T regulatory cells mediate immunosuppresion by adenosine in peripheral blood, sentinel lymph node and TILs from melanoma patients.

T regulatory cells (Tregs), involved in tumour tolerance, can generate Adenosine by CD39/CD73 surface enzymes, which identify four Tregs subsets: CD39+CD73- nTregs, CD39+CD73+ iTregs, CD39-CD73+ oTregs and CD39-CD73- xTregs. In melanoma patients, increased Tregs levels are detected in peripheral blood (PB), sentinel lymph node (SLN) and tumour infiltrating lymphocytes (TILs), but Adenosine role was not investigated yet. We examined total Tregs and Adenosine subsets in PB, SLN and TILs from melanoma patients (n = 32) and PB from healthy donors (HD; n = 10) by flow cytometry. Total Tregs significantly increased in stage III-IV patients PB, in SLN and TILs, as compared to HD/stage I-II patients. Tregs subsets analyses showed that: 1) PB nTregs significantly increased in SLN and decreased in TILs; 2) iTregs significantly increased in stage III-IV patients PB and further significantly increased in SLN and TILs; 3) PB oTregs and xTregs significantly decreased in SLN and TILs. Patients clinical features did not significantly influence total Tregs, except SLN excision order. Results confirmed Tregs role in melanoma progression and indicate Adenosine generation as a novel escape mechanism, being nTregs and iTregs increased in PB/SLN/TILs.

CD4+FOXP3+ T regulatory cells decrease and CD3+CD8+ T cells recruitment in TILs from melanoma metastases after electrochemotherapy.

Electrochemotherapy (ECT) represents an effective local treatment for skin unresectable melanoma metastases with high overall objective response rate. ECT is based on the combination of anti-neoplastic drugs administration and cancer cells electroporation. Whether ECT can also activate the immune system is a matter of debate, however a significant recruitment of dendritic cells in melanoma treated metastases has been described. Herein we investigated immediate and late effects of ECT treatment on T cell subsets in ECT-treated lesions by fluorescent immunohistochemistry. Biopsies from melanoma patients (n = 10) were taken before ECT (t0), at d1 and d14 from treatment. At t0, CD3+CD4+ T cells were the most represented T cells, well detected in the perilesional dermis, particularly at tumour margin, while CD3+CD8+ T cells were less represented. CD4+FOXP3+ T regulatory (Treg) cells were present in the perilesional dermis and within the lesion. ECT induced a significant decrease of CD4+FOXP3+ Treg cells percentage in the perilesional dermis, observed at d1 and at d14 (p < 0.001). CD3+CD8+ T cells frequency significantly increased at d14 from treatment in the perilesional dermis (p < 0.001). Furthermore calreticulin translocation to the plasma membrane, a hallmark of immunogenic cell death, was observed in metastatic cells after ECT. The data reported here confirm that ECT induces a local response, with a lymphoid infiltrate characterized by CD4+FOXP3+ Treg cells decrease and CD3+CD8+ T cells recruitment in the treated lesions. These results might contribute to design novel combinational therapeutic approaches with ECT and immunotherapy in order to generate a systemic long-lasting anti-melanoma immunity.

'Animal-type' melanoma of the scalp with satellitosis and positive sentinel nodes in a 4-year-old child: case report and review of the literature.

BACKGROUND: Although showing a rapidly rising incidence, paediatric melanoma is relatively rare, accounting for 1-4% of all cases of melanoma and for 1-3% of all paediatric malignancies. The overall survival rate in paediatric patients seems to be similar to that recorded in adults. 'Animal-type' melanoma (ATM) is a rare melanoma subtype, occurring both in childhood and in adults, that shows a close histological resemblance to the heavily pigmented melanocytic tumours observed in grey and white horses. CASE PRESENTATION: We present a case of ATM of the scalp with satellitosis and two positive sentinel nodes in a 4-year-old male child. No other tumour deposits were found in the subsequent regional lymphadenectomy; the patient has been tumour free for 30 months. CONCLUSIONS: We treated our case of ATM in a child as the other types of paediatric melanoma, therefore as an adult melanoma. ATM is generally considered a neoplasm with an indolent course, that occasionally shows an aggressive behaviour, and patient deaths of ATM have been reported. Due to the rarity of ATM, further studies are needed to better define the biological behaviour of this particular melanoma subtype and the therapeutic and follow-up strategies.

Dermatofibrosarcoma protuberans in childhood: two case reports and review of the literature.

Dermatofibrosarcoma protuberans (DFSP) is a rare skin cancer with intermediate malignancy, characterized by a progressive local growth and a propensity for local recurrence. DFSP is most frequent in adults; however, in recent years, DFSP in childhood emerged to be more common than previously believed. Unfortunately DFSP in children may be misdiagnosed, leading to a delay in the treatment. The authors report two cases of childhood DFSP with unusual clinical presentation: a congenital nodular variant and an atrophic variant developed at 2 years of age, both with acral localization. They highlight the importance of an early diagnosis by pediatricians and dermatologists to ensure an appropriate complete excision and reduce the risks of recurrences.

Specific immune therapy-related cutaneous B-cell pseudolymphoma with following dissemination.

We report two cases of cutaneous B-cell pseudolymphoma (PCBCL) induced by intradermal antigen injections for specific immune therapy (SIT). In both cases, the lesions had first developed on the area of injection; years later, new lesions appeared far from the original site. The histological, immunohistochemical, and molecular findings of the lesions showed features consistent with the diagnosis of PCBCL in both cases. In particular, the staining for sIg light chains showed a polyclonal pattern, and the molecular analysis by reverse transcriptase-polymerase chain reaction (RT-PCR) and Southern blot showed a germ-line configuration of the Ig heavy chain genes. While the development of PCBCL related to a specific stimulus is well known and widely reported, the development of histologically and immunohistologically identical lesions far from the injection site is definitely worthy of note. This behaviour might be due to the presence of retained antigens in the injection site. This chronic antigenic stimulation could induce the progressive selection--and subsequent dissemination--of antigen-specific B cell clones.

Cd1d is expressed on dermal dendritic cells and monocyte-derived dendritic cells.

CD1 proteins are a family of cell surface molecules that present lipid antigens to T cells. We investigated skin dendritic cells and monocyte-derived dendritic cells for expression of CD1 molecules using a panel of 10 different monoclonal antibodies focusing on the recently described CD1d molecule. By immunohistochemical analysis, CD1d expression in normal human skin was restricted to dendritic appearing cells in the papillary dermis mainly located in a perivascular localization. Langerhans cells did not show detectable CD1d expression in situ. Epidermal/dermal cell suspensions analyzed by flow cytometry demonstrated distinct subpopulations of HLA-DR positive dermal dendritic cells expressing CD1a, CD1b, and CD1c. CD1d was expressed on HLA-DRbright dermal antigen-presenting cells in dermal suspensions (16% +/- 3.6%), as well as on highly enriched dermal dendritic cells migrating out of skin explants (60.5% +/- 8.0%). Migrated mature dermal dendritic cells coexpressed CD83 and CD1d. Western blot analysis on microdissected skin sections revealed the presence of a 50-55 kDa CD1d molecule in dermis, suggesting that CD1d is highly glycosylated in skin. Both immature and mature monocyte-derived dendritic cells cultured in autologous plasma expressed CD1d molecules. In contrast, culture in fetal bovine serum downregulated CD1d expression. In conclusion, antigen-presenting cells in skin express different sets of CD1 molecules including CD1d and might play a role in lipid antigen presentation in various skin diseases. Differential expression of CD1 molecules depending on culture conditions might have an impact on clinical applications of dendritic cells for immunotherapy.

Control of the differentiation state and function of human epidermal Langerhans cells by cytokines in vitro.

Langerhans cells can originate in vitro from immature precursors stimulated with granulocyte macrophage-colony-stimulating factor (GM-CSF), tumour necrosis factor (TNF)-alpha and stem cell factor (SCF). We asked whether these cytokines also control the differentiation state of Langerhans cells within the epidermis and upon leaving this tissue. We harvested sheets of human epidermis by controlled dispase hydrolysis of keratomes, cultured them in RPMI and 10% fetal calf serum for 48 h and analysed the sheets and the cells migrated spontaneously into the medium, most of which were Langerhans cells containing Birbeck granules. By flow cytometry, the intensity of CD1a expression was reduced quite evenly among Langerhans cells migrated from sheets within 48 h. The cells in the sheets underwent loss of dendrites, with a significant reduction in the cell perimeter that was prevented by GM-CSF and TNF-alpha together. Either of these cytokines induced expression of CD18 by cells in the sheets and those in the medium. Moreover, TNF-alpha induced expression of CD54 by cells in the medium, but not by those retained in the sheets, whereas human SCF induced, dose dependently, expression of CD54 by cells in the sheets, but not from those in the medium. The proliferation of allogeneic lymphocytes was much higher when stimulating Langerhans cells were harvested from cultures with any cytokine, rather than from cultures without cytokines. We conclude the following: (i) GM-CSF and TNF-alpha help to maintain full differentiation of Langerhans cells within the epidermis; (ii) cytokine influence on Langerhans cells adhesiveness is in part context dependent; and (iii) pretreatment with cytokines influences positively the number or accessory activity of Langerhans cells on lymphocytes during subsequent mixed leucocyte reaction.

Human keratinocytes cultured without a feeder layer undergo progressive loss of differentiation markers.

Culture of keratinocytes in conventional medium without a mesenchyme-derived feeder layer leads to poor growth and impaired differentiation; however, the exact pathway and degree of differentiation achieved in such conditions is unclear. We have cultured normal human keratinocytes in Rheinwald and Green's medium, on plastic without a feeder layer, in order to investigate the degree of differentiation that they achieve in these conditions. Intermediate filament proteins, tonofibrils and desmosomes were assumed as markers of differentiation and their expression was analyzed by immunohistochemistry and electron microscopy. Before reaching confluence, keratinocytes expressed keratin molecules, as well as vimentin, and formed tonofibrils and desmosomes. The expression of these markers was progressively reduced until confluence and was totally lost thereafter, while cultures could be propagated for at least six passages. On the contrary, reseeding on a feeder layer after the first passage led to rapid cell death. It could be concluded that signals from a feeder layer are relevant to support continuous synthesis of intermediate filaments proteins and formation of tonofibils and desmosomes, and that the derangement of the cytoskeleton in these conditions leads to altered, not simply defective, response to delayed stimulation by a feeder layer.

Immunophenotypic analysis of normal human dendritic cells isolated from epidermis and dermis.

BACKGROUND: The skin immune system comprises two types of dendritic cells, i.e. CD1a-positive Langerhans cells in the epidermis and CD36-positive dendritic macrophages in the dermis. Dendritic cells can migrate from skin explants into a culture medium. METHODS: We have examined the morphology and immunophenotype of the dendritic cells migrating from epidermal and dermal sheets in vitro. The epidermis and dermis of keratomes of normal human skin were separated with dispase and cultured for 72 h. At this time, the non-adherent cells in the medium were removed, enriched on a metrizamide or Lymphoprep gradient, counted, prepared by cytospin, and labeled for CD1a, CD36, and HLADr. RESULTS: Cells migrating from the epidermis and dermis show many thin projections or a few veils from the cell surface. Approximately four times more cells migrate from epidermal than dermal sheets from the same keratome. CONCLUSIONS: Using methods to separate the epidermis from the dermis, both CD1a-positive Langerhans cells and CD36-positive dendritic macrophages can be obtained from both tissues, although in different numbers.

Cyclosporin-A affects the organization of cytoskeleton of normal human keratinocytes in culture.

Cyclosporin-A (CsA) is a potent immunoregulatory molecule which has been widely used in many immunomediated and inflammatory skin diseases. It inhibits the proliferation of keratinocytes, but its possible effects(s) on cell differentiation are poorly known. To address this issue, we have studied the influence of CsA on the assembly of intermediate filaments by normal human keratinocytes in culture. Control keratinocytes were flat; the cells which had not reached confluence stained intensely for vimentin and weakly for cytokeratins; confluent cells stained with intermediate intensity for both types of proteins and the cells adhering on the top of others, interpreted as the best differentiated ones, stained for cytokeratins but not for vimentin. CsA (1.6 micrograms/ml for 10 days) inhibited the growth of keratinocytes, which never reached confluence; most cells appeared small and roundish, only some stained for cytokeratins and few for vimentin. By electron microscopy, a well organized meshwork of tonofibrils was recognized in many control keratinocytes, but never in CsA-treated keratinocytes. We propose that the cytoskeleton could be a target of CsA and that its alteration mediates other effects of CsA on keratinocytes, including those on cell growth.

Phosphotyrosine protein phosphatases and diabetic pregnancy: an association between low molecular weight acid phosphatase and degree of glycemic control.

Low molecular weight acid phosphatase encoded by the highly polymorphic locus ACP1 is a member of the protein-tyrosin phosphatase family (PTPases) which plays an essential role in the control of receptor signalling through phosphotyrosine pathways. Recent experiments have shown that purified rat liver ACP, corresponding to human ACP1, is able to hydrolyze a phosphotyrosine-containing synthetic peptide corresponding to the 1146-1158 sequence of the human insulin receptor, and shows a high affinity for it. This prompted us to analyze the degree of glycemic control in relation to ACP1 genetic variability in a sample of 214 diabetic pregnant women including IDDM, NIDDM and gestational diabetes. The ACP1 genotype was also determined in 482 non-diabetic pregnant women. In diabetic women glycemic levels in the last trimester of pregnancy appear to be significantly associated with the ACP1 genotype, and correlate positively with ACP1 enzymatic activity. The data suggest that quantitative variations of ACP1 may influence the clinical manifestations of diabetic disorders, and call for further studies on the role of this enzyme in the modulation of insulin-receptor phosphotyrosine pathways.

Evidence of selection on PGM1 polymorphism in diabetic pregnancy.

Ninety-nine pregnant women with insulin-dependent diabetes mellitus, 83 women with gestational diabetes, and a control sample of 315 nondiabetic consecutive puerperae have been studied along with their newborn infants. Neonatal macrosomia is less frequent among diabetic mothers with phosphoglucomutase (PGM1) genotype PGM1*1/*2 than among mothers with other PGM1 genotypes. In gestational diabetes the association is more evident among younger women than among older women. Diabetic women with the PGM1*1/*2 genotype show a reduced proportion of heterozygous PGM1*1/*2 offspring. The phenomenon is much more evident among women under 28 years of age and does not depend on the quality of metabolic control. The data suggest that when both mother and fetus share the PGM1*1/*2 genotype, the deleterious effects of a diabetic environment on fetal development are more severe, leading to an early loss of zygotes. This may contribute to a decrease incidence of macromosomia among live-born infants delivered by PGM1*1/*2 mothers.

Is there a genetic basis for gestational diabetes?

Rh E-PGM1 (phosphoglucomutase locus 1) joint genotype frequencies (chromosome 1) have been determined in 90 women with gestational diabetes mellitus (GDM) and in 140 pregnant women with preexisting insulin-dependent diabetes mellitus (IDDM) and non-insulin-dependent diabetes mellitus (NIDDM). The pattern of Rh E-PGM1 association differs among types of diabetes. The distortion of the pattern tends to assume opposite values in GDM and NIDDM and this is particularly evident in PGM1*2,2 subjects. The proportion of PGM1*2,2 subjects homozygous for the Rh e allele is higher in GDM than in IDDM and NIDDM. IDDM women show an intermediate proportion and NIDDM women show the lowest proportion of this genotype. The opposite pattern is observed among PGM1*2,2 subjects carrying the Rh E allele. Genotype frequencies of the hypervariable region flanking the insulin gene (chromosome 11) have been determined in 77 women with GDM, in 52 pregnant women with preexisting diabetes (IDDM and NIDDM), and in 62 normal adults. In GDM and NIDDM subjects the frequencies are similar. In IDDM women the frequency of homozygotes for the class 1 allele is higher than the frequency found in GDM and NIDDM women. The data suggest a possible genetic basis for the differentiation of a subclass of GDM from both IDDM and NIDDM.

Haptoglobin development in newborn infants from diabetic mothers.

Haptoglobin (Hp) development during the neonatal period has been studied in 325 newborn infants from normal pregnancies and in 242 infants from diabetic mothers. In infants from diabetic mothers Hp development is delayed as compared to infants from normal pregnancies. This delay is associated with a change in the pattern of relationship between Hp development and the polymorphism of acid phosphatase (ACP1) (an enzyme which shows phosphotyrosine phosphatase (PTPase) activity). In infants from normal pregnancies who show ACP1 phenotypes with the highest activity, the appearance of Hp is accelerated as compared to other infants. In contrast, infants from diabetic pregnancies who have ACP1 phenotypes with the highest activity, show delayed Hp development.

Genetic and non genetic factors in the outcome of diabetic pregnancy.

In insulin-dependent diabetes mellitus, maternal Phosphoglucomutase genotype is a predictor of fetal macrosomia much more important than quality of metabolic control during pregnancy. In gestational diabetes and in non insulin-dependent diabetes, on the contrary, the most important predictor is the metabolic control of diabetes.

Both maternal and foetal genetic factors contribute to macrosomia of diabetic pregnancy.

The study of 230 diabetic mothers along with their newborn babies has shown that foetal macrosomia is associated with two specific genomic sites: phosphoglucomutase locus 1 (PGM1)-Rhesus blood group (Rh) linkage group (chromosome 1) and HindIII restriction fragment length polymorphism (RFLP) linked to insulin-like growth factor 1 (IGF1) (chromosome 12). In PGM(1)2-1 mothers carrying the E allele, there is a proportion of 8.7% of macrosomic newborns as compared with 39.6% in mothers with other genotypes. The relationship between the maternal PGM1-RhE genotype and neonatal macrosomia does not depend on the type of diabetes. The proportion of macrosomic infants is much lower among newborns carrying the IGF1HS allele of the HindIII RFLP linked to IGF1 (20%) than among IGF1F/IGF1HF newborns (55%).

Erythrocyte transmembrane Na and K fluxes in pseudohypoaldosteronism.

Pseudohypoaldosteronism (PHA) is a disease characterized by hyponatremia, hypotension, and dehydratation, despite the presence of hyperreninemic hyperaldosteronism. The membrane-bound Na,K ATPase activity and the transmembrane Na and K transport systems have been studied in vitro in red blood cells of two subjects, son and mother, affected by pseudohypoaldosteronism with different degrees of clinical involvement. Both parameters were significantly altered suggesting that the refractory response to mineralocorticoids is detectable, not only in kidneys and salivary and sweat glands, but also in red blood cells. Since pseudohypoaldosteronism, in its asymptomatic form, may be much more common than expected, we suggest the use of the tests described herein as a practical approach to the early diagnosis of pseudohypoaldosteronism in the investigation of sodium wasting syndromes.

Diabetic pregnancy: evidence of selection on the Rh blood group system.

The blood glucose levels of pregnant women with insulin-dependent diabetes mellitus and the blood glucose levels of newborns during the first few hours of life show an association with maternal Rh genotype. Distortions of joint maternal-fetal Rh phenotype distribution have also been observed. Because a cluster of genes involved in glycide metabolism is located on the short arm of chromosome 1, the present observations may reflect the action of these genes.