J. Gliemann

A procedure for measurement of distribution spaces in isolated fat cells.

1. 1. Aliquots of homogenous suspensions of isolated rat epididymal fat cells were rapidly transferred to polyethylene tubes containing an oil lighter than buffer and heavier than fat cells. After centrifugation for 30 s the tube was cut through the oil layer which clearly separated the buffer from the cells. 2. 2. 5 different oils were tested of which dinonyl phthalate satisfied all of the following criteria: (a) fat cells were recovered quantitatively; (b) little quenching of 3H radioactivity occured; (c) labelled glucose, inulin, insulin, water and triglyceride did not dissolve in the oil to any significant extent. (d) it was convenient to handle. 3. 3. Fat cells which had passed dinonyl phthalate exhibited normal lipogenesis from glucose and normal sensitivity to insulin. 4. 4. After incubation at 37°C for 2–10 min followed by passage of the cells through dinonyl phthalate, the distribution spaces of the packed cells for [G-3H]-inulin, [I-14C]mannitol and [U-14C]sucrose were about 1.0 μl/100 μl cells and the distribution spaces for 3H2O and 3-O-[14C]methylglucose were about 2.3 μl/100 μl cells. The coefficient of variation for the measurement of these spaces was about 5%. 5. 5. 3H and 14C activity could be removed completely after washing of cells incubated with 3H2O and 3-O-[14C]methylglucose (40 mM). The intracellular 3H activity was almost zero after 30 s whereas the 14C activity had decreased to about half at this point.

Binding and degradation of 125I-labelled insulin by isolated rat fat cells

Abstract 1. 1. Isolated rat fat cells were incubated at 37°C with 125I-labelled insulin and separated from the incubation medium by centrifugation through dinonyl phthalate. At low 125I-labelled insulin concentrations (7 · 10−11−7 · 10−10 M) the uptake of 125I-labelled insulin by fat cells reache equilibrium after 45 min of incubation. The concentration dependence of equilibrium binding of 125I-labelled insulin showed that one part of the binding could be saturated, whereas another part was proportional to the concentration in the range investigated (7 · 10−12−7 · 10−6 M). The saturable part of the binding of 125I-labelled insulin could be explained by a reaction with a cellular receptor resulting in the formation of a receptor-insulin complex. The dissociation constant of the receptor-insulin complex was determined as 3 · 10−9 M and the amount of receptors as 8 · 10−10 moles/lcell, or approximately 5 · 104 per cell. The rate constants of dissociation and association were 7 · 10−2 min and 2.5 · 107 min−1 · M−1, respectively. Neither vasopressin, glucagon nor des-B23-B30-octapeptide insulin, in concentrations of 7 · 10−7 M, reduced the binding of 125I-labelled insulin, whereas proinsulin inhibited the binding with an inhibition constant of 1.3 · 10−7 M. 2. 2. 125I-labelled insulin in the medium was degraded by fat cells. The Michaelis constant of the degrading enzyme system was 3.5 · 10−74 M and the maximal velocity was 45 nmoles/min per cell. Proinsulin was degraded less rapidly than insulin and the degradation of 125I-labelled insulin was inhibited by proinsulin (7.3 · 10−7 M), whereas glucagon (6 · 10−7 M) had no effect. 3. 3. Binding of 125I-labelled insulin and lipogenesis from [2-3h]glucose were measured on the same cells. Fat cells with a dissociation constant of binding of 3 · 1010−9 M exhibited half-maximal stimulation of lipogenesis at a 125I-labelled insulin concentration of about 7 · 10−11 M. 4. 4. The results show that the principle binding of insulin by fat cells occurs to a group of receptors with a Kd of 3 · 10−9 M and that the degradation of insulin and the receptor binding are processes independent of each other. It is suggested that the stimulatory effect of insulin on lipogenesis from glucose in fat cells is mediated by these receptors and that the half-maximal effect of insulin is obtained when about 2% of the receptors are occupied.

Biological potency and binding affinity of monoiodoinsulin with iodine in tyrosine A14 or tyrosine A19.

Summary Monoiodoinsulin with 99% of the iodine in the A chain was separated in two bands by disc-electrophoresis using long polyacrylamide gel rods. The bands contained monoiodoinsulin with the iodine in tyrosine A14 and tyrosine A19, respectively. The biological potency and binding affinity of A14 [ 127 I]monoiodoinsulin on rat adipocytes was indistinguishable from that of insulin, whereas the A19 derivative was only half as potent. Similarly, the maximum binding of “tracer” A19 [ 125 I]monoiodoinsulin to adipocyte receptors was only half of that obtained with A14 [ 125 I]monoiodoinsulin.

Trypsin treatment of adipocytes: Effect on sensitivity to insulin

Abstract 1. 1. Collagenase-isolated epididymal rat fat cells were treated for 18 min with 1 mg/ml and 10 μg/ml trypsin, respectively. The action of trypsin was determinated by soybean trypsin inhibitor, the cells were incubated with [U-14C]glucose and insulin in various concentrations, and finally the synthesized 14C-labelled lipid and 14CO2 were measured. 2. 2. Cells treated with 1 mg/ml trypsin showed no response to insulin in concentrations up to 107 μunits/ml for 30–60 min. After this period, the cells responded maximally to insulin in a very high concentration (104 μunits/ml), whereas they did not respond to insulin in the lowest concentration which gives an almost maximal response on control cells (20 μunits/ml). 3. 3. Cells treated with 10 μg/ml trypsin responded maximally to insulin in very high concentrations (104 μunits/ml) shortly after the trypsin treatment, whereas the responsiveness to 20 μunits/ml was restored after about 90 min. 4. 4. Cells treated with 10 μg/ml trypsin, incubated in buffer without additions for 60 min and then with [14C]glucose for 12 min exhibited a log insulin dose-response curve parallel to that of control cells. The dose-response range was about 80–1280 μunits/ml as compared to about 1.25–20 μunits/ml for control cells. 5. 5. [125I]Insulin (20 μunits/ml) was bound to specific receptor sites of normal cells, whereas no specific binding to trypsin-treated cells was observed. Restoration of responsiveness to insulin was accompanied by at least a partial restoration of the ability to bind [125I]insulin. 6. 6. Various factors with insulin-like activity on fatt cells were tested on trypsin-treated cells. The action of some factors was reduced to the same extent as that of insulin, whereas other factors were fully active. Factors of the first group (A) were: Proinsulin, non-suppressible insulin-like activity of serum (native and purified forms) and Streptomyces priseus protease. Factors of the second group (B) were: p-chloromercuribenzene sulphonic acid, cysteine, glutathion, oxytocin, vasopressin and hyperosmolarity (mannitol). Phospholipase C was more effective on trypsin-treated cells than on control cells. 7. 7. The following is concluded: (i) Fat cells treated with 10 μg/ml trypsin exhibit a markedly reduced sensitivity to insulin. (ii) The sensitivity of the treated cells gradually returns back to normal within about 2 h. (iii) The action of some factors (Group A) may be mediated through interaction with insulin receptors. (iv) Other factors (Group B, including phospholipase C) appear not to act via this mechanism.

Stimulating effect of hyperosmolarity on glucose transport in adipocytes and muscle cells

Abstract 1. 1. The effect of hyperosmolarity on sugar permeability was assessed in isolated fat cells, epididynaml fat pads and soleus muscles of the rat. 2. 2. Hyperosmolarity, whether induced by the addition of sucrose, mannitol or sorbitol, gave an up to 15-fold increase in glucose metabolism of isolated fat cells. The stimulating effect of mannitol (400 mM) and a submaximal concentration of insulin (5 μunits/ml) were similar with respect to the rates of onsent and reversal, kinetics and sensitivity towards 3-O- methylglucose or phlorizin. 3. 3. In whole epididymal fat pads and soleus muscles hyperosmolarity produced a marked increase in the release of 3-O- methylglucose without causing any decrease in K+ content. This effect could be abolished by phlorizin (5 mM) and showed nearly the same time-course as that induced by a submaximal concentration of insulin. 4. 4. It is concluded that hyperosmolarity stimulates the carrier system mediating glucose transport without impairing the overall integrity of the plasma membrane.

Receptor binding and biological effect of insulin in human adipocytes.

SummaryThe binding of125I-labelled insulin to human adipocytes was studied at 37° C. The precipitability of the125I-labelled insulin preparation (0.03 nmol/l) in trichloroacetic acid and the concentration of biologically active insulin (7.5 nmol/l) remained constant in buffer incubated with human adipocytes (100 μl cells/ml suspension) for 30–60 minutes at 37° C, whereas more than half of the insulin was inactivated by rat fat cells under the same conditions. A constant level of binding of125I-labelled insulin (0.03 nmol/l) to human adipocytes was obtained after 45 minutes. The apparent dissociation constant of receptor binding was about 0.2 nmol/l as compared to about 2 nmol/l for rat adipocytes. Conversion of [U-14C]glucose to lipids was stimulated half-maximally by about 0.05 nmol/l of insulin (similar to rat adipocytes). Thus, half-maximal stimulation of human adipocytes was obtained with a receptor occupancy of about 20–30 per cent.

Receptor binding of glucagon and adenosine 3',5'-monophosphate accumulation in isolated rat fat cells.

Abstract • The binding of glucagon and its effect on adenosine 3′,5′-monophosphate (cyclic AMP) accumulation and glycerol release was assessed in isolated rat fat cells at 37°C. • The dissociation constant of the displaceable (saturable, specific) binding was about 1.5 nM. The rate constant of dissociation ( k −1 ) varied from 8 8.6 · 10 −4 s −1 to 3.3 · 10 −3 s −1 , i.e. T 1 2 varied from 800 s to 200 s. Within an experiment, k −1 was the same when cells with prebound 125 I-labelled glucagon were suspended in medium containing either no glucagon or 1 μM of unlabelled glucagon. Negative homotropic cooperativity therefore seemed absent. The half-time of association of 70 pM 125 I-labelled glucagon was of the same order of magnitude as that of dissociation. • The conversion of prelabelled ATP to labelled cyclic AMP was enhanced by glucagon with a time lag of less than 20 s. The maximal accumulation was reached at 2–5 min both in the absence and in the presence of 2.5 mM theophylline. The concentration of glucagon causing half-maximal accumulation of cyclic AMP at 5 min was about 2 nM both in the absence and in the presence of theophylline. Glucagon stimulated glycerol release half-maximally at a concentration of about 2 nM in 1 h incubations. • The des-His 1 -glucagon exhibited about half of the maximal effect of glucagon on cyclic AMP accumulation even though it was able to inhibit binding of 125 I-labelled glucagon to the same extent as glucagon. Half-maximal displacement and half-maximal effect were obtained with about 200 nM of des-His 1 -glucagon. • The results are compatible with the following model for the action of glucagon on adipocytes. The adenylyl cyclase is stimulated in a dose-dependent fashion at the low receptor occupancies obtained 20 s after the addition of glucagon in submaximally stimulating concentrations. The increase in cyclic AMP concentration is antagonized so that the maximal accumulation with a given concentration of glucagon is obtained after a few minutes even though the receptor occupancy continues to increase. The histidine residue is necessary for maximal activation of the adenylyl cyclase.

The excretion of insulin in urine.

SummaryThe concentration of biologically active insulin was measured by the isolated fat cell method in serum and urine from healthy subjects and compared with the concentration of immunoreactive insulin. In both urine and serum the values obtained by the two methods correlated closely. In addition, there was a close correlation between the concentration of biologically active and immunoreactive insulin in urine from maturity onset diabetics. Therefore, conclusions on the excretion in the urine and the urinary clearance of insulin, which are based on measurements of immunoreactive insulin, are also valid for biologically active insulin.