Somatomedins, circulating levels

A wide variety of data firom many laboratories indicates that the liver is a major source of somatomedin peptides. This has been demonstrated directly in studies of isolated perfused livers (F3, K13, MIO, P9, S9, S2I, W6), fetal (D14, R5) and adult (B25, S13) liver in organ culture, rat liver cell lines (M5, M26, S27), and primary hepatocyte cultures (K14, S22, S28). Tliese studies are supported by the observations that partial hepatectomy (U3) or liver disease (S14, T5) results in low circulating somatomedin activity. It has not always been clear, however, which members of the somatomedin family were being assayed in some of these studies, due to the broad specificity of the assays used (see Section 5). For example, whereas it has been well established that the BRL (buffalo rat liver) cell line, and fetal rat liver in organ culture (R5), produce peptides of the MSA or IGF-II family (M5, M26), it is not known whether normal adult liver produces IGF-II. Indeed, production of SM-G/IGF-I by adult liver has been demonstrated specifically in only two studies. In these, Schwander et al. (S21) found that a S-labeled product from perfused liver could be immunoprecipitated with an IGF-I antiserum, and Scott et al. (S22) used a specific SM-C/IGF-I RIA to demonstrate production of the peptide by adult hepatocytes in primary culture. In both studies, the measured hepatic production rate was calculated to be sufficient to account fully for circulating SM-C/IGF-I levels. 

The observations that fetal tissues in culture produce somatomedins (Section 2.1), and that receptors for somatomedins (especially IGF-II) are abundant in membranes of fetal tissues (Section 4.3), both suggest that somatomedins may play a role in fetal development. Hus is supported by measurements of the peptides in the fetal circulation. While some early studies in this area are inconclusive due to the broad cross-reactivities of the assay methods employed, specific determination of SM-C/ICF-I and IGF-II levels has been possible in recent studies, although much of the fetal data is derived from nonhuman species. 

Because dependence upon circulating CH levels is one of the definitive characteristics of the somatomedins (see Section 1.2), new assays have generally been characterized by their ability to distinguish somatomedin levels in patients with CH disorders from those in healthy subjects. Several studies have compared values in GH-deficient children with those in healthy adults and, while an assay method that could not make this distinction would be invalid, dependence upon CH cannot be established by such comparisons, since somatomedin levels (particularly SM-C/ICF-I) in children without CH deficiency are lower than adult values. 

It is becoming increasingly clear that, of the factors apart from GH which influence circulating somatomedin levels, nutritional status is among the most important. Bioassay measurements have shown reduced somatomedin activity in children with protein-calorie malnutrition (G8, H22), and in rats either acutely (P6) or chronically (S24) undernourished. The interpretation of these findings, however, is complicated by the presence of bioassay inhibitors in fiisting serum specimens (see Section 3.5), raising uncertainty as to whether the concentrations of somatomedin peptides were actually decreased. 

In rats with experimentally induced diabetes, circulating somatomedin levels are very low compared with values in control animals (F4, P4). As with the investigations of the effects of undernutrition, this was originally observed using bioassay methods to measure somatomedin levels and, similarly, the discovery of inhibitors of somatomedin bioactivity in the circulation of diabetic animals (see Section 3.5) raised the possibility that the apparently reduced somatomedin levels might simply be due to interference of these substances in the bioassays. More recently, RRA (B12) and RIA (Ml) studies have confirmed that the concentration of somatomedin peptides is indeed reduced in experimental diabetes. This is perhaps not surprising, since both circulating GH levels (T7), and the concentration of GH receptors on liver membranes (B13, B15), are very low in these animals. 

Consistent wdth a major role of the liver in maintaining normal somatomedin levels in the circulation, low levels of somatomedin-A activity have been found in unextracted plasma of patients with cirrhosis of the liver and chronic hepatitis (S14, T5). Significant correlations were seen between somatomedin-A and albumin, cholinesterase, and other indicators of liver function (T5). The decrease measured in this RRA appears to be due to low levels of both somatomedins and binding protein, since Zapf et al. (Z5) have shown an 89% decrease in immunoreactive IGF-I, a 74% decrease in total IGF by protein binding assay (which preferentially measures IGF-II), and a 57% decrease in specific binding of somatomedin tracer to stripped serum, in patients with cirrhosis. 

P12. Philips, L. S., and Kopple, J., Circulating somatomedin activity and sulfate levels in adults with normal and impaired kidney function. Metab., Clin. Exp. 30, 1091-1095 (1981). 

---