UVB-induced leucocyte trafficking in the epidermis of photosensitive lupus erythematosus patients: normal depletion of Langerhans cells

BACKGROUND: The pathogenic mechanisms of UV-induced skin lesions of lupus erythematosus (LE) are unknown. In a recent study of pathogenic mechanisms of polymorphic light eruption (PLE), significantly more Langerhans cells (LCs) persisted in the epidermis after UVB overexposure than in healthy individuals. Interestingly, the same phenomenon was observed in one subacute cutaneous lupus erythematosus (SCLE) patient. It could therefore be hypothesized that both photodermatoses share a common pathogenic mechanism of photosensitivity. In the present study, we tested this hypothesis by investigating leucocyte trafficking in the initial phase of cutaneous LE after intense UVB exposure. METHODS: In 22 photosensitive LE patients (12 chronic discoid lupus erythematosus, seven systemic lupus erythematosus and three SCLE) and nine age/sex-matched controls, uninvolved buttock skin was exposed to six minimal erythemal dose (MED) UVB radiation. Subsequently, biopsies were taken after 24, 48 and 72 h, and one control biopsy was taken from unirradiated skin. Skin sections were stained for the presence of LCs, neutrophils and macrophages. Areal percentages of positively stained cells within the epidermis were quantified and compared between the patients and controls. RESULTS: A gradual decrease of epidermal LCs and a gradual increase of epidermal neutrophils and macrophages at several timepoints after six MED irradiation was observed equally in both LE patients and controls. CONCLUSION: Immunohistopathology of irradiated uninvolved skin of photosensitive LE patients did not reveal the same pathologic trafficking of LCs and neutrophils as described for PLE patients. We conclude that different mechanisms are operative in the pathogenesis of PLE and photosensitive LE.     

The expression pattern of interferon-inducible proteins reflects the characteristic histological distribution of infiltrating immune cells in different cutaneous lupus erythematosus subsets

BACKGROUND: Plasmacytoid dendritic cells and type I interferons (IFNs) are supposed to play a central proinflammatory role in the pathogenesis of cutaneous lupus erythematosus (LE). The IFN-inducible chemokines CXCL9 and CXCL10 are involved in recruiting CXCR3+ effector lymphocytes from the peripheral blood into skin lesions of LE. We hypothesized that the expression pattern of IFN-inducible proteins reflects the characteristic distribution of the inflammatory infiltrate in different subsets of cutaneous LE. OBJECTIVES: To test this hypothesis in patients with LE. METHODS: Lesional skin biopsies taken from patients with different subsets of LE [chronic discoid LE (CDLE), n = 12; subacute cutaneous LE (SCLE), n = 5; LE tumidus (LET), n = 4; LE profundus (LEP), n = 6] were investigated by immunohistochemistry using monoclonal antibodies to the lymphocyte surface markers CD3, CD4, CD8, CD20 and CD68, the cytotoxic proteins Tia1 and granzyme B, the chemokine receptor CXCR3, the specifically type I IFN-inducible protein myxovirus protein A (MxA) and the chemokines CXCL9 and CXCL10. RESULTS: The expression pattern of MxA followed the distribution of the inflammatory infiltrate typically seen in the investigated cutaneous LE subsets. In CDLE and SCLE, expression was focused in the epidermis and upper dermis, while in LET a perivascular and in LEP a subcutaneous pattern was found. Similar findings were obtained for CXCL9 and CXCL10. CONCLUSIONS: Our results demonstrate a close morphological association between the expression pattern of IFN-inducible proteins and the distribution of CXCR3+ CD3+ lymphocytes in all investigated subsets of cutaneous LE. This supports the importance of an IFN-driven inflammation in this condition. Infiltrating lymphocytes carrying CXCL10 in their granules might amplify the lesional inflammation and be responsible for the chronic course of this disease.     

Apoptosis in different cutaneous manifestations of lupus erythematosus

BACKGROUND: It has been shown that dysregulation of apoptosis may play an important part in the development of autoimmune diseases such as lupus erythematosus (LE). OBJECTIVES: The aim of the study was to investigate whether apoptosis may contribute to the pathogenesis of cutaneous changes in LE and whether certain apoptosis-related markers such as Fas antigen, Fas ligand (FasL), Bcl-2 and Bax proteins take part in this process. METHODS: Cryostat sections of lesional skin from 31 patients with LE-specific skin lesions [discoid (DLE) n = 17, subacute cutaneous (SCLE) n = 10 and acute cutaneous (ACLE) n = 4] and normal skin samples (n = 10) were stained with monoclonal antibodies against Fas antigen, FasL, Bcl-2 and Bax proteins. For the detection of apoptotic nuclei, terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labelling technique (TUNEL) was employed. RESULTS: The keratinocytes of the epidermis and mononuclear cells of the dermal infiltrate in LE showed a significant increase in Fas antigen expression as compared with normal skin. The anti-Bcl-2 staining intensity within the epidermis was markedly decreased, while the mononuclear infiltrate stained for this protein just like single lymphocytes scattered in normal skin. There was an increase of FasL and/or Bax-positive cells within the dermal infiltrate, but not in the epidermis or in the hair follicles. Accumulation of FasL-bearing cells around hair follicles was a hallmark of DLE. Numerous cells with pyknotic, TUNEL-positive nuclei were demonstrated in the epidermis, in hair follicles and among the cells of the dermal infiltrate. The extent of apoptosis in lesional skin was greater in acute and subacute cutaneous than in discoid forms of LE. CONCLUSIONS: The exact stimulus triggering apoptosis in cutaneous LE still remains unknown but the present study provides strong evidence that the reduction of Bcl-2 expression in the basal cells is associated with the overexpression of Fas antigen and correlates directly with the extent of apoptosis in the epidermis. Furthermore, the current study indicates that the extent of apoptosis in both the epidermis and dermal infiltrate may correlate with the chronological development of LE skin lesions.     

Use of indirect immunofluorescence in the lupus erythematosus/lichen planus overlap syndrome: An additional diagnostic clue

The lupus erythematosus (LE)/lichen planus (LP) overlap syndrome comprises a heterogeneous group of patients who demonstrate clinical, histologic, and immunopathologic characteristics of two diseases. We report six patients with the syndrome who were evaluated by a double-layer indirect immunofluorescence (IF) technic using patient serum and autologous lesional skin as substrate followed by conjugate. This test demonstrated intense staining of the stratum granulosum in two patients, a finding previously shown to be consistent with LP. A third patient developed criteria for the diagnosis of systemic LE with corroborating direct IF findings and a negative indirect IF assay. This preliminary study provides evidence for a possible way of distinguishing LE from LP in some patients with the overlap syndrome.     

Cutaneous Immunofluorescence in Dermatomyositis

ABSTRACT: Seven patients with dermatomyositis who displayed severe skin and muscle disease, and in whom coexistent systemic lupus erythematosus (LE) was excluded, were evaluated by direct immunofluorescent biopsies of skin lesions. Specimens from six showed deposits of immunoglobulins and complement in small to moderate amounts at the dermal-epidermal junction zone. These deposits were usually in the form of focal granular accumulations which lacked continuity and were not well developed as are those seen in LE, or fluorescent subepidermal hyaline bodies. One patient, however, had a more well developed hand of immunoglobulins at the dermal-epidermal junction. All normal skin specimens in these patients were negative by immunofluorescence.
These findings were helpful in clarifying differences in cutaneous immunofluorescence between dermatomyositis and LE, indicating that dermatomyositis specimens can usually be differentiated from those of LE patients. but demonstrating the possibility that confusion might rarely occur in interpreting a lesional immunofluorescent biopsy in dermatomyositis.     

Cytokine and chemokine ligand expression in cutaneous lupus erythematosus

There is increasing evidence that cytokines as well as chemokines are important players in the pathogenesis of lupus erythematosus (LE). We aimed to compare cytokine and chemokine profiles in different types of cutaneous LE. We investigated lesional mRNA and protein expression of various cytokines and chemokines in patients with chronic discoid LE (CDLE, n=15), subacute cutaneous LE (SCLE, n=11), and lupus erythematosus tumidus (LET, n=21). TNF-α, INF-γ, TGF-β, IL-6, IL-10, IL-12p40, CXCL9, and CXCL10 mRNA expression were significantly increased in SCLE when compared to CDLE. Moreover, LET also showed significantly increased mRNA expression of TNF-α, TGF-β, IL-10, IL-12p40 and CXCL9, as compared to CDLE. In all LE subtypes, CXCL9 and CXCL10 mRNA expression significantly correlated with INF-γ mRNA expression, as indicated by r-values ranging from 0.71 - 0.87. Immunohistochemistry for TNF-α, INF-γ, and IL-10 gave support to our RT-PCR results. In conclusion, our results suggest that T helper 1, as well as T helper 2 cytokines are differentially expressed in CDLE, SCLE, and LET. Compared to CDLE, the highest cytokine and chemokine ligand profiles are found in SCLE followed by LET. Our correlation studies also support the importance of an IFN-driven inflammation in cutaneous LE.     

VCAM-1 expression on endothelium in lesions from cutaneous lupus erythematosus is increased compared with systemic and localized scleroderma

Summary The cutaneous lesions in systemic sclerosis (SSc) and lupus erythematosus (LE) are pathologically distinct and may display separate cell adhesion receptors. We have scored lesional skin for the presence of cell adhesion molecules that may influence inflammatory and fibrotic processes in five patients with LE, six patients with diffuse scleroderma and four patients with morphoea. The immunohistological distribution, and the number and intensity of cells staining, were recorded for VCAM-1, ICAM-1, E-selectin, α2 to α6 and β2 integrins and HLA-DR, VCAM-1 staining intensity was increased on endothelium from lesions in LE compared with SSc (P=0.05). Low-level VCAM-1 and E-selectin expression was present on endothelium from uninvolved skin including that from patients with morphoea. HLA-DR expression was increased on infiltrating mononuclear cells (P < 0.05) and keratinocytes in LE (P < 0.05) and the number of fibroblasts staining for ICAM-1 was increased in lesions from patients with SSc, although this did not reach statistical significance. Overall, with respect to endothelial adhesion events, our findings support an important role for VCAM-1 in sustaining chronic inflammation in cutaneous LE.     

Scarring skin lesions of discoid lupus erythematosus are characterized by high numbers of skin-homing cytotoxic lymphocytes associated with strong expression of the type I interferon-induced protein MxA

BACKGROUND: Infiltrating T lymphocytes are considered to play a major pathological role in skin lesions of cutaneous lupus erythematosus (CLE), a cutaneous autoimmune disease of unknown aetiology. Earlier histological studies revealed that the inflammatory infiltrate in CLE skin lesions is predominantly composed of T lymphocytes, with a slight predominance of CD4+ over CD8+ T cells, but failed to explain the development of scarring skin lesions, characteristic for chronic discoid lupus erythematosus (CDLE). Because recent investigations have highlighted the relevance of cytotoxic lymphocytes in autoimmune tissue destruction, we hypothesized that the scarring CDLE lesions might be caused by cytotoxic T lymphocytes. OBJECTIVES: To analyse skin biopsies of 15 patients with CLE [10 female, five male; localized CDLE (lCDLE), n = 5; disseminated CDLE (dCDLE), n = 5, subacute CLE (SCLE), n = 5] and five control biopsies taken from healthy controls and to characterize the inflammatory infiltrate. Methods We used immunohistochemistry, including staining for the cytotoxic molecule granzyme B, the skin-homing molecule cutaneous lymphocyte antigen (CLA) and the protein MxA, which is specifically induced by type I interferons (IFNs). RESULTS: We found a strong coexpression of granzyme B and CLA on lesional lymphocytes of patients with scarring lCDLE and dCDLE, which was significantly enhanced when compared with nonscarring SCLE and healthy controls. The increased expression of granzyme B was closely associated with the lesional expression of the type I IFN-induced protein MxA. CONCLUSIONS: Our results provide evidence that type I IFNs and potentially autoreactive cytotoxic lymphocytes targeting adnexal structures are highly associated with scarring lupus erythematosus lesions and might be responsible for their scarring character.     

Interleukin-6 in the epidermis of patients with psoriasis before and during PUVA treatment

Biopsies from lesional and unaffected skin of 6 patients with psoriasis, taken before and during treatment with psoralen plus UVA (PUVA) were examined immunohistologically, using partially purified polyclonal antibodies to crude supernatants of activated human blood monocytes. By absorption with recombinant derived human monokines, we were able to demonstrate that interleukin-6 (IL-6) (but not IL-1 alpha or IL-1 beta) was located in a laminar and granular pattern in stratum corneum, and on epidermal cell membranes in the viable cellular epidermis. Before PUVA treatment, the intensity and the extension of staining for IL-6 were both markedly increased in lesional skin compared with uninvolved skin. A weaker staining for IL-6 was observed in lesional skin, simultaneous with the clinical improvement of psoriasis. The staining patterns for IL-6 in biopsies from cleared lesional skin and uninvolved psoriatic skin were identical at the conclusion of therapy.     

Plasmacytoid dendritic cells are present in cutaneous dermatomyositis lesions in a pattern distinct from lupus erythematosus

Background:  Plasmacytoid dendritic cells (PDCs) are CD123-positive dendritic cells (DCs) capable of producing interferon-α (IFN). They are thought to play a role in anti-viral immunity and the pathogenesis of lupus erythematosus (LE). Given the histologic similarities between LE and dermatomyositis (DM), we evaluated the presence and distribution of PDCs in lesional skin of both diseases.
Methods:  Twenty-eight biopsies of DM and 27 biopsies of LE were labeled with antibodies to CD123 to identify PDCs. The presence and relative distribution of CD123+ cells was recorded.
Results:  PDCs are present in LE and DM, albeit in greater numbers in LE lesions. Interestingly, PDCs are preferentially located in the epidermis of DM, and are primarily located in the dermis of LE.
Conclusions:  Skin lesions of DM contain PDCs. Because PDCs are thought to participate in the pathogenesis of LE, this finding suggests that they may play a role in DM as well. The preferential epidermal localization of PDCs in DM suggests that their contribution to disease pathogenesis or maintenance in DM may be distinct from LE. Finally, although not pathognomonic, the different patterns of PDC localization in LE and DM may prove a helpful criterion for distinguishing between these two entities histologically.     

Proliferation and effects of UVA irradiation in cultured fibroblasts from lesions in cutaneous lupus erythematosus

Cutaneous lupus erythematosus (LE) often presents clinically as chronic cutaneous lesions, healing with scar formation, and acute cutaneous lesions that are seen in systemic and subacute LE and heal without scarring. UV-light plays a role in the pathogenesis of the skin lesions, but the pathomechanism is still unclear. The aim of this study was to compare fibroblast proliferation and response to UV-light by cultured fibroblasts from scarring and non-scarring LE lesions. Fibroblasts were cultured from skin lesions from 5 patients with classic discoid LE, 5 patients with subacute cutaneous LE and healthy, age-matched donors. Proliferation rate was assessed by cell counts at days 3, 6 and 9. The fibroblast cultures were irradiated with UVA and the supernatants were analysed for IL-6, TGF-beta, IL-4, soluble ICAM-1 and soluble VCAM-1. Fibroblast cultures from scarring lesions showed significantly lower cell-counts at days 3 (P = 0.01) and 9 (P = 0.009), than cultures from nonscarring lesions or controls. There were no significant differences in levels of IL-6 or TGF-beta in supernatants of irradiated fibroblasts from patients compared to controls and IL-4 and the soluble forms of ICAM-1 and VCAM-1 were below detection level. The response to UV-irradiation was similar to that of normal cells in the parameters studied. In summary, cultured fibroblasts from scarring LE lesions displayed significantly decreased proliferation rates compared to non-scarring LE lesions and controls. This may be secondary to inflammatory factors, or due to a functional defect in LE fibroblasts.     

CUTANEOUS LUPUS ERYTHEMATOSUS IN INDIA: IMMUNOFLUORESCENCE PROFILE

The clinical profile and cutaneous lesions of 65 patients with lupus erythematosus (LE) are described. This included 28 discoid LE (13 disseminated, 15 localized), five subacute cutaneous LE, and 32 systemic LE. The need to recognize a pigmented macular form constituting 25% of discoid LE is emphasized. Increased incidence of involvement of the lower lip in discoid LE and pigmentation in systemic LE is noted. Lupus band test was found to be highly sensitive; it was positive for lesional skin of all untreated patients with subacute cutaneous LE and systemic LE, it was, however, not useful on nonlesional skin.     

Type 1 IFN-induced protein MxA and plasmacytoid dendritic cells in lesions of morphea

Morphea is an autoimmune sclerotic skin disease of unknown pathogenesis. As type 1 interferons (IFN) have been implicated in the pathogenesis of systemic sclerosis, we proposed that type 1 IFN promote localized inflammation and fibrosis in morphea. To investigate the expression of the type 1 IFN-inducible protein myxovirus A (MxA) and the presence of plasmacytoid dendritic cells (pDC) in lesions of morphea, lesional skin of 10 patients with morphea was examined by immunohistochemistry for the presence of the type 1 IFN-inducible protein, myxovirus A (MxA), and the pDC markers, CD123 and BDCA-2, and was compared with lesional skin of cutaneous lupus erythematosus, lichen planus and keloid. Lesional and non-lesional morphea skin was compared. MxA was expressed in the epidermis as well as the reticular dermis and subcutis in morphea. pDCs were abundant around vessels and between fibrous bundles. Non-lesional biopsies demonstrated little or no expression of MxA and pDC markers. Keloid showed minimal expression of MxA and pDC markers. We demonstrate the expression of type 1 IFN-related protein MxA and plasmacytoid DCs in lesional but not in non-lesional biopsies of morphea. These findings suggest a potential role for type 1 interferons in the pathogenesis of morphea.     

Interleukin-8 immunoreactivity in the skin of healthy subjects and patients with palmoplantar pustulosis and psoriasis.

Previous studies have shown that neutrophil-activating peptide 1/interleukin-8 (IL-8) is present in psoriatic scales and to a lesser extent in normal human epidermis. A panel of monoclonal antibodies and polyclonal antisera raised against IL-8 was used to localize IL-8 with immunoperoxidase techniques in non-lesional and lesional skin of patients with psoriasis and palmo-plantar pustulosis (PPP), and in corresponding sites from healthy subjects. Intracellular IL-8 immunoreactivity was found in all epidermal cell layers in biopsies of healthy subjects and in non-lesional and lesional skin in both PPP and psoriasis. The most intense immunolabeling was regularly found in the basal cell layer. Intercellular epidermal IL-8 immunolabeling was regularly detected in lesional biopsies in PPP and psoriasis, but not in healthy subjects or non-lesional skin in PPP and psoriasis. No intercellular immunolabeling was detected after successful treatment of lesional skin. The majority of cells along the eccrine sweat glands, dermal mononuclear cell infiltrates, and endothelial cells were IL-8 immunoreactive in all biopsies studied. The present study suggests that IL-8, its precursor form, or, alternatively, a degradation product is present in normal human epidermis.     

Plasmacytoid dendritic cells are absent in skin lesions of polymorphic light eruption

BACKGROUND/PURPOSE: Polymorphic light eruption (PLE) is a common photodermatosis of potential autoimmune origin, and an overlap with lupus erythematosus (LE) has been described. Plasmacytoid dendritic cell (PDC)-induced expression of interferon (IFN)-alpha has been found to be present in LE skin lesions and plays a pivotal role in the pathogenesis of LE by promoting autoimmunity. We therefore asked whether PDCs may also be involved in the pathogenesis of PLE and searched for those cells [which can be identified by their high levels of interleukin (IL)-3 receptor alpha chain (CD123), combined with other cell markers such as CD68] in skin lesions. METHODS: Paraffin-embedded biopsy specimens from a total of 27 patients with clinically and histologically confirmed PLE (nine women, mean age 32.7 years, age range 18-43), LE (seven women, four men, CCLE: n=4, SCLE: n=2, lupus tumidus: n=5, mean age 48.5 years, age range 41-65) or psoriasis (four women, three men, mean age 43.3 years, age range 19-54) (as control group) were analyzed by immunohistochemical CD68/CD123 double staining. Quantification of the immunohistochemical staining was performed by visual cell counting of CD68-/CD123+, CD68+/123-, and CD68+/CD123+ cells separately in the epidermis and dermis of the samples in at least 10 random fields per sample at x 400 microscopic magnification by two of the investigators in a blinded fashion. RESULTS: Microscopic examination of the immunohistochemically stained sections revealed that CD68+/CD123+ cells were present in most specimens obtained from LE [10/11 (91%)] and psoriasis [6/7 (86%)] patients but not at all in those obtained from PLE patients. Quantification and statistical analysis of the dermal infiltrate revealed that CD68+/CD123+ cells were present at a mean+/-SEM field density of 5.6+/-1.3 in LE, 1.6+/-0.6 in psoriasis but totally absent in PLE (P=0.0010 vs. LE, P=0.0135 vs. psoriasis by an unpaired Student's t-test). CONCLUSION: The results confirm the potential significance of PDCs in LE and psoriasis, however the absence of PDCs in PLE contradicts the hypothesis that these cells might play a role in the latter disease.     

Characterization of the changes in matrix molecules at the dermoepidermal junction in lupus erythematosus

Electron microscopy has revealed that the deposition of immunoglobulin in the skin of lupus erythematosus (LE) patients occurs on and below the basal lamina of the basement membrane (BM). The composition of the BM is now to some extent known, and antibodies have been developed against several of its individual components. In this study, we attempt to elucidate the status of some matrix molecules in the dermoepidermal junction in LE. Lesional and nonlesional skin from LE patients was examined using immunofluorescence microscopy with monoclonal and polyclonal antibodies against 6 matrix molecules. Immuno-electron microscopy using monoclonal antibodies was used to discern changes in type IV and type VII collagen. By immuno-fluorescence microscopy, type IV collagen, type VII collagen, and fibronectin were altered in lesional skin. There was a statistically significant correlation between the presence of immunoglobulin and alteration of type IV collagen and type VII collagen in lesional skin. The alterations in type IV and type VII collagens were confirmed on immuno-electron microscopy which showed fragmentation of staining of both antigenic components, particularly type IV collagen.     

Bullous eruption: a manifestation of lupus erythematosus

Bullous lupus erythematosus (BSLE) is a rare subset of systemic lupus erythematosus (SLE), often associated with autoimmunity to type VII collagen. Generally, patients with BSLE meet the criteria for SLE as defined by the American College of Rheumatology. We present a case of a 17-year-old adolescent girl who presented with a vesiculobullous eruption without detectable type VII collagen antibodies and without full criteria for SLE. Differential staining was characteristic for lupus erythematosus (LE), suggesting her eruption is related to LE. We review the spectrum of bullous disease in patients with LE and discuss the pathogenesis and histology of these eruptions, as well as current therapeutic options.     

银屑病皮损中γ干扰素的免疫组化分析

采用γ干扰素（ＩＦＮ－ ）单克隆抗体、双层ＡＰＡＡＰ染色法及显微分光光度测定技术对３０例银屑病患者皮损、１５例正常对照皮肤进行ＩＦＮ－ 抗原的测定。结果显示：除表皮基底层及表皮突下部 １～３层基层外，大量的ＩＦＮ－　抗原阳性染色弥漫分布于皮损全层表皮的角肮细胞间隙；正常对照皮肤的表皮中无明确的ＩＦＮ－　阳性反应；银屑病皮损表皮中ＩＦＮ－　的含量①与银屑病的活动性有关，进行期为２４．３８０８．２５０，静止期为１５．９５５　５．３２７（Ｐ＜０．０１）；②与皮损表皮中Ｔ细胞、ＨＬＡ－ＤＲ细胞的数量呈直线正相关（ｒ１＝０，６９４，ｒ２＝０．４３６，Ｐ＜０．０５）。     

Plasminogen activator inhibitor type-2 in the lesional epidermis of lupus erythematosus

Under certain pathophysiological conditions epidermal keratinocytes produce urokinase-type plasminogen activator (LIPA) or tissue-type PA (tPA). These PAs are subject to regulation by PA inhibitors (PAI). including PAl type-2 (PAI-2). In the normal epidermis. PAI-2 is present in the differentiating suprabasal layers, albeit in the apparent absence of PAs. It has, therefore, been suggested that PAI-2 plays a role in epidermal differentiation not linked to its ability lo inhibit PAs. In line with this hypothesis, we have studied, by immunohistochemistry. the distribution of PAI-2. uPA and tPA in the normal and in the lesional epidermis of patients with lupus erythematosus (LE). a disease in which epidermal differentiation is disturbed. The PAI-2 antigen was detectable in the normal epidermis and in the lesional epidermis of LE. In the normal epidermis, the PAI-2 antigen was most pronounced in the granular layer. In the hyperkeratotic epidermal lesions of LE. the PAI-2 antigen was increased. In normal and lesional skin. PAI-2 was distributed along the cell periphery. Indicating its association with the cornified envelope. Neither uPA nor tPA was detectable in normal or lesional epidermis. Our findings show that PAI-2 is a major type of PAI in normal epidermis and in the lesional epidermis of LE, and that increased epidermal PAI-2 is observed in a disease which is not associated with an increase in epidermal PAs. The data support the hypothesis that epidermal PAI-2 may have other functions than the regulation of PA activity.