In this study, we demonstrated that HCV core, NS3 and NS4 proteins specifically and selectively induce IL-10 production when incubated for 16 hours with PBMC. The cells producing IL-10 in response to the HCV proteins were identified by flow cytometry as CD14-CD36+ with lymphoid light scatter characteristics. We previously reported constitutive ex vivo production of IL-10 by this PBMC subset and that this production rapidly stops in vitro . These cells do not fit into one of the major subpopulations of circulating immune cells, but can be reliably isolated from all healthy individuals tested at a frequency of up to 1%. The ex vivo frequency of IL-10-producing PBMC was generally equivalent in the four distinct groups we studied, indicating that the differences seen after 16 hour culture were caused by differential responses to the HCV proteins. While 16 hour culture with HCV proteins increased the number of IL-10-producing cells in all groups, the frequency of IL-10+ cells was higher in PBMC from HIV-infected individuals and uninfected controls compared to PBMC from HIV/HCV coinfected individuals or HCV-infected individuals. This demonstrates a propensity for HCV-induced IL-10 production by PBMC from uninfected individuals that is exaggerated by HIV infection and attenuated by chronic HCV infection, even in the setting of HIV coinfection.
We used recombinant HCV core, NS3 and NS4 to investigate cytokine production in response to HCV proteins. Although these three proteins have little sequence or structural similarity and their known functions in HCV replication are very different [23, 24], we found that, collectively or individually, HCV core, NS3 and NS4 proteins induced IL-10 production in CD14-CD36+ PBMC. Others have also reported that recombinant HCV core, NS3 and NS4 proteins all induce IL-10 production [16, 25, 26]. However, in these studies, monocytes were identified as the responding cells and, in direct contrast to our results, there was a greater effect on monocytes from HCV-infected individuals than on monocytes from uninfected controls [16, 27]. These studies utilized monocytes isolated by overnight adherence to plastic, rather than freshly isolated PBMC. Despite varying protocols and different characteristics of the responding cells, the central point is that multiple distinct HCV proteins stimulate IL-10 production by PBMC. There are other examples of redundancy in HCV protein effects as mice transgenic for hepatic HCV structural or non-structural polyproteins develop hepatic steatosis characteristic of chronic HCV infection . Such functional redundancy of viral proteins is not unique to HCV, as three different HIV proteins (nef, tat and vpr) have all been reported to induce IL-10 production by various PBMC subsets [29–33].
In our study, structural and non-structural HCV proteins induced IL-10. Having both structural and non-structural proteins with redundant immunosuppressive functions may be important in sustaining infection. The availability of these proteins in circulation is temporally separated, and not concurrent. Non-structural proteins are only expressed in actively infected cells, while the core protein can be found in plasma at moderately high levels throughout chronic infection [34, 35]. Having multiple proteins that induce IL-10 supports production of an immunosuppressive cytokine whether or not HCV replication is upregulated or infectious virions are present.
In parallel with the redundancy in HCV protein-induced IL-10 production, additional HCV escape mechanisms within both the innate and adaptive immune system have been extensively described. Subversion of the eukaryotic anti-viral interferon-inducible system is a potent evasion strategy employed by many viruses, and several HCV proteins suppress host proteins in this system. HCV E2 inactivates PKR , and the HCV serine protease NS3/4a causes proteolysis of Toll-IL-1 receptor domain-containing adaptor inducing IFN-β (TRIF) , and inhibits phosphorylation of IRF-3, preventing nuclear translocation and IFN-β gene activation . HCV E2 has also been associated with decreased NK cell function and HCV core stabilizes HLA-E expression and subsequently downregulates NK-mediated cytotoxicity [39–41]. HCV core, E1, NS3, and NS4 are all associated with impaired dendritic cell maturation, thus potentially modulating the function of this integral link to adaptive cellular immunity [16, 27, 42]. HCV core also downregulates in vitro anti-HCV T cell responses [17, 43, 44]. While the exact mechanism of HCV protein interaction is unclear, IL-10 inhibits dendritic cell maturation, resulting in an immature phenotype similar to that seen after exposure to HCV core . Not only do HCV proteins modulate immune responses, it is clear that more than one protein has evolved to share similar functions. Production of IL-10 by the cell population described in this paper may stunt dendritic cell maturation, and impair development of anti-HCV cellular immunity through this pathway as well as through a direct effect on responding T cells.
The mechanism for IL-10 induction in PBMC by HCV proteins is unknown, but processing of soluble HCV antigens through the endocytic pathway is unnecessary as blocking antigen processing had no effect on constitutive or induced IL-10 production. A higher percentage of cells from HIV-infected individuals than uninfected controls produced IL-10 in response to HCV proteins. This appears unrelated to lower T cell function in HIV infection as depletion of CD3+ PBMC also did not affect constitutive or HCV-induced IL-10 production by the CD14-CD36+ PBMC.
The direct correlation between intracellular IL-10 production and intensity of cell surface CD36 expression supports a possible connection between these molecules. Ligands of CD36, a scavenger receptor, include oxidized low density lipoproteins (oxLDL), thrombospondin, collagen types I and IV, apoptotic cells, long chain fatty acids and plasmodium-infected erythrocytes [46–54]. At least 2 of these ligands, thrombospondin and apoptotic cells, have immunosuppressive or anti-inflammatory effects [55, 56]. In addition, peroxisomal proliferator associated receptor-γ (PPAR-γ) mediated differentiation of monocytes towards a less inflammatory status is accompanied by up-regulation of CD36, which has been used to clinical advantage in cerebral malaria . While there is circumstantial evidence for a link between CD36 and induction of IL-10, there has been no reported evidence of interactions between HCV or any of its individual components and CD36. HCV associates with low density lipoproteins in circulation, and uses the LDL receptor and CD81 to enter cells . CD36 is not an HCV cellular receptor however, recent studies suggest that HCV and the CD36 ligand oxLDL do interact . HCV proteins may indirectly activate IL-10 production through a CD36 related mechanism.
IL-10 is an important immunomodulatory cytokine, and recent experiments in the LCMV model of chronic viral infection showed that manipulating levels of IL-10 alone was sufficient to dictate LCMV clearance or persistence . However, studies comparing plasma IL-10 levels in HCV-infected and healthy individuals have been less definitive, with some studies describing control levels of serum IL-10 in chronic HCV infection, while other groups describe a marked increase in IL-10 in chronic infection [61–63]. This raises questions about the value of serum IL-10 measurements, as IL-10 has important localized paracrine effects that are not reflected by these generalized IL-10 determinations. As well, none of these studies determine IL-10 levels during acute infection when it may be pivotal in the establishment of chronic infection. We demonstrated that exposure to HCV antigens in chronically infected individuals produced less IL-10 than seen in HCV naïve individuals. Longer term exposure to HCV antigens apparently conditions the CD36+ population of IL-10+ cells, attenuating HCV protein specific IL-10 induction. Thus, IL-10 levels during acute infection may be more relevant to the establishment of persistent infection than to maintaining chronic infection.
The transience of IL-10 responses to HCV proteins may allow emergence of cellular immune responses that are pathological in the context of chronic HCV infection. Several studies indicate a protective role for IL-10 in the immunopathology of HCV infection including a clinical trial of exogenous IL-10 administration, accumulation of IL-10 producing CD5+ B cells in HCV-infected individuals with mild disease and an association between high IL-10 gene polymorphism and decreased rates of HCV-related liver fibrosis [64–67]. The diminution in IL-10 induction seen in chronic HCV after HCV protein exposure appears to parallel, and possibly support, insidiously progressive liver disease.
Coinfection with HIV is an important issue in HCV infection because of its relatively high prevalence, higher HCV virus loads in HIV/HCV coinfected individuals and because of increased HCV-related morbidity in HIV coinfected individuals [68–70]. The higher level of IL-10 induction in PBMC from HIV-infected individuals may allow the establishment of more aggressive HCV infection, with downregulation of anti-HCV responses permitting increased HCV replication and higher HCV virus loads. However, this could depend on the order of HCV and HIV infection, with stronger anti-HCV cellular immunity developing when HCV infection precedes HIV-associated CD8+ T cell activation during the asymptomatic phase of HIV infection. Enhanced CD8+ T cell immunity against HCV in HCV/HIV coinfected individuals is likely to accelerate liver disease, while those who acquire HCV following the generalized decline in cellular immunity associated with progressive HIV infection are likely to have especially weak HCV-specific cellular immunity. Thus, there are 2 distinct groups of HIV/HCV coinfected individuals with respect to HCV-specific cellular immunity and its potential role in viral suppression and in the pathogenesis of liver disease.
Chronic HCV infection mitigates IL-10 production induced by HCV proteins. This is true in the group of three HCV monoinfected individuals, as well as the 10 HIV/HCV coinfected people. Though the monoinfected group is small, the two groups with chronic HCV infection have a total of thirteen individuals, and both have similar results. The limited IL-10 response is even more marked considering that the HIV monoinfected group has the highest IL-10 induction of all groups. Therefore, the lack of IL-10 induction in the setting of chronic infection is likely not an artifact of the small sample size, but a real phenomenon that can even supplant the expected HIV-associated IL-10 induction.
Consistent with previous reports, we found that HCV specific T cell proliferation is rare in chronic HCV infection with HIV coinfection a[71, 72]. In fact, the only individual with anti-HCV proliferative responses is also one of the individuals with the least number of IL-10 producing cells after HCV protein exposure.