Brain Development & Tumours

Prof. Sebastian Brandner, Division of Neuropathology and Department of Neurodegenerative Disease, and Dr. Tom Jacques, Division of Neuropathology

Research focus on wnt signalling in stem cell differentiation and in vivo model systems to study PTEN and other oncogenic pathways in brain development and tumourigenesis. For further details see Head of Division.

Current research projects

• Modelling intrinsic brain tumours: oncogenic signalling in neural stem cells
• Establishing human brain cancer cell lines
• Research on human brain cancer

1. Modelling intrinsic brain tumours: Oncogenic signalling in neural stem cells:

Intrinsic benign and malignant brain tumours, such as astrocytomas and glioblastoma are thought to originate from a population of stem cells in the adult brain. Stem cells divide in the adult brain to form more mature cells such as nerve cells (neurones) astrocytes (supporting cells) or oligodendrocytes (myelin forming cells that wrap processes of neurones in the white matter). We are working on the mechanisms of how mutations in genes that control cell division and migration of stem cells can cause brain tumours.

A novel aspect of our work is that the type of tumours depends on the pattern of mutations in stem cells, even when mutated stem cells are taken away from their normal environment and are placed in other areas of the brain. Knowledge of this mechanism is an important step towards the understanding where brain tumours originate from and how a certain type of brain tumour forms. 

Development of diffusely infiltrating tumours in mice lacking PTEN and p53 in the SVZ stem cell compartment (upper panel). Instead, deletion of the tumour suppressor genes RB and p53 in the same cell population gives rise to a completely different type of tumours, so-called primitive neuroectodermal tumours. This difference highlights the importance of the combination of genotypes in determining he histological appearance of the brain tumour. How does the model work?

We inactivate tumour suppressor genes (PTEN, RB, p53) in the neural stem cell compartment. By injecting Cre-expressing virus into the ventricles of conditional knockout mice (PTENlox/lox, p53lox/lox and Rblox/lox in various combinations) the targeted genes are recombined only in cells located near the sub-ventricular zone (SVZ) which contains the largest known population of neural stem cells. In mice where Rb and p53, or the trio of the genes, PTEN, P53 and Rb are recombined, after several months, tumours of a specific phenotype resembling that of a human PNET develop. Instead, a tumour resembling human glioma (similar to oligoastrocytomas) is induced in mice where PTEN and P53 are targeted. 

Microneoplasia develop in the subventricular zone: Cumulative maps showing the localisation of small lesions. Each dot represents microneoplasia or a focal cluster of tumour cells (ca 200 mm). Each colour represents an individual animal. The map summarises lesions in 14 Rb/p53 mice, 6 Rb/p53/PTEN mice and 14 PTEN/p53 mice.

Several weeks to months prior to the development of large tumours, small neoplastic lesions can be observed which we call microneoplasia. The assumption is that during the considerable time it takes to develop evident tumours additional genetic lesions are accumulated on top of the ones induced by Cre-recombination.

We are currently profiling these brain tumours using Affymetrix Exon arrays and compare the tumour signatures with stem cell signatures.

Comparison of the molecular signature of primary experimental tumours with those of in vitro transformed neural stem cells, grafted into a recipient ,mouse and resulting in a tumour graft with striking similarity to the primary brain tumour of the same genotype (see Jacques et al., EMBO J 2010).

2) Establishing human brain cancer cell lines:

The aim of our brain tumour bank is to collect specimens for research and to derive cell lines from living tumour cells. This collection will be used for brain cancer research and material can be made available to researchers at UCL, in the UK and internationally

The brain tumour bank was established in 2008, and we are banking 20 samples every month, and one cell line is successfully established every week. In parallel, all gliomas are being characterised by state of the art molecular pathology tests, which allows us to correlate clinical findings, pathological features and the growth and biological behaviour of these tumours. The availability of these tests is also essential for a selection of defined tumour subgroups for research projects.

3) Research on human brain cancer:

We have recently started pilot experiments to establish neurografting conditions, growth dynamic and histological appearance of such neurografts.

Characterisation of the xenografts (Histology, Affymetrix Exon expression micro array and Illumina DNA SNP microarray). Comparison with the primary tumour counterpart of the same patient: this will be possible using frozen tumour tissue from the same patients. Comparison of expression levels will be done with Affymetrix Exon expression arrays and potential genomic changes will be rested with Illumina SNP arrays. Both technologies are established and have preliminary data

Examples of cultured glioma cells, following recently established protocols: The primary tumour is first cultured as free floating tumour sphere (A) and then transferred to an extracellular matrix (ECM) environment (B), allowing for a rapid expansion of primary cells. Importantly, reverting back to primary conditions can always revert the adherent cultures back to tumour spheres as shown in (A). C: Example of a tumour graft: Histology shows an expansive tumour in the forebrain of a mouse. D: Ex vivo MRI of a formalin fixed brain do demonstrate the resolution of the 9.4T MRI.