Coinfection with hepatitis C has become a major health problem in HIV-infected patients [1,2]. However, even in the setting of HIV coinfection, a considerable proportion of patients can eliminate hepatitis C virus (HCV) via their residual immune system.
Recently, we provided first data suggesting that natural killer (NK) cells, a central component of the innate immune system, may play an important role in this context .
Activity of NK cells is tightly regulated by signals delivered by activating and inhibitory NK cell receptors. However, recent data clearly indicate that costimulatory molecules such as the tumour necrosis factor receptor (TNFR) family member CD27 also are involved in modulating NK cell functions.
In CD27-deficient mice, the absence of CD27 has been shown to result in decreased cytokine production by NK cells following in-vivo activation . Accordingly, Vossen et al.  showed that expression of CD27 defines specific NK cell subsets with CD27(+)CD56BrightNK cells displaying the strongest interferon-gamma (IFN-γ) secretion.
Given the important role of IFN-γ in NK cell-mediated control of HCV replication, we analysed the potential impact of CD27(+)CD56Bright in modulating the natural course of acute hepatitis C in HIV-positive individuals.
Methods and patients
Twenty-eight HIV-positive patients with acute hepatitis C, all from the Bonn/Cologne area in Germany, were studied, including 13 patients with self-limited course of HCV infection [men 13/13 (100%); age: 39 (31–46) years; CD4+ cell count: 646 (358–873) cells/μl; HCV load: 14.8 × 106 (<0.1 – 69 × 106) IU/ml; alanine aminotransferae (ALT): 784 (23–1182) IU/ml] and 15 patients who subsequently developed chronic hepatitis C [male 15/15 (100%); age: 42 (34–56) years; CD4+ cell count: 686 (348–1377) cells/μl; HCV load: 12.5 × 106 (<0.1 – 69 × 106) IU/ml; ALT 634 (82–2178) IU/ml]. All patients were studied during the acute phase of HCV infection. As controls, 17 HIV-positive patients with chronic hepatitis C [male 15/17 (88.2%); age: 51 (28–72) years; HCV load: 1.7 × 106 (<0.1 – 5.8 × 106) IU/ml; ALT: 114 (33–303) IU/ml] and 12 individuals with HIV mono-infection [men 11/12 (91.7%); age: 47 (34–64) years; CD4+ cell count: 700 (421–1199) cells/μl; ALT: 74 (19–40) IU/ml] were analysed. All patients were HIV RNA-negative under effective HAART (Abbott RealTime HIV-1 assay). As additional controls, we studied 23 HCV mono-infected patients [men 14/23 (60.8%); age: 47 (34–64) years; ALT: 74 (19–40) IU/ml)] as well as 30 healthy, HIV-negative/HCV-negative donors [men 16/30 (53.3%); age: 30 (21–51) years].
Informed consent was obtained from all patients. The study had been approved by the local ethics committee of the University of Bonn.
Flow cytometric analysis
For FACS analysis, the following antibodies were used: anti-CD56, anti-CD3, anti-CD27, anti-CD70 (Biolegend, Fell, Germany) and anti-INF-γ (R&D Systems, Wiesbaden, Germany). After staining, cells were washed and analysed on a FACSCalibur flow cytometer using the Flowjo 7.2.2 software package (Treestar, Ashland, Oregon, USA).
Natural killer cells were immunomagnetically separated from total peripheral blood mononuclear cells (PBMC) by depletion of non-NK cells using MACS cell separation technology (Miltenyi Biotech Bergisch Galdbach, Germany). Purified NK cells were cultured in RPMI1640 medium (PAA, Germany) supplemented with 10% FCS and 1% penicillin/streptomycin at 37°C and 5% CO2 overnight and were stimulated with 25IU recombinant human interleukin (IL)-2 (R&D Systems, Wiesbaden, Germany) or rhIL-12 (0.5 ng/ml; ebioscience, Frankfurt, Germany).
Isolation of CD27(+) and CD27(−) natural killer cell subsets
Purified NK cells were stained with phycoerythrin-conjugated anti-CD27 and allophycocyanin-conjugated anti-CD56. Then, CD27(+)CD56Bright and CD27(−)CD56BrightNK cells, respectively, were sorted using a BD FACSARIA III cell sorter (BD Biosciences, Heidelberg, Germany).
HUH7A2 HCV Replicon cells
HUH7A2HCV Replicon cells  were kindly provided by V. Lohmann and R. Bartenschlager (University of Heidelberg, Germany). Cells were grown in high glucose (4.5 g/l) Dulbecco's modified Eagle's medium supplemented with glutamine (PAA), 10% fetal calf serum (FCS), nonessential amino acids (Biochrom, Berlin, Germany) and 1% penicillin/streptomycin (PAA). Blasticidin S hydrochloride (3 μg/ml) and G418 (1 mg/ml) (PAA) were added to cells containing subgenomic replicons. Huh7A2HCV cells were passaged twice a week and were seeded at a dilution of 1 : 4.
Intracellular staining of interferon-gamma
IL-2-stimulated NK cells were cocultured with HUH7Replicon cells at 1 : 1 effector:target (E : T) ratio at 37°C for 5 h. Brefeldin A (10 μg/ml; Sigma-Aldrich, Munich, Germany) was added after 1 h of coculture. Next, cells were harvested and washed. Finally, cells were stained with anti-CD27, anti-CD56 and anti-CD3 mAbs, fixed and permeabilized using Cytofix/Cytoperm (BD Biosciences), followed by intracellular staining with anti-IFN-γ mAb and FACS analysis.
Blocking and stimulation experiments with mAbs
CD27(+)CD56BrightNK cells were coincubated with HUH7Replicon cells at 1 : 5 (E : T) ratio in the presence of anti-IFN-γ (clone NIB42; 10 μg/ml; Biolegend). Blockade of CD70 on HUH7Replicon cells was achieved by addition of anti-CD70 antibody (9.5 μg/ml, clone BU 69; Abcam, Cambridge, UK) 1.5 h prior to coincubation with purified NK cells. Alternatively, NK cells were coincubated with soluble anti-CD27 (Biolegend) prior to coculturing with HUH7Replicon cells.
In stimulation experiments, IL-12 activated NK cells were cultured in the presence of plate-bound anti-CD27 (20 μg/ml, clone O323; Biolegend) for 24 h. The resulting NK cell supernatants were harvested and added to HUH7Replicon cells for another 24 h. Plate-bound anti-IgG1 (clone MOPC-2; Biolegend) was used as isotype control.
HUH7Replicon cells were seeded in 48-well plates. After 3 h, medium was removed and replicon cells were cocultured with PBMC at an E : T ratio of 1 : 1, sorted NK cells at 1 : 5 or with supernatants from the stimulation experiment for 24 h. The assay was performed using the Steady-Glo Luciferase assay System (Promega, Mannheim, Germany) and luminescence was measured with Tecan infinite M200 (Tecan, Männedorf, Switzerland).
Statistical analyses were performed using GraphPad Prism Version 5.0a (GraphPad Software Inc, San Diego, California, USA) and the SPSS 17.0 (SPSS, Chicago, Illinois, USA) statistical package. Mann–Whitney U tests were used to compare NK cell phenotype and cytolytic response between samples. Paired Student's t-tests were used to estimate paired samples. A two-sided P value of less than 0.05 was considered significant. Linear regression was used to examine correlation.
First, we compared frequency of circulating CD27(+)CD56BrightNK cells among the analysed patient groups. As is shown in Fig. 1a, the frequency of CD27(+)CD56BrightNK cells was significantly lower in HIV-positive individuals than in healthy or HCV mono-infected individuals, irrespective of HCV coinfection.
When HIV-positive patients with chronic and self-limited course of acute hepatitis C were analysed separately, we found high frequency of CD27(+)CD56Bright NK cells as well as high CD27expression to be associated with spontaneous clearance of acute HCV infection (Fig. 1b).
Moreover, we found CD27(+)CD56Bright NK cells to be significantly more effective in blocking viral replication in vitro than in the CD27(−) subpopulation (Fig. 1c), suggesting a role for CD27-expressing NK cells in inhibition of HCV replication. Due to the low number of cells available from HIV-positive patients with acute hepatitis C, these experiments could only be performed in healthy controls.
However, we found that in HIV-positive patients, the frequency of CD27(+)CD56Bright NK cells was positively correlated with NK cell-mediated inhibition of viral replication (Fig. 1d). Of note, when HIV-positive patients who were able to spontaneously clear HCV and those developing chronic infection were analysed separately, we found frequency of CD27(+)CD56Bright NK cells to be correlated with inhibition of viral replication only in patients with self-limiting course of HCV infection (Fig. 1e). No such association was observed in HIV-negative individuals (Fig. 1f).
Secretion of IFN-γ is considered to represent a major mechanism of NK cell antiviral efficacy [3,7]. Comparing IFN-γ production of CD27(+) and CD27(−) NK cells following coincubation with HUH7Replicon cells, we found CD27(+)CD56Bright NK cells to produce significantly more IFN-γ than CD27(−)CD56Bright NK cells in both healthy individuals (Fig. 2a) and HIV-positive patients with acute hepatitis C (Fig. 2b), suggesting that higher IFN γ production might be involved in effective anti-HCV activity of CD27(+) NK cells.
Accordingly, we found that blocking of IFN-γ significantly reduced the ability of CD27(+)CD56Bright NK cells to block viral replication. Interestingly, we found that CD27(+)CD56Bright NK cells and CD27(−)CD56Bright NK cells from healthy controls did not differ significantly with respect to in-vitro inhibition of HCV replication after blocking of IFN-γ (Fig. 2c). An important role for IFN-γ could also be confirmed in HIV-positive patients with acute hepatitis C (Fig. 2d).
Finally, we studied whether CD27/CD70 interactions were directly involved in antiviral activity of CD27(+)CD56Bright NK cells. To this end, we first blocked CD27 using a specific antibody. As is shown in Fig. 2e, blocking of CD27 significantly reduced antiviral activity of healthy NK cells. Similar observations were made in HIV-positive patients with self-limiting course of acute hepatitis C but not in those who subsequently developed chronic hepatitis C (Fig. 2f). Moreover, we found that prestimulation of NK cells with an immobilized CD27-specific mAb significantly increased the capacity of CD56Bright NK cells to block viral replication (Fig. 2g). Finally, we foundHUH7HCVReplicon cells to display a robust expression of CD70, the natural ligand for CD27. Of note, blocking of CD70 also reduced antiviral NK cell activity (Supplemental Figure 1, http://links.lww.com/QAD/A540).
Recent reports suggest that costimulation via CD27, a member of the tumour necrosis factor (TNF) receptor family, may play an important role in shaping NK cell responses [4,8].
In line with this concept, we present first data indicating that CD27 may be involved in modulation of anti-HCV NK cell activity and show that high expression of CD27 is associated with self-limited course of acute HCV infection in HIV-positive patients.
First, we demonstrate that CD56BrightNK cells expressing CD27 are significantly more effective in inhibiting viral replication in vitro than their CD27-negative counterpart. Accordingly, frequency of CD27-expressing CD56Bright NK cells was positively correlated with in-vitro antiviral NK cell activity. Moreover, we found CD27(+) CD56Bright NK cells to produce significantly more IFN-γ than the CD27(−) subset, which is in line with earlier reports by other groups [5,9]. IFN-γ secretion has been shown to be critical for in-vitro antiviral activity of both NK cells [3,7] and HCV-specific CD8+ T cells . Confirming these results, we found that blocking of IFN-γ significantly impaired the ability of CD27(+)CD56Bright NK cells to inhibit viral replication, suggesting that strong IFN-γ production was involved in antiviral activity of CD27-expressing NK cells. Moreover, we observed a high frequency of CD27(+)CD56Bright NK cells in the acute phase of hepatitis C to be associated with a self-limited course of infection in HIV-positive patients, supporting an in-vivo relevance of our findings.
CD27 expression has been postulated to define a phenotypically and functionally distinct NK cell population [8,11], which would be in line with our observation of CD27(+)CD56Bright NK cells being characterized by a strong IFN-γ secretion.
However, several of our findings suggest that CD27 may also play a direct role in antiviral NK cell activity. First, we found that triggering CD27 significantly increased the ability of CD56Bright NK cells to block viral replication. Second, we found that blocking of CD27 on NK cells significantly reduced NK cell-mediated inhibition of HCV replication. Finally, we found that interrupting CD27/CD70 interactions with anti-CD70 also impaired antiviral NK cell activity.
Taken together, our data suggest that CD27 expression on CD56Bright NK cells defines an NK cell subpopulation with strong anti-HCV capacity that can be modulated via costimulatory signals received via CD27.
The authors thank Claudia Zwank for perfect technical assistance.
This work was supported by the German Research Foundation (DFG SFB/TR 57), the H. W. and J. Hector Foundation [grant number M42], by a grant from the BMBF (German Ministry for Science and Education) (01KI0791), a NEAT Gilead research grant [programme number LSHP-CT-2006–037570], the DZIF TTU Hepatitis Project 8.3 and the German Center for Infection Research (DZIF).
Conflicts of interest
The authors who have taken part in this study declared that they do not have anything to disclose regarding funding or conflict of interest with respect to this manuscript.