Proteasomes are the main proteases responsible for cytosolic protein degradation and the production of major histocompatibility complex class I ligands. Incorporation of the interferon γ–inducible subunits low molecular weight protein (LMP)-2, LMP-7, and multicatalytic endopeptidase complex–like (MECL)-1 leads to the formation of immunoproteasomes which have been associated with more efficient class I antigen processing. Although differences in cleavage specificities of constitutive and immunoproteasomes have been observed frequently, cleavage motifs have not been described previously.We now report that cells expressing immunoproteasomes display a different peptide repertoire changing the overall cytotoxic T cell–specificity as indicated by the observation that LMP-7−/− mice react against cells of LMP-7 wild-type mice. Moreover, using the 436 amino acid protein enolase-1 as an unmodified model substrate in combination with a quantitative approach, we analyzed a large collection of peptides generated by either set of proteasomes. Inspection of the amino acids flanking proteasomal cleavage sites allowed the description of two different cleavage motifs. These motifs finally explain recent findings describing differential processing of epitopes by constitutive and immunoproteasomes and are important to the understanding of peripheral T cell tolerization/activation as well as for effective vaccine development.

We present a predictive method that can simulate an essential step in the antigen presentation in higher vertebrates, namely the step involving the proteasomal degradation of polypeptides into fragments which have the potential to bind to MHC Class I molecules. Proteasomal cleavage prediction algorithms published so far were trained on data from in vitro digestion experiments with constitutive proteasomes. As a result, they did not take into account the characteristics of the structurally modified proteasomes--often called immunoproteasomes--found in cells stimulated by gamma-interferon under physiological conditions. Our algorithm has been trained not only on in vitro data, but also on MHC Class I ligand data, which reflect a combination of immunoproteasome and constitutive proteasome specificity. This feature, together with the use of neural networks, a non-linear classification technique, make the prediction of MHC Class I ligand boundaries more accurate: 65% of the cleavage sites and 85% of the non-cleavage sites are correctly determined. Moreover, we show that the neural networks trained on the constitutive proteasome data learns a specificity that differs from that of the networks trained on MHC Class I ligands, i.e. the specificity of the immunoproteasome is different than the constitutive proteasome. The tools developed in this study in combination with a predictor of MHC and TAP binding capacity should give a more complete prediction of the generation and presentation of peptides on MHC Class I molecules. Here we demonstrate that such an approach produces an accurate prediction of the CTL the epitopes in HIV Nef. The method is available at www.cbs.dtu.dk/services/NetChop/.

The 436-amino acid protein enolase 1 from yeast was degraded in vitro by purified wild-type and mutant yeast 20S proteasome particles. Analysis of the cleavage products at different times revealed a processive degradation mechanism and a length distribution of fragments ranging from 3 to 25 amino acids with an average length of 7 to 8 amino acids. Surprisingly, the average fragment length was very similar between wild-type and mutant 20S proteasomes with reduced numbers of active sites. This implies that the fragment length is not influenced by the distance between the active sites, as previously postulated. A detailed analysis of the cleavages also allowed the identification of certain amino acid characteristics in positions flanking the cleavage site that guide the selection of the P1 residues by the three active ␤ subunits. Because yeast and mammalian proteasomes are highly homologous, similar cleavage motifs might be used by mammalian proteasomes. Therefore, our data provide a basis for predicting proteasomal degradation products from which peptides are sampled by major histocompatibility complex class I molecules for presentation to cytotoxic T cells.The eukaryotic 20S proteasome represents the catalytic core particle of the 26S proteasome, which is an essential component of the ubiquitin-dependent protein degradation pathway. The 20S particle is composed of 14 different but related subunits. Two outer disks, each containing seven ␣ subunits, guard the two inner heptameric ␤ subunit rings, which contain three proteolytically active sites each (1). How substrate molecules gain access to the active sites at the inner surface of the ␤ rings is not known. It has been hypothesized that on association of the 20S particle with regulatory complexes, such as the 19S cap structure or PA28, an opening in the middle of the ␣ subunit rings is induced allowing unfolded proteins to be fed into the 20S particle.Probably as a consequence of this closed structure, very few proteins have been found to be degraded by 20S proteasomes in vitro, and only very few individual fragments and cleavage sites generated during the degradation of proteins by eukaryotic 20S proteasomes have been analyzed until now. This kind of analysis is of interest for several reasons.First, it still has to be confirmed that the specificities of the proteasomal ␤ subunits observed in experiments using peptide substrates (3-8), summarized as trypsin-like, chymotrypsin (ChT)-like, and peptidylglutamyl-peptide hydrolyzing activity, correspond to cleavages performed in intact proteins, a situation physiologically more relevant.Second, the mechanism of protein degradation by the proteasome is not known. For the archebacterial proteasome, which is a less complex structure consisting of 14 identical ␣ and ␤ subunits (9), a processive model has been proposed (2, 10). Because of the high structural homology with the archebacterial particle a similar mechanism can be expected for eukaryotic proteasomes. Furthermore, it has been postulated that the 20S pro...

Proteasomes generate peptides that can be presented by major histocompatibility complex (MHC) class I molecules in vertebrate cells. Using yeast 20 S proteasomes carrying different inactivated ␤-subunits, we investigated the specificities and contributions of the different ␤-subunits to the degradation of polypeptide substrates containing MHC class I ligands and addressed the question of additional proteolytically active sites apart from the active ␤-subunits. We found a clear correlation between the contribution of the different subunits to the cleavage of fluorogenic and long peptide substrates, with ␤5/Pre2 cleaving after hydrophobic, ␤2/Pup1 after basic, and ␤1/Pre3 after acidic residues, but with the exception that ␤2/Pup1 and ␤1/Pre3 can also cleave after some hydrophobic residues. All proteolytic activities including the "branched chain amino acid-preferring" component are associated with ␤5/Pre2, ␤1/Pre3, or ␤2/ Pup1, arguing against additional proteolytic sites. Because of the high homology between yeast and mammalian 20 S proteasomes in sequence and subunit topology and the conservation of cleavage specificity between mammalian and yeast proteasomes, our results can be expected to also describe most of the proteolytic activity of mammalian 20 S proteasomes leading to the generation of MHC class I ligands.