ATG16L1 in myeloid cells limits colorectal tumor growth in ApcMin/+ mice infected with colibactin-producing Escherichia coli via decreasing inflammasome activation

Abstract

Escherichia coli strains producing the genotoxin colibactin, designated as CoPEC (colibactin-producing E. coli), have emerged as an important player in the etiology of colorectal cancer (CRC). Here, we investigated the role of macroautophagy/autophagy in myeloid cells, an important component of the tumor microenvironment, in the tumorigenesis of a susceptible mouse model infected with CoPEC. For that, a preclinical mouse model of CRC, the Apc^Min/+ mice, with Atg16l1 deficiency specifically in myeloid cells (Apc^Min/+/Atg16l1[∆MC]) and the corresponding control mice (Apc^Min/+), were infected with a clinical CoPEC strain 11G5 or its isogenic mutant 11G5∆clbQ that does not produce colibactin. We showed that myeloid cell-specific Atg16l1 deficiency led to an increase in the volume of colonic tumors in Apc^Min/+ mice under infection with 11G5, but not with 11G5∆clbQ. This was accompanied by increased colonocyte proliferation, enhanced inflammasome activation and IL1B/IL-1β secretion, increased neutrophil number and decreased total T cell and cytotoxic CD8⁺ T cell numbers in the colonic mucosa and tumors. In bone marrow-derived macrophages (BMDMs), compared to uninfected and 11G5∆clbQ-infected conditions, 11G5 infection increased inflammasome activation and IL1B secretion, and this was further enhanced by autophagy deficiency. These data indicate that ATG16L1 in myeloid cells was necessary to inhibit colonic tumor growth in CoPEC-infected Apc^Min/+ mice via inhibiting colibactin-induced inflammasome activation and modulating immune cell response in the tumor microenvironment. Abbreviation: AOM, azoxymethane; APC, APC regulator of WNT signaling pathway; ATG, autophagy related; Atg16l1[∆MC] mice, mice deficient for Atg16l1 specifically in myeloid cells; CASP1, caspase 1; BMDM, bone marrow-derived macrophage; CFU, colony-forming unit; CoPEC, colibactin-producing Escherichia coli; CRC, colorectal cancer; CXCL1/KC, C-X-C motif chemokine ligand 1; ELISA, enzyme-linked immunosorbent assay; IL, interleukin; MC, myeloid cell; MOI, multiplicity of infection; PBS, phosphate-buffered saline; pks, polyketide synthase; qRT-PCR, quantitative real-time reverse-transcription polymerase chain reaction; siRNA, small interfering RNA; TME, tumor microenvironment; TNF/TNF-α, tumor necrosis factor.

Keywords: Autophagy; bacterial infection; colibactin-producing E. coli; colorectal cancer; toxin.

Figure 1.

MC-specific Atg16l1 deficiency led to increased colonic tumor volume in Apc^Min/+ mice infected with the CoPEC 11G5 strain. Apc^Min/+/Atg16l1[ΔMC] and control (Apc^Min/+/Atg16l1^flox/flox or Apc^Min/+) mice were treated with streptomycin for 3 days, then received H₂O for 24 h. The mice were orally administered (day 0) with PBS or with 10⁹ 11G5 or 11G5∆clbQ bacteria. Mice were sacrificed at day 65 post-administration. (A) Body weight determined at different time points was presented as percentage of body weight of the same mouse at day 0, defined as 100%. Data are means ± SEM. ^ap < 0.05 vs Apc^Min/+ + PBS; ^bp < 0.05 vs Apc^Min/+ +11G5∆clbQ; ^cp < 0.05 vs Apc^Min/+/Atg16l1[ΔMC] + PBS; ^dp < 0.05 vs Apc^Min/+/Atg16l1[ΔMC] +11G5∆clbQ. Statistical analysis was performed using two-way Anova test followed by Bonferroni post-test. (B) Representative photos of the colons taken at the day of sacrifice. The number (C) and volume (D) of colorectal tumors were determined using a binocular loupe. (E) The histological evaluation of the colonic sections was performed under blinded conditions by an expert anatomopathologist according to the criteria in Table S2. Statistical analysis was performed using one-way Anova test followed by Bonferroni post-test. *p < 0.05; **p ≤ 0.01; ***p ≤ 0.001; NS: not significant (C-E).

Figure 2.

H&E-stained colonic sections of Apc^Min/+ and Apc^Min/+/Atg16l1[ΔMC] mice infected with 11G5 or 11G5∆clbQ bacteria. Apc^Min/+/Atg16l1[ΔMC] and control (Apc^Min/+/Atg16l1^flox/flox or Apc^Min/+) mice were treated with streptomycin for 3 days, then received H₂O for 24 h. The mice were orally administered (day 0) with PBS or with 10⁹ 11G5 or 11G5∆clbQ bacteria. Mice were sacrificed at day 65 post-administration. Representative photos of H&E-stained mouse colonic sections at low (A) and high (B) magnifications. Bars: 500 μm.

Figure 3.

MC-specific Atg16l1 deficiency led to increased colonic epithelial cell proliferation in Apc^Min/+ mice under 11G5-infected condition. Apc^Min/+/Atg16l1[ΔMC] and Apc^Min/+ mice were orally administered with PBS or with 10⁹ 11G5 or 11G5∆clbQ bacteria, and sacrificed at day 65 post-administration. (A) Ccnd1 (cyclin D1) mRNA level in the non-tumoral colonic mucosa was quantified by qRT-PCR. (B) Representative images of immunohistochemical staining for MKI67/Ki67 in the colonic mucosa and tumors. Boxes show magnified fields. Bars: 100 μm. (C) Quantification of MKI67-positive cell number/crypt determined from 40 crypts/mouse. (D) Quantification of MKI67-positive cell number/mm² of tumor. Values represent means ± SEM. Statistical analysis was performed using one-way Anova test followed by Bonferroni post-test. *p < 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001.

Figure 4.

MC-specific Atg16l1 deficiency did not affect 11G5-induced DNA damage in colonic epithelial cells in Apc^Min/+ mice. Apc^Min/+/Atg16l1[ΔMC] and Apc^Min/+ mice were orally administered with PBS or with 10⁹ 11G5 or 11G5∆clbQ bacteria, and sacrificed at day 65 post-administration. (A) Representative images of immunohistochemical staining for γH2AX in the non-tumoral colonic mucosa. Boxes show magnified fields. Bars: 100 μm. (B) Quantification of γH2AX foci number/crypt determined from 40 crypts/mouse. Values represent means ± SEM. Statistical analysis was performed using one-way Anova test followed by Bonferroni posttest. **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001.

Figure 5.

Autophagy was induced in macrophages upon CoPEC infection independently of colibactin, and this was necessary to eliminate intramacrophagic CoPEC bacteria. Human THP-1 macrophages (A, B, E, F) or BMDMs prepared from wild-type (Atg16l1^flox/flox) and Atg16l1[ΔMC] mice (C, D, G) were uninfected or infected at a MOI of 100 with one of the following strains: the nonpathogenic commensal E. coli strain (MG1655 or HS), the clinical CoPEC 11G5 strain, the isogenic mutant 11G5∆clbQ, the transcomplemented 11G5 strain. After 10 min of centrifugation at 1 000 g and a 10 min of incubation at 37°C and 5% CO₂, the infected macrophages were washed three times with PBS and incubated with the culture media containing 100 μg/ml of gentamicin for 40 min, and then changed to culture media containing 50 μg/mL of gentamicin for 8 h. Representative western blot analysis (A, C) and quantification of LC3-II:ACTB/β-actin band intensity from 3 independent blots (B, D). (E) 40 μg/mL rapamycin or vehicle (DMSO) was added together with infection. (F) Cells were transfected with 70 nM of control siRNA or ATG16L1 siRNA for 2 days before infection. (E-G) to determine the number of intracellular bacteria, the cells were lysed with 1% Triton X-100 in deionized water and plated onto LB agar plates. After 24 h at 37°C, the bacterial CFU were counted. (B, D) (*p < 0.05; **p ≤ 0.01; ***p ≤ 0.001) vs uninfected condition; (^#p < 0.05; ^##p ≤ 0.01) vs HS-infected condition; (^$p < 0.05; ^$$p ≤ 0.01) vs MG1655-infected condition. (E-G) *p < 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001. Values represent means ± SEM. Statistical analysis was performed using one-way Anova test followed by Bonferroni posttest.

Figure 6.

Atg16l1 deficiency in macrophages led to increases in cleaved CASP1 and secreted IL1B amount upon CoPEC infection dependently of colibactin. BMDMs prepared from wild-type (Atg16l1^flox/flox) and Atg16l1[ΔMC] mice were infected with MG1655, 11G5, 11G5ΔclbQ or the transcomplemented 11G5 strain at a MOI of 100. After 10 min of centrifugation at 1 000 g and a 10 min of incubation at 37°C and 5% CO₂, the infected macrophages were washed three times with PBS and incubated with the culture media containing 100 μg/mL of gentamicin for 40 min, and then changed to culture media containing 50 μg/mL of gentamicin for 8 h. (A-D) Secreted IL6, IL1B, TNF and CXCL1 levels in cell culture supernatant were quantified by ELISA. Representative western blot analysis (E) and quantification of cleaved CASP1 (p10 subunit):ACTB/β-actin band intensity from 3 independent blots (F). (A-D) (*p < 0.05; **p ≤ 0.01; ***p ≤ 0.001) vs uninfected condition; (^#p < 0.05; ^##p ≤ 0.01: ^###p ≤ 0.001) vs MG1655-infected condition; ^$p < 0.05; ^$$p ≤ 0.01; ^$$$p ≤ 0.001; p ≤ 0.0001. (F) ^$$p ≤ 0.01; ^$$$$p ≤ 0.0001. Values represent means ± SEM. Statistical analysis was performed using one-way Anova test followed by Bonferroni posttest.

Figure 7.

MC-specific Atg16l1 deficiency increased secreted IL1B amount in the colon of Apc^Min/+ mice upon CoPEC infection dependently of colibactin. Apc^Min/+/Atg16l1[ΔMC] and Apc^Min/+ mice were orally administered with PBS or with 10⁹ 11G5 or 11G5∆clbQ bacteria, and sacrificed at day 65 post-administration. Secreted IL1B, IL6, TNF and CXCL1 amounts in the colonic culture supernatant were quantified by ELISA. Values represent means ± SEM. Statistical analysis was performed using one-way Anova test followed by Bonferroni posttest. *p < 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001.

Figure 8.

MC-specific Atg16l1 deficiency increased cleaved CASP1 levels in the colon of Apc^Min/+ mice upon CoPEC infection dependently of colibactin. Apc^Min/+/Atg16l1[ΔMC] and Apc^Min/+ mice were orally administered with PBS or with 10⁹ 11G5 or 11G5∆clbQ bacteria, and sacrificed at day 65 post-administration. Representative images of immunofluorescent staining for cleaved CASP1 (red) in the non-tumoral colonic mucosa (A) and tumors (B). Nuclei were stained with Hoechst (blue). Boxes show magnified fields. Bars: 200 μm. Quantification of cleaved CASP1 fluorescence intensity, normalized to Hoechst intensity, in the colonic mucosa (C) and tumor (D), was presented as a percentage of the value defined for the Apc^Min/+ + PBS group (100%). Values represent means ± SEM. Statistical analysis was performed using one-way Anova test followed by Bonferroni posttest. *p < 0.05; ***p ≤ 0.001; ****p ≤ 0.0001.

Figure 9.

MC-specific Atg16l1 deficiency decreased CD3⁺ total T cell and cytotoxic CD8⁺ T cell numbers in the colon of Apc^Min/+ mice upon CoPEC infection dependently of colibactin. Apc^Min/+/Atg16l1[ΔMC] and Apc^Min/+ mice were orally administered with PBS or with 10⁹ 11G5 or 11G5∆clbQ bacteria, and sacrificed at day 65 post-administration. Representative images of CD8 (red) and CD3 (green) immunofluorescent double staining in the colonic mucosa (A) and tumors (B). Nuclei were stained with Hoechst (blue). Bars: 200 μm. Boxes show magnified fields. Quantification of CD3⁺ total T cell and CD8⁺ CD3⁺ T cell number/mm² of colonic mucosa (C) or tumor (D). Values represent means ± SEM. Statistical analysis was performed using one-way Anova test followed by Bonferroni posttest. *p < 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001.

Figure 10.

MC-specific Atg16l1 deficiency increased neutrophil infiltration in the colonic mucosa and tumor of Apc^Min/+ mice upon CoPEC infection dependently of colibactin. Apc^Min/+/Atg16l1[ΔMC] and Apc^Min/+ mice were orally administered with PBS or with 10⁹ 11G5 or 11G5∆clbQ bacteria, and sacrificed at day 65 post-administration. Representative images of LY6G (red) and ITGAM (green) immunofluorescent double staining in the colonic mucosa (A) and tumors (B). Nuclei were stained with Hoechst (blue). Bars: 100 μm. Boxes show magnified fields. Quantification of LY6G⁺ ITGAM⁺ neutrophil number/mm² of colonic mucosa (C) or tumor (D). Statistical analysis was performed using one-way Anova test followed by Bonferroni posttest. **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001.

Figure 11.

Proposed model for the role of ATG16L1 in MCs in colorectal carcinogenesis associated with CoPEC colonization. ATG16L1 in MCs is necessary to inhibit the colonic tumor growth in a mouse model of CRC infected with CoPEC. Mechanistically, ATG16L1 dysfunction in MCs leads to increased colorectal tumor volume, which is accompanied with increased colibactin-induced inflammasome activation and IL1B secretion, increased neutrophil infiltration and decreased anti-tumor CD8⁺ T cells in the colonic mucosa and tumors, and increased colonic epithelial cell and cancer cell proliferation.