Mater, a maternal effect gene required for early embryonic development in mice.

Maternal effect genes produce mRNA or proteins that accumulate in the egg during oogenesis. We show here that Mater, a mouse oocyte protein dependent on the maternal genome, is essential for embryonic development beyond the two-cell stage. Females lacking the maternal effect gene Mater are sterile. Null males are fertile.

Growth hormone treatment induces mammary gland hyperplasia in aging primates.

The decline of growth hormone (GH) and insulin-like growth factor I (IGF-I) production during aging has been likened to the decrease in gonadal steroids in menopause. The repletion of GH/IGF-I levels in aging individuals is suggested to restore the lean tissue anabolism characteristic of youth. In addition to anabolic effects on musculo-skeletal tissues, GH also stimulates mammary glandular growth in some species, although its effects on primate mammary growth remain unclear. Some clinical observations implicate GH in human mammary growth, for example, gynecomastia occurs in some children treated with GH (ref. 6), and tall stature and acromegaly are associated with an increased incidence of breast cancer. To investigate the effects of GH/IGF-I augmentation on mammary tissue in a model relevant to aging humans, we treated aged female rhesus monkeys with GH, IGF-I, GH + IGF-I or saline diluent for 7 weeks. IGF-I treatment was associated with a twofold increase, GH with a three- to fourfold increase, and GH + IGF-I with a four'-to fivefold increase in mammary glandular size and epithelial proliferation index. These mitogenic effects were directly correlated with circulating GH and IGF-I levels, suggesting that either GH or its downstream effector IGF-I stimulates primate mammary epithelial proliferation.

Placental glucose transporter gene expression and metabolism in the rat.

In situ hybridization was used to evaluate patterns of gene expression for glucose transporters 1-4 (GT1-4) in the rat uteroplacenta from the time of implantation through term, and in vivo regional placental glucose metabolism was measured by 14C-labeled 2-deoxyglucose uptake. GT1 mRNA was highly abundant and GT3 was barely detected in the postimplantation decidual reaction. GT1 and 3 mRNAs were colocalized in the labyrinthine syncitiotrophoblast layer of the chorioallantoic placenta, which forms the membranous barrier between maternal and fetal circulations. The level of labyrinthine GT3 mRNA showed no change from midgestation through term; however, the volume of the labyrinth and hence total GT 3 gene expression increased greatly during this period. Labyrinthine GT1 mRNA levels, in contrast, showed significant diminution near term. GT1 mRNA was also localized in the placental growth plate, or junctional zone, where it was most abundant during the period of rapid placental growth and was decreased at term. Placental glucose metabolism, as reflected by steady-state 2-deoxyglucose uptake, was highest in the junctional zone during the rapid growth phase during midgestation, and decreased significantly at term, in parallel with GT1 gene expression. These findings suggest that GT1 is responsible for supplying glucose for use as a placental fuel and that GT3 is important for glucose transfer to the embryo.

Distinctive anatomical patterns of gene expression for cGMP-inhibited cyclic nucleotide phosphodiesterases.

Type III cGMP-inhibited phosphodiesterases (PDE3s) play important roles in hormonal regulation of lipolysis, platelet aggregation, myocardial contractility, and smooth muscle relaxation. We have recently characterized two PDE3 subtypes (PDE3A and PDE3B) as products of distinct but related genes. To elucidate their biological roles, in this study we compare cellular patterns of gene expression for these two enzymes during rat embryonic and postnatal development using in situ hybridization. PDE3B [corrected] mRNA is abundant in adipose tissue and is also expressed in hepatocytes throughout development. This mRNA is also highly abundant in embryonic neuroepithelium including the neural retina, but expression is greatly reduced in the mature nervous system. Finally, PDE3B [corrected] mRNA is localized in spermatocytes and renal collecting duct epithelium in adult rats. PDE3B mRNA is highly expressed in the cardiovascular system, including myocardium and arterial and venous smooth muscle, throughout development. It is also abundant in bronchial, genitourinary and gastrointestinal smooth muscle and epithelium, megakaryocytes, and oocytes. PDE3A [corrected] mRNA demonstrates a complex, developmentally regulated pattern of gene expression in the central nervous system. In summary, the two different PDE3s show distinctive tissue-specific patterns of gene expression suggesting that PDE3B [corrected] is involved in hormonal regulation of lipolysis and glycogenolysis, while regulation of myocardial and smooth muscle contractility appears to be a function of PDE3A [corrected]. In addition, the present findings suggest previously unsuspected roles for these enzymes in gametogenesis and neural development.

Cellular distribution of insulin-degrading enzyme gene expression. Comparison with insulin and insulin-like growth factor receptors.

Insulin-degrading enzyme (IDE) hydrolyzes both insulin and IGFs and has been proposed to play a role in signal termination after binding of these peptides to their receptors. In situ hybridization was used to investigate the cellular distribution of IDE mRNA and to compare it with insulin receptor (IR) and IGF-I receptor (IGFR) gene expression in serial thin sections from a variety of tissues in embryonic and adult rats. IDE mRNA is highly abundant in kidney and liver, tissues known to play a role in insulin degradation. IDE and IR mRNAs are highly coexpressed in brown fat and liver. The highest level IDE gene expression, on a per cell basis, is found in germinal epithelium. IDE and IGFR mRNAs are colocalized in oocytes, while IDE is colocalized with the IGF-II receptor in spermatocytes, suggesting that IDE may be involved with degradation of IGF-II in the testis. In summary, IDE expression demonstrates significant anatomical correlation with insulin/IGF receptors. These data are compatible with a role for IDE in degrading insulin and IGFs after they bind to and are internalized with their respective receptors and may also suggest a novel role for IDE in germ cells.

Androgens stimulate early stages of follicular growth in the primate ovary.

The concept that androgens are atretogenic, derived from murine ovary studies, is difficult to reconcile with the fact that hyperandrogenic women have more developing follicles than normal-cycling women. To evaluate androgen's effects on primate follicular growth and survival, normal-cycling rhesus monkeys were treated with placebo-, testosterone-(T), or dihydrotestosterone-sustained release implants, and ovaries were taken for histological analysis after 3-10 d of treatment. Growing preantral and small antral follicles up to 1 mm in diameter were significantly and progressively increased in number and thecal layer thickness in T-treated monkeys from 3-10 d. Granulosa and thecal cell proliferation, as determined by immunodetection of the Ki67 antigen, were significantly increased in these follicles. Preovulatory follicles (> 1 mm), however, were not increased in number in androgen-treated animals. Follicular atresia was not increased and there were actually significantly fewer apoptotic granulosa cells in the T-treated groups. Dihydrotestosterone treatment had identical effects, indicating that these growth-promoting actions are mediated by the androgen receptor. These findings show that, over the short term at least, androgens are not atretogenic and actually enhance follicular growth and survival in the primate. These new data provide a plausible explanation for the pathogenesis of "polycystic" ovaries in hyperandrogenism.

Heart disease in Turner syndrome.

Turner syndrome (TS) is a relatively common disorder of female development caused by loss of all or part of one sex chromosome. Because short stature and premature ovarian failure are cardinal features of the syndrome, pediatric endocrinologists have taken the lead in care for these girls. Congenital cardiovascular disease affects approximately 50% of individuals and is the major cause of premature mortality in adults. Unfortunately, teenage girls are often lost to follow up after discharge from pediatric clinic. This review describes the spectrum of cardiovascular defects in TS with emphasis on identifying patients at risk for aortic dissection/rupture. Updated consensus guidelines for cardiac screening and care are reviewed and genetic pathways implicated in Turner cardiovascular disease, including premature coronary artery disease, are discussed. This material is of particular importance because cardiac care for adults with TS appears seriously deficient at present.

Expression of Catenin family members CTNNA1, CTNNA2, CTNNB1 and JUP in the primate prefrontal cortex and hippocampus.

Members of the catenin family of proteins are thought to play a major role in the folding and lamination of the cerebral cortex. We have used in situ hybridization to determine the cellular expression patterns of four members of this family, Alpha-E-, Alpha-N-, Beta-, and Gamma-catenins (CTNNA1, CTNNA2, CTNNB1, and JUP respectively) in the adult primate dorsolateral prefrontal cortex (DLPFC) and hippocampus. CTNNA2, CTNNB1, and JUP mRNAs were detected in all layers of the DLPFC and in all neuronal subregions of the hippocampal formation, however CTNNA1 mRNA, coding for an 'epithelial' specific catenin, was not detected in any region of the cortex or hippocampus. CTNNA2, a 'neuronal-specific' catenin, and CTNNB1 mRNAs were abundant in both DLPFC and hippocampus, with a distinct neuronal localization. CTNNA2 mRNA was concentrated in both granular/stellate cells and large pyramidal cell bodies, while CTNNB1 expression was more strongly associated with granular cell bodies throughout the DLPFC, with expression in pyramidal cells confined mainly to cortical Layers III and VI. CTNNA2 and CTNNB1 mRNAs were also abundant in the granule cells of the dentate gyrus and pyramidal cells of Ammon's horn, apparently co-expressed in the same neurons. JUP mRNA was rather diffusely localized in the DLPFC without the distinct laminar patterns seen for CTNNA2 and CTNNB1 but was distinctly localized in the granule cells of the dentate gyrus and pyramidal cells of Ammon's horn. These studies demonstrate a distinct neuronal pattern of gene expression for catenin family members in primate brain structures characterized by high degrees of folding and strong lamination. The high level expression of these transcripts supports the notion of a major role for catenins even in the adult brain. Such an understanding is also important in view of the multiple interactions that catenins have with many other proteins in the adult and ageing brain. This may also have implications for understanding the pathogenesis of neurodegenerative diseases such as Alzheimer's disease, as well as emerging neuronal stem cell therapies.

Developmental and physiological pattern of aldose reductase mRNA expression in lens and retina.

Aldose reductase (AR), an enzyme which converts glucose to sorbitol, has been implicated in the pathogenesis of diabetic cataracts and retinopathy. The normal physiological role of this enzyme in ocular tissue, however, remains unclear. In a developmental study in the rat using in situ and Northern hybridization analyses, we have found that there is a high level of AR mRNA expression in optic cup and lens as early as embryonic day 13. Serial sections through whole embryos at this stage showed that the eye was the only site of AR mRNA hybridization. Levels of AR mRNA declined in the retina as differentiation proceeded and were very sparse there postnatally. As lens development progressed, epithelial AR mRNA levels remained high, especially in the germinative zone, which is the source of the cells that will become lens fibers, and in the bow region, where these cells undergo a dramatic morphogenetic differentiation into lens fibers. AR mRNA was undetectable in terminally differentiated lens fibers. Since it has been suggested that AR-catalyzed sorbitol production could be an osmoprotective device of lens epithelium during systemic hyperosmolar stress, AR mRNA levels from dehydrated hyperosmolar rats were compared with euvolemic control values, and no difference was found. In summary, AR appears to be of particular importance in the development of the eye, with its retinal role receding relative to lens as differentiation is completed. A continued high level of expression in lens epithelium in adulthood may be explained by the fact that lens tissue, unlike retina, normally continues to proliferate and differentiate after birth. The temporal and spatial pattern of distribution of AR mRNA is strongly suggestive of a role for this enzyme in lens fiber morphogenesis.

Regulation of carboxypeptidase H gene expression in magnocellular neurons: response to osmotic stimulation.

Carboxypeptidase H (CPH) is a peptide-processing enzyme thought to be involved in the synthesis of many neuropeptides, including vasopressin (VP) and oxytocin (OT). In this study, employing in situ hybridization histochemistry, we have shown that CPH mRNA is abundantly expressed in the magnocellular paraventricular and supraoptic nuclei of the hypothalamus, the primary sites of OT and VP synthesis. Since this enzyme is copackaged in secretory vesicles and hence coreleased with the neurohypophysial hormones, enzyme stores are depleted in parallel with the peptide hormones during states of hypersecretion. Chronic osmotic stimulation, such as occurs in long-term salt-loading or in diabetes insipidus in the Brattleboro rat, causes depletion of neurohypophysial hormone stores and is accompanied by increased rates of neurohypophysial hormone transcription and translation. This study has shown that the expression of CPH mRNA is also significantly increased in oxytocin and vasopressin producing magnocellular neurons during chronic osmotic stimulation of the hypothalamic-neurohypophysial system. CPH mRNA levels in other peptidergic areas of the brain are not significantly changed by osmotic stimulation. These findings illustrate a coordinate regulation of the transcription of peptide hormones and an enzyme required for the hormones' posttranslational processing.

Developmental and physiological regulation of aldose reductase mRNA expression in renal medulla.

Aldose reductase (AR), an enzyme that catalyzes the conversion of glucose to sorbitol, has been implicated in the pathogenesis of many of the complications of diabetes mellitus, but its normal physiological function in various tissues remains uncertain. It has been suggested that in the kidney, sorbitol production may be an important cellular protection against medullary intersitital hypertonicity. Using in situ and Northern hybridization analyses, we found that at the time of birth, AR mRNA expression in the kidney was very low and seen only in the papilla. By 12 days of age, at about the time a corticopapillary osmotic gradient and the capacity for urinary concentration have developed, a striking increase in renal AR mRNA levels was seen. It was confined to the inner medulla and was characterized by a dramatic gradient of expression paralleling the corticopapillary osmotic gradient. Levels of expression were somewhat lower in adults, but showed the same inner medullary boundary and gradient. Under these hybridization and exposure conditions, no AR transcripts were detected in the outer medulla or cortex. Homozygous Brattleboro rats with congenital diabetes insipidus have relatively dilute corticopapillary osmotic gradients, and their level of medullary AR mRNA was significantly lower than that of controls. Conversely, normal rats made hyperosmotic and, hence, antidiuretic by salt loading showed a large increase in medullary AR mRNA. These changes in renal medullary AR gene expression in correlation with changes in medullary tonicity support the hypothesis that renal AR plays a role in cellular adaption to osmotic stress and suggest that local medullary osmolarity may regulate the level of AR gene expression.

Cellular pattern of insulin-like growth factor-I (IGF-I) and type I IGF receptor gene expression in early organogenesis: comparison with IGF-II gene expression.

To investigate the potential role(s) of the insulin-like growth factors (IGFs) in embryogenesis, we have used in situ hybridization histochemistry to localize mRNAs for IGF-I, IGF-II, and the type I IGF receptor during an early period in rat embryonic development (embryonic days 14 and 15). IGF-I and IGF-II mRNAs were found in distinctly different patterns of cellular distribution. IGF-I mRNA was particularly abundant in undifferentiated mesenchymal tissue in the vicinity of sprouting nerves and spinal ganglia, and in circumscribed regions of the developing face that corresponded to the target zones of the trigeminal nerve. IGF-I mRNA was also found in aggregations of mesenchyme surrounding, but not in developing muscle and cartilage. IGF-I mRNA was selectively concentrated in areas of active tissue remodeling, such as the cardiac outflow tract, and was undetectable in liver, pituitary, and nervous system at this early stage of organogenesis. IGF-II mRNA was abundant in developing muscle, cartilage, and vascular tissue, and in the embryonic liver and pituitary. IGF-II mRNA was also conspicuous in areas of vascular interface with the brain, such as the choroid plexus and the organum vasculosum of the lamina terminalis. Messenger RNA for the type I IGF receptor was widely distributed in embryonic tissues, but the highest level were seen in the ventral floorplate of the hindbrain, where specialized neuroepithelial cells act as guides for axonal targeting. In conclusion, the different cellular patterns of expression of genes for IGF-I and IGF-II indicate that these two IGFs are differently regulated and, thus, may have significantly different roles in the process of embryonic development. Furthermore, the early and widespread expression of the type-I IGF receptor gene, in contrast to the relatively limited and localized pattern of IGF-I gene expression, is consistent with the view that this receptor may mediate the effects of IGF-II as well as IGF-I during embryogenesis.

Anatomy of the pituitary insulin-like growth factor system.

In situ hybridization was used to map patterns of gene expression for components of the insulin-like growth factor (IGF) system, including IGF-I and -II, IGF-binding proteins 1-5 (IGFBP1-5), the IGF-I receptor, and GH in the rat pituitary. IGF-I mRNA was concentrated in isolated cells scattered throughout the gland with features typical of nonendocrine folliculo-stellate cells. IGF-II mRNA was abundant in neural (NL) and intermediate lobe (IL) capillaries, and low levels were present in endocrine cells of anterior lobe (AL) and IL. IGFBP1 mRNA was not detected in the pituitary. IGFBP2 mRNA was concentrated in epithelial cells lining AL follicles and in astroglial-like cells (pituicytes) of the NL. IGFBP3 mRNA was localized in isolated cells scattered throughout the AL and NL. IGFBP4 mRNA was relatively abundant in NL pituicytes and was diffusely expressed in the AL. IGFBP5 mRNA was equally abundant in NL and AL, and was localized in folliculo-stellate and epithelial cells of the AL and pituicytes and capillaries of the NL. Neither IGF-I nor IGFBP1-5 were detected in the IL. IGF-I receptor mRNA was abundant and homogeneously distributed throughout the AL and IL, compatible with expression by endocrine cells. There was overlap, but no particular correlation, between IGF system gene expression and GH-producing cells, which were clustered in the dorsal-lateral wings of the AL. In summary, IGF system gene expression is bountiful in the rat pituitary, but does not correlate with sites of GH synthesis. IGF-I receptor mRNA, which might have been expected to localize to somatotrophs, appears to be equally abundant in all of the endocrine cells of both AL and IL; the other constituents of the IGF system are localized in connective tissue and support elements that demonstrate no special anatomical relationship to somatotrophs. Finally, there is remarkably abundant gene expression for IGFBP2, -4, and -5 in the NL.

Insulin-like growth factors cross the blood-brain barrier.

Although evidence exists that insulin may cross the blood-brain barrier, little is known about the ability of insulin-like growth factors (IGF-I and -II) to cross this barrier. In the present studies, equimolar concentrations of equal specific activity 125I-labeled IGF-I, IGF-II, or insulin were infused into the carotid artery of anesthetized adult rats. The perfusions were carried out for 3 min in the presence or absence of excess unlabeled ligand or insulin, with three or more animals in each group. Immediately after the perfusion, brains were frozen and sectioned for autoradiography. All ligands were detected in choroid plexus, median eminence, and blood vessels, but [125I]IGF-I and -II were also prominently localized in brain parenchyma. Densitometric analysis of film autoradiographs (28-day exposure for all ligands) revealed that radiolabeled IGFs, especially IGF-I, were significantly more abundant throughout the forebrain than [125I]insulin, especially in the paraventricular nucleus, where [125I]IGF-I was 10-fold and [125I]IGF-II was 5-fold more abundant than [125I]insulin. The difference in [125I]IGF-I vs. [125I]insulin accumulation was confirmed by parallel measurements of radioactivity in anatomically matched brain sections using a gamma-spectrometer. The uptake of radiolabeled IGF-I, IGF-II, and insulin by brain parenchyma and vasculature was completely inhibited by excess (1,000-fold) unlabeled ligand; however, insulin (10,000-fold excess) did not completely abolish [125I]IGF-I and -II accumulation. Microscopic evaluation of nuclear emulsion-coated brain sections revealed that radioactivity associated with [125I]IGF-I and -II perfusions was selectively concentrated in capillaries and medium-sized parenchymal cells in the paraventricular nucleus and, to a lesser extent, the supraoptic nucleus and anterior nucleus of the thalamus, whereas in other brain regions the radioligands were mostly bound to capillaries. These results suggest that radiolabeled IGF-I and -II bind to brain capillaries and cross the blood-brain barrier into brain parenchyma more readily than radiolabeled insulin.

Ischemic injury induces brain glucose transporter gene expression.

As neurons rely almost exclusively on glucose as an energy substrate, glucose transport is of critical importance to cerebral function. Two specific facilitative glucose transporters, GT1 and -3, predominate in brain, with the latter exclusively expressed by neurons, whereas GT1 is expressed by astrocytes and vascular elements. Little is known about the regulation of these transporters at the genetic level or the extent to which their expression may change in response to acute or chronic changes in metabolic demands. Thus, we employed in situ hybridization to evaluate changes in glucose transporter gene expression in the rat brain in response to ischemia induced by middle cerebral artery occlusion (MCAO). The most remarkable responses were demonstrated by GT1, which within an hour of ischemic insult demonstrated a global increase in gene expression throughout the forebrain. In the ensuing hours, GT1 expression further intensified and became lateralized to the lesioned hemisphere, with normalization of expression contralaterally. Increased GT1 mRNA levels were found in astroglia and microvessels and were also present in distinct neuronal populations, including the piriform cortex, dentate gyrus, and medial habenula, which normally do not express GT1 mRNA. By 24 h post-MCAO, glial cells of the ipsilateral cortex surrounding the infarct zone still demonstrated elevated GT1 mRNA levels, but expression had returned to baseline in neurons. Interestingly, it was not until GT1 expression had subsided (24 h post-MCAO), that there was a modest increase in neuronal GT3 gene expression in the affected hemisphere. GT2 and GT4 mRNAs were not detected in the rat brain under normal conditions or after ischemia. These data demonstrate that ischemia induces an immediate and sustained increase in brain GT1 gene expression in both glial cells and neurons. This augmentation of GT1 expression could represent a defensive strategy aimed at repletion of the brain's energy stores and stabilization of neuronal membrane potential.

Dynorphin A inhibits and naloxone increases the electrically stimulated release of oxytocin but not vasopressin from the terminals of the neural lobe.

Oxytocin release from the rat neurohypophysis is under endogenous opioid inhibition. It has recently been established that dynorphin precursor-derived peptides are colocalized with vasopressin (VP) in the secretory granules in nerve terminals of the neural lobe, and that the opiate receptors in the neural lobe are restricted to the kappa-subtype. Therefore, we hypothesized that dynorphin, which is copackaged and thus coreleased with VP, is the endogenous opioid that inhibits release from neighboring oxytocin (OT) terminals. To test this hypothesis we examined the effects of dynorphin-(1-8), dynorphin-(1-17), and naloxone on the electrically stimulated release of OT and VP from isolated rat neurointermediate lobes throughout a range of stimulus frequencies. Both dynorphin-(1-8) and -(1-17) (2 microM) produced a substantial reduction in OT release during a 4-Hz stimulus, and this effect was abolished by naloxone (10 microM). Neither form of dynorphin, however, affected OT secretion at a stimulus frequency of 12 or 30 Hz at concentrations up to 10 microM. Naloxone (10 microM) by itself did not affect OT release during the 4-Hz stimulus, but it produced a substantial increase in OT release at a stimulus frequency of 12 Hz. In contrast, neither form of dynorphin produced inhibition, nor did naloxone augment VP secretion at any frequency tested. Frequency-dependent secretion curves (4, 8, 12, 20, and 30 Hz) for OT and VP in the presence and absence of naloxone indicated that the degree of naloxone augmentation of OT release at a given stimulus frequency was positively correlated with the amount of VP release at that frequency. These data support the hypothesis that dynorphin released in parallel with VP during in vitro stimulations of the rat neurohypophysis simultaneously inhibits stimulated OT release.

Renal growth hormone receptor gene expression: relationship to renal insulin-like growth factor system.

In order to elucidate potential sites of direct GH action on the kidney, we used in situ hybridization to localize GH receptor (GHR) gene expression during the course of development and in the adult rat. In order to illuminate potential interactions between GH and insulin-like growth factor-I (IGF-I) in regulating renal function, we compared the anatomical localization of GHR messenger RNA (mRNA) with that for the IGF-I receptor and for IGF-I in the rat kidney. Low levels of GHR mRNA were present in the kidney from before birth and increased in abundance until postnatal day 40. Hypophysectomy resulted in a decrease and GH treatment resulted in an increase in renal GHR mRNA levels. Renal GHR mRNA was most abundant in the proximal straight tubule, with lesser levels present in the medullary thick ascending limb (MTAL), and it was not detected in the glomerulus or inner medulla. In contrast, IGF-I receptor mRNA was concentrated in the glomerulus, distal nephron and collecting system. The only point of convergence for GHR and IGF-I receptor mRNAs was in the MTAL, where IGF-I mRNA was localized. This segregation of GHR and IGF-I receptor gene expression in the kidney suggests that each hormone has distinct spheres of action along the nephron, with GH acting directly on the proximal straight tubule, whereas IGF-I may act on the glomerulus, distal nephron, and collecting duct. GHR expression in the MTAL, which is the site of renal IGF-I synthesis, supports the view that GH has a direct effect on renal IGF-I synthesis. Finally, it appears that in the kidney, as in other GH-sensitive tissues, GH may regulate its receptor levels.

Cellular patterns of insulin-like growth factor system gene expression in murine chondrogenesis and osteogenesis.

In situ hybridization histochemistry was used to map cellular patterns of gene expression for the insulin-like growth factor (IGF) system in developing murine skeleton from embryonic day 15 (E15) through postnatal day 25 (P25). IGF-I receptor and IGF-II receptor messenger RNAs (mRNAs) are both selectively concentrated in developing chondrocytes and osteoblasts. IGF-II and IGF-binding protein-5 and -6(IGFBP-5 and -6) mRNAs are abundant in mesenchymal condensations and chondroblasts on E15. Chondrocyte IGF-II mRNA levels remain high, but IGFBP-5 and -6 mRNAs decline significantly as cartilage matures. Low levels of IGFBP-6 mRNA are detected in postnatal chondrocytes up to at least P25, but IGFBP-5 mRNA is no longer detected in chondrocytes after E18. IGF-I and IGFBP-2, -3, and -4 mRNAs are detected in surrounding mesenchymal tissue, but are not detected in mesenchymal condensations or chondrocytes at any stage of development. IGFBP-3 mRNA is localized in sprouting capillaries invading the perichondrium and periosteum throughout development. IGF-I, IGF-II, and IGFBP-2, -4, -5, and -6 mRNAs are detected in osteoblasts localized in zones of endochondral ossification from E18 to at least P25. IGFBP-1 mRNA is not detected in cartilage or bone cells at any stage of development. These data confirm the recent report by Shinar et al. that IGF-II, but not IGF-I, mRNA is detected in rat chondrocytes in vivo and show that this pattern also applies to the mouse. The present study demonstrates, for the first time, the cell-specific patterns of IGF-I and -II receptor and IGFBP-2 to -6 gene expression during the processes of chondro- and osteogenesis in vivo. Interestingly, IGF-II, both IGF receptors, and IGFBP-5 and -6 are simultaneously coexpressed in chondrocyte precursors early in skeletal development, suggesting functional interactions between these specific factors in chondrogenesis. Both IGFs, both IGF receptors, and IGFBP-2, -4, -5, and -6 are all expressed in osteoblasts, providing evidence for potential local interactions between these IGF system components in osteogenesis. Thus, 9 of 10 known components of the IGF system demonstrate dynamic cell-specific patterns of gene expression during chondro- and osteogenesis, supporting the view that the IGF system has a complex and integral role within the developing skeleton.

Cellular localization and regulation of gene expression for components of the insulin-like growth factor ternary binding protein complex.

Insulin-like growth factors (IGFs) are present in the circulation, largely as part of a high mol wt complex including IGF-binding protein-3 (IGFBP-3) and an acid-labile subunit (ALS). This study used in situ hybridization to investigate the cellular sites of synthesis of these factors in the rat and to evaluate changes in transcript levels during development and after hypophysectomy and GH treatment. IGFBP-3 transcripts are considerably more abundant and widely expressed than ALS at birth, but both are present in liver and kidney. Hepatic IGFBP-3 gene expression increases slightly, whereas ALS increases dramatically in the first few weeks after birth. IGFBP-3 mRNA is concentrated in portal venous and sinusoidal endothelium, but is not detected in hepatocytes, whereas ALS mRNA is diffusely expressed by hepatocytes, but is not detected in nonparenchymal cells. Both transcripts are localized in the renal cortex; however, IGFBP-3 mRNA is concentrated in interstitial cells, whereas ALS is expressed in proximal tubule epithelium. Hypophysectomy results in a 90% reduction in hepatic ALS and an approximately 50% decrease in IGFBP-3 mRNA level. ALS, but not IGFBP-3, transcripts were also reduced in the kidney. GH receptor mRNA is coexpressed with ALS in liver and kidney, suggesting that the effects of GH on ALS gene expression may be direct. In summary, the fact that IGFBP-3 gene expression is far more widespread than that of ALS in both spatial and temporal parameters suggests that IGFBP-3 has a role apart from contribution to the ternary complex. We have also shown that IGFBP-3 and ALS are synthesized by distinct hepatic cell types in an anatomical organization that may serve to ensure efficient formation of the ternary complex in the blood passing through the sinusoids. Finally, the present data suggest that regulation of ALS synthesis may be the primary site of GH regulation of ternary complex formation.

Up-regulation of pituitary [125I]insulin-like growth factor-I (IGF-I) binding and IGF binding protein-2 and IGF-I gene expression by estrogen.

To evaluate the effects of estrogen on the rat pituitary insulin-like growth factor (IGF) system, binding of [125I]IGF-I and in situ hybridization for IGF-I, the IGF-I receptor and IGF binding protein-2 (IGFBP-2) were coupled with quantitative autoradiography. The groups included intact cycling females, intact males, and gonadectomized males and females with or without estrogen pellet implants. Binding of [125I]IGF-I in the anterior lobe of the pituitary occurred in dense clusters over a diffuse lower density background. [125I]IGF-I binding was significantly increased in the estrogen-treated groups and was highest at proestrus compared to the rest of the estrous cycle. IGF-I receptor messenger RNA (mRNA) was distributed diffusely through the anterior pituitary and was not different between the respective gonadectomized and estrogen-treated groups. IGFBP-2 mRNA was clustered throughout the anterior pituitary and was significantly higher in the estrogen-treated groups as noted above for [125I]IGF-I binding. IGF-I mRNA was diffuse throughout the anterior pituitary and was also significantly higher in the estrogen-treated groups. In the neural lobe, [125I]IGF-I binding, IGFBP-2 mRNA, IGF-I receptor mRNA, and IGF-I mRNA were all uniformly distributed and did not differ between groups. The results show that circulating estrogen differentially regulates components of the pituitary IGF-I system in a region-specific manner and suggest that a portion of IGF binding in the anterior pituitary may be to IGFBP-2.