Potential functions of plasma steroid-binding proteins

Expression and regulation of human sex hormone-binding globulin transgenes in mice during development

Human sex hormone-binding globulin (SHBG) is produced by hepatocytes and transports sex steroids in the blood. The rat gene encoding SHBG is expressed transiently in the liver during fetal life, but it is not expressed in the liver postnatally, and the small amounts of SHBG in rat blood are derived from gonadal sources. To study the biosynthesis and function of human SHBG in an in vivo context, we have produced several lines of transgenic mice that contain either 11 kb (shbg11) or 4.3 kb (shbg4) portions of the human shbg locus. The expression and regulation of these transgenes have now been studied during fetal and postnatal development. In situ hybridization of an shbg11 transgenic mouse fetus at 17.5 days postcoitus located human shbg transcripts only in duodenal epithelial cells and hepatocytes. Temporal differences in the hepatic expression of mouse shbg and human shbg transgenes during late fetal development were reflected in corresponding differences in mouse and human SHBG levels in fetal and neonatal mouse blood. Serum concentrations of human SHBG increased during the first weeks of life regardless of gender until about 20 days of age in shbg11 mice, but after this time they continued to increase only in the males. This sexual dimorphism was reflected in corresponding differences in human SHBG messenger RNA (mRNA) abundance in the livers of these animals. However, it was not observed in shbg4 mice, in which hepatic production of plasma SHBG continued to increase after puberty regardless of gender. Serum testosterone and SHBG levels correlated in all sexually mature shbg transgenic mice. Human shbg transcripts were detectable only in testes of shbg11 mice and increased progressively in abundance from 10 days of age until the animal reached sexual maturity at 30 days of age, with appreciable increases occurring well before any changes in serum testosterone concentration. In the kidney, SHBG mRNA levels accumulated earlier in shbg11 than in shbg4 mice, and the expression of both types of transgenes was sexually dimorphic, with much higher SHBG mRNA levels in the kidneys of male mice. As increases in SHBG mRNA in the male kidneys coincided with increases in serum testosterone during sexual maturation, we reasoned that shbg transgene expression is androgen dependent in the kidney. This was confirmed by demonstrating that a decrease in SHBG mRNA abundance in male mouse kidneys after castration could be reversed by 5alpha-dihydrotestosterone treatment. Moreover, exogenous androgen increased human SHBG mRNA levels in the kidneys of female mice. In summary, comparisons of how different human shbg transgenes are expressed in vivo provides information about the positions of potential regulatory sequences that may control the hormonal regulation and tissue-specific expression of this gene during development.

Xenoestrogen interaction with human sex hormone-binding globulin (hSHBG)

This study reports on some environmental chemicals with estrogenic activity (xenoestrogens) and their binding interaction for human plasma sex-hormone binding globulin (hSHBG). The binding affinity constant of these xenoestrogens was measured in equilibrium conditions by solid phase binding assay, and their ability to displace endogenous testosterone and estradiol from hSHBG binding sites was determined with an ammonium sulfate precipitation assay in native plasma from normal men and women. The data showed that some of these xenoestrogens bind hSHBG, with a reversible and competitive binding activity for both [3H]testosterone and [3H]17beta-estradiol and with no apparent decrease in the number of hSHBG binding sites. Their respective binding affinity constants were low, ranging from 0.02 to 7.8 10(5) 1 x mol(-1). However, in native plasma from normal men and women, they were able to dose-dependently increase concentrations of hSHBG-unbound testosterone and/or estradiol. In this study, 4-nonylphenol and 4-tertoctylphenol, two alkylphenols used as surfactants in many commercial products, and bisphenol A and O-hydroxybiphenyl, widely used in the plastics industry, were identified as potent hSHBG-ligands. Additionally, the flavonoid phytoestrogens genistein and naringenin were also identified as hSHBG ligands, whereas their glucoside derivatives, genistin and naringin, had no binding activity for hSHBG. From these data, it is suggested that hSHBG binding may transport some contaminant xenoestrogens into the plasma and modulate their bioavailability to cell tissues. On the other hand, xenoestrogens may also displace endogenous sex steroid hormones from hSHBG binding sites and disrupt the androgen-to-estrogen balance. Whether xenoestrogen SHBG ligands could reach high enough concentrations in the blood to expose humans to any such effect merits further investigation.

Gene expression in the epididymis

The epididymis is a tubular organ exhibiting vectorial functions of sperm concentration, maturation, transport, and storage. The molecular basis for these functions is poorly understood. However, it has become increasingly clear that regional differences along the length of the duct play a role in epididymal physiology and that region-specific gene expression is involved in the formation of these differences. Although not an overtly segmented organ, the epididymis consists of a series of highly coiled "zones," separated by connective tissue septulae and distinct by cell morphology and their pattern of gene expression. Thus, it constitutes an interesting mammalian model to study how pattern formation is achieved by differential gene activity. A large number of epididymis-expressed genes have been cloned and analyzed at the molecular level, most of them have been characterized by a distinct temporal and spatial expression pattern within the organ. Only recently have theories been developed about how and when during ontogenesis this pattern formation takes place and what its significance might be. This review summarizes the current knowledge on regionalized gene expression in the epididymis and presents hypotheses concerning its ontogenetic origin and regulation in the adult.

The additionally glycosylated variant of human sex hormone-binding globulin (SHBG) is linked to estrogen-dependence of breast cancer

Sex Hormone-Binding Globulin (SHBG), the plasma carrier for androgens and estradiol, inhibits the estradiol-induced proliferation of breast cancer cells through its membrane receptor, cAMP, and PKA. In addition, the SHBG membrane receptor is preferentially expressed in estrogen-dependent (ER+/PR+) breast cancers which are also characterized by a lower proliferative rate than tumors negative for the SHBG receptor. A variant SHBG with a point mutation in exon 8, causing an aminoacid substitution (Asp 327-->Asn) and thus, the introduction of an additional N-glycosylation site, has been reported. In this work, the distribution of the SHBG variant was studied in 255 breast cancer patients, 32 benign mammary disease patients, and 120 healthy women. The presence of the SHBG mutation was evaluated with PCR amplification of SHBG exon 8 and Hinf I restriction fragment length polymorphism (RFLP) procedure. This technique allowed us to identify 54 SHBG variants (53 W/v and 1 v/v) in breast cancer patients (21.2%), 5 variants (4 W/v and 1 v/v) in benign mammary disease patients (15.6%), and 14 variants (W/v) in the control group (11.6%). The results of PCR and RFLP were confirmed both by nucleotide sequence of SHBG exon 8 and western blot of the plasma SHBG. No differences in the mean plasma level of the protein were observed in the three populations. The frequency of the SHBG variant was significantly higher in ER+/PR+ tumors and in tumors diagnosed in patients over 50 years of age than in the control group. This observation suggests the existence of a close link between the estrogen-dependence of breast cancer and the additionally glycosylated SHBG, further supporting a critical role of the protein in the neoplasm.

Competitive binding of xenobiotic oestrogens to rat alpha-fetoprotein and to sex steroid binding proteins in human and rainbow trout (Oncorhynchus mykiss) plasma

The ability of a variety of "environmental oestrogens" to compete with radiolabelled steroids to rat alpha-fetoprotein (AFP) and to sex steroid binding proteins was investigated in human and rainbow trout (Oncorhynchus mykiss) plasma. For [3H]oestradiol binding to AFP, diethylstilbestrol and 4-nonylphenoxyacetic acid showed significant competition at concentrations about 100-fold greater than oestradiol (relative binding affinities approximately 1% c.f. oestradiol). All other compounds (phytooestrogens: coumestrol, daidzein, genistein; others: 4-nonylphenol, 4-tert-octylphenol, 4-nonylphenoldiethoxylate, 4-tert-butylphenol, bisphenol-A (Bis-A), bis(2- ethylhexl)phthalate, dioctylphthalate, dibutyl phthalate, 2, 4'DDT (op' enantiomer), 2,4'-DDE (mixed enantiomers), kepone) showed only very weak or no competition (relative binding affinities <<0.1% c.f. oestradiol). The situation for both human and fish plasma was very similar, with only very high concentrations (>>1000 fold more than the natural ligand) of a few of the compounds showing any ability to displace the natural ligand. These results suggest that environmental oestrogenic agents are unlikely to produce biological effects by displacing endogenous steroids from plasma steroid binding proteins unless they are present in very high concentrations.

Brain corticosteroid receptor balance in health and disease

In this review, we have described the function of MR and GR in hippocampal neurons. The balance in actions mediated by the two corticosteroid receptor types in these neurons appears critical for neuronal excitability, stress responsiveness, and behavioral adaptation. Dysregulation of this MR/GR balance brings neurons in a vulnerable state with consequences for regulation of the stress response and enhanced vulnerability to disease in genetically predisposed individuals. The following specific inferences can be made on the basis of the currently available facts. 1. Corticosterone binds with high affinity to MRs predominantly localized in limbic brain (hippocampus) and with a 10-fold lower affinity to GRs that are widely distributed in brain. MRs are close to saturated with low basal concentrations of corticosterone, while high corticosterone concentrations during stress occupy both MRs and GRs. 2. The neuronal effects of corticosterone, mediated by MRs and GRs, are long-lasting, site-specific, and conditional. The action depends on cellular context, which is in part determined by other signals that can activate their own transcription factors interacting with MR and GR. These interactions provide an impressive diversity and complexity to corticosteroid modulation of gene expression. 3. Conditions of predominant MR activation, i.e., at the circadian trough at rest, are associated with the maintenance of excitability so that steady excitatory inputs to the hippocampal CA1 area result in considerable excitatory hippocampal output. By contrast, additional GR activation, e.g., after acute stress, generally depresses the CA1 hippocampal output. A similar effect is seen after adrenalectomy, indicating a U-shaped dose-response dependency of these cellular responses after the exposure to corticosterone. 4. Corticosterone through GR blocks the stress-induced HPA activation in hypothalamic CRH neurons and modulates the activity of the excitatory and inhibitory neural inputs to these neurons. Limbic (e.g., hippocampal) MRs mediate the effect of corticosterone on the maintenance of basal HPA activity and are of relevance for the sensitivity or threshold of the central stress response system. How this control occurs is not known, but it probably involves a steady excitatory hippocampal output, which regulates a GABA-ergic inhibitory tone on PVN neurons. Colocalized hippocampal GRs mediate a counteracting (i.e., disinhibitory) influence. Through GRs in ascending aminergic pathways, corticosterone potentiates the effect of stressors and arousal on HPA activation. The functional interaction between these corticosteroid-responsive inputs at the level of the PVN is probably the key to understanding HPA dysregulation associated with stress-related brain disorders. 5. Fine-tuning of HPA regulation occurs through MR- and GR-mediated effects on the processing of information in higher brain structures. Under healthy conditions, hippocampal MRs are involved in processes underlying integration of sensory information, interpretation of environmental information, and execution of appropriate behavioral reactions. Activation of hippocampal GRs facilitates storage of information and promotes elimination of inadequate behavioral responses. These behavioral effects mediated by MR and GR are linked, but how they influence endocrine regulation is not well understood. 6. Dexamethasone preferentially targets the pituitary in the blockade of stress-induced HPA activation. The brain penetration of this synthetic glucocorticoid is hampered by the mdr1a P-glycoprotein in the blood-brain barrier. Administration of moderate amounts of dexamethasone partially depletes the brain of corticosterone, and this has destabilizing consequences for excitability and information processing. 7. The set points of HPA regulation and MR/GR balance are genetically programmed, but can be reset by early life experiences involving mother-infant interaction. 8. (ABSTRACT TRUNCATED)

Corticotroph axis sensitivity after exercise in endurance-trained athletes

OBJECTIVE

The present study was conducted in order to describe human hypothalamo-pituitary adrenal (HPA) axis adaptation in a model of repeated physical stress (endurance training) that causes a moderate increase in cortisol levels.

SUBJECTS

We performed the same stimulation tests (adrenal stimulation with ACTH or pituitary stimulation with combined CRH/LVP) in a population of 8 endurance-trained athletes in two distinct situations: resting (baseline cortisol values) and 2 h after the end of strenuous exercise (increased cortisol values) to evaluate the HPA axis sensitivity to endogenous sustained increases in cortisol concentrations.

MEASUREMENTS

During these tests, saliva and plasma cortisol (Fs and Fp, respectively) were assessed and compared.

RESULT

Cortisol values in both plasma and saliva at the end of 2 h of exercise were significantly higher than in rested controls: Fs 11.5 +/- 1.3 vs 6.5 +/- 0.8 nmol.l-1 and Fp 428 +/- 36 vs 279 +/- 27 nmol.l-1 (post exercise vs post rest sessions, respectively, P < 0.001 for both). After either hormone test (CRH/LVP or ACTH), cortisol levels in plasma and saliva increased similarly when rest was compared to post exercise. Saliva variations (delta %) under exogenous hormone stimulation were dramatically greater than plasma variations. For example, under ACTH stimulation, the relative increments in cortisol were on control day: delta Fs 980 +/- 139 vs delta Fp 218 +/- 43% (saliva vs plasma, respectively, P < 0.05) and on exercise day: delta Fs 605 +/- 89 vs delta Fp 102 +/- 14% (saliva vs plasma, respectively, P < 0.05).

CONCLUSIONS

In endurance-trained athletes, displaying a moderate but sustained endogenous cortisol increase: (1) ACTH responses following pituitary stimulation are not blunted, (2) cortisol responses following maximal adrenal stimulation are not blunted. Our results favour the hypothesis of a decreased pituitary sensitivity to cortisol negative feedback whereas the hypothesis of a major decreased adrenal sensitivity to ACTH was discarded. The greater ability of saliva assays to detect a cortisol increase strongly supports its use in the study of HPA physiology, whether under basal or dynamic conditions.

Identification and characterization of functional nongenomic progesterone receptors on human sperm membrane

The presence of functional nongenomic progesterone (P) receptors in human spermatozoa has been investigated by equilibrium binding studies in intact spermatozoa, ligand blot and Western blot analysis of sperm lysates, as well as determination of the effects of the steroid on sperm intracellular Ca2+ concentrations. Binding experiments were performed using progesterone-11alpha-glucuronide-[125I]iodotyramine as tracer. Computer analysis of competition curves using different steroids as competitors indicated the presence of two distinct binding sites for P. The high affinity site (Kd in the nanomolar range) appears to be specific for P, whereas the low affinity one (Kd in the micromolar range) binds with equal affinity 11beta-hydroxyprogesterone (11betaOHP) and 17alpha-hydroxyprogesterone (17alphaOHP). A significant correlation exists among affinity constants (as determined by binding studies) and EC50 values for the effects of P, 11betaOHP, and 17alphaOHP on intracellular Ca2+ in fura-2-loaded spermatozoa, strongly indicating the involvement of P-binding sites in the biological effect of the steroid. In particular, dose-response curves for P were biphasic, with an EC50 in the nanomolar range and another in the micromolar range. Conversely, curves for 11betaOHP and 17alphaOHP were monophasic, with an EC50 just in the micromolar range. Ligand blot analysis of sperm total lysates performed with peroxidase-conjugated P revealed the presence of two binding proteins of 54 and 57 kDa that were specific for P. Indeed, peroxidase-conjugated P binding was blocked by the simultaneous presence of the unconjugated steroid. Using alpha c262 antibody, which is directed against the P-binding domain of genomic receptor, we detected two proteins of similar molecular mass (54 and 57 kDa), whereas using antibodies directed against the DNA-binding and N-terminal domains of the genomic P receptors, the two proteins were not detected. In addition, p54 and p57 appear to be mostly localized in sperm membranes and virtually absent in the cytoplasm. The involvement of these proteins in the biological effects of P is indicated by the strong inhibitory effect of alpha c262 on P-induced acrosome reaction of capacitated human spermatozoa.

Sequence and functional relationships between androgen-binding protein/sex hormone-binding globulin and its homologs protein S, Gas6, laminin, and agrin

Androgen-binding protein/sex hormone-binding globulin (ABP/SHBG) is an extracellular binding protein that regulates the bioavailability of sex steroids. ABP/SHBG is closely related to the globular (G) domain of vitamin K-dependent protein S family of proteins and more distantly related to the G domains of several extracellular matrix proteins. ABP/SHBG appears to have evolved from the fusion of two ancestral G domains. Expanding evidence suggests that ABP/SHBG has other functions that are mediated through membrane binding, including signal transduction; however, the types of binding proteins (receptors) have not been identified. Sequence comparisons of ABP/SHBG with G domains of its homologs protein S, Gas6, laminin, and agrin have identified regions of ABP/SHBG that may bind receptors related to homolog receptors. These membrane receptors include beta-integrins, alpha-dystroglycan, and receptor tyrosine kinases. The G domains of laminin and related proteins have clearly evolved from a common ancestor to interact with specific receptors and binding proteins. It remains to be determined if ABP/SHBG followed this evolutionary pathway.

Phyto-oestrogens and Western diseases

Incidences of breast, colorectal and prostate cancer are high in the Western world compared to countries in Asia. We have postulated that the Western diet compared to the semivegetarian diet in some Asian countries may alter hormone production, metabolism or action at the cellular level by some biochemical mechanisms. Our interest has been focused on two groups of hormone-like diphenolic phyto-oestrogens of dietary origin, the lignans and isoflavonoids abundant in plasma of subjects living in areas with low cancer incidence. The precursors of the biologically active compounds detected in man are found in soybean products, whole-grain cereal food, seeds, and berries. The plant lignan and isoflavonoid glycosides are converted by intestinal bacteria to hormone-like compounds. The weakly oestrogenic diphenols formed influence sex-hormone production, metabolism and biological activity, intracellular enzymes, protein synthesis, growth factor action, malignant cell proliferation, differentiation, cell adhesion and angiogenesis in such a way as to make them strong candidates for a role as natural cancer-protective compounds. Their effect on some of the most important steroid biosynthetic enzymes may result in beneficial modulation of hormone concentrations and action in the cells preventing development of cancer. Owing to their oestrogenic activity they reduce hot flushes and vaginal dryness in postmenopausal women and may to some degree inhibit osteoporosis, but alone they may be insufficient for complete protection. Soy intake prevents oxidation of the low-density lipoproteins in vitro when isolated from soy-treated individuals and affect favourably plasma lipid concentrations. Animal experiments provide evidence suggesting that both lignans and isoflavonoids may prevent the development of cancer as well as atherosclerosis. However, in some of these experiments it has not been possible to separate the phyto-oestrogen effect from the effect of other components in the food. The isoflavonoids and lignans may play a significant inhibitory role in cancer development particularly in the promotional phase of the disease, but recent evidence points also to a role in the initiation stage of carcinogenesis. At present, however, no definite recommendations can be made as to the dietary amounts needed for prevention of disease. This review deals with all the above-mentioned aspects of phyto-oestrogens.

Hepatic nuclear proteins that bind cis-regulatory elements in the proximal promoter of the rat corticosteroid-binding globulin gene

The plasma transport protein for glucocorticoids, corticosteroid-binding globulin (CBG), is produced by hepatocytes, and expression of the CBG gene (Cbg) in the liver is controlled by a variety of hormones, environmental stimuli, and developmental cues. The rat Cbg proximal promoter, including 145 base pairs (bp) from the transcription start site, contains two cis-regulatory elements (designated P1 and P2), and is as transcriptionally active as a much more extended region (approximately 1.2 kbp) of the promoter. We have now characterized the rat liver nuclear proteins that interact with P1 and P2. Several proteins interacted specifically with P1 during an electrophoretic mobility shift assay (EMSA), and based on ultraviolet (UV) cross-linking and Southwestern blot analyses, as well as an antibody-supershifting EMSA, one of these has been identified as hepatic nuclear factor (HNF)1 beta. The major band shift formed with P2 in an EMSA appears to comprise a protein that migrates as a doublet of 58 and 62 KDa on sodium dodecylsulfate-polyacrylamide gel ectrophoresis (SDS-PAGE) after UV cross-linking with an oligonucleotide containing P2, as well as during Southwestern blot analyses. Mutations of the CCAAT sequence within P2 also prevent binding to this protein, the physicochemical properties of which resemble the CCAAT-binding protein CP2. Functional analyses of this region of the rat Cbg proximal promoter fused to a luciferase reporter gene demonstrated that mutations within P1, which prevent its interaction with NHF1, do not influence adversely its transcriptional activity. Thus, although members of the HNF1 family of nuclear proteins play an essential role in the transcriptional activation of several other related genes (e.g., thyroxin-binding globulin and alphal-antitrypsin) in hepatocytes, HNF1 beta does not appear to be required for the basal activity of the rat Cbg promoter. In addition, deletion of P2 from the proximal promoter abolishes transcriptional activity and the CCAAT-binding protein that interacts with P2 probably represents an important determinant of Cbg expression under different physiological conditions.

High-affinity binding of corticosterone to mammalian neuronal membranes: possible role of corticosteroid binding globulin

The signal transduction mechanisms mediating rapid steroid actions are poorly understood. To characterize corticosteroid interaction with neuronal membranes in a species with rapid behavioral responses to corticosterone, we examined [3H]corticosterone binding to membranes prepared from prairie vole brains. At 22 degrees C, the rates of association and dissociation of [3H]corticosterone with well-washed synaptosomal membranes were very rapid. Specific binding was characterized by high affinity (Kd = 6.01 nM) and low density (Bmax = 63.1 fmol/mg protein). The binding sites were highly specific for naturally occurring glucocorticoids and the density of binding sites appeared to vary by neuroanatomical region. Unlike most G-protein-coupled receptors, the high-affinity binding of [3H]corticosterone to vole brain membranes was unaffected by the addition of Mg2+ or guanyl nucleotides. Surprisingly, saline perfusion of vole brains before tissue homogenization greatly reduced high-affinity binding. In addition, the affinity and specificity of corticosteroid binding sites were similar in vole neuronal membranes and vole plasma. These data suggest that corticosteroid binding globulins may facilitate [3H]corticosterone binding to neuronal membranes. However, the addition of blood to perfused brains before homogenization did not restore high-affinity binding, so the role of plasma binding globulins is unclear. Whether these binding phenomena represent a technical artifact or a regulatory mechanism for corticosteroid action has yet to be determined.