Sidechain-linked inhibitors of HIV-1 protease dimerization

Article abstract of DOI:10.1016/j.bmc.2008.02.060

There is a great need for alternative modes of inhibition for the design of anti-HIV therapies, due to the increased resistance of HIV to currently approved drugs. A novel strategy for generating potent dimerization inhibitors of HIV-1 protease is describ

Full text of DOI:10.1016/j.bmc.2008.02.060

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 17 (2009) 967–976
Sidechain-linked inhibitors of HIV-1 protease dimerization
Michael J. Bowman and Jean Chmielewski*
Department of Chemistry, 560 Oval Drive, Purdue University, West Lafayette, IN 47907, USA
Received 13 November 2007; revised 15 February 2008; accepted 16 February 2008
Available online 11 March 2008
Abstract—There is a great need for alternative modes of inhibition for the design of anti-HIV therapies, due to the increased resis-
tance of HIV to currently approved drugs. A novel strategy for generating potent dimerization inhibitors of HIV-1 protease is
described based on sidechain-linked interfacial peptides. In a number of cases the activity of these agents against HIV-1 protease
was found to be among the most potent reported, with inhibitory constants in the low nM range.
ꢀ 2008 Elsevier Ltd. All rights reserved.
1. Introduction
2. Results and discussion
In its catalytically active form HIV-1 protease (HIV-1
PR) exists as a homodimer1 with half of its active site
and substrate binding site contributed from each mono-
mer.2,3 A number of agents that target the active site of
HIV PR have been approved for the treatment of HIV
infection. Although these inhibitors have been shown
to be potent in vivo, the effect is often short-lived as
strains containing mutations around the active site of
HIV PR become more prevalent.4 Homodimeric HIV-
1 PR can serve as a template for developing methods
to inhibit protein–protein interactions because it con-
tains a relatively small, highly conserved region5 that
provides the bulk of the stabilization of HIV PR upon
dimerization.6 We have previously reported a minimal
structure (1), consisting of crosslinked interfacial pep-
tides, that is required for dimerization inhibition.7 To
date all efforts to link interfacial HIV-1 PR peptides
and generate dimerization inhibitors of HIV-1 PR have
employed the amino-terminus of the peptides as the site
for crosslinking.8 The prerequisite that the linking agent
be fairly hydrophobic8 has hindered the solubility of
these agents and subsequent structural and cell-based
investigations. In this current study we report a novel
crosslinking strategy that allows for increased structural
diversity and aqueous solubility of the designed agents
while maintaining dimerization inhibition.
2.1. Design of sidechain-linked agents
In our design we replaced the hydrophobic, amino-ter-
minal crosslink of 1 with a more hydrophilic tether that
links the sidechains of each interfacial peptide fragment.
This approach serves to free up the amino-termini for
diverse modification through acylation. To facilitate
the sidechain crosslinking of the interfacial peptides,
the threonine and asparagine residues of 1 were replaced
with cysteine residues (Fig. 1). These two residues were
chosen for modification because modeling had indicated
that these residues are directed to the solvent exposed
face of the interfacial b-sheet structure of HIV PR
(Fig. 1). The crosslinking agent, 1,2-bis-(2-chloroeth-
˚
oxy)ethane was designed to span the 10 A distance be-
tween the backbones of both peptide fragments.9 This
design strategy would allow the amino termini to be fur-
ther functionalized with hydrophobic groups to provide
potential interactions with the hydrophobic cleft be-
tween residues Leu97 and Phe99 of an HIV PR mono-
mer (Fig. 1B).
2.2. Synthesis of sidechain-linked agents
The initial synthesis of the Northern and Southern pep-
tides was initiated with Boc-cystine which was elongated
with additional amino acids using HBTU to provide the
disulfides of the desired peptide fragments. Reduction of
the disulfides with DTT in chloroform provided the de-
sired peptide components 2 and 4 (Scheme 1). The cys-
teine-containing dipeptide, Boc-Cys-Phe(Ot-Bu) (2),
was treated with bis-chloroethoxyethane and potassium
carbonate in DMF at 60 ꢁC to provide the mono-chlori-
Keywords: HIV protease; Inhibition; Protein–protein interaction.
*
Corresponding author. Tel.: +1 765 494 0132; fax: +1 765 494
0239; e-mail: chml@purdue.edu
0968-0896/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.02.060

968
M. J. Bowman, J. Chmielewski / Bioorg. Med. Chem. 17 (2009) 967–976
nated intermediate 3 (Scheme 1). A 10-fold excess of bis-
chloroethoxyethane was used to prevent the undesired
bis-substituted adduct.
Compound 3 was treated with the cysteine-containing
tripeptide (4) and potassium carbonate in DMF at
80 ꢁC to provide crosslinked and protected 5. Com-
pound 5 was treated with TFA containing TIPS
(2.5%) and water (2.5%) to remove the tert-butyl pro-
tecting groups, followed by acylation of the free ami-
no-termini with a series of activated carboxylic acids
to provide the desired agents 6–10.
2.3. Evaluation of potency of initially designed agents
The extent of HIV-1 PR inhibition of 6–10 was evalu-
ated using the fluorogenic substrate developed by Toth
and Marshall.10 We found that compounds with the
alternative crosslinking strategy (6–10) demonstrated
significant inhibition against HIV-1 PR when the appro-
priate acyl group was used. A trend of increasing po-
tency with increasing acyl chain length was observed,
with the most potent compound, 10, containing hepta-
noyl groups at the amino termini (Fig. 2). Interestingly,
compound 10 was found to be slightly more potent than
the original inhibitor with an amino-terminal crosslink
(1). These data demonstrate that moderately potent
inhibitors of HIV PR can be obtained from a sidechain
crosslinking strategy. The results are consistent with our
binding model in which the alkyl chains, R, are more in-
volved in interactions with the hydrophobic cleft exist-
ing between Leu97 and Phe99 of HIV-1 PR as their
length is increased.
Further kinetic analyses were required in order to deter-
mine the mechanism of inhibition for compounds 6–10.
The assay method of Zhang and Poorman11 was used to
determine if these compounds inhibited the activity of
HIV-1 PR by blocking dimerization. The parallel lines
Figure 1. (A) Compound 1 and the proposed structure for sidechain
crosslinked inhibitors. (B) Schematic representations of an inhibitor
(red) binding to the dimerization interface of an HIV-1 PR monomer
(blue) viewed from the hydrophobic interior of HIV-1 PR (R = n-
hexyl).
Scheme 1. Synthesis of sidechain crosslinked compounds.

M. J. Bowman, J. Chmielewski / Bioorg. Med. Chem. 17 (2009) 967–976
969
10. Inhibitor 10 was found to be approximately 2-fold
more potent than compound 1. The flexibility of the R
groups within 10 may allow for improved binding inter-
actions with the interfacial region of HIV-1 PR as com-
pared to the more structurally constrained aliphatic
tether of 1. In addition, the solubility of 9 (90 lM)
and 10 (80 lM) in water was found to be significantly
greater than that observed with 1 (30 lM), presumably
due to the shorter overall alkyl chains lengths and the
inclusion of the ethylene glycol-based ether.
The compound with the shortest acyl chain, 6, however,
was found to not display dimerization inhibition kinetics
as shown by the non-parallel line obtained from the
Zhang–Poorman analysis (Fig. 3A). Interestingly, com-
pound 6 was shown by Lineweaver–Burk analysis to
act by
a
competitive mechanism of inhibition
(Fig. 3B). We believe that 6 does not bind to the active
site of HIV-1 PR because it lacks the heptapeptide struc-
ture associated with recognition and active site binding12
or the equivalent tetrapeptide mimic required for active
site inhibition.13 Alternatively 6 may target the dimer-
ization interface, but may lack enough binding interac-
tions to disrupt the dimer. This hypothesis is
supported by a cross-competition plot using compound
6 and acetyl-pepstatin, a known active site inhibitor of
HIV-1 PR.14,15 If the two inhibitors are not able to bind
simultaneously to HIV-1 PR, then this plot would pro-
vide parallel lines. Figure 4 demonstrates that the lines
of the cross-competition plot for constant acetyl-pepsta-
tin and variable 6 (Fig. 4A), as well constant 6 with var-
iable acetyl-pepstatin (Fig. 4B), are not parallel.
Therefore both inhibitors can occupy the enzyme at
the same time, indicating that 6 is not binding to the ac-
tive site of HIV-1 PR. The plots indicate that the effect
of 6 binding to HIV-1 PR has a positive impact on the
binding of acetyl-pepstatin, perhaps due to the change
in the substrate recognition of HIV-1 PR. Interestingly,
acetyl-pepstatin binding has the opposite effect on 6
binding possibly due to the 4.5 kcal/mol stabilization
of the dimer caused by the binding of acetyl-pepstatin
to the active site.16 With this information in hand, we
have evidence that 6 is not binding to the active site of
the enzyme, but rather acts by binding to an allosteric
site, perhaps at or near the dimerization interface.
Figure 2. Results of the kinetics data for compounds 6–10. Parenthetic
value indicates a competitive mode Ki value.
obtained for 7–10, in relation to uninhibited HIV-1 PR,
demonstrate that these agents are dimerization inhibi-
tors (Fig. 3A). Compounds 7–10 were found again to
be progressively more potent with the subsequent addi-
tion of methylene units, with an overall increase in inhi-
bition of 6-fold going from compound 7 to compound
2.4. A focused library for optimization
With the knowledge that the placement of the crosslink
within HIV PR dimerization inhibitors can be modified
while maintaining potency and a dimerization mecha-
nism of inhibition, the synthesis of a small, focused li-
brary of compounds was undertaken to better
understand the binding requirements for optimized po-
tency. Design modifications attempted to more fully oc-
cupy the hydrophobic cleft between residues Leu97 and
Phe99. Therefore, a series of compounds containing cyc-
lic aliphatic functionality within the acyl moiety (11–14)
was evaluated (Fig. 5). Compound 11, modified with a
cyclohexyl group, was found to be 5-fold less potent
than n-hexyl-containing 10, presumably due to the
diminished occupation of the hydrophobic cleft within
HIV-1 PR with the shorter cyclohexane ring. These data
Figure 3. (A) Zhang–Poorman analysis of novel cross-linked com-
pounds 6–10. s Uninhibited HIV-1 PR; e 65.3 lM 6; , 37.5 lM 7;
d33.7 lM 8; n 11.5 lM 9; h 2 lM 10. (B) Lineweaver–Burk analysis
of compound 6. d Uninhibited HIV-1 PR; s 130 lM compound 6.

970
M. J. Bowman, J. Chmielewski / Bioorg. Med. Chem. 17 (2009) 967–976
bition. With this in mind, the introduction of a methy-
lene spacer before the cyclohexyl-moiety (compound
12) led to a 2-fold increase in potency. Lengthening
the spacer by additional methylene units (13 and 14)
provided approximately a 4-fold increase in potency as
compared to their shorter counterparts (12 and 13).
Inhibitor 14 is 6-fold more potent than 10 with an
IC50 value in the high nM range. Evaluation of the
mechanism of inhibition of HIV-1 PR for 14 using the
Zhang–Poorman analysis demonstrated the compound
to be a potent dimerization inhibitor (Ki of 88 nM),
nearly 15-fold more potent than the n-hexyl-containing
10 and 34-fold more potent than the previously reported
compound 1 (see Fig. 6).7
The incorporation of a phenyl group within the acyl
chain, with one or two methylene spacers, was also
investigated for optimization of activity (Fig. 5). Com-
pound 15, containing phenyl acetyl groups, was nearly
4-fold more potent than the cyclohexyl-containing 12,
demonstrating that the cleft may better accommodate
the aromatic ring. Up to this point, increasing the chain
length had led to increased potency. However, an addi-
tional methylene unit within the acyl chain (16) caused a
2.5-fold decrease in potency as compared to compound
15. This trend was reversed, however, with the incorpo-
ration of an ether linkage (17) in place of a methylene
unit to provide an inhibitor that is nearly equipotent
to 13. Zhang–Poorman analysis also demonstrated that
the dimerization inhibition mechanism was maintained
with agent 17 (Fig. 5).
Figure 4. (A) Yonatani–Theorell plot of v/vi versus acetyl-pepstatin
concentration with constant concentrations of HIV-1 PR and com-
pound 6. d 0 lM compound6; s 22 lM compound 6; . 30 lM
compound 6; n 39 lM compound 6. (B) Yonatani–Theorell plot of v/vi
versus concentration of compound 6 with constant concentrations of
acetyl-pepstatin and HIV-1 PR. d0 nM acetyl-pepstatin; s 7.2 nM
A number of point mutations within compound 1 have
indicated that modifications within the sidechains can
lead to greater potency, including Leu to Ile changes.17
To evaluate whether this trend would hold with the
alternately crosslinked agents, we mutated the Leu resi-
due of compounds 10 and 14 to an Ile residue to provide
compounds 18 and 19 (Fig. 5). Compounds 18 and 19
were found to be 2- to 2.5-fold more potent than their
leucine-containing counterparts 10 and 14, while main-
taining the mechanism of dimerization inhibition. Com-
pound 19 was found to be a highly potent analog with a
Ki value of 40 nM. These results indicate that the side-
chains of this crosslinked inhibitor series can be altered
to provide further increases in potency.
acetyl-pepstatin;
pepstatin.
. 14.4 nM acetyl-pepstatin; n 43.2 nM acetyl-
indicate that the hydrophobicity of the amino-terminal
functionality alone is not a major determinant for inhi-
3. Conclusions
To date all efforts to link interfacial HIV-1 PR peptides
and generate dimerization inhibitors of HIV-1 PR have
employed the amino-terminus of the peptides as the site
for crosslinking. The prerequisite that the linking agent
be fairly hydrophobic has hindered the solubility of
these agents and subsequent structural and cell-based
investigations. Herein we disclose a novel series of po-
tent inhibitors of HIV-1 PR dimerization that contains
a hydrophilic tether spanning the sidechains of interfa-
cial peptides. Complete remodeling of the position of
the crosslink also allowed for the diverse acylation of
the amino-termini of the peptides. With this new strat-
egy, inhibitory potencies against HIV-1 PR among the
Figure 5. Cycloalkyl and aromatic modifications at the R position,
and single amino acid modifications.

M. J. Bowman, J. Chmielewski / Bioorg. Med. Chem. 17 (2009) 967–976
971
Figure 6. Progression of inhibitor design and observed increases in potency.
strongest reported were observed.18–23 Differential mod-
ification of the acyl groups and sidechain functionalities
provides a rich diversity of structures for further modi-
fication and optimization of dimerization inhibition of
HIV-1 PR.
solid was redissolved in 30 ml ethyl acetate and washed
with 10% citric acid (3·), 10% NaHCO3 (3·), and brine
(3·). Purification was performed by silica gel chroma-
tography (3:7 EtOAc/Hexane Rf = 0.4) providing
Bis-Boc-Cys-Phe-O-tBu as
a white solid (834 mg,
1
0.95 mmol, 95% yield). H NMR, 200 MHz (CDCl3) d
(ppm) = 1.38 (s, 18H), 1.46 (s, 18H), 2.86–3.22 (m,
8H), 4.69–4.80 (m, 8H), 5.48 (d, 2H, J = 4.6 Hz), 7.17–
7.28 (m, 10H), 7.65 (d, 2H, J = 4.0 Hz).
4. Experimental procedures
4.1. Materials
Reduction of the disulfide of 2 (400 mg, 0.46 mmol) was
performed in treatment with dithiothreitol (DTT,
284 mg, 1.84 mmol) and triethylamine (184 mg,
1.84 mmol) in degassed CHCl3 (20 ml) under an argon
atmosphere for 24 h at room temperature. After 24 h,
the chloroform was removed in vacuo and the reaction
mixture was redissolved in argon saturated ethyl acetate
(20 ml) and washed with degassed 10% citric acid (3·
10 ml) and degassed ddH2O (3· 10 ml). The ethyl ace-
tate was removed in vacuo, providing Boc-Cys-Phe-O-
t-But (2) as a white solid in quantitative yield without
further purification (400 mg, 0.46 mmol). Boc-Cys-Phe-
O-t-But was positive when reacted with Elman’s re-
agent24 (5,50 dithio-bis-(2-nitrobenzoic acid)) indicating
the presence of a free thiol. 1H NMR, 200 MHz (CDCl3)
d (ppm) = 1.38 (s, 9H), 1.46 (s, 9H), 2.86–3.22 (m, 4H),
4.69–4.80 (m, 4H), 5.48 (d, 1H, J = 4.6 Hz), 7.17–7.28
(m, 5H), 7.65 (d, 1H, J = 4.0 Hz). ESI-MS: calculated,
423.3; experimental 425.2 (M+H+).
4.1.1. General. HIV-1 PR was purchased from Bachem
Biosciences (affinity purified grade). Amino acids, cou-
pling reagents and resins were purchased from either
Advanced Chemtech or ChemImpex. All other reagents
were purchased from Aldrich and used without purifica-
tion. The inhibitors were purified to homogeneity by re-
verse phase HPLC under the following conditions—
column: Vydac reverse phase C8 column, flow rate:
8 ml/min using a linear gradient over 60 min. The eluant
was monitored at 214 nm and 254 nm. Peptides and
inhibitors were characterized by nuclear magnetic reso-
nance (NMR) using an Inova 300 MHz spectrometer.
Electrospray ionization mass spectrometry (ESI-MS)
using a Waters LCQ and matrix assisted laser desorp-
tion ionization mass spectrometry (MALDI-MS) using
a PerSeptive Voyager was performed to confirm iden-
tity. Fluorescence assays were performed on a Tecan
SpectrafluorPlus 96-well plate reader. The amino acid
content of inhibitors was analyzed by amino acid analy-
sis (AAA) using the Waters AccQ-Tag method without
modification, concentration was determined by compar-
ison to internal standards of 2-aminoisobutyric acid,
norleucine, and/or alanine.
4.3. Synthesis of (3)
Boc-Cys-Phe-O-t-Bu (2) (173 mg, 0.40 mmol), 10 equiv-
alents of 1,2-bis-(2-chloroethoxy)ethane (748 mg,
4.0 mmol), 4 equiv of K2CO3 (221 mg, 1.6 mmol) and
triethylbenzylammonium chloride (TEBACl) (18.2 mg,
0.08 mmol) were added to a round bottom flask under
argon. Degassed DMF (20 ml) under argon was added
and the resulting solution was additionally degassed
and replaced with argon (4 · 2 min). The reaction was
heated to 60 ꢁC. After 24 h the DMF was removed in
vacuo and the reaction was dried overnight under high
vacuum. The product was purified by silica gel chroma-
4.2. Synthesis of (2)
The chloride salt of +NH3-Phe(O-tBu) (540 mg,
2.1 mmol) was treated with Bis-Boc-Cystine (440.5 mg,
1.0 mmol), HOBt (321 mg, 2.1 mmol), HBTU (796 mg,
2.1 mmol), and DIEA (516 mg, 4.0 mmol) in CH2Cl2/
THF (1:1) (40 ml) at room temperature for 2 h. After
2 h, the solvent was removed in vacuo and the resulting

972
M. J. Bowman, J. Chmielewski / Bioorg. Med. Chem. 17 (2009) 967–976
tography (3:7 ethyl acetate/hexane (Rf = 0.14)). The iso-
lated product (3) (199 mg, 0.336 mmol, 84% yield) was a
clear oil. H NMR, 300 MHz (CDCl3) d (ppm) = 2.79
(d, 1 H, J = 0.9 Hz), 2.80 (d, 1H, J = 1.65 Hz), 2.82–
3.03 (m, 2H), 3.12 (d, 2H, J = 1.1 Hz), 3.13 (d, 2H,
J = 1.8 Hz), 3.63–3.82 (m, 10H), 4.32 (s, 1H), 4.74 (t,
1H, J = 4.5 Hz), 5.51 (d,1H, J = 3.2 Hz), 6.98 (s, 1H),
7.20–7.31 (m, 5H). ESI: Calculated 574.5; experimental:
575.0, 577 (M+H+); 597, 599 (M+Na+).
9.1% yield) was a yellow oil. 1H NMR, 300 MHz
(CD3CN) (ppm) = 0.79 (d/d, 6H J = 4.8 Hz,
d
1
J = 7.8 Hz), 1.43 (s, 9H), 1.46 (s, 18H), 1.50–1.60 (m,
3H), 2.52–2.90 (m, 12H), 3.57–3.76 (m, 10H), 4.20 (m,
3H), 4.59 (m, 2H), 5.77–5.97 (m, 3H), 7.12 (d, 1H,
J = 3.7 Hz), 7.25–7.40 (m, 10H), 7.97 (d, 1H,
J = 3.7 Hz). ESI: Calculated 1018; experimental: 1019
(M+H+); 1041 (M+Na+); 1057 (M+K+); Negative Ions:
1017 (MꢀHꢀ).
4.4. Synthesis of (4)
4.6. Synthesis of compounds (6–17)
Bis-Boc-cystine-OH (330 mg, 0.75 mmol) was treated
with Leu-Phe-NH2 (620 mg, 1.87 mmol), HOBt
(240 mg, 1.57 mmol), HBTU (595 mg, 1.57 mmol), and
DIEA (387 mg, 3.0 mmol) in CH2Cl2/THF (1:1)
(50 ml) at room temperature for 4 h. The CH2Cl2/THF
was removed in vacuo and the solid was dissolved in
75 ml ethyl acetate and washed with 10% citric acid
(3·), 10% NaHCO3 (3·), and brine (3·). Purification
by silica gel chromatography (95:5 CH2Cl2/MeOH
Rf = 0.34) yielded Bis-Boc-Cys-Leu-Phe-NH2 as a white
solid (605 mg, 0.56 mmol, 74% yield). 1H NMR,
Compound 5 (25 mg) was treated with 95% TFA (9 ml):
2.5% TIPS (500 ll): 2.5% H2O (500 ll) for 2 h. The
cleaving solvent was removed in vacuo, followed by pre-
cipitation with cold ether (3· 20 ml, ꢀ20 ꢁC, 4 h). The
resulting dried pellet was used without further
purification.
Deprotected 5 (10 mg, 0.013 mmol) was dissolved in dry
CH3CN followed by treatment with the desired acid
chloride (20 equiv, 0.26 mmol), HOBt (20 equiv,
0.26 mmol) and triethylamine (6.6 mg, 0.066 mmol).
After 2 h the reaction was quenched with 75 equivalents
ddH2O and allowed to stir for an additional two hours.
The solvent was removed in vacuo and the residue was
dissolved in 500 lL methanol. The sample was purified
to homogeneity using RP-HPLC to provide compounds
6–17, depending on the acid chloride used. Structures
were confirmed by NMR, amino acid analysis, and mass
spectrometry. Concentrations were determined from
amino acid analysis as compared to a norleucine and
alanine internal standard.
300 MHz (DMSO)
d
(ppm) = 0.83 (d/d, 12H
J = 6.0 Hz, J = 6.6 Hz), 1.39 (s, 18H), 1.50–1.60 (m,
3H), 2.82–2.93 (m, 4H), 3.01–3.05 (m, 4H), 4.18–4.28
(m, 4H), 4.44 (s, 2H), 7.209–7.379 (m, 10H).
Reduction of the disulfide of 4 (200 mg, 0.193 mmol)
was performed by dithiothreitol (DTT) (0.8 mmol,
125 mg) and triethylamine (0.8 mmol, 80 mg) in CHCl3
(50 ml). After 24 h the chloroform was removed in va-
cuo, and the reaction mixture was dissolved in argon
saturated ethyl acetate (50 ml) and washed with argon
saturated 10% citric acid (3· 15 ml) and ddH2O (3·
15 ml). The Boc-Cys-Leu-Phe-NH2 (4) was isolated in
quantitative yield (200 mg, 0.385 mmol) and was posi-
tive for when reacted with Elman’s reagent (5,50-
dithio-bis-(2-nitrobenzoic acid)). 1H NMR, 300 MHz
4.6.1. Compound (6). RP-HPLC: 5% CH3CN/95% H2O/
0.1% TFA to 75% CH3CN/25% H2O/0.1% TFA. Elu-
tion time: 43.2 min. MALDI-MS: calcd 874.4, found
897.8 (M+Na+); 913.8 (M+K+). AAA: Leu 1.0 (1);
Phe 2.1 (2). 1H NMR, 300 MHz (CD3CN)
d
(DMSO)
d
(ppm) = 0.83 (d/d, 6H J = 6.0 Hz,
(ppm) = 0.84 (d/d (J = 4.8, J = 7.8 Hz), 6H), 1.50 (bm,
2H), 1.56 (bm, 1H), 1.61( t (J = 7.2 Hz), 6H), 2.18 (t
(J = 4.2 Hz), 4H), 2.74 (t (J = 6.6 Hz), 4H), 2.94 (bm,
4H), 3.12 (m, 2H), 3.74 (m, 8H), 4.24 (d (J = 6.6 Hz),
2H), 4.29 (m, 1H), 4.59 (m, 2H), 5.85 (d (J = 7.2 Hz),
2H), 5.93 (d (J = 6.8 Hz), 1H), 6.55 (d (J = 7.3 Hz),
1H), 7.2 (d (J = 8.1 Hz), 2H), 7.33 (bm, 10H).
J = 6.6 Hz), 1.39 (s, 9H), 1.50–1.60 (m, 3H), 2.82–2.93
(m, 2H), 3.01–3.05 (m, 2H), 4.18–4.28 (m, 2H), 4.44 (s,
1H), 7.209–7.379 (m, 5H). ESI-MS: calculated, 543.5;
experimental 544.8 (M+H+).
4.5. Synthesis of (5)
Compound 3 (147 mg, 0.256 mmol), compound 4
(200 mg, 0.385 mmol), 4 equiv of K2CO3 (142 mg,
1.03 mmol) and triethylbenzylammonium chloride (TE-
BACl) (11.8 mg, 0.07 mmol) were added to a round bot-
tom flask under argon atmosphere. Degassed DMF
(25 ml) under argon was added and the resulting solu-
tion was additionally degassed and placed under an ar-
gon atmosphere (4 · 2 min). The reaction was heated
to 60 ꢁC. After 24 h the DMF was removed in vacuo
and the reaction was dried overnight under high vac-
uum. The product was purified by silica gel chromatog-
raphy (95:5 CH2Cl2/MeOH (Rf = 0.21)), followed by
RP-HPLC 50% CH3CN/0.1% TFA/50% H2O/0.1%
TFA to 90% CH3CN/0.1% TFA/10% H2O/0.1% TFA
using a linear gradient over 60 min. Retention time:
22.7 min. The isolated product 5 (23.6 mg, 0.023 mmol,
4.6.2. Compound (7). RP-HPLC: 5% CH3CN/95% H2O/
0.1% TFA to 75% CH3CN/25% H2O/0.1% TFA. Elu-
tion time: 46.5 min. MALDI-MS: calcd 902.4, found
925.9 (M+Na+); 941.9 (M+K+). AAA: Leu 1.0 (1);
Phe 2.2 (2). 1H NMR, 300 MHz (CD3CN)
d
(ppm) = 0.84 (d/d (J = 4.8, J = 7.8 Hz), 6H), 1.29 (t (
J = 5.4 Hz), 6H), 1.50 (bm, 2H), 1.56 (bm, 1H), 1.61
(m, 4H), 2.18 (m, 4H), 2.74 (t (J = 6.6 Hz), 4H), 2.94
(bm, 4H), 3.12 (m, 2H), 3.74 (m, 8H), 4.24 (d
(J = 6.6 Hz), 2H), 4.29 (m, 1H), 4.59 (m, 2H), 5.85 (d
(J = 7.2 Hz), 2H), 5.93 (d (J = 6.8 Hz), 1H), 6.55 (d
(J = 7.3 Hz), 1 H), 7.2 (d (J = 8.1 Hz), 2H), 7.33 (bm,
10H).
4.6.3. Compound (8). RP-HPLC: 15% CH3CN/85%
H2O/0.1% TFA to 85% CH3CN/15% H2O/0.1% TFA.

M. J. Bowman, J. Chmielewski / Bioorg. Med. Chem. 17 (2009) 967–976
973
Elution time: 43.2 min. MALDI-MS: calcd 930.5, found
953.9 (M+Na+); 969.9 (M+K+). AAA: Leu 1.0 (1); Phe
2.2 (2). 1H NMR, 300 MHz (CD3CN) d (ppm) = 0.84 (d/
d (J = 4.8, J = 7.8 Hz), 6H), 0.90 (t (J = 6.6 Hz), 6H),
1.29 (m, 4H), 1.50 (bm, 2H), 1.56 (bm, 1H), 1.61 (m,
4H), 2.18 (m, 4H), 2.74 (t (J = 6.6 Hz), 4H), 2.94 (bm,
4H), 3.12 (m, 2H), 3.74 (m, 8 H), 4.24 (d (J = 6.6 Hz),
2H), 4.29 (m, 1H), 4.59 (m, 2H), 5.85 (d (J = 7.2 Hz),
2H), 5.93 (d (J = 6.8 Hz), 1H), 6.55 (d (J = 7.3 Hz),
1H), 7.2 (d (J = 8.1 Hz), 2H), 7.33 (bm, 10H).
4.6.8. Compound (13). RP-HPLC: 30% CH3CN/70%
H2O/0.1% TFA to 100% CH3CN/0.1% TFA. Elution
time: 34.0 min. MALDI-MS: calcd 1038.6, found
1
1059.8 (M+Na+). AAA: Leu 1.0 (1); Phe 2.04 (2). H
NMR, 300 MHz (CD3CN)
d
(ppm) = 0.84 (d/d
(J = 4.8, J = 7.8 Hz), 6H), 1.5–1.9(m, 31H), 2.08 (m,
2H), 2.18 (bm, 4H), 2.74 (t (J = 6.6 Hz), 4H), 2.94
(bm, 4 H), 3.12 (m, 2H), 3.74 (m, 8H), 4.24 (d
(J = 6.6 Hz), 2H), 4.29 (m, 1H), 4.59 (m, 2H), 5.85 (d
(J = 7.2 Hz), 2H), 5.93 (d (J = 6.8 Hz), 1H), 6.55 (d
(J = 7.3 Hz), 1H), 7.2 (d (J = 8.1 Hz), 2H), 7.33 (bm,
10H).
4.6.4. Compound (9). RP-HPLC: 20% CH3CN/80%
H2O/0.1%T FA to 90% CH3CN/10% H2O/0.1% TFA.
Elution time: 42.5 min. MALDI-MS: calcd 958.5, found
981.7 (M+Na+); 997.8 (M+K+). AAA: Leu 1.0 (1); Phe
2.2 (2). 1H NMR, 300 MHz (CD3CN) d (ppm) = 0.84 (d/
d (J = 4.8, J = 7.8 Hz), 6 H), 0.90 (t (J = 6.6 Hz), 6H),
1.29 (m, 8H), 1.50 (bm, 2H), 1.56 (bm, 1H), 1.61 (m,
4H), 2.18 (m, 4H), 2.74 (t (J = 6.6 Hz), 4 H), 2.94 (bm,
4H), 3.12 (m, 2H), 3.74 (m, 8H), 4.24 (d (J = 6.6 Hz),
2H), 4.29 (m, 1H), 4.59 (m, 2H), 5.85 (d (J = 7.2 Hz),
2H), 5.93 (d (J = 6.8 Hz), 1H), 6.55 (d (J = 7.3 Hz),
1H), 7.2 (d (J = 8.1 Hz), 2H), 7.33 (bm,10H).
4.6.9. Compound (14). RP-HPLC: 30% CH3CN/70%
H2O/0.1% TFA to 100% CH3CN/0.1% TFA. Elution
time: 49.1 min. MALDI-MS: calcd 1094.6, found
1118.0 (M+Na+); 1133.9 (M+K+). AAA: Leu 1.0 (1);
Phe 1.8 (2). 1H NMR, 300 MHz (CD3CN)
d
(ppm) = 0.84 (d/d (J = 4.8, J = 7.8 Hz), 6H), 1.5–1.9(m,
39H), 2.08 (m, 2H), 2.18 (bm, 4H), 2.74 (t
(J = 6.6 Hz), 4H), 2.94 (bm, 4H), 3.12 (m, 2H), 3.74
(m, 8H), 4.24 (d (J = 6.6 Hz), 2H), 4.29 (m, 1H), 4.59
(m, 2H), 5.85 (d (J = 7.2 Hz), 2H), 5.93 (d
(J = 6.8 Hz), 1H), 6.55 (d (J = 7.3 Hz), 1H), 7.2 (d
(J = 8.1 Hz), 2H), 7.33 (bm, 10H).
4.6.5. Compound (10). RP-HPLC: 25% CH3CN/75%
H2O/0.1% TFA to 95% CH3CN/5% H2O/0.1% TFA.
Elution time: 44.2 min. MALDI-MS: calcd 986.5, found
1010.3 (M+Na+); 1026.4 (M+K+). AAA: Leu 1.0 (1);
4.6.10. Compound (15). RP-HPLC: 25% CH3CN/75%
H2O/0.1% TFA to 95% CH3CN/5% H2O/0.1% TFA.
Elution time: 34.2 min. MALDI-MS: calcd 998.4, found
1022.4 (M+Na+); 1038.4 (M+K+). AAA: Leu 1.0 (1);
Phe 2.0 (2). 1H NMR, 300 MHz (CD3CN)
d
(ppm) = 0.84 (d/d (J = 4.8, J = 7.8 Hz), 6H), 0.90 (t
(J = 6.6 Hz), 6H), 1.29 (m, 12H), 1.50 (bm, 2H), 1.56
(bm, 1H), 1.61 (m, 4H), 2.18 (m, 4H), 2.74 (t
(J = 6.6 Hz), 4H), 2.94 (bm, 4H), 3.12 (m, 2H), 3.74
(m, 8 H), 4.24 (d (J = 6.6 Hz), 2H), 4.29 (m, 1H), 4.59
(m, 2H), 5.85 (d (J = 7.2 Hz), 2H), 5.93 (d
(J = 6.8 Hz), 1H), 6.55 (d (J = 7.3 Hz), 1H), 7.2 (d
(J = 8.1 Hz), 2H), 7.33 (bm, 10H).
Phe 1.84 (2). 1H NMR, 300 MHz (CD3CN)
d
(ppm) = 0.84 (d/d (J = 4.8, J = 7.8 Hz), 6H), 1.50 (bm,
2H), 1.56 (bm, 1H), 2.74 (t (J = 6.6 Hz), 4H), 2.94
(bm, 4H), 3.12 (m, 2H), 3.74 (m, 8H), 4.24 (d
(J = 6.6 Hz), 2H), 4.29 (m, 1H), 4.59 (m, 2H), 4.83 (m,
4 H), 5.85 (d (J = 7.2 Hz), 2H), 5.93 (d (J = 6.8 Hz), 1
H), 6.55 (d (J = 7.3 Hz), 1H), 7.2 (d (J = 8.1 Hz), 2H),
7.33 (bm, 20H)
4.6.6. Compound (11). RP-HPLC: 30% CH3CN/70%
H2O/0.1% TFA to 80% CH3CN/20% H2O/0.1% TFA.
Elution time: 27.5 min. MALDI-MS: calcd 982.5, found
1005.4 (M+Na+); 1022.2 (M+K+). AAA: Leu 1.0 (1);
4.6.11. Compound (16). RP-HPLC: 25% CH3CN/75%
H2O/0.1% TFA to 95% CH3CN/5% H2O/0.1% TFA.
Elution time: 37.0 min. MALDI-MS: calcd 1026.5,
found 1050.4 (M+Na+); 1066.7 (M+K+). AAA: Leu
Phe 1.9 (2). 1H NMR, 300 MHz (CD3CN)
d
ppm) = 0.84 (d/d (J = 4.8, J = 7.8 Hz), 6H), 1.5–1.9 (m,
23H), 2.08 (m, 2H), 2.74 (t (J = 6.6 Hz), 4H), 2.94
(bm, 4H), 3.12 (m, 2H), 3.74 (m, 8 H), 4.24 (d
(J = 6.6 Hz), 2H), 4.29 (m, 1H), 4.59 (m, 2H), 5.85 (d
(J = 7.2 Hz), 2H), 5.93 (d (J = 6.8 Hz), 1H), 6.55 (d
(J = 7.3 Hz), 1H), 7.2 (d (J = 8.1 Hz), 2H), 7.33 (bm,
10H).
1
1.0 (1); Phe 1.7 (2). H NMR, 300 MHz (CD3CN) d
(ppm) = 0.84 (d/d (J = 4.8, J = 7.8 Hz), 6H), 1.50 (bm,
2H), 1.56 (bm, 1H), 2.74 (t (J = 6.6 Hz), 4H), 2.94
(bm, 4H), 3.12 (m, 2H), 3.74 (m, 8H), 3.92 (m, 4H),
4.12 (m, 4H), 4.24 (d (J = 6.6 Hz), 2H), 4.29 (m, 1 H),
4.59 (m, 2H), 4.83 (m, 4H), 5.85 (d (J = 7.2 Hz), 2H),
5.93 (d (J = 6.8 Hz), 1H), 6.55 (d (J = 7.3 Hz), 1H), 7.2
(d (J = 8.1 Hz), 2H), 7.33 (bm, 20H).
4.6.7. Compound (12). RP-HPLC: 30% CH3CN/70%
H2O/0.1% TFA to 100% CH3CN/0.1% TFA. Elution
time: 29.4 min. MALDI-MS: calcd 1010.5, found
1060.0 (M+K+). AAA: Leu 1.0 (1); Phe 1.6 (2). 1H
4.6.12. Compound (17). RP-HPLC: 60 min. Linear gradi-
ent from 25% CH3CN/75% H2O/0.1% TFA to 95%
CH3CN/5% H2O/0.1% TFA. Elution time: 34.2 min.
MALDI-MS: calcd 1030.4, found 1057.0 (M+Na+).
AAA: Leu 1.0 (1); Phe 1.6 (2). 1H NMR, 300 MHz
(CD3CN) d (ppm) = 0.84 (d/d (J = 4.8, J = 7.8 Hz),
6H), 1.50 (bm, 2H), 1.56 (bm, 1H), 2.74 (t
(J = 6.6 Hz), 4H), 2.94 (bm, 4H), 3.12 (m, 2H), 3.74
(m, 8H), 4.24 (d (J = 6.6 Hz), 2H), 4.29 (m, 1H), 4.41
(s, 4H), 4.59 (m, 2 H), 5.85 (d (J = 7.2 Hz), 2H), 5.93
NMR, 300 MHz (CD3CN)
d
(ppm) = 0.84 (d/d
(J = 4.8, J = 7.8 Hz), 6H), 1.5–1.9 (m, 27H), 2.08 (m,
2H), 2.18 (t (J = 4.2 Hz), 4H), 2.74 (t (J = 6.6 Hz),
4H), 2.94 (bm, 4H), 3.12 (m, 2H), 3.74 (m, 8H), 4.24
(d (J = 6.6 Hz), 2H), 4.29 (m, 1H), 4.59 (m, 2H), 5.85
(d (J = 7.2 Hz), 2H), 5.93 (d (J = 6.8 Hz), 1H), 6.55 (d
(J = 7.3 Hz), 1H), 7.2 (d (J = 8.1 Hz), 2H), 7.33 (bm,
10H).

974
M. J. Bowman, J. Chmielewski / Bioorg. Med. Chem. 17 (2009) 967–976
1
(d (J = 6.8 Hz), 1 H), 6.55 (d (J = 7.3 Hz), 1H), 7.2 (d
(J = 8.1 Hz), 2H), 7.33 (bm, 20H)
free thiol. H NMR, 300MHz (DMSO) d (ppm) = 0.83
(m, 6H), 1.39 (s, 9H), 1.50–1.60 (m, 3H), 2.82–2.93 (m,
2H), 3.01–3.05 (m, 2H), 4.18–4.28 (m, 2H), 4.44 (s,
1H), 7.21–7.38 (m, 5H), 7.64 (d, 1H, J = 4.1 Hz), 8.08
(d, 1H, J = 3.8 Hz). ESI-MS: calculated, 543.5; experi-
mental 544.4 (M+H+).
4.7. Synthesis of (4 (Ile))
Synthesis of peptide Ile-Phe-NH2 was performed in
solution utilizing a Boc protection strategy. To a stirred
solution of Boc-isoleucine (0.75 mmol, 180 mg) in
CH2Cl2 (30 ml), was added HOBt (0.83 mmol,
112 mg), and HBTU (0.83 mmol, 315 mg) at room tem-
perature. After 10 min the hydrochloride salt of phenyl-
alanine amide (0.83 mmol, 163 mg) was added with
diisopropylethylamine (DIEA) (1.1 mmol, 192 mg) in
5 ml DMF. The reaction was stirred for 3 h at room
temperature. The CH2Cl2 was removed in vacuo, the
remaining DMF solution was diluted in 20 ml ethyl ace-
tate and washed with LiCl (sat) (2·), 10% citric acid
(3·), 10% NaHCO3 (3·), and brine (3·). The solvent
was removed in vacuo and the resulting solid was fur-
ther purified by silica gel chromatography (1:19
MeOH/CH2Cl2 Rf = 0.35), to yield Boc-Ile-Phe-NH2 as
a white crystalline solid (232 mg, 80% yield). 1H
NMR, 300 MHz (CDCl3) d(ppm) 0.89 (m, 6H), 1.38
(s, 9H), 1.50.–1.60 (m, 3H), 3.12 (d, 2H, J = 6.6 Hz),
4.04 (t, 1H, J = 5.7 Hz), 4.74 (m, 2H), 5.33 (bs, 1H),
6.34 (bs, 1H), 6.42 (bs, 1H), 7.20–7.34 (m, 5H).
4.8. Synthesis of (5 (Ile))
Compound 3 (81 mg, 0.150 mmol), Boc-Cys-Ile-Phe-
NH2 (285 mg, 0.60 mmol), 4 equiv of K2CO3 (82 mg,
0.57 mmol), 1 equiv of NaBr (15 mg, 0.15 mmol) and
triethylbenzylammonium chloride (TEBACl) (13.5 mg,
0.06 mmol) were added to a round bottom flask under
an argon atmosphere. Degassed DMF (15 ml) under ar-
gon was added and the resulting solution was addition-
ally degassed and placed under an argon atmosphere
(4 · 2 min). The reaction was heated to 60 ꢁC. After
24 h the DMF was removed in vacuo and the reaction
was dried overnight under high vacuum. The desired
product was purified by silica gel chromatography
(95:5 CH2Cl2/MeOH (Rf = 0.21)). Followed by RP-
HPLC 50% CH3CN/0.1% TFA/50% H2O/0.1% TFA
to 90% CH3CN/0.1% TFA/10% H2O/0.1% TFA using
a
linear gradient over 60 min. Retention time:
26.7 min. The isolated product, (80 mg, 0.079 mmol,
52% yield) was white solid. 1H NMR, 300 MHz
(CD3CN) d (ppm) = 0.82 (m, 6H), 1.43 (s, 9H), 1.46 (s,
18H), 1.50–1.60 (m, 3H), 2.52–2.90 (m, 12H), 3.57–
3.76 (m, 10H), 4.20 (m, 3H), 4.59 (m, 2H), 5.77–5.97
(m, 3H), 7.12 (d, 1H, J = 3.7 Hz), 7.25–7.40 (m, 10H),
7.97 (d, 1H, J = 3.7 Hz). ESI-MS: Calculated 1018;
experimental: 1019 (M+H+); 1041 (M+Na+); 1057
(M+K+); Negative Ions: 1017 (MꢀHꢀ).
The Boc-Ile-Phe-NH2 (232 mg, 0.6 mmol) was treated
with 3 ml trifluoroacetic acid (TFA)/CH2Cl2 (1:1) for
30 min at room temperature. The cleaving solvent was
removed in vacuo, followed by precipitation with cold
ether (3· 20 ml, ꢀ20 ꢁC, 4 h). NH2-Ile-Phe-NH2
(170 mg, 0.6 mmol) was collected by centrifugation
and used without further purification.
Bis-Boc-cystine-OH (199 mg, 0.45 mmol) was treated
with NH2-Ile-Phe-NH2 (315 mg, 0.95 mmol), HOBt
(128 mg, 0.95 mmol), HBTU (360 mg, 0.95 mmol),
and DIEA (239 mg, 1.85 mmol) in DMF (7 ml) at room
temperature for 4 h. The DMF was removed in vacuo
and the solid was dissolved in 25 ml ethyl acetate and
washed with 10% citric acid (3·), 10% NaHCO3 (3·),
and brine (3·). Purification by silica gel chromatogra-
phy (95:5 CH2Cl2/MeOH Rf = 0.32) yielded Bis-Boc-
Cys-Ile-Phe-NH2 as a white solid (400 mg, 0.37 mmol,
4.9. Synthesis of compounds (18–19)
5 (Ile) (80 mg) was treated with 95% TFA (9 ml): 2.5%
TIPS (500 ll): 2.5% H2O (500 ll) for 2 h. The cleaving
solvent was removed in vacuo, followed by precipitation
with cold ether (3· 20 ml, ꢀ20 ꢁC, 4 h). Residual ether
was removed in vacuo and the resulting solid was used
without further purification.
The deprotected compound (10 mg, 0.013 mmol) was
dissolved in dry CH3CN followed by treatment with
the desired acid chloride (20 equiv, 0.26 mmol), HOBt
(20 equiv, 0.26 equiv) and triethylamine (6.62 mg,
0.0656 mmol). After 2 h the reaction was quenched with
75 equivalents ddH2O and allowed to stir for an addi-
tional two hours. The solvent was removed in vacuo
and the residues were dissolved in 500 lL DMSO. The
samples were then purified to homogeneity using RP-
HPLC to provide compounds 18–19, depending on the
acid chloride used. The structures were confirmed by
NMR, amino acid analysis, and mass spectrometry.
Concentrations were determined from amino acid anal-
ysis as compared to a norleucine and alanine internal
standard.
82% yield). 1H NMR, 300MHz (DMSO)
d
(ppm) = 0.83 (m, 12H), 1.39 (s, 18H), 1.50–1.60 (m,
3H), 2.82–2.93 (m, 4H), 3.01–3.05 (m, 4H), 4.18–4.28
(m, 4H), 4.44 (s, 2H), 7.21–7.38 (m, 10H), 7.64 (d, 1H,
J = 4.1 Hz), 8.08 (d, 1H, J = 3.8 Hz).
Reduction of the disulfide of the desired tripeptide
(250 mg, 0.23 mmol) was performed by treatment with
dithiothreitol (DTT) (0.92 mmol, 144 mg) and triethyl-
amine (0.92 mmol, 92 mg) in CHCl3. After 24 h the
chloroform was removed in vacuo and the reaction mix-
ture was dissolved in argon saturated ethyl acetate
(30 ml) and washed with argon saturated 10% citric acid
(3· 15 ml) and ddH2O (3· 15 ml). The 4 (Ile) was iso-
lated in quantitative yield (250 mg, 0.23 mmol) without
further purification. 4 (Ile) demonstrated a positive re-
sult when reacted with Elman’s reagent (5,50-dithio-
bis-(2-nitrobenzoic acid)) indicating the presence of a
4.9.1. Compound (18). RP-HPLC: 25% CH3CN/75%
H2O/0.1% TFA to 95% CH3CN/5% H2O/0.1% TFA.
Elution time: 34.5 min. MALDI-MS: calcd 986.5, found

M. J. Bowman, J. Chmielewski / Bioorg. Med. Chem. 17 (2009) 967–976
975
1010.3 (M+Na+); AAA: Ile 1.0 (1); Phe 2.0 (2). 1H
NMR, 300 MHz (CD3CN) d (ppm) = 0.84 (m, 6H),
0.90 (t (J = 6.6 Hz), 6H), 1.29 (m, 12 H), 1.50 (bm,
2H), 1.56 (bm, 1H), 1.61 (m, 4H), 2.18 (m, 4H), 2.74 (t
(J = 6.6 Hz), 4H), 2.94 (bm, 4H), 3.12 (m, 2H), 3.74
(m, 8 H), 4.24 (d (J = 6.6 Hz), 2H), 4.29 (m, 1H), 4.59
(m, 2H), 5.85 (d (J = 7.2 Hz), 2H), 5.93 (d
(J = 6.8 Hz), 1H), 6.55 (d (J = 7.3 Hz), 1H), 7.2 (d
(J = 8.1 Hz), 2H), 7.33 (bm, 10H).
solutions a constant concentration of a known active
site competitive inhibitor (acetyl-pepstatin, 7.2, 14.4,
43.2 nM) with varying concentrations of compound 6
(0–10 lM) in DMSO for one hour at room temperature
in 1.2 ml polypropylene tubes (Costar). These solutions
(3· 60 ll) were added to three different 40 ll aliquots of
a 30 lM substrate solution10(10% DMSO and 90% as-
say buffer) in a polypropylene 96 well plate (Griegner).
The final concentration of DMSO was maintained at
14%. The change in fluorescence intensity at 465 nm
(ex: 360 nm) was immediately measured upon the addi-
tion of the protease to the substrate solution over a per-
iod of 14 min.
4.9.2. Compound (19). RP-HPLC: 30% CH3CN/70%
H2O/0.1% TFA to 100% CH3CN/0% H2O/0.1% TFA.
Elution time: 42.0 min. ESI-MS: calcd 1094.6, found
1118.3 (M+Na+); 1094.4 (MꢀHꢀ) Neg Ions. AAA: Ile
1
1.0 (1); Phe 2.0 (2). H NMR, 300 MHz (CD3CN) d
(ppm) = 0.84 (m, 6H), 1.50–1.9 (m, 36H), 2.18 (m,
4H), 2.74 (t (J = 6.6 Hz), 4H), 2.94 (bm, 4H), 3.12 (m,
2H), 3.74 (m, 8H), 4.24 (d (J = 6.6 Hz), 2H), 4.29 (m,
1H), 4.59 (m, 2H), 5.85 (d (J = 7.2 Hz), 2H), 5.93 (d
(J = 6.8 Hz), 1H), 6.55 (d (J = 7.3 Hz), 1H), 7.2 (d
(J = 8.1 Hz), 2H), 7.33 (bm, 10H).
Acknowledgment
This research was supported by the National Institutes
of Health (GM52379).
References and notes
4.10. Zhang–Poorman kinetic assay
1. Miller, M.; Schneider, J.; Sathyanarayana, B. K.; Toth,
M. V.; Marshall, G. R.; Clawson, L.; Selk, L.; Kent, S. B.;
Wlodawer, A. Science 1989, 246, 1149.
2. Davies, D. R. Annu. Rev. Biophys. Biophys. Chem. 1990,
19, 189.
3. Pearl, L. H.; Taylor, W. R. Nature 1987, 328, 482.
4. Deeks, S. G.; Smith, M.; Holodniy, M.; Kahn, J. O.
JAMA 1997, 277, 145.
5. Gustchina, A.; Weber, I. T. FEBS Lett. 1990, 269, 269.
6. Todd, M. J.; Semo, N.; Freire, E. J. Mol. Biol. 1998, 283,
475.
7. Shultz, M. D.; Bowman, M. J.; Ham, Y. W.; Zhao, X.;
Tora, G.; Chmielewski, J. Angew. Chem. Int. Ed. Engl.
2000, 39, 2710.
8. Bowman, M. J.; Chmielewski, J. Biopolymers 2002, 66,
126.
9. Wlodawer, A.; Miller, M.; Jaskolski, M.; Sathyanarayana,
B. K.; Baldwin, E.; Weber, I. T.; Selk, L. M.; Clawson, L.;
Schneider, J.; Kent, S. B. Science 1989, 245, 616.
10. Toth, M. V.; Marshall, G. R. Int. J. Pept. Protein Res.
1990, 36, 544.
11. Zhang, Z. Y.; Poorman, R. A.; Maggiora, L. L.;
Heinrikson, R. L.; Kezdy, F. J. J. Biol. Chem. 1991, 266,
15591.
To perform this assay, varying concentrations of HIV-1
PR were treated with constant concentrations of inhibi-
tor and substrate while monitoring the initial velocity
(V) of the proteolytic cleavage of the fluorogenic sub-
strate. For Zhang–Poorman kinetic assay, 180 ll of
varying concentrations of HIV-1 protease solution (5–
45 nM) in assay buffer (20 mM phosphate, 1 mM
DTT, 10% glycerol, and 0.1% CHAPS at pH 5.5) was
incubated with 36 ll of inhibitor solution at a concentra-
tion equal to the IC50 and 0.75 · IC50 in DMSO for one
hour at room temperature in 1.2 ml polypropylene tubes
(Costar). This solution (3 · 60 ll) was added to three
different 40 ll aliquots of a 62.5 lM substrate solution
(10% DMSO and 90% assay buffer) in a polypropylene
96-well plate (Griegner). The final concentration of
DMSO was maintained at 14%. The change in fluores-
cence intensity at 465 nm (ex: 360 nm) was immediately
measured upon the addition of the protease to the sub-
strate solution over a period of 14 min. Initial velocity
(V) was calculated from D [hydrolyzed substrate]/D time
in seconds determined when <5% of substrate was
consumed.
12. Loeb, D. D.; Hutchison, C. A., 3rd; Edgell, M. H.;
Farmerie, W. G.; Swanstrom, R. J. Virol. 1989, 63,
111.
13. Wlodawer, A.; Erickson, J. W. Annu. Rev. Biochem. 1993,
62, 543.
4.11. Cross-competition kinetic assay
For cross-competition assay, solutions of 180 lL of
HIV-1 protease in buffer (20 mM phosphate, 1 mM
DTT, 1 mM EDTA, 20% glycerol, and 0.1% CHAPS
at pH 5.52; final concentration 10 nM protease) were
incubated with 36 ll of each inhibitor solutions varying
in the concentration of a known active site competitive
inhibitor acetyl-pepstatin (0–43.2 nM) with constant
concentrations of compound 7 (0, 21.9, 29.9, 39 lM)
in DMSO for one hour at room temperature in 1.2 ml
polypropylene tubes (Costar). Alternatively, solutions
of 180 lL of HIV-1 protease in buffer (20 mM phos-
phate, 1 mM DTT, 1 mM EDTA, 20% glycerol, and
0.1% CHAPS at pH 5.52; final concentration 10 nM
protease) were incubated with 36 ll of each inhibitor
14. Richards, A. D.; Roberts, R.; Dunn, B. M.; Graves, M.
C.; Kay, J. FEBS Lett. 1989, 247, 113.
15. Fitzgerald, P. M.; McKeever, B. M.; VanMiddlesworth, J.
F.; Springer, J. P.; Heimbach, J. C.; Leu, C. T.; Herber,
W. K.; Dixon, R. A.; Darke, P. L. J. Biol. Chem. 1990,
265, 14209.
16. Todd, M. J.; Freire, E. Proteins 1999, 36, 147.
17. Shultz, M. D.; Ham, Y. W.; Lee, S. G.; Davis, D. A.;
Brown, C.; Chmielewski, J. J. Am. Chem. Soc. 2004, 126,
9886.
18. Bowman, M. J.; Byrne, S.; Chmielewski, J. Chem. Biol.
2005, 12, 439.
19. Hwang, Y. S.; Chmielewski, J. J. Med. Chem. 2005, 48,
2239.
20. Lee, S. G.; Chmielewski, J. Chem. Biol. 2006, 13, 421.

976
M. J. Bowman, J. Chmielewski / Bioorg. Med. Chem. 17 (2009) 967–976
21. Merabet, N.; Dumond, J.; Collinet, B.; Van Baeling-
hem, L.; Boggetto, N.; Ongeri, S.; Ressad, F.;
Reboud-Ravaux, M.; Sicsic, S. J. Med. Chem. 2004,
47, 6392.
22. Breccia, P.; Boggetto, N.; Perez-Fernandez, R.; Van Gool,
M.; Takahashi, M.; Rene, L.; Prados, P.; Badet, B.;
Reboud-Ravaux, M.; de Mendoza, J. J. Med. Chem. 2003,
46, 5196.
23. Dumond, J.; Boggetto, N.; Schramm, H. J.; Schramm, W.;
Takahashi, M.; Reboud-Ravaux, M. Biochem. Pharmacol.
2003, 65, 1097.
24. Ellman, G. L. Arch. Biochem. Biophy. 1959, 82, 70.