Small Molecule Destabilizer of β-Catenin and Ras Proteins Antagonizes Growth of K-Ras Mutation-Driven Colorectal Cancers Resistant to EGFR Inhibitors

Abstract

Background
Oncogenic K-Ras mutations in colorectal cancer (CRC) combined with APC mutations worsen CRC prognosis and lower drug effectiveness. Thus, inhibition of both Wnt/β-catenin and Ras-MAPK signaling may be a rational strategy to improve the treatment of this cancer.

Objective
To identify a novel compound inhibiting both Wnt/β-catenin and Ras-MAPK signaling in CRC.

Methods and Patients
We developed a two-part screening system consisting of analysis of TOP flash reporter cells and then potential toxicity effects on primary neural stem cells (NSCs). We then screened 2000 chemical compounds and tested efficacy of candidates against isogenic colon cancer cells harboring wild-type or mutant K-Ras. We employed immunohistochemistry and immunocytochemistry to determine marker signatures associated with development of disease phenotypes.

Results
We identified CPD0857, a compound that inactivates Wnt/β-catenin signaling and promotes ubiquitin-dependent proteasomal degradation of β-catenin and Ras proteins. CPD0857 effectively decreased proliferation and increased apoptosis of CRC cell lines, and overcame resistance of CRC harboring APC and K-Ras mutations to treatment with an EGFR monoclonal antibody (mAb). Moreover, CPD0857 attenuated invasiveness of highly migratory CRC cells in vitro. Accordingly, xenograft mice treated with CPD0857 showed slower tumor growth and significant decreases in both β-catenin and Ras protein expression.

Conclusions
CPD0857 may be a potential drug for treating aggressive CRC carrying mutations that aberrantly activate Wnt/β-catenin and Ras-ERK pathways.

Full text

1 23
Targeted Oncology
ISSN 1776-2596
Targ Oncol
DOI 10.1007/s11523-020-00755-5
Small Molecule Destabilizer of β-Catenin
and Ras Proteins Antagonizes Growth of K-
Ras Mutation-Driven Colorectal Cancers
Resistant to EGFR Inhibitors
Jung Kyu Choi, Heeyeong Cho &
Byoung-San Moon

1 23
Your article is protected by copyright and
all rights are held exclusively by Springer
Nature Switzerland AG. This e-offprint is
for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.

Vol.:(0123456789)
Targeted Oncology
https://doi.org/10.1007/s11523-020-00755-5
ORIGINAL RESEARCH ARTICLE
Small Molecule Destabilizer ofβ‑Catenin andRas Proteins Antagonizes
Growth ofK‑Ras Mutation‑Driven Colorectal Cancers Resistant toEGFR
Inhibitors
JungKyuChoi1· HeeyeongCho2· Byoung‑SanMoon2,3
© Springer Nature Switzerland AG 2020
Abstract
Background Oncogenic K-Ras mutations in colorectal cancer (CRC) combined with APC mutations worsen CRC prognosis
and lower drug effectiveness. Thus, inhibition of both Wnt/β-catenin and Ras-MAPK signaling may be a rational strategy
to improve the treatment of this cancer.
Objective To identify a novel compound inhibiting both Wnt/β-catenin and Ras-MAPK signaling in CRC.
Methods and Patients We developed a two-part screening system consisting of analysis of TOP flash reporter cells and
then potential toxicity effects on primary neural stem cells (NSCs). We then screened 2000 chemical compounds and tested
efficacy of candidates against isogenic colon cancer cells harboring wild-type or mutant K-Ras. We employed immunohis-
tochemistry and immunocytochemistry to determine marker signatures associated with development of disease phenotypes.
Results We identified CPD0857, a compound that inactivates Wnt/β-catenin signaling and promotes ubiquitin-dependent
proteasomal degradation of β-catenin and Ras proteins. CPD0857 effectively decreased proliferation and increased apop-
tosis of CRC cell lines, and overcame resistance of CRC harboring APC and K-Ras mutations to treatment with an EGFR
monoclonal antibody (mAb). Moreover, CPD0857 attenuated invasiveness of highly migratory CRC cells invitro. Accord-
ingly, xenograft mice treated with CPD0857 showed slower tumor growth and significant decreases in both β-catenin and
Ras protein expression.
Conclusions CPD0857 may be a potential drug for treating aggressive CRC carrying mutations that aberrantly activate
Wnt/β-catenin and Ras-ERK pathways.
Key Points
CPD0857 was identified as a non-toxic inhibitor sup-
pressing both Wnt/β-catenin and Ras-ERK signaling
pathways in colorectal cancer cells (CRCs).
CPD0857 increases the level of Axin protein and reduces
the level of β-catenin and Ras protein via a ubiquitin-
dependent proteasomal degradation mechanism.
CPD0857 overcomes resistance to the anti-EGFR
cetuximab therapy seen in CRCs harboring K-Ras gene
mutations.
CPD0857 decreases cell proliferation and increases
apoptosis via suppressing Wnt/β-catenin, Ras-ERK, and
PI3K/AKT pathways.
Electronic supplementary material The online version of this
article (https ://doi.org/10.1007/s1152 3-020-00755 -5) contains
supplementary material, which is available to authorized users.
* Byoung-San Moon
moonbyoungsan@gmail.com
1 Department ofBiotechnology, College ofLife andApplied
Sciences, Yeungnam University, Gyeongsan38541, Korea
2 Therapeutics andBiotechnology Division, Drug Discovery
Platform Research Center, Korea Research Institute
ofChemical Technology (KRICT), Daejeon34114, Korea
3 Department ofBiotechnology, Chonnam National University,
Yeosu59626, Korea
Author's personal copy

J.K.Choi et al.
1 Introduction
Colorectal cancer (CRC) is the third most common cancer
worldwide, and the second leading cause of cancer death [1].
Development of therapies, including chemotherapy and tar-
geted approaches, has improved median overall survival time
for patients with aggressive and metastatic CRC. Nonetheless,
treatment remains challenging due to drug toxicity and devel-
opment of tumor resistance [2].
Wnt/β-catenin and Ras/extracellular-signal-regulated
kinase (ERK) pathways play a critical role in tumorigenesis
and metastasis of cancers, especially CRC. CRC arises through
frequent genetic abnormalities, such as loss-of-function of the
tumor suppressor adenomatous polyposis coli (APC) in the
Wnt/β-catenin pathway and gain-of-function K-Ras mutations
in the Ras/ERK pathway, both of which function in initiation
and progression of tumorigenesis. Recent studies indicate that
APC and K-Ras mutations are concurrent during different
stages of CRC tumorigenesis and metastasis [2, 3].
Generally, K-Ras functions as a molecular switch to regu-
late ERK and phosphatidyl inositol 3-kinase (PI3K)-Akt sign-
aling [3, 4]. Oncogenic K-Ras mutations lead to hyperplasia
crypt architecture that is not normal, but have low risk of
progression to CRC [5]. Moreover, patients with metastatic
CRC-bearing oncogenic K-Ras mutations are resistant to
EGFR monoclonal antibodies, including cetuximab, a current
standard of care in treating metastatic colorectal cancer [6].
Ras protein stability and signaling are also regulated by the
Wnt/β-catenin pathway [3]. These findings provide a rationale
for identifying drugs that target both Wnt/β-catenin and Ras/
ERK pathways simultaneously.
Here, to define such compounds, we screened a small-
molecule library to identify compounds that reduce levels of
both β-catenin and Ras proteins by inhibiting Wnt/β-catenin
signaling. Our analysis identified CPD0857, which markedly
reduced proliferation and transforming capacity of various
CRC cell lines, most likely by destabilizing β-catenin and Ras
proteins. CPD0857 also overcame cetuximab chemoresist-
ance in CRC cells harboring APC and K-Ras mutations and
promoted tumor cell apoptosis. Moreover, CPD0857 showed
anti-metastatic properties invitro. Finally, CPD0857 signifi-
cantly suppressed tumor growth and progression in mouse
CRC xenograft models. Our study identifies a novel compound
by chemical library screening that may be applicable to vari-
ous cancer types with activated Wnt/β-catenin as well as Ras/
ERK pathways.
2 Materials andMethods
2.1 Experimental Animals andEthics Statement
Animal experimental procedures were approved by the
Institutional Animal Care and Use Committee of the Yon-
sei Laboratory Animal Research Center. Xenograft experi-
ments and tissue processing for immunostaining were
previously described [3]. Briefly, BALB/c nu/nu mice
were purchased from the Central Lab Animal Inc. (Seoul
Korea). All animals were housed in filter-topped shoebox
cages equipped with a computerized environmental con-
trol system. Room temperature was maintained at 24°C
with 40–70% relative humidity. After acclimatization for
1week, mice were subcutaneously injected with 1 × 106
DLD-1-K-Ras mutant (D-K-Ras MT) cells in 200 μL phos-
phate-buffered saline/Matrigel (1:1) in the dorsal flank.
When mean tumor volumes reached ~ 200 mm3, mice were
randomly divided into two groups (four per group) and
administered either CPD0857 suspended in 0.5% methyl
cellulose/0.5% Tween 80 or vehicle intraperitoneally at a
drug dose of 25mg/kg, twice a week. Tumor volume was
measured every 3–4days using Vernier calipers, apply-
ing the formula: π/6 × length × width × height. Animals
were euthanized when tumor volume exceeded 1500 mm3.
Tumors were then excised, weighed, and fixed in 4% para-
formaldehyde for further analysis.
2.2 Dual‑Cell‑Based High‑Throughput Screening
Screening for chemical compounds that destabilize both
β-catenin and Ras by inhibiting Wnt/β-catenin signaling
was previously described [2]. Briefly, from 2000 com-
pounds (Chemdiv chemical library), we initially identi-
fied 100 that efficiently inhibit Wnt/β-catenin signaling
using the HEK293-TOP flash reporter stable cell line. For
primary screening, reporter cells were seeded into 96-well
plates (black polystyrene plate; Greiner Bio-One) at
2 × 104 cells per well and grown for 24h. Each compound
or control (DMSO) along with Wnt3a conditioned media
(CM) was added to wells at 10μM, and luciferase activ-
ity was measured 24h later by FLUOstar OPTIMA. As a
secondary screening, we treated neural stem cells (NSCs)
with 100 compounds selected to check toxicity. NSCs had
been surgically extracted from the forebrain of E14.5 rats
and maintained in an undifferentiated state by culture in
medium [DMEM/F12 with 10ng/mL bFGF (Peprotech)].
Among phenotypes analyzed were cell number and mor-
phology, as assessed by capturing phase contrast images
of cells after 48h. Nine of the original 100 compounds
showed no toxicity.
Author's personal copy

Destabilizer of β-Catenin and Ras Proteins Overcomes Anti-Cancer Drug Resistance
2.3 Cell Culture
CRC cell lines, including HCT15, RKO, SW480, HCT116,
and LoVo, were purchased from the American Type Cul-
ture Collection (ATCC, Manassas, VA, USA). Isogenic cell
lines (DLD-1 (D)-K-Ras WT, D-K-Ras MT & PI3K WT, and
D-K-Ras MT & PI3K MT) were provided by B. Vogelstein
(John Hopkins University School of Medicine, Baltimore,
MD, USA) [7]. HEK293 and HEK293-TOP flash reporter
cells were grown in DMEM (Gibco) containing 10% fetal
bovine serum (FBS), 100 U/mL penicillin, and 10μg/mL
streptomycin at 37 ˚C. Human CRC lines were grown in
RPMI 1640 (Gibco) containing 10% fetal bovine serum
(FBS), 100 U/mL penicillin, and 10μg/mL streptomycin at
37°C. Cells were stored in CRYO-GOLD (Revive Organ-
tech Inc, Irvine, CA, USA), checked based on morphology,
and Mycoplasma-tested using MycoFluor Mycoplasma
Detection kit (Invitrogen) over a period of 6months. In
analyses of protein stability, cultured cells were treated with
MG132 (carbobenzoxy-Leu-Leu-leucinal) (Sigma-Aldrich,
St. Louis, MO, USA).
2.4 Plasmids andConstructs
FLAG-WT-β-catenin-pcDNA3.0 and FLAG-S33Y-β-
catenin-pcDNA3.0 plasmids were kindly provided by Eric
R. Fearon (University of Michigan, Ann Arbor, MI, USA).
Constructs were confirmed by nucleotide sequencing analy-
sis (Cosmogenetech).
2.5 Immunoblotting Assay
Cells or tissues were gently lysed in RadioImmunoPre-
cipitation Assay (RIPA) buffer (Upstate Biotechnology,
Lake Placid, NY, USA) for 1h on ice and centrifuged at
12,000rpm at 4°C for 15min. Lysates were boiled for 5min
at 95°C in SDS sample buffer and separated on 10% SDS-
PAGE. After blocking, membranes were incubated first with
primary antibodies and then with a peroxidase-conjugated
secondary antibody. Bound secondary antibody (anti-mouse
or anti-rabbit 1:10,000) was detected using the enhanced
chemiluminescence (ECL) reagent (Santa Cruz Biotechnol-
ogy). Images were taken using the luminescent image ana-
lyzer LAS-3000 (Fujifirm, Tokyo, Japan).
2.6 Antibodies andReagents
Antibodies used in this study were anti-β-catenin (Santa
Cruz Biotechnology, Santa Cruz, CA, USA), anti-panRas
(Upstate Biotechnology, Lake Placid, NY, USA; Abcam Inc.,
Cambridge, MA, USA), anti-phospho-β-catenin (Ser33/37/
Thr41) (Cell Signaling Biotechnology, Beverly, MA, USA),
anti-Axin (Santa Cruz Biotechnology), anti-β-actin (Abcam
Inc.), anti-Flag (Sigma-Aldrich, St. Louis, MO, USA),
anti-BrdU (Sigma-Aldrich), anti-PCNA (Santa Cruz Bio-
technology), anti-pERK (Thr202/Tyr204) (Cell Signaling
Biotechnology), anti-pAKT (Ser473), and HRP-conjugated
anti-mouse (Bio-Rad Laboratories, Hercules, CA, USA).
Fluorescent anti-rabbit secondary antibodies (Calbiochem,
La Jolla, CA, USA) were used for detection by a luminescent
image analyzer, LAS-3000 (Fujifilm, Tokyo, Japan).
2.7 Immunohistochemistry
andImmunocytochemistry
For immunohistochemistry, xenograft tissues were dissected
and fixed in 4% paraformaldehyde (PFA) at 4°C. Paraffin
sections were incubated with primary antibody at 4°C for
18h. For immunocytochemistry, cells cultured on coverslips
were fixed with 4% PFA/PBS for 2h and immunostained
after permeabilizing with 0.2% Triton X-100. Cells and tis-
sues were then incubated with indicated primary antibod-
ies overnight at 4°C, followed by Alexa Fluor 488 (Life
Technologies, Carlsbad, CA, USA) or Alexa Fluor 555 (Life
Technologies) secondary antibodies at room temperature for
1h and counterstained in 4′-6′diamidino-2-phenylindole
(DAPI) (Boehringer Mannheim, Mannheim, Germany).
Images were visualized using confocal microscopy (LSM5
PASCAL; Zeiss, Jena, Germany). Values obtained from
at least three independent experiments were averaged and
reported as mean ± SD. Student’s two-tailed t test was used
to compare two experimental groups.
2.8 Wound‑Healing Assay
CRC cell lines (D-K-Ras WT, D-K-Ras MT & PI3K WT,
D-K-Ras MT & PI3K MT, SW480, HCT15, LoVo, and
RKO) were seeded at a density of 2.5 × 105 cells in six-well
plates. After cells reached confluence, they were scratched
with a p200 pipet tip and then the media was changed to
DMEM either with or without CPD0857 (at 10, 25, or
50μM). When required, Wnt3a CM was used to stimulate
cells. Scratched cells were then grown in a tissue culture
incubator at 37°C and imaged using an ECLIPSE TE2000-
U fluorescence microscope (Nikon) 12 and 24h later.
2.9 Automated Cell Migration Assay
We used Electrical Cell-substrate Impedance Sensing
(ECIS, Applied BioPhysics) to electronically measure drug
effects on migration of cultured CRC cells in real time. D-K-
Ras MT cells were grown to confluence on polyethylene
eight-well (8W) 10E + arrays (Applied BioPhysics. NY,
USA). After wounding, cells were treated with CPD0857
(25μM) or DMSO and monitored for changes in resist-
ance at 1000Hz using an ECIS Zθ instrument. Data were
Author's personal copy

J.K.Choi et al.
normalized to resistance values at the point of wounding and
subtracted from DMSO-treated resistance values.
2.10 Invasion Assay
D-K-Ras MT (3 × 104) cells were seeded on matrigel-coated
chambers, and either CPD0857 or DMSO was added to
lower chambers. Cells were allowed to invade for 24h.
After clearing cells on the inner surface of the chamber,
cells on the outer surface were fixed using 4% paraformal-
dehyde (PFA) for 15min and stained with crystal violet for
15min. Chambers were dipped in distilled water to remove
excess stain and allowed to dry. Photographs were taken
using ECLIPSE TE2000-U fluorescent microscope (Nikon).
2.11 Statistical Analysis
Statistical analyses were performed using the Excel sta-
tistical tools or Prism 5 (GraphPad Software). Group dif-
ferences were determined with Student’s t test (*P < 0.05,
**P < 0.005, and ***P < 0.0005). One-way ANOVA tests
(Tukey’s multiple comparison test) and two-way ANOVA
tests (Bonferroni post-tests) were used to analyze data from
multiple groups.
3 Results
3.1 Identication ofaCompound thatDecreases
Intracellular β‑Catenin andRas Protein Levels
High expression of both β-catenin and Ras proteins due to
relevant gene mutations is common in aggressive and meta-
static CRC [2, 3]. To screen for novel compounds facili-
tating degradation of both β-catenin and Ras proteins, we
used a dual-cell-based high-throughput screening system
previously described [2]. Of 2000 chemical compounds
from the ChemDiv drug libraries, we selected nine of the
most effective based on lack of toxicity toward neural stem
cells (NSCs) (Fig.1a). To determine whether treatment with
compounds reduced levels of β-catenin and Ras proteins,
we performed western blot analysis of NSCs treated with
candidate compounds (Fig.1b). We observed significantly
reduced β-catenin protein expression in cells treated with
Fig. 1 Screen for drugs that promote β-catenin and Ras protein deg-
radation. a A dual-cell-based high-throughput screening system using
the HEK293 TOP flash line and primary neural stem cells (NSCs)
was used to identify candidate chemical compounds, as described
previously [2]. Among 2000 chemical compounds screened for inhi-
bition of TOP flash reporter activity, 100 showed > 25% inhibition.
Those were subjected to secondary screening using primary NSCs
to assess toxicity. b Immunoblot (IB) analysis of NSC lysates with
indicated antibodies. c, d Relative intensity of bands shown in b, as
quantified using Image J software (n = 3)
Author's personal copy

Destabilizer of β-Catenin and Ras Proteins Overcomes Anti-Cancer Drug Resistance
CPD0022 (40 ± 45%), CPD0857 (37 ± 21%), and CPD0180
(66 ± 7.5%), but not other CPDs (Fig.1c). Among those
three, only CPD0857 (50.8 ± 9.6%) also reduced Ras protein
expression (Fig.1d). Thus, we chose CPD0857 for further
analysis.
3.2 CPD0857 Promotes Ubiquitin‑Dependent
Proteasomal Degradation ofBoth β‑Catenin
andRas Protein
We found that CPD0857 has a half-maximal inhibitory con-
centration (IC50) of 1.4μM in the HEK293 reporter cell line
(Fig.2a, right). Immunocytochemical analysis of HEK293
cells grown in Wnt3a CM confirmed depletion of both cyto-
solic and nuclear β-catenin following CPD0857 treatment
(Fig.2b). Moreover, CPD0857 treatment of HEK293 cells
also promotes dose-dependent decreases of both β-catenin
and Ras proteins expression by Wnt3a CM stimulation
(Fig.2c). Given that β-catenin and Ras are reportedly regu-
lated at the protein level in cancer cells [8, 9], we focused on
β-catenin and Ras protein stability by treating HEK293 cells
with CPD0857 and the proteasome inhibitor MG132 (10μg/
mL) and assessing β-catenin and Ras protein levels. Treat-
ment with CPD0857 alone reduced levels of β-catenin and
Ras protein, an effect significantly rescued in the presence
of MG132 (10μg/mL) (Fig.2d). To confirm these effects,
we transiently overexpressed HA-Ubiquitin (HA-Ub) vector
in the D-K-Ras MT CRC cell line and then treated cells with
CPD0857 plus MG132 (10μg/mL). Immunoprecipitation
analysis of endogenous β-catenin and Ras proteins followed
Fig. 2 CPD0857 effects on β-catenin and Ras protein destabilization
in CRC cells. a (Left) structure of CPD0857; (right) relative reporter
activity in the HEK293 TOP-flash reporter line grown in Wnt3a CM
and with increasing concentrations of CPD0857. b Immunofluores-
cence analysis of 293T cells grown in the presence and/or absence
of Wnt3a CM and CPD0857 using indicated antibodies; nuclei are
counter-stained with DAPI. Scale bars, 20μm (n = 3). c (Left) Immu-
noblot (IB) of HEK293T cells with indicated antibodies; (middle and
right) quantification of respective β-catenin and Ras proteins in the
presence of increasing CPD0857 doses. d Immunoblot (IB) analysis
as in (c) but with some samples treated for 12h with MG132 (10μM)
(n = 3). e Ubiquitylation assay of endogenous β-catenin and Ras pro-
teins in lysates of D-K-Ras MT cells transfected with HA-Ub plas-
mids and treated 1day later with MG132 (10μM) for 6h before har-
vest. IP was performed with β-catenin or Ras antibody. WCLs were
analyzed by IB for indicated antibodies (n = 3)
Author's personal copy

J.K.Choi et al.
by anti-HA immunoblot indicated that both endogenous pro-
teins were polyubiquitylated, and that polyubiquitylated pro-
teins accumulated in MG132-treated cells (Fig.2e).
3.3 CPD0857 Promotes Proteasomal β‑Catenin
andRas Degradation Independently ofGSK3β
To address the role of CPD0857 in CRC cells, we cultured
CRC cell lines with diverse genetic backgrounds includ-
ing APC mutant isogenic D-K-Ras WT and MT cells.
We observed that β-catenin, pan-Ras, and pAKT protein
expressions were significantly reduced upon treatment with
CPD0857 in various CRC cell lines; however, expression of
both β-catenin and pan-Ras proteins except pAKT were not
significant different in HCT116 cell line harboring mutant
β-catenin (Fig.3a). Because Ras protein degradation by
controlling Wnt/β-catenin signaling does not occur in cells
harboring mutant β-catenin [10], we also further assessed
the effect of CPD0857 in NIH3T3 cells expressing wild-
type or mutant β-catenin (S33Y) constructs, respectively.
CPD0857 significantly reduced the expression of both pro-
teins in cells expressing wild-type β-catenin, but this effect
was not observed in cells expressing mutant β-catenin
(S33Y) (Fig.3b, c). To determine whether regulation of
Ras and β-catenin protein stability occurs through the epi-
dermal growth factor receptor (EGFR), which is upstream
of Ras, we tested CPD0857 effects in EGFR knockout
(EGFR−/−) MEFs or in HEK293 cells stimulated with EGF.
Interestingly, CPD0857 significantly reduced the expression
of Ras and β-catenin proteins (Fig.3d, e). Overall, these
findings suggest that CPD0857 treatment decreases Ras
and β-catenin protein levels in CRC cells and likely acts
upstream of β-catenin but downstream of the EGFR.
β-Catenin destruction complex components, such as
Axin, APC, glycogen synthase kinase-3 beta (GSK3β),
and casein kinase-1 (CK1), bind to β-catenin, triggering
its phosphorylation, ubiquitination, and subsequent deg-
radation [2]. Among these components, GSK3β enhances
β-catenin degradation via phosphorylation at S33, S37, and
T41 [2], and also catalyzes Ras protein phosphorylation at
T-144 and T-148 [8, 10–12]. To investigate whether GSK3β
is a CPD0857 target, we used TOP Flash reporter cells in
which GSK3β kinase activity was blocked by treatment
with GSK3β inhibitors such as LiCl, BIO, or valproic acid
(VPA). CPD0857 co-treatment significantly decreased TOP
Flash activity in cells also treated with LiCl, BIO, or VPA
(Fig.3f), results confirmed by dose-dependent decreases
in expression of β-catenin and Ras proteins in CPD0857-
treated GSK3β wild-type or knock-out cells (Fig.3g, h).
These results suggest that CPD0857-mediated β-catenin
and Ras destabilization in CRCs occurs independently of
GSK3β.
3.4 Axin Loss Rescues CPD0857‑Mediated
Polyubiquitination ofβ‑Catenin andRas Protein
inColorectal Cancers (CRCs)
Previously, we identified a compound that destabilizes both
β-catenin and Ras protein by targeting regulators of the
G-protein signaling domain of Axin [2]. Therefore, we asked
whether CPD0857 treatment altered Axin expression. To
do so, we treated RKO, SW480, and LoVo cell lines, which
harbor APC, K-Ras (or B-Raf), and PI3K mutations with
varying CPD0857 doses and monitored endogenous Axin
protein levels using immunofluorescence analysis. Interest-
ingly, CPD0857 treatment increased Axin protein levels,
dose-dependently, and reduced β-catenin, Ras, pERK, and
pAKT protein level in all CRC cell lines tested (Fig.4a–c),
results confirmed by immunoblot analysis (Fig.4d). We
asked whether such increases in Axin protein levels regu-
lated Wnt/β-catenin signaling by knocking down Axin in
293T cells, treating them with Axin siRNA, and then per-
forming immunoblot analysis. CPD0857-dependent Wnt/β-
catenin pathway inhibition was abolished in Axin knock-
down compared to control siRNA-treated cells (Fig.4e).
To determine whether Axin regulates β-catenin and Ras
protein stability, we carried out a ubiquitination assay. We
found that CPD0857-mediated β-catenin and Ras protein
polyubiquitination was abolished in Axin knockdown RKO
cells (Fig.4f), suggesting overall that CPD0857-dependent
increases in Axin protein levels promote β-catenin and Ras
protein degradation in CRC cells.
3.5 CPD0857 Treatment Inhibits Tumor Cell Growth
andOvercomes Resistance toAnti‑EGFR
Treatment inCRC Cells Harboring Mutant K‑Ras
To determine whether CPD0857 blocks CRC growth, we
treated various CRC cell lines with CPD0857 and assessed
proliferation and cellular transformation capacity. Prolifera-
tion of D-K-Ras WT, D-K-Ras MT, SW480, and LoVo cell
lines was efficiently inhibited dose dependently following
CPD0857 treatment (Fig.5a). Moreover, CPD0857 treat-
ment of all CRC cells tested significantly decreased the num-
ber and size of foci (Fig.5b–d).
mAbs targeting EGFR have been used to treat CRC, but
have no effect against cancers harboring K-Ras mutations
[6]. To assess whether CPD0857 treatment could eradicate
CRCs harboring mutant K-Ras, we compared the effect of
CPD0857, the EGFR mAb cetuximab, or co-treatment with
both on proliferation of D-K-Ras WT and D-K-Ras MT, and
SW480 cells, which bear mutant K-Ras (Fig.6a–c). Based on
an MTT assay performed 24, 48, and 72h after drug treat-
ment, cetuximab alone reduced proliferation of D-K-Ras WT
but not of D-K-Ras MT or SW480 cells (Fig.6a). However,
treatment with cetuximab plus CPD0857 efficiently reduced
Author's personal copy

Destabilizer of β-Catenin and Ras Proteins Overcomes Anti-Cancer Drug Resistance
proliferation in all CRC cells tested (Fig.6a). Moreover, cel-
lular-transforming capacity decreased in D-K-Ras WT, D-K-
Ras MT, and SW480 cells following CPD0857 treatment
(Fig.6b, c). We observed no synergistic effects of cetuximab
and CPD0857 in terms of proliferation or cell transforma-
tion, although some synergy was seen in terms of cellular
transforming capacity in D-K-Ras WT cells (Fig.6a–c).
CRC cells harboring K-Ras mutations exhibit more inva-
sive or metastatic capacity than those with WT K-Ras [3].
To test whether CPD0857 antagonized migration of CRC
cells harboring mutant K-Ras, we monitored wound healing
capacities of cells treated with vehicle (DMSO) or varying
Fig. 3 CPD0857 destabilizes β-catenin and Ras proteins independ-
ent of GSK3β. a Immunoblot analysis of indicated CRC cell lines
with indicated antibodies. b Immunoblot of NIH3T3 cells transfected
with WT β-catenin-Flag or mutant β-catenin (S33Y)-Flag and probed
with indicated antibodies. c Relative band intensity in blot shown in
(b). Data are means ± SD (n = 3). d EGFR−/− MEFs were treated
with indicated CPD0857 doses for 24 h. WCLs were immunoblot-
ted using indicated antibodies. Data are means ± SD (n = 3). e D-K-
Ras WT cells were grown in the presence or absence of EGF (20ng/
mL) and immunoblotted (IB) with indicated antibodies. f Relative
band intensity of blot shown in (d). Data are means ± SD (n = 3). g
GSK3β+/+ or GSK3β−/− MEF cells grown in L-cell-conditioned
medium (L-CM) or Wnt3a-CM were treated with indicated amounts
of CPD0857 for 24 h, and WCLs were immunoblotted using indi-
cated antibodies. h Relative band intensity of blot shown in (g). Data
are means ± SD (n = 3)
Author's personal copy

J.K.Choi et al.
CPD0857 doses at the different time points. CPD0857 treat-
ment significantly decreased migratory capacity of D-K-Ras
MT cell line even at low (10μM) concentrations (Fig.6d,
e). The inhibitory effect of cell migration by CPD0857 was
further confirmed by the live cell wound-healing assay.
These effects were confirmed using a live wound-healing
assay of LoVo cells in which CPD0857 treatment signifi-
cantly reduced the normalized impedance value, which indi-
cates resistance of electrodes placed under a plate, a value
that increases as cell migration increases (Fig.6f). Next,
to determine whether CPD0857 inhibited cell invasion, we
performed matrigel invasion assays using D-K-Ras MT cells.
We observed significantly reduced numbers of invading cells
in samples treated with two doses of CPD0857 (Fig.6g, h).
Overall, CPD0857 exhibited an anti-proliferation effect
in cells resistant to EGFR mAb therapy and antagonized
growth and invasiveness of CRC cells harboring mutant
K-Ras.
3.6 CPD0857 Inhibits Tumor Growth ofCRC
Harboring Mutants ofbothAPC andK‑Ras
To further investigate CPD0857 effects on cell proliferation,
we undertook BrdU incorporation analysis and compared
PCNA expression in D-K-Ras WT and MT cells following
treatment with CPD0857 (at 25 or 50μM). We observed
Fig. 4 CPD0857 destabilization of β-catenin and Ras proteins is
Axin-dependent. a (Left) Immunofluorescence analysis of RKO,
SW480, and LoVo CRCs treated for 24h with increasing CPD0857
doses and incubated with Axin (red) antibodies; DAPI serves as
nuclear stain; (right) quantification of Axin levels. b (Left) Immu-
nofluorescence of RKO cells treated 24h with increasing CPD0857
doses and stained with β-catenin and pan-Ras antibodies; (right)
quantification of analysis at left. c Expression of pERK, and pAKT
in the immunostained cells was quantified using Image J software.
All data are shown as means ± SD (n = 3). Significance was deter-
mined by one-way ANOVA (*P < 0.05, **P < 0.005, ***P < 0.005).
d Immunoblot analysis of RKO cells using indicated antibodies. e
Reporter activity of HEK293-TOP flash reporter cells transduced
with Axin or control siRNAs (50 nM) and treated with indicated
CPD0857 doses. f Ubiquitylation assay of β-catenin and Ras pro-
teins in indicated lysates of RKO cells. Cells were transduced with
HA-Ub plasmids and control or Axin siRNAs and then treated a
day later with or without CPD0857 for 24h. Cells were also treated
with MG132 6h before harvest. IP was performed with β-catenin or
Ras antibodies. WCLs were analyzed by IB for indicated antibodies
(n = 3)
Author's personal copy

Destabilizer of β-Catenin and Ras Proteins Overcomes Anti-Cancer Drug Resistance
a decrease in the number of BrdU-positive cells as well a
significant reduction in PCNA expression (Fig.7a, b). Next,
to analyze potential effects of CPD0857 on apoptosis, we
performed apoptosis assay by FACS analysis using various
CRC cell lines. Interestingly, we observed the presence of a
distinct sub-G1 peak (sub-diploid DNA content), suggestive
of apoptotic cells, following CPD0857 treatment in D-K-
Ras MT cells (Supplementary Fig. S1A, see Online Supple-
mentary Material (OSM)). The height of this peak increased
dose dependently with CPD0857 treatment (Supplementary
Fig. S1A (OSM)). These findings were confirmed by per-
forming a DNA fragmentation assay (Supplementary Fig.
S1B (OSM)).
Next, to investigate the invivo effects, we injected
CPD0857 intraperitonially (i.p.) into mice carrying xeno-
graft tumors from the D-K-Ras MT cell line. CPD0857
administration (25mg/kg) significantly reduced the vol-
ume of tumors by 40% (Fig.7c, d). To investigate mecha-
nisms underlying CPD0857 tumor suppressive activity,
we assessed expression of markers of Wnt/β-catenin,
Ras/ERK, and PI3K/AKT signaling in tumors excised
from xenograft mice. Interestingly, CPD0857 treatment
significantly reduced levels of active proteins such as
β-catenin, Ras, p-ERK, and p-AKT as shown by immu-
noblot analysis (Fig.7e). Immunohistochemical analysis
of tumors further revealed decreased levels of β-catenin
Fig. 5 Effects of CPD0857 on CRC proliferation and transform-
ing activity. a Analysis of cell proliferation as measured by an MTT
assay in indicated lines treated 72 h with or without CPD0857. b
Colony-forming assay of indicated D-R-Ras WT and MT lines treated
2 weeks with CPD0857. c, d Colony number (c) and size (d), as
quantified using Image J software. Data is presented as average ± SD
(n = 3). Significance was determined by one-way ANOVA (*P < 0.05,
**P < 0.005, ***P < 0.005)
Author's personal copy

J.K.Choi et al.
(by ~ 44.3%), panRas (by ~ 48.4%), p-ERK (by ~ 64.2%),
and p-AKT (by ~ 36.4%) in CPD0857-treated tumors rela-
tive to vehicle controls (Fig.7f, g). Moreover, we observed
a ~ 20% increase in levels of Axin protein expression,
suggesting that CPD0857 effects may be mediated by an
increase of Axin protein (Fig.7f, g). To further exam-
ine the effect of CPD0857 on proliferation, we counted
PCNA-positive cells in tumors from mice treated with
or without CPD0857. Consistent with invitro results,
the number of PCNA-positive cells in CPD0857-treated
tumors decreased by 41% relative to vehicle-treated mice
(Fig.7g). Taken together, these data suggest that CPD0857
inhibits tumor and that these effects are due to inhibition
of Wnt/β-catenin, Ras/ERK, and AKT signaling (Fig.7h).
4 Discussion
In this study, we assessed how a small molecule that down-
regulates Wnt/β-catenin and Ras/ERK signaling pathways
can suppress CRC tumorigenesis. CPD0857, the novel
chemical compound identified here, reduced levels of both
β-catenin and Ras protein in multiple CRC cell lines via
the ubiquitin-dependent proteasomal degradation pathway
and subsequently suppressed tumor progression and inva-
sive capacity.
We also found that CPD0857 significantly inhib-
ited tumor progression without downregulating either
β-catenin or Ras protein in the HCT116 colon cancer line
Fig. 6 CPD0857 effects on proliferation of Cetuximab-resistant CRCs
harboring K-Ras mutations and on CRC cell migration. a Analysis of
proliferation in indicated lines as determined by MTT assays. Cetuxi-
mab, 5μg/mL; CPD0857, 25 μM (n = 3). b Colony-formation assay
of indicated lines using various drug combinations. c Colony number
(left) and size (right), as quantified using Image J software (n = 3). d
Wound-healing assay of D-K-Ras MT cells in the presence or absence
(DMSO) of CPD0857. e Quantification of analysis shown in (d). f
Automated live wound-healing assay of LoVo cells in the presence or
absence (DMSO) of CPD0857. g, h Matrigel invasion assay of D-K-
Ras MT cells. h Quantification of analysis shown in (g). All data are
presented as average ± SD (n = 3). Significance was determined by
one-way ANOVA (*P < 0.05, **P < 0.005, ***P < 0.005)
Author's personal copy

Destabilizer of β-Catenin and Ras Proteins Overcomes Anti-Cancer Drug Resistance
expressing a non-degradable mutant β-catenin (Figs.3a,
5b–d), suggesting that the effects of CPD0857 in this line
occur by an alternative mechanism. The phosphatidylin-
ositide-3-kinase (PI3K)/protein kinase B (AKT) pathway
is a major signaling cascade downstream of the EGFR in
colon cancers [13] and is known to be important for pro-
gression of many solid cancers [14–19]. The PI3K/AKT
pathway also plays an important role in cell survival by
inactivating apoptogenic factors in many cell types [20].
Interestingly, CPD0857 treatment of CRC cells, including
HCT116, decreased pAKT expression and increased apop-
tosis, based on our observation of increased populations
of early and late apoptotic cells (Fig.3a and Supplemen-
tary Fig. S2A–C (OSM)). This result strongly suggests
that CPD0857 inhibits growth of HCT116 cells by inhib-
iting the PI3K/AKT pathway. To assess this possibility,
we treated CRC cells harboring wild-type or mutant PI3K
with CPD0857 and observed decreased pAKT levels in
all CRC cells bearing wild-type or mutant PI3K (Figs.3a,
4c, d). Treatment also reduced tumor growth, migration
and invasion invitro and invivo (Fig.6d–h). PI3K/AKT
pathway activation is reported in 60–70% of CRCs, and
inhibitors targeting pathway components have been sug-
gested as therapeutic agents in several studies [21–23].
Therefore, CPD0857 could also serve as an inhibitor of
PI3K/AKT signaling to promote cellular apoptosis and
subsequent tumor suppression.
Fig. 7 Effect of CPD0857 on progression of CRC tumors with
mutant K-Ras. a BrdU incorporation assay was performed in D-K-
Ras WT and MT cell treated with varying doses of CPD0857 (n = 3).
b Immunoblot analysis of D-K-Ras WT, MT, HCT15, and SW480
cells with indicated antibodies. c Representative tumors in nude
mice injected in the flank with D-K-Ras MT cells and then treated
for 21 days by IP injection of vehicle or 25mg/kg CPD0857 once
every 3days (n = 6). d Tumor volumes of xenograft mice treated as
in (c) (n = 5). e Immunoblot analysis excised tumors described in
(c) with indicated antibodies. f Fixed sections from excised tumors
described in (c) were incubated with indicated antibodies. g Quanti-
fication of immunostaining of tumor tissues shown in (f), based on
analysis with Image J software. Representative images were selected
from at least three different fields. h Model of CPD0857 inhibition
of Wnt/β-catenin, Ras-MAPK, and PI3K/AKT pathways. CPD0857
favors activity of the β-catenin destruction complex by increasing
Axin protein levels and enhancing β-catenin and Ras degradation via
the proteasome. All data are shown as means ± SD for at least three
independent specimens. Significance was determined by one-way
ANOVA (*P < 0.05, **P < 0.005, ***P < 0.005)
Author's personal copy

J.K.Choi et al.
We found that CPD0857 significantly reduced levels of
both β-catenin and Ras protein in EGFR knock-out cells
(Fig.3d), suggesting that the effects of CPD0857 in CRC
cell lines do not occur via EGFR. Moreover, we previously
revealed that the α-interface of Ras protein can directly inter-
act with the c-term region of β-catenin, and prior degrada-
tion of β-catenin promotes initiation of Ras protein degrada-
tion by exposure of phosphorylation sites on the α-interface
region of the Ras protein [2]. Based on these observations,
we hypothesize that CPD0857 may first reduce β-catenin
protein by inhibiting the Wnt/β-catenin pathway and sub-
sequently decreases Ras, and these could strongly support
recent stepwise model for sequential degradation of both
β-catenin and Ras in CRCs [10].
TGFβ-Smad signaling is often perturbed in human can-
cers, including CRC [24]. Others have suggested that Axin
negatively regulates basal stability of Smads by promoting
their ubiquitination and thereby inhibiting the TGFβ-Smad
pathway [25]. Axin negatively regulates the Wnt/β-catenin
pathway and functions in β-catenin degradation [11]. Axin
has also recently received attention as a small-molecule drug
target for modulating assembly of the destruction complex
for β-catenin degradation [2]. Importantly, we observed
increased Axin protein levels in cells and xenograft tumors
administered CPD0857 (Fig.4a, d). Therefore, CPD0857
likely promotes Axin expression and induces subsequent
inhibition of TGFβ-Smad signaling by destabilizing Smad3
protein, although mechanisms underlying that activity
remain unclear. Nonetheless, our results suggest that Axin
may be a potential target of CPD0857 in CRC, a possibility
that warrants further investigation in future studies.
5 Conclusions
In summary, we used a dual-cell-based high-throughput
screening system to identify CPD0857, which significantly
suppresses tumorigenesis and metastatic properties by inhib-
iting multiple signaling pathways such as Wnt/β-catenin,
Ras/ERK and PI3K/AKT pathways. Moreover, CPD0857
overcame chemoresistance to standard therapeutics such as
EGFR mAbs.
Acknowledgements We thank Dr. Kang-Yell Choi for insightful dis-
cussions (Department of Biotechnology, College of Life Science and
Biotechnology, Yonsei University, Seoul) for technical advice. Byoung-
San Moonand Heeyeong Cho were supported by the Korea Research
Institute of Chemical Technology (KRICT) (SI2031-50)
Declarations
Funding This work was supported by the Korea Research Institute of
Chemical Technology (KRICT) (SI2031-50).
Conflict of interest The authors Jung Kyu Choi, Heeyeong Cho, and
Byoung-San Moon declare that they have no conflicts of interest that
might be relevant to the contents of this manuscript.
Ethics approval Not applicable.
Consent to participate Not applicable.
Consent for publication Not applicable.
Availability of data and material Not applicable.
Code availability Not applicable.
Author contributions JKC and B-SMdesigned and performed the
all experiments. HCsupported materials, edited themanuscript and
performed data analysis. B-SM wrote the manuscript andorganized
the project.
References
1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA-Can-
cer J Clin. 2018;68(1):7–30. https ://doi.org/10.3322/caac.21442 .
2. Cha PH, Cho YH, Lee SK, Lee J, Jeong WJ, Moon BS, et al.
Small-molecule binding of the axin RGS domain promotes beta-
catenin and Ras degradation. Nat Chem Biol. 2016;12(8):593–
600. https ://doi.org/10.1038/nchem bio.2103.
3. Moon BS, Jeong WJ, Park J, Kim TI, Min-do S, Choi KY. Role of
oncogenic K-Ras in cancer stem cell activation by aberrant Wnt/
beta-catenin signaling. J Natl Cancer Inst. 2014;106(2):djt373.
https ://doi.org/10.1093/jnci/djt37 3.
4. Janssen KP, Alberici P, Fsihi H, Gaspar C, Breukel C, Franken
P, etal. APC and oncogenic KRAS are synergistic in enhanc-
ing Wnt signaling in intestinal tumor formation and progression.
Gastroenterol. 2006;131(4):1096–109. https ://doi.org/10.1053/j.
gastr o.2006.08.011.
5. Feng Y, Bommer GT, Zhao J, Green M, Sands E, Zhai Y, etal.
Mutant KRAS promotes hyperplasia and alters differentiation in
the colon epithelium but does not expand the presumptive stem
cell pool. Gastroenterology. 2011;141(3):1003-13e1–1003-13e10.
https ://doi.org/10.1053/j.gastr o.2011.05.007.
6. Yoon J, Koo KH, Choi KY. MEK1/2 inhibitors AS703026 and
AZD6244 may be potential therapies for KRAS mutated colorec-
tal cancer that is resistant to EGFR monoclonal antibody therapy.
Cancer Res. 2011;71(2):445–53. https ://doi.org/10.1158/0008-
5472.CAN-10-3058.
7. Yun J, Rago C, Cheong I, Pagliarini R, Angenendt P, Rajago-
palan H, etal. Glucose deprivation contributes to the develop-
ment of KRAS pathway mutations in tumor cells. Science.
2009;325(5947):1555–9. https ://doi.org/10.1126/scien ce.11742
29.
8. Jeong WJ, Yoon J, Park JC, Lee SH, Kaduwal S, Kim H, etal.
Ras stabilization through aberrant activation of Wnt/beta-
catenin signaling promotes intestinal tumorigenesis. Sci Signal.
2012;5(219):ra30. https ://doi.org/10.1126/scisi gnal.20022 42.
9. Koo KH, Jeong WJ, Cho YH, Park JC, Min do S, Choi KY. K-Ras
stabilization by estrogen via PKCdelta is involved in endometrial
tumorigenesis. Oncotarget. 2015;6(25):21328–40. https ://doi.
org/10.18632 /oncot arget .4049.
10. Lee SK, Jeong WJ, Cho YH, Cha PH, Yoon JS, Ro EJ, etal.
beta-Catenin-RAS interaction serves as a molecular switch for
RAS degradation via GSK3beta. EMBO Rep. 2018. https ://doi.
org/10.15252 /embr.20184 6060.
Author's personal copy

Destabilizer of β-Catenin and Ras Proteins Overcomes Anti-Cancer Drug Resistance
11. Jeon SH, Yoon JY, Park YN, Jeong WJ, Kim S, Jho EH, etal. Axin
inhibits extracellular signal-regulated kinase pathway by Ras deg-
radation via beta-catenin. J Biol Chem. 2007;282(19):14482–92.
https ://doi.org/10.1074/jbc.M6111 29200 .
12. Park KS, Jeon SH, Kim SE, Bahk YY, Holmen SL, Williams BO,
etal. APC inhibits ERK pathway activation and cellular prolifera-
tion induced by RAS. J Cell Sci. 2006;119(Pt 5):819–27. https ://
doi.org/10.1242/jcs.02779 .
13. Efferth T. Signal transduction pathways of the epidermal growth
factor receptor in colorectal cancer and their inhibition by small
molecules. Curr Med Chem. 2012;19(33):5735–44.
14. Dozmorov MG, Azzarello JT, Wren JD, Fung KM, Yang Q, Davis
JS, etal. Elevated AKR1C3 expression promotes prostate cancer
cell survival and prostate cell-mediated endothelial cell tube for-
mation: implications for prostate cancer progression. BMC Can-
cer. 2010;10:672. https ://doi.org/10.1186/1471-2407-10-672.
15. Song L, Xiong H, Li J, Liao W, Wang L, Wu J, etal. Sphin-
gosine kinase-1 enhances resistance to apoptosis through activa-
tion of PI3K/Akt/NF-kappaB pathway in human non-small cell
lung cancer. Clin Cancer Res. 2011;17(7):1839–49. https ://doi.
org/10.1158/1078-0432.CCR-10-0720.
16. Jiao M, Nan KJ. Activation of PI3 kinase/Akt/HIF-1alpha pathway
contributes to hypoxia-induced epithelial-mesenchymal transition
and chemoresistance in hepatocellular carcinoma. Int J Oncol.
2012;40(2):461–8. https ://doi.org/10.3892/ijo.2011.1197.
17. Vredeveld LC, Possik PA, Smit MA, Meissl K, Michaloglou C,
Horlings HM, etal. Abrogation of BRAFV600E-induced senes-
cence by PI3K pathway activation contributes to melanomagen-
esis. Genes Dev. 2012;26(10):1055–69. https ://doi.org/10.1101/
gad.18725 2.112.
18. Wang D, Chen J, Chen H, Duan Z, Xu Q, Wei M, etal. Lep-
tin regulates proliferation and apoptosis of colorectal
carcinoma through PI3K/Akt/mTOR signalling pathway. J Biosci.
2012;37(1):91–101.
19. Christgen M, Noskowicz M, Schipper E, Christgen H, Heil C,
Krech T, etal. Oncogenic PIK3CA mutations in lobular breast
cancer progression. Gene Chromosome Canc. 2013;52(1):69–80.
https ://doi.org/10.1002/gcc.22007 .
20. Zhuang Z, Zhao X, Wu Y, Huang R, Zhu L, Zhang Y, etal.
The anti-apoptotic effect of PI3K-Akt signaling pathway
after subarachnoid hemorrhage in rats. Ann Clin Lab Sci.
2011;41(4):364–72.
21. Malinowsky K, Nitsche U, Janssen KP, Bader FG, Spath C,
Drecoll E, etal. Activation of the PI3K/AKT pathway cor-
relates with prognosis in stage II colon cancer. Br J Cancer.
2014;110(8):2081–9. https ://doi.org/10.1038/bjc.2014.100.
22. Bahrami A, Khazaei M, Hasanzadeh M, ShahidSales S, Joudi
Mashhad M, Farazestanian M, etal. Therapeutic potential of
targeting PI3K/AKT pathway in treatment of colorectal cancer:
rational and progress. J Cell Biochem. 2018;119(3):2460–9. https
://doi.org/10.1002/jcb.25950 .
23. Zhang J, Roberts TM, Shivdasani RA. Targeting PI3K signaling
as a therapeutic approach for colorectal cancer. Gastroenterol.
2011;141(1):50–61. https ://doi.org/10.1053/j.gastr o.2011.05.010.
24. Jiang Z, Cao Q, Dai G, Wang J, Liu C, Lv L, etal. Celastrol
inhibits colorectal cancer through TGF-beta1/Smad signaling.
OncoTargets Ther. 2019;12:509–18. https ://doi.org/10.2147/OTT.
S1878 17.
25. Guo X, Ramirez A, Waddell DS, Li Z, Liu X, Wang XF. Axin and
GSK3- control Smad3 protein stability and modulate TGF- sign-
aling. Genes Dev. 2008;22(1):106–20. https ://doi.org/10.1101/
gad.15909 08.
Author's personal copy