Intestinal epithelial autophagy is essential for host defense against invasive bacteria

Abstract

The mammalian intestine is colonized with a diverse community of bacteria that perform many beneficial functions but can threaten host health upon tissue invasion. Epithelial cell-intrinsic innate immune responses are essential to limit the invasion of both commensal and pathogenic bacteria and maintain beneficial host-bacterial relationships; however, little is known about the role of various cellular processes, notably autophagy, in controlling bacterial interactions with the intestinal epithelium in vivo. We demonstrate that intestinal epithelial cell autophagy protects against tissue invasion by both opportunistically invasive commensals and the invasive intestinal pathogen Salmonella Typhimurium. Autophagy is activated following bacterial invasion of epithelial cells through a process requiring epithelial cell-intrinsic signaling via the innate immune adaptor protein MyD88. Additionally, mice deficient in intestinal epithelial cell autophagy exhibit increased dissemination of invasive bacteria to extraintestinal sites. Thus, autophagy is an important epithelial cell-autonomous mechanism of antibacterial defense that protects against dissemination of intestinal bacteria.

Figure 1. Salmonella typhimurium triggers autophagosome formation in small intestinal epithelial cells

(A) Germ-free and conventional mice were orally inoculated with 10⁹ CFU of S. typhimurium. Sections of distal small intestine (ileum) were stained with anti-LC3 and anti-rabbit IgG-Cy3 (red). Uninfected mice show diffuse cytoplasmic LC3, while LC3⁺ autophagosomes are visible as distinct puncta in the infected mice (arrowheads). Tissues were counterstained with DAPI (blue). Scale bar=10 μm. (B) Autophagosome number as a function of small intestinal location (proximal to distal). Tissue sections were from the experiment shown in (A). Data are represented as mean±SEM; ***, p<0.001, n=6 mice. (C) Time course of LC3⁺ autophagosome formation after oral challenge with 10⁹ CFU of S. typhimurium. Numbers of LC3⁺ puncta in ileal sections were counted. Data are represented as mean±SEM; n=6 mice/time point). (D) Western blot of LC3 and actin (loading control) from ileal epithelial cells isolated from germ-free mice before and after colonization with S. typhimurium for 24 and 72 hr. Epithelial cell protein extracts were blotted and probed with anti-LC3 and anti-rabbit IgG-HRP. LC3-I and II denote the non-lipidated and lipidated forms of MAP1LC3, respectively. Results are representative of three independent experiments. (E) The LC3-I and LC3-II band intensities in (D) were quantified by scanning densitometry and normalized to the actin band intensity. Data are represented as mean±SEM; *, p<0.05; n=3 independent experiments. (F) Colocalization of autophagosomes and S. typhimurium expressing green fluorescent protein (S. typhimurium-GFP). Fixed tissue sections were stained with anti-GFP (green), anti-LC3 (red), and counterstained with DAPI (blue). Note that intrinsic GFP fluorescence is destroyed by tissue fixation, necessitating antibody detection of the GFP. The merged image shows GFP-labeled bacteria that colocalize with LC3⁺ puncta in ileal epithelial cells (arrowheads). Scale bar=10 μm. (G) Pearson’s colocalization coefficient for LC3 and S. typhimurium-GFP. The coefficient was calculated from three independent experiments with 3 fields per experiment, for a total of 9 fields. Data are represented as mean±SEM. (H,I) Quantification of LC3 and GFP colocalization in ileal epithelial cells from mice challenged with S. typhimurium-GFP and stained with anti-LC3. Results are the means±SEM of three independent experiments. See also Figure S1 and Movie S1.

Figure 2. Enterococcus faecalis induces formation of autophagosomes in small intestinal epithelial cells

(A) Quantification of intracellular Lactobacillus salivarius and Enterococcus faecalis in the ileal epithelium. 10⁹ cfu of GFP-expressing L. salivarius or E. faecalis were introduced into germ-free mice and tissue-associated bacteria (green) were visualized by fluorescence microscopy of fresh-frozen ileal tissues after 24 hr (arrowheads). GFP was visualized on the basis of its intrinsic fluorescence. Tissues were also stained with phalloidin (red) to visualize filamentous actin at the epithelial cell border, and were counterstained with DAPI (blue) to visualize cell nuclei. Scale bars=10 μm. (B) Ileal tissues were stained with anti-LC3 (red) to visualize LC3⁺ autophagosomes (arrows). Scale bar=10 μm. (C) L. salivarius and E. faecalis small intestinal colonization levels 24 hr after oral inoculation. ns, not significant. Data are represented as mean±SEM. (D) Enumeration of intracellular bacteria in (A). Epithelial cell-associated bacteria were counted in 200 well-oriented crypt-villus units from 5 mice per group. Data are represented as mean±SEM; ***, p<0.001 (E) LC3⁺ puncta were enumerated in the epithelial cells of 200 well-oriented ileal crypt-villus units from 5 mice per group. Data are represented as mean±SEM; *, p<0.05; nd, not detected. (F) Western blot of LC3 from ileal epithelial cells isolated from germ-free mice before and after colonization with E. faecalis for 24 hr. The experiment was performed as described for Figure 1D and results are representative of three independent experiments. (G) The LC3-I and LC3-II band intensities in (F) were quantified by scanning densitometry and normalized to actin band intensity. Data are represented as mean±SEM; **, p<0.01, n=3 independent experiments.

Figure 3. Ultrastructure of autophagosomes induced by bacteria in small intestinal epithelial cells

(A) Transmission electron microscopy of an ileal epithelial cell of a germ-free mouse. Key morphological features such as the nucleus, microvilli, tight junctions (TJ), epithelial cell border (ECB), and mitochondria (mt) are indicated. Scale bar=2 μm. (B,C) Lower and higher magnification images of an ileal epithelial cell 24 hr after S. typhimurium colonization of a germ-free mouse. Arrows indicate double membrane-bound compartments surrounding bacteria (St). Scale bars=1 μm (B) and 0.5 μm (C). (D,E) Ileal epithelial cell 24 hr after E. faecalis colonization of a germ-free mouse. Arrows indicate examples of double membrane-bound compartments enclosing bacteria (Ef). Scale bars=2 μm (D) and 1 μm (E).

Figure 4. Invasive bacteria elicit autophagy in intestinal epithelial cells

(A) 10⁹ CFU of wild-type or isogenic mutant S. typhimurium (ΔSPI-1 or ΔinvA) were introduced into germ-free mice and LC3⁺ puncta in ileal epithelial cells were visualized by immunofluorescence 24 hr later. Scale bar=10 μm. (B) Quantification of LC3⁺ puncta in (A). Data are represented as mean±SEM; *, p<0.05; n=4 mice/group. See also Figure S2.

Figure 5. Intestinal epithelial autophagy requires epithelial cell-intrinsic MyD88

(A,B) MyD88 is required for intestinal epithelial autophagy. Immunofluorescence detection of LC3 (red) in ileal epithelial cells from wild-type and MyD88^-/- mice (A). Tissues were counterstained with DAPI (blue). Mice were orally infected with S. typhimurium for 24 hr or were left uninfected. Arrowheads indicate examples of LC3⁺ puncta within intestinal epithelial cells. LC3⁺ autophagosomes were quantified in (B). Data are represented as mean±SEM; *, p<0.05; n=5-8 mice/group; nd, not detected; scale bar=10 μm. (C,D) Epithelial cell-intrinsic MyD88 is required for intestinal epithelial cell autophagy. Myd88^ΔIEC mice or Myd88^fl/fl littermates were analyzed by immunofluorescence detection of LC3 in ileal epithelial cells. LC3⁺ puncta were quantified in (D). Data are represented as mean±SEM; *, p<0.05; n=5-8 mice/group; nd, not detected; scale bar=10 μm. (E,F) Western blot analysis of LC3 from ileal epithelial cells isolated from Myd88^fl/fl or Myd88^ΔIEC mice before and after colonization with S. typhimurium for 24 hr (E). The experiment was performed as described in Figure 1D, and band intensities from three independent experiments were quantified in (F). Data are represented as mean±SEM; *, p<0.05. (G,H) Intestinal epithelial autophagy does not require Nod2. Ileal tissue sections from conventionally-raised Nod2^-/- mice were analyzed by immunofluorescence detection of LC3 (G). LC3⁺ puncta formation was reversible by antibiotic (Abx) treatment to reduce the microflora. LC3⁺ puncta were quantified in (H). Data are represented as mean±SEM; *, p<0.05; n=5-8 mice/group; scale bar=10 μm. (I,J) Increased bacterial invasion into the epithelial cells of Nod2^-/- mice. (I) Bacteria were localized by FISH using a fluorescent probe against the 16S rRNA genes of all bacteria (universal 16S, red). (J) Cell-associated bacteria in (I) were counted. Data are represented as mean±SEM; ***, p<0.001; n=3 mice/group; scale bar=10 μm. See also Figure S3.

Figure 6. Intestinal epithelial ATG5 is required for control of S. typhimurium dissemination in vivo

(A,B) Atg5^fl/fl and Atg5^ΔIEC littermates were orally infected with S. typhimurium-GFP. Ileal sections were stained with anti-LC3 (red) to visualize LC3⁺ autophagosomes (arrows)(A). Scale bar=10 μm. Numbers of LC3⁺ autophagosomes in (A) were counted (B). Data are represented as mean±SEM; ***, p<0.001; n=5 mice/group. (C,D) Western blot analysis of LC3 from ileal epithelial cells isolated from Atg5^fl/fl and Atg5^ΔIEC mice before and after oral infection with S. typhimurium for 24 hr (C). The experiment was performed as described in Figure 1D, and band intensities from three independent experiments are quantified in (D). Data are represented as mean±SEM; *, p<0.05; **, p<0.01; nd, not detected. (E,F) Intracellular bacteria (green) were visualized by immunofluorescence microscopy in the ileums of Atg5^fl/fl and Atg5^ΔIEC mice 24 hr post-infection with S. typhimurium-GFP (E). Fixed tissue sections were counterstained with DAPI (blue) to visualize cell nuclei. Arrowheads point to examples of intracellular bacteria. Scale bar=10 μm. (F) Intracellular bacteria were counted. Data are represented as mean±SEM; ***, p<0.001; n=5 mice/group. (G-I) Bacterial burdens (CFU) in the spleen (G), liver (H), and small intestines (I) of Atg5^fl/fl and Atg5^ΔIEC littermates 24 hr after oral infection with 10⁹ CFU of wild-type S. typhimurium or the non-invasive mutant strains ΔSPI-1 and ΔinvA. Each point represents an individual mouse; data are from three independent experiments. Data are represented as mean±SEM; *, p<0.05, ns=not significant. Dashed lines indicate limits of detection. (J) Bacterial burden (CFU) in the spleen 24 hr after intraperitoneal infection of Atg5^ΔIEC and Atg5^fl/fl littermates with 10³ CFU of wild-type S. typhimurium. ns, not significant. Data are represented as mean±SEM. See also Figure S4.