Epigenetic activation of the FLT3 gene by ZNF384 fusion confers a therapeutic susceptibility in acute lymphoblastic leukemia

Abstract

FLT3 is an attractive therapeutic target in acute lymphoblastic leukemia (ALL) but the mechanism for its activation in this cancer is incompletely understood. Profiling global gene expression in large ALL cohorts, we identify over-expression of FLT3 in ZNF384-rearranged ALL, consistently across cases harboring different fusion partners with ZNF384. Mechanistically, we discover an intergenic enhancer element at the FLT3 locus that is exclusively activated in ZNF384-rearranged ALL, with the enhancer-promoter looping directly mediated by the fusion protein. There is also a global enrichment of active enhancers within ZNF384 binding sites across the genome in ZNF384-rearranged ALL cells. Downregulation of ZNF384 blunts FLT3 activation and decreases ALL cell sensitivity to FLT3 inhibitor gilteritinib in vitro. In patient-derived xenograft models of ZNF384-rearranged ALL, gilteritinib exhibits significant anti-leukemia efficacy as a monotherapy in vivo. Collectively, our results provide insights into FLT3 regulation in ALL and point to potential genomics-guided targeted therapy for this patient population.

Fig. 1. FLT3 expression by subtype in B-ALL patients of diverse ancestries.

a FLT3 expression across different B-ALL subtypes in the US cohort of 1988 children and adults (Cohort 1), with the highest expression of FLT3 in ZNF384-r (ZNF384-rearrangement) ALL (n = 1988). b FLT3 expression in ALL cases (n = 49) with various ZNF384 fusions in cohort 1. Subgroups were defined by fusion partners of ZNF384. Other: fusion partners include ARID1B, CLTC, CREBBP, EWSR1, NIPBL, and SMARCA2. c FLT3 expression across different B-ALL subtypes in cohort 2 (n = 377), with the highest expression of FLT3 in ZNF384-r ALL. d FLT3 expression across ZNF384 fusions (n = 17) in the Asian cohort (Cohort 2). Subgroups were defined by fusion partners of ZNF384. Other: fusion partners include USP25 and CREBBP. e FLT3 mutations in ZNF383-r ALL from each ALL cohort. A.U. arbitrary units. Center lines indicate median values of FLT3 expression. Source data are provided as a Source Data file.

Fig. 2. ZNF384 fusion-specific activation of a distal enhancer at the FLT3 locus in ALL.

a, b ZNF384 CUT&RUN (Cleavage Under Targets & Release Using Nuclease) assay using a TCF3-ZNF384 ALL PDX sample (ID TCZ). In total, 10,470 peaks were identified, with 4820 in promoter regions (±2 kb from TSS) and the other 6650 in enhancers. 6750 peaks overlapped with H3K27ac peaks, while 4602 peaks overlapped with H3K4me3. c CUT&RUN and ATAC-seq (Assay for Transposase-Accessible Chromatin using sequencing) of ALL cell lines and primary ALL blast samples with or without ZNF384-rearrangements. Each track represents a type of assay in a given sample as indicated on the left. As shown in the CUT&RUN tracks, a unique ZNF384 binding site was observed 25 kb upstream of FLT3 with prominent H3K27ac and H3K4me3 marks. This peak also overlapped with an open chromatin region identified by ATAC-seq in EP300-ZNF384 ALL primary samples (n = 2). In contrast, this region was void of ATAC-seq signals in ALL cell lines or ALL primary samples of other subtypes. Bottom panel: two ZNF384 binding motifs (AAAAAAAA) were identified by footprint analysis using the ZNF384 CUT&RUN data derived from TCF3-ZNF384 ALL PDX cells. d Enhancer activity of z-FLT3 enhancer was confirmed by luciferase assay. A 500 bp segment (hg38, chr13:28,124,365–28,124,868) covered by this z-FLT3 enhancer increased transcription activity by 12.7-fold compared to vector control in JIH5 cells. Enhancer activity was normalized to empty vector control. Data are shown as mean values ± SEM of three biological replicates (center of the error bar) and the results are representative of three independent experiments. Source data are provided as a Source Data file.

Fig. 3. 3D chromatin interaction at 5′-UTR of FLT3 in ZNF384-r ALL.

ChIA-PET (chromatin interaction analysis with paired-end tags) (CTCF and RNAP II) was performed in both JIH5 cells and an ALL PDX sample harboring the TCF3-ZNF384 fusion. As shown in the top four panels, DNA looping mediated by CTCF and RNAP II was detected between ZNF384-binding site (“z-FLT3 enhancer”) and FLT3 promoter, indicated by shaded box. In addition, we also performed ZNF384 ChIA-PET in both JIH5 cells and TCF3-ZNF384 ALL PDX sample to map ZNF384-mediated chromatin interactions, in comparison to lymphoblastoid cell line GM12878 which does not harbor ZNF384 fusion gene. Red curves indicated chromatin interactions called using ChIA-PIPE (paired-end tag reads ≥2) and were highlighted in red in Supplementary Data 1 and 2. Additionally, the bottom panels show protein binding sites (ZNF384 and RNAP II) identified from inter-ligation and self-ligation reads of ChIA-PET using MACS2 calling algorithm. Overlap of ZNF384 binding sites and chromatin looping anchors suggests enhancer-promoter interaction specifically mediated by ZNF384 (indicated by green and red arrows).

Fig. 4. Anti-leukemia effect of gilteritinib in vitro and in vivo.

a In vitro sensitivity of a panel of ALL cell lines as well as PDX-derived ALL cell with TCF3-ZNF384 fusion to gilteritinib was determined using MTT assay. Data are shown as mean % viability relative to vehicle ± SEM of three biological replicates (center of the error bar) and results are representative of three independent experiments. b Ex vivo sensitivity of a panel of 47 primary ALL cases including three with EP300-ZNF384 fusion as well as those with ETV6-RUNX1, DUX4-r, BCR-ABL1, hyperdiploidy, and T-ALL. Box plots show summary of data in terms of minimum, maximum, median, and first and third quartiles. Each data point represents two technical replicate. c, d Expression of EPZ (EP300-ZNF384) fusion gene was downregulated by CRISPR Cas9 editing in JIH5 cells, which led to decreased expression of FLT3 and drug sensitivity to gilteritinib. Data are shown as mean values ± SEM of three biological replicates (center of the error bar) and results are representative of three independent experiments. P values (P = 0.006) were estimated using two-sided t test. e–h Gilteritinib efficacy was evaluated in vivo using xenograft models. e, f Show leukemia progression, and survival in mice transplanted with JIH5 cells. g, h Describe results of mice transplanted with PDX-derived ALL cells with TCF3-ZNF384 fusion. Gilteritinib was given daily at a dosage of 10 mg/kg, and therapy started three days following leukemia engraftment. Leukemia burden was monitored weekly, and P values (P < 0.005) were estimated using a two-sided analysis of deviance based on mixed effect models with cubic splines. Leukemia-free survival was plotted as Kaplan–Meier curves and P values were estimated using two-sided log-rank test (P = 0.0027 for f, P = 0.0026 for h). Source data are provided as a Source Data file.