Pre‑clinical evidence that salinomycin is active against retinoblastoma
via inducing mitochondrial dysfunction, oxidative damage and AMPK
activation
Jing Li1
· Yao Min1
Received: 22 February 2021 / Accepted: 20 July 2021
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2021
Abstract
The poor outcomes in retinoblastoma necessitate new treatments. Salinomycin is an attractive candidate, and has demon￾strated selective anti-cancer properties in diferent cancer types. This work addressed the efcacy of salinomycin in retino￾blastoma models and probe the associated mechanisms. Cellular functional assays were conducted to determine the efects
salinomycin in vitro. Xenograft retinoblastoma mouse model was established to investigate the efcacy of salinomycin
in vivo. Biochemical assays were conducted to analyze the mechanism of salinomycin’s action focusing on mitochondrial
functions, energy reduction-related signaling pathways. Salinomycin has positive efects towards retinoblastoma cells regard￾less of heterogeneity through suppressing growth and inducing apoptosis. Salinomycin also specifcally inhibits cells dis￾playing stemness and highly invasive phenotypes. Using retinoblastoma xenograft mouse model, we show that salinomycin
at non-toxic dose efectively inhibits growth and induces apoptosis. Mechanistic studies show that salinomycin inhibits
mitochondrial respiration via specifcally suppressing complex I and II activities, reduces mitochondrial membrane potential
and decreases energy reduction, followed by induction of oxidative stress and damage, AMPK activation and mTOR inhibi￾tion. Our study highlights that adding salinomycin to the existing treatment armamentarium for retinoblastoma is benefcial.
Keywords Salinomycin · Retinoblastoma · Mitochondria respiration · AMPK/mTOR
Introduction
Retinoblastoma is one of many diferent malignant intraocu￾lar diseases afecting children. This is the most common
condition in infants that has very poor prognosis (Kivela
2009). Standard of care (SOC) options include enuclea￾tion, laser photocoagulation, cryotherapy, transpupillary
thermotherapy and chemotherapy (Cassoux et al. 2017).
Although the loss of retinoblastoma 1 (RB1) is a key fac￾tor, which has a critical role in retinoblastoma initialization
and progression, there exists other genetic and epigenetic
modifcations that afects RB1. This may in turn produce
diferent clinical outcomes (Stenfelt et al. 2017). Further￾more, a signifcant fraction of aggressive tumor results from
MYCN amplifcation independent of RB1 (Theriault et al.
2014). Transcriptomics analysis together with known genetic
mutations show that retinoblastoma is extensively hetero￾geneous (Winter et al. 2020). Therefore, targeting common
rather than diferential molecular “driver” mutations may
represent a promising therapeutic strategy for the treatment
of retinoblastoma.
Disrupting the metabolic pathways to aid in cancer
treatment has garnered attention with discovery of aerobic
glycolysis (Luengo et al. 2017). Further work reveal that
metabolic activities in tumor cells are particularly depend￾ent on oxidative phosphorylation (mitochondrial respira￾tion) to generate ATP for energy production (Zheng 2012).
This is because mitochondrial metabolic properties difer
signifcantly among cancerous and normal cells. Many can￾cers, including leukemia, breast cancer and retinoblastoma,
exhibit increased mitochondrial biogenesis as well as oxygen
consumption (Lagadinou et al. 2013; Skrtic et al. 2011; Yu
et al. 2016). Of note, stem-like tumor initiating cells are
rely on mitochondrial metabolism (Lagadinou et al. 2013;
Luca et al. 2015). Salinomycin, an antibiotic, is identifed
* Yao Min
[email protected]
1 Department of Ophthalmology, The Central Hospital
of Wuhan, Tongji Medical College, Huazhong
University of Science and Technology, 26 Shengli Street,
Wuhan 430014, Hubei, China
/ Published online: 8 August 2021
Journal of Bioenergetics and Biomembranes (2021) 53:513–523
1 3
to be an attractive candidate for cancer chemoprevention
and therapy with board mechanisms of action against cancer
(Tung et al. 2017; Ko et al. 2016; Markowska et al. 2019).
Salinomycin has been shown to have negative impact on
mitochondrial bioenergetic function (Manago et al. 2015;
Mitani et al. 1976). We hypothesized that salinomycin is a
promising drug for retinoblastoma treatment.
In the current study, we examined the efects of salinomy￾cin on retinoblastoma using both in vitro and in vivo retino￾blastoma models. We systematically analysed mechanisms
of the action of salinomycin. We demonstrate that salinomy￾cin inhibits retinoblastoma growth, survival and colony for￾mation, and furthermore that these inhibitory properties of
salinomycin in retinoblastoma are contributed to inhibition
of mitochondrial respiration and energy production, induc￾tion of oxidative stress and damage, activation of AMPK and
inhibition of mTOR.
Materials and methods
Cell culture
RB 383, WERI-Rb-1 and RB116 cell lines (the Cell Bank of
Type Culture Collection of Chinese Academy of Sciences)
were grown as suspension cultures using DMEM media
containing 10% FBS (Hyclone), 50 µM β-mercaptoethanoal
(Sigma), 10 µg/ml insulin, 1% penicillin–streptomycin and
4 mM l-glutamine and. All related culturing reagents were
procured from Invitrogen unless otherwise indicated.
Mitochondrial DNA‑defcient cell line generation (ρ0
cells)
The RB 383 ρ0
cell line was derived using processes detailed
by Hashiguchi and Zhang (Hashiguchi and Zhang-Akiyama
2009). Culturing of RB 383 cells were performed and fur￾ther selected using 2 µg/ml ethidium bromide (EtBr). Uri￾dine at 50 µg/ml and sodium pyruvate (Sigma) at 2 mM were
added to the selection medium. Cells were cultured in the
above media for a duration of 8 weeks. Cells were thereafter
cultured in normal culture media in the absence of EtBr.
Bromodeoxyuridine (BrdU) assay
A total of 104
cells/well were incubated with dimethyl Sul￾foxide (DMSO; Sigma) or salinomycin (MedChemExpress)
at diferent concentrations. After drug treatment for 3 days,
proliferative activities were assessed by BrdU Cell Prolifera￾tion Colorimetri Assay kit (Abcam, Catalog No. ab126556)
and quantifed on a microplate reader based on the absorb￾ance at 450 nm.
Flow cytometry
To determine cell apoptosis, cells at 105
/well were treated
with drugs. After 3 days, these cells were counterstained
with Annexin V-FITC and 7-AAD (Abcam, Catalog No.
ab214663). Each sample was incubated with 10  µl of
Annexin V-FITC and 5 µl of 7-AAD for 15 min prior to
being analysed. For determining the cell intracellular reac￾tive oxygen species (ROS) and mitochondrial membrane
potential, cells at 105
/well were treated with DMSO or
salinomycin for 24 h. These cells were counterstained
with 10 µM CM-H2DCFDA (Life Technologies, Catalog
No. C6827) at 37 °C for 30 min for ROS and 10 µg/ml
5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethyl benzimidazolylcar￾bocyanine iodide (JC-1, Invitrogen, Catalog No. T3168) at
37 °C for 20 min for mitochondrial membrane potential,
respectively. The percentage of Annexin V, intensity of
CM-H2DCFDA and JC-1 dye were determined by fow
cytometry analysis on Beckman Coulter FC500.
Anchorage‑independent growth in soft agar
Anchorage-independent growth assay was conducted
using method previously described (Gao et al. 2005). In
brief, 1000 tumor cells-salinomycin mixture was seeded
in 12-well plate. The top layer consists of 0.3% Bacto agar
and the bottom layer consists of 0.7% Bacto agar. Cell
culture media was subsequently added onto the top layer.
Culture media in the wells were removed and fresh solu￾tions were added twice each week. Associated cell colo￾nies were counterstained with crystal violet and colony
numbers were counted after 2 weeks.
Mito stress and glycolytic stress test assays
104
cells/well were seeded in each XF24 cell culture plate
(Seahorse Bioscience, Catalog No. 101848-400) and
treated with drugs for 24 h. Media were replaced with the
XF assay medium. For pH stabilization, the plate was then
placed at 37 °C incubation chamber in a CO2-free envi￾ronment. Oxygen consumption rate (OCR) and extracel￾lular acidifcation rate (ECAR) was then measured using
a Seahorse XF24 extracellular fux analyser (Seahorse
Bioscience). Respiratory capacity was examined after
injection of 1 µg/ml oligomycin, 0.4 µM Carbonyl cya￾nide-p-trifuoromethoxyphenylhydrazone (FCCP), 2.5 µM
antimycin A and rotenone. Optimal FCCP concentration
was determined based on its efcacy in stimulating OCR
level. Glucose (10 nM), oligomycin (1 µM) and 2deoxy￾d-glucose (2-DG, 50 mM) were injected sequentially for
ECAR measurement. OCR and ECAR values were normal￾ized to protein mass.
Measurement of mitochondrial complex I, II, IV
and V activities
Cells at 105
/well were treated with DMSO or salinomycin
for 24 h. In vitro mitochondrial complex I (Abcam, Cata￾log No. ab109721), II (Abcam, Catolog No. ab109908), III
(Sigma, Catalog No. MAK360), IV (Abcam, Catalog No.
ab109909) and V (Abcam, Catalog No. ab109907) activities
were assessed using total cell lysates and were measured
using kits as per the manufacturer’s instructions. Protein
lysates of each sample were adjusted to the concentration
of 5 mg/ml prior to being added to microplates coated with
antibodies against each electron transport chain complex.
The complex activities were determined calorimetrically on
microplate reader.
Western blot (WB) anlaysis
Cells at 106
/well were incubated with DMSO or salinomy￾cin for 24 h. Samples of cell lysate were prepared, resolved
using denaturing sodium dodecyl sulfate–polyacrylamide gel
electrophoresis (SDS-PAGE) and analyzed by WB using the
standard protocol (Mahmood and Yang 2012). Antibodies
against p-AMPK (Catalog No. #2531), AMPK (Catalog
No. 2532), p-ACC (Catalog No. 11818), ACC (Catalog No.
3662), p-mTOR (Catalog No. 2971), mTOR (Catalog No.
2983), p-rS6 (Catalog No. 2211), rS6 (Catalog No. 2217),
p-4EBP1 (Catalog No. 2885), 4EBP1 (Catalog No. 9644)
and β-actin (Catalog No. 4970) were obtained from Cell
Signaling.
Measurement of cellular ATP
105
cells/well were seeded in a 24-well and exposed with
drug for a duration of 24 h. ATP levels were quantifed using
ATPlite Luminiescent Assay kit (Perkin Elmer, Catalog No.
6016941) as per manufacturer’s instructions. ATP values
were normalized to protein mass.
Measurement of oxidative DNA damage assays
cells/well were seeded in a 24-well and exposed with
drug for a duration of 24 h. Oxidative DNA damage was
assessed by quantifying 8-hydroxy-2′-deoxyguanosine
(8-OHdG) levels using the OxiSelect Oxidative DNA Dam￾age ELISA kit (Cell Biolabs, Catalog No. STA-320-T) as per
manufacture’s instructions.
Retinoblastoma xenograft in SCID mouse
The xenograft experiments were approved by the Insti￾tutional Animal Care and Use Committee of Huazhong
University of Science and Technology. RB 383 cells sus￾pended in 100 µl 50%/50% PBS/Matrigel were subcutane￾ously injected into fank of 6-week-old NOD/SCID mice
(Hunan SJA Laboratory Animal Co., Ltd). When tumor
reached ~200 mm3
, mice were randomized into two groups
and administrated for vehicle (20%/80% DMSO/PBS) and
salinomycin (2  mg/kg/day) by intraperitoneal injection
(n=10 per group). After 2 weeks of drug treatment, mice
were euthanatized using CO2 inhalation. Tumours were
excised and sectioned for immunostaining. Sections were
fxed with 4% paraformaldehyde (Sigma). The sections were
subsequently exposed to Ki67 and active caspase 3 (Cell
Signalling) primary antibodies, and counterstained using
secondary antibody conjugated with horseradish peroxidase￾DAB (3,3′-diaminobenzidine). Cell nuclei were stained with
hematoxylin.
Results
Salinomycin efectively inhibits biological functions
of retinoblastoma cells
Using multiple retinoblastoma cell lines (RB383, RB116 and
WERI-Rb1) that represent diferent cell origins and covers
a broad range of genetic profles (Busch et al. 2015), we
frst studied the efects of salinomycin on retinoblastoma
cell growth, associated forming of colonies and survival.
We showed that salinomycin at 5, 10 and 15 µM exhibited
antiproliferative efects against three retinoblastoma cell
lines dose-dependently (p<0.05; Fig. 1A). The IC50 for
salinomycin as determined by the BrdU assay is comparable
among retinoblastoma cell lines (Supplementary Table 1),
suggesting that salinomycin is efective against retinoblas￾toma regardless of cell heterogeneity.
Retinoblastoma cell lines selected in our study present
with certain stem cell (eg, Nanog and Oct4) and retinal pro￾genitor markers (eg, PAX6) (Hu et al. 2012; Bejjani et al.
2012). This subpopulation with stemness and highly inva￾sive phenotypes can be examined by anchorage-independ￾ent colony formation assay (Gao et al. 2005). We showed
salinomycin potently inhibited the anchorage-independent
growth of retinoblastoma cells (p<0.05; Fig. 1B). Of note,
we observed that salinomycin at the same concentration (eg,
10 µM) decreased growth and colony formation by ~ 60%
and ~ 90%, respectively (Fig. 1A, B), suggesting that this
subpopulation is more sensitive to salinomycin than the
whole cell population.
Apoptosis assay was performed by fow cytometry of
Annexin V/7-AAD. The results were shown in Fig. 1C and
supplementary Fig.  1. Pro-apoptotic efects of salinomycin
were demonstrated in all cell lines at high concentration.
Salinomycin impairs mitochondrial function,
induces oxidative damage, activates AMPK
and inhibits mTOR in retinoblastoma cells
We analysed the efects of salinomycin on mitochondrial
bioenergetic performance in RB 383 cells. We measured
the rate of oxygen consumption by the cells, an indicator of
mitochondrial respiration, under basal and electron transport
chain (ETC) accelerator-stimulated conditions. We observed
a reduced basal respiration rates and decreased ATP coupler
response in salinomycin-treated cells (Fig. 2A–C). Trifuoro￾carbonylcyanide phenylhydrazone (FCCP), an ETC accel￾erator, is often used to assess the maximal respiratory capac￾ity of cells. We found that salinomycin-treated cells were
not responsive to FCCP stimulation and there was a dose￾dependent reduction of ETC accelerator response (Fig. 2A,
D), indicative of reduction in the maximal respiratory capac￾ity and potential rate for ATP production. We performed
glycolytic stress test assay in cells after salinomycin treat￾ment. We found that there was no signifcant diference in
the basal and maximal glycolytic capacity between control
and salinomycin-treated cells (Fig. 2E), indicating that salin￾omycin does not compromise glycolysis. We next measured
mitochondrial complexes activities in cells after salinomycin
treatment. We found that salinomycin signifcantly decreased
activities of complex I and II without afecting activities of
complex III, IV and V (Fig. 3), suggesting that salinomycin
inhibits mitochondrial respiration through suppressing mito￾chondrial complex I and II in retinoblastoma cells.
Mitochondrial membrane potential was quantifed using
fow cytometry of JC-1 dye. We showed that salinomycin
signifcantly decreased mitochondrial membrane potential
in retinoblastoma cells (Fig. 4A and supplementary Fig. 2).
As expected, ATP levels were observed in cells after salino￾mycin treatment (Fig. 4B). To understand the sequences of
mitochondrial membrane potential disruption, ATP reduc￾tion and apoptosis, we performed time course analysis.
Reduction of membrane potential and ATP levels were ini￾tialized after 0.5 h and 6 h whereas apoptosis was initial￾ized after 24 h addition of salinomycin (Fig. 4C–E and sup￾plementary Fig. 3), suggesting that salinomycin treatment
Fig. 1 Salinomycin is active
against retinoblastoma cells
in vitro. Salinomycin at 5, 10
and 15 µM inhibits proliferation
(A), suppresses anchorage-inde￾pendent colony formation (B)
and induces apoptosis (C) in RB
383, RB116 and WERI-Rb1 cell
lines. The results were obtained
from at least three independ￾ent experiments with triplicate.
The results were presented as
average±SD and were shown
as relative to 0 µM salinomycin.
*p<0.05, compared to 0 µM
salinomycin
leads to membrane potential disruption, followed by ATP
reduction and apoptosis.
As a consequence of mitochondrial dysfunction and
energy crisis, we observed increased intracellular DCF and
8-OHdG (an oxidized DNA byproduct) levels in salinomy￾cin-treated retinoblastoma cells (Fig. 5A, B and supplemen￾tary Fig. 4), indicating that salinomycin induces oxidative
stress and damage. We noted that the inhibitory efects of
salinomycin on mitochondria respiration reached saturation
at 10 µM. Although there is diference on the average levels
of ROS and 8-DG between 10 and 15 µM, the diference
is not statistically signifcant (p>0.05). This suggests that
mitochondrial respiration inhibition is likely the main cause
of oxidative stress and damage in salinomycin-treated cells.
AMP-dependent kinase (AMPK) activity correlates
to ATP levels and negatively regulates the mammalian
target of rapamycin (mTOR) pathway (Cork et al. 2018).
To investigate the consequence of mitochondrial dys￾function and energy reduction by salinomycin, we per￾formed immunoblotting analysis of essential molecules
associated in AMPK activity and mTOR signaling path￾way. Indeed, treatment of RB383 cells with salinomycin
presented elevations in phosphorylated AMPK at T127
(Fig. 5C). Consistent with increased AMPK activity, we
observed increased inhibitory phosphorylation of acety￾coA carboxylase (ACC) at Ser79, an AMPK phospho￾rylation site. In addition, decreased phosphorylation of
mTOR was observed in RB383 cells after 24 h treatment
Fig. 2 Salinomycin inhibits mitochondrial respiration in retinoblas￾toma cells. A OCR profle of RB 383 cells after salinomycin treat￾ment. OCRs were measured under basal condition frst, followed by
that after sequential addition of oligomycin, FCCP, and antimycin A
and rotenone. Salinomycin signifcantly decreases basal respiration
(B), ATP coupler response (C) and ETC accelerator response (D) in
RB 383 cells. The basal respiration value is the average of measure￾ments 1–3 before oligomycin injection. ATP coupler response value
is the average of measurements 4–6 after oligomycin injection. ETC
accelerator response value is the average of measurements 7–9 after
FCCP injection. D ECAR profle of RB 383 cells after salinomycin
treatment. ECAR tracings were obtained at baseline, and then after
glucose injection, addition of oligomycin and injection of 2-DG. All
the OCR and ECAR values presented were normalized to protein
mass. The results were presented as average±SD. *p<0.05, com￾pared to 0 µM salinomycin
of salinomycin, and so do its downstream efectors rS6 and
eukaryotic translation initiation factor 4E-binding protein
1 (4EBP1). Taken together, our results demonstrate that
salinomycin treatment impairs mitochondrial functions,
induces oxidative damage, activates AMPK and inhibits
mTOR in retinoblastoma cells.
Salinomycin is inefective in mitochondrial
respiration‑defcient retinoblastoma ρ0 cells
To confrm that mitochondrial respiration plays an important
role in the action of salinomycin in retinoblastoma, we gen￾erated RB383 ρ0 cells lacking mitochondrial DNA and thus
Fig. 3 Salinomycin inhibits
mitochondrial complex I and
II activities in retinoblastoma
cells. Salinomycin signifcant
decreases mitochondrial com￾plex I (A), II (B) but not III (C),
IV (D) or V (E) activities in
RB 383 cells. The results were
presented as average±SD and
were shown as relative to 0 µM
salinomycin. *p<0.05, com￾pared to 0 µM salinomycin
Fig. 4 Salinomycin decreases
mitochondrial membrane
potential and ATP levels
in retinoblastoma cells. A
Salinomycin at 5, 10 and
15 µM reduces mitochondrial
membrane potential in RB 383
cells. B Salinomycin decreases
ATP levels in RB 383 cells.
Membrane potential and ATP
level were measured after 24-h
salinomycin treatment. 1 ug/
ml oligomycin and 1 uM FCCP
were used. Time course analysis
of mitochondrial membrane
potential (C), ATP level (D) and
apoptosis (E) in RB 383 cells.
The results were presented as
average±SD and were shown
as relative to 0 µM salinomycin.
*p<0.05, compared to 0 µM
salinomycin
incapable of mitochondrial respiration processes (Chandel
and Schumacker 1999). We have attempted to generate ρ0
cells from RB383, RB116 and WERI-Rb1 cell lines but it
was only successful on RB383 cells. RB116 and WERI-Rb1
did not survive after mitochondrial respiration depletion.
This is consistent with Yasuo et al.’s fndings that generation
of ρ0 cells were failed in most of tested cell lines (Harada
et al. 2012). We validated RB383 ρ0 cells by showing that
ρ0 cells display a minimal level of baseline OCR and are
non-responsive to FCCP stimulation (Fig. 6A). There was
no signifcant diference on mitochondrial membrane poten￾tial, intracellular DCF and 8-OHdG levels between ρ0 and
parental cells (Fig. 6B–D). As expected, we observed a
remarkable ATP reduction in ρ0 cells compared to paren￾tal cells (Fig. 6E). Of note, salinomycin was inefective in
reducing mitochondrial membrane potential, increasing DCF
and 8-OHdG levels, and decreasing ATP production in ρ0
cells (Fig. 6B–E). These results suggest that mitochondrial
membrane potential disruption, ATP reduction and oxidative
stress/damage are the consequence of mitochondrial respi￾ration inhibition by salinomycin. In addition, these ρ0 cells
were resistant to salinomycin-induced apoptosis (Fig. 6F).
The inhibitory efect of salinomycin on the proliferation of
RB 383 ρ0 cells was not determined due to inadequate pro￾liferation of ρ0 cells (Fig. 6G).
Salinomycin inhibits growth and induces apoptosis
of retinoblastoma in vivo
Tumor xenograft model in mice is widely applied to under￾stand the anticancer activities of novel compounds. Fur￾ther examination of the in vivo efcacy of salinomycin
uses SCID mice to establish the correlation with in vitro
experiments. We established retinoblastoma xenograft
mouse model by implanting RB383 cells into the fank
of SCID mice. Salinomycin treatment was initialized
Fig. 5 Salinomycin induces oxidative damage, activates AMPK and
inhibits mTOR in retinoblastoma cells. Salinomycin signifcantly
increases intracellular DCF levels (A) and increases 8-OHdG (B) in
RB 383 cells. C Representative western blot image of RB 383 cells
treated with salinomycin for 24 h. The results were obtained from at
least three independent experiments with triplicate. The results were
presented as average±SD and were shown as relative to 0 µM salino￾mycin. *p<0.05, compared to 0 µM salinomycin
after the development of palpable tumors. After 2 weeks
treatment, we did not observe a signifcant reduction on
body weight and abnormal appearance in mice treated
with salinomycin compared to control (data not shown),
suggesting that salinomycin at 2 mg/kg per day via intra￾peritoneal administration is not toxic to mice. In contrast,
we observed the signifcant reduction of tumor size and
weight in mice treated with salinomycin (Fig. 7A, B). In
addition, salinomycin delayed tumor growth starting at
2 days post treatment.
We next performed immunohistochemistry staining of
Ki67 and active caspase 3 on tumor sections to determine
if the reduced tumor growth was the result of declines in
cell proliferation, increased apoptosis or both. The results
showed ~60% decrease in tumor cell proliferation and ~10
times increase in tumor cell apoptosis in mice receiving
salinomycin, as compared to the control (Fig. 7C–E).
Discussion
Higher levels of mitochondrial biogenesis is associated with
retinoblastoma, and this includes elevated levels of mito￾chondrial respiration, membrane potential and mass when
compared with normal retina cells (Ke et al. 2018). The
unique dependency for retinoblastoma on mitochondrial
metabolism to maintain cell survival and growth can be
exploited therapeutically. In the current work, we demon￾strate that mitochondrial functions can be impaired by salin￾omycin, leading to retinoblastoma models with decreased
growth and increased apoptosis. Although salinomycin is
a widely used antibiotic in veterinary medicine, substantial
pre-clinical evidence has highlighted salinomycin as a poten￾tial anti-cancer agent (Markowska et al. 2019).
In agreement with numerous reports on inhibitory efects
of salinomycin on colorectal and lung cancer, head and neck
Fig. 6 ρ0 retinoblastoma cells are resistant to salinomycin treat￾ment. a RB 383 ρ0 cells displays minimal basal and maximal OCR.
Oligomycin, FCCP and antimycin A and rotenone combination were
added as indicated time points. Salinomycin is inefective in decreas￾ing membrane potential (B), increasing DCF (C) and 8-OHdG (D)
levels, decreasing ATP levels (E). F Salinomycin is inefective in
inducing apoptosis in RB 383 ρ0 cells. Salinomycin at 15  µM was
used. G Minimal growth rate in RB 383 ρ0 cells. The results were
obtained from at least three independent experiments with triplicate.
The results were presented as average±SD. *p<0.05, compared to
control
squamous cell carcinoma (Hochmair et al. 2020; Klose et al.
2019; Gehrke et al. 2018), our fndings show that salinomy￾cin displays pro-apoptotic and anti-proliferative activities
in a panel of retinoblastoma cell lines that represent a board
range of cellular origin and genetic profles. In addition, our
anchorage-independent colony formation analysis indicates
that salinomycin preferentially inhibits retinoblastoma sub￾populations with stemness and highly invasive phenotypes.
This is consistent with the previous work that salinomycin
targets human cancer stem cells via reducing stemness and
motility, and inducing death (Mai et al. 2017; Sun et al.
2017; Lee et al. 2017). We further validated our fndings
using retinoblastoma xenograft mouse model and found
that the in vitro fndings were reproducible in in vivo. We
show that salinomycin at non-toxic dose efectively reduces
retinoblastoma size and weight via suppressing tumor cell
growth and survival. Our fndings add retinoblastoma to the
long list of salinomycin-targeted cancers and demonstrate
the therapeutic window of salinomycin in retinoblastoma.
Increasing evidence has shown the synergistic efect of salin￾omycin with anti-cancer agents and the ability of salinomy￾cin in overcoming therapy-resistant cancer cells (Zhou et al.
2019; Skeberdyte et al. 2018; Manmuan et al. 2017; Dewan￾gan et al. 2017; Gruber et al. 2020). To further characterize
the potential application of salinomycin in retinoblastoma, it
is worthy of investigating the combinatory efect of salino￾mycin with chemotherapy on retinoblastoma models.
The underlying mechanisms of salinomycin’s anti-can￾cer action are complex, including inhibition of amplifed
breast 1 (AIB1) and focal adhesion kinase (FAK), induction
of NF-kB degradation, suppression of extracellular signal￾regulated kinase (ERK1/2) (Sun et al. 2017; Manmuan
et al. 2017; Tyagi and Patro 2019). Although Manago et al.
have demonstrated that salinomycin exerts its efect at the
mitochondrial level that subsequently changes the bioener￾getic performance in B cell chronic lymphocytic leukemia
Fig. 7 Salinomycin inhibits retinoblastoma growth in mice. A Salin￾omycin signifcantly inhibits retinoblastoma growth in xenograft
mouse model (n=10 mice in each group). B Salinomycin signif￾cantly decreases retinoblastoma weight. C Representative photos of
immunostaining of Ki67 and active caspase 3. Salinomycin signif￾cantly decreases Ki67 staining levels (D) and increase active caspase
3 staining levels (E) in retinoblastoma. Intraperitoneal salinomycin
at 2  mg/kg once per day were given to the mice. Cell proliferation
and apoptosis were shown by the average number of proliferative and
apoptotic cells per microscopic fled. *p<0.05, compared to control
Journal of Bioenergetics and Biomembranes (2021) 53:513–523 521
(Manago et al. 2015), our study further extends the previ￾ous fndings by showing that besides mitochondrial dysfunc￾tion, ATP reduction, oxidative stress and damage, AMPK
activation and mTOR inhibition contribute to salinomycin’s
action in retinoblastoma cells. Salinomycin is inefective in
decreasing ATP levels, increasing oxidative damage and
inducing apoptosis in mitochondrial respiration-defcient
cells. As a consequence of mitochondrial dysfunction and
energy crisis by salinomycin, AMPK is activated and mTOR
signaling pathway is inhibited, leading to inhibition of cel￾lular functional activities. Together with the mechanism of
action of salinomycin, our observation that the efective dose
of salinomycin in retinoblastoma xenograft mouse model is
not toxic to mice correlates well with the previous fnding
that retinoblastoma cells are more dependent on mitochon￾drial respiration than normal cells (Ke et al. 2018).
In summary, we report that salinomycin is active against
retinoblastoma. We show that the activity of salinomycin in
retinoblastoma is dependent on its ability to inhibit mito￾chondrial respiration. Our fndings highlight that targeting
mitochondrial respiration is an efective therapeutic strategy
for the treatment of retinoblastoma.
Supplementary Information The online version contains supplemen￾tary material available at https://doi.org/10.1007/s10863-021-09915-2.
Acknowledgements This work was supported by Nature Science Foun￾dation of Hubei Province (2016-A06).
Authors’ contribution JL and YM designed the experiments; JL per￾formed the experiments and wrote the manuscript; All authors inter￾preted the data, supervised the project and revised the manuscript.
Data availability The datasets generated during and/or analysed dur￾ing the current study are available from the corresponding author on
reasonable request.
Declarations
Research involving human participants and/or animals This work was
approved by the institutional review board of Tongji Medical College,
Huazhong University of Science and Technology on 26 August 2018
(Approval No. 2018-C05). This study was conducted in accordance
with the Declaration of Helsinki.
Conflict of interest All authors declare no confict of interest.
References
Bejjani A, Choi MR, Cassidy L, Collins DW, O’Brien JM, Murray T,
Ksander BR, Seigel GM (2012) RB116: an RB1+ retinoblastoma
cell line expressing primitive markers. Mol vis 18:2805–2813
Busch M, Philippeit C, Weise A, Dunker N (2015) Re-characterization
of established human retinoblastoma cell lines. Histochem Cell
Biol 143:325–338
Cassoux N, Lumbroso L, Levy-Gabriel C, Aerts I, Doz F, Desjardins
L (2017) Retinoblastoma: update on current management. Asia
Pac J Ophthalmol (Philaos) 6:290–295
Chandel NS, Schumacker PT (1999) Cells depleted of mitochondrial
DNA (rho0) yield insight into physiological mechanisms. FEBS
Lett 454:173–176
Cork GK, Thompson J, Slawson C (2018) Real talk: the inter-play
between the mTOR, AMPK, and hexosamine biosynthetic path￾ways in cell signaling. Front Endocrinol (lausanne) 9:522
De Luca A, Fiorillo M, Peiris-Pages M, Ozsvari B, Smith DL,
Sanchez-Alvarez R, Martinez-Outschoorn UE, Cappello AR,
Pezzi V, Lisanti MP, Sotgia F (2015) Mitochondrial biogenesis
is required for the anchorage-independent survival and propaga￾tion of stem-like cancer cells. Oncotarget 6:14777–14795
Dewangan J, Tandon D, Srivastava S, Verma AK, Yapuri A, Rath
SK (2017) Novel combination of salinomycin and resveratrol
synergistically enhances the anti-proliferative and pro-apoptotic
efects on human breast cancer cells. Apoptosis 22:1246–1259
Gao CF, Xie Q, Su YL, Koeman J, Khoo SK, Gustafson M, Knudsen
BS, Hay R, Shinomiya N, Vande Woude GF (2005) Proliferation
and invasion: plasticity in tumor cells. Proc Natl Acad Sci USA
102:10528–10533
Gehrke T, Hackenberg S, Polat B, Wohlleben G, Hagen R, Klein￾sasser N, Scherzad A (2018) Combination of salinomycin and
radiation efectively eliminates head and neck squamous cell
carcinoma cells in vitro. Oncol Rep 39:1991–1998
Gruber M, Handle F, Culig Z (2020) The stem cell inhibitor salino￾mycin decreases colony formation potential and tumor-initiating
population in docetaxel-sensitive and docetaxel-resistant pros￾tate cancer cells. Prostate 80:267–273
Harada Y, Ishii I, Hatake K, Kasahara T (2012) Pyrvinium pamo￾ate inhibits proliferation of myeloma/erythroleukemia cells by
suppressing mitochondrial respiratory complex I and STAT3.
Cancer Lett 319:83–88
Hashiguchi K, Zhang-Akiyama QM (2009) Establishment of human
cell lines lacking mitochondrial DNA. Methods Mol Biol
554:383–391
Hochmair M, Rath B, Klameth L, Ulsperger E, Weinlinger C, Faze￾kas A, Plangger A, Zeillinger R, Hamilton G (2020) Efects
of salinomycin and niclosamide on small cell lung cancer and
small cell lung cancer circulating tumor cell lines. Investig New
Drugs 38:946–955
Hu H, Deng F, Liu Y, Chen M, Zhang X, Sun X, Dong Z, Liu X, Ge
J (2012) Characterization and retinal neuron diferentiation of
WERI-Rb1 cancer stem cells. Mol vis 18:2388–2397
Ke F, Yu J, Chen W, Si X, Li X, Yang F, Liao Y, Zuo Z (2018) The
anti-malarial atovaquone selectively increases chemosensitiv￾ity in retinoblastoma via mitochondrial dysfunction-dependent
oxidative damage and Akt/AMPK/mTOR inhibition. Biochem
Biophys Res Commun 504:374–379
Kivela T (2009) The epidemiological challenge of the most frequent
eye cancer: retinoblastoma, an issue of birth and death. Br J
Ophthalmol 93:1129–1131
Klose J, Trefz S, Wagner T, Stefen L, Preissendorfer Charrier A,
Radhakrishnan P, Volz C, Schmidt T, Ulrich A, Dieter SM,
Ball C, Glimm H, Schneider M (2019) Salinomycin: anti-tumor
activity in a pre-clinical colorectal cancer model. PLoS ONE
14:e0211916
Ko JC, Zheng HY, Chen WC, Peng YS, Wu CH, Wei CL, Chen JC,
Lin YW (2016) Salinomycin enhances cisplatin-induced cytotox￾icity in human lung cancer cells via down-regulation of AKT￾dependent thymidylate synthase expression. Biochem Pharmacol
122:90–98
Lagadinou ED, Sach A, Callahan K, Rossi RM, Neering SJ, Minhajud￾din M, Ashton JM, Pei S, Grose V, O’Dwyer KM, Liesveld JL,
Brookes PS, Becker MW, Jordan CT (2013) BCL-2 inhibition
522 Journal of Bioenergetics and Biomembranes (2021) 53:513–523
1 3
targets oxidative phosphorylation and selectively eradicates qui￾escent human leukemia stem cells. Cell Stem Cell 12:329–341
Lee HG, Shin SJ, Chung HW, Kwon SH, Cha SD, Lee JE, Cho CH
(2017) Salinomycin reduces stemness and induces apoptosis on
human ovarian cancer stem cell. J Gynecol Oncol 28:e14
Luengo A, Gui DY, Vander Heiden MG (2017) Targeting metabolism
for cancer therapy. Cell Chem Biol 24:1161–1180
Mahmood T, Yang PC (2012) Western blot: technique, theory, and
trouble shooting. N Am J Med Sci 4:429–434
Mai TT, Hamai A, Hienzsch A, Caneque T, Muller S, Wicinski J,
Cabaud O, Leroy C, David A, Acevedo V, Ryo A, Ginestier C,
Birnbaum D, Charafe-Jaufret E, Codogno P, Mehrpour M, Rodri￾guez R (2017) Salinomycin kills cancer stem cells by sequestering
iron in lysosomes. Nat Chem 9:1025–1033
Manago A, Leanza L, Carraretto L, Sassi N, Grancara S, Quintana￾Cabrera R, Trimarco V, Toninello A, Scorrano L, Trentin L,
Semenzato G, Gulbins E, Zoratti M, Szabo I (2015) Early efects
of the antineoplastic agent salinomycin on mitochondrial function.
Cell Death Dis 6:e1930
Manmuan S, Sakunrangsit N, Ketchart W (2017) Salinomycin over￾comes acquired tamoxifen resistance through AIB1 and inhibits
cancer cell invasion in endocrine resistant breast cancer. Clin Exp
Pharmacol Physiol 44:1042–1052
Markowska A, Kaysiewicz J, Markowska J, Huczynski A (2019) Doxy￾cycline, salinomycin, monensin and ivermectin repositioned as
cancer drugs. Bioorg Med Chem Lett 29:1549–1554
Mitani M, Yamanishi T, Miyazaki Y, Otake N (1976) Salinomycin
efects on mitochondrial ion translocation and respiration. Anti￾microb Agents Chemother 9:655–660
Skeberdyte A, Sarapiniene I, Aleksander-Krasko J, Stankevicius V,
Suziedelis K, Jarmalaite S (2018) Dichloroacetate and salinomy￾cin exert a synergistic cytotoxic efect in colorectal cancer cell
lines. Sci Rep 8:17744
Skrtic M, Sriskanthadevan S, Jhas B, Gebbia M, Wang X, Wang Z,
Hurren R, Jitkova Y, Gronda M, Maclean N, Lai CK, Eberhard
Y, Bartoszko J, Spagnuolo P, Rutledge AC, Datti A, Ketela T,
Mofat J, Robinson BH, Cameron JH, Wrana J, Eaves CJ, Minden
MD, Wang JC, Dick JE, Humphries K, Nislow C, Giaever G,
Schimmer AD (2011) Inhibition of mitochondrial translation as
a therapeutic strategy for human acute myeloid leukemia. Cancer
Cell 20:674–688
Stenfelt S, Blixt MKE, All-Ericsson C, Hallbook F, Boije H (2017)
Heterogeneity in retinoblastoma: a tale of molecules and models.
Clin Transl Med 6:42
Sun J, Luo Q, Liu L, Yang X, Zhu S, Song G (2017) Salinomycin atten￾uates liver cancer stem cell motility by enhancing cell stifness
and increasing F-actin formation via the FAK-ERK1/2 signalling
pathway. Toxicology 384:1–10
Theriault BL, Dimaras H, Gallie BL, Corson TW (2014) The genomic
landscape of retinoblastoma: a review. Clin Exp Ophthalmol
42:33–52
Tung CL, Chen JC, Wu CH, Peng YS, Chen WC, Zheng HY, Jian YJ,
Wei CL, Cheng YT, Lin YW (2017) Salinomycin acts through
reducing AKT-dependent thymidylate synthase expression to
enhance erlotinib-induced cytotoxicity in human lung cancer cells.
Exp Cell Res 357:59–66
Tyagi M, Patro BS (2019) Salinomycin reduces growth, proliferation
and metastasis of cisplatin resistant breast cancer cells via NF-kB
deregulation. Toxicol in Vitro 60:125–133
Winter U, Ganiewich D, Ottaviani D, Zugbi S, Aschero R, Sendoya
JM, Caferata EG, Mena M, Sgroi M, Sampor C, Lubieniecki
F, Fandino A, Abba MC, Doz F, Podhjacer O, Carcaboso AM,
Letouze E, Radvanyi F, Chantada GL, Llera AS, Schaiquevich P
(2020) Genomic and transcriptomic tumor heterogeneity in bilat￾eral retinoblastoma. JAMA Ophthalmol 138:569–574
Yu M, Li R, Zhang J (2016) Repositioning of antibiotic levofoxacin
as a mitochondrial biogenesis inhibitor to target breast cancer.
Biochem Biophys Res Commun 471:639–645
Zheng J (2012) Energy metabolism of cancer: glycolysis versus oxida￾tive phosphorylation (review). Oncol Lett 4:1151–1157
Zhou J, Sun M, Jin S, Fan L, Zhu W, Sui X, Cao L, Yang C, Han C
(2019) Combined using of paclitaxel and salinomycin active tar￾geting nanostructured lipid carriers against non-small cell lung
cancer and cancer stem cells. Drug Deliv 26:281–289
Publisher’s note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional afliations.
Journal of Bioenergetics and Biomembranes (2021) 53:513–523 523