Synthesis and biological activity of 2-anilino-4-(1H-pyrrol-3-yl) pyrimidine CDK inhibitors

Article abstract of DOI:10.1016/j.bmcl.2004.06.012

A series of 2-anilino-4-(1H-pyrrol-3-yl)pyrimidines were prepared and evaluated for their ability to inhibit cyclin-dependent kinases (CDKs). A number of analogues were found to be potent CDK2 and CDK4 inhibitors and to exhibit anti-proliferative activity

Full text of DOI:10.1016/j.bmcl.2004.06.012

Bioorganic & Medicinal Chemistry Letters 14 (2004) 4237–4240
Synthesis and biological activity of 2-anilino-4-(1H-pyrrol-3-yl)
pyrimidine CDK inhibitors
Shudong Wang,a,* Gavin Wood,a Christopher Meades,a Gary Griffiths,a Carol Midgley,b
Iain McNae,c Campbell McInnes,a Sian Anderson,a Wayne Jackson,a Mokdad Mezna,a
Rhoda Yuill,a Malcolm Walkinshawc and Peter M. Fischera
aCyclacel Limited, James Lindsay Place, Dundee DD1 5JJ, Scotland, UK
bCyclacel Limited, Polgen Division, Babraham Bioincubator, Babraham Hall, Cambridge CB2 4AT, UK
cStructural Biochemistry, The University of Edinburgh, Michael Swann Building, King’s Buildings, Edinburgh EH9 3JR, Scotland, UK
Received 23 March 2004; revised 1 June 2004; accepted 7 June 2004
Available online
Abstract—A series of 2-anilino-4-(1H-pyrrol-3-yl)pyrimidines were prepared and evaluated for their ability to inhibit cyclin-
dependent kinases (CDKs). A number of analogues were found to be potent CDK2 and CDK4 inhibitors and to exhibit
anti-proliferative activity against human tumour cell lines. Structure–activity relationships and biochemical characterization are
presented.
Ó 2004 Elsevier Ltd. All rights reserved.
Abnormalities in cell cycle regulation are responsible for
the majority of human neoplasias.1–4 Cyclin-dependent
kinases (CDKs) are key regulators of cell cycle pro-
gression.5 These enzymes are activated by the formation
of periodic complexes with cyclins, proteins that are
present at specific stages of the cell cycle. CDK4 and
CDK6, coupled with their respective cyclin D partners,
are responsible for progression through G1, whereas
CDK2 in combination with cyclin E is required for
normal progress from G1 into S-phase, where DNA
replication takes place. CDK1 and CDK2 (in associa-
tion with cyclins A and B) functions, on the other hand,
are essential for cells to pass through S-phase into G2
and mitosis. The CDK/cyclin complexes are regulated
by stoichiometric associations with small proteins, the
so-called cyclin-dependent kinase inhibitors (CDKIs),
such as p16, p15, p21, and p27, whose loss or inactiva-
tion participates in the development of cancer. CDKs
phosphorylate and regulate the activity of a variety of
cellular proteins that include tumour suppressors (e.g.,
pRb, p53), transcription factors (e.g., E2F-DP1, RNA
pol II), replication factors (e.g., DNA pola, replication
protein A), and organizational factors, which influence
cellular and chromatin structures (e.g., histone H1,
lamin A, MAP4). The recent understanding of the role
of CDKs in cell cycle regulations and the discovery that
high rates of neoplasias are the result of CDK hyper-
activation provide the main impetus to the search for
their pharmacological inhibitors.
Our efforts to develop small molecule ATP-antagonistic
CDK inhibitors as cancer therapeutics has resulted in
the discovery of 2-anilino-4-(thiazol-5-yl)pyrimidine cell
cycle inhibitors.6–8 We now report the discovery of
another related class of CDK inhibitors, 2-anilino-4-
(1H-pyrrol-3-yl)pyrimidines, obtained through struc-
ture-guided analogue design and optimization. Here we
described the synthesis, SAR analysis, and biochemical
characterization of these 2-anilino-4-(1H-pyrrol-3-
yl)pyrimidine CDK inhibitors.
The chemistry adopted for the synthesis of the target
compounds is summarized in Scheme 1. Treatment of 3-
acetyl-1H-pyrroles 3–5 with tert-butoxybis(dimethyl-
amino)methane (Bredereck’s reagent) afforded the
enaminone intermediates 6. These were then condensed
with the appropriate phenyl guanidines 7 to the desired
pyrimidines 8 at elevated temperature in alcoholic alkali.
The former were derived from the corresponding
Keywords: CDKs; Inhibitors; SAR; Crystal structures.
* Corresponding author. Tel.: +44-1382-206062; fax: +44-1382-2060-
67; e-mail: swang@cyclacel.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.06.012

4238
S. Wang et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4237–4240
R3
R2
H
N
H
H2N
O2N
N
a
b
H2N
N
O
O
1
3
R3
R2
R1
NH
NH
N
R1
N
N
H2N
N
H
tBuO
NH
O
NC
H
H
N
R3
R2
NC
7
e
N
d
N
N
H
O
N
2
4
6
8
c
O
H
N
H2N
O
5
Scheme 1. Reagents and conditions: (a) HNO3, H2SO4, AcOH/H2O, DCM; (b) AlCl3, AcCl, DCM, 0 °C–rt; (c) NH3, MeOH, H2O2/H2O, rt; (d)
neat, 75 °C; (e) K2CO3, 2-methoxyethanol, 125 °C.
anilines by the treatment of their nitrate or hydrochlo-
ride salts with cyanamide.8
the thiazole ring overlaps roughly with the space occu-
pied by the ribose group of ATP. However, it makes
better contacts with adjacent residues, particularly
Phe80 in the ATP-binding pocket of CDK2. A number
of potent CDK2 inhibitors were obtained through
introduction of polar substituents to the thiazole sys-
tem.8 One aspect of our analogue design program has
therefore been to evaluate the effect of replacing the
thiazole ring with other heterocyclic systems, 2,4-di-
methyl-1H-pyrrol-3-yl in the present case, in the context
of the optimal meta- or para-substituted anilino pyrim-
idin-2-yl system. A number of analogues were synthe-
sized and screened against CDKs; the SARs are
presented in Table 1. The most potently anti-prolifera-
tive compound 8b was also screened in a panel of 16
representative Ser/Thr and Tyr protein kinases. Apart
from CDKs, only the closely CDK-related glycogen
synthase kinase-3 was inhibited at submicromolar Ki. Of
the CDKs not shown in Table 1, CDKs 1, 7, and 9 were
3-Acetyl-2,4-dimethyl-5-nitro-1H-pyrrole
3 was ob-
tained by nitration of 3-acetyl-2,4-dimethylpyrrole in the
usual manner. Friedel–Crafts acylation of 3,5-dimethyl-
1H-pyrrole-2-carbonitrile 29 with acetyl chloride in the
presence of aluminium chloride gave 4-acetyl-3,5-di-
methyl-1H-pyrrole-2-carbonitrile 4, whose hydrolysis
afforded 4-acetyl-3,5-dimethyl-1H-pyrrole-2-carboxam-
ide 5. 2-Halo-3,5-dimethyl-1H-pyrrol-4-yl)-pyrimidine
analogues 8k and 8l (Table 1) were obtained by treat-
ment of [4-(2,4-dimethyl-1H-pyrrol-3-yl)pyrimidin-2-
yl](4-fluorophenyl)amine with N-chloro-succinimide and
N-bromo-succinimide, respectively.10
In the SARs and CDK2 complex crystal structures of
our
yl)pyrimidine cell cycle inhibitors, it was observed that
previously
described
2-anilino-4-(thiazol-5-
Table 1. CDK inhibitory and in vitro anti-proliferative activities of 2-anilino-4-(1H-pyrrol-3-yl)pyrimidines
Compound
R1
R2
R3
Kinase inhibition Ki, (lM)a
Anti-proliferative
IC50, lMb
CDK2
CDK4
8a
8b
8c
8d
8e
8f
H
NO2
NO2
NO2
H
H
0.05
0.14
0.05
0.05
0.03
0.13
0.09
0.10
0.08
0.03
0.36
0.27
4.17
0.96
0.18
0.26
0.18
0.22
0.30
0.90
0.42
0.12
1.91
1.60
3.30
0.36
0.38
1.36
1.84
1.17
0.89
2.44
1.64
3.12
13.03
15.98
CN
CN
CN
CN
CN
CN
CN
CONH2
NO2
Cl
H
Me
OH
H
OH
OH
Me
H
Me
F
8g
8h
8i
F
H
F
8j
H
F
8k
8l
H
F
Br
H
F
a Values are means of at least three independent determinations.
b Mean values in 72 h MTT assays11 of A549 and H29 human tumour cell lines.

S. Wang et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4237–4240
4239
also inhibited significantly by compound 8b; the same
was true for compounds 8 in general.
involving the aminopyrimidine system with Glu81 and
Leu83 is observed. Furthermore, the hydrophobic ani-
line rings lie against a hydrophobic surface formed by
the Ile10, Phe82, and Leu134 side chains. These inter-
actions are consistent with our 2-anilino-4-(thiazol-5-
yl)pyrimidine pharmacophore. Interestingly, the pyrrole
NH1 forms a strong H-bond interaction with the
Asp145 carboxyl group, whereas a polar interaction of
the thiazol-2-yl N atom with Asn132 was previously
Compound 8a is a potent and comparatively selective
CDK2 inhibitor. Despite its potency against the isolated
CDK2 enzyme, compound 8a showed modest cellular
activity in cell proliferation assays against A549 lung
carcinoma and HT 29 colon cancer cells.
This was dramatically improved by the introduction of a
nitrile group at position 5 of the pyrrole ring and com-
pound 8b exhibited ꢀ10-fold higher cellular activity.
Interestingly, introduction of an additional methyl
group at the para-position of the anilino moiety in
compound 8c was able to restore CDK2 inhibitory
activity (with respect to 8a), while maintaining the cel-
lular anti-proliferative activity of compound 8b. While
there are clearly different profiles in the CDK2/4 selec-
tivity amongst 1H-pyrrol-3-yl nitrile, amide, nitro, and
halo pyrimidin-2-yl-aniline analogues 8b–l, these sub-
stitutions are well tolerated as far as CDK activity is
concerned. It is notable, however, that the halogenated
compounds 8k and 8l are much less effective against the
tumour cells than other analogues of this series. As is
commonly observed with many ATP antagonist kinase
inhibitors, the anti-proliferative, that is cellular potency
of the 2-anilino-4-(1H-pyrrol-3-yl)pyrimidines is signifi-
cantly lower than the potency against isolated CDK
enzymes. In part this is probably due to high intracel-
lular ATP concentrations.
observed
for
2-anilino-4-(thiazol-5-yl)pyrimidine
ligands. Replacement of the pyrrol-5-yl hydrogen
(compound 8a) with a nitrile (compound 8g) function
does not result in an additional interaction with CDK2,
an observation consistent with the similar potency of
these two analogues.
The retinoblastoma tumour suppressor protein pRb is
the main physiological substrate for the G1/S CDK
complexes.12 The entry of cells into S-phase requires
CDK-mediated phosphorylation of pRb, releasing and
activating E2F family transcription factors, which are
inactive while bound to pRb during the G0 and M-
phases of the cell cycle. Cellular CDK inhibitory activity
of our inhibitors was confirmed by the examination of
pRb phosphorylation status following treatment of
A549 lung carcinoma cells with compound 8b (Fig. 2). A
dose- and time-dependent decrease in phosphorylation
of pRb at Ser249/Thr252, Ser807/Ser811, and Thr821,
the preferential CDK2 and CDK4 phosphorylation
sites, were observed. The treatment also resulted in
induction of p53 and in decreased phosphorylation at
Ser-2 and Ser-5 of RNA polymerase II (data not
shown), another substrate of several CDK isoforms,
implying a possible transcriptional inhibition with this
compound.
The X-ray crystal structures of CDK2 complexes with
8a and 8g were determined; the positions of these
inhibitors in the ATP-binding pocket are shown in
Figure 1. The binding modes are similar for both
ligands. The substituted aniline moiety is positioned at
the edge of the ATP binding site and the pyrrole ring is
orientated to occupy the deep pocket of the ribose site.
In each case, the characteristic H-bonding network
In summary, a series of 2-anilino-4-(pyrrol-3-yl)-pyrim-
idine CDK inhibitors were discovered.13 Many of them
exhibited potent CDK enzyme inhibitory activity and
Figure 1. A stereo view of the X-ray crystal structure superimpositions of the CDK2/8a (yellow) and CDK2/8g (orange) complexes in the ATP
binding site. H-bonds are indicated with broken lines. Monomeric CDK2 crystals were used for complex formation and structure calculations.6;8

4240
S. Wang et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4237–4240
1 mL/min, H2O/MeCN solvent system containing 0.1%
CF3COOH) tR 14.3 min (10–70% MeCN linear gradient
over 20 min, purity 99%); 1H NMR (500 MHz, DMSO-d6)
d 2.19 (s, 3H, CH3), 2.42 (s, 3H, CH3), 6.48 (s, 1H, pyrrole-
H), 6.87 (d, 1H, J ¼ 5.5 Hz, pyrimidinyl-H), 7.52 (m, 1H,
Ph-H), 7.73 (m, 1H, Ph-H), 8.08 (m, 1H, Ph-H), 8.38 (d,
1H, J ¼ 5.5 Hz, pyrimidinyl-H), 8.93 (s, 1H, Ph-H), 9.83
(br s, 1H, NH), 10.76 (br s, 1H, NH); 13C NMR (DMSO-
d6) d 13.34, 14.21, 112.02, 112.89, 115.80, 115.82, 117.65,
117.90, 125.25, 130.25, 130.86, 143.01, 148.82, 157.62,
160.05, 164.42; MS (ESIþ) m=z 310.07 (M + H)þ; anal.
(C16H15N5O2) C, H, N. Compound 8b: pale yellow solid,
mp 258–259 °C; anal. RP-HPLC tR 17.1 min (purity
100%); 1H NMR (DMSO-d6) d 2.33 (s, 3H, CH3), 2.43
(s, 3H, CH3), 6.96 (d, 1H, J ¼ 4.5 Hz, pyrimidinyl-H),
7.55 (t, 1H, J ¼ 8.3 Hz, Ph-H), 7.76 (m, 1H, Ph-H), 8.05
(m, 1H, Ph-H), 8.51 (d, 1H, J ¼ 5.0 Hz, pyrimidinyl-H),
8.90 (s, 1H, Ph-H), 10.01 (s, 1H, NH), 12.25 (br s, 1H,
NH); 13C NMR (DMSO-d6) d 12.32, 13.94, 98.87, 112.66,
112.99, 115.09, 116.18, 119.51, 125.39, 130.41, 131.18,
136.34, 142.68, 148.81, 158.65, 160.12, 162.18; MS (ESIþ)
m=z 334.22. Compound 8d: pale tan solid, mp 272–276 °C;
anal. RP-HPLC tR 10.8 min (purity 97%); 1H NMR
(DMSO-d6) d 2.27 (s, 3H, CH3), 2.37 (s, 3H, CH3), 6.67
(m, 2H, Ph-H), 6.73 (d, 1H, J ¼ 5.5 Hz, pyrimidinyl-H),
7.43 (d, 2H, J ¼ 8.5 Hz, Ph-H), 8.32 (d, 1H, J ¼ 5.4 Hz,
pyrimidinyl-H), 9.01 (s, 1H, NH), 12.16 (br s, 1H, NH);
13C NMR (DMSO-d6) d 12.44, 14.05, 98.57, 110.61,
115.21, 115.52, 122.31, 131.09, 132.71, 136.03, 153.03,
158.30, 161.01, 162.20; MS (ESIþ) m=z 305.80. Compound
8e: off-white solid, anal. RP-HPLC tR 16.4 min (0–60%
Figure 2. Western blot analysis of A549 cells treated with compound
8b at 0.5 and 5 lM for 2, 6, and 24 h. DMSO treatment as control.
Experimental procedures were as described previously.8
anti-proliferative activity in tumour cells. Cellular
CDK2/4 inhibitory activities were confirmed.
Acknowledgements
We thank Katy Fishburne for excellent technical assis-
tance in chemical characterizations.
References and notes
1
MeCN, purity 97%); H NMR (DMSO-d6) d 2.30 (s, 3H,
CH3), 2.40 (s, 3H, CH3), 6.83 (d, 1H, J ¼ 5.0 Hz,
pyrimidinyl-H), 7.11 (t, 1H, J ¼ 8.1 Hz, Ph-H), 7.73 (m,
1H, Ph-H), 8.40 (d, 1H, J ¼ 5.5 Hz, pyrimidinyl-H), 9.46
(s, 1H, NH), 12.20 (br s, 1H, NH); MS (ESIþ) m=z 307.70.
Compound 8i: light yellow solid, mp 94–96 °C; anal. RP-
HPLC tR 10.64 min (10–70% MeCN, purity 100%); 1H
NMR (DMSO-d6) d 2.35 (s, 3H, CH3), 2.39 (s, 3H, CH3),
6.76 (d, 1H, J ¼ 6.0 Hz, pyrimidinyl-H), 7.09 (m, 2H, Ph-
H), 7.77 (m, 2H, Ph-H), 8.37 (d, 1H, J ¼ 6.0 Hz,
pyrimidinyl-H), 9.41 (s, 1H, NH), 11.25 (br s, 1H, NH);
13C NMR (DMSO-d6) d 12.58, 13.57, 112.20, 115.42,
115.60, 120.49, 121.12, 121.18, 121.19, 122.15, 122.89,
131.99, 137.86, 137.88, 157.94, 160.53, 163.55; MS (ESIþ)
m=z 325.87. Compound 8k: yellow solid, mp 200–203 °C;
anal. RP-HPLC: tR 15.4 min (purity 98%); 1H NMR
(DMSO-d6) d 2.09 (s, 3H, CH3), 2.34 (s, 3H, CH3), 6.77 (d,
1H, J ¼ 5.5 Hz, pyrimidinyl-H), 7.10 (m, 2H, Ph-H), 7.74
(m, 2H, Ph-H), 8.32 (d, 1H, J ¼ 5.5 Hz, pyrimidinyl-H),
9.42 (s, 1H, NH), 11.54 (br s, 1H, NH); 13C NMR
(DMSO-d6) d 11.41, 13.82, 110.80, 111.00, 113.55, 115.49,
115.66, 121.67, 121.73, 129.61, 137.50, 156.93, 156.94,
158.82, 159.96, 163.62; MS (ESIþ) m=z 317.72 (M + H)þ.
8l: yellow solid, mp 181–183 °C; anal. RP-HPLC tR
15.8 min (purity 95%); 1H NMR (DMSO-d6) d 2.08 (s,
3H, CH3), 2.34 (s, 3H, CH3), 6.77 (d, 1H, J ¼ 5.5 Hz,
pyrimidinyl-H), 7.11 (m, 2H, Ph-H), 7.74 (m, 2H, Ph-H),
8.33 (d, 1H, J ¼ 5.5 Hz, pyrimidinyl-H), 9.39 (s, 1H, NH),
11.50 (br s, 1H, NH); 13C NMR (DMSO-d6) d 12.56,
13.87, 110.00, 111.12, 115.46, 115.64, 116.75, 119.06,
121.49, 121.51, 131.34, 137.63, 157.41, 158.74, 160.15,
163.41; MS (ESIþ) m=z 362.90 (M + H)þ.
1. Hartwell, L. H.; Kastan, M. B. Science 1994, 266, 1821–
1828.
2. Webster, K. R.; Kimball, S. D. Emerg. Drugs 2000, 5, 45–
59.
3. Toogood, P. L. Med. Res. Rev. 2001, 21, 487–498.
4. Sherr, C. J. Science 1996, 274, 1672–1677.
5. Senderowicz, A. M. Oncogene 2003, 22, 6609–6620.
6. Wu, S. Y.; McNae, I.; Kontopidis, G.; McClue, S. J.;
McInnes, C.; Stewart, K. J.; Wang, S.; Zheleva, D. I.;
Marriage, H.; Lane, D. P.; Taylor, P.; Fischer, P. M.;
Walkinshaw, M. D. Structure 2003, 11, 399–410.
7. Wang, S.; Anderson, S.; Clarke, R.; Duncan, K.; Eland,
S.; Jackson, W.; Lyon, C.; McClue, S.; Meades, C.;
Osnowski, A.; Perry, A.; Renachowski, S.; Stuart, I.;
Thomas, M.; Westwood, B.; Wood, G.; Yuill, R.; Lane,
D.; Jackson, R.; Fischer, P. Proc. Am. Assoc. Cancer Res.
2003, 44, Abs. 3974.
8. Wang, S.; Meades, C.; Wood, G.; Osnowski, A.; Ander-
son, S.; Yuill, R.; Thomas, M.; Mezna, M.; Jackson, W.;
Midgley, C.; Griffiths, G.; Fleming, I.; Green, S.; McNae,
I.; Wu, S.-Y.; McInnes, C.; Zheleva, D.; Walkinshaw, M.
D.; Fischer, P. M. J. Med. Chem. 2004, 47, 1662–1675.
9. Cheng, L. J.; Lightner, D. A. Synthesis 1999, 46–48.
10. Fischer, P. M.; Wang, S.; Wood, G. In PCT Intl. Patent
Appl. Publ. WO 02/079193; Cyclacel Limited, UK, 2002.
11. Haselsberger, K.; Peterson, D. C.; Thomas, D. G.;
Darling, J. L. Anti-Cancer Drugs 1996, 7, 331–338.
12. Ortega, S.; Malumbres, M.; Barbacid, M. Curr. Genomics
2002, 3, 245–263.
13. Analytical data for representative compounds. 8a: yellow
solid, mp 198–200 °C; anal. RP-HPLC (Vydac 218TP54,