Potent and selective stimulation of memory-phenotype CD8+ T cells in vivo by IL-15

Proliferation of memory-phenotype (CD44hi) CD8+ cells induced by infectious agents can be mimicked by injection of type I interferon (IFN I) and by IFN I-inducing agents such as lipopolysaccharide and Poly I:C; such proliferation does not affect naive T cells and appears to be TCR independent. Since IFN I inhibits proliferation in vitro, IFN I-induced proliferation of CD8+ cells in vivo presumably occurs indirectly through production of secondary cytokines, e.g., interleukin-2 (IL-2) or IL-15. We show here that, unlike IL-2, IL-15 closely mimics the effects of IFN I in causing strong and selective stimulation of memory-phenotype CD44hi CD8+ (but not CD4+) cells in vivo; similar specificity applies to purified T cells in vitro and correlates with much higher expression of IL-2Rbeta on CD8+ cells than on CD4+ cells.

Induction of bystander T cell proliferation by viruses and type I interferon in vivo

T cell proliferation in vivo is presumed to reflect a T cell receptor (TCR)-mediated polyclonal response directed to various environmental antigens. However, the massive proliferation of T cells seen in viral infections is suggestive of a bystander reaction driven by cytokines instead of the TCR. In mice, T cell proliferation in viral infections preferentially affected the CD44hi subset of CD8+ cells and was mimicked by injection of polyinosinic-polycytidylic acid [poly(I:C)], an inducer of type I interferon (IFN I), and also by purified IFN I; such proliferation was not associated with up-regulation of CD69 or CD25 expression, which implies that TCR signaling was not involved. IFN I [poly(I:C)]-stimulated CD8+ cells survived for prolonged periods in vivo and displayed the same phenotype as did long-lived antigen-specific CD8+ cells. IFN I also potentiated the clonal expansion and survival of CD8+ cells responding to specific antigen. Production of IFN I may thus play an important role in the generation and maintenance of specific memory.

Turnover of naive- and memory-phenotype T cells

On the basis of their surface markers, T lymphocytes are divided into subsets of "naive" and "memory cells". We have defined the interrelationship and relative life spans of naive and memory T cells by examining the surface markers on murine T cells incorporating bromodeoxyuridine, a DNA precursor, given in the drinking water. Three findings are reported. First, using a new method we show that the release of newly formed naive T cells from the unmanipulated thymus is very low (confirming the findings of others with surgical approaches). Second, in thymectomized mice, T cells with a naive phenotype remain in interphase for prolonged periods; however, some of these cells divide and retain (or regain) their "naive" markers. Third, most T cells with a memory phenotype divide rapidly, but others remain in interphase for many weeks. Collectively, the data indicate that long-lived T cells have multiple phenotypes and contain a mixture of memory cells, naive (virgin) cells, and memory cells masquerading as naive cells.

Type i interferons potently enhance humoral immunity and can promote isotype switching by stimulating dendritic cells in vivo

Type I interferons (IFN-I) are rapidly induced following infection and play a key role in nonspecific inhibition of virus replication. Here we have investigated the effects of IFN-I on the generation of antigen-specific antibody responses. The data show that IFN-I potently enhance the primary antibody response to a soluble protein, stimulating the production of all subclasses of IgG, and induce long-lived antibody production and immunological memory. In addition, endogenous production of IFN-I was shown to be essential for the adjuvant activity of CFA. Finally, IFN-I enhanced the antibody response and induced isotype switching when dendritic cells were the only cell type responding to IFN-I. The data reveal the potent adjuvant activity of IFN-I and their important role in linking innate and adaptive immunity.

IL-15 is expressed by dendritic cells in response to type I IFN, double-stranded RNA, or lipopolysaccharide and promotes dendritic cell activation

Cytokines that are induced by infection may contribute to the initiation of immune responses through their ability to stimulate dendritic cells (DCs). In this paper, we have addressed the role of IL-15 in DC activation, investigating its expression by DCs in response to three different signals of infection and examining its ability to stimulate DCs. We report that the expression of both IL-15 and the IL-15 receptor alpha-chain are increased in splenic DCs from mice inoculated with dsRNA (poly(I:C)), LPS, or IFN-alphabeta, and in purified murine splenic DCs treated with IFN-alphabeta in vitro. Furthermore, IL-15 itself was able to activate DCs, as in vivo or in vitro exposure of splenic DCs to IL-15 resulted in an up-regulation of costimulatory molecules, markedly increased production of IFN-gamma by DC and an enhanced ability of DCs to stimulate Ag-specific CD8(+) T cell proliferation. The magnitude of all of the IL-15-induced changes in DCs was reduced in mice deficient for the IFN-alphabeta receptor, suggesting a role for IFN-alphabeta in the stimulation of DCs by IL-15. These results identify IL-15 as a stimulatory cytokine for DCs with the potential for autocrine activity and link its effects to expression of IFN-alphabeta.

The development, maturation, and turnover rate of mouse spleen dendritic cell populations

Three distinct subtypes of dendritic cells (DC) are present in mouse spleen, separable as CD4(-)8alpha(-), CD4(+)8alpha(-), and CD4(-)8alpha(+) DC. We have tested whether these represent stages of development or activation within one DC lineage, or whether they represent separate DC lineages. All three DC subtypes appear relatively mature by many criteria, but all retain a capacity to phagocytose particulate material in vivo. Although further maturation or activation could be induced by bacterially derived stimuli, phagocytic capacity was retained, and no DC subtype was converted to the other. Continuous elimination of CD4(+)8(-) DC by Ab depletion had no effect on the levels of the other DC subtypes. Bromodeoxyuridine labeling experiments indicated that all three DC subtypes have a rapid turnover (half-life, 1.5-2.9 days) in the spleen, with none being the precursor of another. The three DC subtypes showed different kinetics of development from bone marrow precursors. The CD8alpha(+) spleen DC, apparently the most mature, displayed an extremely rapid turnover based on bromodeoxyuridine uptake and the fastest generation from bone marrow precursors. In conclusion, the three splenic DC subtypes behave as rapidly turning over products of three independent developmental streams.

Lymphocyte life-span and memory

Differentiation of immature T and B cells in the primary lymphoid organs gives rise to a pool of long-lived lymphocytes that recirculate through the secondary lymphoid tissues. On the basis of their surface markers, T and B cells comprise a mixture of naïve and memory cells with differing life-spans. Immunization (and vaccination) causes naïve lymphocytes to proliferate and differentiate into effector cells and memory cells. Whether the survival of memory cells is innate or requires persistent contact with residual antigen is controversial. Resolving this issue may be crucial for designing optimal vaccines.

Type I interferon-mediated stimulation of T cells by CpG DNA

Immunostimulatory DNA and oligodeoxynucleotides containing unmethylated CpG motifs (CpG DNA) are strongly stimulatory for B cells and antigen-presenting cells (APCs). We report here that, as manifested by CD69 and B7-2 upregulation, CpG DNA also induces partial activation of T cells, including naive-phenotype T cells, both in vivo and in vitro. Under in vitro conditions, CpG DNA caused activation of T cells in spleen cell suspensions but failed to stimulate highly purified T cells unless these cells were supplemented with APCs. Three lines of evidence suggested that APC-dependent stimulation of T cells by CpG DNA was mediated by type I interferons (IFN-I). First, T cell activation by CpG DNA was undetectable in IFN-IR-/- mice. Second, in contrast to normal T cells, the failure of purified IFN-IR-/- T cells to respond to CpG DNA could not be overcome by adding normal IFN-IR+ APCs. Third, IFN-I (but not IFN-gamma) caused the same pattern of partial T cell activation as CpG DNA. Significantly, T cell activation by IFN-I was APC independent. Thus, CpG DNA appeared to stimulate T cells by inducing APCs to synthesize IFN-I, which then acted directly on T cells via IFN-IR. Functional studies suggested that activation of T cells by IFN-I was inhibitory. Thus, exposing normal (but not IFN-IR-/-) T cells to CpG DNA in vivo led to reduced T proliferative responses after TCR ligation in vitro.

T cell stimulation in vivo by lipopolysaccharide (LPS)

Lipopolysaccharide (LPS) from gram-negative bacteria causes polyclonal activation of B cells and stimulation of macrophages and other APC. We show here that, under in vivo conditions, LPS also induces strong stimulation of T cells. As manifested by CD69 upregulation, LPS injection stimulates both CD4 and CD8(+) T cells, and, at high doses, stimulates naive (CD44(lo)) cells as well as memory (CD44(hi)) cells. However, in terms of cell division, the response of T cells after LPS injection is limited to the CD44(hi) subset of CD8(+) cells. In contrast with B cells, proliferative responses of CD44(hi) CD8(+) cells require only very low doses of LPS (10 ng). Based on studies with LPS-nonresponder and gene-knockout mice, LPS-induced proliferation of CD44(hi) CD8(+) cells appears to operate via an indirect pathway involving LPS stimulation of APC and release of type I (alpha, beta) interferon (IFN-I). Similar selective stimulation of CD44(hi) CD8(+) cells occurs in viral infections and after injection of IFN-I, implying a common mechanism. Hence, intermittent exposure to pathogens (gram-negative bacteria and viruses) could contribute to the high background proliferation of memory-phenotype CD8(+) cells found in normal animals.

T cell death and memory

In typical immune responses, contact with antigen causes naive T cells to proliferate and differentiate into effector cells. After the pathogen is destroyed, most effector T cells are eliminated-thereby preserving the primary T cell repertoire-but some cells survive and form long-lived memory cells. During each stage of this process, the life or death fate of T cells is strictly regulated.

CD40 ligand-mediated interactions are involved in the generation of memory CD8(+) cytotoxic T lymphocytes (CTL) but are not required for the maintenance of CTL memory following virus infection

CD8(+) cytotoxic T lymphocytes (CTL) play a key role in the control of many virus infections, and the need for vaccines to elicit strong CD8(+) T-cell responses in order to provide optimal protection in such infections is increasingly apparent. However, the mechanisms involved in the induction and maintenance of CD8(+) CTL memory are currently poorly understood. In this study, we investigated the involvement of CD40 ligand (CD40L)-mediated interactions in these processes by analyzing the memory CTL response of CD40L-deficient mice following infection with lymphocytic choriomeningitis virus (LCMV). The maintenance of memory CD8(+) CTL precursors (CTLp) at stable frequencies over time was not impaired in CD40L-deficient mice. By contrast, the initial generation of memory CTLp was affected. CD40L-deficient mice produced lower levels of CD8(+) CTLp during the primary immune response to LCMV than did wild-type controls, despite the fact that the LCMV-specific effector CTL response of CD40L-deficient mice was indistinguishable from that of control animals. The differentiation of naïve CD8(+) T cells into effector and memory CTL thus involves pathways that can be discriminated from each other by their requirement for CD40L-mediated interactions. Expression of CD40L by CTLp themselves was not an essential step during their expansion and differentiation from naïve CD8(+) cells into memory CTLp; instead, the reduction in memory CTLp generation in CD40L-deficient mice was likely a consequence of defects in the CD4(+) T-cell response mounted by these animals. These results thus suggest a previously unappreciated role for CD40L in the generation of CD8(+) memory CTLp, the probable nature of which is discussed.

An IFN-gamma-dependent pathway controls stimulation of memory phenotype CD8+ T cell turnover in vivo by IL-12, IL-18, and IFN-gamma

Unlike naive T cells, memory phenotype (CD44(high)) T cells exhibit a high background rate of turnover in vivo. Previous studies showed that the turnover of memory phenotype CD8(+) (but not CD4(+)) cells in vivo can be considerably enhanced by products of infectious agents such as LPS. Such stimulation is TCR independent and hinges on the release of type I IFNs (IFN-I) which leads to the production of an effector cytokine, probably IL-15. In this study, we describe a second pathway of CD44(high) CD8(+) stimulation in vivo. This pathway is IFN-gamma rather than IFN-I dependent and is mediated by at least three cytokines, IL-12, IL-18, and IFN-gamma. As for IFN-I, these three cytokines are nonstimulatory for purified T cells and under in vivo conditions probably act via production of IL-15.

Lifespan of gamma/delta T cells

Information on the turnover and lifespan of murine gamma/delta cells was obtained by administering the DNA precursor, bromodeoxyuridine (BrdU), in the drinking water and staining lymphoid cells for BrdU incorporation. For TCR-gamma/delta (Vgamma2) transgenic mice, nearly all gamma/delta thymocytes became BrdU+ within 2 d and were released rapidly into the peripheral lymphoid tissues. These recent thymic emigrants (RTEs) underwent phenotypic maturation in the periphery for several days, but most of these cells died within 4 wk. In adult thymectomized (ATx) transgenic mice, only a small proportion of gamma/delta cells survived as long-lived cells; most of these cells had a slow turnover and retained a naive phenotype. As in transgenic mice, the majority of RTEs generated in normal mice (C57BL/6) appeared to have a restricted lifespan as naive cells. However, in marked contrast to TCR transgenic mice, most of the gamma/delta cells surviving in ATx normal mice had a rapid turnover and displayed an activated/memory phenotype, implying a chronic response to environmental antigens. Hence, in normal mice many gamma/delta RTEs did not die but switched to memory cells.

Stimulation of naïve and memory T cells by cytokines

On the basis of cell surface markers, mature T cells are considered to have either a naïve or a memory phenotype. These cells exhibit distinct types of kinetic behaviour in vivo. While naïve-phenotype cells persist long term in a non-dividing state, memory-phenotype T cells include cycling cells and exhibit a more rapid rate of turnover; this has also been shown to be true for cells that can be definitively identified as naïve or memory T cells respectively. The number of memory-phenotype (CD44hi) CD8+ T cells entering cell cycle is greatly increased after in vivo exposure to viruses, bacteria or components of bacteria. Accelerated turnover of memory T cells also occurs after the injection of a variety cytokines that are induced by infectious agents, including type I interferon (IFN-I). Although naïve-phenotype T cells do not divide in response to these cytokines, they do exhibit signs of activation, including upregulation of CD69 after exposure to IFN-I. These findings suggest that the dissimilar in vivo kinetics of naïve- and memory-phenotype T cells might reflect their divergent responses to cytokines. Furthermore, the ability of infection-induced cytokines to stimulate non-specific proliferation of memory-phenotype T cells and partial activation of naïve-phenotype T cells implies that they play a complex role during primary immune responses to infectious agents.

Lifespan of lymphocytes

T and B lymphocytes comprise heterogeneous populations of cells at various stages of differentiation and activation. T- and B-cell subsets have different roles in the maintenance of immune homeostasis, and their functional differences are reflected by their respective lifespans. This review briefly summarizes the available data on lymphocyte lifespan, including the kinetics of T- and B-cell development in the primary lymphoid organs and the proliferative behavior of naive, effector and memory lymphocytes in the peripheral lymphoid compartment.

Factors controlling the turnover of T memory cells

Most of the T cells participating in the primary immune response are rapidly eliminated, but small numbers of these cells survive and differentiate into long-lived memory cells. Information on the life history of memory cells can be obtained by studying the component of memory-phenotype T cells found in normal animals; these cells are presumed to represent memory cells specific for various environmental antigens. For CD8+ cells, in vivo exposure to viruses and certain other infectious agents causes a large proportion of memory-phenotype (CD44hi) cells to enter the cell cycle. In this situation, stimulation of CD44hi CD8+ cells does not seem to require T-cell receptor ligation and appears to reflect release of various cytokines, especially type I interferon. The capacity of infectious agents to induce non-antigen-specific stimulation of T cells may play a role in boosting the survival of memory cells and perhaps also in providing an adjuvant function during the primary response.

Life span of naive and memory T cells

The life span of mature T cells is reviewed. Peripheral T lymphocytes are a heterogeneous population and comprise a mixture of naive, effector and memory cells. The recirculating pool of mature T cells is formed during young life through gradual release of naive T cells from the thymus. In adults, the pool of mature T cells is relatively self-sufficient, and input of new T cells from the thymus declines to low levels. Studies on T cell turnover indicate that most peripheral T cells can remain in a resting state for long periods (months in rodents and years in humans). Examination of the phenotype of dividing versus nondividing cells suggests that typical naive T cells are long-lived resting cells whereas the majority of effector and memory T cells have a much more rapid turnover. However, some memory T cells appear to divide very infrequently and eventually return to a resting state. The factors controlling the generation and maintenance of memory T cells are discussed.

T-cell proliferation in vivo and the role of cytokines

Unlike typical naive T cells, T cells with an activated (CD44hi) memory phenotype show a rapid rate of proliferation in vivo. The turnover of memory-phenotype CD8+ T cells can be considerably augmented by injecting mice with various compounds, including polyinosinic polycytidylic acid, lipopolysaccharide and immunostimulatory DNA (CpG DNA). Certain cytokines, notably type I (alpha, beta) interferons (IFN-I), have a similar effect. These agents appear to induce proliferation of CD44hi CD8+ cells in vivo by an indirect process involving production of effector cytokines, possibly interleukin-15, by antigen-presenting cells. Although none of the agents tested induces proliferation of naive-phenotype T cells, IFN-I has the capacity to cause upregulation of surface markers on purified naive T cells. Depending upon the experimental conditions used, IFN-I can either inhibit or enhance primary responses of naive T cells to specific antigen.

T-cell turnover in vivo and the role of cytokines

In addition to responding to specific antigen, CD8+ T-cells with a memory (CD44hi) phenotype undergo bystander proliferation when exposed to certain cytokines, notably type I interferons (IFN I), in vivo; such proliferation does not require T-cell receptor ligation. Since IFN I is unable to induce proliferation of purified CD44hi CD8+ cells in vitro, stimulation of these cells in vivo may reflect IFN I-dependent release of other cytokines. Evidence is presented that IFN I induces macrophages to synthesize IL-15 mRNA and that, at the protein level, IL-15 causes selective stimulation of CD44hi CD8+ (but not CD4+) cells, both in vitro and in vivo. This finding raises the possibility that synthesis of IL-15 during infection may induce bystander proliferation of memory-phenotype CD8+ cells.

Stimulation of memory T cells by cytokines

Mature T cells can be classified on the basis of cell surface markers into naïve- and memory-phenotype cells. These phenotypically-defined subsets exhibit distinct kinetic behaviour in vivo. Thus, naïve-phenotype T cells persist long-term in a non-dividing state, while memory-phenotype T cells include cycling cells and have a more rapid rate of turnover. We have investigated the possibility that the different kinetic behaviour of naïve- and memory-phenotype T cells reflects a differential responsiveness to cytokines. It was discovered that memory-, but not naïve-, phenotype T cells were stimulated to proliferate by a variety of infection-induced cytokines. These results suggest that cytokines contribute to the high background rate of turnover exhibited by memory T cells.

Bystander stimulation of T cells in vivo by cytokines

Immune responses to infectious agents, especially viruses, are often associated with extensive proliferation of T cells and transient enlargement of the lymphoid tissues. Since the precursor frequency of T cells for specific antigen is low, the bulk of the T cells proliferating in the primary response are presumably stimulated via non-antigen-specific mechanisms, e.g. via cytokines elicited by the infectious agent concerned. Such 'bystander' stimulation of T cells occurs in mice injected with agents that elicit production of type I interferon (IFN I). Induction of IFN I in vivo causes marked stimulation of the CD44hi subset of CD8+ T cells and is prominent after injection of live viruses or products of bacteria such as lipopolysaccharide. Cytokines elicited by infectious agents may act as adjuvants during the primary response and could serve to boost the survival of long-lived memory cells.

Multiple effects of immunostimulatory DNA on T cells and the role of type I interferons

In addition to stimulating antigen-specific immune responses, infectious agents cause nonspecific activation of the innate immune system, notably up-regulation of costimulatory/adhesion molecules on APCs and cytokine production. In recent years it has become apparent that stimulation of the immune system by microorganisms is a property of a number of different cellular components, including DNA. As discussed earlier and elsewhere in this volume, the DNA of infectious agents--and indeed of all non-vertebrates tested--differs from mammalian DNA in being enriched for unmethylated CpG motifs. With appropriate flanking sequences, CpG DNA and synthetic CpG ODNs cause strong activation of APCs and other cells. In this article we have focussed on the capacity of CpG DNA/ODNs to alter T cell function. Whether these compounds act directly on T cells or function indirectly by activating other cells, especially APCs, is controversial [7, 8, 13, 14]. In contrast to other workers [8], we have yet to find definitive evidence that CpG DNA/ODNs can provide a co-stimulatory signal for purified T cells subjected to TCR ligation ([14] and unpublished data of authors). For this reason we lean to the notion that CpG DNA/ODNs modulate T cell function by inducing activation of APC rather than by acting directly on T cells. When injected in vivo in the absence of specific antigen, CpG DNA/ODNs have two striking effects on T cells, namely (1) induction of overt activation (proliferation) of memory-phenotype CD8+ cells, and (2) partial activation of all T cells, including naïve-phenotype T cells. Both actions of CpG DNA/ODNs are heavily dependent on the production of IFN-I by APC. For memory-phenotype (CD44hi) CD8+ cells, neither CpG DNA nor IFN-I can cause proliferation of purified APC-depleted T cells in vitro. Hence, under in vivo conditions, CpG DNA-induced proliferation of CD44hi CD8+ cells is probably mediated through the production of a secondary cytokine, i.e., by a cytokine that is directly stimulatory for CD44hi CD8+ cells. Based on the available evidence, it is highly likely that the effector cytokine is IL-15. With this assumption, our current model is that proliferation of CD44hi CD8+ cells induced by injection of CpG DNA/ODNs reflects production of IFN-I which, in turn, leads to synthesis of IL-15. Which particular cell types produce these two cytokines is unclear, although APCs are probably of prime importance. In addition to inducing proliferation of memory-phenotype CD8+ cells via IL-15, the IFN-I induced by CpG DNA/ODNs can also induce partial activation of naive T cells. This form of activation leads to up-regulation of CD69 and other molecules but does not cause entry into cell cycle. It is of interest that the partial activation of naive T cells induced by IFN-I is associated with decreased T proliferative responses. Thus, proliferation of purified naïve T cells elicited by combined TCR/CD28 ligation in vitro is greatly reduced by addition of IFN-I. This inhibitory effect of IFN-I does not influence cytokine production and probably reflects production of cell cycle inhibitors. Surprisingly, except at high doses, IFN-I fails to exert an anti-proliferative effect when T proliferative responses are driven by viable APCs. Indeed, in this situation, IFN-I enhances antigen-specific T proliferative responses, both in vivo and in vitro. This adjuvant effect of IFN-I is presumably a reflection of APC activation, but direct evidence on this issue is still lacking. In this article we have emphasized that contact with CpG DNA/ODNs has multiple effects on T cell function in vivo. Many of these effects seem to be related to the production of certain cytokines by APCs, notably IFN-I and IL-15. It should be stressed, however, that CpG DNA/ODNs probably lead to the production of many other cytokines. Hence, our current models of how CpG DNA/ODNs influence T cell function are undoubtedly oversimplified.

Anti-viral immunity: spotting virus-specific T cells

Historically, quantitation of virus-specific CD8+ T cells has been accomplished by limiting dilution analysis of cytotoxic precursor cells. Recent studies have shown that this technique greatly underestimates the actual number of antigen-specific cells and have provided new insight into anti-viral immune responses.