Faculty Bibliography

Genome level information coupled with phylogenetic analysis of specific genes and gene families allow for a better understanding of the structure and function of their protein products. In this study, we examine the mammalian uroplakins (UPs) Ia and Ib, members of the tetraspanin superfamily, that interact with uroplakins UPII and UPIIIa/IIIb, respectively, using a phylogenetic approach of these genes from whole genome sequences. These proteins interact to form urothelial plaques that play a central role in the permeability barrier function of the apical urothelial surface of the urinary bladder. Since these plaques are found exclusively in mammalian urothelium, it is enigmatic that UP-like genomic sequences were recently found in lower vertebrates without a typical urothelium. We have cloned full-length UP-related cDNAs from frog (Xenopus laevis), chicken (Gallus gallus), and zebrafish (Danio rerio), and combined these data with sequence information from their orthologs in all the available fully sequenced and annotated animal genomes. Phylogenetic analyses of all the available uroplakin sequences, and an understanding of their distribution in several animal taxa, suggest that: (i) the UPIa/UPIb and UPII/UPIII genes evolved by gene duplication in the common ancestor of vertebrates; (ii) uroplakins can be lost in different combinations in vertebrate lineages; and (iii) there is a strong co-evolutionary relationship between UPIa and UPIb and their partners UPII and UPIIIa/IIIb, respectively. The co-evolution of the tetraspanin UPs and their associated proteins may fine-tune the structure and function of uroplakin complexes enabling them to perform diverse species- and tissue-specific functions. The structure and function of uroplakins, which are also expressed in Xenopus kidney, oocytes and fat body, are much more versatile than hitherto appreciated

The Ras superfamily of GTP-binding proteins is involved in many cellular processes, including cell proliferation, movement, and morphology. One such member, Ral GTPase, activates downstream signaling molecules after a conversion to the active state on GTP binding. The RalGDS-related (Rgr) oncogene belongs to the RalGDS family of guanine nucleotide exchange factors (GEFs). RalGEFs activate Ral by stimulating the dissociation of GDP, allowing the binding of GTP and the initiation of downstream signaling events by Ral effectors. Rgr was first identified as a fusion between the rabbit homolog of the Rad 23 gene and the Rgr gene in a rabbit squamous cell carcinoma. The Rgr portion of the fusion was demonstrated to contain the oncogenic activity. The human form of the Rgr oncogene was identified recently, and expression was detected in human T-cell malignancies. This chapter describes the analysis of rabbit and human Rgr function using various methods. These assays may be used for the study of oncogene function in other systems

The INK4 family of proteins negatively regulates cell cycle progression at the G(1)-S transition by inhibiting cyclin-dependent kinases. Two of these cell cycle inhibitors, p16(INK4A) and p15(INK4B), have tumor suppressor activities and are inactivated in human cancer. Interestingly, both INK4 genes express alternative splicing variants. In addition to p16(INK4A), the INK4A locus encodes a splice variant, termed p12--specifically expressed in human pancreas--and ARF, a protein encoded by an alternative reading frame that acts as a tumor suppressor through the p53 pathway. Similarly, the human INK4B locus encodes the p15(INK4B) tumor suppressor and one alternatively spliced form, termed as p10. We show here that p10, which arises from the use of an alternative splice donor site within intron 1, is conserved in the mouse genome and is widely expressed in mouse tissues. Similarly to mouse p15(INK4B), p10 expression is also induced by oncogenic insults and transforming growth factor-beta treatment and acts as a cell cycle inhibitor. Importantly, we show that mouse p10 is able to induce cell cycle arrest in a p53-dependent manner. We also show that mouse p10 is able to inhibit foci formation and anchorage-independent growth in wild-type mouse embryonic fibroblasts, and that these antitransforming properties of mouse p10 are also p53-dependent. These results indicate that the INK4B locus, similarly to INK4A-ARF, harbors two different splicing variants that can be involved in the regulation of both the p53 and retinoblastoma pathways, the two major molecular pathways in tumor suppression.

Proteins controlling cell growth, differentiation, apoptosis, and oncogenic stress are often deregulated in tumor cells. However, whether such deregulations affect tumor behavior remains poorly understood in many tumor types. We recently showed that the urothelium-specific expression of activated H-ras and SV40 T antigen in transgenic mice produced two distinctive types of tumors strongly resembling the human superficial papillary tumors and carcinoma in situ of the bladder, respectively. Here we assessed the expression of a key set of cell cycle regulators in these mouse tumors and in a new transgenic line expressing a cyclin D1 oncogene in the urothelium. We found that urothelia of the wild-type and cyclin D1 transgenic mice exhibited a profile of cell cycle regulators found in quiescent (G(0)) cells, indicating that urothelium overexpressing the cyclin D1 (an 8-fold increase) is reminiscent of normal urothelium and remains slow-cycling. Low-grade superficial papillary tumors induced by activated H-ras had no detectable Rb family proteins (Rb, p107, and p130) and late cell cycle cyclins and kinases (cyclin A, E, and CDK1), but had increased level of p16, p53, and MDM2. These data suggest that the inactivation of the Rb pathway plays an important role in H-ras-induced superficial papillary tumors and that oncogenic H-ras can induce a compensatory activation of alternative tumor suppressor pathways. In contrast, carcinoma in situ of the bladder induced by SV40 T antigen had increased expression of cell cycle regulators mainly active in post-G(1) phases. The fact that phenotypically different bladder tumors exhibit different patterns of cell cycle regulators may explain why these tumors have different propensity to progress to invasive tumors. Our results indicate that the transgenic mouse models can be used not only for studying tumorigenesis but also for evaluating therapeutic strategies that target specific cell cycle regulators

Point mutations in ras genes have been found in a large number and wide variety of human tumors. These oncogenic Ras mutants are locked in an active GTP-bound state that leads to a constitutive and deregulated activation of Ras function. The dogma that ras oncogenes are dominant, whereby the mutation of a single allele in a cell will predispose the host cell to transformation regardless of the presence of the normal allele, is being challenged. We have seen that increasing amounts of Ras protooncogenes are able to inhibit the activity of the N-Ras oncogene in the activation of Elk in NIH 3T3 cells and in the formation of foci. We have been able to determine that the inhibitory effect is by competition between Ras protooncogenes and the N-Ras oncogene that occurs first at the effector level at the membranes, then at the processing level and lastly at the effector level in the cytosol. In addition, coexpression of the N-Ras protooncogene in thymic lymphomas induced by the N-Ras oncogene is associated with increased levels of p107, p130 and cyclin A and decreased levels of Rb. In the present report, we have shown that the N-Ras oncogene is not truly dominant over Ras protooncogenes and their competing activities might be depending on cellular context