Synthesis of a series of caffeic acid phenethyl amide (CAPA) fluorinated derivatives: Comparison of cytoprotective effects to caffeic acid phenethyl ester (CAPE)

Article abstract of DOI:10.1016/j.bmc.2010.05.080

A series of catechol ring-fluorinated derivatives of caffeic acid phenethyl amide (CAPA) were synthesized and screened for cytoprotective activity against H₂O₂ induced oxidative stress in human umbilical vein endothelial cells (HUVEC

Full text of DOI:10.1016/j.bmc.2010.05.080

Bioorganic & Medicinal Chemistry 18 (2010) 5032–5038
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis of a series of caffeic acid phenethyl amide (CAPA) fluorinated
derivatives: Comparison of cytoprotective effects to caffeic acid phenethyl
ester (CAPE)
John Yang a, Gwendolyn A. Marriner c, Xinyu Wang a, Phillip D. Bowman b, Sean M. Kerwin c,
,
*
Salomon Stavchansky a,
*
a Division of Pharmaceutics, College of Pharmacy, The University of Texas, Austin, TX 78712, USA
b US Army Institute of Surgical Research, San Antonio, TX 78234, USA
c Division of Medicinal Chemistry and Institute for Cellular and Molecular Biology, The University of Texas, Austin, TX 78712, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of catechol ring-fluorinated derivatives of caffeic acid phenethyl amide (CAPA) were synthesized
and screened for cytoprotective activity against H2O2 induced oxidative stress in human umbilical vein
endothelial cells (HUVEC). CAPA and three fluorinated analogs were found to be significantly cytoprotec-
tive when compared to control, with no significant difference in cytoprotection between caffeic acid
phenethyl ester (CAPE) and CAPA.
Received 22 March 2010
Revised 27 May 2010
Accepted 31 May 2010
Available online 11 June 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Oxidative stress
Hydrogen peroxide
Ischemia/reperfusion injury
CAPE
1. Introduction
Cytoprotection to oxidative stress by CAPE and analogs has been
correlated with their ability to up-regulate heme oxygenase 1
Oxidative stress plays a significant role in the development of a
variety of disease states such as inflammation, ischemia reperfu-
sion injury (I/R), and cardiovascular complications.1–4 Efforts to
ameliorate oxidative stress have led to the discovery of a number
of natural and novel compounds that have antioxidative properties
or are able to induce genes with the downstream effect of counter-
acting oxidative stress.5,6
(HMOX 1) gene expression.15,16 Despite demonstrating significant
cytoprotection against oxidative damage both in vivo and in vitro,
CAPE is known to be readily hydrolyzed in plasma.17,18 Studies
have suggested that esterase activity in blood and cells is respon-
sible for the rapid degradation of CAPE.18 Pharmacokinetic studies
using a rat model have also shown that CAPE is cleared rapidly
after intravenous administration to rats.19
Caffeic acid phenethyl ester (CAPE), a plant polyphenolic con-
centrated in honeybee propolis, has been found to be biologically
active in a variety of pathways including cytoprotection against
oxidative stress. CAPE has been observed to exhibit anti-inflamma-
tory,7 anti-viral,8 anti-carcinogenic,9 and immunomodulatory10 ef-
fects as well as protection against I/R injury in vivo.11,12 CAPE has
also been investigated for its antioxidative and radical scavenging
properties.13 Previous studies have reported the synthesis and
investigation of catechol ring-fluorinated derivatives of CAPE with
regard to cytoprotective ability against oxidative stress in vitro.14
The purpose of this study was to synthesize and investigate the
cytoprotective activity of a series of amide derivatives of CAPE and
previously reported CAPE analogs. Amides are generally associated
with higher hydrolytic energies of activation compared to esters.
Previous work on CAPA has described its ability to act as an anti-
oxidant against lipid peroxidation20 as well as a potential anti-
inflammatory agent through its inhibition of 5-lipoxygenase.21
CAPA has also been shown to exhibit significant radical scavenging
activity using a 2,2-diphenyl-1-picrylhydrazyl assay.22 Although
various CAPA analogues have been investigated for both radical
scavenging activity as well as
a
-glucosidase inhibition,23 no cate-
chol ring-fluorinated CAPA analogs have been studied. The cyto-
protectant ability of CAPA in vitro has also not been previously
addressed. Due to the importance of endothelial cells as a target
of oxidative stress in I/R injury and other vascular complications,
human umbilical vein endothelial cells (HUVEC) were chosen as
* Corresponding authors. Tel.: +1 512 471 1407; fax: +1 512 471 7474 (S.S.), tel.:
+1 512 471 5074; fax: +1 512 232 2606 (S.M.K.).
E-mail addresses: skerwin@mail.utexas.edu (S.M. Kerwin), stavchansky@mai-
l.utexas.edu (S. Stavchansky).
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.05.080

J. Yang et al. / Bioorg. Med. Chem. 18 (2010) 5032–5038
5033
Table 1
the model in which to study the cytoprotective effects of these
Synthesis of CAPA and fluorinated CAPA analogues
compounds against hydrogen peroxide induced oxidative stress.
1. TBDMSCl, imidazole, DMAP (cat)
DMF, rt, 1 h
2. Results
2. 2, Cs2CO3
dioxane/CHCl3, 60 °C, 18 h
O
2.1. Synthesis
O
N
H
3. TBAF
THF, 0 °C, 5 min
R5
R5
H
CAPA and five additional fluorinated amide analogues of CAPE
were prepared using a Wittig coupling approach. The known chlo-
roacetamide 1,24,25 was reacted with triphenyl phosphine to give
the phosphonium chloride 2 (Scheme 1). Wittig coupling of 2 with
unprotected hydroxybenzaldehydes 3a–e proved problematic, in
contrast to previous studies employing the analogous ester phos-
phonium chloride.14 Thus, the hydroxybenzaldehydes 3a–e, which
were either commercially available or obtained via demethylation
of the corresponding methoxybenzaldehydes with boron tribro-
mide, were first transiently protected as the t-butyldimethylsilyl
ethers by treatment with t-butyldimethylsilyl chloride and imidaz-
R4O
R2
R4O
R2
R3
R3
3a-f
4a-f
Compound
R2
R3
R4
R5
Yielda (%)
a (CAPA)
H
F
H
H
H
F
H
H
OH
OMe
H
H
H
H
H
H
Me
OH
OH
F
F
F
14
7
b
c
d
e
f
15
22
8b
H
OMe
63c
ole prior to Wittig coupling (Table 1). The resulting
a,b-unsatu-
a
Isolated overall yield from benzaldehyde 3 after column chromatography and
rated amides were subjected to deprotection with TBAF, to afford
CAPA and the desired amides 4b–e in modest overall yields
(Table 1). For aldehydes that do not require transient protection
of the catechol functionality, the Wittig coupling proceeds in high-
er yield. The dimethoxybenzaldehyde 3f was used directly in the
Wittig coupling to afford 4f in reasonable yield. With the exception
of amide 4e which was isolated as a ꢀ3:1 mixture of (E)-/(Z)-iso-
mers after column chromatography, the amides 4 were obtained
as >90% pure (E)-isomers after column chromatography and, for
4a–c and 4f, recrystallization.
recrystallization.
b
Isolated as ꢀ3:1 mixture of (E)-/(Z)-isomers.
c
Step 2 only.
cell viability to approximately 20% of control, which was provided
by 2 mM H2O2. The results are shown in Figure 2.
2.4. Cytoprotection against H2O2 induced oxidative stress in
HUVEC
2.2. Cytotoxicity of amide derivatives compared to CAPE in
HUVEC
To evaluate oxidative stress in vitro, we employed a model
using H2O2 as the inducer of oxidative damage. HUVEC were trea-
ted with CAPE or the amide derivatives 4a–f at 20 lM concentra-
CAPE and certain catechol ring-fluorinated CAPE analogs have
been reported to be cytotoxic to HUVEC at higher concentrations.14
CAPA and the CAPA derivatives were screened along with CAPE for
toxicity in HUVEC. Each of the compounds were incubated with
tion for 5 h. The cells were rinsed and then treated with 2 mM
H2O2. After 1 h, the H2O2 containing medium was replaced with
cell culture media and the cells were allowed to recover for 18 h.
At the end of the 18 h period, cell viability was assessed with the
CellTiter-BlueÒ Cell Viability assay and compared to cells treated
only with vehicle and H2O2, as well as with those that were not ex-
posed to H2O2. The results are shown in Figure 3. CAPA and com-
pounds 4b, 4c, and 4e exhibited significant cytoprotection
against H2O2 when compared to vehicle only pre-treatment. CAPE
was also significantly cytoprotective against H2O2. There was no
significant difference in cytoprotective activity between CAPE
and CAPA (P >0.05).
HUVEC at 10, 20, 40, and 60 lM concentrations for 24 h at 37 °C.
Cell viability was measured using the CellTiter-BlueÒ assay and
compared to a vehicle control. Cell viability less than 90% of control
was considered toxic. The results, shown in Figure 1 demonstrate
that CAPA and amides 4b and 4d–f showed no toxicity at any of
the tested concentrations. CAPE exhibited cytotoxicity at 40 and
60 lM. The amide 4c showed cytotoxicity at all concentrations.
2.3. Cytotoxicity of H2O2 in HUVEC
In a dose dependent cytoprotection assay, HUVEC were treated
with CAPE and CAPA at 1, 5, 20, 40, and 60 lM concentrations prior
Hydrogen peroxide is one of the principle reactive oxygen spe-
cies produced in various vascular complications including ischemia
reperfusion injury,26 and has also been used in other in vitro mod-
els as an inducer of oxidative stress in endothelial cells.27,28 To
determine a suitable dose, HUVEC were treated with H2O2 at con-
centrations ranging from 0.01 to 5 mM for 1 h. Following the 1 h
period the culture media was replaced and the cells were allowed
to recover for 18 h. Cell viability was assessed with CellTiter-BlueÒ
following the 18 h period. The target dosage was one that reduced
to the induction of oxidative stress with H2O2. The results are
shown in Figure 4. The EC50 was calculated for both compounds
by linear regression using the first three data points. The EC50
was found to be 8 lM for CAPE and 2 lM for CAPA.
3. Discussion
Introducing a fluorine group on the catechol ring increases the
electronic density of the conjugated system, can decrease the inter-
action with catechol methyltransferase, and may also have a signif-
icant effect on receptor binding or selectivity.29 The hydroxyl
groups on the CAPA catechol may contribute to the antioxidative
activity of the compound. We were interested in seeing the effect
of replacing one of these hydroxyls with a fluorine, hydrogen or
methoxy group on the cytoprotective activity of the compound.
Prior to evaluating the cytoprotective activity of CAPE and the
CAPA derivatives, each compound was screened for toxicity in HU-
PPh3
THF, 85 °C, 72 h
Cl
H
N
H
N
Cl
Ph3P
O
O
82 %
1
2
Scheme 1.
VEC. CAPE was found to be toxic at 40 and 60 lM, in accord with

5034
J. Yang et al. / Bioorg. Med. Chem. 18 (2010) 5032–5038
Figure 1. Toxicity of CAPE, CAPA, and CAPA derivatives toward HUVEC. Compounds were incubated in HUVEC for 24 h at 37 °C. Cell viability was determined by the Alamar
Blue assay. Values are reported as a percentage of the vehicle control (0.1% DMSO).
Figure 2. Toxicity of H2O2 in HUVEC. HUVEC were incubated in culture media containing the indicated concentration of H2O2 for 1 h at 37 °C. The culture media was replaced
and cells were allowed 18 h to recover, then were assessed for viability with the CellTiter-BlueÒ
.
sponding ester analogues. Amides 4e and 4f were not cytotoxic
at any concentration up to 60 M, the highest concentration exam-
ined. The corresponding ester derivatives were similarly reported
to be non-cytotoxic at concentrations up to 15 g/mL (ca.
l
l
50 lM). Amide 4c was toxic at all the concentrations tested; simi-
lar to the corresponding ester analogue.14 However, whereas CAPE
and the esters corresponding to 4b and 4d are cytotoxic at concen-
trations greater than 40
lM, CAPA and the amides 4b and 4d are
not cytotoxic even at concentrations as high as 60
l
M. The origin
of this difference in cytotoxicity between CAPE and certain fluori-
nated CAPE analogues versus CAPA and the corresponding fluori-
nated CAPA analogues is not clear.
CAPE was significantly cytoprotective against H2O2 induced oxi-
dative stress in HUVEC. This was also demonstrated previously in a
similar model,30 as well as a study which used menadione to gener-
ate an oxidative stress.14 Four of the amide derivatives of CAPE were
also found to be significantly cytoprotective. These four compounds
all contained either one or two hydroxyl groups on the cinnamic acid
phenyl ring. Although compound 4c proved to be very cytotoxic in
HUVEC over a 24 h period, the toxicity is less apparent over a 5 h
incubation time, as the compound was found to be significantly
Figure 3. Cytoprotection of HUVEC against 2 mM H2O2 by CAPE, CAPA, and CAPA
analogues. All compound concentrations were at 20 M. CAPE, CAPA, 4B, 4C and 4E
all showed significant cytoprotection when compared to untreated (H2O2 only)
l
(P <0.05). CAPA derivatives 4D and 4F provided no cytoprotection.
previous studies.14 There are interesting differences in cytotoxicity
of the certain amide derivatives when compared to their corre-

J. Yang et al. / Bioorg. Med. Chem. 18 (2010) 5032–5038
5035
Figure 4. Dose response cytoprotection relationship of CAPE and CAPA against 2 mM H2O2. Concentrations above 40
l
M are cytotoxic for both CAPE and CAPA and gave lower
cell viability than untreated HUVEC as shown. CAPE and CAPA showed significant cytoprotection at concentrations from 1 through 40 lM.
cytoprotective against H2O2, and exhibited significantly higher cell
viability over the vehicle control. While the mechanism behind this
cytoprotective activity is not completely known, it is suggested that
the antioxidative and radical scavenging properties of the catechol
group are correlated with the protection against H2O2. The catechols
CAPE, CAPA, 4b, and 4c all display cytoprotective effects; whereas,
the monomethylated and dimethylated analogues 4d and 4f, respec-
tively, were not cytoprotective. The ortho-fluorophenol 4e demon-
strated intermediate cytoprotection, which may be due to the
ability of the ortho-fluorine substituent to stabilize the phenol radi-
cal formed upon hydrogen atom donation.31 Interestingly, the cyto-
protective activity for these amides is quite different from that
reported for the corresponding esters.14 The ester corresponding
to 4c is not cytoprotective, despite the presence of the catechol func-
tionality, and the ester corresponding to 4e is cytoprotective, despite
the lack of any free phenolic hydroxyl groups. In the dose dependent
cytoprotection assay, a biphasic response was observed for both
CAPE and CAPA. The cytoprotection percentage increases from 1
The findings demonstrate that CAPA is less toxic than CAPE, and
that there is no significant difference in cytoprotection between
the two when tested at 5 and 20 lM concentrations against
2 mM hydrogen peroxide. It is anticipated that CAPA is more stable
than CAPE in plasma. Previous studies have shown that ester com-
pounds are very prone to hydrolysis via esterases, and that CAPE
exhibits a short half life in plasma.16 Plasma stability is an impor-
tant issue in the drug development process, as it determines how
much of the initial dose actually reaches the target site. Cinnamic
acid-derived amides are known to be particularly stable to hydro-
lysis.33 In preliminary studies, the expected increased stability of
CAPA has been confirmed. CAPA is surprising stable to acid
(0.1 M HCl, pH ꢀ1), although it undergoes hydrolysis readily at
pH ꢀ10. While incubation of CAPE (100
lM) in rat plasma for
18 h at 37 °C results in complete hydrolysis, CAPA remains largely
intact under these conditions (see Supplementary data).
4. Conclusions
up to 20
The drop off in cytoprotection at higher CAPE concentration had
been attributed to CAPE’s cytotoxicity above 20
M,13 as indicated
in Figure 1. However, a similar effect is observed for CAPA, even
lM of CAPE and CAPA, then starts to decline at 40 lM.
CAPA and catechol ring-fluorinated derivatives of CAPA were
synthesized and investigated for cytoprotective activity against
hydrogen peroxide induced oxidative stress in HUVEC. All but
one of the CAPA derivatives synthesized were non-toxic up to
the maximally tested concentration, with 4c being toxic at all
tested concentrations. The results here also show that CAPA, 4b,
4c, and 4e are all significantly cytoprotective in this model. The
only two analogs which were not cytoprotective were the methyl-
ated compounds 4d and 4f. Although the mechanism of cytopro-
tection for these amides is not well understood, cytoprotection is
correlated with the presence of free catechol hydroxyl groups in
the analogs examined. CAPA was less toxic in HUVEC when com-
pared to CAPE, however, there was no significant difference found
in cytoprotection between the two compounds. This is significant
as CAPA retains CAPE’s cytoprotective activity yet is more stable
in plasma.
l
though it is not cytotoxic at 40
nomenon occurs.
lM. It is unclear as to why this phe-
It is not well understood how cytoprotection is provided by pre-
treatment with these cytoprotective agents. Cytoprotection proba-
bly involves an interplay between direct anti-oxidant activity and
indirect anti-oxidant activity through effects on the transcription
of anti-oxidant genes.32 Studies previously performed in our group
have shown that the cytoprotective activity of CAPE was correlated
with the levels of the heme oxygenase 1 (HMOX 1) gene expres-
sion. CAPE is a potent inducer of the HMOX 1 gene transcription
and has been shown to up-regulate it as much as eightfold over
control.15 Studies have also shown that when heme oxygenase
activity is inhibited, the cytoprotective effect of CAPE against men-
adione induced oxidative stress is abolished.15 Within a series of
catechol ring-fluorinated CAPE analogs, cytoprotection is corre-
lated with up-regulation of HMOX 1, but not direct anti-oxidant
activity.16 However, for CAPA and the CAPA analogs examined
here, it is not clear how the balance of direct and indirect anti-oxi-
dant effects contribute to cytoprotection. Work is currently being
done to evaluate the role of HMOX 1 and to determine if CAPA
exhibits similar response in regards to activation of HMOX 1.
5. Experimental
5.1. Materials and apparatus
The reagents chloroacetyl chloride, phenethylamine, chloro-
tert-butyldimethylsilane (TBDMSCl), 3,4-dihydroxybenzaldehyde,
2-fluoro-4,5-dimethoxybenzaldehyde, 3-fluoro-4-methoxybenzal-

5036
J. Yang et al. / Bioorg. Med. Chem. 18 (2010) 5032–5038
dehyde, 3-fluoro-4-hydroxy-5-methoxybenzaldehyde, tetra-butyl-
ammonium fluoride (TBAF), hydrogen peroxide, and boron tribro-
mide were purchased from Sigma Aldrich (St Louis, MO) and
used without further purification. All solvents were distilled prior
to use. Nuclear magnetic resonance (NMR) spectroscopy was per-
formed with a Varian Unity + 300 (300 MHz). Melting points were
obtained using a Buchi B-540 apparatus and are uncorrected. Mass
spectrometry services were provided by the Mass Spectrometry
Facility at the University of Texas at Austin. Carbon, hydrogen,
and nitrogen (CHN) elemental analysis was conducted by Quanti-
tative Technologies Inc. (Whitehouse, NJ). HPLC was performed
on a Varian Prostar 320 system. Purity of the final compounds
was assessed by both normal (Si) and reverse phase (C8) HPLC.
raphy (2% EtOAc in hexane) of the residue after evaporation of the
solvent afforded 540 mg of the protected benzaldehyde, which was
combined with the phosphonium chloride (2) (1.8 mmol, 828 mg)
and Cs2CO3 (3.9 mmol, 1651 mg) and then 5 mL of dioxane and
5 mL of CHCl3. The resulting mixture was heated to 60 °C for
18 h. The reaction solution was separated, and the solid washed
with CHCl3. The combined organics were washed with water, dried
over Na2SO4, and evaporated. Column chromatography (3:1 hex-
ane/EtOAc) of the residue gave 550 mg of yellow oil. The oil was
dissolved in 5 mL of THF and TBAF (2.5 mL, 1 M in THF) was then
added and the mixture was stirred for 5 min at 0 °C. The reaction
mixture was concentrated on a rotary evaporator and subjected
to chromatography on a silica gel column (4:3 EtOAc/hexane).
Recrystallization (CH2Cl2 and hexane) afforded 115 mg of 4a (14%
overall yield from 3,4-dihydroxybenzaldehyde) as a white solid:
mp 145 °C (lit,19 138–140 °C); 1H NMR matches literature; 13C
NMR (CDCl3): d 35.48, 41.09, 113.89, 115.30, 117.17, 120.95,
126.2, 127.13, 128.35, 128.65, 139.39, 141.07, 145.56, 147.60,
168.14. CI-MS m/z 284 (MH+, 100). HRCI-MS: Calcd for
5.2. Preparation of Wittig reagent
5.2.1. 2-Chloro-N-phenethylacetamide (1)24,25
To a solution of phenethylamine (20 mmol, 2.52 mL) in CH2Cl2
(40 mL) was added K2CO3 (24 mmol, 3.32 g). Chloroacetylchloride
(22 mmol, 1.75 mL) was slowly added to the reaction mixture.
The reaction mixture was stirred at 45 °C under argon for 18 h.
The mixture was diluted with CH2Cl2, washed with water and
brine, and then dried over Na2SO4. The resulting solution was con-
centrated under a rotary evaporator and the resulting solid filtered
to give 2.93 g of white crystals (74% yield); mp 63.6–64.6 °C (lit,25
65 °C); 1H NMR spectrum matches literature.25
C17H18NO3; 284.1287. Found: 284.1288.
The following compounds were prepared following the same
procedure:
5.4.2. 3-(2-Fluoro-4,5-dihydroxyphenyl)-N-phenethylacrylamide
(4b)
Recrystallization (CH2Cl2 and hexane) afforded 80 mg of 4b (7%
overall yield from 2-fluoro-4,5-dihydroxybenzaldehyde) as a white
solid: mp 145 °C; 1H NMR (DMSO-d6) d (ppm): 2.77 (t, J = 6.9 Hz,
2H), 3.39 (q, J = 6.6 Hz, 2H), 6.40 (d, J = 15.9 Hz, 1H), 6.60 (d,
J = 12.3 Hz, 1H), 6.91 (d, J = 7.8 Hz, 1H), 7.28 (m, 5H), 8.18 (s, 1H),
9.17 (s, 1H), 9.89 (s, 1H); 13C NMR (DMSO-d6) d: 35.44, 41.11,
102.81 (JC–F = 26 Hz), 113.07 (JC–F = 5 Hz), 113.29, 119.27
(JC–F = 6 Hz), 126.20, 128.35, 128.64, 133.36, 139.37, 142.09,
148.67 (JC–F = 11.55 Hz), 155.77 (JC–F = 243 Hz), 167.88; CI-MS m/z
302 (MH+, 100). HRCI-MS: Calcd for C17H17NO3F; 302.1192. Found:
302.1194.
5.2.2. Phenethylcarbamoylmethyl-triphenylphosphonium
chloride (2)
To a solution of triphenylphosphine (18.9 mmol, 4.96 g) in THF
(50 mL) was added 2-chloro-N-phenethylacetamide (1) (12.6 mmol,
2.5 g). The mixture was stirred at 85 °C for 72 h under argon. The
reaction mixture was diluted with diethyl ether and filtered, giving
4.75 g of white solid (82% yield); mp 220.9–222.8 °C; 1H NMR
(CDCl3) d: 2.67 (2H, t, J = 8.1 Hz), 3.31 (2H, q), 5.05 (2H, d,
J = 14.4 Hz), 7.15–7.25 (5H, m), 7.59–7.68 (5H, m), 7.72–7.88 (10H,
m). 13C NMR (CDCl3) d: 32.30 (JC–P = 54.94 Hz), 35.35, 41.73,
118.63 (JC–P = 88.45 Hz), 126.33, 128.73 (JC–P = 30.79 Hz), 130.31
(JC–P = 12.68 Hz), 134.24 (JC–P = 10.49 Hz), 135.16 (JC–P = 2.72 Hz),
139.17, 162.661 (JC–P = 4.98 Hz); CI-MS m/z 424 (MÀClÀ, 100).
HRCI-MS: Calcd for C28H27NOP: 424.1830. Found: 424.1825.
5.4.3. 3-(3-Fluoro-4,5-dihydroxyphenyl)-N-phenethylacrylamide
(4c)
Recrystallization (CH2Cl2 and hexane) afforded 135 mg of 4c
(15% overall yield from 3-fluoro-4,5-dihydroxybenzaldehyde) as a
white solid: mp 154 °C; 1H NMR (DMSO-d6) d (ppm): 2.76 (d,
J = 7.2 Hz, 2H), 3.39 (d, J = 6.6 Hz, 2H), 6.36 (d, J = 15.6 Hz, 1H),
6.81 (d, J = 6 Hz, 1H), 6.86 (s, 1H), 7.23 (m, 5H), 8.09 (t, J = 5.1 Hz,
1H), 9.46 (s, 1H), 9.71 (s, 1H); 13C NMR (DMSO-d6) d: 35.43, 41.1,
106.41 (JC–F = 20 Hz), 110.59, 118.79, 126.10, 126.22, 128.36,
5.3. General procedure for the demethylation of benzaldehydes
5.3.1. 2-Fluoro-4,5-dihydroxybenzaldehyde
2-Fluoro-4,5-dimethoxybenzaldehyde (4 mmol, 736 mg) was
dissolved in 10 mL of CH2Cl2. The mixture was placed in a À78 °C
acetone and dry ice bath and 10 mL of a 1 M solution of BBr3 in
CH2Cl2 was added slowly under argon. The reaction mixture was
allowed to warm to room temperature and stirred for 18 h.
Methanol was added to the resulting mixture, and the solvent
evaporated. This process was repeated three times. Column chro-
matography (5:1 CH2Cl2/EtOAc) afforded 590 mg (94.5% yield) of
2-fluoro-4,5-dihydroxybenzaldehyde as a white solid which was
carried forward without further purification.
128.64, 135.23, 139.34, 140.06, 147.57 (JC–F = 6 Hz), 152.28 (JC–F
238 Hz), 167.67; CI-MS m/z 302 (MH+, 100). HRCI-MS: Calcd for
17H17NO3F: 302.1192. Found: 302.1188.
=
C
5.4.4. 3-(3-Fluoro-4-hydroxy-5-methoxyphenyl)-N-phenethyl
acrylamide (4d)
An additional column chromatography purification (1:1.5
EtOAc/hexane) to remove traces of the Z-isomer, afforded 100 mg
of 4d (22% overall yield from 3-fluoro-4-hydroxy-5-methoxybenz-
aldehyde) as a white foam: 1H NMR (CDCl3) d (ppm): 2.91 (t,
J = 6.9 Hz, 2H), 3.68 (q, J = 6.2 Hz, 2H), 5.68 (s, 1H), 5.82 (s, 1H),
6.20 (d, J = 15.4 Hz, 1H), 6.79 (s, 1H), 6.65 (dd, J = 1.6 Hz,
J = 10.8 Hz, 1H), 7.28 (m, 5H), 7.50 (d, J = 15.4 Hz, 1H); 13C NMR
(CDCl3) d: 31.92, 37.13, 52.74, 102.72, 104.76 (JC–F = 19 Hz),
115.81, 122.65 (JC–F = 9 Hz), 122.87, 125.00, 125.09, 131.72
(JC–F = 14 Hz), 135.13, 136.53, 144.69 (JC–F = 6 Hz), 147.02
(JC–F = 242 Hz), 162.22; CI-MS m/z 316 (MH+, 100). HRCI-MS: Calcd
for C18H19NO3F; 316.1349. Found: 316.1351.
5.4. General procedure for the Wittig reaction
5.4.1. 3-(3,4-Dihydroxyphenyl)-N-phenethylacrylamide (4a,
CAPA)
A mixture of 3,4-dihydroxybenzaldehyde (3 mmol, 414 mg),
imidazole (9 mmol, 612 mg), TBDMSCl (9 mmol, 1356 mg), and
DMAP (0.3 mmol, 36 mg) were dissolved in 5 mL of DMF and
allowed to react at room temperature under argon for 1 h. The
reaction mixture was extracted with diethyl ether, washed with
deionized water, and then dried over Na2SO4. Column chromatog-

J. Yang et al. / Bioorg. Med. Chem. 18 (2010) 5032–5038
5037
5.4.5. 3-(3-Fluoro-4-hydroxyphenyl)-N-phenethyl-acrylamide
(4e)
peroxide for 1 h at 37 °C. The hydrogen peroxide was then removed.
The cells were washed once with MCDB 131 media, and were then
incubated in complete MCDB 131 media for 18 h at 37 °C. Following
the 18 h period, the cells were treated with 10% CellTiter-BlueÒ solu-
tion and analyzed for viability. In the dose–response cytoprotection
assay, percent cytoprotection for each compound was calculated by
subtracting the average fluorescent reading of the negative control
(HUVEC treated only with DMSO and hydrogen peroxide) from the
fluorescent values of each well. This was then divided by the average
fluorescence of the positive control (HUVEC treated only with
DMSO) to obtain percent cytoprotection.
This process afforded 90 mg of 4e (8% overall yield from 3-flu-
oro-4-hydroxybenzaldehyde) as a yellow foam with a 3:1 (E)-/
(Z)-isomer ratio by 1H NMR: 1H NMR (CDCl3) d (ppm) major iso-
mer: 2.91 (t, J = 6.9 Hz, 2H), 3.68 (q, J = 6.1 Hz, 2H), 5.66 (s, 1H),
6.18 (d, J = 15.6 Hz, 1H), 7.00 (t, J = 8.5 Hz, 1H), 7.12 (d, J = 8.5 Hz,
1H), 7.18 (d, J = 1.8 Hz, 1H), 7.28 (m, 5H), 7.52 (d, J = 15.6 Hz,
1H); 13C NMR (CDCl3) d: 35.80, 41.30, 114.85 (JC–F = 19 Hz),
118.31, 118.66, 125.50, 126.95 (JC–F = 20 Hz), 127.18, 128.92,
128.99, 138.89, 140.85, 146.99 (JC–F = 14 Hz), 151.86 (JC–F
241 Hz), 167.13; CI-MS m/z 286 (MH+, 100). HRCI-MS: Calcd for
17H17NO2F; 286.1243. Found: 286.1242.
=
C
5.8. Statistical analysis
5.4.6. 3-(2-Fluoro-4,5-dimethoxyphenyl)-N-phenethyl-acrylamide
(4f)
Data are reported as means standard deviation as a percent-
age of the control. Differences between the groups were first ana-
lyzed by ANOVA, and then evaluated by the Tukey-Kramer post
hoc analysis. O’Brien’s and Bartlett’s tests showed that variances
were equal among groups. P <0.05 was considered significant. All
statistical analysis was performed using the JMP program (SAS).
Recrystallization from EtOAc and hexane gave 48 mg of 4f (63%
yield from 2-fluoro-4,5-dimethoxybenzaldehyde) as white crys-
tals: mp 149 °C; 1H NMR (CDCl3) d (ppm): 2.92 (t, J = 6.9 Hz, 2H),
3.68 (q, J = 6.5, 2H), 3.88 (d, J = 8.7, 6H), 5.90 (br s, 1H), 6.38 (d,
J = 15.6 Hz, 1H), 6.64 (d, J = 12.0, 1H), 6.90 (d, J = 7.20 Hz, 1H),
7.29 (m, 5H), 7.66 (d, J = 15.6 Hz, 1H); 13C NMR (CDCl3): d 35.93,
Acknowledgments
41.09, 56.53 (JC–F = 9.96 Hz), 100.45 (JC–F = 28 Hz), 110.54 (JC–F
=
4 Hz), 114.01 (JC–F = 13 Hz), 121.15 (JC–F = 7 Hz), 126.77, 128.99
(JC–F = 11 Hz), 134.03, 139.15, 145.67, 151.41 (JC–F = 10 Hz),
156.54 (JC–F = 248 Hz), 166.40; CI-MS m/z 330 (MH+, 100). HRCI-
MS: Calcd for C19H21NO3F; 330.1505. Found: 330.1506; elemental
Anal. Calcd for C19H20NO3F: C, 69.29; H, 6.12; N, 4.25. Found: C,
69.03; H, 6.12; N, 4.21.
This project was supported by the US Army Institute of Surgical
Research, the Robert Welch Foundation (F-1298 and H-F-0032) and
the TI-3D.
Supplementary data
5.5. Cell culture
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2010.05.080.
Human umbilical vein endothelial cells (HUVEC) were obtained
from Lifeline Technologies (Walkersville, MD) and cultivated on
75 cm2 1% gelatin coated culture flasks using MCDB 131 cell cul-
ture media (Invitrogen, Carlsbad CA) supplemented with 2% fetal
bovine serum, ascorbic acid, heparin, VEGF, hydrocortisone bFGF
and heparin (Lifeline Technologies). The cells were grown to con-
fluency at 37 °C in humidified atmosphere with 5% CO2. HUVEC
were then treated with Trypsin/EDTA and subcultivated onto gela-
tin coated 96 well multi-plates and used when confluent. Popula-
tion doubling levels 2 through 5 were used in the described
experiments.
References and notes
1. Kondo, T.; Hirose, M.; Kageyama, K. J. Atheroscler. Thromb. 2009, 16, 532.
2. Kobayashi, S.; Inoue, N.; Azumi, H.; Seno, T.; Hirata, K.; Kawashima, S.;
Hayashi, Y.; Itoh, H.; Yokozaki, H.; Yokoyama, M. J. Atheroscler. Thromb.
2002, 9, 184.
3. Robin, E.; Guzy, R. D.; Loor, G.; Iwase, H.; Waypa, G. B.; Marks, J. D.; Hoek, T. L.;
Schumacker, P. T. J. Biol. Chem. 2007, 282, 19133.
4. Jaeschke, H.; Farhood, A., et al Am. J. Physiol. 1991, 260, G355.
5. Yang, L.; Gong, J.; Wang, F.; Zhang, Y.; Wang, Y.; Hao, X.; Wu, X.; Bai, H.;
Stockigt, J.; Zhao, Y. J. Enzyme Inhib. Med. Chem. 2006, 21, 399.
6. Bandgar, B. P.; Gawande, S. S.; Bodade, R. G.; Gawande, N. M.; Khobragade, C. N.
Bioorg. Med. Chem. 2009, 17, 8168.
7. Fitzpatrick, L. R.; Wang, J.; Le, T. J. Pharmacol. Exp. Ther. 2001, 299, 915.
8. Fesen, M. R.; Pommier, Y.; Leteurtre, F.; Hiroguchi, S.; Yung, J.; Kohn, K. W.
Biochem. Pharmacol. 1994, 48, 595.
5.6. Cytotoxicity assay
9. Lee, Y. J.; Liao, P. H.; Chen, W. K.; Yang, C. Y. Cancer Lett. 2000, 153, 51.
10. Nam, J. H.; Shin, D. H.; Zheng, H.; Kang, J. S.; Kim, W. K.; Kim, S. J. Eur. J.
Pharmacol. 2009, 612, 153.
11. Cagli, K.; Bagci, C.; Gulec, M.; Cengiz, B.; Akyol, O.; Sari, I.; Cavdar, S.; Pence, S.;
Dinckan, H. Ann. Clin. Lab. Sci. 2005, 35, 440.
12. Kart, A.; Cigremis, Y.; Ozen, H.; Dogan, O. Food Chem. Toxicol. 2009, 47, 1980.
13. Russo, A.; Longo, R.; Vanella, A. Fitoterapia 2002, 73, S21.
14. Wang, X.; Stavchansky, S.; Bowman, P. D.; Kerwin, S. M. Bioorg. Med. Chem.
2006, 14, 4879.
15. Wang, X.; Stavchansky, S.; Zhao, B.; Bynum, J. A.; Kerwin, S. M.; Bowman, P. D.
Eur. J. Pharmacol. 2008, 591, 28.
Stock CAPE and CAPE amide derivative solutions were dissolved
in DMSO then diluted in MCDB 131 tissue culture media for use in
the assays. Confluent HUVEC were treated with CAPE and the
amide derivatives for 24 h at 37 °C at concentrations ranging from
10 to 60 lM. Following the 24 h incubation, the media was re-
placed with 10% CellTiter-BlueÒ Blue solution (Promega, Madison
WI). HUVEC were incubated for 2 h at 37 °C then analyzed for fluo-
rescence. The readings were taken at 545 nm excitation and
590 nm emission wavelengths on a Spectramax M2 microplate
reader (Molecular Devices, Sunnyvale CA). Cell viability was calcu-
lated from these fluorescence readings.
16. Wang, X.; Stavchansky, S.; Zhao, B.; Bynum, J. A.; Bowman, P. D.; Kerwin, S. M.
Eur. J. Pharmacol. 2010, 635, 16.
17. Celli, N.; Dragani, L. K.; Murzilli, S.; Pagliani, T.; Poggi, A. J. Agric. Food Chem.
2007, 55, 3398.
18. Wang, X.; Bowman, P. D.; Kerwin, S. M.; Stavchansky, S. Biomed. Chromatogr.
2007, 21, 343.
5.7. Cytoprotection assay
19. Wang, X.; Pang, J.; Maffucci, J. A.; Pade, D. S.; Newman, R. A.; Kerwin, S. M.;
Bowman, P. D.; Stavchansky, S. Biopharm. Drug Dispos. 2009, 30, 221.
20. Rajan, P.; Vedernikova, I.; Cos, P.; Berghe, D. V.; Augustyns, K.; Haemers, A.
Bioorg. Med. Chem. Lett. 2001, 11, 215.
21. Naito, Y.; Sugiura, M.; Yamaura, Y.; Fukaya, C.; Yokoyama, K.; Nakagawa, Y.;
Ikeda, T.; Senda, M.; Fujita, T. Chem. Pharm. Bull. 1991, 39, 1736.
22. Son, S.; Lewis, B. J. Agric. Food Chem. 2002, 50, 468.
23. Nishioka, T.; Watanabe, J.; Kawabata, J.; Niki, R. Biosci. Biotech. Biochem. 1997,
61, 1138.
Confluent HUVEC were treated with CAPE and the amide deriva-
tives for 5 h at 37 °C. After the 5 h incubation, the compounds were
removed from the wells, and the cells were washed twice with
MCDB 131 buffer. Stock hydrogen peroxidesolution (50%, Sigma–Al-
drich) was diluted in MCDB 131 buffer, and incubated in the cells fol-
lowing the buffer wash. HUVEC were incubated in the hydrogen

5038
J. Yang et al. / Bioorg. Med. Chem. 18 (2010) 5032–5038
24. Von Braun, M. Chem. Ber. 1927, 60, 351.
25. Laurent, S. A.-L.; Robert, A.; Meunier, B.; Boissier, J.; Cosledan, F.; Gornitzka, H.
Eur. J. Org. Chem. 2008.
26. Kunduzova, O. R.; Bianchi, P.; Parini, A.; Cambon, C. Eur. J. Pharmacol. 2002, 448,
225.
27. Li, W. M.; Liu, H. T.; Li, X. Y.; Wu, J. Y.; Xu, G.; Teng, Y. Z.; Ding, S. T.; Yu, C. Basic
Clin. Pharmacol. Toxicol. 2010, 106, 45.
29. Kirk, K. L.; Olubajo, O.; Buchhold, K.; Lewandowski, G. A.; Gusovsky, F.;
McCulloh, D.; Daly, J. W.; Creveling, C. R. J. Med. Chem. 1986, 29, 1982.
30. Wang, T.; Chen, L.; Wu, W.; Long, Y.; Wang, R. Can. J. Physiol. Pharmacol. 2008,
86, 279.
31. Bakalbassis, E. G.; Lithoxoidou, A. T.; Vafiadis, A. P. J. Phys. Chem. A 2006, 110,
11151.
32. Dinkova-Kostova, A. T.; Talalay, P. Mol. Nutr. Food Res. 2008, 52, S128.
33. Kakemi, K.; Sezaki, H.; Nahano, M.; Ohsuga, K.; Mitsunaga, T. Chem. Pharm. Bull.
1969, 17, 901.
28. Liu, H. T.; Li, W. M.; Xu, G.; Li, X. Y.; Bai, X. F.; Wei, P.; Yu, C.; Du, Y. G. Pharmacol.
Res. 2009, 59, 167.