Novel benzoate-lipophilic cations selectively induce cell death in human colorectal cancer cell lines

Article abstract of DOI:10.1016/j.tiv.2020.104814

Introduction: Colorectal cancer (CRC) is a critical health issue worldwide. The high rate of liver and lung metastasis associated with CRC creates a significant barrier to effective and efficient therapy. Tumour cells, including CRC cells, have metabolic alterations, such as high levels of glycolytic activity, increased cell proliferation and invasiveness, and chemo- and radio-resistance. However, the abnormally elevated mitochondrial transmembrane potential of these cells also provides an opportunity to develop drugs that selectively target the mitochondrial functions of tumour cells. Methods: In this work, we used a new batch of benzoic acid esters with cytotoxic activities attached to the triphenylphosphonium group as a vehicle to target tumour mitochondria and improve their activity. We evaluated the cytotoxicity, selectivity, and mechanism of action of these derivatives, including the effects on energy stress-induced apoptosis and metabolic behaviour in the human CRC cell lines HCT-15 and COLO-205. Results: The benzoic acid derivatives selectively targeted the tumour cells with high potency and efficacy. The derivatives induced the uncoupling of the oxidative phosphorylation system, decreased the transmembrane potential, and reduced ATP levels while increasing AMPK activation, thereby triggering tumour cell apoptosis in both tumour cell lines tested. Conclusion: The benzoic acid derivatives studied here are promising candidates for assessing in vivo models of CRC, despite the diverse metabolic characteristics of these tumour cells.

Full text of DOI:10.1016/j.tiv.2020.104814

Journal Pre-proof
Novel benzoate-lipophilic cations selectively induce cell death in
human colorectal cancer cell lines
José Antonio Jara, Diego Rojas, Vicente Castro-Castillo,
Sebastián Fuentes-Retamal, Cristian Sandoval-Acuña, Eduardo
Parra, Mario Pavani, Juan Diego Maya, Jorge Ferreira, Mabel
Catalán
PII:
S0887-2333(19)30776-3
https://doi.org/10.1016/j.tiv.2020.104814
TIV 104814
DOI:
Reference:
To appear in:
Toxicology in Vitro
Received date:
Revised date:
Accepted date:
16 October 2019
1 February 2020
25 February 2020
Please cite this article as: J.A. Jara, D. Rojas, V. Castro-Castillo, et al., Novel benzoate-
lipophilic cations selectively induce cell death in human colorectal cancer cell lines,
Toxicology in Vitro(2020), https://doi.org/10.1016/j.tiv.2020.104814
This is a PDF file of an article that has undergone enhancements after acceptance, such
as the addition of a cover page and metadata, and formatting for readability, but it is
not yet the definitive version of record. This version will undergo additional copyediting,
typesetting and review before it is published in its final form, but we are providing this
version to give early visibility of the article. Please note that, during the production
process, errors may be discovered which could affect the content, and all legal disclaimers
that apply to the journal pertain.
©
2020 Published by Elsevier.

Journal Pre-proof
Novel benzoate-lipophilic cations selectively induce cell death in
human colorectal cancer cell lines
a
b
c
b
José Antonio Jara , Diego Rojas , Vicente Castro-Castillo , Sebastián Fuentes-Retamal ,
b
d
b
b
Cristian Sandoval-Acuña , Eduardo Parra , Mario Pavani , Juan Diego Maya , Jorge Ferreira
b*
b*
,
Mabel Catalán .
Running title: Benzoate-lipophilic cations induce cell death in colorectal cancer cells
Key words: Targeting mitochondria, benzoic acid derivatives, uncoupling agents, metabolism
stress, triphenylphosphonium moiety, colorectal cancer, antitumour-mitochondrial agents
a
Institute for Research in Dental Sciences (ICOD), Faculty of Dentistry, Universidad de Chile,
Santiago, Chile.
b
Clinical and Molecular Pharmacology Program, Institute of Biomedical Sciences (ICBM),
Faculty of Medicine, Universidad de Chile, Santiago, Chile.
c
Department of Physical Chemistry and Chemistry, Faculty of Chemical and Pharmaceutical
Sciences, Universidad de Chile, Santiago, Chile.
d
School of Medicine, Faculty of Health Sciences, University of Tarapacá, Av. General
Velásquez 1775, Arica 1000007, Chile.
*
Corresponding authors: Jorge Ferreira and Mabel Catalán
Clinical and Molecular Pharmacology Program, Institute of Biomedical Sciences (ICBM),
Faculty of Medicine, Universidad de Chile, Independencia 1027, Santiago, Chile
Telephone: +56 2 29789691
Fax: +56 2 27372783
emails: jferreir@med.uchile.cl (Jorge Ferreira)
mabelcatalan@med.uchile.cl (Mabel Catalán)
Both corresponding authors have contributed to read and approved the final manuscript.
1

Journal Pre-proof
Abstract
Introduction: Colorectal cancer (CRC) is a critical health issue worldwide. The high rate of
liver and lung metastasis associated with CRC creates a significant barrier to effective and
efficient therapy. Tumour cells, including CRC cells, have metabolic alterations, such as high
levels of glycolytic activity, increased cell proliferation and invasiveness, and chemo- and radio-
resistance. However, the abnormally elevated mitochondrial transmembrane potential of these
cells also provides an opportunity to develop drugs that selectively target the mitochondrial
functions of tumour cells.
Methods: In this work, we used a new batch of benzoic acid esters with cytotoxic activities
attached to the triphenylphosphonium group as a vehicle to target tumour mitochondria and
improve their activity. We evaluated the cytotoxicity, selectivity, and mechanism of action of
these derivatives, including the effects on energy stress-induced apoptosis and metabolic
behaviour in the human CRC cell lines HCT-15 and COLO-205.
Results: The benzoic acid derivatives selectively targeted the tumour cells with high potency
and efficacy. The derivatives induced the uncoupling of the oxidative phosphorylation system,
decreased the transmembrane potential, and reduced ATP levels while increasing AMPK
activation, thereby triggering tumour cell apoptosis in both tumour cell lines tested.
Conclusion: The benzoic acid derivatives studied here are promising candidates for assessing in
vivo models of CRC, despite the diverse metabolic characteristics of these tumour cells.
2

Journal Pre-proof
1
. Introduction
Colorectal cancer (CRC) is one of the most common cancers worldwide, causing approximately
00,000 deaths annually and representing a significant burden of disease. In Chile, CRC
prevalence is on the rise, accounting for almost 20% of cancer deaths during 2009 (Okada et al.,
016). The complete resection of the primary tumour is often a curative CRC treatment if it is
7
2
detected early. However, in advanced stages, adjuvant chemotherapy is usually required to
prevent local recurrence and metastasis. Few conventional chemotherapy drugs are available to
treat CRC, all of which have drawbacks, including drug resistance, loss of efficacy, and
numerous side effects (Granci et al., 2013). Furthermore, monoclonal antibodies for specific
targets are useful in only a portion of the CRC population, and numerous drug resistance
mechanisms have been reported for these therapies (Khelwatty et al., 2015; Mirone et al., 2016)
and have high associated costs.
Tumour cells exhibit profound metabolic alterations, such as high glucose uptake, high ATP
production through glycolysis at normal oxygen concentrations and the generation of large
quantities of lactate (Pavlova and Thompson, 2016; Warburg et al., 1927). This metabolic shift
towards a glycolytic profile confers bioenergy and biosynthetic advantages to proliferating
cancer cells by increasing the non-oxidative ATP production and generating intermediate
glucose metabolites that are routed to the tumour cells for biosynthetic processes (Zhang et al.,
2
015). Therefore, mitochondrial function is also altered in cancer cells. Cancer cells have a
significantly more negative mitochondrial transmembrane potential (∆m), a lower respiration
rate, and a lower mitochondrial mass than normal cells (Hoye et al., 2008; Moreno-Sanchez et
al., 2007; Ralph and Neuzil, 2009). This metabolic reprogramming facilitates tumour cell
proliferation, invasiveness, and chemo- and radio-resistance (Wahdan-Alaswad et al., 2013).
The differences between normal cell and tumour cell mitochondria provide a chemical
opportunity to selectively target this organelle. Given the mitochondrial hyperpolarization of
cancer cells, positively charged lipophilic molecules can be designed to accumulate within the
mitochondria (Kalyanaraman et al., 2017; Neuzil et al., 2013; Reily et al., 2013).
+
In addition, the triphenylphosphonium (TPP ) moiety can be used to enhance the transport of
+
molecules into the mitochondria (Murphy, 2008; Yang et al., 2016). Moreover, the TPP moiety
has been added to various pharmacologically active groups to produce new compounds intended
to target cancer cell mitochondria. These compounds have been found to show remarkable
cytotoxic activity in cancer cell lines, as well as anti-tumour effects in vivo (Millard et al., 2010;
Rokitskaya et al., 2016; Yasui et al., 2017).
+
Recently, our group described several new gallic acid (GA) derivatives linked to the TPP
moiety through an aliphatic chain. These GA derivatives had a cytotoxic effect, triggered
apoptosis in vitro, and showed an antitumoural effect in vivo in a murine carcinoma model (Jara
et al., 2014; Peredo-Silva et al., 2017).
This work expands upon these discoveries by evaluating the cytotoxicity and modes of action of
+
+
a new set of benzoic acid esters linked to TPP (benzoate-TPP compounds) in colorectal cancer
cell lines, including the effects on metastatic and drug-resistant behaviours.
3

Journal Pre-proof
2
2
. Materials and methods
.1. Materials
Roswell Park Memorial Institute (RPMI) 1640 culture medium, FCCP (carbonyl cyanide 4-
trifluoromethoxy)phenylhydrazone), glutamine, glucose, MTT, DMSO, foetal bovine serum
FBS), and Trypan Blue 0.4% solution were purchased from Sigma Chemical Co. (St. Louis,
(
(
MO). Minimum Essential Medium (MEM) with Earle's Balanced Salts (MEM/EBSS),
penicillin/streptomycin, and trypsin were purchased from HyClone Laboratories (South Logan,
UT). All other organic compounds and inorganic salts, acids, and solvents were purchased from
Merck (Darmstadt, Germany). The primary antibodies for AMPK-P, AMPK-T, and β-actin were
purchased from Cell Signalling Technologies (Boston, MA).
+
2
.2. Synthesis of benzoate-TPP derivatives
+
Benzoate-TPP derivatives were synthesized and characterized as previously described
Sandoval-Acuna et al., 2016). In detail, the first step consisted in the synthesis of (10-
(
hydroxydecyl)triphenylphosphonium bromide performed as follows: A solution of 10-
bromodecan-1-ol in dry acetonitrile (100 mL) was treated with triphenylphosphine. The solution
was refluxed with stirring for 48 h. The solvent was removed in a vacuum pump, and the crude
product was subjected to chromatography on silica gel (EtOAc, MeOH) to yield (10-
hydroxydecyl)triphenylphosphonium bromide as a colorless oil (63%). 1H NMR (400 MHz,
DMSO-d6): δ 1.01−1.51 (m, 18H, CH2), 3.57 (t, J = 7.1 Hz, 2H, CH2), 7.73−7.91 (m, 15H,
ArH). HRMS: m/z 485.4312 (calcd for C27H33BrOP 485.4357). The second step involved a
Steglich esterification,
using N,N-dicyclohexylcarbodiimide (DCC) as reagent, 4-
Dimethylaminopyridine (DMAP) as catalyst and N,N- dimethylformamide (DMF) as solvent.
Briefly, in an atmosphere of N2, a solution of each respective hydroxybenzoic acid in dry DMF
was treated with a solution of DCC in dry DMF. The mixture was cooled to 0 °C, and a solution
of (10-hydroxydecyl) triphenylphosphonium bromide and DMAP in DMF was added. The
reaction was stopped the next day, and any resulting precipitate was removed by filtering. The
solvent was then removed, producing a residue that was subjected to chromatography on silica
+
gel (DCM, MeOH). Figure S1 showed the benzoate-TPP C10 derivatives chemical
characterization.
2
.3. Cell lines and cultures
The human colon cancer cell lines HCT-15 and COLO 205 and the normal colon epithelial cell
line CCD841 CoN were acquired from the American Type Culture Collection (ATCC). Cells
were amplified from passages 0 to 10 and kept in liquid nitrogen until use. Colon cancer cells
were grown in RPMI 1640 medium. CCD 841 CoN cells were grown in MEM/EBSS. All cells
were supplemented with 10% FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin in a
humidified 5% CO2 / 95% air atmosphere at 37°C. For passages (no more than 20), cells were
detached using trypsin and seeded in plates for the various experiments.
2
.4. MTT cell viability
4
+
Cells were seeded into 96-well plates (1×10 per well). After 24 h, the benzoate-TPP
compounds were added in increasing concentrations. Cells were incubated for 24, 48, or 72 h.
4

Journal Pre-proof
After incubation, cells were washed twice with PBS, and 100 µL of 0.5 mg/mL MTT solution
was added to each well. After 2 h of incubation for the CRC cell lines and 4 h for the normal cell
line (as additional time is required to generate formazan crystals in normal vs. tumour cell lines),
the MTT was removed and the formazan crystals were dissolved in 50 µL of DMSO per well.
The absorbance was measured at 570 nm using a microplate ELISA reader (Bio-Rad, Hercules,
CA).
2
.5. Oxygen consumption and extracellular acidification rates
The oxygen consumption rates (OCR) of the COLO 205 and HCT-15 cell lines were measured
polarographically using Clark electrodes (Yellow Spring Instruments Co. Ohio, USa), as
described previously (Cordano et al., 2002; Plaza et al., 2008). The extracellular acidification
rate was measured using pH electrodes (Cole-Palmer, IL, USA), as described previously (Bruner
et al., 2008). CRC cells were cultured at 80% confluence. The cells were trypsinized and counted
6
6
using 7×10 cells from the HCT-15 cell line and 4×10 cells from the COLO 205 cell line for
each analysis. Briefly, 0.6 mL of cell suspension in medium consisting of 150 mM NaCl, 5 mM
KCl, and 1 mM Tris hydrochloride pH 7.0 along with 4 mM glutamine and 5 mM glucose as
substrates were added to the chamber containing the electrodes. For basal measurements,
+
respiration and H release rates were measured in the presence of substrates until complete
oxygen consumption was achieved (anoxia). The respiration rate was inhibited with 1 µg/mL+
oligomycin. OXPHOS was fully uncoupled with 0.155 µM FCCP as a control. OCR and H
release rates were evaluated by adding increasing concentrations of each benzoate-TPP
+
compound. The results were normalized to OCR in the presence of oligomycin and compared
with the effects of FCCP. Data analysis was performed using SigmaPlot 12 (San Jose, CA,
USA).
2
.6. Mitochondrial mass
To measure mitochondrial mass, 100 nM MitoTracker Green FM solution in RPMI 1640
5
medium without FBS was prepared. Detached cells (1×10 ) were deposited in a 1.5-mL
Eppendorf tube with 100 µL of 100 nM MitoTracker Green FM solution and incubated for 45
min. The cells were centrifuged, and the pellets were washed with PBS. After another
centrifugation, the cell pellets were resuspended in FBS-free medium. Cells were analysed using
Flow Cytometry FACS (FACS Aria® III, BD Biosciences) at a wavelength of 490 nm Ex/516
nm Em.
2
.7. Intracellular ATP levels
Intracellular ATP levels were measured using the CellTiter-Glo Luminescent Assay (Promega,
Madison, WI) according to the manufacturer’s protocol. Cells were seeded in 96-well plates
4
+
(
1×10 cells per well) and incubated for 24 h. Each well was stimulated with benzoate-TPP
compounds at the given concentrations for 4 h. Next, an aliquot of the cell suspension was
transferred to an opaque 96-well plate and maintained in the dark at room temperature for 10
min. Luminescence was measured using a Thermo Scientific Varioskan Flash spectral scanning
multimode reader.
2
.8. AMPK-P/AMPK-T Western blot analysis
+
Cells were treated with benzoate-TPP compounds for 4 h and then washed in cold PBS and
lysed with 40×-diluted RIPA buffer (50 mM Tris-HCl; 150 mM NaCl; 0.1% sodium dodecyl
5

Journal Pre-proof
sulfate) containing a proteinase and phosphatase inhibitor cocktail (Cell Signaling Technologies,
Boston, MA). A 40 µg sample of each protein was separated by 10% SDS-polyacrylamide gel
electrophoresis (SDS-PAGE), and the resolved proteins were transferred to a methylcellulose
membrane (Millipore, Billerica, MA). The membranes were blocked with 5% nonfat milk in
Tris-buffered saline with 0.1% Tween-20 at room temperature for 1 h and then incubated with
the primary antibody for AMPK-P (dilution 1:1000), AMPK-T (dilution 1:1000), or β-actin
(
dilution 1:1000) at 4°C overnight. Membranes were washed and then incubated with anti-rabbit
horseradish peroxidase-conjugated secondary antibody for 2 h at room temperature (dilution
:5000). After washing, the membranes were exposed to an enhanced chemiluminescent reagent
EZ-ECL). A ChemiDoc MP System was used for Western blotting. Digital blots were
quantified using ImageJ software.
1
(
2
.9. Changes in mitochondrial transmembrane potential
Changes in the transmembrane potential (∆Ψm) were determined using tetramethyl rhodamine
5
methyl ester (TMRM) as a probe. Cells were seeded into 24-well plates (1×10 cells per well) for
2
4 h and then incubated with 200 nM TMRM probe. Cells were immediately exposed to
increasing concentrations of each compound (2.5 to 30 µM) for 30 min. After washing with PBS,
cells were detached and resuspended in cold PBS for analysis using flow cytometry FACS
(FACSAria® III, BD Biosciences) at a wavelength of 540 nm Ex/595 nm Em.
2
.10. Flow cytometry measurement of apoptosis
The apoptotic events induced by the compounds were determined using an Annexin V-FITC
Apoptosis Detection Kit (Abcam, Cambridge, MA) according to the manufacturer’s protocol.
5
Cells were seeded in 24-well plates (5×10 cells per well) for 24 h. Cells were then incubated
with various concentrations of the compounds for 48 h. Cells were detached and resuspended in
2
50 µL of 1× Annexin buffer with 2.5 µL of Annexin V probe and 2.5 µL of propidium iodide
for each sample. Finally, cells were incubated for 5 min with the probes, light-protected, and
further analysed using flow cytometry (FACS Canto A, BD Biosciences). The laser wavelength
used was 488 nm and the emission wavelengths were FITC-A 530 nm for Annexin V and PerCP-
Cy5-5-A 617 nm for propidium iodide. A total of 10,000 events per sample were acquired, and
the data were analysed using Cyflogic software (non-commercial version, CyFlo Ltd.).
2
.11. Caspase 3 activation
Activation of caspase 3 by the compounds was performed using a PE Active Caspase-3
Apoptosis Kit (BD, Franklin Lakes, NJ) according to the manufacturer’s protocol. Briefly, cells
5
were seeded in 24-well plates (5×10 cells per well) for 24 h. Cells were then incubated with
various concentrations of the compounds for 24 h. Cells were detached and resuspended in cell
permeability/fixation buffer and incubated for 30 min in the dark at 4°C. Cells were centrifuged
and washed once with cell-staining buffer, centrifuged and resuspended in anti-caspase PE
antibody and incubated for 30 min in the dark. Finally, cells were washed once with cell-staining
buffer and analysed using flow cytometry at a wavelength of 488 nm Ex/617 nm Em. The data
were analysed using Cyflogic software (non-commercial version, CyFlo Ltd.).
2
.12. Statistical analysis
6

Journal Pre-proof
The data were analysed using one-way or two-way ANOVA and Bonferroni’s or Tukey’s post-
test. The IC50 was calculated using a dose-response curve with a variable slope in GraphPad
Prism 6. The differences were considered significant at p<0.05.
3
3
. Results
.1. Cytotoxicity and selectivity of benzoate-TPP compounds
+
In this work, we evaluated a series of lipophilic cations derived from 3,4,5 tri-hydroxybenzoic
+
acid (gallic acid, Figure S2A) esters with a 10-carbon aliphatic unit coupled to the TPP moiety
+
(
TPP C10) (Figure 1A), which were previously synthesized and characterized (Sandoval-Acuna
+
et al., 2016). Figure 1B and 1C illustrate the structures of these 2,5-di-hydroxy (GA-TPP C10)
and 2-hydroxy (SA-TPP C10) benzoic acid derivatives. The synthesis of the intermediates was
also evaluated as comparative molecules to characterize the TPP effect. The OH-C10, OH-
+
+
C10GA and OH-C10SA molecules (Figure S2D, S2E and S2F, respectively) correspond to the
+
+
esters of benzoic acid without the TPP moiety. Furthermore, the decyl ester (DA-TPP ) (Figure
S2G) was evaluated to establish the importance of benzoic acid on the cytotoxic effects and was
used as a control to characterize these effects.
Cytotoxicity was evaluated in two colorectal cancer cell lines with different tumour behaviours,
HCT-15 and COLO 205, derived from a non-metastatic tumour with resistance to standard drug
therapy and from a metastatic site, respectively. Figure 2 shows that the compounds SA-
+
+
+
TPP C10, GA-TPP C10, and TPP C10 induce a cytotoxic effect in a concentration- and time-
dependent manner in both HCT-15 (Figure 2A, 2B, and 2C, respectively) and COLO 205 cell
lines (Figure 2D, 2E, and 2F, respectively). Cytotoxicity was quantified by measuring the IC50
values of the three compounds in both cell lines. Table 1 shows the IC50 values for all
+
+
+
compounds at 24, 48, and 72 h. The compounds SA-TPP C10, GA-TPP C10, and TPP C10 had
IC50 values of 5–16 µM at 24 h, 3–14 µM at 48 h, and 3–12 µM at 72 h in both cell lines. All
compounds showed similar IC50 values at each time point evaluated (Table 1), with no
significant difference in terms of cytotoxic activity among any of the compounds. In addition, the
+
cytotoxic effect of the control molecules without the TPP moiety showed limited potency and
+
efficacy, with high IC50 values. However, DA-TPP produced a strong, non-selective cytotoxic
effect (Table 1; Figure S3A and S3B).
+
To measure selectivity, we evaluated the cytotoxicity of the benzoate-TPP compounds in CCD
8
41 CoN cells, a non-tumour colorectal epithelial cell line, and then calculated the selectivity
index by comparing the IC50 for non-tumour vs. tumour cells at 48 h and 72 h of incubation.
+
+
Table 2 shows that GA-TPP C10 had higher IC50 values for CCD 841 CoN than SA-TPP C10 or
+
TPP C10, a selectivity index of 10 for HCT-15 cells and 41 for COLO 205 cells after 48 h, and a
selectivity index of 3 for HCT-15 cells and 10 for COLO 205 cells after 72 h (Table 2). These
+
+
+
results show decreased viability after exposure to SA-TPP C10, GA-TPP C10, and TPP C10,
respectively, with high selectivity for the COLO 205 and HCT-15 cell lines vs. the non-tumour
cell line at 48 h. However, at 72 h, the differences were observed between the tumour and non-
+
+
+
tumour cell lines for SA-TPP C and GA-TPP C , but not TPP C , suggesting that SA-
1
0
10
10
+
+
+
TPP C10 and GA-TPP C10 were more selective for tumour cells than TPP C10.
7

Journal Pre-proof
Moreover, the results were supported by PI cell incorporation studies (Figure S2C). We showed
that none of the compounds significantly induced cell death, even at 100 µM, which represents a
1
0- and 40-fold improvement in IC50 in HCT-15 and COLO 205, respectively.
+
3
.2. Mechanism of action of benzoate-TPP compounds
Figure 3A and 3B show the results for the OCR and media acidification in the basal state. The
HCT-15 cells showed a lower OCR than the COLO 205 cells. In contrast, media acidification, as
measured by proton release into the medium, was considerably greater in the HCT-15 cells than
in the COLO 205 cells. The latter cells only showed increased rates of proton release when the
oxygen was depleted.
To evaluate OCR, tumour cells were deposited into the electrode chamber at time zero using
oligomycin to inhibit OXPHOS. The compounds or DMSO (as a control) were then added at
various concentrations (Figure 3C). The results were compared with values for the classical
uncoupler FCCP. Figure 3D, E, and F show that in HCT-15 cells, all three compounds increased
+
the oxygen consumption rate (that is, there was an uncoupling effect). SA-TPP C10 had an
uncoupling effect at concentrations above 10 µM, reaching the full uncoupling effect at 50 µM
+
+
(
Figure 3D). However, while GA-TPP C10 showed a pattern similar to that of SA-TPP C10, this
+
compound did not achieve the full uncoupling effect at any concentration (Figure 3E). TPP C10
also induced an uncoupling effect, but this compound was less potent than GA-TPP C10 or SA-
TPP C10 and failed to achieve the full uncoupling effect at any concentration (Figure 3F).
+
+
The three compounds were also able to exert an uncoupling effect in COLO 205 cells at lower
+
concentrations. SA-TPP C10 showed
a
concentration-dependent uncoupling effect at
concentrations above 5 µM, reaching the full uncoupling effect at 20 µM (Figure 3G). GA-
+
+
TPP C10 showed the same pattern as SA-TPP C10, reaching the full uncoupling effect at 10 µM
(
+
Figure 3H). TPP C10 produced similar results in the COLO 205 cells as in the HCT-15 cells,
showing a concentration-dependent uncoupling of OXPHOS (Figure 3F and 3I). This compound
+
+
was again less potent than GA-TPP C10 or SA-TPP C10, achieving the full uncoupling effect at
0 µM (Figure 3I).
3
Figure 3J and 3K show the total mitochondrial fluorescence visualized using MitoTracker Green
FM. This probe is incorporated into the mitochondria independent of transmembrane potential in
tumour and non-tumour cell lines. The results suggest that non-tumour cells (CCD 841 CoN)
contain a greater number of mitochondria than COLO 205 or HCT-15 cells.
+
3
.3. Metabolic effects induced by benzoate-TPP compounds
+
To evaluate the acute consequences induced by the uncoupling effect of benzoate-TPP
derivatives, we observed the loss of ∆m as an early and critical event that triggers cell death
after exposure to an uncoupling agent (Kroemer et al., 2007). All three compounds induced a
significant decrease in ∆m in both the HCT-15 (Figure 4A) and COLO 205 (Figure 4B) cell
+
lines. However, benzoate-TPP derivatives did not produce significant changes in ∆m in non-
tumour cells (Figure S4G), although these were estimated to have a larger number of
mitochondria compared to tumour cells (Figure 3J-K).
8

Journal Pre-proof
Additionally, as a direct consequence of the uncoupling effect on OXPHOS, the electron
transport chain detached from ATP synthesis. We measured intracellular ATP levels. HCT-15
and COLO 205 cells were incubated with various concentrations of each compound. Figure 4C
+
shows that the compounds decreased ATP levels in the HCT-15 cell line. Both SA-TPP C10 and
+
GA-TPP C10 induced a significant decrease in ATP levels at concentrations above 20 µM, while
+
TPP C10 only produced a modest albeit significant reduction above 30 µM. Similar results were
observed in COLO 205 cells (Figure 4D) without affecting cell viability (Figure S4A-F). To
evaluate the impact of the decreased intracellular ATP levels, we also evaluated the activity
phosphorylation of the essential metabolic sensor AMPK (AMP-activated protein kinase). We
+
+
+
used 20 µM SA-TPP C10, GA-TPP C10 and TPP C10 for this experiment, a concentration at
which a significant decrease in ATP level was observed in both CRC cell lines after 4 h of
incubation. Figure 4E and 4F show the results for AMPK phosphorylation induced by all
+
benzoate-TPP derivatives after 4 h of incubation in HCT-15 and COLO 205 cell lines,
respectively, compared to control values.
+
The results suggest that the uncoupling effect of benzoate-TPP compounds induces a loss of
∆
m, leading to decreased cellular ATP synthesis and activation of AMPK related to energetic
imbalance; these early events are associated with the induction of apoptosis and cellular death.
+
3
.4. Apoptosis induced by benzoate-TPP compounds
Loss of ∆m induces cell death. This apoptotic event may occur when cytochrome C is released
from the mitochondria into the cytoplasm and engages with the apoptosome to trigger the
apoptosis mechanism (Kroemer et al., 2007). We previously demonstrated that the compounds
studied here can induce a decrease in ∆m. However, it remained uncertain whether this loss of
∆m was pivotal in inducing apoptosis. In Figure 5, representative dot plots show that all
compounds induced apoptosis after 48 h of incubation in both the HCT-15 and COLO 205 cell
lines at the assessed concentrations (Figures 5A and 5B, respectively). The data show that higher
concentrations of the compounds were required to induce the apoptotic effect in HCT-15 cells
than in COLO 205 cells (graphical quantification). Moreover, the induction of apoptosis was
measured in both cell lines at an early time point (24 h). The differences between the effects of
the compounds compared to the control conditions were significant only in the COLO 205 cell
line. The HCT-15 cell line showed greater resistance to the effects of the compounds (Figure
S6). Furthermore, the results showed that the compounds assessed did not induce apoptotic death
in the non-tumour cell line (Figure 5C). In Figure S7, shows cell population selected for the three
cell lines to measure the apoptosis death and cell debris not considered for data analysis.
Figure 6 shows that all compounds induced caspase 3 activation after 24 h of incubation in both
cell lines. However, the HCT-15 cell line (Figure 6A) showed more resistance to this caspase
activation than the COLO 205 cell line (Figure 6B).
+
The results suggest that benzoate-TPP derivatives are capable of inducing apoptosis, an event
that occurs after the onset of the uncoupling effect and the decrease in ∆m. In spite of the
resistance of the HCT-15 cell line to apoptotic death, the compounds may be able to induce
apoptosis at increased concentrations of the drug.
9

Journal Pre-proof
4
. Discussion
Hydroxybenzoic acid derivatives produce cytotoxic effects in tumour cells
Several recent reports have provided evidence of the cytotoxic activity of gallic acid (GA), a
polyphenolic compound found in many foods common to the Mediterranean diet. GA has
antioxidant effects in many types of cells, as well as a broad range of biological activities,
including anticancer and anti-inflammatory properties (Inoue et al., 1995; Kim et al., 2006; Park
et al., 2008). Specifically, GA alone has a selective cytotoxic effect on many human tumour cell
lines, including mammary, colon (such as HCT-15 and COLO 205), and prostate (Inoue et al.,
1
995; Russell et al., 2012; Salucci et al., 2002; Subramanian et al., 2016). However, the effects
of GA alone have been documented in vitro at high concentrations, which are difficult to achieve
in vivo (Shahrzad et al., 2001). On the other hand, several GA ester derivatives have been
synthetized, showing cytotoxicity with respect to their lipophilicity and decreasing the IC50 value
compared to GA in various tumour cell lines (Frey et al., 2007).
To improve efficacy in some pharmacophore groups and to target mitochondria, cationic
molecules attached to the main structure have been shown to be a potential tool to increase the
+
amount of drug in these organelles (Yang et al., 2016). The TPP group may have the potential to
improve the cytotoxicity and selectivity of several pharmacophore groups, such as GA esters.
+
Moreover, TPP is a very widely used moiety to concentrate molecules in the mitochondria,
especially due to its delocalized charge between the phenyl groups present in its structure
+
+
(
Trendeleva et al., 2013). Our results showed that the TPP moiety in our GA ester (TPP C10)
improves its cytotoxic effects, increasing the potency and efficacy with respect to the GA ester,
as we show with a comparison of IC50 values (Table 1), and these concentrations may be
+
achievable in vivo. We previously reported high TPP C10 efficacy in vitro and in vivo in murine
mammary adenocarcinoma (Jara et al., 2014; Peredo-Silva et al., 2017). Our data show that
+
TPP C10 produces more potent cytotoxic effects than GA in CRC cells as reported in other
investigations (Subramanian et al., 2016).
Additionally, salicylic acid and gentisic acid (Figure S2B and S2C) (mono- and
dihydroxybenzoic acid, respectively) have been reported to be chemopreventive agents (Hua et
al., 2019). Gentisic acid has been described as the effector of aspirin-antitumour effects in brain
and breast cancer, with a non-specified target (Altinoz et al., 2018). Otherwise, both acids alone
have been described as CDK inhibitors in CRC, but at higher concentrations (200-300 µM)
+
(
Dachineni et al., 2017). Our results showed that mono- and dihydroxy benzoate-TPP
+
+
derivatives from salicylic and gentisic acid (SA-TPP C10 and GA-TPP C10, respectively)
+
produce similar cytotoxicity to the trihydroxy benzoic derivative (TPP C10) in HCT-15 and
+
COLO 205 cells, and these effects were more potent than the ester derivatives without TPP
+
moiety (Table 1). Altogether, our data reveal that TPP improves the cytotoxic effects with
respect to the acids and esters, comparing our results with the previously described potencies of
salicylic and gentisic acids.
+
Moreover, the results suggest that GA-TPP C has a higher selectivity for tumour cells than
1
0
+
+
TPP C10 or SA-TPP C10 (Table 2). Higher concentrations of all of the compounds assessed were
required to produce cytotoxic effects in non-tumour cells (Table 1, Figure S2).
10

Journal Pre-proof
+
Benzoate-TPP derivatives targeting mitochondria
The cytotoxic effects of GA and its esters are induced through the generation of reactive oxygen
species (ROS) and the inhibition of the electron transport chain complex (Chen et al., 2009; Frey
+
et al., 2007). In this study, we show that conjugation of a GA ester with TPP confers the ability
to induce mitochondrial uncoupling (Figure 2) without increasing ROS levels (Figure S5A and
S5B). This event represents a novel mode of action that is different than the mechanisms
previously described for GA or GA esters (Chen et al., 2009; Frey et al., 2007). These
uncoupling effects can be explained by the protonophore characteristics of the hydroxy groups
present on the phenyl rings of GA. These hydroxy groups are protonated on the outer side of the
internal mitochondrial membrane (intermembrane space) and are deprotonated on the inner side
of the internal mitochondrial membrane (matrix), according to the respective pKa values of the
hydroxy groups (Sandoval-Acuna et al., 2016). This mechanism of action is also present for SA-
+
+
+
TPP C10 and GA-TPP C10, showing similar uncoupling effects to TPP C10 in CRC cell lines.
Therefore, the results suggest that the uncoupling effect is independent of the number of hydroxy
groups present on the benzoic acid ring. In addition, mono- and dihydroxybenzoic acids are
described as CDK inhibitors (Dachineni et al., 2017); however, our results in CRC cells show
+
that esterification and TPP incorporation into the structure may modified their mechanism of
action.
The 10-carbon length of the aliphatic chain has been described as an important characteristic to
penetrate the mitochondria and produce the uncoupling effect. These data are consistent with
findings from Antonenko and colleagues (Antonenko et al., 2016), who reported that fluorescein
+
linked to TPP with a 10-carbon aliphatic chain induces the uncoupling of OXPHOS, decreasing
ROS and providing a neuroprotective effect in trauma models (Antonenko et al., 2016; Denisov
+
et al., 2014). Interestingly, the aliphatic chain attached to TPP alone, without the
+
+
pharmacophore group (DA-TPP ), has higher cytotoxicity than benzoate-TPP compounds
Table 1, Figure S1G). However, this compound induces necrotic cell death and targets tumour
(
and non-tumour cells in a nonspecific manner, as shown in our results and those results
previously reported by our group in breast tumour cells (Sandoval-Acuna et al., 2016).
+
Moreover, the lack of a TPP moiety on the benzoate with 10-carbon aliphatic chain derivatives
(OH-C10, OH-C10GA and OH-C10SA) reduces the potency and efficacy of the pharmacophore
group, suggesting a decrease in the mitochondrial target. Further experiments may be performed
to confirm this affirmation.
+
Benzoate-TPP derivatives act independent of cellular metabolism
+
Our data show similar cytotoxic results for each of the benzoate TPP derivatives in the CRC
lines, independent of cell metabolic characteristics, for both glycolytic and oxidative behaviours
(Figure 3A and 3B). We show that HCT-15 cells behave as a glycolytic cell type. This result is
consistent with the finding that HCT-15 cells show significantly greater rates of proton
production at baseline than the COLO 205 cells. In contrast, the HCT-15 cells show a lower
cellular respiration rate than the COLO 205 cells, which seems to rely more on functional
oxidative phosphorylation. It has been reported that metastatic cells present more mitochondrial
mass and OXPHOS activity surrounded by a glycolytic environment that has mitochondrial
dysfunction and loss of mitochondrial mass, which is associated with drug resistance (Sotgia et
al., 2012). Moreover, the increased drug resistance has been suggested to be the result of less
11

Journal Pre-proof
mitochondrial DNA. We showed that the HCT-15 cell line has a significantly lower
mitochondrial mass than the COLO 205 cell line (Figure 3J-K), which may partly explain the
slight resistance of the HCT-15 cell line to our compounds, which likely influences drug
targeting. However, these molecules can generate a potent cytotoxic effect despite the glycolytic
behaviours of the cells. Altogether, our data suggest that the three compounds are able to induce
tumour cell death regardless of the metabolic features of each type of cell, which is in agreement
with our previous results in breast cancer cells (Sandoval-Acuna et al., 2016). In this sense, the
compounds showed similar effects and selectivity in CRC cells when compared to murine and
human breast cancer cells, leading to the speculation that these new molecules are nonspecific
among tumour cell lines and have a high selectivity for them, and the principal feature involved
is the mitochondrial differences in cancer cells.
+
Benzoate-TPP derivatives induce apoptotic death by mitochondrial stress
As a direct consequence of this mitochondrial disturbance produced by the uncoupling effect,
∆
m and ATP levels are reduced. These changes trigger an increased AMP/ATP ratio, which
activates AMPK phosphorylation, a pivotal metabolic sensor (Lim et al., 2009; Sergent et al.,
002). Our results are in agreement with this mechanism, which implies that our compounds
2
+
+
+
(SA-TPP C10, GA-TPP C10 and TPP C10) are able to reach the mitochondria and induce an
uncoupling effect, which triggers the collapse of the function of these organelles, causes them
enter metabolic stress, decreases cellular ATP levels, and leads the cells to apoptosis.
It should be noted that while the IC50 values in HCT-15 cells were higher than in COLO 205
cells (Table 1, 8-15 in HCT-15 cells), these values are still much lower than the IC50 value for 5-
FU, a classical CRC drug treatment (47.5 µM), in this cell line (Bracht et al., 2010). In addition,
our own results showed that 5-FU induces cytotoxic effects in COLO 205 cells with a higher
+
+
+
IC50 than our compounds (Figure S8), suggesting that SA-TPP C10, GA-TPP C10 and TPP C10
are improved molecules, inducing cell death selectivity compared with a classical treatment for
CRC.
On the other hand, the compounds did not induce a loss in ∆m or apoptosis (Figures S2G and
5
E) in non-tumour cells. These results confirm the higher selectivity of these compounds for
tumour cells vs. normal cells.
5
. Conclusion
Our results show that benzoate-derivative compounds show potential as anti-colorectal cancer
drugs, operating through different modes of action than the current therapies for this disease.
These new molecules selectively target mitochondria with high potency and efficacy, inducing
the uncoupling of oxidative phosphorylation, which in turn triggers apoptotic cell death. The
+
mechanism of action of these compounds is associated with the TPP moiety added to the
pharmacophore, which produces mitochondrial tropism. The benzoic acid derivatives studied
here are promising candidates for assessing in vivo models of CRC.
12

Journal Pre-proof
Declarations
Ethics approval and consent to participate: Not applicable
Consent to publish: Not applicable
Availability of data and materials: Not applicable. Our manuscript did not contain any data.
Competing interests: The authors declare no competing financial interest.
Funding: This work was supported by the following grants from the Consejo Nacional de
Ciencia y Tecnología (CONICYT: FONDECYT 11160281 (MC), FONDECYT 1180296 (JF),
FONDECYT 1130189 (JDM), the Academy Insertion Grant 791220004 (MC-JAJ); from the
Vicerrectoría de Investigación y Desarrollo, Universidad de Chile (Enlace ENL022/16) (JF).
The grants CONICYT: FONDECYT 11160281 (MC) and the Academy Insertion Grant
7
91220004 allowed the development of the project, permeating the purchase of supplies and cell
lines, the design of the study and collection, analysis, and interpretation of data.
The grants FONDECYT 1180296 (JF), FONDECYT 1130189 (JDM) and Universidad de Chile
(Enlace ENL022/16) (JF) allowed the operation of the laboratory and the writing and revision of
the manuscript.
Authors' Contributions: J.A.J., M.C, and J.F. designed the study, analyzed the data, and wrote
the manuscript. J.A.J., D.R., M.P., M.C. and J. F. performed the experimental work. V.C.C.
performed the organic synthesis. S.F.R., C.S.A., E.P. and J.D.M. contributed to project
development and data interpretation.
We declare that all authors have read and approved the manuscript.
Acknowledgements: This work was supported by the following grants from the Consejo
Nacional de Ciencia y Tecnología (CONICYT: FONDECYT 11160281 (MC), FONDECYT
1
180296 (JF), FONDECYT 1130189 (JDM), the Academy Insertion Grant 791220004 (MC-
JAJ)); from the Vicerrectoría de Investigación y Desarrollo, Universidad de Chile (Enlace
ENL022/16) (JF).
Authors' Information:
e-mails:
José Antonio Jara: jsandovalj@u.uchile.cl
Diego Rojas: diegorojas@umce.cl
Vicente Castro-Castillo: vicente.castro@ciq.uchile.cl
Sebastián Fuentes-Retamal: sebastianfuentes@ug.uchile.cl
Cristian Sandoval-Acuña: casacuna@ciq.uchile.cl
Eduardo Parra: eparaarica@gmail.com
Mario Pavani: mpavani@med.uchile.cl
Juan Diego Maya: jmaya@med.uchile.cl
*
Corresponding authors: Jorge Ferreira and Mabel Catalán
Clinical and Molecular Pharmacology Program, Institute of Biomedical Sciences (ICBM),
Faculty of Medicine, Universidad de Chile, Independencia 1027, Santiago, Chile
Telephone: +56 2 29789691
Fax: +56 2 27372783
emails: jferreir@med.uchile.cl (Jorge Ferreira)
mabelcatalan@med.uchile.cl (Mabel Catalán)
Both corresponding authors have contributed to read and approved the final manuscript.
13

Journal Pre-proof
Table legend
+
Table 1. Cytotoxicity and selectivity of benzoate-TPP compounds in the colon cancer cell lines HCT-15 and COLO 205.
IC = concentration necessary to achieve 50% toxicity in the cell culture after 24, 48, or 72 h of treatment. Values were calculated
5
0
from the respective sigmoidal dose-response curves, representing the mean ± SD values of at least three independent experiments.
a
Each assay was performed in triplicate. indicates a significant difference (p<0.0001) between compounds for CCD 841 CoN; ** or
**** indicate a significant difference (p<0.001 or p<0.0001) between tumor and non-tumor cells for a given compound. p-values were
calculated using a two-way ANOVA and Tukey’s post-hoc test.
N/A: not analyze
+
IC50 values (µM) for benzoates-TPP derivatives and control molecules in colorectal cancer cells
24 h
48 h
72 h
CCD841
(non-tumorcell
line)
CCD841
(non-tumorcell
line)
Compound
HCT-15
10.90± 1.24
13.02 ± 2.51
15.27 ± 1.06
>200
COLO205
5.8 ± 0.64
5.3 ± 0.49
9.54 ± 0.46
>200
HCT-15
COLO205
HCT15
COLO205
+
SA-TPP C10
8.35 ± 0.73**
11.32 ± 1.3****
14.23 ± 1.3**
>200
4.07 ± 0.61**** 27.45 ± 3.73
2.77 ± 0.61**** 114.72 ± 25.1
3.64 ± 0.14**** 38.49 ± 7.52
8.69 ± 0.82*
10.22 ± 0.47**
12.20 ± 0.43
>200
3.24 ± 0.83****
3.01 ± 0.28****
4.01 ± 0.57
>200
17.86 ± 5.94
+
a
a
GA-TPP C10
31.02 ± 13.6
+
TPP C10
8.24 ± 3.06
N/A
OH-C10SA
OH-C10GA
>200
N/A
N/A
N/A
>200
46,75 ± 1,58
>400
>200
40,60 ± 3,62
>400
>200
32,85 ± 5,88
>400
N/A
OH-C10
>400
>400
>400
N/A
+
C
10TPP
N/A
2.24 ± 0.11
N/A
0.68 ± 0.08
N/A
0.63 ± 0.02
0.58 ± 0.10
N/A
14

Journal Pre-proof
Table 2.
+
Selectivity index for benzoates-TPP derivatives at 48 h and 72 h.
Values were calculated from IC50 ratio between non-tumor cell and tumor cell lines from the
respective sigmoidal dose-response curves.
Selectivity Index for benzoates-TPP+ derivatives at 48 h and 72 h
4
8 h
72 h
CCD 841 CoN/
CCD 841 CoN/
CCD 841 CoN/
CCD 841 CoN/
Compound
HCT-15
3.3
COLO 205
6.7
HCT-15
2.1
COLO 205
5.5
+
SA-TPP C10
+
10.1
2.7
41.4
3.0
10.3
GA-TPP C10
+
10.6
0.68
2.1
TPP C10
15

Journal Pre-proof
Figure legends
+
Figure 1. Structures of benzoate-TPP compounds linked to the triphenylphosphonium
+
moiety (TPP ), allowing for mitochondrial tropism. Structures were obtained using
ChemDraw 9.0 software. A) (10-((3,4,5-trihydroxybenzoyl)oxy)decyl)triphenylphosphonium
+
bromide (TPP C10), B) (10-((2,5-dihydroxybenzoyl)oxy)decyl)triphenylphosphonium bromide
+
(
GA-TPP C10), C) (10-((2-hydroxybenzoyl)oxy)decyl)triphenylphosphonium bromide (SA-
+
TPP C10), D) 10-hydroxydecil-3,4,5-trihydroxybenzoate, (OH-C10) E) 10-hydroxydecil-2,5-
dihydroxybenzoate (OH-C10GA), F) 10-hydroxydecil-2-hydroxybenzoate (OH-C10SA) and G)
+
decyltriphenylphosphonium bromide (C10TPP ).
+
Figure 2. Cell viability of colon cancer cell lines after exposure to benzoate-TPP
compounds.
The MTT assay was performed at various incubation times (24, 48, and 72 h). A), B), And C) are
sigmoidal dose-response curves (variable slope), providing a graphical representation of the
+
+
+
cytotoxicity of SA-TPP C10, GA-TPP C10, and TPP C10 in HCT-15 cells, respectively. D), E),
+
+
And F) are sigmoidal dose-response curves for the benzoate-TPP compounds SA-TPP C10, GA-
TPP C10, and TPP C10 in COLO 205 cells, respectively. The values represent the mean of at
least three independent experiments ± SD. Each assay was performed in triplicate.
+
+
+
Figure 3. Uncoupling effects of benzoate-TPP compounds in colon cancer cell lines.
Curves A) and B) show representative traces for the baseline O2 consumption rate (OCR) and H
+
release rate of the HCT-15 and COLO 205 cell lines, respectively. The numbers above the curve
+
represent nmol O/min*mg protein, and the numbers under the curve represent nmol H /min*mg
protein. The traces were fitted using SigmaPlot 12 software. C) Schematic graphic of cell
respiration and procedures. D), E) And F) show the OCR of HCT-15 cells after exposure to SA-
+
+
+
TPP C10, GA-TPP C10, and TPP C10, respectively. G), H), And I) show the OCR of COLO 205
+
+
+
cells after exposure to SA-TPP C10, GA-TPP C10 or TPP C10, respectively. The OCR in the
presence of the experimental compounds was normalized to the OCR in the presence of
oligomycin and is compared with the OCR in the presence of FCCP. The data shown represent
the mean ± SD values of at least five independent experiments. Significant differences between
the experimental conditions vs. control (FCCP) are indicated by asterisks (one-way ANOVA,
Bonferroni post hoc test); *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001. J) Plot of
mitochondrial mass from tumour and non-tumour cells by using MitoTracker Green FM. K)
Quantification of mitochondrial mass of tumour and non-tumour cell lines. The mitochondrial
mass values shown represent the mean ± SD of three independent experiments. The data were
normalized to the non-tumour cells (CCD 841 CoN). Significant differences between the tumour
and non-tumour cells are indicated by asterisks (one-way ANOVA, Bonferroni’s post hoc test);
*
*p<0.01; ***p<0.001.
+
Figure 4. Metabolic alterations produced by benzoate-TPP compounds in colon cancer cell
lines.
+
+
+
The effects of various concentrations of SA-TPP C , GA-TPP C , and TPP C on ∆Ψm were
1
0
10
10
measured using a TMRM probe, sensitive to mitochondrial-transmembrane potential, after 30
min of incubation with the compounds in A) HCT-15 and B) COLO 205 cells. The data showed
the dissipation of intensity of the TMRM probe, signal of mitochondrial potential loss. The data
16

Journal Pre-proof
were normalized to the control values. The ∆Ψm values shown represent the mean of four
+
independent experiments ± SD. Besides, effects of various concentrations of SA-TPP C10, GA-
+
+
TPP C10, and TPP C10 after 4 h of incubation on intracellular ATP levels, measured by CellTiter-
Glo Luminescent Assay, in C) HCT-15 and D) COLO 205 cells. The values showed the average
ATP levels for each condition and the comparison with the control. Finally, AMPK
phosphorylation induced by each compound (20 µM) after 4 h of treatment was evaluated using
western blot analysis in E) HCT-15 and F) COLO 205 cells. The AMPK-P and AMPK-T values
shown represent the mean of five independent experiments. Significant differences between the
experimental conditions and control are indicated by asterisks (one-way ANOVA, Bonferroni
post hoc test); *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.
+
Figure 5. Benzoate-TPP compounds induce apoptosis in colon cancer cell lines.
Apoptosis levels induced by the compounds after 48 h of incubation were assessed in HCT-15,
COLO 205 and CCD841 CoN cell lines. Representative scatterplots of higher concentrations
tested are shown for A) HCT-15 and B) COLO 205 C) CCD841 CoN cells. Apoptosis
measurement quantifications are graphically HCT-15, D) COLO 205. Assessments were
performed using Cyflogic software (version 1.2.1). Staurosporine (STS; 10 μM) was used as a
positive apoptosis-induction control. Apoptosis was assessed by quantifying Annexin V-FITC
positive cells using flow cytometry (FACS Canto A, BD Biosciences). The data shown represent
the mean ± SD of four independent experiments. Significant differences between the
experimental conditions and the control are indicated by asterisks (one-way ANOVA,
Bonferroni’s post hoc test); *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.
+
Figure 6. Benzoate-TPP compounds induce the activation of caspase 3.
Compound-induced caspase 3 activation after 24 h of incubation was assessed in A) HCT-15 and
B) COLO 205 cell lines using PE anti-caspase 3 antibody. Assessments were performed using
Cyflogic software (version 1.2.1). Staurosporine (STS; 10 μM) was used as a caspase 3-
induction control. Caspase 3 activation was assessed by quantifying PE-positive cells using flow
cytometry (FACS Canto, BD Biosciences). The data shown represent the mean ± SD of four
independent experiments. Significant differences between the experimental conditions and the
control are indicated by asterisks (one-way ANOVA, Bonferroni’s post hoc test); *p<0.05;
*
*p<0.01; ***p<0.001; ****p<0.0001.
17

Journal Pre-proof
References
Altinoz, M.A., Elmaci, I., Cengiz, S., Emekli-Alturfan, E., Ozpinar, A., 2018. From
epidemiology to treatment: Aspirin's prevention of brain and breast-cancer and cardioprotection
may associate with its metabolite gentisic acid. Chemico-biological interactions 291, 29-39.
Antonenko, Y.N., Denisov, S.S., Silachev, D.N., Khailova, L.S., Jankauskas, S.S., Rokitskaya,
T.I., Danilina, T.I., Kotova, E.A., Korshunova, G.A., Plotnikov, E.Y., Zorov, D.B., 2016. A
long-linker conjugate of fluorescein and triphenylphosphonium as mitochondria-targeted
uncoupler and fluorescent neuro- and nephroprotector. Biochim Biophys Acta 1860, 2463-2473.
Bracht, K., Nicholls, A.M., Liu, Y., Bodmer, W.F., 2010. 5-Fluorouracil response in a large
panel of colorectal cancer cell lines is associated with mismatch repair deficiency. Br J Cancer
1
03, 340-346.
Cordano, G., Pezoa, J., Munoz, S., Rivera, E., Medina, J., Nunez-Vergara, L.J., Pavani, M.,
Guerrero, A., Ferreira, J., 2002. Inhibitory effect of vanillin-like compounds on respiration and
growth of adenocarcinoma TA3 and its multiresistant variant TA3-MTX-R. Eur J Pharm Sci 16,
2
55-263.
Chen, H.M., Wu, Y.C., Chia, Y.C., Chang, F.R., Hsu, H.K., Hsieh, Y.C., Chen, C.C., Yuan, S.S.,
009. Gallic acid, a major component of Toona sinensis leaf extracts, contains a ROS-mediated
2
anti-cancer activity in human prostate cancer cells. Cancer Lett 286, 161-171.
Dachineni, R., Kumar, D.R., Callegari, E., Kesharwani, S.S., Sankaranarayanan, R., Seefeldt, T.,
Tummala, H., Bhat, G.J., 2017. Salicylic acid metabolites and derivatives inhibit CDK activity:
Novel insights into aspirin's chemopreventive effects against colorectal cancer. International
journal of oncology 51, 1661-1673.
Denisov, S.S., Kotova, E.A., Plotnikov, E.Y., Tikhonov, A.A., Zorov, D.B., Korshunova, G.A.,
Antonenko, Y.N., 2014. A mitochondria-targeted protonophoric uncoupler derived from
fluorescein. Chem Commun (Camb) 50, 15366-15369.
Frey, C., Pavani, M., Cordano, G., Munoz, S., Rivera, E., Medina, J., Morello, A., Diego Maya,
J., Ferreira, J., 2007. Comparative cytotoxicity of alkyl gallates on mouse tumor cell lines and
isolated rat hepatocytes. Comp Biochem Physiol A Mol Integr Physiol 146, 520-527.
Granci, V., Cai, F., Lecumberri, E., Clerc, A., Dupertuis, Y.M., Pichard, C., 2013. Colon cancer
cell chemosensitisation by fish oil emulsion involves apoptotic mitochondria pathway. Br J Nutr
1
09, 1188-1195.
Hoye, A.T., Davoren, J.E., Wipf, P., Fink, M.P., Kagan, V.E., 2008. Targeting mitochondria.
Acc Chem Res 41, 87-97.
18

Journal Pre-proof
Hua, H., Zhang, H., Kong, Q., Wang, J., Jiang, Y., 2019. Complex roles of the old drug aspirin in
cancer chemoprevention and therapy. Medicinal research reviews 39, 114-145.
Inoue, M., Suzuki, R., Sakaguchi, N., Li, Z., Takeda, T., Ogihara, Y., Jiang, B.Y., Chen, Y.,
1
995. Selective induction of cell death in cancer cells by gallic acid. Biol Pharm Bull 18, 1526-
1
530.
Jara, J.A., Castro-Castillo, V., Saavedra-Olavarria, J., Peredo, L., Pavanni, M., Jana, F., Letelier,
M.E., Parra, E., Becker, M.I., Morello, A., Kemmerling, U., Maya, J.D., Ferreira, J., 2014.
Antiproliferative and uncoupling effects of delocalized, lipophilic, cationic gallic acid derivatives
on cancer cell lines. Validation in vivo in singenic mice. J Med Chem 57, 2440-2454.
Kalyanaraman, B., Cheng, G., Hardy, M., Ouari, O., Sikora, A., Zielonka, J., Dwinell, M.B.,
2
017. Modified Metformin as a More Potent Anticancer Drug: Mitochondrial Inhibition, Redox
Signaling, Antiproliferative Effects and Future EPR Studies. Cell biochemistry and biophysics
5, 311-317.
7
Khelwatty, S.A., Essapen, S., Seddon, A.M., Fan, Z., Modjtahedi, H., 2015. Acquired resistance
to anti-EGFR mAb ICR62 in cancer cells is accompanied by an increased EGFR expression,
HER-2/HER-3 signalling and sensitivity to pan HER blockers. Br J Cancer 113, 1010-1019.
Kim, S.H., Jun, C.D., Suk, K., Choi, B.J., Lim, H., Park, S., Lee, S.H., Shin, H.Y., Kim, D.K.,
Shin, T.Y., 2006. Gallic acid inhibits histamine release and pro-inflammatory cytokine
production in mast cells. Toxicol Sci 91, 123-131.
Kroemer, G., Galluzzi, L., Brenner, C., 2007. Mitochondrial membrane permeabilization in cell
death. Physiological reviews 87, 99-163.
Lim, H.W., Lim, H.Y., Wong, K.P., 2009. Uncoupling of oxidative phosphorylation by
curcumin: implication of its cellular mechanism of action. Biochem Biophys Res Commun 389,
1
87-192.
Millard, M., Pathania, D., Shabaik, Y., Taheri, L., Deng, J., Neamati, N., 2010. Preclinical
evaluation of novel triphenylphosphonium salts with broad-spectrum activity. PLoS One 5.
Mirone, G., Shukla, A., Marfe, G., 2016. Signaling mechanisms of resistance to EGFR- and
Anti-Angiogenic Inhibitors cancer. Crit Rev Oncol Hematol 97, 85-95.
Moreno-Sanchez, R., Rodriguez-Enriquez, S., Marin-Hernandez, A., Saavedra, E., 2007. Energy
metabolism in tumor cells. FEBS J 274, 1393-1418.
Murphy, M.P., 2008. Targeting lipophilic cations to mitochondria. Biochim Biophys Acta 1777,
1
028-1031.
19

Journal Pre-proof
Neuzil, J., Dong, L.F., Rohlena, J., Truksa, J., Ralph, S.J., 2013. Classification of mitocans, anti-
cancer drugs acting on mitochondria. Mitochondrion 13, 199-208.
Okada, T., Tanaka, K., Kawachi, H., Ito, T., Nishikage, T., Odagaki, T., Zarate, A.J., Kronberg,
U., Lopez-Kostner, F., Karelovic, S., Flores, S., Estela, R., Tsubaki, M., Uetake, H., Eishi, Y.,
Kawano, T., 2016. International collaboration between Japan and Chile to improve detection
rates in colorectal cancer screening. Cancer 122, 71-77.
Park, W., Chang, M.S., Kim, H., Choi, H.Y., Yang, W.M., Kim, D.R., Park, E.H., Park, S.K.,
2
008. Cytotoxic effect of gallic acid on testicular cell lines with increasing H2O2 level in GC-1
spg cells. Toxicol In Vitro 22, 159-163.
Pavlova, N.N., Thompson, C.B., 2016. The Emerging Hallmarks of Cancer Metabolism. Cell
Metab 23, 27-47.
Peredo-Silva, L., Fuentes-Retamal, S., Sandoval-Acuna, C., Pavani, M., Maya, J.D., Castro-
Castillo, V., Madrid-Rojas, M., Rebolledo, S., Kemmerling, U., Parra, E., Ferreira, J., 2017.
Derivatives of alkyl gallate triphenylphosphonium exhibit antitumor activity in a syngeneic
murine model of mammary adenocarcinoma. Toxicol Appl Pharmacol 329, 334-346.
Plaza, C., Pavani, M., Faundez, M., Maya, J.D., Morello, A., Becker, M.I., De Ioannes, A.,
Cumsille, M.A., Ferreira, J., 2008. Inhibitory effect of nordihydroguaiaretic acid and its tetra-
acetylated derivative on respiration and growth of adenocarcinoma TA3 and its multiresistant
variant TA3MTX-R. In Vivo 22, 353-361.
Ralph, S.J., Neuzil, J., 2009. Mitochondria as targets for cancer therapy. Mol Nutr Food Res 53,
9
-28.
Reily, C., Mitchell, T., Chacko, B.K., Benavides, G., Murphy, M.P., Darley-Usmar, V., 2013.
Mitochondrially targeted compounds and their impact on cellular bioenergetics. Redox Biol 1,
8
6-93.
Rokitskaya, T.I., Murphy, M.P., Skulachev, V.P., Antonenko, Y.N., 2016. Ubiquinol and
plastoquinol triphenylphosphonium conjugates can carry electrons through phospholipid
membranes. Bioelectrochemistry 111, 23-30.
Russell, L.H., Jr., Mazzio, E., Badisa, R.B., Zhu, Z.P., Agharahimi, M., Oriaku, E.T., Goodman,
C.B., 2012. Autoxidation of gallic acid induces ROS-dependent death in human prostate cancer
LNCaP cells. Anticancer Res 32, 1595-1602.
Salucci, M., Stivala, L.A., Maiani, G., Bugianesi, R., Vannini, V., 2002. Flavonoids uptake and
their effect on cell cycle of human colon adenocarcinoma cells (Caco2). Br J Cancer 86, 1645-
1
651.
20

Journal Pre-proof
Sandoval-Acuna, C., Fuentes-Retamal, S., Guzman-Rivera, D., Peredo-Silva, L., Madrid-Rojas,
M., Rebolledo, S., Castro-Castillo, V., Pavani, M., Catalan, M., Maya, J.D., Jara, J.A., Parra, E.,
Calaf, G.M., Speisky, H., Ferreira, J., 2016. Destabilization of mitochondrial functions as a target
against breast cancer progression: Role of TPP(+)-linked-polyhydroxybenzoates. Toxicol Appl
Pharmacol 309, 2-14.
Sergent, C., Franco, N., Chapusot, C., Lizard-Nacol, S., Isambert, N., Correia, M., Chauffert, B.,
2
002. Human colon cancer cells surviving high doses of cisplatin or oxaliplatin in vitro are not
defective in DNA mismatch repair proteins. Cancer Chemother Pharmacol 49, 445-452.
Shahrzad, S., Aoyagi, K., Winter, A., Koyama, A., Bitsch, I., 2001. Pharmacokinetics of gallic
acid and its relative bioavailability from tea in healthy humans. J Nutr 131, 1207-1210.
Sotgia, F., Whitaker-Menezes, D., Martinez-Outschoorn, U.E., Flomenberg, N., Birbe, R.C.,
Witkiewicz, A.K., Howell, A., Philp, N.J., Pestell, R.G., Lisanti, M.P., 2012. Mitochondrial
metabolism in cancer metastasis: visualizing tumor cell mitochondria and the "reverse Warburg
effect" in positive lymph node tissue. Cell Cycle 11, 1445-1454.
Subramanian, A.P., Jaganathan, S.K., Mandal, M., Supriyanto, E., Muhamad, II, 2016. Gallic
acid induced apoptotic events in HCT-15 colon cancer cells. World J Gastroenterol 22, 3952-
3
961.
Trendeleva, T.A., Sukhanova, E.I., Rogov, A.G., Zvyagilskaya, R.A., Seveina, II, Ilyasova,
T.M., Cherepanov, D.A., Skulachev, V.P., 2013. Role of charge screening and delocalization for
lipophilic cation permeability of model and mitochondrial membranes. Mitochondrion 13, 500-
5
06.
Wahdan-Alaswad, R., Fan, Z., Edgerton, S.M., Liu, B., Deng, X.S., Arnadottir, S.S., Richer,
J.K., Anderson, S.M., Thor, A.D., 2013. Glucose promotes breast cancer aggression and reduces
metformin efficacy. Cell Cycle 12, 3759-3769.
Warburg, O., Wind, F., Negelein, E., 1927. The Metabolism of Tumors in the Body. J Gen
Physiol 8, 519-530.
Yang, Y., Karakhanova, S., Hartwig, W., D'Haese, J.G., Philippov, P.P., Werner, J., Bazhin,
A.V., 2016. Mitochondria and Mitochondrial ROS in Cancer: Novel Targets for Anticancer
Therapy. J Cell Physiol 231, 2570-2581.
Yasui, H., Yamamoto, K., Suzuki, M., Sakai, Y., Bo, T., Nagane, M., Nishimura, E., Yamamori,
T., Yamasaki, T., Yamada, K.I., Inanami, O., 2017. Lipophilic triphenylphosphonium derivatives
enhance radiation-induced cell killing via inhibition of mitochondrial energy metabolism in
tumor cells. Cancer Lett 390, 160-167.
21

Journal Pre-proof
Zhang, W., Zhang, S.L., Hu, X., Tam, K.Y., 2015. Targeting Tumor Metabolism for Cancer
Treatment: Is Pyruvate Dehydrogenase Kinases (PDKs) a Viable Anticancer Target? Int J Biol
Sci 11, 1390-1400.
22

Journal Pre-proof
Table legend
+
Table 1. Cytotoxicity and selectivity of benzoate-TPP compounds in the colon cancer cell
lines HCT-15 and COLO 205.
IC50 = concentration necessary to achieve 50% toxicity in the cell culture after 24, 48, or 72 h of
treatment. Values were calculated from the respective sigmoidal dose-response curves,
representing the mean ± SD values of at least three independent experiments. Each assay was
a
performed in triplicate. indicates a significant difference (p<0.0001) between compounds for
CCD 841 CoN; ** or **** indicate a significant difference (p<0.001 or p<0.0001) between
tumor and non-tumor cells for a given compound. p-values were calculated using a two-way
ANOVA and Tukey’s post-hoc test.
N/A: not analyze
+
C
50 values (µM) for benzoates-TPP derivatives and control molecules in colorectal cancer cells
24 h
48 h
72 h
CCD841
CCD841
ompound
HCT-15
10.90± 1.24
13.02 ± 2.51
15.27 ± 1.06
>200
COLO205
5.8 ± 0.64
5.3 ± 0.49
9.54 ± 0.46
>200
HCT-15
COLO205
(non-tumourcell HCT15
line)
COLO205
(non-tumourc
line)
+
A-TPP C10
8.35 ± 0.73**
11.32 ± 1.3****
14.23 ± 1.3**
>200
4.07 ± 0.61**** 27.45 ± 3.73
2.77 ± 0.61**** 114.72 ± 25.1
3.64 ± 0.14**** 38.49 ± 7.52
8.69 ± 0.82*
10.22 ± 0.47**
12.20 ± 0.43
>200
3.24 ± 0.83****
3.01 ± 0.28****
4.01 ± 0.57
>200
17.86 ± 5.94
+
a
a
A-TPP C10
31.02 ± 13.6
+
PP C10
8.24 ± 3.06
N/A
H-C10SA
H-C10GA
>200
N/A
N/A
N/A
>200
46,75 ± 1,58
>400
>200
40,60 ± 3,62
>400
>200
32,85 ± 5,88
>400
N/A
H-C10
>400
>400
>400
N/A
+
0
TPP
N/A
2.24 ± 0.11
N/A
0.68 ± 0.08
N/A
0.63 ± 0.02
0.58 ± 0.10
N/A
23

Journal Pre-proof
24

Journal Pre-proof
Table 2.
+
Selectivity index for benzoates-TPP derivatives at 48 h and 72 h.
Values were calculated from IC50 ratio between non-tumor cell and tumor cell lines from the
respective sigmoidal dose-response curves.
Selectivity Index for benzoates-TPP+ derivatives at 48 h and 72 h
4
8 h
72 h
CCD 841 CoN/
CCD 841 CoN/
CCD 841 CoN/
CCD 841 CoN/
Compound
HCT-15
3.3
COLO 205
6.7
HCT-15
2.1
COLO 205
5.5
+
SA-TPP C10
+
10.1
2.7
41.4
3.0
10.3
GA-TPP C10
+
10.6
0.68
2.1
TPP C10
25

Journal Pre-proof
Declarationof interests
☒
The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.
☐
Theauthors declare the following financialinterests/personal relationships whichmay be
considered as potential competing interests:
26

Journal Pre-proof
Highlights
Benzoic acid esters attached to the triphenylphosphonium group
Evaluating the cytotoxicity and modes of action in colorectal cancer cells
The new compounds were able to exert an uncoupling effect in colorectal cancer cells
+
Benzoate-TPP derivatives are capable of inducing apoptosis
27

Figure 1

Figure 2