Discovery of chiral cyclopropyl dihydro-alkylthio-benzyl-oxopyrimidine (5-DABO) derivatives as potent HIV-1 reverse transcriptase inhibitors with high activity against clinically relevant mutants

Article abstract of DOI:10.1021/jm801330n

The role played by stereochemistry in the C2-substituent (left part) on the S-DABO scaffold for anti-HIV-1 activity has been investigated for the first time. A series of S-DABO analogues, where the double bond in the C2-substituent is replaced by an enant

Full text of DOI:10.1021/jm801330n

840
J. Med. Chem. 2009, 52, 840–851
Discovery of Chiral Cyclopropyl Dihydro-Alkylthio-Benzyl-Oxopyrimidine (S-DABO)
Derivatives as Potent HIV-1 Reverse Transcriptase Inhibitors with High Activity Against
Clinically Relevant Mutants
Marco Radi,†,‡ Giovanni Maga,†,§ Maddalena Alongi,‡ Lucilla Angeli,‡ Alberta Samuele,§ Samantha Zanoli,§ Luca Bellucci,‡
Andrea Tafi,‡ Gianni Casaluce,‡ Gianluca Giorgi,| Mercedes Armand-Ugon, Emmanuel Gonzalez, Jose´ A. Este´,§
Mireille Baltzinger, Guillaume Bec,# Philippe Dumas,# Eric Ennifar,# and Maurizio Botta*,‡
Dipartimento Farmaco Chimico Tecnologico, UniVersity of Siena, Via Alcide de Gasperi 2, I-53100 Siena, Italy, Istituto di Genetica
Molecolare, IGM-CNR, Via Abbiategrasso 207, I-27100 PaVia, Italy, Dipartimento di Chimica, UniVersity of Siena, Via Alcide de Gasperi 2,
I-53100 Siena, Italy, RetroVirology Laboratory irsiCaixa, Hospital UniVersitari Germans Trias i Pujol, UniVersitat Auto`noma de Barcelona,
E-08916 Badalona, Spain, Architecture et Re´actiVite´ de l’ARN, UPR 9002 CNRS/UniVersite´ Louis Pasteur, 15 rue Rene´ Descartes,
67084 Strasbourg, France, UMR 7175 CNRS/UniVersite´ Louis Pasteur, Ecole Supe´rieure de Biotechnologie Strasbourg, BouleVard Se´bastien
Brandt, F-67400 Illkirch, France
ReceiVed October 21, 2008
The role played by stereochemistry in the C2-substituent (left part) on the S-DABO scaffold for anti-HIV-1
activity has been investigated for the first time. A series of S-DABO analogues, where the double bond in
the C2-substituent is replaced by an enantiopure isosteric cyclopropyl moiety, has been synthesized, leading
to the identification of a potent lead compound endowed with picomolar activity against RT (wt) and
nanomolar activity against selected drug-resistant mutants. Molecular modeling calculation, enzymatic studies,
and surface plasmon resonance experiments allowed us to rationalize the biological behavior of the synthesized
compounds, which act as mixed-type inhibitors of HIV-1 RT K103N, with a preferential association to the
enzyme-substrate complex. Taken together, our data show that the right combination of stereochemistry
on the left and right parts (C6-substituent) of the S-DABO scaffold plays a key role in the inhibition of both
wild-type and drug-resistant enzymes, especially the K103N mutant.
Chart 1. Structure of Common NNRTIs
Introduction
The drug arsenal for the war on AIDS currently comprises
25 approved drugs belonging to six different classes of anti-
HIV inhibitors1 that target viral enzymes and proteins. The only
exception is Maraviroc (UK-427,857),2 which targets a cellular
cofactor (CCR5). The development of drugs that are active
against different targets represents an important step forward
in the eradication of HIV, especially drug-resistant mutations.
However, recent clinical studies demonstrated that therapies
based on Maraviroc and Raltegravir have comparable efficacy
* To whom correspondence should be addressed. Phone: (+39)0577-
234306. Fax: (+39)0577-234333. E-mail: botta@unisi.it.
†
These authors equally contributed to this work.
Dipartimento Farmaco Chimico Tecnologico, University of Siena.
Istituto di Genetica Molecolare, IGM-CNR.
‡
§
Retrovirology Laboratory irsiCaixa, Hospital Universitari Germans
Trias i Pujol, Universitat Auto`noma de Barcelona.
|
to regimens based on the non-nucleoside reverse transcriptase
inhibitor (NNRTIa) Efavirenz.3 Reverse transcriptase (RT)
therefore represents an old but still important target for the
identification of novel NNRTIs endowed with a better activity
profile against clinically relevant mutant strains.4 Non-nucleoside
reverse transcriptase inhibitors include more than 30 structurally
different classes of molecules such as Nevirapine, TIBO, DABO,
HEPT, TNK-651, ITU, DATA and DAPY (Chart 1).5 Among
the NNRTIs reported to date, the members of the DABO family
(namely S-DABOs, DABOs, NH-DABOs, F2-N,N-DABOs,
DATNOs) present a very similar binding mode within the non-
nucleoside inhibitor binding pocket (NNBP) characterized by
a key hydrogen-bond contact between the NH at position 3 of
the pyrimidinone and Lys101 (Figure 1, zone C).6 In addition,
C5 and C6 substituents establish important hydrophobic interac-
tions with a large pocket defined by the residues of zone A and
Dipartimento di Chimica, University of Siena.
UMR 7175 CNRS/Universite´ Louis Pasteur, Ecole Supe´rieure de
Biotechnologie Strasbourg.
#
Architecture et re´activite´ de l’ARN, UPR 9002 CNRS/Universite´ Louis
Pasteur.
a Abbreviations: S-DABO, dihydro-alkylthio-benzyl-oxopyrimidines;
HIV-1, human immunodeficiency virus type-1; WT, wild type; NNRTIs,
nonnucleoside reverse transcriptase inhibitors; DABOs, dihydro-alkyloxy-
benzyl-oxopyrimidines; DATNOs, dihydroalkylthio-naphthylmethyl-ox-
opyrimidines; EFV, efavirenz; F2-N,N-DABOs, 5-alkyl-2-(N,N-disubsti-
tuted)amino-6-(2,6-difluorophenylalkyl) pyrimidin-4(3H)ones; HEPT, 1-[(2-
hydroxyethoxy)methyl]-6-(phenylthio)thymine; NH-DABOs, dihydro-
alkylamino-benzyl-oxopyrimidines;NVP,nevirapine;TIBO,tetrahydroimidazo[4,5,1-
jk][1,4]benzodiazepinone; DAPY, diarylpyrimidine; DATA, diaryltriazine;
ITU, imidoylthiourea; SAR, structure-activity relationship; NNBP, non-
nucleoside binding pocket; PDB, protein data bank; MD, molecular
dynamics; rmsd, root-mean-square deviation; PBC, periodic boundary
conditions; PME, particle-mesh-Ewald; NPT, number of molecules pressure
and temperature; SPR, surface plasmon resonance; NA, nucleic acid; dNTP,
deoxyribonucleotide triphosphate.
10.1021/jm801330n CCC: $40.75
2009 American Chemical Society
Published on Web 01/13/2009

Chiral Cyclopropyl S-DABO DeriVatiVes
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3 841
Figure 1. Interactions of a generic C2-arylalkyl S-DABO (I) within the NNBP and structures of compound 1 and target analogue 2.
B depicted in Figure 1. The substituent at position C2, which
is characteristic for each DABO family, establishes a series of
interactions within the NNBP for which a clear structure-activity
relationship (SAR) has not been established yet.
macophoric sites of a promising RT-inhibitor such as 1 could
allow us to generate a collection of enantiopure drugs with
improved activity and potentially usefulness in overcoming
future drug-induced mutations.
During the course of a study aimed at exploring the
relationship between the C2-functionalization of the S-DABO
scaffold and the anti-HIV activity, our research group first
reported how the introduction of an arylalkyl moiety in C2
significantly increased the antiviral activity due to profitable
hydrophobic interaction with a large pocket (zone D) of the
allosteric site.7 Subsequent molecular dynamics studies showed
that the right combination of the C2 linker length and the R4
substituent on the phenyl ring played a key role in the opening
of the NNBP of the K103N mutant.8 Compound 1 (Figure 1)
emerged as the most interesting inhibitor for its activity on MT-4
cells infected with both HIV-1 wild type (NL-43 wt) and
clinically relevant mutants, especially the one bearing the K103N
mutation. In the search for new C2-substituted S-DABO
analogues, Nawrozkij et al. have recently reported the synthesis
and biological investigation of many C2-arylalkyl S-DABO
analogues. Among them, the only interesting compounds were
represented by the C2-oxophenethyl derivatives. However, these
showed low antiviral potency against clinically relevant drug-
resistant mutants.9 It is well-known that, to identify high quality
lead compounds, it is important to address the relationship
between drug-chirality and activity early in the drug discovery
process.10 From this point of view, while the role played by
the stereochemistry in the right part of the S-DABO scaffold
for the activity on RT (wt) has already been established,11 no
reports have been published so far on the role played by the
stereochemistry of the left part of the molecule.
Considering the importance of the C2 substituent of 1 for
the antiviral activity, we decided to introduce a chiral moiety
on the left part of the molecule 1, trying not to disrupt the
topology/flexibility of the parent compound. Accordingly, we
planned to convert compound 1 into its cyclopropyl bioisoster
2: the cyclopropyl moiety would not only allow us to introduce
two new stereocenters while maintaining the optimal length of
the C2-linker but it is also a bulky lipophilic substituent that
could give additional interactions with the hydrophobic residues
of zone D. Furthermore, as described in our previous simulations
on the K103N mutant,8 the side chains of residues Lys101 and
Val179 shaped a sort of crevice on the external surface of the
closed enzyme. This hydrophobic area lies in close proximity
to the C2-linker and could therefore be exploited by the
lipophilic cyclopropyl group, giving an additional contribution
to the generation of the allosteric binding pocket. In addition,
the introduction of different stereocenters on important phar-
Result and Discussion
The synthesis of the title compounds (Scheme 1) started with
the reduction of the commercially available p-methoxy cinna-
maldehyde 3 with NaBH4 to give the cinnamyl alcohol 4. The
cyclopropane ring was then built on the double bond using the
Furukawa reagent (diethyl zinc and diiodomethane) alone12 or
in the presence of a chiral promoter (1,2-bis(methanesulfony-
lamino)cyclohexane)13 to give rac-5, (S,S)-5, and (R,R)-5,
respectively (Scheme 1). The last two compounds were obtained
in 98% ee as determined by high-performance liquid chroma-
tography (HPLC) on Chiralcel OD column (see Experimental
Section for details). The racemic and enantiopure cyclopropyl
derivatives 5 were then coupled with 6-(1-(2,6-difluorophenyl)-
ethyl)-5-methyl-3,4-dihydro-2-thioxopyrimidin-4(3H)-one via a
microwave-assisted Mitsunobu reaction to give the S-DABO
derivatives rac-2, 2a, and 2b.
The diastereoisomeric mixtures 2a and 2b were finally
resolved by semipreparative HPLC on ALLTIMA C18 column
thus obtaining each single diasteroisomer (2c-f) with high
diasteromeric excess (>98%) (see Experimental Section for
details). The absolute stereochemistry of the four isomers was
determined by X-ray crystallography of 2d (R,R,S),14 which
showed four crystallographic independent molecules constituting
the asymmetric unit of 2d (see Supporting Information). Each
molecule shows a “folded” conformation with the difluorophenyl
moiety pointing toward the methoxyphenyl ring while the latter
is almost perpendicular to the cyclopropane with a dihedral angle
that means 83.8(7)° in the four molecules (Figure 2).
The pure compounds 2c-f were then submitted to biological
evaluation in cell-free assay using Nevirapine and Efavirenz as
reference compounds. As reported in Table 1, the results of the
biological screening clearly highlighted the importance of the
right combination of stereochemistry on the left and right part
of the S-DABO scaffold for the inhibition of both RT(wt) and
drug resistant mutants (especially the K103N). Comparing the
activity data of compounds 2c and 2f, it was clear that the (R,R)
configuration of the cyclopropane was fundamental for the better
activity profile of 2c against the clinically relevant K103N drug
resistant mutant. This consideration was further reinforced by
the biological evaluation of the two enantiomers of compound
1 obtained via semipreparative HPLC resolution (see Experi-
mental Section for details) and for which absolute configuration
was determined by converting them into the known compounds

842 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3
Radi et al.
Scheme 1a
a (i) NaBH4, MeOH, rt, 1h; (ii) Et2Zn, CH2I2, 0 °C. (iii) (a) (R,R)-1,2-bis(methanesulfonylamino)cyclohexane, Et2Zn, ZnI2; (b) Et2Zn, CH2I2. (iv) (a)
(S,S)-1,2-bis(methanesulfonylamino)cyclohexane, Et2Zn, ZnI2; (b) Et2Zn, CH2I2; (v) 6-(1-(2,6-difluorophenyl)-ethyl)-5-methyl-3,4-dihydro-2-thioxopyrimidin-
4(3H)-one, (CH3)3P, DIAD, DMF, MW 40 °C, 10 min.
Table 2. Anti-HIV-1 Activity of Compounds 2c-f in MT-4 Cells
EC50 (µM)a,b
c
compd
rac-2
2c (R,R,R)
2d (R,R,S)
2e (S,S,S)
2f (S,S,R)
Nevirapine
Efavirenz
NL4-3 wt
K103N
Y181C
Y188L
CC50
0.0006
0.00007
0.0049
0.090
0.010
0.08
0.52
0.036
2.55
5.51
3.71
3.9
0.16
0.013
0.52
7.24
2.70
0.045
0.0015
0.19
1.08
0.70
37.3
14.64
47.17
>56.25
>56.25
>7.5
>7..5
0.008
>7.5
0.003
0.001
0.057
>0.32
a Data represent mean values of at least two experiments; b EC50:effective
concentration 50 or needed concentration to inhibit 50% HIV-induced cell
death, evaluated with MTT method in MT-4 cells. c CC50: cytotoxic
concentration 50 or needed concentration to induce 50% death of noninfected
cells.
Figure 2. X-ray crystal structure of 2d. Only one of the four molecules
constituting the asymmetric unit is reported. Ellipsoids enclose 50%
probability.
inhibitory activity. The pure diasteroisomers 2c-f were finally
evaluated on MT-4 cells for their cytotoxicity and anti-HIV
activity in comparison with Nevirapine and Efavirenz, used as
reference drugs (Table 2). Compound 2c emerged as the most
active S-DABO derivative reported so far with a picomolar
activity against RT (wt), nanomolar activity against all the tested
mutants, and an antiviral profile better than that of Nevirapine
and Efavirenz. Taken together, these biological results confirm
our previous findings on the SAR for the S-DABO inhibitors:
while the right part of the molecule seems to modulate the
activity against the Y181C and Y188L mutants, the left part
seems to play a key role for the activity on the K103N mutant.8
Molecular Modeling Calculations. In the attempt to ratio-
nalize the biological data obtained for compounds 2c-f,
molecular docking simulations (using Autodock 3.0)16 were
initially performed on RT wild type (PDB entry 1RT2)17
following a computational protocol previously described by us.8
The docking simulations clearly showed that only the inhibitors
2c and 2f bind the NNBP in the S-DABOs’ “canonical” way:7a
two hydrogen bonds between the amide moiety of the pyrimi-
dinone ring and the backbone of Lys101 and an additional
H-bond with protonated N atom of Lys101 side-chain, hydro-
Table 1. Anti-HIV-1 Activity of Compounds 1-2 in Cell Free Assay
ID50 (µM)a,b
compd
wt
K103N
Y181I
rac-2
0.04
0.02
0.30
0.12
0.04
0.60
0.01
0.40
0.04
3.00
0.04
0.60
0.83
0.40
7.01
0.47
8.00
0.40
6.00
0.80
6.00
2.30
0.90
27.83
2.99
20.00
0.10
2c (R,R,R)
2d (R,R,S)
2e (S,S,S)
2f (S,S,R)
1 (S)
1 (R)
Nevirapine
Efavirenz
a Data represent mean values of at least two experiments; b ID50:
inhibiting dose 50 or needed dose to inhibit 50% of the enzyme.
2c and 2d via enantioselective cyclopropanation.15 Compound
1(R) was in fact 10-fold more active than 1(S) both against RT
(wt) and mutants but 10-fold less active than 2c against K103N
and Y181I: the presence of a (R,R) cyclopropyl moiety on the
left part of the molecule coupled with a C6′-(R)methyl group
(as in compound 2c) determined a marked increase in the

Chiral Cyclopropyl S-DABO DeriVatiVes
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3 843
Figure 3. (A) Binding mode of compound 2c (carbon atoms in yellow) as found by docking simulations into the NNBP; compound 2f shows the
same binding mode. (B) Compound 2d (carbon atoms in purple) as found by docking simulations into the NNBP; compound 2e shows the same
binding mode. (C) Both compounds 2c and 2d in the allosteric site.
phobic interactions of the C6-benzyl group with the aromatic
cage formed by Tyr181, Tyr188, Phe227, and Trp229 (accessory
interactions also with Leu100 and Leu234), and finally very
profitable lipophilic interactions between the C6′-(R)methyl
group and a hydrophobic pocket defined by Val106, Tyr181,
Tyr188, Val189, and Gly190 (Figure 3A). On the contrary,
compounds 2d and 2e, showed a deep modification of the
“canonical” binding mode in order to allow the C6′-(S)methyl
group to establish the above-mentioned hydrophobic interac-
tions. The pyrimidinone ring tilts and loses the hydrogen bond
with N-amidic of Lys101 backbone, while the benzyl ring,
shifting toward Tyr188, loses favorable contacts with the
aromatic ring of Tyr181 (Figure 3B,C).
To confirm the importance of the C6′-methyl group for the
lipophilic interactions within the NNBP, Grid18 maps were
calculated for CH3 group (C3 probe) to evaluate the regions of
best interaction between this group and the macromolecule. The
most favorable minimum point (minimum 1) was located in
the lipophilic pocket defined by Val106, Tyr181, Tyr188,
Val189, and Gly190, while another one was found between
Leu100, Val179, and Tyr181 (minimum 2). The docking
calculations clearly demonstrated the importance of mapping
the minimum 1 because, among all the first ranked docked
conformations of 2c-f, a methyl group fits this minimum. The
two methyl groups in C5 and C6′ of compounds 2c and 2f
exactly mapped minima 1 and 2, while compounds 2d and 2f
are forced to rearrange within the binding site in order to map
minimum 1, losing some other important interactions inside the
NNBP (Figure 4).
Although our docking calculations allowed us to rationalize
the activity profile of compound 2c-f against RT wild type,
these experiments did not give any useful information on the
role played by chirality of the left part of the molecule: in fact,
the cyclopropyl moiety of the docked inhibitors 2c-f showed
comparable interaction within the NNBP.
A striking feature of RT is its considerable conformational
flexibility. This has complicated the traditional structure-based
drug design approach for the discovery of novel NNRTIs. To
overcome this problem, the inclusion of structural variability
in a docking study and the concept of ligand-induced fit have
been taken into account.19
Accordingly, a cross-docking approach (the process of
docking each ligand into the binding pocket of a series of
different ligand-receptor complexes), followed by a molecular
dynamics protocol, were applied to further investigate the role
played by the C2 side chain in the antiviral activity.20
Figure 4. Superposition of the lower energy docked conformation of
compounds 2c (green), 2d (yellow), 2e (red), and 2f (white). Grid
minima 1 and 2 are shown as light blue spheres. Compounds 2c and
2f exactly mapped both the minima, while 2d and 2e lose interactions
with minimum 2 in order to map minimum 1.
For this purpose, available X-ray structures of reverse
transcriptase bound to various NNRTIs were superimposed on
the crystallographic complex RT:TNK-651 (1RT2)17 and the
root-mean-square deViation (rmsd) values for the residues
forming the NNBP were calculated (see Supporting Informa-
tion). Among the considered 44 RT:NNRTIs complexes, the
crystallographic structures having the most different conforma-
tions of NNBP (chosen on the basis of rmsd value and visual
inspection) with respect to 1RT2, were used to perform cross-
docking simulations. The chosen X-ray structures were 1BQM,
1S9E, and 3HVT.21-23 The performed cross-docking protocol
confirmed that the initial conformation of the NNBP strongly
influences the ligands’ binding: in none of the first-ranked
conformations did compounds 2c-f show the “canonical”
binding mode observed within the 1RT2 structure. The classical
binding mode was, however, found at a higher energy level with
respect to the first ranked conformation, and it was used as a
starting point for the subsequent molecular dynamic simulations
in order to evaluate the ability of the title compounds to reshape
the NNBP.
The behavior of the most (2c) and least active (2e) compounds
in the allosteric site of RT (wt) complexes was investigated by
MD simulations in explicit solvent with the aim of observing

844 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3
Radi et al.
completely fit inside the NNBP of 1BQM and that it was more
exposed to the solvent than 2c. Therefore the observed highest
rmsd value for the 1BQM:2e complex did not rely on the
induced fit property of the ligand but reflected a sort of closure
of the NNBP that was only partially occupied by the ligand.
The superposition of the first and the last minimized structures
taken from the MD simulations of 1BQM and 1RT2 in complex
with both 2c and 2e clearly shows what has been reported in
Table 3 (Figure 6). In the case of 1BQM, the different behavior
of 2c versus 2e can be clearly appreciated: while 2c was able
to fit into the allosteric site during the course of the dynamics
simulation (Figure 6A), compound 2e maintained substantially
its starting pose and was unsuccessful in reshaping the allosteric
cavity (Figure 6B). For comparison, the behavior of 2c,e within
the NNBP of 1RT2 is reported in Figure 6C,D: in this case,
due to the structural similarity between the original ligand
(TNK651) and 2c,e, the first and last frame of the MD
simulation are almost superimposable, no major reshaping of
the allosteric cavity can be appreciated, and the inhibitors
completely fit into the NNBP.
The energetic contribution of the only cyclopropyl moiety
was also estimated in 1RT2:2c and 1RT2:2e in order to quantify
its contribution to the total binding energy. A value of about
7-9% of the total protein-ligand interaction energy was found
for both complexes (Table 3). Taken together, these results
indicate that the right combination of chirality on C2 and C6
of the S-DABO scaffold is responsible for the high activity of
compound 2c, which, contrary to 2e, seems to possess marked
induced-fit properties that can be translated in the ability to
reshape the NNBP in order to establish a more profitable
interactions pattern (see MD movie in the Supporting Informa-
tion).
Figure 5. rmsd time evolution along the MD simulations.
the mutual rearrangement of both the ligands and the NNBP. It
should be mentioned that, while the structure 1RT2 was
cocrystallized with a compound (TNK-651) structurally similar
to 2c and 2e, 1BQM was cocrystallized with inhibitor HBY
097, very dissimilar to 2c and 2e. As reported above, the best-
scored docking solution of 2c,e within the two crystals were
not superimposable, but the docking procedure allowed us to
select two similar geometries of the bound inhibitors as a
reasonable starting point for the MD study (see above in the
text “canonical binding mode”). Supramolecular assemblies
were built by adding TIP3P water molecules and then they were
submitted to a MD protocol. All heavy protein atoms were
harmonically restrained with
a force constant of 100
(kcal·mol-1) Å2 to the crystallographic position. Energy mini-
mization followed by 0.250 ns of molecular dynamics under
NPT (see Experimental Section, Molecular Dynamic Simula-
tions) ensemble conditions were carried out. Subsequently,
harmonic force constant was scaled by a factor of 10 every 50
ps and completely released after 100 ps of simulation. MD
simulations were protracted for 1.6 ns. To describe the induced-
fit properties of the ligands, the root-mean-square deviations
(rmsd) of the atomic coordinates of the NNBP residues (see
Experimental Section, Structure Preparation) were calculated
for every frame of each MD simulation with respect to the
starting point for that run. The rmsd values were used to quantify
the drift of the NNBP structure from its initial conformation,
therefore providing a crude measure of the “ligand’s ability” to
reshape the allosteric cavity (Figure 5).
In the attempt to rationalize the biological data on the K103N
mutant (PDB entry 1IKY),24 docking calculations highlighted
again the importance of the C6′-(R)methyl group for the activity:
while compounds 2c and 2f display a “canonical” binding mode
inside the allosteric site, compounds 2d and 2e seem to bind
the NNBP in a flipped conformation, with the 2,6-difluorophenyl
ring in place of p-methoxyphenyl moiety and vice versa
(Figure 7).
However, even if compound 2f showed profitable interactions
with the NNBP, a remarkable decrease of potency against the
K103N mutant was found both in enzymatic and cellular assays.
It is clear that the only discriminant between 2c and 2f is
represented by the stereochemistry of the cyclopropyl moiety
on the C2 chain, but, again, the docking studies on K103N were
not able to find any major difference in the binding mode of
these two compounds, which showed the same interaction
pattern found for the parent compound 1 (Figure 7). As already
reported by us,8 the simple study of the interactions between
ligand and protein does not seem to be sufficient to rationalize
the biological data in the case K103N mutation, and other
elements must be taken into account like the capacity of the
molecules to disrupt the H-bond between Asn103 and Tyr188,
allowing the opening and so the entrance into the NNBP.
Unfortunately, molecular dynamics simulations conducted on
compounds 2c-f were not helpful in understanding this
behavior. Accordingly, we decided to start a thorough kinetic
study on both RT wild type and K103N mutant in order to get
further information on the mechanism of action of 2c-f, which
could clarify the computational results.
As expected, the calculated rmsd for the NNBP of 1RT2 (blue
and cyan lines in Figure 5) is less pronounced than that observed
for the NNBP of 1BQM (orange and red lines in Figure 5) due
to the similarity of 2c,e with the 1RT2 cocrystallized ligand
(TNK651). In fact, during the simulations, 2c,e did not show
any major rearrangement and did not induce a substantial
remodeling of the allosteric site. On the contrary, both inhibitors
in 1BQM showed a higher rmsd value. However, conversely
to our expectations, the less active compound 2e showed the
highest rmsd trend (orange line in Figure 5), indicating a major
rearrangement of the NNBP. To clarify these results, the
protein-ligand and water-ligand average interaction energies
were calculated over the last 200 ps of the MD simulations for
all complexes and are reported in Table 3. In both crystal-
lographic structures, 2c showed the highest protein-ligand
interaction energy (in absolute value) in agreement with the
experimental activity data. Notably, we observed that the 1BQM:
2e complex showed not only the lower protein-ligand interac-
tion energy but also the highest water-ligand interaction energy
in absolute value. These energies suggested that 2e did not
Kinetic Studies. Kinetic parameters of the highly active
compound 2c were characterized toward RT (wt) using surface
plasmon resonance (SPR) and compared to compound 1 and to

Chiral Cyclopropyl S-DABO DeriVatiVes
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3 845
Table 3. rmsd Values for the Residues of the NNBP of 1RT2 and 1BQM from the Initial Minimized Structures Calculated in the Last 200 ps of MDa
entry
X-ray
ligand
rms (Å)
protein-ligand (kcal ·mol-1
)
water-ligand (kcal ·mol-1
)
cyclopropyl-protein (kcal ·mol-1
)
1
2
3
4
1RT2
1RT2
1BQM
1BQM
2c (R,R,R)
2e (S,S,S)
2c (R,R,R)
2e (S,S,S)
1.27
1.28
1.56
1.96
-80.32
-76.36
-78.21
-63.57
-12.26
-15.54
-16.42
-25.01
-5.89
-6.57
a Protein-ligand, water-ligand, and ligand cyclopropane moiety-protein interaction energies, calculated in the last 200 ps of MD, are kcal ·mol-1
.
Figure 6. (A) Superposition of first (green) and last (purple/pink) minimized frames coming from dynamic simulations of 1BQM:2c. (B) Superposition
of first (green/white) and last (purple/pink) minimized frames of 1BQM:2e. (C) Superposition of first (yellow) and last (blue/cyan) minimized
frames of 1RT2:2c. (D) Superposition of first (yellow) and last (blue/cyan) minimized frames of 1RT2:2e. In (A) and (B), the distances between
amidic N atom of pyrimidinone ring and the carbonylic C atom of Lys101 are reported as black dashed lines. Notably, only compound 2c is able
to form the typical H bond (shown as purple line in A) with the backbone of Lys101. In (C) and (D), the hydrogen bonds are reported as yellow
(for the first minimized frame) or blue (for the last frame) dashed lines.
the recently approved anti-HIV drug Etravirine (TMC125)25
used as a reference of high-affinity compound. Our results show
that all the inhibitors dissociate very slowly if at all from a RT
surface (Figure 8) as shown by similar flat curves (at time >2400
s) and cannot be quantitatively differentiated by SPR. These
observations further confirm the results of our MD studies: as
can be clearly appreciated from the movie in the Supporting
Information, during the course of the dynamics simulation, the
most active compound 2c is deeply incorporated within the
NNBP, which, at the end, completely surround the inhibitor.
Data analysis revealed that compound 2c binds very tightly
to the RT with an apparent association rate constant ka ) (1.0
( 0.4)104 M-1 s-1 and a dissociation rate constant kd ) (2.1 (
3.5)10-5 s-1 (beyond the measurable limit of the instrument).
Kinetic parameters evaluated for 1 were ka ) 1 × 104 M-1s-1
and kd e 6 × 10-5 s-1, and those for TMC125 were ka ) 0.8
× 104 M-1s-1 and kd e 4 × 10-5 s-1. These data reveal the
small dissociation rates for the three compounds and are
significantly lower than values obtained by the same technique
for Delavirdine, Efavirenz, and Nevirapine, which range from

846 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3
Radi et al.
Figure 7. (A) Binding mode of compounds 2c (reported in ball and stick and colored in green), 2f (ball and stick, blue), and 1(R) (ball and stick,
purple) into K103N NNBP (PDB entry 1IKY). (B) Binding mode of 2d (ball and stick, orange), 2e (ball and stick, light gray), and 1(R) (ball and
stick, purple) into K103N NNBP (PDB entry 1IKY).
were obtained for the other diasteroisomers 2d-f (data not
shown). These results showed that 2c is a noncompetitive
inhibitor of HIV-1 RT wt. The same experiments were repeated
in the presence of the HIV-1 RT K103N mutant. As shown in
Figure 9D, the Km values for the nucleic acid substrate were
significantly decreased by increasing concentrations of 2c,
whereas the corresponding values for the nucleotide substrate
were unaffected (Figure 9E). The Vmax values were decreased,
as expected (Figure 9F). Similar results were obtained for the
other diasteroisomers 2d-f (data not shown).
By analyzing the variation of the apparent Km and Vmax values
as a function of the inhibitor concentrations, it was possible to
determine the equilibrium dissociation constants (Ki) for the
three enzymatic forms along the reaction pathway, namely the
free enzyme (E), the binary complex with the nucleic acid (E:
NA), and the ternary complex with both substrates (E:NA:
dNTP). In addition, equilibrium binding studies were applied
to derive the respective association (kon) and dissociation (koff)
rates. The corresponding values are listed in Table 4. These
results indicated that compounds 2c-f are mixed-type inhibitors
of HIV-1 RT K103N, with a more preferential association to
the enzyme-substrate complex than to the free RT. These
findings further support an induce-fit mechanism of action for
these compounds and can justify the failure of our molecular
modeling approach to rationalize the activity of compounds 2c-f
because all the calculations were performed on the free enzyme
and, unfortunately, no crystal structure of the K103N in the
binary or ternary complex is at present available to repeat the
calculations. In the end, by comparing the Ki for the ternary
complex of the wild type and K103N enzymes, it appeared that
compound 2c showed only a 3.3-fold reduction in its inhibition
potency due to the mutation. In comparison, the clinically used
NNRTIs Efavirenz and Nevirapine showed a loss of potency
of 50- and 10- fold, respectively. As a result, 2c was 6- fold
more potent than Efavirenz and 125-fold more potent than
Nevirapine against the K103N mutant.
Figure 8. Kinetic titrations of RT interactions with compounds 2c, 1,
and TMC125. The three sensorgrams were obtained in independent
experiments. The scale for the SPR response signal was normalized
assuming a 1:1 binding stoichiometry. Original responses ranged
between 0 and 70 RU for 2c (green line) and 0-40 RU for 1 (blue
line) and TMC125 (red line). A 2-fold increase in sample concentration
was used for each compound during successive injections (0.075, 0.15,
0.3, 0.6, and 1.2 µM).
26
1 × 10-2 to 2 × 10-3 s-1
.
It can also be appreciated that, at
saturating amount of inhibitor 2c (g1.2 µM), the signal was
observed to exceed the maximum for a 1:1 interaction and this
could be due to (i) the presence of two binding sites for 2c on
RT or (ii) a slow conformational alteration induced by the
binding of the inhibitor.27
While the interaction of 2c with two different binding sites
on RT cannot be either confirmed or excluded with the data at
present in our hands, the pronounced induced-fit properties of
2c, found with the MD simulations herein reported, suggest that
the conformational alteration of RT could be a possible
explanation for the atypical binding stoichiometry.
Finally, to determine the mode of action of compound 2c,
the dependence of the inhibition on both the nucleic acid (NA)
and the nucleotide substrate (dNTP) was studied. The kinetic
parameters Km and Vmax were calculated at increasing concentra-
tion of the inhibitor 2c and compared to the values obtained in
the absence of the inhibitor. As shown in Figure 9, in the
presence of 2c, no significant variations were noted in the Km
values for both substrates (Figure 9A,B), while the Vmax was
reduced in a dose-dependent manner (Figure 9C). Similar results
Conclusions
In the present study, the role played by the stereochemistry
on the C2-substituent (left part) of the S-DABO analogues for
anti-HIV-1 activity was investigated for the first time. The
introduction of different stereocenters on important pharma-
cophoric site of a promising RT-inhibitor such as 1 could also

Chiral Cyclopropyl S-DABO DeriVatiVes
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3 847
Figure 9. (A) Variation of the apparent affinity of RT wt for the nucleic acid (Km[NA]) as a function of the inhibitor 2c concentration. (B) Variation
of the apparent affinity of RT wt for the nucleotide substrate (Km[dNTP]) as a function of the inhibitor 2c concentration. (C) Variation of the Vmax of
the reaction with RT wt obtained varying the nucleic acid (circles) or the nucleotide (triangles) substrates as a function of the inhibitor 2c concentration.
(D-F) The same as in (A-C), but in the presence of the K103N mutant RT. The Km and Vmax values were determined as described in the Experimental
Section. Values are the mean of three independent estimates. Error bars represent ( SD.
Table 4. Kinetics of the Binding of 2c and 2f to RT (wt) and K103N
E
E:DNA
E:DNA:dNTP
compd Kia µM kon M-1 s-1 × 104 koff s-1 × 10-4
Ki µM
kon M-1 s-1 × 104 koff s-1 × 10-4
Ki µM
kon M-1 s-1 × 104 koff s-1 × 10-4
HIV-1 RT Wild Type
2c
0.01
((0.002)
0.04
((0.005)
0.03
((0.008)
0.4
0.8
((0.1)
1.2
((0.1)
1
((0.1)
0.4
8
((1)
5
((0.5)
3
((0.3)
16
0.008
((0.001)
0.03
((0.004)
0.03
((0.007)
0.5
1.5
((0.2)
1.3
((0.1)
1
((0.1)
0.3
1.3
0.009
((0.001)
0.04
((0.003)
0.004
((0.001)
0.4
4.1
((0.3)
1.2
((0.1)
4
((0.2)
0.4
4
((0.5)
5
((0.5)
1.6
((0.2)
16
((1)
2f
5
((0.5)
3
((0.3)
15
EFZ
NVP
((0.06)
((0.1)
((1)
((0.1)
((0.02)
((0.01)
((0.08)
((0.1)
((1)
HIV-1 RT K103N
2c
0.065
((0.005)
0.6
((0.01)
1.5
((0.2)
4
1.2
((0.2)
0.1
((0.02)
2
((0.3)
0.04
8
((1)
6
((0.7)
30
((3)
16
0.04
((0.003)
0.4
((0.05)
1.6
((0.2)
4
1.3
((0.2)
0.1
((0.02)
2.5
((0.3)
0.04
5
((0.5)
4
((0.4)
40
((5)
16
0.03
((0.003)
0.3
((0.04)
0.2
((0.01)
5
1.5
((0.1)
0.1
((0.02)
6
5
((0.5)
3
((0.3)
12
((1)
15
2f
EFZ
NVP
((1)
0.03
((0.2)
((0.005)
((1)
((0.2)
(0.005)
((2)
((0.5)
((0.005)
((2)
a E, free Enzyme; E:NA, binary complex of the enzyme and the nucleic acid (NA); E:NA:dNTP, ternary complex of the enzyme, the nucleic acid, and
the nucleotide substrate (dNTP).
allow us to generate a collection of enantiopure drugs with
improved activity and potential usefulness in overcoming future
drug-induced mutation. Accordingly, the C2-cyclopropyl ana-
logue (2) of compound 1 was synthesized and its four diaste-
reoisomers obtained via semipreparative HPLC. Results from
enzymatic and cellular assays revealed the importance of the
right combination of the stereochemistry on the left (C2-
substituent) and right part (C6-substituent) of the S-DABO
scaffold for the inhibition of both wild type and drug resistant
enzymes, particularly the K103N mutant. The (R,R) configu-
ration of the cyclopropane ring in C2, coupled with a C6′-
(R)methyl group gave the promising lead compound 2c, which
exhibited picomolar activity against RT (wt) and nanomolar
activity against all the tested mutants. A computational protocol
comprising docking studies, grid mapping, cross-docking ex-
periments, and dynamics simulation was successfully applied
for the rationalization of the biological results on wild type RT.
However, the same approach failed when applied to the K103N
mutant, possibly because of a preferential association of 2c-f
to the enzyme-substrate complex as indicated by a thorough
kinetic enzymatic study. Real-time binding kinetics of our
inhibitors was characterized using surface plasmon resonance
(SPR), revealing that compound 2c has a similar behavior to the
recently approved anti-HIV drugs Etravirine (TMC125), with an
extremely low dissociation constant. These data confirm the results
of our MD studies which show that the lead compound 2c is more
effective in reshaping the allosteric cavity, which, at the end,
completely surrounds the inhibitor. The SPR study also revealed
that, at saturating amount of inhibitor 2c (g1.2 µM), the signal
was observed to exceed the maximum for a 1:1 interaction; further
studies are ongoing in order to completely exclude the interaction
of 2c with two different binding sites on RT (wt).
Experimental Section
General Considerations. Reagents were obtained from com-
mercial suppliers and used without further purifications. According

848 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3
Radi et al.
to standard procedures, CH2Cl2 was dried over calcium hydride
and DMF was bought already anhydrous. Anhydrous reactions were
run under a positive pressure of dry N2 or argon. IR spectra were
recorded on a Perkin-Elmer BX FT-IR system using CHCl3 as
the solvent. TLC was carried out using Merck TLC plates Kieselgel
60 F254. Chromatographic purifications were performed on columns
packed with Merck 60 silica gel, 23-400 mesh, for flash technique.
Melting points were measured using a Gallenkamp melting point
apparatus and are uncorrected. 1H NMR spectra were recorded with
a Bruker AC200F spectrometer at 200 MHz, and chemical shifts
are reported in values relative to TMS at 0.00 ppm. CDCl3 was
used as solvent. Optical rotations have been measured on a Perkin-
Elmer 343 polarimeter equipped with a sodium lamp (λ ) 589 nm)
and a 10 cm microcell.
resolution of the computation: the GRID points were 0.333 Å apart.
A value of 0 was used for the GRID keyword MOVE, thus allowing
lone pairs and tautomeric hydrogen atoms to move in response to
the probe. To achieve the points of minimum of molecular
interaction fields (MIFs), Grid calculations were performed using
the export GRIDKONT option. The kont file was then processed
by means of the minim and filmap programs (both implemented in
the Grid package) that extract all the points of minimum of MIF
and retain only those falling under a user-defined cutoff, fixed to 0
kcal·mol-1. The result of this procedure is a file with a .pdb
extension containing the 3D coordinates of the points of minimum
of the interaction energy, together with the corresponding energy
values.
Structure Preparation. Forty-four X-ray crystal structures of
RT complexed with a NNRTI were retrieved from the Protein Data
Bank (Table S1, Supporting Information). For the cross-docking
studies, complexes were prepared as previously reported8 and
backbone atoms of each crystal structure were superimposed with
RT/TNK-651 PDB entry 1RT2)17 as reference. The superposition
of the RT complexes was performed with the program Pymol
version 0.99 using the implemented Shindyalov and Bourne
algorithm.30,31 The rmsd values for all residues backbone atoms
and of heavy atoms of residues (forming the NNBP) Pro95A,
Leu100A-Lys103A, Val106A, Val179A-Tyr181A, Tyr188A-
Gly190A, Phe227A, Trp229A, Leu234A-Pro236A, and Tyr318A
were calculated and are reported in Table S1, Supporting Informa-
tion. The complexes with the highest rmsd values (arbitrarily chosen
over 1.85 Å and reported in bold in Table S1, Supporting
Information) were subjected to visual inspection. Complexes with
PDB codes 1TV6, 2B5J, and 2BE232,33 were discarded since the
peculiar orientation of the aromatic side chain of Tyr181 that points
away from the polymerase active site as in the apo-RT structure.
Among X-ray structures 1S6P, 1S9G, 1S9E, and 1SUQ, which show
very similar NNBP, 1S9E was chosen as representative complex
for subsequent calculations. So, complexes 1BQM, 1S9E, and
3HVT were used to perform cross-docking calculations, applying
the same protocol previously described.8
HPLC and MS Analysis. Mass spectral (MS) data were obtained
using an Agilent 1100 LC/MSD VL system (G1946C) with a 0.4
mL/min flow rate using a binary solvent system of 95:5 MeOH/
water and an electrospray ionization source (ESI). UV detection
was monitored at 254 nm. Mass spectra were acquired in positive
mode scanning over the mass range of 50-1500. The following
ion source parameters were used: drying gas flow, 9 mL ·min-1
nebulizer pressure, 40 psi; drying gas temperature, 350 °C.
;
The purity of compound rac-2 was assessed by HPLC (Agilent
1100 LC/MSD VL system (G1946C)) using a Zorbax Eclipse XDB,
4.6 mm × 150 mm, 5 µm C8 column with a mobile phase
composed of MeOH/H2O (80:20) at a flow rate of 1 mL ·min-1
(all solvents were HPLC grade, Fluka).
Enantiomeric excess of (R,R)-5 and (S,S)-5 was established with
a Varian ProStar HPLC system (Varian Analytical Instruments,
Perkin-Elmer Inc. USA equipped with a binary pump with manual
injection valve and model ProStar 325 UV-vis detector) using a
Chiralcel OD 0.46 cm × 25 cm column, with a mobile phase
composed of hexane/iPrOH, 95:5 at a flow rate of 0.8 mL ·min-1
.
The two diastereoisomeric mixtures 2a and 2b were resolved
on a Varian ProStar HPLC system using a semipreparative Alltima
C18 250 mm × 10 mm column with a mobile phase composed of
CH3CN/formic acid 0.1% 70:30 at a flow rate of 2.5 mL ·min-1
.
Molecular Dynamics Simulations. Atom types, covalent and
nonbonded parameters were assigned to inhibitors from those
already present in the AMBER force field, by analogy or through
interpolation.34 Charges were calculated as previously described
(see Molecular Docking Studies). System setup, molecular dynamics
(MD), and postdynamic analysis were performed with NAMD2
(v2.6)35 and VMD (v1.8.6)36 software packages. Supramolecular
assemblies to be subjected to MD studies were built starting from
appropriate AutoDock output complexes (see docking section). Each
complex were surrounded by a periodic box of TIP3P water
molecules37 that extended about 9.0 A from the protein atoms and
12 Cl- ions were added to conserve neutrality of the system. For
each system, the number of water molecules was 30170 and the
initial dimensions of the rectangular periodic box were about 120
× 87 × 115 Å. The simulations were conducted using periodic
boundary conditions (PBC) and particle mesh Ewald (PME)
method38 with grid size of 100 × 80 × 100 Å. The nonbonded
cutoff was set to 10.0 Å, the switching distance to 9 Å and the
nonbonded pair list distance to 14 Å. The r-RESPA multiple time
step method39 was employed with 1 fs for bonded, 2 fs short-range
for nonbonded, and 4 fs for long-range electrostatic forces.
Hydrogen atoms were free to move. Molecules were simulated for
a total of 2.1 ns in the NPT ensemble. The temperature was set to
310 K and controlled via Langevin thermostat with a dumping factor
of 5 ps-1. The pressure of 1 atm was controlled via isotropic
Langevin piston barostat with oscillation period of 200 fs, damping
time scale of 100 fs, and noise temperature of 310 K. At this stage,
water molecules, ligand and hydrogen atoms were free to move,
while heavy protein atoms were restrained to the crystallographic
position using a value of 4 in the harmonic energy function exponent
and with constant of force of 100.0 kcal ·mol-1 Å-1. The starting
assemblies were relaxed with 2000 steps of conjugate gradient
energy minimization to remove unfavorable contacts. The obtained
structures were then subjected to 250 ps of MD with same harmonic
Diasteromeric excess of 2c-f was established by reinjecting
every single compound in the semipreparative Alltima C18 250
mm × 10 mm column with a mobile phase composed of CH3CN/
formic acid 0.1% 70:30 at a flow rate of 2.5 mL ·min-1 (Varian
ProStar HPLC system).
The enantiomeric mixture rac-1 was resolved on a Varian ProStar
HPLC system using a semipreparative CHIRALPAK A, 250 mm
× 10 mm column with a mobile phase composed of hexane/iPrOH,
80/20, at a flow rate of 2.5 mL/min.
Microwave Irradiation Experiments. Microwave reactions
were conducted using a CEM Discover Synthesis Unit (CEM Corp.,
Matthews, NC). The machine consists of a continuous focused
microwave-power delivery system with operator-selectable power
output from 0 to 300 W. The temperature of the contents of the
vessel was monitored using a calibrated infrared temperature control
mounted under the reaction vessel. All experiments were performed
using a stirring option whereby the contents of the vessel are stirred
by means of a rotating magnetic plate located below the floor of
the microwave cavity and a Teflon-coated magnetic stir bar in the
vessel.
Molecular Docking Studies. All calculations and manipulations
were performed following a previously reported protocol.8 The
geometry of each inhibitor was optimized using the ab initio
quantum chemistry program GAUSSIAN28 and the HF/6-31G*
basis set, in order to calculate partial atomic charges. Consequently,
a set of atom-centered RHF HF/6-31G* charges was obtained for
each inhibitor, by application of RESP methodology.29
GRID-Based Binding Site Analysis. Computation of Molecular
Interaction Fields (MIFs) over the NNBP was carried out with
software GRID, version 22.18 Box dimensions were defined so as
to accommodate all the residues constituting the NNBP and the
NPLA (Number of Planes of grid points per Å) parameter was set
to 3. A value of 3 was used for the GRID keyword NPLA, which
is the number of planes of GRID points per Å and determines the

Chiral Cyclopropyl S-DABO DeriVatiVes
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3 849
restraints and same constant of force. Subsequently, harmonic force
constant were scaled by a factor of 10 every 50 ps and they were
completely released after 100 ps of dynamic simulation. Random
velocities were reassigned according to the temperature system
every 50 ps over the first 0.5 ns of the simulation with the purpose
to enhance disruption in the NNBP. MD simulations were protracted
for 1.6 ns in the NPT ensemble and trajectories were stored each
400 fs.
concentrations of dNTPs was analyzed, at the steady-state, all of
the input RT was in the form of the RT/NA binary complex and
only two forms of the enzyme (the binary complex and the ternary
complex with dNTP) could react with the inhibitor. Similarly, when
the dNTP concentration was kept constant at saturating levels and
the inhibition at various NA concentrations was analyzed, RT was
present either as a free enzyme or in the ternary complex with NA
and dNTP. The steady-state rate equation used for the partially
mixed inhibition was the one describing a reaction involving only
two mechanistic forms of the enzyme:
Biological Evaluation. Expression, Purification, and
Cloning of Recombinant HIV-1 RT Forms. Recombinant het-
erodimeric RT, either wild type or the K103N variant, were
expressed and purified as briefly described below. The HIV-1 RT
gene fragment spanning codons 2 to 261 from pHXB2D2-261RT
constructs carrying the K103N mutation was amplified by PCR,
digested with Acc1 and Pvu2, and cloned into the expression
plasmid p6HRT (∆Xho1/Bgl2), containing the wild type RT gene.
The resulting p6HRT expression vectors with the K103N mutation
were used for the production in E. coli (BL21) and purification of
recombinant His-tagged RT enzymes in a fast protein liquid
chromatography (F-PLC) system, using Ni-NTA superflow column
(QIAGEN, USA) (Abdullah, N.; Chase, H. A. Biotechnol. Bioeng.
2005, 92, 501–513). All the enzymes were purified to a >95%
purity, as confirmed by sodium dodecyl sulfate-polyacrylamide gel
electrophoresis and Gelcode Blue stain, and had a specific activity
on poly(rA):oligo(dT). One Unit of DNA polymerase activity
corresponds to the incorporation of 1 nmol of dNMP into acid-
precipitable material in 60 min at 37 °C.Western Blotting confirmed
the identity of the polypeptides present in the final preparation with
anti-RT monoclonal antibodies.
V ) {[Vmax⁄(1 + I ⁄ K’’i)] ⁄ [1 + (Km⁄S)] · [(1 + I ⁄ K’i) ⁄ (1 +
I ⁄ (K’’i)} (1)
When NA was saturating, K′i ) Kifree, whereas at saturating dNTP
K′i ) Kibin. In all cases, K′′i ) Kiter. According to the mixed-type
mechanism of eq (1), it follows that, if K′i > K′′i, then both Km
and Vmax values will decrease at increasing inhibitor concentrations.
When K′i ) K′′i, then the eq (1) can be simplified to the one
describing a fully noncompetitive mechanism.
The equilibrium dissociation constants (K′i and K′′i) were
calculated from the variations of the Km and Vmax values as a
function of the inhibitor concentrations according to the equations:
K’i)I ⁄ {[Kp(1 + I ⁄ K’i) ⁄ Km] - 1}
K’’i)I ⁄ [(Vmax⁄Vp) - 1]
(2)
(3)
where Kp and Vp are the apparent Km and Vmax values, respectively,
at each given inhibitor concentration.
Kinetics of inhibitor binding experiments were as described
previously. Briefly, HIV-1 RT (20-40 nM) was incubated for 2
min at 37 °C in a final volume of 4 µL in the presence of TDB
buffer, with 10 mM MgCl2 alone or with 100 nM 3′-OH ends (for
the formation of the RT/TP complex), or in the same mixture
complemented with 10 µM unlabeled dTTP (for the formation of
the RT/TP/dNTP complex). The inhibitor to be tested was then
added to a final volume of 5 µL at a concentration at which [EI]/
[E0] ) [1 - 1/(1 + [I]/Ki)] > 0.9. Then, 145 µL of a mix containing
TDB buffer, 10 mM MgCl2, and 10 µM [3H]dTTP (5 Ci/mmol)
was added at different time points. After an additional 10 min of
incubation at 37 °C, 50 µL aliquots were spotted on GF/C filters
and acid-precipitable radioactivity was measured as described for
the HIV-1 RT RNA-dependent DNA polymerase activity assay.
The Vt/V0 ratio, representing the normalized difference between the
amount of dTTP incorporated at the zero time point and at different
time points, was then plotted against time.
HIV-1 RT RNA-Dependent DNA Polymerase Activity
Assay. Poly(rA)/oligo(dT) was used as a template for the RNA-
dependent DNA polymerase reaction by HIV-1 RT, either wt or
carrying the mutations. For the activity assay, a 25 µL final reaction
volume contained TDB buffer (50 mM Tris-HCl (pH 8.0), 1 mM
dithiothreitol (DTT), 0.2 mg/mL bovine serum albumin (BSA), 2%
glycerol), 10 mM MgCl2, 0.5 mg of poly(rA):oligo(dT)10:1 (0.3 mM
3′ -OH ends), 10 mM [3H]dTTP 1 Ci ·mmol-1 and, finally,
introduced into tubes containing aliquots of different enzyme
concentrations (5-10 nM RT). After incubation at 37 °C for
indicated time, 20 µL from each reaction tube were spiked on glass
fiber filters GF/C and immediately immersed in 5% ice-cold
trichloroacetic acid (TCA) (AppliChem GmbH, Darmstadt). Filters
were washed three times with 5% TCA and once with ethanol for
5 min, then dried and, finally, added with EcoLume Scintillation
cocktail (ICN, Research Products Division, Costa Mesa, CA), to
detect the acid-precipitable radioactivity by PerkinElmer Trilux
MicroBeta 1450 Counter.
Anti-HIV Activity in Lymphoid Cells. Biological activity of
the compounds was tested in the lymphoid MT-4 cell line (received
from the NIH AIDS Reagent Program) against the wt HIV-1 NL4-3
strain and three different HIV-1 strains, as described before.40
Briefly, MT-4 cells were infected with the appropriate HIV-1 strain
(or mock-infected to determine cytotoxicity) in the presence of
different drug concentrations. At day five postinfection, a tetrazo-
lium-based colorimetric method (MTT method) was used to
evaluate the number of viable cells. The HIV-1 strains containing
the multi-NNRTI mutation, K103N or the Y188L mutant, were
received from the Medical Research Council Centralised Facility
for AIDS Reagents, Herfordshire, UK.
The apparent binding rate (kapp) values were determined by fitting
the experimental data to the single-exponential equation:
Vt ⁄ V0 ) e-kappt(4)
(4)
where t is time. If [E]0 is the input enzyme concentration, [E]t is
the enzyme available for the reaction at time t, and [E:I]t is the
enzyme bound to the inhibitor at time t, it follows that
[E]t)[E]0-[E:I]t
(5)
Because V0 ) kcat[E]0 and Vt ) kcat[E]t, then Vt/v0 ) 1 - [E:I]t/
[E]0. Thus, the Vt/V0 value is proportional to the fraction of enzyme
bound to the inhibitor.
The true association (kon) and dissociation (koff) rates were
calculated from the equations:
Kinetic Analysis. The mechanism of action of the compounds
was found to be either fully noncompetitive or partially mixed.
According to the ordered mechanism of the polymerization reaction,
whereby template-primer (TP) binds first followed by the addition
of dNTP, HIV-1 RT can be present in three different catalytic forms:
as a free enzyme, in a binary complex with the nucleic acid (NA),
and in a ternary complex with NA and dNTP. The resulting rate
equation for such a system is very complex and impractical to use.
For these reasons, the general steady-state kinetic analysis was
simplified by varying one of the substrates (either NA or dNTP)
while the other was kept constant. When the NA substrate was
held constant at saturating concentration and the inhibition at various
kapp)kon([I] + Ki)
koff)konKi
(6)
(7)
Data analysis and statistics: Data obtained were analyzed by
nonlinear regression analysis using GraphPad Software (San Diego,
CA).
Surface Plasmon Resonance. The interactions between HIV-1
RT and inhibitors were studied using a BIAcore 2000 instrument
(GE Healthcare). Interaction between a protein immobilized on a

850 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3
Radi et al.
biosensor chip and a compound flowed over the surface was
monitored in real time as a change in surface plasmon resonance
units. CM5 sensor chips (research grade) were used for the
experiments. For immobilization, HIV-RT sensor surfaces were
prepared using the amine coupling chemistry as described in refs
26 and 27. Final immobilization levels were between 10000 and
12500 resonance units (RU), corresponding to approximately 10
to 12.5 ng of reverse transcriptase ·mm-2. A dextran surface treated
in the same way but without protein injection was used as a
reference.
Kinetic titration experiments were performed at 25 °C using 10
mM Hepes, 150 mM NaCl, 3.4 mM EDTA, 0,01% surfactant P20
(polysorbate Tween-20), 4% DMSO, pH 7.5 as running buffer. The
inhibitor was serially diluted in running buffer to working concen-
trations. Within a single binding cycle, the samples were injected
sequentially in order of increasing concentration over both the
reference and RT protein surfaces. Also, prior to and after the
binding cycle, buffer was injected for ,blank. responses essential
to double-reference the data.
movie of the MD simulation for compounds 2e (left side) and 2c
(right side) within the NNBP of 1BQM (AVI). This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Greenea, W. C.; Debyserb, Z.; Ikedac, Y.; Freedd, E. O.; Stephense,
E.; Yonemotoa, W.; Buckheitf, R. W.; Este´, J. A.; Cihlarh, T. Novel
targets for HIV therapy. AntiVir. Res. 2008, in press.
(2) (a) Este, J. A.; Telenti, A. HIV entry inhibitors. Lancet 2007, 370,
81–88. (b) Fatkenheuer, G.; Pozniak, A. L.; Johnson, M. A.;
Plettenberg, A.; Staszewski, S.; Hoepelman, A. I.; Saag, M. S.; Goebel,
F. D.; Rockstroh, J. K.; Dezube, B. J.; Jenkins, T. M.; Medhurst, C.;
Sullivan, J. F.; Ridgway, C.; Abel, S.; James, I. T.; Youle, M.; van
der Ryst, E. Efficacy of short-term monotherapy with maraviroc, a
new CCR5 antagonist, in patients infected with HIV-1. Nat. Med. 2005,
11, 1170–1172.
(3) (a) Markowitz, M.; Nguyen, B. Y.; Gotuzzo, E.; Mendo, F.; Ratana-
suwan, W.; Kovacs, C.; Prada, G.;, Morales-Ramirez, J. O.; Crump-
acker, C. S.; Isaacs, R. D., Gilde, L. R.; Wan, H.; Miller, M. D.;
Wenning, L. A.; Teppler, H. Rapid and Durable Antiretroviral Effect
of the HIV-1 Integrase Inhibitor Raltegravir as Part of Combination
Therapy in Treatment-Naive Patients With HIV-1 Infection: Results
of a 48-Week Controlled Study. JAIDS, J. Acquir. Immune Defic.
Syndr. 2007, 46, 125-133. (b) Saag, M.; Ive, P.; Heera, J.; Tawadrous,
M.; DeJesus, E.; Clumeck, N.; Cooper, D.; Horban, A.; Mohapi, L.;
Mingrone, H.; Reyes-Teran, G.; Walmsley, S.; Hackman, F.; van der
Ryst, E.; Mayer, H. 4th International AIDS Society Conference on
HIV Pathogenesis, Treatment and Prevention, Sydney, Australia, 2007;
WESS104.
Compounds 1, 2c, or TMC-125 were injected sequentially for 3
min at a flow rate of 50 µL·min-1 over the reference and reverse
transcriptase surfaces. At the end of each injection, the dissociation
phase was set to 2 min. Data were also collected during the washing
steps, providing a total of 4.5 min of dissociation-phase information
for each injection. The final dissociation phase after the last injection
of the cycle lasted for 20 min.
For data processing, data were double-referenced by subtracting
both the data obtained from the reference surface, and the buffer
response on the protein surface to remove systematic noise and
instrument drift. Data recorded during the injection phase were not
corrected for DMSO solvent effect. For easy comparison between
data sets, all experiments were normalized by dividing the response
(RU) by the theoretical Rmax value, corresponding to the amount
of protein immobilized for each inhibitor cycle. The data were fitted
with the software BIAevaluation assuming a simple 1:1 association
model (see Supporting Information).
Because of the observed extremely slow off-rates of the
compounds and the impossibility to correctly regenerate the surface
after a first experiment, the interaction was analyzed using the
kinetic titration method described in Karlsson and colleagues41 and
for which the evaluation model was recently made available
commercially. This method is particularly adapted in the case where
the complex is very stable, making it difficult to identify regenera-
tion conditions that dissociate the complex while keeping the
immobilized protein active. It allows the determination of the
kinetics parameters as accurately as the classical multi steps
analysis.
(4) Ren, J.; Stammers, D. K. Structural basis for drug resistance
mechanisms for non-nucleoside inhibitors of HIV reverse transcriptase.
Virus Res. 2008, 134, 157–170.
(5) (a) Boone, L. R. Next-generation HIV-1 non-nucleoside reverse
transcriptase inhibitors. Curr. Opin. InVest. Drugs 2006, 7, 128–135.
(b) Barbaro, G.; Scozzafava, A.; Mastrolorenzo, A.; Supuran, C. T.
Highly Active Antiretroviral Therapy: Current State of the Art, New
Agents and Their Pharmacological Interactions Useful for Improving
Therapeutic Outcome. Curr. Pharm. Des. 2005, 11, 1805–1843.
(6) (a) Brito, M. A.; Rodrigues, C. R.; Cirino, J. J. V.; de Alencastro,
R. B.; Castro, H. C.; Albuquerque, M. G. 3D-QSAR CoMFA of a
Series of DABO Derivatives as HIV-1 Reverse Transcriptase Non-
Nucleoside Inhibitors. J. Chem. Inf. Model. 2008, 48, 1706–1715. (b)
Barreiro, G.; Guimaraes, C. R. W.; Tubert-Brohman, I.; Lyons, T. M.;
Tirado-Rives, J.; Jorgensen, W. L. Search for Non-Nucleoside
Inhibitors of HIV-1 Reverse Transcriptase Using Chemical Similarity,
Molecular Docking, and MM-GB/SA Scoring. J. Chem. Inf. Model.
2007, 47, 2416–2428. (c) Mai, A.; Artico, M.; Rotili, D.; Tarantino,
D.; Clotet-Codina, I.; Armand-Ugon, M.; Ragno, R.; Simeoni, S.;
Sbardella, G.; Nawrozkij, M. B.; Samuele, A.; Maga, G.; Este, J. A.
Synthesis and biological properties of novel 2-aminopyrimidin-4(3H)-
ones highly potent against HIV-1 mutant strains. J. Med. Chem. 2007,
50, 5412–5424. (d) Mai, A.; Sbardella, G.; Artico, M.; Ragno, R.;
Massa, S.; Novellino, E.; Greco, G.; Lavecchia, A.; Musiu, C.; La
Colla, M.; Murgioni, C.; La Colla, P.; Loddo, R. Structure-based
design, synthesis, and biological evaluation of conformationally
restricted novel 2-alkylthio-6-[1-(2,6-difluorophenyl)alkyl]-3,4-dihy-
dro-5-alkylpyrimidin-4(3H)-ones as non-nucleoside inhibitors of HIV-1
reverse transcriptase. J. Med. Chem. 2001, 44, 2544–2554.
(7) (a) Manetti, F.; Este, J. A.; Clotet-Codina, I.; Armand-Ugon, M.; Maga,
G.; Crespan, E.; Cancio, R.; Mugnaini, C.; Bernardini, C.; Togninelli,
A.; Carmi, C.; Alongi, M.; Petricci, E.; Massa, S.; Corelli, F.; Botta,
M. Parallel solution-phase and microwave-assisted synthesis of new
S-DABO derivatives endowed with subnanomolar anti-HIV-1 activity.
J. Med. Chem. 2005, 48, 8000–8008. (b) Botta, M.; Corelli, F.; Petricci,
E.; Radi, M.; Maga, G.; Este´, J. A.; Mai, A. Patent WO 2007/043094
A2, 2007.
(8) Mugnaini, C.; Alongi, M.; Togninelli, A.; Gevariya, H.; Brizzi, A.;
Manetti, F.; Bernardini, C.; Angeli, L.; Tafi, A.; Bellucci, L.; Corelli,
F.; Massa, S.; Maga, G.; Samuele, A.; Facchini, M.; Clotet-Codina,
I.; Armand-Ugo´n, M.; Este´, J. A.; Botta, M. Dihydro-alkylthio-benzyl-
oxopyrimidines as inhibitors of reverse transcriptase: synthesis and
rationalization of the biological data on both wild-type enzyme and
relevant clinical mutants. J. Med. Chem. 2007, 50, 6580–6595.
(9) Nawrozkij, M. B.; Rotili, D.; Tarantino, D.; Botta, G.; Eremiychuk,
A. S.; Musmuca, I.; Ragno, R.; Samuele, A.; Zanoli, S.; Armand-
Ugo´n, M.; Clotet-Codina, I.; Novakov, I. A.; Orlinson, B. S.; Maga,
G.; Este´, J. A.; Artico, M.; Mai, A. 5-Alkyl-6-benzyl-2-(2-oxo-2-
phenylethylsulfanyl)pyrimidin-4(3H)-ones, a series of anti-HIV-1
agents of the dihydro-alkoxy-benzyl-oxopyrimidine family with
peculiar structure-activity relationship profile. J. Med. Chem. 2008,
51, 4641–4652.
X-ray Crystallography. X-ray diffraction data were collected
on a single crystal of 2d using a Siemens P4 four-circle diffrac-
tometer with graphite monochromated Mo KR radiation (λ )
0.71073 Å). The structure was solved by direct methods imple-
mented in the SHELXS-97 program.42 The refinement was carried
out by full-matrix anisotropic least-squares on F2 for all reflections
for non-H atoms by using the SHELXL-97 program.43
Acknowledgment. This study was supported by grants from
the European THiNC and ExCellENT-HIT Consortiums. The
University of Siena and the Spanish BFU2006-00966 and FIS
PI060624 are also acknowledged for financial support. S.Z. is
the recipient of a Buzzati-Traverso Fellowship. This work was
also supported by grants from the French National Agency for
AIDS Research (ANRS) (to E.E.). The following reagent was
obtained through the NIH AIDS Research and Reference
Reagent Program, Division AIDS, NIAID, NIH: TMC-125
(Etravirine), catalogue no. 11609, from Tibotec Pharmaceuticals,
Inc. Prof. Gabriele Cruciani (University of Perugia, Italy) is also
acknowledged for kindly providing us with the GRID program.
Supporting Information Available: Details of synthesis and
analytical data, additional molecular modeling information and a

Chiral Cyclopropyl S-DABO DeriVatiVes
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 3 851
(10) Beroza, P.; Suto, M. J. Designing chiral libraries for drug discovery.
Drug DiscoVery Today 2000, 5, 364–372.
Azijn, H.; Pauwels, R.; Lewi, P. J.; Heeres, J.; Koymans, L. M. H.;
de Jonge, M. R.; Van Aken, K. J. A.; Daeyaert, F. F. D.; Das, K.;
Arnold, E.; Janssen, P. A. J. Evolution of anti-HIV drug candidates.
Part 3: diarylpyrimidine (DAPY) analogues. Bioorg. Med. Chem. Lett.
2001, 11, 2235–2239.
(11) Quaglia, M. G.; Mai, A.; Sbardella, G.; Artico, M.; Ragno, R.; Massa,
S.; Del Piano, D.; Setzu, G.; Dorattiotto, S.; Cotichini, V. Chiral
resolution and molecular modeling investigation of rac-2-cyclopen-
tylthio-6-[1-(2,6-difluorophenyl)ethyl]-3,4-dihydro-5-methylpyrimidin-
4(3H)-one (MC-1047), a potent anti-HIV-1 reverse transcriptase agent
of the DABO class. Chirality 2001, 13, 75–80.
(26) Geitmann, M.; Unge, T.; Danielson, U. H. Interaction Kinetic
Characterization of HIV-1 Reverse Transcriptase Non-nucleoside
Inhibitor Resistance. J. Med. Chem. 2006, 49, 2375–2387.
(27) Geitmann, M.; Unge, T.; Danielson, U. H. Biosensor-Based Kinetic
Characterization of the Interaction between HIV-1 Reverse Tran-
scriptase and Non-nucleoside Inhibitors. J. Med. Chem. 2006, 49,
2367–2374.
(28) Frisch, M. J. ; Trucks, G. W.; Schlegel, H. B.; Scusera, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; . Montgomery, J. A.,
Jr.; Stratmann, R. E., Burant, J. C.; Dapprich, S.; Millam, J. M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara,
A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.;
Chen, W.; Wong, M. W.; Andres, J. L.; Gonzale, C.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98, eVision A.11.2; Gaussian,
Inc.; Pittsburgh, PA, 2001.
(29) Bayly, C. I.; Cieplak, P.; Cornell, W. D.; Kollman, P. A. A well-
behaved electrostatic potential based method using charge restraints
for deriving atomic charges: the RESP model. J. Phys. Chem. 1993,
97, 10269–80.
(30) DeLano, W. L. The PyMOL Molecular Graphics System; DeLano
Scientific LLC: San Carlos, CA, 2002; http://www.pymol.org.
(31) Shindyalov, N.; Bourne, P. E. Protein structure aligment by incremental
combinatorial extension (CE) of the optimal path. Protein Eng. 1998,
11, 739–747.
(32) Pata, J. D.; Stirtan, W. G.; Goldstein, S. W.; Steitz, T. A. Structure of
HIV-1 reverse transcriptase bound to an inhibitor active against mutant
RTs resistant to other non-nucleoside inhibitors. Proc. Natl. Acad. Sci.
U.S.A. 2004, 101, 10548–10553.
(33) Himmel, D. M.; Das, K., Jr.; Hughes, S. H.; Benjahad, A.; Oumouch,
S.; Guillemont, J.; Coupa, S.; Poncelet, A.; Csoka, I.; Meyer, C.;
Andries, K.; Nguyen, C. H.; Grierson, D. S.; Arnold, E. Crystal
Structures for HIV-1 Reverse Transcriptase in Complexes with Three
Pyridinone Derivatives: A New Class of Non-Nucleoside Inhibitors
Effective against a Broad Range of Drug-Resistant Strains. J. Med.
Chem. 2005, 48, 7582–7591.
(34) Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; MerzJr., K. M.;
Ferguson, D. M.; Spellmeyer, D. C.; Fox, T.; Caldwell, J. W.; Kollman,
P. A. A Second Generation Force Field for the Simulation of Proteins,
Nucleic Acids, and Organic Molecules. J. Am. Chem. Soc. 1995, 117,
5179–51977.
(35) Phillips, J. C.; Braun, R.; Wang, W.; Gumbart, J.; Tajkhorshid, E.;
Villa, E.; Chipot, C.; Skeel, R. D.; Kale, L.; Schulten, K. Scalable
molecular dynamics with NAMD. J. Comput. Chem. 2005, 26, 1781–
1802.
(12) Denmark, S. E.; James, E. A. A comparison of (chloromethyl)- and
(iodomethyl)zinc cyclopropanation reagents. J. Org. Chem. 1991, 56,
6974–6981.
(13) (a) Denmark, S. E.; O‘Connor, S. P. Catalytic, Enantioselective
Cyclopropanation of Allylic Alcohols. Substrate Generality. J. Org.
Chem. 1997, 62, 584–594. (b) Denmark, S. E.; O’Connor, S. P.
Enantioselective Cyclopropanation of Allylic Alcohols. The Effect of
Zinc Iodide. J. Org. Chem. 1997, 62, 3390–3401.
(14) C24H24F2N2O2S, Mr 442.52, crystal dimensions: 0.6 mm × 0.5 mm
× 0.3 mm, Triclinic, P1, a ) 9.744(5)Å, b ) 15.490(2) Å, c )
15.970(1) Å, R ) 71.863(5)°, ꢀ ) 83.636(17)°, γ ) 87.130(14)°, Z
) 4, Fcalcd ) 1.291 g cm-3, graphite monochromated Mo KR radiation
(λ ) 0.71073 Å), µ (Mo-KR) ) 0.181 mm-1, T ) 293(2) K, 2θmax
)
50.0°, 8400 measured reflections with Rint ) 0.001, R ) 0.0680 and
wR ) 0.1430 for all reflections, last ∆F map of-0.29 and 0.39 eÅ-3
.
The crystallographic data of the structure have been deposited at the
Cambridge Crystallographic Data Centre with deposit number CCDC
695593. These data can be obtained free of charge via www.ccdc-
.cam.ac.uk/conts/retrieving.html (or from the Cambridge Crystal-
lographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK;
fax: (+44) 1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk).
(15) Data not shown.
(16) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart, W. E.;
Belew, R. K.; Olson, A. J. Automated Docking Using a Lamarckian
Genetic Algorithm and and Empirical Binding Free Energy Function.
J. Comput. Chem. 1998, 19, 1639–1662.
(17) Hopkins, A. L.; Ren, J.; Esnouf, R. M.; Willcox, B. E.; Jones, 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.
(18) Grid, Version 22; Molecular Discovery, Ltd: Marsh Road, Pinner,
Middlesex, UK, 2004.
(19) Ragno, R.; Frasca, S.; Manetti, F.; Brizzi, A.; Massa, S. Docking and
3-D QSAR Studies on Indolyl Aryl Sulfones. Binding Mode Explora-
tion at the HIHIV-Reverse Transcriptase Inhibition: Inclusion of
Ligand-Induced Fit by Cross-Docking Studies. J. Med. Chem. 2005,
48, 200–212.
(20) Kroeger Smith, M. B.; Rouzer, C. A.; Taneyhill, L. A.; Smith, N. A.;
Hughes, S. H.; Boyer, P. L.; Janssen, P. A.; Moereels, H.; Koymans,
L.; Arnold, E.; Ding, J.; Das, K.; Zhang, W.; Michejda, C. J.; Smith,
R. H., Jr. Molecular modeling studies of HIV-1 reverse transcriptase
nonnucleoside inhibitors: total energy of complexation as a predictor
of drug placement and activity. Protein Sci. 1995, 4, 2203–2222.
(21) Hsiou, Y.; Das, K.; Ding, J.; Clark, A. D., Jr.; Kleim, J. P.; Rosner,
M.; Winkler, I.; Riess, G.; Hughes, S. H.; Arnold, E. Structures of
Tyr188Leu mutant and wild-type HIV-1 reverse transcriptase com-
plexed with the non-nucleoside inhibitor HBY 097: inhibitor flexibility
is a useful design feature for reducing drug resistance. J. Mol. Biol.
1998, 284, 313–323.
(22) Das, K.; Clark, A. D.; Lewi, P. J.; Heeres, J.; De Jonge, M. R.;
Koymans, L. M.; Vinkers, H. M.; Daeyaert, F.; Ludovici, D. W.;
Kukla, M. J.; De Corte, B.; Kavash, R. W.; Ho, C. Y.; Ye, H.;
Lichtenstein, M. A.; Andries, K.; Pauwels, R.; Boyer, P. L.; Clark,
P.; Hughes, S. H.; Janssen, P. A.; Arnold, E. Roles of Conformational
and Positional Adaptability in Structure-Based Design of TMC125-
R165335 (Etravirine) and Related Non-nucleoside Reverse Tran-
scriptase Inhibitors That Are Highly Potent and Effective against Wild-
Type and Drug-Resistant HIV-1 Variants. J. Med. Chem. 2004, 47,
2550–2560.
(36) Humphrey, W.; Dalke, A.; Schulten, K. VMDsVisual Molecular
Dynamics. J. Mol. Graphics 1996, 14, 33–38.
(37) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.;
Klein, M. L. Comparison of simple potential functions for simulating
liquid water. J. Chem. Phys. 1983, 79, 926–935.
(38) Essman, U.; Perera, L.; Berkowitz, M. L.; Darden, T.; Lee, H.;
Pedersen, L. A smooth particle mesh Ewald method. J. Chem. Phys.
1995, 103, 8577–8593.
(39) Tuckerman, M.; Berne, B.; Martyna, G. Reversible multiple time scale
molecular dynamics. J. Chem. Phys. 1992, 97, 1990–2001.
(40) Armand-Ugon, M.; Gutierrez, A.; Clotet, B.; Este, J. A. HIV-1
Resistance to the gp41-Dependent Fusion Inhibitor C-34. AntiViral
Res. 2003, 59, 137–142.
(23) Wang, J.; Smerdon, S. J.; Jager, J.; Kohlstaedt, L. A.; Rice, P. A.;
Friedman, J. M.; Steitz, T. A. Structural basis of asymmetry in the
human immunodeficiency virus type 1 reverse transcriptase het-
erodimer. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 7242–7246.
(24) Lindberg, J.; Sigurdsson, S.; Lowgren, S.; Andersson, H. O.; Sahlberg,
C.; Noreen, R.; Fridborg, K.; Zhang, H.; Unge, T. Structural basis for
the inhibitory efficacy of efavirenz (DMP-266), MSC194 and
PNU142721 towards the HIV-1 RT K103N mutant. Eur. J. Biochem.
2002, 269, 1670–1677.
(41) Karlsson, R.; Katsamba, P. S.; Nordin, H.; Pol, E.; Myszka, D. G.
Analyzing a kinetic titration series using affinity biosensors. Anal.
Biochem. 2006, 349, 136–147.
(42) Sheldrick, G. M., SHELXS-97, Release 97-2, Program for the Solution
of Crystal Structures; University of Go¨ttingen: Go¨ttingen, Germany,
1997.
(43) Sheldrick G. M. SHELXL-97, Release 97-2, Program for Crystal
Structure Refinement; University of Go¨ttingen: Go¨ttingen, Germany,
1997.
(25) Ludovici, D. W.; De Corte, B. L.; Kukla, M. J.; Ye, H.; Ho, C. Y.;
Lichtenstein, M. A.; Kavash, R. W.; Andries, K.; de Bethune, M. P.;
JM801330N