Prompted by the predisposition to leiomyomata and leiomyosarcoma seen in carriers of germline mutations in FH we have screened the full coding sequence and splice junctions of FH for somatic mutation in a series of sporadic leiomyomas and leiomyosarcomas. Although we cannot exclude the possibility that some mutations may have been missed or cannot be detected by a PCR-based approach, the results suggest that mutations in this gene do not play a major role in the genesis of sporadic forms of these smooth muscle tumours. Although in the leiomyosarcomas LOH was common, this reflected extensive loss of chromosome 1q rather than being limited to the vicinity of FH.
The tumour suppressor gene/recessive oncogene hypothesis articulated by Knudson (1993)
predicts that genes that confer an increased risk of neoplasms as a result of germline mutations are likely to be somatically mutated in sporadic cancers of the same type. This has proved to be the paradigm of the RB1
gene and for several other genes, such as APC
does not appear to conform to this model and is similar to a number of other cancer susceptibility genes in this respect. For example, truncating mutations in BRCA1
confer a high risk of breast cancer, but rarely have somatic mutations in sporadic tumours been reported in either gene (Futreal et al, 1994
; Lancaster et al, 1996
; Teng et al, 1996
). Similarly germline mutations in PTEN
confer an elevated risk of breast cancer in individuals with Cowden's disease, but somatic mutations in PTEN
are rarely observed in breast cancers outside this context (Rhei et al, 1997
; Sakurada et al, 1997
There are several possible reasons for the low frequency of FH mutations in sporadic leiomyomas and leiomyosarcomas detected in our study. It may be that sporadic lesions develop through a route distinct from familial tumours. Alternatively, FH expression is altered in ways other than somatic mutation, such as regulation of mRNA levels. To date there are no studies to our knowledge that suggest this is a possibility either in sporadic leiomyomas or leiomyosarcomas. We are, however, in the process of evaluating this possibility.
The mechanism by which FH can act as a tumour suppressor is unclear. It is possible that hypermutability results from oxidative damage. Alternatively, accumulation of fumarate leads to feedback effects on oxidative metabolism and thus on the cell cycle. Smooth muscle cells harbouring ‘two hits’ in FH
would exhibit inhibition of normal progression of the cell cycle, causing hyperproliferation or a failure of apoptosis. Feedback from fumarate would have tissue-specific effects, hence the growth of smooth muscle tumours (Tomlinson et al, 2002
). Although mutations in FH
do not appear to be involved in the development of sporadic smooth muscle tumours it is conceivable that other mitochondrial defects may be involved.