Unravelling the antitumoral potential of novel bis(thiosemicarbazonato) Zn(II) complexes: structural and cellular studies

JBIC Journal of Biological Inorganic Chemistry

https://doi.org/10.1007/s00775-018-1629-6

ORIGINAL PAPER

Unravelling the antitumoral potential of novel

bis(thiosemicarbazonato) Zn(II) complexes: structural and cellular

studies

Elisa Palma^1,2· Hugo M. Botelho³· Goreti Ribeiro Morais^1,4· Inês Rodrigues¹· Isabel Cordeiro Santos¹·

Maria Paula Cabral Campello¹· Paula Raposinho¹· Ana Belchior¹· Susana Sousa Gomes¹· Maria Fátima Araújo¹·

Isabel Correia²· Nadia Ribeiro²· Sofa Gama^1,5· Filipa Mendes¹· António Paulo¹

Received: 27 August 2018 / Accepted: 7 November 2018

Abstract

The development of pharmacologically active compounds based on bis(thiosemicarbazones) (BTSC) and on their coordi-

nation to metal centers constitutes a promising feld of research. We have recently explored this class of ligands and their

Cu(II) complexes for the design of cancer theranostics agents with enhanced uptake by tumoral cells. In the present work,

we expand our focus to aliphatic and aromatic BTSC Zn(II) complexes bearing piperidine/morpholine pendant arms. The

new complexes ZnL¹–ZnL⁴were characterized by a variety of analytical techniques, which included single-crystal X-ray

crystallography for ZnL²and ZnL³. Taking advantage of the fuorescent properties of the aromatic complexes, we investi-

gated their cellular uptake kinetics and subcellular localization. Furthermore, we tried to elucidate the mechanism of action

of the cytotoxic efect observed in human cancer cell line models. The results show that the aliphatic complexes (ZnL¹and

ZnL²) have a symmetrical structure, while the aromatic counterparts (ZnL³and ZnL⁴) have an asymmetrical nature. The

cytotoxic activity was higher for the aromatic BTSC complexes, as well as the cellular uptake, evaluated by measurement of

intracellular Zn accumulation. Among the most active complexes, ZnL³presented the fastest uptake kinetics and lysosomal

localization assessed by live-cell microscopy. Detailed studies of its impact on cellular production of reactive oxygen spe-

cies and impairment of lysosomal membrane integrity reinforced the infuence of the pendant piperidine in the biological

performance of aromatic BTSC Zn(II) complexes.

Keywords Zinc · Bis(thiosemicarbazones) · X-ray difraction · Fluorescence imaging · Cellular studies

Introduction

Thiosemicarbazones (TSC) and bis(thiosemicarbazones)

(BTSC), as well as their metal complexes, have received

considerable attention in the design of new drugs against a

range of diseases due to their broad pharmacological activity

Electronic supplementary material The online version of this

article (https://doi.org/10.1007/s00775-018-1629-6) contains

supplementary material, which is available to authorized users.

3

* Filipa Mendes

BioISI—Biosystems and Integrative Sciences Institute,

fmendes@ctn.tecnico.ulisboa.pt

University of Lisboa, Faculty of Sciences, Campo Grande,

C8, 1749-016 Lisbon, Portugal

* António Paulo

4

apaulo@ctn.tecnico.ulisboa.pt

Present Address: Institute of Cancer Therapeutics, Faculty

of Life Sciences, University of Bradford, Bradford BD7 1DP,

UK

1

Centro de Ciências e Tecnologias Nucleares, Instituto

Superior Técnico, Universidade de Lisboa, Estrada Nacional

10 (km 139,7), 2695-066 Bobadela LRS, Portugal

5

Present Address: Institut für Anorganische und Analytische

Chemie, Friedrich-Schiller-Universität Jena, Humboldtstr. 8,

07743 Jena, Germany

2

Centro Química Estrutural, Instituto Superior Técnico,

Universidade de Lisboa, Av. Rovisco Pais, 1049-001 Lisbon,

Portugal

Vol.:(0123456789)

1 3

JBIC Journal of Biological Inorganic Chemistry

that includes antibacterial, antimalarial, antiviral, and antitu-

moral properties [1–4]. In addition, bis(thiosemicarbazones)

also present a strong potential for cancer theranostics, since

they can form complexes with diferent di- or trivalent met-

als relevant in the biomedical imaging feld. This comprises

complexes with radiometals suitable for in vivo clinical

imaging by PET (positron emission tomography) (e.g.,

⁶⁴Cu or ⁶⁸Ga) or SPECT (single-photon emission computed

tomography) (e.g., ¹¹¹In), as well as fuorescent Zn(II) com-

plexes for preclinical studies using optical imaging modali-

ties (e.g., fuorescence microscopy) [5].

a new family of BTSC copper(II) complexes [14]. These new

compounds were designed based on the functionalization of

the CuATSM scafold with pendant cyclic amines (piperi-

dine and morpholine) introduced at the exocyclic nitrogen

atoms, using appropriate methylenic linkers. We envisaged

that the presence of the cyclic amine side chains could have

a positive efect on the water solubility of the complexes, a

crucial issue in the design of metal-based anticancer drugs,

without compromising their ability to accumulate in tumoral

cells and induce cytotoxic efects. Moreover, the presence

of the piperidine and morpholine rings could also infuence

the intracellular distribution of the com plexes, namely, by

promoting its lysosomal entrapping [15, 16].

The intense research eforts on the chemistry of TSC and

BTSC derivatives and their metal complexes have already

led to achievements with clinical translational potential. The

most prominent example is the Cu(II) complex CuATSM

(ATSM=diacetyl-bis(N⁴-methylthiosemicarbazonate)) that,

when prepared with the positron-emitting ⁶⁴Cu, has been

thoroughly investigated as a PET tracer to image hypoxia

in tumors and cardiac ischemia. After initial clinical tri-

als, ⁶⁴CuATSM was successfully evaluated in patients with

diferent tumors (e.g., lung, cervical, and rectal cancers)

providing useful clinical data [6–8]. More recently, DpC

(di-2-pyridylketone 4-cyclohexyl-4-methyl-3-thiosemicarba-

zone), which belongs to the di-2-pyridylketone thiosemicar-

bazone class of compounds, has also entered Phase 1 multi-

center clinical trials as an anticancer agent to treat advanced

and resistant tumors [3]. Like other TSC derivatives, DpC

has a polypharmacological profle, which comprises the

ability to bind iron and copper ions with consequent gen-

eration of intracellular reactive oxygen species (ROS) that

ultimately lead to lysoso m al membrane permeabilization

(LMP) and cell death [3, 9–13].

Our previous work proved that piperidinyl/morpholinyl

derivatives of CuATSM present a significant cytotoxic

activity against a panel of human tumoral cell lines, showing

a remarkably higher uptake than CuATSM in most tumoral

cells studied [14]. To further evaluate how the presence of

the pendant cyclic amines infuences the cellular uptake,

subcellular localization, and the possible mechanism of

action for this class of BTSC M(II) complexes, we have

extended the studies to Zn(II) congeners. BSTC Cu(II) com-

plexes typically show poor fuorescence properties, since

Cu(II) is a d⁹paramagnetic metal center that tends to quench

the emission of fuorescence [17]. In opposition, BTSC

Zn(II) complexes usually present fuorescence properties

suitable for the follow-up of their cell uptake kinetics and

subcellular localization by fuorescence microscopy [17, 18].

For the studies with the BTSC Zn(II) complexes bear ing

piperidine/morpholine pendant arms, we have focused on

derivatives of ZnATSM and ZnNqTSM (NqTSM=bis[4-

N-methyl-3-thiosemicarbazone] acenaphthenequinone), as

shown in Fig. 1. By studying ZnNqTSM derivatives con-

taining a naphthalene-based aromatic backbone, we expected

Inspired by these encouraging progresses, we have

recently reported the synthesis and biological evaluation of

Fig. 1 General structures of

the Zn^IIATSM and Zn^IINqTS-

M

^asymderivatives

N

Zn

N

H

N

H

S

N

H

N

H

S

Zn^IIATSM

Zn^IINqTSM^asym

N

Zn

Z

n

N

H

N

H

S

N

H

N

S

N

H

=

N

O

or

N

1 3

JBIC Journal of Biological Inorganic Chemistry

compounds with enhanced fuorescence properties and,

therefore, more suited for live-cell imaging studies than the

aliphatic counterparts. Pascu et al. showed previously that

ZnNqTSM and its congener containing allyl groups at the

exocyclic nitrogen atoms have more favorable fuorescence

properties than ZnATSM due to the presence of the fat

and aromatic naphthalene backbone. This feature allowed

an easier monitoring of the cellular uptake and subcellular

localization of the compounds [17].

Results and discussion

Synthesis

The synthesis of Zn(II) complexes with aliphatic

bis(thiosemicarbazone) ligands, namely, ATSM deriva-

tives, is commonly described in the literature by react-

ing the appropriate ligands with Zn(II) salts.⁵By con-

trast, the reported aromatic congeners of the Zn^IINqTSM

type have been obtained by template reactions between

acenaphthenequinone with the desir ed thiosemicarbazide

in the presence of a zinc salt [17, 18]. In our case, the

frst approach was used to obtain bo th the aliphatic or

aromatic Zn(II) complexes; therefore, the work was ini-

tiated with the synthesis of the diferent chelators, i.e.,

ATSM, NqTSM, and the corresponding derivatives (H

₂L¹–H₂L⁴) carrying piperidine and morpholine pendant

arms (Scheme 1).

In this contribution, we describe the synthesis, charac-

terization, and in vitro biological evaluation of this new

family of BTSC Zn(II) complexes. The in vitro biological

evaluation included a variety of studies on a panel of human

tumoral cell line models, namely, the assessment of their

cytotoxic properties and the quantifcation of their cellular

uptake by ICP-MS measurement of intracellular Zn. For the

aromatic complexes, i.e., those derived from Zn^IINqTS-

M

^asym, their biological evaluation was also performed with

live-cell imaging assays to study their cellular uptake kinet-

ics and for subcellular localization. The ability of selected

aromatic compounds to induce lysosomal membrane per-

meabilization (LMP) and generate intracellular ROS was

also evaluated, to gain some insight on their mechanism

of action. As a comparison, some of these studies were

also conducted for the parental compounds, ZnATSM and

ZnNqTSM^asym, searching for a better understanding of the

infuence of diferent structural features (presence and type

of cyclic amine pendant arms or aliphatic vs aromatic ligand

backbone) on the biological performance of the compounds.

The previously reported aliphatic chelators H₂ATSM,

H₂L¹, and H₂L²were obtained by an acetic acid-catalyzed

condensation reaction between 2,3-butanedione and the

respective 4-substituted thiosemicarbazides (1-3), as

described elsewhere [14]. As shown in Scheme 1, the

aromatic counterparts H₂L³and H₂L⁴were synthesized

using a similar methodology by refuxing acenaphthene-

quinone with the corresponding thiosemicarbazide but

in the presence of hydrochloric acid. H₂L³and H₂L⁴are

new compounds that were obtained with moderate yields

Scheme 1 Synthesis of the BTSC chelators and respective Zn(II) complexes. (i) 2,3-butanedione, 5% acetic acid, 60 °C; (ii) acenaphthenequi-

none, HCl 2 N, ethanol, refux; and (iii) Zn(OAc)₂.2H₂O, MeOH

1 3

JBIC Journal of Biological Inorganic Chemistry

of 53 and 65%, respectively. Recently, Pascu et al. have

used a microwave-assisted reaction to prepare aromatic

bis(thiosemicarbazones), including the H₂NqTSM che-

lator that was obtained in a very high yield (95%) [19].

According to the authors, this microwave-assisted proce-

dure presents the advantage of requiring very short reac-

tion times to produce almost quantitative amounts of the

aromatic BTSCs. The shorter reaction time minimizes the

occurrence of side reactions, namely, the cyclization of the

starting thiosemicarbazides that result in the formation of

impurities and render the purifcation of the fnal chelators

more demanding. Although being synthesized by a con-

ventional acid-catalyzed reaction, H₂L³and H₂L⁴could

be purifed in a straightforward way by column chromatog-

raphy and were obtained with reasonable yields. For this

reason, we did not proceed with the optimization of their

synthesis using microwave-assisted reactions.

Each type of CH₂and CH₃groups from the two alkyl link-

ers and two amine rings is also chemically equivalent and

gives raise to resonances integrating to four or six protons,

respectively, being observed a few occasional overlaps. Con-

sistently, the ¹³C NMR spectra of ZnL¹and ZnL²exhibit a

single resonance for each set of equivalent of carbon atoms

of the two thiosemicarbazone units and ethylenic bridge. The

coordination of H₂L¹and H₂L²as dianionic and tetradentate

N₂S₂chelators was further corroborated by the absence of

the resonance due to the acidic NH protons that appear in

the spectra of the free ligands at ca. 10.4 ppm.

Contrarily to the aliphatic compounds, ZnNqTSM^asym

,

ZnL³and ZnL⁴, and the corresponding free ligands

1

(H₂NqTSM, H₂L³, and H₂L⁴) present H and ¹³C NMR

spectra with splitting patterns compatible with an asymmet-

ric structure for the aromatic BTSCs. This is clearly seen

by the presence of two resonances for the NH hydrazone

The new Zn^IIBTSC complexes (ZnL¹to ZnL⁴) were

obtained in moderately high yields (70–76%) by reacting

hydrated zinc acetate with one equivalent of the correspond-

ing ligand in methanol, as depicted in Scheme 1. The same

methodology was used to synthesize the Zn(II) complex of

NqTSM containing an asymmetrically coordinated ligand

(see discussion below). This complex is herein designated as

ZnNqTSM^asymto distinguish from the previously reported

ZnNqTSM complex obtained based on the template method,

which contains a symmetrically coordinated BTSC.¹⁸All

these Zn(II) complexes, except ZnL³, precipitate from the

respective reaction mixtures as yellow–orange microcrystal-

line powders. For ZnL³, the solvent of the reaction mixture

was evaporated in vacuum and the yellow–orange complex

was obtained in a pure form after recrystallization in a mix-

ture of dichloromethane and diethyl ether.

protons in the H NMR spectra of the free ligands H₂L³

1

and H₂L⁴, spanning between 11.30 and 12.86 ppm, and the

presence of two resonances in the range 8.55–9.06 ppm due

to the exocyclic NH protons (Figures S1 and S3). These

results can be explained by the presence of E and Z pairs

for each type of NH protons, as previously described for the

H₂NqTSM chelator [19]. Moreover, H₂L³and H₂L⁴were

obtained exclusively as (E,Z) isomers. The preference for the

(E,Z) geometry might eventually refect the minimization of

the repulsion between the lone pairs from the imino-nitrogen

atoms. The assignment of the diferent NH protons was per-

formed by comparison with the chemical shifts exhibited by

the same protons in H₂NqTSM. For each type of NH group,

the E protons are upfeld shifted when compared with their

Z counterparts (see SI, Figures S1 and S3), as proposed for

H₂NqTSM [19].

The characterization of the BTSC chelators and their zinc

complexes was done by ESI–MS, elemental analysis (CHN),

¹H, and ¹³C NMR and IR spectroscopy (see the experimental

part). Single crystals were obtained for ZnL²and ZnL³,

which allowed the determination of their solid-state struc-

tures by X-ray crystallography analysis.

For the aromatic BTSC Zn(II) complexes (ZnNqTS-

M

^asym, ZnL³, and ZnL⁴), each pair of H1/1’, H2/2’, and

H3/3′ protons and corresponding carbon atoms gives rise

to two independent resonances, except in a few occasional

overlaps, as shown in Fig. 2, for ZnL⁴. This splitting pat-

tern is less evident for the proton and carbon atoms from

the cyclic amine pendant arms which, being farther away

from the metal center, are less afected by the asymmetric

coordination environment around Zn(II). The asymmetric

tetradentate coordination of the aromatic BTSC chelator

through the two sulfur atoms, one inner and one outer imine

nitrogen atoms was confrmed by X-ray structural analysis in

the case of ZnL³, as discussed below. The ¹H and ¹³C NMR

data collected for the aromatic Zn(II) complexes indicate

that this asymmetric coordination mode is retained in solu-

tion for all the compounds.

NMR spectroscopy

For the aliphatic BTSC Zn(II) complexes, ZnL¹and ZnL²,

the splitting pattern and relative areas of the peaks pre-

1

sent in their H NMR spectra show that the BTSC chela-

tors are coordinated in a symmetric way, by the two inner

imine nitrogen atoms and the two thiolate sulfur atoms. This

mode of coordination was confrmed in the X-ray structure

of ZnL², as described below. The symmetric coordination of

the aliphatic chelators in ZnL¹and ZnL²was evident by the

presence of a singlet at 2.18 ppm, integrating for six protons

(CH₃groups in the ethylenic bridge), and a broad singlet at

6.98–7.35 ppm, integrating for two protons (N–H groups).

The H NMR spectra of ZnL³, ZnL⁴, and ZnNqTS-

1

M

^asymshowed well-defned and rather narrow resonances,

whose profle did not change with time, indicating that the

asymmetric Zn(II) complexes do not undergo dynamic

1 3

JBIC Journal of Biological Inorganic Chemistry

Fig. 2 ¹H and ¹³C NMR spectra

of ZnL₄in DMSO-d₆in the

N–H and aromatic regions

processes in solution, contrarily to the symmetric counter-

parts reported earlier by Pascu et al. [17]. Altogether, these

results strongly suggest that the ligands H₂L³, H₂L⁴, and

H₂NqTSM, in the (E,Z) conformation are pre-organized to

coordinate the metal center through one inner and one outer

imine nitrogen atoms, and consequently, the formation of

asymmetric Zn(II) complexes is kinetically favored.

revealed that the complexes exist in the solid state as discrete

monomeric or dimeric neutral molecules (Figs. 3 and 4).

In both compounds, the Zn(II) ion is pentacoor dinated

with a slightly distorted square-pyramidal geometry

(τ = 0.103 and τ = 0.100 for ZnL²and ZnL³, respec-

tively) [20]. In monomeric ZnL², the square-pyramidal

coordination around the Zn²⁺ion is ensured by the N1,

N4, S1, and S2 atoms occupying the basal plane, and by

the oxygen atom O3 from the water molecule at the axial

position (Fig. 3). The Zn atom lies 0.549(1) Å above the

basal plan towards the coordinated oxygen atom. The coor-

dination of the aliphatic bis(thiosemicarbazone) ligand

afords the formation of three fused fve-membered chelate

rings. The dihedral angle between the least-square (l.s.)

plane of the ethylenic bridge C3–C2–C5–C6 and the l.s.

plane of the central chelate ring C2–N1–Zn1–N4–C3 is

5.71(14)º, whereas the dihedral angles with the l.s. planes

X‑ray crystallography and mass spectrometry

Crystals of ZnL²and ZnL³were obtained by slow difusion

of diethyl ether into concentrated methanolic and dichlo-

romethane solutions of the complexes, respectively. A sum-

mary of the crystallographic data is presented in Table S1

and the most relevant bond distances and angles are listed

in Table S2 (SI). Single-crystal X-ray difraction analysis

Fig. 3 ORTEP diagram of ZnL²with thermal displacement ellipsoids at 40% probability level

1 3

JBIC Journal of Biological Inorganic Chemistry

Fig. 4 ORTEP diagram of ZnL³

with thermal displacement ellip-

soids at 40% probability level

of the other two chelate rings, Zn1–S1–C1–N2–N1 and

Zn1–S2–C4–N5–N4, are 10.59(15)° and 28.50(17)°,

respectively (see Table S3, SI). The ZnL²structure is sta-

bilized by a signifcant network of intermolecular hydro-

gen bonds being the most relevant those formed between

the water molecule at the axial position and the nitrogen

atom of the morpholine moiety, O(3)–H(3AW)…N(7),

with a bond length of 2.753(5) Å (Table S4 and Figure

S7, SI).

ZnL²and ZnL³present bond angles and distances

comparable (within the experimental error) to the values

reported for other aliphatic or aromatic Zn^IIBTSC com-

plexes, with the C–S bond lengths (1.735Å–1.749Å) l ying

wit hin the characteristic range for thiolate bonds [17, 18,

23–30]. The exceptions are the M–S and M–N bond lengt hs

in t he asymmetric ZnL³species, with a strong increase of

the equatorial Zn–S2 bond involving the bridging sulfur

atom (2.9679(11) Å) and a slight reduction in the Zn–N

bonds [Zn–N1 2.068(3) Å and Zn–N5 1.982(3) Å]. Prob-

ably, this trend is due to the symmetry breakdown in the

coordination geometry around the Zn atom and to the mini-

mization of repulsive interactions within the dimer. For the

bridging sulfur atom in ZnL³, the diference between its

equatorial and axial Zn–S bonds (2.9679 Å and 2.4002 Å,

respectively) is much larger than the one found for the bridg-

ing sulfurs in a related dimeric structure of a Zn(II) complex

with a tridentate thiosemicarbazone (av. 2.555 Å and 2.410

Å for the equatorial and axial Zn–S bonds, respectively) [30,

31]. The most relevant intermolecular hydrogen bond inter-

actions present in ZnL³are listed in Table S4 (SI).

In the structure of ZnL³, each unit lies close to an inver-

sion center giving rise to a dimeric structure, thereby ensur-

ing a distorted square-pyramidal coordination around each

Zn atom through the bonding to the sulfur (S2#1) atom

from the related symmetric unit at the axial position and to

the S1, N1, N5, and S2 atoms on the basal plane (Fig. 4).

The aromatic bis(thiosemicarbazone) ligand coordinates

to the Zn atoms in a non-symmetrical fashion forming a

system of four-, six-, and fve-membered chelate rings. The

dihedral angles formed by the l.s. plane of the aromatic

acenaphthene moiety and the l.s. planes of the four-, six-,

and fve-membered chelate rings are 12.93(6)°, 1.68(7)°,

and 35.22(4)°, respectively (see Table S3, SI). The value

of 35.22(4)° observed for the dihedral angle involving the

fve-membered chelate ring is bigger than the corresponding

values found in the ZnL²structure, certainly to minimize

the steric constraints around the metal due to the greatest

bulkiness and rigidity of L³[21, 22]. Most probably, this is

the reason justifying the unusual asymmetric coordination

observed for L³, which was reported previously for a related

aromatic Zn(II) bis(thiosemicarbazonato) complex showing

a bulky 1,4-diazabicyclo[2.2.2]octane ligand coordinated at

the axial position [17].

All the Zn(II) complexes (ZnL¹–ZnL⁴) were analyzed

by electrospray ionization mass spectrometry (ESI–MS)

at room temperature, using dilute acetonitrile solutions of

the pure compounds. The obtained ESI–MS spectra in the

positive mode showed peaks and isotopic patterns consist-

ent with the presence of [M+H]⁺or [M+Na]⁺ions corre-

sponding to monomeric species (see experimental section).

We should emphasize that this behavior was also observed

for ZnL³despite the fact that it crystallized in the solid state

as a dimeric compound. The positive ESI–MS spectrum of

ZnL³showed a prominent [M + H]⁺peak at m/z = 613.2

1 3

JBIC Journal of Biological Inorganic Chemistry

with a splitting pattern superimposable to the one calculated

for monoprotonated ZnL³monomers (ESI, Figure S13). The

peak due to monoprotonated dimer adducts at m/z 1229.38

or the peak at m/z 615.19 attributable to the double proto-

nation of the dimeric ZnL³species was not present in the

experimental spectrum.

superimposable, showing a strong absorption band centered

at around 360 nm and a set of weaker bands between 410 nm

and 510 nm. The shape of these spectra and the energy of

their bands are very similar to those described previously

for symmetric ZnNqTSM [18]. The high-energy bands can

be assigned to the transfer of electron density from metal-

based orbitals to the naphthyl moieties of the ligands. The

bands at lower energy are probably associated with π→π*

and n → π* [HOMO–LUMO (type)] transitions involving

the sulfur lone pair and orbitals associated with the diimine

bonds and naphthyl backbone. Complexes ZnL¹and ZnL²

show absorption bands at ca. 315 and 437 nm.

UV/Vis and fuorescence spectroscopy

As mentioned before, other authors showed that BTSC

Zn(II) complexes containing an aromatic backbone, namely,

the symmetric ZnNqTSM, are more efcient fuorophores

than ZnATSM and related aliphatic complexes [17]. There-

fore, we expected the aromatic BTSC Zn(II) complexes

described in this work to present fuorescence properties

suitable for live-cell imaging studies. For a reliable interpre-

tation of the live-cell imaging results (see below), we started

by characterizing the complexes by UV/Vis absorption and

fuorescence spectroscopies.

The three aromatic Zn(II) complexes, ZnNqTSM^asym

,

ZnL³, and ZnL⁴, emit fuorescence in DMSO when irra-

diated at 370 nm that corresponds approximately to λ_max

of the most intense absorption band, giving rise to fuo-

rescence spectra presenting emission bands centered at

563 nm (Fig. 5). The same bands also appear by irradiating

at 445 nm, although less intense (Figure S16, SI). Com-

plexes ZnL¹and ZnL²show very weak fuorescence emis-

sion (see SI).

As shown in Fig. 5, the absorption spectra of ZnNqTS-

M

^asym, ZnL³, and ZnL⁴solutions in DMSO are almost

Fig. 5 UV–visible absorption

spectra (top) and fuorescence

emission spectra (λ_exc370 nm,

r.t., bottom) of the zinc(II) com-

plexes in DMSO (15 µM)

1 3

JBIC Journal of Biological Inorganic Chemistry

Table 1 Photophysical properties of the Zn(II) complexes in DMSO

(τ) for the complexes (Table 1). In summary, the three

aromatic BTSC Zn(II) complexes show fuorescence emis-

sion around 563 nm, with quantum yields between 0.060

and 0.074 and average lifetimes in the range 0.11–0.14 ns,

possessing, therefore, favorable features for fuorescence

microscopy studies.

solution at room temperature

λ_abs/nm (ε/M⁻¹cm⁻¹

)

λ_em/nm τ/ns

Φ^a

0.14 0.060

ZnNqTSM^asym

359 (4.24×10⁴)

402 (1.23×10⁴)

424 (1.23×10⁴)

454 (1.10×10⁴)

483 (1.20×10⁴)

314 (8.25×10³)

438 (8.81×10³)

315 (1.03×10⁴)

437 (1.08×10⁴)

361 (2.99×10⁴)

404 (9.71×10³)

425 (9.60×10³)

456 (9.11×10³)

485 (9.75×10³)

361 (4.12×10⁴)

403 (1.21×10⁴)

424 (1.22×10⁴)

453 (1.14×10⁴)

486 (1.29×10⁴)

560

Biological evaluation

ZnL¹

ZnL²

ZnL³

545

560

n.d

Cytotoxic activity in human tumoral cell lines

To have insight on the antitumoral properties of the new

Zn^IIBTSC complexes, their cytotoxic activities were stud-

ied using a panel of human cancer cell lines, and compared

with those of the respective chelators and ZnATSM. For

comparative purposes, all the compounds were also tested in

the human non-tumoral HEK293 cell line. For these assays,

cells were incubated with increasing concentrations of the

compounds for 48 h at 37 °C, and the cellular viability was

evaluated by the MTT assay. The inhibition of growth (%)

was calculated and the IC₅₀values (i.e., concentration which

reduces the growth by 50%) were determined and are pre-

sented in Table 2.

0.14 0.074

ZnL⁴

561

0.11 0.074

^aMeasured by comparison wit h the emission of quinine sulphate in

0.5 M H₂SO₄(Φ=0.55) [32, 33]

For the aliphatic BTSCs, there is a clear efect of metal

complexation on their cytotoxic profle in most of the cell

lines studied. This trend is in agreement with the results

reported in the literature for other Zn(II) complexes of thio-

semicarbazone-based ligands, with the complexes displaying

in general a signifcantly greater cytotoxicity than the cor-

responding thiosemicarbazones [3].

The quantum yields (Φ) in DMSO were estimated by

the relativ e me thod, using quinine sulphate as the reference

(Table 1) [32, 33]. The values obtained are in good agree-

ment wit h results previously reported for similar compounds,

namely, symmetrical ZnNqTSM [18]. Time-resolved fuo-

rescence enabled the determination of the average lifetime

Nevertheless, the presence of the pendant cyclic amine

groups in the aliphatic BTSC has a negative efect on the

antitumoral activity of the respective Zn(II) complexes for

Table 2 IC₅₀(μM) values of the Zn^IIBTSC complexes and respec-

tive ligands, as determined by the MTT assay after 48 h of incubation

of the compounds at 37 °C with human non-tumoral HEK293 and

tumoral cell lines: ovarian—A2780 and A2780cisR, breast—MCF7,

cervical—HeLa, and prostate—PC3 and LNCap

IC₅₀(µM)

A2780

A2780cisR

MCF7

HeLa

PC3

LNCap

HEK293

ATSM

H₂L¹

60.8 10.7

>200

20.6 4.9

59.5 22.7

>200

100.1 27.9

130.4 70.8

106.7 20.5

17.6 7.8

4.92 0.70

5.55 0.86

24.4 7.1

79.9 35.3

89.5 18.8

11.0 2.5

4.15 1.00

7.08 1.97

112.4 40.6

117.0 39.9

145.0 143.8

59.4 24.7

7.81 2.34

16.1 4.3

14.65 4.62

18.62 4.55

54.7 21.3

13.4 4.3

6.69 2.68

6.38 1.43

9.33 2.96

47.3 35.0

46.5 14.8

3.49 0.43

3.16 1.73

8.46 2.72

25.9 3.99

108.4 44.7

161.0 61.0

1.44 0.46

0.67 0.15

1.87 0.42

2.78 1.49

31.8 13.5

42.4 15.0

0.93 0.17

2.98 0.89

3.97 0.91

H₂L²

H₂NqTSM

36.9 11.0

1.68 0.37

3.00 0.42

6.62 3.07

50.5 32.1

66.8 43.9

2.53 0.78

2.21 0.76

2.51 0.46

35.5 9.23

3.50 1.41

12.3 2.07

8.49 4.81

33.5 12.8

57.8 25.0

4.27 1.36

4.94 1.91

11.9 4.5

4.29 0.88

1.22 0.28

5.68 1.59

8.90 4.07

85.6 57.3

58.8 20,1

4.70 1.12

5.26 1.80

20.6 7.3

H₂L³

H₂L⁴

ZnATSM

10.7 2.8

ZnL¹

47.6 18.6

59.7 26.4

5.79 1.95

4.78 1.43

9.35 4.93

ZnL²

ZnNqTSM^asym

ZnL³

ZnL⁴

1 3

JBIC Journal of Biological Inorganic Chemistry

all the cell lines studied. In fact, ZnL¹and ZnL²present

IC₅₀values that are roughly 4–10 times higher than those

exhibited by the parental ZnATSM. For instance, in A2780

cells, IC₅₀values are 50.5 and 66.8 μM for ZnL¹and ZnL²,

respectively, and 6.62 μM for ZnATSM. In contrast, and as

reported previously by us [14], the congener Cu(II) com-

plexes (CuL¹and CuL²), carrying the same pendant cyclic

amine rings (piperidine or morpholine, respectively), exhib-

ited lower IC₅₀values than CuATSM.

accumulation in given organelles. MCF7 cells were observed

in the fuorescence microscope for 1 h after adding Zn(II)

complexes or DMSO while monitoring the fuorescence

co-localizing with the cells. Figure 6a shows representative

microscopy images of MCF7 cells incubated with 20 μM

of each complex. A control experiment, where cells were

incubated with DMSO, was used to assess background fuo-

rescence and cellular autofuorescence. Figure 6b presents a

quantitative kinetic analysis of the compounds accumulation.

Overall, ZnL³was the complex that presented the fastest

uptake, measured by the detected intracellular fuorescence

that increased steadily for ~ 30 min. ZnL⁴and ZnNqTS-

M^asymdemonstrated a slower and moderate cell uptake

(~ 2,5 times lower than ZnL³), reaching similar levels of

fuorescence at 60 min of incubation.

In general, each aliphatic and aromatic complex shares

rather similar anti-proliferative properties in the tumoral cell

lines when compared with the non-tumoral cell line. How-

ever, regarding the complexes containing the aromatic naph-

thyl core and pendant cyclic amine groups, ZnL³and ZnL⁴,

they presented a much stronger anti-proliferative action than

their aliphatic counterparts, ZnL¹and ZnL². For example,

IC₅₀values of 2.21 0.76 and 2.51 0.46 μM were obtained

in A2780 cells for ZnL³and ZnL⁴, respectively, while

20–25 times higher IC₅₀values were obtained for ZnL¹and

ZnL²(50.5 32.1 and 66.8 43.9 μM, respectively). Com-

plex ZnNqTSM^asymalso showed higher cytotoxic activity

(ca. twofold lower IC₅₀values) than ZnATSM (A2780:

2.53 0.78 vs 6.62 3.07 μM; A2780cisR: 4.27 1.36 vs

8.49 4.81 μM; MCF-7: 4.70 1.12 vs 8.90 4.07 μM).

The same trend was previously reported for the cytotoxic

activities of symmetrical ZnNqTSM derivatives and related

aliphatic congeners [17, 18].

Importantly, when cells were incubated with 10 μM of

ZnL³(corresponding to half the concentration represented

in Fig. 6), the intracellular fuorescence detected was still

higher than for the rest of the complexes at 20 μM (data not

shown). The positive infuence of the pendant cyclic amine

groups in the kinetics of cell uptake, particularly in the case

of the piperidine ring, is consistent with the behavior that

we found previously for Cu(II) complexes (CuL¹and CuL²)

with aliphatic BTSCs [14].

After 30 min of live-cell imaging and particularly in the

case of cells incubated with ZnL³, cells started presenting

signs of membrane blebbing. This suggests the occurrence

of apoptosis, which might lead to membrane permeabiliza-

tion and consequent loss of the intracellular fuorescence

due to ZnL³. This efect is most likely due to the cytotoxic

properties of the complex, as the excitation was short and

performed in mild conditions and the nuclear stain was also

used at a non-toxic concentration. In addition, this efect

was not observed in the DMSO control. Considering that the

evaluated Zn(II) complexes present similar spectral charac-

teristics (see Fig. 5 and Table 1), the diferences observed

for the intracellular fuorescence are most likely the result

of diferent cellular uptake.

Altogether, these dat a corroborate the crucial infuence

of the aromatic backbone on the biological performance of

BTSC Zn(II) complexes. This infuence was also noticed

in the biological activity of the respective free ligands, as

the new aromatic BTSCs (H₂L³and H₂L⁴) are much more

cytotoxic than the aliphatic counterparts (H₂L¹and H₂L²).

Moreover, the aromatic BTSCs and the respective Zn(II)

complexes showed similar IC₅₀values in almost all tested

cell lines. The infuence of the piperidine and morpholine

pendant arms in the cytotoxic activity of the compounds is

less striking than that of the aromatic vs aliphatic backbone.

However, as a general trend, it is evident that the piperidine

groups confer a higher cytotoxicity to the compounds when

compared with the morpholine congeners, both for the aro-

matic and aliphatic systems. This trend is well illustrated

by the highest cytotoxicity values exhibited by ZnL³in dif-

ferent human tumoral cell lines, among all the complexes

carrying pendant cyclic amines.

The observed intracellular distribution of the fluo-

rescence, and hence of the compounds, was diffuse in

the cytoplasm without accumulation in the nucleus. As

mentioned in the introduction, we have hypothesized that

the presence of the protonable cyclic amines could pro-

vide the BTSC Zn(II) complexes with lysosomotropic

properties, as described in the literature for other com-

pounds [15, 16]. Therefore, we next evaluated if the pres-

ence of the piperidine ring could promote the lysosomal

entrapping of ZnL³. The lysosomal-specific fluorescent

probe, lysotracker red DND-99, was used in co-localiza-

tion experiments in live cells incubated with ZnL³and

ZnNqTSM^asym. In accordance with the previous experi-

ment, a stronger green fluorescence was detected in cells

incubated with ZnL³compared with cells incubated with

Live‑cell fuorescence imaging

Taking advantage of the intrinsic fuorescence of the aro-

matic BTSC Zn(II) complexes (ZnNqTSM^asym, ZnL³,

and ZnL⁴), a live-cell fuorescence microscopy assay was

performed to assess their ability to enter human tumoral

cells, visualize their subcellular localization and eventual

1 3

JBIC Journal of Biological Inorganic Chemistry

Fig. 6 Cell uptake kinetics

of the aromatic BTSC Zn(II)

complexes in MCF7 cells,

as measured by fuorescence

microscopy. a Representa-

tive images of MCF7 cells

incubated with the compounds

for 30 min. Images correspond

to the intrinsic compound

fuorescence, bright feld

(overall cell morphology),

Hoechst 33342 fuorescence

(cell nuclei) and a merged

image, with the compound

fuorescence in green, nuclei

in cyan and brightfeld in gray.

Scale bar=50 μm. b Quantif-

cation of compounds intrinsic

fuorescence co-localizing with

cells during the live-cell kinetic

assay. Increasing measurements

indicate cellular accumulation

of the compounds. Values are

average standard deviation

ZnNqTSM^asym. Some of that intracellular fluorescence,

particularly for ZnL³, was strongly co-localized with the

lysosomal marker, indicating the lysosomal entrapping

of this aromatic Zn complex. When cells were incubated

with the model ZnNqTSM^asymand lysotracker a less

intense co-staining was detected (Fig. 7). These results

suggest that there is an influence of the piperidine arm in

the subcellular distribution of the Zn complexes, namely,

in their lysosomal trapping.

Cellular uptake studies by ICP‑MS

Searching to correlate the cytotoxic activity of the ali-

phatic and aromatic BTSC Zn(II) complexes with their

ability to enter the cells, the quantitative assessment of the

uptake of the compounds in MCF7 cells was performed by

determining their total Zn content using ICP-MS. In this

assay, the amount of total Zn in cells incubated with the

compounds was determined, and compared with the total

1 3

JBIC Journal of Biological Inorganic Chemistry

Fig. 7 Co-localization of aro-

matic BTSC Zn(II) complexes

with the lysosomes in MCF7

cells. MCF7 cells were incu-

bated with Lysotracker, as well

as ZnL³(a) or ZnNqTSM^asym

(b) for 30 min in a live-cell

fuorescence microscopy experi-

ment. Scale bar: 50 μm

Zn in same number of cells incubated only with growth

media, taking into consideration that Zn is an endogenous

element involved in numerous processes of cellular metab-

olism [34].

Cells were incubated with ZnATSM, ZnNqTSM^asym

,

and ZnL¹⁻⁴at 10 μM for 30 min and 2 h. After appropriate

processing, the cellular pellets were analyzed by ICP-MS,

as detailed in the experimental section. As shown in Fig. 8,

the incubation of MCF7 cells with the compounds led to an

increase of the intracellular Zn concentration in comparison

with the control, showing that all the Zn(II) complexes have

the ability to accumulate in this tumoral cell line model. For

both incubation times, the aromatic BTSC Zn(II) complexes

showed higher cellular uptake than the aliphatic counter-

parts with the same N-terminal substituents, i.e., ZnNqTS-

M

^asym>ZnATSM, ZnL³>ZnL¹, and ZnL⁴>ZnL².

The determination of the zinc cellular content by ICP-MS

at 30 min of incubation showed that ZnL³presented the

highest uptake (Fig. 8a). This result is in agreement with

the live-cell imaging studies that showed a faster uptake

kinetics for ZnL³in comparison with ZnNqTSM^asym(see

Fig. 6). In contrast, after 2 h of incubation, ZnNqTSM^asym

exhibited higher total Zn uptake than ZnL³. This trend can

be accounted by the steadily increasing uptake observed

for ZnNqTSM^asymand the faster cellular release found for

ZnL³, up to 60 min, which was the fnal timepoint of the

live-cell imaging studies. As mentioned above, after 30 min

cells incubated with ZnL³presented signs suggestive of

apoptosis that could lead to loss of the intracellularly accu-

mulated ZnL³.

Fig. 8 Uptake of the BTSC Zn(II) complexes in MCF7 cells. Cells

were incubated with the tested compounds (10 μM) or with DMSO as

control for 30 min (a) and 2 h (b) at 37 °C. Thereafter, the concentra-

tion of Zn in the cellular pellet was determined by ICP-MS and data

expressed as metal (Zn) content per million cells

As a general trend, the cytotoxic activity of the complexes

in the MCF7 cell line followed the trend observed for their

cellular uptake. Furthermore, the presence of the aromatic

naphthalene backbone appears to have an important role in

1 3

JBIC Journal of Biological Inorganic Chemistry

the cellular uptake of the Zn(II) complexes and in their cyto-

toxic activity.

37 °C, following a 15 min pre-incubation with AO. Next,

the cells were imaged by fuorescence microscopy and the

results obtained are presented in Fig. 9.

Efects on lysosomal integrity and ROS generation

Control MCF7 cells, treated only with AO, showed gran-

ular red fuorescence consistent with the concentration of

AO in lysosomes (Fig. 9b). Treatment of the cells with H₂L³

and ZnL³resulted in increased cytosolic green fuorescence

and reduced lysosomal red fuorescence (Fig. 9c, d). Con-

sequently, a signifcant (p<0.01) reduction of ca. 40% was

observed for the red:green fuorescence ratio. This efect

is due to the relocation of AO from the lysosomes to the

cytosol, indicating that both H₂L³and ZnL³were able to

induce LMP.

Among the new compounds carrying cyclic amine groups,

ZnL³and the corresponding free ligand, H₂L³, showed

the greatest cytotoxic activity; moreover, ZnL³showed

the fastest uptake kinetics and highest cellular uptake, as

well as lysosomal localization, according to the results of

live-cell imaging experiments and ICP-MS measurements.

Thus, this complex and respective ligand were chosen to

proceed with additional in vitro studies aiming to obtain

some insight on the mechanism of action of the compounds.

As mentioned in the introduction, thiosemicarbazones and

their Zn(II) complexes have shown multiple mechanisms of

action, and it has been demonstrated in several instances that

the lysosomal membrane permeabilization (LMP) giv es an

important contribution to their biological efects [3, 9–13].

Therefore, next, we examined whether ZnL³and H₂L³could

also induce LMP.

It is also important to notice that the LMP activity

observed for H₂L³and ZnL³does not strictly depend on

their lysosomal localization, but also might refect their

ability to mediate ROS generation intracellularly. Thus,

this efect was evaluated in MCF7 cells using the redox-

sensitive probe 2′,7′-dichlorodihydrofuorescein diacetate

(H₂DCF-DA), which is oxidized by ROS to the highly fuo-

rescent product 2′,7′-dichlorofuorescein (DCF) [35]. To

assess the generation of ROS, MCF7 cells were treated with

H₂L³and ZnL³(2, 10, 20 μM) for 30 min and then incu-

bated with 1 µM H₂DCF-DA for a further 30 min. As shown

in Fig. 10, cells incubated with the compounds were imaged

by fuorescence microscopy, in parallel with a positive con-

trol consisting on cells exposed to H₂DCF-DA and H₂O₂

and a negative control consisting on cells treated uniquely

with DMSO vehicle.

The studies were performed using acridine orange (AO)

as a lysosomotropic fuorophore, which when concentrated

in lysosomes emits granular red fuorescence, whereas in the

cytosol or nucleus emits difuse green fuorescence. Reduc-

tion in granular red fuorescence combined with increased

difuse cytosolic green fuorescence typically indicates relo-

cation of AO from the lysosomes to the cytosol, following

changes in lysosomal membrane permeability. To examine

the ability of ZnL³and respective ligand to afect LMP,

MCF7 cells were incubated with the compounds for 1 h at

The positive control, H₂O₂, led to a substantial increase

(~ 2.5-fold of negative control) in DCF fuorescence. The

Fig. 9 Evaluation of LMP

induced by ZnL³and H₂L³

using the AO lysosomal

integrity assay in MCF7 cells.

Cells were pre-incubated with

AO (5 µg/mL) for 15 min,

followed by an 1 h incubation

with the tested compounds

(10 μM). a Quantifcation of

the red:green fuorescence ratio

(expressed data are presented

as average standard deviation

normalized to the AO control).

b–d Images of cells incubated

with AO only (control) and

treated with H₂L³and ZnL³,

respectively

1 3

JBIC Journal of Biological Inorganic Chemistry

Fig. 10 Redox activity of ZnL³and respective ligand H₂L³(2,

10 and 20 μM, 30 min) in MCF7 cells incubated with H₂DCF-DA

(1 μM, 30 min). H₂O₂was included as a positive control. a Quan-

tifcation of DCF fuorescence normalized to the negative control

(H₂DCF-DA alone) presented as average standard deviation. b Rep-

resentative images of DCF fuorescence in cells incubated with H₂O₂

and treated with diferent concentrations of ZnL³

treatment of cells with increasing concentrations of ZnL³

resulted in the incremental generation of ROS, with the

highest concentration (20 μM) inducing ROS production

to levels 2.5 times higher than the negative control and

even slightly superior to the positive control.

Conclusions

New BTSC Zn(II) complexes bearing piperidine and mor-

pholine pendant arms, based on aliphatic and aromatic

scafolds were synthesized and thoroughly characterized.

In the aliphatic derivatives, ZnL¹and ZnL², the BTSC

chelators present symmetrical N₂S₂coordination mode.

The aromatic counterparts, ZnL³and ZnL⁴, formed asym-

metric Zn(II) complexes, which most probably refected

the (E,Z) conformation of the H₂L³and H₂L⁴compounds,

that are pre-organized to coordinate the metal center

through one inner and one outer imine nitrogen atoms.

The aromatic complexes showed fuorescence proper-

ties, which were characterized and allowed live-cell imag-

ing studies. The biological evaluation of the complexes

demonstrated that ZnL³and ZnL⁴showed much stronger

anti-proliferative action than the aliphatic ZnL¹and ZnL²,

which could be correlated with their higher cell uptake as

demonstrated by ICP-MS. Complex ZnL³presented the

fastest cell uptake kinetics in tumoral cells, showed lyso-

somal entrapping and was able to generate intracellular

ROS, which induced lysosomal membrane damage. Alto-

gether these data indicate that the presence of the pendant

cyclic amines might infuence the cell uptake mechanisms

of BTSC Zn(II) complexes, most likely endowing them

with lysosomotropic properties. However, more detailed

mechanistic studies are needed to fully elucidate the efect

of the pendant cyclic amines in the cell uptake kinetics of

In agreement with the cytotoxic and the LMP efects

observed for both compounds, similar values of DCF fuo-

rescence (ROS production) were detected for H₂L³and

ZnL³, in particular for the lowest concentrations. These

results indicate that the LMP efect induced by H₂L³and

ZnL³is probably due to ROS production. ROS can desta-

bilize the lysosomal membrane by causing lipid peroxi-

dation and damaging lysosomal membrane proteins [36].

However, other processes can account for the cytotoxicity

of the compounds, since a variety of cellular structures and

pathways can be afected by ROS generation. Moreover,

it has been recently demonstrated that the mechanism of

action of Zn(II) complexes with di-2-pyridylketone (DpT)

and 2-acetylpyridine (ApT) thiosemicarbazone chelators

involves sequestration within the lysosomes and trans-met-

alation with intralysosomal Cu²⁺to produce redox active

complexes that induce LMP-mediated cytotoxicity by ROS

[3]. The similar IC₅₀values exhibited by H₂L³and ZnL³in

the tested cell lines, as well as the similar ability to induce

LMP and produce ROS, point out for the involvement of

the same type of mechanisms with a possible contribution

of trans-metalation with copper ions. Although justifying

further investigation, we have considered that this aspect

was out of the scope of the present contribution.

1 3

JBIC Journal of Biological Inorganic Chemistry

this class of Zn(II) complexes and on their intracellular

localization.

2 M HCl (2 mL) was refuxed overnight. Then, triethylamine

was added until basic pH. The aqueous phase was diluted

with water (50 mL) and extracted with dichloromethane

(2 × 50 mL). The organic phase was dried over Na₂SO₄,

fltered, concentrated under vacuum and then submitted

to column chromatography on silica gel (CH₂Cl₂/MeOH/

NEt₃1:0.05:0.01) to give the desired BTSC as a reddish

solid, which was further washed with dichloromethane and

n-hexane.

Experimental section

Materials and methods

All chemicals were p.a. grade and were used without puri-

fcation unless stated otherwise. H₂L¹, H₂L², H₂NqTSM

and ZnA TSM were synthesized as described in the litera-

ture [14, 18, 19]. The chemical reactions were followed by

thin-lay er c hromatography (TLC). TLC was performed on

pre-coated silica plates 60 F₂₅₄(Merck). Visualization of the

plates was carried out using UV light (254 nm) and/or iodine

chamber. Gravity column chromatography was carried out

on silica gel (Merck, 70–230 mesh).

bis[4-N-(2´-(piperidine-1-yl)ethyl)-3-thiosemicar-

bazone] acenaphthenequinone (H₂L³)—yield = 53%; R_f

(DCM:MeOH:TFA 1:0.1:0.01)=0.3; ¹H NMR (DMSO-d₆,

300 MHz) δ 1.46 (m, 4H), 1.58 (m, 8H), 2.65 (m, 8H), 2.76

(m, 4H), 3.78 (m, 4H), 7.82 (m, 2H), 8.03 (m, 2H), 8.11 (m,

1H), 8.33 (m, 1H), 8.55 (br s, 1H, NH), 9.01 (br s, 1H, NH),

11.09 (br s, 1H, NH), 12.86 (s, 1H, NH); ¹³C NMR (DMSO-

d₆, 75 MHz) δ 24.46, 24.78, 25.98, 26.45, 41.88, 42.26,

54.73, 54.80, 57.39, 57.80, 118.73, 125.31, 127.59, 129.22,

130.92, 133.62, 133.94, 136.60 (C=N), 178.33 (NHC=S);

ES⁺MS C₂₈H₃₈N₈S₂(550.27) m/z 551.5 [M + H]⁺; Anal.

calcd. for C₂₈H₃₈N₈S₂.H₂O: C 59.12, H 7.09, N 19.74; found

C 59.08, H 7.39, N 19.80. IR (KBr, ν/cm⁻¹): 3287 (w), 2934

(s, N–H), 2083 (w), 1533 (vs, C=N), 1476 (vs, C=N), 1183

(s, thioamide), 1112 (s, N–N), 822, 772. (vs, very strong; s,

strong; m, medium; w, weak; sh, sharp). UV–Vis (DMSO)

λ/nm (ε/M⁻¹cm⁻¹): 335 (1.95×10⁴); 438 (8.55×10³); 464

(8.34×10³); 490 (5.73×10³).

Elemental analyses (CHN) were recorded on an EA

110CE automated instrument. Zn determinations were per-

formed by inductively coupled plasma mass spectrometry

(ICP-MS) using a quadrupole mass flter ICP-MS, the ELAN

DRC-e from PerkinElmer Sciex (Axial Field Technology).

Electrospray ionization mass spectrometry (ESI–MS)

was performed on a QITMS instrument in positive and

1

negative ionization mode. H and ¹³C NMR spectra were

recorded on a Varian Unity 300 MHz or 400 MHz spec-

trometer at the frequencies of 300 or 400 MHz (¹H) and

1

75 or 100 MHz (¹³C), respectively. H and ¹³C chemical

bis[4-N-(3´-morpholinoethyl)-3-thiosemicarbazone]

acenaphthenequinone (H²L⁴)—yield = 65%; R_f

(DCM:MeOH:TFA 1:0.1:0.01)=0.3; ¹H NMR (DMSO-d₆,

300 MHz) δ~2.50 (8H under DMSO-d₆signal, confrmed

by COSY), 2.62 (m, 4H), 3.59 (m, 8H), 3.77 (m, 4H), 7.80

shifts (δ) are reported in ppm relative to residual solvent

signals (DMSO-d₆: 2.50 ppm for ¹H NMR and 39.52 ppm

for ¹³C NMR; C₃OD: 3.31 ppm for ¹H NMR and 49.00 ppm

for ¹³C NMR). The compounds were characterized by elec-

trospray ionization mass spectrometry (ESI–MS) using a

Bruker model Esquire 3000 plus. The spectra were acquired

at room temperature, using acetonitrile diluted solutions of

the pure compounds. IR spectra were recorded in KBr pellets

on a Bruker Tensor 27 spectrometer. UV–visible absorption

spectra were recorded on Perkin Elmer Lambda 35 UV–Vis

spectrophotometer with 1 cm quartz cuvettes. Steady-state

fuorescence measurements were carried out on a SPEX^®

Fluorolog spectrofuorimeter (Horiba Jobin–Yvon) in an

FL3–11 confguration, equipped with a Xenon lamp and in

a 1 cm quartz cuvette. The instrumental response was cor-

rected by means of a correction function provided by the

manufacturer. Fluorescence average lifetimes were deter-

mined using a 372 nm HORIBA Jobin–Yvon NanoLED as

light source.

3

(overlapped triplets, J = 8.0 Hz, 2H), 8.04 (m, 2H), 8.14

3

(d, J = 8.3 Hz, 1H), 8.24 (d, J = 7.2 Hz, 1H), 8.72 (br t,

1H, NH), 9.06 (br t, 1H, NH), 11.30 (br s, 1H, NH), 12.62

(s, 1H, NH); ¹³C NMR (DMSO-d₆, 75 MHz) δ 40.86,

41.41, 53.27, 56.45, 66.30, 118.23, 124.68, 126.89, 127.81,

128.29, 128.59, 128.97, 130.03, 132.78, 136.14, 138.04,

142.23, 146.44. ES⁺MS C₂₆H₃₄N₈O₂S₂(554.2) m/z: 555.3

[M+H]⁺; Anal. calcd. for C₂₆H₃₄N₈O₂S₂.H₂O: C 54.52, H

6.34, N 19.56; found C 54.36, H 6.46, N 19.62. IR (KBr, ν/

cm⁻¹): 3443 (s, NH), 3315 (s, NH), 1536 (vs, C=N), 1478

(vs, C=N), 1298 (s, thioamide), 1133 (s, N–N), 976, 768.

UV–Vis (DMSO) λ/nm (ε/M⁻¹cm⁻¹): 334 (2.89×10⁴); 462

(1.53×10⁴); 488 (1.38×10⁴); 604 (4.67×10²).

General procedure for the synthesis of BTSC zinc(II)

complexes

General procedure for the synthesis of aromatic

BTSCs

To a suspension of the BTSC ligand (H₂NqTSM, H₂

L¹–H₂L⁴; 0.2–0.4 mmol) in methanol (5 mL) was added

one equivalent of zinc acetate dihydrate and the mixture

was stirred overnight at RT (ZnL¹and ZnL²) or at refux

A solution of 4-substituted-3-thiosemicarbazide (3 mmol)

and acenaphthenequinone (1 mmol) in ethanol (2 mL) and

1 3

JBIC Journal of Biological Inorganic Chemistry

(ZnNqTSM^asym, ZnL³and ZnL⁴). The yellow/orange pre-

cipitates of ZnNqTSM^asym, ZnL¹and ZnL²were recov-

ered by fltration and washed with 20 mL of methanol and

3 × 10 mL of diethyl ether. For ZnL³the reaction mixture

was dried in vacuum and the yellow–orange complex was

obtained in a pure form after recrystallization in a mixture

of dichloromethane and diethyl ether. ZnL⁴precipitated

from the reaction mixture and was recovered by fltration

as a yellow–orange solid.

Zinc bis[4-N-(2´-(piperidine-1-yl)ethyl)-3-thiosemi-

carbazone] acenaphthenequinone (ZnL³)—yield=70%;

¹H NMR (DMSO-d₆, 300 MHz) δ 1.41 (m, 4H), 1.54

(m, 8H), 2.45 (m, 8H), 2.59 (q, 4H), 3.56 (m, 4H), 7.67

3

(m, 2H), 7.91–7.85 (m, 2H + NH), 7.97 (d J = 15 Hz,

3

1H), 8.44 (br t, 1H, NH), 8.56 (d, J = 6 Hz, 1H); ¹³C

NMR (DMSO-d₆, 75 MHz) δ 21.34, 24.00, 25.60, 25.68,

40.96, 42.61, 54.02, 54.13, 56.72, 57.32, 116.68, 125.38,

125.50, 127.31, 128.16, 128.28, 129.89, 130.61, 135.28,

140.31, 146.25, 172.55, 175.94, 182.64. ES⁺MS for

C₂₈H₃₆N₈S₂Zn (612.18) m/z: 613.3 [M+H]⁺. Anal. calcd.

for C₂₈H₃₆N₈S₂Zn × 2CH₃COOH.2H₂O: C 49.80, H 6.28,

N 14.55; found C 49.68, H 6.43, N 14.53; IR (KBr, ν/

cm⁻¹): 3277 (s broad, N–H), 2933 (w), 2860 (w), 1575

(s, C=N), 1513 (vs, C=N), 1444 (ring), 1417 (ring), 1263

(s, thioamide), 1155 (vs, N–N), 1110, 780 (w, C–S), 667.

UV–Vis (DMSO) λ/nm (ε/M⁻¹cm⁻¹): 361 (2.99 × 10⁴);

404 (9.71 × 10³); 425 (9.60 × 10³); 456 (9.11 × 10³); 485

(9.75×10³).

Zinc bis(4-N-methyl-3-thiosemicarbazone) acenaph-

1

thenequinone (ZnNqTSM^asym)—yield = 80%; H NMR

(DMSO-d₆, 300 MHz) δ 3.09 (d, 3H, Me), 3.12 (d, 3H,

Me), 7.79 (overlapped triplets, 2H), 8.02 (d, 1H), 8.07 (d,

1H), 8.12 (d, 1H), 8.20 (d, 1H), 8.84 (br q, NH), 9.15 (br

q, NH); ¹³C NMR (CD₃OD, 75.4 MHz) δ 32.44, 116.87,

125.24, 125.80, 127.24, 128.16, 128.48, 129.84, 130.65,

135.24, 135.38, 140.17, 146.15, 176.13, 183.64; ES⁺

MS C₁₆H₁₄N₆S₂Zn (418.0) m/z 419.3 [M + H]⁺; Anal.

calcd. for C₁₆H₁₄N₆S₂Zn.H₂O: C 43.89, H 3.68, N 19.19;

found C 44.01, H 3.88, N 19.18; IR (KBr, ν/cm⁻¹): 3435

(s broad, NH) 3347 (s broad, N–H), 2931 (w), 1569 (vs,

C=N), 1384 (s, thioamide), 1064 (N–N), 774 (w, C–S).

UV–Vis (DMSO) λ/nm (ε/M⁻¹cm⁻¹): 359 (4.24 × 10⁴);

402 (1.23 × 10⁴); 424 (1.23 × 10⁴); 454 (1.10 × 10⁴); 483

(1.20 × 10⁴).

Z i n c b i s [ 4 - N - ( 3 ´ - m o r p h o l i n o e t hy l ) - 3 -

thiosemicarbazone] acenaphthenequinone (ZnL⁴)—

yield=76%; ¹H NMR (DMSO-d₆, 300 MHz) δ 2.46 (m, 8H),

3

2.61 (m, 4H), 3.63 (m, 8H), 3.72 (q, J = 4 Hz, J = 8 Hz,

3

4H), 7.68 (t, J = 8 Hz, 2H), 7.89 (m, 2H + NH), 7.98 (d,

³J=8 Hz, 1H), 8.46 (br t, NH), 8.56 (d, ³J=8 Hz, 1H). ¹³

C

Zinc diacetyl-2-bis[4-N-(2´-(piperidine-1-yl)

ethyl)-3-thiosemicarbazone] (ZnL¹)—yield = 74%;

¹H NMR (DMSO-d₆, 300 MHz) δ 1.37 (m, 4H), 1.48

(t, 8H), 2.18 (s, 6H), 2.35 (br s, 8H), 2.43 (t, 4H),

6.98 (br s, 2H, NH), 4H under residual peak of H₂O;

¹³C NMR (DMSO-d₆, 75.4 MHz) δ 14.31, 23.17,

24.89, 39.02, 54.92, 58.01, 148.27, 178.77; ES⁺MS

C₂₀H₃₆N₈S₂Zn (516.17) m/z 517.3 [M + H]⁺; Anal. calcd.

for C₂₀H₃₆N₈S₂Zn.2H₂O.1CH₃COOH: C 43.02, H 7.22,

N 18.24; found C 43.20, H 6.67, N 18.34; IR (KBr, ν/

cm⁻¹): 3335 (br s, N–H), 2936 (s, sharp), 1422 (vs, C=N),

1215 (s, thioamide), 1127 (N–N), 839 (w, C–S), 665 (w).

UV–Vis (DMSO) λ/nm (ε/M⁻¹cm⁻¹): 263 (5.60×10³); 314

(8.25×10³); 438 (8.81×10³).

NMR (DMSO-d₆, 75 MHz) δ 39.52 (signal under resid-

ual DMSO as confrmed by COSY), 42.57, 53.54, 56.62,

57.42, 66.51, 93.31, 116.97, 125.60, 127.59, 128.42, 130.05,

130.76, 135.40, 140.55, 146.33, 175.98, 182.78; ES⁺MS

for C₂₆H₃₂N₈O₂S₂Zn (616.14) m/z: 617.2 [M+H]⁺; Anal.

calcd. for C₂₆H₃₂N₈O₂S₂Zn: C 50.52, H 5.22, N 18.13; found

C 50.47, H 5.61, N 18.18; IR (KBr, ν/cm⁻¹): 3390 (s broad,

N–H), 2960 (w), 2817 (w), 1575 (s, C=N), 1540 (vs, C=N),

1479 (ring), 1472 (ring), 1335 (vs, thioamide), 1119 (vs,

N–N), 777 (s, C–S). UV–Vis (DMSO) λ/nm (ε/M⁻¹cm⁻¹):

361 (4.12 × 10⁴); 403 (1.21 × 10⁴); 424 (1.22 × 10⁴); 453

(1.14×10⁴); 486 (1.29×10⁴).

Crystal structure determination

Zinc diacetyl-2-bis[4-N-(3´-morpholinoethyl)-3-

thiosemicarbazone] (ZnL²)—yield= 76%; ¹H NMR

(DMSO-d₆, 300 MHz) δ 2.18 (s, 6H), 2.39 (br t, 8H), 2.47

(m, 4H), 3.45 (m, 4H), 3.56 (br t, 8H), 7.01 (br s, 2H, NH);

¹³C NMR (DMSO-d₆, 75 MHz) δ 14.02, 53.34, 57.16, 66.36,

145.30, 175.26 (gHSQC experiment revealed the presence

of a signal under DMSO); ES⁺MS for C₁₈H₃₂N₈O₂S₂Zn

(520.14) m/z 543.1 [M + Na]⁺; Anal. calcd. for

C₁₈H₃₂N₈S₂O₂Zn.4H₂O: C 39.24, H 6.79, N 18.86; found C

39,20, H 7.25, N 18.79; IR (KBr, ν/cm⁻¹): 3356 (vs, broad,

N–H), 2936 (w), 2860 (w), 1456 (vs, C=N), 1230 (m, thio-

amide), 1116 (vs, N–N), 865 (w, C–S). UV–Vis (DMSO)

λ/nm (ε/M⁻¹cm⁻¹): 265 (7.06×10³); 315 (1.03×10⁴); 437

(1.08×10⁴).

Brown crystals of ZnL²and yellow–brown crystals of ZnL³

suitable for X-ray difraction studies were obtained by slow

difusion of diethyl ether into concentrated methanolic or

dichloromethane solutions of the complexes, respectively,

after standing for several days at RT. The crystals were

mounted on a loop with protective oil. X-ray data were

collected at 150 K on a Bruker APEX II CCD diffrac-

tometer using graphite monochromated Mo Kα radiation

(0.71073°A) and operating at 50 kV and 30 mA. The X-ray

data collection was monitored by the APEX2 program. All

data were corrected for Lorentzian, polar ization, and absorp-

tion efects using SAINT [37] and SADAB [38] programs.

Structure solution and ref nement were performed using

1 3

JBIC Journal of Biological Inorganic Chemistry

direct methods with program SIR9 [39] and SHELXL97

[40] both included in the package of programs WINGX-Ver-

sion 2014.1 [41]. A full-matrix least-squares refnement was

used for the non-hydrogen atoms with anisotropic thermal

parameters. All hydrogen atoms were inserted in idealized

positions and allowed to refne riding in the parent atom,

except the nitrogen (N3 and N6) hydrogen atoms that were

fnding in the electron density map and refned isotropically.

Molecular graphics were prepared using Mercury [42].

CCDC—ZnL²-1478469 and ZnL³-565108—contain the

supplementary crystallographic data for this paper. These

data can be obtained from The Cambridge Crystallographic

Data Centre via www.ccdc.cam.ac.uk/data_request/cif.

prostate carcinoma PC3 and LNCap cell lines were main-

tained in RPMI 1640 Medium. Human breast carcinoma

MCF7, cervical carcinoma HeLa cells and the non-tumoral

HEK 293 (human embryonic kidney) cells were grown in

DMEM. Both culture media were supplemented with 10%

heat-inactivated fetal bovine serum (FBS). All culture media

and supplements were from Gibco, Invitrogen, UK. Cells

were cultured in a humidifed atmosphere of 95% air and 5%

CO₂at 37 °C (Heraeus, Germany).

Cytotoxicity assay

The cytotoxicity of the BTSC chelators and corresponding

Zn(II) complexes was assessed by evaluating their efects on

cellular proliferation using the [1-(4,5-dimethylthiazol-2-yl)-

2,5-diphenyl tetrazolium] (MTT) assay. Cells were seeded

in 96-well culture plates at a density of 1.5×10⁴to 2.5×10⁴

cells/well (depending of the cell line) and left to adhere

overnight at 37°C. Cells were then incubated with the Zn

complexes and respective ligands at diferent concentrations

(0–50 µM) during 48 h at 37°C. All tested compounds were

frst solubilized in DMSO (20 mM stock solution) and then

diluted in culture medium for the assay, with the percentage

of organic solvent in the fnal solution never exceeding 0.1%.

After incubation, the medium was removed, cells washed

with PBS and then incubated with MTT (200 µL of 0.5 mg/

mL solution in Modifed Eagle’s Medium without phenol

red) for 3 h at 37°C. The MTT solution was then removed

and the insoluble blue formazan crystals formed were dis-

solved in DMSO. The absorbance of this colored (purple)

solution was measured at 570 nm, using a plate spectropho-

tometer (Power Wave Xs; Bio-Tek). Each test was performed

with at least six replicates. The results were expressed as

percentage of the surviving cells in relation with the control,

which was incubated without any compound. IC₅₀values

were determined using Graph Pad Prism.

Fluorescence spectroscopy

Dimethyl sulfoxide (DMSO) was purchased from Riedel-de

Haën, spectroscopy grade, and used without further purifca-

tion. The complexes were dissolved in DMSO ensuring that

the absorbance values remained below 0.4 at the excitation

and emission wavelengths. Fluorescence measurements were

carried out in 1 cm quartz cuvettes. The experiments were

carried out at room temperature and using the software Flu-

orEssence V3.

The fuorescence quantum yield of each compound was

determined by the relative method, using quinine sulphate as

standard [32]. To minimize instrumental errors, sample and

standard solutions were prepared with very similar absorp-

tion values at the common excitation wavelength (370 nm).

The fuorescence quantum yield of each complex was meas-

ured in DMSO, at room temperature. It can be calculated

from Eq. 1:

Fⁱf_Sꢁ_i²

ꢀ_fⁱ=

ꢀ_f^S,

(1)

F^Sf_iꢁ_S²

where ꢀ_fⁱand ꢀ_f^sare the fuorescence quanta yield of the

Zn determination by ICP‑MS

sample and standard, respectively. Fⁱand F^sare the inte-

grated intensities (areas) of sample and standard spectra,

respectively (in units of photons); f_xis the absorption factor,

or the fraction of the light reaching the sample that is

absorbed (f_x=1–10^–Ax, where A=absorbance); η_iand η_sare

the refractive indices of the sample and reference solution,

respectively [32].

Cellular uptake of the zinc complexes was assessed by

ICP-MS quantifcation of total zinc. Adherent and confu-

ent MCF7 cells (T25 culture fask) were incubated with the

tested compounds (10 μM) or with DMSO as control in

culture medium for 30 min and 2 h at 37 °C and 5% CO₂.

The harvested cell suspension was centrifuged at 200 g

for 3 min at 4º C (Centrifuge 5804R, Eppendorf) and cells

washed twice with cold PBS to remove the unbound com-

plex. An aliquot of 10 mL of cells suspension was separated

for counting the number of cells with trypan blue. Cell pel-

lets were resuspended in 100 μL of concentrated nitric acid

(65% HNO_3,Merck) purifed by sub-boiling distillation (bi-

distilled) and left to digest overnight. Samples were then

heated for 20 min at 90 °C (in a sand bath) to complete

Fluorescence average lifetimes were determined by sin-

gle-photon counting technique. Good to ft values (χ²) were

found between 0.79 and 0.99.

Cell culture

Human ovarian epithelial cancer A2780 (cisplatin sensi-

tive) and A2780R (acquired cisplatin resistance) as well as

1 3

JBIC Journal of Biological Inorganic Chemistry

the digestion. To each sample was added 1.5 mL of 1%

(v/v) nitric acid diluted with water ultra-pure (18.2 MΩcm)

obtained with a MilliQ water purifcation system (Milli-

pore). All samples were fltered with a 0.45 μm PVDF mem-

brane syringe flter. Samples were 4 times diluted with H₂O

ultra-pure before their Zn content was determined by ICP-

MS. SmartTune Solution-Std Elan & DRC-e and ELAN 100

Set up/Stab/MassCal standards solutions by Perkin Elmer

Pure Atomic Spectroscopy Standard were used for daily and

monthly optimization and tuning. Measurement sequences

began with blank and standard solutions.

subtraction. Each condition was imaged in duplicate,

with~60–80 cells being analyzed for each condition (mini-

mum of 40 cells per condition). Values were reported as the

average compound fuorescence standard deviation across

the 2 time lapses at each timepoint.

Live‑cell fuorescence microscopy—co‑localization

studies

For the co-localization studies, cells were plated as men-

tioned above. The day of the assay, cells were loaded with

the vital dye Hoechst 33342 (0.2 μg/ml) to stain the nuclei

and Lysotracker Red DND-99 (50 nM) to stain the lys-

osomes in DMEM Fluorobrite for 30 min. After that period,

the culture medium was replaced by fresh medium contain-

ing ZnL³or ZnNqTSM^asym(20 μM) or an identical amount

of DMSO (0.1%) as control. Then, cells were immediately

placed in an automated Leica DMI 6000B widefeld fuores-

cence microscope for live-cell imaging. Cells were imaged

at 0, 15, and 30 min using a 10x objective with numerical

aperture 0.4. Three image channels were acquired: Bright

field (transmission), Hoechst fluorescence (excitation

band: 340–380 nm; emission band: 450–490 nm), ZnL³, or

ZnNqTSM^asymfuorescence (excitation band: 490–510 nm;

emission band: 520–550 nm) and Lysotracker fuorescence

(excitation band: 515–560 nm; emission band:>590 nm).

Cells were kept at 37 °C and 5% CO₂throughout all steps of

the protocol, except when adding compounds or the nuclear

dye.

The instrument was calibrated using 0, 5, 10, 50, 100 and

500 parts per billion (ppb) of certifed Calibration Stand-

ard 3 solution (Perkin Elmer) prepared in 1% (v/v) nitric

acid. To minimize drift efects, instruments instability and

matrix efects, a certifed internal standard solution contain-

ing 200 ppb of rhodium (Rh) was added to each sample and

standard solution. Detection limit (0.22 ppb) was determined

based on the low range concentrations of the standards used.

The concentration of Zn was calculated as ng of metal per

million of cells.

Live‑cell fuorescence microscopy—cellular uptake

studies

Cellular uptake of the zinc complexes was measured by

fuorescence microscopy using the compounds’ intrinsic

fuorescence. MCF7 cells were plated in 8-well chambered

slides (Millicell EZ slide, Millipore) (20000 cells/well) in

DMEM Fluorobrite supplemented with 10% FBS (Gibco)

and allowed to attach overnight. The following day cell

nuclei were stained by incubating the vital dye Hoechst

33342 (0.2 μg/ml) in the culture media for 30 min. After that

period, the culture medium was replaced by fresh medium

containing the compounds (10 or 20 μM) or an identical

amount of DMSO (0.1%) as control. Then, cells were imme-

diately placed in an automated Leica DMI 6000B widefeld

fuorescence microscope for live-cell imaging. Cells were

imaged every 5 min during approximately 60 min using a

10 × objective with numerical aperture 0.4. Three image

channels were acquired: Bright feld (transmission), Hoechst

fuorescence (excitation band: 340–380 nm; emission band:

450–490 nm) and compound fuorescence (excitation band:

490–510 nm; emission band: 520–550 nm). Cells were kept

at 37 °C and 5% CO₂throughout all steps of the protocol,

except when adding compounds or the nuclear dye.

Lysosomal membrane permeabilization assay

The ability of the selected compounds to induce LMP was

assessed according to Acridine Orange (AO; Sigma-Aldrich)

distribution. MCF7 cells were seeded on sterile coverslips

to give 50% confuency and left to adhere overnight. Cells

were then incubated with AO (5 μg/mL) for 15 min at 37 °C,

washed 3 times with PBS, and then incubated for 1 h at

37 °C with H₂L³or ZnL³(10 μM). The cells were then

washed 3 times with PBS, fxed with 4% formaldehyde in

PBS for 20 min at room temperature and coverslips mounted

in mounting media (Fluoroshield with DAPI, Sigma).

Images were captured using a Zeiss Axioplan2 microscope

equipped with FITC and Texas Red flters. Several images

were randomly collected in each slide. The red and green

fuorescence intensities were quantifed with image process-

ing and analysis software (ImageJ v1.51 k), and presented as

red/green ratio normalized to the control.

Accumulation of the compounds by individual cells was

measured with an automatic image analysis algorithm imple-

mented in CellProfler [43]. Briefy, the Hoechst staining

was used to segment individual nuclei. Then, the nuclear

shape was dilated by 7.8 μm to estimate the location of the

whole cell. Finally, the mean compound fuorescence in

the cell objects was measured, after adequate background

Intracellular ROS production assay

Intracellular ROS generation was measured using

2′,7′-dichlorodihydrofluorescein diacetate (H₂DCF-DA,

1 3

JBIC Journal of Biological Inorganic Chemistry

4. Park KC, Fouani L, Jansson PJ, Wooi D, Sahni S, Lane DJR,

Palanimuthu D, Lok HC, Kovacevic Z, Huang MLH, Kalinow-

ski DS, Richardson DR (2016) Copper and conquer: copper

complexes of di-2-pyridylketone thiosemicarbazones as novel

anti-cancer therapeutics. Metallomics 8(9):874–886

ThermoFisher Scientific). H₂DCF-DA is hydrolyzed by

intracellular esterases to the membrane impermeable analog,

2′,7′-dichlorodihydrofuorescein (H₂DCF), which leads to

its accumulation within the cytosol. Cellular oxidants local-

ized in the cytosol, such as H₂O₂, HO., ROO., or ONOO⁻,

oxidize non-fuorescent H₂DCF to the fuorescent product,

dichlorofuorescin (DCF).

5. Cortezon-Tamarit F, Sarpaki S, Calatayud DG, Mira-

bello V, Pascu SI (2016) Applications of “hot” and “cold”

bis(thiosemicarbazonato) metal complexes in multimodal imag-

ing. Chem Rec 16(3):1380–1397

H₂DCF-DA stock solution (10 mM) in DMSO was stored

at –20 °C. MCF7 cells were seeded on sterile coverslips

to give 50% confuency and left to adhere overnight. Cells

were incubated with H₂L³and ZnL³(2, 10, and 20 μM)

in MEM (without phenol red, Gibco) supplemented with

10% FBS (assay medium) for 30 min at 37 °C, in the dark,

and then washed once with assay medium. Subsequently,

H₂DCF-DA (1 µM) in assay medium was added for further

30 min. Hydrogen peroxide (H₂O₂, 3% (10 vol)), 5 times

diluted in assay medium, was used as a positive control.

After incubation, cells were washed with MEM, and fxed

with 4% formaldehyde in PBS, for 20 min at room tem-

perature. Cells were fnally mounted in mounting media

(Fluoroshield with DAPI, Sigma). Images were captured

using a Zeiss Axioplan2 microscope equipped with FITC

flter. Several images were randomly collected in each slide.

Green cytosolic DCF fuorescence intensity is directly pro-

portional to intracellular ROS generated and was quantifed

with ImageJ v1.51 k. Fluorescence intensity data were nor-

malized to the control (H₂DCF-DA only).

6. Dehdashti F, Mintun MA, Lewis JS, Bradley J, Govindan R,

Laforest R, Welch MJ, Siegel BA (2003) In vivo assessment

of tumor hypoxia in lung cancer with 60Cu-ATSM. Eur J Nucl

Med Mol Imaging 30(6):844–850

7. Holland JP, Lewis JS, Dehdashti F (2009) Assessing tumor

hypoxia by positron emission tomography with Cu-ATSM. Q J

Nucl Med Mol Imaging 53(2):193–200

8. Carlin S, Humm JL (2012) PET of hypoxia: current and future

perspectives. J Nucl Med 53(8):1171–1174

9. Jansson PJ, Kalinowski DS, Lane DJR, Kovacevic Z, Seebacher

NA, Fouani L, Sahni S, Merlot AM, Richardson DR (2015) The

renaissance of polypharmacology in the development of anti-

cancer therapeutics: inhibition of the “triad of death” in cancer

by Di-2-pyridylketone thiosemicarbazones. Pharmacol Res 100

(Supplement:C):255–260

10. Guo Z-L, Richardson DR, Kalinowski DS, Kovacevic Z, Tan-

Un KC, Chan GC-F (2016) The novel thiosemicarbazone, di-

2-pyridylketone 4-cyclohexyl-4-methyl-3-thiosemicarbazone

(DpC), inhibits neuroblastoma growth in vitro and in vivo via

multiple mechanisms. J Hematol Oncol 9(1):98

11. Stacy AE, Palanimuthu D, Bernhardt PV, Kalinowski DS, Jans-

son PJ, Richardson DR (2016) Structure-activity relationships

of Di-2-pyridylketone, 2-Benzoylpyridine, and 2-Acetylpyridine

thiosemicarbazones for overcoming Pgp-mediated drug resist-

ance. J Med Chem 59(18):8601–8620

12. Gutierrez EM, Seebacher NA, Arzuman L, Kovacevic Z, Lane

DJ, Richardson V (2016) Lysosomal membrane stability plays

a major role in the cytotoxic activity of the anti-proliferative

agent, Di-2-pyridylketone 4,4-dimethyl-3-thiosemicarbazone

(Dp44mT). Biochim Biophys Acta 1863:1665–1681

Acknowledgements This work was supported by Fundação para

a Ciência e Tecnologia (projects PTDC/QUI–QUI/114139/2009,

RECI/QEQ-QIN/0189/2012, EXCL/QEQ-MED/0233/2012, UID/

Multi/04349/2013 (to C²TN) and UID/Multi/04046/2013 (to BioISI);

fellowships SFRH/BPD/29564/2006, SFRH/BPD/80758/2011, SFRH/

BPD/93017/2013 and SFRH/BPD/112654/2015 to S. Gama, E. Palma

and H.M: Botelho, respectively, and Grants Ciência 2008 to G. Ribeiro

Morais and FCT Investigator to F. Mendes and I. Correia) and Col-

laborative Research Centre ChemBioSys (CRC 1127) funded by the

Deutsche Forschungsgemeinschaft (DFG). The authors would also like

to thank Pedro Reis for the elemental analyses measurements. We also

thank the Fundação para a Ciência e a Tecnologia for fnancial support

through RNEM—Portuguese Mass Spectrometry Network. The authors

declare no competing fnancial interest.

13. Stefani C, Al-Eisawi Z, Jansson PJ, Kalinowski DS, Richardson

DR (2015) Identifcation of diferential anti-neoplastic activity

of copper bis(thiosemicarbazones) that is mediated by intracel-

lular reactive oxygen species generation and lysosomal mem-

brane permeabilization. J Inorg Biochem 152:20–37

14. Palma E, Mendes F, Morais GR, Rodrigues I, Santos IC,

Campello MP, Raposinho P, Correia I, Gama S, Belo D,

Alves V, Abrunhosa AJ, Santos I, Paulo A (2017) Bio-

physical characterization and antineoplastic activity of new

bis(thiosemicarbazonato) Cu(II) complexes. J Inorg Biochem

167:68–79

15. Fan J, Han Z, Kang Y, Peng X (2016) A two-photon fuores-

cent probe for lysosomal thiols in live cells and tissues. Sci Rep

6:19562

References

16. Daum S, Reshetnikov MSV, Sisa M, Dumych T, Lootsik MD,

Bilyy R, Bila E, Janko C, Alexiou C, Herrmann M, Sellner L,

Mokhir A (2017) Lysosome-targeting amplifers of reactive oxy-

gen species as anticancer prodrugs. Angew Chem Int Ed Engl

56(49):15545–15549

1. Santini C, Pellei M, Gandin V, Porchia M, Tisato F, Marzano C

(2014) Advances in copper complexes as anticancer agents. Chem

Rev 114(1):815–862

2. Shuhong C, Xiahui C, Ligen C, Jingwen C (2016) α(N)-

Heterocyclic thiosemicarbazones: iron chelators that are promis-

ing for revival of gallium in cancer chemotherapy. Anti-Cancer

Agents Med Chem 16(8):973–991

17. Pascu SI, Waghorn PA, Conry TD, Betts HM, Dilworth JR,

Churchill GC, Pokrovska T, Christlieb M, Aigbirhio FI, Warren

JE (2007) Designing Zn(ii) and Cu(ii) derivatives as probes for

in vitro fuorescence imaging. Dalton Trans 43:4988–4997

18. Pascu SI, Waghorn PA, Conry TD, Lin B, Betts HM, Dilworth JR,

Sim RB, Churchill GC, Aigbirhio FI, Warren JE (2008) Cellular

confocal fuorescence studies and cytotoxic activity of new Zn(II)

bis(thiosemicarbazonato) complexes. Dalton Trans 16:2107–2110

3. Stacy AE, Palanimuthu D, Bernhardt PV, Kalinowski DS, Jansson

PJ, Richardson DR (2016) Zinc(II)–thiosemicarbazone complexes

are localized to the lysosomal compartment where they trans-

metallate with copper ions to induce cytotoxicity. J Med Chem

59(10):4965–4984

1 3

JBIC Journal of Biological Inorganic Chemistry

of copper(Ii)—structural and biological studies. Coordin Chem

Rev 123(1–2):49–71

19. Alam IS, Arrowsmith RL, Cortezon-Tamarit F, Twyman F,

Kociok-Kohn G, Botchway SW, Dilworth JR, Carroll L, Aboagye

EO, Pascu SI (2016) Microwave gallium-68 radiochemistry for

kinetically stable bis(thiosemicarbazone) complexes: structural

investigations and cellular uptake under hypoxia. Dalton Trans

45(1):144–155

30. Gonzalez-Garcia C, Mendiola MA, Perles J, Lopez-Torres E

(2017) Structural diversity and supramolecular architectures of

Zn(ii), Cu(ii) and Ni(ii) complexes by selective control of the

degree of deprotonation of diacetyl bis(4-isopropyl-3-thiosemi-

carbazone). CrystEngComm 19(7):1035–1044

20. Addison AW, Rao TN, Reedijk J, van Rijn J, Verschoor GC (1984)

Synthesis, structure, and spectroscopic properties of copper(II)

compounds containing nitrogen-sulphur donor ligands; the crystal

and molecular structure of aqua[1,7-bis(N-methylbenzimidazol-

2[prime or minute]-yl)-2,6-dithiaheptane]copper(II) perchlorate.

J Chem Soc, Dalton Trans 7:1349–1356

31. Shao J, Ma ZY, Li A, Liu YH, Xie CZ, Qiang ZY, Xu JY (2014)

Thiosemicarbazone Cu(II) and Zn(II) complexes as potential

anticancer agents: syntheses, crystal structure, DNA cleavage,

cytotoxicity and apoptosis induction activity. J Inorg Biochem

136:13–23

32. Brouwer Albert M (2011) Standards for photoluminescence quan-

tum yield measurements in solution (IUPAC Technical Report).

Pure Appl Chem 83(12):2213

21. Yang W, Ma Z, Yi J, Sun W-H (2017) Quantitative structure-

thermostability relationship of late transition metal catalysts in

ethylene oligo/polymerization. Catalysts 7(4):120

33. Eaton DF (1988) Reference materials for fuorescence measure-

ment. Pure Appl Chem 60(7):1107

22. López-Torres E, Mendiola MA, Rodrı́ guez-Procopio J, Sevilla

MT, Colacio E, Ma Moreno J, Sobrados I (2001) Synthesis and

characterisation of zinc, cadmium and mercury complexes of ben-

zilbisthiosemicarbazone. Structure of cadmium derivative. Inorg

Chim Acta 323(1):130–138

34. Hashemi M, Ghavami S, Eshraghi M, Booy EP, Los MJ (2007)

Cytotoxic efects of intra and extracellular zinc chelation on

human breast cancer cells. Eur J Pharmacol 557(1):9–19

35. Gomes A, Fernandes E, Lima JL (2005) Fluorescence probes

used for detection of reactive oxygen species. J Biochem Biophys

Methods 65(2–3):45–80

23. Christlieb M, Holland JP, Dilworth JR (2010) Investigation of

the UV–Vis absorption of bis(N-methylthiosemicarbazonato) zinc

Zn[ATSM]. Inorg Chim Acta 363(6):1133–1139

36. Cesen MH, Pegan K, Spes A, Turk B (2012) Lysosomal path-

ways to cell death and their therapeutic applications. Exp Cell Res

318(11):1245–1251

24. Cowley AR, Davis J, Dilworth JR, Donnelly PS, Dobson R,

Nightingale A, Peach JM, Shore B, Kerr D, Seymour L (2005)

Fluorescence studies of the intra-cellular distribution of zinc

bis(thiosemicarbazone) complexes in human cancer cells. Chem

Commun 7:845–847

37. SMART, SAINT and SADABS (2005) Bruker AXS Inc., Madi-

son, Wisconsin, USA. SMART, SAINT and SADABS

38. Sheldrick SGM (2004) Bruker AXS Inc. Madison, Wisconsin,

USA

25. Holland JP, Aigbirhio FI, Betts HM, Bonnitcha PD, Burke P,

Christlieb M, Churchill GC, Cowley AR, Dilworth JR, Donnelly

PS, Green JC, Peach JM, Vasudevan SR, Warren JE (2007) Func-

tionalized Bis(thiosemicarbazonato) complexes of zinc and cop-

per: synthetic platforms toward site-specifc radiopharmaceuticals.

Inorg Chem 46(2):465–485

39. Altomare A, Burla MC, Camalli M, Cascarano GL, Giacovazzo

C, Guagliardi A, Moliterni AGG, Polidori G, Spagna R (1999)

SIR97: a new tool for crystal structure determination and refne-

ment. J Appl Crystallogr 32(1):115–119

40. Sheldrick GM (2008) A short history of SHELX. Acta Crystallogr

Sect A: Found Crystallogr 64(Pt 1):112–122

26. Brown OC, Tocher DA, Blower PJ, Went MJ (2015) Crystal struc-

ture of [butane-2,3-dione bis-(4-methyl-thio-semicarbazonato)-

kappa(4) S, N (1), N (1′), S’](pyridine-kappaN)zinc(II). Acta

Crystallogr, Sect E: Crystallogr Commun 71(Pt 11):1349–1351

27. Blower PJ, Castle TC, Cowley AR, Dilworth JR, Donnelly PS,

Labisbal E, Sowrey FE, Teat SJ, Went MJ (2003) Structural trends

in copper(II) bis(thiosemicarbazone) radiopharmaceuticals. Dal-

ton Trans 23:4416–4425

41. Farrugia L (2012) WinGX and ORTEP for Windows: an update.

J Appl Crystallogr 45(4):849–854

42. Macrae CF, Bruno IJ, Chisholm JA, Edgington PR, McCabe P,

Pidcock E, Rodriguez-Monge L, Taylor R, van de Streek J, Wood

PA (2008) Mercury CSD 2.0—new features for the visualiza-

tion and investigation of crystal structures. J Appl Crystallogr

41(2):466–470

28. West DX, Ives JS, Bain GA, Liberta AE, ValdesMartinez J, Ebert

KH, HernandezOrtega S (1997) Copper(II) and nickel(II) com-

plexes of 2,3-butanedione bis(N(3)-substituted thiosemicarba-

zones). Polyhedron 16(11):1895–1905

43. Carpenter AE, Jones TR, Lamprecht MR, Clarke C, Kang IH,

Friman O, Guertin DA, Chang JH, Lindquist RA, Mofat J, Gol-

land P, Sabatini DM (2006) Cell profler: image analysis software

for identifying and quantifying cell phenotypes. Genome Biol

7(10):R100

29. West DX, Liberta AE, Padhye SB, Chikate RC, Sonawane PB,

Kumbhar AS, Yerande RG (1993) Thiosemicarbazone complexes

1 3

Article Doi

Unravelling the antitumoral potential of novel bis(thiosemicarbazonato) Zn(II) complexes: structural and cellular studies

DOI: 10.1007/s00775-018-1629-6

Source and publish data:

Authors:

Article abstract of DOI:10.1007/s00775-018-1629-6

Full text of DOI:10.1007/s00775-018-1629-6

Products guided by the article

R&D Labs maybe for 13456-08-1

Relevant to this article

Hot Product