Fig 1.
Recruitment of ATG8 by the TBSV p33 and the CIRV p36 replication proteins into VROs in N. benthamiana.
(A) Confocal microscopy images show efficient co-localization of TBSV p33-BFP replication protein and GFP-ATG8f within VROs consisting of clustered peroxisomes, marked by RFP-SKL peroxisomal matrix marker in N. benthamiana leaves. The expression of these proteins, driven by the 35S promoter, was achieved through co-agroinfiltration into N. benthamiana leaves. The plant leaves were either TBSV-infected and applied darkness treatment (to induce bulk autophagy) or mock-inoculated as shown. The VROs are marked with arrows. The enlarged images of VROs (boxed) are shown on the right. Scale bars represent 10 μm. The graph shows the number of ATG8f puncta / VRO. Error bars represent SD (n = 10). Student t-test was used to statistically analyze the data (****P < 0.0001). (B) Confocal microscopy images show efficient co-localization of CIRV p36-BFP replication protein and GFP-ATG8f within VROs consisting of clustered mitochondria, marked by RFP-AtTim21 mitochondrial marker in N. benthamiana leaves. See further details in panel A. (C) Confocal microscopy images show co-localization of TBSV p33-BFP replication protein and GFP-ATG8a within VROs in N. benthamiana leaves. Note that ATG8a localized in both the cytosol and nucleus in the absence of tombusvirus infection. See further details in panel A. Each experiment was repeated.
Fig 2.
Interaction between ATG8 and tombusvirus replication proteins within VROs in N. benthamiana.
(A) Interaction between TBSV p33-cYFP replication protein and the nYFP-ATG8f protein was detected by BiFC. The merged images show the co-localization of RFP-SKL with the BiFC signal, indicating that the interaction between p33 replication protein and ATG8f occurs in VROs in clustered peroxisomal membranes. The VROs are marked with arrows. Scale bars represent 10 μm. Each experiment was repeated three times. (B) Interaction between CIRV p36-cYFP replication protein and the nYFP-ATG8f protein were detected by BiFC. The merged images show the co-localization of RFP-AtTim21 with the BiFC signal, indicating that the interaction between p36 replication protein and ATG8f occurs in VROs consisting of aggregated mitochondria. See further details in panel A. (C) Interactions between TBSV p33-cYFP or CIRV p36-cYFP replication proteins and the nYFP-ATG8a protein were detected by BiFC. See further details in panel A. (D) Protein proximity-labeling with biotin in N. benthamiana. N. benthamiana leaves were agroinfiltrated to express p33 replication protein, which was fused to BirA biotin ligase, and Avi-tagged ATG8f. Note that both fusion proteins carry a 6xHis tag. Biotin treatment lasted for 40 min. The image shows the western blot analysis of the biotinylated Avi-ATG8f detected with streptavidin-conjugated AP in total protein extracts. p33-myc was used as a negative control to show plant does not have biotin ligase activity. GFP-BirA was expressed (lanes 8–10) to measure basal level of biotinylation of Avi-ATG8f. The experiments were repeated. (E) Co-purification of YFP-ATG8f with TBSV Flag-p33 replication protein from subcellular membranes. N. benthamiana leaves were agroinfiltrated to express YFP-ATG8f with TBSV Flag-p33. Top two panels: western blot analysis of co-purified YFP-ATG8f detected with anti-YFP antibody, while Flag-p33 was detected with anti-Flag antibody. The negative control was from plants expressing YFP purified on a Flag-affinity column. Bottom two panels: western blot of input YFP-ATG8f, YFP and Flag-p33 in the total protein extracts. (F) Pull-down assay including His6-NbATG8a and the MBP-tagged TBSV p33 or CIRV p36 replication proteins. Note that we used the soluble C-terminal region of TBSV p33 or CIRV p36 replication proteins, which lacked the N-terminal domain. Top panel: western blot analysis of the captured His6-NbATG8a with the MBP-affinity purified p33C/p36C was performed with anti-His antibody. The negative control was the MBP (lanes 1). Middle panel: Coomassie-blue stained SDS-PAGE of the captured MBP-p33C, MBP-p36C and MBP. Bottom panels: western blot analysis of His6-NbATG8a in total extracts. Coomassie-blue stained SDS-PAGE of the MBP-p33C, MBP-p36C and MBP in total extracts. Each experiment was repeated three times.
Fig 3.
The effect of ATG8 on tombusvirus replication in N. benthamiana plants.
(A-B) Top panel: The accumulation of the CNV and TBSV RNAs in ATG8f-silenced (ATG8f KD) N. benthamiana plants 2.5 and 2 dpi, respectively. The genomic (g)RNA levels in the inoculated leaves were measured by northern blot analysis. Agroinfiltration of pGD-CNV20Kstop or inoculation with TBSV sap was done 10 days after silencing of ATG8f expression. Agroinfiltration of tobacco rattle virus (TRV) vectors carrying either ATG8f or 3′-terminal GFP (as a control) sequences was used to induce VIGS. Second panel: RT-PCR analysis of ATG8f mRNA level in the silenced and control plants. Third panel: RT-PCR analysis of tubulin mRNA level in the silenced and control plants. (C) Top panel: The accumulation of the CIRV RNAs in ATG8f-silenced (ATG8f KD) N. benthamiana leaves at 3 dpi. See more details in panel A. Bottom panel: Ribosomal (r)RNA is shown as a loading control in an ethidium-bromide-stained agarose gel. (D) The accumulation of the TBSV RNAs in ATG8a-silenced (ATG8a KD) N. benthamiana leaves at 2 dpi. See more details in panel A. (E) Real time RT-qPCR analysis of the expression of ATG8f mRNA in the TBSV-inoculated leaves (2 dpi) or systemically infected leaves (5 dpi) of N. benthamiana plants. The same amounts of total RNA extracts were used in RT-qPCR analysis. (F) Accumulation of the TBSV RNAs in ATG5-silenced (ATG5 KD) N. benthamiana leaves at 2 dpi. See more details in panel A. (G) The effect of Legionella RavZ effector on autophagy flux. The plants expressed GFP-ATG8f and either exposed to darkness to induce bulk autophagy or not as shown. The released ‘free’ GFP band is marked with an arrowhead. The left two lanes include samples from plants co-expressing RavZ and GFP-ATG8f, which show lack of released ‘free’ GFP band. Total protein extracts were immunoblotted with anti-GFP antibody. The RavZ and GFP-ATG8f proteins were expressed via agroinfiltration. (H) The effect of Legionella RavZ effector on ATG8f-PE conjugation. The two lanes on the right display plants expressing Flag-ATG8f, where both Flag-ATG8f and Flag-ATG8f-PE are present as marked by arrowhead. The left two lanes show that RavZ expression eliminated the Flag-ATG8f-PE band, but not the Flag-ATG8f band. Total protein extracts were loaded on 15% polyacrylamide gels containing 6 M urea and immunoblotted with anti-Flag antibody. (I) Top panel: The accumulation of the TBSV RNAs in N. benthamiana leaves expressing RavZ at 2 dpi. The RavZ protein was expressed via agroinfiltration. The control leaves were agroinfiltrated with the same amounts of bacteria with ‘empty’ plasmid. See more details in panel A. Bottom panel: Ribosomal RNA is shown as a loading control in an ethidium-bromide-stained agarose gel. Each experiment was repeated three times.
Fig 4.
Recruitment of NBR1 selective autophagy receptor by the TBSV p33 and the CIRV p36 replication proteins into VROs in N. benthamiana.
(A) Confocal microscopy images show co-localization of TBSV p33-BFP replication protein and the eGFP-NBR1 within a VRO consisting of clustered peroxisomes, marked by RFP-SKL peroxisomal matrix marker in N. benthamiana leaves. The expression of these proteins, driven by the 35S promoter, was achieved through co-agroinfiltration into N. benthamiana leaves. The plant leaves were either TBSV-infected, or mock-inoculated as shown. Note that eGFP-NBR1 formed different patterns as visible in the enlarged images on the right. Scale bars represent 10 μm. See further details in Fig 1A. (B) Confocal microscopy images show co-localization of CIRV p36-BFP replication protein and the eGFP-NBR1 within a VRO consisting of clustered mitochondria, marked by RFP-AtTim21 mitochondrial marker in N. benthamiana leaves. See further details in panel A. (C) Interaction between TBSV p33-cYFP replication protein and the nYFP-NBR1 protein was detected by BiFC. The merged images show the co-localization of RFP-SKL with the BiFC signals, indicating that the interaction between p33 replication protein and NBR1 occurs in VROs. The interacting proteins formed different patterns as visible in the enlarged images on the right. (D) Interaction between CIRV p36-cYFP replication protein and the nYFP-NBR1 protein was detected by BiFC. See further details in panel C above. Each experiment was repeated.
Fig 5.
Interaction between the co-opted ATG8f and NBR1 proteins within VROs in N. benthamiana.
(A) Confocal microscopy images show co-localization of TBSV p33-BFP replication protein and the co-opted RFP-ATG8f and eGFP-NBR1 within a VRO consisting of clustered peroxisomes in N. benthamiana leaves. The expression of these proteins, driven by the 35S promoter, was achieved through co-agroinfiltration into N. benthamiana leaves. The VROs are marked with arrows. Scale bars represent 10 μm. (B) BiFC approach was employed to demonstrate interaction between nYFP-NBR1 and cYFP-ATG8f proteins associated with the TBSV p33-BFP-positive VROs. Co-agroinfiltration into N. benthamiana leaves was used for protein expression. ‘EV’ represents plant cells not expressing p33-BFP. Scale bars represent 10 μm. Each experiment was repeated.
Fig 6.
Separate subversion of ATG8f or NBR1 by TBSV p33 replication protein into VROs.
(A) Confocal microscopy images show co-localization of eGFP-NBR1 and p33-RFP (top two panels), or eGFP-NBR1 and RFP-SKL co-expressing cYFP-p33 (bottom panels), within VROs in ATG8f-silenced (ATG8f KD) or control N. benthamiana cells. The expression of these proteins, driven by the 35S promoter, was achieved through co-agroinfiltration into N. benthamiana leaves. TRV vector carrying MBP sequences was used as a VIGS control. Scale bars represent 10 μm. The VROs are marked with arrows. (B) Confocal microscopy images show co-localization of RFP-NBR1 and p33-BFP in N. benthamiana cells. The plants either expressed GFP-RavZ effector (top panel), or pGD vector as control (bottom panel). See further details in panel A. (C) Confocal microscopy images show co-localization of GFP-ATG8f and p33-BFP in NBR1-silenced (NBR1 KD) N. benthamiana cells. See more detail in panel A. Scale bars represent 10 μm. Each experiment was repeated.
Fig 7.
The effect of NBR1 on tombusvirus replication in N. benthamiana plants.
(A) Top panel: The accumulation of TBSV RNAs in NBR1-silenced (NBR1 KD) inoculated leaves of N. benthamiana at 2 dpi was measured by northern blot analysis. Sap inoculation with TBSV was done 10 days after silencing of NBR1 expression. Agroinfiltration of TRV vector carrying NBR1 or 3’-terminal GFP (as a control) sequences was used to induce VIGS. Second panel: Ribosomal RNA is shown as a loading control in an ethidium-bromide stained agarose gel. Third panel: RT-PCR analysis of NBR1 mRNA level in the silenced and control plants. Fourth panel: RT-PCR analysis of tubulin mRNA level in the silenced and control plants. Each experiment was repeated three times. (B) The accumulation of CIRV RNAs in NBR1-silenced (NBR1 KD) inoculated leaves of N. benthamiana at 2.5 dpi was measured by northern blot analysis. Sap inoculation with CIRV was done 10 days after silencing of NBR1 expression. See further details in panel A above. (C) Real time RT-qPCR analysis of the induction of NBR1 mRNA expression in the TBSV-inoculated leaves (2 dpi) of N. benthamiana plants. The same amounts of total RNA extracts were used in RT-qPCR analysis. Each experiment was repeated three times.
Fig 8.
Contributions of ATG8f and NBR1 to PE enrichment in the viral replication compartment in N. benthamiana protoplasts.
(A) Top panel: Confocal microscopy images reveal poor PE enrichment in VROs marked with p33-RFP in protoplasts prepared from ATG8f-silenced (ATG8f-KD) N. benthamiana (top image). Central panel: PE enrichment is visible in VROs in control N. benthamiana protoplasts. Differential interference contrast (DIC) images are shown on the right. PE distribution is detected by a staining probe using biotinylated duramycin peptide and streptavidin conjugated with Alexa Fluor 405. Bottom panel: Confocal microscopy images show poor PE re-distribution into VROs in N. benthamiana protoplasts co-expressing RFP-p33 and RavZ. Scale bars represent 10 μm. The fluorescence intensity of Duramycin was quantified within the VROs marked with arrows using Image J. Error bars represent SD (n = 10). Student t-test was used for statistical analysis (***P < 0.001, ****P < 0.0001). (B) Confocal microscopy images show poor PE re-distribution into VROs marked with p33-RFP in NBR1-silenced (NBR1-KD) N. benthamiana protoplasts (top image). TRV vector carrying partial GFP sequences was used as a VIGS control. See more details in panel A above. Each experiment was repeated three times.
Fig 9.
ATG8 and NBR1 contribute to PI(3)P enrichment within the viral replication compartment in N. benthamiana plants and protoplasts.
(A) Confocal microscopy images reveal poor PI(3)P enrichment in VROs marked with p33-BFP in ATG8f-silenced (ATG8f-KD) N. benthamiana protoplasts. PI(3)P enrichment is visible in VROs in control N. benthamiana protoplasts. PI(3)P distribution is detected by PI(3)P antibody and then incubated with anti-mouse secondary antibody conjugated with Alexa Fluor 568. Scale bars represent 10 μm. The fluorescence intensity for PI(3)P was quantified within the VROs marked with arrows using Image J. Error bars represent SD (n = 10). Student t-test was used for statistical analysis (***P < 0.001, ****P < 0.0001). (B) Confocal microscopy images show reduced enrichment of PI(3)P in VROs in N. benthamiana leaves co-expressing RFP-2xFYVE protein, p33-BFP and RavZ. Note that RFP-2xFYVE selectively binds to PI(3)P. PI(3)P enrichment is visible in VROs in control N. benthamiana cells. Scale bars represent 10 μm. See more details in panel A above. (C) Confocal microscopy images show PI(3)P distribution in NBR1-silenced (NBR1-KD) or TRV2-cGFP control (bottom image) N. benthamiana protoplasts expressing p33-BFP. See more details in panel A above.
Fig 10.
ATG8f promotes the enrichment of VPS34 PI3 kinase within VROs in N. benthamiana plants and yeast.
(A) Confocal microscopy images show co-localization of TBSV p33-BFP replication protein and the AtVPS34-GFP, the PI(3) kinase, in ATG8f-silenced (ATG8f-KD) (top image) or TRV2-MBP control (bottom image) N. benthamiana cells. (B) Quantitative GFP fluorescence intensity values were measured for ~50 samples to calculate relative AtVPS34-GFP levels in VROs. The statistical analysis was performed using a t-test, and the results showed a significant difference between the two groups (p < 0.05). Each experiment was repeated. (C) BiFC-based assay was used to measure the effect of ATG8f knockdown on enrichment of VPS34 in VROs. N. benthamiana plants expressed TBSV p33-cYFP and AtVps34-nYFP in ATG8f-silenced (ATG8f-KD) (left panels) or TRV2-MBP control (right panels). (D) YFP fluorescence intensity values were quantified using Image J. YFP fluorescence intensity in control was arbitrarily set as 1. Error bars represent SD (n = 60). Student t-test was used for statistical analysis of data (****P < 0.0001). (E) Co-purification of HA-VPS34 with Flag-p33 from either atg8Δ or WT (BY4741) yeasts. The membrane fraction of yeasts was detergent-solubilized and Flag-p33 was purified on the anti-Flag column, followed by elution. Western blotting was used to detect the co-purified HA-VPS34 using anti-HA antibody. Bottom panels: western blot of input HA-VPS34 in the total yeast extracts.
Fig 11.
TBSV replication protein inhibits autophagic flux in N. benthamiana.
(A) Agroinfiltrated N. benthamiana plants expressing GFP-p33 were exposed to darkness for 16 h to induce bulk autophagy (lanes 3 and 4) or kept under normal condition (lane 2). Total protein extracts were probed using western blot with anti-GFP antibody. N. benthamiana plants expressing GFP-ATG8f were used as a control (lane 1). The released ‘free’ GFP is marked with an arrowhead. (B) TBSV p33-Flag was expressed in N. benthamiana plants via agroinfiltration. 2.5 days latter, the agroinfiltrated leaves were treated with either 1 μM ConA or 100 μM E64d to inhibit autophagic degradation and samples were taken after 16 h. Western blotting was done with anti-Flag antibody. (C) Expression of GFP-ATG8f in N. benthamiana infected with CNV20kstop (lanes 4–6) or pGD vector as the control (lanes 1–3) was done through agroinfiltration. At 1.5 dpi, two plants from each group were exposed to darkness for 16 h. Total protein extracts were probed using western blot with anti-GFP antibody. The released ‘free’ GFP is marked with an arrowhead. Autophagic flux was measured based on the ratio of GFP/GFP-ATG8f using Image J software. (D) Expression of GFP-ATG8f in N. benthamiana infected with TBSV (lanes 5–8) or pGD vector as the control (lanes 1–4) was done through agroinfiltration. The same leaves were infiltrated with 10 μM AZD8055 to induce bulk autophagy at 1.5 dpi or with 10 μM DMSO. Total protein extracts were obtained 8 h later and probed using western blot with anti-GFP antibody. The released ‘free’ GFP is marked with an arrowhead. Autophagic flux was measured based on the ratio of GFP/GFP-ATG8f using Image J software. (E) Expression of GFP-ATG8f (i) with the pGD vector as the control (lanes 1–2), (ii) with TBSV p33/p92/repRNA combination (lanes 3–6), (iii) with the TBSV coat protein (CP) (lanes 7–10) or (iv) with the movement protein (MP) (lanes 11–14) in N. benthamiana leaves was done through agroinfiltration. At 1.5 dpi all plants underwent 16 h darkness treatment. See more details in panel B above. The statistical analysis was performed using an ANOVA-test, and the results showed a significant difference between the control and ‘p33/p92/repRNA combination group’ (p < 0.05), while no significant difference between the control and CP or MP. (F) Expression of Flag-ATG8f (i) with the pGD vector as the control (lanes 4–6), (ii) with TBSV (lanes 1–3) (iii) with p33/p92/repRNA combination (lanes 7–9), or (iv) with TBSV p33 in N. benthamiana leaves was done through agroinfiltration. At 2 dpai, total protein extracts were obtained and probed using western blot with anti-Flag antibody. Autophagic flux was measured based on the ratio of ATG8f-PE/ATG8f using Image J software. The statistical analysis was performed using an ANOVA-test. ** (p < 0.01), **** (p < 0.0001). Each experiment was repeated.
Fig 12.
NBR1 affects inhibition of autophagic flux by TBSV in N. benthamiana.
(A) Agroinfiltrated N. benthamiana plants co-expressing eGFP-NBR1 and either TBSV (lanes 1–3) or pGD vector control (lanes 4–6) were exposed to darkness for 16 h to induce bulk autophagy. Total protein extracts were obtained at 2 dpi and probed using western blot with anti-GFP antibody. The released ‘free’ GFP is marked with an arrowhead. (B) VIGS was used to obtain NBR1-silenced (NBR1-KD) or control (TRV2-MBP) N. benthamiana plants. Then, 10 days later, agroinfiltration was used to co-express GFP-ATG8f and either TBSV (lanes 3–4 and 7–8) or pGD vector control (lanes 1–2 and 5–6). Plants were exposed to darkness for 16 h to induce bulk autophagy. Total protein extracts were obtained and probed using western blot with anti-GFP antibody. The released ‘free’ GFP is marked with an arrowhead. Autophagic flux was measured based on the ratio of GFP/GFP-ATG8f using Image J software. We used the controls (1.0 value) for normalization of inhibition of autophagy by TBSV for each group. (C) VIGS was used to obtain NBR1-silenced (NBR1-KD) or control (TRV2-MBP) N. benthamiana plants. Then, 10 days later, agroinfiltration was used to co-express Flag-ATG8f and either TBSV (lanes 3–4 and 7–8) or pGD vector control (lanes 1–2 and 5–6). At 2 dpi, total protein extracts were obtained and probed using western blot with anti-Flag antibody. Autophagic flux was measured based on the ratio of ATG8f-PE/ATG8f using Image J software. We used the controls (1.0 value) for normalization of inhibition of autophagy by TBSV for each group. Each experiment was repeated. The statistical analysis was performed using an ANOVA-test. * (p < 0.05).
Fig 13.
ATG8f and NBR1 are present in condensates associated with the TBSV VROs.
(A) Agroinfiltrated N. benthamiana leaves co-expressing p33-BFP, eGFP-NBR1 and RFP-ATG8f were used in fluorescence recovery after photobleaching (FRAP) assay. Confocal images were taken before and after photobleaching for 180 sec. Time ‘0 s’ indicates the time of photobleaching. Scale bars represent 10 μm. Right panel: Quantification of FRAP signals of p33-BFP, eGFP-NBR1 and RFP-ATG8f in the photobleached area was done at the indicated time points after photobleaching. (B) FRAP analysis shows the fluorescence recovery of the BiFC signals after photobleaching within a single VRO induced by TBSV p33-BFP in a N. benthamiana cell. Agroinfiltrated N. benthamiana plants co-expressed nYFP-NBR1 and cYFP-ATG8f and p33-BFP. The graph shows the extent of fluorescence recovery in four individual VROs. (C) Agroinfiltrated N. benthamiana leaves co-expressing p33-BFP, eGFP-NBR1 and RFP-ATG8f for 1.5 d were treated with 10% 1,6-hexanediol or digitonin (control) for 30 min with 10 min intervals. Confocal images of four individual VROs were taken every 10 min. Scale bars represent 10 μm. Each experiment was repeated three times.
Fig 14.
A model on subversion of selective autophagy for tombusvirus replication and inhibition of antiviral autophagy via sequestration of NBR1 and ATG8f in VRO-associated condensates.
(#1) The TBSV p33 replication protein recruits ATG8f and NBR1 and other core autophagy proteins into VROs formed from clustered peroxisomes. The CIRV p36 replication protein performs comparable recruitment into VROs formed from clustered mitochondria. The co-opted ATG8f and NBR1 are sequestered and ‘trapped’ in condensates associated with VROs. (#2) Hijacking the autophagy pathway results in recruitment of VPS34 PI3 kinase and enrichment of phospholipids, such as PE and PI(3)P phosphoinositide needed for VROs biogenesis. (#3) Sequestration of ATG8f and NBR1 in VRO-associated condensate leads to inhibition of antiviral cellular autophagy, thus facilitating tombusvirus replication.