Junctions on the road to cancer


There are two main types of tumor suppressor genes (TSGs): ‘gatekeepers’ that encode proteins that provide direct protection from tumorigenesis, and ‘caretakers’ that encode proteins necessary for maintaining genome stability and consequently guard against somatic mutations arising in oncogenes and other TSGs. A subset of the caretaker TSGs encodes DNA repair proteins, including the Bloom syndrome DNA helicase (BLM), the mismatch repair proteins MSH2 and MLH1, and the breast cancer susceptibility protein BRCA2. The recent report by Hakem and colleagues1 in Science that mus81 mutant mice are profoundly predisposed to cancer (especially non-Hodgkin’s lymphoma), together with the observation that mouse cells deficient in Mus81 show genome instability, places another known DNA repair protein on the list of caretaker tumor suppressors. Although it is not yet known whether loss of Mus81 predisposes humans to cancer, this study clearly implicates human MUS81, on chromosome 11q13.1, as a strong candidate TSG. Interestingly, when only one copy of the MUS81 gene is disrupted, cancer predisposition is as severe as it is when both copies are disrupted1. This is in contradiction with Knudon’s two-hit model, which stipulates that TSGs are recessive and that both alleles must be inactivated for tumorigenesis. However, haploinsufficient TSGs are not without precedence2, and in the case of Mus81 the data suggest that expression from both alleles may be necessary to generate enough protein to perform all of its tasks. Exactly what these tasks are remains the subject of research in a number of laboratories, although a combination of biochemical and genetic studies have already provided significant clues as to what Mus81 does in vivo and how a failure to execute its functions could result in genome instability and ultimately tumorigenesis. Potential targets of Mus81 Mus81 is a member of the XPF family of DNA structure-specific endonucleases, which share a highly conserved motif (V/IERKX3D) that is an integral part of the endonuclease catalytic site3–5. It functions as a heterodimer with a protein called Eme1 (also known as Mms4 in a different organism) for simplicity I will refer to this as the Mus81 complex. Genetic studies in yeast indicate that the complex has a role in processing recombination intermediates during the tolerance and repair of DNA damage3–8. The types of DNA damage repaired by homologous recombination (HR) include double-strand breaks (DSBs), broken replication forks and lesion-containing single-strand gaps. HR may also mediate template switching for DNA polymerases to bypass DNA lesions, and can help restart stalled replication forks at non-replication origin sites9,10. There are a plethora of models that describe how HR may promote these processes, and they variously involve the formation and processing of 3′ flaps, forks, D-loops and nicked, gapped and intact Holliday junctions (HJs). Biochemical studies have therefore focused on testing the ability of the Mus81 complex to cleave mimics of these junctions (Fig. 1)7,11–17. These studies have shown that its preferred substrates are fourway DNA junctions that contain an exposed 5′ DNA strand end at or close to the junction crossover point (for example, nicked HJs and D-loops). These are cleaved ~3 times more efficiently than forks and 3′ flaps and ~70 times more efficiently than intact HJs. In each case, cleavage typically occurs 3–6 bp 5′ of the junction crossover point in a way that would be productive for processing the junction in vivo. Comparisons between human, budding yeast and fission yeast Mus81 complexes have confirmed that the junction specificity is conserved among different organisms. With the kinds of DNA junctions that the Mus81 complex can cleave in vitro thus established, one challenge has been to determine which of them are targeted in vivo. An important observation here has been the ability to suppress mus81 mutant phenotypes in yeast and human cells by the expression of the bacterial HJ resolvase RusA6–8,11,18. RusA is highly specific for cleaving HJs, so its ability to suppress these mutant phenotypes has been used as evidence that the Mus81 complex primarily cleaves HJs in vivo. This idea has proved somewhat controversial and seems to be in contradiction with the complex’s substrate specificity in vitro. However, proponents of this idea have argued that the Mus81 complex could be made an efficient cleaver of HJs by its post-translational modification or interaction with other proteins11,13. Although this is possible, a more parsimonious, and widely accepted, explanation for these data is that the Mus81 complex

DOI: 10.1038/nsmb0804-693

Cite this paper

@article{Whitby2004JunctionsOT, title={Junctions on the road to cancer}, author={Matthew C Whitby}, journal={Nature Structural &Molecular Biology}, year={2004}, volume={11}, pages={693-695} }