Progesterone stimulates Krox-20 gene expression in Schwann cells.

The gene of the zinc finger transcription factor Krox-20 (Egr-2) is expressed in Schwann cells and plays an important role in myelination of peripheral nerves. We have shown that progesterone promotes myelination in the regenerating sciatic nerve and in cocultures of Schwann cells and sensory neurones. To determine whether progesterone regulates Krox-20 expression, we measured its effects on Krox-20 mRNA levels in the MSC80 mouse Schwann cell line by semi-quantitative RT-PCR. Although low levels of Krox-20 mRNA are detectable in MSC80 cells cultured in defined medium, treatment with 10(-6) M progesterone induces a rapid (15 min) and transient increase in the levels of Krox-20 mRNA. Lower doses of progesterone (10(-9), 10(-8) and 10(-7) M) are also effective in increasing Krox-20 mRNA. Other steroids including testosterone, dexamethasone, and estradiol are ineffective when added to the culture medium at 10(-6) M for 1 h. The induction of Krox-20 mRNA was also observed with the selective progesterone agonist Organon 2058 and was abolished by treating the MSC80 Schwann cells with the progesterone antagonist RU486, indicating that progesterone induces Krox-20 mRNA expression by binding to its intracellular receptor. The induction of Krox-20 by progesterone was also demonstrated in primary cultures of Schwann cells isolated from neonatal rat sciatic nerves, at the mRNA level by RT-PCR and at the protein level by immunohistochemistry. As Krox-20 is a necessary step for the initiation of myelin formation in peripheral nerves, its stimulation by progesterone suggests an important signalling function for this steroid in myelination.

Sex differences in the activation of tamoxifen to DNA binding species in rat liver in vivo and in rat hepatocytes in vitro: role of sulfotransferase induction.

Previous work has indicated that metabolic activation of tamoxifen in rat liver cells involves cytochrome P450-mediated alpha-hydroxylation, followed by sulfate ester formation, mediated by hydroxysteroid sulfotransferase a (rHSTa), a member of the SULT2A subfamily, which efficiently metabolizes dehydroepiandrosterone. Because it is known that the expression of rHSTa and other SULT2A forms is substantially higher in female rats than in males, it might be predicted that tamoxifen would be a more potent liver carcinogen in females than in males. Yet tamoxifen has been shown to be equipotent in both sexes. To investigate this paradox, primary cultures of hepatocytes were prepared from Fischer F-344 rats and treated with tamoxifen (10 microM) or alpha-hydroxytamoxifen (1 microM). Rats were also treated with tamoxifen daily by gavage (0.12 mmol/kg/day) for up to 14 days. DNA was isolated from hepatocytes and liver and analyzed by 32P-postlabeling. Liver cytosol fractions were prepared and analyzed for dehydroepiandrosterone sulfotransferase activity and SULT2A protein levels. In tamoxifen-treated hepatocytes and after a single dose of tamoxifen in vivo, DNA adduct formation in male cells was significantly lower than in female cells, 11- and 6-fold, respectively. However, with increasing daily doses of rats with tamoxifen, the adduct level in males increased to a level 89% of that in females by 14 days. Dehydroepiandrosterone sulfotransferase activity in male rat liver cytosols was only 17% of the activity of female cytosols after one dose of tamoxifen but 64% after 14 days of exposure to the compound. This increase in activity correlated with increases in the levels of SULT2A protein, detected by Western blotting. Western blotting did not allow the unambiguous identification of the induced SULT2A form(s). However, by using a specific reverse transcriptase/PCR technique, it was found that it was primarily rHSTa that was induced. Thus, after prolonged exposure to tamoxifen, DNA adduct formation and rHSTa expression in males are significantly closer to the levels in females than they are after initial exposure. These changes explain the similar susceptibility of male and female rats to tamoxifen carcinogenesis.

Hydroxysteroid sulfotransferase activity in the rat brain and liver as a function of age and sex.

The high concentrations of dehydroepiandrosterone sulfate and pregnenolone sulfate in the mammalian brain, despite the blood-brain barrier's impermeability to these compounds, and the apparent independence of these concentrations from those in plasma prompted us to investigate whether enzymatic sulfation of dehydroepiandrosterone was detectable in the rat brain. Low hydroxysteroid sulfotransferase activities were detectable in in vitro incubations of homogenates from all rat brain regions except the cerebellum, being highest in the hypothalamus and pons. This activity was not ascribable to enzyme in brain capillary blood. The activity was mainly cytosolic, although there was also significant activity in the partially purified nuclear fraction. The enzyme had different properties from those of hepatic isozymes, with a pH optimum of 6.5 and a high Km of approximately 2 mM for dehydroepiandrosterone. The enzyme was also active with pregnenolone as substrate. Activities in the brain were approximately 300-fold lower than in the liver but, as in the liver, these were higher in females than in males. The variations in brain activity as a function of age did not parallel those in the liver. Relatively high activities were found in the fetal brain and declined at birth, while activities were insignificant in the fetal liver and rose following birth. There was a major peak in activity in pubertal female brains, but this peak was less important, and later, in males. No evidence was found to indicate that the low brain enzyme activities and high Km were attributable either to the presence of an inhibitor or to the steroid sulfation actually being a secondary activity of another brain sulfotransferase. We discuss whether the sulfotransferase activities found are adequate to synthesize the dehydroepiandrosterone and pregnenolone sulfate found in brain.

Investigation of a 17 beta-estradiol-monoclonal antiestradiol antibody binding mechanism using dilute solutions of organic solvents.

Quantitative understanding of steroid hormone transport and receptor-mediated action requires knowledge of the bonding forces involved in each steroid-protein complex and the effects of a biological environment on these forces. An approach to these problems using dilute solutions of water-miscible organic solvents, with a range of polarity, dielectric and hydrogen bonding properties, was tested on an estradiol-antiestradiol antibody binding system on the basis that comparing the effects of the solvents would both permit the importance of hydrophobic and hydrogen bonding to be differentiated and give information on the effects of the environment on the reaction. The results were compared with thermodynamic measurements. All the solvents reduced the Gibbs free energy of binding as a function of their concentration in the medium. The decreases were virtually a monotonic function of their dielectric constant, indicating reduced hydrogen bonding. Analysis of the decreases in terms of the solvents' hydrogen bonding and polarity properties supported this. Thermodynamic measurement showed the binding reaction was enthalpy-driven with, overall, a slightly unfavorable entropy contribution. This again showed the hydrophobic effect was not the main bonding force. The most deleterious solvent, iso-propanol, not only decreased the enthalpic contribution to binding but rendered the entropic contribution more favorable. This approach still does not allow the relative importance of hydrogen bonding and van der Waals contacts in the actual binding to be differentiated but it does give indications on how a biological environment may affect a steroid-protein binding reaction in vivo.

Estrogen-binding properties of rat serum alpha 1-fetoprotein and its isoforms. Investigation of the apparent non-integrality of sites on the unfractionated protein.

Rat fetal serum alpha 1-fetoprotein (AFP), a heterogeneous glycoprotein, binds estrogens with high affinity but at a fractional number of sites even after treatment with charcoal (n = 0.6), which may mean 60% of the protein has 1 site and the remainder none. To investigate the origin of this fractional number of sites the "native" protein (purified by negative affinity chromatography) was further purified (step 1) and fractionated (step 2) into its two main charge variants (electrophoretically "slow" and "fast") by a two-step fast-protein liquid chromatography method. The binding parameters for estrone and estradiol-17 beta of the "native" and "repurified" proteins and of each charge variant were determined by equilibrium microdialysis. The molar extinction coefficient at 278 nm of each sample was also determined. (1) The "repurified" AFP and each charge variant had a number of binding sites for estrogens close to unity. This increase in the number of sites could neither be explained by the loss of a non-binding isoform (corresponding to 40% of the protein) during chromatography, nor by the existence of complex negative modulatory interactions between isoforms. (2) The affinities for estrogens of the "repurified" protein and the two charge variants were slightly decreased compared to that of "native" AFP, except that the "fast" form had the "native" protein's high affinity for estrone--but not for estradiol-17 beta. (3) The molar extinction coefficients at 278 nm of the "repurified" AFP and the isoforms were much lower than that of the "native" protein. These results suggest that the presence of (an) inhibitor(s) of estrogen binding on the "native" protein which is/are removed by the ion-exchange fast protein liquid chromatography (FPLC) column. A ligand absorbing at 278 nm, which may or may not be the inhibitor, is also removed. The isoform heterogeneity with respect to estrone binding is discussed.

Comparison of graphical procedures for estimating the intrinsic molar fluorescence of protein-bound drugs for drug-binding studies. A reevaluation of existing plots and introduction of two inverse hyperbolic plots.

A theoretical evaluation of existing and new graphical procedures for estimating the intrinsic molar fluorescence (psi B) of a protein-bound drug has been undertaken. The results do not concord with a recent proposition that psi B should be obtained by direct reading from a graph of emitted fluorescence intensity (I) against the logarithm of the binding protein concentration ([P]) rather than by extrapolation of a double reciprocal plot. The calculated errors in estimates of psi B obtained by direct reading or by extrapolation of standard plots and of three new inverse hyperbolic plots showed that, independently of binding affinity: 1. Direct reading from the logarithmic plot gave the least accurate estimates. 2. The single reciprocal plot gave more accurate estimates than the double reciprocal plot providing the (constant) drug concentration was similar to or greater than its dissociation constant in the binding system. At lower drug concentrations the double reciprocal plot gave more accurate estimates. 3. Extrapolation of an inverse hyperbolic sine plot (sinh-1 (1/I) against 1/[P]) did not give more accurate estimates than the standard reciprocal plots. 4. If the drug concentration was close to its dissociation constant the most accurate estimates were obtained with an inverse hyperbolic cosine plot of cosh-1 (I + 1) against 1/[P]. For a low affinity binding system in which non-specific binding is significant an inverse hyperbolic sine plot of 1/sinh-1I against 1/[P] gave the most accurate estimates at low drug concentrations. An experimental and theoretical procedure for optimizing the estimation of psi B is proposed on this basis.

A competitive microtitre plate enzyme immunoassay for plasma testosterone using polyclonal anti-testosterone immunoglobulins.

An enzyme immunoassay for plasma testosterone was developed based on competition between an immobilised testosterone-casein conjugate and the analyte for polyclonal anti-testosterone immunoglobulins, followed by the use of enzyme-labelled second antibodies to determine the degree of competition. The quantity of immobilised testosterone-casein conjugate was optimised so that the lower affinity anti-testosterone antibody populations present did not affect the assay. The assay standard curve covered a range of 11-300 fmol/well. Testosterone levels in small amounts of male and female plasma could be assayed with good reproducibility and correlated well with results obtained by radioimmunoassay. By comparison with an analogous assay using monoclonal antibodies it appears that, given that the assay sensitivity is the most important criterion for choice, the use a polyclonal antiserum with this type of reactive antibody selection is preferable to the use of monoclonal antibodies.

A competitive microtitre plate enzyme immunoassay for plasma aldosterone using a monoclonal antibody.

A competitive microtitre plate enzyme immunoassay for plasma aldosterone was developed using an immobilised aldosterone-bovine serum albumin conjugate and a monoclonal anti-aldosterone preparation, followed by the use of enzyme-labelled second antibodies to determine the degree of competition. The quantity of immobilised aldosterone-protein conjugate was adjusted to give optimum assay sensitivity with respect to the antibodies used. The lower limit of detection of aldosterone (55 fmol) was much less than that of an ELISA for aldosterone, using identical reagents but with an excess of immobilised aldosterone-protein conjugate, and up to 1400 fmol could be determined. Aldosterone levels in small amounts of male and female plasma could be assayed with good reproducibility.

The interpretation of ligand displacement experiments: calculations for multisite acceptors.

A method is described for calculating the degree of competition for binding between two ligands which are bound at any number of site classes on a binding protein from a generalization of the equilibrium competitive binding equations, the protein's binding parameters for each of the ligands, and the total protein and ligand concentrations. Theoretical displacement curves thus obtained for each of the possible competitive binding models with a multisite protein can then be compared with experimentally determined ligand displacements in order to find which model is most realistic or if measured displacements are due rather to negative cooperativity effects. The binding parameters used for the calculations have a statistical error attached to them, since they have been obtained experimentally, so here we also propose a method for calculating the standard deviations of the theoretical displacement curves deriving from these errors. This permits the use of statistical hypothesis testing in the comparison of theoretical and experimental results. An example is shown in which this method permits the verification that two drugs (phenylbutazone and azapropazone) are both bound by the same high- and low-affinity sites of a protein (alpha-fetoprotein).

Enzyme-linked immunosorbent assay (ELISA) for steroid hormones with polyclonal and monoclonal antibodies: an assay for urinary aldosterone.

The development of non-competitive microtitre plate enzyme-linked immunosorbent assays for steroids was investigated using immobilised steroid as immunosorbent and enzyme-labelled second antibodies. An assay for testosterone with polyclonal anti-testosterone immunoglobulins was optimised with respect to a number of parameters but remained unsatisfactory for clinical assays. The results were applied to developing an aldosterone assay using monoclonal anti-aldosterone immunoglobulins. The latter method was used for the determination of urinary aldosterone and the results are compared with those obtained by a classical radioimmunoassay. Problems concerned with the use of sulphur-containing proteins and with the presence of low affinity antibodies in polyclonal preparations are discussed.

The use of disulfonatonaphthalimide fluorescent dyes for the fluorescence polarization immunoassay of steroids.

A series of fluorescent disulfonatonaphthalimide derivatives of testosterone and estriol have been synthesized and their fluorescent properties investigated. The fluorescence lifetimes of these derivatives were higher than that of the unreacted fluorescent dye while the quantum yields were of the same order. The compounds were therefore compared in terms of their utilizability in steroid fluorescence polarization immunoassays. The assay sensitivity and precision with each compound is discussed in terms of the position, type, and length of the chemical "bridge" linking the steroid to the fluorescent dye. It is proposed that these fluorescent labels are highly appropriate to this type of immunoassay.

Drug-binding properties of rat alpha-foetoprotein. Specificities of the phenylbutazone-binding and warfarin-binding sites.

Rat alpha-foetoprotein (alpha-FP) strongly binds the drugs warfarin and phenylbutazone, as does albumin; however, the binding sites for the two drugs seemed to be different. This possibility and the specificity of this/these drug-binding site(s) of rat alpha-FP were investigated by competitive protein-binding experiments with a variety of drugs, representing different pharmacological groups, and biomolecules that are strongly bound by the foetal protein and that are suspected to play a specific role during foetal development. The binding mechanisms were further investigated by using comparisons between computer-derived theoretical displacement curves and experimental points in order to distinguish different possible binding models. The results indicate: that warfarin and phenylbutazone are bound at two distinct sites on rat alpha-FP and that a negative modulatory effect is exerted between the two sites; that the degree of specificity of these two drug-binding sites is different, since the warfarin-binding site appears to be specific for the binding of coumarinic and anthranilic drugs whereas that for phenylbutazone is able to bind substances of very varied chemical structure and is more hydrophobic; that the phenylbutazone-binding site is the site that binds oestrogens that thyroid hormones and, probably, fatty acids and bilirubin are bound at (an)other site(s) but exert negative modulatory effects on phenylbutazone binding. The nature of the different binding areas of rat alpha-FP is compared with that of those already proposed for albumin. The potential risks of toxicity of such interactions between drugs and/or biomolecules on foetal development are also discussed.

The preparation of three fluorescence-labeled derivatives of testosterone.

Three fluorescence-labelled derivatives of testosterone were prepared consisting of the steroid separated from the fluorochrome by a hydrocarbon "bridge". "Bridges" of different lengths (C2 to C7) were used as the length required to avoid steric hindrance effects by the fluorochrome in studies on steroid-protein binding was unknown. The three derivatives prepared were: 17 beta-hydroxy--4-androsten-3-one 3-(O-(N-(2'-mercapto)ethyl)carbamoylmethyl)oxime, 17 beta-hydroxy-4-androsten-3-one 3-(O-(N-(3'-amino)propyl)carbamoylmethyl)oxime and 17 beta-hydroxy-4-androsten-3-one 3-(O(N-(7'-amino)heptyl)carbamoylmethyl)oxime. These were then coupled with either a dansyl or a fluorescein molecule. Overall yields were sufficient and the products immunoreactive with anti-testosterone antiserum.

An enzyme-linked immunoassay of testosterone.

An enzyme-linked immunoassay of testosterone is described. The conjugation of testosterone with high specific activity horseradish per-oxidase and a highly sensitive assay for this enzyme having previously been studied, here we describe the immobilization of the anti-testosterone antibody and the development of assay conditions permitting the determination of 50 pg to 1.5 ng testosterone in one working day. In this work attention has been paid to keeping the assay method as simple as possible. The method is discussed in terms of other enzyme-linked immunoassays for steroids.