The design and synthesis of 9-phenylcyclohepta[d]pyrimidine-2,4-dione derivatives as potent non-nucleoside inhibitors of HIV reverse transcriptase

Article abstract of DOI:10.1039/b607972p

Novel compounds, which can be considered as conformationally restricted analogues of MKC-442, have been synthesized and tested as inhibitors of the reverse transcriptase of human immunodeficiency virus type-1 (HIV-1). Reaction of urea with a β-ketoester furnished 6,7,8,9-tetrahydro-9-phenyl-1H- cyclohepta[d]pyrimidine-2,4-(3H,5H)-dione (6a) and 6,7,8,9-tetrahydro-9-p-tolyl- 1H-cyclohepta[d]pyrimidine-2,4-(3H,5H)-dione (6b) which were then alkylated at the N-1 position with chloromethyl ether, allyl bromide and benzyl bromide to afford the target compounds 7a-b, 8a-b, 9 and 10, respectively. The seven-membered, annelated compounds have a relatively rigid structures and can lock the orientation of the aromatic ring. Chemical modification at N-1 of the pyrinidine ring and the 9-phenyl ring was attempted, with the aim of improving the antiretroviral activity. In particular, replacement of the aliphatic group with the phenyl moiety at the terminus of N-1 side chain can enhance the activity. The most active compounds showed activity in the low micromolar range with IC₅₀ values comparable to that of nevirapine. The biological activity results are in accordance with the docking results. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b607972p

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
The design and synthesis of 9-phenylcyclohepta[d]pyrimidine-2,4-dione
derivatives as potent non-nucleoside inhibitors of HIV reverse transcriptase
Xiaowei Wang,*a Qinghua Lou,a Ying Guo,b Yang Xu,a Zhili Zhanga and Junyi Liu*a
Received 5th June 2006, Accepted 19th June 2006
First published as an Advance Article on the web 28th July 2006
DOI: 10.1039/b607972p
Novel compounds, which can be considered as conformationally restricted analogues of MKC-442,
have been synthesized and tested as inhibitors of the reverse transcriptase of human immunodeficiency
virus type-1 (HIV-1). Reaction of urea with a b-ketoester furnished 6,7,8,9-tetrahydro-9-
phenyl-1H-cyclohepta[d]pyrimidine-2,4-(3H,5H)-dione (6a) and 6,7,8,9-tetrahydro-9-p-tolyl-1H-
cyclohepta[d]pyrimidine-2,4-(3H,5H)-dione (6b) which were then alkylated at the N-1 position with
chloromethyl ether, allyl bromide and benzyl bromide to afford the target compounds 7a–b, 8a–b, 9 and
10, respectively. The seven-membered, annelated compounds have a relatively rigid structures and can
lock the orientation of the aromatic ring. Chemical modification at N-1 of the pyrinidine ring and the
9-phenyl ring was attempted, with the aim of improving the antiretroviral activity. In particular,
replacement of the aliphatic group with the phenyl moiety at the terminus of N-1 side chain can
enhance the activity. The most active compounds showed activity in the low micromolar range with
IC50 values comparable to that of nevirapine. The biological activity results are in accordance with the
docking results.
Introduction
Reverse transcriptase (RT), being the pivot in HIV replication,
is one of the most attractive targets for the development of new
antiretroviral agents.1–3 Two functionally-distinct classes of HIV-
1 RT inhibitors have been described so far: nucleoside reverse
transcriptase inhibitors (NRTIs)4–9 and non-nucleoside reverse
transcriptase inhibitors (NNRTIs).10–16 In recent years, NNRTIs
have gained an increasingly important role in the therapy of HIV
infection. To date, more than 30 different classes of NNRTIs have
been reported. Among the representatives of the NNRTIs, 1-[(2-
hydroxyethoxy)methyl]-6-(phenylthio)thymine (HEPT,17–18 Fig. 1)
was considered as an interesting lead compound for the synthesis
of new analogues, among them the 6-benzyl-1-(ethoxymethyl)-
5-isopropyl pyrimidine-2,4-dione (emivirine, formerly known
as MKC-442)19–21 and 6-benzyl-1-(benzyloxymethyl)-5-isopropyl
pyrimidine-2,4-dione (TNK-651)22 showed high activity against
HIV-1, and MKC-442 was chosen as a candidate for clinical trials
with AIDS patients.23 However, Triangle Pharmaceuticals halted
development of emivirine in January 2002 for being less potent
than other antiretrovirals.24 Like almost all NNRTIs, the HEPT
analogues rapidly develop drug resistance by mutation of residues
that line the NNRTI binding pocket of the viral RT and reduce
drug binding. Their clinical use is limited, even though they are
more active against HIV-1 RT with generally low toxicity and
favorable pharmacokinetic properties. Therefore, in the NNRTIs
field, interest is focused on finding new analogues with higher
Fig. 1 Structure of HEPT, MKC-442, TNK-651. Compounds I and II:
structures of previously synthesized by Claus Nielsen et al.; compounds
7a–b and 8a–b: structures of target molecules.
aDepartment of Chemical Biology, School of Pharmaceutical Science, Peking
University, Beijing, 100083, China. E-mail: xiaowei0301@china.com.cn;
Tel: 86 010 82801504
bState Key Laboratory of Natural and Biomimetic Drugs, Peking University,
Beijing, 100083, China. E-mail: jyliu@bjmu.edu.cn; Tel: 86 010 82801706
binding affinity and the capability of inhibiting clinically resistant
mutants.25–28
3252 | Org. Biomol. Chem., 2006, 4, 3252–3258
This journal is
The Royal Society of Chemistry 2006
©

View Online
Results and discussions
Design
According to structure–activity relationships (SARs), studies of
several crystal structures of the RT complex with inhibitors
suggest that NNRTIs share a common mode of action and
interact with a hydrophobic pocket that is distinct from the NRTI
binding site, the polymerase catalytic site.29–34 Upon binding,
NNRTIs distort the polymerase active site and alter the con-
formation of Tyr181, Tyr183 and Tyr188 residues with respect
to that of the unliganded RT. Interestingly, the two HEPT
analogues (MKC-442, TNK-651), with their bulky isopropyl
groups, force Tyr181 into a position where strong interactions
with the 6-benzyl group of the inhibitor can take place (“trigger”
action).22,31,33 On the contrary, HEPT does not produce any
“trigger” effect. Its weak anti-HIV-1 activity can be explained
by a slight perturbation of the Tyr181 side-chain conformation
produced by the C-5 methyl group. This revealed that the C-5
substituent of the pyrimidine ring is very importance for antiviral
activity.22
Although extensive SARs for non-cyclic analogues (such as
MKC-442 and TNK-651) have been obtained, little or no in-
formation was available about the annelated series. In the year
2000, Claus Nielsen et al. synthesized the five- and six-annelated
compounds I and II (Fig. 1), the conformation of which was
restricted by locking the orientation of the aromatic ring, but
the activity was ca. 103-fold lower than MKC-442.35 Later, they
synthesized compounds with extra methyl groups next to the
C5 position of the uracil to increase the steric bulkiness, but
the biological activities were only slightly better than those for
I and II36 (Fig. 1). In addition, Macchia et al. have reported that
tetrahydrobenzocycloheptenuracils do not have higher anti HIV-1
virus activities.37
After inspection of the reports mentioned above, we decided
that the key point for the development of such compounds is the
right size of the annelated rings, which should not only posses
conformationally flexibility, but should also restrict effectively the
rotation and position of the aromatic ring to the plane of the
pyrimidine ring. The cycloheptyl group represents just these two
characteristics.
Scheme 1 Reagents and conditions: (i) ClCOOC2H5, NaOH–H2O,
10–15 ◦C (ii) NaNO2, H2SO4, rt.; (iii) freshly distilled cyclohexanone,
anhydrate K2CO3 in dry CH3OH, 20 ◦C; (iv) CO(OC2H5)2, NaH,
benzene, reflux
5 h; (v) NH2CONH2, EtONa–EtOH, reflux 6 h;
To support our design, seven-membered ring analogues 7a–b,
8a–b (Fig. 1), 9 and 10 (Scheme 1) were designed and synthesized.
These compounds were flexibly docked into the binding site of
HIV-1 RT complexed with TNK-651, using the AutoDock 3.0
program.38 The first-ranking docked conformations are shown
in Fig. 2. It can be seen that the binding modes proposed by
Autodock for 7a and 8a (with the R configuration) are fairly close
to each other and have approximately the same conformation as
in the TNK-651-RT complex. The docking results revealed that
the target compounds bind to a hydrophobic pocket in the RT
with an conformation forming hydrogen bonds with the main-
chain of Lys103, while the 9-phenyl ring of 7a and 8a adopts a
fashion similar to that of the TNK-651, making favorable p–p
interactions with the residues of the RT allosteric pocket formed
by the side chains of Try181, Try188, Phe227, and Trp229. The
N-1 side chain of 8a has an extended conformation and occupies
a more hydrophobic pocket. Thus the bulky substituted groups
on the side chain of N-1 may be accommodated and improve the
activity.
(vi) BSA, chloromethyl ether, CHCl3, rt.; (vii) allyl bromide (benzyl
bromide), anhydrate K2CO3 in DMF, rt.
Furthermore, we have synthesized compounds 7b and 8b which
have a methyl substituent on the 9-phenyl ring. This substitution
pattern is similar to that of GCA-186. The structural change
of GCA-186 tolerates the presence of Y181C or K103N RT
mutations better than MKC-442 itself.39 The target compounds
(7a–b, 8a–b) with a locked torsion angle may decrease the toxicity.
Chemistry
A new conventional strategy of synthesizing seven-membered,
annelated analogues is described in Scheme 1, in which the b-
ketoesters 5a–b are the key intermediates. Compound 5a was
synthesized in four steps from benzylamine 1a.40 First 1a was
converted easily to the ethyl N-nitrosocarbarnate 3a by treatment
with ethyl chlorocarbonate followed by nitrosation with nitrous
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 3252–3258 | 3253
©

View Online
reacted with thiourea in the presence of sodium ethoxide to
give a very low yield (4%). All efforts were made to improve
the yield, but without any success. Then they looked for an
alternative synthetic route in three steps by reaction of 2-(s-
methylthio)isourea with the b-ketoester in aqueous KOH to
afford the 2-amino-1,3-oxazine-4(3H)-ones and under acidic con-
ditions to give the 7-phenyl-6,7-dihydrocyclopenta[e][1,3]oxazine-
2,4(3H,5H)-dione which was then converted to the corresponding
pyrimidine with concentrated aqueous ammonia.
We thought the key problem may be the ring strain. Therefore,
application of the Danel method to our compounds with a
seven-numbered ring would be feasible. In order to achieve the
uracil ring, we have investigated a one-step process to synthesize
cyclohepta[d]pyrimidine-2,4-dione (6). In our knowledge, the
use of a one-step process to synthesize pyrimidine rings is
infrequent. The b-ketoester 5 condensed directly with urea in the
presence of sodium methoxide and the corresponding pyrimidine
6 was precipated by acidifying with dilute hydrochlororide in
an aqueous solution. and recrystalized from ethanol in yields
of 46.0–52.0%. The advantage of the above-mentioned reac-
tion pathway was its simplicity, because the procedures were
undertaken conveniently and the materials were commercially
available.
The preparation of N-1 substituted uracils deserves some
comment. When 6a–b were treated with N,O-bis(trimethyl-
silyl)acetamide (BSA) in CHCl3 at room temperature, reaction
with alkyl chloromethyl ether gave 7a–b or 8a–b in good yield. In
contrast, a similar reaction with other alkyl halides, such as allyl
bromide and benzyl bromide, gave the corresponding products in
low yield (<10%).
The yields were improved by applying an alternative route.
Treatment of 6a–b with allyl bromide and benzyl bromide,
respectively, using potassium carbonate as an alkali in anhydrous
DMF, gave the mixture of N-1, N-3 substituted and disubstituted
products. Because N-1 is more nucleophilic than N-3, the N-
1 substituted compound can be made the major product by
controlling the ratio of reagents and the reaction temperature and
time. The addition of a catalytic amount of KI can shorten the
time and improve the yield. The structure assignments of these
compounds were identified by NMR and mass spectral data as
well as a single crystal X-ray structure analysis for compound 7a
(Fig. 3).
The N-1 alkylated compound was confirmed by 1H-nuclear
Overhauser effects (NOE) (Fig. 4), as irradiation of 9-H resulted
in NOE in the N–CH2–Rꢀ groups. Furthermore, being affected by
the 9-phenyl ring and the 9-chiral carbon, the 1H NMR spectrum
showed N–CHaHb–Rꢀ with different chemical shifts at d 4.74, 5.54
(d, each, J = 11.4 Hz).
Fig. 2 AutoDock-predicted binding mode of compounds 7a (a), 8a (b)
(green) compared to the crystallized position of TNK-651 (yellow). Only
closely interacting residues are shown.
acid in a quantitative yield. Reaction of 3a with freshly distilled
cyclohexanone in the presence of methanol and potassium carbon-
ate was then carried out, and the product was chromatographed by
silica gel column to give 4a in a 59.0% yield, which is higher than
that of the literature reported 41%. The side product was methyl
benzyl ether, the yield of which was affected by the conditions.
The preparation of several medium- and large-sized b-ketoesters
has been accomplished by treatment of the cycloalkanone with
sodium triphenylmethyl, followed by carbonation with dry ice,
and esterification with diazomethane. The procedure is laborious.
Another synthetic method, via the Dieckmann cyclization of
dimethyl azelate with sodium hydride, yields 48% of this product.
Using potassium t-butoxide and diethyl carbonate in benzene,
the preparation of 2-carbethoxycycloheptanone has been reported
only in 40% yield. With a modified method of Paul Krapcho
et al.,41 we prepared the intermediate cyclic b-ketoester 5a by
reaction of 2-phenylcycloheptanone 4a (racemate) with diethyl
carbonate in the presence of sodium hydride under reflux, followed
by hydrolysis (with dilute HCl preferred to glacial acetic acid) after
which the yield was increased to 94.9%.
Biological activity
The compounds 6a–b, 7a–b, 8a–b, 9 and 10 were tested for their
activity against HIV in RT assay, using a poly(rA)/oligo(dT)15
homopolymer template with the HIV antigen detection ELISA
method42 and nevirapine as a reference compound. It is particu-
larly noteworthy that compound 8a (IC50 = 1.51 lM) was more
active than nevirapine (IC50 = 3.67 lM) at concentrations 2-fold
and 102-fold compared to 7a (IC50 = 100.52 lM) by introducing
the terminal phenyl moiety at the N-1 side chain. However,
Claus Nielsen et al.35 reported the preparation of annelated
analogues I and II, using the Dannel et al. method, the b-ketoester
3254 | Org. Biomol. Chem., 2006, 4, 3252–3258
This journal is
The Royal Society of Chemistry 2006
©

View Online
solid residue was washed with cold water and obtained the white
needles solid.
Ethyl N-benzylc◦arbamate (2a). Yield 96.7% mp 44–45 ◦C
(H2O) (lit.40 45–47 C).
(4-Meth◦yl-benzyl)carbamic aci◦d ethyl ester (2b). Yield 90.0%,
mp 55–56 C (H2O) (lit.43 55–57 C).
General procedure for the preparation of compounds 3a–b
A mixture of 20 mmol of 2 in 20 mL ethyl ether and 174 mmol of
sodium nitrite in 20 mL H2O was stirred, and then treated with
another solution of 10 ml concentrated nitric acid and water for
each. The dropping rate was adjusted to keep the aqueous phase
green. The reaction mixture was allowed to stand for another
20 min. The ether layer was separated and the resulting solution
was washed with 5 mL 10% potassium carbonate solution, dried
and evaporated under reduced pressure to give the nitroso product
as orange–red oil in a nearly quantitative yield.
Ethyl N-nitroso-N-benzylcarbamate (3a). Yield 98.0%.
Ethyl N-nitroso-N-(4ꢀ-methyl)benzylcarbamate (3b). Quanti-
tative yield.
General procedure for the preparation of compounds 4a–b
Fig. 3 X-Ray crystal structure of 7a.
To a stirred mixture which include 40 mmol of freshly distilled
cyclohexanone, 0.3 g of finely powdered anhydrous potassium
carbonate and 8 mL absolute methanol, was added, over a period
of 10 min, 20 mmol of the ethyl N-nitroso compound 3 during
which time the temperature was maintained at 20–25 ◦C by means
of a cold water bath. The reaction mixture was then allowed to
stand by overnight. The potassium carbonate was removed by
filtration, the solvents were evaporated under reduced pressure, an
oily residue was obtained which was purified by silica gel column
chromatography with EtOAc–petroleum ether (60–90 ◦C) to give
the products.
Fig. 4 NOE between 9-H with the N–CH2–Rꢀ groups.
an increase of the steric bulkyness at the 9-phenyl ring with
4-methyl substitution showed lower activity for 8b (IC50
9.31 lM) compared to 8a.
=
2-Phenylcycloheptanone (4a). Yield 59.0%, light yellow oil,
(lit.40 41%).
2-p-Tolylcycloheptanone (4b). Yield 31◦.5%, mp 52–54 ◦C
Experimental
(EtOAc–petroleum ether) (lit.43 26%, 57–58 C).
Melting points were determined with an electrothermal capillary
melting point apparatus and are uncorrected. H and 13C NMR
1
General procedure for the preparation of compounds 5a–b
spectra were recorded on a JEOL-AL-300 FT spectrometer with
TMS as the internal standard. The ESI-TOF spectra were taken
on a Liner Scieneific LDI-1700 mass spectrometer. IR spectra were
recorded with Avatar 360 FT-IR and reported in cm−1. Elemental
analyses were performed on VarVario EL III. Silica gel (0.040–
0.064 mm) was used for column chromatography. Some reagents
were purchased from Aldrich.
A 8.5 mmol amount of sodium hydride was washed with freshly
distilled hexane. After most of the mineral oil had been removed,
8 mL of benzene and 6.0 mmol of diethyl carbonate were added.
The mixture was heated to reflux and a solution of 3.0 mmol 4 in
1 mL benzene was added dropwise. After addition, this mixture
was refluxed for 5 h, cooled to room temperature and hydrolyzed
with dilute HCl. The benzene layer was separated and the aqueous
layer was extracted with ethyl ether (3 × 10 mL). The combined
organic layer was washed with cold water, dried and evaporated in
vacuo to afford the crude product which was purified by silica gel
column chromatography with EtOAc–petroleum ether (60–90 ◦C)
as eluent.
General procedure for the preparation of compounds 2a–b
Compound 1 (37.3 mmol) was mixed with 2 mL H2O and 6 g
chopped ice. To the stirred mixture 40 mmol of ethyl chlorocar-
bonate was added dropwise. Simultaneously, an ice-cold solution
of 41 mmol of sodium hydroxide in 5.2 mL water was added
dropwise, at the same rate with that of the ethyl◦chlorocarbonate
in order to maintain the temperature at 10–15 C. The reaction
mixture was stirred for an additional 20 min and then filtered. The
Ethyl-2-oxo-3-phenylcycloheptanecarboxylate
(5a). Yield
(741 mg, 94.9%), yellow oil; 1H NMR (300 MHz, CDCl3) d
1.30 (3H, t, CH2CH3), 1.46–2.18 (8H, m, CH2CH2CH2CH2),
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 3252–3258 | 3255
©

View Online
3.67 (1H, m, COCHCO), 3.94 (1H, m, ArCHCO), 4.13 (q, 2H
OCH2CH3), 7.23–7.32 (5H, m, Ar); 13C NMR (75 MHz, CDCl3)
d 14.1 (CH2CH3), 26.3, 27.9, 28.7, 32.7 (CH2CH2CH2CH2),
57.2 (COCHAr), 58.4 (COCHCO), 61.0 (COOCH2CH3), 126.9,
127.0, 127.9, 128.2, 128.3, 139.7 (Ar), 170.2 (COOC2H5), 207.9
(CO); ESI-TOF+: 261.1748 (M + H)+.
residue was chromatographed on silica gel with EtOAc–petroleum
ether as eluent to afford the pure N-1 alkylated product.
1-(Ethoxymethyl)-6,7,8,9-tetrahydro-9-phenyl-1H-cyclohepta-
[d]pyrimidine-2,4(3H,5H)-dione (7a). Yield (32 mg, 92.6%); mp
180–182 ◦C (10% EtOAc–petroleum ether), white solid; (found:
C, 68.73; H, 7.10; N, 9.00%, C18H22N2O3, requires C, 68.77;
1
H, 7.05; N, 8.91); mmax/cm−1 3349.99, 1669.80 and 1608.77; H
Ethyl-2-oxo-3-p-tolylcycloheptanecarboxylate
(5b). Yield
(740 mg, 89.9%); yellow oil; 1H NMR (300 MHz, CDCl3) d
1.22 (3H, t, CH2CH3), 1.35–2.23 (8H, m, CH2CH2CH2CH2),
2.32 (3H, s, ArCH3), 3.65 (1H, m, COCHCO), 3.91 (1H, m,
ArCHCO), 4.15 (q, 2H, OCH2CH3), 7.10–7.25 (4H, m, Ar);
13C NMR (75 MHz, CDCl3) d 14.0 (CH2CH3), 21.0 (ArCH3),
26.2, 27.7, 28.3, 32.9 (CH2CH2CH2CH2), 56.9 (COCHAr), 58.2
(COCHCO), 61.1 (COOCH2CH3), 127.7, 128.2, 128.9, 129.1,
136.6,136.7 (Ar), 170.2 (COOC2H5), 208.1 (CO); ESI-TOF+
275.1558 (M+ H)+.
NMR (300 MHz, CDCl3) d 1.20 (3H, t, CH2CH3), 1.42–2.98
(8H, m, CH2CH2CH2CH2), 3.60 (2H, m, OCH2CH3), 4.71(1H,
m, ArCH), 4.74, 5.54 (2H, 2 × d, J 11.4, NCH2O), 7.17–7.38
(5H, m, Ar), 8.51 (1H, s, NH3); 13C NMR (75 MHz, CDCl3) d
15.1 (CH2CH3), 21.7, 23.6, 24.7, 30.9 (CH2CH2CH2CH2), 44.5
(ArCH), 64.9 (OCH2CH3), 72.8 (NCH2O), 116.0 (C5),126.9,
127.0, 127.0, 129.1, 129.1, 138.5 (Ar), 151.6 (C6), 154.8 (C2), 163.2
(C4); ESI-TOF+ 315.1571 (M + H)+.
1-[(Benzyloxy)methyl]-6,7,8,9-tetrahydro-9-phenyl-1H-cyclo-
hepta[d]pyrimidine-2,4(3H,5H)-dione (8a). Yield (29 mg,
70.5%); mp 50–52 ◦C (10% EtOAc–petroleum ether), white
solid; (found: C, 73.40; H, 6.47; N, 7.49%, C23H24N2O3, requires
C, 73.38; H, 6.43; N, 7.44%); mmax/cm−1 3340.18, 1670.23 and
1617.65; 1H NMR (300 MHz, CDCl3) d 1.60–2.95 (8H, m,
CH2CH2CH2CH2), 4.69(1H, m, ArCH), 4.75(2H, m, PhCH2O),
4.79, 5.66 (2 H, 2 × d, J 11.1, NCH2O), 7.25–7.37 (10H, m, 2 ×
Ar), 9.2 (1H, s, NH3); 13C NMR (75 MHz, CDCl3) d 21.5, 23.5,
24.6, 30.9 (CH2CH2CH2CH2), 44.7 (ArCH), 65.2 (OCH2Ph),
72.7 (NCH2O), 116.0 (C5),126.8, 126.8, 126.9, 126.9, 127.5, 127.9,
128.5, 128.5, 129.1, 129.1, 137.2, 138.4 (2 × Ar), 152.0 (C6), 154.5
(C2), 163.6 (C4); ESI-TOF+: 377.1569 (M + H)+.
General procedure for the preparation of compounds 6a–b
Urea (3.3 mmol) and compound 5 (3.0 mmol) were added to a
solution of sodium metal (4.0 mmol) in 9 mL of absolute ethanol,
and the mixture was heated at reflux for 6 h. After evaporation
in vacuo at 40–50 ◦C until nearly dryness, the residue was dissolved
in 5 mL water and then precipitated by addition of dilute aqueous
HCl to pH 4. The crude product was recrystalized from ethanol
to afford pure 6a–b.
6,7,8,9-Tetrahydro-9-phenyl-1H -cyclohepta[d]pyrimidine-2,4-
(3H,5H)-dione (6a). Yield (400 mg, 52.0%); mp 230–232 ◦C
(EtOH), white solid; (found: C, 70.31; H, 6.25; N, 10.87%,
C15H16N2O2, requires C, 70.29; H, 6.29; N, 10.93%); mmax/cm−1
3452.17, 3229.26, 1716.58 and 1634.28;1H NMR (300 MHz,
DMSO-d6) d 1.35–2.83 (8H, m, CH2CH2CH2CH2), 4.05 (1H,
m, CHAr), 7.19–7.39 (5H, m, Ar), 10.53 (1H, s, NH1), 11.10
(1H, s, NH3); 13C NMR (75 MHz, DMSO-d6) d 21.5, 24.0, 25.6,
29.9 (CH2CH2CH2CH2), 46.7 (ArCH), 111.1 (C5), 126.5, 127.1,
127.1, 128.7, 128.7, 138.5 (Ar), 150.8 (C6), 153.4 (C2), 164.8 (C4);
ESI-TOF+ 257.1845 (M + H)+.
1-(Ethoxymethyl)-6,7,8,9-tetrahydro-9-p-tolyl-1H-cyclohepta
[d]pyrimidine-2,4(3H,5H)-dione (7b). Yield (32 mg, 88.2%); mp
136–138 ◦C (10% EtOAc–petroleum ether), white solid; (found:
C, 69.44; H, 7.41 N, 8.60%, C19H24N2O3, requires C, 69.49;
H, 7.37; N, 8.53%); mmax/cm−1 3338.15, 1668.54, 1600.98; 1H
NMR (300 MHz, CDCl3) d 1.16 (3H, t, CH2CH3), 1.45–2.98
(8H, m, CH2CH2CH2CH2), 2.34 (3 H, s, ArCH3), 3.61 (2H, m,
OCH2CH3), 4.66 (1 H, m, PhCH), 4.73, 5.56 (2 H, 2 × d, J 11.4,
NCH2O), 7.04–7.17 (4H, m, Ar), 8.94 (1H, s, NH3); 13C NMR
(75 MHz, CDCl3) d 15.1 (CH2CH3), 21.0 (PhCH3), 21.7, 23.6,
24.7, 30.9 (CH2CH2CH2CH2), 44.2 (ArCH), 64.8 (OCH2CH3),
72.7 (NCH2O), 115.8 (C5), 126.8, 126.8, 129.8, 129.8, 135.4, 136.6
(Ar), 151.8 (C6), 155.0 (C2), 163.4 (C4); ESI-TOF+: 329.1687 (M +
H)+.
6,7,8,9-Tetrahydro-9-p-tolyl-1H-cyclohepta[d]pyrimidine-2,4-
(3H,5H)-dione (6b). Yield (373 mg, 46.0%); mp 236–238 ◦C
(EtOH), white solid; (found: C, 71.11; H, 6.75; N, 10.30%,
C16H18N2O2, requires C, 71.09; H, 6.71; N, 10.36%); mmax/cm−1
3410.89, 3219.36, 1709.85 and 1640.31; 1H NMR (300 MHz,
DMSO-d6) d 1.22–2.83 (8H, m, CH2CH2CH2CH2), 2.28 (3H, s,
ArCH3), 4.00 (1H, m, CHAr), 7.05–7.17 (4H, m, Ar), 10.52 (1H, s,
NH1), 11.10 (1H, s, NH3); 13C NMR (75 MHz, DMSO-d6) d 20.6
(PhCH3), 21.5, 24.1, 25.7, 29.9 (CH2CH2CH2CH2), 46.3 (ArCH),
111.5 (C5), 127.0, 127.0, 129.3, 129.3, 131.5, 135.4 (Ar), 151.0
(C6), 153.7 (C2), 164.8 (C4); ESI-TOF+ 271.1654 (M + H)+.
1-[(Benzyloxy)methyl]-6,7,8,9-tetrahydro-9-p-tolyl-1H-cyclo-
hepta[d]pyrimidine-2,4(3H,5H)-dione (8b). Yield (30 mg,
70.0%); mp 60–62 ◦C (10% EtOAc–petroleum ether), white
solid; (found: C, 73.79; H, 6.68; N, 7.12%, C24H26N2O3, requires
C, 73.82; H, 6.71; N, 7.17%); mmax/cm−1 3343.28, 1675.36 and
1619.21; 1H NMR (300 MHz, CDCl3) d 1.60–2.95 (8H, m,
CH2CH2CH2CH2), 2.33 (3H, s, ArCH3), 4.64(1H, m, ArCH),
4.70 (2H, m, PhCH2O), 4.77, 5.66 (2 H, 2 × d, J 11.1, NCH2O),
7.00–7.38 (9H, m, 2 × Ar), 8.89 (1H, s, NH3); 13C NMR (75 MHz,
CDCl3) d 21.0 (ArCH3), 21.6, 23.5, 24.6, 31.0 (CH2CH2CH2CH2),
44.4 (ArCH), 71.5 (OCH2Ph), 72.7 (NCH2O), 115.9 (C5), 126.8,
127.9, 127.9, 128.0, 128.0, 128.4, 128.4, 129.8, 129.8, 135.3, 136.6,
137.2 (2 × Ar), 151.8 (C6), 154.7 (C2), 163.3 (C4); ESI-TOF+:
391.1543 (M + H)+.
General procedure for the preparation of compounds 7a–b and 8a–b
To a suspension of 5 (0.11 mmol) in 2 mL dry CHCl3 was
added N,O-bis(trimethylsilyl)acetamide (BSA) 0.24 mmol and
the stirring was continued to obtain a clear solution. Then
chloromethyl ether (0.13 mmol) was added and the reaction
mixture was stirred until TLC showed no change in amount of
starting material. After evaporation of the solvent in vacuo, the
3256 | Org. Biomol. Chem., 2006, 4, 3252–3258
This journal is
The Royal Society of Chemistry 2006
©

View Online
General procedure for the preparation of compounds 9 and 10
directly from a b-ketoester with urea; (c) for diverse alkyl agents,
different methods were used in forming the side chain of N-1.
As seen from examination of the results, structural variation in
the TMs have resulted in a wide range of biological activities with
IC50 values from 1 to 200 lM against wild-type HIV-1 RT. When
comparing trends among the series of compounds, we found that
8a (IC50 = 1.51 lM) substituted at Rꢀ by a phenyl group was 2 times
more potent than nevirapine (IC50 = 3.67 lM) and 100 times more
active than 7a (IC50 = 100.52 lM). From the factors responsible
for the significant activity of compound 8a, it can be elucidated
that the terminal phenyl moiety at the N-1 side chain of 8a extends
deeper into and occupies a more hydrophobic pocket, comprised
of Tyr318 and Pro236. Particularly, a Tyr318 situated within a
short distance of the N-1 group provides the p–p interaction with
the phenyl ring. The bulky N-1 substituents may be accommodated
because of the flexibility of the Pro236 loop region. To confirm
the above-mentioned findings, we shortened the side chain of N1
with allyl and benzyl moieties to 9 and 10, respectively, which
led to a significant drop in activity compared to 8a. However, the
activity of compound 10 (IC50 = 101.28 lM) is better than that of
9 (IC50 > 200 lM). Thus, an appropriately substituted benzyloxy
methyl ether may improve the activity.
We next investigated the influence of the substituent at the
9-phenyl ring for activity against HIV-1. It is known from the
literature that the corresponding GCA-186 with 3,5-dimethyl
substituents on its phenyl ring was better able to tolerate the
presence of Y181C or K103N mutations than MKC-442. In our
study compounds, 7b and 8b, the increase in steric bulkiness at
the 9-phenyl ring with 4-methyl substitution seems to be not
particularly favorable; the biological activity obtained for 8b
(IC50 = 9.31 lM) is lower than that of 8a. The 4-methyl group
probably forced the 9-phenyl ring away from the hydrophobic
pocket formed by the side chains of Try181, Try188, Phe227,
Trp229 and reduced the p–p interactions. In contrast, the 3,5-
dimethyl groups of GCA-186 could correlate with an overall
tighter binding in the hydrophobic pocket, and thus enhance the
affinity to the RT. Encouraged by these results, more detailed SAR
studies on the annelated compounds are underway, with a focus
on exploring the important role of this unique structure feature.
A mixture of 6 (0.12 mmol), proper alkyl bromide (0.14 mmol),
potassium carbonate (20 mg, 0.15 mmol) and catalytic amount of
KI in 1 mL of anhydrous N,N-dimethyl formamide (DMF) was
stirred at room temperature for 8 h. After treatment with cold
water (10 mL), the solution was extracted with ethyl acetate (3 ×
15 mL). The organic layers were collected, dried and evaporated
to furnish the crude product, which was purified by column
chromatography on silica gel (eluent: 20% EtOAc–petroleum
ether).
1-Allyl-6,7,8,9-tetrahydro-9-phenyl-1H -cyclohepta[d ]pyrimi-
dine-2,4(3H,5H)-dione (9). Yield (32 mg, 90%); mp 175–177 ◦C
(10% ethyl acetate–petroleum ether), light yellow solid; (found:
C, 73.35; H, 6.90; N, 9.50%. C18H20N2O2, requires C, 72.95; H,
1
6.80; N, 9.45%); H NMR (300 MHz; CDCl3) d 1.25–3.18 (8H,
m, CH2CH2CH2CH2), 4.05 (1H, m, ArCH), 4.50 (2H, d, J 6.0,
=
=
NCH2CH ), 5.17, 5.24 (2H, 2 × d, J 10.5 and 18.0 CH CH2),
=
5.87 (1H, m, CH2CH CH2), 7.37–7.54 (5H, m, Ar), 8.87 (1H, s,
NH3); 13C NMR (75 MHz; CDCl3) d 21.5, 24.1, 25.7, 29.9
=
(CH2CH2CH2CH2), 46.3 (ArCH), 47.8 (NCH2CH ), 111.1 (C5),
=
117.2 (CH CH2), 126.0, 127.2, 127.2, 129.3, 129.3, 135.4 (Ar),
+
=
135.6 (CH CH2), 150.8 (C6), 153.7 (C2), 164.8 (C4); ESI-TOF
297.1496 [M + H]+.
1-Benzyl-6,7,8,9-tetrahydro-9-phenyl-1H-cyclohepta[d]pyrimi-
dine-2,4(3H,5H)-dione (10). Yield (36 mg, 87%); mp 162–164 ◦C
(10% ethyl acetate–petroleum ether), light yellow solid; (found: C,
76.35; H, 6.58; N, 8.28%. C22H22N2O2, requires C, 76.28; H, 6.40;
1
N, 8.09%); H NMR (300 MHz; CDCl3) d 1.42–3.14 (8H, m,
CH2CH2CH2CH2), 4.03 (1H, m, ArCH), 5.06 (2H, s, NCH2Ph)
7.20–7.47 (10H, m, 2 × Ar), 8.91 (1H, s, NH); 13C NMR (75 MHz;
CDCl3) d 23.6, 25.8, 30.1, 32.3 (CH2CH2CH2CH2), 44.2 (PhCH),
48.2 (NHCH2Ph), 111.9 (C5), 127.5, 128.3, 128.3, 128.4, 128.4,
128.4, 128.4, 129.2, 129.7, 129.7, 136.9, 138.4 (2 × Ar), 150.7,
151.5 (C6 and C2), 163.6 (C4); ESI-TOF+ 347.1481 [M + H]+.
Crystal data for 7a†
C18H22N2O3, M = 314.38, triclinic, space group P-1, a = 8.2013
˚
(12), b = 10.5938 (16), c = 10.9160 (17) A, unit-cell volume =
810.1(2), at 297(2) K, Z = 2, absorption coefficient l = 0.088,
total reflections = 3286, refined final R values for all = 0.0509.
Two of the carbon atoms (C10, C11) of the cycloheptene ring
are disordered unequally over two adjacent sites with occupancies
0.732(7) and 0.268(7).
Acknowledgements
We thank the National Science Foundation of China and 985
program of the Ministry of Education of China for financial
support.
Conclusions
References
In the course of synthesizing target molecules (TMs) 7a–b,
8a–b, 9 and 10, we have developed a new synthetic route in
reaching a C5 and C6 annelated structure of uracil with a seven-
membered ring. These include (a) using a modified method of
Paul Krapcho et al., the intermediate cyclic b-ketoester 5 was
prepared with high yield (90–95%); (b) with a one-step process,
the cyclohepta[d]pyrimidine-2,4-diones (6a–b) were condensed
1 S. P. Goff, J. Acquired Immune Defic. Syndr., 1990, 3, 817–831.
2 H. Mitsuya and S. Broder, Nature, 1987, 325, 773–778.
3 M. Baba, Z. Debyser, S. Shigeta and E. D. Clercq, Drugs Future, 1992,
17, 891.
4 P. A. Furman, J. A. Fyfe, M. H. St. Clair, K. J. Weinhold, , J. L. Rideout,
G. A. Freeman, S. N. Lehrman, D. P. Bolognesi, S. Broder, H. Mitsuya
and D. W. Barry, Proc. Natl. Acad. Sci. U. S. A., 1986, 83, 8333–8337.
5 M. H. St. Clair, C. A. Richards, T. Spector, K. J. Weinhold, V. H. Miller,
A. J. Langlois and P. A. Furman, Antimicrob. Agents Chemother., 1987,
31, 1972.
6 R. Yachoan, H. Mitsuya, R. V. Thomas, J. M. Pioda, N. R. Hartman,
C. F. Perno, K. S. Marczky, J. P. Allain, D. G. Johns and S. Broder,
Science, 1989, 245, 412–415.
† CCDC reference number 296985. For crystallographic data in CIF
format see DOI: 10.1039/b607972p
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 3252–3258 | 3257
©

View Online
7 H. Mitsuya, K. J. Weinhold, P. A. Furman, M. H. St. Clair, S. Nusinoff-
Lehrman, R. C. Gallo, D. Bolognesi and D. W. Barry, Proc. Natl. Acad.
Sci. U. S. A., 1985, 82, 7096–7100.
8 H. Mitsuya and S. Broder, Proc. Natl. Acad. Sci. U. S. A., 1986, 83,
1911–1915.
9 T. S. Lin, F. R. Schinazi and W. H. Prusoff, Biochem. Pharmacol., 1987,
36, 2713–2718.
10 V. J. Merluzzi, K. D. Hargrave, S. Hattox, J. Adams, A. S. Rosenthal,
R. Faanes, R. J. Eckner, R. A. Koup and J. L. Sullivan, Science, 1990,
250, 1411–1413.
11 R. Pauwels, K. Andries, J. Desmyter, D. K. Schols, M. J. Kukla, H. J.
Bresdin, A. Raemaekers, J. Von Gelder, R. Woestenborghs, J. Heyrants,
K. Schellekens, E. D. Clercq and P. A. Janssen, Nature, 1990, 343, 470–
474.
12 T. Miyasaka, H. Tanaka, M. Baba, H. Hayakawa, R. T. Walker, J.
Balzarini and E. D. Clercq, J. Med. Chem., 1989, 32, 2507–2509.
13 D. J. Romero, M. Busso, C. K. Tan, F. Reusser, J. R. Palmer, S. M.
Poppe, P. A. Aristoff, K. M. Downey, L. Resnik and W. G. Tarpley;,
Proc. Natl. Acad. Sci. U. S. A., 1991, 88, 8806–8810.
14 J. H. Nunberg, W. A. Schleif, E. J. Boots, J. A. O’Brien, J. C. Quintero,
J. M. Hoffman, E. A. Emini and M. E. Goldman, J. Virol., 1991, 65,
4887–4892.
23 R. T. Walker and E. D. Clercq, Nucleosides Nucleotides, 1995, 14, 575–
583.
24 Press statement from Triangle Pharmaceuticals Inc., 17 January
2002.
25 C. Catalanotti, B. Morelli and E. Novellino, Curr. Pharm. Des., 2002,
8, 615–657.
26 E. D. Clercq and J. Balzarini, Farmaco, 1995, 50, 735–747.
27 E. D. Clercq, Antiviral Res., 1998, 38, 153–179.
28 E. D. Clercq, Farmaco, 1999, 54, 26–45.
29 J. Ren, R. Esnouf, E. Garman, D. Somers, C. Ross, I. Kirby, J. Keeling,
G. Darby, Y. Jones, D. Stuart and D. Stammers, Nat. Struct. Biol., 1995,
2, 293–302.
30 J. Ding, K. Das, C. Tantillo, W. Zhang, A. D. Clark, S. J. Jessen, X.
Lu, Y. Hsiou, A. Jacobo-Molina, K. Andries, R. Pauwels, H. Moereels,
L. Koymans, P. A. Janseen, R. H. Smith, M. Kroeger Koepke, C. J.
Michejda, S. H. Hughes and E. Arnold, Structure, 1995, 3, 365–
379.
31 J. Ding, K. Das, H. Moereels, L. Koymans, K. Andries, P. A. Janssen,
S. H. Hughes and E. Arnold, Nat. Struct. Biol., 1995, 2, 407–415.
32 J. Ren, R. Esnouf, A. Hopkins, C. Ross, Y. Jones, D. Stammers and D.
Stuart, Structure, 1995, 3, 915–926.
33 R. M. Esnouf, J. Ren, A. L. Hopkins, C. K. Ross, E. Y. Jones, D. K.
Stammers and D. I. Stuart, Proc. Natl. Acad. Sci. U. S. A., 1997, 94,
3984–3989.
34 R. M. Esnouf, D. I. Stuart, E. D. Clercq, E. Schwartz and J. Balzarini,
Biochem. Biophys. Res. Commun., 1997, 234, 458–464.
35 J. S. Larsen, L. Christensen, G. Ludvig, P. T. Jørgensen, E. B. Pedersen
and C. Nielsen, J. Chem. Soc., Perkin Trans. 1, 2000, 3035–3038.
36 F. D. Therkelsen, A. L. L. Hansen, E. B. Pedersen and C. Nielsen, Org.
Biomol. Chem., 2003, 1, 2908.
15 J. W. Mellors, G. E. Dutschman, G. J. Im, E. Tramontano, S. R. Winkler
and Y. C. Cheng, Mol. Pharmacol., 1992, 41, 446–451.
16 J. W. Mellors, G. J. Im, E. Tramontano, S. R. Winkler, D. J. Medina,
G. E. Dutschman, H. Z. Bazmi, G. Piras, C. J. Gonzalez and Y. C.
Cheng, Mol. Pharmacol., 1993, 43, 11–16.
17 M. Baba, H. Tanaka, E. D. Clercq, R. Pauwels, J. Balzarini, D. Schols,
H. Nakashima, C. F. Perno, R. T. Walker and T. Miyasaka, Biochem.
Biophys. Res. Commun., 1989, 165, 1375–1381.
18 T. Miyasaka, H. Tanaka, M. Baba, H. Hayakawa, R. T. Walker, J.
Balzarini and E. D. Clercq, J. Med. Chem., 1989, 32, 2507–2509.
19 S. Yuasa, Y. Sadakata, H. Takashima, K. Sekiya, N. Inouye, M.
Ubasawa and M. Baba, Mol. Pharmacol., 1993, 44, 895–900.
20 M. Baba, S. Shigeta, S. Yuasa, H. Takashima, K. Sekiya, M. Ubasawa,
H. Tanaka, T. Miyasaka, R. T. Walker and E. D. Clercq, Antimicrob.
Agents Chemother., 1994, 38, 688–692.
21 H. Tanaka, H. Takashima, M. Ubasawa, K. Sekiya, N. Inouye, M.
Baba, S. Shigeta, R. T. Walker, E. D. Clercq and T. Miyasaka, J. Med.
Chem., 1995, 38, 2860–2865.
37 M. Macchia, G. Antonelli, A. Balsamo, S. Barontini, F. Calvani,
D. Gentili, A. Martinelli, A. Rossello, O. Turriziani and R. Tesoro,
Farmaco, 1999, 54, 242–247.
38 G. M. Morris, D. S. Goodsell, R. S. Halliday, R. Huey, W. E. Hart,
R. K. Belew and A. J. Olson, J. Comput. Chem., 1998, 19, 1639.
39 A. L. Hopkins, J. Ren, H. Tanaka, M. Baba, M. Okamato, D. I. Stuart
and D. K. Stammers, J. Med. Chem., 1999, 42, 4500.
40 Gutsche Johnson, Org. Synth., 1963, Coll. Vol. IV, 781.
41 A. P. Krapcho, J. Diamanti, C. Cayen and R. Bingham, Org. Synth.,
1967, Coll. Vol. V, 20.
22 A. L. Hopkins, J. Ren, R. M. Esnouf, B. E. Willcox, E. Y. Jones, C.
Ross, T. Miyasaka, R. T. Walker, H. Tanaka, D. K. Stammers and D. I.
Stuart, J. Med. Chem., 1996, 39, 1589–1600.
42 K. Danel, E. Larsen and E. B. Pedersen, Synthesis, 1995, 934–936.
43 C. D. Gutsche and H. E. Johnson;, J. Am. Chem. Soc., 1955, 77, 109–
112.
3258 | Org. Biomol. Chem., 2006, 4, 3252–3258
This journal is
The Royal Society of Chemistry 2006
©