We first examined the response to a heterologous antigen during Plasmodium chabaudi (AS strain) infection (Figure 1a) to determine whether this murine model reflected the clinical immunosuppression observed with P. falciparum infection 18192021. Mice were immunized with the model antigen ovalbumin (OVA) and lipopolysaccharide (LPS) to act as adjuvant at various times after infection, and OVA-specific serum immunoglobulin G (IgG) was measured 21 days later.

DC activation is central to induction of adaptive immunity 35, and previous studies have suggested that several protozoan pathogens have evolved mechanisms to suppress this response and consequently to reduce immune-mediated protection 44. Human DCs cultured with P. falciparum-infected erythrocytes are hyporesponsive to stimulation with LPS and less capable of stimulating CD4⁺ T-cell responses 42. This observation remains controversial, however, as studies using murine models have suggested that DCs may be activated during increasing parasitemia in vivo 45 and following culture in vitro with parasite schizont-infected erythrocytes 40.

As the results above indicated that immune responsiveness in vivo is dynamically regulated during infection, we studied the ability of P. chabaudi-infected erythrocytes to modulate DCs directly, by examining the expression of MHC class II and co-stimulatory molecules on DCs. Bone-marrow-derived DCs were incubated with infected erythrocytes and the expression of surface markers examined at various times over the following 24 hours. Our results show that DCs expressed very low levels of surface MHC class II and the co-stimulatory molecules CD40, CD80 and CD86 when cultured in growth medium alone, thus confirming the immature state of these DCs in culture (Figure 2a-d). Stimulation with LPS promoted a significant increase in the expression level of all co-stimulatory molecules within 6 hours. DCs incubated with RBCs or pRBCs did not, however, increase expression of MHC class II, CD40, CD80 and CD86, indicating that malaria parasites do not induce DC activation directly. Analysis of cytokine production showed that DCs exposed to RBC or pRBCs produce small, though detectable, amounts of both interleukins IL-12 and IL-10 that are a thousand-fold lower than those observed after LPS stimulation (Figure 2e,f), suggesting that the presence of parasites does not result in DC activation. The viability of treated DCs and control cells was quantified after 24 hours of culture by trypan blue exclusion (Figure 2g) and propidium iodide (PI) and annexin V staining (data not shown) and was not significantly affected by pRBCs.

Having established that pRBCs at a ratio of 100:1 do not directly induce DC maturation, we examined whether pRBC-treated DCs retained their ability to mature in response to LPS treatment in vitro. DCs were exposed to RBCs or pRBCs for 24 hours and subsequently challenged with LPS. After 18 hours of LPS stimulation, the expression levels of MHC class II, CD40, CD80 and CD86 increased significantly (Figure 3a-d). DCs pre-incubated with pRBCs and subsequently challenged with LPS, however, showed significantly lower levels of expression of MHC class II, CD40, and CD86 compared with those observed when cells were not treated with pRBCs or were pre-incubated with RBCs before the LPS challenge (Figure 3a-d). Kinetic studies demonstrated that the ability of pRBCs to induce this hyporesponsive state in DCs required at least 6 hours pre-incubation before the addition of LPS (data not shown). Cytokine production following treatment with LPS showed that, although DCs treated with pRBCs could still produce appreciable levels of IL-12 and IL-10 in response to LPS, the amount produced was significantly lower than that produced by DCs pre-incubated with RBCs (Figure 3e,f). As the interaction between CD40 on DCs and its ligand CD40L on T cells in vivo is known to be crucial in the production of bioactive IL-12 and upregulation of adhesion and co-stimulatory molecules 4647, we stimulated bone-marrow-derived DCs with CD40L-transfected fibroblasts (Figure 3g,h). DCs treated with RBCs significantly upregulated CD40 expression in response to CD40L and produced high levels of the inducible IL-12p40 subunit. CD40 ligation, however, did not rescue the reduced maturation of DCs treated with pRBCs, although, as previously observed with the LPS treatment, these cells still produced IL-12 p40 but to a lesser extent than the control groups.

As our results suggested that malaria-infected erythrocytes might modulate the responsiveness of DCs in vitro, we next investigated the activation status of splenic DCs in vivo during a time-course of infection with P. chabaudi. DCs isolated from spleens of mice 4 days after infection showed a moderately activated phenotype, as demonstrated by increased expression of CD40 and CD80 (Figure 4a), confirming previous reports 45. DCs isolated from infected animals 12 and 20 days after infection, however, showed a reduced level of activation, with lower levels of CD40, CD80, CD86 and MHC class II molecules on their surface compared with DCs from uninfected animals (Figure 4b,c). Whereas DCs from uninfected mice upregulated CD40, CD80 and CD86 following LPS stimulation (Figure 4d-g), DCs isolated from the spleens of P. chabaudi-infected mice remained refractory to in vitro LPS-induced maturation, with reduced levels of these molecules following stimulation. Thus it seems that, in vivo, DCs are activated soon after infection, and the level of activation on DCs is reduced following the peak of infection (days 12-20), and this cannot be abrogated by microbial stimulation ex vivo.

As we had demonstrated that malaria infection modulates key aspects of DC function, we wanted to examine the possible mechanisms involved. Initially, to determine whether maturation of the parasite from the trophozoite stage to the schizont stage in vitro or some metabolic process within the pRBCs was required for induction of DC hyporesponsiveness, pRBCs were fixed with paraformaldehyde and incubated with DCs for 24 hours before the addition of LPS (Figure 5a-c). The expression levels of MHC class II, CD40 and CD86 were significantly reduced when DCs were co-cultured with fixed pRBCs before LPS challenge, confirming that trophozoite-infected erythrocytes downregulate DC activation in response to LPS treatment and suggesting that growth and maturation of the parasite into schizonts is not required for suppression. In an attempt to understand the mechanisms involved in the parasite-mediated modulation of DC function, we investigated the role of selected parasite components on the LPS-induced maturation of DCs. We addressed this question initially by analyzing the effect that parasite proteins expressed on the erythrocyte surface membrane have on DC function. DCs were exposed to RBC membranes (ghosts) isolated from infected or uninfected erythrocytes. After 24 hours of culture, DCs were challenged with LPS and the expression of co-stimulatory molecules analyzed 18 hours later (Figure 5d,e). Our results clearly show that ghosts isolated from pRBCs did not alter the ability of DCs to respond to LPS treatment in vitro, as observed by the levels of MHC class II and CD40, demonstrating that proteins expressed on the surface membranes of pRBCs are apparently not essential for the modulation of DC function.

Having established that parasite proteins expressed on the erythrocyte cell membrane are not responsible for the modulation of LPS-induced maturation of DCs in vitro, we focused our attention on HZ, a by-product of hemoglobin digestion. We observed that bone-marrow-derived DCs cultured in vitro with infected erythrocytes accumulated intracellular malarial pigment (Figure 6a). Flow cytometric analysis of splenic DCs, identified by CD11c expression, also demonstrated an increase in the size and granularity of DCs during infection (Figure 6b), and DCs isolated ex vivo as well as DCs in spleen sections showed conspicuous HZ deposition (Figure 6a).

In order to assess the role of HZ in the parasite-induced modulation of DC function, we initially analyzed its ability to activate DCs directly in vitro (Figure 6c-e). Even at the highest dose (20 μM), HZ did not induce DC maturation, as the levels of MHC class II, CD40 and CD86 were the same as the levels expressed by untreated controls. We then examined whether HZ-treated DCs still responded to LPS treatment in vitro (Figure 6f-h). Our data clearly demonstrate a dose-dependent inhibition of the LPS-induced maturation of DCs by HZ, as seen by the reduced levels of MHC class II, CD40 and CD86. Taken together, these results indicate that HZ, rather than pRBC membranes, is a key factor involved in the suppression of murine DC function in vitro and in vivo.

The above experiments clearly implicated HZ-mediated suppression of DC function in the failure of antibody production and in the delayed acquisition of protective immunity seen during malaria infection in vivo. To examine the functional consequences of malaria on DC function directly, we examined the ability of affected DCs to activate naive, OVA-specific T-cell receptor-transgenic T cells. These cells allow monitoring of the antigen-specific CD4⁺ T-cell response as their antigen specificity is known and they can be tracked using a clonotypic antibody directed against their T-cell receptor. One of the earliest cell-surface antigens expressed by T cells following activation is CD69, which is detectable within an hour of ligation of the T-cell receptor complex 48. Interestingly, there was no significant difference in the percentage of OVA-specific T cells expressing CD69 between RBC-treated and pRBC-treated groups (Figure 7a), showing that T cells interacting with modulated DCs are equally activated.

To investigate whether treatment of DCs with pRBCs could alter the dynamics of the T-cell proliferative response, we harvested the T cells at 48, 72, 96 and 120 hours of culture (Figure 7b). The ability of pRBC-treated DCs to induce T-cell proliferation was dramatically reduced compared to the control group throughout the observation period (see Figure 7b). Analysis of T-cell production of IL-2, IL-5, IL-10 and IFN-γ (Figure 7c-f) revealed that each cytokine was downregulated in the pRBC-treated groups compared with RBC-treated controls. The observed reduction in T-cell proliferation and cytokine production could not be explained by T-cell death, as similar levels of necrotic and apoptotic T cells were detected in both conditions by FACS analysis of propidium iodide and annexin staining (data not shown). Thus, although DCs pre-treated with pRBCs in vitro can induce initial activation of naive CD4⁺ T cells, causing upregulation of CD69, these T cells fail to proliferate effectively and have a reduced ability to secrete effector cytokines.

To characterize fully the downstream effects of HZ-induced DC hyporesponsiveness on heterologous immune responsiveness, we directly investigated antigen-specific T-cell responses in vivo by transferring traceable OVA-specific CD4⁺ T cells into recipient mice 49. This allows us to follow the response of a small, but detectable, number of antigen-specific CD4⁺ T cells during the induction of an adaptive immune response. Cells were transferred into recipients following the initial peak of parasitemia (day 12 of infection) and immunized 1 day later. In P. chabaudi-infected immunized mice, OVA-specific CD4⁺ T cells underwent a similar initial activation to that of antigen-specific T cells in uninfected mice, as estimated by upregulation of CD69 (Figure 8a) and increased size/blastogenesis (also an indicator of activation; data not shown), confirming the in vitro observation (see Figure 7a). Antigen-specific CD4⁺ T cells in lymph nodes and in spleen failed to expand to the same extent in infected mice as in uninfected controls (Figure 8b), however, partly because the antigen-specific CD4⁺ T cells made a reduced number of divisions (Figure 8c). Interestingly, we did not detect increased apoptosis (determined by annexin staining) of OVA-specific T cells transferred into malaria-infected individuals (data not shown).

One of the most significant components of CD4⁺ T-cell effector function is migration into primary lymphoid follicles to interact with, and provide help for, antigen-specific B cells 50. To track the effect of malaria infection on these populations, we transferred B-cell-receptor transgenic B cells specific for hen egg-white lysozyme (HEL) taken from the MD4 transgenic mouse, together with the OVA-specific DO11.10 T cells and immunized with OVA coupled to HEL 51. B-cell expansion in uninfected animals peaked 5 days after immunization (Figure 9a). Expansion of HEL-specific B cells was almost completely ablated in P. chabaudi-infected animals immunized with OVA-HEL/LPS (Figure 9a), however, suggesting a defect in B-cell activation and/or T-cell help in infected mice.

Optimal expansion of antigen-specific B cells requires their cognate interaction with CD4⁺ T cells 5253; we therefore examined the localization of OVA-specific CD4⁺ T cells following immunization. Five days after immunization of uninfected mice, clonal expansion of antigen-specific T cells was evident by immunohistochemistry, and these cells had begun to migrate into B-cell follicles (Figure 9b). In P. chabaudi-infected mice, however, not only were there reduced numbers of OVA-specific T cells, but these cells were almost completely excluded from B-cell follicles (see Figure 9b). Migration was quantified using laser-scanning cytometry 5455. As shown in Figure 9c, the average proportion of OVA-specific CD4⁺ T cells in B-cell follicles was significantly reduced in P. chabaudi-infected mice following immunization compared with uninfected, immunized animals. This demonstrates that in infected animals, CD4⁺ T cells fail to migrate into follicles and therefore fail to interact with antigen-specific B cells, both essential steps in the induction of protective, adaptive immunity. Importantly, these failures in T- and B-cell function are similar to situations in which immune tolerance is induced 565758, although few studies have directly examined the interaction of T and B cells in vivo during the course of infections. It should be noted that, in contrast to the early inflammatory stage of the infection, normal lymphoid architecture was apparent when these defects in migration were observed, with clearly distinguishable, intact B- and T-cell areas (see Figure 9b).

To investigate whether the failure of T-cell migration into follicles was simply due to a potential alteration in lymphoid architecture or chemokine gradients caused by malaria infection, OVA-specific CD4⁺ T cells were initially activated in vitro with bone-marrow-derived DCs and pRBCs and then transferred into uninfected recipient mice. Expansion of T cells stimulated in vitro with DCs cultured with infected RBCs was reduced following transfer compared with expansion of OVA-specific T cells activated in the presence of uninfected erythrocytes (Figure 9d), and these cells also failed to migrate into B-cell follicles (data not shown).

To address whether the defect in T-cell migration was due to their activation in the context of parasite infection or to modulation of DC function, highly purified DCs from the spleens of malaria-infected animals or uninfected controls were pulsed with OVA and transferred into naive BALB/c recipient mice along with DO11.10 T cells labeled with the fluorescent dye 5,6-carboxy-succinimidyl-fluorescein ester (CFSE). Following transfer of OVA-pulsed DCs from uninfected mice, T cells divided efficiently, as seen by CFSE dilution (Figure 9e). In mice transferred with OVA-pulsed DCs purified from the spleens of malaria-infected mice, however, OVA-specific CD4⁺ T cells failed to divide to the same extent (see Figure 9e), suggesting that DCs exposed to malaria parasites are less capable of inducing effective T-cell responses. Furthermore, DCs purified from in vitro culture with pRBCs and pulsed with antigen were less able to induce an optimal T-cell response upon transfer to uninfected recipients (data not shown). Thus the defect in T-cell function and migration is primarily due to the modulation of DC function by the malaria parasite, and not simply the result of activation of T cells in the context of infection.

Female BALB/c mice were purchased from Harlan Olac (Bicester, UK). DO11.10 mice, with CD4⁺ T cells specific for the OVA_323-339 peptide in the context of the MHC class II molecule I-A^d recognized by the KJ1.26 clonotypic antibody 96 were obtained originally from N. Lycke, University of Göteborg, Sweden. MD4 mice containing HEL-specific B cells 51 were backcrossed onto the BALB/c background. All mice were maintained at Biological Services, University of Glasgow, under specific pathogen-free conditions and first used between 6 and 8 weeks of age in accordance with local and UK Home Office regulations.

To initiate a malaria infection, mice were inoculated with 1 × 10⁶ P. chabaudi AS-infected erythrocytes intra-peritoneally. Parasitemia was monitored by thin blood smears stained with Giemsa's stain. Peak parasitemia occurred at 5-6 days post-infection, after which time parasite levels declined and remained at low but usually detectable levels for the remainder of the experiments (see Figure 1a), as previously described 97. Infected mice were held in a reverse light/dark cycle so that parasites harvested at 08:00 h were at the late trophozoite stage. For studies in vitro, blood was collected when parasitemia was 30-40%. Infected blood was recovered into heparin (10 IU/ml) by cardiac puncture and diluted in phosphate-buffered saline (PBS; Invitrogen, Paisley, UK) to the required concentration of pRBCs.

At various times following malaria infection, mice were immunized intravenously with 500 μg OVA (Sigma-Aldrich, Poole, UK), or a conjugate of OVA and HEL (Biozyme, Gwent, UK) 50, along with 50 ng LPS (from Salmonella equi-abortus; Sigma-Aldrich).

DCs were prepared from bone marrow as previously described 98. Cell suspensions were obtained from femurs and tibias of female BALB/c mice. The bone-marrow cell concentration was adjusted to 5 × 10⁵ cells/ml and cultured in six-well plates (Corning Costar, New York, USA) in complete RPMI (cRPMI: RPMI 1640 supplemented with L-glutamine (2 mM), penicillin (100 U/ml), streptomycin (100 μg/ml) (all from Invitrogen) and 10% fetal calf serum (FCS; Labtech International, Ringmer, UK) containing 10% of culture supernatant from X63 myeloma cells transfected with mouse granulocyte-macrophage colony stimulating factor (GM-CSF) cDNA. Fresh medium was added to the cell cultures every 3 days. On day 6, DCs were harvested and cultured at the required concentration for each individual experimental procedure, as described below. This technique generated a large number of CD11c⁺ DCs largely free from granulocyte and monocyte contamination, as previously described 98.

Blood from P. chabaudi AS-infected mice was washed twice in PBS before being resuspended in cRPMI for addition to DCs. For fixation, infected blood was washed three times in PBS and resuspended in 0.5% paraformaldehyde for 30 min at 4°C. Fixed erythrocytes were then washed in PBS, resuspended in 0.06% Gly-Gly (Sigma-Aldrich) for 5 min at 4°C and washed twice more in PBS before being resuspended in cRPMI for addition to DC. After 24 h culture, DCs were stimulated with 1 μg/ml LPS and the expression of cell-surface molecules was analyzed 18 h later by flow cytometry. To confirm complete fixation, we showed that 2 × 10⁷ fixed, infected erythrocytes could not establish infection when injected intra-peritoneally into a female BALB/c mouse.

The cell lines 3T3-CD40L and 3T3-SAMEN 46 were kind gifts from P. Hwu (NCI, Bethesda, USA). Cells were grown in cRPMI in T75 tissue culture flasks (Helena Biosciences, Gateshead, UK) and, when confluent, harvested and distributed in six-well plates at 2.5 × 10⁵ cells/ml of cRPMI. Bone-marrow-derived DCs were cultured with infected or uninfected erythrocytes at a ratio of 1:100. After 24 h, DCs were harvested, resuspended at 1 × 10⁶ cells/ml and cultured in a 1:1 ratio with either 3T3-CD40L or 3T3-SAMEN cells for a further 24 h. The level of CD40 expression on DCs was analyzed by flow cytometry and culture supernatants collected for IL-12 cytokine analysis.

Bone-marrow DCs were centrifuged at 450 × g, resuspended at 1 × 10⁶ cells/ml and 500 μl aliquots were distributed into 24-well tissue culture plates (Corning Costar) with pRBCs or RBCs. After 24 h incubation at 37°C in 5% CO₂, DCs were antigen-loaded for 6 h with 5 mg/ml OVA (Worthington Biochemical, Freehold, USA). OVA-specific T cells were isolated from the mesenteric and peripheral lymph nodes of DO11.10 transgenic mice 96 on the SCID background and cultured at a 1:1 ratio with DCs. T-cell proliferation was assessed after 48, 72, 96 and 120 h of culture and assessed by incorporation of [³H]thymidine (0.5 μCi/well) for the last 24 h of culture. Cells were harvested using a Betaplate 96-well harvester (Wallac Oy, Turku, Finland) and [³H]thymidine incorporation measured on a Betaplate liquid scintillation counter (Wallac).

For the detection of IL-12 (p40 and p70) and IL-10, OptEIA™ enzyme-linked immunosorbent assay (ELISA) kits (Becton Dickinson, Oxford, UK) were used according to the manufacturer's instructions. For T-cell cytokines, Mouse Th1/Th2 6-Plex kit (Biosource, Nivelles, Belgium) was used according to the manufacturer's instructions. For analysis of cytokine production ex vivo, single-cell suspensions of spleen cells were prepared by rubbing through Nitex mesh (Cadisch & Sons, London, UK) in RPMI 1640 medium. After washing, cells were resuspended at 4 × 10⁶ cells/ml in cRPMI, either alone or with 1 mg/ml OVA or 5 μg/ml concanavalin A (ConA; Sigma-Aldrich) and supernatants sampled after 48 h. These were stored at -20°C until analysis by standard sandwich ELISA protocol (antibodies used: for IFN-γ capture, R4-6A2; for IFN-γ detection, XMG1.2; for IL-5 capture, TRKF5; for IL-5 detection, TRKF4; Pharmingen, Oxford, UK) and the levels of cytokine in supernatants calculated by comparison with recombinant cytokine standards (R & D Systems, Abingdon, UK).

Aliquots of 1 × 10⁶ cells in 12 × 75 mm polystyrene tubes (Falcon BD, Oxford, UK) were resuspended in 100 μl FACS buffer (PBS, 2% FCS and 0.05% NaN₃) containing Fc Block (2.4G2 hybridoma supernatant) as well as the appropriate combinations of the following antibodies: anti-CD4-PerCP (clone RM4-5), anti-CD11c-PE (clone HL3), anti-CD40-FITC (clone 3/23), anti-CD69-PE (clone H1.2F3), anti-CD80-FITC (clone 16-10A1), anti-CD86-FITC (clone GL1), anti-MHC-II (clone 2G9), anti-B220-PE (clone RA3-6B2), PE-hamster IgG isotype control, FITC-rat IgG2a, κ isotype control and FITC-hamster IgG1, λ isotype control (anti-TNP) (all Pharmingen), biotinylated KJ1.26 antibody or biotinylated HEL. Biotinylated antibodies were detected by incubation with fluorochrome-conjugated streptavidin (Pharmingen). After washing, samples were analyzed using a FACSCalibur flow cytometer equipped with a 488 nm argon laser and a 635 nm red diode laser and analyzed using CellQuest software (both BD BioSciences, Oxford, UK).

Ghosts from infected and uninfected erythrocytes were generated as previously described 67. Briefly, blood was collected into heparin by cardiac puncture and washed three times in PBS. Infected and uninfected erythrocytes were concentrated in PBS supplemented with 113 mM glucose (Sigma-Aldrich) and 3% FCS. Infected erythrocytes were incubated in an equal volume of glycerol buffer (10% glycerol (Sigma-Aldrich) supplemented with 5% FCS in PBS) for 1 h at 4°C. Parasites and ghosts were separated in a continuous Percoll (Amersham Biosciences, Little Chalfont, UK) gradient (ρ: 1.02-1.10 g/cm³) in intracellular medium buffer (IM: 20 mM NaCl, 120 mM KCl, 1 mM MgCl₂, 10 mM glucose, 5 mM Hepes pH 6.7) by centrifugation at 5,000 × g for 30 min. Ghosts were then washed in IM buffer and layered on a two-step Percoll gradient (ρ: 1.01+1.02 g/cm³) to separate them from ghosts that might still contain parasites. Ghosts from uninfected erythrocytes were obtained by adding a 40-fold volume of phosphate buffer (5 mM NaH₂PO₄/Na₂HPO₄, 1 mM PMSF, 0.01% azide, pH 8.5). The suspension was centrifuged at 32,000 × g for 30 min. Ghosts from infected and uninfected erythrocytes were then washed three times in PBS before being resuspended in cRPMI for addition to DCs at a ratio of 100:1.

HZ was isolated from supernatants obtained from cultures of P. falciparum gametocytes, kindly provided by Lisa Ranford-Cartwright, (Division of Infection and Immunity, University of Glasgow, UK). Endotoxin-free buffers and solutions were used throughout. Supernatants were centrifuged for 20 min at 450 × g. The pellet was washed three times in 2% SLS and resuspended in 6 M guanidine HCl. Following 5-7 washes in PBS, the pellet was resuspended in PBS and sonicated for 90 min using Soniprep 150 (Sanyo Scientific, Bensenville, USA) at an amplitude of 5-8 μm to minimize aggregation and maintain the HZ in suspension. Total heme content was determined as previously described 99 by depolymerizing heme polymer in 1 ml of 20 mM NaOH and 2% SDS, incubating the suspension at room temperature for 2 h and then reading the optical density at 400 nm using a UV-visible Helios spectrophotometer (Thermo Spectronic, Cambridge, UK). DCs were pulsed with 1-20 μM HZ - a concentration range similar to that seen when DCs were cultured at a 1:100 ratio with pRBCs.

Peripheral blood was collected and the plasma was separated by centrifugation at 450 × g for 10 min and stored at -20°C until analysis. OVA-specific IgG was measured by standard sandwich ELISA using a peroxidase-conjugated anti-mouse total IgG (Sigma-Aldrich).

Lymph nodes and spleens were homogenized and the resulting cell suspensions washed twice by centrifugation at 400 × g for 5 min and resuspended in RPMI. The proportions of antigen-specific T cells were evaluated by flow cytometry, and syngeneic recipients received 3 × 10⁶ antigen-specific cells. In some experiments, cells were labeled with the fluorescent dye CFSE (Molecular Probes, Oregon, USA) immediately before use 100. The level of CFSE in cells was analyzed by flow cytometry and expressed as the mean proportion of antigen-specific T cells under each CFSE peak.

Spleens were frozen in liquid nitrogen in OCT embedding medium (Miles, Elkart, USA) in cryomoulds (Miles) and stored at -70°C. Tissue sections (8 μm) were cut on a cryostat (ThermoShandon, Cheshire, UK) and stored at -20°C. Sections were blocked and stained as previously described 55, using B220-FITC to stain B-cell areas and biotinylated-KJ1.26 to detect OVA-specific DO11.10 cells, and visualized using Streptavidin-Alexa Fluor 647 (Molecular Probes). All photographs were taken at 20× magnification.

To visualize HZ deposition in DCs, cells were photographed using an Axiovert S-100 Zeiss microscope using a 63× oil-immersion lens by normal bright-field imaging. To image HZ in splenic DC, 8 μm sections were cut as described above and stained with biotinylated-CD11c followed by Streptavidin-HRP and finally tyramide-488 (PerkinElmer, Boston, USA). Images were then taken of bright-field and green fluorescence and the images merged by inverting and then false coloring the bright-field image such that deposited HZ appeared red and CD11c appeared green.

Sections were stained as described above. Sections were then scanned on a laser-scanning cytometer equipped with argon, helium, neon, and ultraviolet lasers (Compucyte, Cambridge, USA) and visualized with the Openlab imaging system (Improvision, Coventry, UK). The localization of transgenic T cells and B-cell follicles were plotted. Using these tissue maps the number of transgenic T cells in defined gates was calculated for three gates in periarteriolar lymphoid sheath (PALS) and three B-cell follicle gates per section. Data are plotted as the mean proportion of transgenic T cells in each gate relative to the number of transgenic T cells in the entire section and are the mean of triplicate readings from three mice per group.

Spleens were excised and single-cell suspensions obtained as described above. In some experiments, cells were stimulated with 1 μg/ml LPS for 18 h before analysis by flow cytometry. To obtain purified DCs from spleens of mice, single-cell suspensions were labeled using a CD11c isolation kit (Miltenyi Biotec, Bisley, UK) according to the manufacturer's instructions. DCs were then purified using two MS magnetic columns (Miltenyi Biotec) and found to be 85-95% pure by flow cytometric analysis.

Results are expressed as mean ± standard error or standard deviation as indicated. Significance was determined by one-way ANOVA in conjunction with the Tukey test using Minitab. A p-value of less than 0.05 was considered significant.