The role of melatonin in animals which do not show marked seasonal changes in reproduction is disputed, in part because of the wide variation in reported concentrations. One reason for this may be the difficulties associated with the measurements of low molar concentrations of melatonin and the presence of a wide variety of potentially cross-reacting substances. The availability of a high affinity antiserum has allowed an assay, with low cross-reactivity and good sensitivity, to be established for the direct measurement of melatonin in a wide range of biological fluids, in particular serum, plasma and follicular fluid from man and rat. The high affinity of the antiserum enabled a tritium label of high specific activity to be used, removing the problems associated with the iodination of a small molecular weight compound. Melatonin concentrations in the assay were evaluated by four different methods: UV absorbance, gas chromatography, comparison of the immunoreactive concentrations of the label with the expected concentration by dilution and by comparison with a previously established assay which uses the same antiserum. Melatonin was measured in serum from twelve healthy women over two 24-h periods; eight women with normal menstrual cycles and four taking the contraceptive pill. Concentrations were found to range from 19.8 to 215 pmol/l during the day in both groups. In women with normal menstrual cycles peak concentrations of 513.2 +/- 54.1 (S.E.M.) pmol/l were recorded at 04.00 h, whereas higher concentrations were found in women taking the pill, reaching a peak of 849.12 +/- 21.8 (S.E.M.) pmol/l at 04.00 h.(ABSTRACT TRUNCATED AT 250 WORDS)

Covalently linked insulin dimers have been prepared by cross-linking two insulin monomers with a flexible suberoyl chain at either the B1 phenylalanine or the B29 lysine residue. Binding potencies of dimers determined by inhibition of binding of 125I-insulin to isolated rat liver plasma membranes or adipocytes were 2.5-7-fold greater than their abilities to stimulate lipogenesis in adipocytes. Rates of liver plasma-membrane-associated degradation of labelled insulin and dimers, measured by gel filtration, were similar at 37 degrees C. Binding and lipogenesis potencies of dimers prepared by substitution of each monomeric half of an asymmetrical dimer with desoctapeptide insulin, an almost inactive derivative, implicated the B1-cross-linked monomeric half as predominantly interacting with the insulin receptor. These results suggest that (1) dimers bind univalently to a bivalent insulin-receptor complex, in which the two individual binding subunits are arranged with anti-parallel symmetry and (2) the mechanism by which insulin binds and initiates its biological responses requires a conformational change within the insulin-receptor complex and/or in the insulin molecule for full biological expression.

The crescendo of events leading first to ovulation and subsequently to birth is orchestrated by a broad repertoire of hormones. The major hormones of the ovulatory cycle are representatives of four hormone classes: neurotransmitters, releasing factors, trophic hormones acting on target tissues, and steroid-like molecules released by the target tissues. The punctuate and staccato rhythm of the neurotransmitters and releasing hormones relentlessly drive the swelling and protracted wave of activity by the luteotrophic and steroid hormones. Carbohydrates alone are notably absent as hormones and the predominant role for glycosylation appears to be the conferment of increased solubility to endocrine molecules, either during their manufacture or by modulating circulatory half-life. Rarely considered examples of the importance of glycosylation in reproductive hormones include adenosine, important for spermatozoan activity, and the hormone-binding globulins, which ensure the aqueous transport of hydrophobic steroids. The archetype glycoprotein hormones, especially human chorionic gonadotrophin (HCG), are discussed more extensively, as the structural and functional roles of carbohydrate in these hormones have been studied exhaustively. Conversely, the direct involvement of HCG and the importance of its carbohydrate for autonomous growth, in both placental invasion and tumorigenesis, has received little attention in the literature.

The thyrotropin receptor is of fundamental importance to normal thyroid function and is considered to be the predominant antigen affected by the autoantibodies of Graves' autoimmune hyperthyroidism. The identification of the epitopes on the receptor to which the autoantibodies bind or the mechanism by which the autoantibodies arise remain to be established. In this report we have analysed in detail thein vivo transcription of the human TSH receptor gene (hTSH-R), demonstrating the presence of numerous novel TSH receptor transcripts. Northern blot analysis of mRNA from human thyroid tissue using a radiolabelled cDNA probe specific for the extracellular domain of the hTSH-R revealed the presence of small polyadenylated mRNAs, in addition to the full-length hTSH-R mRNA. A PCR strategy devised to clone transcripts with 3' polyadenylation and 5' hTSH-R specific sequences was used to clone five different hTSH-R transcripts (hTSH-R. ST1 to ST5; 250bp-1.7 kb) from human thyroid tissue. Sequence analysis demonstrated that the small transcripts arose by alternative splicing of the hTSH-R mRNA. The transcripts were associated with polysomes and were demonstrated in human thyroid tissue from patients suffering from Graves' disease, sporadic goiter as well as in healthy lobes of thyroid tissue.In situ hybridization demonstrated that two of the alternative transcripts adopted a tissue distribution pattern identical to that of the full-length hTSH-R transcript. The two major truncated transcripts ST4 and ST5 contained unique sequences at the 3' end of the mRNAs and thus potentially represent the molecular origin of soluble TSH receptor variants which have been postulated on numerous occasions.

A specific homologous radioligand receptor assay for thyroid-stimulating hormone (TSH) using bovine thyroid membranes was adapted for use with human thyroid. Specific 125I-labelled TSH binding was detected in the 3000 g membrane pellet from bovine thyroid but predominantly in the 3000 g supernatant of the human thyroid homogenate. Both assays required incubation in the presence of 10% serum, whilst the assay using human thyroid could only be precipitated using polyethylene glycol (PEG). The serum requirement transcended a possible role as carrier protein and unmasked specific TSH binding. Molecular sieving determined that the active fraction of the serum had an apparent size of 30,000-100,000. The requirement for PEG-assisted precipitation of the TSH receptor assay was a consequence of the TSH-binding entity from Graves' thyroid behaving like a soluble 'receptor': it did not sediment with the membranes, passed a 0.2 microns filter and, upon molecular sieving, had an apparent size of 300,000-1,000,000. A full-length TSH receptor cDNA was cloned from a human Graves' thyroid library and stably transfected cell lines expressing the TSH-receptor protein were constructed using human HeLa and murine 3T3 cells. Specific TSH binding was unmasked by serum in the human cell lines, as observed for the human thyroid TSH receptor, whereas serum hindered TSH binding in the murine cell lines. A soluble form of the receptor was not released from the cells and was not produced in conditions which demonstrated a soluble receptor-like binding component in human thyroid tissue.