The AIEC reference strain LF82 was isolated from a chronic ileal lesion of a CD patient (Darfeuille-Michaud et al., 1998). The clinical AIEC strains CEA501S and CEA614S were isolated from the ileal lesions of CD patients from the CEALIVE cohort (Buisson et al., 2021). The LF82-GFP bacteria was used to visualize the bacteria by fluorescent microscopy as previously described (Nguyen et al., 2014; Bretin et al., 2016). The mCherry-labeled AIEC LF82 strain was obtained by bacterial conjugation in solid medium. Briefly, the mCherry donor strain, the AIEC LF82 recipient strain and the strain carrying the transposase were grown overnight in Luria-Broth (LB) with corresponding antibiotics at 37°C. These 3 strains were mixed together, centrifuged for 2 min at 1,900 g, re-suspended in 1.5 mL of LB to eliminate antibiotics and centrifuged again. Then, the pellet was re-suspended in 50 µL of LB, and this solution was filtered through a 0.45 µm filter membrane on LB agar plate. The plate was incubated for 2 to 6 hours at 37°C, and then, the filter membrane was transferred in a 50 mL tube to resuspend the bacteria in 1 mL of LB. It was spread on LB agar plates and incubated overnight at 37°C. The E. coli K12 C600 strain (Appleyard, 1954) was used as a non-invasive strain. All bacterial strains were grown overnight in LB at 37°C without agitation.

The NDP52 knock-out (KO) and p62 KO HeLa cells, as well as their corresponding control HeLa cells, which express NDP52 or p62, respectively, were kindly provided by Prof. Richard J. Youle (National Institutes of Health, Bethesda, Maryland, USA) (Lazarou et al., 2015). HeLa cells expressing the GFP-LC3 construct was kindly given by Drs. Aurore Rozières and Christophe Viret (CIRI, Centre International de Recherche en Infectiologie, Lyon, France). HeLa cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) with L-glutamine (Gibco) supplemented with 10% fetal bovine serum (FBS) (Dutscher) and 1% antibiotic/antimycotic solution (Penicillin G/Streptomycin/Amphotericin B) (Cytiva). The intestinal epithelial cell line T84 (ATCC, CCL-248) was cultured in DMEM/F-12 Nutrient Mixture (Ham) (Gibco) supplemented with 10% FBS, 1% Glutamine (Gibco), 1% antibiotic/antimycotic solution (Penicillin G/Streptomycin/Amphotericin B) and 1% Hepes (Dutscher). All cells were maintained in an atmosphere with saturated humidity and containing 5% CO₂ at 37°C.

The constructs expressing wild-type NDP52, or mutated NDP52^Val248Ala, or NDP52-GFP or NDP52^Val248Ala-GFP were generated. NDP52 cDNA was cloned from T84 cells into the pEGFP-C2 vector (Clontech) using the SuperFi II DNA polymerase (Invitrogen) and the following primers:

All transfections were performed using Lipofectamine 3000 RNAimax (Invitrogen) according to the manufacturer’s instructions. Briefly, T84 cells or NDP52 KO HeLa cells were seeded in 24-well tissue culture plates. The following day, lipofectamine 3000 reagent was diluted in Opti-MEM medium (Gibco), and nucleotide mix was prepared by mixing siRNAs (Dharmacon ON-TARGETplus SMARTPool; final concentration: 50 nM) against NDP52 (#L-010637-00-0005), Optineurin (#L-016269-00-0005), p62 (#L-010230-00-0005), NBR1 (#L-010522-00-0005) or TAX1BP1 (#L-016892-00-0005) or 500 ng of plasmid in Opti-MEM medium. Then, the lipofectamine 3000 reagent was added to the nucleotide mix and incubated for 5 min at room temperature. The mixed solution was added to the cells and incubated at 37°C in an atmosphere containing 5% CO₂. After 6 h, the transfection medium was removed, cells were washed once with PBS and incubated with antibiotic-free culture medium.

Bacterial strains were cultured overnight in LB at 37°C without agitation. Bacterial concentrations were estimated by measuring the optical density at 600 nm. Bacteria were re-suspended in a proper volume of fresh antibiotic-free culture medium to allow infection of host cells at a MOI (multiplicity of infection) of 10, 100 or 200. The number of intracellular bacteria was determined using the gentamicin protection assay as we described previously (Nguyen et al., 2014; Bretin et al., 2016). Briefly, cell monolayers were incubated with bacteria for 3 h, washed for three times with PBS 1X and then incubated with antibiotic-free culture medium containing 100 µg/mL of gentamicin for the indicated time. To determine the number of intracellular bacteria, cell monolayers were washed once with PBS and lysed with 1% Triton X-100 (Euromedex) in deionized water. Samples were serially diluted and plated onto LB agar plates to determine the number of CFU (colony-forming unit) of the bacteria.

GFP-LC3-expressing HeLa cells were seeded on coverslips in 24-well tissue culture plates 2 days before infection. Cells were infected with AIEC LF82-mCherry at a MOI of 100 and centrifugated at 160 g for 10 min to promote the contact between the bacteria and the cells. The cells were then incubated with the bacteria for 3 h at 37°C in an atmosphere containing 5% CO₂. Cells were then washed three times with PBS 1X and incubated with antibiotic-free culture media containing 100 µg/mL of gentamicin for the indicated time. Then, cells were fixed with 3.7% paraformaldhehyde (PFA) (Sigma) in PBS 1X at room temperature for 10 min, and then permeabilized with 0.1% Triton X-100 in PBS 1X at room temperature for 10 min. Cells were incubated with the saturation buffer (PBS 1X containing 5% FBS, 3% Bovine Serum Albumin (BSA) and 0.025% Triton X-100) at room temperature for 1 h. Cells were then incubated overnight at 4°C with a primary antibody diluted in the saturation buffer (rabbit anti-NDP52 antibody, #ab68588, Abcam, dilution: 1/100) in a humid box. The following day, after 6 washes with PBS 1X, cells were incubated with the corresponding secondary antibody diluted in the saturation buffer (anti-rabbit Cy5, #A10523, Invitrogen, dilution: 1/400) in a humid box for 45 min. Nuclei were stained with Hoescht. Coverslips were mounted with a Mowiol solution (Interchim). Images were taken using Zeiss LSM 800 with Airyscan confocal microscope.

NDP52 KO HeLa cells were seeded in 6-well tissue culture plates. The following day, cells were transfected with wild-type NDP52-GFP or NDP52^Val248Ala-GFP constructs, or the empty plasmid. Then, cells were infected with AIEC LF82-mCherry at a MOI of 100 for 3 h. Cells were then washed three times with PBS 1X and incubated with antibiotic-free culture media containing 100 µg/mL of gentamicin for 1 h. Cells were trypsinized, resuspended in fresh antibiotic-free culture media and centrifugated for 5 min at 130 g. The cells were finally resuspended in PBS 1X containing 1% BSA and 2 mM EDTA. The flow cytometry analysis was performed on Attune flow cytometer (Life Technologies) using the Attune NXT software.

Cell culture supernatants were collected at different time points during invasion assay (before infection and at 4 h, 10 h and 24 h post-infection). The amounts of secreted IL-8 and IL-6 levels were determined by enzyme-linked immunosorbent assay (ELISA) according to the manufacturer’s instructions (Duoset, R&D systems).

Cells were washed with PBS 1X and lysed with RIPA buffer (20 mM Tris-HCl, 150 mM NaCl, 2 mM EDTA, 1% NP-40, pH 7.4) containing protease and phosphatase inhibitor cocktail (Roche) and 20 mM Na-orthovanadate. Lysates were recovered in an eppendorf tube and disrupted for 6 min at 4°C. Then, lysates were centrifuged at 14,000 g for 15 min at 4°C, and the supernatants were used for immunoblotting. Protein concentrations were measured using the DC protein assay kit (Bio-Rad). The samples were diluted with 4X Laemmli sample buffer (277.8 mM Tris-HCl, pH 6.8, 44.4% (v/v) glycerol, 4.4% SDS, 0.02% bromophenol blue; Bio-Rad) containing β-mercaptoethanol, followed by heating for 5 min at 95°C. Whole-cell lysates were separated by acrylamide gel electrophoresis (10 µg of proteins per well), and transferred to a nitrocellulose membrane (Amersham™ Protran™ supported 0.45 µm). The membrane was then incubated with the blocking buffer (PBS 1X containing 0.1% Tween 20 and 5% BSA) for 1 h at room temperature. Then, the membrane was incubated with primary antibodies [rabbit anti-NDP52 antibody (#ab68588, Abcam, dilution: 1/1000); rabbit anti-optineurin antibody (#ab151240, Abcam, dilution: 1/1000); anti-p62/SQSTM1 (#sc-28359, Santa Cruz Biotechnology, dilution: 1/1000); anti-NBR1 (#20145, Cell Signaling Technology, dilution: 1/1000); anti-TAX1BP1 (#5105, Cell Signaling Technology, dilution: 1/1000); anti-phospho-IκB-α (#2859, Cell Signaling Technology, dilution: 1/1000); anti-β-actin rabbit antibody (#4970, Cell Signaling Technology, dilution: 1/1000)] diluted with the blocking buffer overnight at 4°C. After washes, membranes were incubated with the appropriate HRP-conjugated secondary antibodies (Cell Signaling Technology) in blocking buffer for 1 h at room temperature. After washes, blots were detected using the Clarity Western ECL Substrate (Bio-Rad) and revealed using the ChemiDOcTM XRS System (Bio-Rad).

Results were presented as means ± SEM. Statistical analyses between 2 or several groups were performed using the Student t test (Mann-Whitney if not parametric) or analysis of variance (ANOVA) followed by a post-test Bonferroni correction (Kruskal-Wallis if not parametric), respectively, with GraphPad Prism version 9.4.0 software. A P value less than 0.05 was considered statistically significant. *P < 0.05; **P ≤ 0.01; ***P ≤ 0.001; ****P ≤ 0.0001.

Some autophagy receptors, such as Optineurin, NBR1, TAX1BP1, p62 and NDP52, have been shown to participate in the elimination of invading bacteria (Lamark and Johansen, 2021). To investigate which autophagy receptors are involved in the control of intracellular AIEC LF82 by autophagy, the intestinal epithelial T84 cell line was transfected with siRNAs against Optineurin, NBR1, TAX1BP1, p62 or NDP52. The efficiency of siRNA transfection was analyzed by western blot and was shown in Supplementary Figure 1 . Among them, a significant increase in the number of intracellular AIEC LF82 was observed in cells transfected with siRNAs against p62 or NDP52, compared to cells transfected with vehicle or a siRNA control at 4, 10 and 24 h post-infection ( Figure 1A ). In the cells transfected with siRNAs against TAX1BP1, Optineurin or NBR1, no significant difference in the AIEC LF82 intracellular number was detected compared to vehicle and siRNA control conditions ( Figures 1A, B ). Together, these results suggest that p62 and NDP52 might be involved in the control of the AIEC LF82 strain by autophagy.

To determine whether p62 and NDP52 could have a synergistic effect to target AIEC, T84 cells were transfected with both siRNAs against p62 and NDP52, and the number of intracellular LF82 in these cells was compared to that in cells transfected with siRNA against p62 or siRNA against NDP52 separately. As shown in Figure 1C , a significant increase in the number of intracellular AIEC LF82 in cells transfected with both siRNAs against p62 and NDP52 compared to cells transfected with each siRNA separately was observed at 4, 10 and 24 h post-infection. This result suggested that p62 and NDP52 could have a synergistic effect to target AIEC bacteria.

To confirm the involvement of p62 in autophagy-mediated control of AIEC intracellular number, p62 KO and the control HeLa cells were infected with the AIEC LF82 strain. A significant increase in the number of intracellular AIEC LF82 was observed in p62 KO HeLa cells compared to control cells at 4, 10 et 24 h post-infection ( Figure 1D ). The same experiment was performed using NDP52 KO HeLa cells and their corresponding control cells, and the same results were obtained ( Figure 1E ). Indeed, the intracellular number of AIEC LF82 was significantly higher in NDP52 KO HeLa cells compared to control cells ( Figure 1E ). Thus, these results confirm the involvement of p62 and NDP52 in the control of AIEC LF82 intracellular number.

Next, to further investigate the role of these two receptors in the recognition of other clinical AIEC strains, p62 KO HeLa cells, NDP52 KO HeLa cells and their respective control cells were infected with two other clinical AIEC strains (CEA501S and CEA614S) isolated from the ileal mucosa of CD patients [CEALIVE cohort (Buisson et al., 2021)]. The non-pathogenic and non-invasive E. coli K12 C600 strain and the AIEC LF82 strain were also used. In p62 KO HeLa cells, the intracellular number of the AIEC CEA501S and CEA614S strains was significantly higher compared to that in control cells at 4h post-infection ( Figure 2A ). This was also observed for the AIEC LF82 strain but not the K12 C600 strain ( Figure 2A ). Similarly, the intracellular number of the CEA501S and CEA614S strains was significantly higher in NDP52 KO HeLa cells compared to control cells at 4h post-infection ( Figure 2B ). This increase was also observed for the AIEC LF82 strain, but not the K12 C600 strain ( Figure 2B ). Taken together, these results suggest the role of p62 and NDP52 in controlling the intracellular number of clinical AIEC strains.

It has been shown that autophagy receptors can bind the autophagic protein LC3 via their LC3-binding domain and can also bind ubiquitin on the bacterial surface via their ubiquitin-binding domain (Lamark and Johansen, 2021). To demonstrate that p62 and NDP52 act as autophagy receptors for AIEC, we performed immunofluorescent staining to observe a colocalization between NDP52 or p62 with AIEC LF82 bacteria and the autophagic protein LC3. Figure 3A shows a direct colocalization between LF82-GFP bacteria, p62 (blue) and LC3 (red). We also observed a colocalization between LF82-mCherry, NDP52 (magenta) and GFP-LC3 ( Figure 3B ). Together, these results demonstrate that p62 and NDP52 are autophagy receptors responsible for the recognition and targeting AIEC LF82 bacteria to LC3-positive autophagosome.

Data for bacterial invasion assay performed with the same MOI and time post-infection used for immunofluorescent staining was shown in Supplementary Figure 2 . This result showed that depletion in p62 or NDP52 led to increase in intracellular number of the AIEC LF82 strain.

It has been shown that AIEC LF82 bacteria are able to induce the production of pro-inflammatory cytokines by host cells (Nguyen et al., 2014; Bretin et al., 2016). Thus, we sought to analyze the role of NDP52 and p62 in the control of AIEC-induced pro-inflammatory cytokine production. In p62 KO HeLa cells, a significant increase in the amount of secreted IL-6 and IL-8 was detected at 4h, 10h and 24h after LF82 infection at a MOI of 10 compared to control cells ( Figures 4A, B ). Similarly, the amount of secreted IL-6 and IL-8 was increased in NDP52 KO HeLa cells compared to control cells after 4h, 10h and 24h of LF82 infection at a MOI of 10 ( Figures 4C, D ). Similar results were obtained under LF82 infection at a MOI of 100 ( Supplementary Figure 3 ). Together, these results showed that p62 and NDP52 are also required to limit AIEC LF82-induced pro-inflammatory cytokine production.

As the NDP52^Val248Ala polymorphism has been associated with an increased susceptibility to develop CD (Ellinghaus et al., 2013), we investigated its potential impact on host response to AIEC infection. For this, NDP52 KO HeLa cells were transfected with a construct that expresses the wild-type NDP52 or the mutated NDP52^Val248Ala protein. Western blot analysis showed that the level of NDP52 protein in these cells were similar ( Figure 5A ). We also examined whether expression of the NDP52^Val248Ala risk variant modifies expression of p62. As shown in Supplementary Figure S4 , p62 protein level was similar between HeLa cells expressing the wild-type NDP52 and those expressing the mutated NDP52^Val248Ala protein.

A flow cytometry experiment was performed to investigate the impact of the mutated NDP52^Val248Ala protein on the control of AIEC LF82 infection by host cells. For this, NDP52 KO HeLa cells were transfected with the GFP plasmid expressing wild-type NDP52 or the mutated NDP52^Val248Ala protein, and infected with the AIEC LF82-mCherry bacteria. Then, the percentage of mCherry-positive cells (LF82-infected cells) among GFP-positive cells (transfected cells) was determined by flow cytometry. This allowed to select only the transfected cells, and to count the transfected cells that are infected with LF82-mCherry. Our results showed that 67.3% of the cells expressing wild-type NDP52-GFP were infected, and 63% of the cells expressing NDP52^Val248Ala-GFP were infected ( Figure 5B ). The gating strategy was represented in Supplementary Figure 5 , and all control conditions (transfected with empty plasmid, uninfected) were represented in Supplementary Figure 6 .

This result suggested that expression of the mutated NDP52^Val248Ala protein did not seem to impact the percentage of LF82-infected cells. Furthermore, fluorescent microscopy analysis showed that the mutated NDP52^Val248Ala-GFP protein was able to bind to the LF82-mCherry bacteria and the autophagic protein LC3 (magenta) as the wild-type NDP52 did ( Figure 5C ). Finally, we investigated the impact of the NDP52^Val248Ala variant on AIEC LF82 intracellular replication by gentamicin protection assay. No significant difference in the AIEC LF82 intracellular number at 4, 10 and 24 h post-infection was observed between HeLa cells expressing the wild-type NDP52 or the mutated NDP52^Val248Ala protein ( Figure 5D ). Together, these results suggested that in HeLa cells, the CD-associated NDP52^Val248Ala polymorphism has no impact on AIEC recognition and targeting to autophagy, thus does not influence the control of AIEC intracellular replication.

To further examine the impact of the CD-associated NDP52^Val248Ala polymorphism on the control of AIEC LF82-induced inflammation, NDP52 KO HeLa cells were transfected with the construct expressing the wild-type NDP52 or the mutated NDP52^Val248Ala protein, and infected with the AIEC LF82 strain with a MOI of 10 ( Figure 6 ) or 100 or 200 ( Supplementary Figure 7 ). No significant difference was observed in the amount of IL-6 and IL-8 cytokines secreted by HeLa cells expressing the wild-type NDP52 and those expressing the NDP52^Val248Ala protein ( Figures 6A, B ). Furthermore, Western blot analysis showed a similar level of phospho-IκB in cells expressing the mutated NDP52^Val248Ala protein compared to those expressing the wild-type NDP52 under both uninfected and LF82-infected conditions ( Figure 6C ). These results suggested that the CD-associated NDP52^Val248Ala polymorphism has no impact on AIEC-induced inflammation in HeLa cells.