Infection with the hepatitis C virus often results in chronic liver disease and subsequent development of liver fibrosis/cirrhosis.
The exact mechanisms leading to liver injury are only partly understood. However, there is clear evidence indicating that activation of the immune response is a critical factor for the pathogenetic processes leading to progressive tissue injury and ultimately cirrhosis. Accordingly, liver cell damage has been shown to be associated with the presence of an intra-hepatic inflammatory infiltrate. On the other hand, there is increasing data suggesting that natural killer cells, a major component of the intra-hepatic lymphocyte pool, may mediate anti-fibrotic effects. In mouse models NK cells have been shown to exert anti-fibrotic activity by killing activated stellate cells. Moreover, impaired activity of murine NK cells has been associated with progressive liver fibrosis 
Hepatic recruitment of lymphocytes such as NK cells is regulated by chemokines and their respective receptors. Lymphocytes sense chemokine concentration gradients and move toward increasing concentrations. Thus, both differential chemokine secretion in the inflamed liver and selective expression of different chemokine receptors on distinct lymphocyte subsets regulate hepatic recruitment of lymphocytes 
. Of note, recent data indicate that the chemokines receptor CXCR3 and its ligands, especially CXCL10 (IP-10) may play an important role in the regulation of HCV-associated liver fibrosis 
by yet incompletely understood mechanisms. Given the observed anti-fibrotic potential of NK cells we speculated that CXCR3-mediated recruitment of functionally distinct NK cell subpopulation might play a role.
To clarify this issue, we first analyzed the ex vivo phenotypic characteristics of circulating CXCR3(+) and CXCR3(−) NK cells. CXCR3-expressing NK cells were found in both the CD56Dim and the CD56Bright subset although frequency of CXCR3(+) NK cells was higher among CD56Bright cells. More importantly, comparing CD56BrightCXCR3(+), CD56BrightCXCR3(−), CD56DimCXCR3(+), and CD56DimCXCR3(−) NK cells separately we observed a progressive decrease in surface expression of the maturation markers CD27, CD62L, and CD127. In addition, the CXCR3(+) subset displayed significantly higher co-expression of the NK cell receptors NKG2A, NKG2C, and NKp44. Thus, our data suggest that expression of CXCR3 defines distinct NK cell subsets.
This concept was confirmed at the functional level, because in healthy controls CXCR3(+) and CXCR3(−) NK cell subsets differed significantly with respect to cytolytic activity and IFN-γ production after exposure to human hepatic stellate cells, with CD56BrightCXCR3(+) NK cells displaying the strongest activity against HSC. Activated HSC are considered to critically contribute to the establishment of hepatic fibrosis via excessive production of collagen. Thus, effective killing of HSC by CD56BrightCXCR3(+) NK cells suggests that this specific lymphocyte subset may play an important role in the regulation of HCV-associated liver fibrosis. However, in hepatitis C high levels of CXCR3 ligands have been associated with progressive liver cell damage and fibrosis 
. Recently, these counterintuitive data could – at least in part – be explained by the identification of a chemokine antagonism in HCV infection which prevents CXCR3-mediated recruitment of immunocompetent cells into the liver thereby resulting in extra-hepatic accumulation of CXCR3-expressing cells 
. Accordingly, we found chronic hepatitis C to be associated with a significantly increased frequency of circulating CXCR3-expressing CD56Bright NK cells. However, we also observed a significantly increased frequency of CXCR3(+)CD56Bright NK cells in livers with advanced stages of fibrosis, suggesting that mechanism(s) other than dys-regulated hepatic migration may also play a role.
Indeed, our data indicate that hepatitis C may be associated with impaired activity of peripheral CXCR3(+) NK cells because we found that in HCV infected patients_specifically the CD56BrightCXCR3(+) subset displayed decreased degranulation and IFN-γ secretion in response to human HSC. Unfortunately, number of intra-hepatic NK cells in our study was insufficient to study activity of liver NK cells against HSC. Thus, it remains to be clarified whether intra-hepatic CXCR3-expressing NK cells also show dys-regulated activity against activated HSC or even exert pro-fibrotic effects.
The exact mechanisms responsible for this impaired functional activity of circulating CD56BrightCXCR3(+) NK cells in HCV infection remain incompletely understood but may involve altered surface expression of the NK cell receptor NKG2D. Mouse models indicate that NK cell killing of activated hepatic stellate cells is mediated via activating NK cell receptor NKG2D 
. Accordingly, we found that blocking of NKG2D significantly reduced NK cell degranulation and IFN-γ secretion following co-incubation with HSC. Thus, our finding of decreased NKG2D surface expression in hepatitis C would be an intriguing explanation for our observation of impaired NK cell activity against HSC. However, the role of NKG2D in HCV infection is discussed controversially as some authors reported down-regulated expression of this NK cell receptor 
whereas other studies found increased surface expression in hepatitis C 
. In line with data presented by Sène and colleagues we found that differences in HCV(+) patients and healthy controls only affected expression levels of NKG2D, but not the frequency of NKG2D-positive cells 
However, reduced expression of NKG2D was not specific for the CXCR3(+)CD56Bright subset, indicating that other mechanisms may also play a role.
Taken together, we show that distinct NK cell subsets can be distinguished based on CXCR3 surface expression. Intra-hepatic accumulation of functionally impaired CD56BrightCXCR3(+) might be involved in the progression of HCV-induced liver fibrosis.