Synthetic cajanin stilbene acid derivatives inhibit c‑MYC
in breast cancer cells
Onat Kadioglu · Yujie Fu · Benjamin Wiench ·
Yuangang Zu · Thomas Efferth
Received: 10 September 2014 / Accepted: 12 February 2015
© Springer-Verlag Berlin Heidelberg 2015
compounds showed cytotoxicities in the micromolar range.
Microarray analyses pointed to cell cycle, DNA damage,
and DNA repair as mainly affected cellular functions. Pro￾moter motif analysis of the deregulated genes further sup￾ported the microarray gene expression analysis results
emphasizing the relevance of transcription factors regulating
cell cycle and proliferation, with MYC as being the most
pronounced one. Luciferase-based reporter cell line experi￾ments and molecular docking studies yielded supportive
results emphasizing the inhibitory activity of CSA and its
derivatives on MYC. CSA and its derivatives are shown to
be promising anticancer compounds with low toxicity. They
inhibit MYC activity comparable to 10058-F4 and 10074-
G5. Further studies are warranted to analyze the therapeutic
applicability of these compounds in more detail.
Keywords Breast cancer · Cajanin stilbene acid ·
c-MYC · Microarray · Pharmacogenomics
Introduction
Despite tremendous advances in diagnosis and therapy,
breast cancer is still the leading cause of cancer-related
death in women worldwide (Kim et al. 2013). A deeper
understanding of the molecular mechanisms involved in
breast cancer progression and identification of new active
compounds should be helpful in developing more effective
treatments for breast cancer. Numerous investigations on
the biology of breast cancer during the past decades dem￾onstrated the relevance of several oncogenes and tumor
suppressor genes (e.g., BRCA 1/2, HER2, MYC, RAS,
TP53, and others) for carcinogenesis of breast cancer and
survival prognosis of patients (Addou-Klouche et al. 2012;
Bhardwaj et al. 2014; Byler et al. 2014; Parris et al. 2014).
Abstract In the present study, we investigated the activity
and modes of action of cajanin stilbene acid (CSA) and its
derivatives in terms of cytotoxicity, gene expression profile,
and transcription factor activity. XTT assays on MCF7 cells
were performed upon treatment with CSA or derivatives.
After the determination of IC50 values, gene expression pro￾filing was performed with Agilent microarray experiments.
Deregulated genes were determined with Chipster soft￾ware, pathway and functional analyses were performed with
Ingenuity pathway software. In order to identify the poten￾tial upstream regulators, MatInspector software was used
to perform transcription factor binding motif search in the
promoter regions of the deregulated genes. Molecular dock￾ing on MYC/MAX complex and reporter cell line experi￾ments were performed to validate the MYC inhibitory activ￾ity of CSA and its derivatives. Two known MYC inhibitors:
10058-F4 and 10074-G5 were used as positive control. All
Onat Kadioglu and Yujie Fu have contributed equally to this
work.
Electronic supplementary material The online version of this
article contains supplementary
material, which is available to authorized users.
O. Kadioglu · B. Wiench · T. Efferth (*)
Department of Pharmaceutical Biology, Institute of Pharmacy
and Biochemistry, Johannes Gutenberg University, Staudinger
Weg 5, 55128 Mainz, Germany
e-mail: [email protected]
Y. Fu · Y. Zu
Key Laboratory of Forest Plant Ecology, Ministry of Education,
Northeast Forestry University, Harbin 150040, China
Y. Fu · Y. Zu
Engineering Research Center of Forest Bio-Preparation, Ministry
of Education, Northeast Forestry University,
Harbin 150040, China
Arch Toxicol
MYC genes encode a family of basic helix-loop￾helix leucine zipper (bHLH-LZ) proteins. MYC proteins
(c-MYC, N-MYC, and L-MYC) act as transcription fac￾tors and form heterodimers with another bHLH-LZ pro￾tein, MAX, which is ubiquitously expressed (Kim et al.
2008; Prendergast and Ziff 1992). MYC–MAX complexes
activate transcription of various genes carrying a spe￾cific E-box hexamer, CACGTG or CACGCG sequences
in their promoter sequences (Blackwell et al. 1990; Kim
et al. 2008). The c-MYC oncogene was discovered in the
late 1970s as the cellular homolog of the retroviral v-MYC
oncogene (Vennstrom et al. 1982), and its role in tumori￾genesis in vivo was confirmed in transgenic mice studies
(Adams et al. 1985). c-MYC has been connected with vari￾ous cancer types, including breast and colon cancer, neu￾roblastoma, osteosarcoma, and melanoma (Chen and Olo￾pade 2008; Pelengaris et al. 2002a). It is overexpressed in a
wide variety of human tumors including one-third of breast
cancers (Escot et al. 1986; Guerin et al. 1988) and activated
through several possible mechanisms (Schuldiner and Ben￾venisty 2001). It is a DNA-binding, nuclear transcription
factor involved in the regulation of cell cycle, programmed
cell death, and tumorigenesis (Amundadottir et al. 1995;
Harrington et al. 1994). MCF-7 acquired resistance to all￾trans-retinoic acid plus interferon-α upon ectopically over￾expression of c-MYC (Shang et al. 1998). Elevation of c￾MYC levels is a fundamental mechanism of anti-estrogen
resistance in human breast cancer as exemplified by the
observation that c-MYC expression alone conferred resist￾ance to the growth inhibitory effects of the estrogen antago￾nist, ICI 182,780 (Venditti et al. 2002). Inducible expres￾sion of c-MYC in antiestrogen-arrested cells induced effects
similar to estrogen. This approach has been used as a strat￾egy to identify estrogen-regulated genes that are also targets
of c-MYC (Musgrove et al. 2008). Moreover, c-MYC- and
ERα-binding sites co-localize near the transcription start
site of a subset of estrogen-responsive genes, i.e., APLP1,
CR2, DLX1, GABPB2, HK2, LACTB2, LOC153364, MEA,
RBBP8, RCC2, SAMHD1, SYVN1, and VDP. This suggests
that ERα and c-MYC may physically interact to stabilize
the ERα-coactivator complex permitting other signal trans￾duction pathways to fine-tune estrogen-mediated signal￾ing networks (Cheng et al. 2006). Interestingly, almost all
estrogen-regulated genes with roles in cell growth are also
c-MYC targets, and c-MYC plays role in the activation
of ribosome biogenesis and protein synthesis by estrogen
as predicted by pathway analysis (Musgrove et al. 2008).
Patients with ER-positive tumors do not show sustained
responses to therapy targeting estrogen receptor pathway,
which is probably due to the acquisition of endocrine resist￾ance (Ali and Coombes 2002; Clarke et al. 2003; Nicholson
et al. 2005). Overexpression of the estrogen-targeted cell
cycle regulatory molecules c-MYC and cyclin D1 has been
linked with sensitivity to endocrine therapy (Butt et al.
2005). Estrogen induction of c-MYC in the rodent uterus
is not prevented by progesterone, which inhibits DNA syn￾thesis but not cell growth (Kirkland et al. 1992). This is in
concordance with the concept that estrogen regulates cell
growth by c-MYC. Decreasing the c-MYC protein level in
MCF-7 cells by RNAi significantly inhibited tumor growth
both in vitro and in vivo, implying the therapeutic potential
of RNAi for breast cancer treatment by targeting overex￾pressed oncogenes such as c-MYC (Wang et al. 2005).
Taking into mind the eminent role of MYC for cancer
development and progression (Yap et al. 2013), this onco￾gene represents an interesting target to expand the arsenal
of anticancer therapeutics. The induction of oncogenic
MYC leads to full-blown malignancies (Arnold and Watt
2001; D’Cruz et al. 2001; Felsher and Bishop 1999; Pelen￾garis et al. 2002b; Shachaf et al. 2004), while blocking
MYC activation in most cases results in tumor regression
(Jonkers and Berns 2004). Currently, there is no MYC￾targeting drug available in the clinic, and effective MYC
inhibitors may be valuable new weapons in the armory to
fight cancer (Ponzielli et al. 2005).
Cajanin stilbene acid (CSA), isolated from Pigeon Pea
(Cajanus cajan), revealed antioxidant and antimicrobial
activities (Wu et al. 2011; Zu et al. 2010) as well as cyto￾toxicity toward cancer cells (Eichhorn 2011). In the pre￾sent study, the effect of three synthetic derivatives were
compared with CSA as lead compound and investigated
in terms of cytotoxicity, microarray-based gene expression
profiling, and transcription factor activity in MCF-7 breast
cancer cells. Together with in silico-based transcription fac￾tor binding motif searches in promoter sequences of differ￾entially expressed genes upon treatment with these com￾pounds as well as in silico molecular docking of the four
cajanin stilbene acids to MYC, our results revealed inhibi￾tion of MYC as a mechanism of cytotoxicity of this class of
compounds toward breast cancer cells.
Methods
Chemicals
CSA, CSA6, CSA9, and CSA19 were synthesized by one
of the authors (Y.J.F., Key Laboratory of Forest Plant Ecol￾ogy, Ministry of Education, Northeast Forestry Univer￾sity, Harbin, China). Two known MYC inhibitors 10074-
G5 and 10058-F4 (Sigma-Aldrich, Germany) are used as
positive control (Clausen et al. 2010; Yap et al. 2013). The
chemical structures are depicted in Fig. 1. The compounds
were dissolved in DMSO (Sigma-Aldrich, Germany) at a
concentration of 100 mM. 2,3-bis-(2-methoxy-4-nitro-
5-sulfophenyl)-2H-tetrazolium-5-carboxanilide inner salt
Arch Toxicol
(XTT, MP Biomedicals, Germany) was dissolved in ddH2O
at a concentration of 5 mM. N-methyl dibenzopyrazine
methyl sulfate (PMS, Sigma-Aldrich, Germany) was dis￾solved in PBS at a concentration of 5 mM.
Cell culture
The MCF-7 cell line was obtained from the German Cancer
Research Center (DKFZ, Heidelberg, Germany). The origi￾nal source of the cell line is the American Type Culture
Collection (ATCC, USA). MCF-7 cells were cultured in
complete DMEM medium with GlutaMAX™ (Invitrogen,
Germany) supplemented with 10 % FBS and 1 % penicillin
(100 U/mL)-streptomycin (100 μg/mL). The MCF-7 cell
line was authenticated using Multiplex Cell Authentication
(MCA) based on single nucleotide polymorphism (SNP)
profiling by Multiplexion GmbH (Heidelberg, Germany) as
described recently (Castro et al. 2013).
The HEK293 cells were cultured in Opti-MEM medium
(Invitrogen, Germany) supplemented with 5 % FBS and
1 % penicillin (100 U/mL)-streptomycin (100 μg/mL).
Cells were maintained in a humidified environment at
37 °C with 5 % CO2 and subcultured twice per week. All
experiments were performed on cells in the logarithmic
growth phase.
XTT proliferation assay
The XTT proliferation assay was performed to assess
cytotoxicity of CSA and its derivatives toward MCF-7.
The assay was also performed for CSA, its derivatives,
10058-F4 and 10074-G5 in HEK293 cells in order to deter￾mine the concentration to be used for the reporter cell line
experiment. The assay is based on the cleavage of the yel￾low tetrazolium salt XTT in the presence of the electron￾coupling reagent PMS, producing an orange formazan dye.
This conversion only occurs in viable cells, and the amount
of formazan formed directly correlates to the number of liv￾ing cells. A spectrophotometric quantification of the color
formation can be performed using a microplate reader.
Briefly, adherent MCF-7 cells were detached by treatment
with 0.25 % trypsin/EDTA (Invitrogen, Germany), and
an aliquot of 1 × 104
cells was placed in each well of a
96-well cell culture plate (Thermo Scientific, Germany)
in a total volume of 100 µL. Cells were allowed to attach
overnight and then were treated with different concentra￾tions of CSA or its derivatives. After 72 h, 25 µL PMS
(5 mM) were added to 5 mL XTT sodium salt solution
(1.5 mM) to obtain a XTT-PMS staining solution. 50 µL of
this staining solution was added to each well, and the plates
were incubated at 37 °C for 4 h. Absorbance was measured
on an Infinite M2000 Pro™ plate reader (Tecan, Germany)
at 490 nm (reference wavelength: 655 nm). Each assay was
done at least two times, with six replicate each. The via￾bility was evaluated based on a comparison with untreated
cells. IC50 values represent the concentrations required to
inhibit 50 % of cell proliferation and were calculated from
a calibration curve by linear regression using Microsoft
Excel.
Gene expression profiling
Total RNA from MCF-7 cells after 72 h of treatment with
CSA or one of its derivatives at IC50 concentration or
DMSO solvent control was isolated using the RNeasy Kit
from Qiagen (Hilden, Germany) according to the manufac￾ture’s instruction. The quality of total RNA was checked
by gel analysis using the total RNA Nano chip assay on an
Agilent 2100 Bioanalyzer (Agilent Technologies GmbH,
Germany). Microarrays were performed at the Institute of
Molecular Biology (Mainz, Germany). Gene expression
profiling was performed with Whole Human Genome RNA
chips (8 × 60 K Agilent) for CSA and CSA derivatives sep￾arately with two independent experiments for each. Twelve
chips were used in total: control (x2), CSA (x2), control
(x2), CSA6 (x2), CSA9 (x2), CSA19 (x2). Probe labeling
and hybridization were carried out following the One-Color
Microarray-Based Gene Expression Analysis Protocol (Agi￾lent Technologies GmbH). Briefly, total RNA was labeled
and converted to cDNA. Then fluorescent cRNA (Cyanine
3-CTP) was synthesized and purified using QIAgen RNeasy
Kit. After fragmentation of the cRNA, samples were hybrid￾ized for 17 h at 65 °C. Microarray slides were washed and
10058-F4 10074-G5
CSA CSA6
CSA9 CSA19
Fig. 1 Molecular structure of CSA, its derivatives, 10058-F4 and
10074-G5
Arch Toxicol
scanned with Agilent Microarray Scanning system. Images
were analyzed and data were extracted, background sub￾tracted and normalized using the standard procedures of
Agilent Feature Extraction Software. Normalization was
performed according to the 75th percentile signal value by
using the Agilent one-color microarray signals for inter￾array comparisons as the default normalization scheme.
The expression data obtained were filtered with Chipster
data analysis platform (CSC, Espoo, Finland) (Kallio et al.
2011). These steps include filtering of genes by two (CSA,
CSA6, CSA9) to three (CSA19) times standard deviation
of deregulated genes and a subsequent assessment of sig￾nificance using empirical Bayes t-test coupled with Benja￾mini Hochberg FDR adjustment (p < 0.05). Filtered genes
were fed into Ingenuity Pathway Analysis software (IPA) to
determine cellular networks and functions affected by drug
treatment (Ingenuity Systems, Redwood City, CA, USA).
The microarray data were uploaded to GEO (gene expres￾sion omnibus) database with the ID GSE64430.
Motif search
Motif search was performed for the 2-kb upstream region
of the deregulated genes using the MatInspector software
(Cartharius et al. 2005). The software is based on a weight
matrix pattern definition representing the complete nucleo￾tide distribution for each single position, and thus it is supe￾rior to a simple IUPAC consensus sequence representa￾tion. It allows the quantification of the similarity between
the weight matrix and a potential TFBS detected in the
sequence (Cartharius et al. 2005). There are various studies
relying on MatInspector software to identify potential tran￾scription factor binding sites (Plourde et al. 2013; Yan et al.
2012; Ziv-Av et al. 2011). The MatInspector library contains
185 motif family and 757 individual matrixes for human
transcription factors. Each individual matrix is assigned to
a family consisting of matrices that represent similar DNA
patterns. Deregulated genes were screened for all the motifs
in the MatInspector library. A threshold of 85 % was applied
for percentage relative score, which represents how well
the certain region on the promoter sequence fits the model
built for the motifs (Pavesi and Pesole 2006). The number
of occurrences of each motif family in each gene was found
out, and the average of occurrences was calculated for each
motif family. Percentage of genes carrying the motif family
recognition site was multiplied with the average motif fam￾ily occurrences to calculate the weighted motif score to rank
the motif families.
Reporter cell line experiment
HEK293 cell lines were transfected with c-MYC-luciferase
reporter construct (CCS-012L, Qiagen, Germantown, MD,
USA). The cells were cultured, according to the manufac￾turer’s recommendations. The cells were treated with vary￾ing concentrations of CSA derivatives for 18 h. c-MYC pro￾moter activity was quantified with Dual-Luciferase Reporter
Assay System (Promega, Madison, WI, USA) by measuring
the firefly and renilla luciferase luminescences on an Infi￾nite M2000 Pro™ plate reader (Tecan, Germany). The ratio
of firefly luciferase intensity to renilla luciferase intensity
yielded a measure for c-MYC activity. The relative luciferase
for each sample was calculated as: 1000 × firefly luciferase
luminescence/renilla luciferase luminescence. DMSO treat￾ment served as control. Normalized c-MYC activity was cal￾culated by the formula: relative luciferase of sample/relative
luciferase of the DMSO control. The MYC inhibitors 10058-
F4 and 10074-G5 were used as positive controls.
Molecular docking
Calculations were performed with AutoDock 4.2 algo￾rithm, an automated and robust procedure based on the
Lamarckian genetic algorithm (GA) for virtual screening of
ligand binding to target protein (Sousa et al. 2006). Bind￾ing mode and affinity of CSA and its derivatives on MYC/
MAX complex (PDB ID: 1NKP) was investigated by blind
and defined molecular docking. Two known MYC inhibi￾tors, 10074-G5 and 10058-F4, were used as control drugs
(Clausen et al. 2010; Yap et al. 2013). They observed to
interact with MYC with Kd values of 4.4 µM and 2.5 µM,
respectively (Follis et al. 2008; Wang et al. 2007). The
bHLH and leucine zipper regions of the proteins are cov￾ered in the PDB file. Blind docking was performed with
25,000.000 energy evaluations and 250 runs, whereas the
defined docking was performed with 2,500.000 energy
evaluations and 250 runs as triplicates. The average of the
lowest binding energy and mean binding energy of three
runs for blind and defined docking were taken into account.
DNA-binding domain on MYC and MAX were the regions of
interest, and the covered residues for defined docking cal￾culations were Lys902, Arg903, His906, Asn907, Glu910,
Arg911, Arg913, Arg914, Lys936, Lys939, Ile942, Leu943,
Ala946, Tyr949, Ile950, Val953, Leu960, Leu967, and
Leu974 on MYC and Lys203, Arg204, His207, Asn208,
Glu211, Arg212, Arg214, Arg215, Lys236, Lys239, Ile242,
Leu243, Ala246, Tyr249, Ile250, Met253, His260, Leu267,
Leu274, and Leu281 on MAX (Nair and Burley 2003).
Results
Cytotoxicity against MCF-7 cells
XTT assays were performed to evaluate the cytotoxic
effect of CSA and three synthetic CSA derivatives against
Arch Toxicol
MCF-7 cells (Fig. 2). CSA6 and CSA19 showed a stronger
effect against MCF-7 cells than the lead compound CSA.
CSA9 was less toxic than CSA. XTT assays were also
performed in HEK293 cells. IC50 values of CSA and its
derivatives are shown in Table 1. In addition, 10058-F4
revealed an IC50 value of 27.97 ± 2.21 µM and 10074-G5
of 11.29 ± 0.98 µM in HEK293 cells.
Gene expression profiling
CSA treatment
Microarray-based gene expression analyses were per￾formed to identify possible targets and mechanisms of
CSA in MCF-7 breast cancer cells. MCF-7 cells were
treated with 54.8 µM (IC50) CSA or DMSO solvent control
for 72 h, before total RNA was isolated for whole human
genome mRNA gene expression profiling. Bioinformatic
analyses identified 659 genes significantly deregulated
(p < 0.05) after CSA treatment. The most deregulated genes
following CSA treatment are summarized in Supplemen￾tary Table 1. Using the Ingenuity Pathway Analysis tool,
the deregulated genes were correlated with cellular func￾tions including cell cycle, cellular assembly and organiza￾tion, and DNA replication, recombination and repair (Sup￾plementary Fig. 1a). Furthermore, several genetic networks
were found to be significantly deregulated in MCF-7 cells
after CSA treatment and the most deregulated network is
shown in Supplementary Fig. 1b. The genetic network was
correlated with discrete cellular functions such as cellular
growth, proliferation, and movement. Upstream regulator
analysis was performed to identify transcription regula￾tors that may be responsible for gene expression changes
observed in MCF-7 cells upon CSA treatment. TP53 was
observed to be the most significant activated transcription
factor with a p value of 1.85 × 10−26 influencing various
deregulated genes. ERBB2 was found to be activated with a
p value of 4.17 × 10−22 (Supplementary Table 5).
CSA6 treatment
A gene expression analysis was performed to identify pos￾sible targets and mechanisms of CSA6 in MCF-7 breast
cancer cells. MCF-7 cells were treated with 3.0 µM (IC50)
CSA6 or DMSO solvent control for 72 h before total RNA
was isolated for a whole human genome mRNA gene
expression microarray. Bioinformatic analyses identified 30
genes significantly deregulated (p < 0.05) after CSA6 treat￾ment. The most deregulated genes in consequence to CSA6
treatment are summarized in Supplementary Table 2. Using
the Ingenuity Pathway Analysis tool, the deregulated genes
were correlated with cellular functions including cell cycle,
DNA replication, DNA repair, and cellular movement
(Supplementary Fig. 2a). Furthermore, several genetic net￾works were significantly deregulated in MCF-7 cells after
CSA treatment. The most deregulated network is shown in
Supplementary Fig. 2b. The genetic network was correlated
to discrete cellular functions such as cellular movement.
Finally, an upstream regulator analysis was performed to
identify transcription regulators that may be responsible
for gene expression changes observed in MCF-7 cells after
CSA6 treatment. The androgen receptor was predicted to
be inhibited upon CSA6 treatment with a p value of 0.014
(Supplementary Table 5).
CSA9 treatment
A gene expression analysis was performed to identify pos￾sible targets and mechanisms of CSA9 in MCF-7 breast
cancer cells. MCF-7 cells were treated with 65.3 µM (IC50)
CSA9 or DMSO solvent control for 72 h before total RNA
was isolated for a whole human genome mRNA gene
expression microarray. Bioinformatic analyses identified
Fig. 2 Dose–response curves of MCF-7 cells after treatment with
different concentrations of CSA or CSA derivatives. XTT assays
were performed to determine dose–response curves. Viability of
MCF-7 cells are represented by mean ± SEM of two independent
experiments, and it is expressed as percentage survival of control
Table 1 IC50 values (mean ± SEM) of CSA and its derivatives for
MCF-7 and HEK293 cells
Cells were treated for 72 h, and cell viability was analyzed by XTT
assay (two independent experiments with six parallel measurements)
Substance MCF7 HEK293
CSA 54.77 ± 2.83 22.76 ± 8.54
CSA6 2.96 ± 0.12 2.45 ± 0.28
CSA9 65.32 ± 0.89 6.29 ± 2.11
CSA19 22.6 ± 0.03 1.67 ± 0.06
Arch Toxicol
589 genes significantly deregulated (p < 0.05) after CSA9
treatment. The most deregulated genes in consequence to
CSA9 treatment are summarized in Supplementary Table 3.
Using the Ingenuity Pathway Analysis tool, the deregulated
genes were correlated with cellular functions including cel￾lular movement and cell-to-cell signaling (Supplementary
Fig. 3a). Furthermore, several genetic networks were sig￾nificantly deregulated in MCF-7 cells after CSA9 treatment
and the most deregulated network is shown in Supplemen￾tary Fig. 3b. The genetic network was correlated to discrete
cellular functions such as protein synthesis and cellular
movement. An upstream regulator analysis was performed
to identify transcription regulators that may be responsible
for gene expression changes in MCF-7 cells after CSA9
treatment. Thereby, the histone-lysine N-methyltransferase
MLL2 (MLL2) with a p value of 1.61 × 10−6
and the
cyclic AMP-dependent transcription factor ATF-4 (ATF4)
with a p value of 5.97 × 10−6
were predicted to be acti￾vated after treatment (Supplementary Table 5).
CSA19 treatment
A gene expression analysis was performed to identify pos￾sible targets and mechanisms of CSA19 in MCF-7 breast
cancer cells. MCF-7 cells were treated with 22.6 µM (IC50)
CSA19 or DMSO solvent control for 72 h before total
RNA was isolated for a whole human genome mRNA
gene expression microarray. Bioinformatic analyses iden￾tified 142 genes significantly deregulated (p < 0.05) after
CSA19 treatment. The most deregulated genes in conse￾quence to CSA19 treatment are summarized in Supplemen￾tary Table 4. Using the Ingenuity Pathway Analysis tool,
the deregulated genes were correlated with cellular func￾tions including amino acid and carbohydrate metabolism,
cellular transport, and cellular movement (Supplementary
Fig. 4a). Furthermore, several genetic networks were sig￾nificantly deregulated in MCF-7 cells after CSA19 treat￾ment and the most deregulated network is shown in Sup￾plementary Fig. 4b. The genetic network was correlated to
discrete cellular functions such as cell death. An upstream
regulator analysis did not identify any transcription regula￾tors that may be activated or inhibited in MCF-7 cells after
CSA19 treatment.
Comparison of microarray results
To identify common mechanisms induced by CSA and its
derivatives, a comparison analysis of the microarray results
was performed using the Ingenuity Pathway Analysis tool.
Table 2 summarizes all genes consistently deregulated by
at least three CSA compounds. The TGM2 gene encoding
transglutaminase 2 was the only gene consistently deregu￾lated after treatment with all four CSA compounds. Top cel￾lular functions affected by all four compounds are shown in
Fig. 3a. All compounds revealed strong effects on cell cycle
progression and DNA damage response. Interestingly, cellular
movement was among the top five affected cellular functions
after treatment with the four derivatives. Upstream regulator
Table 2 Genes consistently deregulated by at least three different CSA derivatives (Whole Human Genome RNA chips 8 × 60 K, Agilent One￾Color Microarray-Based Gene Expression Analysis Platform)
Symbol Gene name CSA CSA6 CSA9 CSA19
ARHGAP11A Rho GTPase-activating protein 11A −5.3 −1.3 −1.9
C15orf52 Chromosome 15 open reading frame 52 1.4 3.2 1.8
CTTNBP2 Cortactin-binding protein 2 −3.6 −2.2 −1.5
EBLN2 Endogenous bornavirus-like nucleoprotein 2 1.3 2.0 2.3
GPA33 Glycoprotein A33 (transmembrane) −1.4 −5.1 −2.7
HERPUD1 Homocysteine-inducible, endoplasmic reticulum stress inducible,
ubiquitin-like domain member 1
4.1 1.3 2.4
JUN Jun proto-oncogene 4.0 2.5 1.5
MALAT1 Metastasis associated lung adenocarcinoma transcript 1 (nonprotein coding) 1.5 2.2 3.1
PBX3 Pre-B-cell leukemia homeobox 3 −1.3 −2.5 −1.4
RAD51AP1 RAD51 associated protein 1 −10.1 −1.4 −2.3
RAMP1 Receptor (G protein-coupled) activity-modifying protein 1 −6.8 −1.9 −1.5
S100P S100 calcium-binding protein P 4.2 10.2 1.5
SCARA3 Scavenger receptor class A, member 3 −3.8 −1.4 −1.9
SPON2 Spondin 2, extracellular matrix protein 1.3 4.8 1.5
SPTAN1 Spectrin, alpha, non-erythrocytic 1 1.4 1.8 1.5
SRGAP2 SLIT-ROBO Rho GTPase-activating protein 2 1.3 1.8 1.9
TGM2 Transglutaminase 2 6.3 1.3 4.2 1.4
Arch Toxicol
CSA CSA6 CSA9 CSA19 Cell Cycle Cellular Assembly and Organization DNA Replication, Recombination and Repair Cellular Growth and Proliferation
Cellular Movement
Cellular Development
Cell Death and Survival
Cellular Function and
Maintenance
Cellular Compromise
Lipid Metabolis
Fig. 3 a Top cellular functions significantly affected by CSA and CSA
derivative treatment in MCF-7 cells. b Deregulated genes under the
influence of TP53 and TGFB1 in MCF-7 cells treated with CSA deriva￾tives TP53 and TGFB1 were predicted to be activated after CSA deriv￾ative treatment. Red colored genes were up-regulated, green colored
ones were down-regulated after treatment. The arrows indicate effects
of deregulated genes on other genes. Continuous lines show a direct
interaction, dotted lines an indirect interaction (color figure online)
Arch Toxicol
analyses were performed to identify transcription regulators
that may be commonly affected by CSA derivatives. For this
purpose, genes that were consistently deregulated by at least
two CSA derivatives were combined to a new dataset contain￾ing 208 genes. This dataset was used for upstream regulator
analyses. Thereby, ten upstream regulators were predicted to
be activated and four regulators were expected to be inhib￾ited after treatment (Table 3). The cellular tumor antigen p53
(TP53) and the transforming growth factor beta-1 (TGFB1)
control the highest number of deregulated target molecules
in the microarray experiments. Figure 3b shows a network of
deregulated genes controlled by TP53 and TGFB1.
Motif search
Deregulated genes were identified after application of a
fold change threshold of ±1.5. In total, 659 genes were
deregulated upon CSA treatment, 30 genes upon CSA6
treatment, 589 genes upon CSA9 treatment, and 142 genes
upon CSA19 treatment comprising 1270 genes in sum.
MatInspector analysis was performed separately for each
deregulated gene list. Motif family occurrences for each
gene were mainly considered, and the weighted motif score
was referred to rank the motif families. Most pronounced
motif families for each deregulated gene lists are depicted
in Table 4. E2F-MYC activator/cell cycle regulator and
MYC-associated zinc fingers were commonly observed
motif families in each gene expression profile. The mem￾bers of MYC motif families are listed in Table 5. Those
MYC motif family elements were used to evaluate the
percentage of MYC motif element presence in each gene
expression profile. MYC motif elements were observed
to be in high percentage and the genes carrying these ele￾ments were mostly down-regulated upon CSA treatment
(66 %) (Table 6). Motif analyses were in concordance with
the transcription factor analyses performed during micro￾array evaluation since E2F and E2F1 were significantly
affected transcriptional regulators in addition to AR upon
CSA6 treatment, whereas E2F4 was a significantly affected
transcriptional regulator in addition to TP53 and ERBB2
upon CSA treatment (Supplementary Table 2).
Reporter cell line experiments
The mRNA expression profiles indicated that CSA and its
derivatives affected Myc-related gene expression. Therefore,
Table 3 Upstream regulators presumably affected by all CSA derivatives
Genes that were consistently deregulated by at least two CSA derivatives were combined to a new dataset containing 208 genes. This dataset was
used for an upstream regulator analysis to identify transcription regulators that are affected by all CSA compounds
Upstream regulators Predicted activation state Deregulated target molecules
Symbol Description
ATF4 Cyclic AMP-dependent transcription factor ATF-4 Activated ASNS, ATF3, DDIT3, DDIT4, JUN, PSAT1, VEGFA
TP53 Cellular tumor antigen p53 Activated ABCB1, ATF3, AURKA, CCNA2, COL18A1,
DDIT3, DDIT4, FAM3C, FYN, HMOX1,
MRPL46, NDRG1, SCP2, SESN2, SGTB, SNRK,
TNFRSF10B, VCAN, VEGFA
MAP2K1/2 Dual specificity mitogen-activated protein kinase
kinase 1/2
Activated ATF3, DDIT3, HERPUD1, PHLDA1, VEGFA
TGFB1 Transforming growth factor beta-1 Activated AKR1C1/AKR1C2, ANGPTL4, CCNA2, CDKN2C,
COL18A1, HAS3, HMOX1, JUN, RAD51AP1,
TGM2, VEGFA
NFκB Nuclear factor NF-kappa-B Activated ABCB1, CFTR, HMOX1, IL23A, LCN2, NAMPT,
TGM2, TNFRSF10B, TRIB3
PPRC1 Peroxisome proliferator-activated receptor gamma
coactivator-related protein 1
Activated DDIT4, LCN2, NAMPT, PHLDA1
ERK Mitogen-activated protein kinases Activated HAS3, JUN, LCN2, NDRG1, TGM2
IFNG Interferon gamma Activated ATF3, CARD17, CCNA2, CFTR, CHAC1, IL23A,
LAMP3, STAT2, TNFRSF10B
PDGF Platelet-derived growth factor receptor alpha/beta Activated ATF3, JUN, NAMPT, PHLDA1
IL13 Interleukin-13 Activated ATF3, HPSE, PHLDA1, TGM2
TRIB3 Tribbles homolog 3 Inhibited ASNS, DDIT3, DDIT4, HERPUD1, PSAT1, TRIB3
GNE Bifunctional UDP-N-acetylglucosamine
2-epimerase/N-acetylmannosamine kinase
Inhibited ASNS, CHAC1, DDIT3, TRIB3
COL18A1 Collagen alpha-1(XVIII) chain Inhibited CDKN2C, DDIT4, JUN, VEGFA
IL1RN Interleukin-1 receptor antagonist protein Inhibited ATF3, CHAC1, LAMP3, STAT2
Arch Toxicol
Table 4 Most pronounced motif families depending on MatInspector analysis for deregulated genes upon CSA and derivatives treatment
Motif family Gene # (out of 659) % Motif frequency average Weighted motif score
CSA
Human and murine ETS1 factors 223 0.338 21.148 7.148
E2F-MYC activator/cell cycle regulator 430 0.653 8.963 5.853
Fork head domain factors 386 0.586 9.591 5.620
Octamer-binding protein 459 0.697 7.651 5.333
Homeobox transcription factors 422 0.640 8.296 5.309
C2H2 zinc finger transcription factors 2 335 0.508 10.415 5.291
Cart-1 (cartilage homeoprotein 1) 490 0.744 7.033 5.233
Twist subfamily of class B bHLH transcription factors 606 0.920 5.422 4.988
Abdominal-B type homeodomain transcription factors 484 0.734 6.372 4.677
Heat-shock factors 542 0.822 5.308 4.363
ZF5 POZ domain zinc finger 471 0.715 5.964 4.264
Hepatic Nuclear Factor 1 588 0.892 4.760 4.246
GATA-binding factors 495 0.751 5.495 4.127
MYC-associated zinc fingers 518 0.786 4.905 3.855
Brn POU domain factors 258 0.392 9.663 3.788
RXR heterodimer-binding sites 523 0.794 4.495 3.569
Distal-less homeodomain transcription factors 524 0.795 4.460 3.546
AT-rich interactive domain factor 403 0.612 5.648 3.457
Pleomorphic adenoma gene 423 0.642 5.348 3.433
NKX homeodomain factors 327 0.496 6.869 3.407
EGR/nerve growth factor-induced protein C and related factors 290 0.440 7.497 3.299
cAMP-responsive element-binding proteins 422 0.640 5.152 3.297
EVI1-myleoid-transforming protein 214 0.325 9.551 3.104
SOX/SRY-sex/testis-determining and related HMG box factors 194 0.294 10.479 3.081
Homeodomain transcription factors 127 0.193 13.630 2.631
Krueppel-like transcription factors 148 0.224 10.986 2.461
GC-Box factors SP1/GC 186 0.282 7.258 2.047
Lim homeodomain factors 137 0.208 9.328 1.940
Paralog hox genes 1-8 from the four hox clusters A, B, C, D 125 0.190 9.688 1.841
FAST-1 SMAD-interacting proteins 190 0.288 5.410 1.558
CSA6
Myeloid zinc finger 1 factors 17 0.5667 4.588 2.560
MYC-associated zinc fingers 16 0.533 4.750 2.533
GLI zinc finger family 17 0.5667 4.471 2.534
Vertebrate SMAD family of transcription factors 16 0.533 4.625 2.467
C2H2 zinc finger transcription factors 7 15 0.500 4.930 2.465
CTCF and BORIS gene family, transcriptional regulators with 11
highly conserved zinc finger domains
16 0.533 4.125 2.200
Distal-less homeodomain transcription factors 13 0.433 4.486 1.944
GA-boxes 5 0.167 7.000 1.167
E2F-MYC activator/cell cycle regulator 1 0.033 16.000 0.533
CSA9
Distal-less homeodomain transcription factors 241 0.409 4.278 1.750
C2H2 zinc finger transcription factors 2 73 0.124 11.520 1.428
E2F-MYC activator/cell cycle regulator 87 0.148 9.299 1.374
Krueppel-like transcription factors 69 0.117 11.638 1.3634
Homeodomain transcription factors 69 0.117 11.145 1.306
Arch Toxicol
Table 4 continued
Motif family Gene # (out of 659) % Motif frequency average Weighted motif score
Paralog hox genes 1-8 from the four hox clusters A, B, C, D 75 0.127 9.533 1.214
Lim homeodomain factors 72 0.122 9.486 1.160
Fork head domain factors 71 0.121 9.296 1.121
SOX/SRY-sex/testis-determining and related HMG box factors 66 0.112 9.682 1.085
MYC-associated zinc fingers 109 0.185 5.734 1.061
Cart-1 (cartilage homeoprotein 1) 83 0.141 7.482 1.054
Human and murine ETS1 factors 59 0.100 10.491 1.051
Brn POU domain factors 72 0.122 7.958 0.973
Pleomorphic adenoma gene 82 0.139 6.817 0.949
NKX homeodomain factors 79 0.134 6.861 0.920
ZF5 POZ domain zinc finger 70 0.119 6.600 0.784
AT-rich interactive domain factor 80 0.136 5.613 0.762
EGR/nerve growth factor-induced protein C and related factors 56 0.095 7.839 0.745
GC-Box factors SP1/GC 65 0.110 6.200 0.684
Activator protein 2 78 0.132 4.231 0.560
GA-boxes 66 0.112 4.197 0.470
CSA19
CTCF and BORIS gene family, transcriptional regulators with 11
highly conserved zinc finger domains
51 0.360 5.294 1.901
Homeodomain transcription factors 9 0.063 14.667 0.930
MYC-associated zinc fingers 18 0.127 7.167 0.908
C2H2 zinc finger transcription factors 2 9 0.063 14.220 0.901
Krueppel-like transcription factors 9 0.063 14.000 0.887
E2F-MYC activator/cell cycle regulator 12 0.085 8.750 0.739
SOX/SRY-sex/testis-determining and related HMG box factors 8 0.056 12.250 0.690
Pleomorphic adenoma gene 12 0.085 7.500 0.634
EGR/nerve growth factor-induced protein C and related factors 8 0.056 9.250 0.521
Lim homeodomain factors 6 0.042 12.167 0.514
ZF5 POZ domain zinc finger 9 0.063 8.000 0.507
Paralog hox genes 1-8 from the four hox clusters A, B, C, D 8 0.056 9.000 0.507
Brn POU domain factors 6 0.042 11.833 0.450
GC-Box factors SP1/GC 7 0.049 9.000 0.444
Table 5 Members of MYC
motif families E2F-MYC activator/cell cycle regulator
E2F transcription factor 2
E2F transcription factor 3
E2F transcription factor 3 (secondary DNA-binding preference)
E2F transcription factor 4, p107/p130-binding protein
E2F, involved in cell cycle regulation, interacts with Rb p107 protein
E2F-4/DP-1 heterodimeric complex
E2F-4/DP-2 heterodimeric complex
E2F-1/DP-1 heterodimeric complex
RB/E2F-1/DP-1 heterotrimeric complex
E2F-1/DP-2 heterodimeric complex
MYC-associated zinc fingers
MYC-associated zinc finger protein (MAZ)
MYC-associated zinc finger protein-related transcription factor
MYC-interacting Zn finger protein 1
Myc-interacting Zn finger protein 1, zinc finger and BTB domain containing 17 (ZBTB17)
Arch Toxicol
we hypothesized that these compounds may target MYC.
To prove this, we generated a MYC reporter cell line by
transfecting HEK293 cells with a c-MYC-luciferase reporter
construct. This cell line was treated with each of the four
compounds to see whether or not the transcriptional activity
of the c-Myc promoter was inhibited. Indeed, CSA and its
derivatives showed inhibitory effects on c-MYC in a compa￾rable manner as two known MYC inhibitors, 10058-F4 and
10074-G5 did (Fig. 4a). CSA, CSA9, and CSA19 suppressed
the c-MYC activity in a dose-dependent manner. However,
CSA6 did not reveal a significant inhibition. These results
are in concordance with our microarray analysis, since CSA6
treatment of MCF7 caused deregulated expression values in
only 30 genes, eight of which were down-regulated and have
MYC-binding motifs at their promoter sequences (Table 6).
The concentrations required to achieve a 50 % MYC inhibi￾tion are shown in Fig. 4b. CSA revelaed a lower EC50 con￾centration than the control compound, 10058-F4. CSA6,
CSA9, and CSA19 had lower EC50 concentrations than both
10058-F4 and 10074-G5, indicating that CSA and its deriva￾tives possess MYC inhibitory activities.
Molecular docking
To further investigate the possible interaction of the four
CSA compounds with MYC, we performed molecular
docking studies in silico. CSA and its derivatives docked in
close proximity to the DNA-binding domain of MYC and
MAX with considerably high binding energies. When the
average values of the lowest binding energies are consid￾ered, CSA revealed binding energies of −6.34 ± 0.01 kcal/
mol in blind docking and −6.60 ± 0.05 kcal/mol in
defined docking. It formed hydrogen bond with resi￾dues residing at the DNA-binding regions, i.e., Arg214
on MAX and Lys939 on MYC. CSA6 revealed binding
energies of −5.79 ± 0.10 kcal/mol in blind docking and
−6.27 ± 0.13 kcal/mol in the defined docking approach.
CSA9 showed binding energies of −5.71 ± 0.02 kcal/
mol in blind docking and −6.06 ± 0.05 kcal/mol in
defined docking. CSA19 showed binding energies of
−6.15 ± 0.01 kcal/mol using the blind docking mode and
−6.24 ± 0.05 kcal/mol in the defined docking. 10058-F4
showed −4.74 ± 0.00 kcal/mol in the blind docking and
−4.93 ± 0.02 kcal/mol in the defined docking. 10074-G5
showed −7.82 ± 0.01 kcal/mol in the blind docking and
−8.07 ± 0.04 kcal/mol in the defined docking. When
compared to the control inhibitors 10074-G5 and 10058-
F4, CSA derivatives showed comparable binding ener￾gies and docked to the same binding site. The results are
Table 6 Number of down- and up-regulated genes carrying MYC motif elements upon treatment with CSA derivatives
Number of
deregulated genes
Number of deregulated genes
carrying MYC motif elements
Number of down-regulated genes car￾rying MYC motif elements
Number of up-regulated genes
carrying MYC motif elements
CSA 659 608 400 208
CSA6 30 17 8 9
CSA9 589 391 165 226
CSA19 142 87 20 67
CSA CSA6 CSA9 CSA19 10058F4 10074G5
Drug concentration (µM)
EC50 of Myc inhibition
% normalized Myc activity
EC50 (µM)
CSA 15.48 ± 6.84
CSA6 5.20 ± 0.27
CSA9 5.79 ± 0.21
CSA19 1.40 ± 0.64
10058F4 52.63 ± 4.55
10074G5 12.80 ± 0.59
Fig. 4 Effect of CSA, its derivatives, 10058-F4 and 10074-G5 on
c-MYC transcription factor activity. a Normalized c-MYC activ￾ity after treatment with CSA, CSA6, CSA9, CSA19, 10058-F4, and
10074-G5 (two independent experiments with three parallel measure￾ments). Concentrations of IC50, IC50/10 and IC50/100 are tested for
each compound on transiently transfected HEK293 cells with the
MYC luciferase reporter construct. (*p value <0.05) b EC50 concen￾trations of c-MYC inhibition for CSA, CSA6, CSA9, CSA19, 10058-
F4 and 10074-G5
Arch Toxicol
summarized in Fig. 5. Residues marked bold reside at the
DNA-binding region of the proteins indicating that CSA
and its derivatives may indeed target the c-MYC complex
and induce MYC-related changes in the gene expression
profiles.
Discussion
Cajanin stilbene acid (CSA) was previously found to exert
antioxidant, antimicrobial, and anticancer effects (Eich￾horn 2011; Wu et al. 2011; Zu et al. 2010). In the present
study, we investigated the molecular modes of action and
observed MYC inhibitory activities of CSA and three
derivatives toward MCF-7 breast cancer cells.
These compounds revealed cytotoxicity against MCF-7
cells in the micromolar range. We applied mRNA micro￾array analyses to investigate the compounds’ molecu￾lar modes of action. Differentially expressed genes were
subjected to Ingenuity Pathway analysis to identify pos￾sible signaling routes. A comparative analysis of gene
expression data of CSA and its derivatives suggested cell
cycle, cell proliferation, and DNA replication as common
cellular mechanisms involved in the cytotoxic effect of
these compounds.
The microarray analyses were accompanied by motif
analyses of the promoter sequences of the deregulated
genes. We have chosen this approach to identify transcrip￾tion factors possibly responsible for the observed gene
expression profiles. Cell cycle and proliferation-related
transcription factors as well as MYC family transcription
factors were commonly identified from the gene expression
profiles. The percentage of the down-regulated genes carry￾ing MYC motif elements at their promoter sequences was
remarkably high, indicating that CSA and its derivatives
may be potential MYC inhibitors.
In order to prove this hypothesis, we generated a
HEK293 reporter cell line carrying a c-MYC responsive
reporter construct. A sensitive dual-luciferase assay based
on the luminescence of firefly and renilla luciferase was
used to measure MYC activity. CSA, CSA9, and CSA19
inhibited MYC activity in a dose-dependent manner. Two
known MYC inhibitors (10058-F4 and 10074-G5) were
used as positive control. CSA and its derivatives showed
better MYC inhibition than 10058-F4 and in addition, the
CSA derivatives inhibited MYC better than 10074-G5.
CSA
CSA6
CSA9
CSA19
10058-F4
10074-G5
Blind
Docking
Lowest
Binding
Energy
(kcal/mol)
Mean
Binding
Energy
(kcal/mol)
Residues Making
H-Bond
Residues Involved In
Hydrophobic Interactions
pKi (µM)
CSA -6.34±0.01 -5.68±0.06 on MAX: Arg214
on MYC: Lys939
on MAX: Arg214
on MYC: Lys918, Phe921,
Phe922, Pro938, Lys939
22.60±0.27
CSA6 -5.79±0.10 -5.13±0.01 on MYC: Asp926,
Glu932
on MAX: Arg254
on MYC: Arg925, Asp926,
Pro929, Glu932, Glu935
57.60±10.10
CSA9 -5.71±0.02 -5.51±0.01 – on MAX: Arg254
on MYC: Arg925, Asp926,
Gln927, Pro929, Glu932
65.10±1.83
CSA19 -6.15±0.01 -5.78±0.02 – on MYC: Arg925, Asp926,
Gln927, Ile928, Pro929, Leu931,
Glu932, Glu935
31.20±0.25
10058F4 -4.74±0.00 -4.72±0.00 – on MAX: Arg254
on MYC: Arg925, Asp926, Ile928,
Pro929, Leu931, Glu932
337.00±0.71
10074G5 -7.82±0.01 -7.44±0.06 on MAX: Arg214
on MYC: Lys939
on MAX: Arg214
on MYC: Phe921, Lys936, Ala937,
Pro938, Lys939, Ile942
1.86±0.02
Defined
Docking
Lowest
Binding
Energy
(kcal/mol)
Mean
Binding
Energy
(kcal/mol)
Residues Making
H-Bond
Residues Involved In
Hydrophobic Interactions
pKi (µM)
CSA -6.60±0.05 -5.82±0.04 on MAX: Arg214
on MYC: Lys939
on MAX: Arg214
on MYC: Lys918, Phe921,
Phe922, Pro938, Lys939, Ile942
14.60±1.06
CSA6 -6.27±0.13 -5.26±0.13 on MYC: Glu932 on MAX: Arg254
on MYC: Arg925, Asp926,
Pro929, Leu931, Glu932,
Asn933, Glu935
25.60±5.14
CSA9 -6.06±0.05 -5.68±0.07 on MYC: Asp926 on MYC: Arg925, Asp926,
Gln927, Pro929, Glu932, Glu935
36.10±2.79
CSA19 -6.24±0.05 -5.87±0.07 on MYC: Asp926 on MAX: Arg254
on MYC: Arg925, Asp926,
Gln927, Pro929, Glu932, Glu935
26.80±2.34
10058F4 -4.93±0.02 -4.80±0.02 on MYC: Arg925 on MAX: Arg254
on MYC: Arg925, Asp926,
Ile928, Pro929, Leu931, Glu932
242.31±4.73
10074G5 -8.07±0.04 -7.38±0.02 on MYC: Lys936,
Lys939
on MYC: Lys936, Pro938,
Lys939
1.22±0.08
CSA
CSA6
CSA9
CSA19
10058-F4
10074-G5
Blind docking Defined docking
Fig. 5 Binding modes and energies of CSA, its derivatives, 10058-F4 and 10074-G5 on MYC/MAX complex (PDB ID: 1NKP). CSA and its
derivatives occupy the same binding sites on MYC/MAX complex as 10058-F4 and 10074-G5
Arch Toxicol
This is in concordance with the hypothesis that CSA and
its derivatives exert their effect by targeting the cell cycle
and proliferation pathways on cancer cells by inhibiting
c-MYC activity.
This point of view was also supported by bioinformatics,
which molecular docking studies indicated that CSA and
its derivatives interact with the MYC/MAX complex with
comparable binding energies as two known MYC inhibi￾tors, 10058-F4 and 10074-G5. This result can be taken
as another clue that CSA and its derivatives are potential
MYC inhibitors.
In conclusion, CSA and its derivatives showed cyto￾toxic activity toward MCF-7 breast cancer cells and inhib￾ited MYC activity in a luciferase-based HEK293 reporter
cell line in a dose-dependent manner. Microarray profil￾ing and motif analyses of promoter regions of deregulated
genes pointed to MYC as the most pronounced transcrip￾tion factor involved in the action of CSA and its deriva￾tives. Molecular docking calculations further supported the
hypothesis that CSA and its derivatives inhibit MYC activ￾ity. Further analyses in preclinical and clinical studies are
warranted to clarify the therapeutic potential of CSA and
its derivatives for clinical use.
Conflict of interest We declare that there is no conflict of interest.
References
Adams JM, Harris AW, Pinkert CA et al (1985) The c-MYC oncogene
driven by immunoglobulin enhancers induces lymphoid malig￾nancy in transgenic mice. Nature 318(6046):533–538
Addou-Klouche L, Adelaide J, Cornen S et al (2012) Integrated
genomic analysis of breast cancers. Balkan J Med Genet BJMG
15(Suppl):71–74. doi:10.2478/v10034-012-0023-x
Ali S, Coombes RC (2002) Endocrine-responsive breast cancer and
strategies for combating resistance. Nat Rev Cancer 2(2):101–
112. doi:10.1038/nrc721
Amundadottir LT, Johnson MD, Merlino G, Smith GH, Dickson RB
(1995) Synergistic interaction of transforming growth factor alpha
and c-MYC in mouse mammary and salivary gland tumorigenesis.
Cell Growth Differ Mol Biol J Am Assoc Cancer Res 6(6):737–748
Arnold I, Watt FM (2001) c-MYC activation in transgenic mouse epi￾dermis results in mobilization of stem cells and differentiation of
their progeny. Curr Biol CB 11(8):558–568
Bhardwaj A, Rosen D, Liu M et al (2014) suppression of Akt-mTOR
pathway—a novel component of oncogene induced DNA damage
response barrier in breast tumorigenesis. PLoS ONE 9(5):e97076.
doi:10.1371/journal.pone.0097076
Blackwell TK, Kretzner L, Blackwood EM, Eisenman RN, Weintraub
H (1990) Sequence-specific DNA binding by the c-MYC protein.
Science 250(4984):1149–1151
Butt AJ, McNeil CM, Musgrove EA, Sutherland RL (2005) Downstream
targets of growth factor and oestrogen signalling and endocrine
resistance: the potential roles of c-MYC, cyclin D1 and cyclin E.
Endocr Relat Cancer 12(Suppl 1):S47–S59. doi:10.1677/erc.1.00993
Byler S, Goldgar S, Heerboth S et al (2014) Genetic and epigenetic
aspects of breast cancer progression and therapy. Anticancer Res
34(3):1071–1077
Cartharius K, Frech K, Grote K et al (2005) MatInspector and beyond:
promoter analysis based on transcription factor binding sites.
Bioinformatics 21(13):2933–2942. doi:10.1093/bioinformatics/
bti473
Castro F, Dirks WG, Fahnrich S, Hotz-Wagenblatt A, Pawlita M,
Schmitt M (2013) High-throughput SNP-based authentication
of human cell lines. Int J Cancer 132(2):308–314. doi:10.1002/
ijc.27675
Chen Y, Olopade OI (2008) MYC in breast tumor progression. Expert Rev
Anticancer Ther 8(10):1689–1698. doi:10.1586/14737140.8.10.1689
Cheng AS, Jin VX, Fan M et al (2006) Combinatorial analysis of tran￾scription factor partners reveals recruitment of c-MYC to estro￾gen receptor-alpha responsive promoters. Mol Cell 21(3):393–
404. doi:10.1016/j.molcel.2005.12.016
Clarke R, Liu MC, Bouker KB et al (2003) Antiestrogen resistance in
breast cancer and the role of estrogen receptor signaling. Onco￾gene 22(47):7316–7339. doi:10.1038/sj.onc.1206937
Clausen DM, Guo J, Parise RA et al (2010) In vitro cytotoxicity and
in vivo efficacy, pharmacokinetics, and metabolism of 10074-
G5, a novel small-molecule inhibitor of c-MYC/MAX dimeri￾zation. J Pharmaco Exp Ther 335(3):715–727. doi:10.1124/
jpet.110.170555
D’Cruz CM, Gunther EJ, Boxer RB et al (2001) c-MYC induces
mammary tumorigenesis by means of a preferred pathway
involving spontaneous Kras2 mutations. Nat Med 7(2):235–239.
doi:10.1038/84691
Eichhorn T (2011) Molecular biology, pharmacogenomics and bio￾informatics of natural compounds and synthetic compounds for
cancer therapy. Ph.D. Thesis. Ruperto-Carola University of Hei￾delberg, Germany
Escot C, Theillet C, Lidereau R et al (1986) Genetic alteration of the
c-MYC protooncogene (MYC) in human primary breast carcino￾mas. Proc Natl Acad Sci USA 83(13):4834–4838
Felsher DW, Bishop JM (1999) Reversible tumorigenesis by MYC in
hematopoietic lineages. Mol Cell 4(2):199–207
Follis AV, Hammoudeh DI, Wang H, Prochownik EV, Metallo SJ
(2008) Structural rationale for the coupled binding and unfold￾ing of the c-MYC oncoprotein by small molecules. Chem Biol
15(11):1149–1155. doi:10.1016/j.chembiol.2008.09.011
Guerin M, Barrois M, Terrier MJ, Spielmann M, Riou G (1988) Over￾expression of either c-MYC or c-erbB-2/neu proto-oncogenes
in human breast carcinomas: correlation with poor prognosis.
Oncog Res 3(1):21–31
Harrington EA, Bennett MR, Fanidi A, Evan GI (1994) c-MYC￾induced apoptosis in fibroblasts is inhibited by specific cytokines.
EMBO J 13(14):3286–3295
Jonkers J, Berns A (2004) Oncogene addiction: sometimes a tem￾porary slavery. Cancer Cell 6(6):535–538. doi:10.1016/j.
ccr.2004.12.002
Kallio MA, Tuimala JT, Hupponen T et al (2011) Chipster: user￾friendly analysis software for microarray and other high-through￾put data. BMC Genom 12:507. doi:10.1186/1471-2164-12-507
Kim J, Lee JH, Iyer VR (2008) Global identification of Myc target
genes reveals its direct role in mitochondrial biogenesis and its
E-box usage in vivo. PLoS ONE 3(3):e1798. doi:10.1371/journal.
pone.0001798
Kim B, Fatayer H, Hanby AM et al (2013) Neoadjuvant chemother￾apy induces expression levels of breast cancer resistance protein
that predict disease-free survival in breast cancer. PLoS ONE
8(5):e62766. doi:10.1371/journal.pone.0062766
Kirkland JL, Murthy L, Stancel GM (1992) Progesterone inhibits the
estrogen-induced expression of c-fos messenger ribonucleic acid
in the uterus. Endocrinology 130(6):3223–3230. doi:10.1210/
endo.130.6.1375896
Musgrove EA, Sergio CM, Loi S et al (2008) Identification of func￾tional networks of estrogen- and c-MYC-responsive genes and
Arch Toxicol
their relationship to response to tamoxifen therapy in breast can￾cer. PLoS ONE 3(8):e2987. doi:10.1371/journal.pone.0002987
Nair SK, Burley SK (2003) X-ray structures of Myc-Max and Mad￾Max recognizing DNA. Molecular bases of regulation by proto￾oncogenic transcription factors. Cell 112(2):193–205
Nicholson RI, Hutcheson IR, Britton D et al (2005) Growth factor
signalling networks in breast cancer and resistance to endocrine
agents: new therapeutic strategies. J Steroid Biochem Mol Biol
93(2–5):257–262. doi:10.1016/j.jsbmb.2004.12.006
Parris TZ, Kovacs A, Hajizadeh S et al (2014) Frequent MYC coam￾plification and DNA hypomethylation of multiple genes on 8q
in 8p11–p12-amplified breast carcinomas. Oncogenesis 3:e95.
doi:10.1038/oncsis.2014.8
Pavesi G, Pesole G (2006) Using Weeder for the discovery of con￾served transcription factor binding sites. Current protocols in bio￾informatics/editoral board, Andreas D Baxevanis et al. Chapter 2:
Unit 2 11. doi:10.1002/0471250953.bi0211s15
Pelengaris S, Khan M, Evan G (2002a) c-MYC: more than just
a matter of life and death. Nat Rev Cancer 2(10):764–776.
doi:10.1038/nrc904
Pelengaris S, Khan M, Evan GI (2002b) Suppression of Myc-induced
apoptosis in beta cells exposes multiple oncogenic properties of
Myc and triggers carcinogenic progression. Cell 109(3):321–334
Plourde KV, Labrie Y, Desjardins S et al (2013) Analysis of ZNF350/
ZBRK1 promoter variants and breast cancer susceptibility in
non-BRCA1/2 French Canadian breast cancer families. J Hum
Genet 58(2):59–66. doi:10.1038/jhg.2012.127
Ponzielli R, Katz S, Barsyte-Lovejoy D, Penn LZ (2005) Can￾cer therapeutics: targeting the dark side of Myc. Eur J Cancer
41(16):2485–2501. doi:10.1016/j.ejca.2005.08.017
Prendergast GC, Ziff EB (1992) A new bind for Myc. Trends Genet
TIG 8(3):91–96
Schuldiner O, Benvenisty N (2001) A DNA microarray screen
for genes involved in c-MYC and N-MYC oncogenesis in
human tumors. Oncogene 20(36):4984–4994. doi:10.1038/
sj.onc.1204459
Shachaf CM, Kopelman AM, Arvanitis C et al (2004) MYC inacti￾vation uncovers pluripotent differentiation and tumour dor￾mancy in hepatocellular cancer. Nature 431(7012):1112–1117.
doi:10.1038/nature03043
Shang Y, Baumrucker CR, Green MH (1998) c-MYC is a major
mediator of the synergistic growth inhibitory effects of
retinoic acid and interferon in breast cancer cells. J Biol Chem
273(46):30608–30613
Sousa SF, Fernandes PA, Ramos MJ (2006) Protein-ligand dock￾ing: current status and future challenges. Proteins 65(1):15–26.
doi:10.1002/prot.21082
Venditti M, Iwasiow B, Orr FW, Shiu RP (2002) c-MYC gene expres￾sion alone is sufficient to confer resistance to antiestrogen in
human breast cancer cells. Int J Cancer 99(1):35–42
Vennstrom B, Sheiness D, Zabielski J, Bishop JM (1982) Isolation
and characterization of c-MYC, a cellular homolog of the onco￾gene (v-MYC) of avian myelocytomatosis virus strain 29. J Virol
42(3):773–779
Wang YH, Liu S, Zhang G et al (2005) Knockdown of c-MYC
expression by RNAi inhibits MCF-7 breast tumor cells growth
in vitro and in vivo. Breast Cancer Res BCR 7(2):R220–R228.
doi:10.1186/bcr975
Wang H, Hammoudeh DI, Follis AV et al (2007) Improved low
molecular weight Myc-Max inhibitors. Mol Cancer Ther
6(9):2399–2408. doi:10.1158/1535-7163.MCT-07-0005
Wu N, Kong Y, Fu Y et al (2011) In vitro antioxidant properties, DNA
damage protective activity, and xanthine oxidase inhibitory effect
of cajaninstilbene acid, a stilbene compound derived from pigeon
pea [Cajanus cajan (L.) Millsp.] leaves. J Agric Food Chem
59(1):437–443. doi:10.1021/jf103970b
Yan Y, Zhang GX, Williams MS et al (2012) TCR stimulation upregu￾lates MS4a4B expression through induction of AP-1 transcrip￾tion factor during T cell activation. Mol Immunol 52(2):71–78.
doi:10.1016/j.molimm.2012.04.011
Yap JL, Wang H, Hu A et al (2013) Pharmacophore identification of
c-MYC inhibitor 10074-G5. Bioorg Med Chem Lett 23(1):370–
374. doi:10.1016/j.bmcl.2012.10.013
Ziv-Av A, Taller D, Attia M et al (2011) RTVP-1 expression is
regulated by SRF downstream of protein kinase C and contrib￾utes to the effect of SRF on glioma cell migration. Cell Signal
23(12):1936–1943. doi:10.1016/j.cellsig.2011.07.001
Zu YG, Liu XL, Fu YJ, Wu N, Kong Y, Wink M (2010) Chemical
composition of the SFE-CO extracts from Cajanus cajan (L.)
Huth and their antimicrobial activity in vitro and in vivo. Phy￾tomedicine Int J Phytotherapy Phytopharm 17(14):1095–1101.
doi:10.1016/j.phymed.2010.04.005