Cyclic sulfamide HIV-1 protease inhibitors, with sidechains spanning from P2/P2′ to P1/P1′

Article abstract of DOI:10.1016/j.bmc.2004.10.042

Previous studies of HIV protease inhibitors have shown that it is possible to elongate the P1/P1′ sidechains to reach the S3/S3′ binding sites. By analogy, we expected that it would be possible to design inhibitors reaching between the S1/S1′ and S2/S2′ b

Full text of DOI:10.1016/j.bmc.2004.10.042

Bioorganic & Medicinal Chemistry 13 (2005) 755–764
Cyclic sulfamide HIV-1 protease inhibitors, with
sidechains spanning from P2/P20 to P1/P10
Anna Ax,a Wesley Schaal,a Lotta Vrang,b Bertil Samuelsson,b
a
Anders Hallberg and Anders Karlen
a,
*
´
aDepartment of Medicinal Chemistry, Uppsala University, BMC, Box 574, SE-751 23 Uppsala, Sweden
bMedivir AB, Lunastigen 7, SE-141 44 Huddinge, Sweden
Received 16 July 2004; revised 14 October 2004; accepted 19 October 2004
Abstract—Previous studies of HIV protease inhibitors have shown that it is possible to elongate the P1/P10 sidechains to reach the
S3/S30 binding sites. By analogy, we expected that it would be possible to design inhibitors reaching between the S1/S10 and S2/S20
binding sites. Molecular modeling suggested that this could be achieved with appropriate ortho-substitution of the P2/P20 benzyl
groups in our cyclic sulfamide inhibitors. Four different spacer groups were investigated. The compounds were smoothly prepared
from tartaric acid in five steps and exhibit low to moderate activity, the most potent inhibitor possessing a Ki value of 0.53lM.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
1994,5 we embarked on a study of related cyclic sulf-
amides, where carbohydrates were employed as precur-
sors and chiral pools.6 These cyclic sulfamides, for
example, the lead compound 2, prepared from L-man-
nonic-c-lactone in nine steps and with a Ki value of
23nM, was found to bind to HIV protease in an non-
symmetric binding mode.6,7 Specifically, the P10 and
P20 sidechains were transposed relative to the expected
binding mode observed for the corresponding cyclic
urea-based inhibitor (1).8 All cyclic sulfamide—HIV
protease X-ray complexes to date have presented this
same binding mode (PDB codes: ajv1, 1gk2, and
1g35).7–9
The number of people living with HIV/AIDS is esti-
mated to 38 million and approximately 3 million AIDS
deaths were reported in 2003.1 The inhibitors of HIV
protease, which entered the market in 1995, have proven
to be valuable therapeutics in combination with nucleo-
side and non-nucleoside reverse transcriptase inhibitors
(NRTIꢀs and NNRTIꢀs, respectively) in the treatment
of AIDS. This effective drug combination therapy is
known as highly active antiretroviral therapy
(HAART).2,3 Unfortunately this therapy is often attrib-
uted to problems with adherence and to the emergence
of HIV drug resistance.4 As a result, intensive research
efforts have been devoted to the search for new chemical
entities such as potent HIV protease inhibitors exhibit-
ing a minimum of cross-resistance and that are antici-
pated to be structurally distinct from those used in
therapy today.
Previous studies of HIV protease inhibitors have shown
that it is possible to elongate the P1/P10 sidechains to
reach the S3/S30 binding sites.10 By analogy, inspection
of the X-ray structure of our cyclic sulfamide inhibitors
in complex with HIV-1 protease,7–9 predicted that it
could also be possible to reach between the S1/S10 and
S2/S20 binding sites. Molecular modeling suggested that
this design could be achieved with appropriate ortho-
substitution of the P2/P20 benzyl groups. We herein re-
port compounds with the generic structure 3 (Fig. 1)
that exhibited low to moderate activity and that could
easily be prepared via a short synthetic route that in-
cludes bisbenzylation and cross-couplings as key
reactions.
Inspired by the elegant design of cyclic ureas as HIV
protease inhibitors that was reported by Lam et al. in
Keywords: AIDS; HIV-1 protease inhibitors; Cyclic sulfamide; Mole-
cular modeling.
*
Corresponding author. Tel.: +46 18 471 4293; fax: +46 18 471
4474; e-mail: anders.karlen@orgfarm.uu.se
0968-0896/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.10.042

756
A. Ax et al. / Bioorg. Med. Chem. 13 (2005) 755–764
Table 1. Inhibitory activity of biaryl analogs
S2'
S2
O
N
O
N
O
S
N
N
R
R
HO
OH
O
O
OH
HO
Compd
R
Yield (%)
37
Ki (lM)
1
S1'
S1'
S1
S2
Ki = 12 nM
H
9
>20
10
11
53
28
2.0
3.0
O
N
O
N
S
O
O
OH
HO
2
S2'
S1'
S1
S2
Ki = 23 nM
12
13
51
53
2.5
2.5
O
N
O
N
S
0-2[ ]
[ ]0-2
OH
HO
R
that 10 and 11 appeared to be touching the edge of
S3. In contrast, the biphenyl 13 (magenta) seemed to
prefer a significantly different mode of binding. The bulk
of the phenyl ring in the ortho-position could simply not
be accommodated. As a result, the prime side benzyl
group of 13 got positioned in between S10 and S20 and
the terminal phenyl group on the opposite side was posi-
tioned between the S2 and S3 binding sites. Therefore,
13 as well as the unsubstituted 9 were predicted to
occupy only two out of the four possible binding sites
and would likely result in poor activity. The promising
results for the docking of 10–12 encouraged us to pro-
ceed with their synthesis. Compounds 9 and 13 were
included as negative controls.
R
3
S1
S2'
Figure 1. Structure of a cyclic urea inhibitor 1 with symmetric binding
to HIV-1 protease and a cyclic sulfamide inhibitor 2 with nonsym-
metric binding. The generic structure 3 and the proposed nonsymmet-
ric binding to HIV-1 protease.
2. Results
2.1. Modeling
Compounds 9–13 (Table 1) were subjected to a con-
formational search in the active site of the HIV-1 pro-
tease. The results from the docking are depicted in
Figure 2 with the X-ray structure of 2 (yellow) for com-
parison. The unsubstituted compound 9 (gray), stilbene
analog 10 (green), ethylene analog 11 (cyan), and meth-
ylene analog 12 (orange) all adopted binding modes,
which were quite similar to that observed for 2, except
2.2. Enzyme inhibition
The inhibitory activities for compounds 9–13 are sum-
marized in Table 1. Consistent with the predictions from
modeling, compound 9 was found to be inactive.
Figure 2. Results from the Monte Carlo conformational search. Depicted are the low energy conformations of 9 (gray), 10 (green), 11 (cyan), 12
(orange), and 13 (magenta), the X-ray crystal structure of 2 (yellow) is included for comparison. The molecular graphics image was produced using
the UCSF Chimera.32

A. Ax et al. / Bioorg. Med. Chem. 13 (2005) 755–764
757
O
HN
O
NH
S
OH
O
O
O
NH2
O
O
O
N
O
H2
b
c
a
O
O
O
O
O
O
94%
20%
80%
O
OH
O
O
4
6
5
7
O
N
O
N
O
N
O
N
S
S
e
d
R
R
72%
O
O
HO
OH
8a: H 50%
8b: Br 99%
9
Scheme 1. Reagents and conditions: (a) 2,2-dimethoxypropane and DL-10-camphorsulfonic acid in acetone; (b) NH3/methanol; (c) i. LiAlH4 in THF,
80ꢁC, ii. NH2SO2NH2, Et3N, toluene, 120ꢁC; (d) for 8a benzyl bromide, DMF, K2CO3 and for 8b 2-bromobenzyl bromide, DMF, K2CO3; (e) HCl/
ether and methanol.
Compounds 10–12 were found to have reasonably good
inhibitory potency but unfortunately did not attain the
potency of the lead compound 2. Compound 13, which
according to the modeling studies should only interact
with two binding sites, turned out to be as potent as
10–12.
O
N
O
N
S
a
10
11
HO
OH
To explore this unexpected result, we decided to focus
our further research efforts on the derivatives of 13.
Small groups in the para- or meta-position of the at-
tached aryl were anticipated by modeling to be favor-
ably accommodated in the protein by reaching further
into or filling, the S1 and S20. A large variety of building
blocks were commercially available.
O
N
O
N
S
b
c
HO
OH
8b
O
N
O
N
S
S
2.3. Chemistry
12
13
Dimethyl L-tartrate (4) served as the chiral starting
material for the target molecules (Schemes 1 and 2).
The protection of the two hydroxyl groups of the tar-
trate and the transformation of the ester groups to
amides were performed essentially as described in the lit-
HO
OH
O
N
O
N
d
erature.11–13 Reduction of the amides with LiAlH4
14,15
HO
OH
and subsequent cyclization with sulfamide with either
diisopropylamine or triethylamine delivered the seven-
membered cyclic scaffold. The scaffold was thereafter
alkylated using either benzyl bromide or 2-bromobenzyl
bromide to obtain compounds 8a and 8b, respectively.
Deprotection of 8a, using HCl/ether in methanol, deliv-
ered compound 9. Compound 8b served as the precursor
for the subsequent palladium catalyzed cross-coupling
reactions.
Scheme 2. Reagents and conditions: (a) i. styrene, DIEA, Pd(OAc)2,
PPh3, DMF, H2O, microwaves 150ꢁC, 20min, ii. HCl/ether and
methanol; (b) i. styrene, DIEA, Pd(OAc)2, PPh3, DMF, H2O,
microwaves 150ꢁC, 20min, ii. H2, Pd/C, ethyl acetate, iii. HCl/ether
and methanol; (c) i. benzyl zinc, PdCl2, DPPF, THF under N2, 80ꢁC,
88h, ii. HCl/ether and methanol; (d) i. phenyl boronic acid, Pd(OAc)2,
PPh3, Na2CO3, DME, ethanol, microwaves 150ꢁC, 5min, ii. HCl/ether
and methanol.
Compounds 10 and 11 were prepared using a microwave
mediated Heck coupling (150ꢁC, 5min) with styrene and
triethylamine as base.16,17 The double bond of 11 was
reduced using palladium on charcoal and hydrogen
gas. Finally, deprotection was achieved with HCl/ether
in methanol to deliver 10 and 11 in 53% and 28% yield,
respectively. Compound 12 was prepared using a Nigishi
coupling18 with benzyl zinc bromide (6equiv), palladium
chloride, and DPPF in a heating block at 80ꢁC.
After 24h, an additional 2equiv of benzyl zinc bromide
was added and heating continued for another 64h.
After deprotection, 51% of 12 was isolated. Compound
13 was prepared using a Suzuki–Miyaura coupling19,20
with palladium acetate, triphenylphosphine, 5equiv of
phenylboronic acid, and sodium carbonate as the
base. Microwave irradiation for 5min at 150ꢁC17
followed by deprotection afforded compound 13 in
53% yield.

758
A. Ax et al. / Bioorg. Med. Chem. 13 (2005) 755–764
A series of 15 substituted biaryl compounds were pre-
pared from 8b. A representative set of aryl boronic acids
and esters with desirable properties were selected. The
Suzuki–Miyaura reaction conditions were identical to
those described for the synthesis of compound 13 except
for compounds 25 and 26 where a slight modification of
the reaction times and temperature were required
(140ꢁC, 20min). All of the products from the Suzuki–
Miyaura couplings were purified only by extraction
and, in all but one case, provided pure compounds
according to NMR and elemental analysis.
5
4
3
2
1
0
-1
-2
-3
3. Discussion
-7
-6
-5
-4
-3
-2
-1
0
1
2
3
4
(a)
t[1]
To investigate whether it would be possible to fill all
four binding sites using only two substituents on the
scaffold, by attaching the P1/P10 substituents to the
P2/P20 sidechains, we designed four target compounds
with different spacer groups. All four compounds 10–
13 turned out to be about equally potent, with Ki values
of 2.0–3.0lM. A compound lacking ortho-substituents
(9) was also synthesized and tested for comparison. As
anticipated, it was found to be inactive. The most inter-
esting result from this series of compounds was the
activity and proposed binding mode for compound 13.
This biphenyl compound was modeled to adopt the least
optimal interaction with the enzyme but was shown to
have similar activity. This unanticipated finding cer-
tainly called into question the validity of the modeling
results. On the other hand, if the modeling could be
trusted and the activity of 13 has been achieved by only
filling two sites, we anticipated that adding groups,
which could reach the other two binding sites (S2 to
S1 and S10 to S20), would result in an improved inhibi-
tory potency.
PSA
0.6
0.5
0.4
0.3
0.2
0.1
0.0
HBD
HBA
MW
VOL
AREA
-0.1
-0.2
-0.3
-0.4
ClogP
0.0
0.1
0.2
0.3
0.4
0.5
(b)
p[1]
Figure 3. Results from the principal component analysis (PCA) for the
selected boronic acids and esters. In score plot (a), the building blocks
used are depicted (black dots) among the commercially available
building blocks (open diamond). Analyzing the loading plot (b), a
preference for lipophilic compounds with low molecular weight can be
seen, which corresponds to our design criterion.
A series of 15 biaryl derivatives of 13 were prepared
using Suzuki–Miyaura coupling. The aryl boronic acids
that were used in the coupling reactions were selected
based on those that were available in house. We aimed
for moderately lipophilic building blocks that would
produce inhibitors with a molecular weight below
700g/mol. To insure that the chosen reactants well rep-
resented the desired region of the chemical space, we
searched the Available Chemicals Directory (ACD)21
for boronic acids and esters. Seven physicochemical
descriptors were calculated for each reactant and a prin-
cipal component analysis was performed. A two-compo-
nent model representing 71% of the variance was
produced (Fig. 3). Inspection of the score and loading
plots shows that the selected boronic acids were spread
throughout the desired chemical space. We were there-
fore satisfied with our selection.
any conclusions regarding compounds with an ortho-
substituent (25 and 26) since one compound was a some-
what poorer inhibitor than 13 and the other, with a thio-
methyl group, exhibited the second best activity. The
thiophene derivative 27 was equipotent to phenyl deriv-
ative 13, which could be expected by bioisosterism.22
Notably, the benzofurane derivative 28 was seen to be
the most potent inhibitor. Modeling suggested that the
benzofurane oxygen might favorably interact with back-
bone NH of Gly 48 and the S3 subsite would be better
filled.
4. Conclusion
A series of HIV protease inhibitors utilizing an unusual
design has been synthesized. The P2/P20 substituents are
elongated and substituted with the P1/P10 sidechain in
an attempt to reach between the binding sites. The syn-
thesized inhibitors possess moderate activity with Ki val-
ues of 0.53–9.7lM. The concept has proven promising
though the potency of the parent, tetra-substituted
compound 2 has not yet been attained. The reason for
this may be that the binding mode of the investigated
The Ki values of 14–28 ranged between 0.53 and 9.7lM
(Table 2). Compounds 14 and 16, with lipophilic substit-
uents in the para-position, were more potent than the
parent biphenyl compound 13. When other groups were
introduced into the para-position (18, 20, 21, 22, and 24)
the activity dropped. Compounds with a meta-substitu-
ent (15, 17, 19, 23, and 24) have a similar or slightly
poorer activity compared to 13. It was difficult to draw

A. Ax et al. / Bioorg. Med. Chem. 13 (2005) 755–764
Table 2. Inhibitory activity of the biphenyl analogsa
759
O
N
O
N
S
R
R
HO
OH
Compd
R
Yield (%)
53
Ki (lM)
Compd
R
Yield (%)
65
Ki (lM)
O
13
2.5
21
5.3
O
S
O
14
15
16
60
63
60
1.5
2.7
1.9
22
23
24
26
60
57
5.2
7.4
9.4
NO2
O
O
17
18
19
20
56
62
52
50
3.2
5.3
2.4
9.7
25
26
27
28
54
68
45
31
5.1
1.1
2.7
0.53
O
S
F
CF3
S
O
O
a Compounds 14–28 were prepared as described for 13 (Scheme 2), except for 25 and 26 where a slight modification of time and temperature were
necessary (140ꢁC, 20min).
inhibitors are different from the ones suggested by the
modeling; this will be addressed in future investigations.
Considering the modeled differences in the occupation
of the S1 versus S10 subsites, we are also investigating
nonsymmetric derivatives, which may have improved
activity.
tease (PDB code 1ajv). The enzyme–ligand complex
was prepared using the protein preparation feature in
Maestro. Both of the catalytic aspartic acids were
protonated, as has been suggested for the cyclic ureas;24
this protonation pattern seems to be suitable for the
sulfamide class as well (unpublished results).
Each compound was first minimized (MacroModel
8.525) in the active site using the OPLS-AA force field26
and GB/SA27 solvation model. The torsions and bond
angles of the sulfamide ring were constrained to preserve
the conformation observed in all available X-ray com-
plexes for the tetra-substituted inhibitors. Inhibitors
5. Experimental
5.1. Modeling
The compounds were built in Maestro23 starting from
the crystal structure of 2 in complex with HIV-1 pro-
˚
were allowed to reposition and relax in a 6A shell from

760
A. Ax et al. / Bioorg. Med. Chem. 13 (2005) 755–764
the 2–enzyme complex (extended to full residues), which
and were within 0.4% of calculated values. Column
chromatography was preformed on silica gel 60,
0.040–0.063mm (E. Merck). Thin-layer chromatogra-
phy was preformed on precoated silica gel F-254 plates,
0.25mm (E. Merck) plates and visualized with UV light,
phosphomolybdic acid, or ninhydrin. Analytical RP-
LC/MS was performed on a Gilson HPLC system with
a Chromolith Performance RP-18e, 4.6 · 100mm col-
umn, with a Finnigan AQA quadropole mass spectro-
meter at a flow rate of 4mL/min (H2O/CH3CN/0.05%
HCOOH). All microwave reactions were carried out in
heavy-walled glass Smith process vials sealed with alum-
inum crimp caps fitted with a silicon septum. The micro-
wave heating was performed in a Smith SynthesizerTM
single-mode cavity producing continuous irradiation at
2450MHz (Biotage AB, Uppsala, Sweden). Reaction
mixtures were stirred during irradiation. The tempera-
ture, pressure, and irradiation power were monitored
during the course of the reaction. After the irradiation
was completed, the reaction tube was cooled with
high-pressure air until the temperature was below
39ꢁC. A Reacto-StationTM (STEM Electrothermal) was
used for conventional heating. Centrifugation was per-
formed using a Centra EC-4 (International Equipment
Company). Vacuum centrifugation was performed using
a SpeedVac Plus SC250DDA (Savant).
2
was restrained with a force constant of 100kJ/molA .
˚
Debug switch 17 was set to include constrained-atom
mutual interactions in the calculations. Additionally,
distance constraints were applied between the0hydroxyl
˚
groups and the carbonyl oxygen in Asp25/25 ( 1.0A,
2
˚
force constant 100kJ/molA ) in order to focus our
docking efforts to the active site.
The conformational preferences of compounds 9–13
were explored in the environment described above with
a 5000-step Monte Carlo search (MCMM) and a 2000-
step minimization at each step. Excluding the con-
strained seven-membered ring, the rotatable bonds were
varied. The lowest energy conformer of each of the four
compounds was subjected to a final (2000-step) minimi-
zation in the fully unrestrained enzyme; the sulfamide
ring conformation was maintained as described above
(Fig. 2).
5.2. Descriptor calculation and analysis
The ACD (Version 2003.3)21 was used to search for
boronic acids and esters to be used in the syntheses of
14–28. In total 986 were found. Logic filtration was used
to select boronic acids compatible with the chemistry to
be used. Thus, only activated boronic acids, that is, aryl,
benzyl, or vinyl, which did not contain halogens (besides
F), carboxyl, ester, or formyl groups, were used. Com-
pounds with more than one boronic acid function as
well as resin bound boronic acids were also excluded.
The boronic esters were ꢁhydrolyzedꢀ and duplicates
were removed to yield a total of 583 unique boronic
5.4.2. General procedure for the Suzuki–Miyaura cross-
coupling reactions. A Smith-vial was charged with 8b
(0.088mmol, 1equiv), Pd(OAc)2 (8.8lmol, 0,1equiv),
PPh3 (0.026mmol, 0.3equiv), boronic acid (0.44mmol,
5equiv), DME (480lL), ethanol (120lL), and 1M
Na2CO3 (300lL) and heated to 150ꢁC for 5min (except
for compound 25 and 26, which were heated to 140ꢁC
for 20min) using microwave irradiation. The reaction
mixture was diluted with petroleum ether (2mL) and
50% acetonitrile/water (1mL) and transferred to a cen-
trifuge tube, equipped with a filter. The phases were sep-
arated, the petroleum ether phase was washed with 50%
acetonitrile/water (1mL) and concentrated in vacuo. (In
some cases, the product precipitates as a red-brown oil,
which was then collected and washed together with the
petroleum ether phase.) The crude product was dis-
solved in methanol (1mL) and then HCl in ether
(1mL) was added. The reaction mixture was stirred
for 2h before the solvent was removed in vacuo. The
product was dissolved in acetonitrile (1mL), washed
with petroleum ether (2 · 1mL), dried by filtration
through MgSO4, and concentrated in vacuo. In some
cases, white crystals (salt) precipitate during extraction
or when the sample was dissolved for NMR. These were
then filtered off.
acids. Gasteiger–Huckel charges were applied and
¨
physicochemical descriptors were calculated using in-
house and standard SPL scripts (Sybyl 6.928). Size
descriptors (molecular weight, surface area, and vol-
ume), as well as ClogP, polar surface area (PSA), and
H-bonding (acceptors and donors) descriptors were
used. Finally, boronic acids that would result in an
inhibitor with molecular weight above 700g/mol were
excluded to result in 416 suitable reagents. Simca-P+
(version 10)29 was used to generate the principle compo-
nent analysis (PCA) with all settings kept at default. The
generated model consisted of two principal components
with a cumulative r2 of 0.71.
5.3. HIV protease inhibition
The HIV-1 protease was cloned and heterologously ex-
pressed in Esherichia coli and purified as described else-
where.30 The Ki values for the synthesized compounds
were determined by fluorometric assay31 (Tables 1 and
2).
5.4.3. (4S,5S)-O-Isopropylidene-1,2,7-thiadiazepine 1,1-di-
oxide (7). Compound 6 (2.1g, 0.011mol) was placed in
a Soxhlet thimble and extracted down to a refluxing sus-
pension of LiAlH4 (1.0g, 0.26mmol) in dry THF
(350mL). Refluxing was continued overnight and the
reaction mixture was then stirred at room temperature
for another 16h. Water (3mL) was added dropwise to
the reaction mixture followed by 15% NaOH (3mL)
and water (10mL). The solution was filtered through
Celite and K2CO3, the solid was washed with THF
5.4. Synthesis
1
5.4.1. General information. H and 13C NMR spectra
were recorded on a Jeol JNM-EX270 spectrometer at
270.2 and 67.8MHz, respectively. Chemical shifts are re-
ported as d values (ppm) indirectly referred to TMS by
the solvent residual signal. Elemental analyses were pre-
formed by Analytische Laboratorium Lindlar, Germany

A. Ax et al. / Bioorg. Med. Chem. 13 (2005) 755–764
761
and the combined filtrates were concentrated to afford
the crude diamine as a pale brown oil (1.34g, 77%).
The crude amine (1.34g, 7.6mmol) and an equimolar
quantity of sulfamide (0.73g, 7.62mmol) were dissolved
in toluene (100mL) and triethylamine (3mL). The reac-
tion mixture was refluxed at 120ꢁC overnight. The
reaction mixture was concentrated in vacuo and purified
using dry-flash, followed by column chromatography,
(86lL, 0.91mmol, 10equiv), DIEA (57lL, 0.055mmol,
6equiv), Pd(OAc)2 (2mg, 9.1lmol, 0.1equiv), PPh3
(7.2mg, 0.027mmol, 0.3equiv), DMF (1mL), and water
(200lL). The reaction was heated to 150ꢁC for 20min
using microwave irradiation. The two reactions were
combined, transferred to a centrifuge tube, equipped
with a filter, centrifuged to remove palladium, and were
then concentrated in vacuo. The crude product was
purified using column chromatography to yield >98%
(115mg) of white solid. This intermediate was used for
the synthesis of 10 and 11.
for a total yield of 20% (0.49g) of white solid:
24
D
1
½aꢀ +72.19 (c 0.525, CH3OH, 23ꢁC); IR (cmꢁ1): 3300
(NH); H NMR (270MHz, acetone-d6): 1.31 (s, 6H),
3.01 (m, 2H), 3.48 (m, 2H), 4.12 (m, 2H), 6.16 (s, 2H);
13C NMR (270MHz, acetone-d6): 27.1, 44.3, 80.0,
109.5. Anal. (C7H14N2O4S) C, H, N.
The intermediate (53.2mg, 0.086mmol, 1equiv) was
deprotected using HCl/ether in methanol. The reaction
mixture was concentrated in vacuo and the crude prod-
uct was purified using column chromatography to yield
5.4.4. (4S,5S)-2,7-Bisbenzyl-4,5-O-isopropylidene-1,2,7-
thiadiazepine 1,1-dioxide (8a). To a solution of 7
(50.0mg, 0.22mmol) in DMF (10mL), benzyl bromide
(192.4mg, 1.1mmol, 5equiv) and potassium carbonate
(310mg, 2.2mmol, 10equiv) were added. The reaction
mixture was left to stir overnight at room temperature.
The reaction mixture was concentrated in vacuo and
the product was purified using flash chromatography to
1
53% (26.2mg) of white solid: H NMR (270MHz, ace-
tone-d6): 2.97 (m, 2H), 3.37 (m, 4H), 4.12 (br s, 2H),
4.67 (d, J = 15.1, 2H), 4.89 (d, J = 15.1, 2H), 7.15–7.43
(m, 14H), 7.49 (m, 2H), 7.72 (m, 4H), 7.81 (m, 2H);
13C NMR (270MHz, acetone-d6): 48.5, 51.1, 73.1,
125.9, 126.4, 127.8, 128.5, 128.5, 129.1, 129.4, 130.8,
131.6, 134.6, 137.7, 138.4. Anal. (C34H34N2O4S) C, H,
N.
1
yield 51% (45.8mg) of white solid: H NMR (270MHz,
CDCl3): 1.28 (s, 6H), 2.92 (m, 2H), 3.37 (m, 2H), 4.11
(m, 2H), 4.28 (d, J = 14.52, 2H), 4.40 (d, J = 14.52,
2H), 7.17–7.23 (m, 10H); 13C NMR (270MHz, ace-
tone-d6): 29.7, 49.0, 55.3, 78.1, 110.1, 128.6, 129.1,
129.5, 137.7. Anal. (C21H26N2O4S) C, H, N.
5.4.8. (4S,5S)-2,7-Bis[2-(2-phenylethyl)benzyl]-4,5-dihydr-
oxy-1,2,7-thiadiazepine 1,1-dioxide (11). Compound 11
was prepared using the intermediate described in the
synthesis of 10. The intermediate (61.9mg, 0.10mmol,
1equiv) was hydrogenated to remove the double bond
using H2 and palladium on charcoal suspended in ethyl
acetate. The reaction mixture was stirred overnight. The
reaction mixture was filtered and concentrated in vacuo.
The protecting group was cleaved using HCl/ether and
methanol and the crude product was purified using col-
umn chromatography to yield 28% (16.4mg) of white
5.4.5. (4S,5S)-2,7-Bis(2-bromobenzyl)-4,5-O-isopropylid-
ene-1,2,7-thiadiazepine 1,1-dioxide (8b). To a solution of
7 (1.6g, 7.5mmol) in DMF (100mL), 2-bromobenzyl-
bromide (5.6g, 0.023mol, 3equiv) and potassium car-
bonate (5.2g, 0.037mol, 5equiv) were added. The
reaction mixture was left to stir overnight at room
temperature. The reaction mixture was concentrated in
1
solid: H NMR (270MHz, acetone-d6): 2.01 (m, 2H),
vacuo and the product was purified using flash chroma-
2.74–3.20 (m, 10H), 3.23 (m, 2H), 3.36 (m, 2H), 4.31
(d, J = 15.18, 2H), 4.48 (d, J = 15.18, 2H), 7.07–7.25
(m, 16H), 7.31 (m, 2H); 13C NMR (270MHz, acetone-
d6): 34.1, 37.6, 47.6, 50.3, 72.7, 126.1, 126.8, 128.4,
128.5, 128.8, 129.2, 130.1, 133.6, 140.5, 141.5. Anal.
(C34H38N2O4S) C, H, N.
24
D
tography to yield 99% (4.2g) of white solid: ½aꢀ 51.35
1
(c 0.534, CHCl3, 22ꢁC); H NMR (270MHz, CDCl3):
1.31 (s, 6H), 3.05 (m, 2H), 3.42 (m, 2H), 4.23 (m, 2H),
4.39 (d, J = 15.51, 2H), 4.55 (d, J = 15.51, 2H), 7.10
(m, 2H), 7.27 (m, 2H), 7.43 (m, 2H), 7.49 (m, 2H); 13C
NMR (270MHz, CDCl3): 27.2, 49.3, 54.9, 77.7, 110.1,
123.8, 128.1, 129.7, 130.4, 133.2, 135.3. Anal.
(C21H24Br2N2O4S) C, H, N.
5.4.9. (4S,5S)-2,7-Bis[2-(phenylmethyl)benzyl]-4,5-dihydr-
oxy-1,2,7-thiadiazepine 1,1-dioxide (12). Compound 8b
(50.0mg, 0.089mmol, 1equiv), palladium chloride
(1.1mg, 6.20lmol, 0.07equiv), DPPF (1.5mg, 2.7lmol,
0.03equiv), and THF (2mL) were placed in a dry vial
and flushed with nitrogen gas. After a few minutes, ben-
zyl zinc bromide (1.1mL, 0.053mmol, 6equiv) was
added to the reaction mixture and the vial was placed
in an 80ꢁC heating block and stirred overnight. Another
2equiv of benzyl zinc was added and the reaction mix-
ture was heated an additional 64h. The reaction mixture
was diluted with petroleum ether (2mL) and 50% aceto-
nitrile/water (1mL) and then transferred to a centrifuge
tube, equipped with a filter and the phases were sepa-
rated. The acetonitrile/water phase was extracted two
times with dichloromethane and the organic phase was
concentrated in vacuo. The crude product was dissolved
in methanol (2.5mL) and 2M HCl in ether (1mL) was
added. The reaction mixture was left to stir at room
5.4.6. (4S,5S)-2,7-Bisbenzyl-4,5-dihydroxy-1,2,7-thiadi-
azepine 1,1-dioxide (9). To a solution of 8a (35.6mg,
0.088mmol) in methanol (5mL) was added 2M HCl in
ether (2mL) and then the reaction mixture was stirred
for 3h. The reaction mixture was concentrated and the
product was purified by flash chromatography to yield
72% (23.1mg) of light-yellow solid: 1H NMR
(270MHz, acetone-d6): 3.13 (m, 2H), 3.32 (m, 2H),
3.58 (m, 2H), 4.28 (br s, 2H), 4.53 (d, J = 15.84, 2H),
4.70 (d, J = 15.84, 2H), 7.26–7.49 (m, 10H); 13C NMR
(270MHz, CD3OD): 49.1, 53.6, 73.1, 129.0, 129.3,
129.9, 138.5. Anal. (C18H22N2O4S) C, H, N.
5.4.7. (4S,5S)-2,7-Bis(2-styrylbenzyl)-4,5-dihydroxy-1,2,7-thia-
diazepine 1,1-dioxide (10). Two Smith-vials were each
charged with 8b (51mg, 0.091mmol, 1equiv), styrene

762
A. Ax et al. / Bioorg. Med. Chem. 13 (2005) 755–764
temperature overnight. The reaction mixture was con-
centrated in vacuo and the product was purified using
column chromatography to yield 51% (24.7mg): 1H
NMR (270MHz, acetone-d6): 3.07 (m, 2H), 3.31 (m,
2H), 3.54 (m, 2H), 4.14 (s, 4H), 4.25 (m, 2H), 4.57 (d,
J = 15.84, 2H), 4.69 (d, J = 15.84, 2H), 7.14–7.22 (m,
8H), 7.22–7.31 (m, 8H), 7.51 (m, 2H); 13C NMR
(270MHz, acetone-d6): 38.6, 48.5, 50.6, 73.0, 126.8,
127.4, 128.5, 129.2, 129.5, 129.7, 131.4, 135.7, 140.3,
141.3; Anal. (C32H34N2O4S) C, H, N.
1.33 (m, 4H), 1.60 (m, 4H), 2.62 (m, 4H), 2.84 (m,
2H), 3.10 (m, 2H), 3.26 (m, 2H), 4.07 (br s, 2H), 4.41
(d, J = 16.50, 2H), 4.63 (d, J = 16.50, 2H), 7.2 (m,
8H), 3.32 (m, 6H), 7.59 (m, 2H); 13C NMR (270MHz,
acetone-d6): 14.2, 23.0, 34.4, 35.8, 48.7, 50.8, 72.8,
128.0, 128.4, 128.6, 129.2, 129.8, 130.8, 135.4, 138.7,
142.6. Anal. (C38H36N2O4S) C, H, N.
5.4.14. (4S,5S)-2,7-Bis[2-(3-isopropylphenyl)benzyl]-4,5-
dihydroxy-1,2,7-thiadiazepine 1,1-dioxide (17). Com-
pound 17 was prepared according to the general proce-
1
5.4.10. (4S,5S)-2,7-Bis(2-phenylbenzyl)-4,5-dihydroxy-
1,2,7-thiadiazepine 1,1-dioxide (13). A Smith-vial was
charged with 8b (49.3mg, 0.088mmol, 1equiv),
Pd(OAc)2 (2.0mg, 8.8lmol, 0.1equiv), PPh3 (8.0mg,
0.026mmol, 0.3equiv), phenyl boronic acid (53.7mg,
0.44mmol, 5equiv), DME (480lL), ethanol (120lL),
and 1M Na2CO3 (300lL) and heated to 150ꢁC for
5min. The reaction mixture was transferred to a centri-
fuge tube, equipped with a filter. Solids were filtered off
and the reaction mixture was concentrated in vacuo.
The crude product was purified using column chroma-
tography to yield 53% (24mg) of white solid. The com-
pound was later remade using the general procedure for
Suzuki–Miyaura cross-coupling described above (81%
dure (Section 5.4.2) in 56% yield (29.7mg): H NMR
(270MHz, acetone-d6): 1.22 (d, J = 6.93, 12H), 2.83
(m, 2H), 2.92 (m, J = 6.93, 2H), 3.08 (m, 2H), 3.23 (m,
2H), 4.08 (br s, 2H), 4.40 (d, J = 16.17, 2H), 4.59 (d,
J = 16.17, 2H), 7.08 (m, 2H), 7.11–7.4 (m, 12H) 7.59
(m, 2H); 13C NMR (270MHz, acetone-d6): 24.2, 30.1,
34.7, 48.8, 50.7, 72.9, 126.9, 127.4, 128.0, 128.1, 128.5,
128.8, 129.2, 130.7, 135.5, 141.4, 143.0, 149. Anal.
(C36H42N2O4S) C, H, N.
5.4.15. (4S,5S)-2,7-Bis[2-(4-fluorophenyl)benzyl]-4,5-di-
hydroxy-1,2,7-thiadiazepine 1,1-dioxide (18). Compound
18 was prepared according to the general procedure
(Section 5.4.2) in 62% yield (30.0mg): 1H NMR
(270MHz, acetone-d6): 2.90 (m, 2H), 1.13 (m, 2H),
3.33 (m, 2H), 4.26 (s, 2H), 4.44 (d, J = 16.50, 2H),
4.66 (d, J = 16.50, 2H), 7.15–7.28 (m, 6H), 7.28–7.47
(m, 8H), 7.63 (m, 2H); 13C NMR (270MHz, acetone-
d6): 48.5, 50.8, 72.6, 115.9 (d, J = 21.97), 128.1, 128.7,
128.8, 130.9, 131.9 (d, J = 8.54), 135.6, 137.6 (d,
J = 3.66), 141.5, 161.5 (d, J = 244.14). Anal.
(C30H28F2N2O4S) C, H, N.
1
yield (36.7mg) the product was >90% pure): H NMR
(270MHz, acetone-d6): 2.86 (m, 2H), 3.14 (m, 2H),
3.29 (m, 2H), 4.15 (m, 2H), 4.43 (d, J = 16.50, 2H),
4.67 (d, J = 16.50, 2H), 7.22 (m, 2H), 7.27–7.49 (m,
14H), 7.64 (m, 2H); 13C NMR (270MHz, CDCl3):
47.5, 49.7, 71.7, 127.3, 127.4, 128.0, 128.2, 128.2,
128.3, 129.1, 130.0, 133.5, 140.3, 141.7. Anal.
(C30H30N2O4S) C, H, N.
5.4.11. (4S,5S)-2,7-Bis[2-(4-methylphenyl)benzyl]-4,5-di-
hydroxy-1,2,7-thiadiazepine 1,1-dioxide (14). Compound
14 was prepared according to the general procedure
(Section 5.4.2) in 60% yield (28.6mg): 1H NMR
(270MHz, acetone-d6): 2.87 (m, 2H), 3.11 (m, 2H),
3.26 (m, 2H), 4.43 (d, J = 16.50, 2H), 4.68 (d,
J = 16.50, 2H), 7.15 (m, 4H), 7.20 (m, 6H), 7.29 (m,
2H), 7.36 (m, 2H), 7.59 (m, 2H); 13C NMR (270MHz,
acetone-d6): 21.1, 48.7, 50.7, 72.7, 127.9, 128.4, 128.4,
129.8, 129.8, 130.8, 135.4, 137.6, 138.4, 142.5. Anal.
(C32H34N2O4S) C, H, N.
5.4.16. (4S,5S)-2,7-Bis{2-[(3-trifluoromethyl)phenyl]benz-
yl}-4,5-dihydroxy-1,2,7-thiadiazepine 1,1-dioxide (19).
Compound 19 was prepared according to the general
procedure (Section 5.4.2) in 52% yield (29.5mg): 1H
NMR (270MHz, acetone-d6): 2.99 (m, 2H), 3.12 (m,
2H), 2.36 (m, 2H), 4.26 (s, 2H), 4.39 (d, J = 16.50,
2H), 4.55 (d, J = 16.50, 2H), 7.29 (m, 2H), 7.44 (m,
4H), 7.70 (m, 10H); 13C NMR (270MHz, CDCl3):
48.5, 50.9, 72.8, 124.8 (q, J = 3.66), 126.5 (q, J = 3.66),
128.2, 129.0, 129.2, 130.1, 130.9, 130.9 (q, J = 32),
133.9, 135.6, 141.0, 142.5. Anal. (C32H28F6N2O4S) C,
H, N.
5.4.12. (4S,5S)-2,7-Bis[2-(3-methylphenyl)benzyl]-4,5-di-
hydroxy-1,2,7-thiadiazepine 1,1-dioxide (15). Compound
15 was prepared according to the general procedure
(Section 5.4.2) in 63% yield (30.1mg): 1H NMR
(270MHz, acetone-d6): 2.36 (s, 6H), 2.89 (m, 2H), 3.17
(m, 2H), 3.29 (m, 2H), 4.1 (br s, 2H), 4.40 (d, J = 16.4,
2H), 4.61 (d, J = 16.4, 2H), 7.19 (m, 8H), 7.38 (m,
6H), 7.64 (m, 2H); 13C NMR (270MHz, acetone-d6):
20.2, 48.6, 50.6, 72.8, 126.9, 127.9, 128.5, 128.6, 128.7,
129.0, 129.4, 130.6, 130.7, 135.3, 138.7, 141.4, 142.8.
Anal. (C32H34N2O4S) C, H, N.
5.4.17. (4S,5S)-2,7-Bis[2-(4-methoxyphenyl)benzyl]-4,5-di-
hydroxy-1,2,7-thiadiazepine 1,1-dioxide (20). Compound
20 was prepared according to the general procedure
(Section 5.4.2) in 50% yield (25.3mg): 1H NMR
(270MHz, acetone-d6): 2.68 (m, 2H), 3.13 (m, 2H),
3.76–3.90 (m, 10H), 4.28 (d, J = 15.01, 2H), 4.41 (d,
J = 15.01, 2H), 7.02 (m, 4H), 7.23 (m, 6H), 7.32–7.43
(m, 4H), 7.52 (m, 2H); 13C NMR (270MHz, acetone-
d6): 49.1, 52.3, 55.5, 77.8, 114.6, 128.27, 128.29, 129.5,
131.0, 131.2, 133.5, 134.8, 142.6, 159.9. Anal.
(C32H34N2O6S) C, H, N.
5.4.13.
(4S,5S)-2,7-Bis[2-(4-n-butylphenyl)benzyl]-4,5-
dihydroxy-1,2,7-thiadiazepine 1,1-dioxide (16). Com-
pound 16 was prepared according to the general proce-
dure (Section 5.4.2) to a yield of 60% (33.3mg): 1H
NMR (270MHz, acetone-d6): 0.89 (t, J = 7.26, 6H),
5.4.18. (4S,5S)-2,7-Bis[2-(4-acetylphenyl)benzyl]-4,5-di-
hydroxy-1,2,7-thiadiazepine 1,1-dioxide (21). Compound
21 was prepared according to the general procedure
(Section 5.4.2) in 65% yield (34.3mg): 1H NMR

A. Ax et al. / Bioorg. Med. Chem. 13 (2005) 755–764
763
1
(270MHz, acetone-d6): 2.91 (m, 2H), 3.10 (m, 2H), 3.32
(m, 2H), 4.30 (br s, 2H), 4.30 (d, J = 16.50, 2H), 4.69 (d,
J = 16.50, 2H), 7.25 (m, 2H), 7.30–7.51 (m, 8H), 7.66 (m,
2H), 8.07 (m, 4H); 13C NMR (270MHz, acetone-d6):
26.9, 48.6, 50.9, 72.7, 128.3, 128.9, 129.2, 129.5, 130.4,
130.7, 135.6, 137.0, 141.7, 146.2, 197.8. Anal.
(C34H34N2O6S) C, H, N.
dure (Section 5.4.2) in 68% yield (34.1mg): H NMR
(270MHz, CDCl3): 2.25 (m, 6H), 2.26–3.29 (m, 6H),
3.68–4.49 (m, 4H), 6.93–7.62 (m, 16H) OH-peak not vis-
ible; 13C NMR (270MHz, CDCl3): (isomers in spectra;
major peaks indicated when possible) 15.0 (major),
15.6, 47.2, 47.7 (major), 48.2, 49.5, 51.5, 51.7 (major),
52.1, 71.5, 71.8 (major), 72.1, 109.1, 123.8, 123.9 (ma-
jor), 124.4 (major), 124.8, 126.5, 127.6 (major), 127.8,
128.6 (major), 128.4 (major), 128.7 (major), 129.1,
129.5, 129.7, 130.0 (major), 134.2 (major) 134.3, 134.8
(major), 138.1 (major), 139.2, 139.5 (major). Anal.
(C32H34N2O4S3) C, H, N.
5.4.19. (4S,5S)-2,7-Bis{2-[(4-ethylsulfonyl)phenyl]benz-
yl}-4,5-dihydroxy-1,2,7-thiadiazepine 1,1-dioxide (22).
Compound 22 was prepared according to the general
procedure (Section 5.4.2) in 26% yield (16.1mg): 1H
NMR (270MHz, acetone-d6): 1.14 (t, J = 7.26, 6H),
2.79 (m, 2H), 3.98 (m, 2H), 3.15 (q, J = 7.26, 4H), 2.14
(m, 2H), 4.09 (s, 2H), 4.38 (d, J = 16.17, 2H), 4.58 (d,
J = 16.17, 2H), 7.2 (m, 2H), 7.34 (m, 2H), 7.40 (m,
2H), 7.55 (m, 4H), 7.89 (m, 4H); 13C NMR (270MHz,
acetone-d6): 7.7, 48.3, 50.8, 72.7, 128.4, 129.0, 129.4,
129.5, 130.7, 130.9, 135.4, 138.9, 141.0, 146.8. Anal.
(C34H38N2O8S3) C, H, N.
5.4.24. (4S,5S)-2,7-Bis[2-(3-thienyl)benzyl]-4,5-dihydr-
oxy-1,2,7-thiadiazepine 1,1-dioxide (27). Compound 27
was prepared according to the general procedure (Sec-
1
tion 5.4.2) in 45% yield (20.9mg): H NMR (270MHz,
acetone-d6): 2.90 (m, 2H), 3.21 (m, 2H), 3.40 (m, 2H),
4.26 (br s, 2H), 4.55 (d, J = 16.50, 2H), 4.77 (d,
J = 16.50, 2H), 7.21 (m, 2H), 7.25–7.52 (m, 8H), 7.57
(m, 2H), 7.65 (m, 2H); 13C NMR (270MHz, acetone-
d6): 48.7, 50.9, 72.8, 124.1, 126.6, 127.1, 128.0, 128.6,
128.6, 129.7, 130.8, 135.9, 137.2, 141.5. Anal.
(C28H26N2O4S3) C, H, N.
5.4.20. (4S,5S)-2,7-Bis[2-(3-nitrophenyl)benzyl]-4,5-dihydr-
oxy-1,2,7-thiadiazepine 1,1-dioxide (23). Compound 23
was prepared according to the general procedure (Sec-
1
tion 5.4.2) in 60% yield (31.9mg): H NMR (270MHz,
acetone-d6): 2.28 (m, 4H), 3.34 (m, 2H), 4.48 (d,
J = 16.17, 2H), 4.67 (d, J = 16.17, 2H), 4.24 (s, 2H),
7.30 (m, 2H), 7.35–7.51 (m, 4H), 7.56–7.88 (m, 6H),
8.11–8.29 (m, 4H); 13C NMR (270MHz, acetone-d6):
48.3, 50.9, 72.7, 122.9, 124.0, 124.6, 128.4, 129.4,
129.5, 130.3, 130.8, 135.6, 136.4, 140.3, 143.0, 149.0.
Anal. (C30H28N4O8S) C, H, N.
5.4.25. (4S,5S)-2,7-Bis[2-(benzo[b]furan-2-yl)benzyl]-4,5-
dihydroxy-1,2,7-thiadiazepine 1,1-dioxide (28). Com-
pound 28 was prepared according to the general proce-
1
dure (Section 5.4.2) in 31% yield (16.1mg); H NMR
(270MHz, acetone-d6): 3.19 (m, 2H), 3.43 (m, 2H),
3.67 (m, 2H), 4.22 (m, 2H), 4.94 (d, J = 16.99, 2H),
5.17 (d, J = 16.99, 2H), 7.17 (m, 2H), 7.24–7.40 (m,
4H), 7.40–7.58 (m, 4H), 7.58–7.75 (m, 4H), 7.75–7.90
(m, 4H); 13C NMR (270MHz, acetone-d6): 48.7, 51.7,
72.8, 106.8, 112.0, 122.0, 124.0, 125.4, 128.4, 128.9,
129.8, 129.9, 130.1, 130.2, 136.3, 155.5, 155.6. Anal.
(C34H30N2O6S) C, H, N.
5.4.21. (4S,5S)-2,7-Bis[2-(3,4-dimethoxyphenyl)benzyl]-
4,5-dihydroxy-1,2,7-thiadiazepine 1,1-dioxide (24). Com-
pound 24 was prepared according to the general proce-
1
dure (Section 5.4.2) in 57% yield (31.9mg): H NMR
(270MHz, acetone-d6): 2.87 (m, 2H), 3.15 (m, 2H),
3.31 (m, 2H), 4.24 (s, 2H), 4.47 (d, J = 16.50, 2H),
4.69 (d, J = 16.50, 2H), 6.82 (m, 2H), 6.9 (s, 2H), 7.00
(m, 2H), 7.21 (m, 2H), 7.35 (m, 4H), 7.61 (m, 2H); 13C
NMR (270MHz, acetone-d6): 48.8, 50.8, 56.1, 56.2,
72.8, 112.5, 113.8, 122.2, 127.9, 128.3, 128.7, 131.0,
134.1, 135.8, 142.8, 149.7, 150.1. Anal. (C34H38N2O8S)
C, H, N.
Acknowledgements
We thank the Swedish Foundation for Strategic Re-
search (SSF) and the Swedish Research Council (VR)
for financial support. We also thank Biotage AB for
´
providing a Smith Synthesizer. Dr. Johan Hulten is
acknowledged for early work in this project.
5.4.22. (4S,5S)-2,7-Bis[2-(2-methoxyphenyl)benzyl]-4,5-di-
hydroxy-1,2,7-thiadiazepine 1,1-dioxide (25). Compound
25 was prepared according to the general procedure (Sec-
tion 5.4.2) in 54% yield (27.5mg) the product was >80%
pure after extraction but was further purified by column
chromatography before being sent to elemental analysis:
1H NMR (270MHz, acetone-d6): 2.76–38 (m, 8H), 3.77
(m, 4H), 4.11 (m, 2H), 4.37 (br s, 2H), 6.98–7.20 (m,
8H), 7.29–7.47 (m, 6H), 7.61 (m, 2H); 13C NMR
(270MHz, acetone-d6): (isomers in the spectra, major
peaks reported) 48.6, 50.6, 55.6, 72.7, 111.6, 121.4,
127.7, 127.9, 128.4, 130.0, 131.1, 131.1, 131.4, 131.7,
136.6, 139.1. Anal. (C32H34N2O6SÆ2H2O) C, H, N.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmc.2004.10.042.
References and notes
1. UNAIDS. 2004 Report on the Global AIDS Epidemic,
4th Global Report; Joint United Nations Programme on
HIV/AIDS, World Health Organization; Geneva, 2004.
2. Jacobson, M. A.; French, M. AIDS 1998, 12 Suppl A,
S157.
5.4.23. (4S,5S)-2,7-Bis{2-[2-(methylthio)phenyl]benzyl}-
4,5-dihydroxy-1,2,7-thiadiazepine 1,1-dioxide (26). Com-
pound 26 was prepared according to the general proce-
3. Collier, A. C.; Coombs, R. W.; Schoenfeld, D. A.; Bassett,
R.; Baruch, A.; Corey, L. Antiviral Res. 1996, 29, 99.

764
A. Ax et al. / Bioorg. Med. Chem. 13 (2005) 755–764
4. Clavel, F.; Hance, A. J. New Engl. J. Med. 2004, 350,
1023.
5. Lam, P. Y. S.; Jadhav, P. K.; Eyermann, C. J.; Hodge, C.
N.; Ru, Y.; Bacheler, L. T.; Meek, O. M. J.; Rayner, M.
M. Science 1994, 263, 380.
all, T., Cacci, S., Hayashi, T. et al., Eds.; Wiley-Inter-
science: New York, 2002; Vol. I, pp 229–248.
19. Kotha, S.; Lahiri, K.; Kashinath, D. Tetrahedron 2002, 58,
9633.
20. Suzuki, A. In Handbook of Organopalladium for Organic
Synthesis; Negishi, E., de Meijere, A., Ba¨ckvall, J. E.,
Cacci, S., Hayashi, T. et al., Eds.; Wiley-Interscience:
New York, 2002; Vol. I, pp 249–262.
´
6. Hulten, J.; Bonham, N. M.; Nillroth, U.; Hansson, T.;
Zuccarello, G.; Bouzide, A.; Aqvist, J.; Classon, B.;
˚
´
Danielson, U. H.; Karlen, A.; Kvarnsto¨m, I.; Samuelsson,
B.; Hallberg, A. J. Med. Chem. 1997, 40, 885.
21. Available Chemicals Directory; MDL Information Sys-
tems Inc., 14600 Catalina St., San Leandro, CA 94577,
USA, www.mdli.com.
´
7. Hulten, J.; Andersson, H. O.; Schaal, W.; Danielson, H.
U.; Classon, B.; Kvarnstro¨m, I.; Karlen, A.; Unge, T.;
´
Samuelsson, B.; Hallberg, A. J. Med. Chem. 1999, 42,
4054.
¨
8. Ba¨ckbro, K.; Lo¨wgren, S.; Osterlund, K.; Atepo, J.; Unge,
22. Wermuth, C. G. In The Practice of Medicinal Chemistry;
Wermuth, C. G., Ed.; Academic Press: London, 2000; pp
203–237.
´
T.; Hulten, J.; Bonham, N. M.; Schaal, W.; Karlen, A.;
Hallberg, A. J. Med. Chem. 1997, 40, 898.
´
23. Maestro, 6.0 ed.; Schro¨dinger: 1500 S.W. First Avenue,
USA;
Suite
www.schrodinger.com.
1180
Portland,
OR
97201,
´
9. Schaal, W.; Karlsson, A.; Ahlsen, G.; Lindberg, J.;
Andersson, H. O.; Danielson, U. H.; Classon, B.; Unge,
24. Yamazaki, T.; Nicholson, L. K.; Wingfield, P.; Stahl, S. J.;
Kaufman, J. D.; Eyermann, C. J.; Hodge, C. N.; Lam, P.
Y. S.; Torchia, D. A., et al. J. Am. Chem. Soc. 1994, 116,
10791.
25. MacroModel, 8.5 ed.; Schro¨dinger: 1500 S.W. First
Avenue, Suite 1180 Portland, OR 97201, USA;
www.schrodinger.com.
´
T.; Samuelsson, B.; Hulten, J.; Hallberg, A.; Karlen, A. J.
Med. Chem. 2001, 44, 155.
´
10. Kaldor, S. W.; Appelt, K.; Fritz, J.; Hammond, M.;
Crowell, T. A.; Baxter, A. J.; Hatch, S. D.; Wiskerchen,
M. A.; Muesing, M. A. Bioorg. Med. Chem. Lett. 1995, 5,
715.
11. Carroll, F. I. J. Org. Chem. 1966, 31, 366.
12. Carmack, M.; Kelley, C. J. J. Org. Chem. 1968, 33, 2171.
13. Briggs, M. A.; Haines, A. H.; Jones, H. F. J. Chem. Soc.
Perkin Trans. 1 1985, 795.
26. Jorgensen, W. L.; Maxwell, D. S.; Tirado-Rives, J. J. Am.
Chem. Soc. 1996, 118, 11225.
27. Qui, D.; Shenkin, P. S.; Hollinger, F. P.; Still, W. C. J.
Phys. Chem. 1997, 101, 3005.
14. Shainyan, B. A.; Ustinov, M. V.; Belꢀskii, V. K.; Nindak-
ova, L. O. Russ. J. Org. Chem. 2002, 38, 104.
15. Haines, A. H.; Morley, C.; Murrer, B. A. J. Med. Chem.
1989, 32, 742.
28. Sybyl, 6.9 ed.; Tripos Inc.: 1699 South Hanley Rd., St.
Louis, MO 63144, USA, www.tripos.com.
29. Simca-P+, 10.0.2.0, Sept 30, 2002 ed.; Umetrics: Box 7960,
˚
SE-907 19 Umea, Sweden, www.umetrics.com.
16. Hallberg, A.; Larhed, M. In Handbook of Organopalla-
dium Chemistry for Organic Synthesis; Negishi, E., de
Meijere, A., Ba¨ckvall, J. E., Cacci, S., Hayashi, T., et al.,
Eds.; Wiley-Interscience: New York, 2002; Vol. I, pp
1133–1178.
30. Danielson, U. H.; Lindgren, M. T.; Markgren, P.-O.;
Nillroth, U. Adv. Exp. Med. Biol. 1998, 463, 99.
31. Nillroth, U.; Vrang, L.; Markgren, P.-O.; Hulten, J.;
Hallberg, A.; Danielson, U. H. Antimicrob. Agents Chem-
other. 1997, 41, 2383.
17. Larhed, M.; Hallberg, A. J. Org. Chem. 1996, 61, 9582.
18. Negishi, E. In Handbook of Organopalladium Chemistry
for Organic Synthesis; Negishi, E., de Meijere, A., Ba¨ckv-
32. Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G.
S.; Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. J.
Comput. Chem. 2004, 25, 1605, www.cgl.ucsf.edu/chimera.