Structure-based design, synthesis, and biological evaluation of novel pyrrolyl aryl sulfones: HIV-1 non-nucleoside reverse transcriptase inhibitors active at nanomolar concentrations

Article abstract of DOI:10.1021/jm9901125

Pyrrolyl aryl sulfones (PASs) have been recently reported as a new class of human immunodeficiency virus type 1 (HIV-1) reverse transcriptase (RT) inhibitors acting at the non-nucleoside binding site of this enzyme (Artico, M.; et al. J. Med. Chem. 1996, 39, 522-530). Compound 3, the most potent inhibitor within the series (EC₅₀ = 0.14/μM, IC₅₀ = 0.4/μM, and SI > 1429), was then selected as a lead compound for a synthetic project based on molecular modeling studies. Using the three-dimensional structure of RT cocrystallized with the α-APA derivative R95845, we derived a model of the RT/3 complex by taking into account previously developed structure-activity relationships. Inspection of this model and docking calculations on virtual compounds prompted the design of novel PAS derivatives and related analogues. Our computational approach proved to be effective in making qualitative predictions, that is in discriminating active versus inactive compounds. Among the compounds synthesized and tested, 20 was the most active one, with EC₅₀ = 0.045/μM, IC₅₀ = 0.05/μM, and SI = 5333. Compared with the lead 3, these values represent a 3- and 8-fold improvement in the cell-based and enzyme assays, respectively, together with the highest selectivity achieved so far in the PAS series.

Full text of DOI:10.1021/jm9901125

1886
J . Med. Chem. 2000, 43, 1886-1891
Brief Articles
Str u ctu r e-Ba sed Design , Syn th esis, a n d Biologica l Eva lu a tion of Novel P yr r olyl
Ar yl Su lfon es: HIV-1 Non -Nu cleosid e Rever se Tr a n scr ip ta se In h ibitor s Active
a t Na n om ola r Con cen tr a tion s
Marino Artico,*,# Romano Silvestri,# Eugenia Pagnozzi,# Biancamaria Bruno,# Ettore Novellino,‡
Giovanni Greco,‡ Silvio Massa,§ Alessandro Ettorre,† Anna Giulia Loi,| Franca Scintu,| and Paolo La Colla*,|
Istituto Pasteur-Fondazione Cenci Bolognetti, Dipartimento di Studi Farmaceutici, Universita` degli Studi di Roma “La
Sapienza”, P.le A. Moro 5, BOX 36 - ROMA 62, I-00185 Roma, Italy, Dipartimento di Chimica Farmaceutica e Tossicologica,
Universita` degli Studi di Napoli “Federico II”, Via D. Montesano 49, I-80131 Napoli, Italy, Dipartimento Farmaco Chimico
Tecnologico, Universita` degli Studi di Siena, Banchi di Sotto 55, I-53100 Siena, Italy, Menarini Ricerche SpA, Via T. Speri 10,
00040 Pomezia, Roma, Italy, and Dipartimento di Biologia Sperimentale, Sezione di Microbiologia, Universita` degli Studi di
Cagliari, Cittadella Universitaria, SS 554, I-09042 Monserrato (Cagliari), Italy
Received March 12, 1999
Pyrrolyl aryl sulfones (PASs) have been recently reported as a new class of human immuno-
deficiency virus type 1 (HIV-1) reverse transcriptase (RT) inhibitors acting at the non-nucleoside
binding site of this enzyme (Artico, M.; et al. J . Med. Chem. 1996, 39, 522-530). Compound 3,
the most potent inhibitor within the series (EC50 ) 0.14 µM, IC50 ) 0.4 µM, and SI > 1429),
was then selected as a lead compound for a synthetic project based on molecular modeling
studies. Using the three-dimensional structure of RT cocrystallized with the R-APA derivative
R95845, we derived a model of the RT/3 complex by taking into account previously developed
structure-activity relationships. Inspection of this model and docking calculations on virtual
compounds prompted the design of novel PAS derivatives and related analogues. Our
computational approach proved to be effective in making qualitative predictions, that is in
discriminating active versus inactive compounds. Among the compounds synthesized and tested,
20 was the most active one, with EC50 ) 0.045 µM, IC50 ) 0.05 µM, and SI ) 5333. Compared
with the lead 3, these values represent a 3- and 8-fold improvement in the cell-based and enzyme
assays, respectively, together with the highest selectivity achieved so far in the PAS series.
Ch a r t 1. Sulfone Anti-HIV-1 Agents
In tr od u ction
Anti-HIV activity of sulfones was discovered at the
National Cancer Institute (NCI) in 1993 following a
large-scale drug-screening program on synthetic and
natural products from a wide variety of sources. Mc-
Mahon et al. reported the inhibition of HIV-1 reverse
transcription by various diaryl sulfones and selected
2-nitrophenyl phenyl sulfone (1, NPPS; Chart 1), a
potent inhibitor of reverse transcriptase (RT), as a lead
compound for further investigations.1 At the same time,
screening of the drug repositories of Merck Research
Laboratories led to the identification of 2-[(phenylsul-
finyl)methyl]-3-(phenylthio)indole as a potent and spe-
cific inhibitor of HIV-1 RT. Attempts to optimize this
compound with introduction of a chlorine atom and by
oxidation of the sulfur atom to the corresponding sulfone
gave the highly potent 5-chloro-3-(phenylsulfonyl)indole-
2-carboxamide (2, L-737,126).2 These findings indicated
diaryl sulfones as a new emerging class of non-nucleo-
side reverse transcriptase inhibitors (NNRTIs) and gave
great support to the idea of developing novel synthetic
sulfones as inhibitors of RT. We therefore synthesized
a number of pyrrolyl aryl sulfones (PASs) and reported
on the anti-HIV-1 activities of these novel derivatives.3
2-Nitroaryl 2-ethoxycarbonyl-1H-pyrrolyl sulfones were
identified as the most active among the first-generation
series studied by us in 1995.
More recently, a second generation of PASs, charac-
terized by a p-chloroaniline pharmacophore and tar-
geted at RT, was found to exert higher and more specific
anti-HIV-1 activity in vitro.4 Compound 3 exhibited the
highest activity within the series (EC50 ) 0.14 µM, CC50
> 200 µM, IC50 ) 0.4 µM, and SI > 1429). As a further
development of the PAS series, we selected 3 as a lead
compound for a synthetic project guided by structure-
based molecular design.
* To whom correspondence should be sent. For M.A.: tel, 39-6-446-
2731; e-mail, artico@uniroma1.it.
#
Universita` degli Studi di Roma “La Sapienza”.
‡ Universita` degli Studi di Napoli “Federico II”.
§ Universita` degli Studi di Siena.
† Menarini Ricerche SpA.
| Universita` degli Studi di Cagliari.
10.1021/jm9901125 CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/14/2000

Brief Articles
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 9 1887
F igu r e 2. Lead compound 3 (yellow) docked into the NNBS
of RT. Only a subset of residues closest to the ligand is
displayed for sake of clarity.
F igu r e 1. Crystal structure of compound 3 showing the
thermal ellipsoids at 40% probability and the labeling of non-H
atoms.
the non-nucleoside binding site (NNBS) of RT. All
calculations were performed using molecular mechanics
(Tripos force field5), conformational analysis, and dock-
ing and graphics routines available within the SYBYL
program.6
X-r a y Cr ysta llogr a p h y of th e Lea d Com p ou n d 3
It is commonly recognized that the bioactive confor-
mation of a ligand does not necessarily correspond to
the lowest-energy conformation either in vacuo, in
solution, or in a crystal. On the other hand, it can be
reasonably assumed that the solid-state conformation
is energetically stable and therefore not too far from the
energy of the global minimum conformer in vacuo. On
the basis of this assumption, strain energies of a given
ligand can be estimated by taking the corresponding
crystal structure as a bench mark. Accordingly, we
determined the crystal geometry of the lead compound
3 to evaluate the strain energy of the enzyme-bound
conformation of this inhibitor resulting from docking
calculations. Strain energies not greater than 3-5 kcal/
mol, typically calculated using molecular mechanics
force fields or quantum-mechanics semiempirical meth-
ods, are usually regarded as “acceptable” values for a
biological relevant conformation. A further advantage
of having available the crystal structure of 3 was the
possibility to check the calculated geometries of PAS
derivatives and analogues about the sulfonamide moi-
ety. In fact, this fragment is not adequately param-
etrized within the Tripos6 and other molecular mechan-
ics force fields.
The experimentally determined RT-bound conforma-
tions of R-APA R903857 served as a template to select a
trial conformation of 3 for docking studies. By scanning
the N1-S1 and S1-C1 rotatable bonds of the crystal
structure of 3 (Figure 1), we identified a low-energy
conformation of this compound superimposable on R-APA
R95845 about the aromatic rings and the COOEt/COMe
and SO2/CONH2 moieties. It will be shown that the
proposed alignment enables the NH2 of 3 to make a
H-bond with the carbonyl oxygen of Lys101 within the
NNBS. The same H-bond is crucial for the binding of
NNRTIs such as HEPT7,8 and TIBO9,10 analogues.
The pharmacophore-based conformation of 3 was then
placed into the NNBS taken from the crystal structure
of the RT/R-APA R90385 complex7 (filed in the Brook-
haven Protein Data Bank11 with the entry code 1VRU).
A limited energy minimization of the RT/3 complex,
aimed at relieving steric repulsive contacts, did not
modify significantly the starting geometry. The internal
energy of the docked inhibitor resulted in 2.3 kcal/mol
greater than that of the corresponding crystal structure.
Our model of the RT/3 complex is depicted in Figure
2. As stated, the NH2 group of the inhibitor forms a
H-bond with the carbonyl oxygen of Lys101, thus
rationalizing SARs of previously reported PAS deriva-
tives:4 (i) removal of NH2 or its replacement with a NO2
strongly lowers antiviral activity; (ii) removal of p-
chlorine as well as its replacement with a p-methyl
disfavors activity probably owing to electronic effects
exerted by the para-substituent on the H-bond donor
character of the NH2. Two intramolecular H-bonds,
donated by the NH2 to the sulfone and ester oxygens,
stabilize the bound conformation of 3.
Figure 1 shows the crystal structure of 3, solved by
X-ray diffraction methods as detailed in the Experimen-
tal Section. The values of the most relevant torsion
angles defining the conformation of 3 are the following
(see atom labeling in Figure 2): τ(C2-C1-S1-N1) )
-128.7°, τ(C2-C1-S1-O1) ) -12.8°, τ(C1-S1-N1-
C7) ) 62.8°, τ(N1-C7-C11-O4) ) -179.5°, and τ(C11-
O4-C12-C13) ) 84.5°. The two aromatic rings of this
compound are arranged in a propeller-like orientation.
Str u ctu r e-Ba sed Design
The solid-state and docked conformations of 3 differ
in the values of the torsion angles τ(C2-C1-S1-N1),
τ(C1-S1-N1-C7), τ(N1-C7-C11-O4), and τ(C11-
The design of new PAS derivatives was guided by a
model of interaction between the lead compound 3 and

1888 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 9
Brief Articles
Ch a r t 2. Newly Synthesized Pyrrolyl Aryl Sulfones
pyrrole ring. Concerning 28, it was anticipated that an
intermolecular steric repulsion between the bulky 2-pyr-
rolyl substituent on the phenyl ring and the carbonyl
group of Lys101 would prevent the binding to RT.
F igu r e 3. Theoretical model of the RT/24 complex character-
ized by attractive interactions between the terminal phenyl
ring of the docked molecule (gray) and the side chains of
Tyr188 and Trp 229. Only four residues of the NNBS (black)
are displayed.
Ch em istr y
Sulfones 4-8 (Chart 2) were prepared according to
previously reported procedures by phase-transfer con-
densation of arylsulfonyl chlorides with the appropriate
pyrroles in the presence of 18-crown-6 and potassium
tert-butoxide.
Sodium borohydride reduction of 5-7 afforded the
corresponding alcohols 9-11. Treatment of 9 with
phosphorus pentachloride gave 12, which in turn was
transformed into the more reactive iodide 13. Reaction
of the last compound with phenol in alkaline medium
gave the required ethyl 1-[(5-chloro-2-nitrophenyl)-
sulfonyl]-4-phenoxymethyl-1H-pyrrole-2-carboxylate (14).
Similarly 13 was transformed into 15 by reaction with
methanol.
Condensation of 5 and 6 with aniline followed by
sodium cyanoborohydride reduction of the intermediate
azomethines afforded ethyl (16) and isopropyl 1-[(5-
chloro-2-nitrophenyl)sulfonyl]-4-phenylaminomethyl-
1H-pyrrole-2-carboxylate (17), respectively.
Transformation of nitro derivatives 4, 8-11, and 14-
17 into the corresponding amino sulfones 18-22 and
24-27, respectively, was reached by heating at 60 °C
with iron powder in glacial acetic acid. Ethyl 1-[(2-
amino-5-chlorophenyl)sulfonyl]-1H-pyrrole-2-carboxy-
late (3)4 was reacted with 2,5-dimethoxytetrahydrofuran
via the Clauson-Kaas pyrrole synthesis to give 28.
1-[(2-Amino-5-chlorophenyl)sulfonyl]-4-(1-hydroxy-
ethyl)-1H-pyrrole-2-carboxylate (22) was transformed
into ethyl 1-[(2-amino-5-chlorophenyl)sulfonyl]-4-vinyl-
1H-pyrrole-2-carboxylate (23) by heating at 100 °C in
the presence of phosphorus pentoxide.
O4-C12-C13): -128.7° vs 98.9°, 62.8° vs -104.2°,
-179.5° vs -162.9°, and 84.5° vs -71.5°, respectively
(see Figure 1 for atom labeling). In both conformations
one of the SO2 oxygens is eclipsed with the plane of the
benzene ring and accepts a H-bond from the NH2. In
the crystal geometry the two aromatic rings are in a
propeller-like disposition, whereas the same rings as-
sume a butterfly-like arrangement in the proposed
bioactive conformation.
An inspection of the RT/3 complex (Figure 3) revealed
a region of the NNBS, delimited by Tyr181, Tyr188, and
Trp229 side chains, which could be filled by substituents
at the 4-position of the inhibitor pyrrole ring. Accord-
ingly, we introduced at this position small to medium
size groups such as methyl, hydroxymethyl, R-hydroxy-
ethyl, vinyl, and methoxymethyl (compounds 18, 20-
23, and 25). Phenoxymethyl and anilinomethyl substit-
uents at the 4-position of the pyrrole were also considered
prompted by results of docking calculations on the
molecular models of 24, 26, and 27. As shown in Figure
3, the terminal phenyl ring of 24 was hypothesized to
favorably contact the Tyr188 (with a π-stacking interac-
tion) and Trp229 side chains.
To evaluate our theoretical model to discriminate
between active versus inactive PAS derivatives, two
structures predicted to be inactive (“negative controls”)
were synthesized and tested as well: the 6-chloro-2-
aminophenyl and 5-chloro-2-(1H-pyrrol-1-yl)phenyl de-
rivatives 19 and 28, respectively. Molecular mechanics
calculations on 19 revealed that this structure could not
attain the pharmacophoric butterfly-like conformation
due to an intramolecular steric clash between the
chlorine and the hydrogen on the 5-position of the
The chemical and physical data of novel pyrrolyl aryl
sulfones are reported in Table 1.
Resu lts a n d Discu ssion
Aryl pyrrolyl sulfones 18-28 were assayed for cyto-
toxicity and anti-HIV-1 activity. Most of the compounds

Brief Articles
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 9 1889
Ta ble 1. Chemical and Physical Data of Compounds 4-28a
with nevirapine, this compound was more active in the
cell-based assay against HIV-1 (8-fold) and as a RT
inhibitor (12-fold).
crystalb
from
chromatc
yield (%) system
compd
formula
mp (°C)
Derivatives substituted at the same position with
methoxymethyl (25) or vinyl (23) were as active as 3.
Introduction of phenoxymethyl (24) and anilinomethyl
groups (26, 27) at the pyrrole 4-position produced RT
inhibitors with IC50 values comparable with that of the
lead compound 3.
Finally, the importance of the p-chloroaniline phar-
macophore was confirmed by the inactivity of the
m-chloroaniline and 5-chloro-2-pyrrolylphenyl deriva-
tives 19 and 28, respectively. Indeed, we predicted that
these two compounds would be inactive: the former is
unable to attain the required “butterfly-like” conforma-
tion,7-9 while the latter does not make the supposedly
vital H-bond with the Lys101 carbonyl.
In conclusion, a theoretical model of interaction
between the lead compound 3 and the NNBS of RT,
consistent with previously developed SARs, was em-
ployed to design novel PAS derivatives and analogues.
Our approach succeeded in optimizing activity of the
lead as well as in discriminating active versus inactive
molecules prior to their synthesis and biological evalu-
ation. Among the compounds synthesized and tested,
20 turned out as the most active, with a 3- and 8-fold
improvement over the lead 3 in the cell-based and
enzyme assays, respectively.
4
5
6
7
8
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
14H13N2O6SCl 150-151
14H11N2O7SCl 191-192
15H13N2O7SCl 179
A
B
B
B
B
C
B
C
B
C
C
A
C
C
B
A
B
B
C
D
C
A
C
C
D
63
64
70
72
28
68
69
69
78
100
43
35
96
92
73
90
95
69
80
32
98
68
98
61
98
a
a
a
a
a
b
b
c
a
a
a
a
a
a
a
a
b
c
c
a
a
b
a
a
a
15H13N2O7SCl 163-164
13H11N2O6SCl 151-153
14H13N2O7SCl 147-150
15H15N2O7SCl 155-156
15H15N2O7SCl 107-108
14H12N2O6SCl2 135-136
14H12N2O6SClI 144-145
20H17N2O7SCl 124-125
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
15H15N2O7SCl
20H18N3O6SCl
92-93
98-99
21H20N3O6SCl 105-106
14H15N2O4SCl 148-149
13H13N2O4SCl 103-104
14H15N2O5SCl 151-152
15H17N2O5SCl 140-142
15H17N2O5SCl 123-124
15H15N2O4SCl 123
20H19N2O5SCl 129-131
15H17N2O5SCl
95-98
20H20N3O4SCl 147-148
21H22N3O4SCl 129-130
17H16N2O4S
102-103
a
Analyzed elements: C, H, N, S and, when present, Cl, F.
Analytical results where within (0.4% of the theoretical values.
A, cylohexane; B, toluene-cyclohexane; C, toluene-ligroin. c a,
b
silica gel/chloroform; b, silica gel/chloroform-ethanol (9:1); c, silica
gel/ethyl acetate.
Exp er im en ta l Section
Ta ble 2. Cytotoxicity, Anti-HIV-1 Activity in MT-4 Cells,
Selectivity Index, and RT Inhibitory Activity of PASs 18-28a
For general experimental information, see ref 5. Usual
workup: the mixture reaction was concentrated and extracted
with ethyl acetate; the collected extracts were washed with
brine, dried on anhydrous sodium sulfate, and evaporated
under reduced pressure.
Gen er a l P r oced u r e for th e Con d en sa tion of Ar ylsu l-
fon yl Ch lor id es w ith 1H-P yr r ole-2-ca r boxyla tes: Ex-
a m p le - Eth yl 1-[(5-Ch lor o-2-n itr op h en yl)su lfon yl]-4-
m eth yl-1H-p yr r ole-2-ca r boxyla te (4). Sulfones 4-8 were
prepared by phase-transfer reaction as we reported in a
previous paper4 (Table 1).
Gen er a l P r oced u r e for th e Red u ction of Nitr o Gr ou p
to Am in o: Exa m p le - Eth yl 1-[(2-Am in o-5-ch lor op h en -
yl)su lfon yl]-4-m eth yl-1H-p yr r ole-2-ca r boxyla te (18). Sul-
fones 18-22 and 24-27 were prepared by reduction of the
corresponding nitro derivatives as we reported in a previous
paper4 (Table 1).
b
c
e
compd
CC50
EC50
0.09
72
SId
IC50
0.08
>100
18
19
20
21
22
23
24
25
>200
180
240
116
115
>2222
2.5
0.045
0.1
0.25
0.17
4.1
0.4
3.1
0.2
5333
1160
460
274
>49
>500
>55.5
200
0.05
0.11
0.17
0.28
0.6
0.6
0.5
0.3
47
>200
>200
>200
40
26
27
28
3
110
>200
>200
>110
>100
0.14
0.35
>1429
>571
0.4
nevirapine
0.64
a
Data represent mean values for three separate experiments.
Variation among triplicate samples was less than 15%. Com-
b
Gen er a l P r oced u r e for th e Red u ction of Ald eh yd e or
Keton e to Alcoh ol: Exa m p le - Eth yl 1-[(5-Ch lor o-2-
n it r op h en yl)su lfon yl]-4-h yd r oxym et h yl-1H -p yr r ole-2-
ca r boxyla te (9). Sodium borohydride (0.098 g, 0.0026 mol)
was added in portions to a solution of 5 (1.00 g, 0.0026 mol) in
THF (7.5 mL) containing 0.4 mL of water, then the mixture
was refluxed for 2 h. Water (5 mL) was added to the cooled
mixture while stirring for a few min. After concentration, the
solution was extracted with ethyl acetate, washed with brine
and dried. Removal of the solvent afforded a residue which
was purified on silica gel column, eluting with chloroform-
ethanol, 9:1. First fractions were collected and evaporated to
give 9. Further elution with the same solvent gave ethyl
4-hydroxymethyl-1H-pyrrole-2-carboxylate (18%). By this pro-
cedure were also prepared 10 and 11 (Table 1).
E t h yl 1-[(5-Ch lor o-2-n it r op h en yl)su lfon yl]-4-ch lor o-
m eth yl-1H-p yr r ole-2-ca r boxyla te (12). Phosphorus pen-
tachloride (95%, 4.14 g, 0.0188 mol) was added in small
portions over a period of 15 min to an ice-cooled solution of 9
(2.65 g, 0.0068 mol) in chloroform (60 mL). After stirring at 0
°C for 1 h, then at room temperature overnight the reaction
mixture was poured onto crushed ice and extracted with
chloroform. Usual workup afforded 12.
pound dose (µM) required to reduce the viability of mock-infected
cells by 50% as determined by the MTT method. c Compound dose
(µM) required to achieve 50% protection of MT-4 cells from HIV-
d
1-induced cytopathicity as determined by MTT method. Selec-
tivity index, CC50/EC50 ratio. e Compound dose (mM) required to
inhibit the HIV-1 rRT activity by 50%.
were also tested against wild-type recombinant reverse
transcriptase (rRT). Table 2 summarizes the results of
the biological evaluations, expressed as CC50 (cytotox-
icity), EC50 (anti-HIV-1 activity), SI (selectivity), and
IC50 (RT inhibitory activity) values.
As predicted using our model of the RT/3 complex,
the insertion of small to medium size substituents, such
as methyl (18), hydroxymethyl (20, 21), or R-hydroxy-
ethyl (22) on the pyrrole 4-position led to inhibitors more
active than their parent compound 3. The most active
one was 20, which exhibited anti-HIV-1 and enzymatic
inhibitory activities at nanomolar concentrations (EC50
) 45 nM and IC50 ) 50 nM), with low cytotoxicity (CC50
) 240 µM) and high selectivity (SI ) 5333). Compared

1890 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 9
Brief Articles
Eth yl 1-[(5-Ch lor o-2-n itr op h en yl)su lfon yl]-4-iod om e-
th yl-1H-p yr r ole-2-ca r boxyla te (13). A solution of sodium
iodide (0.194 g, 0.0013 mol) in acetone (5 mL) was added to a
solution of 12 (0.53 g, 0.0013 mol) in the same solvent (5 mL),
then the mixture was refluxed for 30 min. After concentration
the residue was extracted with chloroform, washed with brine
and dried. Removal of the solvent afforded pure 13.
Eth yl 1-[(5-Ch lor o-2-n itr op h en yl)su lfon yl]-4-p h en oxy-
m eth yl-1H-p yr r ole-2-ca r boxyla te (14). A mixture of 13 (0.5
g, 0.001 mol), phenol (0.094 g, 0.001 mol), potassium carbonate
(0.138 g, 0.001 mol) and acetone (10 mL) was refluxed for 4 h.
After cooling ethyl acetate and water were added by shaking.
Usual workup gave a residue which was purified on silica gel
column. First eluates (dichloromethane-petroleum ether, 1:1)
were discarded; further elution with chloroform afforded 14.
Eth yl 1-[(5-Ch lor o-2-n itr op h en yl)su lfon yl]-4-m eth oxy-
m eth yl-1H-p yr r ole-2-ca r boxyla te (15). Prepared as re-
ported for 14 from 13 (1.0 g, 0.002 mol), methanol (40 mL),
diethyl ether (40 mL), anhydrous tetrahydrofuran (20 mL) and
sodium hydrogen carbonate (0.19 g, 0.0026 mol) with stirring
at room temperature for 3 days.
E t h yl 1-[(5-Ch lor o-2-n it r op h en yl)su lfon yl]-4-p h en yl-
a m in om eth yl-1H-p yr r ole-2-ca r boxyla te (16) Sodium cy-
anoborohydride (0.36 g, 0.0057 mol) was added to an ice-cooled
solution of 5 (1.47 g, 0.0038 mol) and aniline (0.46 g, 0.0049
mol) in methanol (50 mL) and THF (50 mL) containing 0.63
mL of a 6 N HCl/MeOH (1:1) solution. The reaction was stirred
at 0 °C for 2 h, then at room temperature overnight. Usual
workup gave a crude product which was chromatographed to
afford 16. By this procedure was also prepared 17 starting from
6.
E t h yl 1-[(2-Am in o-5-ch lor op h en yl)su lfon yl]-4-vin yl-
1H-p yr r ole-2-ca r boxyla te (23). Phosphorus pentoxide (1.65
g, 0.0116 mol) was added by small portions to a stirred mixture
of 22 (1.00 g, 0.0027 mol) in anhydrous benzene (30 mL), then
reaction was heated at 100 °C for 45 min. After cooling the
organic solution was separated and the gummy residue was
extracted with benzene. Combined cloudy extracts were fil-
tered, washed with 5% NaHCO3 solution and then with brine
and dried. After concentration the residue was chromato-
graphed to yield 23.
Eth yl 1-{[5-Ch lor o-2-(1H-p yr r ol-1-yl)p h en yl]su lfon yl}-
1H-p yr r ole-2-ca r boxyla te (28). A solution of ethyl 1-(2-
amino-5-chlorophenylsulfonyl)-1H-pyrrole-2-carboxylate (3) (1.00
g, 0.0034 mol) and 2,5-dimethoxytetrahydrofuran (90%, 0.90
g, 0.0061 mol) in glacial acetic acid (20 mL) was refluxed for
1 h, then evaporated to dryness. Usual workup gave 28 which
was purified by chromatography.
Isop r op yl 4-F or m yl-1H-p yr r ole-2-ca r boxyla te. A solu-
tion of dichloromethyl methyl ether (13.80 g, 0.12 mol) in 1,2-
dichloroethane (50 mL) was quickly added to a stirred mixture
of isopropyl 1H-pyrrole-2-carboxylate (15.31 g, 0.10 mol) and
anhydrous aluminum trichloride (32.00 g, 0.24 mol) in 1,2-
dichloroethane (200 mL) while cooling at -30 °C, and the
reaction was maintained at -30 °C overnight. After reaching
0 °C, the mixture was poured on ice water and extracted with
chloroform. Organic extracts were combined, washed with
brine and dried. Removal of the solvent gave a residue which
was purified on a silica gel column (chloroform): yield 50%;
mp 85-86 °C (toluene/cyclohexane). Anal. (C9H11NO3) (181.19)
C, H, N.
g/cm-3, λ(Cu KR) ) 1.54178 Å, T ) 293 K. A total of 2009
reflections were measured, 1853 of which were independent.
The anisotropic refinement of the non-hydrogen atoms used
1468 observed reflections [I > 3σ(I)] and converged to R )
0.049, RW ) 0.060, S ) 1.98. Parameters refined 163, Dr in
the final difference map was within +0.50 and -0.41 Å3.
Molecu la r Mod elin g. Molecular modeling studies were
performed using the SYBYL6 software package running on a
Silicon Graphics R8000 workstation. Intramolecular and
intermolecular energies were calculated using the molecular
mechanics Tripos force field5 including the electrostatic con-
tribution. Atom centered partial charges were calculated
according to the Gasteiger-Hu¨ckel method.13,14 Geometry
optimizations were realized with the SYBYL/MAXIMIN2
minimizer by applying the BFGS algorithm15 and setting a
root-mean-square gradient of the forces acting on each atom
at 0.05 kcal/mol Å as a convergence criterion.
Hydrogens were added to the unfilled valences of the crystal
structure of compound 3 and their positions optimized while
keeping fixed the rest of the molecule. The putative bioactive
conformation of 3 was sought following a two-step approach:
(i) selection of a pharmacophore-consistent conformation and
(ii) docking of this latter into the NNBS of RT.
First, we identified a low-energy conformation of 3 super-
imposable on the experimentally determined bioactive confor-
mation of the R-APA type inhibitor R90385 extracted from the
RT/R90385 complex solved at 2.4 Å resolution.7 The coordi-
nates of this complex are filed in the Brookhaven Protein Data
Bank11 with the entry code 1VRU. The search for the phar-
macophore-based conformation of 3 was performed by manual
adjustment of the torsion angles about the N1-S1 and S1-
C1 and rotatable bonds (see Figure 2 for atom labeling) aimed
at finding a butterfly-like orientation of the two aromatic rings.
As further constraints, we wanted to match the COOEt and
COMe carbonyl oxygens of 3 and R90385, respectively, and
orient the NH2 group of 3 toward the Lys101 carbonyl oxygen
as part of the NNBS. Compound 3 was overlayed on R90385
by minimizing the root-mean-square distance between the
following five points: (i) pseudoatoms placed at 1.0 Å along
the normals to the plane of each aromatic ring passing through
the ring centroid; (ii) the SO2 sulfur and the CONH2 carbon.
The energy of the resulting conformation of 3 was only 3.0 kcal/
mol above that of the corresponding crystal geometry (this
value was computed considering the sum of the nonbonded
components of the force field, i.e., van der Waals plus electro-
statics).
The resulting pharmacophore-based conformation of 3 was
then placed into the NNBS extracted from the crystal structure
of the RT/R90385 complex. The NNBS comprised 71 amino
acids located within a distance of 10 Å from any non-hydrogen
atom of the bound inhibitor. Water molecules were deleted.
Hydrogens were added to the unfilled valences of the amino
acids and the Lys, Asp and Glu side chains were modeled in
their ionized forms. The torsion angle τ(C11-O4-C12-C13)
of 3 was manually changed from 84.5° to -75.0° to avoid a
steric clash between the ester group and the Leu100 side chain.
The resulting NNBS/3 complex was geometry-optimized by
keeping fixed the coordinates of the protein backbone until an
attractive intermolecular interaction energy of 25 kcal/mol was
reached. The internal energy of the docked conformation rose
by 2.3 kcal/mol with respect to the crystal structure.
Meth ods. X-r ay Cr ystallogr aph y of Com pou n d 3. Single
crystals of 3 suitable for X-ray diffraction analysis were
obtained by recrystallization from a hot ethanol solution. All
measurements were made on a Rigaku AFC5R diffractometer
with graphite monochromated Cu KR radiation and a rotating
anode generator. The ω/2θ scan tecnique was used. The
structure was solved by direct methods; the positions of
hydrogens were calculated and not refined. All calculations
were performed using the TEXSAN crystallographic software
package of Molecular Structure Corp.12 The data for 3 were
as follows: C13H13N2O4SCl, Mr 282.77, a ) 7.9558(9) Å, b )
21.069(5) Å, c ) 7.633(1) Å, â ) 93.69(1)°, V ) 1138.5(3) Å3, Z
) 4, space group P21/a (No. 14), monoclinic, DC ) 1.649
Given the lack of adequate parametrization for the sulfona-
mide moiety featured by all the PAS derivatives, bond dis-
tances and bond angles of this fragment were kept fixed to
the values measured in the crystal structure of 3 throughout
the docking calculations. This was done by applying specific
penalty functions characterized by strong force constants. The
torsion angles of the sulfonamide residue were, in contrast,
left free to vary as though they had no impact on the internal
energy of the molecule (this was accomplished by setting the
corresponding torsional force constants to zero). However, the
geometry of the sulfonamide moiety in the docked conforma-
tion of the ligand was checked a posteriori against the crystal
structures of 3 and of 31 pyrrolyl sulfone derivatives retrieved

Brief Articles
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 9 1891
from the Cambridge Structural Database (CSD)16 by a sub-
structure search. The 3D graphics 5.13 version of the CSD for
UNIX platforms (April 1997 release) was employed. Analysis
of these structures and of a published survey on sulfonamide
crystal geometries17 revealed that the combination of values
assumed by the Caryl-N-S-Caryl and Caryl-Caryl-S-N torsion
angles in the docked conformation of 3 was within the range
of experimental values.
Models of derivatives and analogues of compound 3 were
constructed starting from the docked conformation of 3 and
submitted to geometry optimization of the resulting NNBS/
inhibitor complex as described above.
V.; Schleif, W. A.; Theoharides, A. D.; Anderson, P. S. 5-Chloro-
3-(phenylsulfonyl)indole-2-carboxamide: A Novel, Non-Nucleo-
side Inhibitor of HIV-1 Reverse Transcriptase. J . Med. Chem.
1993, 36, 1291-1294.
(3) Artico, M.; Silvestri, R.; Stefancich, G.; Massa, S.; Pagnozzi, E.;
Musiu, D.; Scintu, F.; Pinna, E.; Tinti, E.; La Colla, P. Synthesis
of Pyrryl Aryl Sulfones Targeted at the HIV-1 Reverse Tran-
scriptase. Arch. Pharm. (Weinheim) 1995, 328, 223-229.
(4) Artico, M.; Silvestri, R.; Massa, S.; Loi, A. G.; Corrias, S.; Piras,
G.; La Colla, P. 2-Sulfonyl-4-chloroanilino Moiety: A Potent
Pharmacophore for the Anti-Human Immunodeficiency Virus
Type 1 Activity of Pyrrolyl Aryl Sulfones. J . Med. Chem. 1996,
39, 522-530.
(5) Clark, M.; Cramer, R. D., III; Van Opdenbosch, N. Validation
of the General Purpose Tripos 5.2 Force Field. J . Comput. Chem.
1989, 10, 982-1012.
(6) SYBYL Molecular Modeling System (version 6.2); Tripos Inc.,
St. Louis, MO.
(7) Ren, J .; Esnouf, R.; Garman, E.; Somers, D.; Ross, C.; Kirby, I.;
Keeling, J .; Darby, G.; J ones, Y.; Stuart, D.; Stammers, D. High-
Resolution Structures of HIV-1 RT from four RT-Inhibitor
Complexes. Nat. Struct. Biol. 1995, 2, 293-302.
(8) Hopkins, A. L.; Ren, J .; Esnouf, R. M.; Willcox, B. E.; J ones, E.
Y.; Ross, C.; Miyasaka, T.; Walker, R. T.; Tanaka, H.; Stammers,
D. K.; Stuart, D. I. Complexes of HIV-1 Reverse Transcriptase
with Inhibitors of the HEPT Series Reveal Conformational
Changes Relevant to the Design of Potent Non-Nucleoside
Inhibitors. J . Med. Chem. 1996, 39, 1589-1600.
(9) Ding, J .; Das, K.; Moereels, H.; Koymans, L.; Andries, K.;
J anssen, P. A. J .; Hughes, S. H., Arnold, E. Structure of HIV-1
RT/TIBO R 86183 Complex Reveals Similarity in the Binding
of Diverse Nonnucleoside Inhibitors. Nat. Struct. Biol. 1995, 2,
407-415.
(10) Ren, J .; Esnouf, R.; Hopkins, A.; Ross, C.; J ones, Y.; Stammers,
D.; Stuart, D. The structure of HIV-1 Reverse Transcriptase
Complexed with 9-Chloro-TIBO: Lessons for Inhibitor Design.
Structure 1995, 3, 915-926.
(11) Bernstein, F. C.; Koetzle, T. F.; Williams, G. J . B.; Meyer J r., E.
F.; Brice, M. D.; Rodgers, J . R.; Kennard, O.; Shimanouchi, T.;
Tasumi, T. The Protein Data Bank: A Computer Based Archival
File for Macromolecular Structures. J . Mol. Biol. 1977, 112,
535-542.
(12) TEXAN-TEXRAY Single-Crystal Structure Analysis Software
(version 5.0, 1989); Molecular Structure Corp., The Woodlands,
TX 77381.
In the case of the phenoxymethyl (24) and anilinomethyl
(26) derivatives, molecular dynamics (MD) simulations of the
corresponding complexes were performed with the SYBYL/
DYNAMICS module under default settings. These simulations,
in which the protein backbone was treated as a rigid aggregate,
were run at 300 K for 50 ps after reaching equilibration. Five
structures extracted at 10, 20, 30, 40 and 50 ps were selected
and energy-minimized. Figure 3, displaying the interaction
between the terminal phenyl ring of compound 24 and the side
chains of Tyr188 and Trp229, refers to the last MD snapshot.
An tivir a l Assa ys. HIV Titr a tion . As previously reported4
titration of HIV was performed in C8166 cells by the standard
limiting dilution method (dilution 1:2, four replica wells/
dilution) in 96-well plates. The infectious virus titer was
determined by light microscope scoring of cytopathicity after
4 days of incubation and the virus titers were expressed as
CCID50/mL.
An ti-HIV Assa ys. Activity of the compounds against HIV-1
multiplication in acutely infected cells was based on the
inhibition of virus-induced cytopathicity in MT-4 and C 8166
cells, respectively. According to the previously reported pro-
cedure,4 the number of viable cells was determined by the
3-(4,5-dimethylthiazol-1-yl)-2,5-diphenyltetrazolium bromide
(MTT) method. Cytotoxicity of the compounds was evaluated
in parallel with their antiviral activity. It was based on the
viability of mock-infected cells, as monitored by the MTT
method.
RT Assa ys. Assays were performed as previously de-
scribed.4
(13) Gasteiger, J .; Marsili, M. Iterative Partial Equalization of Orbital
Electronegativity. Tetrahedron 1980, 36, 3219-3228.
(14) Purcel, V. P.; Singer, J . A. A Brief Review and Table of
Semiepirical Parameters Used in the Hu¨ckel Molecular Orbital
Method. J . Chem. Eng. Data 1967, 12, 235-246.
(15) Head, J .; Zerner, M. C. A Broyden-Fletcher-Goldfarb-Shanno
Optimization Procedure for Molecular Geometries. Chem. Phys.
Lett. 1985, 122, 264-274.
Ack n ow led gm en t. We thank the Italian Ministero
della Sanita` - Istituto Superiore di Sanita` - XI Progetto
AIDS 1997 (Grants 40A.0.06 and 40A.0.55) for financial
support. The Italian MURST (40%) and Regione Au-
tonoma della Sardegna (Progetto Biotecnologie) are also
acknowledged for partial support.
(16) Allen, F. H.; Bellard, S. Brice, M. D.; Cartwright, B. A.;
Doubleday: A.; Higgs, H.; Hummelink, T.; Hummelink-Peterd,
B. G.; Kennard, O.; Motherwell, W. D. S.; Rodgers, J . R.; Watson,
D. G. The Cambridge Crystallographic Data Centre: Computer-
Based Search, Retrival, Analysis and Display of Information.
Acta Crystallogr. 1979, B35, 2331-2339.
(17) Krystek, J r., S. R.; Hunt, J . T.; Stein, P. D.; Stouch, T. R. Three-
Dimensional Quantitative Structure-Activity Relationships of
Sulfonamide Endothelin Inhibitors. J . Med. Chem. 1995, 38,
659-668.
Refer en ces
(1) McMahon, J . B.; Gulakowsky, R. J .; Weislow, O. S.; Schoktz, R.
J .; Narayanan, V. L.; Clanton, D. J .; Pedemonte, R.; Wassmundt,
F. W.; Buckheit, R. W., J r.; Decker, W. D.; White, E. L.; Bader,
J . P.; Boyd, M. R. Diaryl sulfones, a New Chemical Class of Non-
Nucleoside Antiviral Inhibitors of Human Immunodeficiency
Virus Type 1 Reverse Transcriptase. Antimicrob. Agents Chemo-
ther. 1993, 37, 754-760.
(2) Williams, T. M.; Ciccarone, T. M.; MacTough, S. C.; Rooney, C.
S.; Balani, S. K.; Condra, J . H.; Emini, E. A.; Goldman, M. E.;
Greenlee, W. J .; Kauffman, L. R.; O’Brien, J . A.; Sardana, V.
J M9901125