Drug-Drug Interaction Studies With Oral Contraceptives: Pharmacokinetic/Pharmacodynamic and Study Design Considerations.

Oral contraceptives (OCs) are the most widely used form of birth control among women of childbearing potential. Knowledge of potential drug-drug interactions (DDIs) with OCs becomes imperative to provide information on the medication to women of childbearing potential and enable their inclusion in clinical trials, especially if the new molecular entity is a teratogen. Although a number of DDI guidance documents are available, they do not provide recommendations for the design and conduct of OC DDI studies. The evaluation of DDI potential of a new molecular entity and OCs is particularly challenging because of the availability of a wide variety of combinations of hormonal contraceptives, different doses of the ethinyl estradiol, and different metabolic profiles of the progestin component. The aim of this review is to comprehensively discuss factors to be considered such as pharmacokinetics (PK), pharmacodynamics (PD), choice of OC, and study population for the conduct of in vivo OC DDI studies. In this context, metabolic pathways of OCs, the effect of enzyme inhibitors and inducers, the role of sex hormone-binding globulin in the PK of progestins, current evidence on OC DDIs, and the interpretation of PD end points are reviewed. With the emergence of new tools like physiologically based PK modeling, the decision to conduct an in vivo study can be made with much more confidence. This review provides a comprehensive overview of various factors that need to be considered in designing OC DDI studies and recommends PK-based DDI studies with PK end points as adequate measures to establish clinical drug interaction and measurement of PD end points when there is basis for PD interaction.

Internalization of sex hormone-binding globulin into neurons and brain cells in vitro and in vivo.

BACKGROUND: Sex hormone-binding globulin (SHBG) is a 94-kDa homodimer that binds steroids and is made in the hypothalamus. We have demonstrated that infusions of SHBG into the hypothalami of rats increase their female sexual receptivity except when SHBG is coupled to dihydrotestosterone (DHT) suggesting that SHBG has an active function in behavioral neuroendocrinology. METHODS: This study examines the possibility that SHBG is internalized by neuronal and/or non-neuronal brain cells as one possible mode of action using in vitro and in vivo techniques. RESULTS: First, analysis of the uptake of radiolabeled SHBG ((125)I-SHBG) found (125)I-SHBG uptake in HT22 hippocampal cells stably transfected with cDNA for ER beta (HT22-ER beta). The addition of DHT to (125)I-SHBG significantly inhibited (125)I-SHBG uptake in HT22-ER beta cells but not in HT22-ER alpha or HT22 wild-type cells. SHBG internalization was specific as it did not occur in either the human neuroblastoma cell line SK-N-SH or the glioma cell line C6. Second, SHBG was labeled with a fluor (Alexa-555), and infused into the lateral cerebroventricles of ovariectomized rats. Optimal SHBG uptake was seen 10 min after these infusions. SHBG uptake was seen in specific parts of the choroid plexus and periventricular cells as well as into cells in the paraventricular nucleus, the medial forebrain bundle, and the habenula. CONCLUSIONS: These studies suggest that SHBG is internalized by brain cells, which may be affected by the presence of ER beta. The gonadal steroids have numerous effects in brain and the discovery that the steroid-binding protein SHBG is taken up into neurons and brain cells may demand a change in thinking about how steroids are delivered to brain cells to affect neurophysiology.

Influence of glycosylation on the clearance of recombinant human sex hormone-binding globulin from rabbit blood.

Human sex hormone-binding globulin (hSHBG) is a plasma glycoprotein that binds sex steroids with high affinity. Variations in hSHBG glycosylation contribute to its electrophoretic microheterogeneity, but the functional significance of different SHBG glycoforms is unknown. Carbohydrates may influence the biological activities and half-lives of glycoproteins and we have examined how oligosaccharides at specific sites influence the plasma clearance of hSHBG in vivo. To accomplish this, fully-glycosylated hSHBG, or hSHBG mutants lacking specific oligosaccharides chains, were expressed in Chinese hamster ovary (CHO) cells and purified by immunoaffinity chromatography. The purified recombinant proteins were then biotinylated to study their plasma half-lives after intravenous injection into rabbits. When compared to hSHBG isolated from serum, recombinant hSHBG migrates with a slightly larger average molecular size during denaturing polyacrylamide gel electrophoresis. This is due to a greater proportion (33-39% vs. 3%) of more highly branched N-linked oligosaccharides on the recombinant proteins. When injected into rabbits, the disappearance of recombinant hSHBG showed two exponential components, as previously shown for natural hSHBG in the same animal model. The mean +/- S.E.M. plasma half-lives of recombinant hSHBG (t 1/2alpha 0.11+/-0.03 h and t 1/2beta 18.94+/-1.65 h) are shorter than previously measured for natural hSHBG (t 1/2alpha 3.43+/-0.72 h and t 1/2beta 38.18+/-7.22 h) and this is likely due to differences in the composition of their N-linked oligosaccharides. An O-linked chain at Thr7 does not influence the plasma clearance of hSHBG in the presence or absence of N-linked carbohydrates at Asn351 and Asn367. However, a 1.5-1.6 fold (p<0.03) increase in plasma half-life of variants lacking both N-glycosylation sites was observed and this is probably due to the fact these variants are not recognized by the asialoglycoprotein receptor-mediated clearance system. Removal of either N-glycosylation consensus site also increased (p<0.0001) the plasma half-life of hSHBG by 2.3 2.4 fold. Thus, the metabolic clearance of hSHBG appears to be determined by the number of N-linked oligosaccharides rather than their location.

Cellular bioavailability of natural hormones and environmental contaminants as a function of serum and cytosolic binding factors.

Environmental contaminants have been reported to function as hormone mimics in various wildlife species. To investigate a potential mechanism for the interaction of contaminants with the endocrine system, we evaluated the cellular bioavailability of numerous chemicals. Hormone binding proteins from oviductal cytosol of the American alligator (Alligator mississippiensis) and yellow-bellied turtle (Trachemys scripta) were used in competitive binding assays with [3H] 17 beta-estradiol. Most of the environmental contaminants, and the potent, synthetic estrogen diethylstilbestrol (DES), did not interact with the cytosolic binding proteins. Among the compounds tested, o,p'-DDT and toxaphene exhibited the greatest affinity for the binding proteins. The functional consequence of the apparent lack of interaction of most contaminants with binding proteins was studied in a strain of yeast containing the human estrogen receptor (YES assay). The activation of YES with estradiol was reduced 30% in the presence of a physiological concentration (0.01 mg/mL) of human sex hormone binding globulin (SHBG), a hormone binding protein found in the blood. In contrast, the activity of DES was not inhibited by 0.01 mg/mL SHBG. Interestingly, ethinyl estradiol, a major component of contraceptives, did not appear to appreciably interact with SHBG in the YES system. Together, these data suggest that cytosolic and circulating binding proteins bind many environmental contaminants with much less affinity than native steroids. Therefore, such contaminants may be more hormonally active than previously hypothesized.

Human variant sex hormone-binding globulin (SHBG) with an additional carbohydrate chain has a reduced clearance rate in rabbit.

Sex hormone-binding globulin (SHBG) is the specific plasma transport protein for sex steroid hormones in humans. Considerable variation in SHBG plasma concentration exists between individuals, irrespective of gender, body weight, or thyroid status. In the present work, the influence of carbohydrate chains on the half-life of human SHBG (hSHBG) was investigated using a rabbit model. A variant hSHBG, with a point mutation in exon 8 (GAC --> AAC) encoding an amino acid substitution (Asp327Asn), which introduces an additional consensus site for N-glycosylation, has recently been identified. This mutation suppresses a recognition site for the restriction enzyme Bbs-I, allowing the development of a simple restriction-fragments length polymorphism (RFLP) screening procedure. In a population of patients (272 female and 49 male) consulting in our Endocrinology Clinic, 48 patients (41 female and 7 male) were heterozygous for the variant hSHBG allele and 3 (2 female and 1 male) were homozygous. In this population, the total variant allele frequency was 0.083. The hSHBG genotype, as determined by RFLP, corresponded in all cases to the phenotype as determined by the migration profile of hSHBG by Western blot analysis. The influence of such an additional glycosylation site on the biological half-life of variant hSHBG was investigated. SHBG from serum of patients carrying one of the three hSHBG genotypes was purified and labeled with biotin, then injected into rabbits, as we have recently described for rabbit SHBG. Biotinylated hSHBG was captured from rabbit serum samples on tubes coated with an anti-hSHBG antibody and detected by luminometry with the streptavidine-alkaline phosphatase-dioxetane (AMPPD) system. The results showed that the half-life value was significantly higher (P < 0.05) for SHBG purified from homozygous variant serum (t1/2 beta = 51.43 +/- 1.15 and 63.63 +/- 3.92 h, for male and a female subjects SHBG respectively) than for SHBG purified from heterozygous variant serum (t1/2 beta = 40.19 +/- 0.12 h) or wild-type (t1/2 beta = 38.18 +/- 7.22 h). This study demonstrated that an additional carbohydrate chain on hSHBG decreases the clearance rate of this protein. The low frequency of this variant allele means that further study will be required to determine whether it is associated with higher serum SHBG concentration.

Tissue uptake of human sex hormone-binding globulin and its influence on ligand kinetics in the adult female rat.

The distribution of human sex hormone-binding globulin (hSHBG) and its influence upon the kinetics of its ligands were assessed in the adult female rat, which lacks a comparable protein in serum. Purified hSHBG was administered i.v. to adult female rats as a single bolus. The plasma disappearance rate of immunoreactive hSHBG had one component with a half-life of 15 h. The estradiol (E2) binding activity of serum attributable to hSHBG was elevated 2-fold; during a continuous infusion of E2, hSHBG increased E2 serum levels above those of control infused animals. Treatment with hSHBG did not modify the plasma clearance of endogenous E2, but accelerated the disappearance rate of 5 alpha-dihydrotestosterone (DHT). In animals injected with a tracer dose of radioactive steroids, pretreatment with hSHBG increased uterine and oviductal accumulation of E2- but not DHT-associated radioactivity. This effect was specific to some E2 target tissues since hSHBG did not alter the concentration of E2- or DHT-associated radioactivity in the hypophysis, liver, diaphragm, or brain. Treatment with anti-E2 antibodies, which elevated E2 binding activity in serum, decreased the accumulation of E2-associated radioactivity in uterus and oviduct. Immunofluorescent localization of hSHBG revealed intense labeling of the uterine and oviductal epithelium. We conclude that this foreign hormone-binding globulin introduced in serum at concentrations that have minimal circulating reservoir effect on E2 can reach intracellular domains and affect the concentration of this ligand in target tissues.

The metabolism of human sex hormone-binding globulin in the rhesus monkey.

The metabolism of human sex hormone-binding globulin (hSHBG) was studied in eight female rhesus monkeys (Macaca mulatta) after the pulse injection of [125I]-hSHBG. hSHBG was iodinated with 125I using a chloramine T technique, and the [125I]-hSHBG was separated from other constituents by molecular sieve chromatography with a Sephadex G-25 column. The [125I]-hSHBG was administered intravenously as a pulse in 2 ml of phosphate buffer, pH 7.4, to each of eight rhesus monkeys. Blood samples (2.0 ml) were obtained at 2, 4, 6, 8, 24, 30, 45, and 54 hr after the injection. The glycoproteins were precipitated with concanavalin A-Sepharose, and the radioactivity was measured. The concentration of radioactivity as fraction of dose/ml of serum was plotted on a semilog scale against time. The disappearance of radioactivity could be expressed best as the sum of two exponentials, with a mean +/- SE t1/2 of 2.5 +/- 0.4 and 33.1 +/- 3.7 hr, respectively. The initial volume of distribution was 461 +/- 78 ml and the metabolic clearance rate was 559 +/- 66 ml/day. The very low clearance rate and prolonged t1/2 are compatible with a relative stability in the circulating mass of SHBG. Rapid changes in concentration of SHBG could be due to changes in serum volume, reversible changes in tissue distribution of SHBG, or the secretion of variable forms of desialylated SHBG.

Pharmacokinetics of oral 17 beta-estradiol.

The pharmacokinetics of oral 17 beta-estradiol (E2) were evaluated: only a limited amount of information is available on the subject. Because of the first passage hepatic effect, the blood levels of estrone (E1) are greater than those of E2; similar profiles exist for oral E1 sulfate, micronized E2 and E2 valerate. However, the short-term effects of oral E2 versus E1 on hepatic parameters may vary somewhat. Peak levels of E1 and E2 are achieved four hours after the administration of 1 mg of E2 and average 200 and 40-50 pg/mL, respectively. A dose-response relationship exists for serum levels achieved after oral E2 administration. Twelve-hour values are representative of the 24-hour profile. With prolonged use, the 24-hour levels may be equally representative and serum E2 levels increase, suggesting some cumulative effects. Smoking enhances the hepatic metabolism of oral estrogen and results primarily in a lower unbound E2 level.

Internalization of human sex steroid-binding protein in the monkey epididymis.

Human sex steroid-binding protein (hSBP) has been purified from late-pregnancy serum and labelled either by iodination (125I) or by photoaffinity with [3H]delta 6-testosterone. Using a micromanipulator, each labelled protein was separately injected into the lumen of epididymal tubules isolated from the head epididymis of the cynomologus monkey (Macaca fascicularis). Tubules were sampled from 3 to 90 min after the injection and processed for electron microscope autoradiography. Localization of the label occurred over the epididymal epithelium whichever tracer was used. The labelling was not randomly distributed over the different cell types constituting the epithelium, since only the 'principal cells' exhibited a silver grain count significantly greater than the background count. In these cells, labelled protein was found over endocytic organelles (coated structures, endosomes, multivesicular bodies and the trans Golgi network) and nuclei (including the nuclear envelope). Quantitative analysis demonstrated the same pattern of cellular and subcellular distribution for each tracer. Pretreatment with excess unlabelled protein significantly reduced the uptake of radioactivity by the principal cells, demonstrating the specificity of this phenomenon. This is the first study to show direct histological evidence for the internalization of hSBP in the primate epididymis, consistent with earlier immunohistochemical or biochemical localization of this protein. It is concluded that head epididymal cells are able to take up labelled hSBP across their apical membrane. The mechanism of internalization seems to involve endocytosis by the principal cells and leads to labelling of the nuclear compartment. This is strikingly similar to the pattern of uptake of rat androgen-binding protein (rABP) by rat epididymal cells previously demonstrated by our group. To what extent the chemical and structural homology between hSBP and rABP can be held responsible for the common cytophysiological behaviour of these sex steroid-binding proteins remains to be determined.

Sex hormone binding globulin in clinical perspective.

The plasma steroid-binding proteins play a fundamental role in the modification and delivery of the steroid "signal" to its target tissue. While their precise function requires further evaluation, the measurement of these proteins, particularly of SHBG, provides clinically useful information. This article reviews the current theories of sex steroid transport and uptake, and documents the changes in SHBG status relevant to health, disease and drug therapy.