Different MLL Breakpoints Influence Leukemia Characteristics of MLL-Rearranged Cells in a Human CRISPR/Cas9-Based Model

MLL-rearranged (MLLr) leukemia is caused by translocations of the human chromosome 11 leading to fusions of the MLL gene to different partner genes, such as AF4 and AF9. The resulting MLL chimera induces malignant transformation of cells and the transcriptional dysregulation of MLL target genes, such as the homeobox (HOX) genes. The majority of fusions occurs in a breakpoint cluster region ranging from MLL exon 8 to exon 14. Leukemias with breakpoint in MLL intron 11 are most frequently observed in children less than 6 months old and present with a more aggressive clinical course leading to worse prognosis compared to leukemias with other MLL breakpoints. However, the reason for preferential translocation of some partner genes and the consequences of the breakpoint location for leukemia outcome remains unknown.

We established a model system for MLL-AF4 and -AF9 leukemia using CRISPR/Cas9 to induce targeted double-strand breaks in CD34+ hematopoietic stem cells derived from human cord blood. The induced double-strand breaks led to balanced chromosomal translocations and the formation of fusion genes involving the MLL gene. The CRISPR/Cas9 complex was thereby delivered to the cells as ribonucleoprotein with nucleofection. Nucleofected cells were cultured under myeloid conditions to reach lineage commitment. We were able to grow a pure population of cells carrying MLL rearrangements after 50-80 days of culturing as confirmed by fluorescence in situ hybridization (FISH). To study the difference between cells with breakpoint in intron 9 (MLL[9]) and cells with breakpoint in intron 11 (MLL[11]), we generated MLL-AF4- and MLL-AF9-rearranged cells from the same cord blood donor carrying translocations with either of the MLL breakpoints. Confirmation of the fusion at the targeted gene loci was done by PCR and Sanger sequencing. Gene expression of common MLLr leukemia target genes was analyzed with quantitative RT-PCR, differentiation was analyzed with flow cytometry, and morphology was determined by May-Gruenwald-Giemsa staining of cytospins. The response of cells to chemotherapeutic agents was tested with flow cytometric Annexin V/PI staining after cytarabine treatment.

Translocations with both breakpoints led to leukemic transformation of cells with increased proliferative capacity. Independent of the breakpoint, MLLr cells strongly expressed the MLLr leukemia-associated transcription factors MEIS1 and HOXA9, and the proto-oncogene MYC. Flow cytometric characterization of cells revealed less differentiation of MLL[11]-AF9 cells in terms of CD15, CD38 and CD64 surface expression in comparison to MLL[9]-AF9 cells. At the same time, MLL[11]-AF9 cells expressed more CD117, representing an immature phenotype. Similarly, MLL[11] cells with both fusion partners morphologically showed less maturation as they contained less promyelocytes and myelocytes. Functionally, MLL[11] cells demonstrated more resistance to cytarabine than MLL[9] cells.

In conclusion, we established an MLLr leukemia model with breakpoints in MLL intron 9 and MLL intron 11 demonstrating continuous cell growth in in vitro culture systems. Our model represents clinical aspects of MLLr leukemia by exhibiting less differentiation and maturation of MLL[11] cells with higher resistance to chemotherapy. Further characterization by RNA-sequencing could address the differential impact of distinct MLL breakpoints leading to a deeper understanding of the disease.

No relevant conflicts of interest to declare.