ABCB1 structural models, molecular docking, and synthesis of new oxadiazolothiazin-3-one inhibitors

Article abstract of DOI:10.1021/ml300436x

Docking methods are powerful tools for in silico screening and drug lead generation and optimization. Here, we describe the synthesis of new inhibitors of ABCB1 whose design was based on construction and preliminary confirmation of a model for this membrane transporter of the ATP-binding cassette family. We chose the strategy to build our three-dimensional model of the ABCB1 transporter by homology. Atomic coordinates were then assayed for their reliability using the measured activity of some oxadiazolothiazin-3-one compounds. Once established their performance by docking analysis, we synthesized new compounds whose forecasted activity was tested by MTT and cytofluorimetric assays. Our docking model of MDR1, MONBD1, seems to reliably satisfy our need to design and forecast, on the basis of their LTCC blockers ability, the inhibitory activity of new molecules on the ABCB1 transporter.

Full text of DOI:10.1021/ml300436x

Letter
pubs.acs.org/acsmedchemlett
ABCB1 Structural Models, Molecular Docking, and Synthesis of New
Oxadiazolothiazin-3-one Inhibitors
Camillo Rosano,† Maurizio Viale,*,‡ Barbara Cosimelli,§ Elda Severi,§ Rosaria Gangemi,‡ Alessia Ciogli,∥
Daniela De Totero,⊥ and Domenico Spinelli*,#
†IRCCS SM-IST, U.O.S. Biopolimeri e Proteomica, Largo R. Benzi 10, 16132 Genova, Italy
‡IRCCS SM-IST, U.O.C. Terapia Immunologica, Largo R. Benzi 10, 16132 Genova, Italy
§Dip. di Farmacia, Universita
∥Dip. di Chimica e Tecnologie del Farmaco, Sapienza Universita
̀
degli Studi di Napoli “Federico II”, Via D. Montesano 49, 80131 Napoli, Italy
di Roma, Piazzale A. Moro 5, 00185 Roma, Italy
̀
⊥IRCCS SM-IST, U.O.C. Trasferimento Genico, Largo R. Benzi 10, 16132 Genova, Italy
#Dip. di Chimica “G. Ciamician”, Universita
̀
degli Studi di Bologna, Via F. Selmi 2, 40126 Bologna, Italy
S
* Supporting Information
ABSTRACT: Docking methods are powerful tools for in silico
screening and drug lead generation and optimization. Here, we describe
the synthesis of new inhibitors of ABCB1 whose design was based on
construction and preliminary confirmation of a model for this membrane
transporter of the ATP-binding cassette family. We chose the strategy to
build our three-dimensional model of the ABCB1 transporter by
homology. Atomic coordinates were then assayed for their reliability
using the measured activity of some oxadiazolothiazin-3-one compounds.
Once established their performance by docking analysis, we synthesized
new compounds whose forecasted activity was tested by MTT and
cytofluorimetric assays. Our docking model of MDR1, MONBD1, seems
to reliably satisfy our need to design and forecast, on the basis of their
LTCC blockers ability, the inhibitory activity of new molecules on the
ABCB1 transporter.
KEYWORDS: ABCB1 inhibitors, multidrug resistance, docking, drug design, LTCC blockers compounds, oxadiazolothiazin-3-ones
he overexpression of the multidrug resistance-1 (MDR1
or ABCB1) gene has been associated with the onset of
cokinetic, and toxic activities.9−13 In this framework, we have
considered different LTCC (L-type calcium channels) blocker
compounds (derivatives of [1,2,4]oxadiazolo[3,4-c][1,4]-
thiazin-3-ones 1 and 214), which were already tested for their
ability to compete for excretion with a typical drug excreted by
Pgp-170, doxorubicin, causing its accumulation into multidrug-
resistant A2780/DX3 cells (Figure 2).14
Here, we describe the construction of a 3D model of Pgp-
170 that was used for drug modeling studies followed by the
synthesis of the new compounds so designed (2m−x). In
particular, to perform this study, we chose the strategy of
modeling the Pgp-170 transporter by homology. This method
may be useful to rationally design novel moieties that can act as
inhibitor of this protein and that can be directed to the TMD
and/or the NBD1 and NBD2 domains located within the
cytoplasm. We tested a databank of small molecules by
simulating docking procedures to these three main zones of
Pgp-170.
T
intrinsic or acquired drug resistance in many tumor
histotypes.1−3 This gene encodes for the membrane ATP-
dependent transporter P-glycoprotein-170 (Pgp-170) able to
pumping out of the cells a number of cationic or neutral
hydrophobic xenobiotics and drugs4,5 by an entropically driven
flip-flop mechanism accounting for the broad range of
substrates exported. It acts as bellows that widen the lower
strand of the bilayer hunting for xenobiotics from the
cytoplasmic interface or the lipid bilayer itself and then push
these substrates out of the cell.
Protein structure consist of two protein halves each made by
a cytoplasmatic nucleotide binding domain (NBD1 and
NBD2), which acts synergistically on the substrate, and by a
6 transmembrane helices domain (TMD) forming a channel
(Figure 1A). Both the NBDs share motifs typical of the
domains involved in ATP binding and highly conserved within
the transporter family. They contain the Walker A and B motifs
as well as the ABC signature motif.6
Many compounds have been used to overcome MDR,7,8 but
today no one has shown a significant clinical efficacy because of
a detrimental unbalance between pharmacodynamic, pharma-
Received: January 21, 2013
Accepted: May 13, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ml300436x | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
Figure 1. (A) Ribbon representation of the three-dimensional structure of the MDR1 in open conformation. NBD1 is on the left of the figure;
NBD2 stands on the right, in a darker gray. The transmembrane helices are drawn in a lighter gray. The small molecules binding site area is
represented enclosed by a rectangle. (B) Superposition of 2k (lighter gray sticks) to 2l (darker gray sticks) inside the ligand binding pocket. NBD2 is
drawn as a dark gray ribbon. The relative position of the transmembrane helices TM4 and TM5 of MDR1 in a closed conformation (light gray) is
reported to highlight the sterical clashes with the ligands (which are preventing the pump closure).
Being the result of this simulation not compatible with the
results obtained by in vitro screening, we repeated the docking
procedures using as a target the atomic model MONDB1.
Structural superposition of the two NBD2 models (MOMDR1
and MONBD1) shows a substantial identity among the two
structures. The overall root-mean-square deviation (rmsd)
calculated on the position of 256 Cα atoms, in fact, was 2.21 Å.
Nevertheless an important difference occurs in the proximity of
the supposed binding site where the helix Asn1105-Leu1113
stands closer to loop Tyr1133−Ser1141 in MOMDR1 model
with respect to NBD1 derived model. This difference seems to
be sufficient to inhibit the binding capacity of the docked
molecule to this site rendering MOMDR1 not suitable for our
in silico test. Moreover, the supposed ligand binding pocket of
NBD2 is placed in correspondence to a cleft that in the open
conformation of MDR1 is occupied by the loop connecting TM
helixes TM4 and TM5. The docking procedures adopted
consider the protein as a rigid object, and thus it would be not
possible for a molecule to bind within the considered cavity,
due to steric hindrance. This reason made us think to target our
molecules to the MONBD1. All the data about in vitro
Figure 2. Oxadiazolothiazin-3-one derivative structure. In the box, the
molecules 2l−2x synthesized and/or selected following MONBD1
docking indications.
screening were obtained from our previous paper.14
On the basis of the models built, we docked 32 molecules
(16 moieties, Figure 2: 1a−e and 2a−2k in R and S
configurations; in 1 this approach is significant for interaction
study, really enantiomer pairs spontaneously transform one into
the other)17 using as target the homology model of NBD2
domain. These simulations allowed us to screen our library and
to determine some ligands to be further compared with in vitro
analysis (see Figure S4, Supporting Information, as an example
of the different docking of molecules with different forecasted
inhibitory activities).
The S-isomers appear more effective than the R-ones. This
different biological activity is well in line with expectations. In
fact, while for most chemical reactions (without interactions
with chiral reagents) the two enantiomers are indistinguishable;
in contrast, they can show different biological effects because of
the presence of chiral centers in living organisms. Because the
best fitting in our docking model always corresponded to the S-
configuration, we compared the Pgp-170 inhibiting activity of
our S compounds (calculated as the percent reduction of IC50s
The availability of the crystal structures of the MDR1 from
mouse,15 a protein sharing 80.9% amino acids identity with the
human Pgp-170, allowed us to build a first three-dimensional
model to be used as a target (MOMDR1). The simulations
using MOMDR1 as a target were not supported by in vitro
analysis. In fact, out of the LTCC blockers previously tested,14
the most promising candidates would be 2b, 2d, 2e, and 2f each
bound to NBD2. Acceptable binding modes would also be
found for 2j and 2k but with lower calculated binding affinity
compared to a second model built based on the homology with
the human NBD1 domain (MONBD1).
MONBD1 was built using as a template the three-
dimensional coordinates of human MDR1 Nucleotide Binding
Domain 1, this domain sharing 52.4% sequence identity with
human NBD2, as previously described.16
Eventually, 1c would be bound to NBD1, while no evidence
of activity was found in in vitro analysis.
B
dx.doi.org/10.1021/ml300436x | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
of doxorubicin obtained with compound concentrations giving
per se 5% reduction of cell proliferation)14 separated in three
groups of high, modest, or absent fitting, as suggested by
docking analysis. Our statistical analysis showed a correlation
with p = 0.007, while, in contrast, the MOMDR1 model did not
correlate with the in vitro activity of our molecules (Figure 3)
Visual inspection of the three-dimensional structure of the
protein model MOMDR1 showed that the ligand binding
pocket is a mainly polar pocket in which six hydrophilic
residues, E1084, R1085, R1110, S1117, R1188, and R1192,
contribute to stabilize the ligand through hydrogen bonds with
their polar groups, while residues F1086, L1109, L1113, I1115,
I1121, and A1189 provide hydrophobic contacts. This site is
about 11 Å apart from the ATP binding site as determined by
similarity with the NBD1 domain, evidencing how our
molecules are neither ATP competitors (see Figure S5,
Supporting Information, for details) nor occupy the Pgp-170
drug binding site, which is hosted in a large hydrophobic cavity
in the TMD domain (Figure 1B). Furthermore, the identified
binding site is coincident with a cleft hosting the loop
connecting TM4 and TM5 (residues A255-Q266) in the
open conformation (Figure 1B); the presence of a molecule
bound in this cavity could lead Pgp-170 to freeze in a closed
conformation blocking its unpalatable export of anticancer
drugs from the cell.
According to docking simulations, it appears evident how the
alkyl chain of 2k is pointing toward a deep hydrophobic hole
made by residues V1080, L1083, L1113, I1115, I1196, L1198,
and I1228.
Figure 3. Comparison between data about the relative activity of our
molecules separated in three grouping variables of high [n = 4 and 0,
for MONBD1 (black bars) and MOMDR1 (white bars) docking
models], modest (n = 6 and 11), and absent activity (n = 6 and 5), as
classified by our docking analysis. The Kruskall−Wallis test for
nonparametric data was used for the correlation between groups (p =
0.007 for MONBD1 model), while the nonparametric Mann−Whitney
test was used to compare the groups in pairs (*p = 0.011; **p =
0.109). Data shown as mean SE.
To improve the ligand binding to this protein site, we
modified the structure of 2k into 2l by lengthening acetal chain
by a further CH2. Interestingly, the comparison between
dockings of 2k and 2i suggests a better influence of the Z-
isomers on pharmacological activity, although when we
considered the structures with 6 atom chains (2l vs 2r), the
Z-isomer did not seem to confer significant advantages in terms
of inhibitory activity (2r, the E-isomer was considered a good
ligand although displaying a less favorable clusterization of the
docking simulation results). In this case, the best binding
affinity was 8.7 × 10−8 M, similar to that of 2l, suggesting a
higher influence on activity for the alkyl chain length.
Thus, we considered two others molecules containing a
saturated alkyl chain of six carbon atoms (2m and 2n) with
chlorine or bromine substituted moieties. As described, these
modifications should lead to higher affinities or to under-
standing the influence of some structural modifications on
pharmacological activity and should represent a first step for
identification of more efficient MDR1 inhibitors.
The first of these modifications, which generated the
synthesis of 2l (see above), was suggested by docking analysis
in order to increase the binding affinity (theoretical Ki = 3.28 ×
10−8 M vs 1.11 × 10−7 M) and allowed the synthesis of one
new Z-isomer that caused an increase of the potency, calculated
as a decrease of IC50, of about 71%, i.e., with a 24% increase of
activity compared to 2k (relative activity 44%, Table 1). As for
2m and 2n, they differentiate each other for chlorine or
bromine in para position on the benzene ring. The measured in
vitro activity of these molecules reflected the predictions of
MONBD1. In fact, in vitro assays showed that only 2n
displayed a significant activity of 63 10% of IC50 reduction
(34% increase of activity compared to 2k), underlining the
importance of bromine and of length of the alkyl chain in
determining the MDR1 inhibitory activity. The synthesis of 2o
similar to 2k but with a bromine substituting chlorine
confirmed the importance of halogen, realizing a 72% increase
of activity compared to 2k (44 vs 81%, Table 1).
confirming that our docking model may be considered reliable
as reference for the molecular design of new compounds with
the same mechanism of MDR1 inhibiting activity. The results
of the simulations were in good agreement with the
experiments carried in vitro using the racemic mixtures of all
compounds and allowed to identify molecules with a potential
activity on Pgp-170. The use of data of racemic mixtures could
seem ambiguous, but their use seemed reasonable on the basis
of our docking analysis and helped us to avoid wasting time to
separate all compounds in their S- and R-isomers. However, our
choice was also confirmed and rewarded by the results obtained
with the enantiomers of 2k, 2p, and 2q (see later). Moreover,
the not extremely low theoretical values for Ki constants
evaluated by docking analysis, from 10−6 to 10−8 M, may justify
the fact that when using for comparison the data obtained from
concentrations giving per se 0% reduction of cell proliferation
we did not observe a significant correlation.
Out of docked molecules, 2g and 2h, indicated as good by
simulations, were not considered as ligands due to the long
hydrophobic tail, which was found entangled to the protein
surface and not interfering with the protein activity. Compound
1a showed a favorable clusterization with good calculated
affinity constant not validated in vitro. Interestingly, 2j and 2k
(S-isomers in E- or Z-configuration) were found to bind to
NBD2 in a surface cleft deputy to host the loop connecting
helices TM4 and TM5, 2j (bromine substituted) being a larger
favorite due to a better clusterization of the results (perhaps for
the larger atomic radius of bromine in comparison to chlorine).
The identification of this site as a potential binding site for our
molecules justify that 1a was not found as active by in vitro
experiments, being its putative binding site was differently
located.
The in vitro inhibitory activity was also tested for the single
enantiomers of 2k: 2p and 2q. Looking at their physical
properties (see ECD, αD in Supporting Information) and
C
dx.doi.org/10.1021/ml300436x | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
All these data were also confirmed in vitro. In fact, the
reduction of doxorubicin IC50 on resistant A2780/DX3 cells by
the IC50 of molecules with the configurations R,E (2s), S,E (2t),
R,Z (2u), and S,Z (2v) was 30 9%, 27 12%, 37 15%, and
47 15%, respectively, remembering that the reduction of the
IC50 due to the activity of the reference compound 2k was on
average 44%.
Table 1. Comparison between the Parent Compound 2k and
the New ABCB1 Inhibitors Designed by the Docking Model
MONBD1
gain/loss in
inhibitory
activity
compared
to 2k
gain/loss in
doxorubicin
accumulation
MDR1
inhibitory
compared
a
b
compd
activity
to 2k
Our results suggest, on one hand, that, in the case of our
inhibitors, the simultaneous presence of single structural
modifications, able per se or even when coupled to improve
the inhibition of the MDR1 transporter, may not be able to
cause an additive or synergistic effect in our test conditions. On
the other hand, these results confirm the validity and reliability
of our hybrid structural model for predicting the inhibitory
activity of our ABCB1 inhibitors prompting us for a further
development of new ABCB1 inhibitors.
Therefore, we are designing, synthesizing, and testing new
derivatives of our more active compounds 2o and 2r to still
increase the inhibitory activity of these compounds.
In conclusion, our docking model of MDR1, MONBD1, in
spite of the relatively low number of structures used to confirm
its validity, seems to reliably satisfy our need to forecast,
considering their chemical structure, the inhibitory activity of
our compounds on the Pgp-170 transporter and to design new
molecules with an inhibitory mechanism.
presence of Cl
or Br
2k
2o
2m
2n
2p
2q
2k
2l
−44 6%
−81 8%
−23 10%
−63 10%
−16 13%
−64 10%
−44 6%
−71 7%
−44 6%
−16 7%
−71 7%
−78 9%
100%
184%
52%
100%
139%
18%
143%
36%
72%
absolute
configuration
48%
145%
100%
161%
100%
36%
136%
100%
131%
100%
35%
alkyl chain
length
Z- or E-
isomers
2k
2i
2l
161%
177%
131%
109%
2r
a
The activity was evaluated as percent reduction of doxorubicin IC50.
b
Data shown as mean
accumulation was calculated compared to that of lead compound 2k
(100%).
SD. The gain/loss in doxorubicin
considering some our previous results,18−20 we suggest to
consider the levogire 2p as the R-enantiomer and the dextrogire
2q as the S-one, respectively. From in vitro inhibitory activity
results (Table 1) only 2q (the S-enantiomer) showed a highly
significant pharmacological activity (64% of IC50 reduction,
with a 36% increase of activity compared to the racemic mixture
2k, while 2p was virtually inactive: −16 13%). This result
once again confirms the predictive reliability of our MONBD1
model emphasizing the importance of configuration of
inhibitors and the lower suggested activity of R-enantiomers.
To confirm all these results, we also assayed cytofluorimetri-
cally the ability of newly synthesized compounds to increase the
accumulation of doxorubicin in our A2780/DX3 cell model in
comparison with the original 2k. From results in Table 1, it is
evident that the compounds showing activity in antiproliferative
assay were also more prone to block the efflux of doxorubicin
via the ABCB1 transporter.
ASSOCIATED CONTENT
* Supporting Information
Materials and instruments used, synthesis and spectroscopic
details of compounds, enantiomer separation, description of
docking simulation, and biological assays. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
*E-mail: domenico.spinelli@unibo.it (D.S.); maurizio.viale@
istge.it (M.V.).
Author Contributions
The manuscript was written through contributions of all
authors.
■
Funding
This research was partially supported by funds from MIUR
(PRIN 20078J9L2A_005 and 20085E2LXC_004).
Considering our docking results, indicating that the presence
of the S-configuration, the unsaturated 6C chain, and the
bromine in benzene ring could increase the activity of our
ABCB1 transporter inhibitors, we finally designed four
molecules (2s−2v) all containing the bromine and all possible
combinations of R/S and Z/E conformations. They were
obtained from racemic 2w and 2x by chiral separation. We then
activated the docking analysis to study their possible activity as
MDR1 inhibitors.
Notes
The authors declare no competing financial interest.
ABBREVIATIONS
■
ABCB1, ATP-binding cassette subfamily B member 1; LTCC,
L-type calcium channels; MDR1, multidrug resistance 1; NBD,
nucleotide binding domain; Pgp-170, P-glycoprotein-170;
TMD, transmembrane domain
Once we performed the docking analysis, surprisingly, we
obtained for these molecules and, especially for those in S
configuration, a score contradicting previous hypotheses. In
fact, the sum of the above active chemical configurations was
supposed to improve the score for the inhibiting activity, if
compared to the single or coupled structural combinations.
Instead, the activity predictable by docking analysis was not
much better or even simply better than reference compound
2k. In particular, (S,Z) 2v had a score similar to that of 2k (high
activity), while the other stereoisomers had a lower score
(modest activity).
REFERENCES
■
(1) Kourti, M.; Vavatsi, N.; Gombakis, N.; Sidi, V.; Tzimagiorgis, G.;
Papageorgiou, T.; Koliouskas, D.; Athanassiadou, F. Expression of
multidrug resistance 1 (MDR1), multidrug resistance-related protein 1
(MRP1), lung resistance protein (LRP), and breast cancer resistance
protein (BCRP) genes and clinical outcome in childhood acute
lymphoblastic leukemia. Int. J. Hematol. 2007, 86, 166−173.
(2) Hille, S.; Rein, D. T.; Riffelmann, M.; Neumann, R.; Sartorius, J.;
Pfutzner, A.; Kurbacher, C. M.; Schondorf, T.; Breidenbach, M.
Anticancer drugs induce mdr1 gene expression in recurrent ovarian
cancer. Anticancer Drugs 2006, 17, 1041−1044.
̈
̈
D
dx.doi.org/10.1021/ml300436x | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
(3) Takara, K.; Sakaeda, T.; Okumura, K. An update on overcoming
MDR1-mediated multidrug resistance in cancer chemotherapy. Curr.
Pharm. Des. 2006, 12, 273−286.
chiral oxadiazol-3-one calcium channel blocker, resolved using chiral
chromatography, via concerted density functional theory calculations
of its vibrational circular dichroism, electronic circular dichroism, and
optical rotation. J. Org. Chem. 2007, 72, 4707−4715.
(20) Ioan, P.; Ciogli, A.; Sirci, F.; Budriesi, R.; Cosimelli, B.; Pierini,
M.; Severi, E.; Chiarini, A.; Cruciani, G.; Gasparrini, F.; Spinelli, D.;
Carosati, E. Absolute configuration and biological profile of two
thiazinooxadiazol-3-ones with L-type calcium channel activity: a study
of the structural effects. Org. Biomol. Chem. 2012, 10, 8994−9003.
(4) Sharom, F. J. ABC multidrug transporters: structure, function and
role in chemoresistance. Pharmacogenomics 2008, 9, 105−127.
́ ́
(5) Sarkadi, B.; Homolya, L.; Szakacs, G.; Varadi, A. Human
multidrug resistance ABCB and ABCG transporters: Participation in a
chemoimmunity defense system. Physiol. Rev. 2006, 86, 1179−1236.
(6) Ramaen, O.; Leulliot, N.; Sizun, C.; Ulryck, N.; Pamlard, O.;
Lallemand, J. Y.; Tilbeurgh, H.; Jacquet, E. Structure of the human
multidrug resistance protein 1 nucleotide binding domain 1 bound to
Mg2+/ATP reveals a non-productive catalytic site. J. Mol. Biol. 2006,
359, 940−949.
(7) Dean, M.; Fojo, T.; Bates, S. Tumour stem cells and drug
resistance. Nat. Rev. Cancer 2005, 5, 275−284.
́
(8) Avendano, C.; Menendez, J. C. Inhibitors of multidrug resistance
̃
to antitumour agents (MDR). Curr. Med. Chem. 2002, 9, 159−193.
(9) Fischer, V.; Rodríguez-Gascon, A.; Heitz, F.; Tynes, R.; Hauck,
́
C.; Cohen, D.; Vickers, A. E. The multidrug resistance modulator
Valspodar (PSC 833) is metabolized by human cytochrome P450 3A.
Implications for drug−drug interactions and pharmacological activity
of the main metabolite. Drug Metab. Dispos. 1998, 26, 802−811.
(10) Gottesman, M. M.; Fojo, T.; Bates, S. E. Multidrug resistance in
cancer: role of ATP-dependent transporters. Nat. Rev. Cancer 2002, 2,
48−58.
́
(11) Solary, E.; Drenou, B.; Campos, L.; de Cremoux, P.; Mugneret,
F.; Moreau, P.; Lioure, B.; Falkenrodt, A.; Witz, B.; Bernard, M.;
Hunault-Berger, M.; Delain, M.; Fernandes, J.; Mounier, C.; Guilhot,
F.; Garnache, F.; Berthou, C.; Kara-Slimane, F.; Harousseau, J. L.
Quinine as a multidrug resistance inhibitor: a phase 3 multicentric
randomized study in adult de novo acute myelogenous leukemia. Blood
2003, 102, 1202−1210.
(12) Van der Holt, B.; Lowenberg, B.; Burnett, A. K.; Knauf, W. U.;
̈
Shepherd, J.; Piccaluga, P. P.; Ossenkoppele, G. J.; Verhoef, G. E.;
Ferrant, A.; Crump, M.; Selleslag, D.; Theobald, M.; Fey, M. F.;
Vellenga, E.; Dugan, M.; Sonneveld, P. The value of the MDR1
reversal agent PSC-833 in addition to daunorubicin and cytarabine in
the treatment of elderly patients with previously untreated acute
myeloid leukemia (AML), in relation to MDR1 status at diagnosis.
Blood 2005, 106, 2646−2654.
(13) Nobili, S.; Landini, I.; Mazzei, T.; Mini, E. Overcoming tumor
multidrug resistance using drugs able to evade P-glycoprotein or to
exploit its expression. Med. Res. Rev. 2012, 32, 1220−1262.
(14) Viale, M.; Cordazzo, C.; Cosimelli, B.; de Totero, D.;
Castagnola, P.; Aiello, C.; Severi, E.; Petrillo, G.; Cianfriglia, M.;
Spinelli, D. Inhibition of MDR1 activity in vitro by a novel class of
diltiazem analogues: toward new candidates. J. Med. Chem. 2009, 52,
259−266. Viale, M.; et al. Multidrug resistance 1 gene (MDR1). SciBX
2009, DOI: 10.1038/scibx.2009.44.
(15) Schmitt, L.; Tampe, R. Affinity, specificity, diversity: a challenge
for the ABC transporter TAP in cellular immunity. ChemBioChem
2000, 1, 16−35.
(16) Aller, S. G.; Yu, J.; Ward, A.; Weng, Y.; Chittaboina, S.; Zhuo, R.
P.; Harrell, P. M.; Trinh, Y. T.; Zhang, Q. H.; Urbatsch, I. L.; Chang,
G. Structure of P-glycoprotein reveals a molecular basis for poly-
specific drug binding. Science 2009, 323, 1718−1722.
(17) Hemithioacetals 1 are stereochemically unstable, so recalling the
well known situation observed in hemiacetals. At now, by dynamic
NMR, the racemization barrier of some hemithioacetals 1 has been
calculated as meaning the impracticability to isolate at room
temperature the two relevant enantiomers (Cosimelli, B.; Mazzanti,
A.; Spinelli, D. ; et al.).
(18) Carosati, E.; Cruciani, G.; Chiarini, A.; Budriesi, R.; Ioan, P.;
Spisani, R.; Spinelli, D.; Cosimelli, B.; Fusi, F.; Frosini, M.; Matucci,
R.; Gasparrini, F.; Ciogli, A.; Stephens, P. J.; Devlin, F. J. Calcium
channel antagonists discovered by a multidisciplinar approach. J. Med.
Chem. 2006, 49, 5206−5216.
(19) Stephens, P. J.; Devlin, F. J.; Gasparrini, F.; Ciogli, A.; Spinelli,
D.; Cosimelli, B. Determination of the absolute configuration of a
E
dx.doi.org/10.1021/ml300436x | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX