MHC class I antigen processing distinguishes endogenous antigens based on their translation from cellular vs. viral mRNA

Abstract

To better understand the generation of MHC class I-associated peptides, we used a model antigenic protein whose proteasome-mediated degradation is rapidly and reversibly controlled by Shield-1, a cell-permeant drug. When expressed from a stably transfected gene, the efficiency of antigen presentation is ~2%, that is, one cell-surface MHC class I-peptide complex is generated for every 50 folded source proteins degraded upon Shield-1 withdrawal. By contrast, when the same protein is expressed by vaccinia virus, its antigen presentation efficiency is reduced ~10-fold to values similar to those reported for other vaccinia virus-encoded model antigens. Virus infection per se does not modify the efficiency of antigen processing. Rather, the efficiency difference between cellular and virus-encoded antigens is based on whether the antigen is synthesized from transgene- vs. virus-encoded mRNA. Thus, class I antigen-processing machinery can distinguish folded proteins based on the precise details of their synthesis to modulate antigen presentation efficiency.

Fig. 1.

Model antigens and a schematic of SCRAP presentation and quantification. (A) The various constructs used in this study are depicted. The red “S” box denotes the SIINFEKL peptide, influenza A virus (IAV) nucleoprotein is in orange, and the green “Ub” box represents ubiquitin. (B) EL4/SCRAP cells were washed in a mild citric acid buffer (pH 3.0) to remove existing K^b–SIINFEKL complexes and cultured in the presence or absence of 5 μM Shield-1. At the indicated times, K^b–SIINFEKL complexes and GFP levels were determined by flow cytometry. (C) Same as in A, except Shield-1 was removed following an initial 3.5-h incubation and, following a second acid wash, cells were cultured in the presence of cycloheximide (CHX). (D) An example of a Western blot for GFP using recombinant GFP standards and lysates from EL4/SCRAP cells treated with or without Shield-1 for 3 h.

Fig. 2.

SCRAP protein expressed by rVV is similar to other model antigens. rVV expressing either SCRAP, NP-S-GFP, or Ub-R-NP-S-GFP was used to infect L-K^b cells. Cells were cultured with or without Shield-1 (for SCRAP infection) and analyzed at the indicated times for GFP (Left) or K^b–SIINFEKL (Right) by FACS analysis.

Fig. 3.

Similar antigen presentation kinetics between different cell types and viruses. (A) L-K^b and EL4 cells were infected with rVSV expressing either Venus-Ub-SIINFEKL (Left) or NP-S-GFP proteins (Right), and antigen presentation as well as fluorescent protein expression were monitored by FACS. (B) L-K^b cells were infected with rVV and rVSV viruses expressing NP-S-GFP and monitored as in A.

Fig. 4.

Virus-expressed antigens are presented at lower efficiencies than self-antigens. (A) HeLa K^b cells were either infected with rVV SCRAP or transfected with SCRAP DNA constructs, and antigen presentation as well as GFP expression were determined 4 hpi or post-acid wash. (B) Representative histograms after 4 h of Shield-1 treatment of SCRAP-expressing cells (blue trace) compared with non-SCRAP-expressing cells (red trace). K^b–SIINFEKL staining with 25-D1.16 mAb (Right) is restricted to GFP-positive cells. (C) EL4/SCRAP or HeLa K^b cells transfected with SCRAP DNA were infected with VSV or rVV, respectively, and both K^b–SIINFEKL and GFP levels were determined 4 hpi. The efficiency of presentation was determined by dividing the MFI of K^b–SIINFEKL staining by the MFI of GFP. Values were compared with uninfected cells and are plotted as a percentage. (D) Levels of K^b–SIINFEKL derived from DRiP substrates (in the presence of Shield-1) from SCRAP expressed as transfected DNA or by rVV in HeLa K^b cells were normalized to GFP expression. Cells (both transfected and nontransfected) were briefly washed in mild acid and cultured for 2 h in complete media. Cells were then harvested and were either left uninfected or infected with control rVV at an MOI of 10, followed by an additional 4 h in culture before FACS analysis. Background levels of K^b–SIINFEKL were determined immediately following rVV infection. These data are representative of three independent experiments.