Synthesis and biological evaluation of sulfur-containing cinnamate and salicylate derivatives

Article abstract of DOI:10.1248/cpb.56.369

UV irradiation induced formation of reactive oxygen radical species and matrix metalloproteinases (MMPs) are thought to be involved in photo-damage to the skin. MMP-1 is the major collagenolytic enzyme responsible for collagen destruction in skin tissue. To develop new anti-photoaging agents, a series of 2,2′-dithiocinnamate derivatives and 2,2′-dithio or 2-thiobenzoate derivatives were designed and synthesized. The biological activities of the synthesized compounds were assayed for ABTS [2,2′-azinobis-(3-ethyl-benzo- thiazoline-6-sulfonic acid)] radical scavenging activity, MMP-1 inhibitory activity, and cytotoxicity to human dermal fibroblast cells. Compounds with potential of resistance to UV irradiation were identified. These compounds are expected to be useful for preventing photo-damage to the skin.

Full text of DOI:10.1248/cpb.56.369

March 2008
Notes
Chem. Pharm. Bull. 56(3) 369—373 (2008)
369
Synthesis and Biological Evaluation of Sulfur-Containing Cinnamate and
Salicylate Derivatives
a
,b
a
,c
*
*
Chih-Chia CHIANG, Tsu-Chung CHANG, Hou-Jen TSAI, and Ling-Yih HSU
a Department of Applied Chemistry and Materials Science, Institute of Technology, National Defense University; Ta-shi,
Tao-yuan 335, Taiwan, ROC: b Department of Biochemistry, National Defense Medical Center; Neihu 114, Taiwan, ROC:
and c Department of Biological Science and Technology, China Institute of Technology; Nangang, Taipei 115, Taipei,
Taiwan, ROC. Received August 12, 2007; accepted December 18, 2007; published online December 20, 2007
UV irradiation induced formation of reactive oxygen radical species and matrix metalloproteinases (MMPs)
are thought to be involved in photo-damage to the skin. MMP-1 is the major collagenolytic enzyme responsible
for collagen destruction in skin tissue. To develop new anti-photoaging agents, a series of 2,2ꢀ-dithiocinnamate
derivatives and 2,2ꢀ-dithio or 2-thiobenzoate derivatives were designed and synthesized. The biological activities
of the synthesized compounds were assayed for ABTS [2,2ꢀ-azinobis-(3-ethyl-benzo-thiazoline-6-sulfonic acid)]
radical scavenging activity, MMP-1 inhibitory activity, and cytotoxicity to human dermal fibroblast cells. Com-
pounds with potential of resistance to UV irradiation were identified. These compounds are expected to be useful
for preventing photo-damage to the skin.
Key words free radical scavenger; matrix metalloproteinase inhibitor; UV protectant
Human skin is the most susceptible organ to be damaged the potency of these compounds as UV protective agents was
by ultraviolet (UV) irradiation as it is directly exposed to UV evaluated by analyzing their free radical scavenging and for
light. Increased evidence has shown the generation of reac- MMP-1 inhibitory activities.
tive oxygen species (ROS) in the skin upon UV exposure.1—3)
Chemistry The synthetic routes13—15) to the target com-
Increased ROS generation can overwhelm the intracellular pounds 5a, 5b, 6a, 6b and 7a, 7b are outlined in Charts 1
antioxidant-defense mechanism, resulting in oxidative stress and 2. Reduction of 2-mercaptobenzoic acid with lithium
and oxidative photo-damage to proteins and other macromol- aluminum hydride (LiAlH4) in tetrahydrofuran resulted 2-
ecules in the skin. Oxidative stress is thought to play a cen- mercaptobenzyl alcohol (1), followed by oxidation of 1 with
tral role in initiating and driving the signaling events that pyridinium chlorochromate (PCC) in dichloromethane at
lead to the cell response following exposure to UV irradia- room temperature afforded 2,2ꢀ-dithiodibenzadehyde (2).
tion.4,5) Exposure of human skin to UV radiation causes nu- Triphenyl-phosphanylidene-acetic acid 2-ethyl-hexyl ester
merous biological effects in the skin, including premature (4) was prepared from triphenylphosphine and bromo-acetic
aging characterized by wrinkles, leathery texture and mottled 2-ethyl-hexyl ester, which obtained from the reaction of bro-
pigmentation.6,7) The mechanism of skin wrinkles is prima- moacetic acid with 2-ethyl-hexanol under toluenesulfonic
rily due to the loss of macromolecules making up the dermal acid in toluene at reflux temperature. Thus, converting the
matrix, among which collagen is the major component. The
degradation of collagens is normally tightly controlled by the
activity of a family of zinc-dependent matrix metallopro-
teinases (MMPs) and their natural inhibitors, the tissue in-
hibitors of MMPs (TIMPs).8) Although several MMPs are
expressed in mammalian skin, studies by Brennan et al.3)
showed that the neutralization of MMP-1 removed virtually
all the collagen-degrading activity elaborated by UV-treated
human skin.
Cinnamates and salicylates are the chemical UV filter
widely used throughout the world.9) In this study, the ethyl
and 2-ethylhexyl esters of these compounds were synthe-
sized, as the presence of these groups may diminish the water
solubility of these compounds to various degree, thus, may
help to develop waterproof sunscreen formulation.9) Because
the radical scavenging ability of mercaptans and disulfides is
well-documented, we speculated that the introduction of
these important functional groups into cinnamates or salicy-
lates might yield potent antioxidants and UV protectants. In
addition, we reasoned that the disulfides10,11) and a-mercap-
tocarbonyl group12) would chelate zinc ion at the active-site
of MMP-1, which renders them to be potential MMP in-
hibitors. In this work, the sulfur-containing cinnamates and
Chart 1
salicylates of substituted ester group were synthesized and
∗ To whom correspondence should be addressed. e-mail: tcchang@ndmctsgh.edu.tw; lyh@mail.ndmctsgh.edu.tw © 2008 Pharmaceutical Society of Japan

370
Vol. 56, No. 3
Chart 2
disulfide 2 into the desired products 5a and 5b by a Wittig re-
action of 2 with 4 and triethylphosphonoacetate (TPA) in
tetrahydrofuran (THF) at room temperature, respectively.
Target compounds 6a and 6b were synthesized by acid-cat-
alyzed Fischer esterification of 2-mercaptobenzoic acid with
2-ethyl-hexanol or ethanol, respectively. Finally, the disul-
fides 6a and 6b were converted into thiols products 7a and
7b using a reduction with lithium tri-tert-butoxyaluminohy-
dride (LTBA) in THF at room temperature.
Fig. 1. Viability of HDF Cells Treated with Different Concentrations of
5a—7b
Data were expressed as relative viability and represent meanꢄS.D. (nꢃ6).
Results and Discussion
Cytotoxicity of Synthetic Compounds The cytotoxicity
of compounds 5a—7b was determined by Cell-Counting-
Kit-8 (CCK-8)-assay on human derma fibroblast (HDF)
cells.16) The result presented in Fig. 1 showed the cell viabil-
ity upon treatment with increasing concentrations of test
compounds after 48 h. With less than 50 mM, these com-
pounds displayed no apparent cytotoxicity as compared with
the control cells that was treated with 0.01% DMSO. In-
creased cytotoxicity was observed in cells treated with
100 mM of these compounds.
ABTS Radical Scavenging Assay Scavenging of ABTS
[2,2ꢀ-azinobis-(3-ethyl-benzo-thiazoline-6-sulfonic acid)]
free radical is a convenient method17,18) to evaluate the in
vitro antioxidative activity of antioxidants. In addition to
compounds 5a—7b, trolox and WR2721 were also included
as positive and negative control compounds, respectively, for
this assay (Fig. 2). Trolox, a potent antioxidant, is frequently
used as a positive control in assessing radical scavenging ac-
tivity.19) WR2721 is an inactive prodrug that is activated to
the active thiol metabolite in the cells and that is responsible
for its free radical scavenging activity.20) As shown in Table
1, these compounds displayed various degrees of free radical
scavenging activity, with decreasing activity in the following
order: Troloxꢁ7bꢁ5bꢁ6aꢀ6bꢂ7aꢃ5aꢁWR 2721. Among
the tested compounds, compound 7b showed significant free
radical scavenging activity. The effect may be attributed to
the free sulfhydryl group in the compound. As the ABTS
free radical scavenging assay was carried out in aqueous sys-
Fig. 2. The Chemical Structures of the Control Compounds Trolox,
WR2721, and GM-1489
tem. The weak activity of compound 6a may be due to the these compounds on MMP-1 activity was assessed by resolv-
decreased water solubility of the ethylhexyl group despite the ing the digested collagen with agarose gel electrophoresis.
presence of a free thiol group. For compound 5b, the conju- As shown in Fig. 3, in the absence of MMP-1, the original
gated double bond system in cinnamate moiety and greater form of collagen sample has four major bands (lane 1). In the
water solubility of ethyl ester group may enhance its scav- presence of MMP-1 but with no inhibitory compound (lane
enging activity. Our results also suggested that the disulfide 2, DMSO), the major collagen bands were digested into
linkage may reduce the free radical scavenging activity.
smaller fragments after 18 h of incubation. The starting mate-
Collagen Degradation Assay In the collagen degrada- rial for our synthetic compounds, thiosalicylic aicd, was used
tion assay,3,7) MMP-1 enzyme and rat tail collagen were incu- as a negative control for this assay and which showed no ap-
bated with synthetic compounds. The inhibitory potency of parent MMP-1-inhibitory activity (lane 3, S). The known

March 2008
371
Table 1. ABTS Radical Cation Scavenging Activity of Compounds 5a—
7b
Compound (100 mM)
Scavenging activity (%)
Trolox
WR2721
5a
15.3ꢄ1.3
0.1ꢄ0.1
1.8ꢄ0.9
3.4ꢄ0.7
2.8ꢄ1.0
2.5ꢄ0.4
1.8ꢄ0.3
6.7ꢄ0.7
5b
6a
6b
7a
7b
Results represent meanꢄS.D. (nꢃ3).
Fig. 4. Inhibitory Effect of Compounds 5a—7b on MMP Zymography
(A) Casein-zymography. MMP-1 activity in conditioned medium from UVB
(40 mJ/cm2)-irradiated HDF cells was analyzed in the absence or presence of 1 mM test
compounds. (B) Relative inhibitory potency of compounds 5a—7b on MMP-1 activity.
The gel image in panel A was quantitated and the results are expressed as relative in-
tensity of the clear band in the gel (arrow indicated). Each bar represents the
meanꢄS.D. of three independent experiments. ∗ pꢅ0.05; ∗∗ pꢅ0.01 compared with
control (lane 2) group.
have different action mechanism in inhibition of MMP activ-
ity.
Casein-Zymography In casein-zymography,21) the in-
hibitory activity of our synthetic compounds on MMP activ-
ity was assessed using conditioned medium from UVB-irra-
diated HDF cells that were treated with 1 mM of our synthetic
compounds. The MMP activity in these conditioned medium
was assessed by casein-zymography. As shown in Fig. 4, the
efficacy of these compounds on inhibition of the caseinolytic
activity was in the following order: 5bꢁ5aꢀ6bꢀ6aꢁ7aꢀ
7b.
Fig. 3. Inhibition of MMP-1-Mediated Collagen Degradation by Com-
pounds 5a—7b
(A) Collagen degradation. Type I collagen was cleaved by purified MMP-1 in the
presence of various compounds. S: Thiosalicylic acid (starting material); G: GM-1489
(broad-spectrum inhibitor of MMPs, Fig. 2). (B) Relative inhibitory potency of test
compounds on MMP-1 activity. The results are expressed as relative intensity of the
third band from top of the gel (arrow indicated). Each bar is the meanꢄS.D. of three in-
dependent experiments. ∗ pꢅ0.05; ∗∗ pꢅ0.01 compared with control (lane 1) group.
In this study, two methods were employed to evaluate the
MMP-1 inhibitory activity of these compounds. The collagen
degradation assay showed in vitro inhibition of MMP-1 en-
MMP-1 inhibitor, GM-1489, was used as a positive control zyme activity. In contrast, the casein-zymography did not
for protection of collagen from further degradation (lane 10, show simple MMP-1 inhibition. In fact, the effect as seen
G). To evaluate the potency of these compounds as MMP-1 casein-zymography method includes the inhibition of MMP-
inhibitors, the relative intensity of the third major band from 1 gene transcription or translation as well as the inhibition of
the top of the gel was quantitated (arrow marked). The MMP-1 activity itself. Since the results obtained from these
MMP-1 inhibitory activity of these compounds occurred in two methods appear to be consistent, MMP-1 inhibition by
the following order: GM-1489ꢁ5bꢀ6aꢀ5aꢁ6bꢁ7bꢀ7a. these compounds in zymography may be mainly due to direct
Results from this assay indicate that compounds 5a, 5b, and inhibition of MMP-1 enzyme activity. However, it is also
6a displayed greater MMP-1 inhibitory activity than com- possible that the greater activity of compounds 5a and 5b in
pounds 7a and 7b. These results suggest that the dimeric cin- zymography assay was resulted from their UV-filtering ef-
namate or salicylate structure as linked by disulfide bond fect. As compounds 5a and 5b possess more extensive conju-
may chelate zinc ion and may be necessary for MMP inhibi- gating double system of the cinnamate moiety and may better
tion.10,11) It should be noted that our compounds are different protect the cells against UV-irradiation.
from the peptide-mimicking GM-1489, which displayed
In summary, we have synthesized a series of sulfur-con-
greater efficacy as MMP inhibitor. The difference in chemi- taining cinnamate and salicylate derivatives in this study. Our
cal structure suggests that our compounds and GM-1489 may results demonstrated that these sulfur-containing compounds

372
Vol. 56, No. 3
The product was stored in the refrigerator to give product 0.34 g. Yield:
30%. mp: 136—140 °C. Rf 0.5 (n-hexane). 1H-NMR (CDCl3) d: 7.91—7.83
(6H, m, ArH), 7.78—7.73 (3H, m, ArH), 7.67—7.63 (6H, m, ArH), 5.44
(2H, d, Jꢃ13.8 Hz, CH2), 3.85 (2H, d, Jꢃ7.2 Hz, OCH2), 1.36—1.32 (1H,
m, CH), 1.15—1.06 (8H, m, CH2), 0.82 (3H, t, Jꢃ6.9 Hz, CH3), 0.72 (3H, t,
exhibit evident antioxidative activity that enables them to be
potential antioxidative. Among these compounds, 5a, 5b, and
6a conferred marked MMP-1 inhibitory activity in col-
lagenolytic assay and caseinolytic zymographic analysis. We
found that the free thiol group is important for marked free Jꢃ7.5 Hz, CH3).
General Procedure to Obtain Compounds 5a and 5b Dry nitrogen
radical scavenging activity and the disulfide linkage may be
important for MMP-1 inhibition. These observations also
suggest that the free radical scavenging and MMP inhibitory
activities of these compounds are conveyed by different
mechanisms. Despite that there is no direct correlation be-
tween free radical scavenging and MMP-1 inhibitory activi-
ties, our studies reveal structure features for the further
development of effective compounds as sunscreening and
MMP-1 inhibitors that may have potential applications. Fur-
ther experiments evaluating the UV-protective effects and ac-
tion mechanism of these compounds in normal cells or mice
are currently in progress.
gas was passed through the apparatus for 15 min. Into the flask was placed a
suspension of 4 or triethyl phophonoacetate (TPA) (20 mmol) in THF
(50 ml), and the 2 (10 mmol) was added. The reaction mixture was main-
tained under a positive pressure from dry nitrogen and was stirred, while
being cooled in an ice-water bath, for 20 min. 60% NaH (15 mmol) was
added in a single portion. It was stirred for an additional 20 min in the ice-
water bath. The bath was then removed, and the reaction mixture was al-
lowed to return to room temperature. The reaction mixture was then allowed
to stand at room temperature for 6 h. The reaction mixture was gravity fil-
tered while it was still warm. The precipitate was washed with additional dry
THF. The filtrate was collected and the solvent was removed by vacuum
evaporation to oil. The resultant oil was purified by chromatography on sil-
ica gel (n-hexane/EtOAcꢃ2/1) to give product.
2,2ꢀ-Dithiobiscinnamic Acid Bis(2-ethylhexyl) Ester (5a): Yield: 45%. Rf
0.4 (n-hexane/EtOAcꢃ2/1). 1H-NMR (CDCl3) d: 8.18 (1H, d, Jꢃ15.9 Hz,
CH), 7.55—7.32 (4H, m, Ar-H), 7.30—7.25 (4H, m, Ar-H), 6.36 (1H, d,
Experimental
General Melting points (mp) were taken on a BUCHI 530 apparatus Jꢃ15.9 Hz, CH), 4.12 (2H, d, Jꢃ5.7 Hz, OCH2, each), 1.57—1.43 (1H, m,
and are uncorrected. Merck Art No. 105554 plates precoated with Silica gel
CH, each), 1.42—1.24 (8H, m, CH2, each), 0.94—0.87 (6H, m, CH3, each).
60 containing fluorescent indicator were used for thin-layer chromatography, 13C-NMR (CDCl3) d: 166.23, 141.28, 136.15, 134.71, 132.76, 129.67,
and Silica gel 60 (Merck Art No 109385, 230—400 mesh) was employed for 127.64, 126.76, 120.19, 66.54, 41.36, 38.346, 29.94, 28.34, 22.29, 13.28,
column chromatography. Evaporations were carried out at ꢅ50 °C using a
10.34. IR (KBr) cmꢇ1: 1672. UV lmax (CH2Cl2) nm (log e): 359 (1.0). FAB-
rotary evaporator at reduced pressure (water aspirator). 1H- and 13C-NMR
MS m/z: 583 (MꢈH)ꢈ. Anal. Calcd for C34H46O4S2: C, 70.06; H, 7.95.
spectra were obtained with a Varian 300 NMR spectrometer at 300 and Found: C, 70.39; H, 7.99.
75 MHz, respectively. Where necessary, deuterium exchange experiments
were used to obtained proton shift assignments. Mass spectra were recorded
2,2ꢀ-Dithiobiscinnamic Acid Diethyl Ester (5b): Yield: 73%. Rf 0.2 (n-
hexane/EtOAcꢃ2/1). 1H-NMR (CDCl3) d: 8.09 (1H, d, Jꢃ15.9 Hz, CH),
on a JEOL J.M.S-300 spectrophotometer. Analytical samples were dried 7.56—7.51 (4H, m, Ar-H), 7.26—7.29 (4H, m, Ar-H), 6.29 (1H, d,
under reduced pressure at 78 °C in the presence of P2O5 for at least 12 h un-
Jꢃ15.9 Hz, CH), 4.26 (2H, q, Jꢃ7.2 Hz, CH2, each), 1.33 (3H, t, Jꢃ7.2 Hz,
less otherwise specified. Elemental analyses were obtained using a Perkin- CH3, each). 13C-NMR (CDCl3) d: 165.90, 140.53, 136.27, 135.61, 131.91,
Elmer 2400 Elemental Analyzer.
129.95, 128.30, 126.69, 120.39, 59.92, 13.64. IR (KBr) cmꢇ1: 1672. UV
lmax (CH2Cl2) nm (log e): 359 (1.0). EI-MS m/z: 414 (Mꢈ). Anal. Calcd for
C22H22O4S2: C, 63.74; H, 5.35. Found: C, 63.80; H, 5.40.
2-Mercaptobenzyl Alcohol (1) The suspension of LiAlH4 (9.8 g,
260 mmol) in dry tetrahydrofuran (THF) (80 ml) was added dropwise to a
solution of 2-mercaptobenzoic acid (16.0 g, 104 mmol) in dry THF (120 ml).
General Procedure to Obtain Compounds 6a and 6b A mixture of 2-
After stirring at room temperature for 4 h, 10% H2SO4 (100 ml) and ethyl ac- mercaptobenzoic acid (60 mmol), 2-ethyl-hexanol or ethanol (120 ml) and
etate (120 ml) were added carefully and the reaction mixture was filtered. concentrated sulfuric acid (3 ml) was heated under reflux for 24 h. The sol-
The residue was washed with ethyl acetate (3ꢆ40 ml). The filtrate and wash- vent of reaction mixture was removed and the residue was partitioned be-
ings were combined, dried with magnesium sulfate, and concentrated under
tween EtOAc and saturated NaHCO3. The organic layer was collected,
reduced pressure to give an oily product which became solid after cooling. washed with brine and concentrated. The residue was purified by chromatog-
Crystallization with a mixture of n-hexane and ethyl acetate, gave pure com-
raphy silica gel (n-hexane/EtOAcꢃ9/1) to give oil product.
2,2ꢀ-Dithiobisbenzoic Acid Bis(2-ethylhexyl) Ester (6a): Yield: 52%. Rf
pound (8.0 g, 55%), mp: 31—32 °C, (lit.13) 31—32 °C).
2,2ꢀ-Dithiodibenzaldehyde (2) A mixture of pyridine chlorochromate 0.7 (n-hexane/EtOAcꢃ9/1). 1H-NMR (CDCl3) d: 8.06 (2H, d, Jꢃ7.8 Hz,
(36.0 g, 167.0 mmol) in dry dichloromethane (150 ml) under nitrogen atmo-
sphere was added dropwise to the mixture of 1 (11.7 g, 84 mmol) in dry
dichloromethane (150 ml). After stirring at room temperature for 4 h, the re-
ArH), 7.77 (4H, d, Jꢃ8.1 Hz, ArH), 7.41 (4H, t, Jꢃ7.2 Hz, ArH), 7.24 (4H,
t, Jꢃ7.2 Hz, ArH), 4.31 (4H, q, Jꢃ2.1 Hz, CH2), 1.54—1.52 (1H, m, CH,
each), 1.49—1.33 (8H, m, CH2, each), 0.94—0.91 (3H, m, CH3, each). 13C-
action mixture was filtered. The residue was washed with dichloromethane NMR (CDCl3) d: 166.19, 139.95, 132.39, 130.84, 127.39, 125.44, 124.93,
(3ꢆ40 ml) and ether (40 ml). The filtrate and washings were combined, and 67.36, 38.35, 29.99, 28.36, 23.41, 22.3, 13.31, 10.39. IR (KBr) cmꢇ1: 1674.
then concentrated to give the solid product. Crystallization with ethanol gave UV lmax (CH2Cl2) nm (log e): 358 (1.0). EI-MS m/z: 530 (Mꢈ). Anal. Calcd
pure compound (6.1 g, 52%), mp: 145—148 °C, (lit.13) 145—148 °C).
for C30H42O4S2: C, 67.89; H, 7.98. Found: C, 68.05; H, 7.99.
Bromo-acetic 2-Ethyl-hexyl Ester (3) In a two-necked round bottom
flask equipped with
72 mmol) and p-TsOH (8.6 g, 45 mmol) were added to a solution of bro-
moacetic acid (8.3 g, 60 mmol) in toluene (200 ml). After stirring under re-
2,2ꢀ-Dithiobisbenzoic Acid Diethyl Ester (6b): Yield: 67%. Rf 0.7 (n-
a
Dean–Stark apparatus 2-ethyl-hexanol (9.4 g, hexane/EtOAcꢃ7/1). 1H-NMR (CDCl3) d: 8.07 (2H, d, Jꢃ1.5 Hz, ArH),
7.74 (4H, d, Jꢃ0.9 Hz, ArH), 7.43—7.22 (4H, m, ArH), 4.44 (2H, q,
Jꢃ7.5 Hz, CH2, each), 1.44 (3H, t, Jꢃ1.2 Hz, CH3, each). 13C-NMR
flux for 18 h, toluene was removed under vacuum. The residue dissolved in (CDCl3) d: 165.98, 139.90, 132.32, 130.82, 127.50, 125.51, 124.89, 60.83,
ethyl acetate (3ꢆ40 ml) was washed with saturated aqueous NaHCO3
13.61. IR (KBr) cmꢇ1: 1672. UV lmax (CH2Cl2) nm (log e): 357 (1.0). EI-
(30 ml), water (2ꢆ30 ml), brine (2ꢆ30 ml), dried over MgSO4 and con- MS m/z: 362 (Mꢈ). Anal. Calcd for C18H18O4S2: C, 59.65; H, 5.01. Found:
centrated. Purification by column chromatography on silica (n- C, 59.68; H, 5.1.
hexane/EtOAcꢃ9/1) afforded a colorless oil 13.5 g, 89%. Rf 0.35 (n-
General Procedure to Obtain Compounds 7a and 7b Dry nitrogen
hexane/EtOAcꢃ9/1). 1H-NMR (CDCl3) d: 4.03 (2H, d, Jꢃ6.0 Hz, OCH2), gas was passed through the apparatus for 15 min. The flask was charged with
3.78 (2H, s, CH2), 1.54—1.52 (1H, m, CH), 1.12—1.26 (8H, m, CH2), 6a or 6b (4.1 mmol) and freshly distilled THF (60 ml). To this well-stirred
0.88—0.85 (6H, m, CH3).
solution, maintained at room temperature was added a 1.25 M solution of
2-Ethylhexyloxycarbonyltriphenylphosphonium Bromide (4) Tri-
lithium tri-tert-butoxyaluminohydride (LTBA) (8.2 mmol) in THF by a sy-
phenylphosphine (0.84 g, 3.2 mmol) was dissolved in benzene or toluene ringe needle over a 15-min period. The resulting clear mixture was stirred
(15 ml). 3 (3.8 ml, 26 mmol) was added to the solution. The mixture was for an additional period of 4 h. Water (1 ml) was added dropwise to destroy
swirled well and almost immediately became cloudy. White precipitate
the excess hydride. The mixture was acidified by the addition of 6.0 N hy-
formed within 10 min. The mixture was allowed to stand at room tempera- drochloric acid to attain a pH of ꢅ3. The organic phase was separated, and
ture for at least 4 h before collecting the precipitate. The solid was filtered, the aqueous phase was extracted with ethyl acetate (3ꢆ30 ml). The com-
washed with toluene, then with hexanes, and allowed to dry over vacuum. bined extracts were washed with saturated brine (30 ml) and dried with
The product was not purified further but used directly for Wittig reactions. MgSO4. Removal of ethyl acetate solvents on a rotary evaporator followed

March 2008
373
by vacuum drying yielded 91% and 94% of 7a and 7b, respectively.
incubated with serum free medium for 24 h before harvested for analysis.
2-Ethylhexyl 2-Mercaptobenzoic Acid (7a): Yield: 91%. Rf 0.4 (n- The enzymic activity of MMP-1 in conditioned medium of UVB-irradiated
hexane/EtOAcꢃ7/1). 84—86 °C (CH2Cl2). 1H-NMR (CDCl3) d: 8.03 (1H, HDF-1 cells was assayed by casein-zymography according to the method of
q, Jꢃ1.5 Hz, ArH), 7.33 (1H, q, Jꢃ1.5 Hz, ArH), 7.29 (1H, d, Jꢃ0.9 Hz,
Bachmeier et al.21) The conditioned media were separated by electrophoresis
ArH), 7.17—7.10 (1H, m, ArH), 2.79 (2H, d, Jꢃ6.3 Hz, OCH2), 1.56—1.54 on an 8.5% SDS-polyacrylamide gel containing 0.33 mg/ml casein. The
(1H, m, CH), 1.52—1.29 (8H, m, CH2), 0.93—0.89 (6H, m, CH3). 13C-NMR electrophoresed gel was washed twice with washing buffer containing 2.5%
(CDCl3) d: 170.12, 142.53, 132.49, 132.01, 126.41, 126.14, 123.57, 37.82, triton X-100, followed by a brief rinsing in washing buffer without triton X-
36.55, 32.08, 28.15, 25.2, 22.28, 13.38, 10.03. IR: (KBr) cmꢇ1: 1676. UV
lmax (CH2Cl2) nm (log e): 358 (1.0). FAB-MS m/z: 267 (MꢈHꢈ). Anal.
Calcd for C15H22O2S: C, 67.63; H, 8.32. Found: C, 67.82; H, 8.42.
100. The gel was then immersed for 18 h in incubation buffer of 50 mM
Tris–HCl, pH 7.5, 150 mM NaCl, 10 mM CaCl2, 0.2% Brij 35 at 37 °C. After
incubation, the gel was stained with 0.25% Coomassie blue R-250. The
Ethyl 2-Mercaptobenzoic Acid (7b): Yield: 94%. Rf 0.6 (n- presence of MMP-1 was identified with a clear band of casein digestion.
hexane/EtOAcꢃ7/1). 1H-NMR (CDCl3) d: 7.97 (1H, d, Jꢃ7.8 Hz, ArH),
7.25 (1H, d, Jꢃ1.2 Hz, ArH), 7.10—7.09 (2H, m, ArH), 4.65 (1H, s, SH),
Acknowledgments The authors would like to thank the National De-
4.33 (2H, q, Jꢃ7.2 Hz, OCH2), 1.36 (3H, t, Jꢃ7.2 Hz, CH3). 13C-NMR fense University and National Defense Medical Center for supporting this
(CDCl3) d: 166.29, 137.55, 131.71, 131.14, 130.37, 125.99, 124.05, 60.57, research. This study was supported by grants from the National Science
13.56. IR: (KBr) cmꢇ1: 1668. UV lmax (CH2Cl2) nm (log e): 355 (1.0). EI-
Council (NSC 94-2320-B-016-041 and DOD-95-15 to T-C Chang and NSC
MS m/z: 182 (Mꢈ). Anal. Calcd for C9H10O2S: C, 59.32; H, 5.53. Found: C, 95-2314-B-157-001 to L-Y Hsu), Taiwan, ROC.
59.41; H, 5.62.
Cell Culture Primary human dermal fibroblast (HDF) cells were pur- References
chased from Cascade Biologics (Portland, OR, U.S.A.). HDF cells were
grown at 37 °C in a humidified incubator under 5% CO2/95% air in Medium
106 and supplemented with LSGS Kit (Cascade Biologics) and 10% fetal
bovine serum. Culture medium was replaced every other day. Cells between
passages 2 and 10 were used for studies.
Cell Viability Assay Cell viability was assessed by CCK-8 solution
(Dojindo Molecular Technologies, Inc., Kumamoto, Japan) by combining
WST [2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-
2H-tetrazolium] and 1-methoxy PMS [1-methoxy-phenazine methosulfate]
as described by the manufacturer. Briefly, cells were seeded in 96-well plates
at the density of 1ꢆ104 cells/well 24 h prior to treatment. Compounds were
added to the medium as a concentration stock and the control cells were fed
with medium containing the same amount of drug-free vehicle. After incu-
bation for 48 h, 10 ml of CCK-8 was added and cells were further incubated
for 3 h. Relative cell viability was obtained by measuring absorbance at
450 nm using an ELISA microplate reader (Spectra Max 190, Molecular De-
vices, Sunnyvale, CA, U.S.A.).
1) Kim J. H., Cho Y. H., Lee K. E., Lee J. J., Lee B. C., Pyo H. B., Song
K. S., Park H. D., Yun Y. P., Arch. Pharm. Res., 27, 177—183 (2004).
2) Bergeron R. J., Huang G., Weimar W. R., Smith R. E., Wiegand J., Mc-
Manis J. S., J. Med. Chem., 46, 16—24 (2003).
3) Brennan M., Bhtti H., Nerusu K. C., Bhagavathula N., Kang S., Fisher
G. J., Varani J., Voorhees J. J., Photochem. Photobiol., 78, 43—48
(2003).
4) Ropke C. D., Silva V. V., Kera C. Z., Miranda D. V., Almeida R. L.,
Sawada T. C. H., Barros S. B. M., Photochem. Photobiol., 82, 439—
442 (2006).
5) Lee Y. S., Jin D. Q., Beak S. M., Lee E. S., Kim J. A., Eur. J. Pharma-
col., 476, 173—178 (2003).
6) Fligiel S. E. G., Varani J., Datta S. C., Kang S., Fisher G., Voorhees J.
J., J. Invest. Dermatol., 120, 842—848 (2003).
7) Varani J., Warner R. L., Gharaee-Kermani M., Phan S. H., Kang S.,
Chung J., Wang Z. Q., Datta S. C., Fsher G. J., Voorhees J. J., J. Invest.
Dermatol., 120, 480—486 (2000).
ABTS Radical Scavenging Assay In vitro antioxidant activity of test
8) Nagase K., Biol. Chem., 378, 151—160 (1997).
9) Lowe N. J., Shaath N. A., Madhu A. P., “Sunscreens: Development,
Evaluation, and Regulatory Aspects,” 2nd ed., Marcel Dekker, Inc.,
New York, 1997, pp. 3—58.
compounds was evaluated using an improved ABTS ꢈ scavenging assay de-
·
ꢈ
scribed by Re et al.17) Briefly, the purple ABTS solution (7.5 mM) was
·
prepared by mixing ABTS with potassium persulfate (2.5 mM) in H2O
overnight in the dark at room temperature. The ABTSꢈ solution was diluted 10) Loo J. A., Holler T. P., Sanchez J., Gogliotti R., Maloney L., Reily M.
with ethanol or PBS (pH 7.4) to an A734 of 0.70ꢄ0.02 for assay. Six minutes
D., J. Med. Chem., 39, 4313—4320 (1996).
after the 5 ml of studied compounds was added to 30 ml of ABTS ꢈ dilution,
11) Domagala J. M., Gogliotti R., Sanchez J. P., Stier M. A., Musa K.,
Song Y., Loo J., Reily M., Tummino P., Harvey P., Hupe D., Sharmeen
L., Mack D., Scholten J., Saunders J., Mcquade T., Drug Design and
Discovery, 15, 49—61 (1997).
12) Fujisawa T., Odake S., Ogawa Y., Yasuda J., Morita Y., Morikawa T.,
Chem. Pharm. Bull., 50, 239—252 (2002).
·
ꢈ
·
the amount of ABTS remaining was determined at 734 nm, and the radi-
cal scavenging activity was obtained from the following equation:
radical scavenging activity (%)ꢃ[(AcontrolꢇAsample)/Acontrol]ꢆ100
Collagen Degradation Assay Rat tail collagen (BD Biosciences, San
Jose, CA, U.S.A.) was diluted to 238 mg/ml in Tris–glucose (0.5 M Glucose
and 0.33 M Tris). Purified MMP-1 (Calbiochem, San Diego, CA, U.S.A.) was
13) Jiang J. J., Chang T. C., Hsu W. L., Hwang J. M., Hsu L. Y., Chem.
Pharm. Bull., 51, 1307—1310 (2003).
diluted to 6ꢆ10ꢇ4 mg/ml with Tris–glucose. Degradation of the native colla- 14) Etzenhouser B., Hansch C., Kapur S., Selassie C. D., Bioorg. Med.
gen was carried out by mixing 19 ml of MMP-1 solution, 9 ml of collagen so-
lution and 2 ml of 1 mM synthetic compounds and incubated for 18 h at
37 °C. Intact collagen exposed to buffer alone served as control. Collagen
Chem., 9, 199—209 (2001).
15) Wang Y. C., Lin C. H., Chen C. M., Liou J. P., Tetrahedron Lett., 46,
8103—8104 (2005).
fragments released were resolved by SDS-PAGE (8.5% gel) and staining 16) Ferle-Vidovic A., Jukic I., Skare D., Suman L., Vukovic L., Cancer
with coomassie brilliant blue R-250. The relative intensity of the protein Biother. Radiopharm., 18, 781—790 (2003).
bands were measured by Image analyzing system (Alpha Innotech, St. San 17) Re R., Pellegrini N., Proteggente A., Pannala A., Yang M., Rice-Evans
Leandro, CA, U.S.A.). C., Free Radical Biol. Med., 26, 1231—1237 (1999).
UVB Irradiation and Zymography To characterize the effect of com- 18) Hung C. C., Tsai W. J., Kuo L. M. Y., Kuo Y. H., Bioorg. Med. Chem.,
pounds 5a—7b on the expression level of MMP-1 activity in human cells,
HDF cells were first treated with or without final concentrations of 1 mM
13, 1791—1797 (2005).
19) Miller N., Rice-Evans C., Free Rad. Res., 26, 195—199 (1997).
5a—7b for 24 h. For UV irradiation, the cells were irradiated with 20) Capizzi R. L., Semin. Oncol., 23, 2—17 (1996).
40 mJ/cm2 of UV-B (312 nm) for HDF cells, using the UV light irradiator 21) Bachmeier B. E., Nerlich A. G., Boukamp P., Lichtinghagen R.,
(UVItec unlimited, Cambridge, England). The cells were then washed and
Tschesche H., Fritz H., Fink E., Biol. Chem., 381, 509—516 (2000).