Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE
What do we know about the chemistry of strawberry aroma? Detlef Ulrich, Steffen Kecke, and Klaus Olbricht J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.8b01115 • Publication Date (Web): 13 Mar 2018 Downloaded from http://pubs.acs.org on March 13, 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.
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 36
Journal of Agricultural and Food Chemistry
What do we know about the chemistry of strawberry aroma?
Detlef Ulrich1 Steffen Kecke2and Klaus Olbricht3,4
1
Julius Kühn-Institute, Institute for Ecological Chemistry, Plant Analysis and Stored Product
Protection, Quedlinburg, Germany 2
Julius Kühn-Institute, Data Processing Unit, Quedlinburg, Germany
3
Hansabred GmbH Co. KG, Dresden, Germany
4
Humboldt-Universität zu Berlin, Albrecht Daniel Thaer-Institute, Berlin, Germany
Corresponding Author: 1
(D.U.) Phone: (49) 3946-47231. Fax: (49) 3946-47300. Email:
[email protected]
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 2 of 36
ABSTRACT 1
The strawberry, with its unique aroma, is one of the most popular fruits worldwide. The demand for
2
specific knowledge of metabolism in strawberries is increasing. This knowledge is applicable for
3
genetic studies, plant breeding, resistance research, nutritional science, and the processing industry.
4
The molecular basis of strawberry aroma has been studied for more than 80 years. Thus far, hundreds
5
of volatile organic compounds (VOC) have been identified. The qualitative composition of the
6
strawberry volatilome remains controversial though considerable progress has been made during the
7
past several decades. Between 1997 and 2016, twenty-five significant analytical studies were
8
published. Qualitative VOC data were harmonized and digitized. In total, 979 VOC were identified,
9
590 of which were found since 1997. However, 659 VOC (67 %) were only listed once (single
10
entries). Interestingly, none of the identified compounds were consistently reported in all of the studies
11
analyzed. The present need of data exchange between ‘omic’ technologies requires high quality and
12
robust metabolic data. Such data are unavailable for the strawberry volatilome thus far. This review
13
discusses the divergence of published data regarding both the biological material and the analytical
14
methods. The VOC extraction method is an essential step that restricts inter-laboratory comparability.
15
Finally, standardization of sample preparation and data documentation are suggested to improve
16
consistency for VOC quantification and measurement.
17 18
KEYWORDS: Fragaria ×ananassa Duch., volatile organic compounds, gas chromatography, mass
19
spectrometry, identification
20
2 ACS Paragon Plus Environment
Page 3 of 36
Journal of Agricultural and Food Chemistry
21 22
INTRODUCTION
23
The garden strawberry (Fragaria ×ananassa) is one of the most popular fruits and represents a
24
significant global economic market. In 2014, the worldwide production was 8.1 million tons
25
(FAOSTAT).1 The unique flavor of strawberries is the primary reason for its high popularity.2
26
The garden strawberry emerged in the mid-1700s in Versailles, from an accidental
27
hybridization of the American octoploids, F. chiloensis and F. virginiana.3,4 The hybridization
28
combined some of the most important characteristics of the garden strawberry: large fruits from the
29
Chilean landrace, and a unique, pleasant sweetish aroma deriving from the smaller-fruited, red wild
30
Virginia strawberry. This combination cannot be found in other species of the genus although
31
Fragaria species have unique and diverse aroma patterns.5 Due to the combination of high sensory
32
popularity and high nutrition, the health value of the garden strawberry can provide an important
33
component for healthy human nutrition.
34
The chemical basis of strawberry aroma has been a frequently researched topic. The volatile
35
organic compounds (VOC) and their subset, the aroma compounds, were intensively identified and
36
quantified. Early analytical investigations were performed by Coppens and Hoejenbos6 for F.
37
moschata (syn. F. eliator) at the end of the 1920s, and were published about 1939. In his compendium
38
from 1991, Latrasse7 evaluated 54 studies on garden and wild strawberries and reported about 360
39
aroma compounds. For the garden strawberry, the review of Zabetakis and Holden8 lists over 80
40
studies with about 280 volatiles. Since the end of the 1960s, the number of VOC identified has
41
increased due to improved analytical technology, i.e., gas chromatography-mass spectrometers (GC-
42
MS) coupling. The online database of the Nutrition and Food Research Institute of the Netherlands9
43
(TNO) currently lists 323 VOC for strawberries from 15 substance classes. These have been identified
44
by quadrupole and ion trap mass spectrometers. From the intensity of activities in the field of
45
metabolomics and the introduction of new powerful analytical techniques the identification of more
46
‘new’ metabolites are expected10.
47
3 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
48
A higher demand for metabolite data also results from modern breeding strategies. The goals
49
of many breeding programs now include development of specific individual secondary metabolites or
50
metabolite profiles responsible for the health value or sensory quality11,12 .
51
Page 4 of 36
Few of the strawberry VOC are consistently identified between studies. For example, in Ulrich
52
and Olbricht11 and Schwieterman et al.13, only 28 of 116 total VOC identified were reported in both
53
studies. To date, a comprehensive inventory of the published strawberry volatilome is incomplete,
54
despite more than 80 years of metabolite research. Today's analytical techniques provide detailed
55
chemical profiles but compounds are inconsistent between reports. This lack of reliability is an
56
obstacle for plant genetics, research, and breeding programs. The identification of metabolic
57
quantitative trait loci (QTL), the candidate genes for valuable metabolites and transcription studies, are
58
based on metabolite analysis. Modern breeding strategies for creating new cultivars with a high
59
resistance against diseases, high health value, and high sensory quality depend on the implementation
60
of powerful and consistent chemical analyses.
61
The aims of this review are to provide a detailed overview of strawberry volatilome
62
identification and to highlight potential causes of the apparent irreproducibility. Twenty-five studies
63
from the past 21 years that have not previously been summarized in review have been included here.
64
Based on published methodological information, factors influencing the confirmation of the substance
65
identification are discussed with regard to the sample collection, preparation, and analytical methods.
66 67
Methods
68
From 1997 to 2016, more than 250 papers were published concerning strawberry VOC in the context
69
of sensory quality (Web of Science search with descriptor ‘strawberr*’ AND (‘aroma’ OR ‘volatiles’);
70
www.webofknowledge.com; web access 2017-10-20). Of the published manuscripts, 25 studies were
71
chosen for evaluation10,11,13-35. The studies were chosen based on 1) the VOC analyses performed by
72
the authors and 2) reports of strawberry VOC identification. Studies that described the quantitation of
73
few individual compounds or a compound class were excluded.36 This approach ensured that studies
74
with comparable objectives were included, and that the research focused on the elucidation of the
75
strawberry volatilome. 4 ACS Paragon Plus Environment
Page 5 of 36
Journal of Agricultural and Food Chemistry
76
Because the evaluated literature covered the past 21 years, a review of Zabetakis and Holden8
77
and the TNO database9 were also included to contrast with previous findings. The TNO data contained
78
references from 1956 to 1995, plus one publication from Du et al.23 Du et al. did not contain ‘new’
79
compounds in comparison to Zabetakis and Holden8. An approximation for the state of knowledge
80
about VOC identification previous to 1996 was obtained by constructing an association set from both
81
VOC lists.
82
Most published substance sets listed the VOC by chemical name. Because the spelling of the
83
chemical names was inconsistent, each entry was transferred manually into the internationally
84
common chemical abstract service (CAS) registry numbers (http://www.cas.org/content/chemical-
85
substances/). Online databases were searched for the CAS numbers (The Flavornet,
86
http://www.flavornet.org/flavornet.html, The Good Scents Company,
87
www.thegoodscentscompany.com/data, The NIST WebBook, https: //www.nist.gov/, PubChem,
88
https://pubchem.ncbi.nlm.nih.gov/, ChemSpider, http://www.chemspider.com/). Obvious literal errors
89
in the names were corrected.
90
For substances with stereo isomers, the unspecified substance name was listed if no reliable
91
isomer determination was possible by mass spectrometric identification. In principle, both CAS
92
numbers for the respective stereo isomers as well as a separate number for the unspecified form are
93
available. If only the unspecified form was mentioned, e.g., hexenal, the compound was assigned to
94
the trans-form, i.e., (E)-2-hexenal, to avoid the unspecified substance being counted as a separate
95
entry. In analogy, unspecified lactones were added to the gamma-form. Substance names for which no
96
CAS number was available were included in the overall list under the chemical names given in the
97
original study. For statistical analysis, the association set (set union) was constructed from the 27
98
substance lists using an in-house web application (FindIntersection).
99
From the viewpoint of accurate chemical analysis the data on substance identification were not
100
strictly handled in the evaluated literature. For this review, we have adopted the guidelines from
101
Molyneux and Schieberle (2007)37 for the nomenclature in the text (Table 2), even though these have
102
been used in the original work in a different way. This guideline includes two steps for identification:
103
a) mass spectrometric fragmentation and retention indices must be determined on at least two 5 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 6 of 36
104
separation columns of different polarity (Table 2, footnotes 1 and 2), and b) comparison of the mass
105
spectra and retention indices (RI) with those of authentic reference substances as a so-called co-elution
106
(Table 2, footnote 3). Identification on the basis of a simple search in mass spectrometric libraries
107
cannot be considered sufficient. Therefore, substances were ‘identified’ when the requirements of
108
points a) and b) were met on at least one separation column. Otherwise the substances were described
109
as ‘tentatively identified’.
110
For substance quantitation (which is not the main topic of this review), the published studies
111
were also inconsistent. Using gas chromatography-flame ionization detector (GC-FID) or gas
112
chromatography-mass spectrometry (GC-MS), the term ‘quantitation’ could be used when
113
accompanied by a stable isotope dilution analysis (SIDA) or by the standard addition method. In these
114
cases, the terms ‘quantitation’ or ‘quantified’ were used. In other cases, the terms ‘semi-quantitation’
115
or ‘semi-quantified’ were applied. Thus, the specification of absolute, quantitative values, such as
116
µg/kg or ng/g is inadmissible if based only on a single internal standard.
117 118 119
Objectives of the VOC analyses
120
Strawberry VOC analyses can be grouped into five categories (Table 1)
121
1. Sensory quality (aroma, flavor)
122
2. Interactions of genes with environment (G x E)
123
3. Bioactivity of VOC as signaling or defense substances
124
4. Metabolic analyses for plant genetics and plant cultivation and
125
5. Methodological work in the area of VOC analysis
126
Schieberle and Hofmann34,39 determined the character impact compounds in strawberry juice by means
127
of quantitative measurements (SIDA). Their sensory activity was assessed using the aroma value
128
concept which was based on the comparison of metabolite concentrations with their odor thresholds.58
129
A total of 15 VOC were fully identified, with no new substances being published in comparison with
130
that in the TNO database9 or the review by Zabetakis & Holden.8 For the comparison of the 25
131
publications from the methodological point of view38, Schieberle and Hofmann34 was the only 6 ACS Paragon Plus Environment
Page 7 of 36
Journal of Agricultural and Food Chemistry
132
comprehensive assessment of the sensory quality of the VOC, including complete identification, exact
133
quantitation, flavor concept, and recombination experiments. This study was based on a previous work
134
by Schieberle39 using a gas chromatograph-olfactory (GC-O) study in which identical VOC list was
135
mentioned except for the mesifuran (2,5-dimethyl-4-methoxy-3(2H)-furanone). Further studies by Ulrich et al.35, Gomes da Silva et al.32, Nuzzi et al.28, Fukuhara et al.30,
136 137
Jouquand et al.27, Zhang, Y.T. et al.26, Li et al.25, Du et al.23, Vandendriessche et al.19, Samykanno et
138
al.18, Cannon et al.10, Schwieterman et al.13, and Ulrich and Olbricht11 investigated the VOC patterns
139
with regard to strawberry aroma. The two recent studies by Schwieterman et al.13 and Ulrich and
140
Olbricht11 pursue an extended and comparable objective by using adequate instrumental analysis for
141
VOC and non-VOC (or aggregate parameters). They also correlated data with a consumer test to
142
obtain consumer preference (acceptance).
143
GC is the method of choice for VOC analyses. For this purpose sample preparation is of
144
crucial importance. In sample preparation, the analytes are isolated from a complex, aqueous matrix
145
and transferred to a water-free GC-capable sample. Concentration of the aroma compounds in
146
essential because the VOC occur in the parts per million (ppm) or sub-ppm range. Initially, liquid-
147
liquid extraction with organic solvents was used for VOC extraction. Schieberle & Hofmann34, Ulrich
148
et al.35 and Zhang J.J. et al.24 used this method. The bulk of research was carried out by adsorption
149
methods such a solid phase micro extraction (SPME), purge and trap (P&T) and stirbar-sorptive
150
extraction (SBSE). Li et al.25 and Cannon et al.10 used isolation of the VOC by solid phase extraction
151
(SPE).
152
Ozcan and Barringer22 applied static headspace extraction in combination with selected-ion
153
flow-tube mass spectrometry (SIFT-MS) and without GC separation. However, the static headspace
154
extraction is unsuitable for detecting aroma relevant VOC in the ppm range or below because this
155
method has no concentration step.
156
A further objective for VOC analysis of strawberries was the determination of specific
157
methodological questions in sample preparation, separation, or detection. Nuzzi et al. (2008)28
158
evaluated six strawberry genotypes as a model system for comparative analysis of qualitative and
159
semi-quantitative results on the aroma activity of VOC obtained on the basis of GC-O, and the 7 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
160
calculation of odor activity values. A total of 38 identified VOC were included in the comparison.
161
Both methods for the determination of the aroma patterns yielded comparable results. Deviating from
162
other studies, Ozcan & Barringer (2011)22 used SIFT-MS as a detection and identification method. A
163
total of 41 VOC were examined depending on the variety, the frost storage, and the VOC release in
164
mouthspace and nosespace. Measurement of the VOC in the respiratory air was possible because the
165
SIFT-MS can be used without pre-concentration and removal of water. Vandendriessche et al.20 used
166
an unconventional separation and detection method for investigating the impact of an infection on
167
VOC patterns by using the headspace multi-capillary column-ion mobility spectrometry (HS MCC-
168
IMS). With this approach, 97 VOC were semi-quantified to identify biomarkers for infection by
169
Botrytis cinera.
170
Page 8 of 36
Over the past twenty years, gas chromatography time-of-flight mass spectrometry (GC-
171
TOFMS) systems have become increasingly useful as powerful analyzers for substance identification.
172
For strawberry, three applications have been described using this technique. In the studies of Song33,
173
the application note N. N.17 and Samykanno18, 31, 74, and 124 VOC were identified.
174
A major component of strawberry metabolite patterns is genotypically determined. Therefore,
175
comparative studies of strawberry varieties were performed (Table 3). In the 25 reviewed studies, 76
176
cultivars and 24 breeding clones were analyzed. Fragaria ×ananassa `Camarosa´ (5 times), `Albion´,
177
`Chandler´, `Festival´ and `Toyonoka´ (4 times each) were examined most frequently. The analyses of
178
cultivars and breeding lines shows that VOC analysis played a major role for breeding research and
179
practical cultivation of strawberries.
180
Further questions for VOC analysis that are related to both sensorial quality research and
181
breeding include ontogenic effects22 and the interaction of genotype by environment (G x E). The
182
latter was considered by Jouquand et al.27 and Samykanno et al.18. This work is a prerequisite for a
183
metabolite-directed selection in the breeding process, because environmentally dependent metabolites
184
are unsuitable as separate breeding objectives.
185 186 187
General survey of the identified VOC The TNO database9 and the review by Zabetakis and Holden8 were state of the art 8 ACS Paragon Plus Environment
Page 9 of 36
Journal of Agricultural and Food Chemistry
188
determinations for VOC identification until 1996, with 307 and 275 substances listed. By creating the
189
association set from the lists of both publications, 389 VOC were identified in strawberries previous to
190
this review.
191
The 25 studies evaluated for this review listed between 15 and 124 identified VOC. In
192
contrast, Cannon et al.10, observed 553 substances (Figure 1). For their analysis, 100 kg fruit of the
193
cultivar `Ciflorette´ were extracted by means of dichloromethane and then separated into 125 fractions
194
using solid phase extraction (SPE). Subsequently, fractions were separated by means of two-
195
dimensional GC (2D-GC) on two separation columns of different polarity and detected by quadrupole
196
MS. Out of the 553 VOC, six substances could be fully identified by co-elution and the remaining 547
197
VOC were tentatively identified by means of a library search and retention index (RI) comparison.
198
Thus, Cannon et al.10 provided the most comprehensive identification of the strawberry volatilome,
199
listing 322 compounds which were (tentatively) identified in strawberry for the first time.
200
The complete strawberry volatilome to date, including those compounds found previous to
201
1996 (389 VOC), totals to 979 VOC. Thus, the strawberry is one of the most thoroughly studied fruits
202
in the plant kingdom. The frequency of the 30 most often identified VOC is summarized (Table 4).
203
The most frequently found substances in strawberries are the methyl and ethyl esters of hexanoic and
204
butanoic acids with 24 to 22 entries. The thirty most frequently analyzed compounds included esters
205
(17, 15 straight chain and 2 branched), acids (4), lactones (2), aldehydes (2), furans (2), alcohols (1),
206
ketones (1) and terpenoids (1), while on the other extreme, 670 substances occurred only once.
207
Surprisingly, none of the 979 VOC was co-mentioned in all of the 27 evaluated literature sources; 959
208
compounds were co-mentioned in fewer than half of the reports. Thus, only a partial consensus
209
concerning the qualitative composition of the strawberry volatome was reached among researchers,
210
despite their intensive analytical work. Possible causes for this discrepancy are subsequently
211
discussed.
212 213 214 215
Studies using GC-O and the flavor value concept In seven studies, the aim was to separate the aroma-active VOC, called character impact compounds or key compounds, from a larger number of identified substances10, 21, 23, 28, 30, 34, 35. For this 9 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 10 of 36
216
purpose, the aroma value concept58 including the determination of odor activity values as well as the
217
GC-O approach were applied. Both methods lead to an improvement of the compound identification
218
because reference substances must be used, or because the odor quality is added as an additional filter
219
for identification when using GC-O. Nuzzi et al.28 and Schieberle and Hofmann34 compared the odor
220
activity values (OAV) using GC-O. These two studies showed that the different methods provided
221
similar results for the character impact compounds. Nuzzi et al.28 confirmed this although only semi-
222
quantitative data were used to determine the OAV.
223
Between 12 and 48 VOC were determined as character impact compounds in individual
224
studies (Figure 2). Seven publications listed 105 character impact compounds. Only one substance,
225
ethyl butanoate, was co-listed in complete consensus in these seven studies. While 36 VOC were
226
found more than in one study, 69 compounds were single entries. Thus, the research on character
227
impact compounds also exhibited the same trend as for the general VOC identification without
228
consideration of the olfactory properties. Only a few substances were listed in consensus. Between 66
229
% and 67 % of the identified compounds are single entries, i. e. they were only identified in one study
230
(69 out of 105 in the seven GC-O studies and 659 out of 979 of all 27 reviewed literature sources).
231 232 233
Influencing factors on substance identification The quantitative and qualitative composition of the plant metabolome is subject to complex
234
influencing factors. These include, on the one hand, factors which determine the quality of the test
235
material, such as the genotype and the environmental influences. The environmental influences
236
(‘outside’ influences) include cultivation, harvesting, storage conditions, maturity and
237
phytopathological status. On the other hand, the results of metabolic analyses are also known to be
238
influenced by the analytical method. Important parameters are VOC extraction (cleanup), gas
239
chromatographic separation, detection and data processing. Some of the essential parameters are
240
summarized in Table 2.
241
Genotype. VOC data from 71 cultivars and 30 breeding clones have been published since 1997
242
(Table 3). In all studies that investigate several genotypes, the genotype has been described as a key
243
influencing factor on the quality and quantity of the aroma patterns. The most commonly analyzed 10 ACS Paragon Plus Environment
Page 11 of 36
Journal of Agricultural and Food Chemistry
244
cultivars were F. ×ananassa ‘Albion’11, 13, 16, 22, ‘Chandler’11, 14, 22, 32, ‘Festival’11, 13, 14, 23, 27, and
245
‘Toyonoka’24, 25, 26, 30 which were evaluated in four independent studies and ‘Camarosa’11, 13, 14, 21, 22 in
246
five. However, the lists of the identified VOC for the respective cultivars also show only a low
247
qualitative correspondence. This is an indication that other variables beside the genotype influenced
248
the results.
249
Sampling size. Obtaining a consistent representative sample is an important prerequisite for
250
chemical analysis. The minimum sample size in chemical analysis depends on the accuracy of the
251
method, particle size, and homogeneity.40,41 If guidelines from the area of solids analytics were to be
252
adopted for a typical strawberry fruit size of about 30 mm, a sample in the range of ten kg to several
253
hundred kg would be needed.41 The actual sample sizes reported were between 0.5 g and 100 kg
254
(Table 2). Four of the 25 studies used a sample size in the kilogram range but other sample sizes were
255
in the range of a single fruit or the size of few grams. Fruit-to-fruit variations have been published for
256
VOC content and dry mass.42,43 These effects are more quantitative than qualitative. However,
257
qualitative effects may occur if a sample size was too small and results dropped below detection limit.
258
Maturity. Fruit composition varies during maturation.22,57 Differences in the results may be
259
due to differential fruit maturity. Strawberries are not commonly evaluated using Brix value. Though
260
Nuzzi28 evaluated the total soluble solids (TSS) values as ‘ripening index’, he did not use this as a
261
consistent harvest criteria. In several studies18, 20, 22, 23, 24, 25, 28 fruit color was judged prior to harvest.
262
Harvest influence within the season. Environmental influences cause qualitative and
263
quantitative effects on the metabolite patterns and also on the results of substance identification.44 A
264
partly solution for this special problem can be the analysis of a batch sample containing all mature and
265
healthy fruits of the season.16,44 Thus, the influence of the harvesting time during the season could be
266
eliminated. Investigations on material of unknown provenance, species designation and cultivation site
267
(Table 3) are basically problematic with regard to reproducibility.
268
Storage. Studies on the aroma of the strawberries were mostly performed on fresh fruit. In 9 of
269
24 studies, frozen material was used. Freezing was used to bridge the period between harvest and
270
instrumental analysis. Regarding the VOC patterns, proper freezing at -25 °C is a viable option
11 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
271
because, contrary to the sensory quality, the patterns of the volatiles are less influenced by this
272
process.45
273
Page 12 of 36
Homogenization. For the cleanup step, the fruits were mostly homogenized and mixed with
274
inorganic salts (calcium chloride, sodium chloride, tin chloride, and sodium flouride) to suppress
275
enzyme activity. Some studies used quartered and some undamaged fruits. With whole fruits used for
276
VOC analysis, the analogy to sensory quality in consumption was lost, because the process of
277
homogenization affects the VOC patterns, e. g. by influencing lipoxygenase activity (LOX). In this
278
case 17,33, the VOC pattern corresponds more to the orthonasal perception (smell) rather than the flavor
279
when consumed (retronasal).
280
VOC isolation. The method of choice for VOC analysis is GC. The preparation of a GC-
281
capable sample is associated with isolation and concentration as well as transfer into a water-free
282
matrix. The cleanup process is the decisive and most complex step regarding the result of the analysis.
283
For this reason, a number of attempts were made to develop effective methods for VOC analysis
284
without time-consuming cleanup. These techniques have not been established for a well-founded
285
metabolome analysis, but they can be used for certain purposes, such as the determination of
286
maturity.46,47
287
The classical method for VOC isolation is liquid-liquid extraction (LLE). Here the extraction
288
power depends on the polarity of the extraction solvent. The disadvantages are the high workload, the
289
lack of automation, and the extraction of non-volatile compounds. The LLE is unsuitable for a high
290
throughput method and is therefore used for basic investigations with a small sample number.
291
Nevertheless, the best results with regard to quality (number of extracted VOC) and quantity (high
292
recovery rates) can be achieved by means of LLE.
293
In addition to LLE, adsorption methods are used for VOC isolation. These include purge-and-
294
trap methods (dynamic headspace), solid-phase extraction (SPE), and stirbar-sorptive extraction
295
(SBSE). Since the market launch of solid phase micro extraction (SPME) for water analysis in 1993,
296
this technique has been used in many applications, and is widely used in the VOC analysis of plant
297
material. Ten out of the 25 evaluated studies used this technique for isolation. The wide distribution of
298
the SPME technique is due to its easy handling without solvent use, including the possibility of 12 ACS Paragon Plus Environment
Page 13 of 36
Journal of Agricultural and Food Chemistry
299
automation. A disadvantage of the SPME, as with all adsorption methods, is the strong discrimination
300
effect for individual substances or substance classes in complex matrices. This effect is particularly
301
pronounced for SPME due to its design and leads to insufficient extraction of polar compounds like
302
acids and furanones because of very low recovery rates. Exact quantitation of VOC in complex
303
matrices is impossible due to the bias in combination with the limited adsorption capacity of the
304
SPME fibers. A comparison of SPME with other isolation techniques was published for strawberry
305
and other crops.48-50 Although the SPME technique is subject to severe restrictions for complex
306
systems, such as the strawberry matrix, this technique is often used in a completely uncritical manner,
307
without accounting for the limitations on the recovery rates or substance identification.
308
Separation. Apolar, medium-polar and polar separation columns were used for
309
chromatographic separation. For a complete substance identification by means of MS in the sense of a
310
good analytic practice, independent separation and identification was required on two separation
311
columns of different polarities.37 This approach reduces the likelihood of false identifications due to
312
peak overlapping, particularly in the case of automatic library searches. These requirements were not
313
fully met by any of the 25 studies.
314
Mass spectrometric detection. For identification, quadrupole-, ion trap-, TOF- and a SIFT-MS
315
were used. Due to its high mass resolution and high scan rate, the TOF-MS technique is preferred for
316
the identification of a large number of mass fractions. At the same time, the limitation of the
317
chromatographic separation capacity of the GC column can be counteracted by deconvolution. GC-
318
TOF-MS systems were used in three studies in which 31, 74, and 124 VOC were identified. 17,18,33
319
However, the largest number of identified VOC (553) was obtained by Cannon et al. 10 using
320
quadrupole-MS detector, by means of a complex cleanup method in combination with a 2D GC.
321
Details on the quality of the reported substance identification are given in the following section.
322
Quantitation. An exact quantitation in gas chromatography is only possible by co-elution of
323
isotope-labeled references (stable isotope dilution analysis) or by the standard addition method. In the
324
studies considered here, Schieberle and Hofmann (1997)34 and Li et al. (2009)25achieved these
325
conditions. The remaining reports share semi-quantitative data, even if the original work did not
326
provide any information or (incorrectly) specify a concentration unit. 13 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 14 of 36
327
Artifacts. One problem is the distinction between genuine VOC and artifacts. For this purpose,
328
the execution of suitable blank analyses, possibly using statistical evaluation methods, must be carried
329
out. Some lists contain ethanol and acetone, which are possible artifacts. Thermal reactions can also be
330
a cause of artifact formation in the GC injector. Thus, when the SPME technique is not applied
331
appropriately (fruit particles on the fiber surface), by thermal degradation, furanones maybe formed
332
from sugars51 adhering to the fiber or injection needle of the SPME device. At > 160 °C, furanones
333
may decompose into a variety of small molecules, including acetone and other ketones, as well as
334
alkylfuranones.52 The formation of furans was reported as artifacts when using SPME as sample
335
preparation method.59 The reviewed lists of identified VOC also contain phthalates and biphenyl
336
which may originate from plasticizers, agrochemicals or food additives. Because no information is
337
given on the blank-analyses in most studies, it cannot be ruled out that the list of the 979 identified
338
VOC in strawberry contains a series of artifacts.
339 340
Quality of GC-MS identification in the evaluated studies
341
Coupling GC and MS is the method of choice for VOC identification. Identification
342
techniques provide different confidence levels. The highest level is reached when the identity of the
343
molecule is validated by co-elution of an authentic standard substance and subsequent MS analysis
344
(confirmed identification or fully identified). A lower level is achieved when a structure is proposed
345
only by spectral similarities present in a database (tentative, provisional or putative identification).
346
Table 2 shows the level of mass spectrometric identification in three stages. Using co-elution of
347
authentic references with proof of origin of reference compounds (Table 2, level 3) represents the
348
highest level and corresponds to the requirements of Molyneux and Schieberle37 for an exact
349
identification on one column type. None of the studies provided full results of identifications on two
350
separation columns of different polarities. Furthermore, some of the studies are based solely on the
351
comparison of mass spectrometric fragmentation with library data or do not provide any information
352
for identification (Table 2).10,13,17,22,32,35 The inclusion of retention data for the qualification of the MS
353
identification was carried out in the remaining papers. The co-elution of authentic reference substances
354
is mentioned in 15 studies, but only a part of the published VOC could be covered in some papers. 14 ACS Paragon Plus Environment
Page 15 of 36
Journal of Agricultural and Food Chemistry
355
Information on the reference substances used and their origin or synthesis is only available in two
356
studies.10,34 More detailed identification studies for chiral compounds are not reported. The addition of
357
odor qualities for flavor-active VOC, as discussed for GC-O, is another way to improve the
358
identification of flavor substances.
359 360
Conclusion
361 362
In conclusion the most important reasons for the low conformity in the substance identification which was documented here are the following:
363
-
use of different genotypes
364
-
influence of different environmental settings (G x E)
365
-
sample preparation inconsistencies
366
-
use of different MS types
367
-
uncritical use of identification methods
368
-
artifacts
369
The progress in analytical techniques and bioinformatics has led to the development of
370
metabolomics and thus to the increased application of this approach in many areas. The exchange of
371
data with other "omics approaches" is currently boosting scientific progress. A key issue for chemical
372
analysis is substance identification, which represents a challenge for the analysis of VOC in complex
373
biological systems.
374
In the last twenty years, many more VOC substances in strawberries have been identified
375
(Figure 1). The volatilome of the strawberry is one of the most frequently investigated plants. Prior to
376
2016, more than 979 VOC were identified, As literature has shown, however, publications have little
377
consensus on defining primary compounds. This situation is scientifically unsatisfactory and leads to
378
inconsistent results in analyses using metabolite data. This inconsistent information is
379
disadvantageous, especially for plant genetics and breeding. Metabolic data are increasingly being
380
sought for functional genomics and breeding for flavor. Because approximately 67 % of the published
381
VOC were reported in only one study and no single substance was found in all of the 27 evaluated
382
sources, the reproducibility becomes a serious question. 15 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 16 of 36
383
The qualitative and quantitative factors influencing the results of the VOC analyzes, as
384
discussed above, can be grouped into two classes: a) the biological system through internal (genotype)
385
and external (environment) factors causes a considerable diversity of the metabolite patterns. b) the
386
results are significantly influenced by the analytical method used.
387
a) The influences on biological variability are complex because of the gene-by-environment
388
interactions (GxE) and inherent to the system. Both variables (G x E) can produce both quantitative
389
and qualitative variability, individually and in their combination. This influence cannot be eliminated
390
and must be statistically verified by means of an adequate test design (multi-year, multiple-order,
391
number of biological repetitions). In the 25 studies, 18 used only a single sample from a single harvest
392
date for analysis which can be the reason for qualitative differences.44 A partial solution for this
393
problem was suggested by Ulrich and Olbricht16,44 using the analysis of a batch sample containing all
394
mature and healthy fruits of the season (pooling). Thus, the influence of the harvesting time during the
395
season could be mitigated. Investigations on material of unknown provenance, species designation and
396
cultivation site (Table 3) are basically problematic with regard to reproducibility.
397
b) The influences introduced by the chosen analytical method are significant. All of the
398
influencing factors discussed above can cause quantitative and qualitative effects on the result and can
399
reinforce each other. To minimize the methodological problems an adequate test design has to be
400
chosen. The basis for reproducible analysis is taken during sampling and clean-up. Errors that are
401
introduced in this step cannot be eliminated in the further process of analysis by applying sophisticated
402
detectors and algorithms for data processing. If sufficient material is available, a maximum sample
403
size should be selected with a subsequent sample division. When investigating the strawberry
404
volatiles, sample sizes of 0.5 g to 100 kg were used. However, sometimes analytical experiments with
405
very small sample sizes, especially in genetics and breeding, in single plant experiments of a
406
population or of wild material are necessary. Often, biological replicates may not be possible. If
407
metabolic analysis is carried out, the analytical reliability of the results must be discussed. The
408
extraction method appeared to exert the greatest influence. A reference to this was the large number of
409
identified substances in the work of Cannon et al.10, in which more than 500 VOC could be detected
410
by applying a large sample quantity in combination with an elaborated extraction technique. In 16 ACS Paragon Plus Environment
Page 17 of 36
Journal of Agricultural and Food Chemistry
411
comparison, the use of fast and simple isolation methods such as the headspace SPME must be viewed
412
critically. Even the most widely used detection by MS can be seen as a cause of divergent results. The
413
technically complex MS detectors often have lower stability than GC standard detectors, e.g., flame
414
ionization detector. Even mass spectrometers of the same manufacturer and type can produce different
415
results depending on the degree of pollution and tune-set. In the totality of the methodological
416
variables discussed above, these factors are a cause of the high inter-laboratory variability.
417
Completely standardized methods such as proposed in the Metabolomics Standards Initiative (MSI)
418
have so far not been used in the field of aroma analysis.53,54 However, the scientific methodology
419
requires the development and exchange of reproducible and falsifiable data, which is obviously not the
420
case for the 659 single entries in the strawberry volatilome. The compound patterns recorded by
421
instrumental analysis are highly dependent on the procedures used. Therefore, the use of a
422
standardized method for the production of test material (genotype, cultivation method) can be useful
423
for studies on genetics and breeding (marker, transcriptomics).
424
Distinct analytical techniques cannot cover the full metabolome and any other characteristic of
425
the plant material. Until now standard operating procedures (SOP) for analysis have not been widely
426
accepted.55 The standardization of protocols to guide data production, quality and robustness is central
427
to coordinating efforts between scientists working in different laboratories. In chemical analysis, SOP
428
can help narrow down the divergence. The focus here must be on sample preparation and VOC
429
extraction. The occurrence of artifacts must be examined by careful methodological experiments.
430
Finally, the quality of generated data must be evaluated. The level of substance identification by MS
431
(confirmed or tentative identification) should be indexed. Also results which are cited in secondary
432
literature should indicate the quality of substance identification to prevent an inflationary increase of
433
tentatively identified substances in substance lists and databases. The consequence is repeated
434
misidentification and misapplication in other areas of science. These errors would misdirect the
435
mapping of metabolic QTLs, the study of candidate genes, the transcriptomics, and the marker-
436
assisted selection in plant breeding.
437 438 439
17 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
440
Page 18 of 36
Abbreviations Used
441 442
2D-GC – two-dimensional gas chromatography
443
CAS – chemical abstracts service
444
FID – flame ionization detector
445
GxE – gene by environment interaction
446
GC – gas chromatography
447
GC-O – gas chromatography olfactometry
448
LLE – liquid-liquid-extraction
449
LOX - lipoxygenase
450
MS – mass spectrometry
451
OAV – odor activity values
452
P&T – purge and trap
453
Q-MS – quadrupol mass spectrometry
454
QTL – quantitative trait loci
455
RI – retention index
456
SBSE – stir bar sorptive extraction
457
SIFT-MS – selected ion flow tube-mass spectrometry
458
SPE – solid phase extraction
459
SPME – solid phase microextraction
460
SOP – standard operation procedure
461
SQ – semi-quantitation
462
TOF – time of flight
463
VOC – volatile organic compounds
464 465
Acknowledgment
466
The authors wish to thank Kirsten Weiß for the careful analysis of the original data.
18 ACS Paragon Plus Environment
Page 19 of 36
467
Journal of Agricultural and Food Chemistry
References
468 469
(1) N.N. Internet Communication. http://www.fao.org/faostat/en/#home, access date 2017-10-16.
470 471
(2) Bhat, R.; Geppert, J.; Funken, E.; Stamminger, R. Consumers Perceptions and Preference for Strawberries—A Case Study from Germany. Int. J. Fruit Sci. 2015, 15, 405-424.
472 473
(3) Darrow, G. M. M. The Strawberry: history, breeding and physiology.; 1st ed.; Holt, Rinehart and Winston. New York, Chicago, San Francisco: 1966.
474 475
(4) Staudt, G. The origin and history of the large-fruited garden strawberry, Fragaria x ananassa. Der Züchter 1961, 31, 212-218.
476 477
(5). Ulrich, D.; Komes, D.; Olbricht, K.; Hoberg, E. Diversity of aroma patterns in wild and cultivated Fragaria accessions. Genet.. Resour. Crop Evo. 2007, 54, 1185-1196.
478 479
(6) Coppens, A.; Hoejenbos, L. Investigation of the volatile constituents of strawberry juice (Fragaria elatior Ehrh.). J Rec Trav Chim Pays Bas 1939, 58, 680-690.
480 481
(7) Latrasse, A. Fruits III. In Volatile compounds in Foods and beverages, Maarse, H., Ed.; Marcel Dekker Inc.: 1991; pp 329-387.
482 483
(8) Zabetakis, I.; Holden, M. A. Strawberry flavour: Analysis and biosynthesis. J. Sci. Food Agric. 1997, 74, 421-434.
484 485
(9) N.N. Internet Communication,TNO Nutrition and Food Research Institute, Centraal Instituut voor Voedingsonderzoek Zeist, http://www.vcf-online.nl, access date 2016-09-15.
486 487 488
(10) Cannon, R. J.; Agyemang, D.; Curto, N. L.; Yusuf, A.; Chen, M. Z.; Janczuk, A. J. In-depth analysis of Ciflorette strawberries (Fragaria x ananassa ‘Ciflorette’) by multidimensional gas chromatography and gas chromatography-olfactometry. Flavour and Fragrance J. 2015, 30, 302-319.
489 490 491
(11) Ulrich, D.; Olbricht, K. A search for the ideal flavor of strawberry - Comparison of consumer acceptance and metabolite patterns in Fragaria x ananassa Duch. J. App. Bot. Food Qual. 2016, 89, 223-234.
492 493
(12) Folta, K. M.; Klee, H. Sensory sacrifices when we mass-produce mass produce. Hortic. Res. 2016, 3, 16032.
494 495 496 497
(13) Schwieterman, M. L.; Colquhoun, T. A.; Jaworski, E. A.; Bartoshuk, L. M.; Gilbert, J. L.; Tieman, D. M.; Odabasi, A. Z.; Moskowitz, H. R.; Folta, K. M.; Klee, H. J.; Sims, C. A.; Whitaker, V. M.; Clark, D. G. Strawberry Flavor: Diverse Chemical Compositions, a Seasonal Influence, and Effects on Sensory Perception. Plos One 2014, 9 (2), e88446.
498 499
(14) Oz, A.; Baktemur, G.; Kargi, S. P.; Kafkas, E. Volatile Compounds of Strawberry Varieties. Chem. Nat. Compd. 2016, 52, 507-509.
500 501
(15) Mishra, R.; Kar, A. Effect of Storage on the Physicochemical and Flavour Attributes of Two Cultivars of Strawberry Cultivated in Northern India. Scientific World Journal 2014, 1-7
502 503
(16) Ulrich, D.; Olbricht, K. Diversity of volatile patterns in sixteen Fragaria vesca L. accessions in comparison to cultivars of Fragaria xananassa. J. App. Bot. Food Qual. 2013, 86, 37-46.
19 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 20 of 36
504 505 506
(17) N.N. Enhanced odour profiling of strawberries using TD and TOF MS. Markes International Appl. Note 515 2013, 1-7. https://www.markes.com/Resources/Mass-Spec-applicationnotes/default.aspx
507 508
(18) Samykanno, K.; Pang, E.; Marriott, P. J. Genotypic and environmental effects on flavor attributes of 'Albion' and 'Juliette' strawberry fruits. Sci. Hortic. 2013, 164, 633-642.
509 510 511
(19) Vandendriessche, T.; Geerts, P.; Membrebe, B. N.; Keulemans, J.; Nicola, B. M.; Hertog, M. L. Journeys through aroma space: a novel approach towards the selection of aroma-enriched strawberry cultivars in breeding programmes. Plant Breeding 2013, 132, 217-223.
512 513 514
(20) Vandendriessche, T.; Keulemans, J.; Geeraerd, A.; Nicolai, B. M.; Hertog, M. L. Evaluation of fast volatile analysis for detection of Botrytis cinerea infections in strawberry. Food Microbiol. 2012, 32, 406-414.
515 516 517
(21) Ubeda, C.; San-Juan, F.; Concejero, B.; Callejon, R. M.; Troncoso, A. M.; Lourdes Morales, M.; Ferreira, V.; Hernandez-Orte, P. Glycosidically Bound Aroma Compounds and Impact Odorants of Four Strawberry Varieties. J. Agric. Food Chem. 2012, 60, 6095-6102.
518 519
(22) Ozcan, G.; Barringer, S. Effect of Enzymes on Strawberry Volatiles during Storage, at Different Ripeness Level, in Different Cultivars, and during Eating. J.Food Sci. 2011, 76, C324-C333.
520 521 522
(23) Du, X.; Plotto, A.; Baldwin, E.; Rouseff, R. Evaluation of Volatiles from Two Subtropical Strawberry Cultivars Using GC-Olfactometry, GC-MS Odor Activity Values, and Sensory Analysis. J. Agric. Food Chem. 2011, 59, 12569-12577.
523 524 525
(24) Zhang, J. J.; Wang, X.; Yu, O.; Tang, J. J.; Gu, X. G.; Wan, X. C.; Fang, C. B. Metabolic profiling of strawberry (Fragaria x ananassa Duch.) during fruit development and maturation. J. Exp. Bot. 2011, 62 (3), 1103-1118.
526 527 528
(25) Li, X. X.; Fukuhara, K.; Hayata, Y. Concentrations of Character Impact Odorants in 'Toyonoka' Strawberries Quantified by Standard Addition Method and PQ Column Extraction with GC-MS Analysis. J. Jpn. Soc. Hortic. Sci. 2009, 78, 200-205.
529 530 531
(26) Zhang, Y. T.; Wang, G. X.; Dong, J.; Zhong, C. F.; Kong, J.; Li, T. Z.; Han, Z. H. Analysis of Volatile Components in Strawberry Cultivars Xingdu 1 and Xingdu 2 and Their Parents. Agric. Sci. China 2009, 8, 441-446.
532 533 534
(27) Jouquand, C.; Chandler, C.; Plotto, A.; Goodner, K. A Sensory and Chemical Analysis of Fresh Strawberries Over Harvest Dates and Seasons Reveals Factors That Affect Eating Quality. J. Am. Soc. Hortic. Sci. 2008, 133, 859-867.
535 536 537
(28) Nuzzi, M.; Lo Scalzo, R.; Testoni, A.; Rizzolo, A. Evaluation of Fruit Aroma Quality: Comparison Between Gas Chromatography-Olfactometry (GC-O) and Odour Activity Value (OAV) Aroma Patterns of Strawberries. Food Analytical Methods 2008, 1, 270-282.
538 539
(29) De Boishebert, V.; Giraudel, J. L.; Montury, M. Characterization of strawberry varieties by SPME-GC-MS and Kohonen self-organizing map. Chemom. Intell. Lab. Syst. 2006, 80, 13-23.
540 541 542
(30) Fukuhara, K.; Li, X. X.; Okamura, M.; Nakahara, K.; Hayata, Y. Evaluation of odorants contributing to 'Toyonoka' strawberry aroma in extracts using an adsorptive column and aroma dilution analysis. J. Jpn. Soc. Hortic. Sci. 2005, 74, 300-305.
543 544
(31) Hakala, M. A.; Lapvetelainen, A. T.; Kallio, H. P. Volatile compounds of selected strawberry varieties analyzed by purge-and-trap headspace GC-MS. J. Agric. Food Chem. 2002, 50, 1133-1142. 20 ACS Paragon Plus Environment
Page 21 of 36
Journal of Agricultural and Food Chemistry
545 546 547
(32) Gomes da Silva, M. D.; Chaves das Neves, H. J. Complementary use of hyphenated purgeand-trap gas chromatography techniques and sensory analysis in the aroma profiling of strawberries (Fragaria x ananassa). J. Agric. Food Chem. 1999, 47, 4568-4573.
548 549 550
(33) Song, J.; Fan, L. H.; Beaudry, R. M. Application of solid phase microextraction and gas chromatography time-of-flight mass spectrometry for rapid analysis of flavor volatiles in tomato and strawberry fruits. J. Agric. Food Chem. 1998, 46, 3721-3726.
551 552 553
(34) Schieberle, P.; Hofmann, T. Evaluation of the character impact odorants in fresh strawberry juice by quantitative measurements and sensory studies on model mixtures. J. Agric. Food Chem. 1997, 45, 227-232.
554 555
(35) Ulrich, D.; Hoberg, E.; Rapp, A.; Kecke, S. Analysis of strawberry flavour - discrimination of aroma types by quantification of volatile compounds. Eur. Food Res. Technol. 1997, 205, 218-223.
556 557 558
(36) Olbricht, K.; Ulrich, D.; Weiss, K.; Grafe, C. Variation in the Amounts of Selected Volatiles in a Model Population of Fragaria x ananassa Duch. as Influenced by Harvest Year. J. Agric. Food Chem. 2011, 59, 944-952.
559 560
(37) Molyneux, R. J.; Schieberle, P. Compound Identification: A Journal of Agricultural and Food Chemistry Perspective. J. Agric. Food Chem. 2007; 4625.
561 562
(38) Schieberle, P.; Hofmann, T. It comes down to taste - The molecular world in enjoying food. Chem. Unserer Zeit 2003, 37, 388-401.
563 564 565
(39) Schieberle, P. Heat-induced changes in the most odour-active volatiles of strawberries. In Trends in Flavour Research, Maarse, H.; van der Heij, Eds.; Elsevier: Amsterdam, London, New York, Tokyo, 1994; pp 345-351.
566 567
(40) Proctor, A; Meullenet,J.F. Sampling and Sample Preparation. In Food Analysis Nielsen, S. S., Ed., 2 Ed.; Aspen publishers, 1998, pp 78-79.
568 569
(41) Schwedt, G. Taschenatlas der Analytik.; 3 ed.; Wiley-VCH Verlag GmbH & Co. KGaA, 2007, pp. 20-21
570 571 572
(42) Gunness, P.; Kravchuk, O.; Nottingham, S. M.; D'Arcy, B. R.; Gidley, M. J. Sensory analysis of individual strawberry fruit and comparison with instrumental analysis. Postharvest Biol. Technol. 2009, 52, 164-172.
573 574 575
(43) Vitten, M. D. Breeding Aspects of Freeze-dry Processing in Fragaria L. Dissertation Technical University Munich (TUM), 2008. Identifier: URN: urn:nbn:de:bvb:91-diss-20070914627666-1-7; http://nbn-resolving.de/urn:nbn:de:bvb:91-diss-20070914-627666-1-7
576 577
(44) Ulrich, D.; Olbricht, K. Diversity of metabolite patterns and sensory characters in wild and cultivated strawberries. J. Berry Res. 2014, 4, 11-17.
578 579 580
(45) Hoberg, E.; Ulrich, D. Influence of freezing on flavour and healthy value of strawberry. Proceedings of the Deutsche Gesellschaft fur Qualitatsforschung (Pflanzliche Nahrungsmittel) DGQ e. V. XXXII. Vortragstagung, Wädenswil 1997, 127-138.
581 582
(46) Krumbein, A.; Ulrich, D. Assessment of flavour volatiles in vegetables - from comprehensive to rapid methods. Acta Hortic. 2003, 604, 123-142.
583 584
(47) Baietto, M.; Wilson, A. D. Electronic-Nose Applications for Fruit Identification, Ripeness and Quality Grading. Sensors 2015, 15, 899-931. 21 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 22 of 36
585 586 587
(48) Krumbein, A.; Ulrich, D. Comparison of three sample preparation techniques for the determination of fresh tomato aroma volatiles. In Flavour Science. Recent Developments., Taylor, A. J.; Mottram, A. S., Eds.; The Royal Society of Chemistry, Cambridge: 1996; pp 289-292.
588 589 590
(49) Ulrich, D.; Krumbein, A.; Schonhof, I.; Hoberg, E. Comparison of two sample preparation techniques for sniffing experiments with broccoli (Brassica oleracea var. italica Plenck). NahrungFood 1998, 42, 392-394.
591 592
(50) Komes, D.; Ulrich, D.; Lovric, T.; Schippel, K. Isolation of white wine volatiles using different sample preparation methods. Vitis 2005, 44, 187-193.
593 594 595
(51) Blank, I.; Fay, L. B. Formation of 4-hydroxy-2,5-dimethyl-3(2H)-furanone and 4-hydroxy2(or 5)-ethyl-5(or 2)-methyl-3(2H)-furanone through Maillard reaction based on pentose sugars. J. Agric. Food Chem. 1996, 44, 531-536.
596 597
(52) Shu, C. K.; Mookherjee, B. D.; Ho, C. T. Volatile Components of the Thermal-Degradation of 2,5-Dimethyl-4-Hydroxy-3 (H-2)-Furanone. J. Agric. Food Chem. 1985, 33, 446-448.
598 599 600 601
(53) Fiehn, O.; Robertson, D.; Griffin, J.; van der Werf, M.; Nikolau, B.; Morrison, N.; Sumner, L. W.; Goodacre, R.; Hardy, N. W.; Taylor, C.; Fostel, J.; Kristal, B.; Kaddurah-Daouk, R.; Mendes, P.; van Ommen, B.; Lindon, J. C.; Sansone, S. A. The metabolomics standards initiative (MSI). Metabolomics 2007, 3, 175-178.
602 603 604 605
(54) Sansone, S. A.; Fan, T.; Goodacre, R.; Griffin, J. L.; Hardy, N. W.; Kaddurah-Daouk, R.; Kristal, B. S.; Lindon, J.; Mendes, P.; Morrison, N.; Nikolau, B.; Robertson, D.; Sumner, L. W.; Taylor, C.; van der Werf, M.; van Ommen, B.; Fiehn, O. The metabolomics standards initiative. Nat. Biotechnol. 2007, 25, 844-848.
606 607 608
(55) Manach, C.; Brennan, L.; Dragsted, L. O. Using metabolomics to evaluate food intake: applications in nutritional epidemiology. In Metabolomics as a Tool in Nutrition Research., Sebedio, J. L.; Brennan, L., Eds.; Elsevier: 2014; pp 167-196.
609 610 611
(56) Ferrara, M.; Sebedio, J. L. Challenges in nutritional metabolomics: from experimental design to interpretation of data sets. In Metabolomics as a Tool in Nutrition Research, Sebedio, J. L.; Brennan, L., Eds.; Elsevier: 2014; pp 3-16.
612 613 614
(57) Aharoni, A.; Giri, A. P.; Verstappen, A. P. G.; Bertea, C. M.; Sevenier, R.; Sun, Z.; Jongsma, M. A.; Schwab, W.; Bouwmeester; H.J. Gain and Loss of Fruit Flavor Compounds Produced by Wild and Cultivated Strawberry Species. The Plant Cell 2004, 16, 3110-3131.
615 616 617 618
(58) Rothe , M. Aroma values – a useful concept? Die Nahrung 1976, 20, 259-266. (59) Adams, A.; Van Lancker, F., De Meulenaer, B., Owzarek-Fendor, A.; De-Kimpe, N. On-fiber furan formation from volatile precursors: a critical example of artefact formation during Solid-Phase Microextraction. J Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2012, 897, 37-41.
619 620 621
22 ACS Paragon Plus Environment
Page 23 of 36
Journal of Agricultural and Food Chemistry
622 623
Figure captions
624 625
Figure 1: Cumulative sum of the number of VOC identified in strawberries.
626 627
Figure 2: Character impact compounds identified in strawberries using GC-O. Studies: A - Ulrich et
628
al.35; B - Schieberle and Hofmann34; C - Fukuhara et al.30; D - Nuzzi et al.28; E - Du et al.23; F - Cannon
629
et al.10; G – Ubeda et al.21. Red bars – tentatively or complete identified compounds.
630
23 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 24 of 36
Tables Table 1: Compilation of scientific aims for F. ×ananassa VOC measurements no ref study year aim 1 34 Schieberle & Hofmann 1997 Quantitation of selected key flavor compounds 2 35 Ulrich et al. 1997 VOC analysis for flavor breeding (key compounds) 3 33 Song et al. 1998 Test of SPME as rapid method 4 32 Gomes da Silva et al. 1999 VOC profile of ‘Oso Grande’ in comparison to ‘Selva’ and ‘Chandler’ (aroma properties) 5 31 Hakala et al. 2002 VOC profile of ‘Senga sengana’ in comparison to 5 others. Geographical origin, processing 6 30 Fukuhara et al. 2005 VOC profile of ‘Toyonoka’ by SPE 7 29 de Boishebert 2006 Characterization of varieties (SPME and data processing like Kohonen map) 8 28 Nuzzi et al. 2008 Comparison of GC-O with OAV 9 27 Jouquand et al. 2008 Eating quality and harvest date GxE (genotype & harvest date) 10 26 Zhang,YT et al. 2009 VOC profile comparison (aroma) 11 25 Li, et al. 2009 Estimation of key compounds in ‘Toyonoka’ 12 24 Zhang, JJ et al. 2010 Aroma development during maturation 13 23 Du et al. 2011 VOC-profiles of 2 varieties (aroma) 14 22 Ozcan &Barringer 2011 VOC-profiles depending on varieties, storage, ripening stages. SIFT-MS 15 20 Vandendriessche et al. 2012 SPME IMS and SPME-fastGC-MS 16 21 Ubeda et al. 2012 Study of glycosidic precursors and free aroma compounds 17 19 Vandendriessche et al. 2013 VOC-analysis for flavor breeding 18 17 N. N. 2013 Odour profiling with TOF-MS 19 16 Samykanno et al. 2013 GxE interaction on VOC. Flavor breeding and production 20 15 Ulrich & Olbricht 2013 Metabolic diversity for breeding 21 14 Mishra & Kar 2014 Quality changes during storage 22 10 Cannon et al. 2014 In-depht analysis of VOC, VSC 23 11 Schwieterman et al. 2014 VOC-analysis and acceptance for breeding 24 13 Oz et al. 2016 VOC-profiles of 8 cultivars (aroma) 25 12 Ulrich & Olbricht 2016 VOC-profiles and acceptance for breeding
24
ACS Paragon Plus Environment
Page 25 of 36
Journal of Agricultural and Food Chemistry
Table 2: Compilation of material and methods for F. ×ananassa VOC measurements no ref 1 34 2 35 3 33 4 32 5 31 6 30 7 29 8 28 9 27 10 26 11 25 12 24 13 23 14 22 15 20 16 21 17 19 18 17 19 16 20 15 21 14 23 11 22 10 24 13 25 12
study
year
material
storage
homogenisation
sample size
Schieberle & Hofmann Ulrich et al. Song et al. Gomes da Silva et al. Hakala et al. Fukuhara et al. de Boishebert Nuzzi et al. Jouquand et al. Zhang,YT et al. Li, et al. Zhang, JJ et al. Du et al. Ozcan & Barringer Vandendriessche et al. Ubeda et al. Vandendriessche et al. N. N. Samykanno et al. Ulrich & Olbricht Mishra & Kar Schwieterman et al. Cannon et al. Oz et al. Ulrich & Olbricht
1997 1997 1998 1999 2002 2005 2006 2008 2008 2009 2009 2011 2011 2011 2012 2012 2013 2013 2013 2013 2014 2014 2015 2016 2016
1 unknown cultivar 3 cultivars (and 1 F. vesca accession) unknown cultivar 3 cultivars 7 cultivars 1 cultivar 14 cultivars and 8 breeding clones 4 cultivars and 2 breeding clones 3 cultivars and 5 breeding clones 4 cultivars 1 cultivar 1 cultivar 2 cultivars unknown cultivar 1 cultivar 4 cultivars 1 cultivar 1 cultivar 2 cultivars 5 cultivars (and 16 F. vesca accessions) 2 cultivars 35 cultivars and 3 breeding clones 1 cultivar 8 Cultivars 10 cultivars and 6 breeding clones
fresh fresh fresh frozen frozen frozen frozen fresh fresh frozen frozen frozen fresh fresh fresh fresh fresh fresh frozen frozen fresh fresh fresh ? fresh
yes, with CaCl2 yes, with NaCl no, whole fruit yes yes yes yes quartering yes, with CaCl2 yes yes quartering yes, with NaCl/NaF yes, with SnCl2 no whole fruit yes yes, with NaCl whole fruit ? yes, with NaCl yes yes yes, with CaCl2 ? yes
1 berry to 500 g 200 g 100 g 100 g 200 g 50 g 100 g 1500 g 80 g 8.3 g ? 0.5 g 200 g 55 g 1 fruit 80 g ? 1 fruit 5 fruits 10 g to 300 g 10 000 g 7 fruits or 100 g 100 000 g ? > 1 000 g
VOC isolation LLE (diethyl ether) LLE (Freon) HS-SPME (DVB) P&T (Tenax) P&T (OV-1 and OV-25) SPE (Porapak Q) HS-SPME (DVB) P&T (Anasorb CSC) HS-SPME (DVB/Car/PDMS) HS(?)-SPME SPE (Porapak Q) LE4) HS-SPME (DVB/Car/PDMS) static-HS HS-SPME (DVB/Car/PDMS) P&T (Lichrolut EN) HS-SPME (DVB/Car/PDMS) P&T (Tenax) HS-SPME (PDMS/DVB) imm-SBSE (HS?)-SPME (PDMS) P&T (HaySep) SPE (silica) HS-SPME imm-SBSE
25
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 26 of 36
Table 2 (continued): Compilation of material and methods for strawberry VOC measurements no ref GC separation MS identification quantitation sum of VOC1) new VOC1) IT-MS 1, 2, 3 µg/kg by SIDA 15 0 1 34 polar and mid polar 2 Q-MS 1 SQ, IST 23 1 35 polar 3 TOF-MS 1, 2 SQ in % 31 5 33 mid polar polar and mid polar IT-MS 1 SQ in % 95 31 4 32 5 Q-MS 1, 2, (3) SQ, relative peak areas 39 3 31 mid polar Q-MS 1, 2, (3) no 48 9 6 30 polar 7 IT-MS 1, 2, (3) SQ, peak heights 23 1 29 mid polar 8 Q-MS 1, 2 EST2) 32 6 28 unpolar 9 Q-MS 1, 2 SQ, IST 69 7 27 mid polar 10 26 mid polar Q-MS 1, 2, (3) SQ in % 50 12 11 25 polar 20 1 Q-MS 1, 2, (3) µg/kg by EST5) 12 24 mid polar Q-MS 1, 2, (3) µg/g by IST5) 50 43 13 23 polar Q-MS 1, 2, (3) SQ, relative concentrations 54 5 14 22 no SIFT-MS 16) ppb on the basis of kinetic data 41 4 15 20 unpolar Q-MS 1, 2 SQ, absolute peak areas 97 22 16 21 polar IT-MS 1, 2, (3) µg/kg2) 28 7 17 19 unpolar Q-MS 1, 2 SQ, relative peak areas 62 3 18 17 mid-polar Q-MS and TOF-MS 1 SQ, peak areas and ‘approx. conc.’ in ng/g 74 14 19 16 unpolar and polar (2D) TOF-MS 1, 2, (3) SQ, relative peak areas 124 40 20 15 polar Q-MS 1, 2, (3 partly) SQ, peak areas 65 21 21 14 mid-polar Q-MS 1, 2, (3) SQ, IST 24 2 23 11 mid polar Q-MS 1, 2, (3) ng/gFW*h by EST5) 75 6 22 10 apolar and mid polar (2D) Q-MS 17) SQ, relative percentage 553 322 24 13 ? Q-MS ? SQ in % 63 23 25 12 polar Q-MS 1, 2, (3 partly) SQ, peak areas 64 6 Abbreviations: IT-MS: ion trap MS; Q-MS: quadrupol MS, SIFT-MS: flow tube MS; TOF-MS: time of flight MS: SQ: semi-quantitation. Identification: 1: MS library search, 2: retention indices from literature, 3: co-elution of authentic references with proof of origin of reference compounds. Number in brackets means that no details and/or no origin are reported. 1) Counting of substance numbers is in accordance with the JKI data base. These quantities sometimes differ from those which are given in the original publications.
26
ACS Paragon Plus Environment
Page 27 of 36
Journal of Agricultural and Food Chemistry
2)
No details or recovery rates are given. 3) ‘Troyonoko’ possibly is a scribal error of the Japanes cultivar ’Toyonaka’. 4) Freezing with liquid nitrogen and grinding following by liquid extraction of the frozen powder with petroleum ether and cyclohexane. 5) No recovery rates are given. 6) Identification on the basis of positive charged product ions. 7) Out of 563 VOC 6 new compounds were fully identified by synthesis and NMR characterization.
27
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 28 of 36
Table 3: Cultivars of F. ×ananassa used for VOC analysis no ref study year material 1 34 Schieberle & Hofmann 1997 unknown cultivar (Spain) from local market 2 35 Ulrich et al. 1997 Elsanta, Polka, Senga Gourmella (and 1 accession of F. vesca) 3 33 Song et al. 1998 unknown cultivar 4 32 Gomes da Silva et al. 1999 Chandler, Oso Grande, Selva 5 31 Hakala et al. 2002 Bounty, Honeoye1), Jonsok1), Korona, Polka1), Senga Sengana 6 30 Fukuhara et al. 2005 Toyonaka 7 29 de Boishebert 2006 Cal Giant3, Capitola, Ciflorette, Cifrance, Cigaline, Cigoulette, Cilady, Ciloe, Cireine, Darselect, Earliglow, Madeleine, Naiad, Pajaro and 8 breeding clones 8 28 Nuzzi et al. 2008 Alba, Darselect, Dora, Eva and 2 breeding clones 9 27 Jouquand et al. 2008 Festival, Rubigem, Sugarbaby and 5 breeding clones 10 26 Zhang,YT et al. 2009 Allstar, Toyonoka, Xingdu1, Xingdu2 11 25 Li, et al. 2009 Toyonaka 12 24 Zhang, JJ et al. 2010 Troyonoka2) 13 23 Du et al. 2011 Festival, Radiance 14 22 Ozcan 2011 Albion, Camarosa, Chandler, Sweet Charlie 15 20 Vandendriessche 2012 Elsanta 16 21 Ubeda 2012 Camarosa, Candonga, Fuentepina, Sabrina 17 19 Vandendriessche 2013 Charlotte 18 17 N. N. 2013 unknown cultivar 19 16 Samykanno 2013 Albion, Juliette 20 15 Ulrich & Olbricht 2013 Alba, Elegance, Frau Mieze Schindler, Mara de Bois, Polka (and 16 F. vesca accessions) 21 14 Mishra 2014 Camarosa, Chandler 22 10 Cannon 2014 Ciflorette 23 11 Schwieterman 2014 Albion, Benicia, Camarosa, Camino Real, Chandler, Charlotte, Darselect, Elyana, Evie2, Festival, Galetta, Mara des Bois, Mojave, Monterrey, Portola, Proprietary1, Proprietary2, Proprietary3, Proprietary4, Proprietary5, Proprietary6, Radiance, Red Merlin, Rubygem, San Andreas, Sweet Anne, Sweet Charlie, Treasure, Ventana, Winter Dawn, Winterstar and 3 breeding clones 24 13 Oz 2016 Albion, Camarosa, Festival, Fortuna, Rubygem, Sabrosa, Sweet Ann 25 12 Ulrich & Olbricht 2016 Clery, Daroyal, Elegance, Elianny, Elsanta, Evie2, Frau Mieze Schindler, Honeoye, Rumba, Sonata and 6 breeding clones 28
ACS Paragon Plus Environment
Page 29 of 36
1)
Journal of Agricultural and Food Chemistry
in addition also from organic cultivation, 2) scribal error of ‘Toyonoka’
29
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 30 of 36
Table 4: Most frequently identified VOC in strawberries # 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
substance methyl hexanoate ethyl hexanoate ethyl butanoate methyl butanoate linalool γ-decalactone hexyl acetate γ-dodecalactone DMMF (E)-2-hexenal butyl acetate DMHF ethyl 3-methylbutanoate ethyl 2-methylbutanoate hexanoaic acid methyl octanoate 2-methyl butanoic acid ethyl acetate hexanal butanoic acid (E)-2-hexen-1-ol ethyl octanoate butyl butanoate octyl butanoate 2-heptanone benzyl acetate (E)-2-hexenyl acetate methyl pentanoate pentyl acetate acetic acid
CAS 106-70-7 123-66-0 105-54-4 623-42-7 78-70-6 706-14-9 142-92-7 2305-05-7 4077-47-8 6728-26-3 123-86-4 3658-77-3 108-64-5 7452-79-1 142-62-1 111-11-5 116-53-0 141-78-6 66-25-1 107-92-6 928-95-0 106-32-1 109-21-7 110-39-4 110-43-0 140-11-4 2497-18-9 624-24-8 628-63-7 64-19-7
entries1 24 24 23 22 22 21 19 18 18 18 17 17 16 16 15 14 14 14 14 13 13 12 12 12 12 12 12 12 12 12
sum of entries in 25 studies from 1997 to 2016, in the review of Zabetakis (1997)8 and in the TNOdatabase9. DMMF - 2,5-dimethyl-4-methoxy-3(2H)-furanone , DMHF - 2,5-dimethyl-4-hydroxy3(2H)-furanone. 1)
30 ACS Paragon Plus Environment
Page 31 of 36
Journal of Agricultural and Food Chemistry
Figure 1
31 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 32 of 36
Figure 2
32 ACS Paragon Plus Environment
Page 33 of 36
Journal of Agricultural and Food Chemistry
TOC Graphic
33 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Progress of number of VOCs identified in strawberries. 254x190mm (96 x 96 DPI)
ACS Paragon Plus Environment
Page 34 of 36
Page 35 of 36
Journal of Agricultural and Food Chemistry
Character impact compounds identified in strawberries using GC-O. Studies: A - Ulrich et al. (1997)35; B Schieberle and Hofmann (1997)34; C - Fukuhara et al. (2005)30; D - Nuzzi et al. (2008)28; E - Du et al. (2011)23; F - Cannon et al. (2014)10; G – Ubeda et al. (2012)21. Red bars – tentatively or complete identified compounds. 190x254mm (96 x 96 DPI)
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
TOC Graphic 254x190mm (96 x 96 DPI)
ACS Paragon Plus Environment
Page 36 of 36
