Inhibition of Access of Bound Somatomedin to Membrane Receptor and Immunobinding Sites: A Comparison of Radioreceptor and Radioimmunoassay of Somatomedin in Native and Acid-Ethanol-Extracted Serum*

Abstract

No agreement exists concerning the necessity for extracting serum somatomedin (Sm) before radioreceptor assay (RRA) and RIA. We have developed a simplified system of Sm extraction which has permitted a quantitative comparison of native and extracted Sm. To 0.8 ml of a mixture of 87.5% ethanol and 12.5% 2 N HC1, 0.2 ml serum is added. After centrifugation, 0.5 ml supernatant is neutralized with 0.2 ml 0.855 M Tris base (XT). Recovery of added 125I-labeled insulin-like growth factor I ([125I]IGF-I) was nearly quantitative. Neutral gel filtration studies established that virtually all of the binding complex was removed by acid-ethanol precipitation. The addition of 100 µl Tris-neutralized acid-ethanol buffer (AE B) to the [125I]IGF-I human placental membrane RRA produced an insignificant shift in the displacement curve with partially purified IGF reference preparation; the shift of the [125I]Sm C RIA displacement curve was greater but did not prevent accurate measurement when AE B was added to blank and standard tubes. When partially purified IGF was added to serum and extracted promptly with acid-ethanol, the added IGF was recovered completely in the extract, but when extraction was delayed until after 16 h of incubation at 4 C, there was poor recovery of added IGF, as determined by RRA. However, there appeared to be a similar recovery of endogenous Sms. The loss of added IGF was prevented by the addition of aprotinin. The RRA of XT in 9 normal sera gave results which were 2.72 ± 0.32 times higher than those obtained by direct RRA on native sera; the fold increment was 2.96 ± 0.37 when the measurements were made by Sm C RIA. Five sera from patients with hypopituitarism gave results with the RRA of XT which were only 1.12 ± 0.30 times greater than those obtained with direct assays on serum; the increment with RIA was 1.64 ± 0.23. The results of RRA on XTs in 34 sera from normal individuals, 18 patients with clinical GH deficiency, and 8 patients with acromegaly correlated well with the known GH status. We conclude that although direct serum Sm RRAs and RIAs have provided much useful information, these assays do not quantitatively measure the Sm content of serum. A simplified acid-ethanol extraction permits convenient and accurate measurement of total Sm. We suggest that direct RIA and RRA of Sm measure predominantly bound Sm. The reported lack of parallelism with equilibrium assays of native serum and the failure of quantitative recovery of Sm described here may be the result of steric and other factors which modify access of the Sm serum binding protein complex to membrane and immunobinding sites.