Immunization with radiation (γ)-attenuated Plasmodia sporozoites (γ-spz) confers sterile and long-lasting immunity against malaria liver-stage infection. In the P. berghei γ-spz model, protection is linked to liver CD8+ T cells that express an effector/memory (TEM) phenotype, (CD44hiCD45RBloCD62Llo ), and produce IFN-γ. However, neither the antigen presenting cells (APC) that activate these CD8+ TEM cells nor the site of their induction have been fully investigated. Because conventional (c)CD8α+ DC (a subset of CD11c+ DC) are considered the major inducers of CD8+ T cells, in this study we focused primarily on cCD8α+ DC from livers of mice immunized with Pb γ-spz and asked whether the cCD8α+ DC might be involved in the activation of CD8+ TEM cells. We demonstrate that multiple exposures of mice to Pb γ-spz lead to a progressive and nearly concurrent accumulation in the liver but not the spleen of both the CD11c+NK1.1− DC and CD8+ TEM cells. Upon adoptive transfer, liver CD11c+NK1.1− DC from Pb γ-spz-immunized mice induced protective immunity against sporozoite challenge. Moreover, in an in vitro system, liver cCD8α+ DC induced naïve CD8+ T cells to express the CD8+ TEM phenotype and to secrete IFN-γ. The in vitro induction of functional CD8+ TEM cells by cCD8α+ DC was inhibited by anti-MHC class I and anti-IL-12 mAbs. These data suggest that liver cCD8α+ DC present liver-stage antigens to activate CD8+ TEM cells, the pre-eminent effectors against pre-erythrocytic malaria. These results provide important implications towards a design of anti-malaria vaccines.
Citation: Jobe O, Donofrio G, Sun G, Liepinsh D, Schwenk R, Krzych U (2009) Immunization with Radiation-Attenuated Plasmodium berghei Sporozoites Induces Liver cCD8α+DC that Activate CD8+T Cells against Liver-Stage Malaria. PLoS ONE 4(4): e5075. doi:10.1371/journal.pone.0005075
Editor: Vasee Moorthy, Initiative for Vaccine Research, Switzerland
Received: November 14, 2008; Accepted: March 2, 2009; Published: April 4, 2009
This is an open-access article distributed under the terms of the Creative Commons Public Domain declaration which stipulates that, once placed in the public domain, this work may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose.
Funding: The study was supported by NIH grant AI43411 (UK) and US Medical Research and Materiel Command. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Dendritic cells (DC) are key conductors in the orchestration of the immune responses to a variety of pathogens . A number of DC subsets have been identified, including conventional (c)CD8α+DC, cCD8α−DC and plasmacytoid (p)DC and these appear to perform unique functional activities. For example, cCD8α+ DC efficiently cross present cell-associated Ags particularly to CD8+ T cells , ; cCD8α− DC usually promote the activation of CD4+ T cells ; and pDC tend to respond to viral infections  and may be involved in the activation of Treg cells . In addition, the distribution of the DC subsets is unique for lymphoid and non-lymphoid organs ,  and this can lead to an organ-specific response often dictated by the tropism of infectious agents. Owing to their prominent role in directing the immune system, DC have been shown to be indispensable for the induction of protective immune responses to a variety of infections  , including malaria that is caused by Plasmodia ssp. infected mosquitoes.
Prime-boost immunizations with radiation-attenuated (γ) Plasmodia sporozoites (γ-spz) constitutes the most effective method for inducing long-lived, sterile protection against the pre-erythrocytic stages (sporozoite and liver) of both rodent and human malaria , . We have reported that Plasmodium berghei (Pb) γ-spz-induced protection against liver-stage infection is mediated primarily by intrahepatic IFN-γ-producing effector/memory CD44hiCD45RBloCD62Llo CD8+T cells (CD8+TEM) . However, the mechanisms of activation of intrahepatic CD8+TEM cells require further investigation. For example, the identity of the APC that process and present liver-stage Ags for the induction of the CD8+TEM cells remains to be elucidated. Several studies conducted in the past have revealed a critical role for DC in the induction of anti-Plasmodia protection against pre-erythrocytic stage Ags. It has been demonstrated that DC-depleted mice fail to develop protection following γ-spz immunization . Bone-marrow derived dendritic cells pulsed with Pb γ-spz or Pb-derived antigen induced antigen-specific CD8+ T cell responses . Similary, adoptive-transfer of DC loaded with Plasmodium yoelii (Py) circumsporozoite (CS) protein peptides into BALB/c mice results in the generation of peptide-specific CD8+ T cells and reduction of parasitemia following infectious challenge . More recently, it has been shown that DC prime Py CS protein specific TCR Tg CD8+ T cells in the LN draining the site of the mosquito bite, however, the CD8+ T cell effector function following infectious sporozoite challenge occurs in the liver and is DC independent . These and other studies comprise a significant body of work on the roles of splenic ,  and bone marrow-derived DC ,  but comparable studies of liver DC have not been carried out, in part because of the paucity of these cells. Nonetheless, liver DC have been shown to play an important role in liver immunity  but their involvement in the induction of protective immunity to the liver stage of malaria remains unknown and, hence, requires investigation. In addition, the relative contributions of the various liver DC subsets also remain to be elucidated. Like infectious spz, Pb γ-spz infect hepatocytes but the parasites arrest at the liver-stage of development, which leads to the apoptosis of the infected hepatocytes . In contrast, infectious sporozoites do not cause infected hepatocytes to undergo apoptosis and are ineffective in inducing protective CD8+ T cells , . Because numerous studies , , ,  have shown that cCD8α+ DC in particular can ingest exogenous cell-associated Ag and cross-present it to CD8+ T cells, it is reasonable to assume, therefore, that γ-spz-induced apoptotic hepatocytes are internalized by hepatic DC  and the available liver-stage Ags are cross-presented to hepatic CD8+ T cells. We asked therefore whether the cCD8α+ subset of liver DC is involved in the induction of the effector CD8+ T cells that mediate protection against Pb sporozoite challenge.
Herein, we present evidence that cCD8α+DC accumulate concurrently with CD8+ TEM cells in the livers of Pb γ-spz-immunized mice. We also show that these cells are able to activate naïve CD8+T cells to express the TEM phenotype (CD44hiCD45RBloCD62Llo). Therefore, we propose that cCD8α+DC are involved in the induction of the liver-stage antigen-specific CD8+ T cells, the key effectors against liver-stage malaria.
The numbers of CD11c+NK1.1− DC are low in the livers of naïve mice, but increase following prime-boost immunizations with Pb γ−spz
Liver CD8+ T cells are the key effectors in protection induced by Pb γ− spz ; however, neither the cellular/molecular processes leading to the induction of these cells nor the sites where this induction occurs have been thoroughly investigated. For example, it is not fully understood whether DC and/or other APC are involved in the presentation of liver-stage Ags for the activation of CD8+ T cells. Because γ-spz enter hepatocytes, where they undergo an aborted development, it is reasonable to assume that Ags responsible for the induction of effector CD8+ T cells are expressed primarily in the liver . Because DC are considered the most efficient professional APC, we presumed, therefore, that liver DC might be excellent APC candidates for presentation of liver-stage Ags to CD8+ T cells.
For the initial experiments we used flow cytometry to identify a population of CD11c+NK1.1− DC in the liver and the spleen according to the gating strategy shown in Fig. 1A and Fig. 2A. To exclude all T cells and in particular CD8+ T cells (expressing CD8αβ dimers) as well as NK cells, intrahepatic mononuclear cells (IHMNC) and spleen cells were stained with anti-CD3/CD8β and anti-CD11c mAbs and the CD11c+CD3−CD8β− population was further subdivided based on the expression of NK1.1 into CD11c+NK1.1− DC and CD11c+NK1.1+ DC subsets. Further staining with anti- CD19, -CD3 and -F4/80 mAbs demonstrated that the resulting CD11c+NK1.1− DC were negative for B cell, T cell and macrophage markers (Fig. 1B).
(A) HMNC were identified by FSC/SSC and after exclusion of CD3/CD8β+ T cells, CD11c+ cells are segregated into CD11c+NK1.1+ and CD11c+NK1.1− populations. (B) Dot plots show the level or absence of B cells, T cells and macrophages within the gated CD11c+NK1.1− population. (C) HMNC were isolated from individual C57BL/6 mice before and after prime and boost immunizations with Pb γ−spz or uninfected mosquito debris (sham). Cells were stained with a cocktail of mAbs and analyzed by flow cytometry for identification of CD11c+NK1.1− DC and CD8+ TEM. Panels show representative contour plots of CD11c+NK1.1− DC in the livers of naïve mice and livers of mice 6 days after the primary (1°), secondary (2°) and tertiary (3°) immunizations. The percentages of the CD11c+NK1.1− DC in relation to the total HMNC/liver for each representative mouse are indicated in each panel. (D) The results show the mean %±SD of CD11c+NK1.1− DC in total HMNC of naïve, Pb γ−spz-immunized and sham-immunized mice. (E) Panels show representative contour plots of CD8+ TCM cells (CD44hiCD45RBhi ) and CD8+ TEM cells (CD44hiCD45RBlo) in the livers of naïve, Pb γ−spz-immunized and sham-immunized mice at the time-points indicated in (C). The percentages of the CD8+ TEM and CD8+ TCM cells in relation to the gated liver CD3+CD8+ T cells are indicated in the panels. (F) The results show the mean %±SD of CD8+ TEM in the gated liver CD3+CD8+ T cells in naïve, Pb γ−spz-immunized and sham-immunized mice. The results are representative of three individual mice per group from three independent experiments.
(A) CD11c+NK1.1−DC in Pbγ−spz-immune splenic MNC were identified according to the procedure described for HMNC in Fig 1. After exclusion of T cells, CD11c+ cells were segregated into CD11c+NK1.1+ and CD11c+NK1.1− populations. Splenic mononuclear cells were isolated from individual mice before and after prime and boost immunizations with Pbγ−spz or uninfected mosquito debris (sham). Cells were stained with a cocktail of mAbs and NK1.1− DC and CD8+ TEM cells were identified by flow cytometry. (B) Panels show representative contour plots of DC in the spleens of naïve mice and in spleens of mice 6 days after either the 1°, 2° and 3° immunizations and in sham-immunized mice 6 days after the 3° immunization. The percentages of the CD11c+NK1.1− DC in relation to the total SMNC/spleen for each representative mouse are indicated in each panel. (C) The results show the mean percentage ±SD of CD11c+NK1.1− DC in total spleens of naïve mice, Pb γ−spz-immunized mice at day 6 after each immunization and in sham-immunized mice at day 6 after the 3° immunization. (D) Panels show representative contour plots of CD8+ TCM cells and CD8+ TEM cells in the spleens of naïve mice as well as of Pb γ−spz-immunized mice and sham-immunized mice at the same time-points described in (B). The numbers indicate the percentages of the TCM and TEM cells in the gated splenic CD3+CD8+ T cell population. (E) The results show the mean percentage ±SD of CD8+TEM in the gated splenic CD3+CD8+ T cell population of naïve and immunized mice at day 6 after each immunization. Contour plots and bar graphs are representative of three individual mice per group in three independent experiments.
We next investigated whether repeated immunizations of C57BL/6 mice with Pbγ− spz led to the accumulation/expansion of the CD11c+NK1.1−DC populations in the liver. IHMNC were analyzed at baseline (naïve) and after each immunization for the presence of CD11c+NK1.1− DC. Consistent with another published study , liver CD11c+NK1.1− DC represented 0.02±0.019% (0.06×104 cells) of the total IHMNC in naïve mice (Fig. 1C and D). Priming with Pb γ− spz increased the level of CD11c+NK1.1− DC to 0.04±0.02% (0.14×104 cells) of the total IHMNC; further increases up to 0.09±0.03% (0.75×104 cells) and 0.15±0.04% (1.23×104 cells) were observed after the first and second boost immunizations, respectively (Fig. 1C, D and Table 1). Prime and boost immunizations with uninfected mosquito debris (sham) resulted but in a negligible increase of CD11c+NK1.1− DC in comparison to naïve CD11C+NK1.1−DC (Fig. 1C, D). Clearly, a population of CD11c+NK1.1− DC that was nearly absent in the livers of naïve mice accumulated during the course of repeated immunizations with Pb γ− spz.
Although hepatocytes serve as the primary venue for pre-erythrocytic stage development and replication of the mammalian Plasmodia spp, sporozoites also pass through the spleen. To determine whether multiple immunizations with Pb γ− spz might have caused an accumulation of DC in the spleen, we examined splenic cells at the same time points as IHMNC for the presence of CD11c+NK1.1− DC. We detected CD11c+NK1.1− DC in naïve spleens; however, they did not undergo any significant expansion during repeated immunizations with Pb γ− spz (Fig. 2 B, C).
The increase in liver CD11c+NK1.1− DC during immunization with Pb γ−spz coincides with the activation of liver CD8+ T cells
Although we previously documented the presence of CD8+TEM (CD44hiCD45RBloCD62Llo) and CD8+ TCM (CD44hiCD45RBhiCD62Lint ) cells in the livers of mice immunized with Pb γ−spz , , we repeated these analyses along with that of liver DC to determine whether the induction of the liver CD8+ T cells was linked to the emergence of hepatic CD11c+NK1.1− DC. As shown previously , , ∼1% of CD8+ TEM cells was detected in the livers of naïve mice. Six days after priming, liver CD8+ TEM cells increased significantly to 6.9±0.9% (1.4×105 cells; p = 0.001) and subsequently, six days after the second and third immunizations, to 17.8±1.9% (5.2×105 cells; p = 0.0001) and to 35.8±1.4% (11.1×105 cells; p = 0.0001), respectively (Fig. 1E and F). Liver CD8+ TEM cells increased negligibly (up to 3%) after the third immunization with uninfected mosquito debris (sham) (Fig. 1E and F). Liver CD8+ TCM cells also increased following immunization with Pb γ−spz, but because CD8+ TCM were already present at an elevated level (∼10%) in the livers of naïve mice, the only significant expansion was seen after the first immunization when the levels increased to ∼40% (p = 0.005) (Fig. 1E). Collectively, these results show that the expansion of liver CD8+ TEM cells in particular was concurrent with the accumulation of CD11c+NK1.1− DC in the livers of Pb γ−spz-immunized mice, thus implying a key role for CD11c+NK1.1− DC in the presentation of liver-stage Ags to liver CD8+ T cells that mediate protection induced by Pb γ-spz.
Overall, the percentages of CD8+ TEM cells in the spleens were lower than those in the livers of the Pb γ−spz immunized mice and similar to our previously published studies , , we observed no significant accumulation of CD8+ TEM cells in the spleens of the mice after repeated immunizations with Pb γ−spz. (Fig. 2D, E). These observations strengthen our hypothesis that the presence of CD11c+NK1.1− DC in the liver was coincident with the activation of liver CD8+ TEM and TCM cells after immunization with Pb γ−spz.
cCD8α+ DC numbers are low in the livers of naïve mice but increase following immunization with Pb γ-spz
It has been reported that splenic and liver DC represent heterogeneous populations ,  and based on the expression of cell surface markers they are divided into three major subsets: cCD8α+ DC and cCD8α− DC as well as pDC. Moreover, these various DC subpopulations differ in their capacity to activate T cells , . There is abundant evidence especially from studies of immune responses to viral infections  that cCD8α+ DC are the main APC responsible for cross-priming of CD8+ T cells. Similarly, liver cCD8α+ DC have been shown to produce pro-inflammatory cytokines and prime peptide-specific CD8+ T cells during Salmonella infection . Therefore, we asked whether cCD8α+DC were present in the expanded population of CD11c+NK1.1− DC that accumulated in the livers of Pb γ-spz immunized mice. The CD11c+NK1.1− DC were gated as described in Fig. 1A and then evaluated for the presence of CD8α+B220− (cCD8α+DC), CD8α−B220− (cCD8α−DC) and CD8α−B220+ (pDC) (Fig. 3A).
(A) Hepatic CD11c+NK1.1− DC (see Fig. 1) were stained with Abs specific for B220 and CD8α and analyzed by flow cytometry to reveal 3 subpopulations of DC . (B) HMNC isolated from livers of Pbγ-spz-fully immunized mice 6 days after the 3° immunization were incubated with a cocktail of biotinylated microbeads to deplete T cells, B cells, NK cells, granulocytes and macrophages as described in Materials and Methods. DC subpopulations were further isolated from the enriched CD11c+NK1.1− population by positive magnetic selection for cCD8α+DC and pDC and by negative selection for CD8α−DC, as described in Materials and Methods. Panels show photographs at 100× of Giemsa stained cytospins of cCD8α+DC, cCD8α−DC and pDC. (C and E) HMNC and (D and F) splenic MNC were isolated from individual naïve mice and from individual mice at 6 days after 1°, 2° and 3° immunizations with Pb γ−spz and after 3° immunizations with uninfected mosquito debris (sham). Cells were stained with a cocktail of mAbs for identification of DC subpopulations as described in (A). Bar graphs show the mean % ± SD of the cCD8α+DC (C and D) and cCD8α−DC (E and F) in the gated CD11c+NK1.1− cells. Data are representative of three individual mice per group at each time-point in two independent experiments.
Cytospin preparations of the purified DC subpopulations were stained with Giemsa and examined by microscopy. cCD8α+DC possessed an abundant cytoplasm and displayed the characteristic dendrites whereas cCD8α−DC were characterized by an abundant cytoplasm and an irregular surface. Plasmacytoid DC displayed a round, smooth surface and an oval nucleus (Fig. 3B).
The relative percentages of cCD8α+DC were determined in naïve mice and in Pb γ-spz immunized mice 6 days after each successive immunization. As can be seen from Fig. 3C, cCD8α+ DC represented 0.8±0.3% of the CD11c+NK1.1− population in naïve mice. After priming, we observed an increase of up to 5.4±4.7%; subsequently, after the second and third immunizations with Pb γ-spz the cells remained relatively stable at 5.1±2.0% and 7.4±1.9%, respectively. The same trend was observed when we considered the total number of cCD8α+ DC (Table 1). The percentage of cCD8α+ DC following three immunizations with mosquito debris (sham) was only slightly elevated in comparison to naïve control mice and did not exceed 1.5±0.8% of the CD11c+NK1.1− cells (p = 0.3). Parallel analysis of splenic CD11c+NK1.1− DC showed the presence of cCD8α+ DC at 4.4%±0.5% in naïve mice; after priming there was an increase up to 10.2±3.2%, but the levels remained relatively constant at 9.6±0.2% and 10.1±1.9% of the CD11c+NK1.1− DC following the second and the third immunizations, (Fig. 3D). Similar trends were observed with respect to total numbers of splenic DCs (Table 2). There was no increase in the percentage of cCD8α− DC in either the liver or spleen following immunization with Pb γ-spz (Fig. 3E and F).
To confirm that the liver cCD8α+ DC induced following immunization with Pb γ-spz were not CD8+ T cells that have lost or down-regulated the CD3 marker, we subjected the purified liver cCD8α+ DC (Materials and Methods) to two separate analyses: (1) flow cytometry for both surface and intracellular expression of CD3, CD8β and CD11c and (2) quantitative RT- PCR for the presence of CD8β mRNA. According to the flow cytometry results, liver cCD8α+ DC were >97% pure (Fig. 4A) and this population was negative for CD8β mRNA when assayed at 1 cell per well. (Fig. 4B). These observations not only confirm that the cCD8α+ DC is a subset of the liver CD11c+NK1.1−DC, but they demonstrate a significant accumulation of cCD8α+ DC in the livers of Pbγ-spz-immunized mice (Fig. 3C). We also characterized the liver cCD8α+DC for the expression of CD205, MHC and co-stimulatory molecules as they have been shown  to be expressed on this subset of DC.We found that indeed cCD8α+ DC expressed higher levels of CD205 than CD8α−DC (MFI = 252 vs MFI = 166, respectively). The cCD8α+ DC also exhibited high expression of MHC class I (MFI = 11×103) and class II (MFI = 14×103) and moderate levels of CD40 (MFI = 545), CD80 (MFI = 667) and CD86 (MFI = 512) (Fig. 4C).
HMNC were pooled from livers of Pbγ-spz-fully immunized mice 6 days after the 3° immunization and were incubated with a cocktail of biotinylated microbeads to deplete T cells, B cells, NK cells, granulocytes and macrophages as described in Materials and Methods. cCD8α+DC were further isolated from the enriched CD11c+NK1.1− population by positive magnetic selection as described in Materials and Methods. Staining with anti-CD8α, CD3, CD8b and CD11c was performed on permeabilized cells to reveal both the surface and the intracellular presence of these markers. (A) Dot plot show the relative % of T cells (red) and cCD8α+DC (blue) within the purified cCD8α+DC population. (B) Two-step quantitative real-time PCR was performed on RNA isolated from magnetic-bead purified liver cCD8α+DC (described in A). Ratio of CD8β/CD8α gene expression was calculated using standard curves for each gene. Measurements were done in duplicates in wells containing 1000, 1, 0.1 cells/well, or non-template control (NTC). Representative results of one out of two experiments are shown. (C) Histogram plots show expression of DEC 205, I-Ab and costimulatory molecules on the cCD8α+DC population (black lines). Grey lines represent staining of the isotype controls.
Liver cCD8α+ DC are the primary inducers of CD8+ T cell activation
Results from two preliminary experiments showed that CD8+ T cells acquired the CD44hiCD45RBlo activation phenotype upon in vitro and in vivo exposure to CD11c+NK1.1− DC purified from the livers of Pb γ-spz-immunized mice. Because cCD8α+DC are considered as the primary APC involved in cross-presentation of Ags to CD8+ T cells, we proceeded to test the capacity of liver cCD8α+DC from Pb γ-spz-immunized mice to serve as APC for the activation/differentiation of naïve CD8+ T cells and, by extension, to confirm the involvement of DC in the development of protective immunity. CD11c+NK1.1− DC isolated from pooled livers were further enriched by magnetic depletion using a cocktail of mAbs (anti-CD3, -CD19, -F4/80, -NK1.1, -Gr-1). cCD8α+ DC and pDC cells within the enriched CD11c+NK1.1− DC were each obtained by positive selection whereas the cCD8α− DC were obtained by negative selection. The purity of the enriched cCD8α+ DC ranged between 94–98% based on CD11c, CD8β, CD3 and NK1.1 expression (Fig. 5A and 6A). The purity of the cCD8α− DC and pDC was also >90%. We co-cultured negatively selected CD8+ T cells from naïve mice with either cCD8α+ DC, cCD8α− DC or pDC derived from the livers of mice immunized with Pb γ-spz or mice immunized with Pb γ-spz and challenged with infectious sporozoites and analyzed the CD8+ T cells on day 4 of the co-culture for the expression of the CD44hiCD45RBloactivation phenotype. As is evident from the results in Fig. 5B–E, cCD8α+ DC derived from the livers of either fully immunized or fully immunized-challenged mice were 3–4 fold more effective than cCD8α− DC and pDC in inducing CD8+ T cells to acquire the CD44hiCD45RBlo activation phenotype.
HMNC were pooled from livers of Pbγ-spz-fully immunized mice (n = 20) 6 days after the 3° immunization and were incubated with a cocktail of biotinylated microbeads to deplete T cells, B cells, NK cells, granulocytes and macrophages as described in Materials and Methods. DC subpopulations were further isolated from the enriched CD11c+NK1.1− population by positive magnetic selection for cCD8α+DC and pDC and by negative selection for CD8α−DC, as described in Materials and Methods. (A) Dot plots show the relative % of T cells and NK1.1 cells within the gated CD11c+CD8α+DC. (B) cCD8α+ DC, cCD8α− DC and pDC were purified from CD11c+NK1.1− DC isolated from pooled livers either 6 days after 3° immunization or 3 days after the challenge of Pbγ-spz-immunized as well as naïve mice . CD8+ T cells were isolated from the spleens of naïve mice using magnetic beads. Liver DC subpopulations and splenic CD8+ T cells were co-cultured at a ratio of 1 DC : 2 CD8+T cells for 4 days. Culture supernatants were harvested and analyzed for IL-10 and IL-12p40 by ELISA. Cells were harvested, stained with a cocktail of mAbs and the % of CD3+CD8+CD45RBloCD44hi cells (CD8+ TEM) was analyzed by flow cytometry. Results show representative contour plots of CD8+T cells co-cultured with cDC and pDC subpopulations from Pb γ-spz-immunized-challenged mice. Bar graphs show the mean percentage±SD of CD8+ TEM in the gated CD3+CD8+ T cell population after co-culture with (C) cCD8α+ DC, (D) cCD8α− DC and (E) pDC each isolated from Pbγ-spz-fully-immunized, Pbγ-spz-immunized-challenged and naïve-challenged mice. The experiments were performed twice yielding similar results. (F) IL-12p40 production in the culture supernatants of CD8+ T cells with the liver DC populations from the Pbγ-spz-immunized mice. The results expressed as pg/ml (mean±SD) represent data from two representative experiments. (G) cCD8α+DC were isolated from pooled livers of Pbγ-spz-fully-immunized mice, Pbγ-spz-immunized and challenged mice or naïve-challenged mice and co-cultured with purified naïve splenic CD8+ T cells for 4 days. Supernatants were harvested and analyzed for IL-12p40 and IL-10 by ELISA.
HMNC were pooled from livers of Pbγ-spz-fully immunized and challenged mice (n = 18) and were incubated with a cocktail of biotinylated microbeads to deplete T cells, B cells, NK cells, granulocytes and macrophages as described in Materials and Methods. DC subpopulations were further isolated from the enriched CD11c+NK1.1− population by positive magnetic selection for cCD8α+DC and pDC and by negative selection for CD8α−DC, as described in Materials and Methods. (A) Dot plots show the relative % of T cells and NK1.1 cells within the purified cCD8α+DC population. (B) CD8+ T cells were isolated from the spleens of naïve mice using magnetic beads and labeled with 2 µM CFSE. Dot plots show the gating scheme for the analysis of CFSE-labeled CD3+CD8+T cells for expression of the CD45RBloCD44hi phenotype. (C) cCD8α+ DC, cCD8α− DC and pDC were purified from CD11c+NK1.1− DC isolated from pooled livers 3 days after the challenge of Pbγ-spz-immunized mice. Liver DC subpopulations and CFSE-labeled splenic CD8+ T cells were co-cultured at a ratio of 1 DC : 2 CD8+T cells for 4 days. Cells were harvested, stained with a cocktail of mAbs and the % of CD3+CD8+CD45RBloCD44hi cells (CD8+ TEM) was analyzed by flow cytometry. Results show contour plots of CD8+T cells co-cultured with cDC and pDC subpopulations. (D) Bar graphs show the percentage of CD8+ TEM in the gated CD3+CD8+ T cell population after co-culture with cCD8α+ DC, cCD8α− DC and pDC each isolated from Pbγ-spz-immunized-challenged mice.
As was shown in Figs. 4, 5A and 6A, approximately 2–5% CD8+ T cells were present within the purified cCD8α+DC subpopulation. To minimize the potential effect of these contaminating immune CD8+ T cells in the co-cultures of naïve CD8+ T cells and cCD8α+DC, we labeled naïve CD8+T cells with CFSE before adding them to DC. At the conclusion of a 4-day culture, the CFSE-labeled CD8+ T cells were analyzed for the expression of the CD44hiCD45RBlo phenotype. The gating scheme for the analysis of the CFSE-labeled splenic CD3+CD8+T cells is shown in Fig. 6B. Similar to the findings shown in Fig. 5, these results indicate that cCD8α+ DC derived from the livers of fully immunized-challenged mice were 4-fold more effective than cCD8α− DC and pDC in inducing CD8+ T cells to acquire the CD44hiCD45RBlo phenotype (Fig. 6C and D).
cCD8α+ DC are the major IL-12 producers following challenge of Pb γ-spz-immunized mice
IL-12 is a key cytokine promoting the activation of CD8+ T cells , therefore, we evaluated the different liver DC subpopulations to assess if the superior APC properties of the cCD8α+ DC might be related to a greater capacity for producing IL-12. Levels of IL-12p40 were measured in day 4 supernatants from co-cultures of naïve CD8+ T cells with the three DC subpopulations from fully-immunized mice (Fig. 5F) and the results revealed that the cCD8α+ DC were the highest (>100 pg/ml), the cCD8α− DC were the intermediate (<50 pg/ml) and pDC were the lowest (<10 pg/ml) producers of IL-12p40. IL-10 was not detected in any of the culture supernatants (data not shown). It should be pointed out that PCR analysis of cytokine specific mRNA isolated from each DC subset confirmed the ELISA results (data not shown).
We also compared the levels of IL-12p40 and IL-10 in the day-4 supernatants co-cultures of naïve CD8+ T cells with the cCD8α+ DC subpopulations from fully-immunized or fully-immunized and challenged or naïve-challenged mice. As shown in Fig. 5G, high levels of IL-12p40 were detected in the culture supernatants with cCD8α+ DC from fully-immunized or from fully-immunized and challenged mice. In contrast, IL-12p40 was not detected in the supernatants of co-cultures containing cCD8α+ DC from naïve-challenged mice. IL-10 was not detected in any of the culture supernatants from any of the groups tested.
Activation of CD8+ T cells by liver cCD8α+ DC from Pb γ−spz-immunized mice is MHC class I- and IL-12-dependent
Ag-specific activation of CD8+ T cells is mediated via the sine qua non ligation of the TCR by peptide:MHC class I complexes and it also requires secondary signals delivered via co-stimulatory molecules  and tertiary signals from cytokines . To test whether this CD8+ T cell activation was mediated by the occupancy of the TCR by MHC:peptide complexes and if it was also dependent upon IL-12, we established parallel co-cultures consisting of liver or spleen cCD8α+ DC from mice fully immunized with Pb γ-spz and liver or splenic CD8+ T cells from naïve mice in the presence or absence of either anti-MHC class I mAb or anti-IL-12 mAb. After 4 days, the acquisition of the CD44hiCD45RBlo phenotype by CD8+ T cells was analyzed by flow cytometry and in parallel, culture supernatants were assayed for IFN-γ.
The APC activity of the liver cCD8α+ DC from Pb γ-spz immunized mice was markedly superior to that of the same subpopulation of splenic DC, regardless of the source of naïve CD8+ T cells. Whereas the splenic DC induced approximately 5% of CD8+ T cells to express the CD44hiCD45RBlo phenotype (Fig. 7A), the liver DC induced at least 2–3-fold higher percentage of these cells (Fig. 7B). The same superior APC activity of liver DC was also indicated by the fact that the level of IFN-γ was nearly three-fold higher (60 pg/ml vs 20 pg/ml) (Fig. 7D) in the supernatants of co-cultures of CD8+ T cells and liver cCD8α+ DC as compared to co-cultures with splenic cCD8α+ DC (Fig. 7C). IFNγ was not detected in the supernatants when cCD8α+ DC from either source were cultured alone (Fig.7 C, D).
Differentiation and induction of CD8+ T cell function is MHC class I- and IL-12 dependent. HMNC or splenic MNC were prepared after the 3° immunization as described in Fig. 3. CD8α+ DC were purified as described in Figs 1 and 3. Splenic (A and C) or hepatic (B and D) cCD8α+ DC were co-cultured for 4 days either alone or with purified CD8+ T cells from the livers (open bars) or spleen (filled bars) of naïve mice. Cells were harvested, stained with the appropriate mAb and analyzed by flow cytometry. Culture supernatants were analyzed for IFNγ by ELISA. (A and B) Results show the mean % of CD8+ TEM in the gated CD3+CD8+ T cell population and (C and D) the amount of IFNγ in the culture supernatant. Data are representative of two individual experiments. (E and F) Liver cCD8α+ DC were co-cultured with naïve splenic CD8+ T cells in the presence or absence of anti-IL-12 (clone C17.1) and/or anti-MHC class I (clone 28-8-6) mAbs for 4 days. (E) Results show the mean % of CD8+ TEM in the gated CD3+CD8+ T cell population and (F).the amount of IFNγ. Data is representative of two individual experiments.
The capacity of immune liver cCD8α+ DC to induce CD8+ T cells to acquire an activation phenotype and produce IFN-γ was reduced by nearly 70% in the presence of anti-MHC class I mAb. Although the inclusion of anti-IL-12 mAb had a somewhat lower inhibitory effect than anti-MHC class I mAb, it also significantly diminished both responses (Fig. 7E and F). Addition of both blocking antibodies to the co-culture abrogated the activation of CD8+ T cells by reducing both the numbers of CD44hiCD45RBlo CD8+ T cells and IFN-γ production to nearly control levels (response with CD8+ T cells alone), thus suggesting an additive or even synergistic effect of TCR ligation by MHC class I:peptide complexes and IL-12 signaling in the activation of the CD8+ T cells by liver cCD8α+ DC.
Adoptively transferred liver CD11c+NK1.1− DC obtained from Pb γ−spz-immunized mice confer protection against liver-stage infection
It has been reported that adoptive transfer of bone marrow-derived DC pulsed with Py CS protein peptide significantly reduces parasitemia in Balb/c mice challenged with Py sporozoites . Based on our preliminary observations that liver CD11c+NK1.1− DC from Pb γ-spz immunized mice were able to activate CD8+ T cells both in vitro and in vivo (data not shown), we hypothesized that adoptive transfer of these DC into naïve mice might confer protection or at least reduce the level of parasitemia in sporozoite-challenged recipients. In an initial experiment, 1×106 liver CD11c+NK1.1− DC from Pb γ-spz−immunized mice were adoptively transferred (i.v.) into naïve recipients and this group as well as a control group not receiving any DC were subsequently challenged (i.v.) with 10K infectious Pb sporozoites. The results showed that even though the transferred hepatic CD11c+NK1.1− DC significantly decreased parasitemia (Fig. 8A) and prolonged survival following infectious challenge, they were unable to confer sterile protection. Because 10K infectious sporozoites constitute a very high challenge dose, we asked whether protection could be achieved against a lower challenge dose. Thus, in the next set of experiments CD11c+NK1.1− DC from the livers of Pb γ-spz immunized-challenged or naïve-challenged mice were transferred separately into naïve recipients that were subsequently challenged with 250 sporozoites. As shown in Fig. 8B, recipients of hepatic CD11c+NK1.1− DC from Pb γ-spz-immunized-challenged mice were sterilely protected when challenged with 250 infectious sporozoites. However, protection was lost upon re-challenge at 6 months with 250 infectious sporozoites, implying that sterile protective immunity induced by the transferred CD11c+NK1.1− DC was short-lived. Notably, hepatic CD11c+NK1.1− DC from naïve-challenged mice did not confer protection even against the low-dose challenge. Collectively, these results strongly support our contention that liver CD11c+NK1.1− DC are actively involved in mediating protective immunity induced by Pb γ-spz and that their protective function depends upon their capacity to present Ag to liver CD8+ T cells that leads to the induction of effector cells against infectious sporozoites.
(A) 1×106 CD11c+NK1.1− cells, isolated from the livers of Pbγ-spz-fully-immunized mice 6 days after the 3° immunization, were adoptively transferred (i.v.) into naïve recipients. Seven days later the adoptively transferred recipients (n = 3) as well as naïve infectivity control mice (n = 3) were challenged with 10K infectious sporozoites. The results show the level of parasitemia assessed in each individual mouse and expressed as the mean of parasitemia per mice/group at 4, 7 and 10 days following challenge. (B) 1×106 hepatic CD11c+NK1.1− cells, purified as described in (A) were isolated from Pbγ-spz-immunized-challenged mice and from naïve-challenged mice and adoptively transferred into naïve syngeneic recipients (n = 13) that were challenged 7 days later with either 250 (n = 6) or 1000 (n = 7) infectious sporozoites. The protected group (250 sporozoites) were re-challenged 60 days later along with another group of naïve infectivity control mice (n = 3). Pb γ-spz-immunized mice (n = 3), used as positive controls, were sterily protected at challenge and re-challenge. Parasitemia and survival were evaluated from day 2 post-challenge.
In the present study we demonstrated that CD8α+CD11c+NK1.1− DC, although infrequent in the livers of naïve mice, increased in number in the livers of mice undergoing prime-boost immunizations with Pb γ-spz. The numbers as well as percentages of the liver cCD8α+DC increased significantly after priming and were maintained following boost immunizations, but declined after challenge with infectious sporozoites. The emergence of liver DC was concurrent with the appearance of hepatic CD8+ TEM cells (CD44hiCD45RBlo), that in agreement with our previous findings , , accumulate slowly, with the peak response occurring only after the 3rd immunization. This temporal relationship between the CD11c+NK1.1− DC and CD8+ T cells suggests that liver DC serve as key Pb liver-stage Ag-presenting cells that induce liver CD8+ T cells to differentiate into the major effector T cells that confer protection against infectious sporozoite challenge. This contention is further supported by our demonstration that liver cCD8α+ DC from either Pb γ-spz-immunized or Pb γ-spz-immunized-challenged mice induced CD8+ T cells from naïve mice to express the CD44hiCD45RBlo activation phenotype and to produce IFN-γ in an in vitro system. Most importantly, adoptive transfer of purified liver CD11c+NK1.1− DC isolated from Pb γ-spz immunized donors conferred protective immunity to congenic recipient mice challenged with infectious sporozoites.
Like infectious Plasmodia sporozoites, Pb γ-spz also invade hepatocytes but they undergo developmental arrest at the liver-stage. The aborted forms of liver-stage parasites are thought to provide a repository of liver-stage Ags needed for the induction and the maintenance of sterile and long lasting protective immunity . However, the mechanisms involved in the processing of these Ags and their presentation to T cells remain to be fully elucidated. Recent studies ,  suggest that infection with either Pb γ-spz or genetically-attenuated Plasmodia sporozoites causes hepatocytes to undergo apoptosis, a process that is currently viewed as a potent trigger of host innate and adaptive immune responses , . Therefore, it is reasonable to suggest that the attenuated sporozoite-infected apoptotic hepatocytes provide a rich repertoire of liver-stage Ags that are taken up by the liver DC and cross-presented to CD8+ T cells in the liver.
Numerous studies , ,  have shown that cCD8α+ DC are the predominant subpopulation of DC that mediate cross-presentation of cell-associated bacterial and viral Ags for the activation of CD8+ T cells. Data from the present study demonstrate for the first time that cCD8α+ DC indeed were represented in the hepatic CD11c+NK1.1− DC of Pb γ-spz-immunized mice. Although nearly absent in the livers of naïve mice, this subset increased significantly after the priming dose of Pb γ-spz and it was the most effective APC that induced CD8+ T cells to both acquire the CD44hiCD45RBlo activation phenotype and to produce IFN-γ in vitro. On the basis of these observations, we propose that liver cCD8α+ DC cross-present P. berghei liver-stage Ags derived from apoptotic Pb-infected hepatocytes to liver CD8+T cells.
Based on their percentages and numbers, the cCD8α+ DC were not the dominant subset in the livers of Pb γ-spz immunized or Pb γ-spz immunized-challenged mice. Nevertheless, the liver cCD8α+ DC exhibited superior APC activity relative to the other two subsets of liver DC and to splenic cCD8α+ DC derived from the same animals. The splenic cCD8α+ DC remained at a steady level (∼10%) during the entire prime-boost immunization with Pb γ-spz and their capacity to activate CD8+ T cells from naïve mice to express the CD44hiCD45Rblo phenotype and to produce IFN-γ was 4-fold lower than that of liver cCD8α+ DC. This dichotomy between APC activities of liver and spleen cCD8α+ DC may be explained by the fact that Pb γ-spz primarily invade and undergo a partial development within hepatocytes. Consequently, the availability of liver-stage Ags for cross-presentation would likely be higher in the liver than the spleen. Thus, the site of the induction of CD8+ T cells (liver versus spleen) might be related to the local availability of Ags for presentation to the responding CD8+ T cells.
Infectious sporozoites develop into liver merozoites that are released into the blood stream to infect red blood cells and thus, initiate clinical malaria. Challenging naive mice with infectious sporozoites also led to an accumulation of cCD8α+ DC in the liver; however the APC activity of these cells was rather ineffective in comparison to that of liver cCD8α+ DC from Pbγ-spz immunized or Pbγ-spz immunized and challenged mice. Several mechanisms may be proposed to explain the difference between the two populations of liver cCD8α+ DC. First, two recent reports ,  showed that, in contrast to attenuated sporozoites, infectious Pb sporozoites do not induce infected hepatocytes to undergo apoptosis and hence are able to complete their differentiation into liver-stage merozoites that are released into the blood stream without prior alarming or inducing the immune system. Consequently, there is likely a paucity of available early liver stage Ags in mice with active liver stage infection. Secondly, our previous results demonstrated that MHC class I molecules are down-regulated on both Kupffer cells (KC)  and liver DC (unpublished data) 24–48 hrs after the challenge of naive mice. The decreased intensity of MHC class I on the surface of KC leads to a diminution in their APC function , . The reduction of MHC class I on cCD8α+ DC in infected animals (data not shown) also might have reduced their capacity to present liver-stage Ag to CD8+ T cells. Finally, cCD8α+ DC from livers of naïve challenged mice do not produce detectable levels of IL-12 and this also could have compromised their capacity to activate CD8+ T cells. These results extend the recently published observations that IL-12-producing DC are found only in mice infected with a non-lethal strain of P. yoelii but not in mice infected with the lethal strain . It should also be noted that in contrast to our observations, a recently published study showed that human pDC that produce IFN-α following viral infection also were able to cross-prime CD8+ T cells .
DC-mediated differentiation of Ag-specific CD8+ T cells involves the requisite signal 1 delivered through the ligation of the TCR by peptide-MHC class I complexes as well as the necessary provision of co-stimulation (signal 2) and pro-inflammatory cytokines such as IL-12 (signal 3). Although the characterization of Pb liver-stage Ags is in progress, the exact nature of these Ags is currently unknown. Consequently, we were unable to use defined tetramers or transgenic CD8+ T cells to verify the Pb Ag specificity of the Pb γ-spz-immune-CD8+ T cells. However, the Ag-specificity of the responding CD8+ T cells is strongly implied by the observation that IFN-γ production and the expression of the CD44hiCD45RBlo on CD8+ T cells were abrogated by nearly 70% by anti-MHC class I mAb. The importance of this finding is supported by the fact that MHC class I KO mice are not protected by immunization with Pb γ-spz when exposed to a high challenge dose . It has also been shown previously  that the activation of CD8+ T cells and polarization toward Th1 activity is dependent on IL-12 as it induces T cell expansion and is required for IFN-γ production. In the present study, an essential role for IL-12 in the DC-mediated activation of hepatic CD8+ T cells is suggested by the fact that the hepatic cCD8α+ DC were the main producers of IL-12. Moreover, anti-IL-12 antibody inhibited both the in vitro induced expression of the CD44hiCD45RBlo activation phenotype on CD8+ T cells and IFN-γ production in co-cultures of CD8+ T cells and cCD8α+ DC. The additive/synergistic inhibitory activities of the anti-MHC class-I and anti-IL-12 antibodies indicate that both MHC class I:peptide-complexes and IL-12 were required for the activation of the hepatic CD8+ T cells by cCD8α+ DC. Our observations are firmly supported by many previous findings demonstrating the importance of IL-12 in protection against malaria liver-stage infection via its ability to directly induce IFN-γ production necessary for protection against liver-stage infection , , .
It is now well accepted that protective immunity induced by attenuated Plasmodia sporozoites is multi-factorial involving antibodies, CD4+ T cells as well as other lymphocytes, but that the CD8+ T cells are the main effectors against pre-erythrocytic infection. An essential question remains, therefore, whether the activation of liver-stage Ag-specific CD8+ T cells by DC occurs in the liver or in an immune organ, such as the draining celiac LN, after which the effector CD8+ T cells are recruited to the liver. Although our study addressed neither the issue of CD11c+NK1.1− DC migration into the liver from other tissues nor their potential differentiation from cells such as CD11c+NK1.1+ DC that are constitutively present in the livers of naïve mice , nonetheless such discussion is warranted in order to reach a better understanding of the individual immune events and their integration into the overall mechanisms that culminate in protective immunity.
According to evidence from studies of immune responses elicited by model protein Ags, viral and microbial Ags as well as from studies of responses to Plasmodia Ags, three different scenarios can be considered: (1) DC acquire and present parasite Ag to CD8+ T cells in the draining LN and subsequently the CD8+ T cells migrate to the liver; (2) DC acquire parasite Ags in the liver after which they migrate to the LN to activate CD8+ T cells and then both cells migrate back to liver; and (3) CD8+ T cell:DC interactions occur exclusively in the liver, where DC cross-present liver stage Ag to liver CD8+ T cells.
According to the first scenario, immature celiac LN DC with enhanced capacity for Ag uptake, encounter Plasmodia Ags, including those associated with sporozoites that are en route to invade the liver. Among the available Ag for uptake by DC, the CS protein might be the primary candidate as it readily sloughs off but is also quickly replaced on sporozoites. Although uptake of other candidate Plasmodia Ags is quite possible, priming of CD8+ T cells specific for Py CS protein has recently been observed in the draining LN after vaccination with Py γ-spz . Following Ag uptake, the DC mature and activate LN CD8+ T cells that then migrate to the liver to perform their effector function during infection. Although our observation of the late emergence of CD44hiCD45RBlo CD8+ T cells in the liver could be accommodated by this scenario, the incompatibility lies with the Ag specificity as liver CD8+ T cells do not recognize peptides from Pb CS protein (Krzych U, unpublished data). Other sporozoite or early liver-stage Ag specificities are possible, but they have not been tested.
According to the second scenario, liver resident immature DC take up liver-stage Ag from the apoptotic hepatocytes invaded by Pb γ-spz. The Ag-loaded and matured DC then migrate to the liver draining LN, where they engage in the induction of CD8+ T cells that subsequently migrate to the liver during infection to perform the effector function. In support of this scenario, influenza-specific CD8+T cells that have been activated in the LN perform optimal effector function only in the presence of Ag-loaded DC in the lung . Although quite attractive, it has been shown that migratory DC do not normally express the CD8α+ phenotype  and the cCD8α+ DC were the major activators of CD8+ T cells in our system. In addition, according to the results from a preliminary study, the draining LN contains a very small proportion of activated CD8+ T cells following i.d. or i.v. immunizations with Pb γ-spz.
The third model proposes that cross presentation of the aborted liver-stage Ags by liver DC and the subsequent activation of liver CD8+ T cells occurs uniquely in the liver. The possibility of the local activation of CD8+ T cells by liver-stage Ag is quite attractive for several reasons, but especially because it enables a rapid response in the event of parasite invasion of the liver that occurs within 2 mins following i.v. challenge . Furthermore, although activated CD8+ T cells home from other regions of the body to the liver where they become eliminated as a way of purging their destructive effector mechanism , there is also evidence from other studies showing that direct activation of naïve CD8+ T cells can indeed occur in the liver , , . The accumulation of CD11c+NK1.1− DC in the liver could be explained on the basis of an inflammatory response, i.e., production of IL-12 by KC shortly after priming with Pb γ-spz . Others have also shown that recruitment of DC to the liver is mediated by soluble factors such as MIP-1α and IL-12, possibly released from KC as DC recruitment is absent in KC cell-depleted animals . KC also trap and maintain DC in the liver via selective adhesion involving N-acetylgalactosamine-mediated interactions . The co-presence of cCD8α+ DC and CD8+ T cells in the liver and the proximity of liver-stage Ags from the apoptotic Pb γ-spz invaded hepatocytes provide all the necessary elements for the initiation of adoptive immune responses in a non-lymphoid organ that is the venue for infection with Plasmodia parasites.
Materials and Methods
Female C57BL/6 mice (6–8 weeks old) were purchased from The Jackson Laboratory (Bar Harbor, ME) and were housed at the Walter Reed Army Institute of Research (WRAIR) animal facility and handled according to institutional guidelines. All procedures were reviewed and approved by the Animal Care and Use Committee of the institute and were performed in a facility accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International.
Sporozoites, immunizations and parasitemia
Pb sporozoites were isolated from mosquitoes as described previously  Sporozoites were attenuated by exposure to 15,000 rad of γ-radiation (Cesium-137 source Mark 1 series or Cobalt-60 Model 109; JL Shepard & Associates, San Fernando, CA). As previously described , , mice were primed i.v. with 75K Pbγ−spz followed by two i.v. boost immunizations of 20K Pbγ−spz administered 1 week apart. Control mice were primed and boosted with material from sham dissected mosquitoes that represented an equivalent number of infected mosquitoes. For challenge experiments, mice were infected i.v. with a low dose of either 250 or 1000 infectious sporozoites or a high dose of 10, 000 infectious sporozoites approximately one week after the last boost immunization. Thin blood smears were prepared from individual mice starting on day 2 after the challenge and the percentage of parasitized red blood cells was determined microscopically.
The following mAbs anti-NK1.1 FITC, NK1.1 PE, CD40 FITC, CD80 FITC, CD86 FITC, H-2Kb FITC, I-Ab FITC, DEC-205 FITC (Serotec, Raleigh, NC.), CD45RB FITC, CD44 PE, B220/CD45RB PE, B220/CD45RB PerCP, CD8α PerCP, CD8α Pacific Orange (Caltag, Burlingame, CA.), CD11c APC, CD3 biotin, CD3e PE, CD3e Pacific Blue (eBioscience, San Diego, CA.), CD8b PE, CD19 biotin, CD19 PE, CD11c biotin, Gr-1 FITC, F4/80 PerCP (Biolegend, San Diego, CA.), F4/80 biotin, Gr-1 biotin, NK1.1 biotin, CD16/CD32 (Fc block), purified anti-IL12p40 (C17.8), purified anti-mouse H-2Kb/H-2Db (28-8-6) and isotype controls and anti-biotin microbeads (Miltenyi Biotec, Auburn, CA.) were (unless stated otherwise) obtained from (BD Pharmingen, San Diego, CA)
Isolation of IHMC
Mice were euthanized by CO2 inhalation. Livers were exposed and the inferior vena cava was cut for blood outflow. The livers were perfused with 10ml PBS , removed and pressed through a 70 µM nylon cell strainer (BD Labware, Franklin Lakes, NJ) and the liver cell suspension was processed as previously described , . Briefly, the liver cells were washed 1× in PBS, resuspended in PBS containing 35% Percoll (Amersham Pharmacia Biotec, Uppsala, Sweden) and centrifuged at 2,000 rpm for 20 min. Erythrocytes in the cell pellet were lysed with RBC lysis buffer (Sigma, St. Louis, MO) and the remaining hepatic mononuclear cells (HMNC) were washed and resuspended in complete RPMI 1640 medium.
Isolation of hepatic CD11c+NK1.1− DC
CD11c+NK1.1− cells were isolated as previously described . Briefly, HMNC were incubated with a cocktail of biotinylated (anti-CD3e, CD19, NK1.1, Gr-1, F4/80) mAb for 15 minutes at 4°C. Following washing in buffer (PBS containing 2mM EDTA and 0.5% BSA), the cells were magnetically labeled with anti-biotin microbeads for 15 minutes at 4°C. The CD11c+ cells were enriched by negative isolation on a MS column (Miltenyi Biotec) according to the manufacturer's instructions. The purity and viability of the enriched CD11c+NK1.1− preparation were assessed by flow cytometry and trypan blue exclusion,.and were found to be ∼94% and >96% respectively.
Isolation of CD11c+NK1.1− DC subpopulations
For isolation of cCD8α+ DC or pDC, enriched hepatic CD11c+NK1.1− cells were labeled with CD8α (Ly-2) microbeads or CD45R (B220) microbeads and the labeled cells isolated by positive selection on MS columns. For isolation of cCD8α−DC, CD11c+NK1.1− cells were labeled with a cocktail of CD8α (Ly-2) and CD45R (B220) microbeads and isolated by negative selection on MS columns. The purity of the DC subpopulations was >95% and viability was >97%.
Total RNA was extracted from purified splenic CD8+T cells and purified liver CD8α+CD11c+ cells using an RNeasy Plus Mini Kit (Qiagen, Valencia, CA) and treated with DNase. RNA concentration and quality control were determined by UV spectrophotometry and denaturing agarose gel electrophoresis. For all samples, cDNA was synthesized from total RNA using High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA) according to the manufacturer's instructions. The qPCR for CD8α chain mRNA and CD8β chain mRNA was performed using SuperArray's RT2 Real-Time SYBR Green/Rox PCR master kit, CD8α (PPM04031A), CD8b1 (PPM04032E) primer mixes, and Applied Biosystems 7500 Fast Real-Time PCR System. The temperature profile was 10 min at 95°C followed by 40 cycles of (95°C, 15 sec; 60°C, 60 sec). Melting curve analysis was performed after qPCR for quality control purposes.
For evaluation of T cell activation, four-color staining of HMNC and splenic MNC was performed using a combination of the following mAb: FITC-anti-CD45RB (16A), PE-anti-CD3 (17A2), PerCP-anti-CD8α (53-6.7), and APC-anti-CD44 (IM7). For identification of CD11c+NK1.1− DC; HMNC and splenic MNC were stained using a combination of the following mAb: FITC-anti-NK1.1 (PK136), PE-anti-CD3e (17A2), PE-anti-CD8b (53-5.8), PerCP-anti-CD45R/B220 (RA3-6B2), Pacific Orange-anti-CD8α, and APC-anti-CD11c (HL3). For evaluation of DEC-205, MHC and costimulatory molecules, HMNC were analyzed using a combination of the following mAb: FITC-anti-DEC-205 (NLDC-145), FITC-anti-CD40, FITC-anti-CD80 (16-10A1), FITC-anti-CD86 (GL-2), FITC-anti-IAb (AF6-120.1), FITC-anti H-2Kb, PE-anti-CD3e, PE-anti-CD8b, PerCP-anti-CD45R/B220, Pacific Orange-anti-CD8a, and APC-anti-CD11c. For evaluation of the purity of the CD11c+CD8α+ cells, the positively isolated cells were stained using a combination of the following mAb: FITC-anti-NK1.1, PE-anti-CD11c, PerCP-anti-CD19, Pacific Blue-anti-CD3e and Pacific Orange-anti-CD8α. Briefly, 2–10×105 cells were resuspended in cold assay buffer (PBS containing 1% BSA (Sigma) and 0.01% sodium azide) and incubated with anti-FcR 24G2 and 0.5 µg of the relevant mAb at 4°C for 30 min. Cells were washed and resuspended in cold assay buffer. Flow cytometry analyses of T cells and of DC were performed on a FACSCalibur (BD Biosciences) and an LSRII (BD Biosciences) flow cytometer, respectively. Data analysis was performed using FlowJo 8.1.0 software (Tree Star, Inc., Ashland, OR).
In vitro cultures
CD11c+NK1.1− DC (5×105) or purified DC subpopulations (5×105) from livers or spleens of Pbγ-spz-immunized mice were co-cultured with naïve hepatic or splenic CD8+ T cells (1×106) in a 1ml volume of complete culture media (RPMI 1640 supplemented with 10% heat-inactivated FCS, 10mM Hepes, 100U of penicillin, 100 µg of streptomycin and 4mM L-glutamine) in a 48-well tissue culture plate for 4 days. A ratio of 1∶2 of hepatic DC:CD8+ T cells yielded the highest percentage of recovered viable CD8+ T cells and was used in all experiments. For blocking experiments, antibodies against lL-12p40 (clone C17.8) and H-2Kb/H-2Db (clone 28-8-6) were used at 5 µg/ml. Cells and culture supernatants were harvested at day 4 for determination of CD8+ T cell activation and cytokine production, respectively.
Purified splenic CD8+ T cells were incubated with 2 µM CFSE (Molecular Probes – Invitrogen, Eugene, OR.) for 7 min at room temperature. Cells were washed three times with 10ml complete culture medium. After the final wash, the cells were resuspended in complete culture medium and viability was >98%. 1×106 CFSE-labeled splenic CD8+T cells were co-cultured with 5×105 purified cCD8α+DC, cCD8α−DC and pDC isolated from livers of Pbγ-spz-immunized-challenged mice in a 1ml volume of complete culture media in a 48-well tissue culture plate for 4 days. Cells were harvested, washed and stained with PE-anti-CD45RB (16A), PerCP-anti-CD3 (17A2), APC-anti-CD44 (IM7) and Pacific Orange-anti-CD8a for evaluation of T cell activation.
Freshly purified hepatic cCD8α+DC, cCD8α−DC and pDC were cytospun onto glass slides (1 min at 1000rpm) and fixed in methanol for 15 min at room temperature. Slides were stained with Giemsa (Sigma Diagnostics, St Louis, MO) for 20 min and photographed.
IL-10 and IL-12p40 determination
The presence of IL-10 and IL-12p40 were determined using IL-10 and IL-12p40 ELISA kits (BD Biosciences, San Diego, CA) according to the manufacturer's instructions.
IFN-γ was determined using an R&D ELISA kit (R&D Systems, Minneapolis, MN) and according to the manufacturer's instructions.
To evaluate the protective effect of hepatic DC against liver-stage infection, purified CD11c+NK1.1− DC (1×106) from livers of Pb γ−spz-immunized mice were transferred (i.v.) into naïve C57BL/6 mice. Seven days after the adoptive transfer, recipient mice were challenged with 250 or 1000 infectious sporozoites. Parasitemia and survival were evaluated from day 2 post-challenge.
The authors would like to thank the members of the Department of Entomology at WRAIR and in particular Drs. Jack Williams and Tatyana Savranskaya for the provisions of infected mosquitoes; Mr. Isaac Chalom for hand dissected sporozoites; Drs. Joanne Lumsden, Sofia Casares and Akiko Iwasaki for critical reading of the manuscript; and Dr. D. Gray Heppner for generous support and encouragement in this and other projects. The views of the authors do not purport to reflect the position of the Department of the Army or the Department of Defense.
Conceived and designed the experiments: OJ RJS UK. Performed the experiments: OJ GD GS DL. Analyzed the data: OJ DL RJS UK. Contributed reagents/materials/analysis tools: OJ GS DL. Wrote the paper: OJ RJS UK.
- 1. Villadangos JA, Schnorrer P (2007) Intrinsic and cooperative antigen-presenting functions of dendritic-cell subsets in vivo. Nat Rev Immunol 7: 543–555.
- 2. Belz GT, Smith CM, Eichner D, Shortman K, Karupiah G, et al. (2004) Cutting edge: conventional CD8 alpha+ dendritic cells are generally involved in priming CTL immunity to viruses. J Immunol 172: 1996–2000.
- 3. Belz GT, Shortman K, Bevan MJ, Heath WR (2005) CD8alpha+ dendritic cells selectively present MHC class I-restricted noncytolytic viral and intracellular bacterial antigens in vivo. J Immunol 175: 196–200.
- 4. Hammad H, de Vries VC, Maldonado-Lopez R, Moser M, Maliszewski C, et al. (2004) Differential capacity of CD8+ alpha or CD8- alpha dendritic cell subsets to prime for eosinophilic airway inflammation in the T-helper type 2-prone milieu of the lung. Clin Exp Allergy 34: 1834–1840.
- 5. Jewell NA, Vaghefi N, Mertz SE, Akter P, Peebles RS Jr., et al. (2007) Differential type I interferon induction by respiratory syncytial virus and influenza a virus in vivo. J Virol 81: 9790–9800.
- 6. Kawamura K, Kadowaki N, Kitawaki T, Uchiyama T (2006) Virus-stimulated plasmacytoid dendritic cells induce CD4+ cytotoxic regulatory T cells. Blood 107: 1031–1038.
- 7. Villadangos JA, Heath WR (2005) Life cycle, migration and antigen presenting functions of spleen and lymph node dendritic cells: limitations of the Langerhans cells paradigm. Semin Immunol 17: 262–272.
- 8. Chung Y, Chang JH, Kim BS, Lee JM, Kim HY, et al. (2007) Anatomic location defines antigen presentation by dendritic cells to T cells in response to intravenous soluble antigens. Eur J Immunol 37: 1453–1462.
- 9. Wilson NS, Villadangos JA (2004) Lymphoid organ dendritic cells: beyond the Langerhans cells paradigm. Immunol Cell Biol 82: 91–98.
- 10. Nardin E, Zavala F, Nussenzweig V, Nussenzweig RS (1999) Pre-erythrocytic malaria vaccine: mechanisms of protective immunity and human vaccine trials. Parassitologia 41: 397–402.
- 11. Hoffman SL, Goh LM, Luke TC, Schneider I, Le TP, et al. (2002) Protection of humans against malaria by immunization with radiation-attenuated Plasmodium falciparum sporozoites. J Infect Dis 185: 1155–1164.
- 12. Berenzon D, Schwenk RJ, Letellier L, Guebre-Xabier M, Williams J, et al. (2003) Protracted protection to Plasmodium berghei malaria is linked to functionally and phenotypically heterogeneous liver memory CD8+ T cells. J Immunol 171: 2024–2034.
- 13. Jung S, Unutmaz D, Wong P, Sano G, De los Santos K, et al. (2002) In vivo depletion of CD11c(+) dendritic cells abrogates priming of CD8(+) T cells by exogenous cell-associated antigens. Immunity 17: 211–220.
- 14. Plebanski M, Hannan CM, Behboudi S, Flanagan KL, Apostolopoulos V, et al. (2005) Direct processing and presentation of antigen from malaria sporozoites by professional antigen-presenting cells in the induction of CD8 T-cell responses. Immunol Cell Biol 83: 307–312.
- 15. Bruna-Romero O, Rodriguez A (2001) Dendritic cells can initiate protective immune responses against malaria. Infect Immun 69: 5173–5176.
- 16. Chakravarty S, Cockburn IA, Kuk S, Overstreet MG, Sacci JB, et al. (2007) CD8+ T lymphocytes protective against malaria liver stages are primed in skin-draining lymph nodes. Nat Med 13: 1035–1041.
- 17. Anjuere F, Martin P, Ferrero I, Fraga ML, del Hoyo GM, et al. (1999) Definition of dendritic cell subpopulations present in the spleen, Peyer's patches, lymph nodes, and skin of the mouse. Blood 93: 590–598.
- 18. Leenen PJ, Radosevic K, Voerman JS, Salomon B, van Rooijen N, et al. (1998) Heterogeneity of mouse spleen dendritic cells: in vivo phagocytic activity, expression of macrophage markers, and subpopulation turnover. J Immunol 160: 2166–2173.
- 19. Garrigan K, Moroni-Rawson P, McMurray C, Hermans I, Abernethy N, et al. (1996) Functional comparison of spleen dendritic cells and dendritic cells cultured in vitro from bone marrow precursors. Blood 88: 3508–3512.
- 20. Grauer O, Wohlleben G, Seubert S, Weishaupt A, Kampgen E, et al. (2002) Analysis of maturation states of rat bone marrow-derived dendritic cells using an improved culture technique. Histochem Cell Biol 117: 351–362.
- 21. Shu SA, Lian ZX, Chuang YH, Yang GX, Moritoki Y, et al. (2007) The role of CD11c(+) hepatic dendritic cells in the induction of innate immune responses. Clin Exp Immunol 149: 335–343.
- 22. Leiriao P, Mota MM, Rodriguez A (2005) Apoptotic Plasmodium-infected hepatocytes provide antigens to liver dendritic cells. J Infect Dis 191: 1576–1581.
- 23. van de Sand C, Horstmann S, Schmidt A, Sturm A, Bolte S, et al. (2005) The liver stage of Plasmodium berghei inhibits host cell apoptosis. Mol Microbiol 58: 731–742.
- 24. Sturm A, Amino R, van de Sand C, Regen T, Retzlaff S, et al. (2006) Manipulation of host hepatocytes by the malaria parasite for delivery into liver sinusoids. Science 313: 1287–1290.
- 25. den Haan JM, Lehar SM, Bevan MJ (2000) CD8(+) but not CD8(−) dendritic cells cross-prime cytotoxic T cells in vivo. J Exp Med 192: 1685–1696.
- 26. Schulz O, Reis e Sousa C (2002) Cross-presentation of cell-associated antigens by CD8alpha+ dendritic cells is attributable to their ability to internalize dead cells. Immunology 107: 183–189.
- 27. White KL, Snyder HL, Krzych U (1996) MHC class I-dependent presentation of exoerythrocytic antigens to CD8+ T lymphocytes is required for protective immunity against Plasmodium berghei. J Immunol 156: 3374–3381.
- 28. Scheller LF, Azad AF (1995) Maintenance of protective immunity against malaria by persistent hepatic parasites derived from irradiated sporozoites. Proc Natl Acad Sci U S A 92: 4066–4068.
- 29. O'Connell PJ, Morelli AE, Logar AJ, Thomson AW (2000) Phenotypic and functional characterization of mouse hepatic CD8 alpha+ lymphoid-related dendritic cells. J Immunol 165: 795–803.
- 30. Jobe O, Lumsden J, Mueller AK, Williams J, Silva-Rivera H, et al. (2007) Genetically attenuated Plasmodium berghei liver stages induce sterile protracted protection that is mediated by major histocompatibility complex Class I-dependent interferon-gamma-producing CD8+ T cells. J Infect Dis 196: 599–607.
- 31. Pillarisetty VG, Shah AB, Miller G, Bleier JI, DeMatteo RP (2004) Liver dendritic cells are less immunogenic than spleen dendritic cells because of differences in subtype composition. J Immunol 172: 1009–1017.
- 32. Lian ZX, Okada T, He XS, Kita H, Liu YJ, et al. (2003) Heterogeneity of dendritic cells in the mouse liver: identification and characterization of four distinct populations. J Immunol 170: 2323–2330.
- 33. Pooley JL, Heath WR, Shortman K (2001) Cutting edge: intravenous soluble antigen is presented to CD4 T cells by CD8- dendritic cells, but cross-presented to CD8 T cells by CD8+ dendritic cells. J Immunol 166: 5327–5330.
- 34. Backer R, van Leeuwen F, Kraal G, den Haan JM (2008) CD8(−) dendritic cells preferentially cross-present Saccharomyces cerevisiae antigens. Eur J Immunol.
- 35. Johansson C, Wick MJ (2004) Liver dendritic cells present bacterial antigens and produce cytokines upon Salmonella encounter. J Immunol 172: 2496–2503.
- 36. De Smedt T, Butz E, Smith J, Maldonado-Lopez R, Pajak B, et al. (2001) CD8alpha(−) and CD8alpha(+) subclasses of dendritic cells undergo phenotypic and functional maturation in vitro and in vivo. J Leukoc Biol 69: 951–958.
- 37. Wilson DC, Matthews S, Yap GS (2008) IL-12 signaling drives CD8+ T cell IFN-gamma production and differentiation of KLRG1+ effector subpopulations during Toxoplasma gondii Infection. J Immunol 180: 5935–5945.
- 38. Mescher MF, Curtsinger JM, Agarwal P, Casey KA, Gerner M, et al. (2006) Signals required for programming effector and memory development by CD8+ T cells. Immunol Rev 211: 81–92.
- 39. Valenzuela J, Schmidt C, Mescher M (2002) The roles of IL-12 in providing a third signal for clonal expansion of naive CD8 T cells. J Immunol 169: 6842–6849.
- 40. van Dijk MR, Douradinha B, Franke-Fayard B, Heussler V, van Dooren MW, et al. (2005) Genetically attenuated, P36p-deficient malarial sporozoites induce protective immunity and apoptosis of infected liver cells. Proc Natl Acad Sci U S A 102: 12194–12199.
- 41. Schaible UE, Winau F, Sieling PA, Fischer K, Collins HL, et al. (2003) Apoptosis facilitates antigen presentation to T lymphocytes through MHC-I and CD1 in tuberculosis. Nat Med 9: 1039–1046.
- 42. Winau F, Weber S, Sad S, de Diego J, Hoops SL, et al. (2006) Apoptotic vesicles crossprime CD8 T cells and protect against tuberculosis. Immunity 24: 105–117.
- 43. Steers N, Schwenk R, Bacon DJ, Berenzon D, Williams J, et al. (2005) The immune status of Kupffer cells profoundly influences their responses to infectious Plasmodium berghei sporozoites. Eur J Immunol 35: 2335–2346.
- 44. Pradel G, Garapaty S, Frevert U (2004) Kupffer and stellate cell proteoglycans mediate malaria sporozoite targeting to the liver. Comp Hepatol 3: Suppl 1S47.
- 45. Wykes MN, Liu XQ, Beattie L, Stanisic DI, Stacey KJ, et al. (2007) Plasmodium Strain Determines Dendritic Cell Function Essential for Survival from Malaria. PLoS Pathog 3: e96.
- 46. Di Pucchio T, Chatterjee B, Smed-Sorensen A, Clayton S, Palazzo A, et al. (2008) Direct proteasome-independent cross-presentation of viral antigen by plasmacytoid dendritic cells on major histocompatibility complex class I. Nat Immunol 9: 551–557.
- 47. Sedegah M, Finkelman F, Hoffman SL (1994) Interleukin 12 induction of interferon gamma-dependent protection against malaria. Proc Natl Acad Sci U S A 91: 10700–10702.
- 48. Hoffman SL, Crutcher JM, Puri SK, Ansari AA, Villinger F, et al. (1997) Sterile protection of monkeys against malaria after administration of interleukin-12. Nat Med 3: 80–83.
- 49. Doolan DL, Hoffman SL (1999) IL-12 and NK cells are required for antigen-specific adaptive immunity against malaria initiated by CD8+ T cells in the Plasmodium yoelii model. J Immunol 163: 884–892.
- 50. Chen L, Calomeni E, Wen J, Ozato K, Shen R, et al. (2007) Natural killer dendritic cells are an intermediate of developing dendritic cells. J Leukoc Biol 81: 1422–1433.
- 51. McGill J, Van Rooijen N, Legge KL (2008) Protective influenza-specific CD8 T cell responses require interactions with dendritic cells in the lungs. J Exp Med 205: 1635–1646.
- 52. Belz GT, Smith CM, Kleinert L, Reading P, Brooks A, et al. (2004) Distinct migrating and nonmigrating dendritic cell populations are involved in MHC class I-restricted antigen presentation after lung infection with virus. Proc Natl Acad Sci U S A 101: 8670–8675.
- 53. Shin SC, Vanderberg JP, Terzakis JA (1982) Direct infection of hepatocytes by sporozoites of Plasmodium berghei. J Protozool 29: 448–454.
- 54. Mehal WZ, Juedes AE, Crispe IN (1999) Selective retention of activated CD8+ T cells by the normal liver. J Immunol 163: 3202–3210.
- 55. Bertolino P, Bowen DG, McCaughan GW, Fazekas de St Groth B (2001) Antigen-specific primary activation of CD8+ T cells within the liver. J Immunol 166: 5430–5438.
- 56. Klein I, Crispe IN (2006) Complete differentiation of CD8+ T cells activated locally within the transplanted liver. J Exp Med 203: 437–447.
- 57. Wuensch SA, Pierce RH, Crispe IN (2006) Local intrahepatic CD8+ T cell activation by a non-self-antigen results in full functional differentiation. J Immunol 177: 1689–1697.
- 58. Matsuno K, Nomiyama H, Yoneyama H, Uwatoku R (2002) Kupffer cell-mediated recruitment of dendritic cells to the liver crucial for a host defense. Dev Immunol 9: 143–149.
- 59. Matsuno K, Ezaki T (2000) Dendritic cell dynamics in the liver and hepatic lymph. Int Rev Cytol 197: 83–136.