0013-7227/90/1262-1235$02.00/0 Endocrinology Copyright© 1990 by The Endocrine Society
Vol. 126, No. 2 Printed in U.S.A.
Different Expression of Protein Kinase C Isozymes in Pancreatic Islet Cells* KOJI ONODAt, MASATOSHI HAGIWARA, TAKAHISA HACHIYA, NOBUTERU USUDA, TETSUJI NAGATA, AND HIROYOSHI HIDAKAt Department of Pharmacology, Nagoya University School of Medicine (K.O., M.H., T.H., H.H.), Showa-ku, Nagoya, Aichi 466, and Department of Anatomy, Shinshu University School of Medicine (N.U., T.N.), Asahi, Matsumoto, Nagano 390, Japan
ABSTRACT. To elucidate differential roles of protein kinase C isozymes in pancreatic islet cells, the precise localization of the isozymes in rabbit and rat islet endocrine cells was investigated using monoclonal antibodies specific for three types of the enzyme. We detected strong immunoreactivity for the type III protein kinase C in B cells. Immunoreactivity for the type II enzyme was seen in A cells, and no apparent immunoreactivity for type I was observed in the islet cells. The expression of the type III protein kinase C in B cells was confirmed using rat
P
ROTEIN kinase C, Ca2+-activated phospholipid-dependent protein kinase, is believed to be involved in transmembrane signaling to regulate several cellular processes (1). This protein kinase is also known to be a Cellular receptor for the tumor-promoting phorbol esters which elicit various cellular responses including secretion, cell proliferation and differentiation, neurotransduction, and gene expression (2). In the pancreatic islet cells, protein kinase C has already been identified (3, 4). Recent progress in understanding the pathways involved in physiological insulin release indicates that the activation of protein kinase C in pancreatic B cells may be a pivotal event in the stimulus-secretion coupling mechanism (5). On the other hand, cDNA cloning and the sequence analysis of protein kinase C indicated that the enzyme was a family of genes (6-11). Brain protein kinase C was further resolved into three peaks, type I, type II and type III, chromatographically on a hydroxylapatite column (12). The identification of the protein kinase C isozymes in pancreatic B cells is required to elucidate the precise physiological role of the kinase in insulin secretion. HowReceived September 18, 1989. T h i s work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture, Japan. fPresent address: Department of Thoracic Surgery, Mie University School of Medicine, Edobashi, Tsu 514, Japan. JTo whom all correspondence and reprint requests should be addressed.
insulinoma cells. The predominant expression of the type III enzyme in these cells was shown by immunoblotting. Moreover, on the basis of an enzyme-linked sandwich immunoassay, the levels of protein kinase C isozymes were determined in these cells. The significant amounts of the type III enzyme was detected, but the contents of the type I and II enzyme were under detectable level. These results suggest that the type III protein kinase C is involved in the regulation of insulin release in pancreatic B cells. {Endocrinology 126: 1235-1240, 1990)
ever, direct biochemical demonstration of the isozymes in isolated pancreatic B cells has been technically difficult. More recently, we generated monoclonal antibodies, MC-la, MC-2a, and MC-3a, which specifically recognized three types of rabbit brain protein kinase C, respectively (13,14), and we demonstrated that the expression of protein kinase C isozymes in the rabbit cerebellum, spleen and human platelet was dependent on the cell type (13, 15). Using these monoclonal antibodies, we have showed a different expression of protein kinase C isozymes in the rabbit or rat pancreatic islet cells, and demonstrated the presence of an isozyme in a continuous cell line, RINr, which secretes insulin (16). Our results suggest that protein kinase C isozymes play differential roles in the regulation of hormone secretion in the pancreatic islet cells. Materials and Methods Immunohistochemical staining of pancreas Male adult rabbit or rat was anesthetized with an iv injection of pentobarbiturate (Nembutal) at 30 mg/kg. The pancreas was removed, trimmed of excess fat, and cut into small pieces. The tissue was immersed in phosphate-buffered paraformaldehyde with picric acid (17) for 6 h at 4 C, and washed with PBS containing 10%, 15%, or 20% sucrose for 12 h, respectively. The tissue frozen in liquid nitrogen was sectioned at 6 nm, using a cryostat, and placed on slide glass. The sections were 1235
1236
PROTEIN KINASE C ISOZYMES IN PANCREATIC ISLET CELLS
exposed to 10% normal horse serum in PBS for 20 min. The sections were then incubated with monoclonal antibodies (hybridoma supernatant), MC-la, -2a, and -3a, diluted from 1:10 to 1:100 in PBS for 12 h at 4 C. After being washed in PBS, they were processed for avidin-biotinylated peroxidase complex staining. Reaction product was developed by incubation in 0.03% (wt/vol) 3,3'-diaminobenzidine tetrachloride and 0.005% peroxidase in 50 mM Tris buffer, pH 7.5, and these sections were counterstained with 1% methyl green solution. In order to compare the localization of the type III protein kinase C with those of pancreatic hormones, serial paraffin sections were stained by the peroxidase antiperoxidase (PAP) method of Sternberger et al. (18), using MC-3a (anti-type III protein kinase C antibody) and antipancreatic hormones antibodies: anti-insulin, antiglucagon, and antisomatostatin (DAKO Corporation, Copenhagen, Denmark). Immunoblot analysis of protein kinase C isozymes in RINr cells. RINr cells were cultured in RPMI1640 medium (GIBCO, Grand Island, NY) supplemented with 10% fetal calf serum (GIBCO) and antibiotics (16). The whole homogenates of RINr cells (2 X 106) were electrophoresed in sodium dodecyl sulfate (SDS)-polyacrylamide gel (10%) and then transferred to nitrocellulose membrane (Bio-Rad Laboratories, Richmond, CA). The membranes were incubated with 2% normal horse serum in TPBS (PBS containing 0.05% Tween 20) to block nonspecific binding sites. The membranes were then exposed to respective antibody solution (100-fold diluted ascitic fluid) prepared in TPBS for 60 min. After being washed in TPBS, they were incubated in biotinylated horse antimouse immunoglobulin G (IgG) followed by the avidin-biotin complex of the Vectastain kit (Vecter Laboratory, Burlingame, CA). Immunoreactive proteins were visualized by reaction with 0.06% (wt/ vol) 4-chloro-l- naphthol and 0.02% H2O2. Sandwich enzyme immunoassay Three monoclonal antibodies, MC-la, -2a, and -3a, and polyclonal antibody against protein kinase C, PC-40 recognized the regulatory domain (33 or 35K) and the catalytic domain (48K), respectively, resulting from the trypsin-digested protein kinase C. Using these antibodies, the levels of protein kinase C isozymes, type I, II, and III in RINr-cells were determined by the enzyme-linked sandwich immunoassay. 0.5 ng of monoclonal antibodies in 100 /A of buffer A; 0.1 M sodium phosphate (pH 7.4) were adsorbed to each 96-well polystyrene flat-bottom well at 4 C overnight. Wells were washed thoroughly. Washing in this and subsequent steps was carried out after each step by three changes of 350 /xl of TPBS. Each of the monoclonal antibody-coated microtiter plate wells was incubated with 350 n\ of blocking buffer (1% BSA in buffer A) at 4 C overnight. Aliquots of 100 /xl of sample solutions and the solutions containing various amounts of purified protein kinase C isozymes for the standard curve were added to each well, and were left for 2 h at 4 C. Aliquots of 100 /A of PC-40 conjugated with horseradish peroxidase (100 ng in 100 fA PBS) were added to each well and the preparations further incubated for 2 h at 4 C. The enzyme reaction was initiated by the addition of 100 nl of 0.1 M citrate-phosphate buffer (pH 5.1) containing 0.01% H2O2 and 1 mg/ml o-phenyl-enediamine. After 10-min incuba-
Endo • 1990 Vol 126 • No 2
tion at room temperature in the dark, 100 /xl of 2 N sulfuric acid were added to stop the reaction. The optical density (OD) was measured at 492 nm by photometer. The contents of protein kinase C isozymes of sample solutions were calculated by using the respective standard curve of purified protein kinase C isozymes from rabbit brain. Results To investigate expression of protein kinase C isozymes in the rabbit pancreatic islet cells, immunohistochemical staining was undertaken, using monoclonal antibodies (MC-la, -2a, and -3a) which selectively interacted with hydroxylapatite column chromatographically resolved isozymes type I, II, and III of protein kinase C, respectively. Strong immunoreactivity with MC-3a was recognized in the cells of the central portion of the islets (Fig. 1A). Cells immunohistochemically reactive to MC-2a were observed in the peripheral region of the islets where glucagon-positive cells were located (Fig. IB). In all positive-staining cells the immunoreactivity could be observed in the cytoplasms but not in the nuclei. No apparent immunoreactivity was observed with MC-la using various dilutions of antibody solutions (Fig. 1C). By serial sections the localizations of pancreatic hormones and the type III protein kinase C were compared. Immunostaining with MC-3a exhibited a clear cut distribution pattern exactly identical to that of insulinpositive cells in the islets (Fig. 2, A and B). No product reacting with MC-3a was observed in the region where glucagon-and somatostatin-positive cells were located (Fig. 2, C-F). Moreover, we examined the localization of protein kinase C isozymes in the rat pancreatic islets. The staining patterns obtained with three monoclonal antibodies in the rat islets coincided with those in the rabbit islets (Fig. 3). When each antibody was reacted with an excess amount of purified brain protein kinase C before incubation, no immunoreactive products were observed (data not shown). In addition, no staining was observed in preparations in which one of the reagents in the staining protocol was deleted or with hybridoma supernatant fluids containing irrelevant antibody. Furthermore, to confirm the immunohistochemical localization of the type III protein kinase in pancreatic B cells, we used a continuous cell line, RINr, that secreted insulin, and the presence of glucagon in the supernatant fluid of the cell line could not be detectable (16). This cell line was originally established from a serially transplantable, radiation-induced, rat islet cell tumor. The whole homogenates of RINr cells were immunoblotted with our monoclonal antibodies (Fig. 4). A single major immunoreactive species of Mr 80,000 was detected with MC-3a and no other immunoreactive proteins were
PROTEIN KINASE C ISOZYMES IN PANCREATIC ISLET CELLS
1237
observed in whole homogenates of the cells. No crossreacting protein was observed in the cells with MC-la and MC-2a. Moreover, no reactive protein was observed with normal mouse IgG solutions. The levels of protein kinase C isozymes were determined by the sandwich enzyme immunoassay (Table 1). RINr cells contained 69.3 ± 8.6 ng/mg protein of the type III enzyme whereas the type I and II enzyme were under detectable level (4.1 ng and 0.65 ng/mg protein, respectively). These results indicate that the type III protein kinase C is predominantly expressed in this insulin-producing cell line.
Discussion In the present study, we at the first time demonstrated specific localization of protein kinase C isozymes in rabbit and rat pancreatic islet cells, by use of immunohistochemical techniques at the light microscopic level: A cells showed immunoreactivity for type II of the isozymes, and B cells showed that for type III. Moreover, evidence of the existence of the type III enzyme in insulin-secreting cells was confirmed by using insulinoma cells. Several reports have documented that protein kinase C is present in rat pancreatic islets (3, 4), rat insulinoma (19), and hamster clonal B cells (20). Inhibition of the insulin secretory response to glucose by polymyxin B, which inhibited protein kinase C-stimulated phosphorylation, indicated that protein kinase C played a role in glucose-induced insulin secretion (21). Zawalich et a/.(22) proposed that the initial phase of insulin secretion was due to calmodulin-dependent processes and sustained phase was controlled by a protein kinase C pathway. On the other hand, however, activation of Ca2+-calmodulin and cAMP-dependent protein kinase has been suggested to be involved in stimulus-secretion coupling in the pancreatic B cell (23-29). Schubart et al. (30) demonstrated that cAMP-mediated insulin release was not affected by trifluoperazine which inhibited cyclic nucleotide phosphodiesterase by binding to calmodulin, and suggested that cAMP and Ca2+ regulated the secretory process by distinct mechanisms. It is widely accepted that these three protein kinases may be involved in the regulation of glucose-induced insulin secretion from the pancreatic B cells, while its detailed intracellular mechanism is not clear. Therefore, our evidence that B cells contain the type III protein kinase C may throw some lights upon the elucidation of the complex intracellular mechanism in insulin release. The involvement of protein kinase C in the stimulus-
FIG. 1. Immunohistochemical localization of protein kinase C isozymes in rabbit pancreatic islet cells. (A) The strong immunoreactivity
with MC-3a is observed in the central portion of the islet. (B) The reactive products with MC-2a are localized in the periphery of the islet. No apparent staining is observed with MC-la. Bar = 25 nm.
1238
PROTEIN KINASE C ISOZYMES IN PANCREATIC ISLET CELLS
Endo • 1990 Vol 126 • No 2
FIG. 2. Immunohistochemical localization of the type III protein kinase & and pancreatic hormones in serial paraffin sections of rabbit pancreatic islet cells. A, Staining for insulin; C, for glucagon; E, for somatostatin; B, D, and F, for type III protein kinase C (MC-3a). Immunostaining with MC-3a exhibits a clear-cut distribution pattern identical to that of insulin-positive cells in the islets (A and B). No product reacting with MC-3a is observed in the region where glucagonpositive cells are located (arrowhead) (C and D). No reaction products with MC3a are observed in the region where somatostatin-positive cells are located (white arrow) (E and F). Bar = 25 ^m.
secretion coupling mechanism of glucagon release has been reported (31, 32). Recent evidence indicated that a protein kinase C-activated pathway was involved at the level of transcription in the regulation of glucagon gene expression in a rat islet cell line, and that expression of the insulin and somatostatin gene was not regulated (33). This can be interpreted to mean that a protein kinase C isozyme contained in pancreatic A cells is different from that in B or D cells. In our study, we found that A and B cells possessed different isozymes, but protein kinase C could not be detected in D cells. It remains unknown
whether or not protein kinase C is expressed in D cells. The immunoreactivity of the 80K protein on immunoblotting in insulinoma cells was detected only with MC-3a of our three monoclonal antibodies. This result indicates that the type III protein kinase C is predominantly expressed in these cells. The widespread occurrence of protein kinase C in various mammalian tissues was shown by Kuo et al. (34) was present in low levels in pancreas. Based on the enzyme-linked sandwich immunoassay, we investigated the contents of protein kinase C isozymes in several
PROTEIN KINASE C ISOZYMES IN PANCREATIC ISLET CELLS
1
1239
2 3 4 5
ffg 2OOK— 116K— 97K— •
8OK
66 K-
43K—
front—FIG. 4. Immunoblotting with MC-la, MC-2a, and MC-3a of whole homogenates of RINr-cells. Lanes: 1 {left) protein markers (Mr at left); 1 {right) whole homogenate (100 fig of protein) stained with Amido black; 2, whole homogenate reacted with MC-la; 3, whole homogenate reacted with MC-2a; 4, whole homogenate reacted with MC-3a; 5, whole homogenate reacted with normal mouse IgG. A single immunoreactive band of Mr 80,000 is detected only with MC-3a. No crossreactive protein is observed with MC-la, MC-2a, and normal mouse IgG.
TABLE 1. Levels of protein kinase C isozymes in RINr cells determined by the enzyme-linked sandwich immunoassay Type I II III 0 b
PlG. 3. Immunohistochemical localization of protein kinase C isozymes in rat pancreatic islet cells. This islets were stained with MC-3a (A), MC-2a (B), and MC-la. Bar = 50 Mm.
mamalian tissues (M. Hagiwara, T. Hachiya, M. Watanabe, N. Usuda, F. Iida, K. Tamai, H. Hidaka, unpublished data). Brain tissue possessed exceedingly high levels of three types of protein kinase C, and spleen and platelets contained high levels of the type II and III isozymes. In this study, significant amounts of the type III enzyme were detected in rat insulinoma cells, but the presence of the type I and II enzyme could not be unequivocally
Protein kinase C content (ng/mg protein)
Vol. 126, No. 2 Printed in U.S.A.
Different Expression of Protein Kinase C Isozymes in Pancreatic Islet Cells* KOJI ONODAt, MASATOSHI HAGIWARA, TAKAHISA HACHIYA, NOBUTERU USUDA, TETSUJI NAGATA, AND HIROYOSHI HIDAKAt Department of Pharmacology, Nagoya University School of Medicine (K.O., M.H., T.H., H.H.), Showa-ku, Nagoya, Aichi 466, and Department of Anatomy, Shinshu University School of Medicine (N.U., T.N.), Asahi, Matsumoto, Nagano 390, Japan
ABSTRACT. To elucidate differential roles of protein kinase C isozymes in pancreatic islet cells, the precise localization of the isozymes in rabbit and rat islet endocrine cells was investigated using monoclonal antibodies specific for three types of the enzyme. We detected strong immunoreactivity for the type III protein kinase C in B cells. Immunoreactivity for the type II enzyme was seen in A cells, and no apparent immunoreactivity for type I was observed in the islet cells. The expression of the type III protein kinase C in B cells was confirmed using rat
P
ROTEIN kinase C, Ca2+-activated phospholipid-dependent protein kinase, is believed to be involved in transmembrane signaling to regulate several cellular processes (1). This protein kinase is also known to be a Cellular receptor for the tumor-promoting phorbol esters which elicit various cellular responses including secretion, cell proliferation and differentiation, neurotransduction, and gene expression (2). In the pancreatic islet cells, protein kinase C has already been identified (3, 4). Recent progress in understanding the pathways involved in physiological insulin release indicates that the activation of protein kinase C in pancreatic B cells may be a pivotal event in the stimulus-secretion coupling mechanism (5). On the other hand, cDNA cloning and the sequence analysis of protein kinase C indicated that the enzyme was a family of genes (6-11). Brain protein kinase C was further resolved into three peaks, type I, type II and type III, chromatographically on a hydroxylapatite column (12). The identification of the protein kinase C isozymes in pancreatic B cells is required to elucidate the precise physiological role of the kinase in insulin secretion. HowReceived September 18, 1989. T h i s work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture, Japan. fPresent address: Department of Thoracic Surgery, Mie University School of Medicine, Edobashi, Tsu 514, Japan. JTo whom all correspondence and reprint requests should be addressed.
insulinoma cells. The predominant expression of the type III enzyme in these cells was shown by immunoblotting. Moreover, on the basis of an enzyme-linked sandwich immunoassay, the levels of protein kinase C isozymes were determined in these cells. The significant amounts of the type III enzyme was detected, but the contents of the type I and II enzyme were under detectable level. These results suggest that the type III protein kinase C is involved in the regulation of insulin release in pancreatic B cells. {Endocrinology 126: 1235-1240, 1990)
ever, direct biochemical demonstration of the isozymes in isolated pancreatic B cells has been technically difficult. More recently, we generated monoclonal antibodies, MC-la, MC-2a, and MC-3a, which specifically recognized three types of rabbit brain protein kinase C, respectively (13,14), and we demonstrated that the expression of protein kinase C isozymes in the rabbit cerebellum, spleen and human platelet was dependent on the cell type (13, 15). Using these monoclonal antibodies, we have showed a different expression of protein kinase C isozymes in the rabbit or rat pancreatic islet cells, and demonstrated the presence of an isozyme in a continuous cell line, RINr, which secretes insulin (16). Our results suggest that protein kinase C isozymes play differential roles in the regulation of hormone secretion in the pancreatic islet cells. Materials and Methods Immunohistochemical staining of pancreas Male adult rabbit or rat was anesthetized with an iv injection of pentobarbiturate (Nembutal) at 30 mg/kg. The pancreas was removed, trimmed of excess fat, and cut into small pieces. The tissue was immersed in phosphate-buffered paraformaldehyde with picric acid (17) for 6 h at 4 C, and washed with PBS containing 10%, 15%, or 20% sucrose for 12 h, respectively. The tissue frozen in liquid nitrogen was sectioned at 6 nm, using a cryostat, and placed on slide glass. The sections were 1235
1236
PROTEIN KINASE C ISOZYMES IN PANCREATIC ISLET CELLS
exposed to 10% normal horse serum in PBS for 20 min. The sections were then incubated with monoclonal antibodies (hybridoma supernatant), MC-la, -2a, and -3a, diluted from 1:10 to 1:100 in PBS for 12 h at 4 C. After being washed in PBS, they were processed for avidin-biotinylated peroxidase complex staining. Reaction product was developed by incubation in 0.03% (wt/vol) 3,3'-diaminobenzidine tetrachloride and 0.005% peroxidase in 50 mM Tris buffer, pH 7.5, and these sections were counterstained with 1% methyl green solution. In order to compare the localization of the type III protein kinase C with those of pancreatic hormones, serial paraffin sections were stained by the peroxidase antiperoxidase (PAP) method of Sternberger et al. (18), using MC-3a (anti-type III protein kinase C antibody) and antipancreatic hormones antibodies: anti-insulin, antiglucagon, and antisomatostatin (DAKO Corporation, Copenhagen, Denmark). Immunoblot analysis of protein kinase C isozymes in RINr cells. RINr cells were cultured in RPMI1640 medium (GIBCO, Grand Island, NY) supplemented with 10% fetal calf serum (GIBCO) and antibiotics (16). The whole homogenates of RINr cells (2 X 106) were electrophoresed in sodium dodecyl sulfate (SDS)-polyacrylamide gel (10%) and then transferred to nitrocellulose membrane (Bio-Rad Laboratories, Richmond, CA). The membranes were incubated with 2% normal horse serum in TPBS (PBS containing 0.05% Tween 20) to block nonspecific binding sites. The membranes were then exposed to respective antibody solution (100-fold diluted ascitic fluid) prepared in TPBS for 60 min. After being washed in TPBS, they were incubated in biotinylated horse antimouse immunoglobulin G (IgG) followed by the avidin-biotin complex of the Vectastain kit (Vecter Laboratory, Burlingame, CA). Immunoreactive proteins were visualized by reaction with 0.06% (wt/ vol) 4-chloro-l- naphthol and 0.02% H2O2. Sandwich enzyme immunoassay Three monoclonal antibodies, MC-la, -2a, and -3a, and polyclonal antibody against protein kinase C, PC-40 recognized the regulatory domain (33 or 35K) and the catalytic domain (48K), respectively, resulting from the trypsin-digested protein kinase C. Using these antibodies, the levels of protein kinase C isozymes, type I, II, and III in RINr-cells were determined by the enzyme-linked sandwich immunoassay. 0.5 ng of monoclonal antibodies in 100 /A of buffer A; 0.1 M sodium phosphate (pH 7.4) were adsorbed to each 96-well polystyrene flat-bottom well at 4 C overnight. Wells were washed thoroughly. Washing in this and subsequent steps was carried out after each step by three changes of 350 /xl of TPBS. Each of the monoclonal antibody-coated microtiter plate wells was incubated with 350 n\ of blocking buffer (1% BSA in buffer A) at 4 C overnight. Aliquots of 100 /xl of sample solutions and the solutions containing various amounts of purified protein kinase C isozymes for the standard curve were added to each well, and were left for 2 h at 4 C. Aliquots of 100 /A of PC-40 conjugated with horseradish peroxidase (100 ng in 100 fA PBS) were added to each well and the preparations further incubated for 2 h at 4 C. The enzyme reaction was initiated by the addition of 100 nl of 0.1 M citrate-phosphate buffer (pH 5.1) containing 0.01% H2O2 and 1 mg/ml o-phenyl-enediamine. After 10-min incuba-
Endo • 1990 Vol 126 • No 2
tion at room temperature in the dark, 100 /xl of 2 N sulfuric acid were added to stop the reaction. The optical density (OD) was measured at 492 nm by photometer. The contents of protein kinase C isozymes of sample solutions were calculated by using the respective standard curve of purified protein kinase C isozymes from rabbit brain. Results To investigate expression of protein kinase C isozymes in the rabbit pancreatic islet cells, immunohistochemical staining was undertaken, using monoclonal antibodies (MC-la, -2a, and -3a) which selectively interacted with hydroxylapatite column chromatographically resolved isozymes type I, II, and III of protein kinase C, respectively. Strong immunoreactivity with MC-3a was recognized in the cells of the central portion of the islets (Fig. 1A). Cells immunohistochemically reactive to MC-2a were observed in the peripheral region of the islets where glucagon-positive cells were located (Fig. IB). In all positive-staining cells the immunoreactivity could be observed in the cytoplasms but not in the nuclei. No apparent immunoreactivity was observed with MC-la using various dilutions of antibody solutions (Fig. 1C). By serial sections the localizations of pancreatic hormones and the type III protein kinase C were compared. Immunostaining with MC-3a exhibited a clear cut distribution pattern exactly identical to that of insulinpositive cells in the islets (Fig. 2, A and B). No product reacting with MC-3a was observed in the region where glucagon-and somatostatin-positive cells were located (Fig. 2, C-F). Moreover, we examined the localization of protein kinase C isozymes in the rat pancreatic islets. The staining patterns obtained with three monoclonal antibodies in the rat islets coincided with those in the rabbit islets (Fig. 3). When each antibody was reacted with an excess amount of purified brain protein kinase C before incubation, no immunoreactive products were observed (data not shown). In addition, no staining was observed in preparations in which one of the reagents in the staining protocol was deleted or with hybridoma supernatant fluids containing irrelevant antibody. Furthermore, to confirm the immunohistochemical localization of the type III protein kinase in pancreatic B cells, we used a continuous cell line, RINr, that secreted insulin, and the presence of glucagon in the supernatant fluid of the cell line could not be detectable (16). This cell line was originally established from a serially transplantable, radiation-induced, rat islet cell tumor. The whole homogenates of RINr cells were immunoblotted with our monoclonal antibodies (Fig. 4). A single major immunoreactive species of Mr 80,000 was detected with MC-3a and no other immunoreactive proteins were
PROTEIN KINASE C ISOZYMES IN PANCREATIC ISLET CELLS
1237
observed in whole homogenates of the cells. No crossreacting protein was observed in the cells with MC-la and MC-2a. Moreover, no reactive protein was observed with normal mouse IgG solutions. The levels of protein kinase C isozymes were determined by the sandwich enzyme immunoassay (Table 1). RINr cells contained 69.3 ± 8.6 ng/mg protein of the type III enzyme whereas the type I and II enzyme were under detectable level (4.1 ng and 0.65 ng/mg protein, respectively). These results indicate that the type III protein kinase C is predominantly expressed in this insulin-producing cell line.
Discussion In the present study, we at the first time demonstrated specific localization of protein kinase C isozymes in rabbit and rat pancreatic islet cells, by use of immunohistochemical techniques at the light microscopic level: A cells showed immunoreactivity for type II of the isozymes, and B cells showed that for type III. Moreover, evidence of the existence of the type III enzyme in insulin-secreting cells was confirmed by using insulinoma cells. Several reports have documented that protein kinase C is present in rat pancreatic islets (3, 4), rat insulinoma (19), and hamster clonal B cells (20). Inhibition of the insulin secretory response to glucose by polymyxin B, which inhibited protein kinase C-stimulated phosphorylation, indicated that protein kinase C played a role in glucose-induced insulin secretion (21). Zawalich et a/.(22) proposed that the initial phase of insulin secretion was due to calmodulin-dependent processes and sustained phase was controlled by a protein kinase C pathway. On the other hand, however, activation of Ca2+-calmodulin and cAMP-dependent protein kinase has been suggested to be involved in stimulus-secretion coupling in the pancreatic B cell (23-29). Schubart et al. (30) demonstrated that cAMP-mediated insulin release was not affected by trifluoperazine which inhibited cyclic nucleotide phosphodiesterase by binding to calmodulin, and suggested that cAMP and Ca2+ regulated the secretory process by distinct mechanisms. It is widely accepted that these three protein kinases may be involved in the regulation of glucose-induced insulin secretion from the pancreatic B cells, while its detailed intracellular mechanism is not clear. Therefore, our evidence that B cells contain the type III protein kinase C may throw some lights upon the elucidation of the complex intracellular mechanism in insulin release. The involvement of protein kinase C in the stimulus-
FIG. 1. Immunohistochemical localization of protein kinase C isozymes in rabbit pancreatic islet cells. (A) The strong immunoreactivity
with MC-3a is observed in the central portion of the islet. (B) The reactive products with MC-2a are localized in the periphery of the islet. No apparent staining is observed with MC-la. Bar = 25 nm.
1238
PROTEIN KINASE C ISOZYMES IN PANCREATIC ISLET CELLS
Endo • 1990 Vol 126 • No 2
FIG. 2. Immunohistochemical localization of the type III protein kinase & and pancreatic hormones in serial paraffin sections of rabbit pancreatic islet cells. A, Staining for insulin; C, for glucagon; E, for somatostatin; B, D, and F, for type III protein kinase C (MC-3a). Immunostaining with MC-3a exhibits a clear-cut distribution pattern identical to that of insulin-positive cells in the islets (A and B). No product reacting with MC-3a is observed in the region where glucagonpositive cells are located (arrowhead) (C and D). No reaction products with MC3a are observed in the region where somatostatin-positive cells are located (white arrow) (E and F). Bar = 25 ^m.
secretion coupling mechanism of glucagon release has been reported (31, 32). Recent evidence indicated that a protein kinase C-activated pathway was involved at the level of transcription in the regulation of glucagon gene expression in a rat islet cell line, and that expression of the insulin and somatostatin gene was not regulated (33). This can be interpreted to mean that a protein kinase C isozyme contained in pancreatic A cells is different from that in B or D cells. In our study, we found that A and B cells possessed different isozymes, but protein kinase C could not be detected in D cells. It remains unknown
whether or not protein kinase C is expressed in D cells. The immunoreactivity of the 80K protein on immunoblotting in insulinoma cells was detected only with MC-3a of our three monoclonal antibodies. This result indicates that the type III protein kinase C is predominantly expressed in these cells. The widespread occurrence of protein kinase C in various mammalian tissues was shown by Kuo et al. (34) was present in low levels in pancreas. Based on the enzyme-linked sandwich immunoassay, we investigated the contents of protein kinase C isozymes in several
PROTEIN KINASE C ISOZYMES IN PANCREATIC ISLET CELLS
1
1239
2 3 4 5
ffg 2OOK— 116K— 97K— •
8OK
66 K-
43K—
front—FIG. 4. Immunoblotting with MC-la, MC-2a, and MC-3a of whole homogenates of RINr-cells. Lanes: 1 {left) protein markers (Mr at left); 1 {right) whole homogenate (100 fig of protein) stained with Amido black; 2, whole homogenate reacted with MC-la; 3, whole homogenate reacted with MC-2a; 4, whole homogenate reacted with MC-3a; 5, whole homogenate reacted with normal mouse IgG. A single immunoreactive band of Mr 80,000 is detected only with MC-3a. No crossreactive protein is observed with MC-la, MC-2a, and normal mouse IgG.
TABLE 1. Levels of protein kinase C isozymes in RINr cells determined by the enzyme-linked sandwich immunoassay Type I II III 0 b
PlG. 3. Immunohistochemical localization of protein kinase C isozymes in rat pancreatic islet cells. This islets were stained with MC-3a (A), MC-2a (B), and MC-la. Bar = 50 Mm.
mamalian tissues (M. Hagiwara, T. Hachiya, M. Watanabe, N. Usuda, F. Iida, K. Tamai, H. Hidaka, unpublished data). Brain tissue possessed exceedingly high levels of three types of protein kinase C, and spleen and platelets contained high levels of the type II and III isozymes. In this study, significant amounts of the type III enzyme were detected in rat insulinoma cells, but the presence of the type I and II enzyme could not be unequivocally
Protein kinase C content (ng/mg protein)
