Von Hippel-Lindau (VHL) disease is a hereditary syndrome characterized by the development of multiple tumors, both benign and malignant, affecting several different organ systems , including the eyes, spine, inner ear, pancreas, adrenal gland, and kidneys. Retinal angiomas, cerebellar and spinal hemangio-blastomas, and renal cell carcinomas are the hallmark lesions of this disease. Renal cysts, pancreatic cysts, pancreatic carcinomas, pheochromocytomas, epididymal or broad ligament cyst adenomas, and endolymphatic sac tumors may also occur in patients who suffer from this disease. Von Hippel-Lindau disease is estimated to affect approximately 1 in 36,000 individuals and is inherited in an autosomal-dominant fashion, with estimated penetrance of 80% to 90% by the age of 65 [12,13]. Renal cell carcinoma eventually develops in approximately 28% to 45% of those individuals affected with VHL disease . The tumors associated with
VHL disease are typically multicentric and are often bilateral. These tumors are predominantly of the clear cell variety.
The VHL gene has been localized to the short arm of chromosome 3, sub-band 25 (3p25) [14-16]. This mapping was first accomplished by Seizinger et al. , who used genetic linkage analysis to study nine families affected with VHL disease. The germline mutation is transmitted in an autosomal-dominant fashion, with each of the offspring having a 50% risk of inheriting the mutant allele . According to Knudson's  "two-hit" hypothesis, the carriers of mutations in a tumor-suppressor gene have a germline mutation in one allele of the gene, and a second "hit" or somatic mutation occurs in the homologous normal allele, thus leading to tumor formation. The purpose of tumor-suppressor genes, such as the VHL gene, is to inhibit tumor development through regulation of self-proliferation and differentiation, and their inactivation predisposes an individual to cancer through loss of these regulatory processes.
Tory et al.  used restriction fragment length polymorphism (RFLP) analysis to confirm that in the VHL patients the wild-type chromosome 3p25 allele inherited from the unaffected parent had been lost. With the other allele being the abnormal germline copy of the VHL gene inherited from the affected parent, Tory et al. were able to demonstrate this association with the development of renal cell carcinoma. Lubensky et al.  subsequently demonstrated loss of the wild-type 3p allele with evidence of the inherited mutated allele in 25 of 26 renal lesions from patients affected with VHL disease. Also, the loss of heterozygosity (LOH) at 3p25 was detectable in atypical renal cysts and cysts with renal cell carcinoma in situ. These provided strong evidence that the VHL gene was indeed a tumor-suppressor gene, and loss of function of both gene copies appeared to be an important early step toward tumor formation.
In 1988 Seizinger et al.  confirmed that the VHL gene was linked to the locus encoding the human homologue of the RAF1 oncogene mapping to the 3p25 chromosome. In this important report, the authors hypothesized that the defect responsible for the VHL phenotype was not a mutation in the RFA1 gene itself but rather the inactivation of a putative tumor-suppressor gene, namely the VHL gene, and that this led to the development of renal cell carci
Molecular Biology of Kidney Cancer noma. This linkage to RFA1 was confirmed by Hosoe et al. , who also reported linkage of the VHL gene to D3S18, a polymorphic DNA marker located at 3p26. Richards et al.  then demonstrated tight linkage of the VHL gene to the DNA probe D3S601, which was located in the region between RAF1 and D3S18. The VHL gene was subsequently identified by Latif et al.  in 1993 through the use of yeast artificial chromosomes and cosmid phase contigs. Latif et al. demonstrated that the VHL gene was a single-copy gene with evolutionary conservation across several species, pointing to its essential role in cellular processes.
The VHL gene contains three exons with an open reading frame of 852 nucleotides that encode a protein of 213 amino acids [2,3]. Several hundred germline mutations have now been recognized in VHL kindreds [2,3]. These include microdeletions, insertions, large deletions, and missense and nonsense mutations. Chen et al.  studied 114 VHL families and identified mutations throughout the coding region, but clustering occurred at the 3' end of exon 1 and the 5' end of exon 3 with a paucity of mutations in exon 2. The importance of this is that specific mutations have now been correlated with certain phenotypic characteristics in VHL patients. For example, VHL type 1 families most frequently have large deletions, microdele-tions/insertions, or nonsense mutations [12,22,23]. The specific phenotypic characteristic of these families is that they typically do not develop pheochromocytomas. However, in VHL type 2 families that do suffer from pheochromo-cytomas, 96% of the mutations are missense mutations [12,22,23]. Gnarra et al.  evaluated sporadic clear cell carcinomas and found VHL gene mutations in 57% of the cancers, with LOH of the gene in 98%. In these patients, they found that the mutations clustered to the 3' end of exon 1 and at the 5' end of exon 3; however, exon 2 also had a high frequency of mutations (approximately 45%). This high number of mutations, as well as splice-site mutations that would eliminate its translation, suggested that exon 2 may have a role in the function of the protein product. The functions of the VHL protein product have been difficult to predict as there is no important homology to other proteins [12,16]. Further characterization has been performed through cellular localization studies. Immunofluorescence microscopy demonstrated that the protein product is located primarily in the cytoplasm but can also been found in the nucleus [3,12,25-28]. Subsequently, two biologically active VHL protein isoforms, pPVHL30 and pVHL19, have been demonstrated . Furthermore, there is evidence to suggest that the expression of the VHL proteins in either the cytoplasm or the nucleus may be associated with clinical outcome . Coimmunoprecipitation of the VHL protein revealed two proteins of 9 and 16 kilodaltons (kd), which were subsequently identified as elongin C and elongin B, respectively [3,30-32]. However, when certain missense mutations of the VHL gene occurred, this protein interreaction was found to be very weak or nonexistent [3,28]. This relationship of the normal VHL protein and loss of its association with other proteins due to certain mutations has led several investigators to study the protein-protein interactions in VHL much more closely.
Normally the VHL protein product binds tightly to elongin B and C, which are regulatory subunits of elongin SIII [3,30,33-35]. The VHL protein product has not been shown to bind to elongin A. Elongin SIII is known to hasten DNA transcriptional elongation by RNA polymerase II by inhibiting temporary pausing of the poly-merase at certain DNA sites and by controlling its release from DNA [3,12,36,37]. With binding of elongin B and C, the VHL protein is able to abort the formation of the active heterotrimeric protein elongin SIII [3,30]. The transcription of certain genes may be downregulated as a result of these binding sequences [12,34]. As mentioned previously, a number of VHL proteins with missense gene mutations have been found to complex minimally or not at all with the elongin regulatory subunits [3,28]. This inability to inhibit the formation of elongin may lead to the loss of regulation of transcription rates of genes important in tumor suppression [12,28,35].
The association of the VHL protein with elongin B and C may function to promote a complex that is responsible for ubiquitination and proteasome degradation. Cullin-2 (CUL-2), a member of the CDC53 family of proteins, has been found to bind to VHL elongin B and C and form a stable tetrameric complex [3,38]. This complex has been noted to be crucial in the degradation of specific proteins such as hypoxia-induced factor 1a (HIF1a) [33,39]. Studies have
Urological Cancers: Science and Treatment shown that VHL gene mutations have resulted in higher concentrations of HIFla protein expression [14,39]. Accumulation of HIFla has been noted to result in higher levels of vascular endothelial growth factor (VEGF) production . VEGF is a polypeptide growth factor that assists in the migration, proliferation, and differentiation of vascular endothelial cells . It has been shown to be markedly overexpressed in sporadic and VHL-associated renal cell carcinomas [2,41-44]. The typical hypervascular appearance of VHL and renal cell carcinomas, and the significant amount of neoangiogenesis often seen with these tumors, has been related to the overproduction of VEGF [2,40].
The VHL gene may also have a more direct interaction in repression of VEGF through its association with the ubiquitous transcriptional activator Spl . Further investigations have revealed that the Spl transcriptional activator was found to interact with a specific isoform of protein kinase C in renal cell carcinoma causing transcriptional promotion of VEGF . In the presence of the wild-type VHL protein product, the SP1 and protein kinase C interaction is inhibited, resulting in lower levels of VEGF. Furthermore, there may be direct complexation of protein kinase C by the wild-type VHL protein, leading to inhibited VEGF expression. Also the VEGF receptors KDR and Flt-1 are overex-pressed in sporadic and VHL-associated renal cell carcinoma [2,47]. All of these findings suggest an important role for the VHL protein as a tumor suppressor.
Another polypeptide growth factor, transforming growth factor beta-1 (TGF-P1) appears to be regulated by the VHL protein [2,48,49]. This factor seems to function in a proliferative fashion through a paracrine mechanism to promote renal cell carcinoma formation. Tumor development is suppressed with reintroduction of the wild-type VHL protein product, which inhibits TGF-P1 production or through the administration of the anti-TGF-b1 antibodies [14,48]. The VHL protein product has also been associated with fibronectin, the extracellular glycoprotein involved in extracellular matrix cell signaling through intergrins [33,50-52]. Extracellular fibronectin matrix assembly is altered in VHL negative cells, and this alteration is corrected with the introduction of the wild-type VHL protein [14,50]. In mutated cells, neovas-cularity parallels the changes in the extracellu-
lar matrix [14,53]. Tumor suppression, therefore, may be a result of appropriate fibronectin matrix assembly with regulation of neoangiogenesis. The effects of the loss of the VHL gene on its target proteins is summarized in (Figure 16.2).
Knowledge of the VHL gene and its protein as well as its targets has led to an ever-expanding list of genes and proteins involved in the development of renal cell carcinoma. The VHL gene is now known as one of the classic tumor-suppressor genes in medicine. Multiple mutations of the gene leading to its inactivation have been characterized and associated with specific clinical phenotypic results. Furthermore, a strong effort to enroll families afflicted with von Hippel-Lindau syndrome has been made by the National Cancer Institute in order to track these families and their outcomes. With further research into this disease, it is hoped that therapeutic targets might be established based on the understanding of the gene's multiple functions.
Was this article helpful?