Small-sized human immunodeficiency virus type-1 protease inhibitors containing allophenylnorstatine to explore the s2′ pocket

Article abstract of DOI:10.1021/jm9005115

A series of HIV protease inhibitor based on the allophenylnorstatine structure with various P₂′ moieties were synthesized. Among these analogues, we discovered that a small allyl group would maintain potent enzyme inhibitory activity compared t

Full text of DOI:10.1021/jm9005115

7604 J. Med. Chem. 2009, 52, 7604–7617
DOI: 10.1021/jm9005115
Small-Sized Human Immunodeficiency Virus Type-1 Protease Inhibitors Containing
,^
0
Allophenylnorstatine to Explore the S2 Pocket
Koushi Hidaka,† Tooru Kimura,† Hamdy M. Abdel-Rahman,† Jeffrey-Tri Nguyen,† Keith F. McDaniel,‡
William E. Kohlbrenner,‡ Akhteruzzaman Molla,‡ Motoyasu Adachi,§ Taro Tamada,§ Ryota Kuroki,§
Noriko Katsuki,† Yoshiaki Tanaka,† Hikaru Matsumoto,† Jun Wang,† Yoshio Hayashi,† Dale J. Kempf,‡ and
Yoshiaki Kiso*,†
†Department of Medicinal Chemistry, Center for Frontier Research in Medicinal Science, 21st Century COE Program,
Kyoto Pharmaceutical University, Yamashina-ku, Kyoto 607-8412, Japan, ‡Global Pharmaceutical Research and Development,
Abbott Laboratories, Abbott Park, Illinois 60064, and §Molecular Structural Biology Group, Quantum Beam Science Directorate,
Japan Atomic Energy Agency, Tokai, Ibaraki, 319-1195, Japan
Received April 22, 2009
0
A series of HIV protease inhibitor based on the allophenylnorstatine structure with various P2 moieties
were synthesized. Among these analogues, we discovered that a small allyl group would maintain potent
enzyme inhibitory activity compared to the o-methylbenzyl moiety in clinical candidate 1 (KNI-764, also
known as JE-2147, AG-1776, or SM-319777). Introduction of an anilinic amino group to 2 (KNI-727)
improved water-solubility and anti-HIV-1 activity. X-ray crystallographic analysis of 13k (KNI-1689)
0
with a β-methallyl group at P2 position revealed hydrophobic interactions with Ala28, Ile84, and Ile500
similar to that of 1. The presence of an additional methyl group on the allyl group in compound 13k
significantly increased anti-HIV activity over 1 while providing a rational drug design for structural
minimization and improving membrane permeability.
Introduction
to bind with the two catalytic aspartic acids of the enzyme.
Fundamentally, this binding inhibits the processing of the gag
and gag-pol polyproteins that are essential for propagation of
the infectious virion. On the basis of the transition-state mimic
concept, we previously reported a series of potent peptido-
mimetic HIV protease inhibitors containing allophenylnor-
statine5 [Apns: (2S,3S)-3-amino-2-hydroxy-4-phenylbutyric
acid] with a hydroxymethylcarbonyl (HMC) isostere. Inter-
actions between the HMC isostere and HIV-1 protease have
been proven by X-ray crystallography, NMR, and neutron
crystallography analyses to reveal hydrogen bonding interac-
tions between the HMC’s hydroxyl and carbonyl groups with
the corresponding carboxylic/carboxylate groups of the two
catalytic aspartic acids.6-8 This ideal transition-state mimic
forms the basis for the development of highly potent small
HIV protease inhibitors, such as 1 (KNI-764, also known as
JE-2147, AG-1776, or SM-319777)9,10 and 2 (KNI-727)5c,11
shown in Figure 1. The potency of these compounds resides in
Improvements in antiretroviral therapy have changed the
fatal human immunodeficiency virus type-1 (HIV-1a) infec-
tion to a manageable chronic illness.1 Highly active antire-
troviral therapy (HAART) using more than three drugs
ranging from HIV reverse transcriptase to protease inhibitors
brought about this remarkable breakthrough. Among this
effective concoction, HIV protease inhibitors are so essential
for HAART that the FDA has approved 10 different protease
inhibitors. Recent studies on HIV protease inhibitors includ-
ing FDA-approved tipranavir and darunavir focused on
drug-resistant HIV.2,3 However, clinically used HIV protease
inhibitors have been associated with the emergence of drug
resistant viruses,4 unfavorable side effects, and long-term high
dose requirements. These concerns urged us to rekindle our
effort to develop a larger variety of HIV protease inhibitors.
A substrate transition-state mimic thatpossesses one or two
hydroxyl groups is introduced in each HIV protease inhibitor
0
a conformationally constrained P1-P1 mimetic moiety,
Apns-Dmt [Dmt: (R)-5,5-dimethylthiazolidine-4-carboxylic
acid]. Compound 1 has been shown to be effective against
mutant proteases that are resistant to clinically available HIV
protease inhibitors.12 Mimoto and co-workers reported
further modifications from 1 to prevent glucuronidation of
the P2 phenolic hydroxyl group to improve the pharmacoki-
netic profile.13 Compound 1 was designed with an o-methyl-
*To whom correspondence should be addressed. Phone: þ81-75-595-
4635. Fax: þ81-75-591-9900. E-mail: kiso@mb.kyoto-phu.ac.jp.
The atomic coordinates have been deposited in the Protein Data
Bank, www.rscb.org (PDB code 3A2O).
^ On the 100th Anniversary of the Division of Medicinal Chemistry,
we present this work in this commemorative MEDI Centennial issue.
a Abbreviations: HIV, human immunodeficiency virus; AIDS, acquired
immune deficiency syndrome; HAART, highly active antiretroviral ther-
apy; Apns, allophenylnorstatine; HMC, hydroxymethylcarbonyl; Dmt,
(R)-5,5-dimethyl-1,3-thiazolidine-4-carboxylic acid; SAR, structure-activ-
ity relationship; Boc, tert-butyloxycarbonyl; BOP, benzotriazole-1-yloxy-
tris(dimethylamino)phosphonium hexafluorophosphate; EDC, 1-ethyl-
3-(3,3-dimethylaminopropyl)carbodiimide; HOBt, 1-hydroxybenzotri-
azole; QSAR, quantitative structure-activity relationship.
0
benzyl structure at the P2 position to accommodate for the
symmetric character of the HIV protease.10 The flexibility of
this benzyl moiety is thought to compensate for the binding
energy loss found in drug resistant protease mutants.14 Here-
0
in, we focused on the P2 o-methylbenzyl group of 1, mainly
r
pubs.acs.org/jmc
Published on Web 07/17/2009
2009 American Chemical Society

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7605
Figure 1. Structures for 1 and 2.
Chemistry
Introduction of various amines to the P20 position of 1 was
performed as shown in Scheme 1. Commercially available
amines were reacted with Boc-Dmt-OH (3) using BOP
as coupling reagent to afford intermediate 4. 2,6-Dichloro-
benzylamine, 2-chloro-6-fluorobenzylamine, 3-hydroxy-2-
methylbenzylamine, 3-fluoro-2-methylbenzylamine, and cis-
4-hydroxy-2-butenylamine were separately synthesized from
their corresponding substituted benzoic acids, benzyl bro-
mide, and allyl chloride. After deprotection of the Boc group
with HCl, Boc-Apns-OH was coupled by the EDC-HOBt
method (compound 5). Subsequent deprotection of the Boc
group and BOP coupling with 3-acetoxy-2-methylbenzoic
acid, followed by hydrolysis of the acetyl group, afforded
the desired compounds (6a-j, 7a-n). The synthesis of N-
protected 4-amino-2,6-dimethylphenoxyacetic acid and its
derivatives is shown in Scheme 2. 4-Nitro-2,6-dimethylphenol
was alkylated with ethyl bromoacetate and reduced to an
amino group to afford intermediate 8, followed by protection
with (Boc)2O to give intermediate 9. 4-(Boc-amino)-2,6-di-
methylphenoxyacetic acid 10 was obtained by saponification
of intermediate 9. A portion of intermediate 9 was deproto-
nated by sodium hydride and methylated with iodomethane
and simultaneously hydrolyzed to form a monomethylated
analogue 11. The amino group of intermediate 8 was fully
methylated by reductive alkylation using formaldehyde and
subsequently hydrolyzed to give dimethylated analogue 12.
Scheme 3 describes the synthesis of 2 derivatives starting
from the set of intermediates 5. For each intermediate
found in compound set 5, the Boc moiety was removed.
Then 4-(Boc)amino-2,6-dimethylphenoxyacetic acid 10 was
coupled using BOP, followed by a final deprotection to give
target compounds 13a, 13b, and 13e-l. In the case of com-
Figure 2. Schematic representation of the residues surrounding the
2-methylbenzyl group of 1.
0
because P2 optimization is limited to a small number of
0
moieties and substrate specificity at the S2 pocket of the
enzyme is relatively low. In the present study, structure-
activity relationships at the P2 position of allophenylnorsta-
0
tine-containing inhibitors were explored by introducing
various moieties to the structure of 1. To confirm our ob-
0
servations, similar P2 modifications along with some P2
improvements were applied to another reference inhibitor 2.
Strategy behind Structure-Activity Relationship Studies
In recent years, five complexes of 1 with a wild or mutant
type HIV-1 protease have been disclosed.14,15 These com-
plexes revealed hydrophobic int0eractions between the
o-methylbenzyl moiety of 1 and S2 pocket of the enzyme.
The PDB data (entries 1MSM and 1KZK) show proximity
between the methyl group of the o-methylbenzyl moiety and
the side chains of Ala28, Ile84, and Ile500 in a wi0ld type
protease, while the phenyl group occupies the S2 space
(Figure 2). In our previous structure-activity relationship
0
pounds with a P2 o-methylbenzyl group, a monomethylated
0
(SAR) study at the P2 position, ortho-methyl substitution on
amino analogue 11 was introduced using BOP to give com-
pound 13c, while the dimethylated amino analogue 13d was
synthesized using 12 without a final deprotection.
the benzyl structure was favored for inhibitory activity over
meta- or para-substitution.10 Herein, while keeping in mind
the importance of o-methyl group interactions, we introduced
va0riants of the benzyl structure. We also incorporated various
P2 non-benzyl structures to form van der Waals contacts with
the aforementioned key residues of the enzyme.
Results and Discussion
HIV-1 protease inhibitory activity of the synthesized com-
pounds was evaluated. Table 1 summarizes the enzymatic
Although 2 was reported as a potent dipeptide-type
HIV-1 protease inhibitor with low anti-HIV activity,5c we
speculated that the disappointing results in cell-based
assays were due to the compound’s high hydrophobic nature.
In order to improve its hydrophilicity and form hydrogen
bonding interactions with the carboxylic acid side chain found
in Asp30 of the enzyme, we introduced an additional para-
amino group to the aromatic P2 structure of 2 to generate a 4-
amino-2,6-dimethylphenoxyacetyl moiety, which was also
used for a malarial plasmepsin inhibitor study.16 Effective
0
assay results from compounds possessing a P2 benzyl struc-
tureas determined at50 and 1 nMof the inhibitor. A log linear
correlation was observed between the results obtained at 50
and 1 nM (r2 =0.93). This well-fitted log linear correlation
indicatesthatinhibitorydata determined at50nMis almost as
reliable as those obtained at 1 nM. Substitution at the ortho-
position on the phenyl ring afforded compounds 6a-g that
effectively exhibited high inhibitory activity (>90% at
50nM). Compounds with a double substitution on both ortho
positions (6a-e) exhibited potency similar to that of 1 at
50 nM. The most potent activity was exhibited by double
0
P2 moieties with potent HIV protease inhibition among the
derivatives of 1 were selected and applied to this combination.

7606 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Hidaka et al.
Scheme 1a
a Reagents: (a) amine (R-NH2), BOP, Et3N DMF; (b) 4 N HCl/dioxane, anisole; (c) Boc-Apns-OH, EDC, HOBt, Et3N, DMF; (d) 3-acetoxy-2-
methylbenzoic acid, BOP, Et3N, DMF; (e) 1 N aqueous NaOH, MeOH.
Scheme 2a
a Reagents: (a) BrCH2CO2Et, K2CO3 DMF; (b) 10% Pd-C, MeOH, H2; (c) (Boc)2O, THF-H2O; (d) 1 N aqueous NaOH, MeOH; (e) NaH, CH3I,
THF; (f) HCHO, 10% Pd-C, THF, H2.
Scheme 3a
a Reagents: (a) 4 N HCl/dioxane, anisole; (b) 10, 11, or 12, BOP, Et3N, DMF.
methyl-substituted 6a (82% at 1 nM, Ki = 2.4 pM). The
presence of aromatic halogen atoms reduced inhibitory
activity (6b-e) in the general order of methyl > chlorine >
fluorine, down to 24% inhibition at 1 nM for the case of

Article
Table 1. HIV Protease Inhibitory and Anti-HIV Activity of P2 -Benzyl Derivatives
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7607
a
0
HIV-1 PR % inhibition
compd
R1
R2
R3
R4
R5
at 50 nM
at 1 nM
anti-HIV-1IIIB EC50 (μM)
1
CH3
CH3
CH3
Cl
H
H
H
H
99b
99c
99
98
97
91
95
97
95
82d
52
72
0.033
nd
6a (KNI-814)
6b (KNI-1526)
6c (KNI-1365)
6d (KNI-1405)
6e (KNI-1367)
6f (KNI-1098)
6g (KNI-1366)
6h (KNI-1352)
6i (KNI-1174)
6j (KNI-1359)
H
H
H
H
H
OH
F
CH3
Cl
Cl
F
H
H
82
77
66
50
24
41
53
31
nd
nd
H
H
0.032
0.039
0.032
0.051
0.120
0.056
0.079
0.123
0.617
H
H
Cl
F
H
H
F
H
H
CH3
CH3
H
H
H
H
H
H
H
H
H
H
H
CH3
CH3
H
H
H
H
CH3
CH3
H
H
a nd: not determined. b Ki = 0.031 nM. c Ki = 0.0024 nM. d Reference 17.
0
difluoro-substituted 6d. A hydroxyl moiety was introduced
at the meta-position (R2 group) in consideration of the
the S2 space while maintaining interactions with the enzyme’s
hydrophobic residues. Gradually decreasing the ring size from
a seven- to four-membered cyloalkyl ring, we observed that
the five-membered rin0g compound (7g) is most likely to have
the better fit in the S2 space. The relatively small size of the
five-membered ring compared to that of the benzyl structure
0
symmetric nature of the protease (cf. the P2 and P2
residues) to afford inhibitor 6f. However, compound 6f
exhibited lower inhibitory activity than 1, which does not
0
possess a P2 m-hydroxyl moiety. This result suggests a
0
disruption of interactions between the o-methyl group and
the side chains of Ala28, Ile84, and Ile500. Compound 6g
possessing a m-fluorine showed a lessened disruption of
interactions. From the crystal structure of 1 and HIV-1
protease, the R4 group is believed to be near the R1
o-methyl group. As expected, compound 6h maintained
inhibitory activity because the R4 group could alternate for
the R1 group. On the other hand, the R5 group is located in
proximity to the amide group of Gly49. The presence of an
R5 group would essentially interfere with the enzyme’s flap
region. To no big surprise, compounds 6i and 6j possessing
a methyl R5 exhibited relatively reduced potencies (cf. 6h-
j). These results brought to light the importance of side
chain interactions with Ala28, Ile84, and Ile500. Inhibitory
activity against wild-type HIV-1 of these compounds was
examined. Among them, only three compounds, 6b-d,
possessed potencies similar to that of 1.
motivated us to focus on downsizing the P2 group. The
0
S2 pocket seems to prefer an n-propyl moiety (7j) over an
ethyl group (7i). Compound 7k possessing an allyl group
instead of an n-propyl group (7j) exhibited increased inhibi-
tory activity (93% vs 88% at 50 nM, respectively), while a
triple bond slightly decreased activity (7l, 90%). We presume
that a more constrained conformation with a double bond
would promote hydrophobic contacts with the side chains of
Ala28, Ile84, and Ile500. An additional methyl group on the
allyl moiety intensified potency to 98% (7m). Interestingly,
compound 7m with a β-methallyl moiety also exhibited strong
antiviral potency against HIV-1 (EC50=17 nM, using wild-
type pNL4-3) over that of 1 (EC50=60 nM, using wild-type
pNL4-3). These results confirmed the effectiveness of small
ligands on inhibitory activities against HIV protease and the
virus. Compound 7n with an additional hydroxymethyl in-
stead of an allyl group (7k) possessed equipotent enzymatic
inhibitory activity. Anti-HIV activity of inhibitor 7n was
10-fold less effective than inhibitor 7k. This difference sug-
gests that the hydroxymethyl moiety is detrimental to the
delivery of the inhibitor to the HIV protease target.
Table 2 depicts the inhibitory activ0ity of compounds pos-
sessing a non-benzyl structureatthe P2 position. The presence
of a chiral indane (7a and 7b) instead of o-methylbenzyl
structure (1) maintained potent activity against the enzyme.
This result indicates tolerable conformational constraints of
the o-methylbenzyl structure. Chiral compound 7a exhibited
slightly higher activity than the compound with opposite
chirality (7b). An additional hydroxyl group on the indane
structure attenuated potency (7c and 7d). The stereochemistry
on the indanol compound possessing higher potency (7d) is
consistent with that of indinavir.18 The crystal structure of a
complex of indinavir and HIV protease revealed direct inter-
actions between the indanolphenyl ring and side chains of
Ala28, Ile84, and Ile500, thereby pushing the hydroxyl group
toward Asp30. In the case of compound 7d, we assume that
After completing our work on the 1 derivatives,19 we were
made aware that some compounds (6g-j and 7a,b,e-h,k-m)
have been synthesized via different routes, assayed, and
declared as patented inventions by Agouron Pharmaceuticals,
Inc. using1asreference compound.20 Consequently, although
reached by convergent research, we are not claiming the
aforementioned compounds to be novel. However, we believe
that the results from our SAR studies to be of interest.
Although 2 was also reported to exhibit potent inhibi-
tion against HIV protease, it exhibited low activity against
HIV (Table 3).11 We assume that its excessive hydrophobic
character reduces its membrane permeability. Not only to
improve its hydrophilic profile but also to induce additional
0
the interactions of the P2 moiety are similar to that of
indinavir. Cycloalkyl groups (7e-h) were introduced to fill

7608 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Hidaka et al.
0
Table 2. HIV Protease Inhibitory and Anti-HIV Activity of P2 Derivatives
a Ki = 0.19 nM. b pNL4-3 strain.
hydrogen bonding interactions with Asp300, a p-amino group
was inserted on the P2 ligand. The resultant compound
13a exhibited equipotent enzymatic inhibitory activity as 2,
89-fold increase in water solubility, and 10-fold improvement
in anti-HIV activity. This improvement in anti-HIV activity
using a p-amino group was more effective than our previously
reported examples using pseudosymmetric compounds.21
Compound 13b possessing an o-methylbenzyl group at the
P20 position, not unlike 1, exhibited enhanced activity against
both the enzyme and virus. Additional methylation of the
R1 amino group (13c and 13d) failed to improve potency.
Consequently, in order to improve inhibitory profiles of the
compound, we only focused on the 0free R1 amino group for
further modifications. Eight other P2 ligands were selected for
combinatorial study. As expected, the HIV-1 protease inhibi-
tory activity profiles for compounds 13e-l were greater than
90% at 50 nM. Although most derivatives exhibited equal or
slightly lower potencies, only the indanol derivative (13h)
exhibited higher enzyme inhibition than the corresponding
1 analogue 7d. We observed a drastic decrease in anti-HIV
activity as a result of an additional hydroxymethyl group to
the allyl structure (cf. 13j and 13l). This result is consistent in
derivatives of 1 (cf. 7k and 7n). However, an additional methyl
group, i.e., β-methallyl, was significantly effective against
anti-HIV activity (13k, EC50=10 nM), similar to that of
derivative 7m.
fashion as previously reported complexes (Figure 4a,b).7,14,15
The additional P2 p-amino group was aligned near the side
chain of Asp300 with possible hydrogen bonding interactions.
The side chain of Asp300 was more distant than to that of
1 observed in PDB 1MSM (Figure 4c). The equipotency of 2
and compound 13a suggests that additional hydrogen bonding
interactions from the amino group could compensate for the
undesired steric hindrance from this same group. The
β-methallyl moiety in inhibitor 13k was within hydrophobic
contacts with the side chains of Ala28, Ile84, and Ile500 in the
0
S2 pocket (Figure 4d). These hydrophobic interactions are
believed toplay a similar role as the o-methyl group in 1. Onthe
other hand, another part of β-methallyl in compound 13k,
the allyl moiety, compensated for the absence of the phenyl
group found in compound 1, interacting with Val32 and Ile47.
These interactions would explain the relatively sustained ac-
tivity against HIV protease from 13k (Ki=0.83 nM) and 7m
(Ki=0.19 nM) compared to that of inhibitor 1(Ki=0. 031 nM).
In regard to water solubility, these analogues dissolved
moderately with a range of 1-13 mg/mL compared with
previously reported water-soluble prodrug forms (Table 3).22
However, the moderate solubility of these compounds would
be an asset for future pharmaceutical development as anti-HIV
therapeutical medicine. The anti-HIV activity of the com-
pounds seems to correlate to the potency against the enzyme
better than water solubility (r2 = 0.41 vs 0.16, from each
respective log linear equation).
We succeeded in obtaining an X-ray crystallographic struc-
ture of a complex of compound 13k (Figure 3) and HIV-1
protease. The data disclosed that the HMC of the inhi-
bitor interacts with the two catalytic Asp residues in a similar
At first sight there does not seem to be any significant
relationship between HIV-1 PR inhibition and antiviral activ-
ity. Only a coefficient of determination (r2) of 0.41 was observed

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7609
Table 3. Introduction of p-Amino Group into 2 and Its Derivativesc
a 43% inhibition at 1 nM inhibitor, Ki = 0.83 nM. b pNL4-3 strain. c HS: human serum. SBE: serum binding effect.
the enzyme, and intrinsic affinity for the enzyme. Whereas
X-ray crystal structures may be used to explain HIV-1 PR
inhibition, various factors influence the overall antiviral activ-
ity, thus rendering antiviral activity predictions somewhat more
challenging.
We performed a quantitative structure-activity relation-
ships (QSAR) study to correlate the biophysicochemical prop-
erties of the inhibitors with antiviral activity. A reliable
equation, using only four descriptors was derived (eq 1): All
possible permutations of 214 descriptors calculated with the
Molecular Operating Environment (MOE 2005.06, Chemical
Computing Group, Inc., Montreal, Canada) software were
evaluated and correlated with antiviral activity for compounds
Figure 3. Structure for 13k.
in the 2 derived series, as previously mentioned. Indeed, when a
drug is tested for antiviral activity, several factors are involved
including solvation, internalization into the cell, localization to

7610 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Hidaka et al.
Figure 5. Plot of calculated versus measured anti-HIV activity for
eq 1 for 1, 2, 6b-j, 7a-n, and 13a-l.
determined using the pNL4-3 wild-type strain while the other
compounds’ EC50 values were determined with the IIIB wild-
type strain, we observed a relationship between both wild types
(IIIB vs pNL4-3, r2=0.76, log linear equation), and the EC50
values used in deriving eq 1 for compounds 7m and 13k were
extrapolated from the correlation. The scoring criteria for
choosing the appropriate descriptor combination in eq 1 were
that the final equation must have an r2 value greater than 0.80
and the lowest deviation error (Figures 5). Deviation error was
calculated as the sum of squares of the floor values of the
absolute differences between calculated and measured -log-
(EC50) divided by 0.1. The final equation is well fitted (r2 =
0.85) and statistically significant (p < 0.01). A form of K-fold
cross-validation, namely, leave-one-out cross-validation, was
performed using a single observation from the original sample
as the validation data and the remaining observations as the
training data, and the process was repeated such that each
observation in the sample was used once as the validation data.
The equation is valid because the root-mean-square deviations
(rmsd) of the coefficients and intercept are relatively small, and
most importantly, the coefficient of determination did not
greatly vary during cross-validation. The narrow r2 range
(r2 =0.82-0.88) indicates the absence of outlier data in the
training set.
QSAR Equation for 1, 2, 6b-j, 7a-n, and 13a-l:
-logðEC50Þ ¼ 3:147logðInhenz
Þ
Figure 4. X-ray crystal structure of a complex of compound 13k
(green sticks) and dimeric HIV-1 protease (orange and light-blue
ribbons): (a) full structure; (b) hydrogen bonding interactions
(dotted lines); (c) superimposition of 1. Compound 1 and Asp300
from PDB 1MSM is represented in magenta sticks. (d) Residues
þ ð5:775 ꢀ 10-2 VSAp - 0:060 ꢀ 10-2 VSAp2Þ
3
3
þ ð3:070 ASAp - 0:290 ASAp2Þ
3
3
0
surrounding the β-methallyl group in S2 pocket. Figures were
generated using MacPyMOL (DeLano Scientific, LLC, CA).
-ð4:886 ꢀ 10-2 SMR -0:051 ꢀ 10-2 SMR2Þ
3
3
- 9:886
ð1Þ
in which HIV-1 PR inhibition at 50 nM and EC50 have been
determined, namely, compounds 1, 2, 6b-j, 7a-n, and 13a-l.
Although the EC50 values for inhibitors 7m and 13k were
n ¼ 37, r2 ¼ 0:85, F ¼ 23, p < 0:01

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7611
EC50 is the anti-HIV inhibitory activity (μM). Inhenz is the
percent HIV-1 protease inhibition at 50 nM of the test com-
pound, expressed as a number from 0 to 100. VSAp uses the
partial equalization of orbital electronegativity method23 for
calculating the van der Waals surface area of atoms with partial
charges in the range of þ0.05 to þ0.10, i.e., slight positively
charged compounds. ASAp is the fractional, positive partial
charge, weighted, water accessible surface area24 using a probe
Table 4. Inhibitory Activity against Wild-Type and Resistant HIV
Straina
EC50 (μM)
compd
wt (pNL4-3)
IND-Rb
FR
lopinavir
1
6d
0.016c
0.060
0.025
0.162
0.240
0.290
0.141
0.558
0.280
0.103
0.122
0.048
0.450
0.620
0.054
0.385c
0.100
0.157
0.155
1.73
24
1.7
6.3
1.0
7.2
7.6
1.2
3.5
7.3
5.5
3.4
8.7
3.2
4.9
8.7
6f
of 1.4 A. SMR uses the subdivided surface area method25 for
˚
7f
7g
calculating the van der Waals surface area of atoms with
contributions to molar refractivity in the range of 0.485-
0.560 and represents the bulkiness of the compound.
2.20
7k
7l
0.173
1.93
2.03
13a
13b
13e
13g
13h
13i
13j
Optimal values for the appropriate descriptor (VSAp,
ASAp, and SMR) can be determined because each descriptor
is described as a parabolic quadratic. Thus, an inhibitor with
potent enzyme inhibitory activity, a VSAp value of ∼48.41
(extrapolated), ASAp of ∼5.287 (extrapolated), and SMR
of ∼48.36 (intrapolated) is expected to exhibit “optimally”
potent antiviral activity. The contribution of each descriptor
to the equation can be estimated when each descriptor’s
quadratic is normalized: VSAp (40%), ASAp (35%), SMR
(13%), and Inhenz (12%). The low contribution of Inhenz
confirms our previous observation that enzyme affinity plays
a small yet important role in antiviral activity to a similar
extent as SMR. However, because Inhenz is the lowest con-
tributor in the equation, it may be possible to very roughly
predict the antiviral activity of a compound that is yet
synthesized, from the compound’s calculated VSAp, ASAp,
and SMR values while keeping in mind the aforementioned
calculated “optimal” values. The major contributors, VSAp
and ASAp, both suggest that compounds with proportionally
large areas of slightly positively charged surfaces in the range
of þ0.05 to þ0.10 are expected to exhibit potent antiviral
activity. Of interest, there is a correlation between the VSAp
and ASAp descriptors (r2 = 0.76) because both descriptors
only differ in their consideration of either van der Waals or
water accessible surface area. However, exclusion of either
one of VSAp or ASAp descriptor from the overall QSAR
equation used to predict anti-HIVIIIB would form equations
with low correlations (r2 < 0.43). Nonetheless, the contribu-
tion of the VSAp and ASAp descriptors is consistent with the
observed reduced anti-HIV activity in compound 6f posses-
0.568
0.409
0.416
1.42
3.01
0.470
a wt: wild-type. FR: fold resistance. b IND-R: L10R/M46I/L63P/
V82T/I84V. c From ref 31.
(r2=0.83 vs 0.85). Some reports suggested a correlation with
molar refractivity (i.e., SMR), thus re-enforcing our current
equation.27
Attenuation of inhibitors’ potential activity in vivo is an-
other hurdle for developing clinical candidates. Evaluation of
activity in the presence of 50% human serum has been
suggested as an alternative method to provide deeper insights
into the in vivo potency of the inhibitors.28 Serum binding
effect of compounds 13a-l ranged from 2 to 8 (Table 3).
Among the series, compound 13k exhibited the most potent
activity with human serum (EC50=82 nM, 8.2-fold attenu-
ation). These moderate reductions in activity in the presence
of human serum suggest promising potentials for future
clinical studies.
A few compounds were preliminary tested against indinavir-
resistant strain29 IND-R (Table 4). Among these compounds,
1 exhibited the most potent inhibitory activity against IND-R.
Moreover, we identified three compounds (6d, 6f, and 7k)
possessing relatively higher potency than lopinavir. Relative
difference in activity against the IND-R and wild-type (pNL4-
3) strains ranged from 1- to 9-fold. Apparent preferences for
0
sing a partially negatively charged P2 m-hydroxy moiety
0
specific P2 or P2 moieties against IND-R were not observed.
The reason behind the comparably high activities of 6d and 6f
might be derived from hydrogen bonding interaction at the
compared with compound 6g which possessed a m-fluorine
moiety. The same line of reasoning could explain that the
addition of the partially negatively charged hydroxymethyl to
0
P2 position. The activity of the tested compounds was lower
0
a P2 allyl moiety is detrimental to anti-HIV activity (cf. 7k vs
against lopinavir-resistant strain30 A17 (data not shown). The
ineffectiveness against A17 suggests dependencies on the
interacting protease residues. Further extensions on the in-
hibitor, such as a tripeptide-type compounds exhibiting spe-
cific interactions with the S3 pocket of the protease, are in
progress to develop compounds with high activity against
drug-resistant viral strains. Statistically, we observed a correla-
tion between both wild-type strains (IIIB vs pNL4-3, r2=0.76,
log linear equation). Low correlation was observed between
wild-type pNL4-3 and indinavir-resistant strain IND-R (r2=
0.55). These low correlations suggest that predicting anti-HIV
activity against resistant viral strains based on wild-type anti-
HIV activity remains quite a challenge.
7n, and 13j vs 13l). Indeed, compound 13k was accurately
predicted by the equation to exhibit the most potent antiviral
activity and 2 as the least potent. It should, however, be noted
that most of our compounds belong to a predefined subset of
potent enzyme inhibitors withsimilar chemical structures, and
the equation may not accurately predict the antiviral activity
of compounds that are vastly structurally different. In fact,
other research groups26,27 have observed a correlation be-
tween the octanol-water partition coefficient (log P) to pre-
dict lipophilicity and antiviral activity, which we did not
observe as one of the top-three calculated contributors to
the equation in our series of compounds. Although logP is in
fact found in the fifth top QSAR equation, it was omitted
from eq 1 to keep the equation more statistically reliable. It
should, however, be noted that most descriptors are some-
what related to each other because an equation formed
with log P instead of SMR would only be slightly less reliable
Conclusion
0
We investigated structure-activity relationships at the P2
position of allophenylnorstatine-containing HIV protease

7612 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Hidaka et al.
inhibitors. Modifications based on the benzyl structure opti-
0
2.83-2.64 (m, 2H), 2.38 (s, 3H), 1.81 (s, 3H), 1.46 (s, 3H), 1.33 (s,
3H). HRMS (FAB) m/z: calcd for C32H36ClN3O5SNa [Mþ Na]þ
632.1962; found 632.1968.
mized the substitution of the P2 o-methyl group in c0ompound
1. Introduction of non-benzyl moieties to the P2 position
led us to uncover the effectiveness of relatively small allyl
moieties for enzymatic inhibition. An additional methyl
group, more specifically a β-methallyl moiety, marked potent
inhibitory activity against both the HIV protease and virus.
Compound 13k, possessing a β-methallyl group, exhibited
potent anti-HIV activity and moderate water solubility be-
cause of a hydrophilic P2 moiety. These results provide a
rational drug design for size-minimizing a lead compound and
providing new insights on membrane permeability for drug
candidates against other enzymes or receptors. Further eva-
luations against resistant HIV strains of synthetic analogues
are underway in order to develop clinical candidates for HIV
therapy.
(R)-N-(2,6-Dichlorobenzyl)-3-[(2S,3S)-3-(3-hydroxy-2-meth-
ylbenzoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-
thiazolidine-4-carboxamide (6c). A mixture of 5c (95.5 mg, 0.16
mmol), anisole (34.6 μL, 0.32 mmol), and 4 M HCl in dioxane
(1.0 mL) was stirred for 30 min at room temperature. After
removal of the solvent in vacuo, the residue was precipitated
from ether to give the hydrochloride salt. To a solution of
the salt in DMF (2 mL) were added triethylamine (55.9 μL,
0.4 mmol), 3-acetoxy-2-methylbenzoic acid (34.2 mg, 0.18
mmol), and BOP (77.8 mg, 0.18 mmol) in an ice bath, and the
mixture was stirred overnight at room temperature. After
removal of the solvent in vacuo, the residue was dissolved in
EtOAc, washed sequentially with 10% citric acid, 5% NaHCO3,
and brine, dried over MgSO4, and concentrated in vacuo. To a
solution of the residue in methanol (2 mL) was added 1 N
aqueous NaOH (0.32 mL), and the mixture was stirred 1 h at
room temperature. After removal of the methanol in vacuo, the
residue was acidified with citric acid, extracted with EtOAc,
washed with 5% NaHCO3 and then brine, dried over MgSO4,
and concentrated in vacuo. Purification of the product by
preparative TLC gave 59.8 mg of the titled compound as a
white solid (yield 59%). Additional purification of the product
by preparative HPLC gave compound 6c as a white powder. 1H
NMR (300 MHz, DMSO-d6) δ (ppm): 9.34 (s, 1H), 8.30 (t, J=
4.6 Hz, 1H), 8.20 (d, J=8.4 Hz, 1H), 7.51-7.14 (m, 8H), 6.95 (t,
J=7.8 Hz, 1H), 6.77 (d, J=7.2 Hz, 1H), 6.55 (d, J=7.2 Hz, 1H),
5.31 (d, J=6.9 Hz, 1H), 5.13 (d, J=8.7 Hz, 1H), 5.01 (d, J=
8.7 Hz, 1H), 4.64-4.64 (m, 4H), 4.43-4.32 (m, 1H), 2.84-2.64
(m, 2H), 1.81 (s, 3H), 1.46 (s, 3H), 1.34 (s, 3H). HRMS (FAB)
m/z: calcd for C31H33Cl2N3O5SNa [M þ Na]þ 652.1416; found
652.1412.
(R)-N-(2-Chloro-6-fluorobenzyl)-3-[(2S,3S)-3-(3-hydroxy-2-
methylbenzoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-
1,3-thiazolidine-4-carboxamide (6d). Compound 6d was prepared
from compound 5d in a manner similar to that described for
compound 6c. Yield 92%. 1H NMR (300 MHz, DMSO-d6)
δ (ppm): 9.35 (br, 1H), 8.38 (t, J=4.7 Hz, 1H), 8.19 (d, J=8.4
Hz, 1H), 7.42-7.14 (m, 8H), 6.95 (t, J=7.8 Hz, 1H), 6.77 (d, J=
8.1 Hz, 1H), 6.55 (d, J=7.3 Hz, 1H), 5.12 (d, J=9.2 Hz, 1H), 5.00
(d, J=9.2 Hz, 1H), 4.54-4.46 (s, 3H), 4.42-4.29 (m, 2H), 2.83-
2.64 (m, 2H), 1.81 (s, 3H), 1.45 (s, 3H), 1.28 (s, 3H). HRMS
(FAB) m/z: calcd for C31H33ClFN3O5SNa [M þ Na]þ 636.1711;
found 636.1716.
(R)-N-(2,6-Difluorobenzyl)-3-[(2S,3S)-3-(3-hydroxy-2-meth-
ylbenzoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-
thiazolidine-4-carboxamide (6e). Compound 6e was prepared
from compound 5e in a manner similar to that described for
compound 6c. Yield 92%. 1H NMR (300 MHz, DMSO-d6)
δ (ppm): 9.36 (br, 1H), 8.44 (t, J=5.1 Hz, 1H), 8.16 (d, J=8.6 Hz,
1H), 7.38-7.16 (m, 6H), 7.07 (t, J=7.8 Hz, 2H), 6.95 (t, J=
7.7 Hz, 1H), 6.77 (d, J=8.1 Hz, 1H), 6.54 (d, J=7.4 Hz, 1H), 5.11
(d, J=9.0 Hz, 1H), 4.99 (d, J=9.2 Hz, 1H), 4.47 (d, J=4.4 Hz,
1H), 4.45 (s, 1H), 4.43-4.32 (m, 2H), 4.25 (dd, J=13.9 Hz, 4.6
Hz, 1H), 2.84-2.64 (m, 2H), 1.81 (s, 3H), 1.44 (s, 3H), 1.24 (s,
3H). HRMS (FAB) m/z: calcd for C31H33F2N3O5SNa [M þ
Na]þ 620.2007; found 620.2001.
Experimental Section
Reagents and solvents used were purchased from Wako Pure
Chemical Ind., Ltd. (Osaka, Japan), Nacalai Tesque (Kyoto,
Japan), Aldrich Chemical Co. Inc. (Milwaukee, WI), and Tokyo
Kasei Kogyo Co., Ltd. (Tokyo, Japan) and were used without
further purification. TLC was performed using Merck silica gel
60 F254 precoated plates. Column chromatography was per-
formed on Merck 107734 silica gel 60 (70-230 mesh). Melting
points were measured on a Yanagimoto micromelting apparatus
without corrections. The purities of the desired compounds were
confirmed by HPLC or elemental analyses as greater than 95%.
Analytical HPLC was performed using C18 reversed-phase
column (4.6 mm ꢀ 150 mm, YMC Pack ODS AM302) with
binary solvent systems: (A) linear gradient of CH3CN 40-100%
in 0.1% aqueous TFA in 15 min, (B) linear gradient of CH3CN
20-80% in0.1% aqueous TFA in30 min at a flow rateof0.9 mL/
min, detected at 230 nm. Preparative HPLC was carried out on a
C18 reversed-phase column (20 mm ꢀ 250 mm, YMC Pack ODS
SH343-5) with a binary solvent system: a linear gradient of
CH3CN in 0.1% aqueous TFA with a flow rate of 5.0 mL/min
and detection at 230 nm. 1H NMR spectra were obtained on a
JEOL AL300 (300 MHz) spectrometer with TMS as an internal
standard. Mass spectra (electrospray ionization, methanol as
the mobile phase) were obtained from a Finnigan SSQ 7000
spectrometer. FAB-MS was performed on a JEOL JMS-SX102A
spectrometer equipped with the JMA-DA7000 data system.
MALDI-TOF MS was performed on a Voyager-DE RP spectro-
meter (PerSeptive Biosystems, Inc.).
(R)-N-(2,6-Dimethylbenzyl)-3-[(2S,3S)-3-(3-hydroxy-2-meth-
ylbenzoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-
thiazolidine-4-carboxamide (6a). Compound 6a was prepared
from compound 5j in a manner similar to that described for
1
compound 6c. Yield 91%. H NMR (300 MHz, DMSO-d6) δ
(ppm): 9.35 (s, 1H), 8.22 (d, J=8.4 Hz, 1H), 8.08 (br, 1H), 7.38
(d, J=7.0 Hz, 1H), 7.29-7.14 (m, 3H), 7.11-6.92 (m, 4H), 6.78
(d, J=7.5 Hz, 1H), 6.56 (d, J=6.8 Hz, 1H), 5.18 (d, J=9.0 Hz,
1H), 4.99 (d, J=9.0 Hz, 1H), 4.54 (s, 1H), 4.52 (d, J=3.7 Hz,
1H), 4.46 (dd, J=13.9 Hz, 6.1 Hz, 1H), 4.41-4.32 (m, 1H), 4.18
(dd, J=13.9 Hz, 3.3 Hz, 1H), 2.85-2.65 (m, 2H), 2.31 (s, 6H),
1.81 (s, 3H), 1.46 (s, 3H), 1.35 (s, 3H). HRMS (FAB) m/z: calcd
for C33H39N3O5SNa [M þ Na]þ 612.2508; found 612.2502.
(R)-N-(2-Chloro-6-methylbenzyl)-3-[(2S,3S)-3-(3-hdroxy-2-
methylbenzoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-
1,3-thiazolidine-4-carboxamide (6b). Compound 6b was prepared
from compound 5b in a manner similar to that described for
compound 6c. Yield 47%. 1H NMR (300 MHz, DMSO-d6)
δ (ppm): 9.36 (br, 1H), 8.20-8.23 (m, 2H), 7.37 (d, J=7.3 Hz,
1H), 7.31-7.14 (m, 6H), 6.95 (t, J=7.8 Hz, 1H), 6.78 (d, J=8.0
Hz, 1H), 6.55 (d, J=7.3 Hz, 1H), 5.15 (d, J=9.0 Hz, 1H), 5.49 (d,
J=9.0 Hz, 1H), 4.55 (s, 1H), 4.51 (d, J=3.3 Hz, 1H), 4.46 (d, J=
5.0 Hz, 1H), 4.41 (d, J=4.1 Hz, 1H), 4.37 (d, J=3.5 Hz, 1H),
(R)-N-(3-Hydroxy-2-methylbenzyl)-3-[(2S,3S)-3-(3-hydroxy-
2-methylbenzoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-
1,3-thiazolidine-4-carboxamide (6f). Compound 6f was prepared
from compound 5f in a manner similar to that described for
compound 6c. Yield 71%. 1H NMR (300 MHz, DMSO-d6)
δ (ppm): 9.37 (s, 1H), 9.19 (s, 1H), 8.24 (t, J=5.2 Hz, 1H), 8.14
(d, J=8.4 Hz, 1H), 7.34 (d, J=7.1 Hz, 12H), 7.23-7.13 (m, 3H),
6.97-6.87 (m, 2H), 6.77 (dd, J=7.2 Hz, 4.7 Hz, 1H), 6.69 (d, J=7.7
Hz, 1H), 6.55 (d, J=7.3 Hz, 1H), 5.13 (d, J=9.0 Hz, 1H), 5.01 (d,
J= 9.0 Hz, 1H), 4.50 (s, 1H), 4.48 (d, J= 4.6 Hz, 1H), 4.44-
4.31 (m, 2H), 4.07 (dd, J=14.9 Hz, 4.8 Hz, 1H), 2.89-2.67 (m, 2H),

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7613
2.06 (s, 3H), 1.83 (s, 3H), 1.49 (s, 3H), 1.24 (s, 3H). HRMS (FAB)
m/z: calcd for C32H37N3O6SNa [M þ Na]þ 614.2301; found
614.2297.
was prepared from compound 5m in a manner similar to that
described for compound 6c. Yield 62%. 1H NMR (300 MHz,
DMSO-d6) δ (ppm): 9.35 (s, 1H), 8.21 (d, J=8.2 Hz, 1H), 8.16
(d, J=7.9 Hz, 1H), 7.34-7.12 (m, 9H), 6.95 (t, J=7.7 Hz, 1H),
6.67 (d, J=8.0 Hz, 1H), 6.56 (d, J=7.5 Hz, 1H), 5.52 (d, J=
6.2 Hz, 1H), 5.26-5.16 (m, 2H), 5.06 (d, J=9.4 Hz, 1H), 4.64
(s, 1H), 4.60 (d, J=3.1 Hz, 1H), 4.53-4.32 (m, 3H), 3.05 (dd,
J = 16.6 Hz, 5.0 Hz, 1H), 2.88-2.64 (m, 3H), 1.80 (s, 3H),
1.54 (s, 3H), 1.49 (s, 3H). HRMS (FAB) m/z: calcd for
(R)-N-(3-Fluoro-2-methylbenzyl)-3-[(2S,3S)-3-(3-hydroxy-2-
methylbenzoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimeth-
yl-1,3-thiazolidine-4-carboxamide (6g). Compound 6g was pre-
pared from compound 5g in a manner similar to that described
for compound 6c. Yield 69%. 1H NMR (300 MHz, DMSO-d6)
δ (ppm): 9.37 (s, 1H), 8.39 (t, J=5.6 Hz, 1H), 8.14 (d, J=8.3 Hz,
1H), 7.34-6.91 (m, 9H), 6.78 (d, J=7.3 Hz, 1H), 6.55 (d, J=
6.6 Hz, 1H), 5.46 (br, 1H), 5.14 (d, J=9.2 Hz, 1H), 5.01 (d, J=
9.2 Hz, 1H), 4.51-4.36 (m, 4H), 4.14 (dd, J=15.1 Hz, 4.8 Hz,
1H), 2.88-2.66 (m, 2H), 2.17 (s, 3H), 1.82 (s, 3H), 1.50 (s, 3H),
1.34 (s, 3H). HRMS (FAB) m/z: calcd for C32H36FN3O5SNa
[M þ Na]þ 616.2257; found 616.2252.
C
33H37N3O6SNa [M þ Na]þ 626.2301; found 626.2294.
(R)-N-[(1S,2R)-2-Hydroxyindan-1-yl]-3-[(2S,3S)-3-(3-hdroxy-
2-methylbenzoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimeth-
yl-1,3-thiazolidine-4-carboxamide (7d). Compound 7d was pre-
pared from compound 5n in a manner similar to that described
for compound 6c. Yield 70%. 1H NMR (300 MHz, DMSO-d6)
δ (ppm): 9.38 (s, 1H), 8.28 (d, J=8.4 Hz, 1H), 8.10 (d, J=8.7 Hz,
1H), 7.42-7.05 (m, 9H), 6.95 (t, J=7.7 Hz, 1H), 6.78 (d, J=7.8
Hz, 1H), 6.57 (d, J=7.2 Hz, 1H), 5.30 (d, J=4.8 Hz, 1H), 5.27 (d,
J=4.8 Hz, 1H), 5.17 (d, J=9.6 Hz, 1H), 5.03 (d, J=9.0 Hz, 1H),
4.75 (s, 1H), 4.57 (d, J=3.3 Hz, 1H), 4.45-4.32 (m, 2H), 3.05 (dd,
J=16.1 Hz, 5.3 Hz, 1H), 2.91-2.68 (m, 3H), 1.82 (s, 3H), 1.56 (s,
3H), 1.49 (s, 3H). HRMS (FAB) m/z: calcd for C33H37N3O6SNa
[M þ Na]þ 626.2301; found 626.2305.
(R)-N-[(S)-1-Phenylethane-1-yl]-3-[(2S,3S)-3-(3-hydroxy-2-
methylbenzoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-
1,3-thiazolidine-4-carboxamide (6h). Compound 6h was prepared
from compound 5h in a manner similar to that described for
compound 6c. Yield 54%. 1H NMR (300 MHz, DMSO-d6)
δ (ppm): 9.36 (s, 1H), 8.45 (d, J = 7.9 Hz, 1H), 8.19 (d, J =
8.4 Hz, 1H), 7.37-7.16 (m, 10H), 6.95 (t, J=7.8 Hz, 1H), 6.78 (d,
J=7.7 Hz, 1H), 6.55 (d, J=7.2 Hz, 1H), 5.36 (d, J=6.8 Hz, 1H),
5.10 (d, J=9.2 Hz, 1H), 5.01 (d, J=9.2 Hz, 1H), 5.00-4.94 (m,
1H), 4.55-4.52 (m, 1H), 4.50 (s, 1H), 4.46-4.37 (m, 1H), 2.87-
2.66 (m, 2H), 1.81 (s, 3H), 1.49 (s, 3H), 1.39 (d, J=7.0 Hz, 1H),
1.22 (s, 3H). HRMS (FAB) m/z: calcd for C32H37N3O5SNa [M þ
Na]þ 598.2352; found 598.2359.
(R)-N-Cycloheptyl-3-[(2S,3S)-3-(3-hydroxy-2-methylbenzoyl)-
amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-thiazoli-
dine-4-carboxamide (7e). Compound 7e was prepared from
compound 5o in a manner similar to that described for
1
compound 6c. Yield 59%. H NMR (300 MHz, DMSO-d6)
(R)-N-[(R)-1-Phenylethane-1-yl]-3-[(2S,3S)-3-(3-hydroxy-2-
methylbenzoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimeth-
yl-1,3-thiazolidine-4-carboxamide (6j). Compound 6j was pre-
pared from compound 5i in a manner similar to that described
for compound 6c. Yield 83%. 1H NMR (300 MHz, DMSO-d6)
δ (ppm): 9.35 (s, 1H), 8.37 (d, J=7.9 Hz, 1H), 8.15 (d, J=8.4 Hz,
1H), 7.38 (d, J=7.3 Hz, 2H), 7.29-7.12 (m, 8H), 6.94 (t, J=
7.8 Hz, 1H), 6.77 (d, J=7.9 Hz, 1H), 6.54 (d, J=7.3 Hz, 1H),
5.39 (d, J=7.0 Hz, 1H), 5.17 (d, J=8.8 Hz, 1H), 5.04-4.94 (m,
2H), 4.54 (s, 1H), 4.49 (dd, J=6.6 Hz, 4.2 Hz, 1H), 4.43-4.33
(m, 1H), 2.81-2.63 (m, 2H), 1.81 (s, 3H), 1.52 (s, 3H), 1.40 (s,
3H), 1.36 (d, J = 7.0 Hz, 1H). HRMS (FAB) m/z: calcd for
C32H37N3O5SNa [M þ Na]þ 598.2352; found 598.2358.
(R)-N-[(S)-Indan-1-yl]-3-[(2S,3S)-3-(3-hydroxy-2-methylben-
zoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-thia-
zolidine-4-carboxamide (7a). Compound 7a was prepared from
compound 5k in a manner similar to that described for com-
pound 6c. Yield 52%. 1H NMR (300 MHz, DMSO-d6) δ (ppm):
9.38 (s, 1H), 8.35 (d, J=7.8 Hz, 1H), 8.22 (d, J=8.4 Hz, 1H),
7.35-7.12 (m, 9H), 6.95 (t, J=7.8 Hz, 1H), 6.78 (d, J=7.8 Hz,
1H), 6.57 (d, J=7.2 Hz, 1H), 5.36-5.26 (m, 2H), 5.15 (d, J=
9.0 Hz, 1H), 5.04 (d, J=9.3 Hz, 1H), 4.52 (d, J=3.9 Hz, 1H),
4.48 (s, 1H), 4.45-4.35 (m, 1H), 3.02-2.68 (m, 4H), 2.44-2.32
(m, 1H, overlapped with DMSO), 1.81 (s, 3H), 1.92-1.77 (m,
1H), 1.50 (s, 3H), 1.45 (s, 3H). HRMS (FAB) m/z: calcd for
C33H37N3O5SNa [M þ Na]þ 610.2352; found 610.2357.
(R)-N-[(R)-Indan-1-yl]-3-[(2S,3S)-3-(3-hydroxy-2-methylben-
zoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-thia-
zolidine-4-carboxamide (7b). Compound 7b was prepared from
compound 5l in a manner similar to that described for com-
pound 6c. Yield 54%. 1H NMR (300 MHz, DMSO-d6) δ (ppm):
9.37 (s, 1H), 8.38 (d, J=8.7 Hz, 1H), 8.18 (d, J=8.7 Hz, 1H),
7.41-7.07 (m, 9H), 6.94 (t, J=7.7 Hz, 1H), 6.78 (d, J=7.5 Hz,
1H), 6.53 (d, J=7.5 Hz, 1H), 5.39 (d, J=8.1 Hz, 1H), 5.33 (d, J=
8.4 Hz, 1H), 4.47 (s, 1H), 4.44-3.38 (m, 1H), 2.94-2.70 (m,
4H), 2.41-2.26 (m, 1H, overlapped with DMSO), 1.82 (s, 3H),
1.87-1.68 (m, 1H), 1.52 (s, 3H), 1.44 (s, 3H). HRMS (FAB)
m/z: calcd for C33H37N3O5SNa [M þ Na]þ 610.2352; found
610.2358.
δ (ppm): 9.35 (s, 1H), 8.16 (d, J=8.3 Hz, 1H), 7.85 (d, J=
7.7 Hz, 1H), 7.35 (d, J=7.0 Hz, 2H), 7.24 (t, J=7.2 Hz, 2H),
7.17 (d, J=7.0 Hz, 1H), 6.94 (t, J=7.7 Hz, 1H), 6.77 (d, J=
7.9 Hz, 1H), 6.54 (d, J=7.5 Hz, 1H), 5.32 (br, 1H), 5.11 (d, J=
9.0 Hz, 1H), 5.00 (d, J=9.0 Hz, 1H), 4.50 (d, J=3.8 Hz, 1H),
4.44 (s, 1H), 4.42-4.35 (m, 1H), 3.81-3.69 (m, 1H), 2.83-2.65
(m, 2H), 1.81 (s, 3H), 1.79-1.70 (m, 2H), 1.61-1.39 (m, 13H),
1.37 (s, 3H). HRMS (FAB) m/z: calcd for C31H41N3O5SNa
[M þ Na]þ 590.2665; found 590.2671.
(R)-N-Cyclohexyl-3-[(2S,3S)-3-(3-hydroxy-2-methylbenzoyl)-
amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-thiazoli-
dine-4-carboxamide (7f). Compound 7f was prepared from
compound 5p in a manner similar to that described for
1
compound 6c. Yield 32%. H NMR (300 MHz, DMSO-d6)
δ (ppm): 9.36 (s, 1H), 7.80 (d, J=7.9 Hz, 1H), 7.34 (d, J=7.1
Hz, 1H), 7.37-7.13 (m, 5H), 6.94 (t, J=7.8 Hz, 1H), 6.73 (d, J=
7.3 Hz, 1H), 6.54 (d, J=7.1 Hz, 1H), 5.33 (d, J=7.1 Hz, 1H),
5.10 (d, J=9.0 Hz, 1H), 5.01 (d, J=9.2 Hz, 1H), 4.51 (dd, J=
7.0 Hz, 4.2 Hz, 1H), 4.43 (s, 1H), 4.42-4.35 (m, 1H), 3.63-3.44
(m, 1H), 2.84-2.65 (m, 2H), 1.81 (s, 3H), 1.77-1.59 (m, 4H),
1.50 (s, 3H), 1.37 (s, 3H), 1.29-1.06 (m, 6H). HRMS (FAB)
m/z: calcd for C30H39N3O5SNa [Mþ Na]þ 576.2508; found
576.2503.
(R)-N-Cyclopentyl-3-[(2S,3S)-3-(3-hydroxy-2-methylbenzoyl)-
amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-thiazoli-
dine-4-carboxamide (7g). Compound 7g was prepared from
compound 5q in a manner similar to that described for
1
compound 6c. Yield 83%. H NMR (300 MHz, DMSO-d6)
δ (ppm): 9.34 (s, 1H), 8.15 (d, J=8.7 Hz, 1H), 7.88 (d, J=
7.2 Hz, 1H), 7.37-7.15 (m, 5H), 6.93 (t, J=7.5 Hz, 1H), 6.77 (d,
J=7.8 Hz, 1H), 6.54 (d, J=7.5 Hz, 1H), 5.32 (d, J=6.6 Hz,
1H), 5.11 (d, J=9.3 Hz, 1H), 5.01 (d, J=9.3 Hz, 1H), 4.50 (dd,
J=6.8 Hz, 4.4 Hz, 1H), 4.46-4.34 (m, 2H), 4.06-3.96 (m, 1H),
2.84-2.65 (m, 2H), 1.87-1.32 (m, 17H). HRMS (FAB) m/z:
calcd for C29H37N3O5SNa [M þ Na]þ 562.2352; found
562.2357.
(R)-N-Cyclobutyl-3-[(2S,3S)-3-(3-hydroxy-2-methylbenzoyl)-
amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-thiazoli-
dine-4-carboxamide (7h). Compound 7h was prepared from
compound 5r in a manner similar to that described for
compound 6c. Yield 22%. 1H NMR (300 MHz, DMSO-d6) δ
(ppm): 9.36 (s, 1H), 8.18-8.14 (m, 2H), 7.34 (d, J=7.1 Hz, 2H),
(R)-N-[(1R,2S)-2-Hydroxyindan-1-yl]-3-[(2S,3S)-3-(3-hydro-
xy-2-methylbenzoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-
dimethyl-1,3-thiazolidine-4-carboxamide (7c). Compound 7c

7614 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Hidaka et al.
7.25 (t, J=7.1 Hz, 2H), 7.18 (d, J=7.1 Hz, 1H), 7.34 (d, J=
7.1 Hz, 2H), 6.94 (t, J=7.8 Hz, 1H), 6.77 (d, J=7.9 Hz, 1H),
6.54 (d, J=7.5 Hz, 1H), 5.38 (d, J=6.8 Hz, 1H), 5.11 (d, J=
9.0 Hz, 1H), 5.01 (d, J=9.2 Hz, 1H), 4.51-4.34 (m, 3H), 4.26-
4.11 (m, 1H), 2.84-2.65 (m, 2H), 1.99-1.89 (m, 2H), 1.81 (s,
3H), 1.68-1.57 (m, 2H), 1.50 (s, 3H), 1.35 (s, 3H). HRMS
(FAB) m/z: calcd for C28H35N3O5SNa [M þ Na]þ 548.2195;
found 548.2191.
2.88-2.67 (m, 2H), 1.82 (s, 3H), 1.65 (s, 3H), 1.51 (s, 3H), 1.36
(s, 3H). HRMS (FAB) m/z: calcd for C28H35N3O5SNa [M þ
Na]þ 548.2195; found 548.2189.
(R)-N-(cis-4-Hydroxy-2-buten-1-yl)-3-[(2S,3S)-3-(3-hydroxy-
2-methylbenzoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-
1,3-thiazolidine-4-carboxamide (7n). Compound 7n was prepared
from compound 5w in a manner similar to that described for
1
compound 6c. Yield 25%. H NMR (300 MHz, DMSO-d6) δ
(R)-N-Ethyl-3-[(2S,3S)-3-(3-hydroxy-2-methylbenzoyl)amino-
2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-thiazolidine-4-
carboxamide (7i). Compound 7i was prepared from com-
pound 5s in a manner similar to that described for compound
6c. Yield 37%. 1H NMR (300 MHz, DMSO-d6) δ (ppm): 9.36
(s, 1H), 8.11 (d, J=8.2 Hz, 1H), 7.91 (t, J=5.5 Hz, 1H), 7.35-
7.14 (m, 5H), 6.94 (t, J=7.8 Hz, 1H), 6.77 (d, J=7.5 Hz, 1H),
6.53 (d, J=7.3 Hz, 1H), 5.42 (d, J=6.4 Hz, 1H), 5.10 (d, J=
9.2 Hz, 1H), 5.01 (d, J=9.2 Hz, 1H), 4.49-4.38 (m, 2H), 4.36
(s, 1H), 3.13-3.04 (m, 2H), 2.87-2.67 (m, 2H), 1.81 (s, 3H),
1.50 (s, 3H), 1.36 (s, 3H), 1.00 (t, J = 7.1 Hz, 3H). HRMS
(FAB) m/z: calcd for C26H33N3O5SNa [M þ Na]þ 522.2039;
found 522.2043.
(R)-N-Propyl-3-[(2S,3S)-3-(3-hydroxy-2-methylbenzoyl)amino-
2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-thiazolidine-4-
carboxamide (7j). Compound 7j was prepared from com-
pound 5t in a manner similar to that described for compound
6c. Yield 48%. 1H NMR (300 MHz, DMSO-d6) δ (ppm): 9.36
(s, 1H), 8.12 (d, J=8.2 Hz, 1H), 7.90 (t, J=5.7 Hz, 1H), 7.35-
7.14 (m, 5H), 6.94 (t, J=7.7 Hz, 1H), 6.77 (d, J=8.1 Hz, 1H),
6.54 (d, J=7.4 Hz, 1H), 5.41 (d, J=6.8 Hz, 1H), 5.10 (d, J=
9.2 Hz, 1H), 5.01 (d, J=9.2 Hz, 1H), 4.50-4.39 (m, 2H), 4.38
(s, 1H), 3.10-2.95 (m, 2H), 2.87-2.66 (m, 2H), 1.82 (s, 3H),
1.50 (s, 3H), 1.47-1.31 (s, 5H), 0.84 (t, J=7.3 Hz, 3H). HRMS
(FAB) m/z: calcd for C27H35N3O5SNa [M þ Na]þ 536.2195;
found 536.2202.
(R)-N-Allyl-3-[(2S,3S)-3-(3-hydroxy-2-methylbenzoyl)amino-
2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-thiazolidine-4-
carboxamide (7k). Compound 7k was prepared from com-
pound 5a in a manner similar to that described for compound
6c. Yield 51%. 1H NMR (300 MHz, DMSO-d6) δ (ppm): 9.39
(s, 1H), 8.16-8.02 (m, 2H), 7.36-7.13 (m, 8H), 6.95 (t, J=
7.8 Hz, 1H), 6.77 (d, J=8.1 Hz, 1H), 6.54 (d, J=7.5 Hz, 1H),
5.84-5.70 (m, 1H), 5.21 (dd, J=17.1 Hz, 1.5 Hz, 1H), 5.12 (d,
J=9.0 Hz, 1H), 5.02 (dd, J=10.2 Hz, 1.5 Hz, 1H), 5.01 (d, J=
9.0 Hz, 1H), 4.48-4.35 (m, 3H), 3.84-3.56 (m, 2H), 2.88-
2.66 (m, 2H), 1.81 (s, 3H), 1.50 (s, 3H), 1.36 (s, 3H). HRMS
(FAB) m/z: calcd for C27H33N3O5SN [M þ Na]þ 534.2039;
found 534.2043.
(R)-N-Propargyl-3-[(2S,3S)-3-(3-hydroxy-2-methylbenzoyl)-
amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-thiazoli-
dine-4-carboxamide (7l). Compound 7l was prepared from
compound 5u in a manner similar to that described for
compound 6c. Yield 38%. 1H NMR (300 MHz, DMSO-d6) δ
(ppm): 9.37 (s, 1H), 8.40 (t, J=5.4 Hz, 1H), 8.11 (d, J=8.0 Hz,
1H), 7.33 (d, J=7.2 Hz, 2H), 7.26 (t, J=7.2 Hz, 2H), 7.21-7.14
(m, 1H), 6.95 (t, J=7.8 Hz, 1H), 6.77 (d, J=8.1 Hz, 1H), 6.54
(d, J=7.1 Hz, 1H), 5.49 (d, J=6.4 Hz, 1H), 5.11 (d, J=9.2 Hz,
1H), 5.01 (d, J=9.2 Hz, 1H), 4.47-4.40 (m, 2H), 4.39 (s, 1H),
3.87-3.82 (m, 2H), 3.09 (t, J=2.2 Hz, 1H), 2.90-2.66 (m, 2H),
1.81 (s, 3H), 1.50 (s, 3H), 1.36 (s, 3H). HRMS (FAB) m/z: calcd
for C27H31N3O5SNa [M þ Na]þ 532.1882; found 532.1878.
(R)-N-(2-Methylallyl)-3-[(2S,3S)-3-(3-hydroxy-2-methylben-
zoyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-thia-
zolidine-4-carboxamide (7m). Compound 7m was prepared
from compound 5v in a manner similar to that described for
compound 6c. Yield 70%. 1H NMR (300 MHz, DMSO-d6)
δ (ppm): 9.36 (s, H), 8.11-8.03 (m, 2H), 7.36-7.13 (m, 5H),
6.95 (t, J=7.8 Hz, 1H), 6.78 (d, J=7.1 Hz, 1H), 6.54 (d, J=
7.5 Hz, 1H), 5.11 (d, J=9.2 Hz, 1H), 5.00 (d, J=9.2 Hz, 1H),
4.88 (s, 1H), 4.73 (s, 1H), 4.49-4.36 (m, 4H), 3.74 (dd, J=15.7
Hz, 6.1 Hz, 1H), 3.52-3.47 (m, 1H, overlapped with H2O),
(ppm):9.37(s, 1H),8.15-8.05 (m, 2H), 7.34 (d, J=7.1 Hz, 2H), 7.25
(t, J=7.3 Hz, 2H), 7.20-7.13 (m, 1H), 6.95 (t, J=7.3 Hz, 1H), 6.78
(d, J=8.0 Hz, 1H), 6.54 (d, J=7.3 Hz, 1H), 5.58-5.48 (m, 1H), 5.45
(d, J=6.1 Hz, 1H), 5.36-5.26 (m, 1H), 5.11 (d, J=9.0 Hz, 1H), 5.01
(d, J=8.8 Hz, 1H), 4.69-4.67 (m, 1H), 4.49-4.40 (m, 2H), 4.38 (s,
1H), 4.06-3.99 (m, 2H), 3.75-3.68 (m, 2H), 2.90-2.66 (m, 2H),
1.82 (s, 3H), 1.49 (s, 3H), 1.35 (s, 3H). HRMS (FAB) m/z: calcd for
C28H35N3O6SNa [M þ Na]þ 564.2144; found 564.2151.
(R)-N-(tert-Butyl)-3-[(2S,3S)-3-(4-amino-2,6-dimethyphenoxy-
acetyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-
1,3-thiazolidine-4-carboxamide (13a). Compound 13a was prepared
from Boc-Apns-Dmt-NHtBu10 in a manner similar to that described
for compound 13b. Yield 68%. Mp 122-124 ꢀC. 1H NMR
(300 MHz, DMSO-d6) δ (ppm): 8.13 (d, J= 8.8 Hz, 1H), 7.67
(s, 1H), 7.35 (d, J=6.8 Hz, 2H), 7.27-7.13 (m, 5H), 6.82 (s, 2H),
4.98 (d, J=8.4 Hz, 1H), 4.92 (d, J=8.8 Hz, 1H), 4.53 (s, 2H), 4.48
(d, J=3.3 Hz, 1H), 4.37-4.27 (m, 1H), 4.19 (d, J=14.1 Hz, 1H),
3.99 (d, J=13.9 Hz, 1H), 2.81-2.71 (m, 2H), 2.14 (s, 6H), 1.49
(s, 3H), 1.40 (s, 3H), 1.26 (s, 9H). HRMS (FAB) m/z: calcd for
C30H43N4O5S [M þ H]þ 571.2944; found 571.2960.
(R)-N-(2-Methylbenzyl)-3-[(2S,3S)-3-(4-amino-2,6-dimethyphe-
noxyacetyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-
thiazolidine-4-carboxamide (13b). A mixture of Boc-Apns-Dmt-
NH(o-methylbenzyl)10 (54.2 mg, 0.1 mmol), anisole (21.6 μL,
0.2 mmol), and 4 N HCl in dioxane (1.0 mL) was stirred for
30 min at room temperature. After removal of the solvent in
vacuo, the residue was precipitated from ether to give the hydro-
chloride salt. To a solution of the HCl salt in DMF (2 mL) were
added triethylamine (34.9 μL, 0.25 mmol), 4-(Boc-amino)-2,6-
dimethylphenoxyacetic acid (10) (32.5 mg, 0.11 mmol), and BOP
(48.7 mg, 0.11 mmol) in an ice bath, and the mixture was stirred
overnight at room temperature. After removal of the solvent in
vacuo, the residue was added to EtOAc, washed sequentially with
10% citric acid, 5% NaHCO3, and saturated NaCl, dried over
MgSO4, and concentrated in vacuo. The residue was mixed with
anisole (21.6 μL) and 4 N HCl in dioxane (1.0 mL), and then the
mixture was stirred for 1 h at room temperature. After removal of
the solvent in vacuo, the residue was precipitated from ether to give
a product as hydrochloride salt, 46 mg. Purification of the product
by preparative HPLC gave compound 13b as a white powder.
Yield 70%. 1H NMR (300 MHz, DMSO-d6) δ (ppm): 8.38 (t, J=
5.8 Hz, 1H), 8.14 (d, J=8.8 Hz, 1H), 7.33-7.06 (m, 11H), 6.81 (s,
2H), 4.96 (s, 2H), 4.51 (s, 1H), 4.47 (d, J=3.7 Hz, 1H), 4.43-4.34
(m, 2H), 4.22-4.13 (m, 2H), 4.00 (d, J=14.1 Hz, 1H), 2.82-2.71
(m, 2H), 2.26 (s, 3H), 2.14 (s, 6H), 1.51 (s, 3H), 1.36 (s, 3H).
HRMS (FAB) m/z: calcd for C34H42N4O5SNa [M þ Na]þ
641.2774; found 641.2777.
(R)-N-(2-Methylbenzyl)-3-[(2S,3S)-3-(4-methylamino-2,6-di-
methyphenoxyacetyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-
dimethyl-1,3-thiazolidine-4-carboxamide (13c). Compound 13c
was prepared from Boc-Apns-Dmt-NH(o-methylbenzyl) and 11
in a manner similar to that described for compound 13b. Yield
68%. 1H NMR (300 MHz, DMSO-d6) δ (ppm): 8.37 (br, 1H),
8.10 (d, J=9.3 Hz, 1H), 7.34-7.05 (m, 11H), 6.58 (s, 1H), 4.96
(d, J=9.0 Hz, 2H), 4.51 (s, 1H), 4.47 (d, J=3.7 Hz, 1H), 4.44-
4.31 (m, 2H), 4.24-4.03 (m, 2H), 3.95 (d, J=14.1 Hz, 1H), 2.82-
2.67 (m, 5H), 2.26 (s, 3H), 2.11 (s, 6H), 1.50 (s, 3H), 1.36 (s, 3H).
HRMS (FAB) m/z: calcd for C35H44N4O5SNa [M þ Na]þ
655.2930; found 655.2927.
(R)-N-(2-Methylbenzyl)-3-[(2S,3S)-3-(4-dimethylamino-2,6-
dimethyphenoxyacetyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-
dimethyl-1,3-thiazolidine-4-carboxamide (13d). To a solution of

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7615
HCl salt of H-Apns-Dmt-NH(o-methylbenzyl)10 (74 mg, 0.14
mmol) in DMF were added triethylamine (19 μL, 0.14 mmol),
HOBt H2O (23 mg, 0.15 mmol), 4-(dimethylamino)-2,6-di-
compound 5q in a manner similar to that described for compound
13b. Yield 88%. 1H NMR (300 MHz, DMSO-d6) δ (ppm): 8.13 (d,
J=8.7 Hz, 1H), 7.94 (d, J=7.2 Hz, 1H), 7.37-7.14 (m, 7H), 6.76
(s, 2H), 4.94 (dd, J=13.2 Hz, 9.0 Hz, 2H), 4.48 (d, J=3.6 Hz, 1H),
4.43 (s, 1H), 4.40-4.29 (m, 1H), 4.16 (d, J=14.1 Hz, 1H), 4.04-
3.94 (m, 2H), 2.81-2.71 (m, 2H), 2.13 (s, 6H), 1.83-1.71 (m, 2H),
1.40-1.68 (m, 9H), 1.37 (s, 3H). HRMS (FAB): HRMS (FAB)
m/z: calcd for C31H42N4O5SNa) [M þ Na]þ 605.2774; found
605.2769.
3
methylphenoxyacetic acid (12) (89 mg, 0.14 mmol), and EDC
3
HCl (29 mg, 0.15 mmol) in an ice bath, and the mixture was
stirred overnight at room temperature. After removal of the
solvent in vacuo, the residue was added into EtOAc, washed
sequentially with 5% NaHCO3 and saturated NaCl, dried over
MgSO4, and concentrated in vacuo. Purification of the product
by silica gel column chromatography gave 56 mg of compound
13d as a white solid. Yield 63%. 1H NMR (300 MHz, DMSO-d6)
δ (ppm): 8.37 (t, J=5.4 Hz, 1H), 8.09 (d, J=8.3 Hz, 1H), 7.33-
7.07 (m, 11H), 6.53 (br, 1H), 4.98 (d, J=9.3 Hz, 1H), 4.94 (d, J=
9.0 Hz, 1H), 4.50 (s, 1H), 4.47 (d, J=3.5 Hz, 1H), 4.45-4.35 (m,
2H), 4.24-4.01 (m, 2H), 3.95-3.87 (m, 1H), 2.90-2.72 (m,
10H), 2.26 (s, 3H), 2.11 (s, 6H), 1.51 (s, 3H), 1.36 (s, 3H). HRMS
(FAB) m/z: calcd for C36H46N4O5SNa [M þ Na]þ 669.3087;
found 669.3081.
(R)-N-Allyl-3-[(2S,3S)-3-(4-amino-2,6,-dimethyphenoxyacetyl)-
amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-thiazoli-
dine-4-carboxamide (13j). Compound 13j was prepared from
compound 5a in a manner similar to that described for
1
compound 13b. Yield 92%. H NMR (300 MHz, DMSO-d6)
δ (ppm): 8.16 (t, J=5.9 Hz, 1H), 8.11 (d, J=8.6 Hz, 1H), 7.34-
7.14 (m, 7H), 6.86 (s, 2H), 5.47 (br, 1H), 5.84-5.71 (m, 1H),
5.21 (dd, J=17.1 Hz, 1.7 Hz, 1H), 5.02 (dd, J=10.4 Hz, 1.7 Hz,
1H), 4.96 (d, J=4.0 Hz, 2H), 4.47 (d, J=3.8 Hz, 1H), 4.44 (s,
1H), 4.41-4.37 (m, 1H), 4.17 (d, J=14.1 Hz, 1H), 4.01 (d, J=
14.3 Hz, 1H), 3.72 (br, 2H), 2.85-2.71 (m, 2H), 2.15 (s, 6H),
1.51 (s, 3H), 1.37 (s, 3H). HRMS (FAB) m/z: calcd for
(R)-N-(2,6-Dimethylbenzyl)-3-[(2S,3S)-3-(4-amino-2,6-dimethy-
phenoxyacetyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-
1,3-thiazolidine-4-carboxamide (13e). Compound 13e was pre-
pared from compound 5j in a manner similar to that described
for compound 13a. Yield 94%. 1H NMR (300 MHz, DMSO-d6)
δ(ppm):8.18(d, J=9.0 Hz, 1H), 8.10 (br, 1H), 7.39-6.96 (m, 10H),
6.77 (s, 2H), 4.98 (d, J=8.7 Hz, 1H), 4.93 (d, J=8.7 Hz, 1H), 4.54 (s,
1H), 4.51-4.12 (m, 2H), 4.38-4.26 (m, 1H), 4.22-4.11 (m, 2H),
4.00 (d, J=14.4 Hz, 1H), 2.82-2.70 (m, 2H), 2.30 (s, 6H), 2.14 (s,
6H), 1.46 (s, 3H), 1.37 (s, 3H). HRMS (FAB) m/z: calcd for
C35H44N4O5SNa [M þ Na]þ 655.2930; found 655.2934.
C
29H38N4O5SNa [M þ Na]þ 577.2461; found 577.2456.
(R)-N-(2-Methylallyl)-3-[(2S,3S)-3-(4-amino-2,6-dimethyphe-
noxyacetyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-
thiazolidine-4-carboxamide (13k). Compound 13k was prepared
from compound 5v in a manner similar to that described for
1
compound 13b. Yield 45%. H NMR (300 MHz, DMSO-d6)
δ (ppm): 8.15 (t, J=5.9 Hz, 1H), 8.06 (d, J=9.2 Hz, 1H), 7.34-
7.15 (m, 7H), 6.58 (s, 2H), 5.47 (br, 1H), 4.96 (d, J=4.1 Hz, 2H),
4.90 (s, 1H), 4.73 (s, 1H), 4.52-4.29 (m, 3H), 4.12 (d, J=15.0 Hz,
1H), 3.94 (d, J=14.5 Hz, 1H), 3.78 -3.54 (m, 2H, overlapped
with H2O), 2.81-2.69 (m, 2H), 2.09 (s, 6H), 1.66 (s, 3H), 1.52 (s,
3H), 1.38 (s, 3H). HRMS (FAB) m/z: calcd for C30H40N4O5SNa
[M þ Na]þ 591.2617; found 591.2623.
(R)-N-(2,6-Dichlorobenzyl)-3-[(2S,3S)-3-(4-amino-2,6-dimethy-
phenoxyacetyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimeth-
yl-1,3-thiazolidine-4-carboxamide (13e). Compound 13e was
prepared from compound 5c in a manner similar to that
described for compound 13b. Yield 31%. 1H NMR (300
MHz, DMSO-d6) δ (ppm): 8.30 (t, J =5.1 Hz, 1H), 8.12 (d,
J=9.2 Hz, 1H), 7.50-7.14 (m, 10H), 6.65 (s, 2H), 4.94 (d, J=
2.2 Hz, 1H), 4.63-4.57 (m, 1H), 4.55 (s, 1H), 4.52-4.43 (m,
1H), 4.38-4.28 (m, 1H), 4.15 (d, J=14.1 Hz, 1H), 3.96 (d, J=
14.1 Hz, 1H), 2.80-2.71 (m, 2H), 2.11 (s, 6H), 1.46 (s, 3H), 1.35
(s, 3H). HRMS (FAB) m/z: calcd for C33H38Cl2N4O5SNa [M þ
Na]þ 695.1838; found 695.1845.
(R)-N-(cis-4-Hydroxy-2-buten-1-yl)-3-[(2S,3S)-3-(4-amino-2,6-
dimethyphenoxyacetyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-di-
methyl-1,3-thiazolidine-4-carboxamide (13l). Compound 13l was
prepared from compound 5w ina manner similar to that described
for compound 13b. Yield 76%. 1H NMR (300 MHz, DMSO-d6)
δ (ppm): 8.18-8.06 (m, 2H), 7.35-7.16 (m, 7H), 6.75 (s, 2H),
5.58-5.47 (m, 1H), 5.39-5.28 (m, 1H), 4.95 (dd, J=14.7 Hz, 9.1
Hz, 1H), 4.47 (d, J=3.8 Hz, 1H), 4.39 (s, 1H), 4.38-4.31 (m, 1H),
4.15 (d, J=14.1 Hz, 1H), 4.03-3.96 (m, 2H), 3.80-3.69 (m, 2H,
overlapped with H2O), 2.81-2.70 (m, 2H overlapped with
DMSO), 2.13 (s, 6H), 1.50 (s, 3H), 1.36 (s, 3H). HRMS (FAB)
m/z: calcd for C30H40N4O6SNa [M þ Na]þ 607.2566; found
607.2573.
(R)-N-[(S)-Indan-1-yl]-3-[(2S,3S)-3-(4-amino-2,6-dimethyphe-
noxyacetyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-
thiazolidine-4-carboxamide (13g). Compound 13g was prepared
from compound 5k in a manner similar to that described for
1
compound 13b. Yield 92%. H NMR (300 MHz, DMSO-d6)
δ (ppm): 8.34 (d, J=7.9 Hz, 1H), 8.10 (d, J=9.2 Hz, 1H), 7.32-
7.12 (m, 11H), 6.62 (s, 2H), 5.30 (dd, J=15.3 Hz, 8.1 Hz, 1H),
5.04 (d, J=9.2 Hz, 1H), 4.95 (d, J=8.8 Hz, 1H), 4.48 (d, J=
3.7 Hz, 1H), 4.46 (s, 1H), 4.42-4.31 (m, 1H), 4.14 (d, J=14.1 Hz,
1H), 3.94 (d, J=14.1 Hz, 1H), 2.99-2.70 (m, 4H), 2.44-2.30 (m,
1H, overlapped with H2O), 2.11 (s, 6H), 1.91-1.77 (m, 1H), 1.51
(s, 3H), 1.45 (s, 3H). HRMS (FAB) m/z: calcd for C35H42N4O5S-
Na [M þ Na]þ 653.2774; found 653.2780.
Water Solubility Test. Compounds were suspended in pure
water, left under sonication for 30 min, then cooled to room
temperature. The saturated solutions were passed through a
centrifugal filter (0.45 nm filter unit, Ultrafree-MC, Millipore).
The filtrate was analyzed using RP-HPLC.
HIV Protease Inhibition. HIV protease inhibitory activity of
the test compounds was determined on the basis of the inhibition
of the HIV protease substrate (H-Lys-Ala-Arg-Val-Tyr-Phe(p-
NO2)-Glu-Ala-Nle-NH2) cleavage using recombinant HIV-1
protease. HIV protease substrate was synthesized by solid phase
methods. Recombinant HIV-1 protease was purchased from
Bachem AG, Bubendorf, Switzerland. In the inhibition assay,
25 μL of 200 mM 2-(N-morpholino)ethanesulfonic acid (MES)-
NaOH buffer (pH 5.5), containing 2 mM dithiothreitol, 2 mM
EDTA-2Na, and 1 M NaCl was mixed with 5 μL of the inhibitor
(500 nM) dissolved in DMSO and 10 μL of HIV-1 protease
(2 μg/mL) in 50 mM AcOH (pH 5.0) containing 1 mM EDTA-
2Na, 25 mM NaCl, 0.2% 2-mercaptoethanol, 0.2% Nonidet P-
40, and 10% glycerol. The reaction was initiated by adding of
10 μL of a 1.0 mM substrate solution. After incubation for 15 min
at 37 ꢀC, the reaction was terminated by the addition of 1 N HCl,
and the N-terminal cleavage fragment (H-Lys-Ala-Arg-Val-
Tyr-OH) was separated by reversed-phase HPLC on a C18
(R)-N-[(1S,2R)-2-Hydroxyindan-1-yl]-3-[(2S,3S)-3-(4-amino-
2,6-dimethyphenoxyacetyl)amino-2-hydroxy-4-phenylbutanoyl]-
5,5-dimethyl-1,3-thiazolidine-4-carboxamide (13h). Compound
13h was prepared from compound 5n in a manner similar
to that described for compound 13b. Yield 46%. 1H NMR
(300 MHz, DMSO-d6) δ (ppm): 8.21 (d, J=8.8 Hz, 1H), 8.10 (d,
J=8.8 Hz, 1H), 7.37 (d, J=7.3 Hz, 2H), 7.31-7.10 (m, 8H), 7.03
(t, J=7.2 Hz, 1H), 6.78 (s, 2H), 5.28 (dd, J=8.1 Hz, 4.7 Hz, 1H),
4.96 (s, 2H), 4.76 (s, 1H), 4.50 (d, J=2.9 Hz, 1H), 4.45-4.27 (m,
2H), 4.19 (d, J=14.3 Hz, 1H), 4.07-3.96 (m, 1H), 3.05 (dd, J=
16.0 Hz, 5.1 Hz, 1H), 2.92-2.71 (m, 3H), 2.14 (s, 6H), 1.56 (s,
3H), 1.48 (s, 3H). HRMS (FAB) m/z: calcd for C35H42N4O6SNa
[M þ Na]þ 669.2723; found 669.2728.
(R)-N-Cyclopentyl-3-[(2S,3S)-3-(4-amino-2,6-dimethyphenoxy-
acetyl)amino-2-hydroxy-4-phenylbutanoyl]-5,5-dimethyl-1,3-thia-
zolidine-4-carboxamide (13i). Compound 13i was prepared from

7616 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
Hidaka et al.
column (3.0 mm ꢀ 75 mm. YMC Pack ODS AS-3E7) with a
linear gradient of CH3CN in 0.1% aqueous TFA at a flow rate of
1.0 mL/min, and its quantity was determined by monitoring
fluorescence intensity (excitation, 275 nm; emission, 305 nm).
In the case of 1 nM assay, 10 nM inhibitor dissolved in DMSO
and HIV-1 protease (0.04 μg/mL) were used, and the sample was
incubated for 3 h at 37 ꢀC. Ki values were estimated by fitting the
data from several substrate and inhibitor concentrations to
standard equations for tight binding competitive inhibitors with
a similar procedure described in ref 13a.
Anti-HIV Activity. Anti-HIV activity of test compounds was
determined on the basis of inhibition of HIV-induced cytopathic
effect in MT-4 cells in vitro as previously reported.28 HIV-1
wild-type (IIIB, pNL4-3) or IND-R was inoculated to MT-4
cells in a 96-well plate. The resulting culture was treated with an
equal volume of a 1% DMSO solution of each test compound
with several concentrations and 10% fetal bovine serum and was
incubated for 5 days in a CO2 incubator at 37 ꢀC, in triplicate.
After treatment with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-
tetrazolium bromide (MTT), the optical density of the plate was
measured and the percent cytopathic effect reduction was
calculated. Then EC50 values were estimated by fitting the data
to a median-effect equation. Cytotoxicity (TD50) was deter-
mined by incubation in the absence of the virus.
X-ray Crystallographic Analysis. Preparation of HIV pro-
tease for crystallization was performed as reported.8 The hang-
ing drop vapor diffusion method was used for crystallization.
Crystals were set up with a 2-fold molar excess of inhibitors to
protease of 2.0 mg/mL concentration. The reservoir solution
consisted of 126 mM phosphate buffer at pH 5.0, 63 mM sodium
citrate, and 0.2 M ammonium sulfate. The obtained crystal was
soaked into the precipitant solution containing 45% (w/v)
glycerol, then flash-frozen under a N2 gas cryostream (100 K).
Data were collected at the BL41XU beamline in SPring-8 and
processed using HKL2000.32 The crystals of HIV protease/13k
complex belong to the monoclinic space group P2(1)2(1)2, with
(2) Chrusciel, R. A.; Strohbach, J. W. Curr. Top. Med. Chem. 2004, 4,
1097–1114.
(3) Randolph, J. T.; DeGoey, D. Curr. Top. Med. Chem. 2004, 4, 1079–
1095.
(4) Johnson, V. A.; Brun-Vezinet, F; Clotet, B.; Gunthard, H. F.;
ꢀ
Kuritzkes, D. R.; Pillay, D.; Schapiro, J. M.; Richman, D. D.
Update of the drug resistance mutations in HIV-1: 2007. Top. HIV
Med. 2007, 15, 119–125.
(5) (a) Mimoto, T.; Imai, J.; Tanaka, S.; Hattori, N.; Takahashi, O.;
Kisanuki, S.; Nagano, Y.; Shintani, M.; Hayashi, H.; Sakikawa,
K.; Akaji, K.; Kiso, Y. Rational design and synthesis of a novel
class of active site-targeted HIV protease inhibitors containing a
hydroxymethylcarbonyl isostere. Use of phenylnorstatine or allo-
phenylnorstatine as a transition-state mimic. Chem. Pharm. Bull.
1991, 39, 2465–2467. (b) Mimoto, T.; Imai, J.; Tanaka, S.; Hattori, N.;
Kisanuki, S.; Akaji, K.; Kiso, Y. KNI-102, a novel tripeptide HIV
protease inhibitor containing allophenylnorstatine as a transition-state
mimic. Chem. Pharm. Bull. 1991, 39, 3088–3090. (c) Kiso, Y.;
Matsumoto, H.; Mizumoto, S.; Kimura, T.; Fujiwara, Y.; Akaji, K.
Small dipeptide-based HIV protease inhibitors containing the hydro-
xymethylcarbonyl isostere as an ideal transition-state mimic. Biopoly-
mers 1999, 51, 59–68.
(6) Baldwin, E. T.; Bhat, T. N.; Gulnik, S.; Liu, B.; Topol, I. A.; Kiso,
Y.; Mimoto, T.; Mitsuya, H.; Erickson, J. W. Structure of HIV-1
protease with KNI-272, a tight-binding transition-state analog
containing allophenylnorstatine. Structure 1995, 3, 581–590.
(7) Wang, Y.-X.; Freedberg, D. I.; Yamazaki, T.; Wingfield, P. T.;
Stahl, S. J.; Kaufman, J. D.; Kiso, Y.; Torchia, D. A. Solution
NMR evidence that the HIV-1 protease catalytic aspartyl groups
have different ionization states in the complex formed with the
asymmetric drug KNI-272. Biochemistry 1996, 35, 9945–9950.
(8) Adachi, M.; Ohhara, T.; Kurihara, K.; Tamada, T.; Honjo, E.;
Okazaki, N.; Arai, S.; Shoyama, Y.; Kimura, K.; Matsumura, H.;
Sugiyama, S.; Adachi, H.; Takano, K.; Mori, Y.; Hidaka, K.;
Kimura, T.; Hayashi, Y.; Kiso, Y.; Kuroki, R. Structure of HIV-1
protease in complex with potent inhibitor KNI-272 determined by
high resolution X-ray and neutron crystallography. Proc. Natl.
Acad. Sci. U.S.A. 2009, 106, 4641–4646.
(9) Kiso, Y. Design and synthesis of a covalently linked HIV-1
protease dimer analog and peptidomimetic inhibitors. J. Synth.
Org. Chem., Jpn. 1998, 56, 896–907.
(10) Mimoto, 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.
(11) Matsumoto, H.; Kimura, T.; Hamawaki, T.; Kumagai, A.; Goto,
T.; Sano, K.; Hayashi, Y.; Kiso, Y. Design, synthesis, and biolo-
gical evaluation of anti-HIV double-drugs: conjugates of HIV
protease inhibitors with a reverse transcriptase inhibitor through
spontaneously cleavable linkers. Bioorg. Med. Chem. 2001, 9,
1589–1600.
(12) Yoshimura, K.; Kato, R.; Yusa, K.; Kavlick, M. F.; Maroun, V.;
Nguyen, A.; Mimoto, T.; Ueno, T.; Shintani, M.; Falloon, J.;
Masur, H.; Hayashi, H.; Erickson, J.; Mitsuya, H. JE-2147: a
dipeptide protease inhibitor (PI) that potently inhibits multi-PI-
resistant HIV-1. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 8675–8680.
(13) (a) Mimoto, T.; Terashima, K.; Nojima, S.; Takaku, H.; Nakayama,
M.; Shintani, M.; Yamaoka, T.; Hayashi, H. Structure-activity
and structure-metabolism relationships of HIV protease inhibi-
tors containing the 3-hydroxy-2-methylbenzoyl-allophenylnorsta-
tine structure. Bioorg. Med. Chem. 2004, 12, 281–293. (b) Mimoto,
T.; Nojima, S.; Terashima, K.; Takaku, H.; Shintani, M.; Hayashi, H.
Structure-activity relationships of novel HIV-1 protease inhibitors
containing the 3-amino-2-chlorobenzoylallophenylnorstatine structure.
Bioorg. Med. Chem. 2008, 16, 1299–1308.
˚
˚
˚
unit cell dimensions a=58.2 A, b=85.8 A, c=46.5 A, R=90ꢀ, β=
90ꢀ, γ = 90ꢀ. The structure was refined to a crystallographic
R-factor of 10.1% (free R-factor=11.8%) at 0.88A resolution
˚
using the program SHELX-97.33
Acknowledgment. We are sincerely grateful to S. Nishii for
his assistancewith inhibitor synthesis andto Y. Matsui, T. Ito,
H. Tsukamoto, and T. Hamada for the determination of
HIV-1 protease inhibitory activity. This research was sup-
ported in part by the Frontier Research Program and the 21st
Century COE Program from the Ministry of Education,
Culture, Sports, Science and Technology (MEXT), Japan;
the Japan Health Sciences Foundation; and Japan Society for
Promotion of Science’s Post-Doctoral Fellowship for Foreign
Researchers. The synchrotron radiation experiments of the
HIV protease/13k complex were performed at the BL41XU
beamline in SPring-8 with the approval of the Japan Synchro-
tron Radiation Research Institute (JASRI) (Proposal No.
2007A1513). We acknowledge the help provided by beamline
staff at SPring-8 (Drs. N. Shimizu, M. Kawamoto, and
M. Yamamoto).
(14) Vega, S.; Kang, L. W.; Velazquez-Campoy, A.; Kiso, Y.; Amzel, L.
M.; Freire, E. A structural and thermodynamic escape mechanism
from a drug resistant mutation of the HIV-1 protease. Proteins:
Struct., Funct., Bioinf. 2004, 55, 594–602.
(15) Reiling, K. K.; Endres, N. F.; Dauber, D. S.; Craik, C. S.; Stroud,
R. M. Anisotropic dynamics of the JE-2147-HIV protease com-
plex: drug resistance and thermodynamic binding mode examined
Supporting Information Available: Results of HPLC and
elemental analysis of target compounds, QSAR equations,
and synthetic details for preparation of 4a-w and 5a-w. This
material is available free of charge via the Internet at http://
pubs.acs.org.
˚
in a 1.09 A structure. Biochemistry 2002, 41, 4582–4594.
(16) Hidaka, K.; Kimura, T.; Ruben, A. J.; Uemura, T.; Kamiya, M.;
Kiso, A.; Okamoto, T.; Tsuchiya, Y.; Hayashi, Y.; Freire, E.; Kiso,
Y. Antimalarial activity enhancement in hydroxymethylcarbonyl
(HMC) isostere-based dipeptidomimetics targeting malarial aspar-
tic protease plasmepsin. Bioorg. Med. Chem. 2008, 16, 10049–
10060.
References
(1) Hoffmann, C.; Mulcahy, F. Overview of Antiretroviral Agents. In
HIV Medicine 2006; Hoffmann, C., Rockstroh, J. K., Kamps, B. S.,
Eds.; Flying Publisher (FlyingPublisher.com): Paris, 2006; pp 94-129.
(17) Abdel-Rahman, H. M.; El-Koussi, N. A.; Alkaramany, G. S.;
Youssef, A. F.; Kiso, Y. A novel dipeptide-based HIV protease

Article
inhibitor containing allophenylnorstatine. Arch. Pharm.
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7617
tive structure property relationship studies. Anal. Chem. 1990, 62,
2323–2329.
(Weinheim, Ger.) 2004, 337, 587–598.
(18) Munshi, S.; Chen, Z.; Li, Y.; Olsen, D. B.; Fraley, M. E.; Hungate,
R. W.; Kuo, L. C. Rapid X-ray diffraction analysis of HIV-1
protease-inhibitor complexes: inhibitor exchange in single crystals
ofthe boundenzyme. ActaCrystallogr., Sect. D 1998, 54, 1053–1060.
(19) (a) Kimura, T.; Hidaka, K.; Tanaka, Y.; Matsumoto, H.; Matsui,
Y.; Abdel-Rahman, H. M.; Hayashi Y.; Kiso, Y. Identification of
Dipeptide-Type HIV-1 Protease Inhibitors with High Antiviral
Activity. Presented at the 5th AFMC International Medicinal
Chemistry Symposium, Kyoto, Japan, Oct 2003. (b) Kimura, T.;
Hidaka, K.; Tanaka, Y.; Matsumoto, H.; Abdel-Rahman, H. M.; Rajesh,
S.; Wang, J.; Ami, E.; Hayashi, Y.; Kiso, Y. Discovery of Potent HIV-1
Protease Inhibitors Containing Allophenylnorstatine Based on a Transi-
tion-State Mimic. Presented at the 3rd General Meeting of the Interna-
tional Proteolysis Society/International Conference on Protease
Inhibitors, Nagoya, Japan, Nov 2003. (c) Kimura, T.; Hidaka, K.;
Abdel-Rahman, H. M.; Matsumoto, H.; Tanaka, Y.; Matsui, Y.; Hayashi,
Y.; Kiso, Y. Design and Synthesis of Dipeptide-Type HIV-1 Protease
Inhibitors with High Antiviral Activity. In Peptide Science 2003; Ueki,
M., Ed.; The Japanese Peptide Society: Osaka, Japan, 2004; pp241-244.
(20) (a) Canon-Koch, S. S.; Alexander, T. N.; Barvian, M.; Bolton, G.;
Boyer, F. E.; Burke, B. J.; Holler, T.; Jewell, T. M.; Prasad, J. V.;
Kucera, D. J.; Linton, M. A.; Machak, J.; Mitchell, L. J.; Murphey,
S. T.; Reich, S. H.; Skalitzky, D. J.; Tatlock, J. H.; Varney, M. D.;
Virgil, S. C.; Webber, S. E.; Worland, S. T; Melnick, M. Patent WO
02/100844, 2002. (b) Canon-Koch, S. S.; Alexander, T. N.; Barvian,
M.; Bolton, G.; Boyer, F. E.; Burke, B. J.; Holler, T.; Jewell, T. M.;
Prasad, J. V.; Kucera, D. J.; Machak, J.; Mitchell, L. J.; Murphey, S. T.;
Reich, S. H.; Skalitzky, D. J.; Tatlock, J. H.; Varney, M. D.; Virgil, S. C.;
Worland, S. T; Melnick, M.; Linton, M. A.; Webber, S. E. PCT Patent
WO 02/100845, 2002.
(21) (a) Hidaka, K.; Kimura, T.; Hayashi, Y.; McDaniel, K. F.;
Dekhtyar, T.; Colletti, L.; Kiso, Y. Design and synthesis of
pseudo-symmetric HIV protease inhibitors containing a novel
hydroxymethylcarbonyl (HMC)-hydrazide isostere. Bioorg. Med.
Chem. Lett. 2003, 13, 93–96. (b) Chen, X.; Kempf, D. J.; Li, L.; Sham,
H. L.; Vasavanonda, S.; Wideburg, N. E.; Saldivar, A.; Marsh, K. C.;
McDonald, E.; Norbeck, D. W. Synthesis and SAR studies of potent
HIV protease inhibitors containing novel dimethylphenoxyl acetates as
P2 ligands. Bioorg. Med. Chem. Lett. 2003, 13, 3657–3660.
(22) Sohma, Y.; Hayashi, Y.; Ito, T.; Matsumoto, H.; Kimura, T.;
Kiso, Y. Development of water-soluble prodrug of the HIV-1
proteaase inhibitor KNI-727: importance of the conversion
time for higher gastrointestinal absorption of prodrugs based on
spotaneous chemical cleavage. J. Med. Chem. 2003, 46, 4124–
4135.
(25) Wildman, S. A.; Crippen, G. M. Prediction of physiochemical
parameters by atomic contributions. J. Chem. Inf. Comput. Sci.
1999, 39, 868–873.
(26) Miller, J. F.; Andrews, C. W.; Brieger, M.; Furfine, E. S.; Hale, M.
R.; Hanlon, M. H.; Hazen, R. J.; Kaldor, I.; McLean, E. W.;
Reynolds, D.; Sammond, D. M.; Spaltenstein, A.; Tung, R.;
Turner, E. M.; Xu, R. X.; Sherrill, R. G. Ultra-potent P1 modified
arylsulfonamide HIV protease inhibitors: the discovery of
GW0385. Bioorg. Med. Chem. Lett. 2006, 16, 1788–1794.
(27) (a) Leonard, J. T.; Roy, K. QSAR by LFER model of HIV protease
inhibitor mannitol derivatives using FA-MLR, PCRA, and PLS
techniques. Bioorg. Med. Chem. 2006, 14, 1039–1046. (b) Garg, R.;
Patel, D. Hydrophobicity in the design of P2/P20 tetrahydropyrimidi-
none HIV protease inhibitors. Bioorg. Med. Chem. Lett. 2005, 15,
3767–3770. (c) Bhhatarai, B.; Garg, R. From SAR to comparative
QSAR: role of hydrophobicity in the design of 4-hydroxy-5,6-dihy-
dropyran-2-ones HIV-1 protease inhibitors. Bioorg. Med. Chem. 2005,
13, 4078–4084.
(28) Molla, A.; Vasavanonda, S.; Kumar, G.; Sham, H. L.; Johnson,
M.; Grabowski, B.; Denissen, J. F.; Kohlbrenner, W.; Plattner, J.
J.; Leonard, J. M.; Norbeck, D. W.; Kempf, D. J. Human serum
attenuates the activity of protease inhibitors toward wild-type and
mutant human immunodeficiency virus. Virology 1998, 250, 255–
262.
(29) (a) Martinez-Picado, J.; Savara, A. V.; Sutton, L.; D’Aquila, R. T.
Replicative fitness of protease inhibitor-resistant mutants of
human immunodeficiency virus type 1. J. Virol. 1999, 73, 3744–
3752. (b) Condra, J. H.; Schleif, W. A.; Blahy, O. M.; Gabryelski, L. J.;
Graham, D. J.; Quintero, J. C.; Rhodes, A.; Shivaprakash, M.; Titus, D.;
Yang, T.; Teppler, H.; Squires, K. E.; Deutsch, P. J.; Emini, E. A. In vivo
emergence of HIV-1 variants resistant to multiple protease inhibitors.
Nature 1995, 374, 569–571.
(30) Mo, H.; Lu, L.; Dekhtyar, T.; Stewart, K. D.; Sun, E.; Kempf, D.
J.; Molla, A. Characterization of resistant HIV variants generated
by in vitro passage with lopinavir/ritonavir. Antiviral Res. 2003, 59,
173–180.
(31) Zhao, C.; Sham, H. L.; Sun, M.; Stoll, V. S.; Stewart, K. D.; Lin, S.;
Mo, H.; Vasavanonda, S.; Saldivar, A.; Park, C.; McDonald, E. J.;
Marsh, K. C.; Klein, L. L.; Kempf, D. J.; Norbeck, D. W. Synthesis
and activity of N-acylazacyclic urea HIV-1 protease inhibitors with
high potency against multiple drug resistant viral strains. Bioorg.
Med. Chem. Lett. 2005, 15, 5499–5503.
(32) Otwinowski, Z.; Minor, W. Processing of X-ray Diffraction Data
Collected in Oscillation Mode. In Methods in Enzymology; Carter,
C. W., Jr., Sweet, R. M., Eds.; Academic Press: New York, 1997; Vol.
276, pp 307-326.
(33) Sheldrick, G. M.; Schneider, T. R. SHELXL: High-Resolu-
tion Refinement. In Methods in Enzymology; Carter, C. W., Jr.,
Sweet, R. M., Eds.; Academic Press: New York, 1997; Vol. 277, pp
319-343.
(23) Gasteiger, J.; Marsili, M. Iterative partial equalization of orbital
electronegativity. A rapid access to atomic charges. Tetrahedron
1980, 36, 3219–3228.
(24) Stanton, D. T.; Jurs, P. C. Development and use of charged partial
surface area structural descriptors in computer assisted quantita-