Isolation of Flavonoids and Flavonoid Glycosides from Myrsine africana and Their Inhibitory Activities against Mushroom Tyrosinase

Article abstract of DOI:10.1021/acs.jnatprod.7b00564

Bioassay-guided fractionation of the methanol extract of the shoots of Myrsine africana led to the isolation of the new compound myricetin 3-O-(2″,4″-di-O-acetyl)-α-l-rhamnopyranoside (9) and 11 known compounds. The known compounds quercetin 3-O-(3″,4″-di-O-acetyl)-α-l-rhamnopyranoside (8), rutin (10), quercetin 3-O-α-l-rhamnopyranoside (11), and myricetin 3-O-α-l-rhamnopyranoside (12) are reported for the first time from the methanol extract of the shoots of M. africana. Compounds 10 and 12 showed significant inhibition of tyrosinase with 50% inhibition (IC₅₀ values) of the enzyme at 0.13 ± 0.003 and 0.12 ± 0.002 mM, respectively, which was supported by the docking fitness scores obtained through molecular docking analysis. In addition, compounds 1-12 displayed significant antioxidant activity with IC₅₀ values ranging 1.90 to 3.90 μM.

Full text of DOI:10.1021/acs.jnatprod.7b00564

Article
Cite This: J. Nat. Prod. XXXX, XXX, XXX−XXX
pubs.acs.org/jnp
Isolation of Flavonoids and Flavonoid Glycosides from Myrsine
africana and Their Inhibitory Activities against Mushroom Tyrosinase
†
§
†
§
†
†,‡
†,‡
Navneet Kishore, Danielle Twilley, Analike Blom van Staden, Praveen Verma, Bikram Singh,
§
†
,†
Giorgia Cardinali, Daniela Kovacs, Mauro Picardo, Vivek Kumar, and Namrita Lall*
†
Department of Plant and Soil Sciences, University of Pretoria, Pretoria 0002, South Africa
‡
Natural Plant Products Division, CSIR-Institute of Himalayan Bioresources Technology, P.O. Box-6, Palampur, Himachal Pradesh,
India 176061
§
San Gallicano Dermatologic Institute, IRCCS, Via Elio Chianesi, 00144, Rome, Italy
S Supporting Information
*
ABSTRACT: Bioassay-guided fractionation of the methanol extract of the shoots of Myrsine africana led to the isolation of the
new compound myricetin 3-O-(2″,4″-di-O-acetyl)-α-L-rhamnopyranoside (9) and 11 known compounds. The known
compounds quercetin 3-O-(3″,4″-di-O-acetyl)-α-L-rhamnopyranoside (8), rutin (10), quercetin 3-O-α-L-rhamnopyranoside
(
11), and myricetin 3-O-α-L-rhamnopyranoside (12) are reported for the first time from the methanol extract of the shoots of M.
africana. Compounds 10 and 12 showed significant inhibition of tyrosinase with 50% inhibition (IC values) of the enzyme at
50
0
.13 ± 0.003 and 0.12 ± 0.002 mM, respectively, which was supported by the docking fitness scores obtained through molecular
docking analysis. In addition, compounds 1−12 displayed significant antioxidant activity with IC50 values ranging 1.90 to 3.90
μM.
elanin is a pigment produced in the melanocytes and is
Myrsine africana L. (Myrsinaceae) is a small shrub that is
widely distributed in South Africa and Europe. The leaf
decoction of M. africana is traditionally used by the Southern
Sotho, Tswana, and Kwena tribes of South Africa for skin
M
responsible for the variation in the colors of eyes, skin,
1
and hair. Melanin protects against photodamage by absorbing
ultraviolet (UV) rays and scavenging reactive oxygen species
(
ROS). The exposure to sunlight induces an increase in
disorders such as acne, pigmentation, wound healing, and
melanin production, leading to a protective phenomenon
termed skin tanning. However, an overproduction of melanin
9,10
cellulitis and for its properties as a blood purifier.
M.
africana has several other traditional uses, such as the treatment
2
can lead to a condition known as skin hyperpigmentation. UV
of diarrhea, rheumatism, toothache, and pulmonary tuber-
radiation increases the presence of free radicals, which are
involved in the melanin biosynthesis pathway and consequently
10
culosis. It is also used as a flavorant, appetizer, carminative,
11
3
and spice. Local people in the Samburu district in Kenya use
stimulates melanogenesis. In a previous study, Fujiwara et al.
the hot decoction of ground seeds of M. africana for treating
concluded that UVB-induced pigmentation in guinea pigs could
be reduced by the daily consumption of vitamin C, a well-
12
wounds. The inhabitants of Pakistan use a leaf decoction of
13
4
M. africana as a blood purifier and for skin allergies.
The blood purifying properties of M. africana play an
known antioxidant; therefore, the scavenging of free radicals
by antioxidants is an important factor to inhibit melanogenesis.
14
important role in ensuring healthy and beautiful skin. Natural
There are several significant tyrosinase inhibitors obtained from
5
−7
blood purifiers fight skin microorganisms, reduce acne and
natural sources.
These are mainly used to treat skin
8
hyperpigmentation or depigmentation conditions. Several
plant species are being explored in South Africa due to their
traditional use against hyperpigmentation and skin problems.
Received: July 13, 2017
2
©
XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.7b00564
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
1
5
pimples, and prevent aging. Poor blood circulation results in
Table 1. Antityrosinase and Antioxidant Activity of the
16
pigmentation changes, which ultimately leads to age spots.
Partitions Obtained through Liquid−Liquid Partitioning,
Major Fractions, and Isolated Compounds from the
Methanol Extract of the Shoots of Myrsine africana
Therefore, for the present study M. africana was selected to
evaluate its effect on skin hyperpigmentation. These studies
also revealed the presence of flavonoids, benzoquinones, and
triterpenoids in the shoots of the plant that led to the
investigation of their therapeutic use against hyperpigmenta-
tion. The study aimed at determining the antityrosinase activity
of the methanol extract of the shoots of M. africana. The
bioassay-guided fractionation was performed to identify
compounds responsible for tyrosinase inhibition and antiox-
idant activity. This led to the isolation of a new and 11 known
compounds. The biological activity of the methanol extract of
the shoots of M. africana and its constituents on melanin
production and its effect on mushroom tyrosinase activity, the
tyrosinase enzyme (through molecular docking), the 1,2-
diphenyl-1-picrylhydrazyl (DPPH) free radical, and melanin
production were explored. The mechanism of action was
determined by evaluating melanin transfer using immunofluor-
escence on a coculture of human melanocytes with
keratinocytes.
antityrosinase activity
DPPH activity
a
b
compound
IC50 ± SD
IC50 ± SD
Myrsineafricana crude extract 0.12 ± 0.001 mg/mL 8.65 ± 0.230 μg/mL
c
petroleum ether partition
0.29 ± 0.001 mg/mL nt
0.28 ± 0.001 mg/mL nt
0.12 ± 0.001 mg/mL nt
CHCl partition
3
EtAc partition
n-BuOH partition
>0.30 mg/mL
>0.30 mg/mL
>0.30 mg/mL
nt
nt
nt
MF-1
MF-2
MF-3
0.08 ± 0.001 mg/mL nt
0.10 ± 0.001 mg/mL nt
MF-4
MF-5
>0.30 mg/mL
>0.30 mg/mL
>0.30 mg/mL
nt
MF-6
nt
MF-7
nt
1
2
3
4
5
6
7
8
9
3.00 ± 0.007 μM**
3.90 ± 0.462 μM**
3.20 ± 0.040 μM**
4.20 ± 0.003 μM**
2.20 ± 0.019 μM**
1.90 ± 0.021 μM**
3.50 ± 0.040 μM**
3.40 ± 0.751 μM**
3.20 ± 0.009 μM**
2.30 ± 0.002 μM**
2.20 ± 0.022 μM**
2.00 ± 0.006 μM**
0.41 ± 0.002 mM
0.24 ± 0.003 mM
0.21 ± 0.003 mM
0.19 ± 0.004 mM
0.31 ± 0.003 mM
0.28 ± 0.002 mM
0.18 ± 0.002 mM
0.52 ± 0.002 mM
0.13 ± 0.003 mM
0.15 ± 0.003 mM
0.12 ± 0.003 mM
0.01 ± 0.001 mM
0.144 ± 0.004 mM
RESULTS AND DISCUSSION
■
Tyrosinase Enzyme Inhibition Assay and Bioassay-
Guided Fractionation. The inhibitory effect of the methanol
extract of the shoots of M. africana on the rate-limiting enzyme
of melanogenesis, tyrosinase, was determined. The extract
showed good inhibition of tyrosinase with an IC value of 0.12
0.001 mg/mL when L-tyrosine was used as the substrate
Table 1). In a similar study by Momtaz et al. an IC value of
.02 ± 0.42 mg/mL for a methanol extract of the aerial parts
and bark of M. africana was reported.
The methanol extract of the shoots of M. africana was
subjected to bioassay-guided fractionation using various
chromatographic techniques. It resulted in the isolation of
one new and 11 known compounds. The structural assessment
5
0
1
0
±
(
17
11
50
1
2
0
d
kojic acid
e
ascorbic acid (vitamin C)
11.20 ± 1.36 μM
a
Myrsinoside A (1), myrsinoside B (2), quercetin (3), myricetin (4),
mearnsetin 3-O-(4″-O-acetyl)-α-L-rhamnopyranoside (5), mearnsitrin
(
6), myricetin 3-O-(4″-O-acetyl)-α-L-rhamnopyranoside (7), quercetin
1
3-(3″,4″-di-O-acetyl-α-L-rhamnoside) (8), myricetin 3-O-(2″,4″-di-O-
acetyl)-α-L-rhamnopyranoside (9), rutin (10), quercetin 3-O-α-L-
rhamnopyranoside (11), and myricetin 3-O-α-L-rhamnopyranoside
of these compounds was performed using HR-MS and H and
C NMR data. The assignment of signals was facilitated by
COSY, HSQC, and HMBC experiments. The known
compounds were (Chart 1) myrsinoside A (1), myrsinoside
1
3
b
c
d
(
12). 50% inhibitory concentration. nt, not tested. Positive control
e
for antityrosinase activity. Positive control for antioxidant activity.
1
8
19
B (2), quercetin (3), myricetin (4), mearnsetin 3-O-(4″-O-
18
11
acetyl)-α-L-rhamnopyranoside (5), mearnsitrin (6), myr-
icetin 3-O-(4″-O-acetyl)-α-L-rhamnopyranoside (7), quercetin
reagents, indicating the attachment of the sugar residue at C-3
20
21
24
3
-O-(3″,4″-di-O-acetyl)-α-L-rhamnoside (8), rutin (10),
of the myricetin moiety. The acidic hydrolysis of 9 yielded an
22
25
quercetin 3-O-α-L-rhamnopyranoside (11), and myricetin 3-
aglycone and rhamnose as the only sugar moiety (identified
23
O-α-L-rhamnopyranoside (12). The compounds were iden-
tified based on the acid hydrolysis, physical data analysis, and
comparison with reported data. Compounds 1−7 have
previously been identified from the shoots and/or leaves of
M. africana; however, compounds 8 and 10−12 were identified
for the first time from the methanol extract of the shoots of M.
africana. Compounds 8 and 10−12 have previously been
by TLC after comparison with authentic samples). An MS
fragment at m/z 319.04 indicated the loss of the di-O-
acetylrhamnose moiety, which indicated myricetin as the
1
aglycone. The H NMR spectrum (500 MHz in DMSO-d )
6
provided further evidence for myricetin as the aglycone of
compound 9 via the singlet resonance at δ 6.86 attributed to
H
H-2′ and H-6′ and the characteristic two-proton AX system of
the A-ring of 5,7-dihydroxyflavonols. The signals at δH 5.34
(1H, d, J = 2.0 Hz) and 0.88 (3H, d, J = 6.8 Hz) were assigned
to the anomeric and 6″-methyl protons of a rhamnopyranosyl
unit, respectively. The glycosidic linkage of the rhamnopyrano-
side was determined based on the coupling constant of the
anomeric proton. The α-configuration of the L-rhamnose was
deduced from the coupling constant (J = 2.0 Hz) of the
20−23
reported from other plants.
Compound 9 was isolated as a yellowish, amorphous powder.
The molecular formula was established as C H O from the
protonated molecular ion peak [M + H] at m/z 549.1238
calcd for C H O 549.1239) in the HRESIMS and from 25
2
5
24 14
+
(
25
25 14
1
3
carbon resonances in the C NMR spectrum. The IR spectrum
(
KBr) exhibited absorption bands at 3417 and 1719 cm−1 along
with other absorption bands, indicative of the presence of
hydroxy and carbonyl groups, respectively, in the molecule. The
UV spectrum showed absorption bands at 257 and 354 nm
which gave bathochromic shifts with the addition of shift
anomeric proton signal at δ 5.34 (Table 1). The small J values
H
indicated α-glycosidic linkages in all cases. The spectrum also
showed signals for a myricetin 3-O-L-rhamnoside possessing
two acetyl groups resonating at δH 2.07 and 2.04 as three-
B
DOI: 10.1021/acs.jnatprod.7b00564
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Chart 1. Chemical Structures of Isolated Compounds from the Methanol Extract of the Shoots of Myrsine africana
1
1
Figure 1. Key HMBC [H → C] (A) and H− H COSY (B) correlations of compound 9 isolated from the methanol extract of the shoots of Myrsine
africana.
proton singlets. The presence of the two acetyl groups was
highest tyrosinase inhibition with IC50 values of 0.13 ± 0.003,
0.15 ± 0.003, and 0.12 ± 0.002 mM, respectively. Compounds
3−7 showed tyrosinase inhibition with IC values ranging from
1
3
further supported by the C NMR resonances at δ 170.1,
C
170.0, 21.1, and 20.9 assigned to the carbonyl and methyl
5
0
carbons. COSY correlations were observed between H-6 (δ
0
.21 to 0.50 mM. In the present study, the IC50 value of
6
.18) and H-8 (δ 6.36) and between H-5″ (δ 3.34) and H -6″
3
quercetin was found to be 0.24 ± 0.003 mM. According to
(
δ 0.88). COSY correlations were also observed between H-1″
earlier reports, the IC value of quercetin in the oxidation of L-
5
0
(
δ 5.34) and H-2″ (δ 5.01), H-2″ (δ 5.01) and H-3″ (δ 3.95),
26
DOPA by mushroom tyrosinase was found to be 0.13 mM.
H-3″ (δ 3.95) and H-4″ (δ 4.63), and H-4″ (δ 3.95) and H-5″
Rutin has been previously reported to inhibit tyrosinase, and it
(
δ 3.34). In the HMBC spectrum, H-1″ (δ 5.34) correlated to
H
showed an IC value of 0.52 ± 0.002 mM against tyrosinase in
50
C-3 (δ 134.8). Similarly, H-2″ (δ 5.01) showed connectivity
C
H
the present study. On the contrary, another study reported an
to C-2″-COCH (δ 170.1) and H-4″ (δ 4.63) with C-4″-
27
3
C
H
IC50 value of 0.07 ± 0.3 mM for rutin against tyrosinase.
COCH (δ 170.0). HMBC correlations were observed from
3
C
Quercetin 3-O-α-L-rhamnopyranoside and myricetin 3-O-α-L-
rhamnopyranoside were previously investigated for their
the methyl group protons (δ 0.88) to C-5″ (δ 69.9), (δ
H
C
H
2
.07) to 2″-COCH (δ 170.1), and (δ 2.04) to 4″-COCH
3
C
H
3
antityrosinase potential and showed IC50 values of 0.10 and
(
δ 170.0) (Figure 1). On the basis of NMR shift values the
C
28
0
.32 mM, respectively. However, in the present study,
acetyl substituents were located at C-2″ and C-4″, which was
quercetin 3-O-α-L-rhamnopyranoside and myricetin 3-O-α-L-
confirmed by the deshielded H-2″ (δ 5.01) and H-4″ (δ
H
H
rhamnopyranoside showed IC values of 0.13 ± 0.003 and 0.15
4
.63) resonances. Hence, the structure of compound 9 was
50
±
0.003 mM, respectively. It is interesting to note that the
defined as the new myricetin 3-O-(2″,4″-di-O-acetyl)-α-L-
rhamnopyranoside.
All the isolated compounds were tested for antityrosinase
activity (Table 1). Compounds 10, 11, and 12 showed the
antityrosinase activities of myrisinosides A and B, myricetin,
mearnsetin 3-O-(4″-O-acetyl)-α-L-rhamnopyranoside, mearnsi-
trin, myricetin 3-O-(4″-O-acetyl)-α-L-rhamnopyranoside, and
C
DOI: 10.1021/acs.jnatprod.7b00564
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
1
13
quercetin 3-O-(3″,4″-di-O-acetyl)-α-L-rhamnopyranoside have
not previously been reported.
Table 2. H and C NMR Data (DMSO; 500 and 125 MHz)
of Compound 9
Measurement of DPPH Radical Scavenging Activity.
There are a number of intrinsic and extrinsic stresses that
constantly affect the skin and, therefore, contribute to an
increase in the reactive oxygen species. ROS induced through
UV radiation are known to play a role in skin pigmentation by
inducing melanogenesis or increasing melanocyte proliferation.
Antioxidants, which are able to scavenge ROS, could therefore
possibly inhibit messenger molecules, which stimulate melano-
genesis. The copper ion present in the active site of the
tyrosinase enzyme could also provide a binding site for
1
1
H− H
position
δH (J in Hz)
δC, type
HMBC
COSY
2
3
4
5
6
7
8
9
1
1
2
3
4
5
6
1
158.4, C
134.8, C
177.6, C
161.3, C
98.7, CH
164.2, C
93.6, CH
156.4, C
108.6, C
119.6, C
6.18, d (2.1)
6.36, d (2.1)
8, 10
6, 10
8
6
29−31
antioxidants and in turn regulate skin pigmentation.
In
0
the molecular docking study, ascorbic acid and kojic acid
showed similar docking fitness scores of 48.25 and 46.15,
respectively, with the active site of the tyrosinase enzyme. The
well-known antioxidant ascorbic acid is furthermore known to
′
′
′
′
′
′
″
2″
″
″
6.86, s
107.8, CH 2, 4′
145.8, C
136.5, C
32
inhibit tyrosinase monophenolase activity. Therefore, the M.
africana extract and isolated compounds were tested for their
antioxidant activity against the DPPH free radical. The extract
145.8, C
6.86, s
107.8, CH 2, 4′
101.6, CH 3, 2″
5.34, d (2.0)
2″
showed strong DPPH radical scavenging capacity with an IC50
value of 8.65 ± 0.23 μg/mL (Table 1). There are no published
reports on the DPPH activity of the constituents of M. africana.
5.01, dd (3.9, 2.0) 73.2, CH
2″-COCH3
4″-COCH3
5″
1″, 3″
2″, 4″
3″, 5″
5″, 6″
5″
3
3.95, dd(9.8, 4.4)
4.63, t (9.8)
68.1, CH
74.0, CH
4
Compounds 1−12 showed IC values statistically lower (p <
5
0
5″
6″
3.34, dd (9.8, 6.8) 69.9, CH
0
.01) than that of ascorbic acid (IC value of 11.2 μM) with
50
0.88, d (6.8,)
17.3, CH3
170.1, C
mearnsitrin having the highest activity (Table 1). The DPPH
radical scavenging activities of myrisinoside A, myrisinoside B,
mearnsetin 3-O-(4″-O-acetyl)-α-L-rhamnopyranoside, and myr-
icetin 3-O-(4″-O-acetyl)-α-L-rhamnopyranoside have not pre-
viously been reported, whereas quercetin, myricetin, myricetin
2
4
2
4
″-COCH3
″-COCH3
″-COCH3 2.07, s
″-COCH3 2.04, s
170.0, C
20.9, CH3
21.1, CH3
2″-COCH3
4″-COCH3
3-O-α-L-rhamnopyranoside, mearnsitrin, rutin, and quercetin-3-
O-α-L-rhamnopyranoside are well known as antioxidants and
have previously been reported.
short polar interaction distances and large H-bond interactions.
The docking study clearly indicated that distances for polar
interactions with Cu2+ ions play a crucial role in controlling the
tyrosinase inhibitory activity.
Measurement of Melanin Content. The effect of the
methanol extract of the shoots of M. africana on melanin
production was determined by observing the effect on melanin
production in B16F10 cells and melanin transfer between
human melanocytes and keratinocytes. The extract was found
to be nontoxic to melanocyte cells at the highest concentration
(100 μg/mL) tested. At a concentration of 50 μg/mL, the
extract inhibited 50% of melanin production as compared to
untreated cells as determined from a bovine serum albumin
standard curve.
31−35
Molecular Docking Analysis. The molecular docking
study was performed to identify the possible orientations and
binding interactions of the molecules in the active site of the
tyrosinase enzyme (Table 3; Figure 2). Corresponding to the
tyrosinase inhibitory activity, kojic acid showed the highest
docking fitness score of 48.25, followed by ascorbic acid
(
vitamin C) at 46.14, as compared to the isolated compounds.
Despite the presence of only two H-bond interactions (Figure
), there is a minimal distance between the polar interaction of
3
2
+
kojic acid and the Cu ions and the primary hydroxy group of
2
+
ascorbic acid and the Cu ions, which could potentially explain
the high docking score. In contrast to this, despite having three
to four H-bond interactions with the active site residues,
compounds 1, 2, 8, and 9 showed lower docking fitness scores
when compared to kojic acid. This could also potentially be due
to the distances of the polar interactions with the Cu2+ ions.
The distances of the polar interactions in 1, 2, 8, and 9 were
marginally larger than the corresponding distances of kojic acid.
Similar interactions were seen for compound 10; however, five
H-bond interactions with the site residues were observed, of
which four bonds were on the same two residues. Compound
Measurement of Melanin Transfer. During the mecha-
nistic studies, the effect of the methanol extract of the shoots of
M. africana on melanin transfer was analyzed using
immunofluorescence. The cocultures of normal human
melanocytes (NHMs) and normal human keratinocytes
(NHKs) treated with the M. africana extract showed an
increase in contact between NHMs and NHKs as compared to
the DMSO-treated cells as well as an increase in dendrite length
(Figure 3). However, no melanosome transfer was detected at a
1
1
0, however, showed good antityrosinase results that
concentration of 20 μg/mL ( / IC ) as compared to cells
4
50
substantiated its docking fitness scores. Compounds 3−7
showed moderate antityrosinase activity that corresponded to
their docking scores. These molecules showed moderate
distances for polar interactions with the Cu ions and also
showed more H-bond interactions when compared to
compounds 1, 2, 8, and 9. Corresponding to tyrosinase
inhibitory activity, compounds 11 and 12 showed better
docking fitness scores, which was supported by reasonably
treated with α-melanocyte stimulating hormone (α-MSH).
Furthermore, there were no melanosomes present around the
nucleus of the keratinocytes. This confirmed our findings that
the M. africana extract was able to inhibit melanin production.
During the tyrosinase and molecular docking studies it was
evident that kojic acid and ascorbic acid had the highest
docking scores, which correlated with their tyrosinase
inhibition. However, the docking fitness scores and tyrosinase
2
+
D
DOI: 10.1021/acs.jnatprod.7b00564
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 3. Antityrosinase Activity and Docking Scores of Compounds Isolated from the Methanol Extract of the Shoots of Myrsine
africana
distance for polar
interactions with
a
2+
compound
kojic acid
IC50 (mM)
dock score (gold fitness score)
Cu ions (Å)
active site residues involved in H-bond
His61, His259
0.01 ± 0.001
48.25
26.49
31.01
40.40
39.58
33.81
35.69
35.92
30.56
29.38
30.24
37.58
37.54
46.15
2.19
2.74
2.19
2.41
2.03
2.36
2.16
2.20
2.22
1.99
3.43
2.25
2.21
2.12
2.37
2.69
3.32
2.42
2.79
2.54
3.04
2.45
3.15
3.37
2.96
2.41
2.72
2.41
1
2
3
4
5
6
7
8
9
1
1
1
His61, His85, His259, Asn260
0.41 ± 0.002
0.24 ± 0.003
0.21 ± 0.003
0.19 ± 0.004
0.31 ± 0.003
0.28 ± 0.002
0.18 ± 0.002
0.52 ± 0.002
0.13 ± 0.003
0.15 ± 0.003
0.12 ± 0.003
His259, Asn260, His263, Met280
His61, His85, His263, Met280, Gly281
b
b
His61, His259, His263 (2) , Asn260 (2) , Met280
His61, His85, His263, Met280, Val283
His61, His263, Met280, Gly281
His61, His263, Met280, Gly281, Val283
His61, His85, His263
His259, His263, Met280, Val283
b
b
0
1
2
His85 (2) , Met280, Gly281 (2)
His61, His263, Met280, Val283
b
His61, His263 (2) , Met280, Gly281
ascorbic acid (vitamin C)
a
Myrsinoside A (1), myrsinoside B (2), quercetin (3), myricetin (4), mearnsetin 3-O-(4″-O-acetyl)-α-L-rhamnopyranoside (5), mearnsitrin (6),
myricetin 3-O-(4″-O-acetyl)-α-L-rhamnopyranoside (7) quercetin 3-(3″,4″-di-O-acetyl)-α-L-rhamnoside (8), myricetin-3-O-(2″,4″-di-O-acetyl)-α-L-
rhamnopyranoside(9), rutin (10), quercetin 3-O-α-L-rhamnopyranoside (11), and myricetin 3-O-α-L-rhamnopyranoside (12). Number of H-bonds
b
with respective residues.
inhibition of the isolated compounds were not as high as those
of the two positive controls. The compounds, however, had
significant antioxidant activity, even higher than that of ascorbic
acid, and, therefore, the mechanism of these compounds could
be due to the inhibition of melanin transfer and not due to
tyrosinase inhibition. Thus, the activity of compounds to inhibit
melanin transfer should be considered for future studies.
inhibition (Table 1). The EtOAc fraction (18.70 g) showed the
highest tyrosinase inhibition and was subjected to silica gel column
chromatography using n-hexane, MeOH, and distilled H O in ratios of
2
increasing polarity. Similar fractions were pooled based on TLC
profiles to obtain seven major fractions (MF-1 to MF-7). MF-3 and
MF-4 showed the highest tyrosinase inhibition, with IC50 values of
0
.08 ± 0.001 and 0.10 ± 0.001 mg/mL, respectively (Table 1) and,
therefore, were subjected to further scrutiny. MF-2 (1.25 g, 1.42%
yield) was subjected to silica gel CC with an eluent of CH Cl /MeOH
2
2
EXPERIMENTAL SECTION
of increasing polarity (0−100%). Frs. 2−4 (500 mg, 0.57% yield of
■
extract) were chromatographed using silica gel with CH Cl /MeOH
2
2
General Experimental Procedures. Optical rotations were
measured in MeOH using a PerkinElmer polarimeter with a sodium
lamp operating at 598 nm. UV spectra were recorded on a Shimadzu
UV-1800 spectrophotometer, and the λ values are reported in nm. IR
spectra were recorded on a Nexus 670 FT-IR instrument using KBr
(7:1, 1 L), which afforded compound 1 (257 mg, 0.29% yield of
extract). Frs. 5−7 (205 mg, 0.23% yield of extract) were subject to
repeated silica gel column chromatography (CC) with CHCl /MeOH
3
(6:1, 800 mL), to afford compound 2 (62 mg, 0.07% yield of extract).
1
13
MF-3 (3.82 g, 4.34% yield of extract) was subjected to silica gel CC
pellets. H NMR (500 MHz) and C NMR (125 MHz) data were
recorded using an Agilent NMR spectrometer (Central Analytical
Facilities, Stellenbosch, South Africa), and the chemical shifts were
reported as δ values with tetramethylsilane as an internal standard at
using CH Cl /MeOH of increasing polarity (0 to 100%) as an eluent.
2
2
Similar fractions were pooled according to TLC profiles to give six
fractions (1−6). Fr. 2 (625 mg, 0.71% yield of extract) was
chromatographed on a Sephadex LH-20 column using CH Cl /
2
2
3
0 °C (measured in methanol-d and DMSO-d ). Inverse detected
4 6
MeOH. Frs. 4−7 were pooled (210 mg, 0.24% yield of extract) and
heteronuclear correlations were measured using HMQC and HMBC
pulse sequences with a pulsed-field gradient. HR-ESIMS data were
obtained using a Waters UPLC-MS system with PDA detector and
Waters Synapt G2 QTOF mass spectrometer.
chromatographed twice using Sephadex LH-20 with CHCl /MeOH
3
(25:1, 500 mL) to yield compound 3 (21 mg, 0.02% yield of extract).
Subfr. 9 (95 mg, 0.11% yield of extract) of fr. 2 was further purified
using Sephadex LH-20 with CHCl /MeOH (16:1, 500 mL) to yield
Plant Material. The shoots of M. africana were collected from the
Manie van der Schijff botanical garden of the University of Pretoria (S
3
compound 4 (25 mg, 0.03% yield of extract). Fr. 3 (118 mg, 0.13%
yield of extract) was chromatographed on Sephadex LH-20 using
2
5°45′21″, E 28°13′51″) in January 2013 and identified by Ms. Magda
CHCl /MeOH as an eluent. Subfrs. 2−4 from fr. 3 were pooled (80
Nel at the H.G.W.J. Schweickerdt Herbarium (PRU), where a voucher
specimen (MA-S-2013-1) was deposited.
3
mg, 0.09% yield of extract) and chromatographed using Sephadex LH-
2
0
0 with CHCl /MeOH (6:1, 500 mL) to yield compound 5 (10 mg,
.01% yield of extract). Fr. 5 (0.28 mg, 0.23% yield of extract) was
Extraction and Isolation. The air-dried shoots (2.3 kg) were
ground to a fine powder and soaked in MeOH (5 L) on a rotatory
shaker for 3 days. The filtrate was collected using a Buchner funnel
with Whatman No. 1 filter paper and concentrated under reduced
3
further chromatographed on Sephadex LH-20 using CHCl /MeOH
3
(3:1, 800 mL) to yield compound 6 (9 mg, 0.01% yield of extract). Fr.
6 (728 mg, 0.83% yield of extract) was chromatographed on Sephadex
pressure using a rotary evaporator (Bu
̈
chi R-200) at 40 °C. The extract
88 g, 3.83% yield of dried plant material) was redissolved in MeOH
100 mL) and suspended in 800 mL of distilled H O. The aqueous
(
(
LH-20 using CHCl
of extract), which was further purified using Sephadex LH-20 with
CHCl /MeOH (2:1, 800 mL) to yield compounds 7 (26 mg, 0.03%
/MeOH, affording subfr. 2 (225 mg, 0.26% yield
3
2
layer extract (88 g) was suspended in different solvents and
3
successively extracted three times in petroleum ether (pet. ether) (3
yield of extract) and 8 (17 mg, 0.02% yield of extract). Subfr. 3 (110
mg, 0.13% yield of extract) from fr. 6 yielded compound 9 via repeated
×
1.2 L), CHCl (3 × 1.2 L), EtOAc (3 × 1.2 L), and n-BuOH (3 ×
3
1
.2 L). The organic layers were evaporated to give 2.30 g (2.61% yield
of extract) of pet. ether, 4.70 g (5.34% yield of extract) of CHCl3,
8.70 g (21.25% yield of extract) of EtOAc, and 40.00 g (45.45% yield
of extract) of n-BuOH extracts and were tested for tyrosinase
Sephadex LH-20 CC eluted with CHCl
mg, 0.02% yield of extract).
3
/MeOH (2.1:1, 600 mL) (13
1
MF-4 (2.58 g, 2.93% yield of extract) was subjected to a Sephadex
LH-20 column using MeOH/H O of decreasing polarity (100 to 0%
2
E
DOI: 10.1021/acs.jnatprod.7b00564
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 2. H-bond interactions with tyrosinase active site residues and polar interactions with Cu2+ ions in various complexes with (a) kojic acid, (b)
compound 11, (c) compound 3, (d) compound 1, and (e) vitamin C.
Figure 3. Untreated cells (A), cells treated with the positive control α-MSH (B), and cells treated with Myrsine africana extract (C). Analyses with
double immunolabeling. Melanosomes (small green dots) observed around the nucleus (stained blue) confirm melanin transfer. No melanin transfer
was observed between the normal human melanocytes (stained green) and normal human keratinocytes (stained red) treated with M. africana.
H O) as an eluent. Similar fractions were pooled according to TLC
profiles to give three main fractions (1−3). Fr. 1 (0.315 g, 0.36% yield
of extract) was chromatographed on a Sephadex LH-20 column using
2
MeOH/H O and rechromatographed using SephadexLH-20 with
2
F
DOI: 10.1021/acs.jnatprod.7b00564
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
MeOH/H O (1:9, 800 mL) to yield compound 10 (11 mg, 0.01%
yield). Fr. 2 (521 mg) was chromatographed using a Sephadex LH-20
molecules in the active site defined as 6 Å regions around the cocrystal
ligand in tyrosinase protein.
2
column with MeOH/H O. The subfractions were combined (120 mg,
Cell Culture and Chemicals. B16F10 murine melanoma cells
(Highveld Biological, Johannesburg, RSA) were grown in Eagle’s
minimum essential medium (Sigma-Aldrich Co.) supplemented with
10% heat-inactivated fetal bovine serum, 1% antibiotics (100 U/mL
penicillin and 100 μg/mL streptomycin), and 1% fungicide (250 μg/
mL fungizone) (Life Technologies, Johannesburg, RSA) at 37 °C with
2
0
.14% yield of extract) and rechromatographed twice using Sephadex
LH-20 with MeOH/H O (1:6, 500 mL) to yield compounds 11 (31
2
mg, 0.04% yield of extract) and 12 (18 mg, 0.02% yield of extract).
Myricetin 3-O-(2″,4″-di-O-acetyl-α-L-rhamnopyranoside (9): yel-
lowish, amorphous powder; [α]D −187.9 (c 0.020, MeOH), R 0.60,
f
silica gel 60 F254, CH Cl /MeOH (82:18); UV (MeOH) λ 257, 354
5% CO . 2,3-Bis(2-methoxy-4-nitro-5-sulfophenyl]-2H-tetrazolium-5-
2
2
2
max
−1 1
13
nm; IR (KBr) νmax 3417, 1719, 1611, 1439 cm ; H and C NMR
data see Table 2; HR-ESIMS m/z 549.1238 ([M + H] ) (calcd for
C H O , 549.1239); HR-ESIMS m/z 547.1090 [M − H] .
Tyrosinase Enzyme Inhibition Assay. The tyrosinase assay was
carboxanilide salt (XTT), actinomycin D, phenylthiorea, and α-MSH
were obtained from Sigma-Aldrich Co. Cells were subcultured after an
80% confluent monolayer had formed using trypsin−EDTA (0.25%
trypsin containing 0.01% EDTA).
+
−
2
5
25 14
33
conducted as described by Curto et al. with slight modifications. The
enzymatic rate of mushroom tyrosinase (Sigma-Aldrich Co. LLC, St.
Louis, MO, USA) was evaluated spectrophotometrically through the
color change and absorbance increase when L-tyrosine (substrate) was
converted into L-DOPA. A stock solution containing 600 μg/mL of
extract, 0.38 mM DMSO, and 970 μL of K PO buffer (pH 6.5) was
Cell Viability Assay. Cytotoxicity was measured by the XTT
method using the Cell Proliferation Kit II on B16-F10 mouse
34
melanoma cells. The method described by Berrington and Lall was
used to perform the assay. Briefly, 100 μL of cells was seeded in 96-
5
well plates (1 × 10 cells/mL) and incubated for 24 h at 37 °C in 5%
CO for cell adherence. Cells were treated with M. africana extract for
3
4
2
prepared. In a 96-well plate, 70 μL of each sample stock solution was
combined with 30 μL of tyrosinase (333 units/mL in K PO buffer) in
72 h. Actinomycin D (purity >95%) was used as the positive control at
concentrations ranging from 0.00039 to 0.05 μg/mL. After treatment,
XTT (50 μL) was added to a final concentration of 0.30 mg/mL for 2
h. Blank plates were included that were prepared in the same manner,
but did not contain cells. Absorbance was measured at 490 and 690
nm (reference wavelength) using a BIO-TEK Power-Wave XS
multiwell plate reader.
3
4
triplicate, and after 5 min of incubation at room temperature, 110 μL
of substrate (2 mM L-tyrosine) was added to each well. Final
concentrations of the extract, positive control (kojic acid), major
fractions, and pure compounds ranged from 1.5 to 200 μg/mL. The
final concentrations of DMSO present in the samples ranged from
0
.007 to 0.895 μg/mL. The absorbance of the sample wells (Abssample
)
Measurement of Melanin Content. The melanin inhibition
3
6
and the DMSO vehicle control wells (Abscontrol) were determined
kinetically over a period of 30 min at 492 nm using the BIO-TEK
Power-Wave XS multiwell plate reader (A.D.P., Weltevreden Park,
RSA). The percentage tyrosinase inhibition was calculated as follows:
Inhibition percent (%) = 100 − (Abssample/Abscontrol × 100). Kojic acid
assay was conducted as described by Hall and Orlow with slight
modification. The M. africana extract was tested at a concentration of
1
50 μg/mL ( /
IC50). Mouse melanoma B16-F10 cells were plated at 1
2
6
× 10 cells per 500 μL of culture medium, treated with 50 μg/mL
extract, and incubated for 72 h at 5% CO at 37 °C. After incubation,
2
(
purity >98%) was used as the positive control (Sigma-Aldrich Co.).
cells were centrifuged at 1100 rpm for 10 min, and the supernatant was
removed and examined for protein concentration using the Bradford
protein assay as described below. The remaining pellet was washed
twice with phosphate-buffered saline and dissolved in 230 μL of 2 M
NaOH with 20% DMSO at 60 °C. A 200 μL amount of the dissolved
mixture, which contained melanin, was measured for absorbance at
All samples were tested in triplicate, and the IC50 values were
calculated using GraphPad Prism 4 software.
Measurement of DPPH Radical Scavenging Activity. The
3
4
method, described by Berrington and Lall, was followed to
determine the antioxidant (DPPH radical scavenging) activity of the
shoot extract of M. africana and the isolated compounds. The samples
and positive control, ascorbic acid, were prepared at a stock
concentration of 2 mg/mL in EtOH. EtOH was used as the control,
as no DPPH inhibition was expected. In a 96-well plate, serial dilutions
of the samples were prepared, in triplicate, to obtain final
concentrations ranging from 0.78 to 100 μg/mL for the extract,
positive control, and isolated compounds. To all the wells was added
4
90 nm. The melanin content was determined as the absorbance of
melanin per total protein concentration. Phenylthiourea (stock
concentration of 0.100 mM) was used as the positive control
(
(
Sigma-Aldrich Co.).
Measurement of Melanin Transfer. Normal human melanocytes
grown from skin biopsies at the San Gallicano Dermatological
Institute, Rome, Italy) were seeded (2500 cells/well) on coverslips
previously coated with 2% gelatin into 24-well plates and were
incubated at 37 °C for 72 h. Normal human keratinocytes were added
9
0 μL of a DPPH solution in EtOH (0.04 M), except for the blank
plates, where distilled H O was added instead. The plates were
2
2
+
to NHMs and incubated in Medium 154 with a higher Ca
concentration (0.15 mM) and keratinocyte growth supplement for
4 h. The M. africana extract (20 μg/mL) was added to the coculture
incubated for 30 min in a dark room for the DPPH to develop. The
absorbance was measured at a wavelength of 515 nm using the BIO-
TEK PowerWave XS multiwell plate reader. The results given as IC50
values indicated the concentration required to bind 50% of the DPPH
free radicals. The percentage inhibition was calculated as follows:
DPPH radical scavenging activity (%) = [(Abscontrol − Abssample)/
2
and incubated at 37 °C for 24 h. Control cells were treated with
DMSO control (0.1%), the volume equal to that present in the cells
incubated with M. africana extract. α-MSH (purity >97%) (100 mM)
was added as the positive control for 24 h (Sigma-Aldrich Co.).
(
+
Abscontrol)] × 100, where Abscontrol is the absorbance of DPPH radical
EtOH control; Abssample is the absorbance of (DPPH radical +
3
7
Melanin transfer was analyzed by immunofluorescence.
Statistical Analysis. Data recorded are presented as mean ± SD
n = 3). Results were analyzed using one-way analysis of variance
ANOVA) followed by Dunnett’s multiple comparison test using the
sample/positive control) − (blank values of corresponding sample).
All samples were tested in triplicate, and the IC50 values were
calculated using GraphPad Prism 4 software. Ascorbic acid (purity
(
(
GraphPad Prism 4 statistical software. Differences with p < 0.01 (**),
where samples were statistically lower than the positive controls, were
considered to be significant.
>
99%) was used as the positive control (Sigma-Aldrich Co.).
Molecular Docking Analysis. Molecular docking was performed
using the GOLD program. It uses a genetic algorithm considering
35
ligand flexibility and partial protein flexibility. The default docking
parameters were employed for the docking study. The crystal structure
of Agaricus bisporus tyrosinase was used and obtained from the protein
data bank (pdb id: 2Y9X). Protein Preparation Wizard version 1.8.2 of
the GOLD program was used to correct the interaction between
residues and topology. All ligands were corrected for bond order and
minimized to get optimized conformation. The docking protocol was
set by extracting and redocking the tropolone in the tyrosinase crystal
structure with rmsd < 1.0 Å. It was followed by docking of all
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.7b00564.
■
*
S
1H and 13C NMR and COSY spectroscopic data for
compound 9 (PDF)
G
DOI: 10.1021/acs.jnatprod.7b00564
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(
27) Si, Y. X.; Yin, S. J.; Oh, S.; Wang, Z. J.; Ye, S.; Yan, L.; Yang, J.
AUTHOR INFORMATION
■
*
M.; Park, Y. D.; Lee, J.; Qian, G. Y. J. Biomol. Struct. Dyn. 2012, 29,
Corresponding Author
Tel: +27 12 420 2425. Fax: +27 12 420 6668. E-mail: namrita.
lall@up.ac.za.
9
99−1012.
(28) Woo, Y. M.; Kim, A. J.; Kim, J. Y.; Lee, C. H. J. Appl. Biol. Chem.
2011, 54, 47−50.
(29) Briganti, S.; Camera, E.; Picardo, M. Pigm. Cell Res. 2003, 16,
ORCID
1
01−110.
Namrita Lall: 0000-0002-3242-3476
(30) Karg, E.; Odh, G.; Wittbjer, A.; Rosengren, E.; Rorsman, H. J.
Invest. Dermatol. 1993, 100, 209−213.
31) Ebanks, J. P.; Wickett, R. R.; Boissy, R. E. Int. J. Mol. Sci. 2009,
0, 4066−4087.
32) SezerSenol, F.; Tareq Hassan Khan, M.; Orhan, G.; Gurkas, E.;
Erdogan Orhan, I.; SubutayOztekin, N.; Ak, F. Curr. Top. Med. Chem.
014, 14, 1469−1472.
33) Curto, E. V.; Kwong, C.; Hermersdo
Notes
(
1
(
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
2
(
The authors thank the University of Pretoria and the National
Research Foundation for financial support and the Central
Analytical Facilities (CAF), Stellenbosch, South Africa, for
̈
rfer, H.; Glatt, H.; Santis,
C.; Virador, V.; Hearing, V. J.; Dooley, T. P. Biochem. Pharmacol. 1999,
57, 663−672.
1
13
acquiring the IR, H and CNMR, and MS data.
(34) Berrington, D.; Lall, N. Evid. Based Complement. Alternat. Med.
2
(
012, 2012, 1−11.
35) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R. J. Mol.
REFERENCES
■
Biol. 1997, 267, 727−748.
(
(
(
1) Lin, J. Y.; Fisher, D. E. Nature 2007, 445, 843−850.
(36) Hall, A. M.; Orlow, S. Pigm. Cell Res. 2005, 18, 122−129.
(37) Cardinali, G.; Bolasco, G.; Aspite, N.; XLucania, G.; Lotti, L. V.;
Torrisi, M. R.; Picardo, M. J. Invest. Dermatol. 2008, 128, 558−567.
2) Lall, N.; Kishore, N. J. J. Ethnopharmacol. 2014, 153, 61−84.
3) Lan, W. C.; Tzeng, C. W.; Lin, C. C.; Yen, F. L.; Ko, H. H.
Phytochemistry 2013, 89, 78−88.
(
4) Fujiwara, Y.; Sahashi, Y.; Aritro, M.; Hasegawa, S.; Akimoto, K.;
Ninomiya, S.; Sakaguchi, Y.; Seyama, Y. BioFactors 2004, 21, 415−418.
(
(
5) Kim, Y. J.; Uyama, H. Cell. Mol. Life Sci. 2005, 62, 1707−1723.
6) Lee, S. Y.; Baek, N.; Nam, T. G. J. Enzyme Inhib. Med. Chem.
2
(
016, 31, 1−13.
7) Erdogan Orhan, I.; Tareq Hassan Khan, M. Curr. Top. Med.
Chem. 2014, 14, 1486−1493.
(
(
8) Chang, T. S. Int. J. Mol. Sci. 2009, 10, 2440−2475.
9) Watt, J. M.; Breyer-Brandwijk, M. G. The Medicinal and Poisonous
Plants of Southern and Eastern Africa being an Account of their Medicinal
and other Uses, Chemical Composition, Pharmacological Effects and
Toxicology in Man and Animal; E. & S. Livingstone Ltd.: Edinburgh,
1
(
1
(
2
(
962; Vol. 2, p 1457.
10) Kang, L.; Zhou, J. X.; Shen, Z. W. Chin. J. Chem. 2007, 25,
323−1325.
11) Zou, Y.; Tan, C.; Zhu, D. Bull. Korean Chem. Soc. 2009, 30,
111−2113.
12) Nanyingi, M. O.; Mbaria, J. M.; Lanyasunya, A. L.; Wagate, C.
G.; Koros, K. B.; Kaburia, H. F.; Munenge, R. W.; Ogara, W. O. J.
Ethnobiol. Ethnomed. 2008, 4, 14.
(
13) Abbasi, A. M.; Khan, M. A.; Ahmed, M.; Zafar, M. Indian J.
Tradit. Know. 2010, 9, 175−183.
14) Shivanand, P.; Nilam, M.; Viral, D. Int. J. PharmTech Res. 2010,
(
2
(
(
, 632−639.
15) Gupta, P. Int. Interdiscip. Res. J. (Online) 2015, 5, 165−174.
16) Packianathan, N.; Kandasamy, R. Funct. Plant Sci. Biotechnol.
2
011, 5, 94−97.
17) Momtaz, S.; Lall, N.; Basson, A. S. Afr. J. Bot. 2008, 74, 577−
82.
18) Zou, Y. P.; Tan, C. H.; Wang, B. D.; Zhu, D. Y.; Kim, S. K. Helv.
Chim. Acta 2008, 91, 2168−2173.
19) Arot, L. O. M.; Midiwo, J. O.; Kraus, W. Phytochemistry 1996,
3, 1107−1109.
20) Lee, I. S.; Kim, I. S.; Lee, Y. M.; Lee, Y.; Kim, J. H.; Kim, J. S.
(
5
(
(
4
(
Chem. Pharm. Bull. 2013, 61, 662−665.
(
(
21) Saleh, N. A. M.; Towers, G. H. N. Phytochemistry 1971, 10, 897.
22) Facundo, V. A.; Andrade, C. H. S.; Silveira, E. R.; Braz-Filho, R.;
Hufford, C. D. Phytochemistry 1993, 32, 411−415.
23) Costa, S. S.; Jossang, A.; Bodo, B.; Souza, M. L. M.; Moraes, V.
L. G. J. Nat. Prod. 1994, 57, 1503−1510.
24) Bilia, A. R.; Palme, E.; Marsili, A.; Pistelli, L.; Morelli, I.
(
(
Phytochemistry 1993, 32, 1078−1079.
(
(
25) Agrawal, P. K. Phytochemistry 1992, 31, 3307−3330.
26) Chen, Q. X.; Kubo, I. J. Agric. Food Chem. 2002, 50, 4108−4112.
H
DOI: 10.1021/acs.jnatprod.7b00564
J. Nat. Prod. XXXX, XXX, XXX−XXX