Interferon-α-Induced Changes to Natural Killer Cells Are Associated with the Treatment Outcomes in Patients with HCV Infections

Aim. We analyzed the pretreatment natural killer (NK) cell functions with the aim of predicting the sustained virological response (SVR) or the interleukin (IL) 28B polymorphism that is strongly associated with the treatment response. Methods. The peripheral NK cells from chronic hepatitis patients with HCV genotype 1 and high virus titers were activated using a Toll-like receptor (TLR) 4 ligand and IFN-α. The cell surface markers were evaluated using a flow cytometric analysis, and IFN-γ production was evaluated using an enzyme-linked immunosorbent assay (ELISA). The genotyping of the polymorphisms in the IL28B gene region (rs8099917) on chromosome 19 was performed on the DNA collected from each patient. Results. The production of IFN-γ was significantly higher in the SVR patients compared with the no-response (NR) patients, whereas the cell surface markers were similar between the SVR and the NR patients. There were no significant differences found in the IL28B genotype distribution associated with the production of IFN-γ. Conclusion. Differences in the NK cell functions were observed between the SVR patients and the NR patients, suggesting that NK cells play a potential role in the treatment response independent of the IL28B genotype.


Introduction
The hepatitis C virus (HCV) is the major cause of chronic liver disease, with an estimated global prevalence of 2.5%, that is, 170 million people infected worldwide [1], and is a leading cause of cirrhosis, hepatocellular carcinoma, and liver transplantation [2]. Antiviral treatment, which is based on the combination of pegylated interferon-(IFN-) and the nucleoside analog ribavirin (RBV), is associated with a sustained virologic response (SVR), that is, serum HCV RNA negatively for 6 months after the cessation of the antiviral therapy.
The HCV genotype, viral load, age, and fibrosis stage are well known as pretreatment variables [3][4][5][6]; moreover, single nucleotide polymorphisms (SNPs) located in the region of the interleukin (IL) 28B gene have been strongly associated with an SVR [7,8]. Although the IL28B genotype can be useful when making treatment decisions, this variable alone is not a perfect predictor of the treatment outcome.
The natural killer (NK) cells involved in innate immunity play central roles of defense against viral infections through a direct cytotoxic effect in the destruction of the virus-infected target cells and the production of inflammatory cytokines [9].

Hepatitis Research and Treatment
Furthermore, in contrast to T cells, NK cells do not require priming for the recognition of the target cells. There has been growing evidence that NK cells have an important role in mediating the IFN-induced viral clearance of chronic HCV infections when the cell surface markers or the cytotoxicity of NK cells is evaluated [10]. Furthermore, we previously demonstrated that activated NK cells have cytotoxity against autologous biliary epithelial cells in the presence of IFN- [11].
We hypothesize that because NK cells rapidly produce IFN-using several cytokines including IFN-, pretreated changes of NK cells in the presence of an ex vivo IFNstimulation would become an indicator for the virological response; likewise, insulin resistance is another pretreatment predictor as we previously reported [12]. Furthermore, since the IL28B gene has been strongly associated with an acute HCV clearance [13], the relationship between the IL28B genotype and the NK cell functions that play a central role in the early phase viral clearance was investigated.

Study Subjects.
A total of 20 patients with a chronic HCV genotype 1 infection and high viral load (>5.0 log IU/mL) were studied. All of the patients were treatment-naïve prior to their enrollment. The study samples were collected prior to the start of a standard pegylated IFN-and RBV therapy. The patients who had undetectable HCV RNA for at least 6 months after treatment were classified as sustained viral responders (SVR, = 8); those who had undetectable HCV RNA in the end of the treatment, however, detectable after the treatment, were classified as partial responders (PR, = 8), and those who had persistently detectable HCV RNA during the therapy were classified as null responders (NR, = 4).
All of the subjects gave their written informed consent and the experimental protocols were conducted under the Guidelines of the Research Ethics Committee of Kyushu University.

IL28B Polymorphism
Analysis. The single nucleotide polymorphism (SNP) testing of the IL28B gene (rs8099917) was completed for all of the patients using a real-time PCR method on the genomic DNA extracted from the whole blood samples. The heterozygotes (TGs) or the homozygotes (GGs) of the minor allele (G) were described as having the IL28B minor allele, whereas the homozygotes for the major allele (TT) were described as having the IL28B major allele [7].

Isolation of Peripheral Blood Mononuclear Cells and NK Cells. The peripheral blood mononuclear cells (PBMCs)
were separated from the heparinized fresh blood using a Ficoll-Isopaque gradient centrifugation technique. The NK cells were negatively isolated from the PBMC using magnetic beads (NK isolation kit; Miltenyi Biotec, Auburn, CA, USA). A viability of >95% using a trypan blue dye exclusion technique and a purity of >90% using flow cytometry for CD56 positives were considered to be acceptable [11].

IFN-Production from NK Cells.
In an effort to identify the nature of the cytokines that were involved in promoting the NK cell effector function, the supernatants from the NK cells that have been cultured for 2 days with or without pegylated IFN-(3 g/mL, Schering-Plough, Kenilworth, NJ, USA) and lipopolysaccharide (LPS) (10 g/mL, Invitrogen, San Diego, CA, USA) were analyzed for the production of IFN-. Pegylated IFN-instead of IFN-was used as pegylated IFN-working in vivo during the standard pegylated IFN-and RBV therapy. The assays were performed using a sandwich ELISA (R&D Systems, Minneapolis, MN, USA) with a combination of unlabeled and biotin-coupled or enzyme-coupled monoclonal antibodies to each cytokine. Both of the activated and the resting NK cells were evaluated for cell surface markers.

Flow Cytometric Analysis of the Cell Surface Antigens.
In order to evaluate the cell surface antigen expression in the PBMC and NK cells (1 × 10 6 ), the cells were stained at 4 ∘ C in the dark for 30 min, washed twice in 2 mL of a phosphate-buffered saline containing 1% bovine serum albumin and 0.01% sodium azide, and were fixed in 500 L of 1% paraformaldehyde. The cells were stained for CD3, CD16, and CD56 (BD Biosciences, San Diego, CA, USA). A twocolor flow cytometry was performed using an FACSnCalibur Flow Cytometer (BD Biosciences).

Statistical Analysis.
All of the experiments were performed in triplicate and the data points shown are the mean values of the results of these triplicates. The comparisons between the points for certain data sets were expressed as the mean ± standard deviation (SD), and the significance of differences was determined by Student's -test. All of the analyses were 2-tailed and the values <0.05 were considered to be significant. The statistical analyses were performed using the Intercooled Stata 8.0 software program (StataCorp, College Station, TX, USA).

Subjects.
The characteristics of patients who were undergoing the standard pegylated IFN-and RBV combined therapy are summarized in Table 1. Patients who experienced a relapse following the administration of the standard therapy (partial responders) were excluded from this study. As expected, the IL28B TT genotype was dominant in the SVR patients (75%) and not dominant in the PR patients (50%) or the NR patients (50%).

Phenotypes and IFN-Production in NK Cells before and after IFN-Stimulation.
In order to evaluate the role of NK cells, we enriched the NK cells from the PBMC using magnetic beads. The CD3 positive fraction was completely eliminated during the isolation of the NK cells ( Figure 1). The NK cells were incubated with or without LPS and IFNfor 2 days, as the NK cells could not produce detectable amounts of IFN-solely with IFN-. CD56 bright NK cells increased after stimulation (Figures 2(a) and 2(b)).  The NK cells were classified as CD56+CD16−, CD56− CD16+, and CD56+CD16+ phenotypes (Figure 2(c)) and these phenotypes changed after stimulation (Figure 2(d)).

Discussion
Because the NK cell cytotoxity and the presence of cell surface markers correlate with the virological response, NK cells can serve as biomarkers of a patient's IFN-responsiveness [14,15]. In addition, the IFN--induced modulation of the signal transducer and activator of transcription (STAT) 1 and STAT 4 phosphorylation underlies the polarization of the NK cells in an HCV infection [16][17][18]. During the IFN-treatment, the polarized NK cells are recruited from the peripheral blood into the liver [19]. Therefore, it may be inappropriate to evaluate the peripheral NK cells for the general functions of the NK cells that are undergoing treatment. To this end, we collected the peripheral NK cells prior to treatment and evaluated the NK cell functions with IFNex vivo with the aim of predicting the virological responses. Although the interactions between the killer cell Ig-like receptors (KIRs) expressed on NK cells and the HLA expressed on target cells are well known to play a key role in NK cell activation, it is difficult to obtain HCV-infected autologous target cells for the evaluation of the cytotoxity of the NK cells. Therefore, we instead evaluated the IFN-production relating to the NK cell functions. It has been previously reported that a polarized NK cell phenotype can be induced by a chronic exposure to HCVinduced IFN-, contributing to liver injury through tumor necrosis factor-related apoptosis inducing ligand (TRAIL) expression and cytotoxity [20]. Therefore, we hypothesized that we could predict an outcome of SVR or NR according  Figure 3: The characterization of the before and after stimulation NK cells, as classified by the SVR and NR grouping. Between (a) and (c), there were no statistically significant differences between the SVR and NR groups associated with the cell surface markers on the NK cells. However, the NK cells from the SVR group did produce higher amounts of IFN-than those from the NR group ( < 0.05) (d The study size was small to evaluate NK cell function fully; however, we found that we could not predict an outcome of SVR according to the changes in cell surface markers, and, furthermore, the NK cell phenotype and function did not differ among the subgroups with different IL28B genotypes as had been previously reported [21]. These findings are consistent with the fact that NK cells do not respond to IFNvia different signal transductions of type III IFN (including IL28B) [22,23]. Additionally, there were no differences of this IFN-production between the patients with early virologic response (EVR) and no-EVR or rapid virologic response RVR and no-RVR (data not shown). This suggests that IFNproduction from peripheral NK cells does not correlate sharply with the declining dose of HCV-RNA for the initial 2 or 4 weeks of pegylated IFN-and RBV treatment.
As the NK cells kill the target cells through the directed release of perforin, granzyme-containing granules, and TRAIL which is associated with the control of an HCV infection [24], the combined effects of KIR, TRAIL, and IL28B in the context of IFN treatment are possible [25]. However, the evaluation of these effects would require additional studies with a larger number of patients. In addition, while the different inhibitory/activating receptors on the NK cells are well known to be associated with antiviral activity [26][27][28][29], further evaluations of the related cell surface markers must be performed.
In conclusion, these results demonstrate that the ex vivo NK cell response may serve as a biomarker for IFN-treatment responsiveness independent of the IL28B genotype.