Basic research

Although recent research, such as the elucidation of the C-terminal structure of FLCN indicating that it is a DENN domain protein with GEF activity, has provided important insights into FLCN’s function, more research is required to comprehensively catalogue the normal cellular functions of FLCN. Understanding its normal function is crucial to understanding how FLCN mutation or loss can lead to the epidermal, pulmonary and renal phenotypes associated with BHD syndrome.

The majority (95%) of FLCN mutations that are associated with BHD truncate the protein by introducing a frameshift (45%), deleting a splice site (19%), introducing a nonsense mutation (18%), deleting or duplicating a large portion of the gene (10%) or by removing the start codon (3%) (Sections 3).  However, it is unknown whether the resultant aberrant protein is degraded, remains in the cell with altered function or exerts a dominant negative effect. Indeed, it has recently been shown that FLCN is expressed in a BHD renal tumour with a different truncating FLCN mutation in each copy of the gene, suggesting that at least one of these mutations does not lead to the complete inactivation of the protein (Menko et al., 2013). The significance of this finding is currently unknown.

There are six missense mutations that are currently reported to be associated with BHD syndrome (Section 3). Although it is likely that these amino acid substitutions are pathogenic, as they are associated with disease, it is unknown how these mutations affect FLCN function in order to cause the disease. Determining how these missense mutations affect the FLCN protein may shed light on which residues or domains of the protein are particularly important for its correct function and structure.

The identification that the C-terminus binds both FNIP1 and FNIP2 (Baba et al., 2006; Hasumi et al., 2008; Takagi et al., 2008), and has Rab GEF activity (Nookala et al., 2012), demonstrates that this region is important for FLCN function, meaning that truncating mutations that delete the C-terminus are likely to be detrimental. Although initial studies suggest that FLCN acts as a GEF for Rab35 in vitro, it remains to be seen whether FLCN has GEF activity towards other GTPases, although as FLCN does bind the GTPase domain of RagA, making it likely that it acts as a GEF for this and other proteins in vivo (Petit et al., 2013). The first 85 amino acids of the protein contain a metal-ion binding motif (Nookala et al., 2012) although FLCN has thus far not been observed to bind any metals. Thus, the function of the N-terminal region is currently unknown.

Localisation experiments have observed FLCN in both the cell nucleus and the cytoplasm (Takagi et al., 2008). Further experiments by Nahorski et al. (2012) show that FLCN co-localises with binding partner PKP4 at cell junctions during interphase, and at the mid-body during cytokinesis, while Gaur et al. (2013) show that FLCN co-localises in the nucleolus with Rpt4. How FLCN is recruited to the correct location has not yet been elucidated.

Two of the proteins known to bind FLCN – FNIP1 and FNIP2 – are themselves poorly characterised.  PKP4, Rpt4 and the Rag proteins have recently been shown to bind FLCN and it is likely that many additional proteins bind to FLCN. The function of the FLCN-FNIP1, -FNIP2, -PKP4 and -Rpt4 complexes are also an area of future work. Characterising FLCN’s interactions with other proteins, and the functions of FLCN-containing protein complexes, will provide insight into how FLCN mutation or loss affects cellular function. These interacting proteins may modulate the phenotype of BHD syndrome and thus affect its variability.

FLCN has been implicated in numerous signalling pathways and cellular processes, including:

However, FLCN’s role in these processes has not been fully elucidatied. Thus, fully characterising the role of FLCN in these and additional processes, will provide insight into how FLCN happloinsufficiency or loss causes BHD syndrome.