Impact of stereochemistry on ligand binding: X-ray crystallographic analysis of an epoxide-based HIV protease inhibitor

Article abstract of DOI:10.1021/ml500092e

A new pseudopeptide epoxide inhibitor, designed for irreversible binding to HIV protease (HIV-PR), has been synthesized and characterized in solution and in the solid state. However, the crystal structure of the complex obtained by inhibitor-enzyme cocrys

Full text of DOI:10.1021/ml500092e

Letter
pubs.acs.org/acsmedchemlett
Impact of Stereochemistry on Ligand Binding: X‑ray Crystallographic
Analysis of an Epoxide-Based HIV Protease Inhibitor
,†
†
†
†
†
Fabio Benedetti,* Federico Berti, Pietro Campaner, Lidia Fanfoni, Nicola Demitri,
Folasade M. Olajuyigbe,†,‡ Matteo De March, and Silvano Geremia*
†
,†
†Department of Chemical and Pharmaceutical Sciences, Centre of Excellence in Biocrystallography, University of Trieste, Via
Giorgeri 1, 34127 Trieste, Italy
‡
Department of Biochemistry, Federal University of Technology, P.M.B. 704, Akure 340001, Ondo State, Nigeria
*
S Supporting Information
ABSTRACT: A new pseudopeptide epoxide inhibitor, designed for irreversible
binding to HIV protease (HIV-PR), has been synthesized and characterized in
solution and in the solid state. However, the crystal structure of the complex
obtained by inhibitor−enzyme cocrystallization revealed that a minor isomer, with
inverted configuration of the epoxide carbons, has been selected by HIV-PR during
crystallization. The structural characterization of the well-ordered pseudopeptide,
inserted in the catalytic channel with its epoxide group intact, provides deeper
insights into inhibitor binding and HIV-PR stereoselectivity, which aids
development of future epoxide-based HIV inhibitors.
KEYWORDS: Epoxide inhibitor, HIV protease, irreversible inhibition, stereochemistry, structure-based drug design
n the treatment of AIDS patients, the so-called highly active
antiretroviral therapy (HAART) is based primarily on a
inhibitor to the Asp25 residues are the low nucleophilicity of
aspartate that needs a strongly electrophilic alkylating counter-
part in the ligand molecule and the geometrical issues related to
the topology of the active site.
I
16
cocktail of inhibitors of the reverse transcriptase and aspartic
protease of HIV (HIV-PR). HAART has proven to dramatically
improve the quality of life of AIDS patients even for critically
The three-dimensional structures of wild-type HIV-PR and
1
HIV-infected subjects. However, viral strains that have mutated
several drug-resistant mutants bound to various inhibitors have
enzymes less sensitive to the inhibitors may survive the
treatment. Viral drug resistance, resulting from mutations of
the genes that code for the target enzymes, limits the
effectiveness of HAART in an increasing number of patients.
Ten first and second generation reversible HIV protease
17,18
been obtained by X-ray crystallography.
Structural informa-
tion about protease−inhibitor complexes is of paramount
importance to study specific HIV-PR/inhibitor interactions
and to define the binding efficiency and mechanism of selective
drugs. Structure-based drug design may then lead to new,
2
,3
inhibitors (PI) have been approved by the FDA to date. All
these drugs do not covalently bind to the protease and may
undergo loss of activity toward resistant PR mutants in different
19
improved inhibitors with increased potency or to the discovery
20
of new lead compounds. Recently, we have used high-
resolution X-ray crystallography as a powerful tool to identify
3
−6
ways.
21
the most potent HIV-PR inhibitor in an epimeric mixture.
Irreversible inhibitors could represent a valuable tool in the
development of new generation drugs capable of overcoming
resistance, as covalent binding may be less sensitive to mutations
that decrease the enzyme’s affinity for the inhibitor. The catalytic
Asp25 residues of HIV-PR must be conserved in all functional
Here, we report the synthesis and structural characterization of
two new epimeric inhibitors (1a and 1b) containing an
epoxyalcohol moiety within a Phe-Phe pseudodipeptide unit.
7
mutants and thus are logical targets for irreversible inhibitors.
However, irreversible binding in HIV-PR inhibition represents a
relatively new chemical expedient. Irreversible inhibition of HIV-
PR by EPNP [1,2-epoxy-3-(p-nitrophenoxy)propane], a specific
8
,9
10
inhibitor of aspartyl residues, has been reported, and
11
covalent binding has been confirmed by mass spectrometry.
Other epoxide inhibitors have been synthesized showing milli- to
Received: March 2, 2014
Accepted: July 14, 2014
Published: July 14, 2014
12−15
nanomolar inhibition.
The main problems in targeting the
©
2014 American Chemical Society
968
dx.doi.org/10.1021/ml500092e | ACS Med. Chem. Lett. 2014, 5, 968−972

ACS Medicinal Chemistry Letters
Letter
On the basis of our previous work on trans-epoxide
1
6
inhibitors, compound 1a was designed to target the S −S ′
1
3
subsites of the enzyme’s catalytic site. The inhibitor’s P1−P1′
core mimics a Phe-Phe dipeptide; the R configuration of the OH
group has been found optimal in similar compounds, and the
22
stereochemistry of the epoxide reflects synthetic accessibility
(vide infra). Proper placement of the molecule in the active site
should allow nucleophilic attack on the epoxide by one of the
catalytically active aspartyl residues, resulting in ring opening and
covalent binding of the inhibitor. Compound 1a was synthesized
23
from the known enone 2, derived from phenylalanine, as
shown in Scheme 1.
a
Scheme 1. Synthesis of Inhibitor 1a
Figure 1. (A) Comparison between the crystal structures of pure 1a
(left) and 1b as found in the complex with HIV-PR. (B) HIV-PR/1b
complex electron density 2F − F (1.0 sigma level) in the active site.
o
c
The epoxide ring is highlighted in the yellow circle.
this complex is R,R, as in 1b, and not S,S, as in 1a (Figure 1).
There is no reasonable mechanism by which 1a can be converted
into 1b; we must thus conclude that a small amount of 1b was
present in the sample, resulting from anti epoxidation of the
alkene 4 (Scheme 1) and was selected by the protease.
Even if the stereochemistry of the HIV-PR bound inhibitor is
not as expected, the analysis of this structure, the first for an
epoxide-based inhibitor, allows for a better understanding of the
binding mode and offers suggestions for improving the
inhibitor’s design.
a
Reagents and conditions: (a) L-selectride, THF, −78 °C,
diastereoselectivity 9:1; 77% after chromatographic separation; (b)
TFA, dichloromethane, 25 °C; (c) POA-Val-OSU, NEt , dichloro-
methane, 96%; (d) mCPBA, dichloromethane, 25 °C, 62%, d.e. > 90%.
3
24
L-Selectride reduction of 2 gave a 9:1 mixture of syn
aminoalcohol 3 and its anti isomer, from which pure 3 was readily
obtained by column chromatography. Boc-deprotection and
coupling with the N-succinimidoyl ester of phenoxyacetylvaline
Figure 1B shows that, despite the short length of the inhibitor,
which can fill asymmetrically only four of the six subsites present
in the catalytic channel (corresponding to the recognition of the
six amino acid residues of natural substrates), the inhibitor is
correctly located with the epoxide group very close to the
catalytic center. The analysis of the crystal structure further
1
6
25
(
POA-Val-OSU) was then followed by syn epoxidation of the
allylic alcohol 4 giving inhibitor 1a, in which the phenoxyacetyl
group and valine side chain are directed at the enzyme’s S3′ and
S2′ subsites, respectively. No trace of the (R,R)-epoxide 1b,
resulting from anti epoxidation of the allylic alcohol, was detected
at this stage (see Supporting Information), in agreement with
previous findings in the epoxidation of structurally related
compounds, and the structure of 1a has been confirmed by X-ray
crystallography (Figure 1A).
26
shows the flap water bridging contacts between the inhibitor’s
core and the flap tips (Ile50 residues) within the S and S ′
1
1
subsites (Figure 2A).
The inhibitor leaves two empty subsites (S and S ) filled with
2
3
water molecules (Figure 2B). The hydroxy group present in the
inhibitor’s core is hydrogen bonded to the catalytic Asp25
26
In a standard fluorimetric assay, epoxide 1a inhibited wild-type
HIV-PR with a 670 nM IC50 (see Supporting Information).
However, when 1a was incubated with the protease, no evidence
was found for time-dependent, irreversible inhibition, nor could
any covalently modified protease be detected by electrospray
ionization mass spectrometry (ESI-MS). Thus, to obtain
information on the mode of binding of 1a to the protease and
on the mechanism of inhibition, we solved the crystal structure of
the epoxide inhibitor bound to HIV-PR.
residues, a common feature in HIV-PR complexes, contribu-
ting to successfully placing the inhibitor in the active site, with the
epoxide ring close to the catalytic center. Figure 3 summarizes
the hydrogen bond pattern and hydrophobic interactions in the
HIV-PR/1b complex. Figure 2B clearly shows the water-filled S2
and S subsites and suggests that hydrophobic substituents on
3
the aromatic ring adjacent to the epoxide might displace the
water molecules and improve the inhibitor’s binding affinity by
making favorable contacts with the protein.
Orthorhombic single crystals of the protease−inhibitor
complex diffract at 1.45 Å (see Supporting Information), and
the electron density maps show a single, ordered inhibitor
molecule in the active site of HIV-PR (Figure 1B). As expected
from the inhibition data, the crystallographic analysis reveals the
intact epoxide ring in the complex with the protease. However,
much to our surprise, the configuration of the epoxide carbons in
As to the selection of the R,R 1b epoxide in the crystals, which
appears to be preferred with respect to the main product S,S 1a,
an overlay of the crystallographic structure of the HIV-PR/1b
complex and of a docked model of the HIV-PR/1a complex is
reported in Figure 4.
It can be seen that the epoxide oxygen is flipped down due to
the reversed configuration of the oxirane carbons; moreover, the
9
69
dx.doi.org/10.1021/ml500092e | ACS Med. Chem. Lett. 2014, 5, 968−972

ACS Medicinal Chemistry Letters
Letter
Figure 4. Overlay of the experimental structure of the HIV-PR/1b
complex and of the docked model for the HIV-PR/1a complex (in
green).
when added at very low ratio with respect to the other
20
competitive inhibitor.
Epoxides usually show high reactivity toward nucleophiles,
leading to ring opening. The intact epoxide ring observed in the
X-ray structure of HIV-PR/1b was therefore surprising.
The catalytic aspartic 25(B) carboxylate is involved in
hydrogen bonding with the hydroxyl group of the inhibitor
and is actually rather far from the epoxide carbon atoms.
Conversely, the carboxyl group of the other aspartic acid 25(A) is
in very close contact with one of the carbons of the oxirane
(Figure 5).
Figure 2. (A) Flap water (red sphere) in the S −S ′ pockets of the
1
1
catalytic site. (B) Water molecules (red spheres) filling S −S pockets of
2
3
the catalytic site.
Figure 5. Contact between Asp 25(A) and the epoxide ring. The van der
Waals surfaces of the carboxyl oxygen and of the oxirane carbon are
shown in red and dark gray, respectively.
The distance between the two atoms is only 3.09 Å,
comparable to the sum of their van der Waals radii; the
carboxylate oxygen atom is well aligned with the epoxide ring,
and the angle with its carbon−carbon bond is 95°. This value is
close to the optimal trajectory calculated for the ring opening of
an epoxide by oxygen nucleophiles and in particular for the
calculated path of epoxide opening by Asp25 in HIV protease
2
7,28
Figure 3. LIGPLOT of hydrophobic and polar contacts between 1b and
amino acid residues in HIV-PR.
models.
The epoxide ring opening by HIV protease has been
computationally studied by Mavri, who has suggested that the
deptrotonated aspartate should attack the oxirane upon proton
transfer to the leaving oxygen by the other, protonated aspartic
benzyl group linked to the epoxide is unable to fit properly inside
the S1 subsite. This leads to the loss of several hydrophobic
contacts found in the HIV-PR/1b complex and supports the
hypothesis that 1b might have been selected thanks to a higher
affinity. The same phenomenon was earlier observed in
competitive crystallization experiments where the more potent
inhibitor occupies the active site of HIV-PR in the crystal, even
29
residue. This proton transfer is clearly not possible in our case,
as the epoxide oxygen is placed upside and far from the aspartyl
side chains. However, two water molecules are found at
hydrogen bond distance from the oxygen (Figure 3) and could
provide protons to assist the epoxide opening. The reason for the
lack of reactivity remains therefore puzzling. This evidence could
9
70
dx.doi.org/10.1021/ml500092e | ACS Med. Chem. Lett. 2014, 5, 968−972

ACS Medicinal Chemistry Letters
Letter
suggest that Asp25(A) is actually protonated in the catalytic site
and not available as a nucleophile.
In conclusion, the reported structural characterization of 1a
and HIV-PR/1b complex clearly indicated that the configuration
of the epoxide carbons plays a crucial role in the binding affinity
of the inhibitor and in the alignment of the epoxide group within
the active site of HIV-PR.
Rhee, S. Y.; Liu, T. F.; Pillay, D.; Shafer, R. W. Drug resistance mutations
for surveillance of transmitted HIV-1 drug-resistance: 2009 update.
PLoS One 2009, 4, e4724.
(5) Poveda, E.; De Mendoza, E.; Martin-Carbonero, L.; Corral, A. I.;
Briz, V. N.; Gonzalez-Lahoz, J.; Soriano, V. Prevalence of darunavir
resistance mutations in HIV-1-infected patients failing other protease
inhibitors. J. Antimicrob. Chemother. 2007, 60, 885−888.
(6) Freire, E. Overcoming HIV-1 resistance to protease inhibitors.
Drug Discovery Today: Dis. Mech. 2006, 3, 281−286.
ASSOCIATED CONTENT
Supporting Information
■
(7) Kohl, N. E.; Emini, E. A.; Schleif, W. A.; Davis, L. J.; Heimbach, J.
C.; Dixon, R. A.; Scolnick, E. M.; Sigal, I. S. Active human
immunodeficiency virus protease is required for viral infectivity. Proc.
Natl. Acad. Sci. U.S.A. 1988, 85, 4686−4690.
*
S
Experimental data for the synthesis and characterization of
inhibitor 1a. Assay protocol for inhibition of HIV-PR. Protocols
for expression and purification of HIV-PR, crystallization of HIV-
PR/1b complex and pure 1a, and structure determination. This
material is available free of charge via the Internet at http://pubs.
acs.org.
Accession Codes
Coordinates for the HIV-PR/1b crystal structure have been
deposited in Protein Data Bank (PDB code: 3TOF). CCDC
(
8) Tang, J. Specific and irreversible inactivation of pepsin by substrate-
like epoxides. J. Biol. Chem. 1971, 246, 4510−4517.
9) Caldera, P. S.; Yu, Z.; Knegtel, R. M. A.; McPhee, F.; Burlingame, A.
(
L.; Craik, C. S.; Kuntz, I. D.; Ortiz de Montellano, P. R. Alkylation of a
catalytic aspartate group of the SIV protease by an epoxide inhibitor.
Bioorg. Med. Chem. 1997, 5, 2019−2027.
(10) Meek, T. D.; Dayton, B. D.; Metcalf, M. W.; Dreyer, G. B.;
Strickler, J. E.; Gorniak, J. G.; Rosenberg, M.; Moore, M. L.; Magaard, V.
W.; Debouck, C. Human immunodeficiency virus 1 protease expressed
in Escherichia coli behaves as a dimeric aspartic protease. Proc. Natl.
Acad. Sci. U.S.A. 1989, 86, 1841−1845.
8
43044 contains the supplementary crystallographic data for this
letter. These data can be obtained free of charge from the
CCDC) Cambridge Crystallographic Data Centre via www.
(
ccdc.cam.ac.uk/data_request/cif.
(11) Yu, Z.; Caldera, P.; McPhee, F.; De Voss, J. J.; Jones, P. R.;
Burlingame, A. L.; Kuntz, I. D.; Craik, C. S.; Ortiz de Montellano, P. R.
Irreversible inhibition of the HIV-1 protease: Targeting alkylating agents
to the catalytic aspartate groups. J. Am. Chem. Soc. 1996, 118, 5846−
AUTHOR INFORMATION
Corresponding Authors
(S.G.) Tel: +390405583936. Fax: +390405583903. E-mail:
■
*
5
856.
12) Park, C.; Koh, J. S.; Son, Y. C.; Choi, H. I.; Lee, C. S.; Choy, N.;
(
sgeremia@units.it.
(F.B.) Tel: +390405583920. E-mail: benedett@units.it.
Moon, K. Y.; Jung, W. H.; Kim, S. C.; Yoon, H. Rational design of
irreversible pseudo C2-symmetric HIV-I protease inhibitors. Bioorg.
Med. Chem. 1995, 5, 1843−1848.
(13) Abell, A. D.; Hoult, D. A.; Bergman, D. A.; Fairlie, D. P. Simple cis-
epoxide-based inhibitors of HIV-1 protease. Bioorg. Med. Chem. Lett.
*
Author Contributions
The manuscript was written through contributions of all authors.
All authors have given approval to the final version of the
manuscript.
1
(
997, 7, 2853−2856.
14) Lee, S. C.; Choy, N.; Park, C.; Choi, H.; Chan Son, Y.; Kim, S.;
Funding
Ok, J. H.; Yoon, H.; Kim, S. C. Design, synthesis and characterization of
dipeptide isoestere containing cis-epoxide for the irreversible
inactivation of HIV protease. Bioorg. Med. Chem. Lett. 1996, 6, 589−594.
Authors are grateful to Schlumberger Foundation for Faculty for
the Future fellowship awarded to Folasade M. Olajuyigbe and to
the International Centre for Theoretical Physics (ICTP),
Trieste, Italy for STEP Fellowship. We thank MIUR (FIRB
RBRN062BCT and PRIN 20109Z2XRJ), FRA-2012 University
of Trieste, and Friuli-Venezia-Giulia region (DPReg. 120/2007/
Pres.) for financial and scientific support.
(15) Choy, N.; Choi, H.; Jung, W. H.; Kim, C. R.; Yoon, H.; Kim, S. C.;
Lee, T. J.; Koh, J. S. Synthesis of irreversible HIV-I protease inhibitors
containing sulfonamide and sulfone as amide bond isoesteres. Bioorg.
Med. Chem. Lett. 1997, 7, 2635−2638.
(16) Benedetti, F.; Berti, F.; Miertus, S.; Romeo, D.; Schillani, F.; Tossi,
A. Design, synthesis and preliminary evaluation of peptidomimetic
inhibitors of HIV aspartic protease with an epoxyalcohol core.
ARKIVOC 2003, XIV, 140−154.
Notes
The authors declare no competing financial interest.
(17) Miller, M. The early years of retroviral protease crystal structures.
ACKNOWLEDGMENTS
The authors thank the beamline scientists at Elettra Synchrotron
for technical assistance.
■
Biopolymers 2010, 94, 521−529.
(18) Olajuyigbe, F. M.; Demitri, N.; Ajele, J. O.; Maurizio, E.;
Randaccio, L.; Geremia, S. Carbamylation of N-terminal proline. ACS
Med. Chem. Lett. 2010, 1, 254−257.
REFERENCES
(19) Mahalingam, B.; Louis, J. M.; Hung, J.; Harrison, R. W.; Weber, I.
J. Structural implications of drug-resistant mutants of HIV-1 protease:
high-resolution crystal structures of the mutant protease/substrate
analogue complexes. Proteins: Struct. Funct. Bioinf. 2001, 43, 455−464.
■
(
1) Coquet, I.; Pavie, J.; Palmer, P.; Barbier, F.; Legriel, S.; Mayaux, J.;
Molina, J. M.; Schlemmer, B.; Azoulay, E. Survival trends in critically ill
HIV-infected patients in the highly active antiretroviral therapy era. Crit.
Care 2010, 14, R107.
(20) Olajuyigbe, R.; Demitri, N.; Geremia, S. Investigation of 2-fold
disorder of inhibitors and relative potency by crystallizations of HIV-I
protease in ritonavir and saquinavir mixtures. Cryst. Growth Des. 2011,
11, 4378−4385.
(
2) Ghosh, A. K.; Chapsal, B. D.; Weber, I. T.; Mitsuya, H. Design of
HIV protease inhibitors targeting protein backbone: an effective strategy
for combating drug resistance. Acc. Chem. Res. 2008, 41, 78−86.
(
3) Mitsuya, H.; Maeda, K.; Das, D.; Ghosh, A. K. Development of
(21) Geremia, S.; Demitri, N.; Wuerges, J.; Benedetti, F.; Berti, F.; Tell,
G.; Randaccio, L. A potent HIV protease inhibitor identified in an
epimeric mixture by high resolution protein crystallography. Chem-
MedChem. 2006, 1, 186−188.
Protease Inhibitors and the Fight with Drug-Resistant HIV-1 Variants.
HIV-I. Molecular Biology and Pathogenesis; Jeang, K.-T., Ed.; Academic
Press: New York, 2008; Vol. 56, pp 169−197.
(
4) Bennett, D. E.; Camacho, R. J.; Otelea, D.; Kuritzkes, D. R.; Fleury,
(22) Campaner, P. Unpublished work.
H.; Kiuchi, M.; Heneine, W.; Kantor, R.; Jordan, M. R.; Schapiro, J. M.;
(23) Benedetti, F.; Miertus, S.; Norbedo, S.; Tossi, A.; Zlatoidzky, P.
Vandamme, A. M.; Sandstrom, P.; Boucher, C. A. B.; van de Vijver, D.;
Versatile and stereoselective synthesis of diamino diol dipeptide
9
71
dx.doi.org/10.1021/ml500092e | ACS Med. Chem. Lett. 2014, 5, 968−972

ACS Medicinal Chemistry Letters
Letter
isosteres. Core units of pseudopeptide HIV protease inhibitors. J. Org.
Chem. 1997, 62, 9348−9353.
(
́
24) Rodriguez, A. C.; Pico Ramos, A.; Hawkes, G. E.; Berti, F.;
Resmini, M. Stereoselective synthesis of a novel pseudopeptide hapten
for the generation of hydrolytic catalytic antibodies. Tetrahedron:
Asymmetry 2004, 15, 1847−1855.
(
25) Adam, W.; Wirth, T. Hydroxy group directivity in the epoxidation
of chiral allylic alcohols: Control of diastereoselectivity through allylic
strain and hydrogen bonding. Acc. Chem. Res. 1999, 32, 703−707.
(
26) Brik, A.; Wong, C.-H. HIV-I protease: mechanism and drug
discovery. Org. Biomol. Chem. 2003, 1, 5−14.
(
27) Na, J.; Houk, K. N.; Shevlin, C. G.; Janda, K. D.; Lerner, R. A. The
energetic advantage of 5-exo versus 6-endo epoxide openings: A
preference overwhelmed by antibody catalysis. J. Am. Chem. Soc. 1993,
1
(
15, 8453−8454.
28) Kona, J. Theoretical study on the mechanism of a ring-opening
́
̌
reaction of oxirane by the active-site aspartic dyad of HIV-1 protease.
Org. Biomol. Chem. 2008, 6, 359−365.
(
29) Mavri, J. Irreversible inhibition of HIV-protease: a theoretical
study. Int. J. Quantum Chem. 1998, 69, 753−759.
9
72
dx.doi.org/10.1021/ml500092e | ACS Med. Chem. Lett. 2014, 5, 968−972