Cre recombinase promotes leukemogenesis in the presence of both homozygous and heterozygous FLT3-ITD

Straube et al., but also provide the ﬁ rst evidence of enhanced leukemogenesis of FLT3 -ITD heterozygous by Cre. FLT3 -ITD knock-in mice have been used to identify cooperative partners, also in the presence of Cre [10]. Our data strongly support the ﬁ ndings of Straube et al. and indicate the need for a careful study design and interpretation of the data when using the Cre-loxP recombination system. In our hands, ITD/ITD in the presence of Cre activity does not allow to investigate the role of MCSFR or FGFR2 in the pathogenesis of FLT3 -ITD. Whether ITD/o in the presence of Cre is suitable for testing of cooperative events remains to be determined.


TO THE EDITOR:
Cre recombinase murine models allow the expression of mutated genes in a cell type-specific manner or via an inducible mechanism and has revolutionized biomedical research.However, these models may be associated with some caveats, such as off-target effects and lack of fidelity.In the issue 4 of LEUKEMIA in 2023, Straube et al., described an unexpected observation where Cre expression alone was able to drive early acute myeloid leukemia (AML) in the context of FLT3-ITD/ITD (homozygous) [1].Herein, we found that expression of Cre recombinase induced early AML in the presence of homozygous FLT3-ITD in different models, but not in the presence of the Kit D814V mutation (murine homolog of human KIT D816V mutation).Moreover, Cre recombinase also promoted leukemogenesis in the presence of heterozygous FLT3-ITD.
To identify cooperating partners for the FLT3-ITD in the development of AML, we analyzed the activation of ≥ 42 receptor tyrosine kinases in primary samples from AML patients using a phospho-kinase antibody array.The phosphorylated kinases, Macrophage colony stimulating factor receptor (MCSFR) and Fibroblast growth factor receptor 2 (FGFR2) were detected in 78% and 31% of AML patients (n = 90), respectively (Fig. 1A and data not shown).The expression of MCSFR on blasts was confirmed in all analyzed primary samples with phosphorylated kinases (n = 32) by flow cytometric analysis (Fig. 1B).In a separate cohort, we detected MCSFR expression almost in all primary samples from AML patients (n = 125) by flow cytometric analysis (data not shown).Importantly, MCSFR was identified as a therapeutic target in AML leukemia [2].Moreover, MCSFR is crucial for leukemic stem cells (LSC) potential induced by the MOZ-TIF2 fusion [3].Interestingly, FGFR2 has been suggested to be important for leukemic-regenerating cells (LRCs) that are induced by chemotherapy and responsible for disease relapse [4].
At the molecular level, chromatin profiling revealed the existence of a poised enhancer in intron 15 of Flt3 in FLT3-ITD/o mouse hematopoietic stem/progenitor cells but not in wildtype counterparts, which were marked by increased chromatin accessibility, enrichment of H3K4me1, and lower levels of H3K27ac (Fig. 1I).Cre expressions resulted in a fully active enhancer mapping at intron 15 of Flt3 and increased the expression of Flt3 gene (Supplementary Fig. 3A, B), suggesting that Cre-mediated recombination may facilitate chromatin activation.Recently, we described that the FLT3-ITD mutation alone can remodel the chromatin landscape to prime the development of fullblown leukemia in cooperation with other mutations [8].A possible causative mechanism may involve Cre cleavage of genomic sites activated by FLT3-ITD.We focused on 3786 genomic regions that gained chromatin accessibility in the presence of FLT3-ITD [8], and then scanned for three loxP motif patterns to identify pseudo loxP sites [9] (Fig. 1J).Interestingly, we identified two pseudo-loxP sites mapping to the FLT3-ITD open chromatin region (Fig. 1K).Whether Cre enhances leukemogenesis of FLT3-ITD through these pseudo-loxP sites needs to be determined.In ongoing studies we wish to understand the underlying molecular mechanism for the development of AML by Cre and FLT3-ITD in more details.
In summary, our data not only confirm the cooperation between Cre and FLT3-ITD homozygous in the induction of AML described by  1A).E Flow cytometry analysis of spleen samples from mice #7003 and #7013 demonstrated a population of myeloblast/immature cells with lower side scatter (SSC) and CD45 dim expression, which were positive for CD11b, c-Kit, and Gr1, but negative for CD3 and CD19 (Supplemental Fig. 1B).Straube et al., but also provide the first evidence of enhanced leukemogenesis of FLT3-ITD heterozygous by Cre.FLT3-ITD knock-in mice have been used to identify cooperative partners, also in the presence of Cre [10].Our data strongly support the findings of Straube et al. and indicate the need for a careful study design and interpretation of the data when using the Cre-loxP recombination system.In our hands, ITD/ITD in the presence of Cre activity does not allow to investigate the role of MCSFR or FGFR2 in the pathogenesis of FLT3-ITD.Whether ITD/o in the presence of Cre is suitable for testing of cooperative events remains to be determined.

Fig. 1
Fig. 1 Cre recombinase promoted leukemogenesis in the presence of FLT3-ITD.A Representative antibody arrays from two patients with AML (#85 and #186).Phosphorylation of FLT3 and MCSFR was observed in patients #85 (FLT3-TKD) and #186 (FLT3-ITD), but not in LAMA84 cells.LAMA84 cells were isolated from a patient with chronic myeloid leukemia in a blast crisis.We did not observe FLT3 or MCSFR phosphorylation in any of the healthy controls analyzed (n = 8).Hybridization signals at the corners (three or four) served as positive controls.MCSFR and FGFR2 belong to the most phosphorylated receptor in the AML specimens in our analysis.B Representative flow cytometry analyses confirming expression of FLT3 and MCSFR in patient #186.C Survival curves of ITD/ITD; Mcsfr flox ; Mxl-Cre, ITD/ITD; Fgfr2 flox/o ; Mxl-Cre, ITD/ITD; Mxl-Cre (3 strains together ITD/ITD;N;Cre), and ITD/ITD mice.The animals were not treated with polyinosinic-polycytidylic acid (polyIC), as this was scheduled fo around 6 weeks after birth.No phenotypic differences were observed between TD/ITD; Mcsfr flox ; Mxl-Cre, ITD/ITD; Fgfr2 flox/o ; Mxl-Cre, and ITD/ITD; Mxl-Cre, especially there was no additional acceleration or delay of the disease in TD/ITD; Mcsfr flox ; Mxl-Cre, and ITD/ITD; Fgfr2 flox/o ; Mxl-Cre mice.****p < 0.0001.D Representative Pappenheim-stained blood smears and cytospins of bone marrow (BM), spleen and liver, and hematoxylin and eosin (H&E)-stained histopathology of BM, spleen, and liver from #7003 (ITD/ITD; Mcsfr flox ; Mxl-Cre) and #7013 (ITD/ITD; Mxl-Cre mice) mice.Note infiltration of myeloblasts in these organs and in the lung (Supplementary Fig.1A).E Flow cytometry analysis of spleen samples from mice #7003 and #7013 demonstrated a population of myeloblast/immature cells with lower side scatter (SSC) and CD45 dim expression, which were positive for CD11b, c-Kit, and Gr1, but negative for CD3 and CD19 (Supplemental Fig.1B).F Representative Pappenheim-stained blood smears and cytospins of bone marrow (BM), spleen and liver, and hematoxylin and eosin (H&E)stained BM, spleen, and liver samples from #1965 (ITD/o; Fgfr2 flox/flox ; Mxl-Cre).G Flow cytometry analysis of the spleen sample showing a population of myeloblast/immature cells with lower side scatter (SSC) and CD45 dim expression in sample from mouse #1965.These blasts were positive for CD11b, c-Kit, and Gr1, but negative for CD3 and CD19 (data not shown).H Survival curves of ITD/o; Mcsfr flox ; Mxl-Cre, ITD/o; Fgfr2 flox ; Mxl-Cre (2 strains together ITD/o;N;Cre), ITD/o, and wildtype (WT) mice.One mouse from ITD/o; Fgfr2 flox ; Mxl-Cre was treated with polyIC.***p < 0.001, ****p < 0.0001.I Illustration of chromatin profiling on accessibility (ATACseq) and modification states (ChIPseq on H3K4me1 and H3K27ac) in wildtype vs. ITD/o mouse HSPCs.An FLT3-ITD-associated enhancer was identified in the intron 15 of Flt3 gene in ITD/o cells, marked by a high enrichment of H3K4me1, and modest levels of ATACseq and H3K27ac.Interesting, Straube et al. reported the presence of a neomycin resistance cassette (NRC) flanked by loxP sites in intron 15 of the Flt3 gene in FLT3-ITD mice, and the excision of NRC in the presence of Cre [1].J Visualization of motif logos representing pseudo loxP sites using three parameters.K Identification of chromatin regions containing pseudo loxP sites in the context of accessibility gained by FLT3-ITD.

F
Fig. 1 Cre recombinase promoted leukemogenesis in the presence of FLT3-ITD.A Representative antibody arrays from two patients with AML (#85 and #186).Phosphorylation of FLT3 and MCSFR was observed in patients #85 (FLT3-TKD) and #186 (FLT3-ITD), but not in LAMA84 cells.LAMA84 cells were isolated from a patient with chronic myeloid leukemia in a blast crisis.We did not observe FLT3 or MCSFR phosphorylation in any of the healthy controls analyzed (n = 8).Hybridization signals at the corners (three or four) served as positive controls.MCSFR and FGFR2 belong to the most phosphorylated receptor in the AML specimens in our analysis.B Representative flow cytometry analyses confirming expression of FLT3 and MCSFR in patient #186.C Survival curves of ITD/ITD; Mcsfr flox ; Mxl-Cre, ITD/ITD; Fgfr2 flox/o ; Mxl-Cre, ITD/ITD; Mxl-Cre (3 strains together ITD/ITD;N;Cre), and ITD/ITD mice.The animals were not treated with polyinosinic-polycytidylic acid (polyIC), as this was scheduled fo around 6 weeks after birth.No phenotypic differences were observed between TD/ITD; Mcsfr flox ; Mxl-Cre, ITD/ITD; Fgfr2 flox/o ; Mxl-Cre, and ITD/ITD; Mxl-Cre, especially there was no additional acceleration or delay of the disease in TD/ITD; Mcsfr flox ; Mxl-Cre, and ITD/ITD; Fgfr2 flox/o ; Mxl-Cre mice.****p < 0.0001.D Representative Pappenheim-stained blood smears and cytospins of bone marrow (BM), spleen and liver, and hematoxylin and eosin (H&E)-stained histopathology of BM, spleen, and liver from #7003 (ITD/ITD; Mcsfr flox ; Mxl-Cre) and #7013 (ITD/ITD; Mxl-Cre mice) mice.Note infiltration of myeloblasts in these organs and in the lung (Supplementary Fig.1A).E Flow cytometry analysis of spleen samples from mice #7003 and #7013 demonstrated a population of myeloblast/immature cells with lower side scatter (SSC) and CD45 dim expression, which were positive for CD11b, c-Kit, and Gr1, but negative for CD3 and CD19 (Supplemental Fig.1B).F Representative Pappenheim-stained blood smears and cytospins of bone marrow (BM), spleen and liver, and hematoxylin and eosin (H&E)stained BM, spleen, and liver samples from #1965 (ITD/o; Fgfr2 flox/flox ; Mxl-Cre).G Flow cytometry analysis of the spleen sample showing a population of myeloblast/immature cells with lower side scatter (SSC) and CD45 dim expression in sample from mouse #1965.These blasts were positive for CD11b, c-Kit, and Gr1, but negative for CD3 and CD19 (data not shown).H Survival curves of ITD/o; Mcsfr flox ; Mxl-Cre, ITD/o; Fgfr2 flox ; Mxl-Cre (2 strains together ITD/o;N;Cre), ITD/o, and wildtype (WT) mice.One mouse from ITD/o; Fgfr2 flox ; Mxl-Cre was treated with polyIC.***p < 0.001, ****p < 0.0001.I Illustration of chromatin profiling on accessibility (ATACseq) and modification states (ChIPseq on H3K4me1 and H3K27ac) in wildtype vs. ITD/o mouse HSPCs.An FLT3-ITD-associated enhancer was identified in the intron 15 of Flt3 gene in ITD/o cells, marked by a high enrichment of H3K4me1, and modest levels of ATACseq and H3K27ac.Interesting, Straube et al. reported the presence of a neomycin resistance cassette (NRC) flanked by loxP sites in intron 15 of the Flt3 gene in FLT3-ITD mice, and the excision of NRC in the presence of Cre [1].J Visualization of motif logos representing pseudo loxP sites using three parameters.K Identification of chromatin regions containing pseudo loxP sites in the context of accessibility gained by FLT3-ITD.