Synthesis and in vitro biological activity of retinyl polyhydroxybenzoates, novel hybrid retinoid derivatives

Article abstract of DOI:10.1016/j.bmcl.2008.11.028

A new hybrid derived from retinol was designed to improve the stability and anti-oxidant activity of retinol and also to add whitening properties besides its usual anti-aging properties. A variety of polyhydroxybenzoates of retinol were prepared either by

Full text of DOI:10.1016/j.bmcl.2008.11.028

Bioorganic & Medicinal Chemistry Letters 19 (2009) 508–512
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and in vitro biological activity of retinyl polyhydroxybenzoates,
novel hybrid retinoid derivatives
Sungbum Kim a, Youngmin Kim a, Younggyu Kong a, Hyojung Kim b, Jahyo Kang a,
*
a Department of Chemistry, Sogang University, 1 Shinsudong, Mapoku, Seoul 121-742, Republic of Korea
b Department of Skin Science and Cosmetics, ENPRANI, Shinhung-dong, Inchon 400-103, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
A new hybrid derived from retinol was designed to improve the stability and anti-oxidant activity of ret-
inol and also to add whitening properties besides its usual anti-aging properties. A variety of poly-
hydroxybenzoates of retinol were prepared either by base-catalysis or by direct esterification of retinol
and screened for such desirable properties by analyzing the in vitro biological activity of the hybrids.
Some of the retinol derivatives enhanced their thermal stability and decreased photosensitivity, and
exhibited an activity in collagen synthesis similar to that of retinol. In addition, the retinyl gallate 6
showed higher activities in free radical scavenging and melanogenesis inhibition than retinol. Thus,
owing to its excellent stabilities, retinyl gallate 6 may be conveniently used not only as an additive for
cosmetics for prevention and improvement of skin aging and whitening but also as medicine for the
treatment of skin troubles.
Received 4 August 2008
Revised 4 November 2008
Accepted 11 November 2008
Available online 14 November 2008
Keywords:
Polyhydroxybenzoate
Retinol
Gallic acid
Anti-aging
Ó 2008 Elsevier Ltd. All rights reserved.
Whitening
The irreversible photo-aging of skin, which is a principal cause
of skin aging,1,2 activates cell surface growth factors and cytokine
receptors. More specifically, UV irradiation stimulates the tran-
scription factor activator protein 1 (AP-1), which regulates the
expression of matrix-metalloproteinase (MMP) to degrade skin
collagen. Thus, photoaging is associated with increased AP-1 activ-
ity, increased MMP expression, enhanced collagen degradation,
and decreased collagen synthesis, all of which result in changes
within the matrix of the dermis and at the dermal–epidermal junc-
tion.3–7
Retinoids bind to nuclear receptors to activate the transcrip-
tion of specific DNA sequences, which results in modulation of
gene expression. Thanks to this ability to modulate genes in-
volved in cellular differentiation and proliferation, retinoids are
good candidates for both treating and preventing the photoaging
process. Indeed, retinoids such as trans-retinoic acid (RA) 18–14
and more tolerable retinol 215 have been the agents of choice to
repair the skin damaged by chronological aging or photoaging.
However, the fat-soluble RA causes side effects such as skin irri-
tation, skin dryness, wounds, and scraping16–19 attributed to the
carboxyl end group in RA.20–22 Many efforts have been made to
modify the structure of RA. Unfortunately, however, these reti-
noids are unstable towards light, oxygen, heat, peroxides, acid
or water.23
CO2H
OH
1
2
trans-retinoic acid (RA)
retinol
Herein we tried to modify retinol to enhance the stability of the
parent retinoid by attaching a chemically stable moiety for intramo-
lecular energy transfer, by which the retinoid part would remain in-
tact. (Scheme 1)24 Additionally, while maintaining the usual
collagen-enhancing properties of retinoids, one might be able to
add, by proper choice of the attachment G, such new biologically
favorable properties as potent anti-oxidant activity and whitening
activity for the reasons stated below, which cannot be expected
from the unmodified retinoids.
The formation of advanced glycosylation end-products (AGEs)
that darkens the skin with their color produces dimerized imidazo-
lium products. It is thus one of the causes of skin aging triggered by
reactive oxygen species (ROS) besides the usual chronological pro-
duction (Scheme 2)25–28. Consequently, it follows that blocking the
formation of ROS may contribute to slowing of the skin aging
processes.
Moreover, the formation of melanin, which is the main source
of skin pigmentation, is the result of 3 consecutive enzymatic oxi-
dation reactions (Scheme 3)29–31. Thus, appropriate anti-oxidants
are expected to provide tools for reducing the rate of skin aging
and also of pigmentation.
* Corresponding author. Tel.: +82 2 705 8439; fax: +82 2 701 0967.
E-mail address: kangj@sogang.ac.kr (J. Kang).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.11.028

S. Kim et al. / Bioorg. Med. Chem. Lett. 19 (2009) 508–512
509
M
*
hν
M
*
M
O
O
O
O
O
O
hν
: moiety for stabilizer for retinoid
and additional functions
Retinoid
R
R
R
excitation
3
4
5
intramolecular
energy
transfer
Scheme 4. Energy dissipation by excited-state intramolecular transfer of proton
*
Retinoid*
and metal.
Retinoid
excited
excited
energy
degradation
energy
dissipation
The retinyl salicylates 4 were prepared in 25–40% yields by the
base-catalyzed reaction (90 °C, 1–2 h) of retinol 2 with cyanometh-
yl salicylates 3, which were prepared from the corresponding sali-
cylic acid and ClCH2CN in the presence of potassium carbonate in
dimethyl acetamide at reflux.40 On the other hand, the correspond-
ing gallate 6 was prepared in 41% yield directly from retinol 2 and
the carboxylic acid 5 in THF in the presence of DCC and a catalytic
amount of DMAP (Scheme 5).
dissipation
various
product
Retinoid
protected by
energy transfer
Scheme 1. Schematic explanation of internal energy transfer to the modifying
group G for stabilization of retinol.
Thus, the following 13 retinyl polyhydroxybenzoates, 6–18,
were prepared by these routes (Fig. 1), among which salicylates 7
and 8 have already been reported to possess an analgetic, anti-
inflammatory, anti-thrombotic and anti-pyretic effect.41
The 13 retinyl polyhydroxybenzoates, 6–18, were screened
for formation of procollagen (anti-aging activity) and % inhibi-
tion of melanin formation (whitening activity). In order to eval-
uate the activities of the compounds, the relative activities
were calculated as % of the references, retinol and resveratrol,
respectively.
As can be seen in Table 1, all compounds except for 10 and 17
showed an acceptable level of cell toxicity, and the salicylates 7
and 8 exhibited activities inferior to the rest of the compounds.
Thus, by and large, both the anti-aging and the whitening activities
of each compound increased at the increase of the electron density
of the aromatic ring (and thus the increase of oxidizability), which
supports our initial assumption that anti-oxidant activity is essen-
tial for anti-aging and whitening of the skin.
After a considerable amount of work, polyhydroxybenzoates
were selected as the stabilizing attachment G: the choice was
based on the many reports that polyhydroxybenzoates, espe-
cially the o-hydroxyaroyl groups, dissipate the excess energy
by excited-state intramolecular proton-transfer (Scheme 4;
M = H).32,33
Furthermore, the salicylic acid derivatives34 and/or other poly-
hydroxybenzoates35,36 are known to capture ROS by themselves,
and thus known to be potent anti-oxidants. Moreover, these ben-
zoates can bind metals (Scheme 4, M = metal) and the resulting
chelates behave as superoxide dismutase and catalase37–39 and
thus behave as a scavenger of Fenton-active Fe which produces
the biologically damaging hydroxyl radical. All these indications
point to the possibility that the hybrid between retinol and poly-
hydroxybenzoic acid could be used for treatment of skin aging
and pigmentation at the same time.
collagen
N+
dimerization
O
collagen
collagen
wrinkle
CHO
NH
O
N
O
H
HO
H
OH
H
NH2
collagen
O
O
promoted
by ROS
OH
OH
OH
cross-linked
collagen
H
H
H
or
chronologically
skin yellowing
age spots
Skin aging
OH
sugar
advanced glycosylation
end-products (AGEs)
Scheme 2. Generation of AGE and cross-linked collagen by reactive oxygen species.
phenylalanine
CO2H
HO
HO
CO2H
NH2
CO2H
tyrosinase*
oxidation
hydroxylase
NH2
tyrosine
NH2
oxidation
HO
phenylalanine
DOPA
O
CO2H
melanin
(a polymer)
NH2
oxidation
O
polymerization
DOPA-quinone
darkening of skin
Scheme 3. The route to melanin by three consecutive oxidation reactions and polymerization starting from phenylalanine.

510
S. Kim et al. / Bioorg. Med. Chem. Lett. 19 (2009) 508–512
OH
NCCH2O2C
O
O
OH
3
O
O
Y
OH
K2CO3, DMA
90 oC, 1~3 h
2
4
Y
HO2C
OH
OH
OH
OH
5
OH
DCC, DMAP(0.1 eq)
THF, rt, 20 h
6
OH
Scheme 5. Preparation of retinyl polyhydroxybenzoates.
O
O
Ar
OH
OH
OH
OH
OAc
OH
OH
OMe
OH
Ar =
OH
OH
OMe
7
8
9
10
11
12
13
OH
OH
OH
OH
OH
OH
OH
OMe
HO
MeO
HO
OMe
OH
OH
14
NHAc
17
15
16
18
6
Figure 1. Representative retinyl polyhydroxybenzoates prepared in this study.
Although the precise experimental set-ups for close examina-
tions were different, the trends of the compounds in anti-aging
activities (formation of procollagen type I and inhibition of MMP-
1) (Table 2) and whitening activities (inhibition rate of tyrosinase
and melanin formation) (Table 3) were almost the same as in the
Table 1
Relative activities of the retinyl polyhydroxybenzoates by crude screening (relative
activities) for anti-aging and whitening activities
Compound Cell toxicity, IC50
mM
,
Collagen assaya
Melanogenesis
inhibitionb
Ref: Resveratrol
(0.001%)
Ref: Retinol
(0.0005%)
7
8
9
0.15
0.15
0.027–0.057
<<0.01
0.12
0.014ꢀ0.028
0.057ꢀ0.115
0.42ꢀ0.85
0.014ꢀ0.034
0.23
<<0.01
0.014–0.028
0.035
46% (0.002%)
61% (0.002%)
116% (0.0005%)
66% (0.0001%)
26% (0.001%)
55% (0.0005%)
81% (0.001%)
37% (0.0005%)
102% (0.0005%)
54% (0.005%)
15% (0.001%)
55% (0.0005%)
91% (0.0002%)
34% (0.002%)
32% (0.002%)
22% (0.001%)
105% (0.001%)
41% (0.001%)
95% (0.001%)
30% (0.001%)
18% (0.001%)
112% (0.001%)
28% (0.001%)
—
Table 2
Effects of 6 anti-aging candidates on cell viability, procollagen type I and MMP-1 in
normal human dermal fibroblasts
10
11
12
13
14
15
16
17
18
6
Sample
Final
concn
Percentage of negative controla
Cell
Procollagen type MMP-
viability
Ic
1d
9
10
12
15
18
6
10
l
M
125.9
71.8
91.8
88.1
92.6
84.2
95.4
105.2
95.2
70.0
NTe
16.9
1
l
M
NTe
10
l
M
78.7
85.5
89.2
115.9
127.5
NT
À32.2
À63.0
À17.4
À19.3
NT
95% (0.001%)
73% (0.0002%)
1
l
M
10
l
M
5
l
M
a
Expression rate of procollagen as % of the reference, retinol. The values in
L-ascorbic acidb
Retinoic acidb
(À)Epigallocatechin
gallateb
200 lM
parentheses refer to weight % concentration.
10
l
M
59.2
46.7
b
Percentage inhibition rate of melanin formation as % of the reference, resve-
10
l
M
NT
ratrol. The values in parentheses refer to weight % concentration.
DMSOa
1%
100
100
0
a
Dimethyl sulfoxide (DMSO) was used as a negative control.
Positive controls.
Expression rate of procollagen type I.
Inhibition rate of MMP-1 expression.
NT, not tested.
From these results, the six retinyl esters, 9, 10, 12, 15, 18 and 6,
were selected, despite some cell toxicity of 10, and each of them
was scrutinized closely for the both activities of anti-aging and
whitening.
b
c
d
e

S. Kim et al. / Bioorg. Med. Chem. Lett. 19 (2009) 508–512
511
Table 3
100
Effects of 6 whitening candidates on cell viability, tyrosinase activity and melanin
formation in cultured B16/F10 melanoma cells
Retinol
15
18
6
No.
Concn
M)
0
Cell
Tyrosinase
activityb
0
Melanin
formationc
0
(
l
viabilitya
100
80
60
40
DMSOd
9
5
5
5
5
5
5
l
M
M
M
M
M
M
107.1
97.0
99.7
102.5
102.0
83.9
38.1
36.7
31.3
45.4
41.9
49.6
46.9
16.2
18.6
25.8
12.7
9.5
10
12
15
18
6
l
l
l
l
l
27.5
39.3
Melasolve 10 ppm
95.4
a
Percentage of negative control.d
Percentage inhibition rate of tyrosinase activity.
Percentage inhibition rate of melanin formation.
Dimethyl sulfoxide (DMSO) was used as a negative control.
b
c
d
e
2 days
7 days
Time (days)
14 days
28 days
Trade name for thymyl 3,4,5-trimethoxycinnamate (Amore-Pacific) which was
used as a positive control.
Figure 2. Thermal stability test at 40 °C.
crude experiments (Table 1). Thus, among the three compounds
(15, 18, and 6) which showed good activities in anti-aging and
whitening, the gallate 6 seemed to be the best candidate.
Subsequently, retinyl gallate 6 was compared with other com-
monly accepted agents for skin troubles.42,43 First of all, it is less
cell-toxic than retinol, and its free radical scavenging activity
was truly outstanding as expected. Although the degree of promo-
tion of collagen synthesis by 6 was less than a half, it was more
effective in inhibiting elastase, which cleaves elastin, an elastic fi-
ber that determines the mechanical properties of connective tissue.
Moreover, retinyl gallate 6 was shown to be potentially a good
whitening agent as its activities toward inhibition of tyrosinase
and melanogenesis were better than that of arbutin but less than
that of resveratrol (Table 4).
To compare the time-dependent thermal stability of retinyl gal-
late 6 with retinol 2, the phase stability in a thermohygrostat
(humidity: 58%) at 40 °C without exclusion of atmospheric oxygen
was examined for 4 weeks by a quantitative HPLC analysis. As
shown in Figure 2, almost a half of retinyl gallate 6 decomposed
after 4 weeks under this condition, while the thermal stability of
retinol was much less than that of retinyl gallate with almost
70% of the material decomposed. These results clearly indicate that
the new hybrid retinyl polyhydroxylbezoate 6 is much more stable
than retinol itself (Fig. 2).
0.8
0.6
0.4
0.2
0.0
Gallic acid
15
18
6
Retinol
260
280
300
320
340
360
380
400
Wavelength(nm)
Figure 3. UV spectra of retinoids (concentration: 2.5 Â 10À4 M (MeOH)).
In conclusion, we have discovered retinyl gallate 6 as a
new class of retinyl polyhydroxybenzoate anti-oxidant that
exhibits not only retinoid activity but also melanin inhibition
effect.
Retinyl gallate 6 has shorter absorption maxima at 260 and
295 nm than the other retinyl polyhydroxybenzoates (kmax 325–
345 nm), gallic acid (kmax 275 nm) and retinol (kmax 323 nm)
(Fig. 3). Consequently, retinyl gallate 6 is able to absorb light in
the more damaging UVB region (wavelength 290–315 nm) and
thus able to protect the skin from photo-aging, although such util-
ity would seem to have a rather slim chance in reality because the
concentration of the active ingredient as an anti-aging and/or
Acknowledgments
This work was initiated by a grant of the Korea Health 21 R&D
Project, Ministry of Health & Welfare, Republic of Korea. (0405-
CS00-0501-0010) and supported by the Special Research Grant of
Sogang University.
whitening agent is going to be rather low (ꢀ100
lM).
Table 4
Summary of biological activity screening results of retinyl gallate 6
Activities
Concn (
l
M)
6
Retinol
Ursolic acid
Resveratrol
Arbutin
Ascorbic acid
Cell toxicity (IC50
Free radical scavenging activity (%)
,
l
M)
35
50
—
—
33.4
—
—
3.5
—
—
900
5.6
16.7
—
—
—
—
—
—
—
—
—
—
5
10
5
5
10
5
31.1
62.8
13.5
12.3
15.3
2.5
19.1
38.6
—
—
—
Increase of collagen synthesis (%)
Elastase inhibitory activity (%)
—
10.3
14.1
—
Tyrosinase inhibitory activity (%)
Inhibition of melanogenesis (%)
—
—
36.2
25.5
À1.5
—
—
5
18.6
—
5.5

512
S. Kim et al. / Bioorg. Med. Chem. Lett. 19 (2009) 508–512
20. Dawson, M.I., Hobbs, P.D. in: The Retinoids. Biology, Chemistry, and Medicine,,
Reference and notes
Raven Press: New York, 1994, pp. 5–178..
21. Bernard, B. A. Skin Pharmacol. 1993, 6, 61.
1. Kang, S.; Fisher, G. J.; Voorhees, J. J. Clin. Geriatr. Med. 2001, 17, 643.
2. Fisher, G. J.; Wang, Z. Q.; Datta, S. C.; Varani, J.; Kang, S.; Voorhees, J. J. N. Engl. J.
Med. 1997, 337, 1419.
3. Ghersetich, I.; Lotti, T.; Campanile, G.; Grappone, C.; Dini, G. Int. J. Dermatol.
1994, 33, 119.
4. Oksvold, M. P.; Huitfeldt, H. S.; Ostvold, A. C.; Skarpen, E. J. Cell Sci. 2002, 115,
793.
5. Lirvall, M.; Ljungqvist-Hoddelius, P.; Wasteson, A.; Magnusson, K. E. Biosci. Rep.
1996, 16, 227.
6. Pienimaki, J. P.; Rilla, K.; Fulop, C.; Sironen, R. K.; Karvinen, S.; Pasonen, S.;
Lammi, M. J.; Hascall, V. C.; Tammi, M. I. J. Biol. Chem. 2001, 276, 20428.
7. Pasonen-Seppanen, S.; Karvinen, S.; Torronen, K.; Hyttinen, J. T.; Jokela, T.;
Lammi, M. J.; Tammi, M. I.; Tammi, R. J. Invest. Dermatol. 2003, 120, 1038.
8. Varani, J.; Fisher, G. J.; Kang, S.; Voorhees, J. J. J. Invest. Dermatol. Symp. 1998, 3, 57.
9. Brodell, L. P.; Asselin, D.; Brodell, R. T. J. Am. Acad. Dermatol. 1993, 27, 621.
10. Kligman, L. H.; Duo, C. H.; Kligman, A. M. Connect Tissue Res. 1984, 12, 139.
11. Lee, K. H.; Tong, T. G. J. Pharm. Sci. 1970, 59, 1195.
12. Varani, J.; Warner, R. L.; Gharaee-Kermani, M.; Phan, S. H.; Kang, S.; Chung, J.
H.; Wang, Z. Q.; Datta, S. C.; Fisher, G. J.; Voorhees, J. J. J. Invest. Dermatol. 2000,
114, 480.
13. Fisher, G. J.; Voorhees, J. J. FASEB J. 1996, 10, 1002.
14. Orfanos, C. E.; Zoubooulis, C. C.; Almond-Roelser, B.; Geilen, C. C. Drugs 1997,
53, 358.
15. Varani, J.; Warner, R. L.; Gharaee-Kermani, M.; Phan, S. H.; Kang, S.; Chung, J.
H.; Wang, Z. Q.; Datta, S. C.; Fisher, G. J.; Voorhees, J. J. J. Invest. Dermatol. 2000,
114 3, 480.
22. Chandraratna, RAS. Br. J. Dermatol. 1996, 135, 18.
23. Varani, J.; Warner, R. L.; Gharaee-Kermani, M.; Phan, S. H.; Kang, S.; Chung, J.
H.; Wang, Z. Q.; Datta, S. C.; Fisher, G. J.; Voorhees, J. J. J. Invest. Dermatol. 2000,
114, 480.
24. Tonnesen, H. H. Int. J. Pharm. 2001, 225, 1.
25. Cai, W.; He, J. C.; Zhu, L.; Chen, X.; Striker, G. E.; Vlassara, H. Am. J. Physiol. 2008,
294, C145.
26. Ha, H.; Lee, H. B. Nephrology 2005, 10, S7.
27. Au, V.; Madison, S. A. Arch. Biochem. Biophys. 2000, 384, 133.
28. Kakehashi, A.; Akiba, J.; Ueno, N.; Chakrabarti, B. Biochem. Biophys. Res.
Commun. 1993, 196, 1440.
29. Lin, J. Y.; Fisher, D. E. Nature 2007, 445, 843.
30. Tobin, D. J. Chem. Soc. Rev. 2006, 35, 52.
31. Rescigno, A.; Sollai, F.; Pisu, B.; Rinaldi, A.; Sanjust, E. J. Enzyme Inhib. Med.
Chem. 2002, 17, 207.
32. Hagiri, M.; Ichinose, N.; Kinugasa, J.-I.; Iwasa, T.; Nakayama, T. Chem. Lett. 2004,
33, 326.
33. Chou, P.-T.; Chiou, C.-S.; Yu, W.-S.; Wu, G.-R.; Wei, T.-H. Chem. Phys. Lett. 2003,
370, 747.
34. Powell, S. R. Free Radic. Res. 1994, 21, 355.
35. Furuno, K.; Akasako, T.; Sugihara, N. Biol. Pharm. Bull. 2002, 25, 19.
36. Strlic, M.; Radovic, T.; Kolar, J.; Pihlar, B. J. Agric. Food. Chem. 2002, 50, 6313.
37. Jay, D.; Jay, E. G.; Medina, M. A. Arch. Med. Res. 1999, 30, 93.
38. Sorenson, J. R. J. Prog. Med. Chem. 1989, 26, 506.
39. Biology of Copper Complexes; Sorenson, J. R. J. Ed.; Humana Press, 1987.
40. Porco, J. A.; Shen, R.; Lin, C. T.; Bowman, E. J.; Bowman, B. J. J. Am. Chem. Soc.
2003, 125, 7889.
41. Weischer, C.H., Oestreich, W.D.E. Patent 4,415,204, 1995.
42. Stefania, B.; Emanuela, C.; Mauro, P. Pigment Cell Res. 2003, 16, 101.
43. Curto, E. V.; Kwong, C.; Hermersdorfer, H.; Santis, C.; Virador, V.; Vincent, J.;
Hearing, Jr.; Thomas, P. D. Biochem. Pharmacol. 1999, 57, 663.
16. Kim, B. H.; Lee, Y. S.; Kang, K. S. Toxicol. Lett. 2003, 146, 65.
17. Kang, S.; Duell, E. A.; Fisher, G. J.; Datta, S. C.; Wang, Z. Q.; Reddy, A. P.;
Tavakkol, A.; Yi, J. Y.; Griffiths, C. E.; Elder, J. T.; Voorhees, J. J. J. Invest. Dermatol.
1995, 105, 549.
18. Nau, H. Skin Pharmacol. 1993, 6, 35.
19. Mills, O. H., Jr.; Berger, R. S. J. Am. Acad. Dermatol. 1998, 38, S11.