PA28 modulates antigen processing and viral replication during coxsackievirus B3 infection

Figures

• 

  Fig 1. Regulation of PA28αβ during CVB3 infection. (A) 6–8 weeks old, male C57BL/6 mice (n = 4) were infected with CVB3 and sacrificed at day 8 after infection. Age- and gender-matched naive mice (n = 4) served as a control. Tissue homogenates of heart and spleen were analyzed by Western blotting to determine PA28α and PA28β protein levels. Actin served as a loading control. Densitometric analysis was performed and PA28α/PA28β levels are shown as relative intensities with a normalization based on the respective levels of actin (means + SEM, n = 4). (B) Primary embryonic cardiomyocytes (eCMs) obtained from wild-type mice and HeLa cells were infected with CVB3 at MOI 0.1 for 0 or 24 h. PA28α/PA28β mRNA levels were determined by real-time qPCR (means + SEM, n = 3). (C) HeLa cells were infected with CVB3 at MOI 0.1 for indicated lengths of time. PA28α and PA28β protein levels in total protein extracts were determined by Western blotting. The densitometric analysis of PA28α/PA28β protein levels is depicted as relative intensities (means + SEM, n = 3).

• 

  Fig 2. Protein homeostasis in TLR3-activated bone marrow derived macrophages. (A) Splenocytes freshly isolated from 8–12 weeks old wild-type and PA28αβ−/− mice were incubated with MV151 activity based probe (ABP). Binding of ABP was detected by flow cytometry and is shown as mean fluorescence intensity (means + SEM, n = 4 for wild-type, n = 3 for PA28α/β−/−, 2 independent experiments). Bone marrow derived macrophages (BMM) were obtained from 8–10 weeks old wild-type and PA28αβ−/− mice. BMMs were treated with poly(I:C). (B) After 24 hours of stimulation, cells were incubated with DCFH-DA to measure reactive oxygen species (ROS). Resulting DCF fluorescence was detected by flow cytometry and mean fluorescence intensity is shown as fold change relative to untreated cells (means + SEM, n = 3). (C) BMM lysates obtained after 0, 24, and 48 hours of treatment with poly(I:C) were subjected to Oxyblot analysis (depicted results were obtained from the same membrane). Results of a densitometric analysis are shown as relative intensities in comparison to untreated wild-type BMM (means + SEM, n = 3, no significant differences were detected). (D) Total cell lysates obtained after 24 and 48 hours of…

• 

  Fig 3. Effect of PA28α/β on coxsackieviral replication. (A) eCMs were isolated from wild-type as well as PA28α/β−/− mice and cell lysates were subjected to Western blotting of PA28α and PA28β (top). Cells were infected with CVB3 at MOI 0.1 for 16 hours and CVB3 mRNA was determined by real-time qPCR (bottom). Data are means + SEM of n = 4 independent experiments. (B) HeLa cells were transfected with siRNA targeting PA28α/β for 72 hours and protein levels were assessed by Western blotting of PA28α and PA28β (top left). siRNA-treated HeLa cells were infected with CVB3 at indicated MOIs for 16 hours and relative expression of CVB3 mRNA was determined by real-time qPCR (bottom, left graph, means + SEM of n = 2). In addition, lysates of CVB3-infected HeLa cells (MOI 0.01) pre-treated with siRNA were analyzed by Western blotting of CVB3 VP1 structural protein and GAPDH (top right). Densitometric analysis of VP1 levels, normalized to GAPDH, is depicted as relative intensities in comparison to infected HeLa cells pre-treated with off-target siRNA (means + SEM, n = 3). (C) HeLa cells were transiently transfected with human PA28α/β. Cell lysates were subjected to Western blotting to confirm…

• 

  Fig 4. Influence of PA28α/β on MHC class I antigen processing of CVB3 epitopes. (A) Murine embryonic fibroblasts were treated with IFN-β for 24 hours. For assessing MHC class I expression, cells were stained with anti-H-2Kb mAb and analyzed by flow cytometry (mean + SEM, n = 3). (B) For in vitro peptide processing studies PA28 from human erythrocytes, 20S sp from human lymphoblasts (T2 cells) and 20S ip from stably LMP2/LMP7-transfected T2 cells (T27 cells) were purified. The purity of the preparations was examined by Coomassie staining of a SDS-PAGE gel. Two epitope-harboring CVB3 peptides—the 28-mer polypeptide P3D2158-2185 (C) and the 31-mer polypeptide VP2272-302 (D)–were processed in the presence of T2-derived sp (filled black circles) and T27-derived ip (filled green circles) for the indicated incubation periods. Open circles represent results of proteasome fragmentation studies in the presence of PA28 for sp (black) and ip (green). The graphs in the upper panel summarize the generation of the respective CVB3 epitopes (C) P3D2170-2177 and (D) VP2285-293 detected by ESI–MS/MS. The lower graphs depict the substrate degradation as fold changes of the ion counts at point in time =…

• 

  Fig 5. Characterization of PA28-proteasome complexes in wild-type and LMP7−/− macrophages. (A) Cellular extracts were generated from wild-type, LMP7−/− and PA28α/β−/− BMMs. Proteins were separated by native PAGE and subjected to Western blotting of proteasome core (α-α4) and PA28 complex (α-PA28β) co-stained for PA700 complex (α-Rpt6), respectively. 20S proteasome purified from mouse B8 fibroblasts was applied for comparison. Fluorescence dye-coupled secondary antibodies were used for detection. *1 –unspecific signal, *2 –PA700-20S and PA700-20S-PA28, *3 –PA700 intermediate complex [51], 4*–PA200-20S-PA200 [52], 5*–PA28α/β not bound to a proteasome core. (B) Magnified section of the α-α4/α-Rpt6 and α-PA28β/α-Rpt6 double-staining from (A) depicting the high molecular mass proteasome complexes PA700-20S and PA700-20S-PA28. (C) Expression and accessibility of catalytically active proteasome subunits within different proteasome-proteasome regulator complexes was investigated with MV151 pan-reactive ABP binding to all active sites and analyzed by densitometry (means + SEM of n = 3). Amido black staining indicates equal protein loading. (D) Total BMM protein extracts were subjected to…

• 

  Fig 6. Investigation of PA28αβ-/- mice during CVB3 myocarditis. PA28+/+ and PA28-/- mice were infected with CVB3. (A) Body weight was monitored daily and mice were sacrificed at day 8 post infection (means ± SEM). (B) Heart, lung, spleen and thymus weight was determined and normalized to the respective tibia length (means + SEM). (C) Representative haematoxylin/eosin (HE) staining of cardiac tissue sections obtained from CVB3-infected PA28+/+ (n = 10) and PA28−/− (n = 7) mice are shown. (D) Myocardial damage comprising cardiac cell necrosis, inflammation, and scarring was quantified applying a myocarditis score from 0 to 4 [36]. (E) The severity of pancreas tissue destruction upon CVB3 infection was rated on a scale from 0% to 100% on HE stains of pancreatic tissue sections.