Cyclic urea amides: Hiv-1 protease inhibitors with low nanomolar potency against both wild type and protease inhibitor resistant mutants of HIV

Article abstract of DOI:10.1021/jm960586t

Cyclic urea amides, a novel series of HIV-1 protease (HIV PR) inhibitors, have increased activity against drug-resistant mutants of the HIV PR. The design strategy for these inhibitors is based on the hypotheses that (i) the hydrogen-bonding interactions

Full text of DOI:10.1021/jm960586t

J . Med. Chem. 1997, 40, 181-191
181
Cyclic Ur ea Am id es: HIV-1 P r otea se In h ibitor s w ith Low Na n om ola r P oten cy
a ga in st both Wild Typ e a n d P r otea se In h ibitor Resista n t Mu ta n ts of HIV
Prabhakar K. J adhav,* Paul Ala, Francis J . Woerner, Chong-Hwan Chang, Sena S. Garber,
Elizabeth D. Anton, and Lee T. Bacheler*
DuPont Merck Research Laboratories, The DuPont Merck Pharmaceutical Company, Wilmington, Delaware 19880
Received August 13, 1996X
Cyclic urea amides, a novel series of HIV-1 protease (HIV PR) inhibitors, have increased activity
against drug-resistant mutants of the HIV PR. The design strategy for these inhibitors is
based on the hypotheses that (i) the hydrogen-bonding interactions between the inhibitor and
the protease backbone will remain constant for wild-type and mutant enzymes and (ii) inhibitors
which are capable of forming many nonbonded interactions, distributed throughout the active
site, will experience a lower percent change in binding energy as a result of mutation in the
target enzyme than those that form fewer interactions by partial occupation of the active site.
The cyclic urea amide, SD146, forms 14 hydrogen bonds and 191 van der Waals contacts to
HIV PR. SD146 is a very potent antiviral agent (IC90 ) 5.1 nM) against wild-type HIV and
maintains the same or improved level of high potency against a range of mutant strains of
HIV with resistance to a wide variety of HIV protease inhibitors.
In tr od u ction
Inhibition of the virally encoded protease has been
demonstrated to be an effective antiviral drug therapy
against HIV infection.1 However, emergence of resist-
ance to protease inhibitors may ultimately limit the
usefulness of such drugs. Monitoring of clinical trials
of HIV PR inhibitors has documented the emergence of
virus variants that are highly resistant to the prescribed
protease inhibitors and that, in some cases, are also
cross resistant to different structural classes of protease
inhibitors.2 More effective and durable antiviral therapy
using HIV PR inhibitors may depend on developing
highly potent inhibitors which are also equipotent
against mutant strains resistant to major classes of
protease inhibitors. In this article, we report the
discovery of cyclic urea amides that are equipotent
inhibitors of a set of viral strains selected for resistance
to the four HIV PR inhibitors that either are approved
for marketing (saquinavir, ritonavir, indinavir) or are
in advanced clinical development (VX478).
F igu r e 1. The C-R carbon trace of HIV-1 protease complexed
with the cyclic urea inhibitor SD146. The residues associated
with resistance are drawn as blue circles and the numbers
correspond to the sequence positions of the amino acids. The
atoms of each monomer of the homodimer are drawn in red
and green, and the inhibitor was omitted for clarity. The list
of residues drawn in blue circles is not a complete list of known
mutations.
presence of an antiviral drug, viable drug resistant
mutants would be expected to have a replicative ad-
vantage over the wild-type virus and could thus rapidly
dominate the HIV population.
Or igin of Resista n ce
The emergence of HIV resistance to chemotherapeutic
agents can be attributed to two major factors. HIV
reverse transcriptase is a highly error prone polymerase;
virus replication is dependent on this enzyme and may
result in as many as 3.4 × 10-5 mutations per base pair
per virus replication cycle.3 In addition, HIV replication
occurs at an extraordinarily high rate, and virus can
accumulate to a very high total body burden. It has
been estimated that in an infected person plasma virion
particles are produced at the rate of 10.3 × 109 virions
per day.4 These factors facilitate the emergence of HIV
drug resistance. It is very likely that mutations leading
to drug resistance arise spontaneously during HIV
replication but remain undetectable until the selective
pressure of antiviral therapy is initiated.5 In the
Mech a n ism of Dr u g Resista n ce
At the molecular level, a qualitative understanding
of the origin of drug resistance in HIV PR due to amino
acid (AA) changes in the active site of the enzyme may
be developed by examination of three-dimensional (3D)
structures of the enzyme-inhibitor complexes (Figure
1). For example, at position 82, the reduced sensitivity
of A, T, or F mutants to C2 symmetric inhibitors,
relative to the wild type virus, is due to decreased van
der Waals (VDW) interactions between the P1 group of
the inhibitor and residue 82 of the protease.6 However,
the precise structural role of mutations remote from the
active site, often observed in association with active site
mutations, is more difficult to rationalize. Some of these
mutations may represent changes which do not directly
confer drug resistance but compensate for AA changes
in the active site.7 In general, the origin of resistance
can be attributed to the modulation of hydrophobic (e.g.
* To whom correspondence should be addressed. E-mail: jadhavpk@
a1.lldmpc.umc.dupont.com and bachell@a1.lldmpc.umc.dupont.com.
X Abstract published in Advance ACS Abstracts, J anuary 1, 1997.
S0022-2623(96)00586-9 CCC: $14.00 © 1997 American Chemical Society

182 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 2
Sch em e 1. Synthesis of Nonsymmetric Cyclic Urea Amidesa
J adhav et al.
a
(a) 2.5 equiv of KOH, 2 equiv of (bromomethyl)cyclopropane, toluene, reflux, 86%; (b) 2.2 equiv of NaH, 1.6 equiv of 3-cyanobenzyl
bromide, DMF, 83%; (c) 10 equiv of KOH, 140 °C ethylene glycol, 97%; (d) 2 equiv of DDC, 1.5 equiv of HOBT, DMF, 1.5 equiv of
2-aminothiazole; (e) 2 N HCl in 1:1 dioxane/CH3OH, 36.2% through two steps.
48V; 50V; 82A, F, T; 84V) and electrostatic (e.g. 8Q, K,
30N) interactions between the inhibitor and AA side
chains in active site of the enzyme. With the exception
of 8Q and 30N mutants,7 the largest decrease in
sensitivity to the protease inhibitors is caused by
reduction of the hydrophobic interactions between the
inhibitor and AA side chains in the protease. In
contrast, mutations which result in the loss of electro-
static interactions between the backbone of the enzyme
and the inhibitor have not been described.
An inhibitor which derives its binding energy by
occupation of six subsites (eq 1) is likely to have broader
potency against mutant proteases than the inhibitor
which occupies four subsites of the enzyme (eq 2) since
the percent change in binding energy experienced by
the former as a consequence of protease mutation will
be lower than the latter. A solution to the drug
resistance problem may thus depend on the successful
design of inhibitors which are capable of forming the
maximum number H bonds to the protein backbone and
the maximum number of VDW contacts distributed
throughout the active site of HIV PR.
The first generation of protease inhibitors selected for
clinical evaluation was optimized for activity against the
wild-type enzyme. As drug resistant mutants were
identified for each major class of HIV PR inhibitor, the
need for inhibitors designed to be equally potent against
a range of resistant mutants became apparent. Our
approach to this problem is based on the assumption
that the protease of viable drug resistant viruses would
not display major alterations in the protein backbone
of the enzyme. Thus electrostatic interactions between
the backbone of the enzyme and the inhibitor would
remain constant for the wild-type and mutant enzymes,
even though interactions with side chain residues could
vary as a consequence of mutation. Consequently,
inhibitors designed to maximize hydrogen bonds (H
bonds) to the backbone of the wild-type enzyme might
retain effectiveness against mutant strains. Further-
more, we envisioned that inhibitors which are capable
of forming many nonbonded interactions (viz. electro-
static, VDW, ionic, dipole-dipole, π-π, π-cationic, etc.),
distributed throughout the active site, might have
superior resistance characteristics as compared to in-
hibitors which only partially occupy the active site and
therefore form fewer interactions with the enzyme. For
example in the case of an inhibitor which occupies six
subsites (S3, S2, S1, S1′, S2′, and S3′) of the enzyme,
the total binding energy (∆Gtotal) will be a sum of
binding energies of six subsites (e.g. ∆GS3, ∆GS2, etc.;
eq 1). Similarly, eq 2 can be used to represent the total
binding energy of an inhibitor which occupies four
subsites of the enzyme.
Syn th esis of Cyclic Ur ea Am id es
The cyclic urea class of inhibitors was designed to
displace the structural water molecule which has been
observed in complexes formed between HIV PR and
linear inhibitors.8 The synthesis and structure-activity
relationship of cyclic ureas have been recently dis-
closed.9,10 The nonsymmetric cyclic urea amides as
represented by Q8467 were synthesized as shown in
Scheme 1. The cyclic urea9 1 was alkylated sequentially
with (bromomethyl)cyclopropane followed by 3-cyano-
benzyl bromide to provide nonsymmetric bisalkylated
cyclic urea 3. Hydrolysis of the cyano derivative 3 to
carboxylic acid 4 proceeded in excellent yield under
aqueous alkaline conditions. Coupling of the intermedi-
ate 4 with 2-aminothiazole followed by deprotection of
the MEM (methoxyethoxymethyl) protecting group pro-
vided the target compound Q8467. The peptide coupling
reagents11 such as DCC, EDC, and BOP provide amides
in moderate to good yields. Alternatively, the amides
can be prepared first by conversion of the acid (e.g. 4)
to the corresponding acyl chloride by treatment with
oxalyl chloride followed by reaction with 2-aminothi-
azole, 2-aminoimidazole, or 2-aminobenzimidazole with
the resulting acid chloride. Carboxylic acid derivatives
with benzyl or n-butyl as P2 substituent were prepared
in an analogous manner as described for 4. Nonsym-
metric cyclic urea amides (Figure 2) were prepared by
a general procedure described for Q8467.
∆Gtotal ) ∆GS3 + ∆GS2 + ∆GS1 + ∆GS1′
+
Symmetric cyclic urea amides as exemplified by
SD146 were synthesized as shown in Scheme 2. Cyclic
urea 1 was alkylated with 3-cyanobenzyl bromide to
provide symmetric bisalkylated cyclic urea 5. Hydroly-
sis of the biscyano derivative 5 to biscarboxylic acid 6
∆GS2′ + ∆GS3′ (1)
∆Gtotal ) ∆GS2 + ∆GS1 + ∆GS1′ + ∆GS2′
(2)

Cyclic Urea Amides
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 2 183
described in a patient treated with MK639.2i The
resistance of a reconstructed virus containing the same
five AA changes (10R/63P/71V/82T/84V, designated
MK639 virus) to each of the cyclic urea protease
inhibitors was less than that of viruses containing 82F/
84V mutations, although resistance to MK63919 itself
was greater. We constructed viruses resistant to RO31-
8959 (48V/90M) and VX478 (46I/47V/50V), as well as
viruses with protease mutations reported to confer
resistance to A77003 (32I/82I and 8K/46I, Ho et al. and
Kaplan et al.2), and these were included in our panel of
drug resistant viruses (Table 1). The 32I/82I and 8K/
46I viruses showed no resistance to any of the protease
inhibitors we tested and are not described further.
Resista n ce P r ofiles of Cyclic Ur ea Am id es
We used constructed viruses to generate a “resistance
profile”, i.e. potency of a compound against the entire
panel of drug resistant mutant viruses. None of the
cyclic urea amides showed any significant loss of potency
against the 48V/90M (Table 2), the 32I/82I, or the 8Q/
46I viruses (data not shown). Indeed, some inhibitors
were reproducibly at least 5-fold more potent against
the 48V/90M mutant virus than against the HXB2 wild-
type virus. Resistance of other viruses with position
82A virus was low, less than 5-fold, for the cyclic urea
amides. Resistance of other position 82 and/or 84
mutants in general seemed to vary concordantly be-
tween the different cyclic ureas amides, with the 82F
virus the least resistant and the ABT538 virus the most
resistant. The mutant viruses with single AA changes
(82A, F or 84V) exhibited mild (2-8-fold) degrees of
resistance to the nonsymmetrical cyclic urea amides
(Q8467, XV655, SD143, SD145, SD152). However, the
ABT538 virus with five AA changes was more (14-86-
fold) resistant to these nonsymmetrical amides. Among
symmetrical cyclic urea amides, XV638 was 24-fold less
potent against ABT538 virus and 23-fold less potent
against the VX478 resistant virus (46I/47V/50V) as
compared to the wild-type. The resistance profiles of
XV652 and SD146 were exceptional. Both of these
compounds maintained full potency against all of the
mutant virus strains.
F igu r e 2. Chemical structures of cyclic urea amides.
proceeded in excellent yield under basic conditions.
Coupling of the intermediate with 2-aminobenzimid-
azole followed by deprotection of the MEM protecting
group provided the target compound SD146. Symmetric
cyclic urea amides (Figure 2) were prepared in an
analogous manner from 6.
Con str u ction of Recom bin a n t Mu ta n t HIV
Str a in s
A panel of isogenic recombinant mutant HIV strains
was constructed in the HXB2 viral background in order
to test the potency of the inhibitors against drug
resistant viruses (Table 1).12 The drug susceptibility
of these recombinant HIV strains was similar to that
reported2 for drug resistant viruses of similar genotype
isolated by in vitro or in vivo selection from a variety of
genetic backgrounds (Table 1). An 82A virus showed
little resistance to DMP3238,9 and DMP45013 but was
6-7-fold resistant to Q8024,14 a linear diol, and to
ABT538.15 An 82F virus showed moderate resistance
to DMP323 and DMP450 as well as to ABT538 and
Q8024. An 84V virus showed greater resistance to
DMP323 and DMP450 than did the 82F virus and
represents the highest degree resistance to cyclic urea
protease inhibitors we have observed as a result of a
single AA change from the wild-type HXB2 sequence.
Markowitz et al.7 reported the selection of a virus with
five AA changes in the protease gene which demon-
strated significant resistance to ABT538 and other
classes of protease inhibitors. A HXB2-based virus with
the same five AA changes (46I/63P/71V/82F/84V, des-
ignated ABT538 virus) in the protease gene showed
substantial resistance (7.3-93-fold) to the cyclic urea
protease inhibitors tested, as well as significant resist-
ance to ABT538 itself (30 X), MK63916 (6.5 X), and
VX47817 (13 X) but no change in IC90 for RO31-895918
(Table 1). A second combination of five AA changes
including changes at positions 82 and 84 has been
X-r a y Str u ctu r es of Cyclic Ur ea Am id es
In order to gain an understanding of the nature of
the interactions between the protease and the cyclic
urea amides, the structures of HIV PR complexed with
Q8467, XV638, and SD146 were determined. The
overall structures of these complexes were very similar
to those previously described for DMP3238 and DMP45013
complexed with HIV PR except that Q8467, XV638, and
SD146 contain groups that bind in a third subsite (S3).
The conformations of the three inhibitor molecules have
been clearly defined on the basis of the residual electron
density within the active sites of the proteases (Figure
3). A superposition of the bound inhibitor structures
revealed that the seven-membered cyclic urea ring and
the P1 and P1′ groups were not significantly altered by
changes in the size or chemical nature of the P2 and
P2′ groups (Figure 3). In each complex, the urea oxygen
formed two H bonds to the amide nitrogens of residues
Ile50/50′ of the flaps and the two diol oxygens formed a
total of four H bonds to the catalytic aspartates (Asp25/
25′) at the base of the pocket.

184 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 2
Sch em e 2. Synthesis of Symmetric Cyclic Urea Amidesa
J adhav et al.
a
(a) 6 equiv of NaH, 4 equiv of 3-cyanobenzyl bromide, DMF, 89%; (b) 11 equiv of KOH, 140 °C ethylene glycol, 95%; (c) 8 equiv of
EDC, 2.0 equiv of HOBT, Et3N, 8 equiv of 2-aminobenzimidazole, DMF; (d) 2 N HCl in 1:1 dioxane/CH3OH, 15.3% through two steps.
Ta ble 1. Resistance of Reconstructed Mutant Viruses to HIV-1 Protease Inhibitors
resistance of constructed mutant viruses (IC90 mutant/IC90 wt)
HIV
protease inhibitor
Ki (nM)a
IC90wt (nM)a
82A
82F
84V
ABT538 virusb
48V/90M
MK639 virusc
46I/47V/50V
Cyclic Ureas
6.6
XK234d
DMP323
DMP450
1.9
0.33
0.34
2100
95
120
1.4
2.8
2.5
4
7.2
93
49
1.3
0.3
1.5
11
18
27
1.7
11
7.5
7.1
5.3
22
9.9
Other
2.4
0.5
0.7
1.5
ABT 538
MK-639
Ro31-8959
VX478
0.37
0.37
0.25
0.16
0.25
150
28
15
81
195
7.1
2.5
0.9
2.2
6.0
8.9
2.8
0.4
1.9
1.8
30
5
3.6
20
1.1
1.3
5.5
18
5.3
6.0
1.8
ND
39
6.5
0.6
13
3.8
>150
Q8024
2.1
13
6.5
a
Ki’s for protease inhibitors against a purified wild-type HIV-1 protease were determined as previously described.9 IC90’s were measured
b
as described in biological methods. ABT538 virus carries five mutations, 46I/63P/71V/82F/84V, and is named for its significant resistance
to ABT538, while MK639 virus (c) carries five mutations, L10/46I/63P/82T/84V, and shows resistance to MK-639. XK234 is a cyclic urea
d
with cyclopropylmethyl at P2/P2′; also see ref 8, Table 1, compound 2 for the structure of XK234.
Ta ble 2. Resistance of Reconstructed Mutant Viruses to Cyclic Urea Amides
resistance of constructed mutant virusesb (IC90 mutant/IC90 wt)
no. of H-bond
inhibitor
interactionsa
Ki (nM)b
IC90 (nM)b
82A
82F
84V
ABT 538 virus
48V/90M
MK639 virus
46I/47V/50V
Q8467
XV655
XV651
SD143
SD152
SD145
XV638
XV652
SD146
10
10
11
11
11
11
12
14
14
0.058
0.074
0.016
0.033
0.025
0.037
0.027
0.014
0.024
40
74
28
21
18
17
4.2
19
5.1
1.8
4.9
NDc
1
1.4
1.9
2.2
2.7
0.8
4.1
3.0
ND
0.5
14.5
2.1
0.9
6.1
7.2
ND
2.9
4.2
8.4
1.2
0.7
0.6
41
31
83
29
27
86
24
0.2
1.2
0.9
1.9
ND
0.6
1
1.9
0.2
0.1
0.3
18.5
10
12
15
14.5
26
8.7
0.4
0.7
11
10.2
ND
ND
ND
2.1
23
1.6
0.3
0.4
1.0
a
b
Except for Q8467, XV638, and SD146, the number of H bonds were estimated from their models. Ki’s, IC90’s for wild-type virus, and
resistance of mutant viruses were measured as described in Biological Methods. c ND ) not determined.
Examination of the structures of these complexes
revealed differences between the individual inhibitor-
protease complexes in the number of electrostatic and
hydrophobic interactions formed. Q8467 is a nonsym-
metric molecule that binds in two different orientations
related by a pseudo-2-fold axis that is coincident with
the carbonyl double bond of the cyclic urea moiety. The
two nitrogen atoms near the amide junction point
toward the flaps, and the carbonyl oxygen and the sulfur
atom point toward the base of the pocket. The nitrogen
atoms form two H bonds to Gly48′ (CdO; NH) while the
carbonyl oxygen forms two hydrogen bonds to Asp29′
(NH) and Asp30′ (NH) (Figure 4). The sulfur atom of
the thiazole ring is located near the side chain of Asp30′.
However, no hydrogen bond could be formed between
the sulfur atom and carboxy group of Asp30′ (Figure
4). The cyclopropyl group of P2 is much smaller than
that of P2′ and is only involved in five VDW contacts
with the protease. A total of 10 possible H bonds and
111 VDW interactions were formed between Q8467 and
the enzyme (Figure 4 and Table 3). The symmetrical
bisthiazole amide XV638 binds symmetrically to HIV
PR with a pseudo-2-fold axis nearly coincident with the
pseudo-2-fold axis relating the homodimer of HIV PR.
The minor deviations from C2 symmetry are due to
small differences in the torsion angles of the substitu-
ents. The electrophilic atoms of the P2 are arranged in
a linear fashion just below the flaps analogous to an

Cyclic Urea Amides
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 2 185
F igu r e 3. SD146 bound in the active site of HIV PR. The residues that interact with SD146 are drawn in blue (Table 3) and the
dashed lines indicate hydrogen bonds. The C-R carbons of the enzyme are drawn in red, the numbers correspond to the sequence
positions of the amino acids, and a |2Fo - Fc| simulated annealed omit map of SD146 at 1.0σ is shown in pink.
F igu r e 4. Schematic diagram showing the hydrogen bond interactions between the HIV PR and the inhibitors, DMP323, Q8467
(one of two orientations), XV638, and SD146.
additional strand of a â-sheet. This arrangement
generates six H bonds to the flap residues (Gly48/48′,
Ile50/50′), and six H bonds to the residues at the base
of the pocket (Asp25/25′, Asp30/30′) (Figure 4 and Table
3). The bound conformation of SD146 is very similar
to that of XV638 with the addition of solvent-exposed
phenyl moieties at P3/P3′. SD146 forms two additional
hydrogen bonds to Asp30/30′ for a total of 14 H bonds,
which is the largest number described for a HIV PR
inhibitor complex. A detailed tabulation of the electro-
static and hydrophobic interactions formed between the
protease and inhibitors is given in Table 3.
four of six possible subsites of the enzyme (S1/S1′ and
S2/S2′) and formed 10 H bonds to Asp25/25′, Ile50/50′,
Asp29/29′, and Asp30/30′. We thought that it might be
possible to design more potent inhibitors that would
occupy all six subsites of the enzyme and would have
complementary groups for additional H bonds to the
backbone of the enzyme. The structure-activity rela-
tionship in a C2 symmetric diol (Figure 5) series of
protease inhibitors we studied previously suggested that
the amide bond between the P2-P3 junction was critical
for maintaining nanomolar binding affinity.14 The
significance of this amide bond between the P2-P3
junction became evident after examination of HIV PR
complexes with linear inhibitors.21 There is an array
of H bonds donors and acceptors at the S2-S3 junction
of the enzyme active site comprised of Asp29 (NH;
Discu ssion
X-ray crystallographic analysis of HIV PR complexed
with DMP323 revealed that the inhibitor occupied only

186 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 2
Ta ble 3. Interactions between HIV-1 and Inhibitorsa
J adhav et al.
a
The boxed letters identify the wild-type residues that contact (<4.1 Å) the inhibitors. The numbers correspond to the actual number
of contacts between one residue and the inhibitor. The first and second rows of numbers represent the distribution of contacts for each
monomer. (() Amount of accessible surface area, around the inhibitor, excluded on complex formation.
Asp30 suggested that replacing the sulfur atom with
an NH would result in the formation of two additional
H bonds one on each side of the C2 symmetric inhibitor.
Structures of HIV PR‚DMP323,8,20 HIV PR‚Q8467,
HIV PR‚XV638, and HIV PR‚SD146 complexes were
useful in developing the correlation between the non-
bonded interactions between inhibitor and enzyme and
the resistance profiles of cyclic urea amides. The
increase in potency against WT HIV and the retention
of potency against protease inhibitor resistant viruses
parallels the increasing number of H bonds formed in
the inhibitor enzyme complex. There are 10 H bonds
between HIV PR and DMP323, 10 between HIV PR and
Q8467, 12 between HIV PR and XV638, and 14 between
F igu r e 5. Schematic diagram showing the proposed hydrogen
bond interactions between the HIV PR and the C2 symmetric
inhibitors, P9941 and Q8024.
HIV PR and SD146. The carboxamide junction present
in the 3-(2-benzimidazolylaminocarbonyl)benzyl moiety
of number of these cyclic urea amides is unique. Every
heteroatom of the carboxamide junction forms a hydro-
gen bond with a nearby residue in HIV PR. SD146
forms a total of 14 H bonds with HIV PR of which 8 H
bonds are between the inhibitor and the backbone of
the enzyme. These H bonds to the backbone are likely
be maintained even in drug resistant mutant proteases.
There are 4 H bonds between the diol moiety and the
active-site aspartates. The loss of these H bonds in
viable drug resistant mutant viruses is unlikely because
the active-site aspartates are required for enzyme
activity and the formation of fully infectious virus
particles.22 The remaining two H bonds between the
NH of the benzimidazole ring and Asp30 side chains
could be lost as a result of mutation; enzymes with N
or Y substitutions have been reported to retain some
enzymatic activity.23 A viable virus containing a 30N
mutation has recently been reported following in vitro
selection against nelfinavir and observed in patients
experiencing a viral RNA load rebound on nelfinavir
therapy.24 Thus, at least 12 if not all 14 H bonds could
be maintained between SD146 and mutant proteases.
A comparision of the resistance profile of XV652 (14 H
bonds) and SD146 (14 H bonds) with XV638 (12 H
bonds) against ABT538 and MK639 viruses suggests
that the two additional H bonds in XV652 and SD146
contribute to broader potency against mutant viruses.
The relative potency of XV652 in an antiviral assay is
about 4-fold less than SD146; however, the resistance
profile of XV652 is as good as that of SD146. It should
COOH), Asp30 (NH; COOH), and Gly48(CdO; NH)
pointing toward the inhibitor binding cleft. Indeed, the
amide linkage between P2 and P3 of the linear inhibi-
tors made critical H bonds with Gly48 (CdO) and Asp29
(NH).11 We thought that it might be possible to capture
these hydrogen bonding interactions in the cyclic urea
series by attaching a P3 group via an amide linkage.
Attachment of P3 might also enable the formation of
additional VDW interactions.
Molecular modeling suggested that the 3-position of
the benzyl group of DMP323 on the P2 substituent in
the cyclic urea series was the most appropriate place
for attaching the P3 group. This led to the synthesis of
a cyclic urea amide Q8467 containing cyclopropylmethyl
at P2 and 3-(2-thiazolylaminocarbonyl) benzyl at P2′.
It is a potent inhibitor of HIV PR (Ki ) 0.058 nM) and
a potent antiviral agent against the wild-type virus (IC90
) 40 nM; Table 2). Symmetric cyclic urea carboxamide
XV638 containing 3-(2-thiazolylaminocarbonyl)benzyl
groups at both P2 and P2′ is an even more potent
inhibitor of HIV PR (Ki ) 0.027 nM) and the wild-type
virus (IC90 ) 4.2 nM). Structural studies of HIV
PR‚Q8467 and HIV PR‚XV638 complexes indicated that
the heteroatoms of the amide linkage as well as the
heteroatoms of the thiazole ring (CO, NH, S, and N) are
within the hydrogen-bonding distances to Asp30 (NH),
Gly48 (CO), Asp30 (COOH), and Gly48 (NH) respec-
tively (Figure 4). The proximity of the sulfur atom of
the thiazole moiety in XV638 to the carboxyl group of

Cyclic Urea Amides
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 2 187
poly(ethylene glycol) (MW 1000, 3.0 g), and potassium hydrox-
ide (85%) (6.9 g, 105 mmol) in dry toluene (200 mL) was heated
at reflux using a Dean-Stark apparatus until ∼21 mmol of
water was collected. A suspension resulted. The mixture was
cooled to room temperature and the treated with (bromo-
methyl)cyclopropane (25.5 g, 84 mmol) dropwise and heated
at 60 °C for 18 h. The mixture was poured into water, and
the organic layer was separated and washed with water (2 ×
100 mL). The organic layer was dried over MgSO4, filtered,
concentrated, and purified by flash chromatography (1 kg of
silica gel) using 2:3 EtOAc:hexane to afford 20.0 g of 2 as a
colorless oil (86% yield): 1H NMR (CDCl3) δ 0.09-0.10 (m, 2H),
0.36-0.44 (m, 2H), 0.90-0.97 (m, 1H), 1.93-1.98 (m, 1H), 2.88-
2.97 (m, 1H), 2.98-3.04 (m, 1H), 3.09-3.13 (m, 1H), 3.22-
3.55 (m, 3H), 3.37 (s, 6H), 3.57-3.63 (m, 4H), 3.69-3.78 (m,
3H), 3.85-4.09 (m, 4H), 4.60 (d, J ) 7.3 Hz, 1H), 4.86 (d, J )
9.9 Hz, 1H), 4.88 (d, J ) 9.9 Hz, 1H), 4.96 (d, J ) 7.7 Hz, 1H),
4.98 (d, J ) 6.9 Hz, 1H), 7.16-7.35 (m, 10H).
[4R-(4r,5r,6â,7â)]-3-[[3-(Cyclopr opylm eth yl)h exah ydr o-
5,6-b i s [(2-m e t h o x y e t h o x y )m e t h o x y ]-2-o x o -4,7-b i s -
(p h e n y lm e t h y l)-1H -1,3-d ia ze p in -1-y l]m e t h y l]b e n zo -
n itr ile (3). A solution of 2 (20.0 g, 36 mmol) in dry DMF (150
mL) was cooled in a 0 °C ice bath, treated with 60% sodium
hydride in mineral oil (3.2 g, 80 mmol), and stirred in the ice
bath for 15 min. The mixture was then treated with R-bromo-
m-tolunitrile (11.5 g, 58 mmol) and stirred at room tempera-
ture for 18 h. The mixture was poured into ice water and
extracted with EtOAc (2 × 100 mL). The organic layers were
combined, washed with brine, dried over MgSO4, filtered, and
concentrated. The residue was purified by flash chromatog-
raphy (500 g silica gel) using 1:1 EtOAc:hexane to afford 20.0
g of 3 (83% yield): 1H NMR (CDCl3) δ 0.08-0.11 (m, 2H), 0.42-
0.45 (m, 2H), 0.94 (m, 1H), 1.96-2.02 (m, 1H), 2.77-3.21 (m,
4H), 3.27 (d, J ) 14.3 Hz, 1H), 3.36 (d, J ) 6.2 Hz, 6H), 3.47-
3.65 (m, 7H), 3.68-3.78 (m, 4H), 3.88-3.95 (m, 1H), 3.98-
4.03 (m, 1H), 4.61 (d, J ) 14.3 Hz, 1H), 4.72 (d, J ) 7.0 Hz,
1H), 4.81 (d, J ) 6.9 Hz, 1H), 4.82 (d, J ) 6.9 Hz, 1H), 4.94 (d,
J ) 7.0 Hz, 1H), 6.92-6.94 (m, 2H), 7.16-7.52 (m, 12H).
[4R-(4r,5r,6â,7â)]-3-[[3-(Cyclopr opylm eth yl)h exah ydr o-
5,6-b i s [(2-m e t h o x y e t h o x y )m e t h o x y ]-2-o x o -4,7-b i s -
(ph en ylm eth yl)-1H-1,3-diazepin -1-yl]m eth yl]ben zoic Acid
(4). A solution of 3 (14.0 g, 20.9 mmol) in dry ethylene glycol
(200 mL) was treated with potassium hydroxide (14.0 g, 212
mmol) and heated at 140 °C for 18 h. The mixture was diluted
with water and neutralized with concentrated HCl and
extracted with EtOAc (2 × 100 mL). The organic extracts were
combined and washed with water, dried over MgSO4, filtered,
and concentrated to afford 14.0 g of 4 as a tan oil (97% yield):
1H NMR (CDCl3) δ 0.10-0.13 (m, 2H), 0.43-0.45 (m, 2H),
0.95-0.97 (m, 1H), 2.00-2.07 (m, 1H), 2.98-3.23 (m, 5H), 3.35
(s, 6H), 3.41-3.48 (m, 2H), 3.51-3.62 (m, 6H), 3.68-3.76 (m,
3H), 3.88-3.94 (m, 1H), 3.99-4.03 (m, 1H), 4.64 (d, J ) 7.0
Hz, 1H), 4.73 (d, J ) 6.6 Hz, 1H), 4.77 (d, J ) 13.9 Hz, 1H),
4.84 (d, J ) 6.9 Hz, 1H), 4.93 (d, J ) 7.0 Hz, 1H), 7.01-7.03
(m, 2H), 7.13-7.48 (m, 10H), 7.89 (s, 1H), 7.97 (d, J ) 7.7 Hz,
1H).
be pointed out that while the improved potency of SD146
and XV652 is maintained against resistant mutants and
strongly correlates to our design efforts, contributions
from biological components aside from intrinsic enzyme
binding cannot be ruled out.
Compounds with an improved resistance profile have
increased number of VDW contacts between the HIV
PR and the inhibitors. The number of VDW contacts
increases from 111 for Q8467 to 166 for XV638 and to
191 for SD146 (Table 3). Similarly, the buried surface
of the inhibitor upon complex formation increases from
810 Å2 for Q8467 to 941 Å2 for XV638 and to 993 Å2 for
SD146 (Table 3). In comparision to XV638 and SD146,
DMP323 and DMP450 exhibit a relatively poor resist-
ance profile which correlated with a decreased number
of H bonds and VDW interactions. The nonsymmetric
cyclic urea amides (e.g. Q8467) have a reduced number
of H bonds and VDW interactions as compared to the
symmetric cyclic urea amides. Indeed, the resistance
profile of symmetric cyclic urea amides is superior to
that of the nonsymmetric cyclic urea amides. SD146 is
very potent antiviral agent (IC90 ) 5.1 nM) and has an
exceptional resistance profile. Unfortunately, because
of its extreme insolubility in water and oils, to date no
formulation of SD146 could be developed for oral or
intravenous administration to animals.
Con clu sion s
These results demonstrate that the ability of an HIV
PR inhibitor to form the largest number of H bonds to
the backbone and the largest number of VDW interac-
tions in the active site correlates strongly with the
retention of high potency against drug resistant mutant
viruses. We believe that the remarkable resistance
profile of SD146 stems from its ability to occupy all six
enzyme subsites and form a large number of H bonds
and extensive VDW contacts. Structure-based ap-
proaches can be successfully applied to the design of
HIV-1 protease inhibitors to combat drug resistance.
The structure-based design strategies developed during
the course of this study may have broader implications
in therapeutic areas where the potency of the drug
decreases due to the mutation of the drug target.
Exp er im en ta l Section
Ch em ica l Meth od s. All procedures were carried out under
inert gas in oven-dried glassware unless otherwise indicated.
Proton NMR spectra were obtained on VXR or Unity 300 or
400 MHz instruments (Varian Instruments, Palo Alto) with
chemical shifts δ in ppm downfield from TMS as an internal
reference standard. Melting points were determined on a
Electrothermal digital melting point apparatus and are un-
corrected. Elemental analyses were performed by Quantita-
tive Technologies, Inc., Bound Brook, NJ 08805. Mass spectra
were measured with a HP 5988A mass spectrometer with
particle beam interface using NH3 for chemical ionization or
a Finnigan MAT 8230 mass spectrometer with NH3-DCI or
VG TRIO 2000 for ESI. High-resolution mass spectra were
measured on a VG 70-VSE instrument with NH3 chemical
ionization. Optical rotations were obtained on Perkin-Elmer
241 polarimeter at 25 °C. Solvents and reagents were obtained
from commercial vendors in the appropriate grade and used
without further purification unless otherwise indicated.
[4R-(4r,5r,6â,7â)]-1-(Cyclop r op ylm et h yl)h exa h yd r o-
5,6-bis[(2-m eth oxyeth oxy)m eth oxy]-4,7-bis(p h en ylm eth -
yl)-2H-1,3-diazepin -2-on e (2). A solution of [4R-(4R,5R,6â,7â)]-
h e xa h yd r o-5 ,6 -b is [(2 -m e t h oxye t h oxy)m e t h oxy]-4 ,7 -
bis(phenylmethyl)-2H-1,3-diazepin-2-one (1) (21.0 g, 42 mmol),
[4R-(4r,5r,6â,7â)]-3,3′-[[Tetr a h yd r o-5,6-bis[(2-m eth oxy-
eth oxy)m eth oxy]-2-oxo-4,7-bis(p h en ylm eth yl)-1H-1,3-d i-
azepin e-1,3(2H)-diyl]bis(m eth ylen e)]bis[ben zon itr ile] (5).
A solution of 1 (61.6 g, 120 mmol) in dry DMF (300 mL) was
cooled in a 0 °C ice bath. The mixture was treated with 60%
sodium hydride in mineral oil (29.4 g, 735 mmol) which was
added in portions followed by the dropwise addition of bromo-
m-tolunitrile (94.1 g, 480 mmol). The mixture was stirred at
room temperature for 18 h. The mixture was then cooled in a
0 °C ice bath and treated with saturated sodium bicarbonate
solution. The mixture was extracted with EtOAc (3 × 150 mL).
The organic layers were combined and washed with saturated
sodium bicarbonate solution, water, and brine. The organic
layer was dried over MgSO4, filtered, and concentrated and
the residue purified by flash chromatography (1 kg silica gel)
using 1:1 EtOAc:hexane to afford 78.1 g of 5 (88.8% yield): 1H
NMR (CDCl3) δ 2.70-2.78 (m, 2H), 2.99-3.03 (m, 2H), 3.26
(d, J ) 14.3 Hz, 2H), 3.35 (s, 6H), 3.48-3.81 (m, 12H), 4.61 (d,
J ) 14.3 Hz, 2H), 4.71 (d, J ) 6.6 Hz, 2H), 4.79 (d, J ) 6.9 Hz,
2H), 6.88-6.91 (m, 4H), 7.22-7.55 (m, 14H).

188 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 2
J adhav et al.
[4R-(4r,5r,6â,7â)]-3,3′-[[Tetr a h yd r o-5,6-bis[(2-m eth oxy-
eth oxy)m eth oxy]-2-oxo-4,7-bis(p h en ylm eth yl)-1H-1,3-d i-
azepin e-1,3(2H)-diyl]bis(m eth ylen e)]bis[ben zoic acid] (6).
A solution of 5 (70 g, 96 mmol) in dry ethylene glycol (600 mL)
was treated with potassium hydroxide (70 g, 1.06 mol) and
the mixture heated at 140 °C for 18 h. The mixture was
diluted with water, acidified with concentrated HCl, and
extracted with EtOAc (3 × 100 mL). The organic layers were
combined and washed with water. The organic layer was dried
over MgSO4, filtered, and concentrated to afford 70.31 g of 6
(95.0% yield): 1H NMR (CDCl3) δ 2.88-3.06 (m, 4H), 3.20 (d,
J ) 14.3 Hz, 2H), 3.33 (s, 6H), 3.36-3.70 (m, 12H), 4.61 (d, J
) 6.6 Hz, 2H), 4.69 (d, J ) 6.6 Hz, 2H), 4.81 (d, J ) 14.3 Hz,
2H), 7.01-7.03 (m, 4H), 7.19-7.30 (m, 6H), 7.38-7.51 (m, 4H),
7.93 (s, 2H), 7.99 (d, J ) 7.3 Hz, 2H).
[4R-(4r,5r,5â,7â)]-3-[[3-(Cyclopr opylm eth yl)h exah ydr o-
5,6-d ih yd r oxy-2-oxo-4,7-b is(p h e n ylm e t h yl)-1H -1,3-d i-
a zep in -1-yl]m eth yl]-N-2-th ia zolylben za m id e (Q8467). A
solution of 4 (1.4 g, 2.0 mmol) and 1-hydroxybenzotriazole (405
mg, 3 mmol) in dry DMF (10 mL) was cooled in a 0 °C ice
bath. The mixture was treated with 2-aminothiazole (300 mg,
3.0 mmol) followed by dicyclohexylcarbodiimide (825 mg, 4.0
tate formed immediately, and after the addition was complete,
the mixture was stirred at room temperature for 18 h. The
solvents and excess oxalyl chloride were pumped off. The
residue in dry pyridine (15 mL) was treated with 2-amino-
imidazole sulfate (1.704 g, 12.9 mmol) and the mixture stirred
at room temperature for 18 h. The mixture was diluted with
ethyl acetate and washed with water (2 × 50 mL). The residue
after removal of solvent was dissolved in dry methanol (10 mL)
and 4 M HCl in dioxane (10 mL) and stirred at room
temperature for 18 h. The mixture was diluted with ethyl
acetate and washed with saturated sodium bicarbonate. The
organic layer was dried over MgSO4, filtered, and concentrated
and the residue purified by flash chromatography (100 g of
silica gel) using 1:1 EtOAc:hexane followed by 10:1:10 EtOAc:
EtOH:hexane. The solvent was switched to 1% MeOH:CHCl3
followed by 2% and finally 5% to afford 711 mg of white solid
compound XV651 (28.5% yield). mp 156.7 °C; [R]D +91.41° (c
1
0.256, DMSO); H NMR (CDCl3) δ 0.01-0.03 (m, 2H), 0.33-
0.37 (m, 2H), 0.87 (m, 1H); 1.89-1.96 (m, 1H), 2.86-2.94 (m,
1H), 3.04-3.13 (m, 5H), 3.50-3.71 (m, 3H), 3.85-3.88 (m, 1H),
4.04-4.08 (m, 1H), 4.71 (d, J ) 14.2 Hz, 1H), 6.61 (s, 2H),
6.91 (d, J ) 6.6 Hz, 2H), 7.10-7.32 (m, 13H), 7.75 (s, 1H); 13
C
mmol) using dry DMF (5 mL) to aid in the addition.
A
NMR (CDCl3) δ 3.45, 3.91, 10.43, 32.36, 32.61, 55.64, 56.95,
65.84, 66.24, 70.95, 71.05, 117.52, 126.12, 126.35, 126.73,
128.28, 128.47, 128.77, 129.22, 129.38, 133.23, 133.35, 139.13,
139.46, 139.82, 142.43, 161.74, 167.20. HRMS calcd for
C34H38N5O4 [M + H]+ 580.2923, found: 580.2912. Anal.
(C34H37N5O4‚0.6H2O) C, H, N.
precipitate forms within 15 min after the addition, and the
mixture was stirred at room temperature for 18 h. The
mixture was diluted with ethyl acetate and filtered. The
mother liquor was washed with water (75 mL) followed by
dilute sodium bicarbonate (2 × 50 mL) followed by water (50
mL). The organic layer was separated, dried over MgSO4,
filtered, and concentrated. The residue was treated with 2.0
M HCl in 1:1 MeOH:dioxane (10 mL) and the mixture stirred
at room temperature for 18 h. The mixture was neutralized
by pouring into saturated sodium bicarbonate (evolution!) and
extracting with ethyl acetate (2 × 75 mL). The organic layer
was separated, dried over MgSO4, filtered, and concentrated.
The residue was purified by flash chromatography (75 g silica
gel) using 1% MeOH:CHCl3 followed by 2% and 3% to afford
432 mg of Q8467 as a white solid (36.2% yield); mp 135.9 °C;
[4R-(4r,5r,6â,7â)]-3-[[Hexa h yd r o-5,6-d ih yd r oxy-2-oxo-
3,4,7-tr is(p h en ylm eth yl)-1H-1,3-d ia zep in -1-yl]m eth yl]-N-
1H-im id a zol-2-ylben za m id e (SD143). Compound SD143
was prepared in a manner analagous to that described in the
synthesis for XV651: 246 mg as a white solid (23.6% yield);
mp 155.0 °C; 1H NMR (DMSO) δ 2.81-2.89 (m, 3H), 2.98-
3.63 (m, 3H), 3.43-3.55 (m, 4H), 4.67 (d, J ) 13.9 Hz, 1H),
4.71 (d, J ) 14.2 Hz, 1H), 5.11-5.14 (m, 2H), 6.80 (s, 2H),
6.96-7.47 (m, 17H), 7.93 (s, 1H), 7.98 (d, J ) 7.7 Hz, 1H),
11.8 (s, 2H); 13C NMR (DMSO) δ 32.48, 55.65, 55.82, 65.75,
66.42, 70.71, 126.50, 127.14, 127.68, 128.65, 128.83, 129.23,
129.35, 129.62, 132.66, 134.67, 138.70, 139.05, 140.49, 140.55,
143.18, 161.63, 166.61; HRMS calcd for C37H38N5O4 [M + H]+
616.2923, found 616.2918. Anal. (C37H37N5O4) C, H, N.
1
[R]D +87.61° (c 0.226, DMSO); H NMR (CDCl3) δ 0.01-0.03
(m, 2H), 0.31-0.43 (m, 2H), 0.89 (m, 1H), 2.77-2.85 (m, 1H),
3.03-3.19 (m, 3H), 3.34 (d, J ) 14.3 Hz, 1H), 3.50-3.57 (m,
1H), 3.7 (m, 2H), 3.88 (bs, 1H), 4.00 (bs, 1H), 4.10 (bs, 1H),
4.7 (d, J ) 14.3 Hz, 1H), 6.82 (m, 2H), 6.92 (d, J ) 3.6 Hz,
1H), 7.10-7.25 (m, 9H), 7.33-7.39 (m, 1H), 7.46 (d, J ) 7.7
Hz, 1H), 7.75 (s, 1H), 7.84 (d, J ) 7.7 Hz, 1H), 11.7 (s, 1H);
13C NMR (DMSO) δ 4.00, 4.13, 10.96, 32.53, 55.83, 56.78,
66.96, 70.81, 71.03, 114.22, 126.27, 126.55, 127.32, 128.49,
128.74, 129.01, 129.50, 129.59, 129.75, 132.61, 133.55, 137.96,
139.60, 140.63, 140.89, 159.17, 161.41, 165.29; HRMS calcd
for C34H37N4O4S [M + H]+ 597.2535, found 597.2523. Anal.
(C34H36N4O4S): C, H, N, S.
[4R-(4r,5r,6â,7â)]-3-[[3-Bu tylh exa h yd r o-5,6-d ih yd r oxy-
2-oxo-4,7-bis(ph en ylm eth yl)-1H-1,3-diazepin -1-yl]m eth yl]-
N-1H-im id a zol-2-ylben za m id e (SD152). Compound SD152
was prepared in a manner analagous to that described in the
synthesis for XV651: 254 mg as a white solid (22.6% yield);
mp 150.0 °C; [R]D +90.65° (c 0.246, DMSO); 1H NMR (DMSO)
δ 0.75 (t, J ) 7.3 Hz, 3H), 1.11-1.28 (m, 4H), 1.93-1.97 (m,
1H), 2.78-3.08 (m, 5H), 3.40-3.50 (m, 4H), 3.65-3.69 (m, 1H),
4.63 (d, J ) 14.3 Hz, 1H), 5.12-5.23 (m, 2H), 6.78 (s, 2H),
6.89 (d, J ) 6.6 Hz, 2H), 7.10-7.43 (m, 10H), 7.87 (s, 1H),
7.94 (d, J ) 7.7 Hz, 1H), 11.8 (s, 2H); 13C NMR (CDCl3) δ 13.76,
20.27, 31.19, 32.66, 32.78, 49.21, 49.49, 49.78, 52.59, 55.79,
66.18, 66.44, 71.04, 71.26, 117.72, 126.24, 126.47, 126.87,
128.39, 128.48, 128.57, 128.81, 129.34, 129.46, 133.38, 133.43,
139.24, 139.52, 139.77, 142.69, 161.71, 167.34. HRMS calcd
for C34H40N5O4 [M + H]+ 582.3080, found 582.3078. Anal.
(C34H39N5O4) C, H, N.
[4R-(4r,5r,6â,7â)]-3-[[Hexa h yd r o-5,6-d ih yd r oxy-2-oxo-
3,4,7-tr is(p h en ylm eth yl)-1H-1,3-d ia zep in -1-yl]m eth yl]-N-
2-th ia zolylben za m id e (XV655). Compound XV655 was
prepared in an analagous manner as described in the synthesis
for Q8467: 411 mg as a white solid (65.0% yield); mp 246.6
1
°C; [R]D +93.93° (c 0.214, DMSO); H NMR (DMSO) δ 2.70-
2.76 (m, 1H), 2.82-2.90 (m, 2H), 2.95-3.02 (m, 2H), 3.09 (d,
J ) 13.9 Hz, 1H), 3.42-3.59 (m, 4H), 4.67 (d, J ) 13.9 Hz,
1H), 4.70 (d, J ) 13.9 Hz, 1H), 5.14 (d, J ) 12.1 Hz, 1H), 5.15
(d, J ) 12.1 Hz, 1H), 6.92-6.95 (m, 2H), 7.01 (d, J ) 6.6 Hz,
2H), 7.14 (d, J ) 6.6 Hz, 2H), 7.23-7.37 (m, 10H), 7.42-7.53
(m, 2H), 7.57 (d, J ) 3.6 Hz, 1H), 7.96 (s, 1H), 8.02 (d, J ) 7.6
Hz, 1H), 12.7 (s, 1H); 13C NMR (DMSO) δ 32.50, 55.64, 55.98,
65.76, 66.89, 70.66, 114.19, 126.43, 126.48, 127.32, 127.68,
128.55, 128.64, 128.82, 128.98, 129.34, 129.47, 129.58, 129.62,
132.53, 133.53, 138.05, 138.68, 139.42, 140.45, 140.53, 159.20,
161.55, 165.22. HRMS calcd for C37H37N4O4S [M + H]+
633.2535, found 633.2517. Anal. (C37H36N4O4S) C, H, N, S.
[4R-(4r,5r,6â,7â)]-3-[[3-(Cyclopr opylm eth yl)h exah ydr o-
5,6-d ih yd r oxy-2-oxo-4,7-b is(p h e n ylm e t h yl)-1H -1,3-d i-
a ze p in -1-y l]m e t h y l]-N -1H -im id a zo l-2-y lb e n za m id e
(XV651). A solution of 4 (3.0 g, 4.3 mmol) and pyridine (1.29
g, 16.2 mmol) in dry benzene (30 mL) was cooled in a 0 °C ice
bath. The mixture was treated dropwise with 2.0 M oxalyl
chloride in dichloromethane (16.2 mL, 32.4 mmol). A precipi-
[4R-(4r,5r,6â,7â)]-N-1H-Ben zim id a zol-2-yl-3-[[3-(cyclo-
p r op ylm et h yl)h exa h yd r o-5,6-d ih yd r oxy-2-oxo-4,7-b is-
(p h e n y lm e t h y l)-1H -1,3-d ia ze p in -1-y l]m e t h y l]b e n z-
a m id e (SD145). Compound SD145 was prepared in a manner
analagous to that described in the synthesis for 9: 617 mg as
a white solid (67.6% yield); mp 170.4 °C; [R]D +69.16° (c 0.214,
DMSO); 1H NMR (DMSO) δ 0.00 (m, 2H), 0.32-0.34 (m, 2H),
0.84-0.89 (m, 1H), 1.76-1.83 (m, 1H), 2.80-3.07 (m, 6H),
3.45-3.52 (m, 3H), 3.73-3.76 (m, 1H), 4.64 (d, J ) 14.3 Hz,
1H), 5.11-5.21 (m, 2H), 6.90 (d, J ) 6.6 Hz, 2H), 7.05-7.43
(m, 14H), 7.93 (s, 1H), 8.02 (d, J ) 7.7 Hz, 1H), 12.2 (s, 2H);
13C NMR (DMSO) δ 3.91, 4.02, 10.89, 32.40, 55.54, 56.70,
66.25, 66.85, 70.75, 70.98, 121.91, 126.24, 126.51, 127.51,
128.48, 128.70, 129.41, 129.59, 129.71, 132.89, 135.07, 139.19,
140.63, 140.90, 161.47. HRMS calcd for C38H40N5O4 [M + H]+
630.3080, found 630.3060. Anal. (C38H39N5O4) C, H, N.

Cyclic Urea Amides
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 2 189
[4R-(4r,5r,6â,7â)]-3,3′-[[Tet r a h yd r o-5,6-d ih yd r oxy-2-
oxo-4,7-bis(ph en ylm eth yl)-1H-1,3-diazepin e-1,3(2H)-diyl]-
bis(m eth ylen e)]bis[N-2-th ia zolylben za m id e] (XV638). A
solution of 6 (1.08 g, 1.4 mmol) and 1-hydroxybenzotriazole
(570 mg, 4.2 mmol) in dry DMF (10 mL) was cooled in a 0 °C
ice bath and treated with 2-aminothiazole (421 mg, 4.2 mmol).
To this was added a solution of 1,3-dicyclohexylcarbodiimide
(1.15 g, 5.6 mmol) in dry DMF (6 mL). The reaction mixture
was stirred at room temperature for 16 h, diluted with ethyl
acetate, and filtered over Celite. The solution was washed
with water (2 × 50 mL), saturated sodium bicarbonate, and
water. The solution was the dried over MgSO4, filtered, and
concentrated in vacuo. The residue was purified by flash
chromatography (75 g silica gel) using 2:3 EtOAc:hexane,
followed by 1:8 MeOH:CHCl3 to elute 1.2 g of a white solid.
The solid 1.2 g in dry methanol (6 mL) was treated with 4 N
HCl in dioxane (6 mL). The mixture was stirred at room
temperature for 18 h and then poured into chloroform (75 mL).
The organic layer was washed with saturated sodium bicar-
bonate. The aqueous layer was extracted with chloroform (2
× 50 mL). The combined organic layers were dried over
MgSO4, filtered, and concentrated in vacuo. The residue was
purified by flash chromatography (75 g silica gel) using
chloroform, followed by 1% MeOH:CHCl3, 2%, and finally 4%
to elute 880 mg of XV638 as a white solid (82.0% yield); mp
ously described.12 Cloned 5′ half HIV plasmids were linearized
with NcoI, ligated with a complementary NcoI linearized 3′
half HIV plasmid, and used to transfect MT4 cells by lipo-
fection. The culture was frozen in aliquots 7-10 days after
lipofection, when virally induced cytopathic effect had spread
throughout the culture. Virus stocks were not passaged, in
order to reduce the opportunity for accumulation of additional
sequence changes beyond those encoded in the transfected
plasmid.
The drug susceptibility of wild type and protease variant
HIV’s was tested by measuring the effects of each compound
on accumulation of viral p24 antigen following infection of MT4
cells. MT4 cells (2.5 × 105 cells/mL) were infected with
appropriate dilutions of each virus stock and cultured for 24
h at 37 °C and 5% CO2 in RPMI 1640 with 10% fetal calf
serum, 2 mM glutamine, and 50 µg/mL gentamycin (Gibco/
BRL) without protease inhibitors. Infected cultures were then
washed three times in medium and plated into microtiter plate
wells with various concentrations of each inhibitor at a final
cell number of 2.5 × 105 infected cells/mL. Following addition
of drug, infected cell cultures were incubated for an additional
72 h. Appropriate dilutions of each virus stock were deter-
mined in preliminary experiments and were selected to result
in the accumulation of between 1000 and 4000 ng/mL of p24
antigen at the end of the 4 day culture period. P24 antigen
accumulation was quantitated in all cultures using the DuPont
p24 antigen Elisa kit. The potency of each compound is
reported as an IC90, that concentration of compound that
inhibited the accumulation of viral p24 antigen by 90% from
the level accumulated in an infected, untreated culture.
Resistance of mutant viruses was calculated as (IC90 for a
mutant virus/IC90 of wt HXB2 virus). Since differences in viral
growth rates or initial viral innoculum could affect the IC90
values measured, susceptibility to ddC(Sigma), a reverse
transcriptase inhibitor, was monitored for all viruses in each
experiment. IC90 values for mutant viruses were normalized
by the variation in the IC90 value for the ddC internal control.
This normalization was usually less than 2 and always less
than 2.5-fold. Despite this normalization, resistance values
less than 3-fold are not considered significant.
1
179.8 °C; [R]D +76.76° (c 0.284, DMSO); H NMR (DMSO) δ
2.67-2.75 (m, 2H), 2.93-2.97 (m, 2H), 3.05 (d, J ) 13.9 Hz,
2H), 3.54-3.57 (m, 4H), 4.63 (d, J ) 13.9 Hz, 2H), 5.16 (s,
2H), 6.88-6.91 (m, 4H), 7.16-7.23 (m, 6H), 7.24 (d, J ) 3.7
Hz, 2H), 7.39-7.50 (m, 4H), 7.53 (d, J ) 3.7 Hz, 2H), 7.93 (s,
2H), 7.98 (d, J ) 7.3 Hz, 2H), 12.4 (s, 2H); 13C NMR (DMSO)
δ 32.44, 55.89, 66.81, 70.54, 114.03, 126.28, 127.19, 128.42,
128.84, 129.34, 129.41, 132.43, 133.38, 137.76, 139.25, 159.0,
161.23, 165.11; HRMS calcd for C41H39N6O5S2 [M + H]+
759.2423, found 759.2406. Anal. (C41H38N6O5S2) C, H, N, S.
[4R-(4r,5r,6â,7â)]-3,3′-[[Tet r a h yd r o-5,6-d ih yd r oxy-2-
oxo-4,7-bis(ph en ylm eth yl)-1H-1,3-diazepin e-1,3(2H)-diyl]-
b is (m e t h y le n e )]b is [N -1H -im id a zo l-2-y lb e n za m id e ]
(XV652). Compound XV652 was prepared in
a manner
analagous to that described in the synthesis for XV651: 75 mg
as a white solid (6.1% yield); mp 136.0 °C; [R]D +63.55° (c
1
X-r a y Meth od s. A modified form of HIV-1 PR, Cys to Ala
at position 95, was expressed in E. coli and purified from
inclusion bodies using a solid phase refolding procedure.20
Frozen aliquots of active protein were thawed in the presence
of inhibitor. Crystals of the protease‚inhibitor complexes were
grown at 18 °C in hanging drops by vapor diffusion as
reported.28 Hexagonal rods (0.07 mm × 0.07 mm × 1.4 mm)
appeared within 1 week. Diffraction data were collected at
room temperature on a RAXIS IIc imaging plate mounted on
a RU-H2R Rigaku rotating anode (Cu KR) generator using a
0.3 mm cathode and operating at 50 kV and 100 mA. The
unit cell parameters were determined from three still frames
taken at 15° intervals using the RAXIS processing software.
All protease‚inhibitor complexes crystallized in the space group
P61 with a dimer in the asymmetric unit and cell dimensions
a ) b ) 62.8 Å and c ) 83.5 Å, with a dimer in the asymmetric
unit. Full data sets were obtained by collecting between 25
and 30 2° oscillation images with an exposure time of 60 min
at a detector distance of 70 mm. A total of 58 704, 39 076,
and 53 827 raw reflections were measured and reduced to
16 523, 16 197, and 15 429 unique reflections (all the data)
for Q8467, XV638, and SD146, respectively. The merging R
factors for symmetry-related reflection were 11%, on intensi-
ties, for the three complexes. The atomic coordinates of HIV
PR complexed with DMP323 complex were used as the starting
model in refinement. The complexes with Q8467, XV638, and
SD146 were refined to 2.1, 1.8, and 1.8 Å, respectively, by
performing several cycles of simulated annealing, positional,
and restrained B-refinement.29 The conformations of the
inhibitors were determined by calculating |Fo - Fc| and |2Fo
- Fc| maps, and the protein structure was adjusted as
necessary using SA-omit maps. The starting R values were
approximately 44% and were reduced to 19% with an overall
deviations from ideal geometry of 3.07, 2.92, and 2.95 Å for
bond distances, and 0.012, 0.012, and 0.012 Å, respectively.
0.214, DMSO); H NMR (DMSO) δ 2.83-2.92 (m, 2H), 3.08-
3.17 (m, 4H), 3.61-3.71 (m, 4H), 4.77 (d, J ) 14.3 Hz, 2H),
6.76 (s, 4H), 6.99 (d, J ) 7.7 Hz, 4H), 7.20-7.28 (m, 6H), 7.41-
7.43 (m, 4H), 7.77 (s, 2H), 7.83-7.86 (m, 2H); 13C NMR (CDCl3)
δ 34.37, 57.91, 68.22, 72.46, 119.45, 128.22, 128.69, 130.31,
130.63, 131.12, 134.94, 135.70, 140.56, 141.35, 144.44, 163.95,
169.31; HRMS calcd for C41H41N8O5 [M + H]+ 725.3199, found
725.3200. Anal. (C41H40N8O5‚0.5H2O) C, H, N.
[4R-(4q ,5r,6â,7â)]-3,3′-[[Tet r a h yd r o-5,6-d ih yd r oxy-2-
oxo-4,7-bis(ph en ylm eth yl)-1H-1,3-diazepin e-1,3(2H)-diyl]-
bis(m et h ylen e)]b is[N-1H-b en zim id a zol-2-ylb en za m id e]
(SD146). Compound SD146 was prepared in
a manner
analagous to that described in the synthesis for XV638 and
isolated as a bishydrochloride: 200 mg as a white solid (15.3%
yield); mp 266.4 °C; 1H NMR (DMSO) δ 2.82-2.91 (m,2H), 3.01
(d, J ) 14.0 Hz, 4H), 3.51-3.57 (m, 4H), 4.71 (d, J ) 14.2 Hz,
2H), 5.16 (s, 2H), 6.99 (d, J ) 7.0 Hz, 4H), 7.12-7.16 (m, 4H),
7.20-7.51 (m, 14H), 8.02 (s, 2H), 8.08 (d, J ) 7.7 Hz, 2H),
12.3 (s, 4H); 13C NMR (DMSO) δ 32.51, 55.81, 66.31, 70.71,
121.94, 126.49, 127.58, 128.63, 128.76, 129.48, 129.64, 132.93,
139.03, 140.48, 161.59; HRMS calcd for C49H45N8O5 [M + H]+
825.3512, found 825.3520. Anal. (C49H46Cl2N8O5) C, H, Cl,
N.
Biologica l Meth od s. Inhibition of HIV protease was
measured by assaying the cleavage of a fluorescent peptide
substrate using HPLC.25 The lower limit of the detection for
the reported25 assay has been improved by lowering the
enzyme concentration from 62 to 50 pM. The antiviral potency
of compounds was assessed by measuring their effect on the
accumulation of viral RNA transcripts three days after infec-
tion of MT-2 cells with HIV-1 RF.26 Oral bioavailability was
measured as previously described.27 In order to introduce
defined sequence alterations into the protease gene, recombi-
nant DNA plasmids containing the 5′ half of the HIV HXB2
genome were subjected to site directed mutagenesis as previ-

190 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 2
J adhav et al.
M. E.; J ohnson, V. A.; Emini, E. A.; Deutsch, P.; Lifson, J . D.;
Bonhoeffer, S.; Nowak, M. A.; Hahn, B. H.; Saag, M. S.; Shaw,
G. M. Viral dynamics in human immunodeficiency virus type
infection. Nature 1995, 373, 117-122.
Ack n ow led gm en t. We thank Edward R. Holler J r.
for his technical assistance during the synthesis of cyclic
urea amides, Ronald M. Klabe for Ki measurements,
Dean Winslow and Renay Buckery for participation in
the early phases of recombinant virus construction, and
Karen B. Gallagher and Renay Buckery for viral resist-
ance data. We also thank Paul S. Anderson, David A.
J ackson, and Charles A. Kettner for helpful comments
on the manuscript. We are indebted to the HIV Work-
ing Group of Dupont Merck.
(5) Coffin, J . M. HIV Population Dynamics in Vivo: Implications
for Genetic Variation, Pathogenesis, and Therapy. Science 1995,
267, 483-489. (b) Kozal, M. J .; Shah, N.; Yang, R.; Shen, N.;
Fucini, R.; Merigan, T. C.; Richman, D. D.; Chee, M. S.; Gingeras,
T. R. Natural polymorphism of HIV-1 clade B protease gene and
the implications for therapy. Fourth International Workshop on
Drug Resistance, Sardinia, Italy, 1995, Abstract No. 76.
(6) (a) Baldwin, E. T.; Bhat, T. N.; Liu, B.; Pattabiraman, N.;
Erickson, J . W. Structural basis of drug resistance for the V82A
mutant of HIV-1 proteainase. Struct. Biol. 1995, 2, 244-249.
(b) Erickson, J . W. The not-so-great escape. Struct. Biol. 1995,
2, 523-529.
Refer en ces
(7) (a) Markowitz, M. M.; Mo, H.; Kempf, D. J .; Norbeck, D. W.;
Bhat, T. N.; Erickson, J . W.; Ho, D. D. Selection and analysis of
human immunodeficiency virus type 1 variants with increase
resistance to ABT-538, a novel protease inhibitor. J . Virol. 1995,
69, 701-706. (b) Anton, E. D.; Bacheler, L.; Garber, S.; Reid,
C.; Buckery, R.; Scarnati, H.; Korant, B.; Winslow, D. L.
Replicative ability of HIV variants with substitutions in the
active site of protease is dependent on compensatory mutations
outside the active site. Fourth International Workshop on Drug
Resistance, Sardinia, Italy, 1995, Abstract No. 60. (c) Rose, R.
E.; Gong, Y., Greytok, J . A.; Bechtold, C. M.; Terry, B. J .;
Robinson, B. S.; Alam, M.; Colono, R. J .; Lin, P. Human
immunodeficiency virus type 1 viral background plays a major
role in development of resistance to protease inhibitors. Proc.
Natl. Acad. Sci. U.S.A. 1996, 93, 1648-1653.
(8) Lam, P. Y. S.; J adhav, P. K.; Eyermann, C. J .; Hodge, C. N.;
Ru, Y.; Bacheler, L. T.; Meek, J . L.; Otto, M. J .; Rayner, M. M.;
Wong, Y. N.; Chang, C.-H.; Weber, P. C.; J ackson, D. A.; Sharpe,
T. R.; Erickson-Viitanen, S. Rational Design of Potent, Bioavail-
able, Nonpeptide Cyclic Ureas as HIV Protease Inhibitors.
Science 1994, 263, 380-384.
(9) Lam, P. Y. S.; Ru, Y.; J adhav, P. K.; Aldrich, P. E.; DeLucca, G.
V.; Eyermann, C. J .; Chang, C.-H.; Emmett, G.; Holler, E. R.;
Daneker, W. F.; Li, L.; Confalone, P. N.; McHugh, R. J ., Han,
Q.; Markwalder, J . A.; Seitz, S. P.; Bacheler, L. T.; Rayner, M.
M.; Klabe, R. M.; Shum, L.; Winslow, D. L.; Kornhauser, D. M.;
J ackson, D. A.; Erickson-Viitanen, S.; Sharpe, T. R.; Hodge, C.
N. Cyclic HIV Protease Inhibitors: Synthesis, Conformational
Analysis, P2/P2′ Structure/Activity Relationship and Molecular
Recognition of Cyclic Ureas. J . Med. Chem. 1996, 39, 3514-3525.
(10) Nugiel, D. A.; J acobs, K.; Worley, T.; Patel, M.; Kaltenbach, R.
F., III; Meyer, D. T.; J adhav, P. K.; DeLucca, G. V.; Smyser, T.
E.; Klabe, R. M.; Bacheler, L. T.; Rayner, M. M.; Seitz, S. P.
Preparation and Strcuture-Activity Relationship of Novel P1/
P1′-Substituted Cyclic Urea-Based Human Immunodeficiency
Virus Type-1 Protease Inhibitors. J . Med. Chem. 1996, 39, 2156-
2169.
(1) For recent reviews, see: (a) Debouck, C. The HIV-1 Protease as
a Therapeutic Target for AIDS. AIDS Res. Hum. Retroviruses
1992, 8, 153-164. (b) Katz, R. A. and Skalka, A. M. Annu. Rev.
Biochem. 1994, 63, 133-173.
(2) (a) Ho, D. D.; Toyoshima, T.; Mo, H.; Kempf, D. J .; Norbeck, D.
W.; Chen, C. M; Wideburg, N. E.; Burt, S. K.; Erickson, J . W.;
Singh, M. K. Characterization of Human Immunodeficiency
Virus Type 1 Variants with Increased Resistance to a C2-
Symmetric Protease Inhibitor. J . Virol. 1994, 68, 2016-2020.
(b) J acobsen, H.; Yasargil, K.; Winslow, D. L.; Craig, J . C.; Krohn,
A.; Duncan, I. B.; Mous, J . Characterization of Human Immuno-
deficiency Virus Type 1 Mutants with Decreased Sensitivity to
Proteinase Inhibitor Ro 31-8959. Virology 1995, 206,527-534.
(c) Kaplan, A. H.; Michael, S. F.; Wehbie, R. S.; Knigge, M. F.;
Paul, D. A.; Everitt, L.; Kempf, D. J .; Norbeck, D. W.; Erickson,
J . W.; Swanstrom, R. Selection of multiple human immuno-
deficiency virus type 1 variants that encode viral proteases with
decreased sensitivity to an inhibitor of viral protease. Proc. Natl.
Acad. Sci. U.S.A. 1994, 91, 5597-5601. (d) King, R. W.; Garber,
S.; Winslow, D. L.; Reid, C. D.; Bacheler, L. T.; Anton, E.; Otto,
M. J . Multiple mutations in the human immunodeficiency virus
protease gene are responsible for decreased susceptibility to
protease inhibitors. Antiviral Chem. Chemother. 1995, 6, 80-
88. (e) Otto, M. J .; Garber, S.; Winslow, D. L.; Reid, C. D.;
Aldrich, P.; J adhav, P. K.; Patterson, C. E.; Hodge, C. N.; Cheng,
Y.-SE. In vitro isolation and identification of human immuno-
deficiency virus (HIV) variants with reduced sensitivity to C-2
symmetrical inhibitors of HIV type 1 protease. Proc. Natl. Acad.
Sci. U.S.A. 1993, 90, 7543-7547. (f) Tisdale, M.; Myers, R. E.;
Blair, E.; Kohli, A.; Kemp, S. D.; Larder, B. A. In Vitro Studies
Support Combination Therapy with HIV Protease Inhibitors.
Fourth International Workshop on Drug Resistance, Sardinia
Italy, 1995, Abstract No. 61. (g) El-Farrash, M. A.; Kuroda, M.
J .; Kitazaki, T.; Masuda, T.; Kato, K.; Hatanaga, M.; Harada,
S. Generation and characterization of human immunodeficiency
virus type-1 (HIV-1) mutant resistant to HIV-1 protease inhibi-
tor. J . Virol. 1994, 68, 233-239. (h) Partaledis, J . A.; Yamaguchi,
K.; Tisdale, M.; Blair, E.; Falcione, C.; Maschera, B.; Myers, R.
E.; Pazhanisamy, S.; Futer, O.; Cullinan, A. B.; Stuver, C. M.;
Byrn, R. A.; Livingston, D. J . In vitro selection and characteriza-
tion of human immunodeficiency virus type-1 (HIV-1) isolates
with reduced sensitivity to hydroxyethylamino sulfonamide
inhibitors of HIV-1 aspartyl protease. J . Virol. 1995, 69, 5228-
5235. (i) Condra, J . H.; Schleif, W. A.; Blahy, O. M.; Gabryelski,
L. J .; Graham, D. J .; Quintero, J . C.; Rhodes, A.; Robbins, H.
L.; Roth, E.; Shivapraksh, M.; Titus, D.; Yang, T.; Toppler, H.;
Squires, K. E.; Deutsch, P. J .; Emini, E. A. In vivo emergence of
HIV-1 variants resistant to multiple protease inhibitors Nature
1995, 374, 569-571. (j) Norbeck, D.; Hsu, A.; Granneman, R.;
Denissen, J .; Kempf, D.; Molla, A.; Leonard, J . Virologic and
immunologic response to ritonavir (ABT-538), an inhibitor of
HIV protease. Fourth International Workshop on Drug Resist-
ance, Sardinia Italy, 1995, Abstract No. 70.
(3) (a) Mansky, L. M.; Temin, H. M. Lower in vivo mutation rate of
human immunodeficiency virus type 1 than that predicted from
the fidelity of purified reverse transcriptase. J . Virol. 1995, 69,
5087-94. (b) Roberts, J . D.; Bebenek, K.; Kunkel, T. A. The
accuracy of reverse transcriptase from HIV-1. Science 1988, 242,
1171-3. (c) Hubner, A.; Kruhoffer, M.; Grosse, F.; Krauss, G.
Fidelity of human immunodeficiency virus type I reverse tran-
scriptase in copying RNA. J . Mol. Biol. 1992, 223, 595-600. (d)
J i, J .; Loeb, L. A. Fidelity of HIV-1 reverse transcriptase copying
RNA in vitro. Biochemistry 1992, 31, 954-8. (e) Boyer, J . C.;
Bebenek, K.; Kunkel, T. A. Unequal human immunodeficiency
virus type I reverse transcriptase error rate with RNA and DNA
templates. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 6919-6923.
(4) (a) Perelson, A. S.; Neumann, A. U.; Markowitz, M.; Leonard,
J . M.; Ho, D. D HIV-1 Dynamics in Vivo: Virion Clearance Rate,
Infected Cell Life-Span, and Viral Generation Time. Science
1996, 271, 1582-1586. (b) Ho, D. D.; Neumann, A. U.; Perelson,
A. S.; Chen, W.; Leonard, J . M.; Markowitz, M. Rapid turnover
of plasma virions and CD4 lymphocytes in HIV-1 infection.
Nature 1995, 373, 123-126. (c) Wei, X.; Ghosh, S. K.; Taylor,
(11) Bodanszky, M. Peptide Chemistry:
A Practical Textbook;
Springer-Verlag: New York, 1986; pp 63-68.
(12) Winslow, D.; Horlick R. A.; Anton, E. A.; Tritch, R. J .; Zagursky,
R. J .; Bacheler, L. T. A novel method for construction of
infectious molecular clones of HIV-1 containing defined muta-
tions in the HIV protease gene. Biochem. Biophys. Res. Commun.
1994, 205, 1651-1657.
(13) Hodge, C. N.; Aldrich, P. E.; Bacheler, L. T.; Chang, C-H.;
Eyermann, C. J .; Garber, S.; Grubb, M.; J ackson, D. A.; J adhav,
P. K.; Korant, B.; Lam, P. Y. S.; Maurin, M. B.; Meek, J . L.;
Otto, M. J .; Rayner, M. M.; Sharpe, T. R.; Shum, L.; Winslow,
D. L.; Erickson-Viitanen, S. DMP450: Improved cyclic urea
inhibitors of the HIV-1 protease: synthesis, potency, resistance
profile, human pharmacokinetics and x-ray crystal structure of
DMP 450. Chem. Biol. 1996, 3, 301-314.
(14) (a) J adhav, P. K.; Woerner, F. J . Synthesis of C2-Symmetric
HIV-1 Protease Inhibitors from D-Mannitol. Bioorg. Med. Chem.
Lett. 1992, 2, 353-356. Q8024 is the same as compound 3
reported in this reference. (b) J adhav, P. K.; McGee, L. R.;
Shenvi, A.; Hodge, C. N. U.S. Patent 5,294,720, issued March
15, 1994.
(15) Kempf, D. J .; Marsh, K. C.; Denissen, J . F.; McDonald, E.;
Vasavanonda, S.; Flentge, C. A.; Green, B. E.; Fino, L.; Park, C.
H.; et al. ABT-538 is a potent inhibitor of human immunodefi-
ciency virus protease and has high oral bioavailability in
humans. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 2484-2488.
(16) Dorsey, Bruce D.; Levin, R. B.; McDaniel, S. L.; Vacca, J . P.;
Guare, J . P.; Darke, P. L.; Zugay, J . A.; Emini, E. A.; Schleif,
W. A.; et al. L-735,524: The Design of a Potent and Orally
Bioavailable HIV Protease Inhibitor. J . Med. Chem. 1994, 37,
3443-51
(17) Kim, E. E.; Baker, C. T.; Dwyer, M. D.; Murcko, M. A.; Rao, B.
G.; Tung, R. D.; Navia, M. A. Crystal structure of HIV-1 protease
in complex with VX-478,
a potent and orally bioavailable
inhibitor of the enzyme. J . Am. Chem. Soc. 1995, 117, 1181-
1182.

Cyclic Urea Amides
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 2 191
(18) Roberts, N. A.; Martin, J . A.; Kinchington, D.; Broadhurst, A.
V.; Craig, J . C.; Duncan, I. B.; Galpin,S. A.; Handa, B. K.; Kay,
J .; et al. Rational design of peptide-based HIV proteinase
inhibitors. Science 1990, 248, 358-361.
with the protease inhibitor, nelfinavir. Fifth International
Workshop on Drug Resistance, Whistler, Canada, 1996, Abstract
No. 29.
(25) Erickson-Viitanen, S.; Klabe, R. M.; Cawood, P. G.; O’Neal, P.
L.; Meek, J . L. Potency and Selectivity of Inhibition of Human
Immunodeficiency Virus Protease by a Small Nonpeptide Cyclic
Urea, DMP323. Antimicrob. Agents Chemother. 1994, 38, 1628-
1634.
(26) Bacheler, L. T.; Paul, M.; J adhav, P. K.; Otto, M.; Stone, B.;
Miller, J . An Assay for HIV RNA in Infected Cell Lysates, and
Its Use for the Rapid Evalution of Antiviral Efficacy. Antiviral
Chem. Chemother. 1994, 5, 111-121.
(27) Wong, N. Y.; Burcham, D. L.; Saxton, P. L., Erickson-Viitanen,
S.; Grubb, M.F.;Quon, C. Y.; and Huang, S.-M. A Pharmaco-
kinetic Evalution of HIV Protease Inhibitors, Cyclic Ureas, in
Rats and Dogs. Biopharm. Drug Dispos. 1994, 15, 535-544.
(28) Erickson, J .; Neidhart, D. J .; VanDrie, J .; Kempf, D. J .; Wang,
X. C.; Norbeck, D.; Plattner, J . J .; Rittonhouse, J .; Turon, M.;
Wideburg, N.; Kohlbrenner, W. E.; Simmer, R.; Helfrich, R.;
(19) A NL4-3-based reconstructed virus (10R/46I/63P/82T/84V) with
reduced senstivity to multiple protease inhibitors was provided
by W. Schleif of Merck Research Laboratories.
(20) Geng, X.; Chang, X. High-performance hydrophobic interaction
chromatography as a tool for protein folding. J . Chromatogr.
1992, 599, 185-194.
(21) Wlodawer, A.; Erickson, J . W. Structure-based inhibitors of
HIV-1 protease. Annu. Rev. Biochem. 1993, 62, 543-585.
(22) Kohl, N. E.; Emini, E. A.; Schleif, W. A.; Davis, L. J .; Heimbach,
J . C.; Dixon, R. A. F.; Scolnick, E. M.; Sigal, I. S. Active human
immunodeficiency virus protease is required for viral infectivity
Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 4686-4690.
(23) (a) Loeb, D. D.; Swanstrom, R.; Everitt, L.; Manchester, M.;
Stamper, S. E.; Hutchison, C. A., III. Complete mutagenesis of
the HIV-1 protease. Nature 1989, 340, 397-400. (b) Manchester,
M.; Everitt, L.; Loeb, D. D.; Hutchison, C. A., III; Swanstrom,
R. Identification of temperature-sensitive mutants of the human
immunodeficiency virus type 1 protease through saturation
mutagenesis. Amino acid side chain requirements for temper-
ature sensitivity. J . Biol. Chem. 1994, 269, 7689-7695.
(24) Patick, A. K.; Duran, M.; Cao, Y.; Ho, T.; Pei, Z.; Keller, M. R.;
Peterkin, J .; Chapman, S.; Anderson, B.; Ho, D.; Markowitz, M.
Genotypic and phenotypic characterization of HIV-1 variants
isolated from in vitro selection studies and from patients treated
Paul, D.; Knigge, M. Design, Activity, and 2.8
Å Crystal
Structure of a C2 Symmetric Inhibitor Complexed to HIV-1
Protease. Science 1990, 249, 527-533.
(29) Bru¨nger, A. T.; Kuriyan, J .; Karplus, M. Crystallographic R
Factor Refinement by Molecular Dynamics. Science 1987, 235,
458-460.
J M960586T