Discovery of substituted 1,4-dihydroquinolines as novel promising class of P-glycoprotein inhibitors: First structure-activity relationships and bioanalytical studies

Article abstract of DOI:10.1016/j.bmcl.2015.05.018

Multidrug resistance (mdr) is the most important problem in the therapeutical treatment of cancer. One central problem in the resistance proceeding is the expression of transmembrane efflux pumps which transport drugs out of the cells. We developed novel substituted 1,4-dihydroquinolines as inhibitors of the transmembrane efflux pump P-glycoprotein. Structure-activity relationships are discussed for this first series. Promising active inhibitors have been identified and first bioanalytical studies have been carried out to address questions of cellular toxicity, P-gp substrate as well as mdr reversal properties.

Full text of DOI:10.1016/j.bmcl.2015.05.018

Bioorganic & Medicinal Chemistry Letters xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of substituted 1,4-dihydroquinolines as novel promising
class of P-glycoprotein inhibitors: First structure–activity
relationships and bioanalytical studies
⇑
Marc Hemmer, Sören Krawczyk, Ina Simon, Andreas Hilgeroth
Institute of Pharmacy, Martin Luther University Halle-Wittenberg, Wolfgang-Langenbeck-Str. 4, 06120 Halle, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Multidrug resistance (mdr) is the most important problem in the therapeutical treatment of cancer. One
central problem in the resistance proceeding is the expression of transmembrane efflux pumps which
transport drugs out of the cells. We developed novel substituted 1,4-dihydroquinolines as inhibitors of
the transmembrane efflux pump P-glycoprotein. Structure–activity relationships are discussed for this
first series. Promising active inhibitors have been identified and first bioanalytical studies have been car-
ried out to address questions of cellular toxicity, P-gp substrate as well as mdr reversal properties.
Ó 2015 Elsevier Ltd. All rights reserved.
Received 16 March 2015
Revised 7 May 2015
Accepted 11 May 2015
Available online xxxx
Keywords:
1
,4-Dihydroquinolines
P-gp inhibitor
Structure–activity relationships
Substrate properties
Mdr reversal
With the understanding of cell regulating processes in the last
decades the number of identified dysregulating agents in cancer
cells increased, for example, the role of protein kinases which are
from the bisbenzylisoquinoline alkaloid tetrandrine. They showed
cardiovascular effects and some mdr reversal activities that were
discussed to result from the affection of the membrane fluidity.12
Two quinolones isolated from a Chinese medical herb showed
some P-gp inhibiting effects, but were not further investigated.13
We reported some P-gp inhibiting effects for quinolone-related
1
partly involved in several cellular pathways. Novel drugs have been
developed that address such dysregulated agents and preferably act
as protein kinase inhibitors.2 Occurring resistances against protein
kinases led to the establishment of multitargeting protein kinase
inhibitors which target several protein kinases in the cancer pro-
,3
1
4
pyridine-2-ones. The structures of both the quinolones and the
pyridine-2-ones are shown in Figure 1.
4
gression. However, successfully acting small-molecule inhibitors
Alkaloids have a complex structure and compounds derived
from natural products afford complex isolation and purification
processes to provide a suitable drug for therapeutical use.
Our novel 1,4-dihydroquinolines were available in a simple
two-step reaction procedure. The first reaction product could be
used without purification. We found partly promising P-gp inhibi-
tory activities for our compounds and demonstrate a first bioana-
lytical compound profile of nontoxicity, no P-gp substrate
properties and mdr reversing effects.
Our 1,4-dihydroquinoline synthesis started from ethyl quino-
line-carboxylate 1 which was melted in a flask on an oil bath at
a temperature of 80–100 °C. Then the respective benzyl bromide
2a–c was added dropwise under stirring at the maintained temper-
ature (Scheme 1).
like imatinib or monoclonal antibodies were found to be substrates
of transmembrane efflux pumps like P-glycoprotein.5,6 P-glycopro-
tein is found in cancer cells and its cellular expression is mostly
induced by the used anticancer drugs.7 P-gp transports drugs of
various structures, so that a multidrug resistance (mdr) of the
respective cancer cell results. The occurrence of P-gp makes such a
,8
9
cancer treatment difficult. The inhibition of the efflux pump activ-
ity has been studied by the use of potential inhibitors which suffered
from cellular toxicity. That toxicity was partly caused by high doses
1
0,11
of the inhibitors that were substrates of the efflux pumps.
Such
high doses were required to reach cancer cell-relevant effects, but
were toxic for normal cells.
We discovered 1,4-dihydroquinolines as a novel P-gp inhibiting
compound class. Related tetrahydroisoquinolines were derived
The noncommercially available 4-methoxy benzyl bromide 2c
was prepared from 4-methoxy benzyl alcohol via treatment with
bromic acid in benzole at room temperature. The resulting N-ben-
zyl quinolinium salts 3a–c were washed with portions of diethyl
ether and dried over phosphorus pentoxide without further
⇑
Corresponding author. Tel.: +49 345 55 25168; fax: +49 345 55 27207.
E-mail address: andreas.hilgeroth@pharmazie.uni-halle.de (A. Hilgeroth).
http://dx.doi.org/10.1016/j.bmcl.2015.05.018
960-894X/Ó 2015 Elsevier Ltd. All rights reserved.
0
Please cite this article in press as: Hemmer, M.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.05.018

2
M. Hemmer et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
R2
P-gp expressing cell line each with inhibitor, after both fluores-
cence amount values have been related to those of the untreated
control cell lines without inhibitor.
The given FAR values are listed in Table 1. Those calculated FAR
values >1.1 prove an inhibition of P-gp.
O
R1
n
N
N
O
Me
Bn
The N-benzyl and 4-phenyl compound 4a showed activity as P-
n = 7,11
1
gp inhibitor at the higher concentration of 10 lM with a FAR value
R
R
= COOMe, CN
2
of 2.86. The introduction of a methoxy function into the 4-position
of the phenyl residue in derivative 4b mainly increased the activity
at both concentrations. Methoxy groups in mdr modulators are
known to function as hydrogen bond acceptors binding to a poten-
= Ph, Bn, 2-tolyl, 2-MeOPh,
4
-tolyl, 4-MeOPh
quinolones
pyridine-2-ones
1
6
tial P-gp binding site. Compound 4b showed a better activity at
Figure 1. Structures and substitution patterns of cited quinolones and pyridine-2-
the higher concentration than the used standard verapamil which
ones.
1
7
is known to be one of the best in vitro inhibitors. When the meth-
oxy function is moved from the 4- to the 3-position of the phenyl
residue we found further increases in the activity of the respective
compound 4c. We then introduced two methoxy functions into the
purifications. The salts were dissolved in THF for the introduction
of the respective 4-phenyl substituent and catalytic amounts of
copper(I) iodide were added with lithium chloride to improve
the catalyst solubility. The stirred mixture was treated with the
respective aryl magnesium chloride as used Grignard reagent at
4
-phenyl residue and expected increases in activity provided that
such methoxy functions strengthen the bonding of the inhibitor
to a potential P-gp binding region. First, we combined a 3- and a
4-methoxy function in compound 4d. This dimethoxy combination
resulted in an increased activity of more than 100 per cent if com-
pared to the monomethoxy substituted compounds 4b and 4c,
respectively. Compound 4d showed a more than 2.5-fold higher
activity than the standard verapamil. A combination of a 3- and a
15
room temperature to give the target compounds 4a–i which
were yielded after the work-up procedure from diethyl ether.
The P-gp inhibiting properties of the compounds were evalu-
ated in a cell line model of a non P-gp expressing mouse T-lym-
phoma cell line and a respective subline. That subline expresses
human P-gp after a retroviral gene transfection and a cell culturing
under colchicine supplementation of the medium. That procedure
ensured an exclusive survival of the P-gp expressing cells.
We determined the uptake of the fluorescent P-gp substrate
rhodamine 123 in both cell lines using flow cytometry technique
with and without a P-gp inhibitor addition. The inhibition of
P-gp resulted in a higher uptake of the fluorescent substrate in
the P-gp expressing cell line.
5
-methoxy substitution in derivative 4e was not as favourable as
the 3- and 4-dimethoxy substitution of compound 4d. However,
if compared to the only 3-methoxy substitution in derivative 4c
we found a higher activity which confirmed that the two potential
hydrogen bond acceptor functions are much more favourable than
only one of these functional groups.
We then investigated whether the intramolecular positioning of
these two methoxy functions would be of importance for the bio-
logical activity. So we combined the favourable 3-methoxyphenyl
function of derivative 4c with a 3-methoxybenzyl function in com-
pound 4f. If compared to the 3- and 5-dimethoxy substitution of
derivative 4e the activities were found little lower at both concen-
trations. The result indicated that a disubstitution of the 4-phenyl
residue in compound 4e with both methoxy functions is more
favourable than the allocation on both aromatic residues in the
respective meta positions of derivative 4f.
The combination of the 3-methoxyphenyl substitution of com-
pound 4c with a 4-methoxybenzyl substitution in derivative 4g
also improved the activity if compared to the monomethoxy phe-
nyl substitution of derivative 4c. If compared to the 3- and 4-
dimethoxy substituted compound 4d the activity was found lower.
Also in this case the allocation of two methoxy functions on both
aromatic residues at the respective positioning is less favourable
Finally, FAR values for the inhibition of P-gp were calculated by
a division of the uptaken fluorescence amounts in the P-gp
expressing cell line with the fluorescence amounts of the non
O
R
OEt
Br
+
N
1
2
a-c
O
a
OEt
Br-
+
N
3
a, R = H
b, R = 3-OMe
c, R = 4-OMe
R
Table 1
Concentration dependent P-gp inhibition of target compounds 4a–i
R1
O
Compound
R1
R2
FAR valuea
1
lM
10 lM
b
OEt
4a
H
H
H
H
H
0.79 ± 0.11
1.27 ± 0.46
1.55 ± 0.54
3.42 ± 0.79
2.96 ± 0.52
2.26 ± 0.43
2.03 ± 0.41
2.23 ± 0.40
2.55 ± 0.26
1.46 ± 0.15
2.86 ± 0.33
5.52 ± 1.22
7.43 ± 1.14
12.77 ± 2.03
10.50 ± 1.78
9.68 ± 1.12
10.34 ± 2.01
9.95 ± 1.19
12.38 ± 0.87
4.98 ± 0.82
4
4
4
4
b
c
d
e
4-OMe
3-OMe
3-, 4-OMe
3-, 5-OMe
3-OMe
3-OMe
4-OMe
4-OME
N
H
R2
4f
3-OMe
4-OMe
3-OMe
4-OMe
4
4
4
g
h
i
4
a-i
Scheme 1. Reagents and conditions for the preparation of compounds: (a) 80–
00 °C, 20 min, 1 h, rt, 20–85%, (b) aryl magnesium chloride, Cu(I) I, LiCl, THF, rt,
Verapamil
1
3
a
0 min, 18–65%.
Mean of three determinations.
Please cite this article in press as: Hemmer, M.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.05.018

M. Hemmer et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
Table 3
3
than the concentration of both functions at the 4-phenyl residue of
IC50 values of the cellular toxicity and fluorescence ratio values of cellular uptake
the dimethoxy phenyl substituted compound 4d.
rates of 4c and 4d
The combination of the 4-methoxyphenyl substitution of com-
pound 4b with the 3-methoxybenzyl function in derivative 4h
led to almost similar activities than the combination of the
Compound
IC50 value (
lM)
Fluorescence ratio valuea
P-cell line mdr cell line
3
-methoxyphenyl substitution with the 3-methoxybenzyl substi-
24 h
48 h
24 h
48 h
50
lM
100 lM
tution of compound 4f. If combined with the 4-methoxybenzyl
function in compound 4i we found increases in activity reaching
the activity of the 3-, 4-dimethoxy substituted compound 4d.
Next, we investigated the potential of our compounds to reverse
the P-gp mediated mdr using our cell line model. We selected
daunorubicin as known P-gp substrate and determined the toxicity
of the cytostatic agent in both, the P-gp nonexpressing parental cell
line and the P-gp expressing subline. The cellular toxicity was
determined in the MTT assay for a reduced cell viability resulting
from a lowered formazane formation via the mitochondrial dehy-
drogenases. The resulting IC50 values are shown in Table 2.
The IC50 value for daunorubicin in the parental cell line was
4
4
c
d
>160
>160
>160
>160
>160
>160
>160
>160
1.03 ± 0.24
1.06 ± 0.24
0.99 ± 0.18
1.07 ± 0.29
a
Mean of three determinations.
could exclude any effects of the compounds on the observed
daunorubicin toxicity.
The observed unchanged toxicity of our compounds in both cell
lines indicated that they were no substrates of the efflux pump. In
case of such substrate properties the toxicity in the P-gp expressing
subline would have been lower due to the fact that the compounds
were partly transported out of the cells.
As our 1,4-dihydroquinolines themselves are fluorescent com-
pounds we directly determined their uptake in both cell lines to
determine such substrate properties. We calculated fluorescence
ratios by relating the determined compound fluorescence uptakes
in both cell lines using the flow cytometry technique.18 A relation
of the compound fluorescence’s in cells of both cell lines with a value
of about 1 meant that the compounds are no substrates of P-gp so
that the compound uptake in both cell lines is similar and not
affectedby the activityof P-g. We usedconcentrationsof our fluores-
determined with 0.75
lM. In the P-gp expressing subline the IC50
value was found mainly increased with 8.26
lM that is a more
than tenfold increased value. The higher value results from the fact
that the P-gp substrate daunorubicin is transported out of the cell
by the P-gp efflux pump activity.
We selected best compounds of our series of mono- and
dimethoxy substituted derivatives which has been the 3-methoxy-
phenyl substituted derivative 4c and the 3,4-dimethoxy substi-
tuted compound 4d.
We first used the effective 10 lM concentration of both evalu-
cent compounds of 50 lM and of 100 lM. All resulting fluorescence
ated P-gp inhibitors and determined the daunorubicin toxicity
under the given inhibitor concentration. The monomethoxy phenyl
substituted derivative 4c resulted in a mainly reduced IC50 value of
the daunorubicin toxicity in the P-gp expressing subline with a
ratios are displayed in Table 3 and were about a value of 1 which
meant that our compounds were really no substrates of the efflux
pump because their uptake in both, the nonexpressing parental cell
line and the P-gp expressing subline, was almost identical.
In conclusion, we discovered a novel class of promising P-gp
inhibitors. First structure–activity relationships prove that meth-
oxy groups as potential hydrogen bond acceptor functions are
favourably concentrated at the 4-phenyl residue. The compounds
are able to completely reverse the P-gp mediated resistance at
the given concentrations. Moreover, they are nontoxic and do not
own P-gp substrate properties. Thus, they are a really perspective
class of novel mdr modulators for further preclinical studies.
value of 0.86 lM which is nearly the same value that was found
for the daunorubicin toxicity in the non P-gp expressing parental
cell line. The reduction meant a complete reversal of the P-gp
mediated resistance and a restoration of the daunorubicin toxicity.
The dimethoxy substituted derivative 4d led to an IC50 value of
0
.39 in the P-gp expressing subline which meant a slightly
increased toxicity in this subline if compared to the non P-gp
expressing parental cell line. Beside a complete restoration of the
P-gp mediated resistance we found a slightly increased sensitivity
of the P-gp expressing subline to the toxic daunorubicin agent.
Acknowledgments
Next, we investigated the effect of a 5 lM concentration applica-
tion of compound 4d to reverse the P-gp mediated resistance in
the P-gp expressing subline. The resulting IC50 value was
The authors acknowledge the financial support of their work by
the DFG – Germany (German Research Foundation) to Sören
Krawczyk and Andreas Hilgeroth within the project HI 687/6-3
and from the European Union (EU) within the project 1184256 to
Marc Hemmer.
4
.76 lM, so that this concentration led to a nearly 50% restoration
of the daunorubicin toxicity.
We then determined the cellular toxicity of our investigated
inhibitors 4c and 4d in both cell lines to exclude possible toxic
effects of the compounds themselves which may have contributed
to the daunorubicin toxicity (Table 3).
References and notes
We incubated varying concentrations of the inhibitors in the
cell lines and found both compounds to be completely nontoxic
with IC50 values >160 lM. Also after 48 h of incubation the com-
1
.
Liu, K.-Q.; Liu, Z.-P.; Chen, L.; Zhao, X.-M. Bioinformatics 2012, 13, 126.
2. Zhang, J.; Yang, P. L.; Gray, N. S. Cancer 2009, 9, 28.
3
4
5
.
.
.
Tsai, C. J.; Nussinov, R. Sem. Cancer Biol. 2012, 23, 235.
Krug, M.; Hilgeroth, A. Mini-Rev. Med. Chem. 2008, 8, 1312.
Mahon, F.-X.; Belloc, F.; Lagarde, V.; Chollet, C.; Moreau-Gaudry, F.; Reiffers, J.;
Goldman, J. M.; Melo, J. V. Blood 2003, 101, 2368.
pounds remained nontoxic with unchanged IC50 values. So we
6.
Walter, R. B.; Raden, B. W.; Hong, T. C.; Flowers, D. A.; Bernstein, I. D.;
Linenberger, M. L. Blood 2003, 102, 1466.
Table 2
7
8
9
.
.
.
Gottesman, M. M.; Fojo, T.; Bates, S. E. Cancer 2002, 2, 48.
Hilgeroth, A.; Hemmer, M.; Coburger, C. Mini-Rev. Med. Chem. 2012, 12, 1127.
Breier, A.; Gibalove, L.; Seres, M.; Barancik, M.; Sulova, Z. Anti-Cancer Agents
Med. Chem. 2013, 13, 159.
IC50 values of the cellular toxicity of daunorubicin without and with P-gp inhibitor
concentrations
Compound
IC50 value (lM)
1
1
0. Srivalli, K. M. R.; Lakshmi, P. K. Braz. J. Pharm. Sci. 2012, 48, 353.
1. Palmeira, A.; Sousa, E.; Vasconcelos, M. H.; Pinto, M. M. Curr. Med. Chem. 1946,
P-cell line
mdr cell line
2
012, 19.
2. Li, Y.; Zhang, H. B.; Huang, W. L.; Li, Y. M. Bioorg. Med. Chem. Letters 2008, 18,
652.
Without inhibitor
0.75 ± 0.61
8.26 ± 0.62
0.86 ± 0.24
0.39 ± 0.12
4.76 ± 0.92
1
1
4
4
4
c (10
d (10
l
l
M)
M)
M)
—
—
—
3
3. Adams, M.; Mahringer, A.; Kuhnert, O.; Fricker, G.; Efferth, T.; Bauer, R. Planta
Med. 2007, 73, 1554.
d (5
l
Please cite this article in press as: Hemmer, M.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.05.018

4
M. Hemmer et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
1
1
4. Krawczyk, S.; Otto, M.; Coburger, C.; Krug, M.; Seifert, M.; Tell, V.; Molnár, J.;
Hilgeroth, A. Bioorg. Med. Chem. 2011, 19, 6309.
5. Spectroscopical data of substituted N-benzyl 4-phenyl 1,4-dihydroquinoline
compound class representing compound 4c: Yield 0.045 g (18%); mp 100–
CH
(ABX
2
), 4.02 (ABX
3
, J = 10.9, 7.0 Hz, 2H, COCH
2
CH
3
), 3.63 (s, 3H, OCH
); m/z (EI) 399 (100, M ). Elemental Anal. Calcd (%) for
: C, 78.17; H, 6.31; N 3.51. Found: C, 78.34; H, 6.25; 3.36.
3
), 1.12
+
3
, J = 7.0 Hz, 3H, CH
3
C
26
H25NO
3
16. Krishna, R.; Mayer, L. D. Eur. J. Pharm. 2000, 11, 265.
1
1
02 °C; IR (ATR)
305 (CAN) cm
m
= 1665 (C@O), 1636 (NAC@C), 1596, 1464 (C@C), 1386 (CH
3
),
17. Mechetner, E. B.; Roninson, I. B. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 5824.
18. The measurement was carried out at a lower wavelength of 450 nm than that
used for the measurement of the rhodamine 123 uptake which was conducted
at a wavelength of 529 nm.
À1
1
;
H NMR (CDCl ) d = 7.60 (s, 1H, 2-H), 7.22–7.18 (m, 5H,
3
benzylic H), 7.10–7.00 (m, 2H, 6-, 7-H), 6.90 (dt, J = 7.8, 1.6 Hz, 1H, phenylic 5-
H), 6.78–6.76 (m, 3H, phenylic 2-, 4-, 6-H), 6.68 (dd, J = 8.2 Hz, 0.8 Hz, 1H, 5-H),
6
.58 (dd, J = 8.2, 0.8 Hz, 1H, 8-H), 5.10 (s, 1H, 4-H), 4.78 (AB, J = 16.4 Hz, 2H, N-
Please cite this article in press as: Hemmer, M.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.05.018