Echium amoenum - American Chemical Society

Subscriber access provided by Washington University | Libraries

Chemistry and Biology of Aroma and Taste

Elucidation of Infusion Induced Changes in the Key Odorants and Aroma Profile of Iranian Endemic Borage (Echium amoenum) Herbal Tea Asghar Amanpour, Oscar Zannou, Hasim Kelebek, and Serkan Selli J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.9b00531 • Publication Date (Web): 13 Feb 2019 Downloaded from http://pubs.acs.org on February 14, 2019

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.

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 34

Journal of Agricultural and Food Chemistry

Elucidation of Infusion-Induced Changes in the Key Odorants and Aroma Profile of Iranian Endemic Borage (Echium amoenum) Herbal Tea Asghar Amanpoura,b, Oscar Zannoub, Hasim Kelebekc, Serkan Sellia,b*

a Department

of Biotechnology, Institute of Natural and Applied Sciences, Cukurova University, 01330 Adana, Turkey

b Department

of Food Engineering, Faculty of Agriculture, Cukurova University, 01330 Adana, Turkey

c Department

of Food Engineering, Faculty of Engineering, Adana Science and Technology University, 01110 Adana, Turkey

ORCID:

Asghar

Amanpour:

https://orcid.org/0000-0001-9783-691X,

Hasim

Kelebek:

https://orcid.org/0000-0002-8419-3019, Serkan Selli: https://orcid.org/0000-0003-0450-2668

*Corresponding author at: Department of Food Engineering, Faculty of Agriculture, Cukurova University, 01330 Adana, Turkey. Tel.: +90 322 3386173; fax: +90 322 3386614

1 ACS Paragon Plus Environment


1

Abstract

2

Infusion-induced changes in the aroma, key odorants, and their odor activity values of

3

Iranian endemic herbal (Gol-Gavzaban) tea obtained from shade-dried violet-blue petals of

4

borage (Echium amoenum) were studied for the first time. Two hot teas and one cold tea were

5

investigated and coded as 4MN (4 minutes/98°C), 16MN (16 minutes/98°C) and 24HR (24

6

hours/ambient temperature), respectively. Aromatic extracts of the tea samples were isolated

7

by liquid-liquid extraction method and analyzed by gas chromatography-mass spectrometry-

8

olfactometry (GC-MS-O) for the first time. According to the results of the aroma profiling, a

9

total of 35 common aroma compounds comprising alcohols, acids, volatile phenols, lactones,

10

aldehydes, ketone, pyrroles, and furans were identified and quantified in the tea samples.

11

Indeed, it is worth noting that the aroma profiles of the borage teas were similar. However,

12

the effects of the infusion techniques were clearly different as observed on the content of each

13

individual and total compounds in the samples. The highest mean total concentration was

14

detected in 24HR (266.0 mg/kg), followed by 16MN (247.1 mg/kg) and 4MN (216.1 mg/kg).

15

1-Penten-3-ol was the principal volatile component in all borage teas. Based on the result of

16

the flavor dilution (FD) factors, a combined total of 22 different key odorants were detected.

17

The potential key odorants with regard to FD factors in all samples were prevailingly

18

alcohols, acids and terpenes. The highest FD factors were observed in 2-hexanol (2048 in

19

4MN and 24HR; 1024 in 16MN) and 1-penten-3-ol (2048 in 24HR; 1024 in 4MN and 16MN)

20

in samples providing herbal and green notes. Principal component analysis (PCA) showed

21

that the tea samples could clearly be discriminated in terms of their aroma profiles and key

22

odorants. The findings of the current study demonstrate that the tea preparation conditions

23

have a significant impact on the organoleptic quality of borage tea.

24 25

Keywords: Borage, Echium amoenum, tea, key odorants, GC-MS-O, infusions


Page 2 of 34

Page 3 of 34


26

Introduction

27

The genus Echium (Boraginaceae) is composed of 95 species vernacular to Northern and

28

Southern Africa, Europe, Iberian Peninsula, Macaronesian archipelagos and Eastern

29

Mediterranean. In Iran, this genus is represented by four species. There has been an increased

30

interest in Echium species, including E. amoenum, because of their medicinal and nutritional

31

properties. Echium amoenum Fisch. and C. A. Mey. or borage is a biennial or perennial wild

32

herbaceous plant cultivated mostly in the northern part of Iran and some regions of Europe.1

33

E. amoenum is one of the oldest plants, which has been used to treat certain diseases

34

throughout Iranian history. Dried violet-blue petals of E. amoenum which is known as “Gol

35

Gavzaban” (ox-tongue) in the traditional medicine of Iran and consumed as a herbal tea have

36

long been used as a tonic, tranquillizer, diaphoretic and as a remedy to prevent or treat several

37

diseases including cough, stress, sore throat, pneumonia, benign prostate hyperplasia and

38

depression.2 On the other hand, in western medicine, it has been used as antifebrile,

39

antidepressant, anxiolytic, ameliorant of heart and pulmonary disturbances, poultice for

40

inflammatory swellings, diuretic, laxative, emollient and demulcent and recently as a possible

41

protective agent against cancer.3 Based on the estimation of the World Health Organization,

42

herbal drugs are used for treating diseases by 80% of the world population because they are

43

cheaper than chemical drugs. Additionally, around 30% of modern drugs are produced from

44

plants and more than 66% of plant species have medical value. Iran’s ancient civilization has

45

a long history in disease diagnosis and treatment by medicinal plants and Persian scientist

46

Avicenna has tried to develop and subdivide this subject.4 In addition to the health benefits of

47

aromatic herbal plants, they have also been used for centuries because of their aroma

48

properties. Many studies have been carried out in some herbal plants and teas such as Borago

49

officinalis,5 black teas,6 Chinese black fu-brick tea,7 rooibos tea,8 Longjing tea,9 Camellia

50

sinensis black teas,10 tea flowers of Camellia sinensis,11 green teas,12 dill, savory,13 and



51

saffron.14 A literature review revealed that the petals of E. amoenum contain anthocyanidin

52

(13%), flavonoid aglycons (0.15%), and a small amount of alkaloids and volatile oils

53

(0.05%).2

54

Aroma of herbal teas is considered as a key attribute in consumer acceptance and product

55

selection. Aroma is a perception of a large number of low molecular weight volatile

56

compounds whose combination depends on species and often to the cultivar of plants.13 The

57

majority of herbal plants are generally dried because high water content in the fresh plant

58

causes drastic deterioration due to microbial activity and biochemical reactions. If freshly

59

harvested herbal plants including E. amoenum are not dried immediately, their quality

60

deteriorates in a rapid way. Water removal by dehydration stabilizes herbal plants

61

microbiologically by lowering the water activity levels under the threshold for microbial

62

growth (aw=0.6). During air-drying, some of the volatile compounds evaporate; whereas,

63

others are partially retained and also some oxidation products emerge upon drying. Brewing

64

herbal plants in water is the most common and oldest method which is often used for making

65

herbal tea in Iran. Herbal tea from petals of E. amoenum which has attractive odor and taste is

66

popular in Iran specifically in the northern part of the country. However, only one study was

67

reported on the volatile constituents of E. amoenum petals15. In the study, aromatic extract of

68

sun-dried petals was obtained by using steam-distillation extraction technique with pentane as

69

the solvent and its volatile compounds were analyzed and identified by gas chromatography-

70

mass spectrometry. Volatile compounds were dominated by sesquiterpenes comprising

71

mainly α-cadinene, viridiflorol, α-muurolene, and ledene.15 It is worth noting that

72

sesquiterpenes are terpenes or C15-terpenoids constituted by three molecules of isoprenes.

73

They are found in plants, marine organisms, and fungi and are less volatile than other terpenes

74

but have strong odors.


Page 4 of 34

Page 5 of 34


75

There are several papers studying the extraction of volatile compounds directly from

76

herbal plants. According to the European Pharmacopoeia,16 volatile oils are extracts isolated

77

from aromatic plants only by distillation procedures such as hydrodistillation. The extracts

78

separated from aromatic plants by other methods should not be considered as volatile oils

79

even if they contain volatile compounds. Hydrodistillation (Clevenger and/or steam

80

distillation apparatus) and organic solvent extraction (Soxhlet extractor) have both been

81

extensively applied in the separation of volatile oils from herbal plants and considered as

82

conventional methods. Nonetheless, these methods both tend to degrade the original plant

83

flavor and change the volatile compounds because of the elevated applied temperatures and

84

new flavors can be developed from non-volatile precursors by Maillard reaction, carotenoid

85

degradation, and lipid oxidation. These limitations may be overcome by applying suitable

86

extraction methods. Hence, a proper extraction method must be selected with the purpose of

87

producing aromatic extracts with odor as close as possible to that of the studied sample.

88

The volatile profile of most of the foods contains a vast number of compounds but

89

very limited part of these actually gives characteristic aroma to the food. Therefore, the main

90

duty in aroma investigations is to isolate the aroma-active compounds from the less odorous

91

or odorless compounds existing in food.17 Screening techniques, like GC-O and aroma extract

92

dilution analysis (AEDA), have provided main progress in discovering the key aroma

93

compounds in a food extract.18,19 Aroma is one of the main quality factors for tea beverages;

94

however, there is no study on the aroma profile and aroma-active compounds of Iranian

95

herbal tea obtained from dried violet-blue petals of Borage (E. amoenum) in the literature

96

except the one carried out by Ghassemi et al. (2003) which brings limited information only on

97

its aroma compounds.15

98

Despite the attractive flavor and health benefits of the herbal tea from the E. amoenum

99

petals, no work has yet been found that investigates its aroma profile and key odorants. Thus,



100

the goal of the present study was (i) to compare the volatile profile and key odorants of

101

endemic Iranian herbal tea obtained from shade-dried violet-blue petals of borage (E.

102

amoenum) by applying three different infusion processes; (ii) to characterize the most aroma-

103

active compounds using aroma extract dilution analysis (AEDA) and finally to calculate the

104

odor activity values (OAVs: ratio of concentration to odor threshold).

105

Materials and methods

106

Plant Material and Chemicals. Petals (approx. 4 kg) of borage used in this study were

107

handpicked from a farming area of Mazandaran province in the northern part of Iran in 2017

108

season. The collected petals were dried in the shade at about 30°C immediately after

109

harvesting for 4–5 days. Besides, they were sorted out to remove undesirable particles or

110

debris. In addition, each air-dried sample was mixed to obtain a homogeneous fine-grade

111

sample. Afterwards, the shade-dried borage petals were packed into polyethylene bags and

112

kept at 4°C until the analysis. Distilled water used during the analysis was purified through a

113

Millipore-Q system (Milliport Billerica, Massachussets, USA). Dichloromethane was used as

114

solvent after distilling. Dichloromethane and sodium sulfate were purchased from Merck

115

(Darmstadt, Germany). Standard aroma compounds including acetic acid, hexanoic acid, 2-

116

ethyl hexanoic acid, octanoic acid, nonanoic acid, capric acid, dodecanoic acid, 1-penten-3-ol,

117

2-hexanol, 3-methyl-2-buten-1-ol, 1-octanol, benzyl alcohol, phenethyl alcohol, pentadecanol,

118

phenol, 2,5-dimethylphenol, 3-ethylphenol, p-xylene, o-xylene, dl-limonene, styrene, p-

119

cresol, thymol, carvacrol, γ-butyrolactone, 2-pyrrolidinone, nonanal, dodecanal, 2-

120

acetylpyrrole, 5-methylene-2(5H)-furanone, and 2(5H)-furanone were acquired from

121

Millipore-Sigma (St. Louis, Missouri, USA).

122

Sample preparation. Three different tea samples coded as 4MN, 16MN and 24HR

123

were prepared from borage petals. The sample preparation parameters were detailed as: 1)

124

Borage petals (1g) were mixed with 150mL of 98°C distilled water for 4 min coded as 4MN.


Page 6 of 34

Page 7 of 34


125

2) Borage petals (1g) were mixed with 150mL of 98°C distilled water for 16 min coded as

126

16MN. 3) Borage petals (1g) were soaked in 150mL of distilled water for 24 hours at ambient

127

temperature coded as 24HR. After preparing the hot tea samples (4MN and 16MN), they were

128

left to cool at room temperature. The samples of 4MN and 24HR were prepared based on the

129

traditional methods of Iran and the sample 16MN was prepared according to literature.20

130

Liquid-Liquid Extraction (LLE) of Aroma Compounds. Aroma compounds of tea

131

samples prepared from borage petals were extracted using dichloromethane as it was an

132

efficient solvent for isolation of the volatile compounds in fruits and plants.21 The isolation

133

technique was differed from our earlier survey.22 A 40-mL of a borage tea aqueous sample

134

was mixed with a 100-mL of dichloromethane. The mixture was stirred under nitrogen gas at

135

700 rpm for 45 min with a magnetic stirrer at 4°C. Afterwards, it was centrifuged at 4°C and

136

4500 rpm for 15 min. Then, phases (insoluble and aromatic) were separated and the aromatic

137

part was filtered and dehydrated with anhydrous sodium sulfate. Finally, the pooled aromatic

138

part was concentrated to 5 mL in a Kuderna Danish concentrator (Sigma Aldrich St. Louis,

139

Missouri, USA) fitted with a Snyder column at 40°C (Supelco, St. Quentin, France) and then

140

to 200 µl under a flow of purified nitrogen. Extractions were carried out in triplicate.

141

GC-MS-O Analyses of Aroma Compounds. The GC (Agilent 6890) equipped with a

142

flame ionization detector (FID), a mass selective detector (Agilent 5973-MSD) and a Gerstel

143

ODP-2 (Linthicum, MD, USA) sniffing port using a deactivated capillary column (30 cm ×

144

0.3 mm) heated at 240°C and supplied with humidified air at 40°C was used to analyze the

145

aroma and aroma-active compounds. Aroma compounds were separated by means of DB-

146

Wax column (30 m length x 0.25 mm i.d. x 0.5 μm thickness, Agilent J&W, CA, USA).

147

Helium was used at a flow rate of 1.5 mL/min as a carrier gas. The FID detector and injector

148

were set at 280 and 270°C. The oven temperature was held at 40°C for 4 min, increased to

149

90°C at 3°C/min, 130°C at 4°C/min and 240°C at 5°C/min and held at 240°C for 8 min. This



150

same oven temperature programs were used for the MSD. Afterwards, GC efﬂuent was split

151

1:1:1 among the FID, MSD, and sniffing port via a Dean’s switch. A total volume of 3 μL

152

from the aromatic extract was injected each time in pulsed splitless (40 psi; 0.5 min) mode.

153

Mass spectra were obtained in electron impact mode with an energy voltage of 70 eV and

154

quantification was performed in scan mode with a mass range of 30-300 amu. The

155

temperature of GC-MS interface and ionization source were programmed to 250 and 180ºC.

156

Identification of the aroma compounds was carried out by using mass spectral database (NIST

157

98, Wiley 6), retention index and chemical standards.14,23 Retention indices of the compounds

158

were calculated using the retention data of the linear n-alkane (C8-C32) series. The internal

159

standard method was conducted to quantify the volatiles. The potential internal standards such

160

as 2-octanol and (E)-2-butenoic acid were used as internal standards in the extractions

161

because they fulfilled all necessary criteria as internal standards and behaved similarly with

162

the main groups of analytes in studied samples. A quantitative method based on a

163

combination of experimental calibration by internal standards and FID response factors was

164

employed.24

165

Aroma Extract Dilution Analysis (AEDA). The aroma-active compounds were

166

evaluated using GC-MS-O by three experienced sniffers and the results were averaged.

167

Sniffing of borage tea extract was divided into three parts (20 min each). Each panelist

168

participated in the sniffing of the three parts interchangeably to remain alert. An AEDA was

169

used to determine the flavor dilution (FD) factors of the key aroma compounds as it was

170

successfully carried out in other previous studies.25,26 For this analysis, the condensed aroma

171

extracts were diluted one by one in a rate of 1:1, 1:2, 1:4, 1:8, 1:16…, 1:2048 using

172

dichloromethane. Sniffing of diluted extracts was sustained until no odor sense. Each

173

perceived odor was expressed as FD factor such as 2, 4, 8, 16…, 2048 corresponding to the

174

above ratios.


Page 8 of 34

Page 9 of 34


175

Representativeness Test of the Aromatic Extract / Sensory Descriptive Analysis

176

Panel. The panel composed of nine assessors (four females and five males between 23

177

and 47 years of age) from the Biotechnology Laboratory at the Food Engineering Department

178

of Cukurova University. The panelists were trained for the scent distinction and sensorial

179

assessment and had experiences in the GC–MS–O analysis.

180

Preparation and Presentation of the Samples. There are various techniques for the

181

estimation of the representativeness of the aroma extracts. In this work, a cardboard sniffing

182

strip (Granger-Veyron, Lyas, France) was used to investigate the representativeness of the

183

aroma extracts. As a reference, 10 mL of tea was placed in a brown coded flask (25 mL) for

184

these tests. The aroma extracts of the samples obtained by using the LLE technique were

185

adsorbed on the cardboard. The details of the work were given in our previous study.14

186

Similarity test. This test was carried out to estimate the closeness between the odors of

187

aromatic extracts and the tea samples (reference). The panelists were instructed to sniff and

188

memorize the aroma of the reference sample and to sniff the smelling strip for the extracts and

189

determine the similarity of their odors. A 100-mm unstructured scale anchored with “very

190

different from the reference” on the left and “similar to the reference” on the right was used.

191

The position of the sample on the unstructured scale was measured as the distance in

192

millimeters from the left anchor.

193

Intensity test. The panelists were requested to evaluate the odor intensity of the extract.

194

A 100-mm unstructured scale anchored with “no odor” on the left and “very powerful odor”

195

on the right was utilized. The position of the sample on the unstructured scale was measured

196

as the distance in millimeters from the left anchor.

197

Descriptive analysis. Nine descriptors composed of herbal, green, citrusy, phenolic,

198

tea, balsamic, sweet, fatty, and floral that designate the decisive aroma of the borage tea

199

samples were assigned by the trained panelists. The reference sample and aromatic extract



200

were offered to the panelists and they were requested to characterize the odor properties of the

201

samples by judging the consistency of above-mentioned descriptors on a scale of 10 cm. The

202

scale was scored towards the left end if there was no odor and was scored towards the right

203

end if there was very strong odor. The scores given by each panelist were determined on the

204

scale and then the average value of the odor intensity was calculated in centimeters.

205

Odor Activity Value (OAV) determination. In order to determine the impact of

206

aromatic compounds on overall borage tea aroma, the OAVs were calculated by dividing the

207

concentrations of each aroma compound with its odor threshold obtained from the literature.21

208

Statistical analysis. Three different preparation methods of the borage teas were

209

compared by using analysis of variance in SPSS 22 software package (SPSS Inc., Chicago,

210

Illinois, USA) at 95% confidence level (p≤0.05). Furthermore, XL Stat software (Addinsoft,

211

New York City, NY, USA) was used for the principle component analysis (PCA).

212 213

Results and Discussions

214

Sensory Analysis

215

Odor Sensory Profiles. Three different borage tea samples were subjected to sensory

216

analysis. The principal intensity groupings of the tea samples (4MN, 16MN, 24HR) were

217

displayed on a spider graph with nine descriptors (Figure 1). Each rating was separately used

218

by the panelists to describe the characteristic odor of the tea samples.

219

Among all the descriptors of the 4MN tea sample obtained at high temperature and the

220

shortest time, the highest score (7.9) was with herbal odor, followed by sweet (7.0), green

221

(6.8) and citrusy (5.1) while the rest of the descriptors such as tea, phenolic, balsamic, fatty,

222

and floral received the lowest scores (≤4). In the case of the 16MN tea sample that was

223

processed at high temperature for a longer time, the highest score (7.7) was in sweet odor

224

descriptor, followed by herbal (6.7), green (6.3) and citrusy (6.1) while the rest of the


Page 10 of 34

Page 11 of 34


225

descriptors received the lowest scores (≤5). Finally, the 24HR tea sample obtained at ambient

226

temperature with the longest time, a herbal odor had the greatest score (8.4) followed by

227

green (7.9), sweet (5.6) and balsamic (5.2) while the rest of the descriptors demonstrated the

228

lowest scores (≤4).

229

By comparing the descriptors of the three different tea preparation methods, it can be

230

seen that the scores of the herbal, green, balsamic, fatty, and floral descriptors were higher for

231

the 24HR tea sample than the other two tea samples. In contrast, the sweet, citrusy, phenolic,

232

and tea scores of the hot tea samples were higher than those of 24HR tea sample. According

233

to the results of GC-MS-O analysis, the most dominant aroma-active compounds were 2-

234

hexanol, 1-penten-3-ol, styrene, and nonanal which provide herbal, green, balsamic, and

235

citrusy scents and these results were found to be in agreement with the sensory analysis. This

236

result approved that the herbal and green odors caused by aroma-active compounds played a

237

vital role in borage teas’ unique aroma.

238

Similarity and Intensity Evaluation of Aromatic Extract. According to the results,

239

both the similarity and intensity scores were satisfactory when compared to the previous

240

studies.13,14,21 The mean similarity score of the aromatic extract on the smelling strips was

241

found to be 71 mm on a 100 mm scale. The mean intensity score obtained from the same

242

extract was found to be 74 mm.

243

Aroma Profile of Borage Teas. Aroma composition identified and quantified from the

244

borage tea samples and its linear retention index values on the DB-WAX column are given in

245

Table 1 as mean and standard deviation of the concentrations (mg/kg).

246

A total of 35 common aroma compounds belonging to volatile groups including acids

247

(7), alcohols (8), hydrocarbons (3), volatile phenols (4), lactones (2), aldehydes (2),

248

pyrroles-pyrrolidinone (3), terpenes (3), furans (2) and ketone (1) were successfully identified

249

and quantified from each tea sample (Table 1). It was observed that the volatile profiles of all



250

three tea samples were quite similar. However, three infusion methods had influence on the

251

concentration of each compound in the samples. All these aroma compounds were

252

characterized for the first time in borage (E. amoenum) teas. In addition, the majority of these

253

compounds have been identified in previous studies in different herbal plants and teas such as

254

Borago officinalis,5 Chinese black fu-brick tea,7 rooibos tea,8 Longjing tea,9 Camellia sinensis

255

black tea,10 tea flowers of Camellia sinensis,11 green tea,12 dill, savory,13 and saffron.14 In the

256

current study, the highest concentration of aroma compounds was observed in 24HR (266.0

257

mg/kg), followed by 16MN (247.1 mg/kg) and 4MN (216.1 mg/kg) samples. Among the

258

aroma groups, alcohols were the most important class of aroma compounds followed by

259

acids, terpenes, volatile phenols, pyrroles, lactones and aldehydes, quantitatively. In the

260

present study, 1-penten-3-ol was found in all tea samples as a dominant aroma compound.

261

Apart from 1-penten-3-ol, the leading aroma compounds were 2-hexanol, p-xylene and

262

dodecanoic acid in 4MN and 24HR tea samples and p-xylene, 2-hexanol, decanoic acid, and

263

dodecanoic acid in 16MN tea samples. In contrast, volatile compounds of the dried petals of

264

E. amoenum were dominated by sesquiterpenes comprising mainly α-cadinene, viridiflorol, α-

265

muurolene, ledene, and α-calacorene and a small amount of aliphatic alkanes.15 The

266

differences between the volatile compounds of dried petals and the borage tea samples may

267

probably be caused by hydrophilic capacities of aroma groups in borage teas, which were

268

released more efficiently with longer contact with water during infusion.

269

Alcohols. Alcohols composed 52.1%, 40.9% and 49.6% of the total aroma compounds

270

in 4MN, 16MN and 24HR tea samples, respectively. The total amounts of alcohols were

271

found to be 112.6, 101.0 and 132.0 mg/kg in 4MN, 16MN and 24HR tea samples,

272

respectively (Table 1). These compounds are synthesized from unsaturated fatty acids through

273

the lipoxygenase action especially when cell structure is disrupted in the presence of oxygen.

274

Volatile alcohols found in the hot tea samples decreased from 52% in 4MN to 41% in 16MN


Page 12 of 34

Page 13 of 34


275

as a result of increasing the infusion time. The results suggested that longer heating process

276

time reduced the total amount of alcohols from 112.6 to 101.0 mg/kg (Table 1). 1-Penten-3-

277

ol, 2-hexanol, 3-methyl-2-buten-1-ol, 1-octanol, 2-(2-butoxyethoxy) ethanol, benzyl alcohol,

278

phenethyl alcohol and pentadecanol were detected in the present study as volatile alcohol

279

compounds in the borage tea samples for the first time. Nevertheless, the majority of them

280

have formerly been detected in several herbal plants and teas.12,27,28 1-Penten-3-ol was the

281

most abundant compound in the amounts of 71.0, 65.4 and 77.0 mg/kg in 4MN, 16MN and

282

24HR tea samples, respectively (Table 1). It was also previously recorded in the aromatic

283

extracts of other teas including white, green, oolong, black, and Puer teas.27,28 According to

284

the previous literature, the amount of 1-penten-3-ol was detected twice more in the steam-

285

processed green teas than in the roast-processed green teas.29 In addition, its amount showed

286

increasing trend over the storage of green tea at ambient temperature.27 It is often created

287

from the degradation of lipids through production and other processes generating a greenish

288

aroma note which may be considered a quality defect.30 2-Hexanol was the second most

289

abundant aroma compound in the borage tea samples and its amount decreased in 4MN and

290

16MN samples (20.0 and 19.0 mg/kg) compared to the 24HR sample (23.0 mg/kg). This

291

compound was also previously determined in rooibos tea (Aspalathus linearis).6 3-Methyl-2-

292

buten-1-ol was in third place amongst the most abundant alcohol compounds (Table 1). This

293

compound was also detected in pickled tea (Miang).31

294

Acids. Acetic, hexanoic, 2-ethyl hexanoic, octanoic, nonanoic, decanoic and dodecanoic

295

acids were detected in borage tea samples (Table 1). Volatile carboxylic acids have rarely

296

been reported in teas probably due to their conversion into volatile alcohols, aldehydes, and

297

lactones during brewing. However, a considerable amount of volatile carboxylic acids was

298

detected in the present study. This compound have already been reported in Hibiscus

299

sabdariffa teas,20 spices such as saffron, dill and savory.13,14 It was found that the total amount



300

of acids was the highest in the tea samples obtained at high temperature and long infusion

301

time (16MN; 44.1 mg/kg). The larger amount of acid compounds in 24HR tea sample (38.0

302

mg/kg) than in 4MN tea (26.7 mg/kg) could be related to their hydrophilic capacities as it was

303

soaked in water for a long time (24 hours). Dodecanoic acid was the prevailing compound in

304

4MN and 24HR teas while in the case of 16MN tea, decanoic acid was the predominant

305

compound. Dodecanoic and decanoic acids in green, green purple, black and black purple

306

teas were also reported as aroma compounds.32,33

307

Hydrocarbons. Three hydrocarbons comprising p-xylene, ο-xylene and styrene were

308

identified in all borage tea samples (Table 1). 21.1, 37.6 and 28.0 mg/kg of total hydrocarbons

309

were found in 4MN, 16MN and 24HR tea samples, respectively. The high amount of

310

hydrocarbons in 16MN tea sample may be due to the higher temperature and time. The same

311

hydrocarbons were reported in several kinds of teas such as Chinese black fu-brick tea7 and

312

rooibos tea.8 p-Xylene was the most dominant hydrocarbon (Table 1) and its amount was

313

increased in higher temperature and heating time (24HR: 5.0%, 4MN: 6.0%, 16MN: 11.1%).

314

Volatile phenols. Low amounts of volatile phenols ranging from 14.1 to 19.2 mg/kg

315

were observed in borage tea samples (Table 1). Phenol, 2,5-dimethylphenol, p-cresol and 3-

316

ethylphenol were volatile phenolic compounds detected in all borage teas. The variations in

317

the contents of these compounds may result from not only the biochemical composition of the

318

raw materials but also the tea infusion conditions. The highest concentration found in the 4

319

MN samples as 19.2 mg/kg (Table 1). Volatile phenols have also been detected in essential

320

oils of leaves (8.3%) and petals (13.2%) of Borago officinalis, which belong to the

321

Boraginaceae family same as E. amoenum5 and other different tea samples.34-36

322

Pyrroles and pyrrolidinone. 2-Formylpyrrole (1H-pyrrol-2-carboxaldehyde), 2-

323

acetylpyrrole, and 2-pyrrolidinone were detected in all borage teas. 2-Formylpyrrole and 2-

324

acetylpyrrole were largely found in various herbal teas as they were accounted among major


Page 14 of 34

Page 15 of 34


325

odor contributors of green teas.12 The total amount of pyrroles was 4.8, 10.7 and 10.4 mg/kg

326

in 4MN, 16MN and 24HR tea samples, respectively (Table 1). Pyrroles belong to heterocyclic

327

compounds are created by the Maillard reaction during tea processing. The Maillard reaction

328

between reducing carbohydrates and amino acids is an important reaction leading to the

329

development of the unique aroma and taste as well as the typical browning, which contribute

330

to the sensory quality of thermally processed foods.37–39 Interestingly, higher amounts of

331

pyrroles quantified in 24HR and 16MN suggested that both tea preparation time and

332

temperature might affect pyrroles content.

333

Lactones. γ-Butyrolactone and pantolactone were identified as lactones in all borage tea

334

samples (Table 1). Both of them have been previously reported in different teas.9 The total

335

concentrations of lactones quantified in 4MN, 16MN and 24HR tea samples were 1.7, 5.4 and

336

8.6 mg/kg, respectively. In respect to this finding, it may be affirmed that longer infusion time

337

had more significant effect than temperature on the lactone. Among lactones, γ-butyrolactone

338

was found as the prevailing compound in all samples followed by pantolactone.

339

Aldehydes. Two volatile aldehydes including nonanal and dodecanal were quantified

340

(Table 1). It is worth mentioning that aldehydes are widespread in herbal teas as one of the

341

major aroma group which affect their overall aroma profile. For instance, Hibiscus sadariffa

342

teas,20 Camellia sinensis black teas,10 tea flowers of Camellia sinensis,11 green teas,12 and

343

rooibos teas8 contained significant amounts of aldehydes in their aroma profiles.

344

Terpenes. dl-Limonene, thymol and carvacrol were terpene compounds in the borage

345

tea samples. dl-Limonene was the most dominant terpene and its percentage decreased in

346

higher temperature (24HR: 4.5%, 4MN: 4.3%, 16MN: 3.0%). It has also been reported in

347

green, green purple, black and black purple teas.32

348

Other compounds. Two furans encompassing 5-methylene-2(5H)-furanone and 2(5H)-

349

furanone were detected in all borage tea samples (Table 1). These aroma compounds are sugar



350

derivatives produced during tea infusion through Maillard reaction. Piperitenone was the only

351

ketone in the tea samples and its amount in 24HR and 4MN tea samples was higher than

352

16MN tea sample (Table 1). Piperitenone was reported in the essential oil of pennyroyal

353

(Mentha longifolia) as one of the antimicrobial and antioxidant agents.40

354

PCA Results of aroma compounds. Principle component analysis (PCA) was applied

355

to evaluate the possibility of distinguishing the samples based on their aroma properties as

356

influenced by different infusion processes (4MN, 16MN and 24HR tea samples). A total of 35

357

variables (aroma compounds) were used for the PCA and the elucidated variance was

358

100.00% (Factor 1: 63.18%; Factor 2: 36.82%) (Figure 2). The tea samples based on the

359

preparation methods were clearly discriminated into three separate classes in the PCA biplot.

360

24HR tea samples were characterized by 2-hexanol, benzyl alcohol, 3-methyl-2-buten-1-ol, 2-

361

pyrrolidinone, thymol, dodecanal, styrene, 2-ethyl hexanoic acid, acetic acid, dodecanoic acid,

362

γ-butyrolactone, pantolactone, 1-octanol, and o-xylene variables. The second class was the

363

16MN tea sample, which was characterized via 2-formylpyrrole, 2-acetylpyrrole, decanoic

364

acid, carvacrol, p-xylene, and 5-methylene-2(5H)-furanone variables. Finally, the third class

365

was 4MN tea sample and differentiated by nonanal, 3-ethyl phenol, nonanoic acid, p-cresol,

366

2,5-dimethylphenol, and piperitenone variables.

367

Key Odorants and Odor Activity Values (OAVs) in Borage Teas. In order to decode

368

the characteristic key odorants in the tea samples, GC-MS-O with AEDA was used in the

369

study. The GC-MS-O results of the tea samples are summarized in Table 2. The OAV data

370

were used for confirming the contribution of the key odorants to the overall aroma of the tea

371

samples. The flavor dilution (FD) values of the key odorants in the tea samples ranged from 4

372

to 2048. The OAV of these compounds ranged from 1 to 276. A total of 22 different key

373

odorants in all tea samples including alcohols (5), terpenes (3), aldehydes (2), ketone (1),


Page 16 of 34

Page 17 of 34


374

acids (4), volatile phenols (2) hydrocarbon (1) and unknown compounds (4) were found.

375

Unknown compounds were detected by GC-MS-O but could not be identified by GC–MS.

376

To the best of our knowledge, there is no study on the key odorants of this endemic

377

Iranian herbal tea; however, most of these key odorants are prevalent as they have been

378

detected in many other herbal plants and teas such as Dong-Ding oolong tea,41 diverse brewed

379

green tea,12 rooibos tea,6 Longjing tea,9,42 and chrysanthemum flower tea.43 A total of 18, 21

380

and 20 different key odorants were identified in 4MN, 16MN, and 24 HR tea samples,

381

respectively. These variations indicated that the type, number, and power (FD factors) of the

382

key odorants were altered as a result of different tea preparation processes (time and

383

temperature). The powerful key odorants were decoded as 2-hexanol and 1-penten-3-ol in all

384

three the samples with varying FDs and OAVs (Table 2).

385

Alcohols were the principle key odorants in all three tea samples. A total of five

386

different aroma-active alcohols were identified: 1-penten-3-ol (green), 2-hexanol (herbal,

387

fruity), 3-methyl-2-buten-1-ol (fruity), 1-octanol (citrusy), and phenethyl alcohol. Their FD

388

values and OAV levels changed from 8 to 2048 and from 1 to 276, respectively. They can be

389

considered as common compounds since they have been detected as key odorants in other

390

herbal plants and teas such as Dong-Ding oolong tea

391

tea,28 rooibos tea,6 and Longjing tea.42 Among the aroma-active alcohols, 1-penten-3-ol in

392

24HR tea sample and 2-hexanol in 4MN and 24HR tea samples with an FD factor of 2048

393

were the potential key odorants, followed by 1-penten-3-ol in 4MN and 16MN samples and 2-

394

hexanol in 16MN sample with an FD factor of 1024. These powerful aroma-active alcohols

395

provided the distinct green, herbal, and fruity odor notes to these herbal teas. These two main

396

alcohols have been previously identified and quantified as the potential key odorants for the

397

overall aroma of other herbal teas such as Dong-Ding oolong tea,41 diverse brewed green

398

tea,12 oolong tea,28 and rooibos tea.6 3-Methyl-2-buten-1-ol with fruity odor note was another

41,

diverse brewed green tea,12 oolong



399

potential key aroma alcohol in the borage tea samples and its FD value was lower in 4MN and

400

16MN (32) as compared to 24HR (64). Octanol and phenethyl alcohol were other potential

401

key aroma alcohols which were observed together or separately in several teas such as oolong

402

tea28 and Longjing tea.42

403

Terpenes were the other important key odorants in tea samples. A total of three

404

different aroma-active terpenes were detected comprising dl-limonene (citrusy), thymol

405

(spicy) and carvacrol (phenolic). Their FD values and OAV levels changed from 4 to 32 and

406

from 1 to 10, respectively. Among the terpenes, dl-limonene, which gives a citrusy odor to the

407

teas, was perceived with the highest FD value in 24HR (32) and 4MN (16) tea samples. The

408

identified aroma-active terpenes are widespread as they have been detected in other black and

409

green teas.8,44

410

Other potential key odorants identified in the tea samples were carboxylic acids.

411

Hexanoic acid (cheesy), octanoic acid (sweet), nonanoic acid (green), and decanoic acid

412

(sweet) were found as aroma-active carboxylic acids in the all samples. Octanoic acid was the

413

main aroma-active acid among this group and its FD value was 8 in 16MN and 4 in 4MN and

414

24HR tea samples.

415

Two important aroma-active aldehydes including nonanal (citrusy) and dodecanal

416

(fatty) were present in all tea samples. Their FD factors and OAV levels varied from 32 to

417

512 and from 25 to 125, respectively. The odor thresholds of the aldehydes are prevalently

418

lower than those of volatile compounds; hence, they have considerable potential impacts on

419

the overall aroma of food products. Nonanal had higher FD values in all samples as compared

420

to the dodecanal (Table 2). The FD values of nonanal were increased by the increase in

421

heating time (128 in 24HR; 256 in 4MN; 512 in 16MN). Nonanal has been reported as

422

responsible for green, citrus and floral odors in Longjing tea9 and chrysanthemum flower tea43


Page 18 of 34

Page 19 of 34


423

while dodecanal was indicated as a major aroma-active compound in essential oil of

424

Polygonum minus providing pungent oily odor.45

425

p-Cresol and 3-ethylphenol were detected as aroma-active volatile phenol compounds in

426

the samples. The FD value of p-cresol with phenolic odor note rose in the shorter heating time

427

(FD=128 for 4MN). Whereas, 3-ethylphenol (soapy) had higher FD factor in hot teas (FD=16

428

for 4MN and 16MN) than the tea sample prepared in ambient condition (FD=8 for 24HR).

429

γ-Butyrolactone, which is the only aroma-active ketone and gives a sweet odor note

430

showed higher FD factor in both 16MN and 24HR tea samples (16) as compared to 4MN (8)

431

tea sample. Similarly, γ-butyrolactone has been reported to display caramel odor note in

432

Longjing tea.9 Styrene, which gives a balsamic odor to the teas, was perceived with the

433

highest FD value in 24HR (512) and 16MN (128) tea samples. The FD factor of the styrene

434

was increased by the longer infusion time in 24HR and 16MN tea samples. Moreover, four

435

unknown aroma-active compounds were determined contributing to the overall aroma of the

436

tea samples. Table 2 shows the FD factors of these unknown compounds changing between 8

437

and 64 providing tea, sweet, green, and floral odor attributes to the tea samples.

438

PCA results of key odorants. 22 variables were selected for the PCA with two

439

principal components elucidating 100% of the total variance (PC1: 65.65%; PC2: 34.35%)

440

(Figure 3). The application of the PCA exposed three distinctive groups. The ellipses indicate

441

the confidence region of the odor descriptions which can contribute to the characterization of

442

the tea samples obtained from three different preparation methods (4MN, 16MN, 24HR). The

443

odor descriptions inside the ellipse contribute more to identify the tea samples based on their

444

aroma-active compounds than those outside. In Figure 3, quadrant Q4 cluster is composed by

445

16MN tea sample and characterized by the citrusy, phenolic, sweet, and metallic attributes.

446

Attributes of tea, soapy, green, and phenolic were placed inside the quadrant Q1 and Q2

447

which characterize the 4MN tea sample. The last group was in the quadrant Q3 containing



448

24HR tea sample with citrusy, floral, spicy, green, fruity, balsamic, fatty, cheesy, sweet, and

449

metallic attributes. The sweet and herbal attributes were between 4MN and 24HR samples

450

with the same FD factor. From the PCA results, it can be concluded that the desirable odors

451

such as green, herbal, fruity, tea and floral attributes were found to be related to both 4MN

452

and 24HR tea samples more than 16MN tea sample.

453

In conclusion, the aroma-active compounds of the endemic Iranian herbal “Gol-

454

Gavzaban” tea obtained from shade-dried violet-blue petals of borage (Echium amoenum) by

455

three different infusion methods were detected for the first time with the application of aroma

456

extract dilution analysis. Regarding the influence of the infusion method on the total volatile

457

compounds of the borage tea samples, the highest amount was detected in 24HR sample

458

(prepared at ambient temperature) followed by 16MN and 4MN (prepared with water at

459

98°C); hence, it is recommended that the borage tea should be prepared at ambient

460

temperature. Terpenes, alcohols, and acids were responsible for the majority of the aroma

461

compounds in all tea samples. Of these groups, 1-penten-3-ol was found in all tea samples as

462

a dominant aroma compound. The main group of the key odorants in all samples was terpenes

463

followed by alcohols and acids. 2-Hexanol (herbal) and 1-penten-3-ol (green) were found as

464

the most potential key odorants containing the highest FD factors and OAV levels in all

465

samples. In addition, the aroma compounds and the key odorants of 24HR were dominated by

466

the highest total aroma concentration, FD factors and OAV levels and the differences could

467

come from the various infusion temperatures and duration. Thus, the outcomes from the

468

present study might be considered as supplying useful knowledge concerning the presence of

469

organoleptic properties (e.g. volatile profile, key odorants, sensory analysis) of borage tea.

470

Acknowledgments

471

We would like to thank Dr. Muharrem Keskin from Hatay Mustafa Kemal University, Turkey

472

and Pei Zhu from Cornell University, USA for their outstanding editing.


Page 20 of 34

Page 21 of 34


References

473 474

(1)

Compos. III–Cynareae. Graz, Austria Akad. Druck-u. Verlagsanstalt (in Ger. 1979.

475 476

Rechinger, K. H. Flora des Iranischen Hochlandes und der umrahmenden Gebirge.

(2)

Mehrabani, M.; Ghassemi, N.; Ghannadi, E. S. A.; Shams-Ardakani, M. Main phenolic

477

compound of petals of Echium amoenum Fisch. and CA Mey., a famous medicinal

478

plant of Iran. DARU J. Pharm. Sci. 2005, 13 (2), 65–69.

479

(3)

Ranjbar, A.; Khorami, S.; Safarabadi, M.; Shahmoradi, A.; Malekirad, A. A.; Vakilian,

480

K.; Mandegary, A.; Abdollahi, M. Antioxidant activity of Iranian Echium amoenum

481

Fisch & CA Mey flower decoction in humans: a cross-sectional before/after clinical

482

trial. Evidence-based Complement. Altern. Med. 2006, 3 (4), 469–473.

483

(4)

aureus. Brazilian J. Infect. Dis. 2004, 8 (5), 382–385.

484 485

(5) (6)

Habu, T.; Flath, R. A.; Mon, T. R.; Morton, J. F. Volatile components of rooibos tea (Aspalathus linearis). J. Agric. Food Chem. 1985, 33 (2), 249–254.

488 489

Mhamdi, B.; Wannes, W. A.; Dhiffi, W.; Marzouk, B. Volatiles from leaves and flowers of Borage (Borago officinalis L.). J. Essent. Oil Res. 2009, 21 (6), 504–506.

486 487

Abolhassani, M. Antibacterial effect of borage (Echium amoenum) on Staphylococcus

(7)

Cao, L.; Guo, X.; Liu, G.; Song, Y.; Ho, C.-T.; Hou, R.; Zhang, L.; Wan, X. A

490

comparative analysis for the volatile compounds of various Chinese dark teas using

491

combinatory metabolomics and fungal solid-state fermentation. J. Food Drug Anal.

492

2017, 26 (1), 112–123.

493

(8)

Kawakami, M.; Kobayashi, A.; Kator, K. Volatile constituents of rooibos tea

494

(Aspalathus linearis) as affected by extraction process. J. Agric. Food Chem. 1993, 41

495

(4), 633–636.

496

(9)

Cheng, Y.; Huynh-Ba, T.; Blank, I.; Robert, F. Temporal changes in aroma release of

497

Longjing tea infusion: interaction of volatile and nonvolatile tea components and

498

formation of 2-butyl-2-octenal upon aging. J. Agric. Food Chem. 2008, 56 (6), 2160–

499

2169.

500

(10)

Rawat, R.; Gulati, A.; Babu, G. D. K.; Acharya, R.; Kaul, V. K.; Singh, B.

501

Characterization of volatile components of Kangra orthodox black tea by gas

502

chromatography-mass spectrometry. Food Chem. 2007, 105 (1), 229–235.

503 504

(11)

Joshi, R.; Saini, R.; Guleria, S.; Babu, G. D.; Kumari, M.; Gulati, A. Characterization of volatile components of tea flowers (Camellia sinensis) growing in Kangra by



GC/MS. Nat. Prod. Commun. 2011, 6 (8), 1155–1158.

505 506

(12)

compounds in various brewed green teas. Molecules 2013, 18 (8), 10024–10041.

507 508

Lee, J.; Chambers, D. H.; Chambers, E.; Adhikari, K.; Yoon, Y. Volatile aroma

(13)

Amanpour, A.; Kelebek, H.; Selli, S. Aroma constituents of shade-dried aerial parts of

509

Iranian dill (Anethum graveolens L.) and savory (Satureja sahendica Bornm.) by

510

solvent-assisted flavor evaporation technique. J. Food Meas. Charact. 2017, 11 (3),

511

1430–1439.

512

(14)

Amanpour, A.; Sonmezdag, A. S.; Kelebek, H.; Selli, S. GC–MS–olfactometric

513

characterization of the most aroma-active components in a representative aromatic

514

extract from Iranian saffron (Crocus sativus L.). Food Chem. 2015, 182, 251–256.

515

(15)

Ghassemi, N.; Sajjadi, S. E.; Ghannadi, A.; Shams-Ardakani, M.; Mehrabani, M.

516

Volatile Constituents of Amedicinal Plant of Iran, Echium amoenum Fisch. and CA

517

Mey. DARU J. Pharm. Sci. 2003, 11 (1), 32–33.

518

(16)

European Pharmacopoeia, 6th edn; 2007.

519 520

Council of Europe (COE). European Directorate for the Quality of Medicines,

(17)

Schieberle, P.; Gassenmeier, K. F.; Guth, H. U.; Sen, A.; Grosch, W. Character Impact

521

Odour Compounds of Different Kinds of Butter. LWT - Food Sci. Technol. 1993, 26

522

(4), 347–356.

523

(18)

Fickert, B.; Schieberle, P. Identification of the key odorants in barley malt (caramalt)

524

using GC/MS techniques and odour dilution analyses. Food/Nahrung 1998, 42 (06),

525

371–375.

526

(19)

Schieberle, P.; Grosch, W. Evaluation of the flavour of wheat and rye bread crusts by

527

aroma extract dilution analysis. Zeitschrift fuer Leb. und Forsch. 1987, 185 (2), 111–

528

113.

529

(20)

Ramírez‐Rodrigues, M. M.; Balaban, M. O.; Marshall, M. R.; Rouseff, R. L. Hot and

530

cold water infusion aroma profiles of Hibiscus sabdariffa: fresh compared with dried.

531

J. Food Sci. 2011, 76 (2).

532

(21)

Selli, S.; Kelebek, H. Aromatic profile and odour-activity value of blood orange juices

533

obtained from Moro and Sanguinello (Citrus sinensis L. Osbeck). Ind. Crops Prod.

534

2011, 33 (3), 727–733.

535

(22)

Selli, S.; Canbas, A.; Varlet, V.; Kelebek, H.; Prost, C.; Serot, T. Characterization of

536

the most odor-active volatiles of orange wine made from a Turkish cv. Kozan (Citrus

537

sinensis L. Osbeck). J. Agric. Food Chem. 2007, 56 (1), 227–234.


Page 22 of 34

Page 23 of 34

538


(23)

Amanpour, A.; Kelebek, H.; Kesen, S.; Selli, S. Characterization of Aroma-Active

539

Compounds in Iranian cv. Mari Olive Oil by Aroma Extract Dilution Analysis and

540

GC–MS-Olfactometry. J. Am. Oil Chem. Soc. 2016, 93 (12), 1595–1603.

541

(24)

de Saint Laumer, J.-Y.; Cicchetti, E.; Merle, P.; Egger, J.; Chaintreau, A.

542

Quantification in gas chromatography: prediction of flame ionization detector response

543

factors from combustion enthalpies and molecular structures. Anal. Chem. 2010, 82

544

(15), 6457–6462.

545

(25)

Schlutt, B.; Moran, N.; Schieberle, P.; Hofmann, T. Sensory-directed identification of

546

creaminess-enhancing volatiles and semivolatiles in full-fat cream. J. Agric. Food

547

Chem. 2007, 55 (23), 9634–9645.

548

(26)

Frank, S.; Wollmann, N.; Schieberle, P.; Hofmann, T. Reconstitution of the flavor

549

signature of Dornfelder red wine on the basis of the natural concentrations of its key

550

aroma and taste compounds. J. Agric. Food Chem. 2011, 59 (16), 8866–8874.

551

(27)

Kaishi 1982, 56, 625–630.

552 553

Hara, T. Changes in aroma components of green tea after storage. Nippon Nogeikagaku

(28)

Wang, K.; Liu, F.; Liu, Z.; Huang, J.; Xu, Z.; Li, Y.; Chen, J.; Gong, Y.; Yang, X.

554

Comparison of catechins and volatile compounds among different types of tea using

555

high performance liquid chromatograph and gas chromatograph mass spectrometer. Int.

556

J. Food Sci. Technol. 2011, 46 (7), 1406–1412.

557

(29)

Sci. 1995, 5 (2), 20.

558 559

(30) (31)

Kawakami, M.; Chairote, G.; Kobayashi, A. Flavor constituents of pickled tea, miang, in Thailand. Agric. Biol. Chem. 1987, 51 (6), 1683–1687.

562 563

Ravichandran, R.; Parthiban, R. Lipid occurrence, distribution and degradation to flavour volatiles during tea processing. Food Chem. 2000, 68 (1), 7–13.

560 561

Choi, S.-H. The aroma components of commercial green tea picked in August. J. Life

(32)

Joshi, R.; Rana, A.; Gulati, A. Studies on quality of orthodox teas made from

564

anthocyanin-rich tea clones growing in Kangra valley, India. Food Chem. 2015, 176,

565

357–366.

566

(33)

Food Sci. Technol. 1984.

567 568

Mick, W.; Gotz, E. M.; Schreier, P. Volatile, acids of black tea aroma. Leb. Technol.

(34)

Schuh, C.; Schieberle, P. Characterization of the key aroma compounds in the beverage

569

prepared from Darjeeling black tea: quantitative differences between tea leaves and

570

infusion. J. Agric. Food Chem. 2006, 54 (3), 916–924.



571

(35)

Schieberle, P.; Schuh, C. Aroma compounds in black tea powders of different

572

origins—changes induced by preparation of the infusion. In Developments in Food

573

Science; Elsevier, 2006; Vol. 43, pp 151–156.

574

(36)

Lv, H.-P.; Zhong, Q.-S.; Lin, Z.; Wang, L.; Tan, J.-F.; Guo, L. Aroma characterisation

575

of Pu-erh tea using headspace-solid phase microextraction combined with GC/MS and

576

GC–olfactometry. Food Chem. 2012, 130 (4), 1074–1081.

577

(37)

Ottinger, H.; Soldo, T.; Hofmann, T. Discovery and structure determination of a novel

578

Maillard-derived sweetness enhancer by application of the comparative taste dilution

579

analysis (cTDA). J. Agric. Food Chem. 2003, 51 (4), 1035–1041.

580

(38)

Hofmann, T.; Schieberle, P. Evaluation of the key odorants in a thermally treated

581

solution of ribose and cysteine by aroma extract dilution techniques. J. Agric. Food

582

Chem. 1995, 43 (8), 2187–2194.

583

(39)

Hofmann, T.; Schieberle, P. Flavor contribution and formation of the intense roast-

584

smelling odorants 2-propionyl-1-pyrroline and 2-propionyltetrahydropyridine in

585

Maillard-type reactions. J. Agric. Food Chem. 1998, 46 (7), 2721–2726.

586

(40)

Gulluce, M.; Sahin, F.; Sokmen, M.; Ozer, H.; Daferera, D.; Sokmen, A.; Polissiou,

587

M.; Adiguzel, A.; Ozkan, H. Antimicrobial and antioxidant properties of the essential

588

oils and methanol extract from Mentha longifolia L. ssp. longifolia. Food Chem. 2007,

589

103 (4), 1449–1456.

590

(41)

Wang, L.-F.; Lee, J.-Y.; Chung, J.-O.; Baik, J.-H.; So, S.; Park, S.-K. Discrimination of

591

teas with different degrees of fermentation by SPME–GC analysis of the characteristic

592

volatile flavour compounds. Food Chem. 2008, 109 (1), 196–206.

593

(42)

Gong, X.; Han, Y.; Zhu, J.; Hong, L.; Zhu, D.; Liu, J.; Zhang, X.; Niu, Y.; Xiao, Z.

594

Identification of the aroma-active compounds in Longjing tea characterized by odor

595

activity value, gas chromatography-olfactometry, and aroma recombination. Int. J.

596

Food Prop. 2017, 1–15.

597

(43)

Luo, D.; Chen, J.; Gao, L.; Liu, Y.; Wu, J. Geographical origin identification and

598

quality control of Chinese chrysanthemum flower teas using gas chromatography–mass

599

spectrometry and olfactometry and electronic nose combined with principal component

600

analysis. Int. J. Food Sci. Technol. 2017, 52 (3), 714–723.

601

(44)

Kraujalyte, V.; Pelvan, E.; Alasalvar, C. Volatile compounds and sensory

602

characteristics of various instant teas produced from black tea. Food Chem. 2016, 194,

603

864–872.


Page 24 of 34

Page 25 of 34

604


(45)

Rusdi, N. A.; Goh, H.-H.; Baharum, S. N. GC-MS/Olfactometric characterisation and

605

aroma extraction dilution analysis of aroma active compounds in’Polygonum minus’

606

essential oil. Plant Omics 2016, 9 (4), 289.

607

(46)

odour. Water 1971, 4, 367–371.

608 609

(47)

Buttery, R. G.; Teranishi, R.; Ling, L. C.; Turnbaugh, J. G. Quantitative and sensory studies on tomato paste volatiles. J. Agric. Food Chem. 1990, 38 (1), 336–340.

610 611

Zoeteman, B. C. J.; Kraayeveld, A. J. A.; Piet, G. J. Oil pollution and drinking water

(48)

Karl-Otto, S.; Hans-Dieter, B.; Christof, von R.; Schnabel, K.-O.; Belitz, H.-D.; von

612

Ranson, C. Untersuchungen zur Struktur-Aktivitäts-Beziehung bei Geruchsstoffen 1.

613

Mitteilung: Wahrnehmungsschwellenwerte und Geruchsqualitäten von gesättigten

614

aliphatischen und alicyclischen Verbindungen mit Sauerstoff-FunktionInvestigations

615

on the structure-activit. Zeitschrift für Leb. und Forsch. 1988, 187 (3), 215–223.

616

(49)

Yang, R.; Sugisawa, H.; Nakatani, H.; Tamura, H.; Takagi, N. Comparison of odor

617

quality in peel oils of acid citrus. Nippon Shokuhin Kogyo Gakkaishi 1992, 39 (1), 16–

618

24.

619 620 621



622

FIGURE CAPTIONS

623

Figure 1. Odor sensory features of borage (E. amoenum) tea samples.

624

Figure 2. PCA biplot of aroma compounds in borage (E. amoenum) tea samples.

625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641


Page 26 of 34

Page 27 of 34


Herbal 9 8 7

Floral

Green

6 5 4 3 2

Fatty

Citrusy

1 0

Sweet

Phenolic

Balsamic

4MN

Tea

16MN

Figure 1.


24HR


Biplot (axes F1 and F2: 100.00 %) 1

Page 28 of 34

. 16MN

Hexanoic acid Octanoic acid o-Xylene 1-Octanol Pantolactone 24HR γ-Butyrolactone Dodecanoic acid Acetic acid 2-Ethyl hexanoic acid Styrene Dodecanal Thymol 2-Pyrrolidinone

.

1

F2 (36.82 %)

2-Formylpyrrole 2-Acetylpyrrole Decanoic acid Carvacrol p-Xylene 5-Methylene-2(5H)furanone

3-Methyl-2-buten-1-ol Benzyl alcohol 2-Hexanol dl-Limonene Phenethyl alcohol 2(5H)-Furanone 2-(2-Butoxyethoxy)ethanol Phenol 1-Penten-3-ol Pentadecanol

0

Nonanal 3-Ethylphenol Nonanoic acid

Piperitenone

-1

2,5-Dimethylphenol p-Cresol

.

4MN -1 -2

-2

-1

-1

0

F1 (63.18 %) Figure 2.


1

1

2

2

Page 29 of 34


Biplot (axes F1 and F2: 100.00 %) 1

Q1

.

1

Q2 4MN 17-Phenolic 5-Sweet 6-Herbal

18-Green 19-Soapy 3-Tea

F2 (34.35 %)

0

2-Citrusy

0

15-Floral 20-Spicy 1-Green 7-Fruity 14-Floral 4-Balsamic 12-Fatty

8-Citrusy 11-Sweet

0

16-Sweet 21-Phenolic -1

22-Metallic

.

.

13-Cheesy 10-Sweet 9-Citrusy, metal

24MN

16MN

Q4

Q3

-1 -1

-1

-1

-1

0

0

0

1

1

1

1

F1 (65.65 %) Figure 3. PCA biplot of the potent key odorants and their related odor descriptions; 1-Green [1-Penten-3-ol], 2-Citrusy [dl-Limonene], 3-Tea [Unknown I], 4-Balsamic [Styrene], 5-Sweet [Unknown II], 6-Herbal [2-Hexanol], 7-Fruity [3-Methyl-2-buten-1-ol], 8-Citrusy [Nonanal], 9-Citrusy, metal [1-Octanol], 10-Sweet [γ-Butyrolactone], 11-Sweet [Unknown III], 12-Fatty [Dodecanal], 13-Cheesy [Hexanoic acid], 14-Floral [Phenethyl alcohol], 15Floral [Unknown IV], 16-Sweet [Octanoic acid], 17-Phenolic [ρ-Cresol], 18-Green [Nonanoic acid], 19-Soapy [3-Ethylphenol], 20-Spicy [Thymol], 21-Phenolic [Carvacrol], and 22Metallic [Decanoic acid].



Page 30 of 34

Table 1. Aroma compounds in borage (Echium amoenum L.) tea samples No.

Aroma compounds

LRI 1

Concentration (mean ± SD) 2

Identification 3

4MN

16MN

24HR

1400

1.6 ± 0.0c

2.6 ± 0.1b

4.1 ± 0.2a

LRI, MS, Std

1807

1.0 ±

0.0c

4.2 ±

0.1a

3.5 ±

0.2b

LRI, MS, Std

2.0 ±

0.1c

3.6 ±

0.1b

5.9 ±

0.2a

LRI, MS, Std

0.0c

2.7 ±

0.1a

2.4 ±

0.2b

LRI, MS, Std

Acids 1 2 3

Acetic acid Hexanoic acid 2-Ethyl hexanoic acid

1911

4

Octanoic acid

2041

1.4 ±

5

Nonanoic acid

2124

5.8 ± 0.1b

6.0 ± 0.2a

4.4 ± 0.0c

LRI, MS, Std

6

Decanoic acid

2313

3.9 ± 0.0c

13.0 ± 0.2a

4.7 ± 0.1b

LRI, MS, Std

7

Dodecanoic acid

2517

11.0 ± 0.5c

12.0 ± 0.4b

13.0 ± 0.6a

LRI, MS, Std

26.7

44.1

38.0

1158

71.0 ± 1.9b

65.4 ± 2.1c

77.0 ± 2.3a

LRI, MS, Std

1254

0.9b

1.0c

23.0 ±

1.2a

LRI, MS, Std

11.0 ±

0.3a

LRI, MS, Std

2.7 ±

0.1a

LRI, MS, Std

1.9 ±

0.1a

LRI, MS, Tent

4.7 ±

0.2a

LRI, MS, Std

5.3 ±

0.2a

LRI, MS, Std

6.4 ±

0.2a

LRI, MS, Std

Total Alcohols 8 9 10 11 12 13 14 15

1-Penten-3-ol 2-Hexanol 3-Methyl-2-buten-1-ol 1-Octanol 2-(2-Butoxyethoxy)ethanol Benzyl alcohol Phenethyl alcohol Pentadecanol

1325 1546 1749 1823 1859 2241

Total

20.0 ± 8.1 ±

0.3b

1.1 ±

0.0c

1.0 ±

0.0b

1.9 ±

0.0b

3.8 ±

0.2b

5.7 ±

0.3b

19.0 ± 7.1 ±

0.2c

2.2 ±

0.0b

0.3 ±

0.0c

0.6 ±

0.0c

2.6 ±

0.3c

3.8 ±

0.1c

112.6

101.0

132.0

Volatile phenols 16

Phenol

1958

2.6 ± 0.0b

1.9 ± 0.0c

3.6 ± 0.1a

LRI, MS, Std

17

2,5-Dimethylphenol

2068

5.3 ± 0.2a

2.8 ± 0.0c

4.6 ± 0.2b

LRI, MS, Std

18

p-Cresol

2074

5.2 ± 0.1a

2.3 ± 0.0c

3.4 ± 0.0b

LRI, MS, Std

2150

0.3b

0.2a

0.2c

LRI, MS, Std

19

3-Ethylphenol Total

6.1 ±

7.1 ±

3.1 ±

19.2

14.1

14.7

1130

13.0 ± 0.6c

27.0 ± 0.8a

13.0 ± 0.3b

LRI, MS, Std

1169

5.2 ±

0.3c

6.4 ±

0.3b

6.8 ±

0.3a

LRI, MS, Std

2.9 ±

0.0c

4.2 ±

0.0b

8.2 ±

0.1a

LRI, MS, Std

Hydrocarbons 20 21 22

p-Xylene o-Xylene Styrene

1245

Total

21.1

37.6

28.0

1943

2.5 ± 0.0c

5.2 ± 0.1a

4.1 ± 0.2b

LRI, MS, Std

0.0c

0.1a

0.1b

LRI, MS, Tent LRI, MS, Std

Pyrroles 23

2-Acetylpyrrole

24

2-Formylpyrrole

2010

0.8 ±

25

2-Pyrrolidinone

2032

1.5 ± 0.0b

1.5 ± 0.0b

3.4 ± 0.1a

4.8

10.7

10.4

Total

4.0 ±


2.9 ±

Page 31 of 34


Lactones 26 27

γ-Butyrolactone Pantolactone

1647

1.1 ± 0.0c

3.1 ± 0.0b

5.0 ± 0.0a

LRI, MS, Std

2024

0.0c

0.1b

0.0a

LRI, MS, Tent

Total

0.6 ±

2.3 ±

3.6 ±

1.7

5.4

8.6

1391

5.0 ± 0.1b

6.1 ± 0.1a

2.0 ± 0.0c

LRI, MS, Std

1682

0.0c

0.0b

0.0a

LRI, MS, Std

Aldehydes 28 29

Nonanal Dodecanal Total

1.4 ±

1.9 ±

4.3 ±

6.4

8.0

6.2

Furans 30

5-Methylene-2(5H)-furanone

1635

3.5 ± 0.1b

4.3 ± 0.1a

2.8 ± 0.0c

LRI, MS, Std

31

2(5H)-Furanone

1702

2.5 ± 0.0b

2.1 ± 0.0c

3.1 ± 0.0a

LRI, MS, Std

6.0

6.4

5.9

1199

9.2 ± 0.2b

7.5 ± 0.1c

12.0 ± 0.2a

LRI, MS, Std

2167

1.8 ±

0.0b

2.0 ±

0.0b

3.2 ±

0.0a

LRI, MS, Std

4.0 ±

0.1b

8.4 ±

0.1a

4.2 ±

0.0b

LRI, MS, Std

Total Terpenes 32 33 34

dl-limonene Thymol Carvacrol

2246

Total

15.0

17.9

19.4

2.6 ± 0.1a

1.9 ± 0.0b

2.7 ± 0.1a

2.6

1.9

2.7

216.1

247.1

266.0

Ketone 35

Piperitenone Total General total

1905

1

LRI, MS, Tent

LRI: Linear retention index calculated on DB-Wax capillary column. 2 Concentration: results are the means of three repetitions as mg/kg. Different lower-case letters (a,b,c) within the same line indicate significant differences (p

Echium amoenum - American Chemical Society

Recommend Documents