T and B memory cells

Locus-specific somatic hypermutation in germinal centre T cells

Somatic hypermutation and affinity-driven selection of active immunoglobulin genes occur in germinal centres (GCs), resulting in the generation of high-affinity memory B cells. In contrast, T lymphocytes do not require the germinal centre microenvironment to establish memory and the T-cell antigen receptor (TCR) genes, though homologous to immunoglobulin genes, are believed to be incapable of hypermutation. Here we present direct evidence that the small population of antigen-specific T cells that are recruited into splenic GCs acquire mutations in the variable region of genes encoding TCR alpha-chains (V alpha) but not those of beta-chains. These locus-specific mutations reach frequencies comparable to mutated immunoglobulin VH exons recovered from the same site and exhibit similar substitution biases and DNA strand polarity. T cells bearing identical mutations appear in multiple GCs, raising the possibility that some cells bearing mutant TCRs may re-enter the peripheral lymphocyte pool.

Apoptosis, Fas and systemic autoimmunity: the MRL-lpr/lpr model

Proteins encoded by the fas and fas ligand (fasL) genes are involved in apoptotic cell death in lymphocytes. In this article we review the recent elucidation of the role of the Fas-FasL interactions in the maintenance of tolerance to self antigens and in the homeostatic regulation of lymphocyte clonal expansion, and discuss the mechanisms of autoimmunity in Fas- and FasL-deficient mutant mouse strains.

Co-stimulation by anti-immunoglobulin is required for B cell activation by CD40Llow T cells

During cognate B:T interactions, B cells encounter antigen (Ag) through surface immuno-globulin (sIg) and present antigenic peptides to T helper (Th) cells. However, most in vitro systems used to study contact events involved in the delivery of T help for B cells circumvent the requirement for T cell Ag specificity by using anti-CD3/T cell receptor (TcR) monoclonal antibodies (mAb) to activate T cells. To study the role of sIg engagement in the responsiveness of B cells to T help, we pre-treated small resting B cells with soluble anti-kappa mAb prior to contact with an activated Th1 clone. By reducing the concentration of anti-TcR mAb we obtained low levels of CD40 ligand (CD40Llow) on Th cells, comparable to those expressed by lymph node T cells activated in vitro (ex vivo T cells). In contrast to untreated B cells, which did not respond to CD40Llow Th, anti-Ig-treated B cells responded strongly. Low buoyant density B cells also responded to CD40Llow Th cells. There was no B cell response to resting Th cells. mAb against CD54/intercellular adhesion molecule-1 or major histocompatibility complex (MHC) class II completely inhibited B cell responses to CD40Llow Th1 cells, equivalent to the effects of blocking CD40 interactions. This contrasts with mAb blocking responses to CD40Lhigh Th, where CD40 effects predominate. Our data show that sIg engagement is necessary for the induction of B cell response to CD40Llow Th cells. Anti-CD3-activated ex vivo T cells that were also CD40Llow did not provide help to small resting B cells, but did induce responses from sIg-stimulated B cells. Thus, our data support a requirement for sIg signaling in physiological B cell activation, and further confirm previous work showing CD40 ligation to be necessary but not sufficient for delivery of T help to B cells.

Self-replicating Semliki Forest virus RNA as recombinant vaccine

Recombinant RNA based on the Semliki Forest virus (SFV) replicon was used to express the nucleoprotein of influenza virus in mice. Two strategies were employed to deliver the RNA. In the first, recombinant RNA was packaged into infectious suicide SFV particles which were used directly for immunization. The second approach involved injection of in vitro-synthesized RNA directly into the quadriceps muscle. Both approaches resulted in the generation of humoral responses with high antibody titres. Immunization with suicide particles showed that a strong, class I-restricted cytotoxic T-cell response can be obtained using only 100 infectious units. We conclude that the self-replicative recombinant SFV RNA may be quite useful as a nucleic acid vaccine.

Apoptotic elimination of V beta 8.2+ cells from the central nervous system during recovery from experimental autoimmune encephalomyelitis induced by the passive transfer of V beta 8.2+ encephalitogenic T cells

A CD4+V beta 8.2+ T cell clone specific for the peptide 72-89 of guinea pig myelin basic protein (GMBP) was used to induce acute experimental autoimmune encephalomyelitis (EAE) in Lewis rats. To assess apoptosis in inflammatory cells infiltrating the central nervous system (CNS), we extracted cells from the spinal cord, enriched them for T cells and performed flow-cytometric analysis of their DNA stained with propidium iodide. The presence of apoptosis was confirmed by the demonstration of DNA fragmentation on gel electrophoresis. A gradual increase in the proportion of apoptotic cells was observed between 4 and 7 days after the transfer of the encephalitogenic T cells. The highest frequency of apoptotic cells (9.2 +/- 1.2%) was observed 7 days after cell transfer, when clinical recovery commenced. Passive transfer of ovalbumin-specific cells resulted in only a background level (0.8%) of apoptosis in the CNS. We conclude that the apoptotic process selectively eliminates autoreactive T cells from the CNS as: (a) there was a selective disappearance of disease-relevant CD5+V beta 8.2+ cells from the CNS during the course of EAE; (b) there was a decrease in the frequency of CNS-infiltrating T cells reactive to the GMBP 72-89 peptide during the course of EAE, and in a standard proliferation assay there was a loss of in vitro reactivity of CNS-infiltrating cells to this peptide, but not to a non-CNS antigen (ovalbumin); (c) simultaneous surface labeling and DNA analysis of CNS-infiltrating cells revealed that the frequency of V beta 8.2+ cells was about sevenfold higher in the apoptotic T cell population than in the normal (non-apoptotic) T cell population; and (d) we were unable to detect recirculation of the V beta 8.2+ cells to lymphoid organs after their frequency decreased in the CNS. The selective apoptotic elimination of autoreactive T cells from the target organ of this spontaneously resolving autoimmune disease may have implications for the understanding of the mechanism by which an autoimmune attack is terminated and for the design of therapeutic strategies to facilitate this process.

Generation and in vivo persistence of polarized Th1 and Th2 memory cells

In vitro generated CD4 cell lines and effectors often produce either Th1 or Th2 cytokines, but stimulation of resting CD4 cells from animals leads to production mostly of IL-2. To determine whether polarization of CD4 effector cells results in development of polarized memory, I generated antigen-specific Th1 and Th2 effectors in vitro and transferred them to adoptive hosts. The effectors gave rise to long-lived populations of CD4 T cells with the phenotype of resting memory cells. Recovered cells responded vigorously to specific antigen, producing a pattern of cytokines closely related to that of the transferred effectors. Thus, encounter with a peptide antigen and directive cytokines at the initiation of culture can lead to generation of both effector and long-lived memory cell populations that produce restricted patterns of cytokines.

Advances in immunology

Much was accomplished in the last decade in understanding how the adaptive immune system evolved to combat pathogens. Essential features of antigen presentation and T lymphocyte recognition were deciphered, setting the stage for further studies that elucidated basic elements of lymphocyte differentiation (including positive and negative selection during lymphocyte ontogeny) and the major interactions that occur among cells in secondary lymphoid organs in an ongoing immune response. The major challenges of today are found in the burgeoning fields of programmed cells death, enzymology of recombination and somatic mutation, development of memory, and the recognition of pathogens by unconventional lymphocytes.

CD8+ T-cell memory to viruses

Recent experiments show that laboratory mice infected once with an influenza A virus or with the murine parainfluenza type 1 virus, called the Sendai virus, have enhanced numbers of cytotoxic T-lymphocyte precursors ( > 20x background) for life. Neither virus persists at the genome level, and the mice are maintained under conditions where there is no possibility of re-infection. These observations are highly relevant to any understanding of CD8+ cell memory and suggest that the clonal burst size associated with the primary challenge is a key determining factor.

Clonal expansions of V delta 1+ and V delta 2+ cells increase with age and limit the repertoire of human gamma delta T cells

We have investigated the complexity of the human gamma delta T cell repertoire by means of a VJ heteroduplex analysis method. cDNA obtained from peripheral blood mononuclear cells was amplified with V delta 1-C delta or V delta 2-C delta primers. The product was denatured and renatured to allow random reannealing of the strands and the heteroduplexes carrying mismatched junctional sequences were separated from the homoduplexes on polyacrylamide gels. Whenever one or more T cell clones were expanded to over 10% of the polyclonal background, discrete bands of homo- and heteroduplex appeared. This method was applied to the analysis of the peripheral gamma delta compartment from healthy donors and rheumatoid arthritis patients of different ages. While samples from young individuals showed a polyclonal pattern, a clear tendency towards oligoclonality appeared with increasing age, both in normal individuals and rheumatoid arthritis patients. We also show that the VJ junctional sequence derived from the heteroduplex fragments can be successfully used to isolate and characterize the corresponding T cell clones in vitro, even after a period of 1 year. In conclusion, our findings indicate that the complexity of the gamma delta T cell repertoire decreases with age as a consequence of the expansion of a few T cell clones.

The fas antigen is involved in peripheral but not thymic deletion of T lymphocytes in T cell receptor transgenic mice

The role of a cell death-associated gene, fas, in T lymphocyte development and responses to antigen has been analyzed by breeding a transgenic T cell receptor specific for the 81-104 peptide of pigeon cytochrome c into fas-defective MRL-lpr/lpr and control MRL+/+ mice. Transgene-expressing T cells mature normally in both strains and populate peripheral lymphoid tissues in normal numbers. Mature CD4+ T cells from the lpr/lpr mice are resistant to suppression by high doses of antigen and to apoptotic cell death. In vivo administration of peptide antigen causes deletion of thymic T cells in both MRL-lpr/lpr and MRL+/+ strains. By contrast, antigen-induced deletion of peripheral T cells occurs in the MRL+/+ but not in the MRL-lpr/lpr strain. Therefore, the fas gene plays an essential role in activation-induced cell death in mature T lymphocytes, but not in the negative selection of immature cells in the thymus.

Human T-cell repertoire, heterogeneity and memory: relevance to malaria

There is abundant evidence that most non-exposed donors have cells able to respond to malaria antigens. These cells have the properties of memory T cells and probably arise as a result of cross-reactive priming by other microbial antigens. The existence of a high frequency of cross-reactive primed cells would be predicted to bias the response to malaria antigens, both in terms of specificity and the type of effector cells generated. It is possible that this bias may interfere with the development of effective anti-malaria responses.