Ultraviolet B (UVB) photosensitivities of tea catechins and the relevant chemical conversions

Article abstract of DOI:10.3390/molecules21101345

Ultraviolet B (UVB) photosensitivities of eight catechins were screened. In both water and ethanol, epicatechin (EC, 575 μM) and catechin (C, 575M) exhibited low photostabilities under 6 h UVB radiation with the generation of yellow photoproducts, while other catechins (epigallocatechin gallate, epigallocatechin, epicatechin gallate, gallocatechingallate, gallocatechin, catechin gallate) were relatively UVB-insensitive. Photoisomerization and photolysis were two important UVB-induced reactions to EC whereas photolysis was the dominant reaction for C. The influencing factors of time (2-10 h), solvent (water, ethanol) and substrate concentration (71.875-1150μM) on UVB-induced chemical conversions of EC and C were investigated, and eight photoproducts were identified through ultra performance liquid chromatography-diode array detection-Tandem mass spectrometry (UPLC-DAD-MS/MS) and 1H nuclear magnetic resonance (1H-NMR analysis). Photolysis reaction involved two pathways, including radical reaction and photo-induced electron transfer reaction. The 2,2-diphenylpicrylhydrazyl (DPPH) scavenging abilities of eight catechins did not change upon 6 h UVB irradiation. EC and C are photosensitive catechins among eight catechins causing deep color.

Full text of DOI:10.3390/molecules21101345

molecules
Article
Ultraviolet B (UVB) Photosensitivities of Tea
Catechins and the Relevant Chemical Conversions
Meng Shi, Ying Nie, Xin-Qiang Zheng, Jian-Liang Lu, Yue-Rong Liang and Jian-Hui Ye *
Tea Research Institute, Zhejiang University, Hangzhou 310058, China; 21316161@zju.edu.cn (M.S.);
21416073@zju.edu.cn (Y.N.); xqzheng@zju.edu.cn (X.-Q.Z.); jllu@zju.edu.cn (J.-L.L.); yrliang@zju.edu.cn (Y.-R.L.)
* Correspondence: jx0515@163.com; Tel.: +86-571-8898-2704
Academic Editors: Declan P. Naughton and Derek J. McPhee
Received: 13 August 2016; Accepted: 6 October 2016; Published: 10 October 2016
Abstract: Ultraviolet B (UVB) photosensitivities of eight catechins were screened. In both water and
ethanol, epicatechin (EC, 575 µM) and catechin (C, 575 µM) exhibited low photostabilities under 6 h
UVB radiation with the generation of yellow photoproducts, while other catechins (epigallocatechin
gallate, epigallocatechin, epicatechin gallate, gallocatechingallate, gallocatechin, catechin gallate) were
relatively UVB-insensitive. Photoisomerization and photolysis were two important UVB-induced
reactions to EC whereas photolysis was the dominant reaction for C. The influencing factors of time
(2–10 h), solvent (water, ethanol) and substrate concentration (71.875–1150 µM) on UVB-induced
chemical conversions of EC and C were investigated, and eight photoproducts were identified
through ultra performance liquid chromatography-diode array detection-tandem mass spectrometry
(UPLC-DAD-MS/MS) and 1H nuclear magnetic resonance (1H-NMR analysis). Photolysis reaction
involved two pathways, including radical reaction and photo-induced electron transfer reaction.
The 2,2-diphenylpicrylhydrazyl (DPPH) scavenging abilities of eight catechins did not change upon
6 h UVB irradiation. EC and C are photosensitive catechins among eight catechins causing deep color.
Keywords: photostability; catechins; photoisomerization; photolysis; photoproducts
1. Introduction
Tea catechins are the major components of the tea polyphenols present in tea leaves, which
have been associated with many health benefits and functionalities such as anti-radical, anti-bacterial,
anticarcinogenic and anti-aging activities [
gallate (EGCg), ( )-epigallocatechin (EGC), (
their geometric epimers ( )-gallocatechingallate (GCg), (
(Cg), and ( )-catechin (C). In the past decades, the protective effects of green tea or tea extract
against Ultraviolet (UV)-induced skin damage and skin aging have been reported [ ], leading to the
growing interest in utilization of green tea or tea extract for cosmetics and skin creams [ ]. However,
1
,
2
−
]. They are mainly composed of (
)-epicatechin gallate (ECg), ( )-epicatechin(EC), and
)-gallocatechin (GC), ( )-catechin gallate
−
)-epigallocatechin
−
−
−
−
−
−
3–5
6–8
tea catechins or tea extracts inevitably turn brown during storage, which limits their application
not only in the cosmetic industry but also in the food industry due to the undesirable dark color.
The bioactivity changes of catechins upon browning draw great interest [9,10].
Illumination, temperature and pH value are important influential factors for non-enzymatic
browning of catechin-containing products [11]. The stabilities of catechins at different temperatures and
pH values have been excessively studied [12–15]. Epimerization, hydrolysis and oxidation/condensation
reactions are three important reactions accounting for instability or degradation of catechins with the
generation of epimer, gallic acid (GA) and non-gallated catechin moiety, as well as dark polymeric
compounds [12–15]. However, these may not represent the photo-induced chemical reactions of tea
catechins that are driven by an entirely different energy. Illumination is a nonnegligible environmental
Molecules 2016, 21, 1345; doi:10.3390/molecules21101345
www.mdpi.com/journal/molecules

Molecules 2016, 21, 1345
2 of 13
condition for storage, thus the photostabilities of tea catechins deserve a systematic investigation in order
to resolve the browning problem of tea catechins caused by illumination.
Ultraviolet B (UVB) is electromagnetic radiation with a wavelength ranging from 320 nm to
290 nm, and is regarded as the most destructive natural radiation on account of its energy, which is high
enough to cause many photochemical reactions [16]. UVB radiation has been used to investigate the
photosensitivity of natural polyphenols due to the advantages of its constant and adjustable intensity,
easy control and simplicity compared with solar radiation [17
such as apigenin, quercetin, resveratrol, EGCg, C and EC have been reported to be susceptible to
photo-oxidation in the presence or absence of exogenous photosensitizers/free radical initiators [17 22].
Severe photodegradation of EGCg in topical cream was reported with the degradation percentage being
~70% [ 23], and 83% of C (950 M, methanol) was lost after 3 h of ultraviolet C (UVC) exposure [24].
–19]. Various phenolic compounds,
–
9,
µ
Phenoxyl radicals were produced from the catechol or resorcinol rings of C by photo-oxidation, leading
to further chemical evolution [25]. For the eight catechins, substituent groups may impact their
chemical properties, resulting in various responses to UVB radiation and different contributions to
browning process. However, relevant information on the photosensitivities of the eight catechins
is insufficient.
In the present work, the photostabilities of eight catechins under UVB radiation were investigated,
and the UVB-susceptible catechins were selected for further studies in terms of the effects of solvent,
radiation time and substrate concentration. Photoproducts were identified by ultra performance
liquid chromatography-diode array detection-tandem mass spectrometry (UPLC-DAD-MS/MS) and
1H nuclear magnetic resonance (NMR), and the reaction mechanisms of UVB-induced transformations
were discussed. The impact of UVB radiation on the antioxidant activities of the eight catechins was
evaluated by 2,2-diphenylpicrylhydrazyl (DPPH) assay.
2. Results and Discussion
2.1. The Photostabilities of Catechins under UVB Radiation
One molecule of catechin converts to one molecule of its epimer through an epimerization
reaction [26], and one molecule of gallated catechins produces one molecule of GA via hydrolysis of
the ester bond [27]. The percentage of conversion of catechins due to epimerization and hydrolysis
reactions was calculated by the equation in Table 1. The percentage of catechin loss due to other
non-identified reactions was calculated by subtracting the percentage of epimerization and the
percentage of hydrolysis from 100%, and was termed other percentage degradation for representation.
Table 1 shows the photostabilities of individual catechins in both water and ethanol. After 6 h of UVB
radiation, no GA was detected in any of the catechin samples, indicating that no hydrolysis reaction
occurred under UVB radiation. Therefore the GA data was not presented in the following studies.
In Table 1, GC, EGC, GCg, EGCg, Cg and ECg were insensitive to UVB irradiation with percentage
maintained of 90.7%–99.6%, whereas only 58.0%–80.3% of EC and C were preserved after 6 h UVB
radiation. A decrease below 10% is considered as insensitivity to UVB. Hence, EC and C both in water
and ethanol showed high photosensitivities to UVB, which was also reflected by their color change
from colorless to yellow under UVB radiation (Table 1). Epimers were only detected in the aqueous
and ethanol solutions of EC and C but not the other six catechins. For EC, 31.1% and 20.0% converted
to its epimer C in water and ethanol respectively, compared with 2.6% and 5.2% of C converting
to EC accordingly, which suggests that EC and C undergo isomerization under UVB radiation and
that the epi-type catechin is favorable for photoisomerization. Forest et al. [28] also reported that
photoisomerization occurred to C with the generation of unknown yellow compounds as side products.
The sum of the percentage of epimerization and the percentage maintained of EC and C in both water
and ethanol were 64.2%–90.4% (Table 1), indicating that other reactions played an important part in the
UVB-induced chemical transformations of EC and C in addition to photoisomerization. Specifically,
the unidentified reactions respectively cost 9.6% and 22.0% of EC, and 17.1% and 35.8% of C in water

Molecules 2016, 21, 1345
3 of 13
and ethanol, suggesting that C was more susceptible to the unknown UVB-induced reactions. EC and
C showed the highest photosensitivities among the eight catechins and were therefore selected for
further study.
Table 1. The photostabilities of eight catechins under ultraviolet B (UVB) radiation (µM) 1.
Catechins 2
Molecular Structures
After UV Radiation 3
Compositions
Water 4
Ethanol 4
OH
341.0 ± 6.3
333.7 ± 1.1
3'
OH
EC
2'
4'
1
(59.3 ± 3.1%)
(58.0 ± 0.2%)
EC
B
1'
O
HO
5'
2
6'
A
C
178.5 ± 3.6
115.0 ± 0.7
Epimer
C
OH
3
4
(31.1 ± 1.2%)
(20.0 ± 0.1%)
OH
OH
461.5 ± 4.5
339.2 ± 3.4
OH
(80.3 ± 0.0%)
(59.0 ± 0.6%)
C
EGC
GC
O
O
O
HO
HO
HO
14.9 ± 0.1
30.0 ± 0.5
Epimer
EGC
OH
(2.6 ± 0.0%)
(5.2 ± 0.1%)
OH
OH
OH
OH
531.6 ± 9.4
521.6 ± 7.1
OH
OH
(92.5 ± 1.6%)
(90.7 ± 1.2%)
Epimer
GC
UD
UD
OH
OH
557.5 ± 3.3
546.7 ± 1.8
OH
OH
(97.0 ± 3.2%)
(95.1 ± 0.3%)
Epimer
ECg
UD
UD
OH
OH
567.7 ± 1.2
544.1 ± 4.1
OH
ECg
Cg
(98.7 ± 0.2%)
(94.6 ± 0.7%)
O
HO
OH
O
C
OH
Epimer
Cg
UD
UD
OH
O
OH
OH
569.8 ± 3.0
536.9 ± 3.8
OH
(99.1 ± 0.5%)
(93.4 ± 0.7%)
O
HO
OH
OH
O
C
O
OH
Epimer
EGCg
Epimer
GCg
UD
UD
OH
OH
572.4 ± 2.6
525.3 ± 0.7
OH
OH
EGCg
(99.6 ± 0.5%)
(91.4 ± 0.1%)
O
HO
OH
OH
O
C
O
OH
UD
UD
OH
OH
566.2 ± 0.0
524.0 ± 2.8
OH
OH
GCg
(98.5 ± 0.0%)
(91.1 ± 0.5%)
O
HO
OH
OH
O
C
O
OH
Epimer
UD
UD
OH
1
The solutions of individual catechins were prepared at 575
EC-C, EGC-GC, ECg-Cg, EGCg-GCg are geometric isomers; Pictures of the water and ethanol solutions
µ
M with water and ethanol respectively;
2
3
4
of individual catechins after 6 h UVB radiation: W-water; E-ethanol; Data in round blankets were the
maintained percentage of catechins. Data in square blankets were the percentages of epimerization of
individual catechins = concentration of epimer (
EGC, ( )-epigallocatechin; GC, ( )-gallocatechin; ECg, (
EGCg, (−)-epigallocatechin gallate; GCg, (−)-gallocatechin gallate; UD, undetectable.
µ
M)/575
µ
−
M
×
100%. EC, (
−
)-epicatechin; C, (
−
)-catechin;
−
−
)-epicatechin gallate; Cg, (
−
)-catechin gallate;
In our study, EC and C, the basic unit of flavan-3-ols, exhibited high susceptibility to UVB
radiation whereas EGC, GC, EGCg, GCg, ECg and Cg were UVB-insensitive, indicating that the
presence of gallated and pyrogallol moieties hindered the UVB-induced chemical transformations.
This is in accordance with the previous report showing that epimerization reaction was closely
related with chemical structure of flavan-3-ols [29]. Dobashi et al. [17] demonstrated that EC and C
showed higher photo-antioxidant activity than EGC whereas no photo-antioxidant activity of gallated
catechins was observed. Hence, a substituent group has a decisive influence on the photosensitivity of
flavan-3-ols. In addition, a clear solvation effect on the photoisomerization of EC and C was observed
(Table 1). An aqueous solution was favorable to photoisomerization. It was reported that UVB-induced
epimerization of EC to C was via an intermediate quinone methide through ionization [20]. Thus,
higher polarity of water facilitated the ionization process, increasing the generation of intermediate
quinone methide compared with ethanol medium. This explained why more EC converted to C in

Molecules 2016, 21, 1345
4 of 13
water than ethanol. Solvation effect on epimerization reaction of catechins was also reported in other
studies [29,30]. Solvation effect on the unidentified reaction was discussed in a later section.
2.2. The Chemical Transformations of EC and C with UV B Radiation Time
There are different types of reactions occurring to EC and C under UVB radiation, which might
play differently with radiation time. Figure 1 shows the changes in percentages of epimerization and
percentages maintained of EC and C over time. The corresponding percentages of degradation due to
other reactions can be obtained by subtracting the percentage of epimerization and the percentage
maintained from 100%. Upon 2 h of UVB irradiation, 15.4% and 0.5% of EC in water underwent
epimerization and degradation reactions respectively, compared with 17.2% and 15.7% of EC in ethanol
(Figure 1). Upon 2 h of UVB radiation, the percentage of epimerization and percentage of degradation
of C were 1.7% and 8.8% in water, and 1.7% and 22.7% in ethanol (Figure 1). These results indicate that
epimerization was the dominant reaction for EC in both water and ethanol for the first 2 h of UVB
radiation, and other degradation reactions were the principal reactions to C under the same conditions.
As radiation time increased, the percentage of epimerization of EC in water increased rapidly after the
first 8 h and then maintained stable at ~35.4% from 8 h to 10 h, while the percentage of degradation of
EC slowly increased from 0.5% to 9.6% after the first 6 h and then rapidly increased from 9.6% to 29.9%
in the subsequent 4 h (Figure 1A). This suggests an antagonistic relationship between percentage of
epimerization and percentage of degradation of EC in water, which is possibly due to the competition
for UVB energy between epimerization and degradation reactions. The percentages of degradation
of EC and C in ethanol gradually increased with radiation time, and were much higher than those
in water. Thus, ethanol medium exacerbated the photo-degradation of EC and C. The 6 h point was
selected for the following studies on account of the high percentage of conversion in both epimerization
and degradation reactions.
Percentage maintained of EC
Percentage maintained of C
Percentage of epimerization
Percentage maintained of EC
Percentage maintained of C
Percentage of epimerization
100
80
60
40
20
0
B
100
80
60
40
20
0
A
1
2
3
4
5
1
2
3
4
5
8
4
10
2
4
8
10
2
6
6
Time (h)
Time (h)
Figure 1. The changes in percentage of epimerization and the percentage maintained of EC and C with
UVB radiation time: (A) water, 575 µM; (B) ethanol, 575 µM.
2.3. The Effect of Substrate Concentrations of EC and C
Figure 2 shows the effect of substrate concentration on percentage of epimerization and percentage
maintained of EC and C in water and ethanol. As substrate concentration increased from 71.875 to
1150
µM, the percentage of epimerization and percentage of degradation of EC and C in both
water and ethanol decreased and resulted in an elevated percentage of maintained EC and C
at high substrate concentration. This indicates that high substrate concentration suppressed the
UVB-induced chemical transformations of EC and C. From Figure 2, we can see that the solvent
exerted a great influence on the chemical conversions of EC and C under UVB radiation. At the
substrate concentration of 71.875–1150 µM, the percentages of epimerization of EC were 29.0%–52.7%
for water and 5.8%–33.2% for ethanol, while the percentages of degradation of EC in water and ethanol
were 7%–20.2% and 1.5%–34.5% respectively. Hence, in contrast to water, ethanol medium suppressed

Molecules 2016, 21, 1345
5 of 13
the photoisomerization of EC but increased the percentages of degradation of EC at low substrate
concentration (71.875 M to 575 M). The percentages of epimerization of C in water and ethanol were
µ
µ
less than 10.4%, while the percentages of degradation of C were 8.4%–20.4% for water and 12.4%–55.3%
for ethanol, hence other reaction was the dominating reaction to C upon UVB irradiation (Figure 2).
This was consistent with previous result showing that epi-structure is favorable for photoisomerization
of catechin.
Percentage maintained of EC
Percentage maintained of C
Percentage of epimerization
Percentage maintained of EC
Percentage maintained of C
Percentage of epimerization
100
80
60
40
20
0
A
B100
80
60
40
20
0
1
2
4
5
2837.5 575
1150
1
2
3
4
5
71.875
143.75
71.875 143.75
575 1150
287.5
Concentration (µM)
Concentration (µM)
Figure 2. The effect of substrate concentration on the percentage of epimerization and percentage
maintained of EC and C under 6 h UVB radiation: (A) water and (B) ethanol.
2.4. Structural Characterization of Photoproducts
Based on UPLC-DAD-MS analysis, the UVB irradiated aqueous and ethanol solutions of EC
and C had the same types of photoproducts. The UPLC chromatogram and MS information of
concentrated EC sample (287.5 µM in ethanol, 6 h UVB radiation) has been presented in Figure 3 as an
example. Except for the typical peaks corresponding to EC (P4, [M
−
H]−1 m/z 289) and its epimer
C (P3, [M − H]−1 m/z 289), seven new peaks for P1, P2, P5–9 were detected with [M
−
H]−1 m/z at
137, 287, 335, 427, 427, 427, 427 (in order), suggesting that new photoproducts were generated under
UVB radiation due to the unidentified reactions. The photolytic cleavage of flavan-3-ols (fisetinidol
and C) in methanol occurred at the ether linkage C-O of the heterocyclic ring with production
of the corresponding diradicals. These diradicals underwent two pathways: radical reaction and
photo-induced electron transfer reaction, which respectively led to the generation of ortho-quinone
methides and 1,3-diarylpropan-2-ols with an addition reaction of methanol to the diradical [18]. In the
case of EC (molecular weight abbreviated Mr, 290 Da) as reactant and ethanol as medium (Mr 46 Da),
theoretically the molecular weights of ortho-quinone methide-like and 1,3-diarylpropan-2-ols-like
products should be 138 Da and 336 Da, which was observed in Figure 3 corresponding to P1 and
P5 with [M
−
H]−1 m/z 137 and m/z 335 respectively. Compound P2 at [M
−
H]−1 m/z 287 was
considered as quinone on account of the typical transformation from phenol to quinone with a loss of
2H [31], and this was affirmed later by its UV adsorption spectrum. New types of photoproducts P6-9
were found with [M
the sum of 138 and 290 Da, suggesting that these four compounds are possibly the coupling products
of the biradical of EC (Mr 290 Da) and the radical precursor of ortho-quinone methide (Mr 138 Da).
Hence, a daughter ion scan MS experiment with a mass of parent ion set at 427 m/z was carried out
to further investigate the molecular structures of Compounds P6-9, and their MS/MS spectra are
shown in the inset of Figure 3. Obviously, Compounds P6-9 were isomers on account of the same
−
H]−1 m/z 427 in Figure 3. Coincidently, the Mr of new compounds (428 Da) was
[M
−
H]−1 m/z 427 and similar MS/MS spectra. Based on our speculation that Compounds P6-9
might be formed from addition of ortho-quinone methide-like compound to the biradical of EC, the
proposed fragmentation pathway of isomers P6-9 is shown in Figure 4, which was in an agreement
with the MS/MS spectrum. The presence of one double bond and one chiral carbon in the molecule is
the reason for the existence of four isomers.

Molecules 2016, 21, 1345
6 of 13
Figure 3. The ultra performance liquid chromatography (UPLC) and mass spectrometry (MS)
information of UVB irradiated EC in ethanol after concentration. Inset P6-9: the MS/MS spectra
of photoproducts assigned to P6-9.
OH
OH
OH
OH
-
OH
-
-
HO
HO
CH CH
- H
HO
- H
- H
OH
OH
OH
OH
OH
HO
HO
m/z 125
OH
-
OH
OH
HO
m/z 301
m/z 275
OH
OH
OH
OH
- H
HO
m/z 427
-
OH
OH
-
OH
OH
- H
HO
HO
- H
O
OH
OH
m/z 137
m/z 289
−1
Figure 4. Proposed fragmentation pathway of isomeric ions at [M − H] m/z 427.
The UV-visible spectra (190–400 nm) of all compounds are shown in Figure 5. The UV-visible
spectra of Compounds P1 and P2 exhibited two strong absorption peaks at 280 nm/307 nm and
250 nm/283 nm respectively, which was distinctly different from the maximum absorbance at ~280 nm
of Compounds P3-9 due to aromatic ring. The para-Quinone methides have typical UV absorbance at
300 nm and benzoquinone has strong absorbance at 244 nm [32,33]. The UV absorption spectra also
confirmed that Compounds P1 and P2 belonged to quinone methides and quinone respectively, due
to the possession of characteristic absorption bands near 300 nm and 244 nm. The less than 10 nm
difference might be attributable to the substitution of hydroxyl groups on aromatic ring leading to a
shift of λmax value to longer wavelength (bathochromic effect) because of electron donating effect of
hydroxyls. In addition, ortho-Quinone methides and quinone are responsible for the yellow color of
UVB irradiated EC sample, since quinonoid compounds are often yellow to red in color [34].

Molecules 2016, 21, 1345
7 of 13
Figure 5. The ultraviolet (UV) spectra of Compounds P1-9.
The 1H-NMR spectra of EC and its UVB irradiated reaction mixture with higher percentages of
conversion were studied to resolve the characteristic 1H frequency bands of new photoproducts from
EC for further testifying the chemical structures of photoproducts. Since nine compounds are present
in the UVB irradiated reaction mixtures including four isomeric products and several products at a
low level, it is not practical to isolate each photoproduct. The NMR spectra data of EC is as follows:

Molecules 2016, 21, 1345
8 of 13
1H-NMR of EC (600 MHz, Acetone-d6)
δ
ppm: 8.10 (s, H, H-3’), 7.94 (s, H, OH-5), 7.79 (s, H, OH-4’),
7.73 (s, H, OH-7), 7.05 (d, 1H, H-2’, J2’,6’ = 1.7 Hz), 6.83 (dd, 1H, H-6’, J6’,2’ = 1.7 Hz,
J6’,5’ = 8.3 Hz), 6.78
(d, 1H, H-5’, J5’,6’ = 8.3 Hz), 6.02 (d, 1H, H-6, J6,8 = 2.3 Hz), 5.91 (d, 1H, H-8, J8,6 = 2.3 H), 4.88 (br, s,
1H, H-2), 4.21 (m, 1H, H-3), 3.58 (d, 1H, OH-3, JOH-3,3 = 5.5 Hz), 2.86 (dd, 1H, H-4a,
J
4a,4b = 16.6 Hz
,
1
J4a,3 = 4.7 Hz), 2.74 (dd, 1H, H-4b,
consistent with reported results [15
the 1H-NMR spectra of reaction mixture from EC in addition to the original bands of EC, including
1.12 ppm (t, J = 6.8 Hz) corresponding to methyl proton of characteristic branch -O-CH2-C 3 in new
Compound 5, 2.91 ppm (dd, J = 5.4 Hz, J = 16.1 Hz) corresponding to methylene proton of characteristic
branch -O-C 2-CH3 in new Compound P5, 6.89 ppm (d) and 6.85 ppm (d) corresponding to protons
J
4b,4a = 16.6 Hz, J4b,3 = 3.2 Hz). The H-NMR data of EC was
,
35]. After UVB radiation, new 1H frequency bands were found in
H
H
on the carbon-carbon double bond in new Compounds P6-9. This indicated the presence of speculated
functional group or structure in the UVB irradiated reaction mixture of EC.
2.5. Proposed Reaction Mechanisms
The photolytic reaction mechanism of EC/C is given in Figure 6, referring to the reported
two photolysis pathways of flavan-3-ols [18]. Upon UVB irradiation, EC/C molecule was excited at
two positions: -OH bond and heterocyclic ring. The excitation of -OH bond led to the generation of
quinone compound P2 at [M
−
H]−1 m/z 287, while the heterocyclic ring of EC/C was preferentially
opened via photolytic cleavage at the ether linkage C-O with low bond dissociation energies, which
resulted in the generation of free radicals A and B at m/z 137 and 289. Free radical reaction led to the
generation of quinone methides such as Compound P1 at m/z 137 and four isomers P6-9 at m/z 427
through grafting reaction of free radical A m/z 137 onto free radical B m/z 289 after intramolecular
rearrangement of H and -OH (Figure 6). The phloroglucinol grafted derivatives of EC and C were
also photosynthesized by Wilhelm-Mouton et al. [36] via photolytic cleavage of the ether bond on
the heterocyclic ring of flavan-3-ols. Besides, neutral radicals can also be ionized in a polar solvent
due to photo-induced electron transfer reaction [37], which is the other pathway of photolysis in
competition with free radical reaction. The photo-induced electron transfer reaction product at m/z 335
was obtained in our study with an addition of -OCH2CH3 from ethanol, indicating the occurrence
of electron transfer reaction under UVB radiation. Thus, photolysis reaction was responsible for
the percentages of EC and C loss previously termed percentages of degradation. In the present
study, two reaction pathways of radical and photo-induced electron transfer reactions synchronously
occurred to UVB irradiated EC and C with the generation of three types of photoproducts, which
complemented the reported conclusion [18].
2.6. Antioxidant Activities
The antioxidant activities of individual catechins (575
µM, ethanol) determined by DPPH assay
were shown in Table 2. EGCg had the highest DPPH scavenging abilities; 2064
±
29 M trolox,
µ
followed by GCg, ECg, Cg, EGC, GC, EC and C in that order, which indicates that gallated catechins
possessed higher antioxidant capacities than non-gallated catechins and epi-type catechin was more
antioxidant active than the corresponding non-epi type at the same molality. This result was consistent
with the antioxidant activity rank of different catechins reported by Lee et al. [38]. Upon 6 h of UVB
irradiation, no significant difference in DPPH scavenging ability was observed for the eight catechins.
This was in line with previous result that no obvious photo-degradation was observed for catechins
except EC and C under 6 h UVB radiation. For EC and C, photoisomerization and photolysis reaction
may not significantly influence their antioxidant activities in the short-term but could be an important
cause of browning deterioration of catechins-containing products in color. Besides, a yellowish color
change was also observed in the aqueous solutions of EC and C after 15-day-storage under laboratory
illumination (Supplementary Materials Figure S1), indicating that UVB radiation is not a necessary
condition for photo-degradation of EC and C. Nevertheless, the photoproducts of EC and C under
laboratory illumination need identification. The photosensitivities of EC and C provide a new aspect

Molecules 2016, 21, 1345
9 of 13
for solving the browning problem of catechin-containing products via diminishing color deterioration
originating from EC and C.
O
OH
O
O
OH
O
O
O
HO
O
HO
HO
hv
hv
OH
OH
OH
OH
OH
OH
C/EC
m/z 289
Compound P2
m/z 287
hv
Photolysis
OH
OH
OH
OH
OH
OH
HO
O
OH
HO
O
OH
HO
OH
CH3CH2OH
hv
OH
Photo-induced
electron transfer
OH
OCH2CH3
OH
OH
OH
Compound P5
O
OH
m/z 335
HO
OH
H
hv
OH
OH
OH
OH
OH
OH
OH
HO
O
HO
OH
HO
OH
+
H
CH2
OH
OH
OH
O
CH2
Free radical A
m/z 137
Free radical B
m/z 289
OH
HO
Rearrangement of -OH
hv
OH
OH
HO
OH
HO
OH
O
HO
OH
OH
CH2
OH
Compound P1
m/z 137
OH
Isomers P6-9
m/z 427
Figure 6. Proposed photolytic reaction mechanism of EC/C.
Table 2. The 2,2-diphenylpicrylhydrazyl (DPPH) scavenging abilities of individual catechins before
and after UVB radiation (µM trolox) 1.
Treatment
EC
C
EGC
GC
ECg
Cg
EGCg
GCg
Before UVB radiation
After UVB radiation
1219 ± 49
967 ± 48
1376 ± 27
1285 ± 7
1836 ± 28
1736 ± 43
2064 ± 29
1996 ± 53
2026 ± 21
1136 ± 33
921 ± 30
1363 ± 12
1274 ± 30
1808 ± 31
1745 ± 23
2064 ± 22
1
The ethanol solutions of individual catechins (575 µM) were UVB irradiated for 6 h.

Molecules 2016, 21, 1345
10 of 13
3. Materials and Methods
3.1. Chemicals
The individual catechins (
−
)-epigallocatechin gallate (EGCg,
98%), ( )-epicatechin (EC,
98%), ( )-catechin gallate (Cg,
98%) were purchased from Aladdin Industrial Corporation (Shanghai,
China). The DPPH was purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan).
The ( )-6-Hydroxy-2,5,7,8-tetramethylchromane-2-carboxylic acid (Trolox) was purchased from
≥
98%), (
98%), (
98%), (
−
)-epigallocatechin (EGC,
≥
98%), (
(GCg, 98%), (
and gallic acid (GA,
−
)-epicatechin gallate (ECg,
≥
−
≥
−
)-gallocatechingallate
)-catechin (C, 98%),
≥
−
)-gallocatechin (GC,
≥
−
≥
−
≥
≥
±
Sigma-Aldrich (Shanghai, China). Chemical purity grade ethanol was purchased from Sinopharm
Chemical Reagent Co., Ltd. (Beijing, China). Acetic acid glacial (TEDIA company, Fairfield, OH, USA),
methanol and acetonitrile (Avantor performance materials, Inc., Center Valley, PA, USA) were of HPLC
(High performance liquid chromatography) grade. The Milli-Q water was prepared by an EASYPure
II UV UltraPure Water System (Barnstead International, Dubuque, IA, USA).
3.2. Photosensitivities of Eight Catechins to UVB
The solutions of individual catechins were freshly prepared at 575
respectively. Two milliliters of each catechin solution was placed in a 2 mL polypropylene centrifuge
tube which was exposed to UVB for 6 h with a fully opening lid, and was then stored at
20 ◦C to
terminate reactions. The intensity of UVB was achieved at 100
cm−2 by placing samples under
µM with water and ethanol
−
µW
·
six UV lamps (SPECTRONICS BLE-1T158 Tube 15 watt, main output at 312 nm) at a distance of
45 cm, which was close to the UVB intensity in the sunlight. All samples were turned back to room
temperature, and made up to 2 mL prior to HPLC analysis.
In a pretest, the potential UVB absorbing effect of polypropylene tube wall on the
UVB-induced chemical conversion of catechins (575 µM, ethanol) was evaluated by comparing with
UVB-transmissive Q-cuvette. The result showed that the photo-induced conversion rates of EC and
C in polypropylene tubes slightly decreased by ~8% compared with Q-cuvette, while no significant
difference was observed for other catechins. The polypropylene centrifuge tubes were used in the
following studies considering operational convenience and practical use.
3.3. Effect of UVB Radiation Time on Conversions of EC and C
According to the recovery of catechins in Section 3.2, EC and C were UVB-susceptible and selected
for further investigation. The solutions of EC and C were prepared at 575 µM with water and ethanol
respectively. Two milliliters of EC and C solutions were exposed to UVB for 2 h, 4 h, 6 h, 8 h and 10 h,
respectively. All samples were treated as above and analyzed by HPLC.
3.4. Effect of Substrate Concentration on Conversions of EC and C
The solutions of EC and C were respectively prepared at 71.875, 143.75, 287.5, 575 and 1150 µM
by water and ethanol. Two milliliters of C and EC solutions at various concentrations were exposed to
UVB for 6 h. All samples were treated as described in Section 3.2 and analyzed by HPLC.
3.5. HPLC Analysis
All samples were centrifuged at 12,000 rpm for 10 min before HPLC analysis. The concentrations
of catechins and GA were analyzed by HPLC according to a previous paper [39]. The HPLC conditions
were: injection volume 10
µ
L, Agilent TC-C18 column (4.6 mm
×
250 mm, Agilent Technologies,
Santa Clara, CA, USA), column temperature 32 ◦C, mobile phase A = acetonitrile/acetic acid/water
(6:1:193, v), mobile phase B = acetonitrile/acetic acid/water (60:1:139, v), linear gradient elution: from
80% (v) A and 20% (v) B to 35% (v) A and 65% (v) B for the first 35 min and then 80% (v) A and
20% (v) B until 40 min, flow rate 1 mL
min−1, Shimadzu SPD ultraviolet detector at 280 nm. Catechins
were quantified according to external standard calibration by comparing with the corresponding
authentic standards.
·

Molecules 2016, 21, 1345
11 of 13
3.6. Identification of the Photoproducts of EC and C
Ninety milliliters of the aqueous and ethanol solutions of EC and C (287.5
µM) were respectively
exposed to UVB for 6 h as described in Section 3.2. The reaction mixtures of EC and C were submitted to
◦
rotary evaporator, concentrated to 10 mL at 40 C under vacuum and sampled for UPLC-DAD-MS/MS
analysis and the left concentrate was freeze dried for 1H-NMR measurement.
3.6.1. UPLC-DAD-MS/MS Analysis
UPLC-DAD-MS/MS (Waters Corporation, Milford, DE, USA) was employed to provide the
MS information of relevant photoproducts. The UPLC conditions were: Acquity UPLC HSST3
column (2.1 mm
×
150 mm, 1.8 µ µL, mobile phase
m), column temperature 35 ◦C, injection volume 5
A = 0.1% formic acid + 99.9% water (v/v), mobile phase B = 0.1% formic acid + 99.9% acetonitrile (v/v),
linear gradient elution: from 99.9% (v) A/0.1% (v) B to 10% (v) A/90% (v) B for 38 min, flow rate
0.3 mL
min−1. An electrospray ionization (ESI) technique in a negative ion mode was employed for
·
MS analysis. The ion source conditions were set as follows: capillary voltage 3000 V, cone voltage 30 V,
◦
extractor 3.0 V and RF lens 0.2 V, ion source temperature 150 C, desolvation gas nitrogen at a flow rate
of 800 L h
−1 and temperature at 300 ◦C. Full scan ranging from 100 to 1000 atomic mass unit (amu)
·
were recorded. Argon was used as the collision gas. The collision energy was 20 eV. Triple quadrupole
was set-up to daughter ion scan experiment, and the mass of parent ion was set at 427 m/z. UV-visible
spectrum of reaction products was recorded in the 190–400 nm range by a photodiode detector.
1
3.6.2. H Nuclear Magnetic Resonance (NMR) Measurement
1H-NMR spectra were obtained from ~5 mg of EC and the corresponding reaction mixture suspended
in 0.8 mL acetone-d 6. The spectra were recorded on an Agilent 600 MHz DD 2 with a 5 mm One NMR
probe (Agilent Technologies, Santa Clara, CA, USA). Integration of the spectra was performed with
Mnova NMR software (version 6.1.0, Mestrelab Research S. L., Santiago de Compostela, Spain).
3.7. Antioxidant Activity Measurement
Antioxidant activity was evaluated by scavenging activity of DPPH free radical according to
the modified method [40]. One hundred and eighty microliters of DPPH solution (0.4 mM, ethanol)
and 20 µL of individual catechin solution (575 µM, ethanol) with and without 6 h UVB radiation
were loaded onto a microplate, and then the mixture was incubated in dark at 30 ◦C for 30 min and
measured by Epoch Microplate Spectrophotometer (BioTek Instruments Inc., Winooski, VT, USA) at
517 nm. Antioxidant capacity was expressed as µM trolox equivalents, using the linear calibration
curve of trolox (100–1250 µM).
3.8. Data Analysis
All analyses were carried out in triplicate. The mean values of the triplicate analysis
are presented.
±
SD
4. Conclusions
Eight catechins showed different photosensitivities to UVB. EC and C were susceptible under
UVB radiation while other six catechins were insensitive. Photoisomerization and photolysis were the
main reactions in the photo-induced chemical transformations of EC and C, which were influenced
by radiation time, solvent and substrate concentration. Two reaction pathways including radical
reaction and photo-induced electron transfer reaction were involved in the photolysis reaction of EC
and C. No significant change was observed in the antioxidant activities of eight catechins upon 6 h
UVB irradiation.
Supplementary Materials: Supplementary materials can be accessed at: http://www.mdpi.com/1420-3049/21/
10/1345/s1.

Molecules 2016, 21, 1345
12 of 13
Acknowledgments: This work was supported by the Zhejiang Provincial Natural Science Foundation of China
under Grant (LQ16C160001) awarded to Jian-Hui Ye.
Author Contributions: Y.-R.L. and J.-H.Y. conceived and designed the experiments; M.S. performed the
experiments; Y.N. and J.-L.L. analyzed the data; X.-Q.Z. contributed reagents/materials/analysis tools;
J.-H.Y. wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
Chen, L.; Zhang, H.Y. Cancer preventive mechanisms of the green tea polyphenol ( )-epigallocatechin-
3-gallate. Molecules 2007, 12, 946–957. [CrossRef] [PubMed]
−
2.
Nomura, S.; Monobe, M.; Ema, K.; Matsunaga, A.; Maeda-Yamamoto, M.; Horie, H. Effects of flavonol-rich
green tea cultivar (Camellia sinensis L.) on plasma oxidized LDL levels in hypercholesterolemic mice.
Biosci. Biotech. Bioch. 2016, 80, 360–362. [CrossRef] [PubMed]
3.
4.
5.
Chiu, A.E.; Chan, J.L.; Kern, D.G.; Kohler, S.; Rehmus, W.E.; Kimball, A.B. Double-blinded,
placebo-controllecl trial of green tea extracts in the clinical and histologic appearance of photoaging skin.
Dermatol. Surg. 2005, 31, 855–860. [CrossRef] [PubMed]
Huang, C.C.; Wu, W.B.; Fang, J.Y.; Chiang, H.S.; Chen, S.K.; Chen, B.H.; Chen, Y.T.; Hung, C.F.
( )-Epicatechin-3-gallate, a green tea polyphenol is a potent agent against UVB-induced damage in HaCaT
−
keratinocytes. Molecules 2007, 12, 1845–1858. [CrossRef] [PubMed]
Jeon, S.E.; Choi-Kwon, S.; Park, K.A.; Lee, H.J.; Park, M.S.; Lee, J.H.; Kwon, S.B.; Park, K.C. Dietary
supplementation of (+)-catechin protects against UVB-induced skin damage by modulating antioxidant
enzyme activities. Photodermatol. Photoimmunol. Photomed. 2003, 19, 235–241. [CrossRef] [PubMed]
Byun, M.W.; Jo, C.; Lee, J.W.; Jo, S.K.; Kim, K.S. Application of radiation technology to develop green tea leaf
as a natural resource for the cosmetic industry. Radiat. Phys. Chem. 2004, 71, 485–487. [CrossRef]
Frauen, M.; Rode, T.; Rapp, C.; Steinhart, H. Determination of green-tea catechins in cosmetic formulations
and in in-vitro skin extracts by high-performance liquid chromatography coupled with electrospray
ionization mass spectrometry. Chromatographia 2002, 55, 43–48. [CrossRef]
6.
7.
8.
9.
Gianeti, M.D.; Mercurio, D.G.; Maia Campos, P.M. The use of green tea extract in cosmetic formulations:
Not only an antioxidant active ingredient. Dermatol. Ther. 2013, 26, 267–271. [CrossRef] [PubMed]
Scalia, S.; Marchetti, N.; Bianchi, A. Comparative evaluation of different co-antioxidants on the
photochemical-and functional-stability of epigallocatechin-3-gallate in topical creams exposed to simulated
sunlight. Molecules 2013, 18, 574–587. [CrossRef] [PubMed]
10. Onoue, S.; Seto, Y.; Ochi, M.; Inoue, R.; Ito, H.; Hatano, T.; Yamada, S. In vitro photochemical and
phototoxicological characterization of major constituents in St. John’s Wort (Hypericum perforatum) extracts.
Phytochemistry 2011, 72, 1814–1820. [CrossRef] [PubMed]
11. Ye, Q.; Chen, H.; Zhang, L.B.; Ye, J.H.; Lu, J.L.; Liang, Y.R. Effects of temperature, illumination, and sodium
ascorbate on browning of green tea infusion. Food Sci. Biotechnol. 2009, 18, 932–938.
12. Ananingsih, V.K.; Sharma, A.; Zhou, W.B. Green tea catechins during food processing and storage: A review
on stability and detection. Food Res. Int. 2013, 50, 469–479. [CrossRef]
13. Fan, F.Y.; Shi, M.; Nie, Y.; Zhao, Y.; Ye, J.H.; Liang, Y.R. Differential behaviors of tea catechins under thermal
processing: Formation of non-enzymatic oligomers. Food Chem. 2016, 196, 347–354. [CrossRef] [PubMed]
14. Su, Y.L.; Leung, L.K.; Huang, Y.; Chen, Z.Y. Stability of tea theaflavins and catechins. Food Chem. 2003, 83,
189–195.
15. Seto, R.; Nakamura, H.; Nanjo, F.; Hara, Y. Preparation of epimers of tea catechins by heat treatment.
Biosci. Biotech. Bioch. 1997, 61, 1434–1439. [CrossRef]
16. Lavker, R.M.; Gerberick, G.F.; Veres, D.; Irwin, C.J.; Kaidbey, K.H. Cumulative effects from repeated exposures
to suberythemal doses of UV B and UV A in human skin. J. Am. Acad. Dermatol. 1995, 32, 53–62. [CrossRef]
17. Dobashi, Y.; Hirano, T.; Hirano, M.; Ohkatsu, Y. Antioxidant and photo-antioxidant abilities of catechins.
J. Photochem. Photobiol. A 2008, 197, 141–148. [CrossRef]
18. Fourie, T.G.; Ferreira, D.; Roux, D.G. Flavonoid synthesis based on photolysis of flavan-3-ols,
3-hydroxyflavanones, and 2-benzylbenzofuranones. J. Chem. Soc. Perkin Trans. 1 1997, 125–133. [CrossRef]
19. Zhao, Y.; Shi, M.; Ye, J.H.; Zheng, X.Q.; Lu, J.L.; Liang, Y.R. Photo-induced chemical reaction of
trans-resveratrol. Food Chem. 2015, 171, 137–143. [CrossRef] [PubMed]

Molecules 2016, 21, 1345
13 of 13
20. Huvaere, K.; Sinnaeve, B.; Van Bocxlaer, J.; Skibsted, L.H. Flavonoid deactivation of excited state flavins:
Reaction monitoring by mass spectrometry. J. Agric. Food Chem. 2012, 60, 9261–9272. [CrossRef] [PubMed]
21. Huvaere, K.; Skibsted, L.H. Flavonoids protecting food and beverages against light. J. Sci. Food Agric. 2015
95, 20–35. [CrossRef] [PubMed]
,
22. Nakamura, K.; Shirato, M.; Ikai, H.; Kanno, T.; Sasaki, K.; Kohno, M.; Niwano, Y. Photo-irradiation of
proanthocyanidin as a new disinfection technique via reactive oxygen species formation. PLoS ONE 2013
,
8, e60053. [CrossRef] [PubMed]
23. Bianchi, A.; Marchetti, N.; Scalia, S. Photodegradation of ( )-epigallocatechin-3-gallate in topical cream
−
formulations and its photostabilization. J. Pharm. Biomed. Anal. 2011, 56, 692–697. [CrossRef] [PubMed]
24. Volf, I.; Ignat, I.; Neamtu, M.; Popa, V.I. Thermal stability, antioxidant activity, and photo-oxidation of natural
polyphenols. Chem. Pap. 2014, 68, 121–129. [CrossRef]
25. Cren-Olivé, C.; Lebrun, S.; Hapiot, P.; Pinson, J.; Rolando, C. Selective protection of catechin gives access to
the intrinsic reactivity of the two phenol rings during H-abstraction and photo-oxidation. Tetrahedron Lett.
2000, 41, 5847–5851. [CrossRef]
26. Wang, H.F.; Helliwell, K. Epimerisation of catechins in green tea infusions. Food Chem. 2000, 70, 337–344.
[CrossRef]
27. Li, N.; Taylor, L.S.; Ferruzzi, M.G.; Mauer, L.J. Color and chemical stability of tea polyphenol
(−)-epigallocatechin-3-gallate in solution and solid states. Food Res. Int. 2013, 53, 909–921. [CrossRef]
28. Forest, K.; Wan, P.; Preston, C.M. Catechin and hydroxybenzhydrols as models for the environmental
photochemistry of tannins and lignins. Photochem. Photobiol. Sci. 2004, 3, 463–472. [CrossRef] [PubMed]
29. Suzuki, M.; Sano, M.; Yoshida, R.; Degawa, M.; Miyase, T.; Maeda-Yamamoto, M. Epimerization of
tea catechins and O-methylated derivatives of ( )-epigallocatechin-3-O-gallate: Relationship between
−
epimerization and chemical structure. J. Agric. Food Chem. 2003, 51, 510–514. [CrossRef] [PubMed]
30. Kennedy, J.A.; Munro, M.H.G.; Powell, H.K.J.; Porter, L.J.; Foo, L.Y. The protonation reactions of catechin,
epicatechin and related-compounds. Aust. J. Chem. 1984, 37, 885–892. [CrossRef]
31. Sousa, W.R.; da Rocha, C.; Cardoso, C.L.; Silva, D.H.S.; Zanoni, M.V.B. Determination of the relative
contribution of phenolic antioxidants in orange juice by voltammetric methods. J. Food Compos. Anal. 2004
17, 619–633. [CrossRef]
,
32. Lukeman, M. Photochemical generation and characterization of quinone methides. In Quinone Methides;
Rokita, S.E., Ed.; John Wiley & Sons, Inc.: Hoboken, NJ, USA, 2009.
33. Wilke, T.; Schneider, M.; Kleinermanns, K. 1,4-Hydroquinone is a hydrogen reservoir for fuel cells and
recyclable via photocatalytic water splitting. Open J. Phys. Chem. 2013, 3, 97–102. [CrossRef]
34. Hailey, C.; Burks, J.W. Contact dermatitis from currently popular tanning agents. JAMA 1960, 174, 2072–2073.
[CrossRef] [PubMed]
35. Shi, J.Y.; Sun, J.; Wei, X.Y.; Shi, J.; Cheng, G.P.; Zhao, M.M.; Wang, J.S.; Yang, B.; Jiang, Y.M. Identification
of ( )-epicatechin as the direct substrate for polyphenol oxidase from longan fruit pericarp. LWT-Food
−
Sci. Technol. 2008, 41, 1742–1747. [CrossRef]
36. Wilhelm-Mouton, A.; Bonnet, S.L.; Ding, Y.Q.; Li, X.C.; Ferreira, D.; van der Westhuizen, J.H. Photochemistry
synthesis. Part 2: Enantiomerically pure polyhydroxy-1,1,3-triarylpropan-2-ols. J. Photochem. Photobiol. A
2012, 227, 18–24. [CrossRef]
37. Faria, J.L.; Steenken, S. Photoionization (λ = 248 or 308 nm) of triphenylmethyl radical in aqueous-solution-
formation of triphenylmethyl carbocation. J. Am. Chem. Soc. 1990, 112, 1277–1279. [CrossRef]
38. Lee, L.S.; Kim, S.H.; Kim, Y.B.; Kim, Y.C. Quantitative analysis of major constituents in green tea with
different plucking periods and their antioxidant activity. Molecules 2014, 19, 9173–9186. [CrossRef] [PubMed]
39. Liang, H.L.; Liang, Y.R.; Dong, J.J.; Lu, J.L.; Xu, H.R.; Wang, H. Decaffeination of fresh green tea leaf
(Camellia sinensis) by hot water treatment. Food Chem. 2007, 101, 1451–1456. [CrossRef]
40. Jo, C.; Son, J.H.; Lee, H.J.; Byun, M.W. Irradiation application for color removal and purification of green tea
leaves extract. Radiat. Phys. Chem. 2003, 66, 179–184. [CrossRef]
Sample Availability: Samples of UV B irradiated products of EC and C are available from the authors.
©
2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).