Lead optimization of diarylpyrimidines as non-nucleoside inhibitors of HIV-1 reverse transcriptase

Article abstract of DOI:10.1002/cmdc.201000045

Antiviral agents: A series of CN-CH2-DAPY analogues were identified as novel non-nucleoside reverse transcriptase inhibitors (NNRTIs) against HIV-1. Most of the newly synthesized compounds exhibited strong activity against wild-type HIV-1.

Full text of DOI:10.1002/cmdc.201000045

DOI: 10.1002/cmdc.201000045
Lead Optimization of Diarylpyrimidines as Non-nucleoside Inhibitors of
HIV-1 Reverse Transcriptase
Zhao-Sen Zeng,[a] Yong-Hong Liang,[a] Xiao-Qing Feng,[a] Fen-Er Chen,*[a, b] Christophe Pannecouque,[c]
Jan Balzarini,[c] and Erik De Clercq[c]
Over the past few years, considerable efforts have been devot-
ed to the structural modification of diarylpyrimidines (DAPYs),
a family of non-nucleoside reverse transcriptase inhibitors
(NNRTIs) with remarkable anti-HIV-1 activity, leading to the de-
velopment of etravirine (1), rilpivirine (TMC278, 2) and other
highly potent compounds against both wild type and mutant
strains of HIV-1 reverse transcriptase (RT).[1–12]
velop new DAPYs with improved activity against wild-type and
mutant strains of RT,[12] we became interested in the influence
of a cyano substituent, the most strongly electron withdrawing
group,[13] on the CH2 linkage of CH2-DAPYs. In the present
work, the synthesis of the CH2-linker-modified DAPYs (6a–ac)
and their anti-HIV activity are described.
The key intermediates 4-(4-chloro-pyrimidin-2-ylamino)ben-
zonitriles (10a–c) were prepared from the commercially avail-
able thiouracils according to our previously reported proce-
dures.[11] All of the cyano-CH2-DAPYs 6a–ac were synthesized
in 29–81% overall yields (four steps) via the condensation of
10a–c with the appropriately substituted aryl acetonitrile in
the presence of sodium hydride. The synthetic route is depict-
ed in Scheme 1.
These investigations incorporated a linker between the pyri-
midine moiety and the aryl group to the left, and substitution
of the same aryl ring: NH-DAPYs (3), O-DAPYs (4), CH2-DAPYs
(5). However, there are still unexplored possibilities in diversify-
ing the CH2 linker between the pyrimidine core and aryl
moiety of CH2-DAPYs. As part of our ongoing program to de-
Scheme 1. Synthetic route to new cyano-CH2-DAPYs (6a–ac) Reagents and
conditions: a) MeI, NaOH, H2O, RT, 24 h; b) 4-cyanoaniline, 180–1908C, 8 h;
c) POCl3, reflux, 30 min; d) R3-aryl acetonitrile, 60% NaH, Ar, anhyd THF, 48–
72 h.
Cyano-CH2-DAPYs 6a–ac were tested in MT-4 cells to evalu-
ate their cytotoxicity and their ability to inhibit the wild-type
HIV-1 (LAI strain, IIIB), HIV-1 double mutant K103N+Y181C and
HIV-2 strain ROD. Three FDA-approved drugs nevirapine (NEV),
delavirdine (DEV) and efavirenz (EFV) were also tested as refer-
ence drugs. As seen from the results listed in Table 1, most
cyano-CH2-DAPYs (6b, 6e, 6i–m, 6o–y, 6aa–ac) showed
potent inhibition of wild-type HIV-1 replication with EC50
values ranging from 195 to 1.8 nm, which were more potent
than those of nevirapine and delavirdine. In particular, com-
pound 6r displayed the highest potency with an EC50 value of
1.8 nm against HIV-1 and the greatest selectivity for the viral
target (SI=118595). It proved more active by twofold, 113-fold
[a] Dr. Z.-S. Zeng, Dr. Y.-H. Liang, Dr. X.-Q. Feng, Prof. F.-E. Chen
Department of Chemistry, Fudan University
Shanghai 200433 (RR China)
Fax: (+86)21-65643811
E-mail: rfchen@fudan.edu.cn
[b] Prof. F.-E. Chen
Institute of Biomedical Science, Fudan University
Shanghai 200433 (RR China)
[c] Prof. C. Pannecouque, Prof. J. Balzarini, Prof. E. De Clercq
Christophe Pannecouque, Jan Balzarini, Erik De Clercq
Rega Institute for Medical Research, Katholieke Universiteit Leuven
10 Minderbroedersstraat, 3000 Leuven (Belgium)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201000045.
ChemMedChem 2010, 5, 837 – 840
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
837

MED
antiviral activity. The substitution
on the aryl ring plays a signifi-
cant role in the antiviral activity.
The unsubstituted compound
6aa exhibited anti-HIV-1 activity
with an EC50 value of 33.8 nm.
When the para position of the
Table 1. Anti-HIV activity and cytotoxicity of compounds 6a–ac in MT-4 cells.[a]
Compd R1 R2
R3
EC50 [nm][b]
wild type (IIIB) HIV-2
CC50 [mm][c]
SI[d]
phenyl was substituted by
a
103N+181C
methyl group (6c), potency was
nearly the same as for 6aa
(EC50 =35.4 nm). Other para-sub-
stituents, such as fluoro, chloro,
bromo and methoxy, all resulted
in a decrease in potency com-
pared with 6aa. These data sug-
gested that a group with suita-
ble steric bulk in the para posi-
tion of the phenyl ring is re-
quired for a significant level of
anti-HIV-1 potency.
6a
6b
6c
6d
6e
6 f
6g
6h
6i
6j
6k
6l
6m
6n
6o
6p
6q
6r
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
4-F
2-F
4-Me
3-Me
2-Me
4-Cl
4869Æ1050
34.9Æ14.5
218Æ23.6
442Æ295
9.4Æ1.2
>34.8
<35.5
>34.4
>35.7
>36.6
>25.6
>26
>82.3
>25.1
>398
>33.1
ꢀ38.5
>35.9
ꢀ29.6
>44.6
>238.6
>287.3
ꢀ216.6
>3.4
>37.9
>249.9
>30.6
>49.3
>4.1
>18.7
>32.16
ꢀ37.2
>50.8
>58.1
–
>34.8
>35.5
>34.4
>35.7
>36.6
>25.6
>26
>82.3
>25.1
>398
>33.1
ꢀ38.5
>35.9
ꢀ29.6
ꢀ11.4
>238.6
>287.3
ꢀ41.1
>3.4
>37.9
>249.9
>30.6
>49.3
>4.1
>18.7
>32.16
ꢀ139.7
>50.8
ꢀ32.2
>15
34.8Æ4.2
35.5Æ6.1
34.4Æ15.9
35.7Æ15.2
36.6Æ1.1
25.6Æ3.6
26Æ3.2
23
983
157
83
3818
20
1250Æ361
1138Æ148
873Æ507
4-Br
4-MeO
23
93
82.3Æ43.7
25.1Æ9.6
>398Æ4.2
33.1Æ5.2
ꢀ38.5
2,3-CH=CH-CH=CH- 123Æ88
202
>392
43
H
800Æ431
760Æ535
52.3Æ21.5
18.5Æ13.9
469Æ293
118Æ6.4
3,5-dimethyl
4-Me
749
2-Me
4-MeO
H
35.9Æ5.1
ꢀ29.6
1940
ꢀ63
378
In two of the three possible
direct comparisons, a clear posi-
tional preference on the phenyl
ring was observed. The inhibito-
ry potency was higher for the
ortho-substituted derivative 6u
than for the meta-substituted
analogue (6t), which in turn was
more active than the para-sub-
stituted congener (6 s). Optimal
activity was obtained with com-
pound 6r with a 2-fluoro-substi-
tuted aromatic ring. It is also
worthwhile to mention that two
isomers 6y and 6z, with a naph-
thyl ring at the C4-position of
the pyrimidine ring, gave mark-
edly different anti-HIV-1 activi-
ties: 6y exhibited an EC50 value
of 69.3 nm, while the activity of
6z was ninefold less active.
44.6Æ2.8
ꢀ238.6
3,5-dimethyl
174Æ100
195.3Æ99
1.8Æ0.29
35.4Æ23.6
8.6Æ5.6
ꢀ1371
1467
ꢀ118595
93
4441
ꢀ35524
177
Me 4-F
Me 2-F
287.3Æ43.9
ꢀ216.6
6s
H
Me 4-Me
3.4Æ2.1
6t
H
Me 3-Me
37.9Æ5.5
ꢀ249.9
6u
6v
6w
6x
6y
6z
6aa
6ab
6ac
NEV
EFV
DEV
H
H
H
H
H
H
H
H
Me 2-Me
Me 4-Cl
Me 4-Br
Me 4-MeO
Me 2,3-CH=CH-CH=CH-
7.1Æ4.7
172.2Æ116.7
59.4Æ19.8
180.2Æ106.8
69.3Æ21.3
30.6Æ10.1
49.3Æ53.6
4.1Æ1.48
18.7Æ3.9
32.16Æ5.7
139.7Æ149.8
50.8Æ16.5
ꢀ58.1
832
27
273
Me 3,4-CH=CH-CH=CH- 594Æ396
51
Me
H
33.8Æ24.6
14.4Æ3.7
26.3Æ14.4
203Æ112.8
3.5Æ0.6
4044
3548
2189
>75
>1843
>62
Me 3,5-dimethyl
Me 2-Cl
H
>15
>6.3
>43.6
–
0.18
>43.6
719Æ328
–
[a] Data represent mean values of at least three separate experiments. [b] EC50: compound concentration (nm
or mm) required to protect against viral cytopathogenicity by 50% in MT-4 cells. [c] CC50: concentration of com-
pound that reduces the normal uninfected MT-4 cell viability by 50%. [d] SI: selectivity index: ratio CC50/EC50
(wild type).
and 399-fold than efavirenz, nevirapine and delavirdine,
respectively.
The activities of these compounds against the double
mutant HIV-1 strain K103N+Y181C and HIV-2 ROD virus were
also evaluated. Compounds 6o, 6r and 6ac showed weak ac-
tivity against the double mutant strains, and compound 6aa
exhibited low potency against the HIV-2 ROD virus. Other ana-
logues were found to be inactive (Table 1).
In terms of structure–activity relationships within these new
cyano-CH2-DAPYs, a methyl group at the C5-position (R2) of the
pyrimidine core increased the anti-HIV-1 activity; for example,
compounds 6q–ac were more active against HIV-1 RT with
EC50 values ranging from 1.8 to 594 nm than the corresponding
unsubstituted compounds 6l–p or C6-methyl substitution as
in 6a–k. These results suggest that the steric properties of the
methyl group in the C5-position of the pyrimidine moiety
might decrease the molecular flexibility to a small extent, and
this would be beneficial to keep the inhibitors in low-energy
conformations.
With the aim to investigate the binding mode of our newly
synthesized compounds, five compounds were docked into
the HIV-1 RT non-nucleoside binding site (NNBS) using Mole-
gro Virtual Docker (demo version).[14] Coordinates of the NNBS
were taken from the crystal structure of the RT/4-(4-(mesityl-
amino)pyrimidin-2-ylamino)benzonitrile (TMC120) complex
(PDB code: 1S6Q)[15] due to the high degree of structural simi-
larity between TMC120 and cyano-CH2-DAPYs.
Keeping the 5-methylpyrimidine motif as the optimum sub-
unit, a number of compounds with R3 substitutions to the left-
hand aryl ring were prepared (6q–ac) and evaluated for their
Figure 1 shows the theoretical binding mode of 6r to the
NNBS, and the overlay of the predicted binding modes of com-
838
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 837 – 840

In summary, a series of 4-(4-(cyano(alkyl)methyl)pyrimidin-2-
ylamino)benzonitriles was discovered as novel NNRTIs, and
their preliminary structure–activity relationships were estab-
lished through chemical modifications. In addition, modeling
studies were employed to better understand the interactions
between these inhibitors and the NNBS of HIV-1 RT and to
guide the structure–activity relationship studies. This investiga-
tion led to the identification of highly potent compounds
against wild type HIV-1 RT. In particular, compound 6r showed
the most potent activity against with an EC50 value of 1.80 nm
and a selectivity index of 118595.
Experimental Section
General procedure for the synthesis of 6a–ac: A solution of 10a–
c (2 mmol) in anhydrous THF (30 mL) was treated with the appro-
priate aryl acetonitrile (2.2 mmol) at RT and stirred for 0.5 h. NaH
(0.2 g, 4.8 mmol; 60% dispersion in mineral oil) was added portion
wise at RT under Ar and the reaction was stirred at RT for 48–72 h.
The resulting mixture was poured into water and extracted with
EtOAc. The combined organic layers were dried (Na2SO4), filtered
and concentrated in vacuo. Purification by flash chromatography
(silica gel; EtOAc/petroleum ether, 1:3) gave the final products 6a–
ac. Full protocols and characterization data for intermediates and
all final compounds can be found in the Supporting Information.
4-(4-(Cyano-(2-fluorophenyl)methyl)-5-methylpyrimidin-2-yl-
amino)benzonitrile (6r): (57%); mp: 203.9–204.68C; 1H NMR
(400 MHz, [D6]DMSO): d=2.16 (s, 3H, CH3), 6.23 (s, 1H, CH), 7.30–
7.51 (m, 4H, ArH), 7.60 (d, 2H, J=8.8 Hz, ArH2,6), 7.83 (d, 2H, J=
8.8 Hz, ArH3,5), 8.45 (s, 1H, H6), 10.26 (br s, 1H, NH); 13C NMR
(100 MHz, [D6]DMSO): d=14.2, 36.3, 102.8, 116.5, 118.2, 118.5 (2C),
120.0, 120.4, 120.9, 121.1, 125.9, 130.6, 131.7, 133.3 (2C), 145.2,
158.3, 160.9, 161.3; MS (ESI+): m/z 344 [M+H]+; Anal. calcd for
C20H14N5F: C 69.96, H 4.11, F 5.53, N 20.40, found: C 69.92, H 3.98, F
5.71, N 20.39.
Figure 1. a) Predicted binding mode of 6r to the NNBS of HIV-1 RT (PDB
code: 1S6Q); b) Overlay of predicted binding modes for compounds 6a
(gray), 6c (purple), 6 L (cyan), 6q (brown), 6r (nut-brown) and the experi-
mental binding mode of TMC120 (yellow) in the NNBS of HIV-1 RT.
Acknowledgements
We are grateful to the National Natural Science Foundation of
China (grant no. 20872018 and 30672536) for the financial sup-
port of this research.
pounds 6a, 6c, 6 L, 6q, 6r with the experimental binding
mode of TMC120 in the HIV-1 RT NNBS. These molecular mod-
eling studies suggest that compounds 6a, 6c, 6 L, 6q, 6r
probably fit the NNRTI site in a very similar manner to TMC120.
All analogues adopt a horseshoe mode-of-binding similar to
that seen for most of the DAPY compounds.[15] As such, the
critical interactions between inhibitor and binding site were re-
tained: namely the hydrogen bonds between the NH group
linking the pyrimidine and 4-cyanophenyl group and amino
acid residues Lys101 and Lys103, and the p–p interactions be-
tween the left-hand aromatic rings and aromatic amino acids
Tyr181, Tyr188, Phe227, Tyr318, and Trp229. In addition, the
cyano group is proximate to the aromatic amino acid residues
Tyr181 and Tyr188, and this might add further van der Waals
and p–p-stacking interactions within the binding pocket of RT.
These additional interactions would lead to the improved bind-
ing affinity and increased anti-HIV-1 activity of these inhibitors.
Keywords: HIV-1 · inhibitors · NNRTI · reverse transcriptase ·
structure–activity relationships
[1] D. W. Ludovici, R. W. Kavash, M. J. Kukla, C; Y. Ho, H. Ye, B. L. De Corte,
K.; Andries, M.-P. de Bꢁthune, H. Azijn, R. Pauwels, H. E. Moereels, J.
Heeres, L. M. Koymans, M. R. de Jonge, K. J. Van Aken, F. F. Daeyaert, P. J.
Lewi, K. Das, E. Arnold, P. A. J. Janssen, Bioorg. Med. Chem. Lett. 2001,
11, 2229–2234; Y. Ho, H. Ye, B. L. De Corte, K.; Andries, M.-P. de Bꢁthune,
H. Azijn, R. Pauwels, H. E. Moereels, J. Heeres, L. M. Koymans, M. R. de
Jonge, K. J. Van Aken, F. F. Daeyaert, P. J. Lewi, K. Das, E. Arnold, P. A. J.
Janssen, Bioorg. Med. Chem. Lett. 2001, 11, 2229–2234.
[2] S. G. Sarafianos, K. Das, S. H. Hughes, E. Arnold, Curr. Opin. Struct. Biol.
2004, 14, 716–730.
[3] K. Das, P. J. Lewi, S. H. Hughes, E. Arnold, Prog. Biophys. Mol. Biol. 2005,
88, 209–231.
[4] B. L. De Corte, J. Med. Chem. 2005, 48, 1689–1696.
[5] P. A. J. Janssen, P. J. Lewi, E. Arnold, F. Daeyaert, M. de Jonge, J. Heeres,
L. Koymans, M. Vinkers, J. Guillemont, E. Pasquier, M. Kukla, D. Ludovici,
K. Andries, M.-P. de Bethune, R. Pauwels, K. Das, A. D. C. Jr, Y. V. Frenkel,
ChemMedChem 2010, 5, 837 – 840
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
839

MED
S. H. Hughes, B. Medaer, F. D. Knaep, H. Bohets, F. D. Clerck, A. Lampo, P.
Williams, P. Stoffelsb, J. Med. Chem. 2005, 48, 1901–1909.
[12] Y. H. Liang, X. Q. Feng, Z. S. Zeng, F. E. Chen, J. Balzarini, C. Pannecou-
que, E. De Clercq, ChemMedChem 2009, 4, 1537–1545.
[6] J. Heeres, M. R. de Jonge, L. M. H. Koymans, F. F. D. Daeyaert, M. Vinkers,
K. J. A. Van Aken, E. Arnold, K. Das, A. Kilonda, G. J. Hoornaert, F. Com-
pernolle, M. Cegla, R. A. Azzam, K. Andries, M.-P. de Bꢁthune, H. Azijn, R.
Pauwels, P. J. Lewi, P. A. J. Janssen, J. Med. Chem. 2005, 48, 1910–1918.
[7] J. Guillemont, E. Pasquier, P. Palandjian, D. Vernier, S. Gaurrand, P. J.
Lewi, J. Heeres, M. R. de Jonge, L. M. H. Koymans, F. F. D. Daeyaert, M. H.
Vinkers, E. Arnold, K. Das, R. Pauwels, K. Andries, M.-P. de Bꢁthune, E.
Bettens, K. Hertogs, P. Wigerinck, P. Timmerman, P. A. J. Janssen, J. Med.
Chem. 2005, 48, 2072–2079.
[13] M. O. Sinnokrot, C. D. Shrill, J. Am. Chem. Soc. 2004, 126, 7690–7697.
[14] R. Thomsen, M. H. Christensen, J. Med. Chem. 2006, 49, 3315–3321.
[15] K. Das, Clark, A. D., Jr.; Jr. , P. J. Lewi, J. Heeres, M. R. De Jonge, L. M. H.
Koymans, M. Vinkers, F. Daeyaert, D. W. Ludovici, M. J. Kukla, B. De
Corte, R. W. Kavash, C; Y. Ho, H. Ye, M. A. Lichtenstein, K. Andries, R. Pau-
wels, M.-P. de Bꢁthune, P. L. Boyer, P. Clark, S. H. Hughes, P. A. J. Janssen,
E. Arnold, J. Med. Chem. 2004, 47, 2550–2560; Y. Ho, H. Ye, M. A. Lich-
tenstein, K. Andries, R. Pauwels, M.-P. de Bꢁthune, P. L. Boyer, P. Clark,
S. H. Hughes, P. A. J. Janssen, E. Arnold, J. Med. Chem. 2004, 47, 2550–
2560.
[8] C. Mordant, B. Schmitt, E. Pasquier, C. Demestre, L. Queguiner, C. Ma-
sungi, A. Peeters, L. Smeulders, E. Bettens, K. Hertogs, J. Heeres, P. Lewi,
J. Guillemont, Eur. J. Med. Chem. 2007, 42, 567–579.
[9] J. Heeres, P. J. Lewi, Adv. Antiviral Drug Des. 2007, 5, 213–242.
[10] E. De Clercq, Nat. Rev. Drug Discovery 2007, 6, 1001–1018.
[11] X. Q. Feng, Y. H. Liang, Z. S. Zeng, F. E. Chen, J. Balzarini, C. Pannecou-
que, E. De Clercq, ChemMedChem 2009, 4, 219–224.
Received: February 1, 2010
Published online on March 23, 2010
840
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 837 – 840