inhibiting myc binding to the e-box dna motif by me47 decreases …shinjumi/publications/pdfs/31...

SHORT COMMUNICATION

Inhibiting MYC binding to the E-box DNA motif by ME47decreases tumour xenograft growthLC Lustig1, D Dingar1, WB Tu1,2, C Lourenco1,2, M Kalkat1,2, I Inamoto3, R Ponzielli1, WCW Chan4, JA Shin3 and LZ Penn1,2

Developing therapeutics to effectively inhibit the MYC oncoprotein would mark a key advance towards cancer patient care as MYCis deregulated in over 50% of human cancers. MYC deregulation is correlated with aggressive disease and poor patient outcome.Despite strong evidence in mouse models that inhibiting MYC would significantly impact tumour cell growth and patient survival,traditional approaches have not yet yielded the urgently needed therapeutic agents that directly target MYC. MYC functionsthrough its interaction with MAX to regulate gene transcription by binding to E-box DNA response elements of MYC target genes.Here we used a structure-based strategy to design ME47, a small minimalist hybrid protein (MHP) able to disrupt the MAX:E-boxinteraction/binding and block transcriptional MYC activity. We show that inducing ME47 expression in established tumourxenografts inhibits tumour growth and decreases cellular proliferation. Mechanistically, we show by chromatinimmunoprecipitation that ME47 binds to E-box binding sites of MYC target genes. Moreover, ME47 occupancy decreases MYC:DNAinteraction at its cognate E-box binding sites. Taken together, ME47 is a prototypic MHP inhibitor that antagonizes tumour cellgrowth in vitro and in vivo and inhibits the interaction of MYC with DNA E-box elements. These results support ME47’s role as a MYCinhibitor and suggest that MHPs provide an alternative therapeutic targeting system that can be used to target transcription factorsimportant in human diseases, including cancer.

Oncogene advance online publication, 14 August 2017; doi:10.1038/onc.2017.275

INTRODUCTIONThe MYC oncoprotein is estimated to be involved in the etiologyof over 50% of human cancers and is associated with aggressivedisease and poor patient outcome.1–4 MYC is considered a hubprotein as it integrates multiple signalling pathways to function asa master regulator with transcriptional control over biologicalprocesses, such as cellular proliferation and transformation.Mechanistically, MYC interacts with MAX and activates genetranscription by binding to E-box DNA elements, such as thecanonical sequence CACGTG, in the regulatory regions of targetgenes.5 MYC expression and activity are highly regulated innormal cells; however, MYC is often deregulated and constitutivelyactive in tumour cells.2,6 Studies show that when MYC activity isimpaired systemically in cells of tumour-bearing transgenic mice,tumour cell proliferation is blocked, whereas normal cells remainfunctionally intact.7–9 These results provide proof-of-concept thata therapeutic index can be achieved by targeting MYC despite itswidespread expression. Evidence shows that tumour cells aredependent on the MYC oncogene, which provides a vulnerabilityfor us to target.10

Therapeutically restraining MYC activity would significantlyimpact cancer patient care, yet targeting MYC directly appearsto be an intractable challenge due to the lack of structural featurestraditionally associated with the design of small-moleculeinhibitors (SMIs). New strategies to inhibit MYC are thereforerequired and are presently under development. SMIs blockingtranscriptional activators of the MYC gene, such as the

bromodomain and extra-terminal proteins,11–13 can lead todownregulation of MYC transcription in addition to other MYC-independent effects.14 SMIs have garnered much excitement withsome having been fast-tracked to phase I clinical trials.12

Unfortunately, resistance mechanisms to these inhibitors havealready been described.15–17 Additional SMIs have been designedto perturb the MYC:MAX protein− protein interaction; however,the target specificity of these inhibitors remains unclear andsuggests additional strategies are needed.18–22 Dominant negativeproteins have been designed to interfere with MYC:MAX interac-tion, such as ectopic expression of the MAX basic helix-loop-helixleucine zipper (B-HLH-LZ).23,24 Similarly, Omomyc was designedto heterodimerize with MYC and preclude interaction withMAX.7,8,25,26 Both chemical and protein inhibitors are presentlyunder investigation for their potential as novel anti-MYC inhibitors.Developing an arsenal of MYC inhibitors with unique mechan-

isms of action would significantly impact patient care. We haveundertaken a structure-based strategy to design small proteininhibitors that competitively block native transcription factorsfrom binding to their DNA recognition motifs. These minimalisthybrid protein (MHP) inhibitors are small fusion proteins compris-ing functional domains from two or more transcription factors. Wehave previously demonstrated in vitro that the MHP ME47(previously called MAX-E47) can block E-box sequence-specifictranscription factor interaction with DNA.27,28 This MHP comprisesthe DNA-binding basic region of MAX and the helix-loop-helixdimerization domain of the E47 transcription factor27 (Figure 1a).

1Princess Margaret Cancer Centre, University Health Network, Toronto, ON, Canada; 2Department of Medical Biophysics, University of Toronto, Toronto, ON, Canada; 3Departmentof Chemistry, University of Toronto, Toronto, ON, Canada and 4Institute of Biomaterials and Biomedical Engineering, University of Toronto, Donnelly Centre for Cellular andBiomolecular Research, Department of Materials Science and Engineering, Department of Chemical Engineering, Toronto, ON, Canada. Correspondence: Dr LZ Penn, PrincessMargaret Cancer Centre, University Health Network, Princess Margaret Cancer Research Tower; 13-706, 101 College Street, Toronto, ON, M5G 1L7, Canada.E-mail: [email protected] 16 October 2016; revised 21 April 2017; accepted 27 May 2017

Oncogene (2017), 1–8© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved 0950-9232/17

www.nature.com/onc

http://dx.doi.org/10.1038/onc.2017.275

mailto:[email protected]

http://www.nature.com/onc

At 66 amino acids, ME47 has a simplified structure compared tothe larger native MAX protein (160 amino acids), yet binds toE-box elements with low nanomolar affinity comparable to that ofMAX as determined by electrophoretic mobility shift assays.27

Structural and yeast one-hybrid analyses show ME47 exists as adimer and can outcompete MAX for binding to E-box elements,further supporting its utility as a MHP inhibitor of transcriptionfactor binding in cells.27,28

Here we evaluate the mechanism of action and efficacy of ME47in transformed human cell lines, in both cell culture and in vivotumour xenograft settings. We demonstrate that this small proteininhibitor decreases tumour cell proliferation and transformation,including tumour xenograft growth. We further show that ME47binds to E-box-containing promoters of MYC target genes anddecreases MYC binding at these regulatory regions. Our studyhighlights the prototypic MHP, ME47, as a new class of

transcription factor inhibitor designed to target the interactionof MYC:MAX with E-box DNA. Moreover, this study reinforces theconcept of MHPs as a therapeutic platform to target transcriptionfactors.

RESULTS AND DISCUSSIONME47 decreases cellular proliferationTo evaluate the biological and molecular effect of ME47expression, we first evaluated ME47 in comparison with Omomyc.ME47 and Omomyc are designed to block MYC:MAX activity bydistinct mechanisms of action. ME47 targets MYC:MAX binding tothe E-box motif, whereas Omomyc is designed to block MYC andMAX interaction.25 ME47 and Omomyc were tagged with an in-frame N-terminal FLAG and nuclear localization signal (FLAG-NLS)

Vector

OmomycME47

Vector

OmomycME47

Rel

ativ

e C

olon

y N

umbe

rR

elat

ive

DN

A C

onte

nt

Vector

OmomycME47

Vector

OmomycME47

Perc

ent E

dU P

ositi

ve C

ells

Omomyc ME47

4OHT ++−+−

Vector

−

IB: FLAG

IB: ACTIN

MAX

MYC

E47

ME47

439

MBI MBII MBIIIa MBIIIb MBIV BHLH LZ

3541

22 1131 160

B HLH LZ

335 4001 654

B HLH

1 66

FLAG-NLS B HLH

EtOH

4OHT

Vector Omomyc ME47

Blocking MYC: E-box binding inhibits tumour growthLC Lustig et al

2

Oncogene (2017) 1 – 8 © 2017 Macmillan Publishers Limited, part of Springer Nature.

to visualize ectopic expression and facilitate nuclear localization,respectively.29,30 We used two MYC-sensitive transformed cell linesto evaluate the effects of ME47; MDA-MB-231 is a triple negativebasal breast tumour-derived cell line and HEK293Tv is anembryonic kidney cell line transformed with SV40 large T-antigen.Both systems showed MYC sensitivity in that they exhibitedsignificantly decreased cellular proliferation and transformationupon MYC knockdown by shRNA (Supplementary Figures 1 and 2).We generated MDA-MB-231 and HEK293Tv cells to stably express4-hydroxytamoxifen (4OHT)-inducible FLAG-NLS-Vector, FLAG-NLS-Omomyc or FLAG-NLS-ME47. Conditional expression ofME47 and Omomyc was evident in response to 4OHT exposurein the MDA-MB-231 (Figure 1b) and HEK293Tv (SupplementaryFigure 3A) cells, but not with Vector and ethanol (EtOH) controls.Omomyc showed consistently higher expression relative to ME47for reasons that remain unclear.We next evaluated the effect of ME47 and Omomyc expression

on cellular proliferation using a variety of assays. We measuredcellular proliferation by assaying 5’-ethynyl-2’-deoxyuridine (EdU)incorporation as a measure of active DNA synthesis (Figure 1c).Expression of Omomyc (Po0.05) or ME47 (Po0.01) significantlydecreased EdU incorporation compared to Vector control. We alsoassessed whether ME47 exerts an anti-proliferative effect at thelevel of population doubling time. Following 24 h incubation inEtOH control or 4OHT-containing media, MDA-MB-231 cellsexpressing inducible Vector, Omomyc or ME47 were analysedfor relative DNA content over five consecutive days (Figure 1d).ME47 expression significantly decreased DNA content comparedto Vector control at days 4 and 5 (Po0.05). By extension, thedoubling time of cells expressing ME47 was increased to 26.8 hcompared to Vector control cells at 23.9 h (Figure 1d, inset).Previous studies of Omomyc in Rat1 fibroblasts and SH-SY5Yneuroblastoma demonstrated a significant inhibitory effect on cellproliferation.31 In MDA-MB-231 cells, we observed a similar trendwith Omomyc; however, the results did not reach significance atexperimental end point (P= 0.077). These data indicate that ME47negatively impacts the growth of MDA-MB-231 breast cancer cellsand demonstrates an anti-proliferative phenotype that is compar-able to Omomyc.

Inhibiting anchorage-independent cell growth with ME47expressionAs ME47 demonstrated a decrease in cell proliferation, we tested itseffect on anchorage-independent growth in soft agar, a well-established surrogate measure of cellular transformation.32,33 Induc-tion of ME47 resulted in a significant decrease in colony number(Po0.05) relative to the Vector control (Figure 1e). Omomyc did notsignificantly decrease colony number in this assay. Both Omomycand ME47 expression led to a decreasing trend in anchorage-independent growth of HEK293Tv cells (Supplementary Figure 3).These results show that ME47 antagonizes MYC-dependenttransformation more strongly than does Omomyc, despite consis-tently lower ME47 expression. Omomyc has been shown to formhomodimers as well as interact with MYC, MAX and other MYCinteractors.25,31,34 By comparison, structural analysis reveals thatME47 forms stable homodimers.28 ME47 has no activation domainand shows no interaction with the B-HLH-LZ region of MYC.27

Further comparisons between these two protein inhibitors withdifferent mechanisms of action will help address relative mechan-isms of action and biological activities.

ME47 decreases xenograft tumour growthWe further evaluated the anti-proliferative activity of ME47 byassaying the effect of ME47 on tumour xenograft growth. Animalwork was carried out in accordance with the regulations of theCanadian Council on Animal Care. MDA-MB-231 cells withinducible ME47 or Vector control were injected into the flanksof 10-week-old female NOD/SCID mice and allowed to grow untiltumours were approximately 200 mm3 in volume. Mice withtumours at 200 mm3 were switched onto food that was infusedwith tamoxifen to induce ME47 expression.35 Expression of ME47delayed tumour growth compared to Vector control mice. Tumourvolumes after day 22 were not considered as animals beganreaching experimental end point on day 22 (Figure 2a). This isconsistent with the decreased colony growth due to ME47expression in MDA-MB-231 soft agar assays (Figure 1e). Asignificant decrease in tumour volume (Po0.0001; Figure 2b)was evident 22 days after inducing ME47 expression. Moreover,ME47 significantly increased experimental survival (P= 0.0003;Figure 2c). Thus, ME47 demonstrates a potent anti-proliferative

Figure 1. ME47 decreases cellular proliferation and transformation in MDA-MB-231 breast cancer cells. (a) ME47 is an MHP inhibitor designedto homodimerize and inhibit the interaction of the MYC:MAX heterodimer with DNA E-box sequences. This 66-amino acid hybrid proteinconsists of the DNA-binding basic region (B) of MAX and the helix-loop-helix dimerization domain of E47. The N-terminal FLAG-tag andnuclear localization signal (FLAG-NLS) were included for experimental and subcellular localization purposes, respectively. Schematics of MYC,MAX, E47, and ME47 are shown with relevant regions and amino acid numbers. MB, MYC homology box; LZ, leucine zipper. (b) MDA-MB-231cells expressing 4-hydroxytamoxifen (4OHT)-inducible Vector, Omomyc (92 amino acids) or ME47 were generated by lentiviral infection withthe two-plasmid GEV16 system.29,30 ME47, Omomyc (a kind gift from Dr Laura Soucek), and Vector were cloned into this system with theFLAG-NLS tag. In this system, the constitutively expressed GEV16 transcription factor is translocated to the nucleus only in the presence of4OHT, which enables the GAL4-DNA-binding domain fusion to induce the inserted gene of interest. To evaluate protein expression, cells weretreated for 24 h with 12.5 nM 4OHT (+) or EtOH (− ) control, collected in boiling SDS lysis buffer and evaluated by immunoblot with FLAG(F3165, Sigma, St Louis, MO, USA) and ACTIN (A2066, Sigma) antibodies.42 Immunoblots are representative of three independent experiments.(c) The MDA-MB-231 panel was treated with 10 µg/ml EdU for 10 h and percent EdU incorporation was evaluated using the Click-It EdU AlexaFluor 488 Imaging Kit (Life Technologies, Waltham, MA, USA). Individual and mean values from three independent experiments are shown,*Po0.05, **Po0.01, one-way ANOVA with Bonferroni post-test. (d) Cell proliferation was assessed using the CyQUANT® NF Cell ProliferationAssay Kit (Invitrogen, Carlsbad, CA, USA). The panel of MDA-MB-231 cells was seeded at 1.0x103 cells per well in 96-well plates, treated for 24 hwith 12.5 nM 4OHT (+) or EtOH control (− ) and subsequently measured for DNA content over 5 days. Mean± s.d. for three independentexperiments are shown, *Po0.05, one-way ANOVA with Bonferroni post-test for multiple testing. Blue and red asterisks indicate significanceof Vector (+) compared to Omomyc (+) or ME47 (+), respectively. Population doubling times (inset) were calculated using Prism Graphpad v5.0(GraphPad Software, La Jolla, CA, USA). (e) The panel of MDA-MB-231 cells was treated for 24 h with 12.5 nM 4OHT or EtOH control, thenseeded in triplicate at 5.0 ×103 cells per dish to assess anchorage-independent colony growth in soft agar.43 Plates were imaged after threeweeks on a Leica MZ FLIII stereomicroscope, with representative images shown (left; scale bars= 2 mm). Colony numbers were quantifiedusing ImageJ software (NIH, Bethesda, MD, USA). Individual and mean values from three independent experiments are shown, *Po0.05, one-way ANOVA with Bonferroni post-test for multiple testing. Normal distribution and equal variances were confirmed for all datasets using theD'Agostino & Pearson omnibus normality test and Brown− Forsythe tests, respectively.


3

© 2017 Macmillan Publishers Limited, part of Springer Nature. Oncogene (2017) 1 – 8

Figure 2. ME47 inhibits MDA-MB-231 xenograft tumour growth. (a) MDA-MB-231 cells with 4OHT-inducible Vector or ME47 weresubcutaneously injected with 2.0 × 106 cells in 0.2 ml phosphate-buffer saline into the right flank of 10 female NOD/SCID mice per group.10 animals per group was chosen based off of previous experience with this model and to ensure enough animals for statistical analysis. Oncetumours reached approximately 200 mm3 in volume, mice were assigned to standard chow alone or standard chow supplemented with400 mg tamoxifen citrate per kg (Harlan Laboratories, Indianapolis, IN, USA) and measured every 2 days. The first tumour to reach 200 mm3

was put on tamoxifen chow and the second animal was given standard chow with subsequent animals continuing this pattern until allanimals were assigned a group. Measurement of the animals was blinded, with one researcher providing a blinded cage to a secondresearcher for measurement. Animals were euthanized after tumour volume reached humane (experimental) end point of approximately1000 mm3. Only experimental time points having all animals per group were considered for statistical testing. Tumour volumes are shown ateach corresponding day post tamoxifen treatment. *Po0.05, ****Po0.0001, Student’s unpaired, two-tailed, t-test. (b) Individual and meantumour volumes are shown after 22 days of treatment, the time point at which the first control animal reached humane end point (bearing a1000 mm3 tumour), ****Po0.0001, Student’s unpaired, two-tailed, t-test. (c) Percent survival of tamoxifen-treated Vector or ME47 mice.Survival curve comparison was performed using Log− rank (Mantel−Cox) test (P= 0.0003). (d) An independent MDA-MB-231 xenograftexperiment was performed as described in (a), using four animals for Vector tumours and five animals for ME47. All animals were euthanized33 days post treatment, and tumours were excised and weighed. Individual and mean tumour weights are shown, P= 0.0635, Student’sunpaired, two-tailed, t-test. Representative tumour images are shown (scale bars = 5 mm). (e) Expression of ME47 was detected byimmunoblot using antibody to FLAG and ACTIN as loading control. (f, g) Tumours were formalin-fixed, paraffin-embedded and cross-sectioned for immunohistochemical analysis of Ki67 (f) and TUNEL staining (g) to measure proliferation and apoptosis, respectively. Scalebars= 500 μm. Slides were scanned by the Advanced Optical Microscopy Facility at the Princess Margaret Cancer Centre using ScanScope XT(Aperio Technologies, Vista, CA, USA) and quantified using the Aperio Positive Pixel Count Algorithm. Individual and mean values are shown,*Po0.05, Student’s unpaired, two-tailed, t-test. Normal distribution and equal variances were confirmed for all datasets using the D'Agostino& Pearson omnibus normality test and F-test test respectively.


4


effect on tumour xenograft growth leading to a significant impacton animal survival.To interrogate the mechanism of ME47, the experiment was

repeated and tumours expressing Vector control and ME47 werecollected on the same day while tumours were in growth phase.At end point, the average tumour weight of the ME47 xenograftswas decreased in size relative to Vector controls, as expected

(Figure 2d). This was further evident in side-by-side comparison oftumour images (Figure 2d, inset). Evidence of ME47 expressionwas shown by immunoblot using anti-FLAG to detect ME47expression in tumour xenograft extracts prepared at experimentalend point from the Vector control and ME47-treated animals(Figure 2e). Immunohistochemical analyses were performed onthe excised tumours. Based on the tissue culture results

IB: AcH3

IB: TUBULIN

IB: FLAG

MDA-MB-231 ME47

C N C N4OHT − ++−

CAD

CCND1CDK4

NCLch

r6

% In

put

CAD

CCND1CDK4

NCLch

r6

% In

put

Figure 3. ME47 binds to MYC target gene promoters and interferes with MYC binding to DNA. (a) MDA-MB-231 cells with inducible ME47 weretreated with 12.5 nM 4OHT or EtOH control for 24 h. Cellular fractionation was performed as previously described42 and immunoblotted withantibodies to FLAG for ME47 detection, TUBULIN (CP06, Millipore, Billerica, MA, USA) for cytoplasmic fraction (C) control, and acetylated histoneH3 (AcH3) (sc-8655, Santa Cruz, Dallas, TX, USA) for nuclear fraction (N) control. (b) MDA-MB-231 cells with inducible ME47 or Vector control wereinduced with 12.5 nM 4OHT for 24 h. Chromatin immunoprecipitation followed by quantitative real-time PCR (ChIP-qPCR) was performed aspreviously described,44 with a FLAG antibody (F1804, Sigma) or mouse IgG control (sc-2025, Santa Cruz) for E-box-containing promoter regions ofMYC target genes (CAD, CCND1, CDK4, NCL) and a non-E-box negative control (chr6). (c) ChIP-qPCR was performed as in b, with a MYC antibody(N262, homemade) or rabbit IgG control (sc-2027, Santa Cruz). Mean± s.d. for three independent experiments are shown in (b) and (c), *Po0.05,**Po0.01, ***Po0.001, ****Po0.0001, ns, not significant, one-way ANOVA with Bonferroni post-test; CAD, carbamoyl-phosphate synthetase 2,aspartate transcarbamylase and dihydroorotase; CCND1, cyclin D1; CDK4, cyclin-dependent kinase 4; NCL, nucleolin.


5


demonstrating the anti-proliferative effect of ME47, we assayedtumour sections for Ki67 and TUNEL staining to evaluate theproliferative and apoptotic index, respectively (Figures 2f and g).Consistent with the results in culture, Ki67 staining wassignificantly lower (Po0.05) in the ME47-expressing xenograftscompared to control (Figure 2f). Levels of apoptosis did notchange as a result of ME47 induction (Figure 2g). This suggeststhat the ME47-mediated decrease on cellular transformation andxenograft tumour weight is due to an anti-proliferative effect.

ME47 binds to the DNA E-box and antagonizes MYC binding totarget gene promotersTo evaluate mechanism and determine whether ME47 binds tothe E-box as designed,27 we assessed whether ME47 is localized tothe nuclear compartment and whether induction of ME47 affectedendogenous MYC or MAX expression in MDA-MB-231 cells.Immunoblot of extracts collected following exposure to vehiclecontrol or 4OHT showed robust induction of ME47 and no changein endogenous MYC or MAX expression, as expected(Supplementary Figure 4). Cellular fractionation followed byimmunoblot analysis was conducted on MDA-MB-231 cellsexposed to 4OHT or EtOH control (Figure 3a). The results showthat ME47 (detected by FLAG antibody) is expressed exclusively inthe nuclear fraction and not in the cytoplasmic fraction asindicated by acetylated histone H3 (AcH3) and tubulin expression,respectively (Figure 3a).ME47 was designed to bind DNA E-box motifs through its MAX

basic region. We compared the binding specificity of ME47 to MAXon a series of E-box and non-E-box sequences (SupplementaryFigure 5). ME47 and the B-HLH-LZ domain of MAX bound to theMAX E-box sequence with comparable affinities (dissociationconstant Kd 15 nM for both proteins). ME47 and MAX B-HLH-LZalso showed similar affinities for the ARNT E-box sequence (Kd30 nM and 60 nM, respectively). These two E-box sequencesfeatured the central CACGTG hexamer, but differed in theirflanking base pairs. Both proteins had decreased affinity for theXRE sequence (Kd 40 nM and 70 nM, respectively), which containeda half E-box motif. The binding of both proteins was further

diminished with sequences unrelated to the E-box motif (C/EBP,AP-1, and non-specific DNA sequences; ME47 Kd 60 nM for all threesequences and MAX B-HLH-LZ Kd 100 nM for all three). Against aseries of different DNA sequences, ME47 and MAX exhibitedsimilar preferential binding affinities to E-box motifs.To determine whether ME47 is bound to chromatin at E-box-

containing regulatory regions of MYC-activated genes in cells,ChIP-qPCR was performed. MDA-MB-231 cells with inducible ME47or Vector control were treated with 4OHT, and then ChIP-qPCRwas conducted using antibody to FLAG to detect ME47 or IgG as anegative control (Figure 3b). These analyses reveal significantbinding in the ME47-expressing cells compared to Vector control,as measured by the percent input enriched at the E-box-containing promoters of CAD (Po0.001), CCND1 (Po0.0001),CDK4 (Po0.05), and NCL (Po0.0001), but not the establishednon-E-box negative control region on chromosome 6 (chr6).36,37

We also determined MYC occupancy on these promoter regionsby conducting ChIP-qPCR using an anti-MYC antibody, N262(Figure 3c). MYC binding was readily detected in Vector controlcells at MYC-induced gene promoters (CAD, CCND1, CDK4, NCL),but was significantly decreased in cells expressing ME47. IgGcontrols demonstrated negligible binding to the regions underinterrogation under all cellular conditions. Together, these datastrongly suggest that ME47 binds to E-box DNA and antagonizesMYC binding to target gene promoters.To establish the potential utility of ME47 as a therapeutic

inhibitor, we sought to determine the amount of ME47 relative toendogenous MAX when reduced MYC binding and decreasedproliferation were observed upon ME47 induction. Quantifying theamount of ME47 induced in MDA-MB-231 cells after 24 h against astandard curve of purified ME47, we calculated that ME47 proteinis expressed at 0.117 nmol/μg lysate or 1.36 ng/μg lysate(Supplementary Figures 6A and C). Using a similar approach, astandard curve with purified MAX was generated and MAX wasquantified at 0.00273 nmol/μg lysate or 0.0527 ng/μg lysate(Supplementary Figures 6B and C). At a time point when ME47displayed molecular and cellular inhibitory effects, ME47 waspresent at approximately 42.9 times the molality of endogenous

MYC

MAX

E-box

nucleus

cytosol

nucleus

cytosol

ME47 ME47

E-box

MYC

MAX

No ME47 With ME47

Proliferative Anti-proliferative

Figure 4. Working model of ME47 MHP inhibitor mechanism of action. In the absence of ME47, the MYC:MAX complex binds to the E-box DNAsequences and activates transcriptional programs leading to cancer cell growth (left). Introducing ME47 out-competes MYC:MAX for E-boxbinding, which inhibits the transcriptional activity of MYC:MAX, leading to anti-proliferative effects on cancer cells (right).


6


MAX (Supplementary Figure 6C). This suggests that inducibleexpression of the ME47 protein inhibitor has the capacity to targetendogenous MYC:MAX in tumour cells.

IMPACT AND CONCLUSIONSThis study explores the efficacy of our MHP strategy to targettranscription factor binding to DNA in human cells. We use ME47as the prototypic MHP to disrupt MYC:MAX interaction with theE-box DNA element. We demonstrate that ME47 decreases cancercell proliferation and transformation in MDA-MB-231 breast cancercells in cell culture, colony growth and tumour xenograft settings.Mechanistically, ME47 is expressed exclusively in the nuclearcompartment of the cell, binds to the DNA E-box-containingpromoter regions of MYC-activated target genes, and antagonizesMYC binding to these regions (Figure 4).In contrast to other agents, such as those generated to interfere

with MYC expression or the MYC:MAX protein-protein interaction,ME47 is specifically designed to perturb MYC:MAX at the level ofits DNA-binding activity. Given MYC:MAX exhibits genome-widechromatin binding that becomes widespread in oncogenicconditions,38,39 this class of MYC inhibitor has enormous potentialfor further development as anti-cancer agents. Our work alsoshows that the MHP inhibitor platform may be extended to targetother sequence-specific transcription factors that have implica-tions in human diseases.40,41

CONFLICT OF INTERESTThe authors declare no conflict of interest.

ACKNOWLEDGEMENTSWe would like to thank the Penn, Shin, and Chen labs for their helpful reviews andcontributions to this manuscript. Special thanks to Dr Peter Mullen, Dr SamSathiamoorthy and Peter Tang. Funding was provided by way of Operating grantsupport from the Collaborative Health Research Program (JS, WC, LZP). Salary andstipend support from the Canadian Research Chairs Program (LZP), Natural Sciencesand Engineering Research Council (LCL), Canadian Breast Cancer Foundation OntarioRegion Doctoral Fellowship (WBT and MK).

REFERENCES1 Dang CV. MYC on the path to cancer. Cell 2012; 149: 22–35.2 Vita M, Henriksson M. The Myc oncoprotein as a therapeutic target for

human cancer. Semin Cancer Biol 2006; 16: 318–330.3 Soucek L, Whitfield JR, Sodir NM, Masso-Valles D, Serrano E, Karnezis AN et al.

Inhibition of Myc family proteins eradicates KRas-driven lung cancer in mice.Genes Dev 2013; 27: 504–513.

4 Meyer N, Penn LZ. Reflecting on 25 years with MYC. Nat Rev Cancer 2008; 8:976–990.

5 Blackwood EM, Eisenman RN. Max: a helix-loop-helix zipper protein that formsa sequence-specific DNA-binding complex with Myc. Science 1991; 251:1211–1217.

6 Ponzielli R, Katz S, Barsyte-Lovejoy D, Penn LZ. Cancer therapeutics: targeting thedark side of Myc. Eur J Cancer 2005; 41: 2485–2501.

7 Soucek L, Whitfield J, Martins CP, Finch AJ, Murphy DJ, Sodir NM et al. ModellingMyc inhibition as a cancer therapy. Nature 2008; 455: 679–683.

8 Soucek L, Nasi S, Evan GI. Omomyc expression in skin prevents Myc-inducedpapillomatosis. Cell Death Differ 2004; 11: 1038–1045.

9 Jain M, Arvanitis C, Chu K, Dewey W, Leonhardt E, Trinh M et al. Sustained lossof a neoplastic phenotype by brief inactivation of MYC. Science 2002; 297:102–104.

10 Felsher DW. MYC inactivation elicits oncogene addiction through both tumor cell-intrinsic and host-dependent mechanisms. Genes Cancer 2010; 1: 597–604.

11 Delmore JE, Issa GC, Lemieux ME, Rahl PB, Shi J, Jacobs HM et al. BET bromo-domain inhibition as a therapeutic strategy to target c-Myc. Cell 2011; 146:904–917.

12 Fu LL, Tian M, Li X, Li JJ, Huang J, Ouyang L et al. Inhibition of BET bromodomainsas a therapeutic strategy for cancer drug discovery. Oncotarget 2015; 6:5501–5516.

13 Picaud S, Da Costa D, Thanasopoulou A, Filippakopoulos P, Fish PV, Philpott Met al. PFI-1, a highly selective protein interaction inhibitor, targeting BET bro-modomains. Cancer Res 2013; 73: 3336–3346.

14 Posternak V, Cole MD. Strategically targeting MYC in cancer. F1000Res 2016; 5:408.

15 Rathert P, Roth M, Neumann T, Muerdter F, Roe JS, Muhar M et al. Transcriptionalplasticity promotes primary and acquired resistance to BET inhibition. Nature2015; 525: 543–547.

16 Shi X, Mihaylova VT, Kuruvilla L, Chen F, Viviano S, Baldassarre M et al. Loss ofTRIM33 causes resistance to BET bromodomain inhibitors through MYC-and TGF-beta-dependent mechanisms. Proc Natl Acad Sci USA 2016; 113:E4558–E4566.

17 Fong CY, Gilan O, Lam EY, Rubin AF, Ftouni S, Tyler D et al. BET inhibitor resistanceemerges from leukaemia stem cells. Nature 2015; 525: 538–542.

18 Jung KY, Wang H, Teriete P, Yap JL, Chen L, Lanning ME et al. Perturbation of thec-Myc-Max protein-protein interaction via synthetic alpha-helix mimetics. J MedChem 2015; 58: 3002–3024.

19 Soodgupta D, Pan D, Cui G, Senpan A, Yang X, Lu L et al. Small molecule MYCinhibitor conjugated to integrin-targeted nanoparticles extends survival in amouse model of disseminated multiple myeloma. Mol Cancer Ther 2015; 14:1286–1294.

20 Berg T. Small-molecule modulators of c-Myc/Max and Max/Max interactions. CurrTop Microbiol Immunol 2011; 348: 139–149.

21 Follis AV, Hammoudeh DI, Daab AT, Metallo SJ. Small-molecule perturbation ofcompeting interactions between c-Myc and Max. Bioorg Med Chem Lett 2009; 19:807–810.

22 Yin X, Giap C, Lazo JS, Prochownik EV. Low molecular weight inhibitors of Myc-Max interaction and function. Oncogene 2003; 22: 6151–6159.

23 Beaulieu ME, McDuff FO, Frappier V, Montagne M, Naud JF, Lavigne P. Newstructural determinants for c-Myc specific heterodimerization with Max anddevelopment of a novel homodimeric c-Myc b-HLH-LZ. J Mol Recognit 2012; 25:414–426.

24 Montagne M, Beaudoin N, Fortin D, Lavoie CL, Klinck R, Lavigne P. The Maxb-HLH-LZ can transduce into cells and inhibit c-Myc transcriptional activities. PLoSOne 2012; 7: e32172.

25 Soucek L, Helmer-Citterich M, Sacco A, Jucker R, Cesareni G, Nasi S. Design andproperties of a Myc derivative that efficiently homodimerizes. Oncogene 1998; 17:2463–2472.

26 Soucek L, Jucker R, Panacchia L, Ricordy R, Tato F, Nasi S. Omomyc, a potentialMyc dominant negative, enhances Myc-induced apoptosis. Cancer Res 2002; 62:3507–3510.

27 Xu J, Chen G, De Jong AT, Shahravan SH, Shin JA. Max-E47, a designed minimalistprotein that targets the E-box DNA site in vivo and in vitro. J Am Chem Soc 2009;131: 7839–7848.

28 Ahmadpour F, Ghirlando R, De Jong AT, Gloyd M, Shin JA, Guarne A. Crystalstructure of the minimalist Max-E47 protein chimera. PloS One 2012; 7: e32136.

29 Callus BA, Ekert PG, Heraud JE, Jabbour AM, Kotevski A, Vince JE et al. Cytoplasmicp53 is not required for PUMA-induced apoptosis. Cell Death Differ 2008; 15:213–215, author reply 215-216.

30 Vince JE, Wong WW, Khan N, Feltham R, Chau D, Ahmed AU et al. IAP anta-gonists target cIAP1 to induce TNFalpha-dependent apoptosis. Cell 2007; 131:682–693.

31 Savino M, Annibali D, Carucci N, Favuzzi E, Cole MD, Evan GI et al. The actionmechanism of the Myc inhibitor termed Omomyc may give clues on how totarget Myc for cancer therapy. PloS ONE 2011; 6: e22284.

32 Wasylishen AR, Kalkat M, Kim SS, Pandyra A, Chan PK, Oliveri S et al. MYC activityis negatively regulated by a C-terminal lysine cluster. Oncogene 2013; 33:1066–1072.

33 Wasylishen AR, Stojanova A, Oliveri S, Rust AC, Schimmer AD, Penn LZ. Newmodel systems provide insights into Myc-induced transformation. Oncogene2011; 30: 3727–3734.

34 Mongiardi MP, Savino M, Bartoli L, Beji S, Nanni S, Scagnoli F et al. Myc andOmomyc functionally associate with the protein arginine methyltransferase 5(PRMT5) in glioblastoma cells. Sci Rep 2015; 5: 15494.

35 Wilson CH, Gamper I, Perfetto A, Auw J, Littlewood TD, Evan GI. The kinetics of ERfusion protein activation in vivo. Oncogene 2014; 33: 4877–4880.

36 Ponzielli R, Boutros PC, Katz S, Stojanova A, Hanley AP, Khosravi F et al. Optimi-zation of experimental design parameters for high-throughput chromatinimmunoprecipitation studies. Nucleic Acids Res 2008; 36: e144.

37 Wasylishen AR, Chan-Seng-Yue M, Bros C, Dingar D, Tu WB, Kalkat M et al. MYCphosphorylation at novel regulatory regions suppresses transforming activity.Cancer Res 2013; 73: 6504–6515.

38 Lin CY, Loven J, Rahl PB, Paranal RM, Burge CB, Bradner JE et al. Transcriptionalamplification in tumor cells with elevated c-Myc. Cell 2012; 151: 56–67.


7


39 Sabo A, Kress TR, Pelizzola M, de Pretis S, Gorski MM, Tesi A et al. Selectivetranscriptional regulation by Myc in cellular growth control and lymphomagen-esis. Nature 2014; 511: 488–492.

40 Bhagwat AS, Vakoc CR. Targeting transcription factors in cancer. Trends Cancer2015; 1: 53–65.

41 Lee TI, Young RA. Transcriptional regulation and its misregulation in disease. Cell2013; 152: 1237–1251.

42 Kalkat M, Chan PK, Wasylishen AR, Srikumar T, Kim SS, Ponzielli R et al. Identifi-cation of c-MYC SUMOylation by mass spectrometry. PloS One 2014; 9: e115337.

43 Oster SK, Mao DY, Kennedy J, Penn LZ. Functional analysis of the N-terminaldomain of the Myc oncoprotein. Oncogene 2003; 22: 1998–2010.

44 Stojanova A, Tu WB, Ponzielli R, Kotlyar M, Chan PK, Boutros PC et al. MYCinteraction with the tumor suppressive SWI/SNF complex member INI1 reg-ulates transcription and cellular transformation. Cell Cycle 2016; 15: 1–13.

Supplementary Information accompanies this paper on the Oncogene website (http://www.nature.com/onc)


8


inhibiting myc binding to the e-box dna motif by me47 decreases …shinjumi/publications/pdfs/31...

Documents