Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
1
Context-selective death of acute myeloid leukemia cells triggered by the novel hybrid retinoid-
HDAC inhibitor MC2392
Floriana De Bellis1,2, Vincenzo Carafa1, Mariarosaria Conte1, Dante Rotili3, Francesca Petraglia1,
Filomena Matarese2, Kees-Jan Francoijs2, Julien Ablain5, Sergio Valente3, Remy Castellano4,
Armelle Goubard4, Yves Collette4, Amit Mandoli2, Joost H. A. Martens2, Hugues de Thé5, Angela
Nebbioso1*, Antonello Mai3*, Hendrik G. Stunnenberg2*, Lucia Altucci 1,6*.
1 Dipartimento di Biochimica, Biofisica e Patologia Generale, Seconda Università degli Studi di
Napoli, Vico L. De Crecchio 7, 80138, Napoli, IT; 2 NCMLS, Radboud University, Nijmegen, NL;
3 Dipartimento di Chimica e Tecnologie del Farmaco, Sapienza University of Rome, P.le Aldo
Moro 5, 00185, Rome, IT; 4 Inserm, CRCM, U1068, TrGET & ISCB, University of Marseille, F-
13009 France; 5 Laboratoire U944 and UMR 7212, University Paris-Diderot, Paris, FR; 6 Istituto di
Genetica e Biofisica, IGB, Adriano Buzzati Traverso, Via P. Castellino 111, 80131, Naples, IT.
Running title: MC2392 induces PML-RARα context-selective cell-death.
Keywords: HDAC inhibitors, retinoic acid, PML-RARα-HDAC complexes, APL targeted therapy,
cell death1
Financial support: Blueprint (n° 282510) to LA, HS, MA; Italian Flag Project: EPIGEN, AIRC n° 11812, PRIN_2009PX2T2E_004, PON0101227 to LA Corresponding authors: Prof. Lucia Altucci, Dipartimento di Biochimica, Biofisica e Patologia Generale, Seconda Università degli Studi di Napoli, Vico L. De Crecchio 7, 80138, Napoli, IT, Tel: +390815667569 Fax: +39081450169 [email protected]; Prof. Hendrik G. Stunnenberg, Department of Molecular Biology, Faculty of Science, Nijmegen Center for Molecular Life Sciences, Radboud University, 6500 HB Nijmegen Tel: +31243510524�Fax: +31243610520 [email protected]; Prof A. Mai: Dipartimento di Chimica e Tecnologie del Farmaco Sapienza Università di Roma, 00185 Roma, IT; Tel: +390649913392 Fax: +390649693268�[email protected]; Dr Angela Nebbioso Dipartimento di Biochimica, Biofisica e Patologia Generale, Seconda Università degli Studi di Napoli, Vico L. De Crecchio 7, 80138, Napoli, IT Tel: +390815665673 Fax: +39081450169. Disclosure of Potential Conflicts of Interest: No potential conflicts of interest were disclosed. Word count: 4629 Number of figures: 6
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
2
ABSTRACT
HDAC inhibitors (HDACi) are widely used in the clinic to sensitize tumorigenic cells for
treatment with other anticancer compounds. The major drawback of HDACi is the broad
inhibition of the plethora of HDAC-containing complexes. In acute promyelocytic leukemia
(APL), repression by the PML-RARαoncofusion protein is mediated by an HDAC-containing
complex that can be dissociated by pharmacological doses of all trans retinoic acid (ATRA)
inducing differentiation and cell death at the expense of side effects and recurrence. We
hypothesized that the context-specific close physical proximity of a retinoid and HDACi
binding protein in the repressive PML-RARα-HDAC complex may permit selective targeting
by a hybrid molecule of ATRA with a 2-aminoanilide tail of the HDAC inhibitor MS-275,
yielding MC2392. We show that MC2392 elicits weak ATRA and essentially no HDACi
activity in vitro or in vivo. Genome-wide epigenetic analyses revealed that in NB4 cells
expressing PML-RARα, MC2392 induces changes in H3 acetylation at a small subset of
PML-RARα binding sites. RNA-seq reveals that MC2392 alters expression of a number of
stress-responsive and apoptotic genes. Concordantly, MC2392 induced rapid and massive,
caspase 8-dependent cell death accompanied by RIP1 induction and ROS production. Solid
and leukemic tumors are not affected by MC2392, but expression of PML-RARα conveys
efficient MC2392-induced cell death. Our data suggest a model in which MC2392 binds to
the RARα moiety and selectively inhibits the HDACs resident in the repressive complex
responsible for the transcriptional impairment in APLs. Our findings provide proof-of-
principle of the concept of a context-dependent targeted therapy.
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
3
Introduction
Acute promyelocytic leukemia (APL) is characterized by the presence of blasts blocked at the
promyelocytic stage of myeloid differentiation and by the fusion protein of RARα with the
promyelocytic leukemia gene (PML) in more than 95% of the cases (1, 2). Despite that the majority
of APLs carry this t(15;17) translocation, rare alternative RARα-containing fusion proteins have
been described (3-5). In the treatment of APL, the efficacy of pharmacologic doses of ATRA is due
to its ability to release the HDAC-containing repressive complexes bound to PML-RARα (6) and to
recruit the multi-subunit HAT complex on RARE. Besides having this desired therapeutic effect on
APL blasts, ATRA binds to and activates the nuclear receptors (RARα, β or γ) that bind to specific
DNA responsive elements (RARE) (7, 8) thereby affect a variety of biological processes in essential
all cell types such as cell proliferation, differentiation and apoptosis (9, 10).
The PML-RARα mediated epigenetic block of gene transcription (11, 12) can also be overcome by
HDAC inhibitors (HDACi) allowing cells to re-start differentiation or pro-apoptotic events (13).
Activation of differentiation programs, inhibition of the cell cycle and eventually induction of
apoptosis are among the key antitumor activities of HDACi in cancer therapy (7, 11, 14-16).
HDACi cause enzymatic inhibition and/or the release of HDACs from repressive complexes
permitting HAT recruitment, histone acetylation, chromatin decondensation and, ultimately,
transcription activation of tumor-suppressor genes (TSG) or other genes crucial for the normal
functioning of the cells (7). The major drawback of HDACi in treatment is the broad inhibition of
the plethora of HDAC-containing complexes.
MS-275/Entinostat is one of the most potent and widely used HDACi with micromolar affinity for
class I HDACs and with selectivity for HDAC1, HDAC3 and HDAC8 (17-19). MS-275 promotes
differentiation, apoptosis and inhibits the proliferation of multiple cancer cell lines (20). The effects
of MS-275 have been examined in human leukemia and lymphoma cells (U937, HL-60, K562, and
Jurkat). When administered at a low concentration (e.g., 1 μM), MS-275 exhibited potent
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
4
antiproliferative activity inducing p21CIP1/WAF1-mediated growth arrest and expression of
differentiation markers (CD11b) in U937 cells (14, 21). However, at higher concentrations (e.g., 5
μM), MS-275 potently induced cell death and a very early increase in death receptor signaling (22)
as well as in reactive oxygen species (ROS), followed by the loss of mitochondrial membrane
potential and cytosolic release of cytochrome c (23).
The use of HDACi in combination with other anticancer agents (epi-drugs such as 5-aza-2′-
deoxycytidine and retinoic acid) is one way forward to a promising application against cancer. In
patients with leukemia, and particularly in the elderly, clinical studies combining ATRA treatment
with HDACi are in progress (24). For example, Valproic acid (VPA) has for many years been and
still is used as an anti-epileptic drug and inhibits preferably class I HDACs although in the high
micromolar to millimolar range. Moreover, even though there were a lot of concerns regarding toxic
side effects of HDACi in the clinical setting due to the roles of HDACs in multiples pathways, until
now clinical trials mostly showed manageable side effects (25). The combination of ATRA with an
HDACi could result in improved anti-tumorigenic activity (24).
Here, we exploited the synergy between HDACi and ATRA by generating and testing a single
hybrid molecule, named MC2392 on APL cells. Genome-wide epigenetic analysis revealed that
MC2392 is a weak retinoid. In contrast to the genome wide effects of MS-275 on histone H3
acetylation, MC2392 induces acetylation at only a small subset of PML-RARα binding sites in NB4
cells line. RNA-seq analysis showed that MC2392 alters the expression of a number of stress-
responsive and apoptotic genes differently from ATRA. Importantly, the hybrid compound acts in a
context and PML-RARα fusion protein-dependent fashion to induce rapid and massive cell death,
RIP1 induction and ROS production. We propose a model in which this hybrid compound binds to
the RARα via its retinoid-moiety and selectively inhibits the HDACs contained in the repressive
complex via the MS-275 part. Taken together, MC2392 is a promising candidate for apoptosis-
based therapy of APL, representing a new and effective, single hybrid drug able to modulate
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
5
multiple cell death pathways. Our study convincingly shows that the targeting of multiple signaling
pathways by a single hybrid drug is a feasible and attractive paradigm for new cancer therapies.
Material and Methods
Cells. NB4, LnCap and U937 were purchased by DSMZ and ATCC respectively. Cell lines have
been tested and authenticated following manufacture’s instruction.
NB4, NB4-R4 (26, 27), U937, U937 PML/RAR, U937-PLZF/RAR Zn inducible (28, 29) and
LnCap cell lines were grown at 37°C in air and 5% CO2 in RPMI 1640 medium (EUROCLONE),
supplemented with 10% heat-inactivated fetal bovine serum (FBS, Sigma), 1% L-glutamine, 1%
ampicillin/streptomycin and 0.1% gentamicin (SIGMA). Cells were kept at the constant
concentration of 200000 cells per milliliter of culture medium. HeLa cell line, was grown at 37°C in
air and 5% CO2 in Dulbecco’s Modified Eagle Medium (GIBCO) with 10% FBS, 1% L-glutamine,
1% ampicillin/streptomycin and 0.1% gentamicin.
Chemicals. ATRA and MS-275 were dissolved in ethanol; MC2392, MC2677 and MC2678 were
dissolved in DMSO and used at 1 µM and 5 µM respectively. SAHA (Merck), resveratrol, and EX-
527 (Alexis) were dissolved in DMSO and used at 5, 100, and 5 µM, respectively.
Synthesis of MC2392, MC2677, MC2678. See Supplementary Materials and Methods.
Total Protein, Histone extraction and western blot. See Supplementary Materials and Methods.
Colony assay. Colony assay was carried out as in (22).
Cell cycle and differentiation analyses. See Supplementary Materials and Methods.
Caspase-3, -8 and -9 assays. Caspase activity was detected within living cells using B-BRIDGE
Kits supplied with cell-permeable fluorescent substrates. The fluorescent substrates for caspase-3, -
8 and -9 were FAM-DEVD-FMK, FAM-LETD-FMK, FAM-LEHD-FMK, respectively. Cells were
washed twice in cold PBS and incubated for 1h in ice with the corresponding substrates, as
recommended by suppliers. Cells were analyzed using Cell Quest software applied to a
FACScalibur (BD). Experiments were performed in duplicate and values expressed in mean ± SD.
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
6
Caspase 2 assay. Caspase-2 activity was detected within whole living cells, using Caspase-2
Fluorometric Kit following supplier’s instructions (R&D). The plate was incubated at 37°C for 1h
and Fluorescence quantified with a TECAN M200 station at a 400-505 nm wavelength. Results are
expressed as fold increase in caspase activity.
Analysis of mitochondrial membrane potential. Detection of the changes in mitochondrial
membrane potential (ΔΨm) was performed using the MITOPTTM JC-1 kit (Immunochemistry
Technologies) following the supplier’s suggestions and analyzed by flow cytometry with Cell Quest
Pro software.
ROS detection. NB4 cells were resuspended in pre-warmed PBS 1X, containing 5 μM of DCF-DA
(diclorodihydrofluorescein diacetate) probe and incubated at 37°C for 20 minutes. Cells were
analyzed after washing, using the Cell Quest software applied to a FACScalibur (BD). H2O2 was
used as positive control.
RNA extraction, reverse transcription (RT) and quantitative PCR in real time (RT-PCR). See
Supplementary Materials and Methods.
Immunoprecipitation. See Supplementary Materials and Methods.
HDAC assay. HDAC assays have been performed as in (30). Specifically, samples
immunoprecipitated with anti-PML-RAR or with IgG were pooled respectively to homogenize all
samples. The Fluor de Lys Substrate was incubated (1h) with the immunoprecipitated PML-RAR
(in presence or absence of SAHA or MC2392 5 μM) and fluorescence was quantified with a
TECAN M200 station.
SIRT assay. See Supplementary Materials and Methods.
Transfections and luciferase assay. See Supplementary Materials and Methods.
IkB alpha ELISA assay. Assay was performed after 48h from the treatment with ATRA and
MC2392 in NB4 cells following supplier’s instructions (Active Motif).
ChIP and ChIP-seq. NB4 cells were treated for 4, 6, 24 and 48h with 1 μM ATRA, 5 μM MS-275
and 5 μM MC2392. Chromatin was harvested as described previously (31). ChIP experiments were
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
7
performed using the H3K9K14ac (Diagenode), Nrf2, c-Fos and c-Jun (Santa Cruz) antibodies.
ChIPed DNA was analyzed by real time qPCR with specific primers (Biolegio, see Supplementary
Methods) using the 2x SYBR Green mix (BIORAD) in a MyiQ thermocycler (BIORAD). Primers
amplifying myoglobin were used as negative control. In addition, ChIPed DNA was prepared for
sequencing and sequenced according to the manufacturer’s instructions (Illumina) in Supplementary
Materials and Methods and essentially as already described (32, 33).
Peak Detection, Clustering analysis and Motif Search are detailed in the Supplementary
Materials and Methods.
RNA-sequencing. See Supplementary Materials and Methods.
Results
MC2392 inhibits HDAC activity of the PML-RARα complex and maintains retinoid activity.
We set out to exploit the possibility to combine the active parts of the HDACi MS-275 and the
retinoid, ATRA, yielding MC2392 (Fig. 1A). To explore the activities of this hybrid compound,
U937 cells were treated for 24h with 5 μM MC2392, its parent compound MS-275 (Entinostat) a
class I inhibitor and the class I/II HDAC inhibitor SAHA (Vorinostat). In contrast to the reference
compounds, MC2392 did not result in a net increase of histone acetylation (Fig. 1B). Moreover
MC2392 did not induce p21 protein expression (Fig. 1B) corroborating that at this concentration the
molecule has reduced or no HDACi activity as compared to MS-275. MC2392 exerted only
minimal (if any) inhibitory action on HDAC1 and HDAC4, nor did it display SIRT1 modulating
activities in vitro (Fig. 1C). Our premise for the design of MC2392 was that the close physical
proximity of a retinoid (PML-RARα) and an HDACi binding moiety (HDAC1 and/or -4) could
convey context-specific inhibitory activity by a hybrid molecule spanning the physical distance. To
investigate this assertion, PML-RARα was immunoprecipitated from HeLa cells after transient
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
8
transfection with PML-RARα and assayed for HDACi activity. Indeed, MC2392 inhibits the
HDAC(s) resident in the PML-RARα repressive complex similarly to SAHA (Fig. 1D).
To investigate whether MC2392 acts as a retinoid, its potential to activate the ATRA-responsive
luciferase reporter (RARE3tkluc) was first tested in transient transfected HeLa that expresses
endogenous RAR and RXR. MC2392 activates RARE3tkluc weakly compared to ATRA (Fig. 1E).
Consistent with these observations, quantitative real-time PCR in NB4 cells showed a similar low
response to MC2392 for the RNA expression of specific genes known to be retinoid modulated (34,
35) such as HOXA1, IRF1 and, to a lesser extent, TNFα, (Supplementary Fig. 1A) as compared to
ATRA suggesting that MC2392 still maintains at least part of its retinoid activity. Finally, western
blot analyses of PML-RARα following MC2392 treatment showed rapid degradation of PML-
RARα in NB4 cells (Fig. 1F) as has been reported for ATRA (36, 37). Taken together the results
show that MC2392 acts as a weak retinoid when compared to ATRA implying that it binds to
RARα, activates RARα signaling as well as induces the degradation of PML-RARα (Fig. 1F). We
tentatively conclude that MC2392 is an HDACi in the context of the PML-RARα-HDAC repressive
complex.
MC2392 induces local H3 acetylation.
We showed that the HDACi activity of MC2392 is indeed apparent in the context of the PML-
RARα-contained HDAC complexes implying that acetylation maybe altered at PML-RARα
genomic binding sites. To examine whether MC2392 induces local epigenetic alterations, ChIP-seq
profiling was performed using the H3K9K14ac antibody in NB4 cells treated with ATRA, MS-275,
MC2392 or solvent, DMSO. In DMSO treated cells, H3K9K14 acetylation appeared in the typical
sharp narrow peaks. Treatment with the HDACi MS-275 had a dramatic effect on the genomic
acetylation landscape: H3K9K14ac peaks were dampened and spreading of acetylation occurred
(Fig. 2A). In line with the absence of global HDACi activity (Fig. 1B), MC2392 did not affect the
overall distribution of H3K9K14ac. Genomic annotation of the peaks called by MACS (38) with a
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
9
P-value of 10-6 (32) revealed a small percentage of H3K9K14ac binding sites at promoter regions
(8.74%) while the major portion of the peaks were located in the gene body (52.11%) or intergenic
(34.19%) (Fig. 2B). Importantly, MC2392 did induce low but reproducible local changes in
acetylation at some PML-RARα binding sites such as at RARβ promoter or TGM2, ICAM1 gene
(Fig. 2C). The RARβ is avidly transcriptionally induced by ATRA and expectedly H3 acetylation is
induced at promoter region. These data suggest specificity and selectivity of MC2392 towards at
least some PML-RARα binding sites and reinforced the similar, although weaker, retinoid activity
of MC2392 as compared to ATRA (Fig. 1E, Supplementary Fig. 1A).
To substantiate our findings we examined the 1000 most significantly changed acetylation peaks
after ATRA induction in PML-RARα/RXR binding regions. The intensity plots covering the region
of 10 kb up- and downstream for cells treated with ATRA and MC2392 show acetylation increases
following MC2392 treatment although to a lower level and after longer treatment times as
compared to ATRA (Fig. 2D): a vivid increase is already apparent after 4h of ATRA treatment
whereas a similar response amplitude is reach only after 48h of MC2392 treatment.
The normalized tag numbers (in treated and untreated samples) were counted for all binding regions
and clustered using k-means. This analysis identified 20 clusters (Supplementary Fig. 1B), amongst
which in particular cluster 2 and cluster 9 showed H3K9K14ac peaks that responded similarly to
ATRA for 4h and 24h, and to MC2392 for 24h and 48h though with a strong timely delay (Fig. 3A).
Moreover, Gene Ontology (GO) analysis revealed that the regulated genes in cluster 2 are
functionally linked to leukocyte differentiation (Supplementary Fig. 1C). The overlap between
H3K9K14ac peaks in cluster 2 and PML-RARα binding sites (33) revealed a common set of 191
loci suggesting that indeed a subset of H3K9K14ac peaks are associated with PML-RARα/RXR.
The overlap between cluster 9 H3K9K14ac peaks and PML-RARα binding sites revealed a
common set of 183 regions, suggesting that also here a subset of induced H3K9K14ac peaks are
associated with PML-RARα/RXR (Fig. 3B, top).
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
10
Recent studies analyzing the genome-wide PML-RARα binding sites not only identified DR2 and
DR5 elements as the primary PML-RARα response elements but also regions containing DR1, DR3
and DR4 motifs (33, 39). We found that both the DR1 and DR2 motifs are enriched within the
cluster 2 and 9 binding sites (Fig. 3B, middle) whereas DR3, DR4 and DR5 motifs are not enriched.
In addition, a de novo motif search identified the DR1 element in the 191 H3K9K14ac peaks of the
cluster 2 that are regulated by MC2392 (Fig. 3B, bottom). This suggest that a proper configuration,
i.e. spatial position of the PML-RARα with respect to the HDAC(s) in the repressive complex is an
important prerequisite for MC2392 to physically bridge the RAR and HDAC(s) moieties and
consequently cause inhibition of HDAC activity.
MC2392, but not ATRA, induces H3K9K14 acetylation at pro-apoptotic genes.
Our analysis also revealed cluster 8 consisting of 774 regulated genes that showed an opposite
epigenetic response to MC2392 as compared to ATRA treatment (Fig. 3C) that is induction by
MC2392 and reduction by ATRA. The assignment of genes in this cluster to biological functions
(Fig. 3C, bottom) revealed enrichment for regulation of cell death. De novo motif search showed
enrichment of the NF-E2-related factor-2 (NFE2L2, Nrf2) and Fos (AP1) response element each
present at about 10% of H3K9K14ac peaks (Fig. 3D). Note that the stress-responsive transcription
factors Nrf2 is reported to play a role in protection against oxidative stress-induced cellular damage.
ChIP experiments coupled with qPCR showed enrichment of Nrf2 over specific responsive element
(ARE) of target genes after MC2392 treatment, confirming an early Nrf2-mediated transcriptional
response during oxidant-induced cell death, differently to ATRA which reduces the binding of Nrf2
to the ARE enhancer region (Supplementary Fig. 1D). Finally, we found an increase in transcription in the target genes that encode antioxidant proteins or
enzymes to buffer the intracellular redox activities, such as FTH1, GCLM, GCLC, NQO and
HMOX, STK17A. Conversely, ATRA inhibits/reduces the expression of these Nrf2 target genes
(Supplementary Fig. 2A).
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
11
MC2392 and ATRA induce expression of apoptotic genes at transcriptional level.
To analyze the expression changes induced by MC2392, we performed transcriptome profiling on
MC2392 and ATRA-treated NB4 cells. The gene expression values from RNA-seq data were
calculated by quantifying the number of sequence tags for each genes using Genomatix (40, 41).
Scatter plots show the general differences of gene expression of MC2392 versus ATRA after 24 and
48h (Supplementary Fig. 2B). Moreover, functional annotation clustering of differentially expressed
genes by Gene Ontology (GO; PANTHER) analysis revealed that genes up-regulated in response to
MC2392 are enriched for terms associated with protein transport, response to stress and apoptosis
induction. Moreover ATRA up-regulated genes are enriched for terms associated mostly to protein
transport and metabolic processes (Supplementary Fig. 2C). The hierarchical clustering analysis
supports these data (Supplementary Fig. 2D).
Moreover, MC2392 down-regulated genes are enriched for terms associated with metabolic
processes and pathway analysis showed the decreased expression of components that drive cell
growth, proliferation and cell communication such as JAK/STAT and integrin pathway (data not
shown).
Analysis of the transcription level of the 148 genes associated with apoptosis
(http://www.reactome.org) revealed main differences between ATRA and MC2392 (Supplementary
Fig. 2E and Supplementary Materials) such as the fast induction of TRAF2, E2F1, DYNLL2 and
DBNL following MC2392 but not ATRA treatment. Next, we selected several MC2392-modulated genes related to apoptosis and oxidative stress, which are also modulated in the H3K914ac levels (Fig. 3 and Supplementary Fig. 1) upon MC2392 but not ATRA treatment and examined transcription factor binding. This revealed an increased occupancy of c-Jun, c-Fos and NRF2 in a time- and MC2392-dependent manner (Supplementary Fig. 3), corroborating and extending our observations that the apoptotic action of the MC2392 is causally bound to the modulation of cell death and redox pathways in response to modulation of AP1 and NRF2 bindings and Research.
on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
12
their downstream pathways. Taken together the transcriptome and histone H3 acetylome analysis
strongly point to a differential effect of MC2392 as opposed to ATRA on apoptotic genes.
MC2392 induces cell death in APL cells.
Genome-wide epigenetic and transcriptome analysis have provided strong indications that MC2392
effects are distinct from those elicited by ATRA in particular on apoptotic genes. Therefore, we
assessed the activity of MC2392 on NB4 proliferation, cell cycle progression, granulocytic
differentiation and apoptosis induction in comparison to MS-275, SAHA, and ATRA (Fig. 4).
Interestingly, analysis of NB4 clonogenicity in semisolid media upon MC2392 treatment (Fig. 4A,
left) revealed a clear loss of clonogenic activity at 10-5 M. In addition, two MC2392 analogues that
differ for the position of the amino group at the anilide ring (position 2, 3 and 4 in MC2392,
MC2678, and MC2677, respectively) (42) were used. The MC2678 and -2677 compounds do not
display HDAC inhibition (data not shown) and MC2677 has been previously described to inhibit
the growth of a number of tumor cell lines including prostate, head and neck, squamous carcinoma
and neuroblastoma (42, 43). In contrast to MC2392 and MS-275, MC2678 and MC2677 are unable
to induce a proliferation arrest of NB4 cells (Fig. 4A, right). MC2392 increased the pre-G1 fraction
(Fig. 4B), a measure of cell death induction but did not induce major changes in the cell cycle
phases (Fig. 4C). MC2392 did also induce an increase of the CD11c levels to 24%, thus inducing
maturation of APL cells similar to ATRA, MC2677 and MC2678 (Fig. 4D) but the MC2678 and
MC2677 compounds did not induce apoptosis (Fig. 4B).
Induction of cell death by MC2392 is not dependent on differentiation as MC2392 readily induces
cell death but not differentiation of the ATRA-resistant NB4 subclone, NB4R4 (26, 27) (Fig. 4E-F).
Note that as a result of the acquired mutation in the ligand-binding domain, PML-RARα can bind
ATRA but it is thought that the binding does not induce the conformational change needed to
dissociate the repressive complex and recruit activating proteins (26). Thus, the cell death activity
of MC2392 acts independently from maturation and is likely linked to the HDAC inhibition of the
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
13
resident repressive complexes. Taken together, our analysis shows that MC2392 indeed acts as a
weak retinoid in line with the transcriptome and histone H3 acetylome data but clearly has acquired
additional properties or such that were ‘masked’ in ATRA and MS-275. Moreover, our results
corroborate and extend the rapid response to the apoptosis/cell death versus slow differentiation in
response to MC2392 treatment.
MC2392-induced cell death involves ROS and caspase-8.
To gain insight into the mechanism(s) by which MC2392 initiates cell death, caspases-8, -9 and -3/7
activities in NB4 cells were analyzed. Similarly to SAHA, MC2392 notably induced caspase
activation with caspase-8, -9 (initiator caspases), caspase-3/7 (effector caspases) (Fig. 5A-B) and
mitochondrial potential activated in NB4 cells (Fig. 5C), thus indicating that the cell death induced
by MC2392 is caspase dependent. MS-275 induced cell death is caspase-8 dependent and related to
TRAIL induction (22). In order to identify the molecular pathway we evaluated the levels of known
players of apoptosis, such as FAS, TRAIL, FADD, caspase-8 and -3/7, Bcl2 and Bid in NB4 cells
after 24h and 48h treatment. As shown in Fig. 5D and Supplementary Fig. 3A, MC2392 induced the
extrinsic apoptotic pathway, increased the expression of FasL, Fas and TRAIL similarly to what has
been reported for MS-275 (22). MC2392 also induced degradation of Bid into the apoptotic factor t-
Bid, mitochondrial depolarization (Fig. 5D) and Bcl2 down-regulation (Supplementary Fig. 4B).
Next, NB4 cells were treated with MC2392 and H2O2 with and without the antioxidant N-
acetylcysteine (NAC). The data suggest that MC2392 mediates a ROS-dependent cell death
associated with the generation of reactive oxygen species (Fig. 5E and 5F). We further observed
increased H2AX phosphorylation which is a hallmark of ongoing DNA damage as a result of
increased reactive oxygen species (ROS) (44, 45). The response of APL cells to MC2392 includes
the generation of ROS as a consequence of Bid cleavage, which triggers mitochondrial membrane
permeabilization. As both caspase-2 and caspase-8 are capable of cleaving the pro-apoptotic Bid we
set out to investigate whether caspase-2 or caspase-8 are responsible. MC2392 induced caspase-2
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
14
activation, which is fully blocked by the caspase-2 inhibitor Z-VDVAD-FMK (Fig. 6A). However,
cell death mediated by MC2392 stayed unaltered (Fig. 6A, right). MC2392-mediated cell death is
also not NOX-dependent, given that Apocynin - a NOX specific inhibitor - is not able to block
apoptosis (Supplementary Fig. 3C). Furthermore, NF-kB is neither induced nor transcriptionally
activated by MC2392 (Supplementary Fig. 4D-4E). In contrast, when the pan-caspase inhibitor
ZVAD (data not shown) or the caspase-8 inhibitor IETD are used together with MC2392, NB4 cell
death is fully abolished (Fig. 6B), strongly suggesting that the MC2392 mechanism of cell death is
caspase 8-dependent. Surprisingly, necrostatin-1 (NEC-1), the known RIP1 kinase and necroptosis
inhibitor, completely abrogated MC2392 cell death (Fig. 6C), thus suggesting a RIP1 mediated cell
death. Consistent with these observations, analysis for RIPK1 expression showed both a strong up-
regulation upon MC2392 and a full abrogation with NEC-1 (Fig. 6D). Finally, in line with our
hypothesis, MC2392 induces cell death specifically in a PML-RARα but not in RARα or PML
driven way (Fig. 4F and data not shown).
Taken together our data revealed that at the level of acetylation, MC2392 induces much fewer
changes as compared to ATRA. Moreover, MS-275 resulted in genome wide dampening and
broadening of all acetylation sites (Fig. 2A) whereas MC2392 affected acetylation only at a subset
of PML-RARα sites (DR1-2 type). Finally, transcriptome analysis corroborated and extended that
MC2392 affects apoptotic genes.
Discussion
Epigenetic modifications can be reversed by epi-drugs that are promising for antitumor therapy. In
patients with leukemia, HDACi are widely used in combination with ATRA resulting in improved
anti-tumorigenic activity. Here we identified and characterized the mechanism of action of
MC2392, a compound merging the HDACi part of the well-known MS-275 to ATRA. When
compared to both ATRA and MS-275, MC2392 displays similar as well as divergent
characteristics. Despite being inactive as an HDACi standard in vitro biochemical assays, MC2392
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
15
specifically inhibits the HDACs contained in the PML-RARα repressive complex (Fig. 1D).
MC2392 induces degradation of PML-RARα (Fig. 1E) and differentiation of NB4 cells (Fig. 4D).
Genome-wide epigenetic analysis shows that MC2392 induces increased acetylation at a small
subset of PML-RARα sites with preponderance for DR1 and -2 type (Fig. 3B) distinct from
acetylation at PML-RARα sites induced by ATRA. This MC2392 characteristic is particularly
interesting and classifies it as a novel class of drugs that need precise protein-protein ‘positioning’
to function. Moreover a specific configuration, i.e. spatial position of the PML-RARα with respect
to the HDAC(s) in the repressive complex is likely an important prerequisite for MC2392 to bridge
the RARα and HDAC(s) moieties and consequently inhibition of HDAC activity.
The fact that MC2392 may represent the first RARα-driven HDACi is corroborated by i) its
inactivity on human recombinant HDACs in vitro, ii) its inability to modify general histone and/or
tubulin acetylation; iii) its selective HDACi action on PML-RARα complexes; iv) its enhanced
ability to induce apoptosis in APL cells but not in other leukemia or solid cancer models. The fact
that MC2392 induces apoptosis in NB4 cells and increases caspases-8, -2 and -9 activity suggests
that MC2392 activate both mitochondrial and death receptor apoptotic pathways. Although
MC2392 is able to activate the extrinsic pathway by inducing FAS and TRAIL, the death response
to MC2392 includes also the generation of ROS as a possible consequence of the cleavage of Bid
(Fig. 5D, F). Indeed, the cleavage of Bid in APL cells occurs concurrently with early generation of
ROS, thereby contributing to cell death by promoting mitochondrial release of cytochrome c. For
this reason, normally the death receptor pathway (extrinsic pathway) converges with the
mitochondrial pathway (intrinsic pathway) through caspase-mediated cleavage and activation of
Bid. Furthermore, both the induction of RIP1 by MC2392 and the abrogation of cell death operated
by NEC-1, an inhibitor of RIP1, dictate a complex scenario in which, alternatively to apoptosis,
necroptotic pathways may lead to mitochondrial depolarization and cell death of NB4 APL cells. Of
note, both ATRA and MS-275 have been previously shown to function via TRAIL and FAS
induction (16, 22). Moreover, the fact that the MC2392 caspase 8-dependent apoptosis can be
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
16
abrogated by NEC-1 supports a model in which mixed types of cell death can occur, displaying
shared features of both necroptosis and apoptosis. Clearly, MC2392-induced apoptosis occurs in
response to oxidative stress and is orchestrated by stress-responsive transcription factors. Nuclear
translocation of NRf2 and subsequent induction of its target genes via antioxidant stress response
element (ARE) may function to buffer oxidative stress response. Moreover, MC2392 induces the
modulation of a large number of stress-responsive genes (Supplementary Fig. 2A) and apoptotic
genes (see Supplementary Materials and Methods). Finally, the evidence that apoptosis occurs in
APL independently from an ‘classical’ ATRA response is suggested both by the MC2392 action in
Zn-inducible PLZF-RARα expression in U937 cells (28, 29) as well as by its activity on NB4R4
cells (26, 27). Note that in both PLZF-RARα U937 cells and in NB4R4 cells ATRA signaling stays
impaired upon treatment (27, 46, 47). Excitingly, given that MC2392 is not an HDACi in vitro,
MC2392 is able to exert its HDAC inhibition only when driven by its ATRA moiety to PML-RARα
and thereby co-localized to HDAC-containing repressive complexes (Fig. 4). In this way, MC2392
relieves the HDAC repressive complex and allows induction of cell death in ATRA responsive and
in bona fide APL ATRA-resistant cells. Accordingly, MC2392 is not inducing apoptosis in solid
tumor such as prostate or breast cancer (data not shown). Moreover, ChIP-seq profiles show that
MC2392 induces changes in H3 acetylation at a small subset of PML-RARα/RXR binding regions
but also in regions not regulated by ATRA (Fig. 2, 3). Unlike natural retinoids, MC2392 exerts
distinct biological effects, inducing partial differentiation and promoting apoptosis, possibly
similarly to As2O3. MC2392 displays a minimal or no cytotoxicity in other kind of tumor cells.
Collectively these data underpin the specificity and selectivity of the MC2392. In conclusion,
MC2392 represents the creation of a single chemical entity with dual action thus reflecting the
concept of a laboratory-created ad hoc drug, which as a single drug can overcome resistance in
cancer.
Acknowledgements
We apologize to the authors whose work we could not cite due to reference restriction.
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
17
References
1. Grignani F, Ferrucci PF, Testa U, Talamo G, Fagioli M, Alcalay M, et al. The acute promyelocytic leukemia-specific PML-RAR alpha fusion protein inhibits differentiation and promotes survival of myeloid precursor cells. Cell. 1993;74:423-31. 2. Kakizuka A, Miller WH, Jr., Umesono K, Warrell RP, Jr., Frankel SR, Murty VV, et al. Chromosomal translocation t(15;17) in human acute promyelocytic leukemia fuses RAR alpha with a novel putative transcription factor, PML. Cell. 1991;66:663-74. 3. Qiu JJ, Lu X, Zeisig BB, Ma Z, Cai X, Chen S, et al. Leukemic transformation by the APL fusion protein PRKAR1A-RAR{alpha} critically depends on recruitment of RXR{alpha}. Blood. 2010;115:643-52. 4. Chen Z, Guidez F, Rousselot P, Agadir A, Chen SJ, Wang ZY, et al. PLZF-RAR alpha fusion proteins generated from the variant t(11;17)(q23;q21) translocation in acute promyelocytic leukemia inhibit ligand-dependent transactivation of wild-type retinoic acid receptors. Proc Natl Acad Sci U S A. 1994;91:1178-82. 5. Chen SJ, Zelent A, Tong JH, Yu HQ, Wang ZY, Derre J, et al. Rearrangements of the retinoic acid receptor alpha and promyelocytic leukemia zinc finger genes resulting from t(11;17)(q23;q21) in a patient with acute promyelocytic leukemia. J Clin Invest. 1993;91:2260-7. 6. Lin RJ, Nagy L, Inoue S, Shao W, Miller WH, Jr., Evans RM. Role of the histone deacetylase complex in acute promyelocytic leukaemia. Nature. 1998;391:811-4. 7. Marks PA, Rifkind RA, Richon VM, Breslow R. Inhibitors of histone deacetylase are potentially effective anticancer agents. Clin Cancer Res. 2001;7:759-60. 8. Sun SY, Wan H, Yue P, Hong WK, Lotan R. Evidence that retinoic acid receptor beta induction by retinoids is important for tumor cell growth inhibition. J Biol Chem. 2000;275:17149-53. 9. Altucci L, Rossin A, Raffelsberger W, Reitmair A, Chomienne C, Gronemeyer H. Retinoic acid-induced apoptosis in leukemia cells is mediated by paracrine action of tumor-selective death ligand TRAIL. Nat Med. 2001;7:680-6. 10. Altucci L, Gronemeyer H. The promise of retinoids to fight against cancer. Nat Rev Cancer. 2001;1:181-93. 11. Minucci S, Pelicci PG. Histone deacetylase inhibitors and the promise of epigenetic (and more) treatments for cancer. Nat Rev Cancer. 2006;6:38-51. 12. Insinga A, Monestiroli S, Ronzoni S, Carbone R, Pearson M, Pruneri G, et al. Impairment of p53 acetylation, stability and function by an oncogenic transcription factor. The EMBO journal. 2004;23:1144-54. 13. Mai A, Altucci L. Epi-drugs to fight cancer: from chemistry to cancer treatment, the road ahead. Int J Biochem Cell Biol. 2009;41:199-213. 14. Johnstone RW, Ruefli AA, Lowe SW. Apoptosis: a link between cancer genetics and chemotherapy. Cell. 2002;108:153-64. 15. Marks P, Rifkind RA, Richon VM, Breslow R, Miller T, Kelly WK. Histone deacetylases and cancer: causes and therapies. Nat Rev Cancer. 2001;1:194-202. 16. Insinga A, Monestiroli S, Ronzoni S, Gelmetti V, Marchesi F, Viale A, et al. Inhibitors of histone deacetylases induce tumor-selective apoptosis through activation of the death receptor pathway. Nat Med. 2005;11:71-6.
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
18
17. Liu S, Cheng H, Kwan W, Lubieniecka JM, Nielsen TO. Histone deacetylase inhibitors induce growth arrest, apoptosis, and differentiation in clear cell sarcoma models. Mol Cancer Ther. 2008;7:1751-61. 18. Yoshida M, Furumai R, Nishiyama M, Komatsu Y, Nishino N, Horinouchi S. Histone deacetylase as a new target for cancer chemotherapy. Cancer Chemother Pharmacol. 2001;48 Suppl 1:S20-6. 19. Bolden JE, Peart MJ, Johnstone RW. Anticancer activities of histone deacetylase inhibitors. Nat Rev Drug Discov. 2006;5:769-84. 20. Hu E, Dul E, Sung CM, Chen Z, Kirkpatrick R, Zhang GF, et al. Identification of novel isoform-selective inhibitors within class I histone deacetylases. The Journal of pharmacology and experimental therapeutics. 2003;307:720-8. 21. Saito A, Yamashita T, Mariko Y, Nosaka Y, Tsuchiya K, Ando T, et al. A synthetic inhibitor of histone deacetylase, MS-27-275, with marked in vivo antitumor activity against human tumors. Proc Natl Acad Sci U S A. 1999;96:4592-7. 22. Nebbioso A, Clarke N, Voltz E, Germain E, Ambrosino C, Bontempo P, et al. Tumor-selective action of HDAC inhibitors involves TRAIL induction in acute myeloid leukemia cells. Nat Med. 2005;11:77-84. 23. Qiu L, Burgess A, Fairlie DP, Leonard H, Parsons PG, Gabrielli BG. Histone deacetylase inhibitors trigger a G2 checkpoint in normal cells that is defective in tumor cells. Mol Biol Cell. 2000;11:2069-83. 24. Cimino G, Lo-Coco F, Fenu S, Travaglini L, Finolezzi E, Mancini M, et al. Sequential valproic acid/all-trans retinoic acid treatment reprograms differentiation in refractory and high-risk acute myeloid leukemia. Cancer Res. 2006;66:8903-11. 25. Wagner JM, Hackanson B, Lubbert M, Jung M. Histone deacetylase (HDAC) inhibitors in recent clinical trials for cancer therapy. Clinical epigenetics. 2010;1:117-36. 26. Shao W, Benedetti L, Lamph WW, Nervi C, Miller WH, Jr. A retinoid-resistant acute promyelocytic leukemia subclone expresses a dominant negative PML-RAR alpha mutation. Blood. 1997;89:4282-9. 27. Rosenauer A, Raelson JV, Nervi C, Eydoux P, DeBlasio A, Miller WH, Jr. Alterations in expression, binding to ligand and DNA, and transcriptional activity of rearranged and wild-type retinoid receptors in retinoid-resistant acute promyelocytic leukemia cell lines. Blood. 1996;88:2671-82. 28. Ruthardt M, Orleth A, Tomassoni L, Puccetti E, Riganelli D, Alcalay M, et al. The acute promyelocytic leukaemia specific PML and PLZF proteins localize to adjacent and functionally distinct nuclear bodies. Oncogene. 1998;16:1945-53. 29. Grignani F, De Matteis S, Nervi C, Tomassoni L, Gelmetti V, Cioce M, et al. Fusion proteins of the retinoic acid receptor-alpha recruit histone deacetylase in promyelocytic leukaemia. Nature. 1998;391:815-8. 30. Nebbioso A, Manzo F, Miceli M, Conte M, Manente L, Baldi A, et al. Selective class II HDAC inhibitors impair myogenesis by modulating the stability and activity of HDAC-MEF2 complexes. EMBO Rep. 2009;10:776-82. 31. Denissov S, van Driel M, Voit R, Hekkelman M, Hulsen T, Hernandez N, et al. Identification of novel functional TBP-binding sites and general factor repertoires. The EMBO journal. 2007;26:944-54. 32. Nielsen R, Pedersen TA, Hagenbeek D, Moulos P, Siersbaek R, Megens E, et al. Genome-wide profiling of PPARgamma:RXR and RNA polymerase II occupancy reveals temporal activation of distinct metabolic pathways and changes in RXR dimer composition during adipogenesis. Genes & development. 2008;22:2953-67.
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
19
33. Martens JH, Brinkman AB, Simmer F, Francoijs KJ, Nebbioso A, Ferrara F, et al. PML-RARalpha/RXR Alters the Epigenetic Landscape in Acute Promyelocytic Leukemia. Cancer cell. 2010;17:173-85. 34. Matikainen S, Ronni T, Hurme M, Pine R, Julkunen I. Retinoic acid activates interferon regulatory factor-1 gene expression in myeloid cells. Blood. 1996;88:114-23. 35. Nouzova M, Holtan N, Oshiro MM, Isett RB, Munoz-Rodriguez JL, List AF, et al. Epigenomic changes during leukemia cell differentiation: analysis of histone acetylation and cytosine methylation using CpG island microarrays. The Journal of pharmacology and experimental therapeutics. 2004;311:968-81. 36. Jing Y, Xia L, Lu M, Waxman S. The cleavage product deltaPML-RARalpha contributes to all-trans retinoic acid-mediated differentiation in acute promyelocytic leukemia cells. Oncogene. 2003;22:4083-91. 37. Saeed S, Logie C, Stunnenberg HG, Martens JH. Genome-wide functions of PML-RARalpha in acute promyelocytic leukaemia. British journal of cancer. 2011;104:554-8. 38. Zhang Y, Liu T, Meyer CA, Eeckhoute J, Johnson DS, Bernstein BE, et al. Model-based analysis of ChIP-Seq (MACS). Genome biology. 2008;9:R137. 39. Kamashev D, Vitoux D, De The H. PML-RARA-RXR oligomers mediate retinoid and rexinoid/cAMP cross-talk in acute promyelocytic leukemia cell differentiation. J Exp Med. 2004;199:1163-74. 40. Audic S, Claverie JM. The significance of digital gene expression profiles. Genome research. 1997;7:986-95. 41. Mortazavi A, Williams BA, McCue K, Schaeffer L, Wold B. Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nature methods. 2008;5:621-8. 42. Herman D, Jenssen K, Burnett R, Soragni E, Perlman SL, Gottesfeld JM. Histone deacetylase inhibitors reverse gene silencing in Friedreich's ataxia. Nature chemical biology. 2006;2:551-8. 43. Mershon SM, Anding AL, Chapman JS, Clagett-Dame M, Stonerock LA, Curley RW, Jr. Solid phase-assisted synthesis and screening of a small library of N-(4-hydroxyphenyl)retinamide (4-HPR) analogs. Bioorg Med Chem Lett. 2007;17:836-40. 44. Park JH, Kim EJ, Jang HY, Shim H, Lee KK, Jo HJ, et al. Combination treatment with arsenic trioxide and sulindac enhances apoptotic cell death in lung cancer cells via activation of oxidative stress and mitogen-activated protein kinases. Oncology reports. 2008;20:379-84. 45. Zhang HY, Hormi-Carver K, Zhang X, Spechler SJ, Souza RF. In benign Barrett's epithelial cells, acid exposure generates reactive oxygen species that cause DNA double-strand breaks. Cancer Res. 2009;69:9083-9. 46. So CW, Dong S, So CK, Cheng GX, Huang QH, Chen SJ, et al. The impact of differential binding of wild-type RARalpha, PML-, PLZF- and NPM-RARalpha fusion proteins towards transcriptional co-activator, RIP-140, on retinoic acid responses in acute promyelocytic leukemia. Leukemia. 2000;14:77-83. 47. Licht JD, Shaknovich R, English MA, Melnick A, Li JY, Reddy JC, et al. Reduced and altered DNA-binding and transcriptional properties of the PLZF-retinoic acid receptor-alpha chimera generated in t(11;17)-associated acute promyelocytic leukemia. Oncogene. 1996;12:323-36. Figure legends
Figure 1. MC2392 selectively inhibits the HDAC-repressive complexes associated to PML-RARα
in NB4 cells. A, structure of MS-275, ATRA and MC2392. B, acetylation evaluation of histone H3
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
20
by SAHA and MC2392 at 5 μM; p21 induction by MS-275 and MC2392 at 5 μM in U937 cells.
Histone H1 or ERKs were used for loading control. C, HDAC1-4 and SIRT1 assay in presence of
the indicated compounds. SAHA and EX-527 were used at 5 μM, Resveratrol at 100 μM. Data are
expressed as % of activity compared to control. Error bars represent standard deviations (SD) based
on three independent experiments. D, HDAC assay after immunoprecipitation of PML-RARα in
HeLa cells overexpressing PML-RARα. Data are expressed as % of activity compared to control,
error bars represent SD based on three independent experiments. E, activation of RARE3tkluc after
transfection in HeLa cells treated as indicated. pMAX-EGFP co-transfection was used as an internal
control, error bars represent ST based on three independent experiments. F, Western blot analyses
for PML-RARα in nuclei extracts in NB4 cells treated with ATRA 1 μM and MC2392 5 μM.
HDAC2 was used as loading control.
Figure 2. MC2392 selectively regulates acetylation at some PML-RARα binding sites in NB4 cells.
A, genomic tracks of random regions displaying ChIP-seq data for H3K9K14ac in NB4 cells after
24h of treatment with MS-275, ATRA and MC2392. B, distribution of the H3K9K14ac enrichment.
Locations are divided in gene body (gene), upstream near (-5 kb to the transcription start site),
intergenic (upstream far from -25 to -5 kb of the transcription start site, downstream far, end of
untranslated region +5 to +25 kb, and distant (everything else)), downstream near (end of
untranslated region to +5 kb). C, overview of the TGM2, ICAM1, RARβ, PML-RARα/RXR binding
site in NB4 cells. D, tag density maps depicting the pattern of H3K9K14ac occupancy around (+ 10
and –10 kb) the PML-RARα/RXR sites. The intensity plot indicates the level of H3K9K14ac
occupancy (log2 of tag density; see scale on the top) in a 500-bp window. The position of one
example (TGM2) is indicated.
Figure 3. A proper PML-RARα spatial position is a prerequisite for MC2392 mediated HDAC
inhibition at PML-RARα binding sites in NB4 cells. A, k-means clustering using a Pearson
correlation metric and the log2 value of the tag density. Cluster 2 and cluster 9 of H3K9K14ac
changes with typical and similar epigenetic responses to ATRA and to MC2392. The number of
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
21
genes in each cluster is indicated. B, Venn diagram representing the overlap of cluster 2 or 9
regulated genes and PML-RARα/RXR binding sites; analysis of the presence of DR motifs
underlying the overlapped genes; response element for PPAR-RXRα identified as the predominant
motif in the 191 genes of the cluster 2. Motif was identified by a de novo motif search. C, Cluster 8
of H3K9K14ac changes with different epigenetic responses to ATRA and to MC2392. The number
of genes is indicated. Functional annotation clustering (GO) of the clustered genes. D, Response
element for NFE2L2, FOS, AP1 identified like predominant motifs in the H3K9K14ac binding sites
of the regulated genes of Cluster 8. Motifs were identified by a de novo motif search.
Figure 4. MC2392 induces PML-RARα-context specific apoptosis in APL cells. A, clonogenity
assay in NB4 cells upon MC2392; proliferation curves in NB4 cells at the indicated time. B, pre-G1
fraction analysis of NB4 cells treated for 48h with the indicated compounds. C, cell cycle analysis
of NB4 cells treated for 48h with the indicated compounds. MS-275, SAHA, MC2677 and MC2678
were used at 5 μM; ATRA was used at 1 μM. D, granulocytic differentiation in NB4 by the CD11c
positive cells. PI positive cells have been excluded from the analysis. E, granulocytic differentiation
by the CD11c positive cells in NB4R4 cells. PI positive cells, have been excluded from the analysis;
pre-G1 fraction in NB4R4 cells treated with the indicated compounds. F, analysis of the pre-G1
fraction in U937 PML-RARα and PLZF-RARα cells after 24h and 48h of treatment with the
indicated compounds in presence or absence of zinc. All experiments have been performed at least
three times and error bars represent SD based on three independent experiments.
Figure 5. MC2392 leads to caspase activation and ROS production in NB4 cells. A, B, Caspase-8, -
9 and -3/7 assay in NB4 cells treated 48h with the indicated compounds. C, mitochondrial
membrane potential dissipation of NB4 treated with MS-275, SAHA and MC2392 used at 5 μM. D,
Western blot analysis of apoptosis players in NB4 cells treated with ATRA and MC2392 for 24h
and 48h. ERKs and tubulin have been used as control for equal protein loading. E, pre-G1 fraction
in NB4 cells treated with the indicated compounds. F, analysis of ROS production in conditions as
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
22
in A. NAC was used 20 mM and H2O2 500 μM; Western blot analysis for phosphoH2AX(S139).
Error bars shown in A, B, C, E, F represent SD based on three independent experiments.
Figure 6. MC2392-mediated apoptosis is caspase 8 dependent. A, Caspase-2 activity on NB4 cells
treated for 48h with the indicated compounds; pre-G1 pick detection in identical settings; the
caspase-2 inhibitor VDVAD was used at 100 μM. B, Caspase-8 activity on NB4 cells treated for
48h with the indicated compounds; pre-G1 pick detection in identical settings. The caspase-8
inhibitor IETD was used at 100 μM. C, pre-G1 fraction in NB4 cells treated with the indicated
compounds; NEC-1 was used at 50 μM, D, Western blot analysis on RIP1 after 24h and 48h of
treatment with the indicated compounds. Blot against ERKs represents control for equal protein
loading. Error bars shown in A, B, C, represent SD based on three independent experiments.
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
% H
DA
C a
ctivity a
sso
cia
ted
with
PM
L-R
AR
α
De Bellis et al., Figure 1
A
D
MC
23
92
24
h
MC
23
92
48
h
AT
RA
24
h
Ctr
PML-RARα
HDAC2
F
0
10
20
30
40
50
60
70
80
RL
U
MC
23
92
AT
RA
Ctr
E
MS-275 ATRA MC2392
AcH3
H1
p21
ERKs
Ctr
SA
HA
MC
23
92
24h 48h
MC
23
92
MC
23
92
MS
-27
5
Ctr
% a
ctivity
B C
HDAC1
% a
ctivity
Ctr
Re
s
Ex-5
27
MC
23
92
0
100
150
50
MS
-27
5
0
20
40
60
80
100
HDAC4
MC
23
92
SA
HA
Ctr
MS
-27
5
SIRT1
0
20
40
60
80
100
MC
23
92
SA
HA
Ctr
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
De Bellis et al., Figure 2
A
Scalechr3:
CpGIslands
45300000 45400000 45500000 45600000 45700000 45800000 45900000 46000000 46100000 46200000
CpG Islands (Islands < 300 Bases are Light Green)
UCSC Genes Based on RefSeq, UniProt, GenBank, CCDS and Comparative GenomicsCDCP1CDCP1
TMEM158 LARS2LARS2
LIMD1
AK026206KIAA0851
SACM1LSACM1L
SLC6A20
SLC6A20
LZTFL1LZTFL1
CCR9CCR9
CCR9CCR9 CXCR6
AX747832XCR1XCR1XCR1
CCR3CCR1CCR1
CCR3CCR3
CCR3
DMSO35 _
1 _
ATRA36 _
1 _
MS_27511 _
1 _
MC239246 _
1 _
H3
K9
K1
4a
c
C D
ATRA
MC2392
DMSO
ICAM1RARβ
ATRA
MC2392
DMSO
H3
K9
K1
4a
cH
3K
9K
14
ac
1_ _
_ _
TGM2
chr20:34
34
1
34
1
B
gene body
upstream near
dowstream near
4.94 %
8.74 %
52.11 %intergenic
34.19%
chr3:42 _
1 _42 _
1 _42 _
1 _
chr19:50 _
1 _50 _
1 _50 _
1 _
ctr
AT
RA
_4
h
AT
RA
_2
4h
MC
23
92
_2
4h
MC
23
92
_4
8h
TGM2
0 22
20kb
H3K9K14ac occupancy in PML-RARα/RXR binding sites
MC
23
92
_4
h
tag
de
nsity
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
De Bellis et al., Figure 4
MC
23
92
48
h
MC
23
92
24
h
Ctr
PML-RARα
CBA
Ctr
MC
23
92
MS
% c
ells
in
pre
-G1
0
10
20
30
40
50
% C
D11
c+
/PI-
ce
lls
0
10
20
30
40
50
60
MC
23
92
AT
RA
MS
Ctr
0
5
10
15
20
25
30
(time 48h)
0
10
20
30
40
50
60
70
80
PLZF-RARα
+ Zn
0
10
20
30
40
50
PML-RARα
+ ZnLnCap
% c
ells
in
pre
-G1
MC
23
92
48
h
MC
23
92
24
h
Ctr
MC
23
92
48
h
MC
23
92
24
h
Ctr
% c
ells
in
pre
-G1
0
10
20
30
40
50
(time 72h)
0
1
2
3
4
5
6
7
8MC2678
MC2677
MC2392
ATRA
MS-275
Ctr
240time (h)
MC
23
92
SA
HA
MS
-27
5
Ctr(time 48h)
0
10
20
30
40
50
60
70
80 G2/M
S
G1
MC
23
92
AT
RA Ctr(time 48h)
n°
ce
lls (
* 1
0^5
)
% c
ells
% c
ells
in
pre
-G1
D FE
MC
26
77
MC
26
78
0
5
10
15
20
25
30
35
MC
26
77
MC
26
78
AT
RA48
(time 48h)
MC
23
92
Ctr
AT
RA
0
10
20
30
40
50
MC
2 6
78
MC
26
77
200
150
100
50
0
nu
mb
er
of co
lon
ies
un
tre
at
10
^(-
6)M
10
^(-
5)M
SA
HA
MS
-27
5
% C
D11
c+
/PI-
ce
lls
NB4
NB4R4
MC
23
92
Ctr
AT
RA
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
A B C
D E
F
caspase 8 caspase 9
% a
ctivity
0
10
20
30
40
50
60
70
80
MC
26
78
MC
26
77
MC
23
92
AT
RA
SA
HA
Ctr
0
10
20
30
40
50
60
% c
asp
ase
3/7
activity
MC
23
92
AT
RA
SA
HA
Ctr
MC
26
77
MC
26
78
0
5
10
15
20
25
30
%M
TP
dis
sip
atio
n
SA
HA
MS
-27
5
Ctr
MC
23
92
(time 48h)
% c
ells
in
pre
-G1
0
5
10
15
20
25
MC
23
92
+N
AC
H2
O2
+N
AC
H2
O2
NA
C
Ctr
MC
23
92
Ctr
AT
RA
24
h
AT
RA
48
h
MC
23
92
48
h
MC
23
92
24
h
pH2AX
H4
0
20
40
60
80
100
MC
23
92
+N
AC
MC
23
92
H2
O2
+N
AC
NA
C
H2
O2
Ctr(time 48h)
% o
xid
ate
d R
OS
De Bellis et al., Figure 5
casp3
ERKs
Bid
tubtBid
48 24 48 24time (h)MC2392 ATRA Ctr
Fas
ERKs
FasL
TRAILERKs
casp8
ERKs
FADD
ERKs
48 24 48 24
MC2392 ATRA Ctr
time (h)
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
A B
De Bellis et al., Figure 6
0
5
10
15
20
25
% c
ells
in
pre
G1
ne
c-1
+ M
C2
39
2
ne
c-1
MC
23
92
Ctr
C
0,0
0,5
1,0
1,5
2,0
2,5
3,0ca
sp
ase
2 a
ctiva
tio
n (
fold
)
VD
VA
D+
MC
23
92
MC
23
92
SA
HA
Ctr
Ctr(time 48h)
Z-V
DV
AD
MC
23
92
Z-V
DV
AD
VD
VA
D+
MC
23
92
% c
ells
in
pre
-G1
0
10
20
30
40
ca
sp
ase
8 a
ctivita
tio
n (
fold
)
MC
23
92
Ctr(time 48h) D
TEI
IET
D+
MC
23
92
0
1
2
3
0
10
20
30
% c
ells
in
pre
-G1
D
MC
23
92
Ctr D
TEI
IET
D+
MC
23
92
RIPK1ERK1
Ctr MSATRAMC2392
48 24
RIPK1ERK1
Ctr
(time h)
M
C2
39
2
MC
23
92
+n
ec-1
48 24 48 24
(time 48h)
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568
Published OnlineFirst February 24, 2014.Cancer Res Floriana De Bellis, Vincenzo Carafa, Mariarosaria Conte, et al. triggered by the novel hybrid retinoid-HDAC inhibitor MC2392Context-selective death of acute myeloid leukemia cells
Updated version
10.1158/0008-5472.CAN-13-2568doi:
Access the most recent version of this article at:
Material
Supplementary
http://cancerres.aacrjournals.org/content/suppl/2014/02/25/0008-5472.CAN-13-2568.DC1
Access the most recent supplemental material at:
Manuscript
Authoredited. Author manuscripts have been peer reviewed and accepted for publication but have not yet been
E-mail alerts related to this article or journal.Sign up to receive free email-alerts
Subscriptions
Reprints and
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Permissions
Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)
.http://cancerres.aacrjournals.org/content/early/2014/02/22/0008-5472.CAN-13-2568To request permission to re-use all or part of this article, use this link
Research. on September 15, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on February 24, 2014; DOI: 10.1158/0008-5472.CAN-13-2568