Radioimmunoassay that measures serum vitamin B12.

AIMS: To develop a specific radioimmunoassay for the routine determination of serum vitamin B12. METHODS: Antisera were raised in rabbits by immunisation with the monocarboxylic acid derivative of cyanocobalamin coupled to human serum albumin. Antibody titres and affinities were determined and the antiserum giving the highest binding affinity constant, Ka, was used to develop the assay protocol. Donkey-anti-rabbit gamma globulin-coated magnetisable particles were used to separate the bound from free vitamin B12. The considerable cobalamin binding capacity of human serum was destroyed by autoclaving in acetate-cyanide buffer. Sixty samples were assayed by the radioimmunoassay (RIA) and the Lactobacillus leichmannii assay. Recovery and cross-reactivity experiments were performed. RESULTS: Final rabbit antibody titres varied from 1/20,000 to 1/188,000. Scatchard plots did not correlate with the antibody titres. The Ka values varied from 2.6 to 6.7 x 10(10) litres/mol. For maximum sensitivity the highest Ka (titre 1/66,000) was chosen. A tracer concentration of 22 pmol/l, an antiserum dilution of 1/100,000, and a sample volume of 0.1 ml were used. At an antiserum dilution of 1 in 100,000 the cyanocobalamin binding of the rabbit serum was diluted out. The assay showed excellent correlation with the microbiological assay, with 100% recovery of added vitamin B12. Levels of cross-reactivity for dicyanide cobinamide and hydroxocobalamin were 9.8 and 8.1%, respectively. CONCLUSIONS: The development of this immunoassay permits the measurement of serum vitamin B12 without important interference from cobalamin analogues, related corrinoids, and non-specific binders.

Hormonal studies on women with polycystic ovaries diagnosed by ultrasound.

Eighty-six patients were diagnosed by ultrasound to have polycystic ovaries. Twelve have fairly regular but anovular cycles (P1), 56 had oligomenorrhoea (P2) and eight had primary or secondary amenorrhoea (P3). The remaining 10 patients had raised progesterone concentrations and were excluded from the hormonal studies. Gonadotrophins, steroid hormones and sex hormone binding globulin (SHBG) were measured in all patients and the results were compared with those of 29 normal women who acted as controls. A high LH:FSH ratio was the most frequently found abnormality (raised in 68.4% of patients) followed by LH (65.8%), free testosterone (FT, calculated from total testosterone (T) and SHBG) (59.2%) and 17-OHP (48.7%). Each of these four estimations was above the normal ranges in 25% of patients: proportionally most of the results outside the normal range occurred in group P3 followed by P2 and P1 respectively. To obtain a quantitative measure of the abnormalities, results were expressed as percentages of the mean control value for each hormone and these percentages were then summated. The four measurements, LH:FSH ratio, LH, FT and 17-OHP gave abnormality scores ranging from a mean of 667 in P1 to 1189 in P3 compared with 405 in controls. Twenty-one of the 76 patients were hirsute with significantly higher androgens than the nonhirsute patients. Their mean abnormality score was 1141 compared with 855 in the non-hirsute group and they all had at least one hormonal abnormality. The study demonstrates that there is a relationship between the degree of hormonal abnormality and the menstrual irregularities and hirsutism in women with PCOS suggesting that there may be a progressive nature to the syndrome.

Regulation of growth hormone secretion and messenger ribonucleic acid accumulation in human somatotropinoma cells in vitro.

GH secretion and mRNA levels were measured in cultured cells obtained from six human pituitary somatotroph tumors to investigate their hormonal and intracellular regulation. The responses were variable between tumors, but, in general, mRNA levels were less responsive than GH release to in vitro manipulation. GH-releasing factor [GRF-(1-29) amide; 10 nM] increased GH release and mRNA levels in three of four tumors tested to 30-97% above control values, but the fourth tumor was unresponsive. Somatostatin (1 microM) inhibited GH release significantly in four of the six cases, to 35-79% of control levels, but had no inhibitory effect on GH mRNA accumulation, in contrast to earlier studies on rat pituitary tissue. Bromocriptine (100 nM) likewise inhibited GH release (50-75% of control), but not GH mRNA levels, in the four tumors tested. Forskolin (10 microM; used to activate adenylate cyclase) stimulated GH release and mRNA levels in the two cases that responded most clearly to GRF, but had no significant effect in the other tumors; however, the phorbol ester 12-O-tetradecanoyl phorbol-13-acetate (100 nM) had no consistent effect on mRNA levels despite stimulating secretion in four of six cases. Thus, there was considerable variation in responses among the tumors tested; however, the responsiveness to GRF was approximately paralleled by that to forskolin, consistent with the suggestion that adenylate cyclase activity and responsiveness are variable among these tumors. Furthermore, the divergent effects of somatostatin on GH release and mRNA suggest uncoupling between its receptor and transcriptional regulatory mechanisms.

An enhanced chemiluminescent enzyme immunoassay for follicle stimulating hormone.

An enhanced chemiluminescent enzyme immunoassay for serum follicle stimulating hormone is described which involves sequential reaction of anti-follicle stimulating hormone antibody immobilised to the inside surface of an opaque microtitre plate with sample, monoclonal anti-alpha thyroid stimulating hormone antibody, and an anti-mouse IgG - horseradish peroxidase conjungate. Bound peroxidase activity was measured using a p-hydroxycinnamic acid enhanced chemiluminescent luminol-hydrogen peroxide reaction. The assay was sensitive (detection limit 0.01 mU/well) precise (intra-assay precision 2.5-8.1%, inter-assay precision 6.7-11.9%) and results obtained with this assay and a competitive radioimmunoassay were in good agreement (correlation coefficient 0.98).

Bioactivity of growth hormone releasing hormone (1-29) analogues after SC injection in man.

The bioactivity of growth hormone releasing hormone 1-29 [GHRH(1-29)NH2] has been compared with that of an agonist analogue [Ac-D-Tyr1,D-Ala2]-GHRH(1-29)NH2, in normal male volunteers. Using a submaximal dose of 3 micrograms/kg administered subcutaneously, peak growth hormone (GH) response and area under the GH curve were similar for the native and agonist analogue. In addition, no significant differences were found in peak GHRH(1-29) immunoreactivity, area under the GHRH(1-29) curves or plasma disappearance rates of the two peptides. The results suggest that, in keeping with the relative activities of other "superactive" analogues tested so far, the greatly enhanced activity of [Ac-D-Tyr1,D-Ala2]-GHRH(1-29)NH2 observed in the rat is not found in humans. It is possible that this species difference is due to differences in the interaction of GHRH peptides with the rat and the human somatotroph GHRH receptor.

The extraction and purification of human pituitary follicle-stimulating hormone and luteinizing hormone.

Techniques are described for the extraction and purification of FSH and LH from human pituitary glands. The methodology uses acetone and alcohol for dehydration, extraction and precipitation of materials from both acetone preserved and fresh frozen glands. The processing involves the use of anion and cation exchange chromatography, adsorption techniques, and molecular sieve chromatography. Immunoaffinity extraction and purification of the gonadotropic hormones are also described. The specific activities of the LH preparations from acetone-preserved glands and from fresh frozen glands ranged from 11,200 to 12,640 IU/mg (68/40) and contained less than 0.2% TSH and FSH contamination. The yield varied from 9% to 18%. The specific activities of the FSH preparations from fresh frozen glands and from acetone preserved glands by affinity extraction (4990 and 4600 IU/mg 78/549) were higher than that from acetone preserved glands (3235 IU/mg). The LH and TSH contamination was less than 0.5%. The highest yield of FSH (34%) was obtained by affinity extraction. The preparations are of sufficient purity for use as immunogens and standards and for the preparation of radioligands to be used in specific and sensitive immunoassays.

Effect of bromocriptine on LH pulsatility in the polycystic ovary syndrome.

The effects were studied of bromocriptine, 10 mg daily for 1 year, on luteinizing hormone (LH) pulse characteristics in patients with classical polycystic ovarian syndrome (PCOS). All patients were hirsute, had been oligomenorrhoeic since menarche, had LH: FSH ratios of greater than 3:1, and either elevated serum testosterone (T) or dehydroepiandrosterone sulphate (DHAS) concentrations. In 10 subjects who completed the study menstrual frequency increased from an average of 3.6 to 8 per year but few of the cycles were ovulatory. Mean (SE) serum testosterone fell from 4.4 (0.5) nmol/l pretreatment to 2.8 (0.3) nmol/l (P less than 0.01) and DHAS from 7.9 (1.1) mumol/l to 5.4 (1.1) mumol/l (P less than 0.05). Serum delta 4 androstenedione and oestradiol did not change with bromocriptine treatment. Mean serum LH fell from 17.4 (2.4) IU/l to 11.2 (1.8) IU/l (P less than 0.03) after 12 months of bromocriptine. No pattern of LH pulsatility specific to PCOS was detected during 10 min sampling for an 8 h period prior to dopamine agonist treatment. LH interpeak interval (58 (5.2) min) and peak amplitude (156 (7.2%) of mean nadir) in untreated PCOS were similar to that of the mid-follicular stage of ovulatory cycles, and bromocriptine for 1 year did not alter these variables. We conclude that while bromocriptine reduces serum androgen levels and increases menstrual frequency it has no effect centrally to modify hypothalamic GnRH secretion. The reduction in LH levels by bromocriptine may be the result of diminished gonadotroph sensitivity to GnRH or reduced pituitary stores of LH available for release. Despite the return towards normal of various hormonal characteristics of PCOS, bromocriptine has little place in the management of this condition.

Is changing hypothalamic activity important for control of ovulation?

The activity of the hypothalamic gonadotrophin releasing hormone pulse generator in women with regular ovulatory and anovulatory menstrual cycles was assessed to see whether changes therein are important determinants of normal and impaired ovarian function. Endogenous gonadotrophin releasing hormone secretion was inferred by measurement of the pituitary luteinising hormone response by characterisation of pulsatile luteinising hormone release over eight hours on three occasions during the course of follicular development and once during the luteal stage of the same cycles. In 13 ovulatory cycles (serum progesterone concentration greater than 25 nmol/l) confirmed by ovarian ultrasonography a pronounced variability in luteinising hormone pulse patterns among subjects was compatible with ovulation. In the luteal stage of ovulatory cycles the luteinising hormone interpeak interval (85 min, range 42-125) was significantly longer than that during the early follicular (64 min, 40-103), mid-follicular (62 min, 37-107), and late follicular (59 min, 39-80) stages of the same cycles. Thus in ovulatory cycles no increase in frequency of the gonadotrophin releasing hormone pulse generator was detected during follicular development, though this activity decreased in the luteal stage. In five late follicular stage studies in which part of the preovulatory luteinising hormone surge was captured no change in pulse frequency of luteinising hormone was detected compared with the mid-follicular stage of the same cycles or when compared with the late follicular stage of other cycles when no luteinising hormone surge was captured. Though mean luteinising hormone concentrations in luteinising hormone surge series (36 IU/l) were high, the amplitude of luteinising hormone pulses (165%) was only slightly greater than during non-surge late follicular stage studies (145%). Hence no change in hypothalamic gonadotrophin releasing hormone activity is required to generate the preovulatory discharge of luteinising hormone in man, which occurs as a result of the sensitising action of rising oestradiol concentrations on pituitary responsiveness to the same hypothalamic input signal. Luteinising hormone pulse frequency, peak amplitude, and mean serum luteinising hormone concentrations in seven anovulatory cycles (progesterone concentration less than 10 nmol/l) were not different from those at comparable stages of ovulatory cycles. These data suggest that the primary abnormality in this group of regularly menstruating anovulatory women lies in the ovary rather than in the hypothalamic control of the anterior pituitary.

Characteristics and incidence of dysfunctional ovulation patterns detected by ultrasound.

The nature and incidence of normal and abnormal spontaneous ovarian cycles, identified with ultrasound and endocrine tracking, were examined in 45 regularly cycling infertile women with no definitive cause and 15 women who were apparently normal and were receiving donor insemination because of clearly infertile partners. In 136 cycles, four apparently distinct abnormal patterns were detected. The total incidence in the infertile group was 58% compared with 23% in the donor insemination group (P less than 0.005). Twelve of 26 subjects who had at least three cycles tracked showed two different abnormalities, and 1 subject had three different abnormalities in five abnormal cycles. These results suggest that abnormal cycles are a significant factor in unexplained infertility and that diagnosis and treatment cannot be based on the study of a single cycle.

Does growth hormone releasing factor desensitize the somatotroph? Interpretation of responses of growth hormone during and after 10-hour infusion of GRF 1-29 amide in man.

It is unclear whether growth hormone releasing factor (GRF) administration in vivo may desensitize the somatotroph. To investigate this possibility we have examined the effects of 10-h infusion of the equipotent 1-29 amide analogue of hpGRF on serum GH levels and on the subsequent GH response to a bolus dose of GRF (1-29). Four normal adult males received an intravenous infusion of 1-29 GRF (1 microgram/kg/h) from 0800 to 1800 h, with blood samples taken at 10 min intervals. A 100 micrograms intravenous bolus dose of GRF was given at 1800 h, and sampling continued for a further 90 min. GH levels were near or below the limit of detection (0.5 mU/l) throughout the control 10 h period. During GRF infusion there was increased GH release with pulses of irregular frequency and amplitude (up to 80 mU/l) continuing throughout the entire infusion period. There was no apparent reduction in total GH released towards the latter part of the infusion. On the control day, 100 micrograms GRF bolus increased mean (+/- SEM) GH from 0.82 +/- 0.21 mU/l to a peak of 59.0 +/- 44.8 mU/l (P less than 0.002). Following 10-GRF infusion, responses to bolus injection of GRF were reduced, but variable. In two subjects a small rise in GH levels occurred (basal 6.4 and 7.2 rising to peak values of 11.2 and 23.0 mU/l respectively). In the other two subjects, GH levels fell but in these the GRF bolus had coincided with a GH peak. The loss of GRF responsiveness after GRF infusion may be due to 'desensitization'.(ABSTRACT TRUNCATED AT 250 WORDS)

Follicular growth and endocrine profiles in spontaneous and induced conception cycles.

Twenty conception cycles were tracked with ultrasound ovarian scanning from the midfollicular to midluteal phase, and simultaneous hormone profiles were obtained in 18 of the cycles. Twelve cycles were spontaneous and 8 were induced with a variety of regimens. Two of the spontaneous cycles did not have endocrine data, and in one other no follicular growth was observed, despite repeated scanning. The induced cycles did not differ significantly from the spontaneous ones in any respect except the number of follicles and the consequent estradiol (E2) levels. E2 levels per unit total follicular surface area, however, were similar in both groups and remained relatively constant throughout the late follicular phase.

Pharmacokinetic and endocrinological parameters of a slow-release depot preparation of the GnRH analogue ICI 118630 (zoladex) compared with a subcutaneous bolus and continuous subcutaneous infusion of the same drug in patients with prostatic cancer.

Seventeen patients with advanced prostatic cancer were treated with the gonadotrophin-releasing hormone analogue DSer (tBU)6 AzaGly 10 GnRH (ICI 118630), either as a constant SC infusion, or in the form of a monthly SC slowrelease depot formulation, in which case patients were randomised to receive one of three doses. Six of these patients also received a 250-microgram SC bolus of ICI 118630, for pharmacokinetic studies, before starting the infusion or the depot. Drug levels were measured using a double-antibody radioimmunoassay. In contrast to the SC infusion, which gave a smooth serum 118630 level profile, drug release from the depot preparation was not constant, levels varying in a predictable manner throughout each 28-day period, reaching a peak proportional to the dose of ICI 118630 received, between days 15 and 18 of each cycle. With all methods of administration there was an initial rise in LH, usually followed by a rise in testosterone, after which the SC infusion and the depot were both effective in reducing serum LH to basal levels and testosterone into the castrate range within 1 month. It is too early to make any assessment of clinical response; however, depot treatment was well tolerated: Four patients experienced an initial flare in bone pain, probably related to the initial rise in testosterone, and twelve patients experienced flushing; one patient with pre-existing hydronephrosis and hydroureter developed renal failure, possibly related to a tumour flare reaction. No patients have experienced cardiovascular side effects or local reaction.

Study of the effect of estradiol on gonadotrophin levels in untreated male-to-female transsexuals.

This project was an attempt to test Dörner's theory that aspects of gender identity and sexual behavior depend on a defect in a normal imprinting mechanism of testicular testosterone (T) on the male hypothalamus. It has been suggested that such an action by T is incomplete in male-to-female transsexuals and that in this disorder the hypothalamus retains a cyclic pattern of gonadotrophin secretion and a positive feedback response to estrogens, such as is seen in the normal female. Five untreated male-to-female transsexuals and five normal male controls were each given 2 mg IM of estradiol, and T, estradiol, LH, and FSH globulin was measured serially over the following five days. No significant differences were found in the gonadotrophin or any of the other hormone responses in the transsexuals compared to normal controls. If there is a defect in imprinting of the hypothalamus in transsexual men, it does not seem to affect the basal levels of gonadotrophins or the pattern of their response to estrogen, at least in the intact male.

Pulsatile administration of GnRH for the treatment of hypogonadotrophic hypogonadism.

Fourteen patients, aged 22-35 years, complaining of infertility and failing to ovulate on clomiphene, were treated with GnRH administered in pulses at 90 min intervals. Four patients received a total of eight courses of GnRH given subcutaneously and 13 were given a total of 20 months of treatment with GnRH given intravenously. Serum concentrations of immunoreactive GnRH were measured in six patients before administration of the drug and at regular intervals for 60 min after subcutaneous and intravenous injections of 5, 10 and 20 micrograms GnRH. Maximum concentrations of GnRH were reached by 5-10 min after subcutaneous injections and within 2 min after intravenous injections. The peak concentrations were 3.6-6.3 times and the sums of increments were 2.0-3.9 times greater following intravenous injections than after subcutaneous injections. Subcutaneous treatments extended for 15-29 days with doses of 5-20 micrograms per pulse. Only one patient ovulated as judged by the luteal phase progesterone and ultrasonic scanning of the follicle. Intravenous treatments were from 12-22 days with doses of 10 micrograms per pulse and 16 treatments out of 20 were ovulatory with four pregnancies. HCG (5000 i.u.) was given when ultrasonic scanning indicated adequate follicular growth, but in eight of the cycles, including three of the pregnancies, the follicle had ruptured before HCG was given. Pulsatile administration of GnRH proved to be an effective treatment for infertility in hypogonadotrophic hypogonadism. Possible reasons for the better results by intravenous rather than subcutaneous injections are discussed.

Purification and characterization of guinea-pig chorionic gonadotrophin.

A human chorionic gonadotrophin-like protein (GF-1, 1.0 g) from the placentae of 50 guinea-pigs killed at Day 26 of gestation was purified by pH and ammonium salt fractionation followed by column chromatography on DEAE-Sephadex and filtration on Sephadex G-100. Relative to the Second International hCG standard (MRC 61/6) GF-1 had an immunological potency of 21 000 i.u./mg as measured in a specific hCG-beta radioimmunoassay and, using the ovarian ascorbic acid depletion assay, an apparent biological potency of 24 064 i.u./mg. Isoelectric focusing yielded 6 bands between pH 4.4 and 5.7 and the material comprised two non-covalently linked subunits. The Stokes' radii were 3.40 nm for the native preparation, and 2.38 nm and 3.15 nm for GF-1-alpha and GF-1-beta subunits respectively. The guinea-pig placenta therefore produces a chorionic gonadotrophin which on purification has physicochemical, biological and immunological properties similar to those of hCG.

Review of 59 patients with hypergonadotrophic amenorrhoea.

Fifty-nine patients presented with elevated concentrations of gonadotrophins and secondary amenorrhoea before the age of 35 years. Fifty-three underwent laparoscopic examination and primordial follicles were observed in 16. Two others had follicles as they later became pregnant and a third showed biochemical evidence of spontaneous ovulation. There were streak ovaries in 12, two with follicles and three others with chromosomal abnormalities, two being 47XXX and one XO/XX. Two other patients had only one ovary each but no follicles. Chromosomal abnormalities were detected in two further patients one being XO/XX and the other a recombinant. Six patients became pregnant, two of them twice, resulting in four term deliveries and four spontaneous abortions. Three other patients showed biochemical evidence of ovulation; one spontaneously, one after oestrogen therapy and the third after treatment with gonadotrophin releasing hormone analogue.

Maternal plasma concentrations of beta-lipotrophin, beta-endorphin and gamma-lipotrophin throughout pregnancy.

Plasma beta-LPH, beta-EP and gamma-LPH concentrations were measured by radioimmunoassay in 10 pregnant women from 12 weeks gestation until term and in nine women in the early follicular phase of the cycle. There was a progressive and significant rise in the concentration of all three peptides throughout pregnancy and by 32 weeks the concentrations of beta-LPH and beta-EP were greater than the corresponding concentrations in the follicular phase: gamma-LPH was greater than in the follicular phase by the end of pregnancy in those women who were delivered after 40 weeks. The ratio of beta-LPH to gamma-LPH did not change significantly throughout pregnancy, but there was a progressive fall in the beta-LPH/beta-EP ratio. The possible presence of a 'big LPH' to explain this finding is discussed.

Maternal and cord plasma concentrations of beta-lipotrophin, beta-endorphin and gamma-lipotrophin at delivery; effect of analgesia.

Maternal venous plasma concentrations of beta-LPH, beta-EP and gamma-LPH were compared in (i) patients undergoing vaginal delivery, 11 with an epidural block and 13 with pethidine and nitrous oxide or no analgesics; (ii) patients delivered by caesarean section, 7 under epidural block and 8 under general anaesthesia. Patients delivered by either method under epidural block had significantly lower levels of all three peptides than those receiving no epidural. There were significant negative correlations between umbilical vein beta-LPH, beta-EP and gamma-LPH concentrations and umbilical artery pH and positive correlations between beta-LPH and beta-EP but not gamma-LPH and cord PCO2 in 29 patients. There was no relation between cord levels of any of the three peptides and the method of analgesia or the route of delivery. Although concentrations of all three peptides were closely correlated to one another in either maternal or cord plasma, there was no relationship between maternal and fetal levels.

Acth function in women with the polycystic ovarian syndrome.

Serum androgen levels, including dehydroepiandrosterone sulphate (DHAS) which is thought to be solely of adrenal origin, are elevated in women with the polycystic ovarian syndrome. We have investigated the possibility that this may be due to a mild form of congenital adrenal hyperplasia by measuring basal and stimulated levels of ACTH in women with this condition and have compared them to levels in normal women. We found no difference in the diurnal rhythm of ACTH between patients and normal subjects nor any difference in stimulated levels achieved after a single-dose oral metyrapone test. Thus there is no evidence from this study to support the idea that these patients might have congenital adrenal hyperplasia. There are two alternative hypotheses to explain the elevated DHAS levels. They may be associated with the high oestrogen levels, which interfere with the enzyme 3 beta-hydroxysteroid dehydrogenase, or there may be alteration of the factors controlling adrenal androgen secretion.