In order to elucidate the mechanism of disturbances of gonadal hormones secretion in anorexia nervosa 14 female patients were investigated. A control group also consisted of 14 women of the same age. The serum LH, progesterone, oestrogens: oestrone + oestradiol (Oe1 + Oe2), oestriol (Oe3) and testosterone were determined by radioimmunological methods. In patients with anorexia nervosa the serum testosterone and Oe3 concentrations were dramatically elevated, whereas LH, progesterone and Oe1 + Oe2 were decreased as compared with the control group. Considerable weight gain induced by cyproheptadine treatment caused a normalization of the serum testosterone and Oe3 concentrations in all the patients. A negative correlation between the testosterone level and the deficit in body weight was observed. The mechanism causing the dramatically high serum testosterone concentration in the female patients with anorexia nervosa is discussed.

The aetiology of polycystic ovary syndrome (PCOS) is unknown. It is uniquely characterized by oligomenorrhoea or amenorrhoea associated with normal or high oestrogen levels. This prospective clinical study was designed to examine the possible role of the lack of cyclical exposure to progesterone in the development of gonadotrophin and androgen abnormalities in PCOS. Gonadotrophin, androgen and oestrogen levels were measured in 15 PCOS patients and 10 normal subjects untreated and following treatment with the progestogen medroxyprogesterone acetate (MPA). When compared to control subjects, PCOS patients had significantly higher luteinizing hormone (LH) pulse height, pulse amplitude, integrated LH levels, LH response to gonadotrophin-releasing hormone (GnRH) and LH/FSH ratio; LH pulse frequency was similar in the two groups. In addition, the testosterone/sex hormone binding globulin ratio (T/SHBG), androstenedione and oestrone concentrations in the plasma were significantly higher in PCOS than in control subjects. When PCOS patients were treated with MPA for 5 days, there were significant decreases (p < 0.02-0.001) to values no longer different from normal: from 8.7 +/- 1.2 to 5.6 +/- 0.8 IU/l for integrated LH levels (untreated and MPA-treated PCOS); from 31.2 +/- 3.5 to 12.9 +/- 1.5 IU/l for LH response to GnRH; from 2.4 +/- 0.26 to 1.3 +/- 0.2 for LH/FSH ratio; and from 10.4 +/- 0.63 to 8.5 +/- 0.7 nmol/l for androstenedione. Significant decreases (p < 0.05-0.005) to values that still remained significantly higher than in normal subjects occurred for: LH pulse height, 11.05 +/- 1.3 to 6.88 +/- 0.79 IU/l (untreated and MPA-treated PCOS); LH pulse amplitude, 2.8 +/- 0.5 to 1.8 +/- 0.2 IU/l; total testosterone, 2.5 +/- 0.2 to 2.0 +/- 0.2 nmol/l; T/SHBG ratio, 14.1 +/- 1.7 to 11 +/- 1.5; and oestrone, 265 +/- 24 to 208 +/- 29 pmol/l. These results are consistent with the concept that ovulation failure and progesterone deficiency play a facilitatory role in the development of the hypothalamic-pituitary abnormality giving rise to disordered LH secretion in PCOS.

The activity of the enzyme 17 beta-hydroxysteroid dehydrogenase (17 beta-OHSD) and concentrations of oestrone (E1), oestradiol (E2) and progesterone have been measured in leiomyoma and myometrium obtained at different stages of the menstrual cycle. Apart from conversion of E2 to E1 in the proliferative phase, no significant difference in enzyme activity was noted between normal and tumour tissue. However, interconversion in both tissues was shown to be higher in the secretory than the proliferative phase of the menstrual cycle. E1 concentrations were significantly higher (P less than 0.01) in leiomyoma than in myometrium, obtained during the proliferative phase. Concentrations of both oestrogens, in some tumour and normal tissues, were higher in the proliferative than the secretory phase. Secretory phase tissues contained higher concentrations of progesterone than those obtained in the proliferative phase of the menstrual cycle. Considerable differences in both enzyme activity and steroid concentrations were noted in different areas of the same tumour.

Oestradiol was converted to oestrone about ten time more rapidly by subcellular fractions of normal human endometrium of the secretory phase than by tissue of the proliferative phase. In subcellular fractions of endometrial carcinoma the 17beta-hydroxysteroid dehydrogenase (17beta-HSD) activity decreased with decreasing differentiation of the tumour. Most of the 17beta-HSD activity was located in mitochondrial and microsomal fractions of both normal and neoplastic endometrium. After treatment of patients with gestagens only the well differentiated carcinomata significantly increased in 17beta-HSD activity demonstrating that the hormonal stimulus leads to similar effects on the 17beta-HSD activity as in normal endometrium. Furthermore quantitative aspects of the in vitro binding of 3H-oestradiol and 3H-progesterone to receptor components from normal endometrium and endometrial carcinoma cytoplasmic fractions have been studied. In normal tissue the number of cytoplasmic binding sites for both oestradiol and progesterone varied dramatically during the menstrual cycle: number of oestradiol binding sites were highest during the proliferative phase and fell during the secretory phase; for progesterone site the contrary was the case. In all endometrial carcinomata high oestradiol binding activity was observed. In contrast the number of progesterone sites in the tumours was related to the state of differentiation, which paralled the progestional sensitivity of these tumours.

Oestrone, oestradiol and testosterone levels were measured by radioimmunoassay in male rats 0.5, 1, 4, 8 and 16 days after rats were either made artifically cryptorchid or sham-operated. Oestradiol levels were not significantly different between cryptorchid and control rats 12 h or 1 day after surgery, but levels in cryptorchid animals fell to 35% of controls on day 4 (P less than 0.05), 31% on day 8, and 29% on day 16 (P less than 0.01). Conversely, oestrone and the total of the two oestrogens was higher in cryptorchid rats at one-half day (P less than 0.05), but did not differ at any other time. Testosterone levels were generally lower in cryptorchid rats than in controls. The minor contribution of oestradiol to total oestrogen levels and the lack of change of total oestrogens in cryptorchid rats led to the conclusion that oestrogens are probably not the tubular regulator of FSH in the male.

The present study was undertaken to evaluate long-term effects of spironolactone on basal serum oestrone, oestradiol, testosterone, LH and prolactin concentrations in hypertensive male patients. Serum prolactin response to TRH was also evaluated. Patients were divided into two groups: a conventional-dosage group, consisting of six males with essential hypertension who took 75 to 150 mg of spironolactone daily for 12 weeks, and a high-dosage group, consisting of two males with idiopathic hyperaldosteronism who took 300 mg of spironolactone daily for more than 40 weeks. In the conventional-dosage group, serum oestrone concentrations significantly increased (P less than 0.01) at 12 weeks, serum oestradiol concentrations gradually increased throughout the study period, however, the increments were not statistically significant (P less than 0.2). Basal serum testosterone, LH and prolactin concentrations were not significantly changed throughout the study period. Enhancement of serum prolactin response to TRH was not found in any of the patients in the conventional-dosage group. In the high-dosage group, serum oestrone maintained high levels from the beginning of this study, and serum oestradiol concentrations increased with the development of gynaecomastia. Serum testosterone, LH and prolactin concentrations did not show any definite change throughout the study period. Thus, long-term spironolactone treatment increased the serum levels of oestrone and oestradiol in hypertensive men followed by the development of gynaecomastia. The elevation in circulating oestrogens could well explain the oestrogenic side-effects of spironolactone treatment.

The changes in concentration of plasma oestradiol, oestrone, progesterone, androstenedione, testosterone, cortisol and FSH were followed in intact female ferrets brought into oestrus by extension of the photoperiod from 8 to 16 h daily. An additional group of spayed females was similarly exposed to the extended photoperiod. There was no change in the blood oestrone, androstenedione and testosterone levels in the spayed females; the concentration of oestradiol, progesterone and FSH fell, while that of cortisol rose after 6 weeks. The intact females showed no change in plasma oestrone and cortisol concentrations, a rise in plasma oestradiol associated with the onset of oestrus, and falls in the blood levels of testosterone, androstenedione, progesterone and FSH. These results indicate that the changes in plasma gonadal steroid levels after extension of the photoperiod differ markedly from those in rodents or ruminants.

A direct enzyme immunoassay was developed to measure conjugated oestrogens in the plasma of pregnant mares. The antibody was produced in rabbits using oestrone-3-glucuronide (E1G) conjugated to bovine serum albumin. The enzyme conjugate was E1G conjugated to horseradish peroxidase. A sharp increase in plasma E1G concentrations occurred between Days 35 and 40 of gestation. Values declined slightly to Day 45, remained relatively constant to around Day 70 and rose sharply thereafter. Fetal death before Day 35 had no effect on plasma concentrations of E1G. Fetal death after Day 35 in conjunction with endotoxin-induced regression of the corpus luteum (CL) resulted in a decrease in plasma E1G levels to non-pregnant values within 3-4 days. Endotoxaemia without fetal death between Days 35 and 70 resulted in marked, but transient, decreases in plasma E1G concentrations. Fetal death without CL regression after Day 35 did not produce an immediate decline in plasma E1G concentrations and existing levels were maintained for 10-14 days. These findings indicate that between Days 35 and 70 of pregnancy, plasma E1G concentrations are not directly correlated with fetal viability; they appear to reflect increased oestrogen secretion by the CL under the influence of chorionic gonadotropin. After Day 70, E1G concentrations begin to reflect directly the production of oestrogen by the developing feto-placental unit.

Oestrogen sulphamates have increased systemic, but reduced hepatic oestrogenicity which results from their sequestration and transport through the liver by red blood cells. Oestrogen sulphamates act as prodrugs for the release of natural oestrogens but, as yet, there is little information as to how the sulphamoyl moiety is cleaved from the steroid nucleus. In the present investigation we have used the potent steroid sulphatase (STS) inhibitor, 667 COUMATE, to explore the possibility that STS might be responsible for the hydrolysis of oestrogen sulphamates. Administration of oestrone sulphamate (10 microg/day, s.c., for 5 days) to ovariectomised rats resulted in a 3.5-fold increase in the uterine weights of treated animals. Co-administration of oestrone sulphamate and 667 COUMATE (2 mg/kg) completely blocked STS activity in treated animals and completely abrogated the ability of oestrone sulphamate to stimulate uterine growth. In vitro studies, using [(3)H]oestrone sulphamate or [(3)H]oestrone, revealed that the uptake of the sulphamate derivative (95.9+/-2.4%) by red blood cells was considerably higher than that for the non-sulphamoylated oestrogen (25.1+/-1.9%). Results from these studies demonstrate convincingly that STS is the enzyme responsible for the removal of the sulphamoyl group from oestrogen sulphamates. This enzyme therefore has a crucial role in regulating the oestrogenicity associated with this class of drug.

Circulating testosterone is usually reduced in men with cirrhosis, but there has not been a comprehensive analysis of androgen status or circulating oestrogens. Little is known about associations between circulating sex steroids with aspects of health in this population.

We report data from men with cirrhosis and low serum testosterone (<12 nmol/L or calculated free testosterone <230 pmol/L). Comprehensive circulating sex steroid profiles were measured by liquid chromatography-mass spectrometry and compared with age-matched controls. Relationships between sex hormone levels, severity of liver disease, biochemistry and clinical outcomes were assessed.

Serum oestrone and oestradiol were significantly elevated in men with cirrhosis compared with controls (median, 869.1 pmol/L vs. 133.8 pmol/L and 166.7 pmol/L vs. 84.6 pmol/L respectively). Serum oestrone correlated with MELD score (correlation +0.306, P < 0.001) and inversely correlated with serum sodium (correlation -0.208, P = 0.004) and haemoglobin (correlation -0.177, P = 0.012). No such correlations were observed for oestradiol. Serum testosterone levels inversely correlated with MELD score (correlation -0.294, P < 0.001) and positively with handgrip strength (correlation +0.242, P < 0.001), physical activity (correlation +0.276, P = 0.012), haemoglobin (correlation +0.282, P < 0.001) and serum sodium (+0.344, P < 0.001). Dihydrotestosterone inversely correlated with MELD score (correlation -0.225, P = 0.002) and shared similar significant relationships to testosterone.

Low serum androgens and elevated serum oestrone (but not oestradiol) are associated with higher MELD and individual adverse health outcomes in cirrhotic cohort of men selected for low testosterone. Serum oestrone may be a novel marker of ill health in this population. Whether low androgens are markers or mediators of ill health requires further investigation.

Circulating levels of oestrone and progesterone were measured by radioimmunoassays in plasma samles from 5 female owl monkeys on 60 consecutive days. Both steroids exhibited cyclic fluctuations and based on nadir to nadir intervals the ovarian cycle was estimated to be 15.92 +/- 0.26 days. Levels of oestrone and pregnanediol-3 alpha-glucuronide were also measured in daily urine samples. The fluctuations of urinary steroids reflected those observed in plasma. Ketamine sedation had no effect on the length of the cycle. Peak values of plasma progesterone and oestrone were 250.48 +/- 11.37 and 3.59 +/- 0.066 ng/ml respectively. There was no clear hormonal distinction between the follicular and luteal phase of the cycle in these owl monkeys.

Adult baboons (5 males and 5 females) were exposed to immobilization stress by being strapped to a table in a horizontal position for 2 h. In females the experiment was performed during both the follicular and luteal phase. Peripheral blood was withdrawn at frequent intervals, the first sample just before immobilization, and the last one 3 days later. A number of steroids were measured in blood plasma samples by radioimmunoassay (17-hydroxypregnenolone, 17-hydroxyprogesterone, pregnenolone, testosterone, dihydrotestosterone, progesterone, 20alpha-dihydroprogesterone, oestone, oestradiol) or competitive protein binding (cortisol) techniques. The cortisol levels exhibited a marked increase in both sexes. This increase was observed already during the immobilization and lasted for approximately 24 h. A similar, even more pronounced increase was seen in 17-hydroxypregnenolone, 17-hydroxyprogesterone and pregnenolone levels. A marked, long-lasting (72 h) decrease of testosterone and dihydrotestosterone levels was a consistent finding in male baboons. This was not observed in the females which, on the other hand, exhibited a marked decrease (duration 48 h) of progesterone and 20alpha-dihydroprogesterone levels during the luteal phase, and a significant decrease (duration greater than 24 h) of oestradiol and oestrone concentrations during the follicular phase. It is concluded that stress has a marked inhibitory action on gonadal function both in male and female baboons. In females and inhibition of steroidogenetic function is exerted both on the ovarian follicles and on the corpus luteum.

Aromatase is the cytochrome P-450 complex responsible for oestrogen biosynthesis in vivo. Inhibitors of this enzyme complex might therefore serve to modulate oestrogen-dependent processes by interfering with the production of oestrogens. Thus, these agents may be useful in reproductive processes and in treating oestrogen-dependent disease states such as breast and endometrial cancer. We have demonstrated that inhibitors such as the naturally occurring flavonoids having 5,7-dihydroxy substituents can bind to human placental cytochrome P-450 with affinity comparable to their ability to inhibit aromatization of androstenedione and testosterone to oestradiol and oestrone, respectively. It appears that the mechanism of this inhibition requires the flavonoid to bind to the active site of the cytochrome P-450 without prior generation of metabolic intermediate products. Our data also suggest that the presently known differences in potency of inhibition of cytochrome P-450-mediated aromatization of steroids by different hydroxylated derivatives of 5,7-dihydroxyflavones may arise from their different binding affinity to the enzyme, particularly those compounds hydroxylated in the C3 position in ring C of the flavonoid nucleus.

OBJECTIVE

Gross cystic disease (GCD) of the breast is reported to occur in 7% of women in the developed world and, although not premalignant, is thought to be associated with an increased risk of breast cancer. Hormone and growth factor concentration levels were measured in breast cyst fluid (BCF) to correlate them with their mitogenic activity in tumour (MCF-7) or nontransformed (MCF-10A) cells.

RESULTS

Oestradiol (E2), oestrone (E1), E2-sulfate (E2-S), E1-sulfate (E1-S) and epidermal growth factor (EGF) concentrations were, as expected, significantly higher in type I than in type II cysts, while transforming growth factor-beta 2 (TGF-beta2) showed higher levels in type II cysts. Fifty per cent of the BCF samples stimulated [3H]-thymidine incorporation into MCF-7 cells while 34.5% inhibited this parameter. In MCF-10A cells, most BCF samples were stimulatory (85%). E2, E1 and EGF concentrations in BCF samples correlated significantly and positively with cell proliferation in MCF-7 cells, whereas a significant negative correlation was found for TGF-beta2. In MCF-10A cells, only E2-S and E1-S exhibited significant positive correlation, whereas a significant negative correlation was found for TGF-beta2. Progesterone (Pg), E2 and EGF incubated under the same conditions had a stimulatory effect on [3H]-thymidine incorporation into MCF-7 cells, whereas TGF-beta2 inhibited this parameter. Pg, E2, E1 and EGF significantly stimulated this parameter in MCF-10A cells.

CONCLUSIONS

The stimulatory action of BCF on cell proliferation in a model of human breast epithelial cells could partly explain the increased incidence of breast cancer in cyst-bearing women.

The present studies examined whether the pituitary hormones, particularly prolactin, have any role in the regulation of rat placental lactogen (rPL) secretion. Total rPL was measured using a lymphoma cell bioassay. Hypophysectomy on day 13 of pregnancy increased total serum rPL levels above those of intact control rats and delayed for 3 days the decline in rPL usually seen by day 14. To test the effect of early hypophysectomy on rPL secretion, a delayed implantation model was used. Hypophysectomy was carried out on day 3, progesterone (2 mg) was given daily until day 8 and oestrone (1.0 micrograms) was given on day 8. This initiated implantation on day 9, which is 4 days later than normal. Rats hypophysectomized on day 3 had high serum levels of rPL (10-13 mg/l) 7 days after initiation of implantation compared with control values of 2-3 mg/l, and these levels remained raised for the duration of the experiment. Termination of maintenance injections of steroids did not affect the high levels of rPL for several days, even though there was fetal and placental resorption. When a pair of anterior pituitaries was transplanted under the kidney capsule on day 13 (day 9 of development) of rats hypophysectomized on day 3, serum rPL levels still increased for the next 3 days. However, unlike similarly treated rats not receiving transplants, rPL levels fell rapidly thereafter and were only 5% of those in control animals 1 week later.(ABSTRACT TRUNCATED AT 250 WORDS)

Postmenopausal women with metastatic breast cancer were treated with trilostane, initially 240 mg daily increasing after 3 days to 480 mg daily and after a further three days to 960 mg daily. After 3 days at this dose dexamethasone 1 mg daily was added and this combination was continued until disease progression occurred. Partial remission was seen in 26% and stabilization of previously progressive disease in a further 13% of the first twenty-three patients studied. During therapy with trilostane alone significant increases in DHEAS, androstenedione, 17-hydroxypregnenolone, progesterone, testosterone and oestradiol were seen. A significant fall in oestrone concentration occurred at the same time. After dexamethasone was added the elevated steroid concentrations fell back to the baseline while oestrone remained depressed below this and testosterone was also significantly lowered. No change was seen in cortisol or ACTH concentration while patients were on trilostane alone but cortisol levels were undetectable after dexamethasone was added though, in most patients, ACTH remained detectable. There was no change in the ratio of delta 5:delta 4 steroids at any stage of therapy but a highly significant increase in the androstenedione: oestrone ratio was seen. We conclude that in long-term use in vivo it is difficult to demonstrate that trilostane inhibits 3 beta-hydroxysteroid dehydrogenase but it may produce inhibition of aromatase.

During the period 1 month before and 1 month after parturition in the cow, several events take place. The dam has to be prepared for the impending parturition and the uterus and ovaries must return to a certain stage to be prepared for a new pregnancy. Most of these processes are due to or reflected in endocrine changes. A special interest is of course the status of the foetus --"foetal well being". The processes could either be considered as normal in a clinical perspective or as impaired (dystocia, small calves, stillbirth, retained foetal membranes, etc.). The main question for this presentation is if normal and impaired performance could be mirrored in endocrine parameters. Many studies have been performed to follow endocrine changes during the periparturient period in the cow. The following parameters have been shown to be the most important and seem to be the most suitable for an endocrine supervision: Endocrine parameter: progesterone; parameter of: corpus luteum, maternal adrenals, placenta. Endocrine parameter: prostaglandin (PG) metabolite; parameter of: placenta, uterus, inflammation. Endocrine parameter: cortisol; parameter of: regulator of prostaglandin synthesis, stress. Endocrine parameter: free oestrogens; parameter of: placenta, ovaries. Endocrine parameter: oestrone sulphate; parameter of: placenta, calf weight. Endocrine parameter: pregnancy-associated glycoproteins (PAG); parameter of: placenta.

We have measured urinary excretion of free immunoreactive oestrone and oestradiol and their respective glucuronides in relation to creatinine in early morning samples in 132 fit, active postmenopausal women. None of the oestrogen/creatinine ratios was significantly correlated either with age or with years since menopause. However, we did demonstrate significant positive correlations between the levels of all four oestrogens and the body mass index, weight and fat mass. These results are similar to those obtained by other workers for plasma or serum oestrogen levels. Since assessment of oestrogenic status from plasma or serum may call for several samples to allow for the rapid minute-to-minute variation in levels, urinary oestrogenic assays as described may provide a valuable non-invasive measurement of oestrogenic status. Such measurements may have a place in the identification of women at greater risk of developing symptomatic osteoporosis in later life.

The plasma levels of oestrone (Oe1), 17 beta-oestradiol (Oe2), oestriol (Oe3), testosterone (T), 5 alpha-dehydrotestosterone (DHT) and androstenedione (A) were assayed by RIA in plasma obtained from peripheral venous blood. The hormones are isolated from the plasma extract, first by Sephadex LH-20 column chromatography (Oe2/Oe3/Oe1, T, DHT, A) and after Oe1, T, DHT, A by TLC on silica gel 60 F254. The accuracy, reproducibility and sensitivity of the method make it satisfactory for clinical studies.

Eight bilaterally oophorectomized women were given a depot injection of 200 mg DHEA-enanthate to study the effect on endocrine and lipid metabolism. A decrease in sex-hormone binding globulin (SHBG) and an increase in androstenedione was found 14 and 30 days after the injection. No changes could be detected in LH, FSH, oestrone, oestradiol or oestriol. Testosterone showed a tendency towards an increase. As compared to pre-treatment values, plasma lipids were unaltered after 30 days. A decrease in high density lipoproteins (HDL), cholesterol and in very low density lipoproteins (VLDL), free cholesterol, total cholesterol and phospholipids were seen in the lipid composition of the lipoproteins on day 30. These findings are in agreement with previous data reported after the administration of drugs with androgen-like effects. The relative fatty acid composition of plasma lecithin revealed only minor changes while the fatty acid composition of cholesterol esters indicated a decreased portion of essential fatty acids. These results suggest, in agreement with previous studies, an impaired endogenous cholesterol formation in the liver. The results from the analysis of the fatty acid composition of lecithin and cholesterol esters might indicate a decreased percentage of exogenous (dietary) cholesterol ester in plasma.

Microsomal UDPglucuronosyltransferase(1-naphthol), an enzyme form previously shown to be selectively inducible in rat liver by 3-methylcholanthrene-type inducers, was purified to apparent homogeneity. Rabbit antibodies against this enzyme form precipitated UDPglucuronosyltransferase activities towards 1-naphthol and 4-methylumbelliferone faster and to greater extents than enzyme activities towards bilirubin, oestrone and 4-hydroxybiphenyl. Ouchterlony double-diffusion analysis showed immunochemical similarity of the rat liver enzyme with the enzymes from other organs of the rat (kidney, testes) and the mouse liver but not with the enzyme from cat and human liver. Electroimmunochemical quantification of the enzyme indicated that its level was enhanced 1.3-fold and 2.5-fold in liver microsomes from phenobarbital-treated and 3-methylcholanthrene-treated rats, respectively. The results indicate that 3-methylcholanthrene treatment increases the enzyme level of rat liver microsomal UDPglucuronosyltransferase(1-naphthol). Despite phospholipid-dependence of its catalytic activity microsomal enzyme activity appears to be a good index of the enzyme level.

1. A partial purification of kynureninase (L-kynurenine hydrolase, EC 3 . 7. 1 . 3) from rat liver and a total resolution of the apoenzyme have been achieved. The hypothesis that conjugates of oestrogenic steroids compete with pyridoxal phosphate for the cofactor binding site of the enzyme, and so disturb tryptophan metabolism, leading to apparent vitamin B6 deficiency, has been tested. 2. Kynureninase from rat liver was partially purified, and the cofactor-free apoenzyme was prepared. Oestrone sulphate inhibited the enzyme uncompetitively with respect to pyridoxal phosphate, and competitively with respect to kynurenine, with a mean (+/- SE) inhibitor constant (Ki) of 82 +/- 6 microM. 3. The addition of a saturating concentration of pyridoxal phosphate to unfractionated liver homogenates led to an approximately fivefold increase in kynureninase activity, indicating the presence of a relatively large amount of apo-kynureninase in the tissue. 4. It is suggested that the abnormal results of tryptophan load tests in women receiving oestrogens are the result of inhibition of kynureninase by oestrogen conjugates, and that there is no evidence for oestrogen-induced vitamin B deficiency in such cases.

We have tested the hypothesis that androstenedione (administered as 21-day, slow-release pellets) is converted to active sex steroids and reduces bone turnover in the ovariectomised rat model. We found that ovariectomy resulted in a minor but significant reduction in plasma concentrations of androstenedione and testosterone and a more significant reduction in oestrone (E1) and oestradiol (E2). This was associated with the expected substantial loss of metaphyseal cancellous bone volume. Androstenedione (1.5-100 mg) pellets increased the plasma concentrations of androstenedione and testosterone above those in the ovariectomised (ovx) rats in a dose-responsive manner, whereas E2 plasma concentrations were increased to a minor but significant degree above those in the ovx animals. Androstenedione reduced loss of cancellous bone volume in a dose-dependent fashion by reducing bone turnover. The 1.5, 5 and 100 mg androstenedione-induced effect on bone turnover was not abrogated by simultaneous treatment with Arimidex, an aromatase inhibitor. This implies that the skeletal-protective effect of androstenedione was not oestrogen-mediated.

In twenty oligospermic or azoospermic patients with elevated plasma FSH, the mean concentrations of plasma oestrone sulphate (843 +/- 233 pg/ml), oestrone (54 +/- 10.4 pg/ml) and oestradiol (46.6 +/- 12.6 pg/ml) were found to be significantly higher than in twenty-one normal fertile men of comparable age (593 +/- 220 pg/ml, 40.6 +/- 8.8 pg/ml and 33.1 +/- 10.9 pg/ml respectively). SHBG binding capacity was elevated in the infertile group (infertile 3.35 +/- 0.82 x 10(-8) M/l v. normal 2.76 +/- 0.89 x 10(-8) M/l) but the total plasma testosterone concentrations were comparable (infertile 5435 +/- 1578 pg/ml v. normal 5046 +/- 1102 pg/ml). Evidence was cited to support the view that Sertoli cells, in response to an unphysiological FSH stimulation, are a likely source of excessive oestrogen production. The possible significance of increased intra-testicular and circulating oestrogen in the altered state of testicular steroidogenic function in men with primary seminiferous tubular defects was discussed.