UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Aspects of neurohypophysial physiology during fetal development and pregnancy in the fur seal Callorhinus… Vizsolyi, Elizabeth 1972

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1972_A1 V59.pdf [ 10.42MB ]
Metadata
JSON: 831-1.0101469.json
JSON-LD: 831-1.0101469-ld.json
RDF/XML (Pretty): 831-1.0101469-rdf.xml
RDF/JSON: 831-1.0101469-rdf.json
Turtle: 831-1.0101469-turtle.txt
N-Triples: 831-1.0101469-rdf-ntriples.txt
Original Record: 831-1.0101469-source.json
Full Text
831-1.0101469-fulltext.txt
Citation
831-1.0101469.ris

Full Text

ASPECTS OF NEUROHYPOPHYSIAL PHYSIOLOGY DURING FETAL DEVELOPMENT AND PREGNANCY IN THE FUR SEAL CALLORHINUS URSINUS  by  ELIZABETH VIZSOLYI B.Sc. Eotvos Lorand University, Budapest, 1954. M.Sc. University of British Columbia, 1968.  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY  in the Department of Zoology  We accept this thesis as confirming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA May, 1972  In  presenting  an  advanced degree at  the  Library  I further for  this  shall  agree  scholarly  by  his  of  this  written  thesis the  fulfilment of  University  of  make i t f r e e l y that permission  p u r p o s e s may  representatives. thesis  in p a r t i a l  for  be  available  granted  gain  ^ D Q [O  The U n i v e r s i t y o f B r i t i s h V a n c o u v e r 8, C a n a d a  by  the  It i s understood  financial  for  for extensive  permission.  Department o f  British  Columbia  shall  requirements  Columbia,  Head o f my  be  I agree  r e f e r e n c e and copying of  that  not  the  that  study.  this  thesis  Department  copying or  for  or  publication  allowed without  my  ABSTRACT  The neurohypophyses of the fur seal, Callorhinus ursinus were studied in the control non-breeding animals, in the pregnant females at five stages of gestation, and in their fetuses at seven stages of intrauterine l i f e . fur  The neurohypophyses of the  seal, a marine mammal, were found to be exceptionally potent  in rat uterus and vasopressor activities.  Purification procedures  and amino acid analyses have shown the active neurohypophysial peptides of the seal, in common with the majority of mammalian species, to be arginine vasopressin and oxytocin. The neurohypophysial hormone levels of the pregnant females were found to be fluctuating from the control level with the advancement of gestation.  The vasopressor activity  showed an i n i t i a l rise between 0.19 and 0.31 of term, but during the remainder of pregnancy i t was found to f a l l below the control values.  The oxytocic activities were found to  be elevated from the control values a l l through pregnancy, but a fluctuation in the potencies was detected, which parallelled that found for vasopressor activities. Neurosecretory activity in the fetal neurohypophysial system was investigated at seven stages of gestation by histological, pharmacological and biochemical methods.  The  histological studies have shown the two hypothalamic nuclei, the supra-optic and the para-ventricular nucleus, as well as the neurohypophysis  to contain neurosecretory material at 0.5  ii  of term, or mid-gestation.  However, biological activities as  measured by rat uterus, vasopressor and frog bladder assays were detected at a much earlier stage, at 0.19 of term.  In the  vasopressor and rat uterus activities a steady accumulation of activities was recorded with the advancement of gestation, but in the third biological activity, the potency on the isolated frog bladder, the highest value was recorded at 0.68 of term, and declined from there until term. Since the neurohypophysial hormones of the fetal neurohypophysis  have not been previously purified, the crude  posterior pituitary extracts of fetuses were subjected to gel f i l t r a t i o n and ion exchange chromatography.  The pharmacological  characteristics of the purified fetal principles, when assayed against synthetic standards, suggested that they were arginine vasopressin, oxytocin and arginine vasotocin.  These findings  were confirmed by amino acid analyses of the three active neurohypophysial peptides, and i t was concluded that the fetal neurohypophysis  contains arginine vasotocin, the antidiuretic  hormone of sub-mammalian vertebrates, in addition to oxytocin and arginine vasopressin, the commonly occurring mammalian peptides. Preliminary experiments suggested that the embryonic membranes might constitute the possible target organs of the fetal neurohypophysial agents.  TABLE OF CONTENTS Pag ABSTRACT  1  LIST OF TABLES  vi  LIST OF PLATES  ix  LIST OF FIGURES  x  ABBREVIATIONS USED  xi  ACKNOWLEDGEMENTS  xi  INTRODUCTION 1. General Background 2. The neurohypophysis during pregnancy 3. The neurohypophysis during fetal development  1 ... 1 9 11  STATEMENT OF THE PROBLEM  15  GENERAL MATERIALS AND METHODS  18  A. Materials  18  1. Collection procedure  18  2. Lyophilisation and storage  19  3.  19  Extraction  B. The Estimation of Biological Activities 1.  Isolated rat uterus assay  20 20  2. Rat vasopressor assay  23  3.  Frog waterbalance assay  24  4.  Rat antidiuretic assay  26  5. Rat milk ejection assay  29  6.  30  Standards  iv  C.  The Calculation of Potencies  30  D.  Purification Methods  32  E.  1.  Paperchromatography  32  2.  Gel f i l t r a t i o n  34  3.  Ion exchange chromatography  36  Analytical Methods  39  1.  pH and conductivity measurements  39  2.  Preparation of samples for hydrolysis  40  3.  Hydrolysis  41  4. Amino acid analysis SECTION I.  41  THE NEUROHYPOPHYSIAL HORMONES OF THE ADULT PREGNANT SEAL  42  Introduction  42  A.  45  B.  Biological Activities 1.  Control non-breeding animals  2.  Biological activities during pregnancy ....  Purification of the Active Agents  45 ...47 52  a.  The purification of the vasopressor fraction . 52  b.  Purification of the oxytocic agent  Discussion  56 62  a.  Environmental considerations  62  b.  The neurohypophysial hormones during pregnancy 66  c.  The purification of the neurohypophysial hormones  69  V  SECTION II.  THE NEUROHYPOPHYSIS DURING FETAL DEVELOPMENT  A.  Histochemical Studies Introduction  76  2.  Materials and Methods  79  3.  Results  .80  a.  Gross anatomy  80  b.  The hypothalamic nuclei  83  i . Supra-optic nucleus  83  i i . Para-ventricular nucleus  85  The pars nervosa  86  4. Discussion  C.  ... 76  1.  c.  B.  ...74  The Biological Activities of the Fetal Seal Neurohypophysis 1.  Introduction  2.  Results  3.  Discussion  87 99 99 101 ...106  The Purification of the Active Agents from the Fetal Neurohypophysis  Ill  Introduction  Ill  Results  117  1. 2.  Paper chromatography of the crude extracts ...118 Gel f i l t r a t i o n of the neurohypophysial extracts 121  3.  Ion exchange chromatography of oxytocin and arginine vasopressin from fetal seals  121  a-.  Purification of the oxytocic principle  126  b.  Purification of vasopressor peak 2b on Phosphocellulose  128  4. The separation of arginine vasotocin from arginine vasopressin a.  Preliminary experiments with Dowex 50-X2  132  b.  Preliminary experiments with IRC-50  134  c.  The separation of the two basic peptides from fetal extracts on IRC-50 columns  136  Discussion 5.  130  Identification of the fetal peptides by their pharmacological properties and amino acid content a. b.  Pharmacological characterisation of the active agents Amino acid content of the three fetal peptides  139  143 ...143 146  Discussion  149  a.  The structure of arginine vasotocin  151  b.  The structure of the fetal arginine vasopressin  153  c.  The structure of fetal oxytocin  153  Action of the Neurohypophysial Peptides on the Embryonic Membranes  154  1.  154  Introduction  2. Method  154  3.  Results  155  a.  The allantoic membrane  159  b.  The amniotic membrane  160  vii  4. E.  Discussion  Discussion  162 .  1. Neurohypophysial function in fetal seal  164 164  2. The importance of arginine vasotocin in mammalian fetuses  171  SUMMARY  175  LITERATURE QUOTED  178  APPENDIX A  187  APPENDIX B  194  APPENDIX C  196  vi i i  LIST OF TABLES Table I II III IV V VI  VII  VIII  IX X  Page Biological activities of the adult and pregnant seals  44  Amino acid analysis of seal neurohypophysial peptides, purified on Sephadex G-15  55  Amino acid analysis of seal oxytocin from CM Sephadex .  61  Biological activities in some mammalian species  63  Ratios of biological activities in some mammalian species ....  64  The percentage and diameter of cells in the hypothalamus nuclei of seal fetuses and of the newborn pup with neurosecretory material in the cytoplasm  81  The biological activities of the seal neurohypophysis at seven stages of gestation  .100  Biological activities of the two vasopressor peaks against arginine vasotocin and arginine vasopressin  145  Amino acid content of the neurohypophysial peptides of seal fetuses  147  Ratios of biological activities of neurohypophysial peptides  .150  ix  LIST OF PLATES Plate I  II III IV  V  VI  VII  Facing page Sagittal section through the hypothalamus and the pituitary of a seal fetus at 0.25 of term  92  Cross section of the pituitary of the seal fetus  93  Supra-optic nucleus of seal fetuses and newborn seal  94  View of the para-ventricular area in the seal fetus at three different stages of development  95  Para-ventricular nucleus of seal fetuses and newborn seal at three stages of development  .96  Comparison of the appearance of the paraventricular nucleus with two different staining techniques  97  The pars nervosa of seal fetuses and newborn seal  98  X  LIST OF FIGURES Figure 1 2  Facing page The biological activities of the adult seal neurohypophyses  45  The ratios of vasopressor to oxytocic activities (V/0) in the neurohypophyses of the adult seal .  48  3  The biological activities of the seal neurohypophysis during pregnancy  4  Ratios of vasopressor to oxytocic activities in the neurohypophyses of pregnant seal  .  50 51  5  Gel f i l t r a t i o n of the neurohypophysial extracts of the adult seal on G-15 Sephadex .... 53  6  Gel f i l t r a t i o n of the neurohypophysial extracts of pregnant seal on G-15 Sephadex  57  The purification of the oxytocic fraction of adult seal neurohypophyses on CM Sephadex  59  Diagram of the hypothalamo-hypophysial system of the seal fetus  91  7 8 9  Average dry weight of fetal posterior pituitary glands through gestation ...102  10  Biological activities of the fetal neurohypophysis per mg dry tissue  102  11  Ratios of vasopressor to oxytocic (V/RU) and frog bladder to oxytocic (FB/RU) activities of the fetal seal neurohypophyses  105  12  The paper chromatography of fetal seal neurohypophysial extracts at 0.31 of term  13  The gel f i l t r a t i o n of crude neurohypophysial extracts of seal fetuses at 0.56 to 0.68 of term  122  The biological activities in the three peaks eluted from the G-15 Sephadex column ....  124  14  ..119  xi  Figure 15  Facing page The gel f i l t r a t i o n of crude neurohypophysial extracts of seal fetuses at 0.56-0.68 of term  127  16  Purification of the oxytocic fractions on CM Cellulose column  129  17  The purification of vasopressor peak 2b from gel f i l t r a t i o n , on Phosphocellulose column  131  Separation of arginine vasopressin from arginine vasotocin on Dowex 50 X-2 and Amberlite IRC-50 resins  133  9  18  19  Separation of arginine vasotocin from arginine vasopressin on Amberlite IRC-50 resin...137  20  Summary of the purification procedures used for the isolation of the fetal neurohypophysial peptides  140  The response of the guinea pig allantoic membrane to neurohypophysial hormones  158  The response of the guinea pig amniotic membrane to aneurohypophysial hormones  161  The correlation between hypothalamic production of neurohypophysial hormones and the detected biological activities in the gland during fetal development  170  24  Resolution of the oxytocic and vasopressor activities on Sephadex G-l5 columns  188  25  Separation of the major protein fraction from the biologically active peptides on Sephadex G-l5  190  Separation of arginine vasopressin, arginine vasotocin and oxytocin on pyridine washed G-25 Sephadex column  192  21 22 23  26  XI l  a.  DEFINITIONS AND ABBREVIATIONS  hypothalamo-hypophysial  system  is used throughout the manuscript  to include the hypothalamic nuclei, the neurohypophysial stalk, and the pars nervosa. neurohypophysial system  is used interchangeably with hypothalamo-  hypophysial system. neurohypophysis or posterior pituitary  includes the pars nervosa  and the stalk. pars nervosa  the neural component of the pituitary, the place of  storage for the neurohypophysial peptides. oxytocic activity,  the potency on the isolated rat uterus,  sometimes also referred.to as uterotonic activity. vasopressor activity,  the potency on the rat vasopressor  preparation. oxytocin and vasopressin  refer to the neurohypophysial  octapeptides. RU  =  rat uterus activity  V  = vasopressor activity  ADH  =  ME  = milk ejection activity  FB  = frog bladder activity  antidiuretic activity  DEFINITIONS AND ABBREVIATIONS CONT.  Ala  alanine  Arg  arginine  Asp  aspartic acid  Cys or %Cys  cystine  Cys A.  cysteic acid  Glu  glutamine  Gly  glycine  His  histidine  lieu  isoleucine  Leu  leucine  Lys  lysine  Met  methionine  Pro  proline  Phe  phenylalanine  Ser  serine  Tyr  tyrosine  Thr  threonine  Val  valine  xi i i  ACKNOWLEDGEMENTS  I would like to express my sincere gratitude to a l l the people who aided me throughout this work.  I am especially indebted  to my supervisor, Dr. A.M. Perks, for his advice and interest during these investigations, and for his help in the preparation of the final manuscript.  The material was made available through the  courtesy of Mr. Ian McAskie and Dr. M. Bigg of the Biological Station, Nanaimo, and of Mr. Ford Wilke . and Dr. G. Harry, Jr., of the Marine Mammal Research Laboratory in Seattle.  The co-operation and interest  of the entire crew of the M.V. Bellina, and especially of Jack Henderson, Eric Lippman and Chris Whiting was invaluable in collecting the pituitaries. The histological work was carried out by Miss Daphne Hards, whose knowledge of the f i e l d , and interest in the problem were most valuable. For financial support, I would like to thank The H.R. MacMillan family for a Family Fellowship for the years 1968 - 1971. was supported by N.R.C. Grant No. A 2584.  The research  1.  INTRODUCTION  1.  General background The basis of the morphological and functional aspects of the  mammalian hypothalamo-hypophysial early 1950's.  system was established in the  It was found to consist of three main components.  F i r s t l y , there were the neurosecretory cell bodies of the supraoptic and paraventricular nuclei of the hypothalamus. there was the hypothalamo-hypophysial  Secondly,  tract, which was made up of  axons leading from the hypothalamic nuclei to the pars nervosa. Thirdly, and lastly, there was the pars nervosa i t s e l f .  Scharrer  (1940, 1954) and Bargmann (1951) used the chrome-alum haematoxylin phloxin stain of Gomori to show that the pars nervosa was only a storage organ for neurosecretory granules; these granules were elaborated in the supra-optic and the paraventricular nuclei of the hypothalamus, and transported to the pars nervosa by the axons of the hypophysial tract.  The experiments of Sachs and Takabatka (1964)  have indicated that a vasopressin precursor is produced on ribosomal RNA, withis the cell bodies of the hypothalamic nuclei; i t is then incorporated into granules formed by the nearby Golgi apparatus. The newly synthesised vasopressin is released front a precursor either outside the granules, or within the granules, during the axonal transport to the pars nervosa, and i t becomes associated with a new carrier protein, termed neurophysine (Ginsbusng, 1968).  It is  assumed that oxytocin synthesis takes place in a similar manner, although there is no direct experimental data concerning the  2.  synthesis of this peptide (Ginsburg, 1968).  It is probable that  the neurosecretory granules are transported by axoplasmic flow to the nerve endings in the pars nervosa, from which they are released into the bloodstream upon physiological demand. Although our understanding of the morphological arrangement of the hypothalamo-hypophysial  system and of the synthesis of i t s  hormones are relatively recent, the primary pharmacological actions of the neurohypophysial extracts have been known since the turn of the century.  In 1895, Oliver and Shafer reported that injections  of posterior pituitary extracts caused an elevation of blood pressure, and in 1905, Dale recognized that they possessed uterotonic activity.  In 1910 Ott and Scott found that lactating  mammary glands responded to neurohypophysial extracts with letdown of milk.  In the next three years three independent groups of  workers demonstrated antidiuretic activity of the neurohypophysis (Frank, 1912; Farmi, 1913; and Von den Velden, 1913).  Finally, in  1918 Hann pointed out the correlation between injuries of the neurohypophysis and the human disease diabetes insipidus, in which urine flow was profuse; this indicated that the major physiological importance of the gland lay in i t s control of the kidney.  Although  the extensive research which followed these discoveries added a great deal to our understanding of neurohypophysial physiology, and although numerous secondary effects of the neurohypophysial peptides have now been described, those actions which were discovered nearly six decades ago are s t i l l considered to represent the primary actions of the neurohypophysis.  3.  Following the discovery of the biological effects of the neurohypophysis, the isolation and identification of i t s active agents became the center of extensive investigations. In 1928, Kamm and his co-workers isolated two biologically active agents from neurohypophysial  extracts, two agents which differed from one  another in their chemical as well as their  pharmacological  properties (Kamm, Aldrich, Grote, Rowe and Bugbee, 1928).  However,  in 1940, van Dyke and his associates isolated one large, apparently pure protein from ox pituitaries, a protein which exhibited a l l the biological actions attributed to the neurohypophysis.  The question  concerning the existance of one or of two hormones within the neurohypophysis became the subject of a great controversy, which was finally resolved by du Vigneaud and his associates in 1953. In confirmation of the original findings of Kamm and his co-workers, they succeeded in isolating and synthetising two cyclic octapeptides: oxytocin and vasopressin.  The major portion of the van Dyke protein,  was found to be a biologically inert molecule, neurophysin,  important  in the binding and transportation of the active peptides (see Acher, 1966).  Oxytocin and vasopressin themselves share a molecular weight  of approximately  1,000, and differ from one another in only two of  their eight amino acid residues.  The structures of the two peptides  are shown below: OXYTOCIN Cys-Tyr-neu-G1u(NH )-Asp(NH )-Cys-Pro-Leu-Gly(NH ) 2  2  2  4.  ARGININE VASOPRESSIN  r  ~\  Cys-Tyr-Fjie-Gl u (NH )-Asp-(NH ) Cys-Pro-Arg_-G1 y (NH ) 2  2  2  Of these two peptides, oxytocin was found to be the principle primarily responsible for the uterotonic and milk-ejection activities of the crude neurohypophysial Bisset, 1967).  extracts (Munsick, 1968;  Although i t e l i c i t s rhythmic contractions of the  uterus during parturition, normal delivery can proceed in hypophysectomised mammals, and consequently i t s contribution to labour is not considered vital (Fitzpatrick, 1966).  However, the clinical  importance of oxytocin in the acceleration of parturition resulted in extensive investigations of i t s secondary effects on the circulatory system, and in particular on the rate and the contractility of the heart (Saameli, 1968).  It is also proposed  that oxytocin promotes renal excretion of Na during antidiuresis +  (Pickford, 1960, 1964).  In addition, there is evidence that after  the administration of oxytocin, blood glucose levels rise, whilst blood free fatty acid concentrations fall (Mirsky, 1968). The pathways by which these latter actions of oxytocin are elicited are poorly understood, even when the dose levels required for responses are within the physiological range.  In most cases,  however, the large doses used make the physiological significance of many of these secondary effects questionable.  Finally, studies  of the physiology of oxytocin have been extended to the male, where oxytocin has been attributed with a role in the transport of sperm (Fitzpatrick, 1966).  5.  As with oxytocin, vasopressin has been extensively investigated since i t s chemical identification.  Its best-known  action, the constriction of blood vessels, is s t i l l commonly used to detect and assay the peptide in posterior pituitary extracts, but i t is considered to be an essentially pharmacological effect (Thorn, 1968).  The main physiological significance of vasopressin  lies in i t s potent antidiuretic effect; i t promotes water reabsorption by a direct action on the distal tubules and collecting ducts of the kidney (see Pickford, 1966).  Besides its  action on kidney membranes, i t is attributed with the capacity to act in a similar manner on membranes of other organs.  Wakim  (1966) has summarized the evidence for some of the extrarenal actions elicited by the peptide: after injections of vasopressin, there is a marked f a l l in the loss of fluid from the pancreas, the liver and the sweat glands.  Water absorption through the  isolated frog skin is also increased in the presence of this peptide (Ussing and Anderson, 1957).  The accumulated data show  that vasopressin acts on widely different membranes to produce the same final result - water conservation. The wide scope of the investigations of the neurohypophysial peptides was largely the result of the biochemical identification of the principles, with the subsequent availability of synthetic peptides.  However, in addition to their contribution to our  understanding of mammalian neurohypophysial physiology, these biochemical achievements also furnished the sophisticated tools needed for another line of research; in the 1960's, investigations of the comparative aspects of the neurohypophysis came into  6.  increasing prominence.  These studies, although far from complete,  have now demonstrated the presence of active neurohypophysial agents in a l l classes of vertebrates studied, down to the most primitive alive today, the Cyclostomes (Sawyer, 1967).  However,  i t became clear from these studies, that several evolutionary mutations had produced analogues of the mammalian peptides discussed above.  At present, i t seems that most of these  mutations occurred in the neutral neurohypophysial as oxytocin.  peptides, such  Three of these, 4-ser 8-ileu oxytocin (teleost),  8-ileu oxytocin (amphibian).and 4-ser 8-glu oxytocin (elasmobranch) have been identified to date.  The growing number of neurohypo-  physial peptides described, together with the demonstration that the synthesis of these hormones involved ribosomal RNA,  - so that,  in consequence, they were under direct genetic control (Sachs, 1967) - resulted in several schemes to explain the evolutionary sequence of these agents (Acher, 1963, Follett and Heller, 1964, Sawyer, 1961a, 1965b, VIiegenhart;. > Pickering and Heller, 1969).  and Versteeg, 1967 and  However, with the examination of new  species, each of these have become obsolete. In contrast to the number of neutral analogues, the basic peptide 3 ileu, 8 arg oxytocin, the antidiuretic principle, usually termed arginine vasotocin, has been remarkably stable throughout evolution.  Its presence has been demonstrated in a l l  non-mammalian vertebrates so far studied.  Only in the mammals has  arginine vasotocin been replaced by 3-Phe 8-Arg oxytocin and 3-Phe 8-Lys oxytocin, or as they are commonly termed, arginine and  7.  lysine vasopressin, respectively. Alongside the biochemical and pharmacological studies of the various purified neurohypophysial  peptides, considerable  progress has been made in elucidating the state of the hormones within the gland.  The neurohypophysial  hormones are stored  within the gland in association with a large protein moiety (approximate MW 25,000 in cattle, Heller and Ginsburg, 1966). This protein.is termed neurophysin.  The octapeptides are bound  to this carrier by an ionic bond which is formed between the free ionic group of the hemicystinyl residue of the peptide and the carboxyl groups in the protein (Ginsburg, 1968).  However, to  account for the high specificity of the carrier for the neurohypophysial peptides, a secondary binding is postulated to take place by disulfide interchanges or lypophylic interactions with other sites (Ginsburg, 1968). that the two neurohypophysial  Until recently i t was believed peptides had a common carrier  protein, but in 1968 Ginsburg succeeded in separating two components of neurophysin, one binding oxytocin alone, and the other apparently capable of adsorbing both oxytocin and vasopressin. Recently, this concept has been further modified by Ginsburg, who has suggested that the two components are specific for one or other of the two peptides, respectively (personal communication, A. M. Perks). In the early stages of the investigations of neurophysin, the ratio of the two hormones (V/0 ratio) within the gland i t s e l f , and the ratio of their release into the bloodstream, were believed  8.  to be strong supporting evidence for the existence of a common carrier (Acher, 1966).  Indeed, in most mammals the V/0 ratio  approximates to one, both in terms of biological activities, and of molecular content.  However, two groups of animals proved to  be exceptions to this observation.  The f i r s t group comprised  the marsupials, which represented the most primitive living order of mammals; they showed a V/0 ratio which was always in favour of vasopressin, although the range of values extended from a low value of 2.9 in the American opossum to a high figure of 6.2 in the Australian opossum (Heller, 1966).  The second group  of mammals in which the V/0 ratio was greatly in favour of vasopressin included species of widely different taxonomic groups, which nevertheless had in common.an environment restricted in i t s fresh water supply.  For example, the V/0 ratio of the camel is  3.6, that of the llama 3.1, and in the three species of whales studied, the ratio ranges from 3.4 to 12.5 (Heller and Ferguson, 1965, Waring and Langrebe, 1950). It is worth noting that not only is the storage ratio of the two hormones approximately unity, but also the release into the blood occurs in approximately the same ratio, in most instances;  this is independent of the physiological need  (Ginsburg, 1968).  Although a preferential release of vasopressin  was found as a result of haemorrhage in anaesthetised rats, or because of severe dehydration in dogs and rats, (Heller, 1961, Pickford, 1964), neither of these stimuli could be termed "normal".  9.  So far, only one physiological condition, the nursing of the young has been reported to result in a preferential release of oxytocin. However, even prolonged release of one component leaves the V/0 ratio in the gland itself unchanged, as has been shown in the case of 5-10 days of preferential depletion of oxytocin by nursing. This suggests that there is a balance between the circulatory release and the hypothalamic production of the active agents (Ginsburg, 1968). In spite of the advances made in the more sophisticated fields of neurohypophysial physiology, such as those concerning the mechanism of synthesis and action of the hormones, there is s t i l l relatively l i t t l e known of neurohypophysial function under special physiological conditions, such as pregnancy or the embryological development of the system.  These aspects, since  they are the object of the studies presented here, will be discussed in more detail below. 2.  The neurohypophysis during pregnancy The primary endocrine changes associated with pregnancy are  the elevated estrogen and progesterone levels of the blood (Marshall, 1962).  Both of these sex steroids markedly alter the  responsiveness of the target organs to the neurohypophysial peptides, and possibly affect the release of these agents from the hypothalamic nuclei (Munsick, 1968; Farrell et a l , 1965).  In  general, estrogen treatment results in regular spontaneous contractions of the uterine muscle, while high progesterone levels,  10.  in most species, block uterine responses (Csapo, 1961; Fitzpatrick, 1966).  In addition, sex steroids have an action on  the target organs of vasopressin. Cobo (1967) reported a high threshold for antidiuretic responses to vasopressin in pregnant women, and Pickford (1966) found that the administration of progesterone produced a rise in threshold for vasopressin induced salt excretion and vascular responses.  Of even greater  interest, i t was reported, that progesterone blocks the responses of the hypothalamic nuclei themselves to genital stimulation in the rat, so that the region of peptide synthesis also appear to be affected by sex steroids (Labsethwar et a l , 1964). If consideration is given to the influence of these sex steroids on both the target organs of the neurohypophysial hormones and on the hypothalamic nuclei themselves, i t is not surprising that Heller (1957) found fluctuations in the levels of stored neurohypophysial hormones at different stages of the reproductive cycle.  In the rat, he reported a rise in the levels  of both hormones present in the pituitary during the follicular phase of the cycle, and a f a l l during metestrous.  In the same  species, the diuretic response to waterloading was also impaired immediately after the cessation of estrus quoted by Heller, 1961).  (Ginsburg, 1950, as  However, both Acher and his co-workers  (1956) and Heller and Lederis (1959) failed to detect similar changes in rats during pregnancy, when changes in the estrogen and progesterone levels were expected to be more pronounced. Only in the sheep was the neurohypophysial hormone level found to  11.  be depressed during pregnancy;  significant changes in the extent  of this depression for both oxytocin and vasopressin could be correlated with the stage of gestation (Vizsolyi, 1968).  Clearly,  further investigations are needed in more species and at more stages of gestation to assess the overall rate of neurohypophysial function during pregnancy. 3.  The neurohypophysis during fetal development The histochemical appearance of the  hypothalamo-hypophysial  system is the best studied aspect of the otherwise poorly investigated fetal neurohypophysis;  morphological changes have  been described in several species (see Yakovleva, 1965).  The  accumulated data indicates that in some species neurosecretory material could be detected f i r s t in the neurohypophysis, and in others i t appeared f i r s t in the hypothalamic nuclei;  however the  supra-optic nucleus'consistently showed a positive histochemical reaction at a somewhat earlier stage than the paraventricular nucleus (Yakovleva, 1965).  A more detailed discussion of this  topic will be presented with the histological studies (see p. 72 ) There is very l i t t l e information concerning therbiological activities of the fetal neurohypophysis, or demonstrating the nature of its active agents.  However, one striking feature of  the fetal neurohypophysis emerges clearly from the limited evidence;  in early embryonic development the V/0 ratio within the  gland is always many times in favour of vasopressin, and the closer the fetus gets to birth, the nearer this ratio approaches  12.  unity. In 1953, the vasopressor and oxytocic activities were studied by Dicker and Tyler, in human, cat and dog fetuses. They reported an exceedingly high V/0 ratio of 28 in human fetus at 112 days of gestation. The ratio f e l l to 5.0 at 190 days of intrauterine l i f e , and reached unity at birth.  If the intrinsic  oxytocic activity of arginine vasopressin (3-5%) is taken into consideration, the i n i t i a l l y recorded V/0 ratio of 28 not only indicates a great excess of vasopressor activity in the fetal neurohypophysis, but also carries the implication that the early neurohypophysis might contain arginine vasopressin alone - or even principles different from those found in the adult. The same authors, recorded a V/0 ratio of 4.7 in cat fetuses at 54 days of gestation, a few days prior to birth, and the relatively high value of 3.3 at the time of birth.  In addition to their studies on  fetuses, Dicker and Tyler estimated the neurohypophysial hormone content of various newborn animals and found that in contrast to the human newborn, guinea pigs at term exhibit a V/0 ratio as high as 10, but this f a l l s to 4.5 by six days post partum.  Similarly,  the neurohypophysial hormone content in 5 day old rats was s t i l l approximately 8 fold in favour of vasopressin. In the ten years following the original investigations of Dicker and Tyler, their work has come under criticism for the probable variability of human tissues obtained post-mortem, and for their use of acetone dried glands (e. g. Ginsburg, 1968). In 1959, Heller and Lederis found that the ratio of vasopressin to  13.  oxytocin was high in the neurohypophysis of newborn rats only i f the glands were dried in acetone, previous to extraction.  They  showed evidence that in the extracts of fresh, non-dried glands, the V/0 ratios were much lower than those obtained from acetone dried tissues, and often approximated to the normal adult value of one.  Since Heller and Lederis succeeded in recovering  oxytocin but not vasopressin from the acetone used in drying their tissues, they suggested that the discrepent ratios reported earlier might have been due to a differential solubility of fetal oxytocin in acetone, and the unusual fetal ratios could be no more than an artifact.  However, recently Vizsolyi and Perks  (1969) have carried out parallel studies of acetone dried and lyophilised posterior pituitaries of sheep fetuses at around midterm, and they found the V/0 ratios to be high, approximately 15 in the lyophilised tissue as well as in the acetone dried glands. Since the lyophilised glands had never been exposed to any type of organic solvent, i t was not possible to explain the remarkably high ratios obtained by a differential extraction of oxytocin, and their existence as a developmental phenomena can no longer be questioned. The few attempts to find a physiological role for the predominating antidiuretic activity of the fetal neurohypophysis were unsuccessful.  In some species, the immaturity of the kidneys  resulted in their complete inability to respond to acute stimuli. Heller (1949), working with newborn humans and rats, failed to e l i c i t either a diuretic response to water!oading, or an anti-  14.  diuretic response to dehydration.  In some of those species which  are more mature at birth, diuresis can be produced by waterloading, as has been shown in the newborn guinea pig by Dicker and Heller (1951), and in the 120-130 day old sheep fetus by Alexander and Nixon (1962).  However, the only response to vasopressin was  reported by Alexander and Nixon (1962), who were able to reverse the elevated glomerular filtration rate and urine flow of the diuresing sheep fetuses by the administration of the hormone.  In  specimens with normal glomerular filtration rate and urine flow, these authors also found that vasopressin was without effect.  In  common with other workers, Ames (1953) was unable to e l i c i t an antidiuretic response to exogenous vasopressin in infants less than three days old;  nevertheless, she was able to detect an elevated  antidiuretic activity in the urine of these infants i f water was withheld for a period of eight hours.  A concensus of these  findings suggests, that although an antidiuretic agent appears in the urine of dehydrated infants, and vasopressin in some species is effective in reversing the elevated glomerular f i l t r a t i o n rate and urine flow associated with diuresis, the immature kidney, in the absence of a diuresis, is unable to respond to vasopressin with either a drop in urine flow or with the production of hypertonic urine.  It is clear that the attempts to demonstrate a  physiological function of neurohypophysial have been few.  hormones in the fetus  There are even less studies of the nature of the  fetal principles.  Our only knowledge concerning their identity  15.  has come from chromatographic studies of Heller and Lederis (1959) in newborn rats.  Their results indicated that the active agents  of the immature animal were identical to those of the adult, and consequently i t has been assumed that the fetal peptides were also oxytocin and vasopressin.  However, in 1969 Vizsolyi and Perks  subjected neurohypophysial extracts of sheep fetuses at 0.60 of term (90 days) to paper chromatography, and found preliminary evidence for the presence of a third biologically active component.  By the pharmacological properties and chromatographic  behaviour of this agent, they suggested that i t was arginine vasotocin, the antidiuretic agent typical of sub-mammalian vertebrates. STATEMENT OF THE PROBLEM The pharmacological and biochemical  properties of the fetal  neurohypophysial principles are poorly investigated, and at the present time there is l i t t l e indication of their possible physiological role.  With the exception of two species, the human  and the sheep, the investigations of the biological activities were limited to only the last stages of gestation, or to the period immediately following birth.  As to the chemical nature  of the active agents, there is an almost complete lack of information.  The only indication as to the nature of the fetal  peptides comes from the recent paper-chromatographic studies in the sheep, which suggested that the fetal neurohypophysis contained a third active agent in addition to oxytocin and  16.  vasopressin;  this third peptide was tentatively identified as  arginine vasotocin, the antidiuretic hormone of sub-mammalian vertebrates. The principal aim of the studies to be presented here, was the purification and identification of the fetal neurohypophysial peptides.  It was clear that this knowledge was  essential before any reliable studies of embryonic functions could be carried out.' This phase of the investigation was important for a further reason;  the presence of the "primitive"  arginine vasotocin during fetal l i f e could make an important contribution to our understanding of the neurohypophysial peptides.  evolution of the  It also constituted an early warning  against the dogmatic allocation of particular peptides to particular species, since they might well vary at different stages of the l i f e cycle. In addition, there was a scarcity of information concerning all other aspects of fetal neurohypophysial physiology. Therefore, in the study presented here, the neurohypophysis of Callorhinus ursinus, the fur seal, was investigated at seven stages of gestation;  this covered a wide range of intrauterine  l i f e , from 0.19 to 0.93 of term.  The glands were assayed not  only for their vasopressor and rat uterus activities, but also for the presence of arginine vasotocin, as indicated by the frog bladder assay.  In addition, the histochemical appearance of the  hypothalamo-hypophysial  system was studied at five stages of  17.  gestation.  The histological studies, although not carried to  completion, outlined the morphology of the system in this new species, where i t had not been described previously.  They were  also useful in relating the results obtained in the seal with those of other species.  Finally, a few preliminary experiments  on the embryonic membranes were included, in an attempt to find the possible target organs of the fetal neurohypophysial peptides. In these ways, these studies were aimed to improve the overall picture of the function of the neurohypophysis during embryonic development. The species selected for these studies, Callorhinus ursinus, was.chosen for a number of different reasons.  Firstly,  the material was readily available (although i t was a rare opportunity).  Secondly, the fact that the seal is a marine  mammal made the adult alone of particular interest, since i t was possibly subject to water deprivation.  Thirdly, the  closely regulated reproductive pattern made i t possible to study fetal development in a non-domestic animal in its natural environment, without the need for extensive, controlled breeding and maintenance f a c i l i t i e s .  The closely regulated  reproductive  pattern also provided the opportunity to study neurohypophysial hormones during pregnancy, a field also in need of more extensive  research.  18.  GENERAL MATERIALS AND METHODS A. (1)  Materials Collection procedure: The neurohypophyses of Pacific fur seals, Callorhinus  ursinus, were collected as part of the Canadian Fur Seal Research Project, with the kind help and permission of Mr. Ian MacAskie. Specimens were collected in four consecutive breeding seasons, for three months each year, either from January to April, or from April to July.  The vessel remained at sea for the entire  duration of the hunt, and therefore i t was necessary to dissect and store the tissues on board. The animals were shot and the tissues were dissected in the following manner:  the superficial skin of the head was  removed with a hunting knife, and the cranium was severed with pliers, to expose the entire brain.  The frontal end of the brain  was lifted up, and the optic nerves, the infundibular stem and all other connections to the cranial floor were severed, in a rostral-caudal sequence.  The dorsal surface of the pituitary  was exposed within the cranial floor, and the entire pituitary was lifted.out from the sella turcica, and separated into posterior and anterior lobes. The posterior lobes together with a numbered label, were placed in small polyethylene vials (Beem capsules, Fisher.Scientific); they were dropped immediately into a liquid nitrogen refrigerator (-196°C), where they remained until received, s t i l l frozen, in the laboratory. Most fetal neurohypophyses, as well as a l l the adult glands, were collected  19.  in the manner described above, but in the younger specimens, which were particularly small (those collected from January to March), the posterior lobe was not separated from the anterior lobe, and the whole pituitary was placed into the refrigerator. (2)  Lyophilisation and storage: At the end of the three month collection period, the  frozen tissues were taken to the laboratory. The polyethylene capsules were removed from the liquid nitrogen, and the pituitaries of the younger fetuses were separated into the posterior and anterior lobes.  The covers of the capsules were  perforated, and the capsules were placed in a round-bottomed flask, and cooled on a dry ice-methanol mixture.  The flask  containing the capsules was attached to a high vacuum pump (Welch Scientific Co.) and evacuated for 24 hrs at a pressure of approximately 0.01 mm Hg.  At the end of the freeze-drying period  the vials were removed, and stored in vacuo over phosphorus pentoxide at 4°C, until they were extracted for investigation. (3) Extraction: The lyophilised glands were homogenized by means of a Thomas tissue extractor into 0.25% acetic acid, at a concentration of 3-5 mg dry tissue/ml.  Extraction was carried out according to  the description of the British Pharmacopea (1963).  The extractor  tube was lightly stoppered with cotton wool and placed on a boiling waterbath for three minutes. tube was cooled in cold water.  At the end of this time, the  The cold suspension was filtered  20.  through a Whatmann #1 f i l t e r paper, and the f i l t r a t e was stored at 4°C until assayed.  In the case of two extracts, those from small  fetuses collected in January and February, the filtration step was replaced by centrifugation of the boiled homogenates at 1470 g (Clinical Centrifuge International Equipment Co., Mass.); this helped to conserve the limited quantity of material. B.  The Estimation of Biological Activities For the estimation of the biological activities of the crude  and purified extracts three bio-assays were used: the isolated rat uterus assay, the rat vasopressor assay, and the frog-bladder assay.  In addition to these, rat antidiuretic assay, rat milk-  ejection assay and isolated rat uterus assay with Mg  ++  ions were  included for the pharmacological identification of purified peptides. All biological assays were carried out by the four-point statistical method of Holton (1948).  Injections were given in  groups of four, two doses of the unknown (one high and one low) matched for response with two parallel doses of standard, and given in a random sequence. six groups.  A complete assay contained four to  Occasionally, when eluates from  chromatographic  columns were assayed, the method was simplified so that one dose of the unknown was bracketed between two doses of the standard for an approximate estimation of potency. (1)  Isolated rat uterus assay: The estimation of potency on the isolated rat uterus was  carried out following the experimental design of Holton (1948), as  21.  modified by Munsick (I960).  Virgin albino rats of the Wistar  strain were selected in full oestrous or late pro-oestrous, by microscopic examination of their vaginal smear. The rats were killed by concussion and bled through an incision in the neck. The abdomen was opened by a midventral and two transverse incisions, and the uterus was exposed.  Both uterine horns were  dissected, starting from the ovaries; they were freed from fat and connective tissue, and the vaginal end was transected. The uterus was transferred into a Petri dish containing van DykeHastings' solution (see below), and divided into its two constituent horns by cutting longitudinally through the vagina. The vaginal end of one horn was tied with silk thread to a glass hook, which also acted as an air supply. thread was sewn through the ovary.  A longer piece of  The glass hook with the  attached uterus was then immersed in a 5 ml muscle bath containing van Dyke-Hastings' solution, and connected to a supply of 5% C02~95% O2.  The free end of the.thread sewn to the ovary was  secured to a carefully balanced writing lever, which in turn rested against the smoked drum of a Palmer Kymograph. At five minutes intervals, doses of standard were pipetted into the bath, alternated with doses of unknown. The standard was 1/1,000 or 1/2,000 dilution of Syntocinon (10 I.U./ml, Sandoz). After the response had reached i t s maximum, the van Dyke-Hastings' solution in the muscle bath was exchanged for fresh solution from the reservoir, by means of a rubber bulb manual pump.  22.  Two types of van Dyke-Hastings' solution were used to support the uterus; they differed from one another only in the presence or absence of Mg a.  ++  ions (Munsick, 1960):  van Dyke-Hastings' solution without Mg  ++  The van Dyke-Hastings' solution was prepared fresh daily, by mixing two stock solutions, according to the following formula: Stock solution A NaCl  132.400 gm  NaHC0  51.100 gm  3  KC1  9.080 g  Phenol red, Na salt  0.054 g Combined, and made up to two liters in dist H 0. o  Stock solution B 4  22.700 g  up to 1 l i t e r in dist. H 0 2  5.520 g up to 1 l i t e r in dist. H 0 2  The Na HP0^ solution was titrated against the NaH P0 2  2  to give a final stock solution of pH 7.4. The final mixture for the assay was prepared by the addition of:  23.  200 ml of stock solution A 20 ml of stock solution B 1 ml of IM CaCl solution 2  1 g of glucose and made up to two liters with dist. water, b.  van Dyke-Hastings' solution with Mg  ++  The preparation of this solution is identical to the preceding one, with the exception that in addition to the other salts 1 ml of IM MgCl was also added. 2  (2)  Rat vasopressor assay: The vasopressor activity of neurohypophysial extracts was  estimated according to the method of Landgrebe, Macaulay and Waring (1946), as modified by Dekanski  (1952).  Wistar rats of approximately 300 g body weight were anaesthetised by subcutaneous injection of urethane (175 mg/100 g wt, Matheson, Coleman and Bell Co.).  At the same time, 0.5 mg/100 g phenoxy-  benzamine hydrochloride, an adrenergic blocking agent (Dibenzylene, Smith, Kline and French Co.), was injected to lower the blood pressure and to block pressor responses to hypertensive agents, such as adrenalin, nor-adrenalin, and to histamine and nicotine (Dekanski, 1951 , 1952). The rats were ready for surgery 30 to 45 minutes after the injection of the anaesthetic. The skin on the throat was cut, the right jugular vein was located and freed from connective tissue. Two ligatures were placed around the vein, and the one distal to the heart was tied.  The vein was punctured with a 22 gauge needle,  24.  the needle was withdrawn, and a cannula (PE 10 Intermedic Clay Adams Co.) was inserted through the puncture.  The cannula  was secured in place, by tying both ligatures. The carotid artery was cannulated in the same manner described for the jugular vein, with the exception that in place of the PE 10 tubing, PE 50 tubing was used. To prevent the formation.of blood clots during the assay, 0.5 mg/100 g body weight of heparin (155 USP U/mg,  Sigma Co.),  dissolved in 0.9% NaCl solution, was injected intravenously. Injections of the standard and the unknown solutions were given through the jugular cannula and washed in with 0.2 ml of 0.9% NaCl.. The blood pressure was recorded from the carotid cannula, by means of a Statham 23.AA transducer, coupled to a Beckman dynograph recorder.  Hoi ton's four-point experimental  design was used for the estimation of the potencies.  The  standard used was 1/400 dilution of Pitressin (Parke Davies Co.). (3)  Frog water balance assay: Posterior pituitary extracts were assayed for their  effect on the water-transport across the isolated bladders of the bullfrog, Rana catesbiana, following the method of Sawyer (1960).  Large bullfrogs were lightly anaesthetised with ether,  and were decapitated. The abdominal wall was opened to expose the bladder, and the bladder was carefully dissected free of the body.  The dissected bladder was divided into i t s two  constituent lobes, and each half bladder was secured onto the  25.  flared end of an open glass tube.  The tube with the attached  bladder was f i l l e d with 5 ml of distilled water, and suspended in a 50 ml saline bath, which contained 25 ml of frog Ringer's solution (see below).  The outer container was supplied with  compressed air by means of a glass tube, sealed into the bottom of the bath. Every 15 minutes the bladders were removed from the Ringer's bath, their outer surface was dried by brief draining onto f i l t e r paper and rapidly.weighed.  They were considered  ready for the assay when the bladder showed a slow steady loss of weight, not exceeding 1.5 mg/min. Doses of standards and unknown solutions were given into the outer bath.  The rate of water-loss from the bladder was  measured by weighing the glass tube with the attached bladders every 15 minutes throughout the assay.  The weight loss during the  period between 15 and 45 minutes following the addition of the standard, or of the pituitary extract, was considered the response.  After 45 minutes the Ringer's solution in the outer  bath was exchanged with fresh solution and the bladder was allowed 45 minutes to recover, in which time the rate of water loss returned to the baseline level.  The potency was estimated  by either the 4 point method of Holton (1948), or when the material was not sufficient for a full assay, the potency was calculated using a log dose response curve.  The assay was  26.  carried out against small volumes of undiluted Syntocinin (Sandoz, 10 I.U.) . The Ringer's solution used in the assay was prepared from two stock solutions as follows:Stock solution A NaCl  94.0 g ) )  KC1  3.7 g ) up to 1 l i t e r in distilled water ) 5.3 g )  CaCl .2H 0 2  2  Stock solution B NaH P0 .2H 0 2  4  2  NaHC0  3  glucose phenol red  2.0 g ) 40.0 g ) ) up to 1 l i t e r in distilled water 4.0 g ) ) 12.0 mg)  50 ml of stock solution A and 50 ml of stock solution B were mixed, and made up to 1 l i t e r in distilled H 0. 2  (4)  Rat antidiuretic assay: The antidiuretic potency of neurohypophysial peptides was  estimated in ethanol anaesthetised rats, by the method of Sawyer (1961). Male albino rats of approximately 250 g were loaded to 5% of their body weight with 12% ethanol by means of a stomach tube (#8 Fr Urethral Catheter, Clay Adams). The animals were ready for surgery after 25 to 40 minutes.  To ensure complete insensitivity  27.  to surgical procedures, Xylocaine Hydrochloride (2% solution, without epinephrine, Astra Pharmaceuticals) was applied locally to the area of incision. The jugular vein was cannulated as described for the vasopressor assay.  A PE 240 Intermedic cannula was inserted into  the trachea to ensure free breathing. The bladder was cannulated in the following manner:  a h inch mid-ventral incision was made  through the skin and abdominal muscle, approximately h inch anterior to the penis. The bladder was eased through the incision and a small hole was cut in a relatively avascular area. The flared end of a PE 200 Intermedic cannula was inserted through the cut, and tied into place with a ligature close to the urethral exit.  The penis was tied to prevent any loss of urine  during the assay.  The abdominal incision was sutured.  The prepared rat was hydrated up to 8% of its pre-surgery weight with a solution of 1.5% ethanol in 0.05% NaCl, by way of a stomach tube.  The same solution was used throughout the assay  to replace lost urine. The assay was started when the urine flow reached a constant high level of approximately 1 ml/min. The 4-point method of assay was used.  Injections of standard and unknown  were given through the jugular cannula and washed with 0.2 ml of 0.9% NaCl.  The response lasted 15 to 30 minutes, and the next  injection was given when urine flow returned to the basal level.  28.  Responses to the injections were recorded on a smoked Kymograph drum and were also monitored electronically in order to estimate the size of the response;  the method was the following:  The cannulated bladder.was connected to a drop counter (C. F. Palmer).  This drop counter incorporated a motorized writing lever,  which rested against a.smoked Kymograph drum (C. F. Palmer). The writing lever was activated and began to rise slowly when a drop of  urine bridged the two electrodes of the drop counter. As  soon as the drop f e l l from the electrodes, the writing lever f e l l rapidly back to the baseline.  In this way the sweep of the lever  was directly proportional to the length of time that the urine droplet spanned the electrodes;  therefore, a decrease in urine  flow registered on the smoked drum as a rise in the strokes recorded.  Although this method of monitoring produced a permanent  visual record of an antidiuresis, calculation of the magnitude of the antidiuretic response was cumbersome, since i t involved the measurement of the height of each stroke of the writing lever during the response.  Therefore, for ease of evaluation of the  antidiuretic responses, the drop counter was also connected to an electric timer (Precision Scientific Co.), and to an impulse counter (modified from a Beckman fraction collector control unit). The impulse counter and timer were activated in the same way as the writing lever, when a drop of urine rested between the electrodes of the drop counter. the electrodes the timer stopped.  When the drop f e l l from between Therefore, the timer recorded  29.  the accumulative time during which successive drops of urine were in contact with the electrodes of the drop counter, and the impulse counter recorded the number of drops.  The anti-  diuretic response was calculated by a formula: Response = total time of contact x number of drops during antidiuresis / total time x number of drops at resting flow.  The assay was carried  out against Pitressin (Parke Davies Co., 20 I.U.) in a 1:40,000 dilution. (5)  Rat milk ejection assay: The method used was that described by Bisset et a l ,  (1967).  Lactating female rats (250-350 g) of the Wistar strain  were used 8 to 16 days after parturition.  They were separated  from their young during the night preceding the experiment. The animal was anaesthetised with sodium pentabarbitol (Nembutal, Abbott Chemical Co.), by intraperitoneal injection at 4.5 mg/100 g body weight.  The jugular vein was cannulated as  described for the vasopressor assay.  To monitor the milk ejection  responses, one of the teats from the most distal pair on the abdomen (lower and upper inguinal) was used.  The teat was  anaesthetised locally with Xylocaine, pulled out with forceps, and the tip was excised. The primary duct was cannulated with PE 60 tubing, and the tubing was tied into place by a silk ligature; the teat was gently pulled away from the body in order to maintain the natural position of the duct.  A Statham P 23 BB transducer  was connected to the teat, and the whole system was f i l l e d with a  30.  3.8%  sodium citrate solution to.prevent coagulation of the milk.  The assay was carried out following the 4-point design of Hoi ton (1948).  At 5 minute intervals injections of standard or unknown  were given through the jugular vein, washed in with 0.2 ml of 0.9% NaCl, and the responses were recorded as a rise in pressure within the mammary gland.  The assay was carried out against  Syntocinon in 1/2,000 dilution. (6)  Standards: Rat uterus, milk-ejection and frog-bladder assays were  carried out against synthetic.oxytocin (Syntocinon, 10 I.U./ml, Sandoz).  The standard used for vasopressor and antidiuretic  assays was Pitressin (Parke Davies, 20 I.U./ml);  this  constituted a mixture of arginine and lysine vasopressins. For direct pharmacological comparisons of neurohypophysial peptides, two preparations were used;  synthetic arginine  vasopressin (Sandoz, 5 I.U./ml), and arginine vasotocin (0.1 mg/ ml, courtesy of Dr. Berde of Sandoz Pharmaceuticals). C. The calculation of potencies The biological activities were calculated as described by Hoi ton (1948), according to the following formuli:  31.  Potency (mU/ml)  =  ^ ^ volume of D  where R = antilog of M C = the high dose of standard M is calculated by the formula: ( A + D ) - ( B + C ) M =  (log c - log b) (C + D ) - ( A  +  B)  the symbols are: A = the sum of responses to the low dose of standard B = the sum of responses to the low dose of unknown C = the sum of responses to the high dose of standard D = the sum of responses to the high dose of unknown c = high dose of standard b = low dose of standard The fiducal limits for the assays were calculated at the 95% level of confidence, as described by Holton (1948).  This  was usually carried out by means of an IBM 1130 computer.  I am  indebted to Mr. S. Borden, Department of Zoology, for setting up the programme. When ratios between two bioassays were calculated  ( a/b ),  the fiducal limits of the ratios were calculated at the 95% level from the following formula:  32.  Confidence limits = + (t) \ 2 where Sm, a  2 and Sm, b  —  2  2 + Sm,  a  D  Sm  v  are the standard errors of M, from the  calculations of potencies for a and b. Student's t (t) was calculated for F degrees of freedom, taking into account the degrees of freedom associated with both assays, using the following formula: < a S m  p  =  +  S m  b  >  2  (Sm^ / R ) + (Smj; / R ) a  fa  where the additional symbols of R and R^ are the degrees of fl  freedom associated with a and b, from the estimation of potencies. D.  Purification methods 1. Paper chromatography: The paper chromatography techniques used were described  by Heller and Pickering (1961) and by Perks (1966). Butanol, acetic acid and water in a 4:1:5 ratio were shaken in 250 ml separatory funnel for 15 minutes immediately after mixing, and every 15 minutes thereafter for a period of 1% hours. layers.  The mixture then was allowed to separate into two The lower aqueous layer was played into four Petri  dishes in the bottom of a 12" x 12" x 24" glass chromatography tank.  The l i d of the tank was generously greased with "Lubraseal"  to make i t vaporproof, and the system was equilibrated at 20°C for one hour.  33.  Standard and unknown solutions were spotted onto 24 x 54 cm sheets of Whatman 3MM chromatography paper, by the use of 1 ml tuberculin syringes.  Cold air from a portable hairdryer  was allowed to pass over the origin, to speed the drying of the spots. When spotting was completed, the sheet was placed into the chromatography tank where i t was suspended from the solvent trough.  The closed system was equilibrated for a further hour.  At the end of this time, the upper phase of the solvent mixture was run into the trough, through a small opening on the l i d of the tank.  The chromatogram was developed for a period of 12  hours at room temperature (20-22°C).  At the completion of the  run the chromatogram was removed, the solvent front was marked, and the remaining solvent was evaporated by a cool flow of a i r . Longitudinal strips of the dry chromatograms were cut beneath the origins; an extra 0.5 cm was allowed on each side of the origin to include any lateral diffusion. The strips were divided into ten equal parts between the origin and the solvent front, so that each resulting square corresponded to 0.1 R  f  units.  The squares were numbered, folded and placed  individually in to 5 ml beakers, which contained 1 ml of 0.25% acetic acid.  The eluate was recovered from the paper by  strongly compressing each square in a 5 ml polythene syringe. Each beaker was sealed with Parafilm (American Can Co.), and stored at 4°C until assayed.  34.  2.  Gel f i l t r a t i o n : Crude extracts of neurohypophyses were passed through  Sephadex G-l5 dextran columns (Pharmacia, Uppsala).  For the  experimental procedures, the manufacturers' instructions were followed. (a)  Preparation of the gel Three different lot numbers.(2014, 435 and 8507) of gel,  with 40-120u particle size, water regain value of 1.5+0.2 g/g, and bed volume of 3 ml/g were used. large volume of 0.2M magnetic stirrer.  The dry gel was added to a  acetic acid, and mixed for one hour on a  The suspension.then was allowed to settle for  30 minutes, the acetic acid was decanted and discarded along with the fine, non-settling particles. times. 0.2M  This process was repeated four  After the fourth washing, the gel was taken up in enough  acetic acid to form a thick slurry.  (b)  Preparation of the column A glass chromatography column (Kontes Chromaflex, 200 cm  x 2.5 cm) was fixed vertically into a specially built stand, and f i l l e d with 0.2M acetic acid.  A plexiglass wheel was fixed  around the column near its base, a few inches above the outlet; this wheel was connected by a rubber belt to a Dayton (0.1 HP) gearmotor.  The motor was used only during the building process to  rotate the column slowly;  this way i t was possible to settle the  gel with a completely horizontal surface. A half inch layer of glass wool, followed by a layer of small beads, were placed at the  35.  bottom of the column before the pouring of the gel was started. Two alternate methods were used for packing the gel into the column, and this i n i t i a l step proved to be of critical in i t s later performance.  importance  In one.case a 250 ml funnel, f i l l e d  with 0.2M acetic acid was attached to the top of the column, by means of a tight fitting  rubber stopper.  The Sephadex G-15  slurry was added to the funnel and stirred continuously with an electric mixer, as i t slowly settled out into the column.  As the  suspension thinned in the funnel, additional gel slurry was added at approximately half hour intervals.  This procedure for packing  the column was deliberately slow, and took a total of 32 to 38 hours.  At the completion of the packing, the gel surface was  covered with a sintered glass f i l t e r , on top of which f i l t e r paper was placed (Whatmann #1). gel  The f i l t e r s on the top of the  bed served as protection against.disturbing the surface during  manual sample applications, (c)  Purification procedure Before using the column, 150 ml of 0.2M acetic acid was  allowed.to run through i t for a period of 6 hours.  Following the  washing, the liquid above the gel surface was allowed to run down until the meniscus reached the gel.surface. Posterior lobe extracts were carefully applied to the top of the column and permitted to run slowly into the gel surface. The column was then f i l l e d with acetic acid (0.2M) and connected to a reservoir of the same solution.  The unknown was eluted with the 0.2M acetic acid  at a rate of 10-14 ml/hr, and the eluate was collected in 2.8 ml  36.  fractions (60 drops) by means of a fraction collector (LKB Ultrarac, Type 7000). 3.  Ion exchange chromatography: In the course of purification of the neurohypophysial  extracts, the eluates from the gel filtration column were applied to one of the following cation exchange resins:  CM Sephadex  (Pharmacia Uppsala) Phosphocellulose (Selectacel, Brown Co.), Amberlite IRC-50 (Rohm & Haas Co.), or Dowex 50X2 (Bauer Chemicals). a.  Precycling of the resins All the resins, with the exception of phosphocellulose,  which did not require pre-treatment (Wilson, 1968), were precycled in essentially the same manner.  They were subjected to  consecutive acid and alkali washings, to free the resins from heavy metal ions and from incompletely polymerised material (Morris and Morris, 1964).  For the acid alkali cycle, HCI and  NaOH were used in concentrations ranging from 0.5 to 2.0M, depending on the gel. After this treatment, the resin was converted to i t s ammonium salt by washing with ammonium hydroxide. This last step was necessary since a l l the experiments were carried out in ammonium acetate buffers. At the completion of the pre-cycling procedure, the resins were equilibrated in the appropriate starting buffers, and were ready for the building of the columns.  Details of the pre-cycling procedures for each  resin are given in the Appendix.  37.  b.  Buffer solutions In a l l experiments, ammonium acetate buffers were used.  Although sodium buffers promote a better resolution of the peptides (Moore and Stein, 1954), ammonium salts sublimate under vacuum, and therefore they are easy to remove from the peptide preparations.  The buffers were prepared in the following manner.  A stock solution 0.2M acetic acid was titrated to pH 5.0 or pH 7.0 (Radiometer, Type pH 22 Ph meter), and diluted with d i s t i l l e d water to the required molarity, using the measurements of electrical conductivity as a rapid and convenient method to determine concentrations. c.  The preparation of ion exchange columns 25 ml and 10 ml burettes (1.0 and 0.5 cm diameters,  respectively), were cut to be no longer than 15 cm, and the cut end was fire-polished. Next, the burette was f i l l e d with the starting buffer, the bottom one half cm was packed with glass wool, and a few glass beads were layered on top of the glass wool. Then the CM Sephadex gel was stirred into the column from a funnel with the aid of an electric mixer.  All other resins were  poured into the burettes directly from a beaker.  The packed  column was washed with 15-20 times i t s void volume of the starting buffer, until the conductivity of the eluate was identical to that of the starting buffer.  The conductivity of the peptide  solution was adjusted to be identical with the conductivity of the starting buffer;  i t was then loaded onto the column at a flow  rate not exceeding 20 ml/hour.  For the loading and eluting an  38.  LKB peristal tic pump was used. d..  The preparation of the sample The buffer concentrations which ensured strong binding of  the peptides to the resin ranged from 0.002 to 0.2M ammonium acetate, depending on the gel used. G-15 columns contained 0.2M  Since the eluates from the  acetic acid, the ionic concentrations  had to be lowered by as much as 100-fold.  This could be  achieved by dilution of the sample, which produces very large volumes and makes further work cumbersome and often inefficient. However, since acetic acid is equally soluble in ether and in water, and the neurohypophysial  peptides are insoluble in the  former, i t was possible to remove up to 95% of the acid from the samples by the use of this solvent; there was no appreciable loss of biological activities.  The ether extraction was carried  out in the following manner. The sample was washed three times in a separatory funnel with ether (Fisher Scientific Co.), each time using three times the volume of the sample.  The aqueous  layer was recovered after each washing and the ether discarded. After the third washing, the ether remaining in the aqueous phase was removed by blowing nitrogen through the solution.  After the  ether washing, the pH of the sample was adjusted by titration with concentrated (28% for ammonia) ammonium hydroxide, and the concentration was adjusted with distilled water, so that both values corresponded to that of the starting buffer (the concentration was assayed by conductivity measurements).  39.  e. . The elution of the biological activities At the completion of the loading procedure, the column was washed with a volume of starting buffer which corresponded 3 to 4 times its void volume.  Following the washing, a gradient  machine was attached to the column.  This gradient machine  consisted of a 60 ml mixing chamber, stoppered with a rubber stopper, into which two glass tubes were extended.  The chamber  was f i l l e d with the starting buffer and placed onto a magnetic stirrer.  One of the outlets was connected to a buffer reservoir  containing the eluting buffer and the other to the top of the column.  In most instances a peristaltic pump was also connected  between the column and the chamber for regulating the flow through the column.  The outlet of the column was opened, and the  peristaltic pump was started at a rate of 10 ml/hr.  Usually the  eluate was collected by the fraction collector in 2.8 ml fractions (60 drops/tube) but occasionally the volume of the fractions was varied to 2 ml or 4 ml.  The tubes containing the  eluates were covered with Parafilm, and stored at 4°C until they were assayed. E.  Analytical methods 1) pH and conductivity measurements: The pH and conductivity was measured in eluates of ion  exchange chromatography of the trial runs only.  When samples for  amino acid analysis were chromatographed, these measurements were made only in tubes previous to the expected appearance of the peptides, to prevent contamination of the samples.  The pH was  40.  checked with a Radiometer type PHM 22R pH meter, and conductivity was measured with a Radiometer type CDM 2d conductivity meter. 2)  Preparation of samples for hydrolysis: Fractions with identical biological activities were  pooled in acid washed, round-bottomed flasks, and frozen on a methanol-dry ice mixture and dried on a high vacuum pump (Welsh Scientific) at 0.005 mmHg pressure.  The dried residue was  taken up in d i s t i l l e d water and dried again.  This procedure was  repeated three times to free the dry residue from a l l ammonium acetate.  The dry powder was stored in a deep freeze (-18°C)  until hydrolysis. 3)  Hydrolysis: The dry sample was taken up in distilled water, and  transferred into a 10 ml hydrolysis bulb (Kjeldhal flask, 10 ml. Canlab.).  It was dried once more, 2 or 3 ml of 6 N HCI was  added to the dry powder, and N gas was bubbled through the 2  liquid for 30 minutes to free i t from'trapped oxygen.  Following  the N treatment, the sample was frozen in liquid nitrogen and 2  evacuated for 30 minutes on.the vacuum pump at 0.005 mmHg pressure.  The hydrolysis bulbs were sealed while s t i l l under  vacuum, with an oxygen torch.  They were immersed in a constant  temperature o i l bath (Blue M. Electric Co.), at 108°C, for 18 to 24 hrs. At the end of the allotted time, the bulbs were removed from the o i l bath cooled in a deep freeze (-18°C), and opened.  41.  The hydrolysate was transferred to a small pear-shaped flask (25 ml), and the HCI was driven off on a flash-evaporator (Calab, Richmond, Calif.).  The sample was taken up in iced  distilled water, and evaporated at 37 C until dry. This process was repeated three times, or until no further smell of HCI could be detected.  The dry powder was stored in a deep freeze until  amino acid analysis could be undertaken. 4) Amino acid analysis: The dry powder, containing the hydrolysed peptide was taken up in a sodium citrate/HCl buffer at pH 2.2, so that the concentration of each amino acid approximated 5-25 nM/ml. One ml of the sample was applied to a Biocal 200 amino acid analyser (Biocal, Richmond, Calif.).  The amino acid content  was calculated from the areas of the recorded peaks, according to the manufacturers' directions (see Biocal 200 service manual).  42.  SECTION I. THE NEUROHYPOPHYSIAL HORMONES OF THE ADULT AND THE PREGNANT SEAL Introduction: I n i t i a l l y , the investigation of the neurohypophysial hormones in the adult seal was undertaken to provide the necessary background information for later work on fetuses. However, i t was extended into a much more comprehensive study, owing to the fact that previous research concerning the neurohypophysis of Callorhinus ursinus was limited to only one description of i t s gross anatomy (Fuse, 1939).  This lack of  previous work made physiological studies of the mature seal of interest in themselves.  In particular, the marine environment  of the seal - with i t s special constraints on salt and water balance - might have been reflected in the hormone content of the neurohypophysis, since this gland is intimately involved in Water regulation. Therefore, the biological activities present in the neurohypophysis of both male and female seals were estimated, and the active agents themselves were identified.  In  addition, a study of the biological activities at five stages of pregnancy were included. Previous evidence in the sheep had shown that neurohypophysial hormone levels fluctuated with different stages of pregnancy;  therefore i t seemed important to  extend investigations of the neurohypophysis to an additional species, and to a broader span through pregnancy.  43.  The specimens used in this study were collected pelagically by the courtesy of the Canadian Fur Seal Research Project, and particularly by the kindness of Mr. Ian MacAskie. The reproductive pattern of the fur seal is closely regulated, and this made possible the collection of material at known stages of gestation.  The reproductive activity of this species  has been described by Bartholomew and Hoel (1953) in the following manner. Mating takes place in the Pribi1 of Islands between mid and late July, immediately after parturition.  The  fertilised ovum does not implant until mid-November, and the actual gestation period from implantation to parturition is given as 240 days.  Pregnant females, together with non-  breeding specimens of both sexes leave the Islands by November, and remain at sea until the following July.  Virtually a l l  females of reproductive maturity have bred, and therefore the non-gravid specimens among them have suffered either resorbed or aborted pregnancies, or have not reached reproductive maturity.  The parallel timing of the reproductive pattern of  the seal colony ensured that any specimens collected at a particular time of the year were at a closely similar stage of gestation.  In the following results the average time of  implantation, mid-November (Crajg, 1964), is taken as 0.00 of term and parturition is expressed as 1.0 of term;  intermediate  stages are calculated as a proportion of these limits.  ^ y  44. TABLE I Biological activities of the adult and pregnant seals  Tissue  Adult Male  Adult Female  Pregnant Female  Number of glands  Time of col lection  Pr^op.  of  term  Vasopressor activity mU/mg* (V)  Rat uterus activity mU/mg* (0)  V/0 ratio  3  February  5,299.+ 793  3,990 + 291  1.32 + 0.27  3.  Apri 1  5,260 + 536  4,085 + 543  1.30 + 0.15  3  June  5,936 + 545  3,383 + 288  1.75 + 0.20  4  February  4,765 + 765  2,914 + ,239  1.63 + 0.37  5  April  5,373 + 495  3,224 + 230  1.65 + 0.25  3  June  5,860 + 649  3,145 + 328  1.72 + 0.30  8  January  0.19  5,373 + 621  3,929 + 182  1.37 + 0.16  February  0.31  7,592 + 175  5,526 + 563  1.37 + 0.33  5  Apri 1  0.56  3,081 + 435  3,090 + 221  0.99 + 0.15  5  May  0.68  3,755 + 614  3,584 + 606  1.05 + 0.27  3  June  0.82  4,991 + 534  3,810 + 416  1.28 + 0.19  .6  * Biological activities expressed as mU/mg lyophilised powder. at P = 0.05.  Fiducal limits  45.  A.  Biological activities 1.  Control, non-breeding animals: The lyophilised posterior pituitaries of 3 to 5 males  and of 3 to 5 non-gravid females (the numbers depending on availability), were collected at three different times of the year.  They were pooled according to sex and the time of  collection, and each group of glands was extracted in 0.25% acetic acid (see Methods, p. 19 ). The crude extracts were assayed for their vasopressor and oxytocic activities, and the results are shown in Table I and Fig. 1. The vasopressor activity was found to be between 5,000 and 6,000 mU/mg dry weight in both the males and the females, although the pituitaries of the males were somewhat more potent. These values are much higher than those recorded for other mammalian species;  they are approximately three times the amount found  in the standard laboratory rat (Heller and Lederis, 1959), or in the posterior lobe of cattle (the species from which the International Standard Powder i s derived). The oxytocic activity was found to be an average of 3,094 mU/mg in the posterior pituitaries of the females, and 3,818 mU/mg in the males.  In a l l cases the glands were found  to be less potent in oxytocic than in vasopressor activity. In the non-breeding animals, the apparent minor fluctuation in the potencies of both biological activities, in both sexes, were found to be within the range of the confidence limits (see Fig. 1).  This suggests that in the non-breeding animals, neither  Figure 1. The biological activities of the adult seal neurohypophy Values are expressed as mU/mg lyophilised  tissue.  Vertical bars represent fiducal limits at P=0.05. A.  The biological activities in the neurohypophyses of adult males.  B.  The biological activities in the neurohypophyses of adult, non-gravid females.  Open squares and broken lines =  rat uterus activity  Full circles and solid lines  vasopressor activity  =  o  6000-  o  •——•  5000-  r  4000-  3000-  2000-  o> i  ,  1  1  1  1—  i  E  6000-  5000-  4000-  3000  2000-  —!  Jan  ,  Feb  1  " Mor  1  .Apr  1  May  —i  June  Vasopressor Rat uterus  47.  the  time of the year, nor the different environmental conditions  encountered during migration, influenced the neurohypophysial hormone levels of the fur seal in any significant way. The ratios of vasopressor to oxytocic activities (V/0 ratios) were calculated, and the results are shown in Table I and Fig. 2.  In all cases except one the glands were found to  be richer in the vasopressor than in the oxytocic activities, so that the V/0 ratio was greater than 1.0. However, the values for the V/0 ratio never exceeded two, and therefore the predominance of vasopressor activity was not great enough to make this species significantly different from other land mammals (see Table IV). 2. Biological activities during pregnancy: The posterior pituitaries of gravid C_. ursinus were collected over a period of five months, between January and June of 1968.  If mid-November - the average time of  implantation - is taken as 0.00 of term, these tissues represented a range from 0.19 to 0.82 of term, the widest range through pregnancy yet investigated.  Five to ten lyophilised  glands, collected only within the f i r s t five days of each month, were pooled and extracted at each stage. The crude extracts were assayed for their vasopressor and oxytocic activities, and the  results are shown in Table I and Fig. 3A. An i n i t i a l rise  from the non-pregnant control level was detected in the vasopressor activity at 0.31 of term. the  During the next two months,  neurohypophysis appeared to be greatly depleted of its  48.  Figure 2. The ratios of vasopressor to oxytocic activities (V/0) in the neurohypophyses of the adult seal.  The values  were calculated as: vasopressor activity, mU/mg dry tissue  /  rat uterus oxytocic activity, mU/mg dry tissue. Vertical bars represent confidence limits at P 0.05. =  A.  V/0 ratio in the glands of adult males.  B.  V/0 ratios in the adult, non-gravid females.  —i  — I —  Jan  Feb  Mar  Apr  May  J  une  49.  vasopressor activity, and at 0.56 of term the potency was only 3.081 mU/mg dry tissue, which was 40%, or less than one-half of the value recorded at 0.31 of term.  In the next two stages  of pregnancy, the glands became gradually.richer in vasopressor activity, and at 0.82 of term, the hormonal activity once again reached the control level. The rat uterus activity, in contrast to the vasopressor activity, was found to be elevated from the control level in a l l stages of pregnancy except for one (0.56 of term), when the oxytocic activity equalled the value recorded in the controls. However, a fluctuation in the neurohypophysial content of this agent was detected with the different stages of gestation, and the changes parallel those found for the vasopressor activity. The V/0 ratios, shown in Fig. 4, are lower than those of the controls, and this would appear to reflect mainly the elevated rat uterus activities during pregnancy.  At approximately mid-  term the ratios seem to be depressed; however, the confidence limits overlap, and therefore the change cannot be regarded as statistically significant. When the results obtained during pregnancy are compared to those of the controls (Fig. 3A & B), i t would appear that pregnancy e l i c i t s several responses on the neurohypophysial hormone level.. Firstly, a fluctuation which involved as much as 60% of the stored hormones was detected in both active agents at different stages of gestation. Secondly, the oxytocic activity  50.  Figure 3. A.  The biological  activities of the seal neurohypophysis  during pregnancy. B.  The biological  activities of the adult seal neurohypophysis,  Activities are expressed in mU/mg lyophilised Open circles, broken line  tissue.  = rat uterus activity  Full circles and solid line = vasopressor activity Vertical  bars denote fiducal  limits at P=0.05  °  8000H  •  D  Rat .uterus  • Vasopressor  7000 6000" 5000400030002000^ ~oT E  ^  °^  ^ °% OT proportion of t e r m  o'-8-  X E 5000H  4000H  3000H  2 0001  — I —  Jan  Feb  Mar  Apr  May  June  0-9  /To '/ Control 7  gure 4. Ratios of vasopressor to oxytocic activities in the neurohypophyses of pregnant seal.  Ratios were calculated  as mU/mg vasopressor activity over mll/mg oxytocic activity. Vertical bars represent fiducal limits at P=0.05  o  or  o >  i-o  1 0-|  0-2  0 3  0-4  0 5 0 6 0~7 p r o p o r t i o n of t e r m  0^8  0^9  7^0~  52.  was found to be elevated from control levels all through pregnancy, an observation which appeared to be reflected in the somewhat depressed V/0 ratios. B.  Purification of the active agents: The neurohypophyses of the adult seals proved to be  exceptionally rich in hormonal activity.  Since the nature of  the neurohypophysial principles had never been established in this species, i t seemed important to determine, whether these unusually potent glands might contain principles in addition to, or other than the two standard neurohypophysial peptides, oxytocin and arginine vasopressin. For the purification of the active agents a one step purification method was employed;  this  method had been found to be successful in previous investigations of the sheep neurohypophysis (Vizsolyi, 1968).  However, since  the available sheep material had been insufficient for amino acid analysis, the experiments below were also designed to establish the value of this method for obtaining analytically pure peptides. Results: (a) The purification of the vasopressor fraction: 10 ml of 3 mg/ml crude neurohypophysial extract of adult female seals were applied to a specially built 200 cm Sephadex G-l5 column (see Methods).  The total activity applied, in terms  of biological activities, represented 92,400 mU of oxytocic, and 104,000 mU of vasopressor activity.  The unknown was eluted from  the column with 0.2M acetic acid, and the eluate was collected  53.  Figure 5. Gel filtration of the neurohypophysial extracts of the adult seal on G-15 Sephadex.  Loaded 92,400 rat uterus  and 104,000 vasopressor activity 2.8 ml/tube Open circles and broken lines = rat uterus activity Full circles and solid line  - vasopressor activity  o o Rat uterus e——• Vasopressor 5000i  4000H  I  100  1  1 2 0  1  M O  1 160  tube  i ISO  number  i  2 0 0  54.  in 2.8 ml fractions by means of a fraction collector.  The eluate  fractions were assayed for their oxytocic and vasopressor activities.  No attempt was made to measure conductivity or total  peptide concentrations in this experiment;  pilot runs had  already established that no changes in conductivity could be detected in the eluates, and that the major peak of proteins and peptides came off the column well before the appearance of the active agents (see Appendix, p. 189).  Further, each extra  measurement added to the risk of contaminating the samples with amino acid substances.  The results of the passage through the  column are shown in Fig. 5.  The oxytocic and vasopressor  activities were separated from.one another, and the recoveries were good, approximating 100% for oxytocic and 80% for vasopressor activity. The fractions with identical biological activities were pooled, lyophilised, and hydrolysed in 6 N HCI.  The amino acid  content of the hydrolysate was analysed on a Bio-Cal 200 analyser, and the results, together with the theoretical amino acid compositions, are shown in Table II. The analysis of the vasopressor fractions yielded the eight amino acids of arginine vasopressin.  The recovery of tyrosine (Tyr) was somewhat lower  than the theoretical value, and some of the cystein (% Cys) was oxidised to cysteic acid, both common occurrences during hydrolysis of small amounts of neurohypophysial hormones (Pickering, 1967).  When allowance was made for these known  TABLE II Amino acid analysis of seal neurohypophysial peptides, purified on Sephadex G-l5 Asp = 1.0 Amino acids  Seal vasopressor Experimental  Seal oxytocic  Theoretical  Experimental  Theoretical  Asp  1.0  1.0  1.0  1.0  Glu  1.21  1.0  2.21  1.0  Pro  1.18  1.0  0.57  1.0  Gly  1.44  1.0  0.84  1.0  h Cys  0.52 )  2.0  0.46 )  2.0  Cys acid  0.80 ) 1.32  0.48 ) 0.81  lieu  0.33  1.0  Leu  0.09  0.0  1.38  1.0  Tyr  0.62  1.0  0.48  1.0  Phe  1.15  1.0  Arg  0.90  1.0  3.12  3.0  1.78  3.0  Ratio of additional  Ser. 0.16, Ala, 0.19,  Ser. 0.19, Ala, 0.28  residues to Asp.  Val, 0.13, Thr, 0.13 in  Val. 0.17, traces of  0.05 molar ratio  Threonine and Methionine  56.  errors, the molar ratios agreed well with those of arginine vasopressin:  indeed, the analysis appeared particularly good  when compared with similar data given in the literature for other purified neurohypophysial  principles (see Acher, 1966).  It was clear that the preparation was remarkably pure, since the total percentage of amino acid residues, not part of arginine vasopressin, was only 7.2%.  The oxytocic fractions  proved to be less pure than those of vasopressin.  Although  various amino acids which were not part of oxytocin were found to be in low proportion, the molar ratios of those amino acids which should form part of the molecule of oxytocin showed a considerable variation from the theoretical values.  This was the case  especially for glutamine (Glu) and isoleucine (lieu).  However,  the absence of arginine (Arg) and Phenylalanine (Phe) from the hydrolysate confirmed a complete separation of oxytocin from arginine vasopressin. (b)  Purification of the oxytocic agent: To obtain a chemically pure preparation of the oxytocic  agent, a second sample of crude extract was purified by gel f i l t r a t i o n , and was subjected to a further purification step on CM Sephadex ion exchange column, i . Gel filtration 10 ml of 3 mg/ml crude extract of pregnant seal neurohypophyses, which represented a total of 92,400 mU of rat uterus and 104,000 mU of vasopressor activity, was applied to a 200 cm  57.  Figure 6.  Gel filtration of the neurohypophysial extracts of pregnant seal on G-15 Sephadex.  Loaded: 92,400 mil of  oxytocic and 104,000 mU of vasopressor activity. 2.8 ml fractions per tube. Open circles and broken line = rat uterus activity Full circles and solid line  = vasopressor activity  o  o Rot  uterus  •  o Vasopressor  58.  Sephadex G-l5 column.  The experiment was carried out in an  identical manner to that described above.  The eluates were  assayed for rat uterus and vasopressor activities as before, and the results are shown in Fig. 6. Again, a separation of the two biological activities was obtained.  The oxytocic fractions  (tube #120 to 150, Fig. 6) were pooled and further purified on a CM Sephadex ion exchange column as described below, i i . Purification on CM Sephadex 70 ml of the oxytocic activity pooled from gel filtration (35,000 mU) was washed with ether to remove acetic acid from the solution (see Methods).  The ether which remained in the aqueous  phase was driven off by bubbling with N gas. 2  The resulting  solution was titrated with concentrated ammonium hydroxide (25% for ammonia), to pH5.0, and diluted to an electrical conductivity of 0.16 m. mho/cm, identical to that of the starting buffer. The final volume reached 200 ml.  This was then applied to a 15 x 1  cm CM Sephadex column, which had been previously equilibrated with the 0.00 2 M ammonium acetate starting buffer.  Elution was  carried out at a constant pH of 5.0, but using a concentration gradient from 0.002 M to 0.2 M ammonium acetate, at a flow rate of 10 ml/hr.  The biological activity eluted as a single peak  (see Fig. 7). The biologically active fractions were pooled and lyophilised. The dry residue was taken up in 2 ml of 6N HCI, and hydrolysed at 108°C for 18 hrs.  The hydrolysate was analysed for its amino acid  Figure 7. The purification of the oxytocic fraction with adult seal neurohypophyses on CM Sephadex. The activity is expressed as mil/ml, as assayed on the isolated rat uterus; Each tube contains 2.8 ml of eluate.  The gradient  used was from 0.002 M to 0.2 M ammonium acetate, a a pH of 5.0, and is shown at the top of the diagram  0-2  MpH5  tube  number  60.  content as before, and the results, together with the theoretical amino acid composition are shown in Table III. The eight amino acids of oxytocin were detected, in ratios closely approximating to the theoretical values.  Although cysteine (% Cys) was low,  the value became an acceptable approximation to the theoretical ratio when compounded with the quantity of i t s cysteic acid breakdown product.  The ratio of tyrosine was also low, but this  was as expected (see Discussion p..147).  It was clear from these  results, that the adult seal oxytocic moiety corresponded to oxytocin i t s e l f , and that i t was exceptionally pure.  TABLE III Amino acid analysis of seal oxytocin from CM Sephadex  Amino acids  Experimental  Theoretical  Asp  1.00  1.00  Glu  1.18  1.00  Pro  1.06  1.00  Gly  1.11  1.00  h Cys  0.65  Cys acid  0.89 )  lieu  0.70  1.00  Leu  0.86  1.00  Tyr  0.59  1.00  Ratio of additional residues to Asp.  j  2.0 1.55  Val. 0.09  -  62.  DISCUSSION: (a)  Environmental considerations: A survey of the neurohypophysial hormone content of a  great number of mammalian species (Table IV), shows that the species which have restricted fresh water supply differ from the others in two aspects of their neurohypophysial hormone content.  Firstly, the V/0 ratios are many times in favour of  vasopressin as is seen in the llama, the camel and the whales. Secondly, the neurohypophyses of these animals normally under osmotic stress appear to be much richer in hormonal activity, especially for vasopressin. At the onset of-these studies i t was thought possible, that the seal with its limited or non-existent access to fresh water might also exhibit both the high V/0 ratios and the abundant biological activities usually found in species under similar environmental conditions.  However, the results showed,  that the V/0 ratios in the seal were not significantly different from those of standard land mammals. This suggests that such high ratios as have been reported in the llama, the camel and the whales are not necessarily an adaptational requirement for survival in a marine environment.  It appears more likely that  the ratios of the active agents in the neurohypophysis are genetically determined.  A clear pattern of this nature has been  suggested by Heller (1966), and Table V, taken from his paper, illustrates this point well.  Provided that a clear pattern in the  63.  TABLE IV Biological activities in some mammalian species Vasopressor activity mU / gland  Species  Rat uterus activity mU / gland  Ratio V/0  Ref.  Rat  350  320  1.14  1  Guinea pig  600  240  2.43  1  Cat  3,870  3,250  1.19  1  Dog  7,050  6,950  1.03  1  Opossum  1,400-  5,800  330-  1,690  3.43-6.30  2  Wallaby  1,650-  3,700  710-  900  2.32-4.11  2  Red kangaroo  9,500- 15,000  2,400-  3,600  3.95-4.16  2  Hedgehog  2,300  3,800  0.61  2  Badger  1,950  1,400  1.39  2  Sheep  16,000  19,000  0.87  5  Indian elephant  19,600  35,000  0.56  2  African elephant  54,000  51 ,000  1.06  2  Rhinoceros  32,000  44,000  0.72  2  Zebra  12,000  29,000- 72,000  0.41  2  Llama  14,000  4,000- 6,000  Camel  132,000  Finback whale  2.4-3.6  2  40,000  3.3  3  200,000 *  60,000 *  3.3  4  Sperm whale  200,000 *  20,000 *  10.0  4  Harbour seal  55,600  37,000  1.5  6  Fur seal  80,000-140,000  60,000-100,000  1.2-1.7  -  Refs.  1.  Dicker and Tyler, 1953.  2. Ferguson and Heller, 1965  3.  Adamson, et a l . , 1956.  4. Geiling, 1939.  5. Vizsolyi, 1968.  6. Vizsolyi, unpublished.  * Approximate estimations calculated for this table from given wet weights and mU/mg activity of dry powder, on the basis of an assumed 80% water content.  TABLE V  Order or suborder Order: Marsupialia  Order: Perissodactyla  Order: Artiodactyla Suborder: Ruminantia  Suborder: Tylopoda  From Heller, 1966  Species  V/0 rati  American opossum  2.9  Australian opossum  6.2  Wallaby  3.8  Red kangaroo  4.8  Horse  0.93  Zebra  0.44  Tapir  0.72  Black rhinoceros  0.77  African buffalo  0.9  Kongoni  1.5  Topi  1.4  Blue wildbeeste  1.3  Kob  1.4  Bushbuck  1.4  Llama  3.1  Camel  3.6  65.  distribution of V/0.ratios.persists when-further species are investigated, i t would appear that the high ratios in species with restricted fresh water supply might not be a direct result of their environment.  Instead, either a genetically pre-  determined ratio in favour of vasopressin activity enabled these species to adapt to a marine or desert environment, or else the high ratios evolved as one possible response to the environment and became genetically fixed. On the other hand, the posterior pituitaries of the fur seal were found to be exceptionally rich in total hormonal activity even among species under osmotic stress.  The 60,000 -  100,000 mU of rat uterus activity, and the 80,000 - 140,000 mU of vasopressor activity per gland, recorded in the fur seal, is approximated only by the camel, and surpassed only by two species of whales (see Table IV).  These animals are not only  much larger than the fur seal (average weight 55 - 60 kg) but they also have in common with i t an environment with scarce fresh water supply.  From the results of the present studies i t  appears that among a l l the mammalian species so far investigated, for i t s size, the fur seal neurohypophysis potent.  is by far the most  It seems possible that in the fur seal an adaptation to  environmental stress had taken place in the amount of stored hormones, and possibly aided.in the successful adaptation of these animals to a marine l i f e .  66.  (b)  The neurohypophysial hormones during pregnancy: A striking fluctuation in the neurohypophysial hormone  levels was recorded during pregnancy.  The changes in activity  were found to be parallel for both vasopressor and oxytocic activity;  following an i n i t i a l accumulation of the active  agents at 0.31 of term, the level of the stored hormones was greatly depressed throughout most of the second half of gestation.  However, the changes in the two activities differed  from one another in relation to the controls.  At 0.19 and 0.82  of term, the vasopressor activity was the same as recorded in the controls, but i t appeared.to be depressed at mid-pregnancy. On the other hand, oxytocic activity was found to be elevated from control levels at all stages of pregnancy, except at 0.56 of term, the lowest point on the graph. No explanation can be offered for the consistently elevated oxytocic activities of the pregnant seal;  i t is  possible that either i t is a false impression due to an age difference in the controls, or that there is an i n i t i a l accumulation of this agent some time between mating and 0.19 of term.  From the point of view of this discussion, the parallel  nature of the changes in the stored levels of the two neurohypophysial hormones is the more striking observation. When the great fluctuations in their quantity during pregnancy is compared to the virtually unchanged potencies of the controls taken over the same period, there is a strong suggestion that the eliciting factor for these.changes is the state of pregnancy.  In support of  67.  this hypothesis is the evidence in the literature for an interaction between sex steroids and neurohypophysial hormones, which will be outlined below. At the level of the target organs, estrogen raises the resting potential of the uterine muscle and causes spontaneous contractions of the uterus (Munsick, 1968).  Progesterone seems  to produce a conduction block to depolarization, and in the progesterone dominated uterus, oxytocin in the usual doses, is ineffective in producing a contraction (Csapo, 1961 as quoted by Munsick, 1965).  Similarly, the sensitivity of the  target organs for vasopressin is also influenced by estrogen and progesterone;  sensitivity is increased in the presence of the  former and decreased by the latter steroid.  Cobo (1967) reported  a high threshold for antidiuretic responses to vasopressin in pregnant women, and Pickford (1966) found a similar effect of progesterone on the salt excretion and vascular responses to the same principle.  From this evidence i t appears that the target  organs for both neurohypophysial hormones become more sensitive in the presence of estrogen, and their sensitivity is diminished by the presence of progesterone. At the level of the hypothalamic nuclei, progesterone is also inhibitory.  Cross and Siler (1965) reported that the  response of the hypothalamic nuclei to genital stimulation is depressed in the presence of progesterone.  68.  Besides the evidence for an influence of steroid hormones on the production and action of neurohypophysial peptides, there is a suggestion in the literature that the neurohypophysial peptides, in turn, alter the levels of the steroid hormones themselves;  this might take place through the mediation of  gonadotrophs hormones of the adenohypophysis.  The role of  neurohypophysial peptides as a releasing factor for the gonadotrophs hormones have been reviewed by Martini (1966), and only one example will be given here.  It has been found that large  doses of oxytocin will bring cattle into premature estrous, and i t is possible that this effect is due to a controlling influence over gonadotrophin release (Labsethwar, et_ al_, 1964). Further support for a possible relationship between the sex steroids and the neurohypophysial hormones are the results obtained in pregnant sheep (Vizsolyi, 1968);  here, a depression  of the levels of both hormones was found parallel to the elevated progesterone levels at mid-pregnancy.  It seems likely  that in the seal, just as in the sheep, the sex steroids are involved in the changes of neurohypophysial hormone levels. However, there is an almost complete lack of data concerning the estrogen and progesterone levels in the fur seal.  Only indirect  evidence, mainly based on histological studies, suggests that the seal placenta secretes steroids as well as gonadotropins (Harrison, 1969).  In view of the lack of evidence in this  69.  species, the neurohypophysial hormone levels cannot be related with any confidence to the primary endocrine changes in the sex steroids.  Therefore, i t can only be pointed out that the major  fluctuations  in neurohypophysial hormone content correspond to  definite phases of reproduction,.and the parallel appearance of the major shift in neurohypophysial hormones with definite reproductive stages.will  be briefly outlined below.  At 0.31 of term (February), when the neurohypophysial hormone activities are at their highest, the luteal phase in the pregnant ovary has ended, and the degeneration of the corpus luteum begins.  At around 0.5 of term (March), follicular  development starts in the non-gravid ovary, and this corresponds to the lowest activities recorded in the neurohypophysis (0.56 of term, beginning of April).  This observation tends to support the  belief that the neurohypophysial hormone levels are influenced by shifts in the progesterone and estrogen levels, or;with the consequent changes in the adenohypophysial gonadotropins, but no definite conclusions can be drawn until parallel experiments on the levels of a l l three groups of hormones are carried out. (c)  The purification of the neurohypophysial hormones: As i t was pointed out earlier, the purification procedures  on the adult pituitaries were undertaken for two reasons. Firstly, the seal holds a special interest from the point of view of neurohypophysial investigations,  since i t is a marine mammal.  Secondly, i t was essential to establish purification techniques on the more easily available adult glands before attempting to  70.  work with fetal material.  The results of the purification and  amino acid analysis of the active agents have shown that the seal neurohypophysis contains two peptides: arginine vasopressin and oxytocin.  This result is not surprising, since these two peptides  are widely distributed throughout the mammals.  Oxytocin has been  found in a l l mammalian species so far.investigated, and arginine vasopressin was found to be replaced by lysine vasopressin in only the pig family (Suina) and in one strain of mutant mice (Stewart, 1968). neurohypophysial  This general distribution of the mammalian hormones had been found to be unaltered in the  whale species, the only marine.mammal in which the posterior pituitary hormones had been identified (Acher, ejt al_, 1963); the work reported here showed.that the principles of the seal, C. ursinus, were also similar to those of the whales, and also to the majority of land mammals. Although the results yielded nothing beyond those expected as far as the chemical nature of the seal peptides were concerned, they were considered important because of the purification method i t s e l f .  Purification of the neurohypophysial  hormones is hindered by several factors not commonly encountered in most biochemical investigations. In most cases the material available is extremely limited, and therefore detection of the active principles has to rely on sensitive bioassays;  this makes  i t impractical to.use strong solvents or highly concentrated salt solutions for eluting the peptides.  Further, the presence of the  71.  S-S bridge in these compounds makes them unstable on many of the resins commonly used (Morris and Morris, 1965), and as a result, the losses during the course of.purification are large.  The one-  step purification used for the analytical preparation of seal vasopressin in the present studies offered the advantage of small losses (not over 20%), and a lack of interference from the neutralized fractions of the acetic acid medium used.  In  particular, i t offered the advantage of yielding a highly purified material in a single step, so that unnecessary losses due to handling, chemical  treatment, or to storage between  steps could be avoided. The method is based on two side effects of the crosslinked dextrans of the G series of Sephadex. The more highly cross-1inked gels in this series (G-l0, 15, and 25) act as weak ion exchangers under certain conditions, and also they are able to retard the movement of aromatic compounds when chromatography is carried out in the appropriate solvent systems. These properties of G-l0 Sephadex have been exploited by Eaker and Porath (1967) for the separation of amino acids.  In the  experiments discussed here, the use of 0.2 M acetic acid as the eluting medium made use of the ion exchange capacity of the gel, so that the elution of the more basic of the two peptides, arginine vasopressin, was retarded.  In passing, i t should be  pointed out that.the two peptides in question also differ from one another in their aromatic amino acid content;  vasopressin  72.  contains the aromatic ring of phenylalanine in addition to that of tyrosine, which i t shares with oxytocin.  However, the  results to be presented later, during the purification of the fetal material, show beyond doubt that the separation was achieved by the ion exchange properties of the gel. The only critical step in this simple single step purification method proved to be the i n i t i a l pouring of the column.  Although the resolution of the two biologically active  peaks varied somewhat from one batch of gel to the other, good resolutionwas only achieved from columns built slowly, with the frequent additions of small portions of the gel slurry. Some of.the experiments which illustrate this point are shown in the Appendix (p. 187).  In contrast, i t appears that the  addition of the slurry in a single application promoted the settling of the particles in steadily decreasing sizes, so that there was a layering of the gel. This appeared to reduce the efficiency of separation. As the final result of this single-step purification procedure, the two biologically active peptides of the neurohypophysis were separated from one another, and although the oxytocin required further purification"before analysis, the vasopressin fraction eluted with analytical purity. It may be concluded that a satisfactory and rapid purification procedure was established for neurohypophysial  73.  peptides, and.that this method demonstrated that the adult fur seal, £. ursinus, stored oxytocin and arginine vasopressin in the pars nervosa of i t s pituitary.  74.  SECTION II THE NEUROHYPOPHYSIS DURING.FETAL DEVELOPMENT In the following account, results concerning four aspects of fetal neurohypophysial physiology in the fur seal, Callorhinus ursinus, will be presented: A.  Histochemical studies  B.  The pharmacological properties of the crude extracts  C. The purification and identification of the neurohypophysial peptides D.  The action of the neurohypophysial peptides on the embryonic membranes.  Of these, the studies of the pharmacological properties of the gland, and the purification and identification of the active agents were carried to completion.  The histological studies  were designed only as an anatomical background to the physiological studies;  they serve to show that the physiological studies rest  on a valid morphological background in this species, and they . form a useful link to relate the results obtained in the seal to those of other mammalian species. In addition, a few preliminary experiments concerning the action of neurohypophysial hormones on embryonic membranes are included; these introductory studies were carried out because the identification of the possible target organs of the fetal neurohypophysial hormones was considered the next most important phase in the understanding of fetal neurohypophysial physiology.  However, the experiments on  75.  physiological function, which are included here, serve merely as a pointer towards the possible course of future investigations.  76.  A.  Hi stochemi cal studi es of the neurohypophysi s duri ng fetal development 1.  Introduction:  The histochemical aspects of the development of the mammalian hypothalamo-hypophysial system have been best investigated in the human (Bernischke and McKay, 1953; Waidl and Semm, 1959).  The f i r s t detectable neurosecretion appears in the  supra-optic nucleus at the fourth lunar month of gestation (0.4 of term).  At this early stage, approximately 40% of the embryos  show a supra-optic nucleus which possesses several well developed nerve cells;  in these cells the cytoplasm is vacuolated, and  neurosecretory material can be found.  At the same developmental  stage only two out of seven specimens showed neurosecretion in the neurohypophysis i t s e l f .  By the f i f t h lunar month, stainable  neurosecretory material was detected in all elements of the system, and from-the beginning of the sixth month the secretory cells increased steadily in numbers. In the embryonic calf, neurosecretion can be seen in both the neurohypophysis and the hypothalamic nuclei at three months of gestation (0.4 of term), but secretory material is more pronounced in the supra-optic than in the para-ventricular nucleus (Kivalo and Talanti, 1957). In addition, the development of the system has been investigated in two standard laboratory species, the rat and the mouse (Rodeck and Caesar, 1956; Green and Breeman, 1955; Enemar,  77.  1961).  There is no indication in the literature that neuro-  secretory activity in the rat begins before birth. the  However, on  f i r s t day post-partum, the nerve cells of the supra-optic  nucleus were weakly and diffusely stained;  Gomori positive  granules could be seen clearly for the f i r s t time on the twelfth day post-partum.  The f i r s t stainable neurosecretory  material in the neurohypophysis could be detected ten days after birth, but the large droplets of neurosecretion termed Herring bodies - characteristic features of adult neurosecretory activity - could be found for the f i r s t time only on the fourteenth day post-partum.  On the basis of stainability, the  para-ventricular nucleus of the rat was 3 to 4 days behind the supra-optic nucleus in i t s development. In the mouse, the f i r s t neurosecretion was seen in the 18 to 19 day old embryo, and i t was localized in the center of the  neurohypophysis (Enemar, 1961).  Six days post-partum, the  distribution of secretory material in the neurohypophysis corresponded to that of the adult, and Herring bodies could be found in the neurohypophysial stalk and in the median eminence. However, stainable neurosecretory material does not appear in the  hypothalamic nuclei until 2 to 3 weeks after birth. The development of the neurosecretory apparatus in the  dog had been investigated by several groups of workers (Scharrer, 1954; Diepen, Engelhardt and Smith-Agreda, 1954; Bargmann, 1949; Rodeck and Caesar, 1956).  The findings are somewhat  78.  contradictory.  Scharrer (1954) found the f i r s t neurosecretory  activity at 7 days prior to birth in the supra-optic nucleus. Some of the other workers found neurosecretion in the hypothalamus only at a later stage of development, some as late as 7 days postpartum (Rodeck and Caesar, 1956).  Sometimes, as in the work of  Diepen and his co-workers, the hypothalamus was found to be void of neurosecretory material, at 3 to 5 days before birth, but stainable neurosecretory granules were already present in the neurohypophysis i t s e l f .  However, since the dog is reported to  have low hormonal activities in the hypothalamus in the adult stage (Vogt, 1953), the f i r s t appearance of neurosecretion in the hypothalamic nuclei of this unusual species is hard to relate to those found in other mammalian species. In summary i t becomes apparent that there are discrepancies between different species;  in some, the f i r s t stainable  neurosecretory material can be seen in the hypothalamic nuclei, while in others i t appears f i r s t in the pars nervosa i t s e l f .  In  the human and the rat, neurosecretion is found f i r s t in the hypothalamic nuclei, more specifically in the supra-optic nucleus. In the mouse, stainable neurosecretory granules are seen f i r s t in the pars nervosa, and only later in the nuclei. hand there is a general agreement on two points.  On the other Firstly, in  the hypothalamus the supra-optic nucleus i s the f i r s t area to show stainable material, and the para-ventricular nucleus lags  79.  behind i t in development.  Secondly, i t is generally agreed that  neurosecretory material, at the time of i t s f i r s t appearance, is weakly and diffusely staining, and that Herring bodies can be found only at later stages of development. 2. Materials and methods: a.) Materials. Specimens at five stages of development were collected for histological studies.  Three of these specimens, collected  at 0.25 of term, at 0.30 of term and at 0.50 of term respectively, were treated in the following manner. the  The head of  fetus was separated.from the body, immediately following a  Caesarian delivery.  The cranium was cut to expose the brain,  and the specimen was placed immediately into a jar of Bouin's fixative. the  After a 48 hour period, the fixative was decanted and  jar was f i l l e d with 70% ethanol. One fetus at 0.75 of term  was received as a gift through the kindness of Dr. Harry; was already fixed in formalin.  this  On arrival at the laboratory the  brain was post-fixed in Bouin's solution in the manner described above.  The last seal was collected shortly after birth  (between 1 to 3 days) at a time when the placenta was s t i l l attached to the pup.  I am indebted to Dr. H. Dean Fisher for  supplying the specimen.  The pup was killed with concentrated Na  pentobarbital and the brain with the pituitary attached was dissected and further fixed in Bouin's solution for 48 hours.  80.  At the end of the 48 hours the fixative was decanted, and the jar was f i l l e d with 70% ethanol. A block of tissue was cut from each of the fixed specimens; this block extended from the optic chiasma to the posterior tip of the pituitary, and included approximately 1.5 cm of brain tissue at each side of the pituitary.  The block was embedded in  paraffin, and cut into sections of 8u thickness, b.) Methods. The mounted sections were post fixed for 16 hours in Bouin's fixative (Gomori, 1941), and were stained with either aldehyde fuchsin (Dawson, 1953; Hal mi,  1952) or with alcian blue-  Schiff's orange technique (van Oordt, personal communication, 1971).  Both staining procedures involved an i n i t i a l oxidation  step of the SH groups present in the neurohypophysial peptides. In both staining methods, the oxidation was accomplished by the use of potassium permanganate.  In both cases the specificity  of the stain was due to i t s affinity for the SO2OH groups, produced by the oxidation (Sloper, 1966).  The details of the  staining procedures are given in the Appendix. 3. a.)  Results:  Gross anatomy. The gross anatomy of the fetal neurohypophysial system  at 0.25 of term is illustrated diagrammatically in Fig. 8; i t i s also shown in the photograph of a sagittal section through the  TABLE VI The percentage and diameter of cells in the hypothalamic nuclei of seal fetuses and of the newborn pup with stainable neurosecretory material in the cytoplasm  Proportion of term  Para-ventricular nucleus  Supra-optic nucleus (1) (2)  cells with neurosecretion  cells with neurosecretion percentage  (3) K  'diameter  0.30  0  0  0.50  52  8.5  1.0  92  (1)  percentage  diameter  0  0  JU  14  8.2  12.0 u  85  21.5 u  All cells which showed positive reaction, however slight, were considered to contain neurosecretory material.  (2)  All percentages are based on a count of 50 cells.  (3)  Diameter of cells are based on the average of 40 measurements.  w  82.  median eminence and the pituitary of a fetus of the same age in Plate I. All components of the pituitary could be identified clearly at this early stage of development, and only two noteworthy features of the gland will be pointed out here.  Firstly, the pars tuberalis of the adenohypophysis (pt)  is exceptionally extensive and well developed in these specimens.  Secondly, the infundibular stem (s), which is large  in proportion to.the actual pars nervosa, extends intact far into the adenohypophysis.  This second feature of the fetal seal  neurohypophysis is also illustrated in the cross sections of the gland.  Plate II (A and B) shows cross sections of the fetal  pituitary at 0.5 of term, approximately midway along the adenohypophysis in the antero-posterior direction.  With either  the alcian blue (A) or the aldehyde fuchsin staining technique (B), two distinct areas appeared in the neurohypophysis.  The  upper area, highly vascularized, shows no neurosecretory material and is probably equivalent to that anatomic structure present in the sheep and termed the lower infundibular stem by Pritchard (1966).  The ventral area shows a diffuse but marked  staining for neurosecretory material, and represents the pars nervosa i t s e l f . The approximate locations of the two hypothalamic nuclei are shown in Fig. 8.  The supra-optic nucleus formed extensive  sheets of neurosecretory cells, confined to the dorsal surface of the optic chiasma, and extending laterally around i t s outer  83.  limits.  The dorsal surface of the nucleus was closely associated  with the central lining of the ventricle, b.) The hypothalamic nuclei. i.The supra-optic nucleus: The neurosecretory activity of the supra-optic nucleus at three stages of development is shown in Plate III and Table VI. At 0.3 of term (Plate IIIA), the morphological elements of the apparatus were already present.  However, no stainable material  could be detected in any of the cells, as is illustrated by the completely white cytoplasm surrounding the cell nuclei in Plate IIIA. At 0.5 of term (Plate 111B) diffusely staining granular material could be detected in several neurons.  The stainable  granular layer around the cell nuclei varied in size from a narrow rim surrounding i t , to extensive areas of stainable cytoplasm.  In a sample cell count of 50 cells (see Table VI), 52%  of the neurons were found to contain variable amounts of neurosecretory material , while the remaining cells appeared totally empty.  The non-secretory neurons were similar in appearance to  those seen in the fetus at 0.3 of term.  The neurons with neuro-  secretory material in their cytoplasm reached an average size of 8.5u. In the newborn pup, almost a l l the neurons were f i l l e d with densely staining neurosecretory material (Plate 111C). A sample cell count of 50 cells indicated that 92% of the cells showed positive staining for neurosecretion (Table VI).  At the same time  84.  the average diameter of the cells increased to 12u as compared to the smaller (8ju) cells of the fetus at mid-term. The results from these specimens suggest that neurosecretion in the supra-optic nucleus of the fur seal begins between 0.3 and 0.5 of term.  At 0.5 of term, the relatively  large proportion of cells, which were already f i l l e d with stainable material suggests that from the onset of secretory activity, the process to full production takes place within a narrow span of gestation. ii.  The para-ventricular nucleus: The morphological and functional development of the  para-ventricular nucleus is shown in Plates IV and V, and in Table VI.  At 0.3 of term, the area of the para-ventricular  nucleus (Plate IVA) shows only the fibers of a well developed tract when i t is viewed under low power (Plate IVA).  At high  magnification (Plate VA), the fibers of this unidentified tract and some cells are visible, but no neurons could be seen with evidence of neurosecretory activity.  The cell nuclei appear  small and elongated, and the cytoplasm is void of all stainable neurosecretory granules. At 0.5 of term, the general appearance of the paraventricular nucleus is s t i l l very much like that of the younger specimen (Plate IVB).  At a magnification of 1,100 a narrow ring  of alcian blue positive material was found to surround the nucleus of some cells (Plate VB).  However only 14% of the cells  contained even this low level of neurosecretion, whilst the rest  85.  remained empty (Table VI).  The average diameter of the secretory  cells was comparable to those of the supra-optic nucleus at the same stage of gestation, and measured 8.2u. At full term, the para-ventricular nucleus contained numerous cells with their cytoplasm f i l l e d with densely staining neurosecretory material. IVC.  A view of the area is shown in Plate  The same area under high magnification (Plate VC) shows the  positive reaction of the cytoplasm of individual cells.  It is  estimated on the basis of a sample cell count of 50 c e l l s , that approximately 85% of the para-ventricular neurons show a positive reaction.  At the same time the average diameter of the cells  increased to 21.5M, reflecting the great expanse of stained cytoplasm (Table VI). The same results were obtained with either the aldehyde fuchsin or the alcian blue staining techniques.  In Plate VI, two  consecutive slides of the para-ventricular nucleus are shown from the newborn pup. The cytoplasm is extensively stained by either staining method.  Although parallel staining was carried out on  all specimens, the results obtained for the two hypothalamic nuclei are illustrated only by the sections stained with alcian blue-Schiff's orange stain.  This is because the greater contrast  produced by the alcian blue stain allowed for better photographic reproduction. In summary, the results indicated that neurosecretory activity in the para-ventricular nucleus begins at approximately mid-term, but at this time i t is considerably less extensive than  86.  that found in the supra-optic nucleus.  At birth, the majority  of the cells within the nucleus were active, c.) The pars nervosa. The distribution of neurosecretory material in the pars nervosa at three stages of embryonic development is shown in Plate VII.  In the specimen at 0.3 of term (Plate VIIA), the pars  nervosa appears to be completely void of stainable material. Only the pituicytes and one blood vessel at the centre of the field could be positively identified. At.0.5 of term, aldehyde fuchsin positive granules were found to be scattered throughout the gland (Plate VIIB).  However,  no extensive aggregation of colloid droplets were noted, and on the whole, the staining appeared diffuse. By 0.75 of term the pars nervosa was f i l l e d with densely staining colloid material, as is shown in Plate VIIC.  Although  the stainable neurosecretory granules were found to be highly aggregated in this preparation, in the entire course of these investigations no large Herring bodies, the characteristic droplet formation of neurosecretory material found in adults, were encountered. In passing, another noteworthy feature of the neurosecretory apparatus in the seal fetus should be pointed out.  At the frontal  level of the para-ventricular nucleus, located in the proximity of the ventral surface of the brain, a bilateral area with extensive sheets of neurosecretory cells was noted.  Its location suggests  87.  that i t might be the infundibular nucleus, a relatively less conspicuous nucleus in other mammalian species (Sloper, 1966). This same area also contained a few neurons with stainable neurosecretion  in the fetus at 0.3 of term, when a l l other areas  were found to be void of secretory activity. 4.  Discussion:  Although the results obtained in the seal fetuses were in general agreement with studies of the fetal neurohypophysial system in other species, some details of the anatomy of the system, as well as the quantity of neurosecretory material were found to be unusual.  In regard to the anatomical arrangement of the  hypothalamo-hypophysial system in fJ. ursinus, two features were found to be in contrast with that described for the common seal, P_. vitulina, by Amoroso (1965) and Harrison (1969).  Firstly, the  infundibular stem in the fetal fur seal was found to be extensively developed, and i t passed a considerable distance into the adenohypophysis before i t formed the pars nervosa i t s e l f ; in contrast, the common seal i s noted for its exceptionally small, almost indistinguishable infundibular stem. Although the present studies did not include any adults, the appearance of comparable sections in the paper by Fuse (1939) indicates that the same situation exists in the adult fur seal.  However, an ambiguity  remains concerning this point, since at the time of Fuse's investigations no specific staining methods for neurosecretory material were available, and unfortunately general stains do not show clearly any division in the neurohypophysis.  Secondly, the  88.  fur seal also appears different from the common seal in the location of i t s para-ventricular nucleus.  In the common seal i t  is located only slightly posterior to the supra-optic nucleus (see Harrison, 1969);  in the fur seal i t is in a much more  posterior position, located at a level well past the beginning of the adenohypophysis. On the other hand, the fur seal was found to be similar to the common seal in the richness of i t s neurosecretion. It had been observed in the human, the rat and the dog that at birth or shortly after birth, the hypothalamic nuclei stain to a much lesser degree than the pars nervosa (see Harrison, 1969).  Only  in the common seal . (Amoroso et al_, 1965), and in the studies of the fur seal presented here, had the hypothalamus been found to be rich in neurosecretory material at the time of birth.  In the  fur seal, at the time of birth, not only were the supra-optic and para-ventricular nuclei rich in neurosecretory material, but also a secondary hypothalamic nucleus, the infundibular nucleus was found to contain extensive sheets of deeply staining neurons. The infundibular nucleus in other mammalian species is noted to be less marked than the supra-optic and the para-ventricular nuclei (Sloper, 1966).  A further interest in this extensive  staining of the infundibular nucleus is provided by the fact that at 0.3 of term, this area was the only one in the fur seal in which neurosecretory material was detected.  If further research  could prove this site to be the f i r s t to develop during embryo-  89.  genesis, i t might possibly explain why biological activity is commonly detected in the neurohypophysis,  before any stainable  neurosecretory material becomes visible in the classical regions of the neurohypophysis (see Heller, 1959 ). yet no evidence that neurohypophysial  However, there is as  principles are produced in  the infundibular nucleus. As far as the process of maturation of the hypothalamohypophysial system is concerned, the results obtained in the fur seal are in good agreement with those found in the majority of other mammalian species. The beginning of neurosecretory activity in the hypothalamic nuclei of the fur seal takes place sometime between 0.3 and 0.5 of term.  Although at 0.5 of term a l l  components of the system showed stainable neurosecretory material, the relative distribution of this secretion suggested that the time of maturation might have been.different for the two;,nuclei. At mid-term, when one half of the cells in the supra-optic nucleus already contained stainable neurosecretory material, the para-ventricular nucleus showed only the beginnings of secretory activity.  At this developmental  stage, the pars nervosa i t s e l f  contained only sparsely distributed neurosecretory granules.  By  the time of birth, a l l three components of the system were found to be rich in neurosecretion.  These results appear to be parallel  to those found in the human, where neurosecretion was f i r s t seen in a l l three components of the neurohypophysial term.  system at 0.5 of  They are also in reasonable agreement with those results  reported in the cow, where a l l three elements were found to be  90.  functioning for the f i r s t time at approximately 0.4 of term. Although the specimens studied in the fur seal did not include any in which neurosecretion was just beginning, the range for the onset of neurosecretory activity has been narrowed down to between 0.3 and 0.5 of term.  In addition, the relative  distribution of neurosecretory material at mid-term indicated that the supra-optic nucleus was more active than the para-ventricular nucleus, and that i t was also richer in neurosecretion than the pars nervosa.  This finding would suggest that the supra-optic  nucleus might be the f i r s t component of the neurohypophysial system to begin neurosecretory activity.  For definite conclusions,  regarding the maturation of the hypothalamo-hypophysial  system of  the fur seal, considerably more specimens should be examined. However, the experiments described above form a foundation for further research in the subject; they establish the existence of all elements of the neurohypophysial system in the fur seal, and show that neurosecretory material can be found from an early stage of development.  On this basis i t was reasonable to attempt a  study of the pharmacology and chemistry of the fetal principles; indeed the presence:of unusually large'amounts of neurosecretory material during fetal l i f e suggested'that C_. ursinus was a good species on which to base such studies.  Further studies of the  biological activities of the fetal neurohypophysial system will be given later, and these results were in good agreement with the brief histological background presented here.  Figure 8. Diagram of the hypothalamo-hypophysial system of the seal fetus.  Drawn from a sagittal section of the  area, from a fetus at 0.25 of term.  Mammillary  body  r  Paraventricular nucleus Pars nervosa  Pars distalis  d  ventricle  Optic chiasma  92.  Plate I. Sagittal section through the hypothalamus and the pituitary of a seal fetus at 0.25 of term. blue-Schiff's orange, x 33. mam. body = mammillary body pn  = pars nervosa  pd  = pars distal is  pt  = pars tuberalis  v  = ventricle  s  = stalk  Alcian  93.  Plate II. Cross section of the pituitary of the seal fetus. A.  0.5 of term.  B. 0.5 of term.  Alcian blue-Schiff's orange, x 33 Aldehyde fuchsin stain, x 33  C. A more posterior section from the same specimens. Aldehyde fuchsin stain, x 82. s  = stalk  pi  =  pars intermedia  c  =  cleft  pd  =  pars distal is  pn  =  pars nervosa  Plate III. Supra-optic nucleus of seal fetuses and newborn seal. Alcian blue-Schiff s orange stain. 1  A.  0.3 of term.  Magnified 1,100 times.  Arrow points to neuronal cell body.  Note the translucent cytoplasm surrounding the nucleus. B.  0.5 of term.  Arrows  a: indicate neurons with secretory activity b: indicate neurons with translucent cytoplasm.  C.  At birth.  Arrow points to one of the nuclei with deeply  staining neurosecretory material.  95.  Plate IV. View of the paraventricular area in the seal fetus at three different stages of development. Alcian blue-Schiff's orange, x 175. A.  0.3 of term  B.  0.5 of term  C.  At birth  tpv  =  tract transversing the para-ventricular nucleus.  pvn  =  neurosecretory cells of the para-ventricular nucleus.  96.  Plate V. Para-ventricular nucleus of seal fetuses and newborn seal at three stages of development. Alcian blue-Schiff s orange, x 1,100 A.  0.3 of term. Arrow points to the neuronal cell body. Note the complete absence of neurosecretory material.  B.  0.5 of term. Arrow  a: cell body with slight neurosecretory activity b.  cell body with no stainable neurosecretion.  C.  At birth.  Arrow indicates cell body with cytoplasm that  is f i l l e d with alcian blue positive material.  Plate VI. Comparison of the appearance of the para-ventricular nucleus with two different staining techniques. A. Aldehyde fuschin stain.  Magn. 1,100.  Note the granular neurosecretory  material surrounding the cell nucleus. B.  Consecutive slide to the above, stained with alcian blueSchiff 's orange stain. cytoplasm.  Note the densely stained  98.  Plate VII. The pars nervosa of seal fetuses and newborn seal. Aldehyde fuschin stain. A.  0.3 of term.  1,100 magnification.  Note the absence of granular material,  prominent pituicytes and centrally placed blood vessel. B.  0.5 of term.  Granular neurosecretory material  scattered throughout the pars nervosa. C.  0.75 of term. granules.  Aggregated deeply staining neurosecretory  0 75  99.  B.  The biological activities of the fetal neurohypophysis 1.  Introduction:  The limited literature concerning the biological activities of the fetal neurohypophysis has been discussed in detail in the Introduction (p. 11). In summary, i t can be said that the vasopressor and oxytocic activities of the posterior pituitary had been investigated in few species only and except in two of these species, the human and the sheep, the studies were limited to the last trimester of gestation.  Furthermore, the fact that the fetal  neurohypophysis contains a principle with high frog bladder activity was a recent observation  (Vizsolyi and Perks, 1969), and  therefore earlier pharmacological studies did not include the estimation of this biological effect.  The investigations here  were designed to include tissues taken at a wide span of gestation, and to estimate the previously unrecorded frog bladder activities during fetal l i f e .  For this reason, the posterior pituitaries of  seal fetuses were studied at seven stages of gestation, from 0.19 to 0.93 of term.  To ensure that the glands in this series were as  close to the same gestational stage as possible, only the posterior pituitaries of fetuses collected within the f i r s t 3 days of each month were included.  It should be noted that, in the fur seal, the  duration of intrauterine development from implantation until birth is approximately 240 days; this stretches from mid November until mid July each year, with a possible variation of plus or minus two weeks from specimen to specimen (Craig, 1964).  In expressing the  stage of development of each fetus, the average time for  TABLE VII The biological activities of the fetal seal neurohypophysis at seven stages of gestation Proportion of term  Number of glands  Av.Weight of glands*  Vasopressor mU/mg  Rat uterus mU/mg  0.19  10  0.19  115.7  30.3  0.31  15  0.30  280.4 + 54.0  68.4 + 6.8  0.44  12  0.90  760.2 + 74.5  0.56  30  1.5  0.68  10  0.82 0.93  Frog bladder mU/mg  V/RU ratio 3.81  16.4  1,500  4.10 + 1.10  22.0  183.6 + 16.2  1,800  4.14 + 0.53  10.0  1,196 + 173  238.2 + 39.3  2,500  5.02 + 1.00  10.5  2.3  1 ,227.5 + 248  391.0 + 53.0  4,500  3.16 + 0.70  11.6  8  3.5  2,358.0 + 179  1 ,289.6 + 120  3,300  1.96 + 0.25  2.5  6  3.2  2,824.9 + 251  1,201.0 + 58  2,200  2.35 + 0.40  1.8  Biological activities are expressed as mU/mg lyophilised tissue.  500  95% fiducal limits are given only  where estimations are based on at least three groups. *  FB/RU  Average weights of glands were calculated by simple division of the total weight by the number of glands.  101.  implantation (mid November, Craig, 1964) is taken as 0.00 of term and birth is given the value of 1.0;  the proportion of term is  calculated between these limits. 2.  Results:  At each stage of gestation, between 6-12 glands were pooled, depending on their availability.  They were weighed,  extracted in 0.25% acetic acid and assayed for their vasopressor, rat uterus and frog-bladder activities (see Methods, p. 20). The results are shown in Table VII and Figs. 9 and 10.  All three  biological activities were detected from the earliest stage of embryonic development investigated.  At 0.19 of term, the 115 mil/  mg vasopressor, 30 mU/mg oxytocic, and approximately 500 mU/mg frog bladder activity equals 22 mU, 6 mU and 95 mU of activity per gland, respectively.. This is a small amount of hormonal activity, but in comparison to the human, where biological activities were found to be too small for quantitative estimation up to 0.39 of term (110 days, Dicker and Tyler, 1953), this potency is surprisingly high. The oxytocic activity of the neurohypophysis showed a general accumulation with each advancing stage of gestation. However, there appears to be one period during development where a sudden increase in the accumulated activity can be detected; is between 0.68 and 0.82 of term.  this  This sudden rise in potency  brought the oxytocic level to its highest value during gestation, and no further accumulation could be detected at the last stage of fetal l i f e studied here.  4 0 -  30-  1 0  "  -i  0 2  1  1  1  1  1  1  0-3  0 4  OS  0 6  0-7  0 8  i  0-9  i~  10  proportion of term  Figure 9.  (above)  Average dry weight of fetal posterior pituitary glands through gestation.  Weights were calculated by simple division of the  total weight by the number of glands in the sample. Figure 10. (opposite page) Biological activities of the fetal neurohypophysis per mg lyophilised tissue.  Vertical  bars represent fiducal  P = 0.05. Triangles and solid line  =  frog bladder activity  Open circles and broken line =  vasopressor activity  Full circles and solid line  rat uterus activity  =  limits at  103.  The vasopressor activity showed a pattern of accumulation similar to that found for the oxytocic activity, but the amounts present were always greater than those of oxytocic activity. Parallel to the time of rapid accumulation of oxytocic activity (0.68 to 0.85 of term), a sharp rise in vasopressor activity was also detected.  Within this 30 day period the vasopressor  activity doubled, and a further slight accumulation of activity, detected at 0.93 of term, accounted for only a small portion of the total vasopressor activity recorded at the end of fetal  life.  The changes in potency of the extracts in promoting the movement of water across the isolated frog bladder (frog bladder assay), showed a different pattern. This bio-assay is commonly used for the detection of arginine vasotocin in posterior pituitary extracts, due to the high sensitivity of this biological preparation to this particular peptide.  An increasing  accumulation of frog bladder activity was recorded until 0.68 of term, after which time the potency rapidly diminished;  at 0.93  of term i t was less than the vasopressor activity and only slightly higher than the oxytocic potency.  The reversal in the  accumulation of this activity corresponds to the time of the greatest rate of increase in oxytocin and vasopressin. It appears from these results, that in terms of the bioassays, the predominant biological activity in the f i r s t two trimesters of gestation is the frog bladder activity, which is rapidly diminishing in the last trimester of intrauterine l i f e . However, i t should be pointed' out, that in terms of the molar  104.  content of the peptides in the pituitary, vasopressin predominates throughout gestation, and the high frog bladder activities are the result of the exceptionally high potency of arginine vasotocin on the isolated frog bladder.  At the peak of the frog bladder  activity, at 0.68 of term, the 4,500 mU/mg of frog bladder activity can be attributed to the presence of only 0.1 jug of arginine vasotocin, i f i t is assumed that this peptide is responsible for the effect.  At the same stage of gestation, the  1,200 mU/mg of vasopressor activity requires the presence of 2 jug of arginine vasopressin, so that in terms of the weight of the two peptides produced, arginine vasopressin is present in 20 times greater quantity than arginine vasotocin. The ratios of vasopressor to oxytocic activity (V/RU ratio), and of the frog bladder to rat uterus activity (FB/RU ratio) were calculated, and they are shown in Table VII and Fig. 11. The highest V/RU ratio, a value of 5.0, was recorded at 0.56 of term. At 0.19 of term, the ratio was found to be only 3.8, slowly rising in the next three stages of gestation to reach the value of 5.0 at 0.56 of term.  In the second half of gestation the V/RU ratio  declined with each successive stage, until at 0.93 of term i t dropped to a value of 2.0. The FB/RU ratio gave an i n i t i a l rise from 0.19 to 0.31 of term, and then steadily declined, until at 0.93 of term i t approximated  to 1.0; this value was the same as  in the adult, and i t is characteristic of oxytocin alone. Arginine vasopressin shows only weak frog bladder activity, and  105.  Figure 11. Ratios of vasopressor to oxytocic (V/RU) and frog bladder to oxytocic (FB/RU) activities of the fetal seal neurohypophyses.  The V/RU ratios were calculated as  vasopressor activity per mg lyophilised activity per mg lyophilised tissue.  tissue/oxytocic  FB/RU ratio was  calculated as frog bladder activity per mg lyophilised tissue/oxytocic  activity per mg lyophilised tissue.  Vertical bars represent fiducal limits at P = 0.05. Open circles, broken line  =  FB/RU ratio  Full circles, solid line  =  VP/RU ratio  106.  its presence would not have effected this ratio to any significant extent. 3.  Discussion:  The vasopressor and oxytocic activities per mg dry tissue were found to increase at each advancing stage of gestation, a trend of accumulation which is magnified further i f the size of the glands is also taken into consideration.  These results reflect the  growth and the maturation of the fetuses, and are in no way surprising.  However, some of the observations made during these  studies merit further comment.  In the seal, biological activities  were not only detected at an early stage of gestation, but at 0.19 of term the 22 mU of vasopressor activity and the 6 mU of rat uterus activity per gland was already sufficient for quantitative estimation.  In contrast, the pituitaries of human fetuses  contained sufficient biological activities for quantitative estimation only after 110 days of gestation (0.40 of term), when vasopressor activity assayed at 53 mU/gland and oxytocic activity was 1.9 mU/gland (Dicker and Tyler, 1953).  Late cat fetuses at 56  days of gestation (0.85 of term), were also shown to have only 56 mU of vasopressor and 13 mU of oxytocic activity per gland (Dicker and Tyler, 1953).  In the studies presented here, the vasopressor  activity was not only exceptionally high at an early stage of gestation, but was also found to be unusually high in the near term seal fetuses.  The 2,824 mU/mg potency represents a total of 9,040  mU of activity per gland.  In contrast, in the newborn human the  total vasopressor activity was found to be only 375 mU/gland, and in the newborn cat the total activity for this hormone was only  107.  160 mil/gland (Dicker and Tyler, 1953). neurohypophysial  The early appearance of  activities in the seal, and the unusually high  potencies present as birth approaches, could possibly reflect two factors: the advanced maturity of the seal pups at birth, and the unusually high biological activities found in the adult gland of this marine mammal.  In support of the f i r s t contention  i t may be stated that many workers have considered that the relative richness of the neurohypophysis is directly related to the maturity of the specimens at birth (see Heller, 1961). As for the high potencies in early embryonic l i f e , i t is not unreasonable to suppose that a species which is rich in neurosecretory material at the adult stage,.might well be richer than usual in the biological activities present during intrauterine life. The second point of interest, and the most striking feature of these investigations is the apparent predominance of frog bladder activity during the f i r s t half of embryonic development. The agent responsible for this biological activity was rapidly accumulated until 0.68 of term, and i t s quantity diminished sharply after this point.  It is interesting to note that there is a  correspondence between the decline in frog bladder activity and the period of greatest accumulation activities.  of vasopressor and oxytocic  This changeover in biological activities is taking  place at a time of ontogenesis when fetuses have been covered with hair for about a month, and in appearance they are superficially indistinguishable from the newborn pups in a l l but size.  Such a  108.  correspondence between the rapid accumulation of the two biological activities characteristic of the adult, and the rapid disappearance of the "new" fetal principle - a principle formerly thought to be confined to non-mammalian vertebrates - seems more than a chance occurrence.  The factors eliciting this phenomena  cannot be suggested until a possible physiological role of the neurohypophysial agents in the fetus is understood, but the possibility exists that there are physiological stimuli for the production of one or the other antidiuretic agent. The V/RU ratios were found to be unusually low during fetal development.  In the acetone dried tissues of human fetuses  (Dicker and Tyler, 1953), as well as in the lyophilised glands of sheep (Vizsolyi and Perks, 1969) and guinea pigs (Kontor and Perks, unpublished) the vasopressor activity was found to be at least 15 times in excess of the oxytocic activity during the f i r s t half of intrauterine l i f e .  In contrast, the ratio of V/RU in the fur seal  fetuses at no time exceeded the value of 5.0.  Indeed, the results  indicated that in the f i r s t two trimesters of intrauterine l i f e the ratio might be even lower.  From 0.56 of term until near birth,  the V/RU ratio f e l l with each stage', until at 0.93 of term i t reached a value of 1.8, only slightly above that of the adults.  It  appeared from these results, that although vasopressor activity was greater than the oxytocic activity all through gestation, the predominance of this agent was greatest around mid-gestation.  Two  features in this pattern of the V/RU ratios are strikingly different  109.  from those recorded in other mammalian species.  Firstly, contrary  to the findings in other mammalian fetuses (see Results in this section), the ratio of the vasopressor agent to the uterotonic principle had not reached a notably high value at any stage of gestation studied here, and secondly, the change in this ratio of the vasopressor agent to the uterotonic principle did not follow the usual pattern of steadily decreasing values with each progressive stage of gestation.  It is not possible to state that  a rapid f a l l in V/RU ratios did not occur prior to 0.19 of term, but this gestational stage is so early that such a situation would seem unlikely. Further, the trend in the ratios suggested that in earlier fetuses, the ratio was even lower than at mid-term. On the basis of these observations, i t would appear that the fur seal shows a species variation from other mammalian species in the magnitude of i t s V/RU ratio during gestation as well as in the pattern of change in this ratio through gestation. In general, the FB/RU ratios showed a steadily declining pattern.  Since no previous attempt was made to estimate the frog  bladder activity in mammalian fetuses, i t cannot be certain whether this pattern is specific to the fur seal or represents a pattern characteristic to a l l fetal mammals. In summary, the pharmacological  studies have shown a  steadily increasing content of the vasopressor and oxytocic principles in the neurohypophysis during fetal development. The fur  seal was found to differ from other mammalian species in i t s  relatively low V/RU ratios during fetal l i f e , and in the pattern  no.  in which these ratios changed during ontogenesis. Most important, they yielded the estimation"of the frog bladder activities throughout gestation, and these values indicated the rise and f a l l of the arginine vasotocin-1ike principle during fetal l i f e .  This information provided the necessary data on  which the purification of this new neurohypophysial principle should be based.  111.  C.  The purification of the active agents from the fetal neurohypophysis Introduction:  general considerations of purification  procedures. During the studies of the pharmacological properties of the crude extracts, an increase of vasopressor and oxytocic activities were recorded with each advancing stage of development.  However,  in a third biological activity, the action on water-movement across the isolated frog bladder, the glands proved to be most potent at mid-term.  The purification procedures to be described  here, were carried out on fetal seal glands taken between 0.56 and 0.68 of term, and they were designed to identify the peptides responsible for a l l three of the biological activities which could be detected.  The pituitary of the fetal seal appeared to be the  tissue of choice for such purification studies, because of the unusually high levels of neurohypophysial activities found throughout a l l stages of gestation. The purification of the standard octapeptides of the mammalian neurohypophysis has been accomplished previously. Several methods for the isolation of the two standard principles proved successful, from the earliest cumbersome chemical precipitation methods (Kamm, ejt al_, 1928), to the more efficient but s t i l l involved purification by countercurrent distribution (du- Vigneaud.et al_, 1953).  In recent years the purification  methods have been greatly simplified by the use of modern ion exchange resins.  112.  The method of purification of the neurohypophysial  hormones  by ion exchange resins was pioneered by Taylor (1954), who separated arginine vasopressin from both lysine vasopressin and oxytocin on Amberlite IRC-50 resins by the use of a sodium phosphate buffer (0.2 M, pH 6.95).  In the late 1950 s the same 1  resin was used by Acher and his associates to separate the active neurohypophysial  agents of several mammalian species;  these  workers added a preliminary purification step in which larger proteins were precipitated with trichloroacetic acid (see Acher, 1966).  In addition, these authors introduced the use of a gradient  of salt concentration for the elution of the active principles, and they further modified the original method of Taylor by the use of ammonium acetate buffer as the eluant.  The importance of the use  of ammonium acetate lies in the fact that i t will sublime under vacuum, and therefore, i t can be removed and prevented from interfering with further analytical procedures.  In theory most  cation exchange resins should be capable of separating oxytocin from arginine vasopressin, and indeed, several of them have been successfully employed for this purpose. The relative ease by which.arginine vasopressin can be separated from oxytocin is due to the widely different pK of the two molecules.  The presence of arginine in arginine vasopressin  makes i t a strongly basic peptide, while the pK of oxytocin is close to neutrality, with a value of slightly above 6.0. Therefore, this phase of the purification of the seal neurohypophysial peptides did not present any unusual difficulty.  The  113.  particular problem in the purification of the fetal extracts was that arginine vasotocin was expected in addition to the two commonly occurring neurohypophysial arginine vasopressin.  peptides, oxytocin and  While arginine vasopressin differs from  oxytocin in two of i t s eight amino acid residues (see below)',' arginine vasotocin only differs from arginine vasopressin in one amino acid;  the isoleucine in position 3 of vasotocin is replaced  by phenylalanine in arginine vasopressin.  Since the pK of these  two amino acids is practically the same, substitution of one for the other alters the overall charge of the molecule only negligibly. Therefore, the differences in the chemical behaviour of these molecules could be expected to rest on the presence of the additional aromatic group of phenyalanine in vasopressin. Consequently, methods that are based on aromatic adsorption would have to be exploited for their separation. In addition to the similarity in the chemical nature of arginine vasotocin and'arginine vasopressin, a further difficulty in the separation of these two peptides was presented by the minute amounts of arginine vasotocin present in the crude extracts. The small quantity of material limited the choice of purification methods directly, in that only very small losses could be tolerated for a successful purification.  In addition, i t limited  the range of buffer concentrations which could be used, since the small quantity of peptides eluted required the use of sensitive biological assays for their detection, and strong solvents or high  114.  The amino acid sequence of neurohypophysial peptides  1.  ARGININE VASOTOCIN I  1  Cys-Tyr-Ile-Gln-Asn-Cys-Pro-Arg-Gly-NH?  1  2  3  4  5  6  7  8  9  ARGININE VASOPRESSIN  Cys-Tyr-Phe-G1n-Asn-Cys-Pro-Arg-Gly-NH  2  OXYTOCIN  Cys-Tyr-Ileu-Gln-Asn-Cys-Pro-Leu-Gly-NHg  115.  salt concentrations interfered with these preparations.  Due to  these considerations, the method finally used had to meet three main qualifications. exceeding 20%.  Firstly, the losses had to be minimal, not  Secondly, the solvent used had to allow for direct  detection of the active peptides by biological assays.  Thirdly,  the separation of the active peptides from one another had to be nearly complete;  this was because the ratio of arginine vaso-  pressin to arginine vasotocin was so high that without an almost complete separation of the two agents, the identification of arginine vasotocin by amino acid analysis would have been impossible - the amino acid residues of the trace peptide would have been swamped by those of the predominant peptide.  Several  methods have been reported in the literature as being capable of separating these two peptides, or of resolving similar compounds. These potential methods were tried, and their value in the purification of'the neurohypophysial  principles of the fetal seal,  were judged against the above criteria. In 1960, Chauvet, Lenci and Acher, reported the separation of oxytocin, arginine vasotocin and arginine vasopressin from the neurohypophysial  extracts of birds on Amberlite IRC-50 resins.  However, their elution curves show only two major peaks, one with rat uterus and the other with vasopressor activity.  There was no  clear separation of the two basic principles, and only a small depression in the vasopressor peak indicated the possible presence of two peptides.  The authors based their claim on the results of  amino acid analysis, the data for which was not included in their paper.  116.  In 1964, Munsick also claimed to.have succeeded in separating these 3 compounds by the use of a CM cellulose column. However, his conclusions are not altogether clear.  His f i r s t  peak appears to consist of oxytocin, although its almost immediate elution from the column might indicate that this peptide was at best weakly attached to the gel. is stated to consist of arginine vasotocin.  His second peak  Although a high hen  oxytocic/rat oxytocic ratio would tend to support this contention, the V/0 ratio within this peak is ten times too small to be accounted for by arginine vasotocin alone.  Since the vasopressor  and rat uterus assays are more reliable and accurate than the hen oxytocic method, the discrepancy in the ratio of these assays appears to weaken considerably the claim that this peak consisted of arginine vasotocin;  at best, i t might suggest considerable  contamination with oxytocin or arginine vasopressin.  The last  peak would appear to consist of arginine vasopressin, but i t is only poorly separated from the previous peak.  In any case, the  poor resolution of the two basic peptides achieved by both the above methods, at least under the conditions reported by the authors, made these methods unsuitable for analytical preparation of arginine vasotocin from the fetal seal extracts. Outside the neurohypophysial  f i e l d , several chromatographic  gels have been used to separate peptides from protein hydrolysates by the combined properties of their net charge and aromatic amino acid content.  In 1967 Eaker and Porath reported the separation of  117.  tripeptides according to their tryptophan content on Sephadex G-25,  and in 1965 Ruttenberg and his co-workers separated  the  three components of tyrocidine on Sephadex G-l5 by the use of their adsorption properties.  Moore and Stein (1953, 1956) used  the aromatic amino acid content and net charge of peptides for the fractionation of protein from enzyme hydrolysates on both Dowex and Amberlite resins. In the search for a method that would be suitable for the isolation-of fetal arginine vasotocin, most of these potentially successful resins were investigated, and the results from those which proved to be the most promising will be outlined with the results. Results: The purification and identification of the fetal neurohypophysial  agents were carried out by the following general  procedures: 1.  The estimation of the arginine vasotocin content of the extracts by paper chromatography.  2.  Gel filtration of the crude extracts on Sephadex 6-15.  3.  The ion exchange chromatography of oxytocin and arginine vasopressin.  4.  The separation of arginine vasotocin from arginine vasopressi n.  5.  The identification of the peptides by direct comparison with synthetic standards, and by amino acid analysis of their hydrolysates.  118.  In the following sections each of these steps will be considered separately. 1.  The paper chromatography of crude neurohypophysial extracts: The lyophilised neurohypophyses of 13 fetal seals, at 0.31  of term, were extracted in 1 ml of 0.25% acetic acid;  assays  showed that the sample contained a total of 131.0 mU of oxytocic and 538 mU of vasopressor activity.  0.96 ml of the original  extract was subjected to paper chromatography on Whatmann 3MM paper in a solvent system of butanol, acetic acid and water (4:1:5), for 13 hrs, at 25°C.  Standard oxytocin (150 mU;  Syntocinon, Sandoz) and standard vasopressin (500 mU;  Pitressin,  Parke Davies) were applied separately to the same chromatogram, alongside the unknown. The vasopressor, rat uterus and frog bladder activities were assayed in the 0.25% acetic acid eluates of the fetal chromatogram, as well as in those of the standards (see Methods, p. 20).  The results are shown in Fig. 12.  Synthetic  oxytocin eluted mainly in Rf 0.6, and vasopressin showed i t s major peak between Rf 0.2-0.3.  No detectable frog bladder activity was  found in any of the Rf-s of the standard strips. The chromatogram of the fetal material showed vasopressor activity between Rf 0.1 and 0.4;  this was in good agreement with  the position of the standard vasopressin. The highest rat uterus activity was detected in Rf-s 0.5 and 0.6, again in a position identical to that of standard oxytocin.  However, oxytocic  119.  Figure 12. The paper chromatography of fetal seal neurohypophysial extracts at 0.31 of term, in a solvent system of butanol, acetic acid and water, 4:1:5, on Whatmann 3MM  paper.  Standards : Syntocinon (Sandoz, 10 I.II./ml) and Pitressin, Parke Davies Co. 20 I.U./ml. Black bars  =  rat uterus activity  White bars  =  frog bladder activity  Striped bars  =  vasopressor activity  0 200  0.1  0.2  0.3  1  1  1  0.4  0.5  0.6  0.7  0.8  09  i  i  i  ....1  synthetic oxytocin  ;  0  rat uterus  _ _ _ _ r n  ^^L~^*  200 standard  vasopressin  100  H  vasopressor  3600 3  E  3500 500  Z  H  M M  seal foetus 90-100  days  of  gestation  400 I  300  1 frog  bladder  vasopressor rat  uterus  1.0  120.  activity was also found in a relatively large quantity, in Rf 0.3. Although a slight rat uterus effect might be expected in this region i f the weak intrinsic oxytocic activity of arginine vasopressin is taken into consideration, the potency of this eluate on the rat uterus  was 10 times too great to be attributed  to such an intrinsic activity.  The presence of this rat uterus  activity could not be explained by a possible trailing of the oxytocic agent, since i t was separated by a whole Rf unit from the major peak.  Rather, this evidence indicated the presence  of arginine vasotocin in this Rf, and this hypothesis was tested by frog bladder assay.  The frog bladder assay showed the remarkably  high value of 3,600 mU/ml. This was confined to Rf 0.3, which was the same position as the major peak of vasopressor activity, and in the same place as the excess oxytocic effect.  A calculation of  the ratios of the three biological activities within this Rf unit yielded a value of 100 for the FB/RU ratio, and of 3.0 for the V/RU ratio.  The only known naturally occurring neurohypophysial  analogue with such a high FB/RU activity is arginine vasotocin. In our laboratory, pure synthetic arginine vasotocin showed an FB/RU ratio of 100 to 200 - although values as high as 910 have been recorded in the hands of other workers (see Sawyer, 1965). The-V/RU ratio of synthetic arginine vasotocin is 1.8-2.0 (Sawyer, (1965)).  On the basis of these results i t appears that the  vasopressor activity within the fetal neurohypophysis could have been elicited partially by arginine vasotocin, mixed with a greater  121.  proportion of arginine vasopressin. 2.  Gel f i l t r a t i o n of crude neurohypophysial extracts: The chromatographic studies, discussed above, confirmed the  presence of a neurohypophysial agent with high frog bladder activity in the posterior pituitary of seal fetuses at 0.31 of term; agent.  this indicated the presence of a third neurohypophysial However, the hormonal activities at this early stage of  gestation were so small that their successful purification would have been a formidable task.  Therefore the purification  procedures based on column chromatography were carried out on more potent pituitaries collected from seal fetuses at a later stage of gestation, between 0.56 and 0.68 of term (month of April).  The tissues at this stage of gestation offered the  additional advantage that the frog bladder activity was at i t s height around this time (see Fig. 10), although at the same time the ratio between the frog bladder activity and the vasopressor activity also became much less. The posterior pituitaries of 42 fetuses, collected during April of 1968 were pooled and extracted in 0.25% acetic acid at 5 mg dry powder/ml.  The total sample volume was 16 ml.  One half  of this extract, 8 ml, which represented 7,600 mU of rat uterus and 16,000 mU of vasopressor activity was applied to a 200 cm x 2.5 cm column of G^15 Sephadex. were 0.2 M acetic acid.  The suspending and eluting media  After discarding the f i r s t 200 ml as the  void volume, the eluate was collected in 2.8 ml fractions.  The  peaks of biological activities were detected by rat uterus and  122.  Figure 13. The gel filtration of crude neurohypophysial extracts of seal fetuses at 0.56 to 0.68 of term, on a 200 cm Sephadex G-l5 column.  Loaded 7,600 mU of rat uterus and  16,000 mU of vasopressor activity.  The loading and  eluting medium was 0.2 M acetic acid. contained 2.8 ml fractions.  Biological  Each tube activities are  expressed in mU/ml of eluate. Open circles and broken line  = rat uterus activity  Full circles and solid line  = vasopressor activity  Peak  I  peak  2  a  tube  number  o-—o  Rot uterus  ©  Vasopressor  o  123.  vasopressor assays.  No attempt was made to measure the conduct-  ivity or total peptide content, since preliminary experiments had shown no detectable peak in conductivity, and the majority of protein ("Lowry peptide") had preceded the active fractions and been separated from them by at least 60 tubes (see Appendix, p. 189). The results of the gel filtration procedure are shown in Fig.  13.  The f i r s t peak eluted exhibited rat uterus activity,  but no detectable vasopressor potency,  i t appeared to correspond  to oxytocin. The vasopressor activity was eluted second, in a double peak which was well separated from the rat uterus activity. In the f i r s t half of the vasopressor peak, unusually high rat uterus activity was also detected;  this suggested that a mixture  of arginine vasotocin and arginine vasopressin might be present in this portion of the vasopressor peak, and i t was probable that this consisted of arginine vasopressin. The eluates were pooled in three lots, according to their biological activities; rat uterus fractions (peak 1) were pooled from tube numbers 110 to 145;  vasopressor peak 2a fractions were pooled from tubes  number 146 to 165, and finally the fractions of vasopressor peak 2b were combined from tubes 168 to 200.  The pooled eluates were  kept in a refrigerator at 4°C. Each of the three pooled lots were assayed for their rat uterus, vasopressor and frog bladder activities, in order to localize the position of arginine vasotocin. The results are given below and illustrated in Fig. 14.  124.  Figure 14. The biological activities in the three peaks eluted from the G-l5 Sephadex column;  fetal neurohypophysial extract  at 0.56 to 0.68 of term. Black columns  =  rat uterus activity  White columns  =  vasopressor activity  Striped columns  =  frog bladder activity  Incomplete columns, denoted by the sign >  , indicate  that no response could be detected and the value is below that indicated by the column.  ,4  1200  peek  I  peak  2a  peak  2b  125.  Bio-assay  Standard  Peak 1 mU/ml  Rat uterus  Syntocinon  62.1  Vasopressor  Pitressin  Frog bladder  Syntocinon  Peak 2a mU/ml  Peak 2b mU/ml  15.5  < 2.5  < 12.0  146.9  50.0  <200.0  1,167.0  <200.0  Rat uterus activity was localized mainly in the f i r s t peak, which corresponds to oxytocin. A smaller concentration of oxytocic activity was present in the f i r s t half of the vasopressor peak (2a). The rat uterus activity within vasopressor peak 2a was 5 times too great to be attributed to the intrinsic rat uterus activity of arginine vasopressin, and could be explained only by either a trailing of oxytocin into this peak or the presence of an additional agent within the peak.  In the second half of the  vasopressor peak (peak 2b) the rat uterus activity did not exceed the intrinsic activity of arginine vasopressin. The frog bladder activity in both the oxytocic peak (peak 1) and in the second half of the vasopressor peak (peak 2b) was found to be below the sensitivity of the assay.  In contrast, there was a strong frog  bladder activity in the f i r s t vasopressor peak (2a);  this  activity was found to be almost 100 times too great to be explained by oxytocin, i f the rat uterus activity within this peak had been due to a trailing of oxytocin behind i t s major zone.  The frog bladder potency of peak 2a was also found to be  160 times too great to be elicited by arginine vasopressin. The Lowry peptide content of the eluates was below the level of detection, and in consequence, no direct measurement of the  126.  extent of purification could be obtained.. However, on the basis of the ratios of the biological activities, i t was assumed that the f i r s t vasopressor peak (2a) contained a mixture of arginine vasopressin and arginine vasotocin, and further appropriate purification steps were carried out according to this assumption. In a second experiment, the remaining 8 ml of the fetal neurohypophysial  extract was mixed with 2 ml of fetal neuro-  hypophysial extract obtained from fetuses which were collected during the months of January and February.  This additional  material brought the potency for rat uterus activity to 7,900 mU, the vasopressor activity to 18,000 mU and the sample volume to 10 ml.  In a l l details the  experiment was carried out in a  manner identical to that of the previous one, and the results are shown in Fig. 15.  The elution pattern was closely similar to that  already discussed in the f i r s t experiment.  The fractions with  identical biological activities were pooled, and stored at 4°C until further purification procedures. 3.  The ion exchange chromatography of oxytocin and vasopressin from fetal seals: a.  Purification of the oxytocic principle. 0  The oxytocic fractions, obtained from gel f i l t r a t i o n , were subjected to a second purification step before an attempt was made to determine their amino acid content.  35 ml of the pooled rat  uterus fractions from the G-l5 column, a volume containing a total of 2,700 mU of rat uterus activity, was washed three times with ether to remove the majority of acetic acid from the medium (see  127.  Figure 15. The gel filtration of crude neurohypophysial extracts of seal fetuses at 0.56-0.68 of term, on a 200 cm Sephadex G-15 column.  Loaded 7,900 mU of rat uterus and 18,000 mU  of vasopressor activity. was 0.2 M acetic acid.  The loading and eluting medium Each tube contained 2.8 ml fraction.  Biological activities are expressed in mU/ml of eluate. Open circles and broken line  = rat uterus activity  Full circles and solid line  = vasopressor activity  100  150  2 00 tube number  o  o  Rat u t e r u s  ©  •  Vasopressor  250  128.  Methods).  After this, the pH was adjusted to 5.0 with concentrated  ammonium hydroxide, and the sample was diluted to a molarity of 0.002 M with d i s t i l l e d water (a final conductivity of 0.12 mi H i mho).  The total volume of 350 ml was loaded at a rate of 20 ml/hr  onto a 1 x 15 cm CM Sephadex column, by means of an LKB peristaltic pump. The column had been previously equilibrated with 0.002 M ammonium acetate, at pH 5.0. At the completion of the loading procedure, the column was washed with 60 ml of the 0.002 M starting buffer, and a gradient to 0.1 M ammonium acetate was started through a 60 ml mixing chamber.  The elution was carried out at a flow-rate  of 12 ml/hr, which was maintained with a peristaltic pump. The eluates were collected in 2.8 ml fractions and assayed for rat uterus activity.  Conductivity measurements were avoided for fear  of contaminating the fractions. The results are shown in Fig. 16. All the biological activity eluted in a single peak, which spread over only seven tubes, and the recovery approximated  100%. The  pooled fractions containing rat uterus activity were lyophilised to dryness.  They were then taken up in distilled water and  re-lyophilised.  This process was repeated a total of three times  to remove the ammonium acetate buffer.  The dry powder obtained at  the end of these procedures was stored in a deep freezer (-18°C) for later amino acid analysis. b.  Purification of vasopressor peak 2b on Phosphocellulose. A preliminary hydrolysis of the second vasopressor peak  (2b) showed that in contrast to similar fractions from adult glands, these fractions contained not only the eight amino acids of arginine vasopressin, but also'a large amount of glycine,  129.  Figure 16. Purification of the oxytocic fractions on CM cellulose column;  partially purified extracts of seal fetuses at  0.56-0.68 of term.  Loaded 2,700 mU of rat uterus activity.  Elution was carried out with a gradient of ammonium acetate from 0.002 M to 0.1 M at a pH of 5.0. Each tube contained 2.8 ml fractions. expressed in mU/ml of eluate.  Biological activity is  tube  number  130.  threonine and serine, presumably present as contaminants.  It  proved necessary to purify these fractions further before a final amino acid analysis.  The pooled fractions of the second vaso-  pressor peak (2b) were washed with ether to remove acetic acid from the sample; they were then titrated to a pH of 5.0 with concentrated ammonium hydroxide, and diluted to a molarity of 0.02 M (conductivity 0.95 mi H i mho) with distilled water. The final volume of 350 ml was loaded onto a 1 x 16 cm Phosphocellulose column, built and equilibrated in 0.02 M ammonium pH of 5.0.  acetate at a  At the completion of the loading, the column was  washed with 60 ml of the starting buffer (0.02 M, pH 5.0 ammonium acetate) and then elution was started with a concentration gradient to a 0.2 M pH 5.0 ammonium acetate solution.  The eluates  were assayed for vasopressor activity, and the results are shown in Fig. 17. The activity eluted in a single peak, in 16 tubes, with a total volume of 32 ml. The recovery was approximately 80%. The biologically active fractions were pooled, lyophilised three times to remove buffer, and the f i n a l , dry powder was stored in the deep freeze (-18°C) for later amino acid analysis. 4. The separation of arginine vasotocin from arginine vasopressin. The most important, but the most d i f f i c u l t problem to be faced at this stage was the separation of arginine vasotocin from arginine vasopressin.  This was especially d i f f i c u l t because of the  vital necessity of avoiding large losses of the trace quantity of arginine vasotocin which was present.  Several methods which were  reported to be capable of the separation of these peptides, or of similar compounds, were tried and abandoned either because of their  131.  Figure 17. The purification of vasopressor peak 2b from gel filtration on Phosphocellulose column (fetal seals at 0.56-0.68 of term).  Vasopressor activity (rat pressor  assay) expressed in mU/ml of eluate.  Elution was  carried out with a gradient of ammonium acetate from 0.02 M to 0.2 M at pH 5.0. of eluate.  Each tube contained 2.0 ml  132.  poor ability in resolving arginine vasopressin from arginine vasotocin, or because of.the poor recoveries of active agents. The two peptides eluted as a single peak from beaded ion exchange resins such as CM Sephadex.  In contrast, a good  resolution was obtained from pyridine washed G-25 Sephadex columns which were free of ion exchange groups, but which possessed an aromatic absorption capacity (Eaker and Porath, 1967). However, this sytem was rejected because only 20% of the applied potency could be recovered  (see Appendix).  The most promising  resins for this work proved to be Dowex-X2 (Dow Chemicals) and Amberlite a.  IRC-50, and their performance will be described below,  Preliminary experiments with Dowex 50-X2. 8,000 mU of purified arginine vasopressin, together with  800 ml) of synthetic arginine vasotocin, and 5,000 mU of synthetic oxytocin were applied to a 10 x 0.5 cm column of Dowex 50-X2 resin.  The resin was previously equilibrated in 0.2 M pH  4.0  ammonium acetate solution, and the concentration and the pH of the sample was adjusted to be identical to that of this starting buffer.  A total volume of 80 ml of the mixed principles was  loaded on to the column, which was then washed with the 0.2 M pH 4.0 starting.buffer. Elution was carried out by means of a continuous gradient from the starting buffer of 0.2 M pH  4.0  ammonium acetate to.a 2.0 M pH 7.0 solution of the same salt; a 30 ml mixing chamber was used.  The eluates were assayed for their  biological activities, and the results are shown in Fig. 18. three neurohypophysial peptides eluted in three well  separated  The  133.  Figure 18. Separation of arginine vasopressin from arginine vasotocin on Dowex 50-X2 and Amberlite IRC-50 resins. Biological activities expressed as mU/ml of eluate.  Each tube  represents 2.8 ml fractions. Open circles and broken line  = rat uterus activity  Full circles and solid line  = vasopressor  On l e f t :  activity.  Elution of arginine vasopressin and arginine vasotocin from Dowex 50-X2. Gradient used for elution was ammonium acetate from 0.2 M pH 4.0 to 2.0 M pH 7.0.  On right: Separation of the two basic peptides on IRC-50 resin.  The gradient for elution was ammonium  acetate solution from 0.2 M pH 5.0 to 0.65 M pH 6.0 for the arginine vasotocin.  Arginine  vasopressin was recovered by straight replacement with 1.0 M pH 7.7 ammonium acetate solution.  20 MpH7  o  a  Rot uterus  »  a  Vasopressor  134.  peaks;  however, the high buffer concentrations necessary for  elution interfered with the biological assays, and made these estimations almost impossible to carry out.  In consequence, the  estimation of activity in the eluates was unreliable. Although the recoveries appear to be poor, between 20% and 50%, this might well be a false impression created by the unreliability of the measurements.  In any case, the difficulty encountered in  detecting the activities in the eluates made this method impractical for the purification of trace amounts of peptides, b.  Preliminary experiments with IRC-50. A separating procedure used by Chauvet, Lenci and Acher  (1960), was reported to result in aresolution of arginine vasopressin, by the use of an IRC-50 column.  However, this  procedure proved to be unsuccessful.in the present investigation, on account of poor resolution of the peptides, and of poor final recoveries.  Nevertheless, in several consecutive experiments, in  which the size of the column, the pH, and the conductivity of the eluting buffers were altered, a combination of these new factors was shown to result in the separation of two peaks from the applied biological activities. In the f i r s t of these final experiments with  standards,  10,000 mU (vasopressor) of purified seal arginine vasopressin and 1,600 mU (vasopressor) of synthetic arginine vasotocin (equal to 800 mU of rat uterus activity) were applied to a 10 x 0.5 cm IRC50 column.  The column was built in distilled water, after the  135.  method of Chauvet, et a]_, (1960). acetic acid.  The sample was applied in 0.2 M  A total volume of 80 ml was loaded onto the resin,  which was then equilibrated with 0.2 M, pH 5.0 ammonium acetate solution;  equilibration was continued until the eluant from the  column was identical to the equilibrating solution in both conductivity and pH.  For elution, a gradient to 0.65 M, pH 6.0  ammonium acetate was applied through a 60 ml mixing chamber.  The  vasopressor and rat uterus activities were assayed in the eluates immediately after elution, and the results are shown in Fig. 18. When the vasopressor activity was found to begin to rise in a second peak, and the rat uterus activity approached zero, the gradient was disconnected.  Arginine vasopressin was recovered from  the column by direct replacement of the final medium with 1.0 M, pH 7.7 ammonium acetate. Two peaks of biological activities were detected at the completion of this experiment.  The V/RU ratio in the f i r s t ,  smaller peak was approximately 2.0, and in the second peak i t reached a value of 20.0;  this indicated that the f i r s t peak  contained arginine vasotocin and the second peak consisted of arginine vasopressin. Although the two activities separated into two sharp peaks, there was a considerable overlap at their bases (Fig.  18).  After discarding the portion of the eluates which  appeared to consist of mixed peptides, the recovery for arginine vasotocin approximated only 70%. In a final experiment with standards, an attempt was made to improve the resolution of the two peaks, and to increase the  136.  recovery of arginine vasotocin by halving the buffer concentrations used for elution.  The 0.2 M, pH 5.0 starting  buffer used in the previous experiment, was replaced by a 0.1 M pH 5.0 ammonium acetate solution, and the gradient for the elution of arginine vasotocin reached a concentration of only 0.32 M pH 6.0 of ammonium acetate.  Vasopressin was eluted by a  direct replacement with 0.75 M pH 7.7 ammonium acetate solution. In a l l other details the experiment was carried out in an identical manner to the one described above.  The eluated fractions  were assayed for their vasopressor and rat uterus activities, and the results are shown in Fig. 19A. The vasotocin peak, which eluted f i r s t , came off in a shallower curve than before, and gave a more dilute solution of the biologically active agent.  However,  the overlap of the peaks was greatly reduced, and the recovery greatly improved.  In the pooled fractions of the vasotocin peak,  1,600 mU of vasopressor activity and 800 mU of rat uterus activity was measured, and this represented a 100% recovery.  Although the  lowering of the buffer concentrations resulted in a less distinct vasotocin peak, the increased resolution and the consequent good recovery made this method the best available for the final purification of the fetal material. c.  The separation of the two basic peptides from fetal extracts on an IRC-50 column. In two experiments, the pooled eluates of vasopressor peak  2a from the G-l5 Sephadex columns (see Fig. 13 and 15) were  137.  Figure 19. Separation of arginine vasotocin from arginine vasopressin on Amberlite IRC-50 resin. as mU/ml of eluate.  Biological activities expressed  2.8 ml fractions per tube.  Gradient  for the elution of arginine vasotocin from 0.1 M, pH 5.0 ammonium acetate, to 0.32 M pH 6.0 ammonium acetate. Arginine vasopressin recovered by straight replacement with 0.75 M pH 7.7 ammonium acetate. Full circles and solid line  = vasopressor activity  Open circles and broken line  = rat uterus activity.  A.  Elution profile of standard run.  Loaded 10,000 mU  arginine vasopressin and 1,600 mU (vasopressor) of synthetic arginine vasotocin. B.  Elution profile of vasopressor fraction from seal fetuses, at 0.56-0.68 of term.  Loaded 80 ml of  eluate from Sephadex G-15 run (peak 2a. Fig. 15).  » o—  I 01MpH5  OlMpHi  •  « -o  Vosoprrttor Rot  uterus  138.  chromatographed on IRC-50 columns.  Only one experiment, that  which used the eluates from Fig. 15, will be described here, since the two experiments were carried out in an identical manner, and yielded closely similar results. 80 ml of the eluate from peak 2a (Fig. 15) was loaded onto an IRC-50 column (10 x 0.5 cm) in the 0.2 M acetic acid in which i t was eluted from the previous G-l5 column.  The column, with  the unknown loaded, was equilibrated in the 0.1 M pH 5.0 ammonium acetate, until the eluant from i t was identical in conductivity and pH to that of the starting buffer.  At the  completion of the equilibration procedure, a gradient was started, through a 60 ml mixing chamber, to a final ammonium acetate solution of 0.32 M, pH 6.0.  The fractions were assayed for rat  uterus and vasopressor activities immediately upon elution, and the results are shown in Fig. 19b.  When the vasopressor activity  began to rise in a second peak, and the rat uterus activity approached zero, the gradient was disconnected, and the second peak was eluted by straight replacement with an 0.75 M pH 7.7 ammonium acetate solution.  The results of this experiment are  shown alongside the preliminary standard run in Fig. 19.  The  elution profile, as well as the V/RU ratios within the two peaks, are remarkably similar in the standard and the unknown runs.  The  biological activity recovered from the f i r s t peak equalled 1,350 of vasopressor activity and 700 mU of rat uterus activity; values give a V/RU ratio of approximately 2.0.  these  The total  vasopressor activity in the second peak was 8,500 mU, and this  mU  139.  same peak gave 465 mU of activity by rat uterus assay; gives a V/RU  ratio of 19.0.  this  These values are in good agreement  with the assumption that the f i r s t peak eluted was arginine vasotocin, and the second peak consisted of arginine vasopressin. These deductions were.further  supported by additional pharacological  data and by amino acid analysis;  these results will be presented  in later sections. Discussion: The purification steps used to isolate the biologically active agents from the fetal neurohypophysis are summarized in Fig. 20.  As the final result of these procedures, three biologically  active fractions were separated.  They were tentatively identified  as oxytocin, arginine vasopressin and arginine vasotocin.  The  purification of oxytocin and arginine vasopressin was successfully accomplished with methods previously applied in the field of neurohypophysial physiology.  The isolation of arginine vasotocin  from arginine vasopressin, the most important feature of these studies, appeared a more formidable task.  On the basis of the frog  bladder activities detected,during the earlier paper chromatographic studies, and from the assays of the eluates from gel filtration columns, a rough estimate of the quantity of arginine vasotocin present in the fetal tissues could be calculated.  If the value of  40,000 I.U. of frog bladder activity per mg of synthetic arginine vasotocin is accepted - a value consistently obtained in our laboratory - the total available fetal arginine vasotocin worked  140.  Figure 20. Summary of the purification procedures used for the isolation of the fetal neurohypophysial peptides.  CRUDE EXTRACT G-15 Sephadex  V  OXYTOCIC peak 1  VASOPRESSOR peak 2 a  VASOPRESSOR peak 2 b  IRC-50  Phosphocellulose  C M - Sephadex  ARG.VASOTOCIN  a.a. analysis  a  . . analysis a  ARG.VASOPRESSIN  a.a.analysis  141.  out to be approximately 5 jug, or 5nM of peptide (MW = cc 1 ,000).  If  the frog bladder activity of arginine vasotocin is taken as 100,000 I.U. per mg peptide, a value reported by Sawyer (1965) the total arginine vasotocin in the fetal tissues would not have exceeded 2 ^ig or 2nM of the peptide.  Fortunately for these studies, later  amino acid analyses proved that these estimations were too low, and the total amount of arginine vasotocin came closer to lOnM. would suggest that each fetal gland between 0.56  This  and 0.68 of term  contained approximately 0.2nM of arginine vasotocin.  However, even  this final amount of lOnM was an exceptionally small amount of peptide for a successful purification.  In comparison, Pickering (1967) used  1,000nM of each neurohypophysial principle in his purification of the active agents from cobra pituitaries.  To purify oxytocin,  Acher, Chauvet and Lenci (1960) started out with enough biological activity to equal 2,000nM of the peptide. Further difficulty in the experiments reported here arose from the ratio of arginine vasotocin to arginine vasopressin in the crude extracts.  For example, in vasopressor peak 2a (Fig. 15),  after more than one half of the vasopressor activity had been separated away, the ratio on.the basis of molar content was 6 to 1 in favour of vasopressin.  In terms of biological activities  this ratio was even greater, since arginine vasopressin is more than twice as potent a pressor agent as arginine vasotocin.  With  such a ratio of the two active agents, even the smallest contamination of the vasotocin peak with the predominating vasopressin, would have made any amino acid analysis unreliable.  142.  The major breakthrough in the purification  techniques,  and the one which is believed to have made the final purification possible, was the i n i t i a l purification of the crude extracts on G-l5 gel f i l t r a t i o n columns.  This preliminary step  yielded a highly purified mixture of the two basic peptides, and offered the added advantage of separating away a large portion of arginine vasopressin in the process, with only negligible losses of the active agents.  Such a highly efficient one-step  purification allowed.the final separation of the two basic peptides with only one additional step. Amberlite  IRC-50 resin was used for the final  separation  and purification of the two basic peptides because i t offered several advantages over the other resins investigated.  The resin,  used in its hydrogen form, allowed the direct loading of the unknown in the 0.2 M acetic acid in which i t was eluted from the preceding G-l5 column.  Other ion exchange resins considered for  this work required the acetic acid in the solution to be converted into ammonium acetate, at a pH and conductivity identical to that used for the equilibration of the resin.  Such a procedure  involved one or two additional steps, which could be eliminated by the use of IRC-50 resin. A further advantage of both IRC-50 and Dowex gels i s that the elution i s primarily dependent on a pH gradient, rather than on a concentration gradient.  This characteristic of the resin  allowed the elution of arginine vasotocin with a concentration  143.  increment of only slightly more than threefold.  However, the  IRC-50 resin offered the added advantage of being effective in a lower concentration range than the Dowex gel, and this was a great advantage in the performance.of the bioassays.  The adverse  effects of high salt concentrations on biological preparations is a particularly important consideration, since the eluates, which are already dilute for biological activities, cannot be further diluted to eliminate salt effects, without making the peptides themselves undetectable. 5.  Identification of the fetal peptides by their pharmacological properties and amino acid content: a.  PharmacologicaT  characterization of the active agents.  Each of the three biologically active fractions separated from the fetal neurohypophysis was assayed against synthetic standards which were thought to duplicate their structure.  On  the day of the f i r s t assay of the series, the synthetic standards were made up in ammonium acetate buffer of a pH and conductivity identical to that of the unknown solution.  When storage was  necessary, both standards and unknowns were acidified to a pH of 3.0, and stored at 4°C.  The same standard solutions were used  throughout the series of assays, on the assumption that any possible.decay of the unknown would be matched by an equal deterioration of any standard solution with identical potency, pH and.salt concentration. For the pharmacological identification of the oxytocin-1ike fractions eluted from the CM Sephadex column, rat uterus assay,  144.  with and without magnesium ions (Mg ), was carried out against synthetic oxytocin. to be 0.89 +_ 0.2.  The ratio of RU + Mg /Rll - Mg ++  This ratio includes unity;  ++  was found  i t is character-  i s t i c of oxytocin, the only naturally occurring neurohypophysial analogue with a Mg  ++  ratio.not above unity.  In the pharmacological identification of the two vasopressor peaks eluted from the IRC-50 columns, the assumption was made that the f i r s t peak contained arginine vasotocin, while the.second peak consisted of arginine vasopressin;  chromato-  graphic evidence for this was shown in Fig. 19. On this assumption, the two peaks were assayed directly against either synthetic arginine vasotocin or synthetic arginine vasopressin, according to which peptide they appeared to parallel.  The assay  preparations were chosen to include a wide range of sensitivity to the peptides in question, and where the potency of the eluates allowed, the same assays were used for the identification of both peptides.  The results are shown in Table VIII.  The ratios  of biological activities to rat uterus activity closely approximated unity in every case; peak was pharmacologically vasotocin,  this indicated that the f i r s t  indistinguishable from arginine  and that the second peak appeared to parallel  arginine vasopressin.  The close approach to unity of these  ratios gives even greater credibility to their pharmacological identification i f i t is remembered that the frog bladder preparation is several thousand times more sensitive to arginine  TABLE VIII Biological activities of the two vasopressor peaks against arginine vasotocin and arginine vasopressin standards  Biological assay  jug peptide/ml  Rat uterus - Mg  Vasopressor peak II from IRC-50 against AVP  Vasopressor peak I from IRC-50 against AVT  ++  ++  ratio to RU - Mg  mil/ml activity  507.4 + 59.5  0.035 + 0.005  Rat uterus + Mg  0.030 •+ 0.004  1.17 + 0.22  488.2 + 67.5  Frog bladder  0.031 + 0.02  1.11 + 0.61  none detectable *  Antidiuretic  0.035 + 0.02  0.99 + 0.5  Milk ejection  0.034 + 0.01  1.08 + 0.4  Vasopressor  485.3 + 93.6  AVT = synthetic arginine vasotocin (0.1 mg/ml, Sandoz) AVP = synthetic arginine vasopressin (4.65 I.U./ml, Sandoz)  **  less than 250 mU/ml less than  10 mU/ml  1.03 + 0.17 0.98 + 0.22  none detectable ** 525.8 + 79.2  *  Ratio to RU  1.06 + 0.30  146.  vasotocin than to arginine vasopressin;  in this case, any  difference in structure between the standard and unknown was likely to result in a great variation in the ratio, b.  Amino acid content of the three fetal peptides. The solutions which remained after the pharmacological  characterization of the fetal principles were lyophilised to dryness.  The dry residue was washed.in distilied water and re-  lyophilised three times until a l l the buffer was removed.  The  final residues were hydrolysed in 6_ N HCI for 18-24 hrs on an oil bath at 108°C. (see Methods).  At the end of the hydrolysis,  the acid was removed by flash evaporation, and this process was repeated three times with fresh washings of distilled water. The residue was taken up in sodium citrate buffer, pH 2.2, and analysed on a Bio-Cal 200 amino acid analyser.  The results,  together with' the theoretical molar ratios of the peptides, are shown in Table IX. The analysis of fetal seal oxytocin yielded the eight amino acids of oxytocin.  The ratios between the constituent  amino acids and the glutamine content showed a remarkable agreement with the theoretical values, except in the case of cystein (% Cys).  This amino acid is prone to oxidation to cysteic  acid during the hydrolysis of small amounts of peptides (see Pickering, 1967).  Indeed, the combined concentrations of cystein  and cysteic acid found here gave a close approximation to the theoretical value.  In addition to the eight amino acids of oxytocin,  only lysine was detected in any quantity (a ratio of 0.4); however, i t is known to be a common contaminant of neurohypophysial  TABLE IX Amino acid content of the neurohypophysial Oxytocic fraction from CM Sephadex Amino Acid  Vasopressor fraction from Phosphocellulose  theoretical  Vasotocin fraction from IRC-50  Ratio to Glu  Ratio to Glu seal  peptides of seal fetuses  seal  theoretical  Ratio to Glu seal  theoretical  Asp  1.08  1.00  0.82  1.00  1.03  1.00  Glu  1.00  1.00  1.00  1.00  1.00  1.00  Pro  0.97  1.00  1.03  1.00  1.00  1.00  Gly  0.96  1.00  0.98  1.00  1.20  1.00  h Cys  0.86 )  2.00  0.40 )  2.00  0.93 )  2.00  Cys Acid  0.34  lieu  0.75  1.00  Leu  0.96  Tyr  1.00  Phe  -  Arg  j  j  -  1.02  -  0.69  1.00  -  1.00  0.50  1.00  0.43  1.06  1.00  0.15  0.86  1.00  0.80  1.20  1.89  1.29  j  1.95 1.00  0.13 1.00 1.00  Ratio of  Ser. 0.08, Ala. -0.11,  Ser. 0.3, Ala.0.04,  Ser. 0.1, Ala. 0.2, Val.  additional  His. 0.15, Lys. 0.41  His. 0.3  0.07, His. 0.06, Lys.  residues to Glu  0.16  148.  preparations (Perks, personal comm.). All other amino acids were found to be in trace amounts only. The hydrolysate of the vasopressin peak was found to contain the eight constituent amino acids of arginine vasopressin, in ratios which were in good agreement with the theoretical values.  In addition to the oxidation of cystein to cysteic acid,  part of the tyrosine appeared to have been lost during hydrolysis; the loss of this amino acid due to oxidation is a frequent occurrence during the hydrolysis of neurohypophysial peptides (see Pickering, 1967).  In view of the correct biological  activities of the unhydrolysed peptide, i t can be assumed that i t was present in the theoretical ratio.  Only three amino acid  residues were detected in addition to those which form part of arginine vasopressin (Ser. 0.3; Ala. 0.04; His. 0.3), and a l l of these were in a relatively low ratio. The eight amino acids of arginine vasotocin were detected in the hydrolysate of the third peptide, and the ratios were close to those which would be expected for this principle.  The loss of  tyrosine during hydrolysis was somewhat greater than in the cases of the other two peptides, but i t s t i l l remained in a quantity well above the level of the contaminants.  In addition to the  amino acids expected of arginine vasotocin, trace amounts of leucine, and phenylalanine were detected.  Since these two amino  acids are typical of oxytocin and arginine vasopressin respectively, there is a possibility that traces of these two peptides were eluted with the vasotocin.  However, the detection of isoleucine and  149.  arginine in a ratio close to one, bore out the hypothesis that this fraction was composed predominantly of arginine vasotocin, since these two amino acids are characteristic components of this peptide.  This is further supported by the pharmacological  activities of this fraction (see p. 143). Discussion: The amino acid analyses of the three fetal neurohypophysial peptides have shown that they contain the amino acids of oxytocin, arginine vasopressin and arginine vasotocin respectively. In order to assign the structure of these molecules with certainty, a sequence analysis should have followed these investigations. However, the quantity of the fetal peptides, especially of arginine vasotocin, was hardly sufficient for the amino acid analysis performed, let alone for the costly process of sequence determination. A simple alternative method to give an indication of the spatial arrangement of the amino acids, makes use of the specific biological activities of neurohypophysial analogues.  When a  variety of bioassays are carried out on a neurohypophysial peptide, using oxytocin, arginine vasopressin, or the mixture of the two which constitutes the International Standard as the classical standards, the ratios of biological activities differ greatly from one analogue to another.  The.pharmacological spectrum  obtained may be characteristic of the analogue concerned, or nearly so. Table X  shows the spectrum of ratios obtained  150.  against the International Standard, for the three peptides discussed here:TABLE X (Values taken from Sawyer, 1965) Neurohypophysial  peptide  Ratios of biological activities RU + Mg RU - Mg  ++ ++  ME  FB  ADH  VP  RU  RU  RU  RU  Oxytoci n  0.9  0.84  1.0  0.0024  Arginine vasopressin  2.3  5.7  2.5  44.0  Arginine vasotocin  1.9  2.2  910  RU = rat uterus;  ME = milk ejection;  ADH = antidiuretic;  0.006  1.55  44.0 1.96  FB = frog bladder;  VP = vasopressor activities.  Although the difference in the ratios obtained with conventional standards is striking, this method for the identification of the peptides is sometimes unsatisfactory.  Different  specimens of a biological preparation such as the rat uterus, might.vary in their sensitivity to an individual peptide, so that the same type of assay, carried out between dissimilar peptides at different times, could give variable results. Consequently, i t is possible to obtain varied pharmacological ratios from the same pair of dissimilar peptides:  therefore the  ratios obtained would not be a reliable guide to the precise identity of a new analogue.  This difficulty can be overcome by  using a synthetic standard of the same composition as the sample. In this case, the indication of identity lies in a constant value,  151.  which is independent of the method of assay used, so that the pharmacological ratios are consistently unity.  For example, i t  is easy to see that, of the three peptides discussed here, only arginine vasotocin is likely to give a FB/RU ratio of unity against synthetic arginine vasotocin, since the other two  peptides  are nearly a thousand fold less potent on the frog bladder preparation.  The same argument applies to the antidiuretic and  vasopressor activities of arginine vasopressin. The series of pharmacological tests, performed during the course of these studies, allow several more specific conclusions to be made concerning the positions of amino acids within the peptide molecules.  Studies of the pharmacological properties of  a great number of analogues have established  the specific  biological activities associated with certain amino acids at definite locations (Sawyer, 1961; Rudinger and Jost, 1964; Berde and Boissonnas, 1968; Pickering, 1970).  The following consider-  ations, regarding the structure of the fetal peptides, are based on data presented by these authors. a.  The structure of fetal arginine vasotocin. In considering the low recoveries of cystein and tyrosine  i t should be noted that although an intact hemisystynyl ring is not essential for uterotonic and pressor activities, any substitution in i t will diminish the milk-ejection and frog bladder activities of the peptide.  The 100 I.U. of rat uterus  activity per mg fetal peptide - as calculated on the basis of  152.  amino acid content - represented a potency of only 9% less than that of synthetic arginine vasotocin (assayed as 110 I.U.).  The  1:1 ratio of biological activities to synthetic arginine vasotocin, together with the high potency of the fetal arginine vasotocin, comprise a strong evidence against a lowered quantity of cystine within the peptide: i t supports the hypothesis that there was a loss of this amino acid due to oxidation during hydrolysis. Similarly, substitution of tyrosine in position 2 diminishes the uterotonic as well as the pressor activities of the peptides. Here again, the ratios of the biological activities are so near to unity, that i t would be unlikely that this amino acid was present in less than the theoretical proportion. The positions of the remaining amino acids within the molecule can also be indicated by pharmacological tests.  The  presence of ileu in the third position is required for full milk ejection and rat uterus activities.  The arginine present would  need to be placed in.position eight for the strong frog-bladder and antidiuretic activities of the peptide.  Asparagine in position  5 is an essential requirement for biological activitity on any of the assay preparations and replacement of the terminal glycinamide leaves the molecule with a greatly diminished potency. This leaves only the proline in position 7 and glutamine in position 4 unaccounted for. However, since the analogues which contain proline in position 4'are highly inactive on any of the standard biological preparations, a switch in the placement of these two amino acids is unlikely (Sawyer, et_ al_, 1969).  153.  b.  The structure of the fetal arginine vasopressin. Since seven of the eight amino acids in arginine vaso-  pressin are identical to those in arginine vasotocin, much of the argument presented above stands for the positions of the amino acids within the fetal arginine vasopressin.  However, in  addition, i t should be mentioned that the substitution of phenylalanine in position 3 appreciably lowers the pressor as well as the antidiuretic potency of the molecule: in practice, i t was found that both these activities were unaltered, so that i t may be concluded that phenylalanine is most likely to be in position 3 of the fetal peptide. c.  The structure of the fetal oxytocin. The only pharmacological  test carried out for the  identification of oxytocin was the rat uterus assay, with and without Mg  ++  ions, against synthetic oxytocin.  naturally occurring neurohypophysial RU with Mg /RU without Mg ++  oxytocin i t s e l f .  ++  However, the only  peptide with a ratio of  which does not exceed one is  Therefore, the ratio of 0.89 + 0.2, obtained  from the fetal oxytocic fraction, together with the amino acid analysis, indicates that the fetal oxytocin-1ike peptide is oxytocin i t s e l f . In summary, the amino acid analysis of the three fetal neurohypophysial  peptides of the fur seal have shown that they  contain the amino acids of oxytocin, arginine vasopressin and arginine vasotocin, respectively. The pharmacological  behaviour  of the fetal peptides was indistinguishable from that of the  154.  appropriate synthetic standards. It seems safe to conclude, that the fetal neurohypophysial  peptides consist of oxytocin, arginine  vasopressin and arginine vasotocin. D.  The action of the neurohypophysial  peptides on the embryonic  membranes 1.  Introduction:  The possible role of the fetal neurohypophysial  agents  especially that of the predominant antidiuretic principle, remains one of the most interesting and s t i l l unresolved problems of posterior pituitary function during embryonic development.  The available  data concerning the ability of the fetus to regulate the tonicity and the volume of i t s body fluids indicate that the kidneys are of l i t t l e importance in this regard;  the immature animal and the  infant are unable to produce hypertonic urine (Seeds, 1965). Heller (1949), working with newborn humans and rats, failed to e l i c i t either a diuretic response to water!oading or an antidiuretic response to dehydration.  Vasopressin was found to have no effect  in causing a concentration of the urine in immature animals (Hansen and Smith, 1952); 1962).  Barnett and Vesterdal, 1953; Alexander and Nixon,  From these results i t appeared that the antidiuretic  principle of the fetal organism was ineffective in conventional adult functions.  However, when i t is considered that osmotic  equilibrium in the fetal circulation is maintained by the maternal body, by means of the placenta, i t would appear that the regulation of the tonicity of i t s urine is of l i t t l e importance to the fetus. On the other hand, i t seems certain that the fetus contributes to the formation and maintenance of the amniotic and allantoic fluids.  155.  These embryonic fluids, after the f i r s t trimester of gestation, are hypotonic in relation to the maternal as well as to the fetal extracellular compartments and their hypotonicity increases with the advance of gestation (Seeds, 1965; Mellor and Slater, 1971). Besides a lowered total colloid pressure, the embryonic fluids have a reduced sodium concentration, while their potassium concentration rises.  The complex regulatory mechanisms and  physical factors that maintain such a gradient between the extracellular spaces and the embryonic fluids are poorly understood, and virtually.nothing is known of the mechanisms responsible for the formation of the embryonic fluids (Seeds, 1965).  The few  preliminary experiments reported here were performed on the hypothesis that the fetal antidiuretic principle might make an important contribution to the regulation of the volume and perhaps the tonicity of the embryonic fluids.  This hypothesis seemed even  more reasonable after the discovery that the fetal neurohypophysis contained arginine vasotocin, a hormone which is known to alter the permeability of membranes to water and electrolytes (Sawyer, 1967). 2.  Method:  The amniotic and allantoic membranes of guinea pig fetuses at 20 days,  40 days and 60 days of gestation were used.  The  pregnant female was anaesthetized with ether, and the fetuses with their embryonic membranes intact, were dissected from the uterus. The membranes were cut away from the placenta;  they were placed  over a flared end of an open glass tube in such a way that the side which was in contact with the embryonic fluids was facing the  156.  inside of the glass tube.  In the following account, the side of  the membrane which was in contact with the embryonic fluids will be referred to as the "inside" of the membrane, and the side in the direction of the uterus will be termed the "outside" of the membrane. The apparatus and the experimental procedure were the same as for the frog bladder assay (see Methods, p. 24).  However, the  inside and outside bathing solutions were exchanged for solutions which mimicked the ionic composition of the embryonic fluid and of the maternal plasma, respectively.  The ionic composition of  the plasma and of the embryonic fluids were determined by flame photometry, and the compositions of the final solutions are given below.  I am indebted to Dr. J. Phillips and his graduate  students for carrying out these measurements.  157.  Composition of the bathing solutions for the embryonic membranes  Inside solution (fetal side) Days of gestation  Outside solution Ions  (Maternal side)  20 (mEq/1it.)  Na ; +  K c  +  + + a  Mg  ;  ++  150.0  125.0  60 (mEq/1it.)  135.0  5.5  6.2  25.0  4.4  4.5  2.0  2.6  2.3  2.3  In addition, 1 g of glucose and 5 ml of phosphate buffer pH 7.4 (see composition in Methods, p. 23) was added to each l i t e r of solution. The inner bathing solution for membranes of 40 day old fetuses was made up as a 1:1 mixture of solutions for 20 and 60 day old fetal membranes.  158.  Figure 21. The response of the guinea pig~allantoic membranes to neurohypophysial hormones. Symbols: fe  = 0.05 ml of fetal extract, (mixture of 50 mU of vasopressor, 12.5 mU of rat uterus and 135 ml) of frog bladder activity) injected into the inner fluid compartment.  s  = a change of the inner and outer solutions.  avp  = 100 mU of synthetic arginine vasopressin.  c  = injection of 0.05 ml of saline solution into the inner fluid compartment.  A.  Response of the allantoic membrane from a 20 day old fetus  B.  Response of the allantoic membrane from a 40 day old fetus  C.  Response of the allantoic membrane from a 60 day old fetus  40-  40  20-  20  o -TJ  .. 0 -  -20-  20-  40-  40  60-  t fe  80-  t  avp  60-  Lr 2 hours  80 3  40-  i  I  i—  2 hours  20-  allantoic membrane  -20 •  40-  60 •  "Ll  t  avp  80 • 2  hours  3  159.  Results and discussion: 3.  Results. The membranes were set up, and the inner and outer containers  were f i l l e d with the appropriate bathing solutions.  The level of  the inner solution was about 2 cm above that of the solution in the outer container.  In this way, a hydrostatic pressure was  created against the inside of the membrane. An osmotic gradient in the same direction was already established by the different ionic compositions of the two saline solutions (see p. 157}. The purpose of this arrangement was to imitate as closely as possible the physico-chemical conditions surrounding the membranes in the intact animal.  The membranes were weighed until a steady,  slow loss of fluid was recorded from the inner tube, through the membrane. a.  The allantoic membrane  The response of the allantoic membrane to neurohypophysial preparations at three stages of gestation is presented in Fig. 21. At 20 days of gestation (Fig. 21 A) a steady loss of water was noted from the inside of the membrane toward the outside.  The  loss of water was in the direction of the osmotic gradient and the hydrostatic pressure.  When fetal neurohypophysial extract (0.44  of term, a mixture of 50 mU of vasopressor, 12.5 mU of rat uterus and 135 mU of frog bladder activities) was injected into the inside saline solution, a reversal in the flow was recorded, and an uptake of water seemed to take place against an osmotic gradient and a hydrostatic pressure.  A change of the bathing solutions on  160.  both sides of the membrane re-established the i n i t i a l steady loss of water across the membrane. At 60 days of gestation (21C) a similar but somewhat smaller response occurred on the injection of arginine vasopressin into the inside solution. Again a change in the saline solutions appeared to reverse the effect of the hormone. At 40 days of gestation, (21B), the fetal extract and arginine vasopressin, both elicited a response similar to those which occurred at the other two gestational ages.  However, a  change in the bathing solutions, after the response to the fetal extract, also resulted in a reversal of the flow.  This phenomena'  is d i f f i c u l t to explain, but i t should be noted that the water uptake following the introduction of fresh saline solutions was short in duration.  Since the hormones were injected only after a  45 minute equilibrating period, this transient reversal in the flow of the water most likely did not interfere with their effect. b.  The amniotic membrane  The responses of the amniotic membrane at three stages of gestation are illustrated in Fig. 22. The results obtained with the amniotic membranes paralleled those found with the allantoic membranes.  At 20 days of gestation (22A) a water uptake was  recorded as a response to fetal neurohypophysial extracts, and at 60 days of gestation the membrane responded similarly to synthetic arginine vasopressin (22C).  At 40 days of gestation, as with the  161.  Figure 22. The response of the guinea pig amniotic membrane to neurohypophysial hormones. Symbols: fe  = 0.05 mL of fetal extract (mixture of 50 mU of vasopressor 12.5 mU of rat uterus and 135 mU of frog bladder activity) injected into the inner fluid compartment.  s  = a change of the inner and outer solutions.  avp  = 50 mU of arginine vasopressin.  c  = injection of 0.05 ml of saline solution into the inner fluid compartment.  A.  Response of the amniotic membrane from a 20 day old fetus  B.  Response of the amniotic membrane from a 40 day old fetus  C.  Response of the amniotic membrane from a 60 day old fetus  40  20  20  0 -  20 •  t  •20  t  fe 40  40  60'  60  fe  t  2 hours  avp 2 hours  40 CD E 20-  c o u  0 •  -C O)  „  Wei  <b  -20-  n amniotic membrane  avp 40-  60 1  2 hours  162.  allantoic membrane a water uptake was recorded not only as a response to the neurohypophysial preparations, but also to a change in the saline solutions. 4. Discussion. The few preliminary experiments on the embryonic membranes suggested that the neurohypophysial peptides have the capacity to alter the permeability of these membranes to water.  In all the  experiments, the addition of neurohypophysial hormones resulted in a marked uptake of water across the embryonic membranes.  This  observation would suggest that the neurohypophysial principles may cause the passage of water from maternal fluids into those which surround the developing fetus.  Although i t seems presumptious to  draw conclusions from the limited number of experiments performed here,  i t should be pointed out that the enhanced water uptake, as  recorded here, is in good agreement with the reported conditions across the embryonic membranes.  The amniotic fluid is slightly  hypertonic to the maternal plasma during early pregnancy (Seeds, 1965; Mellor and Slater, 1971);  with the advancement of intrauterine  development, the amniotic fluid becomes progressively hypotonic to both the maternal and to the fetal plasma.  During this process, the  sodium concentration in the amniotic fluid is lowered, and the potassium concentration is elevated.  Mellor and Slater (1971)  suggested that the ionic composition and the osmolarity of the embryonic fluids is largely regulated by the fetus.  They based  163.  their conclusion on the fact that the composition and the osmolarity of the embryonic fluids around each of a pair of twin fetuses varied to the same extent as i t did between fetuses of different mothers.  Further, they postulated that the main factor  maintaining the osmotic gradient across the embryonic membranes is an active transport of CI"across either the membranes or the fetal skin. Although i t appears from these reports that the fetus plays a major role in the formation and the regulation of the embryonic fluids, there is very l i t t l e known of the mechanisms through which such a control might take place.  In particular, there were no  reports concerning the possible effect of neurohypophysial agents on the embryonic membranes, except that vasopressin was found to be without effect on the in vitro chorion laeve (Seeds, 1970). The enhanced water uptake in response to antidiuretic hormone recorded in the present studies, appears to be in good agreement wi the maintenance of hypotonic amniotic and allantoic fluids by the fetal organism, since i t promotes the passage of water into thes fluids.  Furthermore, such an action of the neurohypophysial  hormones seems to be in good agreement with the known actions of arginine vasopressin (see Wakim, 1967) and of arginine vasotocin (Sawyer, 1967), the two antidiuretic hormones present in the mammalian fetus.  However, to establish these possibilities as  true physiological functions of the fetal neurohypophysial agents, further, more detailed work is required.  164.  E. Discussion The results obtained in each phase of these investigations have been discussed immediately following their presentation. The validity of experimental methods, as well as the relationship of the data obtained to those reported in the literature have been considered in detail in the appropriate sections.  It remains only  to assess the contribution of each aspect of these studies to the evaluation of fetal neurohypophysial function, and to determine their possible contribution to neurohypophysial physiology in general. 1.  Neurohypophysial  function in the fetal seal.  The combined data from histochemical, pharmacological and paper chromatographic studies may be summarized in the following way: a.  Biological activities were detected at the earliest  stage studied, at 0.19 of term when the vasopressor activity was approximately 4% and the oxytocic activity was 2.5% of that found in the near term fetuses (see Table VII).  No material was  available for histological studies at this early stage. b.  At 0.31 of term, the vasopressor and rat uterus  activities were found to have increased, and reached the level of 10% and 5%, respectively, of that recorded near to term (see Table VII).  Paper chromatographic  studies-  were elicited by  arginine vasopressin, oxytocin and arginine vasotocin.  However,  at about the same stage of gestation (0.30 of term), the hypothalamic nuclei and the neurohypophysis were found to be completely void of stainable neurosecretory material.  165.  c.  At 0.5 of term, the neurohypophysis was found to  contain approximately 30% of the vasopressor and 17% of the rat uterus activity of near term fetuses (see Table VII).  At the  same gestational stage, stainable neurosecretory material was found in the two hypothalami's nuclei and in the neurohypophysis.  In the  supra-optic nucleus (attributed mainly with the production of vasopressin;  Heller, 1966), 52% of the neurons contained neuro-  secretory material, while in the para-ventricular nucleus, where most of the oxytocin is produced (Heller, 1966), 14% of the cells were found to contain stainable neurosecretion.  At this same  stage, the neurohypophysis showed an overall diffuse staining. d.  For the studies around the time of birth, specimens  at three different, but relatively close stages were used, due to the limitation of material available.  The biological activities  were determined in specimens at 0.93 of term, the hypothalamic secretion was estimated.in a pup close to birth, and the stainable neurosecretory material in the neurohypophysis was estimated from fetuses at 0.75 of term.  Thebiological activities in the near  term fetuses were found to be unusually high at 2,800 mU/mg for vasopressor and 1,200 mU/mg for rat uterus activities (termed "100%" in each case).  At around the same developmental time  (i.e. in the newborn), the hypothalamic nuclei showed a high secretory activity.  92% of the neurons in the supra-optic nucleus  and 85% of the neurons in the para-ventricular nucleus contained neurosecretory granules.  The neurohypophysis, at 0.75 of term,  was f i l l e d with densely staining colloid droplets, and from all  166.  available data, the amount of stainable neurosecretion would be expected to increase even further as the time of birth approached. When the results from the histochemical and pharmacological studies are viewed side by side, as shown above, a pattern in their relationship to one another becomes clear. Although in most instances a good correspondance was detected between the histochemical and pharmacological data, at the earliest stages of development, when biological activities were already present in the neurohypophysis, the hypothalamic nuclei as well as the  pituitary appeared void of stainable neurosecretory material.  Both, the discrepency and the correspondance betwen these results give useful indications of certain aspects of neurohypophysial function, and will be considered in the following. A discrepency between the f i r s t stainable neurosecretory activity and the f i r s t assayable biological activity had been noticed by previous authors (e.g. see Heller and Lederis, 1959; Yakovleva, 1965; Wingstrand, 1951). To explain this discrepency between the biological activities and stainable neurosecretion, one hypothesis was suggested by Heller and Lederis (1959).  These authors suggested that the  f i r s t low level of biological activities in the fetal neurohypophysis  were possibly elicited by agents other than  neurohypophysial peptides.  The paper chromatographic studies of  fetal seal neurohypophyses at 0.31 of term suggested that this was not the case.  The active agents in the fetal neurohypophysis  appeared to be arginine vasopressin, oxytocin and arginine vasotocin.  An alternative suggestion has been made by many  167. workers (see Sloper, 1966);  the stains used for detection of  neurosecretory material are specific for the carrier protein and not for the peptides themselves.  In this case, the  discrepency between the biological activities and stainable neurosecretion could be attributed to a possible later appearance of the carrier substance.  However, Rodeck in 1959 (see Sloper, 1966),  succeeded in obtaining the characteristic staining reactions of neurosecretory granules, using synthetic oxytocin.  This would  suggest that the stains used for the detection of neurosecretory material react with oxydized SH residues of both the carrier and the neurohypophysial  peptides themselves.  Although the studies on the seal fetuses invalidate the possibility that the f i r s t biological activities assayed are due to non-neurohypophysial agents, the possibility that the intensity of staining is influenced by changes in the amounts of the carrier, remains a very real one.  However, in the case of  the fetal studies, the most reasonable explanation for the discrepency is the most simple one - that histochemical methods are generally less sensitive than the biological assays.  This  contention seems even more reasonable when i t is considered that at 0.30 of term, the fetal seal neurohypophysis contained only enough biological activities to be accounted for by arginine vasopressin and  5 x 10~^ M oxytocin.  2 x 10"^ M  Although this  small amount of biological activity is sufficient to carry out vasopressor assay (sensitivity 10"^ M) and rat uterus assay -12 (sensitivity  10  M)  on the crude extracts,  be expected that staining methods  would  i t could hardly  detect  such  168.  a low concentration, even i f this concentration was effectively doubled, because of the presence of an equal amount of stainable carrier protein. A similar line of argument might be applied to the discrepencies which have been reported concerning the location of the f i r s t detectable neurosecretion during development.  Some  workers found the f i r s t neurosecretory material in the pars nervosa, while others reported that the f i r s t stainable neurosecretion was present in the hypothalamic nuclei (Yakovleva, 1965). accepts the concept of the hypothalamo-hypophysial  If one  system, as i t  was formulated by Scharrer and Bargmann, then in a system in which the hypothalamic nuclei represent the site of hormone synthesis and the pars nervosa serves merely as a storage  depot, one would  have expected that neurosecretion would arise f i r s t in the synthetic nuclei during the course of development.  The most likely explanation  for the findings, mentioned above, which are in disagreement with this concept, could well be a further question of the sensitivity of the staining methods.  In the seal fetus studied here, there  seemed to be a parallel appearance of neurosecretion in both the hypothalamus and the pars nervosa.  However, this could constitute  a false impression, which was due to the lack of material at some critical early phase of development. At gestational stages after 0.30 of term, a good correspondance was found between hypothalamic activity as determined by the extent of stainability, and neurohypophysial storage as determined by biological activities of the posterior pituitary.  169.  The relative increase in both hypothalamic function and the stored levels of the neurohypophysial  hormones is illustrated in Fig. 23.  The values for each stage of gestation are expressed as a percentage of the values recorded around term;  hypothalamic activity is  assessed by both the number of cells which contain neurosecretion and the size of these cells, as explained in Fig. 23. correspondance  The  between the hypothalamic activity and the neuro-  hypophysial hormone level is indeed striking.  Although i t is not  possible to draw definite conclusions on the basis of the limited number of observations and on the indirect nature of the measurements, a strictly mathematical view of this phenomena offers interesting possibilities.  If the neurohypophysial  agents were not released  from the gland in the 40 to 60 day period between these measurements, the rate of accumulation in the gland would have to exceed the rate of increase in the production.  Conversely, should the rate of loss  from the gland exceed the rate of increase in the production, the two graphs would again have to diverge from one another.  The close  parallel in these measurements would suggest that a relatively constant release of the hormones might take place from the fetal neurohypophysis.  However, a possible release of the fetal neuro-  hypophysial hormones could only be established with certainty after direct measurements of the neurohypophysial peptide content of the fetal blood.  170.  Figure 23. The correlation between hypothalamic production of neurohypophysial hormones and the detected biological activities in the gland during fetal development. Values are expressed as a proportion of those at around the time of birth.  The production of the hypothalamic  nuclei were calculated by multiplying the percentage of cells with neurosecretory activity with the proportional size of the cells (see Table VI). A.  Biological activities in the neurohypophysis activity in the para-ventricular nucleus.  and hypothalamic Solid line and  f u l l circles represent oxytocic activity of the neurohypophysis.  Open circles and broken line neurosecretory  activity in the paraventricular nucleus. B.  Vasopressor activities and the neurosecretory activity of the supra-optic nucleus.  Solid circles and solid line =  = vasopressor activity in the neurohypophysis, open circles and broken line = neurosecretion in the supra-optic nucleus.  171.  2.  The importance of arginine vasotocin in mammalian fetuses.  The discovery of arginine vasotocin in mammalian fetuses is an important step toward the understanding of the evolution of the neurohypophysial peptides;  i t also might prove to be the key to  the understanding of the physiological role of the fetal neurohypophysial  agents.  The evolution of the neurohypophysial peptides has been reviewed by several authors (Sawyer, 1967; Versteeg, 1967; follows:  Pickering and Heller, 1969)  Vliegenthart and and is viewed as  Arginine vasotocin, the oldest of the neurohypophysial  agents is a strongly basic peptide and can be found in the most primitive living vertebrate class, the Cyclostomes.  In the  Chondrychthyes a second neurohypophysial peptide can be found for the f i r s t time;  this is neutral in its chemical properties.  In  all other vertebrate classes one basic and at least one neutral peptide can be found in the neurohypophysis.  It is postulated  that somewhere, in the ancestors of Chondrichthyes a doubling of the vasotocin gene occurred (Vliegenthart and Versteeg, 1966) gave rise to the neutral peptides.  and  In this second line, a number  of point mutations took place which gave rise to several neutral analogues, while arginine vasotocin remained remarkably stable throughout evolution, until the appearance of Mammalia. In the mammals a point mutation in the vasotocin gene was supposed to have taken place, one which gave rise to arginine vasopressin.  A further mutation in the gene producing arginine  vasopressin, resulted in the appearance of lysine vasopressin,  172.  the characteristic antidiuretic hormone of the Suina family.  The  present demonstration that mammals retain the capacity to produce arginine vasotocin, at least in some stages of their l i f e cycle, is incompatible with the earlier theory concerning a point mutation in the vasotocin gene.  To remain within the framework of  the  evolutionary theory of these peptides, as outlined above, a second doubling of the vasotocin gene would have to be postulated. However, i t is possible that the answers to the parallel production of arginine vasopressin and arginine vasotocin in the fetal neurohypophysis will require a much greater adjustment in our thinking of the evolution of these peptides. some members of the Suiformes,  Although in  arginine vasopressin and lysine  vasopressin were found to be present together, this parallel occurence of the two antidiuretic agents is explained as a crossbreeding between two species and the consequent survival of both genes (Ferguson and Pickering, 1969).  It is clear that such an  explanation could not apply to the occurance of arginine vasotocin and arginine vasopressin in the fetal neurohypophysis;  i t can  only be viewed as the survival of the original gene after the appearance of a new mutant.  Two reports published in the last  few years suggest that a similar phenomena exists within the family of neutral analogues.  Pickering (1967) purified two  neutral peptides from the cobra pituitary:  mesotocin and  oxytocin, and in 1972 Acher and his co-workers found a mixture of 8 valine and 4 asparagine oxytocin in the neurohypophysis of the dogfish, Squalus acanthias.  These findings suggest that  with the more sophisticated purification methods available to  173.  neurohypophysial physiologists, previously undetected peptides might be found in the neurohypophysis of other species as well. Such a parallel occurance of several neurohypophysial peptides would require a great number of doubling of the genes throughout evolution, and would also require a considerably more complex control of the synthetic pathways than is presently believed. One highly hypothetical possibility is that the synthetic apparatus for the neurohypophysial peptides might retain the capacity to produce several analogues of the peptides.  In this  case an induction-repression mechanism would control the expression of one gene or ghe other, such as was proposed by Jacob and Monod (1961) for the production of enzymes.  The major drawback,  for developing such a scheme, is that basically very l i t t l e is known of the functions of most neurohypophysial analogues. The physiological role of the basic peptides, the vasopressins and arginine vasotocin has been clearly established as antidiuretic agents.  However, the importance of the neutral analogues is  poorly understood.  Among these, only the role of oxytocin in  childbirth and in nursing is clearly established, but the -importance of this peptide in the male also remains unknown (Sawyer, 1968).  Although the possible mechanisms initiating  the production of any particular neurohypophysial analogue cannot be understood without a knowledge of their physiological function, each description of a new analogue, or the' discovery of one in a species where i t was previously thought to be absent, aids in our understanding of their  evolution.  174The few preliminary experiments in the course of the present studies, suggested that the embryonic membranes might constitute a possible target organ for the fetal neurohypophysial agents. If this proves to be the case, after further experiments have been carried out, this could give new significance to the predominant antidiuretic activity during fetal development;  this might  also provide the key to the presence of arginine vasotocin during the f i r s t half of ontogenesis.  It seems clear, that further  research of the physiological function of the fetal neurohypophysial peptides is important not only from the point of neurohypophysial physiology, but also for the understanding of the evolutionary significance of arginine vasotocin in mammalian fetuses.  175. SUMMARY  The biological activities were estimated in the lyophilised posterior pituitaries of the adult, non-breeding and of the pregnant seal, Callorhinus ursinus, at five stages of gestation.  The pooled  extracts were purified and the composition of the active peptides was determined by amino acid analysis.  The neurohypophysial  activity of fetuses was estimated throughout intrauterine l i f e , by histochemical and pharmacological methods.  The glands were  subjected to purification procedures, and the structure of the active peptides was determined. A. Adult 1. The posterior pituitaries of the adult seals were found to be exceptionally potent for both vasopressor and rat uterus activities.  The ratios of the two biological activities  V/RU were found to be within the limits of those recorded for other, non-marine, mammals. 2. The vasopressor and rat uterus activities in the posterior pituitaries of pregnant seals were found to be fluctuating with the different stages of pregnancy.  Following an i n i t i a l  rise in potency at 0.31 of term, the vasopressor activity was found to be depressed through the major portion of pregnancy.  In contrast,  the oxytocic activity was elevated from the control levels a l l through pregnancy, although a fluctuation in activities from one stage to the other parallelled those for vasopressin.  176.  3.  The purification and amino acid analysis of the  active agents of the seal neurohypophysis have shown them to be arginine vasopressin and oxytocin. B.  Fetal neurohypophysis 1.  In the fetal neurohypophyses the f i r s t stainable  neurosecretory material was detected at around mid-gestation, when the  hypothalamic nuclei and the neurohypophysis have all contained  stainable material.  The intensity of staining in the neurohypophysis,  as well as the number and size of the neurons in the hypothalamic nuclei was found greatly increased around the time of birth. 2.  Vasopressor, rat uterus and frog bladder activities  were detected in the fetal neurohypophysis from the earliest gestational stage, 0.19 of term.  The vasopressor and rat uterus  activities were found to be increasing stage by stage, until the time of birth approached.  The frog bladder activity reached its  maximum at 0.68 of term, and rapidly diminished from this stage, until birth. 3.  The ratios of V/RU activities in the fetal glands  were found to be unusually low.  They reached a maximum value of  5.0 at 0.44 of term, gradually declining to a value of 1.8, by 0.93 of term.  The FB/RU ratios following an i n i t i a l rise, have  shown a steadily declining pattern all the way through gestation, and by 0.93 of term they approached the value of 1.0, attributable to the presence of oxytocin in the samples alone.  177.  4.  The purification, amino acid analysis, and  pharmacological comparisons of the active neurohypophysial agents from the fetal glands have shown them to be arginine vasopressin, arginine vasotocin and oxytocin. 5.  Preliminary experiments on the allantoic and  amniotic membranes of guinea.pigs have suggested that these membranes might constitute the possible target organs for the fetal neurohypophysial peptides.  178. -LITERATURE QUOTED Acher, R., Chauvet, J., Lenci, M.T., 1960. Isolement de l'oxytocine du poulet. Biochim. Biophys. Acta. 38: 344-345. Acher, R., Chauvet, J., Chauvet, M.T., 1963. Les hormones neurohypophysiaires de mammiferes: Isolement et caracterisation de l'oxytocine et de l a vasopressine de l a baleine (Balaenoptera physalus). Bull. Soc. Chim. Biol. 45: 1369-1378. Acher, R., 1963. The comparative chemistry of neurohypophysial hormones. Symp. Zool. Soc. London, 9_: 83-91. Acher, R., 1966. Chemistry of neurohypophysial hormones. In: The Pituitary Gland, ed. G.W. Harris and B.T. Donovan, p. 269-287, Butterworth. Acher, R., Chauvet, J. and Chauvet, M., 1972. Evolution Moleculaire. Identification de deux nouvelles hormones neurohypophysaires la Valitocine (val-8-oxytocin) et 1'Aspartocine (Asn-4oxytocine) chez un poisson selacien, l'Aiguillet (Squalus acanthias). C..R. Acad. Sci. Paris, 274: 313-316. Adamson, K., Engel, S.L., vanDyke, H.B., Schmidt Nielsen, B. and Schmidt Nielsen, K., 1956. The distribution of oxytocin and vasopressin (antidiuretic hormone) in the neurohypophysis of the camel. Endocrinology, 58: 272-278. Alexander, P.D. and Nixon, D.A., 1962. The fetal kidney. Brit. Med. Bull., 17: 112-117. Ames, R.G., 1953. Urinary water excretion and neurohypophysial function in f u l l term and premature infants, shortly after birth. Pediatrics, 12: 272-282. Amoroso, E.C., Bourne, Harrison, R.J., Harrison Matthews, L., Rowlands, I.W. and Sloper, J.C., 1965. Reproductive and endocrine organs of fetal, newborn and adult seals. J. Zool. 147: 430-486. Barnett, H.L. and Vesterdal, J., 1953. The physiologic and c l i n i c a l significance of immaturity of kidney function in young infants. J_. Pediatr. _42: 99-119. Barholomew, G.A., Hoel, P.G., 1953. Reproductive behavior of the Alaska Fur Seal, Callorhinus ursinus, .J. Mammal. 34: 417-436. Bargmann, W. and Scharrer, E., 1951. The site of origin of the hormones of the posterior pituitary. Amer. Scientist, 39: 255-259.  179.  Berde, B. and Boissonnas, R.A., 1968. Basic pharmacological properties of synthetic analogues of the neurohypophysial hormones. In: Handbuch der Experimentallen Pharmakologie, New Series, v.23. 0. Eichler et a l (eds.). pp 802-870. Springer Verlag, Berlin. Bernischke, K. and McKay, D.G., 1953. The Antidiuretic Hormone in Fetus and Infant. Obst. Gynec. JL: 638-649. Bisset, G.W., Clark, B.J., Haldar, J., Harris, M.C., Lewis, G.P., and Rocha e Silva Jr. M., 1967. The assay of milk ejecting activity in the lactating rat. Brit. ^J. Pharmacol. Chemoter. 31: 537-549. Chauvet, J., Lenci, M.T. and Acher, R., 1960. Presence de deux vasopressines dans l a neurohypophyse du poulet. Biochim. Biophys. Acta, 38: 571-573. Cobo, E., de Bernal, M.M., Gaitan, E. and Quintero, C.C., 1967. Neurohypophysial hormone release in human. Amer. J. Obst. & Gynec. 97: 519-529. Craig, A.M., 1964. Histology of Reproduction and the Estrus Cycle in the Female Fur Seal, Callorhinus ursinus. J_. Fish. Res. Bd. Canada, 21: 773-811. Cross, B.A. and Silver, F.A., 1965. Effect of luteal hormone on the behaviour of hypothalamic neurons in pseudopregnant rats. J. Endocrinol. 31: 251-263. Csapo, A., Dale, H.H.,  1961.  Progesterone block. Am. J_. Anat. , 98: 273-291.  1909. Actions of extracts of the pituitary body. Biochem. J. 41: 427-447.  Daniel, P.M. and Pritchard, M.M.L., 1966. Distal Retraction Balls in the Neurohypophysis after transection of the Pituitary Stalk. J. Comp. Neurology, 127: 321-333. Dawson, A.B., 1953. Evidence for the termination of neurosecretory fibers within the pars intermedia of the hypophysis of the frog, Rana pipiens. Anat. Rec. 115: 63-70. Dekanski, J., 1952. The quantitative assay of vasopressin. Brit. J_. Pharmacol. _7_ 567-572. :  Dicker, S.E. and Heller, H., 1951. The mechanism of water diuresis in adult and newborn guinea-pigs. J^. Physiol. 112: 272-282. Dicker, S.E. and Tyler, C , 1953. Vasopressor and Oxytocic activities of the pituitary glands of rats, guinea-pigs and cats and of human foetuses. J^. Physiol. 121: 206-214. Diepen, R.F., Engelhardt, R.F. and Smith Agreda, V., 1954. Uber Ort und Art der Entstehung des Neurosecretes im SupraopticoHypophysaren System bei Hund und Katze. Erg. H. Anat. Anz. 101: 276-286.  180.  du Vigneaud, V., Ressler, C , Swan, J.M., Roberts, C.W., Katsoyannis, P.G. and Gordon, S., 1953. Synthesis of an octapeptide with hormonal activity of oxytocin. J. Amer. Chem. Soc. 75: 4879-4880. Eaker, D. and Porath, J., 1967. Sorption Effects in Gel Filtration: 1. A Survey of Amino Acid Behavior on Sephadex G-10. Separation Science 2: 507-550. Enemar, A., 1961. The structure and development of the hypophysial portal system in the laboratory mouse, with particular regard to the primary plexus. Ark. Zoo!. (2 ser.) 1_3: 203-231. Farmi, F. von, 1913. Ueber Diabetes insipidus Hypophysistherapie. Wien. Klin. Wschr. 26: 1867. Farrell, G., Fabre, L.F., Rauschkolb, E.W., J. Physiol. 180: 846-863.  1965.  The neurohypophysis.  Ferguson, D.R. and Heller, H., 1965. Distribution of neurohypophysial hormones in mammals. J_. Physiol. 180:846-863. Ferguson, D.R. and Pickering,B.T., 1969. Arginine and Lysine vasopressins in the Hyppopotamus Neurohypophysis. Gen. Comp. Endocrinol. T3: 425-429. Fitzpatrick, R.J., 1966. The posterior pituitary gland and the female reproductive tract. J_n: The Pituitary Gland, (ed.). Harris, G.W. and Donovan, B.T., Butterworths, London. Fitzpatrick, R.J., 1966. The neurohypophysis and the male reproductive tract. J_n: The Pituitary Gland, (ed.). Harris, G.W. and Donovan, B.T., Butterworths, London. Follett, B.K. and Heller, H., 1964. The neurohypophysial hormones of bony fish and cyclostomes. J_. Physiol. (London) 172: 74-91. Frank, E., 1912. Ueber Beziehungen der Hypophyse zum Diabetes insipidus. Berliner Klin. Wschr. 49: 393-397. Fuse, G.,  1939. Ueber die Hypophyse beim Seebaren (Callorhinus ursinus, GRAY). Sendal Anatomisches Institut Arbeiten, 22: 137-161.  Geiling, E.M.K., 1937. Hypophysis cerebri.of finback and sperm whale. Bull. J. Hopkins. 57: 123. Ginsburg, M., 1968. Production, Release, Transportation and Elimination of the Neurohypophysial Hormones. In: Handbuch der Experimental 1 en'Pharmakologie, New Series, 0. Eichler et a l (eds.). vol. 23, pp 286-371. Springer-Verlag.  181.  Gomori, G., 1941. Observations with differential stains on human islets of Langerhans. Amer. J_. Pathol. J_7: 395-406. Green, G.D. and van Breemen, V.L., 1955. Electron microscopy of the pituitary and observation of neurosecretion. Am. J_. Anat. 95: 177. Halmi, N.S., 1952. Differentiation of two types of basophils in the adenohypophysis of rat and mouse. Stain Tech. 27: 61-64. Hann, F. von, 1918. Ueber die Bedeutung der Hypophysenveranderungen bei Diabetes Insipidus. Frankfurt. 7_. Path. 21: 337-365. Harrison, R.J., 1969. In: The Biology of Marine Mammals. Ed. Andersen, H.T., Academic Press, N.Y. and London, pp 295-337. Heller, H. and Zaimis, E.J., 1949. The Antidiuretic and Oxytocic Hormones in the Posterior Pituitary Glands of Newborn Infants and Adults. J. Physiol. 109: 162-169. Heller, H. and Lederis, K., 1959. neurohypophysial system.  Maturation of the hypothalamoJ;. Physiol. 147: 299-314.  Heller, H., 1961. Occurance, Storage and Metabolism of Oxytocin, (ed.) Caldeyor-Barcia, R. and Heller, H. p. 3-24. Pergamon Press. Heller, H., 1966. The hormone content of the vertebrate hypothalamoneurohypophysial system. Brit. Med. Bull. 22: 227-231. Heller, H. and Pickering, B.T., 1961. Neurohypophysial hormones of non-mammalian vertebrates. J_. Physiol. (London), 155: 98-114. Heller, H. and Ginsburg, M., 1966. Secretion, metabolism and fate of posterior pituitary hormones. J_n:-The Pituitary Gland, v.3. G.W. Harris and B.T. Donovan (eds.). pp 330-373. Butterworths. Holton, P., 1948. A modification of the method of Dale and Laidlaw for standardization of posterior pituitary extracts. Brit. J_. Pharmacol. 3: 328-334. Jacob,F. and Monod, J., 1961. Genetic Regulatory Mechanism in the synthesis of Proteins. J_. Mol. Biol. 3: 318-356.  182.  Kamm, 0., Aldrich, T.B., Grote, I.W., Rowe, L.W. and Bugbee, E 1928. The active principles of the posterior lobe of the pituitary gland. J_. Amer. Chem. Soc. 50: 573-601. Kivalo, E., Talanti, G., 1957. Foetal development of the hypothalamo-hypophysial neurosecretory system of the cow embryo. Acta Endocrinol. 26: 471-476. Labsethwar, A.P., Collins, W.E., Tyler, W.J. and Casida, L.E., 1964. Effect of progesterone and oxytocin on the pituitary-ovarian relationship in heifers. J_. Reprod. Fert. 8: 77-83. Marshall's Physiology of Reproduction. 3rd Ed., 1962. Longmans, Green and Co. Ltd., p. 964-975.  Parkes, A.S  Martini, L. 1966. Neurohypophysis and Anterior Pituitary Action. In: The Pituitary Gland. Harris, G.W. and Donovan, B.T. (eds.). pp 535-577, Butterworths. Meitner, E.R., 1961. Neurosecretion and the Pituitary. Symposia, 9.: 18-21.  Ciba  Mellor, D.J. and Slater, J.S., 1971. Daily changes in amniotic and allantoic fluid during the last three months of pregnancy in conscious, unstressed ewes, with catheters in their foetal fluid sacs. J. Physiol. 217: 573-604. Mirsky, I.A., 1966. Metabolic Effects of the Neurohypophysial Hormones and Related Polypeptides. I_n: Handbuch der Experimentallen Pharmakologie, New Series. 0. Eichler. et a l (eds.). v. 23, pp 613-624, Springer-Verlag. More, S., Stein, W.H. 1951. Chromatography of amino acids on sulphonated pblysteirene resins. J_. Biol. Chem. 192: 663. Moore, S., Stein, W.H. 1956. Column Chromatography of Peptides and Proteins. Advanc. Protein Chem. ]_]_: 191-236. Morris, C.J.O.R. and Morris, P. 1965. Separation methods in biochemistry. Pitman and Sons, London. Munsick, R.A. 1960. Effect of magnesium ions on the response of the rat uterus to neurohypophysial hormones and analogues. Endocrinol. 66: 451-456. Munsick, R.A. 1964. Neurohypophysial Hormones of Chickens and Turkeys. Endocrinol. 75: 104-112.  183.  Munsick, R.A. 1968. The effect of neurohypophysial hormones and similar polypeptides on the uterus and other extravascular smooth muscle tissue. Jji: Handbuch der Experimental 1 en Pharmakologie, New Series, v. 23. 0. Eichler et al_ (eds.). pp 443-474. Springer-Verlag. Oliver, G, Schafer, E.A., 1895. On the physiological action of extracts of pituitary body and certain other glandular organs. J_. Physiol. (London) 18.: 277-279. Ott, I. and Scott, J . C , 1910. The action of infundibulum upon mammary secretion. Soc. for Exp. Biol. and Med. Proc. 8: 48-49. (N.Y.). Perks, A.M., of of 6:  1966. Pharmacological and chromatographic studies the neurohypophysial activities of the pituitary further elasmobranch species. Gen. Comp. Endocrinol. 428-442.  Pickering, B.T., 1967. The neurohypophysial hormones of a reptile species, the cobra (Naja Naja). J_. Endocrinol. 39: 285-294. Pickering, B.T., 1970. Aspects of the relationship between the chemical structure and biological activity of the neurohypophysial hormones and their synthetic analogues. In: Pharmacology of the Endocrine System. Pickering, B.T. and Heller, H., 1969. Oxytocin as a neurohypophysial hormone in the Holecephalian elasmobranch fish, Hydro!agus collei. J_. Endocrinol. 45: 597-606. Pickford, M., 1961. Some extra-uterine actions of oxytocin. J_n: oxytocin. Caldeyro-Barcia, R. and Heller, H. (eds.). p. 68-87. Pergamon Press Pickford, M., 1964. Effects of the neurohypophysial hormones on the vascular system. J_n: Oxytocin, Vasopressin and their structural analogues, (ed.) Rudinger, J . p. 31-37. Pergamon Press. Pickford, M., 1966. Neurohypophysis and kidney function. Iji: The-Pituitary Gland. p. 374-398. G.W. Harris and B.T. Donovan (eds.). Butterworth. Rodeck, H. and Caesar, R.S., 1956. Zur entwicklung des Neurosecretorischen System bei Saugern und Mensch und der regulationsmechanismen des Wasserhaushaltes. 2. Zellforsch., 44: 666-691.  184.  Ruttenberg, M.A., King, T.P., Craig, L.C., 1965. The Chemistry of Tyrocidine. VI. The Amino Acid Sequence of Tyrocodine C. Biochemistry 4: 11-18. Saameli, K., 1968. The Circulatory Action of the Neurohypophysial Hormones and Similar Peptides. J_n: Handbuch der Experimental 1 en Pharmakologie, New Series, 0. Eichler, e_t al_. (eds.). vol. 23, pp 545-612. Springer-Verlag. Sachs, H., Takabatake, Y., 1964. Evidence for a precursor in vassopressin biosynthesis. Endocrinol. 5_: 943-948. Sachs, H., 1967. Biosynthesis and release of vasopressin. Amer. J. Med. 42: 687-700. Sawyer, W.H., 1960. Increased water permeability of the bull frog (Rans catesbiana) bladder in vitro in responses to synthetic oxytocin and arginine vasotocin extracts from non-mammalian vertebrates. Endocrinol. 66: 112-120. Sawyer, W.H., 1961a. Bioassays of oxytocin and vasopressin. Methods in Med. Res. 9: 210-219. Sawyer, W.H., 1965. Active Neurohypophysial Principles from the Cyclostome (Petromyzon marinus) and Two Gartilaginous Fishes (SqUalus aCanthias and Hydro!agus c o l l e i) . Gen, and Comp. Endocrinol. 5: 427-439'. Sawyer, W.H., 1967. Evolution of Antidiuretic Hormones and Their Functions. Amer. J_. Med. 42_: 678-686. Sawyer, W.H., 1968. Phylogenetic aspects of the neurohypophysial hormones. In: Handbuch der Experimental 1 en Pharmakologie, New Series, vol. 23. 0. Eichler et al (eds.) pp 717-747. Springer-Verlag. Sawyer, W.H., Wuu, T.C., Baxter, J.W.M. and Manning, M., 1969. 4-Proline Analogues of Neurohypophysial Hormones: Hypothetical Intermediates in Peptide Evolution. Endocrinol. 85: 385-388. Sawyer, W.H., 1970. Homologies of structure and function among neurohypophysial peptides. Second Annual PCRI Winter Symposia. Miami Scharrer, E., Scharrer, B., 1954. Hormones produced by neurosecretory cells. Rec. Prog. Hor. Res. ]_0: 183-232. Seeds, E.A., 1965. Water metabolism of the fetus. & Gynec. 92: 727-745.  Am. J_. Obst.  Seeds, E.A., 1970. Osmosis across term human placental membranes. Amer. J_. Physiol. 219: 551-554.  185.  Sloper, J.C., 1966. The Experimental and Cytopathological Investigation of Neurosecretion in the Hypothalamus and the Pituitary. J_n: The Pituitary Gland. Harris, G.W. and Donovan, B.T. (eds.). vol. 3. pp 131-239, Butterworth. Stewart, A.D., 1968. Genetic variation in the neurohypophysial hormones of the mouse. J_. Endocr. 41. Proc. Soc. Endocr. p. XIX. Taylor, S.P., 1954. Ion-exchange Chromatography of Oxytocin, Arginine vasopressin, and Lysine vasopressin. Soc. for Exp. Biol. and Med. 85: 226-228. Thorn, N.A., 1968. The Influence of the Neurohypophysial Hormones and Similar Polypeptides on the Kidney. J_n: Handbuch der Experimentallen Pharmakologie. New Series. 0. Eichler etal_(eds.). vol. 23. pp 372-442. Springer-Verlag. Ussing, H.H. and Anderson, B., 1957. Solvent drag on non-electrolytes during osmotic flow through isolated toad skin and its response to antidiuretic hormone. Acta Physiol. Scandinav. 28: 60. von den Velden, R., 1913. Die Nierenwirkung von Hypophysenextraktenbeim Menschen. Berliner Klin. Wschr. 50: 2083-2086. Vizsolyi, E., 1968. Studies of the sheep neurohypophysia during pregnancy and fetal development. M.Sc. thesis, University of British Columbia. Vizsolyi, E. and Perks, A.M., 1969. A new neurohypophysial principle in foetal mammals. Nature 223: 1169-1171. Vliegenthart, J.F.G. and Versteeg, D.H.G., 1967. The evolution of the vertebrate neurohypophysial hormones in relation to the genetic code. <h Endocrinol. 38: 3-12. Vogt, M.,  1953. Vasopressor antidiuretic and oxytocic activities of extracts of the dog's hypothalamus. Brit. J_. Pharmacol. 8: 193-196.  Waidl, E. and Semm, K., 1959. Der Beginn der Hypothalamischen Neurosecretion in der Fetalzeit. Archiv. fur Gynecologie 192: 269-276.  186.  Wakim, K.G., 1966. Reassessment of the Source, Mode and Locus of Action of Antidiuretic Hormone. Amer. J. Med. 42: 394-411. Waring, H. and Landgrebe, F.W., 1950. Hormones of the posterior pituitary. I_n: The Hormones. Pincus, G. and Thimann, K.F. (eds.). New York, Academic Press. Wilson, N., 1968. Isolation and amino acid sequence of neuro" hypophysial hormones of Pacific chinook salmon (Oncbrhyncbus tschawytscha). PhD thesis, University of British Columbia. Wingstrand, K.G., 1953. Neurosecretion and antidiuretic activity in chick embryos with remarks on the subcommissural organ. Ark. Zool. 6: 41-67. Yakovleva, I.V., 1965. The hypothalamo-hypophysial neurosecretory system in the early ontogenesis of vertebrates including men. Arkhiv Anat. Gist. & Embryo!. 48: 79-90.  187.  APPENDIX A  Preliminary experiments with Sephadex G-l5 and G-25 gels. a. Conditions for the maximal resulution of the active neurohypophysial peptides from Sephadex G-l5 columns. In four experiments, three different lots of Sephadex G-l5 gel were used.  The size of the columns (200 x 2.5 cm), the elution  and suspending media (0.2M acetic acid) and the flowrate (12 ml/hr) were kept identical in these experiments. gel  Only the lots of the  and the method for building the columns were varied.  columns were built by one of two alternate methods.  The  In the f i r s t  case, the gel slurry was introduced in one step into a 2,000 ml funnel, attached to the top of the.column, and the gel was stirred in with an electric mixer from this suspension over a period of 8 - 10 hrs.  In the alternate method, small portions of  Sephadex G-15 slurry were added to a 250 ml funnel (attached to the top  of the column) at half an hour intervals.  The slurry in the  funnel was continuously stirred with an electric mixer, as the gel slowly settled into the column.  The building procedure was  completed over a period of 32-38 hrs.  Purifications of seal  neurohypophysial extract were carried out through these columns, as described in the Methods (p. 36).  The eluates were assayed  for their rat uterus and vasopressor activities, and the results are shown in Fig. 24. In the f i r s t two experiments, lot number 2014 Sephadex G15 gel was used.  The oxytocic and vasopressor activities eluted in  188.  Figure 24. Resolution of the oxytocic and vasopressor activities on Sephadex G-15 columns. Suspension and eluting media was 0.2 M acetic acid.  Column  size 200 x 2.5 cm, and eluates were collected in 2,8 ml fractions. Full circles and solid lines  =  rat uterus activity  Triangles and broken lines.  =  vasopressor activity  A. Lot number 2014.  Column built slowly.  Loaded 8 ml of crude  seal neurohypophysial extract, representing 145,000 mU of vasopressor and 125,000 mil of rat uterus activity. B. Lot number 2014, built in one step.  Loaded 145,000 mU of  vasopressor and 125,000 ml) of rat uterus activity in 8 ml of crude seal neurohypophysial extract. C. Lot number 435, built in one step.  Loaded 104,000 mU of  vasopressor and 98,500 mU of rat uterus activity. D. Lot number 8097, built slowly.  Loaded 104,000 mU of  vasopressor and 98,000 ml) of rat uterus activity in 8 ml of crude seal neurohypophysial extract.  189.  two sharp, well separated peaks from the slowly built column (24 A) but the peaks appeared much broader and overlapped  considerably  when eluted from the column, built with the single addition of the gel (24 B).  The column built by a single addition of the gel  slurry yielded the same unsatisfactory results when another lot of gel was used (# 435) as is illustrated in Fig. 24 C.  However a  good resolution was again obtained from a slowly built column when yet another lot of gels was used (# 8097), as shown in Fig. 24 D.  It seemed clear from these experiments, that although  the resolution of the oxytocic and vasopressor activities varied slightly from one lot of gel to another (Fig. 24 A & D) maximum resolution of these peptides could only be obtained when the columns were built by the deliberately slow method.  This was  possibly due to a layering of the particles according to their size,  when the gel slurry was added in one step, and an uneven  packing resulting from the progressively thinner suspension in the large funnel.  Such an uneven packing and layering the gel  could well account for the decreased resolution of the neurohypophysial peptides in the eluates. b. The extent of purification on Sephadex G-15 columns. 30 mg of lyophilised seal posterior pituitary powder was extracted in 8 ml of 0.25% acetic.acid.  The extract was exposed  to ultraviolet light for 12 h r s t o destroy the biologically active peptides.  To this inactivated solution, 3 ml of synthetic  oxytocin (Syntocinon, lOI.U/ml) and 1 ml of Piterssin  (2(0.U of  190.  Figure 25. Separation of the major protein fraction from the biologically active peptides on Sephadex G-15. Suspending and eluting media was 0.2 M acetic acid.  Loaded  30,000 mU of synthetic oxytocin and 20,000 mU of mixed arginine and lysine vasopressin (Pitressin) onto a 200 x 2.5 cm Sephadex G-15 column.  Eluates collected in 2.8 ml fractions.  Solid triangles and solid line  =  Lowry peptide measurements  Open circles and broken line  =  Rat uterus activity  Solid circles and solid line  =  vasopressor activity  Oxytocic  9  IOOOJ 2 0 0  Vasopressor  800H E \ E  60CH  Lowry  3  \ 100  E 400-4  l I  O\  200H 0  —I lube  100 number  \  150  200  191.  arginine and lysine vasopressin) was added.  The final volume of  12 ml was loaded onto a 200 cm x 2.5.cm Sephadex G-15 column. The Lowry peptide concentration, biological activities and conductivity were measured in each fraction. of protein (Lowry peptide)  The major peak  eluted well ahead of the biological  activities, and was separated from them by 80 fractions (2.8 ml/fraction).  The rat uterus and vasopressor activities eluted  in two sharp well separated peaks.  No increase in conductivity,  above that of the supporting 0.2 M acetic acid was  detected.  The results are as shown in Fig. 25. c. Preliminary experiment with pyridine washed Sephadex G-25 gel. The purpose of this experiment was to eliminate the ion exchange groups from Sephadex G-25.  Without these ion exchange  groups the aromatic absorption capacity of the gel should have determined the relative place of elution of oxytocin,arginine vasotocin and arginine vasopressin.  In the following experiment  Sephadex G-25 gel was poured into a 100 x 2.5 cm column and  was  washed with a volume of pyridine, that equalled five times the void volume of the column.  The pyridine was washed out with  0.2 M acetic acid until the eluent from the column was free of pyridine and its pH was identical to that of the 0.2 M acetic acid. A mixture of 10,000 mU of synthetic oxytocin  (Syntocinon,  10 I.U./ml), 20,000 mU of Pitressin, and 1,000 mU arginine vasotocin, (rat uterus activity) was loaded onto the column, in a medium which was 0.25% for acetic acid.  Elution was carried  192.  Figure 26. Separation of arginine vasopressin, arginine vasotocin and oxytocin on pyridine washed G-25 Sephadex column. Suspending and eluting media was 0.2 M acetic acid.  Loaded,  10,000 mU of synthetic oxytocin (Syntocinon), 20,000 mU of vasopressor activity (Pitressin) and 1,000 mU arginine vasotocin (rat uterus) onto a 100 x 2.5 cm column. collected in 2.8 ml fractions. Open circles and broken line  =  rat uterus activity  Solid circles and solid line  =  vasopressor activity  Eluate  193.  out with 0.2 M acetic acid.  Biological activities eluted in  three peaks (see Fig. 26) after 300 ml of acetic acid was put through the column.  Rat uterus activity was f i r s t eluted in  a peak which overlapped with arginine vasotocin, and arginine vasopressin, as expected, eluted.last.  However the recoveries  for a l l three peptides approached only 20-30%, probably due to the strong adsorption of the peptides to the gel. Consequently the method was abandoned as unsuitable for the present investigations.  194.  APPENDIX B  Precycling of the ion exchange resins. a. CM Sephadex. The resin was swelled in dist. water, and then was washed repeatedly on a Buchner funnel with 0.5 N NaOH.  The excess of  NaOH was removed by repeated washings with d i s t i l l e d water until all the base was washed off. 0.5 N HCI and 0.5 N NH 0H. 4  The procedure was repeated with This last step was necessary to convert  the resin to its ammonium salt.  The ammonium hydroxyde was  washed out with d i s t i l l e d water, and a final washing of the resin was carried out with 0.002 M ammonium acetate solution at a pH of 5.0, the starting buffer.  The gel was taken up in a  large volume (cc 10 times the volume of the resin) in the starting buffer, and stirred on a magnetic stirrer.  The buffer solution  was decanted and exchanged to new buffer, until the conductivity and pH of the solution, after stirring with the resin, was identical to that of the starting buffer.  The resin was then  considered ready for the building of the columns. b. IRC-50. The resin was swelled in d i s t i l l e d water and washed repeatedly on a Buchner funnel with 2 N NaOH.  The NaOH was washed  out with d i s t i l l e d water, and the resin.was washed with 2 N HCI, and rinsed once again in d i s t i l l e d water until a l l the acid was  195.  removed.  The resin was stored in d i s t i l l e d water at 4 C until  poured into the columns, c. Dowex 50 X2. The precycling procedure was carried out according to Moore and Stein, 1951.  The resin was washed in 4 N HCI on a  Buchner funnel, using a gentle suction, until a colourless f i l t r a t e came through, using approximately 2 liters of solution. The resin was washed in d i s t i l l e d water, and with 2 N NaOH until the f i l t r a t e was alkaline.  The resin was suspended in 1 N NaOH,  stirred, and was allowed to settle.  The supernatant was  decanted, and the procedure was repeated three times.  After  the final washing with NaOH, the gel was washed three times in 1 N NH^OH to convert the resin to its NH salt, washed with 4  distilled water, and finally washed in 0.2 M pH 4.0 ammonium acetate starting buffer, until the pH and the conductivity of the washings from the resin were equal to that of the starting buffer.  196.  APPENDIX C Staining methods, a. Aldehyde fuchsin - light green staining (after Gomori, 1941; Halmi, 1952; Dawson, 1953). 1.  De-wax sections and hydrate through descending grades of alcohol to water.  2.  Post-fix in Bouin's fixative for 16-18 hrs.  3.  Wash in tap water.  4.  Oxidize in solution containing 0.3% of  15 minutes  each KMn0 and H S0 . 4  2  1 minute  4  Decolourise in 2.5% aquaous sodium bisulphite. 15 seconds Wash in tap water.  5 minutes  Stain in aldehyde fuchsin. 60% alcohol  15-20 minutes  1000 ml  basic fuchsin  0.5 g  paraldehyde  1 ml  cone. HCI  1.5 ml  Ripens at 20°C in 2-3 days.  Keeps 6-8 weeks,  8.  Rinse slides in 2-3 changes of 95% alcohol.  9.  Wash in tap water.  10.  Counterstain in 0.5% Light Green.  11.  Differentiate counterstain rapidly, in 95% alcohol.  12.  Dehydrate, clear, and mount.  10 minutes 1 minute  197.  b. Alcian blue periodic acid Schiff (PAS - orange G method. Deparaffinize and hydrate through descending grades of alcohol to water. 2.  Post-fix in Bouin's fixative for 16-18 hrs.  3.  Oxidize in Gomori's mixture (2.5% potassium permanganate and 5% sulfuric acid;  1 part each;  distilled water 6 parts.  1% minutes  4.  Bleach in 2% sodium bisulfite.  1  5.  Wash in running tap water.  6.  Stain with AB pH 3.0 (1% alcian blue in 1% glacial  5  minute minutes  acetic acid).  20  minutes  7.  Wash in running tap water.  10  minutes  8.  Oxidize in 5% periodic acid.  1% minutes  9.  Wash in running tap water.  5  minutes  20  minutes  10.  Immerse in S c h i f f s reagent. Pour 200 ml boiling distilled water on Ig basic fuchsin.  Shake well for 5 minutes,  cool to exactly 50°C, and f i l t e r . sodium bisulfite.  Add 1 g  Stopper and store in the  dark at room temperature for 24 hrs.  The  solution has a light orange.colour.  Store at  0 - 5°C.  Can be kept for several months in  a well stoppered v i a l . colour develops.  Discard when a pink  198.  11.  Wash in three changes of sulfige rinse.  2 minutes each  Add 5 ml of 1 N HCI and 100 ml of distilled water to 6 ml of 10% sodium bisulfite solution, shortly before use. 12.  Wash in running tap water.  13.  Counterstain with 2% orange G in 1% phosphotungstic acid.  14.  Do not bring to low graded alcohol, but rinse quickly in 96% alcohol and continue dehydration in absolute alcohol and xylene.  Mount in Caedax.  5 minutes  15 seconds  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0101469/manifest

Comment

Related Items