Programmed cell death of T lymphocytes during acute viral infection: a mechanism for virus-induced immune deficiency.

Acute viral infections induce immune deficiencies, as shown by unresponsiveness to mitogens and unrelated antigens. T lymphocytes isolated from mice acutely infected with lymphocytic choriomeningitis virus (LCMV) were found in this study to undergo activation-induced apoptosis upon signalling through the T-cell receptor (TcR)-CD3 complex. Kinetic studies demonstrated that this sensitivity to apoptosis directly correlated with the induction of immune deficiency, as measured by impaired proliferation in response to anti-CD3 antibody or to concanavalin A. Cell cycling in interleukin-2 (IL-2) alone stimulated proliferation of LCMV-induced T cells without inducing apoptosis, but preculturing of T cells from acutely infected mice in IL-2 accelerated apoptosis upon subsequent TcR-CD3 cross-linking. T lymphocytes isolated from mice after the acute infection were less responsive to IL-2, but those T cells, presumably memory T cells, responding to IL-2 were primed in each case to die a rapid apoptotic death upon TcR-CD3 cross-linking. These results indicate that virus infection-induced unresponsiveness to T-cell mitogens is due to apoptosis of the activated lymphocytes and suggest that the sensitization of memory cells by IL-2 induced during infection will cause them to die upon antigen recognition, thereby impairing specific responses to nonviral antigens.

In vivo state of antiviral CTL precursors. Characterization of a cycling cell population containing CTL precursors in immune mice.

The in vivo state of CD8+ mouse memory CTL specific to lymphocytic choriomeningitis virus (LCMV) was characterized. During acute LCMV infection, the majority of the LCMV-specific CTL activity tested immediately ex vivo was mediated by CD8+ L-selectin- Mac-1+ CTL. The L-selectin- population of CD8+ cells elicited during acute infection also carried > 99% of the restimulatable CD8+ CTL precursors (CTLp) to LCMV, and these required added IL-2 for development into effectors in vitro. In contrast with the acute infection, most of the virus-specific CTLp in immune mice were L-selectin+. Examination of CD8+ T cells in LCMV-immune mice revealed that a L-selectin+ blast-size population of cycling CD8+ cells contained CTLp, which developed into effector CTL in the absence of added IL-2. These cells also expressed Mac-1 and IL-2R. Flow cytometric sorting for IL-2R+ and IL-2R- CD8+ cells in the immune animal revealed, by limiting dilution analysis, similar frequencies of CTLp in both populations. In bulk restimulation assays, the CD25+ CTLp did not require added IL-2 for their in vitro development into effectors, whereas the CD25- CTLp did. Hence, the different requirements for CTLp to effector development in vitro reflect qualitative differences in the in vivo state of the CTLp in the various subpopulations. LCMV-specific memory CTLp that did not require added IL-2 for differentiation were also found in the small-size, noncycling, CD8+L-selectin- cells. In contrast, the small-size, noncycling, CD8+L-selectin+, and CD8+IL-2R- populations also carried CTLp, but these required added IL-2 for development into effector CTL. Hence, T cell memory to LCMV is distributed among various lymphocyte subpopulations in immune animals, and the presence of an activated cycling cell component may account for the long-term perpetuation of antiviral immunologic memory.

T-lymphocyte downregulation after acute viral infection is not dependent on CD95 (Fas) receptor-ligand interactions.

Infection of mice with lymphocytic choriomeningitis virus (LCMV) causes a major expansion of CD8+ T cells followed by a period of immune downregulation that coincides with the induction of lymphocyte apoptosis in the mouse spleen. CD95 (Fas) and its ligand are important for regulating peripheral T-lymphocyte numbers and can mediate apoptosis of mature T lymphocytes. We infected CD95- and CD95L-deficient mice (lpr and gld, respectively) with LCMV to determine if the immune downregulation that occurred following resolution of the LCMV infection was due to a CD95-dependent apoptotic mechanism. Lymphocytes from LCMV-infected lpr and gld mice were capable of normal T-cell expansion and cytolytic function but were, in contrast to activated cells from normal virus-infected mice, relatively more resistant to T-cell receptor-induced apoptosis in vitro. However, in vivo there were significant numbers of apoptotic cells in the spleens of lpr and gld mice recovering from the infection, and the T-cell number and cytolytic activity decreased to normal postinfection levels. Thus, CD95 is not required for the immune downregulation of the CD8+-T-lymphocyte response following acute LCMV infection.

Role of apoptosis in the regulation of virus-induced T cell responses, immune suppression, and memory.

Apoptosis is an important mechanism enabling the selection of the non-self-reactive T cell repertoire and for maintaining homeostasis in the immune system after it has expanded to combat infections. Highly activated, proliferating T cells become susceptible to apoptosis driven by a number of stimuli, and T cells activated during a viral infection become susceptible to "activation induced cell death" after repeated stimulation through the T cell receptor (TcR). This is a major mechanism for the immune deficiencies observed during many viral infections. During infections with a high antigen load this can lead to a selective deletion of virus-specific cytotoxic T lymphocytes (CTL) and to the establishment of persistent infection. More commonly, the CTL control the infection first, and high levels of apoptosis in the expanded lymphocyte population occur after antigen and growth factors become limiting. This cell death does not seem to depend on TcR specificity, as the residual population contains a remarkably stable population of memory CTL precursors that approximate the frequency per CD8 cell of that seen during the peak of the acute infection. Subsequent infections with heterologous viruses result in an expansion and then an apoptotic elimination of T cells, with the consequence being a reduction in precursor CTL specific for the first virus. Thus, apoptosis shapes the quality and quantity of T cell memory.

Lymphocyte apoptosis during the silencing of the immune response to acute viral infections in normal, lpr, and Bcl-2-transgenic mice.

This report examines the mechanisms involved in the down-regulation of the immune response in acute viral infection and documents the presence of apoptotic lymphocytes in situ in the spleens of mice during the resolution of the immune response to acute lymphocytic choriomeningitis virus infection. Apoptotic cells were detected by an in situ nucleotidyl transferase assay. Both T and B lymphocytes were shown to be dying in vivo, the latter in clusters. A biphasic occurrence of apoptosis during the course of the acute infection was observed, with elevated levels occurring at day 3 after infection and a second more pronounced peak at day 11 after infection, coincident with the decline of the cytotoxic T lymphocyte response and with the decrease in total splenic leukocyte number. Apoptosis in vivo was detected in lpr mice, suggesting that Fas expression is not imperative for lymphocyte apoptosis in the context of an acute viral infection. Apoptosis in situ and the decline of the T lymphocyte response to acute lymphocytic choriomeningitis virus infection was unaffected by the enforced lymphocyte-directed expression of Bcl-2, a protein that blocks growth factor deprivation-induced apoptosis of lymphocytes in vitro. These results argue that the silencing of the T cell response to acute infection may not be a result simply of growth factor deprivation. The susceptibility of activated T cells to apoptotic death, which has previously been associated with virus-induced immune deficiency, may therefore also explain the en masse elimination of the expanded lymphocyte pool subsequent to an acute viral infection.