Expression of insulin-like growth factor-I and its receptor by SV40-transformed rat granulosa cells

Cellular proliferation is a dominant aspect of ovarian follicular development in the rat, and insulin-like growth factor I (IGF-I) has been proposed as a mediator of cellular growth and differentiation in the ovary. An SV40-transformed rat granulosa cell line (RGA-41S) has been established as a model for studies on dividing cells of granulosa origin. Granulosa cells from the ovaries of immature diethylstilbestrol-treated rats were infected with the tsA255 mutant of SV40, followed by cloning in serum-free medium to select transformed cell lines which were serum independent. At the permissive temperature (33 C), RGA-41S cells exhibited a transformed phenotype and rapidly formed high density multilayers of compact cells that readily overgrew nontransformed cells. At the nonpermissive temperature (40 C) cell replication declined and division ceased after 4 days. Furthermore, at 40 C the cells grew as a monolayer and assumed a tetrahedral shape with a high cytoplasm-to-nucleus ratio, and displayed reduced ability to overgrow nontransformed cells. The transformed ovarian cells did not express detectable gonadotropin receptors and steroidogenic activity but retained their epithelial phenotype as demonstrated by cytokeratin staining of the cytoskeleton, the presence of microvilli, and the formation of tight junctions between cells. In support of the proposed autocrine-paracrine actions of IGF-I in the ovary, assay of conditioned serum-free culture medium revealed secretion of IGF-I-immunoreactive material by RGA-41S cells. HPLC-purified IGF-I immunoreactivity from these cells eluted with the same retention time as recombinant human IGF-I. When hybridized with a 32P-labeled rat IGF-I cDNA probe, poly(A)+ mRNA prepared from RGA-41S cells grown at both temperatures showed the typical three size classes of IGF-I mRNA on Northern blots (7.5, 1.7, and 0.8-1.2 kilobase kb), although the levels were somewhat higher at 33 C. The presence of IGF-I receptors in transformed cells was demonstrated by specific 125I-IGF-I binding to intact cells. Scatchard analysis indicated a single class of high affinity receptors at a density of 10(5) binding sites per cell and a dissociation constant (Kd) = 0.52 x 10(-9) M. Furthermore, hybridization of a 32P-labeled IGF-I receptor probe to Northern blots of poly(A+) RNA prepared from cells grown at 33 C and 40 C revealed an 11-kilobase rat IGF-I receptor mRNA. Physiological concentrations of IGF-I increased [3H]aminoisobutyric acid uptake by RGA-41S cells grown at either temperature, attesting to the retention of responsiveness to IGF-I in these transformed granulosa cells.(ABSTRACT TRUNCATED AT 400 WORDS)

Expression and stability of insulin-like growth factor-I (IGF-I) mRNA splicing variants in the GH3 rat pituitary cell line

We have employed Northern blot analyses and solution hybridization/RNase protection assays to evaluate the presence and stability of IGF-I mRNA splicing variants in the GH3 rat pituitary cell line. All of the IGF-I mRNA size classes and IGF-I mRNAs with alternately-spliced 5'-untranslated regions and E-peptide coding regions seen in adult rat liver also were present in GH3 cells, although the proportions of the 5' splicing variants were significantly altered. In actinomycin D-treated cells, all IGF-I mRNA splicing variants were equally stable; thus, changes in the levels of some splicing variants were not due to differential mRNA stability. Additionally, all IGF-I mRNA size classes seen on Northern blots were equally stable; this data suggests that the large IGF-I mRNA species is not a precursor of the smaller species.

Regulation by fasting of rat insulin-like growth factor I and its receptor. Effects on gene expression and binding

We have examined, in liver and extrahepatic tissues, the effects of fasting on total insulin-like growth factor I (IGF-I) mRNA levels, on levels of different IGF-I mRNAs generated by alternative splicing of the primary IGF-I transcript, and on IGF-I receptor binding and mRNA levels. A 48-h fast decreased total IGF-I mRNA levels by approximately 80% in lung and liver, approximately 60% in kidney and muscle, and only approximately 30-40% in stomach, brain, and testes. In heart, IGF-I mRNA levels did not change. The levels of the different splicing variants, however, were essentially coordinately regulated within a given tissue. Specific 125I-IGF-I binding in lung, testes, stomach, kidney, and heart was increased by fasting by approximately 30-100%, whereas in brain 125I-IGF-I binding did not change in response to fasting. In tissues in which fasting increased IGF-I receptor number, receptor mRNA levels increased approximately 1.6- to 2.5-fold, whereas when IGF-I receptor number was unchanged in response to fasting, receptor mRNA levels did not change. These data demonstrate that the change in IGF-I and IGF-I receptor mRNA levels during fasting is quantitatively different in different tissues and suggest that regulation of IGF-I and IGF-I receptor gene expression by fasting is discoordinate.

Rat ovarian insulin-like growth factor I (IGF-I) gene expression is granulosa cell-selective: 5'-untranslated mRNA variant representation and hormonal regulation

We, and others, have recently reported that the ovary is a site of insulin-like growth factor (IGF)I gene expression. It was the objective of the present studies to assess the relative ovarian abundance of IGF-I transcripts with alternative 5'-untranslated (UT) regions, their cellular localization, and hormonal regulation. To this end, a solution hybridization/RNase protection assay was employed wherein total rat ovarian RNA was hybridized with a 404-base 32P-labelled rat IGF-I riboprobe corresponding to the Class A 5'UT variant. As in liver, three protected bands [322 (Class A), 297 (Class B), and 242 (Class C) bases long] were noted, in keeping with established alternative 5' UT transcripts. The ovarian (as the hepatic) Class C variant proved the most abundant. The ovarian Class B variant was barely detectable. Cellular localization studies revealed these ovarian IGF-I transcripts to be primarily, if not exclusively, of granulosa but not theca-interstitial cell origin. Treatment of immature (21-23 days old) hypophysectomized rats with a diethylstilbestrol (DES)-containing subcutaneous silastic implant for a total of 5 days resulted in a 2-fold increase in the (densitometrically quantified) abundance of ovarian IGF-I transcripts, a diametrically-opposed effect (2.6-fold decrease) being noted at the level of the liver. Whereas treatment of hypophysectomized rats with oGH by itself (150 micrograms, qd, sc x5 days) resulted in a 5-fold increase in hepatic IGF-I gene expression, a limited, albeit distinct inhibitory effect was observed on the steady-state levels of ovarian IGF-I mRNA. In contrast, combined treatment with oGH and DES yielded a 3-fold increase in the abundance of ovarian IGF-I transcripts, there being no net alteration in hepatic IGF-I gene expression. Taken together, these findings reveal ovarian expression of the 3 known 5'-UT IGF-I mRNA variants, document the granulosa cell as the main somatic ovarian cell of IGF-I mRNA generation, and indicate that hepatic and ovarian IGF-I gene expression are differentially regulated in diametrically opposed directions.

Regulation of rat brain/HepG2 glucose transporter gene expression by insulin and insulin-like growth factor-I in primary cultures of neuronal and glial cells

We have demonstrated the expression of the rat brain/HepG2 glucose transporter gene in primary cultures of rat neuronal and glial cells by Northern blot analysis with a rat brain glucose transporter cDNA probe. Incubation of both neuronal and glial cells with insulin and insulin-like growth factor-I induced a time- and dose-dependent increase in the steady state levels of glucose transporter mRNA. The maximal response was achieved between 2-4 h and subsequently decreased. Both insulin and insulin-like growth factor-I at a dose of 1 ng/ml elicited an approximately 57% increase in glucose transporter mRNA levels in neuronal cultures after 90 min, suggesting that each peptide was acting through its own receptor. On the other hand, insulin stimulated [3H]2-deoxyglucose uptake in glial, but not neuronal, cells. These results suggest that insulin-like peptides regulate the expression of the rat brain/Hep G2 glucose transporter gene at both transcriptional and posttranscriptional levels, and that these regulatory mechanisms may be dissociated from each other. Insulin-like peptides may, therefore, participate in the control of brain energy metabolism.

TPA-induced neurite formation in a neuroblastoma cell line (SH-SY5Y) is associated with increased IGF-I receptor mRNA and binding

The neuroblastoma cell line SH-SY5Y was cultured in the presence of TPA for three days. Increased neurite formation was noted as early as 24 hours after TPA was added. These changes were associated with an increase in IGF-I receptor binding as well as increased mRNA for the IGF-I receptor.

Xenopus laevis contains two nonallelic preproinsulin genes. cDNA cloning and evolutionary perspective

We undertook the cloning of preproinsulin cDNAs from the South African clawed toad, Xenopus laevis, in order to study the role of insulin during embryogenesis in this species. We found that X. laevis contains two different preproinsulin cDNAs, both of which code for peptides containing 106 amino acids of typical structure but which differ by eight amino acids: one in the signal peptide, two in the B-chain, four in the C-peptide, and one in the A-chain. Southern blot analysis indicates that the two preproinsulin cDNAs identified correspond to two different nonallelic genes which we believe arose through a recent gene duplication within the amphibian radiation possibly during the development of tetraploidy in this species. Both genes are expressed, since we have recently identified the two corresponding insulins in pancreatic extracts of adult toads (Shuldiner, A.R., Bennett, C., Robinson, E.A., and Roth, J. (1989) Endocrinology, in press). These cDNAs represent the first amphibian preproinsulin sequences to be elucidated.

Insulin-like growth factor I messenger ribonucleic acids with alternative 5'-untranslated regions are differentially expressed during development of the rat

Solution hybridization/RNase protection assays were used to study the developmental expression of insulin-like growth factor I (IGF-I) mRNA levels in rats. In liver, heart, and kidney, total IGF-I mRNA levels were low at birth and increased during the 50-day postnatal period, with liver levels increasing by over 100-fold. In contrast, stomach, muscle, and testicular IGF-I mRNA levels were highest at the earliest stages examined (late fetal or early neonatal) and declined thereafter to the levels observed in 50-day-old rats. In brain, IGF-I mRNA levels rose 2-fold during the first week of life and declined over the next 6-7 weeks. Lung IGF-I mRNA levels were highest in 20-day-old fetuses and exhibited some fluctuation during the postnatal period. Alternative splicing in the 5'-untranslated region of the primary rat IGF-I transcript gives rise to three transcripts, classes A, B, and C, which have divergent 5'-untranslated region sequences associated with a common region that encodes the mature IGF-I peptide. These sequences contain upstream in-frame translation initiation codons and may, therefore, encode alternate IGF-I prepropeptides. The class C variant was the predominant mRNA species at all stages of development studied and was the only IGF-I transcript in brain, heart, and muscle. In tissues where multiple 5'-untranslated region splicing variants occurred, therefore, changes in total IGF-I mRNA primarily reflected changes in this splicing variant. However, the class C and class A (as well as class B in liver) transcripts exhibited temporally divergent changes over some developmental intervals. Class A transcripts in the liver, stomach, testes, and lung as well as class B transcripts in liver, exhibited sustained increases from 15 or 22 postnatal days to maximal levels at 50 postnatal days. In kidney, class A transcripts also increased steadily, but beginning at an earlier stage, i.e. at 8-15 days of postnatal life. These results demonstrate that the temporal expression of total IGF-I mRNA in the developing rat occurs in a tissue-specific manner, and additionally, that IGF-I mRNA variants are differentially expressed during development.

Coordinate decrease of tissue insulinlike growth factor I posttranscriptional alternative mRNA transcripts in diabetes mellitus

In these studies, we examined the effect of excess levels of growth hormone (GH) on rat insulinlike growth factor I (IGF-I) gene expression in streptozocin-induced diabetes mellitus. A solution hybridization/RNase protection assay was used to simultaneously quantitate the relative tissue content of the variant IGF-I mRNA species arising from alternative splicing in the region encoding the COOH-terminal extension E-peptide (IGF-Ia and IGF-Ib). IGF-Ia and IGF-Ib mRNAs were markedly decreased in liver, kidney, and lung tissues of diabetic rats. Although GF stimulates IGF-I gene expression, chronic GH excess from implanted somatomammotropic tumors did not appropriately induce tissue IGF-I mRNA content in diabetic animals. Treatment of diabetic rats with insulin for 1 wk restored basal and GH-stimulated IGF-Ia and IGF-Ib mRNA content toward that present in tissues of nondiabetic rats. The ratio of IGF-Ia to IGF-Ib mRNA remained relatively constant for each tissue and was not affected by the diabetic state, chronic GH hyperstimulation, or insulin therapy, suggesting that posttranscriptional splicing is not a regulated event in these conditions. Thus, both circulating IGF-I levels and tissue IGF-I gene expression are profoundly decreased in this model of experimental diabetes. Diminished tissue availability of IGF-I from endocrine and/or paracrine sources may be responsible for the growth retardation seen in uncontrolled diabetes mellitus.

Insulin and IGF-I stimulate phosphorylation of their respective receptors in intact neuronal and glial cells in primary culture

Previous studies have shown that insulin and IGF-I bind to their respective receptors and stimulate autophosphorylation of the receptor beta subunits in detergent extracts of neuronal and glial cells. In the present study, intact neuronal and glial cells in primary culture have been utilized to characterize insulin- and IGF-I-stimulated phosphorylation of their receptors. Following [32P]orthophosphate labelling and stimulation by insulin or IGF-I, the cells were solubilized and the phosphorylated receptors were partially purified on wheat germ agglutinin--agarose columns, and immunoprecipitated using anti-phosphotyrosine or anti-insulin receptor antibodies. Insulin stimulated the phosphorylation of its receptor beta subunit (95 kD phosphoprotein) in a dose-dependent manner, within at least 20 seconds in both neuronal and glial cells. Additionally, a 102-kD phosphoprotein was observed in insulin-stimulated neuronal cells. Maximal stimulation of receptor phosphorylation occurred at 1 minute for the glial cells, and 10 minutes for the neuronal cells. IGF-I stimulated the phosphorylation of two phosphoproteins in intact neuronal and glial cells; a 95-kD protein and a 102-kD protein, in a dose-dependent manner. These observations demonstrate that both insulin and IGF-I stimulate the phosphorylation of the beta subunits of their respective receptors in brain cells in a similar fashion to their effects on receptors from nonneural tissues.

Liver insulin receptor tyrosine kinase activity in a rat model of type II diabetes mellitus and obesity

Spontaneous hypertensive-corpulent rats (SHR/N-corpulent), homozygous for the corpulent gene (cp/cp), are obese, hyperinsulinemic and exhibit abnormal glucose tolerance and thus represent a model for type II diabetes and obesity. In view of their overall insulin resistance, we examined liver insulin receptor binding and tyrosine kinase activity from corpulent rats and lean littermates fed purified diets containing 54% sucrose or starch for about 12 wk. Specific 125I-insulin binding to crude liver membranes from female corpulent rats fed either starch or sucrose was reduced to approximately 50% of that seen in lean rats (14 vs. 7%). Affinity of insulin receptors was similar in all groups, suggesting that hyperinsulinemic corpulent rats possess fewer hepatic insulin receptors than do lean rats. Using similar numbers of wheat germ agglutinin-agarose (WGA)-purified insulin receptors with similar affinities for insulin, it was found that basal and insulin-stimulated phosphorylation of the synthetic tyrosine-specific kinase substrate poly(Glu, Tyr)4:1 was similar in lean and obese rats fed sucrose or starch. It is suggested that the contribution of the liver to the insulin resistance in obese SHR/N-cp rats probably lies distal to the insulin receptor tyrosine kinase.

Insulin and insulin-like growth factor receptors in the nervous system

Insulin and the insulin-like growth factors (I and II) are homologous peptides essential to normal metabolism as well as growth. These peptide hormones are present in the brain, and, based on biosynthetic labeling studies as well as evidence for local gene expression, they are synthesized by nervous tissue as well as being taken up by the brain from the peripheral circulation. Furthermore, the presence of insulin and IGF receptors in the brain, on both neuronal and glial cells, also suggests a role for these peptides in the nervous system. Thus, these ligands affect brain electrical activity, either as neurotransmitters or as neuromodulators, altering the release and re-uptake of other neurotransmitters. The insulin and IGF-I and -II receptors found in the brain exhibit a lower molecular weight than corresponding receptors on peripheral tissues, primarily caused by alterations in glycosylation. Despite these alterations, both brain insulin and IGF-I receptors exhibit tyrosine kinase activity in cell-free systems, as do their peripheral counterparts. Brain insulin and IGF-I receptors are developmentally regulated, with the highest levels appearing in fetal or perinatal life. However, the altered glycosylation of brain receptors does not appear until late in fetal development. The receptors are widely distributed in the brain, but especially enriched in the circumventricular organs, choroid plexus, hypothalamus, cerebellum, and olfactory bulb. These studies on the insulin and IGF receptor in brain, add strong support to the suggestion that insulin and IGFs are important neuroactive substances, regulating growth, development, and metabolism in the brain.

Developmental regulation of rat brain/Hep G2 glucose transporter gene expression

The developmental regulation of rat brain-derived/Hep G2 glucose transporter gene expression was studied by means of Northern blot hybridization, using a rat brain glucose transporter cDNA probe, in order to directly quantify steady state glucose transporter mRNA levels. The results obtained showed different tissue-specific patterns of glucose transporter mRNA levels during ontogenesis; while in brain there was a sustained increase in the levels of the message from 20 days embryogenesis until 50 days postnatal, other organs such as heart, lung, liver, and muscle expressed maximal levels of the glucose transporter mRNA in 20-day fetuses and 1-day neonates, decreasing subsequently to very low levels. The relative expression of the glucose transporter mRNA in the different tissues, at both fetal and adult stages, was analyzed using a solution hybridization-RNase protection assay. This approach revealed that, while the heart expresses the highest levels of glucose transporter mRNA at 20 days of fetal life, the brain shows the highest levels at the adult stage. These results indicate a tissue-specific ontogenic pattern of glucose transporter gene expression, suggesting a developmental role for this glucose transporter gene product.

Dexamethasone reduces steady state insulin-like growth factor I messenger ribonucleic acid levels in rat neuronal and glial cells in primary culture

Insulin-like growth factor I (IGF-I) mRNA was demonstrated in primary cultures of neuronal and glial cells from rat brain. On Northern blots, a rat IGF-I cDNA probe hybridized to RNA species of 7.5, 1.7, and 0.8-1.2 kilobases in total and poly(A)+ RNA from both cell types. Solution hybridization/RNase protection assays were performed using an antisense riboprobe complementary to the 5'-untranslated region as well as part of the coding region of rat IGF-I mRNA. These studies indicated that two of the previously described three possible alternative 5'-untranslated splicing variants (classes A and C) were expressed in neuronal and glial cells, with class C transcripts predominating. Neuronal cells also possessed extremely low levels of class B transcripts. Treatment of neuronal cell cultures with the synthetic glucocorticoid dexamethasone reduced IGF-I mRNA levels by 60%. Glial cell IGF-I mRNA levels were reduced by dexamethasone by up to 40%. These results suggest that glucocorticoid-induced reductions in IGF-I production could occur at the level of transcription and may underlie some of the actions of glucocorticoids in causing growth retardation and inhibition of cell proliferation.

Insulin-sensitive tyrosine kinase is increased in livers of adult obese Zucker rats: correction with prolonged fasting

Adult obese Zucker rats (fa,fa) are hyperinsulinemic and insulin resistant. Specific insulin binding to crude membranes prepared from livers was 2.8% (per mg protein) in fatty animals compared with 7.9% in homozygous lean (Fa,Fa) and 9.0% in heterozygous lean (Fa,fa) animals. Insulin binding increased in liver membranes from fatty animals after a 72-h fast to 6.4%. The reduced insulin binding in livers from fatty rats was associated with elevated insulin-sensitive tyrosine kinase activity, which fell towards control values after the fast. The elevated tyrosine kinase activity was associated with an increased maximum velocity (Vmax) without a change in Michaelis-Menten constant (Km) for its substrates, ATP and poly(Glu,Tyr)4:1. These findings suggest that, in adult fatty rats, insulin-sensitive tyrosine kinase has increased intrinsic activity. Further, the effect of the prolonged fast on both insulin binding and kinase activity, suggest that in this model environmental factors, and not necessarily a genetic abnormality, may regulate liver insulin receptors and their kinase. Whether the inverse relationship of the kinase and insulin receptor number is the result of a compensatory mechanism remains to be elucidated.

Insulin-related materials in the nervous system of vertebrates and non-vertebrates: possible extrapancreatic production

Studies from multiple laboratories with a range of methods raised the possibility that insulin production occurs naturally at extrapancreatic sites. Part A covers the presence of insulin-related materials in organisms that do not have an endocrine pancreas, including unicellular prokaryotes and eukaryotes as well as multicellular non-vertebrate animals (insects et al.) and plants. Part B covers possible production of insulin by extrapancreatic tissues of vertebrates that are remote from a source of pancreatic insulin e.g. early chick embryos and mammalian cells in culture. Part C covers possible extrapancreatic insulin production in mammals in vivo. Each section ends with an outline summary with evidence in favor of and against the hypothesis.

Distribution and regulation of rat insulin-like growth factor I messenger ribonucleic acids encoding alternative carboxyterminal E-peptides: evidence for differential processing and regulation in liver

Alternative splicing of insulin-like growth factor I (IGF-I)/somatomedin C mRNAs generates two IGF-I mRNAs coding for IGF-I peptides with different sequences in the E domain of the IGF-I prohormone. These two mRNAs encode alternative E peptides due to the presence (IGF-Ib) or absence (IGF-Ia) of a 52-base insert in the region coding for the E domain. We have used a solution hybridization/RNase protection assay to determine the tissue distribution and regulation by GH of the expression of these alternative IGF-I mRNAs. IGF-Ib mRNAs are present in low abundance (representing approximately 2.5% of the total IGF-I mRNA) in heart, lung, muscle, testes, stomach, kidney, and brain, but represent approximately 13% of the IGF-I mRNA in liver. GH treatment of hypophysectomized rats increased steady-state IGF-I mRNA levels in liver, kidney, lung, and heart. In kidney, lung, and heart, IGF-Ia and IGF-Ib mRNA levels were coordinately regulated by GH, but, in liver, the fold increase in IGF-Ib mRNA levels was approximately three times greater than the fold increase in IGF-Ia mRNA levels. These data suggest that the processing of IGF-I mRNA in liver is different than in nonhepatic tissues. These results also further elucidate the organization of the rat IGF-I gene as well as the generation of multiple IGF-I mRNAs by alternative splicing.

Insulin receptors in the brain: structural and physiological characterization

The present study was conducted to characterize insulin receptors and to determine the effects of insulin in synaptosomes prepared from adult rat brains. Binding of 125I-insulin to synaptosome insulin receptors was highly specific and time dependent: equilibrium binding was obtained within 60 minutes, and a t1/2 of dissociation of 26 minutes. Cross-linking of 125I-insulin to its receptor followed by SDS-PAGE demonstrated that the apparent molecular weight of the alpha subunit of the receptor was 122,000 compared with 134,000 for the liver insulin receptor. In addition, insulin stimulated the dose-dependent phosphorylation of exogenous tyrosine containing substrate and a 95,000 MW plasma membrane associated protein, in a lectin-purified insulin receptor preparation. The membrane associated protein was determined to be the beta subunit of the insulin receptor. Incubation of synaptosomes with insulin caused a dose-dependent inhibition of specific sodium-sensitive [3H]norepinephrine uptake. Insulin inhibition of [3H]norepinephrine uptake was mediated by a decrease in active uptake sites without any effects in the Km, and was specific for insulin since related and unrelated peptides influenced the uptake in proportion to their structural similarity with insulin. These observations indicate that synaptosomes prepared from the adult rat brain possess specific insulin receptors and insulin has inhibitory effects on norepinephrine uptake in the preparation.

Retinal insulin receptors: localization using a polyclonal anti-insulin receptor antibody

Although retinal insulin receptors have recently been described biochemically, the location of these receptors within the retina is unknown. The study presented here used a polyclonal anti-insulin receptor antibody (B10), immunofluorescence and immunoelectron microscopy to determine the location of insulin receptors in bovine, monkey and human retina. It was found that antibody immunofluorescence formed discrete bands localized predominantly to photoreceptor and neuronal cell bodies. In addition to the strong association with neuronal perikarya, a lower level of antibody binding was observed in photoreceptor outer segments. In human retina, some of the antibody immunofluorescence also had a pattern that suggested B10 binding to glial-like cells.

Characterization of the altered oligosaccharide composition of the insulin receptor on neural-derived cells

Typical insulin receptors are present on neuroblastoma cell lines. High affinity binding for insulin was present in membrane preparations from NG108 (a hybrid mouse neuroblastoma-rat glioma) as well as in membranes from SK-N-MC and SK-N-SH, two human neuroblastoma cell lines. Specific [125I]insulin binding was 24.4% for NG108, 16.9% for SK-N-MC and 5.2% for SK-N-SH at membrane protein concentrations of 0.4 mg/ml. IC50 for [125I]insulin binding was 3.4 nM in NG108 membrane preparations and 0.9 nM for SK-N-SH and 1.8 nM in SK-N-MC membranes. Apparent mol. wt. for the alpha subunits (identified by specific immunoprecipitation using the anti-insulin receptor antiserum B10) on SDS PAGE was 134 kDa for NG108; 124 kDa for SK-N-MC and 120 kDa for SK-N-SH. Neuraminidase digestion increased the mobility of the alpha subunit from both NG108 and SK-N-MC receptors to 120 kDa, whereas that from SK-N-SH were unaffected. Endoglycosidase H and endoglycosidase F digestions increased the mobility of the alpha subunits of all 3 cell lines to varying degrees, suggesting the presence of N-linked glycosylation. Insulin induced autophosphorylation of the insulin receptor beta subunit in WGA-purified membranes from all 3 cell lines. In addition, phosphorylation of a protein with an apparent mol. wt. 105 kDa was stimulated by insulin in WGA purified membranes from NG108. Tyrosine-specific kinase activity was present in the membranes from each cell line and was stimulated by insulin in a dose-dependent manner from 10(-9) to 10(-6) M. Proinsulin was about 100 times less potent in stimulating phosphorylation of the artificial substrate poly (Glu, Tyr)4:1 when compared to insulin in accordance with its lower binding affinity to the insulin receptor. Hexose transport was stimulated by insulin in all 3 cell lines. These results indicate that neuroblastoma cells contain specific insulin receptors and that they may be useful as models for studying the role of insulin in nervous tissue.

Functional insulin-like growth factor I receptors are expressed by neural-derived continuous cell lines

High affinity insulin-like growth factor I (IGF-I) receptors are expressed by two human neural derived cell lines, SK-N-SH and SK-N-MC. Specific [125I]IGF-I binding to crude membranes was 23.4% for SK-N-SH and 10.7% for SK-N-MC, with 50% inhibition of binding by unlabeled IGF-I between 0.6-0.7 nM. Scatchard analysis of crude membrane binding was linear, whereas Scatchard analysis after wheat germ agglutinin purification of the receptor became curvilinear. The IGF-I receptor alpha-subunits of SK-N-SH have an apparent Mr of 126K, whereas that for SK-N-MC is 132K. Despite these differences in alpha-subunit structure both cell lines demonstrate IGF-I-induced autophosphorylation of their own beta-subunits as well as specific IGF-I induced tyrosine kinase activity, suggesting normal coupling between the ligand-binding alpha-subunit and the tyrosine kinase-containing beta-subunit. Furthermore, IGF-I stimulated iododeoxyuridine uptake in both SK-N-SH and SK-N-MC in a dose-dependent manner, suggesting that these cells may be used to study the role of IGF-I action on neural tissues.

Retinal insulin receptors. 1. Structural heterogeneity and functional characterization

Neural cells of the bovine retina contain specific, high-affinity receptors for insulin. When solubilized and wheat-germ purified, these receptors exhibit a kinase activity that is capable of phosphorylating the receptor's beta-subunit (autophosphorylation) and a tyrosine-containing exogenous substrate, poly (Glu, Tyr) 4:1. Studies of the structure of retinal insulin receptors revealed the existence of two insulin receptor subpopulations. For these populations, the apparent molecular weights of the alpha-subunit were 120- and 133 kDa. This structural heterogeneity does not appear to be related to the presence of vascular contamination and stands in contrast to the brain and liver where a single alpha-subunit type was found (120 kDa for brain and 133 kDa for liver). In addition to being distinguishable by their molecular weights, the two populations of retinal insulin receptors could be distinguished in terms of (a) their solubility in Triton X-100, (b) glycosylation, and (c) recognition by anti-insulin receptor antibody. Despite these structural differences, the two populations of retinal insulin receptors appear to have similar insulin binding affinities.