Figure 1.
IL-23 expression is increased in hepatitis B liver tissue.
(A) Relative mRNA expression of IL-23 in liver tissue of various groups was analyzed by qPCR. (B) Protein expression of IL-23 was analyzed by Western blot assay. Error bars indicate SD. *P<0.05; **P<0.01. (C) In situ expression of IL-23 in liver tissues was detected by immunohistochemical staining (magnification 100×).
Figure 2.
IL-23 is mainly derived from APCs in hepatitis B liver tissue.
Co-localization (yellow) of IL-23 (p19, green) with mDC (CD11c, red) (A), pDC (CD303, red) (B), macrophage (CD68, red) (C), or IL-23 (p19, red) with hepatocyte (hepatocyte, green) (D) in CHB patients' liver tissues was detected by immunofluorescence staining. The right panels are enlarged images of the area in the left panel that is demarcated by a white dashed-line (A and C). (E and F), The frequencies of intrahepatic IL-23-positive mDCs or macrophages in the livers of patients with hepatitis B and healthy controls are shown. A total of 118 CD11c+ and 89 CD68+ cells were counted in 10 random fields (magnification 400×, 0.066 mm2) per section using a digital camera. Error bars indicate SD. **P<0.01.
Figure 3.
IL-23R expression is enhanced markedly and correlated with IL-23 level in liver tissues of CHB and ACLF patients.
(A) Relative mRNA expression of IL-23R in liver tissue of various groups was analyzed by qRT-PCR. (B) Protein expression of IL-23R was analyzed by Western blot. Error bars indicate SD. **P<0.01. (C) In situ expression of IL-23R in liver tissues was detected by immunohistochemical staining (magnification 100×). (D) The frequencies of intrahepatic IL-23R positive cells in patients with hepatitis B and health controls are shown. Error bars indicate SD. **P<0.01; HPF, high power fields. (E) The correlation was analyzed between the expression of IL-23 and IL-23R in HBV infected liver tissue. r is the correlation coefficient and P-value is shown.
Figure 4.
Hepatic IL-23/IL-23R is closely correlated with elevated IL-17, and IL-23R+ T cells are the main source of IL-17 in liver of CHB and ACLF patients.
Relative mRNA expressions of IL-23, IL-23R and IL-17 in liver tissues infected by HBV were assayed by qPCR, and then Spearman's rank correlation was used to determine the correlation between relative mRNA expression of (A) IL-23 or (B) IL-23R and IL-17 in 51 CHB patients. Co-localization of IL-17 (red) with IL-23R (green) (C) in CHB patients' liver tissues was examined by fluorescence microscope. The right panels are enlarged images of the area in the left panel demarcated with the white dashed-line box. (D) Co-localization of IL-17 and IL-23R in PBMCs from hepatitis B patients is shown. PBMCs from patients with CHB were stimulated by PMA and ionomycin for 4 h in the presence of GolgiStop. A moiety of the stimulated PBMCs was measured by FCM to determine the frequency of IL-17+23R+ T cells in the total CD4+ T cells, with gating on CD3+CD8− T cells. Co-localization of IL-17 (red) and IL-23R (green) in the smear of stimulated PBMCs was examined by confocal microscopy. The data shown are representative from one of the five CHB patients.
Figure 5.
In situ infiltration of IL-23 and IL-17 is significantly correlated with liver injury.
Correlations were analyzed between IL-23 or IL-17 and either plasma log10 HBV loads (copies/mL), serum ALT levels (U/L), HBsAg (IU/mL; concentration above 104), IL-8, or TNF-α. Solid line indicates a linear growth trend. r is the correlation coefficient. P-values are shown.
Figure 6.
HBsAg induces DCs and macrophages to produce IL-23 in a MR-dependent manner.
(A) Monocytes were isolated from PBMCs of healthy blood donors and mDCs were induced by culturing in the presence of GM-CSF and recombinant human IL-4 for five days. Then, the mDCs were stimulated by various doses of HBsAg or HBcAg. ELISA was used to detect the concentrations of IL-23 and IL-12 in the supernatants. (B) mDCs were stimulated by 50 µL serum from healthy controls (HC) or 50 µL serum from ACLF patients (HBsAg concentration: 79876.4 IU/mL, equivalent to 8 µg/mL) in a 200 µL culture system. For the blocking group, the ACLF serum had been pretreated with HBsAg antibody at a final concentration of 50 µg/mL for 30 min. The concentrations of IL-23 in the supernatants were detected by ELISA. The data represent one of three independent experiments with similar results. Error bars indicate SD. **P<0.01 vs. all other groups. (C) mDCs preincubated with mannan (100 µg/mL) or various doses of neutralizing anti-MR antibody for 30 min were stimulated by HBsAg or HBcAg (2 µg/mL, equal to that of HBsAg). Controls were preincubated with medium alone. ELISA was used to detect the concentration of secreted IL-23 in the supernatants. The data represent one of three independent experiments with similar results. Error bars indicate SD. **P<0.01; ***P<0.001 vs. the HBsAg group. (D) Monocytes were isolated from PBMCs of healthy blood donors and mDCs were induced by culturing for five days in the presence of GM-CSF (50 ng/mL) and recombinant human IL-4 (5 ng/mL). Then, mDCs were stimulated by HBsAg (2 µg/mL) alone or with recombinant human IL-10 (10 ng/mL) [27] or with TGF-β (2 ng/mL) [55], [56] for 48 hours. The supernatant was collected and IL-23 level was detected by ELISA. The data represent one of three independent experiments with similar results. Error bars indicate SD. **P<0.01.
Figure 7.
IL-23 production from DCs and macrophages stimulated by HBsAg is dependent upon endocytosis.
(A) Monocytes were isolated from PBMCs of healthy blood donors and mDCs were induced by culturing in the presence of GM-CSF (50 ng/mL) and recombinant human IL-4 (5 ng/mL) for five days. Then, the mDCs were stimulated by HBsAg with or without ammonium chloride (10 µM), chloroquine (10 mM), Dynasore (80 µM), or HBsAg antibody (50 µg/mL) for 40 hours. ELISA was used to detect the concentration of IL-23 in the supernatants. The data represent one of three independent experiments with similar results. Error bars indicate SD. ***P<0.001 vs. all other groups. DCs (B) and macrophages (C) were obtained from PBMCs for different stimulations and stainings. (a) Cells were stimulated with HBsAg (2 µg/mL) for 45 min and stained by isotype IgG (FITC). (b) Cells were directly stained with anti-MR mAb (FITC) without any stimulations. (c) Cells were stimulated with HBsAg (2 µg/mL) for 45 min and stained by anti-HBsAg mAb (FITC). (d) Cells were pretreated with MR-blocking antibody (10 µg/mL) for 30 min before the cells were stimulated by HBsAg (2 µg/mL) for 45 min and stained with anti-HBsAg (FITC). (e) Cells were pretreated with chloroquine (10 mM) for 10 min before stimulation by HBsAg (2 µg/mL) for 45 min and staining with anti-HBsAg (FITC). mDCs were stimulated by HBsAg (2 µg/mL) for 45 minutes, and the respective co-localizations of HBsAg with MR (D) or endosomal marker LysoTracker (E) were detected by confocal microscopy. (magnification 400×).
Table 1.
Clinical, viral and immunological criteria for the selection of study subjects.