ctimcat and Experimentat Dermatotogy 1992; 17: 413-420.
The inter-relation between inflammation and epidermal proliferation in normal skin following epicutaneous application of leukotriene-B4—an immunohistochemical study E.M.G.J.DE JONG, P.E.J.VAN ERP, LM.J.J.VAN VLIJMEN AND P.C.M.VAN DE KERKHOF Department of Dermatology, University Hospital., Nijmegen, The Netherlands Accepted for publication 30 March 1992
Summary Topieai application of ieukotriene-B4 (LTB4) on normai skin has been used as an in-vivo modei to investigate cutaneous inflammation and epidermal proiiferation, which are important phenomena in the pathogenesis of psoriasis. The aim of the present investigation is to further eiueidate the interrelation between infiammation and epidermai proiiferation, using specific monoeionai antibodies as markers for the different cell types invoived. Aiiquots of LTB4 were appiied on the upperarms of eight heaithy voiunteers. After LTB4-application, biopsies were taken at consecutive time intervais. On frozen seetions, epidermai proiiferation was assessed by K,8.12(keratin i6) and Ki-67-binding (cycling eeiis), inflammation was charaeterized using anti-eiastase (PMN), T i l (T-iymphoeytes), pan-B (B-iymphoeytes), WT \A (CDi4-positive ceiis) and OKT 6 (Langerhans eeiis). New observations were that the density of CDi4positive ceiis was increased even at 8 h and decreased siightiy at 72 h. A striking rearrangement of Langerhans cells was seen in ciose vieinity to intra-epidermai accumuiations of PMN. Remarkabiy an inereased density of these ceiis in the dermis at 72 h was seen and a decrease in the epidermis. In iine with previous studies, the accumulation of PMN reached a maximum 24 h after LTB4chaiienge. The identity of the mononueiear infiitrate ceiis which have been reported 48-72 h after LTB4 proved to be T-iymphocytes. No B-lymphoeytes were observed. Ki-67-positive nuciei were maximaiiy increased 72 h after LTB4-appiication, which implies that recruitment of cycling ceiis is of reievance for the LTB4-induced proiiferation in vivo. The hyperproiiferation-related keratin i6 was expressed inconsistentiy in the suprabasai compartment. The present investigation demonstrates that the CD 14 ceiis and Langerhans ceils participate in acute LTB4 induced inflammation. Correspondence: E.M.G.J.de Jong, Department of Dermatology, University Hospital Nijmegen, Postbox GlOl, 6500IIB Nijmegen, The Netherlands.
Topicai appiication of ieukotriene B4 (LTB4), a 5lipoxygenase product from arachidonie acid, on normai skin produces an acute inflammation, dominated by PMN,' foiiowed by a perivascuiar mononueiear ceii infiitrate after 48 h^ and epidermal proiiferation,' reaching a maximum after 72-96 h. A dose-response reiationship has been shown between these phenomena and the concentration of LTB4 appiied to the skin. Therefore this approach has been used as an in-vivo modei to investigate the infiuenee of therapies sueh as corticosteroids, dithranol, photochemotherapy, methotrexate and retinoids on transeutaneous migration of PMN.^ Increased amounts of LTB4 have been found in iesionai skin of atopic dermatitis, aiiergic contact dermatitis, urticaria, and psoriasis.'"*^ In psoriasis, a skin disorder characterized by epidermai proiiferation and inflammation, increased production of 12-HETE, i5-HETE, piateiet activating factor, interleukin 8, and LTB4 has been demonstrated in iesionai skin." '^ In view of the relatively high biologic activity of LTB4 compared to other ieukotrienes, this compound seems of particuiar reievanee in eontrol of inflammation The aim of the present investigation is to further eiueidate the interreiation between infiammation and epidermal proiiferation after epicutaneous appiication of LTB4, using monoeionai antibodies as markers for different ceii types involved on frozen sections. Materials and methods Subjects
A group of eight heaithy subjects participated in the study. They had no history or signs of skin disease. The group consisted of two femaies and six maies. Their age was between 21 and 28 years. Prior to the experiment, approval of the Ethical Committee was obtained. LTB^-application and biopsy procedures
Aiiquots of 100 ng LTB4 (Paesei GmbH, Frankfurt, FRG) diiuted in 10 /ii of ethanoi were applied on the 413
414
E.M.G.J.DE JONG et al.
outer surface of the upperarms of the voiunteers via a glass eylinder (55 mm diameter) and the ethanoi was evaporated under a stream of nitrogen.'^ The test sites were covered with impermeable dressings (Silver patch, van der Bend BV, Brieile, The Netheriands) heid at piace with ieukosiik tape (Beiersdorf, Hamburg, FRG). Before LTB4-appiication and at time intervals of 8, 24, 48 and 72 h biopsies were taken using a razor biade in conjunction with a metai guard. Chloraethyi spray was used as a iocai anaesthetic (Aethyichiorid, Adroka, Switzeriand). The biopsies were snap frozen in iiquid nitrogen, embedded in Tissue Tek OCT compound {Miies Scientific, Napervilie, USA) and stored at — SCC. Staining procedures To assess epidermai proiiferation, staining with the foiiowing monoeionai antibodies was performed: (i) K^8.12 (Sigma, St Louis, USA) to assess keratin 16, which is expressed in hyperproliferative epider14.15 mis, (ii) Ki-67 (Dakopatts, Copenhagen, Danmark) to detect a nuciear antigen present in activeiy cycling nuciei.
200 mg/i 3 amino-9-ethyi-carbazoie soiution and 0-01% H2O2 prepared freshly (AEC-solution) for 15 min at room temperature in the dark. Staining with T H , Pan-B, WT14 and 0KT6 were done using a peroxidase-anti-peroxidase technique (PAP-technique). For this technique, slides were incubated with the monoeionai antibodies for 60 min. After two washes with PBS, an incubation step foiiowed with rabbit-anti-mouse immunoglobuiins (RAM-Ig, Dakopatts, Copenhagen, Danmark) and after two more washes an incubation with peroxidase-anti-peroxidase compiexes (PAP, Dakopatts, Copenhagen, Danmark) was performed. The incubation with RAM-Ig and PAP was repeated and the siides stained with AEC-soiution after two washes with PBS and pre-incubation with sodium acetate buffer pH 4 9. Aii siides werefinaiiywashed with demineraiized water and counterstained with Mayer's haematoxylin (Sigma, St. Louis, USA). Slides were mounted in giycerine gelatin and studied by light microscopy. Histological examinations
The density of PMN, T- and B-iymphocytes, CD14positive ceiis and Langerhans ceiis were assessed biind by two investigators in dermis and epidermis using a fiveThe inflammation was characterized with the foiiow- point scaie: 0 — no positive ceiis observed; 1 == only sporadic staining of cells; 2 = minimal presence; ing antibodies: 3 —moderate presence; 4—pronounced presence. The (iii) Anti-eiastase (Serotec, Oxford, Engiand) a marker dermal infiitrate was subdivided in perivascuiar and for polymorphonuciear ieucocytes,"' diffuse iocaiization. (iv) TI 1 (Dakopatts, Copenhagen, Danmark) to assess The number of actively cyciing ceiis was assessed by T-iymphocytes (CD2), counting Ki-67-positive nuciei per mm iength of section. (v) Pan-B (Dakopatts, Copenhagen, Danmark) to Ks8.12-binding of the suprabasai and basal eompartassess B-iymphocytes (CD22), ments was assessed using a five-point scale: 0 = no (vi) WT14 (Department of Medicine, Division of staining; 1 — sporadic staining; 2 — minimal staining; Nephrology, University Hospital Nijmegen)" to 3 = moderate staining; 4 = pronounced staining. demonstrate CD14-positive ceiis whieh represent activated monocytes and macrophages, (vii) OKT6 (Ortho Diagnostic Systems, Raritan, USA) Statistical analysis to assess Langerhans eeiis (CDl). The Mann-Whitney test was used for statistical anaiysis. Skin sections of 7 /im were cut. The slides were fixed for 10 min in acetone/ether 60/40 vol% for Ki-67, or in acetone for staining with the other antibodies. Results Staining with Ks8.12, Ki-67, and anti-eiastase were performed using a peroxidase-technique. In brief, slides Clinical response were incubated with the monoclonai antibodies for 30 min, and after three washes with phosphate-buffered Apart from a small blister in one of the voiunteers at 72 h saline (PBS) incubated with rabbit-anti-mouse antibody after LTB4 appiication, no signs of inflammation were (RAMPO, Dakopatts, Copenhagen, Danmark) conju- seen. gated with peroxidase for 30 min except for the antieiastase antibody, whieh is directly conjugated with Inflammation peroxidase. After three more washes with PBS and preincubation with sodium acetate buffer, pH 4 9, slides Density and localization of inflammatory cells following were finally stained in sodium acetate buffer containing LTB4-appiication, are summarized in Figure la-d.
IMMUNOHISTOCHEMICAL STUDY OF LTB.
415
Anti-eiastase dermis perivascular
0
8 16 24 36 40 48 56 64 72 Hours after LTB4-application
0
8 16 24 32 40 48 56 64 72 Hours atter LTB4-application
PMN
PMN
WT14 epidermis
tb)
0
8 16 24 32 40 48 56 64 72 Hours otter LTB4-application
WT14
0
8 16 24 36 40 48 56 64 72 80 Hours after LT84-applicotiori
WT14 dermis perivasculaf
8 16 24 32 40 48 56 64 72 Hours after LTB4-applicafiOn
0
-CD14-positiue cells
0KT6 epidermis
0KT6 dermis penvascular
8 16 24 32 40 48 56 64 72 80 Hours after LTB4-application
• Langerhans cells
0
Langerhans cells
Til
8 16 24 32 40 48 56 64 72 60 Hours after LTB4-application T-lymphocytes
0
8 16 24 32 40 46 56 64 72 80 Hours after LTB4-application T-lymphocytes
8 16 24 32 40 48 56 64 72 80 Hours after LTB4-application -Longerhans cells
Tn
dermis diffuse
0
8 !6 24 32 40 48 56 64 72 80 Hours after LTB4-application -CD14-positive cells
0KT6 dermjs diffuse
0
8 16 24 32 40 48 56 64 72 Hours after LTB4-application • PMN
dermis diffuse
-CD14-positive cells
0
0
dermis periuascular
"""
6
8 16 24 32 40 48" 56 64 72 80 Hours atter LTB4-application • T-lymphocytes
fuf ;^r^.; ;f Presentation of the density of infiltrate cdls at time intervals following LTB4-appHcation, n = 8 (mean + SEM)- (a (b) CD14-positive cells; (c) Langerhans cells; (d) T-iymphocytes. ' \ _ /• ta
PMN (Fig. la). A reproducible accumulation of PMN m dermis and epidermis was found, with a maximum at 24 h after appiieation of LTB4. In the epidermis, the accumulation was already present and distributed over basal and suprabasai compartments at 8 h. At 24 h,
maximum accumulation, the microabscesses were predominantly iocaiized in the subeorneai 2one (Fig. 2). After 48 and 72 h the mieroabscesses became progressively flattened and faded away in all patients. The dermal PMN-aeeumuiation was in phase with the epidermai
416
E.M.G.J.DE JONG et al. positive cells were only sporadically present at all time intervals. In contrast, dermal CD14-positive cells already were increased diffusely (P=0-05) at 8 h after LTB4application (Fig. 3a-b). Their numbers remained increased until 48 h, and diminished at 72 h. Langerhans cells (Fig. lc). In the epidermis the CDlpositive Langerhans cells were present at all time
Figure 2. Staining with anti-eiastase, showing maximal accumulation of PMN 24 h after LTB4-application ( x 408).
infiltration, however, at 8 h relatively more cells were seen in the dermal compartment. At 72 h, the PMNextravasation had considerably diminished but was still present. CDM-positive cells {Fig. lb). In the epidermis, CD14-
s*^:'.
;..
^
Figure 3. Staining with WT14, showing CD14-positive-celIs before (a) and 8 h (b) after application of LTB4 ( x 204).
Figure 4. Langerhans cells before application of LTB4 (a); alter 24 h jn the vicinity of micro-abscesses (b); and after 72 h decreased density in epidermis, increased density in dermal compartment (c) ( x 204).
IMMUNOHISTOCHEMICAL STUDY OF LTB4 intervais. The density of these ceils in the epidermai compartment showed a statisticaiiy significant decrease at 72 h {P
The inter-relation between inflammation and epidermal proliferation in normal skin following epicutaneous application of leukotriene-B4—an immunohistochemical study E.M.G.J.DE JONG, P.E.J.VAN ERP, LM.J.J.VAN VLIJMEN AND P.C.M.VAN DE KERKHOF Department of Dermatology, University Hospital., Nijmegen, The Netherlands Accepted for publication 30 March 1992
Summary Topieai application of ieukotriene-B4 (LTB4) on normai skin has been used as an in-vivo modei to investigate cutaneous inflammation and epidermal proiiferation, which are important phenomena in the pathogenesis of psoriasis. The aim of the present investigation is to further eiueidate the interrelation between infiammation and epidermai proiiferation, using specific monoeionai antibodies as markers for the different cell types invoived. Aiiquots of LTB4 were appiied on the upperarms of eight heaithy voiunteers. After LTB4-application, biopsies were taken at consecutive time intervais. On frozen seetions, epidermai proiiferation was assessed by K,8.12(keratin i6) and Ki-67-binding (cycling eeiis), inflammation was charaeterized using anti-eiastase (PMN), T i l (T-iymphoeytes), pan-B (B-iymphoeytes), WT \A (CDi4-positive ceiis) and OKT 6 (Langerhans eeiis). New observations were that the density of CDi4positive ceiis was increased even at 8 h and decreased siightiy at 72 h. A striking rearrangement of Langerhans cells was seen in ciose vieinity to intra-epidermai accumuiations of PMN. Remarkabiy an inereased density of these ceiis in the dermis at 72 h was seen and a decrease in the epidermis. In iine with previous studies, the accumulation of PMN reached a maximum 24 h after LTB4chaiienge. The identity of the mononueiear infiitrate ceiis which have been reported 48-72 h after LTB4 proved to be T-iymphocytes. No B-lymphoeytes were observed. Ki-67-positive nuciei were maximaiiy increased 72 h after LTB4-appiication, which implies that recruitment of cycling ceiis is of reievance for the LTB4-induced proiiferation in vivo. The hyperproiiferation-related keratin i6 was expressed inconsistentiy in the suprabasai compartment. The present investigation demonstrates that the CD 14 ceiis and Langerhans ceils participate in acute LTB4 induced inflammation. Correspondence: E.M.G.J.de Jong, Department of Dermatology, University Hospital Nijmegen, Postbox GlOl, 6500IIB Nijmegen, The Netherlands.
Topicai appiication of ieukotriene B4 (LTB4), a 5lipoxygenase product from arachidonie acid, on normai skin produces an acute inflammation, dominated by PMN,' foiiowed by a perivascuiar mononueiear ceii infiitrate after 48 h^ and epidermal proiiferation,' reaching a maximum after 72-96 h. A dose-response reiationship has been shown between these phenomena and the concentration of LTB4 appiied to the skin. Therefore this approach has been used as an in-vivo modei to investigate the infiuenee of therapies sueh as corticosteroids, dithranol, photochemotherapy, methotrexate and retinoids on transeutaneous migration of PMN.^ Increased amounts of LTB4 have been found in iesionai skin of atopic dermatitis, aiiergic contact dermatitis, urticaria, and psoriasis.'"*^ In psoriasis, a skin disorder characterized by epidermai proiiferation and inflammation, increased production of 12-HETE, i5-HETE, piateiet activating factor, interleukin 8, and LTB4 has been demonstrated in iesionai skin." '^ In view of the relatively high biologic activity of LTB4 compared to other ieukotrienes, this compound seems of particuiar reievanee in eontrol of inflammation The aim of the present investigation is to further eiueidate the interreiation between infiammation and epidermal proiiferation after epicutaneous appiication of LTB4, using monoeionai antibodies as markers for different ceii types involved on frozen sections. Materials and methods Subjects
A group of eight heaithy subjects participated in the study. They had no history or signs of skin disease. The group consisted of two femaies and six maies. Their age was between 21 and 28 years. Prior to the experiment, approval of the Ethical Committee was obtained. LTB^-application and biopsy procedures
Aiiquots of 100 ng LTB4 (Paesei GmbH, Frankfurt, FRG) diiuted in 10 /ii of ethanoi were applied on the 413
414
E.M.G.J.DE JONG et al.
outer surface of the upperarms of the voiunteers via a glass eylinder (55 mm diameter) and the ethanoi was evaporated under a stream of nitrogen.'^ The test sites were covered with impermeable dressings (Silver patch, van der Bend BV, Brieile, The Netheriands) heid at piace with ieukosiik tape (Beiersdorf, Hamburg, FRG). Before LTB4-appiication and at time intervals of 8, 24, 48 and 72 h biopsies were taken using a razor biade in conjunction with a metai guard. Chloraethyi spray was used as a iocai anaesthetic (Aethyichiorid, Adroka, Switzeriand). The biopsies were snap frozen in iiquid nitrogen, embedded in Tissue Tek OCT compound {Miies Scientific, Napervilie, USA) and stored at — SCC. Staining procedures To assess epidermai proiiferation, staining with the foiiowing monoeionai antibodies was performed: (i) K^8.12 (Sigma, St Louis, USA) to assess keratin 16, which is expressed in hyperproliferative epider14.15 mis, (ii) Ki-67 (Dakopatts, Copenhagen, Danmark) to detect a nuciear antigen present in activeiy cycling nuciei.
200 mg/i 3 amino-9-ethyi-carbazoie soiution and 0-01% H2O2 prepared freshly (AEC-solution) for 15 min at room temperature in the dark. Staining with T H , Pan-B, WT14 and 0KT6 were done using a peroxidase-anti-peroxidase technique (PAP-technique). For this technique, slides were incubated with the monoeionai antibodies for 60 min. After two washes with PBS, an incubation step foiiowed with rabbit-anti-mouse immunoglobuiins (RAM-Ig, Dakopatts, Copenhagen, Danmark) and after two more washes an incubation with peroxidase-anti-peroxidase compiexes (PAP, Dakopatts, Copenhagen, Danmark) was performed. The incubation with RAM-Ig and PAP was repeated and the siides stained with AEC-soiution after two washes with PBS and pre-incubation with sodium acetate buffer pH 4 9. Aii siides werefinaiiywashed with demineraiized water and counterstained with Mayer's haematoxylin (Sigma, St. Louis, USA). Slides were mounted in giycerine gelatin and studied by light microscopy. Histological examinations
The density of PMN, T- and B-iymphocytes, CD14positive ceiis and Langerhans ceiis were assessed biind by two investigators in dermis and epidermis using a fiveThe inflammation was characterized with the foiiow- point scaie: 0 — no positive ceiis observed; 1 == only sporadic staining of cells; 2 = minimal presence; ing antibodies: 3 —moderate presence; 4—pronounced presence. The (iii) Anti-eiastase (Serotec, Oxford, Engiand) a marker dermal infiitrate was subdivided in perivascuiar and for polymorphonuciear ieucocytes,"' diffuse iocaiization. (iv) TI 1 (Dakopatts, Copenhagen, Danmark) to assess The number of actively cyciing ceiis was assessed by T-iymphocytes (CD2), counting Ki-67-positive nuciei per mm iength of section. (v) Pan-B (Dakopatts, Copenhagen, Danmark) to Ks8.12-binding of the suprabasai and basal eompartassess B-iymphocytes (CD22), ments was assessed using a five-point scale: 0 = no (vi) WT14 (Department of Medicine, Division of staining; 1 — sporadic staining; 2 — minimal staining; Nephrology, University Hospital Nijmegen)" to 3 = moderate staining; 4 = pronounced staining. demonstrate CD14-positive ceiis whieh represent activated monocytes and macrophages, (vii) OKT6 (Ortho Diagnostic Systems, Raritan, USA) Statistical analysis to assess Langerhans eeiis (CDl). The Mann-Whitney test was used for statistical anaiysis. Skin sections of 7 /im were cut. The slides were fixed for 10 min in acetone/ether 60/40 vol% for Ki-67, or in acetone for staining with the other antibodies. Results Staining with Ks8.12, Ki-67, and anti-eiastase were performed using a peroxidase-technique. In brief, slides Clinical response were incubated with the monoclonai antibodies for 30 min, and after three washes with phosphate-buffered Apart from a small blister in one of the voiunteers at 72 h saline (PBS) incubated with rabbit-anti-mouse antibody after LTB4 appiication, no signs of inflammation were (RAMPO, Dakopatts, Copenhagen, Danmark) conju- seen. gated with peroxidase for 30 min except for the antieiastase antibody, whieh is directly conjugated with Inflammation peroxidase. After three more washes with PBS and preincubation with sodium acetate buffer, pH 4 9, slides Density and localization of inflammatory cells following were finally stained in sodium acetate buffer containing LTB4-appiication, are summarized in Figure la-d.
IMMUNOHISTOCHEMICAL STUDY OF LTB.
415
Anti-eiastase dermis perivascular
0
8 16 24 36 40 48 56 64 72 Hours after LTB4-application
0
8 16 24 32 40 48 56 64 72 Hours atter LTB4-application
PMN
PMN
WT14 epidermis
tb)
0
8 16 24 32 40 48 56 64 72 Hours otter LTB4-application
WT14
0
8 16 24 36 40 48 56 64 72 80 Hours after LT84-applicotiori
WT14 dermis perivasculaf
8 16 24 32 40 48 56 64 72 Hours after LTB4-applicafiOn
0
-CD14-positiue cells
0KT6 epidermis
0KT6 dermis penvascular
8 16 24 32 40 48 56 64 72 80 Hours after LTB4-application
• Langerhans cells
0
Langerhans cells
Til
8 16 24 32 40 48 56 64 72 60 Hours after LTB4-application T-lymphocytes
0
8 16 24 32 40 46 56 64 72 80 Hours after LTB4-application T-lymphocytes
8 16 24 32 40 48 56 64 72 80 Hours after LTB4-application -Longerhans cells
Tn
dermis diffuse
0
8 !6 24 32 40 48 56 64 72 80 Hours after LTB4-application -CD14-positive cells
0KT6 dermjs diffuse
0
8 16 24 32 40 48 56 64 72 Hours after LTB4-application • PMN
dermis diffuse
-CD14-positive cells
0
0
dermis periuascular
"""
6
8 16 24 32 40 48" 56 64 72 80 Hours atter LTB4-application • T-lymphocytes
fuf ;^r^.; ;f Presentation of the density of infiltrate cdls at time intervals following LTB4-appHcation, n = 8 (mean + SEM)- (a (b) CD14-positive cells; (c) Langerhans cells; (d) T-iymphocytes. ' \ _ /• ta
PMN (Fig. la). A reproducible accumulation of PMN m dermis and epidermis was found, with a maximum at 24 h after appiieation of LTB4. In the epidermis, the accumulation was already present and distributed over basal and suprabasai compartments at 8 h. At 24 h,
maximum accumulation, the microabscesses were predominantly iocaiized in the subeorneai 2one (Fig. 2). After 48 and 72 h the mieroabscesses became progressively flattened and faded away in all patients. The dermal PMN-aeeumuiation was in phase with the epidermai
416
E.M.G.J.DE JONG et al. positive cells were only sporadically present at all time intervals. In contrast, dermal CD14-positive cells already were increased diffusely (P=0-05) at 8 h after LTB4application (Fig. 3a-b). Their numbers remained increased until 48 h, and diminished at 72 h. Langerhans cells (Fig. lc). In the epidermis the CDlpositive Langerhans cells were present at all time
Figure 2. Staining with anti-eiastase, showing maximal accumulation of PMN 24 h after LTB4-application ( x 408).
infiltration, however, at 8 h relatively more cells were seen in the dermal compartment. At 72 h, the PMNextravasation had considerably diminished but was still present. CDM-positive cells {Fig. lb). In the epidermis, CD14-
s*^:'.
;..
^
Figure 3. Staining with WT14, showing CD14-positive-celIs before (a) and 8 h (b) after application of LTB4 ( x 204).
Figure 4. Langerhans cells before application of LTB4 (a); alter 24 h jn the vicinity of micro-abscesses (b); and after 72 h decreased density in epidermis, increased density in dermal compartment (c) ( x 204).
IMMUNOHISTOCHEMICAL STUDY OF LTB4 intervais. The density of these ceils in the epidermai compartment showed a statisticaiiy significant decrease at 72 h {P
