Biogenesis of Leishmania major-Harboring Vacuoles in Murine Dendritic Cells

Freshly prepared DC were generated from the bone marrow of 6- to 8-week-old BALB/c mice (Charles River Breeding Laboratories, Sulzfeld, Germany) as described previously (20, 28). Briefly, 2 × 10⁶ bone marrow cells derived from the tibia and femur of the hind legs were incubated in 10 ml RPMI 1640 medium (Invitrogen, Karlsruhe, Germany) supplemented with 10% (vol/vol) heat-inactivated fetal calf serum, 2 mM l-glutamine, 10 mM HEPES buffer, 0.05 mM β-mercaptoethanol, 100 U/ml penicillin, and 20 μg/ml gentamicin (culture medium) in the presence of 200 U/ml granulocyte-macrophage colony-stimulating factor (GM-CSF; PeproTech, London, United Kingdom). Cultures were fed with GM-CSF on day 3 if maintained for a period of 6 days and were fed again on day 6 or on days 6 and 8 if maintained for a period of 8 or 10 days, respectively. The bone marrow-derived DC (BMDC) were harvested on day 6, 8, or 10 and were allowed to adhere to glass chamber slides (Nunc, Wiesbaden, Germany) for 3 to 4 h. The cells were shown to have a typical DC morphology with a myeloid DC phenotype (MHC class II⁺, CD80⁺, CD86⁺, CD11c⁺, CD205⁺, B220⁻, CD8⁻). For activation, BMDC harvested on day 8 were incubated overnight with lipopolysaccharide (LPS; Sigma-Aldrich, Taufkirchen, Germany) at 1 μg/ml, CpG oligodeoxynucleotide (ODN) 1668 (5′-TCC ATG ACG TTC CTG ATG CT-3′) or the control AT-rich ODN (non-CpG ODN, 5′-ATT ATT ATT ATT ATT ATT AT-3′) (QIAGEN Operon, Cologne, Germany) at 100 μg/ml, CD40 ligand (PeproTech) at 1 μg/ml, or interleukin 10 (IL-10; PeproTech) at 20 μg/ml. A mouse DC line derived from fetal skin (FSDC) (12) was passaged in culture medium. All experiments were done with subconfluent cultures.

The cloned virulent L. major isolate (MHOM/IL/81/FE/BNI) was maintained by passage in BALB/c mice. Amastigotes were isolated from skin lesions as described previously (4). Promastigotes were grown in vitro in blood agar cultures. For analysis by confocal microscopy, parasites were stained with green fluorescent cell tracker 5-chloromethylfluorescein diacetate (CMFDA) or red fluorescent cell tracker 5- and 6-4-chloromethylbenzoylaminotetramethylrhodamine (CMTMR) (Molecular Probes, Leiden, Netherlands) prior to their use for infection according to the manufacturer's instructions. Briefly, stationary-phase promastigotes were incubated at 27°C and amastigotes at 37°C for 45 min with 30 μM CMFDA or 10 μM CMTMR. After being washed, they were incubated for an additional 30 min at the appropriate temperature in fresh culture medium. The promastigotes were not complement opsonized. For some experiments, stationary-phase promastigotes were heat killed at 80°C for 30 min after the staining procedure. FSDC and BMDC were treated with stationary-phase promastigotes at a ratio of 2:1 or with amastigotes at a ratio of 4:1. After 30 min of exposure to parasites at 37°C in a humidified atmosphere with 5% CO₂, unbound extracellular parasites were removed by washing the cells three times with phosphate-buffered saline (PBS). To analyze PV biogenesis, DC were examined for the expression of endosomal marker proteins either immediately (0.5-h value) or after the addition of fresh culture medium and incubation for a further 0.5 to 23.5 h (1-, 2-, 4-, and 24-h values), using intracellular immunostaining and confocal microscopy. The average number of parasites per infected DC was determined by counting 100 to 200 immature DC (6 days) or mature DC (10 days) at 24 and 72 h after pulse infection with stationary-phase promastigotes for 30 min. The parasite load was quantified by staining with acridine orange (Sigma-Aldrich) and ethidium bromide (Sigma-Aldrich) and analysis by fluorescence microscopy at 495 nm (26).

For analysis of the fusion ability of PV with lysosomes, DC seeded into chamber slides (6 × 10⁵ cells in 0.4 ml of culture medium per chamber) were exposed to CMFDA-stained L. major parasites. After 30 min, free extracellular parasites were removed by repeated washing with PBS, and 1 μM LysoTracker Red DND-99 (Molecular Probes), preferentially labeling lysosomes as the main acidic compartments, was added for 30 min. Thereafter, the supernatants were removed from each well, and fresh culture medium was added for the remaining culture period of 23 h. Alternatively, DC were incubated for 30 min at 37°C with 1 mg/ml lysine-fixable tetramethylrhodamine-coupled dextran (M_r, 10,000 or 70,000) (Molecular Probes), washed, and then cultured overnight in tracer-free culture medium to allow the accumulation of fluorescent tracer molecules in lysosomes. Subsequently, the cells were exposed to CMFDA-stained parasites at 37°C for 30 min. Thereafter, they were washed with PBS to remove free parasites and were cultured for a further 23.5 h.

At different time points (0.5, 1, 2, 4, or 24 h) after DC exposure to stationary-phase promastigotes for 30 min (see above), the cells were washed with PBS containing 0.25 mM MgCl₂ and 0.35 mM CaCl₂, fixed with 4% paraformaldehyde, and then permeabilized with 0.2% saponin in blocking solution (10% goat serum in PBS containing Mg²⁺ and Ca²⁺). Thereafter, DC were incubated for 45 min at room temperature in a humid chamber with one of the following primary antibodies in saponin-containing blocking solution: monoclonal antibody (MAb) C2, a rat immunoglobulin G1 (IgG1) recognizing mouse CD71 (transferrin receptor) (BD Pharmingen, Heidelberg, Germany), fluorescein isothiocyanate-conjugated rat anti-mouse LAMP-2 MAb (BD Pharmingen), rabbit anti-mouse Rab4 antibodies (Santa Cruz, Santa Cruz, Calif.), rabbit anti-mouse Rab7 antibodies (Santa Cruz), phycoerythrin-conjugated rat anti-mouse LAMP-1 MAb (Santa Cruz), and biotin-conjugated rat anti-mouse CD68 (macrosialin) MAb (Serotec, Oxford, United Kingdom). Isotype-matched control antibodies were purchased from BD Pharmingen and DakoCytomation (Glostrup, Denmark). After treatment with the primary antibodies, the cells were washed three times with PBS containing Ca²⁺ and Mg²⁺ and were stained with the secondary reagents streptavidin-Alexa Fluor 546 (Molecular Probes) or rhodamine-conjugated goat anti-rat or goat anti-rabbit antibody (Dianova, Hamburg, Germany) in saponin-containing blocking solution for 45 min at room temperature. Cells were mounted in fluorescent mounting medium (DakoCytomation) before observation with a LSM 510 Zeiss confocal laser scanning microscope. Confocal microscopy images were acquired by use of LSM 510 software and analyzed by use of LSM 3.2 image browser before being exported to Adobe PhotoShop. For each experiment and at each time point, the percentages of parasite-containing compartments displaying endocytic marker molecules or having fused with lysosomes were determined after 50 to 300 PV were counted.

In freshly prepared BMDC populations, DC were identified by labeling with MAb HL3, a hamster IgG1 directed against mouse CD11c (BD Pharmingen) and cytochrome5-conjugated goat anti-hamster IgG (Dianova) before permeabilization for intracellular staining.

Unpaired Student's t test was used for statistical analyses. Differences were considered significant when P was <0.05.

The phagocytic activity of DC varies with their maturation stage. While mature DC have a strongly reduced ability to internalize particles, DC precursors and immature DC efficiently take up microorganisms, such as L. major. For the characterization of L. major-containing compartments in DC, we therefore used the cell line FSDC (representing myeloid DC progenitors arrested at an early stage of development) and freshly prepared immature DC (BMDC that had been cultured in the presence of GM-CSF for 8 days). The features of PV harboring L. major in DC were evaluated according to the presence of endosomal marker proteins using intracellular immunostaining and confocal immunofluorescence microscopy. Freshly prepared DC were identified by cell surface labeling of CD11c, and parasites were visualized after prestaining with CMFDA or CMTMR.

After infection of DC with L. major promastigotes, all parasites were located in membrane-bound compartments. The outline of the phagosome membrane was largely congruent with the shape of the parasites. Late endosomal/lysosomal molecules could be detected in the majority of PV membranes of both DC precursors and immature DC (Fig. 1). Distinct red rims indicated the presence of CD68 (Fig. 1A, 1 to 5, and Fig. 1B, 1 to 4) and LAMP-1 (Fig. 1A, 6 to 9, and Fig. 1B, 5 to 8) in the membranes of L. major-containing organelles in FSDC and BMDC. For examination of the PV biogenesis, BMDC and FSDC were evaluated at 0.5, 1, 4, or 24 h after infection with L. major. The results of the immunofluorescence analysis showed that the early endosomal marker CD71 was detectable in few PV (about 10%) only within 30 min (BMDC) or 4 h (FSDC) after infection (Fig. 2). At 24 h, virtually none of the PV in BMDC (<1%) and less than 3% of the PV in FSDC contained this early marker. In contrast, the majority of PV in both FSDC and BMDC expressed LAMP-1 and LAMP-2 proteins already in the early phase of infection. Thus, PV maturation in L. major-infected DC seems to occur very fast. The rapid formation of late endosomal compartments was most pronounced in BMDC, as indicated by the early expression of high levels of LAMP-1, LAMP-2, and CD68 (Fig. 2A), while lower amounts of these markers were detected in FSDC (Fig. 2B). On the other hand, a higher proportion of PV in FSDC was positive for Rab7, another late endosomal molecule, which is involved in the fusion of late endosomes with lysosomes (5, 22). Of note, the immortalized line FSDC has been shown to display the morphology and surface marker profile suggestive of a myeloid progenitor cell with features of both DC and macrophages (12).

The results described above demonstrated that L. major-containing PV in a DC precursor line resemble those in immature DC with regard to their ability to fuse rapidly with late endocytic compartments. Next, we compared the characteristics of PV in freshly prepared immature and mature BMDC at different times after infection with L. major.

High levels of the late endosomal markers CD68, LAMP-1, and LAMP-2 were present in the parasite-containing organelles of both immature and mature DC (Fig. 3), even at the earliest time points after infection, indicating that the rapid formation of late endosomal compartments harboring L. major in DC is independent of the stage of DC maturation. The most notable difference between mature and immature BMDC was the rapid acquisition of Rab7 by L. major-containing PV in mature DC (Fig. 3B). While Rab7, a major effector of late endosome/lysosome transformation, could not be detected in PV of immature DC at any of the studied time points after infection (30 min to 4 h), 22% of the parasite-harboring compartments in mature DC displayed this molecule already at 30 min postinfection. At 1 h, the percentage of Rab7-positive PV had decreased considerably, and only background levels were detectable thereafter (Fig. 3B). The absence of Rab7 in immature BMDC suggests an arrest of their PV biogenesis at the stage of late endosomes.

The properties of DC not only change during their maturation but also vary depending on their activation status. To determine the effect of DC activation on the biogenesis of L. major promastigote-containing PV, BMDC were exposed to the stimulus LPS, CpG ODN, or CD40 ligand and were compared with nonstimulated BMDC with regard to the presence of endosomal/lysosomal marker proteins in the PV at 24 h after infection. Furthermore, the effect of IL-10, a cytokine promoting the development of tolerance-inducing DC, was examined. The results demonstrated that none of these DC treatments modified the biogenesis of L. major-harboring PV in BMDC (data not shown). Together, our findings document that PV biogenesis is influenced by the stage of maturation but not the activation status of DC.

The finding that Rab7, a key molecule controlling the formation of phagolysosomes, is not present in PV of immature BMDC and can be detected only during the initial phase of infection in PV of mature BMDC (see above) (Fig. 3) suggested that L. major modulates the fusogenic activity of phagosomes in DC. To test this assumption, BMDC at different stages of maturation were infected with CMFDA-labeled L. major promastigotes, and the lysosomes were stained with LysoTracker dye. Subsequently, the fusion activity of the PV with lysosomes was analyzed by confocal microscopy at 24 h after infection. L. major-harboring PV from immature BMDC rarely fused with lysosomes (Fig. 4, panels A and B1); only about 5% of the lysosomal compartments in BMDC that had been cultured for 6 days colocalized with promastigotes. During further DC maturation, the percentage of fusion events increased and reached a level of about 18% (Fig. 4, panels A and B2 to B5). The poor fusion activity of promastigote-containing organelles with lysosomes in immature DC cannot be explained by a delay in PV maturation, because fusion events were almost absent even after 2 and 3 days of infection (data not shown).

Interestingly, the number of parasites per infected DC at 72 h after pulse infection with promastigotes was slightly higher in immature BMDC than in mature BMDC (immature BMDC, 2.56 parasites per DC; mature BMDC, 2.23 parasites per DC), while it was almost identical at 24 h (immature BMDC, 1.62 parasites per DC; mature BMDC, 1.66 parasites per DC). This result indicates that the maturation stage of DC affects the intracellular survival of parasites.

To evaluate whether the incompetence of L. major-harboring PV to fuse with lysosomes depends on parasite viability or life cycle stage, immature BMDC or FSDC were exposed to live promastigotes, heat-killed promastigotes, or live amastigotes. In contrast to PV containing live promastigotes, a large proportion of PV containing heat-killed promastigotes (about 70%) colocalized with lysosomes in both BMDC and FSDC (Fig. 5). Moreover, the fusogenic activity of PV containing amastigotes was much higher than that of PV containing live promastigotes (34% compared to 10% in freshly prepared DC). The percentage of fusion events was higher (20%) in the DC progenitor line, FSDC, that had been exposed to live promastigotes, a finding that correlated with the higher level of Rab7 in the PV of these cells (see above) (Fig. 2). These observations suggest that after internalization of live promastigotes, the maturation of PV in DC is strongly reduced and the vast majority of PV is arrested at the stage of late endosomes. In contrast, a considerable proportion of PV containing amastigotes fuse with lysosomes.