Regulation of protein degradation by virus derived repeated amino acid sequences

Abstract

The ubiquitin-proteasome pathway plays a central role in the controlled degradation of short-lived and regulatory proteins important m a variety of cellular processes, including antigen presentation and combating viral infections. The presentation of antigenic peptides derived from ubiquitin-proteasome dependent degradation of vital proteins to MHC class I restricted cytotoxic T cells is a central component of antiviral responses. It is therefore not surprising that viruses have developed strategies to block this proteolytic machinery m order to escape detection by the host immune system. A particularly interesting example of viral protein that interferes with the ubiquitinproteasome pathway is the Epstein-Barr virus (EBV) nuclear antigen-1 (EBNA1) that contains a long internal glycine-alanine repeat (GAr) and generates a cis-acting inhibitory signal that interferes with antigen presentation.

The aim of this study was to investigate the mechanism by which the GAr interferes with the ubiquitin-proteasome pathway in protecting proteins from proteolysis. We used the capacity of GAr to act as a transferable element by inhibiting the processing of a wide variety of proteasomal substrates. Four main aspects were explored in the present study. We studied the influence of the GAr on the proteasomal degradation of different viral and cellular proteins. We also analyzed the conformational stability of the chimeric protein. Furthermore, we investigated the impact of amino acid composition and trans-inhibitory effect of the GAr on the ubiquitin-dependent degradation.

First, we compared the proteasomal degradation of native EBNA1, GAr deleted EBNA1 and an EBNA4 chimeric protein. EBNA4 was efficiently degraded whereas EBNA1 was resistant to degradation. EBNA1 degradation was restored by deletion of the GAr, both in vitro and in vivo, whereas insertion of the GAr of various lengths and in different positions prevented the degradation of EBNA4 without considerable effect on ubiquitination. These results proposed that the GAr might affect MHC I restricted responses by inhibiting antigen processing via the ubiquitin/proteasome pathway.

Next, we investigated the impact of the GAr in degradation of a well-known target of the ubiquitin-proteasome pathway, inhibitor of nuclear factor kappa B, I kappa B alpha in vivo. We found that insertion of octamer repeat GGAGAGAG with three alanines was sufficient to block I kappa B alpha from signal-induced degradation. This study demonstrated that chimeras were phosphorylated and ubiquitinated in response to TNF alpha and retained the capacity to bind NF-kappa B, but then were released from NF-kappa B and failed to associate with the proteasome. It suggested that the repeat allects ubiquitin-proteasome pathway at a step that lies downstream of ubiquitination and upstream of the interaction of ubiquitinated proteins with the proteasome. We also used p53 to generate a set of chimeras with different length and site of insertion of the GAr to monitore Mdm2- and E6-induced degradation. p53 chimeras were ubiquitinated and protected from proteolysis, retained the capacity to interact with the S5a ubiquitin-binding subunit of the proteasome and maintained the functional properties of wild-type p53.

Previous work raised a possibility that insertion of the GAr domain in target protein may interfere with direct masking of a putative binding site or increase the stability of chimeric protein against unfolding activity of molecular chaperones or the 19S proteasome. The NMR and CD spectroscopy data demonstrated that glycine-alanine insert assumed a flexible random coil conformation and its presence did not increase the stability of I kappa B alpha against denaturation. These results suggested that protection against proteolytic degradation may not be based on structural differences or stabilization caused by the GAr and excluded a structural effect of the Gly-Ala segment. Next aspect of this study deals with the relationship between the inhibitory activity and the amino acid composition of the repeat. Inhibition of TNF alpha-induced degradation was achieved by insertion of different octamers containing three alanines interspersed by no more then three consecutive glycines.

We also found that inhibitory activity of the GAr was abolished when the length of the glycine spacer was increased and when a single alanine residue of the octamer was substituted with a polar or charged amino acid. A substitution of alanines with valine resulted in the maintenance of the inhibitory effect. These findings suggested a model where inhibition requires an interaction of at least three alanine residues of the GAr in a beta-strand conformation with adjacent hydrophobic binding pocket of a putative receptor.

We also studied in trans inhibition of degradation of proteasomal substrate lysozyme in the presence of the GAr oligopeptide that mimics a sequence of the repeat. The presence of peptide did not affect the ubiquitination of lysozyme, hydrolysis of fluorogenic substrate or binding of a synthetic tetra-ubiquitin complex to the S5a subunit. Our findings therefore propose that the GAr domain may act by weakening the interaction of ubiquitinated substrates with the 19S proteasome regulator and promote the premature release of the substrate.