Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The influence of vasopressin and prolactin on the movement of water and sodium through the isolated amnion… Holt, William Faulkner 1975

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1975_A6_7 H06.pdf [ 6.87MB ]
Metadata
JSON: 831-1.0093429.json
JSON-LD: 831-1.0093429-ld.json
RDF/XML (Pretty): 831-1.0093429-rdf.xml
RDF/JSON: 831-1.0093429-rdf.json
Turtle: 831-1.0093429-turtle.txt
N-Triples: 831-1.0093429-rdf-ntriples.txt
Original Record: 831-1.0093429-source.json
Full Text
831-1.0093429-fulltext.txt
Citation
831-1.0093429.ris

Full Text

THE INFLUENCE OF VASOPRESSIN AND PROLACTIN ON THE MOVEMENT OF WATER AND SODIUM THROUGH THE ISOLATED AMNION OF THE FETAL GUINEA-PIG by WILLIAM FAULKNER HOLT BSc. (Honours) Memorial University of Newfoundland, 1973 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in the Department of Zoology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June, 1975 In present ing th is thes is in p a r t i a l fu l f i lment o f the requirements for an advanced degree at the Un ivers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f ree ly ava i lab le for reference and study. I fur ther agree that permission for extensive copying of th is thes is for s c h o l a r l y purposes may be granted by the Head of my Department or by h is representa t ives . It i s understood that copying or p u b l i c a t i o n of th is thes is for f i n a n c i a l gain sha l l not be allowed without my wri t ten permiss ion. Department of Zoology  The Un ivers i ty of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1WS Date June 20. ABSTRACT Unidirectional flux of tr i t i a t e d water across isolated guinea-pig amnion was studied i n a perfusion c e l l supplied with amniotic saline at 3 7 ° C by means of a continuous double-circulation system. The diffusional permeability in the absence of an osmotic gradient was 3.09 + 0.15 x 10 "^ cm sec for maternal-fetal flow (10 experiments), and 2.62 + 0.1*r x 10""^ em sec" 1 for fetal-maternal flow (13 experiments). The addition of vasopressin (50-500 mU/ml) to the fe t a l side of sixteen membranes set up between amni-otic and maternal salines increased isotopic water flux in the maternal-fetal direction against an osmotic gradient of 28 m0sm/l. Treatment of eight other membranes with 10 ug/ml prolactin reduced fetal-maternal water flow by up to 17.0 % at the end of 3 hours, i n the absence of an osmotic gradient. This contrasted with five control experiments, i n which fetal-maternal flow increased by 5.1 f» at the end of 3 hours. Therefore, the prolactin appeared to reduce diffusional fetal-maternal v/ater flow by up to 22.1 fo within 3 hours. In twelve further experiments, net flow (fetal-maternal) was measured gravimetrically. The membranes were set up between amniotic and maternal salines, i n conditions which paralleled the natural ionic environment, and the natural i i • * • 1 1 1 hydrostatic and osmotic gradients. In five experiments, the addition of prolactin (20 ug/ml) to the fetal surface of the amnion, caused a decrease i n fetal-maternal water flow of almost 60 % at 3 hours. In contrast, seven control studies showed no such effects. Permeability of isolated amnion to sodium was studied over the course of gestation by the use of radioactive sodium ( NaCl). Maternal-fetal sodium movement was found to in -crease by a factor of about 35 fold from day 57 to day 70. The treatment of membranes with vasopressin (500 mU/ml) produced an average decrease in maternal-fetal sodium move-ment of 12.2 + 7.8 # i n the third hour (2 experiments). Thus, water movement i n this direction did not seem to be coupled to sodium flux. When prolactin (10 ug/ml) was added to seven membranes, an increase i n maternal-fetal sodium flux of 21.3 + 8.3 % was recorded i n the third hour. Control membranes, however, showed a similiar increase of 21.3 + 1^.9 % in the third hour (7 experiments). Therefore, com-parison of experimental and control preparations suggests that prolactin probably does not affect maternal-fetal sodium movement. In contrast to this, treatment of five other membranes with prolactin (10 ug/ml) produced an average increase i n fetal-maternal sodium flux of 53*6 + 10.1 % i n the third hour. Since five control membranes showed an increase of only 15.1 ± 15•9 % i n this same iv time period, prolactin seemed to be responsible for producing an overall increase of 3 8 . 5 % in unidirectional flux of sodium. Preliminary experiments indicate that neurohypophyseal hormone can stimulate an increase in unidirectional water flux across other fetal membranes and tissues. Vasopressin (100 mU/ml) increased water flow across the isolated f e t a l urinary bladder in the mucosal-serosal direction by ^ 9 . ^ + 17.7 % at the end of an hour (3 experiments)? a control membrane showed an increase of only 9«7 %• The addition of vasopressin (500-1000 mU/ml) increased serosal-mucosal water movement across isolated fetal skin by up to 3 ° # in 60 minutes (^  experiments). At the present time the effects of vasopressin and prolactin on water and/or sodium movement across isolated amnion, fet a l urinary bladder, and fetal skin must be regarded as pharmacological. It seems probable, however, that some of the responses may be physiological since high levels of hormone are found in fet a l blood (i.e. vasopressin) and amniotic f l u i d (i.e. prolactin). This study suggests that hormonal regulation of fetal hydro-mineral metabolism may explain the enigma of how hydro-osmotic homeostasis i s achieved in the intrauterine compartments-TABLE OF CONTENTS Page LIST OF TABLES x LIST OF FIGURES x i INTRODUCTION 1 1. The intrauterine f l u i d compartments 1 2. Fetal tissues and organs involved in hydro-mineral exchange 2 A. The f e t a l lung and tracheobronchial tract 2 B. The f e t a l salivary glands ^ C. The f e t a l skin surface 5 D. The umbilical cord 8 E. The f e t a l gastrointestinal tract 10 F. The f e t a l kidney 13 G. The f e t a l urinary bladder 16 3. Contribution of f e t a l tissues and organs in regulating amniotic f l u i d volume 17 Comparison of osmoregulatory structures of the f e t a l mammal with those of lower vertebrates 18 5. The f l u i d exchange between intrauterine compartments 19 6. Factors affecting inter-compartmental f l u i d exchange 20 7. The role of hormones i n controlling fe t a l hydro-mineral metabolism 23 STATEMENT OF THE PROBLEM 2^ v v i Page GENERAL METHODS 26 1. Unidirectional water flux experiments 26 A. The isolated amnion preparation 26 (a) Dissection 26 (b) The perfusion ce l l s 26 B. Experimental procedure for studying unidirectional flux of water 30 C. Estimation of unidirectional flux of water 32 2. Net water flux experiments 33 3. Unidirectional sodium flux experiments 34 A. Experimental procedure for studying unidirectional flux of sodium 34 B. Estimation of unidirectional sodium flux 35 4. Hormone preparations 36 A. Vasopressin 36 B. Prolactin 37 5. Salines 38 A. Amniotic saline 38 B. Maternal saline 38 SECTION I 39 THE EFFECT OF VASOPRESSIN ON WATER AND SODIUM MOVEMENT THROUGH THE ISOLATED AMNIOTIC MEMBRANE OF THE GUINEA-PIG 39 INTRODUCTION 39 RESULTS 41 v i i Page 1. Permeability coefficients for the guinea-pig amnion 4-1 2. The action of vasopressin? effects on unidirectional flux of water through the amnion 4-2 3. The action of vasopressin; effects on unidirectional flux of sodium through the amnion • 4-8 DISCUSSION 50 1. The diffusional permeability of the isolated guinea-pig amnion 50 2. The action of vasopressin on water flow through the amnion 53 3. Water-solute coupling 55 4-. Hormones present i n amniotic f l u i d 57 SECTION II 59 THE EFFECT OF PROLACTIN ON WATER AND SODIUM MOVEMENT THROUGH THE ISOLATED AMNIOTIC MEMBRANE OF THE GUINEA-PIG l : ; 59 INTRODUCTION 59 RESULTS 61 1. The action of prolactin? effects on unidirectional fetal-maternal water movement through the amnion 6 l 2. The action of prolactin? effect on net flux of water through the amnion 65 3. The action of prolactin? the comparison of effects on unidirectional and net fetal-maternal water flux through the amnion 70 v i i i Page 4-. The action of prolactin? effect on unidirectional maternal-fetal water flux through the amnion ?4-5. Permeability of the amnion to sodium 77 6. The action of prolactin? effect on unidirectional maternal-fetal sodium flux through the amnion 80 7. The action of prolactin? effect on unidirectional fetal-maternal sodium flux through the amnion 83 DISCUSSION 85 1. The action of prolactin on water flux through the amnion 85 2. The permeability of the amniotic membrane to sodium 88 3. The action of prolactin on sodium flux through the amnion 92 SECTION III 95 PRELIMINARY EXPERIMENTS STUDYING VASOPRESSIN'S EFFECT ON OTHER FETAL TISSUES 95 INTRODUCTION 95 1. The effect of vasopressin on water flux through the isolated f e t a l bladder ....... 96 Introduction 96 Results 96 Discussion 97 2. The effect of vasopressin on water flux through isolated skin of the fet a l guinea-pig 101 ix Page Introduction 101 Results 101 Discussion 102 GENERAL DISCUSSION 106 REFERENCES 117 LIST OF TABLES TABLE Page I. THE EFFECT OF VASOPRESSIN ON MATERNAL-FETAL WATER MOVEMENT THROUGH THE ISOLATED AMNION OF THE GUINEA-PIG 48 II. THE EFFECT OF VASOPRESSIN ON UNIDIRECTIONAL MUCOSAL-SEROSAL WATER MOVEMENT THROUGH THE ISOLATED URINARY BLADDER OF THE FETAL GUINEA-PIG 97 III. THE EFFECT OF VASOPRESSIN ON UNIDIRECTIONAL SEROSAL-MUCOSAL WATER MOVEMENT THROUGH THE ISOLATED SKIN OF THE FETAL GUINEA-PIG 102 x LIST OF FIGURES Figure Page 1. The Perfusion Cell used for studying Radio-isotope Flux through the Isolated Guinea-pig amnion 28 2. The Effect of Different Doses of Vasopressin on Unidirectional Water Flux through the Guinea-Pig Amnion. 4-4 3. The Relationship between the Logarithm of the Dose of Vasopressin added to Amniotic Membranes and the Percentage Increase in - • J. Unidirectional flux of water 4-7 4-. The Effect of Prolactin on Unidirectional Water Flux through the Guinea-Pig Amnion in the Fetal-Maternal Direction °3 5. The Effect of Prolactin on Net Water Flux through the Guinea-Pig Amnion in the , Fetal-Maternal Direction 6 8 6 . The Comparison of the Effects of Prolactin on Net Flux and Unidirectional Flux of Water through the Guinea-Pig Amnion in the Fetal-Maternal Direction 72 7. The Effect of Prolactin on Unidirectional Water Flux through the Guinea-Pig Amnion . in the Maternal-Fetal Direction <b 8. The Permeability of Guinea-Pig Amnion to Sodium at Different Gestational Ages 79 9. The Effect of Prolactin on Percentage Increase in Unidirectional Sodium Flux through Guinea-Pig Amnion ° 2 10?. An Example of the Effect of Vasopressin on Percentage Increase in Unidirectional Water Flux through the Urinary Bladder of the Fetal Guinea-Pig 99 11. Examples of the Effect of Vasopressin on Percentage Increase in Unidirectional Water Flux through Skin of the Fetal Guinea-Pig xi ACKNOWLEDGMENTS I am indebted to the National Research Council of Canada for financial assistance in the form of Postgraduate Scholarships for the years 1973-75. I wish to thank my supervisor, Dr. Anthony M. Perks, for c r i t i c a l advi#e and yery generous financial support of research a c t i v i t i e s . I am grateful to Dr. William S. Hoar and Dr. John E. Phillips for guidance and assistance through-out the course of my studies. I would also l i k e to ac-knowledge the aid and assistance of the following peoplet Ms. Daphne Hards for sectioning and staining the skin taken from fe t a l guinea-pigs; Dr. Harold Nordan, Mr. Armin Tepper, and Mr. Joe Molenda for breeding and care of the guinea-pigs used in this study? Mr. Steve Borden for assistance in designing the computer programmes used to analyze the data, and for s t a t i s t i c a l advise? Mr. Colin Parkinson and Fergus O'Hara for mechanical assistance? Dr. R. Bates, U.S. Public Health Service, National Institute of Health, Bethesda, Md., for the kind g i f t of prolactin? Margaret McDonald, Susan Plath, Harold Bryant, and John Spence for moral support? and fi n a l l y , I would like to thank my wife, Louise, for typing this thesis, and for encouragement of my research endeavors. x i i INTRODUCTION 1. The intrauterine f l u i d compartments In the primate, intrauterine water i s divided into three compartments, viz., the f e t a l , the placental, and the amniotic f l u i d compartments (Kerpel-Pronius, 1970). Out of a total water volume i n the human intrauterine cavity of about 3,533 ml, 2,400 ml are found in the feta l compartment, 7 0 0 ml in the amniotic f l u i d compartment and 433 ml i n the placental compartment at 37 weeks of gestation. In the last trimester of pregnancy i n the human being there are approximately 756 mEq of exchangeable Na + and 171 mEq of exchangeable K + (MacGillivray and Buchan, 1958)* Both water and ions are in a state of dynamic flux between the three compartments. Vosburgh et a l . (1948) showed that human amniotic f l u i d does not form a static and stagnant pool, but one which i s constantly being renewed or replaced. They demonstrated that there i s a rapid exchange of heavy water and various isotopic electrolytes between the amniotic f l u i d and the other intrauterine compartments. Vosburgh et a l . (1948) found that 35 % of the water i n human amni-otic f l u i d i s renewed each hour, and thus, water would be replaced about every 2.9 hours. According to Plentl and Hutchinson (1953) the period of renewal of Na + and K + in human amnioitc f l u i d i s about 8 hours. Hutchinson et a l . 1 2 (1955) calculated that water transfer across human amnion occurs at a rate of about 4-20 ml/hr, while Na + transfer was 12 mEq/hr and K + transfer was 0.6 mEq/hr. A number of fe t a l tissues and organs have been implicated as playing a role in hydro-mineral exchange between the fetal and amniotic compartments. These include a) the fe t a l lung and tracheobronchial tract, b) the fetal salivary glands, c) the fe t a l skin surface, d) the umbilical cord, e) the fe t a l gastrointestinal tract, f) the f e t a l kidney and g) the urinary bladder. Probably only a few of these structures, however, participate i n any large scale exchange of f l u i d between the fetal and the other compartments. 2. Fetal tissues and organs involved i n hydro-mineral  exchange» A. The f e t a l lung and tracheobronchial tract During fetal l i f e the lung can act as an exocrine gland that secretes liquid of f a i r l y large volume in some species of mammals. This liquid i s of relatively constant and unique composition. The flow rate i s quite variable, and the high flow rate seems to be correlated with res-piratory movements in the fetus (Adamson et a l . , 1973)* The importance of the lung as a secretory organ varies considerably amongst species. Fetal lambs are able to secrete tracheal f l u i d at a rate of 50-80 ml per day 3 (Abramovich, 1 9 7 3 ) t while fetal monkeys produce only about 0 . 2 3 ml per day (Reynolds et a l . , 1 9 7 1 ) . Fetal tracheal f l u i d has an osmolality and sodium concentration approx-imately equal to fet a l plasma while i t s pH and pCC-2 are lower and i t s chloride concentration i s 40-50 mEq higher in f e t a l lambs near term (Adamson et a l . , 1 9 6 9 ) . Olver et a l . (1973) found evidence that the liquid of the feta l lung i s secreted by a process involving an active transport of CI- and H C O 3 " by the alveolar epithelium of the lamb. CI" seems to be transported from fet a l plasma in exchange for H C O 3 " . The similarity between alveolar epithelium and gastric mucosa with regards to active transport of Cl~ and HCO^" may be a reflection of the common origin of the two epithelia from the embryonic foregut. Gluck et a l . ( 1 9 7 1 ) demonstrated that tracheobronchial f l u i d i s secreted into the amniotic f l u i d compartment. They found that an active lecithin present in amniotic f l u i d was similiar i n fatty acid composition to lecithin found in tracheal effluents, but different from lecithins found in fetal membranes, fet a l urine, maternal blood, and the placenta of the human being. Alveolar or tracheal secretion probably does not contribute significantly to amniotic f l u i d volume in the human being (Bourne, 1962$ Jeffcoate and Scott, 1 9 5 9 ) . In other species of mammals such as the guinea-pig and goat, the fetal lung may make a 4-f a i r l y large contribution to amniotic f l u i d volume i n the last 1/3 of pregnancy (Adams et a l . , I 9 6 7 ) . The fact that the lungs of many feta l mammals secrete relatively small volumes of f l u i d means that respiratory secretions are probably unimportant i n osmoregulation or i n maintaining acid-base balance. Goodlin and Perry ( I 9 6 6 ) feel that in the fe t a l rabbit, however, the lung can contribute to osmoregulation. B. The fe t a l salivary glands Arthur ( 1 9 6 9 ) found that ungulates actively secrete saliva into the amniotic sac. In the earlier stages of gestation i n f e t a l calves, urine passes out of the urinary bladder by two outlets, one going to the allantois and the other going to the amniotic cavity. At about midterm a sphincter in the f e t a l bladder closes so that urine can no longer pass to the amniotic compartment. Instead i t goes along the urachus into the allantoic sac. As gestation proceeds towards term allantoic f l u i d becomes more urine-l i k e , while amniotic f l u i d becomes a colorless, glairy, liquid strongly resembling saliva. Lind and Hytten ( 1 9 7 2 ) discovered that the human fetus may also contribute s a l i -vary gland secretion to the amniotic compartment since amylase was found in amniotic f l u i d . The significance of salivary gland secretion into the 5 amniotic sac i s not known, but i t may be involved in some form of f l u i d or ion exchange. In adult mammals acetyl-choline induces water and ion transport in the salivary glands by activating a NaCl pump at the luminal membrane or by increasing the calcium influx. There also seems to be a pump that extrudes Na and accumulates K (Peterson, 1971). Although i n some mammals, such as the ungulates, saliva can make a large contribution to amniotic f l u i d volume, i t probably does not play a major role in hydro-mineral transfer i n other f e t a l mammals. C. The fet a l skin surface The skin serves as an outer covering, a storage organ, a regulator of body temperature, a sense organ, and a secretory organ in the adult mammal. Fetal skin differs from adult skin in that transitional and cornified layers are thinner (Gleiss, 1970). Prior to keratinization fe t a l skin i s very .permeable to water. The fact that permeability coefficients for f e t a l skin (0.72x10 cm/sec), amnion (2.88xl0"^cm/sec), and chorion leave (1.3xl0"^cm/sec) in the human being are similar suggest that the skin may serve as a significant exchange site between fetal fluids and amniotic f l u i d (Lloyd et a l . , 1969; Seeds, 1972 b). At 12 to 16 weeks of gestation i n the human fetus (term 6 approx. 44 weeks) the periderm c e l l s contain an abundance of m i c r o v i l l i , Golgi apparatus, mitochondria, endoplasmic reticulum, globular protrusions, and membrane-bound vesicles containing mucopolysaccharides and glycogen (Breathnach and Wyllie, 1965; Hoyes, 1967; 1968 a). The ultra-structural evidence would tend to suggest that fe t a l skin i s capable of either secretion or absorption. In fact, Lind and Hytten (1970) describe the periderm cells of the human fetus as having a strong resemblance to renal tubular ce l l s under the influence of vasopressin (see Ganote et a l . 1968). Lind et a l . (1972) showed that during the f i r s t half of pregnancy, Na + as well as water can readily pass through f e t a l skin. Although no potential difference could be measured in vivo or i n vitro there could be a water solute coupling which causes movement of Na + and water across a membrane without generation of a potential difference (see Diamond, 1971). Prior to midterm i n the human fetus, the fetal skin seems to be capable of altering amniotic f l u i d composition and volume (Lind and Hytten, 1970; 1972; Lind, 1973)» At this stage of gestation amniotic f l u i d appears to be a dialysate of fetal plasma, and might be considered merely an extension of the fe t a l extracellular f l u i d space (Seeds, 1972 a; Lind, 1973). The only barrier between amniotic f l u i d and the fetal extracellular f l u i d space i s the skin, 7 and prior to midterm i t may function as a semipermeable membrane. Since the volume of amniotic f l u i d varies directly with the surface area of the fetus, and since the fetal ectoderm and the amnion are derived from the same germ layer there i s good indication that f e t a l skin may alter the volume, and perhaps composition, of amniotic f l u i d (Harrison and Malpas, 1953? Saunders and Rhodes, 1973). Changes in the composition of liquor amnii have been correlated with changes i n fet a l skin (Lind et a l . , 1969). At 17-20 weeks of gestation cornification begins, and there i s a reduction i n the number of micr o v i l l i and an increase in the number of skin layers in the human fetus. Skin becomes more closely bound at the desmosomes restricting the passage of water and solutes. At 25 weeks human feta l skin i s relatively impermeable to water and solutes due to the presence of a stratum corneum (Parmley and Seeds, 1970). The fact that liquor amnii becomes hypotonic to fetal and maternal plasma suggests that amniotic f l u i d i s no longer an u l t r a f i l t r a t e of plasma (Lind et a l . , 1969). When skin becomes keratinized in the second half of pregnancy the amniotic f l u i d becomes exteriorized because water and solutes can no longer pass between body fluids of the fetus and the amniotic f l u i d (Lind, 1973)-Prior to keratinization, the skin of the mammalian 8 fetus may serve as an important osmoregulatory structure. In large measure i t controls the exchange of water and ions between the extracellular f l u i d space of the fetus and the amniotic f l u i d surrounding i t . D. The umbilical cord Since the lining of the umbilical cord i s derived from the same germ layer as the fetal ectoderm and amnion i t might be expected to show similiar characteristics. The cord epithelium consists of a single layer of cells with l i t t l e differentiation between 8 and 10 weeks of gestation in the human being (Hoyes, 1 9 6 9 ) . At the end of the third month the ce l l s undergo differentiation, and the epithelium becomes bilaminar. The c e l l s at this stage contain endoplasmic reticulum, multivescular bodies, cytoplasmic vesicles, glycogen deposits, Golgi apparatus, villous folds and m i c r o v i l l i . These cells are similiar in appear-ance to periderm cells of fe t a l skin, and likewise, are thought to play a role in water exchange. The functional activity of f e t a l periderm declines steadily i n the process of keratinization and the formation of the stratum corneum (Hoyes, 1968 a). Keratinization does not occur in cord epithelium, however, execpt in a region in close proximity to the fetus; the remainder consists of simple squamous epithelium (Hoyes, 1 9 6 9 ) . 9 Although f e t a l skin readily absorbs isotopic water in early gestation, there i s a decrease i n water transfer as the skin begins to keratinize. Prior to 18 weeks of gestation the human umbilical cord reabsorbs l i t t l e or no water in vivo or in vitro, but as the skin begins to keratinize, the umbilical cord begins to play an increas-ingly important role in reabsorption of water. Between 3 and 59 ml per hour may be exchanged across the human umbilical cord at term (Plentl, 196 l j Abramovich, 1973). Between 22 and 26 weeks of gestation zonulae occludentes begin to disappear, villous folds develop i n the intercellular spaces, and cytoplasmic vesicles appear in the basal c e l l s of cord epithelium (Hoyes, 1969). These morphological changes occuring in cord epithelial c e l l s between 22 and 26 weeks of gestation i n the human fetus may explain the observation that the umbilical cord begins to play a significant role i n water transfer when the skin i s becoming keratinized (i.e. in weeks 17-25 of human ges-tation). Hutchinson et a l . (1959) injected isotopic water into the human amniotic sac, and found that large amounts of the tritium isotope were concentrated i n Wharton's j e l l y of the umbilical cord. Gadd (1966) provided further evidence that the umbilical cord plays a role in water exchange between the f e t a l and amniotic compartments. He injected 10 sulfonamides into the amniotic sac. He then measured the concentration in the fet a l membranes and umbilical cord, and found that the sulfonamides were concentrated to a greater extent i n the umbilical cord. Although the umbil-i c a l cord has been postulated as a site of exchange between f e t a l circulation and amniotic f l u i d i t probably does not represent a major site. The surface area of the umbilical vessels i s small compared to a capillary bed (Seeds, 1972 b). Since the surface area of the cord epithelium i s small compared to that of fetal skin, the cord i s not able to completely compensate for the loss of skin as an exchange site when keratinization occurs. This i s reflected by the fact that after midterm amniotic f l u i d i s no longer an ultrafiltra.te of fet a l plasma. E. The fetal gastrointestinal tract Evidence for fet a l swallowing near term had been provided in the 19th century by the observation that epithelial c e l l s , lanugo hairs, and vernix caesoa had been found in the gut of fetuses (Minot, 1892). In the early twentieth century, Wislocki (1920) discovered that when trypan blue was injected into the amniotic cavity of fe t a l guinea-pigs and cats that the colloidal dye was found i n the gastrointestinal and respiratory tracts. Jeffcoate 11 and Scott ( 1 9 5 9 ) and McLain ( 1 9 6 3 ) provided evidence that not only does the human fetus swallow amniotic f l u i d , but that i t can also absorb f l u i d across the wall of the stomach. They introduced a radio-opaque dye into the amniotic compartment, and X-rayed the fetus i n utero a number of hours later; dye was found to be concentrated in the stomach. McLain ( 1 9 6 3 ) also observed that although i t took the dye 8 hours to reach the colon at 3 4 weeks of gestation, by 37 to 40 weeks only 4 / 2 to 7 hours were required for passage of the dye to the colon. The degree of water reabsorption by the gastro-intestinal tract seems to vary greatly amongst species. Wright and Nixon ( I 9 6 I ) calculated that the gastro-intestinal tract of the fet a l lamb may absorb as much as 3 2 l i t e r s of amniotic f l u i d between 80 days of gestation and term (approx. 145 days). Pritchard ( 1 9 6 5 ) injected chromium-tagged maternal erythrocytes into the amniotic cavity and estimated that the gastrointestinal tract of the human fetus may absorb as much as 2 - 1 0 ml of f l u i d per hour. According to Abramovich (1970), however, the human fetus swallows only about 2 ml of amniotic f l u i d per day at 17 weeks of gestation and 13 ml per day at 2 0 weeks. Wright and Nixon U 9 6 I ) demonstrated that the fet a l lamb can absorb the Na + from swallowed amniotic f l u i d via the intestional tract. Mellor ( 1 9 6 9 ) found that the potential difference across the stomach of the fet a l lamb 12 seems to affect the amniotic f l u i d potential difference. This would suggest that an active ion pump may exist in the fe t a l stomach. In late gestation the stomach of the human fetus may contribute hydrogen ions i n the form of HC1 to amniotic f l u i d (Lind, 1973). The concentration of hydrogen ions i n human amniotic f l u i d increases from 59.34- + 1.25 nEq/l at 10-23 weeks of gestation to a con-centration of 78.4-5 + 2.68 nEq/l at term (Seeds and Hel-legers, 1968). The importance of deglutition in altering amniotic f l u i d composition and volume has not been well established. Abramovich (1970) has provided information gained in a study of 8 anencephalic fetuses and 1 microcephalic fetus that swallowing plays a very minor role i n the polyhydram-nios of anencephaly. Fetal disorders such as diabetes mellitus may be present in the anencephalic and exert a greater influence on amniotic f l u i d volume than swallowing (Abramovich, 1973). When the amniotic f l u i d volume l i e s within the normal range of values, however, swallowing of f l u i d and reabsorption by the gastrointestinal tract can play a major role in regulating f l u i d volume. Pritchard (1966) calculated that when the amniotic f l u i d volume was 84-9 ml that the human fetus swallowed about 4-53 ml of the fl u i d per day. 13 P. The f e t a l kidney Hippocrates was responsible for the oldest and the most simple explanation for the source of amniotic f l u i d ; he suggested that i t was a product of fe t a l micturition. Abramovich ( I 9 6 8 ) provided evidence that the kidneys of the human fetus may be functional at 11 weeks of gestation since urine was found in the bladder of fetuses of this age. MacDonald and Emery ( 1 9 5 9 ) suggested that at 18 to 2 2 weeks of gestation, 8 0 % of the glomeruli i n the human fetal kidney appear to be functional based on histological features. Abramovich ( 1 9 7 0 ) estimated that fetuses of this gestational age void between 7 - 1 1 ml of urine per day, based on a bladder capacity of 0 . 3 - 0 . 7 ml. The kidney glomeruli and tubules appear to be functional at 6 0 days of gestation in the fe t a l lamb (Alexander and Nixon, 1 9 6 1 ) . Petal urine i s hypotonic to plasma, and the hypo-tonicity increases as gestation advances i n animals like the lamb. The osmolarity of fe t a l lamb urine at 81 to 9 3 days i s 2 3 9 mOsm/l, and f a l l s to an osmolarity of 166 mOsm/l at 1 3 0 to 142 days (Alexander and Nixon, I 9 6 I ) . Na + and CI" reabsorption from the glomular f i l t r a t e presented to the renal tubules of fe t a l lambs increases from 6 0 % at 61 days to 9 0 % at term. K + decreased i n a similar fashion from 61 days to term. Carpek et a l . (1968) showed by u t i l i z a t i o n of Ik micropuncture techniques that the proximal renal tubule of the 30 day old rat reabsorbs more Na + than the 20 day old rat post parturn. Although water reabsorption by the kidneys of the fe t a l lamb amounts to 92 % at lk2 days, only 28 % of the water in the glomerular f i l t r a t e was reabsorbed by the tubules at 61 days (Alexander and Nixon, 1961). The hypotonicity of fe t a l urine seems to be the result of reabsorption of electrolytes in excess of water at certain stages of gestation in some mammals (Kleinman, 1970). The hypotonicity of fetal urine may also be attrib-uted to a low solute load rather than an increased volume of water present i n the tubules since some of the solute may be excreted by the human fetus via the placenta (Seeds, I965). Edelman and Barnett (i960) theorized that the poor concentrating a b i l i t y of the fe t a l or neonatal kidney i s due to an unsatisfactory gradient of urea between the cortex and the papilla of the counter-current multiplier system. Since the human fetus or infant has a high anabolic rate i t ex-cretes less nitrogen, and thus, the medullary urea concen-tration i s lower than in the adult (Kleinman, 1970). Edel-man (1967) demonstrated that i f the protein intake of the human infant's diet i s increased so that there i s more urea in the medulla, that the urine osmolarity can be greatly increased. The fact that the fe t a l kidney i s functional in utero 15 suggests that i t could alter amniotic f l u i d volume. Seeds (1972 a) provides the following evidence that the kidney of the human fetus makes a significant contribution to the amniotic compartment in the last half of gestations 1. From the 20th week un t i l term in the human being the total solute concentration in amniotic f l u i d decreases. 2. Fetal urine i n utero decreases in osmolality as gestation advances to a value 80-140 mOsm/l, lower than maternal and fetal plasma at term. 3. Creatinine increases i n both fet a l urine and amniotic f l u i d to values higher than i n maternal or fetal blood. 4. Creatinine and urea can pass through the placental membranes. 5« Kerr et a l . (19&9) found that concentration of taurine i n amniotic f l u i d of the rhesus monkey, increases in the later stages of gestation, and i s high in fetal urine. 6. Amniotic f l u i d becomes increas-ingly acidic from the middle of gestation to term, and fetal urine has been found to have a low pH at term. Klopper (1972) provided further evidence that urine i s passed into the amniotic cavity. He showed that uncon-jugated glucosiduronate, and sulfate fractions of e s t r i o l in fetal urine and amniotic f l u i d were of very similiar levels. Indication that fetal micturition can influence amniotic f l u i d volume was provided by Alexander et a l . (1958) 16 who measured urine flow in fetal lambs. They found that urine flow was about 0.14- ml/min at 61 days, 0.64- ml/min at 117 days, and 0.14- ml/min at term. These flow rates approximate the changes i n amniotic f l u i d volume at these stages of gestation. A number of workers have sited cases of bilateral renal agenesis, cystic dysplasia, and congenital urethral atresia being associated with oligo-hydramnios as evidence that fe t a l micturition can play a major role in controlling amniotic f l u i d volume (Blain and Scott, I 9 6 0 ; Lind et a l . , 197D. Abramovich (1973) cautions against drawing a parallel between fet a l disorders such as renal agenesis and oligohydramnios because the two are not always linked, and sometimes the presence of other fe t a l disorders with or without renal agenesis can lead to a decrease in amniotic f l u i d volume. When amniotic f l u i d volume strays from the expected value for a certain gestational age, as in oligohydramnios, the role of the fetal kidney in controlling amniotic f l u i d volume becomes ambiguous. G. The fe t a l urinary bladder Although kidney urine of the f e t a l guinea-pig i s hypotonic to fe t a l plasma, bladder urine i s slightly hypertonic (Kleinman, 1970). Osmolarity of bladder urine 17 of the fetal guinea-pig i s 368 mOsm/l at one stage of gestation while fe t a l guinea-pig plasma has an osmolarity of 292 mOsm/l. The hypertonicity of bladder urine i n the fetal guinea-pig i s suggestive of water reabsorption by the bladder. Stanier (1971) found that bladder urine of fe t a l pigs i s hypotonic to the urine in the renal pelvis. Pelvic urine had an osmolarity of 120-250 mOsm/l. Bladder urine had 20 % to 50 % the osmolarity of pelvic urine early i n gestation and increased to over 50 % at 100 days of ges-tation. Bladder Na + concentration was found to be 15 mEq/l while Na + concentration i n the renal pelvis was 53 mEq/l. These observations seemed indicative of ion reabsorption by the fe t a l bladder. France et a l . (1972) carried out 22 + 24 + radioisotope experiments using Na and Na in order to measure ion fluxes acrosB the urinary bladder of the feta l pig in vitro. There was a net efflux of Na (i.e. reabsorp-tion), and since the ratio of efflux for t r i t i a t e d water was different from that of sodium, water movement did not seemed to be coupled to sodium movement. 3. Contribution of f e t a l tissues and organs i n regulating amniotic f l u i d volume Abramovich (1973) estimated that at 17 to 19 weeks of human gestation that amniotic f l u i d volume i s increasing 18 at a rate of 11-13 nil per day. Since the fetus swallows about 4-11 ml per day, there must be 11-24- ml of f l u i d added to the amniotic compartment per day in order to account for the volume of amniotic f l u i d present. The kidneys have been estimated to add only 7-14- ml of urine per day to the amniotic sac at this stage of gestation. Thus, 4--10 ml of f l u i d must come from other sources, such as the lungs, the salivary glands, the skin, the umbilical cord, the fetal membranes, or the f e t a l side of the placenta. 4-. Comparison of osmoregulatory structures of the fe t a l  mammal with those of lower vertebrates The mammalian fetus i s comparable to teleost fishes in i t s practice of swallowing f l u i d from the external medium, and reabsorbing water and ions across the gastro-intestinal tract (see Utida et a l . , 1972). I t also seems to resemble anuran amphibians with regards to natriferic and hydro-osmotic interactions of the skin surface and the urinary bladder with the f l u i d environment (see Bentley, 1966| Maetz, 1968; and Turner and Bagnara, 1971). The similarity i n function of various f e t a l structures to osmo-regulatory organs of lower vertebrates may represent an ex-ample of ontogeny recapitulating phylogeny. More probably, however, this may be a case of parallel evolution in which organisms develop similar adaptations i n response to similar problems. 19 5. The f l u i d exchange between intrauterine compartments We have determined the relative contribution various fetal structures make to hydro-mineral exchange between the fetal and other f l u i d compartments i n the intrauterine cavity. We have also considered the influence of various fetal tissues and organs in controlling amniotic f l u i d volume and composition. We can now examine the interplay and exchange between the three compartments, i n order to determine how intrauterine fluids accumulate and are dis-posed of. The amniotic sac becomes the major f l u i d space sur-rounding the human fetus by the end of the ninth week of gestation. The chorionic and yolk sac are almost gone, while an allantoic sac never grows to a significant degree in the human being (Seeds, 1972 a). According to the data of E l l i o t and Inman (1961), amniotic f l u i d volume i s about 5-10 ml at 8 weeks of gestation, ri s i n g to 1000 ml at 38 weeks, and dropping to 244 ml at 4-3 weeks i n the human being. The significance of the changes in f l u i d volume i s s t i l l obscure. Since water increases at a rate of 30-4-0 ml per day in the human intrauterine cavity near term, water must be transferred from the maternal compartment. Intrauterine or f e t a l metabolic sources would be too small to contribute 20 significantly to intrauterine water volume (Seeds, 1973). Exchange rates between mother and fetus increase between mid pregnancy and term, while exchange rates between mother and amniotic f l u i d , and between fetus and amniotic f l u i d increase at a slower rate as gestation proceeds. Na + and K were found to exchange more slowly than water (Kerpel-Fronius, 1970). Schuefer et a l . (1972) provided evidence that there i s a dynamic exchange occuring between the amniotic f l u i d compartment,^and other compartments. When they replaced amniotic f l u i d with solute-free water in the rhesus monkey, there was a reduction in f l u i d volume of the amniotic e !avity and the appearance of a large quantity of solutes. By the end of 20 hours the solute concentration in the amniotic sac had returned to pre-experimental levels. How solute and water exchange occurs between the amniotic and other compartments i s not well understood. 6. Factors affecting inter-compartmental f l u i d exchange Battaglia et a l . (I960) showed that one could experi-mentally stimulate water or sodium transfer between the mother and the fetus, or between the fetus and mother across the placenta. They accomplished such a transfer by infusing the mother with mannitol, hypertonic saline, isotonic 21 glucose or 5 f° glucose. Infusion of the mother with a hypertonic saline solution was found to cause a 2 fo to 4 % loss in total body water of the fe t a l rabbit, and an 18 % decrease i n extracellular f l u i d . Infusion of pregnant rhesus monkeys with hypertonic saline or disaccharide solutions caused a 4 % decrease i n fet a l body water, as did infusion of the amniotic cavity with a disaccharide solution (Seeds et a l . 1 9 6 4 ) . Bruns et a l . ( 1 9 6 3 ) showed that experimental dehydration of the feta l rabbit caused a decrease in amniotic f l u i d volume. Water has also been found to move in bulk flow across the perfused guinea-pig placenta in vivo as a linear function of hydrostatic pressure (Dancis et a l . I 9 6 2 ) . The fact that water transfer can occur i n bulk flow in response to experimental osmotic and hydrostatic gradients between the three compartments suggests that water could be supplied for fetal development by in vivo chemical and physical gradients. The osmolarity of fetal plasma ( 2 9 2 . 7 mOsm/l) and maternal plasma ( 2 8 9 . 1 mOsm/l) are approximately equal in the human being (Kerpel-Fronius, 1970). In early stages of gestation amniotic f l u i d i s essentially isotonic to fet a l plasma. Thus, there would appear to be no osmotic gradients present in utero capable of driving water in bulk form from one compartment to another in early stages of gestation. 22 Net flux of water can be driven by hydrostatic as well as osmotic pressure gradients. Reynolds ( i960) determined that umbilical hydrostatic pressures did not indicate that water could be driven from mother to fetus in the sheep. Ramsay et a l . (1959) measured hydrostatic pressures in the human being by intervillous space sampling, and could find no suggestion that gradients existed that were large enough to drive f l u i d from mother to fetus. Existing hydrostatic gradients would, i f anything, cause water movement from fetus to mother. Thus osmotic and hydrostatic gradients do not explain the accumulation of intrauterine water. Although there i s no gradient in total solute concen-tration a difference in colloid osmotic gradients could cause a net flux of water from the mother to the fetus (Barnes, 1968). Meschia (1955) found a higher colloid osmotic pressure in the maternal plasma of sheep and goats than in the fetal plasma during the last half of pregnancy. Colloid osmotic pressure or plasma protein concentration of fe t a l blood increases throughout gestation from 2.5 i» to 6 % at term in the human fetus. These values are s t i l l lower than maternal values, however, so water movement would be favored in the fetal to maternal direction (Westin, 1959). Therefore existing osmotic, hydrostatic, or colloid osmotic gradients between mother and fetus do not explain how water could accumulate in the intrauterine cavity. There i s also very l i t t l e information available on how 23 hydro-mineral balance i s achieved in the intrauterine cavity, and what controls the rate of water and ion exchange between the three compartments. 7. The role of hormones i n controlling f e t a l hydro-mineral  metabolism In the adult mammal many forms of metabolism, including those involved with hydro-mineral balance, are controlled by hormones. The role of hormones in controlling f e t a l metabolism, however, i s not well understood. It i s only recently that any detailed investigation has been initiated to determine hormonal function i n the fetus. Comline et a l . (1970) have suggested that the fetus may control the level of ions i n allantoic f l u i d by fe t a l adrenal gland secretion of corticosteroids. Basset and Hinks (1969) found that reduction of glucose i n the plasma of the 6 week-old neonatal lamb stimulated the adrenal gland to secrete corticosteroids. An increase in glucose concent„••-tration in maternal plasma leads to an increase in sodium concentration in allantoic f l u i d of the fe t a l lamb (Mellor and Slater, 1971). A decrease i n plasma glucose levels leads to a decrease i n sodium concentration and an increase in potassium concentration in allantoic f l u i d . Since Mellor and Slater (1971) suggest that solute movement in and out of the allantoic sac i s associated with f l u i d volume, 24 corticosteroids may influence water exchange between the allantoic sac and other f l u i d compartments in ungulates. Alexander and Williams (1968), however, found evidence in favor of control of allantoic f l u i d volume by maternal secretion of sex steroids. They observed that allantoic f l u i d accumulated i n ovariectomized sheep maintained on low doses of intramuscularly administered progesterone. Increasing the amount of progesterone injected or injection of oestradiol benzoate reduced allantoic f l u i d volume to normal. Amniotic f l u i d volume was not altered by maternal ovariectomy. Thus, there i s some indication that f e t a l or maternal hormones may directly or indirectly influence hydro-mineral exchange in some species of fe t a l mammals. STATEMENT OF THE PROBLEM The present investigation was undertaken in order to determine whether or not pituitary hormones could be implicated i n influencing hydro-mineral exchange i n the fe t a l guinea-pig. Emphasis was placed on determining whether or not a hormone could affect water and/or sodium movement across fe t a l membranes and tissues. Mechanisms of hormone action were not investigated. In order to avoid compli-cations from possible interactions of hormones with other metabolites, an in vitro preparation was used. Membranes 2 5 and tissues were taken from f e t a l guinea-pigs at various stages of gestation. Thus, membrane sensitivity to hormone could be studied as a function of gestional age. This also allowed one to determine whether or not membrane permea-b i l i t y changed over the course of gestation. GENERAL METHODS 1. Unidirectional water flux experiments A. The isolated amnion preparation (a) Dissection* Pregnant guinea-pigs, between midterm and term, were anaesthesised with ether, and the uterine horns, containing fetuses, were removed and placed in maternal saline at 37°C (see page 38 ) . The uterine wall and yolk sac were removed, and an incision was made in the amnion; a viton o-ring (0.3 cm., internal diameter) was slipped under the membrane, and the section of the amnion adhering to the ring was freed form the rest of the membrane. The ring, with i t s membrane attached, was inserted into a perfusion c e l l . Care was taken to prevent any drying of the membrane, and exposure to air was minimal. Although approximate ages were known, the accepted gestational age of the membrane was determined according to the data of Draper (1920) and Ibsen (1928). Whenever possible, two preparations from the same amnion were set up side by side, i n identical c e l l s , one to act as the experimental preparation, the other as control. (b) The perfusion cells 1 Fig. 1 shows the perfusion cells used in some of the later studies with vasopressin and a l l of the studies with 26 27 Figure 1. The Perfusion Cell used for studying Radio-isotope Flux through the Isolated Guinea-Pig Amnion Stippling represents.the concentration of radio-isotope. The "hot" reservoir receives the i n i t i a l injection of radioisotope; the "cold" reservoir con-tains saline with no radioisotope. The arrows show the direction of flow of saline, with both circuits driven at the same flow rate by a common pump. Radioisotope which passes through the amniotic membrane i s collected in the fraction collector. ' COLD P E R F U S I O N C E L L prolactin. The perfusion c e l l employed in most of the early experiments with vasopressin differed from the setup shown i n Fig. 1 in that the inflow and the out-flow tubings were directed perpendicular to one another. The setup was modified to improve the circulation of f l u i d , and to eliminate dead space and unstirred layers i n front of the isolated membrane. The membranes were supported between two viton O-rings (0.3 cm inner diameter); these were clamped between the two halves of the plexiglass cells (2.5 cm outside diameter, 5.2 cm overall length), which were held together l i g h t l y but firmly, in a vise. The membrane formed the central division of a spherical chamber, 0,3 cm i n diameter, of approximately 20 u l capacity for the perfusion c e l l in Fig. 1. The entire assembly was immersed in a constant temperature bath at o 3 7 + 0 . 0 1 C (Model CTC circulator, Bronwill, Rochester, N.Y.). When the isolated membranes were in place, they were perfused on one side with maternal saline (see page o 38) and on the other side with amniotic saline at 37 C. 22 + In experiments with Na , and in a l l experiments with prolactin, the membranes were perfused on both sides with o . amniotic saline at 37 C. An identical rate of flow (12 ml/hr) was maintained on both sides of the membrane by a Buchler Dekastaltic pump (Buchler Instruments, Fort Lee, 30 N.J.), which was common to both flows. The circulation on one side of the isolated amniotic membrane of the f e t a l guinea-pig was a closed, recirculating system, suitable for the i n i t i a l radioisotope containing saline; i t included . o a 6 ml or 12 ml reservoir (3? C) to which hormone prepara-tions could be added, and this was aerated to ensure both oxygenation and mixing. The circulation on the opposite side of the isolated amniotic membrane was open; saline, free of radioisotope was contained in an aerated 25 ml reservoir at 3 7 ° C (for the vasopressin studies) and aerated 100 ml reservoir at 37° C (for the prolactin studies); i t was pumped past the amnion, where i t received any radio-isotope which permeated the membrane, and the out-flow was collected in either 1.0 ml or 2.0 ml samples in a fraction collector (L.K.B. UltroRac, Type 7000; Stockholm, Sweden). The samples were checked for any variations in perfusion rate, and then measured for radioisotope content. B. Experimental procedure for studying unidirectional  flux of water 100 uCi of t r i t i a t e d water (% 20; New England Nuclear, Boston, Mass.) was diluted into 12.0 ml of maternal saline, and added to the reservoir of the closed-circuit circulation. Amniotic saline, free of radioisotope, was placed in the 31 open-circuit circulation. The membranes were inserted, and the salines allowed to flow for a 60 minute e q u i l i b r i -ation, during which time effluent samples were collected i n the fraction collector. At 60 minutes, vasopressin was added to the reservoir of the open circulation, to give a f i n a l concentration of 50-500 mU/ml; normally i t circulated over the f e t a l side of the amnion. The experiment continued for up to 35 minutes after the addition of vasopressin; during this time samples were collected i n 1 ml amounts from the f e t a l side of the amnion and estimated for tritium content. Experiments dealing with the effects of prolactin on unidirectional flux of water differed in several respects from the procedure outlined above for vasopressini F i r s t l y , t r i t i a t e d water (100 uCi) was diluted in amniotic rather than maternal saline since both sides of the membrane were bathed with amniotic saline in the prolactin studies. Secondly, the salines were allowed to flow for a 90 minute, rather than a 60 minute, equilibriation period. At 90 minutes, prolactin was added to the reservoir to give a f i n a l concentration of either 10 or 20 ug/ml; normally i t circulated over the f e t a l side of the amnion. Thirdly, unlike vasopressin studies which lasted 35 minutes, the prolactin experiments continued for up to 3 hours after 32 addition of hormone. During this time samples were collected in 2 ml aliquots from either the maternal or the fe t a l side of the amnion, and estimated for concen-tration of t r i t i a t e d water. C. Estimation of unidirectional flux of-water Unidirectional flux was obtained by measuring t r i t i a t e d water concentrations on the open ci r c u i t (normally fetal) side of the membrane: the measurements were made on the samples collected in the fraction collector. 200 u l aliquots were removed from each centrifuge tube and placed in s c i n t i l l a t i o n vials containing 10.0 ml of "Aquasol" (New England Nuclear Corp.). The vials were kept in the dark overnight, and counted to over 10,000 counts in a liquid s c i n t i l l a t i o n counter (Isocap 300? Nuclear Chicago Corp.). The channels ratio method was used for quench correction, in order to express data i n d.p.m.. Data was analyzed by a PDP-11/45 computor (Digital Equipment Corp., Maynard, Mass.). The permeability coefficient ( K^ r a n g) f o r "^H2° P a s s i n S across the amnion was calculated by computor from the Pick equation (see Hays and Leaf, 1962)1 rr _ QW *trans ~ TcpcJH Where t Qw = the net increase in isotopic water per unit time on the side opposite to which radioisotope was i n i t i a l l y added. 33 C* and C2= the mean concentrations of radioisotope on the two sides of the membrane during the period of measurement. A= the cross-sectional area of the chamber (0.0? em ). The unidirectional flux was calculated by multiplying Ktrans b y m o l a r concentration of water i n the saline (55*2 mM/cm^ ) and the parti a l molal volume of water (18 uL/mM). Values for unidirectional flux were calculated as ul/cm /hr. 2. Net water flux experiments Net flux of water through amnion was determined by the weight-change method of Vizsolyi and Perks (197*0. The original method was modified i n two waysi f i r s t l y , weight changes were determined over longer, one hour time intervals since prolactin i s often slow acting; secondly, the long-term weight changes were determined by direct weighing on a Mettler balance. The amnion, attached across a glass supporting-tube (0.9 cm diameter; 9.5 cm long), was maintained at 37°C, with amniotic saline on the inner, fe t a l side, and with aerated maternal saline on the outer maternal surface, as described by Vizsolyi and Perks (1974). (For salines, see page 38 ). After one hour of equilibration, the preparation was weighed; care was taken to drain excess f l u i d , and to 34 weigh rapidly. The preparation was replaced, in fresh maternal saline, and the inner amniotic saline was substituted by a new sample which contained prolactin at 20 ug/ml. Weight changes were determined at hourly intervals for three hours. A l l values for net flux were expressed as ul/cm /hr, for comparison with the unidirec-tional measurements. A control preparation, whenever possible from the same fetus, was set up alongside the experimental membrane? i t received equivalent treatment, but the pro-la c t i n was replaced by saline of the same volume subjected to the pH treatment, or serum albumen, at the same concen-tration as the prolactin, and adjusted in the same manner. 3. Unidirectional sodium flux experiments A. Experimental procedure for studying unidirectional  flux of sodium 50 uCi of radioactive sodium ( NaCl? Amersham/Searle Corp., Arlington Heights, I l l i n o i s ) was diluted into 6 . 0 ml of amniotic saline, and was added to the reservoir of the closed-circuit circulation. Amniotic saline, free of radioisotope, was placed in the open circulation. The membranes were inserted, and the salines allowed to flow for a 90 minute equilibriation period, during which time 35 samples were collected i n the fraction collector. At 90 minutes, hormone was added to the saline reservoir bathing the fe t a l side of the amnion to give a f i n a l concentration of either 500 mU/ml vasopressin or 10 ug/ml prolactin. The experiments continued for up to 3 hours after addition of hormone. During this time samples were col-lected i n 2 ml amounts from the f e t a l side of the amnion 22 + and estimated for Na content. B. Estimation of unidirectional sodium flux Unidirectional flux of sodium was determined by 22 + measuring the concentration of Na i n samples collected i n the fraction collector. 1.8 ml aliquots were removed from eaeh centrifuge tube and dispenced into planchets. Saline containing radioisotope was evaporated to a solid counting source i n planchets by means of an infrared lamp. Samples were counted to 10,000 counts for at least two cycles in a gas flow Geiger counter (Model 104-2} Nuclear Chicago Corp.). Sample counts did not exceed 5000 cpm since coincidence loss i n the Geiger-Mueller detector becomes significant over this value (see Wang and W i l l i s , 1965). 22 + The rate of Na movement across the amnion was calculated by means of the following equation (see Snell, Shulman, Spencer and Moos, 1965)« 36 V 9 dC. /dt J 2 i 1 2 A C? /C. Wheret J. = the rate of isotopic flux across the membrane 12 from compartment 1 to compartment 2 of the perfusion c e l l . V ? = the volume of solution 2 i n ml (i.e. that collected i n the fraction collector). dC. /dt = the time of change i n radioisotope concen-2 tration (i.e. the rate of appearance of 2 2Na + i n Cpm into solution 2 ) . A = the cross sectional area of the chamber (0.07 cm2). c i / C i = " f c h e s P e c i f i c activity of 2 2Na + in solution 1 1 1 (i.e. the radioisotope reservoir). C. = the concentration of radioisotope in solution 1 1 (in cpm/ml). C. = the concentration of unlabeled Na + in solution 1 1 1 (i.e. 125 uEq/ml). Values for unidirectional flux of sodium were calculated in terms of uEq/cm /hr. 4. Hormone preparations A. Vasopressin The vasopressin used in the present studies consisted of Pitressin (lots Ck 217 and Ej 112? Parke-Davis & Comp. Ltd. Brockville, Ontario). This i s an aqueous solution of arginine and lysine vasopressin at a concentration of 20 pressor units/cc. Solubility of the hormone in solution i s 37 maintained by adjustment of the pH with acetic acid. Before, the vasopressin solution was added to the membrane prepara-tion, i t s pH was brought to neutrality by addition of NaOH. 6 2 . 5 - 6 2 5 . 0 u l of Pitressin was diluted into the 25.0 ml of 1 1 amniotic saline i n the open reservoir (dilutioni /40- /400). A volume of 0 . 2 5 acetic acid equal to that of the hormone solution was brought to neutral pH by addition of 0 .2 N sodium hydroxide, and was added to the control membranes. The Pitressin solution described above w i l l be refered to, henceforth, as "vasopressin". B. Prolactin A purified prolactin preparation (NIH P-S-ll, 26.4 IU/mg) was kindly supplied by Dr. R. Bates, National Institute of Health, Bethesda, Md. It was dissolved i n amniotic saline at 1 mg/ml by careful adjustment to pH 9 . In the radio-'s 22 isotope experiments, the saline contained -'HgPor NaCl at the same specific activity as that i n the closed-circuit reservoir when studying fetal-maternal flux. In those experiments investigating maternal-fetal water or sodium movement, prolactin solution was added to the open-circuit reservoir, free of radioisotope. In the unidirectional water and sodium flux experiments hormone dilution into the A saline reservoir was /100, while in net water flow experiments the dilution was V 5 0 . Controls consisted of amniotic saline adjusted to the same pH i n the same way as for hormone, or serum albumen at the same concentrations as the prolactin, and adjusted in the same manner. 5. Salines Salines were designed to parallel the appropriate natural fluids, and consisted of 1 A. Amniotic saline Na+, 1 2 5 . 0 ; K+, 6 . 2 ; Ca 2 +, 4 . 5 ; Mg2+, 2 . 3 mEq/l i a l l were added as chlorides. Glucose was added at 1 . 0 g/1 . A phosphate buffer, pH 7 . 8 ( 0 . 2 M NaH 2P0 i/Na 2HP0^) was added at 10 ml buffer/l of saline. The f i n a l pH was 7 .^ and the osmolarity was 2 8 6 mOsm/l. B . Maternal saline Na+, 1 5 0 . 0 ; K+, 5*5; Ca 2 +, 4 . 4 ; Mg2+, 2 . 6 mEq/l; a l l were added as .chlorides. Glucose and phosphate were added as for the amniotic saline. The f i n a l pH was 7 . 4 , and the osmolarity was 3 1 4 mOsm/l. SECTION I THE EFFECT OF VASOPRESSIN ON WATER AND SODIUM MOVEMENT  THROUGH THE ISOLATED AMNIOTIC MEMBRANE OF THE GUINEA-PIG INTRODUCTION! Vizsolyi and Perks(1974) demonstrated that the isolated membrane of the guinea-pig i s sensitive to both arginine vasotocin and arginine vasopressin. When arginine vasotocin (8-100 mU/ml) was added to the media bathing the fet a l side of the membrane i t slowed, and could reverse fetal-maternal water movement through the amnion. Fetal pituitary extracts and synthetic arginine vasopressin also stimulated water movement across the amniotic membrane of the guinea-pig. Since the threshold dose for arginine vaso-pressin was lower than for arginine vasotocin (1.0 mU/ml v.s. 6.4 mU/ml), i t seemed to be more effective in stimu-lating water movement across the membrane. Oxytocin (100 mU/ml) was without effect. Reversals of water flow against an osmotic gradient of 28 mOsm/l and a hydrostatic gradient of 2 cm of water were noted for a number of membrane preparations i n response to neurohypophyseal hormone. This suggested that water was not moving by s t r i c t l y passive means (e.g. bulk flow), but rather by an active process (e.g. perhaps being driven i n a water-solute coupling by a pump). 39 4 0 An investigation was undertaken to determine whether or not vasopressin could stimulate an increase in the unidirectional flux of tr i t i a t e d water from the maternal to the fet a l side of the isolated amniotic membrane of the feta l guinea-pig. If the unidirectional flux of water showed a response to treatment with vasopressin, the next proposed step was to determine whether water movement could be linked to a possible sodium transport. RESULTSi 1. Permeability coefficients for the guinea-pig amnion The permeability of the isolated amnion to water was determined by measuring transmembrane movement of tr i t i a t e d water i n response to a tracer gradient? the perfusion c e l l used in determining permeability coefficients i s shown in Fig. 1 (see methods). Fluxes were measured i n both directions across the membrane by inserting the amnion with either i t s fe t a l or i t s maternal surface in contact with the radioisotope-containing closed circulation. Although the membrane ranged from 0 . 62 of term u n t i l apparently overdue, there was l i t t l e evidence for any obvious relation-ship between permeability to water and gestational age over the period studied. In ten experiments, the permea-b i l i t y coefficient, K-fc r a n s» for water movement from the maternal to the fet a l side of the membrane was calculated -4 -1 to be 3 * 0 9 + 0 . 1 5 x 10 cm sec . In thirteen experiments Ktrans ^ n e reverse direction was 2 . 6 2 + 0.14 x 10 cm sec" 1. The difference i n the permeability coefficient in the two directions was s t a t i s t i c a l l y significant (Student's t test, P<0 . 0 5 ) . This implied that water move-ment would be easier from mother to fetus than from fetus to mother. 41 42 2. The action of vasopressin; effects on unidirectional  flux of water through the amnion The addition of vasopressin (50-500 mU/ml) to the fetal side of the perfusion chamber caused an increase i n water movement through the amnion i n the maternal-fetal • direction as judged by use of tr i t i a t e d water. A total of 18 different membranes were tested. Fig. 2 shows the means (i.e. x) + the standard error of the means (i.e. S.E.M.) for a l l the responses obtained to the various doses of vasopressin, as li s t e d in Table I. The responses were calculated as percentage change in water movement over the 35 minute period after addition of hormone or neutralized acetic acid (i.e. sodium acetate); the equi-l i b r i a t i o n period from 30 to 60 minutes after setting up the membrane was averaged for each preparation, and was used as the "base lin e " (i.e. 0 % change). Just as i n earlier studies concerned with the effect of neuro-hypophyseal hormones on the net flux of water across the amnion (see Vizsolyi and Perks, 1974) , the response to hormone occured within 5-10 minutes of addition of vaso-pressin. The fact that water movement across the amnion from the maternal saline (314 mOsm/litre) to the amniotic saline (286 mOsm/litre) increased in response to hormone 4-3 Figure 2. The Effect of Different Doses of Vaso-pressin on Unidirectional Water Flux through the Guinea-Pig Amnion The amniotic membranes (N = 18) are arranged in order of increasing doses of vasopressin. At the origin, the membranes received either acetic acid adjusted to neutral pH with sodium hydroxide (i.e. forming sodium acetate), or vasopressin ( 5 0 - 5 0 0 mU/ml) on their fetal side. The period just prior to hormone addition (i.e. 6 0 min) was taken as the "base lin e " value (0 % change) from which the percentage change in water flux was calculated after the addition of sodium acetate or vasopressin. The values are expressed as the mean + the standard error of the mean. Ordinatesi percentage change in maternal-fetal water movement from the "base line" value. Absissaet time from addition of sodium acetate or vasopressin, in minutes. 44 T I M E ; M I N U T E S . 45 treatment indicated that water could move against electro-chemical gradients. This supported the observation i n the net flux experiments of Vizsolyi and Perks (1974) that water movement could be reversed from the f e t a l -maternal to the maternal-fetal direction in response to vasopressin. The magnitude of the change i n water movement seemed to be a function of the dose of vasopressin used (see Table I ) . When one plots the log dose of vasopressin against the percentage increase in maternal-fetal water movement one obtains a l i n e a l relationship such as shown i n Fig. 3. The threshold dose appears to be about 46 mU/ml. The fact that one obtains a definite log dose/response relationship for both net and unidirectional flux of water across amnion seems indicative of hormone sensitivity of the membrane being genuine and not an a r t i f a c t . It should be pointed out that these experiments studying the effect of vasopressin were carried out in an early version of the perfusion c e l l shown i n Fig. 1. Dead space and unstirred layers resulted in lower transfer rates than were recorded with the apparatus used for determining permeability values for the amniotic membrane to water. However, the responses were clear, and thus, were calculated as percentage change so that they could be compared directly with later data. 4 6 Figure 3. The Relationship between the Logarithm of the Dose of Vasopressin added to Amniotic Membranes and the Percentage Increase in Unidirectional Flux of Water The curve was calculated from data shown in Table I. Ordinate 1 percentage increase in unidirectional water flux from the "base lin e " value for the 35 min-ute period subsequent to hormone addition. Absissaei logarithm of the dose of vasopressin (in mU/ml) added to the membranes. 47 0 1 2 L O G D O S E O F V A S O P R E S S I N . 48 TABLE I THE EFFECT OF VASOPRESSIN ON MATERNAL-FETAL WATER MOVEMENT THROUGH THE ISOLATED AMNION OF THE GUINEA-PIG Treatment sodium acetate 50 mU/ml vasopressin 100 mU/ml vasopressin 125 mU/ml vasopressin 250 mU/ml vasopressin 500 mU/ml vasopressin Number of preparations 2 1 2 2 3 8 % increase in water movement in 35 min 0.4 + 0.4 a 1.6 " 3.8 + 1.9 4.7 + 2.9 9.1 + 0.6 12.3 + 1.5 a _ x + S.E.M. 3. The action of vasopressin; effects on unidirectional  flux of sodium through the amnion Since water was moving against an osmotic gradient of 28 mOsm/l in response to hormone i t seemed to suggest a water-solute interaction i n which water could be "dragged" across the membrane by an ion such as sodium. Studies were carried out with the perfusion chamber shown i n Fig. 1 in order to determine whether or not vasopressin could increase maternal-fetal movement of sodium across the amnion. In contrast to the previous studies which were concerned with unidirectional flux of water, hormone (500 mU/ml vasopressin) was not added u n t i l after a 90 minute, rather than a 60 minute, equilibriation period. 49 In addition to this, both sides of the amnion were bathed with amniotic saline. At 35 minutes after the addition of vasopressin to the fetal side of two membranes an increase of only 0 . 7 + 0 . 9 % was recorded for sodium movement in the maternal-fetal direction. Since this value i s not s i g n i f i -cantly different from control values over this same time interval i t did not appear that vasopressin could stimulate an increase of sodium flux in this direction under the ionic conditions employed i n the experiment. It seemed possible, however, that the membranes may have been slow in responding to the hormone, and thus i t seemed auspicious to study the preparations over a longer period of time than the usual 35 minutes after hormone addition. Contrary to what was ex-pected, a decrease i n maternal-fetal sodium movement across the amnion was observed over the longer time period. In the two membrane preparations studied, a decrease in water movement of 1.4 + 0.6 % was recorded i n the f i r s t hour, f o l -lowed by a 12.4 + 8.2 % and a 12.2 + 7.8 % decrease in the second and third hours respectively, after hormone treat-ment. Both of the membranes showed a reduction of maternal-fet a l sodium movement after the addition of vasopressin, but the response of one was more marked than the response of the other. 50 DISCUSSION! The results given here confirm the findings of Vizsolyi and Perks (1974) that vasopressin can increase the maternal-fet a l flow of water through the amnion. The water movement does not seem to be coupled to sodium transport. In addition, some of the permeability characteristics of the freshly dissected membranes were determined. 1. The diffusional permeability of the isolated guinea- pig amnion Previous studies on the permeability of the isolated amnion concentrated on human preparations taken after delivery. The permeability coefficients obtained for -4 -1 water movement were 2.88 + 0.19 x 10" cm sec" (Lloyd 4 1 et a l . , 1969) and 0.5 to 4 .0 x 10" cm sec" (Page et a l . , 1974 a). The values obtained with the perfusion c e l l shown i n Fig. 1, for freshly dissected guinea-pig amnion, taken prior to birth, agree well with the human results. These measurements made on guinea-pig amnion with the improved perfusion c e l l shown i n Fig. 1 were considered to be relatively dependable since they were carried out in a small chamber, perfused continuously by a double circulation system designed to reduce dead space, unstirred layers, and back fluxes, a l l of which were known to have effects on the 51 results (see Page et a l . , 1974 a). In addition, and unlike previous studies, the permeabilities were calculated for both directions across the membrane. It was found that diffusional permeability in the maternal-fetal direction (3.09 + 0.15 x 10'^cm sec" 1) was higher than that from fetus to mother (2.62 + 0.14 x 10 cm sec ). Differences of this type have been noted in other membranes, such as the toad bladder, where the discrepancies were of a closely similar order of magnitude (mucosal to serosal = 2.00 x 10"^ cm sec - 1 , serosal to mucosal = 1.48 x 10~^cm sec - 1; Hays, 1972). At present, i t i s not clear whether this assymetry i % of any importance in the functioning of the amnion; however, the s t a t i s t i c a l l y significant greater ease with which water could pass in, towards the fetus, might help to ensure the f l u i d environment of the fetus, whilst the greater d i f f i c u l t y with which water could leave would tend to retain any water which accumulated. The permeability of the amniotic membrane to water did not seem to change significantly over the course of gesta-tion, although there was some indication of a very slight increase in late stages of pregnancy. Gillibrand (1969 a) found a large and linear increase in water transfer between 14 and 26 weeks, and later a plateauing or a leveling off of the transfer rate as term approached i n the human being. 52 The apparent discrepancy between the guinea-pig and human amnion could represent a species difference or a result of differences i n methods of measuring water transfer. Rather than using an in vitro membrane preparation, Gillibrand ( 1 9 6 9 a) introduced deuterium oxide into the amniotic sac and determined water transfer by measuring concentration of deuterium oxide in amniotic f l u i d and maternal blood several hours later. A critism of this method i s that Gillibrand ( 1 9 6 9 a) was not actually measuring membrane permeability of amnion, in a s t r i c t sense. In the i n vivo study the fetus could have altered the amount of deuterium oxide passing from the amniotic sac to maternal circulation by either swallowing or absorption across the skin surface. Complications in interpretations of results were eliminated in the i n vitro studies described here, i n the sense that water movement could occur only across amniotic membrane. Another factor which might explain why Gillibrand ( I 9 6 9 a) observed a large increase in water transfer over gestation, while only a very small increase was observed in the present study, may be a result of the range i n gestational ages studied. In the present investigation membranes were taken only between 0 . 6 2 and term, while i n the study with human amnion transfer, rates were calculated between O .32 and 53 term. Thus, there may be a definite increase i n water movement across guinea-pig amnion prior to 0.62 of term. 2. The action of vasopressin on water flow through  the amnion Vizsolyi and Perks (197*0 found that not only could treatment of amniotic membranes with vasopressin slow net fetal-maternal water movement, but i t could also, on oc-casion, cause a reversal of flow. In the present inves-tigation vasopressin was found to be capable of stimu-lating an increase in unidirectional water flux across amnion in the maternal-fetal direction. Indication that the responses observed represent a true sensitivity of the amnion to hormone can be seen by the fact that membranes showed no significant increase i n water movement (approx. 0.4 %) i n response to control saline containing sodium acetate. Further indication that the response of amnion to vasopressin i s real comes from the fact that, there i s a definite log dose/response relationship showing a linear regression lacking an overlap of the standard errors,of the means. This was also noted in the net flux studies of Vizsolyi and Perks (197*0 • One difference between the two studies i s the fact that in the unidirectional flux studies the threshold dose was calculated to be approxi-54 mately 46 mU/ml, while i n the net flux studies i t was calculated to be about 1 mU/ml. Part of the differences in these values may be explained on the basis of d i f -ferences in the mechanics of diffusional and bulk flow. Page et a l . (1974 b) have shown that net flow through the amnion takes place as the organised movement of water known as bulk flow. In contrast, the radioisotope method measures flow resulting from the random movement of water molecules, or diffusional flow. A number of workers have shown that hormones can have different effects on the two types of flow; for example, neurohypophyseal principles increased net flux of water across amphibian skin by 160 whilst the unidirectional flux increased l i t t l e i f at a l l (see Maetz, 1968). Therefore the relatively small changes seen in the radioisotope studies (maximum - approx. 12.3$) compared to the net flux experiments (maximum = approx. 200 % ) , and the discrepancy between threshold values for vasopressin in the two experiments, are less important than the fact that amnion shows a log dose/response to vaso-pressin i n both types of studies. After having verified the fact that water could move against prevailing osmotic gradients in response to hormone the next step was to de-termine how such a process was accomplished. 55 3« Water-solute coupling One means by which water can move against an osmotic gradient i s by coupling of water transfer to active solute transport. Such a coupling depends on "dead-end channels'* in the form of lateral spaces, basal infoldings, or intra-cellular canaliculi such as seen in ga l l bladder, vertebrate intestine, renal tubules, c i l i a r y body, salivary gland s t r i -ated duct, l i v e r , pancreas, stomach, avian salt gland, and g i l l s of aquatic animals (Hochachka and Somero, 1973). When ions, such as sodium, are pumped into these closed channels, the f l u i d inside becomes hypertonic. Diffusion of ions down the concentration gradient within the channel towards the open end causes water to move across the channel walls to the inside i n response to the osmotic gradient. Depending on such factors as the radius and the length of the channels, permeability of channel walls to water, and orientation of the pumps, the f l u i d emerging from the open end of the chan-nel w i l l be either isotonic or hypertonic (Diamdnd, 1971). Histological and electron microscope studies of Danforth and Hull (1958), Bourne and Lacy ( i 9 6 0 ) , Bourne (1962), and Hoyes (1968 b) indicate that the amnion may be involved in active secretion or absorption of water and ions. Bourne and Lacy ( i960) and Bourne (1962) observed that the amniotic epithelium changes as gestation advances from 56 simple epithelial c e l l s to complex cel l s in later stages of pregnancy. The epithelial c e l l s of the human amnion i n later stages of gestation possess complicated intercellular canals, villous folds in the intercellular spaces, secretory granules, membrane-bound vesicles, endoplasmic reticulum, Golgi apparatus, and surface microvilli (Hoyes, 1968 b). Although i n early pregnancy the epithelial c e l l s could allow an easy passage of f l u i d through them, in later stages of pregnancy the cel l s would probably be more selective due to their more complex make-up. Since the ce l l s of the human amnion possess closed end channels there i s a possibility for water-solute coupling. Unfortunately, Scoggin et a l . (1964) could find no indication that water could cross the human amnion except i n response to osmotic or hydrostatic gradients. If the cells of the guinea-pig amnion were simi-lar to those of the human amnion i n possessing closed end channels then perhaps vasopressin could stimulate water movement against the prevailing osmotic gradient by means of a water-solute coupling. In order to test this hypothesis the effect of vasopressin on maternal-fetal sodium movement was studied. Vasopressin was found to produce a decrease rather than an increase i n sodium movement from the maternal to the fe t a l side of the amnion i n the absence of an osmotic 57 gradient. Although only two membranes were used i n the study of vasopressin's effect on sodium movement, further experiments with sodium or other ions were not carried out. The reason for not elaborating on the vasopressin studies was the fact that vasopressin could not be detected in the amniotic f l u i d of the guinea-pig at levels high enough to elecit a response i n the in vitro water flux experiments (Plath, personal communication). Inability to detect vaso-pressin i n amniotic f l u i d of the guinea-pig does not seem to be the result of vasopressinase activity since this en-zyme has not been found i n amniotic f l u i d of, mammals other than the primates. Vasopressin has not been found in human amniotic f l u i d although very low levels of vasopressinase activity have been detected (Rosenbloom et a l . , 1975). 4. Hormones present in amniotic f l u i d Although i t seems unlikely that vasopressin occurs in amniotic f l u i d of the guinea-pig in levels high enough to affect in vitro water movement, the present studies demonstrate hormone sensitivity of the membrane, This i n -dicated that perhaps some other hormone(s) present in am-niotic f l u i d could mediate water and/or ion movement across the membrane. Some of the hormones that have been detected in amniotic f l u i d include« e s t r i o l (Michie and Robertson, 1 9 7 1 ) , cortisone (Nicholas et a l . , 1 9 6 3 ) » Cortisol 58 (Murphy et a l . , 1975)» pregnanediol and pregnedione (Patti and Stein, 1964) , testosterone (Giles et a l . , 1975)» human chorionic gonadotrophin (Crosignani et a l . , 1970, 1971)» human placental lactogen (Josimovich, 1971)» and prolactin (Tyson et a l . , 1972). At least a few of the hormones found in amniotic f l u i d have been implicated in hydro-mineral balance i n mammals and other vertebrates. Thus i t seemed reasonable that one or more of the hormones might regulate passage of water and/or ions across the membrane. In order to test this hypothesis the effect of prolactin on hydro-mineral movement across the amnion was investigated. SECTION II THE EFFECT OF PROLACTIN ON WATER AND SODIUM MOVEMENT THROUGH  THE ISOLATED AMNIOTIC MEMBRANE OF THE GUINEA-PIG INTRODUCTION! Although prolactin has long been known to have reproductive functions in mammals, there i s much evidence for an effect on osmoregulation i n sub-mammalian verte-brates, such as f i s h . Prolactin promotes survival of certain hypophysectomized teleosts i n fresh water, probably by regulating plasma electrolyte levels (Ball, 1969$ I&toi 1972)$ i t influences water and ion movement across g i l l s (Lam, 1969), renal tubules (Stanley and Fleming, 1967$ Lahlou and Giordan, 1970), intestine (Utida et a l . , 1972), and the urinary bladder of some fresh-water teleosts (Johnson et a l . , 1972, 1974$ Johnson, 1973)* Despite this, the effects in mammals are concerned mainly with the mammary glands and corpus luteum, and any connections with osmoregulation are indirect or uncertain (J/^rgensen, 1968). Recently, two factors emerged that were suggective of a salt-water effect of prolactin in mammals. F i r s t l y , i t has been show&i that prolactin occurs in high amounts i n human amniotic f l u i d (Friesen et a l . , 1972$ Josimovich, 1973» Josimovich et a l . , 1974$ Parke, 1973). Secondly, 59 6 0 the amniotic membrane has been found to be sensitive to hormonal control, since neurohypophyseal peptides are capable of causing an active uptake of water into the amniotic cavity by the amniotic epithelium (Vizsolyi and Perks, 1974). Therefore, i t seemed possible that prolactin might combine i t s importance in reproduction with i t s phylogenetic role in osmoregulation, by a direct action on the amnion during pre-natal l i f e . It raised the inter-esting possibility that prolactin might parallel i t s role in hydro-mineral metabolism in fi s h by influencing water and/or ion movement in the mammal during i t s early period of "aquatic" existance. The present investigation was initiated to determine whether or not mammalian prolactin could influence water and/or sodium movement through the isolated amniotic membrane of the guinea-pig. 61 RESULTS t 1. The action of prolactin; effects on unidirectional  fetal-matemal water movement through the amnion The addition of prolactin (10 ug/ml) to the fet a l side of the amniotic membrane was found to cause a clear decrease i n fetal-maternal water movement (for example see membrane 8 i n Fig. 4). In two other preparations studied subsequently (membranes 1 and 3 in Fig. 4) the decrease in water movement was not as great as for membrane 8. Since the hormone might be slow in exerting an effect, i t seemed advantageous to extend the time interval of study after hormone addition from 90 minutes to 3 hours. Fig. 4 shows the responses obtained, arranged in order of increasing gestational ages. There was a time lag of 30-60- minutes before the onset of a response, so that prolactin was slower acting than vasopressin. The exact form of the responses varied in different preparations, with some showing a clear effect i n the second hour, whilst others gave a smaller response, which became marked in the third hour. In contrast, a l l five control experi-ments showed the opposite effect, with a slow, progressive rise i n flow throughout the three hour period. The five three-hour experiments with prolactin, 6 2 Figure 4-. The Effect of Prolactin on Unidirectional Water Flux through the Guinea-Pig Amnion in the Fetal Maternal Direction The amniotic membranes are arranged i n order of increasing proportion of term, as indicated on each graph (bottom l e f t ) . At the arrows, membranes 1 to 8 received prolactin at 10 ug/ml, on the f e t a l sidej treatment continued u n t i l the end of the experiment. In the same way, membranes 9 to 13 received control saline or albumen (10 ug/ml), adjusted for pH as for prolactin. Water fluxes were measured by the use of tri t i a t e d water. Dotted lines indicate the i n i t i a l period of equilibriation of radioisotope. 0 r d i n a t e s » fetal-maternal water flow, ul of water per cm2 of amnion, per hour. Absissaei time from addition of radioisotope, in minutes. 63 1100 1000 700 600 1300 1200 1100 850 750 900 800 700 900 800 1100 1000 1000 800 u_ 600 Z U J LU > o U J r— < I < Z U J r-< H O R M O N E Prolactin (10 H s / m l ) 1 1 i / 0.85 I 1 l 2 — \ . • 1 J O S S . . \ A _ 3 i • — • — *%*+\ : 0.97 V*-* • i i i 1 r • * -; \ A 0 9 7 . " X 5 \ i / (over term) l-6 \/"s—: 1 / ' • 7 Jt (over term) • *v , T i i i i * l r -v • • j . v-\ . : (over term) \ ' ., i n | 8 i 60 120 180 2 4 0 C O N T R O L S amniotic sal ine 1000 900 1100 1000 • 0.65 / 10 1000 900 900 800 V - s / " V - V . / - / V / 11 ;' 0.82 _J 1 '. I L_ /(i.92 \ • • / \ / B. a lbumin 900 800 700 X.92 _1 L. 13 I 60 120 180 240 T I M E ; M I N U T E S . 64 included i n Fig.4 , could be analyzed s t a t i s t i c a l l y . If the resting flow was taken as the average flow in the 30 minute period prior to prolactin administration, the average f a l l i n flow i n the second and third hours was 3 . 5 $ and 9*7% respectively. S t a t i s t i c a l analysis of a l l individual readings showed that the f a l l was significant in both the second and third hours (Student's t test of paired comparisons, Steel and Torrie, I 9 6 0 ; P < 0 . 0 5 ) . In the third period of one hour, the controls showed a rise of 5*1 f ° i therefore, the comparison between experi-mental and control preparations, during the f i n a l hour of the experiments* suggested that prolactin accounted for an overall reduction in flow of 14.8 $ - and there was no indication that this f a l l would not have increased further with time. However, this estimate of the magnitude of the changes in Fig.4 i s probably too low, since i t l e f t out the three particularly clear responses seen in the shorter 90 minute experiments (membranes 1 , 3 , 8 ) . If a conservative assum-ption i s made that the flow through these three membranes would have remained unchanged, at i t s f i n a l level, during a third hour - and this i s clearly conservative, since a l l five three hour experiments showed a continual f a l l during this additional period - then the average response 6 5 f o r ; a l l eight experiments would be estimated at a 1 3 . 5 % f a l l by the second hour, and & 17 % f a l l by the third hour. This f i n a l value constituted a 2 2 . 1 fo factor between experimental and control values. This value i s the best estimate possible for the effect of prolactin in a l l eight responses shown i n Fig. 4 . In three further experiments, prolactin was added to the maternal side of the amnionj no responses were seen. Fig. 4 suggests that the response to prolactin tended to increase with gestional age, but the relationship was only rough, and further studies are needed. Nevertheless, attention i s drawn to the exceptionally strong response given to membrane number 8 , which was judged to be past i t s expected birth date, and about three days overdue, i f one accepts the data of Draper ( 1 9 2 0 ) that gestation in the guinea-pig i s 6 5 days in length. 2 . The action of prolactin; effect on net flux of water  through the amnion Since the essential importance of changes i n water flux l i e i n the overall movement and distribution of water within the body, the effect of prolactin on net water flow through the amnion was investigated. This was carried out on an isolated amniotic membrane, attached across a 66 supporting tube, and immersed i n fluids which reproduced natural conditions as closely as possible (see Methods). At the onset of the experiments measurements of weight changes showed that there was a movement of water from the amniotic saline within the tube to the aerated saline which surrounded the preparation (average value = 37.0 ul/cm /hr, 10 experiments). This movement followed the small hydrostatic and osmotic gradients which were present (2 cm water: 28 mOsm/l, respectively). In five experiments the resting flow was recorded for one hour, after which time both bathing salines were replaced, and prolactin at 20 ug/ml was included in the inner amniotic saline, which contacted the fetal surface of the amnion. In every case, the prolactin appeared to cause a marked and constant reduction in fetal-maternal water flow (Fig. 5 ) . In four cases, there was an i n i t i a l latency of about one hour, followed by a profound f a l l i n fetal-maternal flow, which remained-depressed at the termination of the experiments (3 hours). This was similar to the findings for unidirectional fluxes. In one case (Fig. 5 membrane 4 ) , the f i n a l flow f e l l to zero. In one ad-ditional preparation, where the resting flow was unusually low (Fig. 5 membrane 1), there was an immediate f a l l to zero during the f i r s t hour after prolactin treatment. 67 Figure 5. The Effect of Prolactin on Net Water Flux through the Guinea-Pig Amnion in the Fetal-Maternal Direction The amniotic membranes are arranged i n order of increasing proportion of term, as indicated on each graph (bottom l e f t ) . Hormone-treated and control membranes 1 and 6, 2 and 7» 3 and 8 are pairs taken from the same fetuses. At the arrows, membranes 1 to 5 (black columes) received prolactin at 20 ug/ml on their f e t a l side; treatment continued u n t i l the end of the experiment. In the same way, membranes 6 to 10 (light hatching) received control saline, adjusted for pH as for prolactin. Fetal-maternal flow was measured gravimetrically. Ordinatest f e t a l -maternal water movement, u l of water per cm2 of amnion per hour. Absissae; time from the onset of the experiment, in one hour intervals. T I M E H O U R S . 6 9 Although most membranes were close to term, one amnion, which was judged to be overdue by two days, showed a remarkably strong response (Fig.5 » membrane 5 )I this i s in agreement with findings obtained during the unidirec-tional flux experiments. In the five experiments, the average f a l l over the three hour period of prolactin was 5 9 . 4 fo. S t a t i s t i c a l analysis showed that the reductions in flow during the second and third hours were significant (Student's t test of paired comparisonss 95 % level of probability). Five control experiments showed no changes parallel to prolactin effects? there was often an increase in f e t a l -maternal flow. For the controls to which pH adjusted saline had been added there was a 3 * 6 + 0 . 5 % decrease in water movement in the second hour, and a 1 . 3 + 0 . 2 % increase i n the third hour. When 2 0 ug/ml albumen solution was added to two other control preparations there was a 6 . 7 + 2 . 3 fo increase i n the second hour and a 4 . 8 + 3 . 9 % increase in the third hour. The comparison of experimental and control data was particularly notable in three cases, where i t was possible to study control membranes from the same fetuses as the experimental preparations (Fig. 5 , membranes 1 and 6$ 2 and 7i 3 and 8 ) . S t a t i s t i c a l analysis confirmed that there were no significant effects i n the 70 control preparations (Student's t test of paired comparisons 9 5 $<» confidence level). 3. The action of prolactin; the comparison of the effects  on unidirectional and net fetal-maternal water flux  through the amnion Fig. 6 shows a comparison of the changes induced by prolactin in both types of experiment, when the results from unidirectional flux were converted into the same form used to express net flow (one hour periods). Five three hour experiments are averaged for each graph. It i s clear that prolactin acts on the fe t a l surface of the amnion to cause a consistent f a l l i n fetal-maternal water flow as judged by either unidirectional or net flow, and the latency and form of the flow i s similiar i n both cases. St a t i s t i c a l analysis (Student's t test of paired compari-sons, 95 % level of probability) showed that i n both types of experiment the changes i n the f i r s t hour were not significant, whilst in the second and third hours were significant at the 95 f» confidence levels. Control experiments showed no significant changes at any time. However, comparison of the results of the two methods showed that the decrease after prolactin was approximately five to six times greater when judged by net flow. This 71 Figure 6 . The Comparison of the Effect of Prolactin on Net Flux and Unidirectional Flux of Water through the Guinea-Pig Amnion i n the Fetal-Maternal Direction Topt Average Values for Net Water Flux (Gravimetric) Bottomt Average Values for Unidirectional Flux (Trit-iated Water) A l l measurements are for the fetal-maternal d i -rection of flow, and are expressed by the same param-eters. Each graph represents an average of five com-plete three-hour experiments. A l l membranes were over 0,62 of term. Black columns show experiments with prolactin treatment (20 ug/ml for net fluxj 10 ug/ml for unidirectional flux). The period of treatment lasted from the arrows to the end of the experiment. Columns with light hatching show corresponding con-t r o l experiments carried out with saline, pH adjusted as for prolactin. The values above the columns show the significance of the change from the resting flow, as determined by Student's t test of paired compari-sons; NS = not significant. Ordinatesi fetal-mater-nal water movement, u l of water per cnr of amnion, per hour. Absissaet time from the onset of the ex-periment, in one hour intervals. 7 2 N E T F L U X P r o l a c t i n C o n t r o l 0 1 2 3 4 0 1 2 3 4 T I M E ; H O U R S -73 difference i s probably due, f i r s t l y , to the higher dose of prolactin used in the net flow experiments (20 ug/ml, net flowi 10 ug/ml, unidirectional flow), and, secondly, to differences between bulk flow and diffusional flow (see page 54 and the discussion). However, i t should be remembered that the two sets of experiments were carried out i n different ionic conditions, and direct comparisons are d i f f i c u l t . Evidence that the decrease i n fetal-maternal water movement i s due to a response of the membrane to prolactin, and not due to a change i n the osmotic pressure, comes from two types of observation. F i r s t l y , the membranes treated with albumen showed an increase rather than a decrease i n fetal-maternal water movement. Secondly, the osmotic pressures of amniotic saline, pH adjusted amniotic saline, pH adjusted amniotic saline containing 10 ug/ml prolactin, and pH adjusted amniotic saline containing 20 ug/ml pro-l a c t i n were not significantly different when measured with an osmometer (Model 31I*AS, Advanced Instruments, Newton, Mass.). The differences were within the precision of error of the machine used. 74 4 . The action of prolactin? effect on unidirectional  maternal-fetal water flux through the amnion In eighteen early experiments i n which prolactin ( 2 0 ug/ml) was added to the f e t a l side of the perfusion •; chamber there was an indication that prolactin might be able to increase maternal-fetal water movement. At the end of 30 minutes after hormone addition there was a 0 . 9 + 1.1 % .increase i n water movement, followed by a 5 . 2 + 4 . 8 $ and 4 . 4 +. 3 . 6 fo increase at 6 0 and 90 minutes respectively. Interpretation of the results was d i f f i c u l t because no control membranes had been set up for these experiments. The study was repeated at a later date using controls, and a longer time interval (3 hours v.s. 90 minutes), since fetal-maternal investigations had indicated that the hormone was often slow acting. Upon addition of 10 ug/ml prolactin to the f e t a l side of five membranes, an increase i n maternal-fetal water flow of 1 0 . 2 + 2 0 . 3 % was recorded i n the second hour and 8 . 5 + 2 0 . 2 % i n the third hour. Six control membranes that had been treated with 10 ug/ml albumen showed an average decrease i n water flow of 2 . 2 + 7 . 6 % in the second hour and an increase of 1 . 0 + 1 0 . 0 % in the third hour (see Fig. 7 ). S t a t i s t i c a l analysis (Student's t test of paired comparisons, 95 $ confidence 7 5 Figure 7 . The Effect of Prolactin on Unidirectional Water Flux through Guinea-Pig Amnion in the Maternal-Fetal Direction At the arrows, membranes received either 10 ug/ml prolactin (N = 5 ) or 10 ug/ml albumen (N=*= 6 ) on their f e t a l side; treatment continued u n t i l the end of the experiment. Data were averaged over one hour periods, and expressed as the mean + the standard error of the mean, so that comparison could be made with Fig. 6 . S t a t i s t i c a l analysis (Student's t test of paired com-parisons, 9 5 % confidence interval) shows that a l l flux values subsequent to treatment with prolactin or albumen are not significant (N S). Ordinatesi mater-nal-fetal water movement, pi of water per cm2 of am-ion, per hour. Absissaei time from the onset of the experiment i n one hour intervals. 76 M A T F R N A L - F E T A L D I R E C T I O N p r o l a c t i n a l b u m e n ( 10 M G / M L ) ( I O M G / V R ) 0 1 2 3 4 0 1 2 3 4 T I M E ; H O U R S . 77 level), however, revealed that none of the changes in water movement recorded for either experimental or control prepa-rations were significant at any time. There was also no indication of a dose/response relationship between the early preparations to which 20 ug/ml prolactin had been added and later ones, which were treated with 10 ug/ml prolactin when equivalent time intervals were compared. 5 . Permeability of the amnion to sodium Permeability of the isolated amniotic membrane to sodium was determined by measuring transmembrane flux of radioactive sodium. In every case the permeability value was taken as the flux rate of sodium across the amnion 90 minutes after the preparation was set up (i.e. just prior to the addition of hormone or albumen). Membrane permea-b i l i t y to sodium was found to change markedly over the course of gestation (see Fig. 8 ) . From day 30 "to day 55 permeability of amnion to sodium in the fetal-maternal direction seemed to decrease by a factor of about 5« The permeability in this direction seemed to increase by a factor of about 13 between days 5 5 to 7 0 , but from day 57 to day 70 the increase was only about 4 fold. Membranes had not been taken from fetuses younger than 40 days when studying maternal-fetal movement of sodium, and thus, i t 78 Figure 8. The Permeability of Guinea-Pig Amnion to Sodium at Different Gestational Ages 22 Unidirectional flux of radioactive sodium (Na ) was studied i n both directions across the isolated amniotic membrane of the guinea-pig at different ges-tational ages. The flux rate at 90 minutes, just prior to hormone addition, was used as an indication of permeability for each membrane. Ordinatesi uni-directional sodium movement, uEq of sodium per cm2 of amnion, per hour. Absissaei gestational age i n days for the fet a l guinea-pigs from which amniotic mem-branes were taken. U N I D I R E C T I O N A L S O D I U M F L U X ; p E q / c m 2 / h r 33 m O -»3 NO 80 was not possible to determine whether permeability i n this direction decreases between days 3 0 to 5 5 . Between days 57 to 6 8 maternal-fetal sodium movement increased by a factor of approximately 8 i and from day 57 to day 70 permeability increased 3 5 fold. More experiments are need-ed to determine i f there i s a significant difference i n membrane permeability to sodium in the two directions. 6 . The action of prolactin? effect on unidirectional  maternal-fetal sodium flux through the amnion In order to make a comparison of prolactin's effect on sodium movement with i t s effect on water movement possible, data were expressed as percentage change in ion flux following hormone treatment. The value for sodium movement at 90 minutes, just prior to the addition of prolactin or albumen was taken as the "base l i n e " (i.e. 0 % change). Upon addition of prolactin ( 10 ug/ml) .to seven membrane preparations increases of sodium flux in the maternal-feta l direction of 6 . 2 + 3 . 5 %* 1 1 . 3 ± 6 . 7 %» and 2 1 . 3 ± 8 . 3 % were recorded in the f i r s t , second, and third hours respectively. Changes of 5 . 9 + 4 .1 % t 1 0 . 8 + 8 . 1 % , and 2 1 . 3 ± 14.9 % wre observed in the f i r s t , second, and third hours for membranes treated with albumen (see Fig. 9 ) . S t a t i s t i c a l analysis (Student's t test of paired comparisons, 95 % confidence interval) revealed that there was a 81 Figure 9« The Effect of Prolactin on Percentage In-crease i n Unidirectional Sodium Flux through Guinea-Pig Amnion At the origin, the membranes received either albumen (10 ug/ml) or prolactin (10 ug/ml), on their fe t a l side. Sodium fluxes were measured by the use of Na 2 . The period just prior to hormone addition (i.e. 90 min) was taken as the "base lin e " value (0 $ change) from which percentage change in sodium move-ment was calculated after the addition of albumen or prolactin. Values were measured in the maternal-f e t a l direction (N = 14) and i n the fetal-maternal direction (N = 10), and expressed as the mean + the standard error of the mean. Ordinatesi percentage change in sodium movement from the "base l i n e " value. Absissaei time from addition of albumen or prolactin, in minutes. 82 M A T E R N A L - F E T A L D I R E C T I O N • F E T A L - M A T E R N A L D I R E C T I O N •—• p r o l a c t i n ( I O U G / M I ) o~o albumen ( I O U G / M L ) -I 1 1 I l I I 90 80 70 60 50 40 30 20 10 0 -10 •—• p r o l a c t i n ( 10 L I G / M L ) a-—o a l b u m e n ( 1 0 U G / M L ) / / x \ -1/ [/ _l L J 1 1 I 0 20 40 60 80 100 120 140 1 60 180 0 20 40 60 80 100 120 140 160 180 T I M E ; M I N U T E S 8 3 significant increase in maternal-fetal sodium movement in the f i r s t , second, and third hours after addition of prolactin. The controls, however, showed a similiar trend of increase after addition of albumen, and s t a t i s t i c a l analysis (Student's t test of unpaired comparisons, 95 f° confidence level) showed that there was no significant difference between experimentals and controls at any time. Thus comparison of prolactin-treated membranes with albumen-treated membranes suggested that prolactin probably does not influence maternal-fetal sodium movement. 7 . The action of prolactin; effect on unidirectional  fetal-maternal sodium flux through the amnion When prolactin ( 10 ug/ml) was added to five isolated amniotic membranes from f e t a l guinea-pigs, between 30 and 57 days old, a much greater change i n fetal-maternal sodium movement was observed than was noted i n the maternal-fet a l direction. The rate of sodium movement increased to 2 2 . 7 + 12.4 % i n the f i r s t hour, 22.4 + 14.4 $ i n the second hour, and 5 3 * 6 + 1 0 . 1 $ i n the third hour. Control membranes treated with albumen showed smaller changes that were similiar to those recorded i n the maternal to the fet a l direction. Values of 8 . 0 + 6.4 %, 8 . 3 + 7 . 3 $» and 1 5 . 1 + 1 5 . 9 f° were recorded i n the f i r s t , second, and third 84 hours, respectively for controls. Although there was no significant increase in fetal-maternal sodium movement i n either the f i r s t or second hours after addition of prolactin there was a very significant increase in the third hour (Student's t test of paired comparisons; p < 0 . 0 1 ) . This value was significantly different from that of control membranes in the third hour (Student's t test of paired comparisons, p < 0 . 0 5 ) . At no time did control membranes treated with albumen show a significant increase over the "base l i n e " value of sodium flow. Comparison of hormone and control preparations suggests that prolactin can cause a 38 .5 % increase i n fetal-maternal sodium movement. A mem-brane that was judged to be from a 70 day-old guinea-pig showed no response to treatment with prolactin. Further experiments are necessary, however, in order to determine whether "overdue" membranes are not able to increase sodium flux i n response to hormone; permeability to sodium at this stage may be maximal, and cannot be increased by hormone addition. 85 DISCUSSIONi The results given here suggest that prolactin i s capable of modifying water and sodium movement through the amnion. It i s able to slow water flow i n the fetal-maternal direction by decreasing osmotic and diffusional permea-b i l i t i e s to water. The hormone also stimulates an increase in fetal-maternal sodium movement. Changes i n membrane permeability to sodium over the course of gestation were also studied. 1. The action of prolactin on water flux through the amnion The present experiments were initiated in order to determine whether or not the amniotic membrane of the guinea-pig could alter i t s permeability to water and ions in response to treatment with prolactin. It had been previously noted that the membrane was sensitive to neuro-hypophyseal hormones, but i t did not appear that vasopressin occured i n amniotic f l u i d i n sufficient quantities to e l i c i t an in vitro response. Although i t i s often dangerous to apply data from one species to another, the fact that prolactin occurs i n human amniotic f l u i d i n amounts similiar to those used here (i.e. 10 ug/ml) suggests that the responses of the guinea-pig amnion may prove to be of physiological significance. The results obtained are in general agreement with 86 the hydro-mineral effects observed in some teleost fishes. Prolactin has been found to reduce net influx of water i n the stickleback by reducing osmotic permeability of the g i l l to water (Lam, 1 9 6 9 ) . It i s able to increase urine flow i n some fish by either decreasing reabsorption of water by the renal tubules, or by increasing glomerular f i l t r a t i o n rate (Stanley and Fleming, 1 9 6 ? ) . Utida et a l . ( 1 9 7 2 ) have shown that injection of prolactin into the sea-water adapted eel, Anguilla japonica decreases water reabsorption across the isolated intestine. In particular the results with the amnion agree with the results of Johnson et a l . ( 1 9 7 2 ) and Johnson ( 1 9 7 3 ) . These workers found that prolactin produced a decrease i n water efflux from the isolated urinary bladder of the starry flounder (Platichthys stellatus); the 6 0 fo decrease i n net flux of water (Johnson, 1 9 7 3 ) was close to the average decrease of 5 9 * 4 % found during the f i n a l hour of the amnion experiments reported here. The main difference in the two sets of experiments i s that prolactin did not act i n vitro in the fi s h studies, but had to be injected into the whole animal prior to the experiment. Maximal effects i n decreasing the permeability of the flounder bladder did not occur u n t i l two days after injection of prolactin, although i t lasted for about four days (Johnson et a l . , . 87 1974). At least twelve hours were required i n order to observe any significant change i n permeability. However, prolactin i s known to act i n vitro i n other preparations, such as the toad bladder (Snart and Dalton, 1973) so that the sensitivity of the isolated amnion i s not unreasonable. Perhaps the strongest evidence that the effects of prolactin noted here are genuine was provided by the very recent find-ings of Josimovich and Merisko (1975) which were announced at the 22nd Annual Meeting of the Society for Gynecological Investigation i n March of this year. They found that water shifts could be induced between the amniotic f l u i d and the fe t a l rhesus monkey within 2 hours after injection of pro-l a c t i n into the amniotic sac. The most striking effect of prolactin in the guinea-pig study reported here was shown on the net flux of water, where, on the average, flows were reduced by almost 60 % by the third hour after treatment. Occasionally values of zero were recorded. These effects were five to six times greater than in the radioisotope experiments. However, this discrepancy was not entirely unexpected, for a number of reasons. The dose of prolactin (20 ug/ml) given i n the net flow experiments was twice that i n the radioisotope studies, so that a larger response was reasonable. However, the apparent log dose/response curve for the action of prolactin on water movement i n other tissues i s nowhere 88 near as steep as this discrepency would require (see data of Johnson et a l . , 1 9 7 2 ) . It i s possible that part of the problem reflects differences in the mechanics of diffusional and bulk flow, as were noted i n the studies with vasopressin (see page 5 4 ) . Therefore, the relatively small changes seen in the radioisotope studies, and the discrepancy between the unidirectional and net flow experiments, are less important than the fact that con-sistent responses were produced by prolactin, even in terms of diffusional flow. In any case, i t i s worth pointing out, that i t i s the net flow, which was strongly influenced by prolactin, which has the greater revelance to the functioning of the fetus i t s e l f . 2 . The permeability of the amniotic membrane to sodium The permeability of the amniotic membrane of the guinea-pig to sodium was found to change markedly over the course of gestation. From day 57 to day 70 maternal-fetal sodium movement increases by a factor of about 3 5 . This agrees with the findings of Flexner and Gellhorn ( 1 9 4 2 ) . These workers found that the transfer rate of radioactive sodium into amniotic f l u i d of the guinea-pig increases by a factor of between 20 and 30 fold from the f i r s t third of pregnancy to term. Although increases of up to 70 fold have been recorded for sodium movement across human placenta 89 from week 9 to 36 (Flexner et a l . , 1 9 4 8 ) , permeability changes of isolated human amnion to sodium seem less dramatic (see Lind et a l . , 1972). This may be a result of the fact that only two stages of gestation were studied in the human investigation (viz. at O.36 term and term). The values given here for sodium movement across the guinea-pig amnion are very similiar at 30 ( 0 . 4 4 term) and 68 days (term), and therefore, the selection of these two periods would not indicate that any clear permeability changes took place. If one assumes that the length of gestation in the guinea-pig i s between 65 days (Draper, 1920) and 69 days (Illingworth et a l . , 1974) one can compare human values with guinea-pig values for sodium movement across amnion at term. Transfer rate for sodium across the isolated amniotic membrane of the guinea-pig was 20 uEq/cm /hr at 64 days and 22 uEq/cm /hr at 68 days. p These values are comparable to the mean of 16.3 uEq/cm /hr reported for sodium movement across isolated human amnion at term (Lind at a l . , 1 9 7 2 ) . It i s quite d i f f i c u l t to hypothesize what might be responsible for the observed changes in permeability of guinea-pig amnion to sodium without knowledge of possible ultrastructural changes occuring. Only the human amnion has been well studied at the ultrastructural level so that 90 considerations of the guinea-pig amnion are very specu-lative. If one makes the assumption that guinea-pig amnion may be similiar to human amnion then there may be a morphological basis for the physiological changes noted in the present investigation. Hoyes (1968) demonstrated that there were sites of open communication between the intercellular spaces and the amniotic cavity, and that the time of formation of the villous folds and their change in width seemed to be correlated with total volume of amniotic f l u i d . The villous folds appear at the time of maximum accumulation of amniotic f l u i d . If one associates the appearance of villous folds with an increase in permeability then this might explain why the guinea-pig amnion showed high fetal-maternal permeability to sodium in early gestation (i*e. at 30 gays). Closure of the intercellular spaces i n human amnion occured at a time when the high volume of amniotic f l u i d i s being maintained. If closure of the spaces would decrease permeability to sodium, then the slow movement of sodium across the guinea-pig amnion between days 4 5 - 6 0 might also have a morpho-logical basis. In the last weeks of human pregnancy the intercellular spaces become wider, and this seems to be related to the reduction i n the to$al volume of f l u i d at this time. This observation might also explain the increased 91 permeability of guinea-pig amnion between day 6 0 and term. The large increase noted for membranes that were judged to be 70 days old might represent merely a degeneration of the membrane and concomittant loss of selectivity. The changes i n rates of sodium movement have been re-ferred to as due to permeability changes rather than being due to changes i n an active pump. Although no attempt was made to determine whether sodium movement was active, i t was assumed to be passive for a number of reasons. F i r s t l y , Scoggin et a l . (1964) and Lind et a l . ( 1 9 7 2 ) could find no evidence of active sodium transport across human amnion i n vitro. Secondly, Mellor ( I 9 6 9 ) could measure no potential difference across the amnion, the uterine wall, or the yolk sac splanchnopleur of the guinea-pig i n vitro. The nega-tive potential difference of guinea-pig amniotic f l u i d ap-pears to arise from electromotive forces contributed by the f e t a l gastric mucosa and the placenta. The secretion of HC1 by the fe t a l stomach into the amniotic sac probably accounts for the high chloride concentration of amniotic f l u i d . The potential difference of guinea-pig amniotic f l u i d decreases by over 50 $ from day 6 0 to term. Mellor ( 1 9 6 9 ) attributed this decrease to either a degeneration of the electrogenic ion pump or to an increase i n ionic permeability that would short-circuit the pump. The second explanation seems to 92 be i n agreement with the findings here, where the per-meability of the amnion to sodium increases on about the 6oth day of gestation (see page 79 ) . 3. The action of prolactin on sodium flux through the  amnion The effect of prolactin in stimulating sodium movement across the amniotic membrane i s i n agreement with results obtained i n lower vertebrates. Prolactin promotes survival of some hypophysectomized f i s h i n fresh water by main-tenance of normal plasma osmolality and concentration of NaCl in blood (Ball and Ensor, 1965; Pickford et a l . , 1970; Johnson et a l . , 197*0. Prolactin stimulates an active uptake of sodium by the g i l l s of hypophysectomized Fundulas  kansae (Ball, 1969). In the stickleback, injection of prolactin causes an increase in urine flow and a decrease in urine sodium (Lam, 1969). Utida et a l . , (1971) found that prolactin decreases water reabsorption across the isot lated intestine of Anguilla japonica, while at the same time, promoting reabsorption of NaCl. Johnson et a l . (1974) determined that a single injection of prolactin into the starry flounder (Platichthys stellatus) caused an increase in sodium absorption by the isolated urinary bladder, and a decrease i n urine osmolality an<& sodium concentration. Effects for sodium movement across the 93 bladder were not observed u n t i l 24 hours after injection. Maximal effects were noted at about 96 hours after injection when sodium movement increased by 101 fo, As i n the case of prolactin's effect on water movement, there was a much shorter lag period for hormone action on sodium movement across the amnion than i s noted for tissues of fi s h . But again i t might be mentioned in defense of the present results that Snart and Dalton (1973) found that sodium movement across the i n vitro toad bladder preparation increases about 30 minutes after addition of prolactin to saline bathing the membrane. There i s some indication that prolactin might play a role i n regulating hydro-mineral balance in the adult mam-mal. Mainoya et a l . (1974) demonstrated that injection of the rat, the hamster, and the guinea-pig with 1 mg of ovine prolactin stimulated enhancement of f l u i d and NaCl absorp-tion across the isolated jejunum. Prolactin was found to increase the mucosal Na + transfer by about 110 $ across the jejunum of the rat, i f experimental and control values are compared. The 4 l , 6 fo increase in sodium movement across the jejunum of the adult guinea-pig i s i n agreement with the 3 8 « 5 % increase in sodium movement across the amnion after prolactin treatment. The results given on prolactin's effect on water and ion movement across the intestine of 94 the adult mammal are suggestive of hormone sensitivity. Whether or not this would represent a physiological action of prolactin i n the adult remains to be proven, however, since rather massive doses of the hormone were required to e l i c i t a response. Since Mainoya et a l . ( 1 9 7 4 ) injected each rat with at least 1 mg of prolactin, levels of the hormone in the blood of a 2 5 0 g rat could approximate 6 0 ug/ml, i f one assumes complete entry of the hormone into the circulatory system and a total blood volume of 1 6 . 7 ml. According to Bast and Melampy ( 1 9 7 2 ) the total concen-tration of prolactin i n the blood of the rat amounts to about 48 ng/ml at diestrous. Making the assumptions.men-tioned above for prolactin injection, the dose administered by Mainoya et a l . ( 1 9 7 4 ) could approach 1 0 0 0 times the physiological level i n the blood. In contrast to this, the dose of prolactin ( 1 0 ug/ml) used i n the present study has been detected in the amniotic f l u i d of some species of mammals (Priesen et a l . , 1 9 7 2 ) . It may be that the tissues of the fe t a l mammal are sensitive to prolactin's hydro-mineral control, but that the sensitivity decreases with age. The sensitivity of the adult intestine to prolactin, i n a gross sense, may represent a carry over from f e t a l development that has lost i t s relevance to the physio-logical well being of the organism as an adult mammal. SECTION III PRELIMINARY EXPERIMENTS STUDYING VASOPRESSIN'S EFFECT ON  OTHER FETAL TISSUES INTRODUCTION! The studies of Vizsolyi and Perks ( 1 9 7 4 ) were under-taken in an attempt to determine the significance of high levels of neurohypophyseal hormones i n the blood of the feta l mammal. Although no detectable levels of vaso-pressin have been found in amniotic f l u i d to date, very high levels have been found i n fet a l circulation. During the last trimester of pregnancy i n the rhesus monkey, the neurohypophysis of the fetus maintains a high rate of vasopressin secretion (Skowsky et a l . , 1 9 7 3 ) . Not only does the fe t a l secretory rate of vasopressin exceed the maternal rate, but also the osmolar and volume receptors controlling vasopressin release are functional at birth (Fisher et a l . , 1 9 & 3 ) . This suggests that although vaso-pressin may not be able, by i t s e l f , to exert a direct effect on the amniotic membrane of the guinea-pig, that i t could influence water and ion movement across fe t a l tissues in contact with the circulatory system. 95 96 1. The effect of vasopressin on water flux through the isolated f e t a l bladder Introduction! It was mentioned that bladder urine of the guinea-pig i s hypertonic to kidney urine (Kleinman, 1970). The bladder of the adult mammal acts as a storage organ for urine but seems to have no reabsorptive capacity. This i s i n contrast to the urinary bladder of some anuran amphibians and f i s h . These animals are able to reabsorb water and/or ions across the bladder wall, and regulate the reabsorptive process hormonally. Since the feta l mammal has been found to have certain characteristics i n common with lower vertebrates i t seemed resonable that water reabsorption by the urinary bladder of the fetal guinea-pig might be under endocrine control. Results: In order to test this hypophysis urinary bladders were removed from fet a l guinea-pigs ranging i n age from 0.75 to 0.91 term, and set up in a perfusion chamber. The bladders were bathed on their serosal side by maternal saline and on their mucosal side by amniotic saline. At the end of a 60 minute equilibration period vasopressin 97 ( 1 0 0 mU/ml) was added to the serosal side of three bladders, and was found to increase water flux by almost 50 $ at the end of an hour. The addition of sodium acetate to a control preparation, however, was followed by less than a 10 $ rise i n water flow (see Table II and Fig. 1 0 ) . TABLE II THE EFFECT OF VASOPRESSIN ON UNIDIRECTIONAL MUCOSAL-SEROSAL WATER MOVEMENT THROUGH THE ISOLATED URINARY BLADDER OF THE FETAL GUINEA-PIG Treatment Number of % increase in water preparations movement in 6 0 min sodium acetate 1 9 . 7 1 0 0 mU/ml vasopressin 3 4 9 . 4 + 1 7 . 8 a a _ x + S.E.M. Discussioni Although the results are only preliminary, and should be verified by further experiments, there i s some indication that vasopressin can influence water movement across the isolated urinary bladder of the f e t a l guinea-pig. If such a role for vasopressin should prove to be physiological then the hormone could aid water conservation i n the f e t a l guinea-pig by promoting reabsorption of water across the bladder wall into the blood. Effects of vasopressin on the fe t a l bladder of the lamb had been noted previously by France 98 Figure 1 0 . An Example of the Effect of Vasopressin on Percentage Increase in Unidirectional Water Flux through the Urinary Bladder of the Fetal Guinea-Pig At the origin, the isolated urinary bladders received either 100 mU/ml of vasopressin (N = 1 ) or sodium acetate (N = 1 ) on their serosal side. Water fluxes were measured by. the use of t r i t i a t e d water. The period just prior to hormone addition (i.e. 6 0 min) was taken as the "base lin e " value (0 fo change) from which percentage change i n water flux was cal-culated after the addition of sodium acetate or vaso-pressin. One membrane from the three described i n Table II was chosen as representative of the effect of vasopressin on water movement across the fe t a l bladder. Ordinatesj percentage change i n mucosal-serosal water movement from the "base lin e " value. Absissaet time from addition of sodium acetate or vasopressin, i n minutes. T I M E 1 M I N U T E S 100 et a l . ( 1 9 7 2 ) . These workers found that addition of the hor-mone to the serosal side of the isolated urinary bladder influenced the flux of isotopic ions across this organ. Vaso-pressin caused a net influx of sodium across the bladder of early fetuses and a net efflux i n late fetuses. The natrif-eric response noted by Prance et a l . ( 1 9 7 2 ) , and the water re-sorptive response described here for fet a l urinary bladder to neurohypophyseal hormone i s typical of some anuran amphib-ians (see Bentley, 1 9 6 6 * and Turner and Bagnara, 1 9 7 1 ) . The fact the frog bladder i s more sensitive to arginine vasotocin than to arginine vasopressin suggests the pos-s i b i l i t y that the feta l bladder may also be more sensitive to vasotocin. This may provide a role for the high levels of arginine vasotocin found in the pituitary of fetal mammals by Vizsolyi and Perks ( 1 9 6 9 ) . During i t s early "aquatic" existence the fetal mammal may resemble frogs and toads in i t s a b i l i t y to regulate water and ion movement across the urinary bladder by secretion of arginine vasotocin. The fact that the fetal urinary bladder demonstrates hormone sensitivity suggests a possible role for prolactin, since this hormone regulates hydro-mineral movement across the bladder wall of certain teleost fishes. The possible roles of arginine vasotocin and prolactin i n an endocrine control of the feta l urinary bladder should be investigated in 101 future experiments. 2. The effect of vasopressin on water flux through isolated skin of the feta l guinea-pig Introductioni Ultrastructural studies of human feta l skin suggests that i t i s capable of either secretion or absorption (Breathnach and Wyllie, 1965; Hoyes, 1967, 1968 a). Unlike adult skin, f e t a l skin i s very permeable to water prior to midterm (Lloyd et a l . , 1969* and Seeds, 1972 a). Lind et a l . (1972) demonstrated that Na+, as well as water, can readily diffuse through human fe t a l skin prior to keratin-ization. The fact that Lind and Hytten (1970) describe the periderm ce l l s of the human fetus as having a strong resemblance to renal tubule ce l l s under the influence of vasopressin suggested a possible hydro-mineral effect of neurohypophyseal hormones. Results 1 Skin was removed from the back region of feta l guinea-pigs ranging i n age from 0 . 4 6 term to 0.51 term by sharp dissection, and was placed i n a perfusion chamber, where i t was bathed by amniotic and maternal salines. At the end of a 90 minute equilibration period vasopressin (500 mU/ml and 1000 mU/ml was added to the amniotic saline bathing 102 the serosal side of four preparations, and was found to increase water flux across f e t a l skin (see Table III and Fig. 1 1 ) . TABLE III THE EFFECT OF VASOPRESSIN ON UNIDIRECTIONAL SEROSAL-MUCOSAL WATER MOVEMENT THROUGH THE ISOLATED SKIN OF THE FETAL GUINEA-PIG Treatment Number of preparations % increase i n water movement i n 6 0 min 5 0 0 mU/ml vasopressin 1 0 0 0 mU/ml vasopressin 2 2 14.3 + 8 . 9 a 3 0 . 4 + 1 3 . 7 a _ x + S.E.M. Histological sections were prepared from portions of skin removed from the back region of each animal used in this study. Sections for histological study were taken i n close proximity to those used i n the transport studies, and were stained with Chevremont-Frederick's stain for SH groups and Mallorie's stain. In this way one could determine whether or not skin used in the flux experiments was keratinized. There was no indication of keratinization. Discussiont Although more experiments are needed, including controls, there seems to be some indication of hormonal influence on serosal-mucosal water movement across isolated skin of the f e t a l guinea-pig. The fact that doubling the dose of vasopressin seemed to produce a doubling of the 103 Figure 11. Examples of the Effect of Vasopressin on Percentage Increase in Unidirectional Water Flux through Skin of the Fetal Guinea-Pig At the origin, f e t a l skin received either 500 mU/ml (N = 1) or 1000 mU/ml (N = 1) of vasopressin on their serosal side. Water fluxes were measured by the use of tr i t i a t e d water. The period just prior to hormone addition (i.e. 90 min) was taken as the "base li n e " value (0 % change) from which percentage change in water flux was calculated after the addition of vasopressin. Two membranes from the four described in Table III were chosen as representative of the effect of vasopressin on water movement through isolated fe-t a l skin. Ordinatesi percentage increase i n serosal-mucosal water movement from the "base lin e " value. Absissaei time from addition of vasopressin, in min-utes. T I M E ; M I N U T E S 1 0 5 response tends to support the hypothesis that fe t a l skin i s sensitive to hormone at the gestational ages studied. If these findings prove to he of physiological significance to the fet a l guinea-pig, then the fetus may be able to increase water reabsorption from the amniotic sac across i t s skin by secretion of vasopressin i n response to fe t a l dehydration. A short note appearing after the experiments mentioned above indicated that vasopressin could stimulate an increase of the influx of sodium across the skin of fe t a l sheep (France, 1 9 7 4 ) . Thus, the present findings with regards to vasopressin control of water movement across fetal skin would seem reasonable. Since the f e t a l skin i s bathed with amniotic f l u i d containing high concentrations of prolactin in some species, i t would be interesting to investigate a possible control of hydro-mineral movement across f e t a l skin by prolactin, in view of hormone-sensitivity of fet a l guinea-pig skin. GENERAL DISCUSSION The experiments described in the preceeding sections of this thesis suggest the possibility that vasopressin and prolactin can influence water and sodium movement across isolated membranes and tissues of the fet a l guinea-pig. Although a number of hormonal effects have been noted that parallel the actions seen in other vertebrates, i t i s prob-ably very dangerous at this time to assume that the find-ings i n the guinea-pig study w i l l apply to other species of fetal mammals. The results described in section I indicate that vaso-pressin can influence water movement through the amniotic membrane. In the subsequent discussion of hormonal influence on the amniotic membrane, however, only prolactin w i l l be dealt with i n detail. It i s s t i l l too early to determine whether or not vasopressin can be released into the amniotic sac of the fet a l mammal in amounts capable of exerting an in vivo response. It may be that i f vasopressin i s found in the amniotic f l u i d of the"fetal guinea-pig in amounts too small to cause an effect on i t s own, that i t may be acting in synergism with another hormone (e.g. prolactin). This i s reasonable i n view of the findings of Horrobin et a. a l . (1973) that prolactin can act in synergism with vaso-pressin i n the production of antidiuresis in Merino ewes. 106 107 1. The physiological significance of the effects of  prolactin on the amnion At the present time, the in vitro effects of prolactin on the amnion must be regarded as pharmacological. However, there i s a strong possibility that they may prove to be physiological. Studies on human amniotic f l u i d , from early stages of pregnancy, have shown unusually high levels of prolactin, sometimes 227 times higher than those of maternal serum (Friesen et a l . , 1972; Josimovich, 1973; Josimovich et a l . , 1974; Parke, 1973). The values range from 1.2 to 10 ug/ml at 20 weeks, and this last dose was shown to be capable of having a positive effect on the isolated amnion, in the present experiments. Nevertheless, the possible importance of prolactin i n normal amniotic control must be a matter of speculation. Certainly the action of prolactin in slowing fetal-maternal water flow could partly aid i n allowing amniotic f l u i d to build up. The actual significance of maintaining a certain volume of liquor amnii i s not well understood. Evidence that a certain volume must be maintained i s demonstrated by the fact that removal of the f l u i d at midterm w i l l cause death of the fetus (Adolph, 1967). Some of the proposed roles for amniotic f l u i d include the followingi 1. It provides thermal insulation and serves as a cushion to 108 protect the fetus from shock and abrasion by the uterine wall (Reynolds, 1972). 2. It may act as a moist spacious environment in which the fetus can grow in a relatively weightless condition. 3 . Amniotic f l u i d may play a role i n causing distention:of the uterus, which may be necessary for not only fe t a l development, but also uterine and placental development (Kerpel-Fronius, 1 9 7 0 ) . 4 . Liquor amnii may act as a safeguard against fetal water imbalance, so that i f large quantities of water accumulate in the fetus excess water may be diverted to the amniotic sac. Water may also be removed from the amniotic sac i n the case of fetal dehydration (Seeds,19^5). Bruns et a l . (1963) showed that experimental dehydration of the fe t a l rabbit caused a decrease in the volume of amniotic f l u i d , sug-gesting that water had been transferred from the amniotic compartment to the fetal compartment. It was mentioned i n the general introduction that retention of water i n the amniotic sac i s d i f f i c u l t to explain; the existing osmotic, colloid osmotic, and hydro-static gradients, as well as Staverman (reflection) coefficients for solutes, a l l tend to favor the passage of water from the amniotic compartment to the mother (Seeds, 1973). The action of vasopressin i n promoting water move-ment into the sac, and the a b i l i t y of prolactin to slow 109 water loss from i t , could help to explain the enigma of how amniotic f l u i d volume i s maintained. This i s not meant to imply that the fetus i s able to directly control amniotic f l u i d volume by secretion of prolactin into the amniotic sac. On the contrary, i t appears that in the rhesus monkey the maternal circulation may be the major source of amniotic f l u i d prolactin (Josimovich et a l . , 1974). No attempt has been made yet to determine whether or not the mother i s able to regulate the release of prolactin into the amniotic compartment in response to stimuli such as change in osmotic pressure of amniotic f l u i d . Prolactin deposition into amniotic f l u i d may occur for reasons other than the physiological control of hydro-mineral balance. The net result, however, would s t i l l be a slowing of water flow out of the amniotic sac. Although the amount of prolactin present in human amniotic f l u i d decreases over the course of gestation (Friesen et a l . , 1972), sensitivity of the amniotic membrane to hormonal stimulation seems to increase in apparently "overdue" membranes (see results, page 65) so that low levels don't rule out the possibility of a response to hormone at this stage of gestation. In addition, i t i s possible that prolactin affects the levels of certain solutes in amniotic f l u i d . Isotopically labeled nitrogenous solutes such as creatinine have a 110 rapid and extensive bidirectional exchange between the three intrauterine compartments of the pregnant rhesus monkey (Pitkin and Reynolds, 1975). In the f i r s t half of human pregnancy, creatinine i s present in amniotic f l u i d at approximately the same level as that of maternal plasma. At about midterm there i s a gradual increase in creatinine concentration, and between 34 to 37 weeks there i s a rapid increase so that levels are 2 to 4 times higher than those of early stages of gestation (Reynolds et a l . , 1954). The relationship between amniotic f l u i d creatinine and gesta-tional age has lead to use of amniotic f l u i d creatinine level as an index of fetal maturity (Gauthier et al.» 1972). Although urea can be rapidly exchanged between amniotic f l u i d and maternal circulation, i t accumulates in amniotic f l u i d near term to levels 3 to 4 times higher than found in the fetus or mother (Reynolds, 1972). How solutes like creatinine and urea accumulate in the amniotic compartment, when the amnion i s relatively permeable to their egress, has been an enigma. Leaf and Hays (19&2) found a close parallel i n the passage of water and urea through the toad bladder, and both were influenced by vasopressin. Seeds (1973) describes the placental membranes as having significant porous channels through which there i s a bulk flow of I l l solvent water in a fashion similar to that for toad bladder. Thus, the action of prolactin (and perhaps, vasopressin) might lead to a slowing of the movement of urea out of the amniotic sac of the guinea-pig. This hypothesis needs to be tested, however, since i n tissues other than the toad bladder, urea and water follow different routes. Rocha and Kokko (1974) found that the pathways of urea movement are not the same as the principal pathways of water movement across the papillary collecting duct epithelium of the rabbit. Vasopressin was found to have no effect on increasing permeability of the nephron to urea. According to Hays and Levine (1974), the solvent drag effect noted in the toad bladder for water and urea movement may be the result of inadequate s t i r r i n g in conventional diffusion chambers. There i s no apparent solvent drag effect noted for acetamide, an amide closely related to urea, when toad bladder i s studied in mechan-i c a l l y stirred chambers. The apparent solvent drag effect may have been due to the accumulation of isotopically labeled solute in unstirred layers near the luminal membrane, when water moves from the mucosal to the serosal side of the toad bladder. Studies should, thus, be carried out in order to determine whether or not prolactin can slow the fetal-maternal movement of isotopically labeled 112 nitrogenous solutes across the amniotic membrane i n a chamber with adequate st i r r i n g . Although no significant changes in maternal serum sodium have been observed during the period of pregnancy, alterations have been reported for concentration of sodium in amniotic f l u i d . Gillibrand (I969 b) found that there was a very significant decrease i n amniotic f l u i d sodium concentration as gestation advanced in the human being. Between 38 and 44 weeks there was a mean d e f i c i t of 8 . 8 mEq/l in amniotic f l u i d compared to maternal serum sodium. In other species, such as the rat, the rabbit, and the guinea-pig, Na + distribution between amniotic f l u i d and maternal plasma i s not according to electrochemical equilibriums(Mellor, 1 9 6 9 ) . In the rat the observed amniotic f l u i d sodium concentration i s higher than the calculated value ( 0 . 8 8 v.s. 0.57)# and in the rabbit and guinea-pig, the abserved concentration i s lower than the calculated value (O.89 v.s. 1.41 and O.92 v.s. 6 . 2 5 respectively). Sensitivity of the amniotic membrane's permeability to sodium under action of prolactin might explain the decrease in sodium concentration in guinea-pig amniotic f l u i d with advancing gestation as reported by Bates (1963) and Mellor ( 1 9 6 9 ) . The results of the present study tend to suggest that prolactin can increase 113 fetal-maternal sodium movement while not affecting maternal-fe t a l movement. Sensitivity of the membrane to hormonal stimulation seemed to increase up to at least 57 days of gestation. This would tend to cause a net movement of sodium out of the amniotic sac and thus decrease the concentration of sodium i n amniotic f l u i d . It i s concluded that prolactin can cause slowing of unidirectional and net flux of water in the fetal-maternal direction across the isolated guinea-pig amnion, since the doses required have been shown to exist i n some amniotic fluids. Prolactin may be partly responsible for the retention of water, and perhaps other substances, within the amniotic sac of the intact mammal. There i s a possibility that prolactin may also interact with other hormones present in amniotic f l u i d . According to Lam (1972) prolactin may require some pituitary synergist(s) for i t s f u l l action on the renal tubules of the eel, Anguilla anguilla. Although vasopressin may not be found in amniotic f l u i d of the guinea-pig i n high enough concentration to evoke responses observed i n the i n vitro studies with amnion, i t may act in synergism with a hormone such as prolactin. The trend from isotonicity to hypo-tonicity of amniotic f l u i d and decrease in amniotic f l u i d sodium concentration, with respect to maternal serum, occurs in spite of the fact that there i s a free and rapid 114 movement of various ions and nitrogenous waste products i n and out of the amniotic sac (Reynolds, 1972). Pro-l a c t i n may be responsible for increasing fetal-maternal sodium movement as gestation advances i n the guinea-pig. Therefore permeability studies on the isolated amnion, in normal salines, free of hormone may give a false impression of the i n vivo situation. Finally, there i s some indication that responsive-ness of guinea-pig amnion to prolactin and/or other hor-mones present in amniotic f l u i d may help to in i t i a t e birth. According to some workers such as Schwarz et a l . (1975 a-K the fe t a l membranes may be the key to the i n i t i a t i o n of labour. Indication for such a role comes from the finding that not only does the amniotic f l u i d contain quite large quantities of Cortisol (Murphy et a l . , 1975). but also the human fe t a l membranes contain steroid receptors (Schwarz et a l . , 1975 b). Steroids such as Cortisol have been implicated i n the i n i t i a t i o n of human labour. Arachidonic acid, the precursor of prostaglandin F 2 Q L ( p G F 2 o ^ h a s keen isolated from the human chorioamnion (Schultz et a l . , 1975 a). Partuition in the guinea-pig does not seem to be under the control of progesterone, estrogen, cortico-steroids or oxytocin (Illingworth et a l . , 1974). Partuition can be induced, however, by infusion of guinea-pigs with 115 P G F 2 o ? P G E 2 * O R I * C * 1 , 80,996 (a potent analogue of P G F 2 d ^ * s i n c e arachidonic acid present in the amniotic membrane can form PGF 2 q L, which i s elevated during human labour (Schultz et a l . , 1975 b), such a conversion may-act as a trigger for partuition in the guinea-pig. Prolactin, which i s known to enhance l i p i d and protein biosynthesis (Winter et a l . , 1975) might be speculated to stimulate an increase in the fatty acid concentration (e.g. arachidonic acid) i n fe t a l membranes, and thereby i n i t i a t e a series of reactions leading to the induction of labour i n the guinea-pig. Strauss et a l . (1975) found that treatment of pregnant rats with PGF2oL o n d a v s 19 and 20 of pregnancy resulted i n premature delivery on day 21 . Bast and Melampy (1972) discovered that serum prolactin levels i n pregnant rats doubled on day 20 and tripled on day 21 of pregnancy, indicating that prolactin might play a role in i n i t i a t i n g labour. Such a speculated role for prolactin in the guinea-pig deserves attention. I t may be that the radical change i n membrane permeability to sodium for overdue amnions may be related to a birth response. The present experiments have suggested that prolactin may combine i t s reproductive functions i n mammals with i t s hydro-mineral a c t i v i t i e s i n lower vertebrates by acting 116 on the amniotic membrane of the fet a l mammal. Preliminary experiments outlined in section III indicate that f e t a l guinea-pigs show similarities to certain anuran amphibians in their a b i l i t y to influence water movement across the urinary bladder and the skin surface. Further studies are required to elaborate the role of hormones i n controlling fe t a l hydro-mineral balance. High levels of vasopressin in f e t a l blood may also exert an influence on water and/or ion movement through the yolk sac of the guinea-pig,.since unlike the amnion, this membrane i s i n contact with the fetal circulatory system. Renfree et a l . ( 1 9 7 5 ) suggest that the yolk sac membrane may be more important than the amniotic membrane in regulating the volume and the bio-chemical composition of amniotic f l u i d i n the mouse.: An investigation of possible endocrine influence on the yolk sac membrane i s l e f t for future study. In conclusion, i t appears that hormonal control of fet a l hydro-mineral metabolism may help to explain the enigma of how fluids accumulate in the intrauterine compartments, and how hydro-osmotic homeostasis in the internal mileu i s achieved. REFERENCES Abramovich, D.R. ( 1 9 7 0 ) . Fetal factors influencing the volume and composition of liquor amnii. J. Obstet. Gynaec. Bri t . Cwlth. 21* 8 6 5 - 8 7 7 . Abramovich, D.R. ( 1 9 7 3 )• The volume of amniotic f l u i d and factors affecting this. In "Amniotic Fluid1 Research and Cl i n i c a l Application" (D.V.I. Fairweather and T.K.A.B. Eskes, eds.), pp. 2 9 - 5 1 • Excerpta Medica, Amsterdam. Adams, F., Desilets, D.T., and Towers, B. ( I 9 6 7 ) . Physiology of the fe t a l larynx and lung. Ann. Otol. Rhinol. Lar. 26, 7 3 5 - 7 4 3 . Adamson, T.M., Boyd, R.D.H., Pratt, H.S., and Strang, L.B. ( I 9 6 9 ) . Composition of alveolar liquid in the fetal lamb. J. Physiol. 204, 159-168. Adamson, T.M., Brodecky, V., Lambert, T.F., Maloney, J.E., Rithi, B.C., and Walker, A. ( 1 9 7 3 ) . The production and composition of lung liquid i n the i n utero foetal lamb. In "Foetal and Neonatal Physiology1 Proceedings of the STr Joseph Barcroft Centenary Symposium" (K.S. Comline, K.W. Cross, G.S. Dawes, and P.W. Nathielsz, eds.), pp. 208 -213. Cambridge University Press, Cambridge. Adolph, E.F. ( 1 9 6 7 ) . Ontogeny of volume regulations in embryonic extracellular fluids. Quart. Rev. Biol. 42, 1 - 3 9 . Alexander, D.P., and Nixon, D.A. ( I 9 6 I ) . The foetal kidney. Brit. Med. Bull. 12t 1 1 2 - 1 1 7 . Alexander, D.P., Nixon, D.A., Widdas, W.F., and Wahlzogen, F.X. ( 1 9 5 8 ) . Renal function i n the sheep foetus. J. Physiol. 140, 14 -22 . Alexander, G., and Williams, D. (1968). Hormonal control of amniotic and allantoic f l u i d volume i n ovariectomized sheep. J. Endocrinol. 41, 477-48$. Arthur, G.H. ( 1 9 6 9 ) . The feta l fluids of domestic animals. J. Reprod. Fert. Suppl. % 4 5 - 5 2 . B a l l , J.N. ( 1 9 6 9 ) . Prolactin and osmoregulation in teleost fishesi a review. Gen. Comp. Endocrinol. Suppl. 2 , 1 0 - 2 5 . 1 1 7 118 B a l l , J.N., and Ensor, G.M. (1965). Effect of prolactin on plasma sodium i n the teleost, Poecilia latipinna. J. Endocrinol. 32, 269-270. Barnes, A.C. (1968). "Intra-uterine Development,H pp. 129-1^2. Lea and Fabiger, Philadelphia. Basset, J.M., and Hinks, N.T. (1969). Micro-determination of corticosteroids i n ovine peripheral plasmaJ effects of venipuncture, corticotrophin, insulin and glucose. £• Endocrinol. 44, 387-403. Bast, J.D., Melampy, R.M. (1972). Luteinizing hormone, prolactin, and ovarian 20d-hydroxysteroid dehydrogenase levels during pregnancy and pseudopregnancy in the rat. Endocrinology $ 1 , 1499-1505. Bates, H.R. (I963). Comparison of pharyngeal and amniotic fluids i n the guinea-pig fetus. Amer. J. Obstet. Gynec. 85, 484-486. Battaglia, F.C., Prystowsky, H., Sissan, C , Hellegers, A.E., and Bruns, P.D. (i960). Effect of the administration of fluids intravenously to mothers upon the concentrations of water and electrolytes in plasma of human fetuses. Pediatrics. 25* 2-10. Bentley, P.J. (1966). The physiology of the urinary bladder of amphibia. Biol. Rev. Cambridge Phil. Soc. 41, 275-316". Bourne, G. (1962). "The Human Amnion and Chorion," pp. 135-141. Yearbook Medical Publishers, Chicago. Breathnach, A.S., and Wyllie, L.M. (1965). Fine structure of cells forming the surface of the epidermis i n human fetuses at 14 weeks. J. Invest. Derm. 45, 179-189. Bruns, P.D., Linder, R.O., Drose, V.E., and Battaglia, F.C. (1963). The placental transfer of water from the fetus to the mother following the intravenous infusion of hyper-tonic mannitol to the maternal rabbit. Amer. J. Obstet. Gynec. 86, I6O-I67. Carpek, K., Dlouha, H., Fernandez, J., and Popp, M. (1968). Regulation of proximal tubular reabsorption i n early post-natal period of infant rats« micropuncture study. Abst. Proc. Int. Union. Physiol. Sciences. Washington D.C. 7* 72. 119 Comline, R.S., Nathanielsz, P.W., Paisey, R.B., and Silver, M. (1970). Cortisol turnover in the sheep foetus immed-iately prior to partuition. J. Physiol. 2 1 0 , 141-142 p. Crosignani, P.G., Nencioni, T., Brambati, B., and Polvani, F. (1970). The prognostic value of HCG and HPL i n Rh-sensitized women. Abstracts of Proceedings Sec. Eur. Congr. Perinatal Med. London, 2 . Dancis, J., Brenner, M.A., and Money, W.L. ( I 9 6 2 ) . Some factors affecting the permeability of the guinea-pig placenta. Amer. J. Obstet. Gynaec. 84, 5 7 0 - 5 7 6 . Danforth, D.N., Hull, R.W. ( 1 9 5 8 ) . The microscopic anatomy of the fe t a l membranes with particular reference to the detailed structure of the amnion. Amer. J. Obstet. Gynec. 7 5 , 5 3 6 - 5 5 0 . " Diamond, J.M. ( 1 9 7 1 ) . Water-solute coupling and ion selec-t i v i t y i n epithelia. Phil. Trans. Roy. Soc. Lond. B. 2 6 2 , 141-151. Draper, R.L. (1920). The perinatal growth of the guinea-pig. Anat. Rec. 18, 3 6 9 - 3 9 2 . Edelman, CM. ( 1 9 6 7 ) . Maturation of the neonatal kidney. Proc. Third Int. Congr. Nephrol. Washington, 2* 1 - 1 2 . Edelman, CM., and Barnett, H.L. ( i 9 6 0 ) . Role of the kidney i n water metabolism i n young infants* physiological and c l i n i c a l considerations. J. Pediat. j>6, 154-179. E l l i o t t , P.M., and Inman, W.H.W. ( 1 9 6 1 ) . Volume of liquor amnii i n normal and abnormal pregnancy. Lancet. 2 , 835-840. Fisher, D.A., Pyle, H.R., Porter, J.C, Beard, A.G., and Panos, T.C ( 1 9 6 3 ) . Control of water balance in the new-born. Amer. J. Dist. Child. 1 0 6 , 137-145. Flexner, L.B., Cowie, D.B., Hellman, L.M., Wilde, W.S., and Vosburgh, J.G. (1948). The permeability of the human placenta to Na in normal and abnormal pregnancies and the supply of sodium to the human fetus as determined by : ^ radioactive Na . Amer. J. Obstet. Gynec. 5 5 , 469-480. Flexner, L.B., and Gellhorn, A. ( 1 9 4 2 ) . The transfer of water and sodium to the amniotic f l u i d of the guinea-pig. Amer. J. Physiol. 136, 757-761, 120 France, V.M. (1974). Sodium uptake by the skin of foetal sheep. J. Physiol. 27P. France, V.M., Saunders, N.R., and Stanier, M.W. ( 1 9 7 2 ) . Sodium transport in foetal sheep urinary bladder. J. Physiol. 224, 23P. Friesen, H., Hwang, P., Guyda, H., Tolis, G., Tyson, J., and Myers, R. (1972). A radioimmunoassay for human prolactin. In "Prolactin and Carcinogenesis" (A.R. Boyns and K. Griffiths ,eds.), pp. 6 4 - 8 0 . Alpha Omega Alpha, Cardiff. Gadd, R.L. ( 1 9 6 6 ) . The volume of liquor amnii in normal and abnormal pregnancies. J. Obstet. Gynaec. Brit. Cwlth. 21* H - 2 2 . Ganote, C.E., Grantham, J.J., Moses, H.L., Burg, M.B., and Orloff, J. (1968). Ultrastructural studies of vasopressin effect on isolated perfused renal collecting tubules of the rabbit. J. C e l l . Biol. 3 6 , 355-367. Gauthier, C , Desjardin, P., and McLean, F. (1972). Fetal maturityt amniotic f l u i d analysis correlated with neo-natal assessment. Amer. J. Obstet. Gynec. 112, 33^-350. Giles, H.R., Lox, CD., Heine, M.W., and Christian, CD. (1975)* Intrauterine fe t a l sex determination by radio-immunoassay of amniotic f l u i d testosterone. Sci. Abst. 2 1 s t . Ann. Meeting Soc. Gynec. Invest. j>, 2 5 . Gillibrand, P.N. (1969 a). The r.ate of water transfer from the amniotic sac with advancing pregnancy. J. Obstet. Brit. Cwlth. £6, 530-533. Gillibrand, P.N. (1969 b). Changes in the electrolytes, urea and osmolality of amniotic f l u i d with advancing pregnancy. J. Obstet. Gynaec. Brit. Cwlth. £ 6 , 8 9 8 - 9 0 5 . Gleiss, J. ( 1 9 7 0 ) . Morphological and functional development of the skin. In "Physiology of the' Perinatal Period" U. Stave, ed.77 pp. 8 8 9 - 9 0 6 . Appleton-Century-Crofts-Meredith Corp., New York. Gluck, L., Kulovich, M.V., Borer, R.C, Brenner, P.H., Anderson, G.G., and Spellacy, W.N. (1971). Diagnosis of respiratory distress syndrome by amniocentesis. Amer. J. Obstet. Gynec. 109, 440-445. 121 Harrison, R.G., and Malpas, P. ( 1 9 5 3 ) . The volume of amni-otic f l u i d . J. Obstet. Gynaec. Brit. Emp. 6 0 , 6 3 2 - 6 3 8 . Hays, R.M. ( 1 9 7 2 ) . The movement of water across vasopressin-sensitive epithelia. In "Current Topics i n Membranes and Transport" (F. Bonner and A. Kleinzeller, eds.), Vol. 3 , PP. 3 3 9 - 3 6 6 . Academic Press, New York. Hays, R.M., and Leaf, A. ( 1 9 6 2 ) . Studies on the movement of water through isolated toad bladder and i t s modification by vasopressin. J. Gen. Physiol. 4 5 , 9 0 5 - 9 1 9 . Hays, R.M., and Levine, S.D. ( 1 9 7 4 ) . Vasopressin. Kidney  International 6 , 3 0 7 - 3 2 2 . Hochachka? P.W., and Somero, G.N. ( 1 9 7 3 ) . "Strategies of Biochemical Adaptation," pp. 97-143. W.B. Saunders Company, Philadelphia. Horrobin, D.F., Manku, M.S., and Robertshaw, D. ( 1 9 7 3 ) . Water-losing action of antidiuretic hormone in the presence of excess Cortisolt restoration of normal action by prolactin or oxytocin. J. Endocrinol. 5 8 , 1 3 5 - 1 3 6 . Hoyes, A.D. ( 1 9 6 7 ) . Acid mucopolysaccharide in fetal epidermis. J. Invest. Derm. 4 8 , 5 9 8 -601 . Hoyes, A.D. ( 1 9 6 8 a). Electron microscopy of the surface layer (periderm) of human foetal skin. J. Anat. 1 0 3 i 3 2 1 - 3 3 6 . Hoyes, A.D. ( 1 9 6 8 b). Fine structure of human amniotic epithelium i n early pregnancy. J. Obstet. Gynaec. Brit. Cwlth. 21* 9 4 9 - 9 6 2 . Hoyes, A.D. ( 1 9 6 9 K Ultrastrueture of the epithelium of the human umbilical cord. J. Anat. 1 0 5 , 149-162. Hutchinson, D.L., Gray, M.J., Plentl, A.A., Alvarez, H., Caldeyro-Barcia, R., Kaplan, B., and Lind, T. ( 1 9 5 9 ) . The role of the fetus i n the water exchange of amniotic f l u i d of normal and hydramniotic patients. J. Clin. Invest. 9 7 1 - 9 8 1 . Hutchinson, D.L., Hunter, C.B., Nelson, E.D., and Plentl, A.A. ( 1 9 5 5 ) . The exchange of water and electrolytes i n the mechanism of amniotic f l u i d formation and relation to hydramnios. Surg. Gynec. Obstet. 1 0 0 , 3 9 1 - 3 9 6 . 122 Ibsen, H.L. (1928). Prenatal growth in guinea-pigs. J. Exp. Zool. $1, 5 1 - 9 4 . Illingworth, D.V., Challis, J.R.G., Ackland, N., Burton, A.M., Heap, R.B., and Perry, J.S. ( 1 9 7 4 ) . Partuition in the guinea-pig? plasma levels of steroid hormones, steroid-binding proteins, and oxytocin, and the effect of corticosteroids, prostaglandins and adrenocorticotrophin. £• Endocrinol. 6 3 , 557-570. Jeffcoate, T.N.A., and Scott, J.S. (1959). Polyhydramnios and oligohydramnios. Can. Med. Ass. J. 80, 77-86. Johnson, D.W. (1973). Endocrine control of hydromineral balance i n teleosts. Amer. Zool. 13, 799-818. Johnson, D.W., Hirano, T., Bern, H.A., and Conte, F.P. (1972). Hormonal control of water and sodium movements in the urinary bladder of the starry flounder, Platichthys  stellatus. Gen. Comp. Endocrinol. 19* 115-128. Johnson, D.W., Hirano, T., Sage, M., Foster, R.C., and Bern, H.A. ( 1 9 7 4 ) . Time course of response of starry flounder (Platichthys stellatus) urinary bladder to prolactin and salinity transfer. Gen. Comp. Endocrinol. 24, 373-380. j/rgensen, C.B. ( 1 9 6 8 ) . Chemical correlation and control. In "Animal Function* Principles and Adaptations" (M.S. Gor-don, ed.), pp. 484-547. MacMillan, New York. Josimovich, J.B. ( 1 9 7 1 ) . Placental lactogen hormone. In "Endocrinology of Pregnancy" (F. Fuchs and A. Klopper, eds.), pp. 184 -197. Harper and Row, New York. Josimovich, J.B. (1973). Hormones in amniotic f l u i d . In "Amniotic Fluidt Research and C l i n i c a l Application" (D.V.I. Fairweather and T.K.A.B. Eskes, eds.), pp. 212-222. Excerpta Medica, Amsterdam. Josimovich, J. and Marisko, K. (1975). Prolactin induced water shifts between amniotic f l u i d and fe t a l rhesus. Sci. Abst. 22nd. Ann. Meeting Soc. Gynec. Invest. 6 , 6 . Josimovich, J.B., Weiss, G., and Hutchinson, D. ( 1 9 7 4 ) . Sources and disposition of pituitary prolactin in maternal circulation, amniotic f l u i d , fetus and placenta in the pregnant rhesus monkey. Endocrinology ^ 4 , 1 3 6 4 - 1 3 7 1 . 123 Kerpel-Fronius, E. (1970). Electrolyte and water metabolism. In "Physiology of the Perinatal Period" (U. Stave, ed.), Vol. 2, pp. 643-679. Appleton-Century-Crofts-Meredith Corp., New York. Kerr, G.R., and Kennan, A.L. ( 1 9 6 9 ) . The free amino acids of amniotic f l u i d during pregnancy of the rhesus monkey. Amer J. Obstet. Gynec. 105, 3 6 3 - 3 6 8 . Kleinman, L.I. (1970). Physiology of the perinatal kidney. In "Physiology of the Perinatal Period" (U. Stave, ed.), Vol. 2 , pp. 6 7 9 - 7 0 2 . Appleton-Century-Crofts-Meredith Corp., New York. Klopper, A. (1972). The es t r i o l in liquor amnii. Amer. J. Obstet. Gynec. 112, 4 5 9 - 4 7 1 . Lahlou,^B., and Giordan, A. (1970). Le controle hormonal des echanges et de l a balance de l'eau chez le Te'leosteen d'eau douce Carassius auratus, intact et hypophysectomise'. Gen. Comp. Endocrinol. 14, 491 -509. Lam, T.J. ( 1 9 6 9 ) . Effects of prolactin on loss of solutes via the head region of the early-winter marine threespine stickleback (Gasterosteus aculeatus L., form trachurus) in fresh waterTTan. J. Zool. 47,"8*65-869. Lam T.J. (1972). Prolactin and hydromineral regulation in fishes. Gen. Comp. Endocrinol. Suppl. 2 , 3 2 8 - 3 3 8 . Leaf, A., and Hays, R.M. ( I 9 6 2 ) . Permeability of the iso-lated toad bladder to solutes and i t s modification by vasopressin. J. Gen. Physiol. 4 5 , 9 2 1 - 9 3 2 . Lind, T. (1973). The biochemistry of amniotic f l u i d . In "Amniotic Fluidt Research and Cl i n i c a l Application" (D.V.I. Fairweather and T.K.A.B. Eskes, eds.), pp. 6 l -81. Excerpta Medica, Amsterdam. Lind, T., and Hytten, F.E. (1970). Relationship of amniotic f l u i d volume to feta l weight in the f i r s t half of preg-nancy. Lancet 1, 1147-1149. Lind, T., and Hytten, F.E. (1972). Fetal control of fet a l fluids. In "Physiological Biochemistry of the FetusJ Proceedings of the International Symposium" (A.A. Hodari and F. Mariona, eds.), pp. 5 4 - 6 5 . Charles C. Thomas, Springfield. 124 Lind, T., Kendall, A., and Hytten, F.E. ( 1 9 7 2 ) . The role of the fetus in formation of amniotic f l u i d . J. Obstet. Gynaec. Bri t . Cwlth. 21* 2 8 9 - 2 9 8 . Lind, T., Parkin, F.M., and Cheyne, G.A. ( 1 9 6 9 ) . Biochemical and cytological changes in liquor amnii with advancing gestation. J. Obstet. Gynaec. Brit. Cwlth. ££» 6 7 3 - 6 8 3 . Lloyd, S.J., Garlidj K.D., Reba, R.C., and Seeds, A.E. ( 1 9 6 9 ) . Permeability of different layers of the human placenta to isotopic water. J. Appl. Physiol. 2 6 , 274-2 7 6 . MacDonald, M.S., and Emery, J.L. ( 1 9 5 9 ) . The late intra-uterine and postnatal development of human renal glom-er u l i . J. Anat. £ 3 , 3 3 1 - 3 4 0 . MacGillivray, I., and Buchanan, T.J. ( 1 9 5 8 ) . Total exchange-able sodium and potassium in nonpregnant women and in normal and preeclamptic pregnancy. Lancet 2 , IO9O-IO93. Maetz, J. ( I 9 6 8 ) . Salt and water metabolism. In "Perspec-tives in Endocrinology« Hormones in the Lives of Lower Vertebrates" (E.J.W. Barrington and G.B. Jj^rgensen, eds.), pp. 4 7 - 1 6 2 . Academic Press, New York. Mainoya, J.R., Bern, H.A., and Regan, J.W. (1974). Influence of ovine prolactin on transport of f l u i d and sodium chloride by the mammalian intestine and g a l l bladder. J. Endocrinol. 6 3 , 3 1 1 - 3 1 7 . McLain, C.R. ( 1 9 6 3 ) . Amniography studies of the gastro-intestinal motility of the human fetus. Amer. J. Obstet. Gynec. 8 6 , 1 0 7 9 - 1 0 8 7 . Mellor, D. ( I 9 6 9 ) . Potential differences between mother and foetus at different gestational ages in the rat, rab-bi t , and guinea-pig. J. Physiol. 204, 3 9 5 - 4 0 5 . Mellor, D.J., and Slater, J.S. ( 1 9 7 1 ) . Daily changes in amniotic and allantoic f l u i d during the last three months of pregnancy in conscious, unstressed ewes, with catheters in their foetal f l u i d sacs. J. Physiol. 2 1 7 , 5 7 3 - 6 0 4 . Meschia, G. ( 1 9 5 5 ) . Colloidal osmotic pressures of fet a l and maternal plasmas of sheep and goats. Amer. J. Physiol. 181. 1 - 8 . 125 Miehie, E.A., and Robertson, J.G. (1971). Amniotic and urinary oestriol assays in pregnancies complicated by rhesus immunization. J. Obstet. Gynaec. Brit. Cwlth. 78, 3^-40. " Minot, C.S. (1892).Quoted by Becker, R.F., Windle, W.S., Barth, E.E., and Schulz, M.D. ( 1 9 4 0 ) . Petal swallowing, gastrointestinal activity, and defication in utero. Surg. Gynec. Obstet. £ 0 , 6 0 3 - 6 1 4 . Murphy, B.E.P., Patrick, J., and Denton, R.L. (1975). Corti-sol i n amniotic f l u i d during human gestation. J. Clin. Endocrinol. Metab. 40, 164-167. Nicholas, H.L., Schrepper, R., Spanos, W., and Hunter, C.&. ( I 9 6 3 ) . Steroid relations confirmed. On the steroid con-tent of amniotic f l u i d . J. Kansas Med. Soc. 64, 2 9 9 - 3 0 7 . Olver, R.E., Reynolds, E.O., and Strang, L.B. (1973). Foetal lung l i q u i d . In "Foetal and Neonatal Physiology 1 Proceed-ings of the Sir Joseph Barcroft Centenary Symposium" (K.S. Comline, K.W. Cross, G.S. Dawes, and P.W. Nathan-ie l s z , eds.), pp. 2 0 8 - 2 1 3 . Cambridge University Press, Cambridge. Page, K.R., Abramovich, D.R., and Smith, M.R. (1974 a). The diffusion of t r i t i a t e d water across isolated term human amnion. J. Membrane Biol. 18, 39-48. Page, K.R., Abramovich, D.R., and Smith, M.R. (1974 b). Water transport across the isolated term human amnion. J. Membrane Biol. 18, 4 9 - 6 0 . Parke, L. (1973). Detection of prolactin activity by bio-assay i n human amniotic f l u i d . J. Endocrinol. 5 8 , 137-138. Parmley, T.H., ans. Seeds, A.E. (1970). Fetal skin permea-b i l i t y to isotopic water (TH0) in early pregnancy. Amer. £• Obstet. Gynec. 108, 1 2 8 - I 3 I . Patti, A.A., and Stein, A.A. ( 1 9 6 4 ) . "Steroid Analysis by Gas Liquid Chromatography," Charles C. Thomas, Spring-f i e l d . Peterson, O.H. (1971). Initiation of salt and water trans-port in the salivary glands by acetylcholine. Phil. Trans. Roy. Soc. Lond. B. 262, 307-314. 126 Pickford, G.E., G r i f f i t h , R.W., Torretti, J., Hendlez, E., Epstein, F.H. (1970). Branchial reduction and renal stimu-lation of (Na , K7)-ATPase by prolactin i n hypophysecto-mized k i l l i f i s h i n fresh water. Nature(London) 228, 3 7 8 - 3 7 9 . Pitkin, R.M., and Reynolds, W.A. ( 1 9 7 5 ) . Creatinine exchange between mother, fetus, and amniotic f l u i d . Amer. J. Physiol. 228, 2 3 1 - 2 3 7 . Pitkin, R.M., Reynolds, W.A., and Hodari, A.A. ( 1 9 7 3 ) . Amniotic f l u i d composition following f e t a l nephrectomy or tracheal ligation. Intern. J. Gynaec. Obstet. 1 1 , 4 5 - 5 0 . ~ ~ Plentl, A.A. ( 1 9 6 1 ) . Transfer of water across the perfused umbilical cord. Proc. Soc. Exper. Biol. Med. 1 0 7 , 6 2 2 - 6 2 6 . Plentl, A.A., and Hutchinson, D.L. ( 1 9 5 3 ) * Determination of deuterium exchange rates between maternal circulation and amniotic f l u i d . Proc. Soc. Exper. Biol. Med. 82, 681-684. "~~ Pritehard, J.A. ( I 9 6 5 K Deglutition by normal and anenceph-a l i c fetuses. Obstet. Gynec. (New York) 25, 2 8 9 - 2 9 7 . Pritehard, J.A. ( 1 9 6 6 ) . Fetal swallowing and amniotic f l u i d volume. Obstet. Gynec. (New York) 2 8 , 6 0 6 - 6 I O . Ramsey, E.M., Corner, G.W., Long, W.N., and Strang, H.M. ( 1 9 5 9 ) * Studies of amniotic f l u i d and intervillous space pressures in the rhesus monkey. Amer. J. Obstet. Gynec. 2Z, 1 0 1 6 - 1 0 2 7 . Renfree, M.B., Hensleigh, H.I., and McLaren, A. ( 1 9 7 5 ) . Developmental changes i n the composition and amount of mouse fe t a l fluids. J. Embryol. Exper. Morphol. 3_3_, 435-446. Reynolds, S.R.M. ( i 9 6 0 ) . Regulation of feta l circulation1 i t s relation to f e t a l distress. Clin. Obstet. Gynec. 2* 8 3 4 - 8 5 1 . Reynolds, S.R.M., Paul, W.M., and Hugget, A. St. G. ( 1 9 5 4 ) . Physiological study of the monkey fetus in utero1 a pro-cedure for blood pressure recording, blood, sampling, and injection of the fetus under normal conditions. Bull. Johns Hopkins Hosp. 95» 2 5 6 - 2 6 5 . 127 Reynolds, W.A. (1972). Fetal sources of amniotic f l u i d i an enigma. In "Physiological Biochemistry of the Fetusi Pro-ceedings of theInternational Symposium" (A.A. Hodari and F. Mariona, eds.), pp. 3 - 3 L Charles C. Thomas, Spring-f i e l d . Reynolds, W.A., Pitkin, R.M., and Hodari, A.A. (1971). Transfer of radioactive iodopyracet from fetus into amniotic f l u i d after tracheal ligation. Amer. J. Obstet. Gynec. 110. 566-571. Rocha, A.S., and Kokko, J.P. (1974). Permeability of medul-lary nephron segments to urea and wateri effect of vaso-pressin. Kidney International 6, 379-3 8 7. Rosenbloom, A.A., Sack, J., and Fisher, D.A. (1975). The circulating vasopressinase of pregnancyi species comparison with radioimmunoassay. Amer. J. Obstet. Gynec. 121, 316-320. Saunders, P., and Rhodes, P. (1973). The origin and circu-lation of the amniotic f l u i d . In "Amniotic Fluidi Research and C l i n i c a l Application" (D.V.I. Fairweather and T.K.A.B. Eskes, eds.), pp. 1-18. Excerpta Medica, Amsterdam. Schruefer, J.J.P., Seeds, A.E., Battaglia, F.C., Hellegers, A.E., Behrman,R.E., and Bruns, P.D. (1972). Changes i n amniotic f l u i d volume and total solute concentration in the rhesus monkey following replacement with d i s t i l l e d water. Measurement of amniotic f l u i d volume by I 1- 3 a l -bumin dilution. Amer. J. Obstet. Gynec. 112, 807-815. Schultz, F.M., MacDonald, P.C., and Johnson, J.M. (1975 a). Initiation of human partuitioni demonstration of a unique phospholipase A2 in chorioamnion. Sci. Abst. 22nd. Ann. Meeting Soc. Gynec. Invest. 6, 11. Schultz, F.M., MacDonald, P.C., and Johnson, J.M. (1975 *>). Arachidonic acid in human amniotic f l u i d and i t s relation-ship to labor. Sci. Abst. 21st. Ann. Meeting Soc. Gynec. Invest. J,, 36. Schwarz, B.E., Schultz, F.M., MacDonald, P.C., and Johnson, J.M. (1975 a). Subcellular localization of phospholipase A2 i n human f e t a l membranes. Sci. Abst. 22nd. Ann. Meeting  Soc. Gynec. Invest. 6, 45. 128 Sehwarz, B.E., Vanatta, P., S i i t e r i , P.K., and MacDonald, P. C ( 1 9 7 5 b)» Steroid receptors i n human f e t a l membranes. Sci. Abst. 21st. Ann. Meeting Soc. Gynec. Invest. 21. Scoggin, W.A., Harbert, G.M., Anslow, W.P., Van't Riet, B., and McGaughey, H.S. (1964). Fetomaternal exchange of water at term. Amer. J. Obstet. Gynec. 90, 7-16. Seeds, A.E. (1965)* Water metabolism of the fetus. Amer. J. Obstet. Gynec. ^2, 727-745. Seeds, A.E. (1972 a). Amniotic f l u i d . In "The Water Metabo-lism of the Fetus" (A.C. Barnes and A.E. Seeds, eds.), pp. 48 - 7 3 . Charles C. Thomas, Springfield. Seeds, A.E. (1972 b). Mechanisms of intrauterine water transfer i n pregnancy. In "The Water Metabolism of the Fetus" (A.C. Barnes and A.E. Seeds, eds.), pp. 32-47. Charles C. Thomas, Springfield. Seeds, A.E. (1973). Water dynamics in amniotic f l u i d . In "Amniotic Fluidi Research and C l i n i c a l Application" (D.V.I. Fairweather and T.K.A.B. Eskes, eds.), pp. 52-59. Excerpta Medica, Amsterdam. Seeds, A.E., Behrman, R.E., Battaglia, F.C., Hellegers, A.E., and Bruns, P.D. (1964). Changes i n amniotic f l u i d total solute concentrations i n the rhesus monkey. Amer. J. Obstet. Gynec. 8£, 476-480. Seeds, A.E., and Hellegers,A.E. ( 1 9 6 8 ) . Acid-base determin-ation i n human amniotic f l u i d throughout pregnancy. Amer. J. Obstet. Gynec. 1 0 1 , 2 5 7 - 2 6 0 . Skowsky, W.R., Bashore, R.A., Smith, F.G., and Fisher, D.A. (1973). Vasopressin metabolism i n the fetus and newborn. In "Foetal and Neonatal Physiology1 Proceedings of the Sir Joseph Barcroft Centary Symposium" (K.S. Comline, K.'W. Cross, G.S. Dawes, and P.W., Nathanielsz, eds.), pp. 439-447. Cambridge University Press, Cambridge. Snart, R.S., and Dalton, T. ( 1 9 7 3 ) . Response of toad bladder to prolactin. Comp. Biochem. Physiol. 45, 307-3*1• Snell, F.M., Shulman, S., Spencer, R.P., and Moos, C. ( 1 9 6 5 ) . "Biophysical Principles of Structure and Function," pp. 300-314. State University of New York, Buffalo. 129 Starrier, M.W. (1971). Osmolarity of urine from renal pelvis and bladder of foetal and post-natal pigs. J. Physiol. 218, 30 P. Stanley, J.E., and Fleming, W.R. (1967). Effects of prolactin and ACTH on the serum and urine sodium levels of Fundulus kansae. Comp. Biochem. Physiol. 20, 199-208. Strauss, J.F., Sokoloski, J., Caploe* P.» Duffy, P., Mintz, G., and Strambaugh, R.L. (1975). On the role of prosta-glandins i n partuition in the rat. Endocrinology 96, 1040-1043. Turner, CD., and Bagnara, J.T. (1971). "General Endocrin-ology," pp. 172-178. W.B. Saunders Company, Philadelphia. Utida, S., Hirano, T., Oide,'H., Ando, M., Johnson, D.W., and Bern, H.A. (1972). Hormonal control of the intestine and urinary bladder in teleost osmoregulation. Gen. Comp. Endocrinol. Suppl. 2 , 317-327. Vizsolyi, E., and Perks, A.M. (1969). New neurohypophyseal principle i n foetal mammals. Nature (London) 223, II69-1171. Vizsolyi, E., and Perks, A.M. (1974). The effect of arginine vasotocin on the isolated amniotic membrane of the guinea-pig. Can. J. Zool. j>2, 371-387. Vosburgh, G.L., Flexner, L.B., Cowie, D.B., Hellman, L.M., Proctor, N.K., and Wilde, W.S. (1948). Rate of renewal i n women of water and sodium of the amniotic f l u i d as deter-mined by tracer techniques. Amer J. Obstet. Gynec. 56, 1156-1159. Wang, C.H., and W i l l i s , D.L. (1965). "Radiotracer Method-ology in Biological Science," pp. 197-199* Prentice-Hall, Inc. Englewood C l i f f s , New Jersey. Westin, B. (1959). Some constituents of umbilical venous blood of previable fetuses. Acta. Paediat. Scand. 48, 609-6I3. Winters, A.J., Porter, J.C, Colston, C , and MacDonald, P.C (1975K Hormones influencing the fe t a l adrenal cortex. Sci. Abst. 22nd. Ann. Meeting Soc. Gynec. Invest. 6, 7. 130 Wislocki, G.B. (1920). Experimental studies i n fetal absorp-tion, Carnegie Inst. Wash. Contrib. Embryol. 11, 46-61. Wright, G.H., and Nixon, D.A. (1961). Absorption of amniotic f l u i d i n the gut of foetal sheep. Nature (London) 190, 816. 

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0093429/manifest

Comment

Related Items