Date Published: July 6, 2016
Publisher: Public Library of Science
Author(s): Stephanie Pfaender, Elena Grabski, Claudia N. Detje, Nina Riebesehl, Stefan Lienenklaus, Eike Steinmann, Ulrich Kalinke, Thomas Pietschmann, Richard J. Kuhn.
Hepatitis C virus (HCV) induces interferon (IFN) stimulated genes in the liver despite of distinct innate immune evasion mechanisms, suggesting that beyond HCV infected cells other cell types contribute to innate immune activation. Upon coculture with HCV replicating cells, human CD141+ myeloid dendritic cells (DC) produce type III IFN, whereas plasmacytoid dendritic cells (pDC) mount type I IFN responses. Due to limitations in the genetic manipulation of primary human DCs, we explored HCV mediated stimulation of murine DC subsets. Coculture of HCV RNA transfected human or murine hepatoma cells with murine bone marrow-derived DC cultures revealed that only Flt3-L DC cultures, but not GM-CSF DC cultures responded with IFN production. Cells transfected with full length or subgenomic viral RNA stimulated IFN release indicating that infectious virus particle formation is not essential in this process. Use of differentiated DC from mice with genetic lesions in innate immune signalling showed that IFN secretion by HCV-stimulated murine DC was independent of MyD88 and CARDIF, but dependent on TRIF and IFNAR signalling. Separating Flt3-L DC cultures into pDC and conventional CD11b-like and CD8α-like DC revealed that the CD8α-like DC, homologous to the human CD141+ DC, release interferon upon stimulation by HCV replicating cells. In contrast, the other cell types and in particular the pDC did not. Injection of human HCV subgenomic replicon cells into IFN-β reporter mice confirmed the interferon induction upon HCV replication in vivo. These results indicate that HCV-replicating cells stimulate IFN secretion from murine CD8α-like DC independent of infectious virus production. Thus, this work defines basic principles of viral recognition by murine DC populations. Moreover, this model should be useful to explore the interaction between dendritic cells during HCV replication and to define how viral signatures are delivered to and recognized by immune cells to trigger IFN release.
Hepatitis C virus (HCV) infection constitutes a major global health problem since more than 140 million people suffer from chronic sequelae of the infection . Once infected, approximately 80% of the individuals are not able to clear the pathogen and develop a chronic infection that often is associated with liver diseases such as fibrosis, cirrhosis and hepatocellular carcinoma, thus resulting in the need of liver transplantation . It is believed that chronic infections are a consequence of a multi-factorial immune failure, due to delayed and weak T-cell responses, as well as dysfunctional B-cell, natural killer (NK) -cell and dendritic cell (DC) responses [3–9]. In addition, it has been shown that a strong pre-stimulation of interferon (IFN) stimulated genes (ISGs) during chronic HCV infection constitutes a marker of decreased responsiveness to IFN-based therapies . Bridging the innate with the adaptive immune responses, DC have an important role in the establishment of a protective immune response and they are crucial for the production of interferons and the activation of immune cells . Based on their distinct phenotype and functional characteristics, human peripheral DC can be classified in 3 major subsets. These include conventional DC (cDC) which encompass the myeloid CD1c+/BDCA1+ DC (mDC1) that are the largest mDC population in the blood and that are known for their antigen presenting capacity and cytokine expression, and the myeloid CD141+/BDCA3+ DC (mDC2) which produce IL-12 and type III IFN and have the ability to cross-present antigens to CD8 T-cells. The third subset is represented by the plasmacytoid dendritic cells (pDC), also known as natural type I IFN-producing cells, which upon activation produce high levels of type I IFN [12–15]. All three subsets have been implicated to be involved in responses to HCV infection [16–20]. However, not much is known about the interplay between the DC subsets among each other and with other immune cells. This is at least in part due to technical as well as practical difficulties of studying human DC. Studies of murine DC principally offer an attractive alternative for dissecting the requirements of DC stimulation. However, due to the restricted species tropism of HCV to humans and the lack of suitable immune competent mouse models  not much research has been performed using murine dendritic cells. A clear advantage of the murine system is the availability of genetically modified animals that allow in depth mechanistic studies. Furthermore, homologs of human dendritic cell subsets can be easily generated upon addition of FMS-like tyrosine 3 ligand (Flt3-L) to murine bone marrow (BM) cells, thus inducing high numbers of BM-derived DC [22–25]. Likewise, murine DC can be generated from BM using granulocyte/macrophage colony-stimulating factor (GM-CSF), whereas these cell cultures lack the pDC counterparts and are believed to yield DC which primarily resemble monocyte-derived DC [24, 26, 27]. To some extent Flt3-L DC cultures reflect the physiological DC development that gives rise to three major DC subsets for which orthologues can similarly also be found in the human peripheral blood: pDC as well as the two subsets of conventional dendritic cells, CD8α-like DC which resemble the human CD141+ (mDC2) subset and CD11b-like DC which are homologous to the human CD1c+ DC (mDC1) [23, 28]. Given the homology between murine and human DC subsets we used murine Flt3-L differentiated DC from wildtype as well as genetically modified mice to dissect the requirements for DC stimulation by HCV.
In this study, we describe basic principles of murine DC activation by HCV replicating cells. In the majority of cases, HCV exposure results in chronic infection with a high risk to develop severe liver diseases like cirrhosis, fibrosis and hepatocellular carcinoma . Despite stimulation of early immune responses within the liver as well as high IFN-stimulated gene expression the virus is able to persist in its host [10, 35]. Dendritic cells act as sentinels of the immune system and constitute a first line of defence against invading pathogens. Although human hepatocytes have been reported to produce type I and type III IFNs in response to HCV infection [36, 37], the virus has evolved evasion mechanisms including the cleavage of the adaptor molecules MAVS and TRIF, thus preventing signal transduction and IFN induction in these cells [38–40]. Therefore, dendritic cells that are not infected by HCV constitute an important defence mechanism to establish an antiviral status. Intrahepatic studies of dendritic cell interactions are technically and practically challenging and the isolation of certain dendritic cell subsets from human peripheral blood mononuclear cells (PBMC) can be difficult. So far only few studies have been conducted using murine derived dendritic cells mainly due to the lack of a suitable immunocompetent murine animal model, which supports HCV replication in vivo [41, 42]. Given the fact, that murine Flt3-L derived dendritic cells comprise subsets that are equivalent to physiological DC subsets found in humans, pDC, CD1c+ DC (with the murine counterparts CD11b-like DC) and CD141+ DC (with the murine equivalents of CD8α-like DC) [23, 43], we used this system to study virus host interactions. Interestingly, only Flt3-L but not GM-CSF derived DC were able to produce type I and type III IFN after co-culture with HCV transfected cells. Similar to what has been described for human pDC and CD141+ DC [17, 19, 31], murine IFN-α production was dependent on active viral replication and endocytosis, as co-culture with a replication incompetent virus mutant (ΔGDD) or treatment with acidification inhibitors prevented type I IFN production of DC. Interestingly, cell-free virus as well as purified supernatant harvested from HCV SGR transfected cells was able to stimulate IFN production in vitro, indicating that viral signatures are sensed independent of direct cell-to-cell contact and independent of infectious virus production. A recent study suggested that exosomes are released from HCV infected cells that stimulate human pDC to produce IFN-α . Other studies confirmed that SGR RNA can be transferred from one cell to another via exosomes resulting in a productive replication in susceptible cells [45–47]. Indeed, after stimulation of DC with preparations of extracellular vesicles we could observe type I IFN induction. Therefore, our data are consistent with these previous reports and indicate that exosomes and not the virus particle itself deliver signals to murine immune cells and thus trigger IFN production. However, further studies are needed to reveal the detailed mechanism and the features of these exosome-like vesicles. Sensing of HCV-transfected cells was dependent on TRIF and IFNAR signalling and sorting of individual DC subtypes showed that the CD8α-like DC, which are the murine counterpart of human CD141+ DC, are the main IFN producers in the murine culture after stimulation with HCV-transfected cells. Interestingly, there are striking parallels between the human and the murine system with the human CD141+ DC being able to sense HCV replication in a TRIF dependent manner, to produce IFN-λ and to amplify IFN-α responses [17, 18]. Furthermore, ex vivo isolated murine CD8+ DC have been described to produce type III IFN in an IPS-1 dependent manner upon stimulation with HCV , further supporting our data. However, unlike human pDC [19, 20], isolated murine pDC were not able to sense HCV replication. The reason for this remains elusive and needs to be further studied. One possible explanation is that pDC, in contrast to the other DC subtypes, do not express TLR3 . However, similar to human pDC, recognition of HCV RNA could still be mediated via TLR7 in these cells . Since co-culture with both human and murine HCV replicating liver cells did not stimulate murine pDC it is unlikely that species-specific recognition mechanisms are responsible for the inability of murine pDC to recognize HCV replicating cells.