Type I IFN exhaustion is a host defence protecting against secondary bacterial infections

Type I interferons (IFN‐I) have been known for decades for their indispensable role in curtailing viral infections. It is, however, now also increasingly recognized that IFN‐I is detrimental to the host in combating a number of bacterial infections. We have previously reported that viral infections induce partial lymphocyte activation, characterized by significant increases in the cell surface expression of CD69 and CD86, but not CD25. This systemic partial activation of lymphocytes, mediated by IFN‐I, is rapid and is followed by a period of IFN‐I unresponsiveness. Here we propose that IFN‐I exhaustion that occurs soon after a primary viral infection may be a host response protecting it from secondary bacterial infections.

Granzyme B is expressed in mouse mast cells in vivo and in vitro and causes delayed cell death independent of perforin

Mast cells respond to pathogens and allergens by secreting a vast array of preformed and newly synthesized mediators, including enzymes, vasoactive amines, lipid mediators, cytokines and chemokines, thereby affecting innate and adaptive immune responses and pathogenesis. Here, we present evidence that skin-, but not lung-associated primary mast cells as well as in vitro-differentiated bone marrow-derived mast cells (BMMC) express granzyme (gzm) B, but not gzmA or perforin (perf). GzmB is associated with cytoplasmic granules of BMMC and secreted after Fcepsilon-receptor-mediated activation. BMMC from wild type but not gzmB-deficient mice cause cell death in susceptible adherent target cells, indicating that the perf-independent cytotoxicity of BMMC is executed by gzmB. Furthermore, gzmB induces a disorganization of endothelial cell-cell contacts. The data suggest that activated mast cells contribute, via secreted gzmB, to cell death, increased vascular permeability, leukocyte extravasation and subsequent inflammatory processes in affected tissues.

Immunogenicity of two peptide determinants in the cytolytic T-cell response to flavivirus infection: inverse correlation between peptide affinity for MHC class I and T-cell precursor frequency

We used the CD8+ cytotoxic T (Tc) cell immune response against the flavivirus, Murray Valley encephalitis virus (MVE), restricted by the H-2Kk major histocompatibility complex (MHC) class I molecule, to investigate immunodominance. Split-clone limiting dilution analysis revealed almost exclusive recognition of two peptides, MVE1785 and MVE1971, derived from the viral NS3 protein. The precursor frequency of MVE-reactive Tc cells was determined by limiting dilution analysis for cytotoxic function and intracellular staining for interferon-gamma; the latter gave a 100-fold higher estimate of MVE-reactive Tc cell precursors. MHC class I cell surface stabilization assays revealed that affinity for H-2Kk as well as halflives of the peptide-H-2Kk-complexes were markedly different for the two peptides. However, a kinetic study of antigen presentation showed that both peptides are presented for recognition by Tc cells with a comparable kinetics during the latent period of virus infection. Nevertheless, the lower affinity peptide MVE1785 elicited roughly twofold more Tc cell clones than the high-affinity peptide MVE1971. While the cytolytic activity against both determinants was similar after in vitro restimulation at the peak of the primary response, the smaller pool of memory anti-MVE1971 Tc cells correlated with an impaired memory response against that determinant, suggesting that the available T-cell repertoire is a major factor influencing the establishment of T-cell memory.

Effector cytolotic function but not IFN-gamma production in cytotoxic T cells triggered by virus-infected target cells in vitro

Cytolysis and interferon(IFN)-gamma production are two independent effector functions of activated cytotoxic T (Tc) cells. We have used the Tc-cell response against the flavivirus, Murray Valley encephalitis virus (MVE), to investigate the requirements for inducing these two functions with regard to antigen-concentration and CD8 coreceptor involvement. Cognate peptide-pulsed target cells triggered cytolysis by primary ex vivo MVE-immune as well as in vitro peptide-restimulated splenocytes at lower peptide concentrations than IFNgamma-production (100-fold lower in the case of primary ex vivo effectors). Little difference was observed in CD8 dependency. Importantly, neither of the effector populations were triggered to produce IFN-gamma by virus-infected target cells, although cytolysis occurred. This result raises the possibility that the levels of presentation of cognate antigen on virus-infected cells in vivo may be below the threshold required for the IFN-gamma production.

Concerted action of the FasL/Fas and perforin/granzyme A and B pathways is mandatory for the development of early viral hepatitis but not for recovery from viral infection

Cytotoxic T lymphocytes (CTL) play a major role in the recovery from primary viral infections and the accompanying tissue injuries. However, it is unclear to what extent the two main cytolytic pathways, perforin-granzyme A and B exocytosis and Fas ligand (FasL)-Fas interaction, contribute to these processes. Here we have employed mouse strains with either spontaneous mutations or targeted gene defects in one or more components of either of the two cytolytic pathways to analyze the molecular basis of viral clearance and induction of hepatitis during lymphocytic choriomeningitis virus infection. Our results reveal that viral clearance is solely dependent on perforin but that virus-induced liver damage only occurs when both the FasL/Fas and the perforin pathways, including granzymes A and B, are simultaneously activated. The finding that development of hepatitis but not viral clearance is dependent on the concomitant activation of FasL-Fas and perforin-granzymes may be helpful in designing novel strategies to prevent hepatic failures during viral infections.

Virulence of mousepox virus is independent of serpin-mediated control of cellular cytotoxicity

We have investigated whether the differential virulence seen of two Ectromelia (Ect) strains, EctMoscow and ECtHampstead egg, is due to mutation or differential regulation of their serpins (SPI). Poxvirus encoded serine proteinase inhibitors (serpins) have been shown to interfere with cytolytic activity of leukocytes and can also determine virulence. We show that the deduced amino acid sequences of SPI-1, 2, and 3 are identical for the highly virulent EctMoscow and the low virulent EctHampstead strains and that the two viruses express similar potential to inhibit T-cell cytotoxicity, in particular, Fas-mediated target cell lysis, by allorective effectors. Virus titres in wild type B6 mice were effectively controlled very early after inoculation with EctHampstead as compared with EctMoscow, but lack of perforin renders B6 mice similarly susceptible to both virus strains. The data demonstrate that in Ect infection the perforin-mediated cytolytic pathway is not the primary target of serpins and suggest that the apparent attenuation of EctHampstead seen in B6 mice is due to control elements distinct from SPI-1, 2, and 3.

Modulation of transporter associated with antigen processing (TAP)-mediated peptide import into the endoplasmic reticulum by flavivirus infection

In contrast to many other viruses that escape the cellular immune response by downregulating major histocompatibility complex (MHC) class I molecules, flavivirus infection can upregulate their cell surface expression. Previously we have presented evidence that during flavivirus infection, peptide supply to the endoplasmic reticulum is increased (A. Müllbacher and M. Lobigs, Immunity 3:207-214, 1995). Here we show that during the early phase of infection with different flaviviruses, the transport activity of the peptide transporter associated with antigen processing (TAP) is augmented by up to 50%. TAP expression is unaltered during infection, and viral but not host macromolecular synthesis is required for enhanced peptide transport. This study is the first demonstration of transient enhancement of TAP-dependent peptide import into the lumen of the endoplasmic reticulum as a consequence of a viral infection. We suggest that the increased supply of peptides for assembly with MHC class I molecules in flavivirus-infected cells accounts for the upregulation of MHC class I cell surface expression with the biological consequence of viral evasion of natural killer cell recognition.

Cell death mediated by alloreactive cytotoxic T cells via the granule exocytosis or the Fas pathway is independent of p34cdc2 kinase: Fas dependent killing of cells arrested in the cell cycle

Inappropriate activation of p34cdc2 kinase has been shown to occur during apoptosis induced by cytotoxic T-cell derived perforin and fragmentin. We analysed the effect of two inhibitors of p34cdc2 kinase on alloreactive Tc-cell-mediated lysis and DNA fragmentation of P815 and L1210 target cells. Olomoucine, a specific inhibitor of cyclin dependent kinases, did not affect DNA fragmentation in the target cells. Lysis of olomoucine-treated target cells as assessed by 51Cr release over a typical 8-h period was also unaffected. We also examined the effects of thapsigargin on target cell death. This toxin causes increased intracellular calcium rises that then result in irreversible inhibition of cyclin dependent kinases, including p34cdc2 kinase. The same extent of specific cell lysis was induced by cytotoxic T cells from perforin(-/-), granzyme B(-/-), granzyme A(-/-), perforin(-/-) X granzymeB(-/-) X granzymeA(-/-) KO mice or normal mice in untreated target cells or target cells treated with either olomoucine or thapsigargin. Similarly DNA fragmentation measured by release of tritiated DNA was also unaffected. Thus inhibition of p34cdc2 kinase affects neither the Fas nor the perforin/granzyme pathways of alloreactive cytotoxic T-cell killing as measured by DNA fragmentation or chromium release. P815 cells treated with olomoucine were arrested in the cell cycle after 12-16 h exposure to the toxin. After cell cycle arrest, target cells now showed enhanced 51Cr release induced by effector cytotoxic T cells (CTL) derived from perforin(-/-) mice compared to untreated cells. This lysis was accompanied by an increase in cell surface Fas expression. Olomoucine induced cell cycle arrest and expression of Fas was reversible and when cells re-entered the cell cycle, surface expression of Fas was lost.

Antiviral cytotoxic T cells cross-reactively recognize disparate peptide determinants from related viruses but ignore more similar self- and foreign determinants

We have investigated the reactivities of cytotoxic T (Tc) cells against the two immunodominant, H-2K(k)-restricted determinants from the FLAVIVIRUS: Murray Valley encephalitis virus (MVE), MVE(1785) (REHSGNEI) and MVE(1971) (DEGEGRVI). The respective Tc cell populations cross-reactively lysed target cells pulsed with determinants from the MVE(1785)- and MVE(1971)-corresponding positions of six other flaviviruses, despite low sequence homology in some cases. Notably, anti-MVE(1785) Tc cells recognized a determinant (TDGEERVI) that shares with the determinant used for stimulation only the carboxyl-terminal amino acid residue, one of two H-2K(k) anchor residues. These reactivity patterns were also observed in peptide-dependent IFN-gamma production and the requirements for in vitro restimulation of memory Tc cells. However, the broad cross-reactivity appeared to be limited to flavivirus-derived determinants, as none of a range of determinants from endogenous mouse-derived sequences, similar to the MVE-determinants, were recognized. Neither were cells infected with a number of unrelated viruses recognized. These results raise the paradox that virus-immune Tc cell responses, which are mostly directed against only a few "immunodominant" viral determinants, are remarkably peptide cross-reactive.

Cytotoxic T cells specifically induce Fas on target cells, thereby facilitating exocytosis-independent induction of apoptosis

Cytotoxic T (Tc) cells deficient in perforin lyse Fas-negative targets after lengthy incubation periods. This process is independent of granzymes, and killing occurs via the Fas pathway for the following reasons. Interaction of perforin-deficient Tc cells with Fas-negative targets leads to an up-regulation of Fas that is dependent on Ag recognition, de novo synthesis, and transport of proteins to the target cell surface. Treatment of effectors with brefeldin A, but not with the exocytosis inhibitor concanamycin, inhibited this process. Lysis of targets is inhibited by anti-Fas Abs, soluble mouse Fas-Fc, and the caspase-cascade inhibitor, crm-A. Targets from Fas-mutant lpr mice are refractory to lysis, and Tc cells from mice deficient in Fas- and perforin-mediated lysis do not lyse Fas-negative targets. The possible relevance of this exocytosis-independent cytolytic process in the regulation of T cell activity and control of pathogens is discussed.

Murine cytomegalovirus replication in salivary glands is controlled by both perforin and granzymes during acute infection

The course of mouse cytomegalovirus (MCMV) infection was compared between wild-type and mutant C57BL / 6 (B6) mice deficient in either RAG-2, perforin, granzyme A, granzyme B or combinations thereof at two time points post infection (p. i.). At day 15 p. i., virus titers were similarly elevated in salivary glands of all mutant, but not wild-type B6 mice and undetectable in lung and spleen tissues of any of the mouse strains. Significant pathological alterations were only seen in salivary glands and spleen from RAG2(- / -), but not in those from other mice whereas few inflammatory foci were observed in lung tissues of all mice except B6. At day 30 p. i., elevated virus titers were observed only in salivary glands, lung and spleen from RAG2(- / -), but in none of the other mice, and were accompanied by extended pathological alterations in all three organs. The data extend previous reports on the critical role of NK / CD8(+) T cells in the early control of MCMV infection by showing that both perforin and granzymes A / B contribute to viral elimination in salivary glands; however, neither of the three molecules alone seem to be indispensable for the final control of infection.

Proteolytic processing of peptides in the lumen of the endoplasmic reticulum for antigen presentation by major histocompatibility class I

We have tested the hypothesis that MHC class I molecules are actively involved as protease in the production of natural MHC class I ligands. First, the structure of a class I molecule was analyzed for homology with catalytic sites of known proteases. While several clusters of amino acids in the restriction element resembled protease active sites, structural discrepancies and the influence of nearby residues suggest that these sites are unlikely to have protease activity. Second, we have tested the presentation of viral cytotoxic T cell determinants with affinity for the same restriction element (H-2K(d) or K(k)), when targeted as tandem peptides into the endoplasmic reticulum. Peptide transporter-defective cells were used to exclude cleavage of the tandem peptides by cytosolic proteases. Cleavage by signal peptidase of the tandem peptides was ascertained. The C-terminal peptides in the tandem arrays were almost exclusively presented, suggesting that an aminopeptidase in the endoplasmic reticulum degraded the N-terminally positioned peptides. This result is inconsistent with an MHC class I-catalyzed cleavage following binding of longer peptides in the cleft of the restriction elements. Finally, we conclusively show that an aminopeptidase in the endoplasmic reticulum is also involved in antigen presentation in cells with a functional peptide transporter.

Granzymes are the essential downstream effector molecules for the control of primary virus infections by cytolytic leukocytes

Analysis of perforin-deficient mice has identified the cytolytic pathway and perforin as the preeminent effector molecule in T cell-mediated control of virus infections. In this paper, we show that mice lacking both granzyme A (gzmA) and granzyme B (gzmB), which are, beside perforin, key constituents of cytolytic vesicles, are as incapable as are perforin-deficient mice of controlling primary infections by the natural mouse pathogen ectromelia, a poxvirus. Death of gzmAxgzmB double knockout mice occurred in a dose-dependent manner, despite the expression of functionally active perforin and the absence of an intrinsic defect to generate splenic cytolytic T cells. These results establish that both gzmA and gzmB are indispensable effector molecules acting in concert with perforin in granule exocytosis-mediated host defense against natural viral pathogens.

High peptide affinity for MHC class I does not correlate with immunodominance

Cytotoxic T (Tc)-cell responses against influenza virus infection in BALB/c (H-2d) mice are dominated by Tc clones reactive to the viral nucleoprotein (NP). Here, we report investigations using recombinant vaccinia viruses (VV) encoding major histocompatibility complex (MHC) class I H-2Kd molecules differing by a single amino acid from glutamine (wild-type, Kdw) to histidine (mutant, Kdm) at position 114 located in the floor of the peptide-binding groove. Influenza-infected target cells expressing Kdw were strongly lysed by Kd-restricted Tc cells against A/WSN influenza virus or the immunodominant peptide of viral NP (NPP147-155), whereas infected Kdm-expressing targets gave little or no lysis, respectively, thus showing the immunodominance of NPP147-155. Kdm-expressing target cells saturated with synthetic NPP147-155 (10-5 M) were lysed similarly to Kdw-expressing targets by NPP147-155-specific Tc cells. Thus the defect in influenza-infected Kdm-expressing targets was quantitative; insufficient Kdm-peptide complexes were expressed. Tc-cell responses against four other viruses or alloantigens showed no effect of Kdm. When peptide transport-defective cells were infected with VV-Kdw or VV-Kdm and co-infected with a recombinant VV encoding an endoplasmic reticulum-targeted viral peptide, two influenza haemaglutinin peptides caused higher expression of Kdw than NPP147-155 indicating their higher affinity for Kdw. These results are inconsistent with the hypothesis that immunodominance in the anti-influenza response reflects high affinity of the immunodominant peptide, but are consistent with skewing of the Tc-cell receptor repertoire.

Antigen presentation in syrian hamster cells: substrate selectivity of TAP controlled by polymorphic residues in TAP1 and differential requirements for loading of H2 class I molecules

Expression of mouse major histocompatibility complex (MHC) class I molecules in different cell lines derived from Syrian hamsters has revealed antigen presentation deficiencies of some H2 allelic products in two cell lines (BHK and NIL-2) which were overcome by transient expression of the rat transporter associated with antigen processing (TAP; Lobigs et al. 1995). Here we show that in both cell lines the endogenous MHC class I cell surface expression was completely down-regulated. Lymphokine treatment induced endogenous and recombinant mouse MHC class I cell surface expression to levels similar to that in other Syrian hamster cell lines competent for antigen presentation through transduced H2 molecules. Accordingly, constitutive downregulation of expression of accessory molecules of the MHC class I pathway can reveal differences between H2 class I alleles in antigen presentation not encountered when the expression levels are augmented. In addition to the differential expression of MHC class I pathway genes, two cell lines representing competent (FF) and defective (BHK) antigen presentation phenotypes for mouse class I MHC restriction elements demonstrated substantial sequence polymorphism in Tap1 but not Tap2. Cytokine-treated FF or BHK cells and human TAP-deficient T2 cells transfected with FF or BHK TAP1 in combination with FF TAP2 differed in their preference for C-terminal peptide residues, as shown by an in vitro peptide transport assay. Thus, polymorphic residues in TAP1 can influence the substrate selectivity of the Syrian hamster peptide transporter.

Cell death induced by the Fas/Fas ligand pathway and its role in pathology

Engagement of the cell death surface receptor Fas by Fas ligand (FasL) results in apoptotic cell death, mediated by caspase activation. Cell death mediated via Fas/FasL interaction is important for homeostasis of cells in the immune system and for maintaining immune-privileged sites in the body. Killing via the Fas/FasL pathway also constitutes an important pathway of killing for cytotoxic T cells. Fas ligand is induced in activated T cells, resulting in activation-induced cell death by the Fas/FasL pathway. Recently it has been shown that the Fas receptor can also be up-regulated following a lesion to the cell, particularly that induced by DNA-damaging agents. This can then result in killing of the cell by a Fas/FasL-dependent pathway. Up-regulation of Fas receptor following DNA damage appears to be p53 dependent.

Adenovirus-induced liver pathology is mediated through TNF receptors I and II but is independent of TNF or lymphotoxin

Mice infected with an adenovirus mutant in which the E3 region is deleted, including TNF-resistance genes, develop fatal liver pathology within 3-4 days after infection. At least 10-fold more wild-type virus was needed to cause comparable pathology. These results indicate that the E3 region is critically involved in modulating the pathogenesis of adenovirus infection and that TNF may play a role in liver damage. To explore the latter possibility, the course of disease was examined in infected mice lacking TNFR-I and/or TNFRII, TNF only, or both TNF and lymphotoxin-alpha. Only mice lacking both TNFRI and TNFRII were protected from the lethal affects of the mutant adenovirus. Mice deficient in TNF or TNF and lymphotoxin-alpha displayed the fatal pathology. This outcome is consistent with the existence of another related ligand that binds TNFRI/II to mediate liver damage during infection with this mutant.

Poxvirus-encoded serpins do not prevent cytolytic T cell-mediated recovery from primary infections

Previous observations that the highly conserved poxvirus-encoded serpins inhibit cytotoxic activities of alloreactive CTL via granule and/or Fas-mediated pathways was taken to indicate their involvement in immune evasion by poxviruses. We now show that interference with 51Cr release from target cells by ectromelia and cowpoxvirus is limited to alloreactive but not MHC-restricted CTL. The data are in support of the paramount importance of CTL and its effector molecule perforin in the recovery from primary ectromelia virus infection and question the role of serpins in the evasion of poxviruses from killing by CTL. Further analysis of poxvirus interference with target cell lysis by alloreactive CTL revealed that suppression primarily affects the Fas-mediated, and to a lesser extent, the granule exocytosis pathway. Serpin-2 is the main contributor to suppression for both killing pathways. In addition, inhibition of lysis was shown to be both target cell type- and MHC allotype-dependent. We hypothesize that differences in TCR affinities and/or state of activation between alloreactive and MHC-restricted CTL as well as the quality (origin) of target cells are responsible for the observed phenomenon.

Alloreactive cytotoxic T-cell function, peptide nonspecific

The recognition requirements necessary for murine alloreactive cytotoxic T-cells to carry out their effector function has been investigated using target cells that express a unique class I major histocompatibility complex (MHC)-peptide pair. The human cell line T2 and the murine cell line RMA-S are defective in peptide transport components needed to effectively express stable MHC class I molecules at the cell surface. When T2 cells were infected with a vaccinia virus that encoded the Kd gene and provided with a Kd-motif peptide from the nucleoprotein of influenza virus (NPP), these cells could be lysed by polyclonal allo Kd-reactive cytotoxic T-lymphocytes (CTL). Similar results were obtained with the murine RMA-S-Kd cell line, transfected with cDNA able to express some 'empty' Kd that is heat-labile. Adding another Kd-motif peptide from influenza virus haemagglutinin (HAP) stabilized the surface expression of Kd and allowed the RMA-S-Kd cells to be lysed before or after heat shock. At 27 degrees C anti-Kd alloreactive CTL-lysed target cells in the presence and absence of HAP peptide. Alloreactive CTL appear to have a more stringent requirement for a high density of MHC class I on cell surfaces relative to peptide-specific MHC-restricted CTL. We conclude that while Kd-restricted CTL activity is strictly peptide-specific, anti-Kd-specific alloreactivity is MHC allele-specific, but peptide-nonspecific. This conclusion is at odds with the Standard Model of T-cell receptor (TCR) function, but consistent with the predictions of a Competing Model of TCR function.

Spontaneous mutation at position 114 in H-2Kd affects cytotoxic T cell responses to influenza virus infection

Vaccinia virus (VV)-encoded MHC class I Kd molecules which differ by a single amino acid change from glutamine (Kdw, wild type) to histidine (Kdm, mutant) at position 114 located in the floor of the peptide binding groove were compared in terms of peptide binding and cytotoxic T (Tc) cell recognition. Most anti-viral Tc cells were not affected or only marginally affected. However, the Kdm molecule did not detectably present the immunodominant peptide (NPP147-155) of influenza virus nucleoprotein (NP), encoded by the full-length NP gene either in influenza A virus or recombinant VV. This defect could be overcome by using exogenous synthetic NPP147-155 or translation from a minigene encoding NPP147-155 in VV. Kdw presented NPP147-155 encoded by the full-length NP gene, but Kdw-NPP147-155 complexes were at least 100-fold less abundant than after translation from a minigene.

Perforin is essential for control of ectromelia virus but not related poxviruses in mice

Lack of perforin renders the relatively resistant mouse strain C57BL/6 highly susceptible to the natural mouse pathogen ectromelia virus, a cytopathic orthopoxvirus. This is indicated by increased mortality, elevated virus titers and pathology in liver and spleen, and increased levels of liver enzymes in blood. Cowpox virus on the other hand is more virulent in the presence of perforin than in its absence. An additional lack of granzyme A which together with perforin is a constituent of cytoplasmic granules from cytotoxic T cells increases the virulence of cowpox virus.

In vitro- and ex vivo-derived cytolytic leukocytes from granzyme A x B double knockout mice are defective in granule-mediated apoptosis but not lysis of target cells

Granzyme (gzm) A and gzmB have been implicated in Fas-independent nucleolytic and cytolytic processes exerted by cytotoxic T (Tc) cells, but the underlying mechanism(s) remains unclear. In this study, we compare the potential of Tc and natural killer (NK) cells of mice deficient in both gzmA and B (gzmAxB-/-) with those from single knockout mice deficient in gzmA (-/-), gzmB (-/-), or perforin (-/-) to induce nuclear damage and lysis in target cells. With the exception of perforin-/-, all in vitro- and ex vivo-derived Tc and NK cell populations from the mutant strains induced 51Cr-release in target cells at levels and with kinetics similar to those of normal mice. This contrasts with their capacity to induce apoptotic nuclear damage in target cells. In gzmAxB-/- mice, Tc/NK-mediated target cell DNA fragmentation was not observed, even after extended incubation periods (10 h), but was normal in gzmA-deficient and only impaired in gzmB-deficient mice in short-term (2-4 h), but not long-term (4-10 h), nucleolytic assays. This suggests that gzmA and B are critical for Tc/NK granule- mediated nucleolysis, with gzmB being the main contributor, while target cell lysis is due solely to perforin and independent of both proteases.

Hypothesis: MHC class I, rather than just a flagpole for CD8+ T cells is also a protease in its own right

Ever since the discovery of MHC class I restriction and the onslaught of the dual receptor hypothesis, MHC class I has been perceived as a passive entity in TCR recognition and the appropriate antigen processing and presentation pathways. However, numerous experimental observations and theoretical considerations are difficult or unable to be explained by the accepted mechanism of class I antigen presentation. Proteases within and outside the endoplasmic reticulum (ER) are evoked to be solely responsible for the generation of the appropriate 8-10 amino acid-long peptides associated with MHC class I. A MHC class I with site-restricted ER protease activity would overcome most of the present difficulties in explaining MHC class I antigen presentation.

The generation of memory antigen-specific cytotoxic T cell responses by CD28/CD80 interactions in the absence of antigen

The interaction of co-stimulatory molecules CD80/CD86 on antigen-presenting cells with CD28 on naive CD8+ cytotoxic T (Tc) cells is understood to be critical in the induction of Tc effectors. CD80 is capable of providing signal 2 for the activation of Tc cells, but has no effect if encountered in the absence of specific peptide/MHC complexes (signal 1). We have found that CD80 presented in vitro to resting memory viral-immune or alloimmune Tc cells can provide sufficient stimulus for the generation of effector Tc cells in the absence of specific antigen, the peptide/MHC class I complex. Effector Tc cells generated in vitro from influenza- or class I alloantigen-primed mice by co-stimulation in the absence of antigen require exogenous interleukin (IL)-2 signaling via the cell surface-expressed IL-2 receptor or, under conditions of IL-2 blockade, exogenous IL-7. Activation of memory Tc cells by signal 1 and 2 is independent of IL-2 and IL-7. Although memory influenza-immune Tc cells did respond to CD80 in the absence of antigen, the presence of antigen +CD80 enabled an earlier induction of these Tc cells and they retained their lytic activity in vitro over a longer time period. The capacity of memory Tc cells to be activated by signal 2 alone provides one explanation for the observed heterogeneity of phenotype of memory T cells in vivo and a possible mechanism for the maintenence of memory in the absence of persisting antigen.

Molecular basis of immune evasion strategies by adenoviruses

Human adenoviruses have provided valuable insights into virus-host interactions at the clinical and experimental levels. In addition to the medical importance of adenoviruses in acute infections and the ability of the virus to persist in the host, adenovirus-based recombinants are being developed as potential vaccine vectors. It is now clear that adenoviruses employ various strategies to modulate the innate and the adaptive host immune defences. Adenovirus genome-coded products that interact with the immune response of the host have been identified, and to a large extent the molecular mechanisms of their functions have been revealed. Such knowledge will no doubt influence our approach to the areas of viral pathogenesis, vaccine development and immune modulation for disease management.

Memory alloreactive cytotoxic T cells do not require costimulation for activation in vitro

We have studied the costimulation requirements for the generation of cytotoxic T (Tc) cells in an in vitro recall response to alloantigens. Firstly, we demonstrate that recombinant vaccinia viruses encoding class I MHC can stimulate primary in vivo responses and prime for secondary in vitro responses specific for the immunizing alloantigen. The secondary in vitro response comprises both naive and memory components that are distinguishable kinetically. Naive alloreactive Tc cell precursors are dependent upon the presence of CD80 on the in vitro stimulating population for activation and generation of effector function, as described previously. However, Tc cells from animals primed in vivo with vaccinia virus (VV) encoding allo-MHC do not require CD28-CD80 interactions to respond to the alloantigen presented in vitro. This finding provides further evidence that memory Tc cells have less stringent activation requirements in vitro than naive cells. From limiting dilution analysis of the relative contribution of naive and memory Tc cell precursors in 'primary' responses, to MHC class I alloantigen, memory alloreactive Tc cell precursors, possibly primed by cross-reactive environmental antigens, contribute approximately one-fifth of the precursors. Memory responses exhibit similar precursor frequencies as primary responses. Thus, we conclude that memory is largely a result of qualitative rather than quantitative changes in Tc cell precursors.

Exacerbation of invasive aspergillosis by the immunosuppressive fungal metabolite, gliotoxin

Invasive aspergillosis is a significant cause of death in immunocompromised individuals. The majority of strains of the main causative agent, Aspergillus fumigatus, produce gliotoxin, a secondary metabolite with demonstrated in vitro immunosuppressive activity. Pretreatment of normally resistant mice with a single injection of a sublethal dose of gliotoxin was sufficient to make them susceptible to infection and subsequent death, after challenge with A. fumigatus spores. Animals infected with the non-gliotoxin producing strain survived significantly longer than those infected with a gliotoxin producer. We propose that the release of gliotoxin by A. fumigatus hyphae during infection can exacerbate the pathogenesis of aspergillosis.

Flavivirus-induced up-regulation of MHC class I antigens; implications for the induction of CD8+ T-cell-mediated autoimmunity

Infection of a wide variety of cells of human, mouse and other species' origin by flaviviruses such as WNV, YF, Den, MVE, KUN and JE, increases the cell-surface expression of MHC class I. This MHC class I up-regulation is not due to increased MHC class I synthesis per se, but the result of increased peptide availability in the ER for MHC class I assembly. This is most likely due to the interaction of the viral polyprotein with the ER membrane during viral replication. Flavivirus infection can overcome peptide deficiency in TAP-deficient or non-permissive cell lines such as RMA-S and Syrian hamster cells, BHK and NIL-2. The consequence of this increased MHC class I expression manifests itself in reduced susceptibility to NK cells and augmented lysis by Tc cells. In mice, long-term flavivirus-immune Tc cell memory formation is impaired, following the appearance of strong anti-self Tc cell reactivity observed in in vitro cultures from splenocytes of flavivirus-primed animals. We hypothesize that flavivirus-induced MHC class I up-regulation leads to transient T-cell autoimmunity, followed by down-regulation of both autoimmunity and virus-specific Tc cell memory. Furthermore, we speculate that flavivirus infections of humans in the tropics may be responsible for the observed lower incidence of overt autoimmunity in these geographic regions than in temperate climates where flaviviruses are not endemic.

Granzyme A is critical for recovery of mice from infection with the natural cytopathic viral pathogen, ectromelia

Cytolytic lymphocytes are of cardinal importance in the recovery from primary viral infections. Both natural killer cells and cytolytic T cells mediate at least part of their effector function by target cell lysis and DNA fragmentation. Two proteins, perforin and granzyme B, contained within the cytoplasmic granules of these cytolytic effector cells have been shown to be directly involved in these processes. A third protein contained within these granules, granzyme A, has so far not been attributed with any biological relevance. Using mice deficient for granzyme A, we show here that granzyme A plays a crucial role in recovery from the natural mouse pathogen, ectromelia, by mechanisms other than cytolytic activity.

Aspects of cytotoxic T cell memory

Immunological T cell memory manifest itself in an accelerated second-set graft, or allogeneic tumour cell, rejection. Memory viral-immune cytotoxic T cells have shortened kinetics of induction in vivo and differentiate into more potent effector cells in vitro. The requirements for induction of memory T cells are less stringent than for naive T cells. Memory T cells can be activated by antigen (signal 1) or interaction with co-stimulatory molecules (CD28/CD80, signal 2) alone. Memory T cells are phenotypically distinguishable from naive T cells by a number of cell surface markers, but not from activated T cells. Persistence of antigen is not required for the maintenance of long-lived memory. Continuous stimulation by signal 2 alone and or longevity is sufficient to explain life-long persistence of T cell memory. All available data on memory T cells are consistent with a deterministic model of T cell memory formation, following a precise pathway of T cell differentiation.

Investigation of the potential use of immunosuppressive agent gliotoxin in organ transplantation

Gliotoxin is an immunosuppressive secondary metabolite produced by several pathogenic fungi. It has previously been shown to prevent graft-versus-host disease in transplantation of allogeneic mouse bone marrow and to reduce the immunogenicity of human fetal pancreas. We here report on the effect of gliotoxin on the prevention of rejection of allografts in two distinct models. Bathing mouse thyroid tissue in gliotoxin solution for 16 hr prolonged graft survival following transplantation into allogeneic recipients. In contrast the perfusion of rat kidneys with gliotoxin followed by 1 hr of incubation before orthotopic transplantation had little success with preventing allograft rejection. This disparity is most likely due to the incubation in the renal model not allowing sufficient time for the elimination of antigen presenting cells in the donor organ. However, the success with the thyroid grafts demonstrates the potential of gliotoxin as an immunomodulating agent in organ transplantation and warrants further investigation in other systems.

Granzyme A-deficient mice retain potent cell-mediated cytotoxicity

Granzyme A, a granule-associated serine proteinase of activated cytotoxic T cells and natural killer cells, has been reported to play a critical role in DNA fragmentation of target cells. To address the question of the biological role of granzyme A, we have now generated a granzyme A-deficient mouse mutant by homologous recombination. Western blot analysis, enzyme assays and reverse transcription-PCR confirmed the absence of granzyme A in activated T cells. In addition, deletion of granzyme A does not alter the expression patterns of other granule components, such as granzymes B-G and perforin. Granzyme A-deficient mice are healthy and show normal hematopoietic development. Most notably, their in vitro- and ex vivo-derived cytotoxic T cells and natural killer cells are indistinguishable from those of normal mice in causing membrane disruption, apoptosis and DNA fragmentation in target cells. Furthermore, granzyme A-deficient mice readily recover from both lymphocytic choriomeningitis virus and Listeria monocytogenes infections and eradicate syngeneic tumors with kinetics similar to the wild-type strain. These results demonstrate that granzyme A does not play a primary role in cell-mediated cytotoxicity, as has been assumed previously.

Polymorphic peptide transporters in MHC class I monomorphic Syrian hamster

We have already shown that in species with highly polymorphic major histocompatibility complex (MHC) class I molecules (human, mouse) no functional polymorphism of the peptide transporters TAP1 and TAP2 is detectable (Lobigs and Müllbacher 1993). Investigating the antigen-presentation machinery of the class I MHC monomorphic Syrian hamster using mouse MHC class I expression via recombinant vaccinia viruses (VV) we found that six hamster cell lines fall into two phenotypic classes. four cell lines (HaK, FF, MF-2, and HT-1) showed no defect in expressing four different H2 class I molecules (Kk, Kd, Kb, Dd) and the appropriate VV peptide recognized by mouse VV-immune cytotoxic T (Tc) cells on the cell surface. Two cell lines (BHK-21 and NIL-2) expressed Dd and Kb in association with VV peptides as recognized by VV-immune, H2-restricted Tc cells but not Kk and Kd. However, Kd was expressed on the cell surface, as shown by fluorescence-activated cell sorter (FACS) analysis and alloreactive Tc-cell recognition. Kk is only surface-expressed in these two cell lines when superinfected with two VV recombinants encoding rat TAP1 (VV-mtp1) and TAP2 (VV-mtp2). Superinfection with VV-mtp1 and VV-mtp2 rendered both cell lines, after infection with either VV-Kk and VV-Kd, susceptible to lysis by either Kk- or Kd-restricted VV-immune Tc cells. Thus Syrian hamster cell lines express functionally polymorphic peptide transporters. The TAP2 gene from FF cells was cloned and sequenced; comparison with human, mouse, and rat TAP2 sequences show 78%, 88% and 87% similarity, respectively.

Up-regulation of MHC class I by flavivirus-induced peptide translocation into the endoplasmic reticulum

Flavivirus infection of mammalian cells increases the cell surface expression of major histocompatibility complex (MHC) class I molecules, the recognition elements for cytotoxic T cells. Here, we show that the mechanism for flavivirus-induced up-regulation of class I MHC involves an increase in peptide supply to the endoplasmic reticulum. Flavivirus-mediated peptide supply for MHC class I assembly is independent of the peptide transporters for class I antigen presentation, since infection of class I MHC peptide transport-deficient cell lines with flaviviruses results in the cell surface expression of biologically functional class I MHC peptide complexes. The flavivirus-induced supply of antigenic peptides to the endoplasmic reticulum is not restricted to flavivirus-encoded peptides and independent of interferon. The data imply that peptide availability regulates surface expression of class I MHC restriction elements and suggests a mechanism for flavivirus-induced immunopathology.

Recombinant expression of vaccinia virus-encoded TAP1 and TAP2 promotes MHC class I-restricted antigen presentation in a Syrian hamster cell line

Transporter associated with antigen presentation (TAP) molecules supply small peptides to the lumen of the endoplasmic reticulum for assembly with MHC class I glycoproteins. Here we report that a Syrian hamster cell line, BHK, which is commonly used in cell biological and virological investigations, is deficient in antigen presentation in a xenogeneic model. Thus when the mouse MHC class I restriction element, Kd, is expressed in BHK cells via a vaccinia virus (VV) recombinant, no significant recognition by VV-immune and Kd-restricted cytotoxic T cells is induced. This defect is apparently a consequence of a deficiency in TAP-dependent peptide transport and can be repaired when rat TAP1 and TAP2 are provided by recombinant VV expression.

Down-regulation of human adenovirus E1a by E3 gene products: evidence for translational control of E1a by E3 14.5K and/or E3 10.4K products

The mechanism for down-regulation of E1a expression by products encoded in the E3 transcription unit of human adenovirus types 2 and 5, that occurs in infected L929 cells, has been investigated further. We show that the phenomenon occurs in different mouse cells and also in some human cells suggesting that the observations have relevance to natural human infections. We also provide evidence that probably all viral proteins are down-regulated by E3 products, although to different extents, but that host proteins are unaffected. Whereas E1a protein levels and synthesis are reduced in the presence of E3 products, E1a protein half-life and polysomal E1a RNA levels and size distribution are not. These data suggest that E3 products down-regulate E1a protein levels by interfering with the translation of E1a-specific mRNA. Studies were additionally carried out with mutant adenoviruses containing different defects in the E3 transcription unit. Based on these studies it seems likely that the E3 14.5K and 10.4K proteins are crucially involved in E1a down-regulation. Our data are discussed in terms of strategies for immune evasion by group C human adenoviruses.

The flavivirus nonstructural protein NS3 is a dominant source of cytotoxic T cell peptide determinants

Vaccinia virus recombinants encoding regions of the Murray Valley encephalitis virus (MVE) genome, which together cover the entire viral coding region, were employed to identify the MVE protein which is the dominant source of CD8+, cytotoxic, T cell antigenic determinant(s) presented by the mouse H-2Kk major histocompatibility antigen. MVE and West Nile virus-immune, H-2k-restricted, effector cells recognized peptides derived from the MVE nonstructural polyprotein segment, and in this region the immunodominant determinant mapped to protein NS3. Interestingly, mapping of cytotoxic T cell antigenic determinants of other flaviviruses also identified the NS3 protein as the dominant source of antigenic peptides (A. B. Hill, A. Müllbacher, C. Parrish, G. Coia, E. G. Westaway, and R. V. Blanden, 1992, J. Gen. Virol. 73, 1115-1123; A. L. Rothman, I. Kurane, C.-J. Lai, M. Bray, B. Falgout, R. Men, and F. A. Ennis, 1993, J. Virol. 67, 801-806). Using an allele-specific peptide motiff for H-2Kk, we predicted 12 peptides in the MVE NS3 protein as ligands for the restriction element and identified three peptides which were recognized in association with H-2Kk by MVE-immune cytotoxic T cells. We also examined the effect of proteolytic processing in the MVE nonstructural polyprotein segment mediated by the viral proteinase NS3 on antigen processing and presentation of the MVE H-2Kk-restricted T cell determinant. Processing of the MVE polyprotein by the viral proteinase did not markedly influence the availability of this peptide determinant.

In vivo immunosuppressive activity of gliotoxin, a metabolite produced by human pathogenic fungi

Aspergillosis is a disease caused by the opportunistic pathogen Aspergillus fumigatus and other related fungi. It occurs mainly in immunosuppressed people and causes very high mortality rates. A fumigatus and other pathogenic fungi have been shown to produce a metabolite, gliotoxin, which has immunosuppressive properties in vitro, but little is known about its in vivo activity. Here we report that gliotoxin has increased toxicity in mice after irradiation. A single injection of gliotoxin delayed the recovery of immune cells after immunosuppression by sublethal irradiation by 2 weeks. Study of the morphology of cells of the thymus, spleen, and mesenteric lymph nodes by light microscopy and electron microscopy and agarose gel electrophoresis of DNA from these organs showed that the injection of gliotoxin induced apoptosis in cells of the immune system in vivo. Thus, gliotoxin does have immunosuppressive activity in vivo and could potentially play a significant role in the pathogenesis of aspergillosis and other fungal diseases.

The long-term maintenance of cytotoxic T cell memory does not require persistence of antigen

I have used the transfer of primed lymphocytes into syngeneic irradiated recipients to investigate whether the persistence of antigen is required in the long-term maintenance of cytolytic T cell memory to influenza virus. Animals were immunized with influenza virus (A/WSN) and used 17 wk later as either donors for T cells or as lethally irradiated recipients. Naive age-matched mice served as controls. At intervals of 4, 8, 16, and 25 wk after T cell transfer, experimental and control groups were immunized with a heterologous virus (A/JAP) and splenocytes tested for lytic activity to influenza virus 3 and 6 d after immunization. Lytic activity 3 d after infection (a property exclusive to a memory cytotoxic T cell response) (Effros, R. B., J. Bennink, and P. C. Doherty. 1978. Cell. Immunol. 36:345.; and Hill, A. B., R. V. Blanden, C. R. Parrish, and A. Müllbacher. 1992. Immunol. Cell Biol. 70:259), was only observed by primed and naive irradiated recipients reconstituted with memory T cells. No day 3 responses were observed when naive T cells were transferred into irradiated primed or unprimed recipients. These observations demonstrate that cytolytic T cell memory to influenza virus is long lived in the absence of antigen.

In vivo administration of major histocompatibility complex class I-specific peptides from influenza virus induces specific cytotoxic T cell hyporesponsiveness

We have been investigating the immunogenicity of two class I major histocompatibility complex-specific peptides with a sequence derived from influenza virus nucleoprotein specific for Kd and one for Db. Peptide-modified splenocytes are unable to immunize for a primary cytotoxic T (Tc) cell response in vivo, or secondary response in vitro. Peptide-modified stimulator cells can boost virus-primed splenocytes for a strong secondary response in vitro. Animals primed with syngeneic peptide-modified splenocytes upon challenge with virus in vivo do not generate strong secondary Tc cell responses on day 3 after challenge in contrast to virus primed animals. Day 6 responses of virus-challenged, peptide-primed animals are reduced as compared to unprimed mice. This hyporesponsiveness is independent of CD8+ T cells in the priming population and can be elicited with tumor cell lines. The data are discussed in the framework of the two-signal model of immune induction.

Ir1 genes, peripheral cross-tolerance and immunodominance in MHC class I-restricted T-cell responses: an old quagmire revisited

Previous studies have shown that peripheral tolerance of one MHC molecule may influence the ability of an animal to respond to a particular antigenic determinant in the context of another MHC molecule (cross-tolerance). We describe here an investigation of the extent to which the presence of MHC molecules other than the one involved in Tc cell recognition affected whether or not a particular "antigen" (MHC/peptide) was recognized during the response of mice to infection with the flavivirus, West Nile (WNV). WNV-immune Tc cells from intra-H-2 recombinant mouse strains and F1 hybrid animals were used. In general, the antigens identified as immunogenic for five prototype mouse strains were immunogenic wherever the F1 or recombinant animal possessed the required MHC gene, but there were several notable exceptions. Firstly, while no responses were associated with Kd and Dk in BALB/c (H-2d) and CBA/H (H-2k) mice, respectively, in C3H.OH mice (Kd, Dk) both of these MHC molecules were associated with detectable responses. Secondly, in B10.A(2R) (Kk, Db) mice responses restricted by Db were not found whereas they were present in B6 mice (Kb, Db). This phenomenon was similar to previously reported phenomena with H-Y, Sendai, influenza and VV. However, it differed in two important ways. Firstly, we have no definitive evidence as yet that peripheral cross-tolerance to Kk influences responses against Db plus flavivirus as it does with H-Y and VV. Secondly, in addition to Kk, genes in the H-2s, H-2q and H-2d haplotypes influence responses to Db plus flavivirus but not responses to the other "antigens" listed above. Alternatively, the data are compatible with the concept of "immunodominance" based on a hierarchy of affinities of TCR's. the responding T-cell population appeared to focus the immune response on a limited number of "antigens"; in the presence of certain more strongly immunogenic "antigens", responses to other apparently more weakly immunogenic "antigens" were not seen.

Recognition of vaccinia virus-encoded major histocompatibility complex class I antigens by virus immune cytotoxic T cells is independent of the polymorphism of the peptide transporters

In the cytotoxic T-cell response to viruses, peptide antigens of cytoplasmic origin are presented at the cell surface by the highly polymorphic major histocompatibility complex (MHC) class I molecules to CD8+ T-lymphocyte receptors. Peptide transporter molecules and other MHC-linked gene products have been implicated in the generation and import of antigenic peptides into the lumen of the endoplasmic reticulum for assembly with MHC class I glycoproteins. These accessory molecules in the antigen-presentation pathway map to a polymorphic region in the class II MHC, and the possibility of their allele-specific selectivity in antigen presentation has been raised. Here we show that additional, functionally polymorphic components are not apparent in an in vitro mouse MHC class I-restricted cytotoxic T-cell response to vaccinia and influenza viruses. When the mouse H-2Kd molecule was expressed via a recombinant vaccinia virus in target cells of different mouse MHC haplotypes or cells of rat, Syrian hamster, monkey, and human origin, efficient Kd-restricted and vaccinia virus-specific lysis was observed as measured with bulk effectors and at the clonal level. In addition, human transporters efficiently processed peptides originating from influenza virus nucleoprotein and hemagglutinin antigens as recognized by mouse influenza immune cytotoxic T cells.

Transfer of CD8+ T cells into SCID mice and activation of memory virus-specific cytotoxic T cells

The requirements for the activation of naive and memory CD8+ cytotoxic T (Tc) cells into effector virus-specific Tc cells after transferring them into SCID mice were investigated. SCID mice reconstituted with splenocytes or purified CD8+ T cells from naive or influenza-immune syngeneic mice and immunized with influenza virus generated effector Tc cells specific for influenza virus-infected target cells in vitro. The kinetics of the response varied between those two populations. The generation of effector Tc cells after transfer of memory CD8+ T cells indicates that there exists no absolute requirement for 'help' in the activation of memory virus-immune T cells. However, under the conditions described here the in vitro immunogenic peptide NPP derived from influenza nucleoprotein is not sufficient to elicit a response in vivo.