Increasing the cytotoxicity of Ru(II) polypyridyl complexes by tuning the electronic structure of dioxo ligands

Article abstract of DOI:10.1021/jacs.9b12464

Due to the great potential expressed by an anticancer drug candidate previously reported by our group, namely, Ru-sq ([Ru(DIP)₂(sq)](PF₆) (DIP, 4,7-diphenyl-1,10-phenanthroline; sq, semiquinonate ligand), we describe in this work a structure-activity relationship (SAR) study that involves a broader range of derivatives resulting from the coordination of different catecholate-type dioxo ligands to the same Ru(DIP)₂ core. In more detail, we chose catechols carrying either an electron-donating group (EDG) or an electron-withdrawing group (EWG) and investigated the physicochemical and biological properties of their complexes. Several pieces of experimental evidences demonstrated that the coordination of catechols bearing EDGs led to deep-red positively charged complexes 1-4 in which the preferred oxidation state of the dioxo ligand is the uninegatively charged semiquinonate. Complexes 5 and 6, on the other hand, are blue/violet neutral complexes, which carry an EWG-substituted dinegatively charged catecholate ligand. The biological investigation of complexes 1-6 led to the conclusion that the difference in their physicochemical properties has a strong impact on their biological activity. Thus, complexes 1-4 expressed much higher cytotoxicities than complexes 5 and 6. Complex 1 constitutes the most promising compound in the series and was selected for a more in depth biological investigation. Apart from its remarkably high cytotoxicity (IC₅₀ = 0.07-0.7 μM in different cancerous cell lines), complex 1 was taken up by HeLa cells very efficiently by a passive transportation mechanism. Moreover, its moderate accumulation in several cellular compartments (i.e., nucleus, lysosomes, mitochondria, and cytoplasm) is extremely advantageous in the search for a potential drug with multiple modes of action. Further DNA metalation and metabolic studies pointed to the direct interaction of complex 1 with DNA and to the severe impairment of the mitochondrial function. Multiple targets, together with its outstanding cytotoxicity, make complex 1 a valuable candidate in the field of chemotherapy research. It is noteworthy that a preliminary biodistribution study on healthy mice demonstrated the suitability of complex 1 for further in vivo studies.

Full text of DOI:10.1021/jacs.9b12464

Subscriber access provided by Karolinska Institutet, University Library
Article
Increasing the Cytotoxicity of Ru(II) Polypyridyl Complexes
by tuning the Electronic Structure of Dioxo Ligands
Anna Notaro, Marta Jakubaszek, Nils Rotthowe, Federica Maschietto, Robin Vinck, Patrick S. Felder,
Bruno Goud, Mickael Tharaud, Ilaria Ciofini, Fethi Bedioui, Rainer F. Winter, and Gilles Gasser
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b12464 • Publication Date (Web): 28 Feb 2020
Downloaded from pubs.acs.org on February 28, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Increasing the Cytotoxicity of Ru(II) Polypyridyl
Complexes by tuning the Electronic Structure of Dioxo
Ligands
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
a,b
c
d
Anna Notaro, Marta Jakubaszek, Nils Rotthowe, Federica Maschietto, Robin
a
a
b
e
d
Vinck, Patrick S. Felder, Bruno Goud, Mickaël Tharaud, Ilaria Ciofini, Fethi
f
c
a,*
Bedioui, Rainer F. Winter, and Gilles Gasser
a
Chimie ParisTech, PSL University, CNRS, Institute of Chemistry for Life and
Health Sciences, Laboratory for Inorganic Chemical Biology, F-75005 Paris,
France.
b
c
Institut Curie, PSL University, CNRS UMR 144, Paris, France.
Department of Chemistry, University of Konstanz, Universitätsstrasse 10, D-78457
Konstanz, Germany.
d
e
f
Chimie ParisTech, PSL University, CNRS, Institute of Chemistry for Life and
Health Sciences, Chemical Theory and Modelling Group, F-75005 Paris, France.
Université de Paris, Institut de physique du globe de Paris, CNRS, F-75005 Paris,
France.
Chimie ParisTech, PSL University, CNRS, Institute of Chemistry for Life and
Health Sciences, Team Synthèse, Electrochimie, Imagerie et Systèmes Analytiques
pour le Diagnostic, F-75005 Paris, France.
*
Corresponding author: E-mail: gilles.gasser@chimeparistech.psl.eu; WWW:
www.gassergroup.com; Phone: +33 1 44 27 56 02
ORCID Number
Anna Notaro: 0000-0003-0148-1160
Marta Jakubaszek: 0000-0001-7590-2330
Nils Rotthowe: 0000-0002-0073-136X
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Federica Maschietto: 0000-0002-5995-2765
Patrick S. Felder: 0000-0003-2325-9026
Bruno Goud: 0000-0003-1227-4159
Fethi Bedioui: 0000-0002-0063-4412
Ilaria Ciofini: 0000-0002-5391-4522
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Rainer F. Winter: 0000-0001-8381-0647
Gilles Gasser: 0000-0002-4244-5097
Keywords: Bioinorganic Chemistry, Cancer, DNA, Medicinal Inorganic Chemistry,
Ruthenium.
2
ACS Paragon Plus Environment

Page 3 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Abstract
Due to the great potential expressed by an anticancer drug candidate previously
reported by our group, namely Ru-sq ([Ru(DIP) (sq)](PF ) (DIP: 4,7-diphenyl-1,10-
2
6
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
phenanthroline, sq: semiquinonate ligand), we describe in this work a structure-activity
relationship (SAR) that involves a broader range of derivatives resulting from the
coordination of different catecholate-like dioxo ligands to the same Ru(DIP) core.
2
More in detail, we chose catechols carrying either electron-donating or electron-
withdrawing groups EDG or EWG and investigated the physico-chemical and
biological properties of their complexes. Several pieces of experimental evidences
demonstrated that the coordination of catechols bearing EDGs led to deep red positively
charged complexes 1–4 in which the preferred oxidation state of the dioxo ligand is the
uninegatively charged semiquinonate. Complexes 5 and 6, on the other hand, are
blue/violet neutral complexes which carry an EWG substituted dinegatively charged
catecholate ligand. The biological investigation of complexes 1–6 led to the conclusion
that the difference in their physico-chemical properties has a strong impact on their
biological activity. Thus, complexes 1–4 expressed much higher cytotoxicities than
complexes 5 and 6. Complex 1 constitutes the most promising compound of the series
and was selected for a more in-depth biological investigation. Apart from its
remarkably high cytotoxicity (IC = 0.07–0.7 µM in different cancerous cell lines)
50
complex 1 was taken up by HeLa cells very efficiently by a passive transportation
mechanism. Moreover, its moderate accumulation in several cellular compartments (i.e.
nucleus, lysosomes, mitochondria and cytoplasm) is extremely advantageous in the
search of a potential drug with multiple modes of action. Further DNA metalation and
metabolic studies pointed to the direct interaction of complex 1 with DNA and to the
severe impairment of the mitochondrial function. Multiple targets, together with its
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
outstanding cytotoxicity, make complex 1 a valuable candidate in the field of
chemotherapy research. Noteworthy, a preliminary biodistribution study on healthy
mice demonstrated the suitability of complex 1 for further in vivo studies.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
ACS Paragon Plus Environment

Page 5 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Introduction
The worldwide approval of the anticancer drug cisplatin and later of carboplatin and
oxaliplatin has made platinum-based drugs the leading compounds in the field of
medicinal inorganic chemistry.1,2 However, the side-effects and resistance associated
with these compounds have spurred numerous investigations into other metal-based
drugs as potential chemotherapeutic agents against cancer.1,3–6 Ruthenium complexes
are presently regarded as the successors/alternatives to the aforementioned platinum
compounds.7–14 Three ruthenium complexes were (i.e., NKP-1019 and NAMI-A,
Figure 1) or are (i.e., IT-139, Figure 1) in clinical trials as chemotherapeutic agents. In
addition, a Ru(II) polypyridyl complex, namely TLD-1433 (Figure 1) has just entered
phase II of clinical trials as a photosensitizer for photodynamic therapy against bladder
cancer.15–20 Ru(II) polypyridyl complexes, have been also extensively studied for their
cytotoxicity and sowed a great potential ad as potential chemotherapeutic agents.21–28
Recently, we reported a detailed study on a very promising Ru(II) polypyridyl complex,
namely [Ru(DIP) (sq)](PF ) (Ru-sq, Scheme 1a), where DIP is 4,7-diphenyl-1,10-
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
6
2
9
phenanthroline and sq is a semiquinonate ligand. Semiquinonate (sq) is the product
of the first one-electron oxidation of catecholate (cat) and can be further oxidized to
quinone (q) (Scheme 1b).30,31 Catechol(ate) and its redox congeners semiquinonate and
3
2
1
,2-benzoquinone are pivotal examples for the class of ‘non-innocent’ ligands. This
definition is more appropriately referring to a pair of a metal ion and a ligand which are
both redox-active and whose frontier orbitals are strongly mixed, defying an
3
2
unambiguous assignment of redox states to either component. Catechols are also
considered as pan-assay interference compounds (PAINS) due to their chelating and
redox properties.33
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
+
X
S
HN
O
H
N
S
S
Ru
N
N
N
H
N
N
N
N
N
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
S
HN
Ru
N
Ru
N
N
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
NH
HN
NAMI-A
IT-139 (formerly NKP-1339) X= Na
TLD-1433
H
N
NKP-1019 X=
HN
Figure 1. Chemical structures of Ruthenium complexes that were (i.e., NKP-1019 and
NAMI-A, Figure 1) or are (i.e., IT-139 and TLD-1433) in clinical trials.
Ru-sq was, to the best of our knowledge, the first Ru(II) polypyridyl complex carrying
a semiquinonate moiety which was investigated in-depth as an anticancer drug
candidate. We could demonstrate that it is a valuable candidate as a chemotherapeutic
agent, both in vitro and in vivo. Interestingly, its mechanisms of action involve more
than one cellular target. This finding could potentially be a key feature to overcome
resistance, which is an inherent problem for platinum-based anticancer drug
2
9
candidates. Driven by the promising properties unveiled for Ru-sq, we undertook a
structure-activity relationship (SAR) study, keeping the same Ru(II) polypyridyl core
(
i.e. Ru(DIP) ), but substituting the catechol-type dioxo ligand. Over the years, many
2
studies have been performed focusing on the non-innocent character of different metal
complexes both from an experimental and theoretical point of view.34–41 In 2006, Wada
and co-workers demonstrated how the oxidation state of the dioxo ligand in a given
metal-coligand environment depends on the nature of its substituents.42 More
specifically, the authors investigated a series of [Ru(OAc)(dioxolene)(terpy)]
complexes with dioxo ligands carrying electron withdrawing (EWG) or electron
6
ACS Paragon Plus Environment

Page 7 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
42
donating (EDG) groups. In particular, they observed that due to the non-innocent
II
nature of the dioxo ligand, all complexes lie on a continuum in between formal Ru (sq)
III
II
and Ru (cat) descriptions, with an increase of the contribution of the Ru (sq) form for
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
42
complexes with an EDG-substituted dioxo ligand and vice versa. Herein, we present,
a new series of complexes carrying dioxo ligands substituted with EDGs or EWGs,
namely [Ru(DIP) (3-methoxysq)](PF ) (1), [Ru(DIP) (3-methylsq)](PF ) (2),
2
6
2
6
[
Ru(DIP) (4-methylsq)](PF )
(3),
[Ru(DIP) (4-tert-butylsq)](PF )
(4),
2
6
2
6
Ru(DIP) (tetrabromocat) (5) and Ru(DIP) (4-nitrocat) (6) (Scheme 1a), where DIP is
2
2
4
,7-diphenyl-1,10-phenanthroline, 3-methoxysq is 3-methoxysemiquinonate, 3-
methylsq is 3-methylsemiquinonate, 4-methylsq is 4-methylsemiquinonate, 4-tert-
butylsq is 4-tert-butylsemiquinonate, tetrabromocat is tetrabromocatecholate and 4-
nitrocat is 4-nitrocatecholate. The physico-chemical and biological properties of these
compounds were investigated in-depth. As described in the literature,34–41 the variation
of the electron density on the dioxo ligand leads to the modification of its oxidation
state when bound as a ligand. The different oxidation states of the dioxo ligands have a
strong bearing on the physical properties of the complexes, affecting, among other
parameters, their electronic structures and hence the charge state. Of note, we could
unveil an interesting correlation between the oxidation state of the dioxo ligand and the
biological activity of the complex. Complex 1 was found to be the most promising
compound investigated. Its remarkable cytotoxicity in 2D models (IC values in the
5
0
low nanomolar range) was confirmed in a more accurate 3D model, which led to further
biological investigation. In the course of this work, we demonstrate how the mechanism
of action of complex 1 involves multicellular targets. This is considered a key factor to
overcome one of the main drawbacks associated to chemotherapy treatments, which is
the occurrence of resistance. Moreover, as most complexes of this kind, complex 1
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
43
displays poor water solubility. Recently, we reported the formation of colloids of Ru-
sq and complex 1 in water-DMSO (1% v/v) mixture, which could have been easily
4
3
mistaken by solutions if no appropriate characterization had been performed. Poor
aqueous solubility is usually considered a limitation for in vivo applications. However,
in this study, we could demonstrate that with an appropriate formulation, complex 1 is
able to distribute in healthy BALB/c mice, which renders it a suitable candidate for
further in vivo studies.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a)
b)
O-
O-
cat
N
Ru(DMSO)4Cl2
I
N
N
N
+ -
-e-
e
Ru
Cl
Cl
O
O
+
PF6-
II
III
sq
+e-
-e-
N
N
O
N
N
Ru
O
O
O
R4
R3
q
R1
N
R2
N
O
N
N
Ru
Ru-sq: R =R =R =R =H
1
2
3
4
1
2:
3
4
:
R =R =R = H
R =OCH
R4
R3
O
2
3
4
1
3
R =R =R = H
2
3
4
R =CH
R =CH
1
2
3
3
:
:
R =R =R =H
1
3
4
R1
R =R =R = H
1
3
4
2 3 3
R =C(CH )
R2
5
: R =R =R =R = Br
1
2
3
4
6: R =R =R = H
2
R =NO2
1
3
4
Scheme 1. a) Syntheses of complexes 1–6. I) DIP, LiCl, DMF, reflux, 24 h, 78%; II) (i)
NaOH, 3-methoxycatechol (1), or 3-methylcatechol (2), or 4-methylcatechol (3) or 4-
tert-butylcatechol (4) 2-propanol, reflux, 24h; (ii) air, 2 h; (iii) NH PF , 2-propanol/H O
4
6
2
(1:8), 19% (Ru-sq), 23% (1), 24% (2), 29% (3), 16% (4). III) (i) NaOH,
tetrabromocatechol (5) or 4-nitrocatechol (6), 2-propanol, reflux, 24h; (ii) air, 2 h; (iii)
NH PF , 2-propanol/H O (1:8), 54% (5), 27% (6). b) Catecholate (cat) and its oxidised
4
6
2
forms, semiquinonate (sq) and quinone (q).
8
ACS Paragon Plus Environment

Page 9 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Results and Discussion
Synthesis and characterization of 1–6
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The synthesis of compounds 1–6 was achieved by adapting a previously reported
2
9
44
procedure. Ru(DIP) Cl was obtained starting from the known Ru(DMSO) Cl as
2
2
2
2
previously reported.45 This precursor complex was then refluxed under nitrogen
atmosphere with the corresponding catechol derivative in the presence of NaOH in 2-
2
9
propanol overnight. Subsequently, the reaction vessel was opened to air for 2 h to
allow for the final oxidation step. Interestingly, oxidation of the catecholate ligand to a
semiquinonate occurred only for those complexes bearing catecholate ligands with
EDG groups (complexes 1–4 in Scheme 1a). When the synthesis involved catechol
derivatives bearing EWG groups, neutral complexes (5, 6) were generated. The drastic
change of the electronic properties of the complexes studied in this work is also
reflected by their color. EDG-containing derivatives are deep red solids while EWG-
containing derivatives are blue/violet. UV/Vis/NIR spectroscopy in DMF solution
(Figure 2) confirms the assignment of the EDG-substituted ligands as semiquinonates
as compounds 1-4 display a band at ca. 885-900 nm. This absorption is characteristic
for ruthenium-bis(diimine) complexes containing a semiquinonate ligand and is due to
46
a Ru(II)  sq transition. A set of bands, which is responsible for the visual color
impression of these compounds, is found in the range of 525 to 470 nm. Based on
literature data on similar compounds, this band can be ascribed to MLCT and LL’CT
transitions (MLCT = metal-to-ligand charge-transfer; LL’CT = ligand-to-ligand’
charge-transfer, electron transfer between a donor and an acceptor ligand). The two
neutral compounds 5 and 6 are both characterized by a broad transition at 675 nm
alongside two additional bands at ca. 600 and 500 nm; both of which are responsible
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
for their blue/violet color. Their similarity to the absorptions in the precursor complex
Ru(DIP) Cl is underlined by its similarly convoluted absorption profile with bands at
2
2
680, 570 and 485 nm. These data are in agreement with a previous report for the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
47
Ru(bpy) Cl analogue whose bands can be assigned as Ru(II)  DIP transitons. An
2
2
overview over the characteristic absorption bands of all compounds is provided in Table
S1.
Figure 2. UV/Vis/NIR spectra in DMF solution of the complexes 1-6 and their
precursor Ru(DIP) Cl .
2
2
The paramagnetic nature of compounds 1–4 was indirectly confirmed by their
1
characteristically broadened H-NMR spectra (Figure S1). In contrast, compounds 5
and 6, whose purity was confirmed by the elemental analysis, are neutral and formally
diamagnetic. However, the proton NMR spectra also provided severely broadened
resonances. The addition of zinc dust to the NMR tube, slightly improves the resolution
1
0
ACS Paragon Plus Environment

Page 11 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
of NMR signals which suggests the formation of paramagnetic species in solution
(Figure S1 (5) and (6)). All complexes were characterized by ESI-MS, and their purity
was verified by HPLC (Figure S2) and elemental microanalysis.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
EPR, (Spectro-)electrochemistry and Electronic Structures
The oxidation state of the catecholate/semiquinonate ligand of complexes 1–6 was
further investigated by electron paramagnetic resonance (EPR) spectroscopy, cyclic
voltammetry (CV), rotating disc electrode voltammetry (RDE) as well as UV/Vis/NIR
spectroelectrochemistry and supported by density functional theory (DFT) calculations.
Compounds 1–4 are, as obtained from the synthesis, EPR active. At room temperature,
they display a rather broad, featureless, isotropic signal (Figure 3A), which becomes
slightly anisotropic at lower temperatures and in the frozen glass (Figure S3). The
substituents on the dioxo ligand and their positions seem to have only little influence
on the locus and the distribution of the unpaired spin density. Thus, the g values of
iso
1.9893, 1.9891, 1.9872 and 1.9840 for 1–4, respectively, are all close to the free
electron value g of 2.0023 and differ only slightly from each other.
e
1
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 3. (A) EPR spectra of (from top to bottom) 1–4 at room temperature. (B) RDE
and CV voltammograms of 1 (from -2.1 to +1.0 V) at a glassy carbon electrode in DMF
(1 mM) containing NBu PF (100 mM) as supporting electrolyte and
4
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
decamethylferrocene as an internal standard (0.25 mM). Data were recorded versus a
saturated calomel electrode (SCE) at a scan rate of 100 mV/s and recalculated versus
0
/+
the Me Fc potential value (E1/2 = 0.040 V vs SCE, feature marked with * in Figure
10
3
B).
These data are also confirmed by the spin density distributions computed for
compounds 1 and 3 reported in Table 1 and graphically depicted in Figure 4 (see the
Experimental Section for computational details). Indeed, roughly 82% of the spin
density is localized on the dioxo ligand with only limited delocalization onto the Ru
center (0.19|e-|). Consequently, the computed g-tensors (reported as Table S2 in the SI)
show very small anisotropies and shifts with respect to the free electron value. These
results are consistent with the assumption that the 1,2-dioxo ligand can be formally
represented by its semiquinonate form in the case of complexes 1–4.
Table 1. Computed Mulliken spin densities on the different fragments of the cationic
forms of the complexes.
1
3
5ox
6ox
dioxo
ligand
0.8200
-0.0090
-0.0093
0.1983
0.8268
-0.0091
-0.0089
0.1913
0.6276 0.6003
-0.0077 -0.0062
-0.0078 -0.0098
0.3880 0.4157
ancillary1
ancillary2
Ru
1
2
ACS Paragon Plus Environment

Page 13 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Isodensity plots of computed spin densities. Positive spin density corresponds
to blue lobes.
To further investigate the electrochemical properties of compounds 1–4 and to obtain a
better insight into the effect of the substituent’s position and nature on the redox
behavior of the complexes, cyclic voltammetry (CV) and rotating disc electrode (RDE)
analyses were performed. Complexes 1–4 display reduction as well as oxidation
features in line with what we have reported in our previous work on Ru-sq, carrying
the non-substituted semiquinonate ligand.29 Figure 3 shows the RDE and cyclic
voltammograms of compound 1 while those of 2–4 are available in the SI (Figure S4).
Like all other complexes bearing EDG-substituted dioxo ligands, complex 1 exhibits
four well-defined reversible redox processes. The RDE experiment shows four features
with very similar limiting currents, which indicates that the same number of electrons
2
9
are exchanged during every process. Based on our previous results, the redox process
0/+
at +0.595 V vs Me Fc can be attributed to the oxidation of Ru(II) to Ru(III) while
1
0
0
/+
the first reduction process at -0.201 V vs Me Fc can be attributed to the sq/cat redox
1
0
process. UV/Vis/NIR-spectroelectrochemistry further supports this hypothesis. As
evident from Figures 5A and 5B, reducing the complexes 1 or 2 at sufficiently negative
potential yields spectra, whose general band structures closely resemble those of
compounds 5 and 6. The observed red-shift with respect to the latter complexes is the
1
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
result of the higher-lying donor orbitals (cat) and (Ru(II)), which is due to electron rich
nature of dioxo ligand in 1. Oxidation to the corresponding dications (Figure 5C and
Figure 5D) 1ox and 2ox leads to a general blue-shift of the Vis/NIR bands. The first
visible transition is now found at 675 nm for both complexes. Compared to the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-1 46
Ru(bpy) (cat) reference, this corresponds to a red-shift by 810 cm . Adhering to the
2
assignment of a Ru(II)  diimine MLCT transition this effect can be explained through
the presence of a more extended -system in the DIP ligand as compared to bpy. We
note, however, that a Ru(III)-sq scenario resulting in a dioxo ligand-to-metal charge-
transfer (LMCT) is also possible. The strong electronic coupling between these two
redox-active entities prohibits a clear assignment of redox states without additional
experimental support or quantum chemical calculations.
Figure 5. UV/Vis/NIR-spectroelectrochemistry data for compounds 1 and 2.
The additional two reduction processes at the more negative potentials of -1.463 V and
-1.758 V, respectively, are due to the stepwise one-electron reductions of the ancillary
1
4
ACS Paragon Plus Environment

Page 15 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
ligands (i. e. the DIP0/- couples). These processes were not further studied by
spectroelectrochemical investigations due to the high reduction potentials and the
irrelevance of these processes to the bioactivities of these complexes. Table 2
summarizes the redox potentials associated with compounds 1–4 and compares them
to those of Ru-sq carrying the unsubstituted semiquinonate ligand (see full data in
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
9
Table S3 of the Supporting Information). Comparing the results, it is clear that the
presence of an EDG at the dioxo ligand increases the electron density on the metal
center, hence shifting its redox process cathodically. These observations are perfectly
in agreement with previous work from Wada and co-workers on the charge distribution
4
2
between the ruthenium ion and the dioxo ligand. The stronger EDGs OMe and tBu
have the predictable effect of shifting the redox potentials of the sq/cat and RuII/III redox
couples cathodically, with however, a different ordering of these couples for complexes
1 and 4.
Table 2. Electrochemical data for 1–4 and Ru-sq.
DIP0/-
DIP0/-
Sq/cat
RuII/III
a
E
1/2 [V]
Ru-sq*
-1.876 ± 0.039
-1.578 ± 0.035
-1.537 ± 0.008
-1.497 ± 0.012
-1.472 ± 0.070
-1.515 ± 0.005
-0.249 ± 0.010
-0.284 ± 0.005
-0.252 ± 0.011
-0.265 ± 0.019
-0.271 ± 0.008
0.647 ± 0.018
0.595 ± 0.011
0.615 ± 0.003
0.636 ± 0.011
0.574 ± 0.001
(
RDE)
a
E
1/2 [V]
-1.848 ± 0.015
-1.833 ± 0.007
-1.836 ± 0.028
-1.839 ± 0.017
1
2
3
4
(RDE)
a
E
1/2 [V]
(RDE)
a
E
1/2 [V]
(RDE)
a
E
1/2 [V]
(RDE)
*
29
Values taken from[ ] We however note that these experiments were performed on the same days.
a
E1/2 = half-wave.
1
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
While compounds 1-4 are easily oxidized by air, their reduction to the respective neutral
forms within the cell is also feasible. The glutathione system (GSSG/2GSH redox
couple) is the most abundant redox couple in a cell and therefore was selected for this
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
48
experiment. Nearly full conversion of 1 to its neutral form could be achieved using 5
equivalents of glutathione in a 1:1 mixture of DMF and water containing 20 mM of
NaHCO at 40°C, strongly suggesting that compounds 1-4 can at least be partially
3
reduced to their neutral form inside the cell (see Figure S5). Slightly alkaline conditions
were thereby necessary in these experiments, most likely to either lower the oxidation
potential of the thiol functional group on the cysteine building block through
deprotonation or to prevent the formation of HPF (note that in a biological setting basic
6
amino acids or common intracellular buffers can easily serve this function). In addition,
these measurements underline the charge-transfer nature of all transitions in the
Vis/NIR region as all corresponding absorptions experience a blue-shift with an
increase in solvent polarity; i.e., these complexes display negative solvatochromism.
Remarkably, the presence of one or more EWGs on the dioxo ligand causes a distinct
change in the electronic structure of the complex. The CV and RDE voltammograms
of 5 are shown in Figure 6A, while the voltammograms of compound 6 are available in
the SI (Figure S4). The voltammograms of 5 were recorded over two distinct potential
ranges, from -1.0 V to -2.0 V and from -1.0 V to 1.0 V to avoid the adsorption of the
complex on the electrode (Figure S4).
1
6
ACS Paragon Plus Environment

Page 17 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. (A) RDE and cyclic voltammograms of complex 5 (from -2.0 to -1.0 V and
from -1.0 to +1.0 V) at a glassy carbon electrode in DMF (1 mM) containing NBu PF
4
6
(100 mM) as supporting electrolyte. Data were recorded versus SCE at scan rate of 100
0
/+
mV/s and recalculated versus Me Fc potential value (-0.001 V and -0.002 V for RDE
10
and CV respectively). (B) EPR spectra of complexes 5ox and 6ox at -140 °C.
Complexes 5, 6 are EPR silent from 20 °C down to -150 °C which confirms that the
dioxo ligand is present in its catecholate form as found by UV/Vis/NIR spectroscopy
and the elemental analysis. To provide an attribution of the observed voltammetric
features in the positive potential range, further EPR experiments were conducted on the
monooxidized forms of 5 and 6 (5ox and 6ox, respectively, in Scheme 2). Complexes
5
and 6 were converted to their monocationc forms by treatment with an excess of
ferrocenium hexafluorophosphate (FcHPF , E = 0.450 V vs SCE in DMF/0.1 M
6
1/2
4
9
NBu PF ).
4
6
1
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
+
+
N
N
N
-e-
N
O
N
N
N
O
N
N
N
O
N
N
Ru
O
Ru
Ru
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
+ -
e
R4
R3
O
R4
R3
R4
R3
O
R1
R1
R1
R2
R2
R2
5: R1=R2=R3=R4= Br
R1=R3=R4= H
5ox: R1=R2=R3=R4= Br
6ox: R1=R3=R4= H
6
:
R2=NO2
R2=NO2
Scheme 2. Structures of 5, 6 and 5ox, 6ox represented as two resonance structures. Ru
oxidation states +II and +III are marked in green and pink respectively.
5ox and 6ox are EPR active, giving slightly anisotropic signals at -140 °C with a g of
av
2
.019 for 5ox and of 2.032 for 6ox (Figure 4 and 6B). The absence of an EPR signal at
room temperature for the oxidized complexes is explained by the electron-poor nature
of the per-brominated- and nitro-substituted ligands, which increases the contribution
of the Ru center to the spin density in the case of the oxidized complexes at the expense
5
0
of the dioxo ligand. Indeed, as reported in Table 1, the spin density on the Ru center
increases to 0.4 |e-| in the case of 5ox and 6ox with a corresponding spin density on the
dioxo ligand of roughly 0.6 |e-|. Therefore, 5ox and 6ox have their spin density nearly
equally shared between the Ru ion and the dioxo ligand, corresponding to a strong
orbital mixing. This can be represented as the two resonance structures of Scheme 2,
where the contribution of the Ru(III)/cat form is nearly as important as the Ru(II)/sq
one. A higher metal contribution to the overall spin densities is also indicated by distinct
anisotropy of the g-tensor, which is more clearly evident for 6ox (i.e., showing the
larger computed spin density) with individual g-tensor components of g = 1.870, g =
x
y
2
.025, g = 2.190 as determined by simulation of the experimental spectrum (see Figure
z
1
8
ACS Paragon Plus Environment

Page 19 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
S3 of the SI). Table 3 and S4 list the potentials related to the redox processes observed
for compounds 5, 6.
Table 3. Electrochemical data for complexes 5 and 6.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Phen0/-
Phen0/-
Ph
2
Ph
2
Ox1
Ox2
a
E
1/2 [V]
5
6
-1.746 ± 0.003
-1.423 ± 0.002
-1.440 ± 0.009
0.182 ± 0.005
0.927 ± 0.008
(
RDE)
a
E
1/2 [V]
-1.737 ± 0.009
0.164 ± 0.013
0.970 ± 0.009
(RDE)
a
E1/2 = half-wave.
To shed additional light on the impact of the substituent on the dioxo ligand, complex
was also subjected to UV/Vis/NIR spectroelectrochemical investigations in the
5
positive potential range. As evident from Figure 7A, generation of the monocation
furnishes a structured, broad band with a maximum at 940 nm, while the features,
characteristic to the neutral forms disappear. The significant red-shift of this band as
compared to compounds 1-4 is a clear token of the electron poor nature of the
corresponding semiquinonate ligand, which lowers the energy difference between the
Ru(II) donor and the dioxo ligand acceptor orbitals. The relatively larger loss of
electron density experienced by the ruthenium atom and the rather electron-poor nature
of the electrogenerated semiquinonate ligands in oxidized 5ox are further underlined
by the blue shift of the RuDIP MLCT and sqDIP LL’CT bands in the visible range
to 465 and 410 nm instead of 510 and 470 nm observed for 1. Further oxidation to the
dication, in agreement with the CV studies, leads to decomposition as rereduction after
electrolysis yielded spectra that considerably differed to those of their monocationic or
neutral forms (see Figure 7B).
1
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 7. UV/Vis/NIR-spectroelectrochemistry data for complex 5.
The change from an EDG to an EWG at the sq/cat ligand has therefore two effects.
First, the overall redox potential of the molecule is increased in such a manner that,
applying the same synthetic procedure and under identical conditions, no longer a
paramagnetic, cationic but a neutral, diamagnetic complex is obtained. Secondly, the
spin density in the monocationic form presents distinctly larger contributions from the
ruthenium center when EWG are present. This occurrence results in a slight increase in
EPR signal anisotropy and, more importantly, the absence of a room temperature EPR
signal in 5ox and 6ox.
Stability in DMSO and human plasma
Next, the stability of compounds 1–6 was investigated in DMSO and in human plasma.
DMSO stability is a key factor to consider in medicinal chemistry as this solvent, in
some cases, was found to be problematic for biological experiments for its coordinative
property.51–53 The stability in DMSO was monitored by H NMR spectroscopy over 96
h at room temperature. Overall, all complexes are stable under these conditions as no
significant new peaks appear in the spectra (Figure S6). Only complex 3 shows a certain
degree of degradation after 96 h, displaying a conversion of about 15% of the initial
complex. The spectrum of complex 3 shows new, more resolved peaks which are a
1
2
0
ACS Paragon Plus Environment

Page 21 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
clear indication of the formation of a non-paramagnetic product. Additionally, for
estimating the stability under the physiological conditions, the stability of complex 1
(the most promising of the series) was investigated in human plasma. Complex 1 was
incubated over 96 h in human plasma at 37 °C in the presence of an internal standard
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(caffeine). Figure S7 reveals a linear trend of concentration of complex 1 over time and
a half-life of 48h.
Cytotoxicity Studies
The first step toward the biological investigation of complexes 1–6 was the evaluation
of cell viability in monolayer cultures of HeLa (human cervical adenocarcinoma) cell
line, A2780 (human ovarian carcinoma), A2780 cis (human cisplatin resistant ovarian
carcinoma), A2780 ADR (human doxorubicin resistant ovarian carcinoma), CT-26
(mouse colon adenocarcinoma), CT-26 LUC (mouse colon adenocarcinoma stably
expressing luciferase) and RPE-1 (human normal retina pigmented epithelial) cell lines
using a fluorometric cell viability assay (single graphs available in Figures S8).54
Doxorubicin and cisplatin, as well as the ligands and the Ru(DIP) Cl precursor were
2
2
tested in the same cell lines as positive and additional controls (Table 4 and Table
S5).55,56 Table 4 displays the IC (the half maximal inhibitory concentration) values of
5
0
29
the tested compounds and the previously reported Ru-sq. Ru-sq is the analogous
complex carrying the unsubstituted semiquinonate ligand. The comparison between
Ru-sq and the derivatives carrying an EDG and EWG substituted dioxo ligand
(
compounds 1–4 and 5, 6, respectively) reveals how the electron density on the organic
moiety impacts the observed cytotoxicity in the tested cell lines. In general, compounds
–4 present higher cytotoxicity in most of the cell lines tested. In contrast, complexes
5 and 6 show much lower cytotoxicity with IC in the micromolar range for all the cell
1
5
0
lines tested. Of particular interest is complex 1 with an IC in the low nanomolar range
5
0
2
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
0.07 μM) against the cisplatin resistant cell line which makes it 10 times more active
than doxorubicin and around 200 times more active than cisplatin (0.54 μM and 18.33
μM for doxorubicin and cisplatin, respectively). Complex 1 presents an IC of 0.7 μM
5
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
against the doxorubicin resistant cell line, which is 10 times lower than the values for
Ru-sq and cisplatin (4.13 μM and 8.32 μM for Ru-sq and cisplatin, respectively).
Overall, complexes 1–4 display a cytotoxicity which is comparable to that of
doxorubicin and much higher than that of cisplatin. The Ru(DIP) Cl precursor (Table
2
2
4
) and the different catechols (Table S4), display a much lower activity suggesting that
the great activity shown by complexes 1–4 is the consequence of the coordination of
electron rich sq ligands to the Ru(II) polypyridyl core.
Table 4. IC50 values for 1–6, Ru-sq, cisplatin, doxorubicin and Ru(DIP) Cl on
2
2
different cell lines.
A2780
ADR
CT-26
LUC
IC50 (μM)
HeLa
A2780
A2780 cis
CT-26
RPE-1
9.28 ±
.20
4.00 ±
0.76
8.32 ±
0.71
18.33 ±
2.92
2.60 ±
0.18
2.42 ±
0.23
30.24 ±
5.11
Cisplatin*
0
0
0
.34 ±
.02
0.19 ±
0.03
5.94 ±
0.58
0.54 ±
0.04
0.082 ±
0.003
0.18 ±
0.006
0.89 ±
0.17
Doxorubicin*
1
5.03 ±
.4
4.69 ±
0.14
78.27 ±
4.9
6.36 ±
0.57
9.20 ±
1.22
3.13 ±
0.07
*
Ru(DIP)
2
Cl
2
6.65 ± 0.5
0
0
0
.50 ±
.01
0.67 ±
0.04
0.45 ±
0.03
1.00 ±
0.03
1.51 ±
0.14
0.90 ±
0.04
Ru-sq*
4.13 ± 0.2
0
0
.124 ±
.004
0.0261 ±
0.0005
0.70 ±
0.05
0.076 ±
0.005
0.067 ±
0.004
0.269 ±
0.007
0.764 ±
0.23
1
2
3
4
5
6
0
0
.353 ±
.006
0.18 ±
0.03
1.05 ±
0.22
0.39 ±
0.07
0.31 ±
0.02
0.24 ±
0.01
0.67 ± 0.2
0
.61 ±
0.07
0.20 ±
0.01
1.45 ±
0.14
0.39 ±
0.03
0.65 ±
0.04
0.42 ±
0.01
0.58 ±
0.01
2
.11 ±
.12
0.53 ±
0.03
1.91 ±
0.08
0.80 ±
0.03
1.167 ±
0.15
1.147 ±
0.224
2.965 ±
0.45
0
1
0.46 ±
.25
10.23 ±
0.14
15.01 ±
0.75
17.17 ±
1.4
13.49 ±
0.5
11.64 ±
0.7
23.15 ±
2.5
0
10.03 ±
0.44
18.63 ±
2.02
16.37 ±
2.04
7.61 ±
0.11
9.01 ±
0.19
16.55 ±
0.98
12.4 ± 0.8
*
29
Values taken from[ ] We however note that these experiments were performed on the same days.
2
2
ACS Paragon Plus Environment

Page 23 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Since complex 1 was found to be the most promising candidate among the series of
complexes investigated in the 2D model due to its remarkably high cytotoxicity and its
great activity towards resistant cell lines, its cytotoxicity was explored in a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
57
MultiCellular Tumour Spheroids (MCTS) model. In 3D spheroids, proper cell to cell
and cell to environment interactions, as well as cellular morphology and polarity is
maintained. Additionally growth pattern, metabolism and gene expression mimic the
complexity of initial stages of solid tumours.57,58 These features allow for a good
estimation of in vivo antitumor activity, qualifying MCTS as a more reliable model than
monolayer cell cultures in cancer research.58–60 In addition to the complex of interest
(1), the Ru(DIP) Cl precursor, the 3-methoxycatechol ligand and positive controls
2
2
(
cisplatin and doxorubicin)59,60 were tested via a luminescent cell viability assay in
HeLa MCTS (single graphs are availabe in Figure S9). Moreover, Ru-sq was tested for
comparative porposes. Table 5 shows the IC values after a 48 h treatment for all the
5
0
compounds tested. Complex 1 displays high cytotoxicity toward HeLa MCTSs with an
IC ≈ 21 µM. This value indicates an activity which is almost the double of the current
5
0
drugs cisplatin and doxorubicin (IC ≈ 46 µM and 39 µM, respectively). Nevertheless,
50
the previously reported Ru-sq still shows a slightly higher cytotoxicity (IC ≈ 14 µM).
5
0
3
-methoxycatechol proved to be nontoxic and the Ru(DIP) Cl precursor showed a
2 2
cytotoxicity comparable to cisplatin.
Table 5. IC50 values for complex 1, cisplatin, doxorubicin, Ru(DIP) Cl and 3-
2
2
methoxycatechol in multicellular HeLa cancer cell spheroids.
3
-
Doxorubic
Ru(DIP)2
Cl2
IC (μM) Cisplatin*
Ru-sq*
1
Methoxyc
atechol
5
0
*
in
HeLa
MCTSs
46.49 ±
4.18
29
38.59 ±
0.43
14.11 ±
0.09
21.01 ±
0.66
59.84 ±
3.05
>
100
*
*
Values taken from[ ] We however note that these experiments were performed on the same days.
2
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
To evaluate the time dependent effect on the growth of MCTS treated with complex 1,
an additional experiment was performed. 400 µm HeLa MCTS were treated with a
range of different concentrations of 1. Every three days, half of the media in the wells
was exchanged (treatment concentration consequently decreased by half) and pictures
of the spheroids were taken (Figure 8a). HeLa MCTS had reduced diameter when
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
treated with concentrations higher than the IC (20, 25 and 30 µM) of complex 1.
50
Additionally, Figure 8 indicates that this effect was maintained even after 13 days for
the highest concentration, while, a slow regrowth is observed for 20 and 25 µM
treatment. These findings are similar to the one previously reported for Ru-sq
complex.29
Figure 8. Changes in growth kinetics of MCTSs treated with complex 1 at different
concentrations (1, 5, 10, 20, 25 and 30 µM). (a) Images collected at day 0 (before
treatment) and at day 3, 6, 9 and 13. b) MCTSs diameter measured at different time
points. Blue dotted line indicates day of seeding, red dashed line indicates day of
treatment, green dotted lines indicate days of washing.
Overall, the outstanding activites shown by complex 1 in the monolayer cells model is
confirmed by a MCTSs model. These findings represent a powerful encouragement to
the further investigation of complexes 1 as potential chemotherapeutic agents.
2
4
ACS Paragon Plus Environment

Page 25 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Cell Death Mechanism.
Many of the novel or existing chemotherapeutic agents are developed to trigger cell
death through apoptosis.63 This is considered a carefully regulated and energy
dependend type of cell death in contrast to necrosis considered a rapid, unregulated,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
64
energy- independent mode of death. The mode of cell death induced by the treatment
with complex 1 was investigated in HeLa cell line via flow cytometry using Annexin
V and propidium iodide (PI) staining method. Figure 9a shows the obtained dot-plots
at t = 24 h (see Figure S10 for the dot plots at each time point). Figure 9b represents the
percentage of cell population in different stages of cell death, at different time points in
7
2
comparison to staurosporin (apoptosis inducer - positive control). Collected data
demonstrated that after 24 h treatment, a large population of HeLa cells was in the late
apoptotic/necrotic stage. It is worth noting that apoptosis or necrosis are induced by
different pathways and mechanisms, and it is highly unlikely for a cell to undergo
necrotic death after several hours of early apoptotic stage. Hence, these data indicate
that complex 1 is most probably inducing cell death through apoptosis.
Figure 9. a) Flow cytometry dot-plots of Annexin V and PI staining in HeLa cells
treated with complex 1 (10 μM) and staurosporine (1 μM) at 24 h. b) Percentage of cell
2
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
population in different stages of cell death for staurosporin (positive control) and
complex
Cellular Uptake, Biodistribution, and DNA Metalation.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The high cytotoxicity of complex 1 in the tested cell lines encouraged its further
biological studies. Firstly, cellular uptake, mechanism of uptake and intracellular
distribution were tested using Inductively Coupled Plasma Mass Spectrometry (ICP-
MS). Working concentrations and incubation times were chosen accordingly to avoid
extended cell mass loss due to the high cytotoxicity of the complexes but considering a
final ruthenium amount that allowed for the determination of the metal content.
Nevertheless, the working conditions (5 µM treatment for 2 h) allowed for a minor
accumulation of the drug cisplatin, which was used as control.65,6667 Complex 1 shows
higher cellular accumulation than the positive control cisplatin and the Ru-sq analogue
2
9
previously reported, (Figure 10a). To clarify whether the mechanism of uptake
involves passive or active mechanisms additional experiments were performed. HeLa
cells were kept at low temperature (4ºC) or were pre-treated with different uptake
pathways inhibitors. Thus, 2-deoxy-D-glucose and oligomycin were used to block
cellular metabolism, chloroquine or ammonium chloride to impede endocytic
pathways, and tetraethylammonium chloride to block cation transporters. After pre-
treatment, cells were incubated with the tested compounds (2 h, 5 µM). The amounts
of ruthenium found in cells were then quantified using ICP-MS. Low temperature
slightly inhibited the uptake of complex 1 while all the other conditions (regulating
active transportation mechanisms) did not affect the total uptake (Figure S11). These
findings strongly suggest that the internalisation of complex 1 is due only to a passive,
energy independent mechanism, unlike [Ru(DIP) (sq)](PF ), whose mechanism of
2
6
29
uptake involves both active and passive transports. Intracellular distribution among
2
6
ACS Paragon Plus Environment

Page 27 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
cytoplasm, mitochondria, nucleus and mitochondria was determined by isolating pure
cellular compartments. Most of the compounds accumulates in nucleus and lysosomes
and in a smaller extent to mitochondria and cytoplasm (Figure 10b). To verify whether
the accumulation in the nucleus lead to direct interaction with DNA, the genetic
material was extracted from treated cells and the metal content analysed via ICP-MS.
Figure 10c shows that complex 1 successfully binds DNA in a much higher extent when
compared to cisplatin, whose results are in perfect agreement with what previously
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
7
reported. Even though the accumulation in the nucleus is lower, the DNA metalation
shows comparable results for complex 1 and Ru-sq, which can be explained by the
overall higher uptake of complex 1.
Figure 10. Cellular uptake (a), cellular fractionation (b) and DNA metalation (c) of
HeLa cells after treatment with tested compounds (5 µM, 2 h). Data are presented as the
mean ± SD of at least 3 technical replicates. All data related to Ru-sq were previously
2
9
reported by our group. We, however, note that these experiments were performed on
the same days.
JC-1 Mitochondrial Membrane Potential Test and Metabolic Studies.
Next, to gain more insights into the possible mechanism of action of complex 1, its
effect on mitochondria membrane potential (MMP) and metabolic pathways was
2
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
68
determined. First, the MMP (directly correlated to mitochondrial function ) was
studied via JC-1 staining method. JC-1 is considered the most reliable fluorescent dye
6
9
to illustrate the MMP in living cells. It is present as an aggregate (red fluorescent) and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
as a monomer (green fluorescent), respectively at high and low MMP.69 The net
6
9
distinction allows for an immediate analysis of the MMP. Figure 11a shows the red
fluorescence signal observed in HeLa cells untreated (negative control) and after 24 h
treatment with complex 1, DMSO (vehicle control) and FCCP (carbonyl cyanide 4-
7
0
(
trifluoromethoxy)phenylhydrazone, an uncoupling agent used as positive control). A
slight decrease in the fluorescence is observed in the cells treated with complex 1 (from
0.05 µM to 0.25µM) in a concentration-dependent manner. However, even at the IC50
concentration, the effect is not as prominent as the one obtained for the positive control.
7
0
It is worth noting that ongoing apoptosis can also generate a drop in MMP. Taken
together, complex 1 is slightly decreasing mitochondria membrane potential in treated
cells. To have more insights into the effect of the complex 1 in the process of oxidative
phospohorylation, further experiments were performed. For this purpose, Seahorse XF
Analyzer was used. The performed Mito Stress Test pointed to very low basal
respiration levels, inhibited ATP production and the loss of the ability to restore proton
balance after FCCP treatment (Figure 11b and Figure S12). These data suggest that the
mitochondria processes are defective in HeLa cells incubated with complex 1. On the
contrary, the same effects were not observed during treatment with the Ru(DIP) Cl
2
2
precursor or the 3-methoxycatechol ligand. Furthermore, no influence of compound 1
on the cytosolic process of glycolysis was detected (Figure S13). Unfortunately, no
direct impact on three main fuel pathways could be determined due to very low oxygen
consumption rates (Figure S14).
2
8
ACS Paragon Plus Environment

Page 29 of 67
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Metabolic studies confirm that the effects of complex 1 on mitochondria respiration can
contribute to cell death, leading to a multiple mode of actions involving at least nucleus
and mitochondria as possible targets.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 11. a) Fluorescence signal of JC-1 dye detected in HeLa cells treated for 24 h
with different concentrations of complex 1 (from 0.05 µM to 0.25µM). The bar marked
in red indicates the IC concentration (0.2 µM). FCCP is used as positive control,
5
0
cisplatin and DMSO (1%) are used as negative controls. b) Mito Stress Test profile
after 24 h treatment; oxygen consumption rate changes after treatment with specific
electron transport chain inhibitors. Oligomycin (inhibitor of ATP synthase (complex
V)), FCCP (uncoupling agent), Antimycin-A (complex III inhibitor) and Rotenone
(complex I inhibitor).
Preliminary in vivo biodistribution studies
In this study, complex 1 demonstrated to be of great interest as potential
chemotherapeutic agent. However, one of its main drawbacks is its scarce water
solubility, which could limit its potential use in vivo. The analogue Ru-sq, previously
reported by our group, showed potential as anticancer drug in immunocompetent mice
2
9
bearing Ehrlich tumors. Nevertheless, its clinical interest is limited by its poor water
2
9
ACS Paragon Plus Environment