Autophagic control of listeria through intracellular innate immune recognition in drosophila

Abstract

Autophagy, an evolutionally conserved homeostatic process for catabolizing cytoplasmic components, has been linked to the elimination of intracellular pathogens during mammalian innate immune responses. However, the mechanisms underlying cytoplasmic infection-induced autophagy and the function of autophagy in host survival after infection with intracellular pathogens remain unknown. Here we report that in drosophila, recognition of diaminopimelic acid-type peptidoglycan by the pattern-recognition receptor PGRP-LE was crucial for the induction of autophagy and that autophagy prevented the intracellular growth of Listeria monocytogenes and promoted host survival after this infection. Autophagy induction occurred independently of the Toll and IMD innate signaling pathways. Our findings define a pathway leading from the intracellular pattern-recognition receptors to the induction of autophagy to host defense.

Figure 1

PGRP-LE in hemocytes is essential for resistance against L. monocytogenes infection in vivo. (a,b) Survival rate of wild-type (Oregon R) flies, yw flies, ywPGRP-LE¹¹² (PGRP-LE¹¹²), w;;PGRP-LC⁷⁴⁵⁴ (PGRP-LC⁷⁴⁵⁴), and ywPGRP-LE¹¹²,PGRP-LC⁷⁴⁵⁴ double-mutant flies (PGRP-LE¹¹²,PGRP-LC⁷⁴⁵⁴), after (a) wild-type or (b) Δhly L. monocytogenes injections. (c) Survival rate of Oregon R, yw, ywPGRP-LE¹¹² (PGRP-LE¹¹²), UAS-PGRP-LE IR/+; hml-GAL4/+, (hml-GAL4>PGRP-LE RNAi) and ywPGRP-LE¹¹²; UAS-PGRP-LE/+;hml-GAL4/+ (PGRP-LE¹¹², hml-GAL4>PGRP-LE) flies after wild-type L. monocytogenes injections. All experiments were performed at 28°C. The average survival rate of four independent experiments is shown. In each experiment, >30 flies of the indicated genotypes were examined at the same time. * P<0.01 (Wilcoxon-Mann-Whitney test) (P=0.0006, 0.0043, 0.0043, yw versus PGRP-LE¹¹², PGRP-LC⁷⁴⁵⁴, PGRP-LE¹¹²,PGRP-LC⁷⁴⁵⁴ flies, respectively in (a) P=0.0095 yw versus hml-GAL4>PGRP-LE RNAi flies in (c))

Figure 2

PGRP-LE and autophagy, but not the Toll and IMD pathways, are required to suppress the intracellular growth of L. monocytogenes in hemocytes. (a) Hemocytes from third instar larvae of the indicated genotype were cultured ex vivo and infected with wild-type L. monocytogenes. Hemocyte nuclei (filled arrowhead) and DNA of L. monocytogenes (open arrowhead) were visualized by DAPI staining (blue or white), and the actin cytoskeleton was stained with rhodamine-labeled phalloidin (red). For the induction of RNAi against Atg5, an inverted repeat sequence of Atg5 (Atg5IR) was expressed using hml-GAL4 (hmlGAL4>Atg5IR). Bars represent 10 µm. (b) The number of intracellular wild-type (WT Lm) or Δhly (Δhly Lm) L. monocytogenes per hemocyte of the indicated genotype was manually counted by DAPI staining of the infected hemocytes. Bars indicate standard deviation of triplicate measurements of at least three independent experiments. *P < 0.001 (t-test) each genotype versus wild-type. Genotype abbreviations used: PGRP-LE¹¹²,ywPGRP-LE¹¹²; LE¹¹²,hml>LE, ywPGRP-LE¹¹²;UAS-PGRP-LE/+;hml-GAL4/+; PGRP-LC⁷⁴⁵⁴, w;;PGRP-LC⁷⁴⁵⁴; Rel^E20, Relish^E20; J4, Df(2L)J4; hml>Atg5IR, ywUAS-Atg5IR/+;;hml-GA4/+; w⁻,hml>Atg1, w⁻;;UAS-Atg1^6B/hml-GAL4, LE¹¹²,hml>Atg1, ywPGRP-LE¹¹²;;UAS-Atg1^6B/hml-GAL4

Figure 3

Autophagy is crucial for host survival against L. monocytogenes infection. (a–c) Survival rate of yw flies, PGRP-LE¹¹², Atg5 RNAi (hml-GAL4>Atg5IR), and control hml-GAL4-carrying flies (hml-GAL4>GFP) after injection of (a) wild-type or (b) Δhly L. monocytogenes, or (c) E. carotovora (Ecc15) into adult flies. The average survival rate of four independent experiments is shown. * P < 0.01 (Wilcoxon-Mann-Whitney test versus hmlGAL4>GFP in (a), versus yw in (c)) (P=0.0024, 0.0007, hml-GAL4>GFP versus hml-GAL4>Atg5IR, PGRP-LE¹¹², respectively in (a) P=0.0055, yw versus PGRP-LC⁷⁴⁵⁴ flies in (c))

Figure 4

Autophagy, but not the IMD or Toll pathways, is crucial for PGRP-LE–mediated suppression of intracellular growth of L. monocytogenes in S2 cells. (a) S2 cells (S2) or S2 cells expressing PGRP-LE (S2-LE) were transfected with double-stranded RNA specific for Atg5, Rel or Dif/dl (RNAi, below graph) and infected with wild-type (WT Lm) or Δhly (Δhly Lm) L. monocytogenes for 1.5 h, followed by 6 h incubation in CuSO₄- and gentamicin-containing medium, and L. monocytogenes growth was quantified by determining colony-forming units by plate assay. Bars indicate the variance of two independent experiments. *P < 0.001 (t-test) (b, c) Co-localization of PGRP-LE with wild-type L. monocytogenes in S2 cells. S2 cells engineered to express YFP-PGRP-LE were infected with (b) WT Lm or (c) Δhly Lm for 0.5 h, followed by 1 h incubation in gentamicin-containing medium, DAPI staining, and visualization by fluorescence confocal microscopy. YFP-PGRP-LE (green) and DAPI (magenta) are shown in the merged panel. Closed arrowheads show L. monocytogenes, and the open arrowhead indicates YFP-PGRP-LE accumulated around the bacteria. Scale bar, 5 µm.

Figure 5

PGRP-LE mediates autophagosome formation in S2 cells and hemocytes. (a) The number of dot- or ring-shaped GFP-LC3 signals per cell was quantified after wild-type (WT) or Δhly L. monocytogenes infection or without infection (−) of S2 cells expressing both PGRP-LE and GFP-LC3 or only GFP-LC3, or after 1.5 h incubation with 5 µM rapamycin (rap). Bars indicate standard deviation of triplicate measurements. (b) The number of GFP-LC3 dots colocalized with wt L. monocytogenes in indicated S2 cells was quantified using confocal microscopy images. Bars indicate standard deviation of triplicate measurements. (c) Confocal microscopy images of WT L. monocytogenes infected S2 cells expressing PGRP-LE and GFP-LC3. GFP-LC3 (green), DAPI (magenta). Scale bar, 5 µM. Arrow indicates co-localization of GFP-LC3 and L. monocytogenes. (d) Indicated S2 cells were infected with WT L. monocytogenes (Lm), or treated with 5 µM rapamycin (rap) and indicated proteins in lysates were detected by immunoblotting. (e–i) Ultrastructural analysis of WT L. monocytogenes-infected S2 cells expressing PGRP-LE and GFP-LC3. (e,f) Fluorescence microscopy (e) and electon microscopy (f) images of cells expressing GFP-LC3 (green) stained with DAPI (magenta). Scale bars, 1 µm. (g) Magnified image of a bacteria-containing vacuole shown in (f). Scale bar, 500 nm. (h, i) Magnified images of the fields indicated in (g). Arrows indicate double-membrane structure that surrounds the bacteria. Arrowheads indicate endoplasmic reticulum–like membrane. (j) Confocal microscopy images of WT L. monocytogenes-infected hemocytes. GFP-LC3 (green), DAPI (magenta). (k) The number of dot- or ring-shaped GFP-LC3 signals per cell in ex vivo-cultured hemocytes expressing GFP-LC3 infected or treated as indicated. Bars indicate standard deviation of triplicate measurements. (l) Hemocytes from third instar larvae of the indicated genotype were cultured ex vivo and infected or treated as indicated and lysates were probed with antibodies specific for the indicated antibodies. Arrowhead indicates processed form of GFP-LC3. (m) S2 cells expressing PGRP-LE and GFP-LC3 under the control of an actin promoter were transfected with double-stranded RNA specific for indicated transcripts (RNAi, below graph) and infected with L. monocytogenes for 0.5 h. After 1 h incubation in gentamicin-containing medium, GFP-LC3 dot formation was quantified by confocal microscopy. Bars indicate standard deviation of at least triplicate measurements. * P < 0.001 (t-test).

Figure 6

PGRP-LE is responsible for TCT and DAP-type PGN-induced autophagy. (a) S2 cells expressing PGRP-LE (act-LE) and GFP-LC3, or GFP-LC3 only, were treated with 100 nM TCT, 100 µg/ml highly purified DAP-type peptidoglycans from L. plantarum (DAP), or lysine-type peptidoglycans from S. epidermidis (Lys). After 2 h incubation, GFP-LC3 dot formation was quantified. Bars indicate the variance of two independent experiments. (b) The number of dot- or ring-shaped GFP-LC3 signals per hemocyte of indicated genotype was quantified after wild-type (WT) or Δhly L. monocytogenes infection, 5 µM rapamycin treatment (rap), or treatment with 100 nM TCT (TCT), 100 µg/ml highly purified DAP-type PGN from L. plantarum (DAP), or lysine-type PGN from S. epidermidis (Lys). Bars indicate variance of two independent experiments. Genotypes: w⁻; UAS-GFP-LC3/UAS-Gal4; heat-shock-GAL4/+ (WT), PGRP-LE¹¹² ; UAS-GFP-LC3/UAS-Gal4; heat-shock-GAL4/+ (LE¹¹²), UAS-Atg5 IR/+, UAS-GFP-LC3/UAS-Gal4; heat-shock-GAL4/+ (Atg5 IR). * P < 0.001 (t-test).