Why the Structure but Not the Activity of the Immunoproteasome Subunit Low Molecular Mass Polypeptide 2 Rescues Antigen Presentation

Abstract

The proteasome is responsible for the generation of most epitopes presented on MHC class I molecules. Treatment of cells with IFN-γ leads to the replacement of the constitutive catalytic subunits β1, β2, and β5 by the inducible subunits low molecular mass polypeptide (LMP) 2 (β1i), multicatalytic endopeptidase complex-like-1 (β2i), and LMP7 (β5i), respectively. The incorporation of these subunits is required for the production of numerous MHC class I-restricted T cell epitopes. The structural features rather than the proteolytic activity of an immunoproteasome subunit are needed for the generation of some epitopes, but the underlying mechanisms have remained elusive. Experiments with LMP2-deficient splenocytes revealed that the generation of the male HY-derived CTL-epitope UTY246–254 was dependent on LMP2. Treatment of male splenocytes with an LMP2-selective inhibitor did not reduce UTY246–254 presentation, whereas silencing of β1 activity increased presentation of UTY246–254. In vitro degradation experiments showed that the caspase-like activity of β1 was responsible for the destruction of this CTL epitope, whereas it was preserved when LMP2 replaced β1. Moreover, inhibition of the β5 subunit rescued the presentation of the influenza matrix 58–66 epitope, thus suggesting that a similar mechanism can apply to the exchange of β5 by LMP7. Taken together, our data provide a rationale why the structural property of an immunoproteasome subunit rather than its activity is required for the generation of a CTL epitope.

Proteasome isoforms exhibit only quantitative differences in cleavage and epitope generation

“…2F) and gp100 208-217 peptides did not inversely correlate with the absolute speed of cleavage events within the peptide sequences (Supporting Information Table 1). Thus, in our experimental setup, the postulated immunoproteasome-dependent destructive cut after residue Ile 209 did not substantially interfere with the generation of the epitope precursors.A similar result was obtained for the synthetic substrate pp89 , where we focused on the generation kinetics of the peptide pp89 [20][21][22][23][24][25][26][27][28][29][30] . This peptide is the precursor of the H-2L d epitope pp89 [22][23][24][25][26][27][28][29][30] .…”

“…Our quantitative approach allowed us to verify another aspect of the epitope generation, which was supposed to lead to qualitative differences in epitope repertoire. By applying QME and measuring the relative prevalence and the absolute velocity of cleavage events within a given sequence, our experiments soundly invalidated the hypothesis that an epitope is 'destroyed' by prevalent cleavages within the epitope sequence [16,28,31,32]. Indeed, studying the peptides gp100 206-217 , pp89 20-30 and LLO 296-304 , we found no correlation between the absolute amount of the peptides and how frequent the substrate is hydrolysed within the peptide sequences (Supporting Information Table 1).…”

“…1) [21,22,24]. Despite this, these cells have been used as immunoproteasome source by various laboratories, including those who described qualitative differences between standardand immunoproteasomes in cleavage-site usage and generated epitope repertoire [12,15,20,[24][25][26][27][28]. Thus, their 20S proteasomes have been used in this study and referred as immunoproteasomes.…”

“…This peptide is the precursor of the H-2L d epitope pp89 [22][23][24][25][26][27][28][29][30] . Although the epitope pp89 [22][23][24][25][26][27][28][29][30] is generated by proteasomes, only the precursor is translocated into the endoplasmic reticulum where it is further trimmed by aminopeptidases prior to MHC class I binding [33].…”

“…We therefore focused on MHC class I epitopes derived from the gp100 201-230 , pp89 and LLO 291-317 sequences. The analysis of the generation of the melanoma gp100 protein-derived HLA-A*02-restricted immuno-dominant epitope gp100 209-217 [29, 30] was of particular interest because destructive cleavages within epitope sequences have been postulated to explain the insufficient generation of specific epitopes by standard-or immunoproteasomes [16,28,31,32]. Like Chapiro and coworkers [31], we were unable to detect by MS the gp100 [209][210][211][212][213][214][215][216][217] epitope among the peptides produced by either standard-or immunoproteasomes.…”