More than a decade has now passed since the first reports that the human placenta contains a peptide displaying corticotrophin-releasing hormone (CRH)-like activity (Shibasaki et al. 1982). This peptide has been shown to be biochemically similar to hypothalamic CRH (Sasaki et al. 1988) and both the CRH gene (Grino et al. 1987) and CRH mRNA (Frim et al. 1988) have been found in placental tissue. Despite some earlier confusion regarding the cellular localisation of CRH (Petraglia et al. 1987, Saijonmaa et al. 1988) the consensus now is that CRH is primarily produced within the syncytiotrophoblast layer of the chorionic villi (Riley et al. 1991, Cooper et al. 1994, Warren & Silverman 1995, Perkins & Linton 1995). The syncytiotrophoblast is bathed in maternal blood providing a direct route for placental CRH to access the maternal circulation in significant amounts. Many studies (Sasaki et al. 1984, Goland et al. 1986, Campbell et

Department of Pharmacology, Chelsea College, Manresa Road, London, SW3 6LX, and *Department of Physiology, St George's Hospital Medical School, Tooting, London, SW17 OQT

(Received 12 May 1977)

There is evidence accumulating to suggest that prostaglandins (PG) have a hypothalamic action in mediating the preovulatory surge of luteinizing hormone from the pituitary gland. Intraventricular injection of inhibitors of PG synthesis has been shown to block ovulation (Behrman, Orczyk & Greep, 1972; Orczyk & Behrman, 1972) and this effect can be overcome by subsequent administration of PG or luteinizing hormone releasing hormone. Similarly, PGE₂, administered intraventricularly, can overcome the ovulatory blockade induced by the α-adrenoceptor antagonist phentolamine (Linton, Perkins & Whitehead, 1977). In this study, we have investigated the action of the prostaglandin antagonist N-0164 (Eakins, Rajadhyaksha & Schroer, 1976) on ovulation in the rat.

Female Wistar rats weighing between 230 and 280 g were maintained under a controlled lighting schedule (12

ABSTRACT

We report the purification of human corticotrophin-releasing factor-binding protein (hCRF-BP) using repeated affinity chromatography (with the aid of synthetic CRF immobilized on Sepharose 4B solid phase) followed by gel filtration. Presence of the binding protein was tracked throughout the procedure by its ability to inhibit binding of ¹²⁵I-labelled hCRF to an hCRF antiserum; normal human plasma exhibits 80% inhibition in this system whereas sheep plasma, which does not contain an hCRF-BP, has no effect. Affinity cross-linking of ¹²⁵I-labelled hCRF to the purified hCRF-BP was performed using disuccinimidyl suberate (1 mmol/l). SDS electrophoresis of the purified CRF cross-linked binding protein followed by radioautography resulted in one major band of M _r 37 000 which corresponded to our original molecular weight estimate based on the elution position of the binding protein on Sephacryl S-200. A 10⁷-fold purification of the hCRF-BP resulted in a preparation estimated to be 95% pure, with an overall yield of 5%.

Journal of Endocrinology (1989) 122, 23–31

ABSTRACT

In a group of 12 adult Soay rams living outdoors near Edinburgh there was a conspicuous seasonal cycle in the peripheral plasma concentrations of β-endorphin, ACTH and cortisol. The concentration of all three hormones increased 5- to 20-fold from winter to summer; the seasonal maximum occurring from May to July for ACTH and cortisol and in August for β-endorphin. At the peak of the cycle the ratio of β-endorphin to N-acetyl-β-endorphin was 22:1. The regulation of the seasonal cycle was investigated in a series of five experiments involving treatments with arginine vasopressin (AVP), corticotrophin-releasing factor (CRF) and the synthetic glucocorticoid, dexamethasone.

Injection of AVP i.v. induced a dose-dependent increase in the plasma concentration of β-endorphin (AVP doses of 0, 0·07, 0·33 and 1·67 μg/kg). AVP (0·33 μg/kg) and CRF (1·67 μg/kg) given alone or in combination (equimolar doses), induced an increase in the plasma concentrations of β-endorphin and ACTH in spring, summer, autumn and winter, and produced a synergistic response when given together. The responses varied with season and were greatest in summer and autumn at the time of the seasonal increase in endogenous secretion. Dexamethasone injected i.v. at 68·04 μg/kg produced a decrease in the plasma concentrations of β-endorphin and ACTH, and the responses were also greatest in summer and autumn. A similar treatment with dexamethasone blocked the AVP-induced increase in the plasma levels of β-endorphin, indicating an action of dexamethasone on the pituitary gland. Administration of ACTH (0·33 μg/kg; i.v.) to rams pretreated with dexamethasone stimulated an increase in the plasma concentration of cortisol; this response varied with season, being greatest in spring at the time of the peak in the seasonal cycle in cortisol secretion. The administration of β-endorphin (0·33 pg/kg) failed to induce an increase in the plasma levels of cortisol at any season. Analysis of the hormone profiles in the control rams based on blood samples collected every 10 min for 8 h revealed pulsatile variations in the plasma concentration of ACTH; some of the spontaneous ACTH peaks were correlated with β-endorphin peaks.

From these results in the Soay ram, we conclude that β-endorphin and ACTH are co-secreted from the pituitary gland following stimulation by AVP and CRF, and that adrenal glucocorticoids stimulated by ACTH can act in a negative feedback role at the level of the pituitary gland to inhibit the release of both β-endorphin and ACTH. These acute studies indicate a parallel control of β-endorphin and ACTH at all stages of the seasonal cycle, and a seasonal change in the secretion of AVP and CRF from the hypothalamus may constitute the 'drive' to the seasonal cycle in both β-endorphin and ACTH. There was, however, a notable difference in the timing of the seasonal cycle in β-endorphin compared with that of ACTH, which indicates some form of differential control of these two pituitary hormones.

Journal of Endocrinology (1990) 124, 443–454

ABSTRACT

Using corticotrophin-releasing hormone-binding protein (CRHBP) purified from human plasma and a 25 amino acid peptide corresponding to the C-terminus of CRHBP we have been able to produce rabbit polyclonal antisera specific for CRHBP. This has allowed the development of a radioimmunoassay which is able to detect CRHBP specifically in human plasma regardless of the presence of endogenous CRH. We have used this assay to estimate the level of CRHBP in non-pregnant human plasma to be approximately 20 nmol/l with a range of 9·1–40·6 nmol/l. We have also examined sequential plasma samples taken from 84 normal pregnant women at fortnightly intervals from 16 weeks gestation through to term. Four women were also sampled during labour and the first week postpartum. The median plasma level of CRHBP at week 16 of normal pregnancy was 21·59 nmol/l, levels rose slightly during the early part of the third trimester (26·76 nmol/l at week 30, (P < 0·01) and fell markedly towards term (19·72 nmol/l, P < 0·01) with only 8·70 nmol/l at labour. CRHBP levels returned to normal non-pregnant levels within 48 h of parturition suggesting a role for the fetoplacental unit in CRHBP production. In eight pregnancies complicated by diabetes, CRHBP levels at each gestational age were similar to those recorded for normal pregnancy. However, in pregnancies complicated by pre-term labour (n = 9) and pre-eclampsia (n = 7), plasma CRHBP levels were significantly reduced (P < 0·01).

Journal of Endocrinology (1993) 138, 149–157

Abstract

During pregnancy maternal plasma corticotrophinreleasing hormone (CRH) levels rise 1000-fold whilst fetal plasma levels are often 100-fold higher than the concentrations seen in normal non-pregnant human plasma. Despite these high CRH levels neither the maternal nor fetal pituitary releases excessive amounts of ACTH. A specific CRH-binding protein (CRHBP) exists in the maternal circulation which is able to bind and inactivate the ACTH releasing activity of CRH. In this study we have used a specific CRHBP radioimmunoassay to determine the level of CRHBP in fetal and maternal plasma samples. Fetal samples were collected by cordocentesis between 20 and 33 weeks gestation and matched maternal samples were taken by venepuncture at the same time. In a second study, plasma samples were collected from 8 women at fortnightly intervals from week 20 to term, at labour and post-partum. A fetal sample, taken from the umbilical vein, was collected immediately post-delivery. The mean maternal CRHBP concentration for the samples collected between 20 and 33 weeks (ⁿ=23) was 8·12 nmol/l and the fetal level was 8·62 nmol/l. Data from the second study showed that at term the maternal CRHBP concentration decreased significantly (P<0·025) to 6·32 nmol/l. The fetal CRHBP level also decreased significantly (P<0·001) at term to a level of 5·84 nmol/l. The CRHBP in both fetal and maternal plasma was shown to be functional by ¹²⁵I-CRH binding and gel permeation chromatography. The capacity of maternal and fetal plasma to bind ¹²⁵I-CRH decreased at term in agreement with the quantitation of plasma CRHBP by radioimmunoassay.

Journal of Endocrinology (1995) 146, 395–401

Abstract

Direct immunoassay of plasma corticotrophin-releasing hormone (CRH) is potentially subject to interference from high levels of CRH-binding protein (CRH-BP) that exist in the human circulation. In this study, we tested the effect of CRH-free, native CRH-BP (6·4 nmol/l) purified from human plasma, CRH-BP diluent alone, normal human plasma (containing 5·8 nmol endogenous CRH-BP/1) and normal sheep plasma (containing no CRH-BP) on the binding of ¹²⁵I-labelled CRH tracer to five N-terminal and four C-terminal CRH antibodies. All anti-(1–20)CRH N-terminal antibody dilution curves displayed marked inhibition of binding in the presence of purified CRH-BP and human plasma in comparison with the curves with the control diluent or sheep plasma. Almost no inhibition of binding was obtained with any of the C-terminal antibodies (all directed against epitopes within the last six amino acids of CRH) and the four dilution curves were nearly superimposable. Liquid-phase CRH IRMAs were then developed with different combinations of two of each of the N- and C-terminal antibodies, using radiolabelled IgG prepared from purified C-terminal antisera as tracer and raw N-terminal antisera as the link antibodies to the separating system. The addition of dilutions of purified CRH-BP over the range 1·25–20 nmol/l to the IRMA standard curve in assay buffer resulted in a dose-dependent reduction in the signal; with 5 nmol CRH-BP/1, a level commonly found in human plasma, the reduction in binding was 67% and 81% in two different IRMAs at a CRH concentration of 631 pmol/l. With endogenous CRH-BP in human plasma, a dose-dependent inhibition of binding similarly resulted, with the plasma containing the most CRH-BP causing the greatest inhibition. Since plasma CRH-BP levels in humans vary widely, direct plasma IRMA using these type of antibodies will give inaccurate results and initial extraction of the CRH is necessary. Methanol extraction of synthetic or endogenous CRH is shown to be both highly efficient and unaffected by variable amounts of endogenous or exogenous CRH-BP; it is therefore suitable as the first step in plasma CRH measurement by IRMA.

Journal of Endocrinology (1995) 146, 45–53

ABSTRACT

A specific binding protein for human corticotrophin-releasing hormone (hCRH), which does not bind to the ovine hormone (oCRH), has recently been demonstrated in human plasma. No such binding protein has been found in sheep plasma. We have investigated the half-life of human and ovine CRH in man and in sheep. Peptides were measured directly in plasma with two-site immunoradiometric assays, as these assays are unaffected by the presence of inactivated peptide fragments. In man, the half-life of hCRH (30·5 ± 3·3 min; mean ± s.e.m.) was significantly (P < 0·001) less than that of oCRH (42·8 ± 6·4 min). In sheep, there was no significant difference between the half-life of hCRH (46·5 ± 7·2 min) and that of oCRH (39·8 ± 10·1 min); these half-lives were also significantly (P < 0·001) longer than that of hCRH in man. One possible explanation for the shorter half-life of hCRH in man is that the clearance of hCRH is enhanced by CRH-binding protein, although other binding proteins often have the opposite effect.

Peak ACTH and cortisol responses occurred earlier in sheep than in man, although no differences were found in the response times to oCRH or hCRH within either species. The responses were more sustained in sheep than in man, and the previously reported biphasic response was only seen in some of the sheep and not in man. Absolute responses to either peptide were greater in sheep than in man; however, in man an 8·1-fold rise in ACTH was measured in response to oCRH, while hCRH gave a significantly (P = 0·043) smaller 4·4-fold response. These experiments are discussed with particular reference to the effect of hCRH-binding protein and its possible role in potentiating the clearance of the peptide.

Journal of Endocrinology (1992) 133, 487–495