Discovery of potent HIV-1 protease inhibitors incorporating sulfoximine functionality

Article abstract of DOI:10.1016/j.bmcl.2007.07.095

Based on the unique property of sulfoximine and the homodimeric C₂ structural symmetry of HIV-1 protease, a novel class of sulfoximine-based pseudosymmetric HIV-1 protease inhibitors was designed and synthesized. The sulfoximine moiety was demonstrated to be important for HIV-1 protease inhibitor potency. The most active stereoisomer (2S,2′S) displays a potency of 2.5 nM (IC₅₀) against HIV-1 protease and an anti-HIV-1 activity of 408 nM (IC₅₀). A possible mode of action is proposed.

Full text of DOI:10.1016/j.bmcl.2007.07.095

Bioorganic & Medicinal Chemistry Letters 17 (2007) 5614–5619
Discovery of potent HIV-1 protease inhibitors incorporating
sulfoximine functionality
Ding Lua and Robert Vincea,b,
*
aDepartment of Medicinal Chemistry, College of Pharmacy, 308 Harvard Street SE, University of Minnesota,
Minneapolis, MN 55455, USA
bCenter for Drug Design, Academic Health Center, University of Minnesota, Minneapolis, MN 55455, USA
Received 9 May 2007; revised 25 July 2007; accepted 26 July 2007
Available online 23 August 2007
Abstract—Based on the unique property of sulfoximine and the homodimeric C2 structural symmetry of HIV-1 protease, a novel
class of sulfoximine-based pseudosymmetric HIV-1 protease inhibitors was designed and synthesized. The sulfoximine moiety
was demonstrated to be important for HIV-1 protease inhibitor potency. The most active stereoisomer (2S,20S) displays a potency
of 2.5 nM (IC50) against HIV-1 protease and an anti-HIV-1 activity of 408 nM (IC50). A possible mode of action is proposed.
Ó 2007 Elsevier Ltd. All rights reserved.
Human immunodeficiency virus type 1 (HIV-1) is the
etiological agent of acquired immunodeficiency syn-
drome (AIDS).1 HIV-1 protease plays a vital role in
the HIV-1 life cycle and its inhibition results in imma-
ture virons which are incapable of replication. The com-
bination of HIV-1 protease inhibitors (PIs) and reverse
transcriptase inhibitors (RTIs), known as highly affec-
tive antiretroviral therapy (HAART), has dramatically
improved the life quality of HIV infected patients.2
However, since HIV can easily generate mutants with
diminished sensitivity to the known PIs,3 there is an ur-
gent need to identify new structural types of inhibitors.
peptide chemistry. Sulfoximine has been successfully
incorporated in designing inhibitors of metalloprotein-
ase such as carboxypeptidase A7 and ATP-dependent
ligases such as asparagine synthetase A.8 It has also been
used in the design of pseudopeptides to alter conforma-
tion, polarity, and metabolic stability in drug discovery.9
To our knowledge, there has been no report on applying
sulfoximine in HIV-1 protease inhibitor design.
During our continuous efforts to develop potent HIV-1
PIs, we have designed and synthesized sulfoximine-
based symmetric molecules as possible scaffolds for
HIV-1 PIs. We reasoned that the tetrahedron structural
feature of sulfoximine and its role as a potential hydro-
gen bond donor/acceptor would render it an ideal candi-
date for a transition state mimic (TSM) for HIV-1 PI.
Currently there are 10 HIV-1 PIs approved by the US
FDA.10 All carry a free hydroxyl group that interacts
with the catalytic aspartic acids in the active site to mi-
mic the hydrated amide of the substrate. Since HIV-1
protease functions as a homodimer which has a C2-sym-
metric active site, it may be desirable to include approx-
imate C2 symmetry through the center of the inhibitor.
In fact, the Merck compound, L700,417 (1, Fig. 1)11
which is a carbinol analog, is a very potent pseudosym-
metric HIV-1 PI with an IC50 value of 0.6 nM against
the enzyme. Other functional groups that have been
studied as effective isosteres of TSMs include silanediol12
and phosphinate.13 To assess the effect of sulfoximine in
HIV-1 PIs, we constructed our target molecule 4 shown
The sulfoximine functional group has the potential for
useful incorporation into enzyme inhibitors as it is
chemically stable and tetrahedrally hybridized.4 It was
first discovered in late 1940s as a key structure of methi-
onine sulfoximine, a toxic substance found in wheat
flour treated with nitrogen trichloride.5 Although this
compound was identified as a potent mechanism-based
inhibitor of glutamine synthetase, most early work
focused on the utilization of sulfoximines and their
derivatives as valuable reagents for asymmetric synthe-
sis.6 Recently, more attention has been directed toward
exploring this functionality as an amide replacement in
Keywords: HIV-1 protease inhibitor; C2 symmetry; Sulfoximine; Tra-
nsition state mimic.
*
Corresponding author. Tel.: +1 612 6249911; fax: +1 612 6240139;
e-mail: vince001@umn.edu
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.07.095

D. Lu, R. Vince / Bioorg. Med. Chem. Lett. 17 (2007) 5614–5619
5615
OH
HO
OH HO
OH
OH
HO
H
H
H
H
N
O
N
N
N
Si
Cbz ValN
P
N
ValCBz
H
O
O
O
O
H
OH
3
2
1, L700,417
OH
HO
NH
1 2
HN
2'
O
O
1'
O
S
3
3'
NH
4
Figure 1. Examples of symmetric HIV-1 PIs and designed target molecule.
in Figure. 1. In this structure, there are two unassigned
stereogenic carbons (C2/C20). Our initial approach was
to find a convenient and quick way to obtain the target
molecules and verify our hypothesis, thus we synthesized
compound 4 as a mixture of diastereomers and tested
them for HIV-1 protease activity. Since the resulting sul-
foximine mixture was shown to be potent against HIV-1
protease, we subsequently prepared each stereoisomer
separately and identified the most potent isomer to be
(2S,20S) in configuration (Fig. 1).
zylmalonic acid 5 via a Mannich reaction,14 was con-
verted to intermediate 7 by refluxing with thioacetic
acid and methyl acrylate
8
by reacting with
TMSCHN2. Michael addition between 7 and 8 affor-
ded intermediate 9 as a mixture of diastereomers.
Hydrolysis of the resulting esters with LiOH in aque-
ous THF afforded the corresponding diacid 10, which
was then coupled with (1S,2R)-(ꢀ)-cis-1-amino-2-inda-
nol using BOP reagent15 to generate diamide com-
pound 11. Subsequent oxidation of 11 by mCPBA
produced sulfoxide 13, while the sulfone analog 12
was obtained with oxone.16 Finally, sulfoxide 13 was
converted into the desired sulfoximine 4 using O-mes-
itysulfonylhydroxylamine (MSH).17
We pursued our initial strategy to obtain sulfoximine
compounds as a mixture of four diastereomers
(Scheme 1). Briefly, Acrylic acid 6, obtained from ben-
Ph
HO
Ph
O
d
S
OMe
AcS
OH
b
c
O
7
O
9
a
HO
OH
OH
e
O
O
O
5
6
Ph
Ph
OH
OMe
HO
S
O
8
O
O
10
f
Ph
Ph
O
Ph
O
Ph
O
OH
OH
O
S
OH
OH
H
H
N
H
H
N
N
S
N
h
O
13
11
g
i
Ph
Ph
Ph
Ph
OH
OH
O
OH
OH
O
S
H
H
H
H
N
S
N
N
N
NH
O
O
O
O
O
12
4
Scheme 1. Reagents and conditions: (a) formalin, Et2NH, reflux, 82%; (b) AcSH, benzene, reflux, 77%; (c) TMSCHN2, 89%; (d) NaOMe, 70%; (e)
LiOH, THF/H2O, 86%; (f) (1S,2R)-(ꢀ)-cis-1-amino 2-indaol, BOP reagent, Et3N, 50%; (g) oxone, 80%; (h) mCPBA, quantitative; (i) MSH, 35%.

5616
D. Lu, R. Vince / Bioorg. Med. Chem. Lett. 17 (2007) 5614–5619
Evaluation of these sulfoximine compounds and their
precursors in the recombinant HIV-1 protease enzyme
assay indicated that both sulfoximine and sulfoxide
compounds are potent inhibitors while the sulfone and
sulfide compounds exhibit much less activity (Table 1).
As a diastereomeric mixture, 13 showed 75% inhibition
at 10 lM, while sulfoxide containing C2-symmetric pep-
tide analogs18 were previously shown to be inactive
against HIV-1 protease within the limit of detection.
On the other hand, the target sulfoximine compound 4
showed 87% inhibition at 10 lM. These preliminary
results suggest that the sulfoxide/sulfoximine moiety
could serve as an effective isostere of TSM for HIV-1
PIs. The activity data from sulfoxide and sulfoximine
compounds indicate that sulfur containing C2-symmet-
ric analogs could be used in designing potent HIV-1 pro-
tease inhibitors, which is contrary to the previous report
that sulfide and its oxidative derivatives are not good
TSM in C2-symmetric HIV-1 PIs. The major difference
between the literature18 molecules and ours is the inser-
tion of one methylene unit between the sulfur and benzyl
groups. This extra carbon boosts the activity of 13 and 4
significantly against the HIV-1 protease suggesting that
the proper length between P1=P01 (as defined by Schech-
ter and Berger’s nomenclature)19 is an important
requirement for high inhibitory potency. This promising
result encouraged us to determine the relationship
between stereochemistry and bioactivity of the diaste-
reomeric sulfoximine compounds.
Table 1. Evaluation of compounds 11, 12, 13, and 4 as HIV-1 protease
inhibitors
Ph
O
Ph
O
OH
OH
H
H
N
R
N
The synthetic strategy leading to each stereoisomer of 4
is outlined in Scheme 2. A modified approach to control
the stereochemistry of the central diacid unit was evalu-
ated. Thus, optically pure amides from each side of the
target molecules were prepared. The number of diaste-
reomers produced by the subsequent Michael Addition
was minimized. Amide coupling of 7 with (1S,2R)-(ꢀ)-
cis-1-amino-2-indanol using BOP reagent afforded 14
as a mixture of two diastereomers which were well
separated in an equal ratio by flash chromatography.
Coupling between acrylic acid 6 and (1S,2R)-(ꢀ)-cis-1-
amino-2-indanol provided 15 in good yield. Michael
addition between 15 and 14a afforded a mixture of
two diastereomers 16a/16b. The two isomers were sepa-
rated by reverse phase HPLC20 in equal amount indicat-
Compound
R
% Inhibition at 10 lM
11
S
15
O
S
12
30
O
O
S
13
4
75
87
O
S
NH
OH
H2N
OH
OH
a
H
H
AcS
OH
AcS
N
+
AcS
N
+
O
O
7
O
14a, 38%
14b, 40%
OH
OH
H2N
H
N
a
+
OH
O
O
15, 89%
6
HO
OH
HO
OH
H
H
H
H
N
b
N
S
N
N
S
+
15
14a
+
(S)
(S)
(R)
(S)
O
O
O
O
16a (2S,2'S), 35%
16b (2
S
,2'R), 37%
HO
OH
HO
OH
H
H
H
N
H
N
S
N
S
N
b
+
(R)
(R)
(S)
(R)
+
15
14b
O
O
O
O
16c (2R,2'R), 36%
16b (2S,2'R), 35%
Scheme 2. Reagents: (a) BOP reagent, Et3N; (b) NaOMe/MeOH.

D. Lu, R. Vince / Bioorg. Med. Chem. Lett. 17 (2007) 5614–5619
5617
O
S
HO
OH
n
O
HO
OH
H
H
a
H
H
b
N
N
N
S
N
C1
C2
C1
C2
NH
n=1
n=0
16a (2 ,2'
16b (2
16c (2
O
O
O
O
17a (2S,2'
S
)
)
)
S
S)
18a (2
S,2'S)
17b (2
S
,2'
R
S
,2'
R
)
)
18b (2
S
,2'
,2'
R
)
17c (2
R
,2'R
R
,2'
R
18c (2R
R
)
Scheme 3. Reagents: (a) mCPBA; (b) MSH.
ing lack of diastereoselectivity in this reaction. The same
procedure was applied to 15 and 14b to yield another
two diastereomers 16b/16c. Because of the structural
symmetry, three diastereomers of sulfide compounds
(16a, 16b, and 16c) were obtained. All isomers were
individually oxidized to sulfoxide by mCPBA. While
16a(2S,20S) and 16c(2R,20R) afforded corresponding
sulfoxide as a single diastereomer, 16b(2R,20S) yielded
sulfoxide 17b as a mixture of two epimers. Since HPLC
separation was not satisfactory for 17b, further reaction
was carried out without separation. Each sulfoxide was
then treated with MSH to produce the corresponding
sulfoximines 18a/18b/18c (Scheme 3).
crystallography studies. COSY spectra of 14a(2S) and
14b(2R) revealed that the chemical shifts from benzyl
and indanol groups are within 6.9–7.4 ppm for 14a,
while for 14b an extra peak at 6.4 ppm was observed.
It was identified from HMQC that the extra peak corre-
sponds to an aromatic proton. This chemical shift pat-
tern reflects an anti-(2S) or syn-(2R) configuration
between benzyl and indanol groups. In the X-ray crystal
structure of 14b,21 one H from amino indanol is situated
in close proximity to the benzyl group and well within
the range of the shielding zone. The corresponding H
of diastereomer 14a is presumably positioned away from
the benzyl group. Given the fact that the absolute
stereochemistry of amino indanol is known, this
observation could be used to determine the absolute ste-
reochemistry of pseudosymmetric sulfur-containing tar-
gets. Similar COSY and HMQC experiments were
carried out on each diastereomer of 16 and the chemical
shift variations associated with changes in absolute con-
figurations of these groups were easily observed. Thus,
The absolute stereochemistry was examined in detail by
means of 2D NMR correlation experiments and X-ray
Table 2. Key chemical shift
Compound
H integration Chemical shift
H
integration
1
in the H NMR spectrum of 16b(2S,20R), one H corre-
in aromatic
region
(shielding from
aromatic region)
ppm
sponding to an aromatic group with a shielding of about
0.8 ppm from the aromatic region was observed when
compared to 16a(2S,20S). In 16c(2R,20R) there are two
such peaks (Table 2). By counting the number of proton
peaks close to 6.3 ppm, the absolute configuration of the
pseudosymmetric sulfide compound could be deter-
mined. The oxidation of 16 to 17 and the imination17
14a(2S)
14b(2R)
16a(2S,20S) 18
16b(2S,20R) 17
16c(2R,20R) 16
9
8
—
6.43 (d,J = 7.8 Hz)
—
0
1
0
1
2
6.27 (d, J = 7.5 Hz)
6.20 (d, J = 7.5 Hz)
O
S
O
O
N
N
(R)
O
(R)
NH
O
19a
Figure 2. X-ray structure of 19a.

5618
D. Lu, R. Vince / Bioorg. Med. Chem. Lett. 17 (2007) 5614–5619
of 17 to 18 should all occur with retention of the chiral-
ity of intermediate 16. Similar chemical shift scattering
patterns were indeed observed for all the described mol-
ecules, which led to the assignment of the stereochemis-
try of each isomer as shown in Schemes 2 and 3.
Acetonide derivative 19a was prepared from 17a and
the crystal structure22 was obtained (Fig. 2) which con-
firms the stereochemical assignment.
The X-ray structure of Merck L700,417 complexed with
the HIV-1 protease indicated that the benzyl groups oc-
cupy S1=S01 pockets while amino indanol occupies S2=S02
pockets in the active site.23 We anticipated that the sul-
foximine moiety would form hydrogen bonds with the
catalytic aspartic acids 25 and 250, while the carbonyl
groups would interact with backbone amide groups of
ILe 50 and 500 on the flaps of the enzyme via an H2O
molecule (Fig. 3). The presence of a hydrogen bond do-
nor is essential for the inhibitory potency where the NH
might be engaged in similar hydrogen bond interaction
as an OH. It has been shown that in the active pH range
one aspartic acid in the active site is deprotonated24 to
serve as a good hydrogen bond acceptor. In the sulfox-
ide case, oxygen might function as a hydrogen bond
acceptor to accommodate one aspartic acid which is
intact. For the sulfone derivative, two oxygen atoms
may cause some negative impact to abolish or
drastically decrease the preferred interactions with either
aspartic acid residue in the active site.
Each sulfoximine as well as sulfoxide compound was
evaluated against the HIV-1 protease (Table 3). In each
class, the order of activity is (2S,20S) > (2R,20S) >
(2R,20R). Configuration switching of either C2 or C20
from S to R resulted in a loss of potency, while switching
of both carbons led to a significant drop in activity. This
is exemplified in the case of sulfoxide analogs, where the
IC50 of 17a(2S,20S) is 21.1 nM while 17b(2R,20S) is
53.1 nM, and 17c (2R,20R) has significantly less activity
(IC50 > 10 lM). It can be concluded that the stereo-
chemistry at C2 and C20 appears to be crucial for
HIV-1 protease inhibition. The importance of the ste-
reochemistry for inhibitory activity observed in this
study is in agreement with that of Merck L-700,417 in
which an anti-relationship between benzyl and amino
indanol moiety is observed. The protease activity data
also indicate that sulfoximine is more potent than sulf-
oxide as exemplified by 17a and 18a. They both have
the same configuration at C2 and C20, but the sulfoxi-
mine analog 18a is 8-fold more potent than the sulfoxide
analog 17a. The same relationship can also be identified
between 17b and 18b as well as 17c and 18c. The most
active compound 18a(2S,20S) was determined to have
an IC50 value of 2.5 nM. These results indicate that
the sulfoximine group plays a significant role in the
binding to the enzyme.
Asp25'
Asp25
OH
O
S1'
O
O
O
Ph
Ph
NH
OH
H
S
H
N
OH
N
O
O
H
H
O
S2'
Ile50
Ile50'
Figure 3. Proposed mode of action of 18a.
Table 3. Evaluation of sulfoxide and sulfoximine analogs against the HIV-1 protease
Compound
IC50 (nM)
Compound
IC50 (nM)
O
S
O
HO
H
OH
OH
OH
HO
H
OH
OH
OH
H
N
H
N
S
N
N
21.1
2.5
(
(
(
S)
(S)
(
S
)
)
)
(S)
NH
O
O
O
O
O
O
O
O
18a
17a
O
S
O
HO
H
HO
H
H
N
H
N
S
N
N
53.1
37.3
(S)
(R)
(
R
S)
NH
O
O
17b*
18b*
O
S
O
HO
H
HO
H
H
N
H
N
S
N
N
>10,000
153.9
(
R
R)
(R)
(R)
NH
O
O
17c
18c
*Mixture of 2 diastereomers.

D. Lu, R. Vince / Bioorg. Med. Chem. Lett. 17 (2007) 5614–5619
5619
7. Mock, W. L.; Tsay, J. T. J. Am. Chem. Soc. 1989, 111,
4467.
8. Koizumi, M.; Hiratake, J.; Nakatsu, T.; Kato, H.; Oda, J.
i. J. Am. Chem. Soc. 1999, 121, 5799.
9. Bolm, C.; Kahmann, J. D.; Moll, G. Tetrahedron Lett.
1997, 38, 1169.
The in vitro antiviral activity of the most potent inhibi-
tor 18a was assessed using a cell viability assay. It dis-
played a clear dose response with an antiviral IC50
value of 408 nM and no toxicity was observed at 10
lM, the highest concentration tested.
10. Whiting, M.; Tripp, J. C.; Lin, Y.-C.; Lindstrom, W.;
Olson, A. J.; Elder, J. H.; Sharpless, K. B.; Fokin, V. V. J.
Med. Chem. 2006, 49, 7697.
11. Askin, D.; Wallace, M. A.; Vacca, J. P.; Reamer, R. A.;
Volante, R. P.; Shinkai, I. J. Org. Chem. 1992, 57, 2771.
12. Chen, C.-A.; Sieburth, S. M.; Glekas, A.; Hewitt, G. W.;
Trainor, G. L.; Erickson-Viitanen, S.; Garber, S. S.;
Cordova, B.; Jeffry, S.; Klabe, R. M. Chem. Biol. 2001, 8,
1161.
13. Abdel-Meguid, S. S.; Zhao, B.; Murthy, K. H. M.;
Winborne, E.; Choi, J. K.; DesJarlais, R. L.; Minnich,
M. D.; Culp, J. S.; Debouck, C., et al. Biochemistry 1993,
32, 7972.
In summary, we have demonstrated for the first time
that sulfoximine is a novel moiety potentially function-
ing as a transition state mimic in HIV-1 PIs. We also
confirmed the importance of proper stereochemistry
and identified 18a(2S,20S) to be the most potent stereo-
isomer with activity comparable to indinavir against
HIV-1 protease. The sulfoximine-based compounds are
of great interest in designing more potent HIV-1 prote-
ase inhibitors. Further studies on asymmetric synthesis,
SAR, and mechanism of inhibition are underway and
will be reported in future accounts.
14. Liu, X.; Hu, X. E.; Tian, X.; Mazur, A.; Ebetino, F. H. J.
Organomet. Chem. 2002, 646, 212.
Acknowledgments
15. Robl, J. A.; Cimarusti, M. P.; Simpkins, L. M.; Brown, B.;
Ryono, D. E.; Bird, J. E.; Asaad, M. M.; Schaeffer, T. R.;
Trippodo, N. C. J. Med. Chem. 1996, 39, 494.
16. Trost, B. M.; Curran, D. P. Tetrahedron Lett. 1981, 22,
1287.
17. Johnson, C. R.; Kirchhoff, R. A.; Corkins, H. G. J. Org.
Chem. 1974, 39, 2458.
18. Spaltenstein, A.; Leban, J. J.; Furfine, E. S. Tetrahedron
Lett. 1993, 34, 1457.
The authors gratefully acknowledge Dr. Steve Patterson
for assistance in HPLC separation, Dr. Victor Young
from X-ray Crystallographic Laboratory, Department
of Chemistry, University of Minnesota for X-ray analy-
sis. We acknowledge Ms. Christine Dreis for performing
the HIV protease assays, Roger Ptak at Southern
Research Institute for the antiviral testing data. We
thank Dr. Abbas Raza for providing X-ray crystal struc-
ture of compound 14b. We also thank Dr. William M.
Shannon for his advice as a consultant for the antiviral
work on this project.
19. Schechter, I.; Berger, A. Biochem. Biophys. Res. Commun.
1967, 27, 157.
20. Reverse-phase HPLC was run on Varian Microsorb
column (8l, C18, 250X41.4 mm) using two solvent systems
with 40 mL/min flow rate and detected at 220 and 230 nm.
Solvent system A: 0.02 M TEAB (triethylammonium
bicarbonate) in water, B: 70% acetonitrile in water, 75–
100% B linear in 40 min.
References and notes
21. Crystallographic data (excluding structure factors) for the
structures of 14b1 have been deposited with the Cam-
bridge Crystallographic Data Centre as supplementary
publication numbers CCDC 655535. Copies of the data
can be obtained, free of charge, on application to CCDC,
12 Union Road, Cambridge CB2 1EZ, UK; fax: +44
(0)1223 336033 or e-mail: deposit@ccdc.cam.ac.uk.
22. Cambridge Crystallographic Data Centre deposition
number for the structure of 19a is CCDC 645798.
23. Bone, R.; Vacca, J. P.; Anderson, P. S.; Holloway, M. K.
J. Am. Chem. Soc. 1991, 113, 9382.
1. Morrow, C. D.; Park, J.; Wakefield, J. K. Am. J. Physiol.
1994, 266, C1135.
2. Palella, F. J., Jr.; Delaney, K. M.; Moorman, A. C.;
Loveless, M. O.; Fuhrer, J.; Satten, G. A.; Aschman, D. J.;
Holmberg, S. D. N. Engl. J. Med. 1998, 338, 853.
3. Condra, J. H.; Schleif, W. A.; Blahy, O. M.; Gabryelski,
L. J.; Graham, D. J.; Quintero, J. C.; Rhodes, A.;
Robbins, H. L.; Roth, E., et al. Nature 1995, 374, 569.
4. Reggelin, M.; Zur, C. Synthesis 2000, 2000, 1.
5. Bentley, H. R.; McDermott, E. E.; Pace, J.; Whitehead, J.
K.; Moran, T. Nature 1949, 164, 438.
24. Hofmann, T.; Hodges, R. S.; James, M. N. Biochemistry
1984, 23, 635.
6. Johnson, C. R. Acc. Chem. Res. 1973, 6, 341.