Activated-leukocyte cell adhesion molecule
DC-specific ICAM-grabbing non-integrin
Dendritic cells (DC) that express the type II C-type lectin DC-SIGN (CD209) are located in the submucosa of tissues, where they mediate HIV-1 entry. Interestingly, the pathogen Candida albicans,the major cause of hospital-acquired fungal infections, penetrates at similar submucosal sites. Here we demonstrate that DC-SIGN is able to bind C. albicans both in DC-SIGN-transfected cell lines and in human monocyte-derived DC. The binding was shown to be time- as well as concentration-dependent, and live as well as heat-inactivated C. albicans were bound to the same extent. Moreover, in immature DC, DC-SIGN was able to internalize C. albicans in specific DC-SIGN-enriched vesicles, distinct from those containing the mannose receptor, the other known C. albicans receptor expressed by DC. Together, these results demonstrate that DC-SIGN is an exquisite pathogen-uptake receptor that captures not only viruses but also fungi.
Epithelial surfaces form the first line of defense against microbes. A small proportion of incoming microbes that enter at sites of microlesions is handled by APC, in particular DC 1. Among pathogens that invade mucosal surfaces, Candida albicans is among the most frequently isolated from humans 2. As C. albicans is the major cause of hospital-acquired fungal infections 3, its recognition by cell surface receptors has major pathogenetic consequences. Nevertheless, only limited information is available on themolecular mechanisms involved in recognition of this fungus. C. albicans can switch from a unicellular yeast form into various filamentous forms, all of which can be found in infected tissues 4. The ability to reversibly switch between these forms is thought to be important for C. albicans virulence. Several studies now demonstrate that DC can bind and phagocytose fungi such as C. albicans5–7. Protection from mucocutaneous candidiasis clearly relies on cell-mediated immunity induced after DC process C. albicans and then present antigens that prime T cells 8.
Recent studies in mice demonstrate that whereas the yeast-form activates DC to produce IL-12 and primes Th1 cells, the hyphal-form inhibits IL-12 and Th1 priming and induces IL-4 production 5. These results indicate that DC fulfill the requirement of a cell uniquely capable of sensing the two forms of C. albicans5. In addition, it was recently reported that human DC are also able to bind C. albicans, and that this interaction is mediated by the mannose receptor (MR, CD206), which is also found on macrophages 8.
However, the observation that C. albicans, both as a commensal as well as a true pathogen, is also found in areas (sub-mucosa) highly enriched in DC-specific ICAM-grabbing non-integrin (DC-SIGN)-positive DC prompted us to investigate whether C. albicans could be bound by DC-SIGN as well. Recently, we isolated this novel C-type lectin (also designated CD209) from monocyte-derived DC. We discovered that DC-SIGN acts as a binding partner for ICAM-3, mediating the early contact between DC and T cell and therefore the initiation of primary immune responses 9. In addition, DC-SIGN displays a high affinity for ICAM-2, supporting transendothelial migration of DC and DC trafficking 10. Moreover, DC-SIGN binds and captures HIV-1 at mucosal sites of initial infection, and protects the virus from degradation for subsequent transport by DC to lymphoid organs 11.
Here we demonstrate for the first time that as well as viruses, fungi are also recognized by DC-SIGN.
2 Results and discussion
2.1 Candida albicans is a ligand of DC-SIGN
We used the erythroleukemic cell line K562 transfectants stably expressing DC-SIGN (K-DC-SIGN) 9 to investigate the potential of C. albicans to bind DC-SIGN in the absence of any other known C. albicans receptors. Binding to ICAM-3-coated fluorescent beads was used as a positive control for DC-SIGN function. K562 cells (K-ALCAM) transfected with the homotypic activated-leukocyte cell adhesion molecule (ALCAM) 12, which is expressed by DC but does not bind any of the known ligand of DC-SIGN, were used as negative control. The K562 transfectants stably express DC-SIGN and ALCAM (Fig. 1A) with expression levels similar to those observed for immature DC 9, 12.
Binding studies demonstrated that DC-SIGN clearly mediates adhesion to both C. albicans and ICAM-3–Fc-coated beads (Fig. 1B). In contrast, K-ALCAM cells bound neither ICAM-3 nor C. albicans, but did bind ALCAM–Fc-coated beads. Blocking-antibodies against DC-SIGN significantly inhibited binding of C. albicans by K-DC-SIGN; in addition, the calcium chelator EGTA completely abrogated binding (Fig. 1C). This Ca2+ dependence confirms that the C-type lectin domain of DC-SIGN mediates binding to C. albicans. With increasing concentrations of C. albicans (Fig. 1D) and increasing incubation time (Fig. 1E), binding of K-DC-SIGN to the yeast cells increased significantly.
2.2 DC-SIGN binds both live and heat-inactivated yeast forms of C. albicans
To exclude the possibility that the binding of DC-SIGN to heat-inactivated C. albicans was due to artifacts derived from the heat treatment, K-DC-SIGN were allowed to interact with both live and heat-inactivated (see Sect. 3) C. albicans yeast. As shown in Fig. 2, the percentage of binding did not increase significantly upon heat inactivation of the yeast, when compared with binding to live yeast cells. In addition, the blocking of binding by Ab against DC-SIGN was not profoundly altered by heat treatment.
2.3 Monocyte-derived DC also bind C. albicans through DC-SIGN
DC are specialized in binding and uptake of antigen 13 and recently it has been shown that the interaction between DC and C. albicans is mediated by the MR 5, 8. Our findings with K-DC-SIGN cells and the observation that C. albicans can be found in areas of the body (sub-mucosa) that are highly enriched in DC-SIGN-positive cells 11 suggested that DC-SIGN could contribute to the binding of C. albicans to immature DC.
In Fig. 3A, it is shown that human monocyte-derived immature DC are able to bind C. albicans, and that this interaction increases with time. This is in agreement with previously published work showing that DC rapidly internalize C. albicans: within 10–20 min of incubation at 37 °C, already 40–50% of the particles are ingested, reaching a maximum after 60 min 5, 8.
To determine the contribution of both MR and DC-SIGN to mediate binding of C. albicans, immature DC were incubated with specific inhibitors before interacting with the yeast particles. As shown in Fig. 3B, antibodies against DC-SIGN significantly blocked binding, though only partially (approximately 25–30%), whereas mannose, which is known to specifically inhibit the MR, was responsible for blocking about 65–70% of the binding. Interestingly, mannose did not show any significant blocking of C. albicans binding to DC-SIGN on K-DC-SIGN cells (not shown). Moreover, by combination of both anti-DC-SIGN Ab and mannose, binding of C. albicans was almost completely inhibited (80–85%). This notion is supported by the observation that EGTA blocked binding of C. albicans to DC to a similar level, also indicating that probably no other C-type lectins were involved. Furthermore, the finding that binding of C. albicans was mediated by DC-SIGN and MR, in a ratio of roughly 1:3, respectively, was in agreement with the observation that the expression of MR on the surface of immature DC appears to be higher than that of DC-SIGN (Fig. 3C). However, no real quantitative evaluation can be made.
In addition, besides DC-SIGN and the MR, DC are known to express high levels of the β2-integrin MAC-1, which has already been implicated in the binding of C. albicans to lymphocytes 6. However, we could not detect any blocking when anti-β2-integrin antibody (NKI-L19) was used, suggesting that MAC-1 is not likely to be involved in C. albicans binding on human immature monocyte-derived DC (data not shown). Moreover, the use of laminarin, reported to interfere with the interaction between another DC-specific C-type lectin, Dectin-1, and C. albicans, did not show any blocking effect either (data not shown). Together, these findings strongly suggest that binding of C. albicans to immature DC is predominantly mediated by the C-type lectins DC-SIGN and MR.
2.4 DC-SIGN mediates phagocytosis of C. albicans
We recently demonstrated that DC-SIGN can act as an antigen-uptake receptor and facilitates phagocytosis within minutes 14. To determine whether DC-SIGN could contribute also to the internalization of C. albicans by immature DC, we incubated DC with FITC-labeled C. albicans particles for 60 min at 37 °C to allow phagocytosis. Subsequently, we fixed, permeabilized, and fluorescently labeled DC with specific Ab against various receptors, and analyzed by confocal microscopy.
In general, we observed that after 1 h of incubation, about 40% of immature DC have ingested C. albicans, ranging from 1 to 9 particles (average of 3) per DC. The results in Fig. 4A show that DC-SIGN clearly co-localizes with C. albicans particles, indicating the involvement of this lectin in binding and uptake of this pathogen. By contrast, neither MAC-1 nor ALCAM were found to localize around the yeast particles (Fig. 4B, C).
Fig. 4D shows colocalization between MR and ingested Candida. In order to determine whether DC-SIGN was also able to internalize C. albicans in presence of inhibitors of MR, DC were allowed to phagocytose the yeast particles in presence of mannose. As shown in Fig. 4E, vesicles containing DC-SIGN colocalizing with FITC-labeled Candida can still be clearly observed. To prove that C. albicans were indeed ingested, we made Z-scans, unequivocally demonstrating that the yeast particles are ingested (Fig. 4F). The presence of EGTA almost completely blocked phagocytosis (data not shown).
We showed that as well as viruses (HIV-1, SIV, Ebola) 11, 15, 16 and parasites (Leishmania) 17, the C-type lectin DC-SIGN can also bind yeast (C. albicans). These observations, together with the discovery that DC-SIGN acts as an antigen-uptake receptor, provide further evidence that C-type lectins on DC are major pathogen-recognition receptors 18. Distinct from the Toll-like receptors 19, they mediate antigen uptake rather than activating DC.
Our findings clearly show that C. albicans has two major receptors on human monocyte-derived DC — DC-SIGN and MR. DC-SIGN is expressed at sites in the skin (dermis) and the mucosa 11 where C. albicans is known to enter the host. Therefore, DC-SIGN-positive DC might, trough these C-type lectin receptors, form the first encounter with these pathogens and the host immune system and, after antigen presentation, initiate a cellular response 5, 8.
It remains to be elucidated whether the destiny of the C. albicans-containing vesicles enriched in DC-SIGN is different from those enriched in MR. Though we observed some vesicles containing both receptors (not shown), it was intriguing that colocalization of both C-type lectins seemed not to occur in most C. albicans-containing vesicles. Therefore, it will be interesting to characterize these vesicles in more detail (this work is in progress in our laboratory) and to investigate if both lectins recognize similar or distinct carbohydrate moieties on C. albicans. This is of particular interest because MR is known as a recycling receptor, whereas DC-SIGN targets much deeper in the endosomal compartments 14.
Detailed knowledge of the recognition receptors for C. albicans on DC, together with the specific downstream cellular events initiated by receptor engagement, may increase our understanding of possible immune dysfunctions in patients with mucocutaneous candidiasis and may offer new targets for immunotherapy of candidal infections and diseases.
3 Material and methods
3.1 Reagents and antibodies
FITC was from Fluka. Mannan and D-mannose were from SIGMA Chemical Co., St. Louis, MO. IL-4 and GM-CSF used for culturing monocyte-derived DC were from Schering-Plough (International, Kenilworth, USA). The following antibodies were used: AZN-D1, AZN-D3 (mouse IgG1, anti-DC-SIGN 11); AZN-L50 (mouse IgG1, anti-ALCAM) 12; NKI-L19 (mouse IgG1 anti-β2 integrins); mAb clone 19.2 against MR was from BD Biosciences PharmIngen; the directly allophycocyanin-conjugated mAb against CD45RO was from Becton Dickinson; Cy5-conjugated goat-anti-mouse IgG was from Molecular Probes.
3.2 C. albicans culture conditions
C. albicans, strain UC820, a clinical isolate that has been well described 20, was maintained on agar slants at 4°C. Previous experiments showed that strain UC820 can develop hyphae and pseudohyphae in vitro and in vivo to the same extent as a panel of virulent control strains. C. albicans UC820 was inoculated into 100 ml of Sabouraud broth and was cultured for 24 h at 37°C. After three washes with pyrogen-free saline by centrifugation at 1500×g, the number of yeast cells was counted in a hemocytometer; occasional strings of two yeasts were counted as one colony-forming unit of C. albicans. The suspension was diluted to the appropriate concentration with pyrogen-free saline. Microscopy confirmed that the suspension consisted of blastoconidia. When necessary, the blastoconidia were heat-killed either at 56°C for 1 hr or at 100°C for 30 min.
Immature DC were generated from human peripheral blood monocytes as described previously 9. Briefly, monocytes were isolated by adherence to plastic and cultured in the presence of IL-4 (500 U/ml) and GM-CSF (800 U/ml) for 6–7 days. K562 transfectants either expressing DC-SIGN or ALCAM were generated by transfection of K562 cells with 10 μg of plasmid by electroporation as described previously 11, 12. Positive cells were sorted several times to obtain stable transfectants with expression levels of DC-SIGN and ALCAM similar to immature DC.
Labeling of Candida cells was performed as follows: yeast cells were resuspended to 2×108/ml, in 0.01 mg/ml FITC in 0.05 M carbonate-bicarbonate buffer (pH 9.5). After incubation for 15 min at room temperature in the dark, FITC-labeled Candida cells were washed twice in PBS containing 1% BSA (PBA buffer), heat-killed for 60 min at 56°C, and subsequently analyzed by flow cytometry.
DC were stained in PBA with primary antibodies and FITC-conjugated secondary antibodies and were analyzed by flow cytometry using the FACScalibur (BD Biosciences, Mountain View, CA). Isotype-matched controls were included.
3.5 C. albicans binding studies
DC or transfected K562 cells were stained with anti-CD45-allophycocyanin prior to exposure to FITC-labeled live or heat-inactivated C. albicans yeast forms. Before adding C. albicans, cells were or were not preincubated for 10 min at room temperature with mannose (100 mM), EGTA (5 mM), isotype control (mouse IgG1) or a mixture of anti-DC-SIGN mAb, AZN-D1 and AZN-D3 (20 μg/ml), in 20 mM Tris pH 8.0, containing 150 mM NaCl, 1 mM CaCl2, 2 mM MgCl2, and 1% BSA (TSA buffer) or, when EGTA was used, in PBS. Subsequently, FITC-labeled C. albicans were resuspended to the appropriate concentrations either in TSA or PBS and added in various cell:Candida ratios. After incubation, cell-Candida conjugates were analyzed by flow cytometry, and the relative difference in mean fluorescence intensity of the double-labeled events in comparison with that of control cells was calculated. Cells were labeled with anti-CD45–allophycocyanin to discriminate cells binding FITC-labeled yeast particles from yeast aggregates.
3.6 Fluorescent-bead adhesion assay
The fluorescent-bead adhesion assay was performed as described earlier 12, 21. Briefly, carboxylate-modified TransFluorSpheres (488/645 nm, 1.0 μm; Molecular Probes, Eugene, OR) were coated with ICAM-3–Fc or ALCAM–Fc, and adhesion was determined by measuring the percentage of cells that had bound fluorescent beads, by flow cytometry. In inhibition studies, the bead-adhesion assay was performed in presence of 0.3 mg/ml mannan, 5 mM EGTA, or 20 μg/ml antibodies against DC-SIGN or ALCAM.
Immature DC (5×105) were incubated with unopsonized heat-inactivated FITC-labeled C. albicans (2.5×106) in a total volume of 500 μl at 37°C in a water bath with orbital shaking at 150 rpm for 60 min. At the end of the incubation period, the DC binding C. albicans were separated from unbound Candida by a Ficoll gradient.
The samples were then mounted on poly-L-lysine-coated glass coverslips by centrifugation at 250 rpm for 3 min. Subsequently, the samples were fixed in 1% PFA in PBS for 15 min at roomtemperature, and permeabilized in cold methanol for 5 min on ice. After a blocking step in PBS/3% BSA for 60 min at room temperature, cells were labeled with monoclonal antibody (10 μg/ml in PBS/3% BSA) for 60 min at room temperature and subsequently incubated with Cy5-conjugated Goat-anti-mouse (Fab′)2 fragments for 30 min at room temperature. Finally, samples were sealed in Mowiol and analyzed using a MRC1024 confocal microscope (Bio-Rad).
This work was supported by grant SLW 33.302P from the Netherlands Organization of Scientific Research, Earth and Life Sciences (to A. C.), by grant 497350, Alexion Pharmaceuticals, Cheshire, USA (to K. G.), by grant AZN/KUN 99/1950 from the Dutch Cancer Society (to J. d. V.), and by grant NWO 901-10-092 (to C. G. F.)