By Katie Nightingale
Mutations in the tumour suppressor gene Folliculin (FLCN) cause Birt-Hogg-Dubé syndrome (BHD). Although the loss of a single allele of FLCN is sufficient to induce fibrofolliculomas and lung cysts, it is thought that a secondary mutation in the wild-type allele of FLCN is required for the development of renal cancer. Despite the wide variety of cellular processes FLCN has been connected to, including cell growth and proliferation, stress resistance, and autophagy, the mechanism by which loss of FLCN can induce tumourigenesis remains largely unknown. A recent study by Glykofridis et al., investigated the effect of loss of FLCN in renal tubular epithelial cells (RPTEC), a widely accepted in vitro model of human kidney function. They demonstrated that loss of FLCN led to the upregulation of several interferon (IFN) responsive genes in addition to activation of the transcription factor TFE3, resulting in the upregulation of genes involved in lysosome function and autophagy.
TFE3 directs the expression of genes involved in autophagy and stress tolerance under growth restrictive conditions and has been associated with the development of renal cell carcinoma in children. To investigate the localisation of TFE3 in the context of FLCN loss, they used CRISPR-Cas9 to generate a FLCN knock-out (FLCN-KO) RPTEC cell line. Comparing this to control RPTEC cells, in which TFE3 remains in the cytoplasm under normal growth conditions, they showed that TFE3 localised to the nucleus in FLCN-KO cells, indicating constitutive activation of TFE3.
To examine the effect of FLCN loss in a wider cellular context, they subsequently performed transcriptomic and proteomic analyses comparing FLCN-KO RPTEC cells to wild-type. They observed the induction of a large number of genes upon FLCN loss, which led them to identify a subset of 115 known TFE targets upregulated in the FLCN-KO cells. As many of these genes function in autophagy and lysosome regulation they next assessed mTOR signalling by looking at the phosphorylation status of the mTORC1 targets 43-BP1 and AKT and, contrary to previous reports, found no evidence for altered mTOR signalling or nutrient sensing in the FLCN-KO RPTEC cells. Analysis of the promoter regions of upregulated genes led to the discovery that the majority of upregulated genes could be assigned to 2 different promoter motifs: the E-box motif (regulated by TFE) and the interferon-stimulated response element (ISRE) motif. The upregulation of many of these genes was validated by quantitative PCR. Moreover, they confirmed the elevated expression of these genes in two independent BHD kidney tumours by mass spectrometry.
FLCN forms a complex with its interacting partners FNIP1 and FNIP2. To investigate the role of these proteins in the context of gene expression, they generated an FNIP1/FNIP2-KO RPTEC cell line and confirmed that upregulation of E-box or ISRE genes is dependent on the FLCN-FNIP1/FNIP2 axis. Re-introduction of FLCN to the FLCN-KO cell lines completely reverted the gene expression observed and established that upregulation of ISRE genes was entirely dependent on STAT1 and STAT2 through siRNA knockdown of these proteins. They could not, however, find a role for FLCN in the canonical IFN signalling pathway and thus concluded FLCN loss in RPTEC leads to non-canonical activation of STAT1/2. An unusual phenotype of FLCN-KO RPTEC is the severely reduced proliferation of these cells. This is perhaps unexpected as the loss of a tumour suppressor gene such as FLCN should result in increased proliferation. Indeed, constitutive activation of TFE3 in an FLCN-independent manner promoted uncontrolled proliferation. They hypothesised that the STAT2-mediated IFN signature observed in FLCN-KO RPTEC results in growth inhibition, which could counteract the hyperproliferative effects of TFE3 activation.
Relating these data to clinical outcomes for BHD patients the authors commented that this database of target genes provides a foundation for further investigation into the discovery of biomarkers for early-stage tumourigenesis as well as therapeutic strategies to prevent renal cell carcinoma in BHD patients. Of particular interest, the authors noted that GPNMB, a TFE3 target gene and protein frequently upregulated in a wide variety of tumours, was strongly upregulated in the absence of FLCN. Although it is unclear whether overexpression of GPNMB alone is sufficient to drive tumourigenesis, GPNMB can be targeted therapeutically using antibody-drug conjugates (such as glembatumumab vedotin) which are currently in clinical trials for cancer therapy warranting the investigation of this in the treatment of BHD tumours. In summary, Glykofridis et al., performed a comprehensive analysis of the transcriptome and proteome of cells lacking FLCN however, to truly carry this work forwards into a clinically relevant setting, analysis of further tumours from BHD patients would have to be carried out to confirm their findings.
Glykofridis IE, Knol JC, Balk JA, Westland D, Pham TV, Piersma SR, Lougheed SM, Derakhshan S, Veen P, Rooimans MA, van Mil SE, Böttger F, Poddighe PJ, van de Beek I, Drost J, Zwartkruis FJ, de Menezes RX, Meijers-Heijboer HE, Houweling AC, Jimenez CR, Wolthuis RM. Loss of FLCN-FNIP1/2 induces a non-canonical interferon response in human renal tubular epithelial cells. Elife. 2021 Jan 18;10:e61630.