The trafficking of retinoids in the retina represents a model to study soluble hormone-binding proteins in a complex system subject to profound evolutionary adaptations. Although a remarkable illustration of convergent evolution, all visual systems detect light in the same way, that is through the photoisomerization of an 11-cis retinoid to a corresponding trans isomer. What is strikingly different between the systems, is the mechanism by which the 11-cis chromophore is reformed and visual pigment regenerated in a process known as the visual cycle. The variations of the cycle address a problem inherent to retinoids themselves. That is, the properties that make these molecules suited for light detection also account for their susceptibility to oxidative and isomeric degradation, and cellular toxicity. The cycle therefore provides an opportunity to examine the role of soluble hormone-binding proteins within an integrative and evolutionary context. The present review focuses on interphotoreceptor retinoid-binding protein (IRBP), a controversial glycolipoprotein that recruits a protein fold common to Cterminal-processing proteases and the crotonase family. This unorthodox retinoid-binding protein is entrapped in the subretinal compartment of those eyes that translocate visual cycle retinoids between the photoreceptors and the retinal pigment epithelium. Recent studies suggest that we should look beyond a strictly carrier function if we are to appreciate the role of IRBP in the visual cycle. Here we draw lessons from other soluble hormone-binding proteins to anticipate avenues of future research likely to provide insight into the structure and function of IRBP in vision.

Diminished GH secretion is a well known association of obesity. As in obese humans, Zucker fatty rats develop a progressive GH deficiency, present at 6 weeks of age and maximal at 10 to 12 weeks. The aim of this study was to investigate the GH dependence of IGF-I gene expression in liver and extrahepatic tissues of the obese Zucker rat as a model of progressive GH reduction during adult life. Six- and 11-week-old obese Zucker rats and their lean littermates were used to compare body weight, glycemia, insulinemia, serum GH and IGF-I levels and IGF-I mRNA expression in liver, heart, aorta, kidney and skeletal muscle. In comparison with lean controls, obese Zucker rats showed at both ages comparable glycemia, severe hyperinsulinemia (mU/ml, mean+/-s.e.m.; 6 weeks 138+/-10 vs 45+/-6 P<0.001; 11 weeks 147+/-14 vs 46+/-3, P<0.001) and lower GH (ng/ml; 6 weeks 1.7+/-0.9 vs 2.7+/-1.1; 11 weeks 1.5+/-0.9 vs 4.2+/-1.2) in the presence of similar circulating IGF-I levels (ng/ml; 6 weeks 774+/-26 vs 694+/-28; 11 weeks 1439+/-182 vs 1516+/-121). Hepatic IGF-I mRNA expression was already reduced at 6 weeks of age due to a significant decrease in the IGF-Ib transcript compared with lean controls (relative units; IGF-Ia: 99+/-2% vs 100+/-5%; IGF-Ib: 69+/-10% vs 100+/-2%, P<0.05) and this reduction was more marked in 11-week-old animals when both IGF-I transcripts were significantly diminished (relative units; IGF-Ia: 80+/-6% vs 100+/-1%, P<0.05; IGF-Ib: 65+/-5% vs 100+/-2%, P<0.01). Extrahepatic tissues expressed almost exclusively the IGF-Ia transcript, the amount of which relative to controls was: (1) similar at 6 weeks and decreased at 11 weeks in kidney and skeletal muscle extracts (relative units; kidney: 6 weeks 88+/-10% vs 100+/-2%; 11 weeks 76+/-3% vs 100+/-4%, P<0.05; vastus lateralis: 6 weeks 95+/-7% vs 100+/-10%; 11 weeks 59+/-4% vs 100+/-2%, P<0.001); (2) similar at both ages in thoracic aorta (relative units; 6 weeks 121+/-6% vs 105+/-5%; 11 weeks: 91+/-14% vs 100+/-4%); and (3) increased at both ages in left ventricle extracts (relative units; 6 weeks 114+/-2% vs 99+/-9%, P<0. 05; 11 weeks 119+/-7% vs 95+/-3%, P<0.05). -specific dependence of IGF-I mRNA on GH levels during adulthood, reflected by the different behavior of IGF-I expression for each tissue in conditions of progressive decrease of GH levels.

ABSTRACT

Pimozide (1 mg/kg per day), bromocriptine (1 mg/kg per day) or domperidone (0·1 mg/kg per day) administered daily to rats from day 21 did not change the age at which vaginal opening occurred, nor did they affect the body weight at that age. Therefore the evolution of prolactin levels was different in these three groups. The pimozide-treated group showed high prolactin levels measured on day 23, at vaginal opening and at first oestrus. In the bromocriptine-treated group, levels were undetectable on the day of vaginal opening. Chronic treatment with domperidone failed to increase prolactin levels on day 23 and at vaginal opening. Nevertheless, large increases were observed after a single injection of domperidone at both 21 and 30 days of age.

A significant increase in LH observed on day 23 in the pimozide-treated group was the only effect on gonadotrophin levels which was detected. Ovarian weights were unaffected by the treatments, whereas adrenal weight was increased in the bromocriptine-treated group and decreased in the pimozide- and domperidone-treated groups.

Female rats grafted on day 21 with one additional pituitary gland from adult (90 days) or young (21 days) donors showed a similar advancement in the time of vaginal opening, although the animals bearing an adult pituitary gland showed higher prolactin levels than those observed in animals grafted with young pituitary glands.

This study suggested that the onset of puberty is not closely linked with the evolution of prolactin levels and that the hormone itself is not indispensible for the process.

J. Endocr. (1984) 101, 63–68

Numerous studies have suggested that ovarian hormones are able to modulate insulin sensitivity, but their exact role remains unclear. We have investigated whether different doses of 17beta-oestradiol mediate changes in insulin sensitivity and if these changes could be related to modifications of insulin receptor substrate-1 (IRS-1). Female rats were ovariectomized and later separated into three groups: untreated; treated with a dose of 17beta-oestradiol sufficient to reproduce gestational plasma concentrations of 17beta-oestradiol (group E); and treated with a dose 100 times greater than that given to group E (group E2). A euglycaemic-hyperinsulinaemic clamp was used to measure insulin sensitivity. Changes in IRS-1 were analysed by Western blotting and RT-PCR assays. In group E we found a decrease in insulin sensitivity between days 11 and 16 of treatment as in late gestation, whereas in the untreated group and group E2, development of insulin resistance was observed throughout the treatment. In contrast, whereas in group E2 insulin resistance throughout the hormonal treatment was related to diminished expression and phosphorylation of IRS-1, in group E the decrease in insulin sensitivity between days 11 and 16 of treatment was not related to a decrease in IRS-1 expression. Our results suggest that the effects of oestradiol on insulin sensitivity were dose-dependent and that the insulin resistance associated with a high dose of 17beta-oestradiol was related to downregulation of IRS-1 expression.

Abstract

To address a possible role of type 1 and 2A serine/threonine protein phosphatases (PP1 and PP2A) in regulating granulosa cell hormonal responses, we investigated the effects of okadaic acid (OA) on FSH- and cAMP-induced steroidogenesis in these cells. When added alone (0·01–1 nmol/l), the cell-permeant phosphatase inhibitor did not affect progesterone and 3β-hydroxysteroid dehydrogenase/Δ^5–4 isomerase (3β-HSD) enzyme activity, whereas when added with FSH it dose-dependently augmented (minimal effective dose, 0·1 nmol/l) gonadotropin-stimulated steroidogenesis in cultured granulosa cells. A similar stimulatory effect of the toxin was observed in cells cultured for 48 h with the cell-permeant analogue dibutyryl cAMP (1 mmol/l), or when granulosa cells were stimulated with the cAMP-inducing agents cholera toxin (1 μg/ml), forskolin (15 μmol/l) or 1-methyl-3-isobutyl-xanthine (0·1 mmol/l). The observed effect of OA on FSH-supported granulosa cell steroidogenesis was not a consequence of increased cAMP generation, and time course experiments also revealed that a minimal time period of 12 h was necessary for OA (0·1 and 1 nmol/l) to significantly enhance FSH-induced progesterone and 3β-HSD enzyme activity. Since OA also inhibits the dephosphorylation of protein kinase C (PKC) substrates, we also compared the effect of OA and the PKC activator 12-0-tetradecanoylphorbol-13-acetate (TPA) on FSH-induced granulosa cell steroidogenic activity. While activation of the PKC pathway with the tumor promoter TPA (10 nmol/l) inhibited progesterone and cAMP accumulation in FSH-stimulated granulosa cells, treatment with OA augmented steroidogenesis and did not affect gonadotropin-induced cAMP generation. Collectively these results suggest that PP1 and PP2A may be important in regulating the phosphorylation state of proteins implicated in the cAMP-protein kinase A-stimulated steroidogenic activity of these cells.

Journal of Endocrinology (1997) 152, 131–139

ABSTRACT

The in-vivo regulatory effect of androgens on steroidogenesis was investigated. Adult (2 to 3 months old) hypophysectomized rats were treated intratesticularly with increasing doses of 5α-dihydrotestosterone (DHT; 10–200 μg/100 g body weight) or vehicle (50 μl dimethyl sulphoxide; DMSO) in the contralateral testis. Intratesticular testosterone concentrations were extremely low in hypophysectomized rats 15–20 days after surgery. Treatment with DHT caused a dose-dependent inhibition of testicular 3β-hydroxysteroid dehydrogenase/Δ^5–4 isomerase (3β-HSD) 2 h later, and this effect was apparent at the dose of 20 μg/100 g body weight (P < 0·01). The inhibitory effect of 3β-HSD was not due to a possible interference of DHT in the enzyme assay, since various concentrations of the androgen (0·1–100 μmol/l) were ineffective as inhibitors of 3β-HSD. The highest dose of DHT used in this study (200 μg/100 g body weight) resulted in a rapid (1–2 h) and transient (4–6 h) inhibition (approximately 80%) of 3β-HSD activity. Pretreatment of rats with the antiandrogen cyproterone acetate (5 mg/rat) or the protein synthesis inhibitor cycloheximide (10 mg/rat) did not affect the enzyme activity of testes injected with DMSO, but counteracted the inhibitory effect of DHT on 3β-HSD activity in the contralateral testis. The results presented suggest that the inhibitory effect of the non-aromatizable androgen DHT is receptor-mediated and involves the synthesis of a factor(s) that modulates 3β-HSD activity.

Journal of Endocrinology (1992) 133, 237–243

ABSTRACT

A hypothalamic site of action has been hypothesized for the inhibitory effect of chronic stress on gonadotrophin secretion. The aim of the present study was to examine the temporal changes in hypothalamic LHRH content and gonadotrophin secretion during restraint stress, and the pituitary responsiveness to LHRH stimulation in chronically stressed rats. Adult male rats were killed after being restrained for 0, 20, 45, 90, 180 and 360 min or for 6 h daily over 2, 3 and 4 days. After 20–45 min of stress there was an increase in plasma concentrations of LH (P<0·01) and a decrease in hypothalamic LHRH content (P<0·01), suggesting a negative correlation between plasma LH and hypothalamic LHRH concentrations. Plasma concentrations of FSH were also increased by restraint, but the FSH response was slower and less than the plasma LH response, being significant after 90 min of restraint. Plasma LH and FSH and hypothalamic LHRH concentrations were decreased in chronically stressed rats. In rats restrained for 6 h daily over 4 days, the response of plasma gonadotrophins to administration of 500 ng LHRH was enhanced 45 min after the injection. On the basis of these observations we concluded that in the intact rat, stress may acutely stimulate LHRH and gonadotrophin secretion, and the inhibitory effect of chronic stress on plasma LH and FSH seems not to be due to a reduction in pituitary responsiveness to LHRH, but rather to a decrease in LHRH secretion.

Journal of Endocrinology (1990) 124, 241–246

The mechanism for the development of insulin resistance in normal pregnancy is complex and is associated with serum levels of both progesterone and 17beta-estradiol. However, it remains unclear whether estrogens alone or progestins alone can cause insulin resistance, or whether it is a combination of both which produces this effect. We attempted to determine the role played by progesterone and/or 17beta-estradiol on the phenomena of sensitivity to insulin action that take place during pregnancy in the rat. Ovariectomized rats were treated with different doses of progesterone and/or 17beta-estradiol in order to simulate the plasma levels in normal pregnant rats. A euglycemic/hyperinsulinemic clamp was used to measure insulin sensitivity. At days 6 and 11, vehicle (V)- and progesterone (P)-treated groups were more insulin resistant than 17beta-estradiol (E)- and 17beta-estradiol+progesterone (EP)-treated groups. Nevertheless, at day 16, the V, EP and E groups were more resistant to insulin action than the P group. On the other hand, the V, EP and E groups were more insulin resistant at day 16 than at day 6, whereas the P group was more insulin resistant at day 6 than at day 16. Our results seem to suggest that the absence of female steroid hormones gives rise to a decreased insulin sensitivity. The rise in insulin sensitivity during early pregnancy, when the plasma concentrations of 17beta-estradiol and progesterone are low, could be due to 17beta-estradiol. However, during late pregnancy when the plasma concentrations of 17beta-estradiol and progesterone are high, the role of 17beta-estradiol could be to antagonize the effect of progesterone, diminishing insulin sensitivity.

The use of GH to treat heart failure has received considerable attention in recent years. Although the mechanisms of its beneficial effects are unknown, it has been implicated in the regulation of apoptosis in several cell types, and cardiomyocyte apoptosis is known to occur in heart failure. We therefore decided to investigate whether GH protects cardiomyocytes from apoptosis. Preliminary experiments confirmed the expression of the GH receptor (GHR) gene in primary cultures of neonatal rat cardiomyocytes (PC), the specific binding of GH by HL-1 cardiomyocytes, and the GH-induced activation of GHR and its classical downstream effectors in the latter. That GH prevented the apoptosis of PC cells deprived of serum for 48 h was shown by DNA electrophoresis and by Hoechst staining assays in which GH reduced the percentage of cells undergoing apoptosis. Similarly, the TUNEL-evaluated pro-apoptotic effect of cytosine arabinoside (AraC) on HL-1 cells was almost totally prevented by pre-treatment with GH. Fluorescence-activated cell sorter (FACS) analysis showed apoptosis in 9.7% of HL-1 cells growing in normal medium, 21.1% of those treated with AraC and 13.9% of those treated with AraC+GH, and that GH increased the percentage of AraC-treated cells in the S/G(2)/M phase from 36.9% to 52.8%. GH did not modify IGF-I mRNA levels or IGF-I secretion in HL-1 cells treated with AraC, and the protection afforded by GH against AraC-induced apoptosis in HL-1 cells was not affected by the presence of anti-IGF-I antibodies, but was largely abolished by the calcineurin-inhibiting combination cyclosporin+FK506. GH also reduced AraC-induced phosphorylation of mitogen-activated protein kinase p38 (MAPK p38) in HL-1 cells. In summary, GH protects PC and HL-1 cells from apoptosis. This effect is not mediated by IGF-I and may involve MAPK p38 as well as calcineurin.

Women with polycystic ovary syndrome (PCOS) are often insulin resistant and have chronic low-level inflammation. The purpose of this study was to determine the effects of hyperglycemia in vitro on tumor necrosis factor (TNF)-α release from mononuclear cells (MNC) in PCOS. Twelve reproductive-age women with PCOS (six lean, six obese) and 12 age-matched controls (six lean, six obese) were studied. Insulin sensitivity (IS_HOMA) was estimated from fasting levels of glucose and insulin and percent truncal fat was determined by dual energy absorptiometry (DEXA). TNFα release was measured from MNC cultured under euglycemic and hyperglycemic conditions. IS_HOMA was higher in obese women with PCOS than in lean women with PCOS (student’s t-test; 73.7 ± 14.8 vs 43.1 ± 8.6, P < 0.05), but similar to that of obese controls. IS_HOMA was positively correlated with percent truncal fat (r=0.57, P < 0.04). Obese women with PCOS exhibited an increase in the percent change in TNFα release from MNC in response to hyperglycemia compared with obese controls (10 mM, 649 ± 208% vs 133 ± 30%, P < 0.003; 15 mM, 799 ± 347% vs 183 ± 59%, P < 0.04). The TNFα response directly correlated with percent truncal fat (r=0.45, P < 0.03) and IS_HOMA (r=0.40, P < 0.05) for the combined groups, and with plasma testosterone (r=0.60, P < 0.05) for women with PCOS. MNC of obese women with PCOS exhibit an increased TNFα response to in vitro physiologic hyperglycemia. MNC-derived TNFα release may contribute to insulin resistance and hyperandrogenism, particularly when the combination of PCOS and increased adiposity is present.