Insulin_I l25 (0.1 ml, 0.1 microCi) was added to the equal volume of serum and fraetionated on 9 x 300 mm column of Sephadex G-I00 into protein-bound (B) and free (F) insulin immediately or after 24-hour incubation. When the fractionation was performed immediately, the B/F ratio was in 10 healthy persons 0.045 ± 0.006 (just like that of label-Ied insulin preparation itself), and in twenty-six insulintreated diabetics 0,605 ± 0,073, After incubation B/F ratio were correspondingly 0.056 ± 0.011 and 2.021 ± 0.103. There was no overlapping between normals and diabetics (P < .00l), and no protein binding of insulin by normal serum was evident. At pH 3 the bound insulin dissociated, and virtually all the radioactivity appeared with free insulin fraction. Free insulin level in the serum of diabetics as determined after serum fractionation on Sephadex G-IOO was 35.8 ± 4.0 micro V/mi only and did not conelate with B/F ratios; ,insulin dose or blood glucose. No (free)-hyperinsulinemia was found in severe insulin-treated diabetics.

No abstract

Polyethylene glycol (PEG) and killed Staphylococcus aureus cells (S. aureus) were used as agents to separate free insulin from antibody-bound insulin in diabetic sera. Insulin was determined by conventional double antibody radioimmunoassay. The free insulin values after PEG treatment were almost half of those after S. aureus treatment. The free insulin levels in high-antibody containing sera preincubated at 37 degrees C, 2 h were double the value of fresh sera. PEG treatment caused about 40% loss of total serum protein. The addition of bovine serum albumin (BSA) to the PEG-treated serum greatly increased the immuno-reactive insulin values. This may suggest that protein concentration plays a role in insulin radioimmunoassay. PEG treatment may also enhance the interaction between free insulin and free antibodies resulting in underestimation of free insulin level.

In rats, a moderately hepatotoxic single dose of diethylnitrosamine (DEN) 100 mg/kg causing depletion of liver glycogen, elevation of aspartate aminotransferase and decreased liver uptake of 3-O-methylglucose, resulted in substantial changes in insulin and glucagon balance. Two days after DEN, insulin binding to liver membranes and insulin removal by the liver were sharply reduced whereas its binding to muscle and adipocyte membranes remained unaltered. Serum insulin (random and after an overnight fast) remained normal. Intravenous (I.V.) insulin (10 U/kg) caused the usual degree of hypoglycemia that, however, lasted longer than in the control animals. Removal of glucagon by liver was also depressed in spite of its normal binding to hepatocytes, and peripheral serum glucagon was increased three-fold. I.V. glucagon (40 micrograms/kg) resulted in a blunted response of plasma glucose. I.V. glucose tolerance test (1 g/kg) remained normal in spite of the insulin increase to a level twice as high as in the controls, and in spite of nonsuppressed glucagon. These changes were still present after 1-3 months, but disappeared by 6 months. The results demonstrate remarkable ability of homeostatic mechanisms to preserve normal plasma glucose and glucose tolerance in spite of dramatic changes in insulin and glucagon.