Fig 1.
Microarray-based identification of differentially expressed autophagy-related genes in response to Helicobacter hepaticus CdtB in intestinal epithelial cells.
The expression of genes was determined in HT29 intestinal cells using the Human GE 4x44K v2 Microarray Kit (Agilent Technologies) after a 72 h transduction with lentiviral particles expressing the CdtB of H. hepaticus strain 3B1 versus the tdTomato fluorescent protein (TFP) as previously described [7]. The relative expression of genes in response to CdtB is reported as a fold change versus the value for cells cultured with lentiviral particles expressing the TFP. Results are presented as the mean of 4 replicates as 4 independent transduction experiments were performed. The data presented for ATG5, HIF1A and NPC1 are the results of 40 replicates as 10 probes for each mRNA of these genes were included on the Microarray Kit. The list of genes related in autophagy to be checked in the Microarray data was first extracted from the Human Autophagy Database (HADb, http://autophagy.lu/clustering/). Then a subsequent selection based on their autophagic status annotation available in The Human Gene Database was applied in order to select the major genes involved in autophagy. The discontinuous line shows the basal rate in cells expressing TFP. Asterisks denote significant results. P1 and P2 represent the 2 probe names (S2 Table) used for mRNA quantification. Details are presented in S2 Table (name and sequence of the probes, the corresponding gene name, the genbank accession number, the locus and the transcript variant). Abbreviations: AMBRA1, Autophagy and Beclin-1 Regulator 1; ATG, Autophagy Related Gene; ATG16L, Autophagy Related 16 Like Gene; BAG3, BAG Cochaperone 3;BECN1/Vps30/ATG6, Beclin 1; CALCOCO2, Calcium Binding And Coiled-Coil Domain 2; DRAM1, DNA-damage Regulated Autophagy Modulator 1; eiF2AK3, Eukaryotic Translation Initiation Factor 2-Alpha Kinase 3; FOXO1, Forkhead Box O1; GABARAPL, GABA(A) Receptor-Associated Protein Like; HIF1A, Hypoxia Inducible Factor 1 Subunit Alpha; ITPR1, Inositol 1,4,5-Triphosphate Receptor, type 1; KIAA0226/RUBCN, Rubicon Autophagy Regulator; LAMP1, Lysosomal Associated Membrane Protein 1; LAMP2, Lysosomal Associated Membrane Protein 2; MAP1LC3A, Microtubule Associated Protein 1 Light Chain 3 Alpha; MAP1LC3B, Microtubule Associated Protein 1 Light Chain 3 Beta; MLST8, MTOR (Mechanistic Target of Rapamycin Kinase) Associated Protein, LST8 Homolog (mammalian lethal with Sec13 protein); MTMR14, MyoTubularin Related Protein 14; NPC1, NPC Intracellular Cholesterol Transporter 1 (Niemann-Pick disease, type C1); PARK2/PARKN, Parkin RBR E3 Ubiquitin Protein Ligase; PEX3, Peroxisomal Biogenesis Factor 3; PIK3C3//Vps34, Phosphoinositide-3-Kinase, class 3; PIK3R4/Vps15, Phosphoinositide-3-Kinase, Regulatory subunit 4; PINK1; Phosphatase and Tensin Homolog (PTEN) Induced Kinase 1; RAB1A, Member RAS Oncogene Family; RAB7A, Member RAS Oncogene Family; RAB24, member RAS oncogene family; RB1CC1/FIP200, RB1 (RetinoBlastoma Transcriptional Corepressor 1) Inducible Coiled-Coil 1; RGS19, Regulator of G-protein Signaling 19; RPS6KB1, Ribosomal Protein S6 Kinase B1; RPTOR, Regulatory Associated Protein Of MTOR Complex 1; SESN2, Sestrin 2; SH3GLB1/BIF1, SH3 Domain Containing Growth Factor Receptor Bound Protein 2 (GRB2) Like, Endophilin B1; SIRT2, Sirtuin 2; SPNS1, Sphingolipid Transporter 1 (Putative); P62/SQSTM1, Sequestosome 1; STK11; Serine/Threonine Kinase 11; TMEM49/VMP1, Vacuole Membrane Protein 1; TP53INP2, Tumor Protein P53 Inducible Nuclear Protein 2; ULK, unc-51-like kinase; WD, tryptophan-aspartic acid; WDR45/WIPI4, WD (tryptophan-aspartic acid) Repeat Domain 45; WIPI, WD Repeat Domain, Phosphoinositide Interacting; ZFYVE1/DFCP1, Zinc Finger FYVE-Type Containing 1.
Fig 2.
Effects of Helicobacter hepaticus cytolethal distending toxin on LC3 expression in human intestinal and hepatic epithelial cells.
A) Analysis of LC3 puncta in hepatic Hep3B cells infected for 3 days with H. hepaticus and its corresponding ΔCDT mutant strain. These cells were processed for fluorescent labeling of LC3 (red) and a counterstaining with DAPI (blue). Fluorescent staining was observed using wide field fluorescence imaging. Transgenic HT29 (B) and Hep3B (C) cells were cultivated with doxycycline for 72 h to induce the expression of the control Red Fluorescent Protein (RFP), the CdtB of H. hepaticus strain 3B1 or the CdtB of H. hepaticus strain 3B1 with the H265L mutation which has no catalytic activity [21]. These cell lines were also treated with bafilomycin A1 (30 nM) or chloroquine (30 μM) 48 h and 24 h after doxycycline induction for a duration of 24 h and 48 h, respectively. Then, cells were processed for fluorescent staining with primary antibodies generated against LC3 associated with fluorescent labeled-secondary antibodies (green) and DAPI to counterstain the nuclei (blue) (S2A, S2B Fig). Autophagic flux was also measured in those cells expressing the tandem-tagged mCherry-GFP-LC3 protein with subsequent yellow (mCherry+/GFP+) and red (mCherry+/GFP-) dot/puncta counting (yellow dots) (S2C Fig) [18]. The number of fluorescent LC3 puncta was quantified using the "Find Maxima" function of ImageJ. The results are presented as the mean in one representative experiment (performed in triplicate) out of three. A minimum of 500 cells were measured. B) Analysis of LC3 puncta in transgenic HT29 cells. B1) RFP-, CdtB- and H265L-expressing cells. B2) CdtB-expressing cells treated with bafilomycin A1 or chloroquine. B3) autophagic flux measured in CdtB- and H265L-expressing cells. C) Analysis of LC3 puncta in transgenic Hep3B cells. C1) RFP-, CdtB- and H265L-expressing cells. C2) CdtB-expressing cells treated with bafilomycin A1 or chloroquine. C3) autophagic flux measured in CdtB- and H265L-expressing cells. *p<0.05, **p< 0.01, ***p< 0.001. Abbreviations: Baf., bafilomycin A1; CdtB, CdtB of H. hepaticus strain 3B1; CQ, chloroquine; DAPI, 4′, 6′-diamidino-2-phenylindol; ΔCDT, CDT isogenic mutant of H. hepaticus strain 3B1; H265L, H. hepaticus CdtB with the mutation His→Leu at residue 265 involved in catalytic activity; NI, non-infected; RFP, Red fluorescent protein; WT, H. hepaticus strain 3B1 = wild type strain.
Fig 3.
Effects of Helicobacter hepaticus cytolethal distending toxin on LC3, P62/SQSTM1 and AMPK expression in human epithelial cells.
Transgenic HT29 and Hep3B cells were cultivated with doxycycline for 72 h to induce the expression of the control Red Fluorescent Protein (RFP), the CdtB of H. hepaticus strain 3B1 or the CdtB of H. hepaticus strain 3B1 with the H265L mutation which has no catalytic activity [21]. Then, cells were processed for western blot analysis or fluorescent staining with primary antibodies generated against P62/SQSTM1 associated with fluorescent labeled-secondary antibodies (green) and DAPI to counterstain the nuclei (blue) (S2A Fig). The number of fluorescent P62/SQSTM1 bodies was quantified using the "Find Maxima" function of ImageJ. The results are presented as the mean in one representative experiment (performed in triplicate) out of three. A minimum of 500 cells were measured. A) Time-course western blot analysis of the protein expression level of LC3 and P62/SQSTM1 in response to the CdtB of H. hepaticus in HT29 cells. B) The protein expression level and phosphorylation status of P62/SQSTM1 and AMPK in response to RFP, CdtB and H265L were analyzed by western blot in transgenic HT29 cells. (C) Quantification of P62/SQSTM1 bodies in transgenic HT29 cells. (D) Quantification of P62/SQSTM1 bodies in transgenic Hep3B cells. E) Confocal image of Hep3B transgenic cells expressing the CdtB of H. hepaticus strain 3B1 (72 h) processed for P62/SQSTM1 fluorescent staining (green) and DAPI (blue). Scale bar, 10 μm. *p<0.05, **p< 0.01, ***p< 0.001. Abbreviations: AMPK, AMP-activated protein kinase; CdtB, CdtB of H. hepaticus strain 3B1; DAPI, 4′, 6′-diamidino-2-phenylindol; H265L, H. hepaticus CdtB with the mutation His→Leu at residue 265 involved in catalytic activity; P-AMPK, phosphorylated AMP-activated protein kinase; P62, P62/SQSTM1; P-P62, phosphorylated P62/SQSTM1; RFP, Red fluorescent protein; Tub., tubulin.
Fig 4.
Effect of Helicobacter hepaticus CdtB expression on LC3 and P62 expression in engrafted cells.
HT29- and Hep3B-transgenic cell lines were engrafted into immunodeficient mice as previously reported [21]. Three μm-tissue sections of the xenograft-derived tumors were prepared from formalin-fixed paraffin-embedded tissues and submitted to standard hematoxylin staining and immunostaining raised against LC3 (A) and P62/SQSTM1 (B). Boxes correspond to enlargement. The yellow, black, green and red arrows represent the murine infiltrates, the giant cells, the mucins (HT29) and the LC3 puncta or P62/SQSTM1 bodies, respectively. Quantification was performed by using the ‘Threshold’ function of ImageJ (v. 1.52n). Scale bar, 50 μm. **p< 0.01. Scale bar, 50 μm. Abbreviations: CdtB of H. hepaticus strain 3B1; P62, P62/SQSTM1; RFP, Red fluorescent protein.
Fig 5.
Effect of Helicobacter hepaticus CdtB expression on nucleus remodeling.
Wide field image of Hep3B transgenic cells expressing the CdtB of H. hepaticus strain 3B1 processed for the nuclear lamina (red) and P62/SQSTM1 fluorescent staining (green), and DAPI to counterstain the nuclei (blue). Yellow arrows indicate DAPI-lacking nucleoplasmic reticulum enclosing P62/SQSTM1 bodies. Boxes correspond to enlargement. Scale bar, 10 μm. Abbreviations: DAPI, 4′, 6′-diamidino-2-phenylindol; P62, P62/SQSTM1.
Fig 6.
Effects of Helicobacter hepaticus CdtB on P62/SQSTM1 localization.
As previously demonstrated, NR formation is primarily observed in response to CDT intoxication, via its active CdtB subunit [6]. Thus, images of non-infected cells are not presented below. (A) Hep3B and (B) SW480 Transgenic cells were cultivated with doxycycline for 72 h to induce the expression of the CdtB of H. hepaticus strain 3B1 [21]. Then cells were processed for staining with primary and fluorescent secondary antibodies: P62/SQSTM1 (red), UNR (green) and DAPI to counterstain the nuclei (blue). Subsequent quantification of P62/SQSTM1 in nucleoplasm, cytoplasm and foci were performed using capture of fluorescent staining (confocal imaging) by measuring the pixel intensity with the “Plot Profile” function of ImageJ (v. 1.52n), each count being performed on 100 NRs. The relative expression rate of P62/SQSTM1 in NR in response to the CdtB was reported as fold increase versus the expression in the cytosol. Scale bar, 20 μm. ***p<0.0001. Abbreviations: Cyto., cytoplasm; DAPI, 4′, 6′-diamidino-2-phenylindol; NR, nucleoplasmic reticulum; Nuc., nucleus; P62, P62/SQSTM1.
Fig 7.
Effects of bacterial genotoxins on P62/SQSTM1 bodies localization and γH2AX in intestinal and hepatic cell lines.
HT29 and Hep3B cells were infected for 3 days with CDT-secreting H. hepaticus or colibactin-secreting extra-intestinal pathogenic E. coli. Then, cells were processed for fluorescent staining with primary antibodies generated against γH2AX (red), P62/SQSTM1 (green), associated with fluorescent-labeled secondary antibodies and DAPI to counterstain the nuclei (blue). Fluorescent staining was observed using wide field fluorescence imaging. (A) Images of HT29 cells: P62/SQSTM1 (green), DAPI (blue). (B) Images of HT29 cells: γH2AX (red), P62/SQSTM1 (green), DAPI (blue). (C) Images of Hep3B cells: γH2AX (red), P62/SQSTM1 (green), DAPI (blue). Scale bars, 20 μm. Yellow arrowheads indicate extra-nuclear structures containing chromatin and P62/SQSTM1 aggregates. Yellow and pink boxes correspond to enlargement in which the DAPI and γH2AX were overexposed, respectively, in order to see the micronucleus-like structures. White boxes on the right correspond to enlargement of the merge. Abbreviations: DAPI, 4′, 6′-diamidino-2-phenylindol.
Fig 8.
Effects of bacterial genotoxins on P62/SQSTM1 bodies localization and LC3 in hepatic cell lines.
Hep3B cells were processed as in Fig 7. Fluorescent staining was observed using wide field fluorescence imaging. (A) Images of Hep3B cells: γH2AX (red), LC3 (green), DAPI (blue). (B) Images of Hep3B cells: nuclear lamina (red), P62/SQSTM1 (green), DAPI (blue). Scale bars, 20 μm. Yellow arrowheads indicate extra-nuclear structures containing chromatin. Blue arrowheads indicate P62/SQSTM1 aggregates tightly connected all along the nuclear membrane. Yellow boxes correspond to enlargement of micronucleus-like structures with DAPI. White boxes on the right correspond to enlargement of the merge. Abbreviations: DAPI, 4′, 6′-diamidino-2-phenylindol; NR, nucleoplasmic reticulum.
Fig 9.
Time-course analysis of the effects of ATG5 and ATG7 silencing on Helicobacter hepaticus CDT-induced effects.
A) CdtB-transgenic Hep3B cell line cultivated with doxycycline for 72 h to induce the expression of the CdtB of H. hepaticus strain 3B1. Cells were also treated with bafilomycin A1 (30 nM) or chloroquine (30 μM) 48 h and 24 h after doxycycline induction for a duration of 24 h and 48 h, respectively. Then cells were processed for fluorescent staining with primary antibodies generated against UNR (red) and P62/SQSTM1 (green) associated with fluorescent labeled-secondary antibodies and DAPI to counterstain the nuclei (blue). Quantification of UNR-NR positive cells (%) was performed on a minimum of 500 nuclei. The results are presented as the mean in one representative experiment (performed in triplicate) out of three. B) to G) Mock-KO, ATG5-KO and ATG7-KO Hep3B cells were infected for 3 days with H. hepaticus and its corresponding ΔCDT mutant strain. Then, the medium was removed, new medium was added and incubation continued until 8 days. These cells were processed daily for fluorescent staining with DAPI to detect the nucleus and fluorescent primary and secondary antibodies targeting γH2AX, UNR, cleaved caspase-3, and P62/SQSTM1. Fluorescent staining was observed using wide field fluorescence imaging (Fig 10). The results are presented as the mean in one representative experiment (performed in triplicate) out of three. A minimum of 500 cells were measured. (B) Nucleus surface (area) was quantified by isolating the DAPI fluorescence for each nucleus by using the ‘Threshold’ function of ImageJ (v. 1.52n). Nucleus size was measured in viable and early apoptotic cells. (C) γH2AX foci quantification was performed in viable and early apoptotic cells by measuring the pixel intensity with the “Integrated density” measure function of ImageJ (v. 1.52n). (D) The percentage of cells presenting UNR-NR was determined by manually counting the number of nuclei displaying UNR spots in the nucleoplasm. (E) Cell number quantification was performed manually. (F) Caspase-3-positive cells were quantified by counting the caspase 3 positive cells on 10 fields. *p<0.05, **p<0.01, ***p<0.001. Abbreviations: AU, arbitrary unit; Baf., bafilomycin A1; CQ, chloroquine; ΔCDT, CDT isogenic mutant of H. hepaticus strain 3B1; DAPI, 4′, 6′-diamidino-2-phenylindol; KO, Knock-Out, NR, nucleoplasmic reticulum; P62, P62/SQSTM1.
Fig 10.
Images of the effects of ATG5 and ATG7 silencing on Helicobacter hepaticus CDT-induced effects.
(Fig 9 continued) Mock-KO, ATG5-KO and ATG7-KO Hep3B cells were processed as in Fig 9B–9G. Cells were stained for nuclei (DAPI), as well as for γH2AX, UNR, cleaved caspase-3, and P62/SQSTM1 (red), along with DAPI to counterstain the nuclei (blue). Wide field images of cocultures experiment at days 4 are presented. Scale bars: 50 μm for cell number (DAPI only), 100 μm for γH2AX and DAPI double staining, and 20 μm for other double staining (UNR, cleaved caspase-3, and P62/SQSTM1 with DAPI). *p<0.05, **p<0.01, ***p<0.001. Abbreviations: ΔCDT, CDT isogenic mutant of H. hepaticus strain 3B1; DAPI, 4′, 6′-diamidino-2-phenylindol; KO, Knock-Out, P62, P62/SQSTM1.
Fig 11.
Analysis of the effects of ATG5 and ATG7 silencing on Escherichia coli colibactin-induced effects.
Mock-KO, ATG5-KO and ATG7-KO Hep3B cells were infected for 4 hours with colibactin-secreting extra-intestinal pathogenic E. coli and its corresponding isogenic mutant and cultivated for 3 days in a bacteria free medium. Then, cells were processed for fluorescent staining with primary antibodies generated against LC3 and γH2AX associated with fluorescent labeled-secondary antibodies (green) and DAPI to counterstain the nuclei (blue). (A) Wide field images of Hep3B cells: LC3 (green), DAPI (blue). Scale bar, 20 μm. (B) The number of fluorescent LC3 puncta was quantified using the "Find Maxima" function of ImageJ. (C) Nucleus surface (area) was quantified in viable and early apoptotic cells by isolating the DAPI fluorescence for each nucleus by using the ‘Threshold’ function of ImageJ (v. 1.52n). (D) γH2AX foci quantification was performed in viable and early apoptotic cells by measuring the pixel intensity with the “Integrated density” measure function of ImageJ (v. 1.52n). A minimum of 500 cells were measured. *p<0.05, **p<0.01, ***p<0.001. Abbreviations: AU, arbitrary unit; DAPI, 4′, 6′-diamidino-2-phenylindol; KO, Knock-Out, pks-, bacterial artificial chromosome vector; pks+, bacterial artificial chromosome vector with pks island encoding colibactin.
Fig 12.
Analysis of the effects of ATG5 and ATG7 silencing on Escherichia coli colibactin-induced effects (Fig 11 continued) Mock-KO, ATG5-KO and ATG7-KO Hep3B cells were processed as in Fig 11.
Then, cells were processed for fluorescent staining with primary antibodies generated against UNR, cleaved Caspase-3 or P62/SQSTM1 or γH2AX associated with fluorescent labeled-secondary antibodies and DAPI to counterstain the nuclei. (A) The percentage of cells presenting UNR-NR was determined by manually counting the number of nuclei displaying UNR spots in the nucleoplasm. (B) Cell number quantification was performed manually. (C) Caspase-3-positive cells were quantified by counting the caspase 3 positive cells on 10 fields. (D) P62/SQSTM1 bodies were quantified using the "Find Maxima" function of ImageJ. The results are presented as the mean in one representative experiment (performed in triplicate) out of three. (E) Quantification of micronucleus-like structures in Hep3B cells was performed manually. (F) Quantification of Hep3B cells with colocalized γH2AX-foci and P62/SQSTM1-bodies was performed manually. A minimum of 500 cells were measured. *p<0.05, **p<0.01, ***p<0.001. Abbreviations: AU, arbitrary unit; DAPI, 4′, 6′-diamidino-2-phenylindol; KO, Knock-Out, NR, nucleoplasmic reticulum; P62, P62/SQSTM1; pks-, bacterial artificial chromosome vector; pks+, bacterial artificial chromosome vector with pks island encoding colibactin.
Fig 13.
Effects of DNA damaging agents on nuclear remodeling and autophagy.
Hep3B cells were cultivated in the presence of etoposide (5 μM) or steptozocin (10 mM) for a duration of 24 h. Then, the medium was removed and incubation was continued for 48 h. The cells were then stained with fluorescent primary and secondary antibodies targeting LC3 (red)/γH2AX (green), P62/SQSTM1 (red)/UNR (green) and DAPI to counterstain the nuclei (blue). Fluorescent staining was observed using wide field fluorescence imaging. γH2AX foci quantification was performed by measuring the pixel intensity with the “Integrated density” measure function of ImageJ (v. 1.52n). LC3 puncta and P62/SQSTM1 bodies were quantified using the "Find Maxima" function of ImageJ. The percentage of cells presenting UNR-NR was determined by manually counting the number of nuclei displaying UNR spots in the nucleoplasm. Quantification was performed on a minimum of 500 cells. The results are presented as the mean in one representative experiment (performed in triplicate) out of three. ***p<0.001. Scale bar, 20 μm. White arrow indicates DAPI-lacking large aggregate positive for LC3 and γH2AX foci. Yellow arrow indicates DAPI-lacking nucleoplasmic reticulum enclosing UNR and P62/SQSTM1 bodies. Abbreviations: AU, arbitrary unit; DAPI, 4′, 6′-diamidino-2-phenylindol; ETP, etoposide; NR, nucleoplasmic reticulum; P62, P62/SQSTM1; STZ, steptozocin.