“w LARGE FORMS OF RAT PROLACTIN: HETEROGENEITY OF THE HORMONE SYNTHESIZED AND SECRET ED IN VITRO Thesis for the Degree of M, S. MICHTGAN STATE UNIVERSTTY OLAN CHARLES 'DOMBROSKE 1976 ABSTRACT LARGE FORMS OF RAT PROLACTIN: HETEROGENEITY OF THE HORMONE SYNTHESIZED AND SECRETED Tfl_VITRO By Olan Charles Dombroske Pituitaries of female rats secreted discrete, large forms of prolactin when cultured jn_vjtrg, Radioimmunoassay of the incubation medium fractionated by gel chromatography indicated a preponderance of monomeric prolactin; a dimeric and a larger polymeric component were Tess abundant. These components and an additional oligomeric form of prolactin Observed by their immunoreactivity were recovered from cultured adenohypophysial tissue following extraction by several procedures. The most abundant form of intrapituitary prolactin was dimeric. These polymeric forms were converted to monomers by the reduction of intermolecular disulfide bonds. When newly synthesized, intrapituitary rat prolactin was initially fractionated on dextran gel, then denatured in sodium dodecyl sulfate detergent with mercapto- ethanol, the large forms migrated as monomeric rat prolactin in polyacrylamide gel electrophoresis. Pulse-labeling Of the intra- pituitary prolacin with radioactive amino acids demonstrated that all forms were rapidly synthesized; over a three-hour interval, a shift from the large forms to the monomer was observed as the abundance Olan Charles Dombroske of all labeled components approached levels observed in total immuno- reactive, intrapituitary prolactin. The stability of the labeled dimeric component indicated that it may represent the storage form of rat prolactin, and/or serve as the intermediate of conversion of polymers to monomer. Since the turnover of total intrapituitary pro- lactin was greater than for any of the polymeric components, this was interpreted to mean that all forms were also secreted before con- version. This study demonstrated the usefulness of systems which separate proteins on the basis of their physical properties for the characterization of labile forms of pituitary hormones. LARGE FORMS OF RAT PROLACTIN: HETEROGENEITY OF THE HORMONE SYNTHESIZED AND SECRETED ;M_VITRO By Olan Charles Dombroske A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Physiology 1976 ACKNOWLEDGMENTS I thank Dr. William L. Frantz for his patient encouragement toward this study. Drs. J. R. Brunner, J. Meites and G. D. Riegle, members of the examination committee, and J. R. Hoffert kindly gave advice and criticism. Charles Brooks and Patricia Payne generously provided instruction in radioimmunoassay and organ culture methods, and self- lessly offered valuable support. I appreciate the assistance of Amylou Davis in the preparation of this manuscript. Financial aid was obtained from National Science Foundation Grant 68-30686 to William L. Frantz, and the Department of Physiology. ii TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES . INTRODUCTION . LITERATURE REVIEW I. II. III. IV. Pituitary Culture in Studies of Prolactin Synthesis Preparative and Analytical Methods for Character- ization of Polypeptides . . Gel Exclusion Chromatography . Polyacrylamide Gel Electrophoresis (PAGE). Assays of Prolactin . . Bioassay . Radioimmunoassay Radioreceptor Assay . Spectrometric Assay of Electrophoretically Separated Prolactin . . . . Multiple Hormonal Forms . . . Heterogeneity of Purified Pituitary Hormones . Clinical Importance of Heterogeneity Extraction of Adenohypophysial Hormones . Synthesis of Adenohypophysial Hormones jg_Vitro . MATERIALS AND METHODS . I. II. III. IV. VII. Animals . . Prolactin RadiOreceptor Assay . Lactoperoxidase Iodination of Prolactin . Preparation of Membgane Receptors for Prolactin . Binding Assay of I Prolactin . Determination of Specific Binding for Prolactin . Pituitary Culture in_VitrO . . . . . Culture Media . . . Procedure of Culture . . Processing of Medium and Tissue Samples Gel Exclusion Chromatography. Prolactin Radioimmunoassay Recovery and Measurement of Proteins Disc Electrophoresis in Polyacrylamide Gel iii Page vi -—l dmNVU'I-b-b N N EXPERIMENTAL RESULTS I. The Binding of 125Iodine-Prolactin to Plasma Membrane Receptors Results and Discussion . II. The Secretion of Prolactin in Vitro Results and Discussion . III. Prolactin Extracted from Cultured Adenohypophysial. Tissue . . . . . Results and Discussion . IV. Electrophoretic Heterogeneity of Immunoreactive Intrapituitary Prolactin . Results and Discussion . V. Heterogeneity of Newly Synthesized Rat Prolactin Results and Discussion . . VI. Determination of the Molecular Weight of Intra-o pituitary and Secreted Rat Prolactin Results and Discussion . DISCUSSION APPENDICES A. Linear Regression Analysis: Estimating Equation of Two Variable Linear Correlation and Confidence Interval B. Linear Regression Analysis: Correlation Coefficient for Two Variable Linear Correlations C. Stock Solutions for Disc Gel Electrophoresis D. Formulas for the Preparation of Electrophretic Gels E. Proteins Used for Molecular Weight Determinations . REFERENCES iv Page 38 86 87 88 89 90 91 Table LIST OF TABLES The Binding of Adjacent Fractions of Iodinated Ovine Prolactin from Sephadex Gel Column to Plasma Membrane Receptors The Abundance of Low Molecular Weight Prolactin: Immunoreactive Hormone with Kav Less or Greater than 0.70 Recovered from Media or Tissue Samples of Cultured Rat Adenohypophyses . The Distribution of Secreted Monomeric and Polymeric Rat Prolactins Separated by Gel Chromatography: Recovery of Immunoreactive Hormone from Incuba- tion Medium 199 . . . . . . . Relative Mobilities (Rf) of Rat Prolactin (PRL) and Somatotropin (GH) Components of Pituitary Extracts in Polyacrylamide Separating Gels . The Distribution of Monomeric and Polymeric Intrapitu- itary Rat Prolactins Separated by Gel Chromatography: Recovery of Untreated, Electrophoresed, Chemically Modified, and 14C-Labeled Hormone . . Protein Standards for the Determination of Molecular Weights: Experimentally Determined Kav in Gel Filtration and KR in SDS- Disc PAGE . The Molecular Weights (M) of Rat Prolactin Determined by KR in Gel Electrophoresis and Kav in Gel Filtration . . . . . . . Page 39 45 46 65 7O 73 74 LIST OF FIGURES An Elution Diagram of Immunoassayable Rat Prolactin Recovered from the Medium at Five Different Hours of Pituitary Culture . . . Parallelism of Displacement of Iodinated Rat Prolactin from Antibody by a Reference Preparation and Vari- ous Immunoreactive Components Recovered from Adenohypophysial Cultures . . . . Appearance of Multiple Forms of Immunoreactive Rat Pro- lactin Recovered by Homogenization, Sonication, and Electrophoresis (PAGE) of Pituitary Tissue, and for Comparison, Culture Medium (M199) . . . . . Electrophoretic Distribution of Immunoreactive Prolactin from Pituitary Extracts Gel Chromatographic Profiles of Immunoreactive Prolac- tin Eluted from Spacer Gels of Pituitary Extracts . The Distribution of Intrapituitary 14C-Proteins Labeled 12.!13I9, and Separated by Non-dissociating Electrophoresis . , , The Gel Filtration Profile of Intrapituitary 14C-Rat Prolactin Immediately After a Radioactive Pulse The Gel Filtration Profile of Intrapituitary 14C-Rat Prolactin Three Hours After a Radioactive Pulse The Determination of the Molecular Weight of Intra- pituitary, Synthesized Immunoreactive Rat Prolactin . . . . . . . . vi Page 43 47 51 55 57 6O 66 68 75 INTRODUCTION The reports of discrepancies among various assay methods for pituitary hormones have paralleTed recent characterization of the polymorphism of these polypeptides. Surveys of the heterogeneity of the hormones imply that differences in the molecular properties will account for the discrepancies among the assays, which test a limited response to hormones known to affect multitudinous biochemical and behavioral processes. Clinically, multiple forms of hormones are associated with certain pathologies or change in reproductive func- tion; resolution of these may prove valuable in diagnosis or manipu- lation of physiological states. The cultured mammalian pituitary provides a model for the control of synthesis and secretion of hormones, particularly for the production of prolactin which is enhanced by ablation Of the chroni- cally inhibitory influence of the hypothalamus. The purpose of this investigation was to establish the size heterogeneity of secreted and synthesized (intrapituitary) rat prolactin and to demonstrate the utility of gel chromatographic and gel electrophoretic methods to characterize these forms. The methods used will be applicable to studies of the production and transport of several polypeptide hor- mones and elucidation of the modification of target cell function. LITERATURE REVIEW 1. Pituitary Culture in Studies of Prolactin Synthesis When the mammalian adenohypophysis is removed from control of the hypothalamus by lesion, or section of the pituitary stalk (Harris, l955), prolactin continues to be secreted in larger amounts while release of gonadotropins, corticotropin, thyrotropin, and somatotropin is suppressed (Gale et_al,, l961; Harris, 1950; McCann and Friedman, T960). Ectopic autografts of rat pituitaries trans- planted below the kidney capsule secrete reduced quantities of other polypeptide hormones (Meites gt_al,, 1963), and on the basis of histological examination, function under circumstances not conducive to production of these other hormones, as evidenced by a preponder- ance of degranulated acidophils, chromophobes, and diminished baso- phils (Meites et_al,, 196l). Mammotropic activity, however, is maintained (Meites and Hopkins, T960). Similarly, adenohypophyses may be cultured in 11:39 as a means of selectively producing prolac- tin in large quantities (Nicoll and Meites, l962); the inhibitory influence of the hypothalamus was demonstrated by the culture of pituitaries with hypothalamic tissue or extract (Talwalker gt_a1,, l963). Thus, jg_yitrg_culture of pituitaries potentially provides a useful method for examination of prolactin metabolism with minimal interference from other hormones synthesized and released jg_yiyg (Deuben and Meites, l965). Meites gt_al. (l96l) have ascribed to jn_vitro cultures of rat anterior pituitary glands a daily net release of prolactin at least twice that extracted from uncultured glands. These cultures were gassed with a mixture of oxygen-carbon dioxide, while pitu- itaries exposed to ambient air produced much less hormone; pure oxygen was toxic (Gala and Reece, T963). Attempting to optimize cul- ture conditions, Gala (T970) showed that medium containing amino acids was significantly more suitable for prolactin production than physiological salt solution, generally reflecting nutritive require- ments of mammalian cells (Eagle, T959). Lack of essential amino acids caused a severe decrease in tissue viability. However, even without exogenous amino acids, net synthesis of rat prolactin (r-PRL) was observed over a three-day period, suggesting the existence of sufficient stores of protein synthetic materials and/or the storage of an intrapituitary precursor exhibiting no activity in bioassay until secretion. Tissue surface area aviTable for the diffusion of nutrients is an important factor in cellular viability and production of prolactin. Neither parameter, however, was improved when pitu- itaries were divided into more than quarters. Tissue viability does not always parallel prolactin production. Gassed cultures, although superior for hormone synthesis and tissue survival for several days, failed to produce prolactin after eight days, while cultures in air environment continued production for over one month (Gala and Kuo, 1972). II. Preparative and Analytical Methods for Characterization of Polypeptides Gel Exclusion Chromatography The first use of a cross-linked dextran gel as a method for size-dependent separation of proteins from electrolytes was reported by Porath and Flodin (l959). The character of the elution pattern, in which the rate of migration through the gel support was directly related to the mass of the species filtered, was attributedto the degree to which molecules can penetrate the gel phase; macromolecules enter the gel minimally, and migrate with little retention in the bed. This process, exclusion chromatography, describes the parti- tion behavior of molecules observed as the relative concentration of the support medium in the whole bed is changed (Pedersen, T962). Using fine glass beads as a model, effective fractionation of higher molecular weight solutes was noted for lower concentrations of the support; conversely, as the mass of the support is increased, the molecules must be smaller for separation. When the concentration of dextran in the packed bed is sufficient to preclude movement of a macromolecule through the spheres, it is excluded and elutes in the void vOlume. Andrews (1964: T965) has applied the permeation behavior of polypeptides to the estimation of molecular weights. For several gels of variable cross-linking, a large number of polypeptides con- form to a curvilinear plot of elution volumes against the logarithms of the molecular weights calculated by other methods. The high cor- relation of eTution volumes to molecular weights for carbohydrate-free proteins implies a common shape and density for most globular pro- teins in aqueous solution. Glycoproteins, because of the greater hydration of the carbohydrate residues, apparently possess expanded forms and exclude to a greater degree than globular proteins. The influence of molecular shape on gel permeation behavior has been dis- carded in favor of hydration effects, on the basis of exclusion of albumin polymers (Squire, T964). Polyacrylamide Gel Electro- phoresis (PAGE) Polyacrylamide is used as a support when effecting migration of a particle by an electromotive force which causes separation of proteins on the basis of charge and size (for review, see Chrambach and Rodbard, 1971). Resolution of proteins in different size ranges can be accomplished by varying the total acrylamide concentration and/or the percentage of cross-linking agent. Similarly, separation on the basis of charge can be controlled by choice of pH-buffer sys- tems. The behavior of a polypeptide on the support is commonly described by its relative electrophoretic mobility (Rf) defined as migration of the protein relative to a marker at the moving boundary front of the separation phase (Rodbard and Chrambach, T970). Deter- mination of Rf in several gel concentrations allows a linear plot of log Rf versus I_(percent acrylamide concentration). The slope of this Ferguson plot is defined as the regardation coefficient (KR), and is an estimate of molecular weight (Ferguson, T964; Hedrick and Smith, T968). The intercept resulting from extrapolation of Rf to zero I, designated Yo, yields the electrophoretic free mobility when its antilog is calculated. Since,in the absence of a filtration matrix, migration is dependent only on charge, Yo is a measure of net charge on the molecule. The square root of KR is linearly related to the radius of globular protein. Estimates of molecular weights at a single gel concentration are possible when proteins are denatured in sodium dodecyl sulfate (505). Since the free mobilities of such proteins are nearly identical (Neville, l97l; Maizel, T970), sieving alone determines migration. This was first shown by measure- ment of protein molecular weight from Rf (Shapiro _t_gl,, 1967). High reproducibility of molecular weight determination of SDS-bound protein at a given gel concentration has been confirmed and applied to several proteins (Weber and Osborn, T969). Polyacrylamide gel electrophoresis (PAGE) of proteins in a discontinuous system of anionic buffers (Davis, T964) produces excel- lent resolution due to the concentration of proteins into "discs" between leading and trailing ions of the buffers. As the thin zone of proteins reaches the sieving gel support and region of higher pH, enhanced mobility of the trailing ion effects its migration ahead of the proteins, imposing a voltage gradient down which the proteins move (Ornstein, T964). Considering Ornstein's arguments, Williams and Reisfeld (T964) have summarized the reduirements for disc PAGE at different pH's to optimize charge separation. Disc PAGE of SDS- denatured proteins also improves separation. Laemmli (T970) used discontinuous anionic buffers with $03 to resolve phage proteins into sharp, well defined bands. In an SDS-disc system, the dependence of the logarithm of molecular weight on Rf is a sigmoid function; therefore, the constant Y0 and the KR (as an estimate of molecular weight) can be substituted into the Ferguson equation (T964) to cal- culate the size of an unknown protein from Rf at any given I (Neville, T97l). For metabolic studies in which proteins are isotopically labeled, liquid scintillation spectrometric methods can be used if macromolecules are extracted from the polyacrylamide bed (Young and Fulhorst, l965). Gentle heating in the presence of hydrogen peroxide has dissolved gels with recoveries of radioactivity exceeding ninety percent, indicating minimal loss in the form of water or carbon dioxide (Tishler and Epstein, T968). III. Assays of Prolactin Bioassay A number of la vitro and jn_vivo assays for prolactin have been developed based on the lactogenic activity in serum samples, pituitary extracts or the culture media of pituitaries. In general, these are often variable or insensitive to fluctuations inherent in various physiopathologies (cf. Forsyth and Parke, T973). Attempts to improve precision have included measurements of mammary gland or pigeon crop-sac growth and differentiation. Nicoll (l969) has refined the assay by weighing the dry mucosal epithelium, and Ben- David (T967) has analyzed the increased mitotic activity in crop-sac by uptake of 3H-thymidine; Frantz g g. (1972) have subjectively rated the stained secretion products in alveolar lumina of cultured 32 mouse mammaries, and Turkington (T972) has compared P-casein production and Néacetyl-lactosamine synthetase induction. These represent various histological and biochemical endpoints used to assess_in vivo and jg_vitro responses to prolactin, respectively. Radioimmunoassay_(RIA) Niswender et_a1, (l969) have described an assay for rat pro- lactin (r-PRL) based on the competitive binding of a radioiodinated, purified hormone from frozen rat pituitaries to rabbit antiserum obtained by injecting the unlabeled preparation. A second antibody (Ab), sheep antirabbit gamma globulin, is used to precipitate the complexed antigen (Ag), r-PRL, and rabbit Ab to r-PRL. By assess- ment of isotopically labeled Ag precipitated, one can infer the quantity of material which contains an identical antigenic site. Thus, a sample containing the greater quantity of r-PRL is charac- terized by the precipitation of less radioligand; a competitive inhibition curve is constructed by plotting the proportion of label precipitated as a function of the antigenic activity added in a known reference standard. Critical assumptions include (l) equivalent affinity of Ab for labeled and unlabeled ligand, and (2) lack of non-specific bind- ing of Ab to other proteins. Comparative studies of methods used to covalently attach radioiodide to prolactin have indicated that products of chloramine T and lactoperoxidase iodination are indis- tinguishably inhibited in their binding to Ab in the presence of low levels of competitive hormone, but show divergent inhibition curves when increasing quantities of unlabeled prolactin are added. Discrimination of antigenic concentration over an extended run was poor for the chloramine T-oxidized hormone, while the lactoperoxidase- labeled product afforded excellent sensitivity (Rogol and Chrambach, T975; Rogol and Rosen, l974a). Therefore, it is apparent that cau— tion should be exercised in reading sample antigenic concentrations at extremes of the inhibition curve; where damage to the radioligand is possible, anomalous interaction of the molecule with Ab should be suspected. Corroboration of molecular integrity should be accomplished by physical methods. These authors noted that the enzymatic method of iodination minimally affects the charge properties (Yo) and par- titioning (KR, Kav) of human pituitary and amniotic fluid prolactin. The difficulties generated by non-specific interaction between Ab and serum or organ-extracted solutes are illustrated by disproportionate increases in concentrations of Ab required to quantitatively precipi- tate the hormone. An early study examining jn_vjtrg_synthesis of rat somatotropin by immunoprecipitation of the tritiated Ag indicated a requirement for separate adsorption of non-hormonal radioactivity with another gamma globulin and a second Ab before addition of the primary antiserum (Burek and Frohman, l970). Generally, the region of the sigmoid inhibition curve where 20-80% of the tracer is bound to Ab proves useful for assessing antigenic activity. Rodbard et_gl, (T968) described an empirically satisfactory method for linearization of the function by logit transformation, simplifying reading of the curve beyond the linear mid-portion. Implicit in the estimation of hormonal activity in samples is the univalent interaction between Ag and Ab, without cooperative effects. Inference of activity TO involves conversion of ng/ml to biological potency based on a highly purified protein; validation of the RIA therefore requires correlation of antigenic determinant to bioactivity in various pathological and physiological states. This factor has been of importance in appraisal of the NIAMD RIA for r-PRL by Neill and Reichert (T97l). Claiming that many assays overestimated the levels of r-PRL in serum, the investigators used preparations of pituitary r-PRL from several sources to demonstrate agreement of their assay with that developed for NIAMD in evaluation of crop-sac stimulatory activity, reporting biologic to immunologic rations of 0.6 to l.3. The severalfold dis- crepancies among reported absolute estimations of serum prolactin were attributed to differences in purity of reference standards and methods of blood collection, suggesting the elevation of serum pro- lactin by ether anesthesia. A third immunologic technique utilizes the inactivation of complement by Ag-Ab complexes evidenced by a reduction of hemolytic activity (mediated by hemolysins coating red blood cells) and has been used to discriminate determinants of clinical grade human somatotropin separated by ion-exchange chroma- tography and characterized by differential prolactin activities (Tashjian §t_gl,, l965). A radioimmunoassay suitable for tracing estrous cycle fluc- tuations in serum prolactin has been developed with antisera to r-PRL isolated from secretion granules of pituitary tumors (Kwa gt_gl,, T969). Parallel inhibition curves were generated by various combina- tions of tumor prolactin, its antisera and r-PRL extracted from frozen normal pituitaries and its antisera. Molecular weights of 11 normal and tumor r-PRL were identical, but tumor r-PRL activity in mouse mammary bioassay jn_yjt§9_was only l0-T2% of its crop-sac activity. It is interesting to note that N-terminal amino acid analysis showed gross distributional differences between normal and tumor r-PRL preparations. These observations, in addition to the report of a cross—reactive material in serum of hypophysectomized rats (Neill and Reichert, l97l), indicate that parallelism of dose or pathology does not guarantee identity of antigenic determinant among assays. Hence, extrapolation of prolactin antigenic activity to biological significance cannot be assumed. Radioreceptor Assay Like the RIA, the radioreceptor assay (RRA) evaluates spe- cific binding of a radioactive tracer in competition with the unla- beled hormone. The binding interaction for a number of polypeptide hormones occurs at a cell-surface site; penetration of the plasma membrane is not necessary. Prolactin-mediated differentiation of the mammary gland (Turkington, T970), insulin stimulation of fat-cell utilization of glucose (Cuatrecasas, T969), incorporation of glucose into liver glycogen (Blatt and Kim, 197T), or amino acid uptake in mammary gland (0ka and Topper, l97l) occurs with the hormone covalently bound to Sepharose. It is possible to prepare iodinated hormones with preserved biological activity (Frantz and Turkington, T972; Goldfine gt_al,, T974; Rae and Schimmer, l974). Synthesis of casein jn_vitro has been correlated with specific binding of prolac- tin (Frantz et_al,, T974), and a time study of insulin receptor 12 binding to fat-cells has discriminated rates of lipogenesis propor- tional to low-level receptor occupancy and steady-state maintenance of the response at high concentrations of bound hormone (Gliemann gt_gl,, l975). The prolactin receptor assays are highly specific, with sig- nificant inhibition of binding effected only by prolactins of other species, or somatotropins and placental hormones with lactogenic activity (Shiu gt_al,, T973). Differential binding of prolactin and somatotropin parallel the hormonal responsiveness of these tissues determined by other methods (Kelly gt_gl,, T974) although high bind- ing of prolactin in adrenals of pregnant rabbits was not anticipated (cf. Friesen et_al,, T973). Although the number of tissue binding sites may change severalfold with the reproductive state, the affinity of the receptor is unchanged (Kelly gt_al,, T975). Estrogen induced prolactin receptor sites in livers in male rats, and in female rats ovariectomy reduced binding. Abundant, high affinity receptor sites for prolactin have been demonstrated in mammary car- cinomas which have little estrogen receptor and are not dependent on prolactin for growth (Costlow et_gl,, 1974). In contrast, another author attributed limited milk protein induction by prolactin in these tumors to fewer binding sites as compared to 7,l2-dimethyl- benzanthracene (DMBA)-induced, estrogen- and prolactin-dependent mammary tumors (Turkington, l974). Prolactin-specific sites were shown in cultured cells derived from normal rat pituitary tissue, and from estrogen-induced, prolactin-secreting pituitary tumors (Frantz gt_al,, T975). 13 The RRA, with excellent correlation to serum prolactin detectable by RIA, has also been used to monitor the separation of several highly active forms of rat placental lactogen by preparative and analytical isoelectric focusing (Robertson and Friesen, T975). Spectrometric Assay of Electropho- retically Separated Prolactin The standard disc electrophoretic technique used by Lewis and Clark (T963) to separate rat pituitary extracts qualitatively compares effects of age, sex, and sex steroid replacement therapy (Jones gt g1,, T965) on stained protein bands which were identified as prolactin and somatotropin by crop-sac and tibial cartilage assays, respectively. Mature female rats had more intrapituitary prolactin and less somatotropin than mature males; castration of males followed by the administration of estradiol produced relative staining intensities characteristic of females. Conversely, the female pattern was lost following ovariectomy (with a resultant male pattern), but was restored by estradiol injections. Patterns for the immature sexes were visually indistinguishable. Kragt and Meites (T966) used non—dissociating PAGE for single step separation of hor- mones in pituitary extracts and recovery of r-PRL prior to crop-sac assay. Cross-contaimination of hormone-containing gel segments was small, although l5-30% of crop-sac stimulatory activity associated with the albumin band. Good correlation between biological potency and the intensity of stain for prolactin and somatotropin bands has been reported. Optical density of alkalai-solubilized gel segments was a linear 14 function of applied protein for purified prolactins and somatotropins as well as pituitary extracts from a number of species (Lewis gt_gl,, l969); slopes related to relative biopotencies of the preparations. Concentrations of prolactin in male, female, and ovariectomized rats determined by crop-sac assay agreed with the colorimetric method (band peak absorbance) (Nicoll gt_al,, T969). The differential ratios of optical density : biopotency establish the need for a standard curve for each protein which is adjusted to different affinities of proteins for stain. These authors also noted a high degree of correlation between biopotency and stain intensity of material obtained from pituitary extraction or organ culture. However, the densitometric reading overestimated crop-sac activity, which they attributed in part to use of ovine prolactin (o-PRL) as the bioassay standard. Inhibitory effects of median eminence on j__vitro secretion of r-PRL, and the reduction of intrapituitary r-PRL and somatotropin content in propylthiouracil-induced hypothyroidism were also confirmed. Using the O-PRL standard, mean adult levels of r-PRL corresponded to l80 mIU/mg for pituitaries of females and 60 mIU/mg tissues in males. Administration of estradiol to males increased tissue activities to levels found in females. Kurcz gt_al, (l969) have assessed bioactive r-PRL recovered from gels. Staining intensity of the major r-PRL band was linearly related to its crop—sac activity, but over 20% of the hormone was distributed in the spacer gel and in an area behind the somatotropin band in the separating gel. 15 IV. Multiple Hormonal Forms Heterogeneity of Purified Pituitary_Hormones The characterization of heterogeneity observed for proteins is facilitated by manipulations of isolated material, where experi- mental results are likely to be restricted by conditions defined in the treatment; influences of contaminants are minimized by the use of purified preparations. Squire §t_gl, (1963) used sedimentation velocity and gel permeation to demonstrate pH-dependent, reversible associations of ovine prolactin (o-PRL) into polymers; these were not in equilibrium with the monomeric form (Sluyser and Li, 1964). By ion-exchange chromatography and PAGE, o-PRL was resolved into three acidic components whose lactogenic activities decreased in order of mobility (Reisfeld gt_al,, 1964). More highly acidic com- ponents of o-PRL can be produced by trace proteolytic activity in purified preparations or by deamidation in an alkaline solution (Lewis and Cheever, T965). Diminished hydrogen bonding by urea enhanced the conversion rate, presumably due to unfolding of the molecule, exposing glutamine and asparagine residues normally buried within the tertiary structure of the protein. Cheever and Lewis (1969) determined the molecular weights of polydisperse forms of purified mammalian prolactins and somatotropins electrophoretically, proving the necessity of separation in several gel concentrations for identification of oligomeric hormones. The accuracy of the method was supported by estimations (KR) of molecular weights which generally agreed well with results obtained by ultracentrifugation, 16 and confirmed that deamidated forms differ in charge, not size. More recently, Singh et_al, (1974) have separated two contaminants from purified human somatotropin (h-GH) preparations with SDS-PAGE. A form of h-GH with a single cleavage in the peptide chain, and a protein of molecular weight ten percent less than that of h-GH were isolated. Both proteins exhibited crop-sac stimulatory activity severalfold greater than that of h-GH, but did not cross-react in RIA for prolactin. Several forms of human prolactin have been iden- tified in amniotic fluid (Ben-David and Chrambach, T974), plasma and the pituitary (Rogol gt_§1,, 1975). Most were charge isomers, but two were characterized by charge and size two and one-half times greater than the majority of isohormones. Clinical Importance of Heterogeneity A combination of methods was used to obtain evidences of a human prolactin (h-PRL) separate from h-GH. Chrambach et_al, (T971) electrophoretically fractionated the serum of a lactating woman and a patient with chromophobe adenoma. By the simultaneous considera- tion of KR and Y0 for immunoreactive h-GH and h-PRL active in a casein bioassay, the two hormones were shown as discrete molecules of similar size. Lewis et_al, (197T) extracted from human pituitaries a lactogen of similar KR but possessing a charge different from h-GH in PAGE. A limited qualitative survey indicated larger quantities of the presumed pituitary prolactin in pregnancy, and resulting from estrogen therapy, with lesser amounts in males. By gel chromatogra- phy, at least two circulating forms of immunoreactive prolactin have 17 been observed in women at various reproductive states (Rogol and Rosen, l974b). The normally small proportion of dimer and/or trimer, and a larger h-PRL may predominate in sera and pituitaries of acromegalics with adenoma. The oligomers have reduced activity in RRA and the largest prolactin exhibits no binding in the RRA (Aubert gt_gl,, T975). Prolactin obtained from culture jn_vitro tended toward a smaller proportion of large hormone (~56,000 mol. wt.) stored and released by the pituitary than seen in circulation. The greatest quantities of large h-PRL appeared in pregnant females. No conversion of large to smaller form was observed after incubation with plasma (Suh and Frantz, l974). Incubation of human pituitaries in vitro produced release of less large h-PRL than observed in the serum. Activities of the monomer and oligomer of prolactin were identical in RIA and RRA (Guyda, 1975), although a large h-GH exhibited reduced receptor actiVity. Heterogeneity of glycoprotein hormones has also been observed. Immunoreactive thyrotropin in sera of humans suffering from primary hypothyroidism is composed of a form seen mainly in purified pitu- itary preparations and another major component showing a significantly decreased Kav in gel chromatography and smaller net charge in anion- exchange separation (Dimond and Rosen, l974). Orchidectomy (Peckham and Knobil, 1976a) and ovariectomy (Peckham and Knobil, l976b) pro- duced gonadotropins of male and female monkeys which apparently con- tained more sialic acid. Estrogen therapy reversed ovariectomy by restoring the molecules of the smaller form. 18 Extraction of Adenohypo- physial Hormones The applicability of physical methods in the resolution of multiple hormonal forms may be limited by the fragility of polypep- tides. Even purified preparations of pituitary prolactin and somatotropin contained a diisopropylfluorophosphate-inhibitable proteinase which degraded the hormones (Lewis, 1962). In alkaline pituitary extracts, Ellis §t_a1, (1968) have identified trypsin-like activity capable of limited cleavage of r-GH to fully bioactive sub- units, and have suggested responsibility of this contamination for low molecular weight determination by ultracentrifugation. Thus, Hummel gt_§l, (1975) used both acidic and alkaline pituitary extracts to identify several forms of monkey prolactin and somatotropin by isoelectric focusing. A superior preparation of proteinase-free r-PRL was susceptible to plasmin degradation (Ellis gt_al,, 1969), empha- sizing the potential loss or modification of these hormones during extraction. A possible corticotropin precursor possessing covalent pep- tide bonds was recovered by gel filtration of rat pituitary homogen- ates in strongly acidic solutions and urea. The large and small forms exhibited parallelism in RIA and in stimulation of adrenal cells (Lang gt_al,, T973). Rohde and DOrner (1969) analyzed protein- staining patterns of immunoreactive h-GH, comparing the consequences of extraction at various pH values and ionic strengths; one component was common to all conditions. A large form of intrapituitary h-GH was partially converted to monomer by mercaptoethanol reduction of 19 disulfide bonds: substitution of hydrogen bonds with urea alone did not affect the large immunoreactive form, but urea and mercapto- ethanol together effected complete conversion without peptide cleavage (Benveniste et_al,, 1975). More stringent procedures are required for solubilization of r-PRL than r-GH. By estimation of stained protein separated in non- dissociating PAGE, homogenization in alkaline buffer or urea treat- ment ruptured prolactin granules, yielding at least twice as much hormone as obtained in water or neutral solutions, but affected somatotropin extraction less dramatically (Zanini and Giannattasio, T972). Samli gt a1, (1972) have shown that (ethylenediaminetetra- acetic acid) is superior to normal saline or Krebs-Ringer bicarbonate for recovery of 3H-r-PRL synthesized in 11519. Enriched prolactin and somatotropin granules obtained by differential and sucrose gradient centrifugation, followed by incubation with puromycin at high ionic strength yielded intact granules of r-PRL, but solubilized r-GH (Zanini and Giannattasio, 1973). The detergents sodium deoxy- cholate or Lubrol PX stripped the membranes of prolactin granules but did not solubilize the hormone, forming membrane-free secretory granules (Giannattasio gt_al,, 1975). Neither urea nor the combina- tion of urea-deoxycholate matched SDS for solubilization of total pituitary r-PRL (Zanini gt_al,, l974a). Synthesis of AdenohypOphysial Hormones_in Vitro The resolution obtained by electrophoresis in polyacrylamide gel and gel chromatography, often in combination with hormone assay 20 techniques, have been adapted to biosynthetic studies of pituitaries. The production of r-PRL and r-GH were examined by continuous incuba- tion of anterior pituitary lobes with radioactive amino acids; dis- parities of charge between those hormones afford excellent separation for subsequent evaluation of samples by liquid scintillation spec- trometry. Rates of synthesis were similar for both hormones, but r-PRL was turned over into the medium far more rapidly than r-GH (MacLeod and Abad, T968). Adenohypophyses from rats bearing prolactin- or somatotropin-secreting tumors were characterized by severely decreased production of that hormone jg_yjtrg_when compared to pituitaries from intact rats. The reduction of r-PRL synthesis jg_vitro resulting from i vivo effects of ectopically grafted prolactin-secreting tumors was reversed, and synthesis by glands of intact rats was enhanced, by administration of estrogen and testos- terone (MacLeod gt al., 1969). Synthesis of r-PRL is similar for both sexes until puberty, then it more than doubles in females, and continues to increase thereafter (Yamamoto §t_gl,, l970). MacLeod and Lehmeyer (1974a) demonstrated that the sum of prolactin and somatotropin production is a constant, and an increase in one will result in a decrease of the other in a reciprocal manner. The rates of total protein, r-GH, and r-PRL production are greater in dissociated pituitary cells than in intact tissue blocks, providing an excellent model for hormone synthesis at the cellular level (Hopkins and Farquhar, l973). Pulse-labeling experiments with evaluation by PAGE to accurately compare the synthesis and secretion of r-GH and r-PRL 21 indicated that prolactin is synthesized within minutes, and 70% is released into medium in five hours, while less than one-third of the somatotropin is turned over in that period (Meldolesi gt a1., 1972). Swearingen (1971) determined the specific activity of r-PRL by den- sitometric quantitation of the stained protein bands. Pituitaries were exposed in vivo or in vitro to radioactive amino acids, then cultured in vitro for evaluation of prolactin production. Specific activity of the secreted r-PRL was greater than that of the hormone in tissues; these results indicated that newly synthesized hormone is secreted from at least two pools. At intervals throughout the incubation, discrepancies between tissue loss and medium gain of labeled r-PRL suggested the presence of a form in tissue not detected by PAGE. During the extended incubation of rat pituitaries, Gala and Kuo (1972) noted the consistent underestimation of secreted prolactin by RIA, and overestimation by electrophoresis-densitometry, when com- pared to crop-sac bioassay. A significant increase in prolactin measured by RIA and densitometry from mid-point to termination of culture was not reflected by the bioassay. Electrophoresis in the presence of SDS detergent has been used to separate labeled, immunoprecipitated hormones, by localizing the hormone in a narrow band. Dannies and Tashjian (T973), studying the synthesis of r-PRL in response to thyrotropin releasing 'factor and and hydrocortisone in an established line of tumor cells, determined that the hormone is produced from two intracellular pools of amino acids, but found no evidence for stable prohormones of r-PRL. By the same technique, cell-free, isolated membrane-bound polyribosomes 22 were confirmed as the site of prolactin assembly in tumor cells (Biswas and Tashjian, l974). Radiolabeled r-PRL and r-GH were first sequentially immunoprecipitated and electrophoresed for comparison of the syntheses of the hormones in several tumor cell lines by Gautvik and Kriz (T976). Sussman gt_a1, (T976) isolated messenger ribonucleic acid (mRNA) from rat pituitary tumor cells and ascertained the syn- thesis of an immunoprecipitable precursor of r-GH which was converted to the predominant form by peptide cleavage. Similarly, a precursor of r-PRL, 25% larger than prolactin, was synthesized in a cell-free system by the mRNA from pituitary tumor cell lines (Evans and Rosen- feld, T976) and whole pituitary glands or tumors (Maurer gt_al,. 1976). Complete separation of r-GH from r-PRL in SDS-disc PAGE has facilitated the evaluation of both hormones without immunoprecipita- tion and is more effective for solubilization/separation of the hormone associated with rough microsomes than are the non-dissociating systems. Covalently bonded precursors of both hormones could not be detected under the dissociating influence of SDS binding (cf. Maizel, 1970) and reduction by mercaptoethanol in pulse-labeling experiments (Zanini §t_gl,, l974b). Gel chromatography of the intra- pituitary or secreted hormones followed by immunoprecipitation has been used to separate polyforms under minimally dissociating condi- tions. Stachura and Frohman (1974) have demonstrated differential release of r-GH jn_yjtrg, During the course of incubation, radio- activity associated with large immunoreactive forms diminished, paralleling increased activity in monomeric r-GH which is presumably 23 the major form secreted. Sequential and dual isotopic labeling with serine and leucine yielded ratios of the two amino acids in somato- tropin which rapidly changed to stable values for large and small forms; this suggests two pathways for large r-GH, either storage followed by conversion to monomer, or direct release. Furthermore, much of the large hormone was monomeric r-GH still associated with messenger and ribosomal RNA. A pool of labeled r-GH was established by the continuous perifusion of pituitaries jg_vitro with medium con- taining isotopic carbon precursor. Then the cultures were perfused with tritiated label. Immunoprecipitates of these fractions indi- cated that monomeric r-GH is Stored; the molecules released early are apparently dimers, and as the perfusion proceeds, secreted r-GH is mostly monomeric, with a reduced percentage recovered as dimers (Stachura and Frohman, 1975). In the light of evidence for poly- morphism of pituitary hormones and the discrepancies among various assays for the detection and the measurement of prolactin, it is apparent that methods for the separation Of polypeptides based on physical properties can be adapted to the resolution of these dis- parities. MATERIALS AND METHODS I. Animals Lactating female Sprague Dawley rats and lactating Swiss- Wester mice were used in radioreceptor assays. For culture studies, multiparous female Sprague Dawley rats were used. The animals were Obtained from Spartan Research Animals (Haslett, MI). Animals were caged and maintained at room temperature on a diet of Purina rat chow (Ralston Purina Co., St. Louis, MO) and tap water ag_libitum. II. Prolactin Radioreceptor Assay Lactoperoxidase Iodination of Prolactin The method used was a modification of Frantz and Turkington 125I-prolactin with (1972) for the preparation of biologically active high specific activity, but omitting further purification to facili- tate rapid assay (Sakai et_gl,, T975). Into a polyethylene vial containing TOO ul of 40 mM sodium diethyT-barbiturate buffer (pH 7.0; Mallinckrodt Chemical Works) and 0.5 mCi Na 1251 (IMS.30; Amersham- Searle, Chicago, IL) ovine prolactin (NIH-PS-ll, 25.6 IU/mg; 5 ug in 25 ul distilled H20), lactoperoxidase (bovine milk, Sigma Chemical Co., or B grade, Calbiochem, LaJolla, CA; 10 ug in 25 ul of barbital buffer), hydrogen peroxide (250 ng in 25 u1 of 30% reagent diluted ,l:30,000 with H20) were rapidly injected. The reaction was allowed to proceed at room temperature either 5 or 10 minutes prior to 24 25 separation; in the latter case, a second 250 ng aliquot of peroxide was added at the end of the first 5 min. At term, 200 pl of cold 16% sucrose were added, and the entire volume of approximately 0.5 ml was chromatographed on a 0.9 x 14 cm column of 6-75 Sephadex (Pharamcia) at 4°C. The column had been pre-washed with 20 mg of bovine serum albumin (BSA, fraction V, Sigma) to reduce protein adsorption to the filtration matrix and washed by 20 ml of buffer. The column was eluted with 40 mM sodium barbital buffer pH 7.0; col- lection of 1.0 ml fractions was initiated as the sample was applied to the column. The polystyrene collection tubes were also coated with BSA to reduce the loss of labeled protein during cold storage. Radioactivity of the iodinated protein peak was determined by count- ing 1 ul aliquots of the fractions in a gamma spectrometer. Labeled prolactin capable of binding to cell surfaces was found in all portions of the protein peak, usually 2-3 fractions. By the deter- mination of TCA-insoluble precipitates formed, specific activity of iodinated o-PRL was 30-60 uCi/ug. Preparation of Membrane Receptors for Prolactin Plasma membranes were enriched by high-speed centrifugation on sucrose-density gradients (Neville, T968). Livers from lactating female Sprague-Dawley rats or mammary glands from Swiss-Webster mice at day 10-16 of lactation and 1-2 days removed from pups were quickly excised after cervical dislocation and rinsed hiice-cold 4% (w/v) sodium citrate buffer, pH 7.4. The following procedures were at 0-5°C. A 10-20% (w/v) homogenate in citrate buffer was made by 26 crushing l-2 g of tissue in a 10 ml Kontes glasszglass hand homoge- nizer, avoiding up and down strokes to minimize protein denaturation by foaming and generation of excessive heat in the suspension. These homogenates were filtered through two layers of coarse gauze, and 2.0 ml were layered above a discontinuous gradient of 1.0 M and 1.411 sucrose (1.5 ml of each) poured in 5 ml cellulose nitrate tubes. 5xg (avg) in a Spinco ,These were centrifuged 60 min at 4°C and 10 SW 39L rotor (Beckman). Membrane-rich fractions which concentrated at the citrate:l.0 M and l.0:l.4 M sucrose interfaces were combined for the assay; the lipid layer and pellet (nuclear debris, blood cells) were discarded. The fractions were sometimes frozen at -20°C for up to 5 d prior to use. 125I-Prolactin Binding Assay of Membrane particles corresponding to 40-100 ug Protein in a volume of 100-200 ul was aliquoted into 200 pl of an incubation solu- tion containing 2 mg BSA, 0.5 umol NaCl, 5 umol Tris-HCl, pH 7.4 125 (4°C) and 70,000-100,000 dpm I-o-PRL (T-5 ng). Tubes also had 20 ug of either o-PRL or o-TSH to determine respectively the dis- 125I-o-PRL by receptor-competitive hormone. or the "00' placement of specific binding (failure to inhibit 125I-o-PRL binding) of other polypeptide hormones. Multiple samples were incubated at 0-5°C in 400 pl polyethylene Beckman Microfuge tubes and at the end of 30-45 min, centrifuged 104xg for 5 min. The supernatant was aspirated, and the top of the pellet was washed by the addition and immediate aspiration of cold citrate buffer. The tip of the soft polyethylene 27 tube containing packed membrane particles was carefully cut and placed in polystyrene gamma-counting vials (Nuclear-Chicago) and the protein dissolved by dispersion in basic Lowry C (Lowry gt_al,, 1951), then quantitated by the colorimetric reaction. Determination of Specific Binding for Prolactin Binding of the radioligand was standardized as follows: background-corrected dpm/100 pg of membrane protein for samples in the group containing TSH (A) and, similarly, the group containing PRL (B) were calculated. Tracer bound in the presence of TSH (a pituitary polypeptide with a molecular weight similar to that of pro- lactin, but possessing no lactogenic activity) is taken as the measure 125I-o-PRL (A). That which is not displaced by an of total bound excess of PRL is defined as non-specific binding (Frantz gt_al,, 1974). The specific binding is expressed by a ratio of (displaced activity)/(total bound) which is calculated from (A-B)/A. III. Pituitary Culture in Vitro Culture Media Tissue culture Medium 199 (M199) was obtained from Difco Laboratories (Detroit, product code 5701). The dry powder was rehy- drated by dissolving 11 g in 990 ml of triple distilled, autoclaved H20 and adjusted to pH 7.2 with 3.5 ml 10% sodium bicarbonate. Ten ml of an antibiotic-antimycotic lOOx-concentrated solution (Grand Island Biological Co., Grand Island, NY) containing 10,000 U peni- cillin, 10,000 ug streptomycin and 25 ug fungizone (E. R. Squibb and 28 Sons) per ml were also added. The solution was sterilized by vacuum filtration through a 0.45 u pore filter (Gelman Instrument Co., Ann Arbor, MI) fitted to a sterile Pyrex Millipore filtering appa- ratus (Millipore Corp., Bedford, MA). Earle tissue culture solution (code 5772) was obtained from Difco as a sterile, ten-fold concentrate in 100 m1 bottles. The isotonic balanced salt solution was prepared by the aseptic addition of 50 ml of Earle Solution (10x) to 439 ml of sterile water followed by the adjustment of pH to 7.2 with 11 ml of 10% sodium bicarbonate. A mixture of uniformly labeled 14C-amino acids (NEC-445) was obtained from New England Nuclear (Boston, MA). This typical algal protein hydrolysate in 0.1 N HCT has a specific activity of 0.1 mCi/m1. These solutions were stable for at least 2 months when stored under refrigeration. Procedure of Culture All instruments and containers were autoclaved or sterile- prepackaged. However, because of the short period of culture, it was not necessary to maintain sterile conditions. Nine to 12 rats were decapitated within 30 seconds after initial handling. Heads were sprayed with 70% ethanol to prevent contamination from hair, and the skin was split with a single mid- dorsal cut extending anteriorly to the frontal bones. Antero- lateral incisions were made in the skull from the foramen magnum to each orbit. The resulting bone flap, and brain hemispheres were 29 retracted. The neurohypophysis was teased away and discarded; the intact anterior pituitary gland was removed and placed into a dis- posable sterile 60 x 15 mm petri dish (Lux Scientific Corp., Thousand Oaks, CA) containing 2 ml M199 to keep the tissue moist and to rinse away residual blood. Several adenohypophyses were trans- ferred to another petri dish and cut into eighths or 1 mm3 tissue blocks. Pituitary fragments were randomized, and tissue equivalent to three rat pituitaries was placed into a siliconized (Siliclad, Clay Adams, Parsippany, NJ) 25 ml Erlenmeyer flask containing 2.0 ml M199 and placed on a reciprocating shaker bath (Dubnoff Labline) set at 37°C and 60-80 strokes per min. The flasks were loosely capped and allowed to exchange with the ambient air. The fragments were thus equilibrated for 30 min. At the end Of this period, the incu- bation medium was replaced with fresh medium (designated time zero). In the pulse-labeling experiments, 60 uCi of a mixture of 14C-amino acids in a volume of 0.6 ml were added to 1.4 ml Earle Solution. The preincubated pituitary fragments were drained of the medium con- taining cold amino acids on Nitex 80 synthetic gauze, rapidly rinsed 3x with 2 ml pre-warmed Earle Solution, then incubated for 15 min in a single 25 ml Erlenmeyer flask containing 2.0 ml of the salt solu- tion with radioactive precursors. Following the labeling pulse, the tissue was again rinsed on gauze with a total of 6 ml M199 in 14C-amino acids. The tissue three washes to eliminate exogenous pieces were assigned to flasks containing 2.0 ml fresh M199. (The beginning of the pulse labeling was designated time zero.) The medium was collected from the flasks and pooled, with fresh changes 30 made each hour after time zero for up to 6 hr. In the labeling experiments, one flask was terminated at end of the pulse and at each successive hourly interval. The hourly change of medium was continued until termination in order to minimize the potential effects of accumulated metabolites or secretory products on the cul- tured tissue. Processing of Medium and Tissue Samples The following procedures were conducted at 4°C. Medium sam- ples were centrifuged at 1000xg for 10 min to remove tissue debris and cells; supernatants were decanted into siliconized 12 x 75 mm glass tubes and frozen at -20°C. Pituitary tissue (~15 mg wet weight) was made into a 1% (w/v) homogenate with 50 mM tris phosphate buffer, pH 8.6 (4°C) containing 1 mM ethylenediaminetetracetic acid (EDTA). Homogenization was performed by hand in a Pyrex apparatus with ground-glass surfaces and 2 ml capacity by use of a grinding motion rather than a vertical stroke, to avoid protein denaturation accompanying foaming. The suspension (1.5 ml) was aliquoted into four 450 pl polyethylene tubes and centrifuged (Beckman model 152 Microfuge) at 10,000xg for 5 min. The supernatants were pooled, adjusted to a pH of 7.2-7.6 with 0.1 M H3P04, then frozen (-20°C) until assay. All tissue and medium samples were stored in silicon- ized glass test tubes. IV. Gel Exclusion Chromatpgraphyg Glass distilled water was used as buffer diluent for all pre- parative procedures. Sephadex G-100 (Pharmacia Fine Chemicals AB, 31 Piscataway, NJ) with a dry particle diameter of 40-120 u was swollen (and eluted) with 50 mM sodium or potassium phosphate buffer, pH 7.4 (4°C) at 80°C for 16-24 hr. The gel was resuspended, and the fine particles decanted at least twice. The suspension was stored at 4°C with 1:8000 Zephiran (Clay Adams), a 17% preparation of benzalkonium chloride, added as an anti-microbial agent. The gel was poured as a thick slurry, avoiding the inclusion of air bubbles. The packed column had either a total volume (Vt) of 53 ml (dimensions: 1.4 x 34 (an) or 42 ml (1.6 x 21 cm). The column was allowed to pack with the continuous flow of four Vt“ Operating pressure was 60 cm H20 and flow rates were 0.10-0.25 m./min. Homogeneity of the bed was visually evaluated by the progress of a uniform band of Blue Dextran 2000 (Pharmacia; 2 mg/ml) as it passed through the gel. Fractions of 1.0-2.0 i 0.05 ml were collected at 4°C by an Isco (Lincoln, NB) fraction collector (model 327A) calibrated to index a preset number of drops corresponding to a volume of eluent. When particulate mat- ter was trapped at the bed surface, approximately 1 cm of bed height was aspirated off and replaced with fresh, hydrated gel. Molecular weight determinations were based on the linear relationship of the log of the solute molecular weight with the par- tition coefficient between the liquid phase and the gel phase (Kav)° the Kav was calculated according to the method of Laurent and Killander (1964); by definition, K is independent of artifacts av resulting from compaction of the gel bed. The equation: v-v K=———° ° av Vt - Vo 32, is read as the elution volume of the sample (Ve) minus the void vol- ume of the column (V0) divided by the total bed volume minus the void volume. The void volume was determined by the exclusion of Blue Dextran 2000. Elution volume was taken at the maximum of solute concentration.* Six protein standards were used for the determination of molecular weights. These were prepared as 400-800 ug of solute in 0.5 ml of buffer. All chromatographed samples, standards or experi- mental, contained 7% sucrose to increase density of the samples. These were carefully layered onto the gel bed surface, beneath a layer of buffer. V. Prolactin Radioimmunoassay The concentrations of r-PRL in the samples obtained from the processed tissue extracts or culture media were determined by a modi- fication of the double antibody method described by Niswender gt_gl, (T969) and used by Charles Brooks, Department of Anatomy, Michigan State University. It utilized the standard radioimmunoassay kit dis- tributed by the National Institute for Arthritis, Metabolism and Digestive Diseases (NIAMD). Samples were assayed at multiple dilutions. Phosphate-buffered saline-1% bovine serum albumin (PBS-BSA) was pipetted into 12 x 75 mm glass culture tubes in the appropriate volume so that the addition of the sample (1-80 ul) brought the total *Sephadex Gel filtration in theory and practice, 1974. Printed in Sweden for Pharmacia Fine Chemicals AB by Upplands Grafiska AB. 33 volume of sample plus PBS-BSA to 500 ml. The first antibody, rabbit anti-rat prolactin diluted 1:15,000, was added to each tube in a volume of 200 pl. Contents of the tube were mixed by rotary agita- tion, and the tubes were refrigerated for 24 hr at 4°C. Approxi- 125I-r-PRL preparation (NIAMD-I-2 labeled by mately 30,000 cpm of a the chloramine T method of iodination) were added to each tube in a volume of 100 pl followed by a second agitation and storage for 24 hr at 4°C. The second antibody, ovine anti-rabbit gamma globulin diluted 1:80, was added in 200 pl. Following a final incubation at 4°C for 72 hr the content of each tube was diluted with 3 ml PBS-BSA and the complex was precipitated by centrifugation at 1000xg for 30 min. The tubes were carefully decanted of supernatant, drained by inversion on absorbent paper for 30 min, and permitted to dry. TO estimate the total count added to each tube, five tubes contained only the radioligand. Estimates of total binding were obtained by incubation of nine tubes with all components except unlabeled hormone. Non-specific binding of 125I-r-PRL to the anti- serum was determined by replacing the first antibody with 200 pl of 0.3% normal rabbit serum. A standard curve was constructed by plot- ting specific binding as a function of unlabeled reference rat pro- lactin (NIAMD-RP-l) added, ranging from 0.2-20 ng. Each concentration was assayed in triplicate. The precipitates were counted on a Nuclear Chicago Auto- gamma Counter (model 1185, Nuclear Chicago-Searle, Chicago, IL) at 50% efficiency. The precipitated radioactivity resulted from the competitive binding of labeled and unlabeled prolactin for each 34 sample; thus, the quantity of r-PRL in an unknown sample was inferred from the sigmoid displacement curve for specific binding of the radioligand. Concentrations of r-PRL were expressed as ng per ml sample. VI. Recovery and Measurement of Proteins Protein concentrations were determined by light absorption resulting from the standard colorimetric reaction described by Lowry pt 11. (1951). A standard curve of absorption at 660 nm was con- structed using 5-500 ug BSA. In the culture studies, duplicate assays were performed for at least two sample sizes. Proteins in the culture samples were precipitated by the formation of insoluble salts with trichloroacetate. One hundred ul of 50% trichloroacetic acid (TCA) were added to 0.5 ml of an aqueous sample in 12 x 75 mm glass tubes. The precipitates formed in 1 hr of incubation at 4°C were recovered by centrifugation at 3000xg for 30 min, 4°C. The supernatant was decanted, and the tubes were inverted and drained on absorbent paper. In samplesobtained by gel chroma- tography, 80 ug BSA was introduced as a carrier before the addition of TCA. These precipitates were either immediately separated by electrophoresis, or stored at -20°C. Samples were evaluated for B-emission by a Nuclear Chicago- Searle Mark I liquid scintillation spectrometer and computer (model 6860). Aqueous samples were counted in dioxane containing naphtha- lene, 2,5-diphenyloxazole (Sigma), and a-naphthylphenyl-oxazole (Packard Instrument Co., LaGrange, IL). Quenching for each samplev 35 133Ba external standard. A quench curve 14 was determined by use of a was constructed so that cpm recorded for C in various samples were converted to dpm using efficiencies ranging from 40-85%. Labeled proteins in polyacrylamide gels were prepared for liquid scintillation counting by dissolving the segment (2 mm) in 0.5 ml 30% H20 14 2 at 52°C for 12-16 hr. The dispersed gel fragments and C-protein were poured into a glass vial containing scintilla- tion fluid, and rinsed once with the fluid to ensure quantitative transfer to the counting vial. VII. Disc Electrophoresis in Polyacrylamide Gel Disc polyacrylamide gel electrophoresis with alkaline tris buffers was performed as described for non-dissociating separation (Davis, 1964) and fractionation under dissociating conditions with SDS (Laemmli, 1970). The composition and preparation of gels and buffers, shown in Appendix C and Appendix D, were as described by Davis (1964) and Maizel (1970). The total acrylamide concentration (I) and percent of cross-linking agent (9) were varied for the appro- priate separation. Non-dissociating electrophoresis was used for 0.3 ml samples of pituitary homogenates. Spacer gels were formed by pipetting 0.4 ml of gel solution on top of 60% sucrose in inverted, stoppered, 10 cm siliconized glass tubes (5 mm i.d., 7 mm o.d.). These were over- layered with water and polymerized for 45 min at room temperature with a fluorescent lamp. A separating gel (6 cm) was formed on top Of each inverted spacer gel by pipetting the gel solution to overflow 36 an attached sleeve of rubber tubing. After polymerization (2 hr), the tube was inverted and placed in a transverse electrophoretic apparatus (A. L. Davis Co., Lincoln NB), and the trays and tubes were filled with diluted electrode buffer. Pituitary extracts were made to contain 12% sucrose and layered on top of the spacer gel. Electrophoresis was carried out in a 4°C coldroom. Samples were stacked at 1 mA/gel with an Isco 490 power source until the sample entered the spacer gel. The current was increased to a constant 2.5 mA/gel until the phenol red marking the ion front had moved 5.5 cm through the separating gel. Electrophoresis under dissociating conditions was performed with reservoir and gel buffers containing 0.1% $03. No spacer gels were used. Separating gels 13 cm in length were poured in 14 cm glass tubes coated with dimethyl dichlorosilane (Sigma) and soaked in 1% 505 between uses. The solution was overlayered with 5 mM tris- HCT pH 7-8 by gravity flow through PE 100 tubing (Clay Adams). Poly- merization was allowed to proceed for 6-12 hr. Samples precipitated by TCA were dissolved at 37°C in 50 ul of 20 mM tris-phosphate pH 7.0 containing 1.5% SDS, 2% mercaptoethanol, 10% glycerol and 1 mM EDTA. Phenol red was added as a pH indicator in the sample and served as a marker for the ion front during electrophoresis. The samples were then placed on a boiling water bath for 2 min. The volume during the incubation of the protein-SDS complexes was reduced to approximately 35 ul. This sample was carefully layered to the top of the gel beneath the electrode buffer. Electrophoresis was performed at room temperature at a constant current of 2 mA/gel for approximately 4 hr, 37 when the ion front usually reached the bottom of the gel tube. At the termination of the run, a maximum voltage of 275 was reached. For the determination of molecular weights in the SDS sys- tem, protein samples and standards were dissolved to a concentration of 0.5-2.0 ug/ul. No more than a total of 100 ug of protein was applied to a single gel. Proteins separated in the non-dissociating system were stained by immersion in a solution of 1% amido black (Sigma Chemical Co.) -7% acetic acid (HAc) and destained in 7% HAc. Gels from the SDS-dissociating system were fixed in 10% TCA-25% isopropanol, and stained in 10% HAc-25% isopropanol containing Coomassie brilliant blue (Sigma). These were destained in 10% HAc-TO% isopropanol. All steps in the staining procedures were accomplished by diffusion for several hours at room temperature. Gels were sectioned into 2 mm segments by a device described by Chrambach (T966). EXPERIMENTAL RESULTS I. The Bindingiof 125Iodine-Prolactin to Plasma Membrane Receptors By the use of the method for rapid, single-step separation of iodinated prolactin well defined, widely disparate peaks were not apparent by the criterion of receptor activity, but differences in adjacent fractions of gel column eluent were Observed. Fractions designated S4 co-eluted with a non-filtering marker molecule, Blue Dextran 2000 (Pharmacia) and therefore indicated the presence of some iodinated fraction (prolactin) in the void volume. Chroma- tography of unlabeled ovine prolactin (o-PRL) under identical condi- tions yielded an asymmetric peak of protein with a maximum 1 corresponding to 55. Within each binding experiment, assays of both fractions were performed simultaneously. Results and Discussion The specific binding ratios for the different fractions for a given iodination of o-PRL are listed in Table l, with evaluation of the binding by Student's t-test. Means are shown with i standard error of the mean (i SEM). For fractions S4 in experimentsl and 2 there was higher mean total binding, but insignificant differences between groups A and B. The succeeding fractions (55) bound spe- cifically 32% of ‘25 I-o-PRL (P < 0.05). In the third experiment, the fraction associated with void volume (S4) yielded greater spe- cific binding (P < 0.05) than $5 (P < 0.2). 38 39 mpe.P n A .~.oe Paeeeece N~4.F em.o Amy emmnmmm_ Rev whmnmmom coco“ mm ALmEEmE mmm.~ m~.o Amv ~m_FnNme~ Amy emceeomop ooomfi em emaoe m mom.~ n “.mo.oe Peeeeeee m_e.~ Nm.o Rev ommnempm Amy om__nmmo~_ cocoa, mm acmEEmE o_m.o mo.o Amv memnmewk_ Rev mmmphmmmm_ cocoa em amzoe N _Nm.m n m.mo.oe _eceeeec emo.m mm.o Aev mmo_nNFANN Amv _mmmhmmmmm cocoo_ mm cmm.o mo.o Aev emmmnmmpmm Amv oemmnmpoem coco“ am ca>eF pee _ .1 sz eesoa 4mm sz eezoa 4mm a=~a> a -H eeaeoe -H ceaeee eaeaa sea eoee peas tint: mm_ mNF - - anamee - e m a a: ooF\5ae a: co_\eQe HmNF Ego pace ecaaxm owewuwamicoz pouch m .< .mgopamumm mangaswz msmmpa op caspoo Fmo xmnmgamm soc» :wpumpoem mcw>o umumcwuoH we mcomuumcd acmumnn< we mcwucmm mghii._ m4m

124000 daltons. Much of the immunoassayable hormone recovered from the medium elutes at Kav > 0.70 with molecular weights < 7500. In any of the chromatograms examined, this material did not elute as any discrete peak(s). Although it increased as a proportion of the total hormone recovered in later hours of the second culture experiment (Table 2), the low molecular weight component was less abundant later (3 hr) in the first experiment. The low molecular weight r-PRL was observed in sonicate and homogenate preparations of tissue which had been cultured. Since these proportions of low molecular weight cross-reactive material were variable, only the hormone fractions with Kav 5.0.70 were considered in expressing the percentages of 43 .muemvcwpm mm cam: unmwmz Lm—sompos czocx mo mewmpoca eo mmwemm m ccm pcmwmz Lepzumpoa mo cowm -mwemme mgp 205m mmpugwu cmao ucm m:m_ :mxoea < .mFQEwm comm cw nmcm>oome AoN.o.w >mxv segue—oea pwpou pcmocwa to» umEEmemmwu we mFQEmm Lao; comm cmcz mpcmcoaeou Go wucmucznm pcmpmcoo mzonm Loecm nemucmpm cpwz AmmFQqu uPFOmv pcwoa comm .me:u_:u zgmuwzuwa mo meso: pcwemeewo m>wu um Eswumz mgp Eoce cmcm>oumm cwpumpoea yam mFmemmmeocaasH mo Ememmmo cowpspm :<-1._ mczmwm 44 - - —£_ou LHOIBM unnoaiou 8??? 8 e A A bIO l j I r T 0-4 05 0-6 07 >07 oz 63 KC" I 00 0-1 201 V ' O K) 151 —' OBUBAOOBU ‘18:! 111.01% 45 TABLE 2.--The Abundance of Low Molecular Weight Prolactin: Immuno- reactive Hormone with Ka Less or Greater than 0.70 Recovered from Media or Tissue Samples of Cultured Rat Adenohypophyses. K 5.0.70 K > 0.70 Source Total av av ng ng % ng % Medium Samples 1 hr 400 299 75 111 25 3 hr 50 50 100 0 0 1 hr 1300 1288 99 12 0.9 2 hr 398 387 97 11 2.8 3 hr 215 176 87 29 13 4 hr 244 221 91 23 9.4 5 hr 95 66 70 29 30 6 hr 144 99 69 45 31 Tissue Samples Sonicate 1528 1457 95 71 4.6 Homogenate 10354 9704 94 650 6.3 prolactin recovered in each fraction. These percentages were constant for a given Kav in the superimposable eTution profiles for hr 1-5 in the second culture, shown with standard errors of the mean in Figure 1. Component I was so designated since a Kav of 0.4 indicated a molecular weight of 23000, apparently monomeric r-PRL; component II with a Kav ~ 0.3 has an apparent weight of 37000, not inconsistent with the weight of dimeric r-PRL. Component IV which elutes near the void volume has a molecular weight > 83000. (Component III is found only in pituitary homogenates.) From the data of Table 3, it 46 TABLE 3.--The Distribution of Secreted Monomeric and Polymeric Rat Prolactins Separated by Gel Chromatography: Recovery of Immunoreactive Hormone from Incubation Medium 199. Component I Component II Component IV Source ng % Kav % Kav % Kav Hour 1 299 71 0.42 29 0.30 -0- -- Hour 3 50 71 0.46 29 0.30 -0- -- Hour 1 1288 74 0.42 24 0.28 2.2 1<0.1 Hour 2 387 66 0.42 28 0.28 6.1 1<0.1 Hour 3 179 71 0.42 26 0.28 3.0 1<0.1 Hour 4 221 63 0.42 25 0.28 11.0 1<0.T Hour 5 66 69 0.42 25 0.28 5.6 .<0.1 it is apparent that the proportions of these components remained stable throughout the period of culture. Component I averaged 69% of immunoreactive hormone with mol. wt. 2.7500; component 11 accounted for an average of 27%, and component IV had a mean of 4%. Components I, II, and polymeric component IV paralleled the displace- 125 ment of I-r-PRL by the reference prolactin in radioimmunoassay. III. Prolactin Extracted from Cultured Adenohypophysial Tissue The radioimmunoassay of culture medium proteins separated by gel filtration demonstrated the considerable size heterogeneity of r-PRL. The literature records differences between intrapituitary and circulating forms of adenohypophysial hormones. Some investigators suggest that consequences of cellular extraction may result in the 47 .mmezu_:o mem»;aoa>;ocmu< Eoee umem>oumm mpcmcoquu m>wpommcoczesH msowew> can cowpmcmamga mocmgmemm 8 >2 xconwpc< Eoce cwpumpoea yam umumauoH we pcmEmumpamwo eo Ewe—mFFmemmii.N mczmwm 4E3 2H90¢4Qmm 94¢ uc REC r r . o...n » C 0*... A3117/1/va $-.CV mmzwnom ‘- 5 .53-.“ I H Amery nu dimminz<~z c N. .7? N... 1/8 £1901 49 appearance of multiple forms of hormones. Thus, this study examined immunoreactive prolactin extracted by several commonly used proce- dures which are known to preserve the molecular integrity of pro- teins. Extracts were obtained from tissue which had been cultured as described in the section on materials and methods. An alkaline homogenate was prepared from a pool of adenohypophysial tissue cul- tured for 4 hr. A second preparation was of pituitary tissue cul- tured 6 hr and sonicated by a cell disruptor (Branson Sonic Power Co.) for 1.5 min in M199 on ice. A third pool of tissue was homoge- nized and electrophoretically separated at 4°C under non-dissociating conditions, using a 2.5% spacer gel and a 6% separating gel (Appendix D). Segments of the separating gel (4 mm) were sliced with a razor blade and eluted in 0.5 ml lmM EDTA for 16 hr at 4°C. Only segments which corresponded to the relative mobility (Rf) of r-PRL (cf. Experimental section III) were chosen for chromatography. Samples derived from all three extraction and isolation pro- cedures were separated on G-lOO Sephadex. Since the first experimen- tal section indicated that low molecular weight (< 7500) immunoreac- tive fraction did not appear in a discrete peak, only the material eluting with Kav 5.0.70 was considered in subsequent evaluations of heterogeneity. Results and Discussion In Table 5 (p. 70) are listed the recoveries and distribu- tions of immunoreactive components with the calculated Kav for samples of the following treatments: 4 hr homogenate, 6 hr sonicate, 50 and 6% separating gel. The chromatographic profiles are shown in Figure 3; the profile of hormone secreted into M199 (taken from Figure l) is also shown for comparison. Each profile shows that components I (Kav ~ 0.4) and II (Kav ~ 0.3) are the major forms of immunoreactive hormone. On the basis of Kav determinations, intrapituitary components I and II are similar to the secreted forms. The minor polymeric com- ponent IV was present in tissue as well as secreted into the culture medium. Thus, the heterogeneity of r-PRL as discrete molecular forms is not an exclusive artifact of secretion jg_vitro or the extraction procedure. However, while component I was the major secreted form, component II was the most abundant intrapituitary form of immunoreactive hormone. Furthermore, a minor immunoreactive component III which eluted with a molecular weight similar to albumin was detected only in the prolactin band of pituitary proteins sepa- rated by PAGE; this component was not separated in samples of culture medium, sonicates, and homogenates. IV. Electrophoretic Heterpgeneity of Immuno- reactive Intrapituitary Prolactin In the previous study, prolactin which electrophoresed as a single component (at a single gel concentration) was shown to contain 4 discrete components. These appeared to be polymers of monomeric prolactin (mol. wt. ~ 20000). These forms, under the combined influ- ences of their charge and the sieving characteristics of the gel, migrated indistinguishably. Therefore, an attempt was made to locate species of r-PRL which possessed greatly different charge 51 >m .czocm mcm HH ucm H mucocoasoo com x moom_:o_mu .mE:_o> cowpzpm mcwmmococw ow mFVCoca owcamcm -omeocco comm coo mmmwomcm och .Amszv Ezwmmz mcappzu .comwcmqeoo coe mcm .mzmmwh acmpwsuwa mo Amuooom :wpompoca umm m>wpommgoczeeH mo mscom mpqmupzz eo mocmcmmaa<-i.m oczmwc 52 ONAu ~¢;o mm_2 OnAu OeIO weda w2340> zo_h34u a NNAV bnAv wh 10 ’ .50 .75 .25 .75 1.0 .50 .25 3 hr 2 hr 57 .mpmom mpmcwmco Axopv mmmcmaxo cm cow: wmmmwomcm mcp mo mocwoa xvm pmcwm wcu cow: czocm ocm mOPQEmm moomocoifizoium-zmv Focmcooopamocoe mcm .moummcoiommzm .Focpcoo .moomcpxm xcmpwzowa mo mFow cmomqm Eocw mmszm :wuomFoca o>wpommcocaeaH we mmpmmoca owcamcmoomsocco meii.m mczmwc 58 : . IO..U.Im I wing—O) 2075...; ooudzm O .x O... «no : see. 3.0 . 3x 40:... 200 4 to B .0. TON ABE/10038 13d 'TVLOJ. % 59 released finished r-PRL, the immunodetectable hormone associated with RNA was mostly monomeric. Reduction of disulfide bonds with B-mercaptoethanol yielded only the monomeric component I. Immuno- reactive components II and IV were not detectable following this treatment, and it was concluded that these polymeric components can be explained at least partly by intermolecular disulfide bonding between prolactin and other protein(s) and/or among prolactin sub- units. A 20% increase in the recovery of immunoreactivity was noted for the chemically reduced hormone as compared to control or enzy- matically treated samples. This has been attributed to the masking Of immunoreactive sites by disulfide bonding (Jacobs and Lee, 1975). The results of intrapituitary labeling of intrapituitary pro- teins synthesized ip_yjtrp_are shown in Figure 6. One gel of a pair prepared for each time interval was evaluated for immunoreactive r-PRL; the other was sliced and counted for 14C activity. In these cultures, somatotropin was not as highly labeled as r-PRL. Further- more, r-PRL was turned over (released) much faster than total pitu- itary protein, as evidenced by the drop in percent of total radioactivity contributed by r-PRL. This was confirmed by a 14C-r-PRL to 14C-TCA-precipitated radio- leveling-off in the ratio of activity, which had logarithmically increased until the second hour of chase incubation, then declined (unpublished observations). Although the upper 1/3 of the gels separating samples from the early culture periods contained immunoreactive molecules, no highly labeled peaks were observed; likewise, there was no evidence of rapidly con- verted protein precursors of r-PRL. The immunoreactive proteins 6O .Aemv meeeeeoe a>eoepac we mmmwomcm mcp mpom mcwpmcmawm ocp Eocm mmcw>oowc xuw>wpom o _ Pmuop we we mmmucmocmg mm momcwmco och .cowumw>mm mcmucmum omcwe oco cumw :zocm ocm mucmc :wpompoca mmcwmpm mcm mmpwcmp mewxme .mfl umpocmc mcwmpocmiowp acmpwspwamcch mo cowozcwcpmwo mc»--.o oczmwd 61 >h_.:moz 0.. mo 90 #0 NO glb D b D U P b b b o a o a a o d he dadma <35Aud004<0pw ‘6 “can“ dflfi a As a O a a a 4 O O 0AU w: a a a a r O O c a. e I 2.3m :o r .- .ON _ arm 8 . x... 4 o. moo: . .o» I 4.x 4a.... a hmoao I z_<.rm Jan. I 0?. 41m '139 N0 ALTAILOVOIOVH 96 62 14C which migrated at the ion front had a low specific activity for and may be the low molecular weight polypeptide(s) which contain(s) the antigenic determinant of r-PRL. From these limited data, it is unknown whether "prolactin" with Rf = 1.0 is equivalent to the several cross-reactive molecules with Kav > 0.70 in gel filtration. Prolactin described by Rf = 1.0 accounted for an average of 11% of the recovered immunoreactivity in the 4 gels examined, while immuno- reactivity in gel filtration of 10 pituitary-extracted or secreted samples averaged 12% for Kav > 0.70 (Table 3, page 46). The relative mobilities (Rf) for maxima in 6% separating gels are listed in Table 4 (page 65). The stained, 14C-protein band peak was found to be indistinguishable from immunoreactive r-PRL by its mobility in the non-dissociating system. V. Heterogeneity of Newly Synthesized Rat Prolactin Previous experimentation using non-dissociating PAGE did not clearly separate the multiple, immunoreactive forms of prolactin. Labeling of cultured pituitary glands yielded only a single, highly labeled peak of prolactin 14C-activity. However, when this gel segment was chromatographed on Sephadex (Experimental section 111), four immunoreactive molecular species were observed. Because the non-dissociating PAGE at a single gel concentration failed to resolve the multiple components observed in gel filtration, separation by gel chromatography followed by liquid scintillation analysis of the pituitary proteins for labeled prolactin was used to examine the heterogeneity of the synthesized hormone. This requires a rigorous 63 14C-r-PRL among other labeled pituitary proteins. means of identifying Immunoprecipitation has been used to selectively recover labeled hor- mones; however, this technique limits the search for multiple hormonal forms to molecules containing the antigenic determinant of a highly purified protein. Precipitation of all proteins with TCA after size- dependent separation permitted me to simultaneously identify and characterize the polymerization of labeled r-PRL by SOS-disc PAGE. After preincubation, rat pituitary tissue fragments were 14C-amino acids in 2.0 ml of Earle Solution. exposed to 60 uCi of The post-pulse (chase) incubation was in a series of flasks of M199, changed each hour from to. One flask was terminated at tO and.at each hour thereafter when the contents of each flask were immediately processed and frozen. The samples were thawed and separated on Sephadex G-100. The fractions collected were precipitated with TCA. These precipitates were separated by SOS-disc PAGE in 10% I_gels (Appendix D). Gels were placed in tightly capped tubes and cooled for a few min to reduce protein diffusion, then sliced into 2 mm segments. Eight of these segments were taken to ensure the inclusion 14C-r-PRL band (Table 4, page 65). Each of these eight segments 14 of a were dissolved and counted for C; the r-PRL was typically located in 1-3 (consecutive) segments. Background radioactivity was defined as the average counts in segments on either side of this peak and 14C-r-PRL, was plotted as a subtracted. The resultant, net dpm of function of Kav calculated for the corresponding fraction of the dextran gel separation. 64 Results and Discussion 1251_r_ A test precipitation of chromatographically purified PRL suitable for use in RIA recovered 71.13 i 1.41% (i SEM, N = 8). Thus, TCA precipitation provided a consistent, reasonably quantita- tive method to recover r-PRL. When pituitary homogenates were sepa- rated On SOS-disc gels and the proteins stained, two heavily stained bands appeared in the lower 1/3 of each gel. The rationale for the 14C-r-PRL was summarized from data shown in Table 14 identification Of 4. When eluates from gel segments containing both the C-protein and immunoassayable r-PRL (Rf ~ 0.8) were precipitated with TCA and re-run in SOS-disc PAGE, peak radioactivity was recovered only in a single peak with a mobility indistinguishable from both the slower migrating, stained band and the major radioactive peak of labeled proteins extracted from cultured pituitary tissue. The prolactin remained well separated from somatotropin when run in gels 12-13 cm in length. This separation was apparent in 14C analysis of Sephadex 14C-r-PRL appear in fractionated pituitary extracts; not only did one of the eight segments counted, but frequently a small peak of radioactivity with higher mobility (GH) was discerned. The identity of r-PRL and r-GH bands was ascertained further when pituitary extracts were run in parallel gels with a partially purified prepara- tion of r-PRL (NIAMD-RP-l) or a highly purified r-GH (NIAMD-GH-I) at 10, 12, and 15% I; the slower band from extracts co-migrated with r-PRL, and the band of higher mobility co-migrated with r-GH. At 8% I, r-GH and r-PRL migrate indistinguishably, and do not appear as well separated bands in stained gels of pituitary extracts. 65 TABLE 4. --Relative Mobilities (Rf) of Rat Prolactin (PRL) and Somato- tropin (GH) Components of Pituitary Extracts in Poly- acrylamide Separating Gels. Component Rf 1 SEM (N) A. Non-Dissociating System (6% I)a PRL RIA 0.795 t .016 (4) PRL ‘40 0.813 t .016 (3) PRL stain 0.803 t .007 (3) GH stain 0.343 t .010 (3) B. Dissociating System with SDS (10% I)b 14 PRL C re-run from RIA peak in A 0.770 t .010 (3) PRL ‘40 0.753 s .010 (6) PRL stain 0.767 s .017 (4) GH ‘40 0.830 1 .006 (6) GH stain 0.837 s .017 (4) aNo significant difference among PRL Rf [K*= 0.5835; c = 3.836, P = 0.150; Table I (Lehmann, 1975)] by Kruskal -Wallis Test for comparison of more than two treatments. bNo significant difference among PRL Rf [K* = 1.4966; c = 1.50; P = 0.4724; Table J (Lehmann, 1975)] by Kruskal -Wallis Test. 14C-r-PRL localized by SDS- The chromatographic profiles for disc PAGE immediately after the pulse-labeling period and hr 3 post- pulse are shown in Figure 7 and Figure B. Extracts of these samples contained 500-800 pg protein in a volume of 0.5 ml. The recoveries and distributions of various components are summarized at the bottom of Table 5 (page 70). For these values at to, total pituitary pro- teins were labeled to an activity of 221 i 22.5, (4) dpm/ pg protein; 66 .m mam < mpmwcp cow mpcwoa mo mommcm>m mcp cmaoccp :zmcm mm: o>c=o < .mm_:a o>woomowmmm m emue< »_aeeemasem cocoa_oca omm-o «F memoeaoeamcoafi ea epeecea eoeeeeo_ee Fem ace--.“ mesmec 67 >3. n6 N6 . O 0.0 I I I I I I I I 800 m I I fins. 4 I >mw>oomm 1.4 . m... WdO z-O" 9171 68 .m age < mpmwcu coo mommem>m casoccu czmcu mm: o>c=o < .mm_=a m>wpomowmmm m cmpe< 9.50: wwLPE. czbmpoc—n— Homio E. Acme_=oeamceac ea appease caeomcopee _eo ace--.m mesmec TRIAL 69 J d 2-0' x 0,, mo 7O .HHH ucmconsoo eo mocmmczcm was» ac» moamseommco>o xpocpp mam .cowmoc :mopmpa m Loo mmumpsoFmo mm: oompcoocmq mopweoea cowuapo ocu co Aonopm m>Ppmmmc o: ..m.pv mm>cmmco mm: xmma opmcommm ozm 85v 2 2.0 2 9.5 8 were a 2% :2 .2 M 882825: 8.3 .2 2.0 mm 85 S 85 mm 5% $3 .2 m .qeoeeemoeaz 35v : 2.0 fl and 3 mad mm 5% at” so 382825: 3.3 E 2.0 2 and S was a 5% $3 a» .«eemeaaaeé 8.3 o; -- -o- and mm 53 Na 9. m2: amaze :8 team -- -o- -- -o- .i -o- 2.0 2: a: 22 to m: :8 beam 3.? am -- -o- and 8 .2; mm 9. 82 :33 :5 team 85 v mm 85 S one on 8.0 mm 9. mm $8 ,8 3:238 3.3 E -- -o- 35 mm 3.0 S 9. $3 .5 m .8823 8.? me 85 mm; and mm «to on 9. 83 2 a .aemeamoea: so. A 2w. e 5o. a so. a sun go mz moczom >H ocmcoaeou HHH ucmcoasou HH ucmcoasou H “cocoasou mam .cmwewmoz appmomsmcu .uomococaocuompm .uopmocucn mo xgm>oomm .ocoEgo: mmpocmctu ”Acamemoomeoscu Pwm xc umumcmaom mcwpom_oc¢ pmm mgmuwzuwamcch omgmszpoa mam owcosocoz mo cowozcwgpmmo ochii.m ucm 8% I) gel concentrations to detect polymers of prolactin and somatotropin was predicted by Cheever and Lewis (1969). These authors noted that the gel systems commonly used to demonstrate homogeneity of hormone preparations inadequately separated the pro- tein components of several highly purified mammalian hormones. In this study, the r-PRL found in spacer gels did not contain a 80 preponderance of immunoreactive polymers; monomeric and dimeric pro- lactin which did not enter 6% separating gels might have been asso- ciated with RNA. Under this assumption, some prolactin assembled on polyribosomes can be detected by immunoassay. Although enzymatic degradation of RNA in these studies did not increase the total of the immunoreactive r-PRL, less of the large prolactins were observed after treatment, while a proportionate increase in the abundance of the monomeric component was noted. The reduction of disulfide bonds with mercaptoethanol effected the recovery of monomeric r-PRL alone. It is interesting to note that the total recovery of prolactin increased, indicating that immunoreactive sites may be inaccessible because of intermolecular bonding of sulfhydryl groups in large hor- monal forms (Jacobs and Lee, 1975). In a similar way the serological activity of brain S-100 protein was increased by the reduction of disulfide bridges (Dannies and Levine, 1971). The low molecular weight r-PRL located at the ion front in non-dissociating PAGE was not examined further to determine its structural properties. How- ever, since it accounted for much of the immunoactivity, it should be investigated as a lactogenic product of the degradation of pro- lactin, capable of penetrating plasma membranes and directly affect- ing nuclei of target cells (Chomczynski and Topper, 1974). This study identified the heterogeneity of immunoreactive r-PRL, and evaluated the synthses of several forms by a combination of gel filtration and SOS-disc PAGE; these are two systems for recovering r-PRL based on its physical properties without the requirement of immunoprecipitation. The identification of labeled a... v. - 81 r-PRL was based upon comparisons of the relative electrophoretic 14C- mobilities of purified hormone preparations with those of protein which cross-reacts in RIA. A shift in the specific activity of intrapituitary r-PRL from large forms to the monomeric hormone was shown. However, the decrease in the relative activity of any of 14c-r-PRL. It these large forms was less than the turnover in total is likely that these forms then are independently released, as well as converted to smaller forms before secretion. ID1these experiments, it was not determined whether an alternate or simultaneous process of the aggregation of smaller forms into polymers contributed to the slow turnover of the larger forms. This would be similar to a mech- anism described by Lukens (1976): trimeric chains of procollagen are linked by disulfide bridges before their secretion from embry- onic cells of chick sternum; inhibition of nascent polypeptide chain elongation did not affect the conversion of single chains into oligomers. In this study, the determinations of the weight of the small- est unit of synthesized, immunoreactive r-PRL by partitioning coeffi- cients (Kav, KR) in dextran and polyacrylamide gel matrices showed fair agreement. Calculated molecular weights for monomeric r-PRL were 23700 by dextran gel chromatography and 27400 by SOS-disc PAGE. The size or r-PRL at physiological pH and low temperatures did not differ from the hormone denatured by detergent and disulfide reduc- tion. This was not necessarily expected. When Chrambach gt_gl, (1971) characterized h-PRL on the basis of the KR Obtained by non- dissociating PAGE in two alkaline pH systems, estimates of molecular 82 weight were 17200 and 43300; the latter determination was inter- mediate between the expected weight of dimer or trimer, assuming a unit weight of the lower figure. We found the gel chromatographic determinations of molecular weights of dimeric component II (35700) and trimeric component III (59400) both fall short of their expected molecular weights as true oligomers of r-PRL (23700). However, Suh and Frantz (1972) identified a large species of h-PRL (56000) also intermediate between dimeric and trimeric forms of h-PRL (21000). Ellis gt 91. (1968) noted that proteolytic degradation of r-GH (45000) to fully bioactive subunits yielded molecules with low molecular weights (18500) and greatly reduced the immunoactivity. The large forms observed in this study are not thought to be pro- hormones, i.e., did not contain a non-hormonal peptide which is cleaved to yield the active forms, as has been described for r-GH and r-PRL (Sussman gt 21,, 1976; Maurer §t_gl,, 1976; Evans and Rosen- feld, 1976). The prohormone of prolactin was not distinguished here possibly because the time of pulse-label was of sufficient length to facilitate conversion before a tissue sample was processed. Addi- tionally, the overlap of components I and II in gel filtration would have hidden the prohormone. The failure to generally Observe the prohormone or large forms in SOS-disc PAGE without prior gel filtra- tion (Zanini et_gl,, l974a; Zanini_gt_gl,, l974b) can be explained by the reduction of intermolecular disulfide bonds with mercapto- ethanol and heat dissociation of macromolecules with SDS. Fairbanks gt_gl, (1971) have also cautioned against temperature-dependent pep- tidase activity which is enhanced by partial denaturation of protein 83 substrates in low concentrations of detergent, and have determined that high concentrations of SDS and heat are necessary to completely block the activity of proteolytic enzymes. The results of these experiments have been interpreted to demonstrate apparent polymers of monomeric r-PRL, but this does not preclude the association of r-PRL with non-hormonal proteins, as Kruse (1973) has reported for the association of bovine prolactin with serum proteins. It is possible that r-PRL binds to ubiquitous proteins with readily inter- changeable sulhydryl groups, such as albumin (Wallevik, 1976) and this fulfills a role in the metabolism of the hormone. However, the presence of at least two polypeptides with molecular weights similar to monomeric or dimeric prolactin are required to account for size heterogeneity through association with other proteins. The investigation of intrapituitary, newly synthesized 14C-r-PRL did not reveal any components not detected by RIA, nor did it fail to monitor production of forms indicated in the immunoassay. We conclude therefore that the chromatography-PAGE technique is a valid means to trace the metabolism of pituitary r-PRL. However, these results have not been compared to the heterogeneity which might be observed by bioassay or receptor assay; they did not obviate the existence of a non-detectable form (Swearingen, 1971) which would be demonstrated by discrepancies among various assay systems (Nicoll, 1975). Furthermore, the general success with RIA using the standard NIH antisera does not guarantee the identity of labeled hormone with purified r-PRL. Yalow (1974) has noted that antisera which react identically (parallelism) to parathyroid hormone of several subjects 84 might react differently to parathyroid hormone from still another subject. Thus, previous success with a particular antiserum is a recommendation for, but not an assurance of, the validity of the RIA in a given circumstance. Therefore, the development of assays for prolactin which differentiate its physical properties offers a greater potential for standardizing the characterization of the populations of prolactin and gives better definition of the physi- ological states involving the control of prolactin production and target cell activity. APPENDICES 85 APPENDIX A LINEAR REGRESSION ANALYSIS: ESTIMATING EQUATION OF TWO VARIABLE LINEAR CORRELATION AND CONFIDENCE INTERVAL Equation for regression line (Dunn, 1964): Y = 17 + b(x-7) b ._. z(x-T<')(Y-17) —-2 2(x-X) Confidence interval (95%) for the population mean: _ — - — 2 2 0*” - v + b(X*-X) : tn_2’97.5SYx /1/n + (X*-X) /z(x-TO (t taken from Table A4, Dunn, 1964) The standard error of estimate: va = /z(Y-T)2 - b2£(X-X)2 / n-2 86 APPENDIX B LINEAR REGRESSION ANALYSIS: CORRELATION COEFFICIENT FOR TWO VARIABLE LINEAR CORRELATIONS Correlation coefficient (Dunn, 1964): 2(X-X) (Y-V) «QIJFuv—mz r = correlation of variables X and Y Significance Of correlation (Croxton, 1953): t = r2(n-2) l-r2 (t taken from Appendix V, Croxton, 1953) n = number of samples considered in correlation Standard error (Sr) of correlation coefficient r: S = Jl-rzln-Z Y‘ 87 APPENDIX C STOCK SOLUTIONS FOR DISC GEL ELECTROPHORESIS Stock Solutions: (1) (2) (3) (4) (5) (6) (7) acrylamide-pj§_acrylamide (30.8% I, 2.6% C) 30.0 9 acrylamide (Canal Industrial Corp., Rockville, MD, or Eastman Kodak, Rochester, NY) 0.80 g N,N'-methylenebisacrylamide (bis) (Canalco) diluted to 100 ml with distilled water sodium dodecyl sulfate (10%) 10.0 g 95% SDS (Sigma Chemical Co., St. Louis) diluted to 100 ml N,N,N',N'-tetramethylenediamine (TEMED) (Canalco): as received ammonium persulfate (10%) 100 mg ammonium persulfate (Canalco) dissolved in 1.0 ml water; made fresh every two weeks, stored at 4°C B-mercaptoethanol (Sigma): as received gel buffer 36.6 g tris (hydroxymethyl) aminomethane (Trizma base, Sigma) 9.6 ml 5N HCT diluted to 100 m1, pH 8.8-9.0 electrode buffer (10X) 6.0 g tris and 28-29 g ammonia-free glycine (Sigma) diluted to 100 ml 88 APPENDIX D FORMULAS FOR THE PREPARATION OF ELECTROPHORETIC GELS Separating Gels: Quantities cM’ stock reagents (Appendix C) to make 20 m1 gel solution. Amounts of reagent (1) to yield the final gel concentration in percentages. Percent I _6_ 8 10 . 12 15 Reagent (1) ml 4.0 5.3 6.7 8.0 10.0 2.5 ml reagent (6) 10 pl reagent (3) 0.20 ml reagent (2), omitted for non-dissociating systems water to 19.9 ml and de-aerate under vacuum 100 pl reagent (4) Spacer Gels (2.5% I, 20% 9) Reagents: fprmplg; (a) 1.8 ml 85% H3P04 1.0 m1 (a) 6.0 g tris 2.0 ml (b) 0.50 ml TEMED ' 2.0 m1 (c) H20 to 100 ml, pH 6.7 3.0 m1 H20 (b) 8.0 9 acrylamide 2.0 9 bis H20 to 100 ml (c) 5 mg riboflavin (Sigma) 80 9 sucrose H20 to 100 ml 89 APPENDIX E PROTEINS USED FOR MOLECULAR WEIGHT DETERMINATIONS Protein Manufacturer's Designation Source Bovine serum albumin Deoxyribonuclease I Gamma globulin Myoglobin Ovalbumin Ovine prolactin Rat prolactin Rat somatotropin Ribonuclease A Trypsin Trypsin inhibitor Fraction V Bovine pancreas, B grade Equine muscle Impalpable powder NIH-PS-ll NIAMD-RP-l NIAMD-GH-I Hirs reagent Pancreas, A grade Soybean, B grade 90 Sigma Chemical Calbiochem Major contaminant of BSA Sigma Chemical Fisher Scientific NIAMD NIAMD NIAMD Calbiochem Calbiochem Calbiochem REFERENCES 91 REFERENCES Andrews, P. 1964. Estimation of the molecular weights of proteins by Sephadex gel-filtration. Biochem. J. 91:222-233. Andrews, P. 1965. The gel-filtration behavior of proteins related to their molecular weights over a wide range. Biochem. J. 96:595-605. Aubert, M. L., P. E. Garnier, S. L. Kaplan, and M. M. Grumbach. 1975. Heterogeneity of circulating human prolactin (hPRL); decreased radioreceptor activity of "big" hPRL. Abstract 59 of the 57th Meeting of the Endocrine Society. Ben-David, M. 1967. Sensitive bioassay for prolactin based on 3H- methyT-thymidine uptake by the pigeon-crop mucous epithelium. Proc. Soc. Exp, Biol. Med. 125:705-708. Ben-David, M., and A. Chrambach. 1974. Isolation of iso-hormones of human prolactin from amniotic fluid. Endocr. Res. Comm. 1:193-210. Benveniste, R., M. E. Stachura, M. Szabo, and L. A. Frohman. 1975. Big growth hormone (GH): conversion to small GH without peptide bond cleavage. J. Clin. Endocrinol. Metab. 41: 422-425. Biswas, D. K. , and A. H. Tashjian, Jr. 1974. Intracellular site of prolactin synthesis in rat pituitary cells in culture. Biochem. Biophys. Res. Comm. 60: 241- 248. Blatt, L. M., and K. Kim. 1971. Regulation of hepatic glycogen synthetase. Stimulation of glycogen synthetase in an in vitro liver system by insulin bound to Sepharose. J. Biol. Chem. 246:4895-4898. Burek, C. L., and L. A. Frohman. 1970. Growth hormone synthesis by rat pituitaries jp_vitro: effect of age and sex. Endocrinol. 86:1361-1367. Cheever, E. V., and U. J. Lewis. 1969. Estimation of the molecular weights of the multiple components of growth hormone and pro- lactin. Endocrinol. 85:465-473. 92 93 Chomczynski, P., and Y. J. Topper. 1974. A direct effect of pro- lactin and placental lactogen on mammary epithelial nuclei. Biochem. Biophys. Res. Comm. 60:56-63. Chrambach, A. 1966. Device for sectioning of polyacrylamide gel cylinders and its use in determining biological activity in the sections. Anal. Biochem. 15:544-548. Chrambach, A., W. E. Bridson, and R. W. Turkington. 1971. Human prolactin: identification and physical characterization of the biologically active hormone by polyacrylamide gel elec- trophoresis. Biochem. Biophys. Res. Comm. 43:1296-1303. Chrambach, A., and D. Rodbard. 1971. Polyacrylamide gel electro- phoresis. Science. 1972:440-451. Costlow, M. E., R. A. Buschow, and W. L. McGuire. 1974. Prolactin receptors in an estrogen receptor-deficient mammary carcinoma. Science. 184:85-86. Croxton, F. E. 1953. Elementary Statistics. Dover Publications, Inc., New York. Cuatrecasas, P. 1969. Interaction of insulin with the cell mem- brane: the primary action of insulin. Proc. Natl. Acad. Sci. 63:450-457. Dannies, P. S., and L. Levine. 1971. The role of sulfhydryl groups in serological prOperties of bovine brain S-100 protein. J. Biol. Chem. 246:6284-6287. Dannies, P. S., and A. H. Tashjian, Jr.’ 1973. Effects of thyrotropin- releasing hormone and hydrocortisone on synthesis and degrada- tion of prolactin in a rat pituitary cell strain. J. Biol. Chem. 248:6174-6179. Davis, B. J. 1964. Disc electrophoresis--II. Method and applica- tion to human serum proteins. Ann. NY Acad. Sci. 121:404- 427. Deuben, R., and J. Meites. T965. Ip_vitro reinitiation of pituitary somatotropin release by an acid extract of hypothalamus. Proc. Soc. Exp. Biol. Med. 118:409-412. Dimond, R. C., and S. W. Rosen. 1974. Chromatographic differences between circulating and pituitary thyrotropins. J. Clin. Endocrinol. Metab. 39:316-325. Dunn, 0. J. 1964. Basic Statistics: A Primer for the Biomedical Sciences. John Wiley and Sons, Inc., New York. 94 Eagle, H. 1959. Amino acid metabolism in mammalian cell cultures. Science. 130:432-537. Ellis, s., J. M. Nuenke, and R. E. Grindeland. 1968. Identity between the growth hormone degrading activity of the pitu- itary gland and plasmin. Endocrinol. 83:1029-1042. Ellis, S., R. E. Grindeland, J. M. Nuenke, and P. K. Callahan. 1969. Purification and properties of rat prolactin. Endocrinol. 85:886-894. Evans, G. A., and M. G. Rosenfeld. T976. Cell-free synthesis of a prolactin precursor directed by mRNA from cultured rat pituitary cells. J. Biol. Chem. 251:2842-2847. Fairbanks, G., T. L. Steck, and D. F. H. Wallach. 1971. Electro- phoretic analysis of the major polypeptides of the human erythrocyte membrane. Biochem. 10:2606-2617. Farmer, S. W., H. Papkoff, S. M. Russell, M. A. Vodian, F. C. Leung, and C. S. Nicoll. 1975. Properties of prolactin and growth hormone isolated from the medium of cultured rat adenohypo- physes. Abstract 61 of the 57th Meeting of the Endocrine Society. Ferguson, K. A. 1964. Starch-gel electrophoresis-application to the classification of pituitary proteins and polypeptides. Metabolism. 13:985-1002. Forsyth, I. A., and L. Parke. 1973. Bioassay of human prolactin. IQ; Human Prolactin, edited by J. L. Pasteels and C. Robyn, 71-81. American Elsevier, New York. Frantz, A. G., 0. L. Kleinberg, and G. L. Noel. 1972. Physiological and pathological secretion of human prolactin studied by in vitro bioassay. IQ: Lactogenic Hormones, edited by G. E. W. Wolstenholme and J. Knight, 137-150. Churchill Livingstone, London. Frantz, W. L., ang R. W. Turkington. 1972. Formation of biologically active 2 I-prolactin by enzymatic radio-iodination. Endo- crinol. 91:1545-1548. Frantz, W. L., J. H. MacIndoe, and R. W. Turkington. 1974. Prolac- tin receptors: Characteristics of the particulate fraction binding activity. J. Endocr. 60:485-497. Frantz, W. L., C. Sonnenscggin, P. Payne, and 0. Dombroske. 1975. Binding Of ovine I-prolactin to cultured anterior pitu- itary tumor cells and normal cells. Nature. 255:636-638. 95 Friesen, H., G. Tolis, R. Shiu, and P. Hwang. 1973. Studies on human prolactin: chemistry, radioreceptor assay and clinical significance. 13; Human Prolactin, edited by J. L. Pasteels and C. Robyn, ll-23. American Elsevier, New York. Gala, R. R., and R. P. Reece. 1963. Influence of type and level of gas environment on anterior pituitary lactogen production jp_ vitro. Proc. Soc. Exp. Biol. Med. 114:426. Gala, R. R. 1970. Optimum culture conditions for the production of prolactin from rat anterior pituitary organ cultures. Acta Endocrinol. 65:466-476. Gala, R. R., and E. Y. H. Kuo. 1972. Prolactin production by rat anterior pituitaries cultured jfl_vitro as estimated by crop gland assay, densitometric analysis and radioimmunoassay. Proc. Soc. Exp. Biol. Med. 139:1349-1352. Gale, G. C., S. Taleisnik, H. M. Friedman, and S. M. McCann. 1961. Hormonal basis for impairment in milk synthesis and milk ejection following hypothalamic lesions. J. Endocr. 23: 303-316. Gautvik, K. M., and M. Kriz. 1976. Measurements of prolactin and growth hormone synthesis and secretion by rat pituitary cells in culture. Endocrinol. 98:344-351. Giannattasio, G., A. Zanini, and J. Meldolesi. 1975. Molecular organization of rat prolactin granules. I. Ig_vitro sta- bility of intact and "membraneless" granules. J. Cell. Biol. 64:246-251. Gliemann, J., S. Gammeltoft, and J. Vinten. 1975. Time course of insulin-receptor binding and insulin-induced lipogenesis in isolated rat fat cells. J. Biol. Chem. 250:3368-3374. Goldfine, I. 0., S. M. Amir, A. W. Petersen1 gnd S. H. Ingbar. 1974. Preparation of biologically active 2 I-TSH. Endocrinol. 95:1228-1233. Groves, W. E., and B. H. Sells. 1968. Purification of rat prolactin and growth hormone using preparative polyacrylamide gel electrophoresis. Biochim. Biophys. Acta. 168:113-121. Guyda, H. J. 1975. Heterogeneity of human growth hormone and pro- lactin secreted jp_vitro: immunoassay and receptor assay correlations. J. Clin. Endocrinol. Metab. 41:953-967. Harris, G. W. 1950. Oestrus rhythm, pseudopregnancy and the pituitary stalk in the rat. J. Physiol. 111:347-360. Harris, Hedrick, Hopkins, Hummel, Jacobs, 96 G. W. 1955. Neural Control of the Pituitary Gland. Edward Arnold Ltd., London. J. L., and A. J. Smith. 1968. Size and charge isomer separation and estimation of molecular weights of proteins by disc gel electrophoresis. Arch. Biochem. BiOphys. 126: 155-164. C. R., and M. G. Farquhar. 1973. Hormone secretion by cells dissociated from rat anterior pituitaries. J. Cell Biol. 59:276-303. B. C. W., G. M. Brown, P. Hwang, and H. G. Friesen. 1975. Human and monkey prolactin and growth hormone: separation of polymorphic forms by isoelectric focusing. Endocrinol. 97. 855- 867. L. S., and Y-C. Lee. 1975. Polymeric growth hormone and prolactin in secretory granules: sulfhydryl bonding masks immunoreactive sites. Abstract 66 of the 57th Meeting of the Endocrine Society. Jones, A. E. J. N. Fisher, U. J. Lewis, and W. P. Vanderlaan. 1965. Electrophoretic comparison of pituitary glands from male and female rats. Endocrinol. 76: 578- 583. Kelly, P. A., B. I. Posner, T. Tsushima, and H. G. Friesen. 1974. Studies of insulin, growth hormone and prolactin binding: ontogenesis, effects of sex and pregnancy. Endocrinol. 95:532-539. Kragt, C. L., and J. Meites. 1966. Separation of rat anterior pitu- Kruse, V. itary hormones by polyacrylamide gel electrophoresis. Proc. Soc. Exp. Biol. Med. 121:805-808. 1973. Binding of bovine prolactin serum proteins and its influence on the catabolism of prolactin. In: Human Pro- lactin, edited by J. L. Pasteels and C. Robyn, 51- 55. American Elsevier, New York. Kurcz, M., I. Nagy. and P. Baranyai. 1969. New biochemical micro- Kwa, H. method for the determination of prolactin in rat adenohypo- physis. Acta Biochim. Biophys. Acad. Sci. Hung, 4:287-295. G., A. A. van der Gugten, and F. Verhofstadt. 1969. Radio- immunoassay of rat prolactin: comparison of rat prolactin preparations isolated from the granular fraction of pitu- itary tumor transplants and from normal pituitary glands. Eur. J. Cancer. 5:559-569. 97 Laemmli, U. K. 1970. Cleavage of structural proteins during assem- bly of the head of bacteriophage T4. Nature. 227:680-685. Lang, R. E., H. L. Fehm, K. H. Voigt, and E. F. Pfeiffer. 1973. Two ACTH species in rat pituitary gland. FEBS Letters. 37:197-199. Laurent, T. C., and J. Killander. 1964. A theory of gel filtration and its experimental verification. J. Chromatogp, 14:317- 330. Lehmann. E. L. 1975. Nonparametrics: Statistical Methods Based on Ranks. Holden-Day, Inc., San Francisco. Lewis, U. J. 1962. Enzymatic transformations of growth hormone and prolactin. J. Biol. Chem. 237:3141-3145. Lewis, U. J., and M. 0. Clark. 1963. Preparative methods for disk electrophoresis with special reference to the isolation of pituitary hormones. Anal. Biochem. 6:303-315. Lewis, U. J., and E. V. Cheever. 1965. Evidence for two types of conversion reactions for prolactin and growth hormone. J. Biol. Chem. 240:247-252. Lewis, U. J., M. Litteria, and E. V. Cheever. 1969. Growth hormone and prolactin components of disc electrophoretic patterns: quantitative determination of the amount of protein in a stained band. Endocrinol. 85:690-697. Lewis, U. J., E. V. Cheever, and W. C. Hopkins. 1970. Kinetic study of the deamidation of growth hormone and prolactin. Biochim. Biophys. Acta. 214:498-508. Lewis, U. J., R. N. P. Singh, and B. K. Seavey. 1971. Human pro- lactin: isolation and some properties. Biochem. Biophys. Res. Comm. 44:1169-1176. Lewis, U. J., R. N. P. Singh, Y. N. Sinha, and W. P. Vanderlaan. 1971. Electrophoretic evidence for human prolactin. J. Clin. Endocrinol. Metab. 33:153-156. Lowry, 0. H., N. H. Rosenbrough, A. L. Facci, and R. J. Randall. 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193:265-275. Lukens, L. N. 1976. Time of occurrence of disulfide linking between procollagen chains. J. Biol. Chem. 251:3530-3538. 98 MacLeod, R. M., and A. Abad. 1968. On the control of prolactin and growth hormone synthesis in rat pituitary glands. Eggp- crinol. 83:799-806. MacLeod, R. M., A. Abad, and L. L. Eidson. 1969. Ip_vivo effect of sex hormones on the jp_vitro synthesis of prolactin and growth hormone in normal and pituitary tumor-bearing rats. Endocrinol. 84:1475-1483. MacLeod, R. M., and J. E. Lehmeyer. l974a. Reciprocal relationship between the jp_vitro synthesis and secretion of prolactin and growth hormone. Proc. Soc. Exp. Biol. Med. 145:1128-1131. MacLeod, R. M., and J. E. Lehmeyer. 1974b. Restoration of prolactin Maizel, Maurer, McCann, Meites, Meites, Meites, synthesis and release by the administration of monoaminergic blocking agents to pituitary tumor-bearing rats. Cancer Res. 34:345-350. ‘ J. V. 1970. Polyacrylamide gel electrophoresis of viral proteins. 1p; Methods in Virology, vol. V, edited by K. Maramorosch and H. Koprowski, 179-246. Academic Press, New York. R. A., R. Stone, and J. Gorski. 1976. Cell-free synthesis of a large translation product of prolactin messenger RNA. J. Biol. Chem. 251:2801-2807. S. M., and H. M. Friedman. 1960. The effect of hypothalamic lesions on the secretion of luteotropin. Endocrinol. 67: 597-608. J., and T. J. Hopkins. 1960. Induction of lactation and mammary growth by pituitary grafts in intact and hypophysec- tomized rats. Proc. Soc. Exp. Biol. Med. 104:263-266. J., R. H. Kahn, and C. S. Nicoll. 1961. Prolactin produc- tion by rat pituitary jp_vitro. Proc. Soc. Exp. Biol. Med. 108:440-443. J., C. S. Nicoll, and P. K. Talwalker. 1963. The central nervous system and the secretion and release of prolactin. IQ; Advances in Neuroendocrinology, edited by A. V. Nalban- dov, 238-277. Univ. of Illinois Press, Urbana. Meldolesi, J., 0. Marini. and L. L. 0. Marini. 1972. Studies on jg vitro synthesis and secretion of growth hormone and prolac- tin. I. Hormone pulse labeling with radioactive leucine. Endocrinol. 91:802-808. Miyachi, Y., J. L. Vaitukaitus, E. Nieschlag, and M. Lipsett. 1972. Enzymatic radioiodination of gonadotropins. J. Clin. Endocr. Metab. 34:23-28. 99 Neill, J. 0., and L. E. Reichert, Jr. 1971. Development of a radioimmunoassay for rat prolactin and evaluation of the NIAMD rat prolactin radioimmunoassay. Endocrinol. 88: 548-555. Neville, D. M., Jr. 1968. Isolation of an organ specific protein antigen from cell-surface membrane of rat liver. Biochim. Biophys. Acta. 154:540-552. Neville, D. M., Jr. 1971. Molecular weight determination of a protein-dodecyl sulfate complexes by gel electrophoresis in a discontinuous buffer system. J. Biol. Chem. 246:6328- 6334. Nicoll, C. S., and J. Meites. T962. Prolactin secretion jp_vitro: Comparative aspects. Nature. 195:606-607. Nicoll, C. S. 1969. Bioassay of prolactin. Analysis of the pigeon crop-sac response to systemic prolactin injection by an improved method of response quantification. Acta Endocrinol. 60:91-100. Nicoll, C. S., J. A. Parsons, R. P. Fiorondo, and C. W. Nichols, Jr. 1969. Estimation of prolactin and growth hormone levels by polyacrylamide disc electrophoresis. J. Endocr. 45:183-196. Nicoll, C. S. 1975. Radioimmunoassay and radioreceptor assays for prolactin and growth hormone: a critical appraisal. Amer. Zool. 15:881-903. Niswender, G. 0., C. L. Chen, A. R. Midgley, Jr., J. Meites, and S. Ellis. 1969. Radioimmunoassay for rat prolactin. Proc. Soc. Exp. Biol. Med. 130:793-797. Oka, T., and Y. J. Topper. T971. Insulin-Sepharose and the dynamics of insulin action. Proc. Natl. Acad. Sci. 68:2066- 2068. Orstein, L. 1964. Disc electrophoresis--I. Background and Theory. Ann. NY Acad. Sci. 121:321-349. Payne, P. A. 1975. Normal rat pituitary cells: study of primary culture, secretion of prolactin, and binding of prolactin-125 iodine. M.S. thesis, Michigan State University, East Lansing. Peckham, W. 0., and E. Knobil. l976a. The effects of ovariectomy, estrogen replacement, and neuraminidase treatment on the properties of the adenohypophysial glycoprotein hormones of the Rhesus monkey. Endocrinol. 98:1054-1060. 100 Peckham, W. 0., and E. Knobil. l976b. Qualitative changes in the pituitary gonadotropins of the male Rhesus monkey following castration. Endocrinol. 98:1061-1064. Pedersen, K. 0. 1962. Exclusion chromatography. Archiv. Biochem. Biophys. Supplement 1:157-168. Porath, J., and P. Flodin. T959. Gel filtration: a method for desalting and group separation. Nature. 183:1657-1659. Rae, P. A., and B. P. Schimmer. T974. Iodinated deriva ives of adrenocorticotropic hormone. Preparation of 1'2 -a-ACTH with full hormonal activity. J. Biol. Chem. 249:5649-5663. Reisfeld, R. A., A. S. Muccilli, D. E. Williams, and S. L. Steelman. 1964. Preparation of rat growth hormone. Nature. 201: 821-823. Robertson, M. C., and H. G. Friesen. 1975. The purification and characterization of rat placental lactogen. Endocrinol. 97:621-629. Rodbard, D., P. L. Rayford, J. A. Cooper, and G. T. Ross. 1968. Statistical control of radioimmunoassays. J. Clin. Endocr. Metab. 28:1412-1418. Rodbard, 0., and A. Chrambach. 1970. Unified theory for gel elec- trophoresis and gel filtration. Proc. Natl. Acad. Sci. 65:970-977. Rogol, A. 0., M. Ben-David, R. Sheats, D. Rodbard, and A. Chrambach. 1975. Charge properties of human pituitary and amniotic fluid prolactins. Endocr. Res. Comm. 2:379-402. Rogol, A. 0., and S. W. Rosen. l974a. Alteration of human and bovine prolactins by a chloramine T radioiodination: comparison with lactoperoxidase-iodinated prolactins. J. Clin. Endocr. Metab. 39:379-382. Rogol, A. 0., and S. W. Rosen. l974b. Prolactin of apparent large molecular size: the major immunoactive prolactin component in plasma of a patient with a pituitary tumor. J. Clin. Endocr. Metab. 38:714-717. Rogol, A. 0., and H. Chrambach. T975. Radioiodinated human pituitary and amniotic fluid prolactins with preserved molecular integ- rity. Endocrinol. 97:406-417. Rohde, W., and G. DOrner. 1969. Immunochemical studies on the heterogeneity of human growth hormone. Acta Endocrinol. 60:101-111. 101 Sakai, S., K. Kohmoto, and T. Johke. 1976. A receptor site for prolactin in lactating mouse mammary tissue. T975. Endo- crinol. Japon. 22:379-387. Samli, M. H., M. F. Lai, and C. A. Barnett. 1972. Protein synthesis in the rat anterior pituitary: II. Solubility studies on total protein, growth hormone and prolactin labeled in an jg_vitro incubation. Endocrinol. 91:227-232. Shapiro, A. L., E. Vinuela, and J. V. Maizel. 1967. Molecular weight estimation of polypeptide chains by electrophoresis in SOS-polyacrylamide gels. Biochem. Biophys. Res. Comm. 28:815-820. Shiu, R. P. C., P. A. Kelly, and H. G. Friesen, 1973. Radio- receptor assay for prolactin and other lactogenic hormones. Science. 180:968-971. Shiu, R. P. C., and H. G. Friesen. T974a. Properties of a prolactin receptor from the rabbit mammary gland. Biochem. J. 140: 301-311. Shiu, R. P. C., and H. G. Friesen. 1974b. Solubilization and puri- fication of a receptor from the rabbit mammary gland. J. Biol. Chem. 249:7902-7911. Singh, R. N. P., B. K. Seavey, and U. J. Lewis. 1974. Heterogeneity of human growth hormone. Endocr. Res. Comm. 1:449-464. Sluyser, M., and C. H. Li. 1964. Studies of pituitary lactogenic hormone. XXIII. Isolation of the monomer of ovine prolac- tin. Archiv. Biochem. Biophys. 104:50-57. Squire, P. G., B. Starman, and C. H. Li. 1963. Studies of pitu- itary lactogenic hormone. XXII. Analysis of the state of aggregation of the ovine hormone by ultracentrifugation and exclusion chromatography. J. Biol. Chem. 238:1389-1395. Squire, P. G. 1964. A relationship between the molecular weights of macromolecules and their eTution volumes based on a model for Sephadex gel filtration. Archiv. Biochem. Biophys. 107:171-178. Stachura, M. E., and L. A. Frohman. 1974. "Large" growth hormone: ribonucleic acid-associated precursor of other growth hormone forms in rat pituitary. Endocrinol. 94:701-712. Stachura, M. E., and L. A. Frohman. 1975. Growth hormone: inde- pendent release of big and small forms from rat pituitary ip_vitro. Science. 187:447-449. 102 Sud, S. C. 1971. Suppression of prolactin release from adeno- hypophysis by prolactin: an jp_vitro study. Ind. J. Exp. Biol. 9:260-261. Suh, H. K., and A. G. Frantz. 1974. Size heterogeneity of human prolactin in plasma and pituitary extracts. J. Clin. Endocr. Metab. 39:928-935. Sussman, P. M., R. J. Tushinski, and F. C. Bancroft. 1976. Pre- growth hormone: product of the translation jp_vitro of messenger RNA coding for growth hormone. Proc. Natl. Acad. Sci. 73:29-33. Swearingen, K. C. 1971. Heterogeneous turnover of adenohypophysial 'prolactin. Endocrinol. 89:1380-1388. Talwalker, P. K., A. Ratner, and J. Meites. 1963. Ip_vitro inhibi- bition of pituitary prolactin synthesis and release by hypothalamic extract. Amer. J. Physiol. 205:213-218. Tashjian, A. H., Jr., L. Levine, and A. E. Wilhelmi. 1965. Immuno- chemical studies with antisera to fractions of human growth hormone which are high or low in pigeon-crop gland-stimulat- ing activity. Endocrinol. 77:1023-1036. Tishler, P. V., and C. J. Epstein. 1968. A convenient method of preparing polyacrylamide gels for liquid scintillation spectrometry. Anal. Biochem. 22:89-98. Turkington, R. W. 1970. Stimulation of RNA synthesis in isolated mammary cells by insulin and prolactin bound to Sepharose. Biochem. Biophys. Res. Comm. 41:1362-1367. Turkington, R. W. 1972. Measurement of prolactin activity in human serum by the induction of specific milk proteins jp_ vitro: results in various clinical disorders. lg; Lacto- genic Hormones, edited by G. E. W. Wolstenholme and J. Knight, 169-196. Churchill Livingstone, London. Turkington, R. W. 1974. Prolactin receptors in mammary carcinoma cells. Cancer Res. 34:758-763. Wallevik, K. 1976. Spontaneous ig_vivo isomerization of bovine serum albumin as a determinant of its normal catabolism. J. Clin. Invest. 57:398-407. Weber, K., and M. Osborn. 1969. The reliability of molecular weight determinations by dodecyl sulfate-polyacrylamide gel electrophoresis. J. Biol. Chem. 44:4406-4412. 103 Williams, 0. E., and R. A. Reisfeld. 1964. Disc electrophoresis in polyacrylamide gels: extension to new conditions of pH and buffer. Ann. NY Acad. Sci. 121:373-381. Yalow, R. S. 1974. Heterogeneity of peptide hormones. Rec. Prog. Horm. Res. 30:597-633. Yamamoto, K., L. M. Taylor, and F. E. Cole. 1970. Syntheses and release of GH and prolactin from the rat anterior pituitary jg_vitro as functions of age and sex. Endocrinol. 87:21-26. Young, R. 11., and H. w. Fu1horst. 1965. Recovery of535 radio- activity from protein-bearing polyacrylamide gel. Anal. Biochem. 11:389-391. Zanini, A., and G. Giannattasio. 1972. Polyacrylamide gel electro- phoresis of rat anterior pituitary gland after different extraction procedures. J. Endocr. 53:177-178. Zanini, A., and G. Giannattasio. 1973. Isolation of prolactin granules from rat anterior pituitary glands. Endocrinol. 92:349-356. Zanini, A., G. Giannattasio, and J. Meldolesi. l974a. Studies on jp_vitro synthesis and secretion of gorwth hormone and prolactin. 11. Evidence against the existence of precursor molecules. Endocrinol. 94:104-111. Zanini, A., G. Giannattasio, and J. Meldolesi. l974b. Separation of rat pituitary growth hormone and prolactin by $05 poly- acrylamide gel electrophoresis. Endocrinol. 94:594-598. MICHIGAN STATE UNIVER 7 3 1293 030 ITY Ll I "11111158 4 2 S 151