Human primary dendritic cell subsets differ in their IL-12 release in response to Leishmania major infection


Esther von Stebut, Department of Dermatology, University Medicine, Johannes-Gutenberg University, Langenbeckstrasse 1, 55131 Mainz, Germany, Tel.: +49-6131-175731, Fax: +49-6131-173470, e-mail:


Abstract:  Immunity against leishmaniasis has primarily been studied in experimental infections of mice. It was shown that infected skin dendritic cells (DC) are critical for the induction of protection against this pathogen, and targeting skin DC in vaccination approaches in mice has proven to be successful. However, little is known about the contribution of human DC subsets from the skin to primary immunity against this pathogen. In this study, we have analysed the interaction between different human DC subsets and Leishmania major. Primary human myeloid and monocyte-derived DC ingested the parasite comparable to that of murine skin DC, and this resulted in DC activation and IL-12 release, a cytokine essential for the induction of Th1/Tc1-dependent protection. Interestingly, both Langerhans cells and plasmacytoid DC did not appear to contribute to protection in humans. Thus, in leishmaniasis, both murine and human data suggest that dermal inflammatory DC appear to be superior in promoting protection.




immature monocyte-derived DC


mature monocyte-derived DC


plasmacytoid DC


myeloid DC


Langerhans cell


In murine experimental leishmaniasis, infected skin dendritic cells (DC) prime naïve T cells and promote protective Th1/Tc1-dependent immunity via IL-12 (1,2). Thus, antigen-loaded DC have been utilized as vaccine against progressive disease (3). In contrast, the role that infected DC play for protection against leishmaniasis in humans remains controversial (4–6).

Question addressed

Because leishmaniasis is still among the top ten of infectious diseases worldwide and no vaccine exists, a translation of our current knowledge about the contribution of murine DC (subsets) to anti-Leishmania immunity to the human system will aid vaccine development. In this study, we investigated the interaction of L. major with human DC generated in vitro and, more importantly, with ex vivo isolated primary DC.

Experimental design

Monocyte-derived DC were generated from buffy coats and harvested either as immature (imDC) or as mature DC (mDC) after incubation with proinflammatory cytokines (7). For comparison, macrophages (MΦ) were generated from monocytes. In addition, primary myeloid DC (myDC) and plasmacytoid DC (pDC) were enriched from buffy coats using anti-CD1c and anti-CD304-coated microbeads, respectively. Finally, Langerhans cells (LCs) were isolated from excess skin from plastic surgery after written patient consent was obtained using CD1a-coated microbeads according to the manufacturers’ protocol (8,9). The study was approved by the local ethics committee. The surface phenotype of all DC subsets was characterized by flow cytometry (compare supplementary Fig. S1). Cell viability was assessed using trypan blue and propidium iodide staining.


First, DC subsets were co-incubated with L. major promastigotes (infectious stage parasites transmitted during sand fly bites) or amastigotes (obligate intracellular form found in vivo in infected hosts) at physiologically relevant parasite/cell ratios of 3:1 and 10:1. The majority of DC ingested both parasite life forms in a time- and dose-dependent fashion (Fig. 1a,b). Similar to murine and human MΦ, imDC phagocytosis of L. major was fast and resulted in high infection rates. In contrast, terminal differentiated mDC ingested L. major parasites much slower with preferential uptake of intracellular amastigotes. Similar to murine pDC, infection was not detected in primary human pDC (10). In addition, LC did not ingest significant numbers of parasites, which is in contrast to some prior murine studies (11–13), but not others (14). Interestingly, primary myDC behaved very similar to murine skin DC in that they preferentially took up the amastigote life form in a comparably slow fashion (Fig. 1a,b). These data suggest that L. major amastigote uptake by human myDC (and most likely mDC) is dependent on IgG/Fcγ receptors (15).

Figure 1.

 Differences in L. major phagocytosis by various human DC subsets generated in vitro or isolated ex vivo. Macrophages (MΦ) were generated from monocytes (Mo) by the addition of M-CSF (50 ng/ml). DC were cultured from Mo in the presence of GM-CSF (400 U/ml) and IL-4 (150 U/ml) and harvested as immature DC (imDC) on day 6. Subsequent imDC subculture with PGE2 (30 μg/ml), IL-1β (10 ng/ml) and TNF-α (10 ng/ml) for 2 days resulted in mature DC (mDC). Primary human cells were isolated ex vivo from peripheral blood (myeloid DC/myDC, plasmacytoid DC/pDC) or excess skin obtained from plastic surgery (Langerhans cells, LC) using CD1c-, CD304- or CD1a-coated microbeads, respectively. All cells (2 × 105/ml) were co-cultured with L. major amastigotes isolated from infected BALB/c footpads or cultured promastigotes at parasite/cell ratios of 3:1 (a,b) or 10:1 (b). Infection rates were determined over time (a) or after 18 h (b) by light microscopy of DiffQuick-stained cytospins. All data are shown as mean ± SEM (n ≥ 3 experiments with independent blood or tissue donors). Statistical differences between the indicated groups are designated as *P ≤ 0.05, **P ≤ 0.005 and ***P ≤ 0.002. n.d. = not determined.

In parallel, we determined whether parasite phagocytosis was associated with cell activation as observed in murine DC (13). Upregulation of CD83, CD80 and CD40 after 18 h in response to L. major infection was strongest by mDC, followed by imDC and myDC (Fig. 2a). However, these effects did not reach statistical significance. Prior studies using imDC either demonstrated that L. major infection led to upregulation of costimulatory and MHC class II molecules (4), or had no effect on DC maturation (6). These differences may be explained by variations in the Leishmania spp. and cells used.

Figure 2.

L. major infection of human DC subsets leads to cell activation and release of bioactive IL-12. Human DC subsets were generated as described in Fig. 1. Cells were plated at 2 × 105/ml and infected with L. major parasites (a: 3:1, b: 10:1 parasites/cell). (a) Surface marker expression was determined after 18 h using flow cytometry and antibodies against CD40, CD80 and CD83. The percentage change of baseline expression of untreated cells was calculated. (b) Infection rates and release of IL-12p40 and p70 into 18- h supernatants was determined by ELISA (R&D Systems, Wiesbaden, Germany). (a,b) Data are expressed as mean ± SEM (n ≥ 3). Statistical differences are designated as *P ≤ 0.05 and ***P ≤ 0.002. n.d. = not determined.

Finally, we detected comparably high IL-12p40/p70 release by L. major-infected imDC, mDC and myDC (parasite/cell ratio 10:1), but not MΦ, LC (Fig. 2b) and pDC (data not shown). Production of IL-4, IL-6, IL-10 or TNF-α in response to L. major infection was not found. Our data are in line with previous observations, which indicated that imDC produced IL-12p70 when infected with L. major in conjunction with CD40L, but not when infected with viscerotropic strains as L. donovani or L. tropica (4,5). Recently, Jayakumar et al. demonstrated that L. major-infected DC results in the early activation of NF-kappaB transcription factors and the upregulation and nuclear translocation of interferon regulatory factor 1 (IRF-1) and IRF-8 ultimately leading to IL-12 transcription (16). In contrast, in our hands, IL-12 release was not dependent on CD40 stimulation in agreement with recent studies showing that resistance to murine infection is independent of CD40 (17). However, similar to CD40L, additional TLR ligation may synergize for IL-12 induction from DC. Infection of primary human myDC, pDC or LC has not been studied previously.


The question which DC subtype is most relevant for future human studies remains still open. Several features of human MΦ/DC subsets in their interaction with L. major were comparable to those of murine cells: MΦ phagocytosis of parasites was rapid with amastigote internalization being finalized within 6 h and promastigote incorporation being slower; after 18 h, comparable numbers of promastigotes and amastigotes were ingested. In addition, MΦ did not respond to infection with release of IL-12 or changes in their activation markers (data not shown). In parallel, imDC behaved similar except for IL-12 release and surface marker upregulation in response to infection, which may be attributable to their close relationship with MΦ and their relatively instable phenotype as they revert to monocyte-like cells upon cytokine withdrawal. Whereas pDC and LC appear not to contribute to protective immunity in both mice and humans, the difference between imDC and mDC was primarily in the degree of parasite internalization. mDC and primary myDC behaved very similar to murine skin DC in that they preferentially ingested amastigotes, became activated and released IL-12p70. Thus, for future studies in humans, these two latter cell types are of particular interest, especially in the light of recent findings in experimental leishmaniasis indicating that dermal DC (phenotypically resembling mDC and myDC), rather than LC, which might act as negative immune regulators, are critical for the induction of protective immunity (18–20).

In summary, in contrast to prior studies indicating that human DC do not contribute to primary immunity against L. major (4–6), our results show that certain primary human DC subsets phagocytose the parasite and release IL-12. Thus, depending on the type of DC that interacts with L. major in vivo, human DC may contribute to the induction of Th1-dependent protective immunity against this important human pathogen. Targeting the ‘right’ DC subtype in vivo will be of critical importance for the development and effectiveness of future vaccines.


We thank Prof. W. Wiest (Department of Gynecology, St. Vincenz and Elisabeth Hospital, Katholisches Klinikum Mainz, Germany) and his team for providing excess skin. This work was supported by grants from the Deutsche Forschungsgemeinschaft (SFB 490, GK1043, and Ste 833/6-1 to EvS)