Catch and Release Photosensitizers: Combining Dual-Action Ruthenium Complexes with Protease Inactivation for Targeting Invasive Cancers

Article abstract of DOI:10.1021/jacs.8b08853

Dual action agents containing a cysteine protease inhibitor and Ru-based photosensitizer for photodynamic therapy (PDT) were designed, synthesized, and validated in 2D culture and 3D functional imaging assays of triple-negative human breast cancer (TNBC). These combination agents deliver and release Ru-based PDT agents to tumor cells and cause cancer cell death upon irradiation with visible light, while at the same time inactivating cathespin B (CTSB), a cysteine protease strongly associated with invasive and metastatic behavior. In total five Ru-based complexes were synthesized with the formula [Ru(bpy)₂(1)](O₂CCF₃)₂ (3), where bpy = 2,2′-bipyridine and 1 = a bipyridine-based epoxysuccinyl inhibitor; [Ru(tpy)(NN)(2)](PF₆)₂, where tpy = terpiridine, 2 = a pyridine-based epoxysuccinyl inhibitor and NN = 2,2′-bipyridine (4); 6,6′-dimethyl-2,2′-bipyridine (5); benzo[i]dipyrido[3,2-a:2′,3′-c]phenazine (6); and 3,6-dimethylbenzo[i]dipyrido[3,2-a:2′,3′-c]phenazine (7). Compound 3 contains a [Ru(bpy)₃]²⁺ fluorophore and was designed to track the subcellular localization of the conjugates, whereas compounds 4-7 were designed to undergo either photoactivated ligand dissociation and/or singlet oxygen generation. Photochemical studies confirmed that complexes 5 and 7 undergo photoactivated ligand dissociation, whereas 6 and 7 generate singlet oxygen. Inhibitors 1-7 all potently and irreversibly inhibit CTSB. Compounds 4-7 were evaluated against MDA-MB-231 TNBC and MCF-10A breast epithelial cells in 2D and 3D culture for effects on proteolysis and cell viability under dark and light conditions. Collectively, these data reveal that 4-7 potently inhibit dye-quenched (DQ) collagen degradation, whereas only compound 7 causes efficient cell death under light conditions, consistent with its ability to release a Ru(II)-based photosensitizer and to also generate ¹O₂.

Full text of DOI:10.1021/jacs.8b08853

Subscriber access provided by University of Sunderland
Article
Catch and Release Photosensitizers: Combining Dual Action Ruthenium
Complexes with Protease Inactivation for Targeting Invasive Cancers
Karan Arora, Mackenzie Herroon, Malik H. Al-Afyouni, Nicholas P. Toupin, Thomas N.
Rohrabaugh, Jr., Lauren M. Loftus, Izabela Podgorski, Claudia Turro, and Jeremy J. Kodanko
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b08853 • Publication Date (Web): 02 Oct 2018
Downloaded from http://pubs.acs.org on October 3, 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 48
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Catch and Release Photosensitizers: Combining Dual Action
Ruthenium Complexes with Protease Inactivation for
Targeting Invasive 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
Karan Arora,1 Mackenzie Herroon,2 Malik H. Al-Afyouni,3 Nicholas P. Toupin,1
Thomas N. Rohrabaugh, Jr.,3 Lauren M. Loftus,3 Izabela Podgorski,2,4* Claudia
Turro,3* and Jeremy J. Kodanko1,4*
1 Department of Chemistry, Wayne State University, 5101 Cass Ave, Detroit, MI 48202
2Department of Pharmacology, School of Medicine, Wayne State University, Detroit, Michigan
48201
3Department of Chemistry and Biochemistry, The Ohio State University, Columbus Ohio 43210,
4Barbara Ann Karmanos Cancer Institute, Detroit, Michigan 48201
Abstract
Dual action agents containing a cysteine protease inhibitor and Ru-based photosensitizer for
photodynamic therapy (PDT) were designed, synthesized and validated in 2D culture and 3D
functional imaging assays of triple-negative human breast cancer (TNBC). These combination
agents deliver and release Ru-based PDT agents to tumor cells and cause cancer cell death upon
irradiation with visible light, while at the same time inactivating cathespin B (CTSB), a cysteine
protease strongly associated with invasive and metastatic behavior. In total five Ru-based
complexes were synthesized with the formula [Ru(bpy)2(1)](O2CCF3)2 (3), where bpy = 2,2'-
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 48
1
2
3
4
5
6
bipyridine and 1 = a bipyridine-based epoxysuccinyl inhibitor, and [Ru(tpy)(NN)(2)](PF6)2 where
tpy = terpiridine, 2 = a pyridine-based epoxysuccinyl inhibitor, NN = 2,2'-bipyridine (4); 6,6'-
7
8
9
dimethyl-2,2'-bipyridine
(5);
benzo[i]dipyrido[3,2-a:2',3'-c]phenazine
(6);
3,6-
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
dimethylbenzo[i]dipyrido[3,2-a:2',3'-c]phenazine (7). Compound 3 contains a Ru(bpy)3
fluorophore and was designed to track the subcellular localization of the conjugates, whereas
compounds 4–7 were designed to undergo either photoactivated ligand dissociation and/or singlet
oxygen generation. Photochemical studies confirmed that complexes 5 and 7 undergo
photoactivated ligand dissociation, whereas 6 and 7 generate singlet oxygen. Inhibitors 1–7 all
potently and irreversibly inhibit CTSB. Compounds 4–7 were evaluated against MDA-MB-231
TNBC and MCF-10A breast epithelial cells in 2D and 3D culture for effects on proteolysis and
cell viability under dark and light conditions. Collectively, these data reveal that 4–7 potently
inhibit dye-quenched (DQ) collagen degradation, whereas only compound 7 causes efficient cell
death under light conditions, consistent with its ability to release a Ru(II)-based photosensitizer
and to also generate 1O2.
Introduction
Cysteine cathepsins are a family of 11 human cysteine proteases that are highly expressed
in a variety of cancers.1-3 Collectively, these enzymes contribute to tumor progression, growth,
invasion, angiogenesis and metastasis. In addition to being found intracellularly in lysosomes,
cysteine cathepsins are secreted and bind to the surface of cancer cells.3 Activity-based probes
(ABPs) that target cysteine cathepsins rapidly differentiate between normal and tumor tissue in
vivo, allowing for visual imaging of cells left behind at the tumor margins after resection.4-11 These
results show that surface-bound cathepsins can be targeted to achieve selective delivery of agents
2
ACS Paragon Plus Environment

Page 3 of 48
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
to the invasive edges of tumors.12 Prodrug activation strategies that take advantage of proteolytic
activity of pericellular cathepsin B have also shown promise, confirming that these enzymes are
active at the surface of tumors.13-15 Relative to this project, inhibitors of cysteine cathepsins have
shown efficacy in preclinical models of cancer, including breast cancer.16-20 Given the crucial role,
secretion, and membrane association of cysteine cathepsins in cancer, designing inhibitor-drug
conjugates that simultaneously block invasive behaviors and deliver drugs is an attractive
approach. In addition to cancer cells, tumor-associated macrophages (TAMs) and stromal cells
that express CTSB and contribute to invasiveness, but are difficult to kill with conventional
approaches, can be targeted with CTSB inhibitors as delivery vehicles.21 It is also important to
note that TAMs play a critical role in breast cancer malignancy.22
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
Photodynamic therapy (PDT) is a clinically approved technique for the treatment of many
cancers.23-33 PDT is commonly used after tumor surgical resection to kill cancer cells that invade
outside the margins and are left behind within the tumor cavity. It is a two-step process that
involves uptake of a photosensitizer by cancer cells, followed by drug activation using low energy
light, resulting in the generation of reactive oxygen species (ROS) that cause oxidative damage
and cell death. The precision of PDT is dependent upon selective uptake of the photosensitizer by
cancer cells, as well as and the region where irradiation occurs. Due to suboptimal tumor targeting
and uptake of current PDT drugs, considerable efforts have been made to enhance PDT drug
efficacy using standard drug delivery strategies.34 However, in all of these approaches, the uptake
of the photosensitizer into the cancer cell is a primary concern. In order for the PDT drug to do the
most damage to cancer cells, it must be able to reach intracellular organelles, especially the
mitochondria; generation of ROS at the cancer cell surface is not sufficient to achieve efficient cell
killing.29 In addition, subcellular localization can vary as a function of the method of conjugation
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 48
1
2
3
4
5
6
7
8
9
to the delivery agent. Importantly, none of the current delivery strategies for PDT drugs inhibit
crucial activity in cancer. Instead these approaches rely on ligand-receptor interactions on the
surface, which on their own show little anticancer activity.35-37 Recently, researchers have taken
advantage of high extracellular concentrations of proteases, including cysteine cathepsins, at tumor
sites to effect the proteolytic release of PDT agents and other drugs.14,15 While the strategy does
provide an advantage in that multiple equivalents of drug can be released through the catalytic
action of tumor-associated proteases, it does nothing to inactivate cathepsins and other proteases
that are associated with invasiveness and metastatic behavior.
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
We chose a different strategy to achieve highly effective PDT, namely the use of Ru(II)-
based photosensitizers that are resistant to photobleaching, are water soluble, cell permeable and
are known to accumulate within the mitochondria of cancer cells due to their lipophilic cationic
nature.38 Ruthenium complexes are a potent class of metallopharmaceuticals that have advanced
to clinical trials (Figure 1).39-42 Due to their rich photochemistry and resistance to
photobleaching,43 a common problem with current photosensitizers,44 ruthenium complexes are
undergoing active development as a new class of PDT drugs.45-48 Most recently the PDT drug
TLD-1433 advanced to Phase Ib clinical trials for the treatment of bladder cancer.49-51 These
complexes show attractive properties, including high stability and cell permeability,52,53 low
toxicity,54-57 and higher light to dark ratios for cell killing than current PDT drugs. The Turro group
recently developed hybrid complexes that show dual reactivity with low energy light, with high
quantum yields observed for photoactivated ligand dissociation and ROS generation.58,59 These
ruthenium complexes provide a perfect lead for the development of highly efficacious agents with
two mechanisms of action, where a ruthenium photosensitizer could be released, taken up by cells
4
ACS Paragon Plus Environment

Page 5 of 48
Journal of the American Chemical Society
1
2
3
4
5
6
and activated with light to generate ROS at a tumor site (Figure 2). However, in order for these
complexes to reach their full potential, methods for tumor-specific delivery must be developed.
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 1. Structures of ruthenium-based drugs
cell-specific
delivery and
cathepsin B inhibition
Cancer cell
O2
Cell Death
hn
photorelease,
cell uptake and
ROS generation
ROS
cathepsin B
inhibitor
Ru Photosensitizer
Figure 2. Dual therapeutic for tumor-specific delivery, cysteine cathepsin inhibition and cell death
through generation of ROS
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 48
1
2
3
4
5
6
7
8
9
In this manuscript we describe the synthesis, biochemical and biological characterization
of a dual action compound that potently inhibits CTSB and releases a Ru(II)-based photosensitizer
upon irradiation with visible light. This complex includes a delivery vector that not only carries
the Ru(II)-based photosensitizer to cancer and tumor-associated cells but also irreversibly
inactivates CTSB. Our lead compound is a derivative of CA-074 and NS-134 (Figure 3), which
are highly potent, selective and irreversible inhibitors of CTSB.60,61 In addition to our lead dual-
action compound, we synthesized four other Ru(II) conjugates that were used to test the role of
the inhibitor, photoactivated release, and generation of 1O2 on compound efficacy, as well as the
subcellular localization of the complexes. These complexes were evaluated in a 3D pathomimetic
model of triple negative human breast cancer (TNBC) in order to probe the efficacy of our
compounds in an environment that mimics the in vivo tumor microenvironment, including
acidification of the extracellular space. Our results confirm that photoactivated ligand release and
generation of 1O2 are vital in achieving efficient cell death.
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. Epoxysuccinyl-based inhibitors of CTSB
Experimental Section
6
ACS Paragon Plus Environment

Page 7 of 48
Journal of the American Chemical Society
1
2
3
4
General Considerations
5
6
7
8
NMR spectra were recorded on a Varian FT-NMR Mercury-400 MHz Spectrometer. Mass
spectra were recorded on a Time-of-Flight Micromass LCT Premier XE Spectrometer. IR spectra
were recorded on a Nicolet IS5 FT-IR spectrophotometer (thin film). UV-vis spectra were recorded
on a Varian Cary 60 spectrophotometer. All reactions were performed under ambient atmosphere
unless otherwise noted. Anaerobic reactions were performed by purging the reaction solutions with
Ar or nitrogen.
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
Methods
Stability Studies. DMSO stock solutions of complexes 3–7 (10.0 μL, 40 mM) were added to
individual wells on a clear microtiter plate containing 190 μL DMSO in the dark. Plates were
loaded into a Tecan Infinite M200 Monochromator-based Modular Multi-function Microplate
Reader preequilibrated at 25.0 °C and absorbance spectra were recorded every 4 h for 24 h.
Similarly DMSO stock solutions of complexes 3–7 (40 mM) were diluted in cell growth media
(phenol red-free Dulbecco’s modified Eagle’s cell growth medium (DMEM) supplemented with
10% FBS, 10mM HEPES, 100 U/ml penicillin-streptomycin, 25 mM PIPES, and acidified to pH
6.8) pre-equilibrated at 37 °C to A ≈ 1 (< 10% final concentration DMSO). Electronic absorption
spectra were recorded every 4 h for 24 h at 37 °C (Fig. S58-67), respectively.
Photochemistry. Electronic absorption spectra were obtained using a Hewlett-Packard 8453 diode
array spectrophotometer and luminescence was measured using a Horiba Fluoromax-4
spectrometer in 1×1 cm quartz cuvettes. Luminescence decay traces were collected using an
Edinburgh EPL-445 pulsed diode laser (λ = 444.4 nm, pulse width = 84.4 ps) focused on a sample
at 90° to an Edinburgh M300 emission monochromator and Bentham DH-00-Te PMT detector.
Sample irradiation was performed using a 150 W Xe arc lamp (USHIO) in a MilliArc lamp housing
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 48
1
2
3
4
5
6
7
8
9
unit with an LPS-220 power supply and an LPS-221 igniter (PTI). The desired irradiation
wavelengths were controlled by selecting the appropriate long pass (CVI Melles Griot) or bandpass
(Thorlabs) filters. In general, the samples were irradiated in water (<5% methanol) or acetonitrile
in a quartz cuvette. For photolysis experiments, electronic absorption spectra were recorded at
various time points of irradiation. The ligand exchange quantum yields (ΦLE) were determined by
irradiating the samples in water or acetonitrile using a 500 nm band pass filter together with a 475
nm long pass filter. Electronic absorption spectra were recorded at early irradiation times (<10%
conversion) during the quantum yield determination, and the rate of consumption of the reactant
was determined from the slope of the line of a plot of the moles of reactant vs. irradiation time.
The photon flux of the lamp was determined using potassium ferrioxalate as the chemical
actinometer.62 The quantum yield for ligand dissociation using 500 nm light was calculated as the
rate of moles of reactant consumed divided by the photon flux and corrected for the mean fraction
of light absorbed by the sample. The quantum yields for 1O2 production (ΦΔ) were measured using
[Ru(bpy)3]2+ as a standard, with ΦΔ = 0.81 in CH3OH, and 1,3-diphenylisobenzofuran (DPBF) as
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
a O2 trap.63 DPBF is emissive in CH3OH, and the resulting product from the reaction of DPBF
1
with O2 is nonemissive. The absorbance values of samples in methanol were matched at the
irradiation wavelength (A = 0.01 at 460 nm) in a 1×1 cm quartz cuvette. The samples were
irradiated in the sample compartment of the fluorimeter (λirr = 460 nm) in the presence of 1.0 μM
DPBF, and the quenching of the DPBF emission (λexc = 405 nm; λem = 479 nm) was monitored as
a function of irradiation time. The quantum yield was determined by comparing the rate of DPBF
emission quenching for the sample with that obtained for the [Ru(bpy)3]2+ standard. The emission
quantum yield was measured using [Ru(bpy)3]2+ in nitrogen sparged H2O as the reference (Φem =
0.042).64
8
ACS Paragon Plus Environment

Page 9 of 48
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Enzyme Inhibition. Thermodynamic and kinetic parameters for inhibition of CTSB were
determined as previously described.65 Briefly, hydrolysis of Z-Arg-Arg-AMC to form the
fluorophore AMC was monitored over time in the presence of compounds 1–7 (0–10 nM).
Solutions of activated CTSB (25 L) were added via multichannel pipet to individual wells
containing Z-Arg-Arg-AMC, inhibitor and buffer that were prepared in the dark. Plates were
immediately loaded into a Tecan plate reader and formation of free AMC over time was
determined using excitation (360 nm) and emission (430 nm) filters. The CTSB stock was diluted
to 16 nM (4nM final concentration) in buffer solution (400 mM sodium acetate, 4 mM EDTA, pH
5.5) containing 8 mM DTT, and activated for 15 min at 37 °C. Inhibitors were prepared as <1%
DMSO stock solutions in assay buffer (400 mM sodium acetate, 4 mM EDTA, 0.01% Triton-X
100, pH 5.5). The substrate Z-Arg-Arg-AMC in assay buffer (200 μM, 50 μL) was diluted to
achieve a final concentration of 100 μM. Data were fit to a two-step irreversible enzyme inhibition
model using Dynafit as previously described.65 Experiments were performed in triplicate; Ki, kinact
and kinact/Ki values are averages of three runs and errors are standard deviations.
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
2D Cell Culture and Cytotoxicity Assays. MDA-MB-231, a well-studied human breast
carcinoma line and MCF-10A, a normal breast epithelial line, were purchased from American
Type Culture Collection (ATCC; Manassas, VA, USA). MDA-MB-231 cells were cultured in
DMEM supplemented with penicillin (10,000 U/mL), streptomycin (10 mg/mL) and 10% fetal
bovine serum. MCF-10A cells were cultured in DMEM supplemented with 5% horse serum, EGF
(20 ng/mL), hydrocortisone (0.5 mg/mL), insulin (10 µg/mL), penicillin (10,000 U/mL), and
streptomycin (10 mg/mL). Cells were incubated at 37°C and kept under 5% CO2 atmosphere. Cells
were cultured in 10 cm2 cell-treated dishes from ThermoFisher Scientific. Cell media was replaced
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 48
1
2
3
4
48h after incubation and cells were removed from the plate and re-seeded 24–48 h after changing
5
6
of the media.
7
8
9
For viability experiments cells were seeded at 5,000cells/well on 96-well plates in 100µL
of the appropriate media. The plates were incubated for 16–18h. After incubation, the media was
aspirated from each well and then the control wells were replaced with 100 µL of fresh media
while the treatment wells were filled with 100 µL of media containing 4–7 (1–10 M). After
addition of the treatment media, plates for irradiation (Light) were transferred to the incubator,
while the plates which would not undergo irradiation were covered in aluminum foil and then
transferred to the incubator (Dark). After a 4 h incubation period the Light plates were removed
and irradiated with an Ealing 250W tungsten halogen lamp with an 8 amp 15V power supply for
20 min. The light beam was filtered through 10 cm of water and the distance between the light
source and the plate was situated so that the entire plate could be irradiated evenly. After
irradiation, the plates were incubated for an additional hour and were then irradiated for an
additional 25 min. The plates were then incubated for 72 h. After incubation, 10 µL of MTT
reagent (5 mg/mL) were added to each well and the plates were incubated at 37 °C for 2h. The
plates were then removed from the incubator and the cell media was carefully aspirated off, as not
to disturb the newly formed formazan crystals. DMSO (100 L) was added to each well and the
plates were rocked for 20 min. Absorbance data were recorded at 570 nm using a plate reader.
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
3D Cell Culture. MDA-MB-231 cells were grown and maintained in phenol red-free DMEM
supplemented with 10% FBS, 10mM HEPES, and 100 U/ml penicillin-streptomycin, pH 7.4. For
experimental conditions, the media was altered to additionally contain 25mM PIPES, and acidified
to pH 6.8. During all conditions, cells were grown in a 37 °C humidified incubator ventilated with
5% CO2.
10
ACS Paragon Plus Environment

Page 11 of 48
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
DQ Collagen-IV 3D Live Cell Proteolysis Assay. Cleavage of DQ-collagen IV substrate (Life
Technologies, Carlsbad, CA, USA) by live MDA-MB-231 cells was assayed in real time and
quantified based on published protocols66-68 with some modifications. Briefly, single cell
suspensions of tumor cells were plated on top of coverslips coated with Cultrex™ (Trevigen,
Gaithersburg, MD, USA) containing DQ-collagen IV (1:40) and overlayed with 2% Cultrex™.
Cells were treated with 4, 5, 6, 7, or vehicle (0.2% DMSO) 24 and 72 h after seeding. With
treatment, cells were incubated under dark and light for 20 min at room temperature, incubated 60
min at 37 °C, exposed to dark or light conditions again for 25 min at room temperature, then placed
back in the incubator. The overlay was replaced with treatment at 72 h. At 96 h, proteolysis of DQ-
collagen IV was imaged by capturing z-stacks through the depth of structures using a Zeiss LSM
780 confocal microscope with a 40x water immersion objective. Intensity of green
fluorescence/tumor spheroid was quantified in each 3D reconstructed spheroid using Volocity
Software (Perkin Elmer, Waltham, MA, USA).
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
3D Cytotoxicity Assay. The cytotoxicity of the compounds was assayed on live MDA-MB-231
cells using Molecular Probes™ Live/Dead Viability/Cytotoxicity Kit (Invitrogen, Carlsbad, CA,
USA). Briefly, single cell suspensions of tumor cells were plated on top of coverslips coated with
Cultrex™ and overlayed with 2% Cultrex™. Cells were treated with 4, 5, 6, 7, or vehicle (0.2%
DMSO) at 24 and 72 h after seeding. With treatment, cells were kept under dark (no irradiation)
and light (250 W, 395–750 nm) conditions for 20 min at room temperature, incubated 60 min at
37 °C, exposed to dark or light conditions again for 25 min at room temperature, then placed back
in the incubator. The Cultrex overlay was replaced and treatments replenished at 72 h. At 96 h,
coverslips were stained with 2 µM Calcein AM and 5 µM Ethidium homodimer-1 (Live/Dead
Viability/Cytotoxicity Kit) for 30 min at room temperature, then rinsed and immediately imaged
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 48
1
2
3
4
5
6
7
8
9
by capturing z-stacks through the depth of structures using a Zeiss LSM 780 confocal microscope
with a 10x water immersion objective. The live cells, stained green by the Calcein AM, were
recorded using excitation at 488 nm, emission at 507 nm. The dead cells, stained red by the
Ethidium homodimer-1, were recorded using excitation at 488 nm, emission at 730 nm. Care was
taken so there was no overlap between emission from red and green channels. Sum of channel
intensity was quantified using Volocity software, and percentage of live cells to total cells was
calculated, indicative of compound toxicity.
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
Fluorescent Cathepsin B Imaging. The ability of 3 to track the localization of CTSB in live
MDA-MB-231 cells was examined by confocal microscopy. Briefly, single cell suspensions of
tumor cells were plated on top of coverslips coated with Cultrex™ and overlayed with 2%
Cultrex™. After 72 h of growth, cells were treated with 3 (5–10 M) or vehicle (0.2% DMSO) for
30 min at 37 °C. During the final 3-5 min, 10 µM of DRAQ5 DNA fluorescent probe was added
(to label the nuclei). Cells were washed with PBS and imaged by capturing z-stacks through the
depth of structures using a Zeiss LSM 780 confocal microscope with a 40x water immersion
objective (Ex. 458 nm/Em. 610 nm). Detection wavelengths of 561–659 nm were used to
determine the localization of CTSB. The intensity of fluorescence/tumor spheroid was quantified
in each 3D reconstructed spheroid using Volocity Software.
Results and Discussion
Compound Design
Dual action conjugates were designed based on known Ru(II) complexes with pyridine as
the leaving ligand (Figure 4). Prior studies revealed that variation of bidentate ligands (NN) in
12
ACS Paragon Plus Environment

Page 13 of 48
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
complexes of the general formula [Ru(tpy)(NN)(py)]2+ , where NN was equal to 2,2'-bipyridine
(bpy), 6,6'-dimethyl-2,2'-bipyridine (Me2bpy), benzo[i]dipyrido[3,2-a:2',3'-c]phenazine (dppn) or
3,6-dimethylbenzo[i]dipyrido[3,2-a:2',3'-c]phenazine (Me2dppn), provided high levels of control
1
over quantum efficiencies of monodentate pyridine photorelease and generation of O2.58,59 The
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
complex [Ru(tpy)(bpy)(py)]2+ represents a control that neither photoreleases pyridine nor produces
1O2. The dppn ligand in [Ru(tpy)(dppn)(py)]2+ produces 1O2 with 98% efficiency due to the lowest
3
energy dppn-centered * excited state, but does not undergo ligand dissociation.58 In contrast,
incorporation of methyl groups that point towards the metal in Me2bpy and Me2dppn provides the
necessary steric bulk to lower the energy of the dissociative 3MC state, promoting ligand release
upon irradiation. Of particular interest was [Ru(tpy)(Me2dppn)(py)]2+, which is able to both
1
photorelease pyridine and generate O2 in water. We surmised that the dual photoreactivity of
[Ru(tpy)(Me2dppn)(py)]2+ could be especially useful in targeting cancer cells for PDT if methods
for cell-specific delivery were developed. In particular, photorelease was postulated to provide an
advantage for achieving cellular uptake and favorable subcellular localization of the Ru(II)
photosensitizer.
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 48
1
2
3
2+
photorelease
2+
4
5
6
7
S
N
N
hn
N
+
Ru
N
N
N
Ru
solvent (S)
N
N
N
N
N
8
N
9
3O2
1O2
[Ru(tpy)(NN)(py)]2+
hn
photosensitization
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
R
R
N
N
N
N
N
N
=
or
N
N
R
R
dppn (R = H)
Me2dppn (R = Me)
bpy (R = H)
Me2bpy (R = Me)
NN
bpy
photorelease
photosensitization
–
+
–
+
–
–
+
+
Me2bpy
dppn
Me2dppn
Figure 4. Dual active Ru(II) complexes of the general formula [Ru(tpy)(NN)(py)]2+ that show
photorelease and 1O2 photosensitization.
CTSB was chosen as a target for cell-specific delivery because it is highly upregulated in
a broad range of cancers (breast, melanoma, GBM, esophageal, pancreatic, colorectal and
prostate), trafficked to the cell surface and secreted by cancer cells and TAMs.69 CTSB is
associated with invasive and metastatic behavior and serves as an essential enzyme for activating
various signaling pathways at tumor sites, including activation of matrix metalloproteases (MMPs)
and urokinase plasmodium activators that cleave the extracellular matrix,1,70,71 as well as an
adaptive response to chemotherapy.20 We chose to use compound 2, a derivative of the potent and
selective CTSB inhibitor CA074,72 as the inhibitor and delivery vector (Figure 3). Derivative 2
(Figure 5) was designed based on NS-134 (Figure 3),61 a CA074 analog that spans the entire active
site cleft of CTSB and has shown promising results as a vector for in vivo delivery of liposomal
14
ACS Paragon Plus Environment

Page 15 of 48
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
nanoparticles to tumor and stromal cells.21 The structure of NS-134 was modified to include Ile
residues and a terminal propyl nicotinamide group for Ru(II) binding to afford 2. In addition,
bipyridyl derivative 1 (Figure 5) was prepared to provide access to an emissive tris-chelated Ru(II)
complex similar to [Ru(bpy)3]2+ for subcellular tracking. This compound was needed because none
of the derivatives from 2 were expected to be luminescent.
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
A series of Ru(II) conjugates (4–7) with the general formula [Ru(tpy)(NN)(2)](PF6)2 were
designed with bidentate ligands NN = bpy (4), Me2bpy (5), dppn (6), and Me2ddpn (7), and their
molecular structures are depicted in Figure 5. In addition, complex 3 was designed to contain a
[Ru(bpy)3]2+ fluorophore to track the subcellular localization of the conjugates, whereas
compounds 4–7 were designed to independently undergo each of the four possible combinations
1
of photoactivated ligand dissociation and/or O2 generation as shown in Figure 4. Compound 7,
containing the Me2dppn ligand, was the only complex expected to exhibit dual behavior, whereas
4–6 would serve as controls for evaluating the role that Ru(II) conjugation, ligand dissociation,
and ROS formation had independently in controlling biological behavior and cytotoxicity. With
each compound, additional control experiments would be performed in cells without irradiation,
which was expected to neither release the Ru(II) fragment nor produce ROS.
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 48
1
2
3
A
4
5
6
7
R
O
O
O
2
H
O
N
H
H
N
H
N
N
N
N
H
H
N
COOH
O
3
H
O
O
8
9
1
2
(R = 2-Pyridyl)
(R = H)
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
B
2+
–
· 2 O2CCF3
N
N
N
N
N
Ru
R
N
D,L-[Ru(bpy)2(1)](O2CCF3)2 (3)
2+
-
. 2 PF6
R
[Ru(tpy)(bpy)(2)](PF6)2 (4)
N
[Ru(tpy)(Me2bpy)(2)](PF6)2 (5)
[Ru(tpy)(dppn)(2)](PF6)2 (6)
[Ru(tpy)(Me2dppn)(2)](PF6)2 (7)
N
Ru
N
N
N
N
O
O
O
2
H
N
H
H
H
N
3
R =
N
N
N
H
H
N
COOH
O
H
O
O
O
Figure 5. Structures of A) epoxysuccinyl inhibitors 1 and 2 and B) Ru(II) epoxysuccinyl
conjugates 3–7.
Compound Synthesis
The syntheses of inhibitors 1 and 2 were accomplished starting from commercially
available 8 (Scheme 1). DCC-HOBT coupling of Z-L-Ile-OH (8) with H-Pro-Ot-Bu (9) gave
dipeptide 10 in 98% yield. Compound 10 was subjected to hydrogenolysis using Pearlman's
16
ACS Paragon Plus Environment

Page 17 of 48
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
catalyst to remove the Cbz protecting group, which was followed by EDC coupling with known
diacid 11, obtained in four steps from D-tartatric acid,73 to give monoacid 12 in a moderate yield
over two steps.
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
Scheme 1. Synthesis of carboxylic acid 12
Synthesis of the pyridyl and bipyridyl linkers 17-18 began from known dipeptide 13
(Scheme 2).74 Dipeptide 13 was subjected to KOH saponification followed by EDC coupling with
ethyl glycinate giving tripeptide 14 in a combined yield of 69% over two steps. Tripeptide 14 was
saponified with KOH and the resultant carboxylic acid was coupled with primary amines 1575 or
1676 using EDC, giving 17 and 18 in 52% and 54% yields over two steps, respectively. Compounds
17 and 18 were subjected to Boc deprotection using 1:1 mixture of TFA: DCM to form TFA salts,
which were combined with a p-nitrophenol ester derived from 12, giving 19 and 20 in moderate
yields (Scheme 3).
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 48
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
Scheme 2. Synthesis of linkers 17 and 18.
18
ACS Paragon Plus Environment

Page 19 of 48
Journal of the American Chemical Society
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
Scheme 3. Synthesis of pyridine-based CTSB inhibitors 19 and 20
The syntheses of [Ru(bpy)2(1)](O2CCF3)2 (3) and [Ru(tpy)(NN)(1)](PF6)2 (4–7) are shown
in Schemes 4A and 4B, respectively. Analog 19 was treated with cis-[Ru(bpy)2Cl2] with heating
at 80 oC in ethanol giving [Ru(bpy)2(1)]Cl2 as an intermediate which was then subjected to t-butyl
ester deprotection using TFA:DCM (1:1) to obtain compound 3 as a mixture of  and 
stereoisomers in 32% yield over the two steps. Counteranion exchange of chloride for
trifluoroacetate occurred during this step through removal of volatile HCl gas. In order to
synthesize ruthenium conjugate complexes 4–7, the respective [Ru(tpy)(NN)Cl]Cl77,78 complexes
o
were converted into [Ru(tpy)(NN)OTf]OTf complexes by overnight heating in ethanol at 80 C
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 48
1
2
3
4
5
6
7
8
9
with AgOTf, followed by filtration over celite to remove AgCl formed in the reaction. Treating
respective [Ru(tpy)(NN)OTf]OTf complexes with 1.1 equiv of inhibitor 2 in EtOH at 80 oC for 6
h, followed by concentration resulted in inhibitor metalation, which upon chromatographic
purification over neutral alumina and subsequent t-butyl ester deprotection using 1:1 TFA:DCM
gave final complexes [Ru(tpy)(NN)(2)](PF6)2 4–7 25% to 34% yield over three steps. 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
1
3–7 were characterized by using UV-Vis absorption, H NMR, and IR spectroscopies, mass
spectrometry and elemental analysis.
Scheme 4. Synthesis of A) ,-[Ru(bpy)2(1)](O2CCF3)2 (3) (A) and B) [Ru(tpy)(NN)(2)](PF6)2
complexes, where NN = bpy (4), Me2bpy (5), dppn (6) and Me2dppn (7), see Figure 4 for bidentate
ligand (NN) structures.
Structural Characterization Data
1H NMR spectroscopic analysis of Ru(II) conjugates 4–7 shows resonances corresponding
1
to the Ru(II) complex and epoxysuccinyl inhibitor fragments of the conjugates. The H NMR
20
ACS Paragon Plus Environment

Page 21 of 48
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
spectrum of compound 7 in methanol-d4 shows an additional peak in the aromatic region at 7.30
ppm which corresponds to the 5-pyridyl C-H of the monodentate nicotinamide donor. The doublet
at 7.93 ppm is assigned to the 4-pyridyl C-H based on its coupling constant. A resonance
corresponding to the 2-pyridyl CH proton is merged with other peaks in the multiplet at ~8.25
ppm. Two singlets at 2.38 and 1.82 ppm, which integrate to three protons each, belong to the two
methyl groups on the Me2dppn ligand that are in unique chemical environments, one above and
one below the plane formed by the Ru(tpy) unit. Similar data were observed for compounds 3–6.
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
Electrospray mass spectra of conjugates 3 and 4 show major peaks at 610.7, along with
expected isotopic distributions, which are consistent with [Ru(bpy)2(1)]2+ and [Ru(tpy)(bpy)(2)]2+
dications derived from 2 and 3, respectively (Figures S41–S44). Similarly, spectra for complexes
5, 6 and 7 show major peaks at m/z values of 624.7, 698.7 and 712.7 respectively, which are
consistent
with
dications
[Ru(tpy)(Me2bpy)(2)]2+,
[Ru(tpy)(dppn)(2)]2+
and
[Ru(tpy)(Me2dppn)(2)]2+(Figures S45–S50).
Complexes 4–7 were evaluated for stability in DMSO at 25 °C and phenol red-free
Dulbecco’s modified Eagle’s cell growth medium (DMEM) at 37 °C over the course of 24 h by
electronic absorption spectroscopy (Figures S58–67). With the exception of 5, which showed
spectral changes consistent with partial release of 2 at 37 °C in DMEM over the course of 24 h, all
complexes were exceptionally stable in the dark, with no observable spectral changes during this
time.
Electronic Absorption Spectroscopy and Photochemistry
The electronic absorption spectra for ruthenium complexes 3–7 are shown in Figure 6.
Complexes 3-7 exhibit maxima between ~ 450 and 490 nm in methanol, consistent with metal-to-
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 48
1
2
3
4
ligand charge transfer (MLCT) bands. The maximum of 4 is observed at 461 nm ( = 8800 M–
5
6
1
1cm–1) and a shoulder is visible at ~415 nm. The MLCT maximum of 5, at 469 nm, is
7
8
9
bathochromically shifted relative to 4, consistent with the corresponding pyridine analogs
previously reported and attributed to the distorted octahedral geometry of the former, thus
destabilizing the HOMO.58 Complex 6 exhibits a 1MLCT maximum at 471 nm ( = 13800 M–1cm–
1) and two peaks at 385 and 405 nm associated with the known dppn-centered 1* transitions.58
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
In a manner similar to the red-shift between complexes 4 and 5, the MLCT maximum of 7 is
observed at 482 nm. This complex also possesses Me2dppn-centered bands at 380 and 402 nm as
previously reported for [Ru(tpy)(Me2dppn)(py)]2+.58
Figure 6. Electronic absorption spectra of 3 (brown), 4 (blue), 5 (green), 6 (black) and 7 (red) in
MeOH.
The irradiation of complexes 5 and 7 with visible light (irr  475 nm) in H2O results in the
release of 2, with the concomitant coordination of solvent to generate the corresponding aqua
complex (Figure 7). Irradiation of 5 in CH3CN results in a blue shift of the 1MLCT transition from
469 nm to 453 nm (Figure S51A), whereas a red shift from 469 nm to 481 nm is observed in H2O,
consistent with the formation of the corresponding [Ru(tpy)(Me2bpy)(L)]2+ (L = CH3CN, H2O)
22
ACS Paragon Plus Environment

Page 23 of 48
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
complex as previously reported for related complexes.65 Similarly, the irradiation of N2 purged
solutions of 7 results in a blue shift of the 1MLCT maximum from 482 nm to 464 nm in CH3CN
and a modest red shift from 482 nm to 487 nm in H2O. The quantum yields of photoinduced ligand
exchange with irr = 500 nm, 500, were determined to be 0.15(1) and 0.11(1) for 5 and 7,
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
respectively, in CH3CN under N2 (Table 1). In H2O, 500 values for 5 and 7 were measured to be
0.092(4) and 0.0070(6), respectively. The lower quantum yields observed in H2O are attributed to
the lower solubility of 2 in this solvent, reducing the escape of 2 from the solvent cage in H2O
relative to CH3CN, as previously observed for related systems.79 To confirm the release of 2 from
5 and 7, changes to the 1H NMR spectrum as a function of irradiation time were recorded in D2O
(irr  475 nm), which resulted in new resonances corresponding to free 2 with the concomitant
appearance
of
resonances
corresponding
to
[Ru(tpy)(Me2bpy)(D2O)]2+
and
[Ru(tpy)(Me2dppn)(D2O)]2+ (Figure S52–S57).80
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 48
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 7. Changes to the electronic absorption spectra of 5 (A) and N2 purged 7 (B) in H2O as a
function of irradiation time ( irr ≥ 475 nm) for 0 – 4 min and 0 – 11 min, respectively.
The presence of the dppn and Me2dppn ligands in 6 and 7, respectively, is known to afford
3
a long-lived dppn-centered * excited state in this class of complexes that is able to undergo
3
energy transfer with ground state O2 in solution.58,59 Therefore, in addition to the release of 2,
1
complex 7 sensitizes O2 with = 0.58(3) with irr = 460 nm. Complex 6, for which ligand
photodissociation is not observed (LE < 10–4), sensitizes 1O2 more efficiently, resulting in  =
1
0.83(3) with irr = 460 nm. The lower yield of O2 sensitization in 7 is attributed to competitive
population of 3MC and 3* states, as previously discussed in detail for the corresponding pyridine
parent complexes, [Ru(tpy)(dppn)(py)]2+ and [Ru(tpy)(diMedppn)(py)]2+.58,59 Complex 3 is
24
ACS Paragon Plus Environment

Page 25 of 48
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
moderately emissive at room temperature with a maximum at 675 nm, luminescence lifetime, em,
of 69.1 ns (exc = 444.4 nm), and emission quantum yield, em, of 0.0029(1) in H2O (exc = 460
nm). In contrast to [Ru(bpy)3]2+, the short excited state lifetime of 3 precludes its bimolecular
sensitization of 1O2, which allows for the use of 3 as a nontoxic fluorescent probe.
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
Table 1. Quantum yields for ligand exchange (500) and O2 production singlet oxygen () by
Ru(II) compounds 3–7.
a
b
c
c
Compound
500
500

--
em
3
4
5
6
7
--
--
0.0031(1)
<10–4
0.092(1)
<10–4
<10–4
0.15(1)
<10–4
--
--
--
--
--
--
0.83(3)
0.0070(6) 0.11(1) 0.58(3)
a In 2% CH3OH/H2O with irr = 500 nm. b In CH3CN with irr = 500 nm. c In CH3OH.
Distribution Coefficients
The cellular uptake and localization of ruthenium complexes rely significantly on their
lipophilicity,81 exhibiting enhanced cellular uptake with an increase in the lipophilic character of
the compounds. Also, the lipophilicity of compounds containing acidic or basic functional groups
varies considerably with pH. The distribution measured between octanol and aqueous buffer at a
particular pH is termed as the distribution coefficient.82 The distribution coefficient between
octanol and buffer at pH = 7.00 preeminently mimics physiological conditions. Therefore, we
evaluated distribution coefficient (logD7.00) for complexes 3–7. As shown in Table 2, the logD7.00
25
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 48
1
2
3
4
5
6
7
8
9
values for 6 and 7 were measured to be –0.11 and –0.06, respectively, and are more lipophilic than
3–5, as expected from the presence of the hydrophobic dppn and Me2dppn ligands in the former.
Contrasting data for [Ru(tpy)(Me2dppn)(py)](PF6)2, with log D7.00 = 0.26, to conjugate 7 reveals
that the CTSB inhibitor plays a significant role in reducing the overall hydrophobicity, driving the
log D7.0 value for 7 to –0.06. The lower logD7.00 value of 7 relative to the py model complex
[Ru(tpy)(Me2dppn)(py)]2+ is expected to favor the targeting of extracellular CTSB. In addition,
these data underscore the importance of controlling the overall charge of the complex; dication
[Ru(tpy)(Me2dppn)(py)]2+ is one order of magnitude less lipophilic than monocation
[Ru(tpy)(Me2dppn)(Cl)]+ (Table 3), which is consistent with literature data.83 Importantly, we do
not yet understand what form the photosensitizer takes after release from 7, i.e. what ligand traps
the vacant coordination site on Ru(II) after photolysis. Given the high extracellular chloride
concentration found in vitro and in vivo (100–140 mM), monocationic chloride derivatives such
as [Ru(tpy)(Me2dppn)(Cl)]+ could be formed in the extracellular space that are taken up by cells.
Indeed, treating [Ru(tpy)(bpy)(H2O)]2+ with 110 mM NaCl in H2O results in formation of
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
+
[Ru(tpy)(bpy)(Cl)] within ~1 h at 37.5 °C (data not shown) and recent data confirm that
[Ru(tpy)(bpy)(Cl)]+ is relatively slow to aquate in aqueous media.84 Treating
[Ru(tpy)(Me2bpy)(H2O)]2+ with 110 mM NaCl in H2O at 37.5 C also results in conversion to the
respective monocationic chloride complex, but with lower efficiency; the steric bulk afforded by
the methyl groups on the bidentate ligand is most likely responsible for the observed shift in the
equilibrium. Similarly, addition of [Ru(tpy)(Me2dppn)(H2O)2]2+ to an aqueous solution containing
110 mM NaCl results in the formation of [Ru(tpy)(Me2dppn)(Cl]+, albeit in low yield (~10%).
Taken together these data strongly suggest that our Ru(II) photosensitizers that undergo chloride
26
ACS Paragon Plus Environment

Page 27 of 48
Journal of the American Chemical Society
1
2
3
4
exchange may accumulate in the mitochondria, which is favored for cations with log P values 1–
5
6
585 and has already been shown for similar Ru(II)-based photosensitizers.86
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
Table 2. Distribution coefficients of complexes 3–7 between octanol and 10 mM sodium
phosphate saline buffer (PBS) measured using “Shake-Flask” method.82
a
Compound
logD7.00
3
4
5
6
7
–1.37 (0.16)
–1.69 (0.01)
–1.12 (0.19)
–0.11 (0.05)
–0.06 (0.02)
[Ru(tpy)(Me2dppn)(py)](PF6)2 0.26 (0.04)
[Ru(tpy)(Me2dppn)Cl]Cl 1.27 (0.06)
aConcentrations of complexes ranged between 55 and 75 M
CTSB Inhibition
Compounds 1–7 were examined for inhibition of purified CTSB. Inhibition was determined
in the dark by progress curve analysis using hydrolysis of the fluorogenic substrate Z-Arg-Arg-
AMC (Table 3). Data were fit to a two-step model for irreversible inhibition, which provides the
equilibrium constant for the reversible association between inhibitor and CTSB (Ki) and the rate
constant for irreversible inactivation (kinact), including covalent bond formation via epoxide
opening. The quotient kinact/Ki represents the second-order rate constant for enzyme inactivation.
Inhibitors 1–7 all potently and irreversibly inhibit CTSB at extremely low concentrations (0.25–
10 nM), with kinact/Ki values ranging from 4.3–6.3  106 M–1 s–1, which are amongst the most
27
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 48
1
2
3
4
5
potent irreversible inhibitors of CTSB known to date.87-91 Taken together, these inhibition data
confirm that a broad range of Ru(II) complexes are tolerated by CTSB.
6
7
8
9
Table 3. Kinetic and thermodynamic parameters for inhibition of CTSB.a
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
kinact (s–1) kinact/Ki (M–1s–1)
Ki (M)
Compound
 10–3
 103
 10–3
1
2
3
4
5
6
7
6.8  0.4
8.5  1.3
4.3  0.8
7.6  0.7
8.8  1.1
7.4  1.2
8.5  1.4
34.5  2.1
44.6  9.7
23.7  3.8
38.0  6.1
37.0  5.6
40.6  12
51.0  11
5100  600
5300  1300
5600  400
5000  1300
4300  800
5500  1400
6300  2100
aConditions: The CTSB stock was diluted to 16 nM (4 nM final concentration) in buffer solution
(400 mM sodium acetate, 4 mM EDTA, pH 5.5) containing 8 mM DTT. Inhibitors were prepared
as <1% DMSO stock solutions in assay buffer (400 mM sodium acetate, 4 mM EDTA, 0.01%
Triton-X 100, pH 5.5) to achieve final concentrations ranging from 0.25-10 nM. The substrate Z-
Arg-Arg-AMC in assay buffer (200 μM, 50 μL) was diluted to achieve a final concentration of
100 μM.
Cell Studies
After confirming that 4–7 potently inhibit purified CTSB, compounds, these complexes were
evaluated against MDA-MB-231 TNBC cells in 2D culture. Cells were treated with 4–7 (1–10
M), incubated for 4 h then irradiated (Light, irr > 395 nm) using a two-stage irradiation sequence
to allow for Ru(II) photodissociation (Stage 1, t = 0–20 min), a 60 min delay to allow uptake of
28
ACS Paragon Plus Environment

Page 29 of 48
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
the photosensitizer, then irradiation for ROS generation (Stage 2, t = 80–105 min). Cells under
dark conditions were treated in the same manner without irradiation. After 72 h, viability was
determined using the MTT assay. Results indicated that no growth inhibitory effects were observed
for compounds 4–7 at concentrations of 1 and 5 M (data not shown). At concentrations of 10 M,
7 was the only complex that reduced viability outside the range of error relative to the control cells
treated with vehicle only (Figure 8). Furthermore, the Me2ddpn derivative 7 showed significantly
enhanced cell killing relative to that observed in the dark, reducing viability to roughly 60% of the
control level under light conditions vs. ~90% in the dark. In order to gain insight into the selectivity
of 7 for cancer vs. non-cancer cell lines, 7 was evaluated against the normal breast epithelial cell
line MCF-10A under the same conditions as those used with the TNBC MDA-MB-231 line.
Viability was determined after 72 h using the MTT assay and results showed that 7 did not reduce
viability in the normal breast epithelial line under light or dark conditions (Figure 8) below the
level of control, confirming that selectivity between cancer vs. non-cancer cell lines can be
obtained.
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
29
ACS Paragon Plus Environment