Tuberous sclerosis complex (TSC), an autosomal dominant disease caused by mutations in either TSC1 or TSC2, is characterized by the development of hamartomas in a variety of organs. Concordant with the tumour-suppressor model, loss of heterozygosity (LOH) is known to occur in these hamartomas at both TSC1 and TSC2 loci. LOH has been documented in renal angiomyolipomas, but loss of the wild-type allele in cortical tubers appears very uncommon. We analysed 24 hamartomas from 10 patients for second-hit mutations by several methods, and found no evidence for the inactivation of the second allele in many of the central nervous system (CNS) lesions, including tumours that appear to be clonally derived. We believe that somatic mutations in TSC1 and TSC2 resulting in the loss of wild-type alleles may not be necessary in some tumour types, and other mechanisms may contribute to tumorigenesis in this setting. We have shown that hamartin interacts with neurofilament light chain (NF-L) and could integrate the neuronal cytoskeleton through its direct interaction with NF-L and ERM (ezrin/radixin/moeisin) proteins. Our unpublished work further documents the binding of tuberin with Pam, a protein associated with c-Myc, which is enriched in brain. All these observations suggest that the tuberin-hamartin complex is likely to have distinct functions in the CNS.