Citrus latifolia

Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES

Article

Targeted Analysis of the Concentration Changes of Phenolic Compounds in Persian Lime (Citrus latifolia) During Fruit Growth Carlos Augusto Ledesma-Escobar, Feliciano Priego-Capote, Victor Jose Robles-Olvera, and María Dolores Luque De Castro J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.7b05535 • Publication Date (Web): 05 Feb 2018 Downloaded from http://pubs.acs.org on February 6, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Agricultural and Food Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28

Journal of Agricultural and Food Chemistry

Targeted Analysis of the Concentration Changes of Phenolic Compounds in Persian Lime (Citrus latifolia) During Fruit Growth Carlos A. Ledesma-Escobar1,2,4, Feliciano Priego-Capote1,2,3*, Víctor J. Robles Olvera4 María D. Luque de Castro1,2,3

1

Department of Analytical Chemistry, Annex C-3, Campus of Rabanales, University of

Córdoba, E-14071, Córdoba, Spain. 2

University of Córdoba Agrifood Campus of International Excellence ceiA3, Campus of

Rabanales, E-14071, Córdoba, Spain. 3

Maimónides Institute of Biomedical Research (IMIBIC), Reina Sofía Hospital, University

of Córdoba, E-14014, Córdoba, Spain. 4

Tecnológico Nacional de México – Instituto Tecnológico de Veracruz. Unidad de

Investigación y Desarrollo en Alimentos, Av. Miguel Ángel de Quevedo 2779, Veracruz, Ver. 91797, México.

*Corresponding author: Feliciano Priego-Capote ([email protected]). Phone and fax number: +34957218615.

1 ACS Paragon Plus Environment


Page 2 of 28

1

ABSTRACT: Citrus possess a high content of phenolic compounds; however, few studies

2

have focused on the changes occurring during fruit growth. The changes in the concentration

3

of 20 flavonoids, 4 phenolic acids and their biosynthetic precursors phenylalanine and

4

tyrosine have been evaluated during fruit maturation (14 weeks). Extracts from all samples,

5

obtained by ultrasound assistance, were analyzed by liquid chromatography coupled to

6

tandem mass spectrometry with a triple quad system (LC–QqQ MS/MS). In general, the

7

concentration of flavanones —which represented over 70% of the studied phenols— and

8

flavones increased during fruit growth, reaching their maximum concentration around week

9

12. In general, flavanols and phenolic acids exhibited their maximum concentration at week

10

5, then decreasing significantly during the rest of maturation. Phenylalanine and tyrosine

11

showed a sinuous behavior during fruit growth. Partial least squares showed a clear

12

differentiation among fruits belonging to different maturation stages, being coumaric acid

13

derivatives the most influential variables on the projection.

14 15

KEYWORDS: Citrus fruits maturation, Persian lime, flavonoid composition, targeted

16

analysis, Citrus pathway, mass spectrometry.

17 18


Page 3 of 28


19

INTRODUCTION

20

Phenols, compounds widely distributed in Citrus, have received great attention because of

21

their nutritional and bioactive properties1 associated to a reduction of the risk for different

22

types of cancer and cardiovascular diseases, and to antioxidant, anti-inflammatory and

23

radical-scavenging activities.2 These properties have promoted both the development of

24

methods for their extraction from Citrus and the study of their bioactive properties.3 Few

25

studies have focused on estimating the changes of given metabolites from Citrus during fruit

26

growth. In this context, both sugars (°Brix) and titratable acidity (% of citric acid) have been

27

the most studied, since a relationship between them is widely used as ripening index.4 The

28

analysis of these metabolites in mandarins revealed that sugars increase during ripening,

29

while the titratable acidity decreases.5 Additionally, the analysis of carotenoids in different

30

Citrus varieties during maturation (based on the fruit color) revealed that the concentration

31

of total carotenoids increases with ripening.6 Concerning phenolic compounds, Moulehi et

32

al. analyzed in 2012 the differences in both total phenolic and flavonoids content in oranges

33

and mandarins considering three maturation degrees: immature (green); semimature

34

(yellow); and commercial mature (orange).7 The maximum concentration of both total

35

phenols and flavonoids in oranges was observed in commercial mature fruits, while the

36

maximum concentration of total phenols and flavonoids in mandarins was observed in

37

immature and semimature samples, respectively. Untargeted metabolomics analysis was

38

used by the authors of the present research to differentiate and characterize Persian lime

39

samples from different ripening stages, as a function of the growth time (from week 1 to

40

week 14) as maturity estimator.8 The results of this study, based on 423 molecular entities

41

common to all samples, showed that the phenolic compounds (i.e. hesperetin, naringenin or



Page 4 of 28

42

apigenin derivatives) have a great influence on the discrimination of samples from different

43

growth weeks. The general biosynthetic pathway of phenols begins with the conversion of

44

either phenylalanine or tyrosine into p-coumaric acid and subsequent reactions can transform

45

this phenolic acid into either flavonoids or other phenolic acids.9 Flavonoids are the most

46

abundant phenolic compounds in Citrus, being flavanones, which actually are precursors of

47

both flavones and flavanols, the most abundant flavonoids subclass.10 The results obtained

48

in our previous study and the lack of reports on the changes in the phenolic composition

49

during Citrus growth have encouraged the targeted analysis of these classes of compounds.

50

The present research was aimed at establishing similarities/dissimilarities among 26 phenolic

51

compounds from Persian lime (Citrus latifolia) sampled at eight ripening stages (from weeks

52

1 to 14 of maturation) by using liquid chromatography–tandem mass spectrometry with a

53

triple quadrupole system (LC–QqQ MS/MS). Selection of the target metabolites was based

54

on: (i) identification we reported previously;8 (ii) availability of commercial standards; and

55

(iii) relevance of phenolic compounds in the metabolic pathway. Therefore, the behavior of

56

20 flavonoids (6 flavanones, 12 flavones and 2 flavanols), 4 phenolic acids (including p-

57

coumaric acid) and both phenylalanine and tyrosine were studied.

58 59

MATERIALS AND METHODS

60

Sampling Methodology. Persian lime (Citrus latifolia) samples at different growth stages

61

(1, 3, 5, 7, 9, 12, and 14 weeks (named as samples W01-to-W14) were collected in Martínez

62

de la Torre (Veracruz, México, geographical coordinates: 20°04′00″N 97°03′00″O) from

63

September 2012 to January 2013. The experimental field had 500 mature trees (at least 4

64

years old) uniformly distributed over an area of 1250 m2 (40 m length and 31 m wide). The 4 ACS Paragon Plus Environment

Page 5 of 28


65

field was monitored weekly since the beginning of flowering and the branches that showed

66

newborn fruits were labelled at the beginning of the fruit growth. Twelve samples were

67

collected from each growth stage; those corresponding to week 14 (full mature fruits) were

68

harvested in duplicate and a half of them were stored at 25 °C for 2 weeks to study the short-

69

time storage behavior of mature fruits. The sampling method was based on random

70

rectangular coordinates within the limits of the experimental field. A total of 96 samples were

71

selected by proximity to the corresponding random coordinate. All samples were manually

72

harvested; then immediately stored under liquid nitrogen to reduce possible enzymatic

73

reactions and transported to the laboratory. The whole frozen fruits were individually grinded

74

with a mortar and pestle under liquid nitrogen and lyophilized until constant weight. Once

75

dehydrated, the samples were ground (particle diameter ≤ 0.5 mm) and stored in the dark at

76

–20 °C until use.

77

Reagents. All solvents were HPLC grade. Ethanol and formic acid were from Scharlab

78

(Barcelona, Spain), and acetonitrile was from Fluka (Buches, Switzerland). Deionized water

79

(18 MΩ•cm) from a Millipore Milli-Q water purification system (Bedford, MA, USA) was

80

used to prepare the mobile chromatographic phases and extractant mixtures. Analytical

81

standards apigenin-7-glucoside, caffeic acid, diosmetin, diosmin, eriocitrin, ferulic acid,

82

hesperetin, hesperidin, homoorientin, hydroxytyrosol, luteolin, luteolin-7-glucoside,

83

naringenin, narirutin, neodiosmin, neohesperidina, orientin, phenylalanine, quercetin,

84

tyrosine, vitexin and vitexin-2-O-rhamnoside) were purchased from Sigma–Aldrich (Saint

85

Louis, MO, USA), and o-coumaric acid, p-coumaric acid, rhoifolin, rutin and tangeretin

86

from Extrasynthese (Genay France). Hydroxytyrosol (5 µg mL–1) was used as external

87

standard.



Page 6 of 28

88

Metabolites Extraction. Persian lime samples (1 g dry weight each) were extracted with

89

the help of a Branson (Branson, Germany) 450 digital sonifier (20 kHz, 450 W) equipped

90

with a cylindrical titanium-alloy probe (12.70 mm diameter). Twenty mL of 60.3% ethanol

91

in water was used as extractant, and 5 min were required for extraction with ultrasound

92

assistance (70% amplitude and 0.8 s s–1 duty cycle). The extraction method was previously

93

developed by the authors using a desirability model to maximize the concentration of

94

phenolic compounds in the extracts from lemon.11

95

Phenols Quantitation by LC–QqQ MS/MS. Chromatographic separation of the extract

96

components was performed by using an Agilent 1200 series LC with an Inertsil ODS-2 C18

97

analytical column (250 × 4.6 mm i.d. 5 µm particle) from GL Science (Tokyo, Japan). The

98

chromatograph was coupled to a 6460 Triple Quad detector equipped with a Jet Stream

99

Technology electrospray ion (ESI) source from Agilent Technologies (Palo Alto, USA). The

100

injection volume was 10 µL, and the mobile phases were 0.1% of formic acid in deionized

101

water (phase A) and in acetonitrile (phase B) at a constant flow rate of 1 mL min–1. The

102

gradient was as follows: from initial 4% to 10% B in 5 min; change from 10% to 25% B in

103

30 min; from 25% to 100% B in 15 min and constant 100% B for 5 min. After analysis, the

104

column was equilibrated to the initial conditions for 5 min. The ESI source operated in both

105

positive and negative ionization modes under the following conditions: nebulizer gas at 45

106

psi, sheath gas flow and temperature were set at 8 mL min–1 and 350 °C, respectively. The

107

temperature of the gas nebulizer and the capillary voltage were set at 350 °C and 3500 V,

108

respectively. The SRM parameters for all analytes are listed in Supplementary Table 1.

109

Calibration curves for each analyzed metabolite were developed by linear regression of the

110

peak area obtained from 10 different concentrations of the corresponding analytical standard


Page 7 of 28


111

injected in triplicate. The mixed standard solution was diluted with methanol to yield a series

112

of standard solutions at appropriate concentrations to construct the calibration curves. The

113

limits of detection (LOD) and quantitation (LOQ) for each metabolite were calculated as the

114

concentration providing signals three and ten times, respectively, higher than the background

115

noise measured at a time close to each chromatographic signal. The precision of the method

116

was evaluated by calculation of both the within-day variability and between-days variability

117

expressed as relative standard deviation. For this purpose, a mixed standard solution was

118

analyzed in duplicate each day, for 7 days.

119

Statistical Analysis. Univariate analysis, ANOVA and Tukey (p≤0.01) tests were

120

performed by Statgraphics Centurion XVII (Statpoint Technologies Inc.) Multivariate

121

analysis, ANOVA, Tukey HSD (p≤0.01) and partial least squares for discriminant analysis

122

(PLS-DA)

123

(http://www.metaboanalyst.ca/).

124

(http://www.genome.jp/kegg/pathway.html) was consulted to support discussion on the

125

metabolic pathways.

were

processed

by

the

Additionally,

MetaboAnalyst KEGG

3.0

software

PATHWAY

Database

126 127

RESULTS

128

Absolute quantitation requires the use of analytical standards, which limits the total number

129

of quantified compounds as a function of their commercial availability. In this study, the

130

targeted analysis of 20 flavonoids (6 flavanones, 12 flavones and 2 flavanols), 4 phenolic

131

acids and 2 amino acids in the Persian lime (Citrus latifolia) samples was carried out by LC–

132

MS/MS. The method was properly validated in terms of analytical features such as precision



Page 8 of 28

133

and sensitivity. LODs and LOQs, listed in Supplementary Table 2, were in the range of pg

134

mL–1, which are considerably low for this application. Concerning the precision of the

135

method, the within-day variability and between-days variability, expressed as relative

136

standard deviation, for all metabolites was less than 5.4% and 7%, respectively (see

137

Supplementary Table 2, while Supplementary Figure 1 shows examples of both total ion

138

and single reaction monitoring chromatograms of the main phenols at week 14).

139 140

Changes in the Concentration of Flavanones during Persian Lime growth. The results

141

showed that flavanones were the most abundant phenolic compounds among the studied

142

metabolites. Depending on the growth stage, this subclass constituted between 69.5 and

143

80.5% of the total content of the quantified phenolic compounds. The predominance of

144

flavanones as major phenols in Citrus was previously revised by González-Molina et al. in

145

2010.12 The maximum concentration of flavanones was reached in this case at week 12, being

146

hesperidin (2005±254 µg g–1), narirutin (1207±315 µg g–1) and eriocitrin (1171±392 µg g–1)

147

the most concentrated —69% or higher with respect to the total content of quantified phenols.

148

The ANOVA and Tukey HSD (p≤0.01) tests revealed significant differences in the

149

concentration of flavanones among the growth stages that increased their concentration

150

gradually from the beginning to fruit maturity and reached their maximum concentration

151

around week 12. In general, the highest differences were observed by comparison between

152

samples W03 and W12. As an example, hesperidin, eriocitrin and narirutin yielded their

153

maximum fold changes (3.4, 31.6 and 88.0, respectively) for the interval week 3–week 12.

154

Figure 1 shows the individual behavior of all quantified flavanones during Persian lime

155

growth, while the changes in concentration (mean, standard deviation and significance) of all 8 ACS Paragon Plus Environment

Page 9 of 28


156

studied flavanones are summarized in Table 1. Chromatograms of the single reaction

157

monitoring of the main flavanones at the different maturation stages can be seen in

158

Supplementary Figure 2.

159

Changes in the Concentration of Flavones During Persian Lime Growth. Flavones

160

were the second most abundant subclass of the studied phenolic compounds, constituting

161

between 7.8 and 17.7% of the total concentration of the measured phenols. Like flavanones,

162

the studied flavones underwent significant changes (p≤0.01) in their concentrations during

163

the studied weeks and showed a gradual increase through the growth time, reaching

164

maximum concentration around week 12. In this study, the behavior of 4 apigenin

165

derivatives, 4 luteolin derivatives, 3 diosmetin derivatives and tangeretin was analyzed.

166

Diosmin (366±105 µg g–1), rhoifolin (285±36 µg g–1) and vitexin (237±80 µg g–1) were the

167

most abundant at week 12; while orientin (12.0±3.7 µg g–1) was, among luteolin derivatives,

168

the most concentrated at week 12. The highest fold changes in the concentration of diosmin

169

(13.0), rhoifolin (54.6) and vitexin (41.2) occurred between weeks 3 and 12. A plot of the

170

changes in concentration of all studied flavones during fruit growth can be seen in Figure 2,

171

while all the results (mean, standard deviation and significance) for this subclass of

172

flavonoids are summarized in Table 2. Chromatograms of the single reaction monitoring of

173

the main flavones at the different maturation stages can be seen in Supplementary Figure

174

3.

175

Changes in the Concentration of Flavonols During Persian Lime Growth. Concerning

176

flavonols, both rutin (quercetin-3-O-rutinoside) and quercetin were analyzed. The results of

177

ANOVA and Tukey HSD tests (p≤0.01) showed significant changes of these metabolites

178

during fruit growth and revealed that the highest concentrations corresponded to the first 5 9 ACS Paragon Plus Environment


Page 10 of 28

179

weeks. Regarding rutin its maximum concentration corresponded to week 1 (334±59 µg g–

180

1

181

flavonols decreased at the end of the growth process. As can be seen in Table 3, the

182

concentration of rutin was much higher than that of its aglycone quercetin at all studied

183

weeks, as also happens for all aglycones and their corresponding glucosides. The changes in

184

concentration of both rutin and quercetin during Persian lime growth can be seen in Figure

185

3, while single reaction monitoring chromatograms of the main flavanols at the different

186

maturation stages can be seen as Supplementary Figure 4.

), while that of quercetin was higher at week 3 (0.64±0.30 µg g–1). The concentration of both

187

Changes in the Concentration of Phenolic Acids and Amino Acids During Persian

188

Lime Growth. Concerning phenolic acids, the most abundant among those under study was

189

p-coumaric acid. The concentration of this acid increased significantly from week 1 (9.8±2.1

190

µg g–1) to week 5 (30.4±12.3 µg g–1), then decreasing drastically up to week 7 (1.55±0.87 µg

191

g–1), and remaining without significant changes in subsequent weeks (see Supplementary

192

Figure 4). On the contrary, the concentration of o-coumaric acid decreased significantly from

193

week 1 (2.21±0.42 µg g–1) to week 5 (0.03±0.01 µg g–1), being its concentration below its

194

quantitation limit from week 7. The concentration of both ferulic and caffeic acids showed

195

an alternating behavior during fruit growth, in such a way that their concentration was below

196

their detection limits at weeks 3, 7 and 9. The average concentration of ferulic acid did not

197

show significant differences at the weeks at which it was quantified; while the maximum

198

concentration of caffeic acid was 1.41±0.44 µg g–1 at week 14.

199

The quantified phenolic amino acids (phenylalanine and tyrosine) showed their maximun

200

concentration in the postharvest stage (W16) when the concentration of most of the studied

201

phenols decreased significantly. Considering only the growth weeks (1-to-14), a sinuous 10 ACS Paragon Plus Environment

Page 11 of 28


202

behavior was observed, being weeks 1 and 5 those with maxima concentration of them, while

203

the lowest concentrations were found in samples from week 7 to 12 (see Supplementary

204

Figure 4). Generally, the concentration of these amino acids was in inverse proportion to the

205

rest of studied phenols, mainly to flavanones.

206 207

The heat map in Figure 3 shows the behavior of both phenolic acids and amino acids during fruit growth, while all the set of quantitation results can be consulted in Table 3.

208

Study of the Changes in Phenolic Composition of Persian Lime During Fruit Growth

209

by Targeted Analysis. A supervised analysis by PLS-DA was used to discriminate samples

210

from different growth stages. The data matrix (26×96) was constituted by the concentration

211

of all 26 quantified metabolites in the 96 evaluated samples. The scores plot (Figure 4A)

212

shows a clear discrimination among samples and revealed that the distance among scores in

213

the plot is larger when the growth time between samples is longer, thus indicating greater

214

differences between them. The two-dimensional PLS-DA explained 60.5% (Component1 =

215

45.4% and Component2 = 15.1%) of the total variability. Also, the plot of variables

216

importance in projection (Figure 4B), which summarizes the individual influence of all

217

variables on discrimination, reveals that the metabolites with higher influence on samples

218

differentiation were p-cumaric and o-cumaric acids (followed by rhoifolin, apigenin-7-

219

glucoside and narirutin), probably due to the dramatic decrease in their concentration from

220

week 5 to the end of the maturation process.

221 222

DISCUSSION



Page 12 of 28

223

A brief review to the metabolic pathways of phenols in Citrus before analyzing the changes

224

in composition of these compounds during Persian lime growth is shown in Figure 5. Our

225

results have revealed that most of the studied phenols are present from the beginning of fruit

226

growth and their changes along the growth process can be helpful to show their bioregulation

227

mechanisms. According to these metabolic pathways, phenylalanine is converted into

228

cinnamic acid by phenylalanine ammonia lyase; and subsequent reactions catalyzed by either

229

cinnamate 4-hydroxylase or cinnamate 2-hydroxylase can convert cinnamic acid into p-

230

coumaric or o-coumaric acids, respectively. Also, tyrosine can be converted into p-coumaric

231

acid by tyrosine ammonia lyase. Therefore, our results suggest a bioregulation of cinnamate

232

2-hydroxylase during the first 5 weeks to prioritize conversion of phenylalanine into p-

233

coumaric acid, which is the intermediate for the production of most phenols.13 This behavior

234

has been confirmed by the analysis of conversion ratios (Supplementary Figure 5A).

235

Additionally, the conversion ratios of both phenylalanine and tyrosine into p-coumaric acid

236

suggest that, despite the differences in concentration of both amino acids, selectivity

237

regulation to produce p-coumaric acid from either phenylalanine or tyrosine did not exist

238

(Supplementary Figure 4A).

239

In the case of Citrus fruits, p-coumaric acid can follow two pathways: (i) transformation

240

into caffeic acid by p-coumarate 3-hydroxylase (then, this acid can be transformed into

241

ferulic acid by caffeic acid 3-O-methyltransferase); (ii) conversion into p-coumaroyl-CoA by

242

cinnamate 4-hydroxylase. Subsequent reactions of the p-coumaroyl-CoA with malonyl-CoA,

243

catalyzed by both chalcone synthase and chalcone isomerase, result in the formation of

244

flavonoids. Both acids (caffeic and ferulic) were not detected at weeks 7 and 9 —which

245

coincided with a significant increase in the concentration of flavanones and flavones.


Page 13 of 28


246

Therefore, it can be assumed that the activity of p-coumarate 3-hydroxylase could be blocked

247

during these growth stages thus increasing the production of major flavonoids; then, the

248

enzyme would be reactivated during the final weeks to reach the maxima concentration of

249

flavonoids.

250

Regarding flavonoids, flavanones such as hesperetin, naringenin or eriodictyol derivatives

251

can be converted into their respective flavones by dehydrogenation catalyzed by flavone

252

synthase or into flavonols by a two-step synthesis which starts with hydroxylation caused by

253

a flavonone hydroxylase, followed by dehydrogenation catalyzed by a flavonol synthase.9,14

254

Thus, naringenin is the first flavanone produced and acts as a wildcard from which both

255

hesperetin, and eriodictyol derivatives can be produced. Also eriodictyol can be produced

256

from the caffeoyl-CoA, thus reinforcing the theory on the reduction of caffeic acid.15

257

Subsequent reactions can convert hesperetin derivatives into diosmetin derivatives, and

258

eriodictyol into either luteolin or quercetin derivatives; also, naringenin can be converted into

259

apigenin derivatives. According to the obtained results, it is evident that a proportional

260

conversion of flavanones into flavones occurs from weeks 7 to 12, caused by flavone

261

synthase; which also suggests a decrease in the activity of flavonol synthase, which can

262

explain the decreased concentration of both rutin and quercetin from week 5 to the final

263

growth stage. Finally, the results obtained in the last growing weeks suggest that the activity

264

of cinnamate 4-hydroxylase, responsible for the production of flavonoids, decreases from

265

weeks 12 to 14, allowing the upturn of phenolic acids at the end of the maturation process.

266

Additionally, the analysis of conversion ratios of narirutin (naringenin 7-O-rutinoside) into

267

either hesperidin (hesperetin 7-O-rutinoside) or eriocitrin (eriodictyol 7-O-rutinoside),



Page 14 of 28

268

revealed differences in the conversion between them; thus suggesting the existence of a

269

selective mechanism for the production of hesperidin (see Supplementary Figure 5B).

270

It is known that flavonoids are naturally synthesized by plants as response to physical

271

damages, infections, stress or UV light; therefore, their composition may vary depending on

272

environmental changes.16 The increase in fruit size make it more susceptible of suffering

273

damages; therefore, a higher concentration of flavonoids at the final growth stage could be

274

interpreted as an increase of fruit defense mechanisms.

275

From the point of view of a potential industrial extraction of flavonoids —and taking into

276

account that the extraction yield is higher when the fruit has been previously dehydrated17—

277

the total dry mass of whole limes should be considered to calculate the amount of extracted

278

phenols. In general, the maximum amount of extracted flavonoids would be obtained using

279

fruits either 12 or 14 growth weeks; however, it is remarkable that the water content in limes

280

at week 12 was between 10% and 15% lower than at week 14. This indicates that the

281

collection of the limes at week 12 would require less energy for dehydration without

282

significant reduction of flavonoids yield. Concerning total phenolic acids, their maxima

283

extraction yield was obtained at week 14. To clarify this point, Supplementary Table 3

284

shows the average yield of each metabolite under study, considering the average dry mass of

285

the whole fruits at each growth stage.

286

In short, targeted analysis of phenolic compounds constitutes a useful tool for

287

classification of samples among Citrus varieties,18 and even among process variables such as

288

different sample pretreatments or extraction methods.19 Similarly, the presented PSL–DA

289

reveals that the concentration of phenols can be used to discriminate different growth stages

290

of Persian lime. 14 ACS Paragon Plus Environment

Page 15 of 28


291 292

Abbreviations Used

293

LC–QqQ MS/MS, liquid chromatography coupled to tandem mass spectrometry with a triple

294

quad system; ESI, electrospray ion source; LOD, limit of detection; LOQ, limit of

295

quantitation; W+number, number of growth weeks; ANOVA, analysis of variance; PLS-DA,

296

partial least square-discriminat analysis; PAL, phenylalanine ammonia-lyase; TAL, tyrosine

297

ammonia-lyase; C4H, cinnamate 4-hydroxylase; C2H, cinnamate 2-hydroxylase; pC3H, p-

298

coumarate 3-hydroxylase; Ca3M, caffeic acid 3-O-methyltransferase; CHS, chalcone

299

synthase; CHI, chalcone isomerase; FNS, flavone synthase; F3M, flavonoid 3'-

300

monooxygenase; FNH, flavonoid 3',5'-hydroxylase; N3D, naringenin 3-dioxygenase; FLS,

301

flavonol synthase.

302 303

Funding

304

The Ministerio de Economía y Competitividad, Junta de Andalucía and FEDER program

305

through projects CTQ-2015-68813R and FQM-1602.

306 307

Acknowledgments

308

The Ministerio de Economía y Competitividad, Junta de Andalucía and FEDER program are

309

thanked for the financial support. C.A.L.E thanks the Mexican National Council for Science

310

and Technology (CONACYT, CVU-252846) for grants to support his research in both

311

countries. L. Landa Armenta is thanked for providing the experimental field used in this

312

study. 15 ACS Paragon Plus Environment


Page 16 of 28

313 314

Supporting Information description

315

Supplementary Table 1. LC–QqQ MS/MS parameters for quantitative determination of

316

phenolic compounds in extracts from Persian lime.

317

Supplementary Table 2. Analytical features of the method.

318

Supplementary Table 3. Average concentration of phenolic compounds in whole dry

319

Persian lime samples at different growth weeks. Values with different letter for the same

320

compound (same row) are significantly different among ripening weeks (Tukey HSD,

321

p≤0.01). W+number indicates the number of growth weeks.

322

Supplementary Figure 1. Conversion ratios of phenylalanine (Phe) and tyrosine (Tyr) into

323

p-coumaric acid (p-CoumAc) and phenylalanin into o-coumaric acid (o-CoumAc) (A); and

324

narirutin into either hesperidin or eriocitrin (B).

325 326

References

327

(1)

328 329

Cano, A.; Medina, A.; Bermejo, A. Bioactive compounds in different citrus varieties. Discrimination among cultivars. J. Food Compos. Anal. 2008, 21 (5), 377–381.

(2)

Kris-Etherton, P. M.; Hecker, K. D.; Bonanome, A.; Coval, S. M.; Binkoski, A. E.;

330

Hilpert, K. F.; Griel, A. E.; Etherton, T. D. Bioactive compounds in foods: their role

331

in the prevention of cardiovascular disease and cancer. Am. J. Med. 2002, 113 Suppl

332

(1), 71–88.

333

(3)

Ledesma-Escobar, C. A.; Luque de Castro, M. D. Towards a comprehensive 16 ACS Paragon Plus Environment

Page 17 of 28


334

exploitation of citrus. Trends in Food Science and Technology. September 2014, 63–

335

75.

336

(4)

337 338

Kader, A. A. Fruit maturity, ripening, and quality relationships. Acta Hortic. 1999, 485 (485), 203–208.

(5)

Lin, Q.; Wang, C.; Dong, W.; Jiang, Q.; Wang, D.; Li, S.; Chen, M.; Liu, C.; Sun, C.;

339

Chen, K. Transcriptome and metabolome analyses of sugar and organic acid

340

metabolism in Ponkan (Citrus reticulata) fruit during fruit maturation. Gene 2015, 554

341

(1), 64–74.

342

(6)

Yoo, K.-M.; Moon, B. Comparative carotenoid compositions during maturation and

343

their antioxidative capacities of three citrus varieties. Food Chem. 2016, 196, 544–

344

549.

345

(7)

Moulehi, I.; Bourgou, S.; Ourghemmi, I.; Tounsi, M. S. Variety and ripening impact

346

on phenolic composition and antioxidant activity of mandarin (Citrus reticulate

347

Blanco) and bitter orange (Citrus aurantium L.) seeds extracts. Ind. Crops Prod. 2012,

348

39, 74–80.

349

(8)

Ledesma-Escobar, C. A.; Priego-Capote, F.; Robles-Olvera, V. J.; Luque de Castro,

350

M. D. Changes in the composition of the polar fraction of Persian lime ( Citrus latifolia

351

) during fruit growth by LC – QTOF MS / MS analysis. Food Chem. 2017, 234, 262–

352

268.

353

(9)

354 355

Martens, S.; Mithöfer, A. Flavones and flavone synthases. Phytochemistry 2005, 66 (20), 2399–2407.

(10)

Benavente-García, O.; Castillo, J. Update on uses and properties of citrus flavonoids: 17 ACS Paragon Plus Environment


Page 18 of 28

356

new findings in anticancer, cardiovascular, and anti-inflammatory activity. J. Agric.

357

Food Chem. 2008, 56 (15), 6185–6205.

358

(11)

Ledesma-Escobar, C. A.; Priego-Capote, F.; Luque de Castro, M. D. Comparative

359

study of the effect of auxiliary energies on the extraction of Citrus fruit components.

360

Talanta 2015, 144, 522–528.

361

(12)

González-Molina, E.; Domínguez-Perles, R.; Moreno, D. a; García-Viguera, C.

362

Natural bioactive compounds of Citrus limon for food and health. J. Pharm. Biomed.

363

Anal. 2010, 51 (2), 327–345.

364

(13)

365 366

metabolic engineering. Curr. Opin. Biotechnol. 2014, 26 (Figure 2), 71–78. (14)

367 368

Lin, Y.; Jain, R.; Yan, Y. Microbial production of antioxidant food ingredients via

Forkmann, G.; Martens, S. Metabolic engineering and applications of flavonoids. Curr. Opin. Biotechnol. 2001, 12 (2), 155–160.

(15)

Ueyama, Y.; Suzuki, K.; Fukuchi-Mizutani, M.; Fukui, Y.; Miyazaki, K.; Ohkawa, H.;

369

Kusumi, T.; Tanaka, Y. Molecular and biochemical characterization of torenia

370

flavonoid 3′-hydroxylase and flavone synthase II and modification of flower color by

371

modulating the expression of these genes. Plant Sci. 2002, 163 (2), 253–263.

372

(16)

373 374

Treutter, D. Significance of flavonoids in plant resistance: a review. Environ. Chem. Lett. 2006, 4 (3), 147–157.

(17)

Ledesma-Escobar, C. A.; Priego-Capote, F.; Luque De Castro, M. D. Effect of sample

375

pretreatment on the extraction of lemon (Citrus limon) components. Talanta 2016,

376

153, 386–391.


Page 19 of 28

377


(18)

Abad-García, B.; Berrueta, L. a; Garmón-Lobato, S.; Urkaregi, A.; Gallo, B.; Vicente,

378

F. Chemometric characterization of fruit juices from Spanish cultivars according to

379

their phenolic compound contents: I. Citrus fruits. J. Agric. Food Chem. 2012, 60 (14),

380

3635–3644.

381

(19)

Ledesma-Escobar, C. A.; Priego-Capote, F.; Luque de Castro, M. D. Comparative

382

Study of the Effect of Sample Pretreatment and Extraction on the Determination of

383

Flavonoids from Lemon (Citrus limon). PLoS One 2016, 11 (1), e0148056.

384

Figure Captions

385 386 387

Figure 1. Changes in the concentration of flavanones in the extracts from Persian lime at different growth weeks. *Flavonoid aglycone.

388 389 390

Figure 2. Changes in the concentration of flavones in the extracts from Persian lime at different growth weeks. *Flavonoid aglycone.

391 392 393

Figure 3. Changes in the concentration of flavanols, phenolic acids and phenolic amino acids in the extracts from Persian lime at different growth weeks. *Flavonoid aglycone.

394 395 396 397

Figure 4. Scores (A) and variable importance projection (B) of PLS-DA comparing the samples of Persian lime at different growth weeks. W+number indicates the number of growth weeks.

398 399 400 401 402 403 404

Figure 5. General scheme of the metabolic pathways of phenolic compounds in Citrus. PAL, phenylalanine ammonia lyase; TAL, tyrosine ammonia lyase; C4H, cinnamate 4hydroxylase; C2H, cinnamate 2-hydroxylase; pC3H, p-coumarate 3-hydroxylase; Ca3M, caffeic acid 3-O-methyltransferase; CHS, chalcone synthase; CHI, chalcone isomerase; FNS, flavone synthase; F3M, flavonoid 3'-monooxygenase; FNH, flavonoid 3',5'-hydroxylase; N3D, naringenin 3-dioxygenase; FLS, flavonol synthase; ----, multip-step reaction.



Page 20 of 28

Table 1. Average Concentration (as µg g–1, mean±SD, n=12) of Flavanones in Extracts from Persian Lime Samples at Different Growth Weeks. Values with Different Letter for the Same Compound (same row) are Significantly Different Among Ripening Weeks (Tukey HSD, p≤0.01). W+number Indicates the Number of Growth Weeks. Compound

W01 µg g

-1

W03 µg g

-1

W05 µg g

-1

W07 µg g

W09

-1

µg g

-1

W12 µg g

-1

W14 µg g

-1

W16 µg g-1

Hesperidin

1091±236 c, d

615±191 e

917±153 d, e

938±214 d

1276±122 c

2005±254 a

1682±229 b

1786±186 a, b

Eriocitrin

316±51 b, c

35.8±5.1 d

66.6±12.7 c, d

166±43 c, d

465±92 b

1171±392 a

1084±224 a

1168.±96 a

Narirutin

32.3±15.4 c

16.7±9.1 c

160±62 c

652±238 b

766±202 b

1207±315 a

722±262 b

700±65 b

Neohesperidin

109±23 a

9.47±2.42 d, e

42.8±13.9 c, d

49.2±15.7 c

60.3±23.5 b, c

84.6±34.7 a, b

91.8±38.3 a, b

4.09±5.36 e

Hesperetin

4.69±2.42 a

0.38±0.29 b

1.37±0.75 b

0.14±0.09 b

< LOD

0.32±0.33 b

0.15±0.21 b

1.57±2.61 b

Naringenin

0.25±0.2 b

0.22±0.12 b

0.55±0.23 a

0.13±0.06 b

0.12±0.04 b

0.13±0.05 b

0.18±0.07 b

0.12±0.07 b


Page 21 of 28


Table 2. Average Concentration (as µg g–1, mean±SD, n=12) of Flavones in Extracts from Persian Lime Samples at Different Growth Weeks. Values with Different Letter for the Same Compound (same row) are Significantly Different Among Ripening Weeks (Tukey HSD, p≤0.01). W+number Indicates the Number of Growth Weeks. Compound

W01 µg g

-1

W03 µg g

-1

W05 µg g

-1

W07 µg g

-1

W09 µg g

-1

W12 µg g

-1

W14 µg g

-1

W16 µg g-1

Diosmin

121±33 c

27.9±6.8 d

96.2±31.2 c, d

111±23 c, d

171±45 c

366±105 a

265±53 b

289±86 a, b

Rhoifolin

5.89±2.31 d

5.97±3.36 d

26.5±7.4 d

118±39 c

163±32 b

285±36 a

188±46 b

171±20 b

Vitexin

22.0±8.9 c

8.42±7.58 c

117±35 b

151±50 b

129±46 b

237±80 a

149±56

181±60 a, b

Neodiosmin

1.30±0.82 c

23.03±9.00 b, c

4.8±1.9 c

1.63±0.71 c

2.24±0.68 c

34±10 b

97±19 a

88.5±39.2 a

Diosmetin

0.79±0.18e

12.3±4.7 c

2.79±0.98 d, e

1.05±0.38 e

1.38±0.36 e

18.0±5.6 c

51.5±10.1 a

37.5±14.0 b

Vitexin-2-O-rhamnoside

1.18±0.35 c

0.48±0.15 d

3.61±1.05 c, d

3.11±1.08 c, d

5.51±2.14 c

14.45±4.65 a, b

15.74±1.97 a

11.53±3.68 b

Orientin

5.07±1.21 b, c

2.33±0.58 c

6.80±0.91 b

5.64±1.39 b

5.17±1.43 b, c

11.96±3.7 a

11.20±1.25 a

5.80±2.84 b

Homoorientin

1.64±0.63 c, d

0.26±0.2 d

1.60±0.49 c, d

1.76±0.39 c, d

3.76±1.13 b

6.16±2.38 a

6.55±1.40 a

3.38±1.41 b, c

Apigenin-7-glucoside

< LOQ

< LOQ

0.12±0.14 e

3.86±0.63 a

2.23±0.88 b, c

2.82±1.11 a, b

0.89±0.70 d, e

1.38±1.13 c, d

Luteolin-7-glucoside

< LOD

1.09±0.64 c

2.01±0.25 a, b, c

3.14±0.75 a

1.68±0.57 b, c

2.64±0.70 a, b

1.48±0.43 c

2.19±1 a, b, c

Luteolin

0.24±0.08 a

0.11±0.04 b

0.20±0.07 a

0.09±0.02 b

0.07±0.01 b

0.09±0.03 b

0.10±0.04 b

0.08±0.03 b

Tangeretin

1.41±0.44 b

0.97±0.42 b, c

1.52±0.52 b

0.31±0.12 c

0.80±0.17 b, c

3.80±1.28 a

4.65±0.59 a

1.58±0.60 b



Page 22 of 28

Table 3. Average Concentration (mean±SD; n=12) of Flavanols, Phenolic Acids and Amino Acids (as µg g–1) in Extracts from Persian Lime Samples at Different Growth Weeks. Values with Different Letter for the Same Compound (same row) are Significantly Different Among Ripening Weeks (Tukey HSD, p≤0.01). W+number Indicates the Number of Growth Weeks. CLASS/Compound

W01 -1

W03 µg g

-1

W05 µg g

-1

W07 µg g

W09

-1

µg g

-1

W12 µg g

-1

W14 µg g

-1

W16 µg g-1

FLAVANOLS

µg g

Rutin

334±59 a

159±38 b, c, d

226±74 b

180±62 b, c

152±35 c, d

138±42 c, d

101±32 d

91±34 d

Quercetin

0.53±0.34 a

0.13±0.04 b

0.64±0.30 a

0.09±0.05 b

0.06±0.02 b

0.10±0.05 b

0.08±0.05 b

0.06±0.03 b

p-Coumaric acid

9.79±2.14 b

14.4±5.8 b

30.4±12.3 a

1.55±0.87 c

0.28±0.13 c

0.08±0.03 c

0.06±0.02 c

< LOQ

o-Coumaric acid

2.21±0.42 a

0.86±0.18 b

0.03±0.01 c

< LOD

< LOD

< LOD

< LOD

< LOD

Ferulic acid

0.92±0.42 a

< LOD

1.06±0.48 a

< LOD

< LOD

1.07±1.53 a

1.71±1.62 a

0.72±0.65 a

Caffeic acid

0.14±0.04 b

< LOD

0.07±0.01 c

< LOD

< LOD

< LOD

1.41±0.44 a

0.15±0.07 b

Phenylalanine

0.17±0.05 b

0.05±0.03 d, e

0.13±0.03 b, c

0.07±0.01 d, e

0.04±0.01 e

0.06±0.02 d, e

0.11±0.02 c, d

0.32±0.08 a

Tyrosine

0.09±0.03 b, c

0.03±0.01 d

0.10±0.04 a, b

0.05±0.01 d

0.06±0.01 c, d

0.04±0.02 d

0.05±0.02 d

0.13±0.05 a

PHENOLIC ACIDS

AMINO ACIDS


Page 23 of 28


Figure 1

Figure 1. Changes in the concentration of flavanones in the extracts from Persian lime at different growth weeks. *Flavonoid aglycone.



Page 24 of 28

Figure 2

Figure 2. Changes in the concentration of flavones in the extracts from Persian lime at different growth weeks. *Flavonoid aglycone. 24 ACS Paragon Plus Environment

Page 25 of 28


Figure 3

Figure 3. Changes in the concentration of flavanols, phenolic acids and phenolic amino acids in the extracts from Persian lime at different growth weeks. *Flavonoid aglycone.



Page 26 of 28

Figure 4

Figure 4. Scores (A) and variable importance projection (B) of PLS-DA comparing the samples of Persian lime at different growth weeks. W+number indicates the number of growth weeks. 26 ACS Paragon Plus Environment

Page 27 of 28


Figure 5

Figure 5. General scheme of the metabolic pathways of phenolic compounds in Citrus. PAL, phenylalanine ammonia-lyase; TAL, tyrosine ammonia-lyase; C4H, cinnamate 4-hydroxylase; C2H, cinnamate 2-hydroxylase; pC3H, p-coumarate 3-hydroxylase; Ca3M, caffeic acid 3-Omethyltransferase; CHS, chalcone synthase; CHI, chalcone isomerase; FNS, flavone synthase; F3M, flavonoid 3'-monooxygenase; FNH, flavonoid 3',5'-hydroxylase; N3D, naringenin 3-dioxygenase; FLS, flavonol synthase; ----, multip-step reaction. 27 ACS Paragon Plus Environment


Page 28 of 28

Graphic for table of contents


Citrus latifolia

Recommend Documents