This is an author version of the contribution - IRIS Uni Torino

This is an author version of the contribution - IRIS Uni Torino

This is an author version of the contribution published on: [Rubiolo, P., Matteodo, M., Bicchi, C., Appendino, G., Gnavi, G., Bertea, C., Maffei, M.] ...

564KB Sizes 0 Downloads 2 Views

Recommend Documents

This is an author version of the contribution - IRIS Uni Torino
income tax-funded grants and state-imposed caps on local property tax rates) arises ... use data on the US Community Dev

Artemisia umbelliformis Lam - IRIS Uni Torino
The alpine wormwoods (Artemisia genipi Weber, A. glacialis L. and A. umbelliformis Lam.), commonly referred to as genepì

This is an author produced version of a paper published in Outlook on
In place of collective and state farms, independent farmers relying on family labour were expected to make food producti

This item is the archived peer-reviewed author-version of - Institutional
calculate estimates of imputed rent and analyse how these benefits are distributed over the population and how they help

This item is the archived peer-reviewed author-version of - Institutional
A priori, viewers may not consider Pottery Barn as a brand .... the Movies” event and cross-checked with those of the

This item is the archived peer-reviewed author-version of - irua
Sep 2, 2016 - 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.

This item is the archived peer-reviewed author-version of - Institutional
example, reported that 51.5% of SNS users always or sometimes accept friend requests from strangers (O'Dea and Campbell,

This item is the archived peer-reviewed author-version of - irua
students in Beijing. Interviews were conducted in various locations, including coffee shops around the Wudaokou area, in

IRIS is the catalog of generally-accepted performance - IRIS (GIIN)
The IRIS catalog compiles standardized metrics that are generally accepted in various social and environmental sectors.

Herder und der Entwurf einer Poetik der Moderne - IRIS Uni Torino
Nach wie vor klassisch der Beitrag von Max Rouché, La philo- sophie de l'histoire ...... bliographie, introd. et notes p

This is an author version of the contribution published on: [Rubiolo, P., Matteodo, M., Bicchi, C., Appendino, G., Gnavi, G., Bertea, C., Maffei, M.] Chemical and Biomolecular Characterization of Artemisia umbelliformis Lam., an Important Ingredient of the Alpine Liqueur "Genepi", JOURNAL OF AGRICULTURAL AND FOOD CHEMISTRY, Volume: 57 Issue: 9 Pages: 3436-3443 Published: MAY 13 2009, DOI: 10.1021/jf803915v, ACS] The definitive version is available at: [http:// pubs.acs.org/doi/abs/10.1021/jf803915v]

1

Chemical and Biomolecular Characterization of Artemisia umbelliformis Lam., an Important

2

Ingredient of the Alpine Liqueur “Genepì”

3 RUBIOLO,†*

MAURA

MATTEODO,†

CARLO

BICCHI,†

4

PATRIZIA

GIOVANNI

5

APPENDINO,‡ GIORGIO GNAVI,§ CINZIA BERTEA§ AND MASSIMO MAFFEI §

6 7

Dipartimento di Scienza e Tecnologia del Farmaco, Università di Torino, Via P. Giuria 9, I-10125

8

Torino, Italy; Dipartimento di Scienze Chimiche, Alimentari, Farmaceutiche e Farmacologiche,

9

Università del Piemonte Orientale, Via Bovio 6, 28100 Novara, Italy; and Unità di Fisiologia

10

Vegetale, Dipartimento di Biologia Vegetale, Università di Torino, Via Quarello 11/A, 10135

11

Torino, Italy.

12 13

Running head: Chemical and Biomolecular Characterization of Artemisia umbelliformis Lam.

14 15 16

*Corresponding author:

17

(Tel:+390116707661; Fax:+390116707687; E-mail: [email protected])

18



Dipartimento di Scienza e Tecnologia del Farmaco, Università di Torino

19



Dipartimento di Scienze Chimiche, Alimentari, Farmaceutiche e Farmacologiche, Università del

20

Piemonte Orientale, Novara

21

§

Dipartimento di Biologia Vegetale, Università di Torino.

22

1

23

ABSTRACT

24

A. umbelliformis Lam., an important Alpine plant used for preparation of flavoured beverages

25

showed a remarkable intraspecific variability, both at genomic and gene products (secondary

26

metabolites) level. The variability of A. umbelliformis Lam. currently cultivated in Piedmont (Italy,

27

Au1) and in Switzerland (Au2) was investigated by combining the chemical analysis of essential oil

28

and sesquiterpene lactones and the molecular characterization of the 5S-rRNA-NTS gene by PCR

29

and PCR-RFLP. Marked differences were observed between the two plants. Au1 essential oil

30

contained α- and β-thujone as main components while Au2 1,8-cineole, borneol and β-pinene. Au1

31

sesquiterpene lactone fractions contained cis-8-eudesmanolide derivatives and Au2 the trans-6-

32

germacranolide costunolide. Specific A. umbelliformis Au1 and Au2 primers were designed on the

33

sequence of the 5S-rRNA gene spacer region. Furthermore, a PCR–restriction fragment length

34

polymorphism (PCR–RFLP) method was applied using RsaI and TaqI restriction enzymes.

35

Chemical and biomolecular data contributed to characterize A. umbeliformis chemotypes.

36 37 38

Keywords: Artemisia umbelliformis Lam.; alpine liqueurs; chemotype discrimination; chemical

39

analysis; biomolecular analysis.

40

2

41

INTRODUCTION

42

Artemisia umbelliformis Lam. is an Alpine species used to prepare “genepì”, a highly praised

43

liqueur characterized by a bitter taste and a peculiar flavour (1). These properties have been traced

44

to the volatile constituents and to the sesquiterpene lactone fraction of the plant, that are

45

characterized by a high contents of α- and β-thujone (2, 3) and by the presence of the cis-

46

eudesmanolide sesquiterpene lactones 1-3, respectively (4-6).

47

Thujone is a natural terpenoid also associated with common wormwood (Artemisia absinthium L.)

48

and Roman wormwood (Artemisia pontica L.), absinthe’s most widely used ingredients (7 and

49

reference cited there in ). There is currently a heated debate on the toxicity of absinthe and thujones

50

(7 and reference cited there in), but the EU legislation have imposed a limit of 35 ppm to the total

51

amount of these compounds in alcoholic beverages (8). To overcome this issue, thujones-free

52

chemotypes of A. umbelliformis have been selected by horticultural techniques (9). Remarkably, an

53

investigation on the sesquiterpene lactone fraction of one of these thujones-free chemotypes showed

54

dramatic differences from the wild plant. Thus, the C-8 cis sesquiterpene lactones typical of A.

55

umbelliformis from Western Alps (1-3) were replaced by the C-6 trans lactones 4-6a,b, while a

56

structurally unique sesterpene lactone (7) was also detected (10) (Figure 1). Chemotypes (or

57

chemical phenotypes) are generally considered the phenotypical expression of a genotype, although

58

different chemotypes may derive from the same genotype. This means that, according to

59

environmental conditions, the same genotype may express different chemical patterns, or, on the

60

other side, that different genotype may respond to the same environmental pressure with the same

61

phenotypic expression. In this context, molecular genetic methods have recently been shown to be

62

very effective in genotypic discrimination. Genetic methods focus on genotype rather than

63

phenotype, and DNA based experiments are now widely used for a rapid identification (and

64

therefore autentication) of medicinal plants. Bertea et al. (11) recently showed that molecular

65

approaches are a powerful tool to distinguish the Acorus calamus diploid β-asarone-free cytotype

66

from the other cytotypes containing it. The same group also used specific Salvia divinorum primers 3

67

designed on the sequence of the 5S-rRNA gene spacer region (12) to develop a Real-Time PCR-

68

based mathematical model to quantify S. divinorum in commercial plant samples or hallucinogenic

69

preparation (13). Given the potential of this approach, it seemed interesting to apply a combination

70

of biomolecular and chemical techniques to characterise the chemotypes of A. umbelliformis

71

currently cultivated in Piedmont (Italy, Au1) and in Switzerland (Au2), complementing the analysis

72

of their essential oil and sesquiterpene lactones with a molecular characterization by PCR and PCR-

73

RFLP of the 5S-rRNA-NTS region of their genome.

74 75

MATERIALS AND METHODS

76

Chemicals - Thujones standard mixture (mixture of α-thujone and β-thujone purity 99.9%) and all

77

other pure reference compounds were from Sigma-Aldrich (St. Louis MO, USA). Sabinene was

78

from Chromadex

79

France); sabinyl ester homologous series were synthesized in the authors’ laboratory (2). HPLC and

80

analytical grade solvents were from Carlo Erba Reagenti, Rodano, Italy. The sesquiterpene lactones

81

1-7 were available from previous studies (4-6, 10)

82

Plant Material - Forty three samples of Artemisia umbelliformis Lam aerial parts from

83

experimental cultivations run in different Alpine valleys [Val Grana (latitude 44°25’N, longitude

84

7°20’E), Valle Stura (44°21’N, 7°26’E), Valle Maira (44°28’N, 7°22’E), Val Chisone (44°57’N,

85

6°52’E)] at an height of least 1300 m a.s.l. Fresh plant material was directly indoor-dried by the

86

farmers under controlled temperature and humidity up to a constant weight, in agreement with the

87

WHO’s guidelines on Good Agricultural and Collection Practices (GACP) for medicinal plants.

88

Voucher specimen representative of the two chemotypes (native, defined Au1 and selected in

89

Switzerland, defined Au2) are deposited at the Dipartimento di Scienza e Tecnologia del Farmaco

90

(n°231 for Au1 and n°232 for Au2). For each chemotype, batches of 1 kg of aerial dried parts (see

91

above) were supplied by the “Associazione Genepì Occitan” (Cuneo - Italy)

TM

(Irvine CA, USA); sabinol was kindly supplied by Robertet SA (Grasse,

4

92

Essential oils and Headspace solid phase-microextraction (HS-SPME) sample preparation -

93

Essential oils (EOs) were prepared according to the method of the European Pharmacopoeia (14).

94

Ten grams of dried aerial parts was suspended in 250 mL of water in a 500 mL flask for 1 h, and

95

then submitted to hydrodistillation in a Clevenger micro-apparatus for 2 h (2). The resulting EO was

96

left to stabilize for 1 h, then recovered with hexane and then analyzed by GC-MS.

97

The SPME device and the three component CAR/PDMS/DVB fused silica fiber (2 cm long, coating

98

volume: 1.000 µm3) were purchased from Supelco (Bellafonte, PA, USA), (15). Before use, the

99

fiber was conditioned as recommended by the manufacturer.

100

Each sample (200 mg of A. umbelliformis dried aerial parts) hermetically sealed in a 2.0 mL vial

101

was introduced in a thermostatic bath at 80 °C for 15 min; the SPME device was inserted in the

102

sealed vial containing the sample, and the CAR/PDMS/DVB fiber exposed to the matrix headspace

103

(30 min). The vial was vibrated for 10 s every 5 min with an electric engraver (Vibro-Graver V74),

104

(Burgess Vibrocrafters Inc., Brayslake, IL). After sampling, the SPME device was immediately

105

inserted into the GC injector and the fiber thermally desorbed. A desorption time of 5 min at 230 °C

106

was used. Before sampling, each fiber was reconditioned for 20 min in the GC injection port at 230

107

°C.

108

Sesquiterpene lactone extraction - One gram of dried aerial parts of both chemotypes were

109

sonicated three times with ethanol 96% (50 mL) for 10 min. The resulting total extract (150 mL)

110

was filtered, and evaporated to dryness under vacuum; the weighed solid residue was dissolved in

111

acetonitrile/water 20/80 at a concentration of 0.1 mg/mL and analyzed by high-performance liquid

112

chromatography-diode array-ultraviolet detection-mass spectrometry (HPLC-DAD-UV-MS).

113

GC and GC-MS Analyses - GC analyses were carried out on a Shimadzu QP2010 system provided

114

with a FID and a MS detector, and the results processed by GC Solution software and GC-MS

115

solution software (2.51version) (Shimadzu Italia, Milano Italy). Capillary GC-FID-MS analyses

116

were carried out on two 25 m, 0.25 mm i.d., 0.25 µm columns from MEGA (Milano – Italy), i.e.

117

Mega5 (95% polydimethyl-siloxane, 5% phenyl) and MegaWax (polyethyleneglycol, PEG20M). 5

118

GC and GC-MS conditions: injection mode: split; split ratio: 1: 20 Temperatures: injector: 230 °C,

119

transfer line: 230 °C; ion source: 200 °C; carrier gas: He flow-rate: 1.0 mL /min in constant flow-

120

mode. MS detector was in electron impact ionization mode (EI) at 70 eV, scan rate was 1111 amu/s

121

and mass range of 35–350 m/z. Temperature program: from 50 °C (1 min) to 220 °C (5 min) at 3 °C

122

min-1.

123

EOs and headspace components were identified by comparison of both their linear retention indices,

124

calculated versus a C8-C25 hydrocarbon mixture, and their mass spectra with those of authentic

125

samples or with data in the literature.

126

Quantitative analysis. Suitable amounts of α + β-thujone commercial standard were diluted with

127

cyclohexane to obtain six concentration levels ranging from 0.5 ng /µL to 6 ng/µL for α-thujone

128

and from 0.04 ng/µL to 0.5 ng/µL for β-thujone. Calibration curves were obtained by analyzing the

129

resulting standard solutions three times by GC-FID using n-nonane as internal standard.

130

HPLC-DAD-UV analysis - HPLC-DAD-UV analyses were carried out on a Shimadzu 2010EV

131

system provided with a PDA detector (Shimadzu, Dusseldorf Germany). A 150 x 4.6 mm i.d., 5

132

μm, Zorbax Stable Bond column (Agilent, Waldbronn Germany) was used. Analysis conditions

133

were: mobile phase: eluent A: 20% acetonitrile/water; eluent B: 100% acetonitrile; mobile phase

134

gradient: from 100% A to 100% B in 25 min. Injection volume: 10 μL, flow rate: 1 mL/min. UV

135

detection wavelengths: 210nm.

136

Quantitative analysis – Suitable amounts of custonolide and umbellifolide were diluted with

137

methanol to obtain concentrations ranging from 0.5, 5, 10, 25, 50 and 100 ng/μL of each marker,

138

respectively. Umbellifolide (3) was adopted as standard for quantitation also for the

139

hydroperoxytelekins 1 and 2, because of the similarity of structures, their trace abundance and

140

chemical instability; the results are expressed as the sum umbellifolide + hydroperoxytelekins. A

141

calibration curve was made by analyzing the resulting standard solutions three times by HPLC-

142

DAD-UV at 210 nm.

6

143

HPLC-MSD analysis - HPLC-MSD analyses were carried out with a single quadrupole Shimadzu

144

2010EV system (Shimadzu, Dusseldorf, Germany) equipped with an orthogonal atmospheric

145

pressure chemical ionization (APCI) and electrospray ionization (ESI) sources. The same column

146

and mobile phase as for HPLC-DAD-UV analysis was used. Flow rate: 0.8 mL/min. MSD

147

conditions: MS-APCI+; temperature: 400 °C; nebulizer’s flow: 2.5mL/min; CDL voltage: 250 °C;

148

Q-Array: MS-ESI+; temp:250 °C; nebulizer’s flow: 1,5mL/min; CDL voltage: 250 °C.

149

MSD analysis conditions were optimized by direct flow injection of pure standards of costunolide

150

(4) and umbellifolide (3) and of a fraction containing a mixture of hydroperoxytelekines

151

Genomic DNA extraction - Plant material employed for the chemical analyses was also used for

152

genomic DNA extraction. . Fifty mg of dried material was frozen in liquid nitrogen and ground to a

153

fine powder with Tissue Lyser (Qiagen, Hilden, Germany). Genomic DNA was extracted from the

154

ground powder by using the Nucleospin Plant Kit (Macherey Nagel, Düren, Germany) following

155

manufacturer’s instruction. The quantity and quality of the DNA were assessed by

156

spectrophotometric analysis by using the Nanodrop ND-1000 (Thermo Scientific, Wilmington, DE,

157

USA) from several samples of the two chemotypes.

158

PCR amplification, subcloning and sequencing - Approximately 20 ng of genomic DNA isolated

159

from powdered leaf material of Au1 and Au2 were used as a template for PCR amplification with

160

forward primer 5S-P1 (5′ - GTGCTTGGGCGAGAGTAGTA-3′) and reverse primer 5S-P2 (5′ -

161

TTAGTGCTGGTATGATCGCA-3′) flanking the NTS of 5S-rRNA gene (11-13, 16). The

162

amplification was carried out in a 50 µL reaction mixture containing 5 µL 10X PCR reaction buffer

163

(Fermentas), 0.2 mM dNTPs, 20 pmol forward and reverse primers and 0.5 U of Taq DNA

164

polymerase (Fermentas, Glen Burnie, MA, USA). The PCR reactions were carried out in a

165

Whatman Biometra T-Gradient Thermalcycler (Whatman Biometra, Goettingen, Germany).

166

Cycling conditions consisted of an initial 4 min at 94 °C, followed by 30 sec denaturing at 94 °C, 1

167

min annealing at 52 °C, and 1 min elongation at 72 °C repeated for 30 cycles and with 5 min final

168

extension at 72 °C. 7

169

One microliter of the amplification reaction was analyzed by capillary gel electrophoresis (CGE)

170

using the Agilent 2100 Bioanalyzer (Agilent Technologies) and the DNA 1000 LabChip Kit

171

(Agilent Technologies) following manufacturer’s instructions. The DNA 1000 LabChip Kit

172

provides sizing and quantitation of dsDNA fragments ranging from 25 to 1000 bp. PCR products

173

were also analyzed by a 2% agarose gel electrophoresis and visualized by ethidium bromide

174

staining under UV. From this gel a band of about 220 bp for Au1 and about 320 bp for Au2 was

175

purified by using the Nucleospin Extract II Kit (Macherey Nagel) and then subcloned into pGEM-T

176

Easy vector (Promega). The ligated products were transformed into the Escherichia coli Subcloning

177

DH5α Efficiency Competent Cells (Invitrogen). Colonies containing DNA inserts of the correct size

178

were picked and grown overnight in 3 mL of Luria–Bertani (LB) liquid medium. The mini-

179

preparation of plasmid DNAs were performed using QIAprep Spin Miniprep Kit (Qiagen),

180

following manufacturer’s instructions. The plasmid DNAs were employed as a template for

181

sequencing. Both strands of DNA were sequenced at least twice and the sequences were aligned by

182

using ClustalX software.

183

PCR amplification using specific primers for Au1 and Au2 - Sequences derived from powdered

184

leaf material of Au1 and Au2 were aligned in a unique sequence that allowed the design of two

185

forward primers AuF 5′ -CTAGGATGGGTGACCTCCTG -3′(which is common to both

186

chemotypes) and Au2F 5′-GCGGTGACAGAGTCGGTAAA-3′ and two reverse specific primers:

187

Au2R1

188

TCCTTTCTCATTGCCTATTTTTC -3′, which corresponded respectively to nucleotides 21-40,

189

168-187, 212-231, 253-275 of Au2 non-transcribed spacer (NTS) sequence. The internal primers

190

were used for amplification also in combination with primer 5S-P1 and 5S-P2.

191

The conditions of the PCR reactions were the same as mentioned above. One microliter of the

192

amplification products were separated by CGE with the Agilent 2100 Bioanalyzer (Agilent

193

Technologies) and DNA 1000 LabChip Kit (Agilent Technologies) following manufacturer’s

194

instructions.

5′

-CGTAAAATTCACCGCCTACG

-3′

and

Au2R2

5′-

8

195

PCR–RFLP - The purified PCR products of the 5S-rRNA gene spacer region of both Au1 and Au2

196

chemotypes were either digested with 10 U of RsaI (Amersham Biosciences) at 37 °C for 1h or, in a

197

separate reaction, with 10 U of TaqI (Sigma) at 65 °C for 1 h. One microliter of both digestion

198

reactions was fractionated by CGE using the Agilent 2100 Bioanalyzer (Agilent Technologies) and

199

DNA 1000 LabChip Kit (Agilent Technologies) following manufacturer’s instructions.

200 201

RESULTS AND DISCUSSION

202

This study aims to characterize the two chemotypes of A. umbelliformis Lam. under investigation

203

by combining results from chemical and genomic analyses of 43 samples from experimental

204

cultivation (Au1, Italian native and Au2, selected in Switzerland).

205

Chemical analyses - Chemical analyses investigated the fractions responsible for plant odor and

206

taste i.e. the composition of the volatile fraction including quantitation of α- and β-thujone, and that

207

of the sesquiterpene lactone fraction (i.e. the components responsible for the liqueur bitter taste).

208

The volatile fraction was studied by analyzing both the essential oil (EO) obtained by

209

hydrodistillation and the headspace (HS) sampled by solid phase microextraction (HS-SPME)

210

combined with GC and GC-MS. HS-SPME sampling was applied in view of developing a fully

211

automatic control method, to be run in combination with GC and GC-profile multivariate analysis

212

(Principal Component Analysis (PCA). Table 1 reports the average percent areas normalized vs. n-

213

nonane as internal standard and percent range of the characteristic components of the EOs of the

214

samples investigated together with their Linear Retention Indices (LRI) on both GC columns. Table

215

2 reports calibration curves, mean and range amounts of α- and β-thujone in the samples

216

investigated. This Table considers only ten samples of chemotype Au2 on seventeen because the

217

alpha and beta-thujone peak areas of the remaining seven samples were too low to be correctly used

218

for quantitative determination. From these results it is clear that the two chemotypes are

219

characterized by a different composition: Au1 was found to contain α- and β-thujone and an

220

homologous series of sabinyl esters as main components, whereas in the chemotype Au2, 1,89

221

cineole, borneol and β-pinene were the major compounds. Moreover, thujones were almost absent

222

from the Au2 chemotype, their total amount accounting from 0.2 to 0.4g/100g of EO, while in the

223

Au1 chemotype thujones ranged from 18 to about 58 g/100g of EO. The results obtained by HS-

224

SPME-GC analysis, although not directly comparable, were in full agreement with those of the

225

EOs, as shown by the PCA scatterplot of Figure 2. Each chemotype is clearly discriminated and the

226

samples belonging to the same chemotype analysed by HS-SPME-GC and through their EO are

227

coherently positioned in the PCA scatterplot (Figure 2).

228

Significant differences can also be found in the composition of the non-volatile bitter fraction.

229

Figure 3 reports the HPLC-DAD-UV profiles of two samples belonging to Au1 and Au2

230

chemotypes, respectively. The bitter taste of the native Au1 chemotype is mainly due to

231

sesquiterpene lactones of the cis-8-eudesmanolide type [5-deoxy-5-hydroperoxy-5-epitelekin (1), 5-

232

deoxy-5-hydroperoxytelekin (2), umbellifolide (3) (4-6)]. On the other hand, the Au2 chemotype is

233

characterized by high amounts of costunolide (4), a germacranolide typical of A. genipi Weber (4),

234

and by the presence of an unusual sesterpene lactone, named genepolide (7) (10). An in depth

235

investigation of the Au2 ethanolic extract composition after fractionation by column

236

chromatography in combination with NMR, and analysis by HPLC-UV and HPLC-single-

237

quadrupole-MS revealed that costunolide (4) was accompanied by a series of related oxygenated

238

sesquiterpene lactones (artemorine (5b), santamarine (6a),and reynosine(6b)) (4). On the other

239

hand, in the Au1 chemotype the presence of both telekine hydroperoxides and umbellifolide was

240

confirmed.

241

Costunolide and the sum of hydroperoxytelekins and umbellifolide were adopted as markers of the

242

two chemotypes to evaluate quantitatively the bitter fraction of the 43 samples under investigation.

243

These analyses showed an average amount of umbellifolide + hydroperoxytelekins expressed as

244

umbellifolide (y = 29.9392x -1.6795; R2: 0.99999) of 0.11 g/100g of dried plant material in a range

245

varying between 0.03 and 0.37 g/100g for the Au1 chemotype, and an average amount of

246

costunolide (y = 101.1709x-31.0592; R2 : 0.99975) of 0.56 g/100g of dried plant material in a range 10

247

between 0.20 and 0.93 g/100g for the Au2 chemotype. Costunolide was also detected in very low

248

amounts in some samples of the Au1 chemotype, its percentage never exceeding 0.05 %

249

Molecular characterization of the two A. umbelliformis chemotypes - In higher eukaryotes, the

250

5S-rRNA gene occurs in tandemly repeated units consisting of an 120 bp coding region separated

251

by a non-transcribed spacer varies from species to species (16). Thus the diversity of the spacer

252

region can be used as an identification basis (17).

253

Here, two primers flanking the spacer region of 5S-rRNA, already successfully employed for

254

differing Acorus calamus chemotypes (16), A. calamus cytotypes (11) and Salvia divinorum both as

255

pure plants (12) and in plant mixtures (13) were used in PCR analysis of genomic DNA isolated

256

from the two chemotypes Au1 and Au2.

257

A single fragment of approximately 220 bp was produced by Au1 (Figure 4, lane 1) and a single

258

fragment of about 320 bp was produced by Au2 (Figure 4, lane 2). Fragments derived from both

259

chemotypes were ligated into pGEM-T Easy vector and the nucleotide sequence was determined.

260

The sequenced region spans 224 bp for Au1 (NCBI GenBank Accession No. EU816950) and 327

261

bp for Au2 (NCBI GenBank Accession No. EU816951).

262

Sequence alignment of the 5S-rRNA spacer region flanked by the 3′-and 5′-ends of the coding

263

region is shown in Figure 5. Surprisingly, Au1 presented a difference of 103 nucleotides with

264

respect to Au2. This difference is quite consistent but not uncommon between chemotypes or

265

cytotypes, as it has been previously demonstrated with other plant species (11, 16).

266

In order to characterize better the two chemotypes and to simplify the identification method,

267

nucleotide sequences of the 5S-rRNA gene spacer region were used to design four specific primers

268

(Figure 6).

269

PCR products derived from all possible combinations of Au1 and Au2 specific primers also used

270

with the primers designed on the coding regions of the plant 5S-rRNA gene, were analyzed. In the

271

chemotype Au1 a single fragment of 204 bp in length was amplified using the primer AuF in

272

combination with the primer 5S-P2 (Figure 4, lane 3). The same strategy used with the chemotype 11

273

Au2 produced a single fragment of 307 bp in length (Figure 4, lane 4). The three additional specific

274

primers designed for Au2 gave a combination of single fragments as follows: 231 bp with 5S-P1

275

and Au2R1 (Figure 4, lane 5), 275 bp with 5S-P1 and Au2R2 (Figure 4, lane 6), 64 bp with Au2F1

276

and Au2R1 (Figure 4, lane 7), 108 bp with Au2F1 and Au2R1 (Figure 4, lane 8), 213 bp with AuF

277

and Au2R1 (Figure 4, lane 9) and 255 bp with AuF and Au2R2 (Figure 4, lane 10). All

278

amplifications occurred only in Au2 when the Au2 specific primers were used and no PCR products

279

were detected when Au1 DNA was employed as a template.

280

In addition, a PCR–RFLP method was applied. From the identified sequences, two RsaI sites could

281

be found in both chemotype 5S-rRNA spacer regions at 18 bp and 141 bp position in Au1 and at 18

282

and 301 bp in Au2 (Figure 5). Purified PCR products obtained by using 5S-P1 and 5S-P2 primers

283

were digested with Rsa I. As expected, PCR products from the chemotype Au1 could be digested by

284

RsaI giving two major fragments of 123 bp and 82 bp respectively, and a minor fragment of 19 bp

285

not visible in the gel because out of the resolution capacity of the instrument (Figure 4, lane 11).

286

When purified PCR products from the chemotype Au2 were digested using RsaI, a completely

287

different RFLP profile was observed. RsaI cleaved Au2 5S-rRNA spacer region giving a major

288

fragment of 283 bp and two minor fragments of 25 bp (barely visible in the gel) and 19 bp (not

289

visible in the gel) (Figure 4, lane 12). A TaqI site was also identified in the sequence of the

290

chemotype Au2 (see Figure 5). As expected, PCR products from the chemotype Au1 could not be

291

digested by Taq I (Figure 4, lane 13), whereas purified PCR products from the chemotype Au2.

292

Thus, our biomolecular characterization provides a useful tool for the unequivocal characterization

293

of the two chemotypes. If some intermediate chemotypes due to meiotic rearrangements were

294

already present in the population, they would be detected by using universal primers (different

295

fragment size or different nucleotide composition), but this was not the case, since all samples gave

296

the same results. Besides, this marker has been successfully used for Acorus calamus chemotype

297

determination (16) and it represents a powerful tool to deduce genetic relationship of medicinal

298

plants, especially at the intra-specific level. 12

299

In conclusion, these results clearly support the view that A. umbelliformis, a valuable plant for

300

Alpine agriculture, shows a remarkable intraspecific variabilty, both at the genomic and gene

301

products (secondary metabolites) levels. This multidisciplinary study, by showing remarkable

302

chemical variation in the terpenoid profile, and consistent genomic difference in the 5S-rRNA

303

spacer regions, has identified two chemotypes of A. umbelliformis. A multidisciplinary approach

304

based on the combination of metabolome- and genome-derived product analysis enabled the

305

unequivocal chemical and biomolecular fingerprinting of these two chemotypes. Combined “omics”

306

approaches are becoming a useful tool not only for basic science but also for industrial plant

307

characterization. Owing to the commercial relevance of A. umbelliformis and to the regulatory

308

issues related to the presence of thujones, the identification of a RsaI and Taq sites can be used for a

309

rapid and precise chemotype identification of the plant chemotypes, complementing the chemical

310

analysis of the essential oil and the sesquiterpene lactones.

311 312

LITERATURE CITED

313 314

1. Mucciarelli, M.; Maffei, M. Introduction to the genus. In Medicinal and Aromatic Plants -

315

Industrial Profiles: Artemisia, Wright, C. W., Ed.; Taylor&Francis: London, 2002; pp 1-50.

316

2. Bicchi, C.; Nano, G. M.; Frattini, C. On the composition of the essential oils of Artemisia

317

genipi Weber and Artemisia umbelliformis Lam. Z. Lebensm.-Unters.-Forsch. 1982, 175, 182-

318

185.

319 320

3. Bicchi, C.; D'Amato, A.; Nano, G. M.; Frattini, C. Capillary GLC controls of some alpine Artemisiae and related liqueurs. Chromatographia 1985, 18, 560-566.

321

4. Appendino, G.; Belliardo, G. M.; Nano, G. M.; Stefenelli, S. Sesquiterpene lactones from

322

Artemisia genepi Weber: isolation and determination in plant material and in liqueurs. J. Agr.

323

Food Chem. 1982, 30, 518-521. 13

324 325

5. Appendino, G.; Gariboldi, P.; Nano, G. M. Isomeric hydroperoxy eudesmanolides from Artemisia umbelliformis. Phytochemistry 1983, 22, 2767-2772.

326

6. Appendino, G.; Gariboldi, P.; Calleri, M.; Chiari, G.; Viterbo, D. The structure and

327

conformation of umbellifolide, a 4,5-secoeudesmane derivative. J. Chem. Soc. Perkin Trans. I

328

1983, (1), 2705-2709.

329

7. Lachenmeier, D. W.; Nathan-Maister, D.; Breaux, T. A.; Sohnius, E. M.; Schoeberl, K.;

330

Kuballa, T. Chemical composition of vintage preban absinthe with special reference to

331

thujone, fenchone, pinocamphone, methanol, copper, and antimony concentrations. J. Agr.

332

Food Chem. 2008, 56 (9), 3073-3081.

333

8. Council Directive (EEC) No 88/388 on the approximation of the laws of the Member States

334

relating to flavourings for use in foodstuffs and to source materials for their production. Off. J.

335

Europ. Comm. 1988, L184, 61-66.

336 337 338

9. Rey, C.; Slacanin, I. Domestication du genépi blanc. Revue Suisse Vitic, Arboric, Hortic. 1997, 39, 1-7.

339

10. Appendino, G.; Taglialatela-Scafati, O.; Romano, A.; Pollastro, F.; Avonto, C.; Rubiolo, P.

340

Genepolide, a sesterpene γ−lactone with a novel carbon skeleton from mountain wormwood

341

(Artemisia umbelliformis). J. Nat. Prod. Publication Date on Web: November 20, 2008

342

11. Bertea, C. M.; Azzolin, C. M. M.; Bossi, S.; Doglia, G.; Maffei, M. E. Identification of an

343

EcoRI restriction site for a rapid and precise determination of beta-asarone-free Acorus

344

calamus cytotypes. Phytochemistry 2005, 66 (5), 507-514.

345

12. Bertea, C. M.; Luciano, P.; Bossi, S.; Leoni, F.; Baiocchi, C.; Medana, C.; Azzolin, C. M.;

346

Temporale, G.; Lombardozzi, M. A.; Maffei, M. E. PCR and PCR-RFLP of the 5S-rRNA-

14

347

NTS region and salvinorin A analyses for the rapid and unequivocal determination of Salvia

348

divinorum. Phytochemistry 2006, 67 (4), 371-378.

349

13. Luciano, P.; Bertea, C. M.; Temporale, G.; Maffei, M. E. DNA internal standard for the

350

quantitative determination of hallucinogenic plants in plant mixtures. Forensic Sci. Int.:

351

Genetics 2007, 1, 262-266.

352

14.

353

European Directorate for the Quality of Medicines (EDQM) European Pharmacopoeia 6th Edition; 2008, 251-252.

354

15. Rubiolo, P.; Belliardo, F.; Cordero, C.; Liberto, E.; Sgorbini, B.; Bicchi, C. Headspace-solid-

355

phase microextraction fast GC in combination with principal component analysis as a tool to

356

classify different chemotypes of chamomile flower-heads (Matricaria recutita L.).

357

Phytochem. Anal. 2006, 17 (4), 217-225.

358

16. Sugimoto, N.; Kiuchi, F.; Mikage, M.; Mori, M.; Mizukami, H.; Tsuda, Y. DNA profiling of

359

Acorus calamus chemotypes differing in essential oil composition. Biol. Pharm. Bull. 1999,

360

22 (5), 481-485.

361

17. Cai, Z. H.; Li, P.; Dong, T. T. X.; Tsim, K. W. K. Molecular diversity of 5S-rRNA spacer

362 363 364

domain in Fritillaria species revealed by PCR analysis. Planta Med. 1999, 65 (4), 360-364. 18.

Adams RP. In Identification of Essential oil Components by Gas Chromatography/Mass Spectrometry 4th edn.; Allured: Carol Stream IL, 2007.

365 366

Note: P. Rubiolo, M. Matteodo and C. Bicchi are indebted with Regione Piemonte (Italy) for the

367

financial support to this study carried out within the project “Progetto genepì - Sviluppo di tecniche

368

innovative a supporto della coltivazione e della trasformazione del genepì in Piemonte”. C .M.

15

369

Bertea and M.E. Maffei are indebted with Centre of Excellence CEBIOVEM of the University of

370

Turin.

371

16

372

FIGURES CAPTIONS

373 374

Figure 1 - Structures of the sesquiterpene lactones identified in both chemotypes of A.umbelliformis

375

Lam. 1: 5-desoxy-5hydroperoxy-5-epitelekin; 2: 5-desoxy-5-hydroperoxytelekin;

3:

376

umbellifolide; 4: costunolide; 5a: verlotorin; 5b: artemorine; 6a: santamarine; 6b: reynosine; 7:

377

genepolide

378 379

Figure 2 - PCA scatterplot of the cumulative elaboration of both EOs and headspaces sampled by

380

HS-SPME of both Artemisia umbelliformis chemotypes. Capital letters: EO GC analysis, lower-

381

case letters: HS-SPME GC analysis.

382

Figure 3 - HPLC-DAD-UV profiles at 210 nm of Artemisia umbelliformis chemotypes Au1 and

383

Au2.

384 385

Figure 4 - PCR products derived from all possible combinations of Au1 and Au2 specific primers

386

designed on the coding and non-transcribing regions of the plant 5S-rRNA gene. Lane 1, a

387

single fragment of approximately 224 bp is produced by Au1. Lane 2, a single fragment of

388

about 327 bp is produced by Au2. Lane 3, a single fragment of 204 bp in length amplified using

389

the primer AuF in combination with the primer 5S-P2 in the chemotype Au1. Lane 4, a single

390

fragment of 307 bp in length using the primer AuF in combination with the primer 5S-P2 in the

391

chemotype Au2. Lane 5, a single fragments of 231 bp with primers 5S-P1 and Au2R1 in the

392

chemotype Au2. Lane 6, a single fragments of 275 bp with primers 5S-P1 and Au2R2 in the

393

chemotype Au2. Lane 7, a single fragments of 64 bp with primers Au2F1 and Au2R1 in the

394

chemotype Au2. Lane 8, a single fragments of 108 bp with primers Au2F1 and Au2R1 in the

395

chemotype Au2. Lane 9, a single fragments of 213 bp with primers AuF and Au2R1 in the

396

chemotype Au2. Lane 10, a single fragments of and 255 bp with primers AuF and Au2R2 in the

397

chemotype Au2. Lane 11, purified PCR products obtained by using 5S-P1 and 5S-P2 primers 17

398

digested with Rsa I give two major fragments of 123 bp and 82 bp respectively, and a minor

399

fragment of 19 bp not visible in the gel because out of the resolution capacity of the instrument

400

in the chemotype Au1. Lane 12, RsaI cleaved Au2 5S-rRNA spacer region gives a major

401

fragment of 283 bp and two minor fragments of 25 bp (barely visible in the gel) and 19 bp (not

402

visible in the gel). Lane 13, PCR products from the chemotype Au1 not digested by Taq I. Lane

403

14, purified PCR products from the chemotype Au2 digested using Taq I producing two

404

fragments of 197 and 130 bp.

405 406

Figure 5 - Aligments of the nucleotide sequences of 5S-rRNA gene spacer region of Artemisia

407

umbelliformis chemotypes Au1 and Au2. Universal primer sequences are indicated in squared

408

solid boxes. Artemisia umbelliformis forward primers are indicated in bold. Identical sequences

409

are indicated by (*). Gaps (-) are introduced for the best alignment. RsaI site is evidenced in the

410

squared large dot box, whereas the TaqI site is evidenced in the squared small dots box.

411

Forward and reverse specific primers of the Au2 chemotype are indicated.

412 413

Figure 6 - Position of the universal primers (5S-P1 and 5S-P2) flanking the spacer region of 5S-

414

rRNA gene and specific Artemisia umbelliformis forward (AuF) and A. umbelliformis

415

chemotype Au2 forward (Au2F1) and reverse (AuR1 and AuR2) specific primers used for PCR

416

amplification of the 5S-rRNA spacer region.

18

Table 1 - Components Characterising the Essential Oils of Artemisia umbelliformis Chemotypes Au1 and Au2.

Au1 a

b

Au2

Compounds

MEGA5

MEGA5

CW

Mean

Range

Mean

Range

α-Thujene α-Pinene Camphene Sabinene β-Pinene β-Myrcene α-Terpinene p-Cymene 1,8-Cineole γ-Terpinene cis-Sabinene hydrate α-Terpinolene trans-Sabinene hydrate α-Thujone β-Thujone Sabinol Camphor Borneol Terpinen-4-ol α-Terpineol Bornyl acetate α-Terpinyl acetate Sabinyl isobutirrate β-Caryophyllene Sabinyl isovalerianate Sabinyl valerianate Caryophyllene oxide Neryl isovalerianate

930 936 954 975 979 991 1017 1025 1031 1060 1070 1089 1098 1102 1114 1142 1143 1169 1177 1189 1289 1349 1416c 1419 1503c 1516c 1583 1584

929 936 951 975 978 993 1017 1026 1033 1061 1069 1089 1097 1108 1118 1141 1145 1168 1177 1190 1286 1351 1416 1418 1503 1516 1581 1585

1038 1031 953 1132 1117 1174 1189 1280 1213 1257 1268 1292 1488 1420 1446 1708 1516 1712 1619 1714 1378 1494 1494 1594 1577 1605 1965 1679

0.05 0.26 0.06 0.35 1.74 0.18 0.18 0.27 4.86 0.35 1.17 0.10 0.65 36.99 9.11 0.90 0.15 1.14 1.21 0.29 0.05 0.58 1.35 0.40 3.94 2.69 1.89 4.11

tr-0.10 tr-0.40 tr-0.30 0.09-1.40 0.10-2.30 tr-0.30 tr-0.30 0.10-0.40 0.30-9.50 0.10-0.50 0.30-1.50 tr-0.10 0.30-1.00 29.70-51.90 4.60-19.53 0.28-4.10 tr-1.54 tr-3.90 0.40-1.48 tr-0.50 tr-0.10 tr-1.10 1.00-3.83 tr-0.56 3.10-7.13 1.30-7.28 0.57-3.30 1.08-6.90

0.13 0.60 0.43 0.15 5.12 0.68 0.34 0.21 7.58 0.75 7.16 0.21 2.19 0.37 0.27 0.20 0.75 11.58 4.75 1.08 0.51 2.30 0.01 2.24 0.16 0.14 3.61 4.72

tr-0.57 0.33-1.20 tr-1.10 tr-0.36 1.73-11.50 0.10-2.90 0.10-0.55 0.10-0.70 4.05-14.00 0.43-1.00 2.70-20.08 0.10-0.30 1.30-3.35 tr-2.00 tr-0.65 0.10-0.50 tr-1.70 0.09-19.43 3.30-8.15 0.40-2.20 tr-1.97 0.70-4.19 tr-0.09 1.00-4.60 tr-0.50 tr-0.30 2.15-5.40 1.40-10.00

a: LRI from Adams library (18)y; b: experimental LRI c: LRI from standards synthesized in the authors’ laboratory (2) Mean and range values are expressed as percent areas normalized vs. nonane as internal standard.

19

Table 2 – Mean and Range Amount of α- + β- thujone Expressed as g/100g of Essential Oil in the Investigated Samples of Artemisia umbelliformis Chemotypes Au1 and Au2.

n

Range

Mean

SD

Chemotype AU1

26

18.0-57.3

36.8

14.7

Chemotype AU2

10

0.2-0.4

0.3

0.1

α-thujone (Y= 0.3105x + 3.311, R2 = 0.9995) β-thujone (Y = 0.2896x + 0.1380 ; R2 = 0.9994)

20

Figure 1

8

8 O

O

8 O

O

O OOH

OOH

1

2

O

O

O 3 OH

OR

6

3

6 6

4

O

O O

R 5a OH 5b H

4

O

15

O O

6a ∆ 6b ∆4(15) 3

6 O O

7

21

Figure 2

6

B B BB

4

p.c. 2 - var. sp. 0.13063

B

B

2 A 0

-2

B

B

A A AA AAA A

AA A A A AaA a aa aa aa Aaa a a a a a a a a a

A A

A

B A

A a

B Bb

B b B

b B BB B b

a

b

b

b b b b b

b bb

b

-4

b -6 -6

-4

-2

0 2 p.c. 1 - var. sp. 0.5273

4

b

6

8

22

Figure 3

uV 700000

Au1 chemotype

650000 600000 550000 500000 450000 400000 350000 300000

Telekines 250000 200000

Umbellifolide

150000 100000 50000 0 2.5

5.0

7.5

10.0

12.5

15.0

17.5

20.0

22.5

25.0

27.5

min

uV 700000

Au2 chemotype

650000

Costunolide

600000 550000 500000 450000 400000 350000 300000 250000

Artemorine

Santamarine

200000 150000 100000 50000 0

2.5

5.0

7.5

10.0

12.5

15.0

17.5

20.0

22.5

25.0

27.5

min

23

Figure 4

24

Figure 5

25

Figure 6

26