|Home | About | Journals | Submit | Contact Us | Français|
During infection, enteropathogenic Escherichia coli (EPEC) translocates effector proteins directly into the cytosol of infected enterocytes using a type III secretion system (T3SS). Once inside the host cell, these effector proteins subvert various immune signaling pathways, including death receptor-induced apoptosis. One such effector protein is the non-locus of enterocyte effacement (LEE)-encoded effector NleB1, which inhibits extrinsic apoptotic signaling via the FAS death receptor. NleB1 transfers a single N-acetylglucosamine (GlcNAc) residue to Arg117 in the death domain of Fas-associated protein with death domain (FADD) and inhibits FAS ligand (FasL)-stimulated caspase-8 cleavage. Another effector secreted by the T3SS is NleF. Previous studies have shown that NleF binds to and inhibits the activity of caspase-4, -8, and -9 in vitro. Here, we investigated a role for NleF in the inhibition of FAS signaling and apoptosis during EPEC infection. We show that NleF prevents the cleavage of caspase-8, caspase-3, and receptor-interacting serine/threonine protein kinase 1 (RIPK1) in response to FasL stimulation. When translocated into host cells by the T3SS or expressed ectopically, NleF also blocked FasL-induced cell death. Using the EPEC-like mouse pathogen Citrobacter rodentium, we found that NleB but not NleF contributed to colonization of mice in the intestine. Hence, despite their shared ability to block FasL/FAS signaling, NleB and NleF have distinct roles during infection.
During infection, the extracellular enteric bacterial pathogens enteropathogenic Escherichia coli (EPEC), enterohemorrhagic E. coli (EHEC), and Citrobacter rodentium utilize a type III secretion system (T3SS) to translocate virulence (effector) proteins directly into the cytosol of infected enterocytes (1). The genes encoding the T3SS, as well as several of these effectors, are encoded on the locus of enterocyte effacement (LEE) pathogenicity island and are required for the formation of intestinal attaching/effacing (A/E) lesions that are characteristic of these pathogens (2). A/E lesions are characterized by intimate attachment of the bacteria to the surface of the host cell and cytoskeletal rearrangements leading to pedestal formation (3, 4).
EPEC, EHEC, and C. rodentium also translocate many effector proteins encoded outside the LEE, termed non-LEE-encoded (Nle) effectors (5). Several of these effectors are involved in suppressing the host innate immune response by subverting various inflammatory pathways, including NF-κB, mitogen-activated protein kinase (MAPK), and FAS signaling (6, 7). For example, the EPEC effector protein NleE is a cysteine methyltransferase that inhibits NF-κB signaling by modifying host adapter proteins TAB2 and TAB3, preventing their recruitment to tumor necrosis factor (TNF)-stimulated signaling complexes (8,–10). NleC is a zinc metalloprotease that also functions to inhibit inflammatory signaling by directly cleaving the NF-κB subunit p65 (11,–14). Together, these effectors synergistically reduce the production of inflammatory cytokines, such as interleukin-8 (IL-8), produced in response to EPEC infection.
The EPEC effector protein NleB1 subverts Fas ligand (FasL)-induced apoptosis (15). Upon engagement of FAS with FasL, receptor multimerization occurs, and the death domain (DD)-containing adapter protein Fas-associated protein with death domain (FADD) is recruited through heterotypic DD interactions. Procaspase-8 is then recruited to the complex via homotypic death effector domain (DED) interactions with FADD to form the death-inducing signaling complex (DISC). At the DISC, caspase-8 undergoes dimerization and autoproteolytic processing. Active caspase-8 acts on a range of cytosolic substrates, including receptor-interacting serine/threonine protein kinase 1 (RIPK1) and caspase-3, an executioner caspase of apoptosis (16).
NleB1 was recently identified as a novel glycosyltransferase that modifies a conserved arginine residue in death domain-containing host proteins, including FADD, tumor necrosis factor receptor type 1-associated death domain protein (TRADD), and RIPK1. Inhibition of FasL-induced apoptosis by NleB1 is mediated by the transfer of a single N-acetylglucosamine (GlcNAc) sugar moiety to Arg117 in the death domain of FADD. This modification renders FADD unable to interact with FAS, thereby preventing assembly of the DISC and caspase-8 cleavage. The enzymatic activity of NleB1 depends on a DXD221–223 catalytic motif, as well as Tyr219 and Glu253 (15, 17,–19). The importance of FAS signaling in the immune response to C. rodentium infection was demonstrated in vivo, where C57BL/6 mice deficient in either FasL or FAS were unable to efficiently clear infection (15).
Another effector secreted by the T3SS of A/E pathogens is NleF. The amino acid sequence of NleF is 100% conserved between EPEC O127:H6 (strain E2348/69) and EHEC O157:H7 (strains EDL933 and Sakai), and this sequence is 85% identical to that of NleF from C. rodentium (20, 21). Caspase-9 was initially identified as an interacting partner of NleF in a yeast-2-hybrid (Y2H) screen (22). Analysis of the crystal structure of NleF bound to caspase-9 revealed that the four C-terminal residues of NleF (Leu186, Gln187, Cys188, and Gly189; referred to as the LQCG motif) occupied the active caspase-9 catalytic pocket. Further analysis of NleF and various caspases identified caspase-4 and caspase-8 as additional NleF interacting partners. The ability of these caspases to cleave synthetic substrates was inhibited in the presence of purified NleF (22). A further study confirmed the interaction of NleF with caspase-4 and the murine orthologue caspase-11 and subsequently demonstrated that this interaction led to the inhibition of IL-18 maturation and secretion from infected intestinal epithelial cells (IECs) (23).
Given the importance of FAS signaling in gut immune defense and the ability of NleF to inhibit caspase-8 activity, we investigated the importance of direct caspase-8 inhibition by NleF to EPEC pathogenesis. Here, we show that the effector protein NleF can inhibit FasL-induced caspase-8 cleavage and subsequent cell death. We demonstrate that during EPEC infection, NleB1 and NleF have complementary activities and can act synergistically to promote cell survival. However, in vivo experiments using C. rodentium suggested that the roles of NleB and NleF were distinct rather than synergistic.
Previously, NleF was shown to inhibit caspase-8 activity (22). In order to compare the roles of the EPEC effector proteins NleB1 and NleF in the inhibition of FasL-induced cell death, lentiviral transduction was employed to create HeLa cell lines that stably express Flag-NleB1, an NleB1 catalytic mutant (Flag-NleB1DXD), Flag-NleF, or an NleF truncation mutant (Flag-NleFΔLQCG) under the control of a doxycycline-dependent promoter. Expression was assessed by immunoblotting over a 24-h period, and robust doxycycline induction of the Flag-tagged fusion proteins was observed (Fig. 1A). A high-molecular-mass product was detected upon the expression of Flag-NleF, as observed previously (Fig. 1A) (23).
To test for inhibition of FAS signaling, the expression of Flag-NleB1, Flag-NleB1DXD, Flag-NleF, and Flag-NleFΔLQCG was induced in the cell lines for 24 h before stimulation with FasL. As determined by immunoblot analysis, the expression of Flag-NleB1 but not the catalytically inactive form (Flag-NleB1DXD) inhibited FasL-induced caspase-8, caspase-3, and RIPK1 cleavage (Fig. 1B). NleF also inhibited the cleavage of caspase-8, caspase-3, and RIPK1, but the expression of NleFΔLQCG had no effect on FasL-induced cleavage of any of these host proteins (Fig. 1B). NleB1 and NleF, but not their inactive mutants, also inhibited cell death, as measured by propidium iodide (PI) uptake compared to the PI uptake in HeLa cells treated with FasL alone (Fig. 2A and andB).B). Untreated HeLa cells displayed a low level of PI staining, which was unaffected by the addition of doxycycline (Fig. 2A and andBB).
In order to observe a functional role for NleF during infection, HeLa cells were infected with various EPEC derivatives prior to stimulation with FasL. As with previous experiments, caspase activity was measured by immunoblotting and cell death was measured by PI staining. Consistent with previous findings (15), cells infected with wild-type EPEC were protected from caspase-8 and caspase-3 processing in response to FasL (Fig. 3). Neither the T3SS mutant (ΔescN) nor the nleB1 mutant (ΔnleB1) was able to block caspase cleavage in response to FasL (Fig. 3), and as expected, this phenotype was restored upon complementation of the ΔnleB1 mutant with a plasmid encoding nleB1 (pNleB1) but not with a plasmid encoding catalytically inactive nleB1 (pNleB1DXD). Cells infected with the NleF deletion mutant (ΔnleF) showed a modest increase in caspase processing compared to the level in wild-type EPEC-infected cells (Fig. 3), and this was restored to wild-type levels with the reintroduction of full-length NleF but not the truncated NleFΔLQCG (Fig. 3).
To compare the relative contributions of NleB1 and NleF to inhibition of FasL-induced cell death, we constructed an nleB1 nleF double mutant (ΔnleB1 ΔnleF). HeLa cells infected with the ΔnleB1 ΔnleF mutant displayed significant increases in caspase-8 and -3 cleavage upon FasL stimulation (Fig. 3), and this phenotype was complemented by the reintroduction of either nleB1 [ΔnleB1 ΔnleF(pNleB1)] or nleF [ΔnleB1 ΔnleF(pNleF)] (Fig. 3) but not their inactive mutants [ΔnleB1 ΔnleF(pNleB1DXD) and ΔnleB1 ΔnleF(pNleFΔLQCG)]. In addition, we examined cell death as determined by the PI uptake of cells infected with derivatives of EPEC E2348/69. As expected, cells infected with wild-type EPEC were protected from FasL-induced cell death compared to the FasL-induced cell death in uninfected cells or cells infected with the ΔescN mutant (Fig. 4A and andB).B). In contrast, a substantial increase in PI uptake was observed in cells infected with the ΔnleB1 ΔnleF mutant and treated with FasL (Fig. 4A and andB).B). Reintroduction of either nleB1 or nleF to the ΔnleB1 ΔnleF mutant restored protection from FasL-induced cell death (Fig. 4A and andB),B), which was consistent with the inhibition of caspase cleavage observed by immunoblotting (Fig. 3). Complementation of the ΔnleB1 ΔnleF mutant with either nleB1DXD or nleFΔLQCG was insufficient to restore inhibition of FasL-induced cell death (Fig. 4A and andBB).
Previous work has shown that nleB is required for full colonization of mice by C. rodentium (24). Given the overlapping functions of NleB and NleF that we observed during EPEC infection in vitro, we hypothesized that a C. rodentium double nleB nleF mutant would be more attenuated for colonization than a single nleB mutant. To test this, C57BL/6 mice were orally infected with C. rodentium ICC169 and its ΔnleB and ΔnleB ΔnleF derivatives. Fecal samples were collected over 18 days, and viable bacteria were measured by counting CFU per gram of feces. As expected, the nleB mutant of C. rodentium (ΔnleB mutant) was attenuated for colonization compared to wild-type C. rodentium ICC169 on days 8, 10, 12, 14, and 16 postinfection (Fig. 5). However, the additional deletion of nleF (ΔnleB ΔnleF double mutant) did not increase this colonization defect (Fig. 5). Indeed, somewhat surprisingly, the ΔnleB ΔnleF mutant was less attenuated than the ΔnleB mutant for up to 14 days after infection, whereas the single ΔnleF mutant was not attenuated for colonization at any time point during infection (Fig. 5). To test for possible redundancy in the functions of NleB and NleF in vivo, either nleB or nleF was reintroduced into the ΔnleB ΔnleF mutant by integration into the attTn7 site in the C. rodentium genome. Gene expression was under the control of the s12 promoter. The reintroduction of nleB but not nleF restored the bacterial load to wild-type levels (Fig. 5), suggesting that the effector proteins have distinct rather than overlapping functions in vivo.
Apoptotic cell death is an important defense against many invading bacterial pathogens. EPEC-infected cells are protected from apoptotic cell death by several T3SS effectors, including NleB1 and NleF. Previously, NleB1 from EPEC was shown to inhibit extrinsic apoptosis by modifying the adapter protein FADD, which is essential for FasL-induced formation of the DISC (15). Here, we investigated the role of NleF in the inhibition of FasL-induced cell death and demonstrated that, in addition to NleB1, NleF can inhibit caspase activity in FasL-treated apoptotic cells. When delivered into cells by the T3SS or when expressed ectopically, NleF reduced the amounts of cleaved caspase-8 and caspase-3 in epithelial cells treated with FasL. As a result, these cells were protected from cell death. We demonstrated that the C-terminal truncation of NleF (NleFΔLQCG) resulted in a loss of antiapoptotic activity, which is consistent with previous findings describing the C-terminal motif of NleF as essential for blocking caspase substrate binding (22).
We found that NleF inhibited the cleavage of caspase-8, as well as the caspase-8 substrates caspase-3 and RIPK1, implying that NleF contributes to the inhibition of extrinsic apoptosis, similar to NleB1. The proapoptotic BH3-only protein, Bid, is also cleaved by caspase-8, and this leads to mitochondrial damage, subsequent cytochrome c release, and the activation of caspase-9, leading to further activation of downstream caspases caspase-3 and -7 (16). Given that NleF was previously shown to inhibit caspase-9 activity (22), it is possible NleF acts on both caspase-8 and caspase-9 during FAS signaling to inhibit apoptosis.
NleF was initially reported to be important to C. rodentium colonization in C57BL/6 mice, as a C. rodentium ΔnleF mutant showed reduced colonization fitness compared to that of the wild-type strain in mixed infections (20). However, in contrast to these initial reports, we and others have shown that nleF mutants of C. rodentium are not defective for colonization compared to wild-type bacteria (23).
In terms of bacterial load, the double ΔnleB ΔnleF mutant appeared to have some increased fitness compared to the single ΔnleB mutant, which may result from dysregulated NF-κB signaling. Recently, increased secretion of the proinflammatory chemokine CXCL1 was observed in the colons of mice infected with an nleF mutant of C. rodentium complemented with NleF (23). In addition, ectopic expression of NleF led to increased transcription of an NF-κB-dependent reporter, as well as an increase in IL-8 mRNA expression. While this activity was dependent on the LQCG motif of NleF, it occurred independently of caspase involvement (25). Although both NleF and NleB1 can block FasL/FAS signaling, their distinct roles in vivo may be due to their ability to target other innate immune signaling pathways either directly or indirectly. For example, NleB1 modifies the death domain (DD)-containing proteins TRADD and RIPK1 to subvert TNF-induced NF-κB activity and necroptotic cell death (18), and NleF inhibits other caspases (22). In addition, different cellular levels of effector proteins will influence their efficacy. It is possible that introducing higher levels of NleF expression in C. rodentium could compensate for the lack of NleB.
Caspase inhibition by bacterial effector proteins is not unique to NleF. Effector proteins of both Shigella spp. and Yersinia spp. have similar reported activities. The effector protein OspC3 from S. flexneri interacts with active caspase-4 and inhibits caspase-4-dependent inflammatory cell death (26). The Yersinia type III effector YopM binds caspase-1, the activity of which is dependent on a motif on an exposed loop that is similar to the caspase-1 substrate YVAD. Despite possessing a motif similar to a caspase-1 cleavage motif, YopM itself is not cleaved by caspase-1, suggesting that it acts as a pseudosubstrate (27).
Direct binding of lipopolysaccharide (LPS) to human caspase-4 or murine caspase-11 activates the noncanonical inflammasome, resulting in the production of inflammatory cytokines IL-18 and IL-1β, as well as pyroptotic cell death (28, 29). EPEC infection of human IECs activates this noncanonical inflammasome (30), and recently, the expression of NleF was shown to dampen this innate response by binding caspase-4, resulting in a decrease of IL-18 secretion from infected cells (23). Furthermore, NleF of C. rodentium binds murine caspase-11 to inhibit IL-18 secretion from colonic explants at an early time point during C. rodentium infection, resulting in decreased neutrophil recruitment to the colon (23). Taken together, we conclude that NleF functions as a caspase inhibitor to dampen apoptosis and inflammasome activation during infection with attaching and effacing pathogens and that NleB1 is the major effector blocking the FasL/FAS signaling pathway in vivo.
The bacterial strains used in this study are listed in Table 1. Bacteria were grown in Luria-Bertani (LB) broth or Dulbecco's modified Eagle's medium (DMEM; Thermo Fisher Scientific) at 37°C, in the presence of ampicillin (100 μg/ml), kanamycin (100 μg/ml), chloramphenicol (25 μg/ml), or nalidixic acid (50 μg/ml) when required.
The plasmids and primers used in this study are listed in Table 2 and Table 3, respectively. DNA-modifying enzymes were used in accordance with the manufacturer's instructions (New England BioLabs). Plasmids were extracted using the AxyPrep plasmid miniprep kit. PCR products and restriction digests were purified using the Wizard SV gel and PCR clean-up system (Promega). To construct the complementation vectors, nleF was amplified from pEGFP-NleF (25) using primer pair NleFF1/NleFR1 or NleFF1/NleFR2. PCR products were digested with EcoRI and BamHI and ligated into pTrc99A to create pTrc99A-NleF and pTrc99A-NleFΔLQCG. To construct N-terminal Flag fusions to NleB1 and NleF, the nleB1 and nleF genes were amplified from EPEC E2348/69 genomic DNA (25) using primer pairs NleB1F1/NleB1R1 and NleFF2/NleFR1, respectively. The PCR products were digested with EcoRI and BamHI and ligated into p3XFlag-Myc-CMV-24 to create pFlag-NleB1 or pFlag-NleF. The NleB1 catalytic mutant, pFlag-NleB1DXD, was generated using the Stratagene QuikChange II site-directed mutagenesis kit. pFlag-NleB1 was used as the template DNA and amplified using the primer pair NleB1(DXD)F/NleB1(DXD)R. Plasmids were digested with DpnI at 37°C overnight before subsequent transformation into XL-1 Blue E. coli.
To construct the lentiviral transduction vectors, Flag fusions to NleB1 or NleB1DXD were amplified from pFlag-NleB1 or pFlag-NleB1DXD using the primer pair FlagF/NleB1R2. The Flag fusion to NleF was amplified from pFlag-NleF with primer pair FlagF/NleFR3 or FlagF/NleFR4. PCR products were digested with BamHI and NheI and ligated into pF_TRE to create pF_TRE-Flag-NleB1, pF_TRE-Flag-NleB1DXD, pF_TRE-Flag-NleF, and pF_TRE-Flag-NleFΔLQCG.
To construct transgene insertion vectors, nleBCR and nleFCR were amplified from C. rodentium ICC169 genomic DNA using primer pairs NleBF2/NleBR3 and NleFF3/NleFR5, respectively. The s12 promoter was amplified from C. rodentium ICC169 genomic DNA using primer pair S12F1/S12R1 or S12F1/S12R2 before being annealed to a complementary region of the nleBCR or nleFCR amplification product. Annealed products were ligated into pCR-Blunt II-TOPO using the Zero Blunt TOPO kit (Thermo Fisher Scientific) to create pCR-Blunt II-TOPO-NleBCR or pCR-Blunt II-TOPO-NleFCR. s12 promoter-gene fusions were then digested with XmaI and XhoI and ligated into pGRG36 to create pGRG36-NleBCR and pGRG36-NleFCR.
The EPEC E2348/69 ΔnleB1 ΔnleF double mutant was created using the lambda Red mutagenesis system (35). Using genomic DNA from EPEC E2348/69, PCR was used to amplify regions from the 5′ and 3′ flanking sites of nleF using primer pairs NleFEP_1/NleFEP_2 and NleFEP_3/NleFEP_4. The plasmid pKD4 was used as the template DNA to amply the kanamycin resistance cassette using primer pair pKDF/pKDR. Overlapping PCR was used to assemble the linear fragment consisting of the kanamycin resistance cassette and the 5′ and 3′ nleF flanking regions using primer pair NleFEP_1/NleFEP_4. This fragment was then electroporated into a previously described ΔnleB1 mutant of EPEC E2348/69 (15) containing the lambda Red helper plasmid pKD46, following the expression of lambda Red recombinase with 10 mM l-arabinose. The deletion of nleF was confirmed by PCR and DNA sequencing using the external primer pair NleFEP_1/NleFEP_4.
To construct the C. rodentium ICC169 ΔnleB ΔnleF double mutant, PCR was used to amplify regions from the 5′ and 3′ flanking sites of nleF from C. rodentium ICC169 using primer pairs NleFCR_1/NleFCR_2 and NleFCR_3/NleFCR_4. The plasmid pKD3 was used as the template DNA to amplify the chloramphenicol resistance cassette using primer pair pKDF/pKDR. Overlapping PCR was used to assemble the linear fragment using primer pair NleFCR_1/NleFCR_4. This fragment was then electroporated into a previously described C. rodentium ΔnleB mutant containing pKD46. The deletion of nleF was confirmed by PCR and DNA sequencing using the external primer pair NleFCR_1/NleFCR_4. The C. rodentium ΔnleB ΔnleF mutant was complemented with either nleBCR or nleFCR using a previously described method of transgene insertion (36). Briefly, pGRG36-NleB or pGRG36-NleF was electroporated into electrocompetent C. rodentium ΔnleB ΔnleF. The insertion of nleB or nleF at the attTn7 insertion site was confirmed by PCR and sequencing using primers CRF/CRR.
HEK293T cells were grown on 10-cm plates to a confluence of 60% and transfected with 2.0 μg pCMVΔR8.2, 0.8 μg pCMV-VSV-G, and 1.2 μg pF_TRE-Flag-NleF, pF_TRE-Flag-NleFΔLQCG, pF_TRE-Flag-NleB1, or pF_TRE-Flag-NleB1DXD. The tissue culture medium was replaced 24 h after transfection. Supernatant containing packaged virus particles was harvested 48 h after transfection and filtered through a 0.45-μm membrane filter. Hexadimethrine bromide (5 μg/ml; Sigma) was added to virus-containing medium. HeLa cells were grown in 6-well plates to a confluence of 50%, and then cell culture medium was replaced with virus-containing medium added to DMEM with 10% fetal calf serum (FCS) in a 1:1 ratio. HeLa cells were centrifuged at 1,800 rpm for 45 min at room temperature. Forty-eight hours after infection, the tissue culture medium was replaced with DMEM with 10% FCS containing puromycin (5 μg/ml). Cells were maintained in puromycin for at least three passages before use.
HeLa cells and inducible HeLa cell line derivatives were seeded into 24-well tissue culture plates 24 h prior to being left untreated or treated with doxycycline (20 ng/ml) to induce effector protein expression. An additional 24 h later, the medium was replaced with DMEM containing FasL (20 ng/ml; gift from Andreas Strasser) for 1.5 h. Cell lysates were collected and analyzed by immunoblotting as described below.
For propidium iodide (PI) staining, HeLa cells and inducible HeLa derivatives were seeded on poly-l-lysine-coated coverslips in 24-well tissue culture plates 24 h prior to being treated with doxycycline and FasL as described above. PI (20 μg/ml; Cayman Chemicals) was added to each well for the final 15 min of FasL treatment. Cells were prepared for confocal microscopy as described below.
HeLa cells were seeded onto 24-well tissue culture trays for 24 h prior to infection. EPEC cultures grown with shaking overnight in LB were inoculated 1:100 into DMEM and subcultured for a further 3 h, including 30 min of induction with 1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) in 5% CO2. HeLa cells were left uninfected or infected with EPEC cultures at an optical density at 600 nm (OD600) of 0.03. After 2 h, the tissue culture medium was replaced with DMEM containing gentamicin (100 μg/ml) with or without FasL (20 ng/ml; gift from Andreas Strasser) for a further 1.5 h. For detection of caspase cleavage, cells were harvested for immunoblotting as described below. For quantification of PI-positive cells, PI was added to each well for the final 15 min of FasL treatment. Cells were prepared for confocal microscopy as described below.
For analysis of cell monolayers, cells were lysed in cold lysis buffer (50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 10 mM NaF, 1 mM phenylmethylsulfonyl fluoride [PMSF], 2 mM Na3VO4 with 1× EDTA-free Complete protease inhibitor cocktail [Roche]). Cell lysates were pelleted, and equal volumes of supernatant were collected and added to 4× lithium dodecyl sulfate (LDS) sample buffer (Thermo Fisher) supplemented with dithiothreitol (DTT; 50 mM), heated to 70°C for 10 min, and resolved on 4 to 12% Bis-Tris gels (Thermo Fisher) by PAGE. Proteins were transferred to nitrocellulose membranes before being incubated for 1 h in 5% skimmed milk powder in Tris-buffered saline (TBS; 50 mM Tris-HCl, pH 7.5, 150 mM NaCl) with 0.1% Tween 20. Membranes were washed and probed with one of the following primary antibodies diluted 1:1,000 in 5% bovine serum albumin (BSA) in TBS with 0.1% Tween 20: horseradish peroxidase (HRP)-conjugated monoclonal mouse anti-FLAG-M2 antibody (Ab) (Sigma-Aldrich), monoclonal mouse anti-caspase-8 Ab (1C12) (Cell Signaling Technology), polyclonal rabbit anti-cleaved caspase-3 Ab (Asp175) (Cell Signaling Technology), monoclonal rabbit anti-RIP (D94C12) XP Ab (Cell Signaling Technology), or monoclonal mouse anti-β-actin Ab (Sigma-Aldrich) diluted 1:5,000. The secondary antibodies used were HRP-conjugated anti-mouse Ab (PerkinElmer) and HRP-conjugated anti-rabbit Ab (PerkinElmer) diluted 1:3,000 in TBS with 5% BSA and 0.1% Tween 20. Immunoblots were developed with Amersham ECL Western blotting detection reagents (GE Healthcare) or Western Lightning ultra (PerkinElmer), and chemiluminescence was detected using the DNR MF-ChemiBIS Bio-Imaging System.
For visualization of PI-positive HeLa cells, treated cells were fixed with paraformaldehyde (4% in phosphate-buffered saline [PBS]; Sigma) for 15 min, followed by thorough washing with PBS. Cells were incubated for a further 15 min with Hoechst (2.4 μg/ml; Thermo Fisher) to stain DNA.
Coverslips were mounted onto microscope slides with Prolong Gold antifade reagent (Invitrogen). Images were acquired using a Zeiss confocal laser scanning microscope with a 20× objective. For quantification of PI-positive cells, three distinct fields per coverslip containing approximately 60 cells each were counted blind. Results were obtained from at least three independent experiments.
All animal experimentation was approved by the University of Melbourne Animal Ethics Committee. Strains of C. rodentium were cultured in LB broth supplemented with appropriate antibiotics overnight at 37°C with shaking. Bacteria were pelleted by centrifugation and resuspended in PBS. C57BL/6 mice (6 to 8 weeks old, male and female) were inoculated by oral gavage with 200 μl containing approximately 1 × 109 CFU of C. rodentium derivatives. The viable count of the inoculum was determined by retrospective serial dilution and plating on Luria-Bertani agar (LA) containing appropriate antibiotic. Fecal samples were collected from each mouse on days 4, 8, 10, 12, 14, 16, and 18 postinfection and emulsified in PBS to a concentration of 100 mg/ml. The number of viable bacteria per gram of feces was determined by plating serial dilutions of fecal samples onto antibiotic-containing LA. The limit of detection was 400 CFU/g feces.
This work was supported by the Australian National Health and Medical Research Council (project grants APP1098826 and APP1044061 to E.L.H. and Early Career Fellowship APP1090108 to J.S.P.). G.L.P. and C.G. were supported by Australian postgraduate awards. T.W.F.L. was supported by a University of Melbourne International Research Scholarship (MIRS).
We declare no financial interests related to this work.