Background: One fifth of T-cell acute lymphoblastic leukemia (T-ALL) cases is defined by the aberrant expression of TLX3. Being similar to the closely related TLX1, its expression is mainly associated with mutations in the JAK/STAT pathway (Liu et al., Nat Genetics 2017), which have been shown to cooperate with TLX1 in driving T-ALL (Vanden Bempt et al., Cancer Cell 2018). However, it is not known whether TLX3 has the same oncogenic role as TLX1. Additionally, TLX3 expression is also regularly found with FLT3-ITD mutations, which are virtually absent in TLX1-positive cases (Liu et al., Nat Genetics 2017), suggesting there are some important functional differences between TLX3 and TLX1. Aims: To set up a model of cooperation between TLX3 and FLT3-ITD and to elucidate the oncogenic role of TLX3 in T-ALL. Methods: We performed in vivo bone marrow transplant (BMT) assays by transducing lineage-negative cells derived from C57BL/6 mice with the respective oncogenes using retroviruses. Pro-T cell growth curves were performed in the absence of interleukin-7 (Il7) as described previously (Van Thillo et al., Nat Communications 2021). Fluorescence activated cell sorting (FACS) staining was performed on the MACSQuant VYB (Miltenyi Biotec) or Fortessa (BD). RNA was prepared at the KU Leuven Genomics Core and 3’ prime end sequencing (QuantSeq) was performed. Differential gene expression analyses were conducted using the DeSeq2 R package (v1.22.0). Results: The combination of TLX3 and FLT3-ITD in a constitutive BMT assay resulted in a rapid leukemia with an immature myeloid immunophenotype (CD11b+, Gr1-, CD4-, CD8-). The leukemia cells were positive for both FLT3-ITD (GFP) and TLX3 (mCHERRY), indicating cooperation between the two oncogenes in vivo. Similarly, in our in vitro pro-T-cell model, the TLX3 and FLT3-ITD double-positive cells outcompeted the single-positive cells after withdrawal of Il7. Next, we performed RNA-sequencing in CreER (estrogen receptor) pro-T cells 24 hours after activation of TLX3 expression by adding tamoxifen. This showed both up- and downregulation of many genes, including Tle4 among the downregulated genes. Interestingly, TLE4 expression was also lowest in the TLX3-subgroup of patients (Liu. et al. Nat Genetics 2017). TLE proteins are transcriptional co-repressors that can interact with the engrailed-homology 1 (Eh1) domain of transcription factors (Riz et al. Biochem Biophys Res Commun 2009). Therefore, we inserted a point mutation in the Eh1 domain of TLX3, resulting in a phenylalanine to glutamic acid substitution at position 18, to impair this interaction. Strikingly, TLX3(F18E) in combination with FLT3-ITD led to a proliferative advantage in pro-T cells after the omission of Il7 compared to wild-type TLX3. On top of that, cells with TLX3(F18E) and FLT3-ITD were able to grow in the absence of stem cell factor (Scf). Summary/Conclusion: We demonstrate that TLX3 cooperates with FLT3-ITD in vivo and in vitro in a pro-T-cell context. TLX3 downregulates the co-repressor Tle4 and inhibiting the interaction between TLX3 and Tle4 reinforces the proliferative capacity of pro-T cells in the absence of growth factors.
Read full abstract