siRNA knockdown of CSN5 increases ubiquitination of exosomal proteins. A: Exosomes were analyzed by Western blotting using a monoclonal antibody, anti-CSN5, and rabbit polyclonal antibodies to CSN1, 3, and 8, of COP9. Exosome markers, including CD63 and Tsg101, were also analyzed. 293 cells stably transfected with pGC93 (siRNA CSN5 expression vector) or 293 cells transfected with scrambled siRNAs were treated with doxytetracycline (1.3 μg/ml) over a 5-day period. The cells were then used for the following experiments (B through F). In other experiments, 293 cells stably transfected with either pGC93 or scrambled siRNA were transfected with pcDNA3-HIVGag-GFP before being treated with doxytetracycline and subsequently used in experiment G. B: Exosomes were purified from supernatants using sucrose gradients and analyzed by electron microscopy (scale bar = 200 nm). The micrograph is representative of two independent experiments. C: Fifty μg of total cell lysate from 293 cells pretreated with doxytetracycline over 5 days were loaded on 10% SDS gels for PAGE separation. The separated proteins were transferred to a nitrocellulose membrane for Western blotting and probed using antibodies as indicated in Figure 1C. Representative Western blots are shown. D: Fifty μg of protein extracted from exosomes isolated from the supernatants of 293 cells (lanes 1 and 2) or total cell lysates (lanes 3 and 4) were used for ubiquitin Western blot analysis. β-actin detection served as a control for Western blotting analysis. E: Five-day doxytetracycline treated 293 cells (E, panels 3 and 4) or exosomes (E, panels 1 and 2) isolated from the supernatants of 293 cells were lysed using immunoprecipitation lysate buffer. Immunoprecipitation assays were performed using 500 μg aliquots of the total protein lysates and an anti-ubiquitin monoclonal antibody (E, panels 1 and 3) or mouse IgG as a control (E, panels 2 and 4). Precipitated proteins were run in western blots probed with anti-CD63 and anti-HSP70. The presence of HSP70 (F, top panel) and CD63 (F, bottom panel) were determined by Western blot analysis of exosomes purified from the same 293 cells as described above. G: The presence of HIV Gag (top panel) and CD63 (bottom panel) in exosomes purified from 5-day doxytetracycline treated 293 cells were determined by Western blot analysis. Data presented in A–D represent three independent experiments that yielded similar results. The signal intensity of each protein analyzed in E–G was quantified using an Odyssey infrared imaging system (LI-COR). The ratios of signal intensity of each protein in exosomes of CSN5 siRNA: scramble siRNA knockdown 293 cells were calculated and plotted (E–G, bar graphs). The data are presented as means of three independent experiments.

JAMM domain of CSN5 plays a role in the regulation of ubiquitination of exosomal proteins. A: 293 cells were transfected with different plasmid DNAs (10 μg) as indicated in Figure (A). Thirty-six hours after the transfection, the cells (1 × 10⁶) were lysed in lysis buffer and 50 μg of total protein from each lysate were resolved on a 10% SDS gel by PAGE. The proteins were then transferred to a nitrocellulose membrane and the blots were probed with a mouse anti-CSN5 antibody. 293 cells were transfected with pLEYFP-CSN5 or pLEYFP-CSN5-JAMMDN for 36 hours. The exosomes isolated from the supernatants of transfected 293 cells were immunoprecipitated with an anti-ubiquitin monoclonal antibody, followed by Western blot analysis for HSP70 (B, top left panels) and CD63 (B, bottom left panels). In other experiments, 293 cells were lysed and used for Western blot analyses of mono- or polyubiquinated proteins (C) or exosomes were lysed and used for Western blot analysis (D). The exosomes isolated from the supernatants of 293 cells transfected with pLEYFP-CSN5 or pLEYFP-CSN5-JAMMDN PLEYFP plus pcDNA3-HIVGag-GFP (E, left panel; F, top left panel) or plus pcDNA3 (E, right panel, F, right panel) were used for ubiquitin immunoprecipitation, followed by Western blot analysis of GFP (E) or Gag (F). An equal concentration of normal mouse IgG was used as a control for immunoprecipitation, followed by Western blot analysis of Gag (F, bottom left panel). 293 cells were transfected with pcDNA3.1-CSN5 or pcDNA3.1-CSN5Mut for 36 hours. The sucrose purified exosomes were then run in immunoprecipitation assays using an anti-ubiquitin monoclonal antibody (G, panels 1 and 3) or mouse IgG (G, panels 2 and 4). Western blot analysis of HSP70 (G, top left panels) and CD63 (G, bottom left panels) followed. The signal intensity of each protein analyzed in (B, D–F, and G) was quantified using an Odyssey infrared imaging system (LI-COR). The ratios of signal intensity of each protein in exosomes of 293 cells stably transfected with CSN5JAMMDN : CSN5 genes (B, D, E, and F, bar graphs) and CSN5Mut:CSN5 genes (G, bar graph) were calculated and plotted. The data are presented as means of three independent experiments.

COP9 associated CSN5 regulates the deubiquitination of exosomal proteins. One microgram of a COP9 complex (A, lanes 1, 2, and 3, labeled 0, 30, and 60, respectively) was added to 500 μg of sonicated exosomes isolated from 293 cells transfected with pGC93 vector treated for 5 days with doxytetracycline (1.3 μg/ml). One-third of the mixture was used for Western blot analysis of HSP70 (A, bottom panel), and the remainder used in a deubiquitin assay. A deubiquitin reaction was performed at 37°C for 0, 30, and 60 minutes. The reaction was stopped by addition of 1% SDS sample buffer and subsequently run in an immunoprecipitation assay using agarose beads conjugated to mouse anti-ubiquitin (A, top panel) or mouse IgG antibodies (middle panel), followed by HSP70 Western blot analysis. Five hundred micrograms of 293 cell exosomal protein, as described in (A) were used in a deubiquitin assay in the presence of 1 μg of recombinant COP9 (B, lane 1), 1 μg of GST-CSN5 (B, lane 2), or PBS as a control. One-third of the reacted proteins was used for Western blot analysis of HSP70 (B, bottom panel). The remainder was used for the deubiquitin reaction conducted at 37°C for 1 hour. The deubiquitinated samples were used for ubiquitin (B, top panel) or mouse IgG (B, middle panel) immunoprecipitation, and then Western-blotted and probed with an anti-HSP70 antibody. One representative experiment is presented, out of three independent experiments. The signal intensity of each protein analyzed in A and B was quantified using an Odyssey infrared imaging system (LI-COR). The ratios of signal intensity of ubiquitinated HSP70 at 30 minutes and 60 minutes against 0 minutes after the reaction was terminated were calculated and plotted (A, bar graph). The data are presented as means of three independent experiments. Using a similar method, the ratios of signal intensity of ubiquitinated HSP70 in the 60 minutes reaction containing either GST-CSN5 or COP9 against PBS as a control were calculated and plotted (B, bar graph). The data are presented as means of three independent experiments. **P < 0.0082 (2A); **P < 0.0057 (2B).

293 cells stably transfected with pLEYFP-CSN5-JAMMDN or pLEYFP-CSN5 were infected with vesicular stomatitis virus G pseudotyped HIV-1 NLENG1-ES-IRES. Forty-eight hours later, the cell culture supernatants were analyzed for de novo produced virus by measuring HIV-1 capsid protein using an ELISA for p24. The relative number of HIV-1 infected cells in each culture (JAMMDN mutant versus wild-type) was determined by flow cytometry, measuring NLENG1-ES-IRES+/GFP⁺ cells (A). The ELISA results are expressed as pg concentration of p24 released per GFP⁺ cell (B). Data are given as the average values ± SEM obtained for three samples in two independent experiments. ^**P < 0.001.