E2F1-INDUCED APOPTOSIS REQUIRES DNA-BINDING BUT NOT TRANSACTIVATION AND IS INHIBITED BY THE RETINOBLASTOMA PROTEIN THROUGH DIRECT INTERACTION

E2F1 overexpression has been shown to induce apoptosis in cooperation with p53. Using Saos-2 cells, which are null for p53 and lack function al Rb, we have demonstrated that E2F1 overexpression can also induce a poptosis in the absence of p53 and retinoblastoma protein (Rb). E2F1-i nduced apoptosis can be specifically inhibited by Rb bur not mdm2, whi ch is known for its ability to inhibit p53-induced apoptosis. Through the study of the apoptotic function of a set of E2F1 mutants, it was c lear that the transactivation and the apoptotic function of E2F1 are u ncoupled. The transactivation-defective E2F1 mutants E2F1(1-374), E2F1 (390-1)DF(Delta mdm2), and E2F1(406-415)(Delta Rb) can induce apoptosi s as effectively as wild-type E2F1. In contrast to E2F1 transactivatio n, the DNA-binding activity of E2F1 was proven to be essential for its apoptotic function, as the DNA-binding-defective mutants E2F1(132) an d E2F1(132)(1-374) failed to induce apoptosis, Therefore Rb may inhibi t E2F1-induced apoptosis by mechanisms other than the suppression of t he transactivation of E2F1, This hypothesis was supported by our obser vation that although Rb overexpression can specifically repress the ap optosis induced by wild-type E2F1 and a Rb-binding-competent E2F1 muta nt E2F1/390-1)DF(Delta mdm2), it failed to inhibit the apoptosis induc ed by mutants E2F1(1-374) and E2F1(Delta 406-415)(Delta Rb), which are defective or reduced in Iib binding and transactivation. All of these points argue for a novel function for E2F1 and Rb in controlling apop tosis. The results also indicate that transcriptional repression rathe r than the transactivation function of E2F1 may be involved in its apo ptotic function. The results presented here may provide us some physio logical implication of the repression function of the Rb-E2F1 complex.