Fig 1.
PCV2 inhibits the activation of type I IFN signaling to facilitates the infection of other DNA viruses by reducing cGAMP production.
(A, B) The infection rate and morbidity of porcine parvovirus (PPV) or porcine pseudorabies virus (PRV) were compared between PCV2 positive pigs and PCV2 negative pigs. The native pig herds (n = 452) were separated into PCV2 positive group (n = 271) and PCV2 negative group (n = 181), then the infection rates (A) and the morbidity rates (B) of PPV or PRV were further analyzed in these two groups. (C) The PPV and PRV loads in native PCV2 positive pigs are higher than that in native PCV2 negative pigs. The viral copy numbers of PPV and PRV in the serum of native PCV2 positive pigs (n = 20 per group) and native PCV2 negative pigs (n = 17 per group) were measured by qPCR. * P < 0.05, ** P < 0.01 (compared with PCV2 negative pigs). (D) PPV and PRV replication levels in PCV2-infected cells are higher than that in mock infection cells. The PK-15 cells were infected with mock or PCV2 for 48 h, then were further infected with PPV or PRV, and the relative viral titers were measured by TCID50. * P < 0.05, ** P < 0.01(compared with mock infection). (E, F) PCV2 inhibits PPV- or PRV-induced IFN-β production and response. The piglets were infected by PCV1 (4×105 TCID50), PCV2 (4×105 TCID50), or mock (same volume of medium) for 1 week, respectively, and then challenged with 105 TCID50 PPV or 105 TCID50 PRV for another 24 h. The serum IFN-β of the infected piglets were measured by ELISA (E); IFN-β, IFIT1, and CXCL10 mRNA levels in lung tissues were determined by qPCR (F). * P < 0.05, ** P < 0.01 (compared with mock infection); # P < 0.05, ## P < 0.01 (compared with PCV1 infection). (G) Comparison of VSV-GFP replication in PK-15 cells pretreated with the cell supernatants from HT-DNA-stimulated mock-infected cells or HT-DNA-stimulated PCV2-infected cells. GFP positive cells were measured by flow cytometry. (H-K) PK-15 cells were infected with different doses (0.1, 1, 10 MOI) of PCV2 for 24 h (H, I), or infected with PCV2 (MOI = 1) for the indicated time (J, K), and then the relative cGAMP production levels (H, J), the levels of p-TBK1, p-IRF3, and nucleoprotein pIRF3 (pIRF3(N)) at 6 h following ISD stimulation or PRV infection were determined by report assay and western blotting, respectively. * P < 0.05, ** P < 0.01(compared with mock infection).
Fig 2.
PCV2 infection results in cGAS degradation through activation of autophagy-lysosome pathway.
(A) PCV2 infection induces porcine cGAS reduction. PK-15 cells were infected with PCV2 (MOI = 5) or mock for the indicated time, and then the protein and mRNA levels of porcine cGAS were determined by western blotting (upper panel) and RT-PCR (lower panel). (B) Ultraviolet-inactivated PCV2 induces porcine cGAS reduction. PK-15 cells were inoculated with UV-PCV2 (MOI = 5) or mock for the indicated time, and then the levels of porcine cGAS and PCV2 capsid were determined by western blotting. (C) Chloroquine (CQ) can effectively prevent the reduction of cGAS protein in PCV2-infected cells. PK-15 cells were pretreated with 10 μM MG132 (proteasome inhibitor) or 20 μM CQ (autophagy inhibitor) for 8 h, then infected with PCV2 (MOI = 5) for indicated times to detect cGAS levels. ** P < 0.01 (compared with infection at 0 h), # P < 0.05, ## P < 0.01 (compared with DMSO treatment in indicated same time). (D, E) knockdown of Atg5 or autophagic-sequestration inhibitor 3-MA treatment alleviates PCV2-induced cGAS degradation. PK-15 cells were transfected with Atg5 specific siRNA (siAtg5) or siRNA negative control (siN.C.) for 24 h, or treated with 3-MA (5 mM) for 8 h before PCV2 infection, and then the levels of porcine cGAS were determined at the indicated times by western blotting. * P < 0.05, ** P < 0.01 (compared with infection at 0 h), # P < 0.05, ## P < 0.01 (compared with siRNA negative control or DMSO treatment in indicated same time) (F, G) cGAS proteins are co-localized with LC3 in PCV2-infected cells. PK-15 cells transfected with GFP-LC3 were infected with PCV2 in the presence of DMSO, CQ or 3-MA along, followed by confocal observation (F, Left). The co-localization signals of targeted proteins were analyzed as intensity profiles of indicated proteins along the plotted lines by Image J line scan analysis (F, Right). The number of yellow puncta was counted and compared (G). Scale bar, 10 μm. ** P < 0.01 (compared with DMSO treatment). (H) PK-15 cells were pretreated with different doses (10, 20, 40 μM) of CQ for 8 h before UV-PCV2 inoculation, and then the levels of porcine cGAS, PCV2 Cap and p62 were determined at 48 hpi.
Fig 3.
PCV2 infection enhances the K48-linked poly-ubiquitination of porcine cGAS at K389 and facilitates the interaction of cGAS with p62 for autophagic degradation.
(A) PCV2 infection enhanced the ubiquitination of porcine cGAS. PK-15 cells were infected with PCV2 (MOI = 5) along with or without Baf for 48 h. Cell lysates were analyzed by immunoprecipitated with anti-porcine cGAS antibody, and ubiquitinated cGAS proteins were immunoblotted using anti-ubiquitin antibodies. (B, C) Porcine cGAS was mainly ubiquitinated with K48 linkage in PCV2-infected cells. PK-15 cells were transfected with different HA-Ub constructs as indicated, then infected with PCV2 in the presence or absence of Baf. Cell lysates were immunoprecipitated with anti-Flag antibody and immunoblotted with anti-HA antibody. (D) PK-15 cells were treated with EBSS together with or without PCV2 infection (MOI = 5) in the presence of BAF for 48 h. The poly-ubiquitination levels and protein levels of cGAS were analyzed. (E) The lysine 389 of porcine cGAS is required for its K48-ubiquitination induced by PCV2 infection. PK-15 cells were transfected with cGAS mutant expression vectors as indicated, then cells were infected with PCV2 (MOI = 5) in the presence of Baf for 48 h, the poly-ubiquitination levels of cGAS were analyzed. (F) Mutation of the lysine 389 of porcine cGAS can block poly-ubiquitination and degradation of porcine cGAS induced by PCV2. PK-15 cells were transfected with plasmids as indicated, then infected with PCV2 (MOI = 5) for the indicated time. Protein lysates were immunoprecipitated with anti-Flag antibody and immunoblotted with anti-Ub antibody. (G, H) Poly-ubiquitination of porcine cGAS at K389 is required for the interaction of cGAS with p62. The cGAS-/- PK-15 cells expressed Flag-cGAS, Flag-cGAS (K389R), then infected with Mock or PCV2 along with Baf (DMSO as a control). Cells were subjected to confocal assay to observe the co-localization of cGAS and p62. The co-localization signals of targeted proteins were analyzed as intensity profiles of indicated proteins along the plotted lines by Image J line scan analysis (G). Scale bar, 10 μm. Statistics of the puncta formation by cGAS-p62 in the indicated samples (H). ** P < 0.01 (compared with Mock or WT).
Fig 4.
PCV2 infection activates HDAC6 to mediate the transport of poly-ubiquitin cGAS to the lysosome via HDAC6 mediation.
(A) PK-15 cells were infected with PCV2 (MOI = 1) for the indicated times and the poly-ubiquitination levels and protein levels of cGAS were analyzed. (B) PK-15 cells were transfected with HDAC6 specific siRNA (siHDAC6) or siRNA negative control (siN.C.) for 24 h before PCV2 infection (MOI = 5), and then the levels of porcine cGAS, PCV2 Cap, HDAC6, Ac-tubulin were determined by western blotting at the indicated time post-infection. (C) PK-15 cells were transfected plasmids indicated, then infected with PCV2 (MOI = 5) in the presence of Baf for 48 h. the interaction of ubiquitinated cGAS with HDAC6 and p62 was analyzed. (D) The cGAS-/- PK-15 cells transfected with Flag-cGAS, Flag-cGAS (K389R) expression constructs were infected with PCV2 in the presence of Baf. The colocalization of porcine cGAS and HDAC6 were observed under confocal microscopy. Scale bar, 10 μm. The co-localization signals of targeted proteins were analyzed as intensity profiles of indicated proteins along the plotted lines by Image J line scan analysis. (E) PK-15 cells were transfected expression vectors as indicated, then infected with PCV2 in the presence of Baf for 48 h. Subsequently, the ubiquitination levels of cGAS and cGAS (K389R), and their ability to bind HDAC6 were measured. (F) PK-15 cells were transfected with HDAC6 specific siRNA (siHDAC6) or siRNA negative control (siN.C.) and indicated plasmids for 24 h. Then these cells were infected with PCV2 (MOI = 5) or mock in the presence of Baf for 48 h; the interaction of ubiquitinated cGAS with P62 was analyzed. (G) PK-15 cells were transfected with HDAC6 specific siRNA (siHDAC6) or siRNA negative control (siN.C.) and indicated plasmids, meanwhile, Myc-HDAC6 (FL), Myc-HDAC6 (ΔUBD) were reconstituted in the cells transfected siHDAC6. Then, these cells were infected with PCV2 (MOI = 5) for 48 h, and the interaction of ubiquitinated cGAS with p62 was analyzed in these cells. (H) The HDAC6-/- PK-15 cells were transfected wild-type HDAC6 (WT) or mutated HDAC6-3M (D538A/D608A/S609A) for 24 h, then infected with PCV2 (MOI = 5) for another 48 h, and the interaction of ubiquitinated cGAS with p62 was analyzed. (I) The HDAC6-/- PK-15 cells transfected wild-type HDAC6 (WT) or mutated HDAC6-3M for 24 h, then infected with PCV2 (MOI = 5) for the indicated time, and then the levels of porcine cGAS, PCV2 Cap, and Ac-Tubulin were determined by western blotting. * P < 0.05, ** P < 0.01 (compared with infection at 0 h), # P < 0.05, ## P < 0.01 (compared with wild type HDAC6 group in indicated same time).
Fig 5.
PCV2 induces phosphorylation of porcine cGAS at Ser278 to negatively regulate cGAS enzymatic activity.
(A, B) Suppression of autophagic flux cannot effectively improve the induction of cGAMP and IFN-β in PCV2-infected cells. PK-15 cells pretreated with 3-MA or CQ were infected with PCV2 (MOI = 5) for 48 h, and then the relative cGAMP production levels and IFN-β mRNA levels at 6 h following ISD stimulation or PRV infection were determined by report assay and qPCR respectively (A). ** P < 0.01 (compared with mock infection). The levels of porcine cGAS and PCV2 capsid were determined by western blotting (B). (C) The cGAS-/- PK-15 cells were reconstituted with the wild type cGAS, mutant cGAS (K389R), respectively, then infected with PCV2 (MOI = 5) for 48 h to detect the relative cGAMP and IFN-β mRNA production levels at 6 h following ISD stimulation or PRV infection. (D) PK-15 cells pretreated with Tub A (DMSO as control) were infected with PCV2 (MOI = 5) for 48 h, then the relative cGAMP and IFN-β mRNA production levels were determined at 6 h following ISD stimulation or PRV infection. (E) Sequence comparison of the cGAS phosphorylation site from the indicated species (upper panel). The potential phosphorylated peptides in the NTase motif of porcine cGAS are highlighted in green and red (lower panel). (F) PK-15 cells were infected with mock or PCV2 for indicated times, and the phosphorylation level of cGAS at the S278 site was detected by western blotting using a rabbit anti-Phospho-Akt Substrate monoclonal antibody that can recognize the motif (RRGS*278) of porcine cGAS. (G) The phosphorylation of porcine cGAS at Ser278 exhibits a weakened enzymatic activity in in vitro assay. LC-MS analysis of cGAMP production from an in vitro cGAMP synthesis assay. Small molecules were extracted from in vitro tubes which contained the same doses of wild-type cGAS, phosphorylation-resistant S278A mutant, and phosphomimetic S278D mutant for analysis of cGAMP isomers by tandem mass spectrometry. (H) The cGAS-/- PK-15 cells reconstituted with the WT cGAS, or cGAS mutant S278A, or cGAS mutant S278D were infected with mock or PCV2 in the presence of Baf for 12 h, and then the relative cGAMP production levels were determined at 6 h following ISD stimulation or PRV infection. * P < 0.05, ** P < 0.01 (compared with mock infection). (I) The cGAS-/- PK-15 cells reconstituted with the WT cGAS, or cGAS mutant S278A were infected with mock or PCV2 for 12 h, and then the phosphorylation level of cGAS at the S278 site was detected by Immunoprecipitation.
Fig 6.
The phosphorylation of porcine cGAS at Ser278 facilitates the K48-linked poly-ubiquitination and degradation of porcine cGAS during PCV2 infection.
(A-B) Akt signaling negatively regulates cGAS-mediated cGAMP production in the PCV2-infected cells via phosphorylation of porcine cGAS at S278. PK-15 cells pretreated with indicated inhibitors were infected with PCV2 (MOI = 5) for 12 h, and then the relative cGAMP production levels were determined at 6 h following ISD stimulation or PRV infection (A); the phosphorylation level of cGAS at the S278 site was detected by western blotting (B). * P < 0.05, ** P < 0.01 (compared with DMSO/mock infection). (C, D) The phosphorylation of cGAS facilitates the ubiquitination and degradation of cGAS during PCV2 infection. PK-15 cells transfected with indicated expression constructs were treated with indicated inhibitors and then infected with PCV2 (MOI = 5) in the presence or absence of Baf to detect the poly-ubiquitination levels and protein levels of cGAS. (E) The cGAS-/- PK-15 cells were reconstituted with the WT cGAS, or S278A mutant, or S278D mutant, then infected with PCV2 (MOI = 5) to detect the poly-ubiquitination levels and phosphorylation levels of cGAS in these cells. (F) The cGAS-/- PK-15 cells were reconstituted with WT cGAS, or S278A mutant, then infected with PCV2 (MOI = 5) or mock to detect the interaction of ubiquitinated cGAS with HDAC6. (G) The wild-type PK-15 cells, and cGAS-/- PK-15 cells were reconstituted with WT cGAS or S278A mutant, then infected with PCV2 to observe the colocalization of cGAS and p62. The co-localization signals of targeted proteins were analyzed as intensity profiles of indicated proteins along the plotted lines by Image J line scan analysis. Scale bar, 10 μm.
Fig 7.
PCV2 Cap plays a predominant role in promoting porcine cGAS phosphorylation and HDAC6 activation depending on gC1qR and/or PKCδ.
(A) Model of construction of PCV mutants. (B-D) Cap is a critical regulator of cGAS phosphorylation. PK-15 cells were infected with PCV mutants (MOI = 5) for 12 h, and then the phosphorylation level of cGAS at the S278 site was detected by western blotting (B). PK-15 cells were infected with rAd-Blank (MOI = 100), rAd-Rep (MOI = 100) and rAd-Cap (MOI = 100) for 24 h, then the phosphorylation (C) and the poly-ubiquitination levels (D) of cGAS were detected by immunoprecipitation. (E) gC1qR-/- PK-15 cells and wild type PK-15 cells were infected with PCV2 (MOI = 5) for 12 h, then the phosphorylation of cGAS was detected by immunoprecipitation. (F) Cap is a critical regulator of HDAC6 activation. PK-15 cells were inoculated with PCV mutants (MOI = 5) for the indicated time, and then the levels of HDAC6 and Ac-tubulin were determined by western blotting. (G) gC1qR-/- PK-15 cells and wild type PK-15 cells were infected with PCV2 (MOI = 5) for 48 h, then the levels of Ac-Tubulin detected by western blotting. (H) Interaction of PKCδ with HDAC6. PK-15 cells were transfected with Myc-HDAC6 and Flag-PKCδ for 24 h. Then these cells were infected with PCV2 (MOI = 5) for indicated time; the interaction of HDAC6 with PKCδ was analyzed. (I) PK-15 cells were pretreated with Rottlerin and infected with PCV2 (MOI = 5) for the indicated time, and then the levels of Ac-Tubulin were determined by western blotting. (J) PK-15 cells were transfected with PKCδ specific siRNA (siPKCδ) or siRNA negative control (siN.C.) for 24 h and then infected with PCV2 (MOI = 5) for the indicated time, followed by western blotting detection of the levels of Ac-Tubulin. (K, L) PK-15 cells were pretreated with Rottlerin (K), or transfected with PKCδ specific siRNA (siPKCδ) or siRNA negative control (siN.C.) (L), and then infected with rAd-Cap (100 MOI) for 24 h, followed by western blotting detection of the levels of Ac-Tubulin.
Fig 8.
gC1qR-binding activity deficient PCV2 mutant (PCV2RmA) shows a weakened inhibitory effect on IFN-β induction and a weaker boost effect for other DNA viruses.
(A) PCV2RmA exhibits a lower inhibitory effect on other DNA-induced cGAMP compared with wild type PCV2. PK-15 cells were infected with wild type PCV2 (MOI = 5) or PCV2RmA (MOI = 5) for the indicated time, and then the relative cGAMP production levels were determined at 6 h following ISD stimulation or PRV infection. * P < 0.05, ** P < 0.01 (compared with PCV2 infection). (B, C) PK-15 cells were transfected with HDAC6 specific siRNA (siHDAC6) or control siRNA (siNC) for 24 h, then treated with Akt inhibitor or DMSO for 6 h, followed by PCV2 (MOI = 5) or PCV2RmA (MOI = 5) infection, and then the IFN-β mRNA levels or cGAMP production were determined at 6 h following PRV or PPV infection. * P < 0.05, ** P < 0.01 (compared with PCV2 infection in indicated same condition). # P < 0.05 (compared with the cells that transfected siNC and treated with DMSO in same infection). (D) PK-15 cells were infected with PCV2 at 0.5~5 MOI or PCV2RmA at 5 MOI, and progeny virion production was determined by TCID50 at 48 h post-infection. # P < 0.05, ## P < 0.01 (compared with 5 MOI PCV2RmA-infected PK-15 cells). (E) PK-15 cells were infected with wild type PCV2 (MOI = 1) or PCV2RmA (MOI = 5) for the indicated time, and then the relative cGAMP production levels and IFN-β mRNA levels were determined at 6 h following ISD stimulation or PRV and PPV infection. * P < 0.05, ** P < 0.01 (compared with PCV2 infection). (F-H) PCV2RmA exhibits a lower induction capability in the phosphorylation, poly-ubiquitination, and degradation of porcine cGAS compared with wild type PCV2. PK-15 cells were infected with wild type PCV2 (MOI = 1) or PCV2RmA (MOI = 5) for 12 h, the phosphorylation of cGAS was detected by immunoprecipitation (F), PK-15 cells were infected with wild type PCV2 (MOI = 1) or PCV2RmA (MOI = 5) for 48 h, then poly-ubiquitination (G), and protein levels of cGAS and Ac-tubulin levels (H) were detected by immunoprecipitation. (I-K) PCV2RmA is a weak strain relative to PCV2 in promotion of DNA virus replication. PK-15 cells were infected with PCV2 (MOI = 1) or PCV2RmA (MOI = 5) for 48 h, then were further infected with PPV or PRV, and the relative viral titers were measured by TCID50 (I). * P < 0.05, ** P < 0.01. (J) Comparison of VSV-GFP replication in PK-15 cells pretreated with cell supernatants from HT-DNA-stimulated mock-infected cells, HT-DNA-stimulated PCV2-infected cells, or HT-DNA-stimulated PCV2RmA-infected cells. GFP positive cells were measured by flow cytometry. (K) IFN-β levels were measured by ELISA. ** P < 0.01. (L-N) PCV2RmA alleviating PPV- or PRV-induced pathological changes. The piglets (n = 3 per group) were infected by PCV2 (4×105 TCID50), PCV2RmA (2×106 TCID50) for 1 week, respectively, and then challenged with 105 TCID50 PPV or 105 TCID50 PRV for another one week. (L) The serum IFN-β of the infected piglets were measured by ELISA at 24 h post-2nd infection. * P < 0.05, ** P < 0.01 (compared with PCV2 infected pigs). (M) The viral load of porcine parvovirus (PPV) or porcine pseudorabies virus (PRV) were compared between PCV2 infected pigs and PCV2RmA infected pigs by qPCR. The PPV and PRV loads in the serum of PCV2-infected pigs are higher than that in the serum of PCV2RmA-infected pigs. * P < 0.05, ** P < 0.01 (compared with PCV2 infected pigs). (N) The histological lesion scores of infection piglets (PCV2 and PCV2RmA).
Fig 9.
Model of PCV2 targeting cGAS to suppress Type I interferon production by gC1qR-mediated catalytic activity inhibition and HDAC6-mediated autophagic degradation.
In the early phase of PCV2-infection, porcine cGAS is phosphorylated at S278 site via Cap binding protein gC1qR-mediated PI3K/AKT signaling activation, which directly silences the catalytic activity of cGAS to inhibit cGAMP production; Subsequently, phosphorylation of cGAS at S278 facilitates the K48-linked poly-ubiquitination of cGAS at K389, while PKCδ is phosphorylated and recruited by gC1qR to promote HDAC6 activation; then K48-ubiquitinated-cGAS proteins are recruited and transported from the cytosol to autolysosome by activated histone deacetylase 6 (HDAC6) depending on its deacetylase activity and ubiquitin-binding function, thereby eventually resulting in a markedly increased cGAS degradation in PCV2 infection-induced autophagic cells.