Novel non-substrate modulators of the transmembrane efflux pump P-glycoprotein (ABCB1)

Article abstract of DOI:10.1039/c4md00506f

Novel N- and 4-substituted 1,4-dihydropyridines with a C₂-symmetric molecular scaffold have been profiled as highly active modulators of the transmembrane efflux pump P-glycoprotein (P-gp, ABCB1) in an exclusively P-gp overexpressing cell line

Full text of DOI:10.1039/c4md00506f

MedChemComm
View Article Online
View Journal
CONCISE ARTICLE
Novel non-substrate modulators of the
transmembrane efflux pump P-glycoprotein
Cite this: DOI: 10.1039/c4md00506f
(ABCB1)†
Sören Krawczyk,a Christiane Baumert,a Joséf Molnár,b Christoph Ritter,c
a
d
a
*
Jens Höpner, Charlotte Kloft and Andreas Hilgeroth
Novel N- and 4-substituted 1,4-dihydropyridines with a C2-symmetric molecular scaffold have been
profiled as highly active modulators of the transmembrane efflux pump P-glycoprotein (P-gp, ABCB1) in an
exclusively P-gp overexpressing cell line model. Structure–activity relationships have been discussed for
varied substituents of both the N- and the 4-residue. The influence of potential hydrogen bond acceptor
functions has been characterized in relation to the number and position of the substituents. Cellular toxicity
has been closely considered and the P-gp substrate properties are suggested as the limiting molecular
properties of known P-gp modulators. The non-toxicity and non-substrate properties of our novel inhibitors
qualify this novel compound class as a prospective tool to effectively combat the efflux pump-mediated
cellular resistance of anticancer drug substrates, as could be demonstrated in the first in vitro studies.
Received 7th November 2014,
Accepted 14th January 2015
DOI: 10.1039/c4md00506f
www.rsc.org/medchemcomm
Therapeutic cancer treatment remains a great challenge
although many promising drugs have been developed in the
past.1 Further, the critical side effects of drugs, which also target
normal cells, can be limited by addressing specific structures in
cancer cells like receptor tyrosine kinases.2,3 Such overexpressed
kinases, which are found dysregulated in tumor cells, can
be blocked by monoclonal antibodies or ATP competitive
inhibitors.1,4,5 Moreover, the use of drug carriers like liposomes
or nanoparticles is investigated to preferably address the
respective tumor cells and avoid effects in normal cells.6
The problem with all these novel drugs is that they mostly
turn out to be substrates of transmembrane efflux pumps
which are found in tumor cells and are partly induced by the
drugs themselves.7–9 Uptaken drugs are transported out of
the cells so that the therapeutically necessary drug levels are
no longer reached.7 The tumor cell becomes resistant, and
moreover the resistance mostly includes a great number of
anticancer drugs of various drug families.10 There is still a
discussion about the cause of this multidrug resistance
(mdr), which may be a multiple binding site for various drug
structures.11 However, the mode of action of the various
transmembrane efflux pumps is still under debate and as
three-dimensional structures are missing, the discussion is
concentrated on the models of the efflux pump function.10,12,13
P-glycoprotein (P-gp, ABCB1) is the most important efflux
pump and is widespread in various types of cancer.14 P-gp is
the longest known efflux pump found in cancer cells and
there have been intense efforts to gain insight into the struc-
ture of this transmembrane protein.14 P-gp is known to con-
sist of two domains, each with six α-helical subunits. A low
resolution image of the structure obtained using fluorescence
spectroscopy suggested a C2 symmetrical molecular structure
with nearly all of the α-helical elements of each domain
being located around the C2 axis which is centered in the
middle of the molecule.15 This closer insight suggested the
favouring of the development of C2 symmetric molecules
which may serve as modulators of the efflux pump activity. Over
the last decades there have been several efforts to influence the
efflux pump activities by the use of such potential inhibitors.
Most candidates turned out to be substrates of the efflux
pumps themselves so that unfavourable high concentrations
were necessary to achieve clinical effects.16–18 However, until
now none of the modulators were successful in clinical trials.
Nifedipine was one of the early modulators.19,20 Structural
modifications helped to reduce the calcium-antagonistic
properties but although the potential to modulate the P-gp
a Department of Pharmaceutical Chemistry, Institute of Pharmacy, Martin Luther
University, Wolfgang-Langenbeck-Strasse 4, 06120 Halle, Germany.
E-mail: andreas.hilgeroth@pharmazie.uni-halle.de; Tel: +49 345 5525168
b Department of Medical Microbiology, University of Szeged, Dom tér 10,
6720 Szeged, Hungary
c Department of Clinical Pharmacy, Institute of Pharmacy, Ernst Moritz Arndt
University of Greifswald, Friedrich-Ludwig-Jahn-Strasse 17, 17489 Greifswald,
Germany
d Department of Clinical Pharmacy, Institute of Pharmacy, Free University of
Berlin, Kelchstrasse 21, 12169 Berlin, Germany
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4md00506f
This journal is © The Royal Society of Chemistry 2015
Med. Chem. Commun.

View Article Online
Concise Article
MedChemComm
activity increased, the derivatives were less active than
verapamil, which is one of the best P-gp modulators in such
cellular in vitro studies.21,22 However, the proven P-gp
substrate properties of the compounds hampered further
developments and caused the clinical ineffectivity of the
early investigated nifedipine.16,20 All the 1,4-dihydropyridines
of the series were so far unsubstituted at the nitrogen. We
firstly investigated a series of N-alkyl substituted 1,4-dihydro-
pyridines with a maintained C2 symmetry of the substituted
1,4-dihydropyridine skeleton. Both the N-alkyl substituent
and the 4-aryl substituent were structurally varied, a number
of hydrogen bond acceptor functions were introduced and
the P-gp modulating activities were characterized in a P-gp
cell model with exclusively overexpressing P-gp. Finally, bio-
analysis of selected compounds was carried out to provide
a compound with no cellular toxicity and to investigate
possible P-gp substrate properties. Thus, novel non-
substrate modulators could be identified for further pre-
clinical studies.
The target compound (4) was obtained in a simple one-
pot reaction of the respective aromatic aldehyde (1), methyl
propiolate (2) as the reacting carbonyl compound, and finally
the aliphatic amine (3) with various substitutions in glacial
acetic acid (Scheme 1). We preferably used methyl propiolate
because the resulting target compound (4) partly crystallized
from the reaction mixture after cooling so that extensive
purification procedures were not necessary. The simple
synthesis work-up procedure is also an advantage of the
compound class.
The P-gp modulating properties were determined in a
mouse T-lymphoma cell line model. The parental cell line
expresses no human P-gp and the subline expresses human
P-gp after mdr1 gene transfection with a retrovirus. This
procedure ensured that only P-gp is expressed in the subline
and no other transporter is available and detectable. The
subline resulted from the consequent collection of surviving
cells under colchicine treatment in the cell culture. Only
those P-gp expressing cells were able to transport the cell-
toxic colchicine out of the cells. We used Rhodamine 123 as
a fluorescent and P-gp specific substrate in our assay and
determined the cellular uptake of the substrate in both cell
lines using a flow cytometry technique at a wavelength of
530 nm using a laser with fluorescence excitation at 488 nm,
so that compound fluorescence bands at lower wavelengths
are not detectable. We then added increasing concentrations
of the inhibitor from stock solutions and determined
the resulting fluorescence after Rhodamine addition and
further cell incubation. In those P-gp expressing cells with a
P-gp inhibiting effect of the inhibitors, the cellular uptake
of the fluorescent substrate increased. The so-called FAR
(fluorescence activity ratio) values were calculated as a rela-
tion of the fluorescence in the P-gp expressing and the non-
expressing cell line, each under inhibitor treatment and with
each fluorescence related to that of the untreated control.
Consequently, FAR values > 1 proved the P-gp modulating
properties of the respective investigated compound. We
used verapamil as an effective in vitro standard inhibitor as
well as tariquidar. Although the concentration-dependent
effects of the P-gp modulation increased, we could not reach
a saturation of the effect due to the observed cellular toxicity
of both compounds at a concentration higher than 10 μM.23
We discuss concentration-dependent FAR values similar
to all publications in this field so far which report partly
different saturation effects for compounds in one com-
pound class. Also in our compound class we reached a satu-
ration of the inhibiting effect for only one compound. In all
the other cases of compound inhibition, no saturation of
the modulating effect was observed, so that it made no
sense to determine the IC50 values. The use of higher con-
centrations to reach such saturation effects would probably
lead to toxic effects. Additionally, a limited compound solu-
bility at high concentrations may prevent the saturation
effects from being reached.
Moreover, the value of 1.1 is the lowest possible FAR
value which complicated such a calculation. So for the dis-
cussion of structure–activity relationships we considered our
observed concentration-dependent effects which were found
for all investigated compounds.
First, we combined
a N-benzyl substituent with a
4-IJ3-halogenphenyl) residue in compounds 4a–c. Such halo-
gen substituents may undergo halogen bonding to amino
acid residues in the potential P-gp binding region. The
3-chlorophenyl compound 4a was active as a P-gp modulator
at the lowest concentration of 1 μM with a FAR value of 2.22
(Table 1). Similar activities were found for the 3-bromophenyl
and the 3-trifluorophenyl compounds 4b and 4c. Verapamil
was less active at this concentration with a FAR value of
1.34, whereas tariquidar showed higher activities with a FAR
value of 5.25.
Increasing compound inhibitor concentrations led to
increased FAR values. At a concentration of 10 μM, the
3-bromophenyl compound 4b showed the highest activities,
again more than verapamil. Also at the highest concentration
used (20 μM) compound 4b remained the most active one,
suggesting the favourable influence of such a halogen bond
substituent on the P-gp inhibitory activity. Attempts to syn-
thesize 2- or 4-halogenphenyl substituted derivatives failed
due to the lowered reactivity of the starting aldehyde. An
alternative introduction of a 3-halogen substituent into the
benzyl residue of the 1,4-dihydropyridine structure also failed
due to the fact that the corresponding benzylamine com-
pounds were not commercially available. In the case when 2-
as well as 4-chloro substituted benzylamines were used, we
could not isolate the respective 1,4-dihydropyridine product,
presumably also because of the lowered amine reactivity.
We then introduced a methoxy function into the 4-phenyl
as well as into the N-benzyl substituent. Methoxy functions
are known as favourable hydrogen bond acceptor functions
in mdr modulators.25 Placed into the 4-position of the
N-benzyl residue of derivative 4e the activities were almost
similar to those of the 4-methoxyphenyl substituted compound
4d with the activity at the highest concentration being mainly
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2015

View Article Online
MedChemComm
Concise Article
Scheme 1 Formation of target structures 4a–p.
increased with a FAR value of 9.93. Next, we investigated the
combined methoxy substitutions of both the 4-phenyl and the
N-benzyl residue. A combination of the two methoxy functions
in the 4-phenyl and the N-benzyl substituent in both 4-posi-
tions of compound 4f led to similar P-gp modulating activities
at the given concentrations. The movement of the 4-methoxy
function in the N-benzyl residue to the 3-position led to almost
unchanged activity data of compound 4g. On the contrary,
when the 4-methoxy function of the 4-phenyl residue was
moved into the 3-position, the 4-methoxybenzyl residue
remained unchanged and the activity of compound 4h
decreased. A combination of both a 3-methoxyphenyl and a
3-methoxybenzyl substitution in derivative 4i led to a further
decrease in activity. So it may be concluded that the overall
symmetric 4-methoxy disubstitution of both residues is most
favourable among the varied aromatic monomethoxy substitu-
tions. The 4-methoxyphenyl substitution was most favourable
in combination with a second methoxy function.
This journal is © The Royal Society of Chemistry 2015
Med. Chem. Commun.

View Article Online
Concise Article
MedChemComm
Table 1 Concentration dependent P-gp modulating properties as calcu-
lated FAR values of our target compounds 4a–r
in the range of those with the 3-methoxy disubstitution
of derivative 4i. Obviously, the number of methoxy functions
to undergo a potential hydrogen bonding is limited. Two
methoxy functions are favourable, three methoxy functions
give the best P-gp modulating results, and finally four
methoxy functions lower the activity data.
We then replaced the 3-methoxy function in the 3,4-
dimethoxyphenyl residue with a 3-benzyloxy function in
compound 4o while the N-benzyl residue was unsubstituted.
The P-gp inhibiting activities were higher than those of the
3,4-dimethoxyphenyl and methoxybenzyl substituted com-
pounds 4l and 4m at almost all concentrations. So the more
lipophilic 3-benzyloxy substituent is more favourable than
the two methoxy functions. A further increase in activity was
observed in derivative 4p with the benzyloxy substituent
being located in the 4-position of the 4-phenyl residue and
the methoxy function bound in the 3-phenyl position. So the
resulting compound had a threefold higher FAR value than
verapamil at a concentration of 10 μM and was a better
modulator than tariquidar at this concentration.
FAR valuea
Compound
1 μM
2 μM
10 μM
20 μM
4a
4b
4c
4d
4e
4f
4g
4h
4i
4j
4k
4l
4m
4n
4o
4p
2.22 0.47
2.08 0.61
2.14 0.30
2.37 0.47
2.06 0.26
2.96 0.59
2.37 0.42
2.13 0.22
2.14 0.16
3.24 0.31
1.98 0.23
3.48 0.58
3.77 0.50
2.18 0.42
3.97 0.40
6.24 0.62
2.42 0.51
2.35 0.83
1.34 0.37
5.25 0.58
2.20 0.34
2.36 0.41
2.37 0.41
2.85 0.42
2.55 0.32
3.21 0.91
3.13 0.52
2.83 0.84
2.47 0.56
2.91 0.85
1.93 0.38
3.66 0.48
3.65 0.40
2.25 0.19
4.59 0.93
6.74 1.48
2.75 0.37
2.66 0.50
n.d.b
4.05 0.86
5.84 0.81
5.10 1.08
6.23 0.93
5.80 1.03
6.54 0.81
6.50 0.41
5.10 0.69
4.76 0.51
6.93 0.96
4.71 0.61
10.5 2.39
12.6 1.45
4.56 0.97
13.1 1.10
15.6 1.56
6.53 1.02
8.19 3.39
5.35 0.79
7.24 0.65
8.89 3.61
10.3 3.84
8.92 2.90
6.55 0.73
9.93 3.54
7.48 2.08
7.89 2.08
7.47 1.93
7.21 2.19
9.11 3.13
6.42 0.98
14.1 2.07
14.1 2.49
7.24 0.35
11.4 2.06
11.3 3.19
11.6 2.86
13.5 3.98
n.d.b
4q
4r
Verapamil
Tariquidar
n.d.b
n.d.b
A benzyloxy function however increases the compound's
lipophilicity besides having the ability to serve as a potential
hydrogen bond acceptor function. Obviously, the increased
compound lipophilicity is more important than an increasing
number of potential hydrogen bond acceptor functions. We
could not alternatively introduce the benzyloxy substituent into
the benzyl residue of the respective 1,4-dihydropyridine due
to the fact that the corresponding benzylamine compounds
were not available. A combination of a benzyloxyphenyl and a
methoxy substitution in the benzyl residue will lead to simi-
lar promising activities when compared to its concentration
on the phenyl residue. However, an increasing number of
hydrogen bond acceptor functions have been discussed to
have a limiting favourable effect on activity.
a
b
Mean of three determinations. Not determined.
Next, we tested whether an additional third methoxy
function could increase the P-gp-modulating activity and
started with a 4-methoxyphenyl group in derivative 4j com-
bined with a 3,4-dimethoxybenzyl residue. Compared to the
disubstituted 4-methoxy derivative 4f we found an increase in
activity at almost all the tested concentrations. When the
4-methoxy function in the 4-phenyl residue was moved to the
3-position the activities of the resulting compound 4k were
found mainly lowered at all concentrations by about 30%.
The 3-methoxyphenyl function also led to lowered activity
data in the series of the phenyl and benzyl dimethoxy-
substituted compound series 4f–i.
We then combined a 3,4-dimethoxyphenyl substitution
in compound 4l with a 4-methoxybenzyl substitution. We
found an increase in activity of up to 50% as compared to
the 3,4-dimethoxybenzyl and 4-methoxyphenyl derivative 4j,
even at higher concentrations. When the most favourable
3,4-dimethoxy function of the 4-phenyl residue is combined
Finally, we prepared a 4-naphthyl and a N-naphthylmethyl
derivative, each combined with an unsubstituted N-benzyl
and 4-phenyl residue, respectively (Scheme 2).
The FAR value of the resulting 4-naphthyl compound 4q at
the highest concentration was even higher than those of the
benzyloxy substituted compounds 4o and 4p. At lower con-
centrations, the activity lay in the ranges of the di- and
trimethoxy substituted derivatives. An increase in activity was
observed for the N-naphthylmethyl derivative 4r with activity
data at higher concentrations in the range of the most active
trimethoxy-substituted derivatives.
So besides the importance of a potential hydrogen bond-
ing via the introduced methoxy functions which are
favourably located at the 4-phenyl residue, lipophilic substitu-
ents significantly increase the P-gp modulating properties of
the respective compounds.25
Toxicity problems occurred with most mdr modulators
like tariquidar, in clinical states at the latest.26,27 While early
mdr modulators were toxic due to their original pharmaco-
logical properties as immunosuppressive or antihypertensive
agents, even those of the novel generations suffered from
toxicity problems which were attributed to the fact that they
with
a 3-methoxybenzyl function in derivative 4m we
observed some increase in activity as compared to the
4-methoxybenzyl compound 4l. The increased FAR values
at both the 1 μM and the 10 μM concentrations were more
than the double of the FAR value of verapamil and almost
the double of the tariquidar activity determined at this
concentration.
So it can be stated that the trimethoxy substitution is
more favourable than the disubstitution with the 3,4-
dimethoxyphenyl function, being the most favourable substit-
uent combination with the highest activities so far.
Surprisingly, a tetramethoxy substitution in compound 4n
was less favourable than the trisubstitution. The activities at
all the concentrations were lowered and we found activities
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2015

View Article Online
MedChemComm
Concise Article
Scheme 2 Formation of target structures 4q and r.
were substrates of the efflux pumps themselves so that
higher concentrations were necessary to achieve the desired
cellular effects.26,28,29 Those concentrations were found partly
toxic for normal cells.
We investigated the cellular toxicity of four of our target
compounds which showed the strongest activities as P-gp
modulators as far as evaluated. We determined the toxic
effects in both the non P-gp expressing parental cell line and
the P-gp expressing cell line to have a direct comparison
of the toxic effects in the target cells and the respective
normal cells.
Table 2 IC50 values of reduced cell viability in the non-P-gp expressing
parental cell line (P-cell line) and the P-gp expressing subline (MDR cell
line) for selected target compounds 4l, 4m, 4p and 4r and calculated FAR
values for fluorescent target compound uptake for relevant concentra-
tions of derivatives 4l and 4m
IC50 valuea (μM)
MDR
FAR valuea
P-cell
MDR cell
Compound P-cell line
cell line
line 50 μM line 100 μM
4l
263 1.22
133 1.35
154 1.32
122 1.22
1.31 0.51 1.14 0.50
0.95 0.47 1.02 0.34
4m
4p
4r
197.2 3.59 >160b
n.d.c
n.d.c
n.d.c
84.6 1.18
71.1 1.33 n.d.c
We determined the cell viability under increasing modula-
tor concentrations up to 160 μM using the MTT assay. In this
assay, the mitochondrial toxicity of a compound is deter-
mined. That compound's toxicity reduces the formation of
the fluorescent formazan dye from the MTT reagent under a
reduced catalysis of the mitochondrial dehydrogenases as a
consequence of the cellular compound's toxicity. We calcu-
lated the IC50 values of the reduced cell viability as shown in
Table 2.
The determined IC50 values of reduced cell viabilities
determined in the non P-gp expressing cell line varied from
84.6 μM for compound 4r to 197.2 μM for compound 4p.
With these values, the compounds are completely non-toxic
in the tested concentration ranges starting with 1 μM as the
effective concentration for all compounds. Even at an inhibi-
tor concentration of 10 μM most compounds caused cell
viability rates of more than 90% (data not shown).
a
b
c
Mean of three determinations.
determined.
Not determinable.
Not
Interestingly, we found a slightly increased toxicity in the
P-gp expressing cell line with IC50 values ranging from 71.1 μM
for derivative 4r to 154 μM for compound 4l. At a concentra-
tion of 10 μM the cell viability rates ranged from 78.4% (4r)
to 95.6% (4p).
If a compound was a substrate of the efflux pump, the
toxicity of the compound in the P-gp expressing cell line
would be lower at the given concentration because parts of
the compound would have been transported out of the cells
by the efflux pump activity which is not found in the non-
expressing cell line.
This journal is © The Royal Society of Chemistry 2015
Med. Chem. Commun.

View Article Online
Concise Article
MedChemComm
This would mean an increase in the cell viability rate at
the given concentration, and thus an increase in the resulting
IC50 value.
However, we observed partly increased toxicity data as
discussed with the resulting lowered IC50 values in the
P-gp-expressing cell line, so that we can exclude such P-gp
substrate properties.
This concluded result should be further confirmed in an
additional study which we carried out with our most active
compounds 4m and 4l.
We incubated both the P-gp non-expressing and the P-gp
expressing cell line with higher concentrations of both com-
pounds. If one of the compounds showed P-gp substrate
properties its uptake into the P-gp expressing cell line would
be lower due to a partial cellular efflux. We directly deter-
mined the uptake of our compounds by determination of their
own fluorescence at 430 nm using a flow cytometry technique
and a laser with excitation at 366 nm, much lower than that of
the laser used for the determination of the P-gp modulating
effects. The resulting fluorescence in cells of both cell lines
was determined and the ratio was calculated. The calculated
FAR values are given in Table 2. Compound 4m showed FAR
values of 0.95 at 50 μM and of 1.02 at 100 μM. For compound
4l FAR values of 1.31 and 1.14 were found at the respective
concentrations. FAR values > 1 mean that the compound has
completely no P-gp substrate properties. The results of the
direct compound measurements correlate with the results
from the toxicity studies as discussed. For comparison we
determined the FAR value for Rhodamine 123 as a P-gp sub-
strate. At a Rhodamine concentration of 5 μM, we found a
FAR value of 0.05 which clearly documents the P-gp substrate
properties of the compound.
cell line and 10.21 μM for the P-gp expressing subline. The
lowered daunorubicin toxicity in the subline resulted from
the fact that it was transported out of the cells by the P-gp
efflux pump activity. We then pre-incubated the cells of the
non-P-gp expressing and the P-gp expressing subline with the
best of our modulators 4l, 4m, 4p and 4r at a concentration
of 10 μM each and determined the daunorubicin toxicity
again. The calculated IC50 values are shown in Table 3. The
IC50 values in the non-P-gp expressing cell line were found
almost unchanged. For compounds 4l and 4r we found IC50
values in the P-gp expressing cell line almost identically equal
to that which we found in the parental cell line without P-gp.
The results meant that we had a complete reversal of the
mdr in the P-gp expressing subline and a restoration of its
daunorubicin sensitivity. Compounds 4m and 4p resulted in
little lowered IC50 values which meant an increased sensitiv-
ity of the treated cell line towards daunorubicin. So we
lowered the inhibitor concentration of these compounds to
5 μM and determined the daunorubicin toxicity again. While
compound 4m resulted in an IC50 value of 1.86 μM, the most
active derivative 4p led to an IC50 value of 0.84 μM. So also
this lowered concentration was able to reverse the mdr and
restore the daunorubicin sensitivity of the P-gp expressing
subline.
Our novel and highly active P-gp modulators are no P-gp
substrates like Rhodamine, as shown, and are able to reverse
the toxicity of a cytostatic drug with clinical relevance. These
promising results encourage further preclinical studies of
selected compounds to combat cancer in P-gp overexpressing
cells by the inhibition of the causative efflux pump P-gp.
Acknowledgements
Finally, we investigated the ability of our P-gp modulators
with the highest activities to reverse the mdr of the clinically
relevant cytostatic drug daunorubicin in our cell line model.
Daunorubicin is a known substrate of P-gp. So we deter-
mined the toxicity of daunorubicin in both the mouse T lym-
phoma cell line without P-gp and the P-gp overexpressing
subline (Table 3). The IC50 values of daunorubicin for the cel-
lular viability were calculated to be 0.93 μM for the parental
The authors acknowledge financial support of their work by
the DFG (German Research Foundation) to Sören Krawczyk
and Andreas Hilgeroth within the project HI 687/6-3.
References
1 M. Dobbelstein and U. Moll, Nat. Rev. Drug Discovery, 2014,
13, 179.
2 K. Strebhardt and A. Ullrich, Nat. Rev. Cancer, 2008, 8, 473.
3 J. Luo, N. L. Solimini and S. J. Eliedge, Cell, 2009, 136, 825.
4 J. Zhang, P. L. Yang and N. S. Gray, Nat. Rev. Cancer, 2009,
9, 28.
Table 3 MDR reversal as restored daunorubicin toxicity with determined
IC50 values of cellular viability for daunorubicin under P-gp modulator
application in the P-gp non-expressing and the P-gp overexpressing cell
lines for compounds 4l, 4m, 4p and 4r
IC50 valuea [μM]
5 C. J. Tsai and R. Nussinov, Semin. Cancer Biol., 2013,
23, 235.
6 F. Perche and V. P. Torchillin, J. Drug Delivery, 2013,
DOI: 10.1155/2013/705265.
7 P. D. W. Eckford and F. J. Sharom, Chem. Rev., 2009, 109,
2989.
8 J. P. Gillet and M. M. Gottesman, Methods Mol. Biol., 2010,
596, 47.
P-gp overexpressing cell line
Parental
cell line
5 μM
10 μM
Compound
Without inhibitor
4l
4m
4p
4r
0.93 0.09
0.85 0.08b
0.95 0.06b
0.75 0.06b
0.80 0.04b
10.21 0.99
n.d.c
1.86 0.93
0.84 0.08
n.d.c
0.97 0.09
0.85 0.09
0.61 0.08
0.93 0.09
a
b
9 A. Hilgeroth, M. Hemmer and C. Coburger, Mini-Rev. Med.
Chem., 2012, 12, 1127.
Mean of three determinations.
Determined at
a
modulator
c
concentration of 10 μM. Not determined.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2015

View Article Online
MedChemComm
Concise Article
10 A. Breier, L. Gibalove, M. Seres, M. Barancik and Z. Sulova,
Anti-Cancer Agents Med. Chem., 2013, 13, 159.
11 M. M. Gottesman, Annu. Rev. Med., 2002, 53, 615.
12 M. M. Gottesman and V. Ling, FEBS Lett., 2006, 580, 998.
13 S. G. Aller, J. Yu, A. Ward, Y. Weng, S. Chittaboina, R. Zhou,
P. M. Harrell, Y. T. Trinh, Q. Zhang, I. L. Urbatsch and
G. Chang, Science, 2009, 323, 1718.
14 M. M. Gottesman, T. Fojo and S. E. Bates, Nat. Rev. Cancer,
2002, 2, 48.
15 M. F. Rosenberg, R. Callaghan, S. Modok, C. F. Higgins and
R. C. Ford, J. Biol. Chem., 2005, 280, 2857.
16 L. M. S. Chan, S. Lowes and B. Hirst, Eur. J. Pharm. Sci.,
2004, 21, 25.
17 R. B. Wang, C. L. Kuo and E. J. Lien, J. Clin. Pharm. Ther.,
2003, 28, 203.
18 C. Cheng, P. M. Bahadduri, J. E. Polli, P. W. Swann and
S. Ekins, Drug Metab. Dispos., 2006, 34, 1976.
19 T. Tsuruo, H. Iida, M. Nojiri, S. Tsukagoshi and Y. Sakurai,
Cancer Res., 1983, 43, 2905.
20 P. A. Philip, S. Joel, S. C. Monkman, E. Dolega-Ossowski,
K. Tonkin, J. Carmichael, J. R. Idle and A. L. Harris, Br. J.
Cancer, 1992, 65, 267.
23 In earlier studies we found similar activities for verapamil
and cyclosporine A as P-gp modulators.24
.
24 C. Coburger, H. Lage, J. Molnár and A. Hilgeroth, Pharm.
Res., 2009, 26, 182.
25 We also determined the nanomolar activities for our most
active P-gp modulating compounds 4l, 4m, 4p and 4r at a
concentration of 100 nM with resulting FAR values of 1.23
(4l), 1.37 (4m), 2.59 (4p) and finally 1.19 (4r).
26 R. Krishna and L. D. Mayer, Eur. J. Pharm. Sci., 2000, 11,
265.
27 R. J. Kelly, D. Draper and C. C. Chen, Clin. Cancer Res., 2011,
17, 569.
28 C. Baumert and A. Hilgeroth, Anti-Cancer Agents Med. Chem.,
2009, 9, 415.
29 Drug–drug interactions have long been discussed to
attribute to the toxic effects of P-gp inhibitors which may
influence the excretion processes of co-administered
drugs so that the resulting higher intracellular levels of
those drugs become toxic. A recent report investigated
such possible in vivo effects of digoxin under P-gp inhibitor
applications. Digoxin becomes toxic in slightly increased
intracellular doses. However, no significant effects on the
digoxin toxicity under the P-gp inhibitor applications were
observed.30
21 M. Kawase, A. Shah, H. Gaveriya, N. Motohashi, H. Sakagami,
A. Varga and J. Molnár, Bioorg. Med. Chem., 2002, 10, 1051.
22 S. Saponara, M. Kawase, A. Shah, N. Motohashi, J. Molnár,
K. Ugocsai, G. Sgaragli and F. Fusi, Br. J. Pharmacol., 2004,
141, 415.
30 K. S. Fenner, M. D. Troutman, S. Kempshall, J. A. Cook,
J. A. Ware, D. A. Smith and C. A. Lee, Clin. Pharmacol.
Ther., 2009, 85, 173.
This journal is © The Royal Society of Chemistry 2015
Med. Chem. Commun.