"Medicine, Faculty of"@en . "Cellular and Physiological Sciences, Department of"@en . "DSpace"@en . "UBCV"@en . "Shami, Yehezkel"@en . "2010-02-01T23:16:03Z"@en . "1974"@en . "Doctor of Philosophy - PhD"@en . "University of British Columbia"@en . "Calcium transport across the placenta is asymmetrical and is believed to be an active transport. An essential step in such a transport\r\nis translocation of the ion across a single plasma membrane. The objective of this thesis was to study the Ca2+ -related properties of the placental plasma membranes and to gain some knowledge of their role in Ca2+ -transport. Three Ca2+ -related properties were studied: 1. Ca2+ -binding to the placental plasma membranes; 2. The membrane bound enzyme Ca -ATPase; and 3. Ca2+ -uptake by the placental plasma membrane vesicles. Ca2+ -binding properties of the membrane preparation were studied by the use of a new method, the flow dialysis system. Two types of sites for Ca were found: 1) high affinity, low capacity sites, and 2) low affinity, high capacity sites. The high affinity sites had 10-fold higher affinity for Ca2+ than for Mg2+ . A calcium-stimulated, membrane-bound enzyme, namely Ca2+ -ATPase, was located in the placental plasma membranes. This enzyme is distinct from the Na+, K+-ATPase and alkaline phosphatase. The enzyme can be activated by Mg2+ but with lower efficiency. Both Ca2+ and Mg2+ activate the enzyme at the same site. A formula was derived, enabling one to predict very precisely the velocity of the enzyme incubated under any combination of Ca2+ and Mg2+ ; this relationship is presented in a three dimensional model. The formula can be used for other enzymes or other substrates, as was demonstrated with ATP and ADP.\r\n\r\n\r\nThe placental plasma membrane vesicles are capable of accumulating Ca2+ . Ca2+ -uptake was defined as the amount of Ca2+ which is not available for rapid exchange and cannot be displaced by a high concentration\r\nof competitor in the presence of ATP. This definition is different from and more accurate than the one which is widely used and cited in the literature. An intravesicular Ca2+ concentration of 190 mM was recorded, which was 24-fold higher than the external Ca2+ concentration (8 mM). Ca2+ -uptake was dependent on ATP hydrolysis by the placental Ca2+ -ATPase. This process was independent of Mg2+ . It is suggested that while the substrate for Ca2+ -ATPase is Ca-ATP, the substrate for Ca2+ -uptake is Ca2+. The overall Ca2+ -related properties of the placental plasma membranes are independent of Mg and the entire process from binding to membrane through activation of the enzyme and finally Ca2+ -uptake is dependent on Ca2+ alone. This situation is unique to the placental plasma membranes. It is tempting to speculate that the link between the maternal and the fetal circulation is achieved by forming vesicles loaded with Ca2+ on the maternal side and unloading them through fusion with the basal plasma membrane on the fetal side. The Ca2+ -related properties of placental plasma membranes described\r\nin this thesis, provide many answers regarding the first step in the asymmetrical transplacental Ca2+ -transport. Further investigation is required before a full understanding of the entire process is achieved."@en . "https://circle.library.ubc.ca/rest/handle/2429/19542?expand=metadata"@en . "CALCIUM RELATED PROPERTIES OF PLASMA MEMBRANES FROM GUINEA PIG PLACENTA by YEHEZKEL SHAMI B.Sc, The Hebrew University of Jerusalem, 1969 M.Sc, The Hebrew University of Jerusalem, 1970 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of PHYSIOLOGY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September, 1974 In presenting th is thesis in pa r t i a l fu l f i lment of 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 shal l make it f ree ly ava i l ab le for reference and study. I fur ther agree that permission for extensive copying of th is thes is for scho la r ly purposes may be granted by the Head of my Department or by his representat ives . It is understood that copying or pub l ica t ion of th is thes is for f inanc ia l gain sha l l not be allowed without my wri t ten permission. Department The Univers i ty of B r i t i s h Columbia Vancouver 8, Canada ABSTRACT Calcium transport across the placenta is asymmetrical and is believed to be an active transport. An essential step in such a trans-port is translocation of the ion across a single plasma membrane. The 2+ objective of this thesis was to study the Ca -related properties of the placental plasma membranes and to gain some knowledge of their role 2+ in Ca -transport. 2+ Three Ca -related properties were studied: 2+ 1. Ca -binding to the placental plasma membranes; 2+ 2. The membrane bound enzyme Ca -ATPase; and 2+ 3. Ca -uptake by the placental plasma membrane vesicles. 2+ Ca -binding properties of the membrane preparation were studied by the use of a new method, the flow dialysis system. Two types of sites 2+ for Ca were found: 1) high a f f i n i t y , low capacity sites, and 2) low a f f i n i t y , high capacity sites. The high a f f i n i t y sites had 10-fold 2+ 2+ higher a f f i n i t y for Ca than for Mg . 2+ A calcium-stimulated, membrane-bound enzyme, namely Ca -ATPase, was located in the placental plasma membranes. This enzyme is distinct from the Na+, K+-ATPase and alkaline phosphatase. The enzyme can be activated by Mg but with lower efficiency. Both Ca and Mg activate the enzyme at the same site. A formula was derived, enabling one to predict very precisely the velocity of the enzyme incubated under any 2+ 2+ combination of Ca and Mg ; this relationship is presented in a three dimensional model. The formula can be used for other enzymes or other substrates, as was demonstrated with ATP and ADP. i i I l l The placental plasma membrane vesicles are capable of accumu-2+ 2+ 2+ lating Ca . Ca -uptake was defined as the amount of Ca which is not available for rapid exchange and cannot be displaced by a high con-centration of competitor in the presence of ATP. This definition is different from and more accurate than the one which is widely used and 2+ cited in the literature. An intravesicular Ca concentration of 190 mM 2+ was recorded, which was 24-fold higher than the external Ca concen-2+ tration (8 mM). Ca -uptake was dependent on ATP hydrolysis by the 2+ 2+ placental Ca -ATPase. This process was independent of Mg . 2+ It is suggested that while the substrate for Ca -ATPase is Ca-2+ 2+ ATP, the substrate for Ca -uptake is Ca . 2+ The overall Ca -related properties of the placental plasma 2+ membranes are independent of Mg and the entire process from binding 2+ to membrane through activation of the enzyme and f i n a l l y Ca -uptake 2+ is dependent on Ca alone. This situation is unique to the placental plasma membranes. It is tempting to speculate that the link between the maternal and the fetal circulation is achieved by forming vesicles loaded with 2+ Ca on the maternal side and unloading them through fusion with the basal plasma membrane on the fetal side. 2+ The Ca -related properties of placental plasma membranes des-cribed in this thesis, provide many answers regarding the f i r s t step in 2+ the asymmetrical transplacental Ca -transport. Further investigation is required before a f u l l understanding of the entire process is achieved. TABLE OF CONTENTS Page ABSTRACT i i LIST OF TABLES vi LIST OF FIGURES v i i LIST OF PLATES x ACKNOWLEDGMENT xi GENERAL INTRODUCTION . 1 DIVISION I CHAPTER I - CALCIUM BINDING TO THE PLACENTAL PLASMA MEMBRANES . . . 10 INTRODUCTION * 10 MATERIALS AND METHODS 13 RESULTS 25 DISCUSSION 36 SUMMARY 45 DIVISION II CHAPTER II - CHARACTERIZATION OF CALCIUM-STIMULATED ATPase 47 INTRODUCTION 47 MATERIALS AND METHODS 48 RESULTS 50 DISCUSSION 69 SUMMARY 76 iv V Page CHAPTER III - THE EFFECT OF Ca 2 +/Mg 2 + CONCENTRATION RATIO ON PLACENTAL (Ca2+-Mg2+)-ATPase ACTIVITY 78 INTRODUCTION 78 MATERIALS AND METHODS 83 RESULTS 84 DISCUSSION 89 SUMMARY 91 CHAPTER IV - COMPARISON BETWEEN ALKALINE PHOSPHATASE AND Ca 2 +-ATPase 92 INTRODUCTION 92 MATERIALS AND METHODS . . 93 RESULTS . . 95 DISCUSSION 99 SUMMARY 104 DIVISION III CHAPTER V - Ca2+-UPTAKE BY PLACENTAL PLASMA MEMBRANE VESICLES . . . 107 INTRODUCTION 107 MATERIALS AND METHODS 109 RESULTS 110 DISCUSSION 119 SUMMARY 124 GENERAL CONCLUSIONS 125 REFERENCES 130 LIST OF TABLES Table Page I Specific activities of marker enzymes in the different fractions 18 II Affinity and capacity of placental plasma membranes and Bovine Albumin 32 45 2+ 40 2+ III Apparent Km's for Ca displacement by Ca , Mg2+, S r 2 + 39 IV Comparison between alkaline phosphatase activity and ATPase activity in the membrane preparation and the purified alkaline phosphatase fraction 101 V The effect of phosphate on Ca -uptake H 7 2+ 2+ VI The effect of 5.4 mM Mg on Ca displacement (diffusion rate) by 3 x 10_1M 4 0 C a 2 + and Ca 2 +-uptake 118 vi LIST OF FIGURES Figure Page 1 Diagram of the apparatus used for measuring calcium binding by rate of dialysis 22 45 2+ 2 Measurement of Ca diffusion rate at various 45 2+ Ca concentrations 26 \u00E2\u0080\u00A245 2+ 3 Measurement of Ca diffusion rate as a function of cpm concentration in the upper chamber 27 4 Flow dialysis profiles of calcium binding at various calcium concentrations . . . \u00E2\u0080\u00A2 29 2+ 2+ 5 Ca concentration effect on Ca -binding levels by placental plasma membranes and Bovine Albumin 30 2+ 6 Scatchard plot of Ca -binding by placental plasma membranes and Bovine Albumin 31 2+ 7 Ca -binding as a function of placental plasma membrane protein concentration , 34 2+ 8 pH effect on Ca -binding 35 9A 4^Ca 2 + displacement by divalent cations . . . 37 45 2+ 9B,C,D Double reciprocal plots for Ca displacement by 4 0 C a 2 + , Mg2+ and S r 2 + 38 10 Stimulation of ATP hydrolysis by divalent cations . . . . 52 11 Variations in Lineweaver-Burk plots of ATPase 2+ activity at various concentrations of ATP, Ca and Mg2+ 53 2+ 2+ 12 Enzyme activation by Ca and Mg 55 vi i v i i i Fi gure Page 13 The effect of incubation time on Pi release and pH of the incubation medium 56 14 Protein concentration effect on Pi release 58 + 2+ 15 Effect of Na on enzyme activation by Ca 59 16 Effect of pH on enzyme activity 60 17 Effect of inhibitors 62 2+ 18 Effect of mersalyl on Ca activation of the enzyme . . . 63 2+ 19 Effect of EDTA on enzyme activation by Ca 64 2+ 2+ 20 EGTA effect on Ca and Mg activation of the enzyme. . . 65 21 The effect of blocking NH3+ groups with maleic anhydride . 67 22 Substrate specificity: Hydrolysis of ATP compared to that of other high energy t r i - and diphosphate nucleotides 68 23 Temperature effect on enzyme activity 70 24 Postulated scheme for ATP hydrolysis in the presence of Ca 2 + and Mg 2 + 80 25 The effects of Ca 2 + and Mg 2 + on ATP hydrolysis 85 26A The effect of changing the concentration ratio of Mg 2 + to Ca 2 + on ATP hydrolysis 86 26B The change in -K^ as a function of [Mg 2 +]/[Ca 2 +] 86 27 A three dimensional model describing the relationship 2+ 2+ between [Mg ]/[Ca ] and the activity of the placental Ca2+-ATPase 88 28 The elution profile of alkaline phosphatase and protein on sephadex G-200 gel f i l t r a t i o n 96 29 The effect of pH on ATP hydrolysis by the purified alkaline phosphatase . 97 ix Figure Page 30 The effect of pH on p-nitrophenyl phosphate hydrolysis by the purified alkaline phosphatase 98 31 The effect of storage at 4\u00C2\u00B0C on enzyme activity 100 2+ 32 Ca -ATPase specific activity as function of embryo age. . 102 33 The effect of the major constituents of the incubation 45 2+ -1 40 2+ medium on Ca displacement by 3 x 10 M Ca . . . . I l l 2+ 34A The effect of incubation time on Ca -uptake and ATP hydrolysis 112 2+ 34B The relationship between Pi release and Ca -uptake by the placental plasma membrane vesicles 114 2+ 2+ 35A The effect of Ca concentration on Ca -uptake 115 2+ 35B The effect of Ca concentration on Pi/Ca ratio 115 2+ 2+ 36 The effect of Ca concentration on Ca -binding and 2+ on the velocities of ATP hydrolysis and Ca uptake . . . 126 LIST OF PLATES Plate Page 1 Electron microscopic picture of the trophoblastic layer of guinea pig placenta ^ 2 Detail of placental membrane between the maternal blood space and the fetal capillary 7 3 Electron microscopic picture of the final membrane preparation (x 24,282) 20 4 Electron microscopic picture of the final membrane preparation (x 61,560) 21 5 Histochemical localization of ATPase activity in the guinea pig placenta 51 x ACKNOWLEDGMENT The assistance of a number of people contributed to the accomp-lishment of this thesis. F i r s t , I would l i k e to thank my research super-visor, Dr. D. H. Copp, for the interest he showed in the subject and for the continuous encouragement which made the conclusion of this study possible. The help of Dr. Harold Messer in editing this thesis, and the helpful advice he provided me throughout the entire period, is greatly appreciated. I thank Dr. C. F. Cramer for his responsiveness while acting as the chairman of my research committee. The just criticism of Dr. V. Palaty, which was done in the tradi-tional manner of a good teacher, contributed to reaching a higher scien-t i f i c 1evel. I am particularly indebted to Dr. I. C. Radde for introducing me to the subject and guiding me in the early stages. The good and encouraging atmosphere which was provided by a l l the people in the laboratory was greatly appreciated. Special thanks are due to Kurt Henze and Ralph Assina for the preparation of the thesis il l u s t r a t i o n s . Finally, I wish to thank my wife, Chava, for her help, under-standing and love, for without these I could not have made i t . xi GENERAL INTRODUCTION Calcium ion has important roles associated with cellular func-tion. These include: 1. Coupling between excitation - secretion at nerve endings (Eccles, 1964). 2. Coupling in excitation - contraction in muscle (Ebashi and Endo, 1968; Bianchi, 1969). 3. Maintenance of membrane integrity (Poste and Allison, 1973). 4. Membrane fusion. Thus a l l cellular events which include a stage of fusion of membrane are calcium dependent (Poste and Allison, 1973). These include cell fusion, endocytosis, exocrine (Hagen, 1959) and endocrine secretion (Hales and Milner, 1968; Curry et al_., 1968) (exocytosis). 2+ 5. Regulation of enzyme activity. Ca acts as an inhibitor of a large number of intracellular enzymes while i t serves as an activator for only a few of them (Bianchi, 1968). 6. Control of hormone secretion (Copp, 1970). Calcium ion concentration in the cytosol is estimated to be in the range of 10~5M-10\"8M (Hodgkin and Keynes, 1957; Nanninga, 1961), _3 in contrast to 10 M in extracellular f l u i d . The necessity for low 2+ Ca concentration in the cytosol was explained some years ago as follows: \u00E2\u0080\u00A2 high Ca in the cytosol may react with the high (as then believed) 1 2 inorganic phosphate in the cytosol to form a calcium phosphate precipi-tate, which w i l l impair cell function (Manery, 1969). This argument lost much of it's basis with the finding that inorganic phosphate concen-tration in the cytosol is very low (Seraydarian et al_., 1961). Some calcium phosphate precipitates can be found in cell organelles which 2+ are involved in maintaining low Ca concentration in the cytosol (Martin 2+ and Matthews, 1970; Borle, 1973). By examining the roles for Ca in cellular function, one can jus t i f i a b l y conclude that calcium ion is a powerful regulator of cell function, and for this reason i t s intracellular concentration is controlled very closely at a low level. The extracellu-lar f l u i d is the most obvious \"sink\" for disposal of the excess intra-cellular calcium. The best available description of such a regulatory mechanism is that of the red blood cell (Vincenzi, 1971). According to 2+ 2+ this model, a membrane bound enzyme, namely a Ca -Mg -ATPase, is respon-sible for the active extrusion of calcium from the red blood c e l l . Since red blood cells in circulation contain no organelles, the plasma membrane \"calcium pumps\" is the only effective mechanism regulating intracellular 2+ 2+ Ca concentration. Though Ca -ATPase is found in plasma membranes of several other tissues (Martin et al_., 1969; Parkinson and Radde, 1971; 2+ Ma e_t al_., 1974), its involvement in Ca extrusion in those tissues remains speculative at this stage. Unlike the red blood c e l l , a l l other cells are equipped with 2+ organelles which can respond effectively to changes in cytosol Ca 2+ concentration, accumulating or releasing Ca as needed. The most e f f i -2+ cient cell organelle in controlling cytosol Ca is the sarcoplasmic 2+ reticulum (Martonosi and Feretos, 1964). Once again Ca -ATPase is implicated. However, sarcoplasmic reticulum is unique to muscle tissue. The less specialized endoplasmic reticulum (microsomal fraction) is 2+ 2+ also capable of accumulating Ca (Alonso and Walser, 1968). This Ca accumulation involves ATP hydrolysis. The mitochondrion is another 2+ cellular organelle which accumulates Ca (Reynafarje and Lehninger, 1969) 2+ The fate of the accumulated Ca within the cellular organelles and the way i t is f i n a l l y extruded from the cel l is not clear. One can speculate that at least the microsomes can fuse with the plasma membrane and secrete their contents into the extracellular fluids. Alternatively 2+ Ca uptake by the cellular organelles may become important only when the plasma membrane calcium pump is unable temporarily to handle a large 2+ influx of Ca . Thus the cellular organelles act as a \"buffer\" mechanism. Borle (1973) proposed a model of cellular calcium regulation in which 2+ Ca influx is passive and the mitochondria act as the main regulator of cytoplasmic calcium activity and of calcium transport. He estimated that the efficiency of the plasma membrane calcium pump is about the same of that of the mitochondria. However, since the surface area of the mitochondria is so much larger than that of the plasma membrane, 2+ the contribution of the mitochondria to cellular Ca regulation is 97% where the plasma membrane contributes only 3%. Thus in this model there 2+ are three major parameters: passive Ca influx, active uptake by the 2+ mitochondria which acts as an ion buffer, and active extrusion of Ca by the plasma membrane calcium pump. Hormones affecting calcium metabolism 4 (PTH, Calcitonin and Vitamin D Metabolite 1-25 DHCC) can act by modi-fying the three parameters that regulate the exchange between the cyto-sol, the mitochondria and the extracellular f l u i d . 2+ The importance of maintaining intracellular Ca at a low level 2+ creates special problems in transport of Ca from one body compartment to another across a cellular barrier, since this must occur without 2+ increasing significantly the Ca concentration in the cytosol of the 2+ cells involved. Such transport, which involves a net transfer of Ca in one direction (asymmetric), occurs in the gut, bone, kidney, and placenta. The placenta provides a suitable system for studying calcium transport from one body compartment to another. It was chosen as the subject of this study for the following reasons: 1) By the 60th day of gestation the placental barrier consists of fetal endothelial and trophoblastic layers. It may be assumed that the latter is involved in active transport of calcium. 2) It is easy to isolate significant quantities of plasma membranes primarily of the trophoblast. 2+ 3) In many mammalian species transfer of Ca across the placenta is asymmetrical (net transfer from dam to fetus) and occurs against a concentration difference (Papadopoulos e_t al_., 1967; Macdonald et a l . , 1965; Twardock and Austin, 1970). 4) This concentration difference can be easily measured. 2+ 5) Because of the large amounts of Ca required by the fetus (Comar, 1956), i t is possible that the transport system is operating at high velocity. 5 6) In vivo studies (Papadopoulos et al_., 1967; Macdonald et a l . , 1965; Twardock and Austin, 1970) have suggested the presence of an active 2+ Ca transport across the placenta. Comar (1956) estimated the net 2+ hourly transfer of Ca from dam to fetus to be 7% of the total maternal plasma calcium in the human, 50% in the cow, 100% in the rabbit and 400% in the guinea pig. In these four species the stress placed on calcium homeostatic mechanisms by pregnancy is far the greatest in guinea pig and the least in man. This fact and the similarity in the structure of the guinea pig placenta to that of man (both are classified as haemochorial), made the guinea pig placenta very attractive for use as a model. The structure of the guinea pig placenta as shown by electron microscopy (Bjorkman, 1970) is presented in Plates 1 and 2. They show that the maternal and the fetal circulation are separated by a continuous layer of syncytiotrophoblast. This would appear to be the logical site 2+ for active Ca transport across the placenta. However, the obvious limitations of in vivo studies preclude answers concerning the transport 2+ of Ca at the cellular level. 2+ Since Ca transport occurs across the trophoblastic layer, i t is assumed that the plasma membranes should demonstrate some calcium-2+ related properties which could assist in Ca transport. 2+ The subject of this thesis is to characterize the Ca -related properties of the placental plasma membranes and to try to u t i l i z e them 2+ 2+ in understanding Ca transport, in the same way that Ca -related pro-perties of the sarcoplasmic reticulum (Martonosi and Feretos, 1964), Plate 1. Survey picture showing the great variation in thickness of the trophoblastic layer (Tr) and i t s relation to the foetal capillaries (FCp). Note the microvilli (Mv) projecting into the maternal blood spaces (MS). Em. x 3600. 7 Plate 2. Detail of placental membrane between a maternal blood space ( M S ) and a foetal capillary (FCp). The intervening layers comprise a simple layer of trophoblast (Tr), a basal lamina (BL), and foetal endothelium (FE). The trophoblast appears to be highly differentiated with a well developed rough surfaced endoplasmic reticulum and infoldings of the basal plasma membrane (BI). Em. x 23,300. 8 mitochondria (Reynafarje and Lehninger, 1969), and red blood cell (Schatz-mann and Rossi, 1971) membranes contributed to understanding of intra-2+ cellular Ca regulation. Since the sarcoplasmic reticulum has been 2+ most intensively studied with respect to i t s Ca -related properties, i t often serves as a reference in this thesis. However, one should keep 2+ 2+ in mind that unlike the placental Ca transport, the sarcoplasmic Ca transport is symmetric and is exposed to large changes in velocity in a very short time as part of i t s role in contraction-relaxation of muscle. 2+ Three Ca -related properties of the placental plasma membranes 2+ were chosen for study: Ca binding to the membranes, membrane bound 2+ 2+ Ca activated ATPase, and Ca uptake by the placental plasma membrane vesicles. This thesis is divided into three divisions, and the studies 2+ on each of the three Ca -related properties are described and discussed in separate divisions. The reasons for choosing these properties are given in the introduction to each division. 2+ To date there has been no f u l l description of these three Ca -related properties of any plasma membranes involved in asymmetrical 2+ Ca transport. Therefore the experiments were designed to f i l l this 2+ gap. An attempt to relate these properties to Ca -transport is made in the general conclusion of the thesis. DIVISION I CALCIUM BINDING TO THE PLACENTAL PLASMA MEMBRANES 9 CHAPTER I CALCIUM BINDING TO THE PLACENTAL PLASMA MEMBRANES INTRODUCTION The present concepts of the mechanism of active transport require that a preliminary and essential step for calcium transport is the binding of the ion to the membrane involved. This binding is assumed to be passive. A knowledge of the number of calcium-binding site s , their specificity, relative a f f i n i t i e s , and capacity is essential to an under-standing of this process (binding to the membrane) and i t s relationship to active transport. Perhaps the most important property is the specificity of the site for the ion to be transported. The a b i l i t y of the membrane to transport ions selectively is dependent on the specificity of the si t e . The best described selectivity in biological membranes, is that between Na+ and K+. The \"Sodium Pump\" system has specific sites for Na+ on the inner surface of the cell membrane and specific sites for K+ on the outer surface of the cell membrane. The apparent a f f i n i t i e s of each site are at least 50 times higher for the one ion than for the other (Garay and Garrahan, 1973). If we exclude the monovalent cations as 2+ competitors for the Ca s i t e , the only divalent cation whose pnysiological 2+ 2+ concentration enables i t to act as a competitor for Ca is Mg . 10 11 The driving force underlying spe c i f i c i t y , is the difference between the free energy of cation-site electrostatic interaction and 2+ 2+ the free energy of hydration of the cation. When Ca and Mg are present in the medium in equimolar concentration, the distribution of 2+ 2+ the occupancy of the sites by Ca and Mg is dependent on the ratio between the -AG (free energy) for the corresponding reactions. The free energies of the reactions can be estimated from the following equa-2+ tions. For Ca : A GCa = A GICa \" AGHCa where A ^jca = t ' i e c n a n 9 e i n ^ r e e e n e r9Y of the system after the 2+ interaction of Ca with the site (x) Ca 2 + + X ' Ca-X 2+ and AGHCa = t h e c n a n 9 e 1\"n t n e f r e e e n e r 9 y \u00C2\u00B0f the system when Ca is hydrated Ca 2 + + n H20 \" Ca 2 + (HgO) n 2+ Since Ca in aqueous solution is mainly hydrated, i t must f i r s t be dehydrated (+AG^) before i t can interact with the si t e . If the AG of the overall reaction is negative, i t w i l l proceed spontaneously. (AG 2+ 2+ Mg can be calculated in the same way as for Ca .) In the presence 2+ 2+ of both Mg and Ca in equimolar concentrations, the final ratio of calcium-occupied sites (Ca - X) to magnesium-occupied sites (Mg - X) wi l l be as follows: Ca - X - A G C a 2 + Mg - X -AGMg2+ 12 Since the diameter of Ca 2 + (0.99 A\u00C2\u00B0) is larger than that of Mg 2 + (0.65 A 0) its dehydration energy wi11 be smaller. Since the dehydration energies are constant under physiological conditions, the only way for the cel l to increase i t s selectivity is by construction of a specific s i t e . Since 2+ 2+ the charge density of Ca is smaller than that of the Mg , i t w i l l react selectively with sites with low f i e l d . Strong f i e l d site (weak 2+ acid) w i l l react favorably with Mg (Diamond and Wright, 1969). 2+ In this chapter, the Ca -binding properties of the placental 2+ plasma membrane are described. Their possible relationship to Ca transport is dealt with in the general conclusion. 2+ The most popular technique for measurement of Ca -binding is ul t r a f i l t r a t i o n . However, in the present study some uncontrolled non-specific binding to the f i l t e r was found. For example, i t was observed 2+ that the binding of Ca to the f i l t e r was time dependent. Thus with increasing membrane concentration of the fil t e r e d sample, the f i l t r a t i o n 2+ time increased as did the uncontrolled non-specific Ca -binding to the f i l t e r . Because of these d i f f i c u l t i e s , a method based on the measurement of the rate of diffusion in a flow dialysis system was used (Colowick and Womack, 1969). This method has not previously been used for measuring 2+ Ca -binding, so i t had to be modified as described in the methods section of this chapter. The advantages of the flow dialysis method are given in the discussion. MATERIALS AND METHODS 13 Isolation of Placental Plasma Membranes The method of Post and Sen (1967) for the isolation of renal plasma membranes was adapted for this study of placental plasma membranes. The purity of the final preparation was checked by electron microscopy and by enzyme markers to ensure that the method was suitable for iso-lation of the placental plasma membranes. Pregnant guinea pigs around the 60th day of gestation were anaes-thetized with Na pentobarbital (100 mg/kg). The abdomen was opened and the amniotic sacs with the fetuses were removed. Each placenta was carefully freed from membranes, gross blood vessels and uterine tissue and placed in ice-cold 0.9% NaCl solution. Using a Thomas tissue grinder, size C with Teflon pestle, for 10 strokes at 1500 rpm,the placentas were homogenized in 25 ml of a solution containing 87 g sucrose, 1.169 g NaCl, 1.860 g Na^EDTA, 0.2 g MgCl26 H 20 s and 0.68 g imidazole per l i t r e . The tissues were then processed in a manner similar to that described by Post and Sen (1967) for the isolation of renal plasma mem-branes, omitting the urea stage. The procedure consists of a series of centrifugations at 35,000 xg for 30 min in different solutions, and is given in detail in the accompanying flow chart. The isolated membranes were f i n a l l y suspended in 5 x 10~4M imidazole-histidine, 5 mM Tris-HCl buffer (pH 7.6) and stored at +4\u00C2\u00B0C until assay. This procedure for isolating the placental plasma membranes was used throughout the study except in certain cases, when i t was modified to serve different pur-poses. 14 Isolation of Placental Plasma Membranes HOMOGENATE cellular and nuclear pellet (discarded) -centrifuged at 300 x g for 10 min supernatant centrifuged at \"35,000 x g for 30 min I Mitochondrial pellet (discarded) Particle free supernatant (discarded) \u00E2\u0080\u00A2Microsomal fraction. The upper layer was scraped free and transferred to 25 ml solution (1), homogenized and centrifuged at 35,000 x g for 30 min. This procedure was repeated 3 times, in solutions (2), (3), and (4), to reduce progressively con-tamination by mitochondria. The final microsomal fraction was suspended in 5 x 10\" M imidazole-histidine, 5 mM Tris-HCL pH 8.0 and stored at +4\u00C2\u00B0C. Composition of Solutions Used in Isolation of the Plasma Membranes: (1) (2) (3) (4) 0.25 M Sucrose 0.25 M Sucrose 15 mM NaCl 10 mM Imidazole 0.02 M NaCl 2 mM Na?EDTA 1 mM Na2EDTA 0.1 mM Na2EDTA 5 mM Na2EDTA 0.1 mM MgCl2 3 mM Imidazole 1 mM MgCl2 4 mM Imidazole 10 mM Imidazole 0.02% (W/V) Na-Heparin 15 Determination of Membrane Concentration The protein concentration of the final preparation was taken as an indicator of membrane content in the medium. The total protein was measured by the procedure of Lowry ejt al_. (1951) using bovine albumin standards. All the results are expressed per mg of membrane protein. Enzyme Marker Assays The supernatants from the 35,000 x g centrifugations were pooled, as were the pellets. These fractions and the final preparations were assayed for alkaline phosphatase, glucose-6-phosphatase and succinate dehydrogenase. Alkaline Phosphatase (EC 3.1.3.1) Alkaline phosphatase as a marker for plasma membranes (Dixon and Webb, 1964a) was assayed by the method of Bessey et al_. (1946) as modi-fied by Kelly and Hamilton (1970). Aliquots (50 yl) from the different fractions were incubated for 30 min at 30\u00C2\u00B0C in 2-amino-2-methyl-l-propanol buffer (Eastman Organic Chemical) 0.75 M pH 10.0 using 5 mM disodium p-nitrophenyl phosphate (Sigma) as substrate. The final volume was 0.55 ml. The reaction was terminated by adding 5 ml of 0.05 N NaOH, and p-nitrophenol release was determined spectrophotometrically by reading absorbence at 410 nm, compared to a standard p-nitrophenol curve. Blank samples (in the absence of enzyme) were incubated under the same con-ditions. 16 Glucose-6-Phosphatase (EC 3.1.3.9) Glucose-6-phosphatase is a predominantly microsomal enzyme (Gins-burg and Hers, 1970) and was used in this study as a marker for the assessment of the purity of the subcellular fractions. The enzyme was tested as described by Hiibscher and West (1965). The incubation mix-ture contained: 43 mM maleate buffer (pH 6.0), 2 mM KF, 4 mM EDTA, 28.5 mM glucose-6-phosphate as substrate and approximately 1 mg protein from the different fractions. The final volume was 0.7 ml and i t was incu-bated for 15 min at 37\u00C2\u00B0C. The reaction was terminated with 1.3 ml 10% (W/V) TCA and the amount of inorganic phosphate liberated was measured by the G0M0RI method (Gomori, 1942) using the auto analyzer (Technicon). The Pi released in the absence of protein was subtracted before calcu-lating the specific activity. Succinate Dehydrogenase (EC 1.3.99.1). Succinate dehydrogenase is suitable as a marker for mitochondria (Dixon and Webb, 1964a). The activity was measured as described by Pen-nington (1961) using sodium succinate (50 mM) as substrate, INT 0.1% [2-(p-iodophenyl)-3-(p-nitrophenyl)-5-phenyl-2H-tetrazolium chloride] (Eastman) as hydrogen acceptor in 50 mM potassium phosphate buffer pH 7.4 in the presence or absence of aliquots (0.1 ml) from the different fractions. The samples (final volume 1 ml) were incubated for 15 min at 37\u00C2\u00B0C and the reaction terminated with 1 ml 10% (W/V) TCA. The for-mazan was extracted with 4 ml ethyl acetate and i t s absorbence measured at 490 nm. Non-enzymatic reduction of INT was measured and subtracted from the total activity. 17 Electron Microscopic Examination of the Final Preparation An aliquot from the final preparation was centrifuged at 90,000 x g for 10 min. The pellet was fixed in phosphate-buffered osmium tetra-oxide (1%), dehydrated, and embedded in Epon. The sections were stained with uranyl acetate and lead citrate. Evaluation of the Method Marker Fnzymes Table I summarizes the distribution of marker enzymes. The specific activity of a l l the enzymes in the supernatant was low. Glucose-6-phosphatase and alkaline phosphatase were distributed in the same ratio of 1.1 between the \"final preparation\" and the pellet, indicating that the final preparation is truly a microsomal fraction. The high specific activity of the two phosphatases in the pellet is due to incomplete recovery of the heavier plasma membranes fragments and microsomes. How-ever, the distribution of succinate dehydrogenase showed a ratio of 0.1, indicating that the \"final preparation\" contains very few mitochondria. No attempt was made to separate the plasma membranes from the microsomes. The electron microscopic picture of the trophoblastic layer (plates 1, 2) showed a continuous network of endoplasmic reticulum with the plasma membranes which project into the maternal circulation as micro-vil1i. Thus, following homogenization, both form fragments with the same density and size, which together comprise the microsomal fraction. This fraction 2+ was used for studying the Ca -related properties of these placental components. 18 TABLE I. Specific activities of marker enzymes in different fractions, and ratio between the specific activity found in the final preparation (F.P.) over the specific activity found in the pellet. Alkaline * Glucose-6- * Succinate De-Phosphatase Phosphatase hydrogenase** Supernatant 0.22 0.015 0 F.P. 2.35 0.688 0.398 Pellet 2.05 0.602 3.76 Ratio F.P./Pellet 1.14 1.14 0.105 * Activity of Phosphatases (ymole Pi/mg protein per 30 min) ** O.D./mg protein/ 15 min. 19 Electron Microscopic Examination The electron microscopic examination of the final preparation revealed no contamination with mitochondria confirming the results with the enzyme markers. The preparation consisted mostly of plasma membranes in the form of vesicles and fragments (Plates 3 and 4). Calcium binding assay The calcium-binding properties were investigated in a flow dialy-sis system based on that described by Colowick and Womack (1969). The method is based on the principle that the rate of diffusion is propor-tional to the concentration of the free diffusible molecule; this rate w i l l be constant when equilibrium is achieved with the macromolecule in the upper chamber of the dialysis c e l l . Equilibrium w i l l be reached within seconds, and a constant diffusion rate w i l l occur within 1.5-2 minutes when the effluent volume pumped through the lower chamber is about 5 times i t s volume (Colowick and Womack, 1969). The dialysis cell (Figure 1) was prepared as described by Colo-wick and Womack (1969) using a standard cellophane dialysis membrane 45 (Fisher Scientific Company). The placental plasma membranes and Ca (New England Nuclear) were premixed, and the pH adjusted to 8.0 i f neces-sary, before addition to the upper chamber. All experiments were con-45 2+ ducted at room temperature (24\u00C2\u00B0C). I n i t i a l Ca concentration was 1.5 x 10\"5M (approximately 0.04mCi/ml in final reaction mixture) unless indicated differently. l_ \u00E2\u0080\u0094 Plate 3. The final membrane preparation (x 24,282). 21 Plate 4. The final membrane preparation (x 61,560). 22 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u0094 \u00E2\u0080\u0094 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 . T T T 4 5 C a 2 + * M e m C a - M e m D I A L Y S I S M E M B R A N E ^ \u00E2\u0080\u0094' S r ff | A A A A A A A A A A A A A A A A A A A A A A | f r U U U Purrfp F R A C T I O N C O L L E C T O R BUFFER Figure 1. Diagram of the apparatus used for measuring calcium binding by rate of dialysis. 23 The dialysis buffer contained 20 mM Tris-HCl (pH 8.0) and 100 mM NaCl. The NaCl concentration was raised to 100 mM (as suggested by Reed (1973)) in order to prevent an excessive non-specific binding to the dialysis membrane. The flow rate was maintained at 8 ml/min and the effluent was collected in 2 ml aliquots with a fraction collector. One ml samples were added to 10 ml s c i n t i l l a t i o n f l u i d (aquasol-New England Nuclear) and were counted in a Beckman model LS-233 liquid scin-t i l l a t i o n counter. Protein was determined by the procedure of Lowry et aV. (1951). Calculation of calcium binding The difference between the control diffusion rate without plasma membranes and the diffusion rate observed in the presence of plasma mem-2+ branes represents the fraction of bound Ca in the medium. However, the control diffusion rate is not constant because dilution occurs with each addition to the upper chamber, and because of cpm ( 4 5Ca 2 +) loss in the effluent during the experiment. For these reasons the i n i t i a l control diffusion rate must be corrected for dilution and cpm loss for each step. The dilution effect can be duplicated in the control simply by adding the same volumes to the upper chamber without plasma membranes present. The cpm loss cannot be duplicated because differences in the diffusion rates in the presence and absence of plasma membranes produce different rates of loss from the upper chamber. For each experiment, the loss of cpm for each step was calcu-lated from the cpm found in the effluent collected in each step. These figures were then subtracted from the i n i t i a l cpm in the upper chamber 24 to give actual cpm concentration for each step. The total loss of cpm during experiments of 25-30 min, with a flow rate of 8 ml/min was up to 10%. The control diffusion rate for each step was calculated by multiplying the experimental i n i t i a l control diffusion rate by the d i l u -tion factor and by the cpm loss factor. C.D.R. = I.C.D.R. x D.F. x Cpm.L.F. Where: C.D.R. = Control Diffusion Rate, I.C.D.R. = I n i t i a l Control Diffusion Rate at time zero in the absence of plasma membranes, D.F. = Dilution Factor (due to volume increase with each addition to the upper chamber), and Cpm L.F. = Cpm Loss Factor (due to loss of Cpm from the upper chamber to the effluent). Thus: % Ca 2 + bound = C - D , ^ D \" R \u00C2\u00B0 - D ' R - x 100 Where O.D.R. = Observed Diffusion Rate in the presence of plasma membranes. Calcium Binding by Bovine Albumin 2+ For comparison, Ca -binding by Bovine Albumin (Fraction V, Armour Pharm. Comp.) was also studied, under the same conditions as described for the placental plasma membranes. It is assumed that Bovine 2+ Albumin is representative of a class of proteins which do not bind Ca specifically. 25 RESULTS Evaluation of the Method 2+ The s u i t a b i l i t y of the flow dialysis method for Ca -binding 45 2+ studies is demonstrated in Figure 2. The addition of Ca to the upper chamber in the absence of plasma membranes (Figure 2, upper curve) caused a linear increase in diffusion rate (Figure 3), suggesting that 45 2+ only a small fraction of Ca was bound. The lower curve in Figure 2 represents the results obtained under the same conditions but in the presence of plasma membranes in the upper chamber. The difference in the diffusion rates in the absence and the presence of plasma membranes 45 2+ is due to Ca binding by the placental plasma membranes. 40 2+ -2 The addition of a large excess of Ca (10 M) in the absence of plasma membranes produced a sharp spike in the diffusion rate as a 45 2+ 40 2+ result of Ca displacement from the dialysis membrane by Ca . However the amount bound was negligible (0.1% of the total 4 5 C a 2 + present) and after a few samples the diffusion rate returned to the same level 40 2+ as before the addition of Ca . In the presence of plasma membranes, 40 2+ -2 excess Ca (10 M) increased the diffusion rate to the level of the 45 2+ control, indicating complete displacement of Ca from the plasma membranes. The absence of a spike in the presence of plasma membranes 45 2+ is due to a reduced binding of Ca to the dialysis membrane because 45 2+ of the higher a f f i n i t y of the plasma membranes for the available Ca . 26 cpm x10~3 5 . 44 3 J C ONTROL without membranes r i 40c<'i 7\u00C2\u00BB10 /' >' anes I | ^ [ 4.7x10- 7M 4 5Ca 2 < * 3.8\"10\"7M I v 27-10 M J / with membranes t \u00E2\u0080\u00A2 1 1 1 r- 1 r\u00E2\u0080\u0094 10 20 30 40 50 60 70 - i \u00E2\u0080\u0094 80 7 4^ ?\u00E2\u0080\u00A2 SAMPLE NO. 1 . 5 x 1 0 \" ' M 4 5 C a ^ Figure 2. Measurement of Ca diffusion rate at various Ca concentrations \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 without plasma membranes \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 with plasma membranes. The medium and dialy-sis buffer contained 20 mM Tris-HCl (pH 8.0), 100 mM NaCl. Protein concentration was 2.52 mg/ml, the effluent was collected in 2 ml fractions of which 1 ml was counted for 45ca. Each step was allowed 15 samples before i n -creasing 45rja2+ concentration. The results are expressed in cpm/ml effluent. 27 C p m / m l Effluent \"1 0 0 .5 1.0 1.5 2.0 Cpm x l O ~ 6 / m l . o f upper chamber fluid Figure 3. Measurement of Ca diffusion rate as a function of cpm concentration in the upper chamber. The data presented here were obtained from Figure 2 (upper curve). 28 45 2+ A typical Ca diffusion rate profile is presented in Figure 4. The control diffusion rate after the f i r s t step (samples 1-15) was 45 2+ corrected for dilution and loss of Ca , as described in the Methods section. For a l l experiments, the data were f i r s t plotted in this manner, 2+ to allow calculation of the extent of Ca -binding for each step. The 2+ Ca -binding was then replotted against the different parameters used in subsequent experiments. Calcium Binding by Placental Plasma Membranes 2+ -7 -7 At low Ca concentrations (10 -5 x 10 M) calcium binding per mg protein was linear (Figure 5A: data derived from Figure 2), indi-cating a constant percentage bound in this concentration range. In a double reciprocal plot the line passed through the origin, suggesting non-saturable kinetics. O i C O 0_|_ At higher Ca concentration (10\" -10~ M) the amount of Ca bound per mg protein increased sigmoidally (Figure 5B). A Scatchard 2+ plot (Figure 6) of Ca -binding revealed two types of sites: high af-f i n i t y sites accommodating 25+2 (mean + S.E.) nmoles per mg protein -5 and dissociation constant 3.1 +_ 0.4 x 10 M; and low a f f i n i t y sites accommodating 266 +_ 27 nmoles per mg protein and dissociation constant 1.1 + 0.1 x 10\"3M (Table I I ) . Figure 4: Flow dialysis profiles of calcium binding at various calcium concentrations: The medium and dialysis buffer contained 20 mM Tris-HCl (pH 8.0), 100 mM NaCl. At time zero 1.6 x 10\"5M 45r,a2+ was added to the medium in the presence or ab-sence of plasma membranes. The lower curve represents the diffusion rate profile of 45ca2+ in the presence of plasma membranes (protein concentration 2.4 mg/ml) at various 40Ca2+ concentrations, as indicated under the arrows. The upper curve is a control curve in the absence of plasma membranes. Up to sample No. 16 the control curve represents experimental values. The remainder is a corrected control curve compensated for dilution and cpm loss for each step. The flow rate was 8 ml/min, each step was allowed 8 samples and results are expressed as cpm per ml effluent. 30 Log [ C a 2 + ] Figure 5. Ca^T concentration effect on Ca^-binding levels by placental plasma membranes and Bovine Albumin. (A)Ca 2 +-binding at low Ca 2 + concentration (1 \u00E2\u0080\u00A2 10\"7-5 \u00E2\u0080\u00A2 10_7M). This curve was derived from Figure 1. The results are expressed in nmoles Ca 2 + bound per mg protein. (B)Ca2+ binding at higher Ca2+ concentrations (1 -10-6-1 . 10\"2M). This curve was derived from four different experiments carried out under the same conditions as described in Figure 2. The Ca2+ concentration was varied slightly from one experiment to another to cover the wide range. The bound calcium is expressed in nmoles Ca2+ per mg protein. 31 3 0 C H 2 *t\" Bound Ca (nmoles/mg Protein) Free C a 2 + ( i o - 5 M ) Figure 6. Scatchard plot of Ca -binding by placental plasma membranes and Bovine Albumin (derived from the steady values which were used to construct Figure 5B). Table II. Affinity and capacity of placental membranes and Bovine Albumin for Ca2+ K * n ** K n ** S l H *s2* 2 Placental c o Plasma 3.1+0.4x10\"^ 26+2 1 .l+O.lxlO^M 266+27 Membranes Albumin 1.4+0.06x10\"4M 28+1 9.1+0.6xlO~4M 99+5 32 All the results are expressed as mean +_ S.E. * Ks = dissociation constant * 2+ nmoles Ca per mg protein 33 Calcium Binding by Bovine Albumin The results are plotted in Figures 5-6 and summarized in Table 2+ II. Two types of sites for Ca were revealed. The high a f f i n i t y sites have the same capacity as the high a f f i n i t y sites of the plasma membranes but the af f i n i t y is almost five-fold lower. The low af f i n i t y sites have the same aff i n i t y with only 40% of the capacity of the low af f i n i t y sites of the plasma membranes. Time, Temperature, Protein Concentration and pH Effects The maximal diffusion rate was achieved after 1.5 min, and when no further additions or treatments were performed, the diffusion rate was constant (after compensating for cpm loss) for at least 30 min. Thus calcium binding by placental plasma membranes reached a maximum within 1.5 min, and thereafter was independent of incubation time. Reducing the temperature from 24\u00C2\u00B0C to 4\u00C2\u00B0C did not change signi-2+ ficantly the extent of Ca -binding. 2+ The binding of Ca increased linearly with protein concentration 2+ -5 over the range 0.25-2.7 mg protein/ml when [Ca ] was 1.5 x 10 M (Figure 7). Calcium binding was dependent on pH, and practically no binding was detected at pH 4.0. The binding increased with increasing pH up to pH 11.0, and leveled off (Figure 8). Since the membranes were sub-jected to each pH for no more than 2.5 min, denaturation effects of extreme pH's were minimized; the pH effect was reversible over the entire pH range (3.0-11.8). / / 1 1 1 1 r 0 1 2 3 4 5 mg Protein/ml Figure 7. Ca binding as a function of placental plasma membrane protein concentration. The medium contained 20 mM Tris-HC1 (pH 8.0), 100 mM NaCl, and 1.5 \u00E2\u0080\u00A2 10\"5M 45ca 2 . The plasma membranes were added to the upper chamber to give the indicated concentration. The results are expressed as total nmoles Ca2+ bound. 35 40 J Figure 8. pH effect on Ca*\"T binding. The pH in the upper chamber was measured with a pH electrode, and controlled by addition of HC1 or NaOH (0.1 M). The plasma membranes were exposed to each pH for not more than 2.5 min. The medium contained 20 mM Tris-HCl, 100 mM NaCl 2.2 \u00E2\u0080\u00A2 10\"3M 45Ca2+ a n d protein concentration was 2.5 mg/ml. The results are expressed as total nmoles Ca2+ bound. 36 2+ 2+ 2+ Ca Displacement from Its Sites by Mg and Sr 4c 40 2+ Figure 9A shows the Ca displacement from i t s sites by Ca , 2+ 2+ 45 2+ Mg , and Sr . The percent Ca bound at the end of the f i r s t step 2+ 2+ was regarded as zero percent displaced. The ab i l i t y of Mg and Sr 45 2+ 40 2+ to displace Ca from i t s sites was much lower than that of Ca (Figure 9A). 45 2+ The double reciprocal plots for displacement of Ca from the AC) 9J- OX. 9 4 -plasma membranes by Ca , Mg , and Sr (Figure 9B, C, D) revealed two apparent Km's for each ion. These values appear in Table II I . The 2+ 2+ 2+ a f f i n i t y of Mg and Sr for the high a f f i n i t y Ca sites was approxi-40 2+ 2+ mately 10-fold lower than that of Ca . The af f i n i t y of Mg for the 2+ 2+ high a f f i n i t y Ca sites was higher than that of Sr . The wide gap of af f i n i t i e s was narrower for the low af f i n i t y sites (4-fold lower for 2+ 2+ Mg and 8-fold lower for Sr ). These results indicate high specificity of these sites for calcium, with greater specificity in the high a f f i n i t y sites. DISCUSSION The flow dialysis method was found in the present study to be a 2+ very useful tool for studying Ca -binding. The process of binding has met the requirements for the use of this method as defined by Colowick and Womack (1969): 1) chemical equilibrium was achieved within a few seconds; 2) a constant diffusion rate was established after 1.5 min as predicted; 3) the dissociation constants of this binding reaction were 37 Figure 9A. Ca displacement by divalent cations. (A) The experi-ments were carried out as described for Figure 2. 40ca2 +, Mg2+, and Sr^+ were added in increasing concentrations to the upper chamber as indicated. The results are expressed as a percentage of ^ 5r;a2+ displaced from the plasma membranes. Zero displacement is defined as the amount of 45ca2+ bound at the end of the f i r s t step. 38 Figure 9B, C and D. Double reciprocal plots for Ca displacement by 4 0 C a 2 + , Mg2+, and S r 2 + , respectively. D is expressed as percent 45ca2+ displaced and [S] as millimolar divalent cation. These plots were derived from data used in Figure 9A. 39 A1^ ?+ 40 2+ 2+ TABLE III. Apparent Kin's for Ca displacement by \u00E2\u0084\u00A2Ca , Mg . S r 2 + . The Km's were calculated from the double reciprocal plots presented in Figure 9B-D. Krm Km2 ION (M+S.E.) (M+S.E. 4 0 C a 2 + (2.2 + 0.3) x 10~4M (2 + 0.2) x 10\"3M Mg 2 + (1.4 + 0.15) x 10~3M (9 + 1) x 10~3M S r 2 + (2.2 + 0.25) x 10\"3M (1. .5 +0.1) x 10\"2M 40 -3 -6 within the effective range of the method (10\" -10\" M). The use of 45 2+ Ca as the diffusible molecule, which has a relatively high diffusion rate through the dialysis membrane, made this system sensitive to changes 45 2+ in free Ca in the upper chamber. 2+ The flow dialysis system offers several advantages for Ca -binding measurements over equilibrium dialysis and u l t r a f i l t r a t i o n . It is faster than the equilibrium dialysis method, since each step can be completed within 2 minutes. Washing the membrane, which is necessary in the u l t r a f i l t r a t i o n method, is avoided here; this eliminates a possible 45 displacement of Ca, particularly from the low af f i n i t y sites. The 45 2+ continuous monitoring of the Ca diffusion rate gives a clearer picture of the kinetic behavior of the investigated reaction. In studying pH effects, the membranes are exposed to each pH for not more than 2.5 min. Thus denaturation effects by extreme pH's can be minimized. Using the other methods, the exposure to each pH is much longer and denaturation 2+ effects cannot be excluded. However in calculating Ca -binding using the flow dialysis method two effects must be taken into account: the dilution effect and cpm loss during the experiments. These two factors should be calculated for each step to give the actual control diffusion 2+ rate for each step. An excessive nonspecific Ca -binding to the dialysis membrane was successfully prevented by raising NaCl concentration up to TOO mM, as suggested by Reed (1973). In contrast to the u l t r a f i l -tration method the flow dialysis method (without modfication) is not sensitive enough to measure calcium uptake. Because uptake is a rela-tively slow process, the effect on the diffusion rate w i l l be small, 41 and d i f f i c u l t to detect. Another limitation of this method is that i t requires high protein concentrations (2-3 mg/ml). 2+ Two types of Ca -binding sites can be distinguished on the placental plasma membranes: high a f f i n i t y sites with a capacity of 26 -5 nmoles per mg protein and Ks = 3.1 x 10 M,and low a f f i n i t y sites with -3 a capacity of 266 nmoles per mg protein and K = 1.1 x 10 M. Bovine Albumin which was used as a reference in this study, was also shown 2+ to possess two types of Ca -binding sites. While the capacity of the high a f f i n i t y sites for albumin was equal to that of the membranes, the af f i n i t y was five-fold lower (K s l = 1.4 x 10~4M). It is tempting to 2+ speculate that the higher a f f i n i t y for Ca of the placental plasma 2+ membranes may be related to a function in Ca transport. Membranes from several sources show more than one class of calcium-binding sites. The sarcoplasmic reticulum has been most intensively 2+ studied with respect to i t s Ca -binding properties, but even with this one type of membrane, different methods have yielded conflicting results. Cohen and Selinger (1969) reported two classes of sites with the same -5 dissociation constant (Kg = 4 x 10 M), while a more recent study by Chevallier and Butow (1971) revealed three classes of sites with the following dissociation constants: 1.3 x 10~6M; 3.2 x 10~5M; 3.2 x 10\"4M. 2+ Despite the d i f f i c u l t i e s one faces trying to compare Ca -binding properties of preparations obtained by the various methods, the existence of high a f f i n i t y sites is typical of membranes that are involved in the active regulation of intracellular calcium. In addition to the sarco-plasmic reticulum, high a f f i n i t y sites for calcium are found in the 42 mitochondria (Reynafarje and Lehninger, 1969) and in cardiac microsomes (Repke and Katz, 1972). Lower af f i n i t y sites are found in rat l i v e r plasma membranes (Schlatz and Marinetti, 1972) (Kg = 2.5 x 10~4M and 3.1 x 10 M), and in red blood cell membranes (Gent e_t al_., 1964) -4 2+ (Ks = 2.8 x 10 M). The placental plasma membranes, in respect to Ca --5 binding a f f i n i t y (Kg = 3.1 x 10 M), ranks with the sarcoplasmic r e t i -culum, mitochondria and cardiac microsomes. It is interesting to note that l i v e r plasma membranes (Schlatz and Marinetti, 1972) have binding sites with a 10-fold higher dissociation constant and a correspondingly lower a f f i n i t y for calcium than those sites in the placental plasma membranes. It is tempting to attribute this difference in a f f i n i t y for 2+ Ca to the fact that the l i v e r plasma membranes, unlike the placental 2+ plasma membranes, are not involved in active Ca -transport between body compartments. 2+ The pH-dependent Ca binding profile by the placental plasma membranes shows a pH optimum of 11.0, and at pH 4.0 there is practically no binding. Up to pH 7.0 the pH profile agrees with that already des-cribed for l i v e r plasma membranes (Schlatz and Marinetti, 1972), sarco-plasmic reticulum (Cohen and Selinger, 1969), and muscle microsomes (Carvalho, 1966). Above pH 7-8 a decrease was reported (Cohen and Selinger, 1969; Gent et al_, 1964). In the present study no such decrease was observed even above pH 11.0. These differences can be at least partly accounted for by the different methods used. The time of exposure to each pH varied from 40 min (10 min incubation and 30 min centrigu-gation) (Cohen and Selinger, 1969) up to 42 hr (Shlatz and Marinetti, 1972) using the equilibrium dialysis method. The use of flow dialysis 43 reduced the exposure time to 2.5 min, minimizing long term pH effects. In addition, no attempt to reverse the pH effect was mentioned in these studies (Cohen and Selinger, 1969; Shlatz .and Marinetti, 1972) so that long term effects cannot be excluded. 2+ 2+ 2+ The a f f i n i t y of Mg and Sr for the Ca high a f f i n i t y sites 2+ 2+ was found to be 10-fold lower than that of Ca . The binding of Mg 2+ and Sr was not studied, so that the possibility of specific sites for + 2+ these ions cannot be excluded. The effect of Na on Ca binding could not be studied in detail since i t was necessary to maintain a high Na+ 2+ concentration to prevent nonspecific Ca -binding to the dialysis mem-brane. However, in several short experiments when NaCl (100 mM) was added after the f i r s t step (before measurable amounts of 4^Ca 2 + were bound to the dialysis membrane, which is time dependent) (Reed, 1973) 45 2+ no significant changes in Ca diffusion rate could be detected. The 45 2+ 2+ 2+ more efficient displacement of Ca by Mg and Sr from the low 2+ affi n i t y sites indicates a reduced specificity for Ca . 2+ The specificity of Ca -binding sites of different membranes shows wide variations. The sites on the l i v e r mitochondria (Reynafarje 2+ 2+ 2-1\" and Lehninger, 1969) are insensitive to Mg , but Sr displaces Ca quite effectively. The sarcoplasmic reticulum (Cohen and Selinger, 1969) sites are unaffected by Mg up to 10~ M, while Sr displaces at a level 2+ close to that found in this study. The Ca -binding sites of another plasma membrane (derived from l i v e r ; Sclatz and Marinetti, 1972) are \u00E2\u0080\u00A2f* \"4~ 2*^~ 24~ insensitive to K and Na while Mg reduced Ca -binding to the low af f i n i t y sites only. The degree of specificity observed in the present 44 study enables us to conclude that at physiological concentrations of 2+ 2+ 2+ Mg , Ca and normal levels of Sr , practically no binding other than 2+ Ca to these sites w i l l be detected. 2+ The concept behind this study is that a passive binding of Ca to the membrane is an essential f i r s t step in the process of an active 2+ Ca transport. The placental plasma membranes were found to contain 2+ sites for Ca with capacity, specificity and aff i n i t y within the range 2+ reported for other membranes involved in active transport of Ca . Unlike the sarcoplasmic reticulum and mitochondria, which are exposed 2+ to only two different Ca concentrations, the placental plasma membranes 2+ are exposed to three different Ca concentrations. The outer plasma 2+ membranes on the maternal side are exposed to a Ca concentration of 2.9 mequiv/L, and on the fetal side to 3.8 mequiv/L (Papadopoulos et a l , 2+ 1967). The inner surface of the membranes is exposed to a low Ca -5 -6 concentration, estimated to be around 10 -10 M. It is premature to assign the distribution of the two classes of sites to the different surfaces of the placental plasma membranes. However, i t should be noticed that the dissociation constants obtained are close to the physio-2+ logical concentrations of Ca to which the membranes are exposed. 2+ Although high a f f i n i t y specific Ca sites are present in placental plasma membranes, the direct involvement of these sites in active trans-2+ port of Ca remains speculative at this stage. 45 SUMMARY The Ca-binding properties of placental plasma membranes were studied using a flow dialysis system. Ca-binding was not detectable at pH 4.0, but increased at higher pH's to a maximum binding at pH 11.0. Two types of Ca-binding sites were identified: high a f f i n i t y -5 sites with dissociation constant Kg = 3.1 x 10 M and a capacity of _3 26 nmoles per mg protein; low aff i n i t y sites with K = 1.1 x 10 M and a capacity of 266 nmoles per mg protein. 2+ 2+ The a f f i n i t i e s of Mg and Sr for the high a f f i n i t y sites 2+ were 10-fold lower than that of Ca , and for the low af f i n i t y sites were 4- and 8-fold lower respectively. 2+ The placental plasma membranes contain sites for Ca with capacity, specificity and af f i n i t y within the range reported for other 2+ membranes involved in an active transport of Ca (mitochondria, sarco-plasmic reticulum, cardiac microsomes). DIVISION II -STIMULATED ATPase OF THE GUINEA PIG PLACENTAL PLASMA MEMBRANES 46 CHAPTER II CHARACTERIZATION OF CALCIUM-STIMULATED ATPase INTRODUCTION It has been proposed that calcium transport across the placenta is an active transport (Papadopoulos et al_., 1967). Most energy-utili-zing processes are driven by the breakdown of ATP (Dixon and Webb, 1964b). To be effective the hydrolysis of ATP must be enzymatic and coupled with the energy-utilizing process, in this case, calcium transport. The coupling between the two events must take place within the membrane, so that the enzyme hydrolyzing the ATP is fixed in space; this makes possible the coupling between the chemical reaction (ATP hydrolysis) and translocation of the molecule (Curie-Prigogine principle) (Katchal-ski and Curran, 1967). Skou (1965) has established the relationship between Na+ and K+ + + 2+ fluxes and ATP hydrolysis by the membrane-bound (Na , K )-activated Mg -dependent ATPase (EC 3.6.1.3).. Other membrane-bound ATPases activated 2+ 2+ 2+ by Ca and Mg , and involved in Ca -transport, have been described in the red blood cell (Schatzmann and Rossi, 1971) and the sarcoplasmic reticulum (Martonosi and Feretos, 1964). 47 48 2+ 2+ In this chapter the properties of such a (Ca , Mg )-activated ATPase located in the placental plasma membranes are described, and 2+ 2+ compared with the properties of other (Ca , Mg ) ATPases. MATERIALS AND METHODS 2+ Histochemical Localization of Ca -ATPase in the Placenta The placenta, after removal, was frozen (to conserve enzymatic activity) and sectioned (20p). The sections were incubated for an hour in 20 mM Tris-HCl buffer pH 8.5, 30 mM CaCl 2 and 12 mM ATP. The sections were stained as described by McManus et_ al_ (I960). The staining method is based on precipitating the released Pi as calcium phosphate, replacing the calcium with cobalt by washing with 2% cobalt acetate and replacing the Pi with sulfide by placing the sections for one minute in 2% aqueous yellow ammonium sulfide. The result is that the ATPase sites are stained brown-black by the precipitate of cobalt-sulfide. Though this method 2+ 2+ is not specific for Ca -ATPase, the presence of high Ca concentration (30 mM) and 12 mM ATP in the incubation medium make the contributions of other phosphatases to ATP hydrolysis negligible. ATP Hydrolysis by the Placental Plasma Membranes To test the various enzymatic properties, 0.1 ml of membrane suspension containing 10-20 ug of protein, was incubated with 1.0 ml of solutions containing calcium or magnesium in appropriate amounts (0.1 to 10 mM); 20 mM Tris-HCl buffer (pH 8.2) and 70 mM Na+ (as NaCl); Na?ATP (Sigma) was added to make 5 mM. Blank specimens did not contain bivalent cations. The samples were incubated for 30 min in a Dubnoff shaking water bath at 37\u00C2\u00B0C. The reaction was terminated by plunging the tubes into an ice-water bath and adding 1.0 ml 10% (W/V) trichloro-acetic acid. The rate of ATP hydrolysis was determined by measuring the amount of Pi released from the samples. Using the AutoAnalyzer inorganic phosphate was measured by G0M0RI (1942) method, total protein by the Lowry et al_. (1951) procedure. Results are expressed as umole Pi released per mg of protein in 30 min. Every value of the various experiments was the mean of t r i p l i c a t e samples and every whole experi-ment was repeated at least three times. The following compounds were obtained from the sources indicated 1. Na\u00C2\u00A3ATP (Sigma No. A-3127); 2. Tris-ATP [Di-Tris (hydroxymethyl)-amino-metane s a l t ] ; 3. ITP-Inosine 5'-triphosphate sodium salt (Sigma No. 1-5000); 4. GTP-Guanosine 5'-triphosphate sodium salt (Sigma No. G-8752); 5. ADP-Adenosine 5' diphosphate disodium salt (Sigma No. A-0127); 6. AMP-Adenosine 5'-monophosphoric acid sodium salt (Sigma No. A-1877); 7. EDTA-disodium ethylenediamine-tetraacetate (Fisher S-311); 8. EGTA - ethyleneglycol-bis-(-aminoethyl ether)n, n'-tetraacetic acid (Sigma No. E-3251); 9. Mersalyl acid (Sigma No. M-8125); 10. Oligomycin (Sigma No. 0-1295); 11. Ruthenium Red (K and K Laboratories No. 2603-A); and 12. Maleic anhydride (Fisher A-168). 50 RESULTS Histochemical Localization of ATPase Activity in the Placenta Plate 5 shows clearly that the ATPase activity (black-brown staining) is localized in the trophoblastic layer between the maternal and the fetal circulations. 2+ Distribution of Ca -ATPase Activity 2+ The Ca -ATPase activity of the different fractions (see Chapter I for definition of the fractions) was assayed in the presence of 5 mM 2+ Ca under standard conditions as described in the Methods section. The distribution was as follows: supernatant, 1.3; final preparation, 16.8; and pellet, 15.0 (nmoles Pi/mg protein per 30 min). This follows the distribution of alkaline phosphatase and glucose-6-phosphatase (Table I ) , with the same ratio (1:1) between the activity in the final pre-paration and the pellet. These results indicate that the three phos-phatases are located in the membranes of the microsomal fraction. Activation of the Enzyme by Divalent Cations Activation of the enzyme by calcium ions was assessed by varying the calcium concentration in the incubation medium from 0.02 to 10 mM. Figure 10 depicts the rate of Pi production at the various calcium con-2+ centrations. The apparent for Ca of placentas from 5 different guinea pigs was 0.26 +0.01 mM (m^ SErn) (Figure 11). The rate of Pi release at peak activity ranged from 15.0 to 22.0 pmole Pi per mg protein in 30 min. 51 Plate 5. Histochemical localization of ATPase activity in the guinea pig placenta. F = fetal circulation, M = maternal circulation, and TR = trophoblast. The brown staining indicates ATPase activity. 52 mM divalent cation Figure 10. Stimulation of ATP hydrolysis by divalent cations. Unless indicated incubation fluids contained 70 mM Na+ (as NaCl), 20 mM Tris-HCl (pH 8.2) and 5 mM Na2ATP. \u00E2\u0080\u00A2 \u00C2\u00A9activation by Ca2+; \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 Ca2+ + Mg2+ (in equimolar concentrations); X X Mg2+; \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 Mn2+; 0 0 Sr2+. 53 \"i 1 1 1 \u00E2\u0080\u0094 \u00E2\u0080\u0094 i 1 1 1 1-0 1 2 3 4 ,5 6 7 8 9 10 1/tS] (mM)\"1 Figure 11. Variations in Lineweaver-Burk plots of ATPase activity at various concentrations of ATP, Ca 2 + and Mg2+. V in ymole Pi formed per mg protein in 30 min [S] in mM. The incubation mixture contained 20 mM Tris HCl (pH 8.2), 70 mM Na+. 9. mM ATP + Mg^ .2+ X 0 mM ATP + Ca' mM Mg2+ + ATP mM Ca 2 + + ATP 54 In the absence of calcium, magnesium ion also activated the en-zyme but always less effectively than calcium (Figure 10). The apparent for magnesium was 0.56 +_ 0.03 mM (m+SEm) (Figure 11). When instead 2+ 2+ of the single cation a 1:1 combination of Ca + Mg was used, the resultant curve lay between the two curves obtained from incubation with 2+ 2+ Ca or Mg alone (Figure 10). 2+ 2+ The effect of other divalent cations (Mn , Sr ) on enzyme a c t i -vation was also tested. Figure 10 shows that manganese activated the enzyme but strontium did not. Maximal activation of the enzyme with 2+ Mn occurred at 2 mM, higher concentrations producing inhibition. 2+ 2+ Activation of enzyme by Ca and Mg was further tested by 2+ using a constant Ca concentration (5 mM) and adding increasing amounts 2+ 2+ of Mg , as well as by using a constant Mg concentration (5 mM) and 2+ 2+ adding increasing amounts of Ca (up to 8 mM). The addition of Ca 2+ 2+ to 5 mM Mg led to increased stimulation, whereas the addition of Mg 2+ to 5 mM Ca led to inhibition of the enzyme. Figure 12 gives the curve so obtained. The point of intersection of the two curves was at 5 mM 2+ 2+ Ca^ + 5 mM Mg . The Effect of Incubation Time on the Concentration of H+ and Pi in the Medium\" Frequent recording of pH and Pi release (Figure 13), revealed that during 4 hrs of incubation the pH dropped only 0.1 pH unit, and Pi release was linear. During the f i r s t 30 min the drop in pH was very small indicating that the selection of Tris as buffer was appropriate. 55 5 5+2 5+4 5+6 5+8 D i v a l e n t c a t i o n m M Figure 12. Enzyme activation by Ca and Mg . 2+ 2+ o oConstant Ca (5 mM) + varying Mg concentrations. 2+ 2+ \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 Constant Mg (5 mM) + varying Ca concentrations. 2+ \u00E2\u0080\u00A2 \"Ca alone \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 Mg2+ alone Incubation fluids contained 20 mM Tris HCl (pH 8.2) 70 mM Na+ (as NaCl) and 5 mM NaJ\TP. I c CO o \u00C2\u00BB_ CL E _0) o 6 lOO-i 90-80-70-60 50 4 OH 30-1 20 IOH r 1 1 1 1 1 0 0.5 I 2 3 4 Incubation Time (Hours) Figure 13. The effect of incubation time on Pi release and pH of the incubation medium. The incubation fl u i d contained 5 mM Ca2+, 70 mM NaCl, 5 mM Na2ATP and 20 mM Tris-HCl (pH 8.2). 57 Protein Concentration Effect on Pi Release Inorganic phosphate release was linear within the protein con-centration range tested 5-60pg/ml incubation f l u i d (Figure 14). Sodium Independence of the Enzyme The requirement of sodium ions for enzyme activation was tested by incubating the enzyme in 0 or 70 mM Na and adding Tris-ATP instead of Na2ATP. Specimens incubated without sodium showed slightly higher activation than those incubated with sodium (Figure 15). The Effect of pH on ATP hydrolysis The calcium and magnesium activation curves of the enzyme were 2+ obtained by incubating the enzyme preparation with 0 or 5 mM Ca or 2+ Mg at a pH ranging from 6.5 to 10.7. In each instance the pH was adjusted both in the incubation f l u i d and Na,>ATP solutions and determined before and after incubation. Figure 16 shows the curves obtained. The 2+ pH optimum of Ca ATPase was between 8.2 and 8.5. At pH 7.1 and 9.6 the enzyme was stimulated by calcium ions to only 50% of peak activity. 2+ The pH optimum of Mg ATPase was between 8.2 and 9.3 with approximately 60% of the activation by Ca 2 +. Below pH 7.0 and above pH 9.7, Mg 2 + 2+ produced higher activation than Ca . Inhibitors The effects of different inhibitors on calcium activation of the enzyme were tested. Ouabain (1 mM) was added to the solutions 58 Figure 14. Protein concentration effect on Pi release. The incubation medium contained 5 mM ATP, 20 mM Tris-HC1 (pH 8.2) and membrane protein at the concentrations as indicated. 59 o CO c Q) O a E aT J D o E 3. 3.0 4.0 8.0 mM Ca2+ Figure 15. Effect of Na on enzyme activation by Ca + e \u00E2\u0080\u00A2 without Na+, Owith 70 mM Na . Incubation medium contained 20 mM Tris-HCl (pH 8.2), and 5 mM Tris-ATP. 60 I c E o c o c o i_ D_ E \u00E2\u0080\u00A2 P1 > 0.97; 0.995 > P 2 > 0.99; P 3 = 0.97; 0.9 > P 4 > 0.75. Since the concentration ratio is the only variable and a l l other 2+ values are constant, i t is possible to plot A (activity) against [Mg ]/ 2+ 2+ 2+ [Ca ] while either Mg or Ca is at optimal concentration (5 mM). The calculated theoretical curve so obtained by using the above equation and the experimental results are shown in Figure 26A. It is also possible to calculate the change in the apparent Km for the total divalent cations 85 Divalent cation mM Figure 25. The effects of Ca^T and Mg^ on ATP hydrolysis. The incu-bation medium contained 70 mM NaCl, 20 mM Tris-HCl (pH 8.2), 5 mM Na2ATP and Ca 2 + or Mg2+ or Ca 2 + + Mg2+ in concentration as indicated. Results expressed as percent of activity at 5 mM Ca 2 + which ranged between 15 and 22 ymole/Pi per mg protein in 30 min at 37\u00C2\u00B0. o o activation by Ca 2 +, \u00E2\u0080\u00A2 \u00C2\u00B0 activation by Mg2+; I,H,m, theoretical curves obtained by using the equation. , \u00E2\u0080\u00A2 , experimental values for 5 mM Ca2+ and increasing Mg 2 + concentrations; , \u00E2\u0080\u00A2 , experi-mental values for 5 mM Mg2+ and increasing Ca2+ concentrations, *, experimental values for 10 mM Mg2+ and increasing Ca2+ concentration. 86 Figure 26. A - The effect of changing the concentration ratio of Mg^ ' to Ca 2 +, while at least one cation is at the optimal concentration of 5 mM. The solid line represents the theoretical, calculated curve and individual dots (\u00E2\u0080\u00A2 \u00E2\u0080\u00A2) are experimental values obtained when the Vmax2 for Mg2+ was 60% of Vmax] for Ca 2 +. 2+ 2+ B - The change in Km as a function of Mg /Ca . V values for the various combinations obtained from Figure 26A. 87 2+ 2+ 2+ as a function of Ca /Mg concentration ratio. The \"Km\" drops from 0.54 mM for Mg2+ to 0.25 mM for Ca 2 + (Figure 26B). Finally a three dimensional model describing the relationship 2+ 2+ 2+ between [Mg ]/[Ca ] and the activity of the placental Ca -ATPase was built. In this model the Y and the X axes are the concentration of Ca 2+ and Mg respectively, and the Z axis is the velocity of the reaction. 2+ 2+ The surface formed by the activation curves at different [Mg ]/[Ca ] represent the velocity of the reaction under any possible combination 2+ 2+ up to the 10 mM Ca +10 mM Mg . The model is shown in Figure 27. To test further the validity of equation (19) for competition between substrates, the formula was used for predicting the velocity of the reaction as a function of ATP/ADP concentration ratio. ADP is a production of ATP hydrolysis by ATPase and can i t s e l f be hydrolyzed further by the enzyme. The hypothesis was that i f ADP and ATP compete for the same s i t e , Pi release w i l l be inhibited. The following pathway was assumed: ADP K a s ATP K 3 C ADP as Ca-ATP-Ca-ADP' Km ATP Km ADP E-Ca-ATP E-Ca-ADP E + P E + P The formation of the Ca-ADP and Ca-ATP complexes is dependent on the association constants. Log K for Ca-ATP = 3.6, and log K for Ca-ADP = 2.78 as quoted by Williams (1959). The Km for ATP is 0.1 mM (Figure 11) 88 Figure 27. A three dimensional model describing the relationship between [Mg2+]/[Ca2+] and the activity of the placental Ca2+-ATPase. The Y and X axes are the concentration of Ca2+ and Mg2+ respectively, and the Z axis is the velocity of the reaction. 89 and for ADP Km = 0.15 mM (calculated from Figure 22). Samples were incu-bated with 5 mM ATP ADP and 2.5 mM ATP + 2.5 mM ADP. V] for ATP was considered as 1.0, and V2 with 5 mM ADP was 0.69. Using a l l the constants mentioned, equation (19) w i l l be as follows: A =[ KasATP(K^ATP)LATPj ' VgADP + MOO - KasATP(K^ATP)|ATPj ) V 1 A T P ^KasADP(KmADP)LADPJ + V \ KgsADP(y\DP) LADPJ + 1 A = ( 7782 (0?15) (2.5) , )\u00C2\u00B0- 6 9 + ( 1 0 0 ~ 7782 (0J5) (2.5T , ) 1 , 0 \ 892 (0.10) (2.5) '/ \ 892 (0.10) (2.5) '/ A = 4.89 + 92.9 = 97.8% Thus the calculated velocity in the presence of 2.5 mM ATP + 2.5 mM ADP is 97.8% of that with 5 mM ATP alone. The actual experimental value was 96% for 1:1 [ATP]/[ADP] concentration ratio, which is in good agree-ment with the predicted value. Thus ATP and ADP appear to be hydrolyzed by the same s i t e , and the inhibitory effect of ADP is very small. Even when [ATP]/[ADP] = 0.1, Pi release w i l l decrease only 13% (calculated value). DISCUSSION The use of the formula presented in this chapter was very success-2+ 2+ ful in predicting the velocity of the placental (Ca -Mg )-ATPase incu-2+ 2+ bated with numerous combinations of [Ca ]/[Mg ]. It is concluded 90 2+ 2+ that Ca and Mg compete for the same s i t e , or more precisely the corresponding complexes with ATP compete for the same site. It has been shown very clearly that the placental ATPase activity can be regu-2+ 2+ lated by the Ca to Mg concentration ratio. The role of these two ions in regulating cellular enzymes has been the subject of several studies. Shikama (1971) calculated the standard free energy for ATP 2+ 2+ hydrolysis as a function of pMg and pCa. He found that the [Ca ]/[Mg ] ratio plays an important role as a sensitive modifier of the thermodynamic potential of the ATP molecule. Thus the energy that can be derived from ATP hydrolysis by ATPase is regulated by at least two steps, both depen-dent on [Ca 2 +]/[Mg 2 +]. Step I, the rate of ATP hydrolysis; Step I I , the change of free energy (-AG\u00C2\u00B0) for ATP hydrolysis. 2+ 2+ The role of Ca and Mg in the ATPase of sarcoplasmic reticulum was studied by Panet et al_. (1971). Their results suggest that phosphory-lation of the enzyme (an intermediate step in ATP hydrolysis) is dependent 2+ 2+ on Ca , while dephosphorylation is increased by Mg and decreased by 2+ 2+ 2+ Ca . In the placental (Ca -Mg )-ATPase i t is unlikely that this is true, since the hydrolysis of ATP proceeds with either ion alone. Though the possibility of formation of a phosphorylated intermediate exists for the placental enzyme as well, no attempt was made to study this aspect. The a b i l i t y to predict accurately the velocity of Pi release in the presence of ATP and ADP indicates the wider use that can be made of this formula. It also suggests that ATP and ADP are hydrolyzed by the same site and that product inhibition due to an increase in ADP concentration is negligible. 91 In addition, i f one is interested in the contribution of each substrate to the total activity, i t can be derived easily since each term in the formula represents the respective contributions. Though i t is possible to adopt this equation to predict the activity of an unsaturated enzyme, the prediction w i l l be less accurate than for the saturated enzyme. SUMMARY 2+ 2+ The velocity of the reaction of a Ca Mg -stimulated ATPase in placental plasma membranes of the guinea pig was found to depend on 2+ 2+ the [Mg ]/[Ca ] concentration ratio. The observed activation curve of the enzyme while changing the concentration ratio agrees with the theoretical equation, derived for the case in which two ions activate the same site while the enzyme is saturated. Because of the good agree-2+ 2+ ment, i t is concluded that the two divalent cations, Ca and Mg , activate this placental ATPase at the same site. The regulatory effect 2+ 2+ of [Mg ]/[Ca ] on the placental ATPase activity, can be visualized by a three dimensional model. The equation may be used to determine whether or not two substrates compete for the same s i t e , and what is the relative contribution to the total activity by each substrate. This was demonstrated with ATP and ADP, which are both hydrolyzed at the same site. The equation is not specific for this enzyme and i t can be used for other enzymes as well, provided the necessary constants are available, or can be calculated. CHAPTER IV COMPARISON BETWEEN ALKALINE PHOSPHATASE AND Ca2+-ATPase INTRODUCTION Alkaline phosphatase is one of the most studied placental en-zymes, and i t s specific activity level serves as a general index of placental function (Curzen and Morris, 1968). However, it s physiolo-gical role and even it s naturally occurring substrate have not been 2+ defined. It has been suggested that in the intestine Ca -ATPase a c t i -vity is just another manifestation of alkaline phosphatase (Haussler ejtal_., 1970; Russel et al_., 1972). Since Ca2+-ATPase is implicated 2+ in Ca -transport, alkaline phosphatase can be matched loosely with 2+ Ca -transport. While this study does not oppose this suggestion re-garding the physiological role of alkaline phosphatase, i t does argue that the two enzymes are not identical. As described in previous chapters, i t was found that the two enzymes are located in the microsomal fraction. Since no specific inhibitor is available for either enzyme, a different approach was taken to separate the two enzymes by selective purification. The dissociation of these two phosphatase activities (ATPase and mono-phosphatase), is made by comparing key properties of the two throughout the process of purification of alkaline phosphatase. If the two enzymes 92 93 are not identical, the purified alkaline phosphatase fraction should 2+ not demonstrate Ca -ATPase activity. An additional subject described in this chapter is the specific 2+ activity profile of Ca -ATPase in developing guinea pig placenta through-out the gestation period. Its relation with the reported transplacental calcium flux and fetal calcium deposition (Twardock and Austin, 1970; Twardock, 1967) is considered. The logic behind this study is that since the placenta is a dynamic organic which responds to the changing requirements of the fetus, i t may be expected that the placental enzy-matic picture w i l l be modified accordingly to meet the changing require-ments of the fetus for calcium. MATERIALS AND METHODS Purification of Guinea Pig Placental Alkaline Phosphatase Alkaline phosphatase was purified by the method of Ghosh and Fishman (1968), in a procedure involving three major steps: 1) Butanol extraction to disrupt the lipoprotein association; 2) Ammonium sulphate precipitation; 3) Gel f i l t r a t i o n using sephadex G-200. Lipids were extracted from the membrane preparation by addition of butanol (20 ml butanol to 50 ml membrane preparation). The enzyme passes to the aqueous phase with slow stirring for 1 hr at 4\u00C2\u00B0C and then for 10 min at 37\u00C2\u00B0C. The extract was centrifuged (Sorvall RC2-B) for 30 min at 14,600 x g and the aqueous phase was centrifuged again for 30 min at 35,000 x g. The supernatant (which contains the enzyme) was 94 dialyzed overnight against 50 mM Tris-HCl pH 8.6 (10 volumes) at 5\u00C2\u00B0C-6\u00C2\u00B0C. The proteins were then precipitated with 90% (NH^)^ SO^ , extracted with 40% (NH 4) 2 S0 4 and reprecipitated with 80% (NH 4) 2 S04- The resul-tant precipitate was suspended in 50 mM Tris-HCl pH 8.6 and dialyzed over-night under the same conditions as the previous dialysis. The column was prepared as described by Ghosh and Fishman (1968) using a 60 x 2.5 cm column and sephadex G-200 gel. The enzyme solution was layered on the top of the column, eluted with 50 mM Tris-HCl pH 8.6 and collected in 10 ml fractions. The protein concentration of the fractions was determined by reading abosrbence at 280 nm (Perkin-Elmer spectrophotometer, Coleman-124). Alkaline phosphatase activity was measured as described in the Methods section of Chapter I, using carbonate-bicarbonate buffer (100 mM) 2+ except when Ca was added to the incubation medium and glycine buffer (50 mM) was used. Estimation of the Gestational Age of the Guinea Pigs Pregnant guinea pigs at different stages of gestation were handled as described previously (Chapter I ) . The fetuses were weighed and the average weight per l i t t e r was used to estimate the placental age by using growth tables given by Draper (1920). Each placenta was processed separately and the membrane preparation of each placenta was 2+ assayed for Ca -ATPase (see Chapters I and III for details). The mean 2+ and the standard error of the specific activity of Ca -ATPase for each l i t t e r (3-5 fetuses) were calculated. 95 RESULTS Purification of Alkaline Phosphatase The elution profile of alkaline phosphatase activity and pro-tein is presented in Figure 28. The alkaline phosphatase activity peak appeared between fractions No. 11-16; the highest total activity and specific activity (500 pmole/mg protein) was found in fraction No. 14. 2+ Comparison Between Alkaline Phosphatase Activity and Ca -ATPase Activity 2+ The eluate was tested for Ca -ATPase activity under the standard 2+ + conditions for this enzyme (5 mM Ca ,5 mM ATP, 70 mM Na , and 20 mM Tris-HCl pH 8.2); no measurable activity could be detected in any fraction. 2+ Studying the Ca -ATPase activity in the previous steps revealed that 2+ the butanol extraction abolished completely the Ca -ATPase activity. 2+ However, though Ca -ATPase activity could not be detected, ATP was hydro-lyzed to a certain extent by the purified alkaline phosphatase fraction. the pH profile of ATP hydrolysis (Figure 29), shows that the pH optimum for ATP hydrolysis shifted from pH 8.2-8.5 (in the membrane preparation) 2+ (Figure 16) to pH 9.0-9.5; and only slight activation by 10 mM Ca 2+ was observed at the peak (pH 9.0-9.5). Ca was completely ineffective 2+ as a stimulator at a lower pH range. Mg showed basically the same profile with somewhat better efficiency as a stimulator of enzyme activity. The pH profile of p-nitrophenyi-phosphate hydrolysis by fractions No. 13-14 was typical of alkaline phosphatase with a pH optimum of 10.0-2+ 10.2, and only Mg stimulated hydrolysis. (Figure 30). The specific 96 U J < - J I Fraction na Figure 28. The elution profile of alkaline phosphatase and protein on Sephadex G-200 gel f i l t r a t i o n carried out by the method of Ghosh and Fishman (1968). X X, protein; \u00C2\u00A9 \u00C2\u00A9 , enzyme activity. The protein was eluted with 50 mM Tris-HCl pH 8.6 and collected in 10 ml fractions. 97 100% =27/ jmoles/mg Protein-hr-o Figure 29. The effect of pH on ATP hydrolysis by the purified alkaline phosphatase (fraction no. 13-14). o o, without diva-lent cations; X X, with 10 mM Ca2+; e \u00C2\u00A9, with 10 mM Mg2+. ATP concentration was 5 mM. 98 ^ 100% = 500 ju:moles/mg Protein \u00E2\u0080\u00A2 hr \u00E2\u0080\u0094 o 2 I o J i 1 1 1 1 1 7 8 9 10 II 12 PH Figure 30. The effect of pH on p-nitrophenyl phosphate hydrolysis by the purified alkaline phosphatase (fractions no. 13-14). o o, without divalent cations; X X, 10 mM Ca 2 +; \u00E2\u0080\u00A2 with 10 mM Mg2+. The samples were incubated in the presence of 5 mM p-m'trophenyl phosphate. 99 activity at the peak was 500 ymole/mg protein compared with only 27 ymole/mg protein (5.4% efficiency) when ATP served as a substrate under 2+ the same conditions (pH 10.0, 10 mM Mg ). The s t a b i l i t y on storage of the two enzyme activities in the membrane preparation was compared. Alkaline phosphatase was very unstable when stored at 4\u00C2\u00B0C, and less than 2+ 10% of the i n i t i a l activity was l e f t after 7 days (Figure 31). Ca -ATPase, on the other hand, lost only 10% of i t s i n i t i a l activity during 2+ the same period (Figure 31) and no further decrease in activity of Ca -ATPase was observed up to 2 months at 4\u00C2\u00B0C. The results of the comparison 2+ study between alkaline phosphatase and Ca -ATPase activities in the membrane preparation and the purified alkaline phosphatase fraction, are summarized in Table'IV. 2+ Ca -ATPase Specific Activity as a Function of Placental Age The results obtained from 21 guinea pigs with 3-5 fetuses per 2+ l i t t e r revealed that Ca -ATPase activity increased up to the 50th day, leveled between the 50th and the 60th day, and dropped after the 60th day (Figure 32). DISCUSSION It has been demonstrated clearly that the purified alkaline 2+ ' phosphatase does not possess any Ca -ATPase activity, and nor do any of the other eluted fractions. The only ATPase activity present is 2+ not dependent on Ca for hydrolyzing ATP. It seems very lik e l y from 100 Figure 31. The effect of storage at 4 C on enzyme activity. 2+ \u00E2\u0080\u00A2 e , Ca -ATPase; A \u00E2\u0080\u00A2 , alkaline phosphatase. Table IV. Comparison between alkaline phosphatase activity and ATPase activity in the membrane preparation and the purified alkaline phosphatase fraction. ymoles Pi release per mg protein per hour at 37\u00C2\u00B0C Net activation by (1) and (2) Substrate Preparation pH optimum No divalent cation (1) (2) 10 mM Mg 2 + 10 mM Ca 2 + Stability at 4\u00C2\u00B0C 5 mM p-nitro-phenyl phosphate Membrane preparation 10.0-10.2 ** 5 ** ** 12 0 Very unstable [Alkaline phos-phatase activity) Purified alkaline phosphatase fraction 10.0-10.2 *** 200 *** ** 300 0 5 mM Na2ATP Membrane preparation 8.2- 8.5 * 0.75 * * 30 50 Very stable (ATPase activity) Purified alkaline phosphatase fraction 9.0- 9.5 * 20 * * 7 3 Tris-HCl buffer (20 mM) Glycine buffer (50 mM) Carbonate-bicarbonate buffer (100 mM) 102 I5H O 25-\ a. E .1 2 0 CL o E S o < CD CO D CL i -\u00E2\u0080\u00A2-CM O o 10-J 5 H - T 1 1 1 1 3 0 4 0 5 0 6 0 7 0 Embryo Age in Days 2+ Figure 32. Ca -ATPase specific activity as a function of embryo age. Each experimental value represents the mean + S.E. for each l i t t e r . 103 the pH profile that ATP is being hydrolyzed by alkaline phosphatase and not by ATPase. Felix and Fleisch (1974) investigated the possibility that purified alkaline phosphatase from calf bone might demonstrate 2+ 2+ (Ca -Mg )-ATPase activity. They found that the only activation of 2+ 2+ ATP hydrolysis by Ca was obtained in the absence of Mg and always 2+ was substantially lower than the activation by Mg . Their conclusion 2+ 2+ was that calf bone alkaline phosphatase is not a \"true\" (Ca -Mg )-ATPase, as i t has been suggested for the intestinal alkaline phosphatase (Curzen and Morris, 1968; Haussler et_ al_., 1970). It has been shown that the extraction of the lipids with butanol 2+ from the placental plasma membranes abolished completely the Ca -ATPase activity of the preparation. Emmelot and Bos (1968) found that (Na +, K+)-ATPase activity depended on the l i p i d component of the preparation. 2+ Similar results were reported by MacLennan (1970) for the Ca -ATPase of the sarcoplasmic reticulum. He was able to restore 50% of the original 2+ activity by adding phospholipid to Ca -ATPase which was previously digested with phospholipase C. The exact role of the l i p i d is not known, however, i t might be that the lipoprotein association keeps the protein in the conformation required for enzymic activity. The lack of depen-dence of alkaline phosphatase on the l i p i d , and i t s inst a b i l i t y on storage 2+ at 4\u00C2\u00B0C compared with the very high s t a b i l i t y of the Ca -ATPase, are 2+ additional indications that Ca -ATPase is not just a manifestation of alkaline phosphatase. Another argument against this hypothesis can be drawn from the work of Manning et al_. (1970), who found that alkaline phosphatase from the guinea pig placenta is very heat-sensitive (16 min 104 2+ at 55\u00C2\u00B0C resulted in 95% inactivation). On the other hand, the Ca -ATPase was inactivated only at temperatures above 70\u00C2\u00B0C (Figure 23). 2+ The peak of Ca -ATPase activity in the developing placenta appears to occur between the 50th and 60th day of pregnancy, with an increase in activity up to the 50th day and a decrease after the 60th day. A similar profile was reported for alkaline phosphatase in the guinea pig (Hard, 1946). Twardock (1967) reported that trans-placental calcium transport in the guinea pig (in vivo study), peaks around the 55th day of pregnancy, and drops thereafter. Thus this profile is in agreement 2+ with the Ca -ATPase activity profile in the developing placenta. However, these profiles cannot account for the continuous increase of calcium deposition in the fetus throughout pregnancy (Twardock, 1967). SUMMARY The purified alkaline phosphatase fraction does not possess any 2+ Ca -ATPase activity. Disruption of the lipoprotein association of 2+ the membranes by butanol destroyed completely the Ca -ATPase activity, while increasing the alkaline phosphatase activity. In the membrane 2+ preparation Ca -ATPase is very stable on storage at 4\u00C2\u00B0C, while alkaline phosphatase activity decayed very rapidly. The specific activity profile 2+ of Ca -ATPase throughout the gestation period revealed a peak between the 50th and the 60th day of pregnancy, similar to the alkaline phos-phatase profile. The peak of enzyme activity corresponds with the re-2+ 2+ ported trans-placental Ca flux,^ but not with Ca -deposition in the fetus. 105 2+ In conclusion, i t was demonstrated very clearly that Ca -ATPase and alkaline phosphatase from the guinea pig placenta are two separate enzymes. DIVISION III -UPTAKE BY PLACENTAL PLASMA MEMBRANE VESICLES 106 CHAPTER V Ca2+-UPTAKE BY PLACENTAL PLASMA MEMBRANE VISICLES INTRODUCTION Transplacental calcium transport is asymmetrical. Presumably the ion must be transferred across at least the two plasma membranes which comprise the trophoblastic layer that separates the maternal and 2+ the fetal circulations. Studying Ca -transport in such a system as one unit is extremely d i f f i c u l t i f not impossible. However, active 2+ Ca -transport can be studied with less d i f f i c u l t y i f i t occurs across a single plasma membrane, which is clearly a necessary step for any asymmetrical active transport. To be able to measure such transport 2+ one must have a system in which Ca w i l l be concentrated. This require-ment is met by the placental membrane preparation which consists pre-dominantly of vesicles (Plates 3, 4) of plasma membrane origin. In previous chapters i t was demonstrated that the placental 2+ plasma membrane vesicles contain Ca -ATPase. This enzyme is s u f f i -2+ 2+ ciently similar to other Ca -ATPases involved in active Ca -transport, to suggest that these vesicles may also be involved. 2+ Ca -uptake by sarcoplasmic reticulum vesicles has been studied by many investigators for almost two decades, but numerous questions 107 108 are s t i l l unresolved, and to date there is no widely accepted model 2+ for Ca -transport. Part of the literature on this subject is reviewed by Inesi (1972). The relevant data w i l l be compared with the findings of this study in the discussion. This chapter was designed to answer very simple questions: 1) Can the placental plasma membrane vesicles accumulate calcium; i f so, how efficiently? 2) Is the uptake dependent on ATP hydrolysis? 2+ 2+ 3) What is the effect of the external Ca concentration on Ca -uptake? It is realized that many more questions must be answered before a firm 2+ conclusion can be made concerning the relationship of the Ca -related properties of the placental plasma membranes to the active transport of 2+ Ca . However, positive answers to the basic questions may suggest that the system is operating generally in the same manner as the sarco-plasmic reticulum (Inesi, 1972) and cardiac microsomes (Repke and Katz, 1972). 2+ Before any measurement of Ca -uptake can be made, the term uptake must be defined and distinguished from binding. The most widely 2+ used working definitions of Ca -uptake and binding are given by Entman et al_. (1973) and Repke and Katz (1972). Binding is defined as Ca 2 + accumulation in the presence of ATP but in the absence of a calcium-precipitating anion (oxalate or phosphate). This process is rapid and 2+ reaches a maximum within 1 min. Uptake is Ca accumulation in the presence of ATP and calcium-precipitating anion. This is a slower process. 109 When calculating uptake the binding should be subtracted. This is done by extrapolating the uptake to time zero. In this study binding and uptake are defined differently. Bound calcium is accessible for an immediate exchange and can be displaced by a competitor. Bound calcium is measured by labeling the membrane with 4 5 C a 2 + (in the absence of ATP) and displacing i t with \"cold\" calcium. 45 2+ Uptake is represented by the fraction of Ca which cannot be rapidly 40 2+ displaced by a large excess of Ca . The flow dialysis system described in Chapter I was adapted for 2+ measuring Ca -uptake. MATERIALS AND METHODS 2+ Measurement of Ca -uptake Placental plasma membrane vesicles were incubated at 26\u00C2\u00B0C in 20 mM Tris-HCl buffer pH 8.0, 100 mM NaCl, 10\"5M 4 5 C a 2 + and 4 0 C a 2 + as indicated, with or without 5 mM ATP (final volume 1.5 ml). At the end of the incubation period 3 x 0.1 ml aliquots were taken for the deter-mination of Pi release. One ml of the incubation medium was introduced 40 2+ to the upper chamber of the flow dialysis c e l l , 0.5 ml of 1 M Ca was added, and the diffusion rate was measured. The diffusion rate of the control without ATP was also measured. The difference in the dif-fusion rates represents the fraction of calcium which is not available 2+ for displacement; thus Ca -uptake can be calculated. After each measure-ment the upper chamber was emptied and washed three times with the effluent buffer (20 mM Tris-HCl pH 8.0, 100 mM NaCl) before the next 45 sample was introduced. Ca counting was carried out as described in Chapter I, protein was determined by the Lowry method (1951), and Pi released by the Gomori (1942) method. RESULTS 2+ Evaluation of the Flow Dialysis Method for Measuring Ca -uptake The usefulness of the method for measuring calcium uptake was tested by incubating vesicles for 2 hours in the presence of ATP. Dis-placement of 4 5 C a 2 + by 3 x 10-1M 4 0 C a 2 + was measured in the flow dialysis c e l l . The results are represented in Figure 33. It is evident that in 2+ the presence of 5 mM ATP, 5 mM Ca and the placental plasma membrane vesicles (Figure 33, column B), part of the 4 5 C a 2 + (25%) is not available for displacement after 2 hrs of incubation at 24\u00C2\u00B0C. From the results of the various controls i t is concluded that this fraction of non-2+ displacable Ca represents uptake by the vesicles and is not the result o f n o n s p e c i f i c u n e x c h a n g a b l e b i n d i n g t o any o f t h e c o n s t i t u e n t s o f t h e incubation f l u i d . Thus the s u i t a b i l i t y of the modified flow dialysis 2+ method for Ca -uptake studies is demonstrated. 2+ The Effect of Incubation Time on Ca -uptake and ATP Hydrolysis Calcium uptake and ATP hydrolysis by the placental plasma mem-2+ brane vesicles showed similar kinetic behaviour (Figure 34A). No Ca -uptake was detected in the absence of ATP. The ratio between ATP m LU w CD IO O g 0 E \u00C2\u00B0-CL O 2.5 2.0 1.5 CL 3 o> E rr E .2 "Thesis/Dissertation"@en . "10.14288/1.0093506"@en . "eng"@en . "Physiology"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "Calcium related properties of plasma membranes from guinea pig placenta"@en . "Text"@en . "http://hdl.handle.net/2429/19542"@en .