Nf1 tumor suppressor in skin:: Expression in response to tissue trauma and in cellular differentiation
Prev	Chapter 2. Review of the literature	Next

2.2. NF1 gene

The NF1 gene is mapped to locate in chromosome 17q11.2 and it spans over 3.5 kb of genomic DNA (Cawthon et al. 1990, Viskochil et al. 1990, Wallace et al. 1990). The NF1 gene is ubiquitously expressed in human, resulting in a 11-13 kb NF1 mRNA and many alternatively spliced variants (Skuse & Cappione 1997). The NF1 gene (illustrated below in Fig. 1) contains 60 exons that are organized into four clusters which are separated by four large introns (Marchuk et al. 1991, Li et al. 1995, Viskochil 1999).

Sequence homology to GAP is located within the exons 21-27a, suggesting that NF1 protein accelerates the GTPase activity of Ras proteins (Viskochill 1999b). The NF1 coding region contains an even longer homology to yeast IRA1 and IRA2 proteins, which are regulators of the Ras-cAMP pathway (Ballester et al. 1990, Buchberg et al. 1990, Martin et al. 1990, Xu et al. 1990a, Xu et al. 1990b). Another region, comprising exons 11-17, contain homology to ATP binding- and cAMP-dependent protein kinase (PKA) recognition sequences (Fahrsold et al. 2000). Sequences coding for tubulin binding sites locate within and in the vicinity of the GAP-related domain (GRD) (Bollag et al. 1993, Scheffzk et al. 1998). Some studies have also suggested sequences that contain homology to nuclear targeting signals (Marchuck et al. 1991, Suzuki et al. 1991, Li et al. 2000).

Figure 1. Schematic representation of NF1 gene structure. The transcription start site is marked by an arrow. The transcription stop site and polyadenylation site are marked by an octagon. The GRD is marked, spanning exons 21-27a. The alternatively spliced exons are shown in gray. The three embedded genes within the intron 27b are transcribed to opposite direction (arrow). The asterisk represents the site of mRNA processing. Bar, 1000 bp (introns are not in scale). The figure has been modified from an article by Viskochil (1999b).

The mutation rate of the NF1 gene is one of the highest known to occur in the human genome, 3.1-6.5x10⁵ (Vogel & Motulsky 1997), and about 50% of all NF1 patients lack a family history of the disease (Huson & Hughes 1994). In general, mutational analysis of the NF1 gene is complex due to the large size of the gene, the existence of pseudogenes and the great diversity of the lesions (Legius et al. 1992, Kehrer-Sawatzki et al. 1997, Regnier et al. 1997, Lujiten et al. 2000). Mutations of the NF1 gene are distributed along the whole genomic region, and only a few possible mutational hot spots have been suggested so far (Shen et al. 1996, Boulandet et al. 2000, Fahsold et al. 2000, Messiaen et al. 2000). A diversity of the mutations affecting the transcript or untranslated regions (UTR) has been described, including e.g. chromosomal abnormalities, deletions, insertions and spot mutations (Upadhyaya et al. 1997, Upadhyaya & Cooper 1998, Ars et al. 2000, NNFF International NF1 genetic Mutation Analysis Consortium, Human Gene Mutation Database Cardiff).

In addition to germ line mutations, somatic mutations or LOH have also been shown in many NF1-associated malignancies as well as in cancers of non-NF1 patients (Xu et al. 1992a, Andersen et al. 1993, Johnson et al. 1993, Legius et al. 1993, Shannon et al. 1994). LOH has also been described in some of the benign neurofibromas (Colman et al. 1995, Sawada et al. 1996, Däschner et al. 1997, Eisenberth et al. 2000). Furthermore, a few cases of somatic mosaicism have been described at the molecular level (Lazaro et al. 1995, Colman et al. 1996, Wu et al. 1997).