Tyrenes: Synthesis of new antiproliferative compounds with an extended conjugation

Article abstract of DOI:10.1016/j.bmc.2004.03.043

A series of substituted styryl-acrylonitriles was designed and synthesized. The new compounds, called tyrenes, were tested for the ability to inhibit acute lymphocytic leukemia (ALL) cancer cell growth, as well as on their toxicity to normal bone marrow (NBM) cells. The results showed that 3,4-dihydroxystyryl- acrylonitriles, in particular CR-4, revealed great potency as antitumor agents, and also exhibited low toxicity to normal cells. The effectiveness of these compounds with extended conjugation may be due to their possible functioning as reactive Michael acceptors.

Full text of DOI:10.1016/j.bmc.2004.03.043

Bioorganic & Medicinal Chemistry 12 (2004) 3019–3026
Tyrenes: synthesis of new antiproliferative compounds with an
extended conjugation
Peter Demin, Olga Rounova, Thomas Grunberger, Lorand Cimpean, Nigel Sharfe
and Chaim M. Roifman*
Division of Immunology and Allergy, Infection, Immunity, Injury and Repair Program, Research Institute, and The Hospital for Sick
Children and University of Toronto, Toronto, Canada M5G 1X8
Received 15 August 2003; accepted 12 March 2004
Available online 30 April 2004
Abstract—A series of substituted styryl-acrylonitriles was designed and synthesized. The new compounds, called tyrenes, were tested
for the ability to inhibit acute lymphocytic leukemia (ALL) cancer cell growth, as well as on their toxicity to normal bone marrow
(NBM) cells. The results showed that 3,4-dihydroxystyryl-acrylonitriles, in particular CR-4, revealed great potency as antitumor
agents, and also exhibited low toxicity to normal cells. The effectiveness of these compounds with extended conjugation may be due
to their possible functioning as reactive Michael acceptors.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
occurring potent PTK inhibitor was found to be erb-
statin.2 Subsequent studies have demonstrated that the
inhibitory ability of erbstatin against PTK is connected
to its ability to compete with tyrosine as a substrate to
phosphorylation, because of its structural similarity to
tyrosine,3 that is because of the presence of two phenolic
hydroxyls in erbstatin. It is possible that erbstatin
resembles the specific conformation of tyrosine moieties
within the peptide.3 Some synthetic erbstatin analogs
with different arrangement of hydroxyls in the benzene
ring were synthesized. Among those, 2,3- and 3,4-dihy-
droxy analogs showed potent inhibitory activity com-
parable to erbstatin.4
Protein kinases and phosphatases are involved in certain
signaling pathways, which regulate various cell func-
tions such as proliferation, differentiation, and apop-
tosis. Protein kinases (PK) catalyze the phosphorylation
of the hydroxyl-containing amino acids, such as tyro-
sine, serine, and threonine.1 PK activity appears to be
the predominant way of passing proliferative signaling
through the cell membrane and into cells. Therefore, an
abnormal activity of PK may cause an extensive cell
proliferation, which is the source of oncogenic diseases.
In particular, phosphorylation of tyrosine catalyzed by
protein tyrosine kinase (PTK) is known as a key process
in unregulated cell growth. Therefore the synthesis of
compounds, which inactivate PTK activity becomes an
important tool for development of perspective anti-
cancer drugs.
In the past two decades, significant progress had been
achieved in design and synthesis of various low molec-
ular compounds based on the structures of erbstatin and
related compounds. In particular, Knoevenagel con-
densation of various hydroxy-substituted benzaldehydes
with malononitrile and its derivatives led to a number of
compounds called tyrphostins that possessed a good
inhibitory activity to various PTKs.5;6 The common
feature of erbstatin and synthetic tyrphostins is the
presence of hydroxystyryl moiety, which mimics tyro-
sine and is apparently responsible for their biological
activity. An enhanced acidity of the phenol or catechol
groups caused by presence of the electrophilic acrylo-
nitrile may promote binding of such molecules to the
One of the existing approaches to block the PTK
activity is the use of low molecular competitive inhibi-
tors of PTK, which would mimic the parent tyrosine
structure. One of the structurally similar naturally
Keywords: Tyrene; Antiproliferative activity; Michael acceptor.
*Corresponding author. Tel.: +1-416-813-8623; fax: +1-416-813-8624;
e-mail: croifman@sickkids.ca
0968-0896/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.03.043

3020
P. Demin et al. / Bioorg. Med. Chem. 12 (2004) 3019–3026
substrate subsite of the PTK domain, making them
effective antiproliferative agents. Indeed blocking the
phenol groups or saturating the double bond resulted in
loss of antiproliferative activity.5
nitro-derivatives) led to the decomposition products
only. To obtain the corresponding adducts, b-alanine
was employed as a milder catalyst, and heating for
several hours was required (Scheme 1).
In the past few years, however, a number of tyrphostin-
like compounds had been designed, where the phenolic
hydroxyl group is absent7 or blocked;8 nevertheless
those compounds retained PTK inhibiting activities. It
should be noticed that all these compounds are all Mi-
chael acceptors in a broader sense, that is they have
olefinic groups conjugated with electron-withdrawing
substituents, having the structures Z0–CH@CH–Z.9
Such functionalities are subject to attack by a suitable
nucleophilic site on a corresponding protein or DNA,
for example, by cysteinyl thiol groups on protein (for
reviews on application of Michael acceptors for anti-
cancer drug design see Refs. 10 and 11). In this study, we
describe a new series of antiproliferative compounds,
that we called ‘tyrenes’, which possess the two important
functions: activated phenolic hydroxyls, combined with
an extended conjugation that may serve as a reactive
Michael acceptor.
In all cases, the single E,E-adducts were obtained, which
was confirmed by HPLC (single peaks were detected)
and H NMR spectroscopy (mutual coupling constant
was 14–15 Hz for all olefinic protons, and the proton at
the a-position to CN group was at the vicinity of 8 ppm,
which is in agreement with the spectra of another
a; b; c; d-unsaturated nitriles of trans-configuration14).
1
To obtain this series of compounds, various hydroxy-
substituted cinnamaldehydes were required. Unlike
homologous benzaldehydes, only few of them, such as
sinapyl- and coniferyl aldehydes are commercially
available. Therefore, a reliable synthetic pathway to
such cinnamaldehydes, in particular, to 3,4-dihydroxy-
cinnamaldehyde (caffeoyl aldehyde) had to be
developed.
It was known, that Wittig condensation of hydroxy-
substituted benzaldehydes with resonance-stabilized
ylide formylmethylenetriphenylphosphorane is certainly
not a reliable process for obtaining 3,4-substituted
cinnamaldehydes, as a mixture of products with different
degree of unsaturation was obtained.15 Due to this, for
obtaining caffeoyl aldehyde 1 and its 3,4-dimethoxy
analog 2 we have chosen a straightforward approach,
involving known transformations of commercially
available substituted cinnamic acids. Following the
described procedures,16;17 the alcohols 3 and 4 were
obtained and oxidized with activated manganese dioxide
into corresponding aldehydes 5 and 2. Deprotection of
the aldehyde 5 led to caffeoyl aldehyde 1 (Scheme 2).
2. Results and discussion
2.1. Design and synthesis
2-Benzylaminocarbonyl-3-arylacrylonitriles were previ-
ously reported as PTK blocking agents.12 The activated
styrylphenolic moiety in those compounds might play a
crucial role as a competitive agent to tyrosine phos-
phorylation. We hypothesized that an extra-annular
double bond added to the existing conjugated system,
might serve as an effective acceptor in a conjugate
Michael addition.
The synthons with activated a-methylene, N-benzyl-2-
cyanoacetamides 6, were obtained by a direct coupling
of various benzylamines 7 with methyl cyanoacetate
(Scheme 3). Cleavage of the vic-methoxy groups in the
Based on this hypothesis, we have designed and syn-
thesized a series of hydroxystyryl-acrylonitriles, shown
on Figure 1. Previously, some compounds with similar
structural features were obtained and used as com-
ponents for nonlinear optical and photoreactive
materials.13
O
CHO
N
R
R'
H
CN
The key step in obtaining these compounds is Knoeve-
nagel condensation of a; b-unsaturated aryl substituted
aldehydes with compounds having an activated methyl-
ene group. We have found that hydroxyl-substituted
cinnamaldehydes showed themselves as very reactive
agents, so the reaction proceeded smoothly in ethanol in
the presence of equimolar amount of piperidine (1 equiv
per hydroxyl) within 0.5 h at room temperature. How-
ever, applying these conditions to the condensation of
protected, or nonphenolic cinnamaldehydes (such as
unsubstituted cinnamaldehyde and its dimethoxy- or
A or B
O
N
R
R'
H
CN
CR
Scheme 1. Reagents and conditions: (A) piperidine, EtOH, 0.5 h, 20 ꢁC
(when R ¼ OH); (B) b-alanine, EtOH, 80 ꢁC, 1–4 h (when R¼OH).
XO
XO
YO
YO
CHO
OH
a,b (for aldehyde 1)
O
1
2
5
Z = H
3 X = BDMS
4 X = Me
(HO)n
N
Z = Me
Ar
Ph
R'
H
Z = BDMS
CN
X
R
Scheme 2. Reagents and conditions: (a) activated MnO2, CH2Cl2; (b)
n-Bu4NF, CHCl3.
Figure 1. Basic structure of tyrenes.

P. Demin et al. / Bioorg. Med. Chem. 12 (2004) 3019–3026
3021
O
O
OH
OH
b
a
N
H
H2N
N
H
R'
R'
CN
X
CN
X
7a, R' = H, X = CH
7b, R' = 3-F, X = CH
7c, R' = 3-CF3, X = CH
7d, R' = H, X = N
6a, R' = H, X = CH
6f, R' = 3,4-diOH
6b, R' = 3-F, X = CH
6c, R' = 3-CF3, X = CH
6d, R' = H, X = N
7e, R' = 3,4-diOMe, X = CH
6e, R' = 3,4-diOMe, X = CH
Scheme 3. Reagents and conditions: (a) NBCCH2COOMe neat, 2 h, 20 ꢁC; (b) 9e, BBr3, CH2Cl2, )78 ꢁC fi 20 ꢁC.
3,4-dimethoxybenzylamide yielding 3,4-dihydroxybenzyl-
amide was achieved with BBr3.
was achieved on an Agilent SB-CN column in the
reverse-phase mode (see Section 5). The full scan Q1
mass spectra of all the peaks recorded in LC–MS mode
showed the molecular ions at m=z 383 (for CR-4-SA)
and at 445 (for CR-4-SB), as well as the prominent
daughter ion m=z 321 ([M ) Et]þ) and ([M ) Bz]þ) for the
two derivatives. To prove that the ion with m=z 321 is a
daughter ion and not the molecular ion of CR-4 itself,
the molecular ions of CR-4 (m=z 321) and the derivatives
CR-4-SA and CR-4-SB (m=z 383 and 445) were used for
the daughter ion scan. Both molecular ions of the
compounds CR-4-SA and CR-4-SB gave a number of
daughter ions, whereas the parent molecular ion of CR-
4 gave almost no fragmentation in the same ionization
conditions.
CR-54, the monosaturated analog of CR-4 was obtained
by treatment of CR-4 with sodium borohydride in
methanol as described18 (Scheme 4). In our case, a; b-
double bond in CR-4 was selectively reduced, leading to
2-cyano-5-(3,4-dihydroxyphenyl)-4E-pentenoic
benzylamide (CR-54).
acid
To prove the hypothesis that tyrenes are Michael
acceptors, CR-4 was treated with simple nucleophiles,
mimicking cysteine residues, such as mercaptans
(Scheme 5). Indeed, CR-4 smoothly and quantitatively
reacted with ethanethiol in DMSO in the presence of
piperidine giving a less polar product with a remarkably
different UV spectrum (kmax 205 and 280 nm). Benzyl
mercaptan appeared to be more reactive in this process,
as no catalyst was required. The products were unstable
in solutions, and easily gave back the parent CR-4 when
they were treated with a base. The NMR spectrum of
the mixture of the ethanethiol derivatives (CR-4-SA)
showed the absence of the signal at 8 ppm indicating
that the system of the two double bonds was partially
saturated. The appearance of the complex multiplet at
6.6–6.8 ppm suggested that the obtained adduct is in fact
a diastereomeric mixture of the products of 1,2- and 1,4-
addition of ethanethiol. Partial separation of the four
possible regio/stereoisomers of CR-4-SA and CR-4-SB
3. Biological activity
The synthesized compounds were tested for their ability
to inhibit abnormal cell proliferation. At the same time,
the compounds were tested for their toxicity to normally
growing cells and the most promising ones showed no
deleterious effect on normal cell growth. In those
experiments, acute lymphoblastic leukemia (ALL) cells,
and normal human bone marrow (NBM) cells were
chosen as models. In a typical experiment, leukemic and
normal bone marrow cells were exposed to various
O
O
HO
HO
a,b
N
N
H
H
CN
CN
HO
HO
CR-4
CR-54
Scheme 4. Reagents and conditions: (a) NaBH4, MeOH, 1 h, 20 ꢁC; (b) HPLC, Nova-Pak C18 19.0 · 300 mm, gradient MeCN–H2O,
20:80 fi MeCN–H2O, 80:20.
SR
O
HO
HO
HO
HO
N
H
O
CN
HO
HO
a or b
N
H
+
SR
O
CN
N
H
CR-4
CN
CR-4-SA, R = Et
CR-4-SB, R = Bz
Scheme 5. Reagents and conditions: (a) EtSH, DMSO, piperidine, 0.5 h, 20 ꢁC; (b) BzSH, DMSO, 0.5 h, 20 ꢁC.

3022
P. Demin et al. / Bioorg. Med. Chem. 12 (2004) 3019–3026
styryl-acrylonitriles and the number of colonies counted
after 9 and 14 days, respectively. The most effective
compounds were not only able to block ALL cells
growth, but also permitted the unaffected growth of
NBM cells. The IC50 values for the compounds against
ALL and NBM cells are listed in Table 1.
values were in low-micromolar range. One of the com-
pounds of this sub-series, namely CR-11, showed 10-
fold decrease in activity in comparison with CR-4 (IC50
1.4 lM). When the only hydroxyl in the Ph-ring was less
sterically hindered (CR-56, CR-57), the compound
became significantly more active (IC50 0.32–0.34 lmol).
On the other hand, CR-11 showed proportionally less
toxicity to normal cells, with TD of 20.
The compounds containing free phenol or catechol
groups were effective against ALL cells; some of them
eliminated ALL cell growth at sub-micromolar doses.
The activity of the compounds where the phenolic
hydroxyls were absent or blocked, was significantly
reduced, however, this fact might also be related to their
high level of hydrophobilicy. The most active com-
pounds toward ALL cells were shown to be the ones
having the catechol group. In particular, the compounds
with the 3,4-dihydroxystyryl moiety showed an excep-
tional activity. One of them, CR-4, significantly inhib-
ited Philadelphia positive ALL cells growth at the level
as low as 0.12 lM (Table 1; the detailed report on CR-4
inhibitory ability toward various cancer cell lines see
Ref. 19). When one of the double bonds in CR-4 was
selectively saturated, the inhibitory activity of such
compound (CR-54) was significantly diminished, how-
ever, not blocked completely (compare to Ref. 5). The
IC50 to NBM growth, exhibited by CR-4, was as high as
9 lM, which makes the compound a possible therapeu-
tical candidate.
Some compounds of another sub-series, where the Ph
ring remained unsubstituted, and Ar ring was a catechol
(CR-19, CR-28), also revealed a significant activity to
ALL cells inhibition. However, those compounds were
particularly toxic to the NBM cells, with TD of 12–16.
Those results suggested that the presence of activated
hydroxyls in the Ph-ring is highly desirable, yet is not a
crucial factor in the anti-cancer ability. We have con-
cluded that the extended conjugation adjacent to elec-
tron-withdrawing substituents in the side chain played
an independent role in making the compounds
remarkably strong cancer cell growth inhibitors. It is
known that substituents at the olefinic function gener-
ally decrease reactivity; therefore tyrenes with totally
exposed c; d-double bond appear to be very reactive
Michael acceptors. It is therefore possible that these
compounds might be more effective than compounds
having only one olefinic block partially substituted with
the cyano groups. In support of this hypothesis, AG-490
demonstrated inhibitory activity to ALL at 5–25 lM
level20 while CR-4 was effective in nM concentration.19
Other compounds of this sub-series with different sub-
stitutions at Ar-ring (CR-21, CR-43, CR-66, CR-72, CR-
74) showed similar, yet lower level of activity toward
ALL cells. Noteworthy, the presence of hydrophilic
amide moiety in CR-21 led to decreasing of ALL
inhibitory activity, suggesting that the hydrophobic
amide group in CR-4 might be important to this func-
tion. An attempt to prepare more hydrophobic halogen-
substituted compounds (CR-66, CR-72) was not
successful, as the compounds demonstrated lower
therapeutic differential (TD) to inhibiting cell growth; in
particular, a significant toxicity to NBM cells was
observed (Table 1).
4. Conclusion
In this report, we have synthesized a new series of
tyrenes––compounds, containing an extended hydroxy-
styryl-acrylonitrile conjugation system. The obtained
compounds were evaluated on their ability to inhibit
ALL cells growth, as well as their effect on the devel-
opment of normal cells. 3,4-Dihydroxy-tyrenes showed
a remarkable activity on ALL cells (IC50 ¼ 0:1–0:2 lM).
O-Methyl derivatives were significantly less active
(IC50 ¼ 0:5–1:0 lM), however, they revealed much lower
When only one hydroxyl in the Ph-ring was exposed, the
activity on the cancer cells was lowered, so the IC50
Table 1. Biological activity of tyrenes
Code
R
R0
IC50, ALL (lM)
IC50, NBM (lM)
CR-1
CR-2
CR-4
H
H
10 0.2
2.28 0.06
0.12 0.04
1.4 0.2
7
29
9
2
1
2
2
3,4-(OMe)2
3,4-(OH)2
4-OH, 3,5-(OMe)2
H
H
H
CR-11
CR-19
CR-21
CR-28
CR-43
CR-54
CR-56
CR-57
CR-66
CR-72
CR-74
3,4-(OH)2
3,4-(OH)2
3,4-(OH)2
3,4-(OH)2
3,4-(OMe)2
28
0.09 0.03
1.5 0.36
0.29 0.07
0.21 0.05
0.6 0.2
2.5 0.3
1
3,4-(OH)2
4-NO2
3,4-(OH)2
18.6
43.8 0.3
6.2 0.8
29.4 0.6
Partially saturated CR-4
4-OH, 3-OMe
4-OH, 3-OMe
3,4-(OH)2
H
0.34 0.06
0.32 0.09
0.33 0.06
0.16 0.02
0.2 0.02
9.8 3
3,4-(OH)2
3-CF3
3-F
19.2 1.7
2.8 0.2
3.3 0.3
16 1.1
3,4-(OH)2
3,4-(OH)2
Py

P. Demin et al. / Bioorg. Med. Chem. 12 (2004) 3019–3026
3023
toxicity toward NBM cells (IC50 > 10–15 lm), and also
promoted normal cells growth. These data suggest that
the extended conjugation, acting as an active Michael
acceptor, promote PTK inhibitory activity of this group
of compounds. Among synthesized compounds, CR-4
was the most promising candidate for development of
antitumor agents. We now continue further chemical
modifications and pharmacological evaluations of these
compounds.
from Aldrich and were used as received. Solvents were
purchased from Caledon (Canada).
The key Knoevenagel condensation was carried out in
ethanol using piperidine (method A) or b-alanine
(method B) as catalysts.
5.2. Synthesis of key intermediates
1
5.2.1. Substituted cinnamaldehydes. H NMR spectra of
cyanoacetamides are listed in Table 2.
5. Experimental
5.1. General procedures
5.2.2. 3,4-Dihydroxycinnamaldehyde (caffeoyl aldehyde)
(1). 3,4-Bis-(t-butyldimethylsilyloxy)cinnamyl alcohol
(3) (7.88 g, 20 mmol) was dissolved in 1000 mL CH2Cl2,
17.2 g activated MnO2 (200 mmol) were added and the
mixture was thoroughly stirred for 24 h at 20 ꢁC. MnO2
was filtered off and the filtrate was taken to dryness. To
the obtained 3,4-diBDMS caffeoyl aldehyde (5) 250 mL
of CHCl3 was added followed by addition of n-Bu4NF
monohydrate (11.6 g, 40 mmol). The mixture was stirred
at 20 ꢁC for 30 min and worked up with 300 mL of 5%
HCl. Chloroform layer was separated, washed with
H2O, dried with Na2SO4, and taken to dryness. The
residue was purified on a silica gel column, eluent
MeOH–CHCl3, 1:4 + 1% AcOH. The solvents were
evaporated and the crystalline residue was washed with
CHCl3 to give caffeoyl aldehyde (1) (1.77 g, 54%).
Melting point 208–210 ꢁC.
1H NMR spectra were obtained on a Varian Unity Plus
spectrometer at 500 MHz in acetone-d6 with Me4Si as an
internal standard (d ¼ 0). Electrospray mass spectra of
the compounds CR-4-SA and CR-4-SB were acquired
on an API 4000 LC/MS/MS mass spectrometer (PE
Sciex, Thornhill, Canada). UV–vis spectra were
recorded in MeCN containing 0.1% H3PO4 on a Beck-
man DU 640 spectrophotometer. High-resolution mass
spectroscopic analyses (HRMS) and elemental analyses
were performed at the Chemistry Department of the
University of Toronto, Canada. Elemental analyses
were within 0.4% of the theoretical value unless
otherwise noted. Melting points were recorded on a
Barnstead Mel-Temp capillary melting point apparatus
and are uncorrected. The final compounds were
re-crystallized from PEG400-H2O. Thin layer chromato-
graphy was performed with UV-254 aluminum-backed
TLC sheets of 0.25 mm thickness (Kieselgel 60 F254
,
5.2.3. 3,4-Dimethoxycinnamaldehyde (2). Obtained in a
similar manner by oxidizing of 970 mg (5 mmol) 3,4-
dimethoxycinnamalcohol (4) with 1720 mg (20 mmol) of
activated MnO2 in 400 mL CH2Cl2. The product was
re-crystallized from ethyl acetate–heptane to give 865 mg
(90%) of the aldehyde (2). Melting point 91 ꢁC.
Merck). HPLC chromatograms were obtained on a
model 600 liquid chromatograph (Waters, USA) with a
model 996 PDA detector. The column was 5 lm Zorbax
SB-CN 4.6 · 250 mm (Agilent Technologies, USA).
Gradient separations of the final compounds were con-
ducted using the following buffers: (A) 0.1% H3PO4 in
H2O; (B) MeCN–H2O, 4:1 + 0.1% H3PO4. The following
gradient system was employed: buffer A for 10 min,
followed by linearly increasing the percentage of the
buffer B from 20% to 90% during 20 min, flow rate
1.0 mL/min. Separations of the derivatives CR-4-SA and
CR-4-SB were achieved using the system MeCN–H2O,
50:50, isocratic mode, flow rate 1.0 mL/min. Vacuum
distillations were done using Kugelrohr apparatus
(Aldrich) at stated temperatures of an oven.
1
5.2.4. Substituted cyanoacetamides. H NMR spectra of
cyanoacetamides are listed in Table 3.
5.2.5. General method of synthesis of compounds 6a–e. In
a typical experiment, equimolar amounts of methyl
cyanoacetate and corresponding amine were mixed and
kept at room temperature for 12 h. The formed white
solid was re-crystallized from ethanol or ethanol–water.
The average yield of cyanoacetamides was 60–80%.
Compound 6a: mp 128 ꢁC; 6b: mp 106 ꢁC; 6c: mp
118 ꢁC; 6d: mp 130 ꢁC; 6e: mp 142 ꢁC.
Unsubstituted cinnamaldehyde and its 4-hydroxy-3-
methoxy-, 3,5-dimethoxy-4-hydroxy-, and 4-nitro-
derivatives, as well as all other reagents were purchased
Table 2. 1H NMR Spectra of substituted cinnamaldehydes (d, ppm; J, Hz)
No
Ha
Hb
CHO
H2
H3
––
H4
––
H5
H6
Other
––
1
6.54 (dd,
J ¼ 7:7, 15.8)
6.69 (dd,
7.52 (d,
J ¼ 15:8)
7.60 (d,
9.62 (d,
J ¼ 7:7)
9.66 (d,
J ¼ 7:6)
9.64 (d,
J ¼ 7:7)
7.21 (br s)
6.91 (d,
J ¼ 8:2)
7.05 (d,
J ¼ 8:3)
7.27 (m)
7.12 (br d,
J ¼ 8:2)
2
5
7.37 (d, J ¼ 1:5) ––
7.27 (m) ––
––
––
7.28 (dd,
J ¼ 1:5, 8.3)
7.00 (dd,
3.90
J ¼ 16:0)
J ¼ 16:0)
7.59 (d,
6.60 (dd,
J ¼ 7:7, 15.9)
0.25, 1.00
J ¼ 15:9)
J ¼ 1:5, 8.2)

3024
P. Demin et al. / Bioorg. Med. Chem. 12 (2004) 3019–3026
Table 3. 1H NMR Spectra of substituted cyanoacetamides (d, ppm; J, Hz)
No
NBCCH2
CH2Ar
H2
H3
H4
H5
H6
Other
6a
6b
6c
6d
6e
3.62 (s)
3.14 (s)
3.14 (s)
3.18 (s)
3.82 (s)
4.49 (s)
4.44 (s)
4.53 (s)
4.49 (s)
4.38 (s)
7.26 (m)
7.11 (m)
7.56–7.68 (m) ––
7.32 (m)
––
7.32 (m)
7.02 (m)
7.32 (m)
7.37 (m)
7.26 (m)
7.16 (m)
––
––
7.56–7.68 (m) 7.56–7.68 (m) 7.56–7.68 (m) ––
7.32 (m)
6.97 (d,
J ¼ 2:1)
6.79 (d,
J ¼ 2:0)
8.50 (m)
––
––
8.50 (m)
6.92 (d,
J ¼ 8:1)
6.75 (d,
J ¼ 8:1)
7.32 (m)
6.87 (dd,
––
2.87
––
––
J ¼ 2:1, 8.1)
6.63 (dd,
6f
3.60 (s)
4.25 (s)
––
––
J ¼ 2:0, 8.1)
5.2.6. 2-Cyano-N-(3,4-dihydroxybenzyl)acetamide 6f. To
2-cyano-N-(3,4-dimethoxybenzyl)acetamide 6e (2.68 g,
11.45 mmol) in 200 mL of CH2Cl2, boron tribromide
was added dropwise at )10 ꢁC (8.75 g, 35 mmol).
5.2.8. General method A. To a solution of 0.1 mmol of
hydroxyl-substituted cinnamaldehyde 1 and 0.1 mmol
of amide 6 in 3–4 mL of ethanol, an equimolar amount
of piperidine was added. The deep red solution was
stirred at 20 ꢁC for 0.5–1.0 h until the starting material
disappeared. 1 N HCl (0.2 mL) was added followed by
addition of 10 mL H2O, and the mixture was kept at
0 ꢁC for 2 h. The precipitated powder was washed with
H2O, re-crystallized from MeCN–H2O, and dried in a
desiccator over NaOH. The average yield of the CR
compounds was 50–70%.
After 0.5 h the reaction was brought to room tempera-
ture and stirred for an additional 1 h. The reaction was
cooled to 0 ꢁC, 200 mL of water was carefully added,
and the organic layer was separated. The aqueous phase
was saturated with NaCl and extracted with 3 · 100 mL
of ethyl acetate. The combined organic phase was dried
with Na2SO4, and taken to dryness. The solidified resi-
due was dissolved in methanol and passed through a
column with silica gel, eluent 5% MeOH in EtOAc. The
solvents were evaporated, and the residue was dissolved
in water and lyophilized to give an off-white powder.
Yield 1.24 g (53%). Melting point 153–154 ꢁC.
5.2.9. General method B. To a solution of 0.1 mmol of
cinnamaldehyde, or its 3,5-dimethoxy-4-hydroxy deriv-
ative, or its 4-nitro derivative, or cinnamaldehyde 2 and
0.1 mmol of amide 6 in 3–4 mL of ethanol, a few crystals
of b-alanine were added. The mixture was stirred at
80 ꢁC for 2.0–4.0 h until the starting material disap-
peared. H2O (10 mL) was added, and the mixture was
kept at 0 ꢁC for 2 h. The precipitated powder was
washed with H2O, re-crystallized from MeCN–H2O,
5.2.7. Synthesis and transformations of tyrenes. 1H NMR
Spectra of tyrenes are listed in Table 4. Physical data of
tyrenes are listed in Table 5.
Table 4. 1H NMR Spectra of tyrenes (d, ppm; J, Hz)
Code
H2
H3
H4
H5
H6
PhCH
PhC@CH
CHCN
Other
CR-1
7.24–7.51 7.24–7.51 7.24–
7.24–7.51 7.24–7.51 (m) 7.24–7.51 7.24–7.51 (m)
(m)
8.05 (d,
J ¼ 11:7)
8.05 (d,
4.55; 7.24–7.51
(m)
(m)
––
7.51 (m) (m)
CR-2
7.39 (d,
J ¼ 2:0)
7.19 (d,
J ¼ 2:1)
––
––
––
7.07 (d,
J ¼ 8:3)
6.87 (d,
J ¼ 8:2)
––
7.30 (dd,
J ¼ 2:0, 8.3)
7.05 (dd,
7.37 (d,
J ¼ 15:3)
7.34 (d,
J ¼ 15:2)
7.37 (d,
J ¼ 15:1)
7.48 (d,
J ¼ 15:7)
7.34 (d,
J ¼ 15:3)
7.64 (d,
J ¼ 15:8)
7.34 (d,
J ¼ 15:2)
6.44 (d,
J ¼ 15:7)
7.35 (d,
J ¼ 15:1)
7.39 (d,
J ¼ 15:3)
7.36 (d,
J ¼ 15:1)
7.35 (d,
J ¼ 15:3)
7.38 (d,
J ¼ 15:3)
7.20 (dd,
J ¼ 11:6, 15.3)
7.02 (dd,
3.90; 3.91; 4.57;
7.26–7.41
J ¼ 11:6)
7.99 (d,
CR-4
––
4.54; 7.24;
7.28–7.36
J ¼ 2:1, 8.2)
7.07 (br s)
J ¼ 11:2, 15.2)
7.16 (dd,
J ¼ 11:2)
7.98 (d,
CR-11
CR-19
CR-21
CR-28
CR-43
CR-54
CR-56
CR-57
CR-66
CR-72
CR-74
7.07 (br s, ––
1H)
2.81; 3.89; 4.39;
6.68; 6.76; 6.86
2.82; 4.40; 6.70
6.76; 6.86
J ¼ 11:7, 15.1)
7.29 (dd,
J ¼ 11:7)
8.04 (d,
7.72 (m)
7.46 (m)
7.46 (m) 7.46 (m)
7.72 (m)
J ¼ 11:7, 15.7)
7.05 (dd,
J ¼ 11:7)
8.00 (d,
7.22 (d,
J ¼ 2:1)
8.03 (d,
J ¼ 8:8)
7.21 (d,
J ¼ 2:1)
6.93 (d,
J ¼ 2:1)
7.37 (d,
J ¼ 2:1)
7.37 (d,
J ¼ 2:1)
7.22 (d,
J ¼ 2:2)
7.22 (d,
J ¼ 2:2)
7.23 (d,
J ¼ 2:2)
––
––
––
––
––
––
––
––
––
––
6.90 (d,
J ¼ 8:6)
8.32 (d,
J ¼ 8:8)
6.89 (d,
J ¼ 8:3)
6.77 (d,
J ¼ 8:1)
6.91 (d,
J ¼ 8:2)
6.91 (d,
J ¼ 8:2)
6.90 (d,
J ¼ 8:2)
6.90 (d,
J ¼ 8:2)
6.91 (d,
J ¼ 8:2)
7.09 (dd,
J ¼ 2:1, 8.6)
8.03 (d,
4.38; 6.68; 6.76;
6.85
J ¼ 11:8, 15.3)
7.46 (dd,
J ¼ 11:8)
8.10 (d,
8.32 (d,
J ¼ 8:8)
––
4.37; 6.73; 6.81;
6.89
J ¼ 8:8)
J ¼ 11:7, 15.8)
7.05 (dd,
J ¼ 11:7)
8.00 (d,
7.07 (dd,
J ¼ 2:1, 8.3)
6.71 (dd,
4.46; 3.79; 3.80;
6.90; 7.01
J ¼ 11:7, 15.2)
6.01 (dt,
J ¼ 11:7)
2.79 (m)
––
––
––
––
––
––
3.15; 4.44;
7.22–7.30
J ¼ 2:1, 8.1)
7.20 (dd,
J ¼ 7:3, 15.7)
7.14 (dd,
8.02 (d,
J ¼ 11:6)
8.00 (d,
3.92; 4.55; 7.25;
7.31–7.42
J ¼ 2:1, 8.2)
7.20 (dd,
J ¼ 11:6, 15.1)
7.13 (dd,
3.92; 4.38; 6.68;
6.76; 6.86
4.65;
J ¼ 2:1, 8.2)
7.08 (dd,
J ¼ 11:6, 15.3)
7.05 (dd,
J ¼ 11:6)
8.02 (d,
J ¼ 2:2, 8.2)
7.07 (dd,
J ¼ 11:6, 15.1)
7.05 (dd,
J ¼ 11:6)
8.01 (d,
7.60; 7.70
4.56; 7.02; 7.14;
7.20; 7.36
J ¼ 2:2, 8.2)
7.09 (dd,
J ¼ 11:7, 15.3)
7.07 (dd,
J ¼ 11:7)
8.04 (d,
4.59; 7.38;
8.51
J ¼ 2:2, 8.2)
J ¼ 11:7, 15.3)
J ¼ 11:7)

P. Demin et al. / Bioorg. Med. Chem. 12 (2004) 3019–3026
3025
Table 5. Physical data of tyrenes
Code
Mp (ꢁC)
kmax (nm) (emax
)
HRMS, m=z, found (calcd)
Microanalysis (C,H,N)
CR-1
CR-2
CR-4
195–196
179–180
245–246
247–248
198–199
261–263
244–245 (dec)
212–213
161–163
171–172
225–226
216–218
229–230
250–252
334 (31,694)
381 (35,570)
380 (32,328)
387 (29,702)
335 (28,149)
378 (28,137)
346 (31,851)
378 (37,288)
263 (9851)
288.1271 (288.1263)
348.1467 (348.1474)
320.1169 (320.1161)
396.1315 (396.1321)
320.1144 (320.1161)
352.1051 (352.1059)
365.1023 (365.1011)
380.1363 (380.1372)
322.1316 (322.1317)
334.1308 (334.1317)
366.1215 (366.1216)
388.1047 (388.1035)
338.1071 (338.1067)
321.1116 (321.1113)
C19H16N2O
C21H20N2O3
C19H16N2O3
C21H20N2O6
CR-11
CR-19
CR-21
CR-28
CR-43
CR-54
CR-56
CR-57
CR-66
CR-72
CR-74
C19H16N2O3
C19H16N2O5
C19H15N3O5Æ1/3H2O
C21H20N2O5ÆH2O
C19H18N2O3
C20H18N2O3
379 (33,680)
378 (31,274)
378 (32,584)
380 (34,654)
384 (29,859)
C20H18N2O5ÆH2O
C20H15F3N2O3
C19H15FN2O3
C18H15N3O3Æ1/4H2O
and dried in a desiccator over NaOH. The average yield
of the CR compounds was 70–75%.
HCl, and the product was extracted with ethyl acetate.
The solvent was evaporated and the residue was purified
on a reverse-phase HPLC. The products CR-4-SA and
CR-4-SB were obtained in a quantitative yield. CR-4-
SA: 1H NMR (d, ppm): 1.20, 1.26 (2 · t, 2 · 3H,
J ¼ 7:3 Hz, SCH2CH3), 2.30–2.40 (m, SCH2CH3), 4.38
(m, NHCH2Ph0), 6.60–6.82 (m, olefinic H + Ph), 7.22–
7.36 (m, Ph0). MS (ESI, full scan Q1, m=z, rel intensity,
%): 383.4 ([M+1]þ, 20), 321.2 ([M)SEt]þ, 100), 237.3
(17), 209.3 (7), 188.4 (32), 175.3 (12). HPLC (retention
times, min): 6.5, 8.2, 8.4. CR-4-SB: MS (ESI, full scan
Q1, m=z, rel intensity, %): 445.4 ([M+1]þ, 59), 321.5
([M)SBz]þ, 100). HPLC (retention times, min): 9.2, 16.8
(a wide peak), 17.4.
5.2.10. (E,E)-2-(Benzylaminocarbonyl)-3-(3,4-dihydroxy-
styryl)acrylonitrile (CR-4) (method A). To a solution of
16 mg of 3,4-dihydroxycinnamaldehyde 1 and 17 mg
of amide 6a in 3 mL of ethanol, an equimolar amount of
piperidine was added. The solution was stirred at 20 ꢁC
for 1.0 h. 1 N HCl (0.2 mL) was added followed by
addition of 10 mL H2O, and the mixture was kept at
0 ꢁC for 2 h. The precipitated powder was washed with
H2O, re-crystallized from MeCN–H2O, and dried in a
desiccator over NaOH. Yield 68%.
5.3. Biological assays
5.2.11. (E,E)-2-(3,4-Dihydroxybenzylaminocarbonyl)-3-
(3,5-dimethoxy-4-hydroxystyryl)acrylonitrile
(CR-11)
5.3.1. Leukemia colony assay. To permit the assessment
of individual colony formation, ALL cell lines and pri-
mary cells were cultured in semi-solid media containing
0.8% (v/v) methylcellulose at 5 · 103 cells/mL supple-
mented with 20% FCS. Primary ALL patient cells were
prepared as previously described.21 Briefly, nonadherent
cells and T cell were depleted from bone marrow or
peripheral blood samples and plated in Iscoves modified
Dulbecco medium supplemented with 10% FCS and
PHA-T cell conditioned medium (PHA-TCM). Cell
lines were grown in standard medium with FCS. All
assays were performed in duplicate dishes.
(method B). To a solution of 21 mg of 3,5-dimethoxy-
4-hydroxycinnamaldehyde and 20 mg of amide 6f in
4 mL of ethanol, a few crystals of b-alanine were added.
The mixture was stirred at 80 ꢁC for 4.0 h until the
starting material disappeared. H2O (10 mL) was added,
and the mixture was kept at 0 ꢁC for 2 h. The precipi-
tated powder was washed with H2O, re-crystallized from
MeCN–H2O, and dried in a desiccator over NaOH.
Yield 75%.
5.2.12. 2-Cyano-5-(3,4-dihydroxyphenyl)-4E-pentenoic
acid benzylamide (CR-54). To a solution of CR-4
(32 mg) in 8 mL MeOH 200 mg of NaBH4 was added,
and the mixture was stirred at room temperature for 1 h.
The mixture was acidified with 1 N HCl, MeOH was
evaporated and the residue was extracted with ethyl
acetate. The solvent was evaporated and the residue was
purified on a reverse-phase HPLC. Yield 25 mg (78%).
5.3.2. CFU-GEMM assay. The CFU-GEMM assay was
performed according to Fauser and Messner22;23 with
some minor variations. In brief, heparinized bone mar-
row cells were layered over Percoll and centrifuged at
400g at 4 ꢁC for 10 min to remove neutrophils and
RBCs. The fractionated BM cells at 2 · 105 cells/mL
were cultured in IMDM containing 0.9% (v/v) methyl-
cellulose supplemented with 20% FCS or normal human
plasma, a cocktail of cytokines containing G-CSF
(10 ng/mL), IL-3 (40 U/mL), MGF (50 ng/mL), erythro-
poietin (2 U/mL), and 5 · 10ꢀ5 M b-2-mercaptoethanol.
All cultures were evaluated at 14 days for the number of
BFU-E colonies (defined as aggregates of more than 500
5.2.13. Modeling experiment. Condensation of CR-4 with
mercaptans. To a solution of CR-4 (16 mg, 0.05 mmol) in
0.3 mL DMSO, 0.5 mmol of mercaptan RSH (R ¼ Et,
Bz) was added. In case of ethanethiol, 20 lL of piper-
idine was added, and the mixture was kept for 0.5 h at
room temperature. The mixture was acidified with 1 N

3026
P. Demin et al. / Bioorg. Med. Chem. 12 (2004) 3019–3026
8. Hori, H.; Nagasawa, H.; Ishibashi, M.; Uto, Y.; Hirata,
A.; Saijo, K.; Ohkura, K.; Kirk, K. L.; Uehara, Y. Bioorg.
Med. Chem. 2002, 10, 3257.
9. Talalay, P.; De Long, M. J.; Prochaska, H. J. Proc. Natl.
Acad. Sci. U.S.A. 1988, 85, 8261.
10. Talalay, P. Adv. Enzyme Regul. 1989, 28, 237.
11. Ahn, B.-Z.; Sok, D.-E. Curr. Pharm. Des. 1996, 2, 247.
12. Meydan, N.; Grunberger, T.; Dadi, H.; Shahar, M.;
Arpaia, E.; Lapidot, Z.; Leeder, J. S.; Freedman, M.;
Cohen, A.; Gazit, A.; Levitzki, A.; Roifman, C. M. Nature
1996, 379, 645.
hemaglobinized cells or, three or more erythroid sub-
colonies), CFU-GM colonies (defined as granulocyte or
monocyte-macrophage cells or both), and CFU-GEMM
colonies (a mixed population comprising of all ele-
ments). To prepare haematopoietic stem cells, normal
bone marrow cells were isolated as above and CD34þ
cells positively selected with anti-CD34 magnetic beads,
utilized the MACS magnetic cell sorting system (Milte-
nyi Biotec Inc, CA).
13. Nesterov, V. N.; Antipin, M. Yu.; Timofeeva, T. V.;
Clark, R. D. Acta Cryst. C, Cryst. Struct. Commun. 2000,
C56, 88.
References and notes
1. Sridhar, R.; Hanson-Painton, O.; Cooper, D. R. Pharm.
Res. 2000, 17, 1345.
2. Umezawa, H.; Imoto, M.; Sawa, T.; Isshiki, K.; Matsuda,
N.; Uchida, T.; Iinuma, H.; Hamada, M. J. Antibiot. 1986,
39, 170.
14. Ho, Y. W.; Wang, I. J. J. Heterocycl. Chem. 1995, 32, 819.
15. Cresp, T. M.; Sargent, M. V.; Vogel, P. J. Chem. Soc.,
Perkin Trans. 1 1974, 37.
16. Potgieter, M.; Wenteler, G. L.; Drewes, S. E. Phytochem-
istry 1988, 27, 1101.
€
17. Burgi, C.; Ruedi, P. Helv. Chim. Acta 1993, 76, 1901.
18. Kadin, S. B. J. Org. Chem. 1966, 31, 620.
19. Grunberger, T.; Demin, P.; Rounova, O.; Sharfe, N.;
Cimpean, L.; Dadi, H.; Freywald, A.; Estrov, Z.; Roif-
man, C. M. Blood 2003, 102, 4153.
20. Miyamoto, N.; Sugita, K.; Goi, K.; Inukai, T.; Iijima, K.;
Tezuka, T.; Kojika, S.; Nakamura, M.; Kagami, K.;
Nakazawa, S. Leukemia 2001, 15, 1758.
€
3. Imoto, M.; Umizawa, K.; Isshiki, K.; Kumimoto, S.;
Sawa, T.; Takeuchi, T.; Umezawa, H. J. Antibiot. 1987,
40, 1471.
4. Isshiki, K.; Imoto, M.; Sawa, T.; Umezawa, K.; Takeuchi,
T.; Umezawa, H.; Tsuchida, T.; Yoshioka, T.; Tatsuta, K.
J. Antibiot. 1987, 40, 1209.
5. Gazit, A.; Yaish, P.; Gilon, C.; Levitzki, A. J. Med. Chem.
1989, 32, 2344.
6. Yaish, P.; Gazit, A.; Gilon, C.; Levitzki, A. Science 1988,
242, 933.
7. Montgomery, G. J.; McKeown, P.; McGown, A. T.;
Robins, D. J. Anti-Cancer Drug Des. 2000, 15, 171.
21. Estrov, Z.; Grunberger, T.; Dube, I. D.; Wang, Y. P.;
Freedman, M. H. N. Engl. J. Med. 1986, 315, 538.
22. Fauser, A. A.; Messner, H. A. Blood 1978, 52, 1243.
23. Messner, H. A.; Fauser, A. A. Blut 1980, 41, 327.