A diacylglycerol-dependent signaling pathway contributes to regulation of antibacterial autophagy

Abstract

Autophagy mediates the degradation of cytoplasmic contents in the lysosome and plays a significant role in immunity. Lipid second messengers have previously been implicated in the regulation of autophagy. Here, we demonstrate a signaling role for diacylglycerol (DAG) in antibacterial autophagy. DAG production was necessary for efficient autophagy of Salmonella, and its localization to bacteria-containing phagosomes preceded autophagy. The actions of phospholipase D and phosphatidic acid phosphatase were required for DAG generation and autophagy. Furthermore, the DAG-responsive delta isoform of protein kinase C was required, as were its downstream targets JNK and NADPH oxidase. Previous studies have revealed a role for the ubiquitin-binding adaptor molecules p62 and NDP52 in autophagy of S. Typhimurium. We observed bacteria-containing autophagosomes colocalizing individually with either DAG or ubiquitinated proteins, indicating that both signals can act independently to promote antibacterial autophagy. These findings reveal an important role for DAG-mediated PKC function in mammalian antibacterial autophagy.

Figure 1. DAG colocalizes with bacteria-containing autophagosomes. (a)

HeLa cells were co-transfected with RFP-LC3 and either 2FYVE-GFP (PI(3)P probe), PLCδ-PH-GFP (PI(4,5)P₂ probe), GFP-PH-AKT (PI(3,5)P₂ and PI(3,4,5)P₃ probe) or PKCδ-C1-GFP (DAG probe). Cells were infected with wild-type S. Typhimurium, fixed at 1 h p.i. and immunostained for S. Typhimurium. Representative confocal z-slices are shown. The inner panels represent a higher magnification of the boxed areas. Size bar, 10 μm. (b) The percentage of RFP-LC3⁺ or RFP-LC3⁻ bacteria colocalizing with the lipid probes in a was determined by fluorescence microscopy. (c) HeLa cells were transfected with PKCδ-C1-GFP and infected as in a. Cells were treated with or without chloramphenicol (CM, 200 μg/mL) at 10 min p.i. for the remainder of the infection. Cells were fixed at the indicated time-points and immunostained for S. Typhimurium. The percentage of PKCδ-C1-GFP⁺ bacteria was enumerated by fluorescence microscopy. Data represent the mean ± standard error (s.e.m.) for three independent experiments.

Figure 2. DAG production on SCVs is independent of autophagy. (a-b)

Wild-type (WT) and autophagy-deficient (atg5 ^−/−) MEFs were transfected with either PKCδ-C1-GFP (a) or GFP-LC3 (b). Cells were infected with S. Typhimurium expressing mRFP (RFP-Sal). Cells were fixed at 45 min a or 1 h b p.i. Representative confocal z-slices are shown. The inner panels represent a higher magnification of the boxed areas. Size bar, 10 μm. (c) WT and atg5^−/− MEFs were transfected and infected as in a-b and fixed at the indicated time-points. PKCδ-C1-GFP⁺ or GFP-LC3⁺ RFP-Sal were determined by fluorescence microscopy. Data represent the mean ± standard error (s.e.m.) for at least three independent experiments.

Figure 3. Autophagy of S. Typhimurium requires phosphatidic acid phosphatase and phospholipase D. (a)

Pathways for the production of DAG. PC (phosphatidylcholine), PLD (phopholipase D), PA (phosphatidic acid), PAP (phosphatidic acid phosphatase), propr. (propranolol hydrochloride), PLC (phospholipase C), SMS (sphingomyelin synthase), SM (sphingomyelin). (b) HeLa cells were transfected with GFP-LC3 and infected with RFP-Sal. Cells were treated with growth medium (GM), U73122(10 μM), DMSO, propranolol hydrochloride (propr., 250 μM) or H₂O at 10 min p.i. for the remainder of the infection, and fixed at 1 h p.i. (c) HeLa cells were transfected with PKCδ-C1-GFP and infected with RFP-Sal. Cells were treated with GM, propranolol hydrochloride (propr., 250 μM) or D609 at the indicated concentrations at 10 min p.i. for the remainder of the infection, and fixed at 45 min p.i.. (d) HeLa cells were co-transfected with GFP-LC3 and either control siRNA (si-CTRL) or siRNA specifically targeting Atg12 (si-ATG12) or PAP2B (si-PAP). Cells were infected with RFP-Sal and fixed at 1 h p.i. (e) HeLa cells were transfected with RFP-LC3 and either GFP, HA-PLD1 DN (dominant negative) or HA-PLD2 DN. Cells were infected with S. Typhimurium, fixed at 1 h p.i. and immunostained for S. Typhimurium and HA tag. The percentage of DAG⁺ or LC3⁺ S. Typhimurium was determined by fluorescence microscopy. Data represent the mean ± standard error (s.e.m.) for at least three independent experiments.

Figure 4. Autophagy of S. Typhimurium requires protein kinase C. (a-b)

HeLa cells were co-transfected with GFP-LC3 and either control siRNA (si-CTRL) or siRNA specifically targeting Atg12 (si-ATG12), PKCδ (si-PKCδ) or PKCα (si-PKCα). Cells were infected with wild-type RFP-Sal and fixed at 1 h p.i. Representative confocal z-slices are shown a. The inner panels represent a higher magnification of the boxed areas. Size bar represents 10 μm. The percentage of GFP-LC3⁺ bacteria was determined by fluorescence microscopy b. (c) WT and PKCδ-deficient (PKCδ^−/−) MEFs were transfected with GFP-LC3 and/or a PKCδ expression plasmid. Cells were infected with RFP-Sal and fixed at 1 h p.i. The percentage of GFP-LC3⁺ bacteria was determined by fluorescence microscopy. (d) WT and PKCδ-deficient (PKCδ ^−/−) MEFs were infected with wild-type S. Typhimurim and lysed at the indicated time points. The number of bacteria was determined by the number of colonies formed on agar plates. Data represent the mean ± standard error (s.e.m.) for at least three independent experiments.

Figure 5. Pkc1 is required for starvation-induced autophagy in yeast. (a)

GFP-Atg8 processing was monitored in wild-type (TN124), atg1Δ (TYY127), pkc1^ts, pkc1-3 and pkc1-4 cells expressing GFP-Atg8. Cells were grown in SD-N medium at the permissive (PT, 24°C) or non-permissive (NPT, 38 C) temperature. Full-length GFP-Atg8 and free GFP were detected by Western blotting using anti-YFP antibodies. (b) Autophagic activity was determined by obtaining extracts from wild-type (TN124), atg1Δ (TYY127), pkc1^ts, pkc1-3 and pkc1-4 S. cerevisiae expressing Pho8Δ60 and analyzed for Pho8Δ60-dependent alkaline phosphatase activity. Synthetic defined media lacking nitrogen (SD-N). Data represent the mean ± standard error (s.e.m.) for at least three independent experiments.

Figure 6. DAG and p62 act in independent signaling pathways for anti-bacterial autophagy. (a)

HeLa cells were co-transfected with PKCδ-C1-GFP and RFP-LC3. Cells were infected with wild-type S. Typhimurium, fixed at 45 min p.i. and immunostained for S. Typhimurium and mono- and polyubiquitinated proteins. Representative confocal z-slices are shown. The inner panels represent a higher magnification of the boxed areas. Size bar, 10 μm. (b) HeLa cells were transfected with PKCδ-C1-GFP, infected with wild-type S. Typhimurium and fixed at 45 and 60 min p.i.. Cells were immunostained for ubiquitinated protein and bacteria as in a. DAG⁺ bacteria were scored for colocalization with ubiquitin. (c) HeLa cells were transfected with PKCδ-C1-GFP, infected with wild-type S. Typhimurium, fixed at 45 min p.i. and immunostained for S. Typhimurium and p62. p62 colocalization was quantified for DAG⁺ and DAG⁻ bacteria (d) HeLa cells were co-transfected with GFP-LC3 and either control siRNA (si-CTRL) or siRNA specifically targeting PKCδ (si-PKCδ) PAP (si-PAP) or p62 (si-p62). Cells were infected with wild-type S. Typhimurium and treated with 15 μM rottlerin where indicated. Cells were fixed at 1 h p.i. and quantified for colocalization with GFP-LC3. Data represent the mean ± standard error (s.e.m.) for three independent experiments.

Figure 7. Two signals target S. Typhimurium to the autophagy pathway

Model depicting the dual pathways by which S. Typhimurium is targeted by autophagy in mammals and the conserved role for PKC in the regulation of this process in both mammals and yeast. Data is consistent with two independent pathways for induction of antibacterial autophagy in mammalian cells (DAG and ubiquitin pathways). Dashed line and question mark between pathways represents the possibility that the two pathways may interact with each other.