Functional analysis of the von Hippel-Lindau tumour suppressor and its role in tumourigenesis

Complex genotype-phenotype relations are a hallmark of VHL disease. Patients develop
a wide range of tumours depending on how and where pVHL malfunctions. Thus it
appears that the VHL tumour suppressor gene product must have multiple and tissue
specific functions.
pVHL interacts with the proteins Elongin C, Elongin B, and Cullin 2 in a complex
referred to as the VCB-Cul2 complex. This complex displays structural analogy to the
Skp1-Cdc53/Cul1-F-box protein (SCF) complex. As with its SCF counterparts, the VCBCul2
complex has been shown to constitute an E3 ligase, which serves to recruit protein
substrates for degradation by the 26S proteasome. To date, only one such target has
been firmly established, the transcription factor Hif-α (hypoxia inducible factor).
Proteolytic degradation of Hif-α reflects a key cellular mechanism in the control of
adaptive gene expression in response to changes in oxygen levels. While identification of
Hif-α as a substrate for the VCB-Cul2 complex proves to be a crucial milestone in VHL
biology, isolation of other proteins that are targeted for ubiquitination by VHL represent a
major challenge.
A proteomics approach was developed in an effort to expose unknown pVHL
protein interactions, and thereby highlighting novel functions of the VHL tumour
suppressor. One candidate protein is a 97-kDa ATPase called p97. p97 is a member of
the AAA family of ATPase’s and is involved in a myriad of distinct cellular functions,
interestingly among which include binding to poly-ubiquitin chains and facilitating
substrate presentation to the proteasome. We show that p97 binds pVHL both in vitro
and in vivo, and propose a model whereby p97 may facilitate presentation of
ubiquitinated Hif-α to the 26S proteasome for subsequent degradation. We also
demonstrate that, similar to pVHL, p97 can promote microtubule stability, and propose a
model through binding to HDAC6, a histone deacetylase known to bind p97 and shown
here to bind pVHL in vitro, that these proteins might be involved in the malfolded protein
response, and that this potential function could be microtubule dependent.
The proteomics approach also uncovered additional protein interactions, namely
two uncharacterised proteins with homologues in S.Cerevisiae known to complex and
constitute a critical RNA methyltransferase. We suggest that this co-precipitating
complex is the mammalian orthologue and we provide preliminary data showing, in
addition to the endogenous interaction, in vitro binding to pVHL. Finally we propose a
model where, by means of its ability to negatively regulate the activity of this complex, pVHL could be involved in the mammalian stress-responsive pathway, thereby helping to
explain, in part, the observation that the inability of renal cell carcinoma cells lacking VHL
to exit the cell cycle upon serum withdrawal can be restored upon reintroduction of pVHL
into these cells.
In a second part to this thesis, pVHL intracellular dynamics have been studied. We
demonstrate that pVHl19 and pVHL30 exhibit different localisation patterns, with pVHL30
residing primarily in the cytoplasm, and pVHl19 in the nucleus. We show that when
pVHL30 is in the cytoplasm, it co-localises with microtubules, and that this localisation is
altered upon microtubule destabilisation, which renders a strong nuclear signal for
pVHL30. We conclude that pVHL intracellular dynamics are reflective, in part, by the
stability of the microtubule network. This work contributed to an article in Nature Cell
Biology.
The third part of the results outlines the targeting of endogenous pVHL by RNA
interference. Optimal conditions for VHL RNAi and a study on the cellular affects are
presented. Hif-α regulation is investigated in the presence of VHL siRNA oligos, and the
lack of Hif-α up-regulation discussed. Finally, positive regulation of phospho-cofilin, an
important component in actin ctyoskeleton rearrangements, is demonstrated as the only
positive read-out for VHL RNAi to date, and the implications this regulation by pVHL
might have an cell shape changes and movement is briefly discussed.