Non-nucleoside HIV-1 reverse transcriptase inhibitors, part 7. Synthesis, antiviral activity, and 3D-QSAR investigations of novel 6-(1-naphthoyl) HEPT analogues

Article abstract of DOI:10.1248/cpb.54.1248

A series of novel 1-[(2-hydroxyethoxy)methyl]-6-(phenylthio)thymine (HEPT) analogues bearing a 6-(1-naphthoyl) group of non-nucleoside human immunodeficiency virus (HIV) reverse transcriptase inhibitors were synthesized and evaluated for their activity against HIV-1 and HIV-2. It was found that most of these compounds showed good activity against HIV-1. Among them, compound 5-isopropyl-6-(1-naphthoyl)-1-[(2E)-3-phenylallyl]-2,4-pyrimidinedione (23) displayed the greatest inhibitory potency (IC₅₀=0.14 μM), which is about 35-fold more active than HEPT and DDI. To rationalize the relationships between structure and activity of these novel compounds, a three-dimensional quantitative structure-activity relationship (3D-QSAR) model was also generated. The results provided a tool for guiding the further design of more potent antiviral agents and for predicting the affinity of related compounds.

Full text of DOI:10.1248/cpb.54.1248

1248
Chem. Pharm. Bull. 54(9) 1248—1253 (2006)
Vol. 54, No. 9
Non-nucleoside HIV-1 Reverse Transcriptase Inhibitors, Part 7.1)
Synthesis, Antiviral Activity, and 3D-QSAR Investigations of Novel
6-(1-Naphthoyl) HEPT Analogues
a
Lei JI, Fen-Er CHEN, ,a Xiao-Qing FENG,a Erik De CLERCQ,b Jan BALZARINI,b and
*
b
Christophe PANNECOUQUE
a Department of Chemistry, Fudan University; Shanghai 200433, People’s Republic of China: and b Rega Institute for
Medical Research, Katholieke Universiteit Leuven; 10 Minderbroedersstraat, B-3000 Leuven, Belgium.
Received February 13, 2006; accepted April 26, 2006
A series of novel 1-[(2-hydroxyethoxy)methyl]-6-(phenylthio)thymine (HEPT) analogues bearing a 6-(1-
naphthoyl) group of non-nucleoside human immunodeficiency virus (HIV) reverse transcriptase inhibitors were
synthesized and evaluated for their activity against HIV-1 and HIV-2. It was found that most of these compounds
showed good activity against HIV-1. Among them, compound 5-isopropyl-6-(1-naphthoyl)-1-[(2E)-3-phenylallyl]-
2,4-pyrimidinedione (23) displayed the greatest inhibitory potency (IC50ꢀ0.14 mM), which is about 35-fold more
active than HEPT and DDI. To rationalize the relationships between structure and activity of these novel com-
pounds, a three-dimensional quantitative structure–activity relationship (3D-QSAR) model was also generated.
The results provided a tool for guiding the further design of more potent antiviral agents and for predicting the
affinity of related compounds.
Key words HIV-1 reverse transcriptase inhibitor; antiviral activity; synthesis; 3D-QSAR; 6-(1-naphthoyl)
Non-nucleoside reverse transcriptase inhibitors (NNR- the aim to examine our assumption and find more potent new
TIs)2) play an essential role in the therapy for AIDS due to anti-HIV agents, in this paper we describe the synthesis and
their potent antiviral activity, high specificity, and low tox- antiviral activity of a series of novel 6-(1-naphthoyl) substi-
icity. Among the representatives of the NNRTIs, 1-[(2-hy- tuted HEPTs (4, Fig. 1).
droxyethoxy)methyl]-6-(phenylthio)thymine (1, HEPT),3—5)
has an interesting structure and highly anti-HIV-1 specific ac- Chemistry
tivity that has induced many kinds of structural modifications
The target compounds in this study were synthesized as
on the skeleton of thymine as a lead compound (Fig. 1). It shown in Chart 1. The barbituric acid derivatives 6a, b were
was found that an aromatic ring structure at the C-6 of the known to be readily available from urea and diethyl malonic
thymine moiety is an important determinant for the anti-HIV esters 5a, b. The barbituric acid derivatives 6a, b were con-
activity.6)
verted to the corresponding trichloropyrimidines 7a, b by re-
In previous papers, we have reported the synthesis of a action of POCl3 and dimethylaniline under reflux for 2 h in
series of 6-naphthylmethyl and 6-naphthylthio substituted 80—85% yields.10)
HEPT analogues (HEPTs), and found that most of them ex-
hibited highly potent anti-HIV-1 activity. For example, 1-ben-
zoxymethyl-5-isopropyl-6-(1-naphthylmethyl)uracil (2)7) and
Treatment of 7a, b with NaOMe in MeOH at ꢀ3 °C for
1-ethoxymethyl-5-isopropyl-6-(1-naphthylthio)uracil
(3)8)
showed highly potent anti-HIV-1 activity with IC50 of 17 and
57 nM, respectively (Fig. 1). Based on our previous research
in this field, we postulate that the introduction of a 6-(1-naph-
thoyl) group as a H-bond acceptor to the C-6 position of
HEPTs may be more beneficial to enhance the interaction be-
tween the inhibitors and reverse transcriptase (RT).9) With
Reagents and conditions: (a) urea, MeONa, MeOH, reflux 10 h, 75—80%; (b)
POCl3, dimethylaniline, 80—85%; (c) MeONa, MeOH, 72—78%; (d) 1-naphthylace-
tonitrile, 60% NaH, DMF, N2; (e) 60% NaH, DMF, air, 80—90% (one-pot synthesis);
(f) conc. HCl, MeOH, reflux 24—48 h; (g) R2X (XꢁBr, Cl), K2CO3, DMF, rt 24—36 h,
35—60%.
Fig. 1
Chart 1. Synthesis of the Target Compounds (12—28)
∗ To whom correspondence should be addressed. e-mail: rfchen@fudan.edu.cn
© 2006 Pharmaceutical Society of Japan

September 2006
1249
with the antiviral activity of HEPT and 2,3-dideoxyinosine
(DDI). All results, expressed as IC50, CC50, and SI, are sum-
marized in Table 1.
As seen from the results listed in Table 1, most of the
tested compounds showed good activity against wild-type
HIV-1 with a wide range of IC50 values from 6.60 to 0.14 mM,
except for compounds 17 and 24. Among them, 5-isopropyl-
6-(1-naphthoyl)-1-[(2E)-3-phenylallyl]-2,4-pyrimidinedione
(23) was the most promising compound and could be used to
summarize as the favorable molecular features for interaction
with RT. It displayed the greatest inhibitory potency activity
against HIV-1 replication with an IC50 value of 0.14 mM,
which is about 35 times more potent than HEPT and DDI,
and a CC50 value of 21.23 mM. The viral selectivity index was
153, which is much better than those found from HEPT and
DDI. Compounds (15, 22) also showed good anti-HIV-1 po-
tency (IC50ꢁ0.19 and 0.17 mM, respectively) and good selec-
tivity indices (SIꢁ144 and 206, respectively).
Chart 2
24—36 h led to the corresponding 5-alkyl-6-chloro-2,4-
Thus it can be seen that some N-1 side chains are suitable
dimethoxypyrimidines 8a, b.11) Condensation of 8a, b with 1- to improve the anti-HIV-1 activity, such as allyl, (2E)-3-
naphthylacetonitrile in the presence of 60% NaH in anhy- phenylallyl, methoxycarbonylallyl, and benzyl. The factors
drous DMF for 48 h at room temperature under N2 gave the responsible for the improved inhibitory potency of these
corresponding nitriles 9a, b, which, without separation, were compounds remain to be elucidated. One explanation may be
immediately oxidized to the ketones 10a, b by passing air that these N-1 substituent groups have suitable length to
into the reaction mixture at room temperature. Demethyla- make interactions with the RT. On the other hand, these
tion of 10a, b with conc. HCl in refluxing MeOH for 24— groups could interact with RT-amino acid Phe227 or Pro236
48 h afforded the 5-alkyl-6-(1-naphthoyl)-2,4-pyrimidine- through p-stacking. The (2E)-3-phenylallyl group was the
diones 11a, b, which were subjected to N-1 alkylation with best, which could be due to its steric interactions with the ad-
various halids in the presence of anhydrous K2CO3 in anhy- jacent amino acids. This is also confirmed by the CoMFA
drous DMF to afford the desirable target compounds 12—28. analysis. However, introduction of some groups at the N-1
It is worth mentioning that in this N-1 alkylation reaction, site resulted in lower activity and higher cytotoxicity, e.g., 4-
some by-products such as N-3 alkylation and N,N-1,3 dialky- methoxybenzoylmethyl, solanesyl, and 2-propynyl. This
lation compounds (Chart 2) were also formed, just as we might be ascribed to their unsuitable length or unsuitable
have described previously.7) The N-1 alkylation compounds configuration to make interaction with RT-amino acid
12—28 were the main products, which could be purified by Phe227 or Pro236.
column chromatography (eluent: ethyl acetate : hexane).
Interestingly, in the N-1 alkylation for preparation of the
All the target compounds 12—28 were characterized by target compound 23, one N,N-1,3 dialkylation by-product
NMR, MS, and elemental analysis. Both analytical and spec- 1,3-di[(2E)-3-phenylallyl]-5-isopropyl-6-(1-naphthoyl)-2,4-
tral data of all the compounds are in full agreement with the pyrimidinedione (29) was also isolated and evaluated against
proposed structures.
HIV-1 and HIV-2. Compared with compound 23, compound
The structure determination of the target compounds of N- 29 exhibited less activity and much higher cytotoxicity
1 alkylation relies heavily on the application of protonde- (IC50>129.39 mM and SI<1). It has been deduced that the 3-
tected heteronuclear NMR experiments. Perhaps the most NH function enhances anti-HIV activity by donation of a hy-
1
useful of these methods is the two-dimensional H–13C het- drogen bond to the carbonyl oxygen of Lys101 in RT.13)
eronuclear multiple-bond correlation (HMBC) experiment, Therefore our results further confirm that the 3-NH function
which provides correlations between protons and carbons is essential for inhibition of HIV RT to HEPT analogues.
over two and three bonds. Thus the N-1 substitutent com-
The triggering effect of 5-alkyl on interaction with the ac-
pound was further confirmed by HMBC. No N-3 regioisomer tive binding site was the same as shown in the previous rela-
was detected.
tive QSAR study.14,15) It showed that inhibitory activity in-
creased proportionally to the modification of the C-5 sub-
stituent in the order Et→i-Pr, where the influence of the iso-
Results and Discussion
All the synthesized compounds were evaluated for anti- propyl substituent is much greater than that of the ethyl sub-
HIV activity and cytotoxicity against wild type HIV-1 strain stituent. Besides the anti-HIV-1 activity, we also evaluated
IIIB and HIV-2 ROD in MT-4 cells using the MTT method.12) activity against HIV-2 ROD, and found that all of the com-
The inhibitory concentration of compounds was expressed as pounds showed less activity against HIV-2 in comparision
the concentration that caused 50% inhibition of viral cyto- with that against HIV-1.
pathogenicity (IC50) without direct toxicity to the cells. The
3D-QSAR Analysis With the aim to rationalize the rela-
cytotoxic concentration (CC50) of the compounds was moni- tionships between structure and activity of these novel com-
tored based on the growth of noninfected cells by the trypan pounds, we also conducted 3D-QSAR analyses. We applied
blue exclusion method and corresponded to the concentration 3D-QSAR methods, comparative molecular field analysis
required to cause 50% cell death. Results were compared (CoMFA) to training our compounds 12—17, 19—27 and

1250
Vol. 54, No. 9
Table 1. Anti-HIV Activity in MT-4 Cells of Compounds 12—29
IC50 (mM)a)
HIV-2 ROD
Compd.
R1
R2
CC50 (mM)b)
SIc)
HIV-1 IIIB
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Et
Et
Et
Et
Et
Et
Et
Et
CH2ꢁCHCH2
0.60ꢂ0.09
6.60ꢂ2.46
0.92ꢂ0.00
0.19ꢂ0.05
1.07ꢂ0.18
ꢃ28.75
ꢃ39.22
ꢃ36.33
ꢃ34.44
ꢃ28.71
ꢃ33.67
ꢃ28.75
ꢃ167.62
ꢃ4.82
ꢃ34.35
ꢃ164.27
ꢃ37.10
ꢃ21.23
ꢃ72.01
ꢃ33.54
ꢃ37.01
ꢃ4.10
39.22ꢂ5.12
36.33ꢂ6.10
34.44ꢂ7.19
28.71ꢂ12.97
33.67ꢂ4.37
28.75ꢂ6.70
167.62ꢂ39.34
4.82ꢂ2.68
34.35ꢂ4.17
164.27ꢂ40.18
37.10ꢂ7.24
21.23ꢂ10.97
72.01ꢂ19.41
33.54ꢂ3.04
37.01ꢂ6.85
4.10ꢂ2.77
42.98ꢂ9.03
ꢆ129.39
64
6
37
144
32
ꢅ1
148
ꢅ1
6
(CH3)2CꢁCHCH2
H3COOCCHꢁCHCH2
trans-PhCHꢁCHCH2
PhCH2
(4ꢄ-OCH3)PhCOCH2
CH3OOCCH2
1.12ꢂ0.08
ꢃ4.82
HCꢀCCH2
MeOCH2
Et
Et
5.89ꢂ0.56
5.71ꢂ1.28
0.17ꢂ0.06
0.14ꢂ0.02
ꢃ72.01
0.30ꢂ0.00
1.42ꢂ0.51
ꢆ1.21
MeOCH2CH2OCH2
CH2ꢁCHCH2
trans-PhCHꢁCHCH2
trans-[CH2CHꢁC(CH3)CH2]9H
PhCH2
EtOOCCH2
HCꢀCCH2
MeOCH2
29
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
i-Pr
206
153
ꢅ1
111
26
ꢇ3
48
ꢅ1
80
28
29
HEPT
DDI
0.88ꢂ0.26
ꢃ129.39
5.06
ꢃ42.98
ꢃ129.39
ꢃ405.8
2.71
405
ꢆ529
5.37
ꢆ98
a) Concentration required to protect the cell against viral cytopathogenicity by 50% in MT-4 cells. b) Concentration that reduces the normal uninfected MT-4 cell viability
by 50%. c) Selectivity index: ratio CC50/IC50, a higher SI means a more selective compound.
Table 2. Summary of CoMFA Model
Model
q2
N
r2
S.E.
F
Static
Elec
CoMFA
0.646
4
0.975
0.118 78.634 0.672
0.328
Table 3. Experimental Activities and Predictive Activities by CoMFA of
the Training Set
Compd.
pIC50
CoMFA
Error
12
13
14
15
16
19
20
21
22
23
25
26
27
6.22
5.18
6.04
6.72
5.97
5.32
5.23
5.24
6.77
6.85
6.52
5.85
5.92
6.26
5.04
6.04
6.69
6.00
5.43
5.17
5.29
6.59
6.94
6.42
5.96
6.00
ꢀ0.04
0.14
0.00
0.03
ꢀ0.03
ꢀ0.11
0.06
ꢀ0.05
0.18
ꢀ0.09
0.10
ꢀ0.11
ꢀ0.08
Fig. 2. 3D-View of All Aligned Compounds
29. Compounds 18 and 28 were used as the test set. The
aligned molecules are shown in Fig. 2.
Partial least squares (PLS), the statistical method used in
deriving 3D-QSAR models, is an extension of multiple re-
gression analysis in which the original variables are replaced
by a small set of their linear combinations.16—18) Leave-one-
out (LOO) cross-validated PLS analyses were used to check
the predictive ability of the models and to determine the opti- moval of outliers, compounds 17, 24, and 29, based on resid-
mal number of components to be used in the final QSAR ual values afforded models with q2 value of 0.646, with the
models. With all the compounds’ inhibitory activity as re- number of partial least squares (PLS) components of 4. The
sponse variable, initial CoMFA PLS analyses of the 16 com- regression coefficients (r2) value was up to 0.975. The results
pounds gave low cross-validated coefficients (q2) values. Re- from the CoMFA studies are summarized in Table 2. The

September 2006
1251
quality of the final CoMFA method is measured by two sta- model, should be greater than 0.40. These results indicate
tistical parameters: r2 and q2. The value of r2 shows the self- stability of our CoMFA model. The predictive ability of the
consistency of the model and should be greater than 0.90. model was further validated using two compounds (18, 28)
The value of q2, which exhibits the predictive capacity of the on account of the number of our compounds is limited. The
experimental activities and the predicted activities using our
Table 4. Experimental Activities and Predictive Activities by CoMFA of
the Test Set
CoMFA model and the residue values of the training set and
the test set are given in Tables 3 and 4, respectively. A graph
of the actual pIC50 vs. the predicted pIC50 values for the train-
ing set compounds by the CoMFA models based on com-
Compd.
pIC50
CoMFA
Error
18
28
5.95
6.06
5.56
6.09
0.39
ꢀ0.03
Table 6. Physical Data of the Target Compounds
Analysis (%), Calcd (Found)
Compd.
Formula
C
H
N
12
13
14
15
16
17
18
19
20
21
22
23
25
26
27
28
C20H18N2O3
C22H22N2O3
C22H20N2O5
C26H22N2O3
C24H20N2O3
C26H22N2O5
C20H18N2O5
C20H16N2O3
C19H18N2O4
C21H22N2O5
C21H20N2O3
C27H24N2O3
C25H22N2O3
C22H22N2O5
C21H18N2O3
C20H20N2O4
71.84
(72.05)
72.91
(72.80)
67.34
(67.46)
76.08
(76.34)
74.98
(74.83)
70.58
(70.81)
65.57
(65.65)
72.28
(72.03)
67.44
(67.60)
65.96
(66.05)
72.40
(72.14)
76.39
(76.68)
75.36
(75.19)
66.99
(67.23)
72.82
(72.64)
68.17
(68.39)
5.43
(5.44)
6.12
(6.13)
5.14
(5.13)
5.40
(5.38)
5.24
(5.25)
5.01
(5.02)
4.95
(4.96)
4.85
(4.84)
5.36
(5.38)
5.80
(5.82)
5.79
(5.78)
5.70
(5.72)
5.57
(5.58)
5.62
(5.60)
5.24
(5.25)
5.72
(5.74)
8.38
(8.35)
7.73
(7.71)
7.14
(7.15)
6.82
(6.81)
7.29
(7.27)
6.33
(6.31)
7.65
(7.68)
8.43
(8.46)
8.28
(8.25)
7.33
(7.34)
8.04
(8.07)
6.60
(6.62)
7.03
(7.02)
7.10
(7.12)
8.09
(8.06)
7.95
(7.98)
Fig. 3. Predictive vs. Experimental pIC50 Values Derived from the CoMFA
Model of the Training Set
Fig. 4. CoMFA Contour Map
Green contours indicate regions where a bulky substituent increases potency, red
contours show areas where electronegative substituent increases potency.
Table 5. Physical Data of the Target Compounds
Compd.
Yield (%)
Property
mp (°C)
ESI-MS (m/z, [MꢈH]ꢈ)
12
13
14
15
16
17
18
19
20
21
22
23
25
26
27
28
57.3
36.5
35.0
59.3
50.7
22.9
37.5
38.5
57.2
55.1
49.2
60.1
50.9
50.6
57.2
59.1
A light yellowish powder
A yellow powder
An orange powder
Light yellowish crystals
A white powder
A yellow powder
A yellow powder
A white powder
A light yellowish powder
A white powder
A white powder
White crystals
A white powder
189.3—190.5
130.0—131.2
108.1—109.3
101.6—102.7
221.2—222.8
232.1—233.2
186.1—187.2
191.3—191.5
131.0—132.4
118.0—119.1
196.0—196.2
130.8—131.9
224.4—225.3
169.1—169.6
192.3—193.6
173.1—174.4
335
363
393
411
385
443
367
333
339
383
349
425
399
395
347
353
A white powder
A white powder
A white powder

1252
Vol. 54, No. 9
pound 23 is shown in Fig. 3.
Conclusion
In summary, we have described the synthesis and anti-HIV
The CoMFA coefficient contour maps for the inhibitory
activity are shown in Fig. 4. A green contour near the 5-posi- activity of novel 6-(1-naphthoyl) substituted HEPT ana-
tion of the HEPT analogues indicates that bulky substituents logues. The results showed that most of these analogues ex-
at this position increase activity. This conclusion is consistent hibited moderate or high activity against HIV-1. The most
with the results of our experiments showing that 5-ethyl sub- active compound 23 exhibited about 35-fold more active
stituted compounds (12—21) are generally less active than 5- anti-HIV-1 activity than HEPT and DDI. These results con-
isopropyl substituted compounds (22—28). A blue contour firmed our assumption that the introduction of a 1-naphthoyl
around the 1-methylene attached to the 2,4-pyrimidinedione substituent to the C-6 position of HEPTs might improve the
ring suggests that electronegative groups at this location antiviral activity since it was more favorable to enhance the
might decrease potency. Both the yellow and the red contours interaction between the inhibitors and RT. We also have done
appearing near the end of the allyl moiety suggest that a a detailed 3D-QSAR study for better understanding their in-
small and electronegative substituent at this position will en- hibitory properties. These results provide a tool for guiding
hance inhibitory potency. Compounds 15 and 23 with an further drug design and for predicting the affinity of related
electronegative group, (2E)-3-phenylallyl, exhibit higher ac- compounds. As a result, our studies provide useful indicators
tivity. A comparison of compounds 13 and 14 shows that a for guiding the further rational design of more potent new
change from two methyl to one methoxycarbonyl group at HEPT analogues.
the end of the allyl group increases the potency, which may
be due to an decrease in the steric bulk of the group. This
also can be seen from compounds 12 and 13.
Experimental
Melting points were measured on a WRS-1 digital melting point instru-
ment and are uncorrected. 1H-, 13C-NMR, and HMBC spectra were recorded
in dimethylsulfoxide-d6 (DMSO-d6) or chloroform (CDCl3) on a Brucker
DMX 500 MHz spectrometer. Chemical shifts are reported in d (ppm) units
Table 7. Physical Data of the Target Compounds
Compd.
NMR d (DMSO-d6 or CDCl3)
12
13
14
15
1H-NMR (DMSO-d6) d: 0.82 (3H, t, Jꢁ7.3 Hz), 1.99—2.08 (2H, m), 3.94—4.28 (2H, m), 4.85—4.98 (2H, m), 5.62—5.70 (1H, m),
7.66—9.13 (7H, m), 11.67 (1H, s). 13C-NMR (DMSO-d6) d: 13.82, 18.31, 47.01, 116.56, 118.03, 124.89—137.12 (11C), 145.86, 150.97,
163.49, 191.45.
1H-NMR (DMSO-d6) d: 0.80 (3H, t, Jꢁ7.3 Hz), 1.05 (3H, s), 1.32 (3H, s), 1.94—2.06 (2H, m), 4.12—4.17 (2H, m), 4.96 (1H, m), 7.66—
9.18 (7H, m), 11.62 (1H, s). 13C-NMR (DMSO-d6) d: 12.84, 17.45, 20.78, 25.98, 39.78, 114.65, 115.02, 125.36—137.72 (11C), 145.17,
150.97, 163.59, 191.20.
1H-NMR (DMSO-d6) d: 0.81 (3H, t, Jꢁ7.2 Hz), 1.99—2.16 (2H, m), 3.60 (3H, s), 4.04—4.18 (2H, m), 5.68 (1H, d, Jꢁ15.9 Hz), 6.61—
6.66 (1H, m), 7.63—9.11 (7H, m), 11.70 (1H, s). 13C-NMR (DMSO-d6) d: 13.59, 18.67, 39.85, 53.48, 116.82, 123.59—137.42 (11C),
142.76, 145.34, 150.68, 163.93, 167.01, 191.47.
1H-NMR (CDCl3) d: 0.98 (3H, t, Jꢁ7.3 Hz), 2.16—2.31 (2H, m), 4.42 (2H, d, Jꢁ6.1 Hz), 6.02—6.05 (1H, m), 6.10 (1H, d, Jꢁ15.9 Hz),
7.02—7.20 (5H, m), 7.45—8.61 (7H, m), 9.34 (1H, s). 13C-NMR (DMSO-d6) d: 11.78, 15.49, 46.17, 118.86, 125.68—137.29 (17C),
145.06, 150.48, 163.75, 191.50.
16
17
18
19
20
21
1H-NMR (DMSO-d6) d: 0.80 (3H, t, Jꢁ7.2 Hz), 1.80—2.10 (2H, m), 4.65—4.87 (2H, m), 7.04—7.18 (5H, m), 7.52—9.04 (7H, m), 11.76
(1H, s,). 13C-NMR (DMSO-d6) d: 12.59, 18.58, 47.68, 117.57, 122.33—141.95 (15C), 146.13, 151.05, 163.97, 191.32.
1H-NMR (DMSO-d6) d: 0.79 (3H, t, Jꢁ7.3 Hz), 1.99—2.02 (2H, m), 3.78 (3H, s), 5.05 (2H, m), 6.91—8.94 (11H, m), 11.79 (s, 1H, NH).
13C-NMR (DMSO-d6) d: 13.00, 19.08, 55.87, 56.11, 119.98—135.98 (15C), 145.93, 151.03, 163.86, 165.20, 191.29, 195.69.
1H-NMR (DMSO-d6) d: 0.77 (3H, t, Jꢁ7.3 Hz), 1.99 (2H, m), 3.53 (3H, s), 4.20—4.41 (2H, m), 7.67—9.06 (7H, m), 11.85 (1H, s). 13C-
NMR (DMSO-d6) d: 13.04, 18.98, 45.53, 51.52, 115.67, 124.17—136.94 (10C), 145.73, 151.13, 163.68, 169.71, 191.42.
1H-NMR (CDCl3-d6) d: 0.94 (3H, t, Jꢁ7.4 Hz), 2.17 (1H, s), 2.25 (2H, m), 4.31—4.60 (2H, m), 7.55—8.53 (7H, m), 9.27 (1H, s). 13C-
NMR (DMSO-d6) d: 13.64, 19.03, 32.01, 70.87, 78.37, 115.92, 125.49—137.01 (10C), 143.65, 151.01, 163.73, 191.44.
1H-NMR (DMSO-d6) d: 0.79 (3H, t, Jꢁ7.3 Hz), 1.99 (2H, m), 3.09 (3H, s), 5.00 (2H, s), 7.65—9.08 (7H, m), 11.72 (1H, s). 13C-NMR
(DMSO-d6) d: 12.92, 19.01, 56.52, 77.11, 114.93, 125.17—136.98 (10C), 146.75, 151.06, 163.23, 191.35.
1H-NMR (DMSO-d6) d: 0.78 (3H, t, Jꢁ7.2 Hz), 1.99 (2H, m), 3.02 (3H, s), 3.11—3.73 (4H, m), 5.10—5.36 (2H, m), 7.70—9.10 (7H, m),
11.70 (1H, s). 13C-NMR (DMSO-d6) d: 12.87, 18.7, 59.56, 67.54, 73.01, 75.33), 115.03, 126.97—137.00 (10C), 146.86, 151.14, 163.34,
191.39.
22
23
25
26
1H-NMR (DMSO-d6) d: 1.14 (6H, d, Jꢁ6.8 Hz), 2.30 (1H, sep, Jꢁ6.8 Hz), 3.90 (1H, d, Jꢁ14.9 Hz), 4.26 (1H, d, Jꢁ14.6 Hz), 4.84—4.97
(2H, m), 5.61—5.68 (1H, m), 7.68—9.20 (7H, m), 11.55 (1H, s). 13C-NMR (DMSO-d6) d: 21.01, 24.79, 45.91, 117.31, 120.25, 126.97—
135.27 (11C), 142.55, 151.00, 163.79, 191.45.
1H-NMR (CDCl3) d: 1.18 (6H, d, Jꢁ6.9 Hz), 2.50 (1H, sep, Jꢁ6.9 Hz), 4.37—4.41 (2H, m), 6.00—6.02 (1H, m), 6.07 (1H, d,
Jꢁ15.9 Hz), 7.00—7.19 (5H, m), 7.46—8.10 (7H, m), 9.41 (1H, s). 13C-NMR (DMSO-d6) d: 20.98, 24.23, 45.69, 119.92, 126.11—
135.91 (17C), 142.05, 151.56, 163.58, 191.37.
1H-NMR (DMSO-d6) d: 1.16 (6H, d, Jꢁ6.8 Hz), 2.29 (1H, sep, Jꢁ6.8 Hz), 4.61 (1H, d, Jꢁ15.5 Hz), 4.85 (1H, d, Jꢁ15.5 Hz), 7.02—7.16
(5H, m), 7.54—9.12 (7H, m), 11.64 (1H, s). 13C-NMR (DMSO-d6) d: 20.65, 24.97, 47.96, 119.68, 126.79—136.76 (14C), 141.32, 141.97,
150.98, 163.88, 191.35.
1H-NMR (DMSO-d6) d: 1.11 (6H, d, Jꢁ6.9 Hz), 1.15 (3H, t, Jꢁ7.0 Hz), 2.30 (1H, sep, Jꢁ6.9 Hz), 3.94 (2H, m), 4.13—4.37 (2H, m),
7.69—9.14 (7H, m), 11.73 (1H, s). 13C-NMR (DMSO-d6) d: 14.32, 21.78 (d), 28.76, 46.94 (NCH2), 119.63, 127.37—136.46 (10C),
141.32, 150.45, 163.90, 170.03, 191.77.
27
28
1H-NMR (DMSO-d6) d: 1.14 (6H, d, Jꢁ6.8 Hz), 2.31 (1H, sep, Jꢁ6.8 Hz), 3.14 (1H, s), 4.22—4.34 (2H, m), 7.69—9.23 (7H, m), 11.66
(1H, s). 13C-NMR (DMSO-d6) d: 21.64 (d), 28.03, 32.01, 70.87, 78.37, 119.92, 125.49—137.01 (10C), 142.65, 151.61, 163.73, 191.44.
1H-NMR (DMSO-d6): d: 1.13 (6H, d, Jꢁ6.8 Hz), 2.31 (1H, sep, Jꢁ6.8 Hz), 3.07 (3H, s), 4.96—5.04 (2H, m), 7.68—9.18 (7H, m), 11.61
(1H, s). 13C-NMR (DMSO-d6) d: 22.00 (d), 27.13, 56.51, 77.90, 119.42, 127.07—135.79 (10C), 141.55, 150.27, 163.36, 191.40

September 2006
1253
relative to the internal standard tetramethylsilane (TMS). Mass spectra were
General Procedure for Compounds 12—29 To a solution of 11a, b
obtained on a MAT-95 mass spectrometer. Elemental analyses were per- (2 mmol) in dry DMF (8 ml), anhydrous K2CO3 (2 mmol) was added. After
formed on a CARLOERBA 1106 instrument and the results of elemental stirring at room temperature for 0.5 h, halids (2.2 mmol) were added. The re-
analyses for C, H, and N were within ꢂ0.4% of the theoretical values. All action mixture was stirred for additional 24—36 h. Following filtering off,
chemicals and solvents used were of reagent grade and were purified and the filtrate was concentrated in vacuo to give a brown residue, which was pu-
dried by standard methods before use. All air-sensitive reactions were run rified by silica column chromatography using 50% ethyl acetate/hexane
under a nitrogen atmosphere. All reactions were monitored by TLC on pre- (compounds 14, 20, 21, 28) and 30% ethyl acetate/hexane (other com-
coated silica gel G plates at 254 nm under a UV lamp using ethyl pounds) as eluents.
acetate/hexane as eluents. Silica column chromatography separations were
obtained on silica gel (300—400 mesh).
5-Isopropyl-1-[(2E,6E,10E,14E,18E,22E,26E,30E)-3,7,11,15,19,23,27,
31,35-nonamethylhexatriaconta-2,6,10,14,18,22,26,30,34-nonaen-1-yl]-6-
Biological Activity Assays The anti-HIV activity and cytotoxicity were (1-naphthoyl)-2,4-pyrimidinedione (24) Yield: 1.07 g (58.3%), colorless oil.
evaluated against wild type HIV-1 strain IIIB and HIV-2 ROD in MT-4 cells 1H-NMR (CDCl3) d: 1.17 (6H, d, Jꢁ6.9 Hz), 1.55 (9H, s), 1.60 (21H, s),
using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide 1.98—2.08 (32H, m), 2.44 (1H, sep, Jꢁ6.9 Hz), 4.29 (2H, s), 4.95—5.13
(MTT) method.12) Briefly, virus stocks were titrated in MT-4 cells and ex- (9H, m), 7.54—8.18 (7H, m), 9.39 (1H, s). HR-MS (MALDI): 943.6682
pressed as the 50% cell culture infective dose (CCID50). MT-4 cells were (Calcd for C63H88N2NaO3 [MꢈNa]ꢈ: 943.6693).
suspended in culture medium at 1ꢉ105 cells/ml and infected with HIV at a
1,3-Di[(2E)-3-phenylallyl]-5-isopropyl-6-(1-naphthoyl)-2,4-pyrimidine-
multiplicity of infection of 0.02. Immediately after viral infection, 100 ml of dione (29) Yield: 0.39 g (36.1%), a yellow powder, 1H-NMR (CDCl3) d:
the cell suspension was placed in each well of a flat-bottomed microtiter tray 1.19 (6H, d, Jꢁ6.9 Hz), 2.52 (1H, sep, Jꢁ6.9 Hz), 4.40—4.43 (2H, m),
containing various concentrations of the test compounds. The test com- 4.77—4.79 (2H, m), 6.04—6.05 (2H, m), 6.38—6.41 (1H, m), 6.79 (1H, m),
pounds were dissolved in DMSO at 50 mM or higher. After 4-d incubation at
37 °C, the number of viable cells was determined using the MTT method.
Compounds were tested in parallel for cytotoxic effects in uninfected MT-4
cells.
7.01—7.33 (10H, m), 7.43—8.10 (7H, m). HR-MS (MALDI): 563.2309
(Calcd for C36H32N2NaO3 [MꢈNa]ꢈ: 563.2311).
References and Notes
3D-QSAR Model The structures were built in SYBYL 6.7 on a Silicon
1) For Part 6, see: Sun G. F., Kuang Y. Y., Chen F. E., Wang Y. P., De
Clercq E., Balzarini J., Pannecouque C., Arch. Pharm. Chem. Life Sci.,
338, 1—5 (2005).
Graphics Octane (R10000) workstation.19)
In our study, all the target compounds 12—29 were minimized using mo-
lecular mechanics with the Gasteiger–Huckel charges with a 0.005 kcal/mol
energy gradient covergence criterion.
Compound 23, the most active compound, was chosen as the template
molecule on which other molecules were aligned as described by Jiang and
Guo et al.20,21)
The anti-HIV-1 activity of the compounds was homogeneously expressed
as IC50 (M) and transformed to the pIC50 (ꢀlog IC50) values, which were used
as the target variable in PLS regression analyses to derive 3D-QSAR models
using the implementation in the SYBYL package.
The overlapped molecules were surrounded by a 3D grid of points in
three dimensions extending at least 4 Å beyond the union volume occupied
by the superimposed molecules. The CoMFA descriptors, steric and electro-
static field energies, were calculated using the SYBYL default parameters:
2 Å grid points spacing, an sp3 carbon probe atom with ꢈ1 charge and a van
der Waals radius of 1.52 Å, and column filtering of 2.0 kcal/mol.
General Procedure for 5-Alkyl-2,4-dimethoxy-6-(1-naphthoyl)pyrimi-
dine 10a, b 5-Alkyl-6-chloro-2,4-dimethoxypyrimidine 8a, b (100 mmol)
2) De Clercq E., Antiviral Res., 38, 153—179 (1998).
3) Baba M., Tanaka H., De Clercq E., Pauwels R., Balzarini J., Schols D.,
Nakashima H., Perno C. F., Walker R. T., Miyasaka T., Biochem. Bio-
phys. Res. Commun., 165, 1375—1381 (1989).
4) Miyasaka T., Tanaka H., Baba M., Hayakawa H., Walker R. T.,
Balzarini J., De Clercq E., J. Med. Chem., 32, 2507—2509 (1989).
5) Danel K., Larsen E., Pedersen E. B., Synthesis, 8, 934—936 (1995).
6) Tanaka H., Baba M., Hayakawa H., Sakamaki T., Miyasaka T., Uba-
sawa M., Takashima H., Sekiya K., Nitta I., Shigeta S., Walker R. T.,
Balzarini J., De Clercq E., J. Med. Chem., 34, 349—351 (1991).
7) Meng G., Chen F. E., De Clercq E., Balzarini J., Pannecouque C.,
Chem. Pharm. Bull., 51, 779—789 (2003).
8) Sun G. F., Chen X. X., Chen F. E., Wang Y. P., De Clercq E., Balzarini
J., Pannecouque C., Chem. Pharm. Bull., 53, 886—892 (2005).
9) Hopkins A. L., Ren J., Esnouf R. M., Willcox B. E., Jones E. Y., Ross
C., Miyasaka T., Walker R. T., Tanka H., Stammers D. K., Stuart D. I.,
J. Med. Chem., 39, 1589—1600 (1996).
and 1-naphthylacetonitrile (20.04 g, 120 mmol) were dissolved in anhydrous 10) Baddiley J., Topham A., J. Chem. Soc., 1944, 678—679 (1944).
DMF (300 ml) under nitrogen atmosphere. The solution was cooled to 11) Merkatz A. V., Ber. Dtsch. Chem. Ges., 52, 869—880 (1919).
ꢀ15 °C and NaH (4.80 g, 120 mmol) (60% in paraffin) was added portion-
wise. The resulting mixture was stirred for about 48 h at room temperature
12) Pauwels R., Balzarini J., Baba M., Snoeck R., De Clercq E., J. Virol.
Methods, 20, 309—322 (1988).
to give the corresponding intermediate nitriles 9a, b. Then without separa- 13) Ragno R., Mai A., Sbardella G., Artico M., Massa S., Cadeddu A.,
tion, air was immediately passed through the reaction mixture at room tem- Colla P., J. Med. Chem., 47, 928—934 (2004).
perature for 48—72 h. The henna solution was neutralized with acetic acid 14) Hannongbua S., Lawtrakul L., Quant. Struct.-Act. Relat., 15, 389—
and concentrated under reduced pressure to give a red black residue 10a, b,
which was used without any further purification for the next step.
General Procedure for 5-Alkyl-6-(1-naphthoyl)-2,4-pyrimidinedione
394 (1996).
15) Alves C. N., Pinheiro J. C., Camargo A. J., Ferreira M. M. C., Da Silva
A. B. F., J. Mol. Struct. (Theochem.), 530, 39—47 (2000).
11a, b To a stirred solution of 10a, b (90 mmol) in methanol (450 ml), 36% 16) Wold S., Rhue A., Wold H., Dunn W. J. I., SAIM J. Sci. Stat. Comput.,
HCl (225 ml) was added. The reaction mixture was refluxed for 24—48 h. A 5, 735—743 (1994).
yellow precipitate, which formed, was filtered off, washed with water, and 17) Wold S., Albano C., Dunn W. J., III, Edlund U., Esbensen K., Geladi
recrystallized from methanol/chloroform (5 : 1).
P., Hellberg S., Johanson E., Lindberg W., Sjostrom M., NATO ASI
Ser., Ser. C, 138, 17—95 (1984).
5-Ethyl-6-(1-naphthoyl)-2,4-pyrimidinedione (11a) Yield: 21.7 g (82.0%),
1
a light yellow powder, mp: 270.9—272.0 °C. H-NMR (DMSO-d6) d: 0.83 18) Clark M., Cramer I. R. D., Quant. Struct.-Act. Relat., 12, 137—145
(3H, t, Jꢁ7.3 Hz), 2.03 (2H, q, Jꢁ7.2 Hz), 7.66—9.00 (7H, m), 11.20 (1H,
s), 11.31 (1H, s). HR-MS m/z: 295.1086 (Calcd for C17H15N2O3 [MꢈH]ꢈ: 19) Sybyl, version 6.7; Tripos Associates; St. Louis, MO, Tripos, Inc.,
295.1083). 2000.
(1993).
5-Isopropyl-6-(1-naphthoyl)-2,4-pyrimidinedione (11b) Yield: 19.3 g 20) Hong L., Huang X. Q., Shen J. H., Luo X. M., Li M. H., Xiong B.,
(69.6%), a light yellow powder, mp: 228.8—231.5 °C. 1H-NMR (DMSO-d6)
d: 1.12 (6H, d, Jꢁ6.4 Hz), 2.49 (1H, sep, Jꢁ6.4 Hz), 7.69—9.22 (7H, m),
Chen G., Shen J. K., Yang Y. M., Jiang H. L., Chen K. X., J. Med.
Chem., 45, 4816—4827 (2002).
11.25 (1H, s), 11.38 (1H, s). HR-MS m/z: 309.1238 (Calcd for C18H17N2O3 21) Guo Y. S., Xiao J. F., Guo Z. R., Chu F. M., Cheng Y. H., Wu S.,
[MꢈH]ꢈ: 309.1239).
Bioorg. Med. Chem., 13, 5424—5434 (2005).