Here we report on a novel PTEN mutation that was found as a germline mutation in a patient with meningioma and glioma. To our knowledge this is the first germline mutation reported for brain tumours of multiple lineages. The mutation changes an arginine residue to a glutamine in the C-2 domain of PTEN. This changes the charge of that region (from positive to neutral). Functions of the C-2 region may involve membrane targeting through interactions with membrane phospholipids or association with as yet unknown proteins (Lee, 1999
). It has also been shown that the C-2 domain functions to position the phosphatase domain correctly with respect to its substrates. Therefore, mutations in the C-2 domain could alter the overall enzymatic activity of PTEN, without necessarily affecting its phosphatase activity.
We demonstrate that this mutation has functional consequences because the mutant molecules allow cells to proliferate at much higher levels and do not induce or prevent apoptosis. Given that mice with a targeted PTEN mutation on one allele and a wild-type PTEN molecule on the other allele show abnormalities (Di Cristofano et al, 1998
), the heterozygous nature of the mutation in the patient does not diminish the importance of the functional findings. The functional consequences of the mutation we have found may help explain why the tumour arose. Of course, it is possible that this mutation has nothing to do with the development of the tumour, or that this is a syndrome with a very low penetrance. However, it seems likely that this mutation also has functional consequences in the tumour. Competition for limited substrates or associated molecules may interrupt normal PTEN function sufficiently to lead to increased proliferation and diminished apoptosis. In fact, the mutant molecule is incapable of inducing apoptosis in BaF/3 cells, which contain endogenous functional PTEN. In some experiments the percentage of apoptotic cells after transfection of R234Q PTEN was even lower than GFP alone, perhaps because endogenous PTEN activity was somewhat inhibited. Another indication that this mutation may not be solely a PTEN null mutant, but can act as a negative competitor, stems from experiments investigating PKB activity. In R234Q PTEN transfectants, PKB activity could not be induced by insulin treatment, in contrast to parental cells or cells transfected with wt PTEN. This result suggests that the R234Q mutation acts to inhibit the residual PTEN activity present in U87MG cells and can therefore be considered as acting as a negative competitor. Thus, the functional consequences of the mutation in transfection studies are consistent with high proliferative activity. Together, these findings suggest that the Arg234Gln missense mutation in PTEN has oncogenic properties.
The main PTEN substrate has proved to be phosphatidyl inositol (3,4,5)-triphosphate (PIP-3) (Myers and Tonks, 1997
). Previously, it has been shown that glioma cells lacking wild-type PTEN are characterised by elevated levels of PIP-3 (Li and Sun, 1998
). As a consequence, the activity of protein kinase B (PKB/Akt) was found to be elevated, resulting in prevention of apoptosis and increased proliferation. Our results fit this picture in that U87MG cells stably transfected with the R234Q mutant PTEN also show increased proliferation and increased constitutive PKB activity.
One of the main functions of PTEN is to keep PIP-3 levels low. Loss of PTEN function results in increased PIP-3 levels, leading to hyperactive PKB/Akt. The mutation we have found may similarly affect PTEN activity, finally resulting in increased PKB activity, increased cell proliferation, and diminished apoptosis.
Not many PTEN mutations described in the literature have been associated with the C-2 domain, although Lee et al (1999)
reported that mutagenesis of the C-2 domain reduced the tumour suppressive activity of PTEN. Our findings are consistent with these results. It is tempting to speculate, based on the X-ray crystallography model of PTEN (see ), that the C-2 loop can associate with other proteins. Interestingly the mutation is exactly in the middle of the loop, at a site readily accessible for interaction with other molecules. Future research will undoubtedly identify such C-2 domain-interacting molecules, most likely further explaining the oncogenic nature of the PTEN mutation we have described here. It has been proposed that mutation of basic residues in the C-2 domain reduces the membrane affinity of PTEN and its ability to suppress the growth of glioma tumour cells (Lee et al, 1999
Figure 6 Three-dimensional structure of PTEN, showing the location of the mutation. The mutated amino acid is depicted in yellow in the C-2 domain (green). The substrate binding site of PTEN, located in the phosphatase domain (blue), is shown in red. The model (more ...)
Germline mutations of the PTEN gene are responsible for several inherited conditions, including Cowden disease, Bannayan-Zonana syndrome and Lhermitte-Duclos disease. Here we present a patient with a heterozygous PTEN germline mutation but with no signs of any of these disorders. The patient had an oligodendroglioma, and, in the past, a benign meningioma. The PTEN gene is a tumour suppressor gene and inactivation of both alleles is required for neoplastic transformation. An inherited mutation in one allele with somatic loss of the second allele may result in oncogenesis. Loss of the second allele in human gliomas often occurs by deletion of a part of chromosome 10. However, no LOH could be detected in tumour-derived DNA samples. Inactivation of PTEN by a somatic point mutation could also be excluded, since sequencing of all PTEN coding exons in tumour DNA revealed no changes other than the Arg234Gln germline mutation. One may hypothesise that promoter hypermethylation might function as the second hit at the PTEN gene in the genesis of the patient's tumour. Alternatively, a somatic point mutation could be present in the regulatory sequences of the gene. On the other hand, it has been reported that in a mouse model, inactivation of one PTEN allele has important consequences for cell survival and proliferation (Di Cristofano et al, 1999
). This suggests that PTEN haplo-insufficiency could be an important factor in cell transformation.