Differential Inhibition of Macrophage Activation by Lymphocytic Choriomeningitis Virus and Pichinde Virus Is Mediated by the Z Protein N-Terminal Domain.

Several arenavirus pathogens, such as Lassa and Junin viruses, inhibit macrophage activation, the molecular mechanism of which is unclear. We show that lymphocytic choriomeningitis virus (LCMV) can also inhibit macrophage activation, in contrast to Pichinde and Tacaribe viruses, which are not known to naturally cause human diseases. Using a recombinant Pichinde virus system, we show that the LCMV Z N-terminal domain (NTD) mediates the inhibition of macrophage activation and immune functions.

In vitro and in vivo characterizations of pichinde viral nucleoprotein exoribonuclease functions.

UNLABELLED: Arenaviruses cause severe hemorrhagic fever diseases in humans, and there are limited preventative and therapeutic measures against these diseases. Previous structural and functional analyses of arenavirus nucleoproteins (NPs) revealed a conserved DEDDH exoribonuclease (RNase) domain that is important for type I interferon (IFN) suppression, but the biological roles of the NP RNase in viral replication and host immune suppression have not been well characterized. Infection of guinea pigs with Pichinde virus (PICV), a prototype arenavirus, can serve as a surrogate small animal model for arenavirus hemorrhagic fevers. In this report, we show that mutation of each of the five RNase catalytic residues of PICV NP diminishes the IFN suppression activity and slightly reduces the viral RNA replication activity. Recombinant PICVs with RNase catalytic mutations can induce high levels of IFNs and barely grow in IFN-competent A549 cells, in sharp contrast to the wild-type (WT) virus, while in IFN-deficient Vero cells, both WT and mutant viruses can replicate at relatively high levels. Upon infection of guinea pigs, the RNase mutant viruses stimulate strong IFN responses, fail to replicate productively, and can become WT revertants. Serial passages of the RNase mutants in vitro can also generate WT revertants. Thus, the NP RNase function is essential for the innate immune suppression that allows the establishment of a productive early viral infection, and it may be partly involved in the process of viral RNA replication. IMPORTANCE: Arenaviruses, such as Lassa, Lujo, and Machupo viruses, can cause severe and deadly hemorrhagic fever diseases in humans, and there are limited preventative and treatment options against these diseases. Development of broad-spectrum antiviral drugs depends on a better mechanistic understanding of the conserved arenavirus proteins in viral infection. The nucleoprotein (NPs) of all arenaviruses carry a unique exoribonuclease (RNase) domain that has been shown to be critical for the suppression of type I interferons. However, the functional roles of the NP RNase in arenavirus replication and host immune suppression have not been characterized systematically. Using a prototype arenavirus, Pichinde virus (PICV), we characterized the viral growth and innate immune suppression of recombinant RNase-defective mutants in both cell culture and guinea pig models. Our study suggests that the NP RNase plays an essential role in the suppression of host innate immunity, and possibly in viral RNA replication, and that it can serve as a novel target for developing antiviral drugs against arenavirus pathogens.

Loss of anti-viral immunity by infection with a virus encoding a cross-reactive pathogenic epitope.

T cell cross-reactivity between different strains of the same virus, between different members of the same virus group, and even between unrelated viruses is a common occurrence. We questioned here how an intervening infection with a virus containing a sub-dominant cross-reactive T cell epitope would affect protective immunity to a previously encountered virus. Pichinde virus (PV) and lymphocytic choriomeningitis virus (LCMV) encode subdominant cross-reactive NP205₋212 CD8 T cell epitopes sharing 6 of 8 amino acids, differing only in the MHC anchoring regions. These pMHC epitopes induce cross-reactive but non-identical T cell receptor (TCR) repertoires, and structural studies showed that the differing anchoring amino acids altered the conformation of the MHC landscape presented to the TCR. PV-immune mice receiving an intervening infection with wild type but not NP205-mutant LCMV developed severe immunopathology in the form of acute fatty necrosis on re-challenge with PV, and this pathology could be predicted by the ratio of NP205-specific to the normally immunodominant PV NP38₋45-specific T cells. Thus, cross-reactive epitopes can exert pathogenic properties that compromise protective immunity by impairing more protective T cell responses.

Virus-induced transient immune suppression and the inhibition of T cell proliferation by type I interferon.

Vaccine-induced memory is necessary for protective immunity to pathogens, but many viruses induce a state of transient immune suppression that might contribute to the inability of a vaccine to elicit immunity. We evaluated here the fate of bystander T cells activated by third party cognate antigens during acute viral infections in vivo, using distinct models to track and specifically activate HY and P14 transgenic bystander CD8 T cells in vivo during acute arenavirus infections of mice. Viral infections acted as stimulatory adjuvants when bystander T cells were exposed to an inflammatory milieu and cognate antigens at the beginning of infections, but bystander CD8 T cell proliferation in response to cognate antigen was inhibited 3 to 9 days after virus infection. Reduced proliferation was not dependent on Fas-FasL- or tumor necrosis factor (TNF)-induced activation-induced cell death or on deficiencies of antigen presentation. Instead, reduced proliferation was associated with a delayed onset of division that was an intrinsic defect of T cells. Inhibition of proliferation could be simulated by exposure of T cells to the Toll-like receptor agonist and type I interferon (IFN) inducer poly(I · C). T cells lacking IFN-α/ß receptors resisted both the suppressive effects of preexposure to poly(I · C) and the stimulatory effects of type I IFN, indicating that the timing of exposure to IFN can have negative or positive effects on T cell proliferation. Inhibition of T cell receptor-stimulated bystander CD8 T cell proliferation during acute viral infections may reflect the reduced ability of vaccines to elicit protective immunity when administered during an acute illness.

IFN-alpha beta and self-MHC divert CD8 T cells into a distinct differentiation pathway characterized by rapid acquisition of effector functions.

Nonvirus-specific bystander CD8 T cells bathe in an inflammatory environment during viral infections. To determine whether bystander CD8 T cells are affected by these environments, we examined P14, HY, and OT-I TCR transgenic CD8 T cells sensitized in vivo by IFN-alphabeta-inducing viral infections or by polyinosinic:polycytidylic acid. These sensitized cells rapidly exerted effector functions, such as IFN-gamma production and degranulation, on contact with their high-affinity cognate Ag. Sensitization required self-MHC I and indirect effects of IFN-alphabeta, which together upregulated the T-box transcription factor Eomesodermin, potentially enabling the T cells to rapidly transcribe CTL effector genes and behave like memory cells rather than naive T cells. IL-12, IL-15, IL-18, and IFN-gamma were not individually required for sensitization to produce IFN-gamma, but IL-15 was required for upregulation of granzyme B. These experiments indicate that naive CD8 T cells receive signals from self-MHC and IFN-alphabeta and that, by this process, CD8 T cell responses to viral infection can undergo distinct differentiation pathways, depending on the timing of Ag encounter during the virus-induced IFN response.

RIG-I and MDA5 Protect Mice From Pichinde Virus Infection by Controlling Viral Replication and Regulating Immune Responses to the Infection.

RIG-I and MDA5 are major cytoplasmic innate-immune sensor proteins that recognize aberrant double-stranded RNAs generated during virus infection to activate type 1 interferon (IFN-I) and IFN-stimulated gene (ISG) expressions to control virus infection. The roles of RIG-I and MDA5 in controlling replication of Pichinde virus (PICV), a mammarenavirus, in mice have not been examined. Here, we showed that MDA5 single knockout (SKO) and RIG-I/MDA5 double knockout (DKO) mice are highly susceptible to PICV infection as evidenced by their significant reduction in body weights during the course of the infection, validating the important roles of these innate-immune sensor proteins in controlling PICV infection. Compared to the wildtype mice, SKO and DKO mice infected with PICV had significantly higher virus titers and lower IFN-I expressions early in the infection but appeared to exhibit a late and heightened level of adaptive immune responses to clear the infection. When a recombinant rPICV mutant virus (rPICV-NPmut) that lacks the ability to suppress IFN-I was used to infect mice, as expected, there were heightened levels of IFN-I and ISG expressions in the wild-type mice, whereas infected SKO and DKO mice showed delayed mouse growth kinetics and relatively low, delayed, and transient levels of innate and adaptive immune responses to this viral infection. Taken together, our data suggest that PICV infection triggers activation of immune sensors that include but might not be necessarily limited to RIG-I and MDA5 to stimulate effective innate and adaptive immune responses to control virus infection in mice.

Heterologous arenavirus vector prime-boost overrules self-tolerance for efficient tumor-specific CD8 T cell attack.

Therapeutic vaccination regimens inducing clinically effective tumor-specific CD8+ T lymphocyte (CTL) responses are an unmet medical need. We engineer two distantly related arenaviruses, Pichinde virus and lymphocytic choriomeningitis virus, for therapeutic cancer vaccination. In mice, life-replicating vector formats of these two viruses delivering a self-antigen in a heterologous prime-boost regimen induce tumor-specific CTL responses up to 50% of the circulating CD8 T cell pool. This CTL attack eliminates established solid tumors in a significant proportion of animals, accompanied by protection against tumor rechallenge. The magnitude of CTL responses is alarmin driven and requires combining two genealogically distantly related arenaviruses. Vector-neutralizing antibodies do not inhibit booster immunizations by the same vector or by closely related vectors. Rather, CTL immunodominance hierarchies favor vector backbone-targeted responses at the expense of self-reactive CTLs. These findings establish an arenavirus-based immunotherapy regimen that allows reshuffling of immunodominance hierarchies and breaking self-directed tolerance for efficient tumor control.

Cross-reactivities in memory cytotoxic T lymphocyte recognition of heterologous viruses.

Analyses of the relationships between different viruses and viral proteins have focused on homologies between linear amino acid sequences, but cross-reactivities at the level of T cell recognition may not be dependent on a conserved linear sequence of several amino acids. The CTL response to Pichinde virus (PV) and vaccinia virus (VV) in C57BL/6 mice previously immunized with lymphocytic choriomeningitis virus (LCMV) included the reactivation of memory cytotoxic T lymphocyte (CTL) specific to LCMV. Limiting dilution assays (LDA) demonstrated that at least part of this reactivation of memory cells in LCMV-immune mice related to cross-reactivity at the clonal level, even though acute infections with these viruses in nonimmune mice elicited CTL responses that did not cross-react in conventional bulk CTL assays. Precursor CTL (pCTL) to LCMV were generated in splenic leukocytes from LCMV-immune mice acutely infected with PV or VV when stimulated in vitro with only the second virus but not with uninfected peritoneal exudate cells (PECs). Cytotoxicity mediated by LCMV-specific CTL clones activated by PV infection was greatly inhibited by anti-CD8 antibody, suggesting that these memory CTL clones recognizing LCMV-infected targets were of low affinity. LCMV-immune splenocytes stimulated in vitro with PV or VV demonstrated a low but significant precursor frequency (p/f) to the heterologous viruses, and splenocytes from PV- or VV-immune mice when stimulated in vitro against LCMV generated a low but significant p/f to LCMV. Short-term CTL clones cross-reactive between LCMV and PV were derived from splenic leukocytes from LCMV-immune mice acutely infected with PV. To distinguish whether the cross-reactivity was directed against a viral peptide or a virus-induced endogenous cellular neoantigen, we demonstrated that a pCTL frequency to PV about 1/4-1/7 that of the frequency to LCMV could be generated from LCMV-immune splenic leukocytes stimulated with the immunodominant LCMV NP peptide. A partially homologous PV peptide generated from the equivalent site to the LCMV NP peptide did not sensitize targets to lysis by either LCMV- or PV-specific CTLs, suggesting that the cross-reactivity in killing was not due to evolutionarily conserved equivalent sequences. Experiments also indicated that prior immunity to one virus could modulate future primary immune responses to a second virus. Elevated pCTL frequencies to PV were seen after acute PV infection of LCMV-immune mice, and elevated pCTL frequencies to LCMV were seen after acute LCMV infection of PV- and VV-immune mice.(ABSTRACT TRUNCATED AT 400 WORDS)

Reduction of otherwise remarkably stable virus-specific cytotoxic T lymphocyte memory by heterologous viral infections.

Experimental analyses of the acute cytotoxic T lymphocyte (CTL) response to viruses have focused on studying these infections in immunologically naive hosts. In the natural environment, however, viral CTL responses occur in hosts that are already immune to other infectious agents. To address which factors contribute to the maintenance and waning of immunological memory, the following study examined the frequencies of virus-specific CTL precursor cells (pCTL) not only using the usual experimental paradigm where mice undergo acute infections with a single virus, and in mice immune to a single virus, but also in immune mice after challenge with various heterologous viruses. As determined by limiting dilution assays, the pCTL frequency (p/f) per CD8+ T cell specific for lymphocytic choriomeningitis virus (LCMV), Pichinde virus (PV), or vaccinia virus (VV) increased during the acute infections, peaking at days 7-8 with frequencies as high as 1/27-1/74. Acute viral infections such as these elicit major expansions in the CD8+ T cell number, which has been reported to undergo apoptosis and decline after most of the viral antigen has been cleared. Although the decline in the total number of virus-specific pCTL after their peak in the acute infection was substantial, for all three viruses the virus-specific p/f per CD8+ T cell decreased only two- to fourfold and remained at these high levels with little fluctuation for well over a year. The ratios of the three immunodominant peptide-specific to total LCMV-specific clones remained unchanged between days 7 and 8 of acute infection and long-term memory, suggesting that the apoptotic events did not discriminate on the basis of T cell receptor specificity, but instead nonspecifically eliminated a large proportion of the activated T cells. However, when one to five heterologous viruses (LCMV, PV, VV, murine cytomegalovirus, and vesicular stomatitis virus) were sequentially introduced into this otherwise stable memory pool, the stability of the memory pool was disrupted. With each successive infection, after the immune system had returned to homeostasis, the memory p/f specific to viruses from earlier infections declined. Reductions in memory p/f were observed in all tested immunological compartments (spleen, peripheral blood, lymph nodes, and peritoneal cavity), and on average in the spleen revealed a 3 +/- 0.4-fold decrease in p/f after one additional viral infection and an 8.4 +/- 3-fold decrease after two additional viral infections. Thus, subsequent challenges with heterologous antigens, which themselves induce memory CTL, may contribute to the waning of CTL memory pool to earlier viruses as the immune system accommodates ever-increasing numbers of new memory cells within a limited lymphoid population. This demonstrates that virus infections do not occur in immunological isolation, and that CD8+ T cell responses are continually being modulated by other infectious agents.

Protective heterologous antiviral immunity and enhanced immunopathogenesis mediated by memory T cell populations.

A basic principle of immunology is that prior immunity results in complete protection against a homologous agent. In this study, we show that memory T cells specific to unrelated viruses may alter the host's primary immune response to a second virus. Studies with a panel of heterologous viruses, including lymphocytic choriomeningitis (LCMV), Pichinde (PV), vaccinia (VV), and murine cytomegalo (MCMV) viruses showed that prior immunity with one of these viruses in many cases enhanced clearance of a second unrelated virus early in infection. Such protective immunity was common, but it depended on the virus sequence and was not necessarily reciprocal. Cell transfer studies showed that both CD4 and CD8 T cell populations from LCMV-immune mice were required to transfer protective immunity to naive hosts challenged with PV or VV. In the case of LCMV-immune versus naive mice challenged with VV, there was an enhanced early recruitment of memory phenotype interferon (IFN) gamma-secreting CD4(+) and CD8(+) cells into the peritoneal cavity and increased IFN-gamma levels in this initial site of virus replication. Studies with IFN-gamma receptor knockout mice confirmed a role for IFN-gamma in mediating the protective effect by LCMV-immune T cell populations when mice were challenged with VV but not PV. In some virus sequences memory cell populations, although clearing the challenge virus more rapidly, elicited enhanced IFN-gamma-dependent immunopathogenesis in the form of acute fatty necrosis. These results indicate that how a host responds to an infectious agent is a function of its history of previous infections and their influence on the memory T cell pool.

Pichinde virus induces microvascular endothelial cell permeability through the production of nitric oxide.

This report is the first to demonstrate infection of human endothelial cells by Pichinde virus (PIC). PIC infection induces an upregulation of the inducible nitric oxide synthase gene; as well as an increase in detectable nitric oxide (NO). PIC induces an increase in permeability in endothelial cell monolayers which can be abrogated at all measured timepoints with the addition of a nitric oxide synthase inhibitor, indicating a role for NO in the alteration of endothelial barrier function. Because NO has shown antiviral activity against some viruses, viral titer was measured after addition of the NO synthase inhibitor and found to have no effect in altering virus load in infected EC. The NO synthase inhibition also has no effect on levels of activated caspases induced by PIC infection. Taken together, these data indicate NO production induced by Pichinde virus infection has a pathogenic effect on endothelial cell monolayer permeability.

Recombinant Tri-Segmented Pichinde Virus as a Novel Live Viral Vaccine Platform.

Pichinde virus (PICV) is a nonpathogenic arenavirus with a bi-segmented RNA genome (L and S segments) that encodes four viral genes. We have developed a reverse genetics system to generate recombinant tri-segmented PICV (rP18tri) that packages three RNA segments (L, S1, and S2) and can encode up to two foreign genes. Using influenza virus HA and NP as model antigens, we show that the rP18tri vector can induce strong humoral and cell-mediated immunity, which further increases upon a booster dose. We propose that this novel rP18tri vector can be developed into a useful vaccine platform for other antigens, particularly when strong cellular immunity and prime-boost vaccination strategy are desired.

MHC basis of T cell-dependent heterologous immunity to arenaviruses.

Having a history of infection with one pathogen may sometimes provide a level of T cell-dependent protective heterologous immunity to another pathogen. This immunity was initially thought due to cross-reactive T cell epitopes, but recent work has suggested that such protective immunity can be initiated nonspecifically by the action of cytokines on memory T cells. We retested this concept using two small and well-defined arenaviruses, lymphocytic choriomeningitis virus (LCMV) and Pichinde virus (PV), and found that heterologous immunity in these systems was indeed linked to T cell epitopes and the major histocompatibility complex.

Genome comparison of virulent and avirulent strains of the Pichinde arenavirus.

A virulent (P18) strain of the Pichinde arenavirus produces a disease in guinea pigs that somewhat mimics human Lassa fever, whereas an avirulent (P2) strain of this virus is attenuated in infected animals. It has been speculated that the composition of viral genomes may confer the degree of virulence in an infected host; the complete sequence of the viral genomes, however, is not known. Here, we provide for the first time genomic sequences of the S and L segments for both the P2 and P18 strains. Sequence comparisons identify three mutations in the GP1 subunit of the viral glycoprotein, one in the nucleoprotein NP, and five in the viral RNA polymerase L protein. These mutations, alone or in combination, may contribute to the acquired virulence of Pichinde virus infection in animals. The three amino acid changes in the variable region of the GP1 glycoprotein subunit may affect viral entry by altering its receptor-binding activity. While NP has previously been shown to modulate host immune responses to viral infection, we found that the R374 K change in this protein does not affect the NP function of suppressing interferon-beta expression. Four out of the five amino acid changes in the L protein occur in a small region of the protein that may contribute to viral virulence by enhancing its function in viral genomic RNA synthesis.

Cytokine patterns in a comparative model of arenavirus haemorrhagic fever in guinea pigs.

Arenaviruses such as Lassa virus cause a spectrum of disease in humans ranging from mild febrile illness to lethal haemorrhagic fever. The contributions of innate immunity to protection or pathogenicity are unknown. We compared patterns of expression of cytokines of innate immunity in mild versus severe arenavirus disease using an established guinea pig model based on the macrophage-tropic arenavirus Pichinde virus (PICV). Cytokine transcripts were measured by using real-time RT-PCR in target organs and blood during mild infection (caused by PICV, P2 variant) and lethal haemorrhagic fever (caused by PICV, P18 variant). In the initial peritoneal target cells, virulent P18 infection was associated with significantly increased gamma interferon (IFN-gamma) and monocyte chemoattractant protein-1 (MCP-1, CCL2) mRNA levels relative to P2 infection. Peritoneal cells from P18-infected animals had decreased tumour necrosis factor alpha (TNF-alpha), interleukin (IL)-8 (CXCL-8) and IL-12p40 transcripts relative to mock-infected animals. Late in infection, P18-infected peripheral blood leukocytes (PBL) had decreased TNF-alpha, IFN-gamma, and regulated upon activation, normal T cell expressed and secreted (RANTES, CCL-5) cytokine transcripts relative to P2-infected PBL. We conclude that, in severe arenavirus disease, patterns of cytokine expression in the initially infected cells favour recruitment of additional target monocytes, while inhibiting some of their pro-inflammatory responses. Suppression rather than overexpression of pro-inflammatory cytokines accompanied the terminal shock in this model of arenavirus haemorrhagic fever.

Targeting inside-out phosphatidylserine as a therapeutic strategy for viral diseases.

There is a pressing need for antiviral agents that are effective against multiple classes of viruses. Broad specificity might be achieved by targeting phospholipids that are widely expressed on infected host cells or viral envelopes. We reasoned that events occurring during virus replication (for example, cell activation or preapoptotic changes) would trigger the exposure of normally intracellular anionic phospholipids on the outer surface of virus-infected cells. A chimeric antibody, bavituximab, was used to identify and target the exposed anionic phospholipids. Infection of cells with Pichinde virus (a model for Lassa fever virus, a potential bioterrorism agent) led to the exposure of anionic phospholipids. Bavituximab treatment cured overt disease in guinea pigs lethally infected with Pichinde virus. Direct clearance of infectious virus from the blood and antibody-dependent cellular cytotoxicity of virus-infected cells seemed to be the major antiviral mechanisms. Combination therapy with bavituximab and ribavirin was more effective than either drug alone. Bavituximab also bound to cells infected with multiple other viruses and rescued mice with lethal mouse cytomegalovirus infections. Targeting exposed anionic phospholipids with bavituximab seems to be safe and effective. Our study demonstrates that anionic phospholipids on infected host cells and virions may provide a new target for the generation of antiviral agents.

Rapid conversion of effector mechanisms from NK to T cells during virus-induced lysis of allogeneic implants in vivo.

Viral infections can strongly stimulate both NK cell and allospecific CD8 T cell responses, and these same effector cells can lyse allogeneic cell lines in vitro. However, the impact of viral infections on the effector systems mediating rejection of allogeneic tissues in vivo has not been fully explored. Using in vivo cytotoxicity assays, we evaluated the effector systems mediating the rejection of CFSE-labeled allogeneic splenocytes after an infection of C57BL/6 (B6) mice with lymphocytic choriomeningitis virus. Naive B6 mice predominantly used a NK cell-effector mechanism to reject allogeneic splenocytes because they rejected BALB/C (H2(d)) splenocytes but not CBA (H2(k)) splenocytes, and the rejection was prevented by immunodepletion of NK1.1(+) or Ly49D(+) NK cells. This rapid and efficient in vivo cytotoxicity assay recapitulated the specificity of NK cell-mediated rejection seen in longer duration in vivo assays. However, as early as 1 day after infection with lymphocytic choriomeningitis virus, a CD8 T cell-dependent mechanism participated in the rejection process and a broader range of tissue haplotypes (e.g., H2(k)) was susceptible. The CD8 T cell-mediated in vivo rejection process was vigorous at a time postinfection (day 3) when NK cell effector functions are peaking, indicating that the effector systems used in vivo differed from those observed with in vitro assays measuring the killing of allogeneic cells. This rapid generation of allospecific CTL activity during a viral infection preceded the peak of viral epitope-specific T cell responses, as detected by in vivo or in vitro cytotoxicity assays.

Roads from vaccines to therapies.

Over the past decade, we have demonstrated that various recombinant fragments of botulinum neurotoxin are highly immunogenic, stimulating notable levels of protective antibodies in mice, guinea pigs, and nonhuman primates. One of the fragments evaluated, the fragment C, is a potential next-generation vaccine candidate to replace the current pentavalent botulinum toxoid vaccine. Synthetic genes encoding the carboxyl-terminal regions (approximately 50 kDa) of toxin types A, B, C1, E, and F were expressed in Pichia pastoris, and manufacturing processes were developed for producing highly purified vaccines. These vaccines were shown to be safe, highly efficacious, stable, and amenable to high-level industrial production. Recombinant vaccines are now being produced in accordance with current Good Manufacturing Practices for use in future clinical trials. As our discovery-based program on vaccine development is diminishing, it is concurrently being replaced with a program focused on developing therapeutic interventions to botulism. Synthetic genes encoding the light chains of botulinum toxin have been expressed in Escherichia coli, and purified. These proteolytically active light chains are being used in high-throughput assays to screen for inhibitors of its catalytic activity. Other resources developed as part of the vaccine initiative, likewise, are finding utility in the quest to develop therapies for botulism.

T cell immunodominance and maintenance of memory regulated by unexpectedly cross-reactive pathogens.

We show here that T cell cross-reactivity between heterologous viruses influences the immunodominance of virus-specific CD8(+) T cells by two mechanisms. First, T cells specific for cross-reactive epitopes dominate acute responses to viral infections; second, within the memory pool, T cells specific for cross-reactive epitopes are maintained while those specific for non-cross-reactive epitopes are selectively lost. These findings suggest an immunological paradigm in which viral infections shape the available T cell repertoire, causing alterations in the hierarchies of both the primary and memory CD8(+) T cell responses elicited by subsequent viral infections. Thus, immunodominance is a function of the host's previous exposure to unrelated pathogens, and this may have an impact on protective immunity and immunopathology.

Attrition of T cell memory: selective loss of LCMV epitope-specific memory CD8 T cells following infections with heterologous viruses.

Using a variety of techniques, including limiting dilution assays (LDA), intracellular IFNgamma assays, and Db-IgG1 MHC dimer staining to measure viral peptide-specific T cell number and function, we show here that heterologous virus infections quantitatively delete and qualitatively alter the memory pool of T cells specific to a previously encountered virus. We also show that a prior history of a virus infection can alter the hierarchy of the immunodominant peptide response to a second virus and that virus infections selectively reactivate memory T cells with distinct specificities to earlier viruses. These results are consistent with a model for the immune system that accommodates memory T cell populations for multiple pathogens over the course of a lifetime.