Showing papers in "Hormone and metabolic research. Supplement series in 1988"

In vivo characteristics of needle-type glucose sensor--measurements of subcutaneous glucose concentrations in human volunteers.

Abstract: To explore the clinical performances of the needle-type glucose sensor, in vivo monitoring was conducted in human subjects. The tissue glucose concentrations measured by glucose sensor were lower than the plasma glucose concentrations by 15%, but a good linear relationship was observed between the tissue and plasma glucose concentrations. The sensing sites between abdomen and forearm did not affect the tissue glucose monitoring. The tissue glucose concentrations after meal intake showed 5 min delayed response to the blood glucose concentrations. After 3 days' continuous monitoring, the "relative" sensor output to the plasma glucose concentration decreased by 26% and the "relative" response time to reach peak value prolonged from 5 min to 13.5 min. These data indicate the clinical usefulness of the needle-type glucose sensor.

Lipolysis in human adipocytes, effects of cell size, age and of regional differences.

Abstract: In man only catecholamines and insulin have pronounced and acute effects on lipolysis in fat cells. In obesity the effects of these hormones seem to be normal or even increased. When the rate of lipolysis is expressed per cell there is a strong association between hormonal effect and cell size. Indicating that catecholamines and insulin may be involved in the regulation of adipocyte volume. There are, however, site variations in the effects of the regulatory hormones, which may be of importance for the development of various regional forms of obesity. The lipolytic effect of catecholamines is more pronounced in the abdominal than in the femoral/gluteal subcutaneous fat depots, partly owing to an increased alpha 2-adrenergic receptor mediated antilipolytic effect of catecholamines in the latter region. This alteration in lipolysis favour accumulation of fat in the femoral/gluteal region and may be of importance for the development of the female type of obesity. Furthermore, in omental fat cells the lipolytic activity is higher than in subcutaneous fat cells owing in part to less marked insulin action and lower alpha 2-adrenergic receptor mediated antilipolytic effect of catecholamines. These alterations may cause elevation of the free fatty acid levels in the portal blood so that the handling of glucose and insulin is impaired in the liver which may be one mechanism behind the increased risk to develop cardiovascular complications in the male type of obesity. Age also influences hormone-induced lipolysis. Catecholamine resistance is observed at birth and at the latest stages of life. Insulin resistance is observed in old as compared to middle-aged subjects.

Differentiation of adipocyte precursor cells from obese and nonobese adult women and from different adipose tissue sites.

Abstract: We studied adipocyte development in primary cultures of stromal-vascular cells from the adipose tissue of obese and nonobese adult women. Adipose conversion was promoted by addition to confluent cultures of 10(-9) mol/l insulin, 10(-7) mol/l cortisol and for the first two days 0.5 mmol/l 1-methyl-3-isobutylxanthine (MIX). Presence of MIX for a longer period resulted in an increasing differentiation rate. No significant difference in the number of developing fat cells was observed between cultures from 8 obese (body mass index 33.6 +/- 4.1 kg/m2) and from 6 lean (body mass index 22.3 +/- 2.4 kg/m2) adults (390 +/- 102 vs. 354 +/- 85 fat cells/cm2, n.s.). The number of new fat cells obtained in cultures from the subcutaneous fat depot was significantly higher as compared to those obtained in cultures from the intraabdominal fat region (360 +/- 68 vs. 79 +/- 35 fat cells/cm2, p less than 0.01), suggesting the existence of regional variations in the differentiation capacity of adipocyte precursor cells.

A potentially implantable enzyme electrode for amperometric measurement of glucose.

Abstract: A membrane limited amperometric enzyme electrode suitable for glucose measurement in biological fluids was developed. The sensor consists of a central platinum wire (0.3 mm) surrounded by a stainless steel tubing (0.8-1.0 mm outer diameter; 2-4 cm length). By successive dipcoating procedures, layers from cellulose acetate, glucose oxidase (crosslinked with glutaraldehyde) and polyurethane are placed on its surface. The platinum is polarized at +700 mV against steel. In vitro results: Electrodes are stable for at least 6 days. They exhibit a linear range extending to 500 mg/dl glucose. Response times are less than 100 sec. The sensors are not dependent on stirring and are relatively insensitive to changes of pH. Dependency of glucose measurement upon dissolved oxygen is negligible at oxygen concentrations above 0.5 mg/l. In vivo results: Preliminary studies in sheep using subcutaneously implanted needles indicate that short term glucose monitoring is feasible.

Visceral obesity and diabetes.

Abstract: The cross-sectional subcutaneous and visceral areas of adipose tissue, measured on computed tomographic scans at abdominal level and the subcutaneous/visceral fat area ratio (S/V ratio) distribution have been calculated in 155 subjects. Using the S/V ratio as discriminator, two groups of patients of both sexes, with subcutaneous or visceral type of obesity have been selected. Metabolic studies demonstrate a significant reduction of glucose tolerance at the oral glucose load and a significant increase of the insulin response in patients with visceral obesity. The size of the omental fat cells was larger in visceral obesity, while the size of subcutaneous fat cells was larger in subcutaneous obesity. A significant correlation was found between the S/V area ratios and the subcutaneous/visceral fat cell weight ratios. In vitro studies on subcutaneous and omental fat obtained during surgery from 9 patients with subcutaneous obesity and 7 patients with visceral obesity demonstrate a significantly higher isoprenaline-stimulated lipolytic response in omental fat of patients with visceral obesity. The lipoprotein lipase activity in subcutaneous fat was higher than in omental fat in both types of obesity.

Progress towards in vivo glucose sensing with a ferrocene-mediated amperometric enzyme electrode.

Abstract: A range of miniature, amperometric enzyme electrodes have been constructed using the principle of mediated transfer of electrons. Ferrocene, or one of its derivatives, can be used in conjunction with immobilised glucose oxidase to produce sensors which are linear to at least 20 mmol/l glucose, have fast response times, are relatively oxygen insensitive and are suitable for implantation studies. In preliminary experiments in non-diabetic pigs, electrode responses followed blood responses, although mean tissue levels were about one fifth of those in blood. The ferrocene technology thus appears to be suitable for adaptation to the construction of implantable glucose electrodes for use in diabetic man.

Issues related to in vitro operation of potentially implantable enzyme electrode glucose sensors.

Abstract: The operation of enzyme electrode glucose sensors based on oxygen, hydrogen peroxide, and electron carrier molecules is reviewed. Issues that are most effectively addressed by in vitro experimentation are identified. An understanding of these issues is prerequisite for unequivocal interpretation of in vivo results.

Pharmacokinetics of biosynthetic human proinsulin following intravenous and subcutaneous administration in metabolically healthy volunteers.

Abstract: 40 pmol of biosynthetic human proinsulin was administered to 8 healthy volunteers by intravenous and by subcutaneous route. Following proinsulin administration, venous blood was collected in regular intervals within which proinsulin was determined by a specific radioimmunometric assay with monoclonal antibodies. The proinsulin concentration was determined simultaneously with the insulin and C-peptide radioimmunoassay. Through this investigation the following kinetic parameters were found: The kinetics of the biosynthetic human proinsulin can be best described by the 3-compartment model. The dominant biological half-life was 92 minutes. In intravenous proinsulin administration a proinsulin mean transit time of 80 minutes was found, whereas in subcutaneous administration a proinsulin retention time of 225 minutes was measured. The mean resorption velocity of the subcutaneously applied proinsulin amounted to 145 minutes. Two lag times for subcutaneous resorption can be described, a short one with 9.4 minutes and a long one with 65 minutes. The initial distribution volume for proinsulin was 3.8 l, whereas the distribution volume after complete distribution was 9.3 l. The mean total metabolic clearance was determined with 120 ml/min. Since no difference for the proinsulin concentration was found using the 3 different determination methods a peripheral proinsulin conversion to insulin and C-peptide is not likely. The basal endogenous secretion rate for proinsulin is 68.7 pmol per hour.

Kinetics of biosynthetic human proinsulin in patients with terminal renal insufficiency.

Abstract: Eight volunteers with terminal renal insufficiency having consented to the investigation, were given an i.v. bolus administration of 40 pmol biosynthetic human proinsulin on their dialysis-free day. Intravenous blood for the determination of blood glucose proinsulin, insulin and C-peptide was collected in short intervals for 6 hours and thereafter in longer intervals for 24 hours. Proinsulin was determined by immunoradiometric assay with monoclonal antibodies. The proinsulin kinetics were compared with the kinetics of normal volunteers. The behaviour of proinsulin concentration-time is best described with a 3-compartment model. The dominant biological half-life in terminal renal insufficiency was 6.8 hours which signifies a 4.4-fold increase of the normal half-life. The distribution volumes (V1) in the central compartment do not differ in the two groups, whereas the distribution volume after complete distribution (Vss) is significantly increased in renal insufficiency. The total metabolic clearance in renal insufficiency namely 0.63 ml/kg/min is 2.6 times lower compared to normal subjects with 1.67 ml/kg/min. The extra-renal clearance is 39% of the total metabolic clearance rate, whereas the renal clearance comprises 61%. Peripheral conversion from proinsulin to insulin and C-peptide does not occur in terminal renal insufficiency. The basal endogenous proinsulin secretion rate in renal insufficiency does not differ from that of normal volunteers. The following conclusions can be drawn: 1) Hyperinsulinism observed in renal insufficiency can be explained by circulating proinsulin. 2) In the potential therapeutic use of biosynthetic human proinsulin in diabetics with renal insufficiency dosis adjustment according to the remaining renal function would probably be required.

Are regional metabolic differences of adipose tissue responsible for different risks of obesity

Abstract: Recent research has shown that statistical correlation of disease is usually stronger to abdominal distribution of adipose tissue than to obesity. Abdominal distribution of adipose tissue in women is associated to other male characteristics of muscle tissue mass and morphology, as well as signs of androgenicity in circulating hormones. Abdominal adipose tissue is sensitive to lipolytic stimuli. It might be considered that the endocrine aberrations in combination with elevated concentrations of circulating free fatty acids might cause the complications associated with abdominal obesity.

The coated wire electrode glucose sensor.

Abstract: Previous techniques for sensing and measuring glucose are briefly reviewed and compared. The genesis of the glucose coated wire electrode lies in liquid membrane technology, and the basic principle of the electrode is described. The electrode utilizes a quaternary ammonium salt with a sparingly soluble metallic salt of glucose, in a matrix such as poly(vinyl chloride). Fabrication procedures are detailed. Typical sensor responses are illustrated with experimental data. The sensor generally shows a voltage that decreases linearly with increasing glucose concentration over approximately 40 mg/dl to 200 mg/dl. The time response of the sensor is of the order of seconds, and the lifetime observed to date is of the order of several months. Initial studies of mechanism are discussed, using liquid NMR. Preliminary interference effect studies to date are summarized, in which common compounds such as ascorbic acid, uric acid, L-cystine, glycine, and bilirubin were used. Singly these compounds show no significant interference effects.

The GOD-H2O2-electrode as an approach to implantable glucose sensors.

Abstract: The long-term function of glucose sensors under in-vivo conditions is the first prerequisite in the development of glucose-controlled insulin infusion systems. Amperometric enzyme glucose sensors measuring H2O2 from the enzymatic glucose oxidation are defeated to certain poisoning effects of the electrochemical system and to alterations of the diffusional properties, both resulting in unpredictable changes of the sensor signal. To interpret the signal of subcutaneously implanted sensors after reaching a certain balance of the altering processes, the wick-technique has been established as an independent reference method applicable in the interstitial fluid. Based on the comparison of the wick-glucose concentration and the circulating plasma glucose concentration the in vivo-calibration of the sensor current according to plasma glucose concentration is proven.

The glucose enzyme electrode: is simple peroxide detection at a needle sensor acceptable?

Abstract: Needle type devices have been fabricated based on Pt anodes (to detect H2O2) mounted with stainless steel needles to act as the reference. Coating of these devices with glucose oxidase allowed glucose measurement, but with high dependence on stirring and background pO2 levels as well as a restricted glucose assay range. By wet dipcoating of microporous polyurethane or the application of preformed porous membranes, the linear range has been extended up to 70 mM glucose, to give minimal pO2 dependence and insensitivity to stirring. With incorporation of polyethersulphone membranes, blood measurement was possible with high selectivity (y = 0.954x + 0.202, r = 0.991, n = 48). This establishes that simple peroxide detection at a needle sensor is acceptable, and can now be focussed on increasing biocompatibility.

A glucose sensor utilising tetracyanoquinodimethane as a mediator.

Abstract: This paper describes an amperometric enzyme electrode for glucose analysis. The electrode utilised the mediator tetracyanoquinodimethane (TCNQ) to facilitate electron transfer from glucose oxidase to a pyrolytic graphite carbon electrode. The electrode demonstrated a response to glucose in a clinically relevant range (0-70 mM). Results concerning the direct current cyclic voltammetry of TCNQ and the electrodes' response to glucose, pH profile and the effect of temperature are presented.

Effect of different APO A-I containing lipoprotein particles on reverse cholesterol transport in fat cells.

Abstract: Cholesterol efflux was studied in cultured mouse adipose cells after preloading with LDL cholesterol. Long term exposure to LpA-I particles isolated from HDL2 and HDL3 promoted cholesterol efflux while LpA-I:A-II particles isolated from both fractions did not. However, LpA-I and LpA-I:A-II particles isolated from both HDL subfractions were effective in competing for the binding of 125I-HDL3 to apo A-I/A-II cell surface receptor sites. These results underline the role of LpA-I particles in the promotion of cholesterol efflux and suggest that the concentrations and/or the relative proportions of LpA-I and LpA-I:A-II particles might be critical for the regulation of cholesterol efflux from adipose cells.

The electrocatalytic glucose sensor.

Abstract: An electrocatalytic glucose sensor is a very useful component of an artificial beta cell in controlling the insulin dosage in the diabetes therapy. The advantage is its potential long-term stability and the disadvantage is its lack of selectivity. The principle and experimental results with different electrolytes will be described. In saline solution the mean deviation in glucose response is up to 15% if the amino acid and urea concentrations raise from minimum to their physiological maximum values.

Closure of the loop by glucose sensing. Physiological and practical considerations.

Abstract: This review considers factors which determine the minimum desirable operational function of a glucose sensor, based on a consideration of the meal, basal and exercising phases of metabolism in the insulin-dependent diabetic. Augmented insulin release ahead of, or in synchrony with, the arrival of nutrients in the blood stream is important in the normal regulatory mechanism associated with meals. Thus glucose sensors are not suited to be the sole means of determining insulin delivery during meals, and this phase of metabolic control may best be handled by open-loop insulin delivery. However during other phases of metabolic control, glucose sensors with modest response times may have a very useful clinical function.

Qualitative dissimilarities of insulin and proinsulin binding and action in vitro at IM 9-lymphocytes.

Abstract: To prove hormone analogy of insulin and proinsulin 1. biochemical characterization of binding, 2. competition-inhibition-assay, 3. up-or-down-regulation, and 4. 3H-thymidine incorporation in IM 9-lymphocytes in vitro were applied. Marked differences in binding and biological activity of insulin and proinsulin in the physiological range (10 pM-1 nM) are obvious. Heterogeneity of specific insulin and proinsulin binding sites is suggested.

Biosynthetic human proinsulin, a new therapeutic compound for diabetics? A comparative study of biosynthetic human proinsulin with biosynthetic human insulin.

Abstract: The biological activities of proinsulin, the endogenous precursor of insulin are different in comparison to insulin. Proinsulin has a longer biological half-life in serum and a much lower hypoglycemic potency. Biosynthetic human proinsulin has been suggested as an intermediate acting "insulin" in the treatment of diabetes mellitus, having the advantage of not including a retardation substance. For this reason it is important to establish a dose relationship in comparison to human insulin. Human proinsulin and human insulin were tested in 2 groups of healthy volunteers after i.v. application on a unit equivalent basis. 0.025, 0.05 and 0.1 units/kg of either compound were applied intravenously in a randomised order. Following human proinsulin application, the nadir of blood glucose showed a delay of 5-10 min. in comparison to insulin, however, the increase of blood glucose following the nadir was much more retarded after human proinsulin application. The antilipolytic effect of proinsulin was significantly stronger in comparison to insulin. No significant differences could be observed in the kinetics of beta-hydroxybutyrate levels and the counterregulatory response. Human proinsulin, based on a unit equivalent dosage, exerts a more pronounced hypoglycemic and antilipolytic effect, which is probably due to the longer biological half-life in comparison to insulin.

Effects of single and combined infusions of human biosynthetic proinsulin and insulin on glucose metabolism and on plasma hormone concentrations in euglycaemic clamp experiments.

Abstract: Euglycaemic clamp experiments with single or combined infusions of human biosynthetic proinsulin and insulin were performed in 6 healthy, normal weight subjects in order to assess the possibility of antagonistic, additive, or synergistic effects on whole body glucose metabolism. Insulin (i: 2.02 pmol x kg-1 x min-1) or proinsulin (p: 9.26 pmol x kg-1 x min-1) were infused under euglycaemic clamp conditions over 240 min. After 120 min, the infusion rates were doubled (protocols ii and pp), or proinsulin (protocol ip) or insulin (protocol pi) infusions were added. After 240 min of infusing insulin or proinsulin alone, euglycaemia was maintained for an additional 60 min period without hormone infusions to measure the decay of hormone concentrations and of effects on glucose metabolism. Effects on glucose uptake were measured as the glucose infusion rate necessary to maintain euglycaemia. IR-insulin, IR-proinsulin, IR-C-peptide and IR-glucagon were determined by specific radioimmunoassays. After 240 min, similar steady state glucose infusion rates were reached for all protocols (mean +/- SEM, mg x kg-1 x min-1: ii: 10.6 +/- 1.0; ip: 9.1 +/- 0.4; pi: 10.0 +/- 0.9; pp: 8.4 +/- 0.7). The infusion rate with proinsulin alone (pp), however, was significantly smaller than with insulin alone (ii), indicating a somewhat lower effectiveness of the proinsulin dose employed. With all protocols, nonesterified fatty acid and IR-glucagon concentrations were decreased to a similar extent. Steady state hormone concentrations were reached within 30 min of each infusion period.(ABSTRACT TRUNCATED AT 250 WORDS)

Evaluation of implantable glucose enzyme-based sensors with extracorporal blood shunt.

Abstract: This paper presents several methods and instruments developed for the characterization of enzymatic membranes and enzyme electrodes: 1. amplitude and response time of steady-state and transient responses to glucose, automatically determined with an Apple II based workstation, 2. enzyme electrode response to in vitro simulated I.V.G.T.T. and to in vivo I.V.G.T.T. when connected to an extracorporal blood shunt of conscious rats.

An amperometric glucose sensor with combined enzyme layers.

Abstract: An amperometric needle-type glucose sensor consisting of a silver tube as the cathode and a platinum wire as the anode was developed The sensor was provided with two enzyme layers (glucose-oxidase and catalase) separated by a polyurethane membrane The catalase caused an accelerated decomposition of H2O2 generated during the enzymatic oxidation of glucose In this way the lifetime of the sensor and the oxygen dependence of the analytical signal could be improved

Very low density lipoprotein. Qualitative abnormalities in patients with premature coronary heart disease and in patients with insulin dependent diabetes.

Abstract: VLDL composition has been investigated in two groups of patients: one with Premature Coronary Heart Disease (PCHD) and another one with Insulin Dependent Diabetes Mellitus. Both groups showed abnormal VLDL composition. PCHD patients had VLDL significantly enriched in free cholesterol and triglycerides in comparison with a group of controls matched with patients for sex, age, BMI, serum cholesterol and TG. This VLDL composition indicates a prevalence of big particles, which can be found at the beginning of the lipolytic cascade. On the other hand VLDL of diabetic patients were enriched in cholesterol, especially esterified cholesterol, and apo B in comparison with the respective control group suggesting an increase of smaller VLDL, considered more atherogenic. In conclusion these two studies demonstrate that VLDL compositional abnormalities can be found in two different groups of patients even in the absence of hyperlipidemia.

Investigation of the prolonged hypoglycemic effect of human proinsulin in diabetes mellitus.

Abstract: Human proinsulin possesses a clear hypoglycemic effect, the extent and kinetics of which were investigated in 8 insulin dependent diabetics. In addition, to adequate therapy with a controlled-release insulin, each patient received i.v. injections of insulin (e U) and proinsulin (9 U) on separate test days. After 9 U of proinsulin, the blood glucose decrease was delayed in comparison to 3 U of insulin. However, it reached identical maximum blood glucose decrease values. The elimination kinetics of proinsulin are delayed compared to insulin. There is a linear relation between the time and the reciprocal value of the measured insulin immuno-reactivity. The hypoglycemic effect of proinsulin which is protracted for hours corresponds to that of an extreme controlled-release insulin. Under certain conditions, proinsulin might therefore prove effective in the long-term treatment of the insulin-dependent diabetes mellitus.

On the in vitro lipolytic activity of peptide hormones in human adipose tissue.

Abstract: Peptide hormones are potent stimulators of lipolysis in incubated adipocytes or adipose tissue of laboratory animals. However, none out of 40 synthetic or highly purified peptide hormones or sequences from the pituitary, the gastrointestinal tract and of different origin was able to stimulate glycerol release from 'incubated' human adipocytes. Yet, this model of the 'incubated adipocyte' offers several disadvantages: rapid decrease of pH in the incubation medium, release and increment of lipolysis inhibiting substances such as FFA and adenosine and rapid unspecific degradation of the peptides. In an improved in vitro model--pH-stat titration--which avoids or diminishes the disadvantages of 'incubated adipocytes', a group of pituitary peptides was tested. One peptide hormone, beta-lipotropin, stimulated lipolysis significantly in human adipose tissue. Two other peptides which were highly active in adipose tissue of laboratory animals were ineffective in this system too. The lipolytic response to beta-lipotropin was comparable to the effect of noradrenaline at equimolar concentrations. The data indicate that results on the lipolytic activity of peptide hormones in adipose tissue of laboratory animals cannot be transferred to human adipose tissue. In advanced in vitro test systems the ability to stimulate lipolysis could also be shown for a peptide hormone suggesting a role of such substances in the endocrine regulation of adipose tissue mass in men.

Substitution of basal delivery of insulin by proinsulin in type I diabetic patients under CSII.

Abstract: In seven C-peptide negative type I diabetic patients conventional insulin therapy was replaced by CSII using biosynthetic human insulin. After a pretest period of 1-3 days duration, the patients received in randomized order basal subcutaneous infusion of either biosynthetic human proinsulin (assumed potency 4 U/mg) or insulin for two days each. For meals, insulin was given throughout the study. Insulin doses were adjusted to the patients requirements during the prephase. Afterwards changes of basal rates were allowed only in case of nocturnal hypoglycemia; bolus doses were modified when premeal glucose concentrations were outside the range from 80 to 120 mg/dl. During the test period with basal proinsulin infusion, plasma glucose control (MBG, MAGE, M-value) was significantly better (P less than 0.05) when compared to periods with basal insulin infusion. Basal rates of insulin and proinsulin as well as bolus doses of insulin were similar at all study periods. It is concluded that subcutaneous basal proinsulin infusion, supplemented by subcutaneous premeal insulin administration, can be used for glucose control of patients with type I diabetes. As basal proinsulin delivery gives significantly better results than basal insulin delivery, the question arises whether the present formulation of proinsulin is equipotent to insulin formulations.

Metabolic effects of intravenous proinsulin.

Abstract: Proinsulin induced hypoglycemia was characterized in eight healthy male volunteers. Proinsulin cleared slower from the circulation than insulin. The metabolic effects on plasma glucose, free fatty acids, glycerol, 3-hydroxybutyrate, potassium, and phosphate occurred slower. The anti-lipolytic effect of proinsulin was longer than that of insulin (2P less than 0.001). The hormonal responses of epinephrine, norepinephrine, prolactin, hGH, and cortisol were attenuated following proinsulin. The amount of epinephrine secreted during counterregulation and the amount of lactate produced in response to beta-stimulation were both correlated to the fall of plasma glucose (2P less than 0.005). Stress symptoms were milder after proinsulin (2P less than 0.01). The data obtained are consistent with the hypothesis that the slower kinetics of proinsulin cause slower metabolic effects. The assumption of non-insulin-like effects of a differing pattern of insulin-like effects is not necessitated by the results of this study.

Human proinsulin: bioactivity and pharmaco-kinetics after intravenous and subcutaneous administration.

Abstract: The hypoglycemic actions of biosynthetic human proinsulin and human insulin (1 IU insulin/kg b.w. and about equimolar amounts of proinsulin) were studied after intravenous and subcutaneous injection in rabbits. Blood samples were taken up to four hours after injection for the determination of blood glucose and immunoreactive levels of both insulin and human C-peptide. For the determination of human C-peptide, serum taken after proinsulin injection was divided into two fractions. One was examined directly by the human C-peptide radioimmunoassay and the other after incubation with a protein-A-sepharose coupled insulin antibody to find "free human C-peptide". In amounts equimolar to 1 IU insulin/kg b.w., proinsulin exerted an about one third stronger hypoglycemic action (area under curve estimation) after s.c. compared to i.v. injection. Proinsulin-induced hypoglycemia did not last longer after intravenous administration than that induced by intravenous insulin. Although subcutaneous proinsulin did not show the same maximum decrease of blood glucose as subcutaneous insulin, its action was significantly prolonged (up to 180 min). Specific measurement of free human C-peptide showed no evidence of conversion of proinsulin to insulin and C-peptide.