Design, synthesis, and evaluation of tetrahydropyrimidinones as an example of a general approach to nonpeptide HIV protease inhibitors

Article abstract of DOI:10.1021/jm970081i

Re-examination of the design of the cyclic urea class of HIV protease (HIVPR) inhibitors suggests a general approach to designing novel nonpeptide cyclic HIVPR inhibitors. This process involves the inversion of the stereochemical centers of the core transition-state isostere of the linear HIVPR inhibitors and cyclization of the resulting core using an appropriate cyclizing reagent. As an example, this process is applied to the diamino alcohol class of HIVPR inhibitors to give tetrahydropyrimidinones. Conformational analysis of the tetrahydropyrimidinones and modeling of its interaction with the active site of HIVPR suggested modifications which led to very potent inhibitors of HIVPR (24 with a K(i) = 0.018 nM). The X-ray crystallographic structure of the complex of 24 with HIVPR confirms the analysis and modeling predictions. The example reported in this study and other examples that are cited indicate that this process may be generally applicable to other linear inhibitors.

Full text of DOI:10.1021/jm970081i

J . Med. Chem. 1997, 40, 1707-1719
1707
Design , Syn th esis, a n d Eva lu a tion of Tetr a h yd r op yr im id in on es a s a n Exa m p le
of a Gen er a l Ap p r oa ch to Non p ep tid e HIV P r otea se In h ibitor s
George V. De Lucca,* J ing Liang, Paul E. Aldrich, J oe Calabrese,1a Beverly Cordova, Ronald M. Klabe,
Marlene M. Rayner, and Chong-Hwan Chang1b
DuPont Merck Pharmaceutical Company, Experimental Station, P.O. Box 80500, Wilmington, Delaware 19880-0500
Received February 6, 1997X
Re-examination of the design of the cyclic urea class of HIV protease (HIVPR) inhibitors suggests
a general approach to designing novel nonpeptide cyclic HIVPR inhibitors. This process involves
the inversion of the stereochemical centers of the core transition-state isostere of the linear
HIVPR inhibitors and cyclization of the resulting core using an appropriate cyclizing reagent.
As an example, this process is applied to the diamino alcohol class of HIVPR inhibitors11 to
give tetrahydropyrimidinones. Conformational analysis of the tetrahydropyrimidinones and
modeling of its interaction with the active site of HIVPR suggested modifications which led to
very potent inhibitors of HIVPR (24 with a Ki ) 0.018 nM). The X-ray crystallographic structure
of the complex of 24 with HIVPR confirms the analysis and modeling predictions. The example
reported in this study and other examples that are cited indicate that this process may be
generally applicable to other linear inhibitors.
In tr od u ction
Since the identification of the human immunodefi-
ciency virus (HIV) as the causative agent of acquired
immune deficiency syndrome (AIDS), there has been an
intense worldwide search to find useful chemothera-
peutic agents. One of the prime targets of this research
has been the essential aspartic protease (PR)2 that
processes the viral gag and gag-pol polyproteins into
structural and functional proteins. Inhibition of HIVPR
in vitro results in the production of progeny virions
which are immature and noninfectious.3 With a wealth
of structural information available, HIVPR has been an
attractive target for computer-aided drug design strate-
F igu r e 1. General hydrogen-bonding scheme found with a
typical peptidomimetic inhibitor binding at the HIVPR active
site via a bridging structural water.
gies,4 and this effort has produced a variety of potent
inhibitors.5
In recent clinical studies, several HIVPR inhibitors
low molecular weight scaffold which is highly preorga-
nized and complementary to the HIVPR active site and
optimally directs substituents into the corresponding
enzyme subsites. This work resulted in the identifi-
cation of two clinical candidates from this series:
DMP3238,9 and DMP450.10
have been shown to reduce the viral load and increase
the number of CD4+ lymphocytes in HIV-infected
patients.6 Saquinavir (Roche), Ritonavir (Abbott), and
Indinavir (Merck) have recently been approved by the
FDA and are being used in AIDS therapy in combina-
tion with reverse transcriptase inhibitors. However, the
ability of the virus to rapidly generate resistant mu-
tants7 suggests that there is an ongoing need for new,
structurally diverse HIVPR inhibitors with superior
potency, pharmacokinetics, and efficacy.
The DuPont Merck group recently described the
rational design of a class of novel and highly potent
cyclic urea inhibitors.8 The 7-membered ring cyclic
ureas were designed8a by first using the structure-
activity relationship (SAR) of the linear diamine diol
series of HIVPR inhibitors to define a 3-D pharmacoph-
ore model. Using this model a 3-D database was
searched for possible lead structures. The hits from this
search were modified using computer-aided drug design
tools to arrive at the 7-membered ring cyclic ureas. The
urea oxygen was designed to displace the unique
structural water molecule usually found in the linear
HIVPR inhibitor complexes (Figure 1). The 7-mem-
bered ring cyclic urea is also an excellent nonpeptidic,
Discovery of the 7-membered ring cyclic ureas promp-
ted our interest in designing other heterocyclic cores
that might also make good scaffolds for HIVPR inhibi-
tors. Thus, it is important to understand the structural
features that make the cyclic ureas such potent inhibi-
tors so that these features can be incorporated into the
design of any new heterocyclic core. These features can
be best exemplified by analyzing the X-ray structure of
X Abstract published in Advance ACS Abstracts, May 1, 1997.
S0022-2623(97)00081-2 CCC: $14.00 © 1997 American Chemical Society

1708 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11
De Lucca et al.
F igu r e 2. X-ray structure of the complex of DMP450 with HIV-1PR showing the hydrogen-bonding interactions. The urea oxygen
accepts hydrogen bonds from the protease flaps Ile50/50′ with the exclusion of the intervening structural water commonly found
in linear peptidomimetic inhibitors.
DMP450 bound to HIVPR as shown in Figure 2. Most
of the enzyme residues have been removed to clarify and
simplify the analysis. The most obvious feature is the
network of hydrogen bonds between the flap and the
urea carbonyl and between the diol and the catalytic
aspartic acids. In the view looking down from the flaps
through the urea carbonyl (Figure 3), one can see that
another important feature is the conformation of the
7-membered ring that places the trans diaxial P1 benzyl
groups in good contact with the lipophilic S1 pocket. The
N-benzyl group serves two important functions. First,
it contributes an important hydrophobic interaction with
the lipophilic S2 enzyme pocket. Second, it provides a
rigid framework for directing groups from the meta and
para positions toward the S2/S3 subsites in the enzyme
where there are several polar residues which are a rich
source of hydrogen-bonding possibilities, in particular
Asp29, Asp30, and Gly48. Thus the 7-membered ring
cyclic urea provides a conformationally constrained
scaffolding which properly places all the important
binding elements of the inhibitor in the corresponding
recognition domains of the enzyme.
bered ring cyclic ureas were “simply” a cyclic version of
the potent linear diamine diol11 class of HIVPR inhibi-
tors. In practice the sequence of events needed to get
to the cyclic ureas starting from the linear diamine diol
inhibitors is (1) invert all the stereocenters of the
transition-state isostere core and (2) cyclize to give the
cyclic ureas (Figure 4). In order to explore the general-
ity and the scope and limitations of this process, we
explored the possibility of cyclizing other known linear
HIVPR inhibitors. We have found that this process is
generally applicable. However, it is important, in this
process of cyclizing linear inhibitors as outlined in
Figure 4, not only to get the correct stereochemistry but
also to understand the conformation of the resulting ring
system and its interaction with the enzyme.
The first example that we wish to report in detail in
this paper is the application of this process to the closely
related linear diamino alcohol class of HIVPR inhibitors
that were first developed by Abbott.11 Inversion of the
stereocenters and cyclization give the corresponding
tetrahydropyrimidinone, as shown in Figure 5. Under-
standing the conformation of the ring is very important
in order to properly analyze its interaction with the
enzyme in the active site. In the 6-membered ring, with
trans-1,3-benzyl/benzyl P1/P1′ substituents, both benzyl
groups can not be axialsif one is axial then the other
has to be equatorial. Thus, the equatorial benzyl group
in the 6-membered ring heterocycle can not interact
with the enzyme in the same way that a benzyl group
on the 7-membered ring cyclic urea can interact with
the enzyme, since in the 7-membered ring cyclic ureas
both P1 benzyl groups are axial (see Figure 3).
Resu lts a n d Discu ssion
Design of Tetr a h yd r op yr im id in on es. After the
7-membered ring cyclic ureas were found to be potent
inhibitors of HIV-1PR, we became interested in finding
a systematic approach to discovering other nonpeptide
scaffoldings which might also produce potent inhibitors.
As shown in Figure 4, a backward analysis of the
hypothetical design steps necessary to give the cyclic
ureas (“retro-design analysis”) revealed that the 7-mem-

General Approach to Nonpeptide HIVPr Inhibitors
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11 1709
F igu r e 3. X-ray structure of the complex of DMP450 with HIV-1PR showing the important trans diaxial conformation of the
P1/P1′ benzyl groups. The figure also shows the interactions of the N-benzyl substituent with the S2 binding pocket.
Sch em e 1a
a
Key: (a) isobutyl chloroformate/THF/0 °C/N,O-dimethylhydroxylamine; (b) vinylmagnesium bromide; (c) NaBH4/CeCl3; (d) MsCl; (e)
BnMgCl/CuCN; (f) MCPBA/CH2Cl2; (g) NH4Cl/NaN3/DMF; (h) MEM-Cl; (i) H2, Pd/C; (j) CDI/THF.
However, if one changes the equatorial P1′ benzyl
group into a phenethyl group, then, although this
substituent is still equatorial, the benzyl portion can
take an axial-like position. The phenethyl group can
then interact with the enzyme in the same way that the
benzyl group on the 7-membered ring cyclic urea
interacts. This becomes apparent when computer mod-
els are examined as shown in Figure 6. In green is the
7-membered cyclic urea DMP450 in its bound conforma-
tion, and overlapped in white is a model of the P1/P1′
benzyl/benzyl (Figure 6A) and P1/P1′ benzyl/phenethyl
tetrahydropyrimidinones (Figure 6B). If the 7-mem-
bered ring cyclic urea serves as a template of an
optimally complementary scaffolding for HIVPR, then
these modeling studies suggested that in the tetrahy-
dropyrimidinone, the P1′ phenethyl analog should be a
better inhibitor than the P1′ benzyl analog. In order
to test these design and modeling predictions, the
synthesis of tetrahydropyrimidinones was undertaken.
Syn t h esis of Tet r a h yd r op yr im id in on es. The
methods used to synthesize the desired analogs are
summarized in Schemes 1 and 2. The synthesis of the
tetrahydropyrimidinone is achieved by modification13 of
the procedure developed by researchers at Abbott11 for
the synthesis of the diamine alcohol isostere core with
defined stereochemistry at the three asymmetric cen-
ters. However, since we desire the opposite stereo-
chemistry, we start with the enantiomeric Cbz-D-
phenylalanine as summarized in Scheme 1. The
unnatural amino acid is converted to the corresponding
Weinreb amide12 and then treated with vinyl Grignard
to give the vinyl ketone. This is reduced with NaBH4
to give the allylic alcohol 1 in 70% yield over the three
steps. The alcohol is converted to the mesylate and then

1710 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11
De Lucca et al.
F igu r e 4. Retro-design analysis of the steps which are needed
to convert to the cyclic HIVPR inhibitor from the linear HIVPR
inhibitor.
F igu r e 5. Outline of steps needed to convert the linear
diamine alcohol class of HIVPR inhibitors to the tetrahydro-
pyrimidinone class of HIVPR inhibitors.
Sch em e 2a
DMF8 and appropriate alkylating agent (Scheme 2).
Although it is a little long, the synthesis provided
sufficient material of defined stereochemistry to enable
the validation of the design ideas.
While this synthesis was progressing we discovered
that the 7-membered ring cyclic urea has a tendency to
undergo a novel rearrangement ring-contraction reac-
tion to give the desired tetrahydropyrimidinone. The
scope and details of this rearrangement will be pub-
lished elsewhere. However, we were able to take
advantage of this tendency to design an efficient syn-
thesis of the desired tetrahydropyrimidinones as shown
in Scheme 2. For example, when the diol 8 is treated
with 2-acetoxyisobutyryl bromide,15 a nearly quantita-
tive yield of the 6-membered ring bromo acetate 7 is
obtained. The bromide of 7 is reduced with zinc dust/
acetic acid and the acetate hydrolyzed to give the
desired corresponding alcohol 6 in an overall 90% yield,
identical in every way to that obtained from the tet-
rahydropyrimidinone 5 synthesized from Cbz-D-phen-
ylalanine (Scheme 1). This rearrangement is a very
efficient method to synthesize the tetrahydropyrimidi-
none analogs starting from the symmetrically N-sub-
stituted 7-membered ring cyclic ureas. This allowed us
to fully explore the SAR of this series of compounds and
was the method used for the synthesis of all the
compounds reported in this paper.
a
Key: (a) NaH/DMF/4-CNBnBr; (b) HCl/MeOH; (c) 2-acetoxy-
isobutyryl bromide/CH2Cl2; (d) Zn/AcOH; (e) NaOH/MeOH.
treated with the cuprate derived from benzyl Grignard
and copper cyanide to give the olefin 2 in 40% yield over
two steps. The olefin is converted to the epoxide and
opened using sodium azide to give the azido alcohol 3
as a major isomer (4:1 by NMR)14 in 58% yield over two
steps. The alcohol is protected as the MEM ether and
then converted to the diamine by catalytic hydrogena-
tion. The diamine is cyclized with CDI to give the
desired tetrahydropyrimidinone 5. The urea nitrogens
are then alkylated using standard conditions of NaH/

General Approach to Nonpeptide HIVPr Inhibitors
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11 1711
F igu r e 6. Bound conformation of DMP450 (green), obtained from X-ray analysis of the complex with HIVPR, overlapped (A)
with a model of the tetrahydropyrimidinone inhibitor having a P1′ benzyl substituent (white) and (B) with a model of the
tetrahydropyrimidinone inhibitor having a P1′ phenethyl substituent (white).
cyclic ureas, only the corresponding tetrahydropyrimi-
dinones with RRR configuration are active. In each case
shown the RRR isomer is at least 800 times more potent
than corresponding SSS isomer. The SSS isomers were
obtained from the corresponding enantiomeric SRRS
7-membered ring cyclic urea.
Table 2 summarizes the effect of N-substitution on
the activity of the tetrahydropyrimidinones that were
synthesized in this study. The parent N,N-unsubsti-
tuted compound 17 is a weak (Ki ) 1 µM) inhibitor of
the enzyme. However, even small N-substituents that
can interact with the lipophilic S2 enzyme sites greatly
increase activity. The cyclopropylmethyl analog 18 is
a very potent inhibitor with a Ki of 3.8 nM. The benzyl
analog 19 shows excellent potency (Ki 15 nM) making
it a good skeleton for further analoging. Substituting
the meta position of the N-benzyl ring of 19 with
hydrogen bond donor/acceptor groups gives further
increases in potency. The addition of one heteroatom
capable of donating/accepting a hydrogen bond gives a
50-fold increase in potency (for the phenol 11 with a Ki
of 0.30 nM). The amide 23 with two heteroatoms
capable of donating/accepting a hydrogen bond gives a
further increase in potency with a Ki of 0.088 nM.
Finally, the amidoxime 24 with more opportunities for
hydrogen bonding has a Ki of 0.018 nM.
F igu r e 7.
This rearrangement can be envisioned as proceeding
first through the acetoxonium ion I (Figure 7), which
has been proposed as an intermediate in the reaction
of diols with acetoxyisobutyryl halides.15 The urea
nitrogen then participates to give the aziridinium cat-
ionic intermediate II. Attack by bromide ion opens the
aziridine ring to give the observed bromo acetate 9
(Figure 7) as the only product in quantitative yield. The
structure and stereochemistry of the bromo acetate
intermediate 9 were confirmed by X-ray analysis (see
Supporting Information).
The crystal structure of the amidoxime 24 complexed
with HIVPR was solved and is shown in Figure 8, top.
The binding mode is, as expected, very similar to that
found with the 7-membered ring cyclic ureas. There is
a network of hydrogen bonds between the urea carbonyl
and the flap residues Ile50/50′ and between the alcohol
and the catalytic Asp25/25′. Analysis of the X-ray
structure shows that in order to maximize the S1/S1′
interactions, the alcohol of the tetrahydropyrimidinone
binds asymmetrically and is displaced closer to one of
the aspartates as shown in Figure 8. The amidoxime
functionality also has hydrogen-bonding interactions
Eva lu a tion of Tetr a h yd r op yr im id in on es. As the
modeling studies suggested, the SAR of the tetrahydro-
pyrimidinone closely parallels the SAR found for the
7-membered ring cyclic ureas. As it is with the cyclic
ureas,8 the stereochemistry of the tetrahydropyrimidi-
nones is of critical importance for activity. Table 1
summarizes the effect of stereochemistry on the HIVPR
inhibitory activity. Analogous to the 7-membered ring

1712 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11
De Lucca et al.
Ta ble 1
K
K
i
i
K
K
i
i
K
K
i
i
i
i
K
K
K
K
i
i
Ta ble 2
Ch a r t 1
compd
R
Ki (nM)
IC90 (nM)
K
K
i
i
17
18
19
6
H
970
3.8
15
11
1.5
>5000
5200
2800
1600
>5000
115
cyclopropylmethyl
benzyl
3-cyanobenzyl
3-(carbomethoxy)benzyl
3-hydroxybenzyl
3-(hydroxymethyl)benzyl
3-amino-4-fluorobenzyl
3-benzamido
20
11
21
22
23
24
25
26
27
28
29
30
0.3
0.49
0.25
0.088
0.018
1.7
4.9
1.4
8.0
2.5
109
36
35
49
365
328
490
540
3-benzamido oxime
3-aminobenzyl
K
i
3-(N-methylamino)benzyl
4-fluoro-3-benzamido
4-amino-3-fluorobenzyl
4-(hydroxymethyl)benzyl
4-carbomethoxybenzyl
3600
>5000
1970
The importance of the phenethyl group for the activity
is further exemplified in Chart 1. Whereas the P1′
phenethyl tetrahydropyrimidinone 11 is nearly equipo-
tent to the analogous 7-membered ring cyclic urea 10,8a
the P1/P1′ benzyl/benzyl analog 12 is about 100 times
less active than the P1/P1′ benzyl/phenethyl analog 11.
If, in addition, the wrong stereochemistry is used, as
exemplified by the tetrahydropyrimidinone 13,16 the
potency is reduced even further.
Although in general the SAR of tetrahydropyrimidi-
nones closely parallels that of the 7-membered ring
cyclic ureas, there are some noticeable differences. In
every case the Ki of the tetrahydropyrimidinone analog
with Asp30 and Gly48. The X-ray structure suggests
that the amidoxime binds in both possible tautomer
forms as shown in Figure 8, bottom. This ability to form
many hydrogen-bonding tautomer forms probably ac-
counts for the great potency of 24. Superposition of the
bound conformation of the tetrahydropyrimidinone 24
with the bound conformation of the 7-membered ring
cyclic urea DMP450 (Figure 9) shows the excellent
overlap of all four phenyl rings and confirms the initial
design considerations.

General Approach to Nonpeptide HIVPr Inhibitors
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11 1713
F igu r e 8. (Top) X-ray structure of the complex of 24 with HIV-1PR showing the hydrogen-bonding interactions. The urea oxygen
accepts hydrogen bonds from the protease flaps Ile50/50′ with the exclusion of the intervening structural water molecule. The
hydroxyl group is asymmetrically bound to the two catalytic aspartate residues. (Bottom) Hydrogen-bonding scheme of 24 with
HIV-1PR with distance shown in Å. Note that the amidoxime substituent seems to bind in two different tautomeric forms based
on the distance to the Gly48 carbonyl.
is always 2-5-fold weaker than that of the correspond-
ing 7-membered cyclic urea with the same (P2) substit-
uents. This may be due to the fact that in order for the
phenethyl side chain of the tetrahydropyrimidinone to
interact optimally with the enzyme it has to be in a
conformation in which there is an eclipsing interaction
that is nearly anticlinal and not in the energy minimum
antiperiplanar conformation. The 7-membered ring
cyclic ureas, on the other hand, bind in their lowest
energy conformer. In addition there is an entropic

1714 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11
De Lucca et al.
Ch a r t 2. Other Cyclized Linear HIVPR Inhibitors
K
K
i
i
F igu r e 9. Overlapped DMP450 (green) and 24 (white) in their
bound conformations obtained from X-ray analysis of their
respective complexes with HIVPR. There is an extremely good
fit of all four phenyl groups suggesting the need to maximize
S1/S1′ interactions.
penalty,25 due to the extra rotatable bond of the phen-
ethyl compared to the benzyl group. Thus, the tetrahy-
dropyrimidinone would always have to pay this small
energy cost even when its interaction with the enzyme
is the same as the 7-membered ring cyclic urea (as
pictured in Figure 9). This energy difference (enthalpic
and entropic) would result in the observed weaker Ki
values.
instead of CDI to yield the 7-membered cyclic oxamide
14, which is also a good inhibitor of HIVPR.17 Since
DuPont Merck disclosed its work on the 7-membered
ring cyclic ureas, there have been other examples of
linear HIVPR inhibitors that have been cyclized to give
potent cyclic compounds (Chart 2). Researchers at
Abbott,18 Ciba,19 and DuPont Merck20 have disclosed the
cyclization of linear hydrazine inhibitors to give the
azacyclic ureas 15. Work published by Abbott shows
these to be potent inhibitors of HIVPR. In addition,
X-ray structures of complexes with HIV protease show
that the azacyclic ureas have a similar conformation and
binding mode as the cyclic ureas.18 Researchers at
Hoechst have cyclized their linear diamino phosphinic
acids to give the 6-membered ring urea phosphinic acids
16.21 Although the cyclic phosphinic acids are mixtures
of isomers and they do not have the optimal phenethyl
P1′ substituent, they show surprisingly good activity.
The other influence on the SAR seems to be due
differences in the enzyme’s ability to tolerate substitu-
tion from the para position from the N-benzyl P2 group
of the tetrahydropyrimidinones. Since the 6-membered
ring is smaller, the enzyme conformation may change
slightly to accommodate the difference. However, the
X-ray structures we have obtained (at the current 2 Å
resolution) do not show significant differences in the
enzyme conformation when complexed with tetrahydro-
pyrimidinones compared to the 7-membered ring cyclic
ureas. The net effect, however, is that substituents at
the para position of the P2 benzyl groups are less well
tolerated in the tetrahydropyrimidinone analogs com-
pared to the 7-membered ring cyclic ureas. For ex-
ample, the p-hydroxymethyl analog 29 is 10 times less
potent than the corresponding 7-membered ring cyclic
urea DMP323.8c In contrast the m-hydroxymethyl
tetrahydropyrimidinone analog 21 is nearly equipotent
to the corresponding 7-membered ring cyclic urea
analog.8c An even more dramatic difference is seen with
the meta- and para-substituted methyl esters 20 and
30, which show a 1000-fold difference in binding. In
the corresponding 7-membered ring cyclic ureas both
isomers are equipotent8c (data not shown). Another
manifestation of this seems to extend to the direction-
ality of hydrogen-bonding interactions at the meta
position. Whereas the m-hydroxy-substituted analogs
10 and 11 are nearly equivalent (Chart 1), the m-amino
analog 25 is 10 times less potent that the cyclic urea
analog DMP450. This may be due to the rotational
freedom of a phenol compared to an aniline in their
hydrogen-bonding capabilities. Thus, although the SAR
of the tetrahydropyrimidinones parallels that of the
cyclic ureas, there are some subtle differences that can
be rationally explained.
Con clu sion
The work reported here and the examples cited show
that in general one can convert a linear HIVPR inhibitor
into a potent cyclic inhibitor as long as one takes into
account the conformation of the resulting cyclic com-
pound and how that conformation will interact with the
enzyme active site. In certain cases that will necessitate
a modification of the substituents on the inhibitor for
optimum interaction with the enzyme (for example,
changing a benzyl into a phenethyl group). This method
may be able to produce a larger diversity of potent,
nonpeptide HIVPR inhibitors than are now available.
We are continuing our efforts in expanding the scope of
this method and will report on other examples in the
near future.26
Exp er im en ta l Section
Biologica l Meth od s. Inhibition of HIVPR was measured
by the assay of the cleavage of a fluorescent substrate using
HPLC as described previously.9b The antiviral activity potency
of compounds was assessed by measuring their effect on the
accumulation of viral RNA transcripts 3 days after infection
of MT-2 cells with HIV-1RF as described previously.22 The
concentration of test compound which reduced the concentra-
tion of HIV viral RNA by 90% from the level measured in an
Various cyclic structures, other than a cyclic urea, can
also be envisioned with the diamino alcohol core. For
example, following the same strategy used for the
tetrahydropyrimidinones discussed above, one can cy-
clize the diamino alcohol core using oxalyl chloride
untreated infected culture is designated IC90
.
Molecu la r Mod elin g. Molecular modeling studies were
performed using SYBYL version 6.0, commercially available
from Tripos Associates, Inc. and viewed on a Silicon Graphics
Indigo2 workstation. Models of tetrahydropyrimidinones were

General Approach to Nonpeptide HIVPr Inhibitors
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11 1715
built using SYBYL’s build feature (Sketch Molecule and the
Fragment library) and energy minimized to convergence using
the MAXIMIN standard options (forcefield, Tripos; method,
Powell; charge, Gasteiger-Huckel; dielectric constant, 1.0;
termination, gradient 0.05 kcal/mol Å). Using the X-ray
structure of DMP323 and/or DMP450 the tetrahydropyrimi-
dinone models were manually docked to give the best overlap.
Then an iterative process of manually adjusting the rotatable
bonds on the models to best overlap with the 7-membered ring
cyclic ureas was used followed by minimization using MAXI-
MIN until the difference was as small as possible. The final
dihedral angle for the phenethyl side chain in the model was
127°, and the energy difference from the energy minimum was
about 3 kcal. In the X-ray structure of 24/HIVPR complex this
angle was found to be 116°.
reaction mixture was stirred at room temperature for 30 min
and then concentrated on a rotary evaporator to one-third of
the volume. The resulting solution was neutralized with 1 N
HCl and the precipitate formed extracted into ethyl ether. The
extracts were washed with water and brine, then dried over
MgSO4, and concentrated to give a white solid. This was
chromatographed (MPLC, silica gel, 30% EtOAc/hexane) to
give 7.5 g (65%, three steps) of the allylic alcohol 1 as a mixture
of diastereomers: CIMS (NH3) m/z 312 (M + H+, 100).
A solution of the allylic alcohol 1 (31 g, 0.1 mol) in CH2Cl2
was cooled to 0 °C (ice bath) and treated with diisopropyleth-
ylamine (30 g, 0.23 mol). Then methanesulfonyl chloride (26
g, 0.23 mol) was added slowly and the mixture stirred for 1 h.
The reaction mixture was transferred to a separatory funnel
and washed with 1 N HCl, water, and brine. The solution was
dried (MgSO4) and concentrated to give the mesylate as an
oil, which was used without further purification.
CuCN (12 g, 0.14 mol) in THF (100 mL) was cooled to -70
°C and treated slowly with benzylmagnesium chloride (360
mL, 2 M THF, 0.72 mol) via syringe. The solution was stirred
at -70 °C for 20 min and then at 0 °C for 30 min and then
cooled at -70 °C again before adding the mesylate above as a
solution in THF (130 mL) via syringe. The reaction mixture
was stirred at -70 °C for 60 min, poured into a mixture of 1
N HCl/ice, and extracted into EtOAc. The extract was washed
with NH4Cl (aqueous), NH4OH, and brine, then dried over
MgSO4, and concentrated. The residue was chromatographed
(MPLC, silica gel, 10% EtOAc/hexane) to give 11.7 g (30%, two
steps) of the trans olefin 2 as a white solid: CIMS (NH3) m/z
386 (M + H+, 98), 403 (M + NH4+, 100).
A solution of the olefin 2 (11.0 g, 0.029 mol) in CH2Cl2 (75
mL) was cooled to 0 °C in an ice bath, treated with MCPBA
(60%, 14 g, 0.048 mol), and stirred at 0 °C for 7 h. After the
reaction was complete the solution was diluted with 100 mL
of CH2Cl2 and washed successively with 1 N Na2S2O3, 1 N
NaOH, water, and brine. The solution was dried over MgSO4
and concentrated to give 11.6 g of the epoxide as a thick oil,
which was used without further purification.
To a solution of the epoxide in DMF (80 mL) were added
NaN3 (20 g, 0.3 mol), NH4Cl (2.6 g, 0.05 mol), and 20 mL of
water, and the resulting solution was heated at 85 °C for 3 h
and then stirred overnight at room temperature. The solution
was concentrated on a rotary evaporator under high vacuum
at 60 °C and the resulting residue partitioned between water
and CH2Cl2. The organic extract was washed with water and
brine, dried over MgSO4, and concentrated. The residue was
chromatographed (MPLC, silica gel, 20% EtOAc/hexane) to
give 7.4 g (58%, two steps) of the azido alcohol 3 as a white
solid as the major isomer (about 4:1 ratio by NMR): CIMS
(NH3) m/z 445 (M + H+).
A solution of the above azido alcohol 3 (7.2 g, 0.016 mol) in
CH2Cl2 was treated with diisopropylamine (4.2 g, 0.032 mol)
and MEM-Cl (4.0 g, 0.032 mol) and heated to reflux overnight.
The solution was concentrated in vacuo, and the residue was
chromatographed (MPLC, silica gel, 20% EtOAc/hexane) to
give 7.7 g (90%) of the azido MEM ether 4 as a colorless oil:
CIMS (NH3) m/z 533 (M + H+, 100).
A solution of azido MEM ether 4 (5.7 g, 0.0107 mol) in
EtOAc was treated with 2 mL of acetic acid and 1 g of
Pearlman’s catalyst, and the mixture was hydrogenated at 55
psi of H2 for 20 h. The solution was filtered (Celite) and then
extracted with 1 N HCl (3 × 100 mL). The aqueous extract
was made basic with 50% NaOH while cooling in an ice bath,
and the resulting precipitate was extracted into EtOAc. The
organic extracts were washed with brine, dried (MgSO4), and
concentrated to give 2.5 g of the diamine as a colorless oil:
CIMS (NH3) m/z 373 (M + H+, 100).
A solution of the diamine (2.5 g, 0.0067 mol) in THF (75
mL) was treated with CDI (1.1 g, 0.0067 mol) and stirred at
room temperature overnight. The reaction mixture was
concentrated, and the residue was chromatographed (HPLC,
silica gel, 5% MeOH/CHCl3) to give 1.3 g (30%, two steps) of
the tetrahydropyrimidinone 5: 1H NMR (CDCl3) δ 7.37-7.15
(m, 10 H), 5.41 (bs, 1 H), 4.87 (d, J ) 8 Hz, 1 H, ab), 4.78 (d,
J ) 8 Hz, 1 H, ab), 4.72 (bs, 1 H), 3.82 (m, 1 H), 3.72 (m, 2 H),
3.60 (bm, 2 H), 3.38 (s, 3 H), 2.97 (dd, J ) 5, 13 Hz, 1 H, abx),
X-r a y Cr ysta llogr a p h y. The complex of 24 with HIV-1
protease was crystallized as described previously.8a The unit
cell dimensions of the complex are a ) b ) 63.2 Å and c )
83.9 Å. The diffraction data were collected with a R-AXIS II
imaging plate mounted on an RU200 Rigaku rotating anode
generator operating at 50 kV and 100 mA. Total 27 445
reflections were collected resulting in 11 937 unique reflec-
tions, and Rsym was 12.7%. The data-extended 2.0 Å resolution
was 92% complete. The difference maps calculated with the
proton coordinate of XK2638a revealed the corresponding
inhibitor position. The structure was refined using the
simulated annealing method, XPLOR.24 The final R-factor
using the reflections with intensities greater than 2s(I) was
0.215.
Gen er a l. All reactions were carried out under an atmo-
sphere of dry nitrogen. Commercial reagents were used
without purther purification. 1H NMR (300 MHz) spectra were
recorded using tetramethylsilane as an internal standard. TLC
was performed on E. Merck 15710 silica gel plates. Medium
pressure liquid chromatography (MPLC) was carried out using
EM Science silica gel 60 (230-400 mesh). All final targets
were obtained as noncrystalline amorphous solids unless
specified otherwise. Mass spectra were measured with a
HP5988A 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. Elemental analysis
was performed by Quantitative Technologies, Inc., Bound
Brook, NJ . For compounds where analysis was not obtained,
HPLC analysis was used and purity was determined to be
>98%.
(4R,5R,6R)-Tetr a h yd r o-5-O-[(2-m eth oxyeth oxy)m eth -
yl]-4-(2-p h en ylet h yl)-6-(p h en ylm et h yl)-2(1H )-p yr im id i-
n on e (5). A solution of D-Z-Phe (11.0 g, 0.037 mol) in THF
was cooled to -20 °C and treated with N-methylmorpholine
(4.0 g, 0.04 mol) followed by dropwise addition of isobutyl
chloroformate (5.0 g, 0.037 mol), and the resulting solution was
allowed to stir for 20 min. Then a slurry of N,O-dimethyl-
hydroxylamine‚HCl (4.9 g, 0.05 mol) in DMF (and neutralized
with N-methylmorpholine) was added, and the resulting
reaction mixture was stirred for 30 min. The volume was
reduced at room temperature on a rotary evaporator and the
residue dissolved in ethyl acetate, washed successively with
water, 10% HCl, 10% NaHCO3, and brine, and dried over
MgSO4. After filtration, the solvent was removed under
reduced pressure to give 12.6 g of the amide.
The crude amide was dissolved in ether, cooled to 0 °C (ice
bath), and treated with vinylmagnesium bromide (120 mL, 1
M THF). The reaction mixture was stirred for 30 min at 0
°C. The reaction mixture was then poured while stirring into
a mixture of HCl (concentrated) (30 mL)/ice/water (200 mL)
and then extracted into ethyl acetate. The organic layer was
washed with water and brine, dried over MgSO4, and concen-
trated to give 11.5 g of the vinyl ketone as a pale yellow oil:
CIMS (NH3) m/z 327 (M + NH4+, 100).
A solution of the crude vinyl ketone in methanol was treated
at room temperature with CeCl3‚7H2O (13.8 g, 0.037 mol).
The resultant solution was stirred vigorously while NaBH4 (1.4
g, 0.037 mol) was added in very small portions (caution:
vigorous foaming). After the addition was complete the

1716 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11
De Lucca et al.
2.81 (m, 2 H, abx), 2.57 (m, 1 H, abx), 1.74 (m, 2 H); CIMS
(NH3) m/z 399 (M + H+, 100); HRMS calcd for C23H32N2O4 (M
+ H+) 399.2284, found 399.2279.
dispersion) in DMF was added tetrahydropyrimidinone 5 (500
mg, 1.3 mmol), and the mixture was stirred at room temper-
ature for 30 min. To the resulting solution was added
3-cyanobenzyl bromide (800 mg, 4 mmol), and the mixture was
stirred at room temperature for 45 min. The reaction mixture
was diluted with 5% HCl (aqueous) and extracted into EtOAc.
The organic extracts were washed with water and brine, then
dried (MgSO4), and concentrated. The residue was chromato-
graphed (HPLC, silica gel, 50% EtOAc/hexane) to give 560 mg
of the N,N-dialkylated tetrahydropyrimidinone: CIMS (NH3)
m/z 629 (M + H+, 100).
A solution of the N,N-dialkylated tetrahydropyrimidinone
(75 mg) in MeOH was treated with 4 N HCl/dioxane and
stirred at room temperature overnight. The solution was
evaporated to dryness in vacuo, and the residue was chro-
matographed (HPLC, silica gel, 50% EtOAc/hexane) to give
35 mg of 6 as a foam.
(4R,5R,6R)-Tetr a h yd r o-5-h yd r oxy-4-(2-p h en yleth yl)-6-
(p h en ylm eth yl)-2(1H)-p yr im id in on e (17). A solution of
tetrahydropyrimidinone 5 (30 mg) in methanol (2 mL) was
treated with a saturated HCl/methanol solution (5 mL) and
stirred at room temperature for 5 h. The mixture was
evaporated to dryness in vacuo and the residue chromato-
graphed (HPLC, Zorbax NH2 column, 7% MeOH/CH2Cl2) to
give 20 mg of a white solid: mp 162-163 °C; 1H NMR (CDCl3)
δ 7.38-7.19 (m, 10 H), 6.69 (bs, 1 H), 5.00 (bs, 1 H), 4.61 (d, J
) 8 Hz, 1 H), 3.60 (t, J ) 7 Hz, 1 H), 3.49 (bm, 1 H), 3.38 (bm,
1 H), 3.30 (dd, J ) 7, 14 Hz, 1 H, abx), 2.82 (dd, J ) 7, 14 Hz,
1 H, abx), 2.78 (m, 1 H, abx), 2.57 (m, 1 H, abx), 1.62 (m, 2 H);
13C NMR (CDCl3) δ 157.1, 141.7, 137.2, 129.7, 129.2, 128.8,
128.8, 127.3, 126.4, 65.9, 57.1, 53.5, 38.4, 37.7, 32.4; CIMS
(NH3) m/z 311 (M + H+, 100). Anal. (C19H22N2O2) C, H, N.
Meth od 2: F r om 7-Mem ber ed Cyclic Ur ea . To a solution
of (4R,5S,6S,7R)-hexahydro-5,6-dihydroxy-1,3-bis[(3-cyano-
phenyl)methyl]-4,7-bis(phenylmethyl)-2H-1,3-diazapin-2-
one8c (1.0 g, 1.8 mmol) in CH2Cl2 (20 mL) at room temperature
was added 2-acetoxyisobutyryl bromide (0.56 g, 2.7 mmol), and
the solution was stirred at room temperature for 15 min at
which time TLC showed complete conversion. The reaction
was quenched with saturated NaHCO3, and the organic layer
was separated, washed with water and brine, dried, and
concentrated to give 1.1 g of the rearranged bromo acetate.
The crude bromo acetate (1.0 g) was dissolved in 50 mL of
acetic acid, treated with 7 g of Zn (dust), and vigorously stirred
at room temperature until TLC analysis showed complete
conversion. The mixture was filtered, and the solid was
washed thoroughly with EtOAc. The filtrate was washed with
water, saturated NaHCO3, and brine, dried, and evaporated
to give the acetate. The crude acetate was chromatographed
(MPLC, silica gel, 65% EtOAc/hexane) to give 820 mg of
acetate as a white foam. The pure acetate (400 mg, 0.68 mol)
was dissolved in MeOH, treated with 1 N NaOH, and stirred
at room temperature. The mixture was concentrated, and the
residue was partitioned between 1 N HCl and EtOAc. The
organic extract was washed with water and brine, dried, and
concentrated and the resulting residue chromatographed
(MPLC, silica gel, 65% EtOAc/hexane) to give 300 mg of 6 as
a foam identical in every respect with that obtained via method
1: 1H NMR (CDCl3) δ 7.59-7.41 (m, 8 H), 7.38-7.21 (m, 6
H), 7.00 (dd, J ) 7, 16 Hz, 4 H), 5.23 (d, J ) 15 Hz, 1 H), 5.13
(d, J ) 16 Hz, 1 H), 4.25 (d, J ) 16 Hz, 1 H), 3.91 (d, J ) 15 Hz,
1 H), 3.56 (m, 2 H), 3.14 (m, 1 H), 2.88 (m, 2 H), 2.40 (m, 2 H),
1.90 (m, 1 H), 1.76 (d, J ) 8 Hz, 1 H), 1.62 (m, 2 H); CIMS
(NH3) m/z 541 (M + H+, 100); HRMS calcd for C35H33N4O2 (M
(4R,5R,6R)-Tetr a h yd r o-1,3-bis(cyclop r op ylm eth yl)-5-
h yd r oxy-4-(2-p h en ylet h yl)-6-(p h en ylm et h yl)-2(1H )-p y-
r im id in on e (18). Meth od 1: F r om Tetr a h yd r op yr im id i-
n on e 5. To a suspension of NaH (60 mg, 1.5 mmol, 60% oil
dispersion) in DMF was added tetrahydropyrimidinone 5 (100
mg, 0.27 mmol), and the mixture was stirred at room temper-
ature for 30 min. To the resulting solution was added
(bromomethyl)cyclopropane (250 mg, 1.8 mmol), and the
mixture was stirred at room temperature for 30 min and then
at 60 °C for 30 min. The reaction mixture was diluted with
5% HCl (aqueous) and extracted into EtOAc. The organic
extracts were washed with water and brine, then dried
(MgSO4), and concentrated. The residue was chromato-
graphed (HPLC, silica gel, 50% EtOAc/hexane) to give 35 mg
of the N,N-dialkylated tetrahydropyrimidinone as a colorless
film: CIMS (NH3) m/z 507 (M + H+, 100).
A solution of the N,N-dialkylated tetrahydropyrimidinone
(25 mg) in MeOH was treated with HCl (gas) for 20 min and
then stirred at room temperature for 1 h. The solution was
evaporated to dryness in vacuo, and the residue was chro-
matographed (HPLC, silica gel, 10% MeOH/CHCl3) to give 20
mg of 18 as a colorless film: 1H NMR (CDCl3) δ 7.38-7.12
(m, 8 H), 7.05 (d, J ) 7 Hz, 2 H), 3.92 (dd, J ) 7, 14 Hz, 1 H,
abx), 3.85 (dd, J ) 7, 14 Hz, 1 H, abx), 3.72 (m, 1 H), 3.62 (m,
1 H), 3.40 (m, 1 H), 3.11 (dd, J ) 6, 13 Hz, 1 H, abx), 2.95 (dd,
J ) 10, 12 Hz, 1 H, abx), 2.85 (dd, J ) 7, 14 Hz, 1 H, abx),
2.67 (dd, J ) 7, 14 Hz, 1 H, abx), 2.51 (m, 2 H), 2.30 (d, 10 Hz,
1 H), 2.96 (m, 1 H), 1.70 (m, 1 H), 1.03 (m, 2 H), 0.60-0.15
(m, 8 H); CIMS (NH3) m/z 419 (M + H+, 100); HRMS calcd for
C
27H34N2O2 (M + H+) 419.2699, found 419.2704.
Meth od 2: F r om 7-Mem ber ed Cyclic Ur ea . To a solution
+ H+) 541.2604, found 541.2587; [R]25 -10.38° (c 0.212,
D
of (4R,5S,6S,7R)-hexahydro-5,6-dihydroxy-1,3-bis(cyclopropyl-
methyl)-4,7-bis(phenylmethyl)-2H-1,3-diazapin-2-one (200 mg,
0.46 mmol) in CH2Cl2 (6 mL) at room temperature was added
2-acetoxyisobutyryl bromide (300 mg, 1.4 mmol), and the
solution was stirred at room temperature for 15 min at which
time TLC showed complete conversion. The reaction was
quenched with saturated NaHCO3, and the organic layer was
separated, washed with water and brine, dried, and concen-
trated to give 250 mg of the corresponding bromo acetate. The
crude bromo acetate was dissolved in 10 mL of acetic acid,
treated with 2 g of Zn (dust), and vigorously stirred at room
temperature until TLC analysis showed complete conversion.
The mixture was filtered and the solid washed thoroughly with
EtOAc. The filtrate was washed with water, saturated
NaHCO3, and brine, dried, and evaporated to give the acetate
as an oil. The crude acetate was dissolved in MeOH, treated
with 1 N NaOH, and stirred at room temperature. The
mixture was concentrated, and the residue was partitioned
between 1 N HCl and EtOAc. The organic extract was washed
with water and brine, dried, and concentrated, and the
resulting residue was chromatographed (HPLC, silica gel, 10%
MeOH/CHCl3) to give 20 mg of 18 as a colorless film identical
in every respect with that obtained via method 1.
CHCl3). Anal. (C35H32N4O2‚0.33H2O) C, H, N.
(4R,5R,6R)-Tetr a h yd r o-1,3,6-tr is(p h en ylm eth yl)-5-h y-
d r oxy-4-(2-p h en yleth yl)-2(1H)-p yr im id in on e (19). To a
solution of (4R,5S,6S,7R)-hexahydro-5,6-dihydroxy-1,3,4,7-tet-
rakis(phenylmethyl)-2H-1,3-diazapin-2-one8c (1.3 g, 2.6 mmol)
in CH2Cl2 (40 mL) at room temperature was added 2-acetoxy-
isobutyryl bromide (1.7 g, 8.1 mmol), and the solution was
stirred at room temperature for 15 min at which time TLC
showed complete conversion. The reaction was quenched with
saturated NaHCO3, and the organic layer was separated,
washed with water and brine, dried, and concentrated to the
rearranged bromo acetate. The crude bromo acetate was
chromatographed (MPLC, silica gel, 30% EtOAc/hexane) to
give 1.4 g of 9. A small fraction of the eluent from this
chromatography was allowed to evaporate slowly to give
crystals of the bromo acetate 9 that were of sufficient quality
for single-crystal X-ray analysis. For 9: mp 178-180 °C.
Anal. (C35H35N2O3Br‚0.2C6H14) C, H, N.
The bromo acetate 9 (1.0 g) was dissolved in 50 mL of acetic
acid, treated with Zn (dust) (7 g), and vigorously stirred at
room temperature until TLC analysis showed complete con-
version. The mixture was filtered and the solid washed
thoroughly with EtOAc. The filtrate was washed with water,
saturated NaHCO3, and brine, dried, and evaporated to give
the acetate. The crude acetate was chromatographed (MPLC,
silic gel, 65% EtOAc/hexane) to give 820 mg of acetate as a
(4R,5R,6R)-Tetr ah ydr o-1,3-bis[(3-cyan oph en yl)m eth yl]-
5-h yd r oxy-4-(2-p h en yleth yl)-6-(p h en ylm eth yl)-2(1H)-p y-
r im id in on e (6). Meth od 1: F r om Tetr a h yd r op yr im id i-
n on e 5. To a suspension of NaH (200 mg, 5 mmol, 60% oil

General Approach to Nonpeptide HIVPr Inhibitors
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11 1717
white foam. The pure acetate (400 mg, 0.68 mol) was dissolved
in MeOH, treated with 1 N NaOH (5 mL), and stirred at room
temperature. The mixture was concentrated, and the residue
was partitioned between 1 N HCl and EtOAc. The organic
extract was washed with water and brine, dried, and concen-
trated, and the resulting residue was chromatographed (MPLC,
silica gel, 65% EtOAc/hexane) to give 300 mg of 19 as a foam:
1H NMR (CDCl3) δ 7.37-7.15 (m, 16 H), 6.97 (m, 4 H), 5.46
(d, J ) 15 Hz, 1 H), 5.42 (d, J ) 14 Hz, 1 H), 3.89 (d, J ) 14
Hz, 1 H), 3.82 (d, J ) 15 Hz, 1 H), 3.41 (m, 1 H), 3.33 (m, 1 H),
3.17 (m, 1 H), 2.95 (m, 2 H), 2.85 (m, 2 H), 2.40 (t, J ) 8 Hz,
2 H), 1.9 (m, 1 H), 1.61 (m, 2 H); CIMS (NH3) m/z 491 (M +
H+, 100). Anal. (C33H34N2O2) C, H, N.
abx), 2.36 (t, J ) 8 Hz, 2 H), 1.90 (m, 1 H), 1.65 (m, 2 H);
ESIMS m/z 304 (M + 2H2+, 100); HRMS calcd for C35H39N6O4
(M + H+) 607.3033, found 607.3016.
(4S,5S,6S)-Tetr ah ydr o-1,3-bis[(3-car bom eth oxyph en yl)-
m eth yl]-5-h yd r oxy-4-(2-p h en yleth yl)-6-(p h en ylm eth yl)-
2(1H)-p yr im id in on e (34). A solution of 32 in methanol was
treated with H2SO4 and heated at reflux for 24 h to give the
methyl ester 34: 1H NMR (CDCl3) δ 7.95 (m, 4 H), 7.56 (m, 1
H), 7.42 (m, 3 H), 7.22 (m, 6 H), 7.00 (dd, J ) 7, 16 Hz, 4 H),
5.39 (d, J ) 15 Hz, 2 H), 3.98 (d, J ) 15 Hz, 1 H), 3.93 (d, J )
15 Hz, 1 H), 3.91 (s, 3 H), 3.84 (s, 3 H), 3.52 (m, 1 H), 3.41 (m,
1 H), 3.19 (m, 1 H), 2.87 (m, 2 H), 2.41 (m, 2 H) 1.90 (m, 1 H),
1.87-1.59 (m, 2 H); CIMS (NH3) m/z 607 (M + H+, 100); HRMS
calcd for C37H39N2O6 (M + H+) 607.2808, found 607.2787.
Anal. (C37H38N2O6) C, H, N.
Cr ysta llogr a p h ic Da ta for 9. Empirical formula: C35H35
-
N2O3Br, from 30% EtOAc/hexane, colorless, flat needle, 0.07
× 0.15 × 0.90 mm, monoclinic, P21 (No. 4); a ) 11.131(1), b )
11.908(1), c ) 12.776(1) Å; â ) 111.747(5)°; T ) -51 °C; V )
1572.9 Å3, Z ) 2; FW ) 611.59, Dc ) 1.291 g/cc, µ(Mo) ) 13.27
cm-1. Data were collected on a Rigaku RU300, R-AXIS image
plate area detector, Mo KR radiation, anode power ) 50 Kv ×
120 ma, crystal to plate distance ) 85.0 mm, 210 µm pixel
raster, number of frames ) 31, oscillation range ) 6.0°/frame,
exposure ) 4.0 min/frame, box sum integration, 8344 data
collected, 3.4° e 2θ e 48.3°, maximum h, k, l ) 12, 13, 14, no
absorption correction, 2332 duplicates, 7.4% R-merge, 1880
unique reflections with I g 3.0σ(I). Structure was solved by
direct methods (MULTAN) [difficult solution complicated by
disorder in one of the phenyl rings. The asymmetric unit
consists of one molecule in a general position. Hydrogen atoms
are idealized with C-H ) 0.95 Å. The acentric space group
indicates an enantiomeric isomer which is in agreement with
the calculated intensity differences for the anomalous terms
of bromine (R, Rw ) 6.29/5.82 vs 6.57/6.31 for the inverted
structure)], refinement by full-matrix least-squares on F,
scattering factors from International Tables for X-ray Crystal-
lography, Vol. IV, including anomalous terms for Br, biweight
R [σ2(I) + 0.0009(I)2]2 (excluded 20); refined anisotropic, all non-
hydrogen atoms; fixed atoms, H; 369 parameters, data/
parameter ratio ) 5.04, R ) 0.063, Rw ) 0.058, error of fit )
1.82, max ∆/σ ) 0.15, largest residual density ) 0.40 e/Å3.
(4S ,5S ,6S )-Te t r a h yd r o-1,3-b is[[3-(h yd r oxym e t h yl)-
p h en yl]m eth yl]-5-h yd r oxy-4-(2-p h en yleth yl)-6-(p h en yl-
m eth yl)-2(1H)-p yr im id in on e (35). A solution of 34 in THF
was treated with LAH at 0 °C for 3 h to give the alcohol 35:
1H NMR (CDCl3) δ 7.36-7.08 (m, 14 H), 6.95 (dd, J ) 7, 16
Hz, 4 H), 5.21 (d, J ) 15 Hz, 1 H), 5.19 (d, J ) 15 Hz, 1 H),
4.61 (bm, 4 H), 4.01 (d, J ) 15 Hz, 1 H), 3.95 (d, J ) 15 Hz, 1
H), 3.41 (m, 2 H), 3.18 (m, 1 H), 2.90 (m, 2 H), 2.79 (bm, 1 H),
2.50 (bm, 1 H), 2.33 (m, 3 H), 1.85 (m, 1 H), 1.57 (m, 1 H);
CIMS (NH3) m/z 551 (M + H+, 100); HRMS calcd for C35H39N2O4
(M + H+) 551.2910, found 551.2894. Anal. (C35H38N2O4‚
0.75H2O) C, H, N.
(4R,5R,6R)-Tetr ah ydr o-1,3-bis[(3-car bom eth oxyph en yl)-
m eth yl]-5-h yd r oxy-4-(2-p h en yleth yl)-6-(p h en ylm eth yl)-
2(1H)-p yr im id in on e (20). The same procedure described for
19 above was used: mp 156-158 °C; 1H NMR (CDCl3) δ 7.95
(m, 4 H), 7.56 (m, 1 H), 7.42 (m, 3 H), 7.22 (m, 6 H), 7.00 (dd,
J ) 7, 16 Hz, 4 H), 5.39 (d, J ) 15 Hz, 2 H), 3.98 (d, J ) 15
Hz, 1 H), 3.93 (d, J ) 15 Hz, 1 H), 3.91 (s, 3 H), 3.84 (s, 3 H),
3.52 (m, 1 H), 3.41 (m, 1 H), 3.19 (m, 1 H), 2.87 (m, 2 H), 2.41
(m, 2 H), 1.90 (m, 1 H), 1.87-1.59 (m, 2 H); CIMS (NH3) m/z
607 (M + H+, 100); HRMS calcd for C37H39N2O6 (M + H+)
607.2808, found 607.2808. Anal. (C37H38N2O6) C, H, N.
(4R ,5R ,6R )-Te t r a h yd r o-1,3-b is[[3-(h yd r oxym e t h yl)-
m eth yl]-5-h yd r oxy-4-(2-p h en yleth yl)-6-(p h en ylm eth yl)-
2(1H)-p yr im id in on e (21). A solution of 20 in THF was
treated with LAH at 0 °C for 3 h to give the alcohol 21: mp
(4S,5S,6S)-Tetr ah ydr o-1,3-bis[(3-cyan oph en yl)m eth yl]-
5-h yd r oxy-4-(2-p h en yleth yl)-6-(p h en ylm eth yl)-2(1H)-p y-
r im id in on e (31). Same procedure as that used to synthesize
6 (using method 2) but starting with the enantiomeric cyclic
urea (4S,5R,6R,7S)-hexahydro-5,6-dihydroxy-1,3-bis[(3-cy-
anophenyl)methyl]-4,7-bis(phenylmethyl)-2H-1,3-diazapin-2-
one was used to give 31: 1H NMR (CDCl3) δ 7.59-7.41 (m, 8
H), 7.38-7.21 (m, 6 H), 7.00 (dd, J ) 7, 16 Hz, 4 H), 5.23 (d,
J ) 15 Hz, 1 H), 5.13 (d, J ) 16 Hz, 1 H), 4.25 (d, J ) 16 Hz,
1 H), 3.91 (d, J ) 15 Hz, 1 H), 3.56 (m, 2 H), 3.14 (m, 1 H),
2.88 (m, 2 H), 2.40 (m, 2 H), 1.90 (m, 1 H), 1.76 (d, J ) 8 Hz,
1 H), 1.62 (m, 2 H); CIMS (NH3) m/z 541 (M + H+, 100); HRMS
1
58-61 °C; H NMR (CDCl3) δ 7.36-7.08 (m, 14 H), 6.95 (dd,
J ) 7, 16 Hz, 4 H), 5.22 (d, J ) 15 Hz, 1 H), 5.18 (d, J ) 15
Hz, 1 H), 4.61 (bm, 4 H), 4.01 (d, J ) 15 Hz, 1 H), 3.95 (d, J )
15 Hz, 1 H), 3.41 (m, 2 H), 3.18 (m, 1 H), 2.90 (m, 2 H), 2.79
(bm, 1 H), 2.50 (bm, 1 H), 2.33 (m, 3 H), 1.85 (m, 1 H), 1.57
(m, 1 H); CIMS (NH3) m/z 551 (M + H+, 100); HRMS calcd for
C35H39N2O4 (M + H+) 551.2910, found 551.2886. Anal.
(C35H38N2O4‚0.5H2O) C, H, N.
(4R,5R,6R)-Tetr a h yd r o-1,3-bis[(3-ben za m id o)m eth yl]-
5-h yd r oxy-4-(2-p h en yleth yl)-6-(p h en ylm eth yl)-2(1H)-p y-
r im id in on e (23). The cyano group of 6 was converted to the
amide using the DMSO/H2O2 procedure to give 23: mp 105-
108 °C; 1H NMR (CDCl3) δ 7.97 (s, 1 H), 7.84 (s, 1 H), 7.66
(bm, 2 H), 7.32-7.12 (m, 12 H), 7.05 (d, J ) 7 Hz, 2 H), 6.93
(d, J ) 16 Hz, 2 H), 6.88 (bs, 1 H), 6.57 (bs, 1 H), 5.23 (d, J )
16 Hz, 1 H), 4.80 (d, J ) 16 Hz, 1 H), 4.58 (d, J ) 16 Hz, 1 H),
4.30 (bs, 1 H), 3.96 (d, J ) 16 Hz, 1 H), 3.60 (m, 2 H), 3.22 (m,
1 H), 2.82 (d, J ) 7 Hz, 2 H), 2.32 (t, J ) 8 Hz, 2 H), 1.94 (m,
1 H), 1.57 (m, 1 H); CIMS (NH3) m/z 577 (M + H+, 100). Anal.
(C35H36N4O4‚H2O) C, H, N.
calcd for C35H33N4O2 (M + H+) 541.2604, found 541.2584; [R]25
D
+6.84° (c 0.190, CHCl3). Anal. (C35H32N4O2‚0.2H2O) C, H, N.
(4S,5S,6S)-Tetr a h yd r o-1,3-bis[(3-ben za m id o)m eth yl]-
5-h yd r oxy-4-(2-p h en yleth yl)-6-(p h en ylm eth yl)-2(1H)-p y-
r im id in on e (32). The cyano group of 31 was converted to
the amide using the DMSO/H2O2 procedure23 to give 32: 1H
NMR (CDCl3) δ 7.97 (s, 1 H), 7.84 (s, 1 H), 7.66 (bm, 2 H),
7.32-7.12 (m, 12 H), 7.05 (d, J ) 7 Hz, 2 H), 6.93 (d, J ) 16
Hz, 2 H), 6.88 (bs, 1 H), 6.57 (bs, 1 H), 5.23 (d, J ) 16 Hz, 1
H), 4.80 (d, J ) 16 Hz, 1 H), 4.58 (d, J ) 16 Hz, 1 H), 4.30 (bs,
1 H), 3.96 (d, J ) 16 Hz, 1 H), 3.60 (m, 2 H), 3.22 (m, 1 H),
2.82 (d, J ) 7 Hz, 2 H), 2.32 (t, J ) 8 Hz, 2 H), 1.94 (m, 1 H),
1.57 (m, 1 H); CIMS (NH3) m/z 577 (M + H+, 100); HRMS calcd
for C35H37N4O4 (M + H+) 577.2815, found 577.2809. Anal.
(C35H36N4O4‚C2H5OH) C, H, N.
(4R,5R,6R)-Tetr a h yd r o-1,3-bis[(4-flu or o-3-ben za m id o)-
m eth yl]-5-h yd r oxy-4-(2-p h en yleth yl)-6-(p h en ylm eth yl)-
2(1H)-p yr im id in on e (27). The cyano group was converted
to the amide using the DMSO/H2O procedure to give 27: mp
1
106-109 °C; H NMR (CDCl3) δ 8.00 (m, 2 H), 7.50 (m, 1 H),
7.40 (m, 1 H), 7.26-6.96 (m, 12 H), 6.70 (bd, J ) 11 Hz, 2 H),
6.13 (bs, 2 H), 5.22 (d, J ) 15 Hz, 1 H), 5.06 (d, J ) 15 Hz, 1
H), 4.09 (d, J ) 15 Hz, 1 H), 3.99 (d, J ) 15 Hz, 1 H), 3.47 (m,
2 H), 3.22 (m, 1 H), 2.85 (m, 2 H), 2.38 (m, 2 H), 1.90 (m, 1 H),
1.86 (bs, 1 H), 1.63 (m, 1 H); CIMS (NH3) m/z 613 (M + H+,
100). Anal. (C35H34N4O4F2) C, H, N.
(4R,5R,6R)-Tetr ah ydr o-1,3-bis[(4-car bom eth oxyph en yl)-
m eth yl]-5-h yd r oxy-4-(2-p h en yleth yl)-6-(p h en ylm eth yl)-
2(1H)-p yr im id in on e (30). The same procedure described for
(4S,5S,6S)-Tet r a h yd r o-1,3-b is[(3-b en za m id e oxim e)-
m eth yl]-5-h yd r oxy-4-(2-p h en yleth yl)-6-(p h en ylm eth yl)-
2(1H)-p yr im id in on e (33). An ethanol solution of 31 was
treated with excess NH2OH‚HCl/Et3N and heated (EtOH) to
reflux for 4 h to give the amidoxime 33: 1H NMR (CDCl3/CD3-
OD) δ 7.67 (bs, 1 H), 7.52 (bs, 1 H), 7.45 (m, 2 H), 7.30-7.12
(m, 10 H), 6.98 (m, 4 H), 6.10 (bs, 2 H), 5.22 (d, J ) 15 Hz, 1
H), 4.87 (d, J ) 15 Hz, 1 H), 4.11 (d, J ) 15 Hz, 1 H), 3.94 (d,
J ) 15 Hz, 1 H), 3.51 (m, 2 H), 3.21 (m, 1 H), 2.78 (m, 2 H,

1718 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11
De Lucca et al.
20 above was used: mp 72-75 °C; 1H NMR (CDCl3) δ 8.0 (m,
4 H), 7.2 (m, 10 H), 7.0 (m, 4 H), 5.4 (d, J ) 16 Hz, 1 H), 5.3
(d, J ) 15 Hz, 1 H), 4.1 (d, J ) 16 Hz, 1 H), 3.96 (d, J ) 15 Hz,
1 H), 3.96 (s, 3 H), 3.93 (s, 3 H), 3.45 (m, 1 H), 3.40 (m, 1 H),
3.15 (m, 1 H), 2.95 (m, 2 H), 2.40 (m, 2 H), 1.90 (m, 1 H), 1.76
(d, J ) 8 Hz, 1 H), 1.61 (m, 2 H); CIMS (NH3) m/z 607 (M +
H+, 100); HRMS calcd for C37H39N2O6 (M + H+) 607.2808,
found 607.2797. Anal. (C37H38N2O6‚0.15H2O) C, H, N.
(4R,5R,6R)-Tetr ah ydr o-1,3-bis[[4-(h ydr oxym eth yl)ph en -
yl]m eth yl]-5-h ydr oxy-4-(2-ph en yleth yl)-6-(ph en ylm eth yl)-
2(1H)-p yr im id in on e (29). A solution of 30 in THF was
treated with LAH at 0 °C for 3 h to give the alcohol 29: 1H
NMR (CDCl3) δ 7.32-7.18 (m, 14 H), 7.03-6.97 (m, 4 H), 5.33
(d, J ) 15 Hz, 2 H), 4.66 (s, 2 H), 4.63 (s, 2 H), 3.83 (d, J ) 15
Hz, 1 H), 3.80 (d, J ) 15 Hz, 1 H), 3.41 (m, 1 H), 3.35 (m, 1 H),
3.15 (m, 1 H), 2.85 (m, 1 H), 2.40 (m, 1 H), 2.14 (bs, 1 H), 1.95
(bs, 1 H), 1.85 (m, 1 H), 1.70 (bs, 1 H), 1.60 (m, 1 H); CIMS
(NH3) m/z 551 (M + H+, 100). Anal. (C35H38N2O4‚0.25H2O)
C, H, N.
3.94 (d, J ) 15 Hz, 1 H), 3.51 (m, 2 H), 3.21 (m, 1 H), 2.78 (m,
2 H, abx), 2.36 (t, J ) 8 Hz, 2 H), 1.90 (m, 1 H), 1.65 (m, 2 H);
ESIMS m/z 304 (M + 2H2+, 100). Anal. (C35H38N6O4‚0.25H2O)
C, H, N.
(4R,5R,6R)-Tetr ah ydr o-1,3-bis[(3-h ydr oxyph en yl)m eth -
yl]-5-h yd r oxy-4,6-b is(p h e n ylm e t h yl)-2(1H )-p yr im id i-
n on e (13). The procedure described by Abbott11b was used to
obtain the azido alcohol, and then following the procedure
described above for 5 and 6 (method 1, using 3-benzoxybenzyl
chloride as the alkylating agent) gave, after hydrogenolysis,
the desired phenol 13: 1H NMR (CDCl3) δ 7.13-6.98 (m, 8
H), 6.93-6.71 (m, 3 H), 6.69 (s, 1 H), 6.67-6.64 (m, 6 H), 5.07
(d, J ) 16 Hz, 1 H), 4.94 (d, J ) 15 Hz, 1 H), 4.13 (d, J ) 16
Hz, 1 H), 3.86 (d, J ) 15 Hz, 1 H), 3.50 (m, 1 H), 3.30 (m, 1 H),
3.21 (m, 1 H), 2.92 (m, 1 H, abx), 2.71 (m, 2 H, 2 overlapping
abx), 2.29 (m, 1 H, abx), 1.70 (bs, 3 H); CIMS (NH3) m/z 509
(M + H+, 100); HRMS calcd for C32H33N2O4 (M + H+) 509.2440,
found 509.2456.
Ack n ow led gm en t. C.-H.C. would like to thank the
contributions of Mr. Richard DeLoskey in HIV crystal
structure experiments. We would like to thank Drs.
Patrick Y. S. Lam, Thomas R. Sharpe, C. Nicholas
Hodge, Alex J ohnson, Soo S. Ko, George L. Trainor,
David A. Jackson, Susan Erickson-Viitanen, Lee Bachel-
er, J im Meek, and Paul S. Anderson and the HIVPR
working group for their valuable contributions to the
HIVPR program.
(4R,5R,6R)-Tetr ah ydr o-1,3-bis[(3-h ydr oxyph en yl)m eth -
yl]-5-h yd r oxy-4-(2-p h en ylet h yl)-6-(p h en ylet h yl)-2(1H )-
p yr im id in on e (11): mp 93-95 °C; H NMR (CDCl3) δ 7.91
(bs, 2 H), 7.25-6.84 (m, 14 H), 6.73-6.59 (m, 4 H), 5.07 (d, J
) 15 Hz, 1 H), 4.98 (d, J ) 16 Hz, 1 H), 3.95 (d, J ) 16 Hz, 1
H), 3.85 (d, J ) 15 Hz, 1 H), 3.40 (m, 2 H), 3.15 (m, 1 H), 2.8
(m, 2 H), 2.3 (bs, 1 H), 2.25 (m, 2 H), 1.78 (m, 1 H), 1.48 (m, 1
H); CIMS (NH3) m/z 523 (M + H+, 100). Anal. (C33H34N2O4‚
0.25H2O) C, H, N.
1
(4R,5R,6R)-Tetr ah ydr o-1,3-bis[(3-am in oph en yl)m eth yl]-
5-h yd r oxy-4-(2-p h en yleth yl)-6-(p h en ylm eth yl)-2(1H)-p y-
r im id in on e (25): mp 89-92 °C; 1H NMR (CDCl3) δ 7.29-
7.00 (m, 14 H), 6.64-6.54 (m, 6 H), 5.37 (d, J ) 15 Hz, 1 H),
5.32 (d, J ) 15 Hz, 1 H), 3.77 (d, J ) 16 Hz, 1 H), 3.70 (d, J )
15 Hz, 1 H), 3.63 (bs, 2 H), 3.41 (bm, 1 H), 3.34 (m, 1 H), 3.18
(m, 1 H), 2.90 (m, 2 H, abx), 2.42 (m, 2 H), 1.95 (bs, 1 H), 1.85
(m, 1 H), 1.62 (m, 1 H); CIMS (NH3) m/z 523 (M + H+, 100).
Anal. (C33H36N4O2‚0.1C4H8O2) C, H, N.
Su p p or tin g In for m a tion Ava ila ble: Text describing
single-crystal X-ray analysis of 9 and tables giving fractional
coordinates, thermal parameters, bond distances and angles,
and nonbonding distances (6 pages). Ordering information is
given on any current masthead page.
Refer en ces
(4R,5R,6R)-Tetr a h yd r o-1,3-bis[(4-a m in o-3-flu or op h en -
yl)m eth yl]-5-h ydr oxy-4-(2-ph en yleth yl)-6-(ph en ylm eth yl)-
2(1H)-p yr im id in on e (28): mp 68-70 °C; 1H NMR (CDCl3) δ
7.29-7.17 (m, 6 H), 7.07-6.95 (m, 4 H), 6.92-6.81 (m, 4 H),
6.76-6.68 (m, 2 H), 5.33 (d, J ) 16 Hz, 1 H), 5.28 (d, J ) 15
Hz, 1 H), 3.76 (d, J ) 16 Hz, 1 H), 3.67 (d, J ) 15 Hz, 1 H),
3.71 (bs, 4 H), 3.40 (m, 1 H), 3.30 (m, 1 H), 3.15 (m, 1 H), 2.88
(m, 2 H, abx), 2.41 (t, J ) 8 Hz, 2 H), 1.85 (m, 1 H), 1.61 (m,
1 H), 1.61 (bs, 1 H); CIMS (NH3) m/z 557 (M + H+, 100). Anal.
(1) (a) Questions regarding the single-crystal X-ray structure of
compound 9 should be addressed to J . Calabrese at DuPont,
CR&D Experimental Station, P.O. Box 80228, Wilmington, DE
19880-0228. (b) Questions regarding the X-ray structure of 24/
HIVPR complex should be addressed to C.-H. Chang at DuPont
Merck Pharmaceutical Co., Experimental Station, P.O. Box
80228, Wilmington, DE 19880-0228.
(2) For recent reviews, see: Katz, R. A.; Skalka, A. M. The Retroviral
Enzymes. Annu. Rev. Biochem. 1994, 63, 133-173.
(3) (a) 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 Immunideficiency Virus Protease is Required for
Viral Infectivity. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 4686-
4690. (b) Peng, C.; Ho, B. K.; Chang, T. W.; Chang, N. T. Role
of Human Immunodeficiency Virus Type 1-Specific Protease in
Core Protein Maturation and Viral Infectivity. J . Virol. 1989,
63, 2550-2556.
(4) (a) Wlodawer, A.; Erickson, J . W. Structure-based inhibitors of
HIV-1 protease. Annu. Rev. Biochem. 1993, 62, 543-585. (b)
Appelt, K. Crystal Structures of HIV-1 Protease-Inhibitor Com-
plexes. Perspect. Drug Discovery Des. 1993, 1, 23-48.
(5) For recent reviews, see: (a) Norbeck, D. W.; Kempf, D. J . HIV
Protease Inhibitors. Annu. Rep. Med. Chem. 1991, 26, 141-
160. (b) Tomasselli, A. G.; Howe, W. J .; Sawyer, T. K.;
Wlodawer, A.; Heinrikson, R. L. The Complexities of AIDS: An
C
33H34N4O2F2) C, H, N.
(4R,5R,6R)-Tetr a h yd r o-1,3-bis[(3-a m in o-4-flu or op h en -
yl)m eth yl]-5-h ydr oxy-4-(2-ph en yleth yl)-6-(ph en ylm eth yl)-
2(1H)-p yr im id in on e (22): mp 68-70 °C; 1H NMR (CDCl3) δ
7.30-7.17 (m, 6 H), 7.06-6.85 (m, 6 H), 6.72-6.64 (m, 2 H),
6.58-6.50 (m, 2 H), 5.32 (d, J ) 15 Hz, 1 H), 5.24 (d, J ) 15
Hz, 1 H), 3.83 (d, J ) 15 Hz, 1 H), 3.68 (d, J ) 15 Hz, 1 H),
3.60 (bs, 4 H), 3.41 (m, 1 H), 3.35 (m, 1 H), 3.16 (m, 1 H), 2.90
(m, 2 H, abx), 2.41 (t, J ) 8 Hz, 2 H), 1.83 (m, 1 H), 1.70 (bs,
1 H), 1.60 (m, 1 H); CIMS (NH3) m/z 557 (M + H+, 100). Anal.
(C33H34N4O2F2‚0.5H2O) C, H, N.
(4R,5R,6R)-Tetr ah ydr o-1,3-bis[[3-(N-m eth ylam in o)ph en -
yl]m eth yl]-5-h ydr oxy-4-(2-ph en yleth yl)-6-(ph en ylm eth yl)-
2(1H)-p yr im id in on e (26): mp 65-66 °C deg; 1H NMR
(CDCl3) δ 7.28-7.00 (m, 12 H), 6.62-6.46 (m, 6 H), 5.44 (d, J
) 15 Hz, 1 H), 5.39 (d, J ) 14 Hz, 1 H), 3.79 (d, J ) 15 Hz, 1
H), 3.72 (d, J ) 14 Hz, 1 H), 3.70 (bs, 2 H), 3.4 (bm, 1 H), 3.35
(m, 1 H), 3.31 (m, 1 H), 3.00-2.83 (m, 2 H, abx), 2.80 (s, 3 H),
2.76 (s, 3 H), 2.44 (t, J ) 7 Hz, 2 H), 1.85 (m, 1 H), 1.75 (bs, 1
H), 1.65 (m, 1 H); CIMS (NH3) m/z 549 (M + H+, 100). Anal.
(C35H40N4O2‚0.25H2O) C, H, N.
(4R,5R,6R)-Tetr a h yd r o-1,3-bis[(3-ben za m id e oxim e)-
m eth yl]-5-h yd r oxy-4-(2-p h en yleth yl)-6-(p h en yleth yl)-2-
(1H)-p yr im id in on e (24). The cyano group of 6 was converted
to the amidoxime by treatment with NH2OH‚HCl/Et3N in
refluxing EtOH to give 24: mp 119-121 °C; 1H NMR (CDCl3/
CD3OD) δ 7.67 (bs, 1 H), 7.52 (bs, 1 H), 7.45 (m, 2 H), 7.30-
7.12 (m, 10 H), 6.98 (m, 4 H), 6.10 (bs, 2 H), 5.22 (d, J ) 15
Hz, 1 H), 4.87 (d, J ) 15 Hz, 1 H), 4.11 (d, J ) 15 Hz, 1 H),
Assessment of the HIV Protease as
a Therapeutic Target.
Chimica Oggi 1991, 6-27. (c) Huff, J . R. HIV Protease A Novel
Chemotherapeutic Target for AIDS. J . Med. Chem. 1991, 34,
2305-2314. (d) Martin, J . A. Recent Advances in the Design of
HIV Proteinase Inhibitors. Antiviral Res. 1992, 17, 265-278.
(e) Thaisrivongs, S. HIV Protease Inhibitors. Annu. Rep. Med.
Chem. 1994, 29, 133-144. Kempf, D. J . Progress in the
Discovery of Orally Bioavailable Inhibitors of HIV Protease.
Perspect. Drug Discovery Des. 1994, 2, 427-436. (g) Kempf, D.
J . Design of Symmetry-Based, Peptidomimetic Inhibitors of
Human Immunodeficiency Virus Protease. Methods Enzymol.
1994, 241, 334-354. (h) Vacca, J . P. Design of Tight-Binding
Human Immunodeficiency Virus Type 1 Protease Inhibitors.
Methods Enzymol. 1994, 241, 311-334. (i) Wlodawer, A.
Rational Drug Design: The Proteinase Inhibitors. Pharmaco-
therapy 1994, 14, Suppl.), 9S-20S. (j) Erickson, J .; Kempf, D.
Structure-Based Design of Symmetric Inhibitors of HIV-1 Pro-
tease. Arch. Virol. 1994 (Suppl. 9), 19-29. (k) Darke, P. L.;
Huff, J . R. HIV Protease as an Inhibitor Target for the Treat-
ment of AIDS. Adv. Pharmacol. 1995, 25, 399-454. (l) Rom-
ines, K. R.; Chrusciel, R. A. 4-Hydroxypyrones and Related

General Approach to Nonpeptide HIVPr Inhibitors
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 11 1719
Templates as Nonpeptic HIV Protease Inhibitors. Curr. Med.
Chem. 1995, 2, 825-838. (m) Gait, M. J .; Karn, J . Progress in
Anti-HIV Structure-Based Drug Design. TiBTECH 1995, 13,
430-437. (n) West, M. L.; Fairlie, D. P. Targeting HIV-1
Protease: A Test of Drug-Design Methodologies. TiPS 1995,
16, 67-74. (o) De Clercq, E. Toward Improved Anti-HIV
Chemotherapy: Therapeutic Strategies for Intervention With
HIV Infections. J . Med. Chem. 1995, 38, 2491-2517. (p)
Boehme, R. E.; Borthwick, A. D.; Wyatt, P. G. Antiviral Agents.
Annu. Rep. Med. Chem. 1995, 30, 139-149. (q) Chong, K. T.
Recent Advances in HIV-1 Protease Inhibitors. Exp. Opin.
Invest. Drugs 1996, 5, 115-124. (r) Kempf, D. J .; Sham, H. L.
HIV Protease Inhibitors. Curr. Pharm. Des. 1996, 2, 225-246.
(s) De Lucca, G. V.; Erickson-Viitanen, S.; Lam, P. Y. S. Cyclic
HIV Protease Inhibitors Capable of Displacing the Active Site
Structural Water Molecule. Drug Discovery Today 1997, 2,
6-18.
535-544. (b) Erickson-Viitanen, S.; Klabe, R. M.; Cadwood, P.
G.; O’Neal, P. L.; Meek, J . L. Potency and Selectivity of Inhibition
of Human Immunodeficiency Virus Protease by a Small Non-
peptide Cyclic Urea, DMP323. Antimicrob. Agents Chemother.
1994, 38, 1628-1634.
(10) 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.; Reid, C.; Sharpe, T. R.; Shum, L.;
Winslow, D. L.; Erickson-Viitanen, S. Improved Cyclic Urea
Inhibitors of the HIV-1 Protease: Synthesis, Potency, Resistance
Profile, Human Pharmacokinetics and X-ray Crystal Structure
of DMP450. Chem. Biol. 1996, 3, 301-314.
(11) (a) Kempf, D. J .; Norbeck, D. W.; Codacovi, L. M.; Wang, X. C.;
Kohlbrenner, W. E.; Wideburg, N. E.; Paul, D. A.; Knigge, M.
F.; Vasavanonda, S.; Craig-Kennard, A.; Saldivar, A.; Rosen-
brook, W.; Clement, J . J .; Plattner, J . J .; Erickson, J . Structure-
Based, C2 Symmetric Inhibitors of HIV Protease. J . Med. Chem.
1990, 33, 2687-2689. (b) Kempf, D. J .; Sowin, T. J .; Doherty,
E. M.; Hannick, S. M.; Codavoci, L.; Henry, R. F.; Green, B. E.;
Spanton, S. G.; Norbeck, D. W. Stereocontrolled Synthesis of C2-
Symmetric and Pseudo-C2-Symmetric Diamino Alcohols and
Diols for Use in HIV Protease Inhibitors. J . Org. Chem. 1992,
57, 5692-5700. (c) Kempf, D. J .; Codacovi, L.; Wang, X. C.;
Kohlbrenner, W. E.; Wideburg, N. E.; Saldivar, A.; Vasavanonda,
S.; Marsh, K. C.; Bryant, P.; Sham, H. L.; Green, B. E.;
Betebenner, D. A.; Erickson, J .; Norbeck, D. W. Symmetry-Based
Inhibitors of HIV Protease. Structure-Activity Studies of Acy-
lated 2,4-Diamino-1,5-diphenyl-3-hydroxypentane and 2,5-Di-
amino-1,6-diphenylhexane-3,4-diol. J . Med. Chem. 1993, 36,
320.
(6) (a) Vella, S. HIV Therapy Advances. Uptake on a Proteinase
Inhibitor. AIDS 1994, 8 (Suppl. 3), S25-29. (b) Pollard, R. B.
Use of Proteinase Inhibitors in Clinical Practice. Pharmaco-
therapy 1994, 14 21S-29S. (c) Vacca, J . P.; Dorsey, B. D.;
Schlief, W. A.; Levin, R. B.; McDaniel, S. L.; Darke, P. L.; Zugay,
J .; Quintero, J . C.; Blahy, O. M.; Roth, E.; Sardana, V. V.;
Schlabach, A. J .; Graham, P. I.; Condra, J . H.; Gotlib, L.;
Holloway, M. K.; Lin, J .; Chen, I.-W.; Ostovic, D.; Anderson, P.
S.; Emini, E. A.; Huff, J R. L-735,524; an Orally Bioavailable
Human Immunodeficiency Virus Type 1 Protease Inhibitor.
Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 4096-4100. (d) Wei, X.;
Ghosh, S. K.; Taylor, 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 1 Infection. Nature 1995, 373,
117-122. (e) Kempf, D.; Marsh, K. C.; Denissen, J . F.; Mc-
Donald, E.; Vasavanonda, S.; Flentge, C. A.; Green, B. G.; Fino,
L.; Park, C. H.; Kong, X.-P.; Wideburg, N. E.; Saldivar, A.; Ruiz,
L.; Kati, W. M.; Sham, H. L.; Robins, T.; Stewart, K. D.; Hsu,
A.; Plattner, J . J .; Leonard, J . M.; Norbeck, D. W. ABT-538 is a
Potent Inhibitor of Human Immunodeficiency Virus Protease
with High Oral Bioavailability in Humans. Proc. Natl. Acad.
Sci. U.S.A. 1995, 92, 2484-2488. (f) Ho, D. D.; Neumann, A.
U.; Perelson, A. S.; Chen, W.; Leonard, J . M.; Markowitz, M.
Rapid Turnover of Plasma Virus and CD4 Lymphocytes in HIV-1
Infection. Nature 1995, 373, 123-126. (g) Kitchen, V. S.;
Skinner, C.; Ariyoshi, K.; Lane, E. A.; Duncan, I. B.; Burckhardt,
J .; Burger, H. U.; Bragman, K.; Pinching, A. J .; Weber, J . N.
Safety and Activity of Saquinavir in HIV Infection. Lancet 1995,
345, 952-955.
(12) Nahm, S.; Weinreb, S. M. N-Methoxy-N-Methylamides as Ef-
fective Acylating Agents. Tetrahydron Lett. 1981, 22, 3815-
3818.
(13) The original Abbott procedure11b
requires the addition of vinyl-
magnesium bromide to N-Boc-phenylalaninal to give the allylic
alcohol. However, since we intended to mask the alcohol with
an acid labile protecting group and we desired the ability to
simultaneously unmask the two amino groups later in the
synthesis, we decided to start with N-Cbz-phenylalaninal.
However, in our hands the addition of vinylmagnesium bromide
to N-Cbz-phenylalaninal did not give a clean product. We found
that the three-step procedure shown in Scheme 1 gave the allylic
alcohol 1 much more cleanly.
(14) This is similar to the ratio reported by the Abbott group.11b
(15) Greenberg, S.; Moffatt, J . G. Reactions of 2-Acyloxyisobutyryl
Halides with Nucleosides. I. Reactions of Model Diols and
Uridine. J . Am. Chem. Soc. 1973, 95, 4016-4025.
(16) Randad, R. S.; Pan, W.; Gulnik, S. V.; Burt, S.; Erickson, J . W.
De Novo Design of Nonpeptide HIV-1 Protease Inhibitors:
Incorporation of Structural Water. Bioorg. Med. Chem. Lett.
1994, 4, 1247-1252.
(17) J adhav, P. K.; Man, H.-W. Synthesis of 7-Membered Cyclic
Oxamides: Novel HIV-1 Protease Inhibitors. Tetrahedron Lett.
1996, 37, 1153-1156.
(18) Sham, H. L.; Zhao, C.; Stewart, K. D.; Betebenner, D. A.; Lin,
S.; Park, C. H.; Kong, X.-P.; Rosenbrook, W., J r.; Herrin, T.;
Madigan, D.; Vasavanonda, S.; Lyons, N.; Molla, A.; Saldivar,
A.; Marsh, K. C.; McDonald, E.; Wideburg, N. E.; Denissen, J .
F.; Robins, T.; Kempf, D. J .; Plattner, J . J .; Norbeck, D. W. A
Novel Picomolar Inhibitor of Human Immunodeficiency Virus
Type 1 Protease. J . Med. Chem. 1996, 39, 392-397.
(19) Bold, G.; Bhagwat, S. S.; Fassler, A.; Lang, M. Cyclic hydrazine
compounds with anti-retroviral activity. PCT Patent Application
WO9502582-A, 1995.
(20) Hogde, C. N.; Fernandez, C. H.; J adhav, P. K.; Lam, P. Y. S.
Preparation of Substituted Caprolactams and Derivatives Useful
for Treatment of HIV Disease. PCT Patent Application WO
9422840, 1994.
(7) (a) J acobsen, H.; Yasargil, K.; Winslow, D. L.; Craig, J . C.; Krohn,
A.; Duncan, I. B.; Mous, J . Characterization of Human Immu-
nodeficiency Virus Type 1 Mutants with Decreased Sensitivity
to Proteinase Inhibitor Ro 31-8959. Virology 1995, 206, 527-
534. (b) Markowitz, M. M.; Mo, H.; Kempf, D. J .; Norbeck, D.
W.; Bhat, T. N.; Erickson, J . W.; Ho, D. Selection and Analysis
of Human Immunodeficiency Virus Type
1 Variants with
Increased Resistance to ABT-538, a Novel Protease Inhibitor.
J . Virol. 1995, 69, 701-706. (c) Condra, J . H.; Schlelf, W. A.;
Blahy, O. M.; Gabryelski, L. J .; Graham, D. J .; Quintero, J . C.;
Rhodes, A.; Robbins, H. L.; Roth, E.; Shivaprakash, M.; Titus,
D.; Yang, T.; Teppler, 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. Ridky, T.;
Leis, J . Development of Drug Resistance to HIV-1 Protease
Inhibitors. J . Biol. Chem. 1995, 270, 29621-29623.
(8) (a) 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, N. Y.; 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. (b) Nugiel, D. A.; J acobs, K.;
Worley, T.; Patel, M.; Kaltenbach, R. F., III; Meyer, D. T.;
J adhav, P. K.; De Lucca, G. V.; Smyser, T. E.; Klabe, R. M.;
Bacheler, L. T.; Rayner, M. M.; Steiz, S. P. Preparation and
Structure-Activity Relationship of Novel P1/P1′ Substituted
Cyclic Urea-Based Human Immunodeficiency Virus Type-1
Protease Inhibitors. J . Med. Chem. 1996, 39, 2156-2169. (c)
Lam, P. Y. S.; Ru, Y.; J adhav, P. K.; Aldrich, P. E.; De Lucca, 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.
(21) Peyman, A.; Stahl, W.; Wagner, K.; Ruppert, D.; Budt, K.-H.
Non-Peptide-Based Inhibitors of Human Immunodeficiency
Virus-1 Protease. Bioorg. Med. Chem. Lett. 1994, 4, 2601-2604.
(22) 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 Evaluation of Antiviral Efficacy. Antiviral
Chem. Chemother. 1994, 5, 111-121.
(23) Katrizky, A. R.; Pilarski, B.; Urogdi, L. Efficient Conversion of
Nitriles to Amides with Basic Hydrogen Peroxide in Dimethyl
Sulfoxide. Synthesis 1989, 949-950.
(24) Bru¨nger, A. T.; Kuriyan, J .; Karplus, M. Crystallographic R
Factor Refinement by Molecular Dynamics. Science 1987, 235,
458-460.
(25) We thank one of the referees for suggesting that we should point
out that there is an entropic factor also present.
(26) De Lucca, G. V. Synthesis and Evaluation of Delta Lactams as
Nonpeptide HIV-Protease Inhibitors. Bioorg. Med. Chem. Lett.
1997, 7, 501-504.
(9) (a) Wong, Y. N.; Burcham, D. L.; Saxten, 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,
J M970081I