Avadomide Induces Degradation of ZMYM2 Fusion Oncoproteins in Hematologic Malignancies

Research Article

Aline Renneville1,2,3, Jessica A. Gasser1,2, Daniel E. Grinshpun1,2, Pierre M. Jean Beltran1, Namrata D. Udeshi1, Mary E. Matyskiela4, Thomas Clayton4, Marie McConkey2, Kaushik Viswanathan2, Alexander Tepper1,2, Andrew A. Guirguis1,2,5, Rob S. Sellar2,6, Sophie Cotteret7, Christophe Marzac7, Véronique Saada7, Stéphane De Botton8, Jean-Jacques Kiladjian9, Jean-Michel Cayuela10, Mark Rolfe4, Philip P. Chamberlain4, Steven A. Carr1, and Benjamin L. Ebert1,2,11

Downloaded from https://bloodcancerdiscov.aacrjournals.org by guest on September 26, 2021. Copyright 2021 American Copyright 2021 by AssociationAmerican for Association Cancer Research. for Cancer Research. abstract Thalidomide analogues exert their therapeutic effects by binding to the CRL4CRBN E3 ubiquitin ligase, promoting ubiquitination and subsequent proteasomal degradation of specific protein substrates. Drug-induced degradation of IKZF1 and IKZF3 in B-cell malignancies demonstrates the clinical utility of targeting disease-relevant transcription factors for degradation. Here, we found that avadomide (CC-122) induces CRBN-dependent ubiquitination and proteasomal degradation of ZMYM2 (ZNF198), a transcription factor involved in balanced chromosomal rearrangements with FGFR1 and FLT3 in aggressive forms of hematologic malignancies. The minimal drug- responsive element of ZMYM2 is a zinc-chelating MYM domain and is contained in the N-terminal portion of ZMYM2 that is universally included in the derived fusion proteins. We demonstrate that avadomide has the ability to induce proteasomal degradation of ZMYM2–FGFR1 and ZMYM2–FLT3 chimeric oncoproteins, both in vitro and in vivo. Our findings suggest that patients with hematologic malignancies harboring these ZMYM2 fusion proteins may benefit from avadomide treatment.

Significance: We extend the potential clinical scope of thalidomide analogues by the identification of a novel avadomide-dependent CRL4CRBN substrate, ZMYM2. Avadomide induces ubiquitination and degradation of ZMYM2–FGFR1 and ZMYM2–FLT3, two chimeric oncoproteins involved in hematologic malignancies, providing a proof of concept for drug-induced degradation of transcription factor fusion proteins by thalidomide analogues.

INTRODUCTION IKZF3, accounting for their therapeutic efficacy in multiple myeloma due to addiction of certain B-cell malignancies to Thalidomide and its analogues, including lenalidomide these transcription factors (2, 4, 9). Lenalidomide-induced and pomalidomide, have demonstrated extraordinary clinical degradation of casein kinase 1 alpha (Ck1α) explains the utility mediated by drug-induced targeted protein degrada- clinical efficacy of lenalidomide in myelodysplastic syn- tion. Thalidomide analogues bind the Cullin-RING E3 ubiq- drome with deletion of chromosome 5q [del(5q); ref. 3]. uitin ligase CUL4–RBX1–DDB1–CRBN (CRL4CRBN) complex, Avadomide (CC-122) is a novel thalidomide analogue with promoting ubiquitination and subsequent proteasomal deg- potent antitumor and immunomodulatory activities, par- radation of nonnative target proteins, called neosubstrates ticularly in diffuse large B-cell lymphoma, through degrada- (1–5). Thalidomide analogues bind a shallow hydrophobic tion of IKZF1/3 (10). Phase I trials revealed that avadomide pocket on the surface of CRBN to create a compound– monotherapy has acceptable safety and favorable pharma- protein interface for substrate recruitment to the CRL4CRBN cokinetics (10–12). complex, thereby acting as a “molecular glue” (6–8). Using quantitative mass spectrometry–based proteomic Thalidomide, lenalidomide, pomalidomide, and other profiling, we found decreased ZMYM2 abundance following thalidomide analogues induce degradation of IKZF1 and avadomide treatment. ZMYM2, formerly known as ZNF198, is a zinc finger protein that may function as part of a transcriptional corepressor complex (13). ZMYM2 has long 1Broad Institute of MIT and Harvard, Cambridge, Massachusetts. 2Depart- been known for its implication in a chimeric fusion gene ment of Medical Oncology, Dana-Farber Cancer Institute, Boston, with fibroblast growth factor receptor-1 FGFR1( ), generated Massachusetts. 3INSERM U1287, Gustave Roussy Cancer Campus, Villejuif, by a t(8;13)(p11;q12) chromosomal translocation (14–19). France. 4Celgene/Bristol-Myers Squibb Corporation, San Diego, California. ZMYM2 is the most frequent FGFR1 partner gene, account- 5 6 Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia. Depart- ing for 40% to 50% of cases of FGFR1-rearranged myeloid/ ment of Haematology, UCL Cancer Institute, London, United Kingdom. 7Département de Biologie et Pathologie, Gustave Roussy Cancer Campus, lymphoid neoplasms (18, 20), a distinct entity in the 2016 Villejuif, France. 8Département d’Hématologie, Gustave Roussy Cancer revision of the WHO classification of myeloid neoplasms Campus, Villejuif, France. 9Université de Paris, AP-HP, Hôpital Saint-Louis, and acute leukemia (21). More recently, another fusion tran- Centre d’Investigations Cliniques CIC 1427, INSERM, Paris, France. script involving ZMYM2, ZMYM2–FLT3, resulting from a 10Hematology Laboratory and EA3518, University Hospital Saint-Louis, Université de Paris, Paris, France. 11Howard Hughes Medical Institute, cytogenetically cryptic inversion on chromosome 13, has Dana-Farber Cancer Institute, Boston, Massachusetts. been identified by RNA sequencing in patients with myelo- Note: Supplementary data for this article are available at Blood Cancer proliferative neoplasms with eosinophilia (22) and Philadel- Discovery Online (https://bloodcancerdiscov.aacrjournals.org/). phia chromosome–like acute lymphoblastic leukemia (23). In Corresponding Author: Benjamin L. Ebert, Dana-Farber Cancer Institute, this study, we report for the first time that ZMYM2–FGFR1 450 Brookline Avenue, D1610A, Boston, MA 02215. Phone: 617–632– and ZMYM2–FLT3 oncoproteins are avadomide-dependent 1902; E-mail: [email protected] CRBN neosubstrates, suggesting patients with hematologic Blood Cancer Discov 2021;2:1–16 malignancies harboring these ZMYM2 rearrangements may doi: 10.1158/2643-3230.BCD-20-0105 benefit from treatment with avadomide or novel thalidomide ©2021 American Association for Cancer Research. analogues that selectively degrade ZMYM2.

MAY 2021 blood CANCER DISCOVERY | OF2

RESULTS Fig. S2I), consistent with a nontranscriptional mechanism of regulation. Avadomide Decreases ZMYM2 Protein Level To identify novel targets of avadomide, we profiled Avadomide-Induced Ubiquitination of ZMYM2 changes in protein abundance in Hep3B cells, a hepato- Depends on CRBN cellular carcinoma cell line, treated with varying concentrations of avadomide, pomalidomide, or vehicle control To investigate the mechanism of avadomide-mediated (Supplementary Fig. S1). Samples were analyzed by liquid decrease in ZMYM2 protein level, we treated Hep3B cells chromatography/tandem mass spectrometry (LC/MS-MS) with avadomide or the vehicle control for 12 hours in the employing isobaric chemical labeling and multiplexing presence or absence of small-molecule inhibitors of the using tandem mass tag (TMT) reagents for precise rela- ubiquitin-proteasome system. Cotreatment with avadomide tive quantification and offline fractionation for deep-scale and the E1 (UBA1) inhibitor MLN7243, which blocks all cellular ubiquitination by inhibiting the initial step of the protein analysis. Of the ∼10,000 proteins quantified in Hep3B cells, 8 had decreased abundance following avado- ubiquitination cascade, abrogated ZMYM2 depletion. Like- mide treatment relative to vehicle control, with a log fold wise, addition of MLN4924, a NEDD8-activating enzyme 2 inhibitor that interferes with the activity of Cullin-RING change < –0.8 and a Padj value < 0.01 (Fig. 1A; Supplemen- tary Fig. S2A–S2C; Supplementary Table S1). Among those E3 ubiquitin ligases, or the proteasome inhibitor MG-132, significant hits were ZFP91, RAB28, and WIZ, previously prevented ZMYM2 from being degraded in the presence of identified as targets of thalidomide analogues (2, 24, 25), avadomide (Fig. 1G). and two other proteins not previously published as drug- We next sought to determine whether ubiquitination and induced CRL4CRBN substrates, PDE6D and ZMYM2. Given degradation of ZMYM2 was dependent on CRBN, the molec- the implication of ZMYM2 in two fusion oncoproteins ular target of thalidomide analogues. We utilized CRISPR/ involved in hematologic malignancies, we decided to prior- Cas9 gene editing in Hep3B cells to generate biallelic inacti- itize this target for validation and follow-up studies. Poma- vation of CRBN. In CRBN-knockout Hep3B cells, avadomide lidomide, whose chemical structure is closely related to that did not induce degradation of ZMYM2 (Fig. 1H). We next examined whether ZMYM2 binds CRBN and is ubiquit- of avadomide, had a less pronounced effect on ZMYM2 CRBN protein abundance (Fig. 1B). inated by CRL4 E3 ubiquitin ligase. We observed coim- To validate the proteomic findings, we performed immu- munoprecipitation of endogenous CRBN with V5-tagged noblot analysis in Hep3B cells and tested a series of tha- ZMYM2 only in the presence of avadomide (Fig. 1I). Analysis lidomide analogues. Avadomide was the only drug tested of V5-tagged ZMYM2 in Hep3B cells with an antibody spe- that robustly decreased ZMYM2 endogenous protein level. cific to K48-linked ubiquitin chains showed that avadomide Pomalidomide had a relatively minor effect; lenalidomide, treatment was associated with an increase in ZMYM2 carry- iberdomide (CC-220), and CC-885 had no detectable effect ing K48-linked polyubiquitin chains (Fig. 1J), the canonical on ZMYM2 protein levels (Fig. 1C). Given the relevance of signal for proteasomal degradation. Collectively, these results ZMYM2 in hematologic malignancies, we confirmed these indicate that avadomide treatment causes CRBN-dependent findings in the acute myeloid leukemia cell line TF-1 (Sup- ubiquitination and subsequent proteasomal degradation plementary Fig. S2D). We next showed that avadomide of ZMYM2. decreased ZMYM2 protein abundance in a dose-dependent manner in multiple cellular contexts (Fig. 1D; Supplemen- ZMYM2 Degron Contains a C4 Motif tary Fig. S2E and S2F). Furthermore, a time-course experi- Thalidomide analogues in complex with CRL4CRBN rec- ment revealed that the degradation kinetics of ZMYM2 ognize substrates through degrons, short stretches of pri- was similar to ZFP91 in Hep3B cells (Fig. 1E), but relatively mary sequence that are necessary and sufficient to mediate slower in TF-1 and JURKAT cells (Supplementary Fig. S2G degradation (26). Many CRBN neosubstrates reported to and S2H). Avadomide decreased ZMYM2 protein levels with- date belong to the family of C2H2 zinc finger proteins, for out decreasing ZMYM2 mRNA levels (Fig. 1F; Supplementary which the drug-inducible degrons are zinc finger domains.

Figure 1. Avadomide promotes ZMYM2 degradation in a CRBN- and proteasome-dependent manner. A and B, Volcano plot illustrating changes in protein abundance assessed by proteomic analysis in Hep3B cells after 12 hours of treatment with 10 μmol/L avadomide (A) or 24 hours of treatment with 1 μmol/L pomalidomide (B). The plots represent statistical significance versus the fold change in protein abundance relative to the DMSO vehicle control. Results from two biological replicates are shown. Average log2 fold change in ZMYM2 protein abundance: –0.858, Padj value: 0.002 (A); –0.45, Padj value: 0.043 (B). C, Immunoblot analysis of ZMYM2, ZFP91, and GSPT1 protein levels after 24-hour treatment with DMSO, 1 μmol/L lenalidomide (LEN), 1 μmol/L pomalidomide (POM), 1 μmol/L avadomide (AVA), 1 μmol/L iberdomide (IBER), or 0.005 μmol/L CC-885 in Hep3B cells. ZFP91 and GSPT1 are shown as positive controls for degradation. D, Immunoblot analysis of ZMYM2 protein level in Hep3B cells treated with the indicated concentrations of avadomide for 24 hours. E, Time course of avadomide treatment in Hep3B cells for ZMYM2 protein level. F, Time course of avadomide treatment in Hep3B cells for ZMYM2 mRNA levels assessed by real-time qPCR. Data are mean ± SD, n = 3 technical replicates. G, Immunoblot analysis of ZMYM2 protein levels in Hep3B cells treated for 12 hours with DMSO or 20 μmol/L avadomide alone or in the presence of 0.5 μmol/L MLN7243, 5 μmol/L MLN4924, or 2 μmol/L MG-132. H, Immunoblot analysis of ZMYM2 protein level in wild-type or CRBN knockout Hep3B cells treated with avadomide for 24 hours. I, Immunoprecipitation of ZMYM2-V5 in Hep3B cells treated for 3 hours with DMSO or 10 μmol/L avadomide in the presence of 10 μmol/L MG-132. J, Ubiquitination analysis of ZMYM2-V5 in Hep3B cells treated with DMSO or the indicated concentrations of avadomide for 6 hours in the presence of 10 μmol/L MG-132. EV, empty vector; IP, immunoprecipitated. Results in C–J are representative of three independent experiments.

OF3 | blood CANCER DISCOVERY MAY 2021 AACRJournals.org

A Avadomide B Pomalidomide

DTWD1 ZFP91 7.5 WIZ 7.5 MVP ZFP91 RAB28 ZMYM2 WIZ

) PDE6D ) RAB28

( P 5.0 ( P 5.0 FIZ1 10 10

− Log − Log ZMYM2 2.5 2.5

0.0 0.0 −202 −202 Log2 (fold change relative to DMSO sample) Log2 (fold change relative to DMSO sample)

C DE DMSOLEN POM AVA IBER CC-885 12 h 24 h 48 h 72 h 11 110.005 (µmol/L) 0 0.01 0.1 110Avadomide 0110 0110 0110 0110 (µmol/L) ZMYM2 ZMYM2

ZFP91 ZFP91

GSTP1 β-Actin

β-Actin G ∅ MLN7243 MLN4924 MG-132 F 0200200200 20 Avadomide 1.5 (µmol/L) ZMYM2

1.0 DMSO CRBN

0.5 Avadomide 1 µmol/L

Avadomide 10 µmol/L β-Actin 0.0 Relative ZMYM2 mRNA levels Relative 12 h 24 h 48 h 72 h HIJ EV ZMYM2-V5 EV ZMYM2-V5 Avadomide 0 0.01 0.1 110A0 0.01 0.1 110 vadomide 0020 Avadomide 10100 (µmol/L) (µmol/L) (µmol/L) ZMYM2 V5 V5 IP

CRBN CRBN IP

V5 K-48 linkage β-Actin polyubiquitin 250 kDa Input CRBN Hep3B wild-type Hep3B CRBN−/−

Tubulin V5 Input

Vinculin

MAY 2021 blood CANCER DISCOVERY | OF4

A conserved feature of this C2H2 zinc finger degrome is Structural Modeling of CRBN Bound to the canonical CXXCG motif (27), including a key glycine Avadomide and ZMYM2 Degron residue that engages the phthalimide moiety of thalidomide To validate our findings and gain insight into the role of analogues (8, 27, 28). MYM-type zinc finger proteins, such specific amino acid residues within the ZMYM2 degron, we as ZMYM2, are characterized by a conserved MYM domain built a model of ZMYM2 recruitment to the DDB1–CRBN with the consensus motif CX2CX19–22CX3CX13–19CX2CX19–25 complex using in silico structural modeling. The model FCX3CX3[F/Y] (15). By analogy to the C2H2 zinc finger revealed the predicted ZMYM2 domain (amino acids 421– proteins, we examined whether the CXXCG motif was also 453) is highly compatible with the surface of CRBN and present in ZMYM2. This motif was found twice in the entire binds in a cleft between the N-terminal Lon-like domain and ZMYM2 protein, both located in the second MYM domain the C-terminal thalidomide-binding domain of CRBN (6, 7). (MYM2; Fig. 2A). To test whether the MYM2 domain con- The ZMYM2 β-hairpin was predicted to bind CRBN through tained the degron, we employed a lentiviral degradation the structural degron reported for other neosubstrates, con- reporter vector that enables comparison of the fluorescence sistent with our deletion mapping and mutagenesis studies of enhanced GFP (EGFP)–tagged ZMYM2 construct to (8, 27, 28). This binding mode places the critical glycine mCherry using flow cytometry (Fig. 2B; ref. 27). This assay residue G429 in the conserved position against the bound revealed that the MYM2 domain (amino acids 425–502) avadomide molecule. The G429A mutant should not be was indeed sufficient to allow degradation in cells treated tolerated in this binding mode, consistent with our degra- with avadomide. The MYM2 N-terminal portion (amino dation reporter and in vitro ubiquitination results. Of the acids 425–466) similarly mediated degradation, whereas other amino acid residues found to be critical for ZMYM2 the C-terminal portion (amino acids 467–502) did not con- degradation in our functional studies, R424 is proximal to tribute. Further deletion mapping led to the identification avadomide in the ternary complex, and I427 makes direct of a ZMYM2 degron that was more sensitive to avadomide contacts with the CRBN surface (Fig. 3). Therefore, both of (amino acids 423–466), with a dose response almost overlap- these residues likely contribute to stabilization of the ternary ping with that of the IKZF1/3 degron (Fig. 2C). We next ana- complex needed for efficient ubiquitin transfer. Overall, this lyzed the effect of different thalidomide analogues on the model confirms the interaction between the MYM2 domain degradation of the ZMYM2 423–466 degron in the reporter of ZMYM2 and avadomide-bound CRBN. The model ration- vector. Only avadomide and pomalidomide induced effi- alizes the importance of the three functionally identified cient degradation, while the other drugs had little or no nonstructural amino acid residues in the ZMYM2 degron for effect, even at high concentrations (Fig. 2D; Supplementary CRBN binding and degradation. Fig. S3). To determine which specific amino acid residues are critical for ZMYM2 degradation, we performed an alanine Avadomide Promotes ZMYM2–FGFR1 and scan, systematically mutating each amino acid within the ZMYM2–FLT3 Fusion Protein Degradation 423–466 degron to alanine. Hep3B and TF-1 cells stably ZMYM2 is implicated in two chimeric fusion oncopro- expressing the respective mutant degron in the degradation teins involved in hematologic malignancies. In both fusion reporter vector were treated with avadomide or the vehicle proteins, the N-terminal portion of ZMYM2, including the control. The alanine scan highlighted seven residues criti- drug-responsive degron, is fused to the C-terminal por- cal for degradation in Hep3B and TF-1 cells, including four tion of either FGFR1 or FLT3, including their respective zinc-coordinating cysteine residues and three nonstructural tyrosine kinase domains (Fig. 4A; refs. 14, 15, 22, 23). No amino acid residues, R424, I427, and the key glycine G429 cancer cell lines harboring ZMYM2–FGFR1 or ZMYM2– (Fig. 2E). Of note, the ZMYM2 RCXICG motif is not found FLT3 rearrangements are documented. To address whether in any other MYM-type zinc finger proteins (https://prosite. the fusion proteins could be degraded in cells in a manner expasy.org/scanprosite/), in accordance with the finding similar to the full-length ZMYM2 protein, we stably overex- that other proteins in this family are not degraded upon pressed the V5-tagged fusion proteins in the leukemia line avadomide treatment in our LC/MS-MS data (Supplemen- TF-1 and treated these cells with avadomide for 24 hours. tary Table S2). Immunoblot analysis using antibodies against the ZMYM2 To confirm that this minimal degron is sufficient for N-terminus, FGFR1 or FLT3 C-terminus, and C-terminal drug-induced ubiquitination by the CRL4CRBN, we performed V5 tag consistently showed that the ZMYM2–FGFR1 and in vitro ubiquitination assays. Indeed, we observed that the ZMYM2–FLT3 proteins were both degraded upon avado- ZMYM2 423–466 degron can be ubiquitinated by CRL4CRBN mide treatment in a dose-dependent manner. The G429A in the presence of avadomide or pomalidomide, demon- point mutation within the ZMYM2 degron rendered both strating that ZMYM2 is a primary target of CRL4CRBN. As fusion proteins resistant to degradation, indicating that expected, the G429A mutant completely prevented ubiqui- ZMYM2–FGFR1 and ZMYM2–FLT3 are degraded through tination (Fig. 2F). Immunoblot validation in Hep3B cells the same mechanism as the full-length ZMYM2 protein confirmed that deletion of the MYM2 domain or introduc- (Fig. 4B and C). tion of the G429A point mutation in full-length ZMYM2 Ba/F3 is a murine lymphoid leukemia cell line dependent protein conferred resistance to avadomide-induced degrada- on IL3, and its removal induces cell apoptosis. Like other tion (Fig. 2G). Collectively, these data demonstrate that the tyrosine kinase oncogenes (29), ZMYM2–FGFR1 transforms MYM2 domain is necessary and sufficient for drug-induced Ba/F3 cells to IL3 independence (30, 31), thereby rendering degradation of ZMYM2. Ba/F3 cells dependent on the fusion oncoprotein for their

OF5 | blood CANCER DISCOVERY MAY 2021 AACRJournals.org

A B

ZMYM2 protein segments 430 440 450 460 470 480 490 500 ZMYM2EGFP IRES mCherry 425–502 CTICGKLTEIRHEVSFKNMTHKLCSDHCFNRYRMANGLIMNCCEQCGEYLPSKGAGNNVLVIDGQQKRFCCQSCVSEY 467–502 CEQCGEYLPSKGAGNNVLVIDGQQKRFCCQSCVSEY 425–466 CTICGKLTEIRHEVSFKNMTHKLCSDHCFNRYRMANGLIMNC ZMYM2 degron 425–452 CTICGKLTEIRHEVSFKNMTHKLCSDHC 423–466 SRCTICGKLTEIRHEVSFKNMTHKLCSDHCFNRYRMANGLIMNC reporter

C D

1.2 1.2

1.0 1.0 IC [ mol/L] ZMYM2 425–502 (78 AA) Drug 50 µ 0.8 ZMYM2 425–466 (42 AA) 0.8 CC-885 n.r. 0.6 ZMYM2 467–502 (36 AA) 0.6 Iberdomide n.r. ZMYM2 425–452 (28 AA) Lenalidomide 3.0 0.4 0.4 ZMYM2 423–466 (44 AA) Pomalidomide 0.0052 0.2 IKZF3 ZF2 146–168 (23 AA) 0.2 Avadomide 0.0044 EGFP/mCherry ratio EGFP/mCherry ratio 0.0 0.0 −4 −3 −2 −10 1 −4 −3 −2 −10 1

Log10 ([avadomide] in µmol/L) Log10 ([drug] in µmol/L) E 1.2

1.0

0.8

0.6 Hep3B TF-1 0.4

EGFP/mCherry ratio 0.2

0.0

I427A I434A I463A S423AR424AC425AT426A C428AG429AK430AL431AT432AE433A R435AH436AE437AV438AS439AF440AK441AN442AM443AT444AH445AK446AL447AC448AS449AD450AH451AC452AF453AN454AR455AY456AR457AM458AA459RN460AG461AL462A M464AN465AC466A Amino acid changes

FG Wild-type G429A Empty del 10 mol/L pomalidomide µ − + − − + − ZMYM2-V5 vector Wild-type 425–502 G429A 10 µmol/L avadomide − − + − − + 10 µmol/L avadomide −+ −+−+−+

Polyubiquitinated V5 ZMYM2 degron

ZMYM2 AA 423–466 ZFP91

β-Actin

Figure 2. Identification of the ZMYM2 degron and amino acids critical for drug-induced degradation. A, Sequence alignment of ZMYM2 MYM2 domain and the different peptides tested for degradation using the reporter assay. The cysteine residues that coordinate the zinc ions are highlighted in gray. The glycine residues following a CXXC motif are highlighted in yellow. B, Schematic of the protein degradation reporter vector used to identify the ZMYM2 degron. IRES, internal ribosome entry site. C, Hep3B cells stably expressing the ZMYM2 constructs in the degradation reporter were treated for 20 hours with DMSO or avadomide and analyzed by flow cytometry to quantify the DMSO-normalized ratio of EGFP/mCherry fluorescence. The dose– response curve for IKZF1/3 zinc finger 2 (ZF2) domain is represented as a reference (mean± SD, n = 4 biological replicates). AA, amino acid. D, Effect of different thalidomide analogues on the degradation of the ZMYM2 423–466 wild-type degron in the reporter vector in Hep3B cells (mean ± SD, n = 4 biological replicates). IC50, inhibitory concentration 50; n.r., not reached. E, Alanine scan of the ZMYM2 423–466 degron using the reporter assay in Hep3B and TF-1 cells treated with 1 μmol/L avadomide for 20 hours. Amino acids that reduced degradation above 0.4 when mutated into an alanine are in red font (mean ± SD, n = 3 biological replicates). F, MBP-tagged ZMYM2 amino acids 423–466 wild-type or G429A were tested for in vitro ubiquitination by CRBN–CRL4 in the absence or presence of 10 μmol/L pomalidomide or avadomide. Ubiquitination reactions were separated by SDS-PAGE, followed by anti-MBP. G, Immunoblot validation of the ZMYM2 degron in Hep3B cells transfected with the empty vector, ZMYM2 wild-type, ZMYM2 with deletion of the MYM2 domain (del 425–502), or ZMYM2-binding mutant G429A and treated with DMSO or 10 μmol/L avadomide for 24 hours. Results in C–G are representative of three independent experiments.

MAY 2021 blood CANCER DISCOVERY | OF6

R424 R424 G429 G429

I427 I427

90°

Figure 3. Structural model of ZMYM2 recruitment to DDB1–CRBN bound to avadomide. A model of ZMYM2 amino acids 421–453 (blue), with the positions of G429 highlighted in red and R424 and I427 in orange, shown bound to the complex of DDB1 (purple), CRBN (green), and avadomide (yellow). Zinc ions are shown as gray spheres. The glycine-containing β-hairpin degron of ZMYM2 was docked on the basis of the structures of other glycine- containing degron substrates recruited to CRBN (8, 27, 28). survival and proliferation. We transformed Ba/F3 cells with transformed by either ZMYM2WT–FGFR1 or ZMYM2G429A– ZMYM2–FGFR1 bearing the wild-type degron (ZMYM2WT– FGFR1 were all sensitive to ponatinib, a multityrosine kinase FGFR1) or its G429A mutant counterpart (ZMYM2G429A– inhibitor with activity against FGFR1 (33), regardless of CRBN FGFR1; Fig. 5A and B). Because mouse cells are intrinsically status (Fig. 5F). resistant to thalidomide and its analogues, transformed Ba/F3 cells were subsequently transduced to express either In Vivo Validation of ZMYM2–FGFR1 Degradation human CRBN or the humanized mouse CrbnI391V mutant, by Avadomide which are both capable of mediating drug-induced degra- To address whether avadomide could inhibit the prolif- dation of Ikzf1/3, Ck1α, and Zfp91 in murine cells (3, 32). eration of ZMYM2–FGFR1-dependent leukemia cells in vivo, Expression of human CRBN allowed robust degradation we performed a syngeneic transplant experiment in BALB/c of ZMYM2WT–FGFR1 upon avadomide treatment. However, mouse recipients (34, 35), the mouse strain from which Ba/ expression of CrbnI391V was not sufficient to promote deg- F3 cells were originally derived. As previously described with radation of human ZMYM2WT–FGFR1 (Fig. 5C), consistent BCR–ABL-transformed BaF3 cells (34, 35), we found that with the lack of cytotoxic effects of avadomide in these cells injection of ZMYM2–FGFR1-transformed Ba/F3 cells into (Supplementary Fig. S4). BALB/c mice caused a highly aggressive disease, with all Ba/F3 transformation by ZMYM2–FGFR1 is mediated, mice dying secondary to a diffuse leukemic process within at least in part, by activation or subversion of signaling 2 weeks after transplant. To assess the effect of avadomide in pathways employed by the Jak–Stat pathway downstream of this model, sublethally irradiated BALB/c mice were injected activated receptor tyrosine kinase signaling (30). Immunob- with ZMYM2WT–FGFR1-transformed Ba/F3 cells expressing lot analysis showed that Stat1 and, to a lesser extent, Stat5 human CRBN (with coexpression of GFP) and dosed twice were phosphorylated in ZMYM2–FGFR1-transformed Ba/ per day with avadomide or the vehicle control for 7 con- F3 cells. In Ba/F3 cells transformed by ZMYM2WT–FGFR1 secutive days (Fig. 6A) Spleen weight, which reflects tumor and expressing human CRBN, avadomide treatment led to burden, was significantly lower in mice who received avado- ZMYM2WT–FGFR1 degradation and a concomitant decrease mide compared with those who received the vehicle control in phospho-Stat1 and Stat5 levels (Fig. 5D). In Ba/F3 cells (Fig. 6B). Compared with the vehicle control, avadomide transformed with the drug-resistant ZMYM2G429A–FGFR1 treatment was associated with a lower white blood cell count fusion, the effects of avadomide on phospho-Stat1 and and a higher hemoglobin level, even though blood count Stat5 were absent or decreased (Fig. 5E). In Ba/F3 cells values at baseline did not differ between the two groups transformed by ZMYM2WT–FGFR1 and expressing human (Fig. 6C–E). In addition, the percentages of GFP-positive cells CRBN, avadomide reduced cell viability and proliferation in peripheral blood, bone marrow, and spleen were all strik- in a dose-dependent manner. This effect was completely ingly lower in the avadomide group compared with the con- rescued in Ba/F3 cells transformed by the ZMYM2G429A– trol group (Fig. 6F–H). Altogether, these results demonstrate FGFR1 mutant, confirming that the cytotoxic effect of avado- the efficacy of avadomide to inhibit the growth of ZMYM2– mide was due to ZMYM2WT–FGFR1 depletion. Ba/F3 cells FGFR1-dependent Ba/F3 leukemia cells in vivo.

OF7 | blood CANCER DISCOVERY MAY 2021 AACRJournals.org

Breakpoints A ZMYM2 degron MYM-type zinc fingers NLS

ZMYM2 NH2 MYM1 2 345CP/V L COOH rich 155 kDa 1 423–466 913 1,059 1,377

FGFR1 NH2 lg lg lg TM TK1 TK2 COOH 92 kDa 1 429 822

FLT3 NH2 lg lg lg lg lg TM JM TK1 TK2 COOH 130–160 kDa 1 594 993

ZMYM2–FGFR1 NH2 MYM1 23 45P/V TK1 TK2 COOH 146 kDa rich 1 423–466 1,307

ZMYM2–FLT3 NH2 MYM1 23 45P/V J TK1 TK2 COOH 164 kDa rich M 1 423–466 1,459

B TF-1 EV ZMYM2WT–FGFR1 V5 EV ZMYM2G429A–FGFR1 V5

000.01 0.1 110 000.01 0.1 110Avadomide (µmol/L)

ZMYM2

FGFR1

ZFP91

β-Actin

C TF-1 EV ZMYM2WT–FLT3 V5 EV ZMYM2G429A–FLT3 V5

000.01 0.1 110 000.01 0.1 110 Avadomide (µmol/L)

ZMYM2

FLT3

ZFP91

β-Actin

Figure 4. Avadomide promotes ZMYM2–FGFR1 and ZMYM2–FLT3 fusion protein degradation. A, Schematic representation of ZMYM2, FGFR1, FLT3, and the chimeric fusion proteins ZMYM2–FGFR1 and ZMYM2–FLT3. The indicated breakpoints reported in previous studies (15, 22) have been used to clone both fusion sequences in expression vectors for immunoblot validation. In the context of the ZMYM2–FGFR1 fusion, the ZMYM2 proline/valine (P-V)-rich domain has been shown to mediate oligomerization and tyrosine kinase constitutive activation (31). CL, Cre-like domain; Ig, immunoglobulin-like domain; JM, juxtamembrane domain; NLS, nuclear localization site; TK, tyrosine kinase domain; TM, transmembrane domain. B and C, Immunoblot validation of the chimeric fusion protein degradation in TF-1 cells stably expressing the V5-tagged ZMYM2–FGFR1 (B) or ZMYM2–FLT3 (C) fusion protein, either with the WT or the G429A mutant degron, and treated with avadomide for 24 hours. Results are representative of three independent experiments. EV, empty vector; WT, wild-type.

MAY 2021 blood CANCER DISCOVERY | OF8

AB C Human Crbn Parental (w mIL3) EV/EV CRBN I391V EV pLX304 EV pLX304 (w mIL3) 010 010010 Avadomide WT (µmol/L) ZMYM2 –FGFR1 in pLX304 ZMYM2WT–FGFR1 pLX304 (w/o mIL3) WT ZMYM2G429A–FGFR1 in pLX304 ZMYM2G429A–FGFR1 pLX304 (w/o mIL3) ZMYM2 – FGFR1-V5 ) 6 6 175 ) 6 ( × 10 150 Zfp91 4 125 100 75 2 50 CRBN

le cell number 25 Cell count ( × 10 0 0 Viab 024 68 012 34 β-Actin Days after mIL3 withdrawal Time (days)

D EV ZWT-F1 EV ZWT-F1 E EV ZG429A-F1 EV ZG429A-F1 pLX304 pLX304 pLX304 pLX304 pLX304 pLX304 pLX304 pLX304 +++−−−+++ −−− mIL3 +++−−−+++ −−− 0 1 10 0 1 10 011 10 0 10 Avadomide (µmol/L) 0 1 10 0 1 10 011 10 0 10

ZMYM2– FGFR1 V5

Stat1

pStat1Y701

Stat5

pStat5Y694

Zfp91

CRBN

β-Actin

EV GW Human CRBN GW EV GW Human CRBN GW F

125 125

100 100 ZMYM2WT–FGFR1 pLX304/EV GW WT 75 ZMYM2 –FGFR1 pLX304/human CRBN GW 75 ZMYM2G429A–FGFR1 pLX304/EV GW G429A 50 ZMYM2 –FGFR1 pLX304/human CRBN GW 50

25 25 CellTiter-Glo (% of control) 0 CellTiter-Glo (% of control) 0 −3 −2 −1012 −3 −2 −1012 Log ([avadomide] in µmol/L) 10 Log10 ([ponatinib] in µmol/L)

Figure 5. Avadomide affects viability of ZMYM2–FGFR1-transformed Ba/F3 cells expressing human CRBN. A, Growth curves assessed by Trypan blue staining after mIL3 withdrawal in Ba/F3 cells transduced with lentiviruses expressing the EV, ZMYM2WT–FGFR1, or ZMYM2G429A–FGFR1 (mean ± SD, n = 3–4 biological replicates). B, Growth curves assessed by Trypan blue staining in ZMYM2–FGFR1-transformed Ba/F3 cells in the absence of mIL3, in comparison with Ba/F3 parental and the EV control in the presence of mIL3 (mean ± SD, n = 3–4 biological replicates). C, Immunoblot analysis in Ba/F3 cells transformed by V5-tagged ZMYM2WT–FGFR1 in pLX304 and subsequently transduced with lentiviruses expressing either human CRBN or the humanized mouse CrbnI391V mutant in GW vector. Ba/F3 cells transduced with both EV pLX304 and EV GW are shown as a negative control. Ba/F3 cells were treated with DMSO or 10 μmol/L avadomide for 24 hours. D and E, Immunoblot analysis in Ba/F3 cells transduced with EV pLX304 and EV GW or human CRBN in GW, and Ba/F3 cells transformed by V5-tagged ZMYM2WT–FGFR1 (D) or ZMYM2G429A–FGFR1 (E) in pLX304 and subsequently transduced with EV GW or human CRBN in GW. Ba/F3 cells were treated with DMSO or 1 μmol/L or 10 μmol/L avadomide for 24 hours. F, Cell viability assessed by CellTiter-Glo in Ba/F3 cells transformed by ZMYM2WT–FGFR1 or ZMYM2G429A–FGFR1 and transduced with EV GW or human CRBN in GW. Ba/F3 cells were treated with different concentrations of avadomide or ponatinib for 48 hours (mean ± SD, n = 6 biological replicates). Results in A–F are representative of three independent experiments. EV, empty vector; GW, Gateway; mIL3, murine IL3; WT, wild-type; Z-F1, ZMYM2–FGFR1.

OF9 | blood CANCER DISCOVERY MAY 2021 AACRJournals.org

A ZMYM2WT–FGFR1- B transformed Ba/F3 cells 100 **** 550 50 mg/kg avadomide or vehicle Rads 80 control bi-daily by oral gavage 60 BALB/c mice 40 n = 30 Posttreatment Day 0 Day 3 Drug dosing Day 10 PB sample

Spleen weight (mg) 20 Transplantation Harvest Bone marrow Spleen 0 Pretreatment Vehicle Avadomide PB sample

C DE ns 1.5 *** 20 800 * ns /L) 9 /L) 9 600 Vehicle control 1.0 ns 15 Avadomide 400

0.5 10 ns 200 Hemoglobin (g/dL) WBC count ( × 10 Platelet count ( × 10 0.0 5 0 Before After Before After Before After treatment treatment treatment treatment treatment treatment

FG H Peripheral blood Bone marrow Spleen 80 **** 50 15 **** **** 40 60 10 30 40 cells (%) cells (%) cells (%) + + 20 + 5 GFP GFP 20 GFP 10

0 0 0

Vehicle Avadomide Vehicle Avadomide Before After treatment treatment

Figure 6. Effects of avadomide treatment on ZMYM2–FGFR1-transformed Ba/F3 cells in vivo. A, Experimental design of the mouse experiment. ZMYM2WT–FGFR1-transformed Ba/F3 cells expressing human CRBN were retro-orbitally injected into sublethally irradiated BALB/c mice. n = 15 mice for both treatment groups. PB, peripheral blood. B, Spleen weight at day 10 posttransplantation. White blood cell count (WBC; C), hemoglobin level (D), and platelet count (E) measured in peripheral blood samples collected before and after treatment. F–H, Percentage of GFP-positive cells assessed by flow cytometry in peripheral blood F( ), bone marrow (G), and spleen (H). P values are from an unpaired two-sided t test. ns, nonstatistically significant. ****, P < 0.0001; ***, P < 0.001; *, P < 0.05.

Validation in Primary Human Hematopoietic row from healthy donors as normal controls. CD34+ stem Stem/Progenitor Cells and progenitor cells from healthy donors and patients To verify our findings from experiments with ectopi- with leukemia were expanded in culture and treated with cally expressed fusion proteins in primary samples with 1 μmol/L avadomide or the vehicle control. In normal endogenous expression of ZMYM2 fusions, we collected CD34+ cells, avadomide treatment did not significantly diagnostic bone marrow samples from patients with a affect cell viability or proliferation (Fig. 7A). In contrast, hematologic malignancy harboring a t(8;13)(p11;q12) in bone marrow CD34+ cells isolated from patients with translocation, which generates the ZMYM2–FGFR1 fusion. ZMYM2–FGFR1-positive hematologic malignancy, ava The clinical characteristics of the four patients are sum- domide treatment significantly reducedthe number of marized in Supplementary Table S3. We used bone mar- viable cells compared with the vehicle control (Fig. 7B).

MAY 2021 blood CANCER DISCOVERY | OF10

A Healthy donor # 1 Healthy donor # 2 Healthy donor # 3 ns 40 50 ns 60 ns 50 30 40 40 30 30 20 ns Vehicle control 20 ns 20 ns 1 µmol/L avadomide 10 ns ns 10 ns to vehicle control to vehicle to vehicle control to vehicle 10 control to vehicle 0 0 0 Viable cell number relative Viable Viable cell number relative Viable 0246 024 60 cell number relative Viable 246 Time (days) Time (days) Time (days)

B Patient # 1 Patient # 2 Patient # 3 Patient # 4 (T-ALL) (AML-5) (MPN-Eo) (MPN-Eo) 60 410 **** 25 **** **** **** 8 20 3 40 **** 6 **** 15 2 **** ** 4 **** 10 **** 20 * ns 1 2 5 to vehicle control to vehicle to vehicle control to vehicle to vehicle control to vehicle to vehicle control to vehicle 0 0 0 0 Viable cell number relative Viable Viable cell number relative Viable Viable cell number relative Viable Viable cell number relative Viable 024 602460246 0246 Time (days) Time (days) Time (days) Time (days)

Figure 7. Effects of avadomide treatment on primary human CD34+ hematopoietic stem and progenitor cells ex vivo. Viable cell counts assessed by Trypan blue staining in bone marrow CD34+ cells collected from healthy donors (A) and patients with ZMYM2–FGFR1-positive hematologic malignancy (B) are represented after 2, 4, or 6 days of treatment by 1 μmol/L avadomide or the vehicle control (mean ± SD, n = 4–6 biological replicates). P values are from an unpaired two-sided t test. ns, nonstatistically significant. ****,P < 0.0001; **, P < 0.01; *, P < 0.05. AML-5, acute myeloid leukemia 5 (monoblastic); MPN-Eo, myeloproliferative neoplasm with eosinophilia; T-ALL, T-cell acute lymphoblastic leukemia.

Altogether, these data indicate that avadomide has antipro- degradation (36). Moreover, our findings raise the possibility liferative effects in primary cells from patients with hemato- that other zinc finger proteins containing a C4 motif, such as logic malignancy harboring ZMYM2–FGFR1 rearrangement, the LIM, PHD, or RING-type zinc finger proteins, may also with a broad therapeutic window. be degraded in the presence of new thalidomide analogues, providing new therapeutic opportunities. Avadomide is a novel therapeutic agent that is currently DISCUSSION in phase I/II clinical trials for patients with lymphomas In this study, we established ZMYM2 as a direct avado- given its IKZF1/3 degradation activity. Our data from cell mide-dependent substrate of the CRL4CRBN E3 ubiquitin lines and primary patient samples suggest that avadomide ligase. ZMYM2 is implicated in two chimeric fusion oncopro- could also be repurposed for the treatment of hematologic teins involved in hematologic malignancies, ZMYM2–FGFR1 malignancies with ZMYM2–FGFR1 or ZMYM2–FLT3 rear- and ZMYM2–FLT3. We demonstrated that avadomide has rangements. Chimeric fusion genes are recognized as strong the ability to induce proteasomal degradation of both fusion genetic drivers in hematologic malignancies, sometimes even oncoproteins utilizing the same mechanism as the ZMYM2 pathognomonic for specific clinical subtypes of disease, and full-length protein. We validated these findings in primary often represent the primary event in the leukemic process, cells from patients with hematologic malignancy harboring which makes them attractive targets for diagnosis and treat- ZMYM2–FGFR1 rearrangements. This is the first description ment (21, 37). In immunodeficient mice, ZMYM2–FGFR1 of fusion oncoproteins that can be targeted for degradation alone is capable of initiating a myeloproliferative-like dis- by a thalidomide analogue. order, with increased cellular proliferation, blast accumula- We characterized a novel type of drug-inducible degron tion, bone marrow fibrosis, and increased extramedullary for thalidomide analogues containing a C4 motif, in which hematopoiesis, consistent with features observed in patients the four cysteines represent the zinc-coordinating residues with ZMYM2–FGFR1 neoplasms (17, 38). Given the recent that belong to the consensus motif of MYM-type zinc finger discovery of ZMYM2–FLT3, no data are available regarding domains. The ZMYM2 degron shares similarities with C2H2 its role in the pathogenesis of hematologic malignancies or zinc finger degrons, including the critical glycine residue, its prognostic significance. However, it is likely that this chi- but is differentiated in terms of drug sensitivity. Unlike meric fusion gene similarly drives the malignant phenotype the IKZF1/3 degron, the ZMYM2 degron is not uniformly (22, 23, 39). degraded by active thalidomide analogues. Novel thalido- The prognosis of FGFR1-rearranged myeloid/lymphoid mide analogues have the potential to induce ZMYM2 degra- neoplasms is poor despite intensive chemotherapy. Tyros- dation with greater potency and selectivity, thereby reducing ine kinase inhibitors targeting FGFR1, such as midostaurin the competition with other neosubstrates for CRBN occu- (40), ponatinib (41), or pemigatinib (42), have been tested in pancy, which is a limiting factor for drug-induced protein monotherapy in patients with FGFR1-rearranged myeloid/

OF11 | blood CANCER DISCOVERY MAY 2021 AACRJournals.org

Downloaded from https://bloodcancerdiscov.aacrjournals.org by guest on September 26, 2021. Copyright 2021 American Association for Cancer Research. Avadomide Induces ZMYM2 Fusion Oncoprotein Degradation RESEARCH ARTICLE lymphoid neoplasms, but hematologic response, if any, is conditions, and data analysis are provided in the Supplementary typically partial and transient. Allogeneic stem cell trans- Data. The original mass spectra and the protein sequence database plantation currently remains the only option to achieve long- used for searches have been deposited in the public proteomics term remission and possibly cure these diseases (20, 41). As repository MassIVE (http://massive.ucsd.edu) under the identi- opposed to reversible small-molecule inhibitors that lose effi- fier MSV000086902 and are accessible at ftp://massive.ucsd.edu/ MSV000086902/. cacy as drug concentration drops, protein destruction is able to drive long-lasting target depletion and strong biological Western Immunoblotting effects (9). Moreover, small-molecule degraders and inhibi- Cells were washed with 1× PBS and lysed for 30 minutes on ice tors will have distinct mechanisms of resistance, providing in the following buffer: 150 mmol/L NaCl, 25 mmol/L Tris-HCl pH additional rationale for combining these drugs. Therefore, 8.0, 1% NP-40, 5% glycerol, 0.02 mmol/L ZnCl, 1× Halt Protease, we speculate that, in patients with hematologic malignan- and Phosphatase Inhibitor Cocktail (Thermo Fisher Scientific). The cies with ZMYM2–FGFR1 or ZMYM2–FLT3, a combination lysates were centrifuged at 13,000 × g at 4°C for 15 minutes, and therapy associating a degrader such as avadomide or a novel protein concentrations were quantified with the Pierce BCA Pro- thalidomide analogue that selectively degrades ZMYM2 and tein Assay Kit (Thermo Fisher Scientific). Samples were boiled and a tyrosine kinase inhibitor targeting FGFR1 or FLT3 could reduced in a mixture containing Laemmli (6X, SDS-Sample Buffer, help to achieve deep and durable responses while preventing Reducing; Boston BioProducts) and lysis buffer. Protein lysates were the emergence of drug resistance. run on Bolt 8% Bis-Tris Plus Gels (Thermo Fisher Scientific) by SDS- ZMYM2–FGFR1 and ZMYM2–FLT3 are thought to be PAGE at a constant voltage. Proteins were transferred onto nitro- cellulose membranes (Thermo Fisher Scientific) utilizing a Mini rare genetic abnormalities, but their prevalence might be Trans-Blot Electrophoretic Transfer Cell (Bio-Rad) in 1× Tris-Glycine underestimated, especially as the latter is not detectable Buffer (Bio-Rad) at 40 V overnight. Membranes were blocked in 5% by conventional cytogenetics. Retrospective and prospec- nonfat dry milk in TBST for 1 hour and probed with the indicated tive screening of patient samples using FISH analysis, RNA primary antibody at 4°C overnight. The primary antibodies used are sequencing, or other assays may reveal additional cases that listed in Supplementary Table S4. The membranes were washed in would not have been identified by cytogenetic and molecu- TBST, four times for 5 minutes, before 1-hour incubation with the lar diagnostic tests used routinely at diagnosis but could secondary anti-rabbit or anti-mouse IgG horseradish peroxidase– benefit from targeted therapies. conjugated antibody (Cell Signaling Technology) at room tem- Hundreds of recurrent fusion genes have been reported perature. After four final washes in TBST for 5 minutes, the blots in hematologic malignancies and solid tumors (43). These were subjected to enhanced chemiluminescence detection using the SuperSignal West Dura Extended Duration Substrate (Thermo fusion genes are often powerfully oncogenic and strong Fisher Scientific). Blots were exposed to CL-XPosure films (Thermo genetic dependencies in the resulting malignancies. Few of Fisher Scientific), which were developed using the Medical Film the corresponding fusion proteins can be specifically targeted Processor SRX-101A (Konica Minolta). For reprobing, blots were by small-molecule therapies. The targeting of BCR–ABL and stripped in Restore PLUS Western Blot stripping buffer (Thermo PML–RARA in chronic myeloid leukemia and acute pro- Fisher Scientific) and reblocked. Uncropped immunoblots can be myelocytic leukemia, respectively, has yielded extraordinarily found in Supplementary Fig. S5. successful therapies (37, 43, 44). However, the vast majority of fusion oncoproteins are still considered undruggable. Our RNA Extraction and Real-time Quantitative PCR studies of avadomide-induced degradation of ZMYM2 fusion The RNeasy Plus Mini Kit (Qiagen) was used to isolate mRNA. proteins highlight the broader potential of targeted protein Reverse transcription was performed using the SuperScript IV VILO degradation as a therapeutic strategy for fusion oncoproteins Master Mix (Thermo Fisher Scientific) according to the manufac- in cancer. turer’s instructions. The qRT-PCR analysis was performed using the TaqMan Gene Expression Master Mix and the GAPDH (Hs99999905_ m1) and ZMYM2 (Hs01547045_g1) primer-probe sets from Thermo METHODS Fisher Scientific. Each RT-qPCR was performed in triplicate in a 384-well plate on a QuantStudio 6 Flex System instrument (Thermo

Quantitative Proteomic Profiling by LC/MS-MS Fisher Scientific). The mean threshold cycle C( t) for each assay was Early-passage parental Hep3B cells were treated with DMSO used for further calculations. Relative ZMYM2 gene expression levels vehicle control, 1 μmol/L pomalidomide for 24 hours, or 1 μmol/L were calculated using the ΔΔCt method. or 10 μmol/L avadomide for 12 or 24 hours, with two biological duplicates for all conditions. Cells were harvested by trypsinization, Cloning washed with 1× PBS, and cell pellets were snap-frozen and stored at Empty backbones pDONR223, pLX302, pLX304, pLX306, as well −80°C until use. Proteins were extracted from cell pellets through as pLX311-Cas9, and human CRBN isoform 2 (ccsbBroadEn_08244) lysis in 8 mol/L urea buffer, followed by reduction, alkylation, and in pDONR223, were acquired from the Broad Genetic Perturba- digestion with endoproteinase LysC and trypsin (Supplementary tion Platform. CRBN gRNA1 in pXPR003 and IKZF1/3 amino Fig. S1). Tryptic peptides were labeled using the TMT 11plex acids 146–168 in Cilantro 2 (Addgene #74450) were generated as reagent (Thermo Fisher Scientific) and combined as described described (27, 47). ZMYM2 double-stranded DNA sequences encod- previously (45). TMT-labeled peptides were fractionated by offline ing for amino acids 425–502, 425–466, 467–502, 425–452, or 423–466 basic reverse-phase chromatography into 24 fractions as described and appended with BsmBI restriction sites were in vitro synthesized previously (46). Multiplexed peptide fractions were analyzed by LC/ by IDT, digested with BsmbI (NEB), purified using the QIAquick MS-MS using an Orbitrap Fusion Lumos Mass Spectrometer. MS/ PCR Purification Kit (Qiagen), and ligated into Artichoke (Addgene MS data were processed using Spectrum Mill v7 prerelease (Agilent #73320) reporter vector using the T4 DNA ligase (Thermo Fisher Technologies), and protein quantitative values were further ana- Scientific). For the alanine scan of the ZMYM2 423–466 degron, lyzed using R. Details on proteomic sample processing, LC/MS-MS double-stranded DNA sequences were in vitro synthesized by Twist

MAY 2021 blood CANCER DISCOVERY | OF12

Bioscience and cloned into Artichoke, as described above. ZMYM2 mide, and in the presence of 10 μmol/L MG-132 for the last 5 hours coding sequence (#BC036372) was PCR amplified from a human before harvest. Cells were harvested 48 hours posttransfection. Cell cDNA clone purchased from Transomic Technologies, cloned into lysates were cleared by centrifugation, normalized, and incubated pDONR223 by BP recombination reaction using the Gateway BP Clo- with anti–V5-tag magnetic beads (MBL International) at 4°C over- nase II Enzyme mix (Thermo Fisher Scientific), and then shuttled into night. After washing the beads three times in wash buffer, proteins pLX302 and pLX304 by LR recombination reactions using the Gate- were eluted by boiling in Laemmli SDS sample buffer at 95°C for way LR Clonase II Enzyme mix (Thermo Fisher Scientific). ZMYM2, 5 minutes and then subjected to immunoblot analysis using a K48- with deletion of the MYM2 domain (amino acids 425–502), was linkage–specific polyubiquitin antibody. in vitro synthesized by IDT and cloned in pDONR223 and pLX302. The FGFR1 3′ sequence encoding amino acids 429–822 was in vitro In Vitro Ubiquitination Assays synthesized by IDT and fused by Gibson assembly (NEB) to the Purified E1, E2, ubiquitin, CUL4A–RBX1, cereblon–DDB1, and ZMYM2 5′ sequence encoding amino acids 1–913 to create the ZMYM2 proteins were used to reconstitute the ubiquitination of ZMYM2–FGFR1 chimeric fusion, using the breakpoints described by ZMYM2 in vitro. The ZMYM2 degron (amino acids 423–466) was Reiter and colleagues (15). Likewise, the FLT3 3′ sequence encoding expressed in BL21 star DE3 Escherichia coli as an MBP fusion. Cultures amino acids 594–993 was in vitro synthesized by IDT and fused by were grown to an optical density of 0.6 in 2XYT media, induced with Gibson assembly to the ZMYM2 5′ sequence encoding amino acids 0.4 mmol/L IPTG, and harvested following a 16-hour incubation at 1–1059 to create the ZMYM2–FLT3 chimeric fusion, as reported by 16°C. Cells were lysed via B-PER Complete (Thermo Fisher Scientific), Jawhar and colleagues (22). Both fusion sequences were then cloned diluted with lysis buffer (50 mmol/L Tris pH 7.5, 200 mmol/L NaCl, into pDONR223 and into the destination vectors pLX302, pLX304, or 3 mmol/L TCEP 10% Gly, 50 μmol/L Zinc Acetate, and 2× protease pLX306. Wild-type mouse Crbn isoform 2 was PCR amplified from a ret- inhibitors), batch bound to maltose affinity resin (NEB), washed roviral backbone previously described (3) and cloned into pDONR223. extensively with lysis buffer, and eluted with lysis buffer supplemented To generate the G429A point mutation in ZMYM2, ZMYM2–FGFR1, with 10 mmol/L maltose. Human CRBN (amino acids 40–442) and I391V or ZMYM2–FLT3, and the mouse Crbn mutant, we performed full-length DDB1 were coexpressed in SF9 insect cells in the presence mutagenesis in pDONR223 vectors using the Q5 Site-Directed Mutagen- of 50 μmol/L zinc acetate, and purified by nickel affinity resin (Qiagen), esis Kit (NEB) with primers designed using the NEBaseChanger (https:// HiTrap ANX column ion exchange (GE Healthcare), and Sephacryl 400 I391V nebasechanger.neb.com/). Wild-type human CRBN and Crbn were 16/60 size-exclusion chromatography (GE Healthcare), as described then cloned into the Gateway destination vector « GW » plenti.SFFV. above. Human full-length CUL4A and RBX1 were coexpressed in SF9 gateway.IRES.EGFP that was generated in-house. insect cells and purified by nickel affinity resin and Superdex 200 16/60 size-exclusion chromatography. Purified recombinant human Ube1 Lentiviral Production and Infection E1 (E-305), UbcH5a E2 (E2–616), and ubiquitin (U-100H) were Lentiviral expression plasmids were transfected together with the purchased from R&D Systems. Components were mixed to final packaging plasmids pVSV-G (Addgene #12259) and psPAX2 (Addgene concentrations of 1 μmol/L Ube1, 25 μmol/L UbcH5a, 200 μmol/L #12260) and using TRANS-LTI (Mirrus) into HEK293T cells. The Ub, 1 μmol/L CUL4–RBX1, 42 μmol/L ZMYM2, and 1 μmol/L viral supernatant was collected 48 hours after transfection, cleared cereblon–DDB1, with or without 10 μmol/L avadomide, or 10 μmol/L by centrifugation at 500 × g for 5 minutes, and then filtered through pomalidomide in ubiquitination assay buffer [20 mmol/L HEPES a Millex-HV Syringe Filter Unit, 0.45 μm (EMD Millipore). Cells (pH 7.5), 150 mmol/L NaCl, 10 mmol/L MgCl2]. After preincubation were transduced by spin infection at 30°C for 2 hours at 1,000 × g. for 10 minutes at room temperature, ubiquitination reactions were To select for infected cells, puromycin (2 μg/mL for at least 4 days) or started by addition of ATP to a final concentration of 10 mmol/L. blasticidin (10 μg/mL for at least 7–10 days) was added to the culture Reactions were incubated at 30°C for 2.5 hours before separation by media 24 hours after transduction. Cell sorting (Sony cell sorter) for SDS-PAGE followed by Coomassie staining or immunoblot analysis GFP was performed to select Ba/F3 cells stably integrated with the for MBP–ZMYM2. GW lentiviral vector. Protein Degradation Reporter Analysis Coimmunoprecipitation of CRBN Cells stably expressing the respective construct in the protein deg- 3 × 106 wild-type Hep3B cells were plated onto 15-cm dishes and radation reporter vector Artichoke or Cilantro 2 were seeded at a den- transfected with 35 μg empty vector or pLX304-ZMYM2, which sity of 10,000 to 20,000 cells per well in 96-well culture plates with encodes ZMYM2-V5. Cells were treated with DMSO or 20 μmol/L 200 μL of media per well. Different concentrations of thalidomide avadomide in the presence of 10 μmol/L MG-132 for the last 3 hours analogues were printed in three to four biological replicates using the before harvest. Cells were harvested 48 hours posttransfection and Tecan D300e Digital Dispenser. After 20 hours of drug treatment, lysed in lysis buffer (as described above) containing 20 μmol/L avado- flow cytometry using the CytoFLEX instrument (Beckman Coulter) mide or DMSO. Whole-cell extracts were cleared by centrifugation was performed to quantify EGFP and mCherry fluorescence. In at 13,000 × g at 4°C for 15 minutes, normalized, and incubated with FlowJo software, a parameter was derived that calculated the EGFP/ anti–V5-tag magnetic beads (MBL International) at 4°C for 16 hours. mCherry fluorescence ratio on a single-cell basis. The geometric After washing the beads three times in wash buffer (150 mmol/L mean of the ratio for a given drug-treated sample was normalized to NaCl, 50 mmol/L Tris-HCl pH 8.0, 1% glycerol, 1× Halt Protease, and the average of DMSO-treated controls. Phosphatase Inhibitor Cocktail) containing 20 μmol/L avadomide or DMSO, anti-V5 immunoprecipitates were eluted by boiling in In Silico Structural Modeling Laemmli SDS sample buffer at 95°C for 5 minutes and then sub- Our findings from the reporter assay suggested ZMYM2 recruit- jected to immunoblot analysis for CRBN. ment to CRBN occurs through the MYM-type zinc finger residues 423–466. In the absence of a ZMYM2 structure, a homology model In Vivo Ubiquitination for the ZMYM2 degron residues 421–453 was built in Maestro soft- For assessment of endogenous ubiquitination of ZMYM2, 3 × ware (Schrödinger; ref. 48) using the structure of the N-terminal 106 wild-type Hep3B cells were plated onto 15-cm dishes and trans- MYM-type zinc finger of human ZMYM5 [Protein Data Bank (PBD) fected with 35-μg empty vector or pLX304-ZMYM2, which encodes 2DAS] as a template. ZMYM2 residues 454–466 could not be mod- ZMYM2-V5. Cells were treated with DMSO, 1 or 10 μmol/L avado- eled due to a lack of structural coverage in PDB 2DAS. A model of

OF13 | blood CANCER DISCOVERY MAY 2021 AACRJournals.org

Downloaded from https://bloodcancerdiscov.aacrjournals.org by guest on September 26, 2021. Copyright 2021 American Association for Cancer Research. Avadomide Induces ZMYM2 Fusion Oncoprotein Degradation RESEARCH ARTICLE avadomide bound to DDB1–CRBN (PDB 4TZ4) was generated in six biological replicates for each condition. Complete media change Maestro. The ZMYM2 ternary complex model was assembled in was performed every 48 hours during drug treatment. Viable cell PyMOL (Schrödinger; ref. 49), superimposing the CRBN–avadomide counts were determined using Trypan blue staining. model with CRBN from PDB 5HXB and then superimposing ZMYM2 homology model residues 424–431 onto GSPT1 residues Statistical Analysis 570–577 from PDB 5HXB. GraphPad Prism 8.3.0 software was used to represent and analyze data from cell viability assays, reporter assays, and mouse experiments. Ba/F3 Cell Proliferation and Viability Assays Comparison of quantitative variables was performed using the two- Cell number and viability were determined using a hemocytometer tailed Student t test. P < 0.05 was considered as statistically significant. by Trypan blue staining. As an assessment of cytotoxicity, Ba/F3 cells were seeded at a density of 1,000 cells per well in 96-well culture Authors’ Disclosures plates with 200 μL of media per well. Different concentrations of J.A. Gasser reports other support from Celgene during the con- avadomide or ponatinib were printed in six biological replicates duct of the study. M.E. Matyskiela reports other support from using the Tecan D300e Digital Dispenser. After 48 hours of drug Bristol-Myers Squibb and Neomorph, Inc. outside the submitted treatment, total cellular ATP content was measured using CellTiter- work. R.S. Sellar reports grants from The Kay Kendall Leukaemia Glo Luminescent Cell Viability Assay (Promega) according to the Fund and Cancer Research UK during the conduct of the study. manufacturer protocol. Luminescence was assessed by an Envision S. De Botton reports grants from Agios and Forma, and personal Microplate Reader (PerkinElmer). Results were normalized to the fees from Celgene, Pfizer, Novartis, Daiichi, Astellas, and Jazz out- vehicle control. side the submitted work. J.-J. Kiladjian reports personal fees from Incyte during the conduct of the study, as well as personal fees from Mouse Experiments Novartis and AbbVie outside the submitted work. J.-M. Cayuela Six- to 8-week-old female BALB/cByJ mice were obtained from The reports personal fees from Novartis Pharma and Incyte outside Jackson laboratory. All mice were housed in a pathogen-free animal the submitted work. P.P. Chamberlain reports other support from facility and experiments were conducted with the ethical approval Celgene/Bristol-Myers Squibb and Neomorph, Inc. during the con- of the Harvard Medical Area Standing Committee on animals and duct of the study. S.A. Carr reports personal fees from Kymera according to an Institutional Animal Care and Use Committee– outside the submitted work. B.L. Ebert reports grants from Celgene approved protocol at Dana-Farber Cancer Institute (Boston, MA). Ba/ during the conduct of the study; grants from Deerfield outside F3 cells transformed by wild-type ZMYM2–FGFR1 in pLX304 and the submitted work; and serves on the scientific advisory boards expressing human CRBN (with co-expression of GFP) were used for in for and holds equity in Skyhawk Therapeutics, Exo Therapeutics, vivo experiments. Mice were sublethally irradiated using a Gamma-cell and Neomorph Therapeutics. No disclosures were reported by the irradiator at a dose of 550 Rads, retro-orbitally injected with 100,000 other authors. Ba/F3 cells/mouse, and dosed with avadomide (50 mg/kg; n = 15) or the DMSO vehicle control (n = 15) suspended in 0.5% carboxymethyl Authors’ Contributions cellulose and 0.25% Tween-80 in deionized water (10) by oral gav- A. Renneville: Conceptualization, formal analysis, supervision, val- age twice a day from day 3 to day 9 posttransplantation. Complete idation, investigation, visualization, methodology, writing–original blood counts were determined using the Element HT5 Hematology draft, writing–review and editing. J.A. Gasser: Resources, investigation. Analyzer (Heska), and flow cytometry was performed to monitor the D.E. Grinshpun: Formal analysis, investigation. P.M. Jean Beltran: percentage of GFP-positive cells in peripheral blood samples collected Formal analysis, investigation, visualization, methodology. N.D. Udeshi: from the retro-orbital cavity on day 3 and day 10 posttransplantation. Resources, methodology. M.E. Matyskiela: Investigation, visuali Mice were euthanized at day 10 posttransplantation. At the time of zation, methodology, writing–review and editing. T. Clayton: sacrifice, spleen weight was recorded, and bone marrow cells harvested Investigation. M. McConkey: Investigation. K. Viswanathan: Inves- from femurs and spleen were analyzed by flow cytometry. tigation. A. Tepper: Investigation. A.A. Guirguis: Resources, writing– review and editing. R.S. Sellar: Resources, writing–review and Primary Human Hematopoietic Cell Experiments editing. S. Cotteret: Resources. C. Marzac: Resources. V. Saada: Cryopreserved primary CD34+ hematopoietic stem and progeni- Resources. S. De Botton: Resources. J.-J. Kiladjian: Resources. tor cells from healthy donors were obtained from Lonza. Cryopre- J.-M. Cayuela: Resources. M. Rolfe: Resources, funding acquisi- served bone marrow samples collected at the time of diagnosis from tion. P.P. Chamberlain: Resources, supervision, writing–review and patients with ZMYM2–FGFR1-positive hematologic malignancy editing. S.A. Carr: Resources, supervision, funding acquisition, were obtained from Gustave Roussy or Paris Saint-Louis Hospital methodology, writing–review and editing. B.L. Ebert: Conceptual- biobanks. The studies were approved by institutional review boards ization, resources, supervision, funding acquisition, methodology, and performed in accordance with the Declaration of Helsinki. project administration, writing–review and editing. All patients gave their informed written consent. Cytogenetic and molecular analyses performed in bone marrow patient samples are Acknowledgments described in the Supplementary Data (Supplementary Methods; The authors thank Anna Molineaux and J. Dora Levin for admin- Supplementary Table S5). Patient CD34+ cells were purified from istrative support and assistance. The authors are grateful to Ebert bone marrow total or mononuclear cells using the AutoMACS and lab and Carr lab (to D.R. Mani and Karsten Krug) members for MACS CD34 microbeads (Miltenyi Biotec). Following enrichment, advice and helpful discussions. This work was supported by the CD34+ cells were cultured in StemSpan Serum-Free Expansion NIH (R01HL082945, P01CA108631, and P50CA206963), the How- Medium II (SFEM II; StemCell Technologies) supplemented with ard Hughes Medical Institute, and Celgene (to B.L. Ebert). S.A. Carr 1% penicillin–streptomycin, 1% l-glutamine (Life Technologies), was supported by grants from the NCI Clinical Proteomic Tumor and human cytokines (25 ng/mL SCF, 50 ng/mL thrombopoietin, Analysis Consortium grants NIH/NCI U24-CA210986 and NIH/NCI 40 ng/mL FLT3 ligand, 10 ng/mL IL-3; Miltenyi Biotec) and incu- U01 CA214125. R.S. Sellar was supported by an Intermediate Fel- bated at 37°C with 5% CO2. After expansion for 3 to 5 days, cells lowship grant from The Kay Kendall Leukaemia Fund. The authors were plated in 96-well plates (200–500 cells/well) and dosed with thank Gustave Roussy’s and Saint-Louis Hospital’s CRB Biobanks 1 μmol/L avadomide or 0.01% DMSO for 2, 4, or 6 days, with four to for managing patient biological samples.

MAY 2021 blood CANCER DISCOVERY | OF14

Received June 16, 2020; revised January 10, 2021; accepted March 21. Arber DA, Orazi A, Hasserjian R, Thiele J, Borowitz MJ, Le Beau MM, 8, 2021; published first March 10, 2021. et al. The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood 2016;127: 2391–405. 22. Jawhar M, Naumann N, Knut M, Score J, Ghazzawi M, Schneider B, References et al. Cytogenetically cryptic ZMYM2–FLT3 and DIAPH1-PDGFRB 1. Ito T, Ando H, Suzuki T, Ogura T, Hotta K, Imamura Y, et al. Iden- gene fusions in myeloid neoplasms with eosinophilia. Leukemia 2017; tification of a primary target of thalidomide teratogenicity. Science 31:2271–3. 2010;327:1345–50. 23. Roberts KG, Gu Z, Payne-Turner D, McCastlain K, Harvey RC, Chen IM, 2. Kronke J, Udeshi ND, Narla A, Grauman P, Hurst SN, McConkey M, et al. High frequency and poor outcome of philadelphia chromo- et al. Lenalidomide causes selective degradation of IKZF1 and IKZF3 some-like acute lymphoblastic leukemia in adults. J Clin Oncol 2017; in multiple myeloma cells. Science 2014;343:301–5. 35:394–401. 3. Kronke J, Fink EC, Hollenbach PW, MacBeth KJ, Hurst SN, Udeshi ND, 24. An J, Ponthier CM, Sack R, Seebacher J, Stadler MB, Donovan KA, et al. Lenalidomide induces ubiquitination and degradation of et al. pSILAC mass spectrometry reveals ZFP91 as IMiD-dependent CK1alpha in del(5q) MDS. Nature 2015;523:183–8. substrate of the CRL4(CRBN) ubiquitin ligase. Nat Commun 2017; 4. Lu G, Middleton RE, Sun H, Naniong M, Ott CJ, Mitsiades CS, et al. 8:15398. The myeloma drug lenalidomide promotes the cereblon-dependent 25. Donovan KA, An J, Nowak RP, Yuan JC, Fink EC, Berry BC, et al. destruction of Ikaros proteins. Science 2014;343:305–9. Thalidomide promotes degradation of SALL4, a transcription factor 5. Chamberlain PP, Cathers BE. Cereblon modulators: low molecular implicated in Duane Radial Ray syndrome. Elife 2018;7:e38430. weight inducers of protein degradation. Drug Discov Today Technol 26. Guharoy M, Bhowmick P, Sallam M, Tompa P. Tripartite degrons 2019;31:29–34. confer diversity and specificity on regulated protein degradation in 6. Fischer ES, Böhm K, Lydeard JR, Yang H, Stadler MB, Cavadini S, the ubiquitin-proteasome system. Nat Commun 2016;7:10239. et al. Structure of the DDB1-CRBN E3 ubiquitin ligase in complex 27. Sievers QL, Petzold G, Bunker RD, Renneville A, Slabicki M, Liddicoat BJ, with thalidomide. Nature 2014;512:49–53. et al. Defining the human C2H2 zinc finger degrome targeted by tha- 7. Chamberlain PP, Lopez-Girona A, Miller K, Carmel G, Pagarigan B, lidomide analogs through CRBN. Science 2018;362:eaat0572. Chie-Leon B, et al. Structure of the human Cereblon-DDB1- 28. Matyskiela ME, Lu G, Ito T, Pagarigan B, Lu CC, Miller K, et al. A lenalidomide complex reveals basis for responsiveness to thalidomide novel cereblon modulator recruits GSPT1 to the CRL4(CRBN) ubiq- analogs. Nat Struct Mol Biol 2014;21:803–9. uitin ligase. Nature 2016;535:252–7. 8. Petzold G, Fischer ES, Thoma NH. Structural basis of lenalidomide- 29. Warmuth M, Kim S, Gu X, Xia G, Adrián F. Ba/F3 cells and their use induced CK1alpha degradation by the CRL4(CRBN) ubiquitin ligase. in kinase drug discovery. Curr Opin Oncol 2007;19:55–60. Nature 2016;532:127–30. 30. Smedley D, Demiroglu A, Abdul-Rauf M, Heath C, Cooper C, Shipley J, 9. Chamberlain PP, Hamann LG. Development of targeted protein deg- et al. ZNF198-FGFR1 transforms Ba/F3 cells to growth factor inde- radation therapeutics. Nat Chem Biol 2019;15:937–44. pendence and results in high level tyrosine phosphorylation of 10. Hagner PR, Man HW, Fontanillo C, Wang M, Couto S, Breider M, et al. STATS 1 and 5. Neoplasia 1999;1:349–55. CC-122, a pleiotropic pathway modifier, mimics an interferon response 31. Xiao S, McCarthy JG, Aster JC, Fletcher JA. ZNF198-FGFR1 trans- and has antitumor activity in DLBCL. Blood 2015;126:779–89. forming activity depends on a novel proline-rich ZNF198 oligomeri- 11. Rasco DW, Papadopoulos KP, Pourdehnad M, Gandhi AK, Hagner PR, zation domain. Blood 2000;96:699–704. Li Y, et al. A first-in-human study of novel cereblon modulator 32. Fink EC, McConkey M, Adams DN, Haldar SD, Kennedy JA, Guirguis AA, avadomide (CC-122) in advanced malignancies. Clin Cancer Res et al. Crbn (I391V) is sufficient to confer in vivo sensitivity to thalido- 2019;25:90–8. mide and its derivatives in mice. Blood 2018;132:1535–44. 12. Carpio C, Bouabdallah R, Ysebaert L, Sancho JM, Salles GA, Cordoba R, 33. Ren M, Qin H, Ren R, Cowell JK. Ponatinib suppresses the develop- et al. Avadomide monotherapy in relapsed/refractory DLBCL: safety, ment of myeloid and lymphoid malignancies associated with FGFR1 efficacy, and a predictive gene classifier. Blood 2020;135:996–1007. abnormalities. Leukemia 2013;27:32–40. 13. Gocke CB, Yu H. ZNF198 stabilizes the LSD1-CoREST-HDAC1 com- 34. Ilaria RLJ, Van Etten RA. The SH2 domain of P210BCR/ABL is not plex on chromatin through its MYM-type zinc fingers. PLoS One required for the transformation of hematopoietic factor-dependent 2008;3:e3255. cells. Blood 1995;86:3897–904. 14. Xiao S, Nalabolu SR, Aster JC, Ma J, Abruzzo L, Jaffe ES, et al. FGFR1 35. Chan WW, Wise SC, Kaufman MD, Ahn YM, Ensinger CL, Haack T, is fused with a novel zinc-finger gene, ZNF198, in the t(8;13) leukae- et al. Conformational control inhibition of the BCR-ABL1 tyrosine mia/lymphoma syndrome. Nat Genet 1998;18:84–7. kinase, including the gatekeeper T315I mutant, by the switch-control 15. Reiter A, Sohal J, Kulkarni S, Chase A, Macdonald DH, Aguiar RC, inhibitor DCC-2036. Cancer Cell 2011;19:556–68. et al. Consistent fusion of ZNF198 to the fibroblast growth fac- 36. Sperling AS, Burgess M, Keshishian H, Gasser JA, Bhatt S, Jan M, et al. tor receptor-1 in the t(8;13)(p11;q12) myeloproliferative syndrome. Patterns of substrate affinity, competition, and degradation kinet- Blood 1998;92:1735–42. ics underlie biological activity of thalidomide analogs. Blood 2019; 16. Roumiantsev S, Krause DS, Neumann CA, Dimitri CA, Asiedu F, 134:160–70. Cross NC, et al. Distinct stem cell myeloproliferative/T lymphoma 37. Wang Y, Wu N, Liu D, Jin Y. Recurrent fusion genes in leukemia: an syndromes induced by ZNF198-FGFR1 and BCR-FGFR1 fusion attractive target for diagnosis and treatment. Curr Genomics 2017; genes from 8p11 translocations. Cancer Cell 2004;5:287–98. 18:378–84. 17. Agerstam H, Jaras M, Andersson A, Johnels P, Hansen N, Lassen C, 38. Ren M, Qin H, Wu Q, Savage NM, George TI, Cowell JK. Development et al. Modeling the human 8p11-myeloproliferative syndrome in of ZMYM2–FGFR1 driven AML in human CD34+ cells in immuno- immunodeficient mice. Blood 2010;116:2103–11. compromised mice. Int J Cancer 2016;139:836–40. 18. Jackson CC, Medeiros LJ, Miranda RN. 8p11 myeloproliferative syn- 39. Munthe-Kaas MC, Forthun RB, Brendehaug A, Eek AK, Høysæter T, drome: a review. Hum Pathol 2010;41:461–76. Osnes LTN, et al. Partial response to sorafenib in a child with a mye- 19. Smedley D, Hamoudi R, Clark J, Warren W, Abdul-Rauf M, Somers G, loid/lymphoid neoplasm, eosinophilia, and a ZMYM2–FLT3 fusion. et al. The t(8;13)(p11;q11-12) rearrangement associated with an atypi- J Pediatr Hematol Oncol 2020 Aug 26 [Epub ahead of print]. cal myeloproliferative disorder fuses the fibroblast growth factor recep- 40. Chen J, Deangelo DJ, Kutok JL, Williams IR, Lee BH, Wadleigh M, tor 1 gene to a novel gene RAMP. Hum Mol Genet 1998;7:637–42. et al. PKC412 inhibits the zinc finger 198-fibroblast growth factor 20. Strati P, Tang G, Duose DY, Mallampati S, Luthra R, Patel KP, et al. receptor 1 fusion tyrosine kinase and is active in treatment of stem Myeloid/lymphoid neoplasms with FGFR1 rearrangement. Leuk cell myeloproliferative disorder. Proc Natl Acad Sci U S A 2004;101: Lymphoma 2018;59:1672–6. 14479–84.

OF15 | blood CANCER DISCOVERY MAY 2021 AACRJournals.org

41. Kreil S, Ades L, Bommer M, Stegelmann F, Ethell ME, Lubking A, 45. Zecha J, Satpathy S, Kanashova T, Avanessian SC, Kane MH, Clauser KR, et al. Limited efficacy of ponatinib in myeloproliferative neoplasms et al. TMT labeling for the masses: a robust and cost-efficient, in- associated with FGFR1 fusion genes. Blood 2015;126:2812. solution labeling approach. Mol Cell Proteomics 2019;18:1468–78. 42. Verstovsek S, Vannucchi AM, Rambaldi A, Gotlib JR, Mead AJ, 46. Mertins P, Tang LC, Krug K, Clark DJ, Gritsenko MA, Chen L, et al. Hochhaus A, et al. Interim results from fight-203, a phase 2, open- Reproducible workflow for multiplexed deep-scale proteome and label, multicenter study evaluating the efficacy and safety of pemi- phosphoproteome analysis of tumor tissues by liquid chromatogra- gatinib (INCB054828) in patients with myeloid/lymphoid neoplasms phy-mass spectrometry. Nat Protoc 2018;13:1632–61. with rearrangement of fibroblast growth factor receptor 1 (FGFR1). 47. Sievers QL, Gasser JA, Cowley GS, Fischer ES, Ebert BL. Genome-wide Blood 2018;132:690. screen identifies cullin-RING ligase machinery required for lenalidomide- 43. Mertens F, Johansson B, Fioretos T, Mitelman F. The emerging dependent CRL4(CRBN) activity. Blood 2018;132:1293–303. complexity of gene fusions in cancer. Nat Rev Cancer 2015;15: 48. Schrödinger. Release 2018–3: Maestro [software]. Schrödinger, LLC. 371–81. Available from: https://www.schrodinger.com/products/maestro. 44. de Thé H, Chen Z. Acute promyelocytic leukaemia: novel insights into 49. Schrödinger. The PyMOL Molecular Graphics System, Version 2.2 [soft- the mechanisms of cure. Nat Rev Cancer 2010;10:775–83. ware]. Schrödinger, LLC. Available from: https://pymol.org/2/.

MAY 2021 blood CANCER DISCOVERY | OF16

Downloaded from https://bloodcancerdiscov.aacrjournals.org by guest on September 26, 2021. Copyright 2021 American Association for Cancer Research. Avadomide Induces Degradation of ZMYM2 Fusion Oncoproteins in Hematologic Malignancies

Aline Renneville, Jessica A. Gasser, Daniel E. Grinshpun, et al.

Blood Cancer Discov Published OnlineFirst March 10, 2021.

Updated version Access the most recent version of this article at: doi: 10.1158/2643-3230.BCD-20-0105

Supplementary Access the most recent supplemental material at: Material http://bloodcancerdiscov.aacrjournals.org/content/suppl/2021/03/09/2643-3230.BCD-20-0105.DC 1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://bloodcancerdiscov.aacrjournals.org/content/early/2021/04/15/2643-3230.BCD-20-0105. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.