Mammary tumorigenesis in chemical carcinogen-treated mice. VII. Prolactin and progesterone levels in BALB/c mice.

PIP: Pituitary and serum levels of prolactin (PRL) and serum levels of progesterone (P) were determined by polyacrylamide gel electrophoresis and radioimmunoassays in BALB/c female mice, 15-17 or 44 weeks old, treated with chemical carcinogens. Neither 1.5 mg 3-methylcholanthrene (MCA) nor 1.5-6 mg 7,12-dimethylbenz(a)anthracene (DMBA) markedly altered pituitary or serum levels of PRL in the younger mice, though DMBA increased the total pituitary content of PRL by about 33% in the 44-week-old mice. However, this increase was not correlated with the incidence of mammary tumors in the group or individuals. MCA increased serum P levels by about 22% within 50 days of the last treatment. This increase was attributable to higher serum levels of P during the diestrous and proestrous phases of the cycle. Adrenalectomy reduced serum P levels by about 60%, wheras ovariectomy had no effect. Serum P levels in 44-week-old rats were not affected by DMBA. The results fail to support the notion that MCA and DMBA promote murine mammary tumorigenesis by increasing pituitary and serum prolactin concentrations.^ieng

Antigenic sites of human growth hormone and related molecules detected by monoclonal antibodies to human growth hormone.

Four major antigenic sites for human growth hormone (hGH) were identified by 27 mouse monoclonal antibodies to hGH. Sites 1 and 2 are spatially close whereas sites 3 and 4 are located in other parts of the molecule. There also appears to be a subdivision of antigenic sites. A panel of 10 monoclonal antibodies, which included representatives from each antigenic site group, were used to determine cross-reactivities between hGH and human placental lactogen (hPL), human prolactin (hPRL), the 20,000 mol. wt variant of hGH (hGH20K) and a disulfide-linked dimer of hGH (diS-dimer). The data suggest a high conformational dependence of antigenic sites in hGH. DiS-dimer retains all four antigenic sites of hGH, although all have been altered. hGH20K retains sites 2-4 but site 1 has been dramatically altered. hPL retains site 3, whereas sites 1 and 4 have been dramatically altered and site 2 may be lacking. The extremely low cross-reactivity observed for hPRL is consistent with the dissimilarity between hGH and hPRL. Antigenic site 3 is the most conserved of all sites. The lack of structural similarity compared with hGH of site 1 in hGH20K and of a portion of site 3 in diS-dimer suggests that it may be possible to develop specific radioimmunoassays for these structural variants of hGH.

Effect on growth of prolactin deficiency induced in infant mice.

The effect of lowering PRL levels in blood during early infancy on subsequent growth and development was studied in mice. PRL was reduced by injecting either an antiserum raised against homologous PRL or a PRL-inhibiting drug, 2-chloro-6-methylergoline-8 beta-acetonitrile methanesulfonate (ergoline), into 4-day-old mice for a period of 4 or 5 days. Both the anti-PRL serum and ergoline rapidly killed some of the injected animals, but the effect of anti-PRL serum was much more severe than that of ergoline (39% vs. 8.7% mortality during the period of injection). Similar administration of an antiserum against mouse GH or the GH-inhibiting peptide somatostatin did not cause a significant number of deaths. The deaths from the anti-PRL serum largely ceased when the antiserum was neutralized with rat PRL (NIH-RP-1) before injection. The gain in body weight of baby mice was markedly retarded within 24 h of injecting anti-PRL serum and ergoline, in contrast to the anti-GH serum and somatostatin injections, which took 3--4 days to inhibit growth perceptibly. The anti-PRL serum, despite having only one eighth the titer of anti-GH serum, was by far the most effective of the two antisera in diminishing tibial epiphyseal cartilage width as well as weights of pituitary glands, testes, and adrenals and retarding sexual maturity. The more severe and generalized developmental abnormalities and the incidence of mortality as a result of anti-PRL serum administration suggest that PRL in mice may be involved in the maintenance of some vital function during infancy.

Serum growth hormone and prolactin during and after the development of the obese-hyperglycemic syndrome in mice.

Mice with the recessively inherited obese-hyperglycemic syndrome (ob/ob) and their nonobese litter mates were studied over a 26-week period. The body weights and serum glucose levels of ob/ob mice began to rise markedly at 5-6 weeks of age and remained elevated throughout the period of study. Obese mice were significantly heavier (P less than .001) and had higher serum glucose levels (P less than .001) than lean mice, but obese mice had variably lower serum growth hormone (GH) and prolactin (PRL) levels (P less than .001) than lean litter mate controls after 4-5 weeks of life. A 24 h rhythm study performed on 15-week-old mice revealed a relatively unaltered but attenuated pattern of GH and PRL secretion in ob/ob mice. During and after the development of the obese-hyperglycemic syndrome, the low levels of these two hormones probably indicates an altered hypothalamic regulation of pituitary function.

Control of prolactin and growth hormone secretion in mice by obesity.

Prolactin (PRL) and growth hormone (GH) secretions in mice rendered obese by the administration of gold thioglucose (GTG) are abnormal. The objective of the present experiments was to determine whether the effects were related to the drug or to the resultant obesity. Perphenazine-induced PRL release in normal mice and in GTG-injected non-obese mice was compared to that of GTG-injected obese mice after the initial development of obesity, after body weight reduction by diet control and after the resumption of obesity by ad lib. feeding. The GTG-injected mice which did not become obese had greater (50%) than normal levels of serum PRL following perphenazine stimulation in 2 of 3 experiments. This suggested that the injection of GTG directly affected the control mechanism for PRL secretion, but that the abnormal PRL secretion was probably not the cause of obesity that develops after GTG treatment. Perphenazine-induced PRL levels in mice rendered obese with GTG were much greater (2-3 times higher than normal). However, the unusually high levels of PRL were totally abolished when the body weights of these mice were brought down to normal by dietary restriction. Conversely, when obesity was permitted to recur by giving the mice free access to food, PRL levels reverted back to the original obese pattern. The concentrations of GH were usually lower than normal in GTG-obese mice, and these levels were also more often associated with the development of obesity than with the injection of GTG. The data show a marked influence of obesity on the control of PRL and GH secretions in the mouse.

Prolactin and growth hormone secretion in chemically induced and genetically obese mice.

Experiments were performed to determine PRL and GH concentrations in mice rendered obese by chemical means and to compare these concentrations with those of mice obese as a result of genetic mutation. Basal levels of serum PRL and GH were generally lower in gold thioglucose (GTG) and bipiperidyl mustard (BPM)-treated obese mice compared with lean controls. In the pituitary gland, the hormonal changes varied from lower or unchanged levels of PRL and GH shortly after drug injection to very high concentrations of PRL (but not of GH) a year later. However, when the mice were challenged with perphenazine, a drug that causes prompt release of PRL, GTG and BPM-obese mice released 2-5 times as much PRL as did lean controls, suggesting an impairment in the hypothalamic control of PRL secretion in GTG/BPM-obese mice. Basal levels of PRL and GH in genetically obese (ob/ob) mice of both sexes were also lower than those in their lean relatives (?/+). This was true for both serum and pituitary concentrations of the two hormones, the only exception being pituitary GH concentrations in females which were higher than or equal to those of controls. However, unlike GTG and BPM-obese mice, genetically obese mice released very little PRL compared with their lean relatives when stimulated with perphenazine, which suggested an insufficiency of pituitary function in ob/ob mice. The results demonstrate abnormalities in the secretion of PRL and GH in obese mice of both types.

Prolactin and growth hormone levels in different inbred strains of mice: patterns in association with estrous cycle, time of day, and perphenazine stimulation.

These experiments were designed to compare prolactin (PRL) and growth hormone (GH) secretion in strains of mice with varying incidences of mammary tumors. In addition to the basal levels, PRL concentrations were compared after stimulation with perphenazine. Although pituitary concentrations of PRL and pituitary serum concentrations of GH appeared to be generally higher in strains with high incidence of mammary tumors, basal PRL levels in single decapitate serum samples seemed to have little correlation with the incidence of mammary tumors in different strains. However, PRL concentrations in sera after perphenazine injection followed a pattern characteristic of the mammary tumor incidence rate of the strain: C3H/St and CBA/St--the two high-incidence strains--had lower levels of PRL; C57BL/St and BALB/cST--the two low-incidence strains--had higher levels; and DBA/2St--the medium-incidence strain--had an intermediate level. PRL contents of the pituitary glands were depleted virtually equally in all strains except the BALB/cSt. These results suggested that the rate of metabolism of PRL in strains with high incidence of mammary tumors may be faster than in those with low incidences. Perphenazine had no influence on GH secretion in most mice. The strain-specific differences in PRL and GH concentrations were usually present even during cyclical and diurnal fluctuations. Serum PRL and GH levels were generally higher during the follicular phase and lower during the luteal phase of the estrous cycle in both C3H/St and C57BL/St strains. There were signs of episodic secretion of PRL together with evidences of circadian periodicity in the secretion of PRL and GH in mice of both strains. The levels of GH were usually high during the morning hours of those of PRL high during the evening hours. The data show a high degree of specificity in the secretion of PRL and GH im mice of different inbred strains, and it is possible that these strain-specific differences may be an important factor in the development of mammary tumors.

Serum and pituitary concentrations of prolactin and growth hormone in mice during a twenty-four hour period.

The objective of this study was to investigate changes in the secretion of prolactin (PRL) and growth hormone (GH) occurring during a 24 h period in the mouse. Adult female mice of the C57BL/St strain and male mice of the C3H/St strain, maintained on a 14 h light and 10 h dark schedule, were used. Serum and pituitary concentrations of PRL and GH were measured by radioimmunoassay in samples collected by decapitation at hourly intervals through 24 h. Serum PRL concentrations in female mice averaged higher during the daylight hours and lower at night. However, the pattern was just the opposite in males: the values were lower during the day time and higher and variable during the night. Pituitary PRL levels dropped significantly after the onset of the dark phase in mice of both sexes. Serum GH concentrations of female mice did not fluctuate significantly with the time of the day, but those of male mice displayed a distinct flux: the levels were low from 0800 h until 1500 h, began to rise in the afternoon, and remained relatively high throughout the night. Pituitary levels of GH did not change with time in mice of either sex. The data suggest the existence of daily rhythms in the secretion of PRL and GH in mice, with marked differences related to sex. In general, the changes were most pronounced for serum PRL in females and for serum GH in males.

Glycosylated human prolactin.

A glycosylated form of human PRL (G-hPRL) was isolated from pituitary glands. The glycoprotein was separated from the major form of PRL on columns of lentil lectin-Sepharose 4B. The major form of PRL did not bind to the lentil lectin, whereas the glycosylated modification did and could be eluted with methyl-alpha-D-mannopyranoside. By gel electrophoresis in sodium dodecyl sulfate, a mol wt of 25,000 was estimated for the glycosylated PRL. The mol wt of hPRL is 23,000. In a RIA for hPRL, the glycosylated hormone was about one third as reactive as the principal form. Since there is only one Asn-X-Ser(Thr) sequence in hPRL, the asparagine at position 31 is the likely point of N-linked glycosylation.

Fluorescence polarization measurement of the hormone-binding site interaction.

Fluorescence polarization methodology has been applied to the binding of fluorescent-labeled prolactin, growth hormone and estradiol to subcellular fractions prepared from rabbit mammary and uterine tissue. Equilibrium measurements treated by Scatchard plots have shown that there are high affinity sites (K approximately 10(9) 1 mol-1), as well as lower affinity sites (K approximately 10(8) 1 mol-1) for both hormones. The binding of the fluorescent labeled hormone to microsomal or cytosol fractions has been shown to be inhibited by the prior addition of native, unlabeled hormone. Kinetic results on the interaction of prolactin with the microsomal fraction are consistent with a bimolecular reaction involving significant structural rearrangements during the reaction (not diffusion controlled). The forward rate constant calculated from data on initial rates was found to be 1.7 X 10(5) 1 mol-1 sec-1. Stopped flow kinetic measurements on the reaction between fluorescent-labeled estradiol and cytosol binding sites show that at low temperatures, the reaction goes in two distinct steps separable in time. The second step may be the reaction found by others (utilizing sedimentation velocity methods) which precedes translocation of the hormone-binding site complex to the nucleus. Fluorescence polarization makes it possible to observe both the formation and dissociation of hormone-binding site complexes over a time scale down to a fraction of a second and at concentrations down to the nanogram per nl range.

Binding characteristics of a biologically active variant of human growth hormone (20K) to growth hormone and lactogen receptors.

The dose-dependent displacement characteristics of a biologically active 20,000 molecular weight human pituitary growth hormone variant (20K) and of human growth hormone (hGH) were compared using hGH liver plasma membrane receptors both from 20-30 day pregnant rabbits and from normal female rats and also using mammary gland plasma membrane receptors from 5-6 day lactating rabbits. Different preparations of 20K were found to be only 3-20% as potent as hGH when compared at the dose necessary to cause 50% displacement of (125I)iodo-hGH from liver receptor and was 22-53% as effective as hGH in the mammary receptor assay system. These findings suggest that 20K and hGH may have separate receptors or that the binding characteristics of the two hormones may be quite different.

Comparative potency of subtilisin-cleaved and intact human growth hormone measured in growth hormone-deficient human subjects.

Eight GH-deficient subjects received both subtilisin-cleaved human GH (hGH-S) and intact hGH (hGH-I) during short term balance studies to compare the potency of these two forms of GH. Both forms caused nitrogen retention, calciuria, postassium retention, and elevation of blood glucose. The effects on plasma insulin concentrations were inconstant at the doses used. hGH-S was more potent than hGH-I, as measured by nitrogen and potassium retention, and the differences reached levels of statistical significance. hGH-S also caused greater calciuria and increases in fasting the postprandial blood glucose and in postprandial insulin in absolute terms, but these differences did not reach levels of statistical significance. In no instance was hGH-I significantly more potent than hGH-S. We conclude hGH-S, a two-chain form of hGH, caused significantly greater nitrogen and potassium retention in human subjects in short term balance studies than hGH-I.

A testosterone-producing adrenal cortical adenoma in an elderly woman.

A 76-year-old woman with virilization had menopausal levels of circulating LH and FSH, and a markedly elevated concentration of plasma testosterone (9130 pg/ml) into the range for adult men. Plasma cortisol and androstenedione levels andurinary 17-ketosteroid secretion were normal. Ethinyl estradiol suppressed plasma testosterone, LH, and FSH levels into the normal range for premenopausal women, but the testosterone concentration was unaffected by the administration of dexamethasone or ACTH. Retrograde venous sampling and angiography localized a right adrenal adenoma preoperatively. Following adrenalectomy, there was a prompt fall in testosterone, but there was no change in the LH concentration. Thus, this patient had an adrenal adenoma which secreted only testosterone and appeared to be gonadotropin-responsive. Testosterone levels in the adult male range failed to suppress gonadotropins. The significance of these findings is discussed.

Production of antibodies by inoculation into lymph nodes.

The intra-lymph node technique used to inoculate rabbits with small quantities of antigen has been described. A variety of antigens in the 20,000-22,000 molecular weight range, as well as a 15-amino acid peptide coupled to BSA, have been inoculated successfully by this procedure. We have made no attempt to compare the success rate of the intra-lymph node inoculation route with other techniques utilizing small (microgram) quantities of antigen.

Detection of multiple PRL- and GH-like proteins in human pituitary by Western blot analysis.

Surgically removed normal and tumorous pituitary tissues from a prolactinoma patient were analyzed by Western blot techniques for PRL and GH variants. Criteria for identification were the Rf of the bands within the gel, immunologic crossreactivity to specific antisera, and structural verification by tyrosine peptide-mapping of individual bands from the gel. The authors found the tumor-tissue to be characterized by the presence of a PRL band greater in concentration than in the normal tissue and the virtual absence of a GH band. Immunoblotting of the electrophoretically resolved proteins form both types of tissues revealed several new bands crossreactive with human PRL and GH antibodies. Some of the new bands were of Mr greater than the monomeric PRL and GH and others were of lower Mr. Relative of two of the low mobility Mr PRL-immunoreactive bands designated as 16K and 8K corresponded to the Rf of the two fragments of cleaved PRL, which suggested that cleaved PRL occurred naturally in the human pituitary gland. The most conspicuous of the new PRL-immunoreactive bands, a 25,000 Mr protein migrating slightly behind PRL, displayed strong crossreactivity to hPRL antibodies and was present in greater concentration in the prolactinoma tissue than in the normal tissue. These properties suggested that it was related to hPRL and perhaps represented its glycosylated variant. However, its tyrosine peptide map did not resemble that of hPRL. Thus, it is not clear whether it represented G-hPRL or a new pituitary protein that cross-reacts with hPRL antibodies. In addition, two other bands of low Mr, designated as unknown 1 and unknown 2, reacted with hPRL antibodies. Immunostaining with hGH antibodies revealed the 20K-hGH variant, the F1 fragment of cleaved hGH, and a pair of new bands immediately behind GH that could represent glycosylated hGH--possibly a product of Seeburg's variant hGH gene. Both PRL and GH antibodies elicited numerous bands of high Mr by the technique employed, far more than ever observed by Sephadex chromatography. The nature of the high Mr bands remains unknown. Further characterization of these new PRL- and GH-immunoreactive proteins might help in the understanding of the multiple physiologic functions of PRL and GH in man.