A novel NOD1- and CagA-independent pathway of interleukin-8 induction mediated by the Helicobacter pylori type IV secretion system
Version of Record online: 21 NOV 2012
© 2012 Blackwell Publishing Ltd
Volume 15, Issue 4, pages 554–570, April 2013
How to Cite
Gorrell, R. J., Guan, J., Xin, Y., Tafreshi, M. A., Hutton, M. L., McGuckin, M. A., Ferrero, R. L. and Kwok, T. (2013), A novel NOD1- and CagA-independent pathway of interleukin-8 induction mediated by the Helicobacter pylori type IV secretion system. Cellular Microbiology, 15: 554–570. doi: 10.1111/cmi.12055
- Issue online: 14 MAR 2013
- Version of Record online: 21 NOV 2012
- Accepted manuscript online: 26 OCT 2012 04:13AM EST
- Manuscript Accepted: 9 OCT 2012
- Manuscript Revised: 24 AUG 2012
- Manuscript Received: 18 APR 2012
- NHMRC. Grant Numbers: APP545983, APP1006010
- Victorian Government
Supplementary materials and methods.
Fig. S1. Strategy for cloning and mutagenesis of H. pylori cagL.
A. cagL and surrounding regions of sufficient length for recombination were amplified from H. pylori strains 26695 or P12 and cloned into pGEM-T easy to produce plasmids p26695cagNI and pP12cagNI respectively.
B. The cagL open reading frame from immediately after the cagI stop codon was replaced with aphA3 to produce plasmids p26695ΔcagL::aphA3 or pP12ΔcagL::aphA3; donor DNA produced by amplification of p26695ΔcagL::aphA3 or pP12ΔcagL::aphA3 with the cagNF/cagIR primer pair was used for transformation of 26695 and P12, respectively, to generate CagL− mutants.
C. p26695cagNI was modified using QuikChange site-directed mutagenesis (Agilent Technologies, La Jolla, CA) to insert a BglII site following the cagL stop codon, into which the cat marker was inserted to produce plasmid p26695cagLRGD/cat; p26695cagLRGD/cat was further modified to introduce G77A or D78A amino acid substitutions thus producing plasmids p26695cagLRAD/cat and p26695cagLRGA/cat respectively; donor DNA produced by amplification of p26695cagLRGD/cat, p26695cagLRAD/cat and p26695cagLRGA/cat with the cagNF/cagIR primer pair was used for transformation of H. pylori P12ΔcagL to restore CagL to the deletion mutants as either wild-type or amino acid substitution forms. Short arrows denote primer binding sites; B = BamHI site; primer names correspond to those listed in supplementary Table S2.
Fig. S2. Effects of substitution of the RGD motif of CagL on CagA translocation. AGS cells were inoculated with various isogenic H. pylori strains as indicated; moi = 100. The effects of substitution of the RGD motif of CagL in P12 wild-type and P12ΔvirD4 mutant are shown in (A) and (B) respectively. Cell lysates harvested at 24 hpi were analysed by immunoblotting. Tyrosine-phosphorylated CagA (translocated CagA) and total CagA were detected using the phosphotyrosine-specific antibody PY99 and anti-CagA antibody respectively.
Fig. S3. Effect of AIIB2 on CagA-dependent and -independent IL-8 secretion by H. pylori strains P12 and P1. AGS cells were treated with RPMI alone or 2 μg ml−1 AIIB2 in RPMI for 1 h prior to inoculation with wt and ΔcagA isogenic mutants of H. pylori strains P12 or P1 (moi = 100). Spent culture medium harvested at 24 hpi was assayed for secreted IL-8; data from three (P12) or one (P1) experiment(s), each in duplicate; mean ± standard error of mean (SEM), one-way anova with Bonferroni post-test, *P < 0.05, **P < 0.01, ****P < 0.0001.
Fig. S4. Recombinant CagL stimulates IL-8 secretion by AGS cells in a dose-dependent manner. AGS cells were incubated with RPMI (0 μM CagL) or recombinant wild-type CagL at the concentrations indicated for 24 h. Culture medium was then collected and assayed for secreted IL-8.
Fig. S5. Effects of integrin α5 and integrin β1 function-blocking antibodies on IL-8 induction by wild-type CagL or CagLRGA. AGS were incubated with RPMI (untreated) or AIIB2 (2 μg ml−1) and BIIG2 (2 μg ml−1) prior to stimulation with 4 μM wild-type CagL (CagLWT) or CagLRGA mutant protein. Culture medium at 24 h post stimulation was assayed for secreted IL-8. Percentage levels of IL-8 secreted compared with that observed with AGS stimulated with CagLWT in the absence of function-blocking antibodies (RPMI control) are shown. Mean ± SEM of data from two independent experiments; **P < 0.01; ***P < 0.001, two-way anova with Bonferroni post-test.
Fig. S6. Dose curves showing the effects of signal transduction inhibitors on IL-8 induction by CagL. AGS cells were pre-treated with (A) Src kinase inhibitor, Src I-1, (B) Ras kinase inhibitor, FTS, (C) MEK1/2 inhibitor, U0126, (D) JNK inhibitor, SP600125, (E) proteasome inhibitor, MG-132 or (F) IKK inhibitor, BMS-345541 at the concentrations indicated. AGS cells were pre-treated with RPMI as negative controls (0 μM inhibitor). Pre-treated AGS cells were stimulated with 4 μM wild-type CagL for 24 h, after which the culture medium was collected and assayed for secreted IL-8. Mean ± SEM of data from three independent experiments shown.
Fig. S7. Specific inhibition of ERK, JNK and p38 activation by the inhibitors U0126, SP600125 and SB203580 respectively. AGS cells were pre-treated with 5 μM SrcI-1, 5 μM U0126, 10 μM SP600125 or 10 μM SB203580 prior to stimulation with 50 nM PMA. Total-cell lysate was then analysed for the relative abundance of activated ERK, JNK and p38, respectively, by immunoblotting using anti-phospho-ERK, anti-phospho-JNK and anti-phospho-p38 antibodies as indicated. Relative abundance of GAPDH (loading control) was determined using anti-GAPDH antibody. In contrast to U0126, SP600125 or SB203580, the Src-specific inhibitor, SrcI-1, did not inhibit the activation of ERK, JNK or p38.
Fig. S8. Activation of ERK, JNK and Src by recombinant CagL.
A. AGS cells were stimulated with recombinant protein CagLWT, mutant protein CagLRGA or BSA. Cell lysates at 90 min post stimulation were analysed for the relative abundance of active ERK, JNK and Src by immunoblotting using anti-phospho-ERK, anti-phospho-JNK and anti-phospho-Src antibodies respectively. The relative abundance of GAPDH (loading control) was determined using anti-GAPDH antibody. The relative band intensities of phospho-ERK, phospho-JNK and phospho-Src normalized against loading controls are shown on the right. Mean ± SD; *P < 0.05; **P < 0.01; ***P < 0.001, two-way anova with Bonferroni post-test. All other comparisons are insignificant (P > 0.05).
B. AGS cells were treated with 8 μM recombinant protein CagLWT or RPMI for 2 h prior to inoculation with H. pylori P12 or P12ΔcagL at a moi of 30, or with BHI. Total-cell lysate at 24 hpi was analysed for the relative abundance of activated ERK by immunoblotting using anti-phospho-ERK antibody. The relative abundance of actin (loading control) was determined using anti-actin antibody.
Fig. S9. IL-8 induction by recombinant wild-type CagL is independent of ADAM17.
A. AGS cells transfected with ADAM17-targeting siRNAs (ADAM17 KD) or negative control siRNA (control) were incubated with RPMI or 4 μM wild-type CagL recombinant protein for 24 h, after which culture medium was collected and assayed for secreted IL-8.
B. The relative ADAM17 mRNA levels in the ADAM17 knock-down (ADAM17 KD) and control cells are shown. Mean ± SEM of two independent experiments performed in duplicate; ***P < 0.001, two-way anova with Bonferroni post-test.
Fig. S10. Complementation of T4SS-dependent IL-8 secretion by recombinant CagL is dose-dependent. AGS cells were pre-treated with 8.0, 0.8 or 0 μM recombinant CagL prior to inoculation with H. pylori strains P12 or P12ΔcagL (moi = 30). Spent culture medium was harvested at 24 hpi and assayed for secreted IL-8 by specific ELISA. Bars (mean ± SEM) represent fold increase in secreted IL-8 from three independent experiments; **P < 0.01, ****P < 0.0001, two-way anova with Bonferroni post-test; P12 v P12ΔcagL comparisons asterisks only, rCagL dose comparisons asterisks and bar, all other comparisons NS (> 0.05).
Fig. S11. Exogenous CagL facilitates P12ΔcagL-induced IL-8 secretion in a T4SS-dependent and RGD-enhanced manner. AGS cells were treated with 8 μM recombinant protein CagLWT, mutant protein CagLRGA or RPMI only for 2 h prior to inoculation with various H. pylori strains (moi = 30), or with BHI; data from two independent experiments, each performed in duplicate.
A. Spent culture media harvested at 7 hpi was assayed by ELISA for secreted IL-8; mean ± SEM.
B. Spent culture media harvested at 24 hpi was assayed by ELISA for secreted IL-8; mean ± SEM.
C. Fold increase in IL-8 secreted in response to rCagL treatment compared with RPMI treatment at 24 hpi; mean ± SEM.
All statistical analysis by two-way anova with Bonferroni post-test: *P < 0.05, **P < 0.01, ****P < 0.0001.
Fig. S12. T4SS-dependent binding of recombinant CagL to H. pylori P12ΔcagL.
A. H. pylori P12ΔcagL or P12ΔcagPAI was incubated for 2 h in wells of a 96-well plate coated with rCagLWT, rCagLRGA or BSA (50 μg ml−1), H. pylori-specific rabbit hyperimmune antisera or pre-immune rabbit serum. Bound H. pylori was immobilized in the well and detected by H. pylori-specific ELISA. Bound H. pylori shown as a percentage of binding to H. pylori-specific antisera. Mean ± SEM of three independent experiments performed in duplicate; ***P < 0.001, ****P < 0.0001, two-way anova with Bonferroni post-test. The two-way anova results of pre-immune antibody attachment data were identical to those shown for BSA but were omitted from the graph for simplification.
B. AGS cells were treated with fluorescein-conjugated rCagL prior to (4 μM) and during (2.7 μM) stimulation with H. pylori P12ΔcagL or P12ΔcagPAI (moi = 30) for 2.5 h. H. pylori was immunolabelled with AF647 and samples were visualized by super-resolution GSDIM microscopy. Bars represent 1 μm, two cells shown for each strain.
Fig. S13. Effects of specific inhibitors of Src, ERK and JNK on IL-8 induction by H. pylori P12 wild-type and P12ΔvirD4. AGS was infected with H. pylori strains P12 or P12ΔvirD4 at a moi of 30 after 1 h of pre-treatment with RPMI or RPMI containing 5 μM SrcI-1, 5 μM U0126 or 10 μM SP600125. Spent culture media was harvested at 24 hpi and assayed by ELISA for secreted IL-8; mean ± SEM, data from two independent experiments; *P < 0.001, compared with IL-8 induction by P12 wild-type in the absence of inhibitors.
Fig. S14. Comparison of the ability of P12 T4SS mutants to stimulate IL-8 secretion in NOD1 knock-down cells. AGS-siNOD1 was infected with H. pylori strains P12, P12ΔvirD4, P12ΔcagL or P12ΔcagPAI at a moi of 30. Spent culture media was harvested at 24 hpi and assayed by ELISA for secreted IL-8; mean ± SEM, data from three independent experiments. **P < 0.01, two-way anova, Bonferroni post-test.
Table S1. H. pylori strains used in this study.
Table S2. PCR primers used in this study.
Table S3. IL-8 secretion by AGS cells in response to multiple independent clones of H. pylori CagL variants.
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.