The Langerhans cell controversy: are they immunostimulatory or immunoregulatory cells of the skin immune system?

Langerhans cells (LCs) were discovered by Paul Langerhans in 1868 and were mistaken as nerve cells of the epidermis.¹ These ominous nerve cells were named after him and were almost forgotten for over 100 years until LCs finally became members of the dendritic cell (DC) family.² This finding opened a new research field to unravel the specific function of LCs in the skin immune system. In the last few years the LC community has grown immensely, still we are far away from understanding the role of LCs in skin immunity. Studies demonstrating that LCs are dispensable for immune responses against viral and parasitic infection of the skin^3,4 shook the foundation of the community and started the discussion whether LCs are involved at all in immune responses against pathogens invading the skin. The definition of Langerin as a specific marker for LCs⁵ accelerated the speed of discoveries dramatically, as it was possible for the first time to identify LCs beyond the skin, for instance in lymph nodes. Moreover, three different groups initiated the generation of mouse models in which LCs can be depleted.^6-8 These mouse models were first presented to the LC community in 2005 at the ‘9th International Workshop on Langerhans cells’, held for the first time in Madeira, Portugal, where Paul Langerhans lived and died in 1888 after a very productive scientific life. The early hopes that these models will instantly clarify the important role of LCs in skin immunity were soon shattered by initial findings that in the three mouse models contact hypersensitivity responses were different. The discovery of a novel subset of Langerin-positive DCs in the dermis^9-11 further complicated matters. Now, 4 years later, we have resolved some of the discrepancies between the mouse models, yet there is still a hot debate about the specific role of LCs in the skin immune system. In this special feature on LCs, we highlight the recent advances in LC research and how they help in understanding the developmental pathway and functional properties of LCs in the skin.

The research on the ontogeny of LCs has made major progress in recent years and we have begun to understand how precursors of LCs seed the epidermis during embryogenesis, and that they form the typical network in the epidermis within the first few days or weeks after birth by vigorous proliferation. Elbe-Bürger¹² and Ginhoux¹³ summarized for this special feature the current knowledge on LC development in mouse and humans. The situation in humans is in some aspects quite similar to that in the mouse, yet differs in some aspects, such as acquisition of the LC phenotype, which occurs after birth in mice but already in utero in humans.

The specific role of the diverse skin DC populations in the induction of immunity and tolerance is currently under discussion, and there is some indication that dermal DCs and LCs might have different roles in the induction of immune responses in skin-draining lymph nodes. LCs appear to be important for the induction of T-cell responses and dermal DCs for humoral responses. This division of tasks by the various cutaneous DCs is discussed in more detail in the review by Ueno.¹⁴ The number of dermal DC subsets is still growing and Malissen and coworkers¹⁵ successfully disentangled the complex network of dermal DCs. Currently, we can distinguish five subsets of DCs in the dermis; however, whether they all exert different functions remains to be determined.¹⁵ What becomes clear is that the recently discovered Langerin-positive dermal DC subsets gains importance and seems to conduct many functions that were so far attributed to LCs.

Further progress will be made with the LC-depletion mouse models developed a few years ago, which allowed for the first time to investigate the function of LCs in vivo.^6-8 In the inducible Langerin-DTR mouse models, the function of LCs and Langerin-positive dermal DCs can be dissected by the timing of diphtheria toxin injection. The Langerin-positive dermal DC subset recovers very quickly within the first week, whereas LCs take several weeks to repopulate the epidermis in the steady state. Recent work performed in Clausen’s and our lab¹⁶ demonstrates clearly that both Langerin-DTR mouse models show diminished immune responses to low concentrations of the contact sensitizer. In contrast, higher hapten concentrations diffuse into the dermis and immune reactions can also be mediated by dermal DCs besides LCs. The importance of LCs in mediating contact hypersensitivity and ultraviolet-induced immunosuppression is discussed in detail in the review by Clausen in this special feature on LCs.¹⁷ Things look different in Kaplan’s Langerin-DTA mouse model, in which the diphtheria toxin itself is expressed under the human Langerin promoter, causing constitutive depletion of LCs throughout life, whereas Langerin-positive dermal DCs are not affected.¹⁸ Contact hypersensitivity reactions are increased in these mice, suggesting an immunoregulatory role for LCs. This may be mediated by interaction of LCs directly with CD4-positive T cells in the draining lymph nodes, with LCs secreting IL-10 and thereby curtailing initiation of the T-cell response.¹⁹ The fact that LCs might have a regulatory function is discussed for several tolerogenic situations by Lutz.²⁰ Surprisingly, LCs are not mandatory for induction of peripheral tolerance against a self-antigen expressed in the keratinocytes of the epidermis. The Langerin-positive dermal DCs are the major antigen-presenting DC population in the lymph node.^15,21 Thus, LCs can exert a dual function, that is, induce either tolerance or immunity, depending on the specific situation in the skin. Besides LCs, a subset of Langerin-positive dermal DCs can take over the role of inducing peripheral tolerance and immunity when they gain access to the antigen.

The role of LCs in immune responses against pathogens invading the skin has been questioned some years ago by studies demonstrating that LCs are not required for T-cell responses against the herpes simplex virus that infects cells in the epidermis.³ This was not the whole story, as was shown by the same group identifying lymph node-resident CD8⁺ DCs and the Langerin-positive dermal DCs as major antigen-presenting cells.²¹ The role of LCs in viral infection is discussed in more detail by Cunningham in this special feature.²² The induction of apoptosis in LCs and DCs by viral infection explains the inability of LCs to actively participate in antigen presentation, but renders them as antigen reservoirs for cross-presentation. Another interesting finding is that Langerin can act as a neutralizing receptor for human immunodeficiency virus, which is reviewed by Geijtenbeek.²³ Incorporation of human immunodeficiency virus into Birbeck granules by Langerin surface molecules results in degradation of the virus and protection from infection. In another disease model, atopic dermatitis, it was demonstrated recently that LCs are important for disease induction.²⁴ The current knowledge on LCs in atopic dermatitis is that LCs are involved in the initial phase of pathogenesis by inducing pro-allergic T cells in humans and mice as reviewed in this special feature by Ebner and Dubrac.²⁵

The finding that LCs are involved in immune responses against various diseases makes them potential targets for immunotherapy. Early work in Ralph Steinman’s lab²⁶ and more recent studies in Nikolaus Romani’s lab²⁷ proved that LCs are very potent stimulators of CD4⁺ and CD8⁺ T-cell responses. As a consequence, we have started testing the possibility of harnessing LCs for immunotherapy. The development of novel concepts for topical immunization through the skin aims at loading LCs directly into the skin with antigens. Epicutaneous immunization with antigens through barrier-disrupted skin led to potent cytotoxic CD8⁺ T-cell responses that can inhibit the growth of melanoma.^28,29 The anti-tumor effect is mediated by LCs and dermal DCs as demonstrated in Langerin-DTR mice, suggesting that LCs can present topically applied antigen to T cells in skin-draining lymph nodes.³⁰ This novel immunotherapeutic approach and its potential as a treatment for skin cancer and infection is discussed in a special feature review by Stoitzner.³¹ Targeting the antigen to skin DCs with the help of DC-specific antibodies dramatically enhances T-cell responses.³² This strategy is currently being developed with special regard to targeting skin DCs, in particular LCs. C-type lectin receptors such as DEC-205 and Langerin are promising tools to load LCs with antigen in situ, as was shown recently.³³ Romani³⁴ summarizes the potential of targeting approaches for immunotherapy in this special feature on LCs.

This series of reviews highlights the challenges and controversies that keep the LC community busy. We have moved a long distance since the discovery of LCs in 1868. Still the complex network of cutaneous DCs clearly impairs our understanding of the functional properties of LCs in vivo. Although the definite function of LCs is still elusive, it is becoming clear that similar to other DC subsets, LCs can exert manifold functions depending on the situation in the skin.