Mouse spleens contain three populations of conventional (CD11c high ) dendritic cells (DCs) that play distinct functions. The CD8 ؉ DC are unique in that they can present exogenous antigens on their MHC class I molecules, a process known as cross-presentation. It is unclear whether this special ability is because only the CD8 ؉ DC can capture the antigens used in cross-presentation assays, or because this is the only DC population that possesses specialized machinery for cross-presentation. To solve this important question we examined the splenic DC subsets for their ability to both present via MHC class II molecules and cross-present via MHC class I using four different forms of the model antigen ovalbumin (OVA). These forms include a cell-associated form, a soluble form, OVA expressed in bacteria, or OVA bound to latex beads. With the exception of bacterial antigen, which was poorly cross-presented by all DC, all antigenic forms were cross-presented much more efficiently by the CD8 ؉ DC. This pattern could not be attributed simply to a difference in antigen capture because all DC subsets presented the antigen via MHC class II. Indeed, direct assessments of endocytosis showed that CD8 ؉ and CD8 ؊ DC captured comparable amounts of soluble and bead-associated antigen, yet only the CD8 ؉ DC cross-presented these antigenic forms. Our results indicate that cross-presentation requires specialized machinery that is expressed by CD8 ؉ DC but largely absent from CD8 ؊ DC. This conclusion has important implications for the design of vaccination strategies based on antigen targeting to DC.antigen presentation ͉ mice ͉ endocytosis ͉ ovalbumin ͉ vaccines D endritic cells (DC) possess several mechanisms that make them highly efficient antigen-presenting cells. DC can endocytose a large variety of exogenous antigens for presentation via MHC class II molecules and for cross-presentation via MHC class I (1). The cross-presenting capacity of DC is unusual, because most other cell types are only able to present endogenous antigens (i.e., antigens synthesized by the antigenpresenting cells themselves) on their MHC class I molecules. Thus, DC possess specialized machinery, as yet not fully defined, that allows delivery of exogenous antigens into the MHC class I presentation pathway for cross-presentation (2, 3).On the other hand, several populations of DC have been described (4, 5), and it is unclear whether all these DC types can cross-present (6). Mouse spleens contain three ''conventional'' (CD11c high ) DC subsets: CD8 ϩ DC, CD4 ϩ DC, and CD4 Ϫ CD8 Ϫ [double negative (DN)] DC. Previous studies have shown that cell-associated antigen was cross-presented by CD8 ϩ but not CD8 Ϫ DC (7-10). The unique capacity of the CD8 ϩ DC at cross-presenting this form of antigen was attributed to their ability to capture dead cells (7-9, 11, 12) because antigens in soluble or immunocomplexed form, or associated to bacteria, were reportedly cross-presented by both CD8 ϩ and CD8 Ϫ DC (8, 9, 13). These findings led to the hypothesis that all DC can...

Of the many dendritic cell (DC) subsets, DCs expressing the monomorphic coreceptor CD8 alpha-chain (CD8alpha) are localized permanently in lymphoid organs, whereas 'tissue-derived DCs' remain in nonlymphoid tissues until they 'capture' antigen and then move to local lymph nodes. Here we show that after lung infection, both naive and memory CD8+ 'killer' T cells responded to influenza virus antigens presented by lymph node-resident CD8alpha+ DCs, but only naive cells responded to antigens presented by lung-derived DCs. This difference provides a mechanism for priming naive T cell responses in conditions in which robust memory predominates. Our findings have implications for immunity to pathogens that can mutate their T cell epitopes, such as influenza virus and human immunodeficiency virus, and challenge the long-held view that memory T cells have less-stringent requirements for activation than naive T cells have.

Antigen presentation within the lymph node draining a site of infection is crucial for initiation of cytotoxic T cell responses. Precisely how this antigen presentation regulates T cell expansion in vivo is unclear. Here, we show that, in primary infection, antigen presentation peaks Ϸ3 days postinfection and then slowly decays until day 12. This prolonged antigen presentation is required for optimal expansion of naive CD8 ؉ T cells, because early ablation of dendritic cells reduces the later CD8 ؉ T cell response. Antigen presentation during secondary infection was 10-fold lower in magnitude and largely terminated by day 4 postinfection. Expansion of memory, but not naive, antigen-specific T cells was tightly controlled by perforin-dependent cytolysis of antigen-presenting cells. The ability of the memory T cells to remove antigenpresenting cells provides a negative-feedback loop to directly limit the duration of antigen presentation in vivo.dendritic cell ͉ immunological memory ͉ influenza virus ͉ T lymphocyte ͉ immunity P rotective immunity against pathogens depends on amplification of rare naive antigen-specific T cells that survive for long periods after initial exposure to the antigen. Antigenspecific naive T cells undergo more than a 1,000-fold expansion during the primary response, but 90-95% of the effector cells are rapidly eliminated by apoptosis, leaving a small number of surviving cells that form the stable memory pool (1-3). These few remaining memory cells are then poised to rapidly respond to a second encounter with the antigen, allowing massive reamplification of individual T cell populations. Despite this, memory T cell expansion also has a size constraint, and the immune system is obliged to sacrifice preexisting memory T cells to accommodate new ones.Part of the solution to regulation of the size of T cell populations lies in the rapid expansion and contraction of antigen-specific populations during a response. This is driven by the magnitude and duration of antigen presentation, but how antigen-presenting cell (APC) populations are themselves regulated in vivo to ensure optimal modulation of the size of T cell populations after primary and secondary infection is less clear (4, 5). One major paradigm, based largely on the coincident emergence of effector cytotoxic CD8 ϩ T cells (CTLs) with the quantitative reduction in antigen presentation, suggests that a regulatory feedback network exists whereby APCs are eliminated by responding T cells, thereby limiting the duration of stimulation (5-7). Restricting T cell activation and amplification would be an essential checkpoint in preventing chronic lymphocyte activation and associated immunopathology (8). However, this mechanism may not universally apply in all immune responses. Ablation of antigen presentation at various time points during primary Listeria monocytogenes and herpes simplex-1 infection has highlighted that more prolonged antigen presentation can positively influence the size of the T cell expansion (4, 9, 10). Interestingly, APCs th...

Summary Although the pro-apoptotic BH3-only protein, Bim, is required for deletion of autoreactive thymocytes, Bim-deficient mice do not succumb to extensive organ-specific autoimmune disease. To determine whether other BH3-only proteins safeguard tolerance in the absence of Bim, we screened mice lacking Bim alongside other BH3-only proteins. Most strains showed no additional defects, however, mice deficient for both Puma and Bim spontaneously developed autoimmunity in multiple organs and their T-cells could transfer organ-specific autoimmunity. Puma/Bim double-deficient mice had a striking accumulation of mature single positive thymocytes, suggesting a further defect in thymic deletion was the basis for disease. Transgenic mouse models of thymocyte deletion to peripheral neo-antigens confirmed that the loss of Bim and Puma allowed increased numbers of autoreactive thymocytes to escape deletion. Our data show that Puma cooperates with Bim to impose a thymic deletion checkpoint to peripheral self-antigens and cement the notion that defects in apoptosis alone are sufficient to cause autoimmune disease.

Dendritic cells (DCs) play major roles in immunosurveillance. In peripheral tissues, 'immature' DCs are dedicated to capturing antigens. Detection of pathogens through Toll-like receptors (TLRs) triggers DC migration to the lymph nodes (LNs), where they acquire a 'mature' phenotype specialized at presenting antigens. However, DCs migrate from tissues and mature even in the absence of overt infections. This has been attributed to detection of commensal flora in the skin, the gut or other peripheral tissues in the steady state. To test this assumption, we have analyzed the DCs contained in the lymphoid organs of germ-free mice and of mice lacking the TLR adapter molecules, MyD88 and TRIF. We show that the proportion and expression of maturation markers in DC immigrants in the LNs of these mice are similar to those in normal mice. These results suggest that DC migration from tissues, followed by their phenotypic maturation, is regulated in the steady state by an inherent program of DC differentiation or by the release of low levels of inflammatory signals from normal tissues.