[in Japanese]

View full abstract

Insulin-like growth factor (IGF) is the genoric designation for serum polypeptides such as somatomedins, multiplication stimulating activity (MSA), and non suppressible insulin-like activity (NSILA), which possess both insulin-like activity and cell growth promoting activity. We have purified one of IGF's from human plasma and established radioreceptor assay (RRA) for this IGF. RRA for insulin, using human placental microsomal membranes, was used to detect insulin-like activity throughout the isolation procedure. The soluble fraction from an acid ethanol extract of Cohn fraction IV was chromatographed on G-75 sephadex in 1% formic acid. The insulin-like activity which binds to insulin receptor (ILA) migrated as a small molecule (Kav=0.56-0.70). The active fraction was subjected to isoelectric focusing. Three peaks of ILA were identified at the position of pH 5.2, 7.0 and 8.5. The major fraction (pI 8.5) designated as ILA-C was further purified by G-50 sephadex gel filtration in 1M acetic acid. The apparent molecular weight of purified ILA-C is around 6,000. ILA-C stimulated ¹⁴C-glucose oxidation in rat epididymal fat cells. It also potently augmented ³H-Thymidine incorporation into DNA of 3T3 cells. Thus ILA-C satisfies the characteristics of IGF, but the identification of ILA-C among known IGFs' is not clear at present.
Subsequently RRA for ILA-C, utilizing human placental membrane as a receptor, was developped. Placental membrane fraction (100-200μg protein), ¹²⁵I-ILA-C (50,000cpm) and ILA-C standard or unknown serum in a total volume of 0.5ml were incubated for 16 hrs at 4°. The membrane-bound and the free labelled ILA-C were separated by centrifugation and the pellet was counted in a gamma scintilation counter. An arbitrary unit of ILA-C was expressed as μU/ml insulin equivalent as determined by RRA for insulin. The bound ¹²⁵I-ILA-C was displaced dosedependently by ILA-C between 0.5μU and 100μU per tube. The binding of labelled ILA-C to the receptor was inhibited partially by somatomedin A and MSA, but only minimally by insulin. Somatomedin B, epidermal growth factor, nerve growth factor, fibroblast growth factor or proinsulin were without effect. Addition 1 to 10μl of sera gave the displacement curve parallel to that obtained by the standard.
Comparison of elution patters of ILA-C in human serum on G-50 sephadex at acidic and neutral pH suggests that ILA-C is associated with serum protein and the complex is dissociated by acid. The presence of binding protein in serum might disturb the assay. The dilution curve of native serum between 1μl and 10μl was parallel to that of acid-dissociated serum, and thus ILA-C can be quantitatively measured in native serum at neutral pH in the range between 1μl and 10μl. Therefore all the following assays were done at neutral pH using serum less than 1μl. The intra and inter-assay variences were 6.7% and 20.4%, respectedly.
The RRA can measure the concentration of ILA-C in the sera of a variety of mammals. Serum concentrations of ILA-C in normal males and females were 575±46 (mean±SEM), 528±27, respectedly. ILA-C levels were high in acromegalics (3, 133±245) and low in hypopituitary patients (286±47). Administration of hGH to patients with GH deficiency elevated ILA-C levels significantly. In cases of anorexia nervosa, ILA-C levels were generally low but increased as the body weight returned to normal. Very low levels were also found in cord bloods and in patients with liver cirrhosis, whereas uremic patients had high levels. In patients with diabetes mellitus and hypoglycemia due to extrapancreatic tumor ILA-C levels were normal or low.
In summary, ILA-C was purified from human plasma using RRA for insulin and sensitive RRA for ILA-C was established. The levels of ILA-C in the serum are under the regulation of GH, as well as nutritional and other factors.

View full abstract

Epidermal growth factor (EGF) is a polypeptide that has been isolated from mouse salivary gland and human urine. EGF stimulates proliferation of epidermal tissues of new born animals and causes precocious eye lid opening. In addition, like other so called growth factors, it is a potent mitogen for fibroblasts from various speicies. Recently, urogastrone, a potent inhibitor of gastirc acid secretion, was isolated from human urine. It bore a marked structual relationship to the EGF and now urogastrone is considered to be the same as human EGF.
Although extensive studies have been carried out with EGF, its physiological role remain uncertain. It binds to the specific plasma membrane receptors, and this is the first step in its biologic action as other peptide hormones. Therefore, tissue distribution of specific binding site for mouse EGF was studied. Tissue homogenate in 50 mM Tris-HCl buffer pH 7.4 were centifuged at at 12,000 xg. The resultant pellet was resuspended in the above buffer containing 0.1% BAS and 10mM MgCl₂ and incubated with 0.5 ng of ¹²⁵I-mouse EGF for 90 min at 20°. The incubation was terminated by adding ice cold buffer, and tissue bound ¹²⁵I-EGF was separated by centrifugation. Specific binding sites for EGF were demonstrated in a variety of tissue, being the greatest per unit of protein in mouse liver. Scatchard analysis revealed that EGF receptor from mouse liver had apparent Kd 1.5×10^-10 M with binding capacity of 25 f moles/mg tissue protein.
A specific radioreceptor assay was developed using the mouse liver capable of detecting 100pg of EGF. RRA values for EGF content in mouse submaxillary gland were in agreement with those obtained by radioimmunoassay. EGF like activity were detected by RRA in the submaillary gland and in urine of various species including human. Physicochemical characterization of EGF like substance in human urine showed it was identical with urogastrone. Thus this simple, rapid RRA for EGF is available for detecting and purifmg EGF of various species.

View full abstract

In order to establish a radioreceptorassay for catecholamines, mainly based on the method of Lefkowitz et al, we have investigated the properties of the microsome fraction of bovine myocardium as catecholamine receptors, and obtained the following results.
1) The binding of catecholamines to the microsome fraction is biphasic and the affmity constants of high and low bindings 1.56×10⁷M^-1 and 1.00×10⁴M^-1, respectively.
2) The binding site of catecholamines to receptors is thought to be the catechol nucleus.
3) Radioreceptorassay is sensitive but not so specific, becuase many compounds with catechol nucleus showed cross-reaction.
4) Cross-reactivity of adrenergic blocking agents suggest that our purified bovine myocardium microsome fraction contains alpha-adrenergic receptors.

View full abstract

Recently the laser-nephelometer, which is especially designed for the measurement of the light scattered from immunoprecipitate in solution, has become available, and nephelometric immunoassay has become widely used for the determination of precipitating antigen such as serum protein. We expected that nephelometry would provide a method also for the determination of hapten (non-precipitating antigen). The assay principle is based on that hapten (monoantigenic small molecule) forms soluble immunocomplex which does not scatter the light whereas polyhaptenic molecule forms insoluble immunoprecipitate which scatters the light. Therefore, when the hapten competes for the binding site of the antibody with the haptenic moiety of the polyhaptenic molecule, the hapten can be determined by the measurement of the decrease of the scattered light due to the inhibition of the immunoprecipitation of the polyhaptenic molecule.
The sensitivity of this “competitive nephelometric immunoassay” method was not expected to be high. However therapeutic blood levels of some drugs were thought to be high enough for the sensitivity, and the monitoring of some drugs in patient body fluid is clinically useful. Therefore we used some drugs as model molecules to be determined, and used serum albumin as the hapten-carrier of the polyhaptenic molecule. The assay conditions of this method were studied, and we have developed an accurate, simple and rapid method for the determination of some drugs in human body fluid. The patients' plasma specimens were assayed, and the values correlated well to those determined by the high-performance liquid chromatography (correlation coefficient: 0.971 for theophylline, 0.983 for phenobarbital, 0.987 for diphenylhydantoin).

View full abstract

A new immunological method of quantitative measurement of antigens and antibodies was developed by applying the near infrared turbidimetry to the latex agglutination reaction. The turbidity of the antibody-sensitized latex (0.1 to 0.8 μm in the diameter) increases very markedly along with the immunoreaction with the corresponding antigens in the near infrared region, not in the visible or ultraviolet region. This phenomenon is supported theoretically by the Mie scattering theory which handles the light scattering by relatively large particles.
Detailed studies by the rate method in the system were done with hCG (human chorionic gonadotropin), and the system was evaluated from the view points of reproducibility, recovery and correlation with conventional RIA method. The characteristics of the system were exhibited as follows:
(1) Both within and between assay variance (C. V.) of the results obtained through the reproducibility experiment of the rate measurement were less than 5% with triplicate specimens in each of three-time experiments.
(2) Data processing was carried out according to the second order equation, log V=a+b (log A)+c (log A)², where V represents the rate of turbidity increase at the early phase of immunoreaction, and A represents the concentration of the antigen. The calibration curve obtained with hCG in saline solution containing 0.2% of BSA showed very narrow width of confidence band.
(3) Interference by turbid serum specimens was negligible. Variance (C. V.) of the results obtained through the reproducibility experiment with clinical specimens was as small as those obtained with the standard samples dissolved in saline.
(4) The recoveries of added hCG in sera and urine were 94-106% and 94-117%, respectively, except for two cases of the lowest concentration of hCG dissolved in urine.
(5) Correlation coefficient between this technics (LPIA) and the conventional RIA was more than 0.97.
(6) The minimum detectable dose of hCG is 2 ng (30 mIU) per ml. with the measurable range up to about 200 ng (31U) per ml., when 100 μl sample is used.
(7) Measurement time required for hCG is in the range of 40 to 120 seconds.
In summary, this study has shown that the LPIA system is a promising approach to detect the minute amount of antigen by simple procedure. An overall agreeement in terms of correlation of the LPIA system with RIA procedure has been obtained with an advantage of negligible interference by any color and turbidity of clinical samples. Further works are requied to expand the system for many other kinds of antigens and to provide the computer-based LPIA with fully automated instrument hardware.

View full abstract

A simple and fast fluoroimmunometric assay for accurate and precise quantitation of immu- noglobulins using FIAX System was evaluated. FIAX System includes StiQ samplers, FIAX Test Kit, FIAX 100 Fluorometer and FIAX 110 Microcomputer.
Repeatability and daily reproducibility of the determinations are reasonably good, and the method is fairly well correlated with single radial immunodiffusion and AIP (automated immune precipitin) methods.

View full abstract

Employing a tequnique which determined the rotary Brownian motion by fluorescene polari- zation, an attempt was made to assay for blood thyroxine.
The principle of the method was briefly as follows. Fluorescent-labelled T-4 reacted with T4 with anti-T-4 fractionated to IgG, so that the weight of the molecules with fluorescence was con- siderably altered; this then resulted in a rotary Brownian motion. Fluorescene polarization was detected by FP analyzer (JIMCO Polarization Spectrofluorometer Model IBF-129)(Fig. 1).
By this method, a correlated relation ship between the alteration in fluorescence polarization and the concentration of thyroxine extracted from serum with methanol was obtained (Fig. 3).
A comparison was made between the assay levels obtained by the present method and the radioimmunoassay (Fig. 5). The specificity of the present method for T-4 was confirmed by examining the cross reactivity with tri, di and mono-iodothyronine (Fig. 4).
For clinical purpose, the present assay method requires only a small amount of serum (less than 100μl is available). The main advantage of our method is that the assay requires only simpleand rapid procedures as well as no use of radioisotope. The procedure of separation for T4 andanti-T-4 complex from free T-4 is eliminated.

View full abstract

A method was developed for fluorescence polarization immunoassay of serum cortisol with cortisol 21-amine, which readily couples with fluorescein isothiocyanate. In this immunoassay, it is not necessary to separate bound and free form. The minimal amount of cortisol detectable was 1.5 ng/tube and serum concentrations of 1.5 to 100μg/dl of cortisol could be measured. A good orrelation was found between the values obtained by this method and by radioimmunoassay. This method is sufficiently sensitive, reproducible, specific, simple and rapid for use in routine determination of serum cortisol.

View full abstract

Glucose intolerance is an important feature of diabetes mellitus. However, the mear abnor-mality of glucose tolerance is inadequate for the accurate diagnosis, since there are so many pathological conditions which impair glucose tolerance. The aim of the present study is to investi-gate whether the maturity onset type diabetes developes from low insulin responders to glucose load or not.
Blood glucose and IRI were assayed before and at 30min, 60min, 90min, 120min and 180 min after oral administration of 100g glucose.(1) In definitely diabetic patients, even in the sub-jects who showed normal type glucose tolerance curve after a certain treatment period, IRI (the sums of six insulin values) remained low throughout the wide range of BG (the sums of six blood glucose values). As the absolute increase in plasma insulin depends on the magnitude of glycemic stimulus, the ratios of increment of insulin (ΔIRI in μU/ml) to that of blood glucose (ΔBG in mg/100ml) 30min after glucose load were calculated. ΔIRI/ΔBG values scattered considerably in patients who visited diabetes clinic. However, the values of definitely diabetic patients were invariably lower than 0.5 whether they were obese or not, and regardless of their glucose intol-erance.(2) The investigation was extended to the studies of insulin responses of the patients who had equivocal diabetes initially but later developed diabetic retinopathy. About half of these patients had a family history of diabetes. Their initial glucose tolerance curves were either diabetic, borderline or normal. However, the initial ΔIRI/ΔBG ratios were all less than 0.5 irrespective of their types of glucose tolerance and remained consistently low during the entire follow-up periods regardless of changes in types of glucose tolerance.(3) The offsprings of two diabetic parents have high chance to get diabetes in the future. In 57 such cases, 23 had normal, 15 had borderline and 19 had diabetic type glucose tolerance. The overall prevalence of low insulin responders (ΔIRI/ ΔBG less than 0.5) was about 60 per cent, and it was fairly constant in different age groups despite an increase in the number of cases with glucose intolerance as an increase of age. When the results were expressed by BG-IRI plots, these siblings with ΔIRI/ΔBG below 0.5 were distributed in the zone of definite diabetes, and those with ΔIRI/ΔBG above 0.5 fell in the zone of various pathological conditions other than diabetes.
The present study supports that the early low insulin response after glucose load is a characteristic feature for true diabetes and presents before the development of overt diabetes.

View full abstract

To assess the effects of pregnancy on secretory function in pancreatic alpha and beta cells, the following was studied during mid and late pregnancy and also 1 week and 4 weeks post partum. Glucose, insulin (IRI), C-peptide (CPR) and glucagon (IRG) were measured during basal condition and after oral challenge with glucose or intravenously with arginine. The carbohydrate metabolic function was classified following to the recommendatory criteria of Japan diabetic society and White's criteria.
Fasting values of IRI, CPR and IRG had a tendency to increase during pregnancy, which were smaller in the degree in diabetic. The CPR responses were sustained in diabetic, class A, but weak in class B, who were treated with exogenous insulin.
From mid to late pregnancy, values of CPR during glucose tolerance tests were significantly elevated, while IRI were slightly and not significantly elevated. At 1 week after delivery, both of IRI and CPR were significantly decreased, particularly in IRI, as compared to late pregnancy. From 1 week to 4 weeks after delivery IRI were slightly but significantly elevated despite of no change for CPR.
The incremental ratios ofserum IRI and CPR to glucose from fasting to 30 minutes after glucose administration were the highest in normal and the lowest in diabetic for pregnant and nonpregnant women.
In diabetic, fasting values of IRG were higher in post partum, but the degrees of these elevations during pregnancy were smaller as compared to normal.
IRI responses following 30 garginine infusion had two peaks at 5min. and 30min. The second peak level was lower in diadetic during pregnancy as compared to nonpregnant normal women. While IRG responses were not different between these two groups. But during and after delivery for same subjects, IRI and IRG responses were higher in 4 weeks and 1 week post partum, respectively.
From the results mentioned above, it is considered that alpha and beta cells of pancreas are loaded with physiological stimulation during pregnancy, which can be well tolerated in normal, but in diabetic tend to decrease the responseness to various metabolic stimuli. Peripheral insulin concentrations are influenced by altered metabolic conditions in liver and tissue during pregnancy. There are some differences of effects on alpha and beta cells between glucose and arginine during pregnancy.

View full abstract

In diabetes mellitus, it has recently been proposed that in addition to absolute or relative lack of insulin, glucagon is the essential for the development of hyperglycemia.
The present study was designed to define the role of insulin and glucagon in regulating gluconeogenesis in dogs. To accomplish this, somatostatin (ST 0.4-0.8μg/kg/min), a potent inhibitor of insulin and glucagon, was infused peripherally alone or concurrently with glucagon in order to produce respectively either acute deficiency of both hormones or of insulin only in normal dogs.
The metabolic effects of acute insulin deficiency in normal dogs were subsequently compared to acute or prolonged insulin deficient depancreatized dogs where non-pancreatic IRG originates exclusively in the gastrointestinal tract and can not be distinguished from pancreatic glucagon using a variety of techniques.
Gluconeogenesis was monitored by tracter (2-H³-glucose). To assess the importance of glucagon in early diabetes, ST was infused for 50min in 5 conscious normal dogs. ST induced sustained decreases in insulin (IRI)(-67±4%) and glucagon (IRG)(-37±4%), glucose concentration decreased temporarily (-8±1%) because glucose production (Ra) fell transiently. This indicates that at low insulin levels IRI and IRG interact on liver. This was corroborated by infusing glucagon with ST, thus decreasing IRI but not IRG. These experiments demonstrate that basal levels of glucagon are essential for the hyperglycemic effect of acute insulin deficiency.
In acutely insulin deficient depancreatized dogs the secretion of gastrointestinal glucagon (G-IRG) was enhanced by arginine (AR, 12.5mg/kg/min) and suppressed by ST, and an increase or a decrease in plasma G-IRG levels induced, respectively, an increase or a decrease in Ra. Thus, G-IRG has a stimulatory effect on hepatic glucose production. The metabolic clearance (M) of glucose did not fall during the time when plasma G-IRG levels were decreased by ST. Indirect evidence that M fell when levels of G-IRG increased. This suggests that, at low insulin levels, G-IRG has an inhibitory effect on Ra. This effect is probably confined to the liver. Acute insulin deficiency, in depancreatized dogs, did not result in an increase in Ra if plasma G-IRG was maintained at low levels. Thus, basal levels of G-IRG are essential for these acute diabetogenic effects of insulin deficiency in depancreatized dogs. The effects of changes in plasma IRG levels in normal and depancreatized dogs indicated that pancreatic and gastrointestinal glucagon are similar with the respect to in vivo action on glucose turnover.
On the other hand, in prolonged insulin deficiency in depancreatized dogs, the secretion of G-IRG was suppressed by ST, but was enhanced a little by ST, and an increase or a decrease in plasma G-IRG levels did not induce respectively, an increase or a decrease in Ra. Thus the stimulatory effect of glucagon (G-IRG) on hepatic glucose production was reduced in prolonged insulin deficiency.
The lack of insulin is clearly a major factor in the abnormality of A cell function (diabetes mellitus). Under this circumstance A cells respones to arginine or glucose stimulation were exaggeracted and as IRG levels increase, blood glucose levels increase.
Thus in addition to lack of insulin, glucagon is the essential for the hyperglycemic effect.

View full abstract

For the object of elucidating the pathophysiology of diabetic coma, patients in this state were placed on continuous low-dose insulin therapy, and changes in blood levels of insulin-antagonistic hormones-glucagon (IRG), growth hormone (h-GH) and cortisol-were pursued in the course of therapy. The subjects for study were complized 9 cases of diabetic ketoacidosis (4 males and 5 females), aged 30 on the average (12-51 years) and 3 patients with non-ketotic hyperosmolar coma (2 males and 1 female), aged 56 on the average (46-67 years). Consciousness disturbance was observed in 7 of the 9 ketotic patients (7/9) and in one of the 3 patients with hyperosmolar coma (1/3). MC-actrapid insulin (0.1 U/kg/hr) was given by continuous intravenous infusion.
For rephydration, saline was chiefly used, being added with glucose and potassium in adequate amounts for prevention of hypoglycemia and hypopotassemia. All clinical symptoms accompanying the coma were abolished in a mean of 16.5 hr, and all the patients recovered the normal state with only one exception, who died, being complicated by DIC.
The duration of intravenous insulin infusion was 18±2 hr (Mean±SEM) for ketoacidosis and 11±4 hr for hyperosmolar coma; the total insulin dose, 94.8±16.5 U/24 hrs for the former and 69.3±12.2 U/24 hrs for the latter; and the total fluid infusion volume, 5.8±0.51/24 hrs for the former and 4.5±1.21/24 hr for the latter.
Plasma-IRG level was 352.9±81.7 pg/ml before the therapy, but 92±33.1 pg/ml at 6 hr of the therapy, thus showing abrupt fall. On admission, however, only 4 cases showed level higher than 500 pg/ml.
Serum h-GH level, which was 6.4±1.6 ng/ml before the therapy was transiently elevated to 13.0±3.1 ng/ml at 2 hr after the start of insulin infusion, but there after it gradually fell, though it was again increased slightly at 24 hr.
Serum cortisol level, which was as high as 461.8±74.4 ng/ml before the therapy, slowly fell after its start, but at 24 hr, it still remained as high as 173.4±18.6 ng/ml. These results suggest the following: 1) The effect of glucagon on diabetic coma may be slight.
2) The effect of cortisol and h-GH may rather be greater.
3) The low-dose insulin therapy was effective for the improvement of abnormal levels of insulin-antagonistic hormons in diabetic coma.

View full abstract

The secretory mechanism of extrapancreatic glucagon (considered mainly as being of gastrointestinal origin) was investigated in depancreatized dogs and a totally depancreatized patient. The basal level of glucagon immunoreactivity (GI, as measured with pancreatic glucagon specific antisera) and of gut glucagon-like immunoreactive materials (gut GLI: calculated as the difference between the values with crossreacting anti serum and GI) were elevated in systemic blood of depancreatized dogs and were reduced by insulin therapy. The secretion of GI was stimulated following arginine infusion but was not by insulin-induced hypoglycemia in depancreatized dogs. The paradoxycal rise of GI was observed during oral glucose tolerance test in a totally depancreatized patient. These findings are similar to the glucagon abnormalities observed in insulin deficient diabetics. Thus, gastrointestinal glucagon may contribute to the overall glucagon abnormalities characteristic of insulin deficient diabetes. However, suppression of GI is not effective for the reduction of blood glucose once an extreme hyperglycemia is brought about by insulin deficiency because blood glucose did not change during somatostatin infusion in depancreatized dogs.

View full abstract

In order to elucidate the significance of extrapancreatic glucagon in diabetes mellitus, experimental and clinical investigations were performed. In the present study, plasma glucagon and total immunoreactive glucagon were measured using an antiserum (G-21) specific for pancreatic glucagon and a non-specific antiserum (G-25), respectively. In a series of experiment with conscious dogs, 30 ml of 10% arginine were administered into the crural vein and blood samples were obtained from the peripheral vein on the other side. Alloxan diabetic dogs revealed not only elevated fasting levels of plasma glucagon but also exaggerated glucagon response to arginine. Insulin treatment in the alloxan diabetic dogs reduced the fasting plasma glucagon levels as well as responsiveness of glucagon to arginine. In the experiments with anesthetized dogs, 60 ml of 10% arginine solution were infused for 10 min before and after total pancreatectomy. Before pancreatectomy, plasma glucagon increased in the normal dogs and alloxan diabetic dogs, whereas that in the alloxan diabetic dogs treated with insulin rose slightly in response to arginine infusion. After total pancreatectomy, plasma glucagon in the portal vein of the normal dogs rose only slightly in response to arginine. In contrast, the alloxan diabetic dogs revealed an increase in plasma glucagon following arginine infusion performed after pancreatectomy. The response of plasma glucagon to arginine was reduced after pancreatectomy in the alloxan diabetic dogs with insulin treatment. Total immunoreactive glucagon increased slightly following arginine infusion before pancreatectomy in the three experimental groups. In the alloxan diabetic dogs, however, it increased markedly in response to arginine infused after pancreatectomy. Here insulin treatment reduced an exaggerated response of total immunoreactive glucagon in alloxan diabetic dogs. In addition, the results obtained in the present study suggest that gut glucagon-like immunoreactivity, which was calculated by subtraction of glucagon from total immunoreactive glucagon, revealed an elevated response to arginine in diabetic states. In two patients with total pancreatectomy and treated with insulin, no increase in plasma glucagon was observed during an arginine infusion test at all, corresponding to the results in the alloxan diabetic dogs with insulin treatment.
From the present experiment it is concluded that alloxan diabetic dogs revealed an increased response of plasma glucagon and extrapancreatic glucagon, which was reduced by insulin treatment.

View full abstract

The aim of this study was to determine whether the unresponsiveness of diabetic A cell to glucose was etiologically connected with the disturbamce of B cell functions or a primary lesion independent of the B cell functions. The relationship between the glucose responsiveness of A and B cells was ilvestigated in 53 diabetics. Responses of plasma glucagon immunoreactivity (ΣΔGI) to oral glucose (OGTT) and intravenous injection of insulin (ITT) were determined as the A cell function. The B cell responsiveness to glucose was assessed by ΣΔIRI or ΣΔC-peptide immunoreactivity (CPR) during OGTT. ΣΔGI was inversely correlated with ΣΔCPR during OGTT (r= -0.36, P=0.05) in 29 diabetics. ΣGI in ITT was well correlated with the respective patient's ΣΔCPR in OGTT (r=0.79, P<0.001), but ΣGI in arginine test (AT) was not (r=0) in 18 insulin-treated diabetics. The A and B cell responses to ITT, AT and OGTT were measured before and 1.5 month (mean) after commencement of treatment in 6 diabetics when their FBS levels fell to nomal levels. ΣΔIRI in OGTT was raised significantly and augmented plasma GI response to AT was normalized after the period of treatment. In addition, af tertreatment, the A cell response to ITT tended to be increased.
It is concluded that the glucose responsiveness of A cells seems to be connected with B cell activity and disturbances of the A cell functions are most likely secorldary.

View full abstract

It is widely accepted that in spontaneous or acquired diabetes mellitus, glucagon secretion is upset. Not only does hyperglycemia after oral glucose loading fail to lower plasma immunoreactive glucagon concentrations (IRG) in diabetics, but also hypoglycemia caused by insulin injection fails to stimulate glucagon secretion. Since these pancreatic alpha cell dysfunctions were not normalized by the conventional insulin injection therapy, they were thought to be the primary defect of diabetes mellitus.
It is so important to elucidate the precise significance of alpha cell hypersecretion in the pathogenesis of diabetes mellitus that we studied the change in IRG responses to 100g oral glucose loads or to meal intakes in diabetics whose blood glucose responses and plasma immunoreactive insulin concentrations (IRI) were simulated the same as in normal subjects with the aid of the artificial beta cell system or with pre-programmable insulin infusion pump.
The artificial beta cell system used in the experiments was originally developed by ourselves, and in this system insulin infusion rate is based on the sum of the prdportional (a component related to blood glucose concentration per se) and derivative (a component related to the rate of change in blood glucose concentration) actions to blood glucose concentration.
In 6 non-obese, adult onset diabetics when insulin was not infused, IRG was not suppressed at all but rose despite pronounced hyperglycemia against 100g oral glucose loads. In those 6 patients, when the normalizations of blood glucose responses were tried with the artificial beta cell system, mean plasma IRI was raised to 58.3μU/ml at 30 min and 61.3μU/ml at 60 min, respectively, and mean insulin infusion amount for 2 hours after the initiation of glucose load was only 4.5 units. Glucose responses to oral glucose loading were thus normalized in all cases. As far as the response in IRG was concerned, insulin administration significantly blunted the glucagon response from the mean pre-stimulated level of 150.0pg/ml to 77.2pg/ml at 90 min (p<0.05) and to 66.0pg/ml at 120 min (p<0.05), about the same response as in non-diabetics.
In 5 insulin dependent diabetics, when subcutaneous insulin injections of their usual regimen were done one hour prior to oral glucose challenges, the IRG response was not suppressed but tended rather to rise, but differences were not statistically significant. In those 5 diabetics who were challenged with a 100g oral glucose under the control of artificial beta cell system, mean IRG values were not significantly different from pre-stimulated value, but in 4 cases significant decreases in IRG after glucose administration were recorded. In one case with high anti-insulin binding capacity, plasma IRG increased during experimental period.
Pre-programmable insulin infusion pump (pre-pro pump) was also developed by ourselves for possible long-term use in diabetics. This pump is open-loop control system, because this does not measure glucose concentrations continuously as the closed-loop control system, i. e. the artificial beta cell, does. But pre-pro pump could infuse insulin according to the memorized insulin infusion rate and pattern.
In the juvenile onset diabetics whose blood glucose concentrations after each meal were remarkably high even with subcutaneous insulin injections, when the rate of intravenous insulin infusion for 2 to 3 hours after each mean was around 5 to 8 times B (B=225μU/kg min), the circadian blood glucose profile of the patients was perfectly normalized. It was revealed that IRG values were significantly lower with pre-pro pump than with subcutaneous insulin injections.

View full abstract

The present investigation was undertaken to clarify interrelationship among A, B and D cells in the islets of Langerhans with special reference to insulin release in vivo and in vitro. Ten islets isolated from Wistar male rats with collagenase method were preincubated in Krebs-Henseleit bicarbonate buffer (pH 7.4) containing 3.3 mM glucose in the gas phase of 95%O₂-5%CO₂ at 37° for 20 minutes, being followed by the incubation in 0.7 ml of Krebs-Henseleit bicarbonate buffer containing various concentration of glucose and anti-somatostatin serum (ASS), antiglucagon serum (AGS) or normal rabbit serum (NRS) for 60 minutes under the similar conditions to those in the preincubation. Besides, blood was sampled before and 30 minutes after the administration of ASS or NRS under pentobarbital anesthesia in rats.
(1) The antiserum used in the incubation studies did not cross-react to the other peptide hormones presumably or definitely existing in the islets than its antigen peptide.
(2) Insulin release from ASS-treated islets was enhanced in the presence of 3.3 and 8.3mM glucose compared with that from NRS-treated islets, whereas in the presence of 16.7mM glucose there was no difference in insulin release between both treatments.
(3) On the contrary, glucagon release from AGS-treated islets was suppressed in 3.3 and 83mM glucose in contrast with that from NRS-treated islets, while in 16.7 mM glucose the suppression disappeared.
(4) Plasma insulin and glucagon levels were elevated in rats 30 minutes after administration compared with those prior to the injection, while plasma glucose revealed no significant changes. On the other hand, plasma insulin and glucagon levels were not elevated after NRS treatment.
It is, therefore, suggested that somatostatin and glucagon play an important role in the regulation of insulin release in the physiologic range of glucose concentration and that somatostatin might be involved in insulin release directly as well as through glucagon suppression.

View full abstract

After an overnight fast, 17 dogs weighing 11-21 Kg received protein-fat meal in a conscious state and three hours after the ingestion, the lymph was obtained through a catheter inserted in the thoracic duct under neuroleptanalgesia. In fasting and postprandial states, the levels of flow rate, triglycerides (TG) concentration, and TG content in the lymph were significantly reduced after cyclic somatostatin infusions at rates of 50ng/min and 5μg/min through the portal and the femoral veins. The attenuated levels of those in fasting were approximately 80-90% of the values in preinfusion period at a physiological infusion rate (50ng/min) and 77-82% at a pharmacological rate (5μg/min), whereas, the levels were 79-90% of those in the preinfusion values at 50 ng/min rate and 56-85% at 5μg/min infusion rate in the postprandial state. The somatostatin infusions at the rates through either the portal or the femoral vein induced almost identical levels of the attenuating effects on the thoracic duct lymph.
These results suggest the possibility that circulating somatostatin from the pancreas or other organs in the portal vein region might have a physiologic influence upon the nutrient entry from the gastrointestinal tract.

View full abstract

In the present study, we examined the direct effect of glucose, especially at physiological levels, catecholamine and tolbutamide on somatostatin secretion from isolated rat pancreatic islets prepared by collagenase digestion.
The rats which had been not fasted prior experiment were anesthetized with Nembutal and the pancreatic islets were isolated by method of Lacy and Kostianovisky with some modification. After the preincubation 10 isolated islets were incubated in Krebs Ringer Bicarbonate buffer 1.0 ml containing 0.1% BAS and 1000 KIE of aprotinin under the gas phase of 95% O₂ and 5% CO₂ for 30 minutes, adding the agents to be studied.
Released somatostatin was determined by radioimmunoassay system utilizing a rabbit antiserum (B-5) obtained following immunization with synthetic somatostatin. Synthetic N^α-tyrosylated somatostatin labelled with ¹²⁵I using the lactoperoxidase method and purified on a Sephadex G-10 column was used as the tracer.
The levels of released somatostatin into medium were above the sensitivity limits of the assay at a glucose concentration of 50 mg/dl (8.4±0.7pg/islet/30 min.). With increasing concentrations of glucose (between 50 and 400 mg/dl), there is a progressive increase in the rate of somatostatin secretion which reaches a maximum rate at 300 mg/dl, about 3 times the basal output. DL-Epinephrine (0.1μM) did not have a significant effect on glucose induced somatostatin secretion, although it significantly inhibited glucose induced insulin release. The inhibitory effect of epinephrine on insulin release was removed by addition of phentolamine, but somatostatin secretion was not affected. Tolbutamide significantly enhanced somatostatin release at glucose concentrations of both 50 and 200mg/dl, as well as the expected release of insulin.
These results indicated that, like insulin, pancreatic somatostatin might respond at approximately 100 mg/dl of glucose and above, perhaps directly stimulated by glucose and exert an important physiological role in glucose homeostasis, that tolbutamide induced a somatostatin secretion both at low and high glucose concentration along with an insulin secretion and might exert as antihyperglycemic action, in part, not only through the insulin release but also through the secretion of somatostatin. The effect of catecholamine on pancreatic somatostatin secretion remained to be elucidated.

View full abstract

Glycogen storage disease Type I, which is called von Gierke's disease, are now divided in two groups. The most patients with Type I show an absence of glucose-6-phosphatase (G6Pase) in liver specimens biopsied (Type Ia). A few patients who are indistinguishable from Type I in clinical symptoms and biochemical findings reveal normal activity of hepatic G6Pase (Type Ib). The defect in Type Ib are unknown. But the accumulation of hepatic glycogen, poor response of blood glucose to glucagon and the result of glycerol loading test in such patients suggest that G6Pase demonstrated in vitro is not functional in vivo.
A new variant of Type I glycogen storage disease which is probably due to a defect in G6P transport system is described.
The patient Y. S. was born in 1975 and a girl of healthy, unrelated parents. The pregnancy and delivery were uneventful. She showed frequent episodes of hypoglycemia at fasting and consequently growth was retarded.
At the age of 10 months, the patient showed all the features of glycogen storage disease Type I, such as hepatomegaly, growth retardation, doll-like appearance and rapid development of hypoglycemia after only 2 hours fasting, no response to glucagon, and a failure of the blood glucose to rise in response to administered galactose or fructose.
Liver biopsy demonstrated an accumulation of glycogen (8.5% of wet weight) with normal structure. Enzyme analysis in the liver specimen, stored at -25° after its collection, revealed normal activities of glucose-6-phosphatase (4.5μmoles/min./g. liver) and fructose diphosphatase.
Therefore, the patient was diagnosed as glycogen storage disease Type Ib.
This studies were performed, with use of liver specimen obtained at the operation of mesenterico-caval shunt at the age of 3 years.
Liver specimens immediately after biopsies were homogenized with 0.25M sucrose solution and centrifuged to remove nuclei and cellular debris. The supernatants were divided in two portions: one is treated with neutral deoxycholate (final, 0.2%) for two hours in cold to destroy microsomal membrane and the other is supplemented with distilled water, as described by Nordlie. G6P phosphohydrolase activities were measured in the homogenates with or without deoxycholate according to the procedure of Nordlie and Anon.
The activity in the patient Y. S. was found to be markedly low (0.8μmoles/g. liver) as compared with those in control subjects and rats, when the assays were made in the absence of deoxycholate. However, the treatment of liver homogenates with detergent produced a restoration of the activity up to normal in the patient Y. S.(10.2μmoles/g. liver).
The results from patients Y. S. suggest that a sufficient activity of G6Pase presents in microsomes of the liver but must be masked in intact microsomes, probably also in vivo as judged by the clinical and laboratory findings. The masking phenomenon observed in intact microsomes of the patient Y. S. can be explained by a lack of the G6P transport system, because the treatment with detergent brought the restoration of G6Pase activity.
Our earlier observation, that the activity of hepatic G6Pase in patient Y. S. was normal, may be explained by a decrease in the latency of the enzyme during the storage at-25° which may destroy microsomal membranes. In fact, we have observed a decrease in the latency of mannose-6- phosphate phosphohydrolase, which is thought to be an index of microsomal integrity by freezing the liver specimens.
This variant could be diagnosed only when the enzyme assay was carried out using both intact and disrupted preparation of microsomes.

View full abstract

The results of comparative enzymatic studies between tissues and peripheral blood samples from 25 patients with hepatic glycogenosis and 2 patients with generalized glycogenosis are described.
The enzymatic studies included determination of glycogen contents in tissues and erythrocytes, assay of glucose-6-phosphatase activity in liver and plateletes, amylo-1, 6-glucosidase activity in liver and leucocytes, or α-1, 4-glucosidase activity in skeletal muscle and leucocytes.
Twenty one patients were hepatic glycogenosis associated with deficiency of glucose-6- phosphatase, amylo-1, 6-glucosidase or hepatic phosphorylase system.(Eleven cases: type I, 6 cases: type III, 4 cases: type VI or VIII)
Two patients were generalized glycogenosis associated with deficiency of α-1, 4-glucosidase. Enzyme defect could not be detected in two patients who have excessive glycogen accumulation in the liver. For diagnosis of type I, determination of liver glucose-6-phosphatase is essential, and plateletes glucose-6-phosphatase assay is not suitable.
Despite the amylo-1, 6-glucosidase activity was markedly reduced in the liver tissue, normal activity was found in the leucocytes from one patient with type III. From the above results, the enzymatic diagnosis of glycogenosis using peripheral blood samples seems to have some limitation.

View full abstract

The control mechanism of glycolysis in human erythrocytes was analysed placing an emphasis upon the relationship between phosphofructokinase (PFK) activity and 2, 3-bisphosphoglycerate (2, 3-DPG) level. The blood sample was directly taken into a syringe containing the perchloric acid in order to minimize the changes in metabolite levels after venopuncture. Alterations in glycolysis in erythrocytes in the patients with hereditary eythrocyte PFK deficiency (PFK activity was 30% of normal control), diabetic ketoacidosis and chronic renal failure in acidosis were investigated by the enzymatic analysis of glycolytic intermediates and adenine nucleotides. The following results were obtained.
1) In PFK deficiency, the levels of glucose-6-phosphate (G6P) and fructose-6-phosphate (F6P) were increased, while those of fructose-1, 6-bisphosphate (FDP), dihydroxyacetone phosphate (DHAP) plus glyceraldehyde-3-phosphate (GAP) and 2, 3-DPG were decreased.
2) In diabetic ketoacidosis, the level of F6P was significantly higher than normal control and levels of FDP, DHAP plus GAP, 2, 3-DPG and 3-phosphoglycerate were significantly lowered. In chronic renal failure, the similar patterns in the levels of glycolytic intermediates were demonstrated when the levels were compared with those of the non-azotemic patients with comparable anemia and reticulocytosis.
These findings indicate that glycolysis in erythrocytes in these genetic and aquired diseases is inhibited at PFK step. It was demonstrated in experiments using human erythrocyte suspension and purified PFK preparations that the inhibition in acidosis was mainly due to the increased hydrogen ion concentration. Thus, a reduction in erythrocyte PFK activity is caused not only by the hereditary defect of enzyme protein but also by the elevated concentration of hydrogen ion in blood. The marked fall in 2, 3-DPG level is caused by the reduction of PFK activity in erythrocytes, indicating an important role of this enzyme in the regulation of the 2, 3-DPG level.

View full abstract

We have experienced an adult case who had been attacked by frequent hypoglycemia on fasting. The onset of this illness was the middle of July 1976 (46 years old). Resection of the body and tail of the pancreas with the splenectomy was done under the diagnosis of insulinoma February 2, 1977. The operation revealed no insulinoma and after the operation hypoglycemia appeared as frequent as before. Oral glucose tolerance test, arginine infusion test or glucagon loading test revealed no hyperinsulinemia. Binding antibody to insulin or abnormal sized insulin was not found in the patient plasma. Oral glycerol loading (50g) could not raise the lowered blood glucose level in fasting state. Biopsied liver showed ballooning of hepatic cells by large lipid containing vacuoles. Liver glycogen was normal. The activity of hepatic fructose-1, 6-diphosphatase of this patient was 0.706±0.135 pmoles FDP/g/min (pH 7.5) which value was lower than those of control livers (1.621±0.802).
Oral administration of folic acid (30 mg daily) was started June 10, 1978. This treatment improved the illness and the patient continues this therapy with almost normal social activity.
The hypoglycemia in this case is thought to be induced by the partially impaired activity of hepatic fructose-1, 6-diphosphatase.

View full abstract

A novel sensitive fluorescence reaction of carbohydrates was devised using ethanolamine and borate as reagents. Application of this reagent to the post column labelling of carbohydrates after separation on high performance liquid chromatography provided a simple tool for the detection and determination of reducing sugars in biological samples. This method was also effective in the separation of oligosaccharides. Since this reaction proceeds without strong acid or alkali, the chromatography does not require special equipments. The present method proved to be 100 times as sensitive as refractometric analysis and highly reproducible. This procedure is expected to be useful in the research on the metabolic deficiency of carbohydrates.

View full abstract

Galactose bears a close resemblance to glucose in the molecular structure. It is stereochemically interesting to compare the changes of metabolites during galactose tolerance test with the those during GTT. We carried out 40g oral and intravenous galactose tolerance tests in the normal, diabetic and hepatopathic subjects. In the normal subjects, the galactose led to a slight and transient elevation in blood glucose concentration, whereas in the diabetic and hepatopathic cases, it produced a persistent elevation in the blood glucose level. At that time increases of the glucose excretion in urine were observed at 2 hours after the administration without the significant elevations over the threshold value of blood glucose. The monosaccharide transports in the kidney have been reported to be related to the stereospecificity of molecular structures. Following the interaction between glucose and galactose, galactose was absorbed through the membrane in preference to glucose, and then glucose was excreted in urine. Plasma NEFA concentrations, on the other hand, diminished markedly after the administration irrespective of the levels of the blood galactose retention. However the decreases of the blood NEFA levels at 2 hours after the loading was more dominant in GTT than in the galactose tolerance test. No significant increases in the blood IRI levels were observed in either the oral or the intravenous galactose tolerance tests.
The galactose tolerance test is considered, in other words, to indicate the degree of hepatic dysfunction at the redox state. The redox potentials in the cytoplasma of hepatic cells has been described to have the significant influence upon the metabolism of galactose. According to our studies, the transient significant elevations in the blood lactate and pyruvate levels were observed after the administration in the normal group of the galactose tolerance tests. In abnormal groups, however, the elevation in the blood lactate and pyruvate levels were not so significant. During the galactose loading the significant positive correlations between the changes in the blood lactate and pyruvate levels were found in the normal subjects and diabetics, but no correlation in the hepatic disorders. While the significant negative correlations between the changes in blood lactate and NEFA levels were recognized only in the abnormal groups of the galactose tolerance tests. The fact that the abnormality in galactose tolerance was noted not only in cases of hepatic disorder but also in cases of diabetes suggests a possible involvement of NADH metabolism. As shown in our studies, the lactate-pyruvate system will be considered to play an important part in the conversion of the redox state, and when the regulation by this system was disturbed, the other system concerning the fatty acid metabolism might be connected with the regulation of them.

View full abstract