Aberrant FLT3 Transcription Arised from Polycomb Repressive Complex 2 Inactivation Indiactes a Potential Therapy Target for Pediatric T Cell Acute Lymphoblastic Leukemia

Pediatric T cell acute lymphoblastic leukemia (T-ALL) is an aggressive subtype of pediatric ALL, and is characterized by recurring genetic and epigenetic alterations. Loss-of-function mutations and deletions of Polycomb repressive complex 2 (PRC2) core components, including EED, EZH2 and SUZ12 ,have been identified in around 25% of T-ALL and are associated with diseases progression and poor prognosis. One presented study has demonstrated the specific loss of the repressive mark Lys27 trimethylation of histone 3 (H3K27me3) by abrogating the activity of PRC2 leads to Notch pathway activation; In a mouse model of early T cell precursor ALL (ETP-ALL), inactivation of PRC2 has been revealed to be associated with increasing transcription of early hematopoietic programs and contribute to hyperactive JAK/STAT signaling pathway. However, the detailed mechanism of pediatric T-ALL tumorigenesis driven by PRC2 remains to be further clarified.

Through targeted sequencing of PRC2 components of 62 diagnostic pediatric T-ALL samples, we discovered mutations in EZH2 (8/62), SUZ12 (4/62), and EED (2/62) in 12 patients. RNA sequence profiling of the sorted primary samples from these 12 patients revealed a surprisingly upregulated expression of FLT3 and this upregulation was confirmed to be specifically presented in PRC2 mutated patients by qRT-PCR analysis in all the 62 T-ALL primary samples.

To explore the interplay between the inactivation of PRC2 and upregulation of FLT3, we constructed EZH2 disrupted Jurkat cell lines by CRISPR-Cas9. QRT-PCR reflected eight-to-twenty-fold increasing of FLT3 mRNA expressions in homozygousknockout clones; ChIP-qPCR showed that EZH2 deletions led to remarkably attenuated H3K27me3 histone mark and prominent enrichment of POLII binding in FLT3 promoter region, which were highly consistent with gene expression levels; Besides, elevation in phosphorylation of STAT5, AKT and ERK also showed high correlation with increased FLT3 expressions, which was reversed by the treatment of FLT3 inhibitors, Sorafeinib and Quizartinib, respectively. This, in turn, suggested that the loss of H3K27me3 caused by PRC2 disruption closely coupled to the activation of FLT3 downstream signaling pathway in human T-ALL.

To examine the regulation of FLT3 transcription by PRC2 in hematologic malignances, we exposed cell lines of T-lineage (Jurkat, Molt4, CEM), B-Lineage (Ramos, Raji), and other myeloid malignancies (K562, HL-60, THP-1) to EZH2 inhibitors and observed only T-lineage cell lines specifically presented response to EZH2 inhibitors with increased expression of FLT3, and therefore the activation of downstream pathway. Rather, inhibitions of PRC2 seemed not to uprgulate FLT3 transcriptions in other cell lines.

In a model of T-lineage differentiation from immature hematopoietic cells on OP9-DL1 cells exposed to EZH2 inhibitors, lineage-negative, Sca1-positive, and C-kit-positive (LSK) cells, and ETP cells derived from C57BL/6 mice displayed markedly differentiation arrest via flow cytometric analysis, accompanied by upregulation of Flt3 expression and accentuation of down stream signaling pathway.

To further test whether the hyperactive signaling pathway mediated by EZH2 -inactivation was a potential therapy target, pediatric T-ALL cells from EZH2 -mutated primary samples were transplanted into NSG mouse and treated with FLT3 inhibitors. Surprisingly, FLT3 inhibitors treated group conferred hypersensitivity to the drugs, and the median survival span from treatment to death in FLT3 inhibitors group was significantly longer than control group treated with PBS (P < 0.0001).

In conclusion, our study provides strong evidence that loss of H3K27me3 arised from PRC2 inactivation leads to high level of FLT3 transcription and downstream signaling pathway upregulation in human T-ALL, indicating both a possible mechanism and a potential therapy target for pediatric T-ALL with PRC2 mutations in future clinical practice.