One partially responded to interferon with persistent erythemato-squamous plaques whereas the other one had advanced disease and was evaluated after he was placed under systemic PCT. with AZ158 monoclonal antibody (moAb), which also recognizes the monomeric CD158e/KIR3DL1 receptor. Although positive in all SS samples, immunohistochemistry appeared not to be enough discriminant between SS and EID. Therefore in all samples disclosing a significant staining with AZ158 moAb, CD158k/KIR3DL2, CD158e/KIR3DL1 and T-plastin mRNA expression were analyzed on the same skin specimen using conventional and/or quantitative real time RT-PCR. Interestingly, only CD158k/KIR3DL2 transcripts were found to be significantly overexpressed in skin biopsies from patients with SS (gene amplification and overexpression (Mao et al.. 2004: Mao et al., 2003), gene deletion (Karenko et al.. 2005) and abnormal expression of various markers at the transcript and/or protein level, such as Twist, and EphA4 (Kari et al., 2003: Nebozhyn et al., 2006). Among them, T-plastin is a member of the Fimbrins/plastins family, which are Taranabant highly conserved actin-bundling proteins. T-plastin is not expressed by normal lymphocytes under physiological conditions and was therefore proposed as a specific molecular marker for neoplastic lymphocytes in SS, but, to the best of our knowledge, was not investigated in lesional skin (Su et al., 2003). Importantly, only few studies to date have focused on cutaneous samples, so that a specific marker for SS in skin biopsies is still lacking. Although KIR3DL2/CD158k actually delineates the malignant clonal T-cell population in SS (Bagot et al., 2001; Ortonne et al., 2006; Poszepczynska-Guigne et al., 2004) and is expressed at the protein level by lymphocytes in erythrodermic skin of patients with SS (Wechsler et al., 2003), it is not known whether it may help for the diagnosis of erythroderma in skin biopsies. To address this issue, we analyzed the expression of CD158k/KIR3DL2 at both protein and transcript levels in a panel of skin biopsies from patients with erythroderma from various causes, and we compared the results obtained in SS benign EID. Results Immunohistochemistry with AZ158 moAb did not clearly discriminate between SS and benign EID We observed a subepidermal band-like or mostly perivascular lymphocytic infiltrate in all 40 skin samples from the SS and the EID groups, associated with epidermotropic lymphocytes in 85%. The density of dermal T-cell infiltrates was categorized as +, ++ and +++ in 0 (0%), 8 (57%) and 6 (43%) samples from the SS group, respectively, Rabbit polyclonal to KIAA0802 and in 11 (42%), 12 (46%) and 3 (12%) samples from the EID group, respectively. This indicated a higher lymphocytic infiltration in SS, as shown in the cutaneous sample of patient 8 in Figure 1c. As previously reported (Wechsler et al., 2003), immunohistochemistry with AZ158 moAb was positive in all skin biopsies from patients with SS (Figure 1e), in agreement with CD158k/KIR3DL2 expression. However, with AZ158 moAb, we also found positive cells in 17 biopsies of the EID group (65%), as shown in figure 1f, showing representative sections of a patient with erythrodermic psoriasis. Mean AZ158/CD3 ratios were similar in both groups in the epidermis. However, in EID we found a lower staining with AZ158 moAb in the dermis, with 33% of SS samples 74% of EID samples showing a AZ158/CD3 ratio 25%, (Students t test: P = 0.0045). SS displayed lower CD8/CD3 ratios Taranabant in both the epidermis and the dermis. Indeed, the mean CD8/CD3 ratio was 12% in SS 38 % in EID (Students t test: P = 0.003) in the epidermis. In the dermis, semi-quantitative evaluation revealed that 83% of SS versus 38% of EID samples Taranabant had a CD8/CD3 ratio 25% (Students t test: P = 0.004). This latter finding is in agreement with the preferential intra-epidermal localization of malignant CD4+ lymphocytes in epidermotropic CTCL such as SS and.