Cytoplastic Glyceraldehyde-3-Phosphate Dehydrogenases Interact with ATG3 to Negatively Regulate Autophagy and Immunity in Nicotiana benthamiana

Abstract

Autophagy as a conserved catabolic pathway can respond to reactive oxygen species (ROS) and plays an important role in degrading oxidized proteins in plants under various stress conditions. However, how ROS regulates autophagy in response to oxidative stresses is largely unknown. Here, we show that autophagy-related protein 3 (ATG3) interacts with the cytosolic glyceraldehyde-3-phosphate dehydrogenases (GAPCs) to regulate autophagy in Nicotiana benthamiana plants. We found that oxidative stress inhibits the interaction of ATG3 with GAPCs. Silencing of GAPCs significantly activates ATG3-dependent autophagy, while overexpression of GAPCs suppresses autophagy in N. benthamiana plants. Moreover, silencing of GAPCs enhances N gene-mediated cell death and plant resistance against both incompatible pathogens Tobacco mosaic virus and Pseudomonas syringae pv tomato DC3000, as well as compatible pathogen P. syringae pv tabaci. These results indicate that GAPCs have multiple functions in the regulation of autophagy, hypersensitive response, and plant innate immunity.

Figure 1.

Nb-ATG3 Interacts with Nb-GAPCs in Vivo and in Vitro. (A) LCI assays showed the interaction of Nb-ATG3 with Nb-GAPCs in plants. Image shown is luminescence of N. benthamiana leaf that was agroinfiltrated with NbATG3-nLUC (left) or negative control HA-nLUC (right) with the cLUC-tagged GAPC1, GAPC2, GAPC3, and a negative control cLUC. The experiments were repeated three times with similar results. (B) Nb-ATG3 coimmunoprecipitated with Nb-ATG8f, Nb-GAPC1, and Nb-GAPC2. NbATG3-Myc was coexpressed with GFP-NbATG8f, NbGAPC1-GFP, or NbGAPC2-GFP in N. benthamiana leaves by agroinfiltration. NbATG3-Myc coexpressed with GFP was introduced as a negative control. At 60 hpi, leaf lysates were immunoprecipitated with anti-GFP beads and then the immunoprecipitates were assessed by immunoblotting (IB) using anti-Myc (upper panel) and anti-GFP antibodies (middle panel). In addition to immunoblotting for co-IP, presence of NbATG3-Myc in each treatment of the cell lysates was also analyzed (lower panel). (C) GST pull-down assay for detection of in vitro interaction of ATG3 with GAPCs. The total soluble proteins of E. coli expressing NbGAPC1-3×FLAG-6×His or NbGAPC2-3×FLAG-6×His were incubated with GST-NbATG3 or GST immobilized on glutathione-Sepharose beads. Beads were washed and proteins analyzed by immunoblot assays using an anti-Flag antibody (upper panel). The middle panel shows inputs of Flag-tagged proteins in pull-down assays. Equal aliquots of glutathione beads loaded with GST-NbATG3 or GST were separated by SDS-PAGE and stained with Coomassie blue (CBB). (D) BiFC assays to show the interaction of ATG3 with GAPCs in plants. NbATG3-nYFP or nYFP was coexpressed transiently with NbGAPC1-cYFP, NbGAPC2-cYFP, or cYFP in N. benthamiana leaves, and fluorescence was detected for mesophyll cells. Combinations of NbATG3-nYFP with NbGAPC1-cYFP or NbGAPC2-cYFP, but not other combinations, gave YFP fluorescence signals (left). Yellow color indicates positive interaction signal, while red color indicates the signals from chloroplasts (middle). The experiments were repeated three times with similar results. Bars = 20 μm.

Figure 2.

BiFC Assays Show That MV Treatments Weaken ATG3-GAPC Interactions. (A) to (C) NbATG3-nYFP, along with CFP, was coexpressed transiently with NbGAPC1-cYFP (A), NbGAPC2-cYFP (B), or cYFP-NbATG8f ([C]; as a control) in N. benthamiana leaves, followed by mock (water) treatment (upper panel) or treatment with 20 μM MV (lower panel). Yellow color indicates the YFP signal. Cyan color indicates the signal of CFP alone, which was used as an internal control. For a view of a large area, a 10× objective lens was used for confocal image. Bars =50 μm. (D) BiFC intensity was quantified by YFP/CFP ratio. Relative BiFC intensity was normalized to the control. Values represent means ± se from three independent experiments. Student’s t test was used to determine significant differences from mock (***P < 0.001 and *P < 0.05, Student’s t test).

Figure 3.

Co-IP Assays Show That H₂O₂ Treatments Weaken ATG3-GAPC Interactions. Total proteins from N. benthamiana leaves coexpressing NbATG3-Myc with GFP-NbATG8f (I), NbGAPC1-GFP (II), or NbGAPC2-GFP (III) were immunoprecipitated with anti-GFP beads, followed by immunoblotting using anti-Myc (upper panel) and anti-GFP antibodies (middle panel) in different redox environments. Different interaction levels were observed in the water, 10 mM H₂O₂, 10 mM H₂O₂ with 10 mM DTT, and 20 mM H₂O₂ treatments as indicated. DTT was added before the addition of H₂O₂ when both were applied. The middle panel shows the input of immunoprecipitates using anti-GFP antibodies. The presence of NbATG3-Myc in lysate was also analyzed (bottom panel).

Figure 4.

Silencing of GAPCs Activates Autophagy. (A) Representative confocal images of dynamic autophagic activity revealed by specific autophagy marker CFP-NbATG8f in GAPC-silenced plants (GAPC-CoVIGS) and control plants. Autophagosomes and autophagic bodies are revealed as CFP-positive puncta in mesophyll cells. CFP-NbATG8f fusion proteins are in cyan, and chloroplasts are in red. Bars = 20 μm. (B) Relative autophagic activity in GAPC-silenced plants was normalized to that of TRV control plants, which was set to 1.0. Quantification of the CFP-NbATG8-labeled autophagic puncta per cell was performed. More than 150 mesophyll cells for each treatment were used for the quantification. Values represent means ± se from three independent experiments. Different letters indicate significant differences (ANOVA, P < 0.05). (C) to (E) Magnification of the mesophyll cells in (A) surrounded by a dashed line (C), a solid line (D), and a yellow line (E). Autophagosomes labeled by CFP-NbATG8f are indicated by arrows. Yellow arrows indicate autophagosomes in the cytoplasm. White arrows indicate autophagic bodies in the vacuole. Bars = 20 μm.

Figure 5.

Representative TEM Images of Autophagic Structures. (A) Representative ultrastructure of autophagic bodies (arrows) observed in the vacuoles of mesophyll cells of GAPC-silenced plants (top panels), TRV alone control (middle panels), or plants cosilencing GAPC1-3 and ATG3 (bottom panels). Cp, chloroplast; M, mitochondrion; V, vacuole. Bars = 2 μm. (B) Representative ultrastructure of autophagosomes observed in the cytoplasm of mesophyll cells. The black arrows show the classic double-membrane autophagosomes, and the black arrowheads indicate the autophagic bodies inside the vacuole. Bars = 500 nm. (C) Relative autophagic activity in GAPC-silenced plants was normalized to that of TRV control plants, which was set to 1.0. Approximately 20 cells were used to quantify autophagic structures in each treatment. Values represent means ± se from three independent experiments. Different letters indicate significant differences (ANOVA, P < 0.05).

Figure 6.

Overexpression of ATG3 Induces Autophagy. (A) and (B) Representative confocal images of autophagic activity revealed by autophagy marker CFP-NbATG8f (A) or MDC staining (B) in cells overexpressing ATG3 (upper panel) and nLUC control (lower panel). CFP-NbATG8f-labeled punctate autophagosomes and autophagic bodies inside vacuoles (cyan), MDC-labeled punctate autophagosomes and autophagic bodies inside vacuoles (green), autofluorescence from chloroplasts (red), and merged images (merge). Bars = 20 μm. (C) and (D) Relative autophagic activity revealed by autophagy marker CFP-NbATG8f (C) or MDC staining (D) in leaves overexpressing ATG3 was normalized to that of leaves overexpressing nLUC. Quantification of the CFP-NbATG8-labeled autophagic puncta or MDC-stained structures per cell was performed to calculate the autophagic activity using more than 150 mesophyll cells for each treatment. Values represent means ± se from three independent experiments (***P <0.001 and ** P < 0.01, Student’s t test).

Figure 7.

Visualization of Autophagy Level in ATG3/GAPC Co-Overexpressing Leaves. (A) and (B) Representative images of autophagic activity revealed by autophagy marker CFP-NbATG8f (A) or MDC staining (B) in leaves coexpressing NbATG3-Myc with HA-nLUC (upper), NbGAPC1-HA (middle), or NbGAPC2-HA (bottom). CFP-NbATG8f fusion proteins are in cyan, MDC are in green, and chloroplasts are in red. Bars = 20 μm. (C) and (D) Relative autophagic activity revealed by autophagy marker CFP-NbATG8f (C) or MDC staining (D) in leaves coexpressing NbATG3-Myc with NbGAPC1-HA or NbGAPC2-HA was normalized to that of leaves coexpressing NbATG3-Myc and HA-nLUC. Quantification of the CFP-NbATG8-labeled autophagic puncta per cell (C) and MDC positive structures per cell (D) were performed to calculate the autophagic activity. More than 150 mesophyll cells for each treatment were used for the quantification. Values represent means ± se from three independent experiments. Different letters indicate significant differences (ANOVA, P < 0.05).

Figure 8.

Overexpression of GAPCs Inhibits MV-Induced Autophagy. (A) and (B) Representative images of autophagic activity revealed by autophagy marker CFP-NbATG8f (A) or MDC staining (B) in leaves expressing NbGAPC1-HA, NbGAPC2-HA, or HA-nLUC, followed by treatment with 10 μM MV or mock (water). Numerous CFP-NbATG8f-labeled positive puncta or MDC-stained structures were observed in leaves expressing HA-nLUC followed by MV treatment, but not in other leaves. CFP-NbATG8f fusion proteins are in cyan, MDC is in green, and chloroplasts are in red. Bars = 20 μm. (C) and (D) Relative autophagic activity revealed by autophagy marker CFP-NbATG8f (C) or MDC staining (D) in leaves expressing NbGAPC1-HA, NbGAPC2-HA, or HA-nLUC, followed by treatment with 10 μM MV was normalized to that of leaves expressing HA-nLUC followed by mock treatment. Quantification of the CFP-NbATG8-labeled autophagic puncta per cell (C) and MDC-positive structures per cell (D) was performed to calculate the autophagic activity. More than 150 mesophyll cells for each treatment were used for the quantification. Values represent means ± se from three independent experiments. Different letters indicate significant differences (ANOVA, P < 0.05).

Figure 9.

Silencing of GAPCs Enhances Resistance and Hypersensitive Response. (A) to (C) Silencing of Nb-GAPCs (GAPCs-VIGS) enhanced N gene-mediated resistance against TMV. (A) Representative photographs of TMV-GFP-inoculated leaves were taken at 3 dpi under UV light to assess TMV infection in nonsilenced TRV control (left) or GAPC-silenced plants (right). Bars = 1 cm. (B) Average TMV-GFP infection foci at 3 dpi are shown. All values in bar graphs represent means with sd (**P < 0.01, Student’s t test, n = 3). The experiments were repeated three times with similar results. (C) Real-time RT-PCR confirmed that silencing of GAPCs reduced TMV-GFP RNA levels in local inoculated leaves at 3 dpi. (D) Silencing of GAPCs accelerated TMV-p50-induced hypersensitive response cell death in NN plants. TMV-p50 was expressed by agroinfiltration in nonsilenced TRV control (upper) or GAPCs-silenced plants (lower). Representative photographs were taken at 2 dpi after trypan blue staining. Bars = 1 cm. (E) Quantitative representation of TMV-p50-induced HR programmed cell death. For quantification of death intensity, images were converted to gray scale, and mean gray value of inoculation area minus noninoculation area was calculated using ImageJ. Values represent means ± se (***P < 0.001, Student’s t test, n = 12). The experiments were repeated three times with similar results. (F) Silencing of GAPCs inhibited the growth of compatible pathogen P. syringae pv tabaci. Control plants (TRV alone) and GAPC-silenced plants (GAPCs-VIGS) were infiltrated with P. syringae pv tabaci at 10³ cfu/mL, and bacterial populations were quantified at 2 dpi (*P < 0.05, Student’s t test, n = 5). The experiments were repeated three times with similar results. (G) Silencing of GAPCs inhibited the growth of incompatible pathogen P. syringae pv tomato DC3000. Control plants (TRV alone) and GAPCs-silenced plants (GAPCs-VIGS) were infiltrated with P. syringae pv tomato DC3000 at 10³ cfu/mL, and bacterial populations were quantified at 3 dpi (*P < 0.05, Student’s t test, n = 3). The experiments were repeated twice with similar results.