gms | German Medical Science

Objective: Pericytes have an important role in controlling angiogenic mechanisms in the developing brain and in primary brain tumours (gliomas). Previous studies suggested that newly generated pericytes derive from mitotically active, mature pericytes, which can be identified by the expression of marker molecules and their close association with endothelial cells. Using a newly established lineage-tracing model we show that mature pericytes largely originate from marker-negative cells that are not attached to endothelia. The tumour-supportive role of these pericyte progenitor-like cells (pericyte-PLC) is shown in transgenic mouse models.

Methods: In our mouse model stable fluorescence-reporter activity was obtained after tamoxifen (TAM) inducible genetic recombination controlled by a modified nestin-promoter element (nestin-creER2). To study neoangiogenesis in the brain we orthotopically inoculated mice with syngeneic glioma cells and observed pericyte expansion during. tumourigenesis. The identity of these cells was further verified by electron microscopy. To study if these cells are resident to the brain or originate from the periphery we produced bone-marrow (BM) chimeric mice. And finally, in a functional experiment we crossed our nestin-reporter mice to inducible Diphteria Toxin A (DTA) mice to specifically ablate these cells.

Results: Throughout tumour growth recombined cells expanded in the tumour mass and labelling-experiments with different thymidine-analogues revealed their continuous proliferation. Strikingly, the vast majority of recombined cells in early gliomagenesis was negative for all pericyte-markers investigated (78% of all recombined cells), but they acquired pericyte markers at later stages of tumour expansion (81% PDGFR-B-, 76% NG2- , 74% desmin- and 84% CD146-positive recombined cells). Combinatorial immunofluorescence labelling showed that mature pericytes simultaneously express multiple markers. Few new pericytes were generated in the tumour-periphery or in normal brain. Using this model we were able to lineage-trace in average 34% of all intra-tumoural pericytes. This suggests that a very large part of newly generated pericytes derive from nestin-positive, PDGFR-B-, NG2-, desmin-, CD146-negative cells. Lineage-tracing experiments in bone-marrow chimeric glioma models indicated that pericyte-PLCs are endogenous to the brain. In a functional experiment with DTA-induced cell death under the nestin-promotor we observed a 50% decrease of pericyte-PLCs leading to a 34% decreased tumor vessel density and a reduced glioma volume by 33%.

Conclusion: All in all, our data indicate that newly generated pericytes origin, to a large extent, from a previously unrecognised (pericyte-marker negative) brain-resident precursor cell and that depletion of this pericyte progenitor can be exploited as a novel therapy against glioma.