Since we have compared the allergenic potency of insulin components using the radioallergosorbent test in clinical research on human insulin allergy, we investigated it in rats and mice and determined the immunogenic properties of the IgE antibodies.
In Wistar strain rats immunized with a-component of beef insulin, high levels of IgE antibodies were demonstrated, whereas purified insulins (beef and pork) produced almost no specific IgE antibody.
Among various inbred strains of mice immunized with b-component of beef insulin, the H-2^d strain was a high responder, whereas the H-2^k and the H-2^s strains were non-responders. The high responding strain of mice immunized with purified insulin (beef and pork) produced low IgE titers but the high responding strain of mice immunized with mouse insulin produced almost no IgE antibodies. IgE antibodies against pork purified insulin cross-reacted with beef purified insulin but did not cross-react with fish insulin. These results suggest that the allergic reaction may derive from the insulin purity and insulin species specificity under genetic control.

View full abstract

In preliminary experiments aimed to simplify the methods for determining plasma glycosylated protein (GP), it was found that conversion of plasma free glucose to 5-hydroxymethylfurfural (HMF) occurred, in proportion to the glucose concentration, during oxalic acid-treatment of the plasma. These findings led us to correct the GP value on the basis of the blood sugar level.
As the next step, determinations of GP and hemoglobin As (HbAI) in diabetic and normal subjects were carried out in order to compare their clinical usefullness. The following results were obtained.
1) Both the values of GP and HbAI in diabetes were higher than normL (1.10±0.20 versus 0.84±0.10 nmoles HMF/mg protein, p<0.001, and 11.4±3.11 versus 7.1±0.65%, p<0.001, respectively) The coefficient of variation for GP was much lower than that for HbAI in diabetic subjects (18% versus 27%).
2) In the case of diabetes, the correlation coefficient between GP and fasting blood sugar (FBS) measured with the same sample was significant and similar to that between HbAI and FBS (r=0.605 and 0.618, respectively). Also, a positive correlatin was found between GP and HbAr (r=0.547).
3) The levels of GP, HbAi and FBS were all high in patients treated with diet, drug and insulin in that order. The correlation between GP and FBS as well as that between HbAI and FBS became significant in the order of the diet, insulin and drug groups, and in particular HbAI in the drug group was highly correlated with FBS (r=0.798). The correlation between GP and HbAI was remarkable in both the cases of the drug and insulin groups (r=0.594 and 0.614, respectively) and less remarkable in that of the diet group (r=0.315).
4) In diabetic subjects, HbAI was more highly correlated with the 2 h-postprandial blood sugar (2h-BS) measured 2 or 3 months before, than with that at the same time as or 1 month before the HbAI assay. On the othr hand, GP tended to be correlated more highly with the 2 h-BS measured 1 or 2 months before.
From the above results, it can be said that HbAi is indeed valuable as an indicator of diabetic control, while the clinical application of GP appears, at present, impractical and of limited usefullness, It is clearly very important therefore to improve further the methods for measuring HbAr so that many samples can be handled more rapidly.

View full abstract

A total of 507 subjects who received oral glucose tolerance tests (OGTT) because of postprandial glycosuria in an epidemiologic survey, were called for retest 7 years later, and the 207 out of 419 subjects (excluding 88 deaths during this period) who responded to the retest are included in the present study. The rate of worsening to diabetes in those with impaired glucose tolerance, and the risk factors were examined.
1) The mean serum glucose' values in OGTT at the follow-up examination were elevated by about 10%, compared to those at the initial examination. The elevation of serum glucose values was much greater in obese subjects than in nonobese subjects.
2) Those with a 2-hour glucose value (2-h) between 110-199 mg/dl at the initial test tended to shift to the lower glucose category, 2-h<110 mg/dl, when they were not obese. On the other hand, there was an obvious trend to move to the higher glucose category, 2-h≥200 mg/gl, when associated with obesity.
3) Among those with a fasting glucose value (F)<140 mg/dl and a 2-hour glucose value <<200 mg/dl, the worsening rate to “diabetes” according to the new criteria recommended by WHO in 1980 (F≥140 mg/dl or 2-h≥200 mg/dl) was increased as the 2-hour level of the subjects at the initial test elevated. When these subjects were further broken down into three subgroups: normal, borderline and diabetic according to the criteria of Japan Diabetes Society, the corresponding worsening rates were 1.1%, 8.5% and 55.6%, respectively.
4) Multivariate discriminant function analysis indicated that three glucose determinations of OGTT, fasting, 1-hour and 2-hour values, sex and obesity were highly related to the worsening.
It was concluded that most “borderline” cases according to the conventional classification can be reclassified as normal for practical purposes. At the same time however, it should be pointed out that elevated serum glucose levels and obesity do represent important risk factors for worsening to diabetes.

View full abstract

Raised mean levels of HDL-cholesterol (HDL-C) have been reported in insulin treated diabetics. However, it is unclear whether the increase of HDL-C is actually due to the effects of insulin itself.
In order to determine whether insulin itself could increase HDL-C, we divided 155 diabetics into three groups with different treatments-insulin therapy (20 males, 24 females), oral therapy (19 males, 30 females), and diet only (24 males, 38 females)-and compared the mean plasma HDL-C levels among the three therapeutic groups, matched for triglyceride (TG), total cholesterol (TC), and obesity index. When the three groups of diabetics were so matched, insulin treated diabetics were found to have significantly higher plasma HDL-C concentrations than diabetics treated either with oral agent or with diet alone. This result suggests that the increase of HDL-C in insulin treated diabetics is not dependent on the levels of TG, TC, or obesity index.
Furthermore, we measured the plasma HDL-C before and after treatment with insulin in 12 diabetics (8 males, 4 females; age 56±14 years). the plasma HDL-C concentrations during insulin therapy were significantly higher than thse before insulin treatment (56.4±15.6 v.s.47. 9±13. mg/dl, p<0.02, Student's t test for palred data).
From the results of this study, it is concluded that insulin itself increases HDL-C, and that the increase in HDL-C is not directly related to changes of TG, TC, and obesity.

View full abstract

In order to elucidate the effect of thyroid hormone on endocrine pancreatic functions, investigations were made of the dynamics of insulin and glucagon in intravenous glucose tolerance tests (IV-GTT), arginine tolerance tests (ATT) and perfusion studies of the isolated pancreas in rats treated with L-thyroxine (80 mg/kg i.p. inj.) for 14 days.
In IV-GTT, insulin secretion was markedly decreased, but glucose intolerance did not occur. In ATT, increased insulin release was observed. However, the insulin responses to glucose and arginine from perfused rat pancreas were not changed. The glucagon secretions in these experiments were not changed at all.
The above results suggest that thyroxine exerts a strong effect on insulin secretion in the rat in vivo, but has no effect on glucagon secretion.

View full abstract

A low response of the renin-angiotensin-aldosteone (RAA) system was reported by Christlieb and co-workers in patients with diabetic nephropathy. However, their clinical studies were based on observations of long-duration juvenile diabetic patients who revealed established hypertension and renal dysfunction. The present investigation was therefore designed to establish the kinetics of the RAA system in the compensated phase of diabetic nephropathy.
Plasma renin activity (PRA) and plasma aldosterone (PA) were measured at bed rest and after stimulation with furosemide in 27 adult-onset diabetics and 15 age-matched controls. Seventeen of the diabetics did not suffer from proteinuria (non-proteinuria group) but the other subjects suffered from persistent proteinuria, but from no hypertension or renal dysfunction (proteinuria group). They were kept on a diet containing 86 mEq of sodium as NaCl In the non-proteinuria group, the PRA levels from bed rest time to the post-stimulation period did not significantly differ from those in the control group. In the proteinuria group, the PRA levels were significantly lower than in the other two groups. The PA levels were slightly higher in the non-proteinuria group than in the control group (p: N.S.). In the proteinuria group, the PA levels were significantly lower than those in the other two groups at bed rest time and after stimulation. These findings suggest that (1) the renin-aldosterone system generally responds normally in diabetics without proteinuria but responds subnormally in diabetics with persistent proteinuria, even when there is no complicating hypertension or decreased renal function, and (2) the phenomen a are more closely related with renal disease itself than elevated blood pressure and renal dysfunction. The results for adult-onset diabetics are consistent with the suggestion that hyporeninemic-hypoaldosteronism is found in juvenile-onset diabetics with nephropathy.

View full abstract

Diabetic neuropathy has been described in the diabetes associated with chronic pancreatitis, but the frequency and clinical features are not yet clear. This report utilizes data from our records in an attempt to shed more light on this problem.
Clinically, diabetic peripheral neuropathy occurred in 13 cases out of 39 diabetic patients with calcifying pancreatitis and in 28 cases out of 59 primary diabetics. Painful diabetic peripheral neuropathy was observed in 5 of the patients with pancreatic diabetes. The calcifying pancreatitis was diagnosed simultaneously with or prior to the diagnosis of diabetes mellitus in 2 out of the 13 cases of calcifying pancreatitis with diabetic peripheral neuropathy, and the other cases were diagnosed after the diagnosis of diabetes mellitus. All the pancreatic diabetics with diabetic peripheral neuropathy were on insulin therapy, and these patients had no family hisotry of diabetes.
Although the diabetic peripheral neuropathy in the patients with primary diabetes revealed no significant correlation with the duration of diabetes, there was a significant correlation with the duration of diabetes in the patients with pancreatic diabetes. On the other hand, the diabetic peripheral neuropathy was seen in moderate and severe diabetes in patients with pancreatic diabetes, whereas the degree of diabetes did not appear to have a favorable influence on the occurrence of diabetic peripheral neuropathy in patients with primary diabetes.
These data suggest that a different etiology may exist between the occurrence of diabetic peripheral neuropathy in pancreatic diabetes and in primary diabetes. Also, primary and pancreatic diabetes mellitus cannot be differentiated with certainty on the basis of whether or not diabetic peripheral neuropathy is present.

View full abstract

Phenyltin compounds have been widely used as agricultural chemicals and marine antifoulants. They have been intensively studied from the stand-point of toxicity. However, the in vivo mechanisms of their toxicity and metabolism remain poorly understood. We previously reported that oral administration of 100 mg/kg body weight of tripheyltin fluoride (TPTF) to rabbits evoked hypertriglyceridemia. The present experiments were conducted to elucidate the mechanism of such TPTFinduced hypertriglyceridemia by following the time course of fasting blood glucose (FBS), glucose tolerance tests, serum IRI, serum lipids and lipoprotein lipase (LPL) activity, and also by determining the influence of insulin on the TPTF-induced hypertriglyceridemia. The pancreatic islets, liver and kidney of the rabbits were also examined histologically. After TPTF administration, the FBS and plasma triglyceride concentration were elevated significantly (p<0.02) for about 10 days. On the other hand, the PHLA (plasma postheparin lipolytic activity) and LPL in adipose tissue were remarkably decreased (p<0.02), while chyromicron and VLDL were increased.
Insulin treatment for 1 or 2 days increased the LPL activity to the normal level in these rabbits, but withdrawal of insulin from the insulin-treated rabbits again resulted in a decrease in LPL activity. These results indicate that a single oral dose of TPTF evoked diabetic lipemia and glucose intolerance due to insulin deficiency. The IRI level was decreased before and even after glucose infusion during these periods. Microscopic examinations revealed no remarkable changes in the pancreas islets, liver and kidney. Pancreatic B-cells contained normal amounts of granules at the maximal hyperlipemic stage. These findings suggest that TPTF interferes with insulin release from rabbit beta-cells, and/or TPTF decreases the sensitivity of islets tQrelease insulin in response to increasing levels of blood glucose.

View full abstract