Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
68:259-264, 1977 Copyright Cl 1977 by The Williams & Wilk ins Co. THE J OURNAL OF iNvEsTIGATIVE DERMATOLOGY, Vol. 68, No.5 Printed in U.S A . REPORTS EXPERIMENTAL PRODUCTION OF RABBIT ANTI-GUINEA-PIG EPIDERMAL CELL SERA. COMPARISON TO PEMPHIGUS ANTIBODIES MASAHIRO TAK1GAWA , M .D ., AND SADAO IMAMURA , M .D Department of Dermatology. Faculty of Medicine, Kyoto University . Kyoto. Japan Anti-epidermal cell sera (AES) were obtained by immunizing rabbits with enzymatically dispersed, viable guinea-pig epidermal cells followed by absorption with guinea-pig red blood cells, spleen and thymus cells, and liver powders. Complement-mediated cytotoxicity and immunofluorescence demonstrated that AES were directed towards cell surface antigens specific for stratified squamous epithelia of guinea pigs, and cross reacted with the corresponding tissues of humans and monkeys. Immunofluorescence revealed that AES reacted with Hassall's corpuscles and surrounding epithelial cells of the guinea-pig thymus which seemed to share common antigens with the epidermis; AES gave no reaction with other organs. While antigens reactive with pemphigus antibodies (PA) were demonstrated by membrane immunofluorescence to be present on the epidermal cell surface, PA showed no cytotoxicity to guinea-pig and human epidermal cells. Re-treatment of isolated epidermal cells with trypsin showed that antigens ractive with PA were more s usceptible to the enzyme than those reactive with AES. These findings suggest that the cell surface antigens binding AES are different from the antigens which bind PA. Studies of the cell surface structure have revealed that certain kinds of cells possess a variety of antigens on their membrane. Some of these antigens seem to be exclusive for a specifi.c population or particular to a certain stage in their differentiation. Such antigens have been well studied in lymphoid cells I11. Although epidermal tissue-specific antigens have been widely investigated 12-5], they have rarely been demonstrated on the cell surface. Recently, the presence of histocompatibility antigens and a new system of alloantigens was demonstrated on the surface of mouse epidermal cells [6-8]. Also, heteroantigens were detected on the epidermal cells of mouse and rat using heterologous antisera; the sera reacted with the surface antigens of epidermal cells having a broader specificity which cuts across species differences [9,1 01. There is. therefore, a similarity between heteroloManuscript received May 17, 1976; accepted for publication December 2, 1976. This work was supported, in part, by a Research Grant for Specific Disease from the Ministry of Health and Welfare. Reprint requests to: Dr. M. Takigawa, Department of Dermatology , Faculty of Medicine, Kyoto University. Kyoto 606, Japan. Abbreviations: AES: anti-epidermal cell sera C: complement FCS: fetal calf serum FITC: fluorescein isoth.iocyanate MEM: minimal essential medium PA: pemphigus antibodies 259 gous antisera and pemphigus antibodies (PA) in that they are both directed towards the tissuespecific but not species-specific antigens of the epidermis D 1]. This report deals with the experimental production of heterologous antisera to guineapig epidermal cells (AES) and a comparison of the reactivities to guinea-pig epidermis between AES and PA. MATERIALS AND METHODS Isolation of epidermal cells . A modification of the methods of Karasek and Charlton [12], and Vaughan and Bernstein [13] was used. The back, flanks, and abdomen of out bred guinea pigs (Hartley strain) of both sexes. weighing 250 to 500 gm. were depilated, shaved. and cleaned with 70'*' ethyl a lcohol. Thin skin fragments were removed with razors and washed briefly in Eagle's minimal essential medium (MEM) containing 100 units/ml of penicillin and 100 p.g/ml of streptomycin. The fragments were incubated at 37•c in a 0.3% solution of trypsin (1:250, Difco) which was prepared in CAh- and Mg++-free phosphate-buffered saline and sterilized by membran filtration . The time of incubation was 45 to 75 min depending on the thickness of the fragments. After incubation the epidermis was separated from the dermis with forceps, cut into about 5-mm rectangles, and gently stirred magnetically in a flask containing MEM for about 30 min at. 37•c to disperse the cells. The resulting cell suspension was filtered through cotton wool. and the cells were washed 3 times by centrifugation with MEM containing 10% heat-inactivated fetal calf serum (MEM- FCS). The yield of cells was 3-5 x 107 cells/guinea pig. According to trypan blue exclusion , more than 95% of the cells were viable. 260 Vol . 68, N o.5 TAJUGAWA AND IMAMURA When the cells were to be used for immunization. they were suspended in pyrogen-free physiologic saline. Preparation of anti-epidermal cell sera. Viable epidermal cells were injected in random-bred white female rabbits, weighing 1.8 to 2.0 kg. Two procedures were employed: (1) Rabbit 1 was injected intramuscularly with 1.5 x 107 cells in 1 ml of saline to which was added an equal volume of Freund's complete adjuvant. Two and 4 weeks later a similar number of cells in 5 ml of saline was given intravenously. (2) Rabbits 2 and 3 were injected intravenously with 1.5 x 107 cells in 5 ml of saline 3 times at intervals of2 weeks. One week after the last injection, blood was drawn from the carotid arteries. The sera were inactivated at 56"C for 30 min and stored frozen at -2o•c until used. The sera from r abbits 1, 2, and 3 were referred to as AES 1, 2, and 3. r espectively. Pemphigus sera. The clinical and immunologic data of 4 patients with pemphigus are described briefly in Table III. All sera were stored at -2o•c . Absorption of sera . An aliquot of AES was absorbed at least 3 times with an equal volume of packed red and spleen cells and 1 / 10 volume of thymus cells at room temperature for 30 min. The cells were obtained from more than 3 guinea pigs each time. AES and pemphigus sera were eventually absorbed in the same manner with the cells of a guinea pig used for cytotpx:icity testing or immunofluorescence and with guinea-pig liver powders (30 mg/ml ) which were prepared according to the method of Coons et al [141. Cytotoxicity test. The method of Masuda et all15) was employed. A cell suspension of 2.5 x 10"/ ml was prepared in MEM. Pooled guinea-pig sera absorbed with guinea-pig epidermal and spleen cells were used without dilution as a source of complement (C). Four hundredths milliliter of a cell suspension, 0.05 ml of serum. serially diluted. and 0.01 ml ofC were incubated at 37"C for 45 min. Cell counts, living or dead , were made after addition of0.1 ml of freshly prepared 0.2% trypan blue. Tests always were performed in duplicate. Controls were included in whkh the test cells were incubated with medium instead of serum <C alone) or C (serum alone). The results are expressd by the following formula; primary filter and 41 and 65 secondary filter s. Photographs were taken on a Fuji or Kodak Tri-X fUm. RESULTS Cytotoxicity Test Figure 1 shows the ability of absor bed AES to lyse guinea-pig and human epidermal cells in the presence of C as measured by trypan blue exclusion. Isolated human epidermal cells were obtained by trypsinization of the epidermis of surgically excised scar tissue from a burned patient by 8:----o -- ' ' -----q. --.6 ' tl) ....:l ....:l 0 co \\ \ \ IJ;l u Cl IJ;l ' "0 \ \ \ \ \ '' 0 1.0 H 0 IJ;l 0.. \ \ \ ' ' ',, ~ z IJ;l up:; \ '~ ....:l ....:l E-< \ \ \ \ ,, ~ \ \ \ \\ \ \ 0 N ~ 1 ' \ '' 0 ''li:'' 'CJ"'.... ..........' ........... ~ ' 25 125 ANTISERUM DILUTION 5 625 (RECIPROCAL) Fie . 1. Complement-mediated cytotoxic activity of rabbit anti-guinea pig epidermal cell sera <AES\ towards guinea-pig epidermal cells (broken lines\ and human epidermal cells (solid line\. 0 , AES 1; 6., AES 2; D, AES 3. Percent killed cells = 100 x 1 living cells with serum + c) TABLE I. Titration of AES on frozen tissue sections" - ( living cells with C alone Fluorescein isothiocyanate fFITCJ conjugates. An antibody to rabbit IgG was prepared in a sheep and conjugated with FITC by the method of Masuda [16]. The conjugate obtained had a F/P molar ratio of2.0 and was used at a protein concentration of 1 mg/ml. The monospecificity was confirmed by double diffusion in agar gel. FITC-conjugated rabbit antihuman IgG was obtained from Miles Lab. , ill., and Dakopatts, Denmark. The commer cial conjugates were used at a dilution of 1:10. Immunofluorescence. An indirect method was used on frozen tissue sections as previously described 117). For membrane staining of viable cells, 1-2 x 10" cells were incubated with 0.2 ml of diluted sera for 30 min at 4•c, washed twice with Dulbecco's phosphate-buffered saline containing 1% bovine serum albumin. and reacted with 0.2 ml of a conjugate for 30 min at 37•c . The cells were washed. r esuspended in the same medium, and wet mounts prepared. The slides were viewed with a Zeiss fluorescence microscope equipped with a UV-5 Guinea pig Oral mucosa Esophagus Epidermis: BaM! layoc ~ Spinous layer Granular layer Horny layer Hair follicle Sebaceous gland Thymus (Hassall's corpuscle) Monkey Esophagus Rabbit, rat, and mouse oral mucosa AES AES AES #1 #2 #3 640 640 640 640 320 640 160 80 l60 80 160 40 40 40 40 20 40 80 80 40 20 20 20 20 20> 20> 20> " Titer of serum at a highest dilution giving a positive reaction by indirect immunofluorescence. May 1977 the procedure described in Materials and Methods. AES 1, 2, and 3 showed similar lytic activities to the guinea-pig cells, while it. was impossible to determine which type of cells, i.e. , basal, spinous, or granular, was more susceptible to lysis by AES. Control (C alone) always contained more than 90% living cells. The samples that were incubated with undiluted AES alone agglutinated to a variable degree, while those incubated with serially diluted AES alone resulted in less than 10% killed cells. AES 2 showed weak cytotoxicity against human cells. AES 1 and 3 were not tested on human cells. It was confirmed that AES did not kill guinea-pig spleen and thymus cells. On the other hand. all pemphigus sera diluted to 1:5 or more were not cytotoxic on guinea-pig or human epidermal cells. Undiluted pemphigus sera ANTI-GUI NEA-PIG EPIDERMAL CELL SERA 261 which were incubated with or without C agglutinated the viable guinea-pig cells to a variable degree. I mmunofiuorescence Frozen tissue sections. Full titrations of AES on a variety of tissues are given in Table I. All AES showed an intercellular staining pattern on sections of guinea-pig skin, oral mucosa, and esophagus. which extended in all layers oft..he epithelium (Fig. 2A ). End titers in the horny layer were lower than those in other layers of the epiderrrus. Hair follicles and sebaceous glands also revealed intercellular staining (Fig. 2B ). No specific fluorescence was revealed in the cytoplasm of cells of the stratified squamous epithelia or the dermal com- FIG . 2. Indirect immunofluorescence staining of frozen sections. A: Guinea-pig oral mucosa <AES 3, 1:40. x 130). B: Guinea-pig hair follicles and sebaceous glands (AES 2, 1:20. x 130). C: Guinea-pig oral mucosa (pemphigus serum, case 1, 1:40, x 130). D : Guinea pig thymus <AES 2, 1:20, x 320). 262 Vol . 68, No .5 TAKIGAWA AND IMAMURA ponents other than hair follicles and sebaceous glands. Sections of monkey (Macaca rhesus) esophagus reacted with AES showed an intercellular staining in all layers of the epithelium. AES did not react to rabbit, rat, and mouse oral mucosa. When the patterns obtained with guinea-pig oral mucosa were compared with those by pemphigus sera (Cases 1 to 4) on the same tissue, the similarities were striking (Fig. 2C). Although absorbed AES did not react with the guinea-pig brain, kidney. spleen, or liver, staining of the thymus revealed whorly, homogenous fluorescent areas corresponding to Hassall's corpuscles and surrounding epithelial cells; no other areas of the thymus were stained lFig. 2D ). Pemphigus sera showed n o fluorescence in the guinea-pig thymus. Control sera included pooled normal rabbit sera and rabbit antifowl lgG antisera which was prepared by two injections of fowl IgG plus Freund's complete adjuvant. Absorbed control sera gave no reactions with any of the above tissues. Viable cell suspensions. The guinea-pig epidermal cells treated with AES or all pemphigus sera at 4°C and then placed with FITC-conjugates at 37°C showed fluorescence all around the cell membrane. Different staining patterns were recognized among the various cell types. The larger cells showed a circumferential ring pattern and the smaller ones displayed a m·ore patchy. interrupted ring type reaction (Figs. 3A , B). The titrations of AES and pemphigus sera on viable cells and the percentage of fluorescence-positive cells are given in Tables II and ill. As the sera were diluted. the number of positive cells was decreased. The present study could not demonstrate whether or not fibroblasts. which might contaminate epidermal cell preparations, reacted with AES; our immuneferritin electron microscopic study disclosed that fibroblasts were antigen-negative.* Effects of Further Trypsinization Dispersed guinea-pig epidermal cells concentrated to 10u were further reacted with 1 ml of 0.03% trypsin solution at 37°C for 30 min with occasional pipetting. After t he incubation . the cells were washed with MEM- FCS. Viability as counted by dye exclusion was more than 90% of untreated cells. The viable cells were stained with AES or pemphigus sera at a dilution of 1:10 as described above. Although the intensity of fluorescence seemed to be unchanged, most of the cells that reacted with AES displayed a ring staining pattern; the interrupted pattern was rarely observed. On the other hand, fluorescent spots on the cell surface were markedly reduced and weakened in the cells that reacted with pemphigus sera. DISCUSSION Surface antigens specific for cells of the epidermis and stratified squamous epithelia were de- * Takigawa M, Imamura S, Ofuji S: Manuscript in preparation. FIG . 3. Membrane immunofluorescence staining of viable guinea-pig epidennal cells. A: Cells treated with AES 3 (1:10, x 320l. B: Cells treated with pemphigus serum (case 5, 1:10. x 320). Il. Percentage of antigen-posiliue cells shou·ing membrane immunofluorescence by AES More than 200 cells were coun ted and the number of fluorescent positive cells was examined. TABLE Dilution of serum AES #1 AES #2 AES #3 1:160 > 90 > 90 1:320 1:640 53 50 0 > 90 33 10 0 tected by both complement-mediated cytotoxicity and immunofluorescence using absorbed AES. Irrespective of the immunization procedures, AES 1. 2, and 3 showed similar cytolytic activity against trypsinized guinea-pig epidermal cells. It was impossible to determine which type of cells was most susceptible to lytic activity. Immunofluorescence likewise did not reveal specific preference for the different cell types even with diluted antisera. It appears that the antigen(s) detected by AES is present on all epidermal cells. Immunofluorescence on frozen sections revealed that AES did not react with other organs of guinea pig except Hassall's corpuscles and surrounding epithelial cells of the thymus. Thymic epithelial cells are now considered to be of ectodermal origin and have morphologic similarities to epidermal cells [18]. The existence of a cross-reactive antigen has been demonstrated in the thymic epithelial May1977 ANTI-GUINEA-PIG EPIDERMAL CELL SERA TABLE Ill. I m m unoflourescence of pemphigus sera on guinea-pig epithelial tissue Clinical diagnosis End titer on froten oral mucosa Percentage of antigen-positive cells showing membrane immunofluorescence Dilution of serum 1:10 1:20 1:40 1:80 1:160 1:320 263 Case 11'4 CIIJ!e #1 Case # 2 Case #3 P . vulgaris P . vulgaris P . foliaceus P . erythematosus 160 160 320 1280 81 64 35 0 71 34 0 64 32 0 > 90 81 47 10 1:640 - = Not done . cells and epidermal cells of the guinea pig using antibodies to the polysaccharide of Group A streptococci I19]. Our study a lso confirmed that the thymic epithelial cells and Hassall's corpuscle share common antigen(s) with epidermal cells. AES showed cross reactivity with the epidermal cells of other species including humans and monkeys, indicating that AES was not species specific. In huma ns, this was confirmed by the cytotoxicity assays. In monkeys. intercellular staining was demonstrated in all layers of the epithelium. AES a re different from PA in that the former are heterologous antisera and the latter are autoantibodies. A comparison between AES and PAis. however , feasible since both are tissue specific but not species specific as revealed by similar immunofluorescent staining patterns. One salient point is that. unlike AES. PA did not show complementmediated cytolysis of guinea-pig and human epidermal cells despite the presence of a substance binding to PA on guinea-pig epidermal cells as demonstrated by membrane immunofluorescence I20 [. It has been shown that P A do not appear capable of complement fixation by in vitro complement staining methodology 121.221. Furthermore. the complement system did not play a role in in vit ro PA-mediated acantholysis wruch was induced by the incubation of the human epidermis with pemphigus sera for more than 6 hr 123]. Thus. PA do not show complement-mediated cytotoxicity to guinea-pig and human epidermal cells during short incubations. The further treatment of isolated guinea-pig epidermal cells with 0.03% trypsin solution resulted in a decrease in the number and intensity of PA fluorescent spots on cell membranes. Susceptibility of antigens reactive with PA to the proteolytic enzyme has been reported previously [24]. The same treatme nt did not change the intensity of fluorescence to AES although the staining pattern was mostly the ring type. These findings imply that the cell surface antigens reactive with AES are different from those reactive with PA. REFERENCES l. Boyse EA, Old LJ: Some aspects of normal and abnormal cell surface genetics. Annu Rev Genet 3:269- 290, 1969 2. Dumonde DC: I n Advances in Immunology, vol 5, Edited by FJ Dixon Jr. New York , Academic, 1966, pp 251 - 253 3. Freedberg IM, Baden HP: Studies of epidermal protein metabolism. I. Incorporation of amino acids in vivo. J Invest Dermatol 39:339-345, 1962 4 . Aoki T, Parker D. Turk JL: Analysis of soluble antigens in guinea-pig epider mis. I. An immunoelectrophoretic study with special r eference to tissue-specific antigens and enzyme antigens. Immunology 16:485-497, 1969 5. Tezuka T , Freedberg IM: Epidermal structural proteins. ill . Isolation and purification of histidineri ch protein of the newborn rat . J Invest Dermato) 63:402--406, 1974 6. Cooper S. Lance EM: A serological method for detecting the surface antigens of epiderma l cells. Transplantation 11:108--110, 1971 7. Worst PKM. Fusenig NE: Histocompatibility antigens on t he s urface of cultivated epidermal cells from mouse embryo skin . Transplantation 15:375-382, 1973 8. Scheid M, Boyse EA, Carswell EA, Old LJ: Serologically demonstrable alloantigens of mouse epidermal cells. J Exp Med 135:938-955. 1972 9. Worst PKM, Fusenig NE: A tissue antigen on the surface of cultivated mouse epidermal cells detected by mixed hemadsorption technique. J Invest Dermatol 61:277-281 , 1973 10. Lloyd KO. Dam ule TV: Properties of rabbit antirat epidermal cell sera. Dem onstration of tissuespecific differentiation antigens. J lmmunol ] 12:311-319, 1974 11. Beutner EH, Chorzelski TP. Jordon RE (eds ): Autosensitization, In Pemphigus and Bullous Pemphigoid. Springfield, Ill , Thomas, 1970 12. Karasek MA, Charlton ME: Growth of postembryonic skin epithelial cells on collagen gels. J In.v est Dermatol 56:205-210, 1971 13. Vaughan FL, Bernstein IA: Studies of proliferative capabilities in isolated epidermal basal and differentiated cells. J Invest Dermatol 56:454--466. 1971 14. Coons AH . Leduc EH. Connolly JM: Studies on antibody production. I. A method for the histo- chemical demonstration of specific antibody and its application to a study of the hyperimmune rabbit. J Exp Med 102:49-60, 1955 15. Masuda T. Usami K , Yodoi J: B cell subsets in mice discerned by heterologous anti-mouse B cell anti- 264 Vol. 68, No .5 TAKIGAWA AND IMAMURA body. Annu Rep lnst Virus Res (Kyoto) 1:91-92, 1973 16. Masuda T: In Immunochemistry. Edited by S Migita. Tokyo, Nakayama Press, 1972, pp 263-270 17. Tagami H, Imamura S: Benign mucous membrane pemphigoid. Demonstr ation of circulating and tissue-bound membrane a ntibodies. Arch Dermato! 109:711-713. 1974 18. von Gaudecker B, Schmale E: Simila rities between Hassall's corpuscles of the human thymus and t he epidermis. An investigation by electron microscopy and histochemistry. Cell Tissue Res 21. 22. 23. 151:347-368, 1974 19. Lyampert IM, Beletskaya LV, Borodiyuk NA: A cross-reactive an tigen of thymus and skin epithelial cells common with the polysaccharide of IP'oup A str eptococci. Immunology 31:47-55. 1976 20. Bngden WD, Amos HE: The differential binding of 24. antibody from the sera of patients with pemphigus and pemphigoid to isolated guinea-pig epidermal cells. Br J Dermatol 93:425-430, 1975 Jordon RE , Sams WM Jr, Diaz G, Beutner EH: Negative complement immunofluorescence in pemphigus. J Invest. Dermatol 57:407-410, 1971 Jordon RE, Schroeter AL, Rogers RS Ill , Perry HO: Classical and alternate pathway activation of complement in pemphigus vulgaris lesions. J Invest Dermatol 63:256-259, 1974 Schlitz JR, Michel B: Production of epidermal acantholysis in normal human skin in vitro by the IgG fraction from pemphigus serum. J Invest Dermatol 67:254-260, 1976 Shu S. Beu tner EH: Isolation and characterization of antigens reactive with pemphigus antibodies. J Invest Dermatol 61:270-276. 1973