Synthesis and evaluation of biological and antitumor activities of 5,7-dimethyl- oxazolo[5,4-d]pyrimidine-4,6(5H,7H)-dione derivatives as novel inhibitors of FGFR1

Article abstract of DOI:10.3109/14756366.2014.1002401

A series of 5,7-dimethyl-oxazolo[5,4-d]pyrimidine-4,6(5H,7H)-dione derivatives, N5a-5l, was designed, synthesized and evaluated for their FGFR1-inhibition ability as well as cytotoxicity against three cancer cell lines (H460, B16F10 and A549) in vitro. Several compounds displayed good-to-excellent potency against these cancer cell lines compared to SU5402. Structure-activity relationship analyses indicated that compounds with a rigid structure and more heteroatoms at the side chain of the parent ring were more effective than those without these substitutions. The compound N5g (37.4% FGFR1 inhibition at 1.0 μM) was identified to have the most potent antitumor activities, with IC₅₀ values of 5.472, 4.260 and 5.837 μM against H460, B16F10 and A549 cell lines, respectively. Together, our results suggest that 5,7-dimethyl-oxazolo[5,4-d]pyrimidine-4,6(5H,7H)-dione derivatives may serve as potential agents for the treatment of FGFR1-mediated cancers.

Full text of DOI:10.3109/14756366.2014.1002401

http://informahealthcare.com/enz
ISSN: 1475-6366 (print), 1475-6374 (electronic)
J Enzyme Inhib Med Chem, Early Online: 1–6
2015 Informa UK Ltd. DOI: 10.3109/14756366.2014.1002401
!
RESEARCH ARTICLE
Synthesis and evaluation of biological and antitumor activities of
,7-dimethyl- oxazolo[5,4-d]pyrimidine-4,6(5H,7H)-dione derivatives
as novel inhibitors of FGFR1
5
1
,2
2
2
2
2
2
Faqing Ye , Yuewu Wang , Siyun Nian , Yu Wang , Di Chen , Shufang Yu , and Sicen Wang
1
1 2
School of Pharmacy, Health Science Center Xi’an Jiaotong University, Xi’an, China and School of Pharmaceutical Sciences, Wenzhou Medical
University, Wenzhou, China
Abstract
Keywords
A series of 5,7-dimethyl-oxazolo[5,4-d]pyrimidine-4,6(5H,7H)-dione derivatives, N5a–5l, was
designed, synthesized and evaluated for their FGFR1-inhibition ability as well as cytotoxicity
against three cancer cell lines (H460, B16F10 and A549) in vitro. Several compounds displayed
good-to-excellent potency against these cancer cell lines compared to SU5402. Structure–
activity relationship analyses indicated that compounds with a rigid structure and more
heteroatoms at the side chain of the parent ring were more effective than those without these
substitutions. The compound N5g (37.4% FGFR1 inhibition at 1.0 mM) was identified to have the
most potent antitumor activities, with IC50 values of 5.472, 4.260 and 5.837 mM against H460,
B16F10 and A549 cell lines, respectively. Together, our results suggest that 5,7-dimethyl-
oxazolo[5,4-d]pyrimidine-4,6(5H,7H)-dione derivatives may serve as potential agents for the
treatment of FGFR1-mediated cancers.
Cancer therapy, FGFR1, kinase inhibitor,
structure–activity relationship, synthesis
History
Received 29 October 2014
Revised 26 November 2014
Accepted 30 November 2014
Published online 16 February 2015
Introduction
aid in the transfer of phosphate from ATP to the tyrosine residues
1
4
of their target proteins . In the presence of tyrosine kinase
inhibitors, cell proliferation would be blocked by preventing the
ATP from binding to the RTKs. Our goal was to design a new
class of tyrosine kinase inhibitors that can mimic ATP and block
the constitutive activation of FGFR1, thereby blocking the
FGFR1-dependent cancer signaling pathway.
Through structural analyses of FGFR1 inhibitors, SU5402,
PD173074 (Figure 1) and others, we discovered certain criteria
that needed to be fulfilled in order to achieve high affinity binding
to the FGFR1 ATP binding pocket. These criteria require the
Cancer is a major health concern in today’s society due to
increasing instances of diagnosis and mortality. Various types of
cancer therapeutic drugs are in clinical use, but many of these
drugs fail to cure the disease because of their low selectivity and
1
,2
high toxicity . In the last decade, receptor tyrosine kinases
RTKs) have become a hot topic in the field of cancer research.
There is growing interest in targeting these receptors for cancer
(
3
therapy . Fibroblast growth factor receptors (FGFR1, FGFR2,
FGFR3 and FGFR4) constitute a major class of RTKs serving as
high-affinity receptors for a large family of fibroblast growth
factors (FGFs) to control key cellular processes, such as cell
parent nucleus of FGFR1 antagonists to be: (i) flat, (ii) aromatic
15–20
4
,5
or ꢀ–electron-rich and (iii) nitrogen-containing heterocycles
.
proliferation, migration and apoptosis . FGFRs and their
associated FGFs have been receiving increasing attention in the
SU5402 and PD173074 serve as important antitumor inhibitors
because of their good selectivity towards FGFR1. As expected,
they both fit within the identified criteria. Molecular docking also
reveals that the parent nuclei of SU5402 and PD173074 occupy
the same site as ATP adenine, and that their nitrogen interacted
with the ATP hinge. However, the 3-position side chain of
SU5402 could not completely take up the hydrophobic pocket,
which plays an important role in the inhibition of FGFR1.
In addition, when testing for inhibition in the growth of BaF3/
ZNF198-FGFR1 cells, the IC50 of SU5402 was reported to be
6
,7
area of antitumor drug discovery
.
FGFR1 is the founding member of FGFR family and is known
to be ectopically expressed in prostate carcinoma cells, breast
8
,9
.
carcinoma cells, lung cancer cells and also other cancer cells
Amplification of FGFR1 occurs in 10% of breast cancers and is
1
0
associated with poor prognosis . Given the pivotal role of
FGFR1 in human cancers, we sought to analyze and target the
pathway of cancer development through FGFR1 overexpression
and/or activation. Previous studies have shown the importance of
21
1
1–13
5.0 mM, whereas that of PD173074 was 1000 times higher .
Furthermore, PD173074 and SU5402 failed to enter phase II
ATP-binding of FGFR1 for its enzymatic activity . The RTKs
are catalyzed upon ATP binding to their enzymatic pockets, which
22,23
clinical trials because of their high toxicity
.
Herein we have adopted a nonconventional strategy to design a
lead compound (Figure 1), which contains more heteroatoms, and
resulted in the formation of an intermolecular hydrogen bond
in the ATP binding site. Such exchange of the side chain position
facilitated the interaction between the hydrophobic pocket of
Address for correspondence: Sicen Wang, Professor, School of Pharmacy,
Health Science Center Xi’an Jiaotong University, Xi’an 710061, China.
Tel: +86 29 82656788. Fax: +86 29 82655451. E-mail: wangsc@
mail.xjtu.edu.cn

2
F. Ye et al.
J Enzyme Inhib Med Chem, Early Online: 1–6
Figure 1. The rationale for the design of the target compounds.
ꢀ
FGFR1 and the compounds resulting in higher binding affinity. A and the solution was refluxed for 1 h at 60 C. After removing the
core scaffold 5,7-dimethyl-oxazolo[5,4-d]pyrimidine-4,6(5H,7H)- thionyl chloride by reduced pressure distillation, the mixture was
dione was selected based on the substructure similarity, which is dissolved in a little of water. Then adjust to pH 7 by ammonia
rarely reported in the field of RTK inhibitors with antitumor water. They were purified by further recrystallization with
activity. Additionally, the 2-position was also substituted with DMSO:ethanol ¼ 1:1 and dried to get the target compound 5,7-
different aromatic derivatives based on SAR. Our main goal was dimethyl-oxazolyl[5,4-d]pyrimidine-4,6(5H,7H)-dione
to determine whether the 5,7-dimethyl-oxazolo[5,4-d]pyrimidine- tives (N5a–l).
deriva-
4
,6(5H,7H)-dione derivatives were able to confer higher potency
2-(Furan-2-yl)-5,7-dimethyl-4H-oxazolo[5,4-d]pyrimidine-4,6
against cancer cells and to search for a new heterocyclic stem (5H,7H)-dione (N5a): Yield: 45.5%. White solid, mp: 185–
ꢀ
ꢁ1
1
nucleus that contributes to the design of novel antitumor drugs.
187 C. IR (KBr), cm : 1717, 1682 (C ¼ O); H NMR (CDCl )
3
5
(ppm): 3.50 (s, 3H, N –CH ), 3.70 (s, 3H, N –CH ), 6.58–6.63
7
ꢁ
(m, 1H, 4 furan-H), 7.14 (s, 1H, 3 furan-H), 7.60 (s, 1H, 5 furan-
3 3
0
0
0
Materials and methods
+
H). EI-MS: 248.2[M+H] .
Experimental section
2
Experimental: Melting points (uncorrected values): METTLER 4,6(5H,7H)-dione (N5b): Yield: 52.1%. White solid, mp: 198–
-(Thiophen-2-yl)-5,7-dimethyl-4H-oxazolo[5,4-d]pyrimidine-
ꢀ
ꢁ1
1
TOLEDO XRC-1 micro-melting point apparatus (Shanghai, 200 C. IR (KBr), cm : 1717, 1679 (C ¼ O). H NMR
5
China). Infrared Spectroscopy (IR): Perkin-Elmer FT-IR 1605 (DMSO-d ) ꢁ(ppm): 3.45 (s, 3H, N –CH ), 3.65 (s, 3H, N –
7
6
3
1
spectrometer (Nicolet, America). H NMR spectra: Bruker CH ), 7.15–7.22 (m, 1H, 4 thiophene-H), 7.52 (d, 1H,
0
3
0
0
Avance-3600 spectrometer (600 MHz) (Waldbronn, Germany). 3 thiophene-H, J ¼ 4.2 Hz), 7.75 (d, 1H, 5 thiophene-H,
+
Samples were dissolved in CDCl3 or d -DMSO, while tetra- J ¼ 4.2 Hz). EI-MS: 264.5[M+H] .
6
methylsilane (TMS) was used as an internal standard. Mass 2-(Pyridin-2-yl)-5,7-dimethyl-4H-oxazolo[5,4-d]pyrimidine-4,
spectra (MS): Agilent-1100 LC mass spectrometer (Waldbronn, 6(5H,7H)-dione (N5c): Yield: 54.7%. White solid, mp: 197–
ꢀ
ꢁ1
1
Germany). Analytical thin layer chromatography (TLC, R values) 200 C. IR (KBr), cm : 1718, 1622 (C ¼ O). H NMR (CDCl )
f
3
5
was performed on Silica Gel plates F254 (0.2-mm thick) and ꢁ(ppm): 3.47 (s, 3H, N –CH ), 3.71 (s, 3H, N –CH ), 7.16–7.22
7
3
3
0
0
visualization of the spots was effected by exposure to UV light. (m, 1H, 5 pyridine-H), 7.43–7.47 (m, 1H, 4 pyridine-H), 7.51–
0
0
These apparatus were provided by Wenzhou Medical University. 7.62 (m, 1H, 3 pyridine-H), 7.65 (s, 1H, 6 pyridine-H). EI-MS:
+
General procedure for the synthesis of 1,3-dimethyl-5-nitroso- 258.7[M+H] .
pyrimidine-2,4,6(1H,3H,5H)-trione (2): 1,3-Dimethyl-pyrimi-
dine-2,4,6(1H,3H,5H)-trione (1) (0.03 mol) was added to a imidine-4,6(5H,7H)-dione (N5d): Yield: 60.3%. White solid, mp:
2-[(4-Benzyloxy)phenyl]-5,7-dimethyl-4H-oxazolo[5,4-d]pyr-
ꢀ
ꢁ1
1
mixture of acetic acid (70 ml) and water (70 ml). The mix 220–222 C. IR (KBr), cm : 1714, 1667 (C ¼ O). H NMR
ꢀ
ꢀ
5
7
solution was stirred for 1 h at 75 C, and then cooled to 50 C. (CDCl ) ꢁ(ppm): 3.45 (s, 3H, N –CH ), 3.64 (s, 3H, N –CH ),
3
3
3
Then NaNO (0.06 mol) was added in batches to the reaction 5.13 (s, 2H, –OCH ), 7.06 (d, 2H, phenyl-H, J ¼ 9.0 Hz), 7.32–
2
2
mixture stirring for 1 h and recooled to room temperature. Then 7.38 (m, 1H, phenyl-H), 7.40 (d, 2H, phenyl-H, J ¼ 7.8 Hz), 7.43
the deposit was filtered and washed with water, dried, (d, 2H, phenyl-H, J ¼ 7.2 Hz), 7.9 (d, 2H, phenyl-H, J ¼ 9.0 Hz).
+
concentrated in vacuo to obtain the white solid product 1,3- EI-MS: 364.0[M+H] .
dimethyl-5-nitroso-pyrimidine-2,4,6 (1H, 3H, 5H)-trione (2) with
yields of 81%.
2-[(4-Piperidin-1-yl)phenyl]-5,7-dimethyl-4H-oxazolo[5,4-d]
pyrimidine-4,6(5H,7H)-dione (N5e): Yield: 46.7%. Light yellow
ꢀ
ꢁ1
1
General procedure for the synthesis of 1,3-dimethyl-5-amino- solid, mp: 221–223 C. IR (KBr), cm : 1717, 1676 (C ¼ O). H
pyrimidin-2,4,6(1H,3H,5H) –trione (3): Compound 2 (0.01 mol) NMR (CDCl ) ꢁ(ppm): 1.65–1.75 (m, 6H, piperidine-H), 3.20–
3
5
was added to a mixture of 14.5% NH OH (14 ml) and then 3.38 (m, 4H, piperidine-H), 3.40 (s, 3H, N –CH ), 3.60 (s, 3H,
4
3
7
Na S O (0.02 mol) was added in batches. The solution was N –CH ), 6.86 (d, 2H, phenyl-H, J ¼ 9.0 Hz), 7.87 (t, 2H, phenyl-
2
2
4
3
ꢀ
+
refluxed for 30 min at 50 C. After removing the ammonia water H, J ¼ 9.0 Hz). EI-MS: 341.2[M+H] .
by reduced pressure distillation, the filtrate was washed with (E)-2-(1-Phenylprop-1-en-2-yl)-5,7-dimethyl-4H-oxazolo[5,4-
water and dried to get the white solid product 1,3-dimethyl-5- d]pyrimidine-4,6(5H,7H)-dione (N5f): Yield: 55.2%. Light yellow
ꢀ
ꢁ1
1
amino-pyrimidin-2,4,6(1H,3H,5H)-–trione (3) in 80% yield.
General procedure for the synthesis of imines (4a–l): 3 NMR (CDCl ) ꢁ(ppm): 2.45 (s, 3H, –CH ), 3.48 (s, 3H, N –CH ),
solid, mp: 185–187 C. IR (KBr), cm : 1682, 1614 (C ¼ O). H
5
3 3 3
7
0.001 mol) and the aromatic aldehyde (0.001 mol) were added to 3.68 (s, 3H, N –CH ), 7.41–7.47 (m, 2H, phenyl-H), 7.52 (s, 1H,
(
3
ꢀ
+
1
cooling to 0 C. The filtrate was washed with water and dried to
5 ml ethanol. Then the mixture was refluxed for 3 h at 80 C and CH ¼ C), 7.54–7.65 (m, 3H, phenyl-H). EI-MS: 298.0[M+H] .
ꢀ
(E)-2-(4-Methyl-1-phenylpent-1-en-1-yl)-5,7-dimethyl-4H-
produce intermediate white solid product (4a–l) in 50–70% yield. oxazolo[5,4-d]pyrimidine-4,6-(5H, 7H)-dione (N5g): Yield:
ꢀ
ꢁ1
General procedure for the synthesis of imines (N5a–l): 43.2%. White solid, mp: 184–185 C. IR (KBr), cm : 1719,
1
Compounds 4a–l (0.01 mol) were added to 1.5 ml thionyl chloride 1638 (C ¼ O). H NMR (DMSO-d ) ꢁ(ppm): 0.97–1.10
6

DOI: 10.3109/14756366.2014.1002401
5,7-Dimethyl-oxazolo[5,4-d]pyrimidine-4,6(5H,7H)-dione derivatives
3
(
CH ), 3.25 (s, 3H, N –CH ), 3.95 (s, 3H, N –CH ), 7.35 kinase was detected by the EnVision Multilabel Reader and
m, 6H, CH ꢂ2), 1.78–1.85 (m, 1H, CH), 2.11–2.18 (m, 2H, assay plate and incubated for 1 h. The activity of the tyrosine
3
5
7
2
3
3
(
d, 2H, phenyl-H, J ¼ 7.8 Hz), 7.48 (t, 1H, phenyl-H, controlled by SPSS.
J ¼ 7.8 Hz), 7.52 (s, 1H, CH ¼ C), 7.88 (t, 2H, phenyl-H,
J ¼ 7.8 Hz).
MTT assay for antitumor activity in cellular level
2-(3a,7a-Dihydro-1H-indol-3-yl)-5,7-dimethyl-4H-oxazolo[5,4-d]
pyrimidine-4,6(5H,7H)-dione (N5h): Yield: 60%. White solid, mp:
All of tested cancer cells were obtained from Pilot base of
Wenzhou Medical College and grown in Dulbecco’s Modified
Eagle Medium with high glucose. And these media were
supplemented with 10% fetal bovine serum, 50 U/ml penicillin
and 50 mg/ml streptomycin. All of the tissue culture reagents were
ꢀ
ꢁ1
2
98–300 C. IR (KBr), cm : 3227.94 (N–H), 1710, 1674 (C ¼ O).
1
5
H NMR (DMSO-d ) ꢁ(ppm): 3.38 (s, 3H, N –CH ), 3.55 (s, 3H,
6
3
7
N –CH ), 7.15–7.35 (m, 2H, indole-H), 7.53 (d, 1H, indole-H,
3
J¼ 7.2 Hz), 8.14 (d, 1H, indole-H, J¼ 7.2 Hz), 8.20 (s, 1H, indole-
+
H), 11.97 (s, 1H, indole-H). EI-MS: 297[M+H] .
purchased from Gino Biomedical Technology Co, Ltd. and
ꢀ
cultivated at 37 C in a humidified incubator with 5% CO . At the
2
2-(Naphthalen-1-yl)-5,7-dimethyl-4H-oxazolo[5,4-d]pyrimidine-
,6(5H,7H)-dione (N5i): Yield: 49.1%. White solid, mp: 276–
beginning of the assay, these cells (5000/well) were, respectively,
cultured in a 96-well microplate for 1 day. Then the cells were
treated with 100 ml of new medium (SU5402 as positive control
group 0.2, 4, 10, 50 mg/ml) containing the test substances. After
incubation for 24 h, 20 mg/ml of 3 -(4,5-dimethylthiazol-2-yl)-2,5-
diphenyl-2-H-tetrazolium bromide (MTT) solution (5 mg/ml) was
added to each well and the cells were incubated for an additional
4
2
ꢀ
ꢁ1
1
79 C. IR (KBr), cm : 1715, 1669 (C ¼ O). H NMR
6 3
5
DMSO-d ) ꢁ(ppm): 3.34 (s, 3H, N –CH ), 3.58 (s, 3H,
(
N –CH ), 7.53–7.69 (m, 2H, naphthnalene-H), 7.99–8.01
(
7
3
m, 1H, naphthnalene-H), 8.11–8.23 (m, 3H, naphthnalene-H),
+
8
.66 (s, 1H, naphthnalene-H). EI-MS: 308[M+H] .
-(2-Hydroxy-naphthalen-1-yl)-5,7-dimethyl-4H-oxazolo[5,4-
d]pyrimidine-4,6(5H,7H)-dione (N5j): Yield: 57.6%. White solid,
2
4
h. Finally, the medium was removed and precipitates were
ꢀ
ꢁ1
1
dissolved in 150 ml DMSO with mechanically shaking for 30 min.
The absorbance values at the wavelength of 570 nm were
collected by the ELX800 microplate reader while IC50 values
were calculated by using percentage of growth versus untreated
control by SPSS (Chicago, IL).
mp: 280–282 C. IR (KBr), cm : 1718, 1673 (C ¼ O). H NMR
5
CDCl ) ꢁ(ppm): 3.50 (s, 3H, N –CH ), 3.80 (s, 3H, N –CH ),
3 3 3
.32 (d, 1H, naphthnalene-H, J ¼ 9.0 Hz), 7.34–7.44 (m, 1H,
7
(
7
naphthnalene-H), 7.63–7.77 (m, 1H, naphthnalene-H), 7.84
d, 1H, naphthnalene-H, J ¼ 9.0 Hz), 7.89 (d, 1H, naphthnalene-
(
H, J ¼ 9.0 Hz), 8.59 (d, 1H, naphthnalene-H, J ¼ 9.0 Hz), 11.85
+
(
s, 1H, –OH). EI-MS: 323.9[M+H] .
-(Anthracen-9-yl)-5,7-dimethyl-4H-oxazolo[5,4-d]pyrimidine-
,6(5H,7H)-dione (N5k): Yield: 56.6%. White solid, mp: 275–
Result and discussion
2
Chemistry
4
2
ꢁ
ꢀ
ꢁ1
1
78 C. IR (KBr), cm : 1697, 1684 (C ¼ O). H NMR (CDCl ) Based on the core scaffold of 5,7-dimethyl-oxazolo[5,4-d]pyrimi-
3
5
(ppm): 3.58 (s, 3H, N –CH ), 3.66 (s, 3H, N –CH ), 7.55–7.68
7
3 3
dine-4,6(5H,7H)-dione, we designed and synthesized 12 novel
(
8
m, 4H, anthracene-H), 8.04 (d, 2H, anthracene-H, J ¼ 7.8 Hz), compounds N5a–N5l, and confirmed their chemical structures by
1
.12 (d, 2H, anthracene-H, J ¼ 7.8 Hz,), 8.71 (s, 1H, anthracene-
H NMR, IR and MS. For the spectral data on N5a–N5l, see the
Supplementary Material. Furthermore, the antitumor activities of
these compounds were evaluated by using LANCE Ultra TR-
+
H); EI-MS: 358.2[M+H] .
,7-Dimethyl-2-quinolin-2-yl-4H-oxazolo[5,4-d]pyrimidine-4,
5
6
(5H,7H)-dione (N5l): Yield: 59.7%. White solid, mp: 297– FRET and MTT Assay.
ꢀ
ꢁ1
1
2
99 C. IR (KBr), cm : 1717, 1665 (C ¼ O). H NMR
6 3
The synthetic route is illustrated in Figure 2. The key
5
7
(DMSO-d ) ꢁ(ppm): 3.40 (s, 3H, N –CH ), 3.66 (s, 3H, N – intermediate of 1,3-dimethyl-5-nitroso-pyrimidine-2,4,6(1H, 3H,
CH ), 7.58–7.65 (m, 2H, quinoline-H), 7.42 (d, 2H, quinoline-H, 5H)-trione (2) was obtained from commercially available 1,3-
3
J ¼ 7.8 Hz), 8.04 (d, 1H, quinoline-H, J ¼ 7.8 Hz), 8.30 (s, 1H, dimethyl- pyrimidine-2,4,6(1H, 3H, 5H)-trione (1) via nitrosation
+
quinoline-H). EI-MS: 308.9[M+H] .
with NaNO2 at 81% yield. Reduction of intermediate 2 with
Na S O
2
to
furnish
1,3-dimethyl-5-amino-pyrimidin-
,4,6(1H,3H,5H)-trione (3) at 80% yield, which was subjected
2
2 4
FGFR kinase inhibition assay
to react with various aromatic aldehydes in the presence of
ethanol to obtain the corresponding imines (4a–4l) at 50–70%
yield. Cyclization of the imines (4a–4l) with thionyl chloride
provided the target compounds, 5,7-dimethyl-oxazolyl[5,4-d]pyr-
imidine-4,6 (5H,7H)-dione derivatives (N5a–5l), at 43.2–60.3%
yield. All of the synthetic products were isolated by conventional
work-up with satisfactory yields. Analytical and spectral data of
all synthesized compounds are in full agreement with the
proposed structures.
FGFR1 kinase activity was determined by LANCE Ultra TR-
FRET Assay. For the determination of IC , the compounds were
5
0
tested in duplicate at three concentrations (10, 1 and 0.1 mM). A
standard enzymatic reaction, initiated by transferring 5 ml of
compound to one tube and adding of 95 ml of 1 ꢂ kinase buffer
(
50 mM HEPES, pH 7.5, 10 mM MgCl , 1 mM EGTA, 2 mM
2
DTT, 0.01% Tween-20) in the same tube. As a control, 5 ml of
1
00% DMSO was transferred to one tube and 95 ml of 1 ꢂ kinase
buffer was added in the same tube (10-fold dilution by 1 x kinase
buffer). Then 2.5 ml from each tube was transferred to the 384-
well assay plate, which contains 4 ꢂ compounds and positive
control SU5402 at the concentration of 10, 1 and 0.1 mM,
Compound N5d, N5e and N5f are potent FGFR1 kinase
inhibitors
respectively. Then 2.5 ml of 4 x enzyme solution was added to each To determine the effect of compounds N5a–N5l on FGFR1 kinase
well of the assay plate. After 10-min incubation at room activity, we performed LANCE Ultra TR-FRET Assay. As shown
temperature, 5 ml of 2.5 ꢂ peptide solution (ULight-JAK-1 peptide in Figure 3, a dose-dependent inhibition was observed for most of
and ATP in 1 ꢂ kinase base buffer) was transferred to each well of the compounds toward FGFR1, among which compounds N5d,
the 384-well assay plate followed by additional 60-min incubation N5e, N5f and N5g showed highest potency. Importantly, these
ꢀ
at 28 C. The reaction was stopped by addition of the termination compounds also exhibited good potency at the concentration of
buffer (5 ml of 40 mM EDTA) for 5 min. Finally, 5 ml of 4X 0.1 and 1 mM comparable to SU5402. We postulate that the result
antibody (Eu-anti-phospho-tyrosine Antibody (PT66) at a final could be due to the space structure of ATP site of FGFR1, which
concentration of 2 nM) was added to each well of the 384-well was big enough to contain larger molecules and formed more

4
F. Ye et al.
J Enzyme Inhib Med Chem, Early Online: 1–6
Figure 2. General route for the synthesis of
,7-dimethyloxazolo[5,4-d]pyrimidine-4,6
5H,7H)-dione derivatives. Regents and con-
5
(
ditions: (a) NaNO
2
, CH
, NH
CH
3
COOH:H
2
O ¼ 1:1,
ꢀ
ꢀ
ꢀ
5
3
0 C, 1 h; (b) Na
2
S
2
O
4
4
OH, 50 C,
0 min; (c) R–CHO, CH
3
2
OH, 80 C, 3 h;
ꢀ
(
d) SOCl , 60 C, 1 h, pH 7.
2
Figure 3. Inhibition of FGFR1 by the tested
compounds in vitro. The activity of all
synthesized compounds at indicated concen-
trations toward in inhibiting FGFR1 kinase
activity was determined by the LANCE Ultra
TR-FRET Assay. The data was collected on
the EnVision Multilabel Reader and con-
trolled by SPSS.
bonds with these molecules. Among the tested compounds, N5e Table 1. The IC50 values for the tested compounds on cell proliferation
showed the best performance at the concentration of 10 mM, based on MTT assay.
which was more potent than SU5402. The piperidyl group of
compound N5e appears to be able to form more hydrogen bonds
with the ATP site and occupy the hydrophobic pocket with the
long side chain. The other compound N5g also showed a strong
activity to the FGFR1 kinase at a comparable level to the positive
control SU5402.
IC50 (mM)
Compound
H460
B16F10
A549
41000
311.9 ± 24.2
166.3 ± 16.4
41000
424.1 ± 20.13
16.8 ± 1.06
5.837 ± 3.20
29.77 ± 1.03
372.9 ± 2.13
132.1 ± 3.58
868.1 ± 12.42
30.94 ± 1.56
3.483 ± 1.47
N5a
N5b
N5c
N5d
N5e
N5f
7.9 ± 1.05
608.5 ± 18.4
41000
57.85 ± 2.54
41000
21.36 ± 0.53
1.47 ± 0.53
59.8 ± 2.57
57.5 ± 5.67
41000
10.8 ± 0.34
60.71 ± 1.03
2.217 ± 2.01
41000
311.7 ± 10.1
41000
41000
40.18 ± 1.53
15.26 ± 0.85
4.260 ± 2.61
35.1 ± 0.24
35.6 ± 1.56
41000
32.8 ± 1.34
292.6 ± 5.41
1.262 ± 0.960
The antitumor activity of compounds N5a–N5l in vitro
These synthesized compounds (N5a–5l) were further screened for N5g
their antitumor activities in vitro. Three cancer cell lines known to N5h
express high levels of FGFR1 were tested in MTT assay including
the mouse melanoma (B16F10) and human lung cancer (H460
and A549) and the result is summarized in Table 1.
It has been reported that high FGFR1 expression contributes
N5i
N5j
N5k
N5l
SU5402
to cell proliferation in both H460 and A549 human lung cancer
2
4,25
cells
. Given that our compounds demonstrated inhibition
effect on FGFR1 kinase activity, we performed MTT assay to (Table 1). Compounds N5f, N5g, N5h and N5i demonstrated
determine the potential effect on cell proliferation. Out results promising activity toward nearly all of the three cells (H460,
revealed that several of the compounds showed antitumor activity, B16F10 and A549). Interestingly, compound N5g showed an
which seemingly corresponded to the inhibition of FGFR1 in vitro anticancer activity with IC50 values below 6.0 mM, closed

DOI: 10.3109/14756366.2014.1002401
5,7-Dimethyl-oxazolo[5,4-d]pyrimidine-4,6(5H,7H)-dione derivatives
5
Figure 4. Computed binding geometry of the new inhibitors N5e (A) and N5g (B) with the FGFR1 ATP-binding pocket.
to that of the positive control drug (SU5402). The reason may be
Varying selectivity toward cancer cell proliferation was
that compound N5g has more hydrophobic groups than N5d–f, so observed for these compounds. N5a and N5d affected only the
that the former could be more membrane permeable. Additionally, H460 cells, whereas N5f, N5g, N5h and N5i exhibited good
N5a and N5k affected only H460 cells, whereas N5e showed good activity against all three lines. Among the compounds tested, N5g
activity toward B16F10.
was found to exhibit the most potent and universal anti-
An evident loss in inhibitory activity was observed when the proliferative activity in all cancer cell lines similar to the positive
side chain was substituted with a single ring, such as in N5a, N5b drug SU5402. We suggest that the 5,7-dimethyl-oxazolo[5,4-
and N5c, suggesting that the group is too small to occupy the d]pyrimidine-4,6(5H,7H)-dione derivatives may serve as potential
hydrophobic pocket or form a bond with FGFR1. SAR agents for the treatment of FGFR1-mediated cancers. Therefore,
also demonstrated that the bulky and flexible groups might N5g could be a more promising candidate compound for further
play the most crucial role both in regulating the kinase activity modification and anti-tumor activity study.
and the cellular activity. Additionally, the bulky group likely
forges a tight link to the hydrophobic pocket to hinder the
Acknowledgements
movement of the compound side chain out of the hydrophobic
1
The IR, H NMR, ESI-MS and elemental analysis were recorded by the
pocket. Consistent with this, N5f, N5g, N5h and N5i exhibit more
Analysis and Test Center of Wenzhou Medical University.
promising activity.
Declaration of interest
Conclusion
This work was supported by the National Natural Science
Foundation of China (81373262).
To explain the results, we proposed a hypothetical model for N5g
and N5e binding to FGFR1 based on computational docking
results (Figure 4). As usual, the oxygen atom of a carbonyl group
on N5g bound to the hydrogen of GLU562 to form hydrogen bond
˚
(
distance cutoff: 2.6 A) in the active pocket, while side chain (Z)- References
(
4-methylpent-1-en-1-yl) benzene in the designed compound
1
.
Van-der-Noll R, Leijen S, Neuteboom GH, et al. Effect of inhibition
of the FGFR-MAPK signaling pathway on the development of ocular
toxicities. Cancer Treat Rev 2013;39:664–72.
could be placed into the hydrophobic pocket formed by
SER565, PHE489, GLY567, LEU630, LEU484, ALA488,
GLY485, GLY487 and GLU486, and generates p–p stacking
interactions with the PHE489. The docking of N5e to FGFR1 was
similar to that of N5g, but form two hydrogen bonds (distance
2. Guagnano V, Furet P, Spanka C, et al. Discovery of 3-(2,6-dichloro-
3,5-dimethoxy-phenyl)-1-{6-[4-(4-ethyl-piperazin-1-yl)-phenyla-
mino]-pyrimidin-4-yl}-1-methyl-urea (NVP-BGJ 398), a potent and
selective inhibitor of the fibroblast growth factor receptor family of
receptor tyrosine kinase. J Med Chem 2011;54:7066–83.
˚
cutoff: 3.0 and 3.1 A) in the pocket. The results of the docking
depicted that the energy of the two compounds, N5e and N5g
binding with FGFR1 were similar (8.16 and 7.48 kcal/mol,
respectively). Therefore the binding geometry provides potential
explanations for the more potent antitumor activity associated
with the multiple hybrid compounds.
3
.
Takeuchi K, Ito F. Receptor tyrosine kinases and targeted cancer
therapeutics. Biol Pharm Bull 2011;34:1774–80.
4. Kottakis F, Polytarchou C, Foltopoulou P, et al. FGF-2 regulates cell
proliferation, migration, and angiogenesis through an NDY1/
KDM2B-miR- 101-EZH2 pathway. Mol Cell 2011;43:285–98.
5
.
Turner N, Grose R. Fibroblast growth factor signalling: from
development to cancer. Nat Rev Cancer 2010;10:116–29.
Gangjee A, Zhao Y, Raghavan S, et al. Design, synthesis and
evaluation of 2-amino-4-m-bromoanilino-6-arylmethyl-7H-pyr-
rolo[2,3-d]pyrimidines as tyrosine kinase inhibitors and antiangio-
genic agents. Bioorg Med Chem 2010;18:5261–73.
In summary, 12 novel 5,7-dimethyl-oxazolo[5,4-d]pyrimidine-
,6(5H,7H)-dione derivatives were designed, synthesized and
4
6
.
evaluated for their antitumor activity both in various cancer cells
with high FGFR1 kinase activity. Although most compounds
showed a linear relationship in inhibiting FGFR1 kinase activity,
the compounds N5e and N5g showed better selectivity toward
FGFR1 suggesting that the compounds with side chains
substituted with stereo-structure groups exhibited better activity
than those with planar structures. We speculate that this could be
due to the space structure of FGFR1 ATP pocket, which was big
enough to accommodate larger molecules and form more bonds
with these molecules.
7. Chen Z, Wang X, Zhu W, et al. Acenaphtho[1,2-b]pyrrole-based
selective fibroblast growth factor receptors 1 (FGFR1) inhibitors:
design, synthesis, and biological activity. J Med Chem 2011;54:
3
732–45.
8
.
Yang F, Zhang Y, Ressler SJ, et al. FGFR1 is essential for prostate
cancer progression and metastasis. Cancer Res 2013;73:3716–24.
9. Weiss J, Sos ML, Seidel D, et al. Frequent and focal FGFR1
amplification associates with therapeutically tractable FGFR1

6
F. Ye et al.
J Enzyme Inhib Med Chem, Early Online: 1–6
dependency in squamous cell lung cancer. Sci Transl Med 2010;2: 18. Skaper SD, Kee WJ, Facci L, et al. The FGFR1 inhibitor PD 173074
6
0. Turner N, Pearson A, Sharpe R, et al. FGFR1 amplification drives
2–93.
selectively and potently antagonizes FGF-2 neurotrophic and
neurotropic effects. Neurochem 2000;75:1520–7.
endocrine therapy resistance and is a therapeutic target in breast 19. Mohammadi M, McMahon G, Sun L, et al. Structures of the tyrosine
1
1
1
cancer. Cancer Res 2010;70:2085–94.
1. Liang G, Chen G, Wei X, et al. Small molecule inhibition of
kinase domain of fibroblast growth factor receptor in complex with
inhibitors. Science 1997;276:955–60.
fibroblast growth factor receptors in cancer. Cytokine Growth Factor 20. Thompson AM, Delaney AM, Hamby JM, et al. Synthesis and
Rev 2013;24:467–75.
structure–activity relationships of soluble 7-substituted 3-(3,5-dim-
ethoxyphenyl)-1,6-naphthyridin-2-amines and related ureas as dual
inhibitors of the fibroblast growth factor receptor-1 and vascular
endothelial growth factor receptor-2 tyrosine kinases. J Med Chem
2005;48:4628–53.
2. Ravindranathan KP, Mandiyan V, Ekkati AR, et al. Discovery of
novel fibroblast growth factor receptor 1 kinase inhibitors by
structure-based virtual screening. J Med Chem 2010;53:1662–72.
3. Fearon AE, Gould CR, Grose RP. Int. FGFR signalling in women’s
cancers. J Biochem Cell Biol 2013;45:2832–42.
4. Dokduang H, Juntana S, Techasen A, et al. Survey of activated
kinase proteins reveals potential targets for cholangiocarcinoma
treatment. Tumour Biol 2013;34:3519–28.
5. Byron SA, Chen H, Wortmann A, et al. The N550K/H mutations in
FGFR2 confer differential resistance to PD173074, dovitinib, and
ponatinib ATP-competitive inhibitors. Neoplasia 2013;15:975–88.
6. Patel A, Tiwari AK, Chufan EE, et al. PD173074, a selective FGFR
1
1
21. Grand EK, Chase AJ, Heath C, et al. Targeting FGFR3 in multiple
myeloma: inhibition of t(4;14)-positive cells by SU5402 and
PD173074. Leukemia 2004;18:962–6.
22. Wang Y, Cai Y, Ji J, et al. Discovery and identification of new non-
ATP competitive FGFR1 inhibitors with therapeutic potential on
non-small-cell lung cancer. Cancer Lett 2014;344:82–9.
1
1
1
23. Kumar SB, Narasu L, Gundla R, Dayam R, JARPS. Fibroblast
growth factor receptor inhibitors. Curr Pharm Des 2013;19:687–701.
inhibitor, reverses ABCB1-mediated drug resistance in cancer cells. 24. Yang J, Zhao H, Xin Y, Fan L. MicroRNA-198 inhibits proliferation
Cancer Chemother Pharmacol 2013;72:189–99.
7. Gryshchenko AA, Levchenko KV, Bdzhola VG, et al. Design,
and induces apoptosis of lung cancer cells via targeting FGFR1.
J Cell Biochem 2014;115:987–95.
synthesis and biological evaluation of naphthostyril derivatives as 25. Wu X, Huang H, Wang C, et al. Identification of a novel peptide that
novel protein kinase FGFR1 inhibitors. J Enzyme Inhib Med Chem
014;18:1–7.
blocks basic fibroblast growth factor-mediated cell proliferation.
Oncotarget 2013;4:1819–28.
2
Supplementary material available online