Journal of Dermatological Science,
1 (1990) 289-296 289
Elsevier
DESC 00034
Immunohistochemical comparison of actinic reticuloid with allergic contact dermatitis Mayumi Fujita, Yoshiki Miyachi, Takeshi Horio and Sadao Imamura Department of Dermatology, Faculty of Medicine, Kyoto University, Kyoto, Japan
(Received 28 September
1989; accepted
Key words: Actinic reticuloid; Immunohistochemistry;
19 March 1990)
Allergic contact dermatitis
Abstract Biopsy specimens of chronic lesions and ultraviolet-induced lesions from actinic reticuloid patients were examined by immunoperoxidase techniques and compared with those of allergic contact dermatitis skin, one of the delayed-type hypersensitivity conditions. Each lesion of actinic reticuloid showed a clear predominance of suppressor/cytotoxic T cells to helper/inducer T cells and an increase of Langerhans cells in the epidermis and the dermis. These findings are generally similar to those in the late phase (on day 7 and 11) but not in the early phase (on day 2) of allergic contact dermatitis and suggest that delayed-type hypersensitivity might be involved in some parts of the pathogenesis of actinic reticuloid. CD36 + DR + epidermal cells were also observed in ultraviolet-induced lesions from actinic reticuloid patients, suggesting a possible role in the modulation of the mechanism.
Introduction Actinic reticuloid (AR) is a chronic disease characterized by extreme photosensitivity extending through the ultraviolet (UV) to the visible light spectrum [ 11. Although the histologic features of Correspondence to: Mayumi Fujita, Department of Dermatology, Faculty of Medicine, Kyoto University, Shogoin, Sakyo-ku, Kyoto 606, Japan. Abbreviations: ACD = allergic contact dermatitis; AR = reticuloid; actinic DNCB = dinitrochlorobenzene; H&E = hematoxyline and eosin; LCs = Langerhans cells; mAb = monoclonal antibody; MRD = minimal response dose; OCI = optimal cutting temperature; PLP = periodate-lysine-paraformaldehyde; Th/i = helper/inducer T; Ts/c = suppressor/cytotoxic T; UV = ultraviolet.
0923-181 l/90/$03.50
0 1990 Elsevier Science Publishers
the lesions resemble those of lymphomas of the mycosis fungoides type [l] and associations of AR with other malignant neoplasms have rarely been reported [2,3], the clinical course is usually benign. The pathogenic mechanism of AR is obscure. Thomson et al. proposed that a persistent T-cell immune response may play a role in its pathogenesis or maintenance [ 31. Horio reported a case of AR via persistent light reaction from photoallergic contact dermatitis and indicated that photoallergic contact dermatitis and a subsequent persistent light reaction can be involved in the condition of at least some of the patients with AR [4]. To further investigate the possible pathogenesis, especially about relationship with a cellu-
B.V. (Biomedical
Division)
290
lar immune response, we examined, in this study, experimentally induced lesions as well as chronic lesions from AR patients using immunohistochemical methods and compared them with eczematous skin and allergic contact dermatitis (ACD) skin.
addition, in patient 1, an edematous skin lesion was induced on the back by 3 daily exposures with each 7 minimal response dose (MRD) of UV-A, and the induced lesion was biopsied at 24 h after the last irradiation. In patient 2, the induced lesion with 3 day-exposures of 1 MRD of UV-B was also examined at 24 h after the irradiation. The specimens of normal human skin and eczematous skin were obtained from resected breast of 5 patients with breast cancer and skin biopsies of 5 outpatients, respectively. The skin specimens of ACD were also obtained from 3 healthy volunteers, 2 females and 1 male, aged 28 to 57 years. They were sensitized on their forearms with 0.05 ml of 1 y0 DNCB (dinitrochlorobenzene) and fourteen days after the sensitization, they were challenged with 0.05 ml of 0.1% DNCB on their backs by 48 h-occulusive application. The challenged sites were biopsied on 2, 7, and 11 days after the challenge.
Materials and Methods Patients Three men with AR, 62, 72, and 77 years of age, were studied. Diagnosis was based on the clinical, histologic, and photobiologic characteristics. Their clinical data are summarized in Table I. Light testing of all patients showed abnormal photosensitivity to both UV-B and UV-A. Photo-patch tests were carried out with 1% tetrachlorosalicylanilide, tribromosalicylanilide, bithionol, triclocarban, cloflucarban, hexachlorophene, prometazine, levomepromazine, thioridazine, mequitazine, diphenhydramine, and musk ambrette in white petrolatum. Patient 2 showed positive to musk ainbrette and patient 3 showed positive to tribromosalicylanilide and bithionol. Each subject had a helper/ suppressor T-cell ratio of 0.62-0.93 (normal 0.90-1.80) and a normal level of erythrocyte sedimentation rate and serum porphyrins.
Monoclonal antibodies Primary monoclonal antibodies (mAbs) used are listed in Table II. The Lag antibody was established in our laboratory and specifically reacts to Birbeck granules and related structures of human Langerhans cells (LCs) [ 51. Immunoperoxidase staining Biopsied specimens were split into two fragments. The halves were fixed in 10% formalin solution, embedded in partin, and cut into five
Skin specimens Three punch biopsies were obtained from welldeveloped lesional skin of three AR patients. In
TABLE I Clinical details of three patients with AR Patient
Age
Sex
UVB sensitivity
UVA sensitivity
Visible light sensitivity
Photopatch test
Duration of the disease (years)
CD4/CD8 ratio in blood
1 2 3
II 12 62
M M M
+ + +
+ + +
-
-
4 3 17
0.67 0.93 0.62
+ = positive,
- = negative.
’ Musk ambrette. 2 Tribromosalicylanilide
and bithionol.
+’ +2
291 TABLE II A list of monoclonal
antibodies
used in the present study
CD classification
mAb used
Ag distribution
CDla
OKT6”
CD3 CD4
anti-Leu4b anti-Leu3ab
Thymocytes, Langerhans cells Mature T cells Suppressor/inducer T cells, helper/ inducer T cells, activated macrophages Suppressor/cytotoxic T cells All B cells, late Bprogenitor Platelets, monocytes Stem cells, plasma cells, activated lymphocytes MHC class II (HLA-
CD8
OKT8”
CD20
Bl”
CD36 CD38
OKM5” OKTlO”
anti-HLA-DRb Lagd
DR) Birbeck granules and related structures
a Ortho Diagnostic Systems, Raritan, NJ; b Becton Dickinson Immunocytometry Systems, Sunnyvale, CA; ’ Coulter Electronics, Bethesda, MD; d Established in our laboratory [5].
pm sections, which were stained with hematoxylin and eosin (H&E). The other halves were fixed in periodate-lysine-paraformaldehyde (PLP) solution, embedded in optimal cutting temperature (OCT) compound Tissue-Tek II (Miles Laboratories, Naperville, IL), and prepared for cryostatsectioning. Six pm cryosections were stained using the avidin-biotin-peroxidase complex procedures [ 61. In brief, they were treated with primary mAb, biotinylated anti-mouse IgG (4 pg/ml, Vector Laboratories Inc., Burlingame, CA) and avidin-biotin-peroxidase complex (Vector). They were then incubated in a solution of diaminobenzidine (0.2 mg/ml) and hydrogen peroxide (0.1 pg/ml) in PBS, and counterstained with methylgreen. Positive cells were counted by 3 dermatopathologists under direct observation at a magnification of 400 x . Only cells with brown precipitates surrounding the green nuclei were
considered to be positive. Dermal infiltrating cells were expressed as the percentage of positive cells to total infiltrating cells. The number of epidermal LCs or epidermal infiltrating cells was expressed as the number of positive cells per mm of epidermal length. Results Routine light microscopy The skin biopsies of AR lesions showed acanthosis and exocytosis in the epidermis and moderate to extensive round cell infiltration in the mid and superficial dermis. The infiltrating cells were mainly mononuclear cells resembling lymphocytes and histiocytes, which occasionally contained hyperchromatic, enlarged, and irregular nuclei (Fig. 1). Less frequently, eosinophils, plasma cells, and neutrophils were observed. The skin biopsies of UV-induced lesions showed essentially similar histologic findings with more exocytosis and spongiosis. The infiltrating cells in the epidermis and the dermis were mainly mononuclear cells with varing degrees of atypism. Immunohistochemical studies The immunohistochemical findings of AR lesions and UV-induced lesions were compared with those of normal human skin, eczematous skin, and ACD skin of day 2, 7, and 11. Most of the infiltrating cells in the epidermis and the dermis were reactive with anti-pan-T-cell antibody (anti-CD3 antibody) and anti-HLA-DR antibody but negative with anti-CD20 antibody in both AR lesions and UV-induced lesions. CD38’ cells were observed only in a small proportion. In the dermis, the infiltrating cells showed a clear predominance of CD8 + cells to CD4’ cells in both lesions (Fig. 2). On the contrary, dermal infiltrates of eczematous skin and ACD skin showed a predominance of CD4 + cells to CD8 + cells, although, in ACD, the ratio of CD4’ cells: CD8 + cells gradually decreased with days. The number of CD4 + cells and that of CD8+ cells in ACD became almost equal on day 11.
292
Fig. 1. Biopsy specimen of AR lesions (patient 3). Lymphoid cells with large, hyperchromatic, and irregular nuclei were observed in the epidermis and the dermis, admixed with eosinophils, plasma cells, and neutrophils. H&E stain, x 400.
In the epidermis, infiltrating cells also showed a low ratio of CD4 + cells : CD8 + cells in both AR lesions and UV-induced lesions compared with those in eczematous skin and ACD skin (Table III). Numerically, however, more infiltrating cells were observed in UV-induced lesions than in AR lesions. In ACD skin, CD8’ cells increased in number with days. In addition to T-lymphocytes, antigen-presenting cells were examined with anti-CD 1a antibody, Lag antibody, anti-HLA-DR antibody, and antiCD36 antibody. In the epidermis, CD la + cells in both AR lesions and UV-induced lesions were increased in number compared with those in normal human skin (Fig. 3). There was statistical difference between AR lesions and normal human skin P = ~0.05. In AR lesions and UV-induced lesions, only half of CD la + cells reacted with Lag antibody, although most of CDla’ cells were Lag+ in normal human skin. In ACD, the numbers of CDla’ cells decreased in the early phase (on day 2) and then increased in the late phase (on day 7 and 11). The numbers of HLA-DR + cells were also increased in UV-induced lesions as compared with normal human skin (P=
1 (1990) 289-296 289
Elsevier
DESC 00034
Immunohistochemical comparison of actinic reticuloid with allergic contact dermatitis Mayumi Fujita, Yoshiki Miyachi, Takeshi Horio and Sadao Imamura Department of Dermatology, Faculty of Medicine, Kyoto University, Kyoto, Japan
(Received 28 September
1989; accepted
Key words: Actinic reticuloid; Immunohistochemistry;
19 March 1990)
Allergic contact dermatitis
Abstract Biopsy specimens of chronic lesions and ultraviolet-induced lesions from actinic reticuloid patients were examined by immunoperoxidase techniques and compared with those of allergic contact dermatitis skin, one of the delayed-type hypersensitivity conditions. Each lesion of actinic reticuloid showed a clear predominance of suppressor/cytotoxic T cells to helper/inducer T cells and an increase of Langerhans cells in the epidermis and the dermis. These findings are generally similar to those in the late phase (on day 7 and 11) but not in the early phase (on day 2) of allergic contact dermatitis and suggest that delayed-type hypersensitivity might be involved in some parts of the pathogenesis of actinic reticuloid. CD36 + DR + epidermal cells were also observed in ultraviolet-induced lesions from actinic reticuloid patients, suggesting a possible role in the modulation of the mechanism.
Introduction Actinic reticuloid (AR) is a chronic disease characterized by extreme photosensitivity extending through the ultraviolet (UV) to the visible light spectrum [ 11. Although the histologic features of Correspondence to: Mayumi Fujita, Department of Dermatology, Faculty of Medicine, Kyoto University, Shogoin, Sakyo-ku, Kyoto 606, Japan. Abbreviations: ACD = allergic contact dermatitis; AR = reticuloid; actinic DNCB = dinitrochlorobenzene; H&E = hematoxyline and eosin; LCs = Langerhans cells; mAb = monoclonal antibody; MRD = minimal response dose; OCI = optimal cutting temperature; PLP = periodate-lysine-paraformaldehyde; Th/i = helper/inducer T; Ts/c = suppressor/cytotoxic T; UV = ultraviolet.
0923-181 l/90/$03.50
0 1990 Elsevier Science Publishers
the lesions resemble those of lymphomas of the mycosis fungoides type [l] and associations of AR with other malignant neoplasms have rarely been reported [2,3], the clinical course is usually benign. The pathogenic mechanism of AR is obscure. Thomson et al. proposed that a persistent T-cell immune response may play a role in its pathogenesis or maintenance [ 31. Horio reported a case of AR via persistent light reaction from photoallergic contact dermatitis and indicated that photoallergic contact dermatitis and a subsequent persistent light reaction can be involved in the condition of at least some of the patients with AR [4]. To further investigate the possible pathogenesis, especially about relationship with a cellu-
B.V. (Biomedical
Division)
290
lar immune response, we examined, in this study, experimentally induced lesions as well as chronic lesions from AR patients using immunohistochemical methods and compared them with eczematous skin and allergic contact dermatitis (ACD) skin.
addition, in patient 1, an edematous skin lesion was induced on the back by 3 daily exposures with each 7 minimal response dose (MRD) of UV-A, and the induced lesion was biopsied at 24 h after the last irradiation. In patient 2, the induced lesion with 3 day-exposures of 1 MRD of UV-B was also examined at 24 h after the irradiation. The specimens of normal human skin and eczematous skin were obtained from resected breast of 5 patients with breast cancer and skin biopsies of 5 outpatients, respectively. The skin specimens of ACD were also obtained from 3 healthy volunteers, 2 females and 1 male, aged 28 to 57 years. They were sensitized on their forearms with 0.05 ml of 1 y0 DNCB (dinitrochlorobenzene) and fourteen days after the sensitization, they were challenged with 0.05 ml of 0.1% DNCB on their backs by 48 h-occulusive application. The challenged sites were biopsied on 2, 7, and 11 days after the challenge.
Materials and Methods Patients Three men with AR, 62, 72, and 77 years of age, were studied. Diagnosis was based on the clinical, histologic, and photobiologic characteristics. Their clinical data are summarized in Table I. Light testing of all patients showed abnormal photosensitivity to both UV-B and UV-A. Photo-patch tests were carried out with 1% tetrachlorosalicylanilide, tribromosalicylanilide, bithionol, triclocarban, cloflucarban, hexachlorophene, prometazine, levomepromazine, thioridazine, mequitazine, diphenhydramine, and musk ambrette in white petrolatum. Patient 2 showed positive to musk ainbrette and patient 3 showed positive to tribromosalicylanilide and bithionol. Each subject had a helper/ suppressor T-cell ratio of 0.62-0.93 (normal 0.90-1.80) and a normal level of erythrocyte sedimentation rate and serum porphyrins.
Monoclonal antibodies Primary monoclonal antibodies (mAbs) used are listed in Table II. The Lag antibody was established in our laboratory and specifically reacts to Birbeck granules and related structures of human Langerhans cells (LCs) [ 51. Immunoperoxidase staining Biopsied specimens were split into two fragments. The halves were fixed in 10% formalin solution, embedded in partin, and cut into five
Skin specimens Three punch biopsies were obtained from welldeveloped lesional skin of three AR patients. In
TABLE I Clinical details of three patients with AR Patient
Age
Sex
UVB sensitivity
UVA sensitivity
Visible light sensitivity
Photopatch test
Duration of the disease (years)
CD4/CD8 ratio in blood
1 2 3
II 12 62
M M M
+ + +
+ + +
-
-
4 3 17
0.67 0.93 0.62
+ = positive,
- = negative.
’ Musk ambrette. 2 Tribromosalicylanilide
and bithionol.
+’ +2
291 TABLE II A list of monoclonal
antibodies
used in the present study
CD classification
mAb used
Ag distribution
CDla
OKT6”
CD3 CD4
anti-Leu4b anti-Leu3ab
Thymocytes, Langerhans cells Mature T cells Suppressor/inducer T cells, helper/ inducer T cells, activated macrophages Suppressor/cytotoxic T cells All B cells, late Bprogenitor Platelets, monocytes Stem cells, plasma cells, activated lymphocytes MHC class II (HLA-
CD8
OKT8”
CD20
Bl”
CD36 CD38
OKM5” OKTlO”
anti-HLA-DRb Lagd
DR) Birbeck granules and related structures
a Ortho Diagnostic Systems, Raritan, NJ; b Becton Dickinson Immunocytometry Systems, Sunnyvale, CA; ’ Coulter Electronics, Bethesda, MD; d Established in our laboratory [5].
pm sections, which were stained with hematoxylin and eosin (H&E). The other halves were fixed in periodate-lysine-paraformaldehyde (PLP) solution, embedded in optimal cutting temperature (OCT) compound Tissue-Tek II (Miles Laboratories, Naperville, IL), and prepared for cryostatsectioning. Six pm cryosections were stained using the avidin-biotin-peroxidase complex procedures [ 61. In brief, they were treated with primary mAb, biotinylated anti-mouse IgG (4 pg/ml, Vector Laboratories Inc., Burlingame, CA) and avidin-biotin-peroxidase complex (Vector). They were then incubated in a solution of diaminobenzidine (0.2 mg/ml) and hydrogen peroxide (0.1 pg/ml) in PBS, and counterstained with methylgreen. Positive cells were counted by 3 dermatopathologists under direct observation at a magnification of 400 x . Only cells with brown precipitates surrounding the green nuclei were
considered to be positive. Dermal infiltrating cells were expressed as the percentage of positive cells to total infiltrating cells. The number of epidermal LCs or epidermal infiltrating cells was expressed as the number of positive cells per mm of epidermal length. Results Routine light microscopy The skin biopsies of AR lesions showed acanthosis and exocytosis in the epidermis and moderate to extensive round cell infiltration in the mid and superficial dermis. The infiltrating cells were mainly mononuclear cells resembling lymphocytes and histiocytes, which occasionally contained hyperchromatic, enlarged, and irregular nuclei (Fig. 1). Less frequently, eosinophils, plasma cells, and neutrophils were observed. The skin biopsies of UV-induced lesions showed essentially similar histologic findings with more exocytosis and spongiosis. The infiltrating cells in the epidermis and the dermis were mainly mononuclear cells with varing degrees of atypism. Immunohistochemical studies The immunohistochemical findings of AR lesions and UV-induced lesions were compared with those of normal human skin, eczematous skin, and ACD skin of day 2, 7, and 11. Most of the infiltrating cells in the epidermis and the dermis were reactive with anti-pan-T-cell antibody (anti-CD3 antibody) and anti-HLA-DR antibody but negative with anti-CD20 antibody in both AR lesions and UV-induced lesions. CD38’ cells were observed only in a small proportion. In the dermis, the infiltrating cells showed a clear predominance of CD8 + cells to CD4’ cells in both lesions (Fig. 2). On the contrary, dermal infiltrates of eczematous skin and ACD skin showed a predominance of CD4 + cells to CD8 + cells, although, in ACD, the ratio of CD4’ cells: CD8 + cells gradually decreased with days. The number of CD4 + cells and that of CD8+ cells in ACD became almost equal on day 11.
292
Fig. 1. Biopsy specimen of AR lesions (patient 3). Lymphoid cells with large, hyperchromatic, and irregular nuclei were observed in the epidermis and the dermis, admixed with eosinophils, plasma cells, and neutrophils. H&E stain, x 400.
In the epidermis, infiltrating cells also showed a low ratio of CD4 + cells : CD8 + cells in both AR lesions and UV-induced lesions compared with those in eczematous skin and ACD skin (Table III). Numerically, however, more infiltrating cells were observed in UV-induced lesions than in AR lesions. In ACD skin, CD8’ cells increased in number with days. In addition to T-lymphocytes, antigen-presenting cells were examined with anti-CD 1a antibody, Lag antibody, anti-HLA-DR antibody, and antiCD36 antibody. In the epidermis, CD la + cells in both AR lesions and UV-induced lesions were increased in number compared with those in normal human skin (Fig. 3). There was statistical difference between AR lesions and normal human skin P = ~0.05. In AR lesions and UV-induced lesions, only half of CD la + cells reacted with Lag antibody, although most of CDla’ cells were Lag+ in normal human skin. In ACD, the numbers of CDla’ cells decreased in the early phase (on day 2) and then increased in the late phase (on day 7 and 11). The numbers of HLA-DR + cells were also increased in UV-induced lesions as compared with normal human skin (P=
