CELLULAR FUNCTION OF THE RETINOBLASTOMA GENE PRODUCT

视网膜母细胞瘤基因，RB，最初被确定为假定的

The retinoblastoma gene, RB, was initially identified as the putative suppressor of retinoblastoma in human. The RB gene is expressed in many cell types throughout development and the RB protein plays an important role in the regulation of cell division and cellular differentiation. The current model of RB function proposes that it binds transcription factors, such as E2F, to inhibit their activity. Phosphorylation of RB by the cyclin-dependent protein kinases inactivates the protein binding function, and leads to the activation of genes important to cell division. Although this model may in principle be correct, it is based on a rather superficial understanding of the RB protein and does not explain the phenotypes of the RB homozygous mutant mice. Two aspects of this model are too simplistic as indicated by our recent results. First, we have found that RB contains two protein binding domains. In addition to the so-called "A/B pocket" that binds viral oncoproteins and E2F, RB contains a C-terminal protein binding domain (CPBD). The CPBD and the A/B pocket, we can show, are functionally independent. Second, we have mapped eight phosphorylation sites in RB. Preliminary results using phosphorylation site mutants have indicated that the different sites may regulate the different binding domains of RB. This finding suggests that RB can exist in one of several functional states depending on which of the sites are phosphorylated. We will pursue these two new insights on the RB protein and our immediate goal includes the testing of a hypothesis that the RB protein functions as a molecular "matchmaker" to promoter the assembly of specific protein complexes in the nucleus. We propose that RB brings together, through its two binding domains, proteins that otherwise may not interact with one another. Each RB-mediated protein complex can be regulated by kinases that phosphorylate RB, and the composition of those complexes can be altered by the phosphorylation of specific sites. To test this hypothesis we will focus on the following four specific aims; (1) To further characterize the C-terminal protein binding domain (CPBD) of RB. We will define the minimal sequences of this domain, determine its solution structure by NMR and investigate whether it binds other nuclear tyrosine kinases. (2). To determine if the CPBD can interfere with the function of the wild type RB protein. The "matchmaking" hypothesis predicts that disruption of either of the two protein binding domains will inactivate RB. If so, overproduction of the CPBD of RB should compete for the binding of cellular proteins and neutralize the RB function. (3). To determine the specific regulatory roles of the C-terminal phosphorylation sites of RB. Specific phosphorylation site mutations will be prepared and their effect on the activity of the A/B pocket and the CPBD will be measured to demonstrate the differential regulation. (4). To study the biological effects of specific Ala and Glu mutations at the phosphorylation sites. We will investigate whether the different phosphorylation sites have different regulatory roles on the biological function of RB.