The neurofibromatoses represent the major group of inherited syndromes that cause predisposition to the development of tumours of the nervous system. Two main forms of neurofibromatosis have been clearly identified as being distinct diseases, each one defined by precise diagnostic criteria and caused by different genetic defects. They are called neurofibromatosis type 1 (NF1) and neurofibromatosis type 2 (NF2). Our strategy for identifying the possible mechanism of action of schwannomin is based in large part on our skills in creating and analysing animal models.

In man, the NF2 gene is expressed in the form of two alternative messenger species, which differ in the optional presence of exon 16. It has been suggested that only isoform 1 of schwannomin has tumour suppressor activity.
To demonstrate this hypothesis, we have used the Cre/loxP system to obtain two different alleles of the Nf2 gene. Our preliminary results show that the expression either of the two isoforms of schwannomin is sufficient to correct embryonic lethality associated with the complete absence of schwannomin.
The phenotypic analysis of these mice will enable us to assess the tumour suppressor role of each isoform.

The ability to produce animals with multiple genetic alterations means that we can study cooperation between specific genes in Schwann cell tumourigenesis.
Recently, two American teams obtained mouse mutants with different disrupted allelic combinations of the Nf1 and p53 genes in the germline, which, like Nf2, are located on the same chromosome. This work confirmed the oncogenic cooperation of the two mutations. We have generated Nf1+/-;Nf2flox/flox;PoCre mice.
These mice develop multiple tumours from the third month of life, which develop from the cranial or peripheral nerves or from the spinal roots.
Genotyping of the tumours revealed double inactivation of the Nf1 (by LOH) and Nf2 (by biallelic Cre/loxP recombination) genes. To date, we have identified several neurofibromas and malignant tumours of Schwann cell origin (MPNST) in older mice. We are currently completing the morphological characterisation of these tumours.

Meningiomas represent the main cause of neurological exacerbation in NF2patients. To date, NF2 gene inactivation is the only genetic alteration known to be involved in meningeal tumourigenesis. Meningiomas develop from arachnoid cells, which are bathed in the cerebrospinal fluid (CSF) that circulates in the sub-arachnoid spaces.
Our strategy for targeting arachnoid cells was to inject adenoCre into the CSF of Nf2flox/flox mice. We have shown that the consequence of inactivating the Nf2 gene in the mouse is the development of meningiomas in 40% of mice from the age of 4 months, suggesting that loss of the Nf2 gene is critical for tumour initiation.
However, in these mice, p53 hemizygosity did not result in the acceleration or the appearance of meningiomas. Recently, a second "tumour suppressor" gene, DAL-1, was identified, that encodes a protein (4.1B) with high structural homology with the NF2 protein, and that appears to be lost in 60% of sporadic meningiomas associated with NF2.
In addition, it was shown that the combined loss of expression of schwannomin and protein 4.1B is often seen in malignant meningiomas, indicating their possible cooperation in the progression of meningioma. These two proteins could be critical regulators of leptomeningeal cell proliferation.
We will test this hypothesise by generating mice with disrupted Protein 4.1B and mice with double inactivation of the Nf2 and Protein 4.1B genes.