Conserved nontypeable Haemophilus influenzae-derived TLR2-binding lipopeptides synergize with IFN-beta to increase cytokine production by resident murine and human alveolar macrophages

Abstract

Nontypeable Haemophilus influenzae (NTHi) is strongly associated with exacerbations of chronic obstructive pulmonary disease, which often coincide with viral respiratory infections. TLR2 contributes importantly to innate immunity to NTHi, but whether this pathway is affected by simultaneous antiviral responses is unknown. To analyze potential interactions, resident murine and human alveolar macrophages (AMphi) were exposed, in the presence or absence of the appropriate rIFN-beta, to synthetic lipopeptides corresponding to the triacylated N-terminal fragments of three outer membrane proteins (OMP) (PCP, P4, and P6) that are highly conserved among different NTHi strains. Synthetic OMP elicited strong release of IL-6, the principal inducer of airway mucin genes, and induced CCL5 and CXCL10 from murine AMphi only when IFN-beta was also present. Surprisingly, combined stimulation by OMPs and IFN-beta also markedly enhanced TNF-alpha release by murine AMphi. Stimulation with PCP plus IFN-beta induced IFN-regulatory factor 1 expression and sustained STAT1 activation, but did not alter the activation of MAPKs or NF-kappaB. AMphi derived from STAT1-deficient mice did not demonstrate increased production of TNF-alpha in response to PCP plus IFN-beta. Analysis of wild-type and STAT1-deficient AMphi using real-time PCR showed that increased TNF-alpha production depended on transcriptional up-regulation, but not on mRNA stabilization. The synergistic effect of synthetic OMP and IFN-beta was conserved between murine AMphi and human AMphi for IL-6, but not for TNF-alpha. Thus, IFN-beta, which is produced by virally infected respiratory epithelial cells, converts normally innocuous NTHi OMP into potent inflammatory stimulants, but does so via different mechanisms in mice and humans.

FIGURE 1

Murine AMφ from TLR2^−/− mice do not respond to the NTHi OMP PCP, P4, and P6. Resident AMφ from TLR2^−/− mice were incubated for 6 h in the presence of medium alone or medium containing 100 ng/ml of each of the three NTHI OMPs (PCP, P4, and P6). As a control for TLR2 stimulation, we used the synthetic lipopeptide Pam₃CysSK₄ (referred to in this and subsequent figures as LP). The sources and structures of the synthetic lipopeptides used in this work are described in detail in Materials and Methods. To test for cell responsiveness, 100 ng/ml OMP-free LPS was used as positive control for TLR4 stimulation. Supernatants were collected and assayed by ELISA for TNF-α. φ, None detected. Data are mean ± SD of triplicate wells in a single experiment.

FIGURE 2

Murine AMφ respond strongly to NTHi OMP only in the presence of IFN-β. Resident AMφ from normal C57BL/6 mice were incubated for 6 h in the presence of medium alone, 1000 U/ml IFN-β, 100 ng/ml of each of four OMPs (PCP, P4, P6, and positive control LP) (□), or 100 ng/ml of each OMPs plus 1000 U/ml IFN-β( ■). Supernatants were collected and assayed by ELISA for (A) IL-6 and (B) TNF-α. φ, None detected. Data are mean ± SD of triplicate wells.*, p <0.05, unpaired Student t test. One of three independent experiments with similar results is shown.

FIGURE 3

Overnight stimulation of murine AMφ with NTHi OMP plus IFN-β increases IL-6 production and induces the chemokines RANTES and IP-10. Resident AMφ from normal C57BL/6 mice were incubated for 18 h in the presence of medium alone, 1000 U/ml IFN-β, 100 ng/ml of each of four OMPs (PCP, P4, P6, and positive control LP) (□), or 100 ng/ml of each OMP plus 1000 U/ml IFN-β (■). Supernatants were collected and assayed by ELISA for (A) IL-6, (B) CCL5, and (C) CXCL10.φ, None detected. Data are mean ± SD of triplicate wells.*, p <0.05, unpaired Student t test. One of two independent experiments with similar results is shown.

FIGURE 4

Combined stimulation of murine AMφ with the NTHi OMP PCP and IFN-β induces long-term activation of STAT1 and increases expression of IRF-1 without affecting activation of MAPKs. Resident AMφ from normal C57BL/6 mice were incubated for the indicated times in the presence of medium alone (lane C′), 1000 U/ml IFN-β (lanes IFN-β), PCP 100 ng/ml or a combination of both (lanes PCP+IFN-β). Cells were then harvested and processed for Western blot analysis as specified in Materials and Methods. The membranes were probed with (from the top) anti-IRF-1, anti-pY701 STAT1 anti-STAT1, anti p-p38, and anti-p38. Note that the pY701 STAT1 film in B has been deliberately overexposed relative to the other rows to emphasize the absence of STAT1 phosphorylation by PCP alone. C–F, Densitometric quantitative analysis of the band intensities of (C and E) IRF-1 and (D and F) pY701 STAT1 in the films shown in A (C and D) and B (E and F), respectively. As a control for loading, band intensities were normalized relative to the corresponding p38 bands. AU, arbitrary units. IFN-β, light gray bars; PCP plus IFN-β, ■. Similar results were obtained in an additional independent experiment of equivalent design.

FIGURE 5

Increased induction of TNF-α by combined stimulation of AMφ with the NTHi OMP PCP and IFN-β is transcriptionally regulated. Resident AMφ from normal C57BL/6 mice were cultured for 3 or 6 h in the presence of medium alone (C′) (light gray bars), 1000 U/ml IFN-β (IFN-β ) (gray bars), 100 ng/ml PCP (□), or a combination of the same doses of PCP and IFN-β (Both) (■). Total cellular RNA was collected and processed for real-time PCR; samples were analyzed for (A) TNF-α or (B) IL-6 mRNA expression. The relative quantities of specific mRNA from treated AMφ are plotted compared with the calibrator mRNA (TNF-α or IL-6 mRNA in control AMφ, which is defined as a value of 1). dRn is baseline-corrected, reference dye-normalized fluorescence.*, p <0.05, unpaired Student t test. Data are mean ± SEM of triplicate wells in each of two independent experiments with similar results.

FIGURE 6

Increased mRNA stability is not responsible for increased TNF-α production by AMφ stimulated with the NTHi OMP PCP and IFN-β. Resident AMφ from normal C57BL/6 mice were cultured for 3 h in the presence of medium alone (C′) (light gray bars), 100 ng/ml PCP (□), or a combination of PCP and IFN-β (1000 U/ml) (■). After 3 h, the RNA polymerase inhibitor actinomycin D (10 μg/ml, ActD) was added to a fraction of the samples and all the samples were incubated for 3 additional hours. Total cellular RNA was collected and processed for real-time PCR and samples were analyzed for TNF-α mRNA expression. The relative quantities of specific mRNA from treated AMφ are plotted compared with the calibrator mRNA (TNF-α mRNA in control AMφ, which is defined as a value of 1). dRn is baseline-corrected, reference dye-normalized fluorescence.*, p <0.05, unpaired Student t test. Data are mean ± SEM of triplicate wells in each of two independent experiments with similar results.

FIGURE 7

Increased induction of TNF-α by combined stimulation of murine AMφ with the NTHi OMP PCP and IFN-β is STAT1 dependent. Resident AMφ from STAT1^−/− or STAT1^−/−. mice were cultured in the presence of medium alone (C′), 100 ng/ml PCP (□) or a combination of 100 ng/ml PCP and 1000 U/ml IFN-β (■). Supernatants were harvested at 6 h and analyzed by ELISA for (A) TNF-α or (B) IL-6 concentration. (C) In parallel cultures, total RNA was prepared at 3 or 6 h after stimulation and analyzed for TNF-α mRNA expression by real-time PCR.*, p <0.05, unpaired Student t test. Data are mean ±SD of triplicate wells in each of two independent experiments with similar results.

FIGURE 8

Human AMφ respond synergistically to NTHi lipopeptides and recombinant human IFN-β. Resident AMφ from three donors were incubated for 6 h in the presence of medium alone, 1000 U/ml IFN-β, 100 ng/ml of each of the indicated synthetic OMPs (○) or a combination of one OMP and IFN-β(●). The supernatants were collected and processed for ELISA for IL-6 (A) and TNF-α (B).φ, None detected. The three independent experiments performed are shown together and assessed for statistical significance by paired Student t test. The mean of all the donors is also shown in each experimental group.