Signaling Pathway Mutations Cooperate with the PICALM/MLLT10 Fusion in a Knock-in AML Mouse Model
Rhea H. Desai,Peter J. Browett,Purvi M. Kakadia,Stefan K. Bohlander
DOI: https://doi.org/10.1182/blood-2023-178915
IF: 20.3
2023-11-28
Blood
Abstract:The PICALM/MLLT10 (or CALM/AF10; C/A) is a rare but recurring fusion gene detected in patients with T-cell acute lymphoblastic leukemia, malignant lymphoma and acute myeloid leukemia (AML). Our lab has previously established a C/A knock-in mouse model, where the fusion gene preceded by a loxP-flanked transcriptional stop cassette was inserted into the Rosa26 locus (the R26LSLCA strain). To achieve tissue-specific expression of C/A, R26LSLCA mice were crossed with three different Cre-inducer lines, namely the Vav-Cre, MB1-Cre and CD19-Cre. In the Vav-Cre strain, Cre recombinase is expressed in the entire hematopoietic system, while in the MB1-Cre and CD19-Cre lines, Cre is expressed in the B-cell compartment. Using this model, Dutta et al. showed that C/A leads to leukemia only when expressed in all hematopoietic cells, including hematopoietic stem cells and does not lead to induction of leukemia when expressed in B-cells alone ( Dutta et al., Leukemia 2016). To further characterise leukemias driven by the C/A fusion, we established new R26LSLCA and Vav-Cre crosses. Similar to what was found previously, the Vav-Cre/R26LSLCA mice developed leukemia with a long median latency of 1 year (366.5 days; N=24) and complete penetrance. On post-mortem analysis, the Vav-Cre/R26LSLCA mice were found to have splenomegaly and leucocytosis. Expression of the fusion gene was confirmed in these mice by RT-PCR on the cDNA from bone marrow (BM) cells. Immunophenotyping analysis of the BM cells and splenocytes of these mice revealed a predominantly myeloid phenotype. To assess the effect of expression of C/A on the transcriptome, we performed RNA-seq analysis using BM cells from 17 leukemic mice and 12 healthy controls. In accordance with reports on C/A patient samples, the Hoxa genes Hoxa7 and Hoxa9 as well as the Hox-cofactor Meis1 were significantly upregulated in our C/A expressing mice. In addition, genes involved in B-cell development, such as Cd19, Pax5 and Ikzf3, and Gata1, which is essential for the development of erythrocytes, were downregulated. To prove that the Vav-Cre/R26LSLCA mice were indeed leukemic and had the potential to propagate the disease, we serially transplanted BM cells from 4 Vav-Cre/R26LSLCA mice into congenic recipients. The secondary (N=20) and tertiary (N=19) transplanted mice developed leukemia with a significantly shorter median latency of 21 and 17 days, respectively. Our group and others have shown that a single genetic lesion is in most cases not sufficient to cause leukemia. Thus, to determine if additional somatic mutations were acquired by the C/A knock-in mice, we performed whole exome sequencing on the DNA from BM cells of the 4 knock-in mice, some of their daughter secondary and tertiary leukemias as well as the corresponding germline DNA samples (N=19). We identified between 2-32 mutations per leukemia exome, including mutations in genes and/or members of gene families that have been implicated in human leukemias. Mutations were detected in the transcription factors Sfpi1, Zeb1, and Sox9; Aff2, a member of the Af4/Fmr2 gene family; and Sf3a2, a component of the RNA-splicing machinery, among others. In addition, in two C/A sister secondary leukemias and the corresponding tertiary leukemias, we detected a missense mutation at amino acid 507 (G507A) of the Ptpn11 gene, which is homologous to G507 of human PTPN11, a known hotspot mutation site in AML and juvenile myelomonocytic leukemia. Interestingly, Dutta et al. also reported a G507 mutation (G507V) in Ptpn11 in a C/A knock-in mouse. PTPN11 regulates the RAS/MAPK signaling pathway. This finding is reminiscent of what we have previously reported in a C/A driven murine bone marrow transplantation model, where we detected somatic mutations in several AML-associated signal transduction genes, including Flt3, Cbl and Kras ( Desai et al., Leukemia 2020). Taken together, our results strongly suggest that mutations in genes involved in signaling cascades, particularly those in the RAS/MAPK pathway, cooperate with the C/A fusion and lead to complete malignant transformation. We aim to functionally validate these findings by establishing mouse models that harbor the C/A fusion as well as mutations in Ptpn11 and other signal transduction genes. These murine models will be immensely valuable for gaining a deeper insight into PICALM/MLLT10-mediated leukemogenesis, studying cooperating mutations and for testing new targeted therapies.
hematology