The 19-kilodalton adenovirus E1B transforming protein inhibits programmed cell death and prevents cytolysis by tumor necrosis factor alpha.

Abstract

The adenovirus E1A and E1B proteins are required for transformation of primary rodent cells. When expressed in the absence of the 19,000-dalton (19K) E1B protein, however, the E1A proteins are acutely cytotoxic and induce host cell chromosomal DNA fragmentation and cytolysis, analogous to cells undergoing programmed cell death (apoptosis). E1A alone can efficiently initiate the formation of foci which subsequently undergo abortive transformation whereby stimulation of cell growth is counteracted by continual cell death. Cell lines with an immortalized growth potential eventually arise with low frequency. Coexpression of the E1B 19K protein with E1A is sufficient to overcome abortive transformation to produce high-frequency transformation. Like E1A, the tumoricidal cytokine tumor necrosis factor alpha (TNF-alpha) evokes a programmed cell death response in many tumor cell lines by inducing DNA fragmentation and cytolysis. Expression of the E1B 19K protein by viral infection, by transient expression, or in transformed cells completely and specifically blocks this TNF-alpha-induced DNA fragmentation and cell death. Cosegregation of 19K protein transforming activity with protection from TNF-alpha-mediated cytolysis demonstrates that both activities are likely the consequence of the same function of the protein. Therefore, we propose that by suppressing an intrinsic cell death mechanism activated by TNF-alpha or E1A, the E1B 19K protein enhances the transforming activity of E1A and enables adenovirus to evade TNF-alpha-dependent immune surveillance.

Bcl-2, Bcl-xL and adenovirus protein E1B19kD are functionally equivalent in their ability to inhibit cell death

“…Moreover, crossing of bcl-2 transgenic mice onto the Fas/APO-1 (CD95)-de®cient lpr background revealed that gain of Bcl-2 and loss of Fas/APO-1 (CD95) synergise in promoting lymphocyte accumulation, indicating that these two molecules regulate distinct pathways to lymphocyte apoptosis (Reap et al, 1995;Strasser et al, 1995c). Although in certain cell lines and hepatocytes of transgenic mice, Bcl-2 and certain homologues have been documented to provide some measure of protection ( Figure 6) and (Hashimoto et al, 1991;White et al, 1992;Itoh et al, 1993;JaÈ aÈ ttelaÈ et al, 1995;Lacronique et al, 1996;Rodriguez et al, 1996), the e ects have been modest at best. In our experiments, it is clear that levels of expression that a ord substantial protection, in a dose-dependent manner, against serum deprivation or treatment with staurosporine only weakly inhibited death induced by the Fas/APO-1 (CD95) ligand or TNF (for example, compare data for WEHI-164 cells between Figure 3f to data in Figure 6c, and for SKW6 and CH1 cells in Figure 5).…”

“…For example, adenovirus E1B19kD protein protected some cultured cell lines against TNF-mediated cytotoxicity (Hashimoto et al, 1991;White et al, 1992), while Bcl-2 could not (Vanhaesebroeck et al, 1993), and Bcl-x L enhanced the survival of WEHI-231 B lymphoma cells treated with the immunosuppressive drug cyclosporin A but Bcl-2 was apparently unable to do so (Gottschalk et al, 1994). Furthermore, some reports documented that neither Bcl-2, Bcl-x L nor E1B19kD can block apoptosis induced via Fas/APO-1 (CD95) in lymphoid cell lines (Memon et al, 1995;Strasser et al, 1995c) while other experiments have demonstrated that all three proteins can provide some protection against Fas/APO-1 (CD95)-and/or TNF receptor-transduced apoptosis in non-lymphoid lines (Hashimoto et al, 1991;White et al, 1992;Itoh et al, 1993;JaÈ aÈ ttelaÈ et al, 1995). It remains unclear whether these discrepancies are the result of di erences in cell death regulation between di erent cell types or re¯ect functional di erences between members of the Bcl-2 family.…”

Viruses with deletions in antiapoptotic genes as potential oncolytic agents

“…Since tumors have various mechanisms blocking the apoptosis pathways, it is likely that defect in apoptosis pathways in some tumor types will complement replication of these mutant viruses, thereby supporting viral replication and spread, whereas other cell types might not be complementing. In fact, although most tumor cells infected with these mutants did not become more apoptosis-prone, some publications have identified a number of cells that do (White et al, 1992;Sauthoff et al, 2000). It seems likely that there are two main types of cells that respond differently to these mutants (Figure 3): the first type of cells (type I cells) are not affected in the apoptosis process.…”

Characterization of the 55K adenovirus type 5 E1B product and related proteins

“…The 19K and 55K E1B proteins can function independently, but transforming efficiency is increased when both are expressed (Bernards et al, 1986;Barker & Berk, 1987;White & Cipriani, 1990;Yew et al, 1990;McLorie et al, 1991;Zhang et al, 1992). The 19K protein appears to protect against programmed cell death induced as a consequence of disruption of growth control pathways by E1A (White et al, 1992;Rao et al, 1992). The 55K protein may derive some or all of its transforming activity from interactions with and inactivation of the cellular p53 tumour suppressor (Sarnow et al, 1982b;Kao et at., 1990;Yew & Berk, 1992 cell metabolism (Babiss & Ginsberg, 1984;Logan et al, 1984;Babiss et al, 1985;Halbert et at., 1985;Pilder et al, 1986: Bernards et al, 1986Barker & Berk, 1987;Sandler & Ketner, 1989;Bridge & Ketner, 1990, Leppard & Shenk, 1989, Yew et al, 1990McLorie et al, 1991).…”