Tripartite motif 25 inhibits protein aggregate degradation during PRRSV infection by suppressing p62-mediated autophagy

Abstract

Viral infection causes endoplasmic reticulum stress and protein metabolism disorder, influencing protein aggregates formation or degradation that originate from misfolded proteins. The mechanism by which host proteins are involved in the above process remains largely unknown. The present study found that porcine reproductive and respiratory syndrome virus (PRRSV) infection promoted the degradation of intracellular ubiquitinated protein aggregates via activating autophagy. The host cell E3 ligase tripartite motif-containing (TRIM)25 promoted the recruitment and aggregation of polyubiquitinated proteins and impeded their degradation caused by PRRSV. TRIM25 interacted with ubiquitinated aggregates and was part of the aggregates complex. Next, the present study investigated the mechanisms by which TRIM25 inhibited the degradation of protein aggregates, and it was found that TRIM25 interacted with both Kelch-like ECH-associated protein 1 (KEAP1) and nuclear factor E2-related factor 2 (Nrf2), facilitated the nuclear translocation of Nrf2 by targeting KEAP1 for K48-linked ubiquitination and proteasome degradation, and activated Nrf2-mediated p62 expression. Further studies indicated that TRIM25 interacted with p62 and promoted its K63-linked ubiquitination via its E3 ligase activity and thus caused impairment of its oligomerization, aggregation, and recruitment for the autophagic protein LC3, leading to the suppression of autophagy activation. Besides, TRIM25 also suppressed the p62-mediated recruitment of ubiquitinated aggregates. Activation of autophagy decreased the accumulation of protein aggregates caused by TRIM25 overexpression, and inhibition of autophagy decreased the degradation of protein aggregates caused by TRIM25 knockdown. The current results also showed that TRIM25 inhibited PRRSV replication by inhibiting the KEAP1-Nrf2-p62 axis-mediated autophagy. Taken together, the present findings showed that the PRRSV replication restriction factor TRIM25 inhibited the degradation of ubiquitinated protein aggregates during viral infection by suppressing p62-mediated autophagy.IMPORTANCESequestration of protein aggregates and their subsequent degradation prevents proteostasis imbalance and cytotoxicity. The mechanisms controlling the turnover of protein aggregates during viral infection are mostly unknown. The present study found that porcine reproductive and respiratory syndrome virus (PRRSV) infection promoted the autophagic degradation of ubiquitinated protein aggregates, whereas tripartite motif-containing (TRIM)25 reversed this process. It was also found that TRIM25 promoted the expression of p62 by activating the Kelch-like ECH-associated protein 1 (KEAP1) and nuclear factor E2-related factor 2 (Nrf2) pathway and simultaneously prevented the oligomerization of p62 by promoting its K63-linked ubiquitination, thus suppressing its recruitment of the autophagic adaptor protein LC3 and ubiquitinated aggregates, leading to the inhibition of PRRSV-induced autophagy activation and the autophagic degradation of protein aggregates. The present study identified a new mechanism of protein aggregate turnover during viral infection and provided new insights for understanding the pathogenic mechanism of PRRSV.

Keywords: aggregates; autophagy; misfolded protein; porcine reproductive and respiratory syndrome virus; tripartite motif-containing 25.

Fig 1

PRRSV infection promotes ubiquitinated protein aggregates degradation in host cells. (A–C) Confocal analysis of the effect of PRRSV infection on aggregates formation. Marc-145 cells were mock-infected or infected with PRRSV (MOI = 0.1) or treated with MG132 (10 µM). At 36 hpi, cells were fixed, permeabilized, and followed by staining with rabbit anti-p62 polyclonal antibody (pAb) and Alexa 488-conjugated goat anti-rabbit IgG (H + L; green), and protein aggregates were detected using an Aggresome Detection Kit (red). Scale bar: 10 µm. (D–F) Effect of PRRSV infection on p62-Ub double positive aggregates formation. PRRSV mock-infected, infected (0.1 MOI), or MG132 treatment Marc-145 cells were fixed and permeabilized. Then cells were visualized using rabbit anti-Ub pAb and Alexa 488-conjugated goat anti-rabbit IgG (H + L; green), mouse anti-p62 pAb, and Alexa 594-conjugated goat anti-mouse IgG (H + L; red). Scale bar: 10 µm. (G and H) Effect of PRRSV infection on intracellular autophagy activity. (I) Effect of PRRSV infection on ubiquitinated protein aggregates formation. Marc-145 cells were infected or not with PRRSV (MOI = 0.1). At 36 hpi, cells were harvested and lysated. After centrifugation, the clear supernatants were used as soluble fraction. Cell debris was dissolved in 8 M urea and was used as insoluble fraction. (J) Effect of different doses of PRRSV infection on aggregates formation. (K) Effect of PRRSV infection on p62 recruitment of ubiquitinated aggregates and LC3. 3-MA treatment group was set as the control. Three independent experiments were performed, and the percentage or volume of aggregates in 20 independent fields of cells was counted and calculated each time, and typical data were shown. ^**, P < 0.01; ^***, P < 0.001.

Fig 2

TRIM25 negatively regulated ubiquitinated aggregates degradation during PRRSV infection. (A) Scheme showing the procedure of identifying ubiquitination-related targets in response to PRRSV infection. (B) Marc-145 cells were transfected with or without HA-Ubiquitin for 12 h, and the supernatants were isolated and followed by IP using anti-HA antibody. Silver staining was performed, and the bands in corresponding Coomassie blue staining gel were analyzed by nano LC-MS/MS. (C and D) Analysis of interaction between TRIM25 and ubiquitinated protein. HA-Ub, -K48-Ub, and -K63-Ub plasmids were co-transfected into HEK293T cells or not with Flag-TRIM25 plasmids. At 48 h post-transfection (hpt), cells were collected, and co-IP was conducted for analysis of TRIM25 interaction with ubiquitinated protein aggregates. (E and F) Effect of TRIM25 overexpression (E) or knockdown (F) on aggregates clearance in the context of PRRSV infection. Vector, Flag-TRIM25, shNC, and shTRIM25 cells were mock-infected or infected with PRRSV (0.1 MOI). At 36 hpi, cells were harvested, and the soluble or insoluble fractions were isolated and subjected to western blotting analysis. (G) Confocal analysis of ubiquitinated aggregates. Vector, Flag-TRIM25, shNC, and shTRIM25 cells were infected with PRRSV at an MOI of 0.1. At 36 hpi, cells were fixed, permeabilized, and analyzed using rabbit anti-Ub pAb and corresponding fluorescent antibody. Scale bars: 10 µm. (H and I) Protein aggregates numbers and area analysis of average of 20 fields of Fig. 2G. Three independent experiments were performed, and typical data were shown. ^*, P < 0.05; ^**, P < 0.01; ^***, P < 0.001. Scale bars: 10 µm. (J) Impact of PRRSV infection on the association of TRIM25 with protein aggregates. Flag-TRIM25 cells were left untreated, infected with a 0.1 MOI of PRRSV, or treated with 10 µM MG132. After infection or treatment for 24 h, cells were fixed, permeabilized, and detected using anti-flag monoclonal antibody (mAb), anti-Ub pAb, and corresponding fluorescent secondary antibody. Scale bars: 10 µm. (K) Co-localization analysis of TRIM25 with aggresome marker proteins. Flag-TRIM25 cells infecting with PRRSV (0.1 MOI) for 36 h were fixed, permeabilized, and detected with anti-Flag mAb, anti-p62, -HSP70 pAb, and corresponding fluorescent antibody. Scale bars: 10 µm. (L and M) Effect of TRIM25 on K48- and K63-linked ubiquitinated protein aggregates formation. Vector, Flag-TRIM25, shNC, and shTRIM25 cells were infected with PRRSV at an MOI of 0.1 for 36 h. The insoluble fractions were isolated and then subjected to western blotting analysis with anti-K48-Ub or -K63-Ub pAb.

Fig 3

TRIM25 interacts with and promotes p62 expression. (A) The interaction between endogenous TRIM25 and p62 was assayed by co-IP. (B) Co-IP analysis of the interaction between exogenous TRIM25 and exogenous p62. (C) Confocal analysis of endogenous or exogenous TRIM25 co-localization with exogenous p62. Scale bar: 10 µm. (D) Mapping the domain(s) of TRIM25 interacting with full-length of p62. (E) Co-localization analysis of TRIM25 mutants with endogenous p62. Marc-145 cells were transfected with Flag-TRIM25-WT, -ΔRING, -PS, and HA-TRIM25-ΔSPRY, respectively, for 36 h. Cells were then fixed, permeabilized, and stained with anti-flag, -HA mAb, anti-p62 pAb, and corresponding fluorescent second antibody. Scale bar: 10 µm. (F) Constructs used for co-IP between full-length of TRIM25 and deletion mutants of p62. (G) Total lysates of HEK293T cells transfected with both p62 deletion mutant plasmids and Flag-TRIM25 plasmid were immuno-precipitated with anti-TRIM25 and immunoblotted with anti-Myc, -GFP, and anti-TRIM25 pAb. (H–K) Effect of TRIM25 overexpression or knockdown on p62 expression. (L–O) Vector, Flag-TRIM25, shNC, and shTRIM25 cells were treated with 100 mg/mL of CHX for 0, 2, 4, 6, 8, and 10 h, respectively, and cells were harvested for analysis of p62 expression.

Fig 4

TRIM25 inhibits p62 oligomerization. (A) Myc-p62, HA-Ub, and Flag-TRIM25 or Flag-TRIM25-ΔRING plasmids were co-transfected into HEK293T cells, and the ubiquitination of p62 was analyzed by co-IP. (B) Co-IP analysis of K48- or K63-linked ubiquitination of p62 by TRIM25. (C) Vector, Flag-TRIM25, shNC, and shTRIM25 were cross-linked with 0.4 mg/mL dithiobis(succinimidylpropionate) at 4°C for 2 h, and the lysates were run under reducing or non-reducing conditions. (D and E) The key TRIM25 domain(s) affected the oligomerization of p62. Marc-145 cells were transfected with Flag-TRIM25-WT (500 ng/well), -ΔRING (500 ng/well), or -ΔSPRY (500 ng/well) mutant plasmids, respectively. At 12 hpt, the cells were infected with 0.1 MOI of PRRSV for 24 h. Cells were then fixed, permeabilized, and assayed for p62 oligomerization in transfection-positive cells. Scale bar: 10 µm. Data are mean ± SD values of three independent results. ^*, P < 0.05; ^**, P < 0.01; ^***, P < 0.001; ns, not significant.

Fig 5

TRIM25 inhibits p62 recruitment of autophagic effector protein LC3. (A) Endogenous p62 co-immunoprecipitated with GFP-LC3B from transfected Vector and Flag-TRIM25 cell extracts infected or not with PRRSV (0.1 MOI). GFP or GFP-LC3B were immunoprecipitated from total cellular extracts of GFP-LC3B transfected (2 µg/well) Vector and Flag-TRIM25 cells and subjected to SDS-PAGE. Immunoprecipitated p62 or GFP-LC3B was detected by anti-p62 or -GFP antibody. (B) Endogenous p62 co-immunoprecipitated with GFP-LC3B from transfected shNC and shTRIM25 cell extracts infected or not with PRRSV (0.1 MOI). (C) Vector, Flag-TRIM25, shNC, and shTRIM25 cells were transfected with reporter plasmid GFP-LC3B (500 ng/well). At 12 hpt, cells were inoculated with 0.1 MOI of PRRSV and fixed at 36 hpi. p62 was stained using anti-p62 mAb and corresponding fluorescent antibody. Co-localization between p62 and GFP-LC3B was analyzed by confocal. Scale bar: 10 µm.

Fig 6

TRIM25 inhibits PRRSV-induced autophagy activation. (A and B) The protein levels of LC-I and LC3-II in Vector and Flag-TRIM25 cells at 24 and 48 h were assayed by western blotting. Data are mean ± SD values of three independent results. ^***, P < 0.001. (C and D) Vector and Flag-TRIM25 cells were cultured for 48 h and then were fixed and assayed for the appearance of autophagosomes by staining with anti-LC3B mAb. ^***, P < 0.001. Scale bars: 20 µm. (E and F) The protein levels of LC-I and LC3-II of shNC and shTRIM25 cells at 24 and 48 h were detected using western blotting. (G and H) Immunofluorescence stain of LC3B in shNC and shTRIM25 cells (G), and LC3B puncta positive cell percentage was analyzed of 20 independent fields, scale bar: 10 µm (H). (I and J) Vector and Flag-TRIM25 cells were inoculated or not with 0.1 MOI of PRRSV for 36 h. Cells were collected to determine the expression level of LC3-I and LC3-II. (K and L) Vector and Flag-TRIM25 cells were mock-infected or infected with 0.1 MOI of PRRSV for 36 h. LC3B puncta were stained using anti-LC3B mAb. (M and N) shNC and shTRIM25 cells mock-infected or infected with PRRSV (0.1 MOI) were analyzed for the expression of LC3-I and LC3-II. (O and P) LC3B puncta in PRRSV mock-infected or infected cells were analyzed using confocal. Scale bar: 10 µm. (Q and R) Vector, Flag-TRIM25, shNC, and shTRIM25 cells were inoculated with 0.1 MOI of PRRSV and cultured for 36 h. The expression of ATG5 and Beclin1 was analyzed by western blotting. For statistical analysis of relevant fluorescence images, three independent experiments were performed, 20 independent fields of cells were calculated each time, and typical data were shown. Data are mean ± SD values of three independent results. ^*, P < 0.05; ^**, P < 0.01; ^***, P < 0.001.

Fig 7

TRIM25 inhibits p62 recruitment of ubiquitinated protein aggregates. (A–C) Effect of TRIM25 overexpression on p62 binding to aggregates. IP of endogenous ubiquitinated aggregates in Vector or Flag-TRIM25 cells with anti-p62 antibody, IP products of polyUb (A), poly-K48-Ub (B), and poly-K63-Ub (C) were detected with anti-HA and -p62 mAb. (D, F, and H) Immuno-staining of p62 and Ub-linked protein (D) and K48- and K63-Ub-linked protein (F and H) in Vector or Flag-TRIM25 cells mock-infected or infected with PRRSV (0.1 MOI). (E, G, and I) Quantification of p62-, p62-K48-, or p62-K63-Ub double positive foci corresponding to group D, F, and H. (J–L) Effect of TRIM25 knockdown on p62 binding to aggregates. IP of endogenous ubiquitinated aggregates using anti-p62 antibody in shNC and shTRIM25 cells infected with PRRSV. IP products of polyUb (J), poly-K48-Ub (K), and poly-K63-Ub (L) were detected using specific antibody against HA or p62. (M, O, and Q) Confocal analysis of endogenous p62 recruiting Ub-linked protein (M) and K48- and K63-Ub-linked ubiquitinated protein (O and Q). (N, P, and R) Quantification of p62-, p62-K48-, or p62-K63-Ub double positive foci corresponding to group M, O, and Q. For statistical analysis of relevant fluorescence images, three independent experiments were performed, the p62-K48/-K63-Ub area in 20 independent fields of cells was calculated each time, and typical data were shown. ^**, P < 0.05; ^***, P < 0.01; ^****, P < 0.001. Scale bar: 10 µm.

Fig 8

Inhibition of autophagy by TRIM25 promotes ubiquitinated aggregates accumulation. (A–C) Vector and Flag-TRIM25 cellls were infected with 0.1 MOI of PRRSV. At 24 hpi, cells were treated with dimethyl sulfoxide (DMSO) or 10 nM of PP242 for another 12 h. Soluble and insoluble fractions of cells were extracted and analyzed using western blotting. Part of the cell samples were fixed and assayed using confocal. ^***, P < 0.001. Scale bar: 10 µm. (D–F) Effect of autophagy inhibition on ubiquitinated aggregates clearance. shNC and shTRIM25 cells infected with 0.1 MOI of PRRSV were treated with 10 mM of 3-MA for 12 h, and then aggregates were analyzed using western blotting and confocal, respectively. ^***, P < 0.001. Scale bar: 10 µm. (H–J) Effect of Beclin1 knockdown on clearance of ubiquitinated aggregates. ^***, P < 0.001. Scale bar: 10 µm. (L–N) Effect of ATG5 knockdown on clearance of ubiquitinated aggregates. ^***, P < 0.001. Scale bar: 10 µm. (G and K) The protein level of Beclin1 and ATG5 was both assayed by western blotting after transfection with specific siRNA. (O) Confocal images of PRRSV-infected Marc-145 cells transfected with Flag-TRIM25, BFP-p62, and untreated or treated with PP242 (10 nM, 12 h) and/or 3-MA (10 mM, 12 h), and confocal was performed with antibodies as indicated. Right panel graphs: fluorescence intensity line is tracing corresponding to a white line in zoom panel. Scale bar: 10 µm. For statistical analysis of relevant fluorescence images, three independent experiments were performed, 20 independent fields of cells were calculated each time, and typical data were shown.

Fig 9

TRIM25 inhibits PRRSV replication. (A) Marc-145 cells infected with PRRSV or not were harvested at 24, 36, and 48 hpi, respectively, and then analyzed by western blotting. (B and C) Vector and Flag-TRIM25 cells were infected with 0.1 MOI of PRRSV. At 24, 36, 48, 60, and 72 hpi, cells and supernatants were collected to detect the expression of PRRSV N protein and progeny viral titers using western blotting and TCID₅₀ assay, respectively. (D and E) shNC and shTRIM25 cells infected with 0.1 MOI of PRRSV were collected at 24, 36, 48, 60, and 72 hpi to analyze the level of PRRSV N protein and progeny viral titers using western blotting and TCID₅₀ assay, respectively. (F and G) Flag-TRIM25 cells were transfected with 100 nM siNC or sip62 for 12 h and then infected with PRRSV (MOI = 0.1). Cells were collected at 36 hpi to detect the expression of RRSV N protein using western blotting; and for viral replication kinetics analysis, cell supernatants were collected at 36 hpi to detect progeny viral titers using TCID₅₀ assay. (H and I) Marc-145 cells infected with 0.1 MOI of PRRSV were treated with DMSO, 5 and 10 mM of 3-MA from 1 hpi and onwards. At 36 hpi, the expression of PRRSV N protein was analyzed using western blotting; for viral replication kinetics analysis, supernatants were collected at 36 hpi and were assayed using TCID₅₀. (J and K) Marc-145 cells were transfected with siNC or LC3 specific siRNA (50, 100 nM) for 12 h and then infected with PRRSV (MOI = 0.1). At 36 hpi, cells were collected, and the expression of PRRSV N protein was detected using western blotting; and for progeny viral titers analysis, supernatants were collected at 36 hpi and were analyzed using TCID₅₀ assay. Data are mean ± SD values of three independent results. ^*, P < 0.05; ^**, P < 0.01; ^***, P < 0.001; ns, not significant.