Search

Patients with Activated-PI3Kd Syndrome (APDS) present with sinopulmonary infections, lymphadenopathy and CMV and/or EBV viremia, yet why patients fail to clear certain viral infections remains poorly understood. Using APDS patient samples and a mouse model (Pik3cdE1020K/+ mice), we demonstrate that, upon activation, Pik3cdE1020K/+ CD8+ T cells exhibit exaggerated features of short-lived effectors both in vitro and post-viral infection, associated with increased Fas-mediated apoptosis due to sustained phosphorylation of FoxO1 and derepression of FasL. In addition, Pik3cdE1020K/+ CD8+ T cells exhibit enhanced mTORC1 and c-Myc signatures; metabolic perturbations; and reorientation of their chromatin landscape. Conversely, Pik3cdE1020K/+ CD8+ T cells failed to sustain expression of proteins critical for maintenance of long-lived memory cells, including TCF1. Strikingly, activated Pik3cdE1020K/+ CD8+ T cells exhibit altered transcriptional and epigenetic circuits characterized by a pronounced IL-2/STAT5 signature associated with heightened IL-2 responses that prevented differentiation to memory-like cells in the presence of IL-15. Our data position PI3Kd as a central driver integrating multiple signaling circuits that promote terminal CD8+ T cell effector differentiation at the expense of memory and long-lived T cell responses. This work was funded in part by the Intramural Research Program of NIAID, NIH.

Stem-like CD8 T cells maintain long-term antiviral immunity during chronic infection and share regulatory pathways with memory precursor cells generated after acute infection. However, it is unclear whether stem-like CD8 T cells require distinct transcriptional and epigenetic regulation for their longevity and adaptation to the immunosuppressive environment of chronic infection. By comparing single-cell transcriptomes and epigenetic profiles of CD8 T cells responding to acute and chronic viral infections, we found that stem-like CD8 T cells became distinct from memory precursors prior to the termination of clonal expansion. We discovered a co-expression gene module containing Toxthat exhibited higher transcriptional activity associated with more active histone marks in stem-like T cells than memory precursors. Moreover, TOX promoted persistence of antiviral CD8 T cells and was required for stem-like CD8 differentiation. Thus, stem-like CD8 T cells require distinct transcriptional and epigenetic programs for their differentiation and persistence during chronic viral infection.

Mild traumatic brain injury (mTBI) causes severe meningeal cell death and vascular injury, leading to a robust peripheral immune response featuring neutrophils and monocytes/macrophages. Complications such as repetitive head injuries or viral infections following mTBI have the potential to significantly alter the immune response and impact wound repair. Using a closed-skull compression model of mTBI, we found that meningeal vasculature can regenerate within 7 days of injury, coinciding with a dynamic, bifurcated myeloid cell response. Macrophages derived from inflammatory monocytes scavenged dead cells while non-classical monocyte-derived macrophages promoted angiogenesis. The CD206+ angiogenic macrophages were positioned around the lesion perimeter where they aided in fibrinogen clearance and matrix remodeling. A peripheral infection by lymphocytic choriomeningitis virus (LCMV) during peak vascular damage delayed wound repair and promoted a significant meningeal CD8+ T cell response. The infected environment also shifted the myeloid cells to a more inflammatory state, reducing the amount of CD206+ macrophages around the lesion. Interestingly, we found that a secondary injury also interrupted wound repair only when experienced during the acute “inflammatory” but not the “wound-healing” phase. In fact, the mTBI lesion and infiltrating macrophages were resistant to further damage when reinjured during the wound-healing phase. Our findings demonstrate that meningeal vasculature can undergo robust regeneration following mTBI orchestrated by distinct myeloid cell populations. However, this wound repair process can be significantly disrupted by both sterile and viral insults following the initial injury.

Persistence of even the stealthiest viruses can perturb immune function either to the benefit or detriment of the host. Lymphocytic choriomeningitis virus (LCMV) establishes lifelong, systemic persistence when introduced in utero or at birth. Despite a highly evolved host-pathogen relationship, LCMV cannot escape detection by the innate immune system, which results in chronic stimulation of the type 1 IFN pathway in adult carrier mice. In this study we demonstrate that IFN-β is chronically up-regulated in peripheral lymphoid and nonlymphoid tissues (but not the CNS) of mice persistently infected from birth with LCMV and that dendritic cells (DCs) represent at least one source of IFN-β. Interestingly, chronic stimulation of this innate pathway significantly elevated MHC class I expression in the CNS as well as the periphery. Elevated MHC I expression was dependent on IFN-αβ receptor but not MyD88-dependent signaling, as only genetic deletion of the former reduced MHC I to normal levels. An increase in circulating virus was also observed in the IFN-αβ receptor deficient carrier mice, signifying that type I IFN continually exerts anti-viral pressure during a LCMV carrier state. Finally, to determine whether heightened CNS MHC I could be therapeutically corrected, we purged LCMV carrier mice of their persistent infection using adoptive immunotherapy. This treatment significantly reduced CNS MHC I expression. Collectively, these data demonstrate that even a well adapted pathogen can chronically stimulate the innate immune system and consequently alter the expression of Ag presenting machinery in an immunologically specialized compartment like the CNS.

Restrictions in the diversity of an adaptive immune repertoire can facilitate viral persistence. Because a host afflicted with an immune deficiency is not likely to purge a persistent infection using endogenous mechanisms, it is important to explore adoptive therapies to supplement the host with a functional immune defense. In this study, we describe a virus carrier state that results from introducing lymphocytic choriomeningitis virus (LCMV) into adult mice possessing a restricted T cell repertoire. On infection of these mice, LCMV establishes systemic persistence, and within the CNS the virus infects astrocytes (and later oligodendrocytes) rather than its traditional parenchymal target neurons. To determine whether LCMV could be purged from a novel target selection in the absence of an endogenous immune repertoire, we adoptively transferred virus-specific memory cells into adult carrier mice. The memory cells purged virus from the periphery as well as the CNS, but they induced fatalities not typically associated with adoptive immunotherapy. When the repertoire of the recipient mice was examined, a deficiency in natural regulatory T cells was noted. We therefore supplemented carrier mice with regulatory T cells and simultaneously performed adoptive immunotherapy. Cotransfer of regulatory T cells significantly reduced mortality while still permitting the antiviral memory cells to purge the persistent infection. These data indicate that regulatory T cells can be used therapeutically to lessen the pathogenicity of virus-specific immune cells in an immunodeficient host. We also propose that the novel carrier state described herein will facilitate the study of immunotherapeutic regimens.

Persistent viral infections continue to present major public health problems. Failure to achieve virus control confronts the immune system with a chronic viral burden that may involve immune cells themselves and directly compromise the functionality of effector lymphocytes and APCs. In this study we use the lymphocytic choriomeningitis virus system for persistent viral infection of its natural murine host and use analytical techniques for direct ex vivo visualization of virus-infected immune cells. We report that virtually all cells of the immune system can be infected, but the distribution of the viral burden is differentially allocated to lymphocyte and APC subsets of defined phenotypes. Importantly, the profile of immune cell infection found in the blood is broadly representative for the pattern of cellular infection in most organs and is independent of the presence of Abs or complement. By direct comparison of virus-infected and uninfected cell subsets, we demonstrate that lymphocytic choriomeningitis virus-infected T cells show preferential activation, skewed cytokine profiles, and increased apoptosis. In contrast, increased activation of APCs is generalized and independent of the presence of viral Ag. Our data indicate that specific patterns of immune cell infection are associated with distinct forms of immunostimulatory and immunosuppressive alterations that may provide insights into autoimmune processes associated with infectious disease and offer clues for therapeutic interventions aimed at restoration of complete immunity.

Infection of the pancreas with lymphocytic choriomeningitis virus results in rapid and differential expression among CXCR3 chemokines. IFN-γ-inducible protein of 10 kDa (IP-10), in contrast with monokine induced by IFN-γ and IFN-inducible T cell-α chemoattractant, is strongly expressed within 24 h postinfection. Blocking of IP-10, but not monokine induced by IFN-γ, aborts severity of Ag-specific injury of pancreatic β cells and abrogates type 1 diabetes. Mechanistically, IP-10 blockade impedes the expansion of peripheral Ag-specific T cells and hinders their migration into the pancreas. IP-10 expression was restricted to viruses infecting the pancreas and that are capable of causing diabetes. Hence, virus-induced organ-specific autoimmune diseases may be dependent on virus tropism and its ability to alter the local milieu by selectively inducing chemokines that prepare the infected tissue for the subsequent destruction by the adaptive immune response.

Mild traumatic brain injury (mTBI) leads to a robust immune response featuring resident microglia/macrophages and recruitment of peripherally-derived monocytes/macrophages. We sought novel insights into the dynamics and function of innate myeloid cells following brain injury by studying a closed-skull focal contusion model of mTBI that permits real-time imaging by intravital two-photon microscopy (TPM). TPM studies of CX3CR1GFP/+ CCR2RFP/+ reporter mice revealed an early microglia response followed by CCR2+ monocytes and CX3CR1+ macrophages that accumulated 1 to 7 days post-TBI. This transition coincided with dynamic changes in gene expression that shifted from pro-inflammatory to wound-healing. Early cell death and vascular destruction led to an impressive re-vascularization of the lesioned meninges, with 80–100% wound closure observed within 7 days of injury. We used histo-cytometry to define the anatomical position of the different myeloid subpopulations within the injured meninges during the repair phase. This revealed the presence of non-classical and resident macrophages that highly expressed the remodeling enzyme matrix metalloproteinase 2 (Mmp2) in the lesion core. Depletion of neutrophils and classical monocytes increased pathology due to inefficient dead cell clearance, but re-vascularization of the wound was not affected. Interestingly, vascular regeneration was significantly impaired only when non-classical monocytes and resident meningeal macrophages were depleted, resulting in reduced numbers of CD206+macrophages in the lesion. These studies demonstrate that neutrophils, monocytes, and resident macrophages participate in coordinated yet divergent wound-healing responses to mTBI.

Studies have established a role for T cells in resolving persistent viral infections, yet emerging evidence indicates that both T and B cells are required to control some viruses. During persistent infection, a marked lag or failure to generate neutralizing antibodies is commonly observed and likely contributes to an inability to control certain pathogens. Using lymphocytic choriomeningitis virus (LCMV) as a model, we have examined how a persistent viral infection can suppress neutralizing humoral immunity. By tracking the fate of virus-specific B cells in vivo, we report that LCMV-specific B cells were rapidly deleted within a few days of persistent infection, and this deletion was completely reversed by blockade of type I interferon (IFN-I) signaling. Early interference with IFN-I signaling promoted survival and differentiation of LCMV-specific B cells, which accelerated the generation of neutralizing antibodies. This marked improvement in antiviral humoral immunity did not rely on the cessation of IFN-I signaling in B cells but on alterations in the virus-specific CD8+ T cell response. Using two-photon microscopy and in vivo calcium imaging, we observed that cytotoxic T lymphocytes (CTLs) productively engaged and killed LCMV-specific B cells in a perforin-dependent manner within the first few days of infection. Blockade of IFN-I signaling protected LCMV-specific B cells by promoting CTL dysfunction. Therapeutic manipulation of this pathway may facilitate efforts to promote humoral immunity during persistent viral infection in humans. Our findings illustrate how events that occur early after infection can disturb the resultant adaptive response and contribute to viral persistence.

Viral infections that enter the central nervous system (CNS) can drive immune-mediated neurological disorders such as meningitis, encephalitis, and meningoencephalitis. Developing therapeutic interventions to ameliorate these disorders is an important area of clinically relevant research. To gain insights into the immunopathogenesis of CNS viral infection, we study the fatal meningitis induced by lymphocytic choriomeningitis virus (LCMV). We have recently shown that this disease is induced by the pathogenic cooperation between virus-specific CD8+ T cells and innate myelomonocytic cells that damage CNS blood vessels. In addition, genomic studies revealed that the onset of disease is associated with a profound local type I interferon (IFN-I) signature that triggers massive immune extravasation into the meninges. In the present study, we explored approaches to therapeutically treat this disorder. Importantly, we observed that a single injection of a therapeutic antibody directed against the IFN-I receptor (IFNAR) prevented fatal disease as well as the development of symptoms in infected mice. Using a recombinant virus expressing iCre recombinase, we conducted in vivo fate mapping studies in IFNAR blocked fluorescent protein reporter mice. These studies revealed evidence of non-cytopathic viral clearance throughout meninges and glial limitans. Mechanistically, microarray studies uncovered that IFNAR blockade promoted noncytopathic viral clearance through widespread reduction of immune signaling within the CNS without altering the composition of the immune infiltrate. These studies demonstrate that it is possible to mitigate a fatal virus-induced CNS disease via modulation of innate immune signaling pathways.

Barrier immunity, particularly at mucosal surfaces, provides a first line of defense against invading pathogens. Unlike the more classical barrier surfaces in the gut, respiratory tract, genital tract, and skin, the olfactory epithelium within the nose is a specialized barrier structure containing layers of olfactory sensory neurons (OSNs) dedicated to our sense of smell. Because OSNs must contact the nasal airways, they are vulnerable to infection. Upper respiratory viral infection is known to damage the olfactory neuroepithelium and is associated with loss of olfaction. Neurotropic viruses, such as vesicular stomatitis virus (VSV), readily infect OSNs, yet the olfactory epithelial response to infection is unknown. Using a newly generated VSV expressing Cre recombinase, we tracked the fate of virus infected cells after clearance. Numerous cell types, including neural precursors and olfactory ensheathing cells, survived infection. In addition, several months after infection, we detected clonal patches of regenerated mature OSNs derived from previously infected neuroepithelial precursors. VSV infection drove a massive influx of CD45+ hematopoietic cells into the olfactory epithelium that established residence after viral clearance. This included antigen presenting cells and tissue resident CD4+ and CD8+ T cells that localized in and around the regenerating neuroepithelium. We postulate that these immune cells provide a barrier defense against viral rechallenge and might play a role in supporting neuronal regeneration. Collectively, these studies offer insights into how the nasal barrier remodels its neural-immune architecture to restore function and provide enhanced protection following a viral challenge.

T cells develop into a dysfunctional state, called T-cell exhaustion, during chronic viral infections and cancer. However, how antigen-specific T cells persist in these situations is poorly understood. Here, we show that CD8 T cells differentiate into a less exhausted TCF1highTim3low subset and a more exhausted TCF1lowTim3high subset in chronic viral infection as well as human and mouse solid tumors. The TCF1highTim3low CD8 T cells, generated early after chronic lymphocytic choriomeningitis virus (LCMV) infection, persist better and mount a far stronger secondary response than TCF1lowTim3high cells. Importantly, TCF1highTim3low cells act as progenitor cells that are capable of both self-renewal and terminally differentiating into TCF1lowTim3high cells. We have further shown that cell-intrinsic TCF1 signaling is necessary and sufficient to program these progenitors early after infection. Cell-intrinsic TCF1 deficiency led to a progressive loss of anti-viral CD4 and CD8 T cell responses during the chronic phase of infection. Several pro-exhaustion factors, including Tim3, Cish, and Blimp1, are suppressed by TCF1. Conversely, TCF1 up-regulates Bcl6, which strongly enforces the progenitor fate. Moreover, cell-intrinsic type I interferon signal suppresses the differentiation of progenitor-like CD8 T cells, and antibody-mediated blockade of IFNAR1 enhances the differentiation of these CD8 T cells in a TCF1-dependent manner. Thus, our study not only identifies a CD8 progenitor population, which is critical for long-term persistence of anti-viral CD8 responses during chronic viral infection, but also uncovers both positive and negative regulators of the early programming of this population.

Discovery of the meningeal lymphatic vessels in the brain meninges (coverings of the brain) led to the hypothesis that they carry macromolecules and immune cells from cerebrospinal fluid (CSF) to cervical lymph nodes (cLN). Our new findings suggest that the lymphatic system controls immune responses through guiding antigen-presenting cells, such as dendritic cells (DC) from the meninges to draining lymph nodes. By using surgical and pharmacological methods to interfere with meningeal lymphatic vessels, we found lymphatics to serve as a major conduit by which DC leave the meninges, traffic to cLN and promote the generation of antigen-specific T cells. Inflammation-induced migration of DC is primarily regulated through a CC-chemokine receptor 7 (CCR7) dependent pathway. Co-injection of equal parts of differentially labeled CCR7-deficient and wild-type (WT) DCs into the CSF of the brain did allow translocation of CCR7-competent DC into the T cell zone of the draining lymph nodes but not CCR7−/− DC. Subsequently, antigenic stimulation by CCR7−/− DC did not induce a T cells response in the cLN, whereas WT DC enforced the majority of antigen-specific T cells to proliferate and increase their effective trafficking through integrin expression. To gain insights into how T cell responses are elicited when the meninges are infected with a virus, we currently set out to understand the lymphatics function in a mouse model of virus-induced meningitis. Injection of Lymphocytic choriomeningitis virus into the CSF induces rapid, uniform and fatal inflammation that is dependent on the generation of viral specific CD8+ T cells and potentially the exit of migratory DC through meningeal lymphatic vessels.

Type I interferons participate in the host defense against most viral infections through induction of innate antiviral proteins like tetherin (BST-2). In vitro studies have shown that tetherin can physically sequester viruses by linking them to infected cells, which in turn reduces virion release. It is not currently understood how this antiviral protein functions in vivo to control an immune response mounted against a persistent viral infection. We sought novel mechanistic insights into the function of tetherin by studying its role in the establishment of a persistent lymphocytic choriomeningitis virus (LCMV) infection. Following viral infection, tetherin was significantly upregulated in B cells, plasmacytoid dendritic cells (DCs), conventional DCs, and brain-resident microglia. Genetic deletion of tetherin resulted in an early failure to contain a persistence-prone strain of LCMV in the splenic marginal zone. This resulted in a rapid re-distribution of LCMV into the red and white pulp at early time points post-infection. This also interfered with the early priming of antiviral T cells by causing them to localize preferentially to the splenic red pulp where they proliferated less and acquired an impaired phenotype characterized by upregulation of exhaustion markers (PD-1, LAG-3, TIGIT), loss of cytokine production, and decreased motility. Importantly, these early alterations impeded the ability of the host to control persistence-prone LCMV. Viral control was delayed in peripheral tissues, and lifelong persistence was established in the brain. These data demonstrate that tetherin plays a critical in vivo role in shaping the anatomical distribution and adaptive immune response against a persistent viral infection.

Macrophage activation syndrome (MAS) is a life-threatening sepsis-like hyperinflammatory state arising in certain infectious and rheumatic diseases. Previous genetic associations have implicated genes critical for perforin-mediated cytotoxicity. However, we recently identified that activating mutations in inflammasome component NLRC4 lead to MAS and early-onset enterocolitis. Patients with NLRC4-MAS also have chronically and extraordinarily elevated serum IL-18. Notably, disease-causing mutations in other inflammasomes do not cause MAS or chronic IL-18 elevation. In vitro, IL-1b production by THP1 cells expressing various NLRC4 patient mutations correlated with disease severity and identified pathogenic clusters in the molecular structure. To further study the function of NLRC4, we created a mouse bearing the T337S point mutation. Like NLRC4-MAS patients, NLRC4-T337S mice demonstrate elevated serum IL-18 in the absence of systemic inflammation. However, macrophages from NLRC4-T337S mice did not demonstrate enhanced IL-1β or IL-18 production. Correspondingly, bone marrow chimeras demonstrated that IL-18 hyperproduction was derived from non-hematopoietic cells. NLRC4-T337S mice did not demonstrate more severe disease in response to repeated systemic stimulation through TLR9 or TLR4, nor upon LCMV infection, but developed colitis with IL-10 receptor blockade. These results suggest intestinal epithelia, which express both NLRC4 and proIL-18, may be a source of pathogenic IL-18 in this model. Future work will clarify the source(s) of excessive IL-18 and identify how chronic IL-18 effects on gut homeostasis and T and NK cell responses may promote systemic hyperinflammation.

Th1 cells and T follicular helper cells (Tfh cells), are generated early after viral infection. However, the molecular pathways that drive the differentiation of these two distinct subsets remain unclear. Here, we demonstrate that a molecular circuit involving TCF1, Blimp1, and IL-2 regulates the differentiation of viral-specific Tfh and Th1 cells. Tcf7 gene (encodes TCF1) is actively transcribed in naïve and Tfh cells but strongly repressed in Th1 cells. TCF1 expression was negatively correlated with CD25 and Blimp1 expression, and was suppressed by IL2 signaling and Blimp1. Notably, loss of Tcf7 in T cells led to defective Tfh-cell and GC responses after viral infection. Tcf7-deficient Tfh cells had reduced Tfh transcription signatures and increased Th1 transcription signatures. Moreover, Tcf7-deficient Tfh cells had lower mitochondrial mass and membrane potential, and showed reduced expression of genes related to oxidative metabolism. Il2ra and Prdm1 (encodes Blimp1) were ectopically expressed in Tcf7-deficient Tfh cells and were bound by TCF1, suggesting the existence of a negative feedback loop linking TCF1 to IL-2 and Blimp1. Over-expression of Bcl6 rescued Tfh defects caused by Tcf7 deficiency, demonstrating that TCF1 is upstream of the Bcl6-Blimp1 axis. Finally, we found that Tcf7-deficiency did not cause evident defect in Tfh responses in Th2 immunization models. Therefore, TCF1 specifically regulates antiviral Tfh responses.

Cerebral malaria (CM) is a severe complication of Plasmodium falciparum infection that results in thousands of deaths each year, mostly in African children. Pathological studies of CM victims reveal vascular hemorrhaging and edema within the brain, yet the underlying mechanisms that lead to this fatal condition are unknown. Using the animal model of experimental cerebral malaria (ECM), we sought mechanistic insights into the factors that drive the pathogenesis of CM. We found that ECM is associated with a large influx of immune cells into the central nervous system (CNS) at the peak of infection. By depleting individual innate and adaptive immune cell populations, we discovered that CD8 T cells alone were capable of mediating the fatal pathology associated with ECM, including breakdown of the blood brain barrier, edema, and extensive neuronal death in the brainstem. Using two-photon microscopy, we observed parasite-specific CD8 T cells slowly crawling and arresting along the CNS vasculature in vivo during ECM. This behavior was facilitated by the IFN-γ-dependent upregulation of antigen presentation and adhesion molecules on CNS endothelial cells (ECs). Antibody therapy against the adhesion molecules VLA-4 and LFA-1 disrupted CD8 T cell interactions with the CNS vasculature and prevented disease. Importantly, infected mice rendered genetically unable to present antigen on ECs had reduced CD8 T cell-brain EC interactions and were resistant to ECM. These results demonstrate that parasite-specific CD8 T cells interact with brain vascular ECs in an antigen-dependent manner to induce fatal vascular and neuronal pathology during ECM.

By definition persistent viral infections are accompanied by immune evasion, and most notably the failure to sustain robust T cell responses, responses that are essential to control of these infections. Identifying and understanding the immunoregulatory mechanisms that direct productive versus abortive T cell responses to viral infections is critical, but currently unclear. Using the model of persistent lymphocytic choriomeningitis virus (LCMV) infection, we found that upregulation of interleukin (IL)-10 early during infection induces T cell exhaustion and directly leads to viral persistence. Blockade of IL-10 activity induced sustained T cell function, the rapid elimination of an otherwise persistent virus infection and development of functional memory T cells capable of preventing re-infection. Antibody blockade of the IL-10 receptor during the chronic phase of infection restored T cell activity and eliminated an established persistent viral infection. Further, by manipulating this immunoregulatory molecule we significantly enhance immune responses to vaccination, indicating a potential “adjuvant” role to anti-IL10 immunotherapy. Thus, we propose a novel therapy to treat persistent viral infection by targeting host immunoregulatory molecules and reactivating antiviral T cells.

Adoptive immunotherapy relies on transfer of exogenously derived pathogen-specific (memory) T cells and is a highly successful approach to purge an established persistent viral infection. For example, administration of adoptive immunotherapy to mice persistently infected with lymphocytic choriomeningitis virus (LCMV) (carrier mice) results in systemic eradication of the virus, including noncytopathic clearance of persistently infected CNS neurons. By visualizing GFP-tagged LCMV-specific memory CD8+ T cells on brain reconstructions, it was revealed that cytotoxic lymphocytes (CTL) distributed throughout the CNS within 8 days after immunotherapy. Flow cytometric and microscopic analyses revealed that conventional dendritic cells arrived in the CNS concurrent with the memory CTL and induced anti-viral cytokine production (e.g., TNF-alpha). When adoptive immunotherapy was performed in carrier mice depleted of DCs or with memory T cells deficient in TNF-alpha, the adoptively administered memory T cells failed to purge the persistent infection. We further demonstrated by employing CTLA-4-Fc blockade and CCR2 deficient recipients that memory CTL rely on costimulation and DC migration, respectively, for optimal secondary expansion and efficient viral clearance. These studies underscore the importance of DC mobility and memory CTL-DC interactions in the clearance of a persistent viral infection.