The interaction between resistance modifiers such as pyrido[3,2-g] quinoline, aza-oxafluorene and pregnane derivatives with DNA, plasmid DNA and tRNA

Article abstract of DOI:10.1016/j.ejmech.2004.10.011

Various resistance mechanisms such as complex formation with DNA, tRNA and MDR1 p-glycoprotein were modified in bacteria and cancer cells in presence of pregnane, pyridoquinoline, and aza-oxafluorene derivatives. Interaction between the compounds, plasmid

Full text of DOI:10.1016/j.ejmech.2004.10.011

European Journal of Medicinal Chemistry 40 (2005) 195–202
www.elsevier.com/locate/ejmech
Original article
The interaction between resistance modifiers
such as pyrido[3,2-g]quinoline, aza-oxafluorene and pregnane derivatives
with DNA, plasmid DNA and tRNA
Derek Sharples a,*, Gabriella Spengler b,c, Joseph Molnár b,c, Zsuzsanna Antal b,c
,
Annamária Molnár b,c, János T. Kiss b,c, József A. Szabó b,c, Andreas Hilgeroth d,
Sandrine Gallo e, Abdallah Mahamoud e, Jacques Barbe e
a School of Pharmacy and Pharmaceutical Sciences, University of Manchester, Manchester, UK
b Department of Medical Microbiology, Faculty of Medicine, University of Szeged, Szeged, Hungary
c Department of Chemistry, Faculty of Medicine, University of Szeged, Szeged, Hungary
d Faculty of Pharmacy, Martin-Luther University Halle–Wittenberg, Halle, Germany
e Fac. pharmacie, GERCTOP-UMR CNRS 6178, university Méditerranée, Marseille, France
Received 13 July 2004; received in revised form 15 October 2004; accepted 21 October 2004
Available online 04 January 2005
Abstract
Various resistance mechanisms such as complex formation with DNA, tRNA and MDR1 p-glycoprotein were modified in bacteria and cancer
cells in presence of pregnane, pyridoquinoline, and aza-oxafluorene derivatives. Interaction between the compounds, plasmid DNA and tRNA
was shown and compared to the interaction with calf thymus DNA. Complex formation with MDR1 p-glycoprotein and drug accumulation
increased in cancer cells. Both plasmid DNA and p-gp complex formation were related to the chemical structures of the resistance modifiers.
© 2005 Elsevier SAS. All rights reserved.
Keywords: Pregnanes; Pyridoquinolines; Aza-fluorenes; DNA interaction; Transfer RNA interaction; MDR1 p-glycoprotein
1. Introduction
In order to confirm the evidence of the binding of antiplas-
mid compounds to nucleic acids, the UV spectrum of the
ligands can be examined. When a ligand possessing a chro-
mophore binds to nucleic acids, its UV spectra will be altered
as a consequence of changing in its molecular environment.
The wavelength of absorption undergoes a shift to longer wave-
lengths (bathochromic shift) with corresponding decrease in
absorbance (hypochromic shift), which is a good indication of
It is well known that the denaturation of nucleic acids
caused by elevated temperature will increase their UV absor-
bance due to the melting of double stranded regions (hyper-
chromic effect). By measuring the UV absorbance of nucleic
acids at increasing temperature, the melting of nucleic acids
can be monitored and the phase-transition temperature (ther-
mal denaturation temperature, Tm) can be determined. Both
positive and negative changes in Tm values can occur: an
increase in Tm indicates stabilisation of the DNA helix and a
decrease in Tm can mean the degradation of the nucleic acid
by different chemical agents [1].
an interaction between the ligand and the nucleic acid [2]
.
2. Chemistry
2.1. General procedures for the preparation
of the pregnane derivatives (Fig. 1)
2.1.1. Pregnane thioethers (P-S1 series) were prepared
by a base catalysed Michael addition reaction
from 3b-acetoxy-5,16-pregnadiene-20-one (PDA)
Abbreviations: PTC, phase transfer catalysis; DMSO, dimethyl sul-
phoxide; Tm, thermal denaturation temperature; Tris, tris(hydroxymethyl)-
aminomethane.
A solution of either H2S or the appropriate mercaptan
(H-SR; R = methyl, cyclohexyl, benzyl, phenyl) in dry ben-
* Corresponding author. Tel.: +44 161 275 2335; fax: +44 161 275 2396.
E-mail address: derek.sharples@manchester.ac.uk (D. Sharples).
0223-5234/$ - see front matter © 2005 Elsevier SAS. All rights reserved.
doi:10.1016/j.ejmech.2004.10.011

196
D. Sharples et al. / European Journal of Medicinal Chemistry 40 (2005) 195–202
Fig. 2. Structures of monosubstituted pyrido[3,2-g]quinolines (4a–d).
cial 2,6-diaminotoluene by a three-step synthesis. The first
step, a Skraup reaction, gave 7-amino-8-methylquinoline
[11,12].
A Michael 1,4-addition of ethyl acetoacetate to the
7-amino-8-methylquinoline affords the intermediate imine
(8-methyl-7-[(2′-carboethoxy-1′-methylvinyl)amino)]-quino-
line).
The pyridoquinolinones were then obtained by a Conrad–
Limpach thermocyclisation of this intermediate imine [13,14].
2.3. Diazatetraasteranes (aza-oxofluorenes Az1, Az2, Az3)
(Fig. 3)
Fig. 1. Structures of pregnane derivatives investigated.
The synthesis of compounds Az1, Az2, and Az3; 3,9-
dibenzyl-1,5,7,11-tetrahydroxymethyl-6,12-diphenyl-3,9-
diazahexacyclo[6.4.0.02.7.04.11.05.10]dodecane (Az1) [15],
1,7-dihydroxymethyl-6,12-diphenyl-3,9-diphenoxycarbonyl
3,9-diazahexacyclo[6.4.0.02.7.04.11.05.10]dodecane (Az2) and
1,7-dihydroxymethyl-3,9-dimethoxycarbonyl-6,12-diphenyl-
3,9-diazahexacyclo[6.4.0.02.7.04.11.05.10]dodecane (Az3) has
been described [16].
zene and piperidine was added to a solution of 3b-acetoxy-
5,16-pregnadiene-20-one at 0 °C in dry benzene. The result-
ing solution was slowly warmed to room temperature and the
mixture was stirred for 24 h. After the reaction had gone to
completion, the solution was washed consecutively with dilute
HCl, NaHCO3 solution and water. The organic phase was
dried and the solvent was removed in vacuo. The residue was
purified by crystallisation to give pure products as white crys-
tals.
3. Results and discussion
2.1.2. Sulphoxides (P-S2 series) and sulphones
(P-S3 series) were prepared by oxidation
3.1. Thermal denaturation results
of the corresponding P-S1 product
3.1.1. Pregnanes (Table 1)
The 16a-thioethers were dissolved in ethanol and oxi-
dised by H2O2 (30% in water) at room temperature. After
completion of the reaction (24–48 h) two products were
detected by thin layer chromatography. The mixture was
poured onto acidic water and the precipitate was filtered,
washed, dried and separated by chromatography. The prod-
ucts of this reaction were the corresponding sulphoxide and
sulphone derivatives.
If the temperature of a solution containing a helical nucleic
acid is raised sufficiently, strand separation, or ‘melting’
2.2. Monosubstituted pyrido[3,2-g]quinolines: 4a, 4b, 4c
and 4d (Fig. 2)
Monosubstituted pyrido[3,2-g]quinoline derivatives were
obtained by alkylation of 2,10-dimethylpyridoquinoline-4-
one under phase transfer catalysis (PTC) conditions. 2,10-
Dimethylpyridoquinoline-4-one was obtained from commer-
Fig. 3. Structure of aza-fluorenes, Az1–3.

D. Sharples et al. / European Journal of Medicinal Chemistry 40 (2005) 195–202
197
Table 1
Chemically the pregnane derivatives investigated consist
of three families with varying side chains within each family
(Fig. 1). An analysis of the DTm values shows that there is no
obvious correlation, either between the three families or within
the families, between the DTm values obtained and the struc-
tural variations within the pregnanes investigated.
Change in Tm of CT DNA, plasmid DNA and transfer RNA in the presence
of pregnane derivatives
Pregnane
DTm calf thymus
DNA
DTm
plasmid
DNA
DTm tRNA
P-SH
-0.02
-1.07
-1.23
-2.50
-0.58
-0.60
-2.80
-0.58
-2.66
-2.50
-2.00
1.95
-5.45
-3.00
-3.64
-3.20
-3.40
-4.54
-6.59
-5.40
-5.20
-2.72
-11.82
-6.36
It must be assumed that the alteration of the oxidation state
of the sulphur atom with the same side chains or the varia-
tions of the latter within the same thioether, sulphoxide or
sulphone family produces both electronic and spatial differ-
ences. These competing variations would be expected to result
in a complex structure activity pattern in these compounds.
The results from log P calculations (Fig. 1) suggest that the
polarity of the pregnanes increases in the order P-S1 < P-S2
< P-S3 and whereas this shows very little effect on the inter-
action with DNA which is anyway weak, it shows a marked
effect on the interaction with tRNA (compare P-S1Ph DTm =
–3.64 °C with P-S3Ph DTm = –6.36 °C). A further contribut-
ing effect is the size of the R group attached to the sulphur.
Where R is small the effect on DTm is minimised (compare
P-S1Me, P-S2Me and P-S3Me). However, where R is large
and/or electronegative in nature the P-S3 series shows a
marked increase in interaction with tRNA (compare P-S1Ph
DTm = – 3.64 °C with P-S3Ph DTm = –6.36 °C and P-S3Am
DTm = –10.68 °C). One possible explanation is that the inter-
action with the DNA/tRNA occurs through the 20-ketone and
the 16-substituent at the same time. When it could form, the
strength of the interaction perhaps depends on the distances
between the interacting groups. Those distances are assumed
essentially determined by the spatial arrangement of the
16-substituent and partly influenced by its hydrophobic/
electronic effect.
P-S1Me
P-S1Ph
P-S1Be
P-S1cHe
P-S2Me
P-S2Ph
P-S2Be
P-S2cHe
P-S3Me
P-S3Na
P-S3Ph
P-S3Be
P-S3cHe
P-S3Am
P-S3SuAm
-4.10
-3.00
-0.83
-0.80
1.34
-1.15
-3.00
-1.60
-1.80
-2.00
-10.68
-8.18
occurs [3]. The temperature that marks the midpoint of the
melting process is called the melting temperature (or Tm). At
the Tm, half the nucleic acid exists in the helical state and the
other in the single-stranded state and the two species are in
equilibrium.
Helical nucleic acid ↔ single-stranded nucleic acid
The difference between the Tm of the nucleic acid and the
Tm observed in the presence of a ligand is termed DTm. The
DTm value is related to the affinity of binding of the ligand to
the nucleic acid. Ligands that bind to nucleic acids generally
bind to the helical state with a greater affinity than to the
single-stranded state. This stabilises the helical nucleic acid,
the equilibrium shifts to the left and the Tm increases since
more heat energy must be applied to the solution to dissoci-
ate the strands [4].
If the ligand binds to the single-stranded state more strongly
than to the helical state, then the equilibrium shifts to the right
and the Tm decreases and DTm is negative. A cationic azonia-
cyclophane, for example, destabilises RNA duplexes because
of the formation of an insertion complex with the bases of the
single strand [5]. Cisplatin destabilises DNA resulting in nega-
tive DTm values due to the formation of intrastrand covalent
linkages between bases [6].
An analysis of the results in Table 1 on the effect of preg-
nane derivatives on the Tm of various nucleic acids shows
some interesting variations. Largely negative DTm values are
observed, indicating a greater affinity for single-stranded
forms of the nucleic acid. The variations in Tm produced both
positive and negative values, which are relatively small
( 2 °C) in the case of both calf thymus and plasmid DNA.
However, with transfer RNA a large negative DTm is observed
(up to 11 °C). This indicates a much stronger affinity by the
pregnane derivatives for transfer RNA as opposed to DNA
and in particular for single-stranded regions of transfer RNA.
3.1.2. Monosubstituted pyrido[3,2-g]quinolines
The interaction of the monosubstituted pyrido[3,2-
g]quinolines with calf thymus DNA and plasmid DNA was
studied by two different UV spectroscopic techniques. The
results may be compared to the standard compounds,
9-aminoacridine, acridine orange, chlorpromazine and triflu-
perazine.
3.1.2.1. Thermal denaturation. All compounds raised the
denaturation temperature of calf thymus DNA and plasmid
DNA (Table 2). The increases in Tm, compared with the ref-
erence compounds, indicates that all four compounds inter-
act strongly with calf thymus DNA and plasmid DNA. The
interaction is probably by intercalation, given the planar struc-
ture and the presence of a protonatable nitrogen atom in the
side chain of these compounds [2].
3.1.2.2. UV spectroscopic shift and binding affınity determi-
nation. All four compounds showed a bathochromic and
hypochromic shift in the presence of calf thymus DNA indi-
cating intercalation into DNA (Table 3). The four com-
pounds were also titrated against calf thymus DNA and the
data fitted to a one site ligand binding model using the Grafit
version 5 curve fitting software (Erithacus Software).

198
D. Sharples et al. / European Journal of Medicinal Chemistry 40 (2005) 195–202
Table 2
Table 4
Change in Tm of CT DNA and plasmid DNA in the presence of pyrido[3,2-
MDR reversal of some pregnane derivatives on mouse T-lymphoma cell lines
g]quinoline derivatives and aza-oxafluorenes derivatives
Compounds
Concentration
Fluorescence activity
Compound
DTm calf thymus
DTm plasmid DNA
(µg ml–1
)
ratio
5.02
0.98
1.03
1.04
4.7
DNA
Verapamil
P-S1cHe
10
4
Pyrido[3,2-g]quinoline derivatives
4a
8.10
10.90
14.40
14.40
9.2
11.5
14.8
15.5
8
4b
16
4
4c
P-S2cHe
P-S3cHe
P-S1Be
P-S2Be
P-S3Be
4d
8
18.61
56.45
0.92
2.75
8.73
1.1
Aza-oxafluorenes
Az1
16
1
-1.50
0.00
2.90
1.50
Az2
2
Az3
1.00
1.50
4
9-Aminoacridine
Acridine orange
Phenothiazines
Promethazine
Trifluoperazine
10.70
12.60
12.00
9.60
4
8
2.47
4.04
6.28
10.66
22.83
3.89
11.27
45.37
0.7
16
4
1.00
2.20
5.10
7.00
8
16
1
Table 3
UV spectral shift and binding affinity data for monosubstituted pyrido[3,2-
g]quinolines binding to calf thymus DNA
2
4
Compound
Bathochromic Isosbestic Binding
Dissociation
DMSO control
Table 5
20
shift (nm)
point
capacity constant
(M)
0.22
0.60
1.70
1.40
0.09
4a
4b
4c
4d
1.0
1.0
1.0
1.0
404
404
405
405
442
0.25
0.29
0.56
0.39
0.14
MDR reversal of some monosubstituted pyrido[3,2-g]quinolines (4a–d) and
aza-oxafluorenes(Az1–3) on mouse T-lymphoma cell lines
Compounds
Concentration
Fluorescence activity
ratio
6.06
(µg ml–1
)
9-aminoacridine 4.0
Verapamil
4a
10
4
40.18
40
4
43.55
All four compounds show strong binding to DNA as com-
pared to 9-aminoacridine. On examining the structures of
these compounds, all four compounds possess the same het-
erocyclic nucleus with variations in the side chain. The weaker
binding of 4a and 4b may be rationalised by the bulky alkyl
groups surrounding the side chain nitrogen that could cause
some steric interference to binding. On the other hand 4c and
4d contain five and six membered pyrrolidine and piperidine
ring structures in the side chain, which are less bulky and
produce less interference. Alternatively the differences seen
could be due to structural constraints of the side chain and
access to the unshared electrons of the extracyclic nitrogen
atom.
4b
4c
4d
43.10
40
4
49.83
38.96
40
4
47.55
53.09
40
10
4
43.12
Verapamil
Az1
7.27
30.19
40
4
24.83
Az2
28.10
40
4
30.87
Az3
1.11
40
20
18.62
DMSO control
0.99
3.1.3. Aza-oxafluorenes
phenothiazines and aza-fluorene, Az3 were able to increase
the Tm value of calf thymus DNA. Pregnane derivatives desta-
bilised the nucleic acid structure by lowering the melting point
(m.p.). The results were similar when the compounds were
tested on pBR 322 and plasmid DNA but the interaction was
more marked on tRNA.
The three aza-oxafluorenes investigated showed only mar-
ginal interaction with calf thymus DNA but a slightly
increased interaction with plasmid DNA (Table 2).
3.2. MDR reversal results (Tables 4 and 5)
The biological activity, the efflux of p-gp 170 was studied
in the presence of compounds mentioned above. Therefore,
the efflux capacity of the human multidrug resistance gene
encoded p-glycoprotein was measured by flow cytometry after
10 min treatment by measurement of the rhodamine accumu-
lation. It was found that the majority of pyridoquinoline
derivatives, aza-oxafluorenes, phenothiazines and the preg-
Resistance modifiers are able to interact with the expres-
sion of particular genes and the protein responsible for drug
resistance of bacteria and cancer cells. In our experiments
several compounds were tested for interaction with both calf
thymus and plasmid DNA and with transfer RNA as in vitro
models. The monosubstituted pyridoquinoline compounds,

D. Sharples et al. / European Journal of Medicinal Chemistry 40 (2005) 195–202
199
nane derivatives were able to reduce the activity of MDR
efflux pump.
Yield: 5.2 g (78%), m.p.: 124–126 °C.
1H NMR (CDCl3) 0.60 (s, 3H, C-18), 0.99 (s, 3H, C-19),
2.02 (s, 3H, C-21), 2.05 (s, 3H, 3Ac), 2.52 (d, 1H, C-17),
3.67–3.73 (m, 1H 16, 2H, BzCH2), 4.60 (m, 1H, C-3), 5.37
(m, 1H, C-6) 7.22–7.31 (m, 5H, aromatic).
In the case of substituted pregnane derivatives, the follow-
ing compounds were extremely effective inhibitors of the p-gp
170: PS3-Be, PS2-cHe, PS2-Be, PS1-Be and PS1-cHe, the
last compounds were ineffective in low concentration from
1 to 16 µg ml–1, however, at 40 µg ml–1 they greatly enhanced
the drug accumulation. However two compounds, PS2-cHe,
PS3-cHe were toxic at this highest concentration. It is diffi-
cult to see a direct relationship between the effects of these
compounds on MDR and their interaction with nucleic acids.
Probably the interaction with DNA is unrelated to the MDR
reversal effect. However, the strong interaction with tRNA
particularly with the pregnane derivatives may suggest either
that the mechanism of MDR reversal is linked to disruption
of the function of tRNA in resistant cells or that similar struc-
tural features are responsible for both the interaction with
tRNA and the MDR reversal effect.
13C NMR (CDCl3) 13.8, 19.2, 20.8, 21.3, 27.6, 31.3, 31.5,
31.6, 34.4, 36.5, 36.9, 37.1, 38.0, 38.6, 41.2, 45.2, 49.6, 54.8,
71.7, 73.7, 122.1, 126.9, 128.7, 138.2, 139.6, 170.4, 207.1.
Oxidation of 3-acetoxy-16-benzylsulfanyl-pregn-5-ene-
20-one (P-S1Be): The reaction was carried out as described
by Romo et al., but without the addition of Na2CO3.
Ten millilitre of H2O2 (30% in water) was added at room
temperature to the P-S1Be (2.0 g, 4.16 mmol) dissolved in
150 ml of ethanol. The reaction was allowed to continue over
24 h and resulted in two products. The mixture was poured
onto acidic water and the precipitate was filtered, washed and
dried. The products were separated by column chromatogra-
phy on silica and identified as:
a) 3-Acetoxy-16-benzylsulfinyl-pregn-5-ene-20-one (P-
S2Be)
4. Conclusions
M.p.: 149–150 °C.
1H NMR (CDCl3) 0.64 (s, 3H, C-18), 1.02 (s, 3H, C-19),
2.03 (s, 3H, C-21), 2.14 (s, 3H, 3Ac), 3.00 (d, 1H, C-17),
3.79 (d, 1H, BzCH2), 3.98 (d, 1H, BzCH2), 4.60 (m, 1H, C-3),
5.38 (m, 1H, C-6), 7.27–7.38 (m, 5H, aromatic).
13C NMR (CDCl3) 14.0, 19.2, 20.8, 21.3, 22.3, 27.7, 31.3,
31.8, 36.5, 36.9, 38.0, 38.2, 45.2, 49.4, 55.3, 56.2, 57.6, 65.3,
73.7, 122.1, 128.3, 128.9, 129.8, 130.5, 130.7, 139.6, 170.4,
206.0.
The MDR reversal activity of a series of novel pregnanes,
pyridoquinolines and aza-fluorenes has been measured and
related to the interaction of these compounds with nucleic
acids. It was found that the majority of these compounds were
able to reverse MDR in sensitive cells. However, significant
differences were shown in the abilities of these types of com-
pounds to interact with nucleic acids. The pyridoquinolines
were able to intercalate strongly into DNA whereas the aza-
fluorenes and the pregnanes showed little or no interaction.
The pregnanes however showed significant ability to interact
with single-stranded regions of transfer RNA, which may sug-
gest that the mechanism of MDR reversal is linked to the
disruption of tRNA function in resistant cells.
b) 3-Acetoxy-16-benzylsulfonyl-pregn-5-ene-20-one
(P-S3Be)
M.p.: 199–200 °C.
1H NMR (CDCl3) 0.61 (s, 3H, C-18), 1.00 (s, 3H, C-19),
2.03 (s, 3H, C-21), 2.18 (s, 3H, 3Ac), 3.17 (d, 1H, C-17),
4.12 (t 1H C-16), 4.16 (s, 2H, BzCH2), 4.60 (m, 1H, C-3),
5.36 (m, 1H, C-6) 7.39–7.87 (m, 5H, aromatic).
13C NMR (CDCl3) 13.9, 19.2, 20.8, 21.3, 27.0, 31.2, 31.3,
31.4, 36.4, 38.2, 38.4, 44.6, 49.3, 55.3, 59.0, 57.6, 62.7, 63.9,
73.6, 121.8, 127.5, 128.5, 128.7, 129.1, 130.7, 139.6, 170.3,
205.1.
5. Experimental protocols
5.1. Chemistry
1H (400 MHz) and 13C (100 MHz) NMR spectra were
recorded with a BrukerAvance DRX 400 instrument at 25 °C
in CDCl3 solution and are reported as chemical shifts in ppm
(d) relative to TMS—m.p.s were determined with a Kofler
instrument and are uncorrected—elemental analyses were per-
formed with a Kovo (Czech Republic) CHN automatic analy-
ser and found to be in good agreements with the calculated
values.
5.1.1.2. 3-Acetoxy-16-phenylsulfanyl-pregn-5-ene-20-one
(P-S1Ph). This derivative was described by Sunthankar et al.
[18]. However, here it was prepared by reacting thiophenol
(7.5 g, 68 mmol) and PDA (5.0 g, 14 mmol) by the method
used by Romo et al. [17].
Yield: 5.0 g (76%), m.p.: 151–154 °C.
1H NMR (CDCl3) 0.66 (s, 3H, C-18), 1.01 (s, 3H, C-19),
2.02 (s, 3H, C-21), 2.03 (s, 3H, 3Ac), 2.61 (d, 1H, C-17),
4.20 (t, 1H, C-16), 4.60 (m, 1H, C-3), 5.35 (m, 1H, C-6),
7.18–7.35 (m, 5H, aromatic).
5.1.1. Pregnanes (Fig. 1)
13C NMR (CDCl3) 14.0, 19.2, 20.8, 21.3, 27.7, 31.2, 31.5,
31.7, 34.6, 36.5, 36.9, 38.0, 38.8, 43.6, 45.3, 49.7, 54.8, 70.9,
73.7, 122.0, 126.5, 128.7, 130.9, 136.1, 139.6, 170.4, 206.6.
Oxidation of 3-acetoxy-16-phenylsulfanyl-pregn-5-ene-
20-one (P-S1Ph): Ten millilitre of H2O2 (30% in water) was
5.1.1.1. 3-Acetoxy-16-benzylsulfanyl-pregn-5-ene-20-one
[17] (P-S1Be). Benzyl mercaptan (9.6 g, 77 mmol) and PDA
(5.0 g, 14 mmol) were reacted as described by Romo et al.
[17].

200
D. Sharples et al. / European Journal of Medicinal Chemistry 40 (2005) 195–202
added at room temperature to the P-S1Ph (2.0 g, 4.29 mmol)
dissolved in 150 ml of ethanol. The reaction was allowed to
continue over 24 h and resulted in two products. The mixture
was poured onto acidic water and the precipitate was filtered,
washed and dried. The two products were separated by col-
umn chromatography on silica and identified as:
a) 3b-Acetoxy-16a-phenylsulfinyl-pregn-5-ene-20-one
(P-S2Ph)
a) 3-Acetoxy-16-cyclohexylsulfinyl-pregn-5-ene-20-one
(P-S2cHe)
M.p.: 157–160 °C.
1H NMR (CDCl3) 0.7 (s, 3H, C-18), 1.02 (s, 3H, C-19),
2.03 (s, 3H, C-21), 2.22 (s, 3H, 3Ac), 3.26 (d, 1H, C-17),
3.92 (t, 1H, C-16), 4.60 (m, 1H, C-3), 5.36 (m, 1H, C-6).
13C NMR (CDCl3) 14.6, 19.3, 20.9, 21.3, 25.1, 25.3, 25.5,
26.4, 27.7, 29.9, 31.2, 31.5, 31.6, 36.6, 36.9, 38.0, 38.3, 44.8,
49.4, 53.2, 55.3, 58.3, 61.3, 73.7, 121.9, 139.7, 170.4, 206.9.
b) 3-Acetoxy-16-cyclohexylsulfonyl-pregn-5-ene-20-one
(P-S3cHe)
M.p.: 164–166 °C.
1H NMR (CDCl3) 0.66 (s, 3H, C-18), 1.00 (s, 3H, C-19),
2.02 (s, 3H, C-21), 2.03 (s, 3H, 3Ac), 2.62 (d, 1H, C-17),
4.22 (m, 1H, C-16), 4.63 (m, 1H, C-3), 5.38 (m, 1H, C-6),
7.19–7.50 (m, 5H, aromatic).
M.p.: 170–172 °C.
1H NMR (CDCl3) 66 (s, 3H, C-18), 1.02 (s, 3H, C-19),
2.03 (s, 3H, C-21), 2.23 (s, 3H, 3Ac), 2.73 (t, 1H, cyclohexyl-
1), 3.19 (d, 1H, C-17), 4.22 (t, 1H, C-16), 4.61 (m, 1H, C-3),
5.37 (m, 1H, C-6).
13C NMR (CDCl3) 14.0, 19.2, 20.8, 21.3, 27.5, 29.6, 31.3,
31.5, 31.6, 34.6, 36.5, 38.8, 43.6, 49.6, 54.8, 59.0, 60.8, 70.9,
73.7, 122.0, 124.4, 126.5, 128.7, 130.9, 136.1, 139.6, 170.4,
213.1.
13C NMR (CDCl3) 14.1, 19.2, 20.8, 21.3, 24.9, 25.1, 25.3,
27.6, 27.8, 31.3, 31.4, 31.5, 36.5, 36.9, 37.9, 38.3, 44.5, 49.3,
55.6, 56.6, 60.6, 62.9, 73.7, 121.9, 139.6, 170.3, 205.6.
Log P values for the pregnane derivatives (Fig. 1) were
calculated using interactive log P and log W predictor (Chem-
Silico http://www.logp.com).
b) 3-Acetoxy-16-phenylsulfonyl-pregn-5-ene-20-one
(P-S3Ph)
M.p.: 272–274 °C.
1H NMR (CDCl3) 0.59 (s, 3H, C-18), 1.00 (s, 3H, C-19),
2.00 (s, 3H, C-21), 2.03 (s, 3H, 3Ac), 3.14 (d, 1H, C-17),
4.28 (t, 1H, C-16), 4.62 (m, 1H, C-3), 5.37 (m, 1H, C-6),
7.52-7.87 (m, 5H, aromatic).
5.1.2. Monosubstituted pyrido[3,2-g]quinolines: 4a, 4b, 4c
and 4d (Fig. 2)
13C NMR (CDCl3) 14.9, 19.9, 21.5, 22.0, 27.8, 28.3, 31.7,
32.0, 32.1, 37.2, 37.6, 38.6, 39.1, 45.9, 50.0, 56.0, 63.2, 64.6,
74.3, 122.5, 129.2, 129.9, 134.3, 139.2, 140.3, 171.1, 205.5.
5.1.2.1. 8-Methyl-7-aminoquinoline. 2,6-Diaminotoluene
(4.51 g, 36.9 mmol), glycerol (28.5 g, 0.41 mol) and a freshly
prepared aqueous solution of arsenic (V) pentoxide (19.04 g
in 15 ml of water) were heated at 100 °C with stirring. A
solution of sulphuric acid (38 ml in 35 ml of water) was then
added to the mixture. The temperature was raised to 150 °C
for 4 h. After cooling, the mixture was poured into a large
flask and ammonia was added until precipitation occured. The
precipitate was filtered off, washed with water and dried.
Yield: 3.9 g (67%), m.p. 34–136 °C. 1H NMR (CDCl3): d = 8.9
(dd, J = 4.2 Hz, J = 1.0 Hz, 1H C-2), 8.0 (dd, J = 8.1 Hz,
J = 1.0 Hz, 1H C-3), 7.6 (d, J = 8.7 Hz, 1H C-4), 7.2 (dd,
J = 8.1 Hz, J = 4.3 Hz, 1H C-6), 7.0 (d, J=8.7 Hz, 1H C-5),
4.0 (m, 2H NH2), 2.6 (s, 3H CH3). 13C NMR(CDCl3): 149.7
(d), d = 144.9 (s), 135.2 (d), 122.6 (s), 126.3 (d), 119.9 (d),
118.5 (d), 117.4 (d), 115.1 (s), 10.4 (q).
5.1.1.3. 3-Acetoxy-16-cyclohexylsulfanyl-pregn-5-ene-20-
one (P-S1cHe). Cyclohexyl mercaptan (8.6 g, 74 mmol) dis-
solved in dry benzene was added to a solution of PDA (5.0 g,
14 mmol) and piperidine (4.0 ml, 40 mmol) in dry benzene at
8 °C. The mixture was warmed to room temperature and
stirred for 48 h. After completion of the reaction, the solution
was washed with dilute HCl, and 5% NaHCO3 solution, and
finally with water. The organic phase was dried over anhy-
drous Na2SO4 and the solvent was removed in vacuo. The
residue was purified by crystallisation from hexane to give
pure product.
Yield: 5.4 g (82%), m.p.: 124–126 °C.
1H NMR (CDCl3) 0.64 (s, 3H, C-18), 1.01 (s, 3H, C-19),
2.03 (s, 3H, C-21), 2.15 (s, 3H, 3Ac), 2.53 (d, 1H, C-17),
2.59 (m, 1H, cyclohexyl-1), 3.76 (t, 1H, C-16), 4.61 (m, 1H,
C-3), 5.37 (m, 1H, C-6).
5.1.2.2. 8-Methyl-7-[(2′-carboethoxy-1′-methylvinyl)amino)]-
quinoline. 7-Amino-8-methylquinoline (4.87 g, 30.7 mmol),
ethyl acetoacetate (4.0 g, 30.7 mmol), absolute ethanol
(35 ml), anhydrous CaSO4 (12 g) and few drops of acetic
acid were heated at 80 °C with stirring for 5 days. After fil-
tration, the solvent was removed in vacuo and the crude prod-
uct was used without further purification for the next step.
1H NMR (CDCl3): 11.5 (s, 1H NH), 8.9 (d, J = 4.2 Hz,
J = 1.6 Hz, 1H C-2), 8.1 (d, J = 8.2 Hz, J = 1.6 Hz, 1H C-6),
7.6 (d, J = 8.6 Hz, 1H C-5), 7.3 (dd, J = 8.2 Hz, J = 4.2 Hz,
1H C-3), 7.2 (d, J = 8.7 Hz, 1H C-4), 4.8 (s, 1H ethene H),
4.2 (q, J = 7.2 Hz, 2H ester CH2), 2.8 (s, 3H ring CH3), 1.9 (s,
3H ethene CH3), 1.25 (t, J = 7.2 Hz, 3H ester CH3).
13C NMR (CDCl3) 13.8, 19.3, 20.9, 21.4, 25.8, 26.1, 26.2,
27.7, 31.4, 31.6, 32.0, 33.7, 34.0, 36.1, 36.6, 36.9, 38.1, 38.7,
39.2, 44.3, 45.1, 49.7, 54.8, 72.1, 73.7, 122.1, 139.6, 170.5,
207.3.
Oxidation of the 3-acetoxy-16-cyclohexylsulfanyl-pregn-
5-ene-20-one (P-S1cHe): Ten millilitre of H2O2 (30% in
water) was added at room temperature to the P-S1cHe (2.0 g,
4.23 mmol) dissolved in 150 ml of ethanol. The reaction was
allowed to continue over 24 h and resulted in two products.
The mixture was poured onto acidic water and the precipitate
was filtered, washed, dried. The products were separated by
column chromatography on silica and identified as:

D. Sharples et al. / European Journal of Medicinal Chemistry 40 (2005) 195–202
201
5.1.2.3. 2,10-Dimethylpyrido-[3,2-g]quinoline-4-one. 8-
Methyl-7-aminoquinoline (6 g, 22 mmol) and diphenylether
(80 ml) were heated rapidly to 250 °C under nitrogen and
held for 1 h at this temperature. After cooling to 60 °C, the
mixture was poured onto petroleum ether (150 ml). The result-
ing red precipitate was filtered, washed thoroughly with
ethylether and dried.
lidinyl CH2). 13C NMR (CDCl3): 161.4 (s), 160.5 (s), 150.6
(d), 145.8 (s), 145.5 (s), 137.2 (s), 134.0 (s), 125.1 (s), 120.0
(s), 119.8 (d), 118.8 (d), 99.7 (d), 66.8 (t), 57.5 (t), 55.1 (t),
26.0 (q), 24.1 (d), 12.3 (q).
For 4d yield: 0.29 g (30%); m.p.: 135–1374 °C. 1H NMR
(CDCl3): 9.0 (dd, J = 3.9 Hz, J = 1.9 Hz, 1H C-2), 8.5 (s,1H
C-8), 8.3 (dd, J = 8.5 Hz, J = 1.9 Hz, 1H C-3), 7.4 (dd,
J = 8.5 Hz, J = 4.0 Hz, 1H C-4), 6.6 (s, 1H C-5), 4.4 (t,
J = 6.2 Hz, 2H C-2′), 3.4 (s, 3H C-12), 3.0 (t, 2H C-1′), 2.8 (s,
3H C-11), 2.6 (m, 4H piperidinyl CH2), 1.6 (m, 6H piperidi-
nyl CH2). 13C NMR (CDCl3): 161.4 (s), 160.5 (s), 150.6 (d),
145.8 (s), 145.5 (s), 137.2 (s), 134.0 (s), 125.1 (s), 120.0 (s),
119.8 (d), 118.8 (d), 99.7 (d), 66.8 (t), 57.5 (t), 55.1 (t), 26.0
(q), 24.1 (d), 12.3 (q).
1
Yield: 1.95 g (56%), m.p.: 255 °C. H NMR (dimethyl
sulphoxide, DMSO-d6): 10.5 (s, 1H OH), 9.0 (dd, J = 3.9 Hz,
J = 1.8 Hz, 1H C-2), 8.6 (s, 1H C-8), 8.5 (dd, J = 8.3 Hz,
J = 1.8 Hz, 1H C-3), 7.5 (dd, J = 8.3 Hz, J = 3.9 Hz, 1H C-4),
5.9 (s, 1H C-5), 3.0 (s, 3H C-12), 2.5 (s, 3H C-11). 13C
NMR(CDCl3): 149.7 (s), 144.9 (s), 135.2 (s), 122.6 (s), 126.3
(s), 119.9 (d), 118.5 (s), 99.9 (d), 67.4 (t), 51.04 (t), 48.1 (t),
27.0 (q), 12.4 (q), 12.1 (q).
5.2. Biology
5.1.2.4. 2,10-Dimethylpyrido-[3,2-g]quinolines (4a–d). 2,10-
Dimethylpyrido-[3,2-g]quinoline-4-one (1 g, 4.46 mmol) was
dissolved in toluene (70 ml) and either 1-chloro-2-
diethylaminoethane hydrochloride (1.54 g, 8.92 mmol, for
4a, 1-chloro-2-diisopropylaminoethane hydrochloride (1.1 g
8.92 mmol) for 4b, 1-chloro-2-pyrrolidinoethane hydrochlo-
ride (1.2 g, 8.92 mmol) for 4c or 1-chloro-2-piperidinoethane
hydrochloride (1.3 g, 8.92 mmol) for 4d was added. To this
solution was added 50% aqueous KOH (35 ml) and tetrabu-
tylammonium bromide (0.3 g, 0.9 mmol) and the solution
was heated at 110 °C with stirring, for 24 h. The reaction
mixture was then filtered whilst still warm. The aqueous phase
was extracted three times with toluene. The organic layers
were dried over anhydrous Na2SO4. After removal of the sol-
vent, the oily product was refrigerated until solid and then
washed with cold toluene.
5.2.1. Materials
5.2.1.1. Nucleic acids. Highly polymerised type 1 calf thy-
mus DNA sodium salt and type XXI transfer RNA were from
the Sigma Chemical Corporation, pBR 322 plasmid DNA was
prepared by Birnboim-Doly rapid alkaline extraction proce-
dure [7,8].
5.2.1.2. Buffer solutions. For CT DNA and plasmid DNA:
tris(hydroxymethyl)aminomethane (Tris, 0.003 M), NaCl
(0.018 M) buffer adjusted to pH 7.0 was used.
For transfer RNA: Tris (0.1 M), NaCl (0.1 M), hydrated
MgCl2 (0.01 M) buffer adjusted to pH 7.5 was used.
5.2.1.3. Acridine-derivatives, phenothiazines. 9-Aminoacri-
dine (Aldrich-Chemie, D-7924 Steinheim), Acridine-orange
(Reanal Budapest, Hungary), Promethazine (EGIS Pharma-
ceutical Company, Budapest, Hungary), trifluoperazine
(Sigma Chemical Co., St. Louis, USA) were used as internal
standards.
Stock soutions of all compounds were prepared at a con-
centration of 1 mg ml–1. 9-Aminoacridine, acridine-orange,
promethazine and trifluoperazine were dissolved in deion-
ised water, all other compounds were dissolved in DMSO.
For 4a yield: 0.29 g (30%); m.p.: 142–144 °C. 1H NMR
(CDCl3): 9.0 (dd, J = 3.9 Hz, J = 1.9 Hz, 1H C-2), 8.5 (s, 1H
C-8), 8.3 (dd, J = 8.5 Hz, J = 1.8 Hz, 1H C-3), 7.3 (dd,
J = 6.5 Hz, J = 3.9 Hz, 1H C-4), 6.5 (s, 1H C-5), 4.25 (t,
J = 6.2 Hz, 2HC-2′), 3.3 (s, 3H C-12), 3.05 (t, 2H C-1′), 2.7
(s, 3H C-11), 2.7 (q, J = 6.3 Hz, 4H ethyl CH2), 1.1 (t,
J = 2.1 Hz, 6H ethyl CH3). 13C NMR (CDCl3): 161.6 (s),
160.7 (s), 150.8 (d), 145.9 (s), 145.6 (s), 137.3 (d), 134.1 (s),
125.2 (s), 120.15 (s), 119.9 (d), 118.9 (d), 99.9 (d), 67.4 (t),
51.04 (t), 48.1 (t), 27.0 (q), 12.4 (q), 12.1 (q).
5.2.2. Methods
1
For 4b yield: 0.19 g (20%); m.p.: 140 °C. H NMR
(CDCl3): 9.0 (dd, 1H C-2), 8.6 (s, 1H C-8), 8.3 (dd, 1H C-3),
7.3 (dd, 1H C-4), 6.6 (s, 1H C-5), 4.2 (t, J = 7.1, 2H isopropyl
CH), 3.4 (s, 3H C-12), 3.1 (m, 4H 1′,2′CH2), 2.8 (s, 3H C-11),
1.1 (d, J = 6.5, 12H isopropyl CH3). 13C NMR(CDCl3): 161.6
(s), 160.7 (s), 150.6 (s), 145.9 (s), 145.5 (s), 137.2 (d), 133.9
(s), 125.1 (s), 120.0 (s), 119.9 (d), 118.9 (d), 99.8 (d), 69.7
(t), 49.6 (d), 44.1(t), 26.9 (q), 20.9 (q), 12.3 (q).
5.2.2.1. Nucleic acid interactions: thermal denaturation stud-
ies. Calf thymus DNA (10 mg) was suspended in 0.03 M Tris
hydrochloride buffer containing 0.018 M NaCl adjusted to
pH 7.0 (10 ml) and the solution was kept at 4 °C for at least
3 days before use. This stock solution was then diluted to the
working concentration immediately before use. Transfer RNA
(10 mg) was suspended in 0.1 M Tris hydrochloride buffer
containing 0.1 M NaCl and 0.01 M hydrated MgCl2 adjusted
to pH 7.5 (10 ml) and the solution was kept at 4 °C for at least
3 days before use. This stock solution was then diluted to the
working concentration immediately before use. Plasmid DNA
(1 mg) was suspended in 0.03 M Tris hydrochloride buffer
containing 0.018 M NaCl adjusted to pH 7.0 (1 ml).
For 4c yield: 0.29 g (30%); m.p.: 148–150 °C. 1H NMR
(CDCl3): 9.0 (dd, J = 3.9 Hz, J = 1.9 Hz, 1H C-2), 8.5 (s, 1H
C-8), 8.3 (dd, J = 8.5 Hz, J = 1.9 Hz, 1H C-3), 7.4 (dd,
J = 8.5 Hz, J = 4.0 Hz, 1H C-4), 6.6 (s, 1H C-5), 4.4 (t,
J = 6.2 Hz, 2H C-2′), 3.4 (s, 3H C-12), 3.0 (t, 2H C-1′), 2.8 (s,
3H C-11), 2.6 (m, 4H pyrrolidinyl CH2), 1.6 (m, 4H pyrro-

202
D. Sharples et al. / European Journal of Medicinal Chemistry 40 (2005) 195–202
The concentrations of the polynucleotide solutions were
cator Rhodamine 123 was added to the samples and the cells
were incubated for further 20 min at 37 °C, washed twice
with phosphate buffered saline (PBS) and resuspended in
0.5 ml PBS for analysis. The fluorescence of the cell popula-
tions was measured by flow cytometry using a Beckton Dick-
inson FACScan fluorimeter. Verapamil was used as a positive
control in the Rhodamine 123 exclusion experiments. The
percentage of mean fluorescence intensity was calculated
using the following equation on the basis of measured fluo-
rescence values:
determined spectrophotometrically in terms of nucleotide
phosphate and calculated from an extinction coefficient at
260 nm of 6600 M–1 cm–1 for calf thymus DNA [9] and of
5.3 × 105 M–1 cm–1 for tRNA [10].
Measurement of the thermal denaturation profiles of ligand:
DNA complexes were made for ligand: polynucleotide ratios
of 1:10 and were recorded on a Cary Varian Model IE spec-
trophotometer using a Cary temperature controller con-
nected to a Cary 1/3 multicell block. The solutions were
allowed to equilibrate for 20 min before increasing the tem-
perature and the temperature then increased at a rate of
0.5 °C min–1.A cell containing polynucleotide solution alone
was always measured along side cells containing ligand/
polynucleotide mixtures to act as an internal standard. The
blank cells in all measurements contained Tris buffer solu-
tion and all cells were stoppered with teflon caps, after debub-
bling and space was allowed for expansion of the solutions.
The mid point of the thermal denaturation profile of the solu-
tions (Tm) was determined by calculating the average absor-
bance using the instruments thermal application software [1].
R (fluorescence activity ratio) =
MDR treated ⁄ MDR control
Parental treated ⁄ Parental control
Acknowledgements
This research has been carried out in the frame of the Euro-
pean Community COST B16 action and Szeged Foundation
for Cancer Research. The authors are grateful to Dr. Peter
Forgo for NMR analytic work.
5.2.2.2. Nucleic acid interactions: UV spectral shifts
and binding affınity determination. In the case of the mono-
substituted pyrido[3,2-g]quinolines: 4a, 4b, 4c and 4d which
all exhibit a absorption maxima in the visible region well away
from the absorption maxima of nucleic acids, it was possible
to investigate the interaction of these compounds with nucleic
acids by UV spectral shift analysis.
The spectra of a series of solutions of the monosubstituted
pyrido[3,2-g]quinolines were measured in the presence of
increasing amounts of either CT DNA or plasmid DNA. The
spectra were recorded on a Cary Varian 1E spectrophotom-
eter in 0.03 M Tris buffer containing 0.018 M NaCl at pH
7.00 for 2.50 × 10–5 M. The bathochromic shift (if any) was
determined and a binding affinity constant (k) calculated from
the hypochromic shift recorded at the kmax of each com-
pound using the Grafit version 5 curve fitting software (Erith-
acus Software).
References
[1] J. Nacsa, L. Nagy, D. Sharples, A. Hever, D. Szabo, I. Ocsovski, et al.,
Anticancer Res. 18 (1998) 3093–3098.
[2] A. Cremieux, J. Chevalier, D. Sharples, H. Berny, A.M. Galy,
P. Brouant, et al., Res. Microbiol. 146 (1995) 73–83.
[3] W. Saenger, in: C.R. Cantor (Ed.), Principles of Nucleic Acid Struc-
ture, vol. 6, Springer-Verlag, New York, 1984, pp. 141–149 pp.
[4] W.D. Wilson, F.A. Tanious, M. Fernandez-Saiz, C.T. Rigl, in:
K.R. Fox (Ed.), Drug-DNA Interaction Protocols, vol. 15, Humana
Press, New Jersey, 1997, pp. 219–240 pp.
[5] M. Fernandez-Saiz, H.-J. Schneider, J. Sartorius, W.D. Wilson, J. Am.
Chem. Soc. 118 (1996) 4739–4745.
[6] V. Brabec, J. Reedijk, M. Leng, Biochemistry 31 (1992) 12397–
12402.
[7] H.C. Birnboim, J. Doly, Nucleic Acids Res. 7 (1979) 1513–1523.
[8] C. Miskolci, I. Labádi, T. Kurihara, N. Motohashi, J. Molnár, Int. J.
Antimicrob. Agents 14 (2000) 243–247.
[9] L.M. Angerer, E.N. Moudrianakis, J. Mol. Biol. 63 (1972) 505–513.
[10] V.V. Filimontov, P.L. Privalov, J. Glangloff, Biochim. Biophys. Acta
521 (1978) 209–218.
[11] L. Bradford, T.J. Elliott, F.M. Rouve, J. Chem. Soc. 69 (1947) 437–
445.
[12] F. Sauter, U. Jordis, P. Martinek, G. Cai, Pharm. Sci. 57 (1989) 7–20.
[13] M. Conrad, L. Limpach, Ber 20 (1887) 949–951.
[14] M. Conrad, L. Limpach, Ber 21 (1888) 524–526.
[15] A. Hilgeroth, M. Wiese, A. Billich, J. Med. Chem. 42 (1999) 4729–
4732.
[16] A. Hilgeroth, H. Lilie, Eur. J. Med. Chem 38 (2003) 495–499.
[17] J. Romo, M. Romero, C. Djerassi, G. Rosenkranz, J. Am. Chem. Soc.
73 (1951) 1528–1538.
5.2.2.3. Assay for reversal of multidrug resistance in tumour
cells. The L2178 (parent) mouse T cell lymphoma cell line
and its transformed subline with human MDR-1 gene were
used. The cells were grown in McCoy’s 5A medium supple-
mented with 10% heat-inactivated horse serum, L-glutamine
and antibiotics. The cells were adjusted to a concentration of
2 × 106 ml–1 and resuspended in serum-free McCoy’s 5A
medium and were distributed into 0.5 ml aliquots to Eppen-
dorf centrifuge tubes. The tested compounds were added at
various concentrations in different volume of the stock solu-
tions and the samples were incubated for 10 min at room tem-
perature. Then 10 µl (5.2 µM final concentration) of the indi-
[18] S.V. Sunthankar, S.G. Telang, J. Indian, Chem. 11 (1973) 303–309.