TP63 fusions drive multicomplex enhancer rewiring, lymphomagenesis, and EZH2 dependence

Abstract

Gene fusions involving tumor protein p63 gene (TP63) occur in multiple T and B cell lymphomas and portend a dismal prognosis for patients. The function and mechanisms of TP63 fusions remain unclear, and there is no target therapy for patients with lymphoma harboring TP63 fusions. Here, we show that TP63 fusions act as bona fide oncogenes and are essential for fusion-positive lymphomas. Transgenic mice expressing TBL1XR1::TP63, the most common TP63 fusion, develop diverse lymphomas that recapitulate multiple human T and B cell lymphomas. Here, we identify that TP63 fusions coordinate the recruitment of two epigenetic modifying complexes, the nuclear receptor corepressor (NCoR)-histone deacetylase 3 (HDAC3) by the N-terminal TP63 fusion partner and the lysine methyltransferase 2D (KMT2D) by the C-terminal TP63 component, which are both required for fusion-dependent survival. TBL1XR1::TP63 localization at enhancers drives a unique cell state that involves up-regulation of MYC and the polycomb repressor complex 2 (PRC2) components EED and EZH2. Inhibiting EZH2 with the therapeutic agent valemetostat is highly effective at treating transgenic lymphoma murine models, xenografts, and patient-derived xenografts harboring TP63 fusions. One patient with TP63-rearranged lymphoma showed a rapid response to valemetostat treatment. In summary, TP63 fusions link partner components that, together, coordinate multiple epigenetic complexes, resulting in therapeutic vulnerability to EZH2 inhibition.

Fig. 1.. TP63 fusions promote lymphoma growth in vitro and in vivo

(A) Schematic of fusions TBL1XR1, FOXK2, BCL6, TP63, ΔNp63, TBL1XR1::TP63, FOXK2::TP63 and BCL6::TP63. Different colors show different domains. (B) Cell viability of TCL cell lines treated with shTP63 or shSeed. (C) Western blot analysis for TBL1XR1::TP63 in Cas9-SMZ1-sgwtTP63 cells expressing Dox-inducible shSeed or shTP63 treated with or without Dox for four days. (D) Cell proliferation assay of cells from (C) using CellTiter-Glo (CTG). (E) Tumor volume at the indicated timepoint after subcutaneous implantation of Cas9-SMZ1-sgwtTP63 cells with Dox-inducible shTP63 into mice (n=5 per group). The red arrow indicates the beginning of treatment with Dox or vehicle in vivo. (F) Rescue experiment of SMZ1 cells expressing shSeed or shTP63 with empty vector (EV) or TBL1XR1::TP63 expressing vector. (G) The proliferation of Ba/F3 cells expressing EV or TBL1XR1::TP63 cultured in the presence of the indicated concentrations of mIL-3 measured using CTG. Data are normalized to the empty vector (EV) group. (H) Western blot analysis using an anti-TP63 antibody in Cas9-SMZ1-sgwtTP63-shTP63 cells expressing Dox-inducible BCL6::TP63 fusion. Cells were treated with or without Dox for four days. (I) Cell proliferation assay of cells in (H) using CTG. (J, K) Normalized CRISPR score (NCS) of each sgRNA (dots) and the smoothed score (line) of the pooled TBL1XR1 and TP63 (J) and TBL1XR1::TP63 (K) survival screen on day 30 in SMZ1-Cas9+ cells. All data unless specified are presented as mean ± SD or ±SEM, n=3 biological replicates. * P<0.05, *** P<0.001 or **** P<0.0001 as compared between indicated groups (Student’s t-test).

Fig. 2.. TBL1XR1::TP63 regulates the T- and B-cells in mice

(A) Strategy for insertion of the CAG-loxP-Stop-LoxP-TBL1XR1::TP63-IRES-EGFP cassette into the mouse Rosa26 locus. (B, C) Representative flow activated cell sorting (FACS) plot (B) and quantification (C) showing TCRb staining of viable cells. (D, E) Representative FACS plot (D) and quantification (E) showing CD4/CD8 staining of TCRb+ T cells; (F, G) Representative FACS plot (F) and quantification (G) showing Treg (Foxp3 staining) of TCRb+ T cells; (H, I) Representative FACS plot (H) and quantification (I) showing B cells (CD19/B220 staining) of viable cells; (J, K) Representative FACS plot (J) and quantification (K) showing FoB and MZB (CD21/CD23 staining) of B220+ B cells; (L, M) Representative FACS plot (L) and quantification (M) showing GCB (CD38/CD95 staining) of B220+ B cells; (N, O) Representative FACS plot (N) and quantification (O) showing naïve B cell (CD27/IgD staining) of B220+ B cells; (P, Q) Representative FACS plot (P) and quantification (Q) showing PC (CD138/B220 staining) of viable cells. All spleen samples were from 11-week-old Cre and Cre/tgTT mice. Data are presented as mean ± SD, *P<0.05, ** P<0.01, *** P<0.001, **** P<0.0001 as compared between indicated groups (Student’s t-test). n=4 mice/arm except for B cells (n=6 mice for Cre and n=7 mice for Cre/tgTT). (R) Quantification of cytokines in peripheral blood serum from 11-week-old Cre and Cre/tgTT mice. Cytokine quantification is presented as the ratio in Cre/tgTT versus Cre mice (n=3 mice/arm).

Fig. 3.. TBL1XR1::TP63 drives the development of diverse T- and B-cell lymphomas in mice

(A) Kaplan-Meier curve of lymphoma-specific survival in Cre and Cre/tgTT mice within 210 days. n=20 mice/arm. P value by Log-rank test. (B) Representative image of the spleen from moribund Cre/tgTT mice and age-matched control wildtype (WT) and Cre mice. (C) Western blot analysis for indicated proteins in spleen samples from (B). (D) H&E staining of spleen tissues in (B) showing representative splenic architecture of each group of mice. Scale bar: 1 mm. (E) Immunohistochemical (IHC) analysis showing the expression of CD3 and CD4 in representative spleen tissues in (B). Scale bars are as indicated. (F) H&E and IHC analysis for the indicated markers in representative spleen tissues harvested from four lymphoma-bearing Cre/tgTT mice. Scale bar: 100 mm. (G) RNAseq analysis of transgenic tumors. Enrichment of individual human lymphoma subtype signatures was assessed. (H) Contribution of CDR3b and IgH sequences to the T-cell receptor repertoire and B-cell receptor repertoire in transgenic tumor samples, respectively. The colors between samples do not indicate the same clone. The top 5 clones are shown. (I) Bar graph showing the cumulative incidence of transgenic tumors. 11 of 12 tumors were characterized. One mouse having an enlarged spleen (tumor) was found dead, and the tumor was not further analyzed. (J, K) Kaplan-Meyer survival curve for C57BL/6 mice after being irradiated with 5.5 Gy and intravenously transplanted with 0.5 million spleen cells from diseased Cre/tgTT mice (Tumor 1 (J) or Tumor 3 (K)) or from Cre mice. n=5 mice/arm. P value by Log-rank test.

Fig. 4.. TBL1XR1::TP63 fusion rewires enhancers to regulate gene expression

(A) GSEA plots of indicated gene sets in SMZ1 cells expressing shTP63 versus shSeed control. Normalized enrichment score (NES); FDR: false discovery rate (FDR). (B) Venn diagram showing genes regulated by TBL1XR1::TP63 from RNA-seq in SMZ1 and genes bound by TBL1XR1::TP63 from ChIP-seq in SMZ1. (C) Pie graph showing the locations of TBL1XR1::TP63 binding peaks based on ChIP-seq of SMZ1 and DL40 cells. (D) Schematic of IP-MS of flag-tagged TP63, ΔNp63, TBL1XR1::TP63, TBL1XR1, and the truncated TBL1XR1 (ΔTBL1XR1). (E-G) IP using anti-TP63 antibody that binds TBL1XR1::TP63 in Cas-SMZ1-sgwtTP63 cells (E), anti-KMT2D antibody in SMZ1 cells and Cas9-SMZ1-sgwtTP63 cells (F), and anti-TP63 in SMZ1 cells and OCI-Ly12 cells (G). (H) IP of KMT2D in SMZ1 cells with 150 mM NaCl (low-salt) IP buffer or 300 mM NaCl (high-salt) IP buffer then immunoblotted for TP63. (I) Analysis of TBL1XR1::TP63-bound enhancers from HiChIP and ChIP-seq. (J) Analysis of PRO-seq signal at the genes associated with TBL1XR1::TP63-bound enhancers in Cas9-SMZ1-sgwtTP63 cells with or without TP63 fusion knockdown using Dox-inducible shRNA. PRO-seq signal is shown as violin plots of summed reads from the TSS to +75nt. P-value is from Wilcoxon test. (K) Motif analysis at TBL1XR1::TP63-bound enhancers by using HOMER.

Fig. 5.. TBL1XR1::TP63 activates enhancers to confer PRC2 dependence

(A) Ranked dependency score of genome-wide CRISPR screen in SMZ1 cells with TP63 and EZH2 highlighted. (B, C) Z scores for TP63 (B) and EZH2 (C) dependency across a pan-cancer set of 391 cell lines (14,30). (D) Cell proliferation assay of SMZ1 cells transduced with shRNAs targeting EZH2 using CTG. (E) IC50 of Valemetostat (VAL) for TCL cell lines. The B-cell lymphoma cell line Karpas-422 (blue), which harbors an EZH2 Y646F gain-of-function mutation serves as a positive control. TCL cell lines with TP63 rearrangements are in red and those with WT TP63 are in black. (F) Cell proliferation assay of SMZ1, DL40, and OCI-Ly12 cell lines treated with VAL using CTG. (G) Venn diagram showing genes regulated by TBL1XR1::TP63 from RNA-seq and genes bound by TBL1XR1::TP63 from ChIP-seq in SMZ1. (H) Possible mechanisms of regulating EED and EZH2 by TBL1XR1::TP63. (I, J) RT-qPCR analysis for EED, MYC, and EZH2 after TP63 knockdown with Dox treatment in SMZ1 cells (I) and MTA cells (J). *** P<0.001 or ****P<0.0001 for shTP63+Dox compared to shSeed+Dox for the same gene. (K, L) Western blot of EED, MYC, EZH2 and H3K27me3 after Dox-induced knockdown in Cas9-SMZ1-sgwtTP63 cells (K) and MTA cells (L). (M) The TBL1XR1-TP63-activated gene MYC is connected to a high-confidence distal TBL1XR1::TP63 binding site through chromatin loops identified from H3K27Ac HiChIP assays in Cas9-SMZ1-sgwtTP63 cells. HiChIP anchors that are looped to promoters of the target genes are presented. ChIP-seq shows the TBL1XR1::TP63 binding site as well as enhancer-associated chromatin marks. (N) ChIP-qPCR of TP63 fusion at the MYC enhancer in Cas9-SMZ1-sgwtTP63 cells. Control (CTR) primer sets are also included. (O) Luciferase reporter assay measuring the MYC enhancer activity in 293T cells expressing indicated proteins. The pGL3 plasmid without the MYC enhancer region (Empty vector, EV) is used as a negative control. (Y-axis) Relative (Rel.) Luciferase units are normalized to Renilla signal ± SD. (P) RT-qPCR analysis of MYC expression in SMZ1 cells with (+DOX) and without (−DOX) KRAB-dCas9 mediated repression of the MYC enhancer. (Q) PRO-seq reads at the MYC locus in Cas9-SMZ1-sgwtTP63 cells with (+DOX) and without (−DOX) TBL1XR1::TP63 knockdown using Dox-inducible shRNA. Data are presented as mean ± SD, n=3 biological replicates. *** P<0.001 or ****P<0.0001 as compared between indicated groups (Student’s t-test).

Fig. 6.. EZH2 inhibition as a promising therapeutic strategy for TP63-rearranged TCLs

(A) Tumor growth curve of NSG mice engrafted with SMZ1 cells injected subcutaneously (n=5/arm). Mice were treated with vehicle (Veh) or Valemetostat (VAL) when tumor size reached to 200 mm³ and sacrificed when subcutaneous tumors reached 2 cm in the longest dimension. (B) Kaplan-Meyer survival curve for mice in A. (C) Mean fluorescence intensity (MFI) of the H3K27me3 abundance in tumor samples from B determined by FACS. (D) Schematic outline of transplantation of Cre/tgTT lymphoma-infiltrated spleen cells (Tumor 1) and treating C57BL/6 recipients with Veh or VAL. C57BL/6 mice are irradiated with 5.5 Gy before receiving 0.5 million cells intravenously and treatments are initiated two days after transplantation. A cohort of mice are sacrificed after 10 days for pharmacodynamic (PD) analysis, and another cohort are observed for survival analysis. (E, F, G) Photo of spleens (E), spleen weight versus body weight ratio (F) and western blot analysis of H3K27me3 in spleen cells (G) from mice treated with VAL or Veh (n=3 mice/arm) for 10 days. (H) Kaplan-Meyer survival curve of C57BL/6 mice transplanted with Cre/tgTT lymphoma-infiltrated spleen cells and treated with Veh or VAL (n=6 mice/arm). (I) Western blot analysis for indicated proteins in spleen cells harvested from H. (J) Schematic of subcutaneous engraftment of ALK- ALCL PDX (COH1) harboring a TBL1XR1-TP63 rearrangement in NSG mice and treating recipients with VAL or Veh when tumor size reached to 200 mm³. (K) FACS analysis of H3K27me3 in tumor cells harvested from mice treated with VAL or Veh (n=3 mice/arm) for 10 days. (L, M) Tumor growth curve (L) and Kaplan-Meyer survival curve (M) of the COH1 PDX model. Mice were treated with Veh or VAL (n=6 mice/arm) by p.o. and sacrificed when the subcutaneous tumor reached 2 cm in the longest dimension. (N, O) Western blot analysis for indicated proteins in tumor cells harvested from M (N) and quantification of relative H3K27me3 abundance normalized to H3 (O). (P) Fluorescence in situ hybridization using a 5’ TP63 probe (green, 314 kb centromeric to TP63) and a 3’ TP63 probe (red, 310 kb telomeric to TP63). The translocated chromosomes are indicated with arrows. (Q, R) FACS showing abnormal T cells (purple) (Q) and associated quantifications (R) from pre-treatment and day 28 of VAL treatment in whole blood cells of Patient 2. Other cells shown are NK cells (blue), CD8+ T cells (green), and CD4+ T cells (red). Data are presented as mean ± SEM. ** P<0.01 or **** P<0.0001 as compared between indicated groups (Student’s t-test). P value for survival by Log-rank test.