Distinctive expression patterns of FLCN and GPNMB in BHD renal tumours

As discussed on this blog previously, developing histological screening techniques for renal cell carcinomas (RCCs) associated with BHD is important for early diagnosis. Individuals with folliculin (FLCN) mutations are more likely to develop multiple bilateral renal tumours (Zbar et al., 2002, Pavlovich et al., 2002). A misdiagnosis of sporadic RCC may compromise future treatment and wellbeing of the patient and other affected family members. Currently there are no known histological markers to distinguish between all subtypes of sporadic and FLCN-associated tumours.

A new report by Furuya et al., (2015) addressed this by analysed expression of FLCN and one of its downstream targets GlycoProtein Non-Metatastic B (GPNMB) in normal and neoplastic tissue to determine if they could be used to aid differential diagnosis in RCC samples. FLCN is expressed in normal kidney tissue, including in those who carry heterozygous FLCN mutations, but is not detectable in BHD tumours (Warren et al., 2004). In comparison GPNMB is not typically expressed in kidney tissue but is expressed in a range of neoplastic tissues (Hong et al. 2010).

Furuya et al. analysed 27 tumours from 18 unrelated, except for a parent and child pair, Japanese BHD patients: 12 chromophobe RCCs; six hybrid oncocytoma/chromophobe tumours (HOCTs); three papillary RCCs; and two clear cell RCCs (ccRCC). For seven of these tumours sections of non-neoplastic kidney were also frozen for comparison. Expression of FLCN and GPNMB in the BHD-associated tumours was compared to 62 sporadic renal tumours from an unreported number of patients. All patients were medically examined for the classical BHD symptoms – pulmonary cysts, pneumothorax, fibrofolliculomas and multifocal/hybrid RCC – and a family history obtained. One of the sporadic group also had pulmonary cysts so a diagnosis of BHD was ruled out by genetic testing. The rest of this group did not show any BHD manifestations or have any family history so were assumed to not carry FLCN mutations.

FLCN expression in the tumour samples was assessed by western blot and immunohistochemistry using a monoclonal (D14G9) and previously unpublished polyclonal (ab93196) antibody respectively. The western blot results showed a clear loss of FLCN in BHD-associated but not sporadic tumours or non-neoplastic BHD tissue. In addition the typical strong nuclear staining for FLCN was seen in the majority of sporadic tumours whereas the majority of BHD-associated tumours showed only cytoplasmic or no staining. It is suggested that the presence of cytoplasmic staining should be considered a potential BHD indicator.

The presence of GPNMB protein was confirmed by western blot and immunohistochemistry, but mRNA levels were also quantified by quantitative RT-PCR. The markedly higher expression (up to 23-fold) detected in BHD tumours by qRT-PCR correlated with the intensity of the western band seen. GPNMB was barely detectable in the sporadic tumour samples and non-neoplastic BHD renal tissue. Immunostaining showed positive staining in BHD-associated tumours excluding the ccRCC samples and one papillary sample. The majority of sporadic samples were in contrast negative with the exception of 50% of sporadic chromophobe RCC samples which showed weak GPNMB staining.

It was also possible to detect low expression of GPNMB in small nodules of “non-neoplastic” BHD kidney tissue which could indicate the sites of future tumour development. As the formation of multiple small tumours is characteristic of BHD this could also help in differential diagnosis.

The germline FLCN mutations for each patient was determined from a peripheral blood sample; the most common (50%) mutation was duplication in the hypermutable cytosine tract in exon 11 and four patients had the same GATG deletion in exon 13. In addition Furuya et al., attempted to detect secondary somatic mutations in a number of the tumour samples. A second mutation was found in six individual tumours and a loss of heterozygosity in a further two. Only one of these secondary mutations has so far been reported as a germline mutation in a BHD patient. The other mutations may therefore not be pathogenic but as all would result in a frameshift it is more likely they are new unique mutations.

The results from this work suggest that staining for both a lack of FLCN and a gain of GPNMB could help to distinguish sporadic oncocytomas, chromophobe and papillary RCC from BHD-associated HOCTS, chromophobe and papillary RCCs. It would not be sufficient however to definitively classify a ccRCC as FLCN-related, but a lack of nuclear FLCN staining could identify cases appropriate for genetic testing.

  • Furuya M, Hong SB, Tanaka R, Kuroda N, Nagashima Y, Nagahama K, Suyama T, Yao M, & Nakatani Y (2015). Distinctive expression patterns of glycoprotein non-metastatic B and folliculin in renal tumors in patients with Birt-Hogg-Dubé syndrome. Cancer science PMID: 25594584
  • Hong SB, Oh H, Valera VA, Baba M, Schmidt LS, Linehan WM. Inactivation of the FLCN tumor suppressor gene induces TFE3 transcriptional activity by increasing its nuclear localization. PLoS One. 2010 Dec 29;5(12):e15793. doi:10.1371/journal.pone.0015793. PMID: 21209915.
  • Pavlovich CP, Walther MM, Eyler RA, Hewitt SM, Zbar B, Linehan WM, Merino MJ. Renal tumors in the Birt-Hogg-Dubé syndrome. Am J Surg Pathol. 2002 Dec;26(12):1542-52. PubMed PMID: 12459621.
  • Warren MB, Torres-Cabala CA, Turner ML, Merino MJ, Matrosova VY, Nickerson ML, Ma W, Linehan WM, Zbar B, Schmidt LS. Expression of Birt-Hogg-Dubé gene mRNA in normal and neoplastic human tissues. Mod Pathol. 2004 Aug;17(8):998-1011. PMID: 15143337.
  • Zbar B, Alvord WG, Glenn G, Turner M, Pavlovich CP, Schmidt L, Walther M, Choyke P, Weirich G, Hewitt SM, Duray P, Gabril F, Greenberg C, Merino MJ, Toro J, Linehan WM. Risk of renal and colonic neoplasms and spontaneous pneumothorax in the Birt-Hogg-Dubé syndrome. Cancer Epidemiol Biomarkers Prev. 2002 Apr;11(4):393-400. PubMed PMID: 11927500.


Leave a Reply