Cinnamaldehydes inhibit cyclin dependent kinase 4/cyclin D1

Article abstract of DOI:10.1016/S0960-894X(00)00357-7

A series of cinnamaldehydes was synthesized for the study of inhibitory activity against cyclin dependent kinases (CDKs). A couple of compounds selectively inhibited cyclin D1-CDK4 with an IC₅₀ value of 7-18 μM. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00357-7

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1819±1822
Cinnamaldehydes Inhibit Cyclin Dependent Kinase 4/Cyclin D1
Ha-Won Jeong,a Mi-Ran Kim,a Kwang-Hee Son,a Mi Young Han,a Ji-Hong Ha,b
Matthieu Garnier,c Laurent Meijerc and Byoung-Mog Kwona,*
aKorea Research Institute of Bioscience and Biotechnology, KIST, PO Box 115, Yusung, Taejon, 305-600 South Korea
bDepartment of Genetic Engineering, Kyungbuk National University, Taegu, 702-701 South Korea
cCNRS, Station Biologique, B.P. 74, 29682 Rosco€ cedex, France
Received 10 April 2000; accepted 19 June 2000
AbstractÐA series of cinnamaldehydes was synthesized for the study of inhibitory activity against cyclin dependent kinases
(CDKs). A couple of compounds selectively inhibited cyclin D1-CDK4 with an IC50 value of 7±18 mM. # 2000 Elsevier Science
Ltd. All rights reserved.
In eukarytotes, cell cycle control is the major regulatory
mechanism of cell growth.1 In the past decade, enor-
mous advances have been made in our understanding of
the molecules that mediate the control of cell cycle.2,3
Cyclins, cyclin dependent kinases (CDKs), CDK inhi-
bitors (CDIs), retinoblastoma protein (pRB), and the
E2F transcription factor family that form the central
core of cell-cycle machinery have been identi®ed and
their encoding genes cloned. Among these are the cyclin
dependent kinases, which phosphorylate a large variety
of substrates directly involved in the cell cycle progres-
sion, such as pRb or lamins.4 These kinases are acti-
vated by association with their speci®c regulatory cyclin
proteins and by various post-translational modi®ca-
tions. Primary regulators of the progression are the D-
type cyclins (D1, D2, D3), cyclin E, CDK2, and CDK4.
CDK4 is activated early in G1 by interactions with D-type
cyclins, whereas CDK2 is activated later in G1 by inter-
actions with cyclin E. Pharmaceutical attention has been
focused on the cyclin D1-CDK4 complexes, because the
cumulative data show that the activity of the cyclin D1-
CDK4 is critical for proliferation of many tumors.
Therefore, selective inhibitors of cyclin D1-CDK4 will be
useful for the development of anticancer drugs.5
a cytotoxicity against tumor cell lines.6 10 Compound
IIa also strongly inhibited the growth of tumors without
loss of body weight in nude mice.9 However, it is very dif-
®cult to identify the antitumor mechanism of the com-
pound. The antitumor activity of IIa is likely to derive
from a combination of various independent and additive
mechanisms. Therefore, compound IIa was measured to
have an inhibition activity against CDKs, the major
regulatory kinases in cell cycle and we found that IIa
fairly inhibited cyclin D1-CDK4 with an IC50 value of
5 mg/mL (Table 1). We thus planned to synthesize a
series of derivatives of IIa and measured the CDK
inhibition activity of the cinnamaldehydes.
Based on the previous results, the propenal and free
phenolic hydroxy groups of the cinnamaldehydes were
identi®ed as key functional groups. However, it was
reported, when the propenal group was converted to
propenol or propanal, that the antitumor activity of the
compounds disappeared.10 Therefore, we have focused
on making a cinnamaldehyde library based upon tem-
plates IIa and IIb, and substitution of 20- or 30-OH in
Scheme 1. In particular, IIc and IId were synthesized
to introduce chloro and nitro groups into the 20-OH
position to compare the inhibition activity between 20-OH
and other substituents.
In the course of a screening of herbal medicines for the
development of new types of anticancer drugs, we found
that 20-hydroxycinnamaldehyde (IIa), isolated from the
stem bark of Cinnamonum cassia Blume (Lauraceae),
showed various in vitro activities on farnesyl transferase,
angiogenesis, immunmodulation, cell±cell adhesion, and
The compounds were synthesized using standard litera-
ture methods as outlined in Scheme 1. 20 or 30-Substituted
cinnamic acids were converted to their corresponding
esters, followed by reduction of the esters with diisobutyl
aluminum hydride to give cinnamyl alcohols. Mangan-
ese(IV) oxide oxidation of these alcohols gave cinna-
maldehydes (II). 20 or 30-Hydroxycinnamaldehydes (IIa
*Corresponding author. Tel.:+82-42-860-4557; fax: 82-42-861-2675;
e-mail: kwonbm@kribb4680.kribb.re.kr
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00357-7

1820
H.-W. Jeong et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1819±1822
and IIb) were treated with a variety of substituted ben-
zyl halides to provide compounds III in a good yield
(80±90%).10
The enzyme inhibitory activity of the synthesized com-
pounds was determined by measuring the phosphoryla-
tion of histone H1 for cyclin E-CDK2 and cyclin B-
CDK1, and Rb protein for cyclin D1-CDK4,
respectively.12 14 Their inhibitory activities towards the
enzymes are presented as percentages of the control con-
dition in which the enzyme assay was carried out in the
absence of inhibitors. The inhibitory activity of cinna-
maldehydes against cyclin D1-CDK4 is summarized in
Table 1 and olomoucine (IC50 10mM for cyclin E-
CDK2)12 and ¯avopiridol (IC50 100 nM for cyclin D1-
CDK4),15,16 well known CDK2 and 4 inhibitors, respec-
tively, were used as standard compounds. 30-Substituted
Structures of the synthesized compounds were identi®ed
by 1H NMR and HRMS. Especially, the propenal group
of the compounds was clearly identi®ed by the observed
1H NMR chemical shifts and coupling constants, for
example of the aldehyde at d 9.6 (d, 7.7 Hz), and pro-
penyl moiety at d 8.0 (d, 16.3 Hz) and d 6.8 (dd, 7.8 and
16.3 Hz) for IIa. The methylene group of benzyl sub-
stituted compound III was also detected by 1H NMR at
d 4.5±5.0.11
Scheme 1.
Table 1. Cyclin D1-CDK4 inhibition activity of cinnamaldehydes
Compound
Inhibitiona
Compound
Inhibitiona
Ia (20-OH)
<10
50
30
85
70
22
43
68
71
59
38
63
72
25
52
50
23
IIb (30-OH)
IIc (20-Cl)
40
51
IIa (20-OH)
IId (20-NO2)
IIIa (20-OCH2Ph)
IIIa0 (30-OCH2Ph)
IIIb0 (30-OCH2Ph-2-CH3)
IIIc0 (30-OCH2Ph-2-CN)
IIId0 (30-OCH2Ph-2-OCH3)
IIIe0 (30-OCH2Ph-4-CH3)
IIIf0 (30-OCH2Ph-4-Br)
IIIg0 (30-OCH2Ph-4-OCH3)
IIIh0 (30-OCH2Naphthalene)
IIIi0 (30-OCH(CH3)Ph)
IIIj0 (30-OCH2CH2Ph)
IV (30-OH)
94
73
45
65
84
88
82
85
77
80
40
58
54
53
34
IIIb (20-OCH2Ph-2-CH3)
IIIc (20-OCH2Ph-2-CN)
IIId (20-OCCH2Ph-2-OCH3)
IIIe (20-OCH2Ph-4-CH3)
IIIf (20-OCH2Ph-4-Br)
IIIg (20-OCH2Ph-4-OCH3)
IIIh (20-OCH2Naphthalene)
IIIi (20-OCH(CH3)Ph )
IIIj (20-OCH2CH2Ph)
IIIk (20-OC(O)Ph-4-F)
IIIl (20-OC(O)Ph)
IVa (30-OCH2Ph-4-Br)
V
VI
IIIm (20-OC(O)Ph-4-CH3)
IIIn (20-OC(O)Ph-2-OCH3)
VII
a% Inhibition at 5 mg/mL.

H.-W. Jeong et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1819±1822
1821
Table 2. IC50 values for cyclin D1-CDK4 and cyclin B-CDK1
information for the design of CDK4 inhibitor and
evaluation of biological activities of cinnamaldehydes.
Compound
IC50 (mM)
(cyclin D1-CDK4)
IC50 (mM)
(cyclin B-CDK1)
IIa
IIb
35
45
11
7.5
18
15
0.1a
130
30
50
50
200
60
Acknowledgements
IIIa
IIIa0
IIIb
IIIb0
We are grateful to Professors J. Wade Harper and Boris
Sarcevic for providing baculovirus cyclin D-CDK4 and
cyclin E-CDK2 constructs. This research was supported
by the Ministry of Science & Technology and the Min-
istry of Agriculture & Forestry-Special Grants Research
Program as well as grants from the `Association pour la
Recherche sur le Cancer' (ARC) and the `Conseil
Regional de Bretagne' (to L.M.).
Flavopiridol
aPositive control.
References and Notes
1. Morgan, D. Annu. Rev. Cell. Dev. Biol. 1997, 13, 261.
2. Shapiro, G. I.; Harper, J. W. J. Clin. Invest. 1999, 104,
1645.
3. Meijer, L., Jezequel, A., Ducommun, B., Ed.; Progress in
Cell Cycle Research; Plenum Press: New York, 2000; Vol. 4,
pp 1±248.
4. Peeper, D. S.; Parker, L. L.; Ewen, M. E.; Toebes, M.;
Hall, F. L.; Xu, M.; Zantema, A.; van der Eb, A.; Piwnica-
Worms, H. EMBO J. 1993, 12, 1947.
5. Sielecki, T. M.; Boylan, J. F.; Ben®eld, P. A.; Trainor, G.
L. J. Med. Chem. 2000, 43, 1.
Figure 1. CDK4 inhibition activities by compound IIIa and IIIa0.
6. Kwon, B. M.; Cho, Y. K.; Lee, S. H.; Nam, J. Y.; Bok, S.
H.; Chun, S. K.; Kim, J. A.; Lee, I. R. Planta Med. 1996, 62,
183.
cinnamaldehydes exhibit higher inhibition activity than
that of the 20-substituted compounds towards cyclin
D1-CDK4 (see Figure 1). When a polar substituent is
introduced at the C2 position of the benzyl group, the
inhibitory activity is also signi®cantly decreased. Some
compounds with bulky groups such as methoxy-
coumarin (VI) and anthraquinone (VII), and with
disubstituted cinnamaldehyde (IV) and pyridyl groups
(V) were also prepared for comparison of the inhibitory
activity with the other cinnamaldehydes. We found that a
signi®cant loss of inhibitory activity is observed from
those compounds. Compounds IIIb0, IIIf 0 and IIIh0
have a high potency for cyclin D1-CDK4 inhibition with
IC50 values of 7.5, 8.2, and 7.8 mM, respectively. However,
these compounds have signi®cantly weaker inhibition
activity for the cyclin E-CDK2 (IC50 >200 mM) (data not
shown).
7. Kwon, B. M.; Lee, S. H.; Cho, Y. K.; Bok, S. H.; So, S. H.;
Chang, S. I. Bioorg. Med. Chem. Lett. 1997, 7, 2473.
8. Koh, W. S.; Yoon, S. Y.; Kwon, B. M.; Jeong, T. C.; Nam,
K. S.; Han, M. Y. Int. J. Immunopharm. 1998, 20, 643.
9. Lee, C. W.; Hong, D. H.; Han, S. B.; Park, S. H.; Kim, H.
K.; Kwon, B. M.; Kim, H. M. Planta Med. 1999, 65, 263.
10. Kwon, B. M.; Lee, S. H.; Choi, S. U.; Park, S. H.; Lee, C.
O.; Cho, Y. K.; Sung, N. D.; Bok, S. H. Arch. Pharm. Res.
1998, 21, 147.
11. Data for compound IIIb: HRMS: 252.1139 (calcd
1
252.1150 for C17H16O2). H NMR (400 MHz, CDCl3): d 9.65
(d, 7.5 Hz, 1H), 7.89 (d, 16.3 Hz, 1H), 7.58 (m, 1H), 7.42 (m,
2H), 7.25 (m, 2H), 7.05 (m, 2H), 6.76 (dd, 7.5, 16.2 Hz), 5.09
(s, 3H), 2.38 (s, 3H). Data for compound V: HRMS: 132.0441
(calcd 132.0449 for C8H6NO). 1H NMR (400 MHz, CDCl3): d
9.74 (d, 7.6 Hz, 1H), 8.90 (s, 1H), 8.64 (d, 4.8 Hz, 1H), 8.17 (d,
8.0 Hz, 1H), 7.75 (d, 16.1 Hz, 1H), 7.48 (dd, 4.8, 8.0 Hz, 1H),
6.90 (dd, 7.5, 16.1 Hz).
12. Vesely, J.; Havlicek, L.; Strnad, M.; Blow, J. J.; Donella-
Deana, A.; Pinna, L.; Letham, D. S.; Kato, J. Y.; Detivaud,
L.; Leclerc, S.; Meijer, L. Eur. J. Biochem. 1994, 224, 771.
13. Meijer, L.; Borgne, A.; Mulner, O.; Chong, J. P. J.; Blow,
J. J.; Inagaki, N.; Inagaki, M.; Delcros, J. G.; Moulinoux, J.
P. Eur. J. Biochem. 1997, 243, 527.
14. Enzyme puri®cation. GST-cyclin D1-CDK4 or GST-
cyclin E-CDK2 constructs in baculovirus were expressed in
sf21 cells. Cell lysates were subjected to glutathione chroma-
tography to obtain enzymatic purity and activity suitable for
kinase assay screening of the compound library. Kinase assay.
Enzyme assays to screen for inhibitors of cyclin D1-CDK4
were performed in 96 well plates. Each reaction contained
In a cyclin B-CDK1 inhibition assay, the synthesized
compounds exhibited low inhibitory activity compared to
cyclin D1-CDK4. Compounds IIIb and b0 show 10 and 8
times the selectivity. The IC50 values of the compounds
are summarized in Table 2.
In conclusion, a series of cinnamaldehydes has been
synthesized to identify the antitumor mechanism of
the compound and characterized as CDK4 inhibitors.
The most potent of this series, compound IIIa0, has
been identi®ed. These results should provide valuable

1822
H.-W. Jeong et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1819±1822
5 mM GST-RB fusion protein corresponding to the amino acid
postion 792±928 of the RB protein, 100 mM ATP, 0.5 mCi
[g-33P ] ATP, and 200 nM enzyme in a ®nal volume of 100 mL/
well. Compounds were added to the reaction mixture and after
30 minutes at 30oC, the reaction was stopped by addition of
ice cold 30% TCA and the phosphorylated products were
adsorbed onto PVDF membranes in 96 well microtiter ®lter
plates (Millipore). The radioactivity in [g-33P] -labeled RB was
measured in a Wallac Microbeta Counter after addition of
25 mL/well scintillant (Wallac Optiphase). cyclin E-CDK2 assay
was performed with histone H1 for the substrate in the same
way as for cyclin D1-CDK4. Cyclin B-CDK1 was puri®ed
from star®sh oocytes and assayed for 10 min, at 30 ꢀC, at a
®nal ATP concentration of 15 mM as described previously (ref
12).
15. Carlson, B. A.; Dubay, M. M.; Sausville, E. A.; Brizuela,
L.; Worland, P. J. Cancer Res. 1996, 56, 2973.
16. Meijer, L. Trends Cell Biol. 1996, 6, 393.