Interaction with plasmid DNA of Hoechst-TACN conjugates

Supramolecular Chemistry

ISSN: 1061-0278 (Print) 1029-0478 (Online) Journal homepage: https://www.tandfonline.com/loi/gsch20

Interaction with plasmid DNA of Hoechst-TACN

conjugates

Massimo Tosolini, Teresa Gianferrara, Giuliana Mion, Luca Dovigo, Fabrizio

Mancin, Claudia Sissi & Paolo Tecilla

To cite this article: Massimo Tosolini, Teresa Gianferrara, Giuliana Mion, Luca Dovigo, Fabrizio

Mancin, Claudia Sissi & Paolo Tecilla (2019): Interaction with plasmid DNA of Hoechst-TACN

conjugates, Supramolecular Chemistry, DOI: 10.1080/10610278.2019.1699657

To link to this article: https://doi.org/10.1080/10610278.2019.1699657

Published online: 09 Dec 2019.

Submit your article to this journal

Article views: 20

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=gsch20

SUPRAMOLECULAR CHEMISTRY

https://doi.org/10.1080/10610278.2019.1699657

ARTICLE

Interaction with plasmid DNA of Hoechst-TACN conjugates

Massimo Tosolini^a, Teresa Gianferrara^a, Giuliana Mion^a, Luca Dovigo^b, Fabrizio Mancin^c, Claudia Sissi^band

d

Paolo Tecilla

^aDepartment of Chemical and Pharmaceutical Sciences, University of Trieste, Trieste, Italy; ^bDepartment of Pharmaceutical and

Pharmacological Sciences, University of Padova, Padova, Italy; ^cDepartment of Chemical Sciences, University of Padova, Padova, Italy;

^dDepartment of Mathematic and Geosciences, University of Trieste, Trieste, Italy

ABSTRACT

ARTICLE HISTORY

Received 24 September 2019

Accepted 19 November 2019

A new family of conjugates between the Hoechst minor groove binder and the TACN metal ion

ligand connected through hydrophobic alkyl or more hydrophilic oxyethyl linkers of diﬀerent

length has been prepared. The linkers are connected to the convex side of the Hoechst skeleton

thus forcing the TACN ligand to exit the minor groove and interact with the phosphate backbone

of DNA. The conjugates preserve the binding mode of Hoechst with an aﬃnity inﬂuenced by the

nature of the linker, the more hydrophobic being the more eﬃcient. Coordination of Cu(II) or Zn(II)

poorly aﬀect these parameters. Nevertheless, the Zn(II) complex bearing a C6 linear alkyl linker

induced a modest but reproducible acceleration of the hydrolytic cleavage of DNA which can be

ascribed to the ability of the conjugate to deliver the hydrolytic subunit close to the DNA

phosphodiester bonds.

KEYWORDS

Synthetic nucleases;

Hoechst; TACN; Zn(II) and

Cu(II) complexes; DNA

cleavage

Introduction

chemists towards the development of ‘artiﬁcial

nucleases’(4) with the ﬁnal aim to develop synthetic

catalysts as new and complementary tools for DNA edit-

ing and new antibiotic and chemotherapeutic drugs(5).

The general design of artiﬁcial nucleases is based on

the exploitation of the Lewis acidity of transition and

lanthanide metal ions and theirs complexes which are

the active catalysts of the phosphodiester bond cleavage

(6, 7). However, reactivity and selectivity of simple com-

plexes are usually modest also because of their low aﬃ-

nity for nucleic acids. During the years, several

approaches have been exploited in order to improve

the reactivity and selectivity of these systems. They were

based on the formation of more active multi-metallic

In Nature, the hydrolytic cleavage of P-O bonds in

nucleic acids, DNA and RNA, is catalysed by a class of

enzymes, named nucleases, which are ubiquitous and

involved in many diﬀerent regulation processes(1).

Nucleases are probably the more proﬁcient enzymes

Nature has evolved and catalyse the cleavage of DNA

with impressive rate acceleration and sequence selectiv-

ity(2). Natural restriction nucleases and engineered

nucleases ﬁnd also broad applications in the ﬁeld of

genomic and biotechnologies as tools for manipulating

DNA(3). Parallel to this, a long-standing and growing

eﬀort have been made by chemists and supramolecular

CONTACT Claudia Sissi

Padova, Italy; Paolo Tecilla

claudia.sissi@unipd.it

ptecilla@units.it

Department of Pharmaceutical and Pharmacological Sciences, University of Padova, via Marzolo 5,

Department of Mathematic and Geosciences, University of Trieste, via Weiss 2, Trieste I-34127, Italy

2

M. TOSOLINI ET AL.

Figure 1. Cartoon of the interaction between metal complex/minor groove binder conjugates and DNA. The linear arrangement of the

minor groove binder/linker/metal complex hampers the interaction of the hydrolytic unit with the phosphates (1a) while the lateral

attachment of the linker to the minor groove forces the metal complex to exit from the groove and favours its interaction with the

phosphates (1b).

catalysts(8) or on the conjugation of the metallic centre to

DNA ligands such as, for example, major groove binders

(9), base pairs intercalators(10) or DNA and PNA frag-

ments (11, 12). Somehow surprisingly, however, minor

groove binders, which are an important class of DNA

ligands(13), have been very little exploited in conjugation

with hydrolytic metal ions(14). Some years ago the group

of Qiao and Zhao reported conjugates between mono(15)

or bimetallic(16) Zn(II) complexes with N-methylpyrrole-

based oligopolyamides minor groove binders which

cleaved DNA with moderate activity and selectivity

towards A-T rich sequences and showed, in the case of

the bimetallic complex, some preference for double-

strand cleavage. More recently, we described a family of

conjugates between (indole)₂or benzofurane-indole

amides and the Zn(II)-triaminocyclohexane complex con-

nected through ﬂexible linkers of diﬀerent length and

nature(17). The conjugates show high aﬃnity for the

DNA minor groove but were practically inactive in the

hydrolytic cleavage of the macromolecule.

The above-cited minor groove binder conjugates

share a common structural topology in which the

metal complex is attached trough a linear spacer to

one extremity of the minor groove binder in a linear

fashion. Molecular modelling of the interaction of these

conjugates with DNA(17) shows the DNA binder tightly

inserted in the DNA minor groove with the linker that

follows the minor groove curvature establishing hydro-

phobic interactions with the DNA. As a consequence, the

metal complex is buried in the DNA minor groove away

from the phosphate moieties (Figure 1(a)) and this con-

formation is unproductive or little productive from

a hydrolytic point of view. A strategy to overcome this

problem is to move the attachment point of the linker

from the extremity to one side of the minor groove

binder. In this way the linker should be forced to exit

the minor groove allowing the interaction of the metal

complex with the phosphate groups (Figure 1(b)). To

text this hypothesis, we designed a new family of con-

jugates using as minor groove binder a derivative of

Hoechst 33258(18), which allows the attachment of the

linker bearing the metal complex at the convex side of

its curved skeleton thus forcing the linker to exit from

the minor groove(19) (Figure 2). Here we report the

synthesis, characterisation and DNA interaction studies

of this new class of compounds.

Results and discussion

Design and synthesis of the Hoechst-TACN

conjugates

The structures of the Hoechst-TACN conjugates investi-

gated in this work are reported in Figure 2. The Hoechst

ligand was chosen because it is a representative minor

groove binder with speciﬁcity to the (A₃T₃)₂sequence

(18) and because it was described the synthesis of

a derivative bearing a carboxylic function on the ﬁrst

benzimidazole ring which is a convenient handle for

connecting the linker bearing the hydrolytic unit(19).

Indeed, the carboxylic function is inserted on the convex

side of the ‘banana shaped’ Hoechst skeleton thus ensur-

ing that the linker attached is forced towards the exter-

nal side of the groove over the phosphate backbone

(Figure 1(b)). As metal ion ligand the triazacyclononane

(TACN) macrocycle was selected because it forms strong

complexes with Zn(II) which are eﬃcient catalysts for the

hydrolytic cleavage of model phosphate esters and plas-

mid DNA(20). The TACN subunit was connected to the

Hoechst skeleton via an amide bond using terminally

SUPRAMOLECULAR CHEMISTRY

3

Figure 2. Structures of the Hoechst-TACN conjugates synthesised.

amino functionalised ﬂexible linkers of diﬀerent lengths

(from 4 to 8 atoms) and polarity (fully carbon or with

some oxygen atoms) to vary the distance of the hydro-

lytic unit from the minor groove binder and, for the more

polar linkers, to decrease the hydrophobic interactions

with the DNA base pairs. Compound 5 diﬀers from the

others because the TACN is connected to the linker via

a second amide bond which confers more rigidity to the

system and may also, upon deprotonation of the amide

nitrogen, be involved in the complexation of the metal

ion. This type of linkage has been described in the pre-

paration of self-assembled hydrolytic catalysts and the

results indicate that the Zn(II) complex maintains its

hydrolytic potency(21).

The general synthetic pathway for the preparation of

compounds 1–5 is reported in Scheme 1. The key inter-

mediate is the Hoechst-COOH derivative which was

prepared in a multistep synthesis starting from 3,4-dini-

trobenzoic acid and 3-amino-5-chloro-2-nitrobenzoic

acid, as reported(19). In parallel, the commercially avail-

able TACN was protected on two amines by statistical

reaction with (Boc)₂O giving the bis-protected macro-

cycle (6) in 50% yield. The non-protected amine of 6

was alkylated with the bromoalkyl(oxy)amines 7–10,

protected at the amino functions as benzyloxycabonyl

derivatives, to give compounds 11–14, or with benzyl-

bromoacetate followed by removal of the benzyl group

by catalytic hydrogenation to give 15, respectively(17).

The carboxylic acid of 15 was then conjugated with the

Cbz-monoprotected cadaverine using EDC and HOBT

as condensing agent giving compounds 16. Removal of

the Cbz protecting group by catalytic hydrogenation

from all the linkers gives the free amine 17–21 which

were coupled with Hoechst-COOH using BOP and HOBT

as condensing agents and DIPEA as a base. The ﬁnal

deprotection of the TACN amines with TFA at room

temperature aﬀorded the target compounds 1–5 as

TFA salts.

The Zn(II) and Cu(II) complexes of the conjugates

1–5 were prepared using a procedure optimised by

us(10) in which the proper amount of the TFA salt of

the conjugates is dissolved in a mixture of MeOH/

4

M. TOSOLINI ET AL.

Scheme 1. Reagents: i: CH₃CN, DIPEA, reﬂux; ii: Benzylbromoacetate, CH₃Cl, TEA, rt; iii: H₂, C-Pd 10%, CH₃OH, rt; iv: EDC, HOBT, DMAP,

DMF, rt; v: H₂, C-Pd 10%, CH₃OH, rt; vi: BOP, HOBT, DIPEA, DMF, rt; vii: CH₂Cl₂/TFA 1:1, rt.

water to obtain a millimolar solution. One equivalent

of Zn(NO₃)₂or Cu(NO₃)₂was then added and the pH

adjusted to about 7 by addition of a water solution of

NaOH (Figure 3). The resulting solutions were directly

used as stock solutions for experiments with DNA.

The formation of the complexes was conﬁrmed by

ESI-MS as illustrated in the case of the Zn(II) complex

of compound 4 in Figure 3. The mass spectra show

the absence of the free ligand at m/z = 663.3 and the

formation of the (4-H⁺)·Zn(II) complex at m/z = 729.3.

The insets of Figure 3 show the good matching

between the experimental (A) and calculated (B) iso-

topic cluster distribution of the metal complex.

Similar results were obtained with all the other

complexes.

Cleavage of plasmid DNA promoted by the

Hoechst-TACN conjugates

Starting from our previous experience with related con-

jugates, the interaction of the TACN conjugates and of

their Zn(II) and Cu(II) complexes with plasmid DNA was

preliminary investigated by agarose gel electrophoretic

experiments. Increasing amounts of the compounds

were added to pBR 322 DNA (24 µM) in 20 mM HEPES,

pH 7.1 and after 30 min incubation at 37°C the mixtures

were loaded on an agarose gel. Figure 4 shows the

results obtained with ligands 1–5 and with their Zn(II)

and Cu(II) complexes.

By increasing the concentration of the conjugates the

bands of the plasmid DNA progressively disappear from

SUPRAMOLECULAR CHEMISTRY

5

b

724

726

728

730

732

734

736

738

a

Figure 3. Formation of the metal complexes of compounds 1–5 (top) and ESI-MS spectra of the complex between compound 4 and

Zn(NO₃)₂(bottom). The insets show the experimental (a) and calculated (b) isotopic cluster distribution for the complex with the

formula (4-H⁺)·Zn(II).

the gel while the nucleic acid appears to be trapped in the

loading wells. This behaviour was already observed with

the (indole)₂/TACH conjugates(17) and was correlated to

an aggregation/precipitation of DNA induced by the bind-

ing of the conjugates to the macromolecule. The eﬃ-

ciency of this DNA fading eﬀect was diﬀerent among the

ﬁve tested Hoechst conjugates and follows the ranking

order: 1 > 5 > 2 ≫ 3 ≈ 4. Signiﬁcantly, it seems to correlate

with the lipophilicity of the linker, with the compounds

bearing only carbon linkers being by far more eﬃcient in

inducing the aggregation of the DNA respect to the com-

pounds bearing oxyethyl linkers. The formation of the

corresponding Cu(II) or Zn(II) metal complexes did not

alter this trend to a signiﬁcant extent. By comparing the

DNA aggregation eﬃciency of the Hoechst/TACN conju-

gates with that of the (indole)₂/TACH conjugates bearing

equivalent linkers these latter are almost twice as eﬃcient

(17). Therefore, the phenomenon is further inﬂuenced, as

expected, by the nature of the minor groove binder and

by the topology of the conjugate.

The observed aggregation eﬀect was reversible. Indeed,

if after the 30 min incubation time at 37°C, the ligands or

their metal complexes are forced to dissociate from the

plasmid by addition of 1% SDS no signiﬁcant impairment

of the DNA electrophoretic mobility was observed. This is

shown in Figure 4 in the case of ligand 5 and its Zn(II) and

Cu(II) complexes and similar results were obtained with all

the other ligands and metal complexes.

After having demonstrated that the conjugates bind

to DNA we investigated the ability of their Cu(II) and

Zn(II) complexes to cleave the macromolecule. To asses

it, we increased the incubation up to 24 h at 37°C and,

after the addition of 1% SDS to the reaction mixture, the

reaction products were resolved by agarose gel electro-

phoresis. In the presence of hydrolytic activity by the

metal complexes, the band of the supercoiled plasmid

DNA should decrease with the concomitant appearance

of the bands of the nicked and, eventually, linear forms.

As shown in Figure 5 for compound 1, the free ligands

showed a modest cleaving ability at the highest tested

6

M. TOSOLINI ET AL.

Figure 4. Agarose gel electrophoresis of pBR 322 DNA (24 µM) after 30 min incubation at 37°C with the indicated concentration of

ligands 1–5 and the corresponding Zn(II) or Cu(II) complexes in 20 mM HEPES, pH 7.1. The panel at the bottom on the right refers to

samples added of 1% SDS before loading. Zn(II) and Cu(II) lanes correspond to plasmid treated with 100 μM metal ion.

concentrations. A concentration-dependent formation

of nicked plasmid was detected also upon plasmid incu-

bation with all Cu(II) complexes of conjugates 1–5.

However, this turned out to be not a speciﬁc cleavage

promoted by the delivery of the TACN complexes

towards the macromolecule. Indeed, as exempliﬁed in

Figure 5 in the case of 1, it is evident that after incuba-

tion with 100 μM Cu(II) the plasmid was completely

nicked whereas, at the same concentration of complex

1-Cu(II), the supercoiled DNA was the most abundant

form. This was conﬁrmed also by kinetic experiment in

which the plasmid was incubated for variable time with

a constant concentration of metal complexes (15 µM) at

37°C (Figure 5, panel B). Thus, it appeared that the com-

plex formation actually reduced the DNA damage.

Comparable results were obtained with all the

complexes.

A diﬀerent picture emerged with the Zn(II) complexes.

Indeed, while the Zn(II) complexes of ligands 2–5 did not

induced any DNA cleavage, complex 1-Zn(II) caused

a modest but reproducible increment of nicked plasmid

which extent was concentration and time-dependent

(Figure 5, panel A and C, respectively). As reported in

Figure 5 in this case the cleavage, although modest, was

detectable even at stoichiometric DNA metal complex

concentrations. A control experiment in presence of

TACN alone (100 μM), TACN-Zn(II) (100 μM) and TACN-Zn

(II) (100 μM) with increasing concentrations of free Hoechst

shows the absence of any DNA cleavage highlighting the

importance of the covalent linkage between the metal

complex and the DNA ligand (Figure 5, panel D).

The diﬀerent DNA cleavage behaviour between the

Cu(II) and the Zn(II) complexes can be the results of the

diﬀerent reactivity of the two metal ions. With Cu(II) the

background reaction is rather fast probably as

a combination of hydrolytic and oxidative processes.

On the contrary, Zn(II) is unable to promote the oxidative

cleavage of DNA and the hydrolytic reaction is rather

slow, almost undetectable in the presence of the metal

ion alone. Therefore, with copper the cleavage is domi-

nated by the background reaction and the partial satura-

tion of the metal ion coordination sites by the

coordination with the TACN ligand results in an inhibi-

tion of the process. On the other hand, the absence of

the background reaction with Zn(II) allows to detect

a small cleavage rate increment. This eﬀect is strongly

dependent on the nature and the length of the linker

and indeed is detectable only with conjugate 1 which

has a long and hydrophobic linker which ensures a tight

binding to DNA and enough conformational freedom to

deliver the chelating unit towards the phosphate back-

bone. Other possible interpretations of this increased

reactivity such as, for example, a proximity eﬀect within

ligand/DNA aggregates in which the metal ion attacks

SUPRAMOLECULAR CHEMISTRY

7

Figure 5. Panel A: Agarose gel electrophoresis of pBR 322 DNA (24 µM) after 24 h incubation at 37°C with increasing concentrations of

1, 1-Zn(II) or 1-Cu(II) in 20 mM HEPES, pH 7.0. In panels B and C, the plasmid was incubated for variable time with 15 µM metal

complexes at 37°C. In Panel D the plasmid was incubated with TACN (100 µM), TACN-Zn(II) (100 µM) and TACN-Zn(II) (100 µM) with

increasing concentrations of Hoechst. Before loading, all samples were added of 1% SDS. S and N refer to the supercoiled and nicked

plasmid DNA, respectively. Zn(II) and Cu(II) lanes correspond to plasmid treated with 100 μM metal ion.

‘intermolecularly’ a phosphate unit of a ﬁlament diﬀer-

ent from that the corresponding Hoechst unit is bound

to, appear less likely because no cleavage activity is

observed with the Zn(II) complexes of 5 and 2, which

are also able to aggregate the DNA, and with the family

of the (indole)₂/TACH conjugates previously investi-

gated(17) which are more eﬀective than 1-Zn(II) in

aggregating the DNA. If an ‘aggregation induced clea-

vage’ would be present, being a non-speciﬁc mechan-

ism, some activity should have been observed also with

the other conjugates. Moreover, cleavage is observed

even at 15 µM 1-Zn(II) concentration at which the aggre-

gation should be not relevant (compare Panel C of

Figure 5 with Figure 4). In any case, this modest but

8

M. TOSOLINI ET AL.

detectable hydrolytic cleavage of DNA promoted by the

complex 1-Zn(II) is a particularly relevant result for us

because it is the ﬁrst evidence of an increased cleavage

eﬃciency of a minor groove binder Zn(II) complex

respect to metal ion alone observed in the families of

conjugates investigated by us.

proﬁle showed multiple transitions corresponding to the

diﬀerent complexes formed in solutions.

We found that the ligand complexation of Zn(II) or

Cu(II) did not aﬀect the binding mode of our conjugate.

To get rid of these problems and to have a better

insight into the DNA binding mode of our conjugate to

DNA, we performed CD titrations where we kept con-

stant the concentration of ctDNA while increasing the

concentration of 1 (Figure 7). Under these conditions,

the ligand is expected to bind ﬁrst to the sites with

Binding of the TACN conjugate 1 to double-stranded

DNA

higher aﬃnity.

A strong positive dichroic signal

In order to properly describe the cleavage results, the

binding of the Hoechst-TACN conjugate 1 to DNA was

further investigated. Up to ligand concentration of 20

µM, the correlation between UV-Vis absorbance and

ligand concentration did not signiﬁcantly deviated

from linearity thus reasonably ruling out severe aggre-

gation of our conjugate in solution. Therefore, in the

following experiments, we used concentrations of 1

below 20 μM at which also the aggregation of DNA

should be negligible (see Figure 4). Nevertheless, UV

titrations of 1 with ctDNA, indicated the occurrence of

multiple equilibria involving species with diﬀerent opti-

cal properties. This was highlighted by the lack of well-

deﬁned isosbestic points and by a multi-phasic variation

of the absorbance vs concentration (Figure 6 panels

A and B). Fluorimetric titrations well paralleled this result

(Figure 6, panels C and D). Indeed, low ctDNA concen-

trations caused a decrement of the intrinsic ﬂuorescence

of 1 that at DNA/1 molar ratio larger than 0.1 it was

followed by a biphasic increment of the emission

intensity.

All the above results well relate with literature data on

the related Hoechst 33258(22). Accordingly, the forma-

tion of not site selective complexes which drive the

aggregation of the DNA can be proposed in the pre-

sence of an excess of the ligand. This ﬁts with the

observed electrophoretic proﬁles. At higher DNA/ligand

ratio, the dye redistributes along the macromolecule

and properly ﬁts within the DNA minor groove. This

binding mode is associated to a remarkable increment

of the ligand ﬂuorescence. Along this step, the ligand

recognises diﬀerent nucleic acid sequences with

a variable aﬃnity (a relevant preference for AT vs GC

sites is reported for the related Hoechst 33258) and this

justify the multi-phasic behaviour of the binding

isotherms.

appeared in the ligand absorption range that was con-

sistent with the insertion of the ligands into the minor

groove of the double helix. The ﬁrst signal transition

apparently reached saturation at a DNA (base pairs)

ratios ≈ 0.05, indicating that the main ligand binding

sites on the DNA span about 20 base pairs. Higher ligand

concentrations produced a further increment of this

signal that ﬁt with the occupancy of the minor groove

at lower aﬃnity sites. Increments of the buﬀer ionic

strength (up to 500 mM NaCl) did not change this pro-

ﬁle, thus conﬁrming that none of these two binding

events are extensively related to electrostatic interac-

tions. Again, this overall picture perfectly match the

one currently available for the interaction of Hoechts

33258 with ctDNA(22).

Noteworthy, the metal complexes 1-Zn(II) and 1-Cu(II)

produced an almost superimposable ﬁngerprint thus

supporting that the coordination of the metal ion does

not rearrange the DNA binding mode. Therefore, we can

conclude that neither the addition of the functionalised

side-chain to the binding moiety nor the formation of

the metal complexes prevent the recognition of the

double helix or change the binding mode to

a signiﬁcant extent.

Inhibition of Topoisomerases activity promoted by

the Hoechst-TACN conjugates

It has been reported that the binding of Hoechst to DNA

can impair its enzymatic processing by enzymes(23).

Thus, we tested if the activity of a class of important

DNA processing enzymes like as Topoisomerases can be

inhibited by our ligands. Figure 8 shows the results

obtained by adding increasing amount of 1 and its

metal complexes to pBR322 (0.125 µg) in the presence

of 1 U of human topoisomerase I or IIα. As expected we

recorded a substantial suppression of Topo I and Topo II

activity by 1 in the low micromolar concentrations

range. Formation of the Zn(II) or Cu(II) metal complexes

did not revert this behaviour although they are modestly

less eﬃcient. This inhibition activity suggests that

This binding model was further supported by the

denaturation proﬁle of a short double helix with

a random sequence acquired in the presence of increas-

ing 1 concentration (Figure 6, panel E). The ligand was

conﬁrmed to strongly bind it and to increase its overall

thermal stability but, consistently, the resulting melting

SUPRAMOLECULAR CHEMISTRY

9

Figure 6. Variation of the absorption (Panel A) or ﬂuorescence spectra (λ_ecc= 356 nm, Panel C) of 1 (15 and 2 µM, respectively) induced

by the addition of ctDNA in 20 mM HEPES, 50 mM NaCl, pH 7.1. Bold solid and dashed lines correspond to the spectra of the free and

fully bound 1, respectively. In Panel B and D, the relative variation of the absorbance or ﬂuorescence recorded at 363 or 460 nm is

reported as a function of the nucleic acid/ligand molar ratio. In Panel E the denaturation proﬁles of a FAM-labelled dsDNA fragment in

the presence/absence of increasing 1 concentration (the concentration of 1 is given in the inset in micromolar units).

conjugate 1 may be cytotoxic because the cytotoxicity

of the Hoechst ligand has been suggested to origin from

the inhibition of this class of enzymes(24). Further stu-

dies will be required to investigate this aspect.

between the conjugates and their Zn(II) and Cu(II) com-

plexes with DNA. The conjugates were designed on the

basis of results obtained with a diﬀerent family of

compounds(17) suggesting that a linear arrangement

between minor groove binder/linker/metal complexes

leads to eﬃcient DNA binding but unproductive from

a hydrolytic point of view because all the structure

remains buried in the groove and the metal ion com-

plex is unable to reach the phosphodiester linkage.

Therefore, the topology of the conjugate was changed

Conclusions

We have here presented the synthesis of a new family

of conjugates between a minor groove binder and

a metal ion ligand and the study of the interaction

10

M. TOSOLINI ET AL.

Figure 7. CD spectra of 88 μM ctDNA alone and in the presence of increasing concentrations of 1 recorded in 20 mM HEPES, pH 7.1.

Bold solid and dashed lines correspond to the spectra of the free and fully bound 1, respectively.

Figure 8. Agarose gel electrophoresis of pBR322 DNA (0.125 µg) in the presence/absence of 1 U of human topoisomerase I or IIα after

1 h incubation at 37°C with increasing concentration of 1, 1-Zn(II) or 1-Cu(II) in reaction buﬀer 1X.

using a derivative of Hoechst as minor groove binder

and attaching the linker laterally so that the metal ion

complex should be able to exit the DNA groove being

free to interact with the hydrolysable site. This hypoth-

esis was veriﬁed at least in part. The new derivatives

bind strongly to DNA preserving the typical mode of

binding of Hoechst and its ability to interfere with the

DNA enzymatic processing of enzymes such as

Topoisomerases. However, despite the lateral attach-

ment of the linker, that should minimise the interaction

of the linker itself with the minor groove, the eﬃciency

of the conjugates to aggregate DNA is strongly inﬂu-

enced by the nature of the linker with the more hydro-

phobic ones being the most eﬀective and this may

suggest, as observed with the (indole)₂/TACH conju-

gates previously investigated(17), a similar trend in

the DNA aﬃnity of the conjugates. On the other hand,

the free, partially protonated, ligand and the metal

complexes contribute almost equally to the binding of

DNA suggesting

a week electrostatic interaction

between the charged moiety and the phosphate back-

bone. From a hydrolytic point of view with the complex

1-Zn(II) a modest but reproducible and real cleavage of

the DNA has been observed. On the contrary, all the

other metal complexes are inactive or in the case of the

Cu(II) complexes less active respect to the free metal

ion. This can be interpreted on the basis that the

hydrolytic eﬀect is small and in the case of copper(II)

the reaction is dominated by a background redox reac-

tivity of the free metal ion while in the case of the less

active Zn(II) ion some reactivity inﬂuenced by the nat-

ure and length of the linker can be evidenced.

Apparently to ensure cleavage the linker needs to be

long and hydrophobic possibly to allow the metal ion

SUPRAMOLECULAR CHEMISTRY

11

to reach the phosphodiester bond and to increase the

interaction with the DNA. In any case this result is

particularly interesting because it demonstrates that

the design is correct in order to properly deliver the

chelating unit towards the macromolecule and that

new development are possible to increase activity and

to overcome the inhibition eﬀect observed with Cu(II).

We have, however, to underline that the activity

observed is low in particular when compared with

similar systems based on intercalating ligands instead

of minor groove binders(10). This may suggest an

active role of the DNA ligand in inﬂuencing the clea-

vage activity of the conjugates. Studies in this direction

are underway to better deﬁne the role of the nature of

the DNA ligand and of the structure of the conjugate in

determining its DNA hydrolytic ability.

1,4,7-triazacyclononane(26) (11) and 4,7-N,N-di-tert-

butyloxy

carbonyl-1-N-(6-aminohexyl)-1,4,7-triazacyclononane

(23) (17) were prepared as reported.

Synthesis

General synthesis of 4,7-N,N-di-tert-Butyloxycarbonyl

-1-N-[ω-(N-benzyloxycarbonyl)-aminoalkyl/alkyloxy]-

1,4,7-triazacyclononane derivatives (12-14)

To a solution of 172 mg of diBoc-TACN (6, 0.5 mmol) in dry

CH₃CN (15 mL) were added under argon ﬂux 1.6 eq. of the

appropriate amino bromo derivative (8–10) and 140 μl of

N-diisopropylethylamine (DIPEA, 0.8 mmol). The obtained

mixture was reﬂuxed in a closed vial for 5 days and the

reaction monitored by TLC (CH₂Cl₂/MeOH/NH₃99:1:1).

Then, the solvent was removed under reduced pressure,

the residue dissolved in CHCl₃(30 mL) and washed three

times with a 5% solution of NaHCO₃(10 mL). The organic

phase was dried over Na₂SO₄, concentrated and the crude

residue was puriﬁed by ﬂash chromatography (SiO₂, CH₂

Cl₂/MeOH/NH₃99:1:1) to aﬀord the ﬁnal compounds.

Experimental

Materials and general methods

All commercially available reagents were purchased

from Aldrich and Alfa Aesar and used without puriﬁca-

tion unless otherwise mentioned. Solvents were pur-

chased from Aldrich, VWR, Fluka and Riedel, and

deuterated solvents from Cambridge Isotope

Laboratories and Aldrich. Reactions were routinely mon-

itored by thin-layer chromatography (TLC) on silica gel

(precoated F₂₅₄Merck plates). Chromatography was

performed on Merck silica gel 60F-254 (230 ÷ 400

Mesh) and the solvents employed were of analytical

grade. NMR spectra were recorded on a Jeol 400 spec-

trometer (operating at 400 MHz for proton and at 100

MHz for carbon) or on Jeol X-270 (operating at 270 MHz

for proton and at 67.8 MHz for carbon). Chemical shifts

(δ) are reported in ppm using the solvent residual

signal as an internal reference. Coupling constants (J)

are quoted in Hz. Electrospray mass spectra were

recorded on a Bruker Esquire 4000 ESI-MS instrument,

and compounds were dissolved in methanol.

Infrared spectra (IR) were measured on a Jasco FT/IR-

200 instrument. Hoechst-COOH(19), di-tert-butyl

4,7-N,N-di-tert-Butyloxycarbonyl-1-N-[4-(N-benzyloxy-

carbonyl)- aminobutyl]-1,4,7-triazacyclononane (12).

Yield 70%; Rf = 0.5 (CH₂Cl₂/MeOH 9.5:0.5); ¹H-NMR,

(CDCl₃) δ: 1.40–1.50 (m, 22H), 2.44 (m, 2H), 2.58 (m,

4H), 3.19–3.24 (m, 6H), 3.38–3.46 (m, 4H), 5.07 (s, 2H),

7.33 (m, 5H); ¹³C-NMR (CDCl₃), δ: 25, 28, 28.5, 40.6,

49.1–55.8 (Boc-N-(CH₂)₂-N-Boc, linker-N-(CH₂)₂-N-Boc, -

CH₂-N-TACN), 66.4, 79.4, 127.8–128.5 (Ph), 136, 155.4.

ESI-MS (m/z): 535.6 [M + H⁺];

4,7-N,N-di-tert-Butyloxycarbonyl-1-N-[2-[2-(N-benzy-

loxycarbonyl-2-aminoethoxy)ethoxy]ethyl]-1,4,7-tria-

zacyclononane (13). Yield 38%; Rf = 0.6 (CHCl₃/MeOH

1

9:1); H-NMR, (CDCl₃) δ: 1.45 (s, 18H), 2.67–2.74 (m, 6H),

3.18–3.24 (m, 4H), 3.40–3.56 (m, 14H), 5.10 (s, 2H), 7.34 (m,

5H); ¹³C-NMR (CDCl₃), δ: 28.5, 41.4, 49.7–56.4 (Boc-N-(CH₂)₂

-N-Boc, linker-N-(CH₂)₂-N-Boc, -CH₂-N-TACN), 66.5, 69.8,

69.9, 79.7, 128.1–128.7 (Ph), 136.3, 156.2. ESI-MS (m/z):

595.4 [M + H⁺];

-[1,4,7]-triazacyclononane-1,4-dicarboxylate

(diBoc-

TACN, 6)(25), 1,4-bis(tert-butyloxycarbonyl)-1,4,7-tria-

zacyclononane-7-acetic acid (15)(22), N-benzyloxy

carbonyl-6-bromo-1-hexylamine (7)(10), N-benzyloxyc

arbonyl-4-bromo-1-butylamine (8)(10), 2-[2-(N-benzyl

oxycarbonyl-2-aminoethoxy)ethoxy]-1-bromo-ethane

(9)(17), 2-(N-benzyloxycarbonyl-2-aminoethoxy)-1-bro

mo-ethane (10)(17), 4,7-N,N-di-tert-butyloxycarbonyl-

1-N-[6-(N-benzyloxycarbonyl)-aminohexyl]-

4,7-N,N-di-tert-Butyloxycarbonyl

-1-N-[2-(N-benzyloxycarbonyl-2-aminoethoxy)ethyl]-

1,4,7-triazacyclononane (14). Yield 89%; Rf = 0.4 (CH₂

1

Cl₂/MeOH 9.5:0.5); H-NMR, (CDCl₃) δ: 1.45 (s, 18H), 2.-

68–2.74 (m, 6H), 3.17–3.25 (m, 4H), 3.35–3.48 (m, 10 H),

5.10 (s, 2H), 7.35 (m, 5H); ¹³C-NMR (CDCl₃), δ: 28.6, 41.0,

49.6–56.1 (Boc-N-(CH₂)₂-N-Boc, linker-N-(CH₂)₂-N-Boc, -

12

M. TOSOLINI ET AL.

CH₂-N-TACN), 66.6, 69.7, 79.5, 128.0–128.5 (Ph), 136.6,

/MeOH/NH₃90:10:1) to aﬀord the ﬁnal compounds 22–

26 in moderate yield.

156.4. ESI-MS (m/z): 551.5 [M + H⁺], 573.5 [M+ Na⁺].

Hoechst-C6-diBoc-TACN, (22). Yield 38%; Rf = 0.45

(CH₂Cl₂/MeOH 7:3); ¹H-NMR, (CD₃OD) δ:1.33–1.40 (m,

22H), 1.56 (m, 2H), 1.74 (m, 2H), 2.36 (s, 3H,), 2.38–2.45

(m, 6H), 2.64 (m, 4H), 3.03–3.09 (m, 4H), 3.22 (m, 4H), 3.56

(m, 2H), 6.94 (d, 2H), 7.13 (s, 1H), 7.62 (d, 1H), 7.65 (s, 1H),

7.96 (d, 2H), 7.98 (d, 1H), 8.27 (s, 1H); ¹³C-NMR (CD₃OD), δ:

168.4, 161.6, 157.5, 157.3, 157.2, 155.9, 153.9, 149.2,

129.8, 125.5, 122.8, 122.6, 121.8, 117.1, 115.4, 81.1, 81.0,

57.8, 57.7, 57.6, 56.0, 55.1, 54.9, 54.8, 54.7, 54.5, 51.7, 51.4,

51.2, 50.8, 50.5, 50.3, 46.0, 40.5, 30.4, 29.4, 29.1, 28.7, 28.3,

28.2. ESI-MS (m/z): 879.7 [M + H⁺], 901.6 [M+ Na⁺].

Di-tert-butyl-7-[2-oxo-2-[5-(N-benzyloxycarbonyl)-

1,5-butyldiamine]ethyl]-[1,4,7]-triazacyclononane

-1,4-dicarboxylate (16)

To a 5 mL DMF solution of 1,4-bis(tert-butyloxycarbonyl)

-1,4,7-triazacyclononane-7-acetic acid (15, 50 mg, 0.13

mmol) were added EDCI (27 mg, 0.14 mmol), HOBt

(22 mg, 0.14 mmol) and DMAP (17 mg, 0.14 mmol). After

stirring at room temperature for 10 min 37 mg of Cbz-

Cadaverine (0.15 mmol) dissolved in 3 mL of DMF were

added and the stirring was maintained at room tempera-

ture for 2 days monitoring the reaction by TLC (CH₂Cl₂

/MeOH 95:5). Then, the reaction mixture was diluted with

50 mL of EtOAc and the organic phase was washed with

water (3 x 30 mL), dried and evaporated at reduced

pressure. The crude residue was puriﬁed by ﬂash chroma-

tography (CH₂Cl₂/MeOH 95:5) to aﬀord 56 mg of pure 16

Hoechst-C4-diBoc-TACN, (23). Yield 29%; Rf = 0.56 (CH₂

Cl₂/MeOH 7:3); ¹H-NMR, (CD₃OD) δ:1.32 (s, 18H), 1.67–1.74

(m, 4H), 2.44 (s, 3H), 2.58(m, 6H), 2.75 (m, 4H), 3.10–3.17 (m,

4H), 3.26 (m, 4H), 3.59 (m, 2H), 6.95 (d, 2H), 7.17 (s, 1H), 7.64

(d, 1H), 7.67 (s, 1H), 7.96 (d, 2H), 7.99 (d, 1H), 8.27 (s, 1H);

¹³C-NMR (CD₃OD), δ: 168.4, 161.6, 157.5, 157.4, 157.2,

155.9, 154.1, 149.1, 129.9, 125.4, 122.8, 122.7, 121.8,

117.1, 115.5, 81.1, 81.0, 57.6, 55.9, 55.3, 55.1, 55.0, 54.8,

54.7, 52.1, 51.4, 51.2, 51.0, 50.3, 49.8, 45.7, 40.5, 28.7, 28.4,

26.8. ESI-MS (m/z): 851.6 [M + H⁺], 873.7 [M+ Na⁺].

1

(yield 72%). Rf = 0.15 (CH₂Cl₂/MeOH 9.5:0.5); H-NMR,

(CDCl₃) δ: 1.47 (m, 24H), 2.63 (m, 4H), 3.15–3.48 (m, 14H),

5.06 (s, 2H), 7.33 (m, 5H); ¹³C-NMR (CDCl₃), δ: 171.8, 156.8,

156.7, 156.0, 155.9, 136.9, 128.7, 128.3, 128.2, 80.4, 80.1,

80.0, 77.4, 66.5, 63.5, 63.1, 54.9, 54.0, 53.8, 51.5, 50.7, 50.1,

48.9, 48.2, 40.7, 38.8, 29.7, 29.2, 28.5, 23.7. ESI-MS (m/z):

606.5 [M + H⁺].

Hoechst-TEG-diBoc-TACN, (24). Yield 25%; Rf = 0.22

1

(CH₂Cl₂/MeOH 7:3); H-NMR, (CD₃OD) δ: 1.31–1.36 (m,

General synthesis of conjugates between

18H), 2.39 (s, 3H), 2.44 (m, 6H), 2.70 (m, 4H), 2.97–3.03

(m, 4H), 3.20 (m, 4H), 3.38 (m, 2H), 3.66 (m, 2H), 3.77 (m,

4H), 3.82 (m, 2H), 6.96 (d, 2H), 7.21 (s, 1H), 7.66 (d, 1H),

7.70 (s, 1H), 7.98 (d, 2H), 8.10 (d, 1H), 8.33 (s, 1H); ¹³C-NMR

(CD₃OD), δ: 168.5, 161.6, 157.5, 157.4, 157.3, 155.8, 153.9,

149.2, 138.0, 129.90 125.4, 122.9, 122.6, 121.8, 117.1,

115.4, 81.0, 80.9, 71.7, 71.5, 70.9, 56.6, 56.0, 55.8, 55.6,

55.2, 54.8, 52.1, 51.4, 51.2, 50.9, 46.0, 40.8, 28.7. ESI-MS

(m/z): 911.6 [M + H⁺], 933.6 [M+ Na⁺].

Hoechst-COOH and diBoc-TACN derivatives (22-26)

To a methanol (5 ml) solution containing 0.15 mmol of

the appropriate Z-protected ligands (11–14, 16)

a catalytic amount (10 mg) of 10% C-Pd was added.

The resulting mixture was poured under a hydrogen

atmosphere overnight. The reaction was monitored by

TLC (CHCl₃/MeOH/NH₃9:1:0.1). Then, the catalyst was

ﬁltered over celite and the solvent removed under

reduced pressure aﬀording the Z-deprotected deriva-

tives 17–21 in quantitative yield. The compounds were

used in the following step without further puriﬁcation

and characterisation.

To a 10 mL DMF solution of Hoechst-COOH x 3HCl

(0.87 mg, 0.15 mmol) BOP (331 mg, 0.75 mmol) and

HOBt (115 mg, 0.75 mmol) were added. After stirring at

room temperature for 10 min under argon ﬂux a 5 mL

DMF solution of 0.15 mmol of the above amines 17–21

and DIPEA (0.200 mL) was added and the stirring was

maintained at room temperature for 2 h monitoring the

reaction by TLC. Then, the reaction mixture was diluted

with 50 mL of CH₂Cl₂and the organic phase was washed

with 5% NaHCO₃(3 x 30 mL) and Brine (3 x 30 mL), dried

and evaporated at reduced pressure. The crude residue

was puriﬁed by ﬂash chromatography (CH₂Cl₂

Hoechst-DEG-diBoc-TACN, (25). Yield 38%; Rf = 0.33

(CH₂Cl₂/MeOH 7:3); ¹H-NMR, (CD₃OD) δ: 1.25–1.33 (m,

18H), 2.39 (s, 3H), 2.54 (m, 4H), 2.69–2.77 (m, 6H), 2.91–3.05

(m, 4H), 3.17–3.20 (m, 8H), 3.63 (m, 2H), 3.75 (m, 4H), 6.95

(d, 2H), 7.17 (s, 1H), 7.65 (d, 1H), 7.68 (s, 1H), 7.97 (d, 2H),

8.03 (d, 1H), 8.30 (s, 1H); ¹³C-NMR (CD₃OD), δ: 168.5, 161.6,

157.4, 157.3, 155.9, 154.0, 149.2, 138.1 129.9, 125.5, 122.8,

121.9, 117.1, 115.5, 103.2, 81.0, 80.9, 72.1, 71.5, 70.8, 57.2,

57.0, 56.8, 56.1, 56.0, 55.7, 55.6, 55.5, 54.8, 52.5, 51.5, 51.2,

50.9, 45.9, 41.0, 28.7, 28.6. ESI-MS (m/z): 867.6 [M + H⁺].

Hoechst-C5-NHCOCH₂-diBoc-TACN, (26). Yield 45%;

1

Rf = 0.30 (CH₂Cl₂/MeOH 7:3); H-NMR, (CD₃OD) δ: 1.33–

1.38 (M, 18H), 1.58 (m, 2H), 1.65 (m, 2H), 1.78 (m, 2H), 2.39

(s, 3H), 2.53 (m, 4H), 2.69 (m, 4H), 3.03 (m, 16H), 3.58 (t,

SUPRAMOLECULAR CHEMISTRY

13

2H), 6.92 (d, 2H), 7.11 (s, 1H), 7.63 (m, 2H), 7.62–7.65 (m,

3H), 8.22 (s, 1H); ¹³C-NMR (CD₃OD), δ: 172.6, 166.7, 159.9,

156.4, 155.9, 155.7, 154.2, 152.3, 147.5, 128.3, 123.8,

120.9, 120.2, 115.5, 113.8, 80.1, 79.9, 79.8, 54.5, 54.3,

54.1, 53.4, 53.4, 50.0, 49.6, 49.5, 48.4, 44.4, 42.3, 39.0,

38.7, 29.0, 28.9, 27.3, 24.4, 17.9, 16.5, 11.7. ESI-MS (m/z):

922.5 [M + H⁺], 944.5 [M+ Na⁺].

7H), 3.18 (m, 6H), 3.57 (m, 4H), 3.67 (m, 2H), 3.75 (m, 2H),

3.79 (m, 4H), 3.87 (m, 2H), 7.00 (d, 2H), 7.30 (s, 1H), 7.64 (s,

1H), 7.83 (d, 1H), 7.96 (d, 2H), 8.23 (d, 1H), 8.34 (s, 1H);

¹³C-NMR (CD₃OD), δ: 166.1, 163.3, 151.6, 150.1, 147.3,

135.7, 133.4, 132.1, 129.9, 124.7, 124.6, 121.6, 115.0,

114.0, 112.5, 112.0, 102.6, 69.6, 67.8, 53.9, 53.2, 48.2,

43.6, 42.2, 39.3, 29.4. ESI-MS (m/z): 667.5 [M + H⁺].

General procedure for N-Boc deprotection (1-5)

0.03 mmol of N-Boc protected derivatives (22–26) were

dissolved in CH₂Cl₂(2 mL) and triﬂuoroacetic acid (2 mL)

was added. The mixture was stirred at room temperature

for 1 h and the reaction monitored by TLC (CH₂Cl₂

/MeOH/NH₃90:10:1). Then, the solvent was removed

under reduced pressure and the residue was co-

evaporated with toluene and ethyl ether several times to

fully remove the TFA residues. The obtained triﬂuoroacetic

salts 1–5 were characterised and utilised without any

further puriﬁcation.

Hoechst-C5-NHCOCH₂-TACN, (5). Yield 94%; Rf = 0

(CH₂Cl₂/MeOH 7:3); H-NMR, (CD₃OD) δ: 1.51 (m, 2H),

1

1.64 (m, 2H), 1.77 (m, 2H), 2.99 (s, 3H), 3.02 (m, 4H),

3.17–3.28 (m, 7H), 3.50 (s, 2H), 3.54 (t, 2H), 3.66 (m, 6H),

3.84 (m, 2H), 6.94 (d, 2H), 7.21 (s, 1H), 7.56 (s, 2H), 7.74 (d,

2H), 7.88 (d, 2H), 8.11 (d, 1H), 8.22 (s, 1H); ¹³C-NMR (CD₃

OD), δ: 172.9, 165.6, 163.2, 151.4, 149.7, 147.3, 135.4,

133.4, 131.9, 129.8, 124.6, 121.7, 117.2, 114.9, 113.9,

112.3, 111.9, 102.2, 63.5, 55.7, 54.4, 53.71, 53.3, 53.2,

49.6, 48.4, 44.7, 43.2, 42.2, 39.3, 39.2, 28.9, 28.5, 24.1,

17.8, 17.3, 11.7. ESI-MS (m/z): 722.5 [M + H⁺].

Hoechst-C6-TACN, (1). Yield 95%; Rf = 0 (CH₂Cl₂/MeOH

7:3); ¹H-NMR, (CD₃OD) δ: 1.40 (m, 2H), 1.54–1.60 (m, 4H),

1.77 (m, 2H), 2.73 (m, 2H), 2.98 (m, 7H), 3.15 (m, 2H),

3.03–3.09 (m, 4H), 3.25 (m, 6H), 3.52–3.57 (m, 6H), 3.65

(m, 2H), 3.82 (m, 2H), 6.96 (d, 2H), 7.21 (s, 1H), 7.56 (s, 1H),

7.73 (d, 1H), 7.89 (d, 2H), 8.14 (d, 1H), 8.22 (s, 1H);

¹³C-NMR (CD₃OD), δ: 165.8, 163.12 151.3, 150.1, 146.9,

135.9, 133.6, 133.2, 129.7, 125.2, 124.3, 121.7, 114.7,

113.8, 112.6, 111.5, 102.4, 55.2, 53.2, 43.0, 42.2, 41.6,

39.3, 29.2, 26.7, 26.6, 24.4. ESI-MS (m/z): 679.5 [M + H⁺].

Electrophoretic procedure for DNA interaction study

Stock solutions of the TACN minor groove binders con-

jugates 1–5 (0.5–1 mmol) were prepared in H₂O/CH₃OH

2:1. The ligand was diluted in the proper solvent mixture,

one equivalent of Zn(NO₃)₂or Cu(NO₃)₂in water was

added and the pH was corrected to 7 by addition of

NaOH. The methanol content in the ﬁnal reaction mix-

tures was always lower than 4%. DNA interaction was

monitored using pBR 322 (Gibco BRL) in 20 mM HEPES,

pH 7.1. Reactions were performed by incubating DNA

(24 μM residues) at 37°C in the presence/absence of

increasing amounts of ligand and corresponding metal

complex for the indicated time. Reaction products were

resolved on a 1% agarose gel in TAE buﬀer (40 mM TRIS

base, 20 mM, acetic acid, 1 mM EDTA). To properly check

for ligand-mediated DNA damages, samples were pre-

pared as above reported but, before loading, the reac-

tion mixture was added of 1% SDS to dissociate the

ligands from DNA. The resolved bands were visualised

by ethidium bromide staining and photographed on

a Geliance apparatus.

Hoechst-C4-TACN, (2). Yield 94%; Rf = 0 (CH₂Cl₂/MeOH

1

7:3); H-NMR, (CD₃OD) δ: 1.73 (m, 4H), 2.83 (m, 2H), 3.0

(m, 7H), 3.18 (m, 2H), 3.27 (m, 4H), 3.55–3.60 (m, 6H), 3.67

(m, 2H), 3.86 (m, 2H), 7.01 (d, 2H), 7.28 (s, 1H), 7.64 (s, 1H),

7.82 (d, 1H), 7.96 (d, 2H), 8.22 (d, 1H), 8.34 (s, 1H);

¹³C-NMR (CD₃OD), δ: 165.9, 163.2, 151.6, 150.3, 147.1,

135.9, 133.4, 129.8, 125.0, 124.6, 121.7, 114.9, 113.9,

112.7, 111.8, 102.6, 65.4, 54.6, 53.3, 48.4, 43.1, 42.2,

41.7, 39.0, 27.1, 21.7, 14.0. ESI-MS (m/z): 651.5 [M + H⁺].

Hoechst-TEG-TACN, (3). Yield 97%; Rf = 0 (CH₂Cl₂

1

/MeOH 7:3); H-NMR, (CD₃OD) δ: 2.78 (m, 2H), 2.95 (m,

4H), 2.99 (s, 3H), 3.19–3.25 (m, 6H), 3.60–3.82 (m, 16H),

6.94 (d, 2H), 7.22 (s, 1H), 7.56 (s, 1H), 7.72 (d, 1H), 7.87 (d,

2H), 8.15 (d, 1H), 8.21 (s, 1H); ¹³C-NMR (CD₃OD), δ: 166.1,

163.2, 151.2, 149.9, 147.1, 135.6, 133.2, 131.9, 129.8,

124.6, 121.4, 117.4, 114.9, 113.9, 112.3, 111.7, 102.5,

69.9, 69.8, 69.6, 67.7, 53.8, 53.2, 48.6, 44.2, 42.8, 42.2,

39.2. ESI-MS (m/z): 711.5 [M + H⁺].

UV-VIS and ﬂuorometric titrations

UV-VIS and ﬂuorometric titrations were performed with

a Perkin-Elmer λ20 and Agilent Cary Eclipse, respectively.

Binding was followed by addition of increasing amounts of

ctDNA to a freshly prepared ligand solution in 20 mM

HEPES, pH 7.1. Experimental data were analysed in terms

of relative variation of the absorbance/ﬂuorescence at

a constant wavelength as a function of DNA concentration.

Hoechst-DEG-TACN, (4). Yield 95%; Rf = 0 (CH₂Cl₂

1

/MeOH 7:3); H-NMR, (CD₃OD) δ: 2.91 (m, 2H), 3.0 (m,

14

M. TOSOLINI ET AL.

Fluorescence melting studies

References

Melting experiments were performed in a Roche

LightCycler, using an excitation source at 488 nm and

recording the ﬂuorescence emission at 520 nm using as

target DNA a double-stranded DNA (5’- GGATGT

GAGTGTGAGTGTGAGG) with Dabcyl at the 5 ′and FAM

at the 3’ end of the complementary strand. Mixtures (20

μL) contained 0.25 μM of target DNA and increasing con-

centrations of tested derivatives (0–10 µM) in 50 mM

potassium buﬀer (10 mM LiOH, pH 7.4 with H₃PO₄,

50 mM KCl). Readings were taken while applying a 0.2°

C/min heating rate. Each curve was repeated at least three

times and errors were 0.4°C.

(1) Yang, W. Q Rev. Biophys.. 2011, 44, 1–93. DOI: 10.1017/

S0033583510000181.

(2) Wolfenden, R. Chem. Rev. 2006, 106, 3379–3396. DOI:

10.1021/cr050311y.

(3) Gaj, T.; Gersbach, C.A.; Barbas, C.F. Trends Biotechnol.

2013, 31, 397–405. DOI: 10.1016/j.tibtech.2013.04.004.

(4) Mancin, F.; Scrimin, P.; Tecilla, P. Chem. Commun. 2012,

48, 5545–5559. DOI: 10.1039/c2cc30952a.

(5) Yu, Z.; Cowan, J. A Curr. Opin. Chem. Biol. 2018, 43, 37–42.

DOI: 10.1016/j.cbpa.2017.10.029

(6) Mancin, F.; Scrimin, P.; Tecilla, P.; Tonellato, U. Chem.

Commun. 2005, 2540–2548. DOI: 10.1039/b418164f.

(7) Mancin, F.; Tecilla, P. New J. Chem. 2007, 31, 800–817.

DOI: 10.1039/b703556j.

(8) Brown, R.S.; Lu, Z.-L.; Liu, C.T.; Tsang, W.Y.; Edwards, D.R.;

Neverov, A. J. Phys. Org. Chem. 2010, 23, 1–15.

(9) Copeland, K.D.; Fitzsimons, M.P.; Houser, R.P.; Barton, J.K.

Biochemistry. 2002, 41, 343–356. DOI: 10.1021/bi011793k.

(10) Boseggia, E.; Gatos, M.; Lucatello, L.; Mancin, F.; Moro, S.;

Palumbo, M.; Sissi, C.; Tecilla, P.; Tonellato, U.; Zagotto, G.

J. Am. Chem. Soc. 2004, 126, 4543–4549. DOI: 10.1021/

ja039465q.

(11) Lönnberg, T.; Aiba, Y.; Hamano, Y.; Miyajima, Y.;

Sumaoka, J.; Komiyama, M. Chem. Eur. J. 2010, 16,

855–859. DOI: 10.1002/chem.200902169.

(12) Zelder, F.H.; Mokhir, A.A.; Krämer, R. Inorg. Chem. 2003,

42, 8618–8620. DOI: 10.1021/ic034896v.

CD measurements

Circular dichroism spectra were recorded at 25°C using

1 cm path length quartz cells on a Jasco J 810 spectro-

polarimeter equipped with a peltier temperature con-

troller. CD spectra were recorded on samples containing

80 µM ctDNA and increasing ligands concentrations in

20 mM HEPES, pH 7.1. The reported spectrum of each

sample represents the average of 3 scans. Observed

ellipticities were converted to mean residue ellipticity

[θ] = deg x cm²x dmol⁻¹(Mol. Ellip.).

(13) Kawamoto, Y.; Bando, T.; Sugiyama, H. Bioorg. Med.

Chem. 2018, 26, 1393–1411. DOI: 10.1016/j.

bmc.2018.01.026.

(14) Wang, J.-T.; Xia, Q.; Zheng, X.-H.; Chen, H.-Y.; Chao, H.;

Mao, Z.-W.; Ji, L.-N. Dalton Trans. 2010, 39, 2128–2136.

DOI: 10.1039/b915392f.

(15) Li, C.; Qiao, R.-Z.; Wang, Y.-Q.; Zhao, Y.-F.; Zeng, R. Bioorg.

Med. Chem. Lett. 2008, 18, 5766–5770. DOI: 10.1016/j.

bmcl.2008.09.074.

(16) Li, C.; Du, C.; Tian, H.; Jiang, C.; Du, M.; Liu, Y.; Qiao, R.-Z.;

Jia, Y.-X.; Zhao, Y.-F. Chem. Eur. J. 2010, 16, 12935–12940.

DOI: 10.1002/chem.201000552.

(17) Sissi, C.; Dovigo, L.; Greco, M.L.; Ciancetta, A.; Moro, S.;

Trzciński, J.W.; Mancin, F.; Rossi, P.; Spalluto, G.; Tecilla, P.

Tetrahedron. 2017, 73, 3014–3024. DOI: 10.1016/j.

tet.2017.04.013.

(18) Spink, N.; Brown, D.G.; Skelly, J.V.; Neidle, S. Nucleic

Acids Res. 1994, 22, 1607–1612. DOI: 10.1093/nar/

22.9.1607.

(19) Tanada, M.; Tsujita, S.; Sasaki, S. J. Org. Chem. 2006, 71,

125–134. DOI: 10.1021/jo051836t.

(20) Bonfá, L.; Gatos, M.; Mancin, F.; Tecilla, P.; Tonellato, U.

Inorg. Chem. 2003, 42, 3943–3947. DOI: 10.1021/

ic034139x.

Topoisomerase inhibition

0.125 µg of pBR322 (Inspiralis) was incubated with

increasing concentrations (0.5–200 µM) of tested com-

pounds for 1 h at 37°C in the presence/absence of 1

U of human topoisomerase I or IIα (Inspiralis) in the

required buﬀer (1X). Reaction products were resolved on

a 1% agarose gel prepared in 1X TAE (10 mM Tris 1 mM

EDTA, 0.1% acetic acid, pH 8.0). After the electrophoretic

run (5 V/cm for about 90 min) the DNA bands were

visualised by ethidium bromide staining, photographed

and quantiﬁed using a Geliance 2000 apparatus.

Acknowledgments

The authors thank dr. Ilaria Furlan for her involvement in a part

of the synthetic work. This work was supported by the MSCA-

ITN MMBIO (721613) project.

(21) Mancin, F.; Prins, L.J.; Pengo, P.; Pasquato, L.; Tecilla, P.;

Scrimin, P. Molecules. 2016, 21, 1014. DOI: 10.3390/

molecules21081014.

Disclosure statement

(22) a) Rahimian, M.; Miao, Y.; Wilson, W.D. J. Phys. Chem. B.

2008, 112, 8770–8778. b) Stokke, T.; Steen,

H. B. J. Histochem. Cytochem. 1985, 33, 333-338; c)

Loontiens, F. G.; Regenfuss, P.; Zechel, A.; Dumortier, L.;

Clegg, R. M. Biochemistry 1990, 29, 9029-9039.

DOI:10.1021/jp801997v.

No potential conﬂict of interest was reported by the authors.

ORCID

Paolo Tecilla

http://orcid.org/0000-0002-0088-1077

SUPRAMOLECULAR CHEMISTRY

15

(23) a) McHugh, M.M.; Woynarowski, J.M.; Sigmund, R.D.;

Beerman, T.A. Biochem. Pharmacol. 1989, 38,

2323–2328. b) Woynarowski, J. M.; Sigmund, R. D.;

Beerman, T. A. Biochemistry 1989, 28, 3850-3855.

DOI: 10.1016/0006-2952(89)90472-3

(24) Baraldi, P.G.; Bovero, A.; Fruttarolo, F.; Preti, D.; Tabrizi, M.

A.; Pavani, M.G.; Romagnoli, R. Med. Res. Rev. 2004, 24,

475–528. DOI: 10.1002/med.20000

(25) Mion, G.; Gianferrara, T.; Bergamo, A.; Gasser, G.;

Pierroz, V.; Rubbiani, R.; Vilar, R.; Leczkowska, A.;

Alessio, E. Chem. Med. Chem. 2015, 10, 1901–1914. DOI:

10.1002/cmdc.201500288.

(26) Bonora, G.M.; Drioli, S.; Felluga, F.; Mancin, F.;

Rossi, P.; Scrimin, P.; Tecilla, P. Tetrahedron Lett.

2003, 44, 535–538. DOI: 10.1016/S0040-4039(02)

02584-4.

Article Doi

Interaction with plasmid DNA of Hoechst-TACN conjugates

DOI: 10.1080/10610278.2019.1699657

Source and publish data:

Authors:

Article abstract of DOI:10.1080/10610278.2019.1699657

Full text of DOI:10.1080/10610278.2019.1699657

Products guided by the article

R&D Labs maybe for 528840-32-6

Relevant to this article

Hot Product