Insulin, Glucagon and Somatostatin Content of the Non-Obese Diabetic (NOD) Mouse Pancreas and Plasma Virus Antibodies to Coxsackie B- and Reoviruses

Transcription

1 Tohoku J. exp. Med., 1985, 146, Insulin, Glucagon and Somatostatin Content of the Non-Obese Diabetic (NOD) Mouse Pancreas and Plasma Virus Antibodies to Coxsackie B- and Reoviruses YOSHIMASA TASAKA, YUKIKO INOUE, KOJI MARUMO, YOSHIATSU MIZUNO, YUKIMASA IIIRATA and YORIYUKI AKAO * Tokyo Women's Medical College, Diabetes Center, Tokyo 102 and * National Institute of Health of Japan, Central Virus Diagnostic Laboratory, Gakuen, Musashimurayama-shi, Tokyo TASAKA, Y., INOUE, Y., MARUMO, K., HIRATA, T. and AKAO, Y. Insulin, Glucagon and Somatostatin Content of the Non-Obese Diabetic (NOD) Mouse Pancreas and Plasma Virus Antibodies to Coxsackie B- and Reoviruses. Tohoku J. exp. Med., 1985, 146 (2), The amounts of insulin, glucagon and somatostatin in the pancreas of NOD mice were determined and the results were compared with those of normal ICR-strain mice, and plasma antibodies to Coxsackie B-3 and reovirus types 1, 2 and 3 were measured. In the pancreas of NOD mice with fasting plasma glucose (FPG) less than 140 mg/100 ml, the insulin content of the male mice was similar to that of the normal controls, ranging to U/g wet weight of pancreas, but it was already significantly decreased to 0.85±0.52 U/g in the female mice. In the NOD mice with FPG more than 201 mg 100 ml, it was U/g. The glucagon content of the pancreas was 7.76±0.89,ug/g in the normal controls and it was decreased slightly in the NOD mice, but the values among the NOD mice were not significantly different. Pancreas somatostatin showed a tendency to be higher in the NOD mice with FPG more than 201 mg/ 100 ml. Histologically cell infiltration into the pancreatic islets was conspicious in the hyperglycemic NOD mice, but it was found even in the normoglycemic NOD. Plasma antibodies to Coxsackie B-3 Virus and reovirus types 1, 2 and 3 were not detected at any stage of FPG. NOD mouse ; pancreas insulin ; glucagon ; somatostatin ; plasma virus antibody The non-obese diabetic (NOD) mouse is an inbred strain established in 1980 by Makino et al. in the Shionogi Research Laboratories (Shionogi & Co., Ltd., Osaka). Two sublines were separated from the CTS strain which was derived from the Jcl-ICR mice characterized by cataracts. A female mouse which happened to have polyuria, loss of body weight and strongly positive urinary Received July 28, 1984; accepted for publication, January 24,

2 154 Y. Tasaka et al. sugar died after one month, and this mouse was proved to be insulin-dependent diabetes and have pathological findings of insulitis in the pancreas. Makino et al. (1980) succeeded in making specific pathogenic free (SPF) in the isolator and established the NOD strain as a new diabetic one. Till now several investigations on NOD mouse have been reported (Fushimi et al. 1980; Makino et al. 1981; Matsushima et al. 1982; Kataoka et al. 1983). In this study, the amounts of insulin, glucagon and somatostatin in the pancreas of NOD mice were determined, and the values of virus antibody to Coxsackie B-3 and reovirus types 1, 2 and 3 were measured in NOD mouse plasma. MATERIALS AND METHODS NOD mice (30 females and 13 males, mean age 7.8±0.9 month, mean ± s.e.) were supplied from the Shionogi Research Laboratories and 8 female mice of ICR-strain were from Clea Japan (Tokyo) as control. They were fed ad libitum till experiment. The overt diabetic mice after the appearance of urinary sugar were kept in the poor diabetic state without insulin treatment. The mean living days till sacrifice after the appearance of overt diabetes were 14 days. The mice were anesthetized with ether after fasting for 24 hr and the blood was taken to the heparinized tube by cardiac puncture and centrifuged. Plasma was frozen at -20 C until the determination of fasting plasma glucose (FPG) or virus antibody titer. After exsanguination, the pancreas was excised, minced and extracted with acid ethanol (Hayashi et al. 1977; Tasaka et al. 1979), and immunoreactive insulin (IRI) or glucagon (GI) in the extract was determined, and their contents in the pancreas were calculated. For the determination of somatostatin in the pancreas, pieces of the pancreas were homogenized and extracted with 0.1 N acetic acid and after lyophilization, the extract was dissolved in 0.1 M phosphate buffer and determined immunologically at Special Reference Laboratory (Tokyo). Recovery was 84±3% and the minimum assay value was 3 pg/ml. C.V. (coefficient of variation) of intraassay value was 14.7% at a high concentration (29.6 pg/ml, N = 10), 17.0% at a middle one (19.2 pg/ml) and 14.1% at a low concentration (7.8 pg/ml). Interassay C.V. was 12.7% at a high concentration (53.6 pg/ml, N=5),18.9% at a medium concentration (21.5 pg/ml) and 27.2% at a low concentration (3.7 pg/ml). Plasma glucose was determined by an autoanalyser using the glucose oxidase method. IRI in the extract of the pancreas was determined by the two-antibody immunoassay (Morgan and Lazarow 1963) using rat standard insulin. Antiglucagon serum 30 K for the determination of GI was obtained from Hoechst Japan (Tokyo). GI in the extract was determined by the polyethylene glycol method (Henquin et al. 1974). Bouin-fixed, decerated paraffin section of the pancreas was used for the histological studies. The plasma antibodies to Coxsackie B-virus (Inoue 1978) and reovirus types 1, 2 and 3 (Akao 1978) were determined, respectively. Statistical analysis was performed by Student's t-test and probability of less than 5% was considered to be significant. RESULTS Fasting plasma glucose and insulin content of the pancreas The insulin content of the pancreas of the NOD mice was determined in relation to their FPG levels (Fig. 1). Pancreatic IRI in the female NOD mice with FPG less than 140 mg/ 100 ml was 1.77±0.56 U/g wet weight, which was significantly lower (p <0.05) than those in the control female ICR-strain

3 IRI, GI and IRS in NOD Mouse Pancreas and Virus Antibody 155 mice( U/g wet weight of pancreas) or the male NOD mice with FPG less than 140 mg/ 100 ml (3.6±0.6 U/g wet weight). The decrease in IRI was more marked in the group with FPG between 141 and 200 mg/100 ml or more than 201 mg/ 100 ml ; each value 0.117± 0.08 U/g or 0.002±0.001 U/g wet weight close to zero, respectively. Fasting plasma glucose and glucagon and somatostatin contents in the pancreas GI contents of the pancreas of the NOD mice were determined in relation to their FPG levels. GI contents in the NOD mice with FPG less than 140 mg/100 ml in either sex were significantly lower than those in the normal control mice (7.76± 0.89,u g/g wet weight), but there was no significant difference in the GI contents among the NOD mice with different FPG levels (Fig. 2). Somatostatin contents of the pancreas of of the NOD mice were 148± 41 ng/g wet weight in the Fig.1. Fasting plasma glucose and insulin content in the pancreas of NOD and control mouse. In inverse proportion to FPG, pancreatic IRI decreased. Even in NOD mouse with FPG less than 140 mg 100 ml, pancreatic IRI already decreased significantly. *p<0.01,~p<0.05

4 156 Y. Tasaka et al. Fig. 2. Fasting plasma glucose and glucagon content in the pancreas of NOD and control mouse. In NOD mouse with FPG less than 140 mg/100 ml, pancreatic GI already decreased significantly. * p < 0.01, t p < 0.05 Fig. 3. Fasting plasma glucose and somatostatin content in pancreas of NOD mouse. The pancreas tissue was extracted with 0.1 N acetic acid and after lyophilization, the extract was dissolved in 0.1 M phosphate buffer and determined immunologically. group with FPG higher than 201 mg/ 100 ml, and 102±32 ng/g wet weight in those with FPG between 141 and 200 mg 100 ml and there was no significant difference in pancreatic somatostatin contents among them (Fig. 3). Histological findings of the pancreas and its insulin content in NOD mice Even in some of the NOD mice which had neither urinary sugar, nor overt diabetic symptoms, the insulin content of the pancreas was low. Therefore, the relation between the histological findings of the pancreas and the insulin content

5 IRI, GI and IRS in NOD Mouse Pancreas and Virus Antibody Fig. 4. (upper) Outstanding small round cell infiltration in pancreatic islet of NOD mouse (hematoxylin-eosin staining, x65). The infiltration is marked at peripheral region of islets, and most of islet cells seems to be intact. Fig. 5. (lower) Moderate small round cell infiltration in pancreatic islet of NOD mouse (hematoxylin-eosin region of islets. staining, x 65). The infiltration is marked at peripheral 157

6 158 Y. Tasaka et al. TABLE 1. Values us type of virus B3 were antibodies to Reovirus types investigated. Values mean 1 to 3 and Coxsackievir- dilution titers of the pancreas was investigated in 4 NOD mice. Fig. 4 shows the histological findings of the pancreas in a female NOD mouse which had strongly positive urinary sugar and FPG 366 mg 100 ml. Marked infiltrations of small round cells were found in pancreatic islets. The insulin content of the pancreas was U/ g wet weight. The second case was the female one which had normal FPG (89 mg 100 ml) and no urinary sugar (Fig. 5). In spite of normal FPG level, mild

7 IRI, GI and IRS in NOD Mouse Pancreas and Virus Antibody small cell infiltration in the islet was already seen and the insulin already decreased to 1.21 U/g wet weight. content 159 was Plasma virus antibody in NOD mice Plasma antibodies to Coxsackie virus type B-3 and reovirus types 1, 2 and 3 were determined in the NOD mice. No groups of the NOD mouse with FPG from less than 100 mg/ 100 ml to more than 201 mg/ 100 ml had increased titer of virus antibody associated with the development of diabetes mellitus (Table 1). Some plasma samples were diluted twice due to sample deficiency. DISCUSSION As shown in this investigation, the insulin content of the pancreas of the female NOD mice was already decreased at the early stage when they had no urinary sugar or diabetic symptoms. There was a clear reciprocal relation between this reduction and the increase in FPG value, and the insulin content in the pancreas was close to zero with FPG more than 201 mg/100 ml, suggesting typical insulin-deficient diabetes. These findings are similar to those in streptozotocin diabetes, human insulin-dependent diabetes or Sekoke disease of carp (Yokote 1970). It is easily understood why the amount of pancreas insulin is already decreased in the NOD mice before the onset of diabetic symptoms if we observe lymphocytic infiltration in the pancreatic islets at an early stage in the NOD mice. Lymphocytic infiltration in the pancreas of the NOD mice has been found in 80% of females and 60% of males five or six weeks after birth (Tochino 1982), showing the pathological findings of insulitis. In this study, although the insulin content of the pancreas of the male NOD mice with FPG less than 140 mg/ 100 ml was not different statistically from that in the normal controls, some morphological changes would already be present. In the NOD mice, loss of body weight and ketosis appear immediately after the onset of overt diabetes, and they die within one or two weeks without insulin treatment. Therefore abnormal glucose intolerance would appear suddenly at some stage together with the striking decrease in pancreas insulin. The glucagon content of the pancreas was already decreased in both the male and the female NOD mice with FPG less than 140 mg 100 ml and showed no further decrease even in those with higher FPG levels. Histologically the infiltration of small round cells into the pancreatic islets causes damage only to B cells of the islets initially and other endocrine cells are intact, but soon A, D, or F cells form lines as if they follow the initial lymphocytic infiltration in the pancreatic islet (Fujita et al. 1982). Eventually all the endocrine cells including A cells in the pancreatic islets would be decreased. The reason why the glucagon content of the pancreas was low in the NOD mice from the beginning of the observations is unknown. One possibility is that pancreatic A cells might be more heavily infiltrated with lymphocytes than B cells.

8 160 Y. Tasaka et al. Hyperglucagonemia and high glucagon content in the pancreas are found in streptozotocin diabetic rats together with a marked decrease in insulin in the pancreas (unpublished observations) and no significant increase in glucagon content in the pancreas of alloxan diabetic dogs was reported (Tasaka et al. 1984). The histological findings in the pancreas of NOD mice would be different from those in animals with such experimental diabetes in that the destruction of the pancreatic islet is not limited to B cells. There are two different reports on the somatostatin content of the pancreas of diabetic mice ; one states that it increased (Dolais-Kitabgi et al. 1979) and the other that it decreased (Makino et al. 1979). Berelowitz et al. (1980) reported that amount of somatostatin in the pancreas is low until 18 weeks after birth in ob-ob mouse and 10 weeks in db-db mouse, thus suggesting the inportance of sampling time. In this study. NOD mice with FPG more than 200 mg/ 100 ml had a tendency toward a high somatostatin content in the pancreas, and Matsushima et al. (1982) reported an increased level of somatostatin in the pancreas. Several viruses have been reported to be implicated in the etiology of diabetes. Coxsackie B virus (Pappenheimer et al. 1951; Burch et al. 1953), encephalomyocarditis virus (Craighead and Steinke 1971) and reovirus (Stanley et al. 1953) play an important role in the etiology of diabetes mellitus. Onodera et al. (1981) induced transient diabetes in mice by infecting them with reovirus I and found virus particles in A, B and D cells of the pancreatic islet. In the serum of these animals, autoantibodies to cytoplasmic antigen of pancreatic islets, to the anterior lobe of the hypophysis and to the gastric mucous membrane have been found, thus showing that they have polyendocrinopathy. In NOD mice lymphocytic infiltration has been reported not only in the pancreatic islets but in the salivary gland, ovarium and thyroid gland, and some autoimmune mechanism could play an important role in the development of diabetes mellitus. Although we cannot completely deny the participation of virus infection in the pathogenesis of diabetes in the NOD mouse, no NOD mice had antibodies to reovirus type 1, 2 or 3, or Coxsackie virus type B-3. Recently Fujita et al. (1984) reported that retrovirus was found in the B cells in the pancreatic islets of NOD mice. Whether this retrovirus really induces diabetes mellitus or not has to await further investigation. The B B/W rat is likewise a model of spontaneous autoimmune diabetes mellitus and in this rat, no virus antibodies including those to pneumonia virus, reovirus types 3, encephalomylitis virus and Sendai virus were found. The exact pathogenesis of the development of diabetes mellitus in the NOD mouse is unknown, but an autoimmune mechanism based on some chromosome abnormality has been proposed. The heavy lymphocyte infiltration in the pancreatic islets and other organs would induce various abnormalities in the NOD mouse and these facts would assume a pathophysiology different from those of experimental diabetes.

9 IRI, GI and IRS in NOD Mouse Pancreas and Virus Antibody 161 Acknowledgments We would like to express our thanks to Dr. Y. Tochino in the Shionogi Research Laboratories, Shionogi & CO., Ltd., Osaka for supplying the NOD mice, Mrs. T. Tamura for her technical assistance and to Mrs. M. Tasaka for preparing the manuscript. References 1) Akao, Y. (1978) Reovirus. In : Biseibutsu Kensa Hikkei, Virus and Riekettia Examination, 2nd ed., edited by K. Yanagisawa, Nihon Koshu Eisei Kyokai, Tokyo, pp (Japanese) 2) Berelowitz,,, M., Coleman, DL. & Frohman, L.A. (1980) Temporal relationship of tissue somatostatinlike immunoreactivity to metabolic changes in genetically obese and diabetic mice. Diabetes, 29, ) Burch, G.E., Ching-Ya, T. & Harb, J.M. (1953) Pathological findings in the pancreas of mice injected with Coxsackie virus B 4. Arch. intern. Med., 128, ) Craighead, J.E. & Steinke, J. (1971) Diabetes mellitus-like syndrome in mice infected with encephalomyocarditis virus. Amer. J. Path., 63, ) Dolais-Kitabgi, I., Marchand-Burstel, Y.L. & Freychet, P. (1979) Somatostatin in the pancreas and hypothalamus of obese mice. Diabetologia, 17, ) Fujita, T., Yui, R., Kusumoto, Y., Serizawa, Y., Makino, S. & Tochino, Y. (1982) Lymphocytic insulitis in a non-obese diabetic (NOD) strain of mice : an immunohistochemical arid electron microscope investigation. Biomed. Res., 3, ) Fujita, H., Fujino, H., Nonaka, K., Tarui, S. & Tochino, Y. (1984) Retrovirus-like particles in pancreatic B-cells of NOD (non-obese diabetic) mice. Biomed. Res., 5, ) Fushimi, H., Nonaka, K., Tarui, S., Tochino, Y. & Kanaya, H. (1980) The effects of parabiosis on serum and kidney glucosidase activities in spontaneously diabetic mice. Diabetologia, 19, ) Hayashi, M., Floyd, J.C. Jr., Pek, S. & Fajans, S.S. (1977) Insulin, proinsulin, glucagon and gastrin in pancreatic tumors and in plasma of patients with organic hyperinsulinism. J. din. Endocr. Metab, 44, ) Henquin, J.C., Malvaux, P. & Lambert, A.E. (1974) Glucagon immunoassay using polyethylene glycol to precipitate antibody-bound hormone. Diabetologia, 10, ) Inoue, S. (1978) Technique of serum reaction necessary for virus and rickettia examination. In : Biseibutsu Kensa Hikkei, Virus and Rickettia Examination, 2nd ed., edited by K. Yanagisawa, Nihon Koshu Eisei Kyokai, Tokyo, pp (Japanese) 12) Kataoka, S., Satoh, J., Fujita, H., Toyota, T., Suzuki, R., Itoh, K. & Kumagai, K. (1983) Immunologic aspects of the nonobese diabetic (NOD) mouse. Abnormalities of cellular immunity. Diabetes, 32, ) Makino, H.. Matsushima, Y., Kanatsuka, A., Yamamoto, M., Kumagai, A. & Nishimura, M. (1979) Changes in pancreatic somatostatin content in spontaneously diabetic mice, as determined by radioimmunoassay and immunohistochemical methods. Endocrinology, 104, ) Makino, S., Kunimoto, K., Muraoka, Y., Mizushima, Y., Katagiri, K. & Tochino, Y. (1980) Breeding of a non-obese, diabetic strain of mice. Exp. Anim., 29, ) Makino, S., Kunimoto, K. & Muraoka, Y. (1981) Effect of castration on the appearance of diabetes in NOD mouse. Exp. Anim., 30, ) Matsushima, Y., Makino, Y., Kanatsuka, A., Osegawa, M., Kumagai, A., Nishimura, M., Tochino, Y. & Makino, S. (1982) Pancreatic somatostatin contents in spontaneously diabetic K K and non-obese diabetic (NOD) mice. Horm. Metab. Res., 14,

10 162 Y. Tasaka et al. 17) Morgan, CR. & Lazarow, A.E.. (1963) Immunoassay of insulin : two antibody system. Diabetes, 12, ) Onodera, T., Toniolo, A., Ray, U.R., Jensen, A.B., Knazek, R.A. & Notkins, A.L. (1981) Virus-induced diabetes mellitus. XX. Polyendocrinopathy and autoimmunity. J. exp. Med., 153, ) Pappenheimer, A.M., Kunz, L.J. & Richardson, S. (1951) Passage of Coxsackie virus (connecticut 5 strain) in adult mice with producing of pancreatic disease. J. exp. Med., 94, ) Stanley, N.F., Dorrman, D.C. & Ponsford, J. (1953) Antibodies to Coxsackie viruses in pooled human serum. Aust. J. exp. biol. Sci., 31, ) Tasaka, Y., Takei, M. & Hirata, Y. (1979) C-peptide, insulin and proinsulinlike component in diabetic and nondiabetic human pancreas. Endocr. jap., 26, ) Tasaka, Y., Inoue, S., Marumo, K. & Hirata, Y. (1984) Plasma responses of pancreatic polypeptide, glucagon and insulin in normal and alloxan diabetic dogs and their regional levels in the pancreas. Acta endocr., 105, ) Tochino, Y. (1982) Problem and significance as experimental animal model of the nonobese diabetic mouse found in ICR strain. In: Development and Utilization of Spontaneous Animal Model, edited by H. Matsushita, Seishi Shoin, Tokyo, pp (Japanese) 24) Yokote, M. (1970) Sekoke disease spontaneous diabetes in carp found in fish farms. Responses to mammalian insulin. Bull. Jap. Sci. Fish., 36,