Induction of the Endoplasmic Reticulum Stress Pathway by Highly Cytotoxic Organoruthenium Schiff-Base Complexes

Article abstract of DOI:10.1021/acs.molpharmaceut.8b00003

Current anticancer drug discovery efforts focus on the identification of first-in-class compounds with a mode-of-action distinct from conventional DNA-targeting agents for chemotherapy. An emerging trend is the identification of endoplasmic reticulum (ER) targeting compounds that induce ER stress in cancer cells, leading to cell death. However, a limited pool of such compounds has been identified to date, and there are limited studies done on such compounds to allow for the rational design of ER stress-inducing agents. In our present study, we present a series of highly cytotoxic, ER stress-inducing Ru(II)-arene Schiff-Base (RAS) complexes, bearing iminoquinoline chelate ligands. We demonstrate that by structural modification to the iminoquinoline ligand, we could tune its -acidity and influence reactive oxygen species (ROS) induction, switching between a ROS-mediated ER stress pathway activation and one that is not mediated by ROS induction. Our current study adds to the available ER stress inducers and shows how structural tuning could be used as a means to modulate the mode-of-action of such compounds.

Full text of DOI:10.1021/acs.molpharmaceut.8b00003

Subscriber access provided by Technical University of Munich University Library
Article
Induction of the Endoplasmic Reticulum Stress Pathway by
Highly Cytotoxic Organoruthenium Schiff-Base Complexes
Mun Juinn Chow, Maria Babak, Kwan Wei Tan, Mei Chi
Cheong, Giorgia Pastorin, Christian Gaiddon, and Wee Han Ang
Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.8b00003 • Publication Date (Web): 06 Jul 2018
Downloaded from http://pubs.acs.org on July 7, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 34
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
Induction of the Endoplasmic Reticulum Stress Pathway by Highly
Cytotoxic Organoruthenium Schiff-Base Complexes
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Mun Juinn Chow,*,1 Maria V. Babak,2 Kwan Wei Tan,3 Mei Chi Cheong,3 Giorgia Pastorin2,4,
Christian Gaiddon5, 6 and Wee Han Ang*,3, 4
1Cancer Science Institute of Singapore, National University of Singapore, 14 Medical Drive, Centre for
Translational Medicine, Singapore 117599
2Department of Pharmacy, National University of Singapore, 3 Science Drive 3, 117543 Singapore
3Department of Chemistry, National University of Singapore, 3 Science Drive 3, 117543 Singapore
4NUS Graduate School for Integrative Sciences and Engineering, Singapore
5U1113 INSERM, 3 Avenue Molière, Strasbourg 67200, France
6Oncology section, FMTS, Université de Strasbourg, Strasbourg, France
KEYWORDS
Ruthenium Arene Schiff-Base Complexes, Anticancer, p53-independent activity, Reactive
Oxygen Species, Endoplasmic Reticulum Stress.
1
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 2 of 34
1
2
3
4
ABSTRACT
5
6
7
8
9
Current anticancer drug discovery efforts focus on the identification of first-in-class compounds
with mode-of-action distinct from conventional DNA-targeting agents for chemotherapy. An
emerging trend is the identification of endoplasmic reticulum (ER) targeting compounds that
induce ER stress in cancer cells, leading to cell death. However, a limited pool of such compounds
has been identified to date and there are limited studies done on such compounds to allow for the
rational design of ER stress inducing agents. In our present study, we present a series of highly
cytotoxic, ER stress-inducing Ru (II)-arene Schiff-Base (RAS) complexes, bearing iminoquinoline
chelate ligands. We demonstrate that by structural modification to the iminoquinoline ligand, we
could tune its π-acidity and influence reactive oxygen species (ROS) induction, switching between
a ROS-mediated ER stress pathway activation and one that is not mediated by ROS induction. Our
current study adds to the available ER stress inducers and shows how structural tuning could be
used as a means to modulate the mode-of-action of such compounds.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Page 3 of 34
Molecular Pharmaceutics
1
2
3
4
INTRODUCTION
5
6
7
8
9
Conventional anti-proliferative agents target nuclear DNA leading to DNA damage that trigger
concomitant activation of p53 and downstream apoptosis factors to signal for apoptotic cell death.
However, phenomenon such as p53 mutation and apoptosis resistance in many cancer types has
led to the decreased effectiveness of such therapeutic agents.1-4 In cancers such as melanoma,
glioblastoma and non-small cell lung cancer (NSCLC), dysregulation of the apoptosis pathway is
the key mechanism responsible for resistance to chemotherapeutic treatments.5-7 As such, p53
mutation and BAX/Bcl-2 expression serve as prognostic biomarkers for several cancer types, and
p53 mutation and dysregulation of apoptotic factors such as BAX/Bcl-2 are often associated with
poorer clinical outcomes from conventional chemotherapy.8-10 Consequently, current drug
discovery efforts focus on divergent strategies such as the identification and development of
anticancer agents that target other cell organelles and act via atypical mode-of-action to circumvent
the multidrug resistance phenomenon brought about by defects in the apoptotic pathway.11-13
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
One emerging trend is the identification of drug candidates that target the endoplasmic reticulum
(ER) with the goal of inducing ER stress and leading to eventual cell death.14-16 FDA-approved
drugs such as Nelfinavir, Sorafenib and gold-based Auranofin are ER stress-inducing agents
currently investigated and repurposed for cancer treatment.17-21 Other novel metal-based ER stress
inducers are also being investigated as anticancer drugs (Figure 1).22-25 The recent surge in interest
could be due to the implication of reactive oxygen species (ROS)-mediated ER stress in several
novel strategies for cancer treatment. For instance, ROS-mediated ER stress is required to trigger
Type-II immunogenic cell death, a concurrent cancer cell killing and activation of the immune
system, which has been shown to significantly reduce cancer recurrence in in vivo models.26-27 ER
stress is also implicated in apoptosis-independent programmed cell death such as autophagy,
3
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 4 of 34
1
2
3
4
5
6
7
8
9
paraptosis and necroptosis, which could be exploited in the treatment of apoptosis-resistant
cancers.28-31 Nevertheless, the discovery of such compounds are rarely reported and there have
been few structural-tuning studies performed on ER stress-inducing compounds that allow for a
targeted design approach.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Cytotoxic metal complexes that induce ER stress in cancer.
We earlier identified lead anticancer complex RAS-1T from a class of Ru-arene Schiff-base
(RAS) complexes (Figure 1).32-34 1 (RAS-1T) is able to induce non-apoptotic cell death in
colorectal and gastric cancers through ROS-mediated ER stress. In addition, we also showed that
minor structural modification to the facially-bound arene ligand on 1 could significantly affect its
mode-of-action, switching from a ROS-dependent to a ROS-independent ER stress pathway.34
Building on previous studies, we further explored how structural modification could affect the
mode-of-action of this class of complexes by incorporating chelating iminoquinoline ligands of
varying degrees of π-acidity. Our study showed that these 2nd generation RAS complexes had
nanomolar cytotoxicity in a panel of drug-sensitive and -resistant ovarian, gastric and colorectal
cancer cell lines. In addition, we demonstrate that by modifying the structural moiety on the
4
ACS Paragon Plus Environment

Page 5 of 34
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
iminoquinoline ligand, we could tune its π-acidity and influence ROS induction, switching
between a ROS-mediated ER stress pathway activation and one that is not mediated by ROS
induction. Given the potential application of ER stress-inducing anticancer compounds for the
treatment of resistant cancer types, our study adds to the limited pool of such compounds and
provide additional information on how structural modification could be used to modulate their
mode-of-action.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL
Materials. All experimental procedures were carried out without additional precautions to exclude
air or moisture unless otherwise specified. All chemicals and solvents were used as received unless
otherwise specified. Dry ethanol and methanol were obtained by drying in molecular sieves 3-4 Å
24 h before use. RuCl3.xH2O was purchased from Precious Metals Online. [(η6-1,3,5-
triisopropylbenzene)RuCl2]2 were synthesized according to previously reported protocols.32
Thiazolyl blue tetrazolium bromide (MTT), IGEPAL CA-630, DL-Dithiothreitol,
Tetramethylethylenediamine (TEMED), Sodium Deoxycholate, N-acetylcysteine, Non-fat Dried
Milk Bovine, Bovine Serum Albumin, TWEEN® 20, Ponceau S and Thioredoxin Reductase
Assay Kit were purchased from Sigma-Aldrich. Nitric acid (65% to 71%, TraceSELECT Ultra)
was obtained from Fluka (Sigma Aldrich). Tris was purchased from Vivantis Technologies. 10%
SDS solution were purchased from Life Technologies. Glycine, HycloneTM Trypsin Protease 2.5%
(10X) solution, RPMI 1640, DMEM medium, Fetal bovine serum (FBS), Hank’s Balanced Salt
Solution (HBSS) and PierceTM Protease and Phosphatase Inhibitor Mini Tablets were purchased
from Thermo Fisher Scientific. HycloneTM Dulbecco's Phosphate-Buffered Saline (10x) and
5
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 6 of 34
1
2
3
4
5
6
7
8
9
Penicillin-Streptomycin (10 000 U/mL) were purchased from Ge Healthcare Life Sciences. Bio-
rad Protein Assay Dye Reagent Concentrate, 30% Acrylamide/Bis solution, 4x Laemmli Sample
Buffer, Nitrocellulose Membrane, 0.2 µm and 0.45 µm were purchased from Bio-rad Laboratories.
LuminataTM Classico and Crescendo Western HRP Substrate were purchased from Merck
Millipore Corporation. All other chemicals used were purchased from Sigma-Aldrich (Singapore).
All new compounds synthesized were shown to be > 95% pure either by RP-HPLC or elemental
analysis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Instrumentation. 1H NMR spectrums were obtained using a Bruker Avance 500 spectrometer and
the chemical shifts (δ) were reported in parts per million with reference to residual solvent peaks.
Electrospray-ionization Mass Spectrometry (ESI-MS) spectra were obtained using Thermo
Finnigan MAT ESI-MS System. UV-vis spectra were obtained using the Shimadzu UV-1800 UV
Spectrophotometer. Cellular Ru content was determined by Agilent 7700 Series ICP-MS (Agilent
Technologies, Santa Clara, CA, USA). Ru drug stock concentrations were determined using
Optima ICP-OES (Perkin-Elmer) operated by CMMAC, NUS. Elemental analyses of selected Ru
complexes were carried out using a Perkin-Elmer PE 2400 elemental analyzer by CMMAS, NUS.
Absorbance on 96-well plates were measured using BioTek® Synergy H1 Hybrid Reader. Western
blot proteins bands were visualized via enhanced chemiluminescence imaging (PXi, Syngene).
Flowcytometry experiments were done with BD LSRFortessa Cell Analyzer. Ultrapure water was
purified by a Milli-Q UV purification system (Sartorius Stedim Biotech SA).
HPLC analysis of compound purity. Determination of the purity of 2 and 3 were done using
analytical HPLC on a Shimadzu Prominence System equipped with a DGU-20A3 Degasser, two
LC-20AD Liquid Chromatography Pump, a SPD-20A UV/Vis Detector and a Shim Pack GVP-
ODS 2.0 mm C18 column (5 µM, 120Å, 250 mm x 4.60 mm i.d.) at r.t. at a flow rate of 1.0 mL/min
6
ACS Paragon Plus Environment

Page 7 of 34
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
with detection at both 214 nm and 254 nm. The gradient elution conditions were as follows: 20-
80% solvent B over 30 min, where solvent A is 10 mM aqueous NH4OAc pH 7.0 and solvent B is
CH3CN.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of RAS-1T (1). Complex 1 was synthesized according to published procedure.32
Synthesis of 2, 3 and 4. A similar protocol was used for the syntheis of 2, 3 and 4. 2-
Quinolinecarboxaldehyde (59.0 mg, 0.375 mmol), and 4-ethylaniline (46.5 µl, 0.375 mmol) or
N,N-dimethyl-p-phenylenediamine (51.1 mg, 0.375 mmol) or 4-isopropylaniline (51.3 µl, 0.375
mmol) was added to dry EtOH (5 mL) and and stirred at r.t. over 24 h. The solvent was removed
in vacuo and the resultant crude products were redissolved in MeOH (15 mL). Subsequently, [(η6-
1,3,5-triisopropylbenzene)RuCl2]2 (127 mg, 0.169 mmol) was added and stirred at r.t. over 12 h.
The solvent was removed in vacuo and the resulting solid purified by column chromatography (1:4
v/v EtOH/CHCl3, Rf = 0.5-0.6). The final products were then dried in vacuo for 1 h to give a final
product.
2: Red-brown solid. Yield: 66 mg (31%). 1H NMR (500 MHz, DMSO-d6): δ 9.13 (s, 1H), 8.95 (d,
J = 9 Hz, 1H), 8.87 (d, J = 9 Hz, 1H), 8.3 (m, 2H), 8.13 (t, J = 8 Hz, 1H), 8.04 (d, J = 8 Hz, 2H),
7.98 (t, J = 8 Hz, 1H), 7.48 (d, J = 8 Hz, 2H), 5.57 (s, 3H), 2.75 (q, J = 8 Hz, 2H), 2.41 (sept, J = 7
Hz, 3H), 1.26 (t, J = 8 Hz, 3H), 1.15 (d, J = 7 Hz, 9H), 0.84 (d, J = 7 Hz, 9H) ppm. ESI-MS (+ve
mode): m/z = 601 [M]+. Purity of the complex was determined to be >95% pure by RP-HPLC and
elemental analysis. RP-HPLC (% Purity): 96.5% at 214 nm and 95.8% at 254 nm; tr = 30.4 min.
Analysis (Calcd., found for C33H40N2Cl2Ru.3.5H2O): C (56.65, 56.75), H (6.77, 6.66), N (4.00,
4.06).
7
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 8 of 34
1
2
3
4
5
6
7
8
9
3: Dark purple solid. Yield: 134 mg (62%). 1H NMR (500 MHz, DMSO-d6): δ 9.00 (s, 1H), 8.95
(d, J = 9 Hz, 1H), 8.77 (d, J = 9 Hz, 1H), 8.19 (m, 2H), 8.05 (m, 2H), 7.92 (t, J = 8 Hz, 1H), 6.87
(d, J = 9 Hz, 2H), 5.49 (s, 3H), 3.09 (s, 6H), 2.50 (m, 3H), 1.17 (d, J = 7 Hz, 9H), 0.86 (d, J = 7
Hz, 9H) ppm. ESI-MS (+ve mode): m/z = 616 [M]+. Purity of the complex was determined to be
>95% pure by RP-HPLC (% Purity): 97.4% at 214 nm and 97.0% at 254 nm; tr = 28.5 min.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4: Red-brown solid. Yield: 108 mg (49%). 1H NMR (500 MHz, DMSO-d6): δ 9.14 (s, 1H), 8.94
(d, J = 9 Hz, 1H), 8.87 (d, J = 9 Hz, 1H), 8.29 (m, 2H), 8.14 (t, J = 7 Hz, 1H), 8.03 (d, J = 9 Hz,
2H), 7.99 (t, J = 8 Hz, 1H), 7.51 (d, J = 8 Hz, 2H), 5.58 (s, 3H), 3.05 (sept, J = 7 Hz, 1H), 2.39
(sept, J = 7 Hz, 3H), 1.29 (dd, J = 7 Hz, 6H), 1.15 (d, J = 7 Hz, 9H), 0.85 (d, J = 7 Hz, 9H) ppm.
ESI-MS (+ve mode): m/z = 615 [M]+. Purity of the complex was determined to be >95% pure by
elemental analysis. Analysis (Calcd., found for C34H42N2Cl2Ru.2.5H2O): C (58.70, 58.55), H
(6.81, 6.67), N (4.03, 4.12).
UV-vis analysis of compound stability. 1 – 4 were dissolved at a final concentration of 50 µM in
1.5 ml of ddH2O, DMSO, aqueous NAC (2 mM) or DMEM containing 10% FBS (without phenol
red). The UV-vis profiles of the samples were monitor by UV-vis over 24 h at 1h-intervals.
Determination of Log P. Log Pow of 2, 3 and 4 were determined using the shake flask method.35
The RAS complex were dissolved in ddH2O that was presaturated with n-octanol (for 24 h and left
to stand until phase separation occurs). The UV-vis spectrum for each samples was obtained and
the absorbances at the λmax of each compound were determined. Equal volume of n-octanol was
added to each sample solution and the heterogeneous mixtures shaken for 2 h before centrifuging
at 4000 rpm for 1 min to achieve phase separation. The final absorbance of the aqueous phase at
8
ACS Paragon Plus Environment

Page 9 of 34
Molecular Pharmaceutics
1
2
3
4
the λmax of each compound were determined and their water-octanol partition coefficient were
5
6
calculated. All experiments were done in triplicate.
7
8
9
Tissue culture. The human colorectal carcinoma cells HCT116 and HCT116 p53-/- pair were gifts
from Professor Shen Han-ming (NUS). Human gastric adenocarcinoma AGS and Human
colorectal adenocarcinoma HT29, were acquired from ATCC® (Manassa, VA). Human ovarian
carcinoma cells A2780 and A2780cisR were obtained as gifts from Professor Paul Dyson (EPFL).
TC7 cells were cloned from parental colorectal adenocarcinoma Caco-2 cells by the limited
dilution technique.36 HT29, HCT116 and HCT116 p53-/- cells were cultured in DMEM medium
containing 10% FBS and 1% Penicillin/Streptomycin. A2780, A2780cisR and AGS cells were
cultured in RPMI 1640 medium containing 10% FBS and 1% Penicillin/Streptomycin. TC7 cells
were cultured in DMEM medium containing 20% FBS, 1% Penicillin/Streptomycin and 1%
NEAA. All cell lines were grown at 37 °C in a humidified atmosphere of 95% air and 5% CO2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experiments were performed on cells within 20 passages.
Inhibition of cell viability assay. The anti-proliferative activity the RAS Complexes on
exponentially growing cancer cells were determined using MTT assay as described previously.37
HT29, AGS, HCT116, HCT116 p53-/- and TC7 were seeded at 5 000 cells per well (100 µL), and
A2780 and A2780cisR were seeded at 6000 cells and 10 000 cells per well (100 µL) respectively
in Corning® Costar® 96-well plates and incubated for 24 h. Thereafter, cancer cells were exposed
to drugs at different concentration in media for 48 h. The final concentration of DMSO in medium
was < 1% (v/v) at which cell viability was not significantly inhibited. The medium was removed
and replaced with MTT solution (100 µL, 0.5 mg/mL) in media and incubated for an additional 45
min. Subsequently, the medium was aspirated, and the purple formazan crystals dissolved in
DMSO (100 µL). The absorbance due to the dissolved purple formazan was then obtained at 570
9
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 10 of 34
1
2
3
4
5
6
7
8
9
nm. Inhibition to cell viability was evaluated with reference to the IC50 value, which is defined as
the concentration needed for a 50% reduction of survival based on the survival curves. IC50 values
were calculated from the dose - response curves (cell viability vs drug concentration) obtained in
repeated experiments and adjusted to actual [Ru] administered, which was determined using ICP-
OES. The experiments were performed in 3 replicates for each drug concentration and were carried
out at least three times independently.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For cell viability assays involving ROS quencher, NAC (2 mM) was added at the same time and
co-incubated with drugs for the entire 48 h duration. Cell viability in the absence and presence of
NAC was normalized against untreated control. Experiments were performed in 3 replicates and
carried out at least three times independently.
ROS detection. HCT116 cells were harvested by trypsinization and 1 mL of cell solution (2 × 105
cells per ml) was centrifuged (5 min, 2500 rpm) and washed with 1 mL of HBSS and centrifuged
again (5 min, 2500 rpm). The supernatant was replaced with 20 µM of 2′,7′-
dichlorodihydrofluorescein diacetate (H2DCFDA) in HBSS and the cells were incubated for 10
min at 37 °C in the absence of light for probe activation. The cells were then centrifuged (5 min,
2500 rpm) and the supernatant was replaced with the drug solutions in colorless DMEM without
FBS at desired concentrations. The cells were then incubated with drug solutions for 4 h at 37 °C
in the absence of light. After 4 h, the cell solutions were immediately strained with a 60 µM cell
strainer prior to analysis by flow cytometry. 0.46 g/L propidium iodide (PI) was added to the
strained samples to identify the dead cells. Trolox (100 µM) was used as ROS scavenger in the
control sample. The data was processed and exported using BD FACSDiva 6.2 and the quantity of
ROS species was normalized to untreated control stained with H2DCFDA probe only. Evaluation
was based on the mean of at least three independent experiments.
10
ACS Paragon Plus Environment

Page 11 of 34
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
Cellular uptake of Ru. HCT116 cells were seeded into Cellstar 6-well plates (Greiner Bio-one)
at a density of 400 000 cells/well. After the cells were allowed to resume exponential growth for
24 h, they were exposed to 1 – 4 at their respective IC50 values for 24 h. The cells were washed
twice with 1 ml of PBS and lysed with lysis buffer [100 µL, 1% IGEPAL CA-630, 150 mM NaCl,
50 mM Tris-HCl (pH 8.0)] for 5–10 min at 4 °C. The cell lysates were scraped from the wells and
transferred to separate 1.5 mL microtubes. The supernatant was then collected after centrifugation
(13 000 rpm, 4 °C for 15 min) and total protein content of each sample was quantified via
Bradford’s assay. Cell lysates were transferred to 2 ml glass vials and then digested with ultrapure
65% HNO3 at 100˚C for 24 h. The resulting solution was diluted to 2-4% v/v HNO3 with ddH2O
water. Ru content of each sample was quantified by ICP-MS using Re as an internal standard. Ru
and Re were measured at m/z 101 and m/z 186, respectively. Metal standards for calibration curve
(0, 0.5, 1, 2, 5, 10, 20, 40 ppb) were freshly prepared before each measurement. Ru and Re
standards for ICP-MS measurements were obtained from CPI international (Amsterdam, The
Netherlands). All readings were made in triplicates in He mode.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Thioredoxin Reductase Inhibition Assay. Assays were performed according to the protocol
provided by manufacturer. Briefly, 180 µl of working buffer [100 mM pH 7.0 potassium
phosphate, 10 mM EDTA and 0.24 mM NADPH], 8 µl of 1x assay buffer [100 mM pH 7.0
potassium phosphate and 10 mM EDTA], 2 µl of Enzyme solution (10 ng) was added to each well
of a Corning® clear 96-well plate. 4 µl of complexes 1 – 4 was added to the appropriate wells to
give a final concentration of 1x or 2x [IC50]. The reaction mixture was incubated with gentle
shaking at room temperature for 30 min. Subsequently, 6 µl of DTNB (100 mM) was added and
incubated for an additional 3 mins. Thereafter, the absorbance at 412 nm were measured every
minute for the next 30 min. A separate set of positive control experiments were performed using
11
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 12 of 34
1
2
3
4
5
6
the inhibitor provided with the kit. % enzyme activity was calculated from the data obtained with
reference to control sample without inhibitor.
7
8
9
Antibodies and Western blot protocol. HCT116 cells were grown at 500 000 cells per well (2
mL) on Corning® Costar® 6-well plates for 24 h before being treated with 1 - 4 at IC50 or IC75 for
6 h and 24 h. OXP was used as a positive control at the same effective concentrations. For
experiments involving ROS quencher, NAC (2mM) was added at the same time and co-incubated
for the entire duration. Subsequently, the cells were lysed with RIPA lysis buffer [100 µL, 0.1%
SDS, 0.5% Sodium Deoxycholate, 1% IGEPAL® CA-630, 150 mM NaCl, 25 mM Tris-HCl (pH
8.0), protease and phosphatase inhibitor cocktail]. The cell lysates were transferred to separate 1.5
mL tubes and sonicated for 3 x 15 s. The samples were then centrifuged at 13000 rpm, 4˚C for 15
min. The liquid supernatant containing the proteins were collected and total protein content of each
sample was quantified via Bradford’s assay. 50 µg of proteins from each sample were reconstituted
in loading buffer [100 mM DTT, 1x Protein Loading Dye] and heated at 95˚C for 5 min. The
protein mixtures were resolved on 10% SDS-PAGE gel by electrophoresis and transferred to a 0.2
µm nitrocellulose membrane. Protein bands were visualized via enhanced chemiluminescence
imaging (PXi, Syngene) after treatment with the appropriate primary and HRP-conjugated
secondary antibodies. Equal loading of protein was confirmed by comparison with actin
expression. The following antibodies were used: p53 (FL-393) (sc-6243) and p21 (F-5) (sc-6246)
from Santa Cruz Biotechnologies. CHOP (D46F1), PERK (D11A8), IRE1a (14C10) and phospho-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
eif2a from Cell Signaling Technologies. β-Actin (ab8229) from Abcam. PierceTM HRP-
conjugated anti-rabbit IgG (H+L) (#31460), anti-mouse IgG (H+L) (#31430) and anti-goat IgG
(H+L) (#31402) from Thermo Fisher Scientific. All antibodies were used at 1:1000 dilutions
except for actin (1:10000), anti-mouse, anti-goat and anti-rabbit (1:5000).
12
ACS Paragon Plus Environment

Page 13 of 34
Molecular Pharmaceutics
1
2
3
4
RESULTS & DISCUSSION
5
6
7
Design consideration, synthesis and characterization
8
9
In our previous work, we identified structural motifs common to the library of p53-independent
RAS complexes with diverse physico-chemical properties that were important for the high efficacy
and p53-independent anticancer activity, namely the triisopropylbenzene (TPB) arene moiety and
the quinoline ligand.33 In the present study, using 1 as the core structure, we perform minor
structural modification to the iminoquinoline chelate ligand to modulate its π-acidity (Figure 2).
This was done by replacing the 4-OMe (1) moiety with more electron-donating 4-NMe2 (3) (less
π-acidic) or with their corresponding isoelectronic counterparts, namely 4-Et (2) and 4-iPr (4),
which do not contain any conjugated lone-pair electrons (more π-acidic). We performed structural
tuning studies on these new complexes alongside 1 to study how the changes in chelate ligand π-
acidity could affect properties such as p53-dependence, ROS and ER stress-induction.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. RAS complexes investigated in the present study. Hydrophobicity and p-acidity of the
chelating ligands were modulated by varying the highlighted functional moiety.
13
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 14 of 34
1
2
3
4
5
6
7
8
9
New complexes 2 – 4 were synthesized using a similar protocol for the synthesis of 1 (Scheme
1).32 Briefly, 2-quinolinecarboxyaldehyde and aniline derivatives, containing either the 4-NMe2,
4-Et or 4-iPr moiety, were added to dry EtOH and allowed to react at r.t. to give the
iminoquinoline ligands with varying degrees of purity. The crude ligands were used directly for
chelation with stoichiometric amount of the TPB-Ru precursor in MeOH to give the corresponding
RAS complex, after purification by silica gel column as crystalline solids. Their identities were
verified by 1H NMR and ESI-MS analysis and their purities were determined to be >95% by RP-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
HPLC or elemental analysis. The H NMR of 2 – 4 showed resonances typical of RAS
complexes.32-34 The signal corresponding to the imine proton at ca. 9.1 ppm, and the additional
splitting pattern of the TPB protons signals indicated the formation and chelation of the
iminoquinoline to the Ru center (Supplementary Information (SI), Figure S1, S3 and S5). Their
ESI-MS spectra showed only the characteristic [M]+ ion with Ru and Cl isotopic pattern (SI, Figure
S2, S4 and S6) and tandem MS-MS gave expected fragmentation patterns.
Scheme 1. General synthetic route for 2nd Generation RAS complexes.
14
ACS Paragon Plus Environment

Page 15 of 34
Molecular Pharmaceutics
1
2
3
4
Influence of Schiff-base ligands functional moiety on π-acidity and stability
5
6
7
8
9
We studied how the structural modifications and chelate ligand π-acidity affected
physiochemical properties such as stability and hydrophobicity, which are important factors that
influence RAS complexes’ biological activity. We first confirmed the order of chelate ligand π-
acidity by comparing the UV-vis spectrums and λmax of the RAS complexes. In principle, the lower
the π-acidity of the ligand (less π-backbonding) the higher the energy of the Ru π-HOMO, which
would result in a smaller π - π* (HOMO-LUMO) energy gap. This means that a lower excitation
energy (higher λmax) would be required for a π - π* transition in a complex with a less π-acidic
ligand. Indeed, complexes containing motif with conjugated lone-pair electrons such as 1 and 3
exhibited λmax at 387 nm and 460 nm, respectively, validating the lower π-acidity of their chelate
ligand (Figure 3). Conversely, complexes having more π-acidic ligands without conjugated lone-
pair electrons, such as 2 and 4, exhibited much lower λmax at 365 nm. Therefore, the degree of
iminoquinoline ligand π-acidity was established to be 2 ≈ 4 > 1 > 3, in agreement to our initial
design considerations and theoretical predictions.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. UV-vis spectrum confirms the ranking order of chelate ligand π-acidity. Lower π-acidity
of the chelate ligand results in lower energy π-π* transition as indicated by a higher λmax. Therefore,
the order of chelate ligand π-acidity: 2 ≈ 4 > 1 > 3.
15
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 16 of 34
1
2
3
4
5
6
7
8
Next, we determined the stability of the RAS complexes in different media such as double-
distilled water (ddH2O), Dulbecco’s Modified Eagle’s Medium (DMEM) with 10% fetal bovine
serum (FBS), and aqueous N-acetylcysteine (NAC, 2 mM) by monitoring their UV-vis profile over
24 h. Any chemical interactions between the complexes and the media leading to the formation of
new chemical species would manifest in a gradual shift in the UV-vis profile over time. All RAS
complexes demonstrated stability towards hydrolysis in ddH2O (SI, Figure S7). In DMEM with
10% FBS, 3 having the least π-acidic chelate ligand exhibited minimal changes in UV-vis profile,
demonstrating the greatest stability (Figure 4a). This was consistent with our previous study
showing that a less π-acidic ligand results in the more electron-rich Ru center, stabilizing the Ru-
Cl bond and making it more resistant to substitution by ‘N’- or ‘S’- containing nucleophiles.33 In
keeping with these observations, 1 was previously shown to be stable in Hank’s Balanced Salt
Solution (HBSS) buffer containing 10% FBS and aqueous solutions of biological nucleophiles
such as glutathione and dGMP.32, 34 Its instability in DMEM with 10% FBS suggest that reaction
with the Ru-Cl bond is still possible in medium containing high concentration of nucleophiles
(DMEM is modified from Basal Medium Eagle (BME) and contains 4x more amino acids and
vitamins. It also has 4.5x more glucose than HBSS). It is noteworthy that the trend in stability was
reversed in aqueous NAC. RAS complexes 1, 2 and 4 had unchanged UV-vis profile after 24 h in
NAC while a significant shift was observed for 3 (Figure 4b). Since 3 demonstrated the greatest
stability towards substitution reaction, we postulated that the shift in UV-vis profile was likely due
to a redox reaction between the electron-rich Ru center of 3 and the reducing NAC. However,
more studies would be required to fully understand the interaction between NAC and 3.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
16
ACS Paragon Plus Environment

Page 17 of 34
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. p-acidity of the chelating ligands affects the stability of the RAS complexes to
nucleophilic substitution or reduction. UV-vis profile of RAS complexes in (a) DMEM media with
10% FBS and (b) aqueous NAC (2 mM). Grey arrows indicate regions of increasing or decreasing
absorbance over time.
The log POW of 1 – 4 were determined by the shake-flask method using UV-vis spectroscopy
analysis.35 In general, log POW of the RAS complexes followed an expected trend and structural
modification from the core RAS-1T structure such as addition of methyl groups (4-Et à 4-iPr)
or replacing hydrogen-bonding heteroatoms (N, O) with isoelectronic CHx groups (4-OMe à 4-
à
Et; 4-NMe2 4-iPr) had a predictable effect in log POW (Table 1; SI, Figure S8). All compounds
had log Pow in the negative range, in keeping with previous studies.33 This was presumably due to
the charged nature of the RAS complexes containing neutral chelate ligands.
Efficacy studies in drug-sensitive and –resistant cancers and p53-independent studies
To ascertain the anti-proliferative activity of 1 – 4, we tested them against of a panel of drug-
sensitive and -resistant cancer cell lines that include ovarian (A2780, A2780cisR), gastric (AGS)
and colorectal (HT29, HCT116, HCT116 p53-/-, TC7) cancers, and measured cell viability 48 h
17
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 18 of 34
1
2
3
4
5
6
7
8
9
after treatment using standard MTT assay. We included cisplatin (CDDP), oxaliplatin (OXP) and
5-fluorouracil (5-FU) as clinical drug controls and compare the efficacies of the RAS complexes
to these controls, particularly in cancer lineages that are drug-resistant. For example, ovarian
A2780cisR is a CDDP-resistant variant of A2780 wild-type and its resistance arises from the
accumulation of mutated p53 and more efficient DNA repair mechanism.38-39 p53-null colorectal
HCT116 p53-/- is less sensitive to p53-dependent anticancer agents such as OXP and 5-FU
compared to its wild type lineage.40 Lastly, the multidrug-resistant TC7 demonstrated low
sensitivity towards most apoptosis-inducing drugs such as OXP, 5-FU and doxorubicin in a panel
of colorectal cancer cell lines,34, 41 due to numerous defects in its apoptotic machinery, such as its
p53-/0 status and increased expression of anti-apoptotic Bcl-2/Bcl-xL.41 Compounds efficacious
against these cell lines would have a greater potential in the treatment of multidrug-resistant
cancers.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Cytotoxicity data for second generation RAS Complexes.
IC50 [µM] b
a
Complex
Log Pow
HCT116
wt
HCT116
p53-/-
A2780
-0.85 ± 0.02 0.53 ± 0.15
0.37 ± 0.09
-0.55 ± 0.01 0.29 ± 0.06
A2780cisR
0.87 ± 0.22
0.49 ± 0.08
AGS
HT29
TC7
1 (RAS-1T)
2
0.74 ± 0.14
0.91 ± 0.07
1.14 ± 0.12
0.53 ± 0.11
0.73 ± 0.12
0.39 ± 0.11
2.95 ± 0.18
1.37 ± 0.07
-0.62 ± 0.01
0.54 ± 0.04
0.42 ± 0.04
3
0.43 ± 0.09
0.46 ± 0.04
n.d.
0.38 ± 0.06
0.45 ± 0.04
n.d.
0.39 ± 0.03
0.38 ± 0.05
1.16 ± 0.08
5.88 ± 1.95
n.d.
0.46 ± 0.05
0.47 ± 0.05
3.23 ± 0.81
7.33 ± 2.25
n.d.
0.27 ± 0.05
0.34 ± 0.05
16.7 ± 5.4
33.6 ± 17.4
n.d.
1.60 ± 0.12
1.02 ± 0.04
15.1 ± 4.1
396 ± 133
n.d.
4
-0.31± 0.03
n.d.
0.35 ± 0.12
n.d.
Oxaliplatin
5-Fluorouracil
Cisplatin
n.d.
n.d.
n.d.
n.d.
n.d.
1.36 ± 0.11
7.73 ± 1.08
29.3 ± 2.1
aLog Pow values determined via the shake-flake method against 1:1 n-octanol:H2O partitioning. bIC50 values is the concentration of Ru complexes
required to inhibit 50% of cell growth with respect to control groups, measured by MTT assay after 48 h of incubation. Data obtained are based on
the average of three independent experiments, and the reported errors are the corresponding standard deviations. The IC50 were corrected using
actual [Ru] determined using ICP-OES.
18
ACS Paragon Plus Environment

Page 19 of 34
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
Compound 1 – 4 displayed nanomolar IC50 values in both drug-sensitive cell lines (A2780, AGS,
HT29, HCT116) and drug-resistant phenotypes (A2780cisR, HCT116 p53-/-); in multidrug-
resistant TC7, 1 – 4 displayed low micromolar IC50 values (Table 1). In all cell lines, 1 – 4 were
several times more efficacious than clinical drugs CDDP, OXP and 5-FU. This was most apparent
when comparing their activities in the drug-resistant cell lineages. For instance, 1 – 4 had resistance
factor between 1.3 – 1.6 in A2780cisR/A2780 pair while CDDP showed resistant factor of 5.7
(Table 2; SI, Figure S9a). In HCT116 p53-/-/HCT116 cell pair, 1 – 4 had p53-dependence factor
< 1 while OXP and 5-FU had much higher values of 5.2 and 4.6 respectively (Figure 5a, Table 2).
Lastly, 1 – 4 demonstrated apoptosis-resistance factor between 2.2 – 3.5 in TC7/HCT116 cell pair
while OXP and 5-FU displayed higher resistance factor of 4.6 and 54 respectively (Table 2; SI,
Figure S9b). Taken together, 1 – 4 showed high efficacies against both drug-sensitive and drug-
resistant cell phenotypes and were much less affected by the same resistance mechanisms
impeding the activities of clinical drugs such as CDDP, OXP and 5-FU in A2780cisR, HCT116
p53-/- and TC7 cells.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. Resistance factor for RAS Complexes.
Apoptosis Resistance
Resistance Factora
p53-Dependence Factorb
Factorc
Complex
Ratio of IC50
(A2780cisR / A2780)
Ratio of IC50
Ratio of IC50
(TC7/HCT116)
(HCT116 p53-/- / HCT116)
1 (RAS-1T)
1.6
1.3
1.5
1.3
n.d.
n.d.
5.7
0.6
0.7
0.6
0.7
5.2
4.6
n.d.
2.6
2.6
3.5
2.2
4.7
54
2
3
4
Oxaliplatin
5-Fluorouracil
Cisplatin
n.d.
a, b, c Resistance Factors were calculated by taking the ratio of IC50 in resistant cell lines and sensitive cell
line; a smaller value represents a greater selectivity towards resistant cell lines.
19
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 20 of 34
1
2
3
4
5
6
7
8
9
We further confirmed the p53-independent activity of 1 – 4 by investigating the induction of p53
and downstream target p21 in HCT116 cells after drug treatment (Figure 5b), in comparison with
OXP as the positive control. In general, 1 – 4 showed a lack of induction of p53 regardless of
concentration or duration of treatment. A slight induction of p21 was observed with 1, 2 and 3
only after 24 h exposure at IC75 concentration, likely via p53-independent pathways.42 In contrast,
p53-dependent OXP induced a strong upregulation of both p53 and p21 in a concentration and
time-dependent manner.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. RAS Complexes display p53-independent activity. (a) IC50 values of RAS complexes,
oxaliplatin and 5-fluorouracil after 48h treatment in both HCT116wt and HCT116 p53-/- cells.
Mean ± s.e.m. (* p < 0.05; two-tailed Student’s t-test). (b) Western blot analysis of p53 and target
p21 in HCT116 cells after treatment with RAS complexes and oxaliplatin at IC50 and IC75 for 6h
and 24h. Homogeneous protein loading determined with reference to actin.
20
ACS Paragon Plus Environment

Page 21 of 34
Molecular Pharmaceutics
1
2
3
4
Tuning π-acidity of chelate ligand allows modulation of ROS-induction
5
6
7
8
9
Many metallodrugs including 1 has been shown to induce ROS leading to cell death, due to the
redox active nature of metal complexes.43 However, there are limited study on how the electron-
donating/accepting ability of ligands affect ROS induction by these metal complexes. Hence, we
investigate if varying the ligand π-acidity in the present series of compounds affects cellular ROS
accumulation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cellular ROS was quantified using flow cytometry after cell staining with commercially
available, cell-permeable ROS probe, H2DCFDA. HCT116 cells were pre-stained with H2DCFDA
before treatment with varying concentrations of 1 – 4 for 4 hours. Thereafter, the treated samples
were analyzed and compared to the stained, untreated control. Compounds 1, 2 and 4 induced ROS
in a concentration-dependent manner (Figure 6). 2 and 4 having the most π-acidic ligands, were
the strongest ROS inducers, increasing cellular ROS by 2.6-fold and 3.5-fold at 5 µM treatment
while ROS induction by 1 was more modest only achieving 2.5-fold at a higher concentration of
15 µM. In contrast, 3 having the least π-acidic ligands, only increased ROS levels modestly when
treated at the highest concentration of 15 µM with no increase in ROS at treatment concentration
less than 10 µM. The results suggested a strong correlation between degree of chelate ligand π-
acidity and ROS induction, following the order 2 ≈ 4 > 1 > 3. It is also noteworthy that complexes
with less stable Ru-Cl bonds demonstrated stronger ROS induction. It is possible that a certain
degree of Ru-Cl bond lability was required for the RAS complexes to participate directly with
cellular redox reactions, disrupting redox homeostasis and resulting in the accumulation of ROS.43
21
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 22 of 34
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Chelate Ligand π-acidity influences degree of ROS-induction. Detection of ROS with
H2DCFDA (20 µM) after treatment with RAS complexes for 4 h using flow cytometry
measurements. Mean ± s.e.m. (* p < 0.05, ** p < 0.01; Student’s t test).
To ascertain if the degree of cellular accumulation affects ROS-induction, we quantify Ru
content in the cell after drug treatment by ICP-MS analysis. Briefly, HCT116 cells were treated
with compounds 1 – 4 at IC50 concentrations for 24 h before being detached from the cell culture
vessels, washed with PBS buffer and digested in 65% nitric acid with heating at 100˚C. Once the
sample has dried up, the remaining residue was reconstituted and diluted in ddH2O before Ru
content in cells was quantified by ICP-MS. Compounds 1 – 4 accumulated in cells to similar extent
with no significant differences (p > 0.05) when treated at their respective IC50 values (SI, Figure
S10). Hence, any observed difference in ROS induction is not due to to differential uptake.
As there have been several reports of metal complexes causing increased cellular ROS
accumulation as a direct consequence of thioredoxin reductase (TrxR) inhibition,44-46 we
investigate if this was the case for our series of complexes. We measured the percentage inhibition
of mammalian TrxR by compound 1 – 4 using a colorimetric thioredoxin reductase inhibition assay
according to the protocol provided by the manufacturer. All compounds did not significantly
inhibit TrxR at 1x or 2x IC50 concentrations, in the micromolar range (SI, Figure S11). This was
22
ACS Paragon Plus Environment

Page 23 of 34
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
in contrast to Au complexes which has been known to inhibit TrxR at low nanomolar range.45
Hence, we concluded that cellular ROS accumulation caused by the RAS complexes was not due
to TrxR inhibition.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ROS functions as signaling molecules in many complex cellular pathways and dysregulation in
any number of these pathways could result in an accumulation of ROS in the cell. Hence,
identifying the exact pathway(s) implicated in ROS-induction by 1 – 4 would require more
extensive studies.
Tuning π-acidity of chelate ligand affects duration and intensity of ER stress
Next, we investigated how tuning π-acidity would influence ER stress induction by 1 – 4.
Induction of ER stress would lead to the unfolded protein response (UPR), which mediates
recovery, cellular dysfunction and cell death.47 Three distinct UPR signaling pathways have been
identified, namely the PERK, IRE1α and ATF6 pathway.48 We focused our investigation on two
of the three UPR pathway, namely the PERK and IRE1a pathway by studying their accumulation
and phosphorylation pattern. In addition, we probed the cellular expression of transcription factor
CHOP, a common and highly-induced downstream biomarker of UPR during ER stress.49
Cellular IRE1α, phopho-IRE1α, PERK, phospho-PERK, p-eif2α and CHOP levels were
ascertained by Western blot analysis, after treatment with 1 – 4 under various conditions. All four
compounds seem to induced ER stress via IRE1α pathway as seen by phosphorylation of IRE1α
either after 6h or 24 h treatment at IC75 concentration (Figure 7). Phosphorylation of IRE1α by 4
is the least intense, with only a slight induction after 6 h treatment. The lack of phosphorylation of
PERK or downstream eif2α suggest that the PERK arm of the UPR pathway is not activated by
treatment with 1 – 4 (SI, Figure S12). In addition, all four compounds induced concentration-
23
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 24 of 34
1
2
3
4
5
6
7
8
9
dependent upregulation of CHOP at 6 h (Figure 7). However, ER stress was more sustained for
compounds with less π-acidic ligands, 1 and 3, as seen by the significant upregulation of CHOP
even at 24 h after treatment. In contrast, compounds with more π-acidic ligands, 2 and 4, a less
sustained ER stress was observed with significantly lower expression of CHOP at 24h after
treatment.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. RAS complexes induce ER stress, activating the IRE1a UPR pathway. Western blot
analysis of various ER stress biomarker after 6 h and 24 h treatment with 1 – 4 at IC50 and IC75
concentrations. Homogeneous protein loading determined with reference to actin.
ROS-mediated versus ROS-independent ER stress induction
Complex 1 was previously shown to induce ROS at early time points resulting in ROS-mediated
ER stress, which could be abolished by co-treatment with a ROS-quencher NAC.34 Thus, we
investigated how co-treatment with NAC influenced cell viability and ER stress induction by
analogous 1 – 4. This was done by measuring cell viability and CHOP expression in HCT116 cells
after treatment with compounds 1 – 4, in the absence and presence of NAC. For compound 1, 2
and 4, quenching ROS with NAC (2 mM) resulted in an almost complete protection of the cells at
all treatment concentrations (Figure 8a), showing the cellular ROS accumulation was important
for the activity of 1, 2 and 4. For compound 3, co-incubation with NAC also provided a cyto-
24
ACS Paragon Plus Environment

Page 25 of 34
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
protective effect to a lesser degree. However, the lack of ROS induction by 3 even at highly
cytotoxic concentrations (Fig 6) suggested that ROS was not critical for its activity. Hence, the
observed cyto-protective effect could be due to a ‘detoxification’ by the direct redox reaction
between NAC and 3 (Figure 4b). Nevertheless, the exact nature of the reaction between NAC and
3 is something that requires more extensive investigation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Based on cellular CHOP levels, NAC significantly reduced its expression in cells co-treated with
1, 2 and 4, suggesting that NAC could protect cells against these compounds, restoring ER
homeostasis and shutting off UPR signaling (Figure 8b). In contrast, NAC had a lesser effect on
CHOP expression in cells co-treated with 3, suggesting that NAC was unable to restore ER
homeostasis as effectively in this instance. This further confirmed the non-oxidative nature of the
ER stress induced by 3.
Figure 8. Tuning chelate ligand π-acidity causes a switch between ROS-mediated ER stress and
ER stress not mediated by ROS induction. (a) Cell viability of HCT116 cells treated with RAS
complexes after 48h treatment in the absence or presence of NAC (2 mM). Mean ± s.e.m. (* p <
0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001; two-tailed Student’s t-test). (b) Western blot
analysis of CHOP in HCT116 cells after treatment with RAS complexes at IC75 for 6h, in the
absence or presence of NAC (2 mM). Homogeneous protein loading determined with reference to
actin.
25
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 26 of 34
1
2
3
4
5
6
7
8
Taken together, the results indicated that 1, 2 and 4 exerted their mode-of-action via ROS-
mediated ER stress while 3 induced ER stress without corresponding ROS induction; ROS and ER
stress induction were important for the activity of 1, 2 and 4 and co-treatment with NAC restored
redox and ER homeostasis and rescued cells from cell death. In contrast, 3 did not increase ROS
levels and co-treatment with NAC did not mediate cell death or ER stress to the same extent. We
surmised that direct tuning of the iminoquinoline ligand’s π-acidity could be used a means to
modulate ROS induction of RAS complexes, thereby switching from a ROS-mediated ER stress
mode-of-action to an ER stress pathway that is not mediated by ROS.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CONCLUSION
We presented a new class of Ru-based ER stress inducers with high efficacies in a panel drug-
sensitive and -resistant cancer types and performed structural tuning studies by varying the chelate
ligand π-acidity and studying its effect on stability, ROS and ER stress induction. Our studies
showed that by varying the chelate ligand π-acidity, the stability of the complexes towards
substitution and redox reaction could be chemically tuned; RAS complexes containing greater
chelate ligand π-acidity (1, 2 and 4) had more labile Ru-Cl bond that undergo substitution while
RAS complex with lower π-acidic ligand (3) had more stable Ru-Cl but was more susceptible to
redox reaction with reductant NAC due to its more electron-rich Ru center. Also, decreasing
chelate ligand π-acidity on the RAS complexes reduced ROS induction and caused a switch from
a ROS-mediated ER stress pathway to an ER stress pathway not mediated by ROS. This current
work adds to the available pool of ER stress inducer and lends a greater understanding on how
structural changes on RAS complexes affect ROS and ER stress pathway activation, which could
26
ACS Paragon Plus Environment

Page 27 of 34
Molecular Pharmaceutics
1
2
3
4
5
6
lead to better design of ER stress inducers with specific mode-of-action and efficacy against drug-
resistant cancer types.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
AUTHOR INFORMATION
Corresponding Authors*
1Email: csicmj@nus.edu.sg
3E-mail: chmawh@nus.edu.sg, Phone: +65 6516 5131.
Author Contributions
M.J.C. conceived the study, designed all experiments, performed chemical synthesis and in vitro
experiments, and analyzed all data. M.V.B. and M.C.C. performed cellular ROS accumulation
analysis. M.V.B. performed Ru cellular accumulation studies, TrxR inhibition assay and analysis,
and western blot experiments. K.W.T. performed cell viability assays. C.G. and G.P advised on
biological experiments. W.H.A. conceived and directed the study. M.J.C, M.V.B and W.H.A wrote
the manuscript with contributions from all authors. All authors have read, edited and given
approval to the final version of the manuscript.
Funding Sources
Financial support from Ministry of Education and the National University of Singapore (R143-
000-638-112) and Ligue contre le cancer, CNRS, European COST action CM1105 is gratefully
acknowledged.
27
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 28 of 34
1
2
3
4
ACKNOWLEDGEMENTS
5
6
7
8
9
The authors thank staff of CMMAC, NUS for performing elemental analysis and ICP-OES
analysis, Paul J. Dyson (EPFL) and Han-Ming Shen (NUS) for their gift of cell lines.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ABBREVIATIONS
ROS, reactive oxygen species; ER, endoplasmic reticulum; CDDP, cisplatin; OXP, oxaliplatin, 5-
FU, 5-fluorouracil; H2DCFDA, 2′,7′-dichlorodihydrofluorescein diacetate; TrxR, thioredoxin
reductase; EDTA, ethylenediaminetetraacetic acid; NEAA, non-essential amino acids; FBS, fetal
bovine serum; DMEM, Dulbecco’s Modified Eagle’s Medium; BME, Basal Medium Eagle; NMR,
nuclear magnetic resonance; ESI-MS, electrospray ionization-mass spectrometry; RP-HPLC,
reverse phase-high performance liquid chromatography; DMSO-d6, deuterated dimethyl
sulfoxide; MeOH-d4, deuterated methanol; ddH2O, ultrapure water; ICP-OES, inductively couple
plasma-optical emission spectrometry; ICP-MS, inductively couple plasma-mass spectrometry
NAC, N-acetylcysteine; NADPH, Nicotinamide adenine dinucleotide phosphate, DTNB, 5,5-
dithio-bis-(2-nitrobenzoic acid); UPR, unfolded protein response; RAS, ruthenium
(II) arene Schiff-base; TPB, triisopropylbenzene; Bcl-2, B-cell lymphoma 2; Bcl-xL,
B-cell lymphoma extra large; BAX, bcl-2-associated X protein; CHOP, CCAAT-
enhancer-binding protein homologous protein; PERK, PKR-like ER kinase; IRE1a,
inositol-requiring enzyme 1A; ATF6, activating transcription factor 6; eif2a,
eukaryotic translation initiation factor 2A; Hank’s Balanced Salt Solution (HBSS); dGMP, 2′-
deoxyguanosine 5′-monophosphate sodium salt hydrate.
28
ACS Paragon Plus Environment

Page 29 of 34
Molecular Pharmaceutics
1
2
3
4
Supporting Information Available: 1H NMR, ESI-MS and UV-Vis spectrums of compounds.
5
6
Thioredoxin reductase inhibition assay data, western blot data.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
REFERENCE
1.
2.
Fulda, S., Tumor Resistance to Apoptosis. Int. J. Cancer 2009, 124, 511-515.
Hervouet, E.; Cheray, M.; Vallette, F. M.; Cartron, P. F., DNA Methylation and Apoptosis
Resistance in Cancer Cells. Cells 2013, 2, 545-573.
3. O'Gorman, D. M.; Cotter, T. G., Molecular Signals in Anti-apoptotic Survival Pathways.
Leukemia 2001, 15, 21-34.
4.
Pommier, Y.; Sordet, O.; Antony, S.; Hayward, R. L.; Kohn, K. W., Apoptosis Defects and
Chemotherapy Resistance: Molecular Interaction Maps and Networks. Oncogene 2004, 23, 2934-
2949.
5.
Soengas, M. S.; Lowe, S. W., Apoptosis and Melanoma Chemoresistance. Oncogene 2003,
22, 3138-3151.
6.
Paul, I.; Jones, J. M., Apoptosis Block as a Barrier to Effective Therapy in Non Small Cell
Lung Cancer. World J. Clin. Oncol. 2014, 5, 588-594.
7.
Glioblastomas: Opportunities for Targeted Therapeutics. Mol. Cancer 2010, 9, 135.
8. Robles, A. I.; Harris, C. C., Clinical Outcomes and Correlates of TP53 Mutations and
Cancer. Cold Spring Harb. Perspect. Biol. 2010, 2, a001016.
9. Michaud, W. A.; Nichols, A. C.; Mroz, E. A.; Faquin, W. C.; Clark, J. R.; Begum, S.;
Krakstad, C.; Chekenya, M., Survival Signalling and Apoptosis Resistance in
Westra, W. H.; Wada, H.; Busse, P. M.; Ellisen, L. W.; Rocco, J. W., Bcl-2 Blocks Cisplatin-
Induced Apoptosis and Predicts Poor Outcome Following Chemoradiation Treatment in Advanced
Oropharyngeal Squamous Cell Carcinoma. Clin Cancer Res. 2009, 15, 1645-1654.
10.
Silvestrini, R.; Benini, E.; Veneroni, S.; Daidone, M.; Tomasic, G.; Squicciarini, P.;
Salvadori, B., p53 and bcl-2 Expression Correlates With Clinical Outcome in a Series of Node-
Positive Breast Cancer Patients. J. Clin. Oncol. 1996, 14, 1604-1610.
29
ACS Paragon Plus Environment