Modulation of cytomegalovirus immune evasion identifies direct antigen presentation as the predominant mode of CD8 T-cell priming during immune reconstitution after hematopoietic cell transplantation

Cytomegalovirus (CMV) infection is the most critical infectious complication in recipients of hematopoietic cell transplantation (HCT) in the period between a therapeutic hematoablative treatment and the hematopoietic reconstitution of the immune system. Clinical investigation as well as the mouse model of experimental HCT have consistently shown that timely reconstitution of antiviral CD8 T cells is critical for preventing CMV disease in HCT recipients. Reconstitution of cells of the T-cell lineage generates naïve CD8 T cells with random specificities among which CMV-specific cells need to be primed by presentation of viral antigen for antigen-specific clonal expansion and generation of protective antiviral effector CD8 T cells. For CD8 T-cell priming two pathways are discussed: "direct antigen presentation" by infected professional antigen-presenting cells (pAPCs) and "antigen cross-presentation" by uninfected pAPCs that take up antigenic material derived from infected tissue cells. Current view in CMV immunology favors the cross-priming hypothesis with the argument that viral immune evasion proteins, known to interfere with the MHC class-I pathway of direct antigen presentation by infected cells, would inhibit the CD8 T-cell response. While the mode of antigen presentation in the mouse model of CMV infection has been studied in the immunocompetent host under genetic or experimental conditions excluding either pathway of antigen presentation, we are not aware of any study addressing the medically relevant question of how newly generated naïve CD8 T cells become primed in the phase of lympho-hematopoietic reconstitution after HCT. Here we used the well-established mouse model of experimental HCT and infection with murine CMV (mCMV) and pursued the recently described approach of up- or down-modulating direct antigen presentation by using recombinant viruses lacking or overexpressing the central immune evasion protein m152 of mCMV, respectively. Our data reveal that the magnitude of the CD8 T-cell response directly reflects the level of direct antigen presentation.

Direct antigen presentation is the canonical pathway of cytomegalovirus CD8 T-cell priming regulated by balanced immune evasion ensuring a strong antiviral response

CD8 T cells are important antiviral effectors in the adaptive immune response to cytomegaloviruses (CMV). Naïve CD8 T cells can be primed by professional antigen-presenting cells (pAPCs) alternatively by "direct antigen presentation" or "antigen cross-presentation". In the case of direct antigen presentation, viral proteins are expressed in infected pAPCs and enter the classical MHC class-I (MHC-I) pathway of antigen processing and presentation of antigenic peptides. In the alternative pathway of antigen cross-presentation, viral antigenic material derived from infected cells of principally any cell type is taken up by uninfected pAPCs and eventually also fed into the MHC class-I pathway. A fundamental difference, which can be used to distinguish between these two mechanisms, is the fact that viral immune evasion proteins that interfere with the cell surface trafficking of peptide-loaded MHC-I (pMHC-I) complexes are absent in cross-presenting uninfected pAPCs. Murine cytomegalovirus (mCMV) models designed to disrupt either of the two presentation pathways revealed that both are possible in principle and can substitute each other. Overall, however, the majority of evidence has led to current opinion favoring cross-presentation as the canonical pathway. To study priming in the normal host genetically competent in both antigen presentation pathways, we took the novel approach of enhancing or inhibiting direct antigen presentation by using recombinant viruses lacking or overexpressing a key mCMV immune evasion protein. Against any prediction, the strongest CD8 T-cell response was elicited under the condition of intermediate direct antigen presentation, as it exists for wild-type virus, whereas the extremes of enhanced or inhibited direct antigen presentation resulted in an identical and weaker response. Our findings are explained by direct antigen presentation combined with a negative feedback regulation exerted by the newly primed antiviral effector CD8 T cells. This insight sheds a completely new light on the acquisition of viral immune evasion genes during virus-host co-evolution.

Immunotherapy of cytomegalovirus infection by low-dose adoptive transfer of antiviral CD8 T cells relies on substantial post-transfer expansion of central memory cells but not effector-memory cells

Cytomegaloviruses (CMVs) are host species-specific in their replication. It is a hallmark of all CMVs that productive primary infection is controlled by concerted innate and adaptive immune responses in the immunocompetent host. As a result, the infection usually passes without overt clinical symptoms and develops into latent infection, referred to as "latency". During latency, the virus is maintained in a non-replicative state from which it can reactivate to productive infection under conditions of waning immune surveillance. In contrast, infection of an immunocompromised host causes CMV disease with viral multiple-organ histopathology resulting in organ failure. Primary or reactivated CMV infection of hematopoietic cell transplantation (HCT) recipients in a "window of risk" between therapeutic hemato-ablative leukemia therapy and immune system reconstitution remains a clinical challenge. Studies in the mouse model of experimental HCT and infection with murine CMV (mCMV), followed by clinical trials in HCT patients with human CMV (hCMV) reactivation, have revealed a protective function of virus-specific CD8 T cells upon adoptive cell transfer (AT). Memory CD8 T cells derived from latently infected hosts are a favored source for immunotherapy by AT. Strikingly low numbers of these cells were found to prevent CMV disease, suggesting either an immediate effector function of few transferred cells or a clonal expansion generating high numbers of effector cells. In the murine model, the memory population consists of resting central memory T cells (TCM), as well as of conventional effector-memory T cells (cTEM) and inflationary effector-memory T cells (iTEM). iTEM increase in numbers over time in the latently infected host, a phenomenon known as 'memory inflation' (MI). They thus appeared to be a promising source for use in immunotherapy. However, we show here that iTEM contribute little to the control of infection after AT, which relies almost entirely on superior proliferative potential of TCM.

Memory CD8 T Cells Protect against Cytomegalovirus Disease by Formation of Nodular Inflammatory Foci Preventing Intra-Tissue Virus Spread

Cytomegaloviruses (CMVs) are controlled by innate and adaptive immune responses in an immunocompetent host while causing multiple organ diseases in an immunocompromised host. A risk group of high clinical relevance comprises transiently immunocompromised recipients of hematopoietic cell transplantation (HCT) in the “window of risk” between eradicative therapy of hematopoietic malignancies and complete reconstitution of the immune system. Cellular immunotherapy by adoptive transfer of CMV-specific CD8 T cells is an option to prevent CMV disease by controlling a primary or reactivated infection. While experimental models have revealed a viral epitope-specific antiviral function of cognate CD8 T cells, the site at which control is exerted remained unidentified. The observation that remarkably few transferred cells protect all organs may indicate an early blockade of virus dissemination from a primary site of productive infection to various target organs. Alternatively, it could indicate clonal expansion of a few transferred CD8 T cells for preventing intra-tissue virus spread after successful initial organ colonization. Our data in the mouse model of murine CMV infection provide evidence in support of the second hypothesis. We show that transferred cells vigorously proliferate to prevent virus spread, and thus viral histopathology, by confining and eventually resolving tissue infection within nodular inflammatory foci.

Mast Cells Meet Cytomegalovirus: A New Example of Protective Mast Cell Involvement in an Infectious Disease

Cytomegaloviruses (CMVs) belong to the β-subfamily of herpesviruses. Their host-to-host transmission involves the airways. As primary infection of an immunocompetent host causes only mild feverish symptoms, human CMV (hCMV) is usually not considered in routine differential diagnostics of common airway infections. Medical relevance results from unrestricted tissue infection in an immunocompromised host. One risk group of concern are patients who receive hematopoietic cell transplantation (HCT) for immune reconstitution following hematoablative therapy of hematopoietic malignancies. In HCT patients, interstitial pneumonia is a frequent cause of death from hCMV strains that have developed resistance against antiviral drugs. Prevention of CMV pneumonia requires efficient reconstitution of antiviral CD8 T cells that infiltrate lung tissue. A role for mast cells (MC) in the immune control of lung infection by a CMV was discovered only recently in a mouse model. MC were shown to be susceptible for productive infection and to secrete the chemokine CCL-5, which recruits antiviral CD8 T cells to the lungs and thereby improves the immune control of pulmonary infection. Here, we review recent data on the mechanism of MC-CMV interaction, a field of science that is new for CMV virologists as well as for immunologists who have specialized in MC.

Localization of Viral Epitope-Specific CD8 T Cells during Cytomegalovirus Latency in the Lungs and Recruitment to Lung Parenchyma by Airway Challenge Infection

Interstitial pneumonia is a life-threatening clinical manifestation of cytomegalovirus infection in recipients of hematopoietic cell transplantation (HCT). The mouse model of experimental HCT and infection with murine cytomegalovirus revealed that reconstitution of virus-specific CD8⁺ T cells is critical for resolving productive lung infection. CD8⁺ T-cell infiltrates persisted in the lungs after the establishment of latent infection. A subset defined by the phenotype KLRG1⁺CD62L⁻ expanded over time, a phenomenon known as memory inflation (MI). Here we studied the localization of these inflationary T effector-memory cells (iTEM) by comparing their frequencies in the intravascular and transmigration compartments, the IVC and TMC, respectively, with their frequency in the extravascular compartment (EVC), the alveolar epithelium. Frequencies of viral epitope-specific iTEM were comparable in the IVC and TMC but were reduced in the EVC, corresponding to an increase in KLRG1⁻CD62L⁻ conventional T effector-memory cells (cTEM) and a decrease in functional IFNγ⁺CD8⁺ T cells. As maintained expression of KLRG1 requires stimulation by antigen, we conclude that iTEM lose KLRG1 and convert to cTEM after transmigration into the EVC because pneumocytes are not latently infected and, therefore, do not express antigens. Accordingly, antigen re-expression upon airway challenge infection recruited virus-specific CD8⁺ T cells to TMC and EVC.

Therapeutic Vaccination of Hematopoietic Cell Transplantation Recipients Improves Protective CD8 T-Cell Immunotherapy of Cytomegalovirus Infection

Reactivation of latent cytomegalovirus (CMV) endangers the therapeutic success of hematopoietic cell transplantation (HCT) in tumor patients due to cytopathogenic virus spread that leads to organ manifestations of CMV disease, to interstitial pneumonia in particular. In cases of virus variants that are refractory to standard antiviral pharmacotherapy, immunotherapy by adoptive cell transfer (ACT) of virus-specific CD8+ T cells is the last resort to bridge the "protection gap" between hematoablative conditioning for HCT and endogenous reconstitution of antiviral immunity. We have used the well-established mouse model of CD8+ T-cell immunotherapy by ACT in a setting of experimental HCT and murine CMV (mCMV) infection to pursue the concept of improving the efficacy of ACT by therapeutic vaccination (TherVac) post-HCT. TherVac aims at restimulation and expansion of limited numbers of transferred antiviral CD8+ T cells within the recipient. Syngeneic HCT was performed with C57BL/6 mice as donors and recipients. Recipients were infected with recombinant mCMV (mCMV-SIINFEKL) that expresses antigenic peptide SIINFEKL presented to CD8+ T cells by the MHC class-I molecule Kb. ACT was performed with transgenic OT-I CD8+ T cells expressing a T-cell receptor specific for SIINFEKL-Kb. Recombinant human CMV dense bodies (DB-SIINFEKL), engineered to contain SIINFEKL within tegument protein pUL83/pp65, served for vaccination. DBs were chosen as they represent non-infectious, enveloped, and thus fusion-competent subviral particles capable of activating dendritic cells and delivering antigens directly into the cytosol for processing and presentation in the MHC class-I pathway. One set of our experiments documents the power of vaccination with DBs in protecting the immunocompetent host against a challenge infection. A further set of experiments revealed a significant improvement of antiviral control in HCT recipients by combining ACT with TherVac. In both settings, the benefit from vaccination with DBs proved to be strictly epitope-specific. The capacity to protect was lost when DBs included the peptide sequence SIINFEKA lacking immunogenicity and antigenicity due to C-terminal residue point mutation L8A, which prevents efficient proteasomal peptide processing and binding to Kb. Our preclinical research data thus provide an argument for using pre-emptive TherVac to enhance antiviral protection by ACT in HCT recipients with diagnosed CMV reactivation.

Stochastic Episodes of Latent Cytomegalovirus Transcription Drive CD8 T-Cell "Memory Inflation" and Avoid Immune Evasion

Acute infection with murine cytomegalovirus (mCMV) is controlled by CD8+ T cells and develops into a state of latent infection, referred to as latency, which is defined by lifelong maintenance of viral genomes but absence of infectious virus in latently infected cell types. Latency is associated with an increase in numbers of viral epitope-specific CD8+ T cells over time, a phenomenon known as "memory inflation" (MI). The "inflationary" subset of CD8+ T cells has been phenotyped as KLRG1+CD62L- effector-memory T cells (iTEM). It is agreed upon that proliferation of iTEM requires repeated episodes of antigen presentation, which implies that antigen-encoding viral genes must be transcribed during latency. Evidence for this has been provided previously for the genes encoding the MI-driving antigenic peptides IE1-YPHFMPTNL and m164-AGPPRYSRI of mCMV in the H-2d haplotype. There exist two competing hypotheses for explaining MI-driving viral transcription. The "reactivation hypothesis" proposes frequent events of productive virus reactivation from latency. Reactivation involves a coordinated gene expression cascade from immediate-early (IE) to early (E) and late phase (L) transcripts, eventually leading to assembly and release of infectious virus. In contrast, the "stochastic transcription hypothesis" proposes that viral genes become transiently de-silenced in latent viral genomes in a stochastic fashion, not following the canonical IE-E-L temporal cascade of reactivation. The reactivation hypothesis, however, is incompatible with the finding that productive virus reactivation is exceedingly rare in immunocompetent mice and observed only under conditions of compromised immunity. In addition, the reactivation hypothesis fails to explain why immune evasion genes, which are regularly expressed during reactivation in the same cells in which epitope-encoding genes are expressed, do not prevent antigen presentation and thus MI. Here we show that IE, E, and L genes are transcribed during latency, though stochastically, not following the IE-E-L temporal cascade. Importantly, transcripts that encode MI-driving antigenic peptides rarely coincide with those that encode immune evasion proteins. As immune evasion can operate only in cis, that is, in a cell that simultaneously expresses antigenic peptides, the stochastic transcription hypothesis explains why immune evasion is not operative in latently infected cells and, therefore, does not interfere with MI.

The Anti-apoptotic Murine Cytomegalovirus Protein vMIA-m38.5 Induces Mast Cell Degranulation

Mast cells (MC) represent “inbetweeners” of the immune system in that they are part of innate immunity by acting as first-line sentinels for environmental antigens but also provide a link to adaptive immunity by secretion of chemokines that recruit CD8 T cells to organ sites of infection. An interrelationship between MC and cytomegalovirus (CMV) has been a blank area in science until recently when the murine model revealed a role for MC in the resolution of pulmonary infection by murine CMV (mCMV). As to the mechanism, MC were identified as a target cell type of mCMV. Infected MC degranulate and synthesize the CC-chemokine ligand-5 (CCL-5), which is released to attract protective virus-specific CD8 T cells to infected host tissue for confining and eventually resolving the productive, cytopathogenic infection. In a step forward in our understanding of how mCMV infection of MC triggers their degranulation, we document here a critical role for the mCMV m38.5 gene product, a mitochondria-localized inhibitor of apoptosis (vMIA). We show an involvement of mCMV vMIA-m38.5 in MC degranulation by two reciprocal approaches: first, by enhanced degranulation after m38.5 gene transfection of bone marrow-derived cell culture-grown MC (BMMC) and, second, by reduced degranulation of MC in peritoneal exudate cell populations infected ex corpore or in corpore with mutant virus mCMV-Δm38.5. These studies thus reveal a so far unknown function of mCMV vMIA-m38.5 and offer a previously unconsidered but biologically relevant cell system for further analyzing functional analogies between vMIAs of different CMV species.

Positive Role of the MHC Class-I Antigen Presentation Regulator m04/gp34 of Murine Cytomegalovirus in Antiviral Protection by CD8 T Cells

Murine cytomegalovirus (mCMV) codes for MHC class-I trafficking modulators m04/gp34, m06/gp48, and m152/gp40. By interacting with the MHC class-Iα chain, these proteins disconnect peptide-loaded MHC class-I (pMHC-I) complexes from the constitutive vesicular flow to the cell surface. Based on the assumption that all three inhibit antigen presentation, and thus the recognition of infected cells by CD8 T cells, they were referred to as “immunoevasins.” Improved antigen presentation mediated by m04 in the presence of m152 after infection with deletion mutant mCMV-Δm06^W, compared to mCMV-Δm04m06 expressing only m152, led us to propose renaming these molecules “viral regulators of antigen presentation” (vRAP) to account for both negative and positive functions. In accordance with a positive function, m04-pMHC-I complexes were found to be displayed on the cell surface, where they are primarily known as ligands for Ly49 family natural killer (NK) cell receptors. Besides the established role of m04 in NK cell silencing or activation, an anti-immunoevasive function by activation of CD8 T cells is conceivable, because the binding site of m04 to MHC class-Iα appears not to mask the peptide binding site for T-cell receptor recognition. However, functional evidence was based on mCMV-Δm06^W, a virus of recently doubted authenticity. Here we show that mCMV-Δm06^W actually represents a mixture of an authentic m06 deletion mutant and a mutant with an accidental additional deletion of a genome region encompassing also gene m152. Reanalysis of previously published experiments for the authentic mutant in the mixture confirms the previously concluded positive vRAP function of m04.

Revisiting CD8 T-cell 'Memory Inflation': New Insights with Implications for Cytomegaloviruses as Vaccine Vectors

Murine models of cytomegalovirus (CMV) infection have revealed an exceptional kinetics of the immune response. After resolution of productive infection, transient contraction of the viral epitope-specific CD8 T-cell pool was found to be followed by a pool expansion specific for certain viral epitopes during non-productive ‘latent’ infection. This phenomenon, known as ‘memory inflation’ (MI), was found to be based on inflationary KLRG1⁺CD62L⁻ effector-memory T cells (iTEM) that depend on repetitive restimulation. MI gained substantial interest for employing CMV as vaccine vector by replacing MI-driving CMV epitopes with foreign epitopes for generating high numbers of protective memory cells specific for unrelated pathogens. The concept of an MI-driving CMV vector is questioned by human studies disputing MI in humans. A bias towards MI in experimental models may have resulted from systemic infection. We have here studied local murine CMV infection as a route that is more closely matching routine human vaccine application. Notably, KLRG1⁻CD62L⁺ central memory T cells (TCM) and conventional KLRG1⁻CD62L⁻ effector memory T cells (cTEM) were found to expand, associated with ‘avidity maturation’, whereas the pool size of iTEM steadily declined over time. The establishment of high avidity CD8 T-cell central memory encourages one to pursue the concept of CMV vector-based vaccines.

Enhancement of Antigen Presentation by Deletion of Viral Immune Evasion Genes Prevents Lethal Cytomegalovirus Disease in Minor Histocompatibility Antigen-Mismatched Hematopoietic Cell Transplantation

Hematoablative treatment followed by hematopoietic cell transplantation (HCT) for reconstituting the co-ablated immune system is a therapeutic option to cure aggressive forms of hematopoietic malignancies. In cases of family donors or unrelated donors, immunogenetic mismatches in major histocompatibility complex (MHC) and/or minor histocompatibility (minor-H) loci are unavoidable and bear a risk of graft-vs.-host reaction and disease (GvHR/D). Transient immunodeficiency inherent to the HCT protocol favors a productive reactivation of latent cytomegalovirus (CMV) that can result in multiple-organ CMV disease. In addition, there exists evidence from a mouse model of MHC class-I-mismatched GvH-HCT to propose that mismatches interfere with an efficient reconstitution of antiviral immunity. Here we used a mouse model of MHC-matched HCT with C57BL/6 donors and MHC-congenic BALB.B recipients that only differ in polymorphic autosomal background genes, including minor-H loci coding for minor-H antigens (minor-HAg). Minor-HAg mismatch is found to promote lethal CMV disease in absence of a detectable GvH response to an immunodominant minor-HAg, the H60 locus-encoded antigenic peptide LYL8. Lethality of infection correlates with inefficient reconstitution of viral epitope-specific CD8⁺ T cells. Notably, lethality is prevented and control of cytopathogenic infection is restored when viral antigen presentation is enhanced by deletion of immune evasion genes from the infecting virus. We hypothesize that any kind of mismatch in GvH-HCT can induce "non-cognate transplantation tolerance" that dampens not only a mismatch-specific GvH response, which is beneficial, but adversely affects also responses to mismatch-unrelated antigens, such as CMV antigens in the specific case, with the consequence of lethal CMV disease.

Cytomegalovirus-Associated Inhibition of Hematopoiesis Is Preventable by Cytoimmunotherapy With Antiviral CD8 T Cells

Reactivation of latent cytomegalovirus (CMV) in recipients of hematopoietic cell transplantation (HCT) not only results in severe organ manifestations, but can also cause “graft failure” resulting in bone marrow (BM) aplasia. This inhibition of hematopoietic stem and progenitor cell engraftment is a manifestation of CMV infection that is long known in clinical hematology as “myelosuppression.” Previous studies in a murine model of sex-chromosome mismatched but otherwise syngeneic HCT and infection with murine CMV have shown that transplanted hematopoietic cells (HC) initially home to the BM stroma of recipients but then fail to further divide and differentiate. Data from this model were in line with the hypothesis that infection of stromal cells, which constitute “hematopoietic niches” where hematopoiesis takes place, causes a local deficiency in essential hematopoietins. Based on this understanding, one must postulate that preventing infection of stromal cells should restore the stroma's capacity to support hematopoiesis. Adoptively-transferred antiviral CD8⁺ T cells prevent lethal CMV disease by controlling viral spread and histopathology in vital organs, such as liver and lungs. It remained to be tested, however, if they can also prevent infection of the BM stroma and thus allow for successful HC engraftment. Here we demonstrate that antiviral CD8⁺ T cells control stromal infection. By tracking male donor-derived sry⁺ HC in the BM of infected female sry⁻ recipients, we show the CD8⁺ T cells allow for successful donor HC engraftment and thereby prevent CMV-associated BM aplasia. These data provide a further argument for cytoimmunotherapy of CMV infection after HCT.

Insufficient Antigen Presentation Due to Viral Immune Evasion Explains Lethal Cytomegalovirus Organ Disease After Allogeneic Hematopoietic Cell Transplantation

Reactivation of latent cytomegalovirus (CMV) poses a clinical problem in transiently immunocompromised recipients of hematopoietic cell (HC) transplantation (HCT) by viral histopathology that results in multiple organ manifestations. Compared to autologous HCT and to syngeneic HCT performed with identical twins as HC donor and recipient, lethal outcome of CMV infection is more frequent in allogeneic HCT with MHC/HLA or minor histocompatibility loci mismatch between donor and recipient. It is an open question if a graft-vs.-host (GvH) reaction exacerbates CMV disease, or if CMV exacerbates GvH disease (GvHD), or if interference is mutual. Here we have used a mouse model of experimental HCT and murine CMV (mCMV) infection with an MHC class-I mismatch by gene deletion, so that either HCT donor or recipient lack a single MHC class-I molecule, specifically H-2 Ld. This particular immunogenetic disparity has the additional advantage that it allows to experimentally separate GvH reaction of donor-derived T cells against recipient's tissues from host-vs.-graft (HvG) reaction of residual recipient-derived T cells against the transplanted HC and their progeny. While in HvG-HCT with Ld-plus donors and Ld-minus recipients almost all infected recipients were found to control the infection and survived, almost all infected recipients died of uncontrolled virus replication and consequent multiple-organ viral histopathology in case of GvH-HCT with Ld-minus donors and Ld-plus recipients. Unexpectedly, although anti-Ld-reactive CD8+ T cells were detected, mortality was not found to be associated with GvHD histopathology. By comparing HvG-HCT and GvH-HCT, investigation into the mechanism revealed an inefficient reconstitution of antiviral high-avidity CD8+ T cells, associated with lack of formation of protective nodular inflammatory foci (NIF) in host tissue, selectively in GvH-HCT. Most notably, mice infected with an immune evasion gene deletion mutant of mCMV survived under otherwise identical GvH-HCT conditions. Survival was associated with enhanced antigen presentation and formation of protective NIF by antiviral CD8+ T cells that control the infection and prevent viral histopathology. This is an impressive example of lethal viral disease in HCT recipients based on a failure of the immune control of CMV infection due to viral immune evasion in concert with an MHC class-I mismatch.

Role of antibodies in confining cytomegalovirus after reactivation from latency: three decades' résumé

Cytomegaloviruses (CMVs) are highly prevalent herpesviruses, characterized by strict species specificity and the ability to establish non-productive latent infection from which reactivation can occur. Reactivation of latent human CMV (HCMV) represents one of the most important clinical challenges in transplant recipients secondary to the strong immunosuppression. In addition, HCMV is the major viral cause of congenital infection with severe sequelae including brain damage. The accumulated evidence clearly shows that cellular immunity plays a major role in the control of primary CMV infection as well as establishment and maintenance of latency. However, the efficiency of antiviral antibodies in virus control, particularly in prevention of congenital infection and virus reactivation from latency in immunosuppressed hosts, is much less understood. Because of a strict species specificity of HCMV, the role of antibodies in controlling CMV disease has been addressed using murine CMV (MCMV) as a model. Here, we review and discuss the role played by the antiviral antibody response during CMV infections with emphasis on latency and reactivation not only in the MCMV model, but also in relevant clinical settings. We provide evidence to conclude that antiviral antibodies do not prevent the initiating molecular event of virus reactivation from latency but operate by preventing intra-organ spread and inter-organ dissemination of recurrent virus.

Adverse immunological imprinting by cytomegalovirus sensitizing for allergic airway disease

Cytomegalovirus (CMV) infection has a profound impact on the host’s immune system. Immunological imprinting by CMV is not restricted to immunity against CMV itself, but can affect immunity against other viral or non-viral infectious agents and also immunopathological responses. One category is heterologous immunity based on molecular mimicry, where antigen recognition receptors specific for a CMV antigen with broad avidity distribution also bind with some avidity to unrelated antigens and exert effector functions against target structures other than those linked to CMV. Another category is induction of cytokines by CMV infection that inhibit or drive immune responses to bystander antigens unrelated to CMV, and a third category is the activation of antigen-presenting cells by CMV from which unrelated antigens profit as “stowaways”. A striking example of the “stowaway” category, actually one that is of medical importance, has been published recently and will be discussed here for the more general reader. Specifically, in a murine model, CMV airway infection and inhaled environmental antigen of poor intrinsic allergenic potential were found to sensitize for allergic airway disease (AAD) only when combined. As to the mechanism, viral activation of CD11b⁺ conventional dendritic cells (CD11b⁺ cDC) that localize to airway mucosa facilitates uptake and processing of inhaled antigen. Thus, CMV serves as a “door opener” for otherwise harmless environmental antigens that have no intrinsic property to activate DC. Antigen-laden CD11b⁺ cDC migrate selectively to the airway draining lymph nodes, where they prime type-2 CD4⁺ T helper (Th-2) cells. Upon airway re-exposure to the inhaled antigen, Th-2 cells secrete interleukins (IL-4, IL-5, IL-9, and IL-25) known to induce goblet cell metaplasia, the lead histopathological manifestation of AAD that is characterized by thickening of airway epithelia and increased numbers of mucus-producing goblet cells, resulting in enhanced mucus secretion and airflow obstruction.

Pediatric roots of cytomegalovirus recurrence and memory inflation in the elderly

The establishment of a lifelong latent infection after resolution of primary infection is a hallmark of cytomegalovirus (CMV) biology. Primary infection with human CMV is possible any time in life, but most frequently, virus transmission occurs already perinatally or in early childhood. Many years or even decades later, severe clinical problems can result from recurrence of infectious virus by reactivation from latency in individuals who undergo immunocompromising medical treatment, for instance, transplant recipients, but also in septic patients without canonical immunosuppression, and in elderly people with a weakened immune system. The diversity of disease manifestations, such as retinitis, pneumonia, hepatitis, gastrointestinal disease, and others, has remained an enigma. In clinical routine, seropositivity for IgG antibodies against human CMV is taken to indicate latent infection and thus to define a qualitative risk of recurrence, but it is insufficient as a predictor for the quantitative risk of recurrence. Early experimental studies in the mouse model, comparing primary infection of neonatal and adult mice, led to the hypothesis that high load of latent viral genomes is a better predictor for the quantitative risk. A prolonged period of virus multiplication in the immunologically immature neonatally infected host increased the risk of virus recurrence by an enhanced copy number of latent virus genomes from which reactivation can initiate. In extension of this hypothesis, one would predict today that a higher incidence of reactivation events will also fuel the expansion of virus-specific T cells observed in the elderly, a phenomenon known as "memory inflation". Notably, the mouse model also indicated a stochastic nature of reactivation, thus offering an explanation for the diversity and organ selectivity of disease manifestations observed in patients. As the infection history is mostly undefined in humans, such predictions from the mouse model are difficult to verify by clinical investigation, and moreover, such questions were actually rarely addressed. Here, we have surveyed the existing literature for reports that may help to retrospectively relate the individual infection history to the risk of virus recurrence and recrudescent organ disease.

Cellular reservoirs of latent cytomegaloviruses

Cytomegaloviruses (CMVs), members of the β-subfamily of the herpesvirus family, have co-speciated with their respective mammalian hosts resulting in a mutual virus-host adaptation reflected by sets of 'private' viral genes that a particular CMV species does not share with other CMVs and that define the host-species specificity of CMVs. Nonetheless, based on "biological convergence" in evolution, fundamental rules in viral pathogenesis and immune control are functionally analogous between different virus-host pairs. Therefore, the mouse model of infection with murine CMV (mCMV) has revealed generally valid principles of CMV-host interactions. Specifically, the mouse model has paved the way to cellular immunotherapy of CMV disease in immunocompromised recipients of hematopoietic cell transplantation (HCT). Precisely in the context of HCT, however, current view assumes that there exists a major difference between hCMV and mCMV regarding "latent virus reservoirs" in that only hCMV establishes latency in hematopoietic lineage cells (HLCs), whereas mCMV establishes latency in endothelial cells. This would imply that only hCMV can reactivate from transplanted HLCs of a latently infected donor. In addition, as viral transcriptional activity during latency is discussed as a driver of clonal T-cell expansion over lifetime, a phenomenon known as "memory inflation", it is important to know if hCMV and mCMV establish latency in the same cell type(s) for imprinting the immune system. Here, we review the currently available evidence to propose that the alleged difference in latent virus reservoirs between hCMV and mCMV may rather relate to a difference in the focus of research. While studies on hCMV latency in HLCs likely described a non-canonical, transient type-2 latency, studies in the mouse model focussed on canonical, lifelong type-1 latency.

Coincident airway exposure to low-potency allergen and cytomegalovirus sensitizes for allergic airway disease by viral activation of migratory dendritic cells

Despite a broad cell-type tropism, cytomegalovirus (CMV) is an evidentially pulmonary pathogen. Predilection for the lungs is of medical relevance in immunocompromised recipients of hematopoietic cell transplantation, in whom interstitial CMV pneumonia is a frequent and, if left untreated, fatal clinical manifestation of human CMV infection. A conceivable contribution of CMV to airway diseases of other etiology is an issue that so far attracted little medical attention. As the route of primary CMV infection upon host-to-host transmission in early childhood involves airway mucosa, coincidence of CMV airway infection and exposure to airborne environmental antigens is almost unavoidable. For investigating possible consequences of such a coincidence, we established a mouse model of airway co-exposure to CMV and ovalbumin (OVA) representing a protein antigen of an inherently low allergenic potential. Accordingly, intratracheal OVA exposure alone failed to sensitize for allergic airway disease (AAD) upon OVA aerosol challenge. In contrast, airway infection at the time of OVA sensitization predisposed for AAD that was characterized by airway inflammation, IgE secretion, thickening of airway epithelia, and goblet cell hyperplasia. This AAD histopathology was associated with a T helper type 2 (Th2) transcription profile in the lungs, including IL-4, IL-5, IL-9, and IL-25, known inducers of Th2-driven AAD. These symptoms were all prevented by a pre-challenge depletion of CD4+ T cells, but not of CD8+ T cells. As to the underlying mechanism, murine CMV activated migratory CD11b+ as well as CD103+ conventional dendritic cells (cDCs), which have been associated with Th2 cytokine-driven AAD and with antigen cross-presentation, respectively. This resulted in an enhanced OVA uptake and recruitment of the OVA-laden cDCs selectively to the draining tracheal lymph nodes for antigen presentation. We thus propose that CMV, through activation of migratory cDCs in the airway mucosa, can enhance the allergenic potential of otherwise poorly allergenic environmental protein antigens.

Mouse Model of Cytomegalovirus Disease and Immunotherapy in the Immunocompromised Host: Predictions for Medical Translation that Survived the "Test of Time"

Human Cytomegalovirus (hCMV), which is the prototype member of the β-subfamily of the herpesvirus family, is a pathogen of high clinical relevance in recipients of hematopoietic cell transplantation (HCT). hCMV causes multiple-organ disease and interstitial pneumonia in particular upon infection during the immunocompromised period before hematopoietic reconstitution restores antiviral immunity. Clinical investigation of pathomechanisms and of strategies for an immune intervention aimed at restoring antiviral immunity earlier than by hematopoietic reconstitution are limited in patients to observational studies mainly because of ethical issues including the imperative medical indication for chemotherapy with antivirals. Aimed experimental studies into mechanisms, thus, require animal models that match the human disease as close as possible. Any model for hCMV disease is, however, constrained by the strict host-species specificity of CMVs that prevents the study of hCMV in any animal model including non-human primates. During eons of co-speciation, CMVs each have evolved a set of "private genes" in adaptation to their specific mammalian host including genes that have no homolog in the CMV virus species of any other host species. With a focus on the mouse model of CD8 T cell-based immunotherapy of CMV disease after experimental HCT and infection with murine CMV (mCMV), we review data in support of the phenomenon of "biological convergence" in virus-host adaptation. This includes shared fundamental principles of immune control and immune evasion, which allows us to at least make reasoned predictions from the animal model as an experimental "proof of concept." The aim of a model primarily is to define questions to be addressed by clinical investigation for verification, falsification, or modification and the results can then give feedback to refine the experimental model for research from "bedside to bench".

IL-33/ST2 pathway drives regulatory T cell dependent suppression of liver damage upon cytomegalovirus infection

Regulatory T (Treg) cells dampen an exaggerated immune response to viral infections in order to avoid immunopathology. Cytomegaloviruses (CMVs) are herpesviruses usually causing asymptomatic infection in immunocompetent hosts and induce strong cellular immunity which provides protection against CMV disease. It remains unclear how these persistent viruses manage to avoid induction of immunopathology not only during the acute infection but also during life-long persistence and virus reactivation. This may be due to numerous viral immunoevasion strategies used to specifically modulate immune responses but also induction of Treg cells by CMV infection. Here we demonstrate that liver Treg cells are strongly induced in mice infected with murine CMV (MCMV). The depletion of Treg cells results in severe hepatitis and liver damage without alterations in the virus load. Moreover, liver Treg cells show a high expression of ST2, a cellular receptor for tissue alarmin IL-33, which is strongly upregulated in the liver of infected mice. We demonstrated that IL-33 signaling is crucial for Treg cell accumulation after MCMV infection and ST2-deficient mice show a more pronounced liver pathology and higher mortality compared to infected control mice. These results illustrate the importance of IL-33 in the suppressive function of liver Treg cells during CMV infection.

Treg cells are crucial for immune homeostasis and for dampening immune response to several diseased conditions, including viral infections. Murine cytomegalovirus (MCMV) is a herpesvirus with pathogenic potential, so that early immune mechanisms are essential in controlling virus and protecting from virus-induced pathology. Studies on Foxp3⁺ Treg cells have revealed their inhibitory role on the early T cell response to MCMV infection and have suggested Treg cells as a target of MCMV’s immunoevasion mechanisms. Here we demonstrate that the number and activation status of liver Treg cells is strongly induced upon MCMV infection in order to protect the host from severe liver damage. They constitutively express high amounts of IL-33 receptor ST2 and their accumulation depends on IL-33, which is released as a tissue alarmin after the cell damage. For the first time, we show an immunoregulatory role of IL-33-dependent Treg cells in the liver of MCMV infected mice and their suppression of MCMV-induced immunopathology.

Peptide Processing Is Critical for T-Cell Memory Inflation and May Be Optimized to Improve Immune Protection by CMV-Based Vaccine Vectors

Cytomegalovirus (CMV) elicits long-term T-cell immunity of unparalleled strength, which has allowed the development of highly protective CMV-based vaccine vectors. Counterintuitively, experimental vaccines encoding a single MHC-I restricted epitope offered better immune protection than those expressing entire proteins, including the same epitope. To clarify this conundrum, we generated recombinant murine CMVs (MCMVs) encoding well-characterized MHC-I epitopes at different positions within viral genes and observed strong immune responses and protection against viruses and tumor growth when the epitopes were expressed at the protein C-terminus. We used the M45-encoded conventional epitope HGIRNASFI to dissect this phenomenon at the molecular level. A recombinant MCMV expressing HGIRNASFI on the C-terminus of M45, in contrast to wild-type MCMV, enabled peptide processing by the constitutive proteasome, direct antigen presentation, and an inflation of antigen-specific effector memory cells. Consequently, our results indicate that constitutive proteasome processing of antigenic epitopes in latently infected cells is required for robust inflationary responses. This insight allows utilizing the epitope positioning in the design of CMV-based vectors as a novel strategy for enhancing their efficacy.

Experimental cytomegalovirus (CMV) based vaccine vectors have provided highly encouraging results as innovative vaccine formulations against deadly virus infections, such as Ebola or AIDS. Nevertheless, it has remained incompletely understood why CMV is so efficient at stimulating T-lymphocytes, the immune cells that recognize pathogens within infected cells. We have generated an array of CMV mutants expressing the same antigen in different genes or in different parts of the same gene. This allowed us to identify that the immediate environment of the antigen, rather than properties of the antigen itself, crucially determine the immune protection conferred by CMV-based vaccines, implying that optimal immunity depends on the ability of host cells to degrade CMV proteins into peptides, short units that are recognized by T-cells. Detailed analysis revealed that strong and sustained T-cell immunity occurs only when their antigenic targets are processed by a primitive cellular machinery that is present in all cells of the body, rather than by its newly-evolved counterpart, which is present only in specialized antigen-presenting cells. Most importantly, our results provide a simple strategy to develop improved CMV vaccines by positioning the antigenic peptides at the right spot in CMV proteins.

Exogenous TNFR2 activation protects from acute GvHD via host T reg cell expansion

Activation of TNFR2 with a novel agonist expands T reg cells in vivo and protects allo-HCT recipients from acute GvHD while sparing antilymphoma and antiinfectious properties of transplanted donor T cells.

Donor CD4⁺Foxp3⁺ regulatory T cells (T reg cells) suppress graft-versus-host disease (GvHD) after allogeneic hematopoietic stem cell transplantation (HCT [allo-HCT]). Current clinical study protocols rely on the ex vivo expansion of donor T reg cells and their infusion in high numbers. In this study, we present a novel strategy for inhibiting GvHD that is based on the in vivo expansion of recipient T reg cells before allo-HCT, exploiting the crucial role of tumor necrosis factor receptor 2 (TNFR2) in T reg cell biology. Expanding radiation-resistant host T reg cells in recipient mice using a mouse TNFR2-selective agonist before allo-HCT significantly prolonged survival and reduced GvHD severity in a TNFR2- and T reg cell–dependent manner. The beneficial effects of transplanted T cells against leukemia cells and infectious pathogens remained unaffected. A corresponding human TNFR2-specific agonist expanded human T reg cells in vitro. These observations indicate the potential of our strategy to protect allo-HCT patients from acute GvHD by expanding T reg cells via selective TNFR2 activation in vivo.

Mutual Interference between Cytomegalovirus and Reconstitution of Protective Immunity after Hematopoietic Cell Transplantation

Hematopoietic cell transplantation (HCT) is a therapy option for aggressive forms of hematopoietic malignancies that are resistant to standard antitumoral therapies. Hematoablative treatment preceding HCT, however, opens a “window of opportunity” for latent Cytomegalovirus (CMV) by releasing it from immune control with the consequence of reactivation of productive viral gene expression and recurrence of infectious virus. A “window of opportunity” for the virus represents a “window of risk” for the patient. In the interim between HCT and reconstitution of antiviral immunity, primarily mediated by CD8⁺ T cells, initially low amounts of reactivated virus can expand exponentially, disseminate to essentially all organs, and cause multiple organ CMV disease, with interstitial pneumonia (CMV-IP) representing the most severe clinical manifestation. Here, I will review predictions originally made in the mouse model of experimental HCT and murine CMV infection, some of which have already paved the way to translational preclinical research and promising clinical trials of a preemptive cytoimmunotherapy of human CMV disease. Specifically, the mouse model has been pivotal in providing “proof of concept” for preventing CMV disease after HCT by adoptive transfer of preselected, virus epitope-specific effector and memory CD8⁺ T cells bridging the critical interim. However, CMV is not a “passive antigen” but is a pathogen that actively interferes with the reconstitution of protective immunity by infecting bone marrow (BM) stromal cells that otherwise form niches for hematopoiesis by providing the structural microenvironment and by producing hematopoietically active cytokines, the hemopoietins. Depending on the precise conditions of HCT, reduced homing of transplanted hematopoietic stem- and progenitor cells to infected BM stroma and impaired colony growth and lineage differentiation can lead to “graft failure.” In consequence, uncontrolled virus spread causes morbidity and mortality. In the race between viral BM pathology and reconstitution of antiviral immunity following HCT, exogenous reconstitution of virus-specific CD8⁺ T cells by adoptive cell transfer as an interventional strategy can turn the balance toward control of CMV.

Reconstitution of CD8 T Cells Protective against Cytomegalovirus in a Mouse Model of Hematopoietic Cell Transplantation: Dynamics and Inessentiality of Epitope Immunodominance

Successful reconstitution of cytomegalovirus (CMV)-specific CD8⁺ T cells by hematopoietic cell transplantation (HCT) gives a favorable prognosis for the control of CMV reactivation and prevention of CMV disease after hematoablative therapy of hematopoietic malignancies. In the transient immunocompromised state after HCT, pre-emptive cytoimmunotherapy with viral epitope-specific effector or memory CD8⁺ T cells is a promising option to speed up antiviral control. Despite high-coding capacity of CMVs and a broad CD8⁺ T-cell response on the population level, which reflects polymorphism in major histocompatibility complex class-I (MHC-I) glycoproteins, the response in terms of quantity of CD8⁺ T cells in any individual is directed against a limited set of CMV-encoded epitopes selected for presentation by the private repertoire of MHC-I molecules. Such epitopes are known as “immunodominant” epitopes (IDEs). Besides host immunogenetics, genetic variance in CMV strains harbored as latent viruses by an individual HCT recipient can also determine the set of IDEs, which complicates a “personalized immunotherapy.” It is, therefore, an important question if IDE-specific CD8⁺ T-cell reconstitution after HCT is critical or dispensable for antiviral control. As viruses with targeted mutations of IDEs cannot be experimentally tested in HCT patients, we employed the well-established mouse model of HCT. Notably, control of murine CMV (mCMV) after HCT was comparably efficient for IDE-deletion mutant mCMV-Δ4IDE and the corresponding IDE-expressing revertant virus mCMV-Δ4IDE-rev. Thus, antigenicity-loss mutations in IDEs do not result in loss-of-function of a polyclonal CD8⁺ T-cell population. Although IDE deletion was not associated with global changes in the response to non-IDE epitopes, the collective of non-IDE-specific CD8⁺ T-cells infiltrates infected tissue and confines infection within nodular inflammatory foci. We conclude from the model, and predict also for human CMV, that there is no need to exclusively aim for IDE-specific immunoreconstitution.

Non-cognate bystander cytolysis by clonal epitope-specific CTL lines through CD28-CD80 interaction inhibits antibody production: A potential caveat to CD8 T-cell immunotherapy

Adoptive transfer of virus epitope-specific CD8 T cells is an immunotherapy option to control cytomegalovirus (CMV) infection and prevent CMV organ disease in immunocompromised solid organ transplantation (SOT) and hematopoietic cell transplantation (HCT) recipients. The therapy aims at an early, selective recognition and cytolysis of infected cells for preventing viral spread in tissues with no adverse immunopathogenic side-effects by attack of uninfected bystander cells. Here we describe that virus epitope-specific, cloned T-cell lines lyse target cells that present the cognate antigenic peptide to the TCR, but simultaneously have the potential to lyse uninfected cells expressing the CD28 ligand CD80 (B7-1). While TCR-mediated cytolysis requires co-receptor CD8 and depends on perforin, the TCR-independent and viral epitope-independent cytolysis through CD28-CD80 signaling does not require CD8 on the effector cells and is perforin-independent. Importantly, this non-cognate cytolysis pathway leads to bystander cytolysis of CD80-expressing B-cell blasts and thereby inhibits pan-specific antibody production.

Evaluating Human T-Cell Therapy of Cytomegalovirus Organ Disease in HLA-Transgenic Mice

Reactivation of human cytomegalovirus (HCMV) can cause severe disease in recipients of hematopoietic stem cell transplantation. Although preclinical research in murine models as well as clinical trials have provided 'proof of concept' for infection control by pre-emptive CD8 T-cell immunotherapy, there exists no predictive model to experimentally evaluate parameters that determine antiviral efficacy of human T cells in terms of virus control in functional organs, prevention of organ disease, and host survival benefit. We here introduce a novel mouse model for testing HCMV epitope-specific human T cells. The HCMV UL83/pp65-derived NLV-peptide was presented by transgenic HLA-A2.1 in the context of a lethal infection of NOD/SCID/IL-2rg^-/- mice with a chimeric murine CMV, mCMV-NLV. Scenarios of HCMV-seropositive and -seronegative human T-cell donors were modeled by testing peptide-restimulated and T-cell receptor-transduced human T cells, respectively. Upon transfer, the T cells infiltrated host tissues in an epitope-specific manner, confining the infection to nodular inflammatory foci. This resulted in a significant reduction of viral load, diminished organ pathology, and prolonged survival. The model has thus proven its potential for a preclinical testing of the protective antiviral efficacy of HCMV epitope-specific human T cells in the evaluation of new approaches to an immunotherapy of CMV disease.

Pre-emptive CD8 T-cell therapy of human cytomegalovirus (HCMV) disease in immunocompromised recipients of hematopoietic stem cell transplantation gave promising results in clinical trials, but limited efficacy and the need of HCMV-seropositive memory cell donors has so far prevented adoptive cell transfer from becoming clinical routine. Further development is currently hampered by the lack of experimental animal models that allow preclinical testing of the protective efficacy of human T cells in functional organs. While humanized mouse models with human tissue implants are technically and statistically demanding, and are limited to studying human T-cell activation and local virus control in the implants, a more feasible model for control of systemic infection and prevention of multiple-organ CMV disease is regrettably missing. Here we introduce such a model based on infection of genetically immunocompromised, HLA-A2.1-transgenic NOD/SCID/IL-2rg^-/- mice with a chimeric murine CMV engineered to express the HCMV NLV-peptide epitope. Mimicking the scenario of HCMV-unexperienced donors, human T cells transduced with a human T-cell receptor specific for HLA-A.2.1-presented NLV peptide controlled systemic infection and moderated organ disease resulting in a survival benefit. The model promises to become instrumental in defining T-cell properties that determine their protective efficacy for a further development of adoptive immunotherapy of post-transplantation CMV infection.

An endocytic YXXΦ (YRRF) cargo sorting motif in the cytoplasmic tail of murine cytomegalovirus AP2 'adapter adapter' protein m04/gp34 antagonizes virus evasion of natural killer cells

Viruses have evolved proteins that bind immunologically relevant cargo molecules at the cell surface for their downmodulation by internalization. Via a tyrosine-based sorting motif YXXΦ in their cytoplasmic tails, they link the bound cargo to the cellular adapter protein-2 (AP2), thereby sorting it into clathrin-triskelion-coated pits for accelerated endocytosis. Downmodulation of CD4 molecules by lentiviral protein NEF represents the most prominent example. Based on connecting cargo to cellular adapter molecules, such specialized viral proteins have been referred to as 'connectors' or 'adapter adapters.' Murine cytomegalovirus glycoprotein m04/gp34 binds stably to MHC class-I (MHC-I) molecules and suspiciously carries a canonical YXXΦ endocytosis motif YRRF in its cytoplasmic tail. Disconnection from AP2 by motif mutation ARRF should retain m04-MHC-I complexes at the cell surface and result in an enhanced silencing of natural killer (NK) cells, which recognize them via inhibitory receptors. We have tested this prediction with a recombinant virus in which the AP2 motif is selectively destroyed by point mutation Y248A, and compared this with the deletion of the complete protein in a Δm04 mutant. Phenotypes were antithetical in that loss of AP2-binding enhanced NK cell silencing, whereas absence of m04-MHC-I released them from silencing. We thus conclude that AP2-binding antagonizes NK cell silencing by enhancing endocytosis of the inhibitory ligand m04-MHC-I. Based on a screen for tyrosine-based endocytic motifs in cytoplasmic tail sequences, we propose here the new hypothesis that most proteins of the m02-m16 gene family serve as 'adapter adapters,' each selecting its specific cell surface cargo for clathrin-assisted internalization.

Mechanism of tumor remission by cytomegalovirus in a murine lymphoma model: evidence for involvement of virally induced cellular interleukin-15

A murine model of B and T cell lymphomas in recipients after hematoablative conditioning for hematopoietic cell transplantation (HCT) has previously revealed a tumor-repressive, metastasis-inhibiting function of murine cytomegalovirus (mCMV). More recently, this prediction from the experimental model was put on trial in several clinical studies that indeed gave evidence for a lower incidence of tumor relapse associated with early reactivation of latent human cytomegalovirus (hCMV) after allogeneic HCT in patients treated against different types of hematopoietic malignancies, including lymphoma and acute as well as chronic leukemias. Due to the limitations inherent to clinical studies, the tumor-repressive role of hCMV remained observational with no approach to clarify mechanisms. Although the tumor-repressive mechanisms of mCMV and hCMV may differ and depend on the type of tumor, experimental approaches in the murine model might give valuable hints for concepts to follow in clinical research. We have previously shown for the liver-adapted A20-derived B cell lymphoma E12E that mCMV does not infect the lymphoma cells for causing cell death by viral cytopathogenicity but triggers tumor-selective apoptosis at a tissue site of tumor metastasis distant from a local site of infection. This finding suggested involvement of a cytokine that triggers apoptosis, directly or indirectly. Here we used a series of differential high-density microarray analyses to identify cellular genes whose expression is specifically upregulated at the site of virus entry only by viruses capable of triggering lymphoma cell apoptosis. This strategy identified interleukin-15 (IL-15) as most promising candidate, eventually confirmed by lymphoma repression with recombinant IL-15.

Identification of an atypical CD8 T cell epitope encoded by murine cytomegalovirus ORF-M54 gaining dominance after deletion of the immunodominant antiviral CD8 T cell specificities

Control of murine cytomegalovirus (mCMV) infection is mediated primarily by CD8 T cells, with four specificities dominating in BALB/c mice. Functional deletion of the respective immunodominant epitopes (IDEs) in mutant virus Δ4IDE revealed a still efficient control of infection. In a murine model of hematopoietic cell transplantation and infection with Δ4IDE, an mCMV-specific open reading frame (ORF) library screening assay indicated a strong CD8 T cell reactivity against the ORF-M54 product, the highly conserved and essential mCMV homolog of human CMV DNA polymerase UL54, which is a known inducer of in vivo protection against mCMV by DNA immunization. Applying bioinformatic algorithms for CD8 T cell epitope prediction, the top-scoring peptides were used to stimulate ex vivo-isolated CD8 T cells and to generate cytolytic T cell lines; yet, this approach failed to identify M54 epitope(s). As an alternative, a peptide library consisting of 549 10-mers with an offset of two amino acids (aa), covering the complete aa-sequence of the M54 protein, was synthesized and used for the stimulation. A region of 12 aa proved to encompass an epitope. An 'alanine walk' over this antigenic 12-mer and all possible 11-, 10- and 9-mers derived thereof revealed aa-residues critical for antigenicity, and terminal truncations identified the H-2D(d) presented 8-mer M5483-90 as the optimal epitope. An increased frequency of the corresponding CD8 T cells in the absence of the 4 IDEs indicated immunodomination by the IDE-specific CD8 T cells as a mechanism by which the generation of M54-specific CD8 T cells is inhibited after infection with wild-type mCMV.

Mast cells: innate attractors recruiting protective CD8 T cells to sites of cytomegalovirus infection

Reactivation of latent cytomegalovirus (CMV) in the transient immunocompromised state after hematoablative treatment is a major concern in patients undergoing hematopoietic cell transplantation (HCT) as a therapy of hematopoietic malignancies. Timely reconstitution of antiviral CD8 T cells and their efficient recruitment to the lungs is crucial for preventing interstitial pneumonia, the most severe disease manifestation of CMV in HCT recipients. Here, we review recent work in a murine model, implicating mast cells (MC) in the control of pulmonary infection. Murine CMV (mCMV) productively infects MC in vivo and triggers their degranulation, resulting in the release of the CC chemokine ligand 5 (CCL5) that attracts CD8 T cells to infiltrate infected tissues. Comparing infection of MC-sufficient C57BL/6 mice and congenic MC-deficient Kit (W-sh/W-sh) "sash" mutants revealed an inverse relation between the number of lung-infiltrating CD8 T cells and viral burden in the lungs. Specifically, reduced lung infiltration by CD8 T cells in "sash" mutants was associated with an impaired infection control. The causal, though indirect, involvement of MC in antiviral control was confirmed by reversion of the deficiency phenotype in "sash" mutants reconstituted with MC. These recent findings predict that efficient MC reconstitution facilitates the control of CMV infection also in immunocompromised HCT recipients.

Mast cells expedite control of pulmonary murine cytomegalovirus infection by enhancing the recruitment of protective CD8 T cells to the lungs

The lungs are a noted predilection site of acute, latent, and reactivated cytomegalovirus (CMV) infections. Interstitial pneumonia is the most dreaded manifestation of CMV disease in the immunocompromised host, whereas in the immunocompetent host lung-infiltrating CD8 T cells confine the infection in nodular inflammatory foci and prevent viral pathology. By using murine CMV infection as a model, we provide evidence for a critical role of mast cells (MC) in the recruitment of protective CD8 T cells to the lungs. Systemic infection triggered degranulation selectively in infected MC. The viral activation of MC was associated with a wave of CC chemokine ligand 5 (CCL5) in the serum of C57BL/6 mice that was MC-derived as verified by infection of MC-deficient Kit^W-sh/W-sh “sash” mutants. In these mutants, CD8 T cells were recruited less efficiently to the lungs, correlating with enhanced viral replication and delayed virus clearance. A causative role for MC was verified by MC reconstitution of “sash” mice restoring both, efficient CD8 T-cell recruitment and infection control. These results reveal a novel crosstalk axis between innate and adaptive immune defense against CMV, and identify MC as a hitherto unconsidered player in the immune surveillance at a relevant site of CMV disease.

Being strategically located beneath endothelial and epithelial surfaces, mast cells (MC) serve as sentinels for invading pathogens at host-environment boundaries as part of the innate defense against infection. Host genetic resistance against cytomegaloviruses (CMV) is largely determined by the innate immune response, but an implication of MC in the adaptive immune defense against CMV has not been considered so far and is almost impossible to address in human infection. Using murine CMV as a model that in the past has already pioneered the discovery of fundamental principles in CMV-host interactions, our data reveal MC as central part of a novel crosstalk-axis between the innate and adaptive immune response to CMV. We found that upon host infection MC become rapidly activated and promote the recruitment of protective CD8 T cells to the lungs, a noted critical site of CMV pathogenesis in humans as well as in the mouse model. Enhanced tissue infiltration of CD8 T cells results in a reduced peak viral load and a faster clearance of productive infection. Realizing the importance of MC in the control of pulmonary CMV infection may help to develop new strategies for preventing CMV pneumonia by MC supplementation in recipients of hematopoietic cell transplantation.

The viral chemokine MCK-2 of murine cytomegalovirus promotes infection as part of a gH/gL/MCK-2 complex

Human cytomegalovirus (HCMV) forms two gH/gL glycoprotein complexes, gH/gL/gO and gH/gL/pUL(128,130,131A), which determine the tropism, the entry pathways and the mode of spread of the virus. For murine cytomegalovirus (MCMV), which serves as a model for HCMV, a gH/gL/gO complex functionally homologous to the HCMV gH/gL/gO complex has been described. Knock-out of MCMV gO does impair, but not abolish, virus spread indicating that also MCMV might form an alternative gH/gL complex. Here, we show that the MCMV CC chemokine MCK-2 forms a complex with the glycoprotein gH, a complex which is incorporated into the virion. We could additionally show that mutants lacking both, gO and MCK-2 are not able to produce infectious virus. Trans-complementation of these double mutants with either gO or MCK-2 showed that both proteins can promote infection of host cells, although through different entry pathways. MCK-2 has been extensively studied in vivo by others. It has been shown to be involved in attracting cells for virus dissemination and in regulating antiviral host responses. We now show that MCK-2, by forming a complex with gH, strongly promotes infection of macrophages in vitro and in vivo. Thus, MCK-2 may play a dual role in MCMV infection, as a chemokine regulating the host response and attracting specific target cells and as part of a glycoprotein complex promoting entry into cells crucial for virus dissemination.

Several human herpesviruses form alternative gH/gL complexes which determine the tropism for different cell types. For murine cytomegalovirus (MCMV), a gH/gL/gO complex has recently been characterized. Here, we present the identification and characterization of an alternative gH/gL/MCK-2 complex which promotes MCMV spread and is important for efficient infection of macrophages in vitro and in vivo. Association of the MCMV CC chemokine MCK-2 with a glycoprotein complex promoting virus entry is a novel function for the well-characterized MCK-2. Virus mutants lacking MCK-2 have been shown to exhibit a reduced capacity to attract leukocytes and a disregulated T cell control of the MCMV infection in vivo. These defects can be attributed to the chemokine function of MCK-2. Yet, the observation that MCK-2 knock-out mutants additionally are impaired in infecting leukocytes in vivo is consistent with our new finding that MCK-2 forms a glycoprotein complex promoting entry into monocytic cells. gH/gL complexes associating with multifunctional proteins add a new level of complexity to the interpretation of infection phenotypes of the respective knock-out herpesviruses.

TCR-ligand koff rate correlates with the protective capacity of antigen-specific CD8+ T cells for adoptive transfer

Adoptive immunotherapy is a promising therapeutic approach for the treatment of chronic infections and cancer. T cells within a certain range of high avidity for their cognate ligand are believed to be most effective. T cell receptor (TCR) transfer experiments indicate that a major part of avidity is hardwired within the structure of the TCR. Unfortunately, rapid measurement of structural avidity of TCRs is difficult on living T cells. We developed a technology where dissociation (koff rate) of truly monomeric peptide-major histocompatibility complex (pMHC) molecules bound to surface-expressed TCRs can be monitored by real-time microscopy in a highly reliable manner. A first evaluation of this method on distinct human cytomegalovirus (CMV)-specific T cell populations revealed unexpected differences in the koff rates. CMV-specific T cells are currently being evaluated in clinical trials for efficacy in adoptive immunotherapy; therefore, determination of koff rates could guide selection of the most effective donor cells. Indeed, in two different murine infection models, we demonstrate that T cell populations with lower koff rates confer significantly better protection than populations with fast koff rates. These data indicate that koff rate measurements can improve the predictability of adoptive immunotherapy and provide diagnostic information on the in vivo quality of T cells.

Viral latency drives 'memory inflation': a unifying hypothesis linking two hallmarks of cytomegalovirus infection

Low public awareness of cytomegalovirus (CMV) results from the only mild and transient symptoms that it causes in the healthy immunocompetent host, so that primary infection usually goes unnoticed. The virus is not cleared, however, but stays for the lifetime of the host in a non-infectious, replicatively dormant state known as 'viral latency'. Medical interest in CMV results from the fact that latent virus can reactivate to cytopathogenic, tissue-destructive infection causing life-threatening end-organ disease in immunocompromised recipients of solid organ transplantation (SOT) or hematopoietic cell transplantation (HCT). It is becoming increasingly clear that CMV latency is not a static state in which the viral genome is silenced at all its genetic loci making the latent virus immunologically invisible, but rather is a dynamic state characterized by stochastic episodes of transient viral gene desilencing. This gene expression can lead to the presentation of antigenic peptides encoded by 'antigenicity-determining transcripts expressed in latency (ADTELs)' sensed by tissue-patrolling effector-memory CD8 T cells for immune surveillance of latency [In Reddehase et al., Murine model of cytomegalovirus latency and reactivation, Current Topics in Microbiology and Immunology, vol 325. Springer, Berlin, pp 315-331, 2008]. A hallmark of the CD8 T cell response to CMV is the observation that with increasing time during latency, CD8 T cells specific for certain viral epitopes increase in numbers, a phenomenon that has gained much attention in recent years and is known under the catchphrase 'memory inflation.' Here, we provide a unifying hypothesis linking stochastic viral gene desilencing during latency to 'memory inflation.'

Parameters determining the efficacy of adoptive CD8 T-cell therapy of cytomegalovirus infection

Reactivation of latent cytomegalovirus (CMV) in the transient state of immunodeficiency after hematopoietic cell transplantation (HCT) is the most frequent and severe viral complication endangering leukemia therapy success. By infecting the bone marrow (BM) stroma of the transplantation recipient, CMV can directly interfere with BM repopulation by the transplanted donor-derived hematopoietic cells and thus delay immune reconstitution of the recipient. Cytopathogenic virus spread in tissues can result in CMV disease with multiple organ manifestations of which interstitial pneumonia is the most feared. There exists a 'window of risk' between hematoablative treatment and reconstitution of antiviral immunity after HCT, whereby timely reconstitution of antiviral CD8 T cells is a recognized positive prognostic parameter for the control of reactivated CMV infection and prevention of CMV disease. Supplementation of endogenous reconstitution by adoptive cell transfer of 'ready-to-go' effector and/or memory virus epitope-specific CD8 T cells is a therapeutic option to bridge the 'window of risk.' Preclinical research in murine models of CMV disease has been pivotal by providing 'proof of concept' for a benefit from CD8 T-cell therapy of HCT-associated CMV disease (reviewed in Holtappels et al. Med Microbiol Immunol 197:125-134, 2008). Here, we give an update of our previous review with focus on parameters that determine the efficacy of adoptive immunotherapy of CMV infection by antiviral CD8 T cells in the murine model.

Antigen presentation under the influence of 'immune evasion' proteins and its modulation by interferon-gamma: implications for immunotherapy of cytomegalovirus infection with antiviral CD8 T cells

Cytomegalovirus (CMV) disease with multiple organ manifestations is the most feared viral complication limiting the success of hematopoietic cell transplantation as a therapy of hematopoietic malignancies. A timely endogenous reconstitution of CD8 T cells controls CMV infection, and adoptive transfer of antiviral CD8 T cells is a therapeutic option to prevent CMV disease by bridging the gap between an early CMV reactivation and delayed endogenous reconstitution of protective immunity. Preclinical research in murine models has provided 'proof of concept' for CD8 T-cell therapy of CMV disease. Protection by CD8 T cells appears to be in conflict with the finding that CMVs encode proteins that inhibit antigen presentation to CD8 T cells by interfering with the constitutive trafficking of peptide-loaded MHC class I molecules (pMHC-I complexes) to the cell surface. Here, we have systematically explored antigen presentation in the presence of the three currently noted immune evasion proteins of murine CMV in all possible combinations and its modulation by pre-treatment of cells with interferon-gamma (IFN-γ). The data reveal improvement in antigen processing by pre-treatment with IFN-γ can almost overrule the inhibitory function of immune evasion molecules in terms of pMHC-I expression levels capable of triggering most of the specific CD8 T cells, though the intensity of stimulation did not retrieve their full functional capacity. Notably, an in vivo conditioning of host tissue cells with IFN-γ in adoptive cell transfer recipients constitutively overexpressing IFN-γ (B6-SAP-IFN-γ mice) enhanced the antiviral efficiency of CD8 T cells in this transgenic cytoimmunotherapy model.

Ablation of the regulatory IE1 protein of murine cytomegalovirus alters in vivo pro-inflammatory TNF-alpha production during acute infection

Little is known about the role of viral genes in modulating host cytokine responses. Here we report a new functional role of the viral encoded IE1 protein of the murine cytomegalovirus in sculpting the inflammatory response in an acute infection. In time course experiments of infected primary macrophages (MΦs) measuring cytokine production levels, genetic ablation of the immediate-early 1 (ie1) gene results in a significant increase in TNFα production. Intracellular staining for cytokine production and viral early gene expression shows that TNFα production is highly associated with the productively infected MΦ population of cells. The ie1- dependent phenotype of enhanced MΦ TNFα production occurs at both protein and RNA levels. Noticeably, we show in a series of in vivo infection experiments that in multiple organs the presence of ie1 potently inhibits the pro-inflammatory cytokine response. From these experiments, levels of TNFα, and to a lesser extent IFNβ, but not the anti-inflammatory cytokine IL10, are moderated in the presence of ie1. The ie1- mediated inhibition of TNFα production has a similar quantitative phenotype profile in infection of susceptible (BALB/c) and resistant (C57BL/6) mouse strains as well as in a severe immuno-ablative model of infection. In vitro experiments with infected macrophages reveal that deletion of ie1 results in increased sensitivity of viral replication to TNFα inhibition. However, in vivo infection studies show that genetic ablation of TNFα or TNFRp55 receptor is not sufficient to rescue the restricted replication phenotype of the ie1 mutant virus. These results provide, for the first time, evidence for a role of IE1 as a regulator of the pro-inflammatory response and demonstrate a specific pathogen gene capable of moderating the host production of TNFα in vivo.

The suppression of the production rather than the blockage of action of the potent inflammatory mediator TNFα is a particular hallmark of anti-TNFα mechanisms associated with microbial and parasitic infections. Whether this mode of counter-regulation is an important feature of infection by viruses is not clear. Also, it remains to be determined whether a specific pathogen gene in the context of an infection in vivo is capable of modulating levels of TNFα production. In this study we disclose a virus-mediated moderation of TNFα production, dependent on the ie1 gene of murine cytomegalovirus (MCMV). The ie1 gene product IE1 is a well-characterized nuclear protein capable of altering levels of host and viral gene expression although its biological role in the context of a natural infection is to date unknown. We provide evidence showing that ie1 is associated with a moderated pro-inflammatory cytokine response, in particular with TNFα production. Further, we show that the viral moderation of this cytokine is not only readily apparent in vitro but also in the natural host. The identification of a viral gene responsible for this mode of regulation in vivo may have therapeutic potential in the future in both anti-viral and anti-inflammatory strategies.

Single cell detection of latent cytomegalovirus reactivation in host tissue

The molecular mechanisms leading to reactivation of latent cytomegalovirus are not well understood. To study reactivation, the few cells in an organ tissue that give rise to reactivated virus need to be identified, ideally at the earliest possible time point in the process. To this end, mouse cytomegalovirus (MCMV) reporter mutants were designed to simultaneously express the red fluorescent protein mCherry and the secreted Gaussia luciferase (Gluc). Whereas Gluc can serve to assess infection at the level of individual mice by measuring luminescence in blood samples or by in vivo imaging, mCherry fluorescence offers the advatage of detection of infection at the single cell level. To visualize cells in which MCMV was being reactivated, precision-cut lung slices (PCLS) that preserve tissue microanatomy were prepared from the lungs of latently infected mice. By day 3 of cultivation of the PCLS, reactivation was revealed by Gluc expression, preceding the detection of infectious virus by approximately 4 days. Reactivation events in PCLS could be identified when they were still confined to single cells. Notably, using fractalkine receptor-GFP reporter mice, we never observed reactivation originating from CX3CR1(+) monocytes or pulmonary dendritic cells derived therefrom. Furthermore, latent viral genome in the lungs was not enriched in sorted bone-marrow-derived cells expressing CD11b. Taken together, these complementary approaches suggest that CD11b(+) and CX3CR1(+) subsets of the myeloid differentiation lineage are not the main reservoirs and cellular sites of MCMV latency and reactivation in the lungs.

Antigen-presenting cells of haematopoietic origin prime cytomegalovirus-specific CD8 T-cells but are not sufficient for driving memory inflation during viral latency

Expansion of the CD8 T-cell memory pool, also known as 'memory inflation', for certain but not all viral epitopes in latently infected host tissues is a special feature of the immune response to cytomegalovirus. The L(d)-presented murine cytomegalovirus (mCMV) immediate-early (IE) 1 peptide is the prototype of an epitope that is associated with memory inflation. Based on the detection of IE1 transcripts in latently infected lungs it was previously proposed that episodes of viral gene expression and antigenic activity due to desilencing of a limited number of viral genes may drive epitope-specific memory inflation. This would imply direct antigen presentation through latently infected host tissue cells rather than cell death-associated cross-presentation of viral antigens derived from productively infected cells through uninfected, professional antigen-presenting cells (profAPCs). To address the role of bone marrow-derived profAPCs in CD8 T-cell priming and memory to mCMV, we have used here a combined sex-mismatched and MHC class-I mismatched dual-marker bone marrow chimera model in which presentation of the IE1 epitope is restricted to donor-derived sry(+)L(d+) cells of haematopoietic differentiation lineages. Successful CD8 T-cell priming specific for the L(d)- and D(d)-presented inflationary epitopes IE1 and m164, respectively, but selective failure in IE1 epitope-specific memory inflation in these chimeras indicates different modes of antigen presentation involved in CD8 T-cell priming and memory inflation. These data suggest that memory inflation during mCMV latency requires expression of the epitope-presenting MHC class-I molecule by latently infected non-haematopoietic host tissue cells and thus predicts a role for direct antigen presentation in memory inflation.

In vivo impact of cytomegalovirus evasion of CD8 T-cell immunity: facts and thoughts based on murine models

Cytomegaloviruses (CMVs) co-exist with their respective host species and have evolved to avoid their elimination by the hosts' immune effector mechanisms and to persist in a non-replicative state, known as viral latency. There is evidence to suggest that latency is nevertheless a highly dynamic condition during which episodes of viral gene desilencing, which can be viewed as incomplete reactivations, cause intermittent antigenic activity that stimulates CD8 memory-effector T cells and drives their clonal expansion. These T cells are supposed to terminate reactivation before completion of the productive viral cycle. In this view, CMVs do not "evade" their respective host's immune response but are actually held in check all the time, unless the host gets immunocompromised. Accordingly, CMV disease is typically a disease of the immunocompromised host only. Here we review current knowledge about the in vivo role of viral proteins involved in subverting the immune recognition of infected cells with focus on the CD8 T-cell response and viral interference with the MHC class-I pathway of antigenic peptide presentation. Whereas the intracellular functions of these "immune-evasion proteins" are known in molecular detail, knowledge of their in vivo role in CMV biology is only beginning to take shape. Experimental studies on the in vivo function of human CMV (hCMV) immune-evasion proteins prohibits, of course. Studying animal CMVs paradigmatically in the corresponding natural host is therefore used to identify principles from which the role of hCMV immune-evasion proteins can hopefully be inferred. Here we summarize recent insights gained primarily from the murine model.

A novel transmembrane domain mediating retention of a highly motile herpesvirus glycoprotein in the endoplasmic reticulum

Gene m164 of murine cytomegalovirus belongs to the large group of 'private' genes that show no homology to those of other cytomegalovirus species and are thought to represent 'host adaptation' genes involved in virus-host interaction. Previous interest in the m164 gene product was based on the presence of an immunodominant CD8 T-cell epitope presented at the surface of infected cells, despite interference by viral immune-evasion proteins. Here, we provide data to reveal that the m164 gene product shows unusual features in its cell biology. A novel strategy of mass-spectrometric analysis was employed to map the N terminus of the mature protein, 107 aa downstream of the start site of the predicted open reading frame. The resulting 36.5 kDa m164 gene product is identified here as an integral type-I membrane glycoprotein with exceptional intracellular trafficking dynamics, moving within the endoplasmic reticulum (ER) and outer nuclear membrane with an outstandingly high lateral membrane motility, actually 100 times higher than those published for cellular ER-resident proteins. Notably, gp36.5/m164 does not contain any typical ER-retention/retrieval signals, such as the C-terminal motifs KKXX or KXKXX, and does not pass the Golgi apparatus. Instead, it belongs to the rare group of viral glycoproteins in which the transmembrane domain (TMD) itself mediates direct ER retention. This is the first report relating TMD usage of an ER-resident transmembrane protein to its lateral membrane motility as a paradigm in cell biology. We propose that TMD usage for ER retention facilitates free and fast floating in ER-related membranes and between ER subdomains.

Virally infected mouse liver endothelial cells trigger CD8+ T-cell immunity

BACKGROUND & AIMS

Dendritic cell activation through ligation of pattern recognition receptors leading to full functional maturation causes induction of CD8(+) T-cell immunity through increased delivery of costimulatory signals instead of tolerance. Here we investigate whether organ-resident antigen-presenting cells, such as liver sinusoidal endothelial cells (LSECs), also switch from tolerogenic to immunogenic CD8(+) T-cell activation upon such stimulation.

METHODS

Murine LSECs were isolated by immunomagnetic separation and analyzed for functional maturation upon triggering pattern recognition receptors or viral infection employing gene expression analysis and T cell coculture assays. In vivo relevance of the findings was confirmed with bone-marrow chimeric animals.

RESULTS

LSECs expressed numerous pattern recognition receptors that allowed for sentinel function, but ligand-induced activation of these receptors was not sufficient to overcome tolerance induction of CD8(+) T cells. Importantly, viral infection with murine cytomegalovirus caused functional maturation of antigen-presenting LSECs and was sufficient to promote antigen-specific differentiation into effector CD8(+) T cells in the absence of dendritic cells and independent of CD80/86.

CONCLUSIONS

These results shed new light on the generation of organ-specific immunity and may contribute to overcoming tolerance in relevant situations, such as cancer.

The major virus-producing cell type during murine cytomegalovirus infection, the hepatocyte, is not the source of virus dissemination in the host

The course of systemic viral infections is determined by the virus productivity of infected cell types and the efficiency of virus dissemination throughout the host. Here, we used a cell-type-specific virus labeling system to quantitatively track virus progeny during murine cytomegalovirus infection. We infected mice that expressed Cre recombinase selectively in vascular endothelial cells or hepatocytes with a murine cytomegalovirus for which Cre-mediated recombination would generate a fluorescently labeled virus. We showed that endothelial cells and hepatocytes produced virus after direct infection. However, in the liver, the main contributor to viral load in the mouse, most viruses were produced by directly infected hepatocytes. Remarkably, although virus produced in hepatocytes spread to hepatic endothelial cells (and vice versa), there was no significant spread from the liver to other organs. Thus, the cell type producing the most viruses was not necessarily the one responsible for virus dissemination within the host.