Identification of novel dynamin-related protein 1 (Drp1) GTPase inhibitors: Therapeutic potential of Drpitor1 and Drpitor1a in cancer and cardiac ischemia-reperfusion injury

Article abstract of DOI:10.1096/fj.201901467R

Mitochondrial fission is important in physiological processes, including coordination of mitochondrial and nuclear division during mitosis, and pathologic processes, such as the production of reactive oxygen species (ROS) during cardiac ischemia-reperfusion injury (IR). Mitochondrial fission is mainly mediated by dynamin-related protein 1 (Drp1), a large GTPase. The GTPase activity of Drp1 is essential for its fissogenic activity. Therefore, we aimed to identify Drp1 inhibitors and evaluate their anti-neoplastic and cardioprotective properties in five cancer cell lines (A549, SK-MES-1, SK-LU-1, SW 900, and MCF7) and an experimental cardiac IR injury model. Virtual screening of a chemical library revealed 17 compounds with high predicted affinity to the GTPase domain of Drp1. In silico screening identified an ellipticine compound, Drpitor1, as a putative, potent Drp1 inhibitor. We also synthesized a congener of Drpitor1 to remove the methoxymethyl group and reduce hydrolytic lability (Drpitor1a). Drpitor1 and Drpitor1a inhibited the GTPase activity of Drp1 without inhibiting the GTPase of dynamin 1. Drpitor1 and Drpitor1a have greater potency than the current standard Drp1 GTPase inhibitor, mdivi-1, (IC50 for mitochondrial fragmentation are 0.09, 0.06, and 10 μM, respectively). Both Drpitors reduced proliferation and induced apoptosis in cancer cells. Drpitor1a suppressed lung cancer tumor growth in a mouse xenograft model. Drpitor1a also inhibited mitochondrial ROS production, prevented mitochondrial fission, and improved right ventricular diastolic dysfunction during IR injury. In conclusion, Drpitors are useful tools for understanding mitochondrial dynamics and have therapeutic potential in treating cancer and cardiac IR injury.

Full text of DOI:10.1096/fj.201901467R

Received: 13 June 2019
Revised: 5 November 2019
Accepted: 19 November 2019
|
|
DOI: 10.1096/fj.201901467R
R E S E A R C H A R T I C L E
Identification of novel dynamin-related protein 1 (Drp1) GTPase
inhibitors: Therapeutic potential of Drpitor1 and Drpitor1a in
cancer and cardiac ischemia-reperfusion injury
Danchen Wu1
Jignesh Patel2
Elahe Alizadeh3
Asish Dasgupta1
Kuang-Hueih Chen1
Monica Neuber-Hess1
|
|
|
|
Timothy E. Hurst2
Jeffrey D. Mewburn1
Patricia D. A. Lima3
|
|
|
|
Ashley Martin1
Michael Wells4
Victor Snieckus2
|
|
|
|
Stephen L. Archer1
1Department of Medicine, Queen's
University, Kingston, ON, Canada
Abstract
2Department of Chemistry, Queen's
University, Kingston, ON, Canada
Mitochondrial fission is important in physiological processes, including coordination
of mitochondrial and nuclear division during mitosis, and pathologic processes,
such as the production of reactive oxygen species (ROS) during cardiac ischemia-
reperfusion injury (IR). Mitochondrial fission is mainly mediated by dynamin-
related protein 1 (Drp1), a large GTPase. The GTPase activity of Drp1 is essential for
its fissogenic activity. Therefore, we aimed to identify Drp1 inhibitors and evaluate
their anti-neoplastic and cardioprotective properties in five cancer cell lines (A549,
SK-MES-1, SK-LU-1, SW 900, and MCF7) and an experimental cardiac IR injury
model. Virtual screening of a chemical library revealed 17 compounds with high
predicted affinity to the GTPase domain of Drp1. In silico screening identified an
ellipticine compound, Drpitor1, as a putative, potent Drp1 inhibitor. We also syn-
thesized a congener of Drpitor1 to remove the methoxymethyl group and reduce
hydrolytic lability (Drpitor1a). Drpitor1 and Drpitor1a inhibited the GTPase activity
of Drp1 without inhibiting the GTPase of dynamin 1. Drpitor1 and Drpitor1a have
greater potency than the current standard Drp1 GTPase inhibitor, mdivi-1, (IC50 for
mitochondrial fragmentation are 0.09, 0.06, and 10 μM, respectively). Both Drpitors
reduced proliferation and induced apoptosis in cancer cells. Drpitor1a suppressed
lung cancer tumor growth in a mouse xenograft model. Drpitor1a also inhibited mi-
tochondrial ROS production, prevented mitochondrial fission, and improved right
ventricular diastolic dysfunction during IR injury. In conclusion, Drpitors are useful
tools for understanding mitochondrial dynamics and have therapeutic potential in
treating cancer and cardiac IR injury.
3Queen's Cardiopulmonary Unit (QCPU),
Translational Institute of Medicine
(TIME), Department of Medicine, Queen's
University, Kingston, ON, Canada
4Office of Partnerships and Innovation,
Queen's University, Kingston, ON, Canada
Correspondence
Stephen L. Archer, Department of
Medicine, Queen's University, Etherington
Hall, Room 3041, 94 Stuart St., Kingston,
ON K7L 3N6, Canada.
Email: stephen.archer@queensu.ca
Funding information
Gouvernement du Canada | Canadian
Institutes of Health Research (CIHR); HHS
| National Institutes of Health (NIH), Grant/
Award Number: NIH-R01-HL071115; HHS
| National Institutes of Health (NIH), Grant/
Award Number: 1RC1HL099462; Canada
Research Chairs (Chaires de recherche du
Canada), Grant/Award Number:
950-229252; Gouvernement du Canada |
Natural Sciences and Engineering Research
Council of Canada (NSERC); Canadian
Vascular Network
Abbreviations: Drp1, dynamin-related protein 1; Drp1 K38A, mutant mouse Drp1 plasmid in which lysine 38 is substituted by alanine; Drp1 WT, wild-type
mouse Drp1 plasmid; Drpitor, Drp1 inhibitor; Fis1, mitochondrial fission 1 protein; GMP-PNP, guanosine 5′-[β,γ-imido]triphosphate; HA, hemagglutinin;
IC50, half maximal inhibitory concentration; IR, ischemia-reperfusion; mdivi-1, mitochondrial division inhibitor 1; MFC, mitochondrial fragmentation
count; MFF, mitochondrial fission factor; MiD49, mitochondrial dynamics protein of 49 kDa; MiD51, mitochondrial dynamics protein of 51 kDa; OMM,
outer mitochondrial membrane; ROS, reactive oxygen species; RV, right ventricle; RVEDP, right ventricular end-diastolic pressure; siDrp1, small
interfering RNA against Drp1.
The FASEB Journal. 2020;34:1447–1464.
wileyonlinelibrary.com/journal/fsb2
© 2019 Federation of American Societies for
Experimental Biology
1447
|

1448
WU et al.
|
K E Y W O R D S
breast cancer, ellipticine, lung cancer, mitochondrial dynamics, mitochondrial division inhibitor 1
(mdivi-1), mitochondrial fission, right ventricle
1
INTRODUCTION
in silico prediction tool to identify putative Drp1 inhibi-
tors. Next, we validated the specificity and effectiveness of
candidate compounds in vitro. Finally, we investigated the
therapeutic efficacy of lead compounds in vivo. We proved
that two analogs of ellipticine, which we named Drp1 in-
hibitor 1 (Drpitor1 and Drpitor1a), are potent and specific
Drp1 inhibitors and showed that Drpitor1a can prevent
right ventricular IR injury and regress human lung cancer
in a xenotransplant model.
|
Mitochondria exist in a dynamic network, continuously
joining together (mitochondrial fusion) and separat-
ing (mitochondrial fission). Fission and fusion, along
with mitochondrial motility, are noncanonical functions
which in aggregate are called mitochondrial dynamics.1
Mitochondrial dynamics are regulated by a variety of en-
zymes, most of which are large GTPases. Mitochondrial
fission is primarily mediated by dynamin-related protein
1 (Drp1). When cytosolic Drp1 is activated, it translocates
to the outer mitochondrial membrane (OMM), where it hy-
drolyzes GTP and multimerizes.1 On the OMM, activated
Drp1 interacts with one or more of its four binding part-
ners, namely mitochondrial dynamics proteins of 49kDa
(MiD49), 51kDa (MiD51), mitochondrial fission 1 protein
(Fis1), and mitochondrial fission factor (MFF).2 These
nuclear-encoded proteins are anchored on the OMM where
they facilitate mitochondrial fission.2 The GTPase activity
of Drp1 is essential for OMM constriction.3,4
2
MATERIALS AND METHODS
Virtual screening of compounds
|
2.1
|
Virtual screening was completed using the published crys-
tal structures for Drp1 (4H1V) and dynamin 1 (5D3Q) with
the screening software PyRx 0.9.1.15 A proprietary chemical
compound library of over 4 000 compounds was screened.
Compounds, in .sdf file format, were converted to .pdbqt file
format by Open Babel in PyRx and used as ligands for virtual
screening.
The current gold standard inhibitor of Drp1 is mitochon-
drial division inhibitor 1 (mdivi-1), which was identified
through screening of a library of 23 000 small molecules
by Cassidy-Stone et al.5 Many groups, including ours, have
used mdivi-1 to understand the physiologic and pathologic
roles of Drp1. Mdivi-1 reduces ischemia-reperfusion (IR)
injury in the kidney6 and heart by inhibiting pathologic fis-
sion, thereby decreasing the production of reactive oxygen
species (ROS).7-9 In addition, mdivi-1 has been studied
in diseases characterized by excessive cell proliferation,
apoptosis resistance, and increased Drp1 expression and/
or activity, such as cancer and pulmonary arterial hyper-
tension.10,11 Coordinated mitochondrial fission that ac-
companies cell division (mitosis) is called mitotic fission
and inhibition of mitotic fission causes cell cycle arrest.10
Mdivi-1 mimics the effects of small interfering RNA tar-
geting Drp1 (siDrp1), inhibiting mitotic fission leading to
cell cycle arrest and apoptosis, ex vivo.1 Likewise, in vivo
therapy with mdivi-1 regresses cancer11,12 and pulmo-
nary arterial hypertension10 in preclinical rodent models.
However, concerns have been raised regarding the spec-
ificity of mdivi-1.13,14 Therefore, we wanted to discover
additional more potent and specific Drp1 GTPase inhibi-
tors that are orally bioavailable and nontoxic for use in the
treatment of cardiovascular diseases and cancer.
2.2
Bioactivity and oral
|
availability of compounds
The bioactivity of the candidate compounds was obtained
from the website Molinspiration (https://www.molin
spiration.com; Cheminformatics, SlovenskyGrob, Slovak
Republic). Lipinski's rule of five was employed to predict the
oral availability of the test compounds.16
2.3
Cell culture
|
The human non-small cell lung cancer cell line, A549, and
human invasive breast carcinoma cell line, MCF7, were pur-
chased from the American Type Culture Collection (ATCC,
Manassas, VA, USA). Human non-small cell lung cancer
cell lines SW 900, SK-MES-1, and SK-LU-1 were a kind
gift from Dr. Xiaolong Yang at Queen's University. All
cells were cultured in Dulbecco's modified Eagle's medium
(DMEM) (D5796, Sigma-Aldrich, Oakville, ON, Canada)
supplemented with 10% fetal bovine serum (12483020,
ThermoFisher Scientific, Burlington, ON, Canada). Cells
were cultured at 37°C in a humidified incubator balanced
with 5% CO2.
The objective of this study is to use knowledge of the
structure and function of Drp1 to find new Drp1 inhibi-
tors which are more potent than mdivi-1. We first used an

WU et al.
1449
|
2.4
Compound preparation for in vitro
and the distribution of these three categories among different
treatment groups was compared.
|
experiments
Candidate compounds were dissolved in anhydrous dime-
thyl sulfoxide (DMSO) (276855, Sigma-Aldrich, Oakville,
ON, Canada), filtered using a 0.20-μm PTFE filter
(SLLG013SL, Millipore Sigma, Etobicoke, ON, Canada),
and added to the culture medium to make a working solu-
tion at designated concentrations. An optimal concentration
was determined and used for inhibition of mitochondrial
fission. Carbonyl cyanide-4-(trifluoromethoxy) phenyl-
hydrazone (FCCP), a mitochondrial oxidative phospho-
rylation uncoupler used to cause Drp1-independent fission,
was dissolved in DMSO to a final concentration of 1μmol.
The final concentration of DMSO in the culture medium
was ≤ 0.1%. Solutions containing equal concentrations of
DMSO were used as controls.
2.6
Immunoprecipitation of Drp1
|
We isolated Drp1 for GTPase activity assays from A549
cells, rather than using recombinant Drp1, which we found
unreliable. Thus, for assessing the effects of putative in-
hibitors, compounds versus vehicle controls were applied
to A549 cells for 20 hours before isolation of total protein.
Protein was extracted using cell lysis buffer (9803, Cell
Signaling Technology, Danvers, MA, USA) followed by
sonication. A total of 800 μg of whole cell protein from
each group was allowed to bind to 2 µg of Drp1 antibody
(611112, BD Transduction Laboratories, San Jose, CA,
USA) at 4°C for 5 hours. This was followed by immu-
noprecipitation with 50 µL of protein A/G-agarose beads
(sc-2003, Santa Cruz Biotechnology, Dallas, TX, USA) at
4°C overnight.
2.5
Mitochondrial morphology analysis
|
Cells were treated with candidate compounds for 4 hours
or with FCCP for 10 minutes prior to mitochondrial mor-
phology imaging. Two objective methods were used to
quantitatively examine mitochondrial morphology: (i)
mitochondrial fragmentation count (MFC); (ii) a machine-
learning approach measuring the percentage area of punc-
tate, intermediate, or filamentous mitochondria. In both
methods, cells were loaded with 20 nM tetramethylrhoda-
mine methyl ester (TMRM, T668, ThermoFisher Scientific,
Burlington, ON, Canada) and were imaged using a Leica
SP8 confocal microscope (Leica Microsystems Inc.,
Concord, ON, Canada). Images were taken under a 1.40
NA, ×63 oil immersion objective with 3× digital zoom
using the Leica LAS X software (λex/em = 561/>575 nm)
(Leica Microsystems Inc., Concord, ON, Canada). All
images were taken by the same microscopist who was
blinded to the treatment groups.
2.7
Immunoprecipitation of dynamin 1
|
A similar strategy was used to isolate another large GTPase,
dynamin 1, allowing us to assess the specificity of putative
Drp1 inhibitors. A549 cells were transfected with 2.5 µg of
wild-type hemagglutinin (HA)-tagged dynamin 1 plasmid
(34682, Addgene, Watertown, MA, USA). Candidate com-
pounds versus vehicle were added to cells 48 hours post-
transfection and cells were collected 20 hours after compound
treatment. Total protein was extracted using cell lysis buffer
(9803, Cell Signaling Technology, Danvers, MA, USA). A
total of 500 μg of whole cell protein from each group was
immunoprecipitated with 10 µL of an anti-HA tag antibody
(3724, Cell Signaling Technology, Danvers, MA, USA) at
4°C for 5 hours. Then the protein-antibody mixture was in-
cubated overnight with 50 µL of protein A/G-agarose beads
(sc-2003, Santa Cruz Biotechnology, Dallas, TX, USA) at
4°C. An empty vector containing HA-tag (53395, Addgene,
Watertown, MA, USA) was used as a negative control.
Mitochondrial fragmentation count (MFC)11 and the per-
centage area of punctate, intermediate, or filamentous mito-
chondria were analyzed, as previously described.17 Briefly,
MFC was calculated as the total mitochondrial number di-
vided by the total mitochondrial area in each confocal mi-
croscopy image. Higher MFC indicates a more fragmented
mitochondrial network. Punctate, intermediate, and fila-
mentous mitochondria were classified manually by their
morphology initially, followed by the application of a ma-
chine-learning algorithm, based on mitochondrial area,
length, and sphericity. Ultimately, these data were expressed
as the percentage area of punctate, intermediate, or filamen-
tous mitochondria, as previously described.17 The percentage
area of each category in each confocal image was calculated
2.8
GTPase activity assay
|
After incubation with the protein-antibody mixture, pro-
tein A/G-agarose beads were centrifuged and washed twice
with RIPA lysis and extraction buffer (89901, ThermoFisher
Scientific, Burlington, ON, Canada) and twice with GTPase
buffer (50 µM Tris-HCl pH 7.5, 2.5 µM MgCl2), and then
incubated with 0.5 mM GTP (602-9999, Expedeon Inc.,
San Diego, CA, USA) at 30°C for 30 minutes. The released

1450
WU et al.
|
free phosphate was quantified using a High Throughput
Colorimetric GTPase assay kit (602-0120, Expedeon Inc.,
San Diego, CA, USA) according to the manufacturer's speci-
fied protocol. OD620 was measured by a SpectraMax M3
microplate reader (Molecular Devices, LLC., San Jose, CA,
USA).
transfection and images were taken 4 hours post−compound
treatment.
2.11
Complex I activity and complex I
|
quantity dipstick assay
Mitochondrial complex I activity was measured 20 hours
post−compound treatment using the Complex I Enzyme
Activity Dipstick Assay Kit (ab109720, abcam, Toronto, ON,
Canada) according to the manufacturer’s protocol. Briefly,
7µg of total protein was loaded to each dipstick and the inten-
sity of the complex I signal was measured using a ChemiDoc
MP system (Bio-Rad Laboratories Ltd., Mississauga, ON,
Canada).
2.9
Immunoblots
|
Aliquots from the protein lysates used for GTPase assay were
also used for immunoblotting. For immunoblotting, 40-60 µg
total protein was electrophoresed on a 4%-12% Bis-Tris
Protein Gel (NP0321, ThermoFisher Scientific, Burlington,
ON, Canada) with 1× MOPS running buffer (NP0001,
ThermoFisher Scientific, Burlington, ON, Canada), followed
by electrotransfer to PVDF blotting membrane (LC2002,
ThermoFisher Scientific, Burlington, ON, Canada). The
gel was then probed with the indicated primary antibodies.
Antibodies against total Drp1 (611112, BD Transduction
Laboratories, San Jose, CA, USA) and dynamin 1 (PA1-660,
ThermoFisher Scientific, Burlington, ON, Canada) were
used. β-actin (A5441, Sigma-Aldrich, Oakville, ON, Canada)
was used as a loading control.
A horseradish peroxidase-conjugated goat anti-rabbit
(NA934, General Electric Healthcare, Milwaukee, WI)
or goat anti-mouse (NA931, General Electric Healthcare,
Milwaukee, WI) secondary antibody was used. Detection
was performed using enhanced chemiluminescence (ECL
Plus, General Electric Healthcare, Milwaukee, WI) and ac-
quired using a ChemiDoc MP system (Bio-Rad Laboratories
Ltd., Mississauga, ON, Canada). Images were analyzed using
ImageJ software (https://imagej.nih.gov/ij/).
2.12
Cell proliferation assay
|
Cell proliferation was measured 48 hours post−compound
treatment using Click-iT EdU Alexa Fluor 488 Imaging
Kit (C10420, ThermoFisher Scientific, Burlington, ON,
Canada) following the manufacturer's protocol. 1.0 × 105
cells were seeded per well in 6-well plates and compounds
were added 24 hours after seeding. EdU was allowed to
incorporate for 10 hours before flow cytometry analysis.
Percentage of cells incorporated with EdU was determined
using a SONY SH800S (SONY Biotechnology, San Jose,
CA, USA) and data were analyzed using FlowJo software
(Becton, Dickinson & Company, Franklin Lakes, NJ,
USA).
2.13
Apoptosis assay
|
2.10
Cell transfection
Cell apoptosis was determined by the Annexin V+ pro-
pidium iodide (PI) method, using Alexa Fluor 488 Annexin
V/DeadCellApoptosiskit(V13241, ThermoFisherScientific,
Burlington, ON, Canada). Cells were collected 48 hours after
compound treatment. The percentage of Annexin V (+) and
PI (+) cells was determined using the SONY SH800S flow
cytometer.
|
To establish whether Drpitor compounds require functional
Drp1 as a target to mediate fusion, we first eliminated endoge-
nous Drp1, using siRNA, and then transfected cells with either
a wild-type Drp1 plasmid or a non-functioning Drp1 mutant
plasmid. A549 cells were seeded in 35-mm glass bottom
dishes (P35G-1.5-14-C, MatTek Corporation, Ashland, MA,
USA) at a density of 3-4 × 104/dish. Drp1 was first knocked
down with small interfering RNA against Drp1 (sense:
2.14
Colony formation assay
|
5′-CCCUAGCUGUAAUCACUAAACUUGA-3′,
anti-
sense:3′-UUGGGAUCGACAUUAGUGAUUUGAACU-5′,
Integrated DNA Technologies, Inc., Coralville, IA, USA).
After 24 hours of knockdown, wild-type mouse Drp1 plas-
mid (34706, Addgene, Watertown, MA, USA) or mutant
mouse Drp1 K38A plasmid (26049, Addgene, Watertown,
MA, USA) was transfected in these cells. Experimental Drp1
inhibitor compounds were added 48 hours post−plasmid
Cells were seeded in 6-well plates and optimal concentra-
tions of compounds were added 24 hours after seeding. After
48 hours of compound treatment, cells were disassociated
with 0.25% trypsin-EDTA (25200114, Gibco, Burlington,
ON, Canada) and reseeded in a 48-well plate at a density of
200 live cells/mL. After 7-10 days of culture, cells were fixed
with 4% paraformaldehyde and then stained with 1% crystal

WU et al.
1451
|
violet (V5265, Millipore Sigma, Etobicoke, ON, Canada).
Images of cell colonies were taken by a Panasonic LX3 digi-
tal camera (Kadoma, Osaka, Japan).
performed. Drpitor1a (0.5 μM) was added to the perfusate
~10 minutes before the first ischemic challenge. The same
volume of DMSO was used as a negative control for the
test compound. RV myocardium was collected at the end of
RV-IR for the measurement of mitochondrial superoxide and
mitochondrial morphology.
3
ANIMAL STUDY
|
3.1
In vivo xenotransplant tumor model
|
3.4
Mitochondrial superoxide imaging and
|
BALB/c Rag2−/−
;
IL2Rgc−/− double knockout mice quantification
were kindly provided by Dr. M Ito, Central Institute for
Experimental Animals, Kawasaki, Japan. They were bred in
the Animal Facility at Queen's University. The animal pro-
tocol (2017-1640) was approved by the University Animal
Care Committee of Queen's University. A xenotransplant
mouse model of human lung cancer was established by sub-
cutaneous injection of A549 cells (4.0 × 106 per tumor) in
the back/flank regions. Drpitor1a was dissolved in DMSO
and mice were treated with 10mg/kg body weight with in-
traperitoneal injections three times a week, starting at day 2.
Equal volumes of solvents were injected as control. When
tumors reached ~5mm in diameter, intratumoral injection
of Drpitor1a or vehicle was also administered three times a
week. Tumor length and width were measured every 2 days.
Animals were euthanized with cervical dislocation under 5%
isoflurane inhalation followed by tumor collection. Blood
samples were collected for liver and kidney toxicity studies
just before euthanasia.
RV tissues collected from Langendorff RV-IR experiments
were cut into pieces of ~1 mm3 and were incubated with
5 μM of MitoSOX Red mitochondrial superoxide indicator
(M36008, ThermoFisher Scientific, Burlington, ON, Canada)
at 37°C for 15 minutes. Tissues were then washed thrice with
HBSS/Ca/Mg buffer before being imaged. Images of RV tis-
sues were taken by a 2-photon microscope (Leica TCS SP8
Confocal, Leica Microsystems Inc., Concord, ON, Canada)
using a 0.85 NA, 40× objective, and 0.75× digital zoom.
Specifications for the 2-photon laser include: 1.44 W at in-
tensity 15.8% (λex/em = 775/580 nm). All images were taken
by the same microscopist who was blinded to the treatment
groups. The intensity of the MitoSOX Red mitochondrial
superoxide indicator was analyzed using the Leica LAS X
software.
3.5
Statistical analyses
|
3.2
Langendorff right ventricle
Quantitative data are presented as mean standard error of
mean (SEM) or box-whisker plots. Student's t-test or Mann-
Whitney U test was used to compare the mean or median
value between two groups as appropriate. One-way ANOVA
was used to compare the means of three or more independent
groups. Two-way ANOVA was used to compare the mean
differences between groups that have two independent vari-
ables. A P value of less than 0.05 was considered statistically
significant.
|
ischemia-reperfusion (RV-IR) model
The Langendorff RV-IR experiment was performed as
previously described.7 Briefly, Sprague Dawley rats were
anesthetized with an intraperitoneal injection of ketamine
(75 mg/kg)/xylazine (10 mg/kg). The hearts were harvested
and perfused with oxygenated Krebs solution at 37°C at a
constant pressure of 85 cmH2O on a Langendorff appara-
tus. A fluid-filled balloon was placed in the RV and was
connected to a pressure transducer (Harvard Apparatus,
Holliston, MA, USA). High fidelity signals were re-
corded using a PowerLab 8/35 data acquisition hardware
4
RESULTS
|
(AD Instruments, Colorado Springs, CO, USA) and ana- 4.1
In silico chemical screen for Drp1
|
lyzed using the LabChart Pro 8 software (AD Instruments, GTPase inhibitors
Colorado Springs, CO, USA).
A proprietary chemical compound library of over 4000
compounds was used for in silico screening. Virtual screen-
ing identified compounds that have specific binding affini-
ties against the GTPase domain of Drp1 protein (PDB ID:
4H1V, https://www.rcsb.org/structure/4H1V). Another
large GTPase with similar structure, dynamin 1 (PDB ID:
5D3Q, https://www.rcsb.org/structure/5d3q), was used as
3.3
IR protocol
|
Hearts were stabilized for 10 minutes before the IR ex-
periment. Two consecutive cycles of 20 minutes of global
ischemia followed by 20 minutes of reperfusion were

1452
WU et al.
|
a negative control to test the specificity of the compounds
for Drp1. For virtual screening, catalytic site residues were
selected based on published crystal structures and gener-
ated into flexible conformations.18 For Drp1, residues in-
cluded Gln34, Ser35, Lys38, Ser39, Ser40, Arg53, Thr59,
Asp146, Gly149, Lys216, Asp218, and Asn246. For dy-
namin 1, residues included Ser45, Arg59, Ser61, Thr65,
Asp136, Asp208, Asn236, and Gln239. The screening grid
was minimized by limiting its external edges to include
flexible residues. After in silico screening, 17 compounds
were selected, based on their high and specific predicted
binding affinities to Drp1 (but not dynamin 1), for in vitro
screening (Figure 1).
17 compounds screened, compounds #12 and #13 showed the
most significant inhibition of mitochondrial fission, evident
as a reduction in MFC, a validated, quantitative, measurement
of mitochondrial fragmentation11 (Figure 2A). However, the
concentration of compound #12, an ellipticine analog that
is hereafter referred to as Drpitor1 (5H,5aH,6H,10bH,11H-
pyrido[4,3-b] carbazole-5,11-dione), was much lower than
that of compound #13 (0.6 μM vs 20 μM). Thus, compound
#12, which was predicted to be a kinase or enzyme inhibitor
(Table 1), was selected for further investigation. Prediction of
oral availability showed it does not violate Lipinski's rule of
5, which indicates it is potentially orally active.16
4.3
Synthesizing a stable and potent
|
4.2
Identifying the lead compound
form of Drpitor1 (Drpitor1a)
|
(Drpitor1) by in vitro screening
Because the methoxymethyl group of the compound #12
(Drpitor 1, Figure 2B) is predicted to be hydrolytically labile,
Drpitor1a was synthesized (Figure 2B). Similar to Drpitor1,
Drpitor1a is also predicted to have at least 1 log order higher
binding affinity to Drp1 than mdivi-1 (Drpitor1 vs mdivi-1:
−8.4 vs −7.2; Drpitor1a vs mdivi-1: −9.1 vs −7.2, Figure 3)
and to be an enzyme inhibitor (Table 1). The predicted interac-
tion between Drpitor1 or Drpitor1a with the GTPase domain
of Drp1 is shown in Figure 2C-F. Comparing the predicted
The inhibitory effect of the compound on the GTPase activity
of Drp1 was studied in the A549 human non-small cell lung
cancer (NSCLC) cell line. These cells have at baseline a frag-
mented mitochondrial network, due to increased Drp1 expres-
sion and activity,11 relative to normal airway epithelial cells.
Mitochondrial morphology was used as an indicator of Drp1
activity in in vitro screening, such that Drp1 inhibitors would
be identified by an inhibition of mitochondrial fission. Of the
FIGURE 1 Identification of 17 candidate compounds from in silico screening. A proprietary chemical compound library of over 4000
compounds was used for in silico screening. Virtual screening identified compounds that have specific binding affinities against the GTPase
domain of Drp1 protein over the GTPase domain of dynamin 1. In silico screening was performed using the screening software PyRx 0.9.1. These
compounds belong to the sulfonamide and indole families

WU et al.
1453
|
binding location of these new compounds to a Drp1 crystal-
bound GTP analog, guanylyl-imidodiphosphate (GMP-
PNP), suggests the binding occurs in the phosphate-binding
region of the binding cavity, ie, the GTPase domain of Drp1
(Figure 2G-I, Supplemental Figure 2A,B).
4.4
Drpitor1 and Drpitor1a inhibit
|
mitochondrial fission
Like mdivi-1, Drpitor1 and Drpitor1a both inhibited mito-
chondrial fission in A549 cells in a dose-dependent manner,
1.6
(A)
1.4
1.2
1.0
**
*
*
0.8
0.6
0.4
0.2
0.0
(B)
(C)
(D)
ꢀƌƉŝƚŽƌϭ
(E)
(G)
(F)
(H)
ꢀƌƉŝƚŽƌϭĂ
ŵĚŝǀŝͲϭ
(I)
ꢀƌƉŝƚŽƌϭĂ
ꢀƌƉŝƚŽƌϭ
'DWͲWEW

1454
WU et al.
|
FIGURE 2 Identification, synthesis, and protein interaction of Drpitor1 and Drpitor1a. A, Mitochondrial fragmentation count (MFC) of
mdivi-1 and test compounds on A549 cells showing significant inhibition of mitochondrial fragmentation achieved by compound #12 (Drpitor1)
and compound #13 treatment. Data collected from three independent experiments. Data are represented as mean SEM. *P < .05 vs Ctrl,
**P < .01 vs Ctrl. B, Synthesis of Drpitor1 and Drpitor1a. Final compound Quinone Drpitor1a was prepared with a slight modification of our
earlier method reported,26 iso-nicotinamide (I) and indole 3-carboxaldehyde (II). Thus, lithiation of compound I with s-BuLi under standard
conditions followed by quench with aldehyde II gave carbinol intermediate III (43-61% yield). Compound III was further subjected to lithiation
and an intramolecular cyclization sequence followed by aerial oxidation which occurred upon work-up to give methoxymethyl (MOM) protected
Quinone IV (Drpitor1) in 64%-78% yield. The deprotection of MOM group with con. HCl afforded final Quinone V (Drpitor1a) in low yield. C,
Chemical structure of compound Drpitor1. D, Predicted interaction between Drpitor1 and the GTPase domain of Drp1. E, Chemical structure of
compound Drpitor1a. F, Predicted interaction between Drpitor1a and the GTPase domain of Drp1. G, Chemical structure of mdivi-1. H, Predicted
interaction between mdivi-1 and the GTPase domain of Drp1. I, Overlaid image showing that the binding locations of Drpitor1 (red), Drpitor1a
(blue), and guanylyl-imidodiphosphate (GMP-PNP) (orange and green) are all in the GTPase domain of Drp1
as shown by a decrease in MFC (Figure 4A,D). Drpitor1
and Drpitor1a also increased percentage area of filamen-
tous mitochondria, as measured by an unbiased machine
learning technique (Figure 4B,C). The half maximal inhibi-
tory concentrations (IC50) of mitochondrial fragmentation
for mdivi-1, Drpitor1, and Drpitor1a in A549 cells are 10,
0.090, and 0.057 μM, respectively (Figure 4D), showing
the higher potency of Drpitor1 and Drpitor1a. Furthermore,
Drpitor1a also inhibited mitochondrial fission in three
different NSCLC cell lines (SW 900, SK-MES-1, and
SK-LU-1), independent of P53 and KRAS status, as
well as in a breast cancer cell line, MCF7 (Figure 4E-H,
Supplemental Table 1).
TABLE 1 Molinspiration bioactivity score showing compounds
Drpitor1 and Drpitor1a are predicted to be kinase inhibitors and
enzyme inhibitors
Molinspiration bioactivity
score
Drpitor1
Drpitor1a
GPCR ligand
Ion channel modulator
Kinase inhibitor
−0.01
0.03
0.49
−0.01
0.21
0.50
Nuclear receptor ligand
Protease inhibitor
Enzyme inhibitor
−0.19
−0.13
0.52
−0.25
−0.19
0.57
a mutant mouse Drp1 plasmid (K38A), in which lysine 38 is
substituted by alanine, resulting in a non-functional GTPase3
(Supplemental Figure 2C). Drpitor1 and Drpitor1a signifi-
cantly inhibited mitochondrial fission in siDrp1-treated,
wild-type Drp1-rescued A549 cells (Figure 5E,G). However,
in cells transfected with siDrp1 and mutant Drp1 plasmid
(Drp1 K38A), the mitochondrial network was fused and
could not be further elongated by Drpitor1 or Drpitor1a, re-
flecting the loss of the target that these compounds act on,
namely a functional Drp1 (Figure 5F-G).
4.5
Drpitor1 and Drpitor1a are selective
|
Drp1 GTPase inhibitors
To investigate whether the observed Drpitor-induced mito-
chondrial fusion results from inhibition of the GTPase activ-
ity of Drp1, total Drp1 from A549 cells treated with Drpitor1
or Drpitor1a was immunoprecipitated (Figure 5A) and the
GTPase activity of Drp1 was measured. Both Drpitor1 and
Drpitor1a significantly inhibited the GTPase activity of Drp1
(Figure 5B) without changing the expression level of total
Drp1 (Supplemental Figure 2A).
To determine the selectivity of these new compounds,
we evaluated their ability to inhibit the GTPase activ-
ity of dynamin 1. Hemagglutinin (HA)-tagged dynamin
1 was immunoprecipitated from A549 cells transfected
with HA-tagged dynamin 1 plasmid and treated with
Drpitor1 or Drpitor1a. Drpitor1 and Drpitor1a did not in-
hibit the GTPase activity of dynamin 1 (Figure 5C-D) nor
did they change the expression level of total dynamin 1
(Supplemental Figure 2B).
To overcome the problem of a maximally fused mito-
chondrial network, caused by Drp1 knockdown, we next
induced fission by a Drp1-independent mechanism. We
added carbonyl cyanide-4-(trifluoromethoxy) phenylhydra-
zone (FCCP), a mitochondrial oxidative phosphorylation
uncoupler, to cells treated with siDrp1. FCCP caused mi-
tochondrial fission in Drp1 knocked-down A549 cells in a
Drp1-independent manner19 (Figure 5H). Drpitor1a did not
rescue mitochondrial fission in these FCCP-treated Drp1
knocked-down cells (Figure 5H), further indicating that the
inhibition of mitochondrial fission by Drpitor1a is solely
Drp1 dependent.
To further confirm the selectivity of these two compounds
on the GTPase domain of Drp1, endogenous Drp1 was first
knocked down in A549 cells using siDrp1. Cells were then
transfected with either a wild-type mouse Drp1 plasmid or
Although a previous study has demonstrated that mdivi-1
can inhibit mitochondrial complex I activity,13 neither mdivi-1
nor Drpitor1a changed the mitochondrial complex I activity

WU et al.
1455
|
FIGURE 3 Predicted binding
energies of compounds Drpitor1, Drpitor1a,
and mdivi-1 to Drp1 and dynamin 1. These
numbers indicate the binding between the
compound and Drp1 quantitatively. The
lower the number, the higher the binding
affinity
in A549 cells (Figure 5I). Silencing Drp1 did not change
mitochondrial complex I activity, either (Supplemental
Figure 2D). Taken together, these results indicate that
Drpitor1 and Drpitor1a inhibit mitochondrial fission by spe-
cifically acting as Drp1 GTPase inhibitors.
4.8
Drpitor1a preserves RV diastolic
|
function in an RV-IR model
Drp1-mediated mitochondrial fission, resulting in ROS genera-
tion and calcium overload, plays an important role in the patho-
genesis of cardiac IR injury. In the right ventricle (RV), IR
results in elevated diastolic pressures, reflecting diastolic dys-
function. Previous studies showed that the inhibition of Drp1
GTPase activity using mdivi-1 or inhibition of the interaction
between Drp1 and Fis1 achieved using a competitive inhibitor
peptide of this interaction, P110, prevented IR-induced myo-
cardial calcium overload and RV diastolic dysfunction.7 Here,
we investigated the cardioprotective effect of Drpitor1a on
RV-IR injury, using the RV Langendorff ex vivo model. A test
compound was added to the perfusate prior to IR. Drpitor1a-
treated hearts did not manifest the increase in right ventricular
end-diastolic pressure (RVEDP) that was seen in control hearts
after two periods of IR challenge (Figure 8A-C). After IR, RV
myocardium was collected and mitochondrial superoxide was
measured. Drpitor1a-treated RVs had lower MitoSOX intensity
during IR compared to control RV, reflecting less production
of mitochondrial-derived ROS (Figure 8D-F). Furthermore,
Drpitor1a also inhibited Drp1-mediated mitochondrial fission
of cardiomyocytes during RV-IR compared to DMSO con-
trol, as shown by the increased mean mitochondrial volume
(Figure 8G-H).
4.6
Drpitor1 and Drpitor1a inhibit cell
|
proliferation and induce apoptosis
Both Drpitor1 and Drpitor1a significantly inhibited cell prolif-
eration in A549 cells after 48 hours of treatment (Figure 6A).
Drpitor1a inhibited A549 cell proliferation in a dose-
dependent manner (Supplemental Figure 3). Drpitor1a also
inhibited cell proliferation in SK-MES-1, SK-LU-1, SW
900, and MCF7 cell lines (Figure 6B-E). Colony formation
assay indicated that Drpitor1a inhibited cell survival in A549,
SK-MES-1, SK-LU-1, SW 900, and MCF7 cell lines
(Figure6F-I). Furthermore, bothDrp1inhibitorsincreasedun-
stimulated apoptosis in A549 cells, as shown by a significant
increase in the percentage of cells with positive Annexin V
staining after 48 hours of treatment (Figure 6J). Drpitor1a
also induced unstimulated apoptosis in SK-MES-1, SK-LU-1,
and SW 900 cell lines (Figure 6K-M).
4.7
Drpitor1a inhibits tumor growth in a
|
mouse xenograft model
5
DISCUSSION
|
In order to evaluate the antiproliferative effect of Drpitor1a
in vivo, a xenograft model of lung cancer was developed in
BALB/c Rag2-/-; IL2Rgc-/- mice. After the subcutaneous injec-
tion of tumor cells, Drpitor1a was injected intraperitoneally
and intratumorally. By the end of the study, Drpitor1a had sig-
nificantly inhibited tumor growth, resulting in a 43.8 % reduc-
tion in tumor volume and a 43.9 % reduction in tumor weight
(Figure 7A-C). Drpitor1a did not result in significant weight
loss, change in tumor density or, liver/kidney toxicity during
22 days of therapy (Figure 7D-F, Supplemental Figure 4).
This study identifies two novel pharmacological inhibitors
of Drp1 GTPase activity, Drpitor1 and Drpitor1a. We dem-
onstrate the mechanism of action and specificity of these
compounds and show their therapeutic efficacy in vitro and
in vivo. The study has four novel findings:
1. Identification of 17 candidate Drp1 inhibitors through
screening of a proprietary chemical compound library
of over 4 000 compounds using in silico analysis.

1456
WU et al.
|
(A)
(B)
(C)
(D)
(F)
(E)
(H)
(G)
FIGURE 4 Drpitor1 and Drpitor1a inhibited mitochondrial fragmentation in NSCLC and breast cancer cells. A, Representative images by
confocal microscopy showing elongation of mitochondria after treatment with mdivi-1 (25 µM), Drpitor1 (1.0 µM), or Drpitor1a (0.5 µM). Red:
TMRM. Scale bar = 10 µm. B, Representative images of the classification of mitochondrial morphology. Red, punctate; green, intermediate;
purple, filamentous. Scale bar = 10 µm. C, Percentage of elongated mitochondria was significantly increased after treatment with compounds.
Data collected from three independent experiments. Data are represented as mean SEM. *P < .05 vs Ctrl, **P < .01 vs Ctrl. D, Dose-response
curve showing the percentage of maximal inhibition of MFC by mdivi-1, Drpitor1, and Drpitor1a at different concentrations. Data collected from
three independent experiments. Data are represented as mean SEM. E, Drpitor1a significantly inhibited mitochondrial fission in SK-MES-1
cells, a NSCLC cell line. n = 15 per group; *P < .05. F, Drpitor1a significantly inhibited mitochondrial fission in SK-LU-1 cells, a NSCLC cell
line. n = 15 per group; *P < .05. G, Drpitor1a significantly inhibited mitochondrial fission in SW 900 cells, a NSCLC cell line. n = 15 per group;
*P < .05. H, Drpitor1a significantly inhibited mitochondrial fission in MCF7 cells, an invasive breast cancer cell line. n=15 per group; *P < .05

WU et al.
1457
|
(C)
(A)
(B)
(D)
(E)
(F)
(G)
(H)
(I)

1458
WU et al.
|
FIGURE 5 Drpitor1 and Drpitor1a are specific Drp1 GTPase inhibitors. A, Immunoblot showing the immunoprecipitation of human Drp1
in A549 cells. B, GTPase activity of Drp1 in A549 cells was significantly decreased after treatment with mdivi-1 (50 µM), Dyngo4a (100 µM),
Drpitor1 (2.0 µM), or Drpitor1a (0.5 µM). n = 3 per group. Data are represented as mean SEM. ****P < .0001. C, Immunoblots showing the
immunoprecipitation of human dynamin 1 in dynamin 1-overexpressed A549 cells. A549 cells overexpressed with empty plasmid containing
HA-tag were used as negative control. D, GTPase activity of dynamin 1 from A549 cells was significantly decreased in cells treated with
Dyngo4a (100 µM), but not decreased in cells treated with mdivi-1 (25 µM), Drpitor1 (1.0 µM), or Drpitor1a (0.5 µM). n = 3-6 per group. Data are
represented as mean SEM. **P < .01, ****P < .0001, ns, not significant. E, Representative images by confocal microscopy showing elongation
of mitochondria in siDrp1+Drp1 WT transfected cells after treatment with mdivi-1 (25 µM), Drpitor1 (1.0 µM), or Drpitor1a (0.25 µM). White:
TMRM. Scale bar = 10 µm. F, Representative images by confocal microscopy showing no elongation of mitochondria in siDrp1+Drp1 K38A
transfected cells after treatment with mdivi-1 (25 µM), Drpitor1(1.0 µM), or Drpitor1a (0.25 µM). White: TMRM. Scale bar = 10 µm. G, MFC was
significantly reduced by mdivi-1, Drpitor1, or Drpitor1a in cells transfected with siDrp1+Drp1 WT, but not in cells transfected with siDrp1+Drp1
K38A. n = 15-20 per group. Data are represented as mean SEM. *P < .05, **P < .01, ***P < .001. H, MFC was significantly reduced after
knocking down Drp1. FCCP increased MFC in Drp1-silenced A549 cells. Drpitor1a did not inhibit mitochondrial fission in siDrp1+FCCP-treated
cells. n = 15 per group. Data are represented as mean SEM. *P<.05, ns, not significant. I, Quantification of band density from complex I activity
assay normalized by Complex I quantity assay. Data are represented as mean SEM from three independent experiments. ns, not significant
2. Demonstration that specific, inhibitory effects of
Drpitor1 and Drpitor1a on Drp1 GTPase activity are re-
sponsible for the observed inhibition of mitochondrial
fission, cell proliferation, and induction of apoptosis in
lung cancer cells.
3. Demonstration of the therapeutic efficacy of Drpitor1a in
a xenograft murine model of lung cancer and demonstra-
tion that tumor regression occurs without hepatic or renal
toxicity.
4. Demonstration that Drpitor1a reduces ROS production
and is cardioprotective in an RV-IR model.
relocalization.3,4 Therefore, targeting Drp1's GTPase activity
should result in the inhibition of Drp1 on the OMM which
in turn should inhibit mitochondrial fission. In this study,
we identified and validated two new compounds, Drpitor1
and Drpitor1a, in two distinct fissogenic conditions−a hy-
perproliferative disease marked by increased mitotic fission
(cancer) and a cell injury model, in which fission causes cell
injury through ROS production (RV-IR).
In order to identify small molecule compounds that
inhibit the GTPase activity of Drp1, we performed extensive
in silico screening based on the crystal structure of Drp1
protein and identified 17 candidate compounds, belonging
to sulfonamide and indole families, with selective binding
affinity to the GTPase domain of Drp1 (Figure 1). In vitro
screening, based on the inhibition of mitochondrial fission
in A549 cells, identified two compounds (compound #12,
ie, Drpitor1, and compound #13) with significant inhibitory
effects on mitochondrial fission (Figure 2A). The effective
concentration of Drpitor1 was over 20-fold lower than that
of compound #13 (0.6 μM vs 20 μM), likely reflecting its
higher binding affinity to Drp1 (Figure 2A). Therefore,
Drpitor1 was chosen for further in vitro and in vivo studies.
However, subsequent testing revealed a time-dependent loss
of the pharmacological activity of Drpitor1 (compound #12)
during compound storage. This was determined to be the re-
sult of decomposition due in part to the hydrolytically labile
methoxymethyl group. To overcome the instability problem
of Drpitor1, a new compound not bearing the methoxymethyl
group, Drpitor1a, was synthesized. Drpitor1a is also pre-
dicted to be a selective GTPase inhibitor of Drp1, as shown
by its much higher affinity to Drp1 over dynamin 1 (Figure 3).
Molinspiration (https://www.molinspiration.com) pre-
dicted that both Drpitors are kinase inhibitor or enzyme
inhibitors (Table 1), consistent with findings in our study.
Neither Drpitor1 nor Drpitor1a violates Lipinski's rule of
The significance of mitochondrial fission is contextual.
Mitochondrial fission and cell cycle progression are highly
coordinated, which ensures equitable distribution of mitochon-
dria to daughter cells.11,20 Mitotic fission is coordinated with
the cell cycle through mediators including ERK1/217 and cyclin
B1/CDK1, which activate Drp1 by phosphorylating Drp1 at
serine 616.1 Drp1 is the key regulator of mitochondrial fission.
Molecular silencing (by siRNA) or pharmacological inhibition
(by a small compound mdivi-1) of Drp1 inhibits mitotic mito-
chondrial fission and leads to cell cycle arrest and apoptosis.10,11
Mitochondrial fission is also critical to many pathologic
processes, including cardiac IR injury, which is commonly
seen in patients with myocardial infarction receiving revascu-
larization treatments. In animal models of pulmonary arterial
hypertension, IR is seen in the RV and this RV-IR injury re-
sults in a deleterious increase in ROS production and calcium
overload.21 Inhibition of Drp1 activity or the interaction be-
tween Drp1 and its binding partner Fis1 is cardioprotective
during experimental IR.7,8
During mitochondrial fission, activated Drp1 moves
from the cytosol to the OMM where it interacts with its
binding partners.2 The GTPase activity of Drp1 is critical
for the constriction and tubulation of the OMM after Drp1

WU et al.
1459
|
5, which suggests they are orally bioavailable; however,
oral bioavailability was not directly assessed in this ini-
tial study. Drpitor1 and Drpitor1a are predicted to be more
potent than mdivi-1 based on their higher binding affinity
(Table 1). Consistent with this, we observed significant in-
hibition of mitochondrial fission at sub-micromolar Drpitor1
(C)
(D)
(A)
(E)
(B)
(F)
(G)
(H)
(I)
(L)
(K)
(J)
(M)

1460
WU et al.
|
FIGURE 6 Drpitor1 and Drpitor1a inhibited cell proliferation and induced unstimulated apoptosis in NSCLC and invasive breast cancer
cells. A, The percentage of EdU-positive A549 cells was significantly reduced after treatment with Drpitor1 (0.75 µM) or Drpitor1a (0.5 µM).
n = 3-6 per group. Data are represented as mean SEM. ****P < .0001. B, The percentage of EdU-positive SK-MES-1 cells was significantly
reduced after treatment with Drpitor1a (0.5 µM). n = 3 per group. Data are represented as mean SEM. ***P < .001, ****P < .0001. C, The
percentage of EdU-positive SK-LU-1 cells was significantly reduced after treatment with Drpitor1a (0.5 µM). n = 3 per group. Data are represented
as mean SEM. ****P < .0001. D, The percentage of EdU-positive SW 900 cells was significantly reduced after treatment with Drpitor1a
(0.5 µM). n = 3 per group. Data are represented as mean SEM. ****P < .0001. E, The percentage of EdU-positive MCF7 cells was significantly
reduced after treatment with Drpitor1a (0.5 µM). n = 3 per group. Data are represented as mean SEM. ***P < .001, ****P < .0001. F, Colony
formation assay showing significantly fewer colonies formed in Drpitor1a (0.5 µM)-treated A549 cells. n = 6 per group. Data are represented as
mean SEM. ****P < .0001. G, Colony formation assay showing significantly fewer colonies formed in Drpitor1a (0.5 µM)-treated SK-MES-1
cells. n = 6 per group. Data are represented as mean SEM. ****P < .0001. H, Colony formation assay showing significantly fewer colonies
formed in Drpitor1a (0.5 µM)-treated SK-LU-1 cells. n = 6 per group. Data are represented as mean SEM. ****P < .0001, ns, not significant.
I, Colony formation assay showing significantly fewer colonies formed in Drpitor1a (0.5 µM)-treated MCF7 cells. n = 6 per group. Data are
represented as mean SEM. **P < .01,****P < .0001. J, Baseline apoptosis rate was significantly higher in A549 cells treated with Drpitor1
(1.0 µM) or Drpitor1a (0.5 µM), as identified by the percentage of Annexin V(+)/PI(−) cells. n = 3 per group. Data are represented as mean
SEM. **P < .01, ****P < .0001. K, Baseline apoptosis rate was significantly higher in SK-MES-1 cells treated with Drpitor1a (0.25 µM), as
identified by the percentage of Annexin V(+)/PI(−) cells. n = 5 per group. Data are represented as mean SEM. ****P < .0001. L, Baseline
apoptosis rate was significantly higher in SK-LU-1 cells treated with Drpitor1a (0.25 µM), as identified by the percentage of Annexin V(+)/PI(−)
cells. n = 5 per group. Data are represented as mean SEM. *P < .05. M, Baseline apoptosis rate was significantly higher in SW 900 cells treated
with Drpitor1a (0.5 µM), as identified by the percentage of Annexin V(+)/PI(−) cells. n = 5 per group. Data are represented as mean SEM.
****P < .0001
doses (0.25 μM -1.0 μM for Drpitor1, 0.1 μM-0.5 μM for
Drpitor1a), whereas mdivi-1 inhibited mitochondrial
fission only at significantly higher doses (12.5 μM -25 μM)
(Figure 4D). This higher potency is also reflected by the
compounds’ respective IC50 for mitochondrial fragmentation
(mdivi-1, Drpitor1, and Drpitor1a are 10 μM, 0.090 μM and
0.057 μM, respectively).
Mdivi-1 inhibits the GTPase activity of the yeast ortho-
log of Drp1.5,13 Consistent with this, we demonstrate the in-
hibitory effects of mdivi-1 on the GTPase activity of human
Drp1 (Figure 5A,B). Like mdivi-1, Drpitor1 and Drpitor1a
inhibit Drp1 GTPase but not the GTPase activity of dynamin
1 (Figure 5C,D). This is an important demonstration of the
specificity of these new compounds since dynamin 1 shares
42% homology with Drp1.
Previous research has shown that mdivi-1 or a small dis-
ruptive peptide that blocks Drp1-Fis1 interaction, P110, re-
duces tumor growth and decreases cardiac IR injury.7,11 In
accordance to these findings, we observed significant sup-
pression of tumor growth in a preclinical model of cancer
(Figure 7). PAH and lung cancer share many similarities
including a hyperproliferative and apoptosis-resistant phe-
notype. While abnormalities of mitochondrial metabolism,
notably a shift to aerobic glycolysis called the Warburg phe-
nomenon, contribute to the proliferation/apoptosis imbalance
in cancer, this study focuses on developing therapeutic agents
that target another, more recently recognized abnormality,
namely dysregulation of mitochondrial dynamics. Previous
studies have explored the concept that dysregulated mito-
chondrial dynamics caused by increased activity of the fis-
sion mediator Drp1, contributing to the hyperproliferative
phenotype of cancer and PAH cells.10,11 Inhibition of Drp1
triggers a cell cycle checkpoint and results in cell cycle ar-
rest and induction of apoptosis. Thus, Drp1 is an appealing
therapeutic target, inhibition of which by a pharmacological
inhibitor mdivi-1 has been shown to suppress cell prolifera-
tion and induce apoptosis in both lung cancer cells and PAH
PASMCs.10,11
We further investigated the therapeutic efficacy of
Drpitor1a in a myocardial IR rodent model. In this model,
cardiomyocyte Drp1 is pathologically activated resulting in
excessive mitochondrial fission and mitochondrial ROS gen-
eration.21 We have shown previously that there is a basal frag-
mentation of mitochondria in RV myocytes of monocrotaline
rats with pulmonary arterial hypertension (PAH), even in the
absence of induced IR.7 This suggests the RV dysfunction,
which is commonly the cause of mortality in PAH,22 may
in part be due to mitochondrial fission. Our current results
confirm prior studies and demonstrates that inhibiting mito-
chondrial fission in RV cardiomyocytes by targeting Drp1
decreases mitochondrial ROS production and is cardiopro-
tective ex vivo, evident as a preservation of RV diastolic func-
tion7,8,23 (Figure 8).
6
LIMITATIONS
|
After completing our initial study, we discovered that
Drpitor1a has the same chemical structure as ellipticine
quinone.24 The in vitro antiproliferative effect of ellipticine
quinone has been reported in human cancer cell lines, includ-
ing HeLa cells,25 HCT-116, and HL-60 cells.24 Our findings

WU et al.
1461
|
(A)
(B)
(C)
(D)
(F)
(E)
FIGURE 7 Drpitor1a suppressed tumor development in a mouse xenograft model of lung cancer. A, Representative images showing
decreased tumor size in the Drpitor1a-treated group compared to the control group. Scale bar = 1 cm. B, Tumor growth rate was lower in the
Drpitor1a-treated group. n = 12 for Ctrl, n = 7 for Drpitor1a. Data are represented as mean + SEM for Ctrl and mean - SEM for Drpitor1a,
respectively. C, Drpitor1a-treated group showed a significant reduction in tumor weight compared to the control group. n = 12 for Ctrl, n = 7 for
Drpitor1a. Data are represented as mean SEM. *P < .05 vs Ctrl. D, Drpitor1a-treated mice did not have significantly higher weight loss than
the control mice. n = 12 for Ctrl, n = 7 for Drpitor1a. Data are represented as mean SEM. E, Drpitor1a did not cause more severe liver function
damage than the control group, as shown by albumin, total bilirubin, and cholesterol levels. n=3-4 per group. Data are represented as mean SEM.
ns, not significant. F, Drpitor1a did not cause more severe kidney function damage than control group, as shown by the blood urea nitrogen (BUN)
level. n = 3-4 per group. Data are represented as mean SEM. ns, not significant
are consistent with these studies and we have extended our
in vitro findings to the animal model of cancer and demon-
strated the mechanism of action of the compound to be inhi-
bition of Drp1.
The superior potency of Drpitor1a versus mdivi-1 on
mitochondrial fission was not assessed in the GTPase assay
nor did we prove that the enhanced potency translates into
therapeutic superiority of these new compounds versus
mdivi-1.
Although Drpitor1a significantly reduced the increase of
RVEDP and mitochondrial ROS production during reperfu-
sion by inhibition of Drp1 activity (Figure 8), other poten-
tially beneficial molecular mechanisms of attenuating IR
injury, such as reduction of cardiomyocyte calcium overload,
were not studied.
Finally, we have not yet assessed the therapeutic efficacy
of Drpitor1a on RV function and pulmonary hemodynamics
in a preclinical model of PAH.

1462
WU et al.
|
(B)
(A)
(C)
(D)
(F)
(E)
(G)
(H)
FIGURE 8 Drpitor1a prevents RV ischemia-reperfusion (IR) injury. A, Summary of the RVSP and RVEDP from the IR experiments with
control hearts. n=5-6 per group. Data are represented as mean SEM. B, Summary of the RVSP and RVEDP from the IR experiments with
Drpitor1a-treated hearts. n=5-6 per group. Data are represented as mean SEM. C, Drpitor1a significantly reduced the elevation of RVEDP
after two periods of IR challenge compared to DMSO treatment. n = 5-6 per group. Data are represented as box-whisker plot. ***P < .001.
D, Representative confocal microscopy images indicating that Drpitor1a reduced mitochondrial superoxide production induced by ischemia-
reperfusion challenge. Red: MitoSOX red mitochondrial superoxide indicator. Scale bar = 100 µm. E, Representative images showing the
quantification of mitochondrial superoxide level. Blue: +, Green: ++, Yellow: +++, Orange: ++++, Red: +++++. Scale bar = 100 µm. F,
Drpitor1a significantly inhibited mitochondrial superoxide production as shown by increased area of lower MitoSOX intensity and reduced area
of high MitoSOX intensity. n = 10-20 per group. Data are represented as mean SEM. ****P < .0001. G, Representative images showing
fragmented mitochondria in DMSO-treated IR heart compared with no IR heart, Drpitor1a, inhibited mitochondrial fragmentation during IR. Scale
bar = 25 µm. H, Drpitor1a significantly inhibited mitochondrial fission in RV cardiomyocytes as shown by increased mean mitochondrial volume
compared to DMSO. n = 10-20 per group. Data are represented as mean SEM. *P < .05

WU et al.
1463
|
4. Yoon Y, Pitts KR, McNiven MA. Mammalian dynamin-like
protein DLP1 tubulates membranes. Mol Biol Cell. 2001;12:
2894-2905.
5. Cassidy-Stone A, Chipuk JE, Ingerman E, et al. Chemical inhibi-
tion of the mitochondrial division dynamin reveals its role in Bax/
Bak-dependent mitochondrial outer membrane permeabilization.
Dev Cell. 2008;14:193-204.
6. Brooks C, Wei Q, Cho SG, Dong Z. Regulation of mitochondrial
dynamics in acute kidney injury in cell culture and rodent models.
J Clin Invest. 2009;119:1275-1285.
7. Tian L, Neuber-Hess M, Mewburn J, et al. Ischemia-induced
Drp1 and Fis1-mediated mitochondrial fission and right ventric-
ular dysfunction in pulmonary hypertension. J Mol Med (Berl).
2017;95:381-393.
7
CONCLUSION
|
We have identified two novel analogs of ellipticine, Drpitor1
and Drpitor1a, that are potent and specific Drp1 GTPase in-
hibitors. These compounds have great potential value as tools
to understand the basic mechanisms of fission and Drp1 biol-
ogy. These compounds also have therapeutic potential in can-
cer, where they inhibit cell proliferation and induce apoptosis
by direct suppression of Drp1-mediated mitochondrial mi-
totic fission. Drpitor1a also prevents ROS production in ex-
perimental IR and may be a useful cardioprotective agent for
cardiopulmonary diseases such as pulmonary hypertension.
8. Sharp WW, Fang YH, Han M, et al. Dynamin-related protein
1 (Drp1)-mediated diastolic dysfunction in myocardial isch-
emia-reperfusion injury: therapeutic benefits of Drp1 inhibition to
reduce mitochondrial fission. FASEB J. 2014;28:316-326.
9. Sumida M, Doi K, Ogasawara E, et al. Regulation of mitochondrial
dynamics by dynamin-related protein-1 in acute cardiorenal syn-
drome. J Am Soc Nephrol. 2015;26:2378-2387.
10. Marsboom G, Toth PT, Ryan JJ, et al. Dynamin-related protein
1-mediated mitochondrial mitotic fission permits hyperprolif-
eration of vascular smooth muscle cells and offers a novel ther-
apeutic target in pulmonary hypertension. Circ Res. 2012;110:
1484-1497.
11. Rehman J, Zhang HJ, Toth PT, et al. Inhibition of mitochondrial
fission prevents cell cycle progression in lung cancer. FASEB J.
2012;26:2175-2186.
12. Peiris-Pages M, Bonuccelli G, Sotgia F, Lisanti MP. Mitochondrial
fission as a driver of stemness in tumor cells: mDIVI1 inhibits
mitochondrial function, cell migration and cancer stem cell (CSC)
signalling. Oncotarget. 2018;9:13254-13275.
13. Bordt EA, Clerc P, Roelofs BA, et al. The putative Drp1 inhibitor
mdivi-1 is a reversible mitochondrial complex I inhibitor that mod-
ulates reactive oxygen species. Dev Cell. 2017;40:583-594 e586.
14. Ruiz A, Alberdi E, Matute C. Mitochondrial division inhibitor 1
(mdivi-1) protects neurons against excitotoxicity through the mod-
ulation of mitochondrial function and intracellular Ca(2+) signal-
ing. Front Mol Neurosci. 2018;11:3.
15. Dallakyan S, Olson AJ. Small-molecule library screening by dock-
ing with PyRx. Methods Mol Biol. 2015;1263:243-250.
16. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Experimental
and computational approaches to estimate solubility and permea-
bility in drug discovery and development settings. Adv Drug Deliv
Rev. 2001;46:3-26.
17. Chen KH, Dasgupta A, Lin J, Potus F, et al. Epigenetic dysregu-
lation of the Drp1 binding partners MiD49 and MiD51 increases
mitotic mitochondrial fission and promotes pulmonary arterial hy-
pertension: mechanistic and therapeutic implications. Circulation.
2018;138:287-304.
18. Wenger J, Klinglmayr E, Frohlich C, et al. Functional mapping of
human dynamin-1-like GTPase domain based on x-ray structure
analyses. PLoS ONE. 2013;8:e71835.
ACKNOWLEDGMENTS
This research is supported by Queen's Cardiopulmonary Unit
(QCPU). The authors would like thank Dr. Peter Greer from
Cancer Research Institute, Queen's University for providing
the BALB/c Rag2-/-; IL2Rgc-/- double knockout mice breed-
ing pairs. This study is supported by the CIHR Foundation
Grant, NIH-R01-HL071115, 1RC1HL099462, a Tier 1
Canada Research Chair in Mitochondrial Dynamics and the
William J Henderson Foundation (S.L.A.), NSERC Canada
Discovery Grant (DG) program (V.S., J.P.), and Canadian
Vascular Network Scholar Award 2016-2018 (D.W.). The
invention of Drpitor1 and Drpitor1a is the subject of a US
patent application. Wu D, Wells M, Archer SL. Inhibitors of
Mitochondrial Fission. Patent Application No. 62/826,247.
CONFLICT OF INTEREST
None.
AUTHOR CONTRIBUTIONS
Danchen Wu, Asish Dasgupta, Kuang-Hueih Chen, and
Stephen L. Archer designed the research; Danchen Wu,
Asish Dasgupta, Kuang-Hueih Chen, Monica Neuber-Hess,
Jeffrey D. Mewburn, Patricia D.A. Lima, Elahe Alizadeh,
and Ashley Martin performed the research; Michael Wells
performed in silico analysis; Timothy E. Hurst and Jignesh
Patel synthesized Drpitor1 and Drpitor1a; Danchen Wu,
Asish Dasgupta, and Patricia D.A. Lima analyzed the data;
Danchen Wu, Asish Dasgupta, and Stephen L. Archer wrote
the paper; Stephen L. Archer supervised the experiments; and
Stephen L. Archer and Victor Snieckus acquired funding.
REFERENCES
1. Archer SL. Mitochondrial dynamics–mitochondrial fission and fu-
sion in human diseases. N Engl J Med. 2013;369:2236-2251.
2. Loson OC, Song Z, Chen H, Chan DC. Fis1, Mff, MiD49, and
MiD51 mediate Drp1 recruitment in mitochondrial fission. Mol
Biol Cell. 2013;24:659-667.
19. Chen Y, Dorn GW, 2nd. PINK1-phosphorylated mitofusin 2 is
a Parkin receptor for culling damaged mitochondria. Science.
2013;340:471-475.
20. Mitra K, Wunder C, Roysam B, Lin G, Lippincott-Schwartz J.
A hyperfused mitochondrial state achieved at G1-S regulates
3. Smirnova E, Griparic L, Shurland DL, van der Bliek AM.
Dynamin-related protein Drp1 is required for mitochondrial divi-
sion in mammalian cells. Mol Biol Cell. 2001;12:2245-2256.

1464
WU et al.
|
cyclin E buildup and entry into S phase. Proc Natl Acad Sci USA.
26. Watanabe M, Snieckus V. Tandem directed metalation reactions
- short syntheses of polycyclic aromatic-hydrocarbons and ellipti-
cine alkaloids. J Am Chem Soc. 1980;102:1457-1460.
2009;106:11960-11965.
21. Maneechote C, Palee S, Chattipakorn SC, Chattipakorn N. Roles of
mitochondrial dynamics modulators in cardiac ischaemia/reperfu-
sion injury. J Cell Mol Med. 2017;21:2643-2653.
22. Thenappan T, Ormiston ML, Ryan JJ, Archer SL. Pulmonary ar-
terial hypertension: pathogenesis and clinical management. BMJ.
2018;360:j5492.
23. Sharp WW, Beiser DG, Fang YH, et al. Inhibition of the mito-
chondrial fission protein dynamin-related protein 1 improves sur-
vival in a murine cardiac arrest model. Crit Care Med. 2015;43:
e38-e47.
SUPPORTING INFORMATION
Additional supporting information may be found online in
the Supporting Information section.
How to cite this article: Wu D, Dasgupta A,
Chen K-H, et al. Identification of novel dynamin-
related protein 1 (Drp1) GTPase inhibitors:
Therapeutic potential of Drpitor1 and Drpitor1a in
cancer and cardiac ischemia-reperfusion injury.
The FASEB Journal. 2020;34:1447–1464.
https://doi.org/10.1096/fj.201901467R
24. Nishiyama T, Hatae N, Mizutani M, et al. Concise synthesis and
antiproliferative activity evaluation of ellipticine quinone and its
analogs. Eur J Med Chem. 2017;136:1-13.
25. Bernardo PH, Chai CL, Heath GA, et al. Synthesis, electrochem-
istry, and bioactivity of the cyanobacterial calothrixins and related
quinones. J Med Chem. 2004;47:4958-4963.