Epstein-Barr virus latent membrane protein LMP-2A is sufficient for transactivation of the human endogenous retrovirus HERV-K18 superantigen

Superantigens are microbial proteins that strongly stimulate T cells. We described previously that the Epstein-Barr virus (EBV) transactivates a superantigen encoded by the human endogenous retrovirus, HERV-K18. We now report that the transactivation is dependent upon the EBV latent cycle proteins. Moreover, LMP-2A is sufficient for induction of HERV-K18 superantigen activity.     

Characterization of a novel H2A(-)E+ transgenic model susceptible to heterologous but not self thyroglobulin in autoimmune thyroiditis: thyroiditis transfer with Vbeta8+ T cells.

Recently we reported on a novel H2E transgenic, IA-negative model of experimental autoimmune thyroiditis (EAT) that excludes reactivity to self in its susceptibility pattern to heterologous thyroglobulin (Tg). In conventional, susceptible mouse strains, EAT is inducible with both homologous and heterologous Tg; e.g., human (h)Tg shares conserved thyroiditogenic epitopes with mouse (m)Tg. However, when an H2Ea(k) transgene is introduced into class II-negative B10.Ab(0) mice, which express neither surface IA (mutant Abeta-chain) nor surface IE (nonfunctional Ea gene), the resultant H2E(b) molecules are permissive for EAT induction by hTg, but not self mTg. Also, the hTg-primed cells do not cross-react with mTg. To explore this unique capacity of E+B10.Ab(0) mice to distinguish self from nonself Tg, we have developed T cell lines to examine the T cell receptor repertoire and observed a consistent Vbeta8+ component after repeated hTg stimulation. Enrichment and activation of Vbeta8+ T cells by either superantigen staphylococcal entertoxin B or anti-Vbeta8 in vitro enabled thyroiditis transfer to untreated A-E+ recipients, similar to hTg activation. Vbeta8+ T cells isolated by FACS from hTg-immunized mice also proliferated to hTg in vitro. These studies support the contribution of Vbeta8 genes to the pathogenicity of hTg in this H2A-E+ transgenic model.     

Expansion of effector memory TCR Vbeta4+ CD8+ T cells is associated with latent infection-mediated resistance to transplantation tolerance

Therapies that control largely T cell-dependent allograft rejection in humans also possess the undesirable effect of impairing T cell function, leaving transplant recipients susceptible to opportunistic viruses. Prime among these opportunists are the ubiquitous herpesviruses. To date, studies are lacking that address the effect of viruses that establish a true latent state on allograft tolerance or the effect of tolerance protocols on the immune control of latent viruses. By using a mixed chimerism-based tolerance-induction protocol, we found that mice undergoing latent infection with gammaHV68, a murine gamma-herpesvirus closely related to human gamma-herpesviruses such as EBV and Kaposi's sarcoma-associated herpesvirus, significantly resist tolerance to allografts. Limiting the degree of virus reactivation or innate immune response did not reconstitute chimerism in latently infected mice. However, gammaHV68-infected mice showed increased frequency of CD8+ T cell alloreactivity and, interestingly, expansion of virus-induced, alloreactive, "effector/effector memory" TCR Vbeta4+CD8+ T cells driven by the gammaHV68-M1 gene was associated with resistance to tolerance induction in studies using gammaHV68-M1 mutant virus. These results define the viral gene and immune cell types involved in latent infection-mediated resistance to allograft tolerance and underscore the influence of latent herpesviruses on allograft survival.     

Functional analysis of the env open reading frame in human endogenous retrovirus IDDMK(1,2)22 encoding superantigen activity

Mice harbor a family of endogenous retroviruses, the mouse mammary tumor viruses (MMTV), which encode superantigens. These superantigens are responsible for the deletion of T cells expressing certain Vbeta chains of the T-cell receptor in the thymus. Human T cells are able to recognize MMTV-encoded superantigens presented by human major histocompatibility complex class II-positive cells. Owing to this and to the similarity of the human and murine immune systems, it was speculated that human endogenous retroviruses might also code for superantigens. Recently, it was reported that a proviral clone (IDDMK(1,2)22) of the human endogenous retrovirus family HTDV/HERV-K encodes a superantigen. The putative superantigen gene was located within the env region of the virus. Stimulated by these findings, we amplified by PCR and cloned into eucaryotic expression vectors open reading frames (ORFs) which were identical or very similar to IDDMK(1,2)22. When we transfected these vectors into A20 cells, a murine B-cell lymphoma, we were able to demonstrate mRNA expression and protein production. However, we did not find any evidence that the ORF stimulated human or murine T cells in a Vbeta-specific fashion, the most prominent feature of superantigens.     

Humanized mice mount specific adaptive and innate immune responses to EBV and TSST-1

Here we show that transplantation of autologous human hematopoietic fetal liver CD34+ cells into NOD/SCID mice previously implanted with human fetal thymic and liver tissues results in long-term, systemic human T-cell homeostasis. In addition, these mice show systemic repopulation with human B cells, monocytes and macrophages, and dendritic cells (DCs). T cells in these mice generate human major histocompatibility complex class I- and class II-restricted adaptive immune responses to Epstein-Barr virus (EBV) infection and are activated by human DCs to mount a potent T-cell immune response to superantigens. Administration of the superantigen toxic shock syndrome toxin 1 (TSST-1) results in the specific systemic expansion of human Vbeta2+ T cells, release of human proinflammatory cytokines and localized, specific activation and maturation of human CD11c+ dendritic cells. This represents the first demonstration of long-term systemic human T-cell reconstitution in vivo allowing for the manifestation of the differential response by human DCs to TSST-1.     

Analysis of T-cell receptor Vbeta gene from infiltrating T cells in insulitis and myocarditis in encephalomyocarditis virus-infected BALB/C mice

Encephalomyocarditis (EMC) virus induces insulin-dependent diabetes and myocarditis in several strains of mice. The T-cell receptor (TCR) Vbeta genes of infiltrating T cells in the pancreas and myocardium of BALB/C mice infected with EMC virus D-variant (EMC-D virus) were analyzed. Using a nested two-step polymerase chain reaction (PCR), TCR Vbeta cDNAs were cloned and sequenced. Two and four kinds of TCR Vbeta clones were obtained from T cells infiltrating into the pancreas and myocardium of BALB/C mice infected with EMC-D virus, respectively. The infiltrating lymphocytes in the diabetic mice expressed Vbeta 8.1, 8.2, and 8.3 genes predominantly. Previously, the use of Vbeta 8.2 has been reported in autoimmune diseases such as murine experimental allergic encephalomyelitis (EAE) and non-obese diabetic (NOD) mouse. This study suggests that mice infected with EMC virus are a useful animal model for autoimmune diseases such as insulin-dependent diabetes.     

HLA class II polymorphisms determine responses to bacterial superantigens

The excessive immunological response triggered by microbial superantigens has been implicated in the etiology of a wide range of human diseases but has been most clearly defined for the staphylococcal and streptococcal toxic shock syndromes. Because MHC class II presentation of superantigens to T cells is not MHC-restricted, the possibility that HLA polymorphisms could influence superantigenicity, and thus clinical susceptibility to the toxicity of individual superantigens, has received little attention. In this study, we demonstrate that binding of streptococcal and staphylococcal superantigens to HLA class II is influenced by allelic differences in class II. For the superantigen streptococcal pyrogenic exotoxin A, class II binding is dependent on DQ alpha-chain polymorphisms such that HLA-DQA1*01 alpha-chains show greater binding than DQA1*03/05 alpha-chains. The functional implications of differential binding on T cell activation were investigated in various experimental systems using human T cells and murine Vbeta8.2 transgenic cells as responders. These studies showed quantitative and qualitative differences resulting from differential HLA-DQ binding. We observed changes in T cell proliferation and cytokine production, and in the Vbeta specific changes in T cell repertoire that have hitherto been regarded as a defining feature of an individual superantigen. Our observations reveal a mechanism for the different outcomes seen following infection by toxigenic bacteria.     

T cell activation by coxsackievirus B4 antigens in type 1 diabetes mellitus: evidence for selective TCR Vbeta usage without superantigenic activity.

Numerous clinical and epidemiological studies link enteroviruses such as the Coxsackie virus group with the autoimmune disease type 1 diabetes mellitus (DM). In addition, there are reports that patients with type 1 DM are characterized by skewing of TCR Vbeta chain selection among peripheral blood and intraislet T lymphocytes. To examine these issues, we analyzed TCR Vbeta chain-specific up-regulation of the early T cell activation marker, CD69, on CD4 T cells after incubation with Coxsackievirus B4 (CVB4) Ags. CD4 T cells bearing the Vbeta chains 2, 7, and 8 were the most frequently activated by CVB4. Up-regulation of CD69 by different TCR families was significantly more frequent in new onset type 1 DM patients (p = 0.04), 100% of whom (n = 8) showed activation of CD4 T cells bearing Vbeta8, compared with 50% of control subjects (n = 8; p = 0.04). T cell proliferation after incubation with CVB4 Ags required live, nonfixed APCs, suggesting that the selective expansion of CD4 T cells with particular Vbeta chains resulted from conventional antigen processing and presentation rather than superantigen activity. Heteroduplex analysis of TCR Vbeta chain usage after CVB4 stimulation indicated a relatively polyclonal, rather than oligo- or monoclonal response to viral Ags. These results provide evidence that new-onset patients with type 1 DM and healthy controls are primed against CVB4, and that CD4 T cell responses to the virus have a selective TCR Vbeta chain usage which is driven by viral Ags rather than a superantigen.     

Productive coupling of accessible Vbeta14 segments and DJbeta complexes determines the frequency of Vbeta14 rearrangement

To elucidate mechanisms that regulate Vbeta rearrangement, we generated and analyzed mice with a V(D)J recombination reporter cassette of germline Dbeta-Jbeta segments inserted into the endogenous Vbeta14 locus (Vbeta14(Rep)). As a control, we first generated and analyzed mice with the same Dbeta-Jbeta cassette targeted into the generally expressed c-myc locus (c-myc(Rep)). Substantial c-myc(Rep) recombination occurred in both T and B cells and initiated concurrently with endogenous Dbeta to Jbeta rearrangements in thymocytes. In contrast, Vbeta14(Rep) recombination was restricted to T cells and initiated after endogenous Dbeta to Jbeta rearrangements, but concurrently with endogenous Vbeta14 rearrangements. Thus, the local chromatin environment imparts lineage and developmental stage-specific accessibility upon the inserted reporter. Although Vbeta14 rearrangements occur on only 5% of endogenous TCRbeta alleles, the Vbeta14(Rep) cassette underwent rearrangement on 80-90% of alleles, supporting the suggestion that productive coupling of accessible Vbeta14 segments and DJbeta complexes influence the frequency of Vbeta14 rearrangements. Strikingly, Vbeta14(Rep) recombination also occurs on TCRbeta alleles lacking endogenous Vbeta to DJbeta rearrangements, indicating that Vbeta14 accessibility per se is not subject to allelic exclusion.     

Distribution of cycling T lymphocytes in blood and lymphoid organs during immune responses

Proliferation of murine T lymphocytes in blood, lymph nodes, and spleen was studied in four in vivo stimulation systems, using BrdU pulse-labeling of DNA-synthesizing cells. The T cell response to the superantigen Staphylococcus enterotoxin B (SEB) was studied in detail. Vbeta8+ T cells showed a peak of DNA synthesis 16-24 h after SEB injection, and the percentage of BrdU+ CD4 and CD8 T cells was higher in blood than in lymph nodes and spleen. DNA synthesis was preceded by massive migration of Vbeta8+ cells from blood to lymphoid organs, in which the early activation marker CD69 was first up-regulated. SEB-nonspecific Vbeta6+ cells showed minimal stimulation but, when cycling, also expressed a high level of CD69. The other systems studied were injection of the IFN-gamma inducer polyinosinic:polycytidylic acid, infection by the BM5 variants of murine leukemia virus (the causative agent of murine AIDS), and T cell expansion after transfer of normal bone marrow and lymph node cells into recombinase-activating gene-2-deficient mice. In each case, a peak of T cell proliferation was observed in blood. These data demonstrate the extensive redistribution of cycling T cells in the first few hours after activation. Kinetic studies of blood lymphocyte status appear crucial for understanding primary immune responses because cycling and redistributing T lymphocytes are enriched in the circulating compartment.     

Induction of TCR Vbeta-specific CD8+ CTLs by TCR Vbeta-derived peptides bound to HLA-E

Previous studies have identified murine and human regulatory CD8+ T cells specific for TCR-Vbeta families expressed on autologous activated CD4+ T cells. In the mouse, these regulatory CD8+ T cells were shown to be restricted by the MHC class Ib molecule, Qa-1. In the present study, we asked whether HLA-E, the human functional equivalent of Qa-1, binds Vbeta peptides and whether the HLA-E/Vbeta-peptide complex induces and restricts human CD8+ CTLs. We first created stable HLA-E gene transfectants of the C1R cell line (C1R-E). Two putative HLA-E binding nonapeptides identified in human TCR Vbeta1 and Vbeta2 chains (SLELGDSAL and LLLGPGSGL, respectively) were shown to bind to HLA-E. CD8+ T cells could be primed in vitro by C1R-E cells loaded with the Vbeta1 (C1R-E/V1) or Vbeta2 (C1R-E/V2) peptide to preferentially kill C1R-E cells loaded with the respective inducing Vbeta peptide, compared with targets loaded with the other peptides. Priming CD8+ T cells with untreated C1R-E cells did not induce Vbeta-specific CTLs. Of perhaps more physiological relevance was the finding that the CD8+ CTLs primed by C1R-E/V1 also preferentially killed activated autologous TCR Vbeta1+. Similar results were observed in reciprocal experiments using C1R-E/V2 for priming. Furthermore, anti-CD8 and anti-MHC class I mAbs inhibited this Vbeta-specific killing of C1R-E and CD4+ T cell targets. Taken together, the data provide evidence that certain TCR-Vbeta peptides can be presented by HLA-E to further induce Vbeta-specific CD8+ CTLs.     

V beta 6+ T cells are obligatory for vaccine-induced immunity to Histoplasma capsulatum

We examined TCR usage to a protective fragment of heat shock protein 60 from the fungus, Histoplasma capsulatum. Nearly 90% of T cell clones from C57BL/6 mice vaccinated with this protein were Vbeta6+; the remainder were Vbeta14+. Amino acid motifs of the CDR3 region from Vbeta6+ cells were predominantly IxGGG, IGG, or SxxGG, whereas it was uniformly SFSGG for Vbeta14+ clones. Short term T cell lines from Vbeta6+-depleted mice failed to recognize Ag, and no T cell clones could be generated. To determine whether Vbeta6+ cells were functionally important, we eliminated them during vaccination. Depletion of Vbeta6+ cells abrogated protection in vivo and upon adoptive transfer of cells into TCR alphabeta(-/-) mice. Transfer of a Vbeta6+, but not a Vbeta14+, clone into TCR alphabeta(-/-) mice prolonged survival. Cytokine generation by Ag-stimulated splenocytes from immunized mice depleted of Vbeta6+ cells was similar to that of controls. The efficacy of the Vbeta6+ clone was associated with elevated production of IFN-gamma, TNF-alpha, and GM-CSF compared with that of the Vbeta14+ clone. More Vbeta6+ cells were present in lungs and spleens of TCR alphabeta(-/-) on day 3 postinfection compared with Vbeta14+ cells. Thus, a single Vbeta family was essential for vaccine-induced immunity. Moreover, the mechanism by which Vbeta6+ contributed to protective immunity differed between unfractionated splenocytes and T cell clones.     

Molecular basis of TCR selectivity, cross-reactivity, and allelic discrimination by a bacterial superantigen: integrative functional and energetic mapping of the SpeC-Vbeta2.1 molecular interface

Superantigens activate large fractions of T cells through unconventional interactions with both TCR beta-chain V domains (Vbetas) and MHC class II molecules. The bacterial superantigen streptococcal pyrogenic exotoxin C (SpeC) primarily stimulates human Vbeta2(+) T cells. Herein, we have analyzed the SpeC-Vbeta2.1 interaction by mutating all SpeC residues that make contact with Vbeta2.1 and have determined the energetic and functional consequences of these mutations. Our comprehensive approach, including mutagenesis, functional readouts from both bulk T cell populations, and an engineered Vbeta2.1(+) Jurkat T cell, as well as surface plasmon resonance binding analysis, has defined the SpeC "functional epitope" for TCR engagement. Although only two SpeC residues (Tyr(15) and Arg(181)) are critical for activation of virtually all human CD3(+) T cells, a larger cluster of four hot spot residues are required for interaction with Vbeta2.1. Three of these residues (Tyr(15), Phe(75), and Arg(181)) concentrate their binding energy on the CDR2 loop residue Ser(52a), a noncanonical residue insertion found only in Vbeta2 and Vbeta4 chains. Plasticity of this loop is important for recognition by SpeC. Although SpeC interacts with the Vbeta2.1 hypervariable CDR3 loop, our data indicate these contacts have little to no influence on the functional interaction with Vbeta2.1. These studies also provide a molecular basis for selectivity and cross-reactivity of SpeC-TCR recognition and reveal a degree of fine specificity in these interactions, whereby certain SpeC mutants are capable of distinguishing between different alleles of the same Vbeta domain subfamily.     

CD28 is required for induction and maintenance of immunological memory in toxin-reactive CD4+ T cells in vivo

We previously reported that Vbeta3+ CD4+ T cells maintained a protracted expansion, with the phenotypes of memory Th2 cells, for 30 days in C57BL/6 (B6) mice implanted with SEA-containing mini-osmotic pumps. In the present study, we followed the fate of Vbeta3+ CD4+ T cells in CD28-/- mice. Vbeta3+ CD4+ T cells increased to a degree similar to that of B6 Vbeta3+ CD4+ T cells until day 10 after implantation, then declined rapidly reaching the control level by 28 days. Remaining Vbeta3+ CD4+ T cells at that time did not exhibit memory phenotypes nor Th2-deviated responses. The rapid drop in Vbeta3+ CD4+ T cells in CD28-/- mice was attributable to upregulated induction of apoptosis owing to marginal inductions of Bcl-2 and Bcl-xL. Collectively, these data indicate CD28 to play critical roles in the generation and maintenance of SEA-reactive CD4+ T cells in vivo.     

Signaling lymphocyte activation molecule-associated protein is a negative regulator of the CD8 T cell response in mice

The primary manifestation of X-linked lymphoproliferative syndrome, caused by a dysfunctional adapter protein, signaling lymphocyte activation molecule-associated protein (SAP), is an excessive T cell response upon EBV infection. Using the SAP-/- mouse as a model system for the human disease, we compared the response of CD8+ T cells from wild-type (wt) and mutant mice to various stimuli. First, we observed that CD8+ T cells from SAP-/- mice proliferate more vigorously than those from wt mice upon CD3/CD28 cross-linking in vitro. Second, we analyzed the consequence of SAP deficiency on CTL effector function and homeostasis. For this purpose, SAP-/- and wt mice were infected with the murine gamma-herpesvirus 68 (MHV-68). At 2 wk postinfection, the level of viral-specific CTL was much higher in mutant than in wt mice, measured both ex vivo and in vivo. In addition, we established that throughout 45 days of MHV-68 infection the frequency of virus-specific CD8+ T cells producing IFN-gamma was significantly higher in SAP-/- mice. Consequently, the level of latent infection by MHV-68 was considerably lower in SAP-/- mice, which indicates that SAP-/- CTL control this infection more efficiently than wt CTL. Finally, we found that the Vbeta4-specific CD8+ T cell expansion triggered by MHV-68 infection is also enhanced and prolonged in SAP-/- mice. Taken together, our data indicate that SAP functions as a negative regulator of CD8+ T cell activation.     

Granulocyte colony-stimulating factor modulates alpha-galactosylceramide-responsive human Valpha24+Vbeta11+NKT cells

Despite more than a 10-fold increase in T cell numbers in G-CSF-mobilized peripheral blood stem cell (PBSC) grafts, incidence and severity of acute graft-vs-host disease (GVHD) are comparable to bone marrow transplantation. As CD1d-restricted, Valpha24+Vbeta11+ NKT cells have pivotal immune regulatory functions and may influence GVHD, we aimed to determine whether G-CSF has any effects on human NKT cells. In this study, we examined the frequency and absolute numbers of peripheral blood NKT cells in healthy stem cell donors (n = 8) before and following G-CSF (filgrastim) treatment. Effects of in vivo and in vitro G-CSF on NKT cell cytokine expression profiles and on responsiveness of NKT cell subpopulations to specific stimulation by alpha-galactosylceramide (alpha-GalCer) were assessed. Contrary to the effects on conventional T cells, the absolute number of peripheral blood NKT cells was unaffected by G-CSF administration. Furthermore, responsiveness of NKT cells to alpha-GalCer stimulation was significantly decreased (p < 0.05) following exposure to G-CSF in vivo. This hyporesponsiveness was predominantly due to a direct effect on NKT cells, with a lesser contribution from G-CSF-mediated changes in APC. G-CSF administration resulted in polarization of NKT cells toward a Th2, IL-4-secreting phenotype following alpha-GalCer stimulation and preferential expansion of the CD4+ NKT cell subset. We conclude that G-CSF has previously unrecognized differential effects in vivo on NKT cells and conventional MHC-restricted T cells, and effects on NKT cells may contribute to the lower than expected incidence of GVHD following allogeneic peripheral blood stem cell transplantation.