What's new in photoimmunology?

Epidermal RANKL controls regulatory T-cell numbers via activation of dendritic cells . Loser, K, Mehling, A, Loeser, S, Apelt, J, Grabbe, S, Schwarz, T, Penninger, JM, Beissert, S. Nature Medicine 2006 ; 12 : 1372–1379 . Tolerance induction by transcutaneous immunization through ultraviolet-irradiated skin is transferable through CD4+CD25+ T regulatory cells and is dependent on host-derived IL-10 . Ghoreishi, M, Dutz, JP. Journal of Immunology 2006 ; 176 : 2635–2644 . Experiments carried out in mice about 30 years ago indicated that the exposure to ultraviolet radiation (UVR) could down-regulate cell-mediated immunity. Since then, considerable progress has been made to elucidate the complex cascade that starts with the absorption of photons at or just below the body surface and ends with the generation of T cells that can suppress immune responses. More recently it has been recognized that these cells belong to a subset of T cells called T-regulatory cells, which are CD4+CD25+ and express the transcription factor Foxp3. T-regulatory cells are involved in the maintenance of peripheral tolerance and have been proposed to be particularly important in avoiding autoimmunity. In the first of the highlighted papers, the group of Stefan Beissert at the University of Muenster in Germany provides fascinating and novel insights into the control of T-regulatory cell numbers via the interaction between keratinocytes and Langerhans cells (LCs), the major antigen-presenting cells of the epidermis, and hence into the mechanisms of UV-induced immunosuppression. It was already known that dendritic cells (DCs) can affect the development and proliferation of CD4+CD25+ T cells and that members of the tumour necrosis factor (TNF) family can regulate DC function. TNF and TNF-receptor family proteins play a role in the development, survival, proliferation and function of immune cells. One member of the TNF family is RANK-L, also known as TRANCE, OPGL and CD254. RANK-L initially exists as a trimeric membrane-bound molecule that is subsequently released from the cell surface as a soluble homotrimeric molecule. Upon interaction of RANK-L with its receptor RANK, the NF-κB, akt/PKB and ERK pathways are activated, thus affecting both survival by anti-apoptotic mechanisms and immune function. The original information regarding RANKL was obtained from studies of bone remodelling where RANKL, produced by activated T cells, induced the differentiation of macrophages into osteoclasts. Here a very different set of cells is involved, based on the skin immune system. The initial experiments revealed that while RANKL was expressed at a very low level in normal keratinocytes, it could be upregulated following an inflammatory stimulus, such UVR exposure. This was shown by irradiation of mice on their shaved backs with 800 mJ/cm2, a rather high dose likely to cause sunburn, with subsequent detection of RANKL mRNA by RT-PCR and RANKL protein by immunofluorescence. The next step was to create transgenic mice that overexpressed RANKL under the transcriptional control of the keratin-14 promoter so that this protein is only present at high concentration in the epidermis. These mice were tested subsequently for their T-cell reactivity by measuring contact hypersensitivity (CHS) to DNFB. The CHS response was suppressed by at least 50% in the transgenics compared with the wild type mice. Furthermore, if the transgenic mice were injected intravenously with RANK-Fc before sensitization (to ‘neutralize’ the action of RANKL), the CHS was enhanced over that seen in the wild-type mice. It was concluded that, when the keratinocytes overexpress RANKL, as occurs following UVR, a reduction in CHS ensues. The second series of experiments showed that epidermal RANKL increases the number of CD4+CD25+ T-regulatory cells expressing Foxp3. This was demonstrated by counting the number of such cells in the spleen and lymph nodes of the transgenic and wild-type mice. The transgenics had a two to threefold higher number, which was reduced to the level found in the wild-type mice if the transgenics were treated intravenously with RANK-Fc. Evidence was obtained to demonstrate that first the RANK–RANKL interactions were involved in the maintenance or peripheral expansion of the T-regulatory cells but not in their thymic development, and secondly the CD4+CD25+ T cells generated in response to the expression of RANKL on keratinocytes had the phenotypic and functional characteristics of T-regulatory cells. DC subsets, in addition to macrophages and osteoclasts, are known to express the RANKL receptor, RANK, constitutively and Loser and colleagues next revealed, using immunofluorescence, that epidermal LCs also express RANK. They demonstrated by in vitro experiments that epidermal LCs and DCs from skin-draining lymph nodes stimulated the proliferation of allogeneic CD4+CD25+ T cells five to 12-fold higher when prepared from the RANKL transgenic mice compared with the wild-type mice. In vivo studies followed in which epidermal LCs were depleted by topical treatment of both wild-type and RANKL transgenic mice with mometason fuorate. This led to a reduction in the number of CD4+CD25+ T cells. Thus it seems that epidermal LCs can control the numbers of peripheral T-regulatory cells. Further evidence was provided to indicate that TNF-α, induced by the stimulation of DCs by RANKL, may be a crucial mediator in the proliferation of the T-regulatory cells. Most importantly from the viewpoint of photoimmunology, the authors demonstrated that the expression of RANKL by keratinocytes in the skin links UVR with immunosuppression. Thus, if RANK-Fc was injected into mice before irradiation and sensitization, the expected suppression in the CHS response was abrogated. In addition, in wild-type mice grafted with skin from transgenic mice that did not express RANKL and sensitized through the grafted skin, no suppression occurred in the CHS response following UVR. Furthermore, if mice were UV-irradiated with an extremely small dose of only 1 mJ/cm2 on four consecutive days, increased numbers of CD4+CD25+ T cells were found in lymph nodes draining the exposed skin compared with unirradiated mice or mice treated with RANK-Fc and UV irradiation. In summary, new evidence has been provided to suggest that UVR induces the production of RANKL by keratinocytes, and RANKL then attaches to its receptor, RANK, on LCs in the epidermis. Some of the LCs migrate subsequently to the draining lymph nodes and, as DCs, then stimulate the proliferation of CD4+CD25+ T-regulatory cells in the periphery, thus leading to immunosuppression. This scheme is shown in outline in Fig. 1. One of the interesting ‘unknowns’ is the mechanism by which UVR up-regulates the synthesis of RANKL by keratinocytes. The authors speculate that vitamin D3 might be involved as this molecule is synthesized in the skin following UVB irradiation, and is known both to induce RANKL expression efficiently and to suppress immune responses in the skin. Other questions remain such as determining the possible contributions of UV-induced DNA damage, cis-urocanic acid and lipid peroxidation in RANKL induction. In addition, the minimal dose of UVR required to up-regulate RANKL production in vivo and its action spectrum are needed, and an assessment of the length of time RANKL continues to be expressed following the UV exposure. CHS was used throughout to determine T-cell reactivity in vivo but it would be important to establish if RANKL has the same role in suppressing delayed-type hypersensitivity where complex antigens such as microorganisms require to be taken up and processed by the LCs before presentation in the context of MHC Class II. What happens in systemic UV-induced immunosuppression is also not clear: here the antigen is applied to a non-irradiated site on an irradiated animal so that there would be no up-regulation of RANKL on the keratinocytes where the antigen is administered. Finally, a role for UV-induced RANKL in altering already established immune responses merits investigation. Proposed mechanism of UV-induced immunosuppression via the production of RANKL by keratinocytes. The second of the highlighted papers also involves UVR and the induction of CD4+CD25+ T regulatory cells, but the novel aspect is that transcutaneous immunization (TCI) was studied. In TCI, the vaccine is delivered through the skin using patches or following tape stripping, together with an adjuvant in some cases. This route utilizes the skin immune system and the LCs in particular to generate the response and is particularly attractive because it avoids the need for needles, could be self-administered and allows the safe use of potent immunostimulants as these are sequestered solely in the skin. Ghoreishi and Dutz from the University of British Colombia in Canada investigated what happened when the TCI took place through UV-irradiated skin. Mice were exposed on their shaved backs to a suberythemal UVB dose on four consecutive days, tape stripped and either whole protein (ovalbumin, OVA) or the MHC-I-restricted immunodominant OVA peptide (SIINFEKL) was applied to the skin under tape occlusion, together with cholera toxin or CpG as adjuvants. The priming of antigen-specific CD8+ cytotoxic T cells (CTLs) was used as the read-out system. Instead of the expected proliferation of the CTLs in response to OVA, almost complete inhibition of their production occurred. This was in contrast to the MHC-I-restricted peptide, which did induce proliferation of the CTLs. The authors then analysed the reduced response to the whole OVA protein and found that the CTL precursor frequency in the peripheral lymph nodes was reduced in the irradiated animals and was similar to that in naive mice. A CHS test involving ear swelling was developed to determine the physiological significance of the change in the CTLs as this response is due predominantly to the CD8+ T cells, with the CD4+ T cells contributing only when the CD8+ T cell pool is depleted. In mice immunized with OVA following UVR, there was a significant suppression in the CHS response to OVA compared with the response in unirradiated, immunized animals. The authors then proceeded to demonstrate that immunization with OVA through UV-irradiated skin induced tolerance. This was shown by adoptive transfer experiments where various populations of cells from lymph nodes draining the skin of previously UV-irradiated and immunized mice were injected intravenously into recipient mice. These mice were then immunized transcutaneously with OVA or CpG as adjuvants. The generation of CTLs was again used as the read-out system. It was found that transfer of the whole lymph node cell population caused a reduced CTL response compared with control mice that had not received any donor cells. However, if the lymph node preparation was depleted of CD4+ T cells before transfer, the CTL response was enhanced compared with the control mice. These results were dependent on the donor mice being UV-irradiated before TCI immunization and were specific to the antigen (OVA) used as the immunogen. Thus it was concluded that antigen-specific CD4+ T cells are generated in the irradiated, immunized mice that inhibit the subsequent priming of the CD8+ T cells and are responsible for the tolerance. Subsequent experiments identified this subset of T cells as CD4+CD25+, expressing high levels of Foxp3, thus indicating their regulatory phenotype. Finally the role of IL-10, recognized to be a key immunosuppressive cytokine following UVR, was investigated. IL-10 knockout mice, which had been irradiated before TCI, showed no decrease in CTL priming specific for OVA, unlike the situation in IL-10-sufficient animals. Adoptive transfer of CD4+ T cells from UV-irradiated and immunized IL-10 knockout and sufficient mice were then carried out and, somewhat surprisingly, showed that donor cells from both types of animals could mediate suppression of the CTL priming. Thus IL-10 is not required for the generation of the UV-induced T-regulatory cells or for their tolerogenic activity. However, it was revealed, by adoptive transfer of CD4+ T cells from irradiated and immunized wild-type mice into IL-10 knockout and sufficient mice, that IL-10 from the host is necessary for maintaining the tolerogenic properties of the T-regulatory cells generated through TCI of irradiated skin. The source of the IL-10 could be from antigen-presenting cells or bystander cells activated by the interaction with the CTLs. Considerable advances in knowledge surrounding UVB-induced immune suppression can be derived from these two papers including a RANKL-dependent mechanism for the generation of T regulatory cells following UVB irradiation and the role of T-regulatory cells in inducing tolerance following TCI through a UV-exposed site. Apart from these new findings, the main points of general interest include the provision of a mechanism by which UVR may influence the incidence of autoimmune disease and the tantalizing prospect of being able to induce antigen-specific tolerance by UVR. The strikingly increased prevalence of the autoimmune disease, multiple sclerosis, as the latitude increases, with presumed lower exposure of the population to solar UVB, may be partly attributed to a failure of T-regulatory cell induction/survival in these conditions. Thus at higher latitudes, RANKL expression by keratinocytes could be reduced, hence leading to reduced T-regulatory cell generation and an increased risk of autoimmune disease. It might even be possible in the future to prevent autoimmune diseases by suppressing immune responses to self-antigens, utilizing the T-regulatory cells generated as a result of UV exposure with subsequent cutaneous application of the relevant antigen, which would require to be whole protein rather than MHC-I-restricted peptides. This would have the huge advantage of being antigen-specific, unlike the immunosuppressive agents in general use currently that affect all immune responses indiscriminately. At the very least, it is already clear that TCI should take place through UV-protected skin if a robust CTL response to a protein vaccine is desired.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.