Use of the peroxidase-antiperoxidase technique for differential staining of multiple cell types in the rat pancreatic islet.код для вставкиСкачать
THE ANATOMICAL RECORD 209:417-422 (1984) Use of the Peroxidase-Antiperoxidase Technique for Differential Staining of Multiple Cell Types in the Rat Pancreatic Islet JOYCE A. DELEO AND MICHAEL R. SCHWEISTHAL Department ofdnatomy, School of Medicine, Oral Roberts University, Tulsq OK 74171 ABSTRACT Multiple staining of the endocrine cells of the pancreatic islet was studied in tissue obtained from adult rats. After fixation in Bouin's fluid and processing for light microscopy, the unlabeled peroxidase-antiperoxidase (PAP)technique was performed. Successful staining procedures used variations of the PAP technique with 3,3'-diaminobenzidine (DAB) and 4-chloro-1-naphthol (CN) as chromagens. The purpose of this study was to stain as many of the four primary cell types (A-cells, B-cells, D-cells, PP-cells) as possible either simultaneously or sequentially using photomicroscopy. At optimum antibody titer, there was minimal nonspecific background staining which made it possible to differentiate cell types by intensity of the chromagen. Any two cell types can be shown by using DAB with the first antibody and CN with the second. To demonstrate three cell types simultaneously, three methods which altered dilutions and chromagens were used. The first method consisted of decreasing dilutions of primary antibody with DAB and CN as the chromagens. The second method involved repetitive DAB applications resulting in three intensities of brown. The third method used a DAB immersion after the second cell type was stained. This produced a color differential so the third cell type could be distinguished with CN. To demonstrate the three cell types sequentially, a masking technique was introduced with photomicroscopy. In order to block the preceding complex, the previous cell type (demonstrated by CN) was restained with DAB at a n increased dilution. The next cell type was then stained with CN. These four methods were tried in attempts to stain four cell types in the same tissue section. Immunocytochemistry is a n invaluable tool in elucidating cell types in various systems, such as in the identification of cell types in pancreatic islet tumors and other applications in the field of pathology (DeLellis et al., 1979; Hutson et al., 1979; Mukai et al., 1982; Polak and Van Noorden, 1983).It has become increasingly desirable to demonstrate more than one antigen in the same tissue section to determine topographic localization and distribution (Orci et al., 1976; El-Salhy et al., 1982, 1983). The unlabeled peroxidase-antiperoxidase (PAP) technique (Sternberger, 1979)is one of the most sensitive and specific of all immunocytochemical methods (Montero, 1981; Childs, 1982; Grzanna, 1982; Vacca, 1982) and for this reason it is the best 0 1984 ALAN R. LISS, INC. available technique to use for multiple staining. In studying the pancreatic islet, it would be advantageous to- simultaneously stain the four principle cell types with different chromagens. Due to the limited number of available insoluble chromagens, alternate methods must be developed to depict more than two cell types in the same tissue section. It is the purpose of this paper to describe some extensions of the PAP technique that were tried in a n effort to determine the optimum conditions that would permit the simultaneous or sequential staining of two, three, and four cell types of the pancreatic islet. Received November 28. 1983; accepted March 2, 1984 418 J.A. DELEO AND M.R. SCHWEISTHAL MATERIALS AND METHODS Pancreatic tissue from adult Sprague-Dawley rats was processed for light microscopy after fixation in Bouin's fluid. The unlabeled PAP technique was performed on 5-pm sections with excess sheep-antirabbit IgG (SARG) a t 1:250 and PAP complex a t 1:400 for 1-hour incubations. The secretory products of the four primary islet cell typesinsulin (Cambridge Labs), somatostatin-14 (Peninsula Labs), glucagon (Dako Labs), and pancreatic polypeptide (R.E. Chance, Eli Li1ly)-were used as the primary antibodies in overnight incubations. The four primary antibodies were diluted to the appropriate concentration using a 1 O : l ratio of phosphate-buffered saline (pH 7.2) and 0.5% bovine serum albumin. Optimal dilutions should be determined for each lot of antiserum (Petrusz, 1983). Dilutions of the primary antibodies vary in different steps of the procedures depending on the sequence chosen and the desired result (Table 1). The antisera were used in combination with the insoluble chromagen, 3,3'-diaminobenzidine (DAB) and the soluble chromagen, 4-chloro-1-naphthol (CN), using a daylight filter to facilitate viewing of positive CN reactions. For multiple staining, the PAP technique was performed, consecutively altering the primary antibodies, their corresponding dilutions, and the sequence of the chromagens. For staining only two cell types, any two primary antibodies may be used with DAB as the first chromagen and CN as the second (Joseph and Sternberger, 1979).For staining more than two cell types, the first primary antibody used must have a high specificity at a high dilution to limit background staining. Four protocols were used for staining more than two cells in the pancreatic islet. Three of these four methods involved various combinations of the primary antibodies and the chromagens in order to obtain simultaneous demonstration of three or four cell types. The fourth method employed the use of photomicroscopy to illustrate the staining of multiple cell types in sequence on the same tissue section. These four methods are outlined below. Method 1 For the simultaneous demonstration of three cell types, three primary antibodies were used at decreasing dilutions with the two chromagens DAB and CN. The first primary antibody was applied at a high dilution with DAB as the chromagen. The slides were then washed in phosphate-buffered saline (PBS); a PBS wash was always used between each application of the primary antibody regardless of the method. The next primary antibody was applied at a decreased dilution using DAB again. The third primary antibody was applied at a further decreased dilution in combination with CN. For the simultaneous demonstration of four cell types, four primary antibodies were used at decreasing dilutions using DAB in the first three stains followed by CN. Method 2 A second procedure makes use of different levels of binding between varying dilutions of primary antibodies with one chromagen. The sequence used was as follows: The first primary antibody was applied a t a high dilution with DAB, followed by the second antibody a t a lower dilution with DAB, and by a third antibody at a n even lower dilution and DAB. This method was varied to include four primary antibodies in a similar sequence with the fourth antibody being the least dilute. Method 3 The third method involves using DAB in a n immersion technique. The first cell type was stained with DAB using the primary antibody a t a high dilution. The second cell type was stained with CN at a lower dilution. The slides were then immersed with agitation in DAB for 2-10 minutes. The third primary antibody was applied at a still lower dilution and the process was repeated with CN as the chromagen. A fourth primary antibody can be applied at a n even lower dilution following a 2-10-minute immersion in DAB, and CN as the final chromagen. Method 4 For the sequential staining of three andlor four cell types, a masking technique was used. This involved reapplying the previous primary antibody but changing the chromagen used to stain it. The first cell type was stained with DAB using a high primary antibody dilution. The second cell type was stained using a lower dilution and CN. The tissue was photographed a t this point. The latter primary antibody was reapplied a t a higher dilution and stained with DAB. The MULTIPLE STAINING OF ISLET CELL TYPES third primary antibody was applied at a decreased dilution using CN as the chromagen. The tissue was photographed again and the process was repeated for a fourth primary antibody. Sequential photomicrographs were used to record color changes a t each application. Other methods were used in a n effort to improve multiple staining. Acidification (HC1) and dimethylformamide (DMF) were used in a n attempt to remove bound antibodies. Normal sheep serum (NSS) and methanolic H202 were tried to reduce background staining. Controls were used a s specified by Sternberger (1979). (Figures 1-6 are color illustrations that appear in the Color Section elsewhere in this issue. These figures are located on pp. 392393.) 419 DAB (glucagon) and CN (insulin), respectively, were then placed in a DAB solution for either 2 or 7 minutes. The third primary antibody was then stained with CN (somatostatin). As illustrated in this figure, the slides were in the solution for 7 minutes, resulting in a brown, blue, and purple. Antisomatostatin was used here at 1:10,000 since the antiglucagon that started the staining sequence was not a high-titer antiserum. The attempts to stain four cell types were unsatisfactory with this method due to poor color differentiation. In the masking procedure, the entire technique was performed again after the second cell type was stained with CN (Fig. 6). The same primary antibody is reapplied at a n increased dilution due to the higher affinity of DAB compared to CN and restained with DAB. This procedure better differentiates RESULTS color but increases the number of times the In regard to dual antigen staining, any two technique is performed on a given section. cell types of the pancreatic islet can be dem- This results in increased background stainonstrated using the two chromagens DAB ing or a complete absence of color reaction. and CN. In Figure 1the dilutions of antiglu- The sequential staining of four cell types was cagon and antiinsulin are 1:lOO. Three cell of questionable value since, in some cases, types cannot be obtained with concentrations positive identification was difficult to ascerthis high because of increased background tain while in others, no staining occurred. To staining when the third antibody is applied. stain four cell types using sequential phoFigure 2 illustrates the decreased intensity tomicrographs, the technique must be perof staining with increasing dilutions of an- formed six times on a given section. This tiinsulin and antisomatostatin, 1:4,000 and much treatment not only hinders staining, 1:15,000respectively. The dilutions we found but also adds the risk of losing the tissue to be most effective for each primary anti- from the slides before all the sequences are serum are summarized in Table 1. completed. In simultaneously staining three cell types, After several attempts to use HC1, DMF, the first method, which used decreasing di- NSS, and methanolic H202 to improve mullutions of primary antibody and two chrom- tiple staining, they were discontinued. When agens, resulted in three different colors: a these procedures were used in dual antigen brown-black DAB reaction product, a light staining, the results were unchanged. When tan DAB product, and a blue CN product they were used during repetitive applica(Fig. 3). We obtained the best results by using tions of the PAP technique, the results were antisomatostatin a t 1:15,000 as the first pri- poor because of tissue release from the slide, mary antibody. Efforts to simultaneously possible mixing of CN and DAB, and reducstain four cell types using this method were tion in CN staining. unsuccessful. DISCUSSION The second method, using repetitive DAB applications with decreasing dilutions of priThe demonstration of two and three cell mary antibody, produced three intensities of types was successfully performed using the brown (Fig. 4). It was difficult to obtain four PAP technique under controlled procedures intensities of one color using one chromagen with specific primary antibody dilutions. We and consequently this method cannot be used concur with Sternberger and Joseph (1979) for simultaneous demonstration of four cell that the best method for dual antigen staintypes. ing is the use of insoluble brown DAB folThe third method using DAB immersion is lowed by soluble blue CN. illustrated in Figure 5. In this procedure, the The most appropriate factors for successful slides with two cell types already stained by triple antigen staining are optimal dilutions 420 J.A. DELEO AND M.R. SCHWEISTHAL TABLE 1. Dilutions for staining the four principle cell types of the rut pancreatic islet Antibody Insulin Dilutions Description Significant background staining, very dark core. Some background staining, dark granular positive cells. Limited background staining, lighter positive cells. No background, light but distinguishable cells. Good for masking technique (Method 4). 1:100 1:1,000-1:4,0001 1:4,000-1:8,000 1:10,000 Glucagon Background staining, very dark positive cells with core staining. Some background and core staining, positive cells lighter. Minimal background and core staining, positive cells can be distinguished. Very little overall staining. 1:lOO 1:1,0001 1:2,000-1:8,000 1:10,000 Somatostatin Intense background and core staining, very black positive cells. Decreased background, black cells. Core is less intense, brown cells. Core is lighter, positive cells are easily visible. Very little background, cells distinguishable. Cells lighter but visible. Cells extremely light. 1:lOO 1:1,000 1:2,000-1:4,000 10,000 1:8,000-1: 1:15,0001 1:20,000 1:40,000 Pancreatic polypeptide Background and core too dark, very dark cells. Significantly better background, dark cells. Cells beginning to get lighter, some background staining. Background has not changed, cells are lighter. Background the same, cells extremelv light. 1:lOO-1:1,000 1:2,000' 1:4,000-1: 15,000 1:20,000 1:40,000 'Optimal dilutions for each cell type for various multiple staining procedures. of primary antibody, adequate washings of slides between steps, and freshly prepared chromagen solutions with or without timed DAB immersion and agitation. The sequential demonstration of three and possibly four cell types can be shown by DAB masking, photomicrographs, proper antibody titer, and relatively pure antisera. Among the problems encountered is that of background staining which will occur if primary antibody concentrations are not optimal, adequate washings are not performed (especially after the application of primary antibody), thick sections are used, and/or deparaffhizing is not complete (Polak and Van Noorden, 1983). These problems become more prevalent as the number of times the technique is used on a section increases. The causes of such problems may be any of the following: 1) reactions of the second- or third-sequence linking antibody (SARG)with the first sequence PAP or primary antibody; 2) reactions of second or third primary antibody or PAP with preceding sequence SARG; and 3) reactions of CN MULTIPLE STAINING OF ISLET CELL TYPES and DAB with the preceding sequence of PAP. In multiple staining of three and/or four cell types, limitations arose at the fourth application of the technique. At times, background staining was too intense to distinguish any cell types and/or showed no blue reaction product. This may be due to a n increase in cross-reactivity and/or a decrease in the number of available binding sites. Exposures to HC1 and DMF were used by other investigators in attempts to elute tissue-bound antibodies and thus improve antigen staining (Vandesande and Dierickx, 1975; Erlandsen et al., 1976; Nakane, 1968; Grzanna, 1982; Polak and Van Noorden, 1983). These methods were tried in accordance with specifications between antibody applications and results were either unchanged or poor. The lower pH can possibly disturb the sensitive tissue and facilitate CN and DAB mixing (Sternberger, personal communication). Multiple staining of three cell types was obtained without the use of HC1 and DMF and consequently elution is not necessary for dual or triple staining. This is in agreement with results obtained in other laboratories (Lechago et al., 1978; Joseph and Sternberger, 1979; Sternberger and Joseph, 1979; Roth, 1982).Pretreatment of slides with 3% normal sheep serum and methanolic H202 was tried to reduce nonspecific background staining and to block endogenous peroxidase activity (Hsu et al., 1981). After attempts to reduce background staining with these methods, our results indicated that there was either no effect or actually reduced CN staining. Although we did not eliminate all possible dilutions and sequences, the successful staining of the four antigens may yet be possible with the following: alternate blocking agent, other insoluble chromagens, optimal dilutions, and/or pure antisera. Additionally, with the development of monoclonal antibodies and increased efficiency in their production (Bosman, 1983; Ormanns and Schaffer, 1983) the staining of four cell types in the pancreatic islet using immunocytochemistry should be possible. ACKNOWLEDGMENTS The authors wish to thank Dr. Ludwig Sternberger for his helpful conversations in suggesting alternate procedures and for his encouragement to publish this work; and Mr. Garv McCaulev for his professional help in doing the colorprinting. 42 1 LITERATURE CITED Bosman, F. (1983) Some recent developments in immunocytochemistry. Histochem. J., 15t189-200. Childs, G.V. (1982) Use of immunocytochemical techniques in cellular endocrinology. In: Electron Microscopy in Biology. J. Grifith, ed. John Wiley & Sons, New York, Vol. 2, pp. 107-177. DeLellis, R.A., L.A. Sternberger, R.B. Mann, P.M. Banks, and P.K. Nakane (1979)Immunoperoxidase technics in diagnostic pathology. Am. J. Clin. Pathol., 71:483-488. El-Salhy, M., E. Wilander, and G. Abu-Sinna (1982)The endocrine pancreas of anuran amphibians: A histological and immunocytochemical study. Biomed. Res., 6.579-589. El-Salhy, M., L. Grimelius, E. Wilander, B. Ryberg, L. Terenius, J.M. Lundberg, and K. Takemoto (1983) Immunocytochemical identification of polypeptide YY (PYY) cell in the human gastrointestinal tract. Histochemistry, 77:15-23. Erlandsen, S.L., O.D. Hegre, J.A. Parsons, R.C. McEvoy, and R.P. Elde (1976) Pancreatic islet cell hormones distribution of cell types in the islet and evidence for the presence of somatostatin and gastrin within the D cell. J. Histochem. Cytochem., 24:883-897. Grzanna, R. (1982) Light microscopic immunocytochemistry with fixed, unembedded tissues. In: Techniques in Immunocytochemistry.G.R. Bullock and P. Petrusz, eds. Academic Press, New York, Vol. 1,pp 183-204. Hsu, S., L. Raine, and H. Fanger (1981) Use of avidinbiotin-peroxidase complex (ABCJin irnmunoperoxidase techniques. J. Histochem. Cytochem., 29t577-580. Hutson, J.C., G.V. Childs, and P.J. Gardner (1979) Considerations for establishing the validity of immunocytochemical studies. J. Histochem. Cytochem. 27:12011202. Joseph, S.A., and L.A. Sternberger (1979)The unlabeled antibody method: Contrasting color staining of p-lipotropin and ACTH-associated hypothalamic peptides without antibody removal. J. Histochem. Cytochem., 27: 1430-1437. Lechago, J., N. Sun, and W. Weinstein (1978) Simultaneous visualization of two antigens in the same tissue section by combining immunoperoxidase with immunofluorescence techniques. J. Histochem. Cytochem., 27:1221-1225. Montero, C. (1981) Immunocytochemical methods and their achievements in pathology. Methods Achiev. Exp. Pathol., 10:l-36. Mukai, K., M.H. Greider, J.C. Grotting, and J. Rosai (1982) Retrospective study of 77 pancreatic endocrine tumors using the immunoperoxidase method. Am. J. Surg. Pathol., 6:387-399. Nakane, P.K. (1968) Simultaneous localization of multiple tissue antigens using the peroxidase-labeled antibody method: A study on pituitary glands of the rat. J. Histochem. Cytochem., 16.557-560. Orci, L., D. Baetens, M. Ravazzola, Y. Stefan, and F. Malaisse-Lagae (1976)Pancreatic polypeptide and glucagon: Non-random distribution in pancreatic islets. Life Sci., 19:1811-1816. Ormanns, W., and R. Schaffer (1983) The use of monoclonal antibodies against thyroglobulin in immunoperoxidase techniques. Histochemistry, 78: 177-180. Petrusz, P. (1983)Essential requirements for the validity of immunocytochemical staining procedures. J. Histochem. Cytochem., 31t177-179. Polak, J.M., and S. Van Noorden (1983) Immunocytochemistry: Practical Applications in Pathology and Biology. J.M. Polak and S. Van Noorden, eds. J. Wright PSG, Boston. 1 422 J.A. DELEO AND M.R. SCHWEISTHAL Roth, J. (1982) Applications of immunocolloids in light microscopy. J. Histochem. Cytochem., 30:691-696. Sternberger, L.A., and S.A. Joseph (1979)The unlabeled antibody method Contrasting color staining of paired pituitary hormones without antibody removal. J. Histochem. Cytochem., 27:1424-1429. Sternberger, L.A. (1979)Immunocytochemistry. S. Cohen and R.T. McCluskey, eds. John Wiley & Sons, New York. Vacca, L.L. (1982) “Double Bridge” techniques of immunocytochemistry. In: Techniques in Immunocytochemistry. G.R. Bullock and P. Petrusz, eds. Academic Press, New York, Vol. 1, pp. 155-182. Vandesande, F., and K. Dierickx (1975) Identification of the vasopressin producing and of the oxytocin producing neurons in the hypothalamic magnocellular neurosecretory system of the rat. Cell Tissue Res., I64:153162.