Oximinoarylsulfonamides as potent HIV protease inhibitors

Article abstract of DOI:10.1016/j.bmcl.2005.03.008

The need for a potent HIV protease inhibitor (PI) to combat emerging PI-resistant viruses is anticipated. Analogs formulated from the combination of structural fragments of Ritonavir, Lopinavir, and Amprenavir were synthesized. Analogs containing the oxime pharmacophore were found to have improved activities against both wild type and resistant (A17) viruses. The synthesis and structure-activity relationships (SAR) based upon the in vitro IC₅₀ of this series of compounds are reported.

Full text of DOI:10.1016/j.bmcl.2005.03.008

Bioorganic & Medicinal Chemistry Letters 15 (2005) 2275–2278
Oximinoarylsulfonamides as potent HIV protease inhibitors
Clinton M. Yeung,* Larry L. Klein, Charles A. Flentge, John T. Randolph, Chen Zhao,
MingHua Sun, Tatyana Dekhtyar, Vincent S. Stoll and Dale J. Kempf
Abbott Laboratories, GPRD, D-47D, Building AP52N, 200 Abbott Park Road, Abbott Park, IL 60064-3501, USA
Received 10 January 2005; revised 25 February 2005; accepted 3 March 2005
Abstract—The need for a potent HIV protease inhibitor (PI) to combat emerging PI-resistant viruses is anticipated. Analogs for-
mulated from the combination of structural fragments of Ritonavir, Lopinavir, and Amprenavir were synthesized. Analogs contain-
ing the oxime pharmacophore were found to have improved activities against both wild type and resistant (A17) viruses. The
synthesis and structure–activity relationships (SAR) based upon the in vitro IC50 of this series of compounds are reported.
Ó 2005 Elsevier Ltd. All rights reserved.
The introduction of the highly active anti-retroviral
therapy (HAART) in 1996 has helped to significantly re-
duce AIDS disease progression and mortality among
people living with HIV infection.1 However, a portion
of patients experience drug failure with resistance devel-
opment, and the need for an effective salvaging agent in
treating the HIV infections that are resistant to the cur-
rent PIs has become increasingly important.2a–c Recent
efforts in our research groups have focused on identify-
ing a universal salvage agent, particularly for the suc-
cessful HIV PI, Kaletra. By combining the structural
elements from Ritonavir, Lopinavir, and Amprenavir
(Fig. 1), we designed a novel series of compounds, which
demonstrate interesting SAR and antiviral activity. This
paper describes the optimization of the arylsulfonamide
substituents in 1 resulting in analogs exhibiting good
antiviral activities against viruses resistant to Lopinavir,
the major constituent of Kaletra.
The synthesis of compounds 8–23 (Table 1) is outlined in
Scheme 1. The sulfonamide bond in all but one (8) of
these analogs was formed by direct sulfonylation of the
secondary amine intermediate 7 with commercially avail-
able benzenesulfonyl chlorides. Amine 3 was prepared
Figure 1. Design of hybrid analog.
via the ring opening reaction of the commercially avail-
able epoxide 2 with isobutyl amine. Following removal
of the Boc group in 3 with TFA, the diamine was selec-
tively coupled under standard conditions with acid 63
to afford amine 7. Sulfonylation of 7 with various para-
substituted sulfonyl chlorides 4b–f gave sulfonamide
derivatives 11–14, and 17, respectively. An alternative
route to 8 was necessary due to the incompatibility of
*
the thiazole ring in 7 with the hydrogenolysis chemistry
required in a later step for this analog. In this case, amine
Corresponding author. Tel.: +1 847 937 1520; fax: +1 847 938
3403; e-mail: clinton.yeung@abbott.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.03.008

2276
C. M. Yeung et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2275–2278
Table 1. Antiviral activity of benzenesulfonamide analogs
O
OH
Ph
H
O
N
N
N
N
N
S
R2
O
O
S
R1
Compdsa
R1
R2
Wild type IC50, lM
A17 IC50, lM
8
NH2
H
H
H
H
H
H
H
H
H
H
H
H
H
H
0.031
0.168
1.19
0.311
2.41
10
9
10
NHCHO
NHSO2CH3
NHCOCH3
CO2H
11
12
0.312
10
1.52
9.44
13
14
COCH3
CN
0.039
0.148
0.254
0.143
0.101
0.051
0.016
0.005
0.653
0.045
0.016
0.035
0.109
0.067
0.68
1.169
0.944
0.523
0.957
0.206
0.099
0.031
3.01
15
16
CONH2
C(NH2)@NOH
CH@CH2
CHO
17
18
19
20b
CH2OH
CH@NOH
CH@NOCH3
H
21b
22
CH2OH
CH@NOH
0.91
23b
H
0.222
1.24
3.72
Lopinavir
Amprenavir
Ritonavir
3.44
a All compounds were tested once except the following compounds: 20 and all three reference compounds-1 duplicate.
b trans-isomer only.
3 was directly sulfonylated with 4a. Reduction with
PearlmanÕs catalyst afforded the Boc-intermediate 5.
The conversion to compound 8 was accomplished by re-
moval of the Boc group with TFA followed by an EDC
coupling with the carboxylic acid 6.
The first hybrid analog 8 that was made in this series
showed good wild type (WT) activity and approximately
4- to 12-fold improvement in IC50 against the resistant
(A17) virus4 over the control compounds, Lopinavir
and Amprenavir (Table 1). The anilino group in Ampre-
navir5 has been suggested to hydrogen bond with the
Asp 30 side chain of HIV protease, and the aniline
group in 8 was shown by X-ray crystallography to be
similarly positioned. The enhancements in resistance
profile are probably due to a combination of this inter-
action and the binding to the P3 elements.
In two cases the product 8 underwent further modifica-
tion. Derivatization of aniline 8 with acetic formic anhy-
dride gave formamide 9, while similar treatment of 8
with methanesulfonyl chloride afforded sulfonamide
10. Derivatization of nitrile 14 with hydroxylamine
hydrochloride gave a 2:3 mixture of compounds 15
and 16, respectively. Oxidative cleavage of olefin 17 with
osmium tetroxide and sodium periodate yielded the
aldehyde 18, which in turn, was reduced with sodium
borohydride to afford alcohol 19. Alternatively, treat-
ment of 18 with hydroxylamine hydrochloride gave
1:20 mixture of the cis and trans oximes 20. The cis
oxime was found to be thermodynamically labile under
acidic conditions relative to isomerization to the trans
isomer, and this conversion takes place during silica
gel purification. Because of the facility of this isomeriza-
tion, pure samples of the cis isomer were not isolable. In
a similar manner, aldehyde 18 was transformed into 21
by reaction with methoxylamine hydrochloride.
Further modifications of the aniline group in 8 resulted
in compounds that were much less active (formamide 9,
sulfonamide 10, and acetamide 11) with analogs exhibit-
ing 5- to 38-fold loss in activity against both WT and
resistant viruses activities. Acyl derivatives such as
methylketone 13 and aldehyde 18 had similar WT and
A17 activities (<2-fold changes) versus the anilino ana-
log 8, whereas the acid 12 showed poor antiviral activity
(10 lM) against both viruses. Amide 15 exhibited re-
duced potency compared to the aldehyde 18, and
showed a 3- to 8-fold loss in WT and A17 antiviral activ-
ities versus 8. Although the nitrile 14 and the olefin 17
are less sterically demanding groups, both show medio-
cre IC50 numbers. In an attempt to study the polarity
and positioning of a hydrogen-bonding group such as
the anilino group in the para-substituted benzenesulf-
onamide series, compound 19, was prepared. This ana-
log exhibited WT and A17 antiviral activities superior
to that of the aniline 8, perhaps reflecting a larger pocket
The meta oxime isomer can be obtained by sulfonylation
of 7 with 4g followed by diisobutylaluminum hydride
reduction to the meta-hydroxymethyl analog 22 which
was, in turn, oxidized with manganese dioxide and treat-
ed with hydroxylamine hydrochloride to give 23.

C. M. Yeung et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2275–2278
2277
Scheme 1. Reagents and conditions: (a) i-PrOH, 3 h, 80 °C, 97%; (b) (i) Et3N, DCM, 16 h, 90%, (ii) Pd(OH)2, H2, EtOAc, 4 h, 95%; (c) (i) 80% TFA/
DCM, 4 h, 96%, (ii) EDC, HOBT, DMF, 16 h, 61%; (d) (i) 80% TFA/DCM, 5 h, 98%, (ii) EDC, 16 h, 74%; (e) Et3N, DCM, 16 h, 22–72%; (f) THF,
1 h, 83%; (g) pyridine, DCM, 5 h, 87%; (h) Et3N, EtOH, 3 h, 50 °C; (i) 5:1 THF/H2O, 16 h, 82%; (j) EtOH, 1 h, 62%; (k) EtOH, 2 h, 41%; (l) EtOH,
DIEA, 3 h, 46%; (m) (i) Et3N, DCM, 16 h, 59%, (ii) DCM, ꢀ78 °C, 1 h, 80%; (n) (i) DCM, (ii) MeOH, 59% for two steps.
than previously thought or a conformational movement
of the protein/Asp 30 side chain.
in that replacement of this fragment with the meth-
yloxime as in 21 resulted in greatly reduced antiviral
activity. Other changes were not tolerated, such as the
hydroxyamidine analog, 16. The vector positioning of
the oxime group was also critical for activity as replace-
ment of the oxime at the meta-position as in 23 resulted
in a 3-fold loss in WT activity and greater than a 7-fold
loss in activity against resistant virus. The meta-hydroxy-
methyl 22 suffered a similar loss of antiviral activity.
The most active analog in this series was the correspond-
ing oxime 20 of aldehyde 18. This group imparted a 6- to
10-fold enhancement in potency over the aniline 8 versus
both the WT and A17 HIV activities and therefore, was
the best group in this series against resistant virus. The
presence of the proton of the oxime was very important
Figure 2. (a) Crystal structure of 20 bound to HIV protease indicating the oxime group projecting into S20 cavity. (b) Crystal structure of 20 bound
to HIV protease showing the H-bond of the oxime-OH with the side chain of Asp 30.

2278
C. M. Yeung et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2275–2278
Reagents and conditions: (a) 60 °C, i-PrOH, 79%; (b) (i)
ethanolamine, 16 h, rt, 73%, (ii) Boc2O, 97%; (c) (i) oxalyl
chloride, DMSO, (ii) L-ValOMe, NaCNBH3, 1 h, rt, 63%,
(iii) 50% TFA, 70%; (d) (i) CDI, THF, reflux, 3 h, 68%, (ii)
LiOH, 59%.
An X-ray crystal structure of oxime 20 bound in the
HIV protease active site (Fig. 2)6 revealed the potential
interaction of the oxime hydroxyl group to form a favor-
able hydrogen bond with the Asp 30 side chain carbonyl
with a distance of 2.66 A. This interaction may be
responsible for the excellent potency against both the
WT and A17 resistant viruses.
˚
4. A17 strain was generated by in vitro 17 passages with
Lopinavir/Ritonavir and contains 6 mutations (L10F/
V32I/M46I/I47V/Q58E/I84V). For assay procedure, see
Ref. 11.
In summary, we have discovered a novel series of hybrid
arylsulfonamide HIV PIs that exhibit good in vitro
potency against both the WT and A17 resistant viruses.
The oximino group was identified as the optimal moiety
in terms of resistant profile, and this activity may be
attributed to its constructive hydrogen bonding with
the Asp 30 side chain of the protease as shown by the
X-ray crystallographic study. Further modifications on
the left-hand side of the molecule are being pursued as
a way to further enhance the antiviral and pharmacoki-
netic properties of this series.
5. Kim, E. E.; Baker, C. T.; Dwyer, M. D.; Murcko, M. A.;
Rao, B. G.; Tung, R. D.; Navia, M. A. J. Am. Chem. Soc.
1995, 117, 1181.
6. HIV protease was purified and crystallized in the presence
of compound 21 according to the procedures described by
Stoll et al.7 Data were collected on a Rigaku RU200 X-ray
generator using a Mar 165 CCDdetector. Data were
processed using HKL2000.8 The co-crystals of HIV
protease and compound X belong to the orthorhombic
space group P21212 with unit cell dimensions
˚
˚
˚
a = 58.041 A, b = 85.954 A, c = 46.033 A. Calculation of
initial electron density maps and refinement were done
using CNX9,10 and refined to a final Rwork = 27.75 and
˚
Rfree = 35.31 to 3.0 A resolution. Coordinates for the
References and notes
crystal structure of protease complexed with 21 have
been deposited in the RCSB protein data bank, PDB ID
1YT9.
1. For a recent review of HIV protease inhibitors in therapy,
see: Randolph, J. T.; DeGoey, D. Curr. Top. Med. Chem.
2004, 4, 1079.
2. (a) Boden, D.; Markowitz, M. Antimicrob. Agents Che-
mother. 1998, 42, 2775; (b) Miller, Vernica J. Acq. Immun.
Def. Synd. 2001, 26, S34; (c) Shafer, R. W. Clin. Microbiol.
Rev. 2002, 15, 247.
7. Stoll, V.; Qin, W.; Stewart, K. D.; Jakob, C.; Park, C.;
Walter, K.; Simmer, R. L.; Helfrich, R.; Bussiere, D.;
Kao, J.; Kempf, D. J.; Sham, H. L.; Norbeck, D. W.
Bioorg. Med. Chem. 2002, 10, 2803.
8. Otwinowski, K. Z.; Minor, W. Methods in Enzymology.
In Macromolecular Crystallography, Part A; Carter, C.
W., Jr., Sweet, R. M., Eds.; Academic: New York, 1997;
Vol. 276, p 307.
9. Brunger, A. T.; Adams, P. D.; Clore, G. M.; DeLano,
J.-S.; Kuszewski, J.; Nilges, M.; Pannu, N. S.; Read, R. J.;
Rice, L. M.; Simonson, T.; Warren, G. L. Acta Crystal-
logr. Sect. D 1998, D54, 905.
10. Badger, J.; Berard, D.; Kumar, R. A.; Szalma, S.; Yip, P.;
Griesinger, C.; Junker, J. CNX software manual; Mole-
cular Simulations Inc.: San Diego, CA, USA, 1999
(www.accelrys.com).
3. Unpublished results.
Cl
Cl
S
N
Cl
b
a
+
NH2
S
O
O
H
N
Boc
c
N
N
H
OMe
N
N
OH
S
S
O
11. Mo, H.; Lu, L.; Dekhtyar, T.; Stewart, K. D.; Sun, E.;
Kempf, D. J.; Molla, A. Antiviral Res. 2003, 59,
173.
OH
d
N
N
N
O
6
S