Determination of binding modes and binding constants for the complexes of 6H-pyrido[4,3-b]carbazole derivatives with DNA

Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Bioorganic & Medicinal Chemistry

journal homepage: www.elsevier.com/locate/bmc

Determination of binding modes and binding constants for the

complexes of 6H-pyrido[4,3-b]carbazole derivatives with DNA

a

b

Akihito Shimazu , Masashi Kawagoshi , Shoichi Takeda , Haruaki Kurasaki , Asako Kato , Nahoko Morii ,

a

a,

⇑

Norio Sakai , Takeo Konakahara

a

Department of Pure and Applied Chemistry, Faculty of Science and Technology, Tokyo University of Science (RIKADAI), Noda, Chiba 278-8510, Japan

Nanomaterials Laboratory, National Institute for Materials Science (NIMS), Tsukuba, Ibaraki 305-0003, Japan

b

a r t i c l e i n f o

a b s t r a c t

Article history:

The binding modes and binding constants for the complexes of forty types of pyridocarbazole derivatives

Received 14 October 2016

Revised 14 December 2016

Accepted 16 December 2016

Available online xxxx

1

–40 with double stranded DNAs (dsDNAs) were reported. The binding modes were determined by a

combination of a deﬂection spectroscopy and orientation of the corresponding molecule in the DNA-

based ﬁlm with chain alignment. All of the compounds exhibited the intercalation-binding mode. Its

binding constants K for the complexes, determined by quartz crystal microbalance (QCM), varied from

a

5

7

ꢁ1

1

.7 ꢀ 10 to 4.5 ꢀ 10 M according to the substituents on the pyridocarbazole framework and the

sequences of dsDNA. The binding constants K of pyridocarbazole derivatives possessing the 2-(

amino)alkyl group and 5-( -amino)alkylcarbamyl group were larger than those of the corresponding

-amino compounds exhibited strong GC base-pair preference in complex-

values decreased with the increasing NaCl concentration. It was clariﬁed by a molecular

-amino derivative was completely overlapped with

Keywords:

Ellipticine

a

x-

x

6H-Pyrido[4,3-b]carbazole

x

-ureido derivatives. These x

DNA

Intercalation

ation. The K

a

Binding constant

QCM

Deﬂection spectroscopy

Molecular modeling

GC base-pair preference

Salt effects

modeling that the framework of the 2-tethered x

the stacking GC base-pairs leading to the formation of the stable intercalative-complex, and that the

framework of the 5-tethered ureido derivative was half overlapped leading to the formation of the unsta-

ble complex. Furthermore, there were good linear relationships between lnK

rel of the complexes. Contrary to our expectation, there was no linear relationship between lnK

against Sarcoma-180, NIH3T3, and HeLa S-3 cell lines.

a

and the relative stabilities

S

a

and IC50

pH dependence

Cytotoxic activity

1

. Introduction

ellipticines and b-carbolines with DNA using deﬂection

8

spectroscopy. Ellipticine is a good intercalator and an antitumor

5

9

A great deal of interest has been generated by indole alkaloids

agent like b-carbolines. The unique structural features, limited

toxic side effects, and complete lack of hematological toxicity of

these compounds have prompted chemists to study the synthesis

1

(

5,11-dimethyl-6H-pyrido[4,3-b]carbazoles (ellipticines), b-carbo-

2

1f,3

lines ) because of their intercalation to DNA,

properties,

polymerase I transcription. Ellipticine was ﬁrst identiﬁed in 1959

antitumor

1

b–d

10

anti-HIV activities, and inhibition of RNA

of a number of their analogues in pharmacological evaluations.

1e

11a

The total synthesis of ellipticine was ﬁrst achieved in 1959.

as a compound in the leaves of a small tropical evergreen Ochrosia

elliptica Labill (Fig. 1). Ellipticine and its derivatives cause the selec-

Since then, many groups have synthesized ellipticine and its

4

11,12

derivatives.

Previously, we developed a novel and efﬁcient

tive inhibitionof p53 proteinphosphorylation in several humancan-

method for the synthesis of ellipticine via Suzuki-Miyaura cou-

5

a

5b

13

cer cell lines, induced apoptosis of cancer cell lines, and this is

correlated with their cytotoxic activity. These compounds also

uncouple mitochondrial oxidative phosphorylation,⁶thereby dis-

rupting the energy balance of cells. Further, these compounds are

pling.

Several simple structural modiﬁcations to ellipticine

derivatives have resulted in compounds that have shown increases

14

in cytotoxic activity. A low level of water solubility at physiolog-

ical pH, as well as systemic toxicity, has prevented the use of ellip-

ticine as a therapeutic agent. The introduction of a positive charge

in ellipticine improved water solubility and made a profound dif-

1f

known to react with DNA via a binding process, which may

7

account for their cytotoxicity, and also inhibits DNA topoiso-

1

a

15

merase-II activity.

We have reported the binding mode of

ference in its biological activity. For example, the quaternization

1

6

of the nitrogen at the N-2 of ellipticine (Fig. 1) and the endocyclic

1

7

⇑

Corresponding author.

E-mail address: konaka@rs.noda.tus.ac.jp (T. Konakahara).

nitrogen at the N-5 of 5-aza-ellipticine with a methyl group was

found to interact with DNA through intercalation, which resulted

http://dx.doi.org/10.1016/j.bmc.2016.12.031

0

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

2

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

Fig. 1. Chemical structures of ellipticine (1) and 2-methylellipticin-2-ium chloride (2).

in a high afﬁnity for binding with DNA. The binding mode of a drug

to DNA has been estimated mostly by sedimentation, ﬂow dichro-

ism, ﬂuorescence-detected ﬂow dichroism, viscosimetry, and crys-

tallography with synthetic DNA.⁸^,⁹^aThe clariﬁcation of the binding

mode of novel pharmaceuticals such as synthetic antibiotics is so

important that a method with simplicity, readiness, and repro-

ducibility is expected in the course of development.

9-Hydroxy-2-methylellipticin-2-ium chloride (6) was reportedly

2

0

synthesized and 17, 19, and 20 were analogously synthesized

in the present work. Firstly, synthetic strategy of 2-(2-((2-ami-

noethyl)amino)ethyl)-5-(methoxycarbonyl)-11-methyl-6H-pyrido

[4,3-b]carbazol-2-ium chloride (17) was illustrated in Scheme 1.

We began with the preparation of an imino tether, which was con-

tained in 17 (Method A in Scheme 1). The starting 2-((2-ami-

noethyl)amino)ethan-1-ol dihydrochloride (41) was allowed to

react with thionyl chloride for 2 h under reﬂux to produce the cor-

responding chloride 42 in a 90% yield. Initially, we selected Boc

group as a protecting group of the amino group. When 42 was trea-

ted with di-tert-butyl dicarbonate in the presence of triethylamine

in chloroform at 0 °C and then stirred at room temperature for 2 h,

the objective singly N-protected compound 43a was obtained in a

poor yield (15%). Therefore, we selected 2,5-dimethylpyrrolyl

group as an alternative protecting group. The compound 42 in

ethanol was successively treated with triethylamine, acetic acid

and 1,2-hexanedione at room temperature for 30 min leading to

Recently, we reported the syntheses of the N-methyl-N-nitro-

sourea conjugates of ellipticin-2-ium chloride, 9-hydroxyellip-

ticin-2-ium chloride, and 5-methoxycarbonyl-11-methyl-6H-

pyrido[4,3-b]carbazol-2-ium chloride derivatives, linked by an

1

8a

oxydiethylene unit and a pentamethylene unit at the 2-

or 5-

position of the framework.¹These compounds exhibited potent

cytotoxic activity against the human cervical cancer cell line

HeLa S-3 and also higher solubility in water than ellipticine itself.

8b

Furthermore,

a

glucose conjugate of 9-hydroxyellipticine

exhibited potent water solubility and cytotoxic activity against

HeLa S-3, and was easily hydrolyzed by carboxyesterase under

19

21

physiological conditions.

the formation of the desired compound 43b in an 84% yield.

The potential for these compounds¹⁸encouraged us to focus on

the determination of the binding modes and the binding constants

of various kinds of 11-methyl-6H-pyrido[4,3-b]carbazole deriva-

tives 1–40 with DNA (Fig. 2).

Next, in order to construct a pyridocarbazole ring, the com-

pound 45 was treated with 43a and 43b in DMF (Scheme 1), and

then with sodium methoxide in the presence of 3-ethoxycar-

1

8a

bonyl-1-methylpyridin-1-ium chloride (44)

leading to 2-

alkylpyridocarbazolium derivatives 46a,b in miserably poor

yields (only 4 and 10%, respectively), because of instability of the

2

. Results and discussion

2

-chloroethylamine moiety of 43a,b. These unexpectedly low

yields were fatal to the synthesis of the compound 17, although

the deprotection of the amino group of 46a,b proceeded

smoothly to give 17 in high to good yields (96 and 73%,

respectively). Therefore, we conducted the synthesis of 17

using another method (Method B in Scheme 1). That is to say;

2-(2-aminoethyl)-5-methoxycarbonyl-11-methyl-6H-pyrido[4,3-b]

2

.1. Synthesis of 11-methyl-6H-pyrido[4,3-b]carbazoles and their

quaternary derivatives 1–40, possessing an alkyl and alkylcarbamoyl

tether linked to an amino or ureido group

Synthesis of the tested compounds 1–40 in Fig. 2 (see also

1

8

Tables 2–5), except 6, 17, 19, and 20, were previously reported.

Fig. 2. 11-Methyl-6H-pyrido[4,3-b]carbazoles and their quaternary derivatives 1–40, possessing an alkyl and carbamoyl tether linked to an amino or ureido group.

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

3

Table 1

Effects of solvent and its content on the preparation of the DNA ﬁlms by the brush application and the natural seasoning methods.

Entry

Organic solvent

Content (%)^a

Orientation of DNA ﬁlm^b

Brush application

Natural seasoning

1

2

3

4

5

6

7

8

9

None

DMSO

MeOH

0

1

2

5

10

20

1

+

ꢁ

+

ꢁ

+

2

+

MeOH

MeOH/DMSO

20

19/1

18/2

1

0

1

+

ꢁ

a

Content of an organic solvent in an aqueous solution of the pyridocarbazole derivative.

b

+: Oriented, ꢁ: not oriented.

Table 2

Determination of binding modes for the DNA-complexes of ellipticine (1), 2-methylellipticin-2-ium chloride (2), 11-methyl-6H-pyrido[4,3-b]carbazoles (type A) 3 and 4; and

their 2-methyl quaternary salts (type B) 5 and 6 as drugs, by a deﬂection spectroscopy of the oriented drug-molecule in DNA-based ﬁlm prepared by a brush application method.^a

Entry

Compound

Type

X

R⁵

R⁶

R⁹

Binding mode

1

2

3

4

5

6

1

2

3

4

5

6

A

B

A

B

None

Cl

None

Cl

CH

COOCH

CH

3

H

CH

H

OH

H

Intercalation

Intercalation^b

Intercalation

3

Cl

3

H

OH

a

Each ellipticine-DNA ﬁlm contained 4 mol% of 1–6 as a drug, respectively.

Quoted from Ref. 8.

b

carbazol-2-ium

synthesized,

chloride

(18),

which

was

reportedly

reportedly prepared,¹⁸^aeliminated not only the Boc group but

also the methoxycarbonyl group leading to the formation of 2-

(2-aminoethyl)-11-methy-6H-pyrido[4,3-b]carbazol-2-ium

chloride (20) in a 97% yield.

1

8a

13

was alkylated by the reductive amination of N-

Boc-3-aminopropionaldehyde (49) followed by the deprotection

of the amino group in a 59% yield. The aldehyde 49 was prepared

by oxidation of N-Boc-ethanolamine (48) with 2-iodoxybenzoic

acid (IBX) in a 69% yield. The overall yield of 17 from 45 (via 18)

in Method B was 12%, which was larger than those in Method A

using 43a and 43b (3.8 and 7.3%, respectively).

2.2. Judgment of DNA-binding modes of pyridocarbazole derivatives

1–40

In the synthesis of 2-(2-aminoethyl)ellipticin-2-ium chloride

19), we selected acetyl group as a protecting group of the amino

First we examined biopolymers that bound with pyridocar-

bazole derivatives. The complexes of ellipticine with DNA duplexes

(

2

2–25

group of 2-chloroethylamine. 2-Methyl-2-oxazoline (50) was trea-

ted with trimethylchlorosilane in methanol at room temperature

for 24 h leading to the formation of N-acethyl-2-chloroethylamine

51) in an 82% yield (Scheme 2). When ellipticine (1) was treated

with 51 in the presence of potassium iodide in dimethyl sulfoxide

DMSO) at 60 °C for 5 h, 2-(N-acethylamino)ethylellipticin-2-ium

chloride (52) was obtained in a 90% yield, followed by the anion

exchange with Amberlite IRA-900 (chloride form). The amino

group of 52 was deprotected by treating with 1 M hydrochloric

acid in methanol at room temperature for 24 h to give 19 in a

2% yield. The reaction without potassium iodide gave a lower

yield of 52.

Upon treating with concentrated hydrochloric acid under reﬂux

24 h, 2-(2-(N-Boc-amino)ethyl)-5-methoxycarbonyl-11-

methy-6H-pyrido[4,3-b]carbazol-2-ium chloride (53), which was

were widely studied.

On the other hand, RNA:DNA hybrids are

important molecules, which are transiently formed in the course of

many vital biological processes, and are also very signiﬁcant for

drug-design. It was reported that ellipticine was one of the best

binders of RNA:DNA hybrids as a result of competition dialysis

(

2

6

(

studies of many compounds. Previously, we reported a method

to determine the DNA-binding mode using a combination of

deﬂection spectroscopy and orientation of the pyridocarbazole

8

molecule in DNA ﬁlm with chain alignment. As an extension of

our previous work, we selected DNA duplexes as biopolymers to

bind with pyridocarbazole derivatives in this work. Commercial

DNA derived from salmon sperm was used for the preparation of

9

2

7

DNA-based ﬁlm. The compounds, used as standards, were ethid-

0

for

ium bromide (EB) for intercalation mode, 4 ,6-diamidino-2-

phenylindole dihydrochloride (DAPI) for minor-groove binding

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

4

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

Table 3

Determination of the binding mode of 2-alkylellipticin-2-ium chlorides 8–11, 13–16, and 19, and 2-alkyl-11-methyl-6H-pyrido[4,3-b]carbazol-2-ium derivatives 7, 12, 17, 18, and

2

0, by a deﬂection spectroscopy of the oriented drug-molecule in DNA-ﬁlm prepared by a brush application method.^a

Entry

Compound

R²

R⁵

R⁶

R⁹

Binding mode

1

2

3

4

5

6

7

8

9

7

8

9

(CH

2 2

)

2 2

)

2 2

)

2 2

)

2 5

)

2 2

)

2 2

)

2 2

)

2 2

)

2 5

)

2 2

)

2 2

)

2 2

)

2 2

)

O(CH

2

)

2

NHCONHCH

3

COOCH

3

H

CH

H

CH

H

OH

H

Intercalation

Not determined^b

Intercalation

CH

3

CH

3

CH

3

CH

3

10

11

12

13

14

15

16

17

18

19

20

3

NHCONHCH

3

O(CH

)

2 2

)

2 2

)

2 2

)

2 2

NH

2

COOCH

3

CH

3

CH

3

CH

3

CH

3

H

OH

H

3

10

11

12

13

14

NH

NH(CH

2

3

2

)

2

NH

2

COOCH

3

NH

2

3

CH

H

3

H

a

Each pyridocarbazole-DNA ﬁlm contained 4 mol% of 7–20, respectively.

The binding mode was unable to determine, because the preparation of the corresponding DNA ﬁlm was difﬁcult.

b

Table 4

Determination of the binding mode of 5-alkylcarbamoyl-2,11-dimethyl-6H-pyrido[4,3-b]-carbazol-2-ium derivatives 21–34, by a deﬂection spectroscopy of the oriented drug-

a

molecule in DNA-ﬁlm prepared by a brush application method.

Entry

Compound

R⁵

R⁹

Binding mode

1

2

3

4

5

6

7

8

9

21

22

23

24

25

26

27

28

29

30

31

32

33

34

(CH

2

)

2

)

3

)

4

)

5

)

2

)

2

)

2

)

2

)

3

)

4

)

5

)

2

)

2

)

2

NHCONHCH

3

H

OCH

H

Intercalation

Not tested

Intercalation

O(CH

NH(CH

NHCONHCH

2

)

2

NHCONHCH

3

2

) NHCONHCH

3

NH

2

1

0

1

2

3

4

O(CH

NH(CH

NH

2

)

2

NH

2

)

2

OCH

3

a

Each pyridocarbazole-DNA ﬁlm contained 4 mol% of 21–34, respectively.

mode, and rhodamine blue (RhoB) for electrostatic interaction

mode, respectively. The structural formulas of these compounds

are shown in Fig. 3. The angular dependent absorbance curve of

each of the standard compounds, determined at its absorption

because the molecule of EB was oriented in the same direction

with the nucleobase pairs. On the other hand, that of

minor-groove binding typed DAPI-DNA ﬁlm indicated minimum

values at 0° and ±180°, respectively. In this ﬁlm, the molecule of

DAPI was oriented in parallel with the vertical axis of DNA. The

RhoB-DNA ﬁlm, however, did not indicate obvious angular depen-

dence, because RhoB did not indicate speciﬁc orientation against

maximum wavelength

max

k , is shown in Fig. 4 together

with that of DNA ﬁlm itself. The angular dependent curve

of the intercalation typed EB-DNA ﬁlm indicated maximum values

at 0° and ±180°, respectively. It was same as that of DNA ﬁlm,

8

the DNA-axis.

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

5

Table 5

Determination of the binding mode of 5-alkylcarbamoyl-11-methyl-6H-pyrido[4,3-b]carbazole derivatives 35–40 by a deﬂection spectroscopy of the oriented drug-molecule in

a

DNA-ﬁlm prepared by a brush application method.

Entry

Compound

R⁵

R⁹

Binding mode

1

2

3

4

5

6

35

36

37

38

39

40

(CH

2 2

)

2 4

)

2 2

)

2 2

)

2 4

)

2 2

)

NHCONHCH

3

H

OCH

H

Intercalation

Not tested

Not determined^b

Intercalation

Not tested

3

NH

2

OCH

Not determined^b

a

Each pyridocarbazole-DNA ﬁlm contained 4 mol% of 35–40, respectively.

The binding mode was unable to determine, because the preparation of the corresponding DNA-based ﬁlm was difﬁcult.

b

Before determining angular dependent curves of DNA-based

DNA-based ﬁlms for the compounds 11, 37, and 40 were difﬁcult

because of its insolubility in water even in the mixed solvent

(methanol/DMSO = 18:2 v/v).

ﬁlms of the compounds 1–40, solvent effects on the alignment

of DNA were examined in order to prepare a DNA-based ﬁlm of

a compound with not only good but also poor water-solubility.

The results are shown in Table 1. Although DMSO more than 5%

inhibited the orientation of DNA (entries 4–6 in Table 1), DMSO

not exceeding 2% did not affect the orientation of DNA in the

brush application method (entries 2 and 3). On the other hand,

methanol up to 20% did not affect the orientation of DNA at all

in both the brush application and the natural seasoning methods

a

2.3. Binding constants K for the complex of the pyridocarbazole

derivatives with the AT- and GC-rich dsDNA

2.3.1. Determination of the binding constants K

a

for the complexes of

pyridocarbazole derivatives 1–40 with DNA; substitutent effects on

the binding constants K and base-pair preference

The binding constants K for the complexes of pyridocarbazole

(

entries 7–9). On the basis of these results, we selected entry 11

methanol/DMSO = 18:2 v/v) as an optimum condition for the

a

preparation of DNA-based ﬁlm of the compound with poor

water-solubility.

derivatives 1–40 with dsDNA were determined using a 27 MHz

quartz crystal microbalance (QCM) in 1 mM phosphate-ethylene-

diaminetetraacetic acid (EDTA) buffer solution (pH 7.4) in the pres-

ence of 0.05 M NaCl at 25 °C. The sequences of the biotinylated

synthetic A- and G-rich 30-mer ssDNAs were shown in Fig. 6 as a

part of the corresponding dsDNA. Immobilization of avidin mole-

First, the binding modes of DNA complexes with both ellipticine

(1) and 2-methylellipticin-2-ium chloride (2) were determined.

The DNA-based ﬁlms of 1 and 2 were prepared in the same manner

as those of the standard compounds. The angular dependent curves

for the 1-DNA and 2-DNA ﬁlms indicated maximum values at 0°

and ±180°, respectively, like that of EB (Figs. 4b and 5a, b). There-

fore, we concluded that the binding modes of both 1 and 2 were

intercalation (Table 2, entries 1 and 2). The deﬂection spectroscopy

of all other simple ellipticine derivatives 3–6 was also obtained to

determine their binding modes, which were intercalation (entries

ꢁ5

2

cules and the biotin-DNAs on the Au electrode (0.49 ꢀ 10 m ),

leading to the formation of the corresponding AT-rich and GC-rich

30-mer dsDNA (see Fig. 6), were conducted in the same manner as

2

8

the reported method. The immobilized dsDNA on the Au elec-

trode was treated with an aliquot of a solution of each compound

1–40 in the buffer solution with stirring. A typical time course of

3

–6).

Next, the binding modes of the 2-alkyl substituted pyridocar-

bazol-2-ium chlorides 7–20, the 5-alkylcarbamoyl substituted

,11-dimethyl-6H-pyrido[4,3-b]carbazol-2-ium chlorides 21–34,

the frequency decreases DF of the QCM is shown for complexation

of 8 with the GC-rich 30-mer dsDNA in Fig. 7a. After ﬁrst satura-

tion, next aliquot of the solution was sequentially added. The fre-

2

quency decrease

D

F of 1.00 Hz corresponds to the mass increase

of 0.606 ng cm in the 27 MHz QCM. The frequency decreases

F were transformed into the mass increases m (increases of

ꢁ2

and the corresponding 5-alkylcarbamoyl substituted un-quater-

nized pyridocarbazole 35–40 were determined. These compounds

possessed the amino or ueido group linking to the pyridocarbazole

framework via the ether, imino, and methylene tethers, respec-

tively, at the N-2 or C-5 atom. The results are shown in Tables 3–

D

the DNA-complex of the compound 8), which showed a saturation

curve against the concentration of the added compound 8 in the

phosphate-EDTA buffer as shown in Fig. 7b. The saturation binding

behavior is expressed as a linear relationship by Eq. (1) as a recip-

5. All compounds indicated the angular dependent curves for the

intercalation mode regardless of the length and the type of the

tether. Furthermore, even the substitution position did not affect

its binding mode. That is, the substituent at the N-2 and C-5 posi-

tion did not inhibit the intercalation. However, it was clear from

rocal plot between the concentration [the compound 8]/Dm and

5

ꢁ1

[the compound 8]. The binding constant K

maximum binding amount

a

(9.92 ꢀ 10 M ) and

ꢁ9

D

m

max (3.08 ꢀ 10 g) of the com-

pound 8 to the GC-rich dsDNA were calculated from the intercept

the binding constant K

a

and the modeling that the substitution at

and the slope of Eq. (1), respectively. Alternatively, the binding

the C-5 position had a disadvantage for the stability of the

intercalated complex as discussed latter. The preparation of the

constant K

a d

was obtained as a reciprocal of K according to Eq.

(2), where [pyridocarbazoles]1/2 indicates a concentration of the

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

6

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

Scheme 1. Synthesis of 2-(2-((2-aminoethyl)amino)ethyl)-5-(methoxycarbonyl)-11-methyl-6H-pyrido[4,3-b]carbazol-2-ium chloride (17).

pyridocarbazole at a half of

In most case, we used the latter.

D

m

max (an asymptotic value) in Fig. 7b.

nization at the N-2 atom by a methyl group (2–2.5-fold of those

of 1). Furthermore, the binding constants K for the complexes of

a

the 2-tethered pyridocarbazolium derivatives 7–20 with the AT-

rich (and GC-rich) dsDNA (Table 6, entries 7–20) were larger than

those for the corresponding complexes of the 5-tethered com-

pounds 21–40 (Table 7, entries 1–20); for example, 13 (2-tethered

½

pyridocarbazoleꢂ ½pyridocarbazoleꢂ

1

¼

þ

ð1Þ

ð2Þ

D

m

D

m

max

D

m

max a

K

1

6

ꢁ1

½

pyridocarbazoleꢂ₁₌₂¼ K

d

¼ _K

compound):

K

a

= 6.51 ꢀ 10 M

(AT-rich;

16-fold)

and

a

7

ꢁ1

5

2

.99 ꢀ 10 M

(GC-rich; 6-fold), 32 (5-tethered compound):

ꢁ1

6

ꢁ1

The thus-obtained binding constants K for the complexes of the

-tethered compounds 1–20 with the AT- and GC-rich dsDNAs and

a

K

a

= 4.42 ꢀ 10 M

(AT-rich) and 5.37 ꢀ 10 M

(GC-rich),

2

respectively. The binding ability against the GC-rich dsDNA was

those of the 5-tethered compounds 21–40 were summarized in

remarkably enhanced by the 2-(

pounds 12–19 (Table 6, entries 12–19). Especially, the compounds

14 and 16 exhibited extremely large binding constants K

a

x-aminoalkyl) tether of the com-

Tables 6 and 7, respectively. Binding ability of 2-methylellipticin-

5

ꢁ1

2

-ium chloride (2) (K

a

= 8.37 ꢀ 10 M

for the AT-rich dsDNA;

6

ꢁ1

7

ꢁ1

.99 ꢀ 10 M for the GC-rich dsDNA) was increased by quater-

(3.87 ꢀ 10 and 4.54 ꢀ 10 M

) against the GC-rich dsDNA

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

7

Scheme 2. Synthesis of 2-(2-aminoethyl)ellipticin-2-ium chloride (19) and 2-(2-aminoethyl)-11-methyl-6H-pyrido[4,3-b]carbazol-2-ium chloride (20).

Fig. 3. Structural formulas of standard dyes for each binding mode; EB: intercalation mode, DAPI: minor-groove binding mode, RhoB: electrostatic interaction mode.

Fig. 4. Angular dependent curves of DNA-based ﬁlms for the standard dyes; (a) DNA itself at 290 nm: control, (b) ethidium-bromide (EB)-DNA at 526 nm: intercalation mode,

8

(

c) DAPI-DNA at 360 nm: minor-groove binding mode, (d) rhodamine blue (RhoB)-DNA at 565 nm : electrostatic interaction mode.

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

8

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

Fig. 5. Angular dependent curves for (a) ellipticine (1)-DNA and (b) 2-methylellipticin-2-ium chloride (2)-DNA ﬁlms at 424 nm.

Fig. 6. Sequences of the AT-rich and GC-rich dsDNAs.

Fig. 7. (a) A typical time course of frequency decreases

DF of the immobilized GC-rich 30-mer dsDNA on an Au electrode of a 27 MHz QCM responding to the addition of the

compound 8 in the 1 mM phosphate-EDTA buffer solution (pH 7.4) in the presence of 0.05 M NaCl at 25 °C, and (b) the mass increase

saturation curve against the concentration of the added compound 8 in the same reaction.

Dm (increase of the DNA-complex of 8)

(

33- and 38-folds in comparison with that of ellipticine (1) as a

hydroxyellipticin-2-ium chloride 6 with the AT-rich dsDNA was

larger than those of 1 and 2. The effects of the positive charge

control). The enhancement effects of the -aminoalkyl group on

x

the binding ability were considered to be attributable to the inter-

action between the protonated amino group and the anionic phos-

phate group in DNA. Indeed, the binding constants K of the

a

compounds possessing the terminal amino group increased with

the decreasing pH as discussed latter.

and the 9-hydroxy group on K

were estimated as 4-folds, respectively, on the basis of K

compounds. Furthermore, the effects of the 2-(2-((2-aminoethyl)

amino)ethyl) group on K for the AT- and GC-rich dsDNA com-

plexes were estimated as 9-fold and 8.5-fold on the basis of K val-

a

for the AT-rich dsDNA complex

a

of these

a

Next, K

a

values of the quaternized compounds 21, 23, 27, 28, 30,

ues of 6 and 14. The result suggested an additional interaction of 2-

(2-((2-aminoethyl)amino)ethyl) group with DNA as mentioned

above. The methyl group at N-6 position did not disturb the forma-

tion of the complex (Table 6, entries 9 and 10).

and 34 were compared with those of the corresponding un-quater-

nized compounds 35–40 in order to evaluate the effects of the pos-

itive charge on the binding ability of the 5-tethered

pyridocarbazole derivatives. It was clear from Table 7 that the

enhancing effects were observed for the quaternized compounds

It was reported that the pyridocarbazole framework of 1 in the

intercalated complex was buried within base pairs and oriented

perpendicular to the DNA-helix axis, with the 9-hyxroxy group

protruding into the minor groove and the N-2 atom lying in the

2

3 (13-fold for the AT-rich and 4-fold for the GC-rich dsDNA,

respectively), 30 (15-fold for the AT-rich and 54-fold for the GC-

rich), and 34 (5-fold for the AT-rich and 6-fold for the GC-rich).

These enhancing effects were larger than those between 1 and 2

2

4

major groove. Therefore, the long tether on the C-5 atom has a

disadvantage for intercalation, and it is supported by smaller K

a

(

2-fold against the AT-rich and 2.5-fold against the GC-rich dsDNA,

values of the 5-tethered compounds (Table 7) compared with those

of the corresponding 2-tethered compounds (Table 6). Contrary to

our expectation, lack of the 5-methyl group, however, decreased

the binding ability of 2-(2-aminoethyl)ellipticin-2-ium chloride

19 (0.1-fold against the AT-rich dsDNA and 0.07-fold against the

GC-rich dsDNA; Table 6, entries 19 and 20).

2

respectively). Although replacement of the central CH group by

an oxygen atom in the tether at the N-2 atom did not affect (or

slightly increased) the binding ability (Table 6, entries 15 and 16;

Table 7, entries 11 and 12), replacement by an NH group dimin-

ished the binding ability in both the 2- and 5-tethered compounds,

1

7 and 33 (Table 6, entries 16 and 17; Table 7, entries 11 and13). It

Extension of the length of the methylene chain of the 5-(2-

aminoalkyl)carbamoyl tether of the un-quaternized compound

from (CH ) of 38 to (CH ) of 39 decreased the binding ability

2 2 2 4

was reported that 9-hydroxyellipticinium derivatives possessing a

positive charge interacted with poly[d(AT)] not only by intercala-

tion but also by interaction between the positive charge (and/or

hydroxy group) and the groups belonging to the backbone, base,

and sugars (an external binding) at high ratio of the compound

(AT: 0.2-fold, GC: 0.05-fold; Table 7, entries 18 and 19).

Finally, the base-pair preference of the compounds in the for-

mation of the complexes with dsDNA was examined. As can be

seen in Table 6, all of the 2-tethered compounds 1–20 exhibited

2

3

a

to DNA ratio. Indeed, K value of the complex for 2-methyl-9-

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

9

Table 6

a

The binding constants K for the DNA complexes of the un-quaternized ellipticine derivatives (type A) 1, 3, 4 and those of the quaternized ellipticine and pyridocarbazole

a

derivatives (type C) 2, 5–20, determined by QCM.

R²

R⁵

R⁶

R⁹

a

K M )

(10⁵

ꢁ1

Entry

Comd.

Type

AT-rich^b

4.20^d

GC-rich^c

1

2

3

4

5

6

7

8

9

1

2

3

4

5

6

7

8

A

C

A

C

None

CH

COOCH

CH

COOCH

CH

3

H

CH

H

CH

H

CH

H

OH

H

OH

H

OH

H

12.3 ± 5.2

29.9 ± 3.3

6.88 ± 0.60

107 ± 16

CH

3

8.37 ± 1.33

None

3

Not determined

10.5 ± 1.3

35.1 ± 2.3

7.41 ± 0.91

3.59 ± 2.63

6.81 ± 1.16

4.50 ± 0.72

15.8 ± 1.3

80.0 ± 17.3

65.1 ± 1.9

311 ± 52

102 ± 42

137 ± 4

16.0 ± 4.1

39.5 ± 9.9

40.3 ± 14.1

4.29^d

CH

3

15.6 ± 2.1

47.2 ± 3.9

3

d

(CH

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

5

2

5

2

O(CH

2

)

2

NHCONHCH

3

14.8

3

9.9 ± 2.2

9

32.8 ± 5.6

21.2 ± 6.5

28.5 ± 0.8

111 ± 26

299 ± 76

387 ± 126

311 ± 190

454 ± 211

10

11

12

13

14

15

16

17

18

19

20

10

11

12

13

14

15

16

17

18

19

20

3

NHCONHCH

3

O(CH

)

2 2

)

2 2

)

2 2

)

2 2

NH

2

COOCH

3

CH

COOCH

3

H

OH

H

3

NH

NH(CH

2

3

d

2

) NH

2

3

82.0

NH

2

3

94.2 ± 24.6

120 ± 13

CH

H

3

d

H

8.33

a

The binding constant K

a

was determined in a 1 mM phosphate-EDTA buffer (pH 7.4) with NaCl (0.05 M) at 25 °C, and are reported as the means of three experiments; the

standard deviations are given together.

b

0

Biotin-5 (CAT GAA AAA AAA AAA AAA AAA AAA AAT GCA)-3 .

0

3

-(GTA C T T T T T T T T T T T T T T T T T T T T T TA CGT)-5 .

c

0

Biotin-5 (CAT GGG GGG GGG GGG GGG GGG GGG GGT GCA)-3 .

0

3

-(GTA CCC CCC CCC CCC CCC CCC CCC CCA CGT)-5 .

Single experiment.

d

preference against the GC-rich dsDNA (1.3–5-fold vs the AT-rich

dsDNA). In the case of the 5-tethered compounds, the structure

of the tether affected the preference in an obvious fashion.

Whereas the compounds 21–27 and 35–37, possessing the termi-

2.3.3. Salt-effects on the binding constants K

a

Salt-effects on the binding constants for the DNA-complexes

with the 2-tethered compounds 8–11, 13–16 (Table 9) and the 5-

tethered compounds 21, 28, 34, 38, and 40 (Table 10) were exam-

0

nal N -methylureido group, exhibited no preference at all, the com-

a

ined at 50 and 200 mM NaCl. The binding constants K decreased at

pounds 28–32, 34, 38, and 40, possessing the terminal amino

group, exhibited strong GC base-pair preference (5–30-fold), as

reported previously for 9-hydroxy-2-methylellipticin-2-ium chlo-

0.4–0.5-fold magniﬁcation with the increasing NaCl concentration

for all tested compounds, except for 34 (Table 10, entry 4), regard-

less of the tether position N-2 and C-5 atoms. This result suggested

that the observed salt-effects arose from DNA rather than from the

2

5

ride (2).

structures of pyridocarbazole derivatives. The binding constant K

of 34 was dramatically decreased with increasing NaCl concentra-

a

2

.3.2. pH dependence of the binding constants K

In order to clarify the effect of pH on the binding, K

a

5

ꢁ1

tion (K

a

= 8.28 ꢀ 10 M at 200 mM NaCl, 0.067-fold of that at

a

was deter-

5

0 mM). As one of the plausible explanations, the depression effect

mined at three different pH’s 7.0, 7.4, and 7.8. The result is shown

in Table 8. The effect of the decreasing pH on the binding constants

K

derivatives, 21, 24, 27, 28 and 34 (type D in Table 8, entries 1–4

and 6; 1.1–1.8-fold) and medium for the quaternized derivative

a

on K may be related to the decreasing electrostatic interaction

between the amino group and the phosphate anion of DNA, which

a

was small for the quaternized 5-(x-ureidoalkyl)carbamoyl

2

9

is neutralized by the increasing sodium cation. The details, how-

ever, are not clear at this time.

3

1 (entry 5; 2.6-fold) at pH 7.0 compared with pH 7.8. As opposed

to this moderate effect, a large enhancement was observed for the

un-quaternized 5-( -aminoalkyl)carbamoyl derivatives 38 and 40

type E in Table 8, entries 9 and 10; 4.6–7.8-fold). This large

2.4. Kinetic analysis of the binding process of ellipticine and

pyridocarbazole derivatives with dsDNA

x

(

enhancement may be mainly caused by the protonation of the ter-

minal amino group. Contrary to our expectation, the binding con-

stants of the un-quaternized derivatives 35 and 37 were reduced

to one-half with decreasing pH (Table 8, entries 7 and 8).

a

In order to clarify the reason why K for the complex of 34 with

7

ꢁ1

the GC-rich dsDNA (1.24 ꢀ 10 M , Table 7, entry 14) was remark-

ably larger than those of the other 5-tethered compounds, kinetic

analysis of the binding process of 34 was conducted using the

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

1

0

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

Table 7

The binding constants K

alkylcarbamylpyridocarbazole derivatives (type E) 35–40, determined by QCM. .

a

for the DNA complexes of the quaternized 5-alkylcarbamyl-2-methylpyridocarbazole derivatives (type D) 21–34 and those of the un-quaternized 5-

a

R⁵

R⁹

K

(10⁵

AT-rich^b

M )

ꢁ1

Entry

Compd.

Type

a

GC-rich^c

1

2

3

4

5

6

7

8

9

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

D

E

(CH

2 2

)

2 3

)

2 4

)

2 5

)

2 2

)

2 2

)

2 2

)

2 2

)

2 3

)

2 4

)

2 5

)

2 2

)

2 2

)

2 2

)

2 2

)

2 4

)

2 2

)

2 2

)

2 4

)

2 2

)

NHCONHCH

3

H

OCH

H

OCH

H

OCH

H

1.87 ± 0.56

1.72 ± 0.41

2.49 ± 0.69

2.14 ± 0.63

1.89 ± 0.36

2.20 ± 0.64

2.04 ± 0.30

3.81 ± 2.44

3.44 ± 1.23

4.31 ± 1.33

2.86 ± 0.74

4.42 ± 1.69

2.64 ± 0.19

19.9 ± 1.4

1.67^d

2.87 ± 0.09

1.50 ± 0.22

1.83

d

2.69 ± 0.64

1.68 ± 0.23

1.66 ± 0.11

3.62 ± 0.83

21.1 ± 8.8

89.2 ± 1.5

39.7 ± 29.1

24.3 ± 9.8

53.7 ± 19.4

4.85 ± 2.45

124 ± 2

2.01 ± 0.23

0.427 ± 0.201

3.03 ± 0.43

14.5 ± 4.8

0.733 ± 0.335

20.5 ± 7.5

O(CH

NH(CH

NHCONHCH

2

)

2

NHCONHCH

3

2

) NHCONHCH

3

NH

2

10

11

12

13

14

15

16

17

18

19

20

O(CH

NH(CH

NH

2

)

2

NH

2

)

2

3

NHCONHCH

3

E

3

0.185 ± 0.042

4.75 ± 0.11

1.82 ± 0.71

0.279 ± 0.021

3.88 ± 0.64

NH

2

E

OCH

a

The binding constant K

a

was determined in a 1 mM phosphate-EDTA buffer (pH 7.4) with NaCl (0.05 M) at 25 °C, and are reported as the means of three experiments; the

standard deviations are given together.

b

0

Biotin-5 (CAT GAA AAA AAA AAA AAA AAA AAA AAT GCA)-3 .

0

3

-(GTA CTT T T T T T T T T T T T T T T T T T T T TA CGT)-5 .

c

0

Biotin-5 (CAT GGG GGG GGG GGG GGG GGG GGG GGT GCA)-3 .

0

3

-(GTA CCC CCC CCC CCC CCC CCC CCC CCA CGT)-5 .

Single experiment.

d

QCM together with those 2-methylellipticin-2-ium chloride (2)

and 9-hydroxy-2-methylellipticin-2-ium chloride (6). Binding

kinetics of the 34 to the GC-rich dsDNA can be calculated from

the relaxation time courses of frequency decreases on QCM

(the reciprocal plot of the relaxation time s against the initial

concentration

of

the

pyridocarbazole

derivative

[pyridocarbazole]

0

; see Eq. (6)). The rate constants k

1

and kꢁ1 were

obtained from the slope and intercept of Fig. 8b (the reciprocal

plot of the relaxation time against the initial concentration of the

pyridocarbazole derivative [pyridocarbazole]

(Fig. 8a), and the binding process is expressed by Eq. (3).

s

0

; see Eq. (6)). The

k

1

4

ꢁ1 ꢁ1

pyridocarbazole þ DNA ¢ pyridocarbazole ꢃ DNA

ð3Þ

obtained rate constants k

1

and kꢁ1 of 34 were 2.14 ꢀ 10 M

s

and

k

ꢁ1

ꢁ2 ꢁ1

4

ꢁ1 ꢁ1

and 1.25 ꢀ 10

s

for the AT-rich dsDNA; 1.82 ꢀ 10 M

s

ꢁ

3

ꢁ1

The amount of the complex of pyridocarbazole with DNA [pyri-

1.64 ꢀ 10

s

for the GC-rich dsDNA, respectively. The kinetic

docarbazole ꢃ DNA] (= ), formed at time t after injection of pyri-

docarbazole, is given by Eqs. (4)–(6) using

pyridocarbazole ꢃ DNA] (= max), the relaxation time of the

binding ( ), and the binding (association) and dissociation rate

constants (k

and kꢁ1), as reported previously.²⁸^a^,³⁰

t

Dm

t

parameters for the compounds 2, 6, and 34 are shown in Table 11

3

0a

together with those of ethidium bromide (EB)

Although the association rate constant k of 34 for the GC-rich

dsDNA was nearly equal to that of EB (1.2 ꢀ 10 M

dissociation rate constant kꢁ1 of 34 for the GC-rich dsDNA

as a control.

[

1

D

m

1

4

ꢁ1 ꢁ1

s

s ), the

1

ꢁ3 ꢁ1

(

1.64 ꢀ 10

s ) was extremely smaller than that of EB

0.210 s ). That is, the dissociation of the complex of 34 with the

½

pyridocarbazole ꢃ DNAꢂ ¼ ½pyridocarbazole ꢃ DNAꢂ₁f1 ꢁ expðꢁð1=

s

ÞtÞg

ꢁ1

t

ð4Þ

GC-rich dsDNA was very slow (0.8% of the EB-DNA complex). The

dissociation rate of the complex of 34 with the AT-rich dsDNA was

also slower than that of ethidium bromide (8%). That is, the

dissociation rate of the GC-rich dsDNA-complex with 34 was

slower than that of the AT-rich dsDNA by 0.1-fold. This may be the

reason for the GC base-pair preference of 34. The rate constants of

D

m

t

¼

Dm

maxf1 ꢁ expðꢁð1=

s

ÞtÞg

ð5Þ

where

1

=s

¼ k

1

½pyridocarbazoleꢂ þ kꢁ1

ð6Þ

0

2

and 6 for the GC-rich dsDNA were nearly equal to those of 34

The binding relaxation time

course of F (transformed into

by a curve-ﬁtting method based on Eq. (5).

and kꢁ1 were obtained from the slope and intercept of Fig. 8b

s

was calculated from the time

and the rate constants of 6 for the AT-rich dsDNA were slightly

smaller than those for 34. The binding constants K_awere also

calculated as the ratio of k to kꢁ1 for the complexes of 2, 6, and 34

1

D

Dm) at each concentration (Fig. 8a)

30a,31

The rate constants

k

1

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

11

Table 8

a

pH dependence of the binding constants K of 5-alkylcarbamyl-2,11-dimethyl-6H-pyrido[4,3-b]carbazol-2-ium derivatives 21, 24, 27, 28, 31, 34, and 5-alkylcarbamyl-2,11-

a,b

dimethyl-6H-pyrido[4,3-b]carbazole derivatives 35, 37, 38 and 40 with the GC-rich DNA oligomer, determined by QCM.

R⁵

R⁹

K

a

(10⁵

M

ꢁ1

)

Ratio^c

Entry

Compd.

Type

pH = 7.0

pH = 7.4

pH = 7.8

1

2

3

4

5

6

7

8

9

21

24

27

28

31

34

35

37

38

40

D

E

(CH

2

)

2

)

5

)

2

)

2

)

5

)

2

)

2

)

2

)

2

)

2

NHCONHCH

3

H

OCH

H

OCH

H

OCH

H

2.90 ± 0.28

2.34 ± 0.65

3.74 ± 1.47

33.2^d

45.0 ± 2.3

147 ± 15

1.05 ± 0.12

2.17 ± 0.56

22.3 ± 7.9

25.1 ± 1.6

2.87 ± 0.09

2.69 ± 0.64

3.62 ± 0.83

21.1 ± 8.8

24.3 ± 9.8

124 ± 2

2.01 ± 0.23

3.03 ± 0.43

14.5 ± 4.8

20.5 ± 7.5

2.31 ± 0.57

2.11 ± 0.02

2.55 ± 0.08

18.2^d

16.7 ± 2.3

108 ± 11

2.20 ± 0.08

3.75^d

1.3

1.1

1.6

1.8

2.6

1.4

0.5

0.6

4.6

7.8

3

NH

2

3

NHCONHCH

3

NH

2

4.83 ± 0.22

3.20 ± 0.65

1

0

2

OCH

a

The binding constant K

a

was determined in a 1 mM phosphate-EDTA buffer with NaCl (0.05 M) at 25 °C, and are reported as the means of three experiments; the standard

deviations are given together.

b

0

Biotin-5 (CAT GGG GGG GGG GGG GGG GGG GGG GGT GCA)-3 .

-(GTA CCC CCC CCC CCC CCC CCC CCC CCA CGT)-5 .

Ratio of Ka(pH7.0) to Ka(pH7.8)

Single experiment.

0

3

c

.

d

Table 9

a,b

Salt-effects on the DNA-binding constants of 11-methyl-6H-pyrido[4,3-b]carbazol-2-ium derivatives 8–11 and 13–16, determined by QCM using the GC-rich DNA oligomer.

R²

R⁵

R⁶

R⁹

K

(10⁵

M

ꢁ1

)

Entry

Comd.

a

+

[

Na ] = 50 mM

[Na ] = 200 mM

1

2

3

4

5

6

7

8

9

(CH

)

2 2

)

2 2

)

2 2

)

2 5

)

2 2

)

2 2

)

2 2

)

2 5

O(CH

2

)

2

NHCONHCH

3

CH

3

H

CH

H

CH

H

9.92 ± 1.94

32.8 ± 5.6

21.2 ± 6.5

28.5 ± 0.8

299 ± 76

387 ± 126

311 ± 190

454 ± 211

5.99 ± 1.81

13.0 ± 3.1

11.5 ± 1.4

17.9 ± 1.2

92.8 ± 21.6

155 ± 16

OH

H

OH

10

11

13

14

15

16

3

NHCONHCH

3

O(CH

2

)

2

)

2

)

2

NH

2

3

162 ± 47

272 ± 20

NH

2

a

The binding constant K

are given together.

a

was determined in a 1 mM phosphate-EDTA buffer (pH 7.4) at 25 °C, and are reported as the means of three experiments; the standard deviations

b

0

Biotin-5 (CAT GGG GGG GGG GGG GGG GGG GGG GGT GCA)-3 .

0

3

-(GTA CCC CCC CCC CCC CCC CCC CCC CCA CGT)-5 .

with the AT- and GC-rich dsDNAs (Table 11). The binding constants

, obtained from the relaxation method according to Eqs. (3)–(5),

2.5. Optimal structures and relative stabilization energies for the DNA

complexes of 11-methyl-6H-pyrido[4,3-b]carbazol-2-ium derivatives

K

a

were fairly consistent with those obtained from the saturation

method according to Eqs. (1) and (2) (Table 6, entries 2 and 6;

Table 7, entry 14).

Previously, we reported that a good linear relationship was

observed between the binding constants of DNA origomers and

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

1

2

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

Table 10

Salt-effects on the DNA-binding constants of 5-alkylcarbamyl-2,11-dimethyl-6H-pyrido[4,3-b]-carbazol-2-ium derivatives 21, 27, 28, and 5-alkylcarbamyl-2,11-dimethyl-6H-

a,b

pyrido[4,3-b]carbazole derivatives 38 and 40, determined by QCM using the GC-rich DNA oligomer.

R⁵

R⁹

K

(10⁵

M

ꢁ1

)

Entry

Compd.

Type

a

+

[

Na ] = 50 mM

[Na+] = 200 mM

1

2

3

4

5

6

21

27

28

34

38

40

D

E

(CH

)

2 2

)

2 2

)

2 2

)

2 2

)

2 2

)

2 2

NHCONHCH

3

H

OCH

H

OCH

H

OCH

2.87 ± 0.09

3.62 ± 0.83

21.1 ± 8.8

124 ± 2

14.5 ± 4.8

20.5 ± 7.5

0.899 ± 0.154

1.46 ± 0.10

7.55 ± 1.83

8.28 ± 1.00

5.31 ± 11.33

7.41 ± 1.11

3

NH

2

E

a

The binding constant K

are given together.

a

was determined in a 1 mM phosphate-EDTA buffer (pH 7.4) at 25 °C, and are reported as the means of three experiments; the standard deviations

b

0

Biotin-5 (CAT GGG GGG GGG GGG GGG GGG GGG GGT GCA)-3 .

0

3

-(GTA CCC CCC CCC CCC CCC CCC CCC CCA CGT)-5 .

Fig. 8. Determination of kinetic parameters for the binding processes of 34 with dsDNA oligomer (GC-rich), determined using a 27 MHz QCM in a 1 mM phosphate-EDTA

buffer (pH 7.4) with NaCl (50 mM) at 25 °C. (a) The relaxation time course of the frequency decrease after injection of 1.0, 0.8, 0.6, 0.4, and 0.1 M of 34; (b) the reciprocal plot

l

4

ꢁ1 ꢁ1

of the relaxation time

s

against the concentration of 34. The obtained kinetic parameters: the binding rate constant k

1

= 1.82 ꢀ 10

M

s ; the dissociation rate constant

ꢁ1 = 1.64 ꢀ 10^ꢁ

3

s

ꢁ1

; the binding constant K

7

M .

ꢁ1

k

a

1

= k /kꢁ1 = 1.12 ꢀ 10

the stabilization energies of optimal structures of EB-DNA inter-

calative complexes, obtained by molecular mechanics. In the pre-

sent study, we analogously simulated optimal structures of the

intercalative complexes of d(GGGCCC) with 6H-pyrido[4,3-b]car-

2

bazol-2-ium derivatives 13–16, 19, 21–25, and 28–34 by a molec-

ular modeling. The optimal structures for the complexes of 13 and

bility Srel was calculated. We selected the complex of 22 as a

reference and the result was summarized in Table 12. The stabi-

32

lization energy

energy of the complex and the sum of the energies of the corre-

sponding components (d(GGGCCC) and the pyridocarbazole

derivative). The complexes of the compounds 13–16, possessing

the 2-( -aminoalkyl) tether (Table 12, entries 1–4), exhibited Srel

DE was calculated as a difference between the

2

3

2

2 with d(GGGCCC)

docarbazole framework of the 2-tethered

2-(2-aminoethoxy)ethyl)ellipticin-2-ium chloride 13, completely

2

are shown in Fig. 9a and b. Whereas the pyri-

x

-amino derivative, 2-

larger (Srel = 2.56–3.03) than the others. Especially, the Srel values

for the complexes of the 5-tethered compounds (Table 12, entries

7–10, 11–16) were smaller (Srel = 1.00–2.26) than those for the 2-

(

overlapped with the stacking GC base-pairs leading to the forma-

tion of the most stable intercalative-complex (Fig. 9a and Table 12),

the pyridocarbazole framework of the 5-tethered ureido derivative,

tethered compounds (Srel = 2.56–3.03). When lnK

against Srel, there was a good linear relationship for the 2-(

amino)alkyl-tethered compounds 13–16 and 19 (Fig. 10a)

although slight deviation was observed for 13

a

were plotted

x

-

2

,11-dimethyl-5-((3-(3-methylureido)propyl)carbamoyl)-6H-pyr-

ido[4,3-b]carbazol-2-ium chloride 22, half overlapped and its

tether bound to the minor grove of the DNA leading to the forma-

tion of the most unstable complex (Fig. 9b). This may be the reason

for the smaller binding constants of the 5-tethered compounds.

In order to compare the stability for the complexes of the pyri-

docarbazole derivatives 13–16, 19–21, 23–25, 28–34, relative sta-

(y = 0.5492x ꢁ 6.8869, relation coefﬁcient r = 0.9998 with the

exception of 13). Analogous linear relationships were also

observed for the 5-(

28–34 (Fig. 10b; y = 0.4919x ꢁ 5.5572, relation coefﬁcient

r = 0.7562 with the exception of 33) and the 5-( -ureidoalkyl)

x-aminoalkyl)carbamoyl-tethered compounds

x

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

13

Table 11

Kinetic parameters for the binding processes of 2-methylellipticin-2-ium chloride (2), 9-hydroxy-2-methylellipticin-2-ium chloride (6), and 5-(2-aminoethylcarbamyl)-2,11-

a

dimethyl-6H-pyrido[4,3-b]carbazol-2-ium chloride (34) with dsDNA oligomers, determined by QCM.

DNA^b

k

1

(10⁴

M

ꢁ1

s

ꢁ1

)

kꢁ1 (10

ꢁ3 ꢁ1

s

)

a

K M )

(10⁶

ꢁ1

Compound

2

6

3

GC-rich

0.51

1.51

3.4

2

6

AT-rich

GC-rich

0.78

1.28

8.05

2.01

0.97

6.37

4

AT-rich

GC-rich

2.14

1.82

12.5

1.64

1.71

11.2

3

4

Ethidium bromide^c

AT

GC

0.17

1.2

150

210

0.011

0.056

a

The kinetic parameters were determined in a 1 mM phosphate-EDTA buffer (pH 7.4) with NaCl (0.05 M) at 25 °C.

b

0

AT-rich: biotin-5 (CAT GAA AAA AAA AAA AAA AAA AAA AAT GCA)-3 3 -(GTA CTT T T T T T T T T T T T T T T T T T T T TA CGT)-5 .

0

GC-rich: biotin-5 (CAT GGG GGG GGG GGG GGG GGG GGG GGT GCA)-3 3 -(GTA CCC CCC CCC CCC CCC CCC CCC CCA CGT)-5 .

c

Quoted from Ref. 30a. AT: dA30 ꢃ dT30 dsDNA, GC: dG30 ꢃ dC30 dsDNA.

Fig. 9. Modeling illustrations of the intercalated complexes for the 2-tethered and 5-tethered pyridocarbazol-2-ium derivatives with d(CCCGGG)

amino)alkyl tethered ellipticine 13 (completely overlapped insertion); (b) a complex of 5-( -amino)alkylcarbamoyl tethered 2,11-dimethyl-6H-pyrido[4,3-b]carbazol-2-ium

2 (half overlapped incomplete insertion).

2

; (a) a complex of 2-(x-

x

2

carbamoyl-tethered compounds 21–25 (Fig. 11; y = 1.8899x ꢁ

1.318, relation coefﬁcient r = 0.8101), respectively. The reason

lines. Their IC50 values were summarized in Table 13 together with

those of the other pyridocarbazole derivatives reported previ-

2

1

8

for the deviation of 13 and 33 from the regression lines may be

attributable to the hydrogen bond between the corresponding

terminal amino group of the tether and some of the functional

groups on the DNA-framework as shown in Fig. 9a for the com-

pound 13.

ously. The compounds 9, 12, 13, 14, 16, and 18 exhibited more

potent cytotoxic activity against Sarcom-180 than against HeLa

S-3 by the power of 13–37-fold and also against NIH3T3 by the

power of 3–24-fold. If the cytotoxic activity of the pyridocarbazole

arises from its intercalation to DNA, the IC50 values of the com-

pounds may be related to the binding constant K

the hypothesis, lnK was plotted against IC50 of some of the pyrido-

carbazole derivatives 1–40. As an example, a scatter plot of lnK for

a

. In order to verify

2

.6. Relationship between Ka for the complexes of the pyridocarbazole

a

derivatives 1–40 with DNA and their cytotoxic activity IC50 against

HeLa S-3, Sarcoma-180, and NIH3T3 cell lines

the complexes of 1–3, 5, 8–16, 18, 19, 23, 27, 35, 37, 38, and 40

with the GC-rich dsDNA vs. their IC50 against HeLa S-3 was shown

in Fig. 12. Contrary to our expectation, there was no linear relation-

Recently, we reported the synthesis and the cytotoxic activity of

3

the pyridocarbazole derivatives.¹

tives exhibited potent cytotoxic activity. In this work, we evaluated

the cytotoxicity of additional pyridocarbazole derivatives 3, 9, 10,

and 12–14, 16–19 against HeLa S-3, Sarcoma-180 and NIH3T3 cell

8,19

The pyridocarbazole deriva-

^s^hⁱ^p^b^e^t^w^e^eⁿ^lⁿ^Ka ^aⁿ^d^I^C50 values.

^T^h^e^p^l^o^t^s^o^f^I^C50 values

against both Sacoma-180 and NIH3T3 cell lines were also same.

Furthermore, there was also no linear relationship for the complex

of the AT-rich dsDNA.

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

1

4

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

Table 12

Relative stability Srel of 5-substituted 11-methyl-6H-pyrido[4,3-b]carbazol-2-ium derivatives 8–11, 13–16, 19, 21–25, 27, and 28–34 with dsDNA oligomer d(GGGCCC)

2

to that of

a

2

2, calculated by a molecular modeling.

R²

R⁵

R⁶

R⁹

S

rel

b

Entry

Compd.

Functional group

1

2

3

4

5

13

14

15

16

19

Amino

(CH

)

2 2

)

2 2

)

2 2

)

2 5

)

2 2

O(CH

2

)

2

)

2

)

2

NH

2

NH

2

NH

2

CH

3

H

CH

H

3.03

2.72

2.56

2.80

2.07

OH

H

3

NH

2

3

2

6

7

8

9

21

22

23

24

25

Ureido

CH

3

CONH(CH

2 2

)

2 3

)

2 4

)

2 5

)

2 2

)

NHCONHCH

3

H

2.80

1.00

1.40

1.87

1.91

3

1

0

O(CH

) NHCONHCH

2 2 3

1

2

3

4

5

6

7

28

29

30

31

32

33

34

Amino

CH

3

CONH(CH

2 2

)

2 3

)

2 4

)

2 5

)

2 2

)

2 2

)

2 2

)

NH

O(CH

NH(CH

NH

2

H

OCH

3

1.62

2.22

1.37

1.94

2.26

2.13

2.62

2

)

2

NH

2

)

2

a

The calculation was performed using DS-modeling^Òfor the molecules in water in the presence of 0.05 M NaCl.

b

rel 0 0 2

S = DE/DE ; DE : the stabilization energy for the complex of 22 with d(GGGCCC) .

Fig. 10. Linear relationships of lnK

y = 0.5492x ꢁ 6.8869, relation coefﬁcient r = 0.9998 with the exception of 13), (b) the 5-(

r = 0.7562 with the exception of 33).

a

with

S

rel values of DNA complexes of ellipticine derivatives possessing (a) the 2-(

x-amino)alkyl tether, 13–16 and 19

(

x

-amino)alkylcarbamoyl tether, 28–34 (y = 0.4919x ꢁ 5.5572, relation coefﬁcient

3

. Conclusion

In conclusion, we successfully synthesized three types of pyrido-

dsDNA than the corresponding 5-tethered compounds. The salt and

pH effects on the binding constants were examined and the kinetic

parameters for the formation of the complex were also determind.

carbazol-2-ium derivatives 17, 19, and 20 and determined the bind-

ing modes of the DNA-complexes of four types of simple ellipticine

Finally, the binding constant K

rel of the complex, which was determined by the molecular model-

ing, as good linear relationships between lnK and Srel. Contrary to

our expectation, there was no linear relationship between lnK and

a

was related to the relative stability

S

derivatives 1–4 and thirty-six types of the

conjugates of 6H-pyrido[4,3-b]carbazole derivatives 5–40, which

were substituted with various types of -aminoalkyl and -urei-

x

-amino and

x

-ureido

a

x

IC50 against Sarcoma-180, NIH3T3, and HeLa S-3 cell lines. The

results obtained in this study will be useful to develop cytotoxic

pyridocarbazole derivatives as drugs. In order to develop a water-

soluble un-quaternized ellipticine analogue which is one of the

promising candidates of anti-cancer drugs, synthesis and cytotoxic

activity of the PEG esters of 5- and 6-carboxilic acid derivatives of

pyridocarbazole are under investigation together with develop-

ment of the 2-substituted 1,2-dihydroellipticine derivatives as the

pro-drugs of ellipticine. In addition, binding constants of the com-

plexes for our ellipticine and pyridocarbazole derivatives with

RNA:DNA hybrids will be determined in the near future.

doalkyl tether on the N-2 atom and the corresponding alkylcar-

bamoyl tether group on the C-5 atom. The binding modes of these

compounds were determined by deﬂection spectroscopy and were

a

all intercalation, and their binding constants K , which were deter-

mined by QCM, widely varied depending on the substitution posi-

tion, the structure of the terminal group of the tethers, and the

5

7

ꢁ1

sequence of DNA from 1.7 ꢀ 10 to 4.5 ꢀ 10 M . In most case,

the binding constants for the complexes of the GC-rich dsDNA were

larger than those of the AT-rich dsDNA and the compounds possess-

ing the 2-(

x-aminoalkyl) tether more strongly bound to the GC-rich

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

15

chromatography (TLC) with hexane-ethyl acetate as the irrigating

system and UV light at shorter wavelengths as the visualizer. Chro-

matography was performed using SiO

ODS). Melting points were determined using a Yanaco micro-

melting-point apparatus and uncorrected values were reported.

2 2

and octadecylsilylated SiO

(

1

13

H and C NMR spectra were recorded using a JNM ECP-500

500 and 125 MHz) and a JNM ECP 300 (300 and 75 MHz), respec-

(

1

13

tively. The H, C chemical shifts were referenced to TMS (d 0.00)

using the solvent residual peak as a reference and J values are

reported in Hz. Mass spectra were recorded on a JEOL JMS-MS

7

00 mass spectrometer using xenon (6 kV, 10 mA) as the fast atom

bombardment (FAB) gas. The dried DNA ﬁlms were characterized

with an Olympus polarization microscope CX31-P equipped with

crossed nicols and a digital camera. The ultraviolet–visible (UV–

vis) light absorption spectra were measured with a JASCO V-550

spectrophotometer. The binding constant was determined with

Ò

the aid of an ULVAC initium AFFINIX Q 27 MHz quartz crystal

microbalance (QCM).

Fig. 11. Linear relationship of lnK

ellipticine derivatives possessing the 5-(

y = 1.8899x ꢁ 21.318, relation coefﬁcient r = 0.8101).

a

with Srel values of the DNA complexes of

x

-ureido)alkyl tether, 21–25

(

4

.2. Synthesis

4

. Experimental

.1. Materials and methods

All solvents and chemicals were purchased from TCI, Nacalai

.2.1. Synthesis of pyridocarbazole derivatives 1–16, 18, 21–40

1

8a

18a

Ellipticine (1),

2-methylellipticine (2),

6-methylellipticine

4

3),¹

8a

9-hydroxyellipticine

(4),

18b

5-ethoxycarbonyl-2,11-

(

18a

dimethyl-6H-pyrido[4,3-b]carbazol-2-ium chloride (5),

and 9-

1

8a

hydroxy-2-methylellipticin-2-ium chloride (6)

synthesized. 2-( -Aminoalkyl)- and 2-(

were reportedly

x-ureidoalkyl)-11-

tesque, and Wako and used without further puriﬁcation. Reaction

progress and compound purity were monitored using thin-layer

x

methyl-6H-pyrido[4,3-b]carbazol-2-ium chloride derivatives 7–16

Table 13

Cytotoxic activity IC50 of the pyridocarbazole derivatives 1–5, 8–19, 21–35, 37, 38, and 40 against Sarcoma-180, NIH3T3, and HeLa S-3 cell lines.^a

Compd

R²

R⁵

R⁶

R⁹

IC50 (lM)

Sarcoma-180

NIH3T3

HeLa S-3

1

2

3

4

5

8

9

None

CH

3

H

CH

H

CH

H

CH

H

OH

H

2.0 ± 0.7^b

0.7 ± 0.2^b

0.5 ± 0.1

0.30 ± 0.08^b

2.0 ± 0.5^b

Not tested

0.9 ± 0.2^b

1.2 ± 0.4^b

Not tested

5.0 ± 0.2

2.1 ± 0.5^b

CH

3

2.90 ± 0.02^b

0.93 ± 0.09

Not tested

None

3

b

0.44 ± 0.07

1.0 ± 0.3^b

Not tested

2.2 ± 0.3

2.7 ± 0.6

Not tested

1.12 ± 0.47

0.8 ± 0.1

1.0 ± 0.4

Not tested

1.4 ± 0.1

1.2 ± 0.3

0.52 ± 0.6

3.2 ± 0.5

>100^b

1.2 ± 0.4

b

CH

3

COOCH

3

(CH

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

)

2

5

2

5

2

O(CH

2

)

2

NH-CONHCH

3

CH

3

CH

3

CH

3

CH

3

H

75.3 ± 1.6^c

c

OH

H

29.4 ± 4.3

c

10

11

12

13

14

15

16

17

18

19

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

37

38

40

3

1.6 ± 0.3

6.2 ± 0.2

NHCONHCH

3

Not tested

1.75 ± 0.78

8.8 ± 1.0

32.5 ± 1.1^c

41.3 ± 1.6^c

O(CH

2

)

2

)

2

)

2

)

2

NH

2

COOCH

3

c

CH

3

CH

3

CH

3

CH

3

H

27.8 ± 2.5

OH

H

OCH

H

2.4 ± 0.4

17.7 ± 0.8^c

c

3

Not tested

1.87 ± 0.58

0.78 ± 0.14

Not tested

>100^b

17.9 ± 2.9

NH

NH(CH

2

3

19.3 ± 1.0^c

Not tested

5.8 ± 1.8^c

2

) NH

2

COOCH

3

NH

2

3

2

CH

3

6.58 ± 0.69

b

CH

3

CONH(CH

)

2 2

)

2 3

)

2 4

)

2 5

)

2 2

)

2 2

)

2 2

)

2 2

)

2 3

)

2 4

)

2 5

)

2 2

)

2 2

)

2 2

)

2 2

)

2 2

)

2 2

)

2 2

NHCO-NHCH

3

>100

b

>22^b

3

34.0 ± 0.6

31 ± 2

>88^b

>100^b

96 ± 2

b

>100^b

80 ± 10

37 ± 4

O(CH

NH(CH

NHCO-NHCH

2

)

2

-NHCONHCH

3

41 ± 4^b

18 ± 5^b

Not tested

2

)

2

-NHCONHCH

3

78 ± 2^b

45 ± 6^b

>100^b

32 ± 4

b

3

Not tested

b

NH

2

55 ± 7^b

70 ± 20

>100

>26

b

26.0 ± 0.8

42 ± 2

>100^b

21 ± 7^b

32 ± 9^b

>100^b

b

>100^b

O(CH

NH(CH

NH

2

)

2

-NH

2

48.0 ± 0.8

60 ± 10

2

)

2

>100^b

86.0 ± 8^b

Not tested

b

>100^b

2

OCH

H

OCH

H

3

58 ± 8

Not tested

11.0 ± 0.6^b

Not tested

1.40 ± 0.05^b

Not tested

15 ± 3^b

15 ± 1

b

None

NHCO-NHCH

3

7.6 ± 0.7^b

3

Not tested

2.2 ± 0.1^b

Not tested

b

NH

2

2.4 ± 0.4

1.04 ± 0.06^b

2

OCH

a

b

c

MTT assay was used and IC50 values are reported as the means of three experiments; the standard deviations are given together.

Quoted from Ref. 18b.

Quoted from Ref. 18a.

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

1

6

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

4.2.4. Synthesis of N-protected 2-(2-((2-aminoethyl)amino)ethyl)-5-

(

methoxycarbonyl)-11-methyl-6H-pyrido[4,3-b]carbazol-2-ium

chloride derivatives 46a,b using 43a and 43b

A ﬂame-dried 100 mL round-bottom ﬂask containing a mag-

netic stirring bar was charged with 45 (0.100 g, 0.342 mmol),¹

DMF (1 mL), and 43a (0.759 g, 3.42 mmol). The reaction mixture

was stirred at 110 °C for 15 h and cooled to rt, and then the

residue was dissolved in MeOH (5 mL). To the solution, NaOMe

8a

(

0.028 g, 0.513 mmol) in MeOH (5 mL) was added drop by drop

and the reaction mixture was stirred at rt for 1 h. Subsequently,

.11 g (0.513 mmol) of 44 in methanol (5 mL) was added and

0

stirring was continued at rt for 24 h. After the volatile materials

were removed under reduced pressure, the crude residue was

puriﬁed by ODS column chromatography (H O/MeOH = 7:3) to

2

yield 46a (4%). When 43b (0.68 g, 3.42 mmol) was similarly used

instead of 43a, 46b was obtained in a 10% yield.

Fig. 12. A scatter plot of lnK

pyridocarbazole derivatives vs. IC50 values against HeLa S-3. The pyridocarbazoles

–3, 5, 8–16, 18, 19, 23, 27, 35, 37, 38, and 40 were used. Minimum IC50 values (>22,

26 and >100) for 21, 22, 24, 26, 28, 29, 30, 31, 32, and 34 were not used.

a

for the complexes of the GC-rich dsDNA with the

+

ꢁ

4

6a; decomp. 220 °C; MS (FAB ): m/z 477 (M ꢁCl ).

1

2

6b; decomp. 220 °C; H NMR (500 MHz, D O): d 8.65 (1H, s, 1-

1

>

H), 8.30 (1H, d, J = 7 Hz, 3-H), 7.54 (1H, d, J = 7 Hz, 4-H), 7.11 (1H, d,

J = 7 Hz, 7-H), 6.94 (1H, t, J = 7 Hz, 8-H), 6.70 (1H, J = 7 Hz, 9-H),

6

.58 (1H, d, J = 7 Hz, 10-H), 5.54 (2H, s, CHCH), CH

2 2

CH ), 4.12

and 18, which were substituted at the N-2 atom with a various

(3H, s, OCH ), 4.22 (2H, t, J = 5.5 Hz, 3.75 (2H, t, J = 5.5 Hz, CH CH ),

3

2

alkyl-tether containing an ether linkage, were also reportedly

3 2 2

3.69 (3H, s, OCH ), 2.94 (2H, t, J = 5.5 Hz, CH CH ), 2.68 (2H, t,

1

8b

synthesized.

2,11-Dimethyl-6H-pyrido[4,3-b]carbazol-2-ium

2 2 3 3

J = 5.5 Hz, CH CH ), 2.07 (3H, s, CH ), 2.00 (6H, s, CH ); MS (FAB):

+

ꢁ

+

ꢁ

chloride derivatives 21–34 and 11-methyl-6H-pyrido[4,3-

b]carbazole derivatives 35–40, which were substituted at the C-5

m/z 455 (M ꢁCl ); HRMS (FAB): Calcd for C₂₈H₃₁N O (M ꢁCl ):

4

2

455.2477, found: 455.2422.

atom with

a

various 5-(x-aminoalkyl)carbamyl or 2-(x-

ureidoalkyl)carbamyl tether containing an ether linkage, were

also reportedly synthesized.¹

8b

2

4.2.5. Deprotection of the NH group of 46a,b; synthesis of 2-(2-((2-

aminoethyl)amino)ethyl)-5-(methoxycarbonyl)-11-methyl-6H-pyrido

4,3-b]carbazol-2-ium chloride (17)

Deprotection of the NH group of 46a: A round-bottom ﬂask was

charged with 46a (20.0 mg, 0.044 mmol) and 4 M HCl (10 mL). The

reaction mixture was stirred at rt for 1 h. After the volatile materi-

als were removed under reduced pressure, puriﬁcation of the crude

[

2

4

.2.2. Synthesis of tert-butyl (2-((2-chloroethyl)amino)

ethyl)carbamate (43a)

To a cold solution of N -(2-chloroethyl)ethane-1,2-diamine

1

_d_i_h_y_d_r_o_c_h_l_o_r_i_d_e₍₄₂₎₍₅_.₀₀_g_,₂_.₆₀_m_m_o_l₎₃4 in CHCl

(30 mL) were

successively added triethylamine (0.50 g, 5.19 mmol) and Boc

0.55 g, 2.60 mmol) in CHCl drop by drop at 0 °C, and stirring

was continued at room temperature (rt) for 2 h. The mixture was

quenched with water and the organic layer was dried (Na SO

and concentrated. The crude residue was puriﬁed by ﬂash chro-

matography (CH Cl :MeOH:Et N = 200:10:1) to yield the com-

pound 43a (15%) as an oil; H NMR (500 MHz, CDCl

2H, t, J = 8 Hz, CH Cl), 3.61 (2H, t, J = 8 Hz, CH NH), 3.31 (2H, t,

J = 6 Hz, CH NH), 3.27 (2H, J = 6 Hz, CH NHBoc), 1.34 (9H, s, C

3

2

residue by ODS column chromatography (H O:MeOH = 80:20) gave

2

O

1

7 (96%).

(

3

Deprotection of the NH

2

group of 46b: A round-bottom ﬂask

equipped with

(

a

reﬂux condenser was charged with 46b

20.0 mg, 0.044 mmol), NH OHÅHCl (15 mg, 0.220 mmol), KOH

6.00 mg, 0.110 mmol), i-PrOH (2 mL), and H O (0.6 mL). The reac-

2

4

)

2

3

1

tion mixture was reﬂuxed for 40 h and acidiﬁed with 1 M HCl (pH

), and then concentrated under reduced pressure. Puriﬁcation of

3

): d 4.26

3

(

2

the crude residue by ODS column chromatography (H O:

2

MeOH = 80:20) gave 17 (73%).

+

(

CH

3

)

3

); MS (FAB ): m/z 223 (M+H).

1

7; decomp. 210 °C; H NMR (500 MHz, D

2

O): d 9.00 (1H, s, 1-

H), 8.69 (1H, d, J = 5 Hz, 3-H), 8.03 (1H, d, J = 5 Hz, 4-H), 7.60 (1H,

d, J = 6 Hz, 7-H), 7.23 (1H, t, J = 7.5 Hz, 8-H), 7.04 (1H, t, J = 7.5 Hz,

4

.2.3. Synthesis of 2-chloroethyl-(2-(2,5-dimethyl-1H-pyrrol-1-yl)

9

3

-H), 6.95 (1H, d, J = 6 Hz, 10-H), 4.50 (2H, t, J = 6 Hz, CH

.86 (3H, s, OCH ), 3.20 (2H, t, J = 6 Hz, CH CH ), 3.03 (2H, t,

CH ), 2.92 (2H, t, J = 6 Hz, CH CH ), 2.49 (3H, s, CH );

MS (FAB ): m/z 377 (M ꢁCl ); HRMS (FAB ): Calcd for

2 2

CH ),

ethyl)amine (43b)

3

2

1

To a solution of N -(2-chloroethyl)ethane-1,2-diamine dihy-

drochloride (42) (0.30 g, 1.56 mmol)³⁴in EtOH (7 mL) were succes-

sively added triethylamine (0.44 g, 4.37 mmol), acetic acid (0.18 g,

J = 6 Hz, CH

2

3

+

ꢁ

+

ꢁ

C

22

H

25

N

4

O

2

(M ꢁCl ): 377.1978, found: 377.1975.

3

.12 mmol), and 2,5-hexandione (0.53 g, 4.68 mmol) drop by drop

at 0 °C under Ar atmosphere, and stirring was continued at rt for

0 min. After the volatile materials were removed under reduced

pressure, the resulting residue was dissolved in CHCl and succes-

sively washed with brine, an aqueous solution of NaHCO , and

again brine. The organic layer was dried (Na SO ) and concen-

trated. The crude residue was puriﬁed by ﬂash chromatography

CH Cl :MeOH:Et N = 200:10:1) to yield the compound 43b (84%)

as an oil; H NMR (500 MHz, CDCl ): d 5.69 (2H, s, CH), 3.90 (2H,

t, J = 8 Hz, CH Cl), 3.61 (2H, t, J = 6 Hz, CH NH), 2.92 (2H, t,

J = 6 Hz, CH N), 2.84 (2H, J = 8 Hz, CH NH), 2.10 (6H, s, CH ); MS

FAB ): m/z 201 (M+H).

3

4.2.6. Synthesis of tert-butyl (2-hydroxyethyl)carbamate (48)

To a cold solution of 2-aminoethanol hydrochloride (47) (6.10 g,

0.10 mol) were successively added pyridine (10 mL) and Boc O

2

(4.60 g, 0.02 mol) drop by drop under argon atmosphere at 0 °C,

and stirring was continued at the same temperature for 30 min.

3

2

4

(

2

3

The mixture was puriﬁed by SiO

2 3

ﬂash chromatography (CHCl :i-

1

35

3

PrOH = 19:1) to yield the compound 48 (90%; lit. 95% ) as an oil;

1

3

H NMR (500 MHz, CDCl ): d 5.13 (1H, br, NH), 3.65 (2H, t,

2

J = 4.5 Hz, CH

CH

2

OH), 3.24 (2H, t, J = 4.5 Hz, CH

2

NH), 3.15 (1H, br,

2

3

+

(

2

OH), 1.41 (9H, s, C(CH

3

)

3

); MS (FAB ): m/z 162 (M+H).

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

17

4

.2.7. Oxidation of 48 by 2-iodoxybenzoic acid (IBX); synthesis of tert-

2

raphy (H O:MeOH = 90:10) to give 19 (92%) as light orange solids;

1

butyl (2-oxoethyl)carbamate (49)

decomp. ꢄ238 °C. H NMR (300 MHz, D

2

O) d 8.66 (1H, s, 1-H), 7.56

(1H, d, J = 7.8 Hz, 3-H), 7.44 (1H, d, J = 7.8 Hz, 4-H), 7.22 (1H, d,

J = 7.8 Hz, 7-H), 6.91 (1H, t, J = 7.8 Hz, 8-H), 6.65 (1H, t, J = 7.8 Hz,

A round-bottom ﬂask was charged with 48 (0.80 g, 10.0 mmol)

3

6

and IBX (2.80 g, 10.0 mmol) under argon atmosphere. DMSO

25 mL) was added to the mixture and then stirring was continued

overnight at rt. The reaction mixture was quenched with water

200 mL) and ﬁltered under reduced pressure. The ﬁltrate was

extracted with ethyl acetate. The organic layer was dried (Na SO

and concentrated under reduced pressure. Puriﬁcation of the crude

residue by SiO column chromatography (CHCl :EtOAc = 7:3) gave

9 (69%; lit. 55% ) as an oil; H NMR (500 MHz, CDCl ): d 9.65 (2H,

br s, HCO), 5.16 (1H, br, NH), 4.06 (2H, s, CH ), 1.45 (9H, s, C(CH ).

(

9-H), 6.56 (1H, d, J = 7.8 Hz, 10-H), 4.46 (2H, t, J = 7.2 Hz, CH

3.41 (2H, t, J = 7.2 Hz, CH ), 2.19 (s, 3H, CH ), 1.89 (3H, s, CH

O) d 145.1, 143.9, 141.6, 132.9, 132.7,

129.9, 128.9, 125.3, 124.1, 121.5, 121.3, 121.0, 120.1, 111.2,

2

),

2

3

);

1

3

(

C NMR (75 MHz, D

2

4

)

+

ꢁ

110.2, 56.6, 39.9, 14.8, 11.8; MS (FAB ): m/z 290 [M ꢁCl ]; HRMS

+

2

3

22 3

(FAB ): Calcd. for C21H N O: 290.1652, Found: 290.1650.

3

7

1

4

3

2

3 3

)

4.2.12. Synthesis of 2-(2-aminoethyl)-11-methyl-6H-pyrido[4,3-b]

carbazol-2-ium chloride (20)

A solution of 53 (12 mg, 0.028 mmol) in conc. HCl (10 mL) was

heated for 24 h under reﬂux and concentrated under reduced pres-

4

.2.8. Reductive amination of 49; synthesis of 2-(2-((2-aminoethyl)

amino)ethyl)-5-(methoxycarbonyl)-11-methyl-6H-pyrido[4,3-b]

carbazol-2-ium chloride (17)

sure. The residue was puriﬁed by ODS column chromatography

A

round-bottom ﬂask was charged with 18 (10 mg,

1

8a

(H

500 MHz, D

.06 (1H, s, 5-H), 6.86 (1H, d, J = 7.7 Hz, 4-H), 6.75 (1H, d,

J = 8.1 Hz, 7-H), 6.46 (1H, d, J = 7.9 Hz, 10-H), 6.29 (1H, dd, J = 7.4,

2

O:MeOH = 90:10) to give 20 (97%). Decomp. ꢄ200 °C; H NMR

0

.023 mmol),

49 (3 mg, 0.023 mmol), molecular sieves 4A, and

(

2

O): d 8.36 (1H, s, 1-H), 7.34 (1H, d, J = 7.7 Hz, 3-H),

MeOH (5 mL) and stirring was continued overnight at rt. To the

reaction mixture, NaBH CN (15 mg, 0.23 mmol) was added at

80 °C, and then stirring was continued at the same temperature

7

3

ꢁ

8

.0 Hz, 8-H), 6.21 (1H, dd, J = 7.4, 8.0 Hz, 9-H), 4.30 (2H, t,

for 1 h. After that, 1 M HCl (appropriate amounts) was added and

allowed to raise the temperature to rt. The reaction mixture was

ﬁltered and the ﬁltrate was concentrated under reduced

pressure. To the residue, 4 M HCl was added and stirred at rt for

+

J = 5.3 Hz, N CH

2

), 3.34 (2H, t, J = 5.3 Hz, CH

O): d 144.9, 144.5, 141.3, 134.7,

30.1, 128.8, 125.4, 124.6, 124.1, 121.0, 119.3, 111.1 102.0, 97.2,

2 2

NH ), 1.85 (3H, s,

1

3

3 2

CH ); C NMR (125 MHz, D

1

5

+

ꢁ

7.0, 49.5, 40.0 14.0; MS (FAB ): m/z 276 (M ꢁCl ); HRMS

1

h. The concentrated reaction mixture was puriﬁed by ODS

+

ꢁ

(

FAB ): Calcd for C18

18

H N

3

(M ꢁCl ): 276.1500, found: 276.1490.

column chromatography (H O:MeOH = 80:20) to give 17 (59%).

2

4

.2.9. Synthesis of N-(2-chloroethyl)acetamide (51)

Trimethylchlorosilane (TMSCl; 1.30 g, 12 mmol) was added to a

solution of 2-methyl-2-oxazoline (0.85 g, 10 mmol) in MeOH

5 mL), and the stirring was continued at rt for 24 h. After the reac-

tion, the mixture was diluted with CHCl (20 mL) and washed with

O (20 mL ꢀ 3). The organic layer was dried (Na SO ) and concen-

trated under reduced pressure to give 51 (82%) as a colorless oil.

4

.3. Determination of the binding mode for the complexes of the

pyridocarbazole derivatives 1–40 with DNA by a combination of a

deﬂection spectroscopy and orientation of the corresponding molecule

in the DNA-based ﬁlm with chain alignment

(

3

H

2

4

4.3.1. Preparation of an oriented DNA-based ﬁlm of pyridocarbazole

3

8

Commercially available ﬁbrous DNA sodium salt (from salmon

1

H NMR (300 MHz, CDCl

CH CH ), 2.02 (3H, s, CH

1.3, 23.1; MS (EI ): m/z 121 (M ).

3

): d 6.15 (1H, br s, NH), 3.60–3.62 (4H, m,

sperm) was dissolved in water (250 mg/6 mL, pH 6.8), and the

solution was used to prepare the oriented DNA-based ﬁlm of pyri-

docarbazole after standing at rt for a month in a dark place. The

pyridocarbazole derivatives were dissolved in water (or a mixed

solvent; methanol/DMSO = 18:2 v/v). The solutions of a pyridocar-

bazole derivative and DNA were mixed to adjust each concentra-

tion to 2 mM and 50 mMbp, respectively (mol fraction of the

13

2

3 3

); C NMR (75 MHz, CDCl ): d 170.4, 43.9,

+

4

.2.10. Synthesis of 2-(2-acetamidoethyl)-5,11-dimethyl-6H-pyrido

[4,3-b]carbazol-2-ium chloride (52)

To a solution of ellipticine (100 mg, 0.36 mmol) and potassium

iodide (179.3 mg, 1.08 mmol) in DMSO (2 mL) was added 51

131.3 mg, 1.08 mmol) in DMSO (0.5 mL), and stirring was contin-

pyridocarbazole = 0.04). An aliquot (20 lL) of the mixed solution

(

was put on a quartz slide plate (76 ꢀ 26 ꢀ 1 mm), which was pre-

ued at 60 °C for 24 h. After the reaction, the mixture was diluted

with ethyl acetate. The suspension was ﬁltered and the solid was

dissolved in water (10 mL) and Amberlite IRA-900 (chloride form)

was added to the solution. The mixture was stirred at rt for 5 h.

After ﬁltration, the ﬁltrate was concentrated under reduced pres-

sure, and the residue was puriﬁed by ODS column chromatography

liminary washed with ultrapure water after soaking in a mixture of

1

2 2

5% H O and 15% HCl at 80 °C for 5 min. The droplet of the solu-

tion was applied to the plate using a pliable glass capillary to pre-

pare a pyridocarbazole-DNA alignment ﬁlm (a brush application

method). The alignment of the ﬁlm was conﬁrmed with a polariza-

tion microscope equipped with crossed nicols and a 530-nm sensi-

tive tint plate. The samples were illuminated with a halogen lamp.

(

MeOH/H

2

O) to give 52 (90%) as orange solids; decomp. ꢄ258 °C.

H NMR (500 MHz, D O): d 8.14 (1H, s, 1-H), 7.25 (1H, m, 3-H),

.91 (1H, m, 4-H), 6.52–6.59 (2H, m, 7,8-H), 6.28 (1H, m, 9-H),

.14 (1H, m, 10-H), 3.98 (2H, m, CH ), 3.38 (2H, m, CH ), 1.75

), 1.52 (3H, s, CH ), 1.26 (3H, s, COCH

125 MHz, CD OD) d 173.7, 146.1, 144.6, 143.0, 133.5, 133.3,

31.6, 129.5, 126.3, 124.7, 122.7, 121.7, 121.0, 120.7, 112.1,

1

2

6

(

1

2

4.3.2. Linear dichroic UV–vis spectroscopy (deﬂection spectroscopy) to

determine the binding mode for the complex of pyridocarbazole with

DNA

1

3

3H, s, CH

3

); C NMR

3

The optical system for the determination of the binding mode

was set in the order of an illuminant, a polarizer, a sample, and a

circular slit 5 mm in diameter. The incident light was polarized

through a Glan-Taylor prism and was radiated from the surface

of the ﬁlm mounted on a quartz or silica plate. The optical polariza-

tion of pyridocarbazoles was evaluated by linear dichroism at the

wavelength of maximum absorbance in the range from 350 to

650 nm. The absorbance data were obtained with and without a

polarizer and collected for the sample at the rotational angles

increased by each 30° increment, using a rotary sample stage. To

+

ꢁ

10.5, 60.6, 41.2, 22.6, 14.8, 11.8; MS (FAB ): m/z 332 (M ꢁCl );

+

22 3

HRMS (FAB ): Calcd for C21H N O: 332.1757, Found: 332.1743.

4

.2.11. Synthesis of 2-(2-aminoethyl)-5,11-dimethyl-6H-pyrido[4,3-

b]carbazol-2-ium chloride (19)

A solution of 52 (19.5 mg, 0.053 mmol) in acidic aqueous

methanol (1 M-HCl/MeOH = 1:1; 10 mL) was heated under reﬂux

for 24 h. The reaction mixture was concentrated under reduced

pressure and the residue was puriﬁed by ODS column chromatog-

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

1

8

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

remove the inﬂuence of heterogeneity in the observed area, the

raw data were reportedly corrected.⁸

identiﬁed the cavity possessing the largest volume as the active

site. The ligand molecule was docked with the receptor molecule

using GOLD program leading to the formation of a DNA-pyridocar-

4

.4. Determination of the binding constant K

a

and kinetic parameters

bazole docking model. The stabilization energy

as a difference between the energy of the complex and the sum of

the energies of the corresponding components (d(GGGCCC) and

the pyridocarbazole derivative). The energy of each molecule

was calculated in the presence of NaCl aqueous solution (50 mM)

as a medium.

DE was calculated

for the DNA-complex of pyridocarbazoles by QCM

2

3

2

4.4.1. Immobilization of DNA to a QCM sensor tip

Immobilization of DNA to a QCM Au sensor tip was reportedly

2

8b

ꢁ5

2

performed.

An Au surface (0.49 ꢀ 10 m ) of the QCM sensor

tip was cleaned with a piranha solution (a mixture of 25%

/75% H SO = 1:3) for 5 min and well rinsed with ultrapure

water several times. The cleaned Au electrode tip was soaked in

mM dithiodipropionic acid solution at rt for 30 min and rinsed

4

.6. Cytotoxic activity

H

2

O

2

4

.6.1. Cell lines and culture

5

Sarcoma-180 and HeLa S-3 cells were provided by Dr. Okada

with ultrapure water. The electrode was successively soaked in

an 1:1 (v/v) mixture of an 100 mg/mL 1-ethyl-3-(3-

dimethylaminopropyl)carbodiimide hydrochloride (EDC) solution

and an 100 mg/mL N-hydroxysuccinimide solution for 15 min, in

a solution of avidin (0.15 mg/mL) for 30 min, and in a solution of

(

Institute for Biological Resources and Functions, National Institute

of Advanced Industrial Science and Technology, Japan), NIH3T3

cells were purchased from Sumitomo Dainippon Pharma Co., Ltd.

The cells were maintained in MEM (Nissui) supplemented with

1

0% FBS (GIBCO), 2% HEPES, 3% NEAA, and 2 mM L-glutamine in a

1

M ethanolamine for 10 min, and ﬁnally rinsed with ultrapure

2

water-saturated atmosphere of 5% CO at 37 °C.

water. The thus-obtained electrode was immersed in an 8 mL of

buffer solution (10 mM Tris-EDTA including 50 mM NaCl, pH 7.8)

in a QCM cell and allowed to stand till the frequency became

4.6.2. Cell viability assay

constant. To the cell, 1

l

L of the biotinylated 30-mer DNA

Cell viability was determined using a 3-[4,5-dimethylthiazol-2-

y]-2,5-diphenyltetrazolium bromide (MTT) assay, which is

solution (100 M) in the same buffer was added and allowed to

l

a

stand till the frequency became constant. After removing the

solution in the cell, the complementary ssDNA was added in the

same manner leading to the formation of the ds DNA.

method for determining cell viability by measuring the mitochon-

drial dehydrogenase action. Cells were seeded in a 96-well cell cul-

ture cluster (Becton Dickinson) at a density of 2 ꢀ 10 cells/mL and

4

cultured 3 h prior to drug treatment. Cells were exposed at 37 °C

for 72 h to pyridocarbazole derivatives. An MTT reagent (Nacalai

tesuque) was prepared at a concentration of 2 mg/mL in Dulbecco’s

PBS without calcium and magnesium, and stored at 4 °C. After

treatment for 72 h, cells were incubated with MTT reagent for

4 h at 37 °C. The plate was centrifuged at 3000 rpm for 10 min,

and the medium was removed. To solubilize the resultant MTT for-

mazan, 200 mL/well of dimethyl sulfoxide (DMSO) was added to

each well followed by thorough mixing with a mechanical plate

mixer. Absorbance at 540 nm was measured on a microplate

reader (MTP-500, CORONA), and the percentage of cell viability

was taken as the percentage of absorbance at 540 nm of pyridocar-

bazole-treated cells against control cells.

4

a

.4.2. Determination of the binding constants K for the complexes of

pyridocarbazoles 1–40 with the AT- and GC-rich dsDNAs by QCM

To the QCM cell was charged 8 mL of the buffer solution (1 mM

phosphate-EDTA including 50 mM NaCl, pH 7.4) and the DNA-

immobilized sensor tip was immersed and allowed to stand till

the frequency became constant. Aliquot of the aqueous stock solu-

tion (10 mM, 8

and the time course of the frequency was recorded. After the fre-

quency became constant, next 8 L of the compound stock solution

was added to the cell recording the frequency time course. The

same operation was repeated till the frequency did not change

any more. The frequency decrease

the mass increase of 0.606 ng cm in the 27 MHz QCM.

lL) of the pyridocarbazole derivative was added

l

D

F of 1.00 Hz corresponds to

ꢁ

2

Acknowledgements

4.4.3. Analysis of the kinetics for the complex formation of

pyridocarbazoles 2, 6, and 34 with the AT- and GC-rich dsDNAs using

QCM

To the QCM cell was charged 8 mL of the buffer solution (1 mM

phosphate-EDTA including 50 mM NaCl, pH 7.4) and the DNA-

immobilized sensor tip was immersed and allowed to stand till

This work was partially supported by a Grant-in-Aid for Scien-

tiﬁc Research from MEXT (Ministry of education, culture, sports,

science and technology – Japan), a matching fund subsidy from

MEXT 2009–2011 (No. 21590025) and 2012–2014 (No.

2

4590025); a Grant from the Japan Private School Promotion Foun-

the frequency became constant. The aqueous solution (8

lL) of

dation (2008–2009); and a grant for the Development of strategic

Research Center in Private Universities supported by MEXT (Center

for Technologies against Cancer, CTC), 2009–2013. We would like

to thank Dr. Okada (Institute for Biological Resources and Func-

tions, National Institute of Advanced Industrial Science and Tech-

nology) for kindly providing the HeLa S-3 and Sarcoma-180 cells.

We thank Mr. Yusuke Nagatuka, Ryota Mori, Tomokazu Hiratuka,

Kousuke Komenoi, Tetsuya Tsukamoto, Toru Iijima, Dr. Yuri Okuno,

and Satoko Kiuchi (Department of Pure and Applied Chemistry,

Tokyo University of Science) for their partial experimental works

on the synthesis of the compounds.

the pyridocarbazole (0.1, 0.4, 0.6, 0.8, and 1.0 M, respectively)

l

was added with stirring and the time course of the frequency

was recorded.

4.5. Molecular modeling for the complexes of the pyridocarbazoles 13–

1

6, 19, 21–25, 28–34 with dsDNA oligomer d(GGGCCC)

2

Docking of the complex was performed on Discovery Studio

Modeling ver. 1.7 (Accelrys) running on an IBM computer. The

coordinate data of the dsDNA oligomer d(GGGCCC) were obtained

2

from PDB. The structures of the molecules used in the calculation

were optimized after removal of the other molecules such as

water. In docking experiment, GOLD (a program for calculating

the docking modes of small molecules into protein binding sites)

was used. The active site of the receptor was automatically

searched from the cavities of the receptor molecule structure and

Supplementary material

Supplementary data associated with this article can be found, in

the online version, at http://dx.doi.org/10.1016/j.bmc.2016.12.031.

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

A. Shimazu et al. / Bioorganic & Medicinal Chemistry xxx (2017) xxx–xxx

19

(

1

c) Charcosset JY, Jacquemin-Sablon A, Le Pecq JB. Biochem Pharmacol .

984;33:2271–2275 .

References

1

2

7. Moody DL, Dyba M, Kosakowska-Cholody T, Tarasova NI, Michejda CJ. Bioorg

Med Chem Lett . 2007;17:2380–2384 .

8. (a) Mori R, Kato A, Komenoi K, et al. Eur J Med Chem . 2014;82:16–35 ;

1

.

(a) Stiborová M, Stiborová-Rupertová M, Bo rˇ ek-Dohalská L, Wiessler M, Frei E.

Chem Res Toxicol . 2003;16:38–47 ;

(

b) Stiborová M, Breuer A, Aimová D, Stiborová-Rupertová M, Wiessler M, Fre E.

(

b) Kato A, Nagatsuka Y, Hiratsuka T, et al. Tetrahedron . 2016;72:4258–4272 .

9. Sato N, Kawai Y, Akaba Y, Honma S, Sakai N, Konakahara T. Heterocycles .

016;92:664–679 .

0. (a) Auclair C, Voisin E, Banoun H, Paoletti C, Bernadou J, Meunier B. J Med Chem .

984;27:1161–1166 ;

b) Bernard S, Schwaller MA, Lévi G, Aubard J. Biospectroscopy .

996;2:377–389 ;

c) Bernadou J, Meunier G, Paoletti C, Meunier B. J Med Chem . 1983;26:574–579 .

1. (a) Bruekelman SP, Leach SE, Meakins GD, Tirel MD. J Chem Soc Perkin Trans .

984;1:2801–2807 ;

b) Quiclet-Sire B, Quintero L, Sanchez-Jimenez G, Zard SZ. Synlett .

003;2003:75–78 ;

c) Wuts PGM, Greene TW. Greene’s Protective Groups in Organic Synthesis . 4th

Int J Cancer . 2003;107:885–890 ;

(

c) Kouadio K, Chenieux JC, Rideau M, Viel C. J Nat Prod . 1984;47:872–874 ;

d) Poljakova J, Hrebackova J, Dvorakova M, et al. Neuroendocrinol Lett .

(e) Gopal M, Shahabuddin MS. Indian J Med Res . 2004;119:198–205 ;

(f) Garbett NC, Graves DE. Curr Med Chem Anti-cancer Agents . 2004;4:149–172 ;

(g) Andrews WJ, Panova T, Normand C, Gadal O, Tikhonova IG, Panov KI. J Biol

Chem . 2013;288:4567–4582 ;

h) Kohn KW, Waring MJ, Glaubiger D, Friedman CA. Cancer Res .

975;35:71–76 .

(a) Herraiz T, Chaparro C. Biochem Biophys Res Commun . 2005;326:378–386 ;

2

2

.

(b) King RS, Teitel CH, Kadlubar FF. Carcinogenesis . 2000;21:1347–1354 ;

(c) Wu J, Cui G, Zhao M, Cui C, Peng S. Mol BioSyst . 2007;3:855–861 ;

(d) Xiao S, Lin W, Wang C, Yang M. Bioorg Med Chem Lett . 2001;11:437–441 .

ed. Wiley; 2007. 796 .

22. Dodin G, Schwaller MA, Aubard J, Paoletti C. Eur J Biochem . 1988;176:371–376 .

23. Monnot M, Mauffret O, Simon V, et al. J Biol Chem . 1991;266:1820–1829 .

24. Ismail MA, Sanders KJ, Fennell GC, Latham HC, Wormell P, Rodger A.

Biopolymers . 1998;46:127–143 .

3

Recently, Nagesh et al. reported interaction of 1-methylindole 1,2,3-triazole

analogues with quadruplex DNA (Raju G, Srinivas R, Reddy MD, Reddy CR,

Nagesh N. Nucleosides Nucleotides Nucleic Acids . 2014;33:489–506).

Goodwin S, Smith AF, Horning EC. J Am Chem Soc . 1959;81:1098–1903 .

(a) Ohashi M, Sugikawa E, Nakanishi N. Jpn J Cancer Res . 1995;86:819–827 ;

4

5

.

25. (a) Feigon J, Denny WA, Leupin W, Kearns DR. J Med Chem . 1984;27:450–465 ;

(

1

(

b) Schwaller MA, Aubard J, Auclair C, Paoletti C, Dodin G. Eur J Biochem .

989;181:129–134 ;

c) Bailly C, OhUigin C, Rivalle C, Bisagni E, Hénichart J-P, Waring MJ. Nucleic

(

b) Kim JY, Lee SG, Chung J-Y, et al. Toxicology . 2011;289:91–102 .

Schwaller MA, Allard B, Lescot E, Moreau F.

995;270:22709–22713 .

(a) Singh MP, Hill GC, Peoch D, Rayner B, Nabach IJL, Lown JW. Biochemistry .

994;33:10271–10285 ;

6

.

J

Biol Chem .

1

Acids Res . 1990;18:6283–6291 .

7

2

6. Ren J, Qu X, Dattagupta N, Chaires JB. J Am Chem Soc . 2001;123:6742–6743 .

7. Morii et al. reported that the oriented DNA-based ﬁlms were able to be

prepared from both Calf thymus DNA and Salmon sperm DNA, and that there

were little practical differences between them (Morii N, Kido G, Suzuki H,

Nimori S, Morii H. Biomacromolecules . 2004;5:2297–2307; Morii N, Kido G,

Suzuki H, Morii H. Biopolymers . 2005;77:163–172). Furthermore, Salmon sperm

DNA was more inexpensive than Calf thymus DNA. Therefore, we employed the

former as the biopolymer .

8. (a) Niikura K, Matsuno H, Okahata Y. Chem Eur J . 1999;5:1609–1616 ;

b) Maftouh M, Besselievre R, Monserrat B, et al. J Med Chem . 1985;28:708–714 ;

c) Ferlin MG, Marzano C, Gandin V, Dall’Acqua S, Dalla VL. Chem Med Chem .

2

009;4:363–377 .

8

9

.

Morii N, Kido G, Konakahara T, Morii H. J Phys Chem B . 2005;109:15636–15644 .

(a) Tamura S, Konakahara T, Komatsu H, Ozaki T, Ohta Y, Takeuchi H.

Heterocycles . 1998;48:2477–2480 ;

(

b) Ikeda R, Iwaki T, Iida T, et al. Eur J Med Chem . 2011;46:636–646 ;

c) Ikeda R, Kurosawa M, Okabayashi T, et al. Bioorg Med Chem Lett .

2

b) Okahata Y, Niikura K, Sugiura Y, Sawada M, Morii T. Biochemistry .

998;37:5666–5672 ;

c) Okahata Y, Kawase M, Niikura K, Ohtake F, Furusawa H, Ebara Y. Anal Chem .

998;70:1288–1296 .

9. (a) Velayudhan A, Horváth C. J Chromatogr A . 1986;367:160–162 ;

b) Kao Y-H, Fitch CA, Bhattacharya S, Sarkisian CJ, Lecomte JTJ, García-Moreno

2

011;21:4784–4787 ;

d) Ikeda R, Kimura T, Tsutsumi T, Tamura S, Sakai N, Konakahara T. Bioorg Med

Chem Lett . 2012;22:3506–3515 .

0. (a) de Melo LC, Braga SF, Barone PMVB. J Mol Graph Model . 2007;25:912–920 ;

b) Monks NR, Blakey DC, East SJ, et al. Eur J Cancer . 2002;11:1543–1552 .

1. (a) Woodward RB, Iacobucci GA, Hochstein FA. Am Chem Soc .

959;81:4434–4435 ;

b) Ruckdeschel JC, Modi SP, Hamouly WE, Portuese E, Archer S. J Med Chem .

992;35:4854–4857 ;

c) Ishikura M, Hino A, Yaginuma T, Agata I, Katagiri N. Tetrahedron .

999;56:193–207 ;

(

1

2

3

(

J

B. Biophys J . 2000;79:1637–1654 .

0. (a) Furusawa H, Nakayama H, Funasaki M, Okahata Y. Anal Biochem .

2016;492:34–42 ;

(

b) Ebara Y, Itakura K, Okahata Y. Langmuir . 1996;12:5165–5170 ;

c) Kazunori M, Yoshio O. Chem Lett . 1996;25:119–120 ;

d) Matsuura K, Ebara Y, Okahata Y. Thin Solid Films . 1996;273:61–65 ;

e) Matsuura K, Ebara Y, Okahata Y. Langmuir . 1997;13:814–820 ;

f) Okahata Y, Matsuura K, Ebara Y. Supramol Sci . 1996;3:165–169 ;

g) Okahata Y, Matsuura K, Ito K, Ebara Y. Langmuir . 1996;12:1023–1026 .

1

(

d) Díaz M, Cobas A, Guitián E, Castedo L. Eur J Org Chem . 2001;4543–4549 ;

e) Knölker H-J, Reddy KR. Chem Rev . 2002;102:4303–4427;

f) Ho T-L, Hsich S-Y. Helv Chim Acta . 2006;89:111–116 ;

g) Dra cˇ ínsk y´ M, Sejbal J, Rygerová B, Stiborová M. Tetrahedron Lett .

3

1. Alternatively, the reciprocal of the relaxation time (1/

slope of the plot of ln{ )} against time t according to the

following equation derived from Equation 5. ln{ )} = (1/ )t.

s) was obtained from the

2

007;48:6893–6895 ;

h) Liu C-Y, Knochel P. J Org Chem . 2007;72:7106–7115 .

2. Konakahara T, Kato A, Kurasaki H, Sakai N. Jpn Kokai Tokkyo Koho (2006) JP

006321724; Chem Abstr . 2006;146:728 .

3. Konakahara T, Kiran YB, Okuno Y, Ikeda R, Sakai N. Tetrahedron Lett .

010;51:2335–2338 .

4. (a) Moinet-Hedin V, Tabka T, Poulain L, et al. Anticancer Drug Des .

000;15:109–118 ;

b) Tylinska B, Jasztold-Howorko R, Mastalarz HK, Szczaurska-Nowak, Wietrzyk

max t

Dmmax/(Dm – Dm

1

3

2. Konakahara T, Komatsu H, Sakai N, Gold B. J Comput Chem Jpn . 2007;6:27–32.

3. When the IC50 values of the un-quaternized 5-tethered pyridocarbazole

derivatives 35, 37, 38, and 40 against HeLa S-3 were plotted against the

2

binding constants K

a

for the corresponding complexes with the GC-rich dsDNA,

a

good linear relationship was obtained (y = 5.1941x + 76.05, relation

2

(

coefﬁcient r = 0.9388).

1

3

4. N -(2-Chloroethyl)ethane-1,2-diamine dihydrochloride (42 ) was synthesized

J. Arch Pharm . 2008;341:351–356 .

5. (a) Moody DL, Dyba M, Kosakowska-Cholody T, Tarasova NI, Michejda C. J

Bioorg Med Chem Lett . 2007;17:2380–2384 ;

from 2-((2-aminoethyl)amino)ethan-1-ol dihydrochloride (41 ) and thionyl

chloride under reﬂux for 2 h in a 90 % yield according to the reported method

(

1

Hansen GR, Burg TE. J Heterocyclic Chem . 1967;4:653–656).

5. de Napoli L, Evidente A, Piccialli G, Santacroce C, Vellone R. Phytochemistry .

990;29:701–703 .

(

b) Garbett NC, Graves DE. Curr Med Chem Anti-Cancer Agents . 2004;4:149–172 ;

c) Liu X-F, Guan W-C, Ke W-S. Can J Chem . 2007;85:157–163 .

1

6. (a) Anderson WK, Gopalsamy A, Reddy PS. J Med Chem . 1994;37:1955–1963 ;

b) Lempereur L, Sautereau AM, Tocanne JF, Laneelle G. Pharmacology .

984;33:2499–2503;

36. Frigerio M, Santagostino M, Sputore S. J Org Chem . 1999;64:4537–4538 .

37. Chhabra SR, Mahajan A, Chan WC. J Org Chem . 2002;67:4017–4029 .

38. Saito S, Tamai H, Usui Y, Inaba M, Moriwake T. Chem Lett . 1984;13:1243–1246 .

(

1

Please cite this article in press as: Shimazu A., et al. Bioorg. Med. Chem. (2017), http://dx.doi.org/10.1016/j.bmc.2016.12.031

Article Doi

Determination of binding modes and binding constants for the complexes of 6H-pyrido[4,3-b]carbazole derivatives with DNA

DOI: 10.1016/j.bmc.2016.12.031

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.bmc.2016.12.031

Full text of DOI:10.1016/j.bmc.2016.12.031

Products guided by the article

R&D Labs maybe for 5590-29-4

Relevant to this article

Hot Product