A direct assay for proinsulin in plasma and its applications in hypoglycaemia.

A direct radioimmunoassay in unextracted plasma is described. The assay has a sensitivity of 4 pmol/l (2 standard deviation from zero). The proinsulin antiserum was immuno-adsorbed against human C-peptide and insulin coupled to glass beads. Cross-reactivity of the antiserum was assessed and shown to be less than 0.01% with both peptides. In normal healthy fasting subjects the plasma proinsulin level was 6.7 +/- 1.7 pmol/l (n = 17) (mean +/- SD). Fasting proinsulin levels in non-insulin dependent diabetics were significantly elevated compared with non diabetics (14.2 +/- 2 pmol/l (n = 11) vs 6.7 +/- 1.7 (n = 17) P less than 0.005). The insulin/proinsulin ratio was 3.4:1 in the non-insulin dependent diabetic compared with 6:1 in non-diabetics. Samples from 21 insulinoma patients were assayed and mean fasting plasma proinsulin level was 255 pmol/l +/- 479 when the patients were hypoglycaemic. The range in pro-insulin levels was large (30-2300 pmol/l). Mean fasting proinsulin level in three hypoglycaemic subjects due to sulphonylurea overdose was 15.7 +/- 2.3 pmol/l. The molar ratio of proinsulin to insulin was 1:6 in healthy subjects, 1:1 in insulinoma patients and 10:1 in sulphonylurea induced hypoglycaemic patients.

Comparison of horseradish peroxidase and alkaline phosphatase-labelled antibodies in enzyme immunoassays.

The periodate method was found to be most effective for preparing horseradish peroxidase-sheep anti-human and horseradish peroxidase-donkey anti-mouse immunoglobulin (IgG) conjugates. The conjugates were improved by carrying out the oxidation of the enzyme at low pH. Anti-human and anti-mouse IgG-peroxidase conjugates (0.5 mg/mL IgG and 0.7 mg/mL IgG, respectively) were used at 1:15,000 and 1:8000 dilutions, respectively, in a sandwich ELISA to detect human and mouse IgG in buffer or in a growth medium containing 20% foetal calf serum. Using the peroxidase conjugates, it was possible to detect human and mouse IgG at concentrations as low as 1 ng/mL. The glutaraldehyde method was found to be much more effective than the periodate method for conjugating alkaline phosphatase to the antibodies. The optimum dilutions for anti/human and anti-mouse IgG-alkaline phosphatase conjugates (0.18 mg/mL IgG and 0.3 mg/mL IgG, respectively) in ELISA were 1:500 and 1:1000, respectively. The detection limit with alkaline phosphatase conjugates was 7 ng/ml for human IgG and 4 ng/ml for mouse IgG.