THE RELATIONSHIP OF ERYTHROCYTE INSULIN RECEPTORS TO RED CELL AGE AND TO MONOCYTE INSULIN RECEPTORS

SUMMARY This study examines the degree to which erythrocyte insulin receptors might be used instead of monocyte insulin receptors to assess overall insulin receptor status. Red blood cells from eight normal subjects were fractionated into older and younger populations by high speed centrifugation. Fractionated red cells (3·5 ± 10 9 /ml) were incubated with 0·14 ng/ml of ‘monoiodo’ 125 I‐insulin for 2·5 h at 15°C. The high reticulocyte fractions (mean 2·1% reticulocytes) had greater insulin binding than the low reticulocyte fractions (mean 0·6%), −9·0% compared with 6·0% specific binding, respectively. The difference ( P > 0·05) appeared to be predominantly due to a greater number of high affinity sites per cell in the high reticulocyte fractions. Culture of high and low reticulocyte fractions, from both normal subjects and patients with high reticulocyte counts, in vitro for 2·5 days at 37°C, markedly reduced reticulocyte count and insulin binding of the high reticulocyte fractions, but produced little change in the low reticulocyte fractions. It appeared the reticulocytes had a ten‐ to twenty‐fold increase in binding sites compared with erythrocytes. There is little degradation of insulin by intact erythrocytes or reticulocytes at 37°C, with a dissociation between insulin binding and insulin degradation. The large degradative capacity of haemolysed red cells implies that insulin receptor assays would be invalid in the presence of haemolysis. Monocyte and erythrocyte receptors were compared in the same blood samples from overnight fasted subjects, with a correlation between insulin binding (r = 0·62). The inexact correlation may be partly due to the non‐homogeneous insulin binding of red cells, but correction for differences between reticulocyte content of the assayed red cells only increased the correlation to r = 0.67. Whilst the non‐homogeneity of red cells means that erythrocyte binding must be interpreted with caution, we cannot exclude similar non‐homogeneity of monocyte populations.