Differential dysfunction in dendritic cell subsets during chronic HCV infection

Abstract

Hepatitis C virus (HCV) infection is a leading cause of chronic liver disease with over 200 million individuals infected worldwide. The vast majority of acutely infected humans develop chronic infection, which is characterized by attenuated antiviral T-cell responses. The mechanisms leading to such attenuation/suppression are poorly understood. It has been proposed that dysfunction of antigen-presenting cells (APC) may underlie the downregulation of antiviral immune responses. However, studies using bulk or in vitro-derived APC populations have resulted in conflicting reports. In this study, we evaluated the functional and immunophenotypic features of ex vivo-purified dendritic cell (DC) subsets during chronic HCV infection. We found that plasmacytoid DC (PDC) from HCV-infected patients (HCV-PDC) showed a striking deficit in IFN-alpha production in response to CpG stimulation. In addition, we found that myeloid DC (MDC) from these patients showed a diminished capacity to induce a mixed lymphocyte response (MLR), correlating with lower levels of HLA-DR and CD86 expression and higher IL-10 production in response to poly-IC stimulation. In contrast, HCV-PDC showed increased ability to stimulate an MLR. Of note, within the HCV-PDC preparation, we noted a distinctly expanded DC subset that expressed some markers of MDC, but showed significantly lower HLA-DR and CD86 expression, suggesting an expansion of DC at an immature/intermediate stage of differentiation. Our studies demonstrate distinct and contrasting dysfunctional features in DC subsets and underscore the importance of evaluating APC subpopulations separately.

Figure 1. Comparable immunophenotype and purity of MDC obtained from HCV patients and healthy subjects

These dotplots represent bead-purified MDC from a healthy subject (top panel) and an untreated chronic HCV patient (bottom panel). These preparations showed the expected CD11c^{bright +}/CD123^{moderate + to −} phenotype with >90% purity. This is representative of 10 subject pairs, evaluated in parallel.

Figure 2. Lower allostimulatory capacity of MDC from HCV-infected patients, compared to healthy donors

Purified MDC from either healthy donors (circles) or from HCV-infected patients (triangle) were used to stimulate purified allogeneic T-cells from a healthy subject (third party). Cultures were pulsed with ³H-thymidine on day 5 and harvested and counted on day 6. One healthy subject and one HCV patient were always evaluated as a pair (as indicated by the lines, n=8 pairs) and exposed to the same third party responder T-cells in all experiments. The data represent a stimulator:responder ratio of 1:6. Similar pattern was observed at other ratios. Ex vivo-purified MDC from HCV subjects showed a lower capacity to induce an MLR (17055 ± 3557 CPM), compared to healthy MDC (24970 ± 3394 CPM; p<0.05).

Figure 3. Slightly lower HLA-DR and CD86 expression on HCV-MDC

The histograms (A and C) depict representative examples of HLA-DR (A) and CD86 (C) expression on healthy and HCV-MDC. The graphs (B and D) represent the average mean fluorescence intensity (MFI) of HLA-DR (B) and CD86 (D) expression on poly-IC-stimulated MDC from 10 healthy-HCV pairs. MDC rested in media alone also showed a similar trend (data not shown). MDC from HCV patients showed slightly but consistently lower HLA-DR and CD86 expression compared to healthy MDC (p<0.04).

Figure 4. Production of IL-10 and IL-12 by MDC from healthy subjects and HCV patients

Purified MDC were cultured for 2 days in media alone or with Poly-IC (50 μg/ml). The levels of IL-10 (A) or IL-12 (B) were measured in 2-day supernatants by ELISA. In contrast to healthy MDC, HCV-MDC showed constitutive IL-10 production in media alone. The level of IL-10 produced by HCV-MDC in response to Poly-IC was significantly higher than that produced by healthy MDC (p<0.05). In contrast, IL-12 production by healthy MDC tended to be higher than HCV-MDC, although the difference was not statistically significant (p<0.09).

Figure 5. A significantly expanded subset of DC in PDC preparations from HCV-infected patients

Panel A shows representative dot-plots of bead-purified PDC preparations from a healthy subject (top panel) and an untreated HCV-infected patient (bottom panel). PDC were isolated from PBMC using a Miltenyi Biotech microbead kit (based on BDCA-4 expression). As expected, healthy PDC showed >90% purity, with a predominant population showing the characteristic CD123^{bright +}/CD11c⁻ phenotype. However, the purified PDC also showed a distinct subpopulation showing a CD11c⁺/CD123^{moderate to dim +} phenotype, which was greatly expanded in HCV-PDC preparation. Panel B shows an analysis of 10 healthy-HCV subject pairs. The bars represent the mean percentage (+ SEM) of this DC-subset in PDC preparations from healthy subjects (6.4% ± 2.9%) vs. HCV patients (31.6% ± 5.4%), showing a significant expansion of this subset in HCV patients (p<0.01).

Figure 6. Dramatically deficient interferon-α production by HCV-PDC

PDC were cultured for 2 days in media alone, or with CpG (6 μg/ml). IFN-α levels in supernatants were quantified by ELISA. HCV-PDC (triangles) showed a remarkably reduced capacity to produce IFN-α in response to CpG, compared to healthy PDC (circles; p<0.01). PDC produced negligible IFN-α in media alone.

Figure 7. Higher allostimulatory capacity of PDC from HCV-infected patients, compared to healthy donors

PDC preparations from either healthy donors (circle) or HCV-infected patients (triangle) were used as stimulators with purified allogeneic healthy T-cells. One healthy subject and one HCV patient were always evaluated as a pair (as indicated by the lines, n=8 pairs) and exposed to the same third party responder T-cells in all experiments. The data represent a stimulator:responder ratio of 1:6. Similar pattern was observed at other ratios. PDC from HCV subjects showed a significantly elevated capacity to induce an MLR (Healthy; 8911 ± 1917 vs. HCV; 13609 ± 2548; p<0.05).

Figure 8. Expression of HLA-DR and CD86 on PDC preparations from healthy subjects and HCV-infected patients

The histograms (A and C) depict representative examples of HLA-DR (A) and CD86 (C) expression on PDC preparations from a healthy subject and an HCV patient. The mature PDC and immature DC subset were evaluated separately (see text and Figure 5). The graphs represent the average mean fluorescence intensity (MFI) of HLA-DR (B) and CD86 (D) expression on CpG-stimulated PDC preparations from 10 healthy-HCV pairs. PDC rested in media alone also showed a similar pattern (data not shown). The mature PDC from healthy subjects or HCV patients showed comparable HLA-DR and CD86 expression. In contrast, the immature DC subset from HCV patients showed lower HLA-DR and CD86 expression. Of note, the HLA-DR and CD86 expression on this subset were significantly lower than that of MDC (Figure 3). CD86 expression on this subset was higher than that of mature PDC (p<0.05).