Subscriber access provided by University of Newcastle, Australia
Article
Fast Separation and Sensitive Quantitation of Polymethoxylated Flavonoids in the Peels of Citrus Using UPLC-Q-TOF-MS Tian Tian Xing, Xi Juan Zhao, Yi Dan Zhang, and Yuan Fang Li J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.6b05821 • Publication Date (Web): 05 Mar 2017 Downloaded from http://pubs.acs.org on March 7, 2017
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 free 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 accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Agricultural and Food Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 33
Journal of Agricultural and Food Chemistry
Fast Separation and Sensitive Quantitation of Polymethoxylated Flavonoids in the Peels of Citrus Using UPLC-Q-TOF-MS Tian Tian Xing,a Xi Juan Zhao,b,* Yi Dan Zhanga and Yuan Fang Li a,* a
Key Laboratory of Luminescence and Real-Time Analytical Chemistry (Southwest
University), Ministry of Education, College of Chemistry and Chemical Engineering, Southwest University, Chongqing 400715, P R China b
College of Horticulture and Landscape Architecture, Southwest University, Key
Laboratory of Horticulture Science for Southern Mountainous Regions, Ministry of Education,
Chongqing 400715, P R China Corresponding Author: *
(ZH. X. J.) Phone: 86-23-68250483; Email:
[email protected] 1
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
1
Page 2 of 33
ABSTRACT
2
A rapid, sensitive and efficient ultra-performance liquid chromatography coupled
3
with quadrupole time-of-flight mass spectrometry (UPLC-Q-TOF-MS) method has
4
been developed to analyze polymethoxylated flavonoids (PMFs) in 14 Citrus peels,
5
including 7 Citrus reticulata (C. reticulata) and 7 Citrus Sinensis (C. sinensis). In this
6
study, fast separation can be achieved within 12 minutes and 42 PMFs have been
7
identified including 33 flavones and 9 flavanones. Most C. reticulata were shown to
8
contain more than 20 PMFs, except Guangxihongpisuanju (GX) containing only 12
9
PMFs. While most C. sinensis contained less than 20 PMFs, except Edangan (EG)
10
containing as many as 32 PMFs. To our knowledge, there are few reports about the
11
quantitation of PMFs using the MS response. Here, a MS quantitative method was
12
established and systematically validated in linearity, precision and recovery. The
13
linearity was from 1.25ng/mL to 1.0 µg/mL with the limit of detection (LOD) as low
14
as 75 pg/mL and the limit of quantitation (LOQ) as low as 0.25 ng/mL. Up to 13
15
PMFs more types than ever before were undoubtedly identified and quantitated
16
according to the PMF standards. The results showed that the contents of PMFs in the
17
C. reticulata were generally higher than that in the C. sinensis. This study is
18
systematic for analyzing PMFs and is of great significance because it can provide
19
guidance on utilization of both PMFs and citrus germplasm resources in future.
20
Keywords: PMFs, Citrus, UPLC-Q-TOF-MS
2
ACS Paragon Plus Environment
Page 3 of 33
Journal of Agricultural and Food Chemistry
21
INTRODUCTION
22
Polymethoxylated flavonoids (PMFs) refer to a group of natural products of
23
flavones attached with four or more methoxy groups on their basic 2-phenyl
24
chromone skeleton. They exist widely in Citrus fruits particularly in Citrus reticulata
25
(C. reticulata) and Citrus sinensis (C. sinensis) possessing a broad spectrum of
26
beneficial properties such as anticancer,1 anti-inflammatory,2 anti-allergic,3 and
27
antifungal activities.4 However, to the best of our knowledge, there have been few
28
systematic researches on the identification and quantitation of PMFs in the peels of
29
Citrus by MS until now.5-8 Therefore, it is of great importance to perform a systematic
30
analysis of PMFs in Citrus fruits.
31
Early reported methods for study of PMFs mainly used high-performance liquid
32
chromatography (HPLC) and HPLC coupled to photodiode array detection and
33
electrospray tandem mass spectrometry (HPLC-DAD–ESI-MS/MS). Curtis O. Green
34
et al.9 separated 6 major PMFs within 45 minutes in the peels of 20 Citrus cultivars,
35
which was carried out by RP-HPLC and UV detection. Jia-Yu Zhang et al.5
36
established an HPLC-DAD-ESI-MS/MS method to identify 32 PMFs in ‘Shatangju’
37
mandarin within 95 minutes. And Yu-Shan Lin et al.6 developed a method using
38
HPLC coupled with linear ion trap mass spectrometry, and quantitated 6 PMFs and 6
39
hydroxylated polymethoxyflavones (OH-PMFs) within 20 minutes in the peels of
40
Citrus. It is obvious that the previous methods needed a long time to perform the
41
analysis and some components could not be detected owing to the low abundance,
42
co-elution and high background of HPLC.10 Recent years, the emergence of UPLC 3
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 4 of 33
43
technology and high resolution mass spectrometry (HRMS) has brought great benefits
44
to the analysis work such as the fast separation, high sensitivity, less organic solvent
45
and accurate mass.11-13 Medina-Remon group developed a rapid method using
46
UHPLC-PDA to allow the simultaneous separation and quantitation of 11 selected
47
flavonoids in only 5.5 minutes,14 which displayed the advantages of UHPLC
48
including rapid, solvent-saving and small sample volume. HRMS can give more
49
specific and much exacter MS and MS2 information, making the identification more
50
accurate. UPLC coupled with quadrupole time-of-flight mass spectrometry
51
(UPLC-Q-TOF-MS) combines the advantages of both UPLC and HRMS effectively
52
and is shown to be a powerful tool to identify compounds in botanic extracts and
53
complex matrix. For example, Yang et al.15 used UPLC-Q-TOF-MS to achieve the
54
separation and identification of 32 chemical compositions within only seven minutes
55
in Ponkan peel.
56
The present study focuses on the identification and quantitation of PMFs in the
57
peels of Citrus including 7 C. reticulata and 7 C. sinensis and the comparison of
58
PMFs with respective to types and contents between them. In order to carry out the
59
objective, an effective UPLC-Q-TOF-MS method was established after a series of
60
parameters optimization and method validation in linearity, precision and accuracy.
61
This work is a systematic investigation on PMFs in Citrus peels and is important
62
because it can provide guidance on efficient utilization of PMFs and citrus germplasm
63
resources in future.
4
ACS Paragon Plus Environment
Page 5 of 33
Journal of Agricultural and Food Chemistry
64
MATERIALS AND METHODS
65
Chemicals 13 PMFs reference compounds (Table S1 in the Supporting Information),
66 67
including
5,
7,
68
purity≥98.0%),5,6,7,4’-tetramethoxyflavone
69
5,7,4’-trimethoxyflavone (S-5, purity≥95.0%), 3,5,6,7,8,3’,4’-heptamethoxyflavone
70
(S-8,
71
5-hydroxy-6,7,8,3’,4’-pentamethoxyflavone
72
5-hydroxy-7,3’,4’-trimethoxyflavone (S-11, purity≥98.0%) were purchased from
73
SinoStandards
74
(S-3,purity≥98.0%),
75
5,6,7,8,3’,4’-Hexamethoxyflavone (S-6, purity≥98.0%), 5,7,8,4’-tetramethoxyflavone
76
(S-7, purity≥98.0%), 5-Hydroxy-6,7,8,3’,4’,5’-Hexamethoxyflavone (S-12, RG),
77
5-hydroxy-6,7,8,4’- tetramethoxyflavone (S-13, purity≥98.0%) were purchased from
78
ChromaDex Inc. (Santa Ana, CA, USA). The structures of the 13 PMFs reference
79
standards were confirmed by MS (Figure S1-S13 in the Supporting Information).
purity≥98.0%),
8,
3’,
4’-pentamethoxyflavone (S-4,
5,6,7,8,4’-pentamethoxyflavone
(Chengdu,
China),
(S-10,
(S-1,
purity≥98.0%),
(S-9,
purity≥98.0%),
purity≥98.0%)
and
5,6,7,3’,4’-pentamethoxyflavone
5,7,3’,4’-tetramethoxyflavone
(S-2,
purity≥97.0%),
80
HPLC grade acetonitrile and formic acid were purchased from Sigma–Aldrich
81
(St. Louis, MO). Ultrapure water (Milli-Q, 18.2 MΩ) was obtained from a Millipore
82
System (Bedford, MA, USA). The filters (PTFE, 0.22µm) were purchased from
83
ANPEL Inc. (Shanghai, China).
84
Materials
85
The materials used in this study were collected on Nov. 20, 2015 from the 5
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 6 of 33
86
National Citrus Germplasm Repository at the commercial maturity stage and
87
identified by Dong Jiang who is a research associate in the Citrus Research Institute
88
of the Chinese Academy of Agricultural Sciences in Chongqing, China. There are 14
89
Citrus, including 7 C. sinensis and 7 C. reticulata (Table S2 in the Supporting
90
Information). Firstly, the peels of the collected citrus fruits were separated from the
91
fruits after washing with the clean water. Secondly, the peels were dried in an
92
electrothermal blowing drier at 40˚C for 48 hours. Then, the dried peels were crushed
93
and screened with a 40 mesh sieve. Finally, the dried powder was stored in a dryer for
94
further use.
95
Sample Preparation
96
0.5g of the powder was weighed accurately, and extracted using 7mL methanol
97
in 300W ultrasonic bath for 30minutes. Later, the extracts were centrifuged at 5000
98
rpm for 15minutes at room temperature and the supernatants were placed in a brown
99
volumetric flask. The above operation was repeated 3 times, and then the supernatants
100
were merged and constant-volumed to 25mL. The extracts were diluted with methanol
101
to a proper concentration and filtered with 13mm syringe filters with 0.22µm PTFE
102
membrane.
103
UPLC-PDA-ESI-MSE Analysis
104
UPLC analysis was performed on an ACQUITY UPLC I-Class (Waters, Milford,
105
MA), including the following components: Binary Solvent Manager (BSM), Sample
106
Manager-FTN (FTN), Photo-Diode Array Detector (PDA Detector) and column heater.
107
Samples were separated on an ACQUITY UPLC BEH C18 column (2.1х100 mm, 1.7 6
ACS Paragon Plus Environment
Page 7 of 33
Journal of Agricultural and Food Chemistry
108
mm, UK). The PDA detector scanned from 240 nm to 400 nm, and the peaks were
109
detected at 330 nm. To obtain the optimal condition, several parameters were
110
investigated such as the organic mobile phase of methanol or acetonitrile, the
111
concentration of formic acid in water including 0, 0.01%, 0.05%, 0.1% and 0.2%, the
112
flow rate of 0.2mL/min, 0.3mL/min, 0.4mL/min and 0.5mL/min and the
113
elution gradient. The optimal conditions are as follows. The binary mobile phases are
114
consisted of solvent A (water containing 0.01% formic acid) and solvent B
115
(acetonitrile), with the following gradient elution: 0-1 min, 25-30% B; 1-9 min, 30-45%
116
B; 9-11 min, 45-60 B; 11-12 min, 60-90% B; 12-14 min; 90% B; 14-16 min; 90-25%
117
B. The data acquisition was finished in the first 12 minutes, and the following 4
118
minutes were used to wash and balance the column. The column temperature was
119
settled at 40˚C. The flowing rate was 0.4mL/min and the injection volume was 1.0µL.
120
Xevo G2-S Q-TOF (Waters MS Technologies, Manchester, UK) equipped with
121
electrospray ionization ion (ESI) source was applied for mass spectrometry data’s
122
acquisition in positive ionization mode ranged from m/z 100 to m/z 1200. The
123
LockSpray™ dual electrospray ion source was used to insure the authenticated exact
124
mass measurement. The lock mass compound is Leucine-enkephalin (m/z 556.2766,
125
concentration: 200 ng/mL, flow rate: 10 µL/min) as the reference. The ion scan mode
126
adopted the MSE methodology to obtain alternating MS and MS/MS spectra. The
127
parameters of the source were set as follows: electrospray capillary voltage 1.0 kV for
128
positive ionization mode; cone voltage 40 V; source temperature 120˚C; desolvation
129
temperature 400˚C; low energy 6V, high energy ramp from 20V to 40V; nitrogen and 7
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 8 of 33
130
argon were used as the nebulizer and the collision gas, respectively; cone gas flow 50
131
L/h and desolvation gas flow 800 L/h, respectively.
132
Method Validation
133
The 13 reference standards were accurately weighed, dissolved in methanol and
134
dimethyl sulphoxide (DMSO), and then diluted to a proper concentration. Calibration
135
curves were fitted by the MS response at least 9 appropriate concentrations in
136
triplicate of each compounds. The limit of detection (LOD) and limit of quantitation
137
(LOQ) were determined at 3 and 10 times of the basic of signal-to-noise (S/N) ratios,
138
respectively. In addition, the inter- and intraday precision were investigated by
139
analyzing a proper calibration sample during a single day and on three successive
140
days, respectively. The recovery experiments were carried out to insure the precision
141
further by spiking the authentic standards in samples directly.
142
RESULTS AND DISCUSSION
143
Optimization of UPLC Analysis Conditions
144
To obtain a better separation efficiency, several parameters were investigated
145
such as the types of mobile phase, the concentration of formic acid in aqueous
146
solution, the flow rate of the mobile phase and the elution gradient. The optimized
147
aqueous phase A is water containing 0.01% formic acid, and the organic phase B is
148
acetonitrile with the flow rate of 0.4mL/min. An optimal gradient program was
149
applied in the elution: 0-1 min, 25-30% B; 1-9 min, 30-45% B; 9-11 min, 45-60 B;
150
11-12 min, 60-90% B; 12-14 min; 90% B; 14-16 min; 90-25% B.
8
ACS Paragon Plus Environment
Page 9 of 33
Journal of Agricultural and Food Chemistry
151
Identification of PMFs
152
PMFs are based on a core aglycone structure that is modified by different numbers
153
and/or positions of methoxyl/hydroxyl substituents. For polymethoxylated flavones,
154
they have the same basic flavone structure as shown in Figure 1 with the molecular
155
weight of 222 Da. And the molecular weights of polymethoxylated flavones can be
156
calculated in advance by adding n×30 and/or n×16. The same principle also applies to
157
polymethoxylated flavanones and chalcones with the basic structures of flavanone and
158
chalcones, respectively (Figure 1). According to the numbers and types of substituent
159
groups, either methoxyl groups or hydroxyl groups, we can designate the chemical
160
formula of every possible PMF isomer and know the corresponding molecular weight,
161
which is very helpful to identify the structures of PMFs in the complex extracts of
162
Citrus. Here, with the optimized UPLC condition, the PMFs in methanolic extracts of
163
14 Citrus peels are well separated in a short analytical time within 12 minutes (Figure 2
164
and Figure S14-S25 in the Supporting Information). Figure 2A shows the base peak
165
ion (BPI) chromatogram of methanolic extracts of Jiaogan (JG), which is a C. reticulata,
166
and 30 PMFs have been identified. Figure 2B displays the BPI chromatogram of
167
8045Tiancheng (8045), a C. sinensis, and 20 PMFs have been identified in it. Actually,
168
there are up to 42 PMFs that have been identified in the peels of 14 Citrus, including
169
33 flavones and 9 flavanones as shown in Table 1. Among them, 13 PMFs were
170
identified undoubtedly compared with the standards (Figure 2C), and the other 29
171
PMFs were identified tentatively according to the MSE fragments and references
172
(Table 1 and Figure S26-S67 in the Supporting Information). 9
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
173
Page 10 of 33
For polymethoxylated flavones, they tend to lose nCH3• preferentially, yielding +
174
the characteristic fragments of [M+H-n×15]
175
fragments such as neutral loss of CH4(16), H2O(18), CO(28), CH4+CH3•(31),
176
H2O+CH3•(33), CO+CH3•(43), CO2(44), H2O+CO(44), and CO+H2O+CH3•(59) from
177
[M+H-n×15]
178
the quasi-molecular ion at m/z 403.1390 was observed by HRMS at the low energy
179
mode (Figure 3A), and then the molecular formula C21H22O8 was obtained by element
180
matching. Next, the compound was identified as a hexamethoxyflavone through
181
comparing with the basic flavone structure (Figure 1). In addition, prominent ions at
182
m/z 373.0919 [M+H-2CH3] + and m/z 388.1156 [M+H-CH3] + by loss of 30 Da (CH3)
183
and 15 Da (2CH3) from the quasi-molecular ion [M+H]
184
characteristic ions of polymethoxylated flavones at the high energy mode as shown in
185
Figure 3B. Further, turn attention to the relative low abundance ions which underwent
186
the fragmentation pathways of Retro-Diels-Alder (RDA) cleavage of C-ring. The ions
187
at m/z 211.0240 [1,3A+-2CH3] and m/z 163.0755 [1,3B+] indicated that there were three
188
methoxy substituents at the ring A and two at the ring B. Finally, compare the
189
retention time, UV absorption and MSE spectra with that of the standard. Based on the
190
above analysis, the compound 27 was identified as nobiletin (Table 1).
+
usually as the base peaks. And other
can also be detected frequently.16 Take nobiletin as an example. First,
+
were regarded as the
191
As for polymethoxylated flavanones, they preferentially underwent the
192
fragmentation pathways of Retro-Diels-Alder (RDA) cleavage from the 1, 3 or 1, 4
193
position of the C-ring producing the prominent ions
194
loss of 15 Da (CH3), 28 Da (CO), 30 Da (2CH3) and 33 Da (CH3+H2O) from the base
1,3or4
A+ and
1,3or4 +
B . And the ions
10
ACS Paragon Plus Environment
Page 11 of 33
Journal of Agricultural and Food Chemistry
195
peak 1,3A+ or 1,3B+ can be pointed as the characteristic peaks of flavanones.10 The ions
196
0or1, 2
197
replaced, although the abundance of them is very low. Similarly, the ions loss of 15
198
Da (CH3), 28 Da (CO), 30 Da (2CH3) from the base peak xB+ and yA+ can be pointed
199
as the characteristic peaks of chalcones. It is noteworthy that chalcones tend to retain
200
the ring-B as the base peak while flavanones retain the ring-A, and the maximum
201
absorption wavelength of flavanones is centered at about 320 nm, while that of
202
chalcones ranges from 330 to 370 nm.5, 10, 17, 18 Take compound 11 as an example.
203
Firstly, the quasi-molecular ion at m/z 375.1436 was obtained from the low energy
204
mode (Figure 4A). And the molecular formula C20H22O7 obtained through elements
205
match indicated that the compound 11 is a pentamethoxyflavanone or a
206
pentamethoxychalcones according to Figure 1. Then, the abundances of the fragment
207
at m/z 211.0601 (1,3A+), m/z 196.0366 (1,3A+-CH3) were 100% and 26.42% which
208
were regarded as the diagnostic ions of flavanones at the high energy mode (Figure
209
4B). Furthermore, the existence of m/z 211.0601(1,3A+) and m/z 191.0705 (1,4B+-H2O)
210
indicated that there are three methoxy substituents at the ring A and two at the ring B.
211
Finally, make the comparison with literatures about their UV absorption and MS
212
spectra.
213
4’-pentamethoxyflavanone (Table 1).
A+ and
0or1, 2
B+ are the key fragments to determine whether the position-3 is
Therefore,
compound
11
was
identified
as
5,
6,
7,
3’,
214
According to the method stated above, the identification work was carried out
215
between C. sinensis and C. reticulata and a total of 42 PMFs were detected. Among
216
them, the structures of 35 PMFs can be identified in comparison with reported 11
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 12 of 33
217
research and compounds 1, 4, 6, 8, 9, 10, 13, and 37 were detected in Citrus for the
218
first time (Figure S68 in the Supporting Information). According to the qualitative
219
analysis of the 14 Citrus, more types of PMFs in C. reticulata were identified than
220
that in C. sinensis. The base peak ion (BPI) currents of the 7 C. reticulata were almost
221
similar expect GX, and the BPI of all the 7 C. sinensis were almost similar expect EG.
222
Most of C. reticulata contained more than 20 PMFs, and among them, JG, NF and EJ
223
contained many types of PMFs up to 30 while GX contains the least types of 12 PMFs.
224
By contrast, most of C. sinensis contained less than 20 PMFs, except EG containing
225
as many as 32 PMFs. In addition, the structures of all the possible PMFs that have
226
been detected were summarized (Table S3 and Figure S68 in the Supporting
227
Information). In brief, a total of 24 OH-PMFs and 18 PMFs were identified according
228
to whether containing hydroxyl substituents or not. And through classification of the
229
basic structures, 33 polymethoxylated flavones and 9 polymethoxylated flavanones
230
were identified.
231
Validation of Analytical Method
232
As displayed in Table 2 and 3, all of 13 standards behaved good linearity
233
between concentrations and the corresponding MS responses with correlation
234
coefficients > 0.99. The linear range is from 1.25ng/mL to 1.0µg/mL with the LOD of
235
75pg/mL and the LOQ of 0.25ng/mL. The developed method showed good
236
repeatability with interday relative standard deviations (RSDs) < 2.79%, and intraday
237
RSDs 0.99, excellent LOD of 75pg/mL and LOQ of 0.25ng/mL, good
267
precision with interday RSDs< 2.79% and intraday RSDs