Synthesis, G-quadruplex stabilisation, docking studies, and effect on cancer cells of indolo[3,2-b]quinolines with one, two, or three basic side chains

CHEMMEDCHEM

FULL PAPERS

DOI: 10.1002/cmdc.201300288

Synthesis, G-Quadruplex Stabilisation, Docking Studies,

and Effect on Cancer Cells of Indolo[3,2-b]quinolines with

One, Two, or Three Basic Side Chains

Jo¼o Lavrado,^[a]Pedro M. Borralho,^{[b, c]}Stephan A. Ohnmacht,^[d]Rui E. Castro,^{[b, c]}

Cecꢀlia M. P. Rodrigues,^{[b, c]}Rui Moreira,^[a]Daniel J. V. A. dos Santos,^{[a, e]}Stephen Neidle,^[d]and

Alexandra Paulo*^[a]

G-quadruplex (G4) DNA structures in telomeres and oncogenic

promoter regions are potential targets for cancer therapy, and

G4 ligands have been shown to modulate telomerase activity

and oncogene transcription. Herein we report the synthesis

and G4 thermal stabilisation effects, determined by FRET melt-

ing assays, of 20 indolo[3,2-b]quinolines mono-, di-, and trisub-

stituted with basic side chains. Molecular modelling studies

were also performed in an attempt to rationalise the ligands’

binding poses with G4. Overall, the results suggest that ligand

binding and G4 DNA thermal stabilisation increase with an N5-

methyl or a 7-carboxylate group and propylamine side chains,

whereas selectivity between G4 and duplex DNA appears to

be modulated by the number and relative position of basic

side chains. From all the indoloquinoline derivatives studied,

the novel trisubstituted compounds 3d and 4d, bearing a 7-

(aminoalkyl)carboxylate side chain, stand out as the most

promising compounds; they show high G4 thermal stabilisa-

tion (DT_mvalues between 17 and 88C) with an inter-G4 DT_m

trend of Hsp90A>KRas21RꢀF21T>c-Kit2, 10-fold selectivity

for G4 over duplex DNA, and 100-fold selectivity for the

HCT116 cancer cell line (IC₅₀and IC₉₀: <10 mm) over primary

rat hepatocytes. Compounds 3d and 4d also decreased pro-

tein expression levels of Hsp90 and KRas in HCT116 cancer

cells.

Introduction

Nucleic acid sequences containing repetitive guanine (G)-rich

tracts can form G-quadruplex higher-order structures (G4), in

which guanine bases form stacked planar G-quartets stabilised

by Hoogsteen hydrogen bonding and cation coordination.^[1]

Bioinformatics analyses of the human genome have revealed

an elevated frequency in the occurrence of these sequences,

notably in telomeres and promoter regions of genes such as

the oncogenes c-Myc, c-Kit, and K-Ras,^[2]thus suggesting that

G4s may naturally act as regulatory elements, particularly in

cell proliferation processes.^[3]Recently, the 27-nucleotide se-

quence located 77 nucleotides upstream of the transcription

start site of Hsp90, a gene that encodes a protein involved in

the regulation of cell proliferation and a validated target for

cancer therapy,^[4]was also shown to form stable quadruplex

DNA structures.^[5]Down-regulation of transcription by small

molecules that are able to induce and stabilise DNA G4 struc-

tures has been reported for oncogenes c-Myc,^[6]c-Kit,^[7]KRas,^[8]

and Hsp90,^[9]reinforcing the concept that G4s are promising

targets for the design of new compounds (G4 ligands) with po-

tentially high cancer cell selectivity and lower toxicity than

standard DNA-targeting anticancer drugs.^[3a]Moreover, target-

ing G-quadruplexes in promoter oncogenes has several advan-

tages over targeting expressed proteins, including the de-

creased likelihood of point mutations and resistance. However,

from a drug-discovery perspective, we are still taking the first

steps, as there is currently restricted diversity in the available

G4 ligands, and these have low affinity and selectivity for G4

relative to protein inhibitors.^[3a]The large planar and aromatic

surface of a terminal G-quartet is a common G4 chemical fea-

ture that has led to the development of mainly three G4

ligand families: the macrocyclic family, of which the porphyrin

derivative TMPyP4 and telomestatin are well-known examples;

the polyaromatic fused family, which includes acridines, anthra-

[a] Dr. J. Lavrado, Prof. R. Moreira, Dr. D. J. V. A. d. . Santos, Dr. A. Paulo

Medicinal Chemistry Group

Research Institute for Medicines and Pharmaceutical Sciences

Faculty of Pharmacy, University of Lisbon

Av. Prof. Gama Pinto, 1649-003 Lisbon (Portugal)

E-mail: mapaulo@ff.ul.pt

[b] Dr. P. M. Borralho, Dr. R. E. Castro, Prof. C. M. P. Rodrigues

Molecular & Cell Biology of Eukaryotic Systems Group

Research Institute for Medicines and Pharmaceutical Sciences

Faculty of Pharmacy, University of Lisbon

Av. Prof. Gama Pinto, 1649-003 Lisbon (Portugal)

[c] Dr. P. M. Borralho, Dr. R. E. Castro, Prof. C. M. P. Rodrigues

Department of Biochemistry and Human Biology

Faculty of Pharmacy, University of Lisbon

Av. Prof. Gama Pinto, 1649-003 Lisbon (Portugal)

[d] Dr. S. A. Ohnmacht, Prof. S. Neidle

The School of Pharmacy, University College London

29/39 Brunswick Square, London WC1N 1AX (UK)

[e] Dr. D. J. V. A. d. . Santos

REQUIMTE, Department of Chemistry & Biochemistry

Faculty of Sciences, University of Porto

R. do Campo Alegre, 4169-007 Porto (Portugal)

Supporting information for this article is available on the WWW under

http://dx.doi.org/10.1002/cmdc.201300288.

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 2013, 8, 1 – 15 1

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

quinones, and indoloquinoline derivatives; and the family of

aromatic non-fused systems such as 1,4-triazole derivatives.^[10]

Many of these G-quadruplex ligands are selective for G4 struc-

tures over duplex DNA, but the design of inter-G4 selective li-

gands remains challenging. However, in view of the high

number of putative G4-forming sequences revealed by the

analysis of the human genome, this issue needs to be ad-

dressed in order to avoid secondary and undesirable effects in

any therapeutic application of G4 targeting. Monomolecular

G4s can adopt a wide diversity of topologies and have various

loops, grooves, and capping structures as a consequence of

differences in their primary DNA sequences. Adenines, thy-

mines, and phosphate groups present in loops, grooves, and

Embracing the challenge of expanding the chemical toolbox

of effective G4 ligands for cancer therapy, herein we describe

the design, synthesis, and G4 thermal stabilisation effects of

a set of mono-, di-, and trisubstituted indolo[3,2-b]quinolines

against three different DNA sequences—F21T human telomeric

quadruplex (F21T), c-Kit2, and Hsp90A quadruplexes—using

a high-throughput fluorescence resonance energy transfer

(FRET) melting assay. Molecular modelling was then used to ra-

tionalise the binding to telomeric quadruplexes, and the anti-

proliferative activity of the most effective and selective ligands

were evaluated in cancer and normal cells.

capping structures of G4 provide distinct patterns of hydrogen Results and Discussion

bonding, hydrophobic p surfaces, and negative charges that

can be selectively targeted.^[11]Crystallographic and NMR stud-

Design of target compounds

ies of G4–ligand complexes have shown that fused polyaro-

matic ligands such as acridines and indoloquinolines bind to

G4s through p–p stacking with guanines of terminal G-quar-

tets, while side chains target the grooves and loops of the G4,

contributing both to enhanced binding and selectivity be-

tween G4s.^[12]

Previous studies have shown that thermal stabilisation of telo-

meric or c-Myc G4 by quindoline (1) derivatives increases with

increasing basicity of the quinoline nitrogen (N5) atom, for ex-

ample, when an electron-donating group (NH) is present at

C11.^[14b]These results have been interpreted as indicating that

a positive charge near the aromatic core, due to protonation

of N5 at physiological pH, is an important feature to improve

binding to external G-quartets of G4 DNA structures.^[14a,b]

Methylation of N5, leading to a stable positive charge, further

increased binding affinity and thermal stability of telomeric G4,

an effect that was ascribed to increased p–p stacking interac-

tions, due to a decrease in electron density in the aromatic

core.^[14c]As reported for other G4 ligands, the basicity of the

quindoline side chains is also important for G4 binding,^[14c]but

other chemical features of basic side chains remain unexploit-

ed. In this context, we first decided to expand the chemical di-

versity of the basic side chain at C11 and to evaluate the effect

of N5-methyl quindoline derivatives 2c–m, containing various

alkyl side chain lengths, terminal amine groups (primary, sec-

ondary, or tertiary), and different aryl side chains on G4 thermal

stabilisation.

Indoloquinolines are tetracyclic aromatic alkaloids with high

potential to be developed into anticancer drugs, but have

been insufficiently exploited.^[13a]Derivatives of natural indolo-

[3,2-b]quinolines 1 (quindoline) and 2a (cryptolepine) were

previously shown to be good G4 ligands, telomerase inhibitors,

oncogene (c-Myc) transcription inhibitors, and are able to

induce cell-growth arrest (Figure 1).^[14]Moreover, the drug-like

It has been reported that appending two basic side chains

to C2 and C7 or N10 and C11 of the quindoline core increases

binding to telomeric^[16]or c-Myc G4 structures.^[14b]Because no

other positions of the asymmetrical quindoline nucleus have

been exploited to modulate G4 binding, we next designed

new indolo[3,2-b]quinolines, 3 and 4, with two and three basic

side chains, to evaluate the effect of various alkylamine substi-

tution patterns and of substituents at positions 3 and 7 on in-

teractions with different G4 structures.

Figure 1. Quindoline (1), cryptolepine (2a), C11 derivatives 2b–m, and quin-

doline derivatives 3 and 4 screened against a small panel of G-quadruplexes.

analogue NSC748393 (2b) showed good in vitro and in vivo

anticancer profiles, as well as strong binding affinity to a telo-

meric G-quadruplex.^[15]However, as for most G-quadruplex li-

gands known to date, C11-substituted or N10,C11-disubstituted

indolo[3,2-b]quinolines are poorly selective G4 ligands.^[14b,15]

Nevertheless, the asymmetric shape of the indolo[3,2-b]quino-

line aromatic core and the substantial derivatisation potential

offered by this structure suggest that it is a very appropriate

scaffold for exploitation to obtain efficient and selective G4 li-

gands.

Chemistry

Indolo[3,2-b]quinolines 2–4 were synthesised via quindolones

8 (indolo[3,2-b]quinolin-11-ones), according to the route illus-

trated in Scheme 1 and based on a procedure developed by

Gçrlitzer and Weber and adapted by Bierer.^[13]In the first step

anthranilic acids 5 were allowed to react with bromoacetyl

bromide to afford compounds 6, which were then converted

into the corresponding 2-(2-(phenylamino)acetyl)benzoic acids

7 by reaction with the appropriate aniline. Acid-promoted bi-

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&

2

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

In addition, acid-promoted bicyclisation of 7e, with a trifluor-

omethyl substituent, gave quindolone 8c, with a carboxylic

acid group, as a bicyclisation product of 7c. Compound 8c

was fully characterised by NMR spectroscopy, which revealed

two carbonyl carbons in the ¹³C spectra at 167.92 and

168.18 ppm. The carbon at 168.18 ppm showed a ¹H–¹³C

HMBC correlation with the singlet resonating at 8.99 ppm as-

signed to the proton at position 6 of the aromatic nucleus.

Analysis of the 2D NMR spectra allowed assignment of the re-

maining carbonyl signal to C11 of the aromatic nucleus. Fur-

thermore, ¹⁹F NMR data of the product from 7e bicyclisation

did not reveal any signal resonating in the spectrum, in con-

trast to what was observed for the starting material 7e, which

showed a singlet at À59.09 ppm. Mass spectrometric analysis

also corroborates the presence of a carboxylic acid instead of

a trifluoromethyl substituent. In this particular system, the elec-

tronic effect of the indole nitrogen atom promoted an S_N1-

type cleavage of the CÀF bond in the trifluoromethyl group, as

described for 3-(trifluoromethyl)-1H-indoles, followed by hy-

drolysis of the difluoromethylene intermediate.^[17]

Scheme 1. Synthesis of the indolo[3,2-b]quinoline nucleus. Reagents and con-

ditions: a) bromoacetyl bromide, DMF/1,4-dioxane (1:1), RT, overnight; b) ap-

propriate aniline; c) PPA, 1308C, 2 h.

cyclisation of 7a–c with polyphosphoric acid (PPA) afforded

the corresponding quindolones 8a–c. However, PPA-promoted

cyclisation of 7d did not afford the corresponding quindolone

(R¹=H, R²=NO₂), but another unexpected compound. The

Cryptolepine (2a) and derivatives 2c–m were obtained from

8a after 11-chlorination with phosphorus oxychloride. Hydro-

genation and N5-methylation with methyl trifluoromethanesul-

fonate gave 1 and N5-methylation of the 11-chloro intermedi-

ate, followed by nucleophilic substitution with the required al-

kyldiamine to yield 2c–m. All cryptolepine derivatives were

fully characterised by NMR, and in all cases, purities were

ꢁ95%, as described elsewhere.^[18]Quindoline derivatives 3 and

4 were obtained by reaction of 8 with the appropriate chlor-

oalkylamine in the presence of a base, as described in a syn-

thetic procedure recently reported by our group (Scheme 3).^[19]

In the case of the intermediate quindolone 8c, the same syn-

thetic procedure afforded the trialkylated derivatives 3d and

1

¹H NMR spectra and H–¹H COSY data showed a total of eight

protons, instead of seven protons, and two inter-correlated

sets of doublets integrating for two protons each at 6.56 and

7.52 ppm, as in 7d for the 4-nitroaniline ring. The total assign-

ment of the ¹H and ¹³C NMR chemical shifts by analysis of

2D HMQC and HMBC allowed the identification of 2-hydroxy-3-

((4-nitrophenyl)amino)quinolin-4(1H)-one (7id) as the product

of PPA-catalysed cyclisation of 7d. This observation led us to

propose the two-step mechanism depicted in Scheme 2. In the

case of 7d, bearing a para-nitro group in the aniline moiety,

the indoloquinoline cyclisation reaction stopped after the clo-

sure of the six-membered ring, possibly due to the strong elec-

tron-withdrawing effect of the nitro substituent which preclud-

ed closure of the five-membered ring and formation of the

final quindolone.

Scheme 3. Synthesis of indolo[3,2-b]quinoline derivatives 3 and 4. Reagents

and conditions: a) chloroalkylamine, K₂CO₃, dry acetone, reflux, overnight.

4d. The positions of N-alkylamine side chains in compounds

1

3a–d and 4a–d were assigned on the basis of H–¹³C correla-

tions between CH₂protons of the side chain and quaternary

carbon atoms of the quindoline aromatic structure, observed

in HMBC spectra and confirmed by NOE experiments. For ex-

Scheme 2. Acid-promoted cyclisation mechanism of 7 to afford quindolones

8.

1

ample, H and ¹³C spectra of derivative 4d displayed signals

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&

3

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

corresponding to the introduction of three side chains in the

quindoline nucleus. The disappearance of the carbonyl carbon

signal resonating at 167.92 ppm in 8c (C11), replaced by a typi-

cal phenoxy carbon signal at 145.15 ppm and an additional

1

ether carbon (74.79 ppm), correlating in the HMQC with a H

triplet (4.31 ppm), indicated the presence of an alkoxy chain at

C11.

This triplet revealed a NOE correlation with a triplet at

4.73 ppm, which correlated in HMBC with C9a and C10a

1

(147.36 and 125.30 ppm, respectively) and in the H–¹H COSY

spectrum with two methylene protons, corroborating the in-

troduction of a second side chain at position N10. The third

1

side chain was characterised in H NMR data by two triplets, in-

tegrating to two protons each, at 2.96 and 4.49 ppm, one

quartet at 2.70 ppm integrating for four protons, and one trip-

let at 1.12 ppm integrating for six protons. The triplet at

4.49 ppm correlated in the HMBC spectrum with the carbonyl

carbon (166.84 ppm), which in turn correlated with the singlet

at 9.24 ppm, assigned to H6, thus confirming the introduction

of the third side chain in the carboxylic acid moiety of 8c and

formation of the 2-(diethylamino)ethyl-11-[2-(diethylamino)-

ethoxy]-10-[2-(diethylamino)ethyl]-10H-indolo[3,2-b]quinolone-

7-carboxylate (4d) (Supporting Information).

G-quadruplex stabilisation

The ability of the indolo[3,2-b]quinolines 2–4 to bind and sta-

bilise G4 structures and a self-complementary hairpin DNA (T-

loop) was evaluated by FRET melting assays with ligand con-

centrations ranging from 0.1 to 5 mm. The FRET melting data,

given as changes in melting temperature (DT_m) and obtained

with fluorescently labelled DNA G4 sequences (0.2 mm) of F21T

human telomeric DNA^[20]and of oncogene promoter sequen-

ces c-Kit2^[7a]and Hsp90A,^[5]as well as with T-loop in potassium

ion buffer, are summarised in Tables 1 and 2. The results show

that the indolo[3,2-b]quinolines 2–4 produce a wide range of

G4 stabilisation effects, and, for all compounds, these effects

are concentration dependent, with much higher stabilisation

effects at 5 mm than at 1 mm, as shown in Figure 2 for 2e and

4d (see Supporting Information for the remaining com-

pounds).

Figure 2. Concentration-dependent FRET melting profile of G4 structures

and T-loop DNA complexed with compounds a) 2e and b) 4d.

ary amine (2e–g) does not significantly change the complex

stabilisation, despite the differences between pK_avalues ob-

served for those amine groups (pK_avalues ranging from 8.7 to

~12).^[21]However, compound 2d, which contains a weak termi-

nal amine with a pK_avalue of 7.7,^[21]or compound 2l with an

N-(2-hydroxybenzyl)piperidine side chain in which the basicity

of the tertiary amine is expected to be lower due to hydrogen

bond interactions with the 2-hydroxy group of benzyl, and

compounds 2j,k,m that have no basic side chains, showed

lower G4 stabilisation (DT_m: 2–158C). This confirms that elec-

trostatic or hydrogen bond interactions between the ligand

protonated side chain and the G4 structures are important fac-

tors for complex stabilisation, as suggested by molecular mod-

elling studies,^[14c]although this could not be confirmed by

NMR analysis of a c-Myc promoter G4 and the 11-((2-dimethyla-

mino)ethylamino)quindoline complex.^[12c]The melting data for

T-loop with compounds 2 showed, in most cases, a similar out-

line in comparison with G4 data (Table 1), and no significant

selectivity for G4 structures would be expected.

Cryptolepine derivatives 2c–m showed DT_mvalues ranging

from 2 to 258C at 2 mm for Hsp90A, F21T, and c-Kit2, in most

cases increased relative to parent compound 2a and are simi-

lar to those reported elsewhere for human telomeric G4.^[14c,15]

Compounds with aliphatic amine side chains (2c–i), which at

physiological pH are expected to be double protonated,^[21]

showed higher stabilisation effects with G4. From those, 2e–g

with propyl side chains showed the highest degree of stabilisa-

tion, with DT_mvalues ranging from 17 to 258C. These results

show for the first time in the case of C11-monosubstituted

indolo[3,2-b]quinolines, that there is an optimal NH-alkylamine

side chain length for G4 stabilisation, in accordance with previ-

ous observations for di- and trisubstituted acridines^[22]and

naphthalene diimides,^[23]that propylamine side chains are su-

perior to ethyl or butylamine. Changing the distal part of the

propylamine side chain from a primary to a secondary or terti-

The indolo[3,2-b]quinolines with two (3a–c, 4a–c) and three

(3d, 4d) alkylamine side chains, which are expected to be di-

and triprotonated at physiological pH, respectively, were found

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&

4

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

binding affinity for G4 structures than the corresponding

N10,C11-disubstituted compound. Remarkably, competition di-

alysis experiments also showed that despite the lower binding

affinity for G4 structures, disubstituted quindolines bind more

selectively to different G4 than monosubstituted com-

pounds,^[14b]a pattern also observed in this study. N5,N10-disub-

stituted indoloquinolines 3a–b show a G4 thermal stability

trend effect of Hsp90A>F21T>c-Kit2. Also, DT_mvalues for T-

loop suggest a weak binding affinity for duplex DNA struc-

tures. Finally, to our surprise, the introduction of an electron-

withdrawing chlorine atom at the 3-position of the indolo[3,2-

b]quinoline aromatic core seems to decrease the G4 binding

capacity of the N5,N10-disubstituted compound, as 3c has

a lower DT_m(58C for Hsp90A and 28C for F21T) than the ethyl

side chain analogue 3a (DT_mvalues of 88C for Hsp90A and 48C

for F21T).

Table 1. Melting temperatures of DNA structures stabilised by 2a–m at

2 mm.

Compd R¹

DT_m[8C]^[a]

Hsp90A^[b]F21T^[c]c-Kit2^[d]T-loop^[e]

2a

2c

H

10.1

18.7

7.0

5.0

3.2

7.8

13.9

12.3

2d

2e

12.1

19.9

11.4

21.3

8.1

9.3

18.6

18.9

2 f

25.0

21.9

16.8

19.3

2g

2h

23.8

22.2

21.3

18.4

17.0

13.5

19.9

13.9

2i

2j

2k

18.0

9.6

15.8

5.5

11.4

14.5

4.7

The introduction of an alkylamine carboxylate side chain at

the 7-position of the indolo[3,2-b]quinoline aromatic core, as in

the case of trisubstituted compounds 3d and 4d, clearly im-

proves binding to G4 DNA structures (DT_mbetween 16 and

78C for both compounds), maintaining the inter-G4 thermal

stability trend effect Hsp90A>F21T>c-Kit2>T-loop (see Fig-

ure 2b for compound 4d) and also suggesting weak binding

affinity for duplex DNA structures. However, in the case of tri-

substituted indoloquinolines 3d (C7,N5,N10-substituted) and

4d (C7,N10,O11-substituted) the relative positions of the three

side chains has no effect on G4 stabilisation, as observed for

disubstituted indoloquinolines 3a–c (N5,N10-substituted) and

4a–c (N10,O11-substituted). Overall, the results suggest that

the electron-withdrawing carboxylate group at the 7-position

can increase p–p stacking interactions between the aromatic

indoloquinoline ring and the terminal G-quartet of G4s by de-

creasing the electron density of the former.

2.78

9.8

5.9

3.5

7.5

2.1

3.5

2l

15.6

6.8

10.9

3.0

6.4

1.5

2m

[a] DT_mÆ0.3. [b] T_m=59.08C. [c] T_m=57.98C. [d] T_m=58.48C. [e] T_m=

53.28C.

to be less effective stabilisers of G4 structures (DT_m<178C)

than monosubstituted N5-methylindoloquinolines 2c–i. This

observation suggests that the positive charge at the aromatic

moiety, which can be ascribed to decreased electron density

and consequently stronger interactions with the electron-rich

p system of a G-quartet, is more important for G4 stabilisation

than positive charges at side chain termini, as in the case of 3

and 4. This is consistent with the type of interactions observed

between 11-((2-dimethylamino)ethylamino)quindoline and the

c-Myc G4 complex studied by NMR.^[12c]These studies showed

that the principal forces stabilising the complex are p–p stack-

ing between the aromatic indoloquinoline ring of the ligand

and two guanines of the terminal G-quartet, whereas no par-

ticular interactions were observed between the side chain and

the G4 structure, although this could be due to the short side

chain length of the ligand.

The trisubstituted derivatives 3d and 4d were also evaluat-

ed for their stabilisation effects on the G4-forming promoter

sequence of KRas^[24](Table 2). Compound 4d is as good a stabil-

iser of the G4 sequence KRas21R (DT_m=138C) as of F21T. The

G4 stabilising effect of compound 4d was further evaluated

with c-Kit1,^[7a]HIF-1a,^[25]and Hsp90C (comprising the full G4 se-

quence in the Hsp90 promoter region)^[5](Figure 2). The results

showed a similar stabilisation of the oncogenic c-Kit1 G4

(DT_m=78C at 2 mm ligand concentration), compared with c-

Kit2 and lower stabilisation of HIF-1a (DT_m=38C) relative to

the remaining G4 structures. Despite 4d being a good stabilis-

er of Hsp90A G4 (DT_m=178C), no significant stabilisation of

the oncogene promoter Hsp90 full sequence G4 was observed

(DT_m=38C). These results further suggest the capacity of tri-

substituted indolo[3,2-b]quinolines to discriminate between G4

structures.

Generally, N5,N10-disubstituted indolo[3,2-b]quinolines 3a–

c showed greater G4 stabilisation, and in particular 3b with

a propylamine side chain (DT_mbetween 12 and 78C), than the

counterparts 4a–c (DT_mof 4b between 5 and 38C), with both

side chains close to each other (at N10 and C11) and directed

to the same side of the indolo[3,2-b]quinoline aromatic core.

Boddupally et al.^[14b]recently reported that the introduction of

a second alkylamine chain at N10 of an 11-piperazinylquindo-

line increases the thermal stability of a c-Myc G4 complex, al-

though subsequent competition dialysis experiments showed

that monosubstituted 11-piperazinylquindoline has higher

To access the selectivity of these ligands for G4 over duplex

DNA, suggested by DT_mdata, competitive FRET experiment

with F21T G4 and 4d was performed in the presence of a non-

fluorescent duplex DNA competitor (26ds)^[26](Figure 3). The re-

sults indicate that stabilisation of the telomeric G4 is only af-

fected in the presence of 50-fold excess of competitor 26ds,

showing that this compound is at least 10-fold more selective

for G4 than for duplex DNA.

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&

5

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

Table 2. DT_mvalues of DNA structures stabilised by compounds 3 and 4 at 2 mm and DG values of the optimised propeller-like parallel (1KF1), antiparallel

basket-type (143D), hybrid-type1 (2HY9), and hybrid-type2 (2JPZ) G4 structures with compounds 3 and 4.

Compd

R¹

R²

R³

DT_m[8C]^[a]

DG [kcalmol^À1

]

Hsp90A^[b]

7.6

F21T^[c]

3.8

c-Kit2^[d]

T-loop^[e]

0.5

KRas21R^[f]

ND

1KF1

143D

2HY9

2JPZ

3a

3b

3c

H

Cl

H

2.3

7.5

1.4

À42.7

À45.4

À65.1

À71.6

À67.6

H

11.8

5.0

7.2

1.7

3.8

0.5

ND

À43.9

À43.3

À40.6

À35.7

À67.9

À68.7

CH₃

3d

H

15.9

10.6

8.7

2.2

7.4

À44.7

À38.1

À72.6

À83.6

4a

4b

4c

H

Cl

H

3.2

5.2

3.2

1.2

2.5

1.4

1.1

2.4

0.9

1.8

2.2

0.3

ND

À40.5

À42.5

À40.4

À36.0

À43.0

À34.2

À53.8

À68.5

À60.2

À63.9

À64.0

À62.2

H

CH₃

4d^[g]

H

16.7

12.0

7.4

2.9

12.9

À50.0

À40.2

À73.5

À77.6

[a] DT_mÆ0.3. [b] T_m=59.08C. [c] T_m=57.98C. [d] T_m=58.48C. [e] T_m=53.28C. [f] T_m=51.58C; ND: not determined. [g] Also evaluated with: c-Kit1 (T_m=

60.28C), DT_m=7.1Æ0.38C; Hsp90C (T_m=78.58C), DT_m=2.8Æ0.18C; HIF-1a (T_m=79.58C), DT_m=2.5Æ0.28C.

In physiologically relevant potassium ion concentrations,

human telomeric DNA can adopt antiparallel, hybrid, or parallel

arrangements.^[28]However, due to water depletion, the antipar-

allel and hybrid arrangements under crowded conditions are

converted into propeller-type parallel strands.^[29]To generate

our models, four experimentally determined structures of G4

human telomere were chosen as starting points: 1) the 22-mer

unimolecular propeller-like parallel with three stacked G-quar-

tets (PDB ID 1KF1);^[30]2) the 22-mer unimolecular antiparallel

basket-type with three stacked G-quartets, which are connect-

ed by two lateral loops and a central diagonal loop (PDB ID

143D);^[31]and 3) and 4) the 26-mer unimolecular hybrid-type1

and -type2 mixed parallel/antiparallel-G-stranded G-quadru-

plexes (PDB IDs 2HY9 and 2JPZ, respectively).^[32]

To achieve the 21-mer common sequence 5’-GGG TTA GGG

TTA GGG TTA GGG-3’ in all topologies, in PDB 1KF1 and 143D,

the nucleotide 5’-A was removed, nucleotides 5’-AAA and AA-

Figure 3. FRET melting competition assay data for the F21T–4d complex in

the absence (0 mm) and presence of increasing concentrations of duplex

(26ds) DNA competitor; the concentration of F21T was 0.2 mm.

3’ were removed in PDB 2HY9, while nucleotides 5’-TTA and

TT-3’ were removed in PDB 2JPZ. The presence of the K⁺ions

are very important for the stability of G4; the absence of coor-

dination cations modifies the structures, causing evident defor-

mation or disruption.^[33]Therefore, K⁺ions were added be-

Molecular modelling studies

tween the guanine stacks to the first single G4 of the ensem-

To understand the structural basis of G4 binding in this class of

ligands, molecular modelling of several compounds with avail-

able G-quadruplexes structures was performed. Based on FRET

melting data, which showed the highest stabilisation of G4

structures in the promoter of Hsp90A and human telomeres by

indolo[3,2-b]quinolines, and in the absence of Hsp90A G4 struc-

ture, several intramolecular G4 structures with different topolo-

gies were selected.^[27]

bles of NMR structures (PDB IDs 143D, 2HY9, and 2JPZ) and

used as starting points. Refinement of the G4 structures was

performed by molecular dynamics (MD) simulations in water

with the appropriate number of counterions for neutrality of

the system and sampling the NpT ensemble (p=1 bar, T=

298 K) for 10 ns. During the complete course of MD simulation

the integrity of the G4 persisted, with the RMSD of all atoms

of the structure stabilising around 1.9, 2.5, 2.6, and 2.5 ꢂ for

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&

6

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

parallel, antiparallel, hybrid-type1, and -type2, respectively

(Supporting Information). The stacked G-quartets are very rigid

and are the most stable segment of the structure due to

strong hydrogen bonds and stacking interactions. However,

without impact on the structure of the central G-tetrad core,

the residue’s root-mean-square fluctuation (RMSF) showed that

loop nucleotides exhibited relatively larger flexibility, (Support-

ing Information), as recently described by Haider and co-work-

ers for the human propeller telomeric DNA quadruplex.^[34]To

evaluate indolo[3,2-b]quinoline complex formation with the

optimised structures, several docking studies were performed

using the MD-optimised G4 structures with different topolo-

gies as starting conformations. The efficiency of the docking

methodology was validated with the crystal structure of the

human telomeric bimolecular parallel-stranded G4, co-crystal-

lised with the acridine BRACO-19 (PDB ID 3CE5),^[12b]using rigid-

and induced-fit docking protocols and several refinement scor-

ing functions included in MOE v.2012.10.^[35]The final docking

pose obtained with rigid docking protocol and alpha HB scor-

ing function was in good agreement with the crystallographic

pose in the 3’ end of structure, showing an RMSD of 1.8 ꢂ for

all heavy atoms, demonstrating the effectiveness of the chosen

methodology (Supporting Information).

due to steric hindrance caused by the TTA loops. The complex

is also stabilised by asymmetric p–p stacking interactions, and

in some cases the N10 side chain in 3a–c also extends over

the G-quartet, as in the parallel topology, interacting with gua-

nine or thymine bases. The N5 side chains of 3a–c point

toward the outside of the G-quartet core, able to interact with

the TTA loops and/or phosphate backbone and probably justi-

fying the higher stabilisation revealed by the DG values in

comparison with propeller-type parallel topology. On the other

hand, docking results for derivatives 4a–c in the hybrid-type

structures showed both N10 and O11 side chains pointing

toward the outside of the G-quartet core, and probably justify-

ing the highest DG values when compared with 3a–c. General-

ly, unlike in the propeller-type parallel topology, the edge

placement of 3 and 4 found in the hybrid-type structures

allows interactions between the side chains and the TTA loops

and/or backbone, in accord with the higher stabilisation prop-

erties predicted by the DG values (Table 2) in the latest struc-

tures.

As already described, trisubstituted derivatives 3d and 4d

showed higher stabilising properties (DT_mand DG) than disub-

stituted derivatives; therefore, the influence of a third aminoal-

kylcarboxylate side chain at position 7 was also modelled.

Compounds 3d and 4d were docked against the four different

G4 topological structures. Docking results in the propeller-like

parallel structure showed derivatives 3d and 4d interacting

with the G-quadruplex structure as already described for 3a–

c and 4a–c, yet with the third side chain at position 7 acting

as a new anchoring point. In both cases, the C7 side chain is

directed outside the G-quartet core, allowing interactions with

the external loops and/or backbone of the G4 structure and

justifying the increased experimental and predicted binding

properties (DT_mand DG), when compared with the disubstitut-

ed derivatives. In the hybrid-type G4 structures, 3d and 4d

showed the same asymmetric p–p stacking interactions of the

indolo[3,2-b]quinoline nucleus with two guanine bases, on the

edge of the G-quartet. For instance, Figure 4 shows the top-

ranked docking pose for 4d in the hybrid-type2 human telo-

mere G4 structure.

Docking studies of the indolo[3,2-b]quinoline derivatives 3

and 4 were performed with the different topologies of the 21-

mer human telomere G4. The final docking poses showed 3

and 4 interacting with the propeller-like parallel, hybrid-type1

and -type2 models in a stacking mode on the top of the 3’ G-

quartet, while in the anti-parallel basket-type model, the li-

gands behave like groove binders due to the steric hindrance

caused by the TTA loop (Supporting Information). The free

energy of complex formation (DG) was estimated from the

final docking poses after energy refinement (Table 2). The high-

est stabilisers of the human telomere F21T G4 structure, 3d

and 4d, showed to be the most favourable predicted stabilis-

ers, with the lowest DG values, as reported elsewhere for quin-

dolines and fluorenone derivatives.^[14c,36]Coefficients of deter-

mination (r²) of 0.74, 0.65, and 0.80 (Supporting Information)

were found between the F21T FRET data and DG for 3 and 4

with the propeller-type parallel, hydrid-type1 and -type2

models, respectively. Analysis of DG values showed the most

stable binding modes converging to the hydrid-type2 human

telomere G4 topology, possibly indicative of the most favoura-

ble conformation in our experimental assays.

The complex is stabilised by asymmetric p–p stacking inter-

actions between the indole ring and guanine DG4. Additional-

ly, hydrogen bond interactions between the O11 terminal ni-

trogen and the phosphate backbone, and between the proton-

ated C7 terminal nitrogen and N3 of the guanine DG12 purine

ring improve complex stability. As described for the propeller-

type parallel structure, in the hybrid-type structures, the third

side chain in C7 adds an additional point of contact with G4,

improving the stability of the complex. As such, this additional

binding feature described for 3d and 4d is probably correlated

with the generalised increase in binding predicted by the DG

values, and is in accord with the DT_mvalues found for the tri-

substituted derivatives for all the G4 structures evaluated.

Our studies showed that for the propeller-like parallel struc-

ture (1KF1), the N5,N10-indolo[3,2-b]quinolines 3a–c stack at

the centre of the G-quartet. The complex is stabilised by asym-

metrical p–p stacking interactions, and, in some cases, with

the side chains extended over the G-quartet face, stabilised via

a p–cation interaction between protonated terminal nitrogen

and a guanine base. On the other hand, N5,O11 derivatives

4a–c also showed stacking interactions with the guanine

bases of propeller-like parallel G4, but due to a less favourable

conformation, both side chains point toward the outside of

the G4, leading to decreased stabilisation, as shown by higher

DG values (Table 2). In the hybrid-type1 and -type2 topologies,

derivatives 3a–c and 4a–c stack on the edge of the G-quartet

Effect on HCT116 cancer cells

To examine the effect of G4-selective ligands 3d and 4d on

cancer cells, the antiproliferative activity of these compounds

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&

7

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

Figure 4. Top view of the predicted pose, upon docking with MOE, of com-

pound 4d and the optimised 21-mer hybrid-type2 G4 structure (PDB ID

2JPZ). The G4 structure is represented as a wireframe and the ligand as

sticks. Non-relevant bases and nonpolar hydrogen atoms were removed for

clarity; grey lines represent p interactions, and black dashed lines indicate

hydrogen bond interactions.

was evaluated by means of MTS assays with the HCT116 colon

cancer cell line after drug exposure for 96 h. HCT116 cells were

selected, as these cells overexpress Hsp90 and KRas proteins.

We also evaluated the selectivity of 3d and 4d for cancer cells

by assessing the effect of these compounds on the viability of

primary rat hepatocytes. Figure 5 clearly shows that com-

pounds 3d and 4d are ~100-fold more selective for HCT116

cells (IC₅₀=4.5Æ0.6 and 5.7Æ0.7 mm for 3d and 4d, respec-

tively) than for normal hepatocytes (IC₅₀>100 mm), a pattern

also observed for the standard cytotoxic drug used for colon

cancer treatment, 5-fluorouracil (5-FU).

Figure 5. Dose–response curves for a) HCT116 cancer cells and b) primary rat

hepatocytes exposure to 3d, 4d, and 5-fluorouracil (5-FU).

However, 5-FU shows a distinct anti-proliferative profile rela-

tive to that of G4 ligands 3d and 4d over HCT116 human

colon cancer cells. Although 5-FU seems to be more effective

against HCT116 cancer cells at lower concentrations, its maxi-

mum effect is only of 75% cancer cell growth inhibition, even

at 100 mm, whereas 3d and 4d are more effective, almost com-

pletely inhibiting the growth of HTC116 colon cancer cells at

low-micromolar concentrations (IC₉₀=7.6Æ0.9 and 11.0Æ

0.9 mm for 3d and 4d, respectively). The antiproliferative activi-

ties of 3d and 4d compare with those of other disubstituted

indolo[3,2-b]quinolines.^[14b]We also found that HCT116 cell line

exposure to 3d and 4d for 72 h decreased the expression

levels of Hsp90 and KRas proteins (Figure 6).

Conclusions

Figure 6. Exposure to compounds 3d and 4d decreases Hsp90 and KRas

protein expression in HCT116 cells. Total proteins were extracted from

HCT116 cells following exposure to 3d, 4d, or vehicle (DMSO) for 72 h and

used to evaluate Hsp90 and KRas steady-state levels by Western blot. Results

are expressed as the mean ÆSEM; *p<0.05 relative to DMSO control.

Targeting G4 DNA structures at telomeres and promoter re-

gions of oncogenes may be an effective strategy to control

cancer cell proliferation, but to date, knowledge of the chemi-

cal requirements for both G4 structures and ligands that

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&

8

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

would lead to high and selective stabilisation of G4 DNA struc-

tures is still sparse. To further explore and understand the G4-

structure stabilising activity relationship of indolo[3,2-b]quino-

lines, several mono- (2), di- (3a–c, 4a–c), and trialkylamine (3d

and 4d) derivatives were designed, synthesised, and evaluated

for thermal stabilisation of three different G4 DNA sequences

and a hairpin duplex DNA sequence. Further molecular model-

ling studies were performed to better understand the basis of

their binding.

From all indoloquinoline derivatives studied, the trisubstitut-

ed compounds 3d and 4d, bearing a 7-(aminoalkyl)carboxylate

side chain are the most promising, showing high G4 thermal

stabilisation (DT_mbetween 17 and 88C) with an inter-G4 DT_m

trend of Hsp90A>KRas21RꢀF21T>c-Kit2, 10-fold selectivity

for G4 over duplex DNA, and 100-fold selectivity for the

HCT116 cancer cell line (IC₅₀and IC₉₀<10 mm) over primary rat

hepatocytes. Additionally, 3d and 4d decreased Hsp90 and

KRas expression levels in HCT116 cancer cells. These results are

relevant to the future design and synthesis of new 7-carboxyl-

ate indoloquinoline derivatives that aim to improve both G4

stabilisation and selectivity between various G4s.

Overall, monosubstituted N5-methyl derivatives, bearing

a full cationic charge at the quinoline nitrogen (N5) atom and

a protonatable amine group at the side chain terminus (com-

pounds 2c–i) are better G4 DNA stabilisers, but seem to be

nonselective for G4 DNA structures. Further analysis of FRET

melting data obtained for indolo[3,2-b]quinolines with one

(2a–m), two (3a–c, 4a–c), or three (3d and 4d) alkylamine

side chains and molecular modelling results for 3a–d and 4a–

d has enabled the following SAR conclusions to be made:

1) Optimal linker length between the distal basic centre and

the indolo[3,2-b]quinoline moiety seems to be that of propyl

linear chains, a pattern also observed with the disubstituted

derivatives (3a–b, 4a–c) and which consistently showed lower

DG values. 2) Aryl substitution at the 11-position has no effect

on G4 thermal stabilisation; compounds 2j–k show DT_mvalues

similar to that of parent compound 2a. 3) The basicity of side

chains seems to be positively correlated with thermal G4 stabi-

lisation up to pK_a~8.5, consistent with increased concentration

of the protonated species and clearly suggesting that cationic

groups are required for active participation in G4 complex sta-

bilisation. Molecular modelling studies have shown that pro-

tonated side chains extend over G-quartets and can form hy-

drogen bonds and electrostatic interactions with the guanine

bases as well as with TTA loops and phosphate backbones.

4) The introduction of a third side chain, an alkylaminoalkylcar-

boxylate (3d and 4d), increases G4 stabilisation relative to dis-

ubstituted derivatives, leading to selectivity for G4 over duplex

DNA structures and suggests an inter-G4 selectivity trend of

Hsp90A>F21TꢀKRas21R>c-Kit2. 5) The relative position of

side chains plays an important role in the thermal stability of

G4 DNA structures. N5,N10-disubstituted indolo[3,2-b]quino-

lines 3a–c are modest G4 stabilisers, whereas N10,O11-disub-

stituted analogues 4a–c showed very poor G4 DNA stabilising

effects. 6) A chlorine atom at the 3-position of the indolo[3,2-

b]quinoline (compound 3c) is detrimental to G4 DNA complex

stabilisation. Analysis of free binding energies obtained from

molecular modelling studies clearly corroborates the FRET ex-

perimental data. Molecular modelling studies showed that the

relative position of the side chain plays an important role in

complex stabilisation, in which the N5,N10-disubstituted

indolo[3,2-b]quinolines 3a–c can participate in more binding

interactions with the G-quartet face, due to a more favourable

conformation. Additionally, the introduction of a third side

chain at position 7 of the aromatic nucleus (compounds 3d

and 4d) seems to favour the formation of additional hydrogen

bonds and electrostatic interactions with the G-quartet and/or

TTA loops, and as such, increases complex stabilisation.

Experimental Section

Chemistry

Chemicals were purchased from Sigma–Aldrich Chemical Co. Ltd.

(Spain) and were used without further purification. All compounds

were characterised by NMR spectroscopy, recorded on a Bruker

Avance 400 spectrometer at 400 (¹H) and 100 MHz (¹³C), with sol-

vent as internal reference. Chemical shifts (d) are expressed in

ppm. Signal splitting patterns are described as singlet (s), doublet

(d), triplet (t), quartet (q), quintet (quint) and multiplet (m), or com-

binations thereof. Coupling constants (J) are given in Hz. The

purity of compounds submitted to biological tests was ꢁ95% in

all cases as determined from elemental analysis (C, H, N), carried

out by the Elemental Analysis Unit, University of Santiago de Com-

postela (Spain), on a LECO model CHNS-932 elemental analyser.

MS data were recorded using a Micromass Quattro Micro API,

Waters, and were obtained by direct infusion on “Full Scan” mode

(m/z 60–800); sample ionisation was made in positive and negative

electrospray ionisation modes (ESI⁺and ESI^À). Melting points (mp)

were determined with a Bock-Monoscop M instrument. Reactions

were monitored by thin-layer chromatography (TLC) using coated

silica gel plates (Merck, aluminium sheets, silica gel 60 F₂₅₄) and alu-

minium oxide matrix plates (Sigma–Aldrich, PET support, F₂₅₄).

Preparative TLC was performed in aluminium oxide 60 F₂₅₄(VWR,

200ꢃ200 mm glass support, F₂₅₄). Fluorescent-conjugated oligonu-

cleotides were purchased from Eurogentec Ltd. (UK). Fluorescence

measurements were made on a DNA Engine Opticon instrument

(MJ Research).

General procedure A. 2-(2-Bromoacetamido)benzoic acid (6a): A

solution of 2-aminobenzoic acid (10.0 g, 72.9 mmol) in DMF

(30 mL) and 1,4-dioxane (30 mL) was cooled to 08C. Bromoacetyl

bromide (8.0 mL, 91.7 mmol, 1.25 equiv) was added dropwise over

a 20 min period, keeping the internal temperature between 0 and

58C. After the addition was complete, the ice bath was removed,

and stirring was continued overnight at room temperature. The re-

action mixture was added to H₂O (300 mL), and the light-yellow

precipitate which formed was filtered, washed with H₂O, until neu-

tral pH, and then dried to give 18.1 g (96%) of 6a as a white solid;

1

mp: 162–1658C. H NMR (400 MHz, DMSO): d=11.23 (s, NH), 8.06

(d, J=8.4 Hz, 1H), 7.62 (dd, J=7.9, 1.4 Hz, 1H), 7.24 (dd, J=8.4,

7.3 Hz, 1H), 6.82 (dd, J=7.9, 7.3 Hz, 1H), 3.87 ppm (s, 2H); ¹³C NMR

(100 MHz, DMSO): d=169.20, 165.04, 139.92, 134.06, 131.11,

123.45, 119.99, 117.06, 30.65 ppm.

2-(2-Bromoacetamido)-4-chlorobenzoic acid (6b): Reaction of 2-

amino-4-chlorobenzoic acid (10 g, 58.2 mmol) and bromoacetyl

bromide (14.7 g, 6.3 mL, 72.8 mmol, 1.25 equiv) according to gen-

eral procedure A gave 17.1 g (99%) of 6b as a light-yellow solid;

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&

9

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

1

mp: 166–1688C. H NMR (400 MHz, DMSO): d=11.73 (s, NH), 8.54

1H), 7.82 (d, J=8.8, 2H), 7.56 (t, J=7.1, 1H), 7.14 (t, J=7.2, 1H),

6.68 (d, J=8.8, 2H), 4.28 (q, J=7.1, 2H), 4.02 (s, 2H), 1.35 ppm (t,

J=7.1, 3H); ¹³C NMR (100 MHz, MeOD): d=170.93, 169.12, 167.27,

152.16, 140.22, 133.66, 131.10, 130.93, 122.84, 119.83, 118.64,

116.41, 111.50, 60.04, 48.45, 13.30 ppm.

(d, J=2.1 Hz, 1H), 7.98 (d, J=8.6 Hz, 1H), 7.24 (dd, J=8.6, 2.1 Hz,

1H), 4.28 ppm (s, 2H); ¹³C NMR (100 MHz, DMSO): d=169.03,

165.92, 141.54, 138.95, 133.26, 123.73, 119.61, 115.91, 30.98 ppm.

General procedure B. Synthesis of 2-[2-(phenylamino)acetami-

do]benzoic acid (7a): A solution of 6a (15.0 g, 58.1 mmol) and ani-

line (19.0 mL, 208.3 mmol, 3.5 equiv) in DMF (30 mL) was heated at

1208C for 18 h. After cooling, the reaction mixture was poured into

ice-water (500 mL), aqueous 5% KOH was added to solubilise the

solid product and adjust the solution to pH 10. Then the mixture

was washed with CH₂Cl₂(3ꢃ200 mL). The combined CH₂Cl₂ex-

tracts were set aside, and the aqueous layer was acidified to pH 3

with 5% HBr. The precipitate which formed was collected, washed

with H₂O, and then dried, yielding 11.0 g (70%) of 7a as a white

General procedure C: 5H-indolo[3,2-b]quinolin-11(10H)-one (8a):

A mixture 7a (6.0 g, 21.5 mmol) and polyphosphoric acid (PPA,

160 g) was heated with mechanical stirring at 1308C for 2 h. The

reaction mixture was added to ice-water (500 mL), neutralised with

saturated KOH solution, and then extracted with EtOAc (3ꢃ

500 mL). The extract was washed with H₂O, dried with brine and

anhydrous Na₂SO₄, and then the solvent was removed at reduced

pressure. The product was recrystallised from EtOAc with Et₂O/

hexane (9:1) to give 3.47 g (67%) of 8a as a light-green solid; mp:

1

solid; mp: 194–1978C. H NMR (400 MHz, DMSO): d=11.96 (s, NH),

>3008C. H NMR (400 MHz, DMSO): d=12.81 (s, 1H), 11.69 (s, 1H),

8.61 (d, J=8.7 Hz, 1H), 7.83 (d, J=7.9 Hz, 1H), 7.47 (dd, J=8.7,

8.0 Hz, 1H), 7.02 (dd, J=8.0, 7.9 Hz, 1H), 6.98 (d, J=7.3 Hz, 2H),

6.50 (d, J=7.7 Hz, 1H), 6.48 (dd, J=7.7, 7.3 Hz, 2H), 6.39 (s, NH),

3.73 ppm (s, 2H); ¹³C NMR (100 MHz, CD₃OD): d=170.82, 169.01,

148.08, 140.52, 134.03, 131.08, 128.92, 122.60, 119.32, 116.98,

116.12, 112.36, 48.89 ppm.

8.35 (d, J=8.0 Hz, 1H), 8.28 (d, J=8.0 Hz, 1H), 7.81 (d, J=8.4 Hz,

1H), 7.68 (dd, J=8.4, 7.5 Hz, 1H), 7.50 (d, J=8.2, 1H), 7.49 (dd, J=

8.2, 7.4 Hz, 1H), 7.28 (dd, J=8.0, 7.5 Hz, 1H), 7.20 ppm (dd, J=8.0,

7.4 Hz, 1H); ¹³C NMR (100 MHz, DMSO): d=167.77, 139.54, 139.05,

130.99, 129.44, 127.84, 125.57, 123.48, 123.24, 121.63, 120.87,

119.25, 118.28, 116.38, 112.98 ppm.

4-Chloro-2-{2-[(4-methylphenyl)amino]acetamido}benzoic acid

(7b): A solution of 6b (4.0 g, 13.6 mmol) and 4-methylaniline

(2.93 g, 27.3 mmol, 2 equiv) in DMF (15 mL) was heated at 1208C

for 48 h and treated according to general procedure B to give

3-Chloro-7-methyl-5H-indolo[3,2-b]quinolin-11(10H)-one

(8b):

Reaction of 7b (1.5 g, 4.7 mmol) with PPA (45 g) according to gen-

eral procedure C gave 0.697 g (52%) of 8b, as a light-green solid;

1

mp: >3008C. H NMR (400 MHz, DMSO): d=12.53 (s, NH), 11.67 (s,

1

3.68 g (83%) of 7b as a light-yellow solid; mp: 187–1898C. H NMR

NH), 8.33 (d, J=8.7 Hz, 1H), 7.92 (s, 1H), 7.72 (d, J=1.7 Hz, 1H),

7.42 (d, J=8.5 Hz, 1H), 7.32 (d, J=8.5 Hz, 1H), 7.28 (dd, J=8.7,

1.7 Hz, 1H), 2.47 ppm (s, 3H); ¹³C NMR (100 MHz, DMSO): d=

167.23, 139.98, 137.57, 135.55, 129.89, 129.05, 128.31, 127.92,

123.86, 121.88, 121.19, 120.33, 117.07, 116.24, 112.95, 21.61 ppm.

(400 MHz, DMSO): d=12.13 (s, 1H), 8.83 (d, J=2.0 Hz, 1H), 7.94 (d,

J=8.5 Hz, 1H), 7.21 (dd, J=8.5, 2.0 Hz, 1H), 6.91 (d, J=8.3 Hz, 2H),

6.49 (d, J=8.3 Hz, 2H), 3.81 (s, 2H), 2.14 ppm (s, 3H); ¹³C NMR

(100 MHz, DMSO): d=172.15, 168.74, 146.19, 142.00, 139.02,

133.32, 129.88, 126.05, 123.04, 119.14, 115.14, 112.96, 49.67,

20.53 ppm.

11-Oxo-10,11-dihydro-5H-indolo[3,2-b]quinoline-7-carboxylic

acid (8c): Reaction of 7c (2.0 g, 6.1 mmol) with PPA (60 g) accord-

ing to general procedure C gave 0.312 g (19%) of 8c as a light-

brown solid. Reaction of 7e (2.0 g, 5.8 mmol) with PPA (60 g) ac-

cording to general procedure C gave 1.01 g (62%) of 8c as a light-

2-{2-[(4-(Trifluoromethyl)phenyl)amino]acetamide}benzoic acid

(7c): A solution of 6a (3.0 g, 11.6 mmol) and 4-trifluoromethylani-

line (2.34 g, 1.82 mL, 14.5 mmol, 1.25 equiv) in CH₃CN (50 mL) was

held at reflux for 18 h. After this period the solution was cooled in

an ice bath to precipitate. The solid was collected by filtration,

washed with ice-CH₃CN and dried to give 2.07 g (53%) of 7c as

1

brown solid; mp: >3008C. H NMR (400 MHz, DMSO): d=12.64 (s,

1H), 12.10 (s, 1H), 8.99 (s, 1H), 8.36 (d, J=7.9 Hz, 1H), 8.04 (dd, J=

8.7, 1.6 Hz, 1H), 7.72 (m, 2H), 7.56 (d, J=8.7 Hz, 1H), 7.32 ppm (dd,

J=7.9, 7.1 Hz, 2H); ¹³C NMR (100 MHz, DMSO): d=168.18, 167.92,

141.01, 139.60, 131.44, 130.07, 128.55, 125.64, 124.60, 124.20,

123.54, 121.80, 121.29, 118.37, 116.18, 112.77 ppm; MS ESI⁺(m/z)

calcd for C₁₆H₁₀N₂O₃: 278.06, found: 278.96.

1

a white solid; mp: 218–2218C. H NMR (400 MHz, DMSO): d=11.94

(s, 1H), 8.70 (d, J=8.4 Hz, 1H), 7.94 (dd, J=7.9, 1.5 Hz, 1H), 7.61

(dd, J=7.9, 7.7 Hz, 1H), 7.43 (d, J=8.6 Hz, 2H), 7.42 (s, 1H) 7.15

(dd, J=7.7, 7.4 Hz, 1H), 6.73 (d, J=8.6 Hz, 2H), 3.95 ppm (s, 2H);

¹³C NMR (100 MHz, DMSO): d=170.39, 169.68, 151.61, 140.92,

134.70, 131.61, 126.81, 123.29, 119.86, 116.38, 112.41, 48.53 ppm;

¹⁹F NMR (376 MHz, DMSO): d=À59.09 ppm (s).

2-Hydroxy-3-((4-nitrophenyl)amino)quinolin-4(1H)-one

(7id):

Compound 7d (2.0 g, 6.34 mmol) was reacted in PPA (60 g) for 2 h

at 1308C according with general procedure C to give a yellow solid

1

2-{2-[(4-(Nitro)phenyl)amino]acetamide}benzoic acid (7d): A solu-

tion of 6a (3.0 g, 11.6 mmol) and 4-nitroaniline (3.21 g, 23.2 mmol,

2 equiv) in DMF (30 mL) was held at reflux for 66 h according to

general procedure B to give 2.22 g (60%) of 7d as a yellow solid;

(0.701 g), to yield 7id at 37%; mp: 225–2298C. H NMR (400 MHz,

DMSO): d=12.86 (s, 1H), 10.54 (s, 1H), 8.70 (d, J=8.4 Hz, 1H), 8.07

(dd, J=7.9, 1.5 Hz, 1H), 7.68 (dd, J=7.9, 7.7 Hz, 1H), 7.52 (d, J=

8.8 Hz, 2H), 7.25 (dd, J=8.4, 7.7 Hz, 1H), 6.56 ppm (d, J=8.8 Hz,

2H); ¹³C NMR (100 MHz, DMSO): d=169.40, 159.29, 157.46, 146.41,

139.76, 134.46, 131.93, 126.93, 124.23, 122.35, 119.90, 118.25,

114.06 ppm; MS ESI⁺(m/z) calcd for C₁₅H₁₁N₃O₄: 297.07, found:

298.10.

1

mp: 231–2358C. H NMR (400 MHz, DMSO): d=11.81 (s, 1H), 8.66

(d, J=8.1 Hz, 1H), 8.04 (d, J=9.3 Hz, 2H), 7.95 (dd, J=8.0, 1.5 Hz,

1H), 7.61 (dd, J=9.3, 7.7 Hz, 1H), 7.16 (dd, J=8.1, 7.7 Hz, 1H), 6.74

(d, J=9.3 Hz, 2H), 4.10 ppm (s, 2H); ¹³C NMR (100 MHz, DMSO):

d=169.79, 169.37, 154.54, 140.85, 137.47, 134.73, 131.63, 126.55,

123.42, 119.98, 116.52, 112.07, 48.06 ppm.

General procedure D. 5,10-Bis(2-(pyrrolidin-1-yl)ethyl)-5H-

indolo[3,2-b]quinolin-11(10H)-one (3a) and 11-(2-(pyrrolidin-1-

yl)ethoxy)-10-(2-(pyrrolidin-1-yl)ethyl)-10H-indolo[3,2-b]quino-

lone (4a): A solution of 8a (70 mg, 0.29 mmol), K₂CO₃(619.8 mg,

4.35 mmol, 15 equiv), in dried acetone (20 mL) and 1-(2-chloroe-

thyl)pyrrolidine hydrochloride (203.4 mg, 1.20 mmol, 4 equiv) was

held at reflux overnight. After this period, the solvent was removed

at reduced pressure, and the remaining solid suspended in H₂O

2-{2-[(4-(Ethoxycarbonyl)phenyl)amino]acetamide}benzoic acid

(7e): A solution of 6a (5.0 g, 19.4 mmol) and ethyl 4-aminoben-

zoate (9.6 g, 58.1 mmol, 3.5 equiv) in DMF (30 mL) was held at

reflux for 96 h according to general procedure B to give 2.70 g

(40%) of 7e as

a

light-yellow solid; mp: >3308C. ¹H NMR

(400 MHz, MeOD): d=8.69 (d, J=8.4, 1H), 8.03 (dd, J=7.9, 1.5,

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&10

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

(30 mL). The aqueous solution was extracted with CH₂Cl₂(3ꢃ

30 mL), and the combined organic extracts were washed with H₂O,

dried with brine and anhydrous Na₂SO₄, and decreased to small

volume. The crude mixture was purified by preparative TLC using

CH₂Cl₂/MeOH (9:1) as eluent to afford compounds 3a (27.3 mg,

21%) and 4a (45.5 mg, 35%). The compounds were recrystallised

as hydrochloride salts with HCl in Et₂O after NMR characterisation

and obtained as light-yellow solids. 3a: mp: 227–2308C; ¹H NMR

(400 MHz, CDCl₃): d=8.68 (d, J=7.8 Hz, 1H), 8.24 (d, J=8.4 Hz,

1H), 7.72 (m, 2H), 7.69 (d, J=8.5 Hz, 1H), 7.57 (dd, J=8.5, 7.2 Hz,

1H), 7.37 (ddd, J=7.8 Hz, 1H), 7.28 (dd, J=8.4, 7.2 Hz, 1H), 5.13 (t,

J=7.7 Hz, 2H), 4.89 (t, J=7.8 Hz, 2H), 3.11 (m, 4H), 2.85 (m, 4H),

2.78 (m, 4H), 1.92 (m, 4H), 1.89 ppm (m, 4H); ¹³C NMR (100 MHz,

CDCl₃): d=169.05, 139.61, 139.51, 131.62, 130.81, 127.70, 126.78,

124.88, 122.50, 122.46, 121.18, 119.95, 115.10, 114.11, 110.65, 55.82,

54.75, 54.24, 53.44, 47.20, 43.21, 23.62, 23.58 ppm; Anal. calcd for

C₂₇H₃₂N₄O·2HCl·2H₂O: C 60.33, H 7.13, N 10.42, found: C 60.76, H

3-Chloro-5,10-bis(2-(diethylamino)ethyl)-7-methyl-5H-indolo[3,2-

b]quinolin-11(10H)-one (3c) and 2-((3-chloro-10-(2-(diethylami-

no)ethyl)-7-methyl-10H-indolo[3,2-b]quinolin-11-yl)oxy)-N,N-di-

ethylethanamine (4c): A solution of 8b (80 mg, 0.28 mmol), K₂CO₃

(580 mg, 4.20 mmol, 15 equiv), in dried acetone (20 mL) and 2-

chloro-N¹,N¹-diethylethanaminium chloride (192.7 mg, 1.12 mmol,

4 equiv) was reacted and purified according to general proced-

ure D to give 3c (26.3 mg, 19%) and 4c (47.2 mg, 35%) as light-

yellow solids. 3c: mp: 220–2228C; ¹H NMR (400 MHz, CDCl₃): d=

8.36 (d, J=8.7 Hz, 1H), 7.77 (s, 1H), 7.48 (d, J=1.6 Hz, 1H), 7.28 (d,

J=8.6 Hz, 1H), 7.17 (d, J=8.6 Hz, 1H), 7.05 (dd, J=8.7, 1.67 Hz,

1H), 4.74 (t, J=7.2 Hz, 2H), 4.54 (t, J=7.2 Hz, 2H), 2.81 (t, J=

7.2 Hz, 2H), 2.68 (t, J=7.2 Hz, 2H), 2.51 (q, J=7.1 Hz, 4H), 2.48 (q,

J=7.1 Hz, 4H), 2.32 (s, 3H), 0.90 (t, J=7.1 Hz, 6H), 0.85 ppm (t, J=

7.1 Hz, 6H); ¹³C NMR (100 MHz, CDCl₃): d=168.55, 140.29, 138.29,

137.60, 130.23, 129.42, 129.11, 128.44, 123.17, 122.96, 121.77,

121.51, 115.10, 114.16, 110.45, 53.13, 50.86, 47.77, 47.57, 43.06,

21.62, 12.05, 11.97 ppm; Anal. calcd for C₂₈H₃₇N₄O·2HCl·3.5H₂O: C

54.50, H 7.51, N 9.08, found: C 54.60, H 7.57, N 9.31. 4c: mp: 254–

1

7.53, N 10.32. 4a: mp: 212–2168C; H NMR (400 MHz, CDCl₃): d=

8.44 (d, J=7.6 Hz, 1H), 8.22 (d, J=8.0 Hz, 1H), 8.20 (d, J=8.0 Hz,

1H), 7.62–7.54 (m, 2H), 7.49 (d, J=8.0 Hz, 1H), 7.48 (dd, J=8.0,

8.0 Hz, 1H) 7.25 (ddd, 1H), 4.79 (t, J=7.5 Hz, 2H), 4.30 (t, J=5.8 Hz,

2H), 3.03 (t, J=5.8 Hz, 2H), 2.86 (t, J=7.5 Hz, 2H), 2.63 (m, 8H),

1.79 (m, 4H), 1.75 ppm (m, 4H); ¹³C NMR (100 MHz, CDCl₃): d=

148.59, 145.92, 144.58, 144.43, 129.95, 129.37, 126.78, 124.88,

124.62, 122.41, 122.14, 122.09, 121.10, 120.13, 109.24, 74.64, 55.81,

54.76, 54.41, 43.43, 23.60, 23.56 ppm; Anal. calcd for

C₂₇H₃₂N₄O·2HCl·3H₂O: C 58.37, H 7.26, N 10.09, found: C 58.79, H

7.74, N 9.72.

1

2568C; H NMR (400 MHz, CDCl₃): d=8.27 (d, J=9.0 Hz, 1H), 8.20

(s, 1H), 8.19 (d, J=2.1 Hz, 1H), 7.40–7.36 (m, 2H)„ 7.29 (d, J=

8.4 Hz, 1H), 4.54 (t, J=7.6 Hz, 2H), 4.16 (t, J=6.1 Hz, 2H), 2.94 (t,

J=6.1 Hz, 2H), 2.66 (t, J=7.6 Hz, 2H), 2.58 (q, J=7.1 Hz, 4H), 2.50

(q, J=7.1 Hz, 4H), 2.47 (s, 3H), 1.01 (t, J=7.1 Hz, 6H), 0.88 ppm (t,

J=7.1 Hz, 6H); ¹³C NMR (100 MHz, CDCl₃): d=149.39, 145.91,

144.55, 143.27, 132.18, 131.37, 129.57, 127.97, 125.41, 125.21,

122.84, 122.32, 121.98, 120.65, 109.02, 74.82, 53.04, 51.71, 47.71,

47.63, 43.60, 21.16, 12.11, 11.91 ppm; Anal. calcd for

C₂₈H₃₇N₄O·2HCl: C 60.70, H 7.10, N 10.11, found: C 60.48, H 6.97, N

9.87.

5,10-Bis(3-(piperidin-1-yl)propyl)-5H-indolo[3,2-b]quinolin-

11(10H)-one (3b) and 11-(3-(piperidin-1-yl)propoxy)-10-(3-(piper-

idin-1-yl)propyl)-10H-indolo[3,2-b]quinoline (4b): A solution of

8a (70 mg, 0.29 mmol), K₂CO₃(619.8 mg, 4.35 mmol, 15 equiv), in

dried acetone (20 mL) and 1-(3-chloropropyl)piperidine hydrochlo-

ride (237.0 mg, 1.20 mmol, 4 equiv) was reacted and purified ac-

cording to general procedure D to give 3b (45.2 mg, 31%) and 4b

2-(Diethylamino)ethyl

5,10-bis(2-(diethylamino)ethyl)-11-oxo-

10,11-dihydro-5H-indolo[3,2-b]quinoline-7-carboxylate (3d) and

2-(diethylamino)ethyl 11-(2-(diethylamino)ethoxy)-10-(2-(diethy-

lamino)ethyl)-10H-indolo[3,2-b]quinoline-7-carboxylate (4d):

A

solution of 8c (70 mg, 0.23 mmol), K₂CO₃(476 mg, 3.45 mmol,

15 equiv), in dried acetone (20 mL) and 2-chloro-N¹,N¹-diethyletha-

naminium chloride (159.3 mg, 0.92 mmol, 4 equiv) was reacted and

purified according to general procedure D to give 3d (36.2 mg,

31%) and 4d (39.6 mg, 34%) as light-yellow solids. 3d: mp: 175–

1788C; ¹H NMR (400 MHz, CDCl₃): d=9.00 (s, 1H), 8.65 (d, J=

8.0 Hz, 1H), 8.19 (d, J=8.9 Hz, 1H), 7.72 (m, 2H), 7.60 (d, J=8.9 Hz,

1H), 7.35 (m, 1H), 5.00 (t, J=6.9 Hz, 2H), 4.87 (t, J=6.7 Hz, 2H),

4.46 (t, J=6.5 Hz, 2H), 3.03 (t, J=6.7 Hz, 2H), 2.94 (t, J=6.9 Hz,

2H), 2.91 (t, J=6.5 Hz, 2H), 2.74 (q, J=7.1 Hz, 4H), 2.68 (m, 8H),

1.10 (m, 12H), 1.04 ppm (t, J=7.2 Hz, 6H); ¹³C NMR (100 MHz,

CDCl₃): d=169.05, 166.79, 141.65, 139.69, 131.74, 131.34, 128.08,

126.75, 125.84, 125.09, 123.16, 121.55, 121.43, 114.88, 114.43,

110.37, 63.18, 53.12, 51.16, 50.81, 47.79, 47.60, 47.49, 47.30, 43.36,

12.01, 11.89, 11.78 ppm; Anal. calcd for C₃₄H₄₉N₅O₃·3HCl·3.5H₂O: C

54.58, H 7.95, N 9.36, found: C 54.67, H 8.01, N 9.27. 4d: mp: 168–

1

(56.3 mg, 38%) as light-yellow solids. 3b: mp: 258–2618C; H NMR

(400 MHz, CDCl₃): d=8.68 (d, J=8.1 Hz, 1H), 8.24 (d, J=8.3 Hz,

1H), 7.80 (d, J=8.6 Hz, 1H), 7.68 (dd, J=8.6, 7.3 Hz, 1H), 7.65 (d,

J=8.4 Hz, 1H), 7.52 (dd, J=8.4, 7.4 Hz, 1H), 7.34 (dd, J=8.1,

7.3 Hz, 1H), 7.22 (dd, J=8.3, 7.4 Hz, 1H), 4.96 (t, J=7.1 Hz, 2H),

4.81 (t, J=7.2 Hz, 2H), 2.53 (t, J=6.3 Hz, 2H), 2.47 (m, 4H), 2.35 (m,

6H), 2.18 (m, 2H), 2.12 (quint, J=7.1 Hz, 2H), 1.69 (quint, J=

5.5 Hz, 4H), 1.58 (quint, J=5.6 Hz, 4H), 1.51 (m, 2H), 1.42 ppm (m,

2H); ¹³C NMR (100 MHz, CDCl₃): d=169.03, 139.79, 139.53, 131.21,

130.66, 127.05, 126.73, 124.91, 122.73, 122.70, 120.84, 119.33,

115.08, 114.37, 110.91, 55.97, 55.91, 55.07, 54.54, 54.46, 53.83,

46.25, 42.78, 28.19, 26.20, 25.98, 24.48 ppm; Anal. calcd for

C₃₁H₄₀N₄O·2HCl·5H₂O: C 57.49, H 8.09, N 8.65, found: C 57.45, H

8.04, N 8.43. 4b: mp: 241–2458C; ¹H NMR (400 MHz, CDCl₃): d=

8.53 (d, J=7.7 Hz, 1H), 8.32 (d, J=7.9 Hz, 1H), 8.31 (d, J=7.8 Hz,

1H), 7.66–7.58 (m, 2H), 7.53 (dd, J=8.4, 7.8 Hz, 1H), 7.51 (d, J=

7.8 Hz, 1H), 7.30 (dd, J=7.7, 7.2 Hz, 1H), 4.57 (t, J=6.9 Hz, 2H),

4.22 (t, J=6.2 Hz, 2H), 2.65 (t, J=7.2 Hz, 2H), 2.47 (m, 4H), 2.26 (m,

6H), 2.20 (quint, J=, 6.2 Hz, 2H), 2.02 (quint, J=6.9 Hz, 2H), 1.64

(quint, J=5.6 Hz, 4H), 1.54 (quint, J=5.6 Hz, 4H), 1.49 (m, 2H),

1.40 ppm (m, 2H); ¹³C NMR (100 MHz, CDCl₃): d=148.58, 145.76,

144.82, 144.53, 129.54, 129.29, 126.55, 124.74, 124.59, 122.34,

122.18, 121.92, 121.25, 119.68, 109.36, 74.13, 56.12, 55.35, 54.68,

54.53, 42.75, 27.58, 26.91, 26.09, 25.97, 24.52, 24.45 ppm; Anal.

calcd for C₃₁H₄₀N₄O·3HCl·3.5H₂O: C 56.36, H 8.01, N 8.55, found: C

56.66, H 7.67, N 8.53.

1

1728C; H NMR (400 MHz, CDCl₃): d=9.24 (d, J=1.5 Hz, 1H), 8.39

(dd, J=8.4, 0.9 Hz, 1H), 8.35 (dd, J=8.6, 1.7 Hz, 1H), 8.33 (d, J=

7.5 Hz, 1H), 7.71 (ddd, J=8.4, 6.7, 1.4 Hz, 1H), 7.59 (ddd, J=8.4,

6.8, 1.4 Hz, 1H), 7.52 (d, J=8.7 Hz, 1H), 4.73 (t, J=6.8 Hz, 2H), 4.49

(t, J=6.6 Hz, 2H), 4.31 (t, J=6.2 Hz, 2H), 3.08 (t, J=6.2 Hz, 2H),

2.96 (t, J=6.6 Hz, 2H), 2.83 t, J=6.8 Hz, 2H), 2.70 (q, J=7.2 Hz,

8H), 2.57 (q, J=7.1 Hz, 4H), 1.13 (t, J=7.2 Hz, 6H), 1.11 (t, J=

7.2 Hz, 6H), 0.93 ppm (t, J=7.1 Hz, 6H); ¹³C NMR (100 MHz, CDCl₃):

d=166.84, 148.13, 147.36, 146.30, 145.15, 131.15, 129.46, 127.18,

125.30, 125.08, 124.39, 122.36, 122.16, 121.84, 121.40, 108.95,

74.79, 62.86, 52.97, 51.96, 51.12, 47.77, 47.65, 47.56, 43.88, 12.02,

11.90, 11.83 ppm; Anal. calcd for C₃₄H₄₉N₅O₃·3HCl·2H₂O: C 56.62, H

7.83, N 9.71, found: C 56.44, H 7.99, N 9.73.

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&

11

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

atoms). Finally, the dynamics of the system was studied by sam-

pling the NpT ensemble (T=298 K, p=1 bar) for 10 ns. In all MD

runs, the particle mesh Ewald (PME) formalism was applied in the

long-range electrostatic interactions. The short-range electrostatic

cutoff was 1.2 nm, and the same was applied for the van der Waals

(vdW) interactions. All bonds were constrained with the Lincs algo-

rithm. Nosꢄ–Hoover and Parrinello–Rahman were applied to con-

trol the temperature and isotropic pressure (t_T=0.2, t_p=5.0 ps,

and b=4.5ꢃ10^À5bar^À1). Energy and pressure corrections for the

vdW cutoff were also applied. The structures obtained at the end

of these production runs were used in the docking studies and

were visually compared with the starting PDB structures by using

the alignment application tool present in MOE.

FRET melting assays

The ability of indolo[3,2-b]quinolines to stabilise G-quadruplex

DNA sequences was investigated by using a fluorescence reso-

nance energy transfer (FRET) assay. The labelled oligonucleotides

(Eurogentec Ltd., UK) contained the donor fluorophore 6-carboxy-

fluorescein (FAM) and the acceptor fluorophore 6-carboxytetrame-

thylrhodamine (TAMRA). Sequences were as follows: F21T (5’-

[FAM]-GGG TTA GGG TAG GGT TAG GG-[TAMRA]-3’), KRas21R (5’-

[FAM]-AGG GCG GTG TGG GAA GAG GGA-[TAMRA]-3’), Hsp90A (5’-

[FAM]-GGG CCA AAG GGA AGG GGT GGG-[TAMRA]-3’), Hsp90C (5’-

[FAM]-AGG GCG GGC CAA AGG GAA GGG GTG GGC-[TAMRA]-3’), c-

Kit1 (5’-[FAM]-AGA GGG AGG GCG CTG GGA GGA GGG GCT-

[TAMRA]-3’), c-Kit2 (5’-[FAM]-CCC GGG CGG GCG CGA GGG AGG

GGA GG-[TAMRA]-3’), HIF-1a (5’-[FAM]-GCG CGG GAG GGG AGA

GGG GGC GGG AGC GCG-[TAMRA]-3’), T-loop (5’-[FAM]-TAT AGC

TAT ATT TTT TTA TAG CTA TA-[TAMRA]-3’). Each oligonucleotide was

initially diluted to 100 mm in nuclease-free water (not DEPC-treat-

ed), purchased from Ambion Applied Biosystems (UK). Stock solu-

tions of 20 mm and subsequent dilutions were obtained in FRET

buffer (60 mm KCl, potassium cacodylate, pH 7.4). The FRET probe

sequences were diluted from stock to the correct concentration

(0.4 mm) and then annealed by heating at 858C for 10 min, fol-

lowed by slow cooling to room temperature in the heating block.

Test compounds were prepared as 10 mm DMSO stock solutions

and diluted to 1 mm using 1 mm HCl in HPLC-grade water. The

rest of the dilutions were performed using FRET buffer. Annealed

DNA (50 mL) and test compound solution (50 mL) were distributed

across 96-well RT-PCR plates (BioRad; MJ Research, Waltham, MA,

USA). Relevant controls were also performed to check for interfer-

ence with the assay. Fluorescence readings were made with excita-

tion at l 450–495 nm and detection at l 515–545 nm, taken at in-

tervals of 0.58C in the range 30–1008C, with a constant tempera-

ture being maintained for 30 s prior to each reading to ensure

a stable value. The melting of the F21T G-quadruplex (0.2 mm) was

also monitored in a competition assay in the presence of 0.4, 2.0,

10, and 25 mm non-fluorescent double-stranded competitor 26ds

DNA (5’-CAA TCG GAT CGA ATT CGA TCC GAT TG-3’) at a ligand

concentration of 1 mm. Experiments were performed in triplicate.

Final analysis of the data was carried out with GraphPad Prism

v.5.0 (GraphPad Software Inc., La Jolla, CA, USA). The advanced

curve-fitting function in GraphPad Prism was used for calculation

of DT_mvalues and associated standard deviations.

Docking: The crystal structure of the parallel-stranded F21T quad-

ruplex (PDB ID 3CE5)^[12b]was prepared with the MOE v.2012.10 soft-

ware package by removing water and adding the missing hydro-

gen atoms and protonation states with Protonate 3D application

tool, at 300 K, pH 7, salt concentration of 0.1m at the GB/VI electro-

static formalism. The co-crystallised ligand found in PDB 3CE5 was

re-docked with the G4 structure using rigid docking protocol, the

triangle matcher placement, the alpha HB free-energy scoring func-

tion, and retaining 100 poses. Final poses were subjected to a final

refinement in the receptor pocket with the AMBER99 force field, at

an RMS gradient threshold of 0.01 and a second rescoring with

alpha HB free-energy scoring function. The best-scoring values

were selected as final poses. Indolo[3,2-b]quinoline ligand struc-

tures were also constructed in MOE and energy minimised with

the MMFF94x force field at an RMS gradient threshold of 0.05. The

refined parallel, antiparallel basket, and hybrid G4 structures (1KF1,

143D, and 2JPZ, respectively) obtained from the MD simulations

were used in docking with the indolo[3,2-b]quinolone ligands as

described above. The G4–ligand interactions were visualised with

the MOE v.2012.10 ligand interactions application tool.

Viability assays

HCT116 human colon carcinoma cells were grown in McCoy’s 5A

medium supplemented with 10% fetal bovine serum (FBS) and 1%

antibiotic/antimycotic (Invitrogen, Grand Island, NY, USA) and main-

tained at 378C in a humidified atmosphere of 5% CO₂. Cells were

seeded in 96-well plates at 5000 cells per well. Primary rat hepato-

cytes were isolated from male Sprague–Dawley rats (100–150 g) by

collagenase perfusion as described previously.^[39]Cell viability was

determined by trypan blue exclusion and was typically 80–85%.

After isolation, hepatocytes were resuspended in William’s E medi-

um and plated on Primaria 96-well culture plates (BD Biosciences,

San Jose, CA, USA) at 5000 cells per well. The cells were maintained

at 378C in a humidified atmosphere of 5% CO₂for 4 h to allow at-

tachment. Plates were then washed with medium to remove dead

cells and incubated in William’s E medium containing 10% heat-in-

activated FBS. Test compounds 3d and 4d, and the positive con-

trol 5-fluorouracil (5-FU, Sigma), a common cytotoxic agent used in

colon cancer treatment,^[40]were dissolved in DMSO; 24 h after cell

plating, media was removed and replaced with fresh media con-

taining 1–100 mm test compounds and 5-FU, or vehicle (DMSO)

control. After compound exposure for 96 h, cell viability was evalu-

ated using CellTiter 96 AQueous Non-Radioactive Cell Proliferation

Assay (Promega, Madison, WI, USA), using 3-(4,5-dimethylthiazol-2-

yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium,

inner salt (MTS) as previously described.^[41]Cell viability data were

expressed as mean ÆSEM or mean ÆSD from at least three inde-

pendent experiments. IC₅₀and IC₉₀values were determined using

GaphPad Prism v.5.00 (GraphPad Software).

Molecular modelling studies

Molecular dynamics: G4 structures obtained from the RSCB Pro-

tein Data Bank (PDB IDs 1KF1, 143D, 2JPZ) were prepared in the

MOE v.2012.10 software package (Chemical Computing Group)^[35]

by removing water molecules, adding the missing hydrogen

atoms, and assigning protonation states with the Protonate 3D ap-

plication tool by using the GB/VI formalism at 300 K, pH 7, and salt

concentration of 0.1m. Addition of the two K⁺ions to the G4

structures resolved by NMR were made by using the solvate appli-

cation tool present in MOE v.2012.10 which, when necessary, were

placed manually in the central cavity of the G4 structure. The sys-

tems were then subjected to MD simulations using the GROMACS

simulation package 4.5.5^[37]and applying the AMBER03^[38]force

field. The G4 structures prepared in MOE were inserted in a cubic

box, with no less than 1 nm between the G4 and the simulation

box edge. The system was solvated and neutralised by adding the

required K⁺ions. After energy minimisation of the MD box with

the steepest descent method, a 100 ps NVT equilibration run fol-

lowed at 298 K (spatially restraining the oligonucleotide’s heavy

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&12

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

guez, J. Dash, G. J. McKenzie, A. R. Venkitaraman, S. Balasubramanian, J.

Am. Chem. Soc. 2011 , 133 , 265 8 –2 663.

Total protein extraction and immunoblotting

HCT116 cells were seeded in 35 mm plates at 150000 cells per

well. Test compounds 3d and 4d were dissolved in DMSO and

added to the cells 24 h after plating, at IC₅₀, 10, and 20 mm. Vehicle

(DMSO) control was also included. After compound exposure for

72 h, cells were collected and processed for total protein extrac-

tion. Briefly, samples were homogenised in ice-cold 1:1 solution of

buffer A [10 mm Tris·HCl pH 7.6, 5 mm MgCl₂, 1.5 mm KOAc, 2 mm

dithiothreitol (DTT), and Halt Protease and Phosphatase inhibitor

cocktail, EDTA-free (#78445, Thermo Scientific)] and buffer 2ꢃ

(10 mm Tris·HCl pH 7.6, 1% Nonidet-P40, and Halt Protease and

Phosphatase inhibitor cocktail), by vigorous vortexing and incubat-

ed on ice for 30 min. Samples were then sonicated (two cycles of

15 s sonication and 30 s ice incubation, using a compact ultrasonic

device with amplitude adjusted to 80% and pulse to 90%; model

UP100H, Hielscher Ultrasonics GmbH, Teltow (Germany); 100 W, ul-

trasonic frequency: 30 kHz) and centrifuged at 10000 g for 10 min

at 48C. The clear supernatants containing the total protein extracts

were transferred to a fresh tube and stored at À808C. Protein con-

centrations were determined using the BioRad protein assay kit ac-

cording to the manufacturer’s instructions.^[40b,42]Steady-state levels

of Hsp90 and KRas proteins were determined by immunoblot anal-

ysis. Briefly, 25 mg of total protein extracts were separated by 10%

SDS-PAGE. After electrophoretic transfer onto nitrocellulose mem-

branes, immunoblots were incubated with 15% H₂O₂for 15 min at

room temperature. After blocking with 5% milk solution, the blots

were incubated overnight at 48C with primary mouse monoclonal

antibody reactive to Hsp90 or to KRas (#13119 and #sc-30, respec-

tively; Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA). Finally,

membranes were incubated with secondary anti-mouse sera conju-

gated with horseradish peroxidase (BioRad) for 3 h at room tem-

perature. The membranes were processed for protein detection

using Super Signal substrate (Pierce, Rockford, IL, USA). b-Actin

(#A-5441, Sigma–Aldrich) was used as a loading control. Protein

levels were expressed as mean ÆSEM from at least three inde-

pendent experiments.

[8] S. Cogoi, L. E. Xodo, Nucleic Acids Res. 2006 , 34 , 253 6 –2 549.

[9] M. Gunaratnam, M. de La Fuente, S. M. Hampel, A. K. Todd, A. P. Reszka,

A. Schatzlein, S. Neidle, Bioorg. Med. Chem. 2011 , 19 , 715 1 –7 157.

[10] a) Y. Xu, Chem. Soc. Rev. 2011 , 40 , 271 9 –2 740; b) J. H. Tan, L. Q. Gu, J. Y.

Wu, Mini-Rev. Med. Chem. 2008, 8, 1163–1178.

[11] D. Z. Yang, K. Okamoto, Future Med. Chem. 2010 , 2 , 61 9 –6 46.

[12] a) N. H. Campbell, M. Patel, A. B. Tofa, R. Ghosh, G. N. Parkinson, S.

Neidle, Biochemistry 2009 , 48 , 167 5 –1 680; b) N. H. Campbell, G. N. Par-

kinson, A. P. Reszka, S. Neidle, J. Am. Chem. Soc. 2008 , 130 , 672 2 –6 724;

c) J. Dai, M. Carver, L. H. Hurley, D. Yang, J. Am. Chem. Soc. 2011 , 133 ,

1767 3 –1 7680.

[13] a) J. Lavrado, R. Moreira, A. Paulo, Curr. Med. Chem. 2010, 17, 2348–

2370; b) J. Lavrado, R. Moreira, A. Paulo, in Heterocyclic Targets in Ad-

vanced Organic Synthesis (Eds.: M. D. Carreiras, J. Marco-Contelles), Re-

search Signpost, Kerala (India), 2011, pp. 109–124.; c) K. Gçrlitzer, J.

Weber, Arch. Pharm. 1981 , 314 , 85 2 –8 61; d) D. E. Bierer, L. G. Dubenko,

P. Zhang, Q. Lu, P. A. Imbach, A. W. Garofalo, P. W. Phuan, D. M. Fort, J.

Litvak, R. E. Gerber, B. Sloan, J. Luo, R. Cooper, G. M. Reaven, J. Med.

Chem. 1998 , 41 , 275 4 –2 764.

[14] a) T. M. Ou, Y. J. Lu, C. Zhang, Z. S. Huang, X. D. Wang, J. H. Tan, Y. Chen,

D. L. Ma, K. Y. Wong, J. C. Tang, A. S. Chan, L. Q. Gu, J. Med. Chem. 2007 ,

50 , 146 5 –1 474; b) P. V. Boddupally, S. Hahn, C. Beman, B. De, T. A.

Brooks, V. Gokhale, L. H. Hurley, J. Med. Chem. 2012 , 55 , 607 6 –6 086;

c) Y. J. Lu, T. M. Ou, J. H. Tan, J. Q. Hou, W. Y. Shao, D. Peng, N. Sun, X. D.

Wang, W. B. Wu, X. Z. Bu, Z. S. Huang, D. L. Ma, K. Y. Wong, L. Q. Gu, J.

Med. Chem. 2008 , 51 , 638 1 –6 392; d) Z. S. Huang, T. M. Ou, J. Lin, Y. J.

Lu, J. Q. Hou, J. H. Tan, S. H. Chen, Z. Li, Y. P. Li, D. Li, L. Q. Gu, J. Med.

Chem. 2011, 54, 5671–5679.

[15] J. Lavrado, A. P. Reszka, R. Moreira, S. Neidle, A. Paulo, Bioorg. Med.

Chem. Lett. 2010 , 20 , 704 2 –7 045.

[16] B. Guyen, C. M. Schultes, P. Hazel, J. Mann, S. Neidle, Org. Biomol. Chem.

2004 , 2 , 98 1 –9 88.

[17] Y. Kobayashi, I. Kumadaki, Acc. Chem. Res. 1978 , 11 , 19 7 –2 04.

[18] J. Lavrado, G. G. Cabal, M. Prudencio, M. M. Mota, J. Gut, P. J. Rosenthal,

C. Diaz, R. C. Guedes, D. J. Dos Santos, E. Bichenkova, K. T. Douglas, R.

Moreira, A. Paulo, J. Med. Chem. 2011 , 54 , 73 4 –7 50.

[19] J. Lavrado, K. Gani, P. A. Nobre, S. A. Santos, P. Figueiredo, D. Lopes, V.

Rosario, J. Gut, P. J. Rosenthal, R. Moreira, A. Paulo, Bioorg. Med. Chem.

Lett. 2010 , 20 , 563 4 –5 637.

[20] C. M. Schultes, W. Guyen, J. Cuesta, S. Neidle, Bioorg. Med. Chem. Lett.

2004 , 14 , 434 7 –4 351.

[21] J. Lavrado, A. Paulo, E. Bichenkova, K. T. Douglas, R. Moreira, Magn.

Reson. Chem. 2012 , 50 , 21 6 –2 20.

Acknowledgements

The authors thank the FundaÅ¼o para a Ciencia e Tecnologia

(FCT), Portugal, for financial support through project grants PEst-

OE/SAU/UI4013/2011, EXPL/QEQ-MED/0502/2012, and PTDC/SAU-

ORG/119842/2010. J.L. also acknowledges the FCT for a postdoc-

toral grant: SFRH/BPD/72903/2010. Work in London was support-

ed by Cancer Research UK.

[22] a) M. J. Moore, C. M. Schultes, J. Cuesta, F. Cuenca, M. Gunaratnam, F. A.

Tanious, W. D. Wilson, S. Neidle, J. Med. Chem. 2006 , 49 , 58 2 –5 99; b) F.

Cuenca, M. J. Moore, K. Johnson, B. Guyen, A. De Cian, S. Neidle, Bioorg.

Med. Chem. Lett. 2009 , 19 , 510 9 –5 113.

[23] G. W. Collie, R. Promontorio, S. M. Hampel, M. Micco, S. Neidle, G. N. Par-

kinson, J. Am. Chem. Soc. 2012 , 134 , 272 3 –2 731.

[24] M. Paramasivam, S. Cogoi, L. E. Xodo, Chem. Commun. 2011 , 47 , 496 5 –

4967.

[25] R. De Armond, S. Wood, D. Y. Sun, L. H. Hurley, S. W. Ebbinghaus, Bio-

chemistry 2005 , 44 , 1634 1 –1 6350.

Keywords: alkaloids · cancer · cryptolepine · G-quadruplexes ·

indoloquinolines

[26] A. De Cian, L. Guittat, K. Shin-ya, J. F. Riou, J. L. Mergny, Nucleic Acids

Symp. Ser. 2005 , 49 , 23 5 –2 36.

[27] A. N. Lane, J. B. Chaires, R. D. Gray, J. O. Trent, Nucleic Acids Res. 2008 ,

36 , 548 2 –5 515.

[28] D. Renciuk, I. Kejnovska, P. Skolakova, K. Bednarova, J. Motlova, M. Vor-

lickova, Nucleic Acids Res. 2009 , 37 , 662 5 –6 634.

[29] B. Heddi, A. T. Phan, J. Am. Chem. Soc. 2011 , 133 , 982 4 –9 833.

[30] G. N. Parkinson, M. P. Lee, S. Neidle, Nature 2002 , 417 , 87 6 –8 80.

[31] Y. Wang, D. J. Patel, Structure 1993 , 1 , 26 3 –2 82.

[32] J. Dai, C. Punchihewa, A. Ambrus, D. Chen, R. A. Jones, D. Yang, Nucleic

Acids Res. 2007 , 35 , 244 0 –2 450.

[1] S. Neidle, Curr. Opin. Struct. Biol. 2009 , 19 , 23 9 –2 50.

[2] A. K. Todd, M. Johnston, S. Neidle, Nucleic Acids Res. 2005 , 33 , 290 1 –

2907.

[3] a) S. Balasubramanian, L. H. Hurley, S. Neidle, Nat. Rev. Drug Discovery

2011 , 10 , 26 1 –2 75; b) S. Neidle, FEBS J. 2010 , 277 , 111 8 –1 125.

[4] L. Neckers, P. Workman, Clin. Cancer Res. 2012 , 18 , 6 4 –7 6.

[5] S. A. Ohnmacht, M. Micco, V. Petrucci, A. K. Todd, A. P. Reszka, M. Gunar-

atnam, M. A. Carvalho, M. Zloh, S. Neidle, Bioorg. Med. Chem. Lett. 2012 ,

22 , 593 0 –5 935.

[33] M. Cavallari, A. Calzolari, A. Garbesi, R. Di Felice, J. Phys. Chem. B 2006 ,

110 , 2633 7 –2 6348.

[34] B. Islam, M. Sgobba, C. Laughton, M. Orozco, J. Sponer, S. Neidle, S.

Haider, Nucleic Acids Res. 2013 , 41 , 272 3 –2 735.

[6] T. A. Brooks, L. H. Hurley, Genes Cancer 2010 , 1 , 64 1 –6 49.

[7] a) M. Gunaratnam, S. Swank, S. M. Haider, K. Galesa, A. P. Reszka, M. Bel-

tran, F. Cuenca, J. A. Fletcher, S. Neidle, J. Med. Chem. 2009 , 52 , 377 4 –

3783; b) K. I. E. McLuckie, Z. A. E. Waller, D. A. Sanders, D. Alves, R. Rodri-

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&13

&

ÞÞ

These are not the final page numbers!

CHEMMEDCHEM

FULL PAPERS

www.chemmedchem.org

[35] Molecular Operating Environment (MOE) v.2012.10, Chemical Comput-

ing Group Inc., 2012, Montreal, QC (Canada).

[36] S. Alcaro, A. Artese, J. N. Iley, S. Missailidis, F. Ortuso, L. Parrotta, R. Pas-

ceri, F. Paduano, C. Sissi, F. Trapasso, M. G. Vigorita, ChemMedChem

2010 , 5 , 57 5 –5 83.

Basis Dis. 2007 , 1772 , 4 0 –4 7; b) P. M. Borralho, B. T. Kren, R. E. Castro,

I. B. da Silva, C. J. Steer, C. M. Rodrigues, FEBS J. 2009 , 276 , 668 9 –6 700.

[41] T. F. Silva, L. M. Martins, M. F. Guedes da Silva, A. R. Fernandes, A. Silva,

P. M. Borralho, S. Santos, C. M. Rodrigues, A. J. Pombeiro, Dalton Trans.

2012 , 41 , 1288 8 –1 2897.

[37] a) H. J. C. Berendsen, D. van der Spoel, R. van Drunen, Comput. Phys.

Commun. 1995 , 91 , 4 3 –5 6; b) E. Lindahl, B. Hess, D. van der Spoel, J.

Mol. Model. 2001, 7, 306–317; c) D. van der Spoel, E. Lindahl, B. Hess, G.

Groenhof, A. E. Mark, H. J. C. Berendsen, J. Comput. Chem. 2005, 26,

1701–1718.

[42] a) P. M. Borralho, A. E. Simoes, S. E. Gomes, R. T. Lima, T. Carvalho, D. M.

Ferreira, M. H. Vasconcelos, R. E. Castro, C. M. Rodrigues, PLoS One 2011 ,

6 , e23787; b) A. E. Simꢅes, D. M. Pereira, J. D. Amaral, A. F. Nunes, S. E.

Gomes, P. M. Rodrigues, A. C. Lo, R. D’Hooge, C. J. Steer, S. N. Thibo-

deau, P. M. Borralho, C. M. Rodrigues, BMC Genomics 2013 , 14 , 181.

[38] L. J. Yang, C. H. Tan, M. J. Hsieh, J. M. Wang, Y. Duan, P. Cieplak, J. Cald-

well, P. A. Kollman, R. Luo, J. Phys. Chem. B 2006 , 110 , 1316 6 –1 3176.

[39] R. E. Castro, S. Sola, X. Ma, R. M. Ramalho, B. T. Kren, C. J. Steer, C. M. Ro-

drigues, J. Hepatol. 2005 , 42 , 89 7 –9 06.

Received: June 28, 2013

[40] a) P. M. Borralho, I. B. Moreira da Silva, M. M. Aranha, C. Albuquerque, C.

Nobre Leitao, C. J. Steer, C. M. Rodrigues, Biochim. Biophys. Acta Mol.

Published online on && &&, 0000

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&14

&

ÞÞ

These are not the final page numbers!

FULL PAPERS

Three is better: Novel trisubstituted

indolo[3,2-b]quinolines bearing a 7-(ami-

noalkyl)carboxylate side chain stand as

the most promising anticancer com-

pounds, showing good G4 thermal sta-

bilising effects, good selectivity for G4

over duplex DNA, and good selectivity

toward the HCT116 cancer cell line.

J. Lavrado, P. M. Borralho,

S. A. Ohnmacht, R. E. Castro,

C. M. P. Rodrigues, R. Moreira,

D. J. V. A. d. . Santos, S. Neidle, A. Paulo*

&& – &&

Synthesis, G-Quadruplex Stabilisation,

Docking Studies, and Effect on Cancer

Cells of Indolo[3,2-b]quinolines with

One, Two, or Three Basic Side Chains

ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemMedChem 0000, 00, 1 – 15

&15

&

ÞÞ

These are not the final page numbers!

Article Doi

Synthesis, G-quadruplex stabilisation, docking studies, and effect on cancer cells of indolo[3,2-b]quinolines with one, two, or three basic side chains

DOI: 10.1002/cmdc.201300288

Source and publish data:

Authors:

Article abstract of DOI:10.1002/cmdc.201300288

Full text of DOI:10.1002/cmdc.201300288

Products guided by the article

R&D Labs maybe for 195393-02-3

Relevant to this article

Hot Product