Structure-activity relationships of anticancer ruthenium(II) complexes with substituted hydroxyquinolines

Article abstract of DOI:10.1016/j.ejmech.2018.04.044

8-Hydroxyquinolines (HQ), including clioquinol, possess cytotoxic properties and are widely used as ligands for metal-based anticancer drug research. The number and identity of substituents on the HQ can have a profound effect on activity for a variety of inorganic compounds. Ruthenium complexes of HQ exhibit radically improved potencies, and operate by a new, currently unknown, mechanism of action. To define structure-activity relationships (SAR), a family of 22 Ru(II) coordination complexes containing mono-, di- and tri-substituted hydroxyquinoline ligands were synthesized and their biological activity evaluated. The complexes exhibited promising cytotoxic activity against a cancer cell line, and the SAR data revealed the 2- and 7-positions as key sites for the incorporation of halogens to improve potency. The Ru(II) complexes potently inhibited translation, as demonstrated by an in-cell translation assay. The effects were seen at 2–15-fold higher concentrations than those required to observe cytotoxicity, suggesting that prevention of protein synthesis may be a primary, but not the exclusive mechanism for the observed cytotoxic activity.

Full text of DOI:10.1016/j.ejmech.2018.04.044

Accepted Manuscript
Structure-activity relationships of anticancer ruthenium(II) complexes with substituted
hydroxyquinolines
Dmytro Havrylyuk, Brock S. Howerton, Leona Nease, Sean Parkin, David K. Heidary,
Edith C. Glazer
PII:
S0223-5234(18)30378-7
10.1016/j.ejmech.2018.04.044
EJMECH 10393
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 21 February 2018
Revised Date: 16 April 2018
Accepted Date: 21 April 2018
Please cite this article as: D. Havrylyuk, B.S. Howerton, L. Nease, S. Parkin, D.K. Heidary, E.C. Glazer,
Structure-activity relationships of anticancer ruthenium(II) complexes with substituted hydroxyquinolines,
European Journal of Medicinal Chemistry (2018), doi: 10.1016/j.ejmech.2018.04.044.
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.

ACCEPTED MANUSCRIPT
Structure-Activity Relationships of Anticancer Ruthenium(II) Complexes
with Substituted Hydroxyquinolines
Dmytro Havrylyuk, Brock S. Howerton, Leona Nease, Sean Parkin, David K. Heidary,* and
Edith C. Glazer*
Department of Chemistry, University of Kentucky, 505 Rose Street, Lexington, Kentucky 40506,
United States
ABSTRACT
8-Hydroxyquinolines (HQ), including clioquinol, possess cytotoxic properties and are
widely used as ligands for metal-based anticancer drug research. The number and identity of
substituents on the HQ can have a profound effect on activity for a variety of inorganic
compounds. Ruthenium complexes of HQ exhibit radically improved potencies, and operate by a
new, currently unknown, mechanism of action. To define structure-activity relationships (SAR),
a family of 22 Ru(II) coordination complexes containing mono-, di- and tri-substituted
hydroxyquinoline ligands were synthesized and their biological activity evaluated. The
complexes exhibited promising cytotoxic activity against a cancer cell line, and the SAR data
revealed the 2- and 7-positions as key sites for the incorporation of halogens to improve potency.
The Ru(II) complexes potently inhibited translation, as demonstrated by an in-cell translation
assay. The effects were seen at 2–15-fold higher concentrations than those required to observe
cytotoxicity, suggesting that prevention of protein synthesis may be a primary, but not the
exclusive mechanism for the observed cytotoxic activity.
1

ACCEPTED MANUSCRIPT
1. INTRODUCTION
Coordination complexes containing 8-hydroxyquinoline ligands (HQ) have shown
promise for the development of small molecule drugs, particularly in anticancer research.[1] Most
notably, tris-8-HQ gallium(III) (KP46) has reached clinical evaluation in phase I trials, and
exhibited activity in the treatment of renal cell carcinoma.[2] This complex was discovered and
patented due to its potential efficacy for treating pancreatic cancer,[3] and was also highly active
against osteosarcoma cells by inducing cancer cell death via a p53 dependent mechanism, and
inhibiting cellular migratory potential.[4]
Various other metal complexes of HQ ligands have been investigated, with a range of
oxidation states and coordination numbers. These include silver (I),[5] copper(II)[6],
platinum(II),[7] cobalt(II),[8] zinc(II),[9] gold(III),[10] and rhodium(III).[11] Both unsubstituted and
substituted HQ ligands have been incorporated into complexes, but often the individual studies
described only a few systems, preventing any conclusive structure-activity relationship (SAR)
conclusions from being drawn. In other cases, conflicting SAR patterns have been reported. For
example, a Pt(II) complex with unsubstituted HQ was identified as the most active in specific
cell lines, while a Pt(II) complex with clioquinol possessed the highest cytotoxicity in others,16
and the complex bearing the 5,7-diiodo-HQ ligand was the more potent entity in a different
study.17 In nearly all cases of homoleptic metal complexes, though, the free HQ ligands were less
potent than corresponding coordination complexes. For example, cobalt(II) complex of 5-chloro-
8-hydroxyquinoline showed higher cytotoxicity than the corresponding metal salt
(Co(NO3)2•6H2O) and the free ligand when tested with five tumor cell lines.[8a] Complexes of
5,7-dihalo-HQs with lanthanides[12], tin(IV)[13], nickel(II)[14], zinc(II), copper(II)[15], cerium(III,
2

ACCEPTED MANUSCRIPT
IV)[16] and iron(III)[17] have been reported, and the complexes exhibited significantly enhanced
cytotoxicities compared to parent HQ ligands, with single micromolar to nanomolar IC50 values.
Less common are metal complexes containing only one HQ ligand, and in these cases
metal coordination can increase or decrease potency, depending on the other components of the
system. A study performed by Hartinger and coworkers investigated coordination of HQs ligands
to a Ru(II)(η6-p-cymene) scaffold, where halogens at the 5- and 7-positions of the HQ ligand
were systematically varied.[18] In this report the metal complexes were less potent than the
corresponding free ligands, and little variation was found with regards to the identity of the
halogen. In other reports, coordination resulted in decreases or only modest improvements in
potency.[19] In contrast, we previously demonstrated that the coordination of HQs to the
[Ru(dmphen)2] scaffold (dmphen = 2,9-dimethyl-1,10-phenanthroline) yielded a significant
improvement in cytotoxicity compared to the parent ligands, with potencies up to 86-fold greater
than clioquinol.[1a] The complexes were also >100-fold more potent than clioquinol in a 3D
tumor spheroid model, with values similar to chemotherapeutics currently used for the treatment
of solid tumors. We observed that the Ru(II) scaffold played a major role in driving the potency
of the complexes, with compounds containing bpy coligands being far less active. Two similar
molecules were investigated in vivo by Liu and coworkers, with the 8-hydroxyquinoline ligand
coordinated to Ru(II) centers containing either 2,2'-bipyridine (bpy) or 1,10-phenanthroline
(phen) coligands. They showed promising inhibition of angiogenesis and tumor growth, with
effects observed for the phen complex at concentrations of 8 mg kg-1d-1.[20] Thus, Ru(II)
heteroleptic complexes containing HQ ligands possess noteworthy activity both in vitro and in
vivo.
3

ACCEPTED MANUSCRIPT
These findings have motivated us to pursue a comprehensive SAR investigation of Ru(II)
complexes with mono-, di- and tri-substituted hydroxyquinoline ligands in order to identify the
optimal structural frameworks for further medicinal chemistry efforts. The main goal of this
study was to answer the following questions: 1) Does the nature of the substituent (halogens vs.
methyl or aryl groups) influence the cytotoxic effect? 2) What positions of HQ should be
modified for enhanced activity?
2. RESULTS AND DISCUSSION
2.1 Chemistry. Our earlier SAR analysis of HQ complexes with the [Ru(dmphen)2] scaffold
revealed that the presence of halogens at the 5- and 7-positions resulted in the most potent
compounds, while incorporation of electron rich substituents such as a nitro group or sulfonic
acids at the 5-position of the hydroxyquinoline reduced potency up to 220-fold.[1a] Therefore, in
this study we focused on halogen-, methyl- and aryl-substituted HQs, generating complexes with
mono-, di- and tri-substituted HQ ligands. The Ru(II) complexes were synthesized from a
racemic mixture of the ∆ and Λ enantiomers of [Ru(dmphen)2Cl2] and form a mixture of
enantiomers upon coordination of the hydroxyquinoline ligand.[1a] All complexes were
exhaustively purified to ensure no contamination of either free ligands or coordinatively
unsaturated Ru(II) center. The yields were moderate or low for some complexes, due in part to
the fact that the [Ru(dmphen)2] scaffold is sterically congested.
Aiming to identify the impact of halogen substitution on the biological activity, the analogous
chloro- and bromo-substituted HQ ligands were used. To allow for comparison of variation is the
radius of the substituent, as well as its electronic nature, a methyl-substituted HQ was also
investigated. In order to introduce a larger substituent, the Ru(II) complexes with 5- (compound
4

ACCEPTED MANUSCRIPT
6) or 7-bromo-HQs (compound 9) were modified via the Suzuki coupling reaction, yielding
ruthenium compounds 8 and 11 with aryl-substituted HQ ligands. Interestingly, the synthesis of
the analogous 2-substituted compound, 5, failed on the metal complex. This was hypothesized to
be due to steric constraints, despite successful reaction at the 7-poistion, which is also partially
occluded by the coodinated Ru(II) center. As a result, the free ligand was first subjected to the
coupling reaction and then coordinated to the Ru(II) center to form compound 5. Both complexes
5 and 11 displayed a single resonance for the Me group in 1H NMR spectra, but resonances for
Me groups of the o-tolyl fragment of complex 8 were resolved as two singlets. This indicated
restricted rotation of the o-tolyl fragment and presence of two rotamers with 2:3 ratio.
Under the reaction conditions for the coordination of 7-chloro-HQ to the [Ru(dmphen)2]
scaffold in ethanol-water medium (1:1), the complex unexpectedly underwent an oxidative
coupling (dimerization), producing the Ru(II) dimer 22 linked at the 5 position of the HQ rings
(Scheme S1). The structure of 22 was confirmed by 1H NMR and ESI MS spectra (Figures S28,
29) and X-ray, as discussed below.
2.2 Crystallography
The structures of complexes
2, 9 and 14 were determined by X-ray crystallography
(Figures 1, S1-3). Selected bond lengths and angles are listed in the Table 1.
5

ACCEPTED MANUSCRIPT
Figure 1. Ellipsoid plot of ruthenium complexes: (A) (∆)-2, (B) (∆)-9, (C) (∆)-14 at 50 %
probability with H atoms omitted for clarity. Right column: side views, highlighting the
distortion of the dmphen ligand.
All complexes exhibited distorted octahedral geometries. The incorporation of two dmphen
ligands and HQ resulted in shortening of the Ru−N(dmphen) bonds to 2.093 Å (average value for
6

ACCEPTED MANUSCRIPT
2), 2.096 Å (average value for 9), and 2.089 Å (average value for 14) in comparison with
analogous complex containing 2,3-dihydro-1,4-dioxino[2,3-f]-1,10-phenanthroline32 or pyridyl-
benzazole 33
on the [Ru(dmphen)2] scaffold, where the average values are 2.103–2.117 Å. Introduction of a
methyl group into position 2 of the HQ (2) caused the Ru-N5 bond to lengthen to 2.162 Å in
comparison to complexes 9 and 14. The Ru-O bonds are longer for compounds 9 and 14, most
likely due to the presence of halogens at the 7-position of HQs. The bond angles between the
dmphen and HQ ligands are nonequivalent, with the largest distortion of L1 (dmphen where L is
N1 and N2) for complex 14 (Figure 1C). The angles change from the ideal 90º to 79.05-105.38º
and 180º to 168.58-177.02º. Both the dmphen ligands (L1 and L2, Figure 1, Table 1) for each
compound are considerably bent from the normal plane, with deviations of 8.3–23.1º. While the
bend angles for both dmphen ligands in complex 9 are similar, the bends of L1 for compounds 2
and 14 are significantly larger than those for L2.
For the dimer 22, the crystals were twinned by non-merohedry and diffracted poorly, giving
diffuse but indexable, Bragg diffraction to not quite 1Å resolution. Although the structure solved
with relative ease, it did not refine to commonly accepted standards. Nevertheless, the
connectivity of the molecule is consistent with the dimerization between the HQ rings at the 5-
position. A cartoon of the structure, inspired by the x-ray data, is shown in Figure S30.
7

ACCEPTED MANUSCRIPT
Table 1. Selected bond lengths [A], bond angles [°] and torsion angles [°] of 2, 9 and 14
2
Bond Lengths (Å)
2.111(2)
9
14
Ru-N1
Ru-N2
Ru-N3
Ru-N4
Ru-N5
Ru-O
2.114(2)
2.095(2)
2.079(2)
2.099(2)
2.063(2)
2.0924(19)
2.1048(16)
2.0859(16)
2.0757(16)
2.0891(16)
2.0718(16)
2.0789(13)
2.0852(19)
2.081(2)
2.097(2)
2.162(2)
2.0537(16)
Bond Angles (°)
80.12(8)
N1-Ru-N2
N1-Ru-N3
N1-Ru-N4
N1-Ru-N5
N1-Ru-O
N2-Ru-N3
N2-Ru-N4
N2-Ru-N5
N2-Ru-O
N3-Ru-N4
N3-Ru-N5
N3-Ru-O
N4-Ru-N5
N4-Ru-O
N5-Ru-O
L1 bend a
L2 bend b
80.03(9)
103.16(9)
176.40(9)
98.10(9)
79.86(8)
93.74(9)
100.58(9)
171.70(9)
91.96(8)
80.36(9)
94.56(9)
173.93(8)
80.79(9)
96.56(8)
79.74(8)
14.2
79.53(6)
105.38(6)
174.49(6)
95.65(6)
79.05(6)
97.39(6)
99.46(6)
168.58(6)
88.79(6)
80.11(7)
93.87(6)
172.92(6)
84.34(6)
95.53(6)
80.09(6)
23.1
103.64(8)
176.24(7)
97.20(7)
79.29(7)
94.36(7)
98.32(8)
166.58(8)
86.68(7)
79.85(8)
99.05(8)
177.02(7)
83.52(7)
97.24(7)
79.90(7)
19.7
8.3
11.4
9.6
aL1 bend = 90º – average angle (O-Ru-C13/C14); L1 – dmphen
where L is N1 and N2, as example for complex 2
b L2 bend = 90º – average angle (N5-Ru-C27/C28); L2 – dmphen
where L is N3 and N4, as example for complex 2
8

ACCEPTED MANUSCRIPT
2.3 Cytotoxicity Studies.
To generate SAR to understand and rationally modulate the biological activity of the
Ru(II) complexes, we studied the cytotoxicity in HL60 cells of nine compounds with mono-
substituted HQs, six that were di-substituted, and four tri-substituted analogs, and compared
them to the corresponding complex with the unsubstituted HQ ligand (Table 2, Figures 2 and 3).
The data for four complexes (1, 2, 14 and 16) have been previously published.[1a]
The initial SAR study for the ruthenium complexes coordinated with HQ was based on
the incorporation of a single substituent (a halogen or Me group) into hydroxyquinoline. The
comparison of complexes with 2-substituted HQ ligands showed that the potency depends on the
nature of substituent. The presence of a halogen at the 2-position of HQ improved the
cytotoxicity of complexes by 3–5 fold in comparison to the complex 1 containing the
unsubstituted HQ. The 2-bromo- and 2-chloro-HQ (3 and 4) were potent cytotoxic agents (IC50 =
110–180 nM). However, the addition of a methyl group at the 2-position of HQ (complex 2) did
not significantly increase the potency compared to complex 1. Incorporation of 5-bromo- and 5-
chloro-substituted HQs (6 and 7) resulted in complexes that were 2–4 fold less potent by than
analogous compounds with halogens at the 2-position (3 and 4), though the compounds were
more potent than compound 1. The highest activities were identified for complexes with 7-
bromo-and 7-chloro-HQs (9 and 10). Incorporation of the halogen at the 7-position resulted in at
least 5-fold increases of activity compared to complex 1, with IC50 values ≈100 nM.
Arylation of the 5- and 7-positions reduced the potencies of the compounds. The IC50
value for the complex containing 7-(o-tolyl)-HQ (11) shifted to 0.96 µM, which is approximately
2-fold less potent than the parent complex 1 and 10-fold less potent than the value observed for
the 7-bromo complex 9. The IC50 value was 0.67 µM for the complex containing 5-(o-tolyl)-HQ
9

ACCEPTED MANUSCRIPT
(8), 2-fold less potent than the value observed for the 5-bromo complex 6. Interestingly, the
opposite trend was observed for arylation at the 2-position, with complex 5 being approximately
2-fold more potent than the parent complex 1 and having the same IC50 value as the 2-bromo
complex 3. Thus, significant variation in potencies were found, with values spanned the range
from 0.09 to 0.96 µM. The trends for potencies of monosubstituted HQs coordinated with Ru(II)
scaffold and were as follows: 7-(o-tolyl)-HQ (11) < 5-(o-tolyl)-HQ (8) < HQ (1) ≈ 2-Me-HQ (2)
< 5-Cl-HQ (7) < 5-Br-HQ (6) < 2-(o-tolyl)-HQ (5) ≈ 2-Br-HQ (3) < 2-Cl-HQ (4) ≈ 7-Br-HQ (9)
≈ 7-Cl-HQ (10).
Further SAR analysis was focused on the incorporation of two or three substituents into
the HQ ligand. In contrast to mono-methyl substituted system (complex 2), the addition of
methyl groups at 5- and 7-positions (complex 12) improved the potency by 3-fold compared to 1,
but the dimethyl complex was less active compared to dihalogenated compounds (13–17).
Despite the absence of a radical improvement in cytotoxicity of 5,7-dihalogen substituted HQs
(13–17) in comparison with 7-bromo- or 7-chloro-analogues (9 and 10), it should be noted that
three of the five compounds possessed activity against the HL60 cell line with IC50 values lower
than 100 nM. Complex 13, with 5,7-dibromo-HQ (IC50 = 70 nM), and 15, with 5-chloro-7-
bromo-HQ (IC50 = 87 nM), possessed the same range of activity as complex containing
clioquinol (16, IC50 = 57 nM; Figure 2B). As clioquinol and its complexes undergo degradation
due to a deiodination reaction,[21] complexes 13 and 15 are preferable lead compounds for
further medicinal chemistry efforts as both were found to be stable as well as potent.
Interestingly, halogen size does not appear to be the primary driver for activity, as both
the dichloro-HQ (14) and diiodo-HQ (17) were slightly less potent than the dibromo-HQ 13 and
15, with IC50 values of 110–120 nM. The addition of a third substituent (Me or Cl) at the 2-
10

ACCEPTED MANUSCRIPT
position resulted in the same IC50 values as analogous disubstituted systems, except for complex
20. Thus, compounds 19 and 21 possessed the same range of activity as complex 13, with IC50
values of 68–77 nM. Finally, while the monomethyl HQ complex was no better than an
unsubstituted HQ, the trimethyl HQ complex 18 was twice as potent, but the presence of the
third methyl group at the 2-position was somewhat deleterious (250 nM for 18 vs. 180 nM for
compound 12).
Notably, the Ru(II) dimer (compound 22) possessed the lowest potency, with IC50 value
of 10.01 µM. This dimer was less potent than the previously reported analogous monomer
containing a nitro group at the 5-position (IC50 = 2.31 µM). Only the complex containing a
sulfonic acid at the same position was less potent, with no toxicity observed at concentrations up
to 30 µM.[1a] Thus, it appears that the 5-position is sensitive to both steric bulk and electron rich
substituents; potent compounds can only be achieved with smaller substituents such as halogens,
methyl groups, or a single aromatic ring.
The main findings of the SAR analysis (Figure 3) illustrated that: 1) incorporation of a
halogen at positions 2 and 7 is crucial for improvement of potency, but the nature of the halogen
does not result in radical shifts; 2) the presence of an additional halogen at position 5 slightly
improved potencies in comparison to 7-monosubstituted analogs, and resulted in IC50 values
lower than 100 nM; 3) arylation of the 5- or 7-position significantly reduced the activity, while
arylation at the 2-position increased activity; 4) a Ru(II) dimer linked at the 5-position of the HQ
ligand possessed the lowest potency among described compounds.
11

ACCEPTED MANUSCRIPT
Figure 2. Cytotoxicity dose responses of ruthenium complexes on HL60 cells: (A) activity of
Ru(II) complexes with mono-substituted HQ ligands containing bromine at 2- (3), 5- (6) and 7-
positions (9), compared to complex containing 7-(o-tolyl)-HQ (11); (B) activity of Ru(II)
complexes with di-substituted HQ (13) and tri-substituted HQ (19), compared to the complex
containing clioquinol (16).
12

ACCEPTED MANUSCRIPT
Table 2. Cytotoxicity IC50 Values (µM) for Ru(II) Complexes in the HL60 Cancer Cell Line
Compound R1
R2
R3
IC50, (µM)
mono-substituted HQs
H
H
H
0.52 ± 0.06a
0.49 ± 0.07a
0.18 ± 0.015
0.11 ± 0.003
0.20 ± 0.069
0.32 ± 0.013
0.43 ± 0.04
0.67 ± 0.19
0.10 ± 0.018
0.09 ± 0.004
0.96 ± 0.11
1
Me
H
H
2
Br
Cl
o-tolyl
H
H
H
3
4
5
H
H
H
H
Br
Cl
o-tolyl
H
H
H
H
H
H
Br
Cl
o-tolyl
6
H
7
H
8
H
9
H
H
10
11
di-substituted HQs
H
H
H
H
H
H
Me
Br
Cl
Cl
Cl
I
Me
Br
Cl
Br
I
0.18 ± 0.015
0.07 ±0.008
0.11 ± 0.006a
0.09 ± 0.034
0.057 ± 0.002
0.12 ± 0.004
12
13
14
15
16
17
I
tri-substituted HQs
Me
Me
Me
Cl
Me
Br
Cl
Br
Me
Br
Cl
0.25 ± 0.051
0.08 ± 0.02
0.12 ± 0.002
0.07 ± 0.009
18
19
20
21
Br
Ru(II) dimer
H
-
Cl
11.14 ± 0.658
22
a previously reported data [1a]
13

ACCEPTED MANUSCRIPT
Figure 3. Structure activity relationships for cytotoxicity based on analysis of Ru(II) complexes
with mono-, di-, and tri-substituted HQ ligands. Potency increases upon coodination, and further
increases with addition of a single halogen, followed by two or three halogen or methyl
substituents. Systems containing two or three substituents show equal potencies.
2.4 In-cell transcription and translation assay. The cytotoxic mechanism of action for
hydroxyquinoline ligands has been previously reported to occur through inhibition of the
proteasome.[22] Recently, it was demonstrated that clioquinol induced pro-death autophagy in
leukemia and myeloma cells by disrupting the mTOR signaling pathway.[23] The mechanistic
effects of various metal complexes containing HQ ligands are diverse, and we previously
demonstrated that the ruthenium complex with clioquinol did not inhibit the proteasome at
concentrations relevant for cell death.[1a] In order to investigate the the effect of the HQ
complexes on essential biological processes, a cell-based transcription and translation assay was
performed using Dendra2 as a reporter for protein synthesis. This allowed for a real-time report
in live cells of any damage to the DNA, RNA, or the ribosome, or inhibition of any essential
components of the cellular machinery responsible for the processes of transcription and
translation.[24] Dendra2 is a photoconvertable protein; upon irradiation, Dendra2 switches from
14

ACCEPTED MANUSCRIPT
green to red fluorescence, while Dendra2 synthesized after irradiation will only show green
fluorescence. Therefore, this assay allows for the real-time observation of newly-synthesized
protein with ratiometric detection compared to previously made protein, providing an assay for
inhibition of protein synthesis that can be assessed in dose response and with kinetic information.
Five potent complexes (4, 13, 15, 19 and 20) and two associated ligands (2-methyl-5,7-
dibromo-HQ and 2-methyl-5,7-dichloro-HQ) were tested for potential effects on protein
synthesis. Rapamycin was used as a positive control, with the results shown in Figure 4A.
Inhibition of protein synthesis was observed for rapamycin with an IC50 of 6.3 µM. The free
ligand 2-methyl-5,7-dichloro-HQ was more potent, with an IC50 of 1.6 µM. Notably, this value
corresponds closely to the cytotoxicity of compound in the HL60 cell line (IC50 = 0.55 µM,
Figure S5). The 2-methyl-5,7-dibromo-HQ was slightly less potent at 3.34 µM (Figure S4).
Complex 20, [Ru(dmphen)2-2-Me-5,7-diCl-HQ], had the same effect, but it occurred at lower
doses and with a steeper dose response, where 0.54 µM was required for 50% inhibition of
translation and 1 µM led to an 80% reduction in Dendra2 production (Figure 4A, S6). The most
potent inhibition in the Dendra2 assay was observed for compound 4 (IC50 = 0.29 µM), while
compounds 13 (IC50 = 1.0 µM), 15 (IC50 = 0.46 µM), and 19 (IC50 = 0.92 µM) were 1.5–3-times
less potent (Figure 4B). No degradation was seen for the photoconverted Dendra2 over the
course of the assay, indicating the compounds did not affect degradation of existing proteins.
15

ACCEPTED MANUSCRIPT
Figure 4. Inhibition of protein synthesis dose responses of ruthenium complexes in Dendra2
assay: (A) activity of Ru(II) complex 20, compared with parent ligand (2-methyl-5,7-dichloro-8-
hydroxyquinoline) and rapamycin; (B) activity of Ru(II) complexes 4, 13, 15 and 19.
In general, the tested complexes exhibited effective inhibition of Dendra2 at IC50 values <
1 µM; however, these concentrations are 2–14 times higher than required for cytotoxicity. These
data suggest that inhibition of translation is likely involved but may not be the exclusive
mechanism that induces the cytotoxicity of Ru(II) HQ complexes.
3. CONCLUSIONS
Quinoline and hydroxylquinoline are considered privileged structures, as these
heterocycles are found in a wide range of naturally occurring and synthetic biologically active
molecules that interact with diverse targets, inducing functional changes of importance in a
variety of disease states. These features suggest a variety of possible mechanisms of action and
biological interaction partners, leading to complex and inconsistent structure activity
relationships, depending on both the functional assay and biological test system chosen. Further
complicating the situation, many hydroxyquinolines under investigation coordinate various
16

ACCEPTED MANUSCRIPT
metals, acting as ionophores to increase cellular uptake, but they can also transport the metals to
difference subcellular compartments or form semi-stable metal complexes that could participate
in redox reactions. Alternatively, the transient metal complexes could directly bind and regulate
the activity of important biomolecules. Stable metal complexes, in contrast, present a simpler
case, as metal transport properties are eliminated, and in most cases, redox cycling or covalent
adduct formation is not possible. This leaves direct, but non-covalent, interactions with
biological targets as the most likely source for the observed activity.
This detailed SAR study for 22 cytotoxic ruthenium complexes containing mono-, di- and
tri-substituted hydroxyquinoline ligands demonstrated complexes that are highly potent. Nearly
all of these complexes were found to possess activity at submicromolar concentrations, with IC50
values ranging from 58–96 nM in the HL60 cell line. Incorporation of a halogen at the 2-, 5-, or
7-position is associated with improvement of the activity, though the greatest impact was seen at
the 2- and 7-positions (with 3–5-fold increases in potency). Placement of a methyl group at the
2-position resulted in a complex with the same potency as the unsubstituted HQ complex,
suggesting that the halogen plays an electronic role rather than exerting some steric influence, as
the van der Waals radius of –CH3 (2.00 Å) is essentially the same as that of –Br (1.95 Å).
However, addition of the large and asymmetric o-tolyl group at the 5- and 7- positions resulted in
up to a 10-fold loss of activity. The 2-position appears to be the only site for incorporation of
larger groups without loss of activity.
What is most striking from the SAR analysis is how distinct the activity profile is from
both free HQ ligands and those contained in organometallic complexes (in contrast to the
coordination complexes discussed here). Substituents at the 5-position are very commonly found
in biologically active HQ free ligands, particularly those that act as neuroprotective or anticancer
17

ACCEPTED MANUSCRIPT
agents through metal coordination;[25] however, this study demonstrates that the addition of
oxygen rich substituents such as a nitro group or sulfonic acid is exceedingly detrimental to the
activity of the HQ Ru(II) coordination complex. Arylation at this position was also disfavored,
and the serendipitous synthesis of the 5-5-linked dimer 22 suggests that loss of activity may also
scale with the size of the substituent, as the dimer was 15-fold less potent than the complex
containing the o-tolyl group. These results combine to suggest that the coordination complex has
a specific binding site on a particular biological target, and that introduction of steric clash or
electron rich substituents on this face of the molecule disrupts key contacts.
There are also many reports of 7-substituted HQ systems with noteworthy biological
activity, but in this study, the addition of groups larger than a halogen at this position reduced
cytotoxicity for the Ru(II) complex. In contrast, the 2-position was found to be the best site for
arylation, and thus, we hypothesize, addition of other substituents. There are comparatively few
reports of biologically active HQ ligands with substituents at the 2-position, and none, to the best
of our knowledge, that are Ru(II) coordination complexes. This provides a relatively unexplored
region of chemical space to exploit.
Rational design of improved systems requires the identification of the biological target.
Previously we explored and eliminated the possibility of DNA binding and proteasome
inhibition, both of which had been hypothesized as mechanisms of action for other metal
complexes containing HQ ligands or the free ligands themselves. Having excluded these as
possible causes for cytotoxicity, we turned to a functional assay recently developed in our
laboratory that monitors the production of a fluorescent protein, Dendra2. This is a global assay
for transcription and translation, and reports on interference with any stage or biological
component that plays a role in these processes. The most potent Ru(II) complexes were tested,
18

ACCEPTED MANUSCRIPT
and all were found to inhibit protein production (compounds 4, 13, 15, 19, and 20; Figure S6).
The free HQ ligands that were incorporated into complexes 19 and 20 were also tested, and also
inhibited protein production. It may be important that the ligands possessed cytotoxicity at 3–4-
fold lower doses than demonstrated inhibition of translation, while the Ru(II) complexes were
cytotoxic at 2–15-fold lower concentrations than those required to observe inhibition of Dendra2.
The inhibition of protein synthesis may not be the exclusive cause for cytotoxicity of ruthenium
complexes, and additional mechanisms might be involved for their antitumor activity.
Alternatively, the discrepancy between the IC50 values may reflect the difference in the time
frame for the two experiments (72 hours for cytotoxicity, while changes in Dendra2 production
were observed at time points less than 15 hours). In either case, inhibition of translation is an
appealing mechanism for anticancer agents, due in part to the fact that cancer cells have greater
needs for ongoing protein synthesis for proliferation and cell survival that depend on specific
regulatory proteins.[26] One translation inhibitor, omacetaxine mepesuccinate, is already in
clinical use for chronic meyloid leukemia (CML).[27] Very recently, cyclometalated Ru(II)
complexes were reported that inhibit proteosynthesis;[28] this is a new mechanism of action for
metal complexes, and the systems described in that report have similar cytotoxic potencies
(albeit in other cell lines) to the compounds described here. While there many notable chemical
differences, the overall charge on the molecules (+1) and general structures are alike, as both are
octahedral complexes containing bidentate ligands. It is possible that both these classes of metal
complexes are selectively inhibiting some components of the protein synthesis machinery.
Studies are underway to further elucidate the target(s) of the HQ complexes.
19

ACCEPTED MANUSCRIPT
4. EXPERIMENTAL SECTION
4.1 Materials and Methods. The starting hydroxyquinoline ligands were obtained from
commercial sources and were used without further purification. The ligands purchased were 5-
bromo-8-hydroxyquinoline (Ark Pharm, 97%), 5-chloro-8-hydroxyquinoline (Aldrich, 95%), 7-
bromo-8-hydroxyquinoline (Aldrich, 97%), 7-chloro-8-hydroxyquinoline (Toronto Research
Chemicals), 5,7-dimethyl-8-hydroxyquinoline (Aldrich, 98%), 5,7-dibromo-8-hydroxyquinoline
(MP Biomedicals), 5,7-dichloro-8-hydroxyquinoline (Acros Organics, 99%), 5,7-diiodo-8-
hydroxyquinoline (Aldrich, 97%), 8-hydroxy-2,5,7-trimethylquinoline (Aurum Pharmatech), 5,7-
dichloro-8-hydroxy-2-methylquinoline (Aldrich, 98%). The ligands 2-bromo-HQ,[29] 2-chloro-
HQ,[30] 2-Cl-5,7-dibromo-HQ and 2-methyl-5,7-dibromo-HQ[31] were synthesized according to
the described methods with minor modification. 2-o-Tolyl-HQ was synthesized using general
Suzuki coupling procedure.[32] Complexes 1, 2, 14, and 16 were synthesized and described
previously.[1a]
All 1H NMR spectra were obtained on a Varian Mercury spectrometer (400 MHz) with chemical
shifts reported relative to the residual solvent peak of acetonitrile at δ 1.94. Electrospray
ionization mass spectra were obtained on a Varian 1200L mass spectrometer. Absorption spectra
were obtained on an Agilent Cary 60 spectrophotometer. Extinction coefficients were determined
from three independent replicates, and reported values are with 5% error. All synthesized
compounds were isolated in >95% purity, as determined by analytical HPLC. For HPLC
analysis, the ruthenium complexes were injected on an Agilent 1100 series HPLC equipped with
a model G1311 quaternary pump, G1315B UV diode array detector, and ChemStation software
version B.01.03. Chromatographic conditions were optimized on a Column Technologies Inc.
C18, 120 Å (250 mm × 4.6 mm inner diameter, 5 µM) fitted with a Phenomenex C18 (4 mm × 3
20

ACCEPTED MANUSCRIPT
mm) guard column. Injection volumes of 15 µL of 100 µM solutions of the complex were used.
The detection wavelength was 280 nm. Mobile phases were: mobile phase A, 0.1% formic acid
in dH2O; mobile phase B, 0.1% formic acid in HPLC grade acetonitrile. The mobile phase flow
rate was 1.0 mL/min. The following mobile phase gradient was used: 98−95% A (containing
2−5% B) from 0 to 5 min; 95−70% A (5−30% B) from 5 to 15 min; 70−40% A (30−60% B)
from 15 to 20 min; 40−5% A (60−95% B) from 20 to 30 min; 5−98% A (95−2% B) from 30 to
35 min; reequilibration at 98% A (2% B) from 35 to 40 min.
4.2 General synthesis of [Ru(dmphen)2L] complexes with HQ ligands:
The synthesis of metal complexes was performed following a previously described procedure.[1a]
[Ru(dmphen)2Cl2] (100 mg, 0.17 mmol) and HQ (0.19 mmol) were added to 4 mL of ethylene
glycol in a 15 mL pressure tube. The mixture was heated at 100–120 °C for 2 h while protected
from light. The purple solution was allowed to cool to room temperature and poured into 50 mL
of dH2O. Addition of a saturated aq. KPF6 solution (ca. 1 mL) produced a purple precipitate that
was collected by vacuum filtration. The purification of the solid was carried out by flash
chromatography (silica gel, loaded in MeCN). A gradient was run, and the pure complex eluted
at 0.2% KNO3, 5–10% H2O in MeCN. The product fractions were concentrated under reduced
pressure, and a saturated aq. solution of KPF6 was added, followed by extraction of the complex
into CH2Cl2. The solvent was removed under reduced pressure to give the product as a solid.
4.2.1 Compound 3. Yield: 78 mg (52%). 1H NMR (CD3CN): δ 8.35-8.39 (m, 2H), 8.28-8.33 (m,
2H), 8.03-8.09 (m, 4H), 7.73 (d, J = 8.8 Hz, 1H), 7.64 (d, J = 8.2 Hz, 1H), 7.41-7.44 (m, 3H),
7.08 (t, J = 8.0 Hz, 1H), 6.94 (d, J = 8.6 Hz, 1H), 6.64 (d, J = 7.6 Hz, 2H), 6.22 (d, J = 8.0 Hz,
1H), 2.74 (s, 3H), 2.51 (s, 3H), 2.23 (s, 3H), 1.52 (s, 3H); purity by HPLC = 97 %; ESI MS calcd
21

ACCEPTED MANUSCRIPT
for C37H29BrN5ORu [M]+ 740.06, found 742.2 [M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 480 nm
(11.5).
4.2.2 Compound 4. Yield: 57 mg (40%). 1H NMR (CD3CN): δ 8.36-8.39 (m, 2H), 8.27-8.32 (m,
2H), 8.01-8.09 (m, 4H), 7.85 (d, J = 8.8 Hz, 1H), 7.64 (d, J = 8.3 Hz, 1H), 7.40-7.46 (m, 3H),
7.07 (t, J = 8.0 Hz, 1H), 6.75 (d, J = 8.6 Hz, 1H), 6.66 (d, J = 7.8 Hz, 2H), 6.24 (d, J = 8.0 Hz,
1H), 2.75 (s, 3H), 2.44 (s, 3H), 2.23 (s, 3H), 1.57 (s, 3H); purity by HPLC = 97 %; ESI MS calcd
for C37H29ClN5ORu [M]+ 696.11, found 696.2 [M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 485 nm
(13.9).
4.2.3 Compound 5. Yield: 25 mg (17%). 1H NMR (CD3CN): 1H NMR (CD3CN): δ 8.36 (d, J =
8.2 Hz, 1H), 8.27 (d, J = 8.2 Hz, 1H), 8.22 (d, J = 8.3 Hz, 1H), 7.97-8.01 (m, 4H), 7.90 (d, J =
8.7 Hz, 1H), 7.70 (d, J = 8.7 Hz, 1H), 7.55 (d, J = 8.2 Hz, 1H), 7.46 (d, J = 8.2 Hz, 1H), 7.22 (d,
J = 8.2 Hz, 2H), 7.04 (t, J = 7.9 Hz, 1H), 6.83 (d, J = 7.5 Hz, 1H), 6.60-6.69 (m, 2H), 6.55 (d, J =
8.4 Hz, 1H), 6.39 (t, J = 7.6 Hz, 1H), 6.31 (d, J = 7.6 Hz, 1H), 5.91 (brs, 1H), 2.73 (s, 3H), 2.34
(s, 3H), 2.26 (s, 3H), 2.16 (s, 6H); purity by HPLC = 99 %; ESI MS calcd for C44H36N5ORu
[M]+ 752.2, found 752.4 [M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 485 nm (11.3).
4.2.4 Compound 6. Yield: 81 mg (54%). 1H NMR (CD3CN): δ 8.44-8.48 (m, 2H), 8.31 (d, J =
8.3 Hz, 1H), 8.19 (d, J = 8.3 Hz, 1H), 8.15 (d, J = 8.7 Hz, 1H), 7.97-8.11 (m, 4H), 7.69 (d, J =
8.4 Hz, 1H), 7.60 (d, J = 8.2 Hz, 1H), 7.36 (d, J = 8.3 Hz, 2H), 7.31 (d, J = 8.6 Hz, 1H), 6.82 (dd,
J = 8.4, 5.1 Hz, 1H), 6.70 (d, J = 5.0 Hz, 1H), 6.17 (d, J = 8.6 Hz, 1H), 2.70 (s, 3H), 2.19 (s, 3H),
1.96 (s, 3H), 1.82 (s, 3H); purity by HPLC = 98 %; C37H29BrN5ORu [M]+ 740.06, found 742.2
[M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 495 nm (10.6).
4.2.5 Compound 7. Yield: 87 mg (61%). 1H NMR (CD3CN): δ 8.43-8.47 (m, 2H), 8.30 (d, J =
8.3 Hz, 1H), 8.18 (d, J = 8.3 Hz, 1H), 8.14 (d, J = 8.8 Hz, 1H), 8.00-8.10 (m, 4H), 7.68 (d, J =
22

ACCEPTED MANUSCRIPT
8.3 Hz, 1H), 7.60 (d, J = 8.3 Hz, 1H), 7.34-7.38 (m, 2H), 7.15 (d, J = 8.7 Hz, 1H), 6.82 (dd, J =
8.6, 5.1 Hz, 1H), 6.71 (dd, J = 5.1, 1.3 Hz, 1H), 6.18 (d, J = 8.6 Hz, 1H), 2.69 (s, 3H), 2.18 (s,
3H), 1.96 (s, 3H), 1.82 (s, 3H); purity by HPLC = 99 %; ESI MS calcd for C37H29ClN5ORu [M]+
696.11, found 696.2 [M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 495 nm (11.3).
4.2.6 Compound 9. Yield: 78 mg (52%). 1H NMR (CD3CN): δ 8.45-8.48 (m, 2H), 8.29 (d, J =
8.2 Hz, 1H), 8.19 (d, J = 8.3 Hz, 1H), 8.15 (d, J = 8.7 Hz, 1H), 8.01-8.11 (m, 3H), 7.81 (d, J =
8.3 Hz, 1H), 7.68 (d, J = 8.3 Hz, 1H), 7.60 (d, J = 8.2 Hz, 1H), 7.29-7.37 (m, 3H), 7.15 (d, J =
8.7 Hz, 1H), 6.74 (dd, J = 8.3, 5.1 Hz, 1H), 6.68 (d, J = 4.9 Hz, 1H), 6.60 (d, J = 8.5 Hz, 1H),
2.65 (s, 3H), 2.19 (s, 3H), 1.92 (s, 3H), 1.83 (s, 3H); purity by HPLC = 97 %; C37H29BrN5ORu
[M]+ 740.06, found 740.1 [M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 490 nm (12.8).
4.2.7 Compound 10. Yield: 69 mg (48%). 1H NMR (CD3CN): δ 8.44-8.48 (m, 2H), 8.29 (d, J =
8.3 Hz, 1H), 8.18 (d, J = 8.3 Hz, 1H), 8.14 (d, J = 8.7 Hz, 1H), 8.00-8.11 (m, 3H), 7.81 (dd, J =
8.3, 1.3 Hz, 1H), 7.68 (d, J = 8.3 Hz, 1H), 7.61 (d, J = 8.2 Hz, 1H), 7.32-7.36 (m, 2H), 7.19 (d, J
= 8.5 Hz, 1H), 6.72 (dd, J = 8.2, 5.1 Hz, 1H), 6.67 (dd, J = 5.1, 1.3 Hz, 1H), 6.62 (d, J = 7.9 Hz,
1H), 2.67 (s, 3H), 2.52 (s, 3H), 2.18 (s, 3H), 1.83 (s, 3H); purity by HPLC = 97 %; ESI MS calcd
for C37H29ClN5ORu [M]+ 696.11, found 696.1 [M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 495 nm
(13.6).
4.2.8 Compound 12. Yield: 60 mg (42%). 1H NMR (CD3CN): δ 8.44 (d, J = 8.2 Hz, 2H), 8.24
(d, J = 8.3 Hz, 1H), 8.17 (d, J = 8.3 Hz, 1H), 8.14 (d, J = 8.8 Hz, 1H), 8.09 (d, J = 8.7 Hz, 1H),
8.00-8.02 (m, 2H), 7.86 (dd, J = 8.5, 1.2 Hz, 1H), 7.68 (d, J = 8.3 Hz, 1H), 7.58 (d, J = 8.3 Hz,
1H), 7.33 (d, J = 8.3 Hz, 1H), 7.26 (d, J = 8.3 Hz, 1H), 6.89 (s, 1H), 6.65 (dd, J = 8.5, 5.1 Hz,
1H), 6.60 (dd, J = 5.1, 1.2 Hz, 1H), 2.67 (s, 3H), 2.32 (s, 3H), 2.21 (s, 3H), 1.88 (s, 3H), 1.83 (s,
23

ACCEPTED MANUSCRIPT
3H), 1.61 (s, 3H); purity by HPLC = 97 %; ESI MS calcd for C39H34N5ORu [M]+ 690.18, found
690.2 [M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 500 nm (9.0).
4.2.9 Compound 13. Yield: 93 mg (57%). 1H NMR (CD3CN): δ 8.46-8.49 (m, 2H), 8.31 (d, J =
8.3 Hz, 1H), 8.21 (d, J = 8.3 Hz, 1H), 8.16 (d, J = 8.8 Hz, 1H), 8.11 (d, J = 8.8 Hz, 1H), 8.02-
8.06 (m, 2H), 7.99 (dd, J = 8.6, 1.2 Hz, 1H), 7.69 (d, J = 8.3 Hz, 1H), 7.60-7.62 (m, 2H), 7.37 (d,
J = 8.3 Hz, 1H), 7.34 (d, J = 8.3 Hz, 1H), 6.86 (dd, J = 8.6, 5.1 Hz, 1H), 6.75 (dd, J = 5.1, 1.2
Hz, 1H), 2.62 (s, 3H), 2.19 (s, 3H), 1.96 (s, 3H), 1.82 (s, 3H); purity by HPLC = 97 %; ESI MS
calcd for C37H28Br2N5ORu [M]+ 817.97, found 820.0 [M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 490
nm (14.0).
4.2.10 Compound 15. Yield: 63 mg (40%). 1H NMR (CD3CN): δ 8.47 (d, J = 8.3 Hz, 2H), 8.30
(d, J = 8.3 Hz, 1H), 8.20 (d, J = 8.3 Hz, 1H), 8.15 (d, J = 8.8 Hz, 1H), 8.02-8.11 (m, 4H), 7.69
(d, J = 8.3 Hz, 1H), 7.61 (d, J = 8.3 Hz, 1H), 7.45 (s, 1H), 7.36 (d, J = 8.3 Hz, 1H), 7.34 (d, J =
8.3 Hz, 1H), 6.86 (dd, J = 8.6, 5.1 Hz, 1H), 6.77 (dd, J = 5.1, 1.2 Hz, 1H), 2.62 (s, 3H), 2.19 (s,
3H), 1.92 (s, 3H), 1.82 (s, 3H); purity by HPLC = 97 %; ESI MS calcd for C37H28BrClN5ORu
[M]+ 774.02, found 776.1 [M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 490 nm (12.9).
4.2.11 Compound 17. Yield: 68 mg (38%). 1H NMR (CD3CN): δ 8.48 (d, J = 8.3 Hz, 2H), 8.28
(d, J = 8.3 Hz, 1H), 8.20 (d, J = 8.3 Hz, 1H), 8.16 (d, J = 8.8 Hz, 1H), 8.10 (d, J = 8.8 Hz, 1H),
8.02-8.05 (m, 2H), 7.90 (s, 1H), 7.85 (dd, J = 8.6, 1.1 Hz, 1H), 7.70 (d, J = 8.3 Hz, 1H), 7.60 (d,
J = 8.3 Hz, 1H), 7.35 (d, J = 8.3 Hz, 1H), 7.30 (d, J = 8.3 Hz, 1H), 6.84 (dd, J = 8.6, 5.1 Hz, 1H),
6.70 (dd, J = 5.1, 1.1 Hz, 1H), 2.57 (s, 3H), 2.23 (s, 3H), 1.88 (s, 3H), 1.81 (s, 3H); purity by
HPLC = 98 %; ESI MS calcd for C37H28I2N5ORu [M]+ 913.94, found 914.0 [M]+; UV/Vis
(CH3CN): λmax (ε × 10-3) 490 nm (15.2).
24

ACCEPTED MANUSCRIPT
4.2.12 Compound 18. Yield: 74 mg (51%). 1H NMR (CD3CN): δ 8.36-8.40 (m, 2H), 8.26 (d, J
= 8.3 Hz, 1H), 8.22 (d, J = 8.3 Hz, 1H), 7.97-8.11 (m, 4H), 7.87 (d, J = 8.7 Hz, 1H), 7.39-7.43
(m, 2H), 7.27 (d, J = 8.2 Hz, 1H), 6.81 (s, 1H), 6.85 (d, J = 8.7 Hz, 1H), 2.65 (s, 3H), 2.32 (s,
3H), 2.28 (s, 3H), 2.17 (s, 3H), 1.50 (s, 3H), 1.46 (s, 3H), 1.18 (s, 3H); purity by HPLC = 99 %;
ESI MS calcd for C40H36N5ORu [M]+ 704.2, found 704.3 [M]+; UV/Vis (CH3CN): λmax (ε × 10-3)
505 nm (11.0).
4.2.13 Compound 19. Yield: 70 mg (42%). 1H NMR (CD3CN): δ 8.40-8.42 (m, 2H), 8.27-8.30
(m, 2H), 8.00-8.12 (m, 5H), 7.70 (d, J = 8.3 Hz, 1H), 7.52 (s, 1H), 7.42-7.45 (m, 2H), 7.35 (d, J
= 8.3 Hz, 1H), 7.85 (d, J = 8.8 Hz, 1H), 2.60 (s, 3H), 2.28 (s, 3H), 2.20 (s, 3H), 1.50 (s, 3H), 1.25
(s, 3H); purity by HPLC = 96 %; ESI MS calcd for C38H30Br2N5ORu [M]+ 831.99, found 834.0
[M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 490 nm (11.2).
4.2.14 Compound 20. Yield: 70 mg (46%). 1H NMR (CD3CN): δ 8.42 (d, J = 8.3 Hz, 2H), 8.31
(d, J = 8.3 Hz, 1H), 8.28 (d, J = 8.3 Hz, 1H), 8.02-8.12 (m, 5H), 7.70 (d, J = 8.3 Hz, 1H), 7.42-
7.47 (m, 2H), 7.38 (d, J = 8.3 Hz, 1H), 7.28 (s, 1H), 7.84 (d, J = 8.8 Hz, 1H), 2.64 (s, 3H), 2.26
(s, 3H), 2.23 (s, 3H), 1.52 (s, 3H), 1.24 (s, 3H); purity by HPLC = 99 %; ESI MS calcd for
C38H30Cl2N5ORu [M]+ 744.09, found 744.2 [M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 490 nm
(12.2).
4.2.15 Compound 21. Yield: 81 mg (48%). 1H NMR (CD3CN): δ 8.42 (d, J = 8.3 Hz, 1H), 8.40
(d, J = 8.3 Hz, 1H), 8.30-8.34 (m, 2H), 8.02-8.11 (m, 5H), 7.67 (d, J = 8.3 Hz, 1H), 7.61 (s, 1H),
7.45-7.48 (m, 2H), 7.40 (d, J = 8.3 Hz, 1H), 6.94 (d, J = 9.0 Hz, 1H), 2.62 (s, 3H), 2.39 (s, 3H),
2.27 (s, 3H), 1.55 (s, 3H); purity by HPLC = 99 %; ESI MS calcd for C37H27Br2ClN5ORu [M]+
851.93, found 854.0 [M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 470 nm (12.5).
25

ACCEPTED MANUSCRIPT
4.2.16 Compound 22. [Ru(dmphen)2Cl2] (100 mg, 0.17 mmol) and 7-chloro-HQ (30.5 mg, 0.17
mmol) were added to 8 mL of ethanol-water mixture (1:1) in a 15 mL pressure tube. The mixture
was heated at 60 °C for 5 h while protected from light. The purple solution was allowed to cool
to room temperature and poured into 50 mL of dH2O. Addition of a saturated aq. KPF6 solution
(ca. 1 mL) produced a purple precipitate that was collected by vacuum filtration. The purification
of the solid was carried out by flash chromatography (silica gel, loaded in MeCN). A gradient
was run, and the pure complex eluted at 0.2% KNO3, 8–10% H2O in MeCN. The product
fractions were concentrated under reduced pressure, and a saturated aq. solution of KPF6 was
added, followed by extraction of the complex into CH2Cl2. The solvent was removed under
reduced pressure to give a purple solid. Yield: 64 mg (45%). 1H NMR (CD3CN): δ 8.43-8.48 (m,
2H), 8.30-8.32 (m, 1H), 7.98-8.19 (m, 5H), 7.67 (dd, J = 8.3, 2.5 Hz, 1H), 7.61 (d, J = 8.9 Hz,
1H), 7.34-7.39 (m, 2H), 6.96-7.25 (m, 2H), 6.63-6.68 (m, 1H), 6.48-6.59 (m, 1H), 2.69-2.70 (m,
3H), 2.16 (s, 3H), 1.93 (s, 3H), 1.82 (s, 3H); purity by HPLC = 96 %; ESI MS calcd for
C74H56Cl2N10O2Ru2 [M]2+ 695.11, found 695.1 [M]2+; UV/Vis (CH3CN): λmax (ε × 10-3) 495 nm
(12).
4.4 Compounds 8 and 11.
The synthesis was performed using general Suzuki coupling procedure[32]. Complex 6 or 9 (34
µmol), o-tolylboronic acid (70 mg, 51 µmol), [Pd(PPh3)4] (3.4 mg, 0.34 µmol) and K2CO3 (14
mg, 102 µmol) were added to a flask under argon. Methanol (3 mL; degassed) was added to the
reaction mixture via cannula. The resulting mixture was refluxed with stirring for 48 h, followed
by removal of the solvent under reduced pressure to give a purple solid. Purification was carried
out by flash chromatography (silica gel, loaded in MeCN, followed by a gradient); the pure
complex eluted at 0.2% KNO3, 1% H2O in MeCN. The product fractions were combined and
26

ACCEPTED MANUSCRIPT
concentrated under reduced pressure. A saturated aq. solution of KPF6 was added, and the
complex was extracted into CH2Cl2, followed by removal of the solvent under reduced pressure
to give a purple solid.
4.4.1 Compound 8. Yield: 16 mg (53%). 1H NMR (CD3CN): 1H NMR (CD3CN): δ 8.39-8.46
(m, 2H), 8.29 (d, J = 8.3 Hz, 1H), 7.99-8.21 (m, 5H), 7.70 (d, J = 8.2 Hz, 1H), 7.58 (d, J = 8.3
Hz, 1H), 7.35-7.42 (m, 2H), 7.17-7.29 (m, 5H), 6.96-7.05 (m, 2H), 6.59-6.66 (m, 1H), 6.54 (brs,
1H), 2.80 (s, 3H), 2.23 (s, 3H), 2.05 (s, 1.2H), 1.98 (s, 1.8H), 1.89 (s, 1.8 H), 1.82 (s, 3 H), 1.78
(s, 1.2 H); purity by HPLC = 97 %; ESI MS calcd for C44H36N5ORu [M]+ 752.2, found 752.3
[M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 495 nm (7.6).
4.4.2 Compound 11. Yield: 13 mg (43%). 1H NMR (CD3CN): δ 8.44 (d, J = 8.2 Hz, 1H), 8.33-
8.37 (m, 2H), 8.10-8.18 (m, 2H), 8.04 (d, J = 8.8 Hz, 1H), 8.01 (d, J = 8.8 Hz, 1H), 7.80 (d, J =
8.3 Hz, 1H), 7.68 (d, J = 8.3 Hz, 1H), 7.48 (d, J = 8.3 Hz, 1H), 7.40 (d, J = 8.2 Hz, 1H), 7.32 (d,
J = 8.3 Hz, 1H), 6.82-7.02 (m, 5H), 6.65-6.73 (m, 3H), 6.28 (brs, 1H), 2.48 (s, 3H), 2.15 (s, 6H),
2.02 (s, 3H), 1.75 (s, 3H); purity by HPLC = 98 %; ESI MS calcd for C44H36N5ORu [M]+ 752.2,
found 752.3 [M]+; UV/Vis (CH3CN): λmax (ε × 10-3) 490 nm (6.8).
4.5 Counter ion exchange
Compounds 1–22 were converted to Cl- salts by dissolving 5–10 mg of product in 1–2 mL
methanol. The dissolved product was loaded onto an Amberlite IRA-410 chloride ion exchange
column, eluted with methanol, and the solvent removed under reduced pressure.
4.6 Cytotoxicity Assay
HL60 cells were plated at 30,000 cell per well in optiMEM (supplemented with 2% FBS, 50
U/ml Penicillin and 50 mg/ml Streptomycin) in 96 well plates. Compounds were serially diluted
27

ACCEPTED MANUSCRIPT
in optiMEM in a 96 well plate and then added to the cells. The cells were incubated with the
compounds for 72 h followed by the addition of resazurin. The plates were incubated for 3 h and
then read on a SpectraFluor Plus plate reader with an excitation filter of 535 nm and emission of
595 nm.
4.7 Dendra 2 Transcription-Translation Assay. 96 well plates were coated with matrigel
followed by the addition of HEK T-Rex cells at a density of 30,000 cells/well and incubated with
1 µg/mL of tetracycline for 16 hours. Media was removed and 50 µL of L-15 media containing 1
µg/mL tetracycline along with compound was added to each well and allowed to incubate for 1
hr. Plates were then illuminated with a 405 nm LED flood array for one 1 min and then read in
kinetic mode on a SpectraFluor Plus (Tecan) set to 37 C. The plates were read every 30 min for
15 hrs with excitation and emission wavelengths of 480 nm and 530 nm for newly translated
Dendra2 and 535 nm and 595 nm for post-translated Dendra2.
4.8 Crystallography
Single crystals of compounds 2, 9 and 14 were grown from methylene chloride or acetone by
vapor diffusion of diethyl ether. They were mounted in inert oil and transferred to the cold gas
stream of the diffractometer. X-ray diffraction data were collected at 90.0(2) K on either a
Nonius kappaCCD diffractometer using MoKα X-rays or on a Bruker-Nonius X8 Proteum
diffractometer with graded-multilayer focused CuKα X-rays. Raw data were integrated, scaled,
merged and corrected for Lorentz-polarization effects using either the HKL-SMN package[33] or
the APEX2 package.[34] Corrections for absorption were applied using SADABS[35] and
XABS2.[36] The structures were solved by SHELXT,[37] and refined against F2 by weighted full-
matrix least-squares using SHELXL-2014.[38] For compound 8 the SQUEEZE routine[39] was
used to treat disordered solvent. Hydrogen atoms were placed at calculated positions and refined
28

ACCEPTED MANUSCRIPT
using a riding model. Non-hydrogen atoms were refined with anisotropic displacement
parameters. Structures were checked using check CIF tools in Platon[40] and by an R-tensor.[41]
Crystal data and relevant details of the structure determinations are summarized below and
selected geometrical parameters are given in Table 1.
4.8.1 Crystal data (2): C39H34Cl2F6N5OPRu, Mr = 905.65, Monoclinic, P21/c, a = 16.2343(3)
Å, b = 13.0460(3) Å, c = 17.6161(3) Å, α = 90.º, β = 105.283(1)º, γ = 90º, V = 4180.29(12) Å3, Z
= 4, ρ = 1.671 mg m-3, µ = 5.951 mm-1, F(000) = 1832, crystal size = 0.100×0.080×0.020 mm,
θ(max) = 68.210º, 46939 reflections collected, 6555 unique reflections (Rint = 0.0397), GOF =
1.079, R1 = 0.0302 and wR2 = 0.081 [I > 2σ(I)], R1 = 0.0316 and wR2 = 0.0826 (all indices),
largest difference peak/hole = 0.726 / -0.681 eÅ-3.
4.8.2 Crystal data (9): C38H31BrCl2F6N5OPRu, Mr = 970.53, Monoclinic, P2(1)/n, a =
15.1996(4) Å, b = 12.8658(3) Å, c = 19.7650(5) Å, α = 90.º, β = 109.65(1)º, γ = 90º, V =
3640.05(16) Å3, Z = 4, ρ = 1.771 mg m-3, µ = 7.171 mm-1, F(000) = 1936, crystal size =
0.290×0.180×0.070 mm, θ(max) = 68.231º, 47643 reflections collected, 6626 unique reflections
(Rint = 0.0506), GOF = 1.087, R1 = 0.0322 and wR2 = 0.0879 [I > 2σ(I)], R1 = 0.0337 and wR2 =
0.0891 (all indices), largest difference peak/hole = 0.689 / -0.611 eÅ-3.
4.8.3 Crystal data (14): C43H41Cl2F6N5O2PRu, Mr = 976.75, Monoclinic, I2/a, a = 23.4088(6)
Å, b = 10.0546(2) Å, c = 34.4516(12) Å, α = 90.º, β = 90.859(1)º, γ = 90º, V = 8107.8(4) Å3, Z =
8, ρ = 1.60 mg m-3, µ = 5.349 mm-1, F(000) = 3976, crystal size = 0.090×0.080×0.045 mm,
θ(max) = 68.321º, 54029 reflections collected, 7380 unique reflections (Rint = 0.0379), GOF =
1.028, R1 = 0.0262 and wR2 = 0.0706 [I > 2σ(I)], R1 = 0.0269 and wR2 = 0.0711 (all indices),
largest difference peak/hole = 0.544 / -0.518 eÅ-3.
ASSOCIATED CONTENT
29