Enamino-oxindole HIV protease inhibitors

Article abstract of DOI:10.1016/j.bmcl.2012.05.120

We have designed and synthesized a series of HIV protease inhibitors (PIs) with enamino-oxindole substituents optimized to interact with the S2′ subsite of the HIV protease binding pocket. Several of these inhibitors have sub-nanomolar K_i and antiviral IC₅₀ in the low nM range against WT HIV and against a panel of multi-drug resistant (MDR) strains.

Full text of DOI:10.1016/j.bmcl.2012.05.120

Bioorganic & Medicinal Chemistry Letters 22 (2012) 5078–5083
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Enamino-oxindole HIV protease inhibitors
Michael Eissenstat , Tanya Guerassina, Sergei Gulnik , Elena Afonina à, Abelardo M. Silva, Douglas Ludtke,
⇑
Hiroko Yokoe, Betty Yu, John Erickson
Sequoia Pharmaceuticals, Inc., 267 Kentlands Blvd, #905, Gaithersburg, MD 10878, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We have designed and synthesized a series of HIV protease inhibitors (PIs) with enamino-oxindole sub-
stituents optimized to interact with the S20 subsite of the HIV protease binding pocket. Several of these
inhibitors have sub-nanomolar Ki and antiviral IC50 in the low nM range against WT HIV and against a
panel of multi-drug resistant (MDR) strains.
Received 29 March 2012
Revised 30 May 2012
Accepted 30 May 2012
Available online 7 June 2012
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
Oxindole
HIV
Protease
Inhibitors
Resistance
HIV protease inhibitors (PIs) comprise a class of potent antiret-
roviral compounds and are a major component of highly active
antiretroviral therapy regimens (HAART) for the treatment of
HIV-AIDS. Since Hoffmann-La Roche introduced the first HIV PI,
saquinavir (SQV) into the market in the mid-1990s a total of 10
HIV PIs were successfully introduced over the next decade culmi-
nating in the approval of darunavir (DRV) in 2006. In spite of their
intrinsic effectiveness in reducing viral burden and preventing dis-
ease progression in patients, the long term effectiveness of PIs has
been limited by drug resistance, toxic side effects and suboptimal
pharmacokinetic properties. Therefore, there is a persistent clinical
need for discovery and development of potent and broad spectrum
PIs with reduced toxicity for the treatment of multidrug resistant
(MDR) HIV strains.
Our efforts have been focused on the design of PIs that maintain
high potency across a broad spectrum of clinically drug resistant
mutant, as well as wild type, viruses. We discovered at the Na-
tional Cancer Institute (NCI) that certain analogs of amprenavir,
including DRV (Fig. 1), that contain a 3(R),3a(S),6a(R)-bis-tetrahy-
drofuranylurethane (bis-THF) group showed excellent resistance
profiles against drug resistant HIV protease mutants including
highly drug resistant clinical isolates.1,2 We participated in the
development of darunavir and its analogs originally when we were
at NCI and later at Tibotec.
The bis-THF group was initially incorporated into the design of
potent SQV analogs by Ghosh et al. at Merck & Co. as a replacement
for the bulky P3-quinaldic and polar P2-asparagine groups,3 and la-
ter extended to amprenavir analogs.4 Modeling and X-ray crystal-
lography studies showed that the potency of bis-THF SQV analogs
was enhanced through interactions between the fused ring oxy-
gens of bis-THF with backbone NH atoms of Asp29 and Asp30.5
Structure–activity relationship (SAR) and X-ray crystal structure
and modeling analyses for DRV and bis-THF DRV analogs indicated
that the bis-THF was critical, but insufficient, to achieve their
excellent resistance profiles.1,2 Careful analysis of wild type and
mutant enzyme structures with amprenavir and bis-THF analogs
revealed that the combination of bis-THF with the sulfonamide
backbone led to a 3-dimensional motif of hydrogen-bonding inter-
actions with the enzyme that was predicted to provide broad spec-
trum binding to drug resistant mutants.6 We therefore set out to
identify new chemical motifs that would maintain or improve on
the excellent potency and resistance properties of DRV.
DRV has a p-aminophenylsulfonamide group as its P20 substitu-
ent.7 Potent analogs that we had studied at the NCI and Tibotec had
p-methoxy substituents and 3,4-methylenedioxy substituents.8
Molecular modeling studies of the interactions made by these
compounds indicated that incorporating the N- or O-atom into a
ring to form a bicycloaromatic group should retain potency and
resistance profile. Furthermore the second ring could serve as a
template for substituents that would allow for additional interac-
tions to be made with the wild-type protease and its mutants. This
Letter will discuss the SAR of enamino-oxindole P20 substituents.9
It should be noted that researchers at Tibotec also recognized the
⇑
Corresponding author. Tel.: +1 301 695 1668.
E-mail address: eissenst@comcast.net (M. Eissenstat).
Current address: Beckman Coulter, Life Sciences Research, Miami, FL 33196, USA.
Current address: Life Sciences Global Assay and Application Development, Miami,
à
FL 33196, USA.
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.05.120

M. Eissenstat et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5078–5083
5079
OH
OH
O
H
N
H
N
O
O
N
O
O
N
S
S
O
O
O
O
NH2
NH2
O
O
Darunavir
Amprenavir
Figure 1. Structures of amprenavir and darunavir.
possible utility of various bicyclic substituents and patented a ser-
ies of alkylidene oxindoles during the course of this work.10
Enamino-oxindoles have been identified as active against vari-
ous therapeutic targets. For example, Sarges et al. produced a series
of enamino-oxindoles as GABAergics.11 Bramson et al. prepared a
series of enamino-oxindoles that were active cyclin dependent ki-
nase inhibitors. They also demonstrated that the dimethylamino
analog could be converted to other enamino-oxindoles.12 Wood
et al. prepared enamino-oxindoles as TrkA kinase inhibitors.13
Hesperadin is an enamino-oxindole that has been identified as an
Aurora kinase inhibitor.14 Moreau’s group prepared enamino-oxin-
doles as antiproliferatives and antibacterials, also using displace-
ment of dimethylamino analogs.15,16 More recently Kumar et al.
prepared a series of enamino-oxindoles as antimalarials, also using
displacement reactions of the dimethylamino analogs,17 and Roth
et al. identified enamino-oxindoles as clinically active VEGFR2
inhibitors.18
It was intriguing to us that aromatic amino and aromatic meth-
oxy substituents would have similar interactions with the enzyme
target since amines are usually protonated at pH <7 where the pro-
tease is most active, and under those circumstances would have to
be a hydrogen bond donor whereas a methoxy group can only
interact as a hydrogen bond acceptor. In DRV the presence of the
p-sulfonyl group and the fact that the amine is aromatic decreases
the basicity so that it can remain unprotonated and can act as a
hydrogen bond acceptor. X-ray crystal structures of DRV with
HIV protease and mutants suggest that the amino group can act
as an H-bond donor or acceptor since it is found to be within H-
bonding distance of some combination of Asp30’ CO, Asp30’
CO2ꢀ, and Asp30’ NH.19 Weber’s group has also published a com-
parison of DRV binding to an inhibitor with an ether O at that posi-
tion, showing a slight shift in position.20 For oxindoles, the NH of
the oxindole can only be an H-bond donor unless the imino tauto-
mer of the oxindole served as the binding species.
The enamine analogs were readily prepared from oxindole 7
(Scheme 1) which was prepared by condensation of bis-THF amino
analog 5 and oxindole sulfonyl chloride 6. Enamino-oxindoles
could then be prepared by heating 7 in the presence of DMF
dimethylacetal or related acetals to form the enamines (Method
A).9 The amine portion of the enamines could be readily varied
by an exchange reaction of dimethamino analog 8 with excess
amine (Method B).21 For primary amines this exchange reaction
proceeded at room temperature in the presence of excess amine
(or amine hydrochloride) and triethylamine; for secondary amines
elevated temperature was required.
The enamines generated are vinylogous amides, thus increasing
their stability greatly over simple enamines. Since the enamines
derived from primary amines could form an additional intramolec-
ular H-bond as well as additional tautomers, they were the more
stable. The enamines generated were generally a mixture of (E)-
and (Z)-isomers. Attempts to separate these isomers were unsuc-
cessful since they appear to readily interconvert. If this equilibrium
is fast relative to binding there would be little impact on the
biological results. In enamines derived from primary amines the
(Z)-form that allows intramolecular H-bonding with the oxindole
carbonyl appeared to predominate.
Recombinant wild type and mutant HIV protease were ex-
pressed, purified and refolded as described previously.22 Inhibition
OH
2
OH
1
OH
4
O
Na2CO3
THF/H2O
77%
H
N
O
1) 4M HCl
O
3
O
O
BocHN
NH
BocHN
NCbz
CbzCl
O
NCbz
+
+
N
O
2) Cmpd 3
DIPEA
O
O
O
O
73%
OH
H
OH
H
O
Cl
O
O
NaHCO3
O
N
N
O
N
NH
H2/Pd/C
EtOH
S
O
S
+
O
O
O
O
NH
O
O
O
O
CH2Cl2/H2O
90%
5
N
H
7
6
97%
R
NR1R2
O
R
R
OMe
OMe
NMe2
O
Me2N
OH
OH
R1R2NH
H
H
N
O
O
O
N
N
O
N
S
S
O
O
NH
O
O
NH
O
Et3N
EtOH
(Method B)
O
CHCl3, reflux
O
O
9
8
75%
(Method A)
Scheme 1. Synthesis of enamino-oxindoles.

5080
M. Eissenstat et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5078–5083
Table 1
Preparation of enamino-oxindoles
Compound
R
R1
R2
Method
Temp °C
Time (h)
Yielda (%)
m/z (M+H)
9a
9b
9c
9d
8
9e
9f
9g
9h
9i
9j
9k
9l
9m
9n
9o
9p
9q
9r
H
H
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
H
H
Me
H
Me
Me
Et
Pr
iPr
Cyclopropyl
Bu
iBu
tBu
Cyclobutyl
Et
H
H
H
H
Me
H
H
H
H
H
H
H
H
Et
H
H
H
H
H
H
H
H
H
H
H
H
H
H
B
B
B
B
A
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
rt
rt
rt
rt
61
rt
rt
rt
rt
rt
rt
rt
rt
80
rt
rt
rt
83
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
2.5
2.5
3
2
1
3
5
23
19
4
83
80
60
51
75
54
30
43
40
51
42
44
48
45
70
46
47
27
47
55
36
30
32
50
29
44
27
30
615.4
629.4
629.4
643.4
643.4
643.4
657.4
657.2
655.2
671.4
671.5
671.5
669.5
671.6
685.3
699.4
711.4
691.2
705.5
706.5
726.5
745.4
756.6
719.4
659.4
673.2
686.5
734.5
2
120
18
29
22
5
22
22
5
23
22
18
22
24
24
20
3
CH2tBu
CH2CH2tBu
CH2C6H11
Ph
CH2Ph
9s
9t
CH2-4-Pyridyl
CH2-4-(2-Me-thiazolyl)
CH2-2-benzimidazolyl
CH2-2-quinolinyl
(CH2)2Ph
CH2CH2OH
CH2CH2OMe
CH2CH2NMe2
CH2CH2NHPh
9u
9v
9w
9x
9y
9z
9aa
18
a
Purified yield after chromatography.
Table 2
HIV protease inhibitory activity and antiviral activity of cyclized darunavir analogs
OH
H
N
O
O
N
X
S
O
O
O
NH
O
7 X=O
11
X= H2
Compound
HIV Protease Inh Ki (nM)
Antiviral IC50 (nM)
WT
M8
M9
M10
WT
M8
M9
M10
130
930
1700
M3
11
M4
30
M5
22
M6
10
M7
3.2
DRV
7
11
<0.10
<0.10
0.12
0.25
0.53
0.63
<0.30
0.95
1.5
1.0
14
10
3.0
82
3.9
16
150
107
25
110
190
80
175
120
30
7.5
constants were determined at pH 6.5 using a fluorogenic assay.
Data were fitted to the Michaelis–Menten equation for competitive
inhibitors. For tight-binding inhibitors the data were fitted by non-
linear regression analysis to the equation:
HIV protease/9d complex (1:5 molar ratio) was crystallized at
a protein concentration of 6 mg/mL in 50 mM sodium acetate,
pH 5.6, and 10 mM dithiothreitol. Crystals were grown at room
temperature by the hanging drop method, with 0.3 M ammo-
nium sulphate, pH 6.0, 5% methanol and 5% ethanol in the well.
Diffraction data collection and data processing used a Raxis
detector and an RU200 X-ray generator. The crystals belong t0o
the P6122 space group, and diffracted to resolution of 2.0 ÅA.
The programs Refmac,27 MAIN28 and Pymol29 were used for
refinement, modeling and visualization. The coordinates have
been deposited with the RCSB Protein Data Bank, with PDB ID
code 4FE6.
Compounds were evaluated for biological activity in enzyme
and cell culture-based antiviral assays using a panel of drug-resis-
tant HIV mutant enzymes and virus strains (M3–M10; Table 1,
Supplementary data) that showed phenotypic resistance to mar-
keted PIs. The number of resistance mutations varied between 10
and 17 amino acids, and all included I84V except for M7. We ob-
served a decline in potency at both the enzyme and cell level for
2
1=2
V ¼ V0=2Etðf½Kið1 þ S=KmÞ þ It ꢀ Etꢁ þ 4Kið1 þ S=KmÞEtg
ꢀ ½Kið1 þ S=KmÞ þ It ꢀ EtꢁÞ
where V and V0 are initial velocities with and without inhibitor, Km
is the Michaelis–Menten constant and S, Et and It are the concentra-
tions of substrate, active enzyme and inhibitor respectively.23
The antiviral assays were run in MT-4 cells obtained from the
AIDS Research and Reference Reagent Program. The potency of test
compound was determined as previously described with minor
modifications.24–26 Cells were exposed to 50% tissue culture infec-
tive doses (TCID50) of virus in the presence of various concentra-
tions of test compound for 5 days. 3-(4,5-dimethylthiazol-2-yl)-
2,5-diphenytetrazolium bromide (MTT) solution was added and
absorbance measured.

M. Eissenstat et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5078–5083
5081
Table 3
Alkyl-enamine SAR
R1
R
N
R2
OH
O
H
N
O
N
O
S
O
O
O
NH
O
Compound
R
R1
R2
HIV Protease Inh Ki (nM)
Antiviral IC50 (nM)
WT
M8
M9
M10
3.0
WT
300
M8
200
60
70
29
90
35
M9
200
M10
M3
M4
M5
M6
M7
9a
9b
9c
9d
8
9e
9f
9g
9h
9i
9j
9k
9l
9m
9n
9o
9p
9q
9r
H
H
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
H
H
Me
H
Me
Me
Et
Pr
iPr
Cyclopropyl
Bu
iBu
tBu
Cyclobutyl
Et
H
H
H
H
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
0.12
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
<0.10
0.10
<0.10
0.11
<0.30
<0.30
<0.30
<0.30
<0.30
<0.30
<0.30
<0.30
<0.30
<0.30
<0.30
0.72
<0.30
<0.30
<0.30
<0.30
<0.30
1.2
<0.30
<0.30
<0.30
<0.30
<0.30
<0.30
<0.30
<0.30
<0.30
<0.30
<0.30
84
180
93
60
56
70
80
120
60
300
40
33
200
115
41
27
32
40
>400
30
110
29
>400
27
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
<1.0
2.1
60
64
10
93
28
30
15
24
13
7.5
9.0
17
38
4.5
11
9.7
13
14
300
45
>400
35
22
50
60
17
65
27
40
19
25
11
9.5
25
31
21
9.5
11
9.0
100
14
220
30
>400
23
51
70
12
<0.10
<0.10
<0.10
<0.10
<0.10
0.11
70
16
75
18
65
20
30
5.0
Me
H
H
H
H
H
H
H
H
Et
H
H
H
H
H
H
H
H
H
H
H
H
H
H
40
15
8.0
31
8.0
30
32
25
16
35
30
40
13
23
27
110
35
270
40
>400
34
61
>400
70
>400
160
16
0.12
<0.10
<0.10
0.35
<0.10
<0.10
<0.10
<0.10
0.16
<1.0
<1.0
14
28
15
18
32
5.0
2.1
9.0
33
7.0
9.0
32
6.0
1.2
36
7.0
<1.0
<1.0
<1.0
<1.0
13
< 1.0
<1.0
<1.0
<1.0
<1.0
1.3
<1.0
1.2
1.7
3.0
1.0
CH2tBu
8.8
25
30
1.5
3.2
CH2CH2tBu
CH2C6H11
Ph
19
21
17
18
24
25
0.46
CH2Ph
<0.10
<0.10
<0.10
<0.10
<0.10
0.17
0.11
<0.10
0.57
13
21
20
40
20
35
18
25
4.1
9s
9t
CH2-4-Pyridyl
CH2-4-(2-Me-thiazolyl)
CH2-2-benzimidazolyl
CH2-2-quinolinyl
CH2CH2Ph
CH2CH2OH
CH2CH2OMe
CH2CH2NMe2
CH2CH2NHPh
9u
9v
9w
9x
9y
9z
9aa
DRV
ATV
<0.10
<0.10
0.11
<0.10
<0.10
<0.10
<0.10
28
800
42
>400
80
25
110
400
48
>400
120
130
3300
>400
70
>400
80
0.39
0.25
2.8
3.0
3.5
11
350
30
600
22
280
10
430
3.2
105
171
750
2500
both the indoline 11 and the oxindole 7 relative to DRV (Table 2).
The potency reductions were more pronounced with the mutants,
with the exception of the oxindole, which was much less potent
against the wild-type virus (82 vs 3.0 nM) than expected based
on its enzyme activity (<0.10 nM).
The antiviral assay allowed us to differentiate compounds that
were too potent to be distinguished in the enzyme assay, and also
indicated a requirement for lipophilicity that was not evident in
the enzyme SAR. For example the unsubstituted enamine 9a, while
very potent at the enzyme level, was relatively weak in the antivi-
ral assay. However, introduction of a methyl substituent (9b, 9c)
increased potency considerably against both wild type and mu-
tants. Potency continued to improve with increasing lipophilicity
until reaching a plateau above three carbons. Similarly, introducing
polar groups onto the amine (9s, 9u, 9x–9aa) caused a reduction of
potency in the antiviral assay. The isobutyl and neopentyl analogs
9j and 9n appear slightly more potent than the rest and appear to
have a slightly better overall resistance profile in the antiviral as-
says than DRV and much better than atazanavir (ATV). The most
potent analog with a potential solubilizing side chain was quinoli-
nyl analog 9v.
Paradoxically, the oxindole analogs that were more potent than
DRV against the drug resistant mutants in the antiviral assays
exhibited similar to weaker potency against wild type HIV, despite
being highly potent against the wild type enzyme. This effect was
even more pronounced for less lipophilic analogs (e.g., 9a–c,
Table 3) and resulted in the appearance of a flat resistance profile
in the cell-based assays. One could thus say that these compounds
have an excellent resistance profile as measured by the ratio of the
potencies against the mutant and wild-type viruses. However, the
Modeling experiments suggested that increasing planarity of
the indoline ring by adding a double bond would be accommo-
dated by both wild type and mutant enzymes, and would facilitate
the design of substituents that could provide additional points of
interaction with the enzyme. We found (Table 3) that a wide vari-
ety of alkyl-enamine analogs exhibited high potency against wild
type and multi-drug resistant HIV strains. Enamines derived from
unsubstituted (9a), monosubstituted (9b, 9e–9l, 9n–9p), and
disubstituted (8, 9m) amines were tolerated. Alkylidene enamines
(9c, 9d) were also tolerated. Of the variety of alkyl-enamines
examined, only the t-butyl analog where the bulky substituent is
placed close to the double bond was significantly less potent
against the mutants at the enzyme level. However, introduction
of alkylene groups between the t-butyl group and the double bond
(9n and 9o) restored potency against the mutants. Similar results
were seen for aryl substituted compounds. The phenyl analog
(9q) was not well tolerated by the mutants, but a variety of aralkyl
and heteroaralkyl substituents (9r–9w) were. Hydroxy and alkoxy
substituents were tolerated at the enzyme level, but amino substit-
uents less so.

5082
M. Eissenstat et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5078–5083
Figure 2a. X-ray crystal structure of the HIV protease complex of 9d showing
Figure 2b. H-bond interactions of 9d with HIV protease.
twofold related images of the inhibitor and 2 Asp30 conformations.
flat profile might be related to potency limitations due to the max-
imum achievable intracellular free drug concentration for these
compounds. In biochemical terms:
IC50 = f ꢂ [Ki ꢂ (1 + S/Km) + 0.5E] where Ki, S, Km and E are inhibi-
tion constant, substrate concentration, Michaelis–Menten con-
stant, and enzyme concentration, and f is a composite correction
factor that accounts for the protein binding, efficiency of intracel-
lular transport, etc. For very potent inhibitors, IC50 ꢃ f ꢂ 0.5E. It
is therefore tempting to speculate that alkyl-enamine oxindoles
represent a series of extremely potent HIV PIs and that the flat
antiviral resistance profile for some is determined by the extracel-
lular concentration that allows achievement of 0.5E intracellular
concentration.30
In order to better understand the interactions that the enamino-
oxindoles were making with HIV protease the structure of its com-
plex with inhibitor 9d was determined by X-ray crystallography
(Fig. 2a). The complex crystallized in space group P6122, with a
crystallographic twofold axis as molecular twofold axis. Thus there
are twofold related images of the inhibitor, but not of the protein,
since it in itself is mostly twofold symmetric. However, as in Tie
et al. (PDB code: 2IEN),31 we observed two conformations for
Asp30, which most likely result from the breakdown of the twofold
symmetry induced by the inhibitor binding. In one of them the dis-
Figure 2c. Superposition of compound 9d (aqua colored carbons) with darunavir
structure (white carbons).
NH2 group could form an H-bond 0with Asp30 for one of the orien-
tations of the side chain (2.8–2.7 ÅA). Thus the most likely interpre-
tation of the structure is that the NH2 group is acting as an H-bond
acceptor for Asp30 NH and an H-bond donor for the Asp30 side
chain carboxyl and the carbonyl oxygen. However, for the other
orientation of Asp30 in the Tie structure and in the Surleraux paper
no interaction of the amino group with Asp 30 side chain is ob-
served. The importance of this interaction to DRV’s binding po-
tency is thus unclear.
tance between the oxindole O and the aspar0tyl O is too close, just
0
1.8 ÅA. In the other one that distance is 3.0 ÅA, which is consistent
with an H-bond interaction, provided that Asp30 is protonated.
Also, in that conformation the Asp30 carbonyl oxygen is closer to
0
0
The NH of the oxindole is within H-bonding distance of the NH
the oxindole NH, 2.7 ÅA, than in the first conformation, 2.9 ÅA. Our
interpretation of the model is that the first conformation of
Asp30 corresponds to the case when the bis-THF group of the
inhibitor is near Asp30 and the second one occurs when the oxin-
dole group is nearby.
0
0
of Asp30 (3.2 ÅA) and the CO of Asp30 (2.9 ÅA). However, as opposed
to the interactions noted above for DRV, the angles are consistent
only with the latter interaction. This suggests that the H remains
on the oxindole N (c or d in Fig. 3) rather than as an imino tauto-
As expected, the bis-THF oxygens made H-bonds with the main
mer. The carbonyl of the oxindole appears to form an H-bond to
0
0
0
0
the Asp30 side chain (3.0 ÅA). Such an interaction should only be fa-
vored if the side chain were protonated or if the oxygen of the
oxindole were protonated to form a hydroxy indole tautomer.
However, in the 9d structure there is a very nice intramolecular
H bond generated between the enamine NH and the oxindole car-
bonyl (Fig. 3 structure c). That H could just as easily be transferred
to the oxindole oxygen to give the tautomeric imino hydroxy in-
dole (Fig. 3 structure d). Both the oxindole form and hydroxy in-
dole form require Asp30 to be protonated to have an
energetically favorable H-bond interaction. Generally oxindoles
have only a very small percentage of hydroxy indole tautomer
present, but such tautomers have been seen in certain cases.32
Thus, Asp30 is likely protonated in our structure, while in the
DRV complex its protonation state is less clear.
chain NHs of Asp29 NH (3.0 ÅA and 3.3 ÅA) and Asp30 NH (3.4 ÅA)
(Fig. 2b). In Surleraux et. al.7 the structure of the protease/DRV
complex0 show interactions with the bis-THF and Asp29 of 3.1
0
and 2.9 ÅA and with Asp30 of 2.9 ÅA. In Tie, et al. the bis-THF O dis-
0
0
tances are 3.0 and 3.1 ÅA to Asp29 and 3.3 ÅA to Asp30. Thus the tight
binding interactions of the bis-THF with the protease main chain
found in DRV were maintained by the enamino-oxindoles.
When the structures of DRV by Tie et al. and compound 9d are
superimposed (Fig. 2c), the oxindole NH overlays closely with the
NH2 of DRV. In the Tie structure there are two possible orientations
of the inhibitor in the binding pocket, making similar interactions
with the protein and having two conformations for Asp30 side
chain. The amino group of DRV0 is at H-bondi0ng distance to the
CO and NH of Asp30 of 3.0–3.3 ÅA and 3.3–3.2 ÅA, respectively. The

M. Eissenstat et al. / Bioorg. Med. Chem. Lett. 22 (2012) 5078–5083
5083
Me
Me
Me
Me
Me
Me
Me
Me
N
O
N
N
O
N
O
H
H
H
H
O
H
N
H
N
N
N
H
b
c
a
d
Figure 3. Depiction of potential tautomers of the enamino-oxindole.
3. Ghosh, A. K.; Thompson, W. J.; Fitzgerald, P. M.; Culberson, J. C.; Axel, M. G.;
McKee, S. P.; Huff, J. R.; Anderson, P. S. J. Med. Chem. 1994, 37, 2506.
4. Ghosh, A. K.; Kincaid, J. F.; Cho, W.; Walters, D. E.; Krishnan, K.; Hussain, K. A.;
Koo, Y.; Cho, H.; Rudall, C.; Holland, L.; Buthod, J. Bioorg. Med. Chem. Lett. 1998,
8, 687.
5. Ghosh, A. K.; Kincaid, J. F.; Walters, D. E.; Chen, Y.; Chaudhuri, N. C.; Thompson,
W. J.; Culberson, C.; Fitzgerald, P. M. D.; Lee, H. Y.; McKee, S. P.; Munson, P. M.;
Duong, T. T.; Darke, P. L.; Zugay, J. A.; Schleif, W. A.; Axel, M. G.; Lin, J.; Huff, J. R.
J. Med. Chem. 1996, 39, 3278.
6. Erickson, J.W.; Eissenstat, M.; Silva, A.; Gulnik, S. International Patent
Application WO 2003057173, 2003.
7. Surleraux, D. L.; Tahri, A.; Verschueren, W. G.; Pille, G. M.; de Kock, H. A.;
Jonckers, T. H.; Peeters, A.; De Meyer, S.; Azijn, H.; Pauwels, R.; de Bethune, M.
P.; King, N. M.; Prabu-Jeyabalan, M.; Schiffer, C. A.; Wigerinck, P. B. J. Med. Chem.
2005, 48, 1813.
The N-alkyl substituent of the enamine (or imine tautomer)
generally points out into solvent but lipophilic substituents could
potentially make hydrophobic interactions with the flap. It is diffi-
cult to speculate on the reason(s) for the reduction in potency for
some of our inhibitors with some of the mutants since the muta-
tions do not occur in close proximity to the N-alkyl substituent.
However, the lesser potency of the t-butyl and phenyl analogs
might be due in part to the strain induced by keeping everything
co-planar in the binding conformation that maintains the intramo-
lecular H-bond network. The lesser potency of the (protonated)
amino substituted side chains might relate to their effect on the
protonation state of Asp30 or disruption of the intramolecular H-
bond network of the inhibitor by introduction of an additional
source of an intramolecular H-bond.
Theoretically the energetically favorable accessibility of tauto-
meric forms of an inhibitor might be useful for favorable binding
to certain mutations of the target protein, such as an Asp to Asn
mutation, by allowing an exchange, for example, of an H-bond do-
nor interaction with an H-bond acceptor interaction. However we
have no clear indication that such an effect pertains in the current
series of inhibitors.
Enamino-oxindoles provide a series of potent HIV protease
inhibitors that are effective against a wide variety of drug resistant
mutants. Several compounds were potent against the mutants in
the enzyme and cell assays and thus have excellent resistance pro-
files. Others had flat resistance profiles in antiviral testing that ap-
pear to be due more to their cell partitioning than to inherently
similar binding potencies against the various mutants at the en-
zyme level. Preliminary pharmacokinetic studies in rats (data not
shown) suggested that this series of compounds had relatively
poor oral absorption and thus they were not pursued further.
8. Wigerinck, P. T. B. P.; Wang, G.; Eissenstat, M.; Erickson, J. W. International
Patent Application WO 2001025240, 2001.
9. This work was reported in part in Eissenstat, M.; Guerassina, T. International
Patent Application WO 2005087728, 2005 and in Gulnik, S. V.; Afonina, E. I.;
Yokoe, H.; Yu, B.; Guerassina, T.; Silva, A. M.; Erickson, J. W.; Eissenstat, M. 17th
International HIV Drug Resistance Workshop, 2008, Sitges, Spain, Antiviral
Therapy; 13 (Suppl. 3): A35.
10. Tahri, A.; Wigerinck. P. T. B. P. International Patent Application WO
2004016619, 2004.
11. Sarges, R.; Howard, H. R.; Koe, B. K.; Weissman, A. J. Med. Chem. 1989, 32, 437.
12. Bramson, H. N.; Corona, J.; Davis, S. T.; Dickerson, S. H.; Edelstein, M.; Frye, S.
V., ; Gampe, R. T., Jr.; Harris, P. A.; Hassell, A.; Holmes, W. D.; Hunter, R. N.;
Lackey, K. E.; Lovejoy, B.; Luzzio, M. J.; Montana, V.; Rocque, W. J.; Rusnak, D.;
Shewchuk, L.; Veal, J. M.; Walker, D. H.; Kuyper, L. F. J. Med. Chem. 2001, 44,
4339.
13. Wood, E. R.; Kuyper, L.; Petrov, K. G.; Hunter, R. N., III; Harris, P. A.; Lackey, K.
Bioorg. Med. Chem. Lett. 2004, 14, 953–957.
14. Hauf, S.; Cole, R. W.; LaTerra, S.; Zimmer, C.; Schnapp, G.; ther Walter, R.;
Heckel, A.; van Meel, J.; Rieder, C. L.; Peters, J.-M. J. Cell Biol. 2003, 161, 281.
15. Sassatelli, M.; Debiton, E.; Aboab, B.; Prudhomme, M.; Moreau, P. Eur. J. Med.
Chem. 2006, 41, 709.
16. Bouchikhi, F.; Rossignol, E.; Sancelme, M.; Aboab, B.; Anizon, F.; Fabbro, D.;
Prudhomme, M.; Moreau, P. Eur. J. Med. Chem. 2008, 43, 2316.
17. Kumar, S. P.; Gut, J.; Guedes, R. C.; Rosenthal, P. J.; Santos, M. M.; Moreira, R.
Eur. J. Med. Chem. 2011, 46, 927.
18. Roth, G. J.; Heckel, A.; Colbatzky, F.; Handschuh, S.; Kley, J.; Lehmann-Lintz, T.;
Lotz, R.; Tontsch-Grunt, U.; Walter, R.; Hilberg, F. J. Med. Chem. 2009, 52, 4466.
19. Kovalevsky, A. Y.; Tie, Y.; Liu, F.; Boross, P. I.; Wang, Y. F.; Leshchenko, S.;
Ghosh, A. K.; Harrison, R. W.; Weber, I. T. J. Med. Chem. 2006, 49, 1379.
20. Wang, Y.-F.; Tie, Y.; Boross, P. I.; Tozser, J.; Ghosh, A. K.; Harrison, R. W.; Weber,
I. T. J. Med. Chem. 2007, 50, 4509.
21. Park, S. J.; Cho, C. G.; Lee, K. I. Bull. Korean Chem. Soc. 2004, 25, 349.
22. Gulnik, S. V.; Suvorov, L. I.; Liu, B.; Yu, B.; Anderson, B.; Mitsuya, H.; Erickson, J.
W. Biochemistry 1995, 34, 9282.
23. Williams, J. W.; Morrison, J. F. Methods Enzymol. 1979, 63, 437.
24. Kodama, E.; Shigeta, S.; Suzuki, T.; De Clerq, E. Antiviral Res. 1996, 31, 159.
25. Pauwels, R.; Balzarini, J.; Baba, M.; Snoeck, R.; Schols, D.; Herdewijn, P.;
Desmyter, J.; De Clerq, E. J. Virol. Methods 1988, 20, 309.
Acknowledgements
Initial quantities of compounds 1 and 2 were provided by Dr.
Pavel Majer. We thank Mr. Greg Delehanty for doing stability mea-
surements on selected enamino-oxindoles. We thank Dr. Andrey
Topin for help in the preparation of compound 11.
Supplementary data
26. Yoshimura, K.; Kato, R.; Kavlick, M. F.; Nguyen, A.; Maroun, V.; Maeda, K.;
Hussain, K. A.; Ghosh, A. K.; Gulnik, S. V.; Erickson, J. W.; Mitsuya, H. J. Virol.
2002, 76, 1349.
27. Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Acta Crystallogr., Sect. D 1997, D53,
240.
Supplementary data (details for the syntheses described in
Scheme 1 and full descriptions of the enzyme, antiviral, and crys-
tallography testing methodology) associated with this article can
be found, in the online version, at http://dx.doi.org/10.1016/
j.bmcl.2012.05.120.
28. Turk, D. Proceedings from the 1996 Meeting of the International Union of
Crystallography
Macromolecular
Computing
School,
http://
www.hhmi.swmed.edu/Manuals/main/MAIN_index.html.
29. DeLano, W. L. The PyMOL Molecular Graphics System, DeLano Scientific, San
Carlos, CA, USA. http://www.pymol.org.
References and notes
30. Gulnik, S. V.; Eissenstat, M. Curr. Opin. HIV AIDS 2008, 3, 633.
31. Tie, Y.; Boross, P. I.; Wang, Y. F.; Gaddis, L.; Hussain, A. K.; Leshchenko, S.;
Ghosh, A. K.; Louis, J. M.; Harrison, R. W.; Weber, I. T. J. Mol. Biol. 2004, 338, 341.
32. Friedrichsen, W.; Traulsen, T.; Elguero, J.; Katritzky, A. R. Adv. Heterocycl. Chem.
2000, 76, 85.
1. Erickson, J. W.; Gulnik, S. V.; Ghosh, A. K.; Hussain, K. A. International Patent
Application WO 9967254, 1999.
2. Yoshimura, K.; Kato, R.; Kavlick, M. F.; Nguyen, A.; Maroun, V.; Maeda, K.;
Hussain, K. A.; Ghosh, A. K.; Gulnik, S. V.; Erickson, J. W.; Mitsuya, H. J. Virol.
2002, 76, 1349.