What is the 'true' function of skin?

Conventional textbook wisdom portrays the skin as an organ that literally enwraps whatever each of us stands for as a more or less functional, individual member of the mammalian species, and has it that the skin primarily establishes, controls and transmits contacts with the external world. In addition, the skin has long been recognized to protect the organism from deleterious environmental impacts (physical, chemical,microbiological), and is well-known as crucial for the maintenance of temperature, electrolyte and fluid balance. Now, ever more studies are being published that show the skin to also operate as a huge and highly active biofactory for the synthesis,processing and/or metabolism of an astounding range of e.g. structural proteins, glycans, lipids and signaling molecules. Increasingly, it becomes appreciated that the skin, furthermore, is an integral component of the immune, nervous and endocrine systems, with numerous lines of cross-talk between these systems established intracutaneously (e.g. Ann NY Acad Sci Vol 885, 1999; Endocrine Rev 21:457-487, 2000; Physiol Rev 80:980-1020, 2001; Exp Dermatol 10: 349-367, 2001). All these emerging cutaneous functions beyond the classical image of the skin as a barrier and sensory organ are immediately relevant for many of the quandaries that clinical dermatology, dermatopathology, and dermatopharmacology are still struggling with to-date, and offer the practising dermatologist attractive new targets for therapeutic intervention. Yet, many of these skin functions are not even mentioned in dermatology textbooks and await systematic therapeutic targeting. Following a suggestion by Enno Christophers, the current 'Controversies' feature brings together an unusually diverse council of biologists and clinicians, who share their thought-provoking views with the readers and allow us to peek into the future of research in cutaneous biology, not the least by reminding us of the -- often ignored -- evolutionary and embryonal origins of our favorite organ. Hopefully, this unique discussion feature will foster an understanding of the 'true' skin functions that is both more comprehensive and more profound than conventional teaching on this topic, and will stimulate more than 'skin-deep' reflections on the full range of skin functions.

Free hapten molecules are dispersed by way of the bloodstream during contact sensitization to fluorescein isothiocyanate.

The fate of the contact sensitizer fluorescein isothiocyanate was traced by means of fluorescence spectrophotometry and flow cytometry. The hapten applied to one ear rapidly entered the circulation by way of local lymphatics and blood vessels. It was dispersed for several hours essentially as free hapten, released from a reservoir left behind at the site. Hapten molecules coupled to plasma proteins while circulating and reacted with white blood cells. Total cells of regional lymph nodes, spleen, and distant lymph nodes became fluorescent in successive order. Fluorescence of CD11c-positive dendritic cells exceeded significantly that of lymphoid cells. Total spleen cells and total nonregional lymph node cells were shown in vitro to drive committed lymph node cells to proliferation. The mechanism disclosed is proposed to counterbalance the action of epidermal Langerhans cells for regulation of contact hypersensitivity.

Contact tolerance.

Contact tolerance describes an immunological state which is caused by ordinary contact allergens painted in doses too low to sensitize, either once or repeatedly, onto healthy intact skin. The tolerance state accomplished by this means in BALB/c and C57BI/6 mice was found to be mediated by hapten-specific T cells that adoptively transferred tolerance to naive recipients. Furthermore, these cells were shown to be sensitive to cyclophosphamide, to express the Lyt2+ (CD8) phenotype, and to produce, upon restimulation in vitro, predominantly anti-inflammatory cytokines such as IL-4, IL-5 and IL-10. These data indicate that contact tolerance of the low zone type is actively mediated by Th2-like CD8 T cells rather than arising as a consequence of clonal anergy. The induction of contact tolerance appeared to be strictly dose-dependent. As opposed to sensitizing doses of allergen, subsensitizing doses did not involve epidermal Langerhans cells discernibly. This was suggested by their normal ultrastructure, their unaffected adenosine triphosphatase system, the inefficacy of functional blocking and excision experiments. Both radiolabeled and fluorescent contact sensitizers were observed to readily enter the bloodstream, thereby being dispersed throughout the body. Presumably, contact tolerance is induced systemically rather than locally. The presence of hapten-specific tolerance can only be uncovered through a subsequent attempt to sensitize. If the attained sensitization turns out to be significantly lower than that of immunologically naive controls, and if sensitization to chemically unrelated sensitizers is not impaired, hapten-specific tolerance does exist. Thus, contact tolerance is a result obtained from experimental sensitization in animals. Nonetheless, it is assumed to occur also in humans, although it is not demonstrable unless different proofs of existence become available.

Induction of low zone tolerance to contact allergens in mice does not require functional Langerhans cells.

Epidermal Langerhans cells are known to be the major controlling element in the development of contact hypersensitivity. Haptenic molecules permeating the skin are taken up locally by Langerhans cells and then presented to T lymphocytes in the regional lymph nodes. Despite the presence of functional Langerhans cells, however, subsensitizing doses of hapten applied epicutaneously induce tolerance. We examined epidermal Langerhans cells at the site of contact with picryl chloride or oxazolone in BALB/c and C57B1/6 mice with regard to their responding to either subsensitizing or sensitizing doses of allergen. Subsensitizing doses did not interfere with the membranous adenosine triphosphatase system on Langerhans cells, known to relate to functional readiness of the cell. Accordingly, on electron microscopy the ultrastructure of Langerhans cells was found to be like that in untreated skin. In contrast, sensitizing doses caused a significant depletion of adenosine triphosphatase-positive Langerhans cells, and electron microscopy revealed marked cellular activation of Langerhans cells, with enlarged nuclei and increased numbers of mitochondria and Birbeck granules. Furthermore, subsensitizing doses induced tolerance regardless of whether Langerhans cells were functionally intact or had their function blocked arbitrarily. Blocking was achieved either by preceding ultraviolet B irradiation at the site of application or by painting of a sensitizer before painting another sensitizer on the same site. Moreover, not even surgical removal of the site within minutes after painting could prevent the induction of tolerance. The data suggest that subsensitizing doses of contact allergens painted on normal murine skin bypass involvement of epidermal Langerhans cells.

Low zone tolerance to contact allergens in mice: a functional role for CD8+ T helper type 2 cells.

Normal skin is permeable to low molecular hydrophobic substances, including allergenic chemicals. Whereas such foreign matter appears to enter the skin naturally, it rarely induces contact hypersensitivity. This suggests that immunological tolerance would be the normal state of affairs. In search of a suitable model, we painted picryl chloride or oxazolone once or repeatedly on normal skin of BALB/c or C57B1/6 mice and found subsensitizing doses to be tolerogenic. The most effective doses in inducing tolerance were doses between those at the point of inflection from no responses to threshold sensitivity. But even doses three orders of magnitude lower than these suppressed subsequent sensitization if applied repeatedly. C57B1/6 mice (low responders) were consistently easier to make tolerant than BALB/c mice (high responders). The tolerant state established by a single painting was found to be fully developed at 48 h after initiation and long-lasting (>14 d). It could be adoptively transferred by intravenous injection of total spleen cells (SC), lymph node cells (LNC), or purified T cells and shown to be hapten specific. Pretreatment with cyclophosphamide (Cy) prevented tolerization. The T cells capable of transferring suppressive activity were found to be generated irrespective of the dose applied. On day 2 after painting, tolerance could be transferred with LNC from both tolerant and sensitized animals. On day 5, however, only cells from tolerant donors transferred tolerance. But by action of Cy, suppression was shown to be part of every sensitization, although masked. Production of hapten-specific antibodies was suppressed as well. Through depletion by monoclonal antibody in vitro the T suppressor cells were shown to belong to the murine CD8+ subset (Lyt2+). Upon restimulation in vitro by haptenized and irradiated normal SC, LNC from tolerant donors produced predominantly interleukin (IL)-4, IL-5, and IL-10. In contrast, LNC from sensitized donors produced preferentially IL-2 and interferon-gamma. Thus we demonstrate that painting subsensitizing doses of contact sensitizers on normal murine skin generates CD8+ Th2-like cells that give rise to hapten-specific tolerance. The model may have broader significance and apply to other species, including humans.

Prognostic value of tuberculin and BCG immunoreactivity in stage I high-risk malignant melanoma (EORTC protocol 18781).

BACKGROUND: Since the immune system is considered to be a major determinant in the outcome of malignant melanoma, vaccination with BCG (bacillus Calmette-Guérin) or more recently with interleukins is used in this condition, although the effects of this immunotherapy are unclear. OBJECTIVE: The present study was to elucidate clinical aspects and the prognostic relevance of the cellular and humoral immune responses of melanoma patients treated with two different BCG vaccines. METHODS: A subgroup of patients from a multicenter adjuvant trial with BCG in stage I (pT3-4N0M0) high-risk melanoma was prospectively subjected to detailed analysis regarding tuberculin (PPD, purified protein derivate) skin test reactivity, local, regional and systemic reactions to BCG vaccination, PPD antibody response and disease-free survival. Patients were randomized into three arms and received either no adjuvant treatment (22 patients), BCG RIV (40 patients) or BCG Pasteur (44 patients). All patients were followed for up to 10 years (median follow-up 6 years). RESULTS: Patients treated with BCG Pasteur mounted a stronger antibody response, experienced stronger regional and systemic reactions to vaccination and converted more frequently to positive PPD skin tests, compared to controls and to patients vaccinated with BCG RIV. All BCG-treated patients who developed an antibody response had a longer disease-free interval (p = 0.05), with slightly higher significance for BCG Pasteur-treated patients (p = 0.02). CONCLUSIONS: Cellular as well as humoral immune responses to PPD and BCG thus identify stage I malignant melanoma patients with an overall better prognosis.

Long-term adjuvant immunotherapy in stage I high risk malignant melanoma, comparing two BCG preparations versus non-treatment in a randomised multicentre study (EORTC Protocol 18781).

The present study reports the results of a multicentre adjuvant trial with BCG (Bacillus Calmette-Guérin) in high risk patients (Breslow thickness > or = 1.5 mm, Clark level > or = III) with malignant melanoma, after surgical removal of their primary tumour. The trial was specifically designed in order to resolve the controversy and to provide some definite answers regarding the value of adjuvant BCG treatment in stage I malignant melanoma. Patients were randomised to either BCG RIV (108 patients) or BCG Pasteur (109 patients) for 3 years or to follow-up only (110 patients). The two vaccines used had greatly divergent properties regarding their mode of preparation, their composition and their immunomodulating activities. Of the 353 randomised patients, 23 were ineligible, 3 refused participation after randomisation and 327 were evaluable for final analysis. Median follow-up time was 6 years (range 0-10 years). The log-rank test comparison showed no statistical difference between the three arms regarding time to progression (P = 0.55) and duration of survival (P = 0.82). Treatment was generally well tolerated, with no major adverse events in either treatment arm. These findings confirm data with different BCG preparations and with stage II melanoma which also demonstrated no benefit regarding patient survival and time to relapse.

Metastasizing eccrine porocarcinoma. Report of two cases with fatal outcome.

The clinical, histopathological and ultrastructural features of a metastasizing eccrine porocarcinoma are described in a 75-year-old woman and a 82-year-old woman. The previously not recognized tumors were identified by its distinct pattern of metastasis. Metastatic spreading was restricted for years to a circumscribed region of the skin. Histology and electron microscopy disclosed pronounced epidermotropism of the PAS-positive tumor cells. The characteristic pattern of dissemination obviously represents a homing phenomenon of the tumor cells derived from the intraepidermal sweat gland duct.

Mutation rate estimates are not compatible with autosomal dominant inheritance of the dysplastic nevus "syndrome".

Dysplastic nevi represent precursor lesions harboring an increased risk of evolving into melanoma. Their association with familial melanoma is usually considered a monogenic syndrome with autosomal dominant transmission. To test this concept we estimated the mutation rates. When derived directly from the sporadic occurrence of the trait, the mutation rate is exceedingly high (0.9%-2.5%), whereas, as estimated with the aid of Haldane's formula it would be 0.007% to 0.02%. Accordingly, newly arising mutation would outnumber eliminated mutations by 100:1. Even if only 80% of all old mutations are passed onto the next generation, this ratio of 100:1 would rapidly change. After only a few generations, 10% of the world population should be affected with the dysplastic nevus "syndrome". The apparent lack of a genetic equilibrium between newly arising and eliminated mutations is not compatible with autosomal dominant inheritance of the dysplastic nevus "syndrome."

Characterization and differential expression of an endothelial cell-specific surface antigen in continuous and sinusoidal endothelial, in skin vascular lesions and in vitro.

Continuous and sinusoidal endothelial cells display marked morphological and functional heterogeneity as to their plasmalemmal vesicle content, to the kind of intercellular junctional complexes, to the existence and kind of fenestrae and gaps, to the existence and character of their basement membrane, to their ability for phagocytosis and to other functional parameters. Monoclonal antibody 1F10, raised against human umbilical vein endothelial cells (HUVE cells), reflects these differences in recognizing--without any nonendothelial side reactions--an endothelial cell surface antigen, abundantly expressed in continuous endothelia, low and inconsistently expressed in liver sinusoidal and dermal lymphatic endothelia and absent from splenic sinusoidal endothelial cells. In differentiated skin vascular tumors, 1F10 antigen is expressed in normal amounts while it is only low and inconsistently expressed in the dedifferentiated endothelial cells of Kaposi's sarcoma and hemangiosarcoma. HUVE cells in culture, in contrast to their in situ ancestors, express variable amounts of 1F10 antigen. When endothelial cell-conditioned medium (ECC medium) is supplied to HUVE cells in culture, no 1F10 antigen is expressed, while supplementation with fresh serum-containing medium (FSC medium) or cytokines, such as bFGF, suffices to maintain 1F10 expression in 10-70% of the cells. From this we conclude that developmental regulation, environmental influences and cytokine supply contribute to the differentiation and maintenance of the 1F10+ and 1F10-endothelial cell phenotypes, both in vivo and in vitro.

[Sensitivity and specificity of sonographic diagnosis of the lymph nodes in malignant melanoma]. / Sensitivität und Spezifität der sonographischen Lymphknotendiagnostik beim malignen Melanom.

A total of 213 melanoma patients were checked perioperatively with a 5-MHz sonographic scanner in order to detect lymph-node metastases; they were also checked in the scope of tumor follow-up. Of the 415 sonographic results, the method yielded a 97% accuracy. The soundness of lymph-node sonography has been proved histologically and/or through clinical observation. Compared to the other diagnostic techniques available for checking surface and subsurface lymph-node groups, lymph-node sonography is an advantageous combination of diagnostic practicability, accuracy, economic feasibility, and patient acceptance. The possible therapeutic implications are also discussed in this paper.

To perfuse or not to perfuse? A retrospective comparative study to evaluate the effect of adjuvant isolated regional perfusion in patients with stage I extremity melanoma with a thickness of 1.5 mm or greater.

The use of isolated regional perfusion in an adjuvant setting for stage I melanoma of the extremity continues to be controversial. The present retrospective study evaluates the past 20 years' experience by comparing 227 perfused patients from Groningen with 238 matched controls from five hospitals in The Netherlands and Westphalia (a region of West Germany bordering the Netherlands). All patients underwent wide local excision for a primary extremity melanoma of 1.5 mm or greater in thickness. A proportional hazards regression analysis for recurrence of disease and survival identified the significant prognostic factors, of which tumor thickness was the most important. Corrected for these factors, it was not possible to demonstrate a statistically significant effect for perfusion in terms of time to limb recurrence (P = .61), time to regional lymph node metastasis (P = .11), time to distant metastasis (P = .73), disease-free interval (P = .42), and survival (P = .90). No statistically significant differences were seen for adjuvant perfusion in any of the subgroups.

Inflammatory cell infiltrates in human melanoma at different stages of tumor progression.

Progression of human melanoma is associated with changes in antigenic phenotypes of tumor cells. To establish whether inflammatory infiltrates in progressing melanoma also change, we studied 146 cutaneous melanomas at different stages of progression. Monoclonal antibodies (MAbs) against lymphocyte and macrophage subpopulations, interleukin-2 receptor (IL-2 R), immune interferon (IFN-gamma), and the IFN-gamma-inducible, progression-associated melanoma antigens HLA-DR and gp89 were applied in situ. During the course of melanoma progression, decreased amounts of peritumoral T cells, IL-2 R-expressing lymphocytes and dermal T6+ dendritic cells were found, while increased numbers of intratumoral T cells, inflammatory (27E10+) and mature (25F9+) macrophages were associated with local progression of primary melanomas. In metastases, most infiltrate components except 25F9+ macrophages were rare. Positive correlations were observed between: (1) dermal T6+ cells and IL-2 R+ lymphocytes, and (2) presence of IFN-gamma in the infiltrate and HLA-DR and gp89 antigens on tumor cells. In all stages, HLA-DR expression on tumor cells was correlated with: (1) a shift towards T8+ lymphocytes in the infiltrates and (2) a loss of IL-2 R expression. Our data suggest mutual influences between melanoma cells and mononuclear cell infiltrates in situ.

[Malignant melanoma as cancer model]. / Das maligne Melanom als Krebsmodell.

Malignant melanoma is a suitable model for studying the growth of malignant tumors in general, because there are no obstacles to observation during its course and it can easily be obtained. Local progression of the tumor implies stepwise changes from low to high malignancy. However, regression can also be recognized, often simultaneously with progression. Both phenomena can be explained with reference to heterogeneity of tumor cells and instability of phenotypes. These are reflected not only in morphologic differences but also in the production of enzymes and factors and the expression of antigens. Phenotypic dynamics are not haphazard but rather according to definite principles. Levels of early tumor-associated and HLA-A,B,C (class I) antigens decline with advancing progression, whereas those of some late "risk" antigens, including HLA-D (class II) antigens, increase. Tumor and host interact, apparently in alternating directions, until biologic preponderance is established.

[Adjuvant chemotherapy in addition to radical surgical treatment of regional lymph node metastases in malignant melanoma]. / Adjuvante Polychemotherapie zusätzlich zur radikalen operativen Behandlung regionaler Lymphknotenmetastasen beim malignen Melanom.

Thirty-three patients with malignant melanoma and regional lymph node metastases who underwent lymph node dissection were additionally given polychemotherapy with carmustine, hydroxycarbamide and dacarbazine immediately before surgery and up to five times postoperatively. Twenty-nine patients were only treated surgically. These two groups were comparable as regards prognostic criteria, in particular tumour size, ulceration and the number of lymph nodes affected, although the individual follow-up periods varied considerably. The group given chemotherapy showed better results than the control group undergoing surgery alone. The log rank test yielded a significant difference (P less than 0.05) with respect to the probability of relapse-free survival but not as regards probability of survival time. Patients with ulcerated primary melanomas and with a large number of affected lymph nodes had a less favourable prognosis. The major side effects of chemotherapy were transient nausea and bone marrow depression.

Topical immunotherapy changes the composition of the peribulbar infiltrate in alopecia areata.

It has previously been shown that, in patients with untreated progressive alopecia areata (AA), the peribulbar T4/T8 ratio is about 4:1. In the present study, the immunohistochemical findings obtained in untreated AA patients were compared to those obtained in patients who had received topical immunotherapy with diphencyprone. The untreated group consisted of 5 patients with progressive AA and 5 patients with inactive AA. The treated group consisted of 5 patients with a good response to diphencyprone and 5 patients with little or no hair regrowth after treatment. In untreated patients with progressive AA, the mean peribulbar T4/T8 ratio was 4:1, whereas in untreated patients with stable AA, the ratio was 2:1. In the treated patients with a good response to diphencyprone, the mean T4/T8 ratio was 1:1, while in the patients with poor or no response to treatment, the ratio was 0.7. In conclusion, topical immunotherapy considerably alters the peribulbar T4/T8 ratio in AA. The results are consistent with, but do not prove, the concept of topical immunomodulation.

BCG immunotherapy in stage I melanoma patients. Does it influence prognosis determined by HLA-DR expression in high-risk primary tumors?

Previously, we have provided evidence for a positive correlation between HLA-DR expression in primary melanoma and early metastasis. In the present study we investigated whether this relationship was modified by adjuvant BCG immunotherapy. The study comprised 107 patients with a stage I high-risk melanoma; 44 patients had been treated with BCG, whereas the remaining patients had not received any adjuvant therapy. There was no difference in disease-free survival between BCG-treated and untreated patients. Disease-free survival was significantly shorter in patients with high expression of HLA-DR antigens in the primary tumor. Subgrouping BCG-treated and control patients according to HLA-DR phenotype of the melanoma revealed a prolongation of disease-free survival in the subgroup of BCG-treated patients with no or low expression of HLA-DR antigens in the primary melanoma. BCG therapy apparently did not influence prognosis of patients with high expression of HLA-DR antigens in the tumor.