The ORIGIN4MB project aimed to define the cell of origin of Group 4 MB and generate the first preclinical model of this disease by targeting MYCN and CDK6 activation in unipolar brush cells (UBC) progenitors.
Specific scientific objectives were:
1. To infer which promoters could be used to target specifically UBCs.
2. To uncover whether activation of MYCN and/or CDK6 in the developing cerebellum leads to Group 4 MB formation.
3. To infer whether activation of MYCN and/or CDK6 specifically in UBC early or late progenitors leads to Group 4 MB formation.
4. To validate whether the generated mouse model for Group 4 MB mimics the human disease.
A brief summary of the progress made towards the achievement of each objective is given below.
Objective 1) Using immunofluorescence and lineage tracing in advanced genetic mouse models, we were able to define the spatial and temporal expression of key UBC progenitor markers, including Math1, Lmx1a, Lhx2, and EOMES, to distinguish early from late UBC progenitors. This was achieved by collecting embryos at all ages and performing immunofluorescence for the above-mentioned markers, showing that LHX2 starts being expressed at embryonic day 11 (E11), whereas Math1 starts being expressed at E12 in the rhombic lip, and Lmx1a and Eomes at E13.
In parallel, lineage tracing was performed by crossing Math1-CREER, Lhx2-CREER, Lmx1a-CREER or EOMES-CREER mice with Rosa-YFP reporter mice. Temporal control of recombination using tamoxifen allowed for an assessment of whether progeny from each promoter specifically gave rise to UBCs. It was possible to infer that Math1+ progenitors during embryonic days 13 – 14 (E13/E14) give rise to UBCs (but not after that, since Math1+ progenitors afterwards become restricted to granule cell progenitors), whereas Lmx1a+ progenitors give rise to UBCs from E13 onwards. In the case of LHX2+ progenitors these don’t seem to give rise to any UBCs. On the other hand, EOMES expression persists in UBC progenitors, suggesting EOMES as a UBC-specific marker at both embryonic and postnatal stages (being expressed in UBCs since their genesis at E13). This allowed for a selection of promoters that label UBCs to be used for the following experiments to drive oncogene activation selectively in UBC progenitors.
Objective 2) Lentiviral vectors encoding MYCN and/or CDK6, coupled to fluorescent reporters (mCherry and mCerulean, respectively), were injected in utero into the cerebellum of mouse embryos at different developmental stages (E15–E19) to identify critical temporal windows of tumor susceptibility. These were injected in the rhombic lip, the embryonic region where MBG4 arises from. The individual and combined contributions of MYCN and CDK6 to tumor initiation and progression were assessed by injecting each oncogene alone or together. Unfortunately, since in utero injections are very delicate and complex, many obstacles were encountered: pregnant females would open the surgery wound post-surgery (leading to their death), some embryos would die from the injection, or pregnant females would have spontaneous abortions post-surgery. Therefore, only a few injected embryos were later on able to be born and survive. However, these did not give rise to tumours – another type of cells (granule cells) were labelled with the reporter, indicating that the wrong cells were targeted by the lentivirus.
Objective 3) UBC progenitor cells were isolated from embryonic mouse cerebella using Math1-based reporter models (taking into consideration the embryonic timings inferred in objective 1). These cells were then sorted, and cultured to generate neurospheres. These neurospheres were characterized by immunocytochemistry, flow cytometry and qPCR to confirm UBC progenitor identity, reporter expression, and proliferative capacity. Lentiviral vectors expressing MYCN and/or CDK6, each coupled to fluorescent reporters and to a luciferase-encoding cassette, were used to infect dissociated neurosphere cells, with successful overexpression validated by fluorescence, qPCR, and proliferation assays.
Cells expressing the oncogenes were then grafted into the cerebellum of postnatal day 3 (P3) pups via stereotaxic surgery to mimic tumor initiation in the juvenile brain. Tumor formation was detected after cell grafting, and its development was then monitored using bioluminescence in vivo, followed by post-mortem analyses including immunostaining for Group 4 MB markers and single-cell RNA sequencing.
