest purity.

130

Water was HPLC grade.

131 Isoflavone methyl transferase activity evaluation

ip t

132 133

The entire procedure was performed at 4°C. Ten g of fresh in vitro tissues, from multiplication,

135

induction, initialization and elongation phases, respectively, were homogenized with a Turrax T 65

136

homogenizer (IKA Werke, Staufe, Germany), at 9.500 rpm, in 0.2 M Tris HCl buffer, pH 7.5,

137

containing 14 mM mercaptoethanol, 5 mM EDTA and 5 g·L-1 insoluble Polyvinylpolypirrolidone

138

(Polyclar). The homogenates were centrifuged for 10 min at 8,000 g and the pellet discarded, while

139

to the supernatant (NH 4 ) 2 SO 4 80% of saturation point was added. The precipitate was separated

140

by centrifugation as above, re-dissolved in 0.2 M Tris HCl buffer, pH 7.5, and desalted overnight in

141

a collodion bag against the same buffer, to yield the crude protein extract (CPE).

142

Two isoflavone substrates were assayed: Genistein (4’,5,7-Trihydroxyisoflavone) and Daidzin

143

(Daidzein-7-O-glucoside), standard pure samples obtained from Sigma (USA). They were dissolved

144

in 12 mM DMSO while 36 mM SAM was prepared in the same solvent.

145

10 mL CPE (with a total protein cncn. of 8 mg · mL-1), was incubated with 0.5 mL of each substrate

146

and 0.5 mL SAM for one hour at 28°C. The reaction was stopped by the addition of 20 mL MeOH

147

and the precipitate was removed by centrifugation at 8,000 g for 2 min; the supernatant was

148

analyzed both through HPLC and column chromatography as described below for a first

149

identification of possible new methylated products. The purified molecules obtained from the

150

enzymatic reaction were submitted to MS analyses for their final identification. Data concerning the

151

isoflavone methyl transferase activity of CPE were obtained according to the protocol of Curir et al.

152

(2003).

Ac ce p

te

d

M

an

us

cr

134

153 154

Isoflavone demethylase activity evaluation Page 6 of 35

155 To cover a wide range of possible plant demethylases (Hagel et al., 2010), two kinds of experiments

157

were set, according to the procedure of Berim et al. (2012). Protein extract was submitted to

158

(NH 4 ) 2 SO 4 80% of saturation point fractionation, as above mentioned. Briefly, in both types of

159

experiments 25 μM substrate were added to 0.5 mL CPE (with a total protein content of 10 mg ·

160

mL-1) in a total volume of 2.0 mL 100 mmolar Tris HCl buffer solution, pH 7.5, and 1 mmolar

161

DTT. Then, 2 mM NADPH and 1 mM dithiocarbamate were added in the first series of

162

experiments, while in a second series of trials 10 mM 2-oxoglutarate, 10 mM sodium ascorbate and

163

250 μM FeSO4 were added. Reactions were allowed to develop for 60 min. Controls were

164

represented by the same solutions as above but without substrates, cofactors or both. The assayed

165

molecules were TMIF, Biochanin A (5,7-Dihydroxy-4’-methoxyisoflavone) and Formononetin (7-

166

Hydroxy-4’-methoxyisoflavone).

167

The possible products formed from enzymatic reactions, and derived from the initial starting

168

substrates, were monitored through HPLC analyses as below described; then, they were separated

169

and purified through column chromatography and submitted to MS analyses to ascertain their

170

respective identity. Data concerning the isoflavone demethylase activity of CPE were obtained

171

according to the protocol of Curir et al. (2003).

173 174

cr

us

an

M

d

te

Ac ce p

172

ip t

156

Protein quantitative determination

175

Total protein in extracts was spectrophotometrically determined according to the Bradford’s method

176

(Bradford et al., 1976), comparing the observed values to those obtained with known concentrations

177

of BSA, plotted to generate a standard, reference curve.

178 179 180

HPLC analysis Page 7 of 35

181 MCE from in vitro material was evaporated to dryness under nitrogen flow, and the residues were

183

redissolved in MeOH and then injected after having been filtered through 1 µm pore size Acrodisc

184

CR PTFE filter (Gelman, St. Louis, USA).

185

The analyses were performed on a Perkin Elmer HPLC system equipped with a Perkin Elmer Ic 200

186

binary pump, 85A UV – Vis detector. PE Nelson NCI900 interface and Turbochrom (version 1.2)

187

computer software were used. Discovery C18 column (250 mm x 4.6 mm i.d.) packed with silica

188

gel (5 μm particle size, 180 Ǻ pore size, 200 m2·g-1 surface area, 3.0 μmole·m-2 calculated bonded

189

phase coverage, end capping) was employed. Analyses were carried out in gradient mode, with a

190

mobile phase consisting of H2O (solvent A) and CH 3 CN (solvent B) with the following elution

191

profile: from 10% B in A to 90% B in A in 50 min, according to a linear gradient. The flow rate was

192

1 mL · min-1. H2O pH was brought to 3.8 with 0.15 N H 3 PO 4 . The injection volume was 20 μl and

193

the eluates were monitored at 250 nm. The identification of an isoflavone already reported in this

194

species (Proestos et al., 2006) was based on the comparison of the respective retention time and UV

195

spectral data with those obtained from pure reference compounds under the same experimental

196

conditions.

198 199

cr

us

an

M

d

te

Ac ce p

197

ip t

182

Nuclear Magnetic Resonance (NMR) and MS (mass spectrum) analyses

200

NMR spectra were obtained with VG Prospec Fisons mass spectrometer (500 MHz, CD3OD, δ ppm

201

and mult. in Hz). FABMS (glycerol matrix, CsI) were measured on a VG Prospec Fisons mass

202

spectrophotometer. 1HNMR and 13CNMR spectra were recorded at 500 and 125 MHz, respectively,

203

on a Bruker AMX-500 spectrometer in pyridine-d5 (Ferracini et al., 2010). Spectral data were

204

assigned by comparison with literature (Miyazawa et al., 2006; Takahashi et al., 2006). Further

205

analyses were performed through a PE-Sciex (Concord, Ontario, Canada) API III triple quadrupole

206

mass spectrometer. The ionspray interface voltage was -4900 V. Negative ion MS over the m/z Page 8 of 35

207

range was 200-800. Selected [M - H]- ions were analyzed by collision-induced dissociation with

208

90% argon, 10% nitrogen. Selected ions detected by CEM (Channel Electron Multiplier).

209 Isoflavone quantitative determination

ip t

210 211

Isoflavones were quantified using the calibration standard method. The calibration curve was

213

established at 5 data points covering the concentration range of analytes according to levels

214

expected in plant samples.

215

For each isoflavone, a stock solution was prepared by dissolving the pure compound in CH3CN at

216

the concentration of 1 mg·mL-1. Calibration solution of 100, 70, 30 and 14.3 μg·mL-1 were prepared

217

by diluting the stock solution with CH3CN. Slope, intercept, correlation coefficients (R2) were

218

calculated on a linear regression model using Excel 2003 Program. Flavonoid content in plant was

219

expressed as mg of compound per gram of fresh tissue calculated according to the formula reported

220

in Fig.1 (Galeotti et al., 2008).

te

d

M

an

us

cr

212

221

Effect of methylated and unmethylated isoflavones on in vitro rooting of S. junceum

223

microcuttings

224

Ac ce p

222

225

Commercial Genistein (4’,5,7-Trihydroxyisoflavone), purchased from Sigma, USA, and its

226

methylated form TMIF (4’, 5, 7-Trimethoxyisoflavone) not commercially available and therefore

227

purified through column chromatography from S. junceum extracts according to the procedure

228

described above, were used in the experiments. These molecules were dissolved in a few drops of

229

absolute EtOH and added, in the needed quantities (10-5, 10-4 and 2·10-4 M, respectively) to the in

230

vitro rooting medium, when still molten after autoclaving, and prepared as described in the Methods

231

section.

Page 9 of 35

232

S. junceum microcuttings, with an average length of 4.0 cm, were transferred onto the isoflavone-

233

supplemented rooting media, according to the scheme reported in Table 1.

234 Effect of methylated and unmethylated isoflavones on root elongation through lettuce seedling

236

bioassay

ip t

235

cr

237

Seeds of Lactuca sativa cv Autan (Syngenta) were allowed to germinate in Petri dishes 150 x 25

239

mm on filter paper disk moistened with distilled water, at 20°C and under continuous light (120 μE

240

m-2 s-1). Three days later, germinated plantlets, with radicles 6 mm long, were transferred under the

241

same conditions as above into Petri dishes containing filter paper dishes moistened with aqueous

242

solutions at various concentrations of the isoflavones to be assayed. One hundred plantlets (10

243

replicates, each of 10 plantlets) formed each group that received the same treatment. Treatments

244

consisted of three different concentrations (10-5, 10-4 and 2·10-4 M, respectively) of both TMFI and

245

Genistein plus an untreated control. After three days, the average root length observed in treated

246

plantlets was compared to that reached by the untreated samples (control), assumed as 100%. Root

247

length scored in treated plantlets was then expressed as a percentage of the length reached by roots

248

of control plantlets.

an

M

d

te

Ac ce p

249

us

238

250

Statistics

251

The statistical analyses were carried out with the IBM SPSS Statistic 20 Softonic program, using

252

the analysis of variance (ANOVA). The difference among means was analyzed using the least

253

significant difference (LSD) according to the Student-Neumann-Keuls test at a probability level of

254

0.05. Percentages were transformed in arcsin before analysis. Correlation coefficients (R2) were

255

calculated on a linear regression model using SPSS 13.0 statistic software.

256 257 Page 10 of 35

258

Results

259 260

S. junceum in vitro multiplication and rooting

ip t

261 To date, few data about S. junceum micropropagation and microcutting rooting have been reported

263

(Morone et al., 2005). We therefore provided an original and reliable multiplication protocol that

264

allowed us to obtain high propagation rates and reproducible results (Fig. 2A in vitro explants

265

during the multiplication phase, Fig. 2B explants after rooting phase).

266

Generally, the presence of cytokinins in the medium was essential to induce shoot proliferation

267

from nodal explants, low concentrations of 6-benzylaminopurine (BA, 0.5 mg·L-1) are

268

recommended to prevent vitrescence and sucker growth of explants. At the end of the experiments,

269

a monthly multiplication rate of about 3.9 was observed for the culture medium used.

270

The in vitro rooting process took about four weeks, with a rooting response around 85% in the

271

absence of hormones. The induced roots were thick, long and strong. In vitro rooted plants were

272

successfully transferred to in vivo growth conditions on a suitable adaptation substrate composed of

273

peat: vermiculite (1:1, w/w), kept under mist in a greenhouse.

275 276

us

an

M

d

te

Ac ce p

274

cr

262

Isoflavone detection and identification in the in vitro cultured tissues

277

Two main isoflavones were detected in the S. junceum tissues during the different phases of

278

micropropagation.

279

Trihydroxyisoflavone). It was recognized through HPLC analyses performed after having run,

280

under the same analytical conditions, the commercial reference compound. A same Rt of 22.68 ±

281

0.02 min for both the unknown sample and pure Genistein was recorded. In addition, the spectral

282

data of this investigated molecule (Table 2), purified and dissolved in MeOH, in the presence of

One,

already

described

in

this

species,

was

Genistein,

(4’,5,7-

Page 11 of 35

suitable diagnostic reagents, were coincident with those obtainable with the pure, commercial

284

Genistein.

285

The second detected isoflavone is an uncommon compound, not yet recognized in this species yet.

286

The NMR spectral data and MS analysis (Table 2), with a FABMS (positive ion): m/z 313 [M+H]+,

287

allowed us to identify this molecule as 4’,5,7-Trimethoxyisoflavone, (Fig. 3). The chemical

288

structure of this compound was identified by comparing its spectral, FABMS data and physical

289

properties with those reported in the literature.

290

HPLC chromatograms reveal the presence of this isoflavone as a peak with a retention time of

291

27.50 ± 0.02 min (Fig. 4).

us

an

292

Isoflavone quali-quantitative variation in micropropagated plant tissues

M

293

cr

ip t

283

294

Constitutive isoflavone fluctuations were observed to occur in S. junceum tissues during different

296

phases of micropropagation. The average concentration of TMIF in the in vitro tissues was about

297

0.750 ± 0.02 mg·g-1 dry matter, during the multiplication and root manifestation stages, with a

298

correlation coefficient (R2) of 0.953. Genistein average content was instead about 1.95 mg·g-1 dry

299

matter during multiplication, induction and initiation phases, with a correlation coefficient (R2) of

300

0.964. Interestingly, while both isoflavones were recognizable during the multiplication phase,

301

Genistein disappeared during the late rooting stage (manifestation), and TMIF was undetectable in

302

the two earlier rooting stages (induction and initiation), but could be found again in the stage of root

303

manifestation (Table 3).

Ac ce p

te

d

295

304 305

Methylating/demethylating activity in the in vitro plant tissues

306 307

Based on the experimental results, the methylating/demethylating activities in the in vitro plant

308

material in the different culture steps can be represented as shown in Table 4. Page 12 of 35

The isoflavone methyl transferase activity in CPE showed a certain level of activity toward

310

Genistein, converted into the 4’-methylated form Biochanin A. The recorded Vmax value of 24· 10-2

311

µmol min-1 mg-1 indicates an average affinity toward Genistein.

312

Some data were likewise obtained about CPE isoflavone demethylase activity towards Biochanin A,

313

during its conversion into Genistein. The relatively low observed Vmax of 6 · 10-2 µmol min-1 mg-1

314

suggests a low enzyme activity towards Biochanin A.

315

At the end of the experiments, performed as described in the Material and Methods section, the

316

initial reaction substrates and their respective possible transformation products, chromatographed

317

(Table 5, Fig. 5) and separated as previously mentioned, were identified by negative MS analyses as

318

shown in Table 6. No reciprocal interconversion between Genistein and TMIF due to

319

methylating/demethylating activities was observed. These two isoflavone metabolites, although

320

structurally closely related, are therefore expected to originate from the activity of a strictly

321

coordinated enzymatic array, whose spatial organization and associated catalytic activities is

322

presumably destroyed during protein extraction procedures.

323

On the other hand, some methylating and demethylating activities have been retained by S. junceum

324

homogenized tissues and were appreciable indeed (Table 4), although they seemed to exclusively

325

affect the 4’ position of the assayed isoflavones. Although TMIF could be conceived to be derived

326

from Genistein through the methylation of all its three hydroxyl groups, in effect Genistein

327

undergoes in vitro methylation only at the level of the hydroxyl group on the B ring of its molecule,

328

giving raise to Biochanin A. On the other hand, Daidzin (Daizein-7-O-glucoside), a Daidzein

329

glycoside, cannot be methylated at all despite the presence of a suitable 4’-OH group in the B ring.

330

At the same time, no demethylation of the 4’-methoxylated Formononetin (7-Hydroxy-4’-

331

methoxyisoflavone) could be obtained with homogenized in vitro tissue extracts, while Biochanin A

332

(5,7-Dihydroxy-4’-methoxyisoflavone) underwent the enzymatic activity, losing the 4’ methoxy

333

group and originating Genistein.

Ac ce p

te

d

M

an

us

cr

ip t

309

Page 13 of 35

334

Biological effects of the methylated and unmethylated constitutive isoflavones on in vitro

335

rooting of S. junceum microcuttings.

336

ip t

337 Taking in consideration the accumulation pattern of the two isoflavones in the S. junceum tissues,

339

which appeared to depend on the different rooting phases (Table 3), a bioassay was conducted to

340

evaluate the effect of both the methylated and the unmethylated isoflavones when exogenously

341

given the in vitro microcuttings facing the different rooting phases (Tables 7 and 8). Genistein

342

showed a promoting activity on rooting percentage independently on the rooting phase the

343

microcuttings entered, although root number per cutting was slightly reduced and root length was

344

shortened in comparison to the control. Its methylated form TMIF lowered rooting percentage and

345

root number per cutting, but stimulated root elongation when administered in the two early rooting

346

phases. The same methylated molecule did not affect rooting percentage and greatly stimulated root

347

elongation when administered in the final rooting phase (manifestation)

us

an

M

d

te

348

cr

338

Biological effects of the methylated and unmethylated constitutive isoflavones on root

350

elongation of lettuce seedlings.

351

Ac ce p

349

352

The results of the assay are presented in Table 9. The unmethylated isoflavone Genistein had a

353

slight positive effect on lettuce root growth that was not significantly different from that displayed

354

by the control. When the methylated form was applied, although no statistically appreciable effects

355

were observed at the lowest dosage, a considerably high stimulatory effect on root growth was

356

recorded when the dosages were increased, with significant differences in comparison to the

357

untreated root samples. These results on lettuce roots show that, even in a species other than S.

358

junceum, root elongation is stimulated in a statistically appreciable way by the methylated

Page 14 of 35

359

isoflavone. This effect seems therefore to be nonspecific, since it took place independently of the

360

assayed plant species.

361

Discussion

ip t

362 Leguminosae isoflavonoids have an ecophysiological importance as defense molecules and signal

364

compounds in the root-rhizobium symbiotic relations (Aoki et al., 2000). Most natural isoflavonoids

365

are O-methylated, at one or more positions (Bisby et al., 1994) and this influences their physico-

366

chemical properties. In particular, a lower methylation degree induces a changed ability of

367

isoflavones to be located in liposomal membrane lipid bi-layer and altered membrane permeability

368

may ensue (Lania – Pietrzak et al., 2005). The methylated/unmethylated isoflavone ratio may

369

therefore regulate some plant physiological processes. In effect, we observed that in the S. junceum

370

manifestation rooting phase, the unmethylated isoflavone Genistein is replaced by its trimethylated

371

form TMIF. In this phase, in vitro isoflavone methylating activity has been observed, although only

372

at the level of the 4’ position of the molecule. The failure to obtain TMIF from Genistein through in

373

vitro experiments with homogenized plant extracts indicates the absence of enzymatic activities

374

responsible for sequential methylations, possibly due to mixing components stored in different cell

375

compartments disrupted by tissue homogenization, with a subsequent enzyme inactivation (Cooke

376

et al., 1980). Therefore, Sequential methylation of Genistein in S. junceum intact tissues cannot be

377

excluded. The enzymatic machinery allowing a channeled sequence of methyl transfers in

378

flavonoids has been on described in several plant species (Cacace et al., 2003; Zhou et al., 2006).

379

Independent of the pathway of formation, however, the highly methylated TMIF, undetectable in

380

the early rooting phases of induction and initiation, re-appears in the root manifestation phase in

381

concomitance with a sudden disappearance of its unmethoxylated form Genistein. Its biosynthesis

382

thus seems to some extent linked to a possible methylation of Genistein, leading to the formation of

383

a highly methylated compound that actively stimulates root elongation, a property observed in

384

polymethylated flavonoids (Yoshioka et al., 2004). Paradoxically, the same beneficial highly

Ac ce p

te

d

M

an

us

cr

363

Page 15 of 35

methylated compound TMIF exerts inhibitory activity on the S. junceum rooting process when

386

supplied to microcuttings in the induction and initiation rooting phases, during which it is

387

constitutively absent within plant tissues. It is interesting to observe that both the unmethylated and

388

the methylated isoflavone form co-exist exclusively during the in vitro multiplication phase,

389

possibly because they are not harmful to tissues that face this phase. On the contrary, when tissues

390

are induced to root, Genistein and TMIF are no longer detectable together, suggesting that a

391

reciprocal conversion may occur in cuttings. The unmethylated form Genistein explicates a positive

392

effect on root induction and initiation phases, corroborating previous analogous data (Osterc et al.,

393

2007), while the methylated form TMIF appears within tissues only when a stimulatory effect on

394

root elongation is needed. The observed occurrence of methylating/demethylating enzymatic

395

activity in tissues seems to indicate that these two forms could be subjected to reciprocal

396

transformations to ensure a constant bulk of vacuole-stored molecules, to be utilized in the different

397

steps of the rooting process.

d Ac ce p

400

Conclusion

te

398 399

M

an

us

cr

ip t

385

401

Data concerning phenolic composition of vegetative tissues of S. junceum are lacking. In addition to

402

describing a successful and easy protocol for the micropropagation of this species, this research

403

indicates that a relationship exists between in vitro rooting phases and isoflavonoid pattern in

404

explant tissue. Our data demonstrate that the ratio between the respective concentration of Genistein

405

(4’,5,7-Trihydroxyisoflavone) and its corresponding highly methylated isoflavone form TMIF (4’,

406

5, 7-Trimethoxyisoflavone) changes during the different phases of the rooting process. The

407

unmethylated isoflavone Genistein shows a promoting effect in the early phases of root

408

induction/formation; the corresponding methylated form TMIF displays its positive effect only on

409

the elongation of the already formed new roots. This indicates that methylating/demethylating

410

activities in plant tissues could be associated with other important physiological processes than Page 16 of 35

those already known of DNA repair, toxin degradation, and metabolism of bioactive metabolites.

412

Although the involvement of polymethoxylated flavonoids in the rooting process has been recently

413

demonstrated (Yoshioka et al., 2004), to our knowledge this is the first report about a close

414

correlation between isoflavone methylation and rooting. Future studies should be aimed to evaluate

415

the possible involvement of methyl group transfer mechanism at the level of flavonoids in further

416

plant physiological processes.

cr

ip t

411

417 Acknowledgements

us

418

an

419

This work was supported by Project n° 67 ALCOTRA 2007-2013, FIORIBIO II. We thank Ms

421

Francesca Mancini for valuable help in performing experiments.

M

420

422 List of references

d

423

426 427 428 429 430 431 432

Aoki T, Akashi T, Ayabe S. Flavonoids of Leguminoseae plants: structure, biological and biosynthesis. J Plant Res 2000; 113: 475-88.

Ac ce p

425

te

424

Berim A, Hyatt DC, Gang DR. A set of regioselective O-methyltransferases gives rise to the complex pattern of methoxylated flavones in sweet basil. J Plant Physiol 2012;160: 1052-69. Bisby FA, Buckingham J, Harborne JB. Phytochemical Dictionary of the Leguminosae, London: Chapman and Hall, 1994.

Bradford MM. Rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976; 72: 248–54.

433

Cacace S, Schroder G, Wehinger E, Strack D, Schimdt J, Schroder J. A flavonol O-

434

methyltransferase from Catharanthus roseus performing two sequential methylations.

435

Phytochem 2003; 62: 127-37.

Page 17 of 35

436

Cerchiara T, Straface SV, Chidichimo G, Belsito EL, Liguori A, Luppi B, F. Bigucci et al. Spartium

437

junceum aromatic water: chemical composition and antitumor activity. J Nat Product Commun

438

2012; 7: 137-40.

ip t

441

Plant Physiol 1980; 66: 119-22.

Curir P, Lanzotti V, Dolci M, Dolci P, Pasini C, Tollin G. Purification and properties of a new S-

442

adenosyl-L-methionine:flavonoid

443

caryophyllus L.). Eur J Biochem 2003; 270: 3422-31.

445

from

carnation

(Dianthus

Da Rocha Correa L, Trolesi J, Mastroberti AA, Mariatch JEA, Fett-Neto AG. Distinct modes of

an

444

4’-O-methyltransferase

cr

440

Cooke RJ, Roberts K, Davies DD. Model of stress-induced protein degradation in Lemna minor.

us

439

adventitious in Arabidopsis thaliana. Plant Biol 2012; 14: 100-09. De Klerk GJ, Guan H, Huisman P, Marinova S. Effects of phenolic compounds on adventitious root

447

formation and oxidative decarboxylation of applied indoleacetic acid in Malus ‘Jork 9’. Plant

448

Growth Regul 2011; 63: 175 – 85.

451 452

d

te

450

Fernandez – Lorenzo JL, Riguero A, Ballester A. Polyphenols as potential markers to differentiate juvenile and mature chestnut shoot cultures. Tree Physiol 1999; 19:461 – 66. Ferracini C, Curir P, Dolci M, Lanzotti V, Alma A. Aesculus pavia foliar saponins: defensive role

Ac ce p

449

M

446

against the leafminer Cameraria ohridella. Pest Manag Sci 2010; 6: 767 – 72.

453

Galeotti F, Barile E, Lanzotti V, Dolci M, Curir P. Quantification of major flavonoids in carnation

454

tissues (Dianthus caryophyllus) as a tool for cultivar discrimination. Z Naturforsch 2008;

455

63c:161– 68.

456 457

Hagel JM, Facchini P. Biochemistry and occurrence of O- demethylation in plant metabolism. Front Physiol 2010;1: 1-7.

458

Lania-Pietrzak B, Hendrich AB, Zugaj J, Michalak K. Metabolic O-demethylation does not alter the

459

influence of isoflavones on the biophysical properties of membranes and MRP1-like protein

460

transport activity. Arch Biochem Biophys 2005; 433: 428-34.

Page 18 of 35

461

Larose G, Chenevert R, Moutoglis P, Gagné S, Piché Y, Vierheilig H. Flavonoid levels in roots of

462

Medicago sativa are modulated by the developmental stage of the symbiosis and the root

463

colonizing arbuscular mycorrhizal fungus. J Plant Physiol 2002; 159: 1329–39.

466 467

ip t

465

Miraldi E, Ferri S, Giorgi G. Identification of volatile constituents from the flower oil of Spartium junceum. J Essent Oil Res 2004; 16: 568–70.

Miyazawa M, Takahashi K, Araki H. Biotransformation of isoflavones by Aspergillus niger as biocatalyst. J Chem Technol Biotechn 2006; 81: 674 – 78.

cr

464

Morone Fortunato I, Ruta C, Tagarelli A. Different species of broom and related propagation

469

protocols [Cytisus scoparius (L.), Spartium junceum L, Genista aspalathoides Lam) ]. In:

470

Italian Horticultural Society, (SOI) National meeting on biodiversity 7, Catania, Italy, 31

471

March – 2 April, 2005.

474 475

an

M

cultures. Physiol Plantarum 1962; 15: 473 – 97.

d

473

Murashige T, Skoog G. A revised medium for rapid growth and bioassay with tobacco tissue

Oggiano N, Angelini LG, Cappelletto P, Pulping and paper properties of some fibre crops. Ind

te

472

us

468

Crops Prod 1997; 7: 59 – 67.

Osterc G, Stefancic M, Solar A, Stampar F. Potential involvement of flavonoids in the rooting

477

response of chestnut hybrid (Castanea crenata x Castanea sativa) clones. Aust J Exp Agr

478

2007; 47: 96-02.

479 480 481 482

Ac ce p

476

Ozyigit I. Phenolic changes during in vitro organogenesis of cotton (Gossypium hirsutum L.) shoot tips. Afr J Biotechnol 2008; 7: 1145-50. Preti F, Giadrossich F. Root reinforcement and slope bioengineering stabilization by Spanish Broom (Spartium junceum L.). Hydrol Earth Syst Sci Discuss 2009; 6: 3993 – 33.

483

Proestos C, Boziaris IS, Nychas GJE, Komaitis M. Analysis of flavonoids and phenolic acids in

484

Greek aromatic plants: investigation of the antioxidant capacity and microbial activity. Food

485

Chem 2006; 95: 664 –71.

Page 19 of 35

488 489 490 491

cutworm (Spodoptera litura). Chem Pharm Bull 2006; 54: 719 – 21. Treutter D. Significance of Flavonoids in Plant Resistance and Enhancement of Their Biosynthesis. Plant Biol 2005; 75: 81–91.

ip t

487

Takahashi K, Araki H, Miyazawa M. Biotrasformation of isoflavones by larvae of the common

Uribe ME, Materan ME, Canal MJ, Rodriguez R. Specific polyamine ratios as indicators of Pinus caribaea microshoot rooting phases. Plant Biosyst 2008; 142: 444-53.

cr

486

Yeşilada E, Tsuchiya K, Takaishi Y, Kawazoe K. Isolation and characterization of free radical

493

scavenging flavonoid glycosides from the flowers of Spartium junceum by activity-guided

494

fractionation. J Ethnopharmacol 2000; 73: 471 – 78.

498

an

M

497

regulators from fruit and leaf of Vitex rotundifolia. Z Naturforsch C 2004; 59c: 509-14. Zhou JM, Gold ND, Martin VJJ, Wollenweber E, Ibrahim RK. Sequencial O-methylation of tricetin by a single gene production in wheat. Biochim Biophys Acta 2006; 1760: 115-24.

d

496

Yoshioka T, Inokuchi T, Fujioka S, Kimura Y. Phenolic compounds and flavonoids as plant growth

499

501

Appendices

Ac ce p

500

te

495

us

492

502

List of all figure captions

503

Fig. 1 Formula used to calculate final quantification of isoflavone [33]

504

Fig. 2 Multiplication (A) and rooting phases during in vitro culture of S. junceum

505

Fig. 3 Chemical structure of 4’,5,7 –Trimethoxyisoflavone isolated from S. junceum

506

Fig. 4 HPLC Chromatogram obtained from extract of in vitro multiplication phase of S. junceum

507

showing isoflavone Rt . 1: Genistein (Rt =22.68 ± 0.02); 2: 4’, 5, 7- TMIF (Rt = 27.50 ± 0.02).

508

Fig. 5 HPLC profile of the substrates used for methylation and demethylation

509

3: Daidzin (Rt= 11.23 ± 0.02); 1: Genistein (Rt =22.68 ± 0.02);

510

± 0.02); 5:Biochanin A (Rt= 30.58 ± 0.02)

4: Formononetin (Rt= 25.65

Page 20 of 35

Table 1

ip t

cr

us an M ed ce pt

514 515 516 517

Table 1. Scheme used to assess the effect of methylated and unmethylated isoflavones on in vitro rooting of S. junceum microcuttings. Rooting phase Microcuttings kept on isoflavone-amended Then transferred evaluated media from: onto isoflavone-free medium day to day on day Induction 1 7 8 a Initiation 8 18 19 b Manifestation 19 40 − a Microcuttings transferred onto the isoflavone-amended media after having been kept 7 days on the isoflavone-free rooting medium. b Microcuttings transferred onto the isoflavone-amended media after having been kept 18 days on the isoflavone-free rooting medium.

Ac

512 513

Page 21 of 35

Table 2

Table 2. 1H NMR and 13C NMR data of TMIF and genistein [500 MHz, CD3OD,  (ppm) and mult. in Hz]. TMIF Position

1

H

Genistein 13

1

C

13

H

C

8.68 (1H, s)

153.2 (CH)

8.65 (1H, s)

153.2 (CH)

3

-

123.5 (C)

-

123.5 (C)

4

-

174.3 (C)

-

180.7 (C)

4a

-

109.3 (C)

-

105.5 (C)

5

-

161.0 (C)

-

161.8 (C)

6

6.33 (1H, bs)

96.1 (CH)

5.94 (1H, bs)

7

-

164.1 (C)

-

8

6.74 (1H, bs)

92.5 (CH)

6.25 (1H, bs)

94.0 (CH)

8a

-

159.2 (C)

-

160.0 (C)

1'a

-

124.8 (C)

2'-6’

7.52 (2H, bd, 7.5)

130.1 (CH)

3'-5’

6.94 (2H, bd, 7.5)

114.2 8 (CH)

6.95 (2H, bd, 7.5)

115.8 (CH)

4'

-

159.8 (C))

-

157.7 (C)

OCH3

3.83 (9H, s)

55.8 (CH3)

98.3 (CH)

an

us

166.4 (C)

125.1 (C)

7.45 (2H, bd, 7.5)

130.5 (CH)

ed

M

-

Ac

ce pt

520

ip t

2

cr

518 519

Page 22 of 35

Table 3

Table 3. Isoflavone presence in the S. junceum plant extracts at different plant physiological stages: in vitro multiplication and rooting phases, respectively. In vitro multiplicationa

Y

N

N

N

Y

ce pt

ed

M

an

us

TMIF Y (4’,5,7Trimethoxyisoflavone) a Y: the isoflavone is present; N: the isoflavone is absent.

Y

Ac

523

Y

ip t

Genistein (4’,5,7Trihydroxyisoflavone)

In vitro rooting phasesa Induction Initiation Manifestation

cr

521 522

Page 23 of 35

Table 4

Table 4. Methylating/demethylating activities observed in micropropagated plant tissues. In vitro Multiplication phasea

b

Assayed Substrate

Expected Product

Y

Y

Y

Meth.

Genistein (4´,5,7Trihydroxyisoflavone)

Biochanin A (5,7-Dihydroxy-4´methoxyisoflavone)

N

N

N

N

Meth.

Genistein (4´,5,7Trihydroxyisoflavone)

TMIF (4’,5,7Trimethoxyisoflavone)

Y

Y

ND

N

Demeth.

Biochanin A (5,7-Dihydroxy-4´methoxyisoflavone)

Genistein (4´,5,7Trihydroxyisoflavone)

N

N

N

N

Demeth.

TMIF (4’,5,7Trimethoxyisoflavone)

Genistein (4´,5,7Trihydroxyisoflavone)

N

N

N

N

Daidzin (Daidzein-7-Oglucoside)

4’-Methoxy-Daidzin

N

ND

N

Formononetin (7-Hydroxy-4´methoxyisoflavone)

Daidzein (4´,7Dihydroxyisoflavone)

N

cr

us

an

M Meth.

Demeth.

ip t

Y

ce pt

a

Tested Activityb

Y: activity is present; N: activity is absent; ND: not determined. Meth: methylating activity; Demeth: demethylating activity.

Ac

526 527 528

In vitro rooting phasesa Induction Initiation Manifestation

ed

524 525

Page 24 of 35

Table 5

Table 5. Structure of isoflavones and their retention time (Rt, min.) on a reverse-phase HPLC column. X1

X2

X3

Rt

Biochanin A

OH

OH

OCH3

30.24 ± 0.02

Daidzin

OH

H

OH

11.18 ± 0.02

Formononetin

OH

H

OCH3

24.65 ± 0.02

Genistein

OH

OH

OH

22.19 ± 0.02

TMIF

OCH3

OCH3

ip t

Compound

cr

529 530

27.50 ± 0.02

Ac

ce pt

ed

M

an

us

531

OCH3

Page 25 of 35

Table 6

Table 6. Negative ions MS identification of the products of in vitro tissue methylating/demethylating activities on the assayed substrates. HPLC Rt (min) [M – H]Major product Corresponding compound ions 252.8

417, 439

Daidzin (Daidzein-7-O-glucoside)

22.68 ± 0.02

269

240, 224, 196, 180, 133

Genistein (4´,5,7-Trihydroxyisoflavone)

25.65 ± 0.02

267

252

Formononetin (7-Hydroxy-4´methoxyisoflavone)

30.58 ± 0.02

283

268, 238, 222, 210

cr

ip t

11.23 ± 0.02

Biochanin A (5,7-Dihydroxy-4´methoxyisoflavone)

us

532 533

Ac

ce pt

ed

M

an

534

Page 26 of 35

Table 7

Table 7. Effects of the two S. junceum constitutive isoflavones, in their methylated and unmethylated form, supplemented in the induction and initiation phases, on the in vitro rooting response of S. junceum microcuttings. Results after 40 days of culture. Supplementation during Supplementation during Induction phase Initiation phase Concentration Rooting Average root Rooting Average root x x (M) percentage number per percentage number per x cutting cuttingx Genistein 10-5 87 by 6.0 ay 87 b 6.3 a -4 (4’,5,710 92 ab 5.3 a 90 ab 6.0 a -4 Trihydroxy 2·10 100 a 6.1 a 98 a 5.9 a isoflavone) 4.4 b 4.0 b 2.3 c

82 b 61 c 28 d

us

84 b 60 c 30 d

3.8 b 4.2 b 2.8 c

Control 85 b 6.0 a 85 b 6.0 a Each treatment in each rooting phase was applied to 20 microcuttings. Each value in a same column followed by a same letter does not significantly differ for P=0.05 according to the StudentNeuman-Keuls method. Percentages were transformed in arcsin before the analysis. y Each number is the average value of 20 observations.

ce pt

ed

M

x

Ac

538 539 540 541

10-5 10-4 2·10-4

an

TMIF (4’,5,7Trimethoxy isoflavone)

cr

ip t

535 536 537

Page 27 of 35

Table 8

86 b 79 c 83 d

105 bc 120 ab 168 a

Control 85 b 100 bc Each treatment was applied to 20 microcuttings. Each value in column followed by a same letter does not significantly differ for P = 0.05 according to the Student-Neuman-Keuls method. Percentages were transformed in arcsin before the analysis. y Each number is the average value of 20 observations.

ce pt

ed

M

an

x

Ac

545 546 547 548

10-5 10-4 2·10-4

us

TMIF (4’,5,7Trimethoxyisoflavone)

ip t

Table 8. Effects of the two S. junceum constitutive isoflavones, in their methylated and unmethylated form, supplemented in the root manifestation phase, on the in vitro rooting response of S. junceum microcuttings. Results after 40 days of culture. Concentration. Rooting percentagex Root growth (M) (% of the control, assumed as 100%) -5 y 10 85 b 88 dy Genistein 10-4 95 ab 96 cd -4 (4’,5,7-Trihydroxyisoflavone) 2·10 100 a 94 cd

cr

542 543 544

Page 28 of 35

Table 9

Table 9. Effects of the two S. junceum constitutive isoflavones, in their methylated and unmethylated form, on root growth of lettuce seedlings. Concentration Root growth (M) (% of the control, assumed as 100%) Genistein (4’,5,7-Trihydroxyisoflavone)

10-5 10-4 2·10-4

95 ± 2 dx 105 ± 1.5 cd 107 ± 2.3 c

ip t

549 550

10-5 103 ± 1.8 cd -4 10 122 ± 2.9 b -4 2·10 175 ± 3.3 a x Each number is the average and SE of 100 observations. Each value in column followed by a same letter does not significantly differ for P = 0.05 according to the Student-Neuman-Keuls method. Percentages have been transformed in arcsin before the statistical analysis.

us

an M ed ce pt Ac

551 552 553

cr

TMIF (4’,5,7-Trimethoxyisoflavone)

Page 29 of 35

Ac

ce

pt

ed

M

an

us

cr

i

Figure 1

Page 30 of 35

Ac ce p

te

d

M

an

us

cr

ip t

Figure 2A

Page 31 of 35

Ac ce p

te

d

M

an

us

cr

ip t

Figure 2B

Page 32 of 35

Ac

ce

pt

ed

M

an

us

cr

i

Figure 3

Page 33 of 35

Ac

ce

pt

ed

M

an

us

cr

i

Figure 4

Page 34 of 35

Ac

ce

pt

ed

M

an

us

cr

i

Figure 5

Page 35 of 35