Langerhans Cell Migration in Murine Cutaneous Leishmaniasis: Regulation by Tumor Necrosis Factor α Interleukin-1β, and Macrophage Inflammatory Protein-1α

After intradermal infection of mice with the obligatory intracellular parasite Leishmania major, Langerhans cells (LC) are intimately involved in the induction of the primary T-cell immune response. LC can phagocytose Leishmania and transport ingested parasites from the infected skin to the regional lymph nodes. Since TNFα and IL-1β have been shown to induce LC migration after epicutaneous exposure to skin-sensitizing chemicals, we investigated the involvement of both cytokines in the migration of Leishmania-infected LC. In addition, the relevance of two chemokines of the β subfamily, macrophage inflammatory protein 1α(MIP-1α) and macrophage chemoattractant protein 1 (MCP-1), was analyzed.In vivo depletion of TNFα significantly reduced the amount of infected LC and the parasite load in the draining lymph nodes. Administration of recombinant TNFα caused the reverse effect. In contrast, the depletion of IL1β enhanced the parasite-induced LC migration, whereas treatment with recombinant IL-1β, as well as recombinant MIP- c, reduced the rate of infected LC in the lymph nodes. MCP- did not influence LC migration. Our data demonstrate that TNFα and IL-1β are regulating the LCmediated transport of Leishmania and also provide evidence for the involvement of macrophage attractant chemokines in this process.


er et   al.,
992; Schreiber et al., 1992).Furthermore, keratinocytes in the skin epidermis release tumor necrosis factor c (TNFc0 and granulocyte/macro- phage-colony stimulating factor (GM-CSF) that are critical for the viability and differentiation of LC (Koch et al., 1990).However, the factors regulating LC migration are not completely understood.In this study, we analyzed the role of TNFa and IL-lfl in Leishmania-induced LC migration.Furthermore, the influence of the chemokines MIP-1 c and macrophage chemoattractant protein 1 (MCP-1) was investigated.

Both chemokines are involved in skin-associated inflammatory processes and were recently shown to be important markers for the classification of different clinical forms of human cutaneous leishmaniasis (Ritter et al., 1996).Recombinant cytokines or neutralizing antibodies were administered in vivo, prior to or simultaneously with infection with the parasite.The amount of infected LC migrating to the lymph no es was estimated by double-staining tech- niques.In addition, the frequency of parasite-infected cells in the regional lymph nodes was determined in a limiting dilution assay.To distinguish the effects of locally produced mediators from those of exogenous sources on LC migration, we additionally used a skin explant culture system.


RESULTS


Detection of L. major-Containing Dendritic Cells

In the first experiments, we defined the time point that is suitable for detection of parasite-bearing LC arriving in lymph nodes draining the site of infection.As early as 2 days after intradermal infection of mice with L. major promastigotes, significant numbers of parasite-containing cells could be identified in lymph node sections using a double-staining technique (Figure 1).This time point was used in all experi- ments to evaluate the effects of TNFc, IL-lfl, MIP-1 a, and MCP-1 on LC migration.It is of particular importance that at this early time point, Leishmania parasites only can be found in lymph node dendritic cells and are not associated ith macrophages (Moll  et al., 1993).

TNFc Enhances the Leishmania-Induced LC Migration To determine whether TNFce has an effect on the migration rate of infected LC, mice were treated with recombinant TNFce and, at the same time, were infected intradermally with L. major parasites.Two days after treatment, the lymph nodes draining the site of infection were removed and immunohistochemical double staining was performed to enumerate infected dendritic cells.Figure 2 illustrates that the administra- tion of recombinant TNFce enhanced the number of parasite-bearing dendritic cells in the lymph nodes nearly twofold as compared to control mice that received normal serum.In order to confirm the involvement of TNFce in the induction of LC migra

ere treated with neutralizing anti- TNFce antibod
es, 2 hr before infection with the parasite.The depleting antibodies were applied intra- peritoneally and not locally to avoid antibody-induced changes in the content of LC at the site of exposure to the parasite (unpublished observation).As shown in Figure 2, the in vivo depletion of TNFce resulted in a pronounced reduction of infected LC in the lymph nodes.It should be noted that LC migration could not be totally inhibited by in vivo depletion of TNFce.

significantly change the frequency of infected lymph node cells.Thus, the cytokine-induced changes in LC migration correlate with those in the parasite load of the draining lymph nodes.

Skin Explant Cultures Show that

e Skin Nex
, we wanted to investigate whether the influence of TNFc and IL-1 fi on LC migration is dependent on the infection with Leishmania parasites.Furthermore, we wanted to analyze the effects of cytokine treat- ment on the local environment in the skin.To exclude the involvement of exogenous mediators, released by infiltrating cells, we harvested the skin areas immedi- ately after administration of recombinant TNFa or ILl f alone or in combination with L. major promasti- gotes.The skin explants were placed onto medium in 3-ml wells and, after 4 days of culture, the emigrated cells were immunostained for dendritic cell markers.

As demonstrated in Table 2, the administration of IL- l f i TNFa, parasites or phoshate-buffered saline (PBS) alone induce an enhanced LC emigration, as compared to untreated controls.Interestingly, the amount of emigrated dendritic cells was significantly lower than the PBS-treated group after combined treatment with L. major parasites and recombinant ILlfl On the other hand, only the combined administra- tion of TNFc with Leishmania parasites significantly enhanced the rate of emigrated LC.Thus, these data confirm the results obtained from the analysis of lymph nodes and, in addition, show that TNFo acts locally to enhance the LC-mediated export of L. The migration of L. major-infected LC is modulated by cytokines and chemokin s.The microscopical quantification of infected LC in lymph node sections after administration of control serum, neutralizing antibodies, or recombinant cytokines/chemokines and intradermal infection with 10 L. major promastigotes.Forty-eight hours after infection, serial sections were subjected to double-staining immunohistology for detection of NLDC-145 cells expressing L. major antigen.The data represent means SD of 10 random fields at 400 magnification.Two additional experiments gave similar results.

12 Control TNFc IL-1I MIP-lc MCP-1  5.3-9.9Note: Limiting dilution analysis of total lymph node cells after administration of neutralizing antibodies, recombinant cytokines/chemokines, or control serum and intradermal infection with 10 L. major promastigotes.Forty-eight hours after infection, single-cell suspensions were prepared and serial dilutions were incubated in blood agar cultures.Seven days later, the proportion of Leishmania-negative cultures was determined.The experiments were repeated three times with similar results.Note: Microscopical evaluation of epidermal dendritic cells that have emigrated from skin explants.Prior to culture in Click's RPMI medium, the skin was intradermally infected with 10 L. major parasites.Simultaneously, recombinant cytokines were administered.Four days after skin explantation, emigrated dendritic cells were harvested from the bottom of the wells and immunocytochemically detected by the APAAP technique.The data represent means SD of 10 random fields at 400 magnification and two sets of experiments.


DISCUSSION

The rapid emigration of LC from the epidermis and their consecutive accumulation in the paracortical areas of the draining lymph nodes after sensitization with alloantigens or contact allergens has been demonstrated previously in different systems (Macatonia et al., 1987; van Wilsem et al.; 1994, Kimber   and Cumberbatch, 1995).Recent studies have extended these observations by showing that not only allergens, but also the skin-associated protozoan parasite L. major can induce this migration (Moll et  al., 1993).Cytokines, in particular TNFc, have been suggested to be involved in this phenomenon.Follow- ing topical exposure to skin-sensitizing chemicals, the administration of TNFce, a potent immunomodulatory molecule, causes a reduction in the density of LC in the skin and an increase

n- bearing
C in the draining lymph nodes in a concentration-and time-dependent manner (Cumberbatch and Kimber, 1992).The administration of neutralizing anti-TNFce antibodies, on the other hand, inhibits dendritic cell accumulation in lymph nodes (Cumberbatch and Kimber, 1995).Furthermore, TNFa was shown to directly influence LC function by maintaining their viability in culture (Koch et al.,  1990).Our results presented here, using the skin- associated parasite L. major, extend these data to an infectious disease model and provide strong evidence that TNFce is an important signal inducing the migration of antigen-bearing LC in cutaneous leish- maniasis.The in vivo depletion of TNFce caused a pronounced reduction in the number of infected dendritic cells and the parasite load in the draining lymph nodes.As demonstrated in vivo and in skin explant cultures, the administration of recombinant TNFa caused the reverse effect.These data show that TNFce is a locally induced mediator of LC migration not only after contact sensitization, but also after intradermal infection with parasites.A recent model to explain LC migration after antigenic stimulation is based on the interaction of LC with keratinocytes, which are known to be the major source of TNFa in the skin (Schreiber et al., 1992).Since LC have been suggested to express the TNF receptor type II (p75), they may respond to TNF released from keratinocytes (Wang et al., 1996).This may result in the activation of protein kinase C, which transduces the signal for LC migration from the epidermis (Halliday and  Lucas, 1993).

Epidermal LC have been identified as the major source of IL-lfi production in the epidermis (Heufler   et al., 1992; Schreiber et al., 1992).Following epicutaneous skin sensitization, IL-lfl secretion can be induced within minutes.Paracrine and autocrine mechanisms cause the upregulation of MHC class II expression on LC and enhancement of their immunos- timulatory potential for resting T cells (Heufler et al.,  1988, Nylander-Lundqvist and Bick, 1990).This is consistent with the finding that the in vivo depletion of IL-lfi by monoclonal antibodies completely prevents sensitization to allergens (Enk et al., 1993).Further- more, there is evidence that IL-lfl can induce LC migration (Kimpgen et al., 1995).These findings were further substantiated by th use of IL-lfldeficient mice, which showed defective contact hypersensitivity responses to topically applied skin sensitizers (Shornick et al., 1996).In the present study, we analyzed the role of IL-lfl in Leishmania- induced LC migration by the use of recombinant ILlf or neutralizing anti-IL-1 fi antibodies.Administra- tion of the recombinant cytokine was expected to enhance the parasite-induced LC migration.Interestingly, however, the injection of IL-lfl reduced the migration rate of L. major-infected LC in vivo.The depletion of the cytokine caused the reverse effect.As demonstrated in skin explant cultures, this phenomenon was strongly dependent on the presence of parasites, because the application of IL-lfl alone resulted in an enhancement of LC emigration and, thus, had an effect that was similar to the nonspecific response after PBS injection.The observed reduction in the number of parasite-containing LC leaving the skin and reaching the lymph node after treatment with IL-lfl can most likely be explained by a decrease in the phagocytic activity of LC.The presence of recombinant IL-lfl significantly reduced the propor- tion of L. major-infected LC in vitro in a dose- dependent manner (data not shown).Blocking this inhibitory effect of IL-1 fl in vivo, by administration of anti-IL-lfl antibodies, may result in an enhanced capacity to phagocytose parasites, and in consequence would increase the percentage of infected LC migrat- ing to the lymph nodes.Therefore, our findings suggest that the detrimental role of IL-1 fi in L. major infection is not only caused by the selective activation of T helper 2 over T helper 1 cells (Chakkalath and  Titus, 1994), but is also due to a decreased rate of parasite phagocytosis and transport by LC in the very early phase of infection, which is critical for the primary stimulation of specific T cells.The chemokine MIP-lce, primarily secreted by stimulated macrophages, additionally is expressed by epidermal LC (Matsue et al., 1992).MIP-1 ce causes local inflammatory reactions and can regulate epi- dermal homeostasis by inhibiting keratinocyte colony formation.Thus, LC, as the major source of MIP-1 ce in the epidermis, can downregulate keratinocyte proliferation (Heufler et al., 1992).Keratinocytes, the overwhelming majority of epidermal cells, have been shown to be important inducers of LC maturation and differentiation (reviewed in Kimpgen et al., 1995).

This implies that inhibition of keratinocyte function results in a reduced ability to stimulate LC and, in consequence, may diminish the antigen-induced differentiation and migration of LC.
his hypothesis is in line with our data demonstrating that intradermal administration of MIP-lce, in combination with L. major parasites, decreased the number of antigenbearing dendritic cells and the parasite load in the draining lymph nodes.In conclusion, our observations extend the impor- tant roles of TNFce, IL-lfl, and MIP-1 ce in the regulation of LC migration to an infectious disease model.They also show that IL-lfl has a differential role in this process, because it supports LC migration in contact sensitization with haptens, whereas it inhibits the tr nsport of the intracellular parasite L. major by reducing the phagocytic activity of LC.


MATERIAL AND METHODS

phatase (APAAP) complexes were purchased from Dako (Hamburg, Germany).An immunogold-conju- gated goat anti-rabbit immunglobulin antibody (Amersham, Braunschweig, Germany) served as sec- ond-stage reagent for silver-enhanced staining.


Mice

Female mice of the inbred strain BALB/c were 8 to 10 weeks of age at the onset o experiments.All mice were purchased from Charles River Breeding Labora- tories (Sulzfeld, Germany) and, during experimenta- tion, were maintained under specific pathogen-free (SPF) conditions in an isolation facility.


Parasites and Infection of Mice

The origin, culture, and propagation of the L. major isolate (MHOM/IL/81/FE/BNI) have been described elsewhere (Solbach et al., 1986).Promastigotes were grown

n vitro in blood agar
cultures.Stationary- phase promastigotes were washed in PBS and, for intradermal infection of mice, 1 10 6 organisms were injected in a volume of 10 /zl on the dorsum, using a 100-/xl Hamilton syringe mounted with a 30-gauge Yale n

monoclonal antibodies (mAb) agai
st nonlymphoid dendritic cells, from hybridoma NLDC-145, as well as the isotype control antibodies, were from Dianova (Hamburg, Germany).Polyclonal antibodies to L. major were raised in rabbits by subcutaneous and intramuscular injections of promastigotes in Complete Freund's adjuvans (CFA) followed by several boosters of promastigotes in Incomplete Freund's adjuvans (IFA) and collection of the serum.Goat anti- rat i

unoglobulin antiserum was
obtained from Dianova, and alkaline phoshatase anti-alkaline phos-


Animal Treatment

All animal manipulations, other than antibody treat- ment, were performed under metofane anesthesia (Janssen, Neuss, Germany).Antibodies against TNFa and IL-1/ were injected intraperitoneally (1 mg/ mouse) 2 hr before infection with L. major parasites.

The recombinant cytokines and chemokines were injected intradermally (50 ng/mouse) simultaneously with the parasites.Control mice received normal serum at equivalent concentrations.All preparations were diluted with sterile PBS and administered in a total volume of 10/1.


Skin Explant Culture

Immediately after intradermal injections of L. major promastigotes and/or recombinant cytokines onto the shaved dorsum of BALB/c mice, skin explants of 10 mm 2 were transferred to 12-

ll plates and cul
ured in Clicks RPMI containing 10% FCS.After 4 days, the emigrated cells were harvested from the bottom of the wells and cytospins were prepared for subsequent immunostaining.

lmmunostaining For immunohistological identification of L. major- containing cells, lymph nodes were snap-frozen in OCT compound (Miles, Naperville, IL).Cryostat sections (4-6 #m) as well as cytospins were fixed in 4% paraformaldehyde for 20 min, followed by extensive washings in PBS.Nonspecific binding sites were blocked by incubation in Blotto (

S containing 10% skim
milk powder and 0.1% bovine serum albumin [BSA], fraction V) supplemented with 10% FCS for 30 min at room temperature.Mixed labeling of tissue sections was performed by incubation with the primary antibodies diluted in Blotto for 1 hr followed by treatment with goat anti-rat immuno- globulin antiserum diluted in Blotto (30 min) and the APAAP complex diluted in PBS (30 min).Afte incubation with immunogold-labeled immuno- globulin antiserum diluted in Blotto for 30 min, the APAAP staining was developed with New Fuchsin (Cordell et al., 1984).After postfixation in 2% glutaraldehyde and extensive washings in aqua bidest, immunogold signals were detected by silver enhancement according to the manufacturer's instructions (Amersham).Finally, slides were coun- terstained with hematoxylin and mounted in Kaisers glycerin gelatine (Merck, Darmstadt, Germany).All immunostainings were controlled by (1) the use of secondary reagents alone, to confirm their species specificity; (2) the development of alkaline phospha- tase alone, to exclude staining due to endogeneous enzyme activity; and (3) the use of rat and rabbit isotype control antibodies.All control stainings yielded negative results.


Limiting Dilution Assay

Groups of three mice were used to determine the parasite load in the draining lymph nodes, 2 days after intradermal infection with L. major.Single-cell suspensions were prepared and lymph node cells were resuspended in Clicks RPMI containing 10% FCS.Serial dilutions ranging from 1 108 to 1 103 cells/ well (20 replicate cultures per dilution) were incu- bated in blood agar microcultures to support the growth of parasites.After 7 days of incubation at 28C, the cultures were scored for the presence of parasites using an inverted microscope.Minimal estimates of the frequencies of L. major-infected cells and the 95% confidence intervals were calculated according to the minimum X z method from the Poisson distribution rel

ionship between the numb
r of lymph node cells and the logarithm of the fraction of Leishmania-negafive cultures by using a computer program (Taswell, 1987).



FIGURE 2The migration of L. major-infected LC is modulated by cytokines and chemokines.The microscopical quantification of infected LC in lymph node sections after administration of control serum, neutralizing antibodies, or recombinant cytokines/chemokines and


TABLE Frequency of
Frequency
Lymph Node Cells Infected with Leishmania Parasites


TABLE II
IIEmigration of LC from Skin Explant CulturesIn vivo treatmentNumber of NLDC-145 cells/100 cellsNone0.5 0.08PBS1.6 0.5rlL-fi1.8 0.3L. major + rlL-lfl0.8 0.1rTNFce2.2 0.4L. major + rTNFce2.6 0.4
AcknowledgmentsThe authors are grateful to R&D systems, Germany, for providing anti-IL-lfl antibodies and to Dr. Jan Kimber, ZENECA Central Toxicology Laboratory, Cheshire, UK, and thank Sandra K6ster