changes in gene expression. Pol II poising mayaccommodate these dynamics by promoting rapidand/or synchronous transcriptional onset (12, 19).Recruitment of poised Pol II is followed bydecompaction of chromatin. One appealing hy-pothesis is that deposition of Pol II at TSS regionsmay participate in chromatin opening, along withPHA-4 (18). This scenario predicts that Pol II bindsto regions that would otherwise contain stable nu-cleosomes, a prediction that is borne out by ourFAIRE analysis, and that Pol II interferes with theconstruction of nucleosomes. It will be of interestto see whether other pioneer factors or FoxA pro-teins also poise Pol II.

REFERENCES AND NOTES

1. M. Iwafuchi-Doi, K. S. Zaret, Genes Dev. 28, 2679–2692(2014).

2. D. L. D. L. Updike, S. E. S. E. Mango, PLOS Genet. 2, e161(2006).

3. S. E. Mango, Annu. Rev. Cell Dev. Biol. 25, 597–628(2009).

4. M. A. Horner et al., Genes Dev. 12, 1947–1952(1998).

5. J. M. Kalb et al., Development 125, 2171–2180(1998).

6. M. Zhong et al., PLOS Genet. 6, e1000848 (2010).7. P. G. Giresi, J. Kim, R. M. McDaniell, V. R. Iyer, J. D. Lieb,

Genome Res. 17, 877–885 (2007).8. J. W. K. Ho et al., Nature 512, 449–452 (2014).9. T. H. I. Fakhouri, J. Stevenson, A. D. Chisholm, S. E. Mango,

PLOS Genet. 6, e1001060 (2010).10. J. M. Caravaca et al., Genes Dev. 27, 251–260

(2013).11. Materials and methods are available as supplementary

materials on Science Online.12. K. Adelman, J. T. Lis, Nat. Rev. Genet. 13, 720–731

(2012).13. L. R. Baugh, J. Demodena, P. W. Sternberg, Science 324,

92–94 (2009).14. C. S. C. S. Maxwell et al., Cell Reports 6, 455–466

(2014).15. C. A. Kuchenthal, W. Chen, P. G. Okkema, Genesis 31, 156–166

(2001).16. B. Gaertner et al., Cell Reports 2, 1670–1683 (2012).17. J. C. Kiefer, P. A. Smith, S. E. Mango, Dev. Biol. 303, 611–624

(2007).18. D. A. Gilchrist et al., Cell 143, 540–551 (2010).19. A. N. Boettiger, M. Levine, Science 325, 471–473

(2009).20. M. Levin, T. Hashimshony, F. Wagner, I. Yanai, Dev. Cell 22,

1101–1108 (2012).

ACKNOWLEDGMENTS

We thank J. Whetstine for ChIP advice, A. Schier and S. von Stetinafor comments, and G. Marnellos and M. Clamp for informatics.Some strains were obtained from the Caenorhabditis GeneticsCenter (CGC), funded by NIHP40OD010440. S.E.M. receivedsupport from the MacArthur Foundation and grant NIHR37GM056264,E.L. received support from grant NSFMCB-1413134, and K.S.Z.received support from grant NIHR37GM36477. Sequencing data areaccessible from the National Center for Biotechnology InformationGene Expression Omnibus: GSM1666978 R1.CE.8E.ChIPGSM1666979, R1.CE.8E.Input GSM1666980, R2.CE.8E.ChIPGSM1666981, R2.CE.8E.Input GSM1666982, R1.CE.BE.ChIPGSM1666983, and R1.CE.BE.Input.

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/348/6241/1372/suppl/DC1Materials and MethodsSupplementary TextFigs. S1 to S7Tables S1 to S3References

13 March 2015; accepted 12 May 201510.1126/science.aab1223

DRUG DEVELOPMENT

Phthalimide conjugation as a strategyfor in vivo target protein degradationGeorg E. Winter,1* Dennis L. Buckley,1* Joshiawa Paulk,1 Justin M. Roberts,1

Amanda Souza,1 Sirano Dhe-Paganon,2 James E. Bradner1,3†

The development of effective pharmacological inhibitors of multidomain scaffoldproteins, notably transcription factors, is a particularly challenging problem. In part,this is because many small-molecule antagonists disrupt the activity of only one domainin the target protein. We devised a chemical strategy that promotes ligand-dependenttarget protein degradation using as an example the transcriptional coactivator BRD4, aprotein critical for cancer cell growth and survival. We appended a competitive antagonistof BET bromodomains to a phthalimide moiety to hijack the cereblon E3 ubiquitin ligasecomplex. The resultant compound, dBET1, induced highly selective cereblon-dependentBET protein degradation in vitro and in vivo and delayed leukemia progression in mice.A second series of probes resulted in selective degradation of the cytosolic proteinFKBP12. This chemical strategy for controlling target protein stability may haveimplications for therapeutically targeting previously intractable proteins.

Phthalimide-based drugs emerged in the1950s. Among themost notable was thalid-omide, developed initially as a sedative butinfamouslywithdrawn fromhumanuse ow-ing to catastrophic teratogenicity (1). More

recently, the phthalimides have been successfullyrepurposed for erythema nodosum leprosum,multiplemyeloma (MM), andmyelodyspasia. Theefficacy of thalidomide, lenalidomide, andpomali-domide inMM(Fig. 1A) has prompted investigationinto the mechanism of action of phthalimide im-munomodulatorydrugs (IMiDs). By ligand-affinitychromatography, cereblon (CRBN)—a componentof a cullin-RINGubiquitin ligase (CRL) complex—was identified as the target of thalidomide (2).Recently, our group and others reported thatphthalimides prompt CRBN-dependent protea-somal degradation of transcription factors (TFs)IKZF1 and IKZF3 (3, 4). Crystallographic studiesnow establish that IMiDs bind CRBN to form acryptic interface that promotes recruitment ofIKZF1 and IKZF3 (5).Ligand-induced target protein destabilization

has proven tobeanefficacious therapeutic strategy,in particular for cancer, as illustrated by arsenictrioxide–mediateddegradation of thePMLproteinin acute promyelocytic leukemia (6) and estrogenreceptor degradation by fulvestrant (7). Histori-cally, target-degrading compounds have emergedserendipitously or through target-specific cam-paigns in medicinal chemistry. Chemical biolo-gists have devised elegant solutions to modulateprotein stability using engineered cellular systems,but these approaches have been limited to non-endogenous fusion proteins (8–11). Others have

achieved the degradation of endogenous proteinsthrough the recruitment of E3 ligases, but theseapproaches have been limited by the requirementof peptidic ligands (12–14), the use of nonspecificinhibitors (15), and by low cellular potency.RING-domainE3ubiquitin-protein ligases lack

enzymatic activity and function as adaptors to E2ubiquitin-conjugating enzymes. Inspired by theretrieval of CRBN using a tethered thalidomide(2), we hypothesized that rational design of bi-functional phthalimide-conjugated ligands couldconfer CRBN-dependent target protein degrada-tion as chemical adapters. We selected BRD4 asan exemplary target. BRD4 is a transcriptionalcoactivator that binds to enhancer and promoterregions by recognition of acetylated lysines on his-tone proteins andTFs (16). Recently,we developeda direct-acting inhibitor of BET bromodomains(JQ1) (17) that displaces BRD4 from chromatinand leads to impaired signal transduction fromTFs to RNA polymerase II (18–20). Silencing ofBRD4 expression by RNA interference inmurineand human models of MM and acute myeloidleukemia (AML) elicited rapid transcriptionaldown-regulation of theMYC oncogene and a po-tent antiproliferative response (19, 21). These andother studies in cancer, inflammation (22), andheart disease (23, 24) establish a desirable mech-anistic and translational purpose to target BRD4for selective degradation.Having shown that the carboxyl group on JQ1

(25) and the aryl ring of thalidomide (5) can tol-erate chemical substitution, we designed the bi-functional dBET1 to have preservedBRD4 affinityand an inactive epimeric dBET1(R) as a stereo-chemical control (Fig. 1, A and B). Selectivity pro-filing confirmed potent and BET-specific targetengagement among 32 bromodomains (BromoScan)(Fig. 1C and tables S1 and S2). A high-resolutioncrystal structure (1.0 Å) of dBET1 bound to BRD4(1)confirmed the mode of molecular recogni-tion, comparable to JQ1 (Fig. 1D, fig. S1, and

1376 19 JUNE 2015 • VOL 348 ISSUE 6241 sciencemag.org SCIENCE

1Department of Medical Oncology, Dana-Farber CancerInstitute, Boston, MA 02115, USA. 2Department of CancerBiology, Dana-Farber Cancer Institute, Boston, MA 02115,USA. 3Department of Medicine, Harvard Medical School,Boston, MA 02115, USA.*These authors contributed equally to this research. †Correspondingauthor: E-mail: james_bradner@dfci.harvard.edu

RESEARCH | REPORTSon A

ugust 2, 2020

http://science.sciencemag.org/

Dow

nloaded from

table S3). Using the dBET1-BRD4(1) crystal struc-ture and the recently reported structure of CRBNbound to thalidomide (5), we have modeled thefeasibility of ternary complex formation in silico.An extended conformation of dBET1was capableof bridging ordered BRD4(1) and CRBN withoutdestructive steric interactions (Fig. 1E). To exper-imentally assess the chemical adaptor function ofdBET1, we established a homogeneous proxim-ity assay for recombinant human CRBN–DNAdamage–binding protein 1 (CRBN-DDB1) andBRD4(1) (Fig. 1F). Luminescence arising fromprox-imity of CRBN-DDB1- and BRD4-bound acceptor-donor beads increases with low concentrations ofdBET1 and decreases at higher concentrations,consistent with saturation of CRBN and BRD4binding sites by excess ligand. Competitive coin-cubation with free JQ1 or thalidomide inhibits

ternary complex formation in a stereo-specificmanner (Fig. 1G).To assess the effect of dBET1 in cells, we treated

a humanAML cell line (MV4;11) for 18 hours withincreasing concentrations of dBET1 and assayedendogenous BRD4 levels by immunoblot. Pro-nounced loss of BRD4 (>85%) was observed withconcentrationsof dBET1 as lowas 100nM(Fig. 1H).The epimeric control dBET1(R) was inactive (Fig.1B), which demonstrated that BRD4 degradationrequires target protein engagement (fig. S2, Aand B). The kinetics of BRD4 degradation werenext determined using 100 nM dBET1 in MV4;11cells. Marked depletion of BRD4 was observed at1 hour, and complete degradationwas observed at2 hours of treatment (Fig. 1I). A partial recovery inBRD4 abundance at 24 hours establishes the pos-sibility of compound instability, a recognized lia-

bility of phthalimides. To quantify dose-responsiveeffects on BRD4 protein stability, we developed acell-count normalized, high-content assay usingadherent SUM149 breast cancer cells (Fig. 1J). De-pletion of BRD4 was observed for dBET1 [half-maximal effective dose (EC50) = 430 nM] withoutapparent activity for dBET1(R), confirmed byimmunoblot (fig. S2, C and D). Additional cul-tured adherent and nonadherent human cancercell lines showed comparable response (SUM159,MOLM13) (fig. S3).To explore the mechanism of dBET1-induced

BRD4 degradation, we studied requirements onproteasome function, BRD4 binding, and CRBNbinding using chemical genetic and gene-editingapproaches. First, we confirmed that treatmentwith either JQ1 or thalidomide alone was insuf-ficient to induce BRD4degradation inMV4;11 cells

SCIENCE sciencemag.org 19 JUNE 2015 • VOL 348 ISSUE 6241 1377

Fig. 1. Design and characterization of dBET1. (A) Chemical structure of JQ1(S), phthalimides, and dBET1. (B) Vehicle-normalized BRD4 displacement byAlphaScreen (triplicate means T SD). (C) Selective displacement of phage-displayed BETs by dBET1 (BromoScan at 1 mM; n = 1). (D) Crystal structure ofdBET1 bound to BRD4 bromodomain 1 (E) Docking of (D) into the publishedDDB1-CRBN structure (F) dBET1-induced ternary complex formation of re-combinant BRD4(1) and CRBN-DDB1 by AlphaScreen [quadruplicate means T

SD; normalized to dimethyl sulfoxide (DMSO)]. (G) Competition of 111 nMdBET1-induced proximity as in (F) in the presence of vehicle (DMSO), JQ1,

thal-(–), JQ1(R), and thal-(+) all at 1 mM. Values represent quadruplicatemeans T SD, normalized to DMSO. (H) Immunoblot for BRD4 and vinculin(VINC) after 18 hours of treatment of MV4;11 cells with the indicated con-centrations of dBET1. (I) Immunoblot for BRD4 and vinculin after treatmentof MV4;11 cells with 100 nM dBET1 for the indicated exposures. (J) Cell count–normalized BRD4 levels as determined by high-content imaging in SUM149cells treated with the indicated concentrations of dBET1 and dBET1(R) for18 hours.Values represent triplicate means T SD, normalized to DMSO-treatedcells and baseline-corrected using immunoblots seen in fig. S2C.

RESEARCH | REPORTSon A

ugust 2, 2020

http://science.sciencemag.org/

Dow

nloaded from

(fig. S4A). Degradation of BRD4 by dBET1 wasrescued by the proteasome inhibitor carfilzomib,which established a requirement for proteasomefunction (Fig. 2A). Pretreatment with excess JQ1or thalidomide abolished dBET1-induced BRD4degradation, consistent with a requirement forboth BRD4 and CRBN engagement (Fig. 2A).Pretreatment with the NAE1 inhibitor MLN4924(26) rescued BRD4 stability and established its de-pendence on CRL activity, as expected for cullin-based ubiquitin ligases that require neddylationfor processive E3 ligase activity (4, 27) (fig. S4B).Finally, to definitively confirm the cellular require-ment for CRBN, we used a previously publishedCRBN-deficient human MM cell line (MM1.S-CRBN−/−) (4). Whereas treatment of wild-typeMM1.SWT cells with dBET1 promoted degrada-tion of BRD4, exposure of MM1.S-CRBN−/− cellsto dBET1 was ineffectual (Fig. 2B). These data pro-vide mechanistic evidence for CRBN-dependentproteasomal degradation of BRD4 by dBET1.To assess the feasibility of extending this strategy

to other protein targets, we designed and syn-thesized phthalimide-conjugated ligands to thecytosolic signaling protein FKBP12 (FKBP1A).FKBP12 has been extensively studied in thechemical biology literature, which includes studiesof engineered target degradation and a growingliterature on chemical dimerizers. At a known per-missive site on the FKBP12-directed ligand steelfactor (SLF), we positioned two chemical spacers

to create the conjugated phthalimides dFKBP-1 and dFKBP-2 (Fig. 2C). Both potently decreasedFKBP12 abundance in MV4;11 cells (Fig. 2D). Aswith dBET1, destabilization of FKBP12 by dFKBP-1 was rescued by pretreatment with carfilzomib,MLN4924, free SLF, or free thalidomide (Fig. 2E).We established CRBN-dependent degradation us-ing previously published isogenic 293FT cell linesthat arewild type (293FT-WT) or deficient (293FT-CRBN−/−) (4) for CRBN (Fig. 2F).An unbiased, proteome-wide approach was se-

lected to assess the cellular consequences of dBET1treatment on protein abundance, in a quantitativeand highly parallel format (28). We compared theimmediate impact of dBET1 treatment (250nM) toJQ1 and vehicle controls in MV4;11 cells. A 2-hourincubation was selected to capture primary, im-mediate consequences of small-molecule actionand to mitigate expected, confounding effects onsuppressed transactivation of BRD4 target genes.We prepared three biological sample replicatesfor each treatment condition using isobaric tag-ging that allowed the detection of 7429 proteins.After JQ1 treatment, few proteomic changes wereobserved (Fig. 3A and table S4). Only MYC wassignificantly depleted more than twofold after2 hours of JQ1 treatment, which confirmed thereported rapid and selective effect of BET bromo-domain inhibition on MYC abundance in AML(Fig. 3, A and C) (21). JQ1 treatment also down-regulated the oncoprotein PIM1 (Fig. 3, A and C).

Treatment with dBET1 elicited a comparable,modest effect on MYC and PIM1 expression. Re-markably, only three additional proteins wereidentified as significantly (P < 0.001) and mark-edly depleted (to one-fifth) in dBET1-treated cells:BRD2, BRD3, and BRD4 (Fig. 3, B and C). Thesefindings are consistent with the anticipated, BET-specific bromodomain target spectrum of the JQ1bromodomain-biasing element on dBET1 (17). Theremaining BET-family member BRDT is not de-tectable in MV4;11 cells. To validate these find-ings, wemeasured BRD2, BRD3, BRD4,MYC, andPIM1 levels by immunoblot after compound treat-ment, as above. BET family members were de-graded only by dBET1, whereas MYC and PIM1abundance was decreased by both dBET1 andJQ1 (and to a lesser degree) (Fig. 3D). No effect onIkaros TF expressionwas observed in either treat-ment condition (fig. S5).MYC and PIM1 are oftenassociatedwithmassive adjacent enhancer loci byepigenomic profiling (18, 20), which suggested atranscriptionalmechanism of down-regulation.Wetherefore measured mRNA transcript abundancefor each depleted gene product (Fig. 3E). Treat-ment with either JQ1 or dBET1 down-regulatedMYC and PIM1 transcription, suggestive of sec-ondary transcriptional effects. Transcription ofBRD4 and BRD3 were unaffected, consistentwith posttranscriptional effects. Note that tran-scription of BRD2was affected by JQ1 and dBET1,whereas protein stability of theBRD2geneproduct

1378 19 JUNE 2015 • VOL 348 ISSUE 6241 sciencemag.org SCIENCE

Fig. 2. Chemical andgenetic rescue ofdBET1- and dFKBP-1–mediated degrada-tion. (A) Immunoblotfor BRD4 and vinculinafter a 4-hour pre-treatment with DMSO,carfilzomib (400 nM),JQ1 (10 mM), or thalid-omide (10 mM),followed by a 2-hourdBET1 treatment(100 nM) in MV4;11cells. (B) Immunoblotfor BRD4, CRBN, andtubulin after treatment ofMM1SWTorMM1SCRBN−/−

cells with dBET1 for18 hours at the indi-cated concentrations.(C) Structures ofdFKBP-1 and dFKBP-2.(D) Immunoblot forFKBP12 and vinculinafter 18 hours of treat-ment with the indicatedcompounds. (E) Immu-noblot for FKBP12 andvinculin after a 4-hourpretreatment withDMSO, carfilzomib(400 nM), MLN4924 (1 mM), SLF (20 mM), or thalidomide (10 mM), followed by a 4-hour dFKBP-1 treatment (1 mM) in MV4;11 cells. (F) Immunoblot for FKBP12,CRBN, and tubulin (Tub) after treatment of 293FTWT or 293FTCRBN−/− cells with dFKBP12 at the indicated concentrations for 18 hours.

RESEARCH | REPORTSon A

ugust 2, 2020

http://science.sciencemag.org/

Dow

nloaded from

was only influenced by dBET1, suggestive of tran-scriptional and posttranscriptional consequences.These data establish a highly selective effect ofdBET1 on target protein stability, proteome-wide.We next explored the differential antiprolif-

erative consequences of BET degradation withdBET1 to BET bromodomain inhibition with JQ1.Degradation of BRD4 by dBET1 was associatedwith an enhanced apoptotic response in MV4;11AML and DHL4 lymphoma cells, as measured by

caspase activation (Caspase-GLO) (Fig. 4A), cleav-age of poly(ADP-ribose) polymerase (PARP), cleav-age of caspase-3 (immunoblot) (Fig. 4B), andAnnexin V staining (flow cytometry) (fig. S6A).Kinetic studies revealed an apoptotic advantagefor dBET1 after pulsed treatment followed bywashout in MV4;11 cells (Fig. 4, C and D). In-deed, dBET1 induced a potent and superior in-hibitory effect on MV4;11 cell proliferation at24 hours (Fig. 4E).

The rapid biochemical activity and robustapoptotic response of cultured cell lines to dBET1established the feasibility of assessing effects onprimary human AML cells, where ex vivo pro-liferation is negligible in short-term cultures.Exposure of primary leukemic patient blasts todBET1 elicited dose-proportionate depletion ofBRD4 (Fig. 4F) and improved apoptotic responsecompared to JQ1 (Annexin V and propidium io-dide staining) (Fig. 4G and fig. S6B). Together,

SCIENCE sciencemag.org 19 JUNE 2015 • VOL 348 ISSUE 6241 1379

7429 quantified proteins

p-value

rela

tive

abun

danc

e

rela

tive

abun

danc

e

p-value

JQ1 dBET12

1

0.5

0.1 1

0.25

0.125

0.0625

2

1

0.5

0.25

0.125

0.0625

0.00

001

0.00

01

0.00

1

0.01 0.1 1

0.00

001

0.00

01

0.00

1

0.01

7429 quantified proteins

Fig. 3. Selective BET bromodomain degradation established by expressionproteomics. MV4;11 cells were treated for 2 hours with DMSO, 250 nM dBET1,or 250 nM JQ1. (A) Fold-change in abundance of 7429 proteins comparing JQ1 tovehicle (DMSO) treatment, versus P value (t-test; triplicate analysis). (B) As for(A), but comparing 250 nM dBET1 to vehicle treatment. (C) Selected proteinsfrom (A) and (B) normalized to vehicle. Values represent triplicate means T SD.

(D) Immunoblot of BRD2, BRD3, BRD4,MYC, PIM1, and vinculin (VINC) after a2-hour treatment of MV4;11 cells with DMSO, 250 nM dBET1, or 250 nM JQ1. (E)Quantitative reverse transcription–polymerase chain reaction analysis of tran-script levels of BRD2, BRD3, BRD4, MYC, and PIM1 after a 2-hour treatment ofMV4;11 cells with DMSO, 250 nM dBET1, or 250 nM JQ1. Values representtriplicate means T SD. *P = 0.01 to 0.05; **P = 0.001 to 0.01; ****P < 0.0001.

RESEARCH | REPORTSon A

ugust 2, 2020

http://science.sciencemag.org/

Dow

nloaded from

these data exemplify that target degradationcan elicit a more pronounced biological conse-quence than domain-specific target inhibition.Tomodel the therapeutic opportunity of BRD4

degradation in vivo, we first evaluated the tol-erability and antitumor efficacy of dBET1 in amu-rine hind-limb xenograft model of humanMV4;11leukemia cells (29). As pharmacokinetic studies ofdBET1 indicated adequate drug exposure in vivo(fig. S7A), tumor-bearing mice were treated withdBET1 administered by intraperitoneal injection(50mg/kg bodyweight daily, ip) or vehicle control.After 14 days of therapy, a first tumor reachedinstitutional limits for tumor size, and the studywas terminated for comparative assessment of

efficacy and modulation of BRD4 stability andfunction. Administration of dBET1 attenuated tu-mor progression, as determined by serial volu-metric measurement (Fig. 4H), and decreasedtumorweightwas assessedpostmortem (fig. S8A).Acute pharmacodynamic degradation of BRD4was observed by immunoblot 4 hours after a firstor second daily treatment with dBET1 (50mg/kgip) (Fig. 4I), accompanied by down-regulation ofMYC (Fig. 4I). These findings were confirmed byimmunohistochemistry at the conclusion of the14-day efficacy study (Fig. 4J). A statistically sig-nificant destabilization of BRD4, down-regulationof MYC, and inhibition of proliferation (Ki67staining) was observed with dBET1 compared

with vehicle control in excised tumors (Fig. 4Jand fig. S8B). Notably, 2 weeks of dBET1 waswell tolerated with preservation of animal weightand normal complete blood counts (fig. S7, C andD). To compare the efficacy of BET inhibition toBET degradation, we selected an aggressive dis-seminated leukemia model (mCherry+ MV4;11)and treated animals with established disease usingequimolar concentrations of JQ1 and dBET1 for 19days. Post mortem analysis of leukemic burdenin bonemarrow by fluorescence-activated cell sort-ing revealed significantly decreased mCherry+ dis-ease with dBET1 administration (Fig. 4K).In summary, we present a mechanism-based

chemical strategy for endogenous target protein

1380 19 JUNE 2015 • VOL 348 ISSUE 6241 sciencemag.org SCIENCE

Fig. 4. The kinetic and antileukemic advantage of BET bromodomaindegradation. (A) Fold increase of apoptosis, assessed via Caspase-Glo assayrelative to DMSO-treated controls, 24 hours of treatment in MV4;11 or DHL4cells.Values represent quadruplicate means T SD. (B) Immunoblot for cleavedcaspase 3, PARPcleavage, and vinculin after treatment with dBET1 and JQ1 for24 hours, as indicated. (C) Fold increase of apoptosis (Caspase-Glo) relative toDMSO-treated controls. MV4;11 cells were treated for 4 or 8 hours with JQ1 ordBET1 at the indicated concentrations. Drug was washed out with phosphate-buffered saline (3 times) before cells were plated in drug-free medium for afinal treatment duration of 24 hours.Values represent quadruplicate means T

SD. (D) Immunoblot for cleaved caspase 3, PARP cleavage, and vinculin aftertreatment conditions as described in (C). (E) Dose-proportional effect ofdBET1 and JQ1 (24 hours) on MV4;11 cellular viability as approximated byadenosine triphosphate–dependent luminescence. Values represent quadru-plicate means T SD. (F) Immunoblot for BRD4 and vinculin after treatment of

primary patient cells with the indicated concentrations of dBET1 for 24 hours.(G) Annexin V–positive primary patient cells after 24 hours of treatment witheither dBET1 or JQ1 at the indicated concentrations. Values represent theaverage of duplicates and the range as error bars (representative scatter plotsin fig. S6). (H) Tumor volume (means T SEM) of vehicle-treatedmice (n = 5) ormice treated with dBET1 (50 mg/kg; n = 6) for 14 days. (I) Immunoblot forBRD4, MYC, and vinculin (VINC) by using tumor lysates from mice treatedeither once for 4 hours or twice for 22 hours and 4 hours, compared with avehicle-treated control. (J) Immunohistochemistry for BRD4,MYC, andKi67 ofa representative tumor of a dBET1-treated and a control-treated mouse(quantification of three independent areas in fig. S8). (K) Percentage ofmCherry+ leukemic cells (means T SEM) in flushed bone marrow fromdisseminated MV4;11 xenografts after daily treatment with dBET1 (n = 8) andJQ1 (n = 8) (both at 63.8 mmol/kg) or formulation control (n = 7) for 19 days.***P = 0.0001 to 0.001; other P values as in Fig. 3 legend.

RESEARCH | REPORTSon A

ugust 2, 2020

http://science.sciencemag.org/

Dow

nloaded from

degradation. Phthalimide conjugation to selec-tive small molecules produces CRBN-dependentposttranslational degradation with exquisitetarget-specific activity. Although this approach isCRBN-dependent, CRBN is ubiquitously expressedin physiologic and pathophysiologic tissues, whichsupports its broad utility in developmental anddisease biology. The increased apoptotic responseof primary AML cells to dBET1, even comparedwith the efficacious tool compound JQ1, highlightsthe potential superiority of BET degradation overBET bromodomain inhibition and prompts consid-eration of therapeutic development. Pharmacologicdestabilization of BRD4 in vivo also resulted inimproved antitumor efficacy in a human leuke-mia xenograft compared with the effects of JQ1.The extension of this approach to new targets(here, FKBP12), lowmolecular mass, high cell per-meability, potent cellular activity, and syntheticsimplicity addresses limitations associated withprior pioneering systems. We anticipate dFKBP1will also serve as a useful tool for the researchcommunity in control of fusion protein stability.Amore general implication of this research is thefeasibility of approaching intractable proteintargets using phthalimide-conjugation of target-binding ligands that may ormay not have target-specific inhibitory activity.

REFERENCES AND NOTES

1. J. B. Bartlett, K. Dredge, A. G. Dalgleish, Nat. Rev. Cancer 4,314–322 (2004).

2. T. Ito et al., Science 327, 1345–1350 (2010).

3. J. Krönke et al., Science 343, 301–305 (2014).4. G. Lu et al., Science 343, 305–309 (2014).5. E. S. Fischer et al., Nature 512, 49–53 (2014).6. V. Lallemand-Breitenbach et al., J. Exp. Med. 193, 1361–1372

(2001).7. Y.-L. Wu et al., Mol. Cell 18, 413–424 (2005).8. M. R. Pratt, E. C. Schwartz, T. W. Muir, Proc. Natl. Acad. Sci.

U.S.A. 104, 11209–11214 (2007).9. L. A. Banaszynski, L. C. Chen, L. A. Maynard-Smith,

A. G. L. Ooi, T. J. Wandless, Cell 126, 995–1004(2006).

10. L. A. Banaszynski, M. A. Sellmyer, C. H. Contag, T. J. Wandless,S. H. Thorne, Nat. Med. 14, 1123–1127 (2008).

11. T. K. Neklesa et al., Nat. Chem. Biol. 7, 538–543(2011).

12. K. M. Sakamoto et al., Proc. Natl. Acad. Sci. U.S.A. 98,8554–8559 (2001).

13. J. S. Schneekloth Jr. et al., J. Am. Chem. Soc. 126, 3748–3754(2004).

14. J. Hines, J. D. Gough, T. W. Corson, C. M. Crews, Proc. Natl.Acad. Sci. U.S.A. 110, 8942–8947 (2013).

15. Y. Itoh, M. Ishikawa, M. Naito, Y. Hashimoto, J. Am. Chem. Soc.132, 5820–5826 (2010).

16. P. Filippakopoulos, S. Knapp, Nat. Rev. Drug Discov. 13, 337–356(2014).

17. P. Filippakopoulos et al., Nature 468, 1067–1073(2010).

18. B. Chapuy et al., Cancer Cell 24, 777–790(2013).

19. J. E. Delmore et al., Cell 146, 904–917(2011).

20. J. Lovén et al., Cell 153, 320–334 (2013).21. J. Zuber et al., Nature 478, 524–528

(2011).22. E. Nicodeme et al., Nature 468, 1119–1123

(2010).23. P. Anand et al., Cell 154, 569–582 (2013).24. J. D. Brown et al., Mol. Cell 56, 219–231 (2014).25. L. Anders et al., Nat. Biotechnol. 32, 92–96

(2014).

26. T. A. Soucy et al., Nature 458, 732–736(2009).

27. R. I. Sufan, M. Ohh, Neoplasia 8, 956–963(2006).

28. E. L. Huttlin et al., Cell 143, 1174–1189(2010).

29. J. A. Mertz et al., Proc. Natl. Acad. Sci. U.S.A. 108,16669–16674 (2011).

ACKNOWLEDGMENTS

We thank W. Kaelin, S. Orkin, and R. Mazitschek for engagingdiscussions and W. Kaelin for cellular reagents; S.-H. Seo andS. Deangelo for assistance in the purification of BRD4; C. Ottfor help with in vivo model studies; and N. Thoma for CRBNexpression reagents. This research was supported by generousphilanthropic gifts from Marc Cohen and Alain Cohen, theWilliam Lawrence and Blanche Hughes Foundation, and theNIH (R01-CA176745 and P01-CA066996 to J.E.B.). G.E.W. issupported by an EMBO long-term fellowship. D.L.B. is a MerckFellow of the Damon Runyon Cancer Research Foundation(DRG-2196-14). Atomic coordinates and structure factors havebeen deposited in the Protein Data Bank (accession code 4ZC9).Quantitative proteomics studies were performed by R. Kunz ofthe Thermo Fisher Scientific Center for Multiplexed Proteomics atHarvard Medical School. Dana-Farber Cancer Institute has filedpatent applications (62/096,318; 62/128,457; 62/149,170) thatinclude the dBET and dFKBP compositions described in thismanuscript. J.E.B. is a Founder of Tensha Therapeutics, abiotechnology company that develops druglike derivatives of JQ1as investigational cancer therapies.

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/348/6241/1376/suppl/DC1Materials and MethodsFigs. S1 to S8Tables S1 to S3References (30–36)

17 March 2015; accepted 6 May 2015Published online 21 May 2015;10.1126/science.aab1433

SCIENCE sciencemag.org 19 JUNE 2015 • VOL 348 ISSUE 6241 1381

RESEARCH | REPORTSon A

ugust 2, 2020

http://science.sciencemag.org/

Dow

nloaded from

Phthalimide conjugation as a strategy for in vivo target protein degradation

BradnerGeorg E. Winter, Dennis L. Buckley, Joshiawa Paulk, Justin M. Roberts, Amanda Souza, Sirano Dhe-Paganon and James E.

originally published online May 21, 2015DOI: 10.1126/science.aab1433 (6241), 1376-1381.348Science 

, this issue p. 1376Sciencethe progression of leukemia in mice.demonstrated selective degradation of a transcriptional coactivator called bromodomain-containing protein 4 and delayed enzyme whose function is to direct proteins to the cell's protein degradation machinery. In a proof-of-concept study, theystrategy involves chemically attaching a ligand known to bind the desired protein to another molecule that hijacks an

devised a chemical strategy that, in principle, permits the selective degradation of any protein of interest. Theet al.Certain classes of proteins that contribute to cancer development are challenging to target therapeutically. Winter

A degrading game plan for cancer therapy

ARTICLE TOOLS http://science.sciencemag.org/content/348/6241/1376

MATERIALSSUPPLEMENTARY http://science.sciencemag.org/content/suppl/2015/05/20/science.aab1433.DC1

REFERENCES

http://science.sciencemag.org/content/348/6241/1376#BIBLThis article cites 36 articles, 8 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.ScienceScience, 1200 New York Avenue NW, Washington, DC 20005. The title (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement ofScience

Copyright © 2015, American Association for the Advancement of Science

on August 2, 2020

http://science.sciencem

ag.org/D

ownloaded from