A Ru(II)-p-cymene compound bearing naproxen-pyridineamide. Synthesis, spectroscopic studies, computational analysis and in vitro anticancer activity against lung cells compared to Ru(II)-p-cymene-naproxen and the corresponding drug ligands

Article abstract of DOI:10.1016/j.ica.2019.01.030

The design of new Ru(II) organometallics is a subject of interest to the field of anticancer metallodrugs. This work reports the interaction of the Ru(II)-η⁶-p-cymene framework with naproxen-pyridineamide (Npxpya, L1), a structurally modified form of the naproxen (HNpx, HL2) drug, to give the new organometallic [Ru(η⁶-p-cymene)(L1)Cl₂] (1) bearing the Npxpya ligand. The reported naproxenate-derived, [Ru(η⁶-p-cymene)(L2)Cl] (2), is re-prepared, also from the precursor [Ru(η⁶-p-cymene)Cl₂]₂ (3), and additional investigation is performed. The two Ru(II)-arenes and the L1 ligand are fully characterized by ESI-MS, NMR, ATR/FT-IR and UV/VIS, and their structures corroborated by DFT computational calculations. Time-dependent ¹H MNR studies show that both Ru(II)-arenes, despite being stable in non-coordinating solvents, undergo distinct step dissociation in dimethylsulfoxide solvent to give the corresponding drug ligands and [Ru(η⁶-p-cymene)(dmso)Cl₂] (4) species. Electronic absorption spectroscopy experimental data show good correlation with DFT calculations. Organometallics 1 and 2, as well as their corresponding parent drug ligands, exhibit luminescence properties mainly associated to the naproxen moiety. Screening in NCI-H460 and A549 lung cancer cells reveals lack of activity for 2 and L2, while the new organometallic 1 is found to inhibit cell proliferation of both types of cell lines in similar way to the L1 drug. The structural modification, by inserting the pyridineamide moiety into the original structure of naproxen to form the Npxpya conjugated drug, is shown to be crucial for the anticancer activity. Compound 1, despite having IC₅₀ close to the IC₅₀ of L1, does not show significant effect on the mitochondrial membrane potential (MMP), in contrast to the behavior of the free L1 parent drug which significantly decreases the MMP in NCI-H460 cells. Interestingly, since ¹H MNR studies indicate that organometallic 1 is completely dissociated in dmso (the solvent used to prepare the drug solutions for cell treatment in the biological assays) to give the L1 free drug and species 4, it is plausible to infer that the presence of Npxpya-free Ru species, probably in the form of species 4, might play a role in inhibiting the mechanism related to the mitochondrial function when the cells are treated with 1, in comparison with the cell treatment with the L1 free drug.

Full text of DOI:10.1016/j.ica.2019.01.030

Accepted Manuscript
Research paper
A Ru(II)-p-cymene compound bearing naproxen-pyridineamide. Synthesis,
spectroscopic studies, computational analysis and in vitro anticancer activity
against lung cells compared to Ru(II)-p-cymene-naproxen and the correspond-
ing drug ligands
Julie Pauline Gaitan Tabares, Rodrigo Luis S.R. Santos, Jefferson Luiz
Cassiano, Marcio H. Zaim, João Honorato, Alzir A. Batista, Sarah F. Teixeira,
Adilson Kleber Ferreira, Rommel B. Viana, Sandra Quispe Martínez, Antonio
Carlos Stábile, Denise de Oliveira Silva
PII:
S0020-1693(18)31792-4
https://doi.org/10.1016/j.ica.2019.01.030
ICA 18753
DOI:
Reference:
Inorganica Chimica Acta
To appear in:
Received Date:
Revised Date:
Accepted Date:
2 December 2018
25 January 2019
25 January 2019
Please cite this article as: J.P.G. Tabares, R.L.S.R. Santos, J.L. Cassiano, M.H. Zaim, J. Honorato, A.A. Batista,
S.F. Teixeira, A.K. Ferreira, R.B. Viana, S.Q. Martínez, A.C. Stábile, D. de Oliveira Silva, A Ru(II)-p-cymene
compound bearing naproxen-pyridineamide. Synthesis, spectroscopic studies, computational analysis and in vitro
anticancer activity against lung cells compared to Ru(II)-p-cymene-naproxen and the corresponding drug ligands,
Inorganica Chimica Acta (2019), doi: https://doi.org/10.1016/j.ica.2019.01.030
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

A Ru(II)-p-cymene compound bearing naproxen-pyridineamide.
Synthesis, spectroscopic studies, computational analysis and in
vitro anticancer activity against lung cells compared to Ru(II)-p-
cymene-naproxen and the corresponding drug ligands
Julie Pauline Gaitan Tabares a, Rodrigo Luis S. R. Santos a,b, Jefferson Luiz
Cassiano a, Marcio H. Zaim c, João Honorato d, Alzir A. Batista d, Sarah F.
e
Teixeira , Adilson Kleber Ferreira e, Rommel B. Viana f, 1, Sandra Quispe
Martínez a Antonio Carlos Stábile a, Denise de Oliveira Silva a, *
,
a
Laboratory for Synthetic and Structural Inorganic Chemistry - Bioinorganic and
Metallodrugs. Department of Fundamental Chemistry, Institute of Chemistry,
University of São Paulo. Av. Prof. Lineu Prestes, 748, 05508-000, São Paulo, SP,
Brazil.
b Department of Exact and Technological Sciences, State University of Santa Cruz. Rod.
Jorge Amado, Km 16, 45662-900, Ilhéus, BA, Brazil.
c
Catalysis and Transfer Phase Laboratory. Department of Fundamental Chemistry,
Institute of Chemistry, University of São Paulo. Av. Prof. Lineu Prestes, 748, 05508-
000, São Paulo, SP, Brazil.
d
Department of Chemistry, Federal University of São Carlos, CP 676,
13565-905, São Carlos, SP, Brazil.
e
Department of Immunology. Laboratory of Tumor Immunology. Institute of
Biomedical Sciences, University of São Paulo. Av. Prof. Lineu Prestes, 1730, 05508-
000, São Paulo, SP, Brazil.
f
Departament of Chemistry and Molecular Physics, São Carlos Chemistry Institute,
University of São Paulo. Av. Trabalhador São Carlense, 400, 13566-590, São Carlos,
SP, Brazil.
*Corresponding author: Email address: deosilva@iq.usp.br (D. de Oliveira Silva).
1 Current address: Chemistry Institute, Federal University of Alfenas, MG, Brazil.
1

Abstract
The design of new Ru(II) organometallics is a subject of interest to the field of
anticancer metallodrugs. This work reports the interaction of the Ru(II)-η6-p-cymene
framework with the naproxen-pyridineamide (Npxpya, L1), a structurally modified form of
the naproxen (HNpx, HL2) drug, to give the new organometallic [Ru(η6-p-cymene)(L1)Cl2]
(1) bearing the Npxpya ligand. The reported naproxenate derived, [Ru(η6-p-cymene)(L2)Cl]
(2), is re-prepared also from the precursor [Ru(η6-p-cymene)Cl2]2 (3) and additional
investigation is performed. The L1 ligand and the two Ru(II)-arenes are fully characterized
by ESI-MS, NMR, ATR/FT-IR and UV/VIS, and their structures corroborated by DFT
1
computational calculations. Time-dependent HMNR studies show that both Ru(II)-arenes,
despite being stable in non-coordinating solvents, undergo distinct step dissociation in
dimethylsulfoxide solvent to give the corresponding drug ligands and [Ru(η6-p-
cymene)dmsoCl2] (4) species. Experimental electronic absorption spectroscopy data show
good correlation with DFT calculations. Organometallics 1 and 2 as well as their
corresponding parent drug ligands exhibit luminescence properties mainly associated to the
naproxen moiety. Screening in NCI-H460 and A549 lung cancer cells reveals lack of activity
of 2 and L2 while the new organometallic 1 is found to inhibit cell proliferation of both types
of cell lines in similar way to the free L1. The structural modification, by the insertion of the
pyridineamide moiety into the original structure of naproxen, to form the Npxpya conjugated
drug, is shown to be crucial for the anticancer activity. Compound 1, in similar way to species
4 (generated from dissolution of 3 in dmso), and despite having IC50 close to the IC50 of L1,
does not show significant effect on the mitochondrial membrane potential (MMP), in contrast
to the behavior of the L1 parent free drug which significantly decreases the MMP in NCI-
1
H460 cells. Interestingly, since HMNR studies indicate that organometallic 1 is completely
dissociated in dmso (the solvent used to prepare the drug solutions for cell treatment in the
biological assays) to give the L1 free drug and species 4, it is plausible to infer that the
presence of Npxpya-free Ru species, probably in the form of species 4, might play a role in
inhibiting the mechanism related to the mitochondrial function when cells are treated with 1,
in comparison with the cell treatment with the free L1.
Keywords: Ruthenium(II) organometallics, p-cymene, naproxen, naproxen-pyridineamide,
metallodrug.
2

1. Introduction
The investigation of chemical and biological properties of potential metallodrugs play
key role in the development of new antitumor agents aiming to broad strategies targeting
cancer therapy. The early success of platinum chemotherapeutics yet accompanied by severe
limitations in the clinical treatment of cancer led to an extensive search for non-platinum
drugs which pointed to the promising anticancer activity of diverse ruthenium compounds [1-
18].
An interesting approach to the development of new pharmaceuticals is the design of
drugs in which metal ions are combined with bioactive compounds already used in clinic. In
this context, the non-steroidal anti-inflammatory drugs (NSAIDs) that are widely prescribed
to treat pain, fever and inflammation, and also show antitumor properties, constitute a unique
class of pharmaceuticals to be explored [19-23]. In fact, NSAIDs have been successfully used
to build metal-based drugs which may show enhanced efficacy and/or reduced side-effects in
relation to the correspondent organic parent drugs [3,5,24,25]. Carboxylic NSAIDs have been
used for many years in de Oliveira Silva’s scientific research to develop new metallo-NSAIDs
[3,5,26-29]. In particular, a unique class of metallodrugs bearing metal-metal multiply bonded
Ru2(II,III) cores stabilized by four NSAID-derived carboxylate ligands was successfully
designed. The combination of the dimetal core and the NSAIDs in [Ru2(O2CR)4]+
paddlewheel type structured units succeeded in promoting synergistic effects leading to the
enhancement of in vitro and in vivo anticancer activity [30-36].
It is noteworthy that the growth in the field of ruthenium-based drugs has been driven
in part by the early promising results from phase trials conducted for the well-known
antimetastatic NAMI-A, (Him)[trans-RuCl4(dmso)im], im = imidazole [37,38], and the
structurally related cytotoxic KP1019, (Hind)[trans-[RuCl4(ind)2], ind = indazole, which was
further replaced by the more stable sodium salt NKP1339 in the clinical trials [39,40]. The
expansion of the ruthenium-based drugs is also clearly marked by the development of a
number of Ru(II) organometallics to which good activity against primary or metastatic
tumors, and lower cytotoxicity than platinum drugs, have been attributed [13,41,42].
Particular attention has been given to organometallics in which three of the coordination sites
are occupied by a η6-coordinated arene stabilizing the Ru(II) oxidation state. Two well-known
representatives of families of these compounds are RAPTA-C, ([Ru(6-p-cymene)(pta)Cl2],
pta
=
1,3,5-triaza-7-phosphatricyclo[3.3.1.1]decane, and RAED-C, ([Ru(η6-p-
cymene)(en)Cl][PF6]), en = ethylenediamine), both of them bearing the p-cymene (cym)
3

ligand [43,44]. More recently, NSAIDs and derivatives has attracted attention as ligands also
in the field of ruthenium organometallics aiming the study of anticancer properties [45,46].
In this context, we have investigated here the interaction of the Ru(II)-η6-p-cymene
framework with the naproxen-pyridinenamide (Npxpya, L1, or 5-N-4(Pyridyl)-2-(6-methoxy-
2-naphthyl)propionamide), a structurally modified form of the naproxen (HNpx or HL2) drug
carrying a moiety from the biologically active 4-aminopyridine [47-50]. Naproxen, besides
being currently used to treat rheumatic and arthritis diseases [51], exhibits activity against
cancer and viral diseases as well as some of its derivatives [52-58] and, moreover, it has been
reported as typical photoactive drug [59-62]. No studies reporting biological activity or
applications as a ligand in metal complexes were found for the modified drug in the form of
Npxpya, although this compound is claimed as a reagent for resolution of racemic mixtures of
optically derivatives of cyclopropane [63].
In this work, we report for the first time the synthesis and characterization of the new
organometallic [Ru(η6-p-cymene)(L1)Cl2] (1) and an alternative synthetic route for Npxpya
(L1). We have also re-prepared and performed additional investigation for the naproxenate-
derived [Ru(η6-p-cymene)(L2)Cl] (2) which was previously reported [64]. The compounds
are characterized by ESI-MS, NMR, ATR/FT-IR and UV/VIS, and their structures
corroborated by DFT computational calculations. Photophysical properties and biological
activity against NCI-H460 and A549 lung cancer cell lines are described.
2. Experimental
2.1 Reagents and solvents
Reagents from Sigma-Aldrich or Merck were used without further purification.
Solvents from Merck or LabSynth were dried by standard methods. Deuterated solvents
(CDCl3, d-methanol (d-MeOH), and d-dmso) were from Sigma-Aldrich. Naproxen (HNpx or
HL2) was purchased from manipulation pharmacy (Purifarma) in São Paulo, Brazil, and
1
characterized by H NMR (Appx. SM.1.1). The sodium naproxenate salt (NaL2) was
1
prepared from neutralization reaction between HL2 and NaOH, and characterized by H
NMR and ATR-FTIR (Appx. SM.1.2). The syntheses of the Ru(II)-arene compounds were
carried out under nitrogen atmosphere using standard Schlenk techniques. Precursor [Ru(η6-p-
cymene)Cl2]2 (3) (Figure 1) was synthesized by a method adapted from the literature [65], and
1
characterized by H NMR, ATR-FTIR, UV/VIS (Appx. SM.1.3). The dmso-derived
compound, [Ru(η6-p-cymene)dmsoCl2] (4) (Figure 1), was prepared by a method adapted
from the literature [66], and characterized by 1H NMR (Appx. SM.1.4).
4

2.2. Syntheses
2.2.1 Npxpya (L1)
The 4-aminopyridine (0.96 g; 0.01 mol) was added to a mixture of HL2 (2.34 g; 0.01
mol) and DCC (N,N'-dicyclohexylcarbodiimide, 2.66 g; 0.013 mol) that has been previously
stirred for 1-2 min in CH2Cl2 (60 mL). This solution was stirred overnight at room
temperature, and then it was filtered to remove small amounts of solid impurities (excess
DCC and residual DCU dicyclohexylurea). To purify the product from more of these
impurities, the filtrate was kept under refrigeration until more solid was formed, and then
submitted to filtration again. This purification procedure (refrigeration/filtration) was repeated
until no more solid was precipitated. The final filtrate was roto-evaporated and purified by
column chromatography (silica gel Merck, 2.6 x 30 cm, 0.063-200 mm; ethyl acetate as
eluent), accompanied by thin layer chromatography (TLC, silica gel) analysis. The purified
fraction collected from the chromatography column had the volume reduced to dryness by
roto-evaporation. Yield 91 %. Anal. calc. (%) for C19H18O2N2 · 1/2 H2O: C, 72.3; H, 6.07; N,
8.89. Found: C, 71.99; H, 5.92; N, 9.00. ESI-MS(+), CH2Cl2/MeOH (m/z): 307.14, [L1+H]+
1
requires 307.14; 329.13, [L1+Na]+ requires 329.13. H NMR, 500 MHz, CDCl3, δ (ppm vs
TMS), Appx. SM.2, Fig. S1a: 8.40 (dd, 3JH-H = 4.9, 4JH-H = 1.5 Hz, 2H, CHpy 1); 7.76 (d, 3JH-H
3
4
= 8.5 Hz, 1H, CHring 10); 7.72 (d, JH-H = 9.0 Hz, 1H, CHring 7); 7.70 (d, JH-H = 1.5 Hz, 1H,
CHring 6); 7.40 (dd, 3JH-H = 8.5 Hz, 4JH-H = 1.8 Hz 1H, CHring 11); 7.36 (dd, 3JH-H = 4.8, 4JH-H
=
1.6 Hz, 2H, CHpy 2); 7.18 (dd, 3JH-H = 8.9 Hz, 4JH-H =2.5 Hz 1H, CHring 8); 7.14 (d, 4JH-H = 2.5
Hz, 1H, CHring 9); 3.92 (s, 3H, O-CH3 12); 3.86 (q, 3JH-H = 7.1 Hz, 1H, CHchiral 4); 1.66 (d, 3JH-
H = 7.1 Hz, 3H, Cchiral-CH3 5). 13C NMR (500 MHz, CDCl3, δ (ppm), Appx. SM.2, Fig. S1b:
173.26 (C=O); 158.01 (C-OCH3); 150.48 (Npy-CH); 145.03 (HN-Cpy); 135.27 (Cchiral-Cring);
133.99 (Cring), 129.22 (CHring), 128.99 (Cring), 128.06 (CHring), 126.40 (CHring), 125.85
(CHring); 119.52 (CHring); 113.41 (CHpy); 105.70 (CHring); 55.37 (O-CH3); 48.20 (CHchiral);
18.48 (Cchiral-CH3). ATR-FTIR major bands, (cm-1): 3322w, 3251w, 3160w, 3066w (pya
NH); 3000sh, 2975w,2930w,2866w (Npx CH); 1702vs (pya C=O); 1589s (pya N-H
bending); 1506s (pya CNring + CCring and Npx CH3); 1416s (pya ring); 1336m,1323sh (pya
CNring); 1400-1370w (Npx CH3 rocking); 1284s (pya CNring); 1264s (Npx in plane CCring
deformation); 1227s,1209sh,1181s (pya CCring and Npx in plane CHring bending + CH3
rocking); 1181s (pya CCring); 1070w,1051w,1029w,997w (pya ring breathing and Npx
5

bands; 900-800w, 778m (pya out of plane CHring bending). UV/VIS, λmax (nm) / ɛ (mol-1Lcm-
1): 275(sh)/ n.d.; 285(sh)/ 5000; 318/ 1800; 333/2000 (in CH2Cl2).
2.2.2 [Ru(η6-p-cymene)(L1)Cl2] (1) (Figure 1)
Compound 3 (0.10 g; 0.16 mmol) and L1 (0.098 g; 0.32 mmol) were stirred in CH2Cl2
(15 mL) for 5 h, at room temperature. After removal of the solvent on a rotary evaporator, the
residue was dissolved in minimum amount of acetone and then precipitated with addition of
hexane. The product was washed with portions of hexane and dried in vacuum to give an
orange solid. Yield 71 %. Anal. calc. (%) for C29H32N2O2ClRu: C, 56.86; H, 5.26; N, 4.57;
Cl, 11.57. Found: C, 56.52; H, 5.25; N, 4.75; Cl, 10.65. ESI-MS(+), CHCl3/CH3CN (m/z):
577.12, [Ru(cym)(L1)Cl]+ requires 577.12; 541.14, [Ru(cym)(L1)-1H]+ requires 541.14;
1
270.98, [Ru(cym)Cl]+ requires 270.98. H NMR, 300 MHz, CDCl3, δ (ppm vs TMS), Appx.
SM.3, Fig. S2a: 7.93-7.83 (m, 3H and 2H, Npxpya CHring 6,7,10 and Npxpya CHpy 1; 7.61 (d,
3JH-H = 8.1 Hz, 1H, Npxpya CHring 11); 7.20 – 7.17 (m, 2H, Npxpya CHring 8,9); 6.96 (d, 3JH-H
= 6.3 Hz, 2H Npxpya CHpy 2); 5.36 (d, 3JH-H = 5.7 Hz, 1H, cym CHb); 5.01 (d, 3JH-H = 5.7 Hz,
1H, CHa); 4.10 (q, 3JH-H = 6.8 1H, Npxpya CHchiral 4); 3.92 (s, 3H, Npxpya O-CH3 12); 2.81
3
(m, 3JH-H = 6.8, 1H, cym CHc-(CH3)2); 1.77 (s, 3H, cym CH3 d); 1.57 (d, JH-H = 6.7 Hz, 3H,
3
Npxpya Cchiral-CH3 5); 1.24 (dd, JH-H = 6.8, 6H, cym CH(CH3e)2). 13C NMR, 500 MHz,
CDCl3, δ (ppm), Appx. SM.3, Fig. S2b: 173.91 (C=O); 157.64 (Npxpya C-OCH3); 154.13
(Npxpya Npy-CH); 146.42 (Npxpya HN-Cpy); 136.12 (Npxpya Cchiral-Cring); 133.81 (Npxpya
Cring); 129.43 (Npxpya CHring), 128.88 (Npxpya Cring); 127.28 (Npxpya CHring); 127.13
(Npxpya CHring); 126.42 (Npxpya CHring), 118.96 (Npxpya CHring), 115.01 (Npxpya CHpy);
105.72 (Npxpya CHring); 102.98 (cym Cring-(CH(CH3)2); 96.95 (cym Cring-CH3); 83.39 (cym
CH); 83.18 (cym CH); 81.61 (cym CH); 81.24 (cym CH); 55.35 (Npxpya O-CH3); 47.34
(Npxpya CHchiral); 30.60 (cym CH(CH3)2); 22.27 (cym CH(CH3)2); 22.23 (cym CH(CH3)2);
19.00 (Npxpya Cchiral-CH3); 18.04 (cym Cring-CH3). NMR HSQC/HMBC (Appx. SM.3, Fig.
S3, TS1), NOESY (Appx. SM.3, Fig. S4). ATR-FTIR major bands, (cm-1): 3313vw,
3245w, 3163vw, 3068w (pya NH); 2964mw, 2932mw, 2873vw, 2847vw (Npx CH);
1702m (pya C=O); 1632sh, 1603 sh, 1590s (pya N-H bending); 1502vs (pya CNring
+
CCring and Npx CH3); 1424m (pya ring); 1339sh, 1327w (pya CNring); 1400-1370w (Npx
CH3 rocking); 1294s (pya CNring); 1263s (Npx in plane CCring deformation); 1227w,1209m
(pya CCring and Npx in plane CHring bending + CH3 rocking); 1174sh,1181s (pya CCring);
1084w,1060w,1027m (pya ring breathing and Npx bands); 853sh, 834m-br, 806sh-m (pya out
of plane CHring bending). UV/VIS, λmax (nm) / ɛ (mol-1Lcm-1): 280 (sh)/ n.d.; 332/4000; 406/
800 (in CH2Cl2).
6

2.2.3 [Ru(η6-p-cymene)(L2)Cl] (2) (Figure 1)
Compound 3 (0.10 g, 0.16 mmol) and NaL2 (0.11 g, 0.42 mmol) were added to deaerated
methanol (30 mL) and stirred for 1 h, at room temperature. The solvent was removed on a
rotary evaporator and the solid was extracted with CH2Cl2. After filtration, the solution was
roto-evaporated and the residue was dissolved in small volume of acetone before addition of
hexane. The product was washed with portions of hexane and dried in vacuum. Yield 87 %.
Anal. calc. (%) for C24H27ClO3Ru ∙ H2O: C, 55.65; H, 5.64; Cl, 6.84. Found: C, 55.63; H,
5.04; Cl, 6.89. ESI-MS(+), CHCl3/CH3CN (m/z): 465.10, [Ru(cym)(L2)]+ requires 465.10;
1
312.01, [Ru(cym) (CH3CN)Cl]+, requires 312.01; 270.98, [Ru(cym)Cl]+ requires 270.98. H
NMR, 500 MHz, CDCl3, δ (ppm vs TMS), Appx. SM.4, Fig. S5a: 7.70-7.65 (m, 3H, Npx
CHring 3,4,7); 7.40 (dd, 3JH-H = 8.5, 4JH-H = 1.6 Hz, 1H, Npx CHring 8); 7.14–7.10 (m, 2H, Npx
CHring 5,6); 5.54 (d, 3JH-H = 5.8 Hz, 2H, cym CHb); 5.32 (d, 3JH-H = 5.8 Hz, 2H, cym CHa);
3.91 (s, 3H, Npx O-CH3 9); 3.52 (q, 3JH-H = 7.1 Hz, 1H, Npx CHchiral 1); 2.87 (m, 3JH-H = 7.0,
3
4JH-H = 3.0 Hz, 1H, cym CHc-(CH3)2); 2.29 (s, 3H, cym CH3 d); 1.43 (d, JH-H = 7.2 Hz, 3H,
3
Npx Cchiral-CH3 2); 1.27 (m, JH-H = 6.9 Hz, 6H, cym CH(CH3e)2). 13C NMR, 500 MHz,
CDCl3, δ (ppm vs TMS), Appx. SM.4, Fig. S5b: 193.60 (Npx COO); 157.50 (Npx C-OCH3);
135.33 (Npx Cchiral-Cring); 133.63 (Npx Cring); 129.31 (Npx CHring); 128.97 (Npx Cring); 126.92
(Npx CHring); 126.73 (Npx CHring); 126.24 (Npx CHring), 118.69 (Npx CHring); 105.59 (Npx
CHring); 100.73 (cym Cring-(CH(CH3)2); 94.33 (cym Cring-CH3); 78.13 (cym CH); 77.69 (cym
CH); 55.30 (Npx O-CH3); 48.11 (Npx CHchiral); 31.48 (cym CH(CH3)2); 22.27 (cym
CH(CH3)2); 18.82 (cym Cring-CH3); 18.09 (Npx Cchiral-CH3). NMR HSQC/HMBC (Appx.
SM.4, Fig. S6, TS2). ATR-FTIR major bands, (cm-1): 1630w, 1605m (Npx CCring + in
plane CHring bending); 1505m, 1501sh (Npx and cym CH3, cym CCring), 1488-1450w (Npx
and cym CCring); 1460s (Npx aCOO); 1437s (Npx sCOO); 1387-1373w (Npx and cym
CH3 rocking); 1267s (Npx in plane CCring deformation + in plane CHring deformation + in
plane CH bending); 1229s (Npx C-O, in plane CCring + CHring deformation + in plane CH
bending ); 1195m, 1179m, 1160m (Npx C-O + C-C + in plane CH bending + CH3); 1028s
(Npx C-O + CH3 rocking and cym CH3 rocking + CH bending); 898m (cym out of plane
CHring bending). UV/VIS, λmax (nm) / ɛ (mol-1Lcm-1): 270(sh)/ n.d.; 317 / 2200; 332 / 2000;
423 / 800 (in CH2Cl2).
7

2.3 General Instrumentation and Analysis
Elemental analyses (C,H,N, Perkin-Elmer 2400 Elemental Analyzer; Cl, volumetric analysis),
ESI-MS (Bruker Daltonics Micro TOF equipment, capillary, 4.5 kV, nebulizer 0.4 bar, dry
gas 4.0 L min−1, 200 °C, samples dissolved in chloroform/further diluted in acetonitrile), and
1H NMR, 13C NMR, HSQC (Heteronuclear Multiple Quantum Correlation), HMBC
(Heteronuclear Multiple Bond Correlation) experiments (INOVA 300 MHz or Bruker AIII
500 MHz spectrometer, at probe temperature) were performed at the Analytical Center of the
Institute of Chemistry, University of São Paulo. The electronic absorption spectra (compound
solution in 1.0 cm quartz cuvettes) were recorded on a Shimadzu UV-1650 PC and the
ATR/FT-IR spectra (solid samples) in a Bruker/alpha FT-IR spectrophotometer.
2.4 Fluorescence studies
The fluorescence measurements were performed in a PC1 photon-counting spectrofluorimeter
(ISS) with a photomultiplier-based photon counting detector, using all-polished-side 1.0 cm
quartz cuvettes. The fluorescence emission and excitation spectra of compounds (2.0 x 10-6
mol L-1 in CH2Cl2) were registered at room temperature (about 25oC). The excitation
wavelength was set at 275 nm. Relative fluorescence quantum yields (Φf) were determined by
the equation:
2
Fu As nu
 f   fs
2
Fs Au
ns
where Φf is the fluorescence quantum yield, A is the absorbance at the excitation wavelength,
F is the area under the corrected emission curve, and n is the refractive index of the solvent.
Subscripts s and x refer to the standard and to the unknown, respectively [67]. Both sample
and standard were excited at the same wavelength and slit condition. The absorbance of
solutions at the excitation wavelength ranged from 0.01 to 0.03 to avoid the inner filter effect
[67]. HNpx was used a reference standard at 2.0 x 10-6 mol L-1 in CH3OH solution (ΦF =
0.53) [68].
2.5 Computational methodology
The electronic calculations were performed using the GAUSSIAN 09 program [69].
Stationary points on the potential energy surface were fully optimized, followed by
evaluations of the harmonic vibration frequencies to characterize their natural minima. The
absence of imaginary frequencies indicated that all optimized structures were true minima.
The PBE1PBE function [70,71] was used in the optimization procedure employing the
8

LANL2TZ basis set for the ruthenium atom [72] and 6-31G(d,p) ones for the other elements
[73,74]. Nevertheless, additional DFT methods were also applied for particular purposes
along this investigation as M05 [75], M05-2X [76], M06L [77], M06 [78], M06-2X [78],
BMK [79], B3LYP [80,81], B97D [82], BP86 [83] and wB97XD [84]. The solvation effect
was examined by the Polarizable Continuum Model (PCM) with the integral equation
formalism variant (IEF-PCM) [85]. In the calculation of the bond order indice, Mayer [86],
Wiberg [87] and NLMO/NPA [88] methods were applied. The Natural Population Analysis
(NPA) [89] was used for the condensed dual descriptor. Natural bond orbital (NBO) analysis
was carried out using the NBO 6.0 program [90]. Time-dependent Density Functional Theory
[91-93] making use of the PBE1PBE functional was applied to simulate UV/VIS spectra.
2.6 Biological Assays
2.6.1 Cell culture
NCI-H460 and A549 cell lines were obtained from American Type Culture Collection (VA,
USA) and Rio de Janeiro Cell Bank (RJ, BRA), respectively. Both cell lines were cultured in
RPMI-1640 medium supplemented with 10% (v/v) fetal bovine serum (BSA), penicillin (100
units mL-1), streptomycin (100 μg mL-1) and 0.5% (w/v) amphotericin B. All cells were
cultured at 37 °C and 5% CO2 and cultures were amplified and cryopreserved in 10% dmso in
BSA at - 80 °C.
2.6.2 Preparation of compound solutions
Stock solutions of compounds were prepared at 20 mmol L-1 in dmso and stored at -20 °C.
Then, for each assay, these solutions were diluted in the culture medium having 10% FBS and
1% antibiotic/antimycotic to give final concentrations varying from 6.5 to 400 μmol L-1.
2.6.3 Cytotoxicity assay
Cells were plated at the density of 104 cells/well in 96-well plates and, after adhesion they
were treated separately with each compound at concentrations varying from 6.5 to 400 μmol
L-1 for 24 h. Then, 10 μL of MTT ([3(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide], Sigma-Aldrich, MO, USA) was added to each well at 5 mg mL-1 and the plates
were incubated for 3 h. After centrifugation at 240 g, for 10 min, the medium was removed
and the formed crystals were solubilized in 100 μL dmso. Cell viability was calculated from
the optical densities at 538 nm in the VERSAmax Tunable Microplate Reader, (Molecular
Devices, CA, USA). The absorbance values from the control wells (exposed only to the
medium with the vehicle) were used to calculate the total cell viability (100%). The
9

cytotoxicity is expressed by the IC50 (i.e., 50% inhibitory concentration of cell proliferation)
values estimated from the dose-response curve of each compound.
2.6.4 Evaluation of mitochondrial membrane potential
NCI-H460 cells (2.5 x 105 cells / well) were treated individually with L1 (75 and 150 μmol L-
1), compound 1 (75 and 150 μmol L-1) and compound 3 (37.5 and 75 μmol L-1) for 24 h to
evaluate mitochondrial membrane potential (MMP, Δψm). CCCP (m-chlorophenylhydrazone
carbonylcyanide, Sigma-Aldrich, MO, USA), a potent decoupler of mitochondrial oxidative
phosphorylation, was used at 100 μmol L-1 as a positive control of mitochondrial
depolarization. After treatment, cells were harvested, washed with PBS/0.5% BSA 0.5% azide
and incubated for 15 min at 37°C with 50 nmol L-1 TMRE (ethyl ester tetramethylrodamine,
Molecular Probes, OR, USA). The cells were washed again, re-suspended in 300 μL of
PBS/BSA solution and analyzed on the FACSCalibur flow cytometer (Becton Dickinson, CA,
USA). A total of 20,000 events were collected per sample and analyzed using the FlowJo
software version 0.7 (Tree Star Inc., OR, USA).
2.6.5 Cell cycle analysis
NCI-H460 cells (2.5 x 105 cells / well) previously synchronized by serum starvation were
treated for 24 h with compound 1 or L1 (at 75 and 150 μmol L-1), removed from the plate and
washed twice with PBS / BSA 0.5% azide 0.2%. Cells were centrifuged for 10 min at 290 g,
fixed and permeabilized with 70% alcohol, and after that they were kept overnight in a
freezer. Finally, the cells were labeled with 0.1 mg mL-1 propidium iodide solution and 0.25
mg mL-1 RNAse. Cell fluorescence was measured on a FACSCalibur flow cytometer (Becton
Dickinson, CA, USA) with CellQuest software and 10,000 events were acquired for each
sample. Further analyzes were performed by using the FlowJo software version 0.7 (Tree Star
Inc., OR, USA).
2.6.6 Western blotting
NCI-H460 cells treated for 24 h with compound 1 or L1 (at 75 and 150 μmol L-1) were lysed
with mammalian protein extraction reagent (Mammalian Protein Extraction Reagent®,
Thermo Scientific Pierce, IL, USA) containing 10% protease inhibitor and phosphatase
inhibitor. Each lysate protein concentration was determined by Bradford assay. According to
the obtained concentration of proteins, cell lysates were diluted in Standard Western blotting
buffer (50 mmol L-1 Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, 5% 2-β-mercaptoethanol,
0.002% blue bromophenol) and denatured by heating. Thereafter, cell lysates were
fractionated by electrophoresis (SDS/PAGE, Biorad, CA, USA) and transferred to
10

polyvinylidene fluoride membranes. Then, the membranes were washed in Tris-saline buffer
(TTBS: 100 mmol L-1 Tris-HCl, 137 mmol L-1 NaCl and 0.05% Tween-20, pH 7.8) and
blocked for 1 h in 5% diluted in TTBS. After washing with TTBS, the membrane was
incubated overnight with the primary antibodies anti-capase 8, capase 3, caspase 9,
cytochrome c, β-actin, BAX, BAD, Bcl-Xl, Bcl-2, CDK4, CDK6, cyclin A, cyclin D1, cyclin
D3, cyclin E2, NF-Ƙb, p-p53, p38, p-cdc2 at 8 °C. The membrane was washed again with
TTBS and then incubated for 1 h with secondary antibody anti-mouse or rabbit conjugated to
horseradish peroxidase. Detection was performed with the Pierce® Western Blotting
Substrate Plus kit (Thermo Scientific Pierce, IL, USA).
c
e
b
10
6
9
7
1
2
11
4
8
a
(1)
5
3
d
c
e
7
6
1
8
b
5
9
4
3
a
d
(2)
2
c
e
b
a
d
(3)
(4)
Figure 1. Structures of Ru(II)-arene compounds 1, 2, 3 and 4 (showing labels of hydrogens
used in the 1H NMR spectra assignment).
3 Results and discussion
11

3.1 Synthesis and characterization
3.1.1 [Ru(η6-p-cymene)(L2)Cl]
The synthesis of the organometallic 2 has been improved by using only methanol,
instead of the CH2Cl2/MeOH mixture, to achieve high yield (85 %) in much shorter time (1 h)
than the overnight period needed before [64]. Single crystal structure data for this compound
(not shown) were in agreement with the reported results [64], thus supporting the typical
piano-stool geometry of the Ru(II)-arene bearing one chloride and bidentate naproxenate.
However, the use of other techniques is crucial for reliable characterization since the crystal
structure of one single crystal is not guarantee for the purity of any bulk solid. Regarding this
part, disagreement in values and/or assignments, in relation to the published data, were found.
The ESI-MS(+) peaks having the isotopic pattern of Ru are found here at m/z: 465.10,
assigned to [Ru(cym)(L2)]+, as previously [64]; 270.98, assigned to a fragment lacking the
Npx, [Ru(cym)Cl]+, not reported before; and 312.01, ascribed to [Ru(cym)(CH3CN)Cl]+,
which seems more reasonable than the reported [Ru(cym)Cl(H2O)Na] fragment [64], since if
this late one is 1+ charged, it would have slightly distinct m/z (311.75) and it would be also
incoherent with the Ru(II) oxidation state of the organometallic 2.
The NMR analysis also shows discrepancies when compared with the reported data.
1
The assignment of the H NMR spectrum of 2 (CDCl3, Appx. SM.4, Fig. S5a) is supported
13
here by the CNMR (Appx. SM.4, Fig. S5b) and HSQC, HMBC spectra (Appx. SM.4, Fig.
S6, TS2), and it is also corroborated by the comparison with spectra of 3 (CDCl3), HL2
(CDCl3 or d-MeOH) and NaL2 (d-MeOH). Typical signals of the p-cymene ring hydrogens in
2 are significantly shifted upon coordination of L2. The shift towards downfield of the cym
ring peaks (CHb, 5.54; CHa, 5.32 ppm) in relation to precursor 3 (5.48; 5.35 ppm,
respectively) support the environment changes around the metal. The coordination of L2 is
corroborated by the chemical shift of the signals of the hydrogens nearly adjacent to the -COO
group. Both the upfield shift of the signal of the hydrogen directly bonded to the drug chiral
carbon (Npx CHchiral 1) in 2 (3.52 ppm) in relation to HL2 (3.86, CDCl3; 3.82 ppm, d-MeOH),
and the lower value of  when compared with the peak shifting of drug salt (NaL2, 3.71
ppm), for which electrostatic interactions exist, support the assignment. Taking into account
this interpretation together with the expected deprotonation of naproxen upon coordination in
2, and also the expected hydrogen shielding upon drug coordination to Ru(II), the previous
assignment of the peak at 3.84 ppm [s, 4H] to Npx CHchiral hydrogen/CH3 hydrogens of
CH3(CH)COOH is questionable [64]. The hydrogen signal of Npx O-CH3 9 is assigned here
12

to the singlet at 3.92 ppm in 2, based on the lack of significant shift when compared to the
parent drug (HL2: 3.91, CDCl3; 3.88, d-MeOH; NaL2: 3.87 ppm). The signals of the methyl
group hydrogens bonded to the chiral carbon (Npx Cchiral-CH3 2) in 2 (1.43 ppm) are assigned
to a doublet (3H) shifted to higher field in relation to free drug (HL2: 1.58, CDCl3; 1.50, d-
MeOH; NaL2 1.49 ppm).
The major ATR-IR bands of 2 (experimental part) are tentatively assigned here by
comparison with the spectra of 3 (Appx. SM.1.2), p-cymene (not shown), HL2 (not shown)
and NaL2 (Appx. SM.1.3). The (C=O) typical vibrational bands of the -COOH group [94]
found at 1728, 1685 cm-1 in HL2 must disappear after deprotonation of the acidic drug to give
the -COO- naproxenate anion (L2). The aCOO asymmetric and sCOO symmetric stretching
modes of ionic L2 appear at 1546 and 1360 cm-1, respectively, in the spectrum of the NaL2
salt, which also shows spectral changes at the region ascribable to C-O stretching of -COO
(1200-1150 cm-1) when compared to the carboxylic acid spectrum. Spectrum of 2 shows
typical bands of both p-cymene and naproxenate ligands. The major spectral changes are
found at the frequency ranges: 1510-1400 and 1270-1000 cm-1. The appearance of new bands
in spectrum of 2, when compared to HL2 and NaL2, suggests the assignment of the aCOO
and sCOO vibrations of the organometallic at 1460 and 1437 cm-1, respectively. Although
the band at 1505 cm-1 in 2 might be not totally ruled out for aCOO (considering the usually
reported a frequencies of other Ru-carboxylates [95-98]), it is worthy to note that the joint
analysis of all spectra here suggests high contribution of vibrational modes of both Npx and p-
cymene ligands (CH3 of both Npx and cym, and CCring of both Npx and cym) rather than
only a new individual aCOO band. The free HL2 shows band at 1505 cm-1 while the
precursor 3 (lacking carboxylate as ligand) shows aromatic ring vibration around 1500 cm-1
(due to lower frequency shift in relation to the free p-cymene (1516 cm-1)). The (COO)
difference from 1460 and 1437 cm-1 supports the bidentate mode of L2 in 2 when compared
to the ionic carboxylate in NaL2. The previous assignment [64] of aCOO at 1502-1594 cm-1
and sCOO at 1379-1385 cm-1 frequency ranges for 2 and other Ru(II)-p-cymene-NSAID
compounds seems to be not characteristic of bidentate carboxylate. The bands at 1270-1000
cm-1 in spectrum of 2 are assigned here to C-O modes of carboxylate. Furthermore, the
spectral changes in spectrum of 2 (1229s, 1195m, 1179m, 1160m) in relation to NaL2 (1213s,
1163m) support the distinct modes of interaction of the -COO group with the metal atoms
(bidentate coordination (Ru) and ionic (Na)). Finally, the band at 898 cm-1, which shifts to
higher frequency in relation to those of 3 (874) and free p-cymene (816 cm-1), is assigned to
cym out of plane CHring bending in 2.
13

3.1.2 Npxpya
The naproxen-pyridineamide (L1) has been prepared here in high purity (see
13
characterization bellow with additional ESI-MS, CNMR data not previously reported), and
also in high yield (91%), by reacting HL2 with 4-aminopyridine in the presence of DCC, in
CH2Cl2 at room temperature, without need of any special gas atmosphere. The present
synthetic procedure is advantageous when compared with the reported method [63] which is
based on the coupling reaction with thionyl chloride (moisture-sensitive chemical) in benzene
(toxic) solvent, conducted under argon atmosphere, and gave only 50% yield (S)-Npxpya. The
ESI-MS(+) supports the formation of the modified drug according to the peaks at m/z: 307.14
and 329.13 ascribable to [L1+H]+ and [L1+Na]+ (Na+ ion coming from the matrix)
1
respectively. The assignment of the HNMR spectrum of L1 (experimental section, Appx.
SM.2, Fig. S1a) is supported by the 13CNMR spectrum (Appx. SM.2, Fig. S1b). The peaks of
the pyridine ring hydrogens (CHpy) are at 8.40 - 7.36 ppm, while the peaks of the naphtalene
ring hydrogens (Npx CHring) appear at 7.8 - 7.1 ppm. A peak at 3.93 ppm is assigned to the
methyl hydrogens belonging to the O-CH3 group (O-CH3 12). The signal of the hydrogens
from the methyl groups bonded to the chiral carbon (C-CH3(chiral) 5) shifts to lower field (1.66
ppm) in relation to HL2 (1.58 ppm) in the same solvent. The ATR-FT spectrum (Appx. SM.5,
Fig. S7) shows the typical bands of both Npx and pyridine ring, in addition to new bands
assignable to the amide group: 1702 cm-1 (pya C=O), 1589 cm-1 (pya N-H bending) and
1506 cm-1 (pya ring and Npx CH3).
3.1.3 [Ru(p-cymene)(L1)Cl2]
Compound 1 was obtained in good yield from the reaction of L1 with precursor 3, in
CH2Cl2. The ESI-MS(+) peaks having isotopic pattern of Ru are at m/z: 577.12, assigned to a
fragment lacking one chloride ligand, [Ru(cym)(L1)Cl]+; 541.14, associated to [Ru(cym)(L1)
- 1H]+; and 270.98, ascribed to a fragment lacking the L1 ligand, [Ru(cym)Cl]+. The typical
piano stool geometry of the new compound is corroborated by spectroscopic characterization.
1
13
The H NMR spectral analysis of 1 (Appx. SM.3, Fig. S2a) is supported by the C NMR
(Appx. SM.3, Fig. S2b), HSQC, HMBC (Appx. SM.3, Fig. S3, TS1) and NOESY spectra
(Appx. SM.3, Fig. S4), and by comparison with the signals of precursor 3 (Appx. SM.3) and
compound 2 (Appx. SM.4, Fig. 4). The spectral profile of 1 confirms the changes on the metal
coordination sphere, in relation to precursor 3, which are accompanied by symmetry reduction
14

of the p-cymene associated to the presence of the Npxpya drug ligand. Shifting towards
higher field upon coordination of L1 is observed for p-cymene CH ring hydrogens (four
doublets: CHa, 5.01, 4.95; CHb, 5.36, 5.29 ppm, compared to single doublets: CHa, 5.35;
CHb, 5.48 ppm in 3) and methyl group (cym CH3 d: 1.77 in 1; 2.16 ppm in 3). The
coordination of L1 to Ru through the nitrogen atom in 1 is supported by the upfield chemical
shift of the pyridine-ring hydrogens adjacent to the nitrogen (CHpy 1 at 7.93-7.83 (m) coupled
to Npxpya CHring 6,7,10 ; CHpy 2 at 6.96 (d), ppm), in relation to the free L1 (8.40 (dd); 7.36
(dd), ppm). The signal of the isopropyl group hydrogens also shifts to higher field (in ppm:
CH(CH3)2(c) 2.81 (m), 1H; CH(CH3)2(e) 1.24 (m), 6H) in comparison with the signals of 3
(2.92 (m), 1H; 1.28 (d), 6H, respectively). The 13C NMR spectrum of 1 shows chemical shifts
of carbons belonging to the pyridine ring towards downfield upon coordination of L1: Npy-CH
(from 150.48 in L1 to 154.13 ppm in 1), HN-Cpy (145.03 to 146.42) and CHpy (113.41 to
115.01 ppm) which corroborate the coordination of the N-pyridine to ruthenium.
The presence of the Npxpya in 1 is corroborated by the ATR-IR spectrum (Appx.
SM.5, Fig. S7). Compound 1 shows the three typical major bands assignable to L1 ((cm-1):
1702, pya C=O; 1590s, pya N-H bending; 1502vs, pya CNring + CCring and Npx CH3);
1424m, pya ring). The band of pyridine ring breath (1417 cm-1 in L1) slightly shifts to higher
frequency (1423 cm-1 in 1), thus confirming the coordination of L1 to Ru through the N-
pyridine atom. Other relevant spectral changes occur at the regions 1230-1100 cm-1 (CCring),
1100-1000 cm-1 (ring breathing) and 900-700 cm-1 (out of plane CH py ring). The bands of p-
cymene were difficult to assign due to overlap with bands of L1 at 1500-1300 cm-1, although
spectral changes could be observed in relation to precursor 3.
3.2 Computational analysis of structures
Computational structural analysis was performed for both Ru(II)-arenes and Npxpya
(Appx. SM.6). The DFT analysis of 2 by PBE1PBE is in good agreement with the crystal
structure reported data [64] (Appx. SM. 6, Fig. S8; SM.6.1, TS3). Good structural description
was also obtained by B3LYP, B97D, BP86, M05, M06, M06L, wB97XD methods, although
these outperformed Ru-Cl and Ru-O bond lengths compared to PBE1PBE (Appx. SM.6.1,
TS3, TS4). DFT bond order data suggests a Ru metal center dominated by -bonds with
NLMO/NPA bond indices: Ru-C, 0.45-0.47; Ru-Cl, 0.52; Ru-O, 0.19-0.25 (Appx. SM.6.1,
TS5). The C-O bond shows partial double character (between  and ) corroborating the
bidentate coordination of the L2 drug ligand. Compound 2 shows C-O NPA charge separation
higher than MeOH and slightly lower than CO2 molecule (Appx. SM.6.1, TS6). The DFT
15

structural analysis of the new organometallic 1, due to the lack of crystallographic data, was
based on the DFT most stable calculated structure for L1 (predominant conformer at 52%, see
discussion Appx. SM.6.2, Fig. S9; TS7, TS8), and has been performed by using different DFT
methods (Appx. SM.6.3, TS9). The relative Gibbs free energy of six possible structures
predicted for 1 by PBE1PBE in CH2Cl2 IEF-PCM solvent pointed to the most stable structure
at 63% contribution (Appx. SM.6.3, Fig. S10). Taking into account all the methods used here
(Appx. SM.6.1, TS3; SM.6.3, TS9; SM.6.4, TS11) for both Ru(II)-arenes, the bond lengths
are at the ranges 2.400-2.445 Å (Ru-Cl) and 2.135-2.217 Å (Ru-N). The Ru-C bond distances
are slightly longer in 1 (2.154-2.300 Å) than in 2 (2.138-2.259 Å). The coordination sphere of
1, represented by bonds, shows low bond order reflecting small covalent character for the
RuN bond (Appx. SM.6.3, TS10) [99]. The molecular electrostatic potential maps (Appx.
SM.6.4, Fig.S11) suggest similar distribution of electron density with the negative regions
mainly located on the naproxen moiety for both Ru(II)-arenes, while extensive positive
regions are predicted along the coordination sphere, and particularly extended to the pyridine
ring in the case of 1.
3.3 Chemical behavior of the Ru(II)-arenes in dmso
Studies about the stability of compounds bearing the moiety [Ru(η6-p-cymene)(L)]
with different types of ligands, as for example, monodentate or bidentate ligands having N-
[100], P- [101, 102] or O- [103-105] donor atoms, indicate dissociation of the L-ligand,
and/or also of the arene, when the organometallic is dissolved in dmso or water (solvents
commonly used to prepare solutions of drugs in biological assays). Taking into account that
both of the Ru(II)-arenes investigated here were dissolved in dmso for the biological
1
experiments, we have monitored their chemical behavior by following the H NMR spectral
changes in d-dmso as a function of the time compared to the spectra of the precursors and
parent ligands (Appx. SM.7).
The 1H NMR spectra of both 1 and 2 compounds show significant changes in relation
to their spectra in CDCl3, suggesting that, differently from the behavior in the non-
coordinating chloroform, they are not stable in dmso. In the case of 2, the spectral changes
(Appx. SM.7, Fig.S12) suggest transformations that start shortly after the dissolution, in
disagreement with the previous report [64] which claims that this compound is stable up to 12
h. The peaks of Npx CHring (CHring 3,4,7 at 7.9-7.6; CHring 8 at 7.5-7.3; CHring 5,6 at 7.3-7.0
ppm) which appear nearest to the signals of NaL2 at the beginning, decrease in intensity in
the subsequent times, while new peaks closest to the signals of HL2 arise. Similar behaviour
16

occurs at the region of Npx Cchiral-CH3 9, what may suggest possible transformation of species
having bidentate- to species bearing monodentate- Npx drug ligand. Subsequent and gradual
spectral changes over the time indicate predominant dissociation of the drug ligand
accompanied by the formation of the substituted Ru(cym)dmso species 4 (5.82 (d), CHb;
5.77, CHa; 2.82 (m), CHc-(CH3)2, 1.16 (d) ppm, CH(CH3e)2). The findings indicate that 2
readily undergoes chemical transformations, apparently involving changes on the
coordination mode of L2, that are followed by the dissociation of the drug ligand. The process
may involve equilibrium among distinct species that seems to be not finished after 48 h (the
maximum time investigated here). The literature proposal for protonation of L2 to HL2,
based on a peak at 9.7 ppm [64] might be unlike to happen since the typical peak of free HL2
in d-dmso is observed here at 12.3 ppm. We have also observed low intensity peaks at 9.7 and
9.8 ppm after 24 h, but although these appear are located at the downfield typical region of
hydrogen-bonded species, they could not be undoubtable assigned.
In contrast to 2, the new organometallic 1 was found to lose its integrity by prompt
1
total dissociation of the L1 ligand. The H NMR spectrum (Appx. SM.7, Fig.S13) registered
shortly after dissolution of 1 in d-dmso gives evidence for the lack of the original
organometallic while it confirms the presence of the substituted Ru(cym)dmso species 4
(CHa, 5.82 (d); CHb, 5.78 (d); CH3d, 2.08 ppm) and the L1 free drug [peaks of the pyridine-
ring hydrogens adjacent to the N atom (in CDCl3, at 7.93-7.83 (m), CHpy 1, coupled to CHring;
and at 6.96 (d) ppm, CHpy 2) shift towards downfield reaching resemblance to the free L1 (in
dmso, 8.40 (d); 7.82-7.12 ppm (m); amide group N-H hydrogen, around 10.5 ppm].
It is worthy to highlight that the time-dependent spectra of the Ru(II)-arenes were
compared with the spectrum of the free p-cymene (in d-dmso, peaks at: 7.12-7.05; 2.89-2.75;
2.25; 1.18,1.16 ppm) to check for possible dissociation of the p-cymene ring, since this has
been recently reported for other Ru-(p-cymene) bearing NSAID-carboxylates [105]). In the
case of compound 2, bearing the bidentate naproxenate, no evidence is found for appreciable
dissociation of the p-cymene ring. Furthermore, in the case of 1, bearing the N-coordinated
Npxpya, the spectrum clearly indicates the presence of only two species, i.e., Npxpya (L1)
and Ru(cym)dmso species (4), shortly after its dissolution in dmso.
3.4 Electronic absorption spectra
The electronic absorption properties of both Ru(II)-arenes are described here for the
first time. Experimental max, DFT PBE1PBE computed wavelengths and the assignment of
electronic transitions based on electron density of the frontier molecular orbitals (Appx. SM.8,
17

Fig. S14, TS12) are presented in Table 1. Compound 2 shows intense bands at 270 (sh), 317
and 332 nm, and a relatively less intense band centered at max 423 nm. Similar UV bands are
found for HL2 (λ (nm) /ɛ (mol-1Lcm-1): 230/n.d.; 260/4400; 270/4500; 317/1300; 330/1500, in
methanol) and an intense band at 270 nm is also observed for precursor 3 (Appx. SM.1.3).
The DFT PBE1PBE calculations suggest contributions of metal-to-ligand charge transfer
(MLCT) and intraligand (LL) transitions at 260-290 nm. The bands at 317, 332 nm are
predicted mainly as LL transitions, although a band at 340 nm is also observed for 3. The
visible band (423 nm) is assigned predominantly as MLCT transition. The shift of this band
when precursor 3 is dissolved in different solvents (450 nm in CH2Cl2; 418 nm in CH3CN;
400 nm in dmso, spectra not shown) may corroborate the contribution of the Ru-p-cymene
framework as the “metal part” in the MLCT transition. A visible band at the same region has
been assigned to metal-to-ligand transition for other ruthenium organometallics [106-108].
Compound 1 exhibits electronic spectrum dominated by high intensity UV shoulders which
extends up to the beginning of the visible region, what makes the interpretation of the
electronic absorptions more complicated. Intense shoulders at 275-285 nm and a relatively
less intense shoulder at 332 nm, in addition to a visible band around 406 nm, are observed.
The UV absorptions are similar to those found for L1 (λ (nm) /ɛ (mol-1L cm-1): 275(sh)/n.d.;
285(sh)/5000; 318/1800; 333/2000, in CH2Cl2). However, computational analysis predicts
contributions of both MLCT and LL transitions for the UV bands of 1. The band at 406 nm is
overlapped with part of the UV absorptions, what might explain the participation of LL
transitions also at the visible. Despite the band overlap, it is interesting to note that the visible
band of 1 shifts to higher energy (lower , 406 nm) when compared with 2 (423 nm) and 3
(450 nm) in the non-coordinating dichloromethane.
Regarding the analysis of electron density of the frontier molecular orbitals (Appx.
SM.8, Fig. S14, TS12), it is interesting to mention that the energy gap between the highest
occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) is
similar for both Ru(II)-arenes (~4.2 eV) suggesting similar kinetic stability [109]. While the
π* LUMO orbitals are distributed in the naproxen moiety in 2, the π* LUMO orbitals of 1 are
predominantly distributed in the pyridine ring having small participation in the amide group.
The data show that the Ru d-orbitals contribute mainly in the H-2 and L+4 molecular orbitals,
in good accordance with other studies [110,111]. The Ru t2g orbitals contribute mainly in H-1
and H-2, whereas the antibonding orbital shows minor contribution in L+1 and L+4
molecular orbitals of 2 and 1, respectively. Notwithstanding an antibonding combination from
oxygen lone pair electrons of the carboxylate moiety is seen in the L+1 molecular orbital of 2,
18

the combination comes delocalized from the antibonding naproxen  part and from oxygen
lone pair electrons in the ligand in the case of the L+4 molecular orbital of compound 1
Table 1. Major bands (λmax) at the experimental electronic absorption spectra of Ru(II)-arenes
in dichloromethane and computed DFT PBE1PBE calculated wavelenghts (λcalc).
Compound
1
λmax (nm) / ɛ (mol-1Lcm-1)*
λcalc (nm)
Transition**
280
291.3
302.0
317.3
351.2
407.5
418.4
482.5
265.5
281.5
318.5
328.2
449.7
LL
MLCT
LL
MLCT
LL
332 / 4000
406 / 800
MLCT / LL
LL
MLCT / LL
MLCT
LL
270
2
317 / 2200
332 / 2000
423 / 800
LL
MLCT
*
The values of molar absorptivity were estimated directly from the absorbance in the spectrum
without deconvolution. ** Assignment: LL = intraligand electronic transition; MLCT = metal to ligand
charge transfer electronic transition.
3.5 Fluorescence studies
The photophysical properties of the Ru(II)-arene compounds are reported here also for
the first time. Both organometallics 1 and 2 show fluorescence spectra similar to those of their
correspondent parent ligands. The emission band assignable to the naphthalene fluorophore
[60] is found at 350 nm (Appx. SM.9, Fig. S15). The large Stokes shift values (Table 2) might
be related to significant structural changes on the ground and the excited states upon photo-
excitation. The lack of self-absorption which may facilitate the distinction between emission
and excitation suggests attractive photophysical property [112]. Similar behavior is observed
in Re(I) [112] and other Ru(II) [113] compounds. L1 shows emission quantum yield (Φf =
0.25, CH2Cl2) lower than those found for the HL2 drug (0.40, aqueous buffer [60]; 0.47,
CH3CN [114]; 0.53, CH3OH [67]). The decrease in the Φf of the Ru(II)-arenes (0.04 (1); 0.22
(2), Table 2), in relation to those of their corresponding parent ligands, might be explained by
the metal coordination to the fluorophore ligand inducing fluorescence quenching
[115,116,117]. Moreover, the pyridineamide moiety in 1 may act as a bridge in the metal-
naproxen donor-acceptor conjugate system promoting additional fluorescence quenching, in
consistency with other reported data [118-120].
19

Table 2. Photophysical data of Ru(II)-arene compounds
Compound
λmax (nm)
Excitation Emission
Stokes*
Φf
L1 (in CH2Cl2)
1 (in CH2Cl2)
260
266
356
356
96
90
0.25
0.04
L2 (in MeOH)
2 (in CH2Cl2)
230, 260 sh, 270
236, 274
354
356
84
82
0.53 [67]
0.22
* Determined from the difference between the maximum wavelengths of the excitation
and the emission bands; sh = shoulder.
These findings also may drive future studies based on fluorescence quenching to
investigate interactions of biomolecules, such as DNA and albumin, with naproxen-containing
metal compounds. Compound 2 exhibits intense emission band at about 350 nm, assignable to
the presence of the naproxen drug, which may overlap with the emission bands of
biomolecules. Therefore, the interference from naproxen emission could only be avoided by
appropriate choice of the emission wavelength in studies with biomolecules [121]. Both
Ru(II)-arenes also show absorption at the UV spectral region what would require correction
for inner-filter effect [122]. None of these problems have been mentioned in the previous
report describing the interaction of 2 with biomolecules [64].
3.6 Biological Studies
3.6.1 Anti-proliferative effects on lung cancer cell lines
The two Ru(II)-arene compounds and their correspondent parent drugs were screened
in vitro against two different ling cancer cell lines, A549 and NCI-H460, to evaluate the
cytotoxicity (in terms of IC50, the concentration of the drug that causes 50% inhibition in cell
viability). No significant anti-proliferative activity was found for compound 2 and the NaL2
drug salt, after 24 h treatment, in both cell types (IC50 > 200 µmol L-1, Table 3). These
findings are in disagreement with the previous report [64] which claims a marked
antiproliferative activity (nanomolar concentration) for 2 against A549 lung cancer cells. The
reason for the discrepancy between these results is difficult to establish. Firstly, compound 2
is unstable in dmso since gradual dissociation of the L2 ligand leads to distinct species which
may coexist in equilibrium, and hence the composition of species in the biological medium
may be different depending on the bioassay experimental conditions. More importantly, we
have conducted the assay by the MTT method [123], while the literature data [64] are based
20

on the Sulforhodamine B assay (SRB) [124]. MTT method relies on measuring metabolic
activity to investigate the anticancer efficacy expressed by the values of IC50, whereas the
SRB method is based on the property of SRB to bind proteins under mild acidic conditions
and the cytotoxicity results are expressed in terms GI50 (concentration of the drug that
produces 50 % inhibition of the cells). Although both methods can be used for in vitro
anticancer drug screening, each one shows its own peculiarities and chemosensitivity, which
depend on the type of cell line. The differences in data from these two methods, however,
should not be highly significant [125,126], and, moreover, cytotoxicity data from distinct
methodologies, by considering all variables, should not exhibit difference much higher than
20%, in general. Therefore, the difference between our results (IC50 > 200 µmol L-1) and the
reported data (GI50 at nanomolar) [64] should not be so high as it is found to be for compound
2. Another relevant aspect to be taken into account is the fact that since the SRB dye stains the
total protein content in the cell, the SRB assay may not discriminate the cells that have
changed their metabolic activity after the treatment but retained the protein content [127].
Furthermore, the arresting of the cells in the S phase of the cell cycle might be wrongly
interpreted as a cell death effect.
The modified drug L1 shows antiproliferative effects in both A549 and NCI-H460
lung cancer cells (IC50 = 158.8 and 136.10 µmol L-1, respectively), despite exhibiting IC50
higher than the paclitaxel anticancer positive control (46.49 and 30.53 µmol L-1, respectively).
The anticancer activity of L1, contrasted to the inactivity of naproxen (Table 3), reveals that
the structural modification, by inserting the amide moiety into the structure of naproxen, to
give the pyridineamide-naproxen conjugated drug, is crucial for the anticancer activity. The
treatment with the new organometallic 1 also led to the inhibition of cell proliferation against
both lung cancer cell lines (IC50: 161.00 and 145.30 µmol L-1, for A549 and NCI-H460
respectively), thus giving support to the key role of L1 to decrease cell viability. It is worth to
mention that although the IC50 values of 1 are slightly higher than those found for the L1 free
drug, the differences between them are small (Table 3). These findings, at the time biological
assays were performed, raised the suspicion that organometallic 1 might undergo ligand
1
dissociation in dmso (phenomenon that was further confirmed by the time-dependent H
NMR studies discussed above), being that the antiproliferative activity could be due only to
the free L1. However, further experiments on the effects on mitochondrial membrane
potential (discussed below) suggest that Ru-species may play key role in mechanisms
associated to drug activity when cells are treated with compound 1.
21

Table 3. In vitro cell viability against A549 and NCI-H460 lung cancer cells.
Compound
IC50 values ± SD (µmol L-1 )
A549
161.00 ± 12.09
158.80 ± 4.47
> 200
NCI-H460
145.30 ± 13.23
136.10 ± 9.02
> 200
1
L1
2
> 200
46.49 ± 13.07
> 200
30.53 ± 7.21
NaL2
Paclitaxel
Data presented as mean ± standard deviation of three independent experiments.
IC50 = half maximal inhibitory concentration. SD = standard deviation.
3.6.2 Effects on Mitochondrial Membrane Potential
The Npxpya drug (L1) and organometallic 1 were investigated for possible mechanisms
associated to drug activity in the NCI-H460 cell line. The mitochondrial membrane potential
(MMP) is a key parameter for investigating mitochondrial function since the MTT assay is
based mainly in mitochondrial metabolic activity and this organelle plays crucial role in cell
death pathways. CCCP and paclitaxel were used as positive controls for mitochondrial
membrane depolarization and apoptosis process, respectively. The effects of the drugs (1 and
L1) were evaluated in two different drug concentrations (75 and 150 µmol L-1). Experiments
were also performed for precursor 3 (37.5 and 75 µmol L-1 which gives 75 and 100 µmol L-1 of
dmso-species 4, respectively). The results show that the MMP is significantly decreased (p <
0.001) in the presence of the L1 free drug, at a concentration (150 µmol L-1) close to the drug
IC50 (Figure 2). The finding indicates that the mechanism related to the Npxpya drug activity
might be mainly associated to anti-proliferative effects. Conversely, no significant effects on
the MMP (Figure 2) are found when the cells are treated with compound 1 or precursor 3
(which generates Ru(cym)-dmso species 4 in dmso). Since we know at this point that
organometallic 1 undergoes complete dissociation in dmso to give species 4 and L1, it is
reasonable to think that the presence of Npxpya-free Ru species, probably in the form of
species 4, might play a role in inhibiting the mechanism related to the mitochondrial function
when cells are treated with 1, in comparison with the cell treatment with the L1 free drug.
22

.
75
100
10
75
150
75
150
150
Control CCCP Paclitaxel
L1
1
4
Figure 2. Flow-cytometry analysis of NCI-H460 cells, after 24 h treatment with L1 and 1 (at
75 and150 µmol L-1) and Ru(cym)-dmso species 4 (from 3 at 37.5 and 75 µmol L-1), stained
with TMRE. CCCP and paclitaxel were used as positive controls for mitochondrial
membrane depolarization and apoptosis process, respectively. The mean fluorescence of
positive cells was used to assess changes in the MMP. The data are the means ± SD from
three independent experiments. * p < 0.05 vs. control and *** p < 0.001 vs. control.
3.6.3 Cell Cycle arrest
The anti-proliferative activities were evaluated by investigating the cellular
distribution among the cell cycle phases by flow cytometry. Nocodazol and paclitaxel were
used as positive controls of G2/M and G0/G1 arrests, respectively. Both L1 and 1 were found
to arrest cells in S-phase at concentration (75 µmol L-1) lower than their IC50 values, and to
arrest G0/G1-phase at 150 µmol L-1 (Figure 3a), what suggests a cytostatic mechanism of
action. The findings are corroborated by Western blotting protein expression analysis by the
monitoring of several proteins related to cell death (Figure 3b) and cell cycle progression
(Figure 3c) (rTRAIL was used as a positive control for apoptosis). Npxpya (L1) shows an
apoptotic feature even at 75 µmol L-1, by inducing caspase 3 cleavage and reduction of anti-
apoptotic proteins such as Bcl-2 and Bcl-xL (Figure 3b). By it turns, compound 1 also leads to
reduction of Bcl-2 and Bcl-X Bcl-2 and Bcl-xL l, at 75 µmol L-1. Nevertheless, both L1 and 1
also promote the reduction of apoptotic proteins, such as BAD and BAX, which probably
makes the scenario not favorable enough to apoptosis, leading to unchanged expression of
caspase 8 and 9 cleaved and even reduction of cytochrome c. Concerning the cell cycle
progression proteins (Figure 3c) L1 and 1 are found to reduce cyclin-dependent kinase 4 and
6, cyclins A, D1, D3 and E2, which are mitogenic proteins, and also p-cdc2, which has
dephosphorylation related to mitosis. In addition to the damage caused by both L1 and 1
23

leading to a reduction of p38 inactive and to an increase of p-p53 by L1 at 150 µmol L-1, the
anti-apoptotic feature mentioned before is confirmed by NF-ƙB reduction.
Conclusions
The new half-sandwich organometallic 1 of formula [Ru(η6-p-cymene)(L1)Cl2],
where L1 is the N-coordinated naproxen-pyridinamide, was successfully synthesized and
characterized, as well as the L1 parent drug ligand. The reported compound 2, [Ru(η6-p-
cymene)(L2)Cl], bearing bidentate naproxenate was re-prepared and additional investigation
was performed. DFT computational calculations corroborate the proposed structures for both
1
Ru(II)-arene compounds. The time-dependent HNMR shows that both organometallics,
despite the good stability in non-coordinating solvents, undergo dissociation in dmso to give
Ru(cym)-dmso species 4 and the corresponding free drug ligand. Distinct behavior indicates
that while compound 2 loses L2 gradually, the dissociation of L1 from compound 1 occurs
instantaneously shortly after dissolution. The experimental data from electronic absorption
spectroscopy shows good correlation with the DFT computational calculations.
Luminescence properties of both Ru(II)-arenes, mainly associated to the naproxen moiety,
suggest potential applications based on photoactivity, and may guide future studies based on
fluorescence quenching to investigate interactions of compounds of this type with
biomolecules. The Ru(II)-arene compounds show distinct behavior against NCI-H460 and
A549 lung cancer cell lines. The treatment of the cells with 2 indicates no anticancer activity,
while the treatment with the new organometallic 1 led to the inhibition of cell proliferation
similarly to the L1 free drug. These findings give evidence for the crucial role of the
structural modification of naproxen (by the insertion of the amide moiety into the original
structure of the drug to form the naproxen-pyridineamide conjugate) in promoting anticancer
activity. The L1 parent free drug was found to decrease significantly the mitochondrial
membrane potential (MMP) in NCI-H460 cells, while compound 1, as well as species 4
(generated from dissolution of 3 in dmso), exhibits no significant effects on the MMP.
1
Interestingly, since HMNR studies indicate that organometallic 1 is completely dissociated
in dmso (the solvent used to prepare the drug solutions for cell treatment in the biological
assays) to give the L1 free drug and species 4, it is plausible to infer that the presence of
Npxpya-free Ru species, probably in the form of species 4, might play a role in inhibiting
the mechanism related to the mitochondrial function when cells are treated with 1, in
comparison with the cell treatment with the L1 free drug.
24

PI
Control
Nocodazol
Paclitaxel
(a)
PI
L1
Npxpya
1
RucymNpxpya
75 µmol L-1 150 µmol L-1
75 µmol L-1 150 µmol L-1
10
10
75
150
75
150
(µmol L-1)
L1
Control Nocodazol Paclitaxel Npxpya
1
RucymNpxpya
Untreated
Untreated
Npxpya
75 µmol L-1
L1
Npxpya
150 µmol L-1
L1
RucymNpxpya 75 µmol L-1
1
1
RucymNpxpya 150 µmol L-1
Untreated
Untreated
Npxpya
rTRArTILRAIL
25 ng mL-1
75 µmol L-1
L1
Npxpya
L1
150 µmol L-1
RucymNpxpya 75 µmol L-1
1
RucymNpxpya 150 µmol L-1
1
rTRAIL
rTRAIL
25 ng mL-1
(b)
(c)
Figure 3. (a) Cell cycle phases G0/G1, S, and G2/M indicated in both the histogram and
bars of % cell population distribution (data are the mean ± SD from three independent
experiments); (b) Western blotting analysis of proteins expression related to cell death; (c)
Cell cycle progression. rTRAIL 25 ng/mL was used as a positive control for apoptosis. β-
Actin levels were used as loading control. Images are representative of at least three
independent experiments.
25

Acknowledgements
The authors gratefully acknowledge the financial support to perform the present work
to the Brazilian agencies: Fundação de Amparo à Pesquisa do Estado de São Paulo
(FAPESP, research grants to D. de Oliveira Silva [2014/23047-5 and 2018/00297-4] and
to A. K. Ferreira [2015/18528-7]), Conselho Nacional de Desenvolvimento Científico e
Tecnológico (CNPq fellowships, for productivity to D. de Oliveira Silva
[305914/2015-4] and to A. K. Ferreira [304255/2017-3], and for Master Project to S.
Q. Martinez), and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
(CAPES, finance code 001; Doctorate fellowships to J. P. G. Tabares, J. L. Cassiano, A.
C. Stábile). The authors sincerely acknowledge Prof. P. S. Santos and J. A. A. Castilho
(doctorate student) for registration of the IR spectra at Laboratory for Molecular Spectroscopy
(LEM), Prof. L. Marzorati, Prof. N. Y. M. Iha and Prof. H. E. Toma for technical support, at
Institute of Chemistry, University of São Paulo; Prof. L. C. Salay and L. A. Santos (bachelor
student) at University of Santa Cruz. The authors are also grateful to Prof. K. P. M. Frin at
Federal University of ABC (for suggestions on fluorescence studies); to the Center for
Scientific Computing (NCC/GRIGUNESP) of the São Paulo State University (UNESP) and
Centro Nacional de Processamento de Alto Desempenho (CENAPAD-SP) (for provision of
computational facilities); and to the Alchemy, Innovation, Research & Development.
Appendix A. Supplementary material (SM.)
References
[1] M. Gielen, E.R.T. Tiekink, Eds. Metallotherapeutic Drugs & Metal-based Diagnostic
Agents - The use of metals in medicine, 2005, John Wiley & Sons Ltd., West Sussex.
[2] E. Alessio, Ed. Bioinorganic Medicinal Chemistry, 2011, Wiley-VCH Verlag & Co.
KGaA, Weinheim.
[3] D. de Oliveira Silva. Ruthenium Compounds Targeting Cancer Therapy. In Atta-ur-
Rahman, M. I. Choudhary, Eds, Frontiers in Anti-Cancer Drug Discovery, 2014, Bentham
Science Publishers Ltd., Sharjah, vol. 4, 88-156.
[4] W. H. Ang, P. J. Dyson, Eur. J. Inorg. Chem., 2006, 20, 4003-4018.
[5] D. de Oliveira Silva, Anti Canc. Agents Med. Chem., 2010, 10, 312-323.
[6] M.A. Jakupec, M. Galanski, V.B. Arion, C.G. Hartinger, B.K. Keppler. Dalton Trans,
2008, 2, 183-194.
[7] I. Kostova, Curr. Med. Chem., 2006, 13, 1085-1107.
26

[8] A. Bergamo, G. Sava, Dalton Trans, 2007, 13, 1267-1272.
[9] A. Levina, A. Mitra, P.A. Lay, Metallomics, 2009, 1, 458-470.
[10] G.S. Fink, Dalton Trans., 2010, 39, 1673-1688.
[11] W.H. Ang, A. Casini, G. Sava, P.J. Dyson, J. Organomet. Chem., 2011, 696, 989-998.
[12] G. Sava, A. Bergamo, P.J. Dyson, Dalton Trans., 2011, 40, 9069-9075.
[13] G. Gasser, I. Ott, N. Metzler-Nolte, J. Med. Chem., 2011, 54, 3-25.
[14] A. Bergamo, C. Gaiddon, J.H.M. Schellens, J.H. Beijnen, G. Sava, J. Inorg. Biochem.,
2012, 106, 90-99.
[15] C.G. Hartinger, N. Metzler-Nolte, P.J. Dyson, Organometallics, 2012, 31, 5677-5685.
[16] S. Komeda, A. Casini, Curr. Top. Med. Chem., 2012, 12, 219-235.
[17] G. S. Fink, J. Organom. Chem., 2014, 751, 2-19.
[18] A. A. Nazarov, C. G. Hartinger, P. J. Dyson, J. Organom. Chem., 2014, 751, 251-260.
[19] M. J. Thun, S. J. Henley, C. Patrono. J. Nat. Cancer Inst., 2002, 94, 252-266.
[20] T. Tanaka, T. Kojima, N. Yoshimi, S. Sugie, H. Mori, Carcinogenesis, 1991, 12, 1949-
1952.
[21] P. J. Hu, J. Yu, Z. R. Zeng, W. K. Leung, H. L. Lin, B. D. Tang, A. H. Bai, J. J. Sung,
Gut, 2004, 53,195-200.
[22] R. E. Harris, S. Kasbari, W.B. Farrar, Oncol. Rep., 1999, 6, 71-73.
[23] T. Hida, J. Leyton, A. N. Makheja, P. Ben-Av, T. Hla, A. Martinez, J.Mulshine, S.
Malkani, P. Chung, T.W. Moody, Anticancer Res., 1998, 18, 775-782.
[24] J. E. Weder, C. T. Dillon, T. W. Hambley, B. J. Kennedy, P. A. Lay, J. R. Biffin, H.
L. N. Regtop, M. Davies, Coord. Chem. Rev., 2002, 232, 95-126.
[25] C. N. Banti and S. K. Hadjikakou, Eur. J. Inorg. Chem., 2016, 3048-3071.
[26] F. A. Cotton, D. de Oliveira Silva, Inorg. Chim. Acta, 1996, 249, 57-61.
[27] A. Andrade, S.F. Namora, R.G. Woisky, G. Wiezel, R. Najjar, J.A.A. Sertié, D.de
Oliveira Silva, Inorg. Biochem., 2000, 81, 23-27.
[28] C. R. Gordijo, C. A. S. Barbosa, A. M. C. Ferreira, V. R. L. Constantino, D. de Oliveira
Silva, J. Pharm. Sci., 2005, 94, 1135-1148.
[29] D. J. Martins, Hanif-Ur-Rehman, S. R. Alves Rico, I. M. Costa, A. C. P. Santos, R. G.
Szszudlowski, D. de Oliveira Silva, RCS Advances, 2015, 5, 90184-90192.
[30] G. Ribeiro, M. Benadiba, A. Colquhoun, D. de Oliveira Silva, Polyhedron, 2008, 27,
1131-1137.
[31] M. Benadiba, R.R.P. Santos, D. de Oliveira Silva, A. Colquhoun, J. Inorg. Biochem.,
2010, 104, 928-935.
27

[32] M. Benadiba, I.M. Costa, R.L.S.R. Santos, F.O. Serachi, D. de Oliveira Silva, A.
Colquhoun, J. Biol. Inorg. Chem., 2014, 19, 1025-1035.
[33] R.L.S.R. Santos, A. Bergamo, G. Sava, D. de Oliveira Silva, Polyhedron, 2012, 42, 175-
181.
[34] R.L.S.R. Santos, R.N.F. Sanches, D. de Oliveira Silva, J. Coord. Chem., 2015, 68, 3209-
3228.
[35] Hanif-Ur-Rehman, T.E. Freitas, R.N. Gomes, A. Colquhoun, D. de Oliveira Silva. J.
Inorg. Biochem., 2016, 165, 181-191.
[36] S.R. Alves Rico, A.Z. Abbasi, G.Ribeiro, T.Ahmed, X.Y. Wu, D. de Oliveira Silva.
Nanoscale, 2017, 9, 10701-10714.
[37] G. Sava, S. Zorzet, C. Turrin, F. Vita, M. R. Soranzo, G. Zabucchi, M. Cocchietto, A.
Bergamo, S. DiGiovine, G. Pezzoni, L. Sartor, S. Garbisa, Clin. Cancer Res., 2003, 9, 1898-
1905.
[38] A. Bergamo, B. Gava, E. Alessio, G. Mestroni, B. Serli, M. Cocchietto, S. Zorzet, G.
Sava, Int. J. Oncol., 2002, 21, 1331-1338.
[39] C.G. Hartinger, S.Z-Seifred, M.A. Jakupec, B. Kynast, H. Zorbas, B. Keppler, J. Inorg.
Biochem., 2006, 100, 891-904.
[40] B. Schoenhacker-Alte, T. Mohr, C. Pirker, K. Kryeziu, P.S. Kuhn, A. Buck, T. Hofmann,
C. Gerner, G. Hermann, G. Koellensperger , B. K. Keppler, W. Berger, P. Heffete, Cancer
Letters, 2017, 404 , 79-88.
[41] C.G. Hartinger, P.J. Dyson, Chem. Soc. Rev., 2009, 38, 391-401.
[42] P. Zhang, P. J. Sadler, J. Organom. Chem., 2017, 839, 5-14.
[43] B.S. Murray, M.V. Babak, C.G. Hartinger, P.J. Dyson, Coord. Chem. Rev., 2016, 306,
86-114.
[44] G. Palermo, A. Magistrato, T. Riedel, T. von Erlach, C. A. Davey, P. J. Dyson, U.
Rothlisberger, Chem. Med. Chem., 2016, 11, 1199-1210.
[45 . unescu, S. McArthur, M. Soudani, R. Scopelliti, P. J. Dyson. Inorg. Chem., 2016,
55, 1788-1808.
[46] A. Ashraf, M. Hanif, M. Kubanik, T. Söhnel, S. M. F. Jamieson, A. Bhattacharyya, C. G.
Hartinger, J. Organom. Chem., 2017, 839, 31-37.
[47] Q. Ru, X. Tian, Y. X. Wu, R.H. Wu, M. S. Pi, C. Y. Li, Oncol. Rep., 2014, 31, 842-848.
[48] E. K. Hoffmann, I. H. Lambert, Phil. Trans. R. Soc. B., 2014, 369, 20130109.
[49] W. Wang, J. Xiao, M. Adachi, Z. Liu, J. Zhou, Cell. Physiol. Biochem., 2011, 28, 199-
208.
28

[50] L. S. Chin, C. C. Park, K. M. Zitnay, M. Sinha, A. J. DiPatri, P. Perillán, J. M. Simard, J.
Neurosci. Res., 1997, 48, 122-127.
[51] F. J. Al-Shammary, N. A. Aziz Mian, M. S. Mian, Anal. Profiles Drug Subst. Excipients,
1992, 21, 345-373.
[52] T. M. K. Motawi, Y. Bustanji, S. EL-Maraghy, M. O. Taha, M. A. S. Al-Ghussein, J.
Enzym. Inhib. Med. Ch. , 2014, 29, 153-161.
[53] M. S. Kim, J. E. Kim, D. Y. Lim, Cancer Prev. Res., 2014, 7, 236-245.
[54] J. Deb, J. Majumder, S. Bhattacharyya, S. S. Jana, BMC Cancer, 2014, 14, 567-574.
[55] B. Tarusa, H. Bertrandb, G. Zeddab, C. D. Primoc, S. Quideaub, A. Slama-Schwok, J.
Biomol.Struct. Dyn., 2015, 33, 1899-1912.
[56] Y. A. Ammar, M. A. Salem, Eman A. Fayed, M. H. Helal, M. S. A. El-Gaby, H. Kh.
Thabeta, Synthetic Commun., 2017, 47, 1341-1367.
[57] K. C. Duggan, D. J. Hermanson, J. Musee, J. J. Prusakiewicz, J. L. Scheib, B. D. Carter,
S. Banerjee, J. A. Oates, L. J. Marnett, Nat. Chem. Biol., 2011, 7, 803-809.
[58] M. S. Kim, J. E. Kim, D.Y. Lim, Z. Huang, H. Chen, A. Langfald, R. A. Lubet, C.J.
Grubbs, Z. Dong, A.M. Bode, Cancer Prev. Res., 2014, 7, 236-245.
[59] V. Lhiaubet-Vallet, M.A. Miranda, Pure Appl. Chem., 2006, 78, 2277-2286.
[60] D. E. Moore, P. P. Chappuis, Photochem. Photobiol., 1988, 47, 173-180.
[61] . Kuś, V. Kozik, M. Rojkiewicz, A. Sochanik, A. Szurko, M. Kempa, . Kozub, M.
Rams-Baron, K. Jarzembek, M. Stefaniak, J. Sakowicz, Dyes Pigments, 2015, 116, 46-51.
[62] A. Catalfo, G. Bracchitta, G. De Guidi, Photochem. Photobiol. Sci., 2009, 8, 1467-1475.
[63] T. M. Georgiadis, M. M. Georgiadis, F. Diederich, J. Org. Chem., 1991, 56, 3362-3369.
[64] P. Mandal, B. K. Kundu, K. Vyas, V. Sabu, A. Helen, S. S. Dhankhar, C. M. Nagaraja,
D. Bhattacherjee, K. P. Bhabake, S. Mukhopadhyay, Dalton Trans., 2018, 47, 517-527.
[65] M. A. Bennett, T. N. Huang, T. W. Matheson, A. K. Smith, J. P. Fackler, Jr., Ed.,
Inorganic Syntheses XXI, 1982, John Wiley & Sons, Inc., New York, vol. XXI, 74.
[66] J. Will, A. Kyas, W. S. Sheldrick, D. Wolters, J. Biol. Inorg. Chem., 2007, 12, 883-894.
[67] S. Fery-Forgues, D. Lavabre, J. Chem. Educ., 1999, 76, 1260-1264.
[68] L.J. Martínez, J.C. Scaiano, Photochem. Photobiol., 1998, 68, 646-651.
[69] M.J. Frisch , G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheesman,
V.G. Zakrzewski, J.A. Montgomery Jr., R.E. Stratmann, J.C. Burant, S. Dapprich, J.M.
Millam, A.D. Daniels, K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi,
R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G.A. Petersson,
P.Y. Ayala, Q. Cui, K. Morokuma, N. Roga, P. Salvador, J.J. Dannenberg, D.K. Malick, A.D.
Rabuck, K. Raghavachari, J.B. Foresman, J. Cioslowski, J.V. Ortiz, A.G. Baboul, B.B.
29