The dynamics of the blood glucose concentration and insulin activity during the intravenous glucose tolerance test in 134 patients were examined mathematically using the so-called Ackerman model. Five unknown parameters could be successfully estimated in 110 out of the 134 cases by a discrete-time linear least square method. The fitting rate was 82%. The model was then evaluated physiologically and clinically based on values for these parameters under various clinical conditions.
The mean value of the rate constant of insulin release in diabetics (mean±1 SD, 0.0086±0.010, N=32) was significantly lower than that in non-diabetics (0.029±0.034, N=78) Also, the mean value of the rate constant of insulin removal from the blood in the diabetics (-0.036±0.019) was significantly higher than that in the non-diabetics (-0.077±0.057). On the other hand, the rate constant related to change of glucose scarcely contributed to the discrimination of the two categories. Moreover, these parameters indicated physiological absurdity in two-thirds of the fitted cases. One paradox was the positive sign of the rate constant of glucose removal from the blood due to insulin or glucose itself. Principal component analysis with 4 variables (4 parameters) yielded two components with which the clinical states of diabetes mellitus and degree of glucose intolerance could be estimated quantitatively.
The physiological impertinence found for the parameters related to glucose concentration hindered a clear understanding of the pathological states in diabetes mellitus and in glucose intolerance.

View full abstract

The effects of diabetic sera on in vitro production of immunoglobulins in normal human peripheral blood lymphocytes were examined to elucidate whether there are any humoral factors that inhibit immunoglobulin prodution in diabetic patients. Forty-three diabetic patients and 15 normal adults were examined by the immunofluorescent detection of intracytoplasmic immunoglobulin in vitro stimulated with pokeweed mitogen (PWM). Cytoplasmic immunoglobulin production of normal B lymphocytes incubated with diabetic serum samples was significantly decreased (p<0.025) compared with that of control serum. Normal serum restored the impaired cytoplasmic IgG production in diabetic lymphocytes derived from patients under treatment with oral agents and insulin. However, there was no significant difference in the cytoplasmic IgG production of normal lymphocytes between those incubated with dialyzed and non-dialyzed serum samples.
It is concluded that there is some inhibitory factor for cytoplasmic immunoglobulin production in diabetic serum. This factor was assumed to be a high molecular substance.

View full abstract

Urinary N-acetyl-beta-D-glucosminidase (NAG), serum and urinary beta 2-microglobulin (sBMG, uBMG) and urinary protein were measured in 103 diabetic patiens with or without clinical diabetic nephropathy.
The results obtained were as follows.
1) NAG which indicates active renal proximal tubular cell damage is thought to be the earliest manifestation of renal dysfunction in diabetic patient, followed by increase of urinary protein excretion and elevation of sBMG.
2) Studies on the ratio of uBMG to urinary protein and BMG clearanc to creatinine clearance reveal that renal tubular dysfunction exist in about 20 % of diabetic patients. And, high ocurrence of renal tubular dysfunction is noted in patients with marked urinary NAG activity.
3) A positive, good correlation is observed between sBMG and serum creatinine. Thus, sBMG is a reliable index for monitoring the progression of diabetic nephropathy.

View full abstract

Insulin (2 mg/ml) was incubated with glucose (36 mg/ml) for 14 days at 37°C in 0.1M sodium phosphate buffer, pH 8.5. Incubation was carried out in the presence and absence of 0.25M sodium cyanoborohydride. D-(¹⁴C)-glucose or NaB³H₃CN was used as a tracer.
Unreacted glucose was removed by gel filtration on a Bio-Gel P-6, and unreacted insulin was separated from glucosylated insulin by affinity chromatogrophy on phenyl boronate coupled to agarose (glyco-Gel B, Pierce, Illinois). Synthetic ¹⁴C-glucosylated insulin reduced with NaBH₃CN and ³H-glucosylated insulin reduced with NaB³H₃CN were eluted in the same fraction. Moreover, we detected sugar in this fraction by periodate oxidation. Thus, we believe that three glucoses, were taken up per mole of insulin monomer.
The immunological reactivity of glucosylated insulin was 20% lower than native insulin.
The hormonal effects on ¹⁴C-glucose oxidation and lipogenesis by rat adipocytes was diminished by 70%.
It was shown that glucose was incorporated into the insulin molecule upon incubation in vitro. The biological and immunological activities of the products were reduced.
Based on these results, it might be suggested that glucosylation of insulin also underwent a modification similar to hemoglobin A. We have considerable interest in the glucosylation of insulin in vivo at the hyperglycemic state.

View full abstract

The importance of physical activity in the treatment of diabetes mellitus is generally acknowledged, mainly because of its potential blood glucose-lowering and insulin-sparing effects. The precise nature as well as the underlying pathophysiological mechanism remains, however, to be clarified. Nine poorly-controlled and 4 well-controlled insulin-dependent diabetic patients and 6 healthy control subjects were studied at rest and during and after 60 min of bicycle exercise (40% of Vo₂ max). The catheter technique was employed to determine the splanchnic exchange of metabolites.
The basal arterial concentrations of glucose (269±13 mg/dl), FFA (0.95±0.10 mEq/l) and total ketone body (KB)(1.11±0.22 mmol/l) were significantly (p<0.001) higher in the poorlycontrolled diabetics than in the well-controlled diabetics and controls. The basal arterial glucose and FFA levels of well-controlled diabetic patients were almost equal to those of the control subjects, while the KB concentrations of these diabetics were significantly (p<0.05-0.001) higher than those of the controls. Arterial FFA remained unchanged or slightly increased during exercise. During the post-exercise recovery period there was an initial distinct rise in FFA concentration followed by a gradual decrease, in the diabetics and the controls. The concentration changes occurred in parallel for all three groups but the levels were 60-100% higher in the poorly-controlled diabetics. Glycerol levels exhibited a continuous rise in response to exercise, decreased promptly after exercise and showed a behavior comparable to that of FFA in all three groups. Splanchnic ketone production (SKP) rose gradually during exercise, increased further, and peaked at 10-20 min after exercise. Although the FFA and glycerol levels of well-controlled diabetics were kept almost equal to those of the controls, the stimulation of SKP during exercise persisted during the recovery period in the diabetics. The arterial KB concentrations were unchanged during exercise, rose promptly after exercise and remained elevated throughout the 60 min of the recovery period in the diabetics, but not in the controls. It is concluded that the stimulation of KB production during exercise persisted during the recovery period in the diabetics, regardless of the controlled condition of diabetes, probably as a result of insulin deficiency in the diabetic patients. These findings confirm the previously reported conclusion of the importance of adequate insulin treatment in connection with exercise.

View full abstract

It has been suggested that hepatic triglyceride lipase (H-TGL) plays a role in the hydrolysis of serum lipid. However, the precise function of the enzyme is not yet clear. Based on two lines of reasoning we emphasize the importance of direct measurement of the plasma membrane-bound HTGL activity for elucidating the function of the enzyme in the rat liver. First, when the H-TGL activity in post-heparin plasma is measured, it covers not only the enzyme activity originating in the liver, but also that from the ovaries and adrenal glands of the rats. Second, recent studies have indicated that the H-TGL in the liver is located mainly on the plasma membranes and is the origin of the lipase released from the liver by heparin.
Plasma membrane-associated H-TGL activity was reduced by fasting and streptozotoein diabetes, whereas is was partially restored by refeeding and insulin treatment, respectively. The H-TGL activity of rats fed a no-fat diet was higher than that of those fed a fat-containing diet. The changes in enzyme activities under all these conditions were found to coincide with those of the plasma IRI (immunoreactive insulin) levels. It is strongly suggested therefore that the H-TGL associated with plasma membranes is under hormonal regulation by insulin.
In the present study, no relationship was observed between the enzyme activity and serum triglyceride or cholesterol levels. The physiological significance of the enzyme in the hydrolysis of serum lipids remains obscure. However, it is suggested that H-TGL may function in order to deliver free fatty acids originating in the serum triglyceride for esterification in the liver and subsequent very low density lipoprotein synthesis in harmony with other enzymes affected by insulin.

View full abstract

We investigated the occurrence of insulin autoimmune syndrome among hypoglycemic subjects in Japan by sending questionnaires to 2049 hospitals during the three-year period from 1979 to 1981.
We received responses from 1033 hospitals (50.4%). No cases of hypoglycemia were reported at 769 (74.4%) of these hospitals, while more than one case was reported at 264 (25.6%). The total number of hypoglycemia cases was 461.
Causes of hypoglycemia were as follows, in the order of frequency: Insulinoma (125 cases, or 27.1%), oral hypoglycemic agents (101 cases, or 21.9%), extrapancreatic neoplasms (92 cases, or 20%) and insulin autoimmune syndrome (41 cases, or 8.9%).
Of the 41 insulin autoimmune syndrome subjects, 18 were male and 23 were female. In seven cases (four males, three females) the onset occurred after the use of thiamazole and, in two (both males) after the use of mercaptopropionyl glycine.
The incidence of insulin autoimmune syndrome during the three years of our study was 32% of all insulinomas. We concluded that insulin autoimmune syndrome incidence ranked fourth as the cause of hypoglycemia in Japan, excluding hypoglycemia caused by insulin injections.

View full abstract

A case with circulating glucagon binding antibody without any previous immunization was described. The patient was a 47-year-old married nurse. At the age of 44 (1977), a diagnosis of hyperthyroidism was made. Subsequently, she began to take methimazole (MMI). Four months of MMI treatment brought about a euthyroid state. She had no history of previous insulin or glucagon injection. The OGTT revealed a normal glucose tolerance. There was no circulating antibody to insulin or pancreatic polypeptide. The treponema pallidum hemagglutination test was negative.
Plasma immunoreactive glucagon could not be assayed in the native plasma taken in April, 1980 because there was a substance that strongly bound ¹²⁵I-glucagon and interfered with glucagon radioimmunoassay. This binding substance was detected by the polyethylene glycol method. The value expressed as specific binding % was 13.2 (control=1.2). This high binding could be displaced only by porcine glucagon and porcine glicentin and became dissociated at acidic pH. The substance was found in the plasma protein fractions on gel filtration and proved to belong to γ-globulin (IgG, L-chain K-type) by means of salting out with ammonium sulfate and specific radioprecipitation.
The patient's history indicates that the glucagon binding antibody might have developed through autoimmunity. However, there were no clinical symptoms caused by the autoantibody.
Aside from the patient reported here, the glucagon binding antibody was detected in plasma samples of three patients with hyperthyroidism under MMI treatment out of 68 patients with thyroid disease (40: hyperthyroidism; 36: treated with MMI). These findings suggest that the development of glucagon binding antibody in hyperthyroidism may be associated with MMI treatment.
A few cases of hyperthyroidism treated with MMI and having insulin autoantibody have already been reported. However, this is the first report showing glucagon autoantibody.

View full abstract

Glycemic control under continuous subcutaneous insulin infusion (CSII) was examined in 24 inpatients, and was compared with conventional treatment using the mean blood glucose, M value, mean amplitude of glycemic excursion (MAGE) and urine glucose.
The mean blood glucose, M value and urine glucose were improved under CSII compared to the conventional treatment. However, the MAGE in 7 out of the 24 patients exceeded 125 mg/100 ml. In 3 cases, persistent high blood glucose levels after breakfast resulted in a poor MAGE. One patient had a high percentage of insulin binding antibody, and the total and fre insulin contents in the serum did not parallel the rate of insulin infusion during CSII. In 2 other patients, the causes were derived from their irregular dietary habits. The last 2 patients, brittle diabetics, could not be controlled owing to wide, fast and unpredictable swings in their blood glucose levels under CSII in spite of long-term studies, in contrast with the treatment of an artificial pancreas.
Using an artificial pancreas, when we claculated the average infusion rate of insulin during a whole day, the ratio of insulin infusion at the time of breakfast, lunch, and evening meal was 100: 70: 81.
It is concluded that CSII improved the dialy profile of hyperglycemia in most of the diabetics when we used a suitable dose of insulin in each subject, but did not do so in the brittle diabetics. The reason for this is not clear. One of the main causes of brittleness may be abnormal absorption and degradation of injected insulin at the injected site.

View full abstract