Somatomedins in pregnancy: a cross-sectional study of insulin-like growth factors I and II and somatomedin peptide content in normal human pregnancies

To explore the role of the somatomedins (SM) during human pregnancy, we have measured plasma levels of insulin-like growth factor I (IGF-I), IGF-II, and SM peptide content (SMPC) in 79 women in various stages of normal pregnancies. IGF-I and IGF-II were measured by specific RIAs, and SMPC was measured by a radioreceptor assay using human placental membranes. IGF-I and SMPC rose during pregnancy, showing a significant positive correlation with the length of gestation. Plasma levels of IGF-I in the third trimester averaged 324 ng/ml, a 33% increase over the first trimester average of 243 ng/ml (P less than 0.05). Although IGF-II did not correlate with the length of gestation, the third trimester average was significantly higher than the first trimester average (780 vs. 630 ng/ml; P less than 0.05). After delivery, both IGF-I and IGF-II levels rapidly dropped to levels significantly below those seen in the third trimester. The gestational rise in SMPC and plasma levels of both IGF-I and IGF-II supports the hypothesis that SM play a role in the regulation of fetal growth.

Characterization of an insulin-like growth factor-I/somatomedin-C radioimmunoassay specific for the C-peptide region

Insulin-like growth factor-I (IGF-I) and somatomedin-C )SM-C) have been shown to be functionally identical by a number of criteria. We have synthesized the 12 amino acid C-peptide region of IGF-I (Gly-Tyr-Gly-Ser-Ser-Ser-Arg-Arg-Ala-Pro-Glu-Thr) and developed a RIA based on antibodies against this synthetic peptide. IGF-I and SM-C were indistinguishable in this RIA. No other peptides competed for this antiserum. The SM-C/IGF-I values of acid-chromatographed serum were strongly age dependent. The mean of children 1-5 yr old was 0.67 +/- 0.033 U/ml (mean +/- SD; n = 23), whereas the mean of children 12-17 yr old was 2.01 +/- 0.66 U/ml (n = 39) and the mean of 38 adults 26-85 yr old was 1.05 +/- 0.34. The SM-C/IGF-I values measured by this RIA were also growth hormone dependent. Thus, this region-specific RIA provides a clinically useful assessment of serum SM-C/IGF-I levels.

Serum lactate dehydrogenase isoenzyme 1 in patients with advanced testicular cancer

Abnormal levels of serum lactic dehydrogenase-1 (LD-1) activity have been observed in 81% (34/42) of patients with stage III germ cell malignancy of the testis. The criteria for evaluation the electrophoretic isoenzyme patterns of these patients were, as follows: For criterion 1 elevations the LD-1 value in absolute units was greater than 52.0 U/I with the LD-1/Total LD ratio greater than 37.2%. Criterion 2 elevations had absolute values of LD-1 less than 52.0 U/I, but the LD-1/Total LD ratio was greater than 37.2%. For criterion 3 elevations of LD-1, the absolute value was greater than 52 U/I and the LD-1/Total LD ratio was less than 37.2% activity with the LD-5/LD-1 ratio less than 0.5; or when the LD-5/LD-1 ratio was greater than 0.5 but the LD-1 is equal to or greater than the LD-2. The frequency of LD-1 elevation correlated well with the extent of the disease (stage II-B-1 and 2, 50%; stage III-B-3, 86%; stage III-B-4, 91%; stage III-B-5, 93%). LD-1 elevation occurred in groups I, II, IV and V histopathologic cell types (Dixon and Moore Classification) and there did not appear to be any correlation between the histologic cell type and the frequency of elevation of LD-1. Interpretation of LD-1 activity only on the basis of its relative ratio to the total LD value (criterion 1 and 2) identified a total of 28 patients (67%). A criterion 3 elevation was demonstrated in 6 (14%) additional patients. All patients with persistent elevations or recurrent elevations of LD-1 have shown progressive or recurrent disease and patients with no clinical evidence of disease have demonstrated normal LD-1 values. In those patients with elevated LD-1 activity, serial measurements of serum. LD-1 isoenzyme reflect the response of the patient to therapy.

Silica fragments from millet bran in mucosa surrounding oesophageal tumours in patients in northern China

Millet bran is a component of the diet in the area of highest oesophageal cancer incidence in northern China. Millet bran was found to contain up to 20% by weight of silica; some of this silica occurs as friable sheets or sharply-pointed fibres. These types of silica in millet bran are the most likely source of an unusual contamination with fragments of silica found in the oesophageal mucosa surrounding tumours in patients in northern China. A group of mucosal samples analysed together contained over 5,000 particles/g (100 parts per million by weight), ten times as many as were found in tissue from normal controls taken at necropsy in London. The modal diameter was 10 microgram (1-70 microgram). The particles were in the body of the mucosa and were not simply a surface contaminant. Silica fragments and fibres of similar size originating from other plant species occur in the diet in the two other regions of greatest incidence of oesophageal cancer, the Transkei and Iran. If such fragments enter the mucosa, they must cause some degree of trauma, and they may also be able to stimulate proliferation by providing anchorage. These findings suggest the possibility that silica particles might be involved in the aetiology of oesophageal cancer.

Somatomedin peptide distribution and somatomedin-binding protein content in cord plasma: comparison to normal and hypopituitary plasma

To characterize the macromolecular forms of somatomedin (SM) in human newborn plasma, we have studied the molecular weight distribution of endogenous SM peptides as well as the content and distribution of the acid-stable and the unsaturated SM-binding proteins (SMBP) in cord blood from 13 normal term infants. The radioreceptor assayable SM peptide content was significantly reduced in newborns compared with that in normal adults. Furthermore, 50% of the SM content of newborns circulated as part of a 50,000 mol wt complex, in contrast to adult plasma where the majority of SM peptide content is found in the 150,000 mol wt range. Unsaturated SMBP was strikingly elevated in newborns (mean +/- SRM, 2.75 +/- 1.73 U/ml) compared to adult values of 0.63 +/- 0.24. Sephacryl-200 chromatography demonstrated that the unsaturated SMBP is found in the 40,000-50,000 mol wt region in newborns, adults, and GH-deficient children, although adults appear to have a secondary peak of unsaturated SMBP in the 150,000 mol wt region. Assay of the acid-stable SMBP indicated similar levels in newborns (1.15 +/- 0.26 U/ml) and adults (1.18 +/- 0.47) and strikingly lower values in GH-deficient subjects. The molecular weight of the acid-stable SMBP of newborns, adults, and hypopituitary dwarfs appeared to be similar, measuring approximately 60,000. We conclude that despite low levels of SM peptides, human cord plasma contains normal levels of the acid-stable SMBP and elevated of the unsaturated SMBP. The role of these binding proteins in cord plasma remains uncertain, but suggests that other SM peptides may be important in fetal growth.

A radioimmunoassay for insulin-like growth factor II specific for the C-peptide region

Insulin-like growth factor II (IGF-II) is a human plasma peptide whose sequence is homologous to both insulin-like growth factor I/somatomedin C (IGF-I/SM-C) and human proinsulin in the A and B regions. However, there is no obvious homology in the C (connecting peptide) region. The synthetic 8-amino acid C-peptide segment of IGF-II (Ser-Arg-Val-Ser-Arg-Arg-Ser-Arg) was covalently linked to thyroglobulin to render it more antigenic. Antiserum against the IGF-II C-peptide was generated which had a titer of 1:2000 determined with [125I]IGF-II C-peptide. Half-maximum displacement was by 350 pg/ml IGF-II C-peptide or 80 ng/ml IGF-II. There was no displacement by IGF-I/SM-C, insulin, or a wide variety of peptides. There was also a high degree of species specificity of this antisera. Isoelectric focusing studies of immunoreactive IGF-II showed an apparent pI of 6-6.5. The mean (+/- SEM) level of IGF-II after acid chromatography of 28 normal adult males was 687.0 +/- 31.9 ng/ml. The mean of 8 acromegalics was 600.5 +/- 57.4 ng/ml, indistinguishable from normal. The IGF-II levels of 21 hypopituitary children were significantly lower (231.5 +/- 32.3 ng/ml). Thus, GH action appears to be necessary for normal levels of IGF-II, but excess GH does not cause an elevation above normal of IGF-II, unlike what is seen with IGF-I/SM-C. These structurally related IGF peptides have different control mechanisms and ultimately may play different functional roles. The availability of specific RIAs for the measurement of IGF-II will help to clarify its role in human physiology and disease states.

Plasma somatomedin-binding proteins in hypopituitarism: changes during growth hormone therapy

The somatomedin (SM) peptides are carried in plasma complexed to specific SM-binding proteins (SMBPs). In addition to the SMBP in the complex, there is an unsaturated SMBP in plasma in which the SMBP is not occupied by SMs. We have compared levels of unsaturated SMBP in 7 normal adults to those in 21 children with GH deficiency before and during treatment with hGH (0.1 U/kg.day) for 4 days. In addition, the SM-C/insulin-like growth factor I (IGF-I) content of each plasma was measured by RIA. There was a significant (P less than 0.01) difference in unsaturated SMBP levels between normal controls (17.8 +/- 0.8% bound/20 microliters) and untreated hypopituitary patients (27.8 +/- 2.2% bound/20 microliters). Thus, a lower SM-C/IGF-I content was associated with a higher unsaturated SMBP level. Furthermore, there was a significant negative correlation between SM-C/IGF-I content and unsaturated SMBP in untreated hypopituitary patients (r = 0.73; P less than 0.0001). Treatment with hGH normalized the mean unsaturated SMBP level in hypopituitary patients within 2 days. Full displacement curves and Scatchard analysis showed that the increased unsaturated SMBP level in hypopituitary plasma was entirely due to an increased affinity (8.4 X 0.9 X 10(-10) M) compared to normal (2.3 X 0.2 X 10(-9) M). A higher affinity form of unsaturated SMBP is uniquely present in hypopituitarism and disappears with hGH treatment. The measurement of this unsaturated SMBP mirrors the SM peptide content.

Immunochemical determination of LDH-1

An evaluation of a new immunochemical method for the heart isoenzyme of lactic dehydrogenase (LDH-1) is described. In this assay, a goat antibody against the M subunit of LDH is added to a patient's serum. Donkey antigoat gamma globulin is then added, which precipitates the M-anti-M complex. LDH activity in the supernatant is thus a measure of the heart isoenzyme, since this isoenzyme is not precipitable by the anti-M antibody. Assays for total CK, CK-MB, and total LDH, and LDH electrophoresis were performed along with an immunochemical LDH-1 assay on 261 serum samples from 49 patients clinically suspected to have acute myocardial infarctions. These were performed on admission and 12, 24, 48, 72, and 96 hours later. The diagnosis of acute infarction was confirmed for 25 of the 49 patients and ruled out for 19. Five patients were classified clinically as borderline cases for the diagnosis of acute myocardial infarction. All 25 patients with the clinical diagnosis of myocardial infarction had positive results by both electrophoresis and immunochemistry; four of the five borderline patients had positive results by electrophoresis. All five borderline patients had positive results by immunochemistry. None of the 19 patients without apparent myocardial injury had positive results by either electrophoresis or immunochemistry. Ten (40%) of the 25 samples collected at the time of admission from 25 patients who had myocardial infarctions had elevated LDH-1 activity by electrophoresis, while 14 patients (56%) had increased LDH-1 by immunochemistry. The immunochemical method for LDH-1 is more sensitive than electrophoresis for the detection of myocardial infarction and offers the further advantages of greater simplicity, precision, and accuracy, while being no less specific.

Acute somatomedin response to growth hormone: radioreceptor assay versus radioimmunoassay

The acute somatomedin (SM) response to GH therapy has been examined in 21 GH-deficient children using a placental membrane radioreceptor assay (RRA) which measures a variety of SMs and a RIA specific for SM-C and insulin-like growth factor I (IGF-I). Plasma for determination of SM peptide content was obtained before initiation of therapy and 13 h after each of four daily injections of GH (0.1 U/kg). An additional SM determination was performed after 6 weeks of GH therapy (0.1 U/kg, three times per week) in seven of the subjects. RRA and RIA SM determinations were performed on the same acid-chromatographed sample and were compared to an acid-chromatographed pooled plasma standard. The 4 days of GH therapy resulted in an increase in SM levels from 0.39 +/- 0.24 to 1.18 +/- 0.62 (+/- SD) U/ml, determined by RIA. A single injection of GH resulted in a significant rise in plasma SM levels, measured by either RRA or RIA (P less than 0.001). Subjects who responded poorly to two injections of GH also had low SM levels after 4 days and even after 6 weeks of GH therapy. The RRA resulted in consistently higher value than the RIA. This difference was even greater when results were compared to a pure IGF-I/SM-C standard. The SM peptide contents determined by RRA and RIA were strongly correlated, not only for the group, but also among the determinations for each individual subject. However, the consistently higher values observed when the SM peptide content was measured by RRA compared to that measured by RIA and the variability in the RRA to RIA ratio among individual subjects suggest that the IGF-I/SM-C RIA measures only one of a number of GH-dependent SM peptides.

Joining of yeast alanine transfer ribonucleic acid half molecules to form a whole molecule by T4 RNA ligase

Covalent joining of the two half molecules of tRNAAla by T4 RNA ligase to form a reconstituted whole molecule was investigated. The two half molecules consisting, respectively, of residues 1-35 and 36-75 were prepared by partial degradation of tRNAAla with RNAase T1. The 5'-half molecule was treated with alkaline phosphatase to remove the 3'-terminal phosphate group, and the 5'-OH group of the 3'-half molecule was phosphorylated with [gamma-32P]ATP by polynucleotide kinase. The two terminal nucletides to be joined were identified as Guo and Cyd. Prior to the covalent joining reaction, the two modified half molecules in an equimolar mixture were annealed, and the rejoined half molecules, separated by gel electrophoresis, served as the substrate for T4 RNA ligase. Optimum conditions for this ligation, such as RNA ligase concentration, pH, Mg2+ concentration, reaction temperature and time of reaction, were investigated. Under the optimum conditions a yield of about 70% joining of the reconstituted whole molecule was obtained as shown by gel electrophoresis, resistance to hydrolysis by alkaline phosphatase, nearest neighbour analysis and alanine acdeptor activity.

Interaction of somatomedin-C with an antibody directed against the synthetic C-peptide region of insulin-like growth factor-I

Insulin-like growth factor-I is a human plasma peptide with strong structural homology to human prosinsulin. This peptide has been classified as a somatomedin on the basis of its biological actions and growth hormone dependence. We have generated an antibody to the synthetic 12 amino acid C-peptide region of insulin-like growth factor-I and used it to compare three somatomedin preparations to insulin-like growth factor-I. We also compared these somatomedin preparations to insulin-like growth factor-I using the standard SM-C RIA. We conclude that insulin-like growth factor-I and somatomedin-C behave identically in these two RIA's and are very similar if not identical molecules.

Comparison of [125I]somatomedin A and [125I]somatomedin C radioreceptor assays for somatomedin peptide content in whole and acid-chromatographed plasma

The placental membrane radioreceptor assay was used to measure the levels of somatomedin (SM) peptides in plasma. Displacement of both [125I]somatomedin A ([125I]SM-A) and [125I]somatomedin C ([125I]SM-C) by normal whole plasma, the peptide fraction of acid-chromatographed plasma, and a partially purified, insulin-free SM preparation were compared. The peptide fraction of plasma was isolated by acid chromatography over Sephadex G-50 in 0.25 M formic acid with a yield of greater than or equal to 90%, as determined by bioassay and [125I]SM. In the case of [125I]SM-A, the dose-response curves for whole plasma, acid-chromatographed plasma, and the standard SM preparation were parallel (P less than 0.2). In contrast, for [125I]SM-C, the dose-response curves for acid-chromatographed plasma and the purified SM preparation were parallel (P less than 0.2), but both differed significantly from that of whole plasma (P less than 0.001). In addition, there was less variability in the assay of acid-chromatographed plasma compared to whole plasma. The results indicate that radioreceptor assay of unextracted normal plasma using [125I]SM-A is a valid measure of SM peptide concentration, while radioreceptor assay of unextracted normal plasma using [125I]SM-C, in our hands, is not. Acid chromatography of plasma before its assay is an uncomplicated procedure which allows valid and precise measurement of SM peptide content using either [125I]SM-A or [125I]SM-C.

Synthesis of heteroaromatic potential beta-adrenergic antagonists by the glycidol route

The synthesis of several 3-alkylamino-2-hydroxypropyl heteroaryl ethers (13-15, 17, and 18) is described. These compounds were prepared by the alkylamination of the corresponding glycidyl ethers (6-8, 10, and 11), which in turn were obtained from the requisite heteroaryl halides and the sodium salt of glycidol. The above basic ethers exhibited beta-blocking activity, but the potency of the tested compounds was considerably less than that of propanolol. Only 3-tert-butylamino-2-hydroxyl-1-(1,2,4-thiadiazol-5-yl) propyl ether (13) showed some selective myocardial beta-blocking activity.

Demonstration of specific plasma protein binding sites for somatomedin

Native somatomedin (SM) has an apparent mol wt of at least 60,000 daltons, while all the purified SM peptides have mol wt 5-8,000 daltsons. We have studied the behavior of somatomedin in plasma during gel filtration at pH 8.1 and pH 2.8, and by the recombination of fractions. Using porcine bioassay for SM, displacement of 125I-insulin by SM, and binding of 125I-SM to plasma proteins, we can demonstrate that the disparity in apparent mol wt of SM is due to reversible binding to plasma proteins which are themselves biologically inactive. This binding site is specific for the SM peptides, saturable, and has high affinity and relatively low capacity.

Comparison of 125-I-insulin binding and degradation to isolated rat hepatocytes and liver membranes

We have compared the ability of rat liver plasma membranes and isolated hepatocytes to bind and degrade insulin. Isolated cells were prepared in two different ways: by mechanical separation of cells and by collagenase digestion of extracellular matric. In all studies the binding and degradative characteristics of both types of hepatocyte preparations were identical. Furthermore, with one exception, the binding characteristics of membranes and cells were also quite similar. The only exception concerned the amount of insulin bound by hepatocytes as compared to liver membranes. Thus, at concentrations of cells (1.2 times 10(6) cells per milliliter) and membranes (150 mug. protein per milliliter) that gave equal binding at insulin concentrations less than 100 ng./ml., the amount of insulin specifically bound at insulin concentrations greater than 100 ng./ml. was greater with use of hepatocytes. Additional studies indicated that, in contrast to membranes, at the higher insulin concentration only 75 per cent of the previously bound insulin could be recovered from hepatocytes. Thus, a nondissociable component exists; which probably represents intracellular radioactivity and appears to account for the higher specific insulin binding by cells at higher insulin concentrations. When insulin degradation was studied at the above hepatocyte and plasma membrane concentrations, cells degraded 30 per cent more insulin than did membranes. Kinetic analysis of these data revealed that the Km for insulin degradation (5 times 10(-7) M at 37 degrees) was the same for both systems wereas the Vmax was greater with use of hepatocytes.

IN CONCLUSION

(1) Preparation of hepatocytes by collagenase digestion does not appear to alter insulin binding or degradation; (2) studies of liver membranes and isolated hepatocytes obtained from normal rats should yield similar information about insulin-receptor interaction as long as insulin concentrations less than 100 ng./ml. are used; (3) at very high insulin concentrations, some of the radioactivity appears to enter the cells; (4) the kinetics of insulin degradation by hepatocytes and liver membranes are similar; and (5) insulin degradation appears to be primarily a membrane phenomenon.

The effects of acute and chronic dexamethasone administration on insulin binding to isolated rat hepatocytes and adipocytes

In an effort to determine the possible relationship between changes in insulin-receptor binding and the glucocorticoid-induced insulin-resistant state, we studied insulin binding to specific receptors located on isolated adipocytes and hepatocytes obtained from dexamethasone (D)-treated rats. Three groups of D-treated rats were studied: (1) acute high-dose treatment (1.5 mg/kg/6 days), (2) acute low-dose treatment (0.125 mg/kg/6 days), and (3) chronic low-dose treatment (0.125 mg/kg/21 days). When insulin binding to isolated hepatocytes was studied, we found that binding to isolated hepatocytes was studied, we found that binding was only 30%-50% of control values when cells from the D-treated animals were used. This decrease in binding was greatest for cells from the acute high-dose group, indicating a dose-response effect, and least for cells from the chronic group, suggesting a tendency toward return of insulin-receptor binding during chronic treatment. When insulin binding to isolated adipocytes was studied, binding was 50%-60% of control values when cells from both acute D-treated groups were used. While the magnitude of the decrease in insulin binding was not as great as that seen with hepatocytes, the decrease was still greatest using cells from the acute high-dose group as compared to the acute low-dose group. Thus, a dose-response effect was suggested in both tissues. On the other hand, the effects of chronic D treatment on insulin binding were strikingly different in the two cell systems. After chronic D treatment, insulin binding to adipocytes returned to near-normal levels, while a 55% decrease in binding to hepatocytes persisted. Thus, the tendency toward return of insulin binding after chronic D treatment seen with hepatocytes was almost fully expressed by adipocytes. This might be related to the amelioration of the corticosteroid-induced insulin-resistant state which has been reported after chronic corticosteroid administration to humans. In conclusion, (1) a decrease in insulin binding is associated with corticosteroid excess, and it is possible that this decreased binding is related to the insulin resistance which results from corticosteroid administration; (2) the return of insulin binding toward normal after chronic D treatment could well be related to the improvement in insulin resistance seen during chronic corticosteroid administration to humans; and (3) the difference in effects of chronic D treatment on insulin binding to hepatocytes versus adipocytes indicates that changes in insulin binding can be tissue specific.

Long term results of gastrectomy with respect to blood lipids, blood pressure, weight and living habits

A sample of ambulant Japanese-American men (ages 45-69 years), was divided into those having a previous partial gastrectomy and a control non-gastrectomy population. Three-hundred-and-forty-seven men with a history of partial gastrectomy weighed less and had lower values for serum cholesterol, triglyceride, and blood pressure than did the control population of 7,598 men. The depressed lipid and blood pressure values could not be entirely explained by the reduced weight. Likewise, none of these differences appeared related to diet or living habits. Those operated on for gastric ulcer had, on the average, lower systolic pressures than duodenal ulcer patients and those with gastrojejunal anastamoses had lower cholesterol levels than patients with a gastroduodenostomy.