Effects of somatomedin and other factors on glycosaminoglycan synthesis in cultured rat chondrocytes

We have investigated the effects of hormones and serum on glycosaminoglycan (GAG) synthesis, using cultured rat chondrocytes isolated from growing cartilage. Somatomedin A stimulated GAG synthesis at a physiological concentration, however in the case of insulin the dose required to stimulate GAG synthesis was 500 times as great as the physiological concentration. Parathyroid hormone also increased GAG synthesis. In contrast, hydrocortisone inhibited GAG synthesis at a pharmacological dosage. None of the following had any effect on GAG synthesis: epidermal growth factor, fibroblast growth factor, triiodothyronine, growth hormone, sex steroid or vitamin D3. Human serum up to a concentration of 1% stimulated GAG synthesis. Serum from patients with acromegaly stimulated GAG synthesis more than that from those with hypopituitarism.     

Decrease in serum receptor-reactive somatomedin in diabetes.

Somatomedin in rat serum has been measured by a sensitive radioreceptor assay using 125I-labelled human somatomedin and human placental membrane. In rats made diabetic with strepotzotocin, receptor-reactive somatomedin levels were decrease by up to 75%. The decrease followed the time course of increasing serum glucose and occurred to the same extent in rats aged between 4 and 40 weeks. Endogenous serum receptor-reactive somatomedin appeared exclusively in high molecular weight fractions on gel chromatography. In diabetes the decreased somatomedin was due to a fall in this high molecular weight activity, but was not accompanied by a fall in somatomedin binding protein. These results suggest a role for insulin in maintaining serum somatomedin levels.     

Binding of somatomedins and insulin to plasma membranes prepared from rat and monkey tissue.

Particulate membranes prepared from a variety of monkey and rat tissues were shown to have specific binding sites for somatomedin A. Binding to talc on the other hand showed no specificity, and the sensitivity was less. The membrane-bound somatomedin A was displaced by somatomedin A in concentrations between 0.01 and 5 U/ml. Of the other hormones tested only insulin in high concentrations could interfere with the binding of labelled somatomedin A to tissue membranes. Membranes prepared from a number of rat and monkey tissues also contained binding sites for insulin. Somatomedin A could interfere with the binding of labelled insulin. No binding was observed for somatomedin B in the tissues studied.     

Somatomedin production in the neonatal rat.

The hormonal stimulus to rat fetal and neonatal somatic and skeletal growth has been investigated by a study of ciruclating somatomedin (SM), growth hormone (GH) and insulin levels in rats from 21 days in utero to 40 days of post natal life. Somatomedin activity could not be detected in the serum of fetal rats in which circulating GH and insulin levels were high. In early post natal life GH and insulin levels remained high but gradually declined reaching normal adult levels at 19 days and 40 days respectively. Somatomedin activity was measurable only at low levels until 11 days after birth and thereafter remained steady throughout the time period studied. These studies suggest that somatomedin alone is not responsible for the rapid growth of the rat in early neonatal life and it is suggested that insulin may also be active as a growth factor in this period.     

Normal somatomedin and somatomedin receptors in achondroplastic dwarfism

It has been proposed that the basic abnormality in achondroplasia may be a quantitative defect in endochondral new bone formation secondary to decreased synthesis of somatomedin (SM) or abnormal binding of SM to specific receptors. To test this hypothesis, we have measured plasma SM levels and SM receptors on circulating mononuclear cells obtained from 5 achondroplastic dwarfs and 5 age-matched controls. Plasma SM levels were 0.82 +/- 0.14 U/ml (mean +/- S.E.M.) for the achondroplastic dwarfs and 0.90 +/- 0.12 U/ml for the controls. The specific binding of 125I-SM to 50 x 10(6) mononuclear cells was 7.66 +/- 1.11% for the dwarf group and 7.66 +/- 1.16% for the controls. Achondroplastic and control cells possessed equal numbers of receptor sites and identical receptor affinity for SM. The data indicate that plasma SM levels and SM binding to circulating mononuclear cells are normal in achondroplastic dwarfs and suggest a primary intrace-lular defect.     

Production and isolation of recombinant somatomedin C

High-level production of a growth promoting peptide hormone somatomedin C (insulin-like growth factor I) has been achieved using recombinant DNA techniques in Escherichia coli. We found a new structural protein, designated as LH, to stabilize somatomedin C in vivo, and constructed expression vectors for somatomedin C fusing to LH through a methionine and through a tryptophan, respectively. Each of the fused proteins was produced at approximately 4.5 X 10(5) molecules per single E. coli cell and comprised more than 20% of the total cellular proteins. Somatomedin C was obtained in high yield by limited cyanogen bromide degradation of the methionine-type fused protein, in spite of somatomedin C itself having a Met at the 59th position, followed by renaturation of the resultant reduced peptide. The tryptophan-type fused protein was also converted to the peptide hormone by treating with 3-bromo-2-nitrophenylsulphenyl skatole or N-chlorosuccinimide. The recombinant somatomedin C obtained by these procedures was identical with the native one in amino acid sequence as well as in biological activity of stimulation of DNA synthesis in BALB/c 3T3 cells.     

Specific 125I-somatomedin receptor on circulating human mononuclear cells

Somatomedin is a growth hormone-stimulated peptide purified from plasma which has $\text{in vitro}$ actions on cartilage and other tissues. Specific receptor sites for somatomedin distinct from insulin receptor sites have been described for many tissues. This study demonstrates the existence of a specific ¹²⁵I-somatomedin receptor site on circulating human mononuclear cells. This will provide an easily accessible human source for the study of changes in somatomedin receptors in disease states.     

Association between serum insulin, serum somatomedin and liver receptors for human growth hormone in streptozotocin diabetes

Streptozotocin-induced diabetes in the female rat caused a decrease in the serum level of somatomedin (Sm), measured by radioreceptor assay. The decrease was reversed by insulin therapy. In diabetes of varying severity, serum insulin and Sm levels showed highly significant association up to the insulin concentration (18 microU/ml) corresponding to normal serum Sm (1 U/ml). Similarly, the hepatic binding of human growth hormone (hGH) showed highly significant association with serum Sm levels up to the degree of binding (7% of tracer) corresponding to normal serum Sm. Binding of hGH to normal liver was about 12% of tracer. These results suggest that insulin might regulate serum Sm via its effect on liver lactogenic receptors, and that about half of these receptors are "spare", or in excess of those required to maintain normal serum Sm levels.     

Comparative studies on human placental insulin and basic somatomedin receptors

The disuccinimidy! suberate, affinity-labeling procedure, and proteolytic mapping techniques have been employed to characterize further the human placental receptors for insulin and basic somatomedin. Electrophoretic analysis of the basic somatomedin receptor, selectively crosslinked to ¹²⁵I basic somatomedin in the presence of excess native insulin revealed, under reducing conditions, major labeled constituents of 270-280 and 125-140 kd, substantiating our previous work employing a photoaffinity labeling reagent. Affinity labeling also demonstrated the presence of less intensely labeled components with apparent molecular weights of 40 and 45 kd but failed to reveal a distinct 90- to 100-kd species observed in parallel experiments with insulin. In the absence of β-mercaptoethanol, all components specifically labeled with ¹²⁵I basic somatomedin migrated in the 300- to 400-kd range. In comparison, selective affinity labeling of the insulin receptor in the presence of excess native basic somatomedin revealed components, upon electrophoresis under reducing conditions, with apparent molecular weights of 270-280, 125-140, 90-100, and 40 kd. The major insulin-labeled component (125-140 kd) comigrated with the major constituent (125-140 kd) selectively labeled with basic somatomedin. When digestion was performed prior to solubilization, chymotryptic and tryptic proteolysis of the membrane-localized selectively labeled insulin, and basic somatomedin receptors yielded quite similar gel electrophoretic maps. However, when digestion was done subsequent to solubilization, chymotryptic and tryptic proteolysis of selectively labeled insulin and basic somatomedin receptors solubilized in SDS yielded similar but not identical gel electrophoretic maps. We conclude that the receptors for basic somatomedin and insulin are highly homologous structures with respect to their disulfide crosslinked composition, and with respect to the size of the major components detected by selective affinity-labeling procedures. Nevertheless, the detection of electrophoretically distinct labeled receptor components upon analysis of specifically labeled intact or proteolytically digested receptors points to subtle differences between the polypeptide compositions of the two receptors.     

Comparative effects of somatomedin C and insulin on the metabolism and growth of cultured human fibroblasts

At concentrations of 25 ng/ml in serum-free medium, somatomedin C (SM-C) and insulin stimulated 3H-thymidine incorporation in adult human fibroblasts 4- and 1.5-fold, respectively. The presence of 0.25% human hypopituitary serum (HHS), which by itself had little effect, enhanced the mitogenicity of both SM-C and insulin. Furthermore, 10(-7)M dexamethasone dramatically potentiated SM-C stimulation (70-fold) and insulin stimulation (28-fold) of 3H-thymidine incorporation. With dexamethasone and 0.25% HHS, significant stimulation of DNA synthesis was seen at 2.5 ng/ml for both SM-C and insulin. The effects of SM-C and insulin on 3H-thymidine incorporation were additive. These 3H-thymidine incorporation results were clearly supported by cell replication studies. On the other hand, SM-C and insulin had equivalent, nonadditive effects on RNA and protein synthesis and protein degradation. Half-maximal effects were seen for both peptides on all three metabolic processes at 2-5 ng/ml. In contrast to their synergism with SM-C in the stimulation of DNA synthesis and cell replication, HHS and dexamethasone did not enhance SM-C stimulation of RNA or protein synthesis or protein degradation. These data indicate that SM-C and insulin stimulate DNA, RNA, and protein synthesis, protein degradation, and cell replication in adult human fibroblasts at nanomolar concentrations, suggesting that each peptide is capable of acting through its own receptor. Both SM-C and insulin are also capable of synergism with low concentrations of serum and dexamethasone in the stimulation of DNA synthesis and cell replication. It is proposed that SM-C and insulin both participate in the regulation of cell growth and metabolism in vivo.     

Aberrant regulation of the metabolism of the insulin receptor in Swarm rat chondrosarcoma chondrocytes.

Treatment of Swarm rat chondrosarcoma chondrocytes for 3 days in media containing either non-recombinant pig or recombinant human insulin (1 micrograms/ml) increased the rate of proteoglycan synthesis approximately 6-fold compared with cells cultured in the absence of insulin. The concentrations of human and pig insulin that stimulated the cells to double their rate of proteoglycan synthesis were approximately 1 ng/ml and approximately 2 ng/ml respectively. Because physiological concentrations of insulin do not influence proteoglycan synthesis in non-transformed chondrocytes, the findings indicated a possible abnormality in the insulin-dependent regulation of the insulin receptor in these tumour cells. Like most cells, chondrosarcoma chondrocytes down-regulated their insulin receptors when incubated with insulin for 30 min. However, the number of plasma-membrane and intracellular insulin receptors did not decrease when the chondrocytes were exposed to insulin chronically for 4 days. Chondrocytes were cultured in media containing 2H-, 13C- and 15N-labelled amino acids, and the heavy-isotope density-shift method was used to investigate both the rate of degradation and the rate of synthesis of the insulin receptor. Although the rate of synthesis of the receptor was slightly faster in the insulin-treated cultures, as assessed by a slightly faster rate of appearance of the 'heavy' receptor, the rate of degradation of the receptor was slower in the insulin-treated cultures. The half-lives for the 'light' receptors were approx. 18 h and 10 h for chondrocytes cultured in insulin-containing and insulin-free media respectively. These studies in vitro indicate that the apparent up-regulation of insulin receptors that occurs in this transformed cell upon long-term exposure to insulin is primarily the result of a decreased rate of receptor degradation.     

Evidence that somatomedin is synthesized by multiple tissues in the fetus

Production of somatomedin-C, a growth hormone-dependent peptide believed to mediate the growth-promoting actions of growth hormone, has been assessed using explants of fetal mouse tissues. Quantitation of this peptide in media of explants cultured for 3 days has been accomplished with a membrane receptor assay for somatomedin and a specific radioimmunoassay for somatomedin-C. Somatomedin-C is produced by the 11-day-gestation fetal mouse liver, increases exponentially in parallel with liver growth until the 16th day of gestation, and falls postnatally. Media somatomedin is believed to be derived by de novo synthesis since saline extracts of liver and most other fetal tissues contain only a small fraction of the activity in culture media. The immunoreactive material secreted into media appears to be closely related to human somatomedin-C since it produces dilution curves which are parallel to those of pure hormone, migrates on Sephacryl 200 at a size similar to that of one of the components of human serum somatomedin-C, dissociates into small molecular weight material with acid treatment, and isofocuses in a range comparable with that of somatomedin-C purified from human serum. Eleven-day limb bud mesenchymal micromass cultures and 17-day-gestation intestine, heart, brain, kidney, and lung also synthesize immunoreactive somatomedin-C in serum-free medium. For these tissues, the media activity was far in excess of the tissue extractable activity. Somatomedin activity in excess of the tissue extractable activity, however, was not found in media from 17-day-gestation placenta. The finding that multiple tissues synthesize somatomedin-C raises the possibility that the primary biological actions of this hormone are exerted locally at its sites of origin. Although a function of this type by a peptide has not been widely suspected, it seems plausible that the cells of fetal tissues are capable of producing local mitogens in much the same manner as the postulated inducers of tissue differentiation.     

Characterization of a somatomedin (insulin-like growth factor) synthesized by fetal rat liver organ cultures.

Explants of 19- to 20-day fetal rat liver synthesize polypeptides biochemically and immunologically related to the well characterized somatomedin (insulin-like growth factor) BRL-MSA, multiplication-stimulating activity. Fetal MSA was purified from media conditioned by fetal liver explants by chromatography on Sephadex G-75 under acid conditions. Partially purified fetal MSA: 1) inhibited the binding of BRL-MSA to the MSA receptor of rat liver plasma membranes, to somatomedin-binding proteins from rat serum, and to rabbit anti-BRL-MSA serum; 2) had a molecular weight of 4,500 to 12,500 determined by polyacrylamide gel electrophoresis in sodium dodecyl sulfate; 3) stimulated the incorporation of [3H]thymidine into the DNA of chick embryo fibroblasts and induced cell multiplication; 4) stimulated glucose oxidation in rat adipocytes and weakly inhibited the binding of insulin to the insulin receptors of IM-9 lymphocytes; and 5) stimulated sulfate uptake in costal cartilage from hypophysectomized rats. These activities were associated with the same molecular species in fetal MSA preparations following disc acrylamide electrophoresis and co-migrated with active BRL-MSA peptides.     

Effects of insulin and somatomedin C on the function of cultured Leydig cells

Porcine cultured Leydig cells (LC) lose hCG receptors and hCG responsiveness (cAMP and testosterone) when they are cultured for three days in a defined medium without insulin or somatomedin C (Sm-C) (Insulin-like growth factor I). In the presence of insulin (50 ng/ml) or of Sm-C (10 ng/ml) the loss of the hCG receptor number and the decreased cAMP response to hCG were prevented, but the steroidogenic response to hCG was only partially prevented. This parameter became normal when cells were pretreated with either Sm-C (10 ng/ml) plus insulin (50 ng/ml) or with insulin alone at high concentrations (5 micrograms/ml). These results indicate that both Sm-C and insulin acting through their own receptors increase Leydig cell steroidogenic capacity by increasing hCG receptor number and improving some step beyond cAMP formation.     

Effect of somatomedin A and insulin on cyclic AMP generation in isolated rat hepatocytes

Insulin and somatomedin A were shown to have inhibitory action on glucagon stimulated but not basal cyclic AMP production in isolated rat hepatocytes. The inhibition was dose-dependent and the potency per mol was about 100 fold higher for insulin than for somatomedin A.     

Insulin down-regulates insulin receptor number and up-regulates insulin receptor affinity in cells expressing a tyrosine kinase-defective insulin receptor

We examined the effect of insulin treatment on HTC cells transfected with large numbers of either normal insulin receptors (HTC-IR) or insulin receptors defective in tyrosine kinase (HTC-IR/M-1030). In both HTC-IR and HTC-IR/M-1030 cells, 20 h of insulin treatment (1 microM) at 37 degrees C resulted in a 65% decrease in the number of binding sites with a reciprocal 6-fold increase in affinity. In contrast, treatment with 10 nM insulin (20 h, 37 degrees C) also increased receptor affinity but had a smaller effect on the number of binding sites. 125I-Insulin binding to soluble receptors from HTC-IR and HTC-IR/M-1030 cells pretreated with insulin showed results similar to those obtained in intact cells. In both HTC-IR and HTC-IR/M-1030 cells, insulin enhanced insulin receptor degradation. In HTC-IR/M-1030 cells a 1-h incubation with insulin did not change receptor number and had only a small effect on receptor affinity; also there was no effect of insulin after a 20-h incubation at 15 degrees C. Inhibiting protein synthesis by pretreatment with cycloheximide (100 microM) did not block either the decrease in receptor number or the increase in receptor affinity. Both HTC-IR and HTC-IR/M-1030 cells exhibited a very slow rate of insulin and insulin receptor internalization and no differences were seen in this parameter when HTC-IR cells were compared to HTC-IR/M-1030 cells. These studies indicate, therefore, that in cells expressing kinase-defective insulin receptors, insulin down-regulates insulin receptor number via enhanced receptor degradation, and up-regulates receptor affinity. These effects were time- and temperature-dependent, but not dependent on new protein synthesis, and suggest that activation of tyrosine kinase may not be a prerequisite for certain mechanisms whereby insulin regulates its receptor.     

The insulin-like growth factor, somatomedin C, induces the synthesis of cholesterol side-chain cleavage cytochrome P-450 and adrenodoxin in ovarian cells

The actions of insulin and somatomedin C (insulin-like growth factor I) on cholesterol side-chain cleavage activity and the synthesis of cytochrome P-450scc and adrenodoxin were investigated in primary cultures of swine ovarian (granulosa) cells. Nanomolar concentrations of pure human somatomedin C stimulated biosynthesis of progesterone and 20 alpha-hydroxypregn-4-en-3-one. Moreover, in the presence of exogenous sterol substrate for cholesterol side-chain cleavage, somatomedin C significantly enhanced pregnenolone biosynthesis in a time- and dose-dependent manner. This augmentation of functional cholesterol side-chain cleavage activity was accompanied by a dose-dependent (2-16-fold) increase in [35S]methionine incorporation into specific immunoprecipitable cytochrome P-450scc and adrenodoxin. Micromolar concentrations of insulin (but not proinsulin or desoctapeptide) also induced synthesis of cholesterol side-chain cleavage constituents by 4-7-fold. These results demonstrate that an insulin-like growth factor, somatomedin C, exerts discrete differentiating effects on ovarian cells characterized by increased synthesis of immunospecific cytochrome P-450scc and adrenodoxin. Thus, we infer that somatomedin C may serve a critical role in the differentiation of steroidogenic cells in the mammalian ovary.     

Comparison of inhibition by a tumor promoter (12-O-tetradecanoylphorbol-13-acetate) of expression of the differentiated phenotype of chondrocytes in rabbit costal chondrocytes in culture with inhibition by retinoic acid

Both retinoids and the tumor promoter 12-O-tetradecanoylphorbol-13-acetate (TPA) inhibit expression of the differentiated phenotype by rabbit costal chondrocytes in culture, as judged by morphological changes and decreased sulfation of glycosaminoglycans (GAG). However, the inhibition of the differentiated phenotype of chondrocytes in TPA-treated cells is restored by parathyroid hormone (PTH), while the inhibition by retinoids is not [Takigawa et al. (1982) Mol. Cell. Biochem. 42, 145-153; Takigawa et al. (1983) Cell Differ. 13, 283-291]. In the present study, we examined the difference between TPA-treated chondrocytes and retinoic acid-treated chondrocytes to determine the mechanism of the restoration of the differentiated phenotype in de-differentiated cells treated with TPA. PTH increased the activity of ornithine decarboxylase [ODC; EC 4.1.1.17], a rate limiting enzyme of polyamine biosynthesis, and proteoglycan synthesis in chondrocytes pretreated with TPA as well as in normal chondrocytes. The maximal stimulations of ODC activity and GAG synthesis were observed 4 h and 24-36 h, respectively, after addition of PTH. The dose-response curve for ODC induction by PTH was parallel to that of PTH-stimulated proteoglycan synthesis both in TPA-treated chondrocytes and in normal chondrocytes. PTH also increased the intracellular cyclic AMP level after 2 min in TPA-treated cells as in normal cells. Addition of dibutyryl cyclic AMP (DBcAMP) induced ODC and restored proteoglycan synthesis in TPA-treated cells. The dose-response curve for induction of ODC by DBcAMP was parallel to that of DBcAMP-stimulated proteoglycan synthesis in both TPA-treated chondrocytes and normal chondrocytes. On the other hand, the increases by PTH in the intracellular cyclic AMP level, ODC activity, and proteoglycan synthesis were inhibited in chondrocytes pretreated with a combination of TPA and retinoic acid as well as in those pretreated with retinoic acid alone. TPA stimulated the syntheses of DNA and RNA in chondrocytes but did not increase the cyclic AMP level or ODC activity. PTH and DBcAMP inhibited the syntheses of DNA and RNA both in TPA-treated cells and in normal cells. These results suggest that ODC induction mediated by elevation of cyclic AMP plays an important role in re-differentiation of de-differentiated cells pretreated with these agents.     

Ontogeny of somatomedin during development in the mouse. Serum concentrations, molecular forms, binding proteins, and tissue receptors

The purpose of this study was to assess the ontogeny of serum concentrations and molecular forms of somatomedin during fetal and postnatal development and to define the changes in serum binding proteins for somatomedin-C during various stages of development. The finding that fetal, placental, and decidual mouse tissues possess receptors for somatomedin suggests a role for somatomedin in fetal growth and possibly in the maintenance of pregnancy. Serum somatomedin-C was measured using a highly specific, heterologous radioimmunoassay (RIA) and a less specific membrane binding assay (MBA) which is more sensitive to the influence of somatomedins other than somatomedin-C. The assays were validated for mouse serum by showing that serum concentrations were reduced in genetically growth hormone-deficient mice and in hypophysectomized mice and were increased by growth hormone therapy. As in the human, the RIA measures only a portion of the somatomedin-C present in mouse serum. This “covering up” of somatomedin is attributed to the presence of serum binding proteins and is corrected by treatment of serum samples with acid. By both RIA and MBA, serum somatomedin concentrations are low in fetal and newborn mice, begin to rise in the fourth postnatal week, and reach adult values by 7 weeks of age. The chromatographic pattern of adult mouse serum on Sephacryl 200 is similar to that observed with human sera: The immunoreactive material elutes at apparent molecular weights of 140,000 and 30,000–40,000. The elution profile of ¹²⁵I-labeled somatomedin-C bound to components of serum is nearly identical to the pattern of endogenous activity. As with human serum, somatomedin-C in acidified mouse serum elutes at a lower molecular weight, coincident with insulin and purified somatomedin-C. Maternal serum somatomedin declines in the last half of gestation at the time when placental lactogen levels rise. Along with the absolute decline in somatomedin content is the appearance of unsaturated sites on somatomedin binding proteins. These findings are unexpected and unexplained since somatomedin rises late in pregnancy in humans and several lines of evidence suggest that placental lactogen has the capacity to stimulate somatomedin production. We previously have presented evidence that explants of multiple fetal mouse tissues synthesize somatomedin-C. The present study shows that the immunoreactive somatomedin-C in fetal mouse serum shares identical characteristics with those reported previously for media obtained from mouse liver explants. It seems possible that somatomedin's actions are exerted primarily at or near its site of production and that circulatory levels do not reflect the importance of somatomedin-C on fetal growth. While elucidation of the dramatic developmental changes in serum content and molecular forms of somatomedin-C and in somatomedin binding proteins may be essential to clarifying the role of somatomedin on fetal growth, proof that somatomedin stimulates fetal growth will depend in large part on studies of its biological actions on fetal tissues.