Nonsymmetrically substituted cyclic urea HIV protease inhibitors

Article abstract of DOI:10.1021/jm970288b

A series of nonsymmetrically substituted cyclic ureacarboxamides was synthesized and evaluated for antiviral activity as a function of the inhibition of HIV-protease. Selected protease inhibitors were also evaluated for oral bioavailability. The synthesis

Full text of DOI:10.1021/jm970288b

J . Med. Chem. 1997, 40, 4079-4088
4079
Non sym m etr ica lly Su bstitu ted Cyclic Ur ea HIV P r otea se In h ibitor s
Wendell W. Wilkerson,*,† Scott Dax,‡ and Walter W. Cheatham
DuPont Merck Pharmaceutical Company, Experimental Station, P.O. Box 80500, Wilmington, Delaware 19880-0500
Received May 2, 1997X
A series of nonsymmetrically substituted cyclic ureacarboxamides was synthesized and
evaluated for antiviral activity as a function of the inhibition of HIV-protease. Selected protease
inhibitors were also evaluated for oral bioavailability. The synthesis, pharmacology, quantita-
tive structure-activity relationship (QSAR), and pharmacokinetics for the series will be
discussed.
Ta ble 1. Pharmacological and Pharmacokinetic Profile for
In tr od u ction
Selected Anti-HIV N,N′-Bis-Substituted Cyclic Ureasa
The viral encoded aspartyl protease of the human
immunodeficiency virus (HIV) is responsible for the
processing of the viral polyprotein precursor to the
mature structural proteins and enzymes that comprise
the virus particle. Since correct processing of the viral
polypeptides is essential for the production of infectious
virus, HIV protease inhibitors represent a potential
target for therapeutic agents that may be effective in
the treatment of autoimmune disease syndrome (AIDS).
During our attempt to discover novel orally bioavail-
able HIV protease (HIV PR) inhibitors that might be
useful in the treatment of AIDS, members of these
laboratories synthesized the cyclic urea DMP450 (X )
3-NH2 MeSO3H). DMP450 was found to inhibit both
HIV PR with a Ki ) 0.31 nM and viral replication with
an IC90 ) 125.0 nM. Additionally, DMP450 was found
to have a desirable bioavailability profile with a F% )
79.2 (see Table 1). Our objective was to synthesize more
a potent cyclic urea (Ki < 0.030 nM and IC90 < 30 nM)
than DMP450 but with comparable pharmacokinetic
profile.
a
iv dosing was at 5.0 mg/kg and po dosing was at 10 mg/kg.
Ch em istr y
Sch em e 1. Synthetic Approach to the Symmetrical
Carboxamides 1a -c
The syntheses of the symmetrically substituted cyclic
ureas (1a -c) were accomplished using a modification
of the method described by Basha et al.1 where the ester
Ia 2 was reacted with excess dimethylaluminum amide
[(CH3)2AlNHR] prepared from equal molar amounts of
trimethylaluminum [(CH3)3Al] and the amine (RNH2)
in dichloroethane, and refluxed until chromatographic
methods indicated no Ia 2 remained. The method is
illustrated in Scheme 1 and is referred to as the Weinreb
Method. The nonsymmetrically substituted monocar-
boxamides were synthesized as shown in Scheme 2.
Generally, the appropriate methylisourea (Ib) was
alkylated with a substituted benzyl halide (P2CH2X˙ )
using NaH in anhydrous DMF. The resulting N-
substituted benzylisourea (II) was treated with methyl
3-(bromomethyl)benzoate in acetonitrile to give the
disubstituted cyclic urea (III). Intermediate III was
treated with a dimethylaluminum amide1 to give the
corresponding amide (IV). Treating IV with dilute HCl
produced the desired mixed pendent cyclic urea V. The
ester III was saponified to give the benzoic acid deriva-
tive IIIb which was coupled with R1NH2 using N,N-
dicyclohexylcarbodiimide-type reagents and conditions
to give IV. The structural components of these com-
pounds are shown in Table 2.
P h a r m a cology
† Current address: Dextron Corporation, P.O. Box 713, Bear, DE
The Ki values were determined with recombinant
single-chain dimeric HIV protease and a fluorescent
substrate (see Cheng et al.3) The use of single-chain
dimeric protease allows enzyme concentrations as low
19710.
‡ Current address: R. W. J ohnson Pharmaceutical Co., Springhouse,
PA.
X Abstract published in Advance ACS Abstracts, November 15, 1997.
S0022-2623(97)00288-4 CCC: $14.00 © 1997 American Chemical Society

4080 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 25
Wilkerson et al.
Ta ble 2. Structural and Physicochemical Data for the
Sch em e 2. Synthetic Approach to the Nonsymmetrical
Carboxamides
Nonsymmetrically Substituted Cyclic Ureas
Sch em e 3. Synthetic Approach to the
Anilinecarboxamides
ported in Table 3. The standard deviation (sd) for the
assay has been found to be <(40%. The Ki values for
a series of reference compounds are also included in
Table 3.
HIV RNA Assa y.5 The assay determines cell associ-
ated viral RNA levels 3 days after infection of suscep-
tible T-cell lines grown in individual microtiter wells.
Viral RNA was quantified by a sandwich hybridization
assay, the first step of which was performed directly in
crude infected cell lysates prepared in quinaldinium
isothiocyanate. Levels of cell-associated viral RNA were
shown to correlate with the yield of infectious virus and
this correlation formed the basis of the test. Antiviral
potencies of a large series of compounds tested in this
RNA hybridization assay correlated closely with potency
values determined by a sensitive but slower and more
labor-intensive yield reduction assay. Both laboratory
strains and selected clinical isolates of HIV can be
as 0.0625 nM to be used. Reaction products were
separated by HPLC with a Pharmacia Mono Q anion-
exchange column, and the product was quantified by
fluorescence. The ability of test compounds to block
cleavage of the HIV-1 gag polyprotein was assessed with
[35S]methionine-labeled in vitro translation product
corresponding to gag p17 plus the first 78 amino acids
of gag p24 and recombinant HIV PR as described by
Erickson-Viitanen et al.4 Ki values were measured with
62.5-250 pM HIV PR dimer and 1-10 nM inhibitor.
Each compound was assayed at least twice, and the
mean values for the experimental compounds are re-

Cyclic Urea HIV Protease Inhibitors
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 25 4081
Ta ble 3. Pharmacological, Chemical Characteristics, and
Physiochemical Data for the Cyclic Ureas and Reference
Compounds
Ki
IC90 log P
MW,
compd
(nM) (nM) HPLC CLOGP CMR anhydrous ϑNH
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
2.511 743.1 4.84
1.900 281.7 5.59
0.700 281.7 5.69
0.860 129.3 5.20
0.038 91.2 4.75
0.069 41.6 5.51
0.053 42.7 5.54
0.047 30.0 5.06
0.075 16.5 4.47
0.096 97.5 5.05
0.410 114.4 5.28
0.016 20.2 3.77
0.023 13.2 4.95
0.012 26.2 3.39
0.024 741.0 2.44
6.10
7.27
7.27
6.77
6.01
7.18
7.18
6.68
6.03
6.50
6.74
4.86
6.42
4.82
3.40
19.49
20.16
20.16
19.70
18.87
19.55
19.55
19.08
19.30
19.19
20.75
18.62
19.23
18.27
19.09
687.80
700.84
700.84
686.82
657.77
670.82
670.82
656.79
655.80
671.75
620.80
642.76
680.81
630.75
689.77
r ef
11.14
10.76
10.78
10.77
11.15
10.98
10.67
10.76
10.66
10.75
6.45
11.12
10.22
11.75
8.93
9.21
DMP323
DMP450
VX478
ABT538
Ro31-8959 0.25
MK-639 0.37
0.32 114.6 3.56
0.31 125.0 3.17
4.75
4.37
3.19
3.77
4.57
2.63
16.53
16.04
13.53
20.12
19.05
17.81
na
na
F igu r e 1. Anti-HIV cyclic ureas.
0.17
0.37
46.0
76.0
10.4
50.0
we chose the computer-calculated lipophilicity (CLOGP)
to represent the hydrophobic component, computer-
calculated molar refractivity (CMR) or the molecular
weight of the anhydrous compound (MW/100)9 to rep-
resent the steric component, and the proton nuclear
resonance chemical shift (ϑNH) of the amide proton
(CONHR, ppm) to represent the electronic component.
The CLOGP and CMR values were obtained using
MedChem Software v3.0, Pomona College, Claremont,
CA. It has been suggested that the CLOGP values for
some of the compounds in this series were “unrealisti-
cally high”. Consequently, a comparison was made
between CLOGP and HPLC measured log P (see Ex-
perimental Section) and found to be in excellent agree-
ment (r2 ) 0.99) where log PHPLC ) 0.83CLOGP-0.39
(see ref 10).
detected in this RNA hybridization assay. Results are
reported as IC90 (the concentration of antiviral com-
pound required to inhibit HIV RNA synthesis by 90%).
Initial results are reported in microgram per milliliter
(µg/mL) and converted to nanomolar (nM) for structure-
activity relationship studies (Table 3). The assay
provides a rapid, high-capacity assay for evaluating the
potency of anti-HIV compounds. The standard devia-
tion (sd) for the assay has been found to be <(37%. The
data obtained on reference compounds are listed in
Table 3.
Cytotoxicity TC50. Compound cytotoxicity was des-
ignated as TC50, which is defined as the concentration
of compound that produced a 50% reduction in the
number of viable cells as determined by the metabolism
of a tetrazolium dye. None of the compounds in this
report had TC50 < 50 µg/mL. A correct interpretation
of the RNA assay requires that test molecules not be
cytotoxicity at RNA assay dose levels.
P h a r m a cok in etic Stu d ies (P K). Pharmacokinetic
methodologies in these laboratories have previously
been reported.2,6,7 The study was conducted in the
female Beagle dog. The oral study (po) at 10.0 mg/kg
was conducted in three dogs, and the intravenous study
(iv) was conducted at 5.0 mg/kg. The vehicle used was
PEG400, water, and EtOH (60/30/10). In the current
study, the concentration was 10.0 mg/mL, and the
dosing volume was 1.0 mL/kg. The iv formulation was
the same as the oral: 10.0 mg/mL with dosing volume
of 0.5 mL/kg. The method was validated in dog plasma
at a concentration range 0.10-50.0 µg/mL.
Sta tistica l Meth od s a n d QSAR P a r a m eter s. Sta-
tistical analyses were conducted using J MP v3.0.2 by
SAS Institute, Cary, NC. The following statistical
measures were used: n, the number of samples in the
regression; r, correlation coefficient; s, root mean square
error of the regression; F-ratio; and the probability of
finding a greater F-ratio.
Hansch and Leo8 have stated that “...the structural
changes that affect the biological activities of a set of
congeners are of three major types: electronic, steric,
and hydrophobic”. In accordance with this philosophy,
The proton nuclear resonance chemical shift (ϑNH) of
the amide proton (CONHR, ppm) was used to represent
potential electronic effects that might be associated with
the R group. The chemical shift of the amide proton
was used as a parameter because previous findings and
molecular modeling studies on the symmetrical CUs11
demonstrated the importance of the presence and
nature of the amide proton for protease inhibition.
Method and values are listed in the Experimental
Section. This parameter has also been calculated by
computer and found to be in excellent agreement with
the observed values (data not included).
Discu ssion
Previous efforts in this laboratory resulted in the
synthesis of a series of bisbenzamides cyclic ureas as
exemplified by 1a -c.11 These bisamides were found to
be some of the most potent inhibitors of HIV PR and
viral replication (see Table 1). Unfortunately, the
bisamides did not have the desired oral bioavailability
profile as shown in Table 1. The previous study
suggested that the “larger” symmetrically substituted
cyclic ureas had poor gastrointestinal dissolution which
resulted in poor absorption which resulted in less than
desirable oral bioavailability. Dissolution and GI ab-
sorption were not problems with the smaller cyclic ureas
such as DMP3236,7,12-15 (X ) 4-CH2OH) or DMP450 (X
) 3-NH2) (see Figure 1). One attempt to enhance the
solubility of the compounds was to synthesize nonsym-
metrical cyclic ureas (increase the entropy, increase
solubility).

4082 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 25
Ta ble 4. Cross-Correlations for the Components of Eq 1a
Wilkerson et al.
Ta ble 5. Cross-Correlations for the Components of Eq 2a
variable
log(1/Ki)
CLOGP
CMR
ϑNH
variable
log(1/IC90
1.000
)
logKi
CLOGP
log(1/Ki)
CLOGP
CMR
1.000
-0.636
1.000
-0.801
0.712
1.000
0.209
-0.087
-0.608
1.000
log(1/IC90
log Ki
CLOGP
)
-0.646
1.000
0.087
0.547
1.000
ϑNH
observed versus predicted Ki values using eq 1a are
shown in Table 5 and 6. Compound 16 is a nonsym-
metrical diamide that was synthesized and added to the
regression analysis as a “bridge” between the bisamides
(1a -c) and the monoamides (2-15).
Our first attempt to prepare nonsymmetrical cyclic
ureas involved the substitution of one of the benzamides
of 1c with 5-(1,3-dimethoxyphenyl) (DMP) to produce 2
resulted in a loss of both protease inhibitory activity and
antiviral replication activity. Similarly, the “half amide”
3 (R ) 2-(5-Me-Pyr) and R′ ) DMP) and 5 (R ) 2-Pyr
and R′ ) DMP) were also less active than the corre-
sponding bisamides 1b and 1a , respectively. These data
clearly indicated that the attempt to alter the symmetry
of the cyclic ureas with the DMP moiety resulted in loss
of protease inhibition and that antiviral activity, and
regardless of bioavailability, would not be of further
interest. Substituting 3-methoxyphenyl (MMP) for
DMP resulted in 6, 7, 8, and 9, respectively. Clearly
the mixed amides of MMP were superior antiviral
agents to the corresponding DMP derivatives and, in
most cases, were superior to the reference compounds
shown in Table 3. However, they did not meet our
pharmacological objective of having a Ki < 0.030 nM
and an IC90 < 30 nM.
Since DMP450 (P2 ) CH2(3-NH2Ph)) was a moderate
inhibitor of viral protease and replication with the
desired PK profile, we synthesized a series of “aniline-
amides” (10, 12, 13, 14, and 15). The aniline-amide 10
met our requirements for viral replication inhibition
(IC90 ) 16.5 nM) but failed as a protease inhibitor (Ki
) 0.075 nM), and 12 failed in both respects. We were
successful in meeting our pharmacological objectives
with the synthesis of the more polar aniline-amides 13
(Ki ) 0.016 nM, IC90 ) 20.2 nM), 14 (Ki ) 0.023 nM,
IC90 ) 13.2 nM), and 15 (Ki ) 0.012 nM, IC90 ) 26.2
nM).
log(1/Ki) ) 0.221((0.076)CLOGP2 -
1.963((0.789)CLOGP - 2.344((0.337)CMR -
0.592((0.118)ϑNH + 56.052((8.195) (1a)
n ) 15
r ) 0.941
F ) 19.406
s ) 0.306
prob > F ) 0.000
CLOGP0 ) 4.44
When (MW/100)2 and (MW/100) replaced CMR2 and
CMR in a stepwise multiple regression, respectively, eq
1b resulted which was statistically less robust than eq
1a. However, (MW/100) was not highly correlated with
CLOGP (r2 ) 0.36) as was observed with CMR, and eq
1b did not contain a term for ϑNH. The only outlier to
eq 1b was 14 (Ki obs ) 0.023 nM vs Ki pred ) 0.198
nM). The reason for this difference is not known, but
when 14 was removed from the regression model, the
correlation was improved (r ) 0.92).
log(1/Ki) ) 0.190((0.123)CLOGP2 -
2.377((1.346)CLOGP - 7.398((2.308)(MW/100)2 +
96.294((30.473)(MW/100) - 304.541((99.365)
(1b)
n ) 15
r ) 0.852
s ) 0.474
prob > F ) 0.007
F ) 6.599
An ti-P r otea se a n d An tivir a l QSAR. In an attempt
to understand the physicochemical requirements for
HIV protease inhibition (log(1/Ki)), a stepwise multiple
regression was conducted on compounds 2-15 (Table
2) using the molecular parameters (CLOGP)2, CLOGP,
(CMR)2, CMR, and ϑNH (see Table 3). The proton NMR
chemical shift (ϑNH) of the amide proton was used as
an electronic parameter.16-20 This study then resulted
in eq 1a and the corresponding correlation matrix in
Table 4. Unfortunately, CLOGP and CMR were highly
cross-correlated (r2 ) 0.71) as was CMR and ϑNH (r2 )
0.61). The quadratic term for CLOGP in eq 1a would
suggest that for this series of compounds, the optimum
molecular lipophilicity occurred at CLOGP0 ) 4.442.
The lack of such a term for size or volume (CMR) would
suggest that the size of the molecules was not near
optimum. The implication of eq 1a was that smaller
more lipophilic carboxamides with less of a chemical
shift of the amide proton produce the better protease
inhibitor. The outliers to eq 1a were 2 (Ki obs ) 2.511
nM vs Ki pred ) 0.948 nM) and 10 (Ki obs ) 0.075 nM,
sd ) (0.010, n ) 3; Ki pred ) 0.197 nM). The reasons
for the differences for 10 are not known. Within the
framework of the error in the assay and the standard
error of the regression, the differences between the
observed versus the predicted Ki values for 2 are not
significant; qualitatively, they are the same. The
A similar approach was applied to viral replication
inhibition (log(1/IC90)) using the parameters (CLOGP)2,
CLOGP, (CMR)2, CMR, and log Ki obs to produce eq 2a
which was a modest descriptor of viral replication
inhibition for this data set. The correlation matrix for
eq 2a is shown in Table 5. The equation was of limited
predictive value since it relied on measured protease
inhibition activity (log Ki obs).
log(1/IC90) ) -0.217((0.080)CLOGP2 +
2.778((0.873)CLOGP + 0.329((0.262)CMR -
0.813((0.191) log Ki obs - 16.440((0.6.479) (2a)
n ) 15
r ) 0.867
F ) 7.589
s ) 0.336
prob > F ) 0.004
CLOGP0 ) 6.40
In an attempt to expand the utility of the relationship
expressed in eq 2a, we conducted a stepwise multiple
regression using (CLOGP)2, CLOGP, (CMR)2, CMR, and
the logarithm of the predicted Ki (log Ki pred) from eq
1a. This study resulted in eq 2b which was statistically
less satisfying than eq 2a. The principle outlier to eq
2b was 10 (R ) 2-(5-MePyr)) which was predicted to be
less active (IC90 ) 77.6 nM) than observed (IC90 ) 16.5
nM, sd ) (1.06, n ) 3). The reason for the failure of

Cyclic Urea HIV Protease Inhibitors
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 25 4083
Ta ble 6. Found Versus Predicted Anti-Protease (Ki) and -Viral (IC90) Activity
IN
no.
obs Ki
(nM)
pred Ki
(eq 1a)
pred Ki
(eq 1b)
obs IC90
(nM)
pred IC90
(eq 2a)
pred IC90
(eq 2b)
pred IC90
(eq 2c)
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
2.511
1.900
0.700
0.860
0.038
0.069
0.053
0.047
0.075
0.096
0.410
0.016
0.023
0.012
0.024
0.948
1.493
1.534
0.458
0.040
0.093
0.061
0.020
0.197
0.052
0.394
0.032
0.038
0.011
0.019
0.440
1.964
1.964
0.351
0.046
0.059
0.059
0.042
0.044
0.089
0.181
0.021
0.198
0.035
0.016
743.1
281.7
281.7
129.3
91.2
41.6
42.7
30.0
16.5
97.5
114.4
20.2
13.2
26.2
503.4
574.3
255.0
250.1
26.3
55.2
44.5
35.0
33.2
51.4
60.3
28.0
14.9
30.4
384.4
198.9
349.9
355.7
132.7
29.5
60.9
47.1
19.2
77.6
33.4
120.0
58.2
27.2
32.0
510.0
297.2
348.0
354.5
151.5
35.3
51.2
38.3
17.8
35.5
137.0
74.8
29.1
37.8
741.0
447.7
Ta ble 7. Pharmacological and Pharmacokinetic Profile for
Selected Anti-HIV Nonsymmetrically Substituted Cyclic Ureas
eq 2b to correctly predict the IC90 for 10 is believed to
be the result of the poorly predicted Ki value from eq
1a (Ki obs ) 0.075 nM, sd ) (0.010, n ) 3; Ki pred )
0.197 nM).
log(1/IC90) ) -0.166((0.080)CLOGP2 +
2.112((0.885)CLOGP - 0.608((
0.180) log Ki pred - 9.019((2.375) (2b)
n ) 15
r ) 0.797
F ) 6.372
s ) 0.389
prob > F ) 0.009
CLOGP0 ) 6.36
When 10 was removed from the regression, eq 2c
resulted which was a marginal improvement over eq 2b.
Equations 2a-c would suggest that the optimum CLOGP
for this class of compounds would lie in the range 6.4-
6.9. The observed versus predicted IC90 values for eqs
2a-c are listed in Table 6.
log(1/IC90) ) -0.114((0.073)CLOGP2 +
1.581((0.799)CLOGP - 0.685((0.159)
log Ki pred - 7.891((2.107) (2c)
n ) 14
r ) 0.851
F ) 8.727
s ) 0.335
overall PK profile represented an improvement over the
bisamides and DMP450.
prob > F ) 0.004
CLOGP0 ) 6.93
Con clu sion
P h a r m a cok in etics
By incorporating the P2 substituent of DMP450
(CH2(3-NH2Ph)) and the P2′ substituent of selected
biscarboxamides (CH2(3-RNHCOPh)), we have been
successful in synthesizing cyclic urea “aniline-amides”
with the desired biochemical properties of both “par-
ents”: excellent inhibitory activity against the protease
(Ki < 0.030 nM) and viral (IC90 < 30.0 nM) and good to
excellent oral bioavailability (see Tables 1 and 5).
Limited QSAR studies on this series of cyclic ureas
suggest that protease inhibition is related to electronic
property of the amide proton as determined by NMR
chemical shift and the size and lipophilicity of the
molecule (eq 1a). A large portion of the antiviral activity
could be explained by the inhibition of the protease; and
the size and, possibly more importantly, the lipophilicity
of the molecule. However, the QSAR studies demon-
strated that the optimum lipophilicity for protease
inhibition (CLOGP0 ) 4.44) was different from that
Two compounds (15 and 14) met our pharmacological
objectives and were subjected to pharmacokinetic stud-
ies in the female Beagle dog. The results are shown in
Table 7. The clearance for 15 was between that
determined for 14 and DMP450. The volume of distri-
bution was ∼4 L/kg which was 3-4 times higher than
that determined for 14 and DMP450. This result
indicated that 15 distributed into tissues much more
than the other two compounds. However, the site of
increased distribution remains to be determined. One
beneficial consequence was the prolongation of half-life.
The observed a half-life for 15 was 5.6 h after iv and 10
after po dosing. At the same time, due to greater
distribution to tissues, we had expected lower plasma
concentrations. The Cmax for 15 , after 10 mg/kg dose,
was 0.9 µg/mL. Although it was lower than that for 14
and DMP450, and considering the longer half-life, the

4084 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 25
Wilkerson et al.
cm-1; MS (NH3-DCI) m/e 747 (M + 1); [R]20D +57.14° (c ) 0.098,
DMSO). Anal. Calcd for C45H42N6O5‚2HCl, MW 819.80: C,
65.93; H, 5.41; N, 10.25. Found: C, 65.72; H, 5.64; N, 10.10.
required to inhibit viral replication (CLOGP0 ) 6.36).
These differences make it more difficult to optimize both
pharmacological activities for the same molecule. Simi-
lar differences and difficulties may exist for other
properties considered in selecting a compound for
further development.
The selection of 14 or 15 as possible development
candidates will also depend on other characteristics such
as protein binding, toxicity, and activity against selected
HIV mutants. All such studies are presently under
investigation
A solution of the dihydrochloride salt (0.240 g, 0.030 mmol)
in 10 mL of acetonitrile was made alkaline with 5% NaHCO3
and stirred at room temperature for an additional 3 h. The
resulting white solid was collected by filtration, washed with
water, and air-dried to give the desired product as the free
base in 91% (0.2046 g) yield: 1H NMR (300 MHz, DMSO-d6,
TMS) δ 2.78 (dd, 2H, Ar′CH), 3.0 (m, 4H, Ar′CHCNCH), 3.5
(m, 4H, OCHCH), 4.64 (d, J ) 13.9 Hz, 2H, NCH), 5.16 (s,
2H, OH), [6.95 (d, 4H), 7.2 (m, 8H), 7.37 (d, 2H), 7.44 (dd, 2H),
7.85 (m, 4H), 7.95 (d, 2H), 8.18 (d, 2H), 8.40 (d, 2H), Ar], 10.77
(s, 2H, NH); IR (Nujol) 3397 (OH), 1678 (CdO), 1635 (CdO)
cm-1; UV-vis (c ) 0.0149 mg/mL, MeOH) λmax 281 (29 686),
246 (25 352), 214 (57 591) nm; MS (NH3-DCI) m/e 747 (M +
1). Anal. Calcd for C45H42N6O5, MW 746.78: C, 72.37; H, 5.67;
N, 11.25. Found: C, 71.98; H, 5.98; N, 11.06.
Exp er im en ta l Section
Melting points were determined with a Thomas-Hoover
melting point apparatus and are uncorrected. The NMR
spectra were recorded with a Varian-300S spectrometer, IR
spectra were recorded with a Perkin-Elmer 1650 FTIR spec-
trophotometer, UV spectra were obtained with a Cary 2415
spectrophotometer, optical rotations (OR) were determined on
a Perkin-Elmer 241 polarimeter, and mass spectra (MS) were
obtained using the Hewlett-Packard HP5988A GC-MS sys-
tem. Analytical HPLC determinations were obtained with a
system composed of two Varian 2510 pumps and a Varian 2550
variable-wavelength detector using a 4.6 × 250 mm Zorbax
ODS column and CH3CN-water mobile phase, HPLC log P
(4r,5r,6â,7â)-3,3′-[[Tetr a h yd r 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 ia ze p in e -1,3(2H )-d iyl]b is-
(m eth ylen e)]bis[N-(5-m eth yl-2-pyr idin yl)ben zam ide] (1b).
By substituting 2-amino-5-picoline in the Weinreb method, the
desired product was obtained in 13% (0.157 g) yield: mp 132-
133 °C; 1H NMR (300 MHz, DMSO-d6, TMS) δ 2.28 (s, 6H,
ArCH3), 2.78 (dd, 2H, Ar′CH), 3.0 (m, 4H, Ar′CHCNCH), 3.45
(2, OH), 3.5 (m, 4H, Ar′CCHCHO), 4.68 (d, J ) 14.3 Hz, 2H,
NCH), [6.96 (d, 4H), 7.22 (m, 6H), 7.37 (d, 2H), 7.46 (dd, 2H),
7.68 (m, 2H), 7.87 (s, 2H), 7.95 (d, 2H), 8.08 (d, 2H), 8.22 (s,
2H), Ar], 10.72 (s, 2H, NH); IR (Nujol) 3392 (OH), 1676 (CdO),
1646 (CdO) cm-1; UV-vis (c ) 0.0170 mg/mL, MeOH) λmax
287 (28 399), 258 (24 980), 215 (50 962) nm; MS (NH3-DCI)
determinations were obtained with
a Varian STAR-9000
system using a YMC C18 column and 55% MeOH and 45%
pH 7.0 phosphate buffer as mobile phase. Thin layer chro-
matography (TLC) was performed on silica gel plates.
Syn th esis of In ter m ed ia te Meth ylisou r ea (Ib). A solu-
tion of [3aS-(3aR,4â,8R,8aâ)]-hexahydro-2,2-dimethyl-4,8-bis-
(phenylmethyl)-6H-1,3-dioxolo[4,5-e][1,3]diazepin-6-one (I2) (3.00
g, 8.19 mmol) in dichloroethane (20 mL) was treated with
methyl trifluoromethanesulfonate (1.02 mL, 9.01 mmol) and
refluxed for 16 h under anhydrous conditions. The solution
was cooled to room temperature, diluted with CH2Cl2 (100 mL),
and washed with saturated NaHCO3. The organic phase was
washed with brine, dried over MgSO4, filtered, and concen-
trated in vacuo to give the desired isourea as a foam (3.10 g,
99% yield). The foam was inspected by TLC (CHCl3-EtOAc,
3:2) and found to be homogeneous, and used without further
purification: IR (Nujol) no CdO; MS (NH3-DCI) m/e 381 (M
+ 1); C23H28N2O3, MW 380.49.
(4r,5r,6â,7â)-3,3′-[[Tetr a h yd r 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 ia ze p in e -1,3(2H )-d iyl]b is-
(m eth ylen e)]bis[N-2-p yr id in ylben za m id e] (1a ). The com-
pound was synthesized using a modification of the Weinreb
method.1 A solution of 2-aminopyridine (100 mmol) in dichlo-
roethane (50 mL) was treated with 2 M trimethylaluminum
(TMA) (50.0 mL, 100 mmol), stirred at room temperature for
10 min, and added to a solution of Ia 2 (5.00 g, 7.54 mmol) in
dichloroethane (150 mL). The mixture was refluxed under dry
nitrogen for 25 h and inspected by TLC (CHCl3-EtOAc, 3:2)
which showed no Ia . The mixture was diluted with 500 mL
of CH2Cl2 and 500 mL of water, stirred for 1 h, and filtered
through a bed of Celite. The organic phase was washed with
water (50 mL), 5% NaHCO3 (2 × 25 mL), water, and brine;
dried over MgSO4; filtered; and concentrated to give the
bisamide acetonide as an off-white solid.
m/e 775 (M + 1), 388 (M + 2H)2+; [R]20 +82.93° (c ) 0.082,
D
MeOH). Anal. Calcd for C47H46N6O5‚0.5H2O, MW 783.93: C,
72.01; H, 6.04; N, 10.72. Found: C, 71.84; H, 6.08; N, 10.47.
(4r,5r,6â,7â)-3,3′-[[Tetr a h yd r 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 ia ze p in e -1,3(2H )-d iyl]b is-
(m eth ylen e)]bis[N-2-p yr a zin ylben za m id e] (1c). By sub-
stituting 2-aminopyrazine in the Weinreb method for 1a , the
desired product was obtained in 58% (0.659 g) yield from Ia :
1
mp 146-148 °C; H NMR (300 MHz, DMSO-d6, TMS) δ 2.77
(dd, 2H, Ar′CH), 2.99 (d, J ) 13.18 Hz, 2H, Ar′CH), 3.07 (d, J
) 14.28 Hz, 2H, NCH), 3.55 (m, 4H, OCHCH), 4.69 (d, J )
14.28 Hz, 2H, NCH), 5.17 (s, 2H, OH), [6.95 (d, 4H), 7.22 (m,
6H), 7.42 (d, 2H), 7.50 (dd, 2H), 7.90 (s, 2H), 7.98 (d, 2H), 8.42
(d, 2H), 8.48 (d, 2H), 9.41 (s, 2H), Ar], 11.13 (s, 2H, NH); IR
(KBr) 4312 (OH, 1686 (CdO), 1642 (CdO) cm-1; UV-vis (c )
0.0210 mg/mL, MeOH) λmax 298 (20 076), 284 (25 140), 245
(25 140), 217 (42 898) nm; MS (NH3-DCI) m/e 749 (M + 1),
375 (M + 2)2+; [R]20D +78.57° (c ) 0.084, MeOH). Anal. Calcd
for C43H40N8O5, MW 748.84: C, 68.97; H, 5.38; N, 14.96.
Found: C, 68.70; H, 5.38; N, 14.72.
(4r,5r,6â,7â)-3-[[3-[(3,5-Dim e t h oxyp h e n yl)m e t h yl]-
h exa h yd r o-5,6-d ih yd r oxy-2-oxo-4,7-b is(p h en ylm et h yl)-
1H-1,3-diazepin -1-yl]m eth yl]-N-2-pyr azin ylben zam ide (2).
A suspension of 60% NaH in mineral oil (0.82 g, 20.46 mmol)
in dry THF (50 mL) was treated with the cyclic urea2 (5.00 g,
13.64 mmol) and stirred under dry nitrogen in an ice bath for
30 min. The suspension was treated with 3,5-dimethoxybenzyl
chloride (3.82 g, 20.46 mmol) in dry THF (10 mL). The mixture
was stirred in the ice bath for 5 min and at room temperature
for 48 h and diluted with EtOAc (150 mL) and saturated NH4-
Cl (50 mL). The organic layer was washed with water, 5%
NaHCO3, water, and brine; dried over MgSO4; filtered; and
concentrated in vacuo to give a crude product containing mono-
and disubstituted cyclic urea and unreacted starting cyclic
urea. The crude product was column chromatographed on
silica gel using EtOAc-hexane (1:1) to remove the disubsti-
tuted urea followed by straight EtOAc to recover the desired
product. Appropriate fractions were combined and concen-
trated in vacuo to give the desired monoalkylated cyclic urea
(6.91 g, 98% yield) as a foam: 1H NMR (300 MHz, CDCl3-
TMS) δ 1.45 (d, 6H, CH3CCH3), 2.8 (m, 2H, Ar′CH and NCH),
3.05 (m, 3H, Ar′CH2 and Ar′CH), [3.55 (m, 1H), 3.8 (m, 1H),
3.88 (m, 1H), 4.25 (m, 1H, CHCH(O)CH(O)CH]), 3.71 (s, 6H,
OCH3), 4.92 (d, 1H, NH), 5.06 (d, J ) 14.7 Hz, 1H, NCH), [6.20
The acetonide (1.93 g, 2.45 mmol) in 20 mL CH3CN was
treated with 10 mL of 1 N HCl and stirred at room temper-
ature until no starting material remained as demonstrated by
TLC (CHCl3-MeOH, 9:1). The mixture was concentrated in
vacuo at 60 °C to give a white solid which was triturated with
10 mL of water and placed in the cold for 16 h. The resulting
crystals were collected by filtration, washed with a small
amount of cold water, and air-dried to give the desired product
1
in 77% (1.415 g) yield: mp 158-161 °C; H NMR (300 MHz,
DMSO-d6 TMS) δ 2.77 (dd, 2H, Ar′CH), 2.99 (d, J ) 12.4 Hz,
2H, Ar′CH), 3.07 (d, J ) 14.3 Hz, 2H, NCH), 3.55 (s, 2H, OCH),
3.60 (d, J ) 11.7 Hz, 2H, Ar′CCH), 4.68 (d, J ) 14.3 Hz, 2H,
NCH), [6.95 (d, 2H), 7.22 (m, 6H), 7.4 (m, 4H), 7.52 (dd, 2H),
7.92 (s, 2H), 8.04 (d, 2H), 8.16 (m, 4H), 8.45 (d, 2H), Ar], 11.59
(s, 2H, NH); IR (Nujol) 3334 (OH), 1676 (CdO), 1640 (CdO)

Cyclic Urea HIV Protease Inhibitors
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 25 4085
(d, 2H), 6.33 (m, 1H), 7.2-7.4 (m, 10H), Ar]; MS (NH3-DCI)
m/e 534 (M + NH4) for C31H36N2O5, MW 516.64.
N, 9.87. Analytical HPLC: Zorbax ODS 4.6 × 250 mm column;
solvent H2O-CH3CN; flow rate of 1 mL/min; detector at 256
nm.
A mixture of the above monoalkylated cyclic urea (2.5 g, 4.84
mmol) and 60% NaH (0.252 g, 6.29 mmol) in dry DMF (25 mL)
was stirred at room temperature for 30 min and treated with
methyl 3-(bromomethyl)benzoate (1.22 g, 5.32 mmol) in dry
DMF (2 mL). The mixture was stirred at room temperature
until TLC (EtOAc-hexane, 1:3) showed no starting material.
The mixture was poured into a mixture of cracked ice (250 g)
and saturated NH4Cl (75 mL). The mixture was stirred until
the ice had melted, and the resulting solid was collected by
filtration. The solid was dissolved in EtOAc (50 mL); washed
with water, 5% NaHCO3, water, and brine; dried over MgSO4;
filtered; and concentrated to an impure solid (TLC, EtOAc-
hexane, 1:3). The crude product was column chromatographed
on silica gel using EtOAc-hexane (1:4) as mobile phase. Two
sets of fractions were collected and concentrated to constant
weight: half ester fractions A, pure product (1.050 g); and
fraction B, impure product (1.95 g) (90+% product by analyti-
cal HPLC). Fraction A was collected as a white foam in 33%
(1.05 g) yield: 1H NMR (300 MHz, DMSO-d6, TMS) δ 1.31 (2s,
6H, CH3CCH3), 2.7 (m, 1H, Ar′CH), 2.85 (m, 3H, Ar′CH), 3.00
(d, J ) 13.9 Hz, 1H, NCH), 3.32 (d, J ) 7.3 Hz, 1H, Ar′CCH),
3.40 (d, J ) 14.3 Hz, 1H, NCH), 3.69 (s, 6H, ArOCH3), 3.81 (s,
3H, CO2CH3), 3.9 (m, 1H, Ar′CCH), 4.03 (d, J ) 8.4 Hz, 2H,
OCH), 4.58 (“d”, J ) 13.9 Hz, 2H, NCH), [6.31 (m, 2H), 6.39
(s, 1H), 6.85 (m, 2H), 7.05 (d, 2H), 7.23 (m, 6H), 7.45 (dd, 1H),
7.48 (d, 1H), 7.84 (m, 2H), Ar]; IR (KBr) 1724 (CdO), 1632
(CdO) cm-1; MS (NH3-DCI) m/e 682 (M + NH4) for C40H44N2O7,
MW 664.80. Analytical HPLC: Zorbax ODS column F44558,
4.6 × 25 cm, flow rate 1 mL/min, water-acetonitrile, wave-
length 256. Fraction B was also used in subsequent syntheses.
The above half ester (0.500 g, 0.752 mmol) was dissolved in
10 mL of dichloroethane and treated with a mixture of
2-aminopyrazine (0.946 g, 9.95 mmol) and 2 M trimethylalu-
minum (4.97 mL) in 10 mL of dichloroethane. The mixture
was refluxed under dry nitrogen for 24 h, cooled to room
temperature, and diluted with 150 mL of methylene chloride
and 50 mL of water. The mixture was stirred for 30 min and
filtered through a bed of Celite. The aqueous phase was
decanted, and the organic phase was evaporated in vacuo. The
residue was dissolved in EtOAc (50 mL), and the solution was
washed with water (2 × 25 mL), 5% NaHCO3 (25 mL), water
(25 mL), and brine (25 mL); dried over MgSO4; filtered; and
concentrated to a gum. The crude intermediate was column
chromatographed on silica gel (50 g) using EtOAc-hexanes
(3:1) as solvent. Appropriate fractions were combined and
concentrated to constant weight to give the desired intermedi-
ate as a pure white foam (0.33 g, 60% yield): 1H NMR (300
MHz, CDCl3, TMS) δ 1.39 (d, 6H, CH3CCH3), 2.88 (dd, 1H,
Ar′CH), 3.0 (m, 4H), 3.36 (d, J ) 14.3 Hz, 1H, NCH), 3.72 (s,
6H, OCH3), 3.78-4.02 (m, 5H), 4.87 (d, J ) 14.3 Hz, 1H, NCH),
4.90 (d, J ) 14.3 Hz, 1H, NCH), [6.25 (d, 2H), 6.33 (d, 1H),
7.02 (d, 2H), 7.17 (d, 2H), 7.23-7.40 (m, 6H), 7.47 (m, 2H),
7.64 (s, 1H), 7.82 (d, 1H), 8.29 (s, 1H), 8.38 (d, 1H), 8.49 (s,
1H), Ar], 9.68 (s, 1H, NH); IR (KBr) 3420 (NH), 1684 (CdO),
1630 (CdO) cm-1; MS (NH3-DCI) m/e 728 (M + 1) for
C43H45N5O6, MW 727.86.
(4r,5r,6â,7â)-3-[[3-[(3,5-Dim e t h oxyp h e n yl)m e t h yl]-
h exa h yd r o-5,6-d ih yd r oxy-2-oxo-4,7-b is(p h en ylm et h yl)-
1H -1,3-d ia zep in -1-yl]m et h yl]-N-(5-m et h yl-2-p yr id in yl)-
ben za m id e Hem ih yd r a te (3). By substituting 2-amino-5-
methylpyridine in the method for 2, the desired product was
obtained in 42% yield from the ester: mp 236-238 °C; 1
(300 MHz, DMSO-d6, TMS) δ 2.28 (s, ArCH3), 2.71 (m, 2H,
Ar′CH), 2.93-3.08 (m, 4H, Ar′CH and NCH), 3.37-3.60 (m,
H NMR
4H, CHCHCHCH), 3.68 (s, 6H, OCH3
), 4.63 (d, J ) 14.3 Hz,
1H, NCH), 4.70 (d, J ) 13.9 Hz, 1H, NCH), 5.05 (broad s, 2H,
OH), [6.23 (2, 2H), 6.39 (s, 1H), 6.94 (d, 2H), 7.1 (d, 2H), 7.2-
7.46 (m, 8H), 7.68 (d, 1H), 7.86 (1, 1H), 7.94 (d, 1H), 8.06 (d,
1H), 8.22 (s, 1H), Ar], 10.76 (s, 1H, NH); IR (KBr) 3400 (OH),
3288 (NH), 1684 (CdO), 1616 (CdO) cm-1
; UV-vis (c ) 0.0219
mg/mL, MeOH) λmax 284 (14 465), 270 (12 481), 261 (11 969)
nm; MS (NH3-DCI) m/e 701 (M + 1); [R]20D +73.81° (c ) 0.210,
MeOH). Anal. Calcd for C42H44N4O6 0.5H2O, MW 709.85: C,
71.07; H, 6.39; N, 7.89. Found: C, 71.27; H, 6.25; N, 7.79.
(4r,5r,6â,7â)-3-[[3-[(3,5-Dim e t h oxyp h e n yl)m e t h yl]-
h exa h yd r o-5,6-d ih yd r oxy-2-oxo-4,7-b is(p h en ylm et h yl)-
1H -1,3-d ia zep in -1-yl]m et h yl]-N-(6-m et h yl-2-p yr id in yl)-
ben za m id e Hem ih yd r a te (4). By substituting 2-amino-6-
methylpyridine in the method for 2, the desired product was
obtained in 43% yield from the half ester: mp 207-209 °C;
1H NMR (300 MHz, DMSO-d6, TMS) δ 2.46 (s, 3H, ArCH3),
2.72 (m, 2H, Ar′CH), 2.89-3.10 (m, 4H, Ar′CH and NCH),
3.43-3.62 (m, 4H, CHCHCHCH), 3.68 (s, 6H, OCH3), 4.63 (d,
J ) 14.3 Hz, 1H, NCH), 4.71 (d, J ) 14.3 Hz, 1H, NCH), 4.90
(broad s, 2H, OH), [6.23 (s, 2H), 6.39 (s, 1H), 6.93 (d, 2H), 7.06
(d, 2H), 7.1-7.48 (m, 7H), 7.75 (m, 1H), 7.89 (s, 1H), 7.97 (m,
2H), Ar], 10.78 (s, 1H, NH); IR (KBr) 3370 (OH), 3304(NH),
1670 (CdO), 1609 (CdO) cm-1; UV-vis (c ) 0.0150 mg/mL,
MeOH) λmax 284 (15 745), 278 (13 783) nm; MS (NH3-DCI) m/e
701 (M + 1); [R]20 +72.53° (c ) 0.142, MeOH). Anal. Calcd
D
for C42H44N4O6 0.5H2O, MW 709.85: C, 71.07; H, 6.39; N, 7.89.
Found: C, 71.05; H, 6.39; N, 7.52.
(4r,5r,6â,7â)-3-[[3-[(3,5-Dim e t h oxyp h e n yl)m e t h yl]-
h exa h yd r o-5,6-d ih yd r oxy-2-oxo-4,7-b is(p h en ylm et h yl)-
1H -1,3-d ia zep in -1-yl]m et h yl]-N-2-p yr id in ylb en za m id e
Hem ih yd r a te (5). By substituting 2-aminopyridine in the
method for 2, the desired product was obtained in 31% yield
from the half ester: mp 215-218 °C;
1H NMR (300 MHz,
DMSO-d6, TMS) δ 2.73 (m, 2H, Ar′CH), 2.87-3.09 (m, 4H,
Ar′CH and NCH), 3.32-3.59 (m, 4H, CHCHCHCH), 3.68 (s,
6H, OCH3), 4.64 (d, J ) 14.3 Hz, 1H, NCH), 4.72 (d, J ) 14.0
Hz, 1H, NCH), 5.15 (2s, 2H, OH), [6.24 (d, 2H), 6.39 (s, 2H),
6.94 (d, 2H), 7.1-7.4 (m, 11H), 7.83 (m, 2H), 7.95 (d, 1H), 8.18
(d, 1H), 8.39 (d, 1H), Ar], 10.77 (s, 1H, NH); IR (KBr) 3410
(OH and NH), 1686 (CdO), 1600 (CdO) cm-1
; UV-vis (c )
0.021 mg/mL, MeOH) λmax 282 (15 371), 277 (14 914) nm; MS
(NH3-DCI) m/e 704 (M + NH4); [R]20 +75.00° (c ) 0.144,
D
MeOH). Anal. Calcd for C41H42N4O6‚0.5H2O, MW 695.82: C,
70.77; H, 6.23; N, 8.05. Found: C, 70.87; H, 6.18; N, 7.66.
(4r,5r,6â,7â)-3-[[Hexa h yd r o-5,6-d ih yd r oxy-3-[(3-m eth -
oxyp h en yl)-m eth yl]-2-oxo-4,7-bis(p h en ylm eth yl)-1H-1,3-
d ia zep in -1-yl]m et h yl]-N-2-p yr a zin ylb en za m id e (6). By
substituting 3-methoxybenzyl chloride in the method used to
synthesize 2, the desired intermediate half ester was obtained.
A solution of half ester (0.500 g, 0.788 mmol) in 10 mL of
dichloroethane was treated with a mixture of 2-aminopyrazine
(0.992 g, 10.43 mmol) and 2 M trimethylaluminum (5.23 mL)
in 10 mL of dichloroethane. The mixture was refluxed under
dry nitrogen for 48 h, cooled to room temperature, and diluted
with 200 mL of CH2Cl2 and 75 mL of water. The resulting
emulsion was filtered through Celite. The organic phase was
separated from the aqueous phase; washed with water, 5%
NaHCO3, water, and brine; dried over MgSO4; filtered; and
concentrated to an impure foam (0.47 g). The crude material
was column chromatographed on silica gel using EtOAc-
hexane (3:2) as mobile phase. Appropriate fractions were
combined and concentrated to give the desired pure intermedi-
ate a foam (0.190 g, 35%): 1H NMR (300 MHz, CDCl3, TMS)
δ 1. 37 (s, 3H, CH3), 1.40 (s, 3H, CH3), 2.81-3.04 (m, 6H,
The above intermediate (0.30 g, 0.412 mmol) was dissolved
in 10 mL of acetonitrile, treated with 10 mL of 1 N HCl, and
stirred at room temperature for 16 h. The mixture was diluted
with 150 mL of water, stirred for 1 h, and filtered. The
resulting white solid was washed with water and dried in
vacuo to give the desired product in 74% (0.209 g) yield or 40%
yield from the half ester: mp 151-153 °C; 1H NMR (300 MHz,
DMSO-d6, TMS) δ 2.7 (m, 2H, Ar′CH), 2.9-3.15 (m, 4H, Ar′CH
and NCH), 3.5 (m, 3H), 3.58 (m, 2H), 3.69 (s, 6H, OCH3), 4.60-
4.74 (2d, 2H, NCH), 5.16 (2s, 2H, OH), [6.23 (d, 2H), 6.39 (m,
1H), 6.94 (d, 2H), 7.11 (d, 2H), 7.16-7.38 (m, 6H), 7.42 (d, 1H),
7.50 (dd, 1H), 7.89 (s, 1H), 7.97 (d, 1H), 8.40 (m, 1H), 8.47 (m,
1H), 9.40 (s, 1H), Ar], 11.14 (s, 1H, NH); IR (KBr) 3412 (OH
and NH), 1684 (CdO), 1640 (CdO) cm-1; UV-vis (c ) 0.0120
mg/mL, MeOH) λmax 302 (10 145), 292 (11 349), 283 (14 959),
251 (13 240) nm; MS (NH3-DCI) m/e 705 (M + NH4); [R]20
D
+75.46° (c ) 0.33, MeOH). Anal. Calcd for C40H41N5O6, MW
687.80: C, 69.85; H, 6.02; N, 10.18. Found: C, 69.59; H, 5.88;

4086 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 25
Wilkerson et al.
Ar′CH2 and NCH), 3.32 (d, J ) 14.6 Hz, 1H, NCH), 3.74 (s,
3H, OCH3), 3.78-3.89 (m, 4H, CHCHCHCH), 4.84 (d, J ) 14.3
Hz, 1H, NCH), 4.91 (d, J ) 14.6 Hz, 1H, NCH), [6.64 (s, 1H),
6.71 (m, 2H), 7.02 (d, 2H), 7.2-7.37 (m, 10H), 7.45 (d, 1H),
7.64 (s, 1H), 7.81 (d, 1H), 8.27 (d, 1H), 8.38 (d, 1H), 8.47 (s,
1H), Ar], 9.69 (s, 1H, NH); MS (NH3-DCI) m/e 698 (M + 1) for
C42H43N5O5, MW 697.84.
7.1-7.35 (m, 8H), 7.39 (d, 1H), 7.44 (dd, 1H), 7.82 (m, 1H),
7.88 (s, 1H), 7.95 (d, 1H), 8.18 (d, 1H), 8.39 (d, 1H), Ar], 10.76
(s, 1H, NH); IR (KBr) 3452 (OH), 3304 (NH), 1692 (CdO), 1612
(CdO) cm-1; UV-vis (c ) 0.12 mg/mL, MeOH) λmax 286
(17 022), 261 (12 588), 221 (40 611) nm; MS (NH3-DCI) m/e
657 (M + 1), 674 (M + NH4); [R]20D +91.25° (c ) 0.080, MeOH).
Anal. Calcd for C40H49N4O5‚0.5H2O: C, 72.18; H, 6.21; N, 8.42.
Found: C, 71.78; H, 5.88; N, 8.15.
A solution of the acetonide (0.170 g, 0.244 mmol) in 10 mL
of acetonitrile was treated with 10 mL of 1 N HCl and stirred
at room temperature for 24 h. The mixture was diluted with
50 mL of water and adjusted to pH 8 with 5% NaHCO3. The
mixture was stirred at 4 °C for 2 h, and the resulting white
solid was collected by filtration, washed with cold water, and
dried in vacuo at 80 °C for 16 h to give the desired product in
90% (0.145 g) yield, or 32% from the half ester: mp 112 °C
(4R,5R,6â,7â)-3-[[3-[(3-Am in oph en yl)m eth yl]h exah ydr o-
5,6-d ih yd r oxy-2-oxo-4,7-b is(p h en ylm et h yl)-1H-1,3-d ia z-
ep in -1-yl]m et h yl]-N-(5-m et h yl-2-p yr id in yl)b en za m id e
Hem ih yd r a te (10). The methylisourea Ib (13.14 mmol) was
alkylated with methyl 3-bromobenzoate using NaH in dry
DMF to give (methyl 3-methylbenzoate)-methylisourea in 86%
(5.96 g) yield as an oil after column chromatography on silica
gel (EtOAc-hexane, 1:4): 1H NMR (300 MHz, DMSO-d6, TMS)
δ 1.46 (s, 6H, CH3CCH3), 2.7-2.86 (m, 2H, Ar′CH), 2.99-3.10
(m, 2H, Ar′CH), 3.18 (d, J ) 14.3 Hz, 1H, NCH), 3.38 (s, 3H,
OCH3), 3.71 (m, 1H, CH), 3.86 (s, 3H, OCH3), 4.18 (m, 3H,
CHCHCH), 4.38 (d, J ) 14.65, 1H, NCH), 6.98-7.92 (m, 14H,
Ar); MS (NH3-DCI) m/e 529 (M + NH4). The methyl benzoate-
methylisourea (11.22 mmol) was alkylated by refluxing with
3-nitrobenzyl bromide in acetonitrile for 3 days. After the
usual workup, the crude material was column chromato-
graphed of silica gel (EtOAc-hexane, 9:1) to give desired
product in 54% yield: MS (NH3-DCI) m/e 667 (M + NH4). The
nitrobenzyl methyl benzoate (1.38 mmol) was reacted with
2-amino-5-methylpyridine under Weinreb conditions to give
the corresponding amide in 35% yield as a foam: MS (NH3-
DCI) m/e 726 (M + NH4). The nitrobenzyl amide acetonide
(0.45 mmol) was reduced with 10% Pd/C in CH2Cl2 to give the
desired anilino amide acetonide in 100% yield.
1
dec; H NMR (300 MHz, DMSO-d6, TMS) δ 2.65-3.1 (m, 6H,
2Ar′CH2 + 2NCH)), 3.3-3.65 (m, 4H, NCHCHCHCHN), 3.59
(s, 3H, OCH3), 4.66 (d, J ) 14.3 Hz, 1H, NCH), 4.71 (d, J )
14.3 Hz, 1H, NCH), 5.15 (s, 2H, 2OH), [6.65 (s, 1H), 6.69 (d,
1H), 6.83 (m, 1H), 6.94 (d, 2H), 7.07 (d, 2H), 7.15-7.35 (m,
7H), 7.42 (d, 1H), 7.49 (dd, 1H), 7.89 (s, 1H), 7.97 (d, 1H), 8.42
(m, 1H), 8.48 (m, 1H), 9.40 (s, 1H), Ar], 11.15 (s, 1H, NH); IR
(KBr) 3412 (OH and NH), 1686 (CdO), 1604 (CdO) cm-1; UV-
vis (c ) 0.0270 mg/mL, MeOH) λmax 303 (10 183), 287 (14 593),
252 (12 888) nm; MS (NH3-DCI) m/e 675 (M + NH4). Anal.
Calcd for C39H39N5O5, MW 657.77: C, 71.21; H, 5.99; N, 10.65.
Found: C, 70.98; H, 6.09; N, 10.34. Analytical HPLC: Zorbax
ODS 4.6 × 250 mm column, solvent H2O-CH3CN, flow rate
of 1 mL/min, detector at 256 nm.
(4r,5r,6â,7â)-3-[[Hexa h yd r o-5,6-d ih yd r oxy-3-[(3-m eth -
oxyp h en yl)m eth yl]-2-oxo-4,7-bis(p h en ylm eth yl)-1H-1,3-
d ia ze p in -1-yl]m e t h yl]-N -(5-m e t h yl-2-p yr id in yl)b e n z-
a m id e Sesqu ih yd r a te (7). By substituting 5-methyl-2-
aminopyridine in the method for 6, the desired product was
obtained in 74% (0.327 g) or 35% from the half ester: mp 238-
240 °C; 1H NMR (300 MHz, DMSO-d6, TMS) δ 2.30 (s, 3H,
CH3), 2.7-3.09 (m, 6H, Ar′CH2 and NCH), 3.40-3.58 (m, 4H,
CHCHCHCH), 3.70 (s, 3H, OCH3), 4.66 (d, J ) 14.3 Hz, 1H,
NCH), 4.71 (d, J ) 14.3 Hz, 1H, NCH), 5.40 (broad s, 2H, OH),
[6.65 (s, 1H), 6.68 (d, 1H), 6.81 (m, 1H), 6.93 (d, 1H), 7.06 (d,
2H), 7.18-7.33 (m, 7H), 7.38 (d, 1H), 7.45 (dd, 1H), 7.58 (d,
1H), 7.88 (s, 1H), 7.96 (d, 1H), 8.04 (d, 1H), 8.24 (s, 1H), Ar],
10.98 (s, 1H, NH); IR (KBr) 3364 (OH and NH), 1686 (CdO),
1612 (CdO) cm-1; UV-vis (c ) 0.0160 mg/mL, MeOH) λmax
287 (16 603), 263 (14 003), 221 (42 597) nm; MS (NH3-DCI)
The acetonide protecting group was removed in the usual
manner to give the desired product in 62% yield (overall yield
1
) 10% for five steps): mp 230-232 °C; H NMR (300 MHz,
DMSO-d6, TMS) δ 2.27 (s, 3H, CH3), 2.59 (d, J ) 14.3 Hz, 1H,
NCH), 2.76 (dd, 1H, Ar′CH), 2.96 (m, 3H, Ar′CH), 3.05 (d, J )
14.3 Hz, 1H, NCH), 3.50 (m, 4H, CHCHCHCH), 4.58 (d, J )
14.3 Hz, 1H, NCH), 4.71 (d, J ) 14.3 Hz, 1H, NCH), 5.07 (m,
4H, OH and NH2), [6.24 (d, 1H), 6.34 (s, 1H), 6.43 (d, 1H),
6.94 (m, 3H), 7.12 (d, 2H), 7.26 (m, 7H), 7.45 (dd, 1H), 7.64
(dd, 1H), 7.86 (s, 1H), 7.94 (d, 1H), 8.07 (d, 1H), 8.21 (s, 1H,)
Ar], 10.66 (s, 1H, NH); IR (KBr) 3366 (OH and NH), 1678
(CdO), 1606 (CdO) cm-1; UV-vis (c ) 0.014 mg/mL, MeOH)
λ
max 310 (19 130), 254 (16 095) nm; MS (NH3-DCI) m/e 656 (M
m/e 671 (M + 1), 688 (M + 14); [R]20 +80.77° (c ) 0.16,
+ 1); [R]20 +78.75° (c ) 0.08, MeOH). Anal. Calcd for
D
D
MeOH). Anal. Calcd for C41H42N4O5‚1.5H2O, MW 697.84: C,
70.58; H, 6.48; N, 8.03. Found: C, 70.82; H, 6.17; N, 8.01.
(4r,5r,6â,7â)-3-[[Hexa h yd r o-5,6-d ih yd r oxy-3-[(3-m eth -
oxyp h en yl)m eth yl]-2-oxo-4,7-bis(p h en ylm eth yl)-1H-1,3-
d ia ze p in -1-yl]m e t h yl]-N -(6-m e t h yl-2-p yr id in yl)b e n z-
a m id e Hem ih yd r a te (8). By substituting 6-methyl-2-ami-
nopyridine in the method for 6, the desired product was
obtained in 82% (0.146 g) or 28% from the half ester: mp 243-
245 °C; 1H NMR (300 MHz, DMSO-d6, TMS) δ 2.45 (s, 3H,
CH3), 2.7-3.08 (m, 6H, Ar′CH2 and NCH), 3.35-3.58 (m, 4H,
CHCHCHCH), 3.70 (s, 3H, OCH3), 4.66 (d, J ) 14.3 Hz, 1H,
NCH), 4.71 (d, J ) 14.6 Hz, 1H, NCH), 5.13 (broad s, 2H, OH),
[6.65 (1, 1H), 6.68 (d, 1H), 6.83 (m, 1H), 6.93 (d, 2H), 7.00 (d,
1H), 7.06 (d, 2H), 7.2-7.34 (m, 7H), 7.36 (d, 1H), 7.42 (dd, 1H),
7.68 (dd, 1H), 7.88 (s, 1H), 7.94 (d, 1H), 7.98 (d, 1H), Ar], 10.67
(s, 1H, NH); IR (KBr) 3448 (OH), 3310 (NH), 1690 (CdO), 1608
(CdO) cm-1; MS (NH3-DCI) m/e 671 (M + 1), 688 (M + NH4);
C40H41N5O4‚0.5H2O, MW 664.81: C, 72.27; H, 6.37; N, 10.53.
Found: C, 72.57; H, 6.33; N, 10.35.
(4r,5r,6â,7â)-3-[[Hexa h yd r o-5,6-d ih yd r oxy-3-[(3-n itr o-
p h en yl)m eth yl]-2-oxo-4,7-bis(p h en ylm eth yl)-1H-1,3-d ia z-
epin -1-yl]m eth yl]-N-2-pyr idin ylben zam ide dih ydr ate (11).
By substituting 2-aminopyridine in the method for 10, the
desired product was obtained in 42% yield after chromatog-
raphy on silica using CHCl3-MeOH (99:1): mp 235-238 °C;
1H NMR (300 MHz, DMSO-d6, TMS) δ 2.65-2.79 (m, 2H,
Ar′CH), 2.98 (m, 3H, Ar′CH and NCH), 3.51-3.64 (m, 4H,
CHCHCHCH), 4.50 (d, J ) 13.9 Hz, 1H, NCH), 4.67 (d, J )
13.9 Hz, 1H, NCH), 5.18 (s, 2H, OH), 6.85-8.38 (m, 22H, Ar),
10.75 (s, 1H, NH); IR (KBr) 3332 (OH), 1678 (CdO), 1640
(CdO) cm-1; UV-vis (c ) 0.017 mg/mL, MeOH) λmax 284
(16 122), 264 (15 647) nm; MS (NH3-DCI) m/e 672 (M + 1);
[R]20D +77.21° (c ) 0.136, MeOH). Anal. Calcd for C39H37N5O6‚
2H2O, MW 707.79: C, 66.18; H, 5.84; N, 9.89. Found: C,
66.51; H, 5.56; N, 9.61.
[R]20 +97.37° (c
) 0.08, MeOH). Anal. Calcd for
D
C41H42N4O5‚0.5H2O, MW 679.83: C, 72.44; H, 6.38; N, 8.24.
Found: C, 72.76; H, 6.31; N, 8.14.
(4r,5r,6â,7â)-3-[[3-[(3-Am in oph en yl)m eth yl]h exah ydr o-
5,6-d ih yd r oxy-2-oxo-4,7-b is(p h en ylm et h yl)-1H-1,3-d ia z-
ep in -1-yl]m eth yl]-N-(1,1-d im eth yleth yl)ben za m id e (12).
By substituting tert-butylamine in the method used to obtain
10, the desired intermediate nitro amide acetonide was
obtained in 103% (1.10 g) yield: 1H NMR (300 MHz, CDCl3,
(4r,5r,6â,7â)-3-[[Hexa h yd r o-5,6-d ih yd r oxy-3-[(3-m eth -
oxyp h en yl)m eth yl]-2-oxo-4,7-bis(p h en ylm eth yl)-1H-1,3-
d ia zep in -1-yl]m eth yl]-N-2-p yr id in ylben za m id e Hem ih y-
d r a te (9). By substituting 2-aminopyridine in the method for
6, the desired product was obtained in 59% yield from the half
t
TMS) δ 1.41 (m, 15H, Bu and iPr), 2.75-3.1 (m, 4H, ArCH2),
1
ester: mp 223-225 °C; H NMR (300 MHz, DMSO-d6, TMS)
3.16 (d, J ) 14.3 Hz, 1H, NCH), 3.40 (d, J ) 14.6 Hz, 1H,
NCH), 3.8 (m, 2H, ArCCH), 3.09-4.05 (m, 2H, OCH), 4.80 (d,
J ) 14.6 Hz, 1H, NCH), 4.89 (d, J ) 14.3 Hz, 1H, NCH), 5.85
(s, 1H, NH), 6.95-8.1 (m, 18H, Ar); MS (NH3-DCI) m/e 708
(M + NH4) for C41H46N4O6, MW 690.85.
δ 2.7-3.1 (m, 6H, Ar′CH2 and NCH), 3.4-3.65 (m, 4H,
CHCHCHCH), 3.70 (s, 3H, OCH3), 4.67 (d, J ) 14.7 Hz, 1H,
NCH), 4.72 (d, J ) 14.6 Hz, 1H, NCH), 5.12 (m, 2H, OH), [6.65
(s, 1H), 6.69 (d, 1H), 6.83 (m, 1H), 6.96 (d, 2H), 7.08 (d, 2H),

Cyclic Urea HIV Protease Inhibitors
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 25 4087
The above intermediate (1.00 g, 1.54 mmol) in 25 mL of
tBuOH and ammonium formate (2 g, mmol) and 10% Pd/c (0.2
g) was stirred at room temperature until no starting material
remained as evidenced by TLC (CHCl3-EtOAc, 3:2). The
reaction mixture was filtered through Celite to remove the
catalyst. The Celite bed was washed with 3 × 50 mL of EtOAc,
and the filtrate and washings were combined, washed with
water and brine, dried over MgSO4, filtered, and concentrated
to a foam. The foam was dissolved in 20 mL of acetonitrile,
treated with 10 mL of 1 N HCl, stirred for 16 h, and
concentrated to a foam. The foam was dissolved in 25 mL of
EtOH and precipitated with 100 mL of 5% NaHCO3. The
resulting precipitate was collected by filtration, washed with
water, and dried in vacuo at 80 °C to give the desired product
in 82% (0.780 g) yield from the nitro ester acetonide: mp 102-
105 °C dec; 1H NMR (300 MHz, DMSO-d6, TMS) δ 1.32 (s, 9H,
tBu), 2.63 (d, J ) 14.3 Hz, 2H, NCH), 2.8 (m, 2H, Ar′CH), 2.95
(m, 4H, Ar′CH and NCH), 3.4 (m, 4H, CHCCHCHCH), 4.61
(d, J ) 14.3 Hz, 1H, NCH), 4.75 (d, J ) 13.9 Hz, 1H, NCH),
5.08 (broad s, 2H, OH), 6.24 (d, 1H, NH), 6.36 (s, 1H, CONH),
6.45 (d, 1H, NH), 6.9-7.71 (m, 18H, Ar); IR (KBr) 3362 (OH
and NH), 1642 (CdO) cm-1; UV-vis (c ) 0.014 mg/mL, MeOH)
(4R,5R,6â,7â)-N-[3-[[3-[[3-[[(Cya n om eth yl)a m in o]ca r bo-
n yl]p h en yl]m eth yl]h exa h yd r o-5,6-d ih yd r oxy-2-oxo-4,7-
bis(p h en ylm eth yl)-1H-1,3-d ia zep in -1-yl]m eth yl]ben zoyl]-
glycin e (16). The ester Ia was saponified to give the bisacid
acetonide (0.635 g, 1.0 mmol) which was dissolved in 25 mL
of CH2Cl2 and treated with oxalyl chloride (0.38 g, 3.0 mmol)
and 1 drop of DMF. The mixture was stirred for 30 min. The
mixture was treated with aminoacetonitrile hydrochloride
(0.278 g, 3.0 mmol), stirred for an additional 10 min, and
treated with diisopropylethylamine (0.78 g, 6.0 mmol). The
mixture was stirred at room temperature for 16 h and diluted
with 100 mL of water and 50 mL of CH2Cl2. The organic layer
was washed with additional water, 5% citric acid, water, and
brine; dried over MgSO4; filtered; and concentrated to a foam
of constant weight: 1H NMR (300 MHz, DMSO-d6, TMS) δ
2.70 (dd, 2H, Ar′CH), 2.84 (m, 2H, Ar′CH), 3.25 (d, J ) 14.0
Hz, 2H, NCH), 4.02 (d, J ) 11.0, 2H, Ar′CCH), 4.07 (s, 2H,
OCH), 4.31 (d, J ) 5.5 Hz, 4H, CH2-CN), 4.52 (d, J ) 14.0 Hz,
2H, NCH), [6.93 (m, 4H), 7.2 (m, 6H), 7.44 (m, 4H), 7.77 (m,
4H), Ar and Ar′], 9.20 (dd, 2H, NH).
The acetonide was dissolved in 10 mL of acetonitrile and
treated with 10 mL of 10.0 N HCl (accidentally used instead
of 1.0 N). The mixture was stirred at room temperature until
no starting acetonide remained. The mixture was diluted with
100 mL of water and triturated. The resulting solid was
collected by filtration, washed with water, and dried in vacuo.
Recrystallization from EtOAc produced 0.201 g of white solid
which was identified by analytical methods to be the “half-
λ
max 284 (3059) nm; MS (NH3-DCI) m/e 621 (M + 1), 638 (M +
NH4); [R]20 +76.06° (c ) 0.14, MeOH). Anal. Calcd for
D
C38H44N4O4, MW 620.80: C, 73.52; H, 7.14; N, 9.02. Found:
C, 73.39; H, 7.16; N, 8.99.
[4R-(4r,5r,6â,7â)]-3-[[3-[(3-Am in oph en yl)m eth yl]h exah y-
d r o-5,6-d ih yd r oxy-2-oxo-4,7-bis(p h en ylm eth yl)-1H-1,3-d i-
a zep in -1-yl]m eth yl]-N-2-p yr a zin ylben za m id e (13). By
substituting 2-aminopyrazine in the procedure used to syn-
thesize 10, the desired product was obtained in an overall yield
of 13% from the isourea: mp 154-157 °C; 1H NMR (300 MHz,
DMSO-d6, TMS) δ 2.58 (d, 2H, Ar′CH), 2.73 (dd, 2H, Ar′CH),
2.93 (d, J ) 13.9 Hz, 1H, NCH), 3.05 (d, J ) 14.3 Hz, 1H,
NCH), 3.4-3.6 (m, 4H, CHCHCHCH), 4.56 (d, J ) 13.9 Hz,
1H, NCH), 4.69 (d, J ) 14.3 Hz, 1H, NCH), 5.09 (d, 2H, OH),
6.4-8.5 (m, 21H, Ar), 9.38 (s, 2H, NH2), 11.12 (s, 1H, NH); IR
(KBr) 3360 (NH and OH), 1682 (CdO), 1606 (CdO) cm-1; MS
(NH3-DCI) m/e calcd for C38H39N6O4: 643.303 279 (M + 1),
found 643.303 224, 690 (M + 1). Anal. Calcd for C38H38N6O4,
MW 642.76: C, 71.01; H, 5.96; N, 13.07. Found: C, 70.92; H,
6.00; N, 12.98.
1
acid, half-nitrile, diol”: mp 154-157 °C; H NMR (300 MHz,
DMSO-d6, TMS) δ 2.7 (m, 2H, Ar′CH), 2.9 (m, 4H, Ar′CH and
NCH), 3.5 (m, 4H, CHCHCHCH), 3.80 (d, J ) 5.9 Hz, 2H,
CHCO2), 4.30 (d, J ) 5.2 Hz, 2H, CHCN), 4.6 (m, 2H, NCH),
5.15 (s, 2H, OH), [6.9 (m, 4H), 7.2 (m, 6H), 7.4 (m, 4H), 7.75
(m, 4H), Ar], 8.63 (t, 1H, HNCCO2), 9.21 (t, 1H, HN-C-CN);
IR (KBr) 3390 (OH and NH), 2254 (CN), 1650 (CdO) cm-1
;
MS (NH3-DCI) m/e 690 (M + 1), 707 (M + NH4). Anal. Calcd
for C39H39N5O7, MW 689.77: C, 67.91; H, 5.70; N, 10.15.
Found: C, 67.74; H, 5.85; N, 10.52.
Ack n ow led gm en t. We thank Gilbert N. Lam and
Dani Timby for providing the pharmacokinetic data for
this study; Ronald Klabe for providing the HIV protease
inhibition data; and Beverly Cordova for providing the
viral replication inhibition data.
[4R-(4r,5r,6â,7â)]-3-[[3-[(3-Am in oph en yl)m eth yl]h exah y-
d r o-5,6-d ih yd r oxy-2-oxo-4,7-bis(p h en ylm eth yl)-1H-1,3-d i-
a zep in -1-yl]m et h yl]-N-1H -b en zim id a zol-2-ylb en za m id e
(14). By substituting 2-aminobenzimidazole in the procedure
for 10, the desired product was obtained in an overall yield of
39%: mp 134-135 °C dec; 1H NMR (300 MHz, DMSO-d6,
TMS) δ 2.59 (d, 1H, Ar′CH), 2.82 (dd, 1H, Ar′CH), 2.9-3.1 (m,
4H, Ar′CH and NCH), 3.5 (m, 4H, CHCHCHCH), 4.58 (d, J )
13.9 Hz, 1H, NCH), 4.72 (d, J ) 1, 14.3 Hz, 1H, NCH), 5.03
(m, 4H, OH and NH2), [6.22 (d, 1H), 6.34 (s, 1H), 6.42 (d, 1H),
6.9 (m, 3H), 7.13 (m, 3H), 7.20 (m, 2H), 7.23 (m, 4H), 7.35 (m,
3H), 7.44 (m, 3H), 8.05 (d, 1H), Ar], 8.00 (s, 1H, NH), 12.23
(broad s, 1 NH); IR (KBr) 3368(NH and OH), 1630 (CdO), 1606
(CdO) cm-1; MS (NH3-DCI) m/e calcd for C41H41N6O4: high-
resolution MS 681.317 592, found 681.317 126, 681 (M + 1).
Anal. Calcd for C41H40N6O4, MW 680.81: C, 72.33; H, 5.92;
N, 12.34. Found: C, 71.27; H, 6.01; N, 12.29.
Refer en ces
(1) Basha, A.; Lipton, M.; Weinreb, S. M. A Mild, General Method
for Conversion of Esters to Amides. Tetrahedron Lett. 1977,
4171-4174.
(2) Wilkerson, W. W.; Hollis, A. Y.; Cheatham, W. W.; Lam, G. N.;
Erickson-Viitanen, S.; Bacheler, L.; Cordova, B. C.; Klabe, R.
M.; Meek, J . L. A Bis-[N-3-(1-Hydroxy-1-methyl-ethyl)-benzyl]-
Cyclic Urea As A HIV Protease Inhibitor. Bioorg. Med. Chem.
Lett. 1995, 5, 3027-3032.
(3) Cheng, Y.-S. E.; Yin, F. H.; Foundling, S.; Blomstrom, D.;
Kettner, C. A. Stability and Activity of Human Immunodefi-
ciency Virus Protease: Comparison of the Natural Dimer with
a Homologous, Single-chain Tethered Dimer. Proc. Natl. Acad.
Sci. U.S.A. 1990, 87, 9660-9664.
(4) Erickson-Viitanen, S.; Manfredi, J .; Viitanen, P.; Tribe, D. E.;
Tritch, R.; Hutchison, C. A., III; Loeb, D. D.; Swanstrom, R.
Cleavage of HIV-1 gag Polyprotein Synthesized In Vitro: Se-
quential Cleavage by Viral Protease. AIDS Res. Hum. Retrovi-
ruses 1989, 5, 577-591.
(5) Bacheler, L.; Paul, M.; Otto, M. J .; J adhav, P. K.; Stone, B. A.;
Miller, J . A. An Assay for HIV RNA in Infected Cell Lysates,
and Its Use for the Rapid Evaluation of Antiviral Efficacy.
Antiviral Chem. Chemother. 1994, 5, 111-121.
(6) Wong, Y. N.; Burcham, D. L.; Saxton, P. L.; Erickson-Viitanen,
S.; Grubb, M. F.; Quon, C. Y.; Huang, S.-M. A Pharmacokinetic
Evaluation of HIV Protease Inhibitors, Cyclic Ureas, in Rats and
Dogs. Biopharm. Drug Dispos. 1994, 15, 535-544.
(7) Grubb, M. F.; Wong, Y. N.; Burcham, D. L.; Saxton, P. L.; Quon,
C. Y.; Huang, S.-M. Pharmacokinetics of HIV Protease Inhibitor
DMP 323 in Rats and Dogs. Drug Met. Disp. 1994, 22, 709-
712.
(8) Hansch, C.; Leo, A. Exploring QSAR. Fundamentals and Ap-
plications in Chemistry and Biology; American Chemical Soci-
ety: Washington, 1995.
(9) Lien, E. J . In Medicinal Research; Grunewald, G. L., Ed.; Marcel
Dekker: New York, 1987; Vol. 11, pp 92-102.
[4R-(4r,5r,6â,7â)]-3-[[3-[(3-Am in oph en yl)m eth yl]h exah y-
d r o-5,6-d ih yd r oxy-2-oxo-4,7-bis(p h en ylm eth yl)-1H-1,3-d i-
a zep in -1-yl]m eth yl]-N-1H-im id a zol-2-ylben za m id e (15).
By substituting 2-aminoimidazole in the method for 10, the
desired product was obtained in 26% overall yield: mp 166
°C dec; 1H NMR (300 MHz, DMSO-d6, TMS) δ 2.58 (d, J )
14.3 Hz, 2H, Ar′CH), 2.8-3.03 (m, 4H, Ar′CH and NCH), 3.44-
3.55 (m, 4H, Ar′CCHCH), 4.58 (d, J ) 13.9 Hz, 1H, NCH),
4.70 (d, J ) 13.9 Hz, 1H, NCH), 5.06 (s, 2H, OH), [6.24 (m,
1H), 6.33 (s, 1H), 6.43 (d, 1H), 6.79 (s, 2H), 6.86-6.96 (m, 4H),
7.10-7.35 (m, 10H), 7.43 (dd, 1H), 7.92 (s, 1H), 7.97 (d, 1H),
Ar], 11.75 (s, 2H, NH); IR (KBr) 3376 (OH and NH), 1604
(CdO) cm-1; UV-vis (MeOH) λmax 285 (13 862) nm; MS (NH˙ -
3
DCI) m/e 631 (M + 1); [R]20D +79.81° (c ) 0.104, MeOH). Anal.
Calcd for C37H38N6O4, MW 630.75: C, 70.46; H, 6.07; N, 13.32.
Found: C, 70.39; H, 6.33; N, 13.28.

4088 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 25
Wilkerson et al.
(10) The log PHPLC and CLOP values for compounds 2-16 were
subjected to regression analysis to produce the following rela-
tionship:
(14) Rayner, M. M.; Cordova, B. C.; Meade, R. P.; Aldrich, P. E.;
J adhav, P. K.; Ru, Y.; Lam, P. Y. DMP 323, A Nonpeptide Cyclic
Urea Inhibitor of Human Immunodeficiency Virus (HIV) Pro-
tease, Specifically and Persistently Blocks Intracellular Process-
ing of HIV Gag Polyprotein. Antimicrob. Agents Chemother.
1994, 38, 1635-1640.
CLOGP ) 1.236((0.034) log PHPLC + 0.294((0.168)
(15) Pierce, M. E.; Harris, G. D.; Islam, Q.; Radesca, L. A.; Storace,
L.; Waltermire, R. E.; Wat, E.; J adhav, P. K.; Emmett, G. C.
Stereoselective Synthesis of HIV-1 Protease Inhibitor DMP 323.
J . Org. Chem. 1996, 61, 444-450.
(16) Kan, R. O. A Correlation of Chemical Shifts with Inductive Effect
Parameters. J . Am. Chem. Soc. 1964, 86, 5180-5183.
(17) Perrin, D. D.; Dempsey, B.; Serjeant, E. P. pKa Prediction for
Organic Acids and Bases; Chapman and Hall: New York, 1981.
(18) Topsom, R. D. In Prog. Phys. Org. Chem.; Taft, R. W., Ed.; J ohn
Wiley & Sons: New York, 1987; Vol. 16, pp 125-191.
(19) Mitchell, T. J .; Tute, M. S.; Webb, G. A. A Theoretical Study of
the pKa Values of Some Clonidine-like Imidazolidines. Eur. J .
Med. Chem. 1990, 25, 117.
n ) 15
r2 ) 0.990
s ) 0.114
(11) Wilkerson, W. W.; Hollis, A. Y.; Cheatham, W. W.; Akamike ,
E.; Collins, R. D.; DeLucca, I.; J adhav, P. K.; Lam, P. Y. S.; Ru,
Y. HIV Protease Inhibitory bis-Benzamides Cyclic Ureas:
QSAR Analysis. J . Med. Chem. 1996, 39, 4299-4312.
A
(12) 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.; Sharp,
T. R.; Erickson-Viitanen, S. Rational Design of Potent, Bioavail-
able, Nonpeptide Cyclic Ureas as HIV Protease Inhibitors.
Science 1994, 263, 380-284.
(20) Brinck, T.; Murray, J . S.; Politzer, P.; Carter, R. E. A Relation-
ship between Experimentally Determined pKas and Molecular
Surface Ionization Energies for Some Azines and Azoles. J . Org.
Chem. 1991, 56, 2934-2936.
(13) Otto, M. J .; Reid, C. D.; Garber, S.; Lam, P. Y.-S.; Scarnati, H.;
Bacheler, L. T.; Rayner, M. M.; Winslow, D. L. In Vitro Anti-
Human Immunodeficiency Virus (HIV) Activity of XM323, A
Novel HIV Protease Inhibitor. Antimicrob. Agents Chemother.
1993, 37.
J M970288B