Design and synthesis of sulfoximine based inhibitors for HIV-1 protease

Article abstract of DOI:10.1016/j.bmcl.2008.09.044

A new class of potent sulfoximine inhibitors for HIV-1 protease has been designed and synthesized. Substitution of the sulfoximine moiety into different parent compounds yields different inhibition effects. While our previously studied sulfoximine-based inhibitors display potency of 2.5 nM (IC₅₀) against HIV-1 protease, introduction of the sulfoximine moiety into the asymmetric Indinavir yielded only micromolar inhibition. Docking studies showed structural variations in their modes of binding which explains this unexpected observation. The implication of these observations in the development of other sulfoximine inhibitors is discussed.

Full text of DOI:10.1016/j.bmcl.2008.09.044

Bioorganic & Medicinal Chemistry Letters 18 (2008) 5406–5410
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of sulfoximine based inhibitors for HIV-1 protease
*
Abbas Raza, Yuk Yin Sham, Robert Vince
Center for Drug Design, Academic Health Center, University of Minnesota, 516 Delaware St SE, Minneapolis, MN 55455, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 July 2008
Revised 9 September 2008
Accepted 10 September 2008
Available online 13 September 2008
A new class of potent sulfoximine inhibitors for HIV-1 protease has been designed and synthesized. Sub-
stitution of the sulfoximine moiety into different parent compounds yields different inhibition effects.
While our previously studied sulfoximine-based inhibitors display potency of 2.5 nM (IC50) against
HIV-1 protease, introduction of the sulfoximine moiety into the asymmetric Indinavir yielded only micro-
molar inhibition. Docking studies showed structural variations in their modes of binding which explains
this unexpected observation. The implication of these observations in the development of other sulfox-
imine inhibitors is discussed.
Keywords:
HIV protease
HAART
Ó 2008 Elsevier Ltd. All rights reserved.
Sulfoximine
Transition state mimic
Indinavir
Docking studies
Crystallographic water
TSM binding
Human immunodeficiency virus (HIV) infection can lead to ac-
quired immunodeficiency syndrome (AIDS).1,2 It appears in two
forms as HIV-1 and HIV-2. Currently, there are over 30 US Food
and Drug Administration (FDA) approved antiviral drugs available1
for the highly active antiretroviral regimen treatment (HAART) for
AIDS.2–4 The treatment targets simultaneously various viral protein
enzymes found only in HIV, thereby, preventing full maturation
during viral replication within the host cells. Due to the persistence
of the viral reservoir within tissues, AIDS remains an incurable dis-
ease.5 As HIV with drug resistant viral strains continues to emerge,
extensive efforts are being spent on the development of novel ther-
apeutic agents that can be used as alternative drug substitutes to
augment HAART antiviral activities.6
One of the vulnerable viral protein targets most studied is the
HIV protease.7 It is a homo dimeric aspartyl enzyme that catalyzes
the proteolytic cleavage of a large polyprotein precursor during vir-
al assembly and maturation.8,9 Since the characterization of its
structure6 and mechanistic function,10,11 innumerous rationally de-
signed inhibitors have been chemically synthesized and clinically
tested.12–15 To-date, there are only 10 commercially available pro-
tease inhibitors (PIs) approved by the FDA.1 Nine of these inhibitors
contain a free hydroxyl group that acts as a transition state mimic
(TSM) for the tetrahedral hybridized amide linkage during catalytic
cleavage. Structurally, this secondary alcohol is shown to interact
directly with the catalytic Asp25 and Asp250 residues in the active
site and has been shown to be crucial in the design of highly potent
HIV PIs. In addition to the TSM, many of the HIV PIs include a hydro-
gen bond acceptor motif, which interact directly with a crystallo-
graphic conserved water molecule that is hydrogen bonded to the
carbonyl backbone of Ile50 and Ile500 residues in the active site.
Compounds designed to include this conserved water mimic have
also been shown to improve on binding affinity.16 Inclusion of mul-
tiple hydrophobic substituents, which project into the various
hydrophobic binding pockets (S1, S2, S1 , and S2 ) of HIV protease
as well as compounds which take advantage of the C2 symmetry
of the homodimeric HIV protease16,17 are also desirable (Fig. 1).
Sulfoximines, which are often referred to as ‘chemical chame-
leons’, have been well studied for their bioactivity as antibiotics,
antithrombotics and tumor metastasis inhibitors. Its tetrahedral
structural feature can act as an alternative substitute for the TSM
of the secondary alcohol. In addition, the sulfoximine has the po-
tential to act simultaneously as a hydrogen bond donor as well
as an acceptor. Recently,18 we have designed and synthesized a
new set of sulfoximine inhibitors for HIV-1 protease based on the
Merck compound, L700,417.19 The most active sulfoximine stereo-
isomer (2S,20S) (1) displays nanomolar inhibition potency of
2.5 nM against HIV-1 protease, similar to that of L700,417 (IC50
of 0.6 nM).
0
0
Encouraged by these results, we proceeded to substitute the
same sulfoximine moiety into an FDA approved HIV PI with the
expectations of retaining the inhibition potency by replacing the
secondary alcohol (TSM). Thus, a novel class of inhibitors similar
to one of the most potent commercially available HIV PI was
* Corresponding author. Tel.: +1 612 624 9911; fax: +1 612 625 2633.
E-mail address: vince001@umn.edu (R. Vince).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.09.044

A. Raza et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5406–5410
5407
Table 1
HIV-1 protease inhibition assay results (in vitro) for compounds 2 and 9
Compound
IC50
0.6 nM
21 nM
2.5 nM
0.4 nM
Figure 1. Binding of Indinavir to HIV-1 protease.
Sequential reaction of amine 6 with p-formaldehyde under re-
flux conditions generated the Schiff base in situ to which the thiol
5 in DMF was added to give the sulfide 7. The sulfide derivative 8
was accessed by the removal of Boc and hydrolysis of the isopro-
pylidine of sulfide 7, followed by the alkylation with 3-picolyl chlo-
ride in the presence of Et3N. Subsequent oxidation of sulfide 8 with
NaIO4 produced sulfoxide 9. Finally, the reaction of sulfoxide 9
with O-mesitylsulfonylhydroxylamine (MSH)23 afforded the corre-
sponding sulfoximine 2.24
We anticipated that incorporating the sulfoximine moiety into
an existing drug molecule such as Indinavir would enhance the po-
tency through a better binding mode of the sulfoximine as pro-
posed in our earlier report. However, the in vitro assay of 2 and
9 in the recombinant HIV-1 protease enzyme provided only micro-
molar inhibition (Table 1). This suggested a major deviation from
the sulfoximine group binding into the active site of the enzyme
resulting in a significant loss of activity.
100 lM
To further understand the effect of replacing the secondary alco-
hol with sulfoximine on inhibition potency, subsequent docking cal-
culations were carried out using Glide.25 The starting X-ray crystal
structure of HIV-1 protease in complex with the inhibitor Indinavir
(PDB ID:1SDT)26 and with L700,417 (PDB ID:4PHV)27 were used.
Both complexes are structurally similar with a RMSD of 0.41 Å with
the TSM secondary alcohol positioned near equidistance to each of
the hydrogens binding to Asp25 and Asp250. The sulfoximine ana-
logues for each of the parent inhibitors were generated by the
replacement of the secondary alcohol moiety. Since Indinavir is an
asymmetric compound consisting of a stereogenic S chiral center
at the secondary alcohol carbon atom, both the R and S forms of
the sulfoximine analogue were considered. All compounds were en-
ergy minimized using OPLS 2005 forcefield in the presence of impli-
cit water solvent. The calculations were carried out with the van der
Waals radii scaled by a factor of 0.8 to compensate for the non-native
binding conformation of the HIV protease binding site for the sulfox-
imine analogues. All compounds were docked and scored using the
Glide extra-precision (XP) mode in the presence of the conserved
crystallographic water. To account for protein flexibility, conforma-
tions with the best G-score were refined by energy minimization
within a 5 Å radius of the ligand inside the protein active site while
restraining the external second outer shell of 3 Å by a force constant
of 50 kcal/mol Å2. The remaining residues were held fixed.28
Based on the best scoring conformation poses, the overall inter-
actions of the hydrophobic groups and the hydrogen bonding
interactions to the crystallographic water were conserved in both
of the sulfoximine analogues from their parent compounds. In
compound 1 (Fig. 3 C), the relative distances between the heavy
atoms of the sulfoximine group to the nearest carboxylate atom
100 lM
expected. The parent compound selected was Indinavir,20 which is
commercially known as CrixivanR (Fig. 2). It is an asymmetric PI
that is closely related to the previously studied compound of
L700,417.21 The strategy to synthesize Indinavir with a sulfoximine
moiety 2 was to first make optically pure thioester 3, the desired
isomer. The absolute stereochemistry of the isomer 3 was assigned
based on 2D NMR correlation experiments and the X-ray crystal
structure. The synthesis of the compound 3 was also carried out
as reported (CCDC 65535).18
A protecting group was appended to compound 3 using 2,20-
DMP to generate the isopropylidine derivative 4 (Scheme 1). The
thioester of 4 was hydrolyzed using K2CO3 in methanol, giving al-
most quantitative conversion to thiol 5 which was used directly in
the next step without further purification. The key intermediate 6
(Scheme 2) was conveniently prepared in optically pure form
according to the reported procedure.22

5408
A. Raza et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5406–5410
Ph
O
Ph
O
Ph
Ph
O
OH
OH
OH
OH
OH
O NH
S
H
N
H
N
H
H
N
N
(2 S)
(2 'S)
O
1
(IC50=2.5 nM)
L700,417 (IC50=0.6 nM)
Ph
H
N
Ph
O
OH
OH
N
OH
CONHBut
N
O
S
NH
H
N
N
N
N
N
O
CONHBut
Indinavir (IC50=0.4 nM)
2
(IC50=100 μM)
Figure 2. Known inhibitors of HIV protease and the designed inhibitors.
Scheme 1. Reagents and conditions: (a) 2,20-dimethoxy propane, PTSA, DCM, 85%; (b) K2CO3, MeOH.
Scheme 2. Reagents and conditions: (a) p-formaldehyde, toluene, reflux, 2 h; (b) 5 in DMF, rt, 65%; (c) 4 N HCl/dioxane, MeOH, 0 °C; (d) 3-picolyl chlorideꢀHCl, Et3N, DMF,
70 °C, 4 h, 61%; (e) m-CPBA (1.1 equiv), DCM, 60%; (f) MSH, DMF, 65%.
of the Asp25 and Asp250 were within 3.3 Å, similar to that of
L700,417 in the original X-ray structure. Conformation of the pseu-
do symmetric compound 1 (2S,2S0) bound in 180 degrees rotated
along the C2 axis of the HIV protease was also observed as an alter-
native pose with similar G-score. The binding of compound 1 did
show interactions of the sulfoximine group interacting both as a
hydrogen bond donor and acceptor as proposed in our earlier work.
This mode of binding was accompanied by small changes in the
dihedral angles along the carbon backbone of the L700,417
(Fig. 3 A) with minor changes in the binding of the hydrophobic
groups in each of the hydrophobic sites. This suggested, the incor-
poration of the sulfoximine moiety requires the parent compound
to possess sufficient degree of conformation flexibility. While it is
desirable to maximize the hydrophobic interactions within the
hydrophobic binding pockets and to retain the crystallographic
water interaction in the design of potent PI, conformational flexi-
bility is essential in order for the sulfoximine moiety to adapt the
desired bound conformation with the dual role of hydrogen bond
donor and acceptor property.
For compound 2, both sulfur stereoiosomers (R and S) bind simi-
larly to that of Indinavir (Fig. 3 B) in the active site of HIV protease
except in the overall orientation of the sulfoximine moiety. In com-
pound S-2, the oxygen atom of the sulfoximine group is at near equi-
distance to the nearest carboxylate atom of the Asp25 and Asp250
residues while the nitrogen atom is positioned within 3.5 Å from
the conserved crystallographic water (Fig. 3 D). For compound R-2,
its vice versa (not shown). No significant change in the overall dihe-
dral angles of scaffold backbone was observed as compared to the
original X-ray structure. While Indinavir does possess a degree of li-
gand flexibility, steric restriction by the piperazine moiety prevents
thesulfoximine to adapttheproperconformationasa dualhydrogen
bond donor/acceptor.

A. Raza et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5406–5410
5409
Figure 3. Binding of L700,417 (A), Indinavir (B), 1 (C), S-2 (D) to HIV-1 protease.
Both of the two conformations showed hydrogen bonding inter-
actions involving only one of the heavy atoms of the sulfoximine
group to both of the Asp25 and Asp250, contrary to our earlier
expectation. For compound S-2, such orientation would result in
a non-TSM binding, which could greatly diminish its overall bind-
ing affinity. Interestingly, both compound 2 and its sulfoxide pre-
Dreis, CDD for HIV protease assay. We thank the Minnesota Super-
computing Institute at the University of Minnesota for computa-
tional resources.
References and notes
cursor 9 exhibit similar inhibition at 100 lM, supporting the
view that both compounds may bind similarly in the HIV protease
active site.
1. http://www.fda.gov/oashi/aids/virals.html.
2. Machouf, N.; Thomas, R.; Nguyen, V. K.; Trottier, B.; Boulassel, M. R.; Wainberg,
M. A.; Routy, J. P. J. Med. Virol. 2006, 78, 608.
3. Potter, S. J. D.; Dwyer, E.; Saksena, N. K. Virology 2003, 305, 339.
4. Stephensen, A. JAMA 1997, 277, 614.
5. Kedzierska, K.; Maslin, C. L. V.; Crowe, S. M. Med. Chem. Rev. 2004, 1, 351.
6. Mastrolenzo, A.; Rusconi, S.; Scozzafava, A.; Barbaro, G.; Supuran, T. C. Curr.
Med. Chem. 2007, 14, 2734.
7. Wlodawer, A.; Miller, M.; Jaskolski, M.; Sathyanarayana, B. K.; Baldwin, E.;
Weber, I. T.; Selk, L. M.; Clawson, L.; Schneider, J.; Kent, S. B. H. Science 1989,
245, 616.
8. Miller, M.; Schneider, J.; Sathyanarayana, B. K.; Toth, M. V.; Marshall, G. R.;
Clawson, L.; Selk, L.; Kent, S. B. H. Science 1989, 246, 1149.
9. Swain, A. L.; Miller, M. M.; Green, J.; Rich, D. H.; Schneider, J.; Kent, S. B. H.;
Wlodawer, A. Proc. Natl. Acad. Sci. 1990, 87, 8805.
10. Wlodawer, A.; Erickson, J. W. Annu. Rev. Biochem. 1993, 62, 543.
11. Wlodawer, A.; Vondrasek, J. Annu. Rev. Biophys. Biomol. Struct. 1998, 27, 249.
12. Bernardi, L.; Bonini, B. F.; Dessole, G.; Fochi, M.; Comes-Franchini, M.; Gavioli,
S.; Ricci, A.; Varchi, G. J. Org. Chem. 2003, 68, 1418.
13. Erickson, J. Perspect. Drug Discov. Des. 1993, 1, 109.
14. Wannberg, J.; Kaiser, N. F. K.; Vrang, L.; Samuelsson, B.; Larhed, M.; Hallberg, A.
J. Comb. Chem. 2005, 7, 611.
15. Wannberg, J.; Sabnis, Y. A.; Vrang, L.; Samuelsson, B.; Karlen, A.; Hallberg, A.;
Larhed, M. Bioorg. Med. Chem. 2006, 14, 5303.
16. Lam, P. Y.; Jadhav, P. K.; Eyermann, C. J.; Hodge, C. N.; Ru, Y.; Bacheler, L. T.;
Meek, J. L.; Otto, M. J.; Rayner, M. M.; Wong, Y. N. Science 1994, 263, 380.
17. Trost, B. M.; Matsuoka, R. T. Synlett 1992, 1, 27.
18. Lu, D.; Vince, R. Bioorg. Med. Chem. Lett. 2007, 17, 5614.
19. Vacca, J. P.; Dorsey, D. B.; Schleif, W. A.; Levin, R. B.; McDaniel, S. L.; Darke, P. L.;
Zugay, J.; Quintero, J. C.; Blahy, O. M.; Roth, E.; Sardana, V. V.; Schlabach, A. J.;
Graham, P. I.; Condra, J. H.; Gotlib, L.; Holloway, M. K.; Lin, J.; Chen, I.; Vastag,
K.; Ostovic, D.; Anderson, P. S.; Emini, E. A.; Huff, J. R. Proc. Natl. Acad. Sci. 1994,
91, 4096.
20. Dorsey, B. D.; Levin, R. B.; McDaniel, S. L.; Vacca, J. P.; Guare, J. P.; Darke, P. L.;
Zugay, J. A.; Emini, E. A.; Schleif, W. A. J. Med. Chem. 1994, 37, 3443.
21. Askin, D.; Wallace, M. A.; Vacca, J. P.; Reamer, R. A.; Volante, R. P.;
Shinkai, I. J. Org. Chem. 1992, 57, 2771.
For successful incorporation of the sulfoximine moiety, the par-
ent compound should possess sufficient conformation flexibility to
accommodate the proper mode of binding. The incorporation of the
sulfoximine moiety into Indinavir most likely resulted in the loss of
key interactions for the binding of compound 2 to HIV protease,
resulting in the observed drastic loss in inhibition. Based on the
predicted binding conformation of the two diastereomers of com-
pound 2, the close proximity of the nitrogen or the oxygen atom of
the sulfoximine group within the crystallographic water may
potentially disrupt the pre-existing hydrogen bonding network of
the crystallographic water since both heavy atoms may act as
either a hydrogen bond donor or acceptor. Alternatively, it is pos-
sible that due to the orientation of the bound conformation, the
sulfoximine moiety displaces the conserved crystallographic water
from the active site. As interactions with the conserved crystallo-
graphic water molecule are crucial in binding, such loss of interac-
tion could drastically diminish the binding affinity.
In conclusion, the introduction of a sulfoximine group to Indina-
vir afforded less potency with in vitro activity against HIV-1 prote-
ase. The docking studies have revealed the potential disruption of
the crystallographic water network of interactions as the cause
for the loss of potency. Further work, to design and synthesize
new classes of sulfoximine inhibitors for HIV-1 protease and
understanding their mechanisms of action in the active site of
the enzyme is currently underway.
22. Rossen, K.; Weissman, S. A.; Sager, J.; Reamer, R. A.; Askin, D.; Volante, R. P.;
Reider, P. J. Tetrahedron Lett. 1995, 36, 6419.
Acknowledgments
23. Jhonson, C. R.; Kirchoff, R. A.; Corkins, H. G. J. Org. Chem. 1974, 39, 2458.
24. Spectral data for compound 2: 1H NMR (600 MHz, CDCl3) d 8.58 (m, 2H), 7.65
(br s, 1H), 7.58–7.15 (m, 11H), 5.35 (m, 1H), 4.63 (m, 1H), 4.15 (br s, 2H), 3.83
(br s, 2H), 3.74 (s, 2H), 3.45 (s, 2H), 3.08 (m, 2H), 3.82–3.86 (m, 1H), 2.78–2.74
This research work was supported by the Center for Drug De-
sign (CDD) at the University of Minnesota. We thank Ms. Christine

5410
A. Raza et al. / Bioorg. Med. Chem. Lett. 18 (2008) 5406–5410
(m, 1H), 2.41–2.58 (m, 7H), 2.31 (m, 2H), 1.32 (s, 9H). 13C NMR (150 MHz,
CDCl3) 173.69, 169.49, 150.51, 149.10, 140.63, 140.46, 139.12, 136.84,
25. Glide. Version 45, Schrödinger, LLC, New York, NY, 2007.
26. Mahalingam, B.; Wang, Y.-F.; Boross, P. I.; Tozser, J.; Louis, J. M.; Harrison, R.
W.; Weber, I. T. Eur. J. Biochem. 2004, 271, 1516.
27. Bone, R.; Vacca, J. P.; Anderson, P. S.; Holloway, M. K. J. Am. Chem. Soc. 1991,
113, 9382.
28. Barreiro, G.; Guimaraes, C. R. W.; Tubert-Brohman, I.; Lyons, T. M.; Tirado-
Rives, J.; Jorgensen, W. L. J. Chem. Inf. Model. 2007, 47, 2416.
d
132.42, 129.02, 128.60, 127.81, 126.67, 126.45, 124.98, 124.20, 123.43, 72.79,
76.83, 67.51, 63.90, 60.13, 59.91, 57.48, 54.68, 52.71, 51.25, 50.77, 50.70, 47.93,
39.57, 38.73, 37.92, 35.44, 28.97; ESI-HRMS (m/z): [M+H]+ calcd for
C35H46N6O4S 647.3379; found, 646.3364. Optical rotation ½a D20
ꢁ
þ 7:5ꢂ (c 1.0,
CHCl3).