Editor—Proteus syndrome (MIM 176920) is a rare, congenital, hamartomatous disorder, which is a member of a group of local overgrowth diseases. Happle1 proposed that some of these disorders are the result of the action of a lethal gene that can only survive in the mosaic state, which arises from an early somatic mutation or from a half chromatid mutation. Such a mechanism has been shown to be the underlying basis of McCune-Albright syndrome (MIM 174800).2 One of the mandatory diagnostic criteria for Proteus syndrome is a mosaic distribution of lesions and sporadic occurrence, entirely consistent with Happle's hypothesis.

Currently, little is known about the molecular causes of Proteus syndrome. It is, however, likely that the overgrowth of tissue involves all germ layers. This may be because of hyperproliferation, an absence of appropriate apoptosis, or alternatively cellular hypertrophy. There have been few investigations into the molecular basis of Proteus syndrome. Zhou et al 3 recently identified PTEN mutations in a patient with a Proteus-like syndrome. Germline PTENmutations are found in a high proportion of patients with Cowden (MIM 158350) and Bannayan-Riley-Ruvalcaba (BRR) syndromes (MIM 153480),4-7 which share many features of Proteus syndrome. These observations make PTEN a strong candidate for a gene mutated in Proteus syndrome. To investigate this possibility, we examined eight patients with Proteus syndrome forPTEN mutations. All were unrelated and had classical Proteus syndrome using published diagnostic criteria.8 Samples were obtained with informed consent and local ethical review board approval. Fibroblasts were cultured from skin biopsies obtained from normal tissue and from regions of overgrowth. Genomic DNA was extracted from cultured cells using a standard sucrose lysis technique. PTENmutational analysis was performed by PCR based conformational specific gel electrophoresis using published oligonucleotides9 and semi-automated sequencing using an ABI 377 Prism sequencer. A common exon 4 polymorphism was observed in three of the patients, but no missense or truncating mutations in any of the eight samples were detected, suggesting that mutation in PTENis unlikely to be a common cause of Proteus syndrome.

We evaluated PTEN as a candidate gene because of its role in the overgrowth syndrome Cowden disease and the recent report of a PTEN mutation in a boy with Proteus-like syndrome.3 PTEN plays a role in the regulation of PI3 kinase signalling, which is involved in the control of apoptosis and cell cycle progression.10 Hence, by removing the regulatory effects of PTEN on PI3 kinase signalling, deregulated cellular growth could occur. PTEN also appears to play a role in the regulation of cell size and a role for the PI3 kinase signalling pathway in the determination of organ size in mammals has been reported.11 The boy reported by Zhouet al 3 with Proteus-like syndrome had a germline single base transversion resulting in an Arg 335 to Ter substitution in PTEN. A secondPTEN mutation resulting in Arg 130 to Ter was found in DNA from a naevus, lipoma, and an arteriovenous mass. The authors postulated that the first germline mutation gave rise to many of the features of BRR and that the second hit occurred early in embryogenesis causing mosaicism. In our study we did not detectPTEN mutations in any of the Proteus syndrome patients we examined. Zhou et al 3 similarly failed to detect anyPTEN mutations in five patients with classical Proteus syndrome; their patient withPTEN mutations did not fulfil the stringent diagnostic criteria for Proteus syndrome.

Mutations in the coding region of PTEN do not appear to be implicated in classical Proteus syndrome.PTEN may still be involved, as our finding does not preclude the possibility that it may be aberrantly imprinted in Proteus syndrome, for example by promoter methylation,12 leading to reduced PTEN expression. Given the innumerable possibilities for a molecular basis of Proteus syndrome, the identification of which genes are disrupted will prove difficult. One strategy for dissecting the molecular pathways of Proteus and other overgrowth syndromes is through examining the expression patterns of genes in affected and unaffected tissues, which is becoming feasible with the advent of microarray technology.13