In order to further assess the glucoreceptor mechanism involved in insulin release and its character, we have studied the dynamics of insulin release from the rat pancreas perfused with a 6.2 mM K⁺and K⁺-depleted media containing α-and β-methyl-D-glucopyranosides (α-MDG and β-MDG), which are homogeneous in solution in contrast to D-glucose. Moreover, we compared them with those induced by glucose, glucosamine and galactose. The results are as follows.
1. Perfusion in a 6.2 mM K⁺ medium.α-and α-MDG (15.5 mM) did not induce insulin release, either alone or in combination with theophylline (10 mM). Insulin response to a combined infusion of glucose (8.3 mM) and a-MDG (15.5 mM) was a little higher than that either to glucose alone or to a combined infusion of glucose and β-MDG. The dose response relationship between the amounts of insulin released and the glucose concentrations was sigmoid with a Km of 8-9 mM glucose.
2. Perfusion in a K⁺-depleted medium. Under K⁺ deprivation as well as in a 6.2 mM K+ medium, Ca⁺⁺ was required for insulin release by glucose, α-MDG, β-MDG, glucosamine and galactose. Glucose did not induce insulin release at 1.4 mM, but triggered a biphasic insulin release at concentrations ranging from 2.8 mM to 16.7 mM. The amount of insulin release was much higher at each glucose concentration than that in a 6.2 mM K⁺ medium. The dose response curve was sigmoid with a Km of 4-5 mM glucose.α-MDG evoked a burst of insulin release at 2.6 mM, and the ratios of the average rate of insulin release and the maximal value in insulin releasingprofile by α-MDG to those by 2.8 mM glucose were 0.96 and 1.1, respectively. The dynamics of α-MDG-induced insulin release were quite similar to those induced by glucose. 2.6 mM β-MDG did not induce insulin release. The average rate of insulin release and the maximal value were 74 ± 6 μU/ml/min and 120±14 pUiml for 2.8 mM glucosamine, and 37±8 pU/ml/min and 53±14 pU/m1 for 2.8 mM galactose. The dynamics of glucosamine-induced insulin release were similar to those of glucose-induced insulin release in which the average rate of insulin release and the maximal value were 91±10 pUjmljmin and 158±8 μU/m1 respectively. Those of galactose-induced insulin release were less than those caused by glucose.
In addition to the fact that 5.7 mM L-arginine evoked a burst of insulin release in the presence of 5.2 mM α-MDG whereas arginine did not induce insulin release in the presence of β-MDG in a 6.2 mM K⁺ medium, the results suggest that the B cells may contain glucoreceptors on the plasma membrane directly controlling insulin release, and moreover, that α-MDG could interact with the glucoreceptors in a 6.2 mM K⁺ medium without any insulinogenic action and then could induce insulin release in a K⁺-depleted medium through a conformational change of the glucoreceptor and depolarization on the B cell plasma membrane.

View full abstract

In order to characterize the glucoreceptor mechanism involved in insulin release and the character of the release, we have studied dynamics of insulin release from the rat pancreas perfused with excess K⁺ media containing glucose, α-or β-methyl-D-glucopyranoside (α-MDG or βMDG). The results are as follows.
1. Perfusion in raising the extracellular K⁺ concentration from 6.2 to 24.8 mM
K⁺ (24.8 mM) provoked a transient enhancement of insulin release. The average rate of insulin release during the first 6-min period (the rate of insuln release) and the maximal value were 115 ± 11, μU/ml/min and 191 ± 17 μU/ml. respectively. The rate of insulin release by 2.8 mM glucose was 1.4 times as largeas that of the control. Insulin releasing profile due to 15.5 mM α-MDG was similar to that induced by 2.8 mM glucose, while insulin response dul to 15.5 mM β-MDG was below that of the control. Therefore, α-MDG specifically enhanced insulin release in a 24.8 mM K⁺ medium.
2. Perfusion in raising the extracellular K concentration from 6.2 to 12.4 mM
K⁺ (12.4 mM) did not enhance insulin release, and the rate of insulin release was only 22 ±8 μU/ml/min. The rate of insulin release and the maximal value caused by 2.8 mM glucose, which were indicated as 49 ± 10 μU/ml/min and 64 ± 22 μU/ml respectively, were significantly higher than those of the control experiment, but were lower than those by 2.8 mM glucose in a 24.8 mM K⁺ medium. The rate of insulin release by 15.5 mM α-MDG, which was indicated as 102 ± 26 μU/ml/min, was significantly higher than that of the control, but 15.5 mM α-MDG did not induce insulin release. Combined infusion of 2.8 mM glucose and 15.5 mM α-MDG or β-MDG caused almost the same insulin releasing profile as that due to α-or β-MDG. Therefore, both agents seemed to bind a glucoreceptor. In the presence of 12.4 mM K⁺, unlike in a 24.8 mM K⁺ medium, β-MDG specifically release insulin.
3. Perfusion in raising a extracellular K⁺ concentration from 6.2 to 12.4, and finally to 24.8 mM
K⁺ (24.8 mM) provoked a transient enhancement of insulin release. The rate of insulin release and the maximal value were 84 ±8 μU/ml/min and 153 ± 26 μU/ml. The rate of insulin release by 2.8 mM glucose was 2.6 times as large as that of the control and that by 15.5 mM μ-MDG, which was indicated as 159 ± 28 μU/ml/min, was significantly higher than that by 15.5 mM α-MDG, which was similar to that of the control. Combined infusion of 2.8 mM glucose and 15.5 mM α-MDG or β-MDG caused almost the same insulin releasing profile as that due to α-or β-MDG alone. Therefore, both agents seemed to bind a glucoreceptor. In the presence of 24.8 mM K⁺ raised from 12.4 mM K⁺, 43-MDG insulin release was specifically enhanced.
The results have led to the suggestion that the B cell may contain glucoreceptors on tne plasma membrane directly controlling insulin release, and moreover, that β-MDG with β configuration could interact with a glucoreceptor to cause insulin release through a change in a glucoreceptor conformation.

View full abstract

The effect of tolbutamide on the oxidation and phosphorylation of isolated rat liver mitochondria was tested under various experimental conditions, and the following results were obtained.
1. The effect of 100 mg % tolbutamide on mitochondrial respiration was not influenced by the addition of uncouplers such as 400 pM Ca* or 100 pM DNP. Uncoupling by the addition of tolbutamide was obserbed with four different substrates, but was not observed by the sonication of mitochondria.
2. In the reaction medium without 10 mM KH2PO4, the respiratin of sonicated mitochondria was inhibited by the addition of 100m % tolbutamide, while the addition of 2.4 DNP had no effect. At a low succinate concentration, mitochondrial respiration wat gradually decreased by 100 mg% tolbutamide.
Addition of ADP, caused a greater decrease in respiratior than did the absence of ADP.
Unlike tolbutamide, at 5 mM-50 mM succinate, DNP did not inhibit respiration, but at less than 3 mM concentration of succinate, respiration was consistently decreased by DNP in either the presence or absence of ADP.
3. Both the above results and previous reports that tolbutamide induced latent ATPase activity of mitochondria, clearly suggest that tolbutamide exerts a DNP like effect on mitochondrial respiration.
Except for the DNP-like effect, tolbutamide has few specific actions on mitochondrial respiration.

View full abstract

Thirty cases of hepatic steatosis associated with varing degrees of hypertriglyceridemia and/or carbohydrate intolerance were studied clinically. We classified the patients into three groups (normal, borderline and diabetic) based on the degree of carbohydrate intolerance, in order to provide a complete profile of insulin response in subjects with hepatic steatosis.
(1) The incidence of obesity (more than 10 % in excess of the standard body weight) in hepatic steatosis subjects was 90.9 %.
(2) The fasting serum triglyceride and free fatty acid levels of hepatic steatosis subjects were significantly higher than those of normal controls, and among hepatic steatosis subjects the fasting serum free fatty acid levels of diabetic type were significantly higher than those of the others.
(3) The normal and borderline type of hepatic steatosis subjects secreted significantly greater amounts of insulin during fasting and during the oral glucose tolerance test than did the normal controls. On the other hand, subjects with the diabetic type of hepatic steatosis showed later hypersecretion when their fasting blood glucose levels were below 160 mg/dl, and showed less secretion when their fasting blood glucose levels were over 160 mg/dl than did the normal controls.
(4) There was a linear correlation between plasma basal IRI and EIRI and fasting serum triglyceride levels, and a linear correlation between plasma basal IRI and LIR' and relative weight in normal and hepatic steatosis subjects. (EIRI is sum of all plasma IRI levels during the oral glucose tolerance test)
(5) Subjects with severe hepatic steatosis showed the following clinical characteristics:(i) hy-perinsulinemia (ii) relatively mild carbohydrate intolerance (iii) hypertriglyceridemia (iv) obesity, and (v) mild disturbance of liver function.
In this study it is postulated that (1) hepatic steatosis is characterized by high insulin levels during the state of mild carbohydrate intolerance and relative or absolute insulin deficiency when decompensation (as characterized by fasting hyperglycemia) appears, although the cause of hyperinsulinemia in hepatic steatosis is as yet unclear, and that (2) hyperinsulinemia and overnutrition might be associated with accelerated hepatic triglyceride synthesis and hepatic steatosis production in normal, borderline and mild diabetic types of hepatic steatosis.

View full abstract

In order to assess the mechanism involved in L-arginine-induced insulin release, we have studied the dynamics of insulin release by L-arginine alone in a K+-depleted, 6.2 mM K⁺, and 24.8 mM K⁺ medium, and by combined infusions of arginine (D, L) and glucose, α-MDG, β-MDG in a 6.2 mM K⁺ medium. The results are as follows.
I. Effect of L-arginine-(1) Perfusion in a K+-depleted medium: Arginine-induced insulin release was biphasic at 2.9 mM, in which the mean rate of insulin release during the fi rst 6-min period and the maximal value were 77 ± 6 μU/ml/min, and 126 ± 31 μU/ml (n=4, M ± SEM), respectively. The amounts of insulin release were augmented by an increase in arginine concentration. The dose response curve was sigmoid, and the Km for arginine was 7.9 mM when calculated by Hill's equation. (2) Perfusion in a 6.2 mM K⁺ medium: Arginine did not induce insulin release at 2.9 mM, and the rate of insulin release was 18 ± 3 μU/ml/min (n=5). (3) Perfusion in a 24.8 mM K⁺ medium: Arginine provoked a transient enhancement of insulin release at 2.9 mM. The rate of insulin release and the maximal value were 163 ± 32 μU/ml/min and 278 ± 64 μU/ml (n=5), respectively. The amount of insulin release was larger than that of the control experiment.
II. Effects of D-and L-arginine in the presence of glucose.α-MDG or β-MDG-(1) Perfusion with L-arginine and glucose in a 6.2 mM K⁺ medium: Combined infusion of arginine (2.9 mM) and glucose (5.2 mM) provoked a transient insulin release. The rate of insulin release and the maximal value were 47 ± 13 μU/ml/min and 81 ± 14 μU/ml (n=4), respectively. Dynamics of insulin release induced with L-arginine over 5.7 mM were biphasic, and the dose response curve was sigmoid with a Km of 4.2 mM arginine. (2) Perfusion with L-arginine and a-MDG in a 6.2 mM K⁺ medium: Combined infusion of arginine (2.9 mM) and a-MDG (5.2 mM) also transiently induced insulin release. The rate of insulin release and the maximal value were 34± 2 μU/ml/min and 48 ± 4 μU/ml (n=5), respectively. The amounts of insulin released by combined infusion of arginine and 5.2 mM a-MDG were increased by increments of arginine concentration. The dose response curve was sigmoid with a Km of 4.0 mM arginine. (3) Perfusion with L-argin-ine and β-MDG in a 6.2 mM K⁺ medium: Combined infusion of arginine (5.7 mM) and 13-MDG (5.2 mM) did not induce insulin release, and the rate of insulin release was 9 ± 3 μU/ml/min (n=3). (4) Perfusion with D-arginine and glucose in a 6.2 mM K⁺ medium: Combined infusion of arginine (5.7 mM) and glucose (2.6 mM) did not induce insulin release, and the rate of insulin release was 12 ± 5 μU/ml/min (n=4). (5) Perfusion with D-arginine and a-MDG in a 6.2 mM K⁺ medium: Combined infusion of arginine (5.7 mM) and a-MDG (5.2 mM) did not induce insulin release, and the rate of insulin release was 14 ± 5 μU/ml/min (n=3).
In conclusion, the specific effect of a combination of L-arginine and α-MDG, in contrast to that of L-arginine and β-MDG, have led to the suggestion that the B cell plasma membrane may contain arginine receptors and glucoreceptors directly controlling insulin release, and that, moreover, both of the mechanisms could be interreleted to enhance the insulin release induced with L-arginine.

View full abstract

In order to study the taste disturbance in diabetes mellitus, electrogustometry was carried out in 158 diabetics and compared with that in 1, 029 normal subjects. Taste abnormalities were noted in 26 % of the diabetic group. Comparison of these results with the knee jerk revealed a significantly more frequent taste abnormality in those with a decreased or absent knee jerk than in those with a normal reflex (p<0.01). In the case of precoma, a marked improvement in taste threshold was noted along with the improvement of blood sugar value. In order to study the relationship between taste abnormality and the disease picture of diabetes mellitus, changes of lingual papillae and lingual blood vessels were photographed, and angiography of the lingual blood vessels was attempted for a more detailed study. According to these results, the lingual papillae of diabetics are frequently flattened which is in agreement with the findings after disruption of the innervating chorda tympani. Such flattening of lingual papillae appears to be due to diabetic neuropathy, in view of the relationship of such findings to the abnormality in the electrogustometry. The blood vessels within the lingual papilla are distributed more coarsely than in normal subjects, with irregular loops, torsion and tortuosity of the blood vessels. Angiography revealed findings suggesting aneurysm. In order to study the diabetic microangiopathy in the tongue further, histological firidings were examined during autopsies. In the small blood vessels of the tongue, hyaline-like intimal thickening with PAS-positive substance was found, confirming the presence of microangiopathy in the tongue.

View full abstract

In diabetics, insulin antibodies in the serum interfere with the action of injected insulin. The capacity of insulin to bind to insulin antibodies has been estimated by various methods according to the single and the double antibody system. In the present study, a quantitative determination of insulin binding capacity was devised at two points of the double antibody method of Andersen. First, free insulin in the patient serum was removed with dexran-coated charcoal, because insulin in the serum interfered with the insulin binding capaciy. Secondly, healthy human serum that had been proved not to contain insulin antibodies was used as the control, although in Andersen's method, buffer was used instead of human serum.
The procedure is as follows: 100 pl of serum that has been pretreated with dextran-coated charcoal and diluted, 100 μl of AIS-GP (anti-insulin serum of guinea pig), 100 μl of porcine 125Iinsulin (0.7-0.8 μU) and 500 μl of borate buffer containing 0.5% bovine albumin are mixed and incubated for 96 hours at 4°C. Then 100 μI of anti-guinea pig serum of rabbit (×4) and 100 μl of normal guinea pig serum (×100) are added to the solution and incubated for 24 hours at 4°C. During incubation, the test tubes are carefully shaken several times to mix the solution. The tubes are then centrifuged at 3000 r. p. m. for five minutes. The radioactivity of the precipitate is measured with a well type scintillation counter. The binding capacity of serum bo is calculated with the following formula: bo=a ([B] a/[B] s-1), where a is the binding capacity of AIS-GP that has been estimated by the method of dextran-coated charcoal.[B] a and [B] are the radioactivity of ¹²⁵I-insulin in the precipitate of the healthy control serum and the patient serum, respectively.
This procedure for the determination of insulin binding capacity is simple. The insulin binding capacity is expressed as μU/ml insulin.

View full abstract

Though MC insulin preparations are said to have been developed from which the heterogeneous peptic contaminants have been removed, for the purpose of lowering the immunogenicity of insulin, a recent report claims the detection of considerable quantities of both substances with molecular weights higher than that of insulin and desamidoinsulin. Hence, a comparative study was made of MC insulin and conventional insulin as to purity and stability, using gel chromatography, polyacrylamide disc-gel electrophoresis, bio-assay (J. P. rabbit method) and other physical/chemical methods.
With gel chromatography applied to three vials of conventional insulin (400 U/10 ml/vial), bcomponent (23%) was detected besides c-component, but no b-component was detected in ten vials of MC-insulin (400 U/10 ml/vial) treated similarly with gel-filtration. With polyacrylamide disc-gel electrophoresis applied to 80 jug of various insulin preparations, it was demonstrated that conventional insulin contained proinsulin like substances (2-3%), arginineinsulin (2.5-3%) and monodesamidoinsulin (10-42%), but MC-insulin contained only traces of monodesamidoinsulin.
The stability of MC-insulin preparations was determined after storage for different periods at 5°C and at 30°C. It was found that there were no significant changes in biological potency, nitrogen content, zinc, or crystalline shapes and sizes after storage for twentyfour months at 5°C or after three months at 30°C. High molecular weight substances were not formed during these periods, but monodesamidoinsulin became slightly increased as storage was prolonged. This tendency was higher after storage at 30°C and was more noticeable in neutral soluble insulin than in an insulin zinc suspension.
Based on the above findings, it was assumed that MC-insulin preparations were more highly purified than was conventional insulin, and were very stable for long periods in cool places. For further development of MC-insulin, studies on the possible immunogenicity of desamidoinsulin and the prevention of its formation and its increase during storage are required.

View full abstract

It is now generally recognized that there has been a remarkable increase of diabetes in the past two decades in Japan. It has also been pointed out recently that in the meantime, changes have taken place in the pattern of diabetic complications associated with its increase. In this study, causes of the deaths of diabetics for the past fifteen years were analyzed to demonstrate such changes. The subjects studied were 8, 908 cases of death from diabetes (diabetes was identified as the underlying cause of death in 4, 974 cases and as a contributory condition in 3, 934 cases), which were extracted from some 570, 000 death certificates reported in Osaka Prefecture during the years 1960-1974.
1) During the period of the study the mean age at death increased from 62.9 yrs to 66.2 yrs indicating a prolongation of life expectancy of more than three years.
2) The contributory conditions of those with diabetes as the underlying cause of death were analyzed by the following two methods;(a) analysis of percent distribution of specific contributory conditions mentioned on the death certificates, and (b) analysis of rate of those with specified contributory conditions among total cases. Both methods indicated a remarkable increase of disease of the circulatory system for the fifteen-year period, especially disease of heart revealed a sharp increase. Cerebrovascular disease and hypertensive disease also exhibited an increasing trend. In the older age group, 65 yrs and over, the most remarkable increase of disease of the circulatory system was observed.
3) Analysis of the distribution of the underlying causes of death of those with diabetes as a contributory condition also indicated an increased trend similar to that of disease of the circulatory system, 38.4% in 1960-1964 to 48.9% in 1970-1974, which accounted for the greatest number of all underlying causes of death. Of diseases of the circulatory system, cerebrovascular disease was most frequent, but an increasing trend was not apparent. On the other hand, disease of heart, which was less frequent than cerebrovascular disease in 1960-1964, showed a rapid increase approaching the level of the latter.
4) In Osaka prefecture the ratio of actual to expected number of deaths from specified underlying causes of death for those with diabetes as a contributory condition was computed on the basis of age and sex distribution of the corresponding causes of death of the general population in Osaka Prefecture, in order to find out the difference in the distribution of causes of death between diabetics and the general population. Tuberculosis and nephritis-and-nephrosis were significantly higher in diabetics. Disease of the circulatory system, however, was as only 1.3 times higher in diabetics than in the general population, although the difference was statistically significant. This indicated that the increase of disease of the circulatory system in diabetics was largely due to that in the general population.
Changes observed here in causes of death of diabetics for the past fifteen years were considered to represent qualitative changes of diabetic complications in Japanese diabetics, strongly suggesting further changes towards the Western pattern of diabetic complications in the future.

View full abstract

It is well known that pregnancy affects diabetes and diabetes affects pregnancy. However, diabetic ketoacidotic coma rarely develops during pregnancy without any history of diabetes or glycosuria. Only 7 cases of diabetic ketoacidotic coma occurring as a sequel to stillbirth without diabetic history have been reported in the past 5 years in Japan.
The authors present two cases which had not been diagnosed as having diabetes mellitus until ketoacidotic coma developed at the time of stillbirth.
Case 1 was a 26-year-old primiparous woman, having neither history nor family history of diabetes. The patient had never had glycosuria before the end of trimester. Soon after glycosuria was detected in the regular examinations in the tocological clinic, the patient complained of thirst and polyuria. Two weeks after these symptoms developed, the patient delivered a dead child and then fell into a coma.
Case 2 was a 25-year-old primiparous woman having no history of diabetes. Her father had a history of glycosuria, but the patient have never had glycosuria in the regular examinations until the beginning of the second trimester. However, early in the seventh month of pregnancy the patient complained of thirst, nausea and vomiting with severe abdominal pain. After this episode the intrauterine death of the fetus was discovered by her doctor. The patient fell rapidly into a coma after the Caesarian operation.
The clinical features of these two cases such as abrupt onset of ketoacidotic coma, insulindependent severe and uncontrollable types of diabetes etc. fulfilled the description of typical juvenile diabetes. For the treatment of both cases a continuous low-dose drip infusion of insulin at the rate of 2 U/hr to 10 U/hr was tried. Case 1 required 70 U of insulin to recover from coma and case 2 required 108 U.
From observation of these cases, it was emphasized that doctors should be alert for the occurrence of ketoacidotic coma during pregnancy, even in women without diabetic history. The effect of low-dose insulin therapy for diabetic coma was discussed briefly.

View full abstract

To study on plasma lipid metabolism in lipoatrophic diabetes, the turnover rate of plasma free fatty acid (FFA) and triglyceride (TG) was measured in two cases of the congenital type of lipoatrophic diabetes using, ¹⁴C-palmitic acid as a tracer.
After an overnight fast, 5 μCi of ¹⁴C-palmitic acid was injected intravenously in 30 seconds into two patients. Venous blood samples were drawn at appropriate intervals during the 32 hours after administration of ¹⁴C-fatty acid. The total lipids were extracted from one-ml aliquots of plasma by the method of Dole, and FFA and TG in the extract were separated according to the modified method of Borgström. Radioactivity of each sample was measured in a gas-flow counter.
The half periods of decay (T 1/2) of plasma ¹⁴C-palmitic acid were 4.5 and 7.5 minutes, which are longer than the values of 1.4-3.4 minutes for the normal subjects reported by Fredrickson. This result indicates that lypolysis is decreased in these cases, and so enhanced lypolysis is not the cause of lipoatrophy in this disorder.
Radioactivity in plasma TG fractions reached the maximum 2 hours after the injection of ¹⁴C-palmitic acid. The fractional disappearance rates af plasma TG were 0.107 and 0.074 rate/hour, which were relatively lower than the values of 0.078-0.347 rate/hour reported by Friedberg for normal subjects.

View full abstract