Characterization of the Helicobacter pylori cysteine-rich protein A as a T-helper cell type 1 polarizing agent

Abstract

Predominant T-helper 1 (Th1) responses with increased gamma interferon (IFN-gamma) levels have been proposed to play an important role in Helicobacter pylori-induced gastritis and peptic ulceration. However, bacterial factors contributing to the initiation of Th1 polarization of H. pylori-specific immune responses have not been characterized in detail thus far. We report here on the identification of Helicobacter cysteine-rich protein A (HcpA) as a novel proinflammatory and Th1-promoting protein. The capacity of HcpA to induce immune activation was studied in splenocyte cultures of naive H. pylori-negative mice. HcpA stimulated the release of high concentrations of the proinflammatory and Th1-promoting cytokines interleukin-6 (IL-6) and IFN-gamma, in addition to significant levels of IL-12, tumor necrosis factor alpha, and IL-10. The observed cytokine profile was comparable to that induced by lipopolysaccharide but differed in the kinetics and maximum levels of cytokine production. In addition, HcpA-induced cytokine release resembled that observed upon incubation with H. pylori except for IL-10, which was only moderately released upon HcpA stimulation. Both HcpA- and H. pylori-mediated IFN-gamma production was drastically reduced by a neutralizing antibody against IL-12 but not by an anti-IL-2 antibody. Thus, HcpA seems to represent a novel bacterial virulence factor triggering the release of a concerted set of cytokines to instruct the adaptive immune system for the initiation of proinflammatory and Th1-biased immunity.

FIG. 1.

Expression and purification of HcpA protein. (A) Flow diagram of HcpA purification process. (B) Coomassie blue-stained SDS-PAGE of HcpA at various purification stages. Lanes: 1, lysate of noninduced cells; 2, lysates of cells 10 h after induction with 1 mM IPTG; 3, periplasmic fraction; 4, heat-precipitated periplasmic protein fraction; 5, HcpA-positive peak fractions obtained from cation-exchange chromatography; 6, immunoblot analysis of purified HcpA by using a HcpA-specific polyclonal rabbit serum. The arrow at the right indicates the position of HcpA. The position of the markers (MW) is given on the left (in kilodaltons).

FIG. 2.

Expression and secretion of HcpA in different H. pylori strains. (A) Expression of the strictly cytoplasmic protein RecA in H. pylori lysates was determined as a loading control. (B and C) HcpA expression and secretion were investigated in H. pylori whole cellular lysates (B) or cell-free culture supernatants (C) by immunoblotting with a HcpA-specific rabbit serum. Lanes 1 to 4 and 7, CagA-positive H. pylori strains from ulcer patients; lane 5, CagA-positive isolate of a carcinoma patient; lanes 6 and 8, CagA-positive strains from MALT lymphoma patients; lanes 9, 10, and 15, CagA-negative strains from gastritis patients; lanes 11 to 14, CagA-negative isolates from carcinoma patients. The band beyond HcpA represents HcpC, which also belongs to the gene family 12 and shows the highest degree of similarity to HcpA among all Hcp members. (D) Schematic illustration of the hcpA ORF indicating the positions of the signal peptide and the putative internal cleavage site.

FIG. 3.

Dose dependency of cytokine production induced by purified HcpA in cultures of splenic cells from naive BALB/c mice. Cells were incubated in triplicate with increasing HcpA concentrations, and after 36 h the cytokine levels were determined from the precleared culture supernatant by ELISA as essentially described in Materials and Methods. One representative experiment out of three is shown. The data represent means ± standard deviations (SD).

FIG. 4.

Comparison of HcpA and LPS for the stimulation of cytokine release from splenic cells from naive BALB/c mice. Cells were incubated in triplicate with purified HcpA (2 μg/ml) or LPS (5 μg/ml), and the levels of TNF-α (A), IL-12 (B), and IFN-γ (C) were measured from the precleared culture supernatant at the indicated time points by ELISA essentially as described in Materials and Methods. (D) Murine splenocytes were stimulated for 36 h with purified HcpA (2 μg/ml), RBI fraction (2 μl), LPS (5 μg/ml or 10 ng/ml), and the concentrations of IFN-γ were determined by ELISA. PBS-stimulated splenocytes served as negative controls. One representative experiment out of three is shown. The data represent the mean values ± SD.

FIG. 5.

Cytokine production from cultured murine splenocytes upon stimulation with H. pylori 2802 or recombinant HcpA. Cells were stimulated with live H. pylori 2802 (MOI = 0.5) or 2 μg of E. coli-derived HcpA/ml, and the levels of the indicated cytokines were determined by ELISA. One representative experiment out of three performed in triplicate cultures is shown. The data represent means ± the SD. ✽, P < 0,0001.

FIG. 6.

Generation and characterization of an isogenic HcpA-deficient H. pylori mutant strain. (A) PCR analysis of two transformants of HcpA mutagenesis (lanes 1 and 2) confirmed a larger amplicon in 2802 hcpA::cat due to the presence of cat resistance gene in the hcpA locus in comparison to the H. pylori parental strain 2802 (lane 3). (B) Amplification of hcpC gene in 2802 hcpA::cat (lane 1) and 2802 (lane 2) was performed to verify the specificity of the hcpA mutagenesis. (C) Comparative immunoblot analysis of HcpA expression in cytoplasmic (CP) and periplasmic (P) fractions of H. pylori 2802 (lane 2) and the isogenic 2802 hcpA::cat H. pylori strain (lane 3) with H. pylori 60190 (lane 1).

FIG. 7.

Cytokine production from cultured murine splenocytes induced by wild-type H. pylori 2802 and the isogenic HcpA knockout strain. Splenocytes from naive C57BL/6 mice were incubated with H. pylori at an MOI of 0.5, and the induced cytokine pattern was assessed after 36 h. PBS-stimulated splenocytes served as negative controls. The levels of the indicated cytokines were determined by ELISA. One representative experiment out of four performed in triplicate cultures is shown. The data represent means ± the SD (P > 0.2).

FIG. 8.

Dependency of IFN-γ production on IL-12 and IL-2. Splenic cell of naive BALB/c mice were stimulated with H. pylori 2802 or E. coli-derived HcpA in presence of increasing concentrations of anti-IL-12 and anti-IL-2 antibodies. The levels of IFN-γ were determined from the cell culture supernatant after 36 h by ELISA. One representative experiment out of three performed in triplicates is shown. The data represent means ± the SD.