Because of significant decline in HIV-related morbidity since the introduction of HAART, liver disease caused by chronic HCV infection has emerged as a major problem in HIV-1/HCV co- infected individuals. HIV co-infection has a negative impact on HCV pathogenesis, and therefore it is recommended to control HIV-1 levels prior to beginning HCV treatment in co-infected individuals. It is also important to note that despite successful response to HAART, many individuals fail to achieve full functional restoration of their immune system. If such individuals get infected with HCV, their ability to mount a successful immune response against HCV could be compromised. HIV-1/HCV co-infection should thus be considered a pathological state with both unique and common features of HIV-1 and HCV mono-infection. Although our immune response can clear HCV [37
], virus exposure often proceeds to chronicity. PEG-IFN and RBV combination therapy has been the mainstay of HCV treatment. IFN is a potent antiviral cytokine but is very unpleasant for the patient because of significant side effects such as depression, anemia, thrombocytopenia, neutropenia, diarrhea, and flulike symptoms such as fever, chills, fatigue, headache, and muscle aches. Besides, in individuals infected with the difficult-to-treat HCV genotypes (genotypes 1 and 4), it is curative only 40%–50% of the time. Although first- generation direct-acting antiviral drugs are now being introduced into the clinic; high cost and emergence of drug resistant HCV variants remain important concerns. IFN/RBV will thus be an essential part of HCV treatment and therefore a thorough understanding of its mechanism of action is necessary in order to identify host factors that predictive IFN/RBV treatment response.
Genome-wide association studies have identified SNPs near IFNL3 locus that can predict both spontaneous HCV clearance [26
] and successful clinical outcome to HCV therapy [38
]. These SNPs associate with altered mRNA expression of IFNL3, an antiviral cytokine, suggesting that IFNL3 expression levels are associated with HCV clearance and response to therapy [39
]. Interestingly, SNPs in the IFNL3 locus have been shown to affect the expression of IFNL3 not only in liver which is the site of infection but in PBMCs and whole blood [39
] with the minor (unfavorable) allele resulting in less IFNL3 expression. Interestingly, unfavorable IFNL3 genotype has been shown to influence DC and NK cells [45
Recent studies have led to identification of two functional SNPs rs4803217 and ss469415590 that are associated with both natural and treatment-based clearance of HCV [27
]. In our study, we did not observe any correlation between rs12979860, rs4803217 and ss469415590 genotypes and DC frequencies/phenotype. Given the small size of our cohort, it is necessary to validate our findings in a larger cohort. It will be interesting to study how different IFNL3 polymorphisms affect IFNL3 expression and resulting induction of Interferon-stimulated genes (ISGs) in DCs. It is possible that IFNL3 polymorphisms are responsible for intrinsic defects in DCs and thus responsible for generation of attenuated antiviral immune response.
The importance of DCs in resolving viral infection has been shown for several viruses including HIV-1 and HCV. Individuals that can control HIV-1 and HCV infection are characterized by strong multi epitope-specific CD4+
T cell response probably reflecting the efficient antigen presentation and T cell activation potential of DCs [49
]. However, chances of viral clearance after HCV and HIV-1 infection are low, owing to the remarkable ability of these viruses to impair DC functions. At the primary level, viruses can modulate the frequency of various DC subsets by causing their aberrant trafficking, inducing apoptosis, or interfering with their development. Compared to uninfected individuals, HIV-infected [18
] and HCV-infected [12
] individuals have lower frequency of mDCs and pDCs in blood. During HIV-1 infection, pDCs migrate to the inflamed lymph nodes, where they become activated, apoptotic, and frequently infected with the virus [56
]. During HCV infection, blood DCs are enriched in liver [24
] suggesting that increased migration of DCs to liver, the primary site of HCV replication, could be the cause of observed drop in DC numbers in blood. Other possibilities are that HCV targets DC precursors [61
] or differentiated DCs. In this regard, it has been shown that HCV core, NS3, and NS5 proteins induce apoptosis in mature DCs [62
]. Our results show that the frequency of mDCs was diminished in NRs (Figure 2A) whereas the frequency of pDCs was diminished in both NRs and SVRs (although to a greater extent in NRs) (Figure 2B). It is worth noting that HCV-infected individuals with high intrahepatic expression of ISGs prior to treatment achieve the lowest therapy response rates [42
]. Since HCV has evolved a potent strategy to preclude IFN production by infected hepatocytes, the cellular source of IFN responsible for inducing these ISGs in NRs has been unknown for a long time. Recent studies have solved this conundrum by demonstrating the ability of HCV-infected cells to release HCV RNA containing exosomes, which upon recognition by hepatic pDCs trigger IFN production [65
]. Therefore, it may be possible that in NRs, the increased expression of ISGs is due to greater frequency/IFN production of pDCs within liver. In fact our results show that prior to treatment, NRs had higher frequency of CCR5+
pDCs (Figures 6B and 6C), which could be responsible for their increased migration to liver, the site of inflammation. Upon IFN/RBV treatment, the mDC frequency increased in NRs (Figure 2C) but remained unchanged in SVRs (Figure 2D). On the other hand, the frequency of pDCs reduced even further in NRs (Figure 2E) but remained unchanged in SVRs (Figure 2F). HIV-1 is known to infect mDCs [66
]; therefore we were interested in investigating whether mDCs in NRs serve as HIV-1 reservoir and thus become functionally compromised. We did not detect any differences in the frequency of CD4+
mDCs between NRs and SVRs (data not shown).
In addition to direct antiviral effects, IFN-α is recognized to play an important role in regulation of innate and adaptive immune responses. Immunomodulatory effects of IFN-α are partly mediated through their ability to regulate maturation, migratory potential and immunostimulatory capacity of DCs [67
]. Differential IFN signaling in DCs of NRs and SVRs could underlie intrinsic differences in modulation of DC functions during treatment. Many studies on HCV-infected individuals have shown that various phenotypic and functional characteristics of DCs associate with the treatment outcome. High CD83 expression and low IL-10 production of DCs [8
] as well as lower chemotaxis index of pDCs to CXCL12 and CXCL10 [35
] prior to therapy are all associated with SVR. Increased frequency of CD86+
mDCs and pDCs upon treatment is also associated with SVR [33
]. All these studies suggest that there is a strong link between DC functionality (before or during treatment) and treatment outcome in HCV mono-infected individuals. But not many studies have investigated the same in HIV-1/HCV co-infected individuals. We show that the frequency of mDCs increased during treatment in NRs. This is interesting given the fact that the frequency of CCR7+
mDCs was reduced in NRs (Figure 3A). Reduced frequency of CCR7+
mDCs could lead to their reduced migration to lymph node resulting in their accumulation in blood. Frequencies of CD54+
mDCs were reduced in both NRs and SVRs (Figure 4) indicating that HIV-1/HCV co-infection impairs the maturation potential of mDCs, which fails to recover even after treatment. Also, NRs exhibited reduced frequency of CCR7+
pDCs. Our cocktail also included HLA-ABC and HLA-DR but we did not observe any differences in HLA-ABC and HLA-DR expression (data not shown).
Loss of T cell polyfunctionality T is one of the major events during the establishment of persistent viral infection [10
]. Binding of PD-L1 (on DCs) to PD-1 (on T cells) delivers a coinhibitory signal to T cells leading to their reduced proliferation and effector functions. Our study shows that NRs to IFN/RBV treatment have higher pDC PD-L1/CD86 ratio (Figure 5), which could be the cause of failed treatment response. Many viruses are known to attenuate the antiviral immune response by upregulating PD-1/PD-L1 expression. PD-1/PD-L1 blockade has shown tremendous promise in restoring the antiviral immune response against many different viruses. In future, it will be important to learn about the exact mechanisms of PD-L1/PD-1 upregulation during chronic viral infections. We also show that prior to treatment, PBMCs from SVRs secrete higher amounts of IFN-γ compared to controls and NRs (Figure 7). IFN-γ, a Th1 cytokine plays an important immunoregulatory role by enhancing the cytotoxic activity of T cells, macrophages and NK cells. Interestingly, IFN-γ levels secreted by PBMCs drop in SVRs at the end of treatment suggesting that infection of PBMCs in SVRs (and not NRs) could be the cause of high pre-treatment levels of IFN-γ and thus could be an important predictor of treatment success. Overall, this study emphasizes the role of DCs in mediating IFN/RBV treatment induced clearance of HCV. It also supports the possibility of using DC-based immunotherapeutic strategies (such as PD-L1 blockade) against HCV.