2',6'-Dimethylphenoxyacetyl: A new achiral high affinity P3-P2 ligand for peptidomimetic-based HIV protease inhibitors

Article abstract of DOI:10.1021/jm990336n

Starting from palinavir (1), our lead HIV protease inhibitor, we have discovered a new series of truncated analogues in which the P₃-P₂ quinaldic-valine portion of 1 was replaced by 2',6'-dimethylphenoxyacetyl. With EC₅₀'s in the 1-2 nM range, some of these compounds are among the most potent inhibitors of HIV replication in vitro, reported to date. One of the most promising members in this series (compound 27, BILA 2185 BS) exhibited a favorable overall pharmacokinetic profile, with 61% apparent oral bioavailability in rat. X-ray crystal structures and molecular modeling were used to rationalize the high potency resulting from incorporation of this structurally simple, achiral ligand into the P₃-P₂ position of hydroxyethylamine-based HIV protease inhibitors.

Full text of DOI:10.1021/jm990336n

1094
J . Med. Chem. 2000, 43, 1094-1108
2′,6′-Dim eth ylp h en oxya cetyl: A New Ach ir a l High Affin ity P 3-P 2 Liga n d for
P ep tid om im etic-Ba sed HIV P r otea se In h ibitor s
Pierre L. Beaulieu,* Paul C. Anderson, Dale R. Cameron, Gilbert Croteau, Vida Gorys, Chantal Grand-Maˆıtre,
Daniel Lamarre, Francine Liard, William Paris, Louis Plamondon,† Franc¸ois Soucy,‡ Diane Thibeault,
Dominik Wernic, and Christiane Yoakim
Boehringer Ingelheim (Canada) Ltd., Bio-Me´ga Research Division, 2100 Cunard Street, Laval (Que´bec), Canada H7S 2G5
Susan Pav and Liang Tong§
Boehringer Ingelheim Pharmaceuticals Inc., 175 Briar Ridge Road, Ridgefield, Connecticut 06877
Received J une 30, 1999
Starting from palinavir (1), our lead HIV protease inhibitor, we have discovered a new series
of truncated analogues in which the P3-P2 quinaldic-valine portion of 1 was replaced by 2′,6′-
dimethylphenoxyacetyl. With EC50’s in the 1-2 nM range, some of these compounds are among
the most potent inhibitors of HIV replication in vitro, reported to date. One of the most
promising members in this series (compound 27, BILA 2185 BS) exhibited a favorable overall
pharmacokinetic profile, with 61% apparent oral bioavailability in rat. X-ray crystal structures
and molecular modeling were used to rationalize the high potency resulting from incorporation
of this structurally simple, achiral ligand into the P3-P2 position of hydroxyethylamine-based
HIV protease inhibitors.
In tr od u ction
Inhibition of the virally encoded protease of the
human immunodeficiency virus (HIV) results in the
production of immature, noninfectious virions.1 The
discovery has led to new strategies for the treatment of
AIDS, based on the inhibition of this essential viral
protein. Most successful to date in this area is the use
of peptidomimetic structures that resemble the transi-
tion state for the cleavage of the enzyme’s natural
substrates.2 Recently, several HIV protease inhibitors
have received FDA approval and are providing clinicians
with new tools for combating the deadly infection.3
Our own efforts in this field have culminated with the
discovery of palinavir (1, Figure 1), a potent inhibitor
S
of the protease and viral replication in vitro.4 This
peptidomimetic structure incorporates a hydroxyethy-
lamine transition state mimic and a novel 4-hydroxy-
pipecolic acid fragment which spans the S1′-S3′ pockets
of the protease active site.5 While palinavir exhibited a
good pharmacological profile in several laboratory ani-
mal species, we felt that it could still benefit from
further simplification and size reduction.6 SAR studies
leading to the discovery of palinavir had revealed that
manipulation of the right-hand side of the molecule (the
4-substituent on the pipecolic ring) could be used to
modulate the overall physicochemical properties of the
inhibitor while maintaining antiviral properties.4 In
designing simpler and smaller versions of 1, we wanted
to retain this very useful feature of this class of
inhibitors, and our efforts thus concentrated on modi-
F igu r e 1.
fications of the left-hand side of the molecule (i.e. the
P3-P2 segment). We describe herein the discovery of 2′,6′-
dimethylphenoxyacetyl, a new achiral, high affinity P3-
P2 ligand for peptidomimetic-based HIV protease
inhibitors.4a,b
Resu lts a n d Discu ssion
Early efforts in the field of HIV protease inhibitors
resulted in the incorporation of rather large peptide-
like fragments in the P3-P2 positions, often resulting in
a disappointing pharmacokinetic profile.7 As a result,
considerable efforts have been directed at the design of
smaller, less peptidic P3-P2 replacements, some of which
are shown in Figure 2. In many instances, substantial
increases in potency were achieved through the design
of ligands which were able to take advantage of hydro-
phobic (Val32, Ile50, Ile84) as well as hydrogen-bonding
(Asp29, Asp30) interactions with the enzyme active site.
Most notable are the mono- (2) and bicyclic tetrahydro-
* E-mail: pbeaulieu@lav.boehringer-ingelheim.com.
† Present address: BASF Corp., Biomedical Research Center,
Worcester, MA.
‡ Present address: Millennium Pharmaceuticals Inc., 75 Sidney
Street, Cambridge, MA 02139.
§ Present address: Department of Biological Sciences, Columbia
University, 1212 Amsterdam Avenue, New York, NY 10027.
10.1021/jm990336n CCC: $19.00 © 2000 American Chemical Society
Published on Web 02/26/2000

A New P3-P2 Ligand for HIV Protease Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6 1095
Ta ble 1
F igu r e 2.
furanyl (3) groups, which, through a combination of
conformational restriction, hydrogen-bonding function-
ality, and molecular weight reduction, conferred an
improved antiviral profile to several classes of HIV
protease inhibitors.8 Along the same lines, cyclic (4, 5)8a
as well as acyclic sulfones (6)9 were designed for
improved potency. While in several instances these
P3-P2 replacements satisfied requirements of improved
antiviral activity, reduced molecular weight, and superior
pharmacokinetic profiles, they often resulted in increased
structural complexity relative to the readily available
amino acid residues they were designed to replace. One
notable exception is the 3-hydroxy-2-methylbenzoyl
moiety (7) found in inhibitors such as Viracept (nelfi-
navir, AG 1343)10a,b and the closely related LY316340,10c
and some C2 symmetric inhibitors.10d A common feature
of P3-P2 ligands introduced to date is the incorporation
of a carbonyl group for linkage to P1 (via amide or
carbamate functions) to ensure conservation of a hy-
drogen bond with the structural water (W301) that
bridges inhibitors to the flaps of the enzyme.2,5
Discover y of a New P 3-P 2 Liga n d . During SAR
studies leading to the discovery of palinavir, we had
noticed that synthetic intermediates spanning P1 to P3′
and capped at the N-terminal with simple carbamate
protecting groups, such as carbobenzyloxycarbonyl (Cbz,
8) and tert-butyloxycarbonyl (Boc, 9), produced modest
inhibitors of the enzyme. The binding affinity of these
compounds could be improved by optimization of the P1′-
P3′ enzyme interactions.4c For example, replacement of
the quinaldic-valine fragment in 1 by Cbz gave com-
pound 10 (Table 1), which, when tested in an HIV-1
protease HPLC-based assay (see Experimental Section
for details), had an IC50 of 400 nM. This 100-fold drop
in potency compared to 1 is not surprising considering
that two critical hydrogen bonds (Val NH to G48′ and
quinaldyl CO to D29′),5 which certainly contribute to
the high potency of 1, have been sacrificed in 10.
Nevertheless, we thought that 10 could serve as a
starting point for the design of new P3-P2 ligands. Our
objectives were to recover and perhaps improve potency
relative to 1, establish a favorable pharmacokinetic
profile, and maintain molecular weight and structural
complexity at a minimum. Table 1 summarizes our
initial attempts at improving the potency of 10.
Hoping to reestablish a hydrogen bond with residues
present in the S2 subsite of the enzyme, the alkoxy
oxygen atom and methylene group in 10 were inter-
changed, leading to phenoxyacetyl derivative 11 suffer-
ing an additional 3-fold decrease in potency (IC50 ) 1100
nM). Replacement of the same oxygen atom in 10 by a
methylene group gave 12 (IC50 ) 1400 nM), equipotent
with 11. The lower potency of amide derivatives 11 and
12 relative to carbamate 10 could be the consequence
of an inductive decrease in basicity of the carbonyl
oxygen, resulting in a weaker hydrogen bond with
W301. This hydrogen bond is important for bringing the
enzyme flaps down onto inhibitors through a water-
mediated hydrogen bond to protein flap residues, ef-
fectively locking them into place inside the active site.
Replacement of the ether linkage of 11 with a thio-
ether function (compound 13) or incorporation of an
additional methylene group in the chain (compound 14)
gave analogues that did not significantly inhibit enzy-
matic activity (IC50 g 4 µM). The increased length of
the chain linking the ring to the carbonyl group pre-
sumably does not allow the aromatic portion to settle
into the S2 pocket, and considerable interference occurs
between the inhibitor and residues in the enzyme active
site. Similarly, cinnamic derivative 15 forces the N-
terminal group into a rigid linear arrangement, which
appears to be detrimental to binding with the protease.
We conclude that the modest activity seen in the case
of 10 is most likely due to weak lipophilic interactions
between the phenyl ring and the S2 pocket and to a
strong hydrogen bond interaction between the carbam-
ate carbonyl and structural W301. These interactions

1096 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6
Beaulieu et al.
Ta ble 2
methyl group is producing beneficial effects that over-
ride the otherwise less efficient binding of amide ligands
relative to carbamates (see Table 1). Encouraged by
these results, we investigated the effect of adding a
second ortho-methyl group to inhibitors 16-18. The
2′,6′-dimethyl-Cbz derivative 19 (IC50 ) 420 nM) suf-
fered a 2-fold loss in potency relative to monosubstituted
carbamate 16, canceling out the beneficial effect of
adding one methyl group (compound 19 now has a
similar activity to unsubstituted 10). Adding a second
ortho-methyl group to phenylpropionic amide 17 had no
additional effect on potency (IC50 ) 2500 and 2400 nM
for 17 and 20, respectively). In the case of phenoxyacetyl
amide 18, however, the presence of an additional ortho-
methyl group resulted in a 15-fold increase in potency
(from 120 nM to 7 nM for inhibitor 21). Compared to
unsubstituted derivative 11, this represents a dramatic
150-fold increase in potency (1100 nM to 7 nM) through
the simple addition of two methyl groups on the
molecule (see also X-ray and modeling section). Based
on enzymatic inhibitory values, 2′,6′-dimethylphenoxy-
acetamide 21 showed similar potency to palinavir 1 (IC50
) 7.0 nM and 4.0 nM, respectively).
In an attempt to further improve on the potency of
21, phenoxypropionamide 22 was prepared as a 1:1
mixture of isomers. Orientation of the ligand through
introduction of an asymmetric center (mimicking a P2
amino acid residue) does not provide increased affinity
(IC50 ) 20 nM for the 1:1 mixture of isomers vs 7.0 nM
for 21). A large number of ring-substituted phenoxy-
acetyl amides, structurally related to 21, were also
evaluated for their ability to inhibit the enzymatic
activity of the protease (results not shown). These
included a variety of ring-polymethylated and haloge-
nated derivatives and a combination thereof. While
some compounds compared favorably, none proved
superior to the original 2′,6′-dimethylphenoxyacetyl
derivative 21. Larger alkyl groups (Et, iPr, nPr) were
also incorporated in the ortho-positions of the phenol
ring, in an attempt to further optimize interactions with
the S2 pocket, but without success.
require flexibility in the linker portion of the ligand to
allow proper positioning.
Conformational restriction has proven successful in
the design of high affinity P2 ligands. For example,
rigidification through incorporation into cyclic systems8
as well as hindered rotational mobility10 are but two
concepts which have allowed the rigidification of P2
ligands into arrangements that more closely resemble
a presumed bioactive conformation. Lowering of the
activation energy required by a peptidomimetic inhibitor
to adopt a bioactive conformation can, in some cases,
result in dramatic increases in potency.11
It was envisaged that incorporation of methyl groups
in the ortho-positions of the N-terminal aromatic ring
of compounds 10-12 might force the otherwise flexible
ligands into a more rigid conformation that would lie
closer in energy to the actual bioactive conformation.
The added methyl groups could also provide some
additional lipophilic interactions within the S2 pocket
of the enzyme, resulting in an increase in potency. In
the event, attachment of a single methyl group to Cbz
derivative 10 resulted in a 2-fold increase in potency
(inhibitor 16, IC50 ) 210 nM, Table 2). In contrast, a
similar modification of phenylpropionic amide 12 re-
sulted in a 2-fold decrease in potency (inhibitor 17, IC50
) 2500 nM). Much more significant, however, was the
effect on phenoxyamide 11, resulting in inhibitor 18
with a 10-fold improvement in potency (IC50 ) 120 nM
compared to 1100 nM for 11). In this last case, the added
Cell Cu ltu r e Activity. Inhibition of HIV-1 replica-
tion in acutely infected C8166 T cells was examined next
(see Experimental Section for details). Compound 21
inhibited replication of the IIIb strain of the virus at
an effective concentration (EC50) of 24 nM (Table 3). No
apparent cytotoxicity was noticed in this cell line at
inhibitor concentrations up to 10 µM. Relative to pali-
navir 1 (EC50 ) 5 nM), 2′,6′-dimethylphenoxyacetyl is
a good replacement for the bulkier N-quinaldylvaline
moiety of this class of inhibitors. To compare our novel
P2-P3 ligand with those found in other hydroxyethyl-
amine-based HIV protease inhibitors, we also prepared
compounds 23 and 24, featuring both the pipecolic P2′-
P
3′ portion of our molecules, and previously reported P2
replacements. The hydroxytoluamide moiety (compound
23) found in Viracept8a conferred similar antiviral
properties to our compounds (compare 21 and 23).
Tetrahydrofuryl carbamate8b 24, on the other hand, was
found significantly less potent, indicating suboptimal
compatibility with our pipecolic amide class of inhibi-
tors.
Op tim iza tion of th e P 3′ P ip ecolic Su bstitu en t. As
mentioned previously, the 4-substituent on the pipecolic

A New P3-P2 Ligand for HIV Protease Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6 1097
Ta ble 3
Ta ble 4
Ta ble 5
ring of our inhibitors can be used to modulate the
physicochemical properties of our compounds, while
maintaining inhibitory potencies against the protease.
This has been rationalized by the fact that most of this
substituent lies outside of the enzyme active site and is
mainly exposed to solvent.5 This feature was used to
successfully improve the pharmacokinetic properties of
our previous series of inhibitors, eventually leading to
the discovery of palinavir.4a,c
During our search for structurally novel P3-P2 re-
placements, all studies were carried out on pipecolic
amide derivatives bearing the 4-picolyl substituent
found in our lead compound, palinavir. Further refine-
ment of the new dimethylphenoxyacetamide series,
described herein, was now directed to optimization of
the 4-substituent of the pipecolic ring. A variety of
heterocyclic substituents were introduced at the 4-posi-
tion of the pipecolic ring of compound 21 in place of the
original 4-picolyl moiety. These were linked to the six-
membered ring through ether or thioether linkages and
included mostly pyridyl, picolyl, and pyrimidyl deriva-
tives. As expected, most derivatives gave IC50 values in
the 1-10 nM range. The best compounds in terms of
antiviral properties are shown in Table 4. Replacement
of the ether linkage in 21 by a thioether bond (com-
pounds 25 and 26) resulted in a 4-6-fold improvement
in EC50. Furthermore, excision of the methylene group
in these derivatives gave mercaptopyridine (e.g. 27) and
mercaptopyrimidine (e.g. 28) derivatives with still
superior antiviral properties. To our knowledge, com-
pounds 27 and 28, with EC50 values in the 1-2 nM
range (no cytotoxicity observed in C8166 T cells at 1
µM), are among the most potent inhibitors of in vitro
HIV-1 replication reported to date that target the
protease enzyme. In addition, the use of the 2′,6′-
dimethylphenoxyacetyl as a P3-P2 ligand provides for
significant reduction in size and structural complexity
as originally planned. We now hoped that these positive
a
Reference compounds (saquinavir and amprenavir) were
synthesized in house and tested under identical assay conditions.
features would translate into favorable pharmacokinetic
properties for this new lead series as represented by
compounds 27 and 28 (see next section).
The compatibility of our novel P3-P2 replacement with
other hydroxyethylamine-based HIV protease inhibitors
was also examined, and the results are presented in
Table 5. Replacement of N-quinaldyl-asparagine in
saquinavir by 2′,6′-dimethylphenoxyacetyl gave com-
pound 29 with an IC50 ) 8.4 nM and EC50 ) 100 nM
(compared to EC50 ) 5 nM for saquinavir). Our new P3-
P2 replacement is thus less effective when combined to
the hexahydroisoquinoline amide found in the Roche

1098 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6
Beaulieu et al.
F igu r e 3. Palinavir (1) in green is superimposed with BILA
2185 BS (27) in yellow as bound to HIV-2 protease. The
superimposition was generated by overlapping the protein
residues from each protease-ligand complex crystal structure
(overall rmsd fit of backbone residues was 0.29 Å). Protein is
not shown. The important flap water is indicated in red
(W301).
F igu r e 4. Palinavir (1, green) and BILA 2185 BS (27, yellow)
are shown superimposed as in Figure 3. The protein (from the
crystal structure in complex with 27) is represented by a
solvent accessible Connolly surface. The residues are labeled
by single letter amino acid codes. A or B before the residue
number indicates which monomer of HIV protease the residue
belongs to.
inhibitor.12 When incorporated in place of the tetrahy-
drofuryl carbamate of amprenavir, however,8b the re-
sulting compound (30) retained most of the antiviral
potency of the original inhibitor (20 nM for 30 vs 7 nM
for amprenavir). These observations are evidence for
cooperative binding of functionalities throughout the
HIV protease active site. Recent literature reports
suggest that 2′,6′-dimethylphenoxyacetyl is also a useful
P3-P2 replacement in the context of other classes of HIV
protease inhibitors. For example, its incorporation into
hydroxyethylene and hydroxymethylcarbonyl peptido-
mimetic inhibitors (such as ritonavir13a and KNI-27213b
)
F igu r e 5.
has led to very potent analogues, allowing at the same
time for significant reduction in size and structural
complexity.
methyl groups for a favorable interaction with the S2
pocket. One can also notice that a methyl group is
probably optimal in size and that larger substituents
in the 2′,6′-positions of the phenyl ring might result in
unfavorable contacts with protein residues (Et and iPr
substituents were not tolerated in those positions).
The effect of added methyl groups on benzyloxycar-
bonyl versus phenylpropionamide versus phenoxyacetyl
derivatives is more difficult to rationalize based on the
crystal structures alone. The presence of an additional
interaction between one of the methyl groups and the
S2 pocket as seen in the X-ray structures (Figure 4),
alone, cannot explain the 150-fold difference in potency
seen between 11 and 21. This is especially evident when
taking into consideration the effect of similar modifica-
tions on carbamate (10, 19) and propionamide (12, 20)
derivatives (Tables 1 and 2). This would suggest that
additional factors must be taken into consideration to
account for the significant boost in potency achieved
with dimethylphenoxyacetyl derivatives.
To probe this phenomenon, we undertook conforma-
tional studies on model fragments (N-methyl amides I,
II, and III, depicted in Figure 5; see Experimental
Section for details on computational methods used)
representing the left-hand portion of inhibitors 19, 20,
and 21. Each fragment was minimized, and energies
were compared to calculated crystal structure positions
on a rotational potential energy plot (Figure 6). The
X-ray conformation for 19 (fragment I) is predicted to
be high in energy and away from the global minimum
by a rotation of 180° about the R-angle. Both 20 and 21
are fairly close to the global minimum (1.1 and 1.3 kcal/
X-r a y Cr ysta l Str u ctu r e a n d Molecu la r Mod el-
in g. Insight into the binding mode of this new series of
inhibitors was gained with the help of a crystal struc-
ture of compound 27 in complex with HIV-2 protease.5,14
Figure 3 shows a comparison between the bound con-
formations of 27 (which is representative of the binding
mode of this class of inhibitors) and palinavir (1).5 A
good overlap is seen between the two structures in the
regions spanning the P1 to P3′ positions. In particular,
the critical transition state mimic hydroxyl groups and
the P2-P1 carbonyls show similar interactions with the
protein active site aspartyl residues and W301, respec-
tively.
Figure 4 shows the dimethylphenoxyacetyl group
lying in a groove between flap residues 48 and 49 and
residues 27-30. In comparison to the palinavir complex,
several protein residues have undergone noticeable
movements to accommodate the dimethylphenoxy moi-
ety.5 Figure 4 also shows that this fragment does not
reach into the S3 pocket of the enzyme as did the
quinaldyl group of 1. Rather, a portion of the structure
binds into the S2 pocket, occupied by the valine residue
in the case of palinavir. One of the methyl groups on
the phenol ring of 27 binds similarly to one of the valine
methyls in 1. The other methyl does not appear to
interact with protein residues and is exposed to solvent.
The 15-fold increase in potency observed between mono-
and dialkylated derivatives 18 and 21 (Table 2) is
probably, in part, the result of a decrease in binding
entropy for the latter, which always positions one of the

A New P3-P2 Ligand for HIV Protease Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6 1099
F igu r e 6. Conformational PM3-SM3 energy profiles of fragments I, II, and III are plotted against angles R and â while angle
γ was held fixed at 90° (see Figure 5). Energies are shown relative to the global minimum (A), and the surface is truncated at 5
kcal/mol. In the case of I and II, a local minimum (B) was also identified. X indicates the position of the observed crystal structure
on this energy plot. The relative energies of each conformer (A, B, and X) following minimization are shown in kcal/mol followed
by the final values for R and â in parentheses. The structures shown for I, II, and III depict the lowest energy conformation (A)
for each fragment for comparison with the crystal structure position (X).
Ta ble 6
Ta ble 7
a
b
For a po dose of 5 mg/kg. After oral administration. c For an
d
iv dose of 1 mg/kg. Apparent bioavailability for an iv dose of 1
mg/kg.15
provided an explanation for the 5-fold difference seen
between compounds 19 and 20.
mol for fragments II and III, respectively). Compound
21, however, is contained in a much larger well and
shows much less rotation about R. It is expected that
20 will undergo a larger entropic penalty upon binding
than 21. The bound conformation of 21 is the closest of
the three to its own solvation phase global minimum
as predicted by PM3-SM3 calculations.15 After consid-
eration of protein-ligand interactions, ligand-water
interactions, ligand conformational effects and electro-
static solvation energies, 21 is predicted to be bound
more tightly than either 19 or 20 by at least 3 kcal/mol
(Table 6). This amount of energy would lead to a
difference in Ki of approximately 170-fold between 21
and 19 or 20. Actual IC50 values indicate a 60-fold
difference for 19 and 21 and 335-fold for 20 and 21.
Bioavailability Stu dies.16 Compounds 25-28 (Table
7) were evaluated for oral bioavailability in rat at a po
dose of 5 mg/kg (formulated in methyl cellulose/Tween-
80, 0.5/0.01% in water) and an iv dose of 1 mg/kg
(formulated in DMA/NaH2PO4 0.04 M in 10% dextrose/
water: 30/50/20).16 Pharmacokinetic parameters are
shown in Table 7. All compounds were able to achieve
plasma concentrations (Cmax) of 800-2300 nM, after oral
administration, several orders of magnitude in excess
of their respective EC50 values. AUC’s after oral admin-
istration at 5 mg/kg were 1700-3200 nM h. Half-lives
for the clearance of the compounds (T1/2) ranged from
0.4 to 1.7 h as determined after iv administration of 1
mg/kg. Analogue 27, however, exhibited a superior
apparent bioavailability of 61%, a 2-3-fold improvement
over that of other compounds investigated.
On computational grounds therefore, the higher af-
finity of dimethylphenoxyacetyl derivatives can be
rationalized by a smaller gap in energy required by such
ligands to adopt the bioactive conformation seen in
X-ray structures of inhibitors such as 27 complexed to
HIV-2 protease. At this stage, our calculations have not
Biologica l P r ofile of Com p ou n d 27. Compound 27
(company code BILA 2185 BS) was identified as the
most promising member of this novel series of P3-P2-
truncated HIV protease inhibitors (Table 8). It inhibited

1100 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6
Beaulieu et al.
Ta ble 8
Sch em e 1
Sch em e 2
a
b
HPLC-based peptide substrate cleavage assay. See ref 4a.
both HIV-1 and HIV-2 protease activities with IC50’s )
1.6 nM and 2.3 nM, respectively. The observed inhibi-
tion constants (Ki) for compound 27 were 6 and 23 pM
for the HIV-1 and HIV-2 enzymes, as determined by the
method of Morrison and Walsh.4a,17,18 Its EC50 for
inhibition of the IIIB strain of HIV-1 in cell culture
experiments was found to be 2 nM (3 nM against HIV-2
ROD). In presence of 20% human serum, a 4-fold
increase in EC50 (from 1.3 nM to 5.3 nM) was observed.
This small shift in EC50 was comparable to that found
for palinavir under similar conditions. To our knowl-
edge, this is one of the most potent in vitro anti-HIV
protease agent reported to date. The more than 300-
fold difference between the observed Ki and the EC50
values for compound 27 is probably the result of a
number of factors including protein binding, metabo-
lism, and its ability to enter cells. Furthermore, experi-
mental conditions used in enzymology studies may not
reflect conditions found in the cell (similar observations
were made previously in the case of palinavir4a and
hydroxyethylamide inhibitors19). The apparent bioavail-
ability of 27 in the rat was 61%, and at 5 mg/kg oral
doses, it achieved plasma concentrations in excess of 900
nM with a clearance half-life of 0.4 h.16a The apparent
oral bioavailability of 27 when administered in the dog
was 39% and that in chimpanzee was 5%, for an oral
dose of 10 mg/kg and iv doses of 1 mg/kg and 0.5 mg/
kg, respectively.20a
Sch em e 3
noxyacetyl moiety itself but is probably related to other
features of the molecule.
Syn th esis of In h ibitor s. All inhibitors were pre-
pared by standard peptide coupling methods. The
synthesis of palinavir 1 via intermediate amine 31
(Scheme 1) was reported recently.21 Carbamate deriva-
tives 10, 16, and 19 were prepared by coupling 31 with
the appropriate chloroformate (Scheme 1), prepared in
turn from commercially available alcohols and phos-
gene. Arylpropionyl, aryloxyacetyl amides, and com-
pounds 13-15 were synthesized by coupling 31 with the
corresponding carboxylic acids using TBTU22 in the
presence of N,N-diisopropylethylamine (DIEA) as shown
in Scheme 1. 3-(2′,6′-Dimethylphenyl)propionic acid 32
was prepared by homologation of 2,6-dimethylbenzoic
acid 33 (Scheme 2). Noncommercially available aryloxy-
acetic acids 34a ,b (R ) H, CH3, Scheme 3) were
prepared by alkylation of 2,6-dimethylphenol 35 with
ethylbromoacetate or ethyl 2-bromopropanoate, followed
by saponification of the ester function. Compounds 23
and 24 were prepared by coupling 31 to 3-hydroxy-2-
methylbenzoic acid23 (DCC/HOBt)24 and to the chloro-
formate derived from 3-(S)-hydroxytetrahydrofuran,
respectively.
BILA 2185 BS (27) is a highly potent inhibitor of HIV
protease activity and viral replication in vitro. Com-
pared to palinavir (1), it is characterized by slightly
reduced structural complexity, resulting in a lower
molecular weight (619 vs 719 for palinavir), better or
equal potency, and a favorable overall pharmacokinetic
profile in several laboratory animal species. BILA 2185
BS was shown to induce multiple mutations in the
HIV-1 protease gene, as well as in gag precursor
cleavage sites, leading to drug resistance in vitro. The
observed variants exhibit decreased viral fitness, sug-
gesting that antiviral therapies using such inhibitors
may maintain some clinical benefits.20b,c Unfortunately,
in preliminary animal toxicology studies, BILA 2185 BS
caused significant reductions in heart rates and blood
pressure, which prevented further development of this
compound. However, recent reports on the development
of ABT-378,13a now undergoing phase III clinical evalu-
ation, would suggest that the toxicity seen with BILA
2185 BS may not be caused by the 2′,6′-dimethylphe-
Inhibitors with modifications at the 4-position of the
pipecolic ring (compounds 25-28) were prepared form
the corresponding pipecolic amide derivatives as shown
in Schemes 4 and 5. N-Boc-(2S,4R)-4-hydroxypipecolic
tert-butylamide (36)21,25 was first converted to the
(2S,4S)-4-iodo or 4-bromo derivatives (37) via mesyla-
tion and displacement with inversion using iodide or
bromide. Intermediate 37 was then treated with sulfur

A New P3-P2 Ligand for HIV Protease Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6 1101
Sch em e 4
Sch em e 6
Sch em e 7
Sch em e 5
nearest 0.5 Hz. Mass spectral data was obtained in FAB mode
on a MF 50 TATC instrument operating at 6 kV and 1 mA
using thioglycerol or nitrobenzyl alcohol (NBA) as a matrix
support.
HPLC homogeneities were determined under reversed-
phase conditions. System A: Waters (Milford, MA) Symmetry
C8 column (3.9 × 150 mm), 10 f 80% CH3CN in 50 mM NaH2-
PO4 (pH 4.4) in 25 min then 5 min isocratic, 1.0 mL/min flow
rate, UV detection at 215 nm. System B: Waters (Milford, MA)
Deltapack C18 column (2 × 150 mm), 0 f 100% CH3CN (0.1%
TFA) in 0.1% TFA in 25 min then 5 min isocratic, 0.5 mL/min
flow rate, UV detection at 215 nm.
nucleophiles to deliver the desired (2S,4R)-4-substituted
pipecolic amide derivatives 38a -c through an overall
double-inversion process. Compound 38d was prepared
from 37 (X ) I) by conversion to its 4-mercapto deriva-
tive and alkylation with 4-chloro-2,6-dimethylpyrimi-
dine. N-Deprotection of 38 gave amines 39 (Scheme 5),
which were then used to open epoxide 4026 to give
protected amino alcohol derivatives 41. Following depro-
tection, amino alcohols 42 were obtained that were
condensed with 2′,6′-dimethylphenoxyacetic acid in the
usual fashion.
Flash chromatography was performed on Merck silica gel
60 (0.040-0.063 mm). Analytical thin-layer chromatography
(TLC) was carried out on precoated (0.25 mm) Merck silica
gel F-254 plates.
Ma ter ia ls. Reagents and solvents were of commercial
sources and used as received without purification. Benzotria-
zol-1-yl-1,1,3,3-tetramethyluronium tetrafluoroborate (TB-
TU)22 and BOP coupling reagent were purchased from Rich-
elieu Biotechnologies Inc. (Montre´al, Canada).
The decapeptide substrate VSFNFPQITL-NH2 was synthe-
sized using standard solid-phase methodology. The fluorogenic
substrate, 2-aminobenzoyl-Thr-Ile-Nle-Phe(p-NO2)-Gln-Arg-
NH2, was purchased from Bachem. Protease-free BSA, MES,
and EDTA were purchased from Sigma.
IC50 Det er m in a t ion s Usin g a n HIV P r ot ea se HP LC-
Ba sed Assa y.28 Recombinant HIV-1 and HIV-2 proteases were
expressed from Escherichia coli and purified as previously
described. Enzymatic assays were performed in 100 mM MES,
300 mM KCl, 5 mM EDTA, 1 mg/mL BSA, pH 5.5. DMSO
(final concentration was brought to 4.5%) was used to aid
solubilization. To 20 µL of a 400 µM solution of VSFNFPQITL-
Compound 29 was prepared by condensation of ep-
oxide 4026 with amine 4327 to give amino alcohol 44
which was elaborated to 29 in the usual way (Scheme
6). Analogue 30 was prepared by condensation of
epoxide 4026 with isobutylamine followed by treatment
with 4-nitrobenzesulfonyl chloride, to give sulfonamide
45 (Scheme 7). Deprotection followed by the usual
coupling to P3-P2 gave nitro derivative 46, which gave
aniline 30 upon hydrogenolysis.
Exp er im en ta l Section
Gen er a l. NMR spectra were recorded on an AMX400
spectrometer (400 MHz for 1H and 100 MHz for 13C unless
otherwise specified) in deuterated solvents and were referenced
to TMS (δ scale). J coupling constants are reported to the

1102 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6
Beaulieu et al.
NH2 substrate were added 10 µL of various concentrations of
) 10, 4 Hz, 1H), 3.22 (broad d, J ) 12 Hz, 1H), 2.97 (dd, J )
14, 3 Hz, 1H), 2.72-2.58 (m, 3H), 2.13 (dd, J ) 13, 6.5 Hz,
1H), 2.11-1.98 (m, 2H), 1.93 (broad d, J ) 11 Hz, 1H), 1.52
(q, J ) 11.5 Hz, 1H), 1.46 (m, 1H), 1.27 (s, 9H). 13C NMR
(DMSO-d6) δ 171.4, 155.8, 149.4, 148.1, 139.8, 137.2, 129.2,
128.1, 127.8, 127.4, 127.1, 125.5, 121.6, 75.0, 70.3, 67.1, 66.8,
64.7, 57.3, 56.1, 50.0, 49.9, 34.8, 34.3, 30.1, 28.3. HRMS (FAB)
m/z C34H45N4O5 (MH+) calcd: 589.3390; found: 589.3406.
HPLC homogeneity: 94.1% (system A); 94.8% (system B).
In h ibitor 16: 2-methylbenzyl alcohol (0.610 g, 5.0 mmol)
was added to an ice-cold solution of phosgene in toluene (1.8
M, 5 mL, 9.0 mmol). The mixture was stirred 30 min in the
cold and then another 3 h at room temperature. Volatiles were
removed under reduced pressure, and the residual chlorofor-
mate was used without purification to prepare the carbamate
the inhibitor tested. The enzymatic reaction was initiated by
the addition of 12 nM of enzyme in 10 µL for
a final
concentration of 3 nM. The assay mixture was then incubated
for 45 (HIV-2 protease assay) or 60 (HIV-1 protease assay) min
at 37 °C and the reaction terminated by addition of 200 µL of
1%TFA/H2O. Cleavage products and substrate were separated
by reversed-phase HPLC on a Perkin-Elmer 3 × 3CR C8
column as previously described.28a A nonlinear curve fit using
the Hill model was applied to the percent inhibition-
concentration data, and 50% effective concentrations (IC50
)
were calculated through the use of SAS (Statistical Software
System, SAS Institute Inc., Cary, NC).
Extr acellu lar p24 Assay.In h ibition ofVir al Replication .4a
The extracellular core protein level, determined by ELISA with
the HIV-1 p24 antigen (Ag) Coulter kits, was used to monitor
the replication of the HIV-1 IIIB strain. C8166 T cells were
infected at a MOI of 0.001 in complete medium and were
incubated in the presence of serially diluted inhibitors for 3
days. The percentage of inhibition was obtained by determi-
nation of the core protein level from pools of supernatants (8
replicas). A nonlinear curve fit using the Hill model was
applied to the percent inhibition-concentration data, and
EC50’s were obtained through the use of the SAS software. No
apparent cytotoxicity was noticed in this cell line at concentra-
tions up to 1 µM (up to 10 µM for less active compounds).
according to the general procedure. Rf 0.47 (9:1 CHCl3/MeOH).
1
[R]23 -28.2° (c 1.0, MeOH). H NMR (DMSO-d6) δ 8.52 (d, J
D
) 5.5 Hz, 2H), 7.36 (s, 1H), 7.32 (d, J ) 5.5 Hz, 2H), 7.29-
7.08 (m, 11H), 4.91 (d, J ) 13 Hz, 1H, part of AB), 4.89 (d, J
) 7 Hz, 1H), 4.83 (d, J ) 13 Hz, 1H, part of AB), 4.60 (d, J )
14 Hz, 1H, part of AB), 4.55 (d, J ) 14 Hz, 1H, part of AB),
3.75 (m, 1H), 3.71 (m, 1H), 3.39 (m, 1H), 3.21 (broad d, J ) 12
Hz, 1H), 2.95 (dd, J ) 13.5, 2 Hz, 1H), 2.70-2.57 (m, 3H), 2.19
(s, 3H), 2.16-1.97 (m, 3H), 1.92 (broad d, J ) 11 Hz, 1H), 1.50
(q, J ) 11.5 Hz, 1H), 1.44 (m, 1H), 1.25 (s, 9H). 13C NMR
(DMSO-d6) δ 171.4, 155.7, 149.4, 148.1, 139.8, 135.9, 135.0,
129.8, 129.2, 127.9, 127.8, 127.6, 125.6, 121.6, 75.0, 70.3, 67.1,
66.7, 63.3, 57.2, 56.0, 49.9, 34.8, 34.3, 30.1, 28.3, 18.2. HRMS
(FAB) m/z C35H47N4O5 (MH+) calcd: 603.3547; found: 603.3528.
HPLC homogeneity: 93.2% (system A); 95.8% (system B).
Com p u t a t ion a l Ch em ist r y Met h od s. Small fragments
(Figure 5, I, II, and III) representing the left-hand portion of
our inhibitors were utilized in quantum mechanics calculations
to determine the rotational minima on each potential energy
surface. Semiempirical calculations using Spartan 5.0 (Wave
function Inc. 18401 Von Karman Ave., suite 370, Irvine, CA,
92612) were performed using the PM3 Hamiltonian.15a Three
angles were restrained while the remaining parameters were
allowed to relax. Angle γ (Figure 5) was held at 90° throughout
(close to X-ray position). Angles R and â were varied by 15°
and 30°, respectively, to generate 288 relaxed points for each
system. Symmetry was exploited when possible.
Each relaxed point was then treated with the SM3 solvation
method15b as a single point calculation, and the energy was
plotted as a function of angles R and â. Each predicted
minimum was investigated by optimization without restraints
as above. The X-ray conformation was determined by freezing
angles R, â, and γ at 350°, 159°, and 87°, respectively,5 followed
by optimization as above.
Preliminary protein-inhibitor complex calculations de-
scribed in the text were performed using the Discover 95.5 and
the CFF95. The crystal structure position for 275 was used as
a starting point, followed by a series of tethered minimizations
to relieve any bad contacts from the initial placement. Crys-
tallographically determined waters were included with a
dielectric constant of 1.0. Interaction energies were computed
and utilized as a crude measure of protein-inhibitor interac-
tion. Delphi was also used to gain an estimate of the electro-
static portion of the solvation energy (finite difference Poisson-
Boltzmann methodology) for each of the ligands, free and in
their bound complexes (Molecular Simulations Inc., 9685
Scranton Road, San Diego, CA 92121-3752).
Gen er a l P r oced u r e for th e P r ep a r a tion of Ca r ba m a te
In h ibitor s 10, 16, a n d 19: amine 31 (1.364 g, 3.0 mmol) and
DIEA (1.05 mL, 6.0 mmol) were dissolved in CH2Cl2 (35 mL)
and the solution was cooled in an ice bath. The chloroformate
(4-5 mmol) was added and the mixture was stirred 1-2 h.
The reaction mixture was then poured into water, washed with
water, dried (MgSO4), and concentrated under reduced pres-
sure to give an oil that was purified by flash chromatography
using 95:5 CHCl3/MeOH as eluent.
In h ibitor 19: 2,6-dimethylbenzoic acid 33 (5.00 g, 33 mmol)
was added in small portions to a suspension of LiAlH4 (1.14
g, 30 mmol) in dry THF (100 mL) under an argon atmosphere.
The mixture was refluxed for 3 days, cooled to room temper-
ature, and quenched by dropwise addition of 10% HCl (100
mL). After extraction with EtOAc, drying (MgSO4), and
removal of volatiles under reduced pressure, crude 2,6-di-
methylbenzyl alcohol was obtained as a white solid (4.30 g,
94% yield) that was used without purification: 1H NMR (200
MHz, CDCl3) δ 7.2-6.9 (m, 3H), 4.70 (s, 2H), 4.05 (broad s,
1H), 2.40 (s, 6H). Following the procedure described above for
16, the alcohol was converted to its chloroformate and used to
prepare carbamate 19 according to the general procedure.
Inhibitor 19 was purified by reversed-phase chromatography
using 0.06% TFA/0.06% TFA in CH3CN gradients and isolated
by lyophilization. Rf 0.44 (95:5 CHCl3/MeOH). [R]23 -2.5° (c
D
1.0, MeOH). 1H NMR (DMSO-d6) δ 9.80 (broad s, 1H), 8.66
(broad d, J ) 4.5 Hz, 2H), 8.26 (broad s, 1H), 7.54 (d, J ) 5.5
Hz, 2H), 7.23-7.11 (m, 5H), 7.08 (t, J ) 7 Hz, 2H), 6.97 (d, J
) 7.5 Hz, 2H), 4.92 (d, J ) 12 Hz, 1H, part of AB), 4.78 (d, J
) 12 Hz, 1H, part of AB), 4.69 (m, 2H), 3.90 (m, 2H), 3.16 (m,
2H), 2.97 (broad d, J ) 13.5 Hz, 1H), 2.90 (broad t, J ) 10.5
Hz, 1H), 2.42 (broad d, J ) 11.5 Hz, 1H), 2.24 (m, 1H), 2.19
(s, 6H), 1.80 (q, J ) 13 Hz, 1H), 1.66 (q, J ) 11.5 Hz, 1H),
1.28 (s, 9H). 13C NMR (DMSO-d6) δ 166.2, 156.2, 147.0, 138.7,
137.6, 132.4, 128.9, 128.2, 127.9, 127.8, 125.9, 122.4, 71.8, 70.0,
67.4, 65.3, 60.3, 58.1, 56.4, 51.4, 51.0, 35.6, 33.9, 28.4, 28.0,
19.0. HRMS (FAB) m/z C36H48N4O5 (MH+) calcd: 617.3703;
found: 617.3690. HPLC homogeneity: 98.5% (system A);
98.6% (system B).
Gen er a l P r oced u r e for Cou p lin g of Am in e 31 w ith
Ca r boxylic Acid s: amine 31 (1 equiv), the carboxylic acid
(1.2-1.3 equiv) and DIEA (3-4 equiv) were dissolved in CH3-
CN, and TBTU22 (1.3-1.5 equiv) was added. The mixture was
stirred for 1-2 h at room temperature. The solvent was
removed under reduced pressure, the residue dissolved in
EtOAc, and the solution washed twice with 2 N NaOH and
water. After drying (MgSO4), the solvent was evaporated under
reduced pressure and the residue purified by flash chroma-
tography using 95:5 CHCl3/MeOH as eluent, to give the pure
inhibitors as off-white foams.
In h ibitor 10: prepared using benzyl chloroformate. Rf 0.36
(95:5 CHCl3/MeOH). [R]23 -32.8° (c 1.0, MeOH). 1H NMR
D
(DMSO-d6) δ 8.52 (d, J ) 5.5 Hz, 2H), 7.36 (s, 1H), 7.34-7.14
(m, 13H), 7.10 (d, J ) 9.5 Hz, 1H), 4.94 (d, J ) 13 Hz, 1H,
part of AB), 4.90 (d, J ) 5.5 Hz, 1H), 4.86 (d, J ) 13 Hz, 1H,
part of AB), 4.60 (d, J ) 14 Hz, 1H, part of AB), 4.55 (d, J )
14 Hz, 1H, part of AB), 3.78 (m, 1H), 3.72 (m, 1H), 3.39 (tt, J
In h ibitor 11: amine 31 was coupled to phenoxyacetic acid
following the general procedure. Rf 0.43 (95:5 CHCl3/MeOH).
1
[R]23 -31.4° (c 1.0, MeOH). H NMR (DMSO-d6) δ 8.51 (d, J
D

A New P3-P2 Ligand for HIV Protease Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6 1103
) 5.5 Hz, 2H), 7.83 (d, J ) 9 Hz, 1H), 7.35 (s, 1H), 7.31 (d, J
) 5.5 Hz, 2H), 7.28-7.19 (m, 6H), 7.15 (m, 1H), 6.92 (t, J )
7.5 Hz, 1H), 6.78 (d, J ) 8.5 Hz, 2H), 4.98 (d, J ) 5 Hz, 1H),
4.60 (d, J ) 14 Hz, 1H, part of AB), 4.54 (d, J ) 14 Hz, 1H,
part of AB), 4.37 (d, J ) 15 Hz, 1H, part of AB), 4.32 (d, J )
15 Hz, 1H, part of AB), 4.14 (m, 1H), 3.76 (m, 1H), 3.39 (tt, J
) 10.5, 4.5 Hz, 1H), 3.20 (broad d, J ) 12 Hz, 1H), 2.96 (dd, J
) 14, 3 Hz, 1H), 2.70 (m, 2H), 2.60 (dd, J ) 12.5, 7 Hz, 1H),
2.15 (dd, J ) 13, 6.5 Hz, 1H), 2.10-2.00 (m, 2H), 1.94 (broad
d, J ) 11.5 Hz, 1H), 1.55 (q, J ) 11.5 Hz, 1H), 1.46 (m, 1H),
1.26 (s, 9H). 13C NMR (DMSO-d6) δ 171.3, 166.9, 157.7, 149.3,
148.0, 139.4, 129.2, 129.1, 127.8, 125.6, 121.5, 120.9, 114.5,
75.0, 70.1, 67.1, 66.6, 66.5, 57.1, 53.4, 49.9, 49.8, 34.2, 34.0,
30.1, 28.3. HRMS (FAB) m/z C34H45N4O5 (MH+) calcd: 589.3390;
found: 589.3406. HPLC homogeneity: 95.7% (system A);
94.4% (system B).
1H), 3.00 (dd, J ) 14, 3.5 Hz, 1H), 2.80-2.68 (m, 2H), 2.63
(dd, J ) 12.5, 7 Hz, 1H), 2.18 (dd, J ) 12.5, 6 Hz, 1H), 2.13-
1.93 (m, 3H), 1.56 (q, J ) 11.5 Hz, 1H), 1.49 (m, 1H), 1.29 (s,
9H). 13C NMR (DMSO-d6) δ 171.7, 164.6, 149.3, 148.1, 139.7,
138.1, 135.0, 129.1, 129.0, 128.7, 127.8, 127.3, 125.6, 122.6,
121.6, 75.0, 69.8, 67.1, 66.8, 57.4, 54.1, 50.0, 49.9, 34.9, 34.8,
30.3, 28.3. HRMS (FAB) m/z C35H45N4O4 (MH+) calcd: 585.3441;
found: 585.3453. HPLC homogeneity: 99.1% (system A);
98.9% (system B).
In h ibitor 17: amine 31 was coupled to 2-methylhydrocin-
namic acid following the general procedure. Rf 0.32 (95:5
CHCl3/MeOH). [R]23 -30.8° (c 1.0, MeOH). 1H NMR (DMSO-
D
d6) δ 8.52 (d, J ) 5 Hz, 2H), 7.80 (d, J ) 9 Hz, 1H), 7.39 (s,
1H), 7.32 (d, J ) 5.5 Hz, 2H), 7.30-7.20 (m, 4H), 7.14 (t, J )
7 Hz, 1H), 7.11-7.00 (m, 4H), 4.92 (d, J ) 5 Hz, 1H), 4.60 (d,
J ) 14 Hz, 1H, part of AB), 4.55 (d, J ) 14 Hz, 1H, part of
AB), 4.07 (m, 1H), 3.71 (m, 1H), 3.40 (m, 1H), 3.21 (broad d, J
) 12 Hz, 1H), 2.92 (dd, J ) 14, 3.5 Hz, 1H), 2.73-2.56 (m,
5H), 2.25 (m, 2H), 2.22 (s, 3H), 2.13 (dd, J ) 13, 6.5 Hz, 1H),
2.11-2.00 (m, 2H), 1.95 (broad d, J ) 11 Hz, 1H), 1.54 (q, J )
11.5 Hz, 1H), 1.46 (m, 1H), 1.27 (s, 9H). 13C NMR (DMSO-d6)
δ 171.5, 171.0, 149.4, 148.1, 139.8, 139.4, 135.4, 129.8, 129.2,
128.0, 127.7, 125.8, 125.5, 121.6, 75.0, 70.1, 67.1, 66.6, 57.1,
53.7, 49.9, 49.8, 35.8, 34.5, 34.4, 30.2, 28.5, 28.3, 18.8. HRMS
(FAB) m/z C36H49N4O4 (MH+) calcd: 601.3754; found: 601.3739.
HPLC homogeneity: 93.7% (system A); 94.9% (system B).
In h ibitor 12: amine 31 was coupled to 3-phenylpropionic
acid following the general procedure. Rf 0.38 (95:5 CHCl3/
MeOH). [R]23 -29.8° (c 1.0, MeOH). 1H NMR (DMSO-d6) δ
D
8.52 (d, J ) 5.5 Hz, 2H), 7.78 (d, J ) 9 Hz, 1H), 7.38 (s, 1H),
7.32 (d, J ) 5.5 Hz, 2H), 7.29-7.20 (m, 6H), 7.18-7.09 (m,
4H), 4.91 (d, J ) 5.5 Hz, 1H), 4.60 (d, J ) 14 Hz, 1H, part of
AB), 4.55 (d, J ) 14 Hz, 1H, part of AB), 4.07 (m, 1H), 3.70
(m, 1H), 3.40 (m, 1H), 3.20 (broad d, J ) 12.5 Hz, 1H), 2.91
(dd, J ) 14, 3 Hz, 1H), 2.73-2.55 (m, 5H), 2.31 (t, J ) 8 Hz,
2H), 2.13 (dd, J ) 12.5, 6.5 Hz, 1H), 2.11-1.91 (m, 3H), 1.53
(q, J ) 11.5 Hz, 1H), 1.51-1.41 (m, 1H), 1.27 (s, 9H). 13C NMR
(DMSO-d6) δ 171.5, 171.0, 149.4, 148.1, 141.4, 139.8, 129.2,
128.1, 127.9, 127.8, 125.7, 125.5, 121.6, 75.0, 70.1, 67.1, 66.6,
57.1, 53.7, 50.0, 49.9, 37.1, 34.5, 34.3, 31.1, 30.2, 28.3. HRMS
(FAB) m/z C35H47N4O4 (MH+) calcd: 587.3597; found: 587.3624.
HPLC homogeneity: 98.2% (system A); 98.5% (system B).
In h ibitor 18: amine 31 was coupled to 2-methylphenoxy-
acetic acid following the general procedure. Rf 0.42 (95:5
CHCl3/MeOH). [R]23 -30.6° (c 1.0, MeOH). 1H NMR (DMSO-
D
d6) δ 8.51 (d, J ) 6 Hz, 2H), 7.67 (d, J ) 9.5 Hz, 1H), 7.35 (s,
1H), 7.31 (d, J ) 5.5 Hz, 2H), 7.29-7.20 (m, 4H), 7.16 (m, 1H),
7.11 (d, J ) 7 Hz, 1H), 7.03 (t, J ) 7.5 Hz, 1H), 6.83 (t, J ) 7.5
Hz, 1H), 6.58 (d, J ) 8.5 Hz, 1H), 4.99 (d, J ) 5.5 Hz, 1H),
4.59 (d, J ) 14 Hz, 1H, part of AB), 4.54 (d, J ) 14 Hz, 1H,
part of AB), 4.38 (d, J ) 14.5 Hz, 1H, part of AB), 4.31 (d, J )
14.5 Hz, 1H, part of AB), 4.16 (m, 1H), 3.77 (m, 1H), 3.39 (tt,
J ) 10.5, 4.5 Hz, 1H), 3.21 (broad d, J ) 12.5 Hz, 1H), 2.98
(dd, J ) 14, 3 Hz, 1H), 2.70 (m, 2H), 2.61 (dd, J ) 12.5, 7 Hz,
1H), 2.19 (s, 3H), 2.15 (dd, J ) 13, 6.5 Hz, 1H), 2.10-2.00 (m,
2H), 1.94 (broad d, J ) 11.5 Hz, 1H), 1.55 (q, J ) 11.5 Hz,
1H), 1.47 (m, 1H), 1.26 (s, 9H). 13C NMR (DMSO-d6) δ 171.3,
167.0, 155.8, 149.3, 148.0, 139.4, 130.3, 129.2, 127.8, 126.7,
125.8, 125.6, 121.5, 120.6, 111.4, 75.0, 70.0, 67.1, 66.9, 66.6,
57.1, 53.2, 49.9, 49.8, 34.3, 33.9, 30.2, 28.2, 16.0. HRMS (FAB)
m/z C35H47N4O5 (MH+) calcd: 603.3547; found: 603.3534.
HPLC homogeneity: 98.4% (system A); >99.5% (system B).
In h ibitor 13: amine 31 was coupled to thiophenoxyacetic
acid following the general procedure. The compound was
further purified by reversed-phase chromatography using
0.06% TFA/0.06% TFA in CH3CN gradients and isolated by
lyophilization: Rf 0.40 (95:5 CHCl3/MeOH). [R]23D -1.6° (c 1.0,
MeOH). 1H NMR (DMSO-d6) δ 8.75 (broad m, 2H), 8.28 (broad
m, 1H), 8.16 (broad d, J ) 9 Hz, 1H), 7.71 (broad, J ) 2.5 Hz,
2H), 7.26-7.12 (m, 10H), 4.77 (broad m, 2H), 3.97 (broad t, J
) 12.5 Hz, 1H), 3.51 (tq, J ) 9, 3 Hz, 1H), 3.15 (t, J ) 12 Hz,
1H), 3.09 (broad d, J ) 14.5 Hz, 1H), 2.95 (broad d, J ) 13.5
Hz, 2H), 2.60 (m, 1H), 2.46 (d, J ) 12.5 Hz, 1H), 2.22 (broad
d, J ) 12 Hz, 1H), 1.81 (broad q, J ) 12 Hz, 1H), 1.68 (broad
q, J ) 12 Hz, 1H), 1.29 (s, 9H). 13C NMR (DMSO-d6) δ 167.8,
166.3, 154.2, 145.0, 138.5, 136.2, 129.0, 128.9, 128.1, 127.5,
126.1, 125.7, 123.2, 72.1, 69.5, 67.4, 65.0, 58.1, 54.5, 51.3, 51.1,
36.4, 35.1, 34.0, 28.3, 28.1. HRMS (FAB) m/z C38H46N4O4S
(MH+) calcd: 605.3162; found: 605.3181. HPLC homogene-
ity: 99.3% (system A); 96. 9% (system B).
In h ibitor 20: 2,6-dimethylbenzyl alcohol (Scheme 2, pre-
pared as described for inhibitor 19, 2.42 g, 18 mmol) was
dissolved in CH2Cl2 (100 mL). Celite (∼ 5 mL) and powdered
4 Å molecular sieves (∼5 mL) were added. The suspension was
cooled in ice, and pyridinium chlorochromate (7.76 g, 36 mmol)
was added in portions. The mixture was stirred until comple-
tion as determined by TLC analysis. The mixture was con-
centrated to half-volume under reduced pressure, diethyl ether
(100 mL) was added, and the suspension was filtered through
Celite. Volatiles were removed under reduced pressure, and
the residual 2,6-dimethylbenzaldehyde was used without
In h ibitor 14: amine 31 was coupled to 4-phenylbutyric acid
following the general procedure. Rf 0.38 (95:5 CHCl3/MeOH).
1
[R]23 -27.4° (c 1.0, MeOH). H NMR (DMSO-d6) δ 8.51 (d, J
D
) 5 Hz, 2H), 7.67 (d, J ) 9 Hz, 1H), 7.34 (s, 1H), 7.31 (d, J )
5 Hz, 2H), 7.28-7.08 (m, 8H), 7.08 (d, J ) 8 Hz, 2H), 4.87 (d,
J ) 5 Hz, 1H), 4.60 (d, J ) 14 Hz, 1H, part of AB), 4.54 (d, J
) 14 Hz, 1H, part of AB), 4.08 (m, 1H), 3.69 (m, 1H), 3.39 (tt,
J ) 10.5, 4 Hz, 1H), 3.21 (broad d, J ) 12.5 Hz, 1H), 2.91 (dd,
J ) 14, 3 Hz, 1H), 2.68 (broad t, J ) 11.5 Hz, 2H), 2.58 (dd, J
) 12.5, 6.5 Hz, 1H), 2.38 (t, J ) 7.5 Hz, 2H), 1.64 (m, 2H),
1.51 (q, J ) 11.5 Hz, 1H), 1.46 (m, 1H), 1.27 (s, 9H). 13C NMR
(DMSO-d6) δ 171.5, 149.4, 148.2, 141.8, 139.9, 129.2, 128.2,
128.1, 127.8, 125.6, 121.7, 75.1, 70.4, 67.2, 66.7, 57.2, 53.6, 50.0,
49.9, 35.0, 34.6, 34.4, 30.2, 28.4, 27.1. HRMS (FAB) m/z
1
purification. H NMR (200 MHz, CDCl3) δ 10.60 (s, 1H), 7.30
(t, J ) 8 Hz, 1H), 7.10 (d, J ) 8 Hz, 2H), 2.65 (s, 6H).
The crude aldehyde from above and methyl (triphenylphos-
phoranylidene)acetate (6.70 g, 20 mmol) were dissolved in
toluene (40 mL), and the solution was refluxed for 4 h. After
being stirred overnight at room temperature, the reaction
mixture was passed through a pad of silica gel (∼100 mL)
using ether as eluent. Solvents were removed under reduced
pressure, and the residue was passed through another pad of
silica gel (30 mL) using ether (150 mL) for elution. Removal
of the solvent gave the crude cinnamate ester (∼2.4 g) as a
yellow oil that was used without further purification in the
next step.
C
36H49N4O4 (MH+) calcd: 601.3754; found: 601.3739. HPLC
homogeneity: 95.8% (system A); 97.0% (system B).
In h ibitor 15: amine 31 was coupled to trans-cinnamic acid
following the general procedure. Rf 0.38 (95:5 CHCl3/MeOH).
1
[R]23 -44.5° (c 1.0, MeOH). H NMR (DMSO-d6) δ 8.51 (d, J
D
) 5.5 Hz, 2H), 8.13 (d, J ) 9.5 Hz, 1H), 7.50 (m, 3H), 7.41-
7.29 (m, 7H), 7.23 (t, J ) 7.5 Hz, 2H), 7.13 (t, J ) 7.5 Hz, 1H),
6.72 (d, J ) 16 Hz, 1H), 5.00 (d, J ) 5 Hz, 1H), 4.59 (d, J ) 14
Hz, 1H, part of AB), 4.55 (d, J ) 14 Hz, 1H, part of AB), 4.28
(m, 1H), 3.83 (m, 1H), 3.40 (m, 1H), 3.26 (broad d, J ) 12 Hz,
The crude cinnamate from above (∼2.4 g) was hydrogenated
(1 atm H2 gas) in MeOH over 20% Pd(OH)2 on carbon (100
mg) for 20 h. Filtration followed by removal of solvent gave
crude methyl 3-(2′,6′-dimethylphenyl)propionate. 1H NMR (200

1104 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6
Beaulieu et al.
MHz, CDCl3) δ 7.00 (s, 3H), 3.70 (s, 3H), 3.00 (m, 2H), 2.49
(m, 2H), 2.35 (s, 6H). The material was dissolved in 3:1 THF-
water (25 mL), and LiOH monohydrate (1.68 g, 40 mmol) was
added. The mixture was stirred 20 h at room temperature.
THF was removed under reduced pressure and the aqueous
phase washed with 1:1 ether-hexane. After acidification to
pH 1 with 6 N HCl, the product was extracted with EtOAc
and washed with water and the solution dried (MgSO4).
Removal of the solvent under reduced pressure gave 3-(2′,6′-
dimethylphenyl)propionic acid 32 as a cream colored solid (1.52
g, 48% yield overall from 2,6-dimethylbenzyl alcohol). Mp 75-
above for 34a : 1H NMR (200 MHz, CDCl3) δ 9.2 (broad s, 1H),
7.0 (m, 3H), 4.60 (q, J ) 7 Hz, 1H), 2.28 (s, 6H), 1.55 (d, J )
7 Hz, 3H).
Amine 31 was coupled to 34b to give 22 as a mixture of
epimers, following the general procedure: Rf 0.46 (95:5 CHCl3/
1
MeOH). H NMR (DMSO-d6) δ 8.51 (d, J ) 5.5 Hz, 2H), 7.73
(d, J ) 9 Hz, 0.5H), 7.72 (d, J ) 9 Hz, 0.5H), 7.36-7.18 (m,
7H), 7.14 (m, 1H), 6.96 (t, J ) 6.5 Hz, 2H), 6.90 (t, J ) 7 Hz,
0.5H), 6.88 (t, J ) 7.5 Hz, 0.5H), 5.02 (d, J ) 5.5 Hz, 0.5H),
5.01 (d, J ) 5.5 Hz, 0.5H), 4.61 (d, J ) 14 Hz, 1H, part of AB),
4.54 (d, J ) 14 Hz, 1H, part of AB), 4.28 (q, J ) 6.5 Hz, 0.5H),
4.20 (m, 1H), 4.16 (q, J ) 6.5 Hz, 0.5H), 3.77 (m, 1H), 3.41 (m,
1H), 3.21 (m, 1H), 2.99 (m, 1H), 2.80-2.60 (m, 3H), 2.23-2.04
(m, 2H), 2.15 (s, 3H), 2.08 (s, 3H), 1.92 (broad d, J ) 10.5 Hz,
1H), 1.62-1.45 (m, 2H), 1.27 (s, 9H). 13C NMR (DMSO-d6) δ
171.2, 170.5, 170.4, 153.8, 153.3, 149.3, 148.0, 139.3, 139.2,
130.4, 129.3, 129.1, 128.7, 127.7, 127.6, 125.5, 123.5, 121.5,
77.5, 77.1, 75.0, 70.5, 70.4, 67.1, 66.2, 56.6, 52.7, 49.9, 49.6,
33.9, 33.7, 33.5, 29.7, 28.2, 18.2, 17.9, 16.6, 16.5. HRMS (FAB)
m/z C37H51N4O5 (MH+) calcd: 631.3881; found: 617.3715.
HPLC homogeneity: 99.1% (system A); 99.6% (system B).
1
77 °C. H NMR (CDCl3) δ 7.03 (s, 3H), 3.00 (m, 2H), 2.51 (m,
2H), 2.35 (s, 6H). 13C NMR (100 MHz, CDCl3) δ 179.5, 137.0,
136.4, 128.6, 126.5, 33.4, 25.0, 19.9. MS (FAB) m/z 179 (MH+).
Anal. Calcd for C11H14O2: C, 74.13; H, 7.92. Found: C, 74.06;
H, 7.78.
Amine 31 was coupled to acid 32 to give inhibitor 20
following the general procedure. Rf 0.40 (95:5 CHCl3/MeOH).
1
[R]23 -44.5° (c 1.0, MeOH). H NMR (400 MHz, DMSO-d6) δ
D
8.52 (dd, J ) 5, 1.5 Hz, 2H), 7.80 (d, J ) 9 Hz, 1H), 7.39 (s,
1H), 7.33-7.28 (m, 4H), 7.24 (t, J ) 7.5 Hz, 2H), 7.15 (m, 1H),
6.94 (s, 3H), 4.94 (d, J ) 5.5 Hz, 1H), 4.60 (d, J ) 14 Hz, 1H,
part of AB), 4.56 (d, J ) 14 Hz, 1H, part of AB), 4.06 (m, 1H),
3.71 (m, 1H), 3.40 (tt, J ) 10.5, 4.5 Hz, 1H), 3.22 (dt, J ) 12,
3.5 Hz, 1H), 2.93 (dd, J ) 13.5, 3.5 Hz, 1H), 2.70 (dd, J ) 11,
2 Hz, 1H), 2.68-2.46 (m, 4H), 2.22 (s, 6H), 2.16-2.00 (m, 6H),
1.95 (broad d, J ) 11.5 Hz, 1H), 1.53 (q, J ) 11.5 Hz, 1H),
1.46 (m, 1H), 1.26 (s, 9H). 13C NMR (DMSO-d6) δ 171.4, 171.0,
149.4, 148.1, 139.8, 138.0, 135.5, 129.3, 127.8, 127.7, 125.6,
121.6, 75.0, 70.2, 67.1, 66.6, 57.1, 53.7, 50.0, 49.9, 34.8, 34.4,
30.2, 28.3, 25.5, 19.2. HRMS (FAB) m/z C37H51N4O4 (MH+)
calcd: 615.3910; found: 615.3931. HPLC homogeneity: 93.8%
(system A); 93.7% (system B).
In h ibitor 23: 3-hydroxy-2-methylbenzoic acid23 (0.452 g, 3.0
mmol) and 1-hydroxybenzotriazole hydrate (0.419 g, 3.1 mmol)
were dissolved in THF (35 mL), and 1,3-dicyclohexylcarbodi-
imide (0.619 g, 3.0 mmol) was added. After 15 min, amine 36
(1.350 g, 3.0 mmol) was added and the mixture stirred 20 h
at room temperature. The precipitated dicyclohexyl urea was
removed by filtration, and the filtrate was washed with water.
After drying (MgSO4) and removal of solvent under reduced
pressure, the residue was purified by flash chromatography
using 95:5 CHCl3/MeOH as eluent. The product was then
precipitated from EtOAc (3 mL) by addition of ether (20 mL)
to give 23 as an off-white solid (0.406 g, 23% yield). Rf 0.40
(9:1 CHCl3/MeOH). [R]23 -4.2° (c 1.0, MeOH). 1H NMR
D
In h ibitor 21: 2,6-dimethylphenol 35 (500.0 g, 4.09 mol),
ethyl bromoacetate (480 mL, 4.3 mol), and anhydrous potas-
sium carbonate (690 g, 5.0 mol) were suspended in acetone (2
L), and the mixture was stirred 56 h at room temperature.
NaOH (1 N, 1 L) was added and the mixture refluxed for 4 h.
NaOH (150 g, 3.75 mol) was added, and reflux continued for
another 48 h. After cooling and removal of acetone under
reduced pressure, the aqueous phase was carefully acidified
to pH 1 with concentrated HCl. The tan-colored precipitate
was collected, washed with water, and dried in air. The crude
material was recrystallized once from 2-propanol and then
from EtOAc to give pure 2′,6′-dimethylphenoxyacetic acid 34a
(R ) H) as white crystals (460 g, 62% yield): mp 137.5-139
°C. 1H NMR (200 MHz, CDCl3) δ 10.4 (broad s, 1H), 7.0 (m,
3H), 4.55 (s, 2H), 2.34 (s, 6H). 13C NMR (CDCl3) δ 172.2, 155.1,
130.7, 129.0, 124.5, 68.7, 16.2. MS (FAB) m/z 181 (MH+). Anal.
Calcd for C10H12O3: C, 66.65; H, 6.71. Found: C, 66.37; H,
6.74.
(DMSO-d6) δ 9.34 (s, 1H), 8.51 (d, J ) 5.5 Hz, 2H), 7.88 (broad
d, J ) 8.5 Hz, 1H), 7.38-7.28 (m, 5H), 7.25 (t, J ) 7.5 Hz,
2H), 7.15 (t, J ) 7 Hz, 1H), 6.92 (t, J ) 8 Hz, 1H), 6.75 (d, J
) 8 Hz, 1H), 6.49 (d, J ) 7.5 Hz, 1H), 4.94 (broad m, 1H), 4.61
(d, J ) 14 Hz, 1H, part of AB), 4.55 (d, J ) 14 Hz, 1H, part of
AB), 4.21 (broad m, 1H), 3.77 (broad m, 1H), 3.41 (broad m,
1H), 3.25 (broad d, J ) 12 Hz, 1H), 3.00 (broad d, J ) 11.5
Hz, 1H), 2.78-2.62 (m, 2H), 2.2-1.89 (m, 3H), 1.81 (s, 3H),
1.51 (q, J ) 11.5 Hz, 1H), 1.23 (s, 9H).13C NMR (DMSO-d6) δ
171.2, 168.8, 155.3, 149.4, 148.2, 139.9, 139.3, 129.3, 127.8,
125.6, 125.5, 121.6, 121.3, 117.3, 115.0, 75.0, 70.9, 67.1, 66.3,
56.7, 53.7, 49.9, 49.8, 34.2, 33.7, 29.8, 28.3, 12.2. HRMS (FAB)
m/z C34H45N4O5 (MH+) calcd: 589.3390; found: 589.3404.
HPLC homogeneity: 90.5% (system A); 90.4% (system B).
In h ibitor 24: 3-(S)-hydroxytetrahydrofuran (1.322 g, 15.0
mmol) in CH2Cl2 (50 mL) was added dropwise to a ice-cooled
solution of phosgene in toluene (1.93 M, 30 mL). After the
mixture was stirred for 1 h at 0 °C, volatiles were removed
under vacuum. The residue was dissolved in CH2Cl2 (20 mL),
N-methylmorpholine (2.41 mL, 22.5 mmol) was added, and the
solution cooled in an ice bath. Amine 31 (2.27 g, 5 mmol) in
CH2Cl2 (20 mL + 20 mL rinse) was added dropwise, and the
mixture stirred for 16 h at room temperature. Volatiles were
then removed under reduced pressure, the residue was dis-
solved in EtOAc, and the solution was washed successively
with saturated aqueous NaHCO3 and brine. After the solution
was dried over MgSO4, evaporation of the solvent gave an oil
that was purified by flash chromatography using 5% EtOH in
Amine 31 was coupled to 34a following the general proce-
dure: Rf 0.44 (95:5 CHCl3/MeOH). [R]23D -23.1° (c 1.0, MeOH).
1H NMR (DMSO-d6) δ 8.51 (d, J ) 5.5 Hz, 2H), 7.78 (d, J )
9.5 Hz, 1H), 7.37 (s, 1H), 7.30 (m, 4H), 7.23 (t, J ) 7.5 Hz,
2H), 7.16 (t, J ) 7 Hz, 1H), 6.99 (d, J ) 7 Hz, 2H), 6.91 (dd, J
) 8, 6.5 Hz, 1H), 5.04 (d, J ) 5 Hz, 1H), 4.60 (d, J ) 14 Hz,
1H, part of AB), 4.54 (d, J ) 14 Hz, 1H, part of AB), 4.25 (broad
t, J ) 10 Hz, 1H), 4.12 (d, J ) 14.5 Hz, 1H, part of AB), 3.93
(d, J ) 14.5 Hz, 1H, part of AB), 3.83 (m, J ) 5 Hz, 1H), 3.40
(tt, J ) 10.5, 4.5 Hz, 1H), 3.23 (broad d, J ) 12.5 Hz, 1H),
3.03 (dd, J ) 14, 3 Hz, 1H), 2.81-2.71 (m, 2H), 2.67 (dd, J )
12.5, 6.5 Hz, 1H), 2.19 (dd, J ) 13, 6.5 Hz, 1H), 2.14 (s, 6H),
2.08 (broad d, J ) 12 Hz, 2H), 1.94 (broad d, J ) 11.5 Hz,
1H), 1.56 (q, J ) 11.5 Hz, 1H), 1.50 (m, 1H), 1.28 (s, 9H). 13C
NMR (DMSO-d6) δ 171.2, 167.0, 154.3, 149.3, 148.0, 139.4,
130.2, 129.2, 128.6, 127.7, 125.6, 124.0, 121.5, 75.0, 70.1, 67.1,
66.5, 57.1, 53.1, 49.9, 49.7, 34.0, 33.8, 29.9, 28.2, 15.7. HRMS
(FAB) m/z C36H49N4O5 (MH+) calcd: 617.3703; found: 617.3715.
HPLC homogeneity: 96.7% (system A); 97.2% (system B).
CHCl3 as eluent. Rf 0.43 (95:5 CHCl3/MeOH). [R]23 -31.7° (c
D
1
1.0, MeOH). H NMR (DMSO-d6) δ 8.52 (d, J ) 5.5 Hz, 2H),
7.37-7.20 (m, 7H), 7.14 (m, 1H), 6.97 (d, J ) 9 Hz, 1H), 4.93
(m, 1H), 4.84 (d, J ) 5 Hz, 1H), 4.60 (d, J ) 13.5 Hz, 1H, part
of AB), 4.55 (d, J ) 14 Hz, 1H, part of AB), 3.75-3.57 (m, 5H),
3.43-3.32 (m, 1H), 3.21 (broad d, J ) 12 Hz, 1H), 2.93 (dd, J
) 13.5, 2.5 Hz, 1H), 2.70-2.53 (m, 2H), 2.12-1.90 (m, 5H),
1.82-1.75 (m, 1H), 1.55-1.37 (m, 2H), 1.26 (s, 9H). 13C NMR
(DMSO-d6) δ 171.4, 155.6, 149.4, 148.1, 139.9, 129.4, 129.1,
127.8, 125.6, 121.6, 74.9, 73.9, 72.7, 70.2, 67.1, 66.8, 66.1, 57.3,
56.0, 50.0, 34.9, 34.5, 32.1, 30.2, 28.3. HRMS (FAB) m/z
In h ibitor 22: carboxylic acid 34b was prepared from 2,6-
dimethylphenol 35 and ethyl 2-bromopropanoate as described

A New P3-P2 Ligand for HIV Protease Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6 1105
C31H45N4O6 (MH+) calcd: 569.3339; found: 569.3328. HPLC
homogeneity: 95.7% (system A); 98.6% (system B).
washed with hexane (200 mL), and dried. Mercaptopyridine
derivative 38c (44.93 g, 80% yield) was obtained as a yellowish
solid: mp 127-129 °C. Rf 0.56 (9:1 CHCl3/MeOH). [R]18
D
P r ep a r a tion of (2S,4R)-4-br om op ip ecolic ter t-bu tyl-
a m id e 37 (X ) Br ): (2S,4R)-4-hydroxypipecolic tert-butyl-
amide 3621,25 (70.10 g, 0.233 mol) was dissolved in dry THF
(700 mL), and Et3N (42 mL, 0.30 mol) was added. The solution
was cooled in an ice-water bath under a nitrogen atmosphere,
and methanesulfonyl chloride (22.5 mL, 0.29 mol) was added
dropwise over 30 min. The resulting white suspension was
stirred at room temperature for 2 h and poured into 20%
aqueous NaCl, and the organic layer was separated. The
organic extract was washed with 20% aqueous NaCl (650 mL),
dried (MgSO4), and concentrated under reduced pressure to a
volume of ca. 200 mL. Hexane (350 mL) was added to the
yellow solution, producing a white precipitate. After cooling
overnight in a refrigerator, the solid was collected by filtration,
washed with hexane, and dried in air. The mesylate of alcohol
36 (84.96 g, 96% yield) was obtained as a white solid: mp 122-
123 °C. Rf 0.67 (1:4 hexane/EtOAc). [R]22D -41.2° (c 1.0, CHCl3).
-64.5° (c 1.0, CHCl3). IR (KBr) ν 3270, 1682 cm-1
.
1H NMR
(CDCl3) δ 8.41 (d, J ) 6 Hz, 2H), 7.11 (d, J ) 6 Hz, 2H), 5.84
(broad s, 1H), 4.53 (dd, J ) 6.5, 4 Hz, 1H), 3.98 (m, 1H), 3.77
(m, J ) 4.5 Hz, 1H), 3.33-3.20 (m, 1H), 2.64 (m, 1H), 2.16
(ddd, J ) 14.5, 7, 5 Hz, 1H), 2.10-1.97 (m, 1H), 1.85-1.75
(m, 1H), 1.48 (s, 9H), 1.38 (s, 9H). 13C NMR (CDCl3) δ 169.6,
155.3, 149.6, 147.8, 122.1, 81.3, 54.5, 51.2, 37.7, 37.6 (broad),
29.1, 28.8, 28.4. MS (FAB) m/z 394 (MH+). Anal. Calcd for
C20H31N3O3S: C, 61.04; H, 7.94; N, 10.68. Found: C, 60.74;
H, 8.29; N, 10.67.
Following the same procedure but using iodide 37 (X ) I)
instead of the bromide, carbamate 38c was obtained in 95%
yield after purification by flash chromatography.
Gen er a l P r oced u r e for Dep r otection of Ca r ba m a tes
38. P r ep a r a tion of 39c: carbamate 38c (5.26 g, 13.4 mmol)
was stirred at room temperature for 20 min with 4 N HCl-
dioxane (45 mL).
1
IR (KBr) ν 3378, 1694, 1520 cm-1. H NMR (CDCl3) δ 5.84-
5.69 (broad s, 1H), 5.07 (m, 1H), 4.68 (broad s, 1H), 4.08 (broad
m, 1H), 3.25-3.13 (m, 1H), 3.07 (s, 3H), 2.92-2.83 (m, 1H),
2.06-1.95 (m, 1H), 1.81 (dd, J ) 7, 3 Hz, 1H, part of ABX),
1.77 (dd, J ) 7, 3 Hz, 1H, part of ABX), 1.78-1.67 (m, 1H),
1.49 (s, 9H), 1.37 (s, 9H). 13C NMR (CDCl3) δ 169.1, 155.1, 81.5,
74.8, 52.7 (broad), 51.3, 39.1, 35.8 (broad), 30.5 (broad), 29.4
(broad), 28.8, 28.5. MS (ES+) m/z 379 (MH+), 279 (MH+ - Boc).
Anal. Calcd for C16H30N2O6S: C, 50.78; H, 7.99; N, 7.40.
Found: C, 51.10; H, 8.34; N, 7.46.
Volatiles were removed under reduced pressure, and the
residual solid was partitioned between EtOAc and water. The
pH of the aqueous phase was adjusted to 9 with 1 N NaOH
and the organic phase separated. The aqueous phase was
extracted with more EtOAc (3 × 75 mL), and the combined
organic extracts were washed with brine (100 mL). After
drying (MgSO4) and removal of solvent under vacuum, crude
39c was obtained as a white foam (3.94 g, quantitative): 1H
NMR (CDCl3) δ 8.41 (dd, J ) 5, 1.5 Hz, 2H), 7.14 (dd, J ) 5,
1.5 Hz, 2H), 6.50 (broad s, 1H), 3.38 (tt, J ) 12, 4 Hz, 1H),
3.22 (m, 1H), 3.18 (dd, J ) 12, 3 Hz, 1H), 2.77 (dt, J ) 12, 2.5
Hz, 1H), 2.48 (dq, J ) 13, 2.5 Hz, 1H), 2.04 (m, 1H), 1.62 (broad
s, 1H), 1.51 (dq, J ) 12, 4 Hz, 1H), 1.45-1.30 (m, 1H), 1.34 (s,
9H).
Gen er a l P r oced u r e for Cou p lin g Am in e 39 to Ep oxid e
40. P r ep a r a tion of Ca r ba m a te 41c: amine 39c (3.92 g, 13.5
mmol, 1 equiv) was dissolved in EtOH (45 mL), and LiCl (1.24
g, 29.5 mmol, 2.2 equiv) was added. Epoxide 4026 (1.76 g, 6.7
mmol, 0.5 equiv) was added, and the mixture was stirred at
60 °C for 45 min. A second portion of epoxide (1.76 g, 6.69
mmol, 0.5 equiv) was added, and stirring at 60 °C continued
for 30 min. A third portion (1.82 g, 6.9 mmol, 0.52 equiv) was
added, and the mixture stirred overnight at 60 °C. One last
portion of epoxide (1.19 g, 4.5 mmol, 0.34 equiv) was added,
and heating resumed for a total reaction time of 22 h. The
reaction mixture was then evaporated under reduced pressure
and the residue dissolved in EtOAc (200 mL). The solution
was washed with water (135 mL) and brine (135 mL), dried
(MgSO4), and evaporated under reduced pressure. The residue
was purified by flash chromatography on silica gel using 4%
and 6% MeOH/EtOAc as eluent. Carbamate 41c was obtained
as a solid (3.97 g, 53% yield). 1H NMR (CDCl3) δ 8.41 (dd, J )
4.5, 1.5 Hz, 2H), 7.38-7.26 (m, 2H), 7.25-7.18 (m, 3H), 7.13
(dd, J ) 4.5, 1.5 Hz, 2H), 6.29 (broad s, 1H), 4.70 (broad m,
1H), 3.95-3.68 (m, 3H), 3.39-3.25 (m, 2H), 2.97-2.80 (m, 3H),
2.72 (dd, J ) 13, 4 Hz, 1H), 2.45-2.27 (m, 3H), 1.99 (m, 1H),
1.73 (dq, J ) 12.5, 3.5 Hz, 1H), 1.65 (q, J ) 12 Hz, 1H), 1.79-
1.58 (m, 2H), 1.36 (broad s, 9H), 1.34 (s, 9H).
The mesylate from above (79.67 g, 0.210 mol) and tetra-n-
butylammonium bromide (271.41 g, 0.842 mol) were dissolved
in THF (1.1 L), and the solution was refluxed for 15 h. After
the mixture cooled to room temperature, ether (1.7 L) was
added and the solution washed with 5% Na2S2O3 (1 L), water
(1.1 L), and brine (1.1 L). The solution was dried (MgSO4),
filtered, and concentrated under reduced pressure to give a
white solid. MeOH (500 mL) was added to dissolve the solid,
and water (220 mL) was added slowly. The mixture was left
to crystallize for 2 days at 4 °C. The product was collected by
filtration and dried in a vacuum over P2O5. Bromide 37 (X )
Br), was obtained as fine white needles (52.05 g, 68% yield):
mp 112-114 °C. Rf 0.78 (2:1 hexane/EtOAc). [R]22 -119.2° (c
D
1.0, CHCl3). IR (KBr) ν 3345-3301, 1663, 1549 cm-1. 1H NMR
(CDCl3, mixture of rotamers) δ 6.31-6.01 (broad s, 1H), 4.81-
4.41 (m, 2H), 4.23-3.82 (m, 1H), 2.92-2.67 (m, 2H), 2.28-
2.15 (m, 1H), 1.97-1.81 (m, 2H), 1.49 (s, 9H), 1.32 (s, 9H).13
C
NMR (CDCl3) δ 169.2, 156.0, 81.5, 55.8, 51.4, 45.1, 42.7, 36.8,
36.6, 28.9, 28.5. MS (ES+) m/z 263 (MH+ - Boc). Anal. Calcd
for C15H27BrN2O3: C, 49.59; H, 7.49; N, 7.710. Found: C, 50.00;
H, 7.78; N, 7.81.
P r ep a r a tion of (2S,4R)-4-iod op ip ecolic ter t-bu tyl-
a m id e 37 (X ) I): following the procedure described above
for the preparation of the bromide, the mesylate of 36 was
converted to the iodo derivative 37 (X ) I) using tetra-n-
butylammonium iodide (71% yield): 1H NMR (CDCl3, mixture
of rotamers) δ 6.27-6.02 (broad s, 1H), 4.80-4.27 (m, 2H),
4.19-3.69 (m, 1H), 2.99-2.70 (m, 2H), 2.37-2.25 (m, 1H),
2.17-1.99 (m, 2H), 1.50 (s, 9H), 1.33 (s, 9H).
Gen er a l P r oced u r e for Dep r otection of 41 a n d Cou -
p lin g of Am in es 42 (Sch em e 5). P r ep a r a tion of in h ibitor
27: carbamate 42 (7.85 g, 14 mmol) was dissolved in 4 N HCl-
dioxane (47 mL) and the mixture stirred for 20 min at room
temperature. Volatiles were then evaporated under reduced
pressure, and the resulting salt was dried under vacuum for
1 h. The crude hydrochloride was used without purification.
The salt from above (assume 14 mmol, 1 equiv), 2′,6′-
dimethylphenoxyacetic acid (2.96 g, 16.5 mmol, 1.16 equiv),
and BOP coupling reagent (7.79 g, 17.6 mmol, 1.25 equiv) were
dissolved in dry DMF (100 mL), and diisopropylethylamine
(14.8 mL, 85 mmol, 6 equiv) was added. The mixture was
stirred for 2.5 h at room temperature. The reaction mixture
was diluted with EtOAc (500 mL) and washed with saturated
NaHCO3 (2 × 500 mL), water (500 mL), and brine (500 mL).
After drying (MgSO4) and evaporation, crude 27 was obtained
Gen er a l P r oced u r e for th e P r ep a r a tion of Th io De-
r iva tives 38a -c. P r ep a r a tion of 38c: to an ice-cold suspen-
sion of 95% NaH (4.24 g, 0.178 mol, 1.2 equiv) in anhydrous
DMF (350 mL) was added 4-mercaptopyridine (22.30 g, 0.200
mol, 1.4 equiv) in small portions. After the mixture was stirred
for 1 h at 0 °C under an atmosphere of nitrogen, bromide 37
(52.05 g, 0.143 mol, 1 equiv) was added (the corresponding
iodide can also be used). The mixture was stirred for 1 h at
room temperature, then 18 h at 43 °C. The reaction was then
cooled to room temperature, diluted with EtOAc (2 L), and
washed successively with water (900 mL), 1 N NaOH (900 mL),
and brine (900 mL). After drying (MgSO4), volatiles were
removed under reduced pressure, and the residual foam (60.16
g) was refluxed with hexane (550 mL) for 20 min. The resulting
suspension was cooled to room temperature and allowed to
stand overnight. The product was collected by filtration,

1106 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6
Beaulieu et al.
as a foam. The inhibitor was purified by flash chromatography
on silica gel using 3-10% iPrOH/EtOAc as eluents. Compound
27 was obtained as an off-white amorphous solid (3.08 g, >98%
homogeneous, and 4.37 g, 88% homogeneous. Total: 7.45 g,
mixture was acidified with 10% aqueous citric acid and
extracted with EtOAc. The extract was washed twice with
saturated NaHCO3 and then with brine. After the solution was
dried over MgSO4, the solvent was evaporated and the residue
dried under vacuum. (2S,4S)-4-Mercaptopipecolic tert-butyl-
amide 38 (R ) H) was used without purification (4.20 g, 99%
yield, crude).
To a suspension of NaH (0.351 g, 14.6 mmol, 1.1 equiv) in
DMF (44 mL) at 0 °C was added the thiol from above (4.20 g,
13.3 mmol), and the mixture was stirred for 0.5 h. 4-Chloro-
2,6-dimethylpyrimidine (2.64 g, 18.6 mmol, 1.4 equiv) was
added, and stirring continued at 0 °C for another hour. The
reaction was then quenched by pouring into saturated NaH-
CO3, and the product was extracted with EtOAc. The extract
was washed twice with 10% aqueous Na2CO3 and then with
brine. After drying (MgSO4), the solvent was evaporated and
the residue purified by flash chromatography on silica gel
using 1:1 and then 3:2 EtOAc/hexane to give 38d (4.1 g, 73%
yield). MS (FAB) m/z 423 (MH+).
85% yield). Rf 0.56 (9:1 CHCl3/MeOH). [R]22 +16.7° (c 1.1,
D
CHCl3). 1H NMR (DMSO-d6) δ 8.39 (dd, J ) 5, 1.5 Hz, 2H),
7.79 (d, J ) 9 Hz, 1H), 7.42 (s, 1H), 7.33-7.27 (m, 4H), 7.24
(t, J ) 7.5 Hz, 2H), 7.16 (m, 1H), 7.00 (d, J ) 7.5 Hz, 2H),
6.92 (dd, J ) 8.5, 6.5 Hz, 1H), 5.05 (d, J ) 5 Hz, 1H), 4.21 (m,
1H), 4.11 (d, J ) 14.5 Hz, 1H, part of AB), 3.92 (d, J ) 14.5
Hz, 1H, part of AB), 3.82 (m, J ) 5 Hz, 1H), 3.24 (broad d, J
) 12.5 Hz, 1H), 3.00 (dd, J ) 14, 3 Hz, 1H), 2.88 (dd, J ) 11,
2 Hz, 1H), 2.73 (dd, J ) 13.5, 11 Hz, 1H), 2.65 (dd, J ) 12.5,
6.5 Hz, 1H), 2.26 (m, 1H), 2.17 (dd, J ) 12.5, 6.5 Hz, 1H), 2.12
(s, 6H), 2.02 (broad d, J ) 12 Hz, 1H), 1.92 (broad d, J ) 12
Hz, 1H), 1.60 (m, 2H), 1.25 (s, 9H). 13C NMR (DMSO-d6) δ
170.6, 166.6, 154.0, 149.0, 146.3, 139.0, 129.8, 128.9, 128.3,
127.4, 125.3, 123.7, 121.2, 69.8, 69.5, 67.4, 56.7, 52.8, 51.1, 49.6,
34.5, 33.5, 29.9, 27.8, 15.3. HRMS (FAB) m/z C35H47N4O4S
(MH+) calcd: 619.3318; found: 619.3308. HPLC homogene-
ity: 96.8% (system A); 98.9% (system B).
Following the general procedures described previously, 38d
was converted into inhibitor 28: mp 90-96 °C. Rf 0.50 (9:1
1
CHCl3/MeOH). [R]20 +4.4° (c 1.0, MeOH). H NMR (DMSO-
In h ibitor 25: alkylation of 4-iodopipecolic derivative 37 (X
) I) with 4-pyridinemethanethiol,29a,c using NaH in DMF as
described for 38c, gave 38a in 68% yield. Following the general
procedures described above, this compound was converted into
D
d6) δ 7.78 (d, J ) 9 Hz, 1H), 7.42 (s, 1H), 7.30 (d, J ) 7 Hz,
2H), 7.24 (t, J ) 7.5 Hz, 2H), 7.16 (t, J ) 7.5 Hz, 1H), 7.09 (s,
1H), 7.00 (d, J ) 7.5 Hz, 2H), 6.92 (dd, J ) 8.5, 6.5 Hz, 1H),
5.04 (d, J ) 5 Hz, 1H), 4.22 (m, 1H), 4.11 (d, J ) 14.5 Hz, 1H,
part of AB), 3.92 (d, J ) 14.5 Hz, 1H, part of AB), 3.82 (m,
2H), 3.28 (s, 1H), 3.24 (m, 1H), 3.00 (dd, J ) 13.5, 3 Hz, 1H),
2.85 (dd, J ) 11, 2.5 Hz, 1H), 2.73 (dd, J ) 13.5, 11 Hz, 1H),
2.65 (dd, J ) 12.5, 7 Hz, 1H), 2.50 (s, 3H), 2.32 (s, 3H), 2.20
(m, 2H), 2.12 (s, 6H), 2.06 (broad d, J ) 12 Hz, 1H), 1.96 (broad
d, J ) 11 Hz, 1H), 1.65 (m, 2 H), 1.25 (s, 9H). 13C NMR (DMSO-
d6) δ 171.1, 168.1, 167.0, 166.4, 165.1, 154.4, 139.5, 130.2,
129.3, 128.7, 127.8, 125.7, 124.2, 114.7, 70.2, 69.9, 68.2, 57.4,
53.2, 51.8, 50.0, 35.1, 33.8, 30.6, 28.3, 25.6, 23.2, 15.8. HRMS
(FAB)m/zC36H50N5O4S (MH+)calcd: 648.3583;found: 648.3570.
HPLC homogeneity: >99.5% (system A); 99.7% (system B).
inhibitor 25. Mp 72-76 °C. Rf 0.52 (9:1 CHCl3/MeOH). [R]20
D
-10.9° (c 1.0, MeOH). 1H NMR (DMSO-d6) δ 8.50 (dd, J ) 4.5,
1.5 Hz, 2H), 7.75 (d, J ) 9 Hz, 1H), 7.36 (m, 3H), 7.31-7.21
(m, 7H), 7.16 (tt, J ) 7.5, 1 Hz, 1H), 7.00 (d, J ) 7.5 Hz, 2H),
6.92 (dd, J ) 8.5, 6.5 Hz, 1H), 4.99 (broad d, J ) 3 Hz, 1H),
4.20 (m, 1H), 4.10 (d, J ) 14.5 Hz, 1H, part of AB), 3.92 (d, J
) 14.5 Hz, 1H, part of AB), 3.81 (s, 2H), 3.79 (m, 1H), 3.18
(broad d, J ) 11 Hz, 1H), 2.99 (dd, J ) 13.5, 3 Hz, 1H), 2.71
(dd, J ) 13.5, 11 Hz, 1H), 2.67-2.55 (m, 2H), 2.12 (s, 6H),
2.05-1.79 (m, 4H), 1.47 (m, 1H), 1.24 (s, 9H). 13C NMR
(DMSO-d6) δ 170.8, 166.6, 154.0, 149.2, 147.7, 139.1, 129.8,
128.9, 128.3, 127.4, 125.3, 123.7, 123.5, 69.8, 69.4, 67.8, 57.1,
52.7, 51.5, 49.6, 35.5, 33.4, 31.7, 31.0, 27.9, 15.4. HRMS (FAB)
m/z C36H49N4O4S (MH+) calcd: 633.3475; found: 633.3465.
HPLC homogeneity: 98.7% (system A); 99.4% (system B).
In h ibitor 29: amine 4327 (0.104 g, 0.44 mmol), epoxide 4026
(0.060 g, 0.23 mmol), and LiCl (0.037 g, 0.87 mmol) were
dissolved in EtOH (1.5 mL), and the mixture was stirred at
60 °C for 1.5 h. Another portion of epoxide 40 (0.057 g, 0.22
mmol) was added, and the mixture was stirred overnight at
60 °C. More epoxide (0.060 g, 0.23 mmol) was added, and
heating continued for a total time of 24 h, at which point the
reaction was judged complete by TLC. The solvent was
removed under reduced pressure and the residue dissolved in
EtOAc (10 mL). The solution was washed with water (10 mL)
and brine (10 mL), dried (MgSO4), and evaporated under
reduced pressure. The crude material was purified by flash
chromatography on silica gel, eluting sequentially with 20%,
25%, 30%, and 40% EtOAc/hexane. Carbamate 44 was ob-
In h ibitor 26: alkylation of 4-iodopipecolic derivative 37 (X
) I) with 3-pyridinemethanethiol,29b,c as described above, gave
38b in 70% yield. Following the general procedures described
previously, this compound was converted into inhibitor 26: mp
68-74 °C. Rf 0.56 (9:1 CHCl3/MeOH). [R]20 - 1.1° (c 1.0,
D
MeOH). 1H NMR (DMSO-d6) δ 8.53 (d, J ) 2 Hz, 1H), 8.45
(dd, J ) 5, 1.5 Hz, 1H), 7.78-7.73 (m, 2H), 7.37 (s, 1H), 7.36-
7.20 (m, 5H), 7.16 (m, 1H), 7.00 (d, J ) 7.5 Hz, 2H), 6.92 (dd,
J ) 8, 6.5 Hz, 1H), 4.99 (broad d, J ) 5 Hz, 1H), 4.21 (m, 1H),
4.10 (d, J ) 14.5 Hz, 1H, part of AB), 3.92 (d, J ) 14 Hz, 1H,
part of AB), 3.83 (s, 2H), 3.80 (m, 1H), 3.18 (broad d, J ) 12
Hz, 1H), 2.99 (dd, J ) 13.5, 3 Hz, 1H), 2.71 (dd, J ) 13.5, 11
Hz, 1H), 2.68-2.55 (m, 2H), 2.12 (s, 6H), 2.05-1.81 (m, 4H),
1.49 (m, 1H), 1.24 (s, 9H). 13C NMR (DMSO-d6) δ 170.8, 166.6,
154.1, 149.2, 147.5, 146.1, 139.1, 135.8, 134.4, 129.8, 128.9,
128.3, 127.4, 125.3, 123.7, 123.1, 69.8, 69.4, 67.9, 57.1, 52.7,
51.5, 49.6, 35.6, 33.3, 31.0, 29.9, 27.9, 15.4. HRMS (FAB) m/z
1
tained as a colorless oil (0.151 g, 69% yield). H NMR (CDCl3)
δ 7.35-7.16 (m, 5H), 5.88 (broad s, 1H), 4.82 (broad m, 1H),
3.88 (broad m, 1H), 3.80 (broad m, 1H), 3.54 (broad m, 1H),
3.11-2.86 (m, 3H), 2.67 (dd, J ) 13, 6 Hz, 1H, part of ABX),
2.61 (dd, J ) 11, 3 Hz, 1H, part of ABX), 2.36-2.22 (m, 2H),
2.02-1.90 (m, 1H), 1.90-1.60 (m, 4H), 1.59-1.38 (m, 7H), 1.35
(s, 18H).
Carbamate 44 (0.143 g, 0.28 mmol) was dissolved in 4 N
HCl-dioxane (2 mL). After the mixture was stirred for 20 min,
volatiles were removed under a stream of nitrogen to give a
white solid (0.128 g, 95% yield).
C
36H49N4O4S (MH+) calcd: 633.3475; found: 633.3458. HPLC
homogeneity: 95.8% (system A); 97.7% (system B).
In h ibitor 28: 4-iodopipecolic amide derivative 37 (X ) I,
8.00 g, 19.5 mmol) and potassium thioacetate (3.35 g, 29.3
mmol) were dissolved in DMF (20 mL), and the mixture was
heated to 45 °C for 2 h. The reaction mixture was then diluted
with EtOAc and washed twice with water and then brine. The
extract was dried (MgSO4) and concentrated to give a solid
residue that was purified by flash chromatography on silica
gel using 4:1 to 2:1 hexane/EtOAc as eluent. (2S,4R)-4-
thioacetoxypipecolic tert-butylamide (38, SR ) SAc) was
isolated as a solid (4.80 g, 68% yield). MS (FAB) m/z 359
(MH+).
The crude salt from above (0.052 g, 0.11 mmol), 2′,6′-
dimethylphenoxyacetic acid (0.023 g, 0.13 mmol), and BOP
coupling reagent (0.060 g, 0.14 mmol) were dissolved in DMF
(2.2 mL). Diisopropylethylamine (95 µL, 0.55 mmol) was added
and the mixture stirred overnight at room temperature. The
reaction mixture was then diluted with EtOAc (25 mL),
washed with saturated aqueous NaHCO3 (2 × 25 mL), water
(25 mL), and brine (25 mL), and dried (MgSO4). Evaporation
of solvents under reduced pressure followed by purification by
flash chromatography on silica gel using 50% EtOAc/hexane
as eluent gave 29 as a white solid (0.030 g, 50% yield): mp
The thioacetate from above (4.80 g, 13.4 mmol) was dis-
solved in 1:1 MeOH/THF (64 mL) and the solution cooled in
an ice-water bath. NaOH (1 N, 16.1 mL, 16.1 mmol) was
added and the mixture stirred 1.5 h at 0 °C. The reaction
119-124.5 °C. Rf 0.67 (9:1 CHCl3/MeOH). [R]20 -5.0° (c 1.0,
D

A New P3-P2 Ligand for HIV Protease Inhibitors
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6 1107
MeOH). 1H NMR δ 7.33-7.25 (m, 4H), 7.25-7.18 (m, 1H),
7.01-6.90 (m, 4H), 5.99 (broad s, 1H), 4.35 (m, 1H), 4.22 (d, J
) 15 Hz, 1H, part of AB), 4.05 (d, J ) 15 Hz, 1H, part of AB),
3.97 (broad m, 1H), 3.83 (broad s, 1H), 3.11-2.95 (m, 3H), 2.73
(dd, J ) 13.5, 5.5 Hz, 1H, part of ABX), 2.67 (dd, J ) 10.5, 3.5
Hz, 1H, part of ABX), 2.37 (dd, J ) 13.5, 7 Hz, 1H, part of
ABX), 2.35 (dd, J ) 11, 3 Hz, 1H, part of ABX), 2.08 (s, 6H),
2.02-1.61 (m, 4H), 1.58-1.21 (m, 8H), 1.33 (s, 9H). 13C NMR
(DMSO-d6) δ 167.7, 167.0, 154.2, 138.6, 136.2, 129.0, 128.8,
128.0, 126.0, 124.2, 70.0, 68.1, 66.1, 58.9, 57.3, 53.7, 51.0, 34.8,
33.5, 31.2, 29.7, 28.3, 28.2, 28.1, 25.5, 24.6, 15.7. HRMS (FAB)
m/z calcd. for C34H50N3O4 (MH+): 564.3801; found: 564.3792.
HPLC homogeneity: 94.1% (system A); 96.6% (system B).
(dd, J ) 8, 6.5 Hz, 1H), 6.61 (d, J ) 9 Hz, 2H), 5.96 (broad s,
2H), 5.08 (broad d, J ) 6 Hz, 1H), 4.11-4.02 (m, 1H), 4.10 (d,
J ) 14 Hz, 1H, part of AB), 3.92 (d, J ) 14 Hz, 1H, part of
AB), 3.76 (m, 1H), 3.34 (dd, J ) 4 Hz, 1H, part of ABX, overlap
with water signal), 3.10 (dd, J ) 14, 3 Hz, 1H, part of ABX),
2.94 (dd, J ) 13.5, 8.5 Hz, 1H, part of ABX), 2.80-2.65 (m,
3H), 2.14 (s, 6H), 1.98 (m, J ) 6.5 Hz, 1H), 0.86 (d, J ) 6.5
Hz, 3H), 0.81 (d, J ) 6.5 Hz, 3H). 13C NMR (DMSO-d6) δ 166.7,
153.8, 152.3, 138.9, 129.9, 128.8, 128.5, 128.3, 127.4, 125.3,
123.8, 123.3, 112.2, 71.6, 69.6, 56.7, 52.6, 52.3, 34.2, 25.9, 19.65,
19.60, 15.2. HRMS (FAB) m/z C30H40N3O5S (MH+) calcd:
554.2689; found: 554.2700. HPLC homogeneity: 99.5% (sys-
tem A); >99.5% (system B).
In h ibitor 30: epoxide 4026 (1.30 g, 5 mmol) was dissolved
in 2:1 EtOH/isobutylamine (3 mL), and the solution was stirred
overnight at 50 °C under an argon atmosphere. Volatiles were
then removed under vacuum, and the residue was dissolved
in CH2Cl2 (35 mL). Saturated aqueous NaHCO3 (15 mL),
4-nitrobenzenesulfonyl chloride (1.06 g, 4.8 mmol), and NaH-
CO3 (0.42 g, 4.8 mmol) were added, and the biphasic mixture
was vigorously stirred for 6 h at room temperature. The
organic layer was separated, washed with water, dried (Mg-
SO4), and concentrated. The residue was crystallized from
EtOAc/hexane to give 45 as a solid (1.85 g, 74% yield): mp
Con clu sion
Incorporation of 2′,6′-dimethylphenoxyacetyl in the
P2-P3 position of peptidomimetic-based HIV protease
inhibitors has led to extremely potent inhibitors of the
enzyme and viral replication in vitro. In particular, in
the context of our palinavir class of protease inhibitors,
substitution of quinaldic-valine by this new achiral,
high-affinity ligand resulted in a decrease in structural
complexity and molecular size, while providing some of
the most potent antiviral agents reported to date in this
field. Furthermore, some of the compounds displayed
promising pharmacokinetic properties, including high
oral bioavailability in rat and dog. On the basis of the
analysis of X-ray structures and molecular modeling,
the high affinity of 2′,6′-dimethylphenoxyacetyl toward
the enzyme active site is attributed, in part, to binding
of one of the methyl groups inside the S2 pocket of the
protease, to favorable entropic factors, and to a ligand
minimal energy conformation which closely resembles
that of the bioactive conformation.
168-169 °C. Rf 0.75 (9:1 CHCl3/MeOH). [R]25 +15.9° (c 1.0,
D
CHCl3). IR (KBr) ν 3475-3220, 1655, 1636, 1595 cm-1 1H
.
NMR (CDCl3) δ 8.33 (d, J ) 9 Hz, 2H), 7.98 (d, J ) 9 Hz, 2H),
7.38-7.21 (m, 5H), 4.61 (broad d, J ) 5 Hz, 1H), 3.80 (m, 3H),
3.21 (d, J ) 5.5 Hz, 2H), 3.01 (d, J ) 9.5 Hz, 2H), 2.98 (m,
1H), 2.90 (m, 1H), 1.89 (m, J ) 7 Hz, 1H), 1.38 (s, 9H), 0.90 (t,
J ) 6 Hz, 6H). 13C NMR (CDCl3) δ 156.4, 150.0, 145.0, 137.5,
129.4, 128.6, 128.5, 126.7, 124.3, 80.1, 72.2, 57.5, 55.2, 52.5,
35.6, 28.2, 26.9, 20.0, 19.8. MS (FAB) m/z 522 (MH+), 422 (MH+
- Boc). Anal. Calcd for C25H35N3O7S: C, 57.76; H, 6.76; N,
8.06. Found: C, 57.59; H, 6.84; N, 8.02.
Carbamate 45 (0.480 g, 0.9 mmol) was stirred for 1 h in 1:1
TFA/CH2Cl2 (5 mL). Volatiles were removed under reduced
pressure and the residue coevaporated twice with EtOAc (5
mL). The residue was dissolved in DMF (5 mL), and N-
methylmorpholine (600 µL, 5.5 mmol, 6 equiv), 2′,6′-dimeth-
ylphenoxyacetic acid (0.180 g, 1.0 mmol), and TBTU (0.353 g,
1.1 mmol) were added. The mixture was stirred for 2 h at room
temperature. The reaction was quenched by pouring into 1 N
NH4OH (100 mL), and the tan-colored precipitate was collected
by filtration. The solid was washed with water and dried in
air. Compound 46 was obtained as a tan-colored solid (0.57 g,
Ack n ow led gm en t. We are grateful to Anick Berg-
eron, Colette Boucher, and Serge Valois for analytical
support and to Dr. J ean-Marie Ferland for the synthesis
of compound 24. We thank Dr. Louise Doyon for EC50
measurements in the presence of human serum.
Refer en ces
quantitative): mp 60-75 °C. [R]25 +9.8° (c 1.0, MeOH). 1H
D
(1) (a) Khol, 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 HIV-1 protease is required for viral infectivity. Proc. Natl.
Acad. Sci. U.S.A. 1988, 85, 4686-4690. (b) Kramer, R. A. et al.
Science 1986, 231, 1580-1584.
(2) Kempf, D. J .; Sham, H. L. HIV protease inhibitors. Curr. Pharm.
Des. 1996, 2, 225-246.
(3) (a) Deeks, S. G.; Smith, M.; Holodniy, M.; Kahn, J . O. HIV-1
protease inhibitors. A review for clinicians. J . Am. Med. Assoc.
1997, 277, 145-153. (b) Flexner, C. HIV-protease inhibitors. N.
Engl. J . Med. 1998, 338, 1281-1292. (c) Glaxo Wellcome/Vertex′
Agenerase wins accelerated FDA approval. Pharma Marketletter
1999, April 26, 18.
(4) (a) Lamarre, D.; Croteau, G.; Wardrop, E.; Bourgon, L.; Thi-
beault, D.; Clouette, C.; Vaillancourt, M.; Cohen, E.; Pargellis,
C.; Yoakim, C.; Anderson, P. C. Antiviral properties of palinavir,
a potent inhibitor of the human immunodeficiency virus type 1
(HIV-1) protease. Antimicrob. Agents Chemother. 1997, 965-
971. (b) Gorys, V.; Soucy, F.; Yoakim, C.; Beaulieu, P. L.
Substituted pyrrolidine derivatives as HIV protease inhibitors.
U.S. Patent 5 552 405, 1996. (c) Anderson, P. C.; Soucy, F.;
Yoakim. C.; Lavalle´e, P.; Beaulieu, P. L. Substituted pipecolinic
acid derivatives as HIV protease inhibitors. U.S. Patent 5 545
640, 1997, and 5 807 870, 1998
NMR (CDCl3) δ 8.38 (d, J ) 8.5 Hz, 2H), 7.99 (d, J ) 8.5 Hz,
2H), 7.38-7.22 (m, 5H), 7.05 (broad d, J ) 8 Hz, 1H), 7.05-
6.93 (m, 3H), 4.25 (m, J ) 4 Hz, 1H), 4.21 (d, J ) 15 Hz, 1H,
part of AB), 4.13 (d, J ) 15 Hz, 1H, part of AB), 4.11 (m, 1H),
3.98 (broad m, J ) 4 Hz, 1H), 3.32 (dd, J ) 15, 4 Hz, 1H, part
of ABX), 3.25 (dd, J ) 15, 8 Hz, 1H, part of ABX), 3.15 (dd, J
) 14, 5 Hz, 1H, part of ABX), 3.05-2.98 (dd, J ) 9.5 Hz, 1H,
part of ABX), 3.04 (d, J 8 Hz, 2H), 2.12 (s, 6H), 1.93 (m, J )
7 Hz, 1H), 0.92 (t, J ) 6.5 Hz, 6H). 13C NMR (CDCl3) δ 169.6,
153.8, 150.1, 144.8, 137.2, 130.2, 129.3, 129.2, 128.8, 128.5,
126.9, 124.9, 124.4, 72.3, 70.0, 57.9, 54.3, 52.7, 35.1, 27.0, 20.0,
19.9, 16.0. MS (FAB) m/z 584 (MH+). Anal. Calcd for
C
30H37N3O7S: C, 61. 73; H, 6.39; N, 7.20. Found: C, 61.88; H,
6.51; N, 7.30.
Nitro derivative 46 (0.490 g, 0.8 mmol) was hydrogenated
(1 atm H2 gas) over 20% Pd(OH)2/C (50 mg) in MeOH (20 mL)
for 6 h. The suspension was filtered through a 0.45 µM filter,
and the filtrate was evaporated to dryness under reduced
pressure. The residue was dissolved in MeOH (3 mL), and
water (25 mL) was added to precipitate the product. Crude
30 was collected by filtration, washed with water, and dried
in air. The material was purified by flash chromatography on
silica gel using 1:1 EtOAc/hexane as eluent. Compound 30 was
obtained as a white solid (0.365 g, 78% yield). Mp 60-75 °C.
(5) Tong, L.; Pav, S.; Mui, S.; Lamarre, D.; Yoakim, C.; Beaulieu,
P.; Anderson, P. C. Crystal structures of HIV-2 protease in
complex with inhibitors containing the hydroxyethylamine
dipeptide isostere. Structure 1995, 3, 33-40.
(6) Navia, M. A. Chaturvedi, P. R. Design principles for orally
bioavailable drugs. Drug Discovery Today 1996, 1, 179-189.
(7) Kempf, D. J . Progress in the discovery of orally bioavailable
inhibitors of HIV protease. Perspect. Drug Discovery Des. 1995,
2, 427-436.
Rf 0.45 (1:2 hexane/EtOAc). [R]25 +19.6° (c 1.0, MeOH). 1H
D
NMR (DMSO-d6) δ 7.89 (d, J ) 9.5 Hz, 1H), 7.42 (d, J ) 8.5
Hz, 2H), 7.28-7.21 (m, 5H), 7.17 (m, 1H), 7.02 (m, 2H), 6.93

1108 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 6
Beaulieu et al.
(8) (a) Ghosh, A. K.; Kincaid, J . F.; Walters, D. E.; Chen, Y.;
Chaudhuri, N. C.; Thompson, W. J .; Culberson, C.; Fitzgerald,
P. M. D.; Lee, H. Y.; Mckee, S. P.; Munson, P. M.; Duong, T. T.;
Darke, P. L.; Zugay, J . A.; Schleif, W. A.; Axel, M. G.; Lin, J .;
Huff, J . R. Nonpeptidal P2 ligands for HIV protease inhibitors:
structure-based design, synthesis, and biological evaluation. J .
Med. Chem. 1996, 39, 3278-3290 and references therein. (b)
Kim, E. E.; Baker, C. T.; Dwyer, M. D.; Murcko, M. A.; Rao, B.
G.; Tung, R. D.; Navia, M. A. Crystal structure of HIV-1 protease
(17) Pargellis, C. A.; Morelock, M. M.; Graham, E. T.; Kinkade, P.;
Pav, S.; Lubbe, K.; Lamarre, D.; Anderson, P. C. Determination
of kinetic rate constants for the binding of inhibitors to HIV-1
protease and for the association and dissociation of active
homodimer. Biochemistry 1994, 33, 12527-12534.
(18) For a competitive inhibitor, the relationship between IC50 and
Ki values is described by the following equation: IC50
) +
1/2Et
Ki (1 + S/Km) in which Et represents the enzyme concentration.
For compound 27, the term Ki (1 + S/Km) is negligible compared
to 1/2Et, such that the IC50 value reflects the enzyme concentra-
tion rather than the Ki, allowing for the 266-fold discrepancy
observed between these values.
in complex with VX-478,
a potent and orally bioavailable
inhibitor of the enzyme. J . Am. Chem. Soc. 1995, 117, 1181-
1182. (c) Chen, X.; Li, L.; Kempf, D. J .; Sham, H.; Wideburg, N.
E.; Saldivar, A.; Vasavanonda, S.; Marsh, K. C.; McDonald, E.;
Norbeck, D. W. Evaluation of furofuran as a P2 ligand for
symmetry-based HIV protease inhibitors. Bioorg. Med. Chem.
Lett. 1996, 6, 2847-2852. (d) Ghosh, A. K.; Kincaid, J . F.; Cho,
W.; Walters, D. E.; Krishnan, K.; Hussain, K. A.; Koo, Y.; Cho,
H.; Rudall, C.; Holland, L.; Buthod, J . Potent HIV protease
inhibitors incorporating high-affinity P2-ligands and (R)-(hy-
droxyethylamino)sulfonamide isostere. Bioorg. Med. Chem. Lett.
1998, 8, 687-690.
(19) Beaulieu, P. L.; Wernic, D. Abraham, N.; Anderson, P. C.; Bogri,
T.; Bousquet, Y.; Croteau, G.; Guse, I.; Lamarre, D.; Liard. F.;
Paris, W.; Thibeault, D.; Pav, S.; Tong, L. Potent HIV protease
inhibitors containing a novel (hydroxyethyl)amide isostere. J .
Med. Chem. 1997, 40, 2164-2176.
(20) (a) Unpublished results. (b) Doyon, L.; Croteau, G.; Thibeault,
D.; Poulin, F.; Pilote, L.; Lamarre, D. Second locus involved in
human immunodeficiency virus type 1 resistance to protease
inhibitors. J . Virol. 1996, 70, 3763-3769. (c) Croteau, G.; Doyon,
L.; Thibeault, D.; McKercher, G.; Pilote, L.; Lamarre, D.
(9) J ungheim, L. N.; Shepherd, T. A.; Baxter, A. J .; Burgess, J .;
Hatch, S. D.; Lubbehusen, P.; Wiskerchen, M.; Muesing, M.
Potent human immunodeficiency virus type 1 protease inhibitors
that utilize noncoded D-amino acids as P2/P3 ligands. J . Med.
Chem. 1996, 39, 96-108.
Impaired fitness of human immunodeficiency virus type
1
variants with high-level resistance to protease inhibitors. J .
Virol. 1997, 71, 1089-1096.
(10) (a) Kaldor, S.; Kalish, V. J .; Davies, II, J . F.; Shetty, B. V.; Fritz,
J . E.; Appelt, K.; Burgess, J . A.; Campanale, K. M.; Chirgadze,
N. Y.; Clawson, D. K.; Dressman, B. A.; Hatch, S. D.; Khalil, D.
A.; Kosa, M. B.; Lubbehusen, P. P.; Muesing, M. A.; Patick, A.
K.; Reich, S. H.; Su, K. S.; Tatlock, J . H. Viracept (nelfinavir
mesylate, AG1343): a potent, orally bioavailable inhibitor of
HIV-1 protease. J . Med. Chem. 1997, 40, 3979-3985. (b) Shetty,
B. V.; Kosa, M. B.; Khalil, D. A.; Webber, S. Antimicrobial Agents
Chemotherapy 1996, 40, 110-114. (c) Munroe, J . E.; Hornback,
W. J .; Campbell, J . B.; Ouellette, M. A.; Hatch, S. D.; Muesing,
M. A.; Wiskerchen, M.; Baxter, A. J .; Su, K.; Campanale, K.
LY316340: a potent HIV-1 protease inhibitor containing a high
affinity octahydrothienopyridine hydroxyethylamine isostere.
Bioorg. Med. Chem. Lett. 1995, 5, 2885-2890. (d) Randad, R.
S.; Lubkowska, L.; Bhat, T. N.; Munshi, S.; Gulnik, S. V.; Yu,
B.; Erickson, J . W. Symmetry-based HIV protease inhibitors:
rational design of 2-methylbenzamides as novel P2/P2′ ligands.
Bioorg. Med. Chem. Lett. 1995, 5, 1707-1712.
(21) Beaulieu, P. L.; Lavalle´e, P.; Abraham, A.; Anderson, P. C.;
Boucher, C.; Bousquet, Y.; Duceppe, J .-S.; Gillard, J .; Gorys, V.;
Grand-Maˆıtre, C.; Grenier, L.; Guindon, Y.; Guse, I.; Plamondon,
L.; Soucy, F.; Valois, S.; Wernic, D.; Yoakim, C. Practical,
stereoselective synthesis of palinavir, a potent HIV protease
inhibitor. J . Org. Chem. 1997, 62, 3440-3448.
(22) Knorr, R.; Trzeciak, A.; Bannwarth, W.; Gillessen, D. New
coupling reagents in peptide chemistry. Tetrahedron Lett. 1989,
30, 1927-1930.
(23) Moreau, C.; Rouessac, F. Le comportement thermique de trois
alce´nylphe´nols. Les m- et p-hydroxyphe´nyl-5 pente`ne-1 et l′o-
me´thyl m-hydroxyphe´nyl-5 pente`ne-1. Bull. Soc. Chim. Fr. 1973,
3427-3432.
(24) Bodansky, M. In Principles of Peptide Synthesis; Springler-
Verlag: New York, 1984.
(25) (a) Gillard, J .; Abraham, A.; Anderson, P. C.; Beaulieu, P. L.;
Bogri, T.; Bousquet, Y.; Grenier, L.; Guse, I.; Lavalle´e, P.
Preparation of (2S,4R)-4-hydroxypipecolic acid and derivatives.
J . Org. Chem. 1996, 61, 2226-2231. (b) Bousquet, Y.; Anderson,
P. C.; Bogri, T.; Duceppe, J .-S.; Grenier, L.; Guse, I. Preparation
of enantiopure 4-oxygenated pipecolic acid derivatives. Tetra-
hedron 1997, 53, 15671-15680.
(26) (a) Beaulieu, P. L.; Wernic, D.; Duceppe, J .-S.; Guindon, Y. Large
scale preparation of (2S,3S)-N-boc-3-amino-1,2-epoxy-4-phe-
nylbutane: a key building block for HIV-protease inhibitors.
Tetrahedron Lett. 1995, 36, 3317-3320. (b) Beaulieu, P. L.;
Wernic, D. Preparation of aminoalkyl chlorohydrin hydrochlo-
rides: key building blocks for hydroxyethylamine-based HIV
protease inhibitors. J . Org. Chem. 1996, 61, 3635-3645.
(27) Houpis, I. N.; Molina, A.; Reamer, R. A.; Lynch, J . E.; Volante,
R. P.; Reider, P. J . Towards the synthesis of HIV-protease
inhibitors. Synthesis of optically pure 3-carboxyl-decahydroiso-
quinolines. Tetrahedron Lett. 1993, 34, 2593-2596.
(28) (a) Betagari, R.; Hopkins, J . L.; Thibeault, D.; Emmanuel, M.
J .; Chow, G. C.; Skoog, M. T.; deDreu, P.; Cohen, K. A. Rapid,
sensitive and efficient HPLC assays for HIV-1 proteinase.
Biochem. Biophys. Methods 1993, 27, 191-197. (b) Pav, S.;
Lubbe, K.; Doˆ, F.; Lamarre, D.; Pargellis, C.; Tong, L. Microtube
batch protein crystallization: applications to human immuno-
deficiency virus type 2 (HIV-2) protease and human renin.
Proteins: Struct., Funct., Genet. 1994, 20, 98-102.
(29) (a) Fischer, A.; King, M. J .; Robinson, F. P. Inductive substituent
constants from dissociation of methyl-substituted 4-picolinium
ions. Can. J . Chem. 1978, 56, 3072-3077. (b) Fischer, A.; King,
M. J .; Robinson, F. P. Inductive substituent constants from
dissociation of methyl-substituted 3-picolinium ions. Can. J .
Chem. 1978, 56, 3068-3071. (c) Fischer, A.; King, M. J .;
Robinson, F. P. Inductive substituent constants from dissociation
of methyl-substituted 2-picolinium ions. Can. J . Chem. 1978, 56,
3059-3067.
(11) Moss, N.; Beaulieu, P.; Duceppe, J .-S.; Ferland, J .-M.; Gauthier,
J .; Ghiro. E.; Goulet, S.; Guse, I.; Llina`s-Brunet, M.; Plante, R.;
Plamondon, L.; Wernic, D.; De´ziel, R. Ureido-based peptidomi-
metic inhibitors of herpes simplex virus ribonucleotide reduc-
tase: an investigation of inhibitor bioactive conformation. J .
Med. Chem. 1996, 39, 2178-2187 and references therein.
(12) Roberts, N. A.; Martin, J . A.; Kinchington, D.; Broadhurst, A.
V.; Craig, J . C.; Duncan, I. B.; Galpin, S. A.; Handa, B. K.; Kay,
J .; Khro¨n, A.; Lambert, R. W.; Merrett, J . H.; Mills, J . S.; Parkes,
K. E. B.; Redshaw, S.; Ritchie, A. J .; Taylor, D. L.; Thomas, G.
J .; Machin, P. J . Rational design of peptide-based HIV proteinase
inhibitors. Science 1990, 248, 358-361.
(13) (a) Sham, H. L.; Kempf, D. J .; Molla, A.; Marsh, K. C.; Kumar,
G. D.; Chen, C.-M.; Kati, W.; Stewart, K.; Lal, R.; Hsu, A.;
Betebenner, D.; Korneyeva, M.; Vasavanonda, S.; McDonald, E.;
Saldivar, A.; Wideburg, N.; Chen, X.; Niu, P.; Park, C.; J ayanti,
V.; Grabowski, B.; Granneman, G. R.; Sun, E.; J apour, A. J .;
Leonard, J . M.; Plattner, J . J .; Norbeck, D. W. ABT-378, a highly
potent inhibitor of the human immunodeficiency virus protease.
Antimicrob. Agents Chemother. 1998, 42, 3218-3224. (b) Mim-
oto, T.; Kato, R.; Takaku, H.; Nojima, S.; Terashima, K.; Misawa,
S.; Fukazawa, T.; Ueno, T.; Sato, H.; Shintani, M.; Kiso, Y.;
Hayashi, H. Structure-activity relationship of small-sized HIV
protease inhibitors containing allophenylnorstatine. J . Med.
Chem. 1999, 42, 1789-1802.
(14) Because of the similar binding affinities of compound 27 for
HIV-1 and HIV-2 proteases (Ki ) 6 and 23 pM, respectively),
we assume similar binding modes with the enzymes.
(15) (a) Stewart, J . J . P. MOPAC: a semiempirical molecular orbital
program. J . Comput.-Aid. Mol. Des. 1990, 4, 1-105. (b) Cramer,
C. J .; Truhlar, D. G. PM3-SM3: a general parametrization for
including aqueous solvation effects in the PM3 molecular orbital
model. J . Comput. Chem. 1992, 13, 1089-1097.
(16) (a) Liard, F.; Ghiro, E.; Paris, W.; Yoakim, C. Determination of
the HIV protease inhibitor BILA 2185 BS in rat plasma by
liquid-liquid extraction and high performance liquid chroma-
tography photodiode array detector. J . Pharm. Biomed. Anal.
1995, 14, 151-154. (b) Liard, F.; J aramillo, J .; Paris, W.;
Yoakim, C. Pharmacokinetic aspects of palinavir, an HIV
protease inhibitor, in Sprague-Dawley rats. J . Pharm. Sc. 1998,
87, 782-785.
J M990336N