Inactivation of the Mitotic Checkpoint Protein UBE2Q2 to Identify Its Function In Oncogenesis and Development

Abstract 5115

UBE2Q2 is a ubiquitin conjugating enzyme that was first identified in a microarray screen for mitotic regulatory proteins. Since UBE2Q2 is expressed in G2- and M-phases, we evaluated it as a possible regulator of mitotic entry or progression. While inactivation of the protein using either siRNA or dominant-negative protein expression caused no overt change in cell cycle progression in unperturbed cultured human cells, it induced a prolonged early mitotic arrest after cells were treated with vincristine or other microtubule inhibitors (MIs). In addition, inhibition of UBE2Q2 caused a pronounced and specific sensitization to the cytotoxicity of MIs, an effect mediated by apoptosis (Banerjee et al, Oncogene 26;6509-17). The cell cycle effect caused by UBE2Q2 inactivation was indistinguishable from that produced by over-expression of the checkpoint protein CHFR, suggesting that the two proteins function in the same mitotic checkpoint, the mitotic stress or prophase checkpoint. Although UBE2Q2 and CHFR appear to function in the same checkpoint, they do so in opposition to one another. In this study, we have explored possible mechanisms for UBE2Q2-mediated cell cycle arrest by assessing the effect of its inhibition on the accumulation of candidate gene products after checkpoint activation. We have found that prophase checkpoint activation in cells lacking functional UBE2Q2 is accompanied by accumulation of several mitotic regulatory proteins including Securin and Aurora-A. Studies are underway to determine if UBE2Q2 inactivation exerts this effect by inhibiting the ubiquitination of these proteins and to determine if proteins upregulated by UBE2Q2 inactivation are mediators of checkpoint function. Since CHFR functions as a tumor suppressor (Yu et al, Nat Genet 37;401-6) and in opposition to UBE2Q2, we hypothesize that UBE2Q2 will be oncogenic when constitutively over-expressed in vivo. In order to test this hypothesis, we have generated transgenic mice using an inducible system that allows expression of the FLAG-tagged UBE2Q2 transgene after Cre-mediated recombination, shown schematically in the accompanying figure. Several independent transgenic lines were generated and screened for expression of β-galactosidase (β-gal), the protein that serves as a marker for transgene expression. This system provides several advantages, allowing us to circumvent potential embryonic lethality in transgenics, providing an opportunity for tissue specific transgene expression, and allowing activation at a specific age or developmental stage. Two founders with abundant β-gal expression have been bred with mice expressing EIIa-Cre, which is constitutively expressed in all tissues. We have confirmed that Cre⁺-offspring from these matings have high level expression of UBE2Q2-FLAG. We are now breeding mice with the activated allele to generate a cohort of UBE2Q2-transgenic mice for evaluation. Since we hypothesize that UBE2Q2 over-expression will promote the development of malignancies, we will initially evaluate mice separately for the development of (1) spontaneous malignancies and (2) carcinogen-induced (DMBA) skin tumors. A group of about 100 transgenic mice and 100 littermate controls will be monitored for the spontaneous development of malignancies and submitted for necropsy in the case of death or an obvious malignancy. Since UBE2Q2 over-expression might have effects other than the promotion of tumors, we will also monitor the mice in these groups for weight gain (by weighing monthly), and metabolic and hematologic abnormalities (by laboratory evaluation). To assess the ability of UBE2Q2 over-expression to promote the development of DMBA-induced malignancies, we will give a single application of DMBA to the skin of about 25 transgenic mice and 25 littermate controls. Using this approach, there should be a low frequency of skin tumors in control mice (Yu et al, Nat Genet 37;401-6). In contrast, we expect a much higher frequency of skin tumors in UBE2Q2-transgenic mice. If UBE2Q2 over-expression promotes malignancies, we can evaluate the tissue specificity of this property by breeding with mice expressing a tissue-specific Cre. Finally, our data suggest that UBE2Q2 is a potentially useful target for the treatment of malignancies (Banerjee et al, Oncogene 26;6509-17). If our transgenic mice develop tumors, these mice will be a very useful tool for testing UBE2Q2 inhibitors that are developed.