The skin immune system and the challenge of tumour immunosurveillance

ARTICLE

Auteur(s) :, Gregory M Woods, Roslyn C Malley, H Konrad Muller^*

Discipline of Pathology, School of Medicine, University of Tasmania, Private Bag 29, Hobart Tasmania Australia 7001 Fax: (+61)3 6226 4833

accepté le 17 Novembre 2004

SIS and the cutaneous immune structure

In this context dermal vascular endothelium and its adhesion molecules have a key role in allowing the influx of dendritic cells, lymphocytes and other inflammatory cells into the skin. Mast cells, which are situated in close proximity to the dermal vasculature, release chemical mediators that maintain homeostasis [10]. For example, when TNFα is released from mast cells it controls leukocyte infiltration as well as upregulation of adhesion molecules [11]. The interaction of the neural network and mast cells has a regulatory role in inflammatory and immune responses through sensory neuropeptides, the tachykinins. A good example is substance P released during an axon reflex from sensory nerves triggering mast cell release of histamine and the flare vasodilatation reaction [10].

Langerhans cells and T cells

Langerhans cells can be identified in the epidermis by the expression of MHC-II, DEC-205 (CD205) and Langerin (CD207) [18]. Other distinguishing phenotypic markers expressed by LC include CD1(a,c), CD11(b,c), CD33, S100 and costimulatory molecules such as CD40, CD86 [19]. Of all the phenotypic markers used to identify LC, it is only Birbeck granules which are unique to this population of cells. These rod/tennis racket shaped organelles remain one of the most distinguishing features of LC. The function of these organelles has yet to be fully resolved, although it has been proposed that they are involved in antigen processing. This notion was derived from experiments utilising Langerin, a mannose specific C-type lectin receptor involved in antigen uptake. When the gene for Langerin was transfected into fibroblasts, Birbeck granules were induced [20]. As antibodies to Langerin also react with Birbeck granules it appears that Langerin and Birbeck granules are closely related, if not the same molecules.

The most important biological role of LC is to initiate an immune response, which will occur following antigen uptake and processing. The external skin environment is sampled by the LC via either pinocytosis of soluble antigen (“drinking”) or endocytosis/phagocytosis of particulate antigen (“eating”). Soluble antigen can be pinocytosed via two distinct mechanisms. For small molecules (~ 0.1 μm) internalisation occurs via clathrin-coated pits (micropinocytosis) whereas uptake of larger particles (0.5-3 μm) internalisation occurs via a cytoskeleton-dependent membrane ruffling (macropinocytosis) [21]. Although macropinocytosis can be performed by a range of cells types, dendritic cells are the only cell type capable of this process without the receipt of external stimuli [22]. Such a process enables cells, like LC, to engulf a large volume of fluid, from which soluble antigens are then concentrated for processing and presentation. Receptor-mediated endocytosis involves receptors such as the lectin-type receptors DEC-205 [23, 24], Fc receptors CD16 and CD32 [25, 26], and complement receptors CR3 and CR4 [27]. Microbial pathogens, are engulfed by phagocytosis [28] a process utilising pattern recognition molecules which are receptors for prokaryotic antigens including lipopolysaccharide and peptidoglycan [29, 30]. Langerhans cells have been shown to have some phagocytic potential against a range of targets such as yeast and bacteria but this is less effective than macrophages and is lost when the cells are cultured in vitro [27].

After antigen uptake the ingested material is then processed and peptide is presented to T cells, a process that is more efficiently performed by LC than by macrophages. Very few studies have specifically analysed freshly isolated LC, due to the difficulty in obtaining pure populations of LC, and most analyses of antigen processing have utilised monocyte-derived dendritic cells. Nonetheless it is clear that LC are potent antigen presenting cells and process antigen via the MHC-I and MHC-II pathways [31]. The class I pathway is designed to present endogenous antigens, including viral and tumour antigens to cytotoxic CD8+ T cells, whereas the class II pathway presents exogenous antigens to cytokine producing CD4+ T cells. Once LC mature they lose their capacity to process antigen via the MHC-II pathway [31]. When dendritic cells endocytose tumour antigens, the antigen is essentially exogenous and therefore CD4+ T cells rather than tumour specific cytotoxic CD8+ T cells would be activated. To overcome this, dendritic cells utilise cross presentation, which occurs when exogenous antigens, such as tumour cells, are phagocytosed and initially processed via the MHC-II pathway but then “cross over” to the MHC-I pathway, thereby presenting to cytotoxic CD8+ T cells [32].

Following uptake of antigen and initial processing, LC undergo a process of maturation which converts them from a cell that can internalise antigen to a cell that can process and present antigenic peptides. Included in this maturation is a down regulation of adhesion molecules such as E-cadherin which allows the LC to detach from neighbouring keratinocytes. These detached cells are then triggered to migrate out of the epidermis by the cytokines IL-18, TNFα and IL-1β produced by keratinocytes and LC [33]. During this process LC downregulate CCR6, the chemokine receptor that attracts these cells into the skin, and upregulate CCR7, which is the receptor for the chemokine CCL21 that directs them into the lymphatics [34]. The production of metalloproteinases is also required to allow LC to penetrate the epidermal basement membrane and access the afferent lymphatics [35]. Once LC reach the lymphatics they are directed via CCL19 into the paracortex [34] to present antigenic peptides to naïve T cells. Within the lymph node they undergo further maturation converting them from antigen processing cells to antigen presenting cells. This is accompanied by an upregulation of molecules associated with antigen presentation namely MHC-I and MHC-II as well as costimulatory molecules including CD40, CD80 and CD86 [36].

The initial antigen presentation is critical to the development of an effective immune response involving the activation of antigen specific naïve T cells. The activation of these T cells is directed by way of an “immunological synapse” which involves critical receptor interactions between the membranes of LC, the antigen presenting cells, and T cells. Inappropriate antigen presentation can lead to an impaired immune response which may result in diverse outcomes such autoimmunity or immunosuppression. Further complex interactions involving regulatory and helper T cells also determine the type of immune response that develops. After T cell activation, proliferation follows and the effector T cells migrate from the lymph nodes into the circulation where they home to the sites of antigen challenge in the skin. The T cells that migrate to the skin belong to a population of skin specific homing cells that can be identified by the expression of cutaneous lymphocyte-associated antigen (CLA) [37]. This homing is mediated by chemokines such as CCL27 (CTACK) [38] that attract the T cells to the appropriate skin site where they interact with the local antigen bearing LC. This interplay results in pathological reactions such as a contact sensitivity response.

LC and interacting T cells are likely to have a key role in a range of pathological reactions including viral infection, leishmaniasis, transplantation rejection, autoimmunity, dermatitis reactions and diseases of altered epidermal growth such as lichen planus and psoriasis. In human contact dermatitis, chemicals such as nickel are presented by LC to the immune system resulting in Th1 cell activation involving cytokines such as IFNγ, IL-12 and TNFα. The migration of CD4 and CD8 cells back to the skin results in further interaction with LC and the classical lesions of contact dermatitis, including spongiosis [39]. In many of the above conditions antigens are cleared and cell death is evident and mediated by immune events. A major issue in dermatology is to understand the interplay of SIS and neoplastic cells. This needs to be understood from the outset of tumour genesis including initiation, promotion/ progression, tumour invasion and metastases. Central to this interaction are mechanisms which explain how cutaneous tumour cells bypass immune attack. An understanding of such issues is likely to lead to better and more targeted therapy in the future.

Immune lessons from experimental cutaneous carcinogenesis

To enhance the understanding of the interplay between LC and T cells, experimental models have been developed to follow these events. One good model uses the fluorescent contact sensitiser fluorescein isothiocyanate (FITC), which, when applied to the skin, is taken up by the resident LC and transported to the draining lymph nodes within 18 hours. These LC bearing antigen that have migrated from the skin can be identified and localised by their fluorescence [7, 41, 42]. Dissecting this further involved studies of the earliest alterations in the SIS after exposure to carcinogenic agents. Analysis of the immune events demonstrated that cutaneous exposure to an antigen following UV-B treatment resulted in antigen specific immunosuppression associated with regulatory T cells [43]. Chemical as well as physical carcinogens also involved the production of antigen specific suppression, as application of the complete carcinogen dimethylbenz(a)anthracene (DMBA) to mouse skin prior to the application of antigen resulted in the generation of antigen specific regulatory T cells [44].

After both UV-B and DMBA treatment of skin, LC depletion occurs due to migration of LC to the regional lymph nodes [45]. These are gradually replaced in the epidermis by immature LC, which, on further exposure to antigen, migrate to the local LN and induce immunosuppression. This has been well documented after both DMBA and UV treatment of skin using the contact sensitivity model outlined above [6, 42, 46]. Further analysis of the immune response after DMBA treatment showed that LC conveyed less antigen to the draining lymph nodes, failed to produce IL-1β and had reduced expression of the co-stimulatory molecule CD86 [42, 47] and they also formed fewer clusters with CD4+ T cells. Likewise, UV-B results in similar changes in LC function [48]. However, UV-B has other immunosuppressive related effects not yet found with DMBA, such as the generation of cis-urocanic acid (cis-UCA) [49] and IL-10 [50], both of which are immunosuppressive molecules. Furthermore, mast cells play a critical role in UV-B induced immunosuppression as the production of cis-UCA results in the release of neuropeptides such as calcitonin gene related protein (CGRP) from the sensory nerves, which then act on mast cells [51] to induce degranulation and systemic suppression of contact hypersensitivity responses [52].

The precise molecular events of UV-B and DMBA induced immune suppressive mechanisms and regulatory T cells requires further study. Yet, what is clear is that both UV-B and chemical carcinogens disrupt the local immune response by altering LC structure and function, resulting in both cases in the generation of tolerance to growing aberrant cells. In the biology of tumour development these events are associated with tumour promotion and the establishment of a tumour [7].

SIS and human skin tumours

Ultraviolet light is well documented as the major aetiological cause of both melanoma and non-melanoma skin cancer and, as already indicated, while causing immunosuppressive effects, it also damages DNA leading to mutations. However, the genesis of skin tumours across time is poorly understood. While molecular events have been defined such as alterations in p53 in squamous cell carcinoma [57] and p16 in melanoma [58], the antigenic changes occurring during cutaneous tumour development are poorly defined. As already indicated, with experimental carcinogenesis the promotional phase of aberrant cells leading to tumour development after UV treatment, or chemical carcinogen application, is associated with immunosuppression and immune tolerance. At what stage similar events occur during the genesis of human skin tumours remains unclear. However, as already noted, established skin tumours can spontaneously regress, which may be a consequence of an active immune attack against the aberrant cells.

At present melanoma antigens are the best characterised. These include tumour specific, shared antigens which are expressed on other tumours besides melanomas. Examples are MAGE, BAGE and GAGE. Antigens specific to normally differentiated melanocytes as well as melanoma cells include tyrosinase, Melan-A/Mart-t, gp100, and gp75. Individual tumours also express unique mutations [59]. The MAGE family is probably the best defined with the genes encoded predominantly on the X chromosome but also on chromosomes 3 and 5 [60]. These melanoma antigens can be associated with both MHC-I and MHC-II in the development of anti-tumour immunity [61, 62]. CD8+ cytotoxic T lymphocytes derived from the peripheral blood of melanoma patients have been used to characterise these tumour antigens and demonstrated that these cells are able to recognise and destroy melanoma cells in vitro [63]. The development of tetramer technology has allowed the more precise determination of T cells which interact with melanoma antigens. However, in the study of Palmoswki and colleagues, only 0.01% of peripheral blood lymphocytes recognised Melan-A; other melanoma related antigens were not detected [64]. This low frequency of melanoma specific antigen reactive T cells probably reflects the inability of the immune system to mount an effective anti-tumour response.

In non-melanoma skin cancers the antigens are not well characterised, except for the potential tumour antigen p53. Mutated p53 is found in > 90% of SCCs and in most BCCs [57]. CD8 cytotoxic T cells responsive to p53 have been described and p53 specific CTLs demonstrated to recognise and kill squamous cell carcinoma cell lines [65]. However, their effectiveness in controlling tumour cell growth is in doubt. In the coming years other antigens recognised by the immune system will be found in melanoma and non-melanoma skin tumours. Micro-array and proteomic technology will further refine the range of antigens altered during the malignant cutaneous process, providing potential immune targets.

For an effective cutaneous immune response to develop, the tumour antigens must be processed and presented to specific reactive T cells by the LC. Alteration in the number of LCs have been documented in pre-malignant, solar keratosis, carcinoma in situ, Bowen’s disease and invasive skin cancers, squamous and basal cell carcinomas. In general the LC numbers in basal and squamous cell carcinoma are higher than those in solar keratosis and Bowen’s disease. Interestingly, those in basal cell carcinomas are present just above the tumour stromal interface [66]. In melanomas LC are found irregularly distributed along the epidermis associated with tumour cells in superficial spreading melanoma [67]. Plasmacytoid dendritic cells have also been detected in melanomas while absent from the adjacent normal skin and naevi [67].

Currently there are critical deficiencies in our knowledge of LC and other dendritic cells associated with cutaneous tumours. Research into the function of LC within skin tumours has received limited attention. A related study showed that high numbers of LC infiltrating cervical tumours was associated with a more favourable prognosis [68], however, the tumours did eventually progress. The lack of mature dendritic cells within tumours has been hypothesised to result in poor anti-tumour immunity [69]. Yet it is the immature dendritic cells which are most likely to have the greatest potential to acquire tumour antigen, but if these cells are unable to mature following activation then effective anti-tumour immunity will not develop. Analysis of the antigen uptake and processing ability of LC from human tumours is problematic as it is difficult to isolate tumour infiltrating LC without inducing phenotypic changes and activation. To achieve T cell activation, LC within skin lesions must migrate to the lymph node. Thus a further issue requiring analysis is the ability of LC to migrate in and out of these lesions. It is likely that many of these cells are held in situ in skin tumours by cytokines such as TGFβ [70]. Newer techniques allowing these cells to be studied in situ should enlighten some of these problems.

To return to the issue of tumour regression and immunosurveillance, an alteration in the immune profile of regressing and non-regressing tumours has been documented. As already indicated, CD8+ cytotoxic T cells can be identified with an ability to kill melanoma and non-melanoma skin cancers. However, what stands out in regressing and non-regressing tumours is the profile of CD4+ T cells. In areas of actively regressing basal cell carcinomas and malignant melanomas, the cytokine profile associated with these CD4+T cells shows an increased expression of IL-2 and TNFβ [71], reflecting a Th1 response, although the precise cytokines involved may differ. In the case of melanoma, regressing lesions show increased IL-2 and TNFβ compared to non-regressing melanomas. With regressing basal cell carcinomas, interferon-γ mRNA is significantly increased relative to non-regressing BCC. Non-regressing lesions where the tumour cells are proliferating and locally invading, fail to show a significant immune response [72]. In an analysis of the Th2 cytokines, IL-10 and IL-13, no difference was demonstrated in regressing melanomas compared to non-regressing melanomas [71]. Similar observations have been made for regressing and non-regressing basal cell carcinomas [73].

On examining malignant lesions of skin, areas of immune attack with localised tumour regression are observed alongside other areas where tumour cells are proliferating and locally invading. The reasons for these distinctively localised events are yet to be clarified but they do have implications on prognostic features of tumours.

Skin tumour escape

Defects in the Fas/Fas-ligand (Fas-L) apoptotic killing pathways may have a key role in the development of UV-induced skin cancer and immune escape. Loss of Fas expression correlates with disease progression in melanoma [76]. Melanoma cells may also by-pass immune-mediated Fas-triggered destruction by a selective blockade of the Fas apoptotic pathway [77].

There is an extensive literature on tumour produced blocking factors that may inhibit immune recognition and destruction of cancer cells. These have also been documented in melanoma, providing a basis for a lack of tumour destruction [6]. Tumours may also produce immunosuppressive factors which prevent cell mediated immunity and tumour destruction. Again, there is now an extensive list of these, but two stand out – IL-10 and TGFβ. IL-10 secreted by melanomas inhibits LC maturation and hence cell mediated immunity [78]. In a mature squamous cell carcinoma model, TGFβ and IL-10 both inhibited LC maturation but only TFGβ prevented LC migration from cultured skin [70, 79]. The antigen uptake, processing and migratory capacity of dendritic cells in skin tumours remains an essentially unchartered field.

The challenges of immunotherapy

References

2 Bos JD, Kapsenberg ML. The skin immune system. Its cellular constituents and their interactions. Immunol Today 1986; 7: 235-40.

3 Streilein JW, Lonsberry LW, Bergstresser PR. Depletion of epidermal Langerhans cells and Ia immunogenicity from tape-stripped mouse skin. J Exp Med 1982; 155: 863-71.

4 Burnet FM. The concept of immunological surveillance. Prog Exp Tumor Res 1970; 13: 1-27.

5 Marshall V. Premalignant and malignant skin tumours in immunosuppressed patients. Transplantation 1974; 17: 272-5.

6 Muller HK, Halliday GM, Woods GM. The skin immune system and tumor immunosurveillance. In: Bos JD, ed. Skin Immune System (SIS). 3rd ed. Boca Raton, Florida: CRC Press, 2004: 475-96.

7 Muller HK, Bucana CD, Kripke ML. Antigen presentation in the skin : modulation by u.v. radiation and chemical carcinogens. Semin Immunol 1992; 4: 205-15.

8 Chu AC, Morris B. The keratinocyte. In: Bos JD, ed. Skin Immune System (SIS). 3rd ed. Boca Raton, Florida: CRC, 2004: 77-100.

9 Steinhoff M, Luger TA. The skin cytokine network. In: Bos JD, ed. Skin Immune System. 3rd ed. Boca Raton, Florida: CRC, 2004: 349-72.

10 Van Loveren H, Redegeld F, Matsuda H, Buckley T, Teppema JS, Garssen J. Mast Cells. In: Bos JD, ed. Skin Immune System (SIS). 2nd ed. Boca Raton, Florida: CRC, 1991: 159-84.

11 Galli SJ. New concepts about the mast cell. N Engl J Med 1993; 328: 257-65.

12 Breathnach SM. Origin, cell lineage, ontogeny, tissue distribution and kinetics of Langerhans cells. In: Schuler G, ed. Epidermal Langerhans Cells. Boca Raton: CRC Press Inc, 1991: 23-48.

13 Katz SI, Tamaki K, Sachs DH. Epidermal Langerhans cells are derived from cells originating in bone marrow. Nature 1979; 282: 324-6.

14 Frelinger JG, Hood L, Hill S, Frelinger JA. Mouse epidermal Ia molecules have a bone marrow origin. Nature 1979; 282: 321-3.

15 Perreault C, Pelletier M, Landry D, Gyger M. Study of Langerhans cells after allogeneic bone marrow transplantation. Blood 1984; 63: 807-11.

16 Hart DN. Dendritic cells : unique leukocyte populations which control the primary immune response. Blood 1997; 90: 3245-87.

17 Merad M, Manz MG, Karsunky H, Wagers A, Peters W, Charo I, et al. Langerhans cells renew in the skin throughout life under steady-state conditions. Nat Immunol 2002; 3: 1135-41.

18 Romani N, Holzmann S, Tripp CH, Koch F, Stoitzner P. Langerhans cells-dendritic cells of the epidermis. APMIS 2003; 111: 725-40.

19 Teunissen MB. Langerhans cells and other skin dendritric cells. In: Bos JD, ed. Skin Immune System (SIS). 3rd ed. Boca Raton, Florida: CRC Press, 2004: 123-82.

20 Valladeau J, Dezutter-Dambuyant C, Saeland S. Langerin/CD207 sheds light on formation of Birbeck granules and their possible function in Langerhans cells. Immunol Res 2003; 28: 93-107.

21 Watts C, Marsh M. Endocytosis : what goes in and how? J Cell Sci 1992; 103: 1-8.

22 Swanson JA, Watts C. Macropinocytosis. Trends Cell Biol 1995; 5: 424-8.

23 Kato M, Neil TK, Clark GJ, Morris CM, Sorg RV, Hart DNJ. cDNA cloning of human DEC-205, a putative antigen-uptake receptor on dendritic cells. Immunogenetics 1998; 47: 442-50.

24 Valladeau J, Ravel O, Dezutter-Dambuyant C, Moore K, Kleijmeer M, Liu Y, et al. Langerin, a novel C-type lectin specific to Langerhans cells, is an endocytic receptor that induces the formation of Birbeck granules. Immunity 2000; 12: 71-81.

25 Fanger NA, Wardwell K, Shen L, Tedder TF, Guyre PM. Type I (CD64) and type II (CD32) Fc gamma receptor-mediated phagocytosis by human blood dendritic cells. J Immunol 1996; 157: 541-8.

26 Esposito-Farese ME, Sautes C, De la Salle H, Latour S, Bieber T, De la Salle C, et al. Membrane and soluble Fc gamma RII/III modulate the antigen-presenting capacity of murine dendritic epidermal Langerhans cells for IgG-complexed antigens. J Immunol 1995; 155: 1725-36.

27 Reis S, Stahl PD, Austyn JM. Phagocytosis of antigens by Langerhans cells in vitro. J Exp Med 1993; 178: 509-19.

28 Aderem A, Underhill DM. Mechanisms of phagocytosis in macrophages. Annu Rev Immunol 1999; 17: 593-623.

29 Lien E, Means TK, Heine H, Yoshimura A, Kusumoto S, Fukase Ket al. Toll-like receptor 4 imparts ligand-specific recognition of bacterial lipopolysaccharide. J Clin Invest 2000; 105: 497-504.

30 Takeuchi O, Hoshino K, Kawai T, Sanjo H, Takada H, Ogawa T, et al. Differential roles of TLR2 and TLR4 in recognition of gram-negative and gram-positive bacterial cell wall components. Immunity 1999; 11: 443-51.

31 Romani N, Koide S, Crowley M, Witmer-Pack M, Livingstone AM, Fathman CG, et al. Presentation of exogenous protein antigens by dendritic cells to T cell clones. Intact protein is presented best by immature, epidermal Langerhans cells. J Exp Med 1989; 169: 1169-78.

32 Heath WR, Carbone FR. Cytotoxic T lymphocyte activation by cross-priming. Curr Opin Immunol 1999; 11: 314-8.

33 Cumberbatch M, Dearman RJ, Griffiths CE, Kimber I. Epidermal Langerhans cell migration and sensitisation to chemical allergens. APMIS 2003; 111: 797-804.

34 Jakob T, Ring J, Udey MC. Multistep navigation of Langerhans/dendritic cells in and out of the skin. J Allergy Clin Immunol 2001; 108: 688-96.

35 Ratzinger G, Stoitzner P, Ebner S, Lutz MB, Layton GT, Rainer Cet al. Matrix metalloproteinases 9 and 2 are necessary for the migration of Langerhans cells and dermal dendritic cells from human and murine skin. J Immunol 2002; 168: 4361-71.

36 Banchereau J, Steinman RM. Dendritic cells and the control of immunity. Nature 1998; 392: 245-52.

37 Picker LJ. Control of lymphocyte homing. Curr Opin Immunol 1994; 6: 394-406.

38 Morales J, Homey B, Vicari AP, Hudak S, Oldham E, Hedrick J, et al. CTACK, a skin-associated chemokine that preferentially attracts skin-homing memory T cells. Proc Natl Acad Sci USA 1999; 96: 14470-5.

39 Saint-Mezard P, Krasteva M, Berard F, Dubois B, Kaiserlian D, Nicolas JF. Allergic Contact Dermatitis. In: Bos JD, ed. Skin Immune System (SIS). Boca Raton, Florida: CRC Press, 2004: 593-614.

40 Kripke ML, Fisher MS. Immunologic parameters of ultraviolet carcinogenesis. J Natl Cancer Inst 1976; 57: 211-5.

41 Macatonia SE, Edwards AJ, Knight SC. Dendritic cells and the initiation of contact sensitivity to fluorescein isothiocyanate. Immunology 1986; 59: 509-14.

42 Woods GM, Doherty KV, Malley RC, Rist MJ, Muller HK. Carcinogen-modified dendritic cells induce immunosuppression by incomplete T-cell activation resulting from impaired antigen uptake and reduced CD86 expression. Immunology 2000; 99: 16-22.

43 Kripke ML. T cell suppressors in ultraviolet carcinogenesis. Transplant Proc 1974; 16: 474-5.

44 Halliday GM, Muller HK. Sensitization through carcinogen induced Langerhans cell deficient skin activates long lived suppressor cells for both cellular and humoral immunity. Cell Immunol 1987; 109: 206-21.

45 Dandie GW, Clydesdale GJ, Radcliff FJ, Muller HK. Migration of Langerhans cells and gammadelta dendritic cells from UV-B-irradiated sheep skin. Immunol Cell Biol 2001; 79: 41-8.

46 Woods GM, Qu M, Ragg SJ, Muller HK. Chemical carcinogens and antigens induce immune suppression via Langerhans’ cell depletion. Immunology 1996; 88: 134-9.

47 Ragg SJ, Woods GM, Egan PJ, Dandie GW, Muller HK. Failure of carcinogen-altered dendritic cells to initiate T cell proliferation is associated with reduced IL-1 beta secretion. Cell Immunol 1997; 178: 17-23.

48 Muller HK, Bucana CD, Kripke ML, Cox PA, Saijo S, Strickland FMUltraviolet irradiation of murine skin alters cluster formation between lymph node dendritic cells and specific T lymphocytes. Cell Immunol 1994; 157: 263-76.

49 De Fabo EC, Noonan FP. Mechanism of immune suppression by ultraviolet irradiation in vivo. J Exp Med 1983; 157: 84-98.

50 Rivas JM, Ullrich SE. Systemic suppression of delayed-type hypersensitivity by supernatants from UV-irradiated keratinocytes. An essential role for keratinocyte-derived IL-10. J Immunol 1992; 149: 3865-71.

51 Khalil Z, Townley SL, Grimbaldeston MA, Finlay-Jones JJ, Hart PHcis-Urocanic acid stimulates neuropeptide release from peripheral sensory nerves. J Invest Dermatol 2001; 117: 886-91.

52 Townley SL, Grimbaldeston MA, Ferguson I, Rush RA, Zhang SH, Zhou XF, et al. Nerve growth factor, neuropeptides, and mast cells in ultraviolet-B-induced systemic suppression of contact hypersensitivity responses in mice. J Invest Dermatol 2002; 118: 396-401.

53 Fairley GH, Lewis MG, Ikonopisov RL, Nairn RC, Alexander P. Detection of tumor specific immune reactions in human melanoma. Ann N Y Acad Sci 1971; 177: 286-9.

54 Nairn RC, Guli EP, Nind AP, Muller HK, Rolland JM, Minty CC. Specific immune response in human skin carcinoma. BMJ 1971; 4: 701-5.

55 Curson C, Weedon D. Spontaneous regression in basal cell carcinomas. J Cutan Pathol 1979; 6: 432-7.

56 McGovern VJ. Spontaneous regression of melanoma. Pathology 1975; 7: 91-9.

57 Urosevic M, Dummer R. Immunotherapy for nonmelanoma skin cancer : does it have a future? Cancer 2002; 94: 477-85.

58 Melnikova VO, Bolshakov SV, Walker C, Ananthaswamy HN. Genomic alterations in spontaneous and carcinogen-induced murine melanoma cell lines. Oncogene 2004; 23: 2347-56.

59 Halliday GM. Skin immunity and melanoma development. In: Thompson JF, Morton DL, Kroon BBR, eds. Textbook of Melanoma. London: Martin Dunitz Publications, 2004: 25-42.

60 Chomez P, De Backer O, Bertrand M, De Plaen E, Boon T, Lucas S. An overview of the MAGE gene family with the identification of all human members of the family. Cancer Res 2001; 61: 5544-51.

61 Storkus WJ, Zeh HJ, Maeurer MJ, Salter RD, Lotze MT. Identification of human melanoma peptides recognized by class I restricted tumor infiltrating T lymphocytes. J Immunol 1993; 151: 3719-27.

62 Slingluff Jr. CL, Cox AL, Henderson RA, Hunt DF, Engelhard VH. Recognition of human melanoma cells by HLA-A2.1-restricted cytotoxic T lymphocytes is mediated by at least six shared peptide epitopes. J Immunol 1993; 150: 2955-63.

63 Degiovanni G, Hainaut P, Lahaye T, Weynants P, Boon T. Antigens recognized on a melanoma cell line by autologous cytolytic T lymphocytes are also expressed on freshly collected tumor cells. Eur J Immunol 1990; 20: 1865-8.

64 Palmowski M, Salio M, Dunbar RP, Cerundolo V. The use of HLA class I tetramers to design a vaccination strategy for melanoma patients. Immunol Rev 2002; 188: 155-63.

65 Black AP, Ogg GS. The role of p53 in the immunobiology of cutaneous squamous cell carcinoma. Clin Exp Immunol 2003; 132: 379-84.

66 McArdle JP, Muller HK. Quantitative assessment of Langerhans’ cells in human cervical intraepithelial neoplasia and wart virus infection. Am J Obstet Gynecol 1986; 154: 509-15.

67 Vermi W, Bonecchi R, Facchetti F, Bianchi D, Sozzani S, Festa S, et al. Recruitment of immature plasmacytoid dendritic cells (plasmacytoid monocytes) and myeloid dendritic cells in primary cutaneous melanomas. J Pathol 2003; 200: 255-68.

68 Zeid NA, Muller HK. S100 positive dendritic cells in human lung tumors associated with cell differentiation and enhanced survival. Pathology 1993; 25: 338-43.

69 Bell D, Chomarat P, Broyles D, Netto G, Harb GM, Lebecque Set al. In breast carcinoma tissue, immature dendritic cells reside within the tumor, whereas mature dendritic cells are located in peritumoral areas. J Exp Med 1999; 190: 1417-26.

70 Halliday GM, Le S. Transforming growth factor-beta produced by progressor tumors inhibits, while IL-10 produced by regressor tumors enhances, Langerhans cell migration from skin. Int Immunol 2001; 13: 1147-54.

71 Lowes MA, Bishop GA, Crotty K, Barnetson RS, Halliday GM. T helper 1 cytokine mRNA is increased in spontaneously regressing primary melanomas. J Invest Dermatol 1997; 108: 914-9.

72 Hunt MJ, Halliday GM, Weedon D, Cooke BE, Barnetson RS. Regression in basal cell carcinoma : an immunohistochemical analysis. Br J Dermatol 1994; 130: 1-8.

73 Wong DA, Bishop GA, Lowes MA, Cooke B, Barnetson RS, Halliday GM. Cytokine profiles in spontaneously regressing basal cell carcinomas. Br J Dermatol 2000; 143: 91-8.

74 Boon T, Gajewski TF, Coulie PG. From defined human tumor antigens to effective immunization. Immunol Today 1995; 16: 334-5.

75 Ugurel S, Reinhold U, Tilgen W. HLA-G in melanoma : A new strategy to escape from immunosurveillance? Onkologie 2002; 25: 129-34.

76 Owen-Schaub LB, Van Golen KL, Hill LL, Price JE. Fas and Fas ligand interactions suppress melanoma lung metastasis. J Exp Med 1998; 188: 1717-23.

77 Ferrarini M, Imro MA, Sciorati C, Heltai S, Protti MP, Pellicciari Cet al. Blockade of the Fas-triggered intracellular signaling pathway in human melanomas is circumvented by cytotoxic lymphocytes. Int J Cancer 1999; 81: 573-9.

78 Chen QY, Daniel V, Maher DW, Hersey P. Production of IL-10 by melanoma cells : examination of its role in immunosuppression mediated by melanoma. Int J Cancer 1994; 56: 755-60.

79 Alleva DG, Burger CJ, Elgert KD. Tumor-induced regulation of suppressor macrophage nitric oxide and TNF-alpha production. Role of tumor-derived IL-10, TGF-beta, and prostaglandin E2. J Immunol 1994; 153: 1674-86.

80 Nestle FO, Alijagic S, Gilliet M, Sun Y, Grabbe S, Dummer R, et al. Vaccination of melanoma patients with peptide- or tumor lysate-pulsed dendritic cells. Nat Med 1998; 4: 328-32.

81 Grabbe S, Bruvers S, Gallo RL, Knisely TL, Nazareno R, Granstein RD. Tumor antigen presentation by murine epidermal cells. J Immunol 1991; 146: 3656-61.