Antiproliferative, antiandrogenic and cytotoxic effects of novel caffeic acid derivatives in LNCaP human androgen-dependent prostate cancer cells

Article abstract of DOI:10.1016/j.bmc.2013.08.057

Caffeic acid and its naturally occurring derivative caffeic acid phenethyl ester (CAPE) have antiproliferative and cytotoxic properties in a variety of cancer cell lines without displaying significant toxicity toward healthy cells, and are considered to be potential anticancer agents. However, little is known about their effects on prostate cancer cells. We synthesized and evaluated the effects of caffeic acid, CAPE (2) and 18 synthetic derivatives on cell viability and androgen-dependent cell proliferation, subcellular localisation and expression of androgen receptor (AR) and secretion of prostate-specific antigen (PSA) in LNCaP human hormone-dependent prostate cancer cells. Several synthetic derivatives of CAPE were strong, concentration-dependent cytotoxic agents in LNCaP cells with IC₅₀ values in the 6.8-26.6 μM range, potencies that were up to five-fold greater than that of CAPE (33.7 ± 4.0 μM). A number of caffeic acid derivatives were inhibitors of androgen-stimulated LNCaP cell proliferation with concomitant inhibition of DHT-stimulated PSA secretion. Compound 24 was the most cytotoxic and antiproliferative caffeic acid derivative (IC₅₀ values of 6.8 ± 0.3 and 2.4 ± 0.8 μM, respectively) inhibiting DHT-stimulated cell proliferation and PSA secretion statistically significantly at concentrations as low as 0.3 μM. Exposure to DHT increased cytoplasmic and nuclear AR levels and co-treatment with increasing concentrations of compound 24 or CAPE (2), notably, further increased these levels. In conclusion, a number of synthetic derivatives of caffeic acid are potent inhibitors of androgen-dependent prostate cancer cell proliferation and viability, acting, at least in part, via an antiandrogenic mechanism that involves increased nuclear accumulation of (presumably inactive) AR.

Full text of DOI:10.1016/j.bmc.2013.08.057

Bioorganic & Medicinal Chemistry 21 (2013) 7182–7193
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Antiproliferative, antiandrogenic and cytotoxic effects of novel
caffeic acid derivatives in LNCaP human androgen-dependent
prostate cancer cells
J. Thomas Sanderson a, , , Hélène Clabault a, Cody Patton a, Grégoire Lassalle-Claux b,
⇑
Jacques Jean-François b, Aurélie F. Paré b, Martin J. G. Hébert b, Marc E. Surette b, Mohamed Touaibia b,
⇑
,à
a INRS-Institut Armand-Frappier, 531, boulevard des Prairies, Laval, QC H7V 1B7, Canada
b Department of Chemistry and Biochemistry, Université de Moncton, Moncton, NB, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 3 July 2013
Revised 19 August 2013
Accepted 27 August 2013
Available online 6 September 2013
Caffeic acid and its naturally occurring derivative caffeic acid phenethyl ester (CAPE) have antiprolifera-
tive and cytotoxic properties in a variety of cancer cell lines without displaying significant toxicity toward
healthy cells, and are considered to be potential anticancer agents. However, little is known about their
effects on prostate cancer cells. We synthesized and evaluated the effects of caffeic acid, CAPE (2) and 18
synthetic derivatives on cell viability and androgen-dependent cell proliferation, subcellular localisation
and expression of androgen receptor (AR) and secretion of prostate-specific antigen (PSA) in LNCaP
human hormone-dependent prostate cancer cells. Several synthetic derivatives of CAPE were strong, con-
Keywords:
Caffeic acid
CAPE
Synthetic derivatives
Androgen receptor
LNCaP
centration-dependent cytotoxic agents in LNCaP cells with IC50 values in the 6.8–26.6
lM range, poten-
cies that were up to five-fold greater than that of CAPE (33.7 4.0 M). A number of caffeic acid
l
derivatives were inhibitors of androgen-stimulated LNCaP cell proliferation with concomitant inhibition
of DHT-stimulated PSA secretion. Compound 24 was the most cytotoxic and antiproliferative caffeic acid
Prostate cancer
Prostate specific antigen
Cytotoxicity
Real-time cell proliferation
Concentration–response
Structure–activity
derivative (IC50 values of 6.8 0.3 and 2.4 0.8
ation and PSA secretion statistically significantly at concentrations as low as 0.3
l
M, respectively) inhibiting DHT-stimulated cell prolifer-
M. Exposure to DHT
l
increased cytoplasmic and nuclear AR levels and co-treatment with increasing concentrations of com-
pound 24 or CAPE (2), notably, further increased these levels. In conclusion, a number of synthetic deriv-
atives of caffeic acid are potent inhibitors of androgen-dependent prostate cancer cell proliferation and
viability, acting, at least in part, via an antiandrogenic mechanism that involves increased nuclear accu-
mulation of (presumably inactive) AR.
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
deaths, to 350,000 by 2025.3 The Canadian Cancer Society predicts
26,500 new cases and 4000 prostate cancer deaths for 2012.4 Poor
The incidence of certain (longevity-corrected) cancers is
increasing in the West, including that of various hormone-depen-
dent cancers such as those of the breast, testis and prostate. In
Western men, prostate cancer is now the most common malig-
nancy (accounting for 30% of newly diagnosed cancers) and the
third leading cause of cancer-related death. The burden of human
suffering and cost to society are expected to increase significantly
in the coming decades due to increased life-expectancy.1,2 In the
United States alone the number of new cases is expected to grow
from 241,740 in 2012, a year that will see 28,170 prostate cancer
nutrition has been implicated as an important risk factor in these
developments. Epidemiological studies consistently associate a
life-time of high plant-derived consumption with reduced risk of
developing cancer.5–7 Unfortunately, a generation of people has
been raised on a high fat, carbohydrate and protein diet, low in fi-
bre and plant-derived nutrients. A plant-derived diet contains a
large variety of phytochemicals purported to have many health
benefits, including protection against cancers such as that of the
prostate.8–10
Caffeic acid (1, Fig. 1) is found in relatively high concentrations
in many plants as a key intermediate in the biosynthesis of lignins.
Caffeic acid and its naturally occurring derivative caffeic acid
phenethyl ester (CAPE) (2, Fig. 1), which is found in high concen-
trations in propolis of the honeybee hive have anticancer proper-
ties, as well as immunomodulatory, anti-inflammatory and
antioxidant effects. Caffeic acid (1) and CAPE (2) have been found
⇑
Corresponding authors.
E-mail addresses: thomas.sanderson@iaf.inrs.ca (J.T. Sanderson),
mohamed.
touaibia@umoncton.ca (M. Touaibia).
Correspondence regarding the synthesis of caffeic acid derivatives to M.T.
Correspondence regarding the biological activities of caffeic acid derivatives to
à
J.T.S.
0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.08.057

J. T. Sanderson et al. / Bioorg. Med. Chem. 21 (2013) 7182–7193
7183
O
O
O
O
R
HO
HO
HO
HO
HO
HO
HO
i
OH
O
O
OH
1
2
1
HO
3 : R = Me
4 : R = Et
Figure 1. Structures of caffeic acid (1) and CAPE (2).
5 : R = n-Pr
6 : R = I-Pr
7 : R = n-But
to have antiproliferative and cytotoxic properties in a variety of
(nonprostate) cancer cell lines11,12 without displaying significant
toxicity toward healthy cells. Only sporadically have the effects
of CAPE been evaluated in prostate cancer cells. CAPE (2) has inhib-
itory effects on steroid 5 alpha-reductase (SRD5A)13 and the tran-
8 : R = I-pentyl
Scheme 1. Synthesis of caffeic acid alkyl esters. Reagents and conditions: (i) HOR,
H2SO4, reflux, 2 h.
scription factor NF-j
b,14 which is involved in the activation of
The substituents were chosen for their electronic and steric
properties: appending progressively larger aliphatic moieties
(R = Me, to R = I-pentyl; R = phenyl to R = CH2CH2Naphtyl) not only
modifies the steric bulk of the molecules but also augments the
general lipophilicity. The addition of insaturation to (R = 3-methyl-
but-3-enyl) modulates flexibility of the side chain while appending
p-substituents to the aromatic moiety should allow investigation
of the effects of additional steric bulk as well as the presence of
an electron withdrawing and donating groups.
many cellular processes including cell proliferation and survival.
Although either of these targets may contribute to the antiprostate
cancer activity of CAPE (2), the relatively high concentrations re-
quired for inhibition suggest that other not yet established targets
are likely of greater importance in explaining the growth inhibitory
effects of CAPE (2), which we have found to be evident at lower
concentrations. Recently, a study showed that a number of tran-
scription factors involved in cell survival were inhibited and that
the tumor suppressor gene KLF6 and cyclin-dependent kinase
inhibitor p21 were upregulated in PC-3 androgen-independent
2.2. Effects of caffeic acid and its derivatives on LNCaP cell
viability
prostate cancer cells treated with 20 l
M of CAPE (2).15 However,
no studies to date have systematically evaluated synthetic analogs
of caffeic acid or CAPE (2) to search for molecules with improved
biological potencies against prostate cancer cell growth. Also, nei-
ther CAPE (2) nor other caffeic acid derivatives have been studied
in detail in hormone-dependent prostate cancer cells, which gener-
ally represents an early stage of the disease. The potential for syn-
thetic variations on the caffeic acid pharmacophore with potent
anticancer activities was emphasised recently in a review by one
of our laboratories.16 Given the biological significance of the caffeic
acid pharmacophore, we hypothesised that certain derivatives of
caffeic acid will have potent anticancer activities against human
prostate tumor cells, and may be suitable for the development of
chemotherapeutic agents effective against hormone-dependent
as well as -independent prostate cancer. To test part of this
hypothesis the objective of the present study was to synthesize a
series of caffeic acid alkyl- and aryl ester-derivatives and deter-
mine their antiproliferative, antiandrogenic and cytotoxic poten-
cies in androgen-dependent LNCaP prostate cancer cells. Also,
because of the presence of a redox-active catechol group in each
molecule, the radical scavenging activity of the newly synthesized
caffeic acid derivatives was evaluated.
Of 19 synthetic derivatives of caffeic acid, all but three 3, 4 and
34 decreased the cell viability of LNCaP cells concentration-depen-
dently after a 24 h exposure, including CAPE (2) (Fig. 2). Seven
derivatives of caffeic acid were potent inhibitors of cell viability,
with lower IC50 values than CAPE (2) (Table 1) and in most cases
resulted in greater than 80% reductions in cell viability at the high-
est concentrations. Compound 24 was the most cytotoxic deriva-
tive with an IC50 value of 6.8 0.3
reduction of cell viability of greater than 95% at 100
acid (1) increased cell viability after a 24 h exposure to the lowest
tested concentration of 1 M but this effect was no longer ob-
lM, (Table 1) and a maximum
lM. Caffeic
l
served at greater concentrations up to 100
lM, where cell viability
was not different from unexposed cells.
2.3. Effects of caffeic acid and its derivatives on LNCaP cell
proliferation
LNCaP cells cultured 2–18 passages in phenol red-free medium
containing 10% stripped FBS and supplemented with 0.1 nM DHT
proliferated with a doubling-time of 24 4 h (or a growth rate of
2.9%), whereas in the absence of DHT the doubling time was
134 12 h (or a growth rate of 0.52%). The doubling-time of
DHT-stimulated LNCaP cells decreased significantly after about
20 passages, whereas that of cells cultured under steroid-free con-
ditions would increase with doubling-times converging to about
80 h (a growth rate of 0.87%) after 25 passages (Fig. 3). All exposure
experiments with caffeic acid derivatives were performed with
cells passaged fewer than 18 times.
2. Results
2.1. Synthesis strategy
The strategy for synthesis of the alkyl esters is outlined in
Scheme 1. Compounds 3–8 were prepared by esterification with
selected alcohol and commercially available caffeic acid (1) in the
presence of sulfuric acid; the overall yield was about 80–90%.
The synthesis of the aryl esters of caffeic acid is depicted in
Scheme 2. First, the two hydroxyl groups of caffeic acid (1) were
protected by acetate groups in acetic anhydride in the presence
of sodium hydroxide. Esters 10–21 and an amide 22 were synthe-
sized from selected alcohol or 2-phenylethanamine and acetylated
caffeic acid 9. The conversion of 9 into the corresponding carbox-
ylic chloride was achieved by the Vilsmeier–Haack adduct17 de-
rived from thionyl chloride with N,N-dimethylformamide (DMF)
as catalyst. Base-induced de-O-acetylation resulted in the desired
CAPE (2) and caffeic acid esters (23–34); the overall yield was
about 40–60% with high purity after purification.18,19
Of 16 tested derivatives of caffeic acid, 12 concentration-depen-
dently decreased the growth rate of DHT-stimulated LNCaP cells at
1 and 10
whereas 3, 4, 23 and 30 did not have a statistically significant effect
on cell proliferation. Compounds 24, 32 and 33 at 10 M resulted
lM (Fig. 4), including CAPE (2), 5, 24–29 and 31–34,
l
in negative growth rates due to cell loss caused by cytotoxicity,
whereas CAPE (2), 24–27 and 33 reduced cell proliferation at con-
centrations below those that induced cytotoxicity in nonstimulat-
ed cells. The effects of low concentrations of CAPE (2) and
compound 24 show that androgen-stimulated prostate cancer cell
proliferation was inhibited at concentrations as low as 0.3
lM
(Fig. 5), which are well below those that caused cell death (Fig. 2).

7184
J. T. Sanderson et al. / Bioorg. Med. Chem. 21 (2013) 7182–7193
O
O
R1
R1
R1
R1
R
ii
X
OH
1 R1 = OH
9 R1 = OAc
10- 22 R1 = OAc; X = O, NH; R
iii
i
23 R1 = H; X = O; R = 3-methylbut-3-enyl
24 R1 = H; X = O; R = CH2CH2cycloHexyl
25 R1 = H; X = O; R = Ph
26 R1 = H; X = O; R = CH2Ph
02 R1 = H; X = O; R = CH2CH2Ph
27 R1 = H; X = O; R = CH2CH2CH2Ph
28 R1 = H; X = O; R = CH2CH2PhF(4-F)
29 R1 = H; X = O; R = CH2CH2PhMe(4-Me)
30 R1 = H; X = O; R = CH2CH2PhOMe(4-OMe)
31 R1 = H; X = O; R = CH2CH2Ph(4-NO2)
32 R1 = H; X = O; R = CH2CH(Ph)2
33 R1 = H; X = O; R = CH2CH2Naph
34 R1 = H; X = NH; R = CH2CH2Ph
Scheme 2. Synthesis of caffeic acid 3-methylbut-3-enyl, hexylethyl and arylethyl esters. Reagents and conditions: (i) Ac2O, NaOH, 0 °C, 1 h; (ii) SOCl2, DMF, 4 h, HOR or H2NR,
Et3N, CH2Cl2, rt, 12 h; (iii) guanidineꢀHCl, Et3N, CH2Cl2, MeO, rt, 4 h.
Figure 2. LNCaP androgen-dependent human prostate cancer cell viability after a 24 h exposure to various concentrations (0.1–100
lM) of caffeic acid (1), CAPE (2) and 18
caffeic acid derivatives. Concentration–response curves (n = 3) were analyzed by nonlinear curve fit to calculate IC50 values (Table 1).
Table 1
Cytotoxicity expressed as IC50
(lM) (n = 3) of caffeic acid, CAPE (2) and 18 caffeic acid derivatives in LNCaP human prostate cancer cells, together
with hydrophobicity constants (ClogP)
a
a
Caffeic acid derivative
IC50 SEM (
l
M)
ClogP
Caffeic acid derivative
IC50 SEM (
lM)
ClogP
Caffeic acid (1)
CAPE (2)
3
4
5
6
7
8
>100
33.7. 4.0
>100
0.975
3.293
1.201
1.730
2.259
2.039
2.788
3.187
2.703
4.380
25
26
27
28
29
30
31
32
33
34
65.9 10.4
45.6 2.4
26.6 2.5
20.8 1.3
14.0 1.0
47.3 2.8
33.5 4.9
20.8 3.9
9.3 1.2
>100
2.84
2.969
3.677
3.441
3.797
3.217
3.041
4.736
4.472
4.380
>100
60.7 8.9
51.9 4.8
45.4 6.2
45.4 2.5
20.6 1.9
6.8 0.3
23
24
a
Hydrophobicity constants (ClogP) were calculated using ChemDraw Ultra (version: 11.0.1, CambridgeSoft 1986–2007).
2.4. Effects of caffeic acid and its derivatives on androgen
receptor expression in LNCaP cells
(2) or compound 24. CAPE (2) and compound 24 were chosen be-
cause they were potent inhibitors of DHT-stimulated LNCaP cell
proliferation at sub-cytotoxic concentrations. DHT at 10 nM in-
creased cytoplasmic and nuclear AR protein levels statistically sig-
nificantly and by about 2–3 fold after a 24 h exposure (Fig. 6).
When LNCaP cells were exposed to a combination of 10 nM DHT
and several concentration of CAPE (2) or compound 24 a further
To determine whether the antiproliferative effects of the caffeic
acid derivatives was related to an antagonistic effect on DHT-med-
iated accumulation of nuclear AR, LNCaP cells were exposed to
DHT in the presence or absence of various concentrations of CAPE

J. T. Sanderson et al. / Bioorg. Med. Chem. 21 (2013) 7182–7193
7185
180
160
140
120
100
80
exposure. DHT (3 nM) increased the amount of PSA detected in
the extracellular medium of LNCaP cells by about four-fold, after
a 24 h exposure (Fig. 7). A combined exposure of LNCaP cells to
DHT and various concentrations of CAPE (2) or compound 24 re-
duced the amount of PSA recovered from the extracellular medium
concentration-dependently (Fig. 7).
DMSO
DHT
2.6. Effects of caffeic acid and its derivatives on radical
scavenging activity
60
Caffeic acid (1) and CAPE (2) are well-known antioxidants and
40
their IC50 values were 15–16
tions. All other caffeic acid derivatives exhibited radical scavenging
activities within the 10–25 M range (Table 2). Hydrophobicity
constants for the caffeic acid derivatives were also calculated and
are listed in Table 1.
lM under our experimental condi-
20
0
l
0
10
20
30
Passage number
Figure 3. The effect of passage number on dihydrotestosterone-(DHT)-stimulated
or basal proliferation of LNCaP cells cultured in steroid-free medium (n = 3).
3. Discussion
Our study is the first to describe the synthesis of a series of no-
vel caffeic acid derivatives and to evaluate their cytotoxic effects
on androgen-dependent prostate cancer cells. We further provide
the first evidence that caffeic acid derivatives including CAPE (2)
are capable of inhibiting androgen-dependent LNCaP prostate can-
cer cell proliferation with certain caffeic acid derivatives having
potencies up to five-fold greater than that of naturally occurring
CAPE (2). Structure–activity relationships for inhibition of cell via-
bility and androgen-dependent cell proliferation were not the
same among caffeic acid derivatives suggesting that each effect is
mediated via (a) different mechanistic pathway(s).
increase of nuclear AR protein levels was observed, which could be
as much as 5–10 fold greater than levels found in the cytoplasm or
nucleus of unexposed cells (Fig. 6). At concentrations of CAPE (2) or
compound 24 above 10 lM, AR protein levels in the cytoplasm de-
creased, whereas in the nucleus they continued to increase or re-
mained at levels well above those in cells treated with DHT
alone (Fig. 6).
2.5. Effects of caffeic acid and its derivatives on prostate specific
antigen release from LNCaP cells
To determine whether the increased levels of AR in the nucleus
resulted in the induction of AR-regulated proteins, the effects of
the caffeic acid derivatives CAPE (2) and compound 24 was
determined on the release of PSA from LNCaP cells during a 24 h
3.1. Cytotoxic effects of caffeic acid derivatives
Our caffeic acid derivatives including CAPE (2) inhibited LNCaP
cell viability with various potencies, with compound 24 being the
1.5
1.4
1.3
1.2
1.1
1
0.9
0.8
*
*
*
0.7
*
*
*
0.6
*
0.5
*
0.4
*
*
0.3
0.2
0.1
0
*
*
*
*
-0.1
1 uM
Caffeic acid derivaꢀve
10 uM
Figure 4. Effect of caffeic acid and synthetic derivatives on dihydrotestosterone-(DHT)-stimulated LNCaP cell proliferation determined by real-time impedance monitoring
(see Section 5.3). (⁄) Statistically significantly different from DHT-treated cells (one-way ANOVA with Dunnett posteriori test for multiple comparisons to control, p < 0.05;
n = 3).

7186
J. T. Sanderson et al. / Bioorg. Med. Chem. 21 (2013) 7182–7193
1.5
1.4
1.3
1.2
1.1
1
1.5
1.4
DMSO
DMSO
DHT (0.1 nM)
DHT (0.1 nM)
24 (0.1 μM)
24 (0.3 μM)
24 (1 μM)
24 (3 μM)
24 (5 μM)
1.3
CAPE (0.1 μM)
CAPE (0.3 μM)
1.2
CAPE (1 μM)
CAPE (3 μM)
1.1
CAPE (5 μM)
1.0
0.9
30
36
42
48
30
36
42
48
Time (h)
Time (h)
Figure 5. Concentration-dependent effect of CAPE (2) (A) or compound 24 (B) on dihydrotestosterone-(DHT)-stimulated LNCaP cell proliferation. Cells were exposed to
0.1 nM DHT in the presence or absence of various concentrations of the caffeic acid derivatives for 24 h in steroid-free culture medium. CAPE (2) and compound 24 had IC50
values for androgen-dependent growth inhibition of 3.0 0.3 and 2.4 0.8
lM, respectively. Each trace is the average signal of duplicate treatments determined by real-time
impedance monitoring (see Section 5.3). One of three experiments is shown.
Figure 6. Concentration-dependent effects of caffeic acid derivatives CAPE (2) and compound 24 on dihydrotestosterone-(DHT)-stimulated accumulation of AR protein in
cytoplasmic and nuclear fractions of LNCaP cells after a 24 h exposure. (a) Statistically significant difference from cytoplasmic or nuclear DMSO controls. (b) Statistically
significant difference from cytoplasmic or nuclear DHT treated cells (one-way ANOVA with Dunnett posteriori test for multiple comparisons to relevant control, p < 0.05;
n = 3).
most cytotoxic. CAPE (2) was toxic to LNCaP cells at considerably
concentration of 17.6
BxPC-3 cells, CAPE (2) induced mitochondrial apoptosis at
35
M.22 In C6 glioma cells CAPE (2) induced apoptosis after a
24 h exposure to 50 M and programmed cell death appeared to
l
M CAPE (2).21 In pancreatic PANC-1 and
lower concentrations (Fig. 1) than reported in PC-3 cells (no effect
at 20
l
M) but appeared to be equally toxic to MCF-7 cells,20 reduc-
l
ing cell viability at concentrations of 10
l
M and greater. The lesser
l
sensitivity of PC-3 cells compared to LNCaP cells may be related to
the absence of p53 protein in the former, but no comparative study
has yet been reported to assess the role of p53 in CAPE (2)-medi-
ated prostate cancer cell death. In MCF-7 cells CAPE (2) induced
cytotoxicity via an apoptotic pathway involving increased BAX
expression, Fas activation and stimulation of the p38 mitogen-acti-
vated protein kinase (MAPK) pathway. A 24 h exposure to CAPE (2)
involve activation of p38 MAPK and tumor suppressor protein
p53.23
3.2. Antiandrogenic effects of caffeic acid derivatives
Several CAPE (2) derivatives inhibited androgen-stimulated pro-
liferation of LNCaP cells at sub-cytotoxic concentrations, with CAPE
(2) and compound 24 among the most potent with IC50 values of 2.4
at 17.6
l
M induced apoptosis in U937 human myeloid leukemia
M and
cells, although cytotoxicity was already observed at 3.5
l
and 3.0 lM, respectively. Inhibition of androgen-dependent pros-
apoptotic markers such as increased cytochrome c release and
tate cancer cell growth may occur via several mechanisms, among
which inhibition of androgen synthesis, antagonism of androgen li-
gand binding to the AR, blocking of androgen-stimulated AR trans-
location to the nucleus and/or inhibition of downstream activation
PARP cleavage and decreased Bcl-2 expression were observed after
exposures as low as 0.35
not observed, although this was determined at a single high
l
M.21 In this study, Fas activation was

J. T. Sanderson et al. / Bioorg. Med. Chem. 21 (2013) 7182–7193
7187
co-activator proteins such as steroid receptor co-activators SRC-1
and -2, thus rendering the AR complex transcriptionally inactive.24
Detailed studies of the interactions between caffeic acid derivatives
and AR will shed light on the nature of the inactivation of AR signal-
ing that occurs in LNCaP cells. CAPE (2) was found to inhibit cell
proliferation in PC-3 cells, which do not express AR, at greater con-
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
l
M)15 than we found in LNCaP cells
*
*
centrations (IC50 = 51.4
(IC50 = 3 M; Fig. 5). The mechanism of CAPE (2)-mediated inhibi-
tion of PC-3 cell growth appeared to involve stimulation of cyclin-
dependent kinase inhibitor p21Cip, which was increased at 20
M.
l
*
*
l
A reduction of the expression of proteins in the Akt signaling path-
way was also observed, although they appeared to occur after long-
er exposures than required for inhibition of cell proliferation.15
*
3.3. Structure–activity
The cytotoxic and antiandrogenic effects of the caffeic acid
derivatives do not follow the same structure–activity pattern and
the rank order of cytotoxic and anti-proliferative potencies are dis-
similar. The caffeic acid derivative 24 was five times more potent
than CAPE (2) at killing LNCaP cells, although their antiandrogenic
activity was similar with CAPE (2) possibly more potent than 24.
Although the ClogP values (Table 1) of the 16 cytotoxic caffeic acid
derivatives were strongly and significantly correlated with their
IC50 values for decreasing cell viability (r = 0.775; p < 0.01; n = 16;
caffeic acid and derivatives 3, 4 and 34 were excluded), this was
not the case for their ability to decrease DHT-stimulated cell prolif-
eration, suggesting that steric factors and a three-dimensional
interaction play an influential role in the antiproliferative/antian-
drogenic effects of CAPE (2) derivatives. The strong correlation be-
tween hydrophobicity and cytotoxicity does not tell us by which
mechanism(s) compound 24 and other hydrophobic caffeic acid
derivatives are killing LNCaP cells more effectively than CAPE (2)
and other more water soluble derivatives. The greater hydropho-
bicity of compound 24 (ClogP = 4.38) than CAPE (2) (ClogP = 3.29)
could result in greater bioavailability to a common target site, or
the different molecular structure of compound 24 could result in
greater affinity of this caffeic acid derivative for the same or an-
other molecular target responsible for triggering LNCaP cell death.
The electronic nature of substitution of the phenethylester moiety
of CAPE (2) does not appear to differentially affect the cytotoxicity
of the derivative, as the fluorinated (28) and methylated (29) deriv-
atives of CAPE (2) both exhibit increased cytotoxic potency com-
pared to CAPE (2). Modification of the phenethylester moiety
with bulkier substituents such as diphenyl (32), naphthyl (33)
and cyclohexyl (24) further increase cytotoxic potency, which gi-
ven the different nature of the substituents appears to be due
+ DHT (3 nM)
Figure 7. Caffeic acid derivatives CAPE (2) and compound 24 reduce dihydrotes-
tosterone-(DHT)-stimulated PSA release from LNCaP cells concentration-depen-
dently after a 24 h exposure. CAPE (2) and compound 24 had IC50 values for
inhibition of DHT-stimulated PSA expression of 6.2 1.0 and 2.0 0.4 lM, respec-
tively. (⁄) Statistically significant difference from PSA secretion by DHT treated
LNCaP cells (one-way ANOVA with Dunnett posteriori test for multiple comparisons
to control, p < 0.05; n = 3).
Table 2
Antioxidant activity as free radical scavengers of caffeic acid and its derivatives
expressed as IC50. Values are means of two independent experiments, each performed
in triplicate
Caffeic acid
derivative
IC50 SEM
M)
Caffeic acid
derivative
IC50 SEM
M)
(l
(l
Caffeic acid (1)
CAPE (2)
3
4
5
6
7
8
15.3 1.1
16.5 4.0
21.3 3.9
13.3 0.9
14.2 1.7
13.0 2.2
12.6 2.3
23.8 4.8
9.5 0.7
25
26
27
28
29
30
31
32
33
34
12.5 2.6
15.3 2.0
11.9 0.6
12.8 1.1
11.3 0.9
17.4 0.8
15.4 0.9
18.5 2.1
13.2 1.1
13.8 0.4
23
24
15.1 2.3
of androgen responsive genes through interactions with participat-
ing transcription factors and/or cross-talk with other signaling
pathways such as that of Akt. Despite this complexity, it is unlikely
that inhibition of androgen-synthesis by inhibition of, for example,
SRD5A plays a role in the antiandrogenic affects observed with our
caffeic acid derivatives in LNCaP cells, because dihydrotestosterone
is added exogenously. Exposure of LNCaP cells to the growth inhib-
itory caffeic acid derivatives 24 and CAPE (2) resulted in decreased
DHT-stimulated PSA secretion (Fig. 7), indicating that AR depen-
dent gene expression was inhibited. The IC50 values for inhibition
of androgen-dependent PSA secretion (Fig. 7) were in the same
range as that for inhibition of androgen-stimulated cell prolifera-
tion (Fig. 5). However, CAPE (2) and compound 24 did not decrease
DHT-stimulated nuclear accumulation of AR but, in fact, further in-
creased it (Fig. 6). This observation would suggest that certain caf-
feic acid derivatives such as compound 24 and CAPE (2) do not
prevent DHT-mediated nuclear AR translocation or nuclear AR pro-
tein stabilization, but may possibly form an inactive nuclear com-
plex with AR. The clinically used antiandrogen bicalutamide
forms an inactive nuclear complex with AR that is still capable of
binding to AREs on the DNA but is no longer able to recruit
ˆ
mainly to an increase in hydrophobicity of the derivatives. More
detailed studies of the mechanisms of cell death induced by CAPE
(2) and other caffeic acid derivatives are needed to understand not
only the specific molecular pathways involved, but also the ob-
served differences in sensitivity among cell types to the toxic ef-
fects of these compounds. Figure
8 summarizes the most
significant results of the structure–activity relationship study on
Figure 8. Structure–activity relationship of a series of caffeic acid derivatives on
LNCaP cell viability and dihydrotestosterone-stimulated proliferation of LNCaP
cells.

7188
J. T. Sanderson et al. / Bioorg. Med. Chem. 21 (2013) 7182–7193
viability and proliferation of LNCaP and dihydrotestosterone-
(DHT)-stimulated LNCaP cells.
MEL-TEMPÒ (model 1001D) melting point apparatus. FTIR spectra
were recorded on a NicoletÒ Impact 400 spectrometer. NMR spec-
tra were recorded on a Bruker Avance III 400 MHz spectrometer.
Where necessary, DEPT, APT, and two dimensional 1H–1H COSY
experiments were performed for complete signal assignments.
Accurate mass measurements were performed on a MicrOTOF
instrument from Bruker Doltonics’ in positive electrospray. Either
protonated ions (M+H+) or sodium adducts (M+Na+) were used
for empirical formula confirmation. All compounds used for the
in vitro tests were 98% pure or more. Full details of synthetic pro-
tocols for all compounds and are provided below.
All caffeic acid derivatives contained a catechol moiety crucial
to antioxidant activity. Varying the chain length or degree of
branching of aliphatically substituted derivatives made little dif-
ference in antioxidant activity. Lengthening (from ethyl to butyl)
or branching of the aliphatic group on the ester moiety do not alter
significantly the measured IC50 (12–17 lM; Table 2). Nevertheless
a minimum length and a maximum level of branching of the ali-
phatic group of the ester moiety seem to be required for good anti-
oxidant activity. Thus, compound 3 with only a methyl group at the
ester position and compound 8 with an isopentyl group were
weaker antioxidants (IC50 values 21–24
lM) than the intermediate
5.1.2. Detailed synthetic procedures
alkylated derivatives. As illustrated by CAPE (2), the introduction in
the ester moiety of an aromatic group, whether it is bicyclic (naph-
talene, 33) or a phenyl group with either electron-donating or elec-
tron-withdrawing groups, did not alter the IC50 values (e.g.,
compounds 28–31). Neither did the introduction of a cyclohexyl
ring in compound 24 or changing the ester group to an amide
group as in compound 34, alter the observed range of IC50 values
5.1.2.1. Procedure 1: Synthesis of caffeic acid alkyl esters.
Caffeic acid (0.5 g, 2.77 mmol) was dissolved in dry appropriate
aliphatic alcohol (20 mL) and H2SO4 (0.5 mL, 96%, v/v) was added.
Reaction mixture was refluxed for 2 h. After cooling at room
temperature, the solution was diluted with ethyl acetate (100 mL)
and washed with an aqueous solution of NaHCO3 (5% w/v) until
neutral pH. The organic layer was then washed with distilled water
and dried over anhydrous MgSO4, and the solvent was removed
under vacuum. The residue was then purified by silica gel circular
chromatography to afford the required ester derivative.
around 12–17 lM. Earlier reports on the subject has underlined
the importance of the presence of catechol(s) and of carbonyl
groups separated from the aromatic ring in dictating free radical
scavenging activity.25 A more recent study by our group on the
same topic26 using propionic, acetic and benzoic acid analogues
of CAPE (2) have shown that the ethylene moiety is as important
as the nature of the link between the phenyl and the carbonyl
group. An ethyl-link promotes better antioxidant activity than a
propionyl group. The antioxidant activity of the caffeic acid deriv-
atives did not correlate with either antiproliferative or cytotoxic ef-
fects in LNCaP prostate cancer cells suggesting it does not play a
decisive role in these potentially chemotherapeutic activities.
5.1.2.2. Procedure 2: Synthesis of diacetylcaffeic acid aryl
esters.
A
mixture of diacetylcaffeic acid,19 5 mL of thionyl
chloride, and two drops of DMF was heated at reflux for 4 h. The
excess of thionyl chloride was removed on a Rotovap, and the
residue was dissolved in 4 mL of dry dichloromethane. To this
solution was slowly added 1 mL of pyridine and the appropriate
alcohol derivative (1.2 equiv). The resulting mixture was stirred
overnight at room temperature. After removal of solvents, the
residue was dissolved in ethyl acetate (50 mL), the organic extract
was washed with water (2 ꢁ 20 mL), brine (2 ꢁ 20 mL), and then
dried over MgSO4. The residue was then purified by silica gel
circular chromatography (Chromatotrons, Model 7924, Harrison
Research) to afford the required ester derivative.
4. Conclusion and perspectives
Our present study points at the androgen receptor and its sig-
naling pathway as one of the targets for inhibition of cell prolifer-
ation of human androgen-dependent prostate cancer cells by
certain derivatives of caffeic acid, including CAPE (2). We have also
demonstrated that certain caffeic acid derivatives and, in particu-
lar, compound 24, a cyclohexyl-substituted analog of CAPE (2),
has potent cytotoxic effects in LNCaP cells. These findings suggest
that caffeic acid derivatives with potent antiandrogenic and cyto-
toxic potentials could be potentially developed as effective chemo-
therapeutic agents against hormone-dependent prostate cancer.
The context in which certain caffeic acid derivative could be used
in the treatment of prostate cancer (as antihormonal or chemo-
therapeutic adjuvant) and the optimal delivery mechanisms (intra-
venously or orally, possibly in nanoparticle-encapsulated
sustained release form) remain to be studied in vivo, as well as
their classic absorption, distribution, metabolism and excretion
characteristics.
5.1.2.3. Procedure 3: Deacetylation of diacetylcaffeoyl deriva-
tives. The appropriate diacetylcaffeoyl derivative was dissolved
in 2 mL anhydrous CH2Cl2 under N2, to which was added 4 mL
MeOH. To the resulting stirred solution was added guanidinium
hydrochloride (3.25 equiv) followed by triethylamine (9.75 equiv).
After consumption of the diacetylated precursor (about 2 h), the
reaction mixture was concentrated and partitioned between 60 mL
AcOEt and 30 mL water. The organic phase was then washed again
with water, twice with saturated ammonium chloride, twice with
brine, treated with charcoal, dried over MgSO4 and concentrated to
give the resulting pure caffeoyl derivative as fine powders.
5.1.2.4. Compound 3: procedure 1. Ester 3 (473 mg, 89%) was
obtained after silica gel circular chromatography (40% EtOAc–
hexane) as a white solid; mp 160–161 °C; 1H NMR (400 MHz,
DMSO-d6, 25 °C) d (ppm): 9.64 (br s, 1H, OH), 9.20 (br s, 1H, OH),
7.49 (d, J = 15.93 Hz, 1H, @CHCar), 7.06 (s, 1H, Har), 7.01 (dd,
J = 8.16, 1.80 Hz, 1H, Har), 6.77 (d, J = 8.12 Hz, 1H, Har), 6.28 (d,
J = 15.93 Hz, 1H, @CHCO), 3.69 (s, 3H, CH3); 13C NMR (101 MHz,
DMSO-d6, 25 °C) d (ppm): 167.47, 148.88, 146.03, 145.64, 125.93,
121.89, 116.17, 115.24, 114.13, 51.68; HRMS m/z calcd for
5. Experimental section
5.1. Synthesis of caffeic acid derivatives
5.1.1. General
C
10H10O4+(H+): 195.0652; found: 195.0643.
All chemicals for the syntheses were purchased from Aldrich.
Purification of compounds was carried out by silica gel circular
chromatography (ChromatotronÒ, model 7924, Harrison Research,
Palo Alto, CA, USA). Thin layer chromatography (TLC) was per-
formed on silica gel coated aluminum sheets (SiliaPlate TLC, Silicy-
cleÒ) with detection by UV light (254 nm, UVS-11, MineralightÒ
shortwave UV lamp). Melting points were obtained using a
5.1.2.5. Compound 4: procedure 1. Ester 4 (382 mg, 81%) was
obtained after silica gel circular chromatography (40% EtOAc–
hexane) as a white solid; mp 128–140 °C; 1H NMR (400 MHz,
DMSO-d6, 25 °C) d (ppm): 9.58 (br s, 1H, OH), 9.14 (br s, 1H, OH),
7.47 (d, J = 15.93 Hz, 1H, @CHCar), 7.05 (s, 1H, Har), 7.00 (dd,

J. T. Sanderson et al. / Bioorg. Med. Chem. 21 (2013) 7182–7193
7189
J = 8.24, 1.16 Hz, 1H, Har), 6.76 (d, J = 8.12 Hz, 1H, Har), 6.26 (d,
J = 15.89 Hz, 1H, @CHCO), 4.16 (q, J = 7.04 Hz, 2H, CH2CH3), 1.24 (t,
J = 7.08 Hz, 3H, CH2CH3); 13C NMR (101 MHz, DMSO-d6, 25 °C) d
(ppm): 167.01, 148.83, 146.02, 145.45, 125.97, 121.80, 116.19,
115.26, 114.51, 60.15, 14.73; HRMS m/z calcd for C10H12O4+(H+):
209.0808; found: 209.0811.
C(CH3)@CH2); 13C NMR (101 MHz, CDCl3, 25 °C) d (ppm): 168.10,
168.01, 166.55, 143.46, 142.79, 142.42, 141.66, 133.32, 126.42,
123.92, 122.72, 119.31, 112.37, 62.88, 36.74, 22.54, 20.67, 20.63;
HRMS m/z calcd for C18H20O6+(H+): 333.1333; found: 333.1317.
5.1.2.11. Compound 11: procedure 2. Ester 11 (180 mg, 68%)
was obtained after silica gel circular chromatography (5–40%
EtOAc–hexane) as a yellow oil; 1H NMR (400 MHz, CDCl3, 25 °C) d
(ppm): 7.63 (d, J = 16.01 Hz, 1H, @CHCar), 7.43 (d, J = 8.48 Hz, 1H,
5.1.2.6. Compound 5: procedure 1. Ester 5 (580 mg, 94%) was
obtained after silica gel circular chromatography (40% EtOAc–
hexane) as a white solid; mp 114–115 °C; 1H NMR (400 MHz,
CDCl3, 25 °C) d (ppm): 7.62 (d, J = 15.93 Hz, 1H, @CHCar), 7.15 (s,
1H, Har), 7.03 (d, J = 8.28 Hz, 1H, Har), 6.91 (d, 8.20 Hz, 1H, Har),
6.50 (br s, 1H, OH), 6.30 (d, J = 15.93 Hz, 1H, @CHCO), 6.20 (br s, 1H,
OH), 4.20 (t, J = 6.68 Hz, 2H, CH2CH2CH3), 1.76 (sext., J = 7.20 Hz,
2H, CH2CH2CH3), 1.02 (t, J = 7.40 Hz, 3H, CH2CH2CH3); 13C NMR
(101 MHz, CDCl3, 25 °C) d (ppm): 167.96, 146.28, 144.84, 143.76,
127.56, 122.41, 115.73, 115.52, 114.46, 66.34, 22.07, 10.45; HRMS
m/z calcd for C12H14O4+(H+): 223.0965; found: 223.0963.
Har), 7.38 (s, 1H, Har), 7.24 (d, J = 8.36 Hz, 1H, Har), 6.40 (d, 1H,
15.97 Hz, @CHCO), 4.26 (t, J = 6.84 Hz, 2H, CH2CH2cyclohexyl), 2.33
(s, 6H, 2ꢁ OAc), 1.78–1.59 (m, 7H, CH2CH2cyclohexyl, CH2CH2-
cyclohexylH), 1.43 (br s, 1H, CH2CH2cyclohexylH), 1.32–1.13 (m,
3H, CH2CH2cyclohexylH), 1.02–0.93 (m, 2H, CH2CH2cyclohexylH);
13C NMR (101 MHz, CDCl3, 25 °C) d (ppm): 168.10, 168.02, 166.71,
143.41, 142.57, 142.41, 133.39, 126.40, 123.91, 122.70, 119.54,
63.01, 36.04, 34.58, 33.19, 26.50, 26.21, 20.68, 20.64; HRMS m/z
calcd for C21H26O6+(H+): 375.1802; found: 375.1834.
5.1.2.7. Compound 6: procedure 1. Ester 6 (392 mg, 63%) was
obtained after silica gel circular chromatography (40% EtOAc–
hexane) as a white solid; mp 138–139 °C; 1H NMR (400 MHz,
CDCl3, 25 °C) d (ppm): 7.60 (d, J = 15.89 Hz, 1H, @CHCar), 7.14 (s,
1H, Har), 7.03 (d, J = 8.24 Hz, 1H, Har), 6.91 (d, J = 8.20 Hz, 1H, Har),
6.27 (d, J = 15.89 Hz, 1H, @CHCO), 5.16 (sept., J = 6.28 Hz, 1H,
CH(CH3)2), 1.34 (d, J = 6.20 Hz, 6H, CH(CH3)2); 13C NMR (101 MHz,
CDCl3, 25 °C) d (ppm): 167.73, 146.49, 144.95, 143.86, 127.44,
122.33, 115.98, 115.48, 114.46, 68.29, 21.93; HRMS m/z calcd for
5.1.2.12. Compound 12: procedure 2. Ester 12 (280 mg, 74%)
was obtained after silica gel circular chromatography (5–40%
EtOAc–hexane) as a white solid; mp 80–81 °C; 1H NMR (400 MHz,
CDCl3, 25 °C) d (ppm): 7.83 (d, J = 16.01 Hz, 1H, @CHCar), 7.50 (d,
J = 8.40 Hz, 1H, Har), 7.45–7.42 (m, 3H, Har), 7.30–7.26 (m, 2H, Har),
7.19 (d, J = 7.88 Hz, 2H, Har), 6.60 (d, J = 15.97 Hz, 1H, @CHCO), 2.35
(s, 6H, 2ꢁ OAc); 13C NMR (101 MHz, CDCl3, 25 °C) d (ppm): 168.08,
167.99, 165.02, 150.70, 144.57, 143.82, 142.52, 133.03, 129.48,
126.64, 125.89, 124.07, 122.98, 121.59, 118.51, 20.70, 20.65; HRMS
m/z calcd for C19H16O6+(H+): 341.1020; found: 341.1007.
C
12H14O4+(H+): 223.0965; found: 223.0959.
5.1.2.8. Compound 7: procedure 1.
was obtained after silica gel circular chromatography (40%
EtOAc–hexane) as
white solid; mp 107–109 °C; 1H NMR
Ester 7 (470 mg, 72%)
5.1.2.13. Compound 13: procedure 2. Ester 13 (240 mg, 69%)
was obtained after silica gel circular chromatography (5–40%
EtOAc–hexane) as a white solid; mp 99–100 °C; 1H NMR (400 MHz,
CDCl3, 25 °C) d (ppm): 7.68 (d, J = 16.01 Hz, 1H, @CHCar), 7.43–7.35
(m, 7H, Har), 7.24 (d, J = 8.36 Hz, 1H, Har), 6.45 (d, J = 16.01 Hz, 1H,
@CHCO), 5.27 (s, 2H, CH2Ph), 2.32 (s, 6H, 2ꢁ OAc); 13C NMR
(101 MHz, CDCl3, 25 °C) d (ppm): 168.04, 167.96, 166.38, 143.54,
143.21, 142.45, 135.91, 133.24, 128.62, 128.31, 128.28, 126.39,
123.94, 122.79, 119.09, 66.50, 20.66, 20.61; HRMS m/z calcd for
a
(400 MHz, CDCl3, 25 °C) d (ppm): 7.62 (d, J = 15.93 Hz, 1H, @CHCar),
7.15 (s, 1H, Har), 7.03 (d, J = 8.20 Hz, 1H, Har), 6.91 (d, J = 8.20 Hz,
1H, Har), 6.37 (br s, 1H, OH), 6.30 (d, J = 15.93 Hz, 1H, @CHCO),
6.07 (br s, 1H, OH), 4.24 (t, J = 6.64 Hz, 2H, CH2(CH2)2CH3), 1.72
(quint., J = 6.88 Hz, 2H, CH2CH2CH2CH3), 1.46 (sext., J = 7.52 Hz,
2H, (CH 2)2CH2CH3), 0.99 (t, J = 7.32 Hz, 3H, (CH2)3CH3); 13C NMR
(101 MHz, CDCl3, 25 °C) d (ppm): 168.32, 146.54, 145.20, 143.88,
127.39, 122.42, 115.49, 115.47, 114.47, 64.77, 30.73, 19.19,
13.74; HRMS m/z calcd for C12H16O4+(H+): 237.1121; found:
237.1111.
C
20H18O6+(NHþ): 372.1442; found: 372.1427.
4
5.1.2.14. Compound 14: procedure 2.
was obtained after silica gel circular chromatography (5–40%
EtOAc–hexane) as
white solid; mp 80–82 °C; 1H NMR
Ester 14 (250 mg, 73%)
a
5.1.2.9. Compound 8: procedure 1. Ester 8 (360 mg, 52%) was
obtained after silica gel circular chromatography (40% EtOAc–
hexane) as a white solid; mp 127–128 °C; 1H NMR (400 MHz,
CDCl3, 25 °C) d (ppm): 7.61 (d, J = 15.93 Hz, 1H, @CHCar), 7.14 (s,
1H, Har), 7.03 (d, J = 8.24 Hz, 1H, Har), 6.91 (d, J = 8.16 Hz, 1H, Har),
6.29 (d, J = 15.89 Hz, br s, 2H, @CHCO, OH), 6.01 (br s, 1H, OH), 4.27
(t, J = 6.80 Hz, 2H, CH2CH2CH(CH3)2), 1.75 (sept., J = 6.48 Hz, 1H,
(CH2)2CH(CH3)2), 1.63 (q, J = 6.80 Hz, 2H, CH2CH2CH(CH3)2), 0.98
(d, J = 6.60 Hz, 6H, (CH2)2CH(CH3)2); 13C NMR (101 MHz, CDCl3,
25 °C) d (ppm): 168.20, 146.47, 145.09, 143.84, 127.44, 122.42,
115.55, 115.49, 114.46, 63.52, 37.38, 25.12, 22.49; HRMS m/z calcd
for C12H18O4+(H+): 251.1278; found: 251.1267.
(400 MHz, CDCl3, 25 °C): d 7.6 (d, 1H, J = 18.6 Hz, CHCar), 7.5–7.3
(m, 8H, Har), 6.35 (d, 1H, J = 18.6 Hz, CHCO), 4.3 (t, 2H, J = 4.7 Hz,
CH2O), 3.1 (t, 2H, J = 4.7 Hz, CH2CH2O), 2.3 (s, 6H, 2ꢁ OAc); 13C
NMR (101 MHz, CDCl3, 25 °C): d 168.1, 167.9, 166.4, 143.5, 142,9,
142.4, 137.8, 133.3, 128.9, 128.5, 126.6, 126.4, 123.9, 122.7,
119.2, 65.1, 35.2, 20.7, 20.6; LRMS: 369.1, 309.1, 265.1, 249.1,
231.1, 205.1, 163.1, 145.1, 117.1, 89.1; HRMS: calcd for C13H12-
O6+(Na+): 391.1152; found: 391.1151.
5.1.2.15. Compound 15: procedure 2. Ester 14 (314 mg, 73%)
was obtained after silica gel circular chromatography (5–40%
EtOAc–hexane) as a white solid; mp 66–67 °C; 1H NMR (400 MHz,
CDCl3, 25 °C) d (ppm): 7.63 (d, J = 15.93 Hz, 1H, @CHCar), 7.43 (d,
J = 8.32 Hz, 1H, Har), 7.39 (s, 1H, Har), 7.34–7.30 (m, 2H, Har), 7.26–
7.21 (m, 4H, Har), 6.41 (d, J = 15.97 Hz, 1H, @CHCO), 4.25 (t,
J = 6.44 Hz, 2H, CH2(CH2)2Ph), 2.77 (t, J = 7.40 Hz, 2H, (CH2)2-
CH2Ph), 2.33 (s, 6H, 2ꢁ OAc), 2.06 (quint., J = 6.68 Hz, 2H, CH2CH2-
CH2Ph); 13C NMR (101 MHz, CDCl3, 25 °C) d (ppm): 168.11, 168.02,
166.61, 143.46, 142.77, 142.43, 141.21, 133.33, 128.47, 128.45,
126.43, 126.04, 123.94, 122.73, 119.33, 64.06, 32.25, 30.27, 20.68,
20.65; HRMS m/z calcd for C22H22O6+(H+): 383.1489; found:
383.1444.
5.1.2.10. Compound 10: procedure 2. Ester 10 (152 mg, 63%)
was obtained after silica gel circular chromatography (0.2% MeOH–
dichloromethane) as a colorless oil; 1H NMR (400 MHz, CDCl3,
25 °C)
J = 8.32 Hz, 1H, Har), 7.38 (s, 1H, Har), 7.24 (d, J = 8.28 Hz, 1H,
ar), 6.40 (d, J = 15.97 Hz, 1H, @CHCO), 4.85 (s, 1H, (CH2)2-
d (ppm): 7.63 (d, J = 15.97 Hz, 1H, @CHCar), 7.42 (d,
H
C(CH3)@CHH), 4.80 (s, 1H, (CH2)2C(CH3)@CHH), 4.34 (t,
J = 6.72 Hz, 2H, CH2CH2C(CH3)@CH2), 2.43 (t, J = 6.62 Hz, 2H,
CH2CH2C(CH3)@CH2), 2.32 (s, 6H, 2ꢁ OAc), 1.81 (s, 3H, (CH2)2-

7190
J. T. Sanderson et al. / Bioorg. Med. Chem. 21 (2013) 7182–7193
5.1.2.16. Compound 16: procedure 2. Ester 15 (210 mg, 66%)
was obtained after silica gel circular chromatography (5–40%
EtOAc–hexane) as a yellow solid; mp 96–97 °C; 1H NMR (400 MHz,
CDCl3, 25 °C) d (ppm): 7.62 (d, J = 16.01 Hz, 1H, @CHCar), 7.42 (d,
J = 8.36 Hz, 1H, Har), 7.37 (s, 1H, Har), 7.23 (t, J = 8.36 Hz, 3H, Har),
7.02 (t, J = 8.56 Hz, 2H, Har), 6.38 (d, J = 15.97 Hz, 1H, @CHCO), 4.41
(t, J = 6.92 Hz, 2H, CH2CH2Ph(4-F)), 3.00 (t, J = 6.88 Hz, 2H, CH2-
CH2Ph(4-F)), 2.33 (s, 6H, 2ꢁ OAc); 13C NMR (101 MHz, CDCl3,
25 °C) d (ppm): 168.11, 168.00, 166.44, 162.95, 160.52, 143.53,
143.03, 142.44, 133.49, 133.46, 133.23, 130.38, 130.31, 126.47,
123.96, 122.72, 119.09, 115.47, 115.26; HRMS m/z calcd for
5.1.2.21. Compound 21: procedure 2. Ester 20 (170 mg, 63%)
was obtained after silica gel circular chromatography (5–40%
EtOAc–hexane) as
a
yellow solid; mp 143–144 °C; 1H NMR
(400 MHz, CDCl3, 25 °C) d (ppm): 8.17 (d, J = 8.32 Hz, 1H, Har),
7.90 (d, J = 8.00 Hz, 1H, Har), 7.80 (d, J = 7.56 Hz, 1H, Har), 7.61 (d,
J = 16.56 Hz, 1H, @CHCar), 7.61–7.51 (m, 2H, Har), 7.48–7.37 (m,
4H, Har), 7.25 (d, J = 9.88 Hz, 1H, Har), 6.39 (d, J = 16.01 Hz, 1H,
@CHCO), 4.58 (t, J = 7.24 Hz, 2H, CH2CH2Naph), 3.52 (t, J = 7.24 Hz,
2H, CH2CH2Naph), 2.34 (s, 6H, 2ꢁ OAc); 13C NMR (101 MHz, CDCl3,
25 °C) d (ppm): 168.11, 168.02, 166.56, 143.51, 142.99, 142.44,
133.88, 133.72, 133.29, 132.09, 128.86, 127.51, 127.05, 126.45,
126.23, 125.69, 125.55, 123.95, 123.66, 122.74, 119.21, 64.73,
32.28, 20.69, 20.65; HRMS m/z calcd for C25H22O6+(H+): 419.1489;
found: 419.1484.
C
21H19FO6+(NHþ): 404.1504; found: 404.1488.
4
5.1.2.17. Compound 17: procedure 2. Ester 16 (320 mg, 72%)
was obtained after silica gel circular chromatography (5–40%
EtOAc–hexane) as a white solid; mp 80–81 °C; 1H NMR (400 MHz,
CDCl3, 25 °C) d (ppm): 7.62 (d, J = 15.97 Hz, 1H, @CHCar), 7.42 (d,
J = 8.36 Hz, 1H, Har), 7.37 (s, 1H, Har), 7.24 (d, J = 8.32 Hz, 1H, Har),
7.18–7.16 (m, 4H, Har), 6.39 (d, J = 15.97 Hz, 1H, @CHCO), 4.42 (t,
J = 7.00 Hz, 2H, CH2CH2Ph(4-CH3)), 3.00 (t, J = 7.00 Hz, 2H, CH2-
CH2Ph(4-CH3)), 2.36 (s, 3H, CH2CH2Ph(4-CH3)), 2.33 (s, 6H, 2ꢁ
OAc); 13C NMR (101 MHz, CDCl3, 25 °C) d (ppm): 168.10, 168.01,
166.51, 143.47, 142.84, 142.43, 136.14, 134.69, 133.33, 129.24,
128.92, 128.81, 126.44, 123.93, 122.71, 119.30, 65.32, 34.73, 21.06,
20.68, 20.64; HRMS m/z calcd for C22H22O6+(Na+): 405.1309;
found: 405.1277.
5.1.2.22. Compound 22: procedure 2. Ester 20 (210 mg, 62%)
was obtained after silica gel circular chromatography (5–40%
EtOAc–hexane) as
(400 MHz, CDCl3, 25 °C)
a
yellow solid; mp 129–131 °C; 1H NMR
d
(ppm): 7.59 (d, J = 15.97 Hz, 1H,
@CHCar), 7.37–7.29 (m, 4H, Har), 7.26–7.18 (m, 4H, Har), 6.27 (d,
J = 15.97 Hz, 1H, @CHCO), 5.6 (br s, 1H, NH), 3.66 (d, J = 7.56 Hz,
2H, CH2CH2Ph), 2.90 (t, J = 7.52 Hz, 1H, CH2CH2Ph), 2.31 (s, 6H, 2ꢁ
OAc); 13C NMR (101 MHz, CDCl3, 25 °C) d (ppm): 168.15, 168.10,
165.37, 142.94, 142.34, 139.30, 139.24, 138.80, 133.80, 128.81,
128.73, 126.61, 126.22, 123.80, 122.27, 121.78, 121.73, 40.83,
35.60, 20.66, 20.64; HRMS m/z calcd for
C
21H21NO5+(H+):
368.1492; found: 445.1489.
5.1.2.18. Compound 18: procedure 2. Ester 17 (140 mg, 68%)
was obtained after silica gel circular chromatography (5–40%
5.1.2.23. Compound 23: procedure 3. Ester 22 (118 mg, 87%)
was obtained as
yellow solid; mp 117–118 °C; 1H NMR
EtOAc–hexane) as
(400 MHz, CDCl3, 25 °C)
@CHCar), 7.42 (d, J = 8.32, 1H, Har), 7.37 (s, 1H, Har), 7.24 (d,
J = 8.36 Hz, 1H, ar), 7.19 (d, J = 8.04 Hz, 2H, Har), 6.88 (d,
a
yellow solid; mp 104–105 °C; 1H NMR
a
d
(ppm): 7.62 (d, J = 15.97 Hz, 1H,
(400 MHz, DMSO-d6, 25 °C) d (ppm): 9.62 (br s, 1H, OH), 9.17 (br
s, 1H, OH), 7.47 (d, J = 15.85 Hz, 1H, @CHCar), 7.05 (s, 1H, Har), 7.00
(d, J = 8.12 Hz, 1H, Har), 6.76 (d, J = 8.00 Hz, 1H, Har), 6.25 (d,
J = 15.85 Hz, 1H, @CHCO), 4.79 (s, 1H, (CH2)2C(CH3)@CHH), 4.75 (s,
1H, (CH2)2C(CH3)@CHH), 4.23 (t, J = 6.48 Hz, 2H, CH2CH2C(CH3)-
@CH2), 2.35 (t, J = 6.40 Hz, 2H, CH2CH2C(CH3)@CH2), 1.74 (s, 3H,
H
J = 8.08 Hz, 2H, Har), 6.39 (d, J = 15.97 Hz, 1H, @CHCO), 4.40 (t,
J = 7.00 Hz, 2H, CH2CH2Ph(4-OCH3)), 3.82 (s, 3H, CH2CH2Ph(4-
OCH3)), 2.97 (t, J = 6.92 Hz, 2H, CH2CH2Ph(4-OCH3)), 2.33 (s, 6H, 2ꢁ
OAc); 13C NMR (101 MHz, CDCl3, 25 °C) d (ppm): 168.11, 168.01,
166.51, 158.33, 143.48, 142.86, 142.43, 133.31, 129.90, 129.81,
126.45, 123.94, 122.71, 119.29, 113.96, 65.41, 55.27, 34.29, 20.68,
20.64; HRMS m/z calcd for C22H22O7+(H+): 399.1438; found:
399.1430.
CH2CH2C(CH3)@CH2); 13C NMR (101 MHz, DMSO-d6, 25 °C)
d
(ppm): 166.95, 148.86, 146.02, 145.61, 142.27, 125.90, 121.84,
116.19, 115.26, 114.33, 112.67, 62.29, 22.72; HRMS m/z calcd for
C
14H16O4+(H+): 249.1121; found: 249.1114.
5.1.2.24. Compound 24: procedure 3. Ester 23 (65 mg, 83%) was
obtained as a yellow solid; mp 147–148 °C; 1H NMR (400 MHz,
DMSO-d6, 25 °C) d (ppm): 9.61 (br s, 1H, OH), 9.16 (br s, 1H, OH),
7.46 (d, J = 15.89 Hz, 1H, @CHCar), 7.05 (s, 1H, Har), 7.01 (d,
5.1.2.19. Compound 19: procedure 2. Ester 18 (180 mg, 56%)
was obtained after silica gel circular chromatography (5–40%
EtOAc–hexane) as a yellow solid; mp 91–93 °C; 1H NMR (400 MHz,
CDCl3, 25 °C) d (ppm): 8.21 (d, J = 8.48 Hz, 2H, Har), 7.61 (d,
J = 15.97 Hz, 1H, @CHCar), 7.45–7.40 (m, 2H, Har), 7.37 (s, 1H, Har),
7.24 (d, J = 8.36 Hz, 1H, Har), 6.35 (d, J = 15.97 Hz, 1H, @CHCO), 4.48
(t, J = 6.60 Hz, 2H, CH2CH2Ph(4-NO2)), 3.14 (t, J = 6.56 Hz, 2H,
CH2CH2Ph(4-NO2)), 2.33 (s, 6H, 2ꢁ OAc); 13C NMR (101 MHz,
CDCl3, 25 °C) d (ppm): 166.12, 168.00, 166.30, 146.92, 145.71,
143.64, 143.42, 142.47, 133.04, 129.78, 126.53, 124.00, 123.81,
122.76, 118.66, 64.04, 35.01, 20.67, 20.64; HRMS m/z calcd for
J = 8.20 Hz, 1H,
Har), 6.76 (d, J = 8.08 Hz, 1H, Har), 6.26 (d,
J = 15.89 Hz, 1H, @CHCO), 4.15 (t, J = 6.52 Hz, 2H, CH2CH2cyclo-
hexyl), 1.72–1.60 (m, 5H, CH2CH2cyclohexylH), 1.52 (q, J = 6.56 Hz,
2H, CH2CH2cyclohexyl), 1.36 (br s, 1H, CH2CH2cyclohexylH), 1.25–
1.11 (m, 3H, CH2CH2cyclohexylH), 0.96–0.88 (m, 2H, CH2CH2cyclo-
hexylH); 13C NMR (101 MHz, DMSO-d6, 25 °C) d (ppm): 167.07,
148.83, 146.01, 145.48, 125.95, 121.81, 116.18, 115.29, 114.48,
62.25, 36.13, 34.48, 33.07, 26.49, 26.15; HRMS m/z calcd for
C
21H19O8+(H+): 414.1183; found: 414.1166.
C
17H22O4+(H+): 291.1591; found: 291.1579.
5.1.2.20. Compound 20: procedure 2. Ester 19 (210 mg, 69%)
was obtained after silica gel circular chromatography (5–40%
EtOAc–hexane) as a yellow oil; 1H NMR (400 MHz, CDCl3, 25 °C) d
(ppm): 7.55 (d, J = 15.97 Hz, 1H, @CHCar), 7.38–7.21 (m, 13H, Har),
6.31 (d, J = 15.97 Hz, 1H, @CHCO), 4.78 (d, J = 7.56 Hz, 2H, CH2-
CH(Ph)2), 4.46 (t, J = 7.52 Hz, 1H, CH2CH(Ph)2), 2.32 (s, 6H, 2ꢁ
OAc); 13C NMR (101 MHz, CDCl3, 25 °C) d (ppm): 168.09, 168.00,
166.46, 143.49, 143.05, 142.40, 141.11, 133.22, 128.62, 128.23,
126.85, 126.49, 123.90, 122.70, 119.05, 66.90, 49.90, 20.67, 20.63;
HRMS m/z calcd for C27H24O6+(H+): 445.1646; found: 445.1628.
5.1.2.25. Compound 25: procedure 3. Ester 24 (166 mg, 82%)
was obtained as a white solid; mp 156–157 °C; 1H NMR (400 MHz,
DMSO-d6, 25 °C) d (ppm): 9.74 (br s, 1H, OH), 9.23 (br s, 1H, OH),
7.69 (d, J = 15.85 Hz, 1H, @CHCar), 7.44 (t, J = 7.72 Hz, 2H, Har), 7.28
(t, J = 7.40 Hz, 1H, Har), 7.19 (d, J = 7.76 Hz, 2H, Har), 7.15 (s, 1H,
H
ar), 7.11 (d, J = 8.20 Hz, 1H, Har), 6.80 (d, J = 8.08 Hz, 1H, Har), 6.51
(d, J = 15.89 Hz, 1H, @CHCO); 13C NMR (101 MHz, DMSO-d6, 25 °C)
d (ppm): 165.74, 151.09, 149.35, 147.60, 146.10, 129.94, 126.12,
125.82, 122.38, 116.24, 115.55, 113.34; HRMS m/z calcd for
C
15H12O4+(H+): 257.0808; found: 257.0796.

J. T. Sanderson et al. / Bioorg. Med. Chem. 21 (2013) 7182–7193
7191
5.1.2.26. Compound 26: procedure 3. Ester 25 (142 mg, 78%)
was obtained as a white solid; mp 150–151 °C; 1H NMR (400 MHz,
DMSO-d6, 25 °C) d (ppm): 9.65 (br s, 1H, OH), 9.18 (br s, 1H, OH),
7.53 (d, J = 15.89 Hz, @CHCar), 7.42–7.34 (m, 5H, Har), 7.07 (s, 1H,
Har), 6.76 (d, J = 8.08 Hz, 1H, Har), 6.22 (d, J = 15.93 Hz, 1H,
@CHCO), 4.39 (t, J = 6.40 Hz, 2H, CH2CH2Ph(4-NO2)), 3.11 (t,
J = 6.36 Hz, 2H, CH2CH2Ph(4-NO2)); 13C NMR (101 MHz, DMSO-d6,
25 °C) d (ppm): 166.86, 148.96, 147.21, 146.73, 146.03, 145.86,
130.73, 125.84, 123.90, 121.92, 116.17, 115.31, 114.05, 63.99,
34.68; HRMS m/z calcd for C17H15NO6+(H+): 330.0972; found:
330.0957.
H
ar), 7.03 (d, J = 8.24 Hz, 1H, Har), 6.77 (d, J = 8.00 Hz, 1H, Har), 6.34
(d, J = 15.93 Hz, @CHCO), 5.20 (s, 2H, CH2Ph); 13C NMR (101 MHz,
DMSO-d6, 25 °C) d (ppm): 166.90, 148.96, 146.02, 136.91, 128.93,
128.48, 125.92, 121.97, 116.18, 115.34, 114.13, 65.75; HRMS m/z
calcd for C16H14O4+(Na+): 293.0784; found: 293.0772.
5.1.2.32. Compound 32: procedure 3. Ester 31 (83 mg mg, 96%)
was obtained as a white solid; mp 185–186 °C;1H NMR (400 MHz,
DMSO-d6, 25 °C) d (ppm): 7.40–7.30 (m, 9H, @CHCar, Har) 7.22 (t,
J = 7.00 Hz, 2H, Har), 6.99 (s, 1H, Har), 6.95 (d, J = 9.44 Hz, 1H, Har),
6.74 (d, J = 8.08 Hz, 1H, Har), 6.16 (d, J = 15.89 Hz, 1H, @CHCO), 4.71
(d, J = 7.64 Hz, 2H, CH2CH(Ph)2), 4.43 (t, J = 7.64 Hz, 1H, CH2-
CH(Ph)2); 13C NMR (101 MHz, DMSO-d6, 25 °C) d (ppm): 166.89,
148.93, 145.99, 145.88, 142.05, 128.98, 128.45, 127.09, 125.78,
121.84, 116.18, 115.34, 114.04, 66.33, 49.86; HRMS m/z calcd for
5.1.2.27. Compound 27: procedure 3. Ester 26 (119 mg, 81%)
was obtained as a white solid; mp 121–122 °C; 1H NMR (400 MHz,
DMSO-d6, 25 °C) d (ppm): 9.63 (br s, 1H, OH), 9.18 (br s, 1H, OH),
7.47 (d, J = 15.85 Hz, 1H, @CHCar), 7.29 (t, J = 7.48 Hz, 2H, Har),
7.24–7.17 (m, 3H, Har), 7.06 (s, 1H, Har), 7.02 (d, J = 9.16 Hz, 1H,
Har), 6.77 (d, J = 8.08 Hz, 1H, Har), 6.28 (d, J = 15.89 Hz, 1H,
@CHCO), 4.11 (t, J = 6.48 Hz, 2H, CH2(CH2)2Ph), 2.68 (t,
J = 7.44 Hz, 2H, (CH2)2CH2Ph), 1.94 (quint., J = 6.80 Hz, 2H, CH2-
CH2CH2Ph); 13C NMR (101 MHz, DMSO-d6, 25 °C) d (ppm): 167.06,
148.85, 146.02, 145.55, 141.68, 128.81, 128.78, 126.33, 125.97,
121.84, 116.18, 115.29, 114.39, 63.58, 31.96, 30.36; HRMS m/z
calcd for C18H18O4+(H+): 299.1278; found: 299.1264.
C
23H20O4+(Na+): 383.1254; found: 383.1255.
5.1.2.33. Compound 33: procedure 3.
Ester 32 (78 mg, 83%)
was obtained as
a
white solid; mp 167–168 °C; 1H NMR
(400 MHz, DMSO-d6, 25 °C) d (ppm): 9.64 (br s, 1H, OH), 9.19 (br
s, 1H, OH), 8.20 (d, J = 8.24 Hz, 1H, Har), 7.94 (d, J = 7.92 Hz, 1H,
Har), 7.83 (t, J = 4.08 Hz, 1H, Har), 7.61–742 (m, 5H, Har, @CHCar),
7.03 (s, 1H, Har), 6.98 (d, J = 8.16 Hz, 1H, Har), 6.77 (d, J = 8.08 Hz,
1H, Har), 6.23 (d, J = 15.89 Hz, 1H, @CHCO), 4.43 (t, J = 6.88 Hz,
2H, CH2CH2Naph), 3.44 (t, J = 6.84 Hz, 2H, CH2CH2Naph); 13C
5.1.2.28. Compound 28: procedure 3. Ester 27 (51 mg, 84%) was
obtained as a white solid; mp 143–144 °C; 1H NMR (400 MHz,
DMSO-d6, 25 °C) d (ppm): 7.45 (d, J = 15.81 Hz, 1H, @CHCar), 7.33
(t, J = 8.08 Hz, 2H, Har), 7.14 (t, J = 8.88 Hz, 2H, Har), 7.04 (s, 1H,
Har), 7.00 (d, J = 7.92 Hz, 1H, Har), 6.76 (d, J = 7.96 Hz, 1H, Har), 6.23
NMR (101 MHz, DMSO-d6, 25 °C) d (ppm): 167.00, 148.92,
(d, J = 15.89 Hz, 1H, @CHCO), 4.30 (t, J = 6.72 Hz, 2H, CH2CH2Ph(4-
F)), 2.94 (t, J = 6.76 Hz, 2H, CH2CH2Ph(4-F)); 13C NMR (101 MHz,
DMSO-d6, 25 °C) d (ppm): 166.91, 162.64, 160.24, 148.91, 146.02,
145.71, 134.78, 134.75, 131.22, 131.14, 125.88, 121.88, 116.18,
115.62, 115.41, 115.28, 114.22, 64.72, 34.06; HRMS m/z calcd for
146.04, 145.76, 134.52, 133.89, 132.11, 129.08, 127.57, 126.65,
126.17, 126.06, 125.89, 124.20, 121.87, 116.20, 115.23, 114.23,
64.38, 32.02; HRMS m/z calcd for C21H18O4+(H+): 335.1278; found:
335.1275.
C
17H15FO4+(H+): 303.1027; found: 303.1013.
5.1.2.34. Compound 34: procedure 3. Amide 33 (87 mg, 78%)
was obtained as a white solid; mp 152–153 °C; 1H NMR (400 MHz,
DMSO-d6, 25 °C) d (ppm): 8.23 (br s, 1H, NH), 7.42 (d, J = 15.56 Hz,
1H, @CHCar), 7.32–7.19 (m, 5H, Har), 7.07 (br s, 1H, Har), 6.95–6.93
(m, 1H, Har), 6.84 (d, J = 8.12 Hz, 1H, Har) 6.41 (d, J = 15.56 Hz, 1H,
@CHCO), 3.57–3.52 (m, 2H, CH2CH2Ph), 2.98–2.84 (m, 2H, CH2-
CH2Ph); 13C NMR (101 MHz, DMSO-d6, 25 °C) d (ppm): 166.63,
146.84, 145.24, 139.50, 138.78, 135.19, 128.73, 128.56, 126.53,
126.34, 124.16, 122.27, 120.76, 119.56, 40.78, 35.84; HRMS m/z
calcd for C17H17NO3+(H+): 284.1281; found: 284.1279.
5.1.2.29. Compound 29: procedure 3.
Ester 28 (128 mg, 89%)
was obtained as a white solid; mp 161–162 °C; 1H NMR (400 MHz,
DMSO-d6, 25 °C) d (ppm): 9.60 (br s, 1H, OH), 9.20 (br s, 1H, OH),
7.45 (d, J = 15.89 Hz, 1H, @CHCar), 7.17 (d, J = 7.72 Hz, 2H, Har),
7.11 (d, J = 7.68 Hz, 2H, Har), 7.04 (s, 1H, Har), 7.00 (d, J = 8.16 Hz,
1H, Har), 6.76 (d, J = 8.08 Hz, 1H, Har), 6.23 (d, J = 15.89 Hz, 1H,
Har), 4.28 (t, J = 6.80 Hz, 2H, CH2CH2Ph(4-CH3)), 2.90 (t,
J = 6.76 Hz, 2H, CH2CH2Ph(4-CH3)), 2.27 (s, 3H, CH2CH2Ph(4-
CH3)); 13C NMR (101 MHz, DMSO-d6, 25 °C) d (ppm): 166.94,
148.89, 146.02, 145.65, 135.78, 135.41, 129.41, 129.22, 125.90,
121.86, 116.19, 115.27, 114.29, 64.89, 34.53, 21.11; HRMS m/z
calcd for C18H18O4+(H+): 299.1278; found: 299.1272.
5.2. Cell culture
LNCaP cells (CRL-1740; American Type Culture Collection, Rock-
land, MD) were cultured in 75 cm2 flasks in RPMI 1640 medium
5.1.2.30. Compound 30: procedure 3. Ester 29 (73 mg, 88%) was
obtained as a white solid; mp 185–186 °C;1H NMR (400 MHz,
DMSO-d6, 25 °C) d (ppm): 9.63 (br s, 1H, OH), 9.17 (br s, 1H, OH),
7.46 (d, J = 15.81 Hz, 1H, @CHCar), 7.20 (d, J = 8.32 Hz, 2H, Har), 7.04
(s, 1H, Har), 7.00 (d, J = 8.20 Hz, 1H, Har), 6.87 (d, J = 8.36 Hz, 2H,
containing 10% fetal bovine serum (FBS), 2 mM L-glutamine,
10 mM HEPES, 1 mM sodium pyruvate, 4500 mg/L glucose,
1500 mg/L sodium bicarbonate, 100 U/L penicillin and 100 mg/L
streptomycin, under an atmosphere of 5% CO2 at 37 °C. Early pas-
sage numbers were cryogenically cooled and preserved at
ꢂ80 °C. Cells were cultured in complete medium for no more than
18 passages, after which androgen-responsiveness was found to
decline (Fig. 3).
Har), 6.76 (d, J = 8.12 Hz, 1H, Har), 6.24 (d, J = 15.93 Hz, 1H,
@CHCO), 4.27 (t, J = 6.84 Hz, 2H, CH2CH2Ph(4-OCH3)), 3.72 (s, 3H,
CH2CH2Ph(4-OCH3)), 2.88 (t, J = 6.84 Hz, 2H, CH2CH2Ph(4-OCH3));
13C NMR (101 MHz, DMSO-d6, 25 °C) d (ppm): 166.95, 158.31,
148.88, 146.02, 145.64, 130.33, 125.90, 121.86, 116.18, 115.27,
114.31, 114.25, 65.03, 55.43, 34.07; HRMS m/z calcd for
5.3. Cell viability and proliferation
C
18H18O5+(K+): 353.0786; found: 353.0772.
LNCaP cells (5.0 ꢁ 104 cells/ml) were added to 24-well plates
(CellBind; Corning LifeSciences, Lowell, MA) in complete culture
medium and allowed to acclimatise for 24 h. Medium was then re-
freshed and various concentrations of caffeic acid or its derivatives
(Table 1) dissolved in DMSO were added (final DMSO concentra-
tion 0.1% in culture medium). Control cells were exposed to 0.1%
DMSO alone. After another 24 h, culture medium was removed
5.1.2.31. Compound 31: procedure 3. Ester 30 (63 mg, 92%) was
obtained as a white solid; mp 168–169 °C; 1H NMR (400 MHz,
DMSO-d6, 25 °C) d (ppm): 9.40 (br s, 1H, OH), 9.38 (br s, 1H, OH),
8.19 (d, J = 8.56 Hz, 2H, Har), 7.60 (d, J = 8.52 Hz, 2H, Har), 7.45 (d,
J = 15.93 Hz, 1H, @CHCar), 7.04 (s, 1H, Har), 6.99 (d, J = 8.20 Hz, 1H,

7192
J. T. Sanderson et al. / Bioorg. Med. Chem. 21 (2013) 7182–7193
and cell viability was determined using a WST-1 mitochondrial
reductive cleavage assay following the enclosed instructions
(Roche Diagnostics, Laval, QC). LNCaP cell proliferation was mea-
sured in real-time using an xCELLigence cell event monitoring sys-
tem (Roche Diagnostics). LNCaP cells were added to 16-well E-
plates (Roche Diagnostics) at a density of 25 ꢁ 103 cells/well in
phenol red-free culture medium supplemented with 10% stripped
fetal bovine serum (FBS; Hyclone, Logan, UT). After 24 h, cells were
exposed to various concentrations of caffeic acid or its derivatives,
in the presence or absence of 0.1 nM dihydrotestosterone (DHT;
Steraloids, Newport, RI) dissolved in DMSO (final DMSO concentra-
tion 0.2% in culture medium). Cells exposed to 0.1 nM DHT or 0.2%
DMSO alone were used as positive and negative controls, respec-
tively. Cell proliferation was determined quantitatively and in
real-time over a period of 72 h by measuring changes in impedance
detected by the gold electrode-microarrays at the bottom of each
of the 16 wells of the Eplate.
the calculation of IC50 values. A solution of 60 mM DPPH (1 mL in
ethanol) was mixed with 1 mL (in ethanol) of increasing concentra-
tions of each caffeic acid derivative or with ethanol alone. Each
mixture was then shaken vigorously and kept in the dark for
30 min at room temperature, after which the absorbance of DPPH
at 520 nm was measured.
5.7. Data analyses
Statistically significant differences between treatments and
control were determined by analysis of variance with correction
for multiple comparisons to control using a posteriori Dunnett test
(p < 0.05). Rates of cell proliferation and doubling-times were cal-
culated using the following formula:
logð2Þ
À
Á
Td ¼
r
log 1 þ 100
where Td is the doubling time (h) and r is the percentage growth
5.4. Androgen receptor expression
rate. The radical scavenging activity was expressed in terms of%
inhibition of DPPH absorbance using the equation
% inhibi-
LNCaP cells were cultured in 25 cm2 flasks (Corning) in standard
culture medium until almost confluent after which medium was
replaced with steroid-free medium. After 24 h, cells were exposed
to various concentrations of caffeic acid or its derivatives in the
presence or absence of 10 nM DHT for and additional 24 h. Cells
were then harvested and nuclear and cytoplasmic protein fractions
were isolated using a NE-PER Nuclear and Cytoplasmic Extraction
Reagent kit (Pierce Biotechnologies, Rockford, IL). Protein concen-
trations were determined using a bicinchoninic acid (BCA) protein
tion = [(Acontrol ꢂ Atest)/Acontrol)] ꢁ 100. All treatments were per-
formed in quadruplicate (cell viability and PSA experiments) or in
duplicate (cell proliferation, androgen receptor expression and rad-
ical scavenging experiments) and each experiment was performed
three times. IC50 values for radical scavenging activity and inhibi-
tion of cell viability were calculated using a sigmoidal concentra-
tion–response curve-fitting model with
a
variable slope
(GraphPad Prism 5, GraphPad Software, San Diego, CA).
assay kit (Pierce). Proteins (40 lg) were separated by sodium dode-
Acknowledgements
cylsulfate/polyacrylamide gel electrophoresis and transferred to
polyvinylidene diflouride (PVDF) membranes using a Trans-BlotÓ
Turbo system (BioRad, Mississauga, ON). The membranes were
then incubated overnight at 4 °C with primary antibodies raised
against human androgen receptor protein (anti-AR PG-21; Mili-
pore, Billerica, MA). The next day membranes were washed three
times in Tris–buffered saline containing 0.1% Tween 20 (TBS-T
0.1%), followed by three rinses with 1ꢁ TBS and an incubation of
60 min with horseradish peroxidase-conjugated goat-anti-rabbit
secondary antibodies (Millipore). Immunoreactive bands were
made visible using an Immobilon Western HRP chemolumines-
cence detection kit (Millipore) and digitally scanned using a Fluor-
Chem FC2 imager (AlphaInnotech, San Leandro, CA).
This work was funded by the Natural Sciences and Engineering
Research Council of Canada (J.T.S.; Grant No. 313313-2012) for the
development of structure–activity relationships of endocrine-ac-
tive molecules as well as by the New Brunswick Health Research
Foundation and the New Brunswick Innovation Foundation (M.T.)
in support of the synthesis of bioactive molecules. C.P. was sup-
ported by an undergraduate bursary from the Fondation Armand-
Frappier, and A.F.P. and M.J.G.H. by fellowships from the New
Brunswick Health Research Foundation. This study presents part
of the research activities of H.C. toward fulfilling a Master’s 2 de-
gree at the Université Paris-Sud 11.
References and notes
5.5. Prostate specific antigen secretion
1. Jemal, A.; Siegel, R.; Ward, E.; Murray, T.; Xu, J.; Thun, M. J. CA Cancer J. Clin.
2007, 57, 43.
2. Crawford, E. D. Urology 2003, 62, 3.
3. Siegel, R.; DeSantis, C.; Virgo, K.; Stein, K.; Mariotto, A.; Smith, T.; Cooper, D.;
Gansler, T.; Lerro, C.; Fedewa, S.; Lin, C.; Leach, C.; Cannady, R. S.; Cho, H.;
Scoppa, S.; Hachey, M.; Kirch, R.; Jemal, A.; Ward, E. CA Cancer J. Clin. 2012, 62,
220.
4. Statistics, C. C. S. s. S. C. o. C.; Canadian Cancer Society: Toronto, ON, 2013, p 65.
5. Riboli, E.; Slimani, N.; Kaaks, R. IARC Sci. Publ. 1996, 23.
6. Murillo, G.; Mehta, R. G. Nutr. Cancer 2001, 41, 17.
LNCaP cells were added to 24-well plates (Corning Cell-Bind) at
a density of 2 ꢁ 105 cells/well in steroid-free medium and incu-
bated for 24 h. Cells were then exposed for 24 h to various concen-
trations of caffeic acid and its derivatives in the presence or
absence of 0.3 nM DHT. Culture medium was removed and aliquots
were prepared for the determination of PSA concentrations using
an enzyme-linked immunosorbent assay kit (EL10050; Anogen,
Mississauga, ON).
7. Van’t Veer, P.; Jansen, M. C.; Klerk, M.; Kok, F. J. Public Health Nutr. 2000, 3, 103.
8. Kristal, A. R.; Lampe, J. W. Nutr. Cancer 2002, 42, 1.
9. Cohen, J. H.; Kristal, A. R.; Stanford, J. L. J. Natl. Cancer Inst. 2000, 92, 61.
10. Verhoeven, D. T.; Goldbohm, R. A.; van Poppel, G.; Verhagen, H.; van den
Brandt, P. A. Cancer Epidemiol. Biomarkers Prev. 1996, 5, 733.
11. Grunberger, D.; Banerjee, R.; Eisinger, K.; Oltz, E. M.; Efros, L.; Caldwell, M.;
Estevez, V.; Nakanishi, K. Experientia 1988, 44, 230.
12. Gomes, C. A.; da Cruz, T. G.; Andrade, J. L.; Milhazes, N.; Borges, F.; Marques, M.
P. J. Med. Chem. 2003, 46, 5395.
13. Hiipakka, R. A.; Zhang, H. Z.; Dai, W.; Dai, Q.; Liao, S. Biochem. Pharmacol. 2002,
63, 1165.
14. Natarajan, K.; Singh, S.; Burke, T. R., Jr.; Grunberger, D.; Aggarwal, B. B. Proc.
Natl. Acad. Sci. U.S.A. 1996, 93, 9090.
15. Lin, H. P.; Jiang, S. S.; Chuu, C. P. PloS One 2012, 7, e31286.
16. Touaibia, M.; Jean-Francois, J.; Doiron, J. Mini-Rev Med. Chem. 2011, 11, 695.
17. Bruckner, R. Advanced Organic Chemistry: Reaction Mechanisms; Harcourt/
Academic Press: San Diego, CA, USA, 2002.
5.6. Radical scavenging activity
The radical scavenging activity of caffeic acid derivatives was
measured as previously described using a 2,2-diphenyl-1-pic-
rylhydrazyl-(DPPH)-based radical generating system27,28 with
slight modifications. This assay measures the capacity of com-
pounds to reduce free radicals through an electron transfer (ET)
mechanism.29 Particular care was taken in the preparation of the
assay control (DPPH reagent with ethanol as diluent without test
compounds). Controls within an optical density range of 0.350–
0.360 at 520 nm were deemed acceptable to avoid variations in

J. T. Sanderson et al. / Bioorg. Med. Chem. 21 (2013) 7182–7193
7193
18. Boudreau, L. H.; Maillet, J.; LeBlanc, L. M.; Jean-Francois, J.; Touaibia, M.;
Flamand, N.; Surette, M. E. PloS One 2012, 7, e31833.
24. Masiello, D.; Cheng, S.; Bubley, G. J.; Lu, M. L.; Balk, S. P. J. Biol. Chem. 2002, 277,
26321.
19. Touaibia, M.; Guay, M. J. Chem. Educ. 2011, 88, 473.
20. Watabe, M.; Hishikawa, K.; Takayanagi, A.; Shimizu, N.; Nakaki, T. J. Biol. Chem.
2004, 279, 6017.
25. Gocer, H.; Gulcin, I. Int. J. Food Sci. Nutr. 2011, 62, 821.
26. LeBlanc, L. M.; Pare, A. F.; Jean-Francois, J.; Hebert, M. J.; Surette, M. E.;
Touaibia, M. Molecules 2012, 17, 14637.
21. Jin, U. H.; Song, K. H.; Motomura, M.; Suzuki, I.; Gu, Y. H.; Kang, Y. J.; Moon, T.
C.; Kim, C. H. Mol. Cell. Biochem. 2008, 310, 43.
27. Hatano, T.; Edamatsu, R.; Mori, A.; Fujita, Y.; Yasuhara, E.; Yoshida, T.; Okuda, T.
Chem. Pharm. Bull. 1989, 37, 2016.
22. Chen, M. J.; Chang, W. H.; Lin, C. C.; Liu, C. Y.; Wang, T. E.; Chu, C. H.; Shih, S. C.;
Chen, Y. J. Pancreatology 2008, 8, 566.
28. Lin, C. F.; Chang, T. C.; Chiang, C. C.; Tsai, H. J.; Hsu, L. Y. Chem. Pharm. Bull. 2005,
53, 1402.
23. Lee, Y. J.; Kuo, H. C.; Chu, C. Y.; Wang, C. J.; Lin, W. C.; Tseng, T. H. Biochem.
Pharmacol. 2003, 66, 2281.
29. Prior, R. L.; Wu, X.; Schaich, K. J. Agric. Food Chem. 2005, 53, 4290.